VOLUME FOUR HUNDRED AND THIRT Y-SEVEN
METHODS
IN
ENZYMOLOGY Globins and Other Nitric Oxide-Reactive Proteins, Part B
METHODS IN ENZYMOLOGY Editors-in-Chief
JOHN N. ABELSON AND MELVIN I. SIMON Division of Biology California Institute of Technology Pasadena, California Founding Editors
SIDNEY P. COLOWICK AND NATHAN O. KAPLAN
VOLUME FOUR HUNDRED AND THIRT Y-SEVEN
METHODS
IN
ENZYMOLOGY Globins and Other Nitric Oxide-Reactive Proteins, Part B EDITED BY
ROBERT K. POOLE Department of Molecular Biology and Biotechnology The University of Sheffield Sheffield, UK
AMSTERDAM • BOSTON • HEIDELBERG • LONDON NEW YORK • OXFORD • PARIS • SAN DIEGO SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO Academic Press is an imprint of Elsevier
Academic Press is an imprint of Elsevier 525 B Street, Suite 1900, San Diego, California 92101-4495, USA 84 Theobald’s Road, London WC1X 8RR, UK
This book is printed on acid-free paper.
⬁
Copyright # 2008, Elsevier Inc. All Rights Reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the Publisher. The appearance of the code at the bottom of the first page of a chapter in this book indicates the Publisher’s consent that copies of the chapter may be made for personal or internal use of specific clients. This consent is given on the condition, however, that the copier pay the stated per copy fee through the Copyright Clearance Center, Inc. (www.copyright.com), for copying beyond that permitted by Sections 107 or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating new collective works, or for resale. Copy fees for pre-2008 chapters are as shown on the title pages. If no fee code appears on the title page, the copy fee is the same as for current chapters. 0076-6879/2008 $35.00 Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone: (þ44) 1865 843830, fax: (þ44) 1865 853333, E-mail: permissions@elsevier. com. You may also complete your request on-line via the Elsevier homepage (http://elsevier.com), by selecting ‘‘Support & Contact’’ then ‘‘Copyright and Permission’’ and then ‘‘Obtaining Permissions.’’ For information on all Elsevier Academic Press publications visit our Web site at www.books.elsevier.com ISBN: 978-0-12-374278-0 PRINTED IN THE UNITED STATES OF AMERICA 08 09 10 11 9 8 7 6 5 4 3 2 1
CONTENTS
Contributors Preface Volumes in Series
xvii xxvii xxix
Section I. Nitric Oxide-Metabolising and Detoxifying Enzymes
1
1. Structural Studies on Flavodiiron Proteins
3
Joa˜o B. Vicente, Maria Arme´nia Carrondo, Miguel Teixeira, and Carlos Fraza˜o 1. Introduction 2. Crystallization of Flavodiiron Proteins 3. Diffraction Data Collection, Structure Determination, and Refinement 4. Overall Description of Structures 5. Conclusion References
2. Biochemical, Spectroscopic, and Thermodynamic Properties of Flavodiiron Proteins
4 4 7 8 16 17
21
Joa˜o B. Vicente, Marta C. Justino, Vera L. Gonc¸alves, Lı´gia M. Saraiva, and Miguel Teixeira 1. Introduction 2. Cloning of Genes Encoding Flavodiiron Proteins and Their Truncated Domains 3. Production and Purification of Recombinant Flavodiiron Proteins 4. Biochemical Characterization of Flavodiiron Proteins 5. Spectroscopic Properties 6. Redox Properties 7. Conclusions Acknowledgments References
22 24 25 26 29 32 37 42 42
v
vi
Contents
3. Kinetic Characterization of the Escherichia coli Nitric Oxide Reductase Flavorubredoxin
47
Joa˜o B. Vicente, Francesca M. Scandurra, Elena Forte, Maurizio Brunori, Paolo Sarti, Miguel Teixeira, and Alessandro Giuffre` 1. Introduction 2. Amperometric Measurements 3. Spectroscopic Measurements 4. Conclusions Acknowledgments References
4. Escherichia coli Cytochrome c Nitrite Reductase NrfA
48 49 51 61 61 61
63
Thomas A. Clarke, Paul C. Mills, Susie R. Poock, Julea N. Butt, Myles R. Cheesman, Jeffrey A. Cole, Jay C. D. Hinton, Andrew M. Hemmings, ¨derberg, Stephen Spiro, Jessica Van Gemma Kemp, Christopher A. G. So Wonderen, and David J. Richardson 1. Introduction 2. Measurement of Cytochrome c Nitrite Reductase-Dependent Consumption of Nitric Oxide in Whole Cells 3. Growth of E. coli Optimized for Cytochrome c Nitrite Reductase Production for Use in Enzyme Purification 4. Purification of Cytochrome c Nitrite Reductase 5. Assaying the Cytochrome c Nitrite Reductase 6. Crystallization of E. coli Cytochrome c Nitrite Reductase 7. Concluding Remarks Acknowledgments References
5. The Respiratory Nitric Oxide Reductase (NorBC) from Paracoccus denitrificans
64 66 66 68 69 73 74 76 76
79
Sarah J. Field, Faye H. Thorndycroft, Andrey D. Matorin, David J. Richardson, and Nicholas J. Watmough 1. Introduction 2. Purification of Native NorBC from Paracoccus denitrificans 3. Purification of Recombinant NorBC 4. Amperometric Assays of NO Consumption 5. Pseudoazurin as an Electron Donor in Assays of NorBC 6. Preparation of NOR for Spectroscopic Investigation 7. Electron Paramagnetic Resonance Spectroscopy 8. Concluding Remarks Acknowledgments References
80 82 85 86 88 91 96 98 99 99
Contents
6. Redox-Controlled Dinitrosyl Formation at the Diiron-Oxo Center of NorA
vii
103
Rainer Cramm and Katja Strube 1. Introduction 2. Genetic Context and Expression of the NorA Gene in R. eutropha 3. Purification of NorA 4. Disulfide Bridges in NorA 5. Iron Analysis and Preparation of Apo-NorA 6. Interconversion of Redox Forms of NorA 7. Generation of NorA-DNIC In Vitro 8. Preparation of NorA-DNIC Formed In Vivo 9. Quantification of NO from NorA-DNIC 10. Outlook References
7. Purification and Functional Analysis of Fungal Nitric Oxide Reductase Cytochrome P450nor
104 105 106 107 108 109 111 113 113 114 114
117
Li Zhang and Hirofumi Shoun 1. Introduction 2. Screening of P450nor Activity 3. Gas Analysis 4. Purification of P450nor 5. Nitric Oxide Reductase Activity Assay 6. Protein Sequencing 7. Isolation of cDNA 8. Subcellular Fractionation of T. cutaneum 9. Site-Directed Mutagenesis 10. Expression of Recombinant Proteins 11. Purification of Recombinant Proteins 12. Titration of NAD Analogs 13. Stopped-Flow Rapid Scan Analysis 14. Other Analysis 15. Conclusion Acknowledgments References
8. A Quantitative Approach to Nitric Oxide Inhibition of Terminal Oxidases of the Respiratory Chain
118 119 119 120 121 122 123 124 125 126 127 128 130 131 131 131 131
135
Maria G. Mason, Rebecca S. Holladay, Peter Nicholls, Mark Shepherd, and Chris E. Cooper 1. Introduction 2. Evaluation of Current Techniques for Measuring pNO, pO2, and KM (O2)
136 137
viii
Contents
3. 4. 5. 6.
Nitric Oxide Donor Compounds Nitric Oxide Kinetics Oxygen Kinetics Optical Detection of Enzyme Intermediates in the Presence of Oxygen and NO Appendices Acknowledgments References
Section II. Sensor Proteins 9. Cloning, Expression, and Purification of the N-terminal Heme-Binding Domain of Globin-Coupled Sensors
138 139 149 151 153 156 156
161 163
Jennifer A. Saito, Tracey Allen K. Freitas, and Maqsudul Alam 1. Introduction 2. Bioinformatic Search of Globin-Coupled Sensors 3. Functional Analysis of Globin-Coupled Sensors Acknowledgments References
10. Oxygen-Sensing Histidine-Protein Kinases: Assays of Ligand Binding and Turnover of Response-Regulator Substrates
164 164 166 171 171
173
Marie-Alda Gilles-Gonzalez, Gonzalo Gonzalez, Eduardo Henrique Silva Sousa, and Jason Tuckerman 1. Introduction 2. Assays Acknowledgments References
11. Reactions of Nitric Oxide and Oxygen with the Regulator of Fumarate and Nitrate Reduction, a Global Transcriptional Regulator, during Anaerobic Growth of Escherichia coli
174 175 187 187
191
Jason C. Crack, Nick E. Le Brun, Andrew J. Thomson, Jeffrey Green, and Adrian J. Jervis 1. Introduction 2. Production of 4Fe-FNR Protein 3. Determination of Iron and Acid-Labile Sulfide Content of FNR 4. UV-Visible Absorbance Spectra of FNR 5. Cluster Reaction with Nitric Oxide and Oxygen 6. Purification of 2Fe-FNR 7. Detection of Other Reaction Products 8. Conclusions References
192 194 197 198 198 204 204 206 207
Contents
12. Genome-Wide Identification of Binding Sites for the Nitric Oxide-Sensitive Transcriptional Regulator NsrR
ix
211
Sam Efromovich, David Grainger, Diane Bodenmiller, and Stephen Spiro 1. Introduction 2. Strain Construction 3. Reference and Control Samples 4. Culture Conditions 5. Immunoprecipitation of DNA Targets Associated with NsrR 6. DNA Labeling, Microarray Hybridization, and Processing 7. Visualization and Analysis of DNA Microarray Data 8. A New Statistical Methodology for Treatment of Chip-on-Chip Data 9. Conclusions Acknowledgments References
13. Characterization of the Nitric Oxide-Reactive Transcriptional Activator NorR
212 214 216 217 218 219 220 222 231 231 231
235
Benoıˆt D’Autre´aux, Nick Tucker, Stephen Spiro, and Ray Dixon 1. 2. 3. 4.
Introduction Measurement of NorR Activity In Vivo Measurement of Transcriptional Activation by NorR In Vitro Detection of the Ferrous-Nitrosyl Form of NorR by In Vivo Electron Paramagnetic Resonance (EPR) 5. In Vitro Reconstitution of the Iron Center in NorR 6. Measurement of NO Affinity 7. Standardization of the NO Electrode 8. Determination of NorRFe(NO) Kd 9. Conclusions Acknowledgment References
Section III. Advanced Spectroscopic Methods 14. Hemoglobins from Mycobacterium tuberculosis and Campylobacter jejuni: A Comparative Study with Resonance Raman Spectroscopy
236 237 238 240 242 243 246 247 248 248 249
253
255
Changyuan Lu, Tsuyoshi Egawa, Masahiro Mukai, Robert K. Poole, and Syun-Ru Yeh 1. Hemoglobin Superfamily: An Overview 2. Microbial Hemoglobins 3. Resonance Raman Spectroscopy: Applications in Hemeproteins
256 257 258
x
Contents
4. Structures and Functions of Microbial Hemoglobins 5. Closing Remarks Acknowledgments References
266 281 282 282
15. The Power of Using Continuous-Wave and Pulsed Electron Paramagnetic Resonance Methods for the Structure Analysis of Ferric Forms and Nitric Oxide-Ligated Ferrous Forms of Globins 287 Sabine Van Doorslaer and Filip Desmet 1. Introduction 2. Electron Paramagnetic Resonance in a Nutshell 3. EPR Studies of NO-Ligated Globins 4. EPR Studies of Ferric globins 5. Spin-Labeling Heme Proteins 6. Future Challenges and Possibilities Acknowledgments References
16. Oxygen Binding to Heme Proteins in Solution, Encapsulated in Silica Gels, and in the Crystalline State
288 289 295 301 304 305 305 306
311
Luca Ronda, Stefano Bruno, Serena Faggiano, Stefano Bettati, and Andrea Mozzarelli 1. Oxygen-Binding Curves to Heme Proteins 2. Determination of OBCs for Hemoglobin in Solution 3. Determination of K1 for Hemoglobin in Solution in the Absence of Allosteric Effectors 4. Determination of OBCs for T State Hemoglobin Gels in the Absence and Presence of Allosteric Effectors 5. Determination of OBCs for T State Hemoglobin Crystals 6. Determination of OBCs for Hemocyanin in Solution and in Silica Gels Acknowledgments References
17. Characterization of Ligand Migration Mechanisms inside Hemoglobins from the Analysis of Geminate Rebinding Kinetics
313 316 318 318 320 323 325 325
329
Stefania Abbruzzetti, Stefano Bruno, Serena Faggiano, Luca Ronda, Elena Grandi, Andrea Mozzarelli, and Cristiano Viappiani 1. 2. 3. 4.
Introduction Principles of Nanosecond Laser Flash Photolysis Basic Experimental Layouts Encapsulation of Hbs in Silica Gels
330 330 331 335
Contents
5. Enhancement of Geminate Rebinding and Advantages of Gel Encapsulation 6. Extraction of Kinetic Information Acknowledgments References
18. Ligand Dynamics in Heme Proteins Observed by Fourier Transform Infrared Spectroscopy at Cryogenic Temperatures
xi
336 337 342 342
347
Karin Nienhaus and G. Ulrich Nienhaus 1. Introduction 2. Materials 3. Fourier Transform Infrared Cryospectroscopy 4. Low-Temperature FTIR Spectroscopy on NO-Ligated Heme Proteins 5. Concluding Remarks Acknowledgments References
19. Time-Resolved X-Ray Crystallography of Heme Proteins
348 349 353 365 373 374 374
379
Vukica Sˇrajer and William E. Royer, Jr. 1. Introduction 2. Experiment 3. Data Processing and Analysis 4. A Case Study: Scapharca Dimeric Hemoglobin 5. Conclusions Acknowledgments References
20. Structural Dynamics of Myoglobin
379 381 385 388 391 393 393
397
M. Brunori, D. Bourgeois, and B. Vallone 1. Background 2. Crystallographic Studies of Myoglobin States 3. Experimental Approaches Acknowledgments References
21. Use of the Conjugate Peak Refinement Algorithm for Identification of Ligand-Binding Pathways in Globins
398 399 400 413 413
417
Stephen D. Golden and Kenneth W. Olsen 1. Introduction 2. Exploration of Oxygen-Binding Pathways in Myoglobin
418 418
xii
Contents
3. Theoretical Models 4. Potential Energy Function 5. Transition Pathways 6. Methods 7. Results 8. Conclusions References
22. Finding Gas Migration Pathways in Proteins Using Implicit Ligand Sampling
419 420 421 425 429 432 433
439
Jordi Cohen, Kenneth W. Olsen, and Klaus Schulten 1. Introduction 2. Methods 3. Example Calculation: Truncated Hemoglobin (trHb) from Paramecium caudatum 4. Discussion Acknowledgments References
440 442 446 449 455 456
23. Identification of Ligand-Binding Pathways in Truncated Hemoglobins Using Locally Enhanced Sampling Molecular Dynamics
459
Stephen D. Golden and Kenneth W. Olsen 1. Introduction 2. Molecular Dynamics 3. Locally Enhanced Sampling Molecular Dynamics 4. Methods 5. Results 6. Conclusions References
24. Nitric Oxide Reactivity with Globins as Investigated Through Computer Simulation
460 462 465 466 468 471 472
477
Marcelo A. Marti, Luciana Capece, Axel Bidon-Chanal, Alejandro Crespo, Victor Guallar, F. Javier Luque, and Dario A. Estrin 1. 2. 3. 4.
Introduction Molecular Dynamics (MD) Methods Quantum Mechanical-Molecular Mechanical Methods Illustrative Examples
478 479 485 488
Contents
5. Ligand Migration Profiles from MSMD and PELE Simulations: Exploring Ligand Entry Pathways in M. tuberculosis trHbN 6. Conclusions Acknowledgments References
25. Microbial Responses to Nitric Oxide and Nitrosative Stress: Growth, ‘‘Omic,’’ and Physiological Methods
xiii
490 494 495 495
499
Steven T. Pullan, Claire E. Monk, Lucy Lee, and Robert K. Poole 1. Introduction 2. Methods 3. Nitric Oxide, NO-Releasing Agents, and Nitrosating Agents 4. Illustrative Results from Applications of These Methods References
26. Analysis of Nitric Oxide-Dependent Antimicrobial Actions in Macrophages and Mice
500 504 512 514 516
521
Andre´s Vazquez-Torres, Tania Stevanin, Jessica Jones-Carson, Margaret Castor, Robert C. Read, and Ferric C. Fang 1. NO-Dependent Antimicrobial Actions of Murine Macrophages 2. NO-Dependent Antimicrobial Actions of Human Macrophages 3. NO-dependent Antimicrobial Actions in Laboratory Mice References
27. Measuring Nitric Oxide Metabolism in the Pathogen Neisseria meningitidis
522 528 532 536
539
Melanie J. Thomson, Tania M. Stevanin, and James W. B. Moir 1. 2. 3. 4.
Introduction Safety Aspects of Handling N. meningitidis in the Laboratory Metabolism of Neisseria sp. Experimental Approaches to Analyzing Nitrogen Metabolism Relevant to NO 5. Simultaneous Measurement of Oxygen and NO during Pure Culture of N. meningitidis 6. Measurement of NO Production/Disappearance in Tissue Culture Using Human Monocyte-Derived Macrophages References
540 541 541 544 547 555 558
xiv
Contents
28. Localization of S-Nitrosothiols and Assay of Nitric Oxide Synthase and S-Nitrosoglutathione Reductase Activity in Plants
561
Francisco J. Corpas, Alfonso Carreras, Francisco J. Esteban, Mounira Chaki, Raquel Valderrama, Luis A. Del Rı´o, and Juan B. Barroso 1. Introduction 2. Determination of L-Arginine-Dependent NOS Activity by Ozone Chemiluminescence in Plant Tissues 3. Assay of GSNOR Activity 4. Localization of S-Nitrosothiols and S-Nitrosoglutathione in Plant Tissues by Confocal Laser-Scanning Microscopy (CLSM) 5. Conclusion Acknowledgments References
29. Methods for Nitric Oxide Detection during Plant–Pathogen Interactions
562 563 566 567 571 572 572
575
E. Vandelle and M. Delledonne 1. 2. 3. 4. 5. 6.
Introduction Nitric Oxide Detection by Mass Spectrometry Nitric Oxide Detection by Laser Photoacoustic Spectroscopy Nitric Oxide Detection by Chemiluminescence Nitric Oxide Detection by Hemoglobin Conversion Nitric Oxide Detection by Electron Paramagnetic Resonance (EPR) Spin Trap 7. Nitric Oxide Detection Using Diaminofluoresceins 8. Conclusion References
30. Bioimaging Techniques for Subcellular Localization of Plant Hemoglobins and Measurement of Hemoglobin-Dependent Nitric Oxide Scavenging In Planta
576 577 579 582 583 585 587 590 591
595
Kim H. Hebelstrup, Erik stergaard-Jensen, and Robert D. Hill 1. Introduction 2. Measuring Hemoglobin-Dependent NO Scavenging 3. Techniques for Determination of Subcellular Localization of Plant Hemoglobins 4. Imaging of Hemoglobin-Dependent NO Scavenging in Arabidopsis Plants 5. Engineering of GLB1-GFP/GLB2-GFP Constructs and Microscopic Analysis of A. thaliana Plants Expressing GFP-Tagged Hemoglobin References
596 596 597 598 600 603
Contents
31. Use of Recombinant Iron-Superoxide Dismutase as A Marker of Nitrative Stress
xv
605
Estı´baliz Larrainzar, Estı´baliz Urarte, In˜igo Auzmendi, Idoia Ariz, Cesar Arrese-Igor, Esther M. Gonza´lez, and Jose F. Moran 1. Introduction 2. Immunodetection of Nitrated Proteins: Metal-Mediated Tyrosine Nitration of BSA 3. Tyrosine Nitration of Purified Recombinant Vu_FeSOD Affects its Enzymatic Activity 4. Tyrosine Nitration in Vu_FeSOD can be Estimated Using Antibodies Against 3-Nitrotyrosine 5. SIN-1-Dependent Vu_FeSOD Nitration can be Detected by the Loss of Enzymatic Activity Acknowledgments References Author Index Subject Index
606 608 610 612 612 616 616 619 647
This page intentionally left blank
CONTRIBUTORS
Stefania Abbruzzetti Dipartimento di Fisica, Universita` degli Studi di Parma, Parma, Italy, and NEST CNR-INFM, Pisa, Italy Maqsudul Alam Department of Microbiology, and Advanced Studies in Genomics, Proteomics, and Bioinformatics, College of Natural Sciences, University of Hawaii, Honolulu, Hawaii Cesar Arrese-Igor Departamento de Ciencias del Medio Natural, Universidad Pu´blica de Navarra, Campus de Arrosadia, E-31006 Pamplona, Navarre, Spain Idoia Ariz Instituto de Agrobiotecnologı´a, Universidad Pu´blica de Navarra-CSIC-Gobierno de Navarra, Campus de Arrosadı´a, E-31006 Pamplona, Navarre, Spain In˜igo Auzmendi Departamento de Ciencias del Medio Natural, Universidad Pu´blica de Navarra, Campus de Arrosadia, E-31006 Pamplona, Navarre, Spain Juan B. Barroso Grupo de Sen˜alizacio´n Molecular y Sistemas Antioxidantes en Plantas, Unidad ´ rea de Bioquı´mica y Biologı´a Molecular, Universidad Asociada al CSIC (EEZ), A de Jae´n, E-23071 Jae´n, Spain Stefano Bettati Department of Biochemistry and Molecular Biology, University of Parma, Parma, Italy Axel Bidon-Chanal Departament de Fisicoquı´mica, Facultat de Farma`cia, Universitat de Barcelona, Barcelona, Spain Diane Bodenmiller School of Biology, Georgia Institute of Technology, Atlanta, Georgia D. Bourgeois Institut de Biologie Structurale Jean-Pierre Ebel, CEA, CNRS, Universite´ Joseph Fourier and European Synchrotron Radiation Facility, Grenoble Cedex, France
xvii
xviii
Contributors
Nick E. Le Brun Centre for Metalloprotein Spectroscopy and Biology, School of Chemical Sciences and Pharmacy, University of East Anglia, Norwich, United Kingdom Stefano Bruno Department of Biochemistry and Molecular Biology, University of Parma, Parma, Italy, and Dipartimento di Biochimica e Biologia Molecolare, Universita` degli Studi di Parma, Parma, Italy M. Brunori Dipartimento di Scienze Biochimiche ‘‘A. Rossi Fanelli,’’ Universita` di Roma ‘‘La Sapienza,’’ Roma, Italy Maurizio Brunori Department of Biochemical Sciences, CNR Institute of Molecular Biology and Pathology and Istituto Pasteur–Fondazione Cenci Bolognetti Sapienza, University of Rome, Rome, Italy Julea N. Butt Centre for Metalloprotein Spectroscopy and Biology, School of Biological Sciences, and School of Chemical Sciences and Pharmacy, University of East Anglia, Norwich, United Kingdom Luciana Capece Departamento de Quı´mica Inorga´nica, Analı´tica y Quı´mica Fı´sica/INQUIMAECONICET, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Ciudad Universitaria, Buenos Aires, Argentina Alfonso Carreras Grupo de Sen˜alizacio´n Molecular y Sistemas Antioxidantes en Plantas, Unidad ´ rea de Bioquı´mica y Biologı´a Molecular, Universidad Asociada al CSIC (EEZ), A de Jae´n, E-23071 Jae´n, Spain Maria Arme´nia Carrondo Instituto de Tecnologia Quı´mica e Biolo´gica, Universidade Nova de Lisboa, Oeiras, Portugal Margaret Castor University of Washington School of Medicine, Seattle, Washington Mounira Chaki Grupo de Sen˜alizacio´n Molecular y Sistemas Antioxidantes en Plantas, Unidad ´ rea de Bioquı´mica y Biologı´a Molecular, Universidad Asociada al CSIC (EEZ), A de Jae´n, E-23071 Jae´n, Spain Myles R. Cheesman Centre for Metalloprotein Spectroscopy and Biology, School of Chemical Sciences and Pharmacy, University of East Anglia, Norwich, United Kingdom
Contributors
xix
Thomas A. Clarke Centre for Metalloprotein Spectroscopy and Biology, School of Biological Sciences, University of East Anglia, Norwich, United Kingdom Jordi Cohen Beckman Institute, University of Illinois, Urbana, Illinois Jeffrey A. Cole School of Biosciences, University of Birmingham, Edgbaston, Birmingham Chris E. Cooper Department of Biological Sciences, University of Essex, Colchester, United Kingdom Francisco J. Corpas Departamento de Bioquimica, Biologı´a Celular y Molecular de plantas, Estacio´n Experimental del Zaidı´n, CSIC, Spain Jason C. Crack Centre for Metalloprotein Spectroscopy and Biology, School of Chemical Sciences and Pharmacy, University of East Anglia, Norwich, United Kingdom Rainer Cramm Institut fu¨r Biologie/Mikrobiologie, Humboldt-Universita¨t zu Berlin, Berlin, Germany Alejandro Crespo Departamento de Quı´mica Inorga´nica, Analı´tica y Quı´mica Fı´sica/INQUIMAECONICET, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Ciudad Universitaria, Buenos Aires, Argentina Benoıˆt D’Autre´aux Laboratoire Stress Oxydant et Cancer, Service de Biologie Inte´grative et Ge´ne´tique Mole´culaire, Institut de Biologie et de Technologies de Saclay, CEA-Saclay, Gif-sur-Yvette Cedex, France M. Delledonne Dipartimento Scientifico e Tecnologico, Universita` degli Studi di Verona, Verona, Italy Filip Desmet University of Antwerp, Department of Physics, SIBAC Laboratory, Antwerp, Belgium Ray Dixon John Innes Centre, Colney, Norwich, United Kingdom Sabine Van Doorslaer University of Antwerp, Department of Physics, SIBAC Laboratory, Antwerp, Belgium
xx
Contributors
Sam Efromovich Department of Mathematical Sciences, University of Texas at Dallas, Richardson, Texas Tsuyoshi Egawa Department of Physiology and Biophysics, Albert Einstein College of Medicine, Bronx, New York Francisco J. Esteban Grupo de Sen˜alizacio´n Molecular y Sistemas Antioxidantes en Plantas, Unidad ´ rea de Bioquı´mica y Biologı´a Molecular, Universidad Asociada al CSIC (EEZ), A de Jae´n, E-23071 Jae´n, Spain Dario A. Estrin Departamento de Quı´mica Inorga´nica, Analı´tica y Quı´mica Fı´sica/INQUIMAECONICET, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Ciudad Universitaria, Buenos Aires, Argentina Serena Faggiano Department of Biochemistry and Molecular Biology, University of Parma, Parma, Italy, and Dipartimento di Biochimica e Biologia Molecolare, Universita` degli Studi di Parma, Parma, Italy Ferric C. Fang University of Washington School of Medicine, Seattle, Washington Sarah J. Field Center for Metalloprotein Spectroscopy and Biology, School of Biological Sciences, University of East Anglia, Norwich, United Kingdom Elena Forte Department of Biochemical Sciences, CNR Institute of Molecular Biology and Pathology and Istituto Pasteur–Fondazione Cenci Bolognetti Sapienza, University of Rome, Rome, Italy Carlos Fraza˜o Instituto de Tecnologia Quı´mica e Biolo´gica, Universidade Nova de Lisboa, Oeiras, Portugal Tracey Allen K. Freitas Department of Microbiology, University of Hawaii, Honolulu, Hawaii Marie-Alda Gilles-Gonzalez Department of Biochemistry, University of Texas Southwestern Medical Center, Dallas, Texas Alessandro Giuffre` Department of Biochemical Sciences, CNR Institute of Molecular Biology and Pathology and Istituto Pasteur–Fondazione Cenci Bolognetti Sapienza, University of Rome, Rome, Italy
Contributors
xxi
Stephen D. Golden Department of Chemistry, Loyola University Chicago, Chicago, Illinois Vera L. Gonc¸alves Instituto de Tecnologia Quı´mica e Biolo´gica, Universidade Nova de Lisboa, Oeiras, Portugal Esther M. Gonza´lez Departamento de Ciencias del Medio Natural, Universidad Pu´blica de Navarra, Campus de Arrosadia, E-31006 Pamplona, Navarre, Spain Gonzalo Gonzalez Department of Biochemistry, University of Texas Southwestern Medical Center, Dallas, Texas David Grainger School of Biosciences, University of Birmingham, Edgbaston, Birmingham, United Kingdom Elena Grandi Dipartimento di Fisica, Universita` degli Studi di Parma, Parma, Italy Jeffrey Green Department of Molecular Biology and Biotechnology, University of Sheffield, Sheffield, United Kingdom Victor Guallar Catalan Institute for Research and Advanced Studies (ICREA), Computational Biology Program, Barcelona Supercomputing Center, Barcelona, Spain Kim H. Hebelstrup Department of Plant Science, University of Manitoba, Winnipeg, Manitoba, Canada Andrew M. Hemmings Centre for Metalloprotein Spectroscopy and Biology, School of Biological Sciences, and School of Chemical Sciences and Pharmacy, University of East Anglia, Norwich, United Kingdom Robert D. Hill Department of Plant Science, University of Manitoba, Winnipeg, Manitoba, Canada Jay C. D. Hinton Institute of Food Research, Norwich, United Kingdom Rebecca S. Holladay Department of Biological Sciences, University of Essex, Colchester, United Kingdom Adrian J. Jervis Department of Molecular Biology and Biotechnology, University of Sheffield, Sheffield, United Kingdom
xxii
Contributors
Jessica Jones-Carson University of Colorado Health Sciences Center, Aurora, Colorado Marta C. Justino Instituto de Tecnologia Quı´mica e Biolo´gica, Universidade Nova de Lisboa, Oeiras, Portugal Gemma Kemp Centre for Metalloprotein Spectroscopy and Biology, School of Chemical Sciences and Pharmacy, University of East Anglia, Norwich, United Kingdom Estı´baliz Larrainzar Departamento de Ciencias del Medio Natural, Universidad Pu´blica de Navarra, Campus de Arrosadia, E-31006 Pamplona, Navarre, Spain Lucy Lee Department of Molecular Biology and Biotechnology, University of Sheffield, Sheffield, United Kingdom Changyuan Lu Department of Physiology and Biophysics, Albert Einstein College of Medicine, Bronx, New York F. Javier Luque Departament de Fisicoquı´mica, Facultat de Farma`cia, Universitat de Barcelona, Barcelona, Spain Marcelo A. Marti Departamento de Quı´mica Inorga´nica, Analı´tica y Quı´mica Fı´sica/INQUIMAECONICET, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Ciudad Universitaria, Buenos Aires, Argentina Maria G. Mason Department of Biological Sciences, University of Essex, Colchester, United Kingdom Andrey D. Matorin Center for Metalloprotein Spectroscopy and Biology, School of Biological Sciences, University of East Anglia, Norwich, United Kingdom Paul C. Mills Centre for Metalloprotein Spectroscopy and Biology, School of Biological Sciences, University of East Anglia, Norwich, United Kingdom James W. B. Moir Department of Biology, University of York, Heslington, York Claire E. Monk Department of Molecular Biology and Biotechnology, University of Sheffield, Sheffield, United Kingdom
Contributors
xxiii
Jose F. Moran Instituto de Agrobiotecnologı´a, Universidad Pu´blica de Navarra-CSIC-Gobierno de Navarra, Campus de Arrosadı´a, E-31006 Pamplona, Navarre, Spain Andrea Mozzarelli Department of Biochemistry and Molecular Biology, University of Parma, Parma, Italy, and Dipartimento di Biochimica e Biologia Molecolare, Universita` degli Studi di Parma, Parma, Italy Masahiro Mukai Mitsubishi Kagaku Institute of Life Sciences, Minamiooya, Machida, Tokyo, Japan Peter Nicholls Department of Biological Sciences, University of Essex, Colchester, United Kingdom Karin Nienhaus Institute of Biophysics, University of Ulm, Ulm, Germany G. Ulrich Nienhaus Institute of Biophysics, University of Ulm, Ulm, Germany, and Department of Physics, University of Illinois at Urbana–Champaign, Urbana, Illinois Erik stergaard-Jensen Department of Molecular Biology, University of Aarhus, Aarhus, Denmark Kenneth W. Olsen Department of Chemistry, Loyola University Chicago, Chicago, Illinois Susie R. Poock Centre for Metalloprotein Spectroscopy and Biology, School of Biological Sciences, University of East Anglia, Norwich, United Kingdom Robert K. Poole Department of Molecular Biology and Biotechnology, University of Sheffield, Sheffield, United Kingdom Steven T. Pullan Department of Molecular Biology and Biotechnology, University of Sheffield, Sheffield, United Kingdom Robert C. Read University of Sheffield, Sheffield, United Kingdom David J. Richardson Center for Metalloprotein Spectroscopy and Biology, School of Biological Sciences, University of East Anglia, Norwich, United Kingdom Luis A. Del Rı´o Departamento de Bioquimica, Biologı´a Celular y Molecular de plantas, Estacio´n Experimental del Zaidı´n (EEZ), CSIC, Apartado 419, E-18080 Grannada, Spain
xxiv
Contributors
Luca Ronda Department of Biochemistry and Molecular Biology, University of Parma, Parma, Italy, and Dipartimento di Biochimica e Biologia Molecolare, Universita` degli Studi di Parma, Parma, Italy William E. Royer, Jr. Department of Biochemistry and Molecular Pharmacology, University of Massachusetts Medical School, Worcester, Massachusetts Jennifer A. Saito Department of Microbiology, University of Hawaii, Honolulu, Hawaii Lı´gia M. Saraiva Instituto de Tecnologia Quı´mica e Biolo´gica, Universidade Nova de Lisboa, Oeiras, Portugal Paolo Sarti Department of Biochemical Sciences, CNR Institute of Molecular Biology and Pathology and Istituto Pasteur–Fondazione Cenci Bolognetti Sapienza, University of Rome, Rome, Italy Francesca M. Scandurra Department of Biochemical Sciences, CNR Institute of Molecular Biology and Pathology and Istituto Pasteur–Fondazione Cenci Bolognetti Sapienza, University of Rome, Rome, Italy Klaus Schulten Beckman Institute, University of Illinois, Urbana, Illinois Vukica Sˇrajer Center for Advanced Radiation Sources, University of Chicago, Illinois ¨derberg Christopher A. G. So Centre for Metalloprotein Spectroscopy and Biology, School of Biological Sciences, University of East Anglia, Norwich, United Kingdom Mark Shepherd University of Sheffield, Western Bank, Sheffield, United Kingdom Hirofumi Shoun Department of Biotechnology, University of Tokyo, Bunkyo-ku, Tokyo, Japan Eduardo Henrique Silva Sousa Department of Biochemistry, University of Texas Southwestern Medical Center, Dallas, Texas Stephen Spiro Department of Molecular and Cell Biology, University of Texas at Dallas, Richardson, Texas
Contributors
xxv
Tania M. Stevanin School of Medicine and Biomedical Science, University of Sheffield, Sheffield, United Kingdom Tania Stevanin University of Sheffield, Sheffield, United Kingdom Katja Strube Institut fu¨r Biologie/Mikrobiologie, Humboldt-Universita¨t zu Berlin, Berlin, Germany Miguel Teixeira Instituto de Tecnologia Quı´mica e Biolo´gica, Universidade Nova de Lisboa, Oeiras, Portugal Andrew J. Thomson Centre for Metalloprotein Spectroscopy and Biology, School of Chemical Sciences and Pharmacy, University of East Anglia, Norwich, United Kingdom Faye H. Thorndycroft Center for Metalloprotein Spectroscopy and Biology, School of Biological Sciences, University of East Anglia, Norwich, United Kingdom Melanie J. Thomson Department of Biology, University of York, Heslington, York Nick Tucker John Innes Centre, Colney, Norwich, United Kingdom Jason Tuckerman Department of Biochemistry, University of Texas Southwestern Medical Center, Dallas, Texas Estı´baliz Urarte Instituto de Agrobiotecnologı´a, Universidad Pu´blica de Navarra-CSIC-Gobierno de Navarra, Campus de Arrosadı´a, E-31006 Pamplona, Navarre, Spain Raquel Valderrama Grupo de Sen˜alizacio´n Molecular y Sistemas Antioxidantes en Plantas, Unidad ´ rea de Bioquı´mica y Biologı´a Molecular, Universidad Asociada al CSIC (EEZ), A de Jae´n, E-23071 Jae´n, Spain B. Vallone Dipartimento di Scienze Biochimiche ‘‘A. Rossi Fanelli,’’ Universita` di Roma ‘‘La Sapienza,’’ Roma, Italy E. Vandelle Dipartimento Scientifico e Tecnologico, Universita` degli Studi di Verona, Verona, Italy
xxvi
Contributors
Andre´s Vazquez-Torres University of Colorado Health Sciences Center, Aurora, Colorado Cristiano Viappiani Dipartimento di Fisica, Universita` degli Studi di Parma, Parma, Italy, and NEST CNR-INFM, Pisa, Italy Joa˜o B. Vicente Instituto de Tecnologia Quı´mica e Biolo´gica, Universidade Nova de Lisboa, Oeiras, Portugal Nicholas J. Watmough Center for Metalloprotein Spectroscopy and Biology, School of Biological Sciences, University of East Anglia, Norwich, United Kingdom Jessica Van Wonderen Centre for Metalloprotein Spectroscopy and Biology, School of Chemical Sciences and Pharmacy, University of East Anglia, Norwich, United Kingdom Syun-Ru Yeh Department of Physiology and Biophysics, Albert Einstein College of Medicine, Bronx, New York Li Zhang Department of Biology, University of Kentucky, Lexington, Kentucky
PREFACE
The genesis of ideas for these two volumes of Methods in Enzymology appears to be a talk (subtitled Bloody Bacteria) that I presented at the Agouron Institute meeting in Santa Fe, New Mexico, in April 2006. The topic of the meeting was Oxygen, but my message was not how microbial hemoglobins manage oxygen but rather how the primary function of many such hemoglobins is nitric oxide detoxification. Despite my straying from my brief, John Abelson and Mel Simon generously invited me to consider editing a volume of Methods in Enzymology to cover these emerging aspects of such a well-studied protein family. Further discussion of the proposal at the XIVth International Conference on Dioxygen Binding and Sensing Proteins at Stazione Zoologica Anton Dohrn in beautiful Napoli later that year—warmly hosted by Cinzia Verde and Guido di Prisco—generated much interest and support. The result was a two-volume heterodimer: I hope cooperativity can be found in Volumes 436 and 437. Just as the organizers of the Agouron Institute conference interpreted Oxygen with commendable flexibility, Methods in Enzymology has allowed some freedom in the definition of an enzyme. In 1994, when the topic Hemoglobins (Part C) was last covered explicitly in this series (Volume 232), some justification for labeling a hemoglobin as an enzyme might have been warranted. But as Maurizio Brunori pointed out in 1999 (Trends in Biochemical Sciences, 24, 158–161), the promotion of hemoglobin to the status of ‘‘honorary enzyme’’ had been conferred decades earlier by Monod, Wyman, and Changeux. In 2007, the idea that certain hemoglobins, even those not displaying allosteric heme–heme interactions, have enzymatic functions is well established, the most obvious examples being those hemoglobins that transform substrates into products, such as nitric oxide into nitrate. Other topics covered in these volumes are not new to the Methods in Enzymology series either. The most recent coverage of overtly related topics was Nitric Oxide (Part E) in Volume 396 (2005) and Oxygen Sensing in Volume 381 (2004). I hope, however, that the particular juxtaposition of topics in these two volumes will draw attention to the intimate links between globins, their gaseous ligands (nitric oxide, oxygen, and carbon monoxide), and the sensing and detoxification of these biologically critical small molecules. There is a strong microbial flavor in these volumes, reflecting some of the most exciting developments in recent years. Volume
xxvii
xxviii
Preface
436 deals with some chemical and analytical aspects of nitric oxide and methods for bacterial and archaeal hemoglobins, as well as diverse (especially ‘‘newer’’) hemoglobins in plants and animals. Volume 437 covers various non-hemoglobin nitric oxide-detoxifying proteins, sensors for gaseous ligands, advanced spectroscopic tools, and aspects of the functions of these proteins in microbial and plant physiology. In each volume, some chapters serve not as methodological recipes but short reviews to place the methods in a proper framework. These volumes would not have been possible without the tremendous enthusiasm of so many colleagues, contributors, and friends around the world. I thank them all, and also Tari Broderick and Cindy Minor (Elsevier, San Diego, California), for their help and encouragement in leading these volumes to a successful and timely outcome. ROBERT K. POOLE
METHODS IN ENZYMOLOGY
VOLUME I. Preparation and Assay of Enzymes Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME II. Preparation and Assay of Enzymes Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME III. Preparation and Assay of Substrates Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME IV. Special Techniques for the Enzymologist Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME V. Preparation and Assay of Enzymes Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME VI. Preparation and Assay of Enzymes (Continued) Preparation and Assay of Substrates Special Techniques Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME VII. Cumulative Subject Index Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME VIII. Complex Carbohydrates Edited by ELIZABETH F. NEUFELD AND VICTOR GINSBURG VOLUME IX. Carbohydrate Metabolism Edited by WILLIS A. WOOD VOLUME X. Oxidation and Phosphorylation Edited by RONALD W. ESTABROOK AND MAYNARD E. PULLMAN VOLUME XI. Enzyme Structure Edited by C. H. W. HIRS VOLUME XII. Nucleic Acids (Parts A and B) Edited by LAWRENCE GROSSMAN AND KIVIE MOLDAVE VOLUME XIII. Citric Acid Cycle Edited by J. M. LOWENSTEIN VOLUME XIV. Lipids Edited by J. M. LOWENSTEIN VOLUME XV. Steroids and Terpenoids Edited by RAYMOND B. CLAYTON xxix
xxx
Methods in Enzymology
VOLUME XVI. Fast Reactions Edited by KENNETH KUSTIN VOLUME XVII. Metabolism of Amino Acids and Amines (Parts A and B) Edited by HERBERT TABOR AND CELIA WHITE TABOR VOLUME XVIII. Vitamins and Coenzymes (Parts A, B, and C) Edited by DONALD B. MCCORMICK AND LEMUEL D. WRIGHT VOLUME XIX. Proteolytic Enzymes Edited by GERTRUDE E. PERLMANN AND LASZLO LORAND VOLUME XX. Nucleic Acids and Protein Synthesis (Part C) Edited by KIVIE MOLDAVE AND LAWRENCE GROSSMAN VOLUME XXI. Nucleic Acids (Part D) Edited by LAWRENCE GROSSMAN AND KIVIE MOLDAVE VOLUME XXII. Enzyme Purification and Related Techniques Edited by WILLIAM B. JAKOBY VOLUME XXIII. Photosynthesis (Part A) Edited by ANTHONY SAN PIETRO VOLUME XXIV. Photosynthesis and Nitrogen Fixation (Part B) Edited by ANTHONY SAN PIETRO VOLUME XXV. Enzyme Structure (Part B) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME XXVI. Enzyme Structure (Part C) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME XXVII. Enzyme Structure (Part D) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME XXVIII. Complex Carbohydrates (Part B) Edited by VICTOR GINSBURG VOLUME XXIX. Nucleic Acids and Protein Synthesis (Part E) Edited by LAWRENCE GROSSMAN AND KIVIE MOLDAVE VOLUME XXX. Nucleic Acids and Protein Synthesis (Part F) Edited by KIVIE MOLDAVE AND LAWRENCE GROSSMAN VOLUME XXXI. Biomembranes (Part A) Edited by SIDNEY FLEISCHER AND LESTER PACKER VOLUME XXXII. Biomembranes (Part B) Edited by SIDNEY FLEISCHER AND LESTER PACKER VOLUME XXXIII. Cumulative Subject Index Volumes I-XXX Edited by MARTHA G. DENNIS AND EDWARD A. DENNIS VOLUME XXXIV. Affinity Techniques (Enzyme Purification: Part B) Edited by WILLIAM B. JAKOBY AND MEIR WILCHEK
Methods in Enzymology
VOLUME XXXV. Lipids (Part B) Edited by JOHN M. LOWENSTEIN VOLUME XXXVI. Hormone Action (Part A: Steroid Hormones) Edited by BERT W. O’MALLEY AND JOEL G. HARDMAN VOLUME XXXVII. Hormone Action (Part B: Peptide Hormones) Edited by BERT W. O’MALLEY AND JOEL G. HARDMAN VOLUME XXXVIII. Hormone Action (Part C: Cyclic Nucleotides) Edited by JOEL G. HARDMAN AND BERT W. O’MALLEY VOLUME XXXIX. Hormone Action (Part D: Isolated Cells, Tissues, and Organ Systems) Edited by JOEL G. HARDMAN AND BERT W. O’MALLEY VOLUME XL. Hormone Action (Part E: Nuclear Structure and Function) Edited by BERT W. O’MALLEY AND JOEL G. HARDMAN VOLUME XLI. Carbohydrate Metabolism (Part B) Edited by W. A. WOOD VOLUME XLII. Carbohydrate Metabolism (Part C) Edited by W. A. WOOD VOLUME XLIII. Antibiotics Edited by JOHN H. HASH VOLUME XLIV. Immobilized Enzymes Edited by KLAUS MOSBACH VOLUME XLV. Proteolytic Enzymes (Part B) Edited by LASZLO LORAND VOLUME XLVI. Affinity Labeling Edited by WILLIAM B. JAKOBY AND MEIR WILCHEK VOLUME XLVII. Enzyme Structure (Part E) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME XLVIII. Enzyme Structure (Part F) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME XLIX. Enzyme Structure (Part G) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME L. Complex Carbohydrates (Part C) Edited by VICTOR GINSBURG VOLUME LI. Purine and Pyrimidine Nucleotide Metabolism Edited by PATRICIA A. HOFFEE AND MARY ELLEN JONES VOLUME LII. Biomembranes (Part C: Biological Oxidations) Edited by SIDNEY FLEISCHER AND LESTER PACKER
xxxi
xxxii
Methods in Enzymology
VOLUME LIII. Biomembranes (Part D: Biological Oxidations) Edited by SIDNEY FLEISCHER AND LESTER PACKER VOLUME LIV. Biomembranes (Part E: Biological Oxidations) Edited by SIDNEY FLEISCHER AND LESTER PACKER VOLUME LV. Biomembranes (Part F: Bioenergetics) Edited by SIDNEY FLEISCHER AND LESTER PACKER VOLUME LVI. Biomembranes (Part G: Bioenergetics) Edited by SIDNEY FLEISCHER AND LESTER PACKER VOLUME LVII. Bioluminescence and Chemiluminescence Edited by MARLENE A. DELUCA VOLUME LVIII. Cell Culture Edited by WILLIAM B. JAKOBY AND IRA PASTAN VOLUME LIX. Nucleic Acids and Protein Synthesis (Part G) Edited by KIVIE MOLDAVE AND LAWRENCE GROSSMAN VOLUME LX. Nucleic Acids and Protein Synthesis (Part H) Edited by KIVIE MOLDAVE AND LAWRENCE GROSSMAN VOLUME 61. Enzyme Structure (Part H) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME 62. Vitamins and Coenzymes (Part D) Edited by DONALD B. MCCORMICK AND LEMUEL D. WRIGHT VOLUME 63. Enzyme Kinetics and Mechanism (Part A: Initial Rate and Inhibitor Methods) Edited by DANIEL L. PURICH VOLUME 64. Enzyme Kinetics and Mechanism (Part B: Isotopic Probes and Complex Enzyme Systems) Edited by DANIEL L. PURICH VOLUME 65. Nucleic Acids (Part I) Edited by LAWRENCE GROSSMAN AND KIVIE MOLDAVE VOLUME 66. Vitamins and Coenzymes (Part E) Edited by DONALD B. MCCORMICK AND LEMUEL D. WRIGHT VOLUME 67. Vitamins and Coenzymes (Part F) Edited by DONALD B. MCCORMICK AND LEMUEL D. WRIGHT VOLUME 68. Recombinant DNA Edited by RAY WU VOLUME 69. Photosynthesis and Nitrogen Fixation (Part C) Edited by ANTHONY SAN PIETRO VOLUME 70. Immunochemical Techniques (Part A) Edited by HELEN VAN VUNAKIS AND JOHN J. LANGONE
Methods in Enzymology
xxxiii
VOLUME 71. Lipids (Part C) Edited by JOHN M. LOWENSTEIN VOLUME 72. Lipids (Part D) Edited by JOHN M. LOWENSTEIN VOLUME 73. Immunochemical Techniques (Part B) Edited by JOHN J. LANGONE AND HELEN VAN VUNAKIS VOLUME 74. Immunochemical Techniques (Part C) Edited by JOHN J. LANGONE AND HELEN VAN VUNAKIS VOLUME 75. Cumulative Subject Index Volumes XXXI, XXXII, XXXIV–LX Edited by EDWARD A. DENNIS AND MARTHA G. DENNIS VOLUME 76. Hemoglobins Edited by ERALDO ANTONINI, LUIGI ROSSI-BERNARDI, AND EMILIA CHIANCONE VOLUME 77. Detoxication and Drug Metabolism Edited by WILLIAM B. JAKOBY VOLUME 78. Interferons (Part A) Edited by SIDNEY PESTKA VOLUME 79. Interferons (Part B) Edited by SIDNEY PESTKA VOLUME 80. Proteolytic Enzymes (Part C) Edited by LASZLO LORAND VOLUME 81. Biomembranes (Part H: Visual Pigments and Purple Membranes, I) Edited by LESTER PACKER VOLUME 82. Structural and Contractile Proteins (Part A: Extracellular Matrix) Edited by LEON W. CUNNINGHAM AND DIXIE W. FREDERIKSEN VOLUME 83. Complex Carbohydrates (Part D) Edited by VICTOR GINSBURG VOLUME 84. Immunochemical Techniques (Part D: Selected Immunoassays) Edited by JOHN J. LANGONE AND HELEN VAN VUNAKIS VOLUME 85. Structural and Contractile Proteins (Part B: The Contractile Apparatus and the Cytoskeleton) Edited by DIXIE W. FREDERIKSEN AND LEON W. CUNNINGHAM VOLUME 86. Prostaglandins and Arachidonate Metabolites Edited by WILLIAM E. M. LANDS AND WILLIAM L. SMITH VOLUME 87. Enzyme Kinetics and Mechanism (Part C: Intermediates, Stereo-chemistry, and Rate Studies) Edited by DANIEL L. PURICH VOLUME 88. Biomembranes (Part I: Visual Pigments and Purple Membranes, II) Edited by LESTER PACKER
xxxiv
Methods in Enzymology
VOLUME 89. Carbohydrate Metabolism (Part D) Edited by WILLIS A. WOOD VOLUME 90. Carbohydrate Metabolism (Part E) Edited by WILLIS A. WOOD VOLUME 91. Enzyme Structure (Part I) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME 92. Immunochemical Techniques (Part E: Monoclonal Antibodies and General Immunoassay Methods) Edited by JOHN J. LANGONE AND HELEN VAN VUNAKIS VOLUME 93. Immunochemical Techniques (Part F: Conventional Antibodies, Fc Receptors, and Cytotoxicity) Edited by JOHN J. LANGONE AND HELEN VAN VUNAKIS VOLUME 94. Polyamines Edited by HERBERT TABOR AND CELIA WHITE TABOR VOLUME 95. Cumulative Subject Index Volumes 61–74, 76–80 Edited by EDWARD A. DENNIS AND MARTHA G. DENNIS VOLUME 96. Biomembranes [Part J: Membrane Biogenesis: Assembly and Targeting (General Methods; Eukaryotes)] Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 97. Biomembranes [Part K: Membrane Biogenesis: Assembly and Targeting (Prokaryotes, Mitochondria, and Chloroplasts)] Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 98. Biomembranes (Part L: Membrane Biogenesis: Processing and Recycling) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 99. Hormone Action (Part F: Protein Kinases) Edited by JACKIE D. CORBIN AND JOEL G. HARDMAN VOLUME 100. Recombinant DNA (Part B) Edited by RAY WU, LAWRENCE GROSSMAN, AND KIVIE MOLDAVE VOLUME 101. Recombinant DNA (Part C) Edited by RAY WU, LAWRENCE GROSSMAN, AND KIVIE MOLDAVE VOLUME 102. Hormone Action (Part G: Calmodulin and Calcium-Binding Proteins) Edited by ANTHONY R. MEANS AND BERT W. O’MALLEY VOLUME 103. Hormone Action (Part H: Neuroendocrine Peptides) Edited by P. MICHAEL CONN VOLUME 104. Enzyme Purification and Related Techniques (Part C) Edited by WILLIAM B. JAKOBY
Methods in Enzymology
xxxv
VOLUME 105. Oxygen Radicals in Biological Systems Edited by LESTER PACKER VOLUME 106. Posttranslational Modifications (Part A) Edited by FINN WOLD AND KIVIE MOLDAVE VOLUME 107. Posttranslational Modifications (Part B) Edited by FINN WOLD AND KIVIE MOLDAVE VOLUME 108. Immunochemical Techniques (Part G: Separation and Characterization of Lymphoid Cells) Edited by GIOVANNI DI SABATO, JOHN J. LANGONE, AND HELEN VAN VUNAKIS VOLUME 109. Hormone Action (Part I: Peptide Hormones) Edited by LUTZ BIRNBAUMER AND BERT W. O’MALLEY VOLUME 110. Steroids and Isoprenoids (Part A) Edited by JOHN H. LAW AND HANS C. RILLING VOLUME 111. Steroids and Isoprenoids (Part B) Edited by JOHN H. LAW AND HANS C. RILLING VOLUME 112. Drug and Enzyme Targeting (Part A) Edited by KENNETH J. WIDDER AND RALPH GREEN VOLUME 113. Glutamate, Glutamine, Glutathione, and Related Compounds Edited by ALTON MEISTER VOLUME 114. Diffraction Methods for Biological Macromolecules (Part A) Edited by HAROLD W. WYCKOFF, C. H. W. HIRS, AND SERGE N. TIMASHEFF VOLUME 115. Diffraction Methods for Biological Macromolecules (Part B) Edited by HAROLD W. WYCKOFF, C. H. W. HIRS, AND SERGE N. TIMASHEFF VOLUME 116. Immunochemical Techniques (Part H: Effectors and Mediators of Lymphoid Cell Functions) Edited by GIOVANNI DI SABATO, JOHN J. LANGONE, AND HELEN VAN VUNAKIS VOLUME 117. Enzyme Structure (Part J) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME 118. Plant Molecular Biology Edited by ARTHUR WEISSBACH AND HERBERT WEISSBACH VOLUME 119. Interferons (Part C) Edited by SIDNEY PESTKA VOLUME 120. Cumulative Subject Index Volumes 81–94, 96–101 VOLUME 121. Immunochemical Techniques (Part I: Hybridoma Technology and Monoclonal Antibodies) Edited by JOHN J. LANGONE AND HELEN VAN VUNAKIS VOLUME 122. Vitamins and Coenzymes (Part G) Edited by FRANK CHYTIL AND DONALD B. MCCORMICK
xxxvi
Methods in Enzymology
VOLUME 123. Vitamins and Coenzymes (Part H) Edited by FRANK CHYTIL AND DONALD B. MCCORMICK VOLUME 124. Hormone Action (Part J: Neuroendocrine Peptides) Edited by P. MICHAEL CONN VOLUME 125. Biomembranes (Part M: Transport in Bacteria, Mitochondria, and Chloroplasts: General Approaches and Transport Systems) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 126. Biomembranes (Part N: Transport in Bacteria, Mitochondria, and Chloroplasts: Protonmotive Force) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 127. Biomembranes (Part O: Protons and Water: Structure and Translocation) Edited by LESTER PACKER VOLUME 128. Plasma Lipoproteins (Part A: Preparation, Structure, and Molecular Biology) Edited by JERE P. SEGREST AND JOHN J. ALBERS VOLUME 129. Plasma Lipoproteins (Part B: Characterization, Cell Biology, and Metabolism) Edited by JOHN J. ALBERS AND JERE P. SEGREST VOLUME 130. Enzyme Structure (Part K) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME 131. Enzyme Structure (Part L) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME 132. Immunochemical Techniques (Part J: Phagocytosis and Cell-Mediated Cytotoxicity) Edited by GIOVANNI DI SABATO AND JOHANNES EVERSE VOLUME 133. Bioluminescence and Chemiluminescence (Part B) Edited by MARLENE DELUCA AND WILLIAM D. MCELROY VOLUME 134. Structural and Contractile Proteins (Part C: The Contractile Apparatus and the Cytoskeleton) Edited by RICHARD B. VALLEE VOLUME 135. Immobilized Enzymes and Cells (Part B) Edited by KLAUS MOSBACH VOLUME 136. Immobilized Enzymes and Cells (Part C) Edited by KLAUS MOSBACH VOLUME 137. Immobilized Enzymes and Cells (Part D) Edited by KLAUS MOSBACH VOLUME 138. Complex Carbohydrates (Part E) Edited by VICTOR GINSBURG
Methods in Enzymology
xxxvii
VOLUME 139. Cellular Regulators (Part A: Calcium- and Calmodulin-Binding Proteins) Edited by ANTHONY R. MEANS AND P. MICHAEL CONN VOLUME 140. Cumulative Subject Index Volumes 102–119, 121–134 VOLUME 141. Cellular Regulators (Part B: Calcium and Lipids) Edited by P. MICHAEL CONN AND ANTHONY R. MEANS VOLUME 142. Metabolism of Aromatic Amino Acids and Amines Edited by SEYMOUR KAUFMAN VOLUME 143. Sulfur and Sulfur Amino Acids Edited by WILLIAM B. JAKOBY AND OWEN GRIFFITH VOLUME 144. Structural and Contractile Proteins (Part D: Extracellular Matrix) Edited by LEON W. CUNNINGHAM VOLUME 145. Structural and Contractile Proteins (Part E: Extracellular Matrix) Edited by LEON W. CUNNINGHAM VOLUME 146. Peptide Growth Factors (Part A) Edited by DAVID BARNES AND DAVID A. SIRBASKU VOLUME 147. Peptide Growth Factors (Part B) Edited by DAVID BARNES AND DAVID A. SIRBASKU VOLUME 148. Plant Cell Membranes Edited by LESTER PACKER AND ROLAND DOUCE VOLUME 149. Drug and Enzyme Targeting (Part B) Edited by RALPH GREEN AND KENNETH J. WIDDER VOLUME 150. Immunochemical Techniques (Part K: In Vitro Models of B and T Cell Functions and Lymphoid Cell Receptors) Edited by GIOVANNI DI SABATO VOLUME 151. Molecular Genetics of Mammalian Cells Edited by MICHAEL M. GOTTESMAN VOLUME 152. Guide to Molecular Cloning Techniques Edited by SHELBY L. BERGER AND ALAN R. KIMMEL VOLUME 153. Recombinant DNA (Part D) Edited by RAY WU AND LAWRENCE GROSSMAN VOLUME 154. Recombinant DNA (Part E) Edited by RAY WU AND LAWRENCE GROSSMAN VOLUME 155. Recombinant DNA (Part F) Edited by RAY WU VOLUME 156. Biomembranes (Part P: ATP-Driven Pumps and Related Transport: The Na, K-Pump) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER
xxxviii
Methods in Enzymology
VOLUME 157. Biomembranes (Part Q: ATP-Driven Pumps and Related Transport: Calcium, Proton, and Potassium Pumps) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 158. Metalloproteins (Part A) Edited by JAMES F. RIORDAN AND BERT L. VALLEE VOLUME 159. Initiation and Termination of Cyclic Nucleotide Action Edited by JACKIE D. CORBIN AND ROGER A. JOHNSON VOLUME 160. Biomass (Part A: Cellulose and Hemicellulose) Edited by WILLIS A. WOOD AND SCOTT T. KELLOGG VOLUME 161. Biomass (Part B: Lignin, Pectin, and Chitin) Edited by WILLIS A. WOOD AND SCOTT T. KELLOGG VOLUME 162. Immunochemical Techniques (Part L: Chemotaxis and Inflammation) Edited by GIOVANNI DI SABATO VOLUME 163. Immunochemical Techniques (Part M: Chemotaxis and Inflammation) Edited by GIOVANNI DI SABATO VOLUME 164. Ribosomes Edited by HARRY F. NOLLER, JR., AND KIVIE MOLDAVE VOLUME 165. Microbial Toxins: Tools for Enzymology Edited by SIDNEY HARSHMAN VOLUME 166. Branched-Chain Amino Acids Edited by ROBERT HARRIS AND JOHN R. SOKATCH VOLUME 167. Cyanobacteria Edited by LESTER PACKER AND ALEXANDER N. GLAZER VOLUME 168. Hormone Action (Part K: Neuroendocrine Peptides) Edited by P. MICHAEL CONN VOLUME 169. Platelets: Receptors, Adhesion, Secretion (Part A) Edited by JACEK HAWIGER VOLUME 170. Nucleosomes Edited by PAUL M. WASSARMAN AND ROGER D. KORNBERG VOLUME 171. Biomembranes (Part R: Transport Theory: Cells and Model Membranes) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 172. Biomembranes (Part S: Transport: Membrane Isolation and Characterization) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER
Methods in Enzymology
xxxix
VOLUME 173. Biomembranes [Part T: Cellular and Subcellular Transport: Eukaryotic (Nonepithelial) Cells] Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 174. Biomembranes [Part U: Cellular and Subcellular Transport: Eukaryotic (Nonepithelial) Cells] Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 175. Cumulative Subject Index Volumes 135–139, 141–167 VOLUME 176. Nuclear Magnetic Resonance (Part A: Spectral Techniques and Dynamics) Edited by NORMAN J. OPPENHEIMER AND THOMAS L. JAMES VOLUME 177. Nuclear Magnetic Resonance (Part B: Structure and Mechanism) Edited by NORMAN J. OPPENHEIMER AND THOMAS L. JAMES VOLUME 178. Antibodies, Antigens, and Molecular Mimicry Edited by JOHN J. LANGONE VOLUME 179. Complex Carbohydrates (Part F) Edited by VICTOR GINSBURG VOLUME 180. RNA Processing (Part A: General Methods) Edited by JAMES E. DAHLBERG AND JOHN N. ABELSON VOLUME 181. RNA Processing (Part B: Specific Methods) Edited by JAMES E. DAHLBERG AND JOHN N. ABELSON VOLUME 182. Guide to Protein Purification Edited by MURRAY P. DEUTSCHER VOLUME 183. Molecular Evolution: Computer Analysis of Protein and Nucleic Acid Sequences Edited by RUSSELL F. DOOLITTLE VOLUME 184. Avidin-Biotin Technology Edited by MEIR WILCHEK AND EDWARD A. BAYER VOLUME 185. Gene Expression Technology Edited by DAVID V. GOEDDEL VOLUME 186. Oxygen Radicals in Biological Systems (Part B: Oxygen Radicals and Antioxidants) Edited by LESTER PACKER AND ALEXANDER N. GLAZER VOLUME 187. Arachidonate Related Lipid Mediators Edited by ROBERT C. MURPHY AND FRANK A. FITZPATRICK VOLUME 188. Hydrocarbons and Methylotrophy Edited by MARY E. LIDSTROM VOLUME 189. Retinoids (Part A: Molecular and Metabolic Aspects) Edited by LESTER PACKER
xl
Methods in Enzymology
VOLUME 190. Retinoids (Part B: Cell Differentiation and Clinical Applications) Edited by LESTER PACKER VOLUME 191. Biomembranes (Part V: Cellular and Subcellular Transport: Epithelial Cells) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 192. Biomembranes (Part W: Cellular and Subcellular Transport: Epithelial Cells) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 193. Mass Spectrometry Edited by JAMES A. MCCLOSKEY VOLUME 194. Guide to Yeast Genetics and Molecular Biology Edited by CHRISTINE GUTHRIE AND GERALD R. FINK VOLUME 195. Adenylyl Cyclase, G Proteins, and Guanylyl Cyclase Edited by ROGER A. JOHNSON AND JACKIE D. CORBIN VOLUME 196. Molecular Motors and the Cytoskeleton Edited by RICHARD B. VALLEE VOLUME 197. Phospholipases Edited by EDWARD A. DENNIS VOLUME 198. Peptide Growth Factors (Part C) Edited by DAVID BARNES, J. P. MATHER, AND GORDON H. SATO VOLUME 199. Cumulative Subject Index Volumes 168–174, 176–194 VOLUME 200. Protein Phosphorylation (Part A: Protein Kinases: Assays, Purification, Antibodies, Functional Analysis, Cloning, and Expression) Edited by TONY HUNTER AND BARTHOLOMEW M. SEFTON VOLUME 201. Protein Phosphorylation (Part B: Analysis of Protein Phosphorylation, Protein Kinase Inhibitors, and Protein Phosphatases) Edited by TONY HUNTER AND BARTHOLOMEW M. SEFTON VOLUME 202. Molecular Design and Modeling: Concepts and Applications (Part A: Proteins, Peptides, and Enzymes) Edited by JOHN J. LANGONE VOLUME 203. Molecular Design and Modeling: Concepts and Applications (Part B: Antibodies and Antigens, Nucleic Acids, Polysaccharides, and Drugs) Edited by JOHN J. LANGONE VOLUME 204. Bacterial Genetic Systems Edited by JEFFREY H. MILLER VOLUME 205. Metallobiochemistry (Part B: Metallothionein and Related Molecules) Edited by JAMES F. RIORDAN AND BERT L. VALLEE
Methods in Enzymology
xli
VOLUME 206. Cytochrome P450 Edited by MICHAEL R. WATERMAN AND ERIC F. JOHNSON VOLUME 207. Ion Channels Edited by BERNARDO RUDY AND LINDA E. IVERSON VOLUME 208. Protein–DNA Interactions Edited by ROBERT T. SAUER VOLUME 209. Phospholipid Biosynthesis Edited by EDWARD A. DENNIS AND DENNIS E. VANCE VOLUME 210. Numerical Computer Methods Edited by LUDWIG BRAND AND MICHAEL L. JOHNSON VOLUME 211. DNA Structures (Part A: Synthesis and Physical Analysis of DNA) Edited by DAVID M. J. LILLEY AND JAMES E. DAHLBERG VOLUME 212. DNA Structures (Part B: Chemical and Electrophoretic Analysis of DNA) Edited by DAVID M. J. LILLEY AND JAMES E. DAHLBERG VOLUME 213. Carotenoids (Part A: Chemistry, Separation, Quantitation, and Antioxidation) Edited by LESTER PACKER VOLUME 214. Carotenoids (Part B: Metabolism, Genetics, and Biosynthesis) Edited by LESTER PACKER VOLUME 215. Platelets: Receptors, Adhesion, Secretion (Part B) Edited by JACEK J. HAWIGER VOLUME 216. Recombinant DNA (Part G) Edited by RAY WU VOLUME 217. Recombinant DNA (Part H) Edited by RAY WU VOLUME 218. Recombinant DNA (Part I) Edited by RAY WU VOLUME 219. Reconstitution of Intracellular Transport Edited by JAMES E. ROTHMAN VOLUME 220. Membrane Fusion Techniques (Part A) Edited by NEJAT DU¨ZGU¨NES VOLUME 221. Membrane Fusion Techniques (Part B) Edited by NEJAT DU¨ZGU¨NES VOLUME 222. Proteolytic Enzymes in Coagulation, Fibrinolysis, and Complement Activation (Part A: Mammalian Blood Coagulation Factors and Inhibitors) Edited by LASZLO LORAND AND KENNETH G. MANN
xlii
Methods in Enzymology
VOLUME 223. Proteolytic Enzymes in Coagulation, Fibrinolysis, and Complement Activation (Part B: Complement Activation, Fibrinolysis, and Nonmammalian Blood Coagulation Factors) Edited by LASZLO LORAND AND KENNETH G. MANN VOLUME 224. Molecular Evolution: Producing the Biochemical Data Edited by ELIZABETH ANNE ZIMMER, THOMAS J. WHITE, REBECCA L. CANN, AND ALLAN C. WILSON VOLUME 225. Guide to Techniques in Mouse Development Edited by PAUL M. WASSARMAN AND MELVIN L. DEPAMPHILIS VOLUME 226. Metallobiochemistry (Part C: Spectroscopic and Physical Methods for Probing Metal Ion Environments in Metalloenzymes and Metalloproteins) Edited by JAMES F. RIORDAN AND BERT L. VALLEE VOLUME 227. Metallobiochemistry (Part D: Physical and Spectroscopic Methods for Probing Metal Ion Environments in Metalloproteins) Edited by JAMES F. RIORDAN AND BERT L. VALLEE VOLUME 228. Aqueous Two-Phase Systems Edited by HARRY WALTER AND GO¨TE JOHANSSON VOLUME 229. Cumulative Subject Index Volumes 195–198, 200–227 VOLUME 230. Guide to Techniques in Glycobiology Edited by WILLIAM J. LENNARZ AND GERALD W. HART VOLUME 231. Hemoglobins (Part B: Biochemical and Analytical Methods) Edited by JOHANNES EVERSE, KIM D. VANDEGRIFF, AND ROBERT M. WINSLOW VOLUME 232. Hemoglobins (Part C: Biophysical Methods) Edited by JOHANNES EVERSE, KIM D. VANDEGRIFF, AND ROBERT M. WINSLOW VOLUME 233. Oxygen Radicals in Biological Systems (Part C) Edited by LESTER PACKER VOLUME 234. Oxygen Radicals in Biological Systems (Part D) Edited by LESTER PACKER VOLUME 235. Bacterial Pathogenesis (Part A: Identification and Regulation of Virulence Factors) Edited by VIRGINIA L. CLARK AND PATRIK M. BAVOIL VOLUME 236. Bacterial Pathogenesis (Part B: Integration of Pathogenic Bacteria with Host Cells) Edited by VIRGINIA L. CLARK AND PATRIK M. BAVOIL VOLUME 237. Heterotrimeric G Proteins Edited by RAVI IYENGAR VOLUME 238. Heterotrimeric G-Protein Effectors Edited by RAVI IYENGAR
Methods in Enzymology
xliii
VOLUME 239. Nuclear Magnetic Resonance (Part C) Edited by THOMAS L. JAMES AND NORMAN J. OPPENHEIMER VOLUME 240. Numerical Computer Methods (Part B) Edited by MICHAEL L. JOHNSON AND LUDWIG BRAND VOLUME 241. Retroviral Proteases Edited by LAWRENCE C. KUO AND JULES A. SHAFER VOLUME 242. Neoglycoconjugates (Part A) Edited by Y. C. LEE AND REIKO T. LEE VOLUME 243. Inorganic Microbial Sulfur Metabolism Edited by HARRY D. PECK, JR., AND JEAN LEGALL VOLUME 244. Proteolytic Enzymes: Serine and Cysteine Peptidases Edited by ALAN J. BARRETT VOLUME 245. Extracellular Matrix Components Edited by E. RUOSLAHTI AND E. ENGVALL VOLUME 246. Biochemical Spectroscopy Edited by KENNETH SAUER VOLUME 247. Neoglycoconjugates (Part B: Biomedical Applications) Edited by Y. C. LEE AND REIKO T. LEE VOLUME 248. Proteolytic Enzymes: Aspartic and Metallo Peptidases Edited by ALAN J. BARRETT VOLUME 249. Enzyme Kinetics and Mechanism (Part D: Developments in Enzyme Dynamics) Edited by DANIEL L. PURICH VOLUME 250. Lipid Modifications of Proteins Edited by PATRICK J. CASEY AND JANICE E. BUSS VOLUME 251. Biothiols (Part A: Monothiols and Dithiols, Protein Thiols, and Thiyl Radicals) Edited by LESTER PACKER VOLUME 252. Biothiols (Part B: Glutathione and Thioredoxin; Thiols in Signal Transduction and Gene Regulation) Edited by LESTER PACKER VOLUME 253. Adhesion of Microbial Pathogens Edited by RON J. DOYLE AND ITZHAK OFEK VOLUME 254. Oncogene Techniques Edited by PETER K. VOGT AND INDER M. VERMA VOLUME 255. Small GTPases and Their Regulators (Part A: Ras Family) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 256. Small GTPases and Their Regulators (Part B: Rho Family) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL
xliv
Methods in Enzymology
VOLUME 257. Small GTPases and Their Regulators (Part C: Proteins Involved in Transport) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 258. Redox-Active Amino Acids in Biology Edited by JUDITH P. KLINMAN VOLUME 259. Energetics of Biological Macromolecules Edited by MICHAEL L. JOHNSON AND GARY K. ACKERS VOLUME 260. Mitochondrial Biogenesis and Genetics (Part A) Edited by GIUSEPPE M. ATTARDI AND ANNE CHOMYN VOLUME 261. Nuclear Magnetic Resonance and Nucleic Acids Edited by THOMAS L. JAMES VOLUME 262. DNA Replication Edited by JUDITH L. CAMPBELL VOLUME 263. Plasma Lipoproteins (Part C: Quantitation) Edited by WILLIAM A. BRADLEY, SANDRA H. GIANTURCO, AND JERE P. SEGREST VOLUME 264. Mitochondrial Biogenesis and Genetics (Part B) Edited by GIUSEPPE M. ATTARDI AND ANNE CHOMYN VOLUME 265. Cumulative Subject Index Volumes 228, 230–262 VOLUME 266. Computer Methods for Macromolecular Sequence Analysis Edited by RUSSELL F. DOOLITTLE VOLUME 267. Combinatorial Chemistry Edited by JOHN N. ABELSON VOLUME 268. Nitric Oxide (Part A: Sources and Detection of NO; NO Synthase) Edited by LESTER PACKER VOLUME 269. Nitric Oxide (Part B: Physiological and Pathological Processes) Edited by LESTER PACKER VOLUME 270. High Resolution Separation and Analysis of Biological Macromolecules (Part A: Fundamentals) Edited by BARRY L. KARGER AND WILLIAM S. HANCOCK VOLUME 271. High Resolution Separation and Analysis of Biological Macromolecules (Part B: Applications) Edited by BARRY L. KARGER AND WILLIAM S. HANCOCK VOLUME 272. Cytochrome P450 (Part B) Edited by ERIC F. JOHNSON AND MICHAEL R. WATERMAN VOLUME 273. RNA Polymerase and Associated Factors (Part A) Edited by SANKAR ADHYA VOLUME 274. RNA Polymerase and Associated Factors (Part B) Edited by SANKAR ADHYA
Methods in Enzymology
xlv
VOLUME 275. Viral Polymerases and Related Proteins Edited by LAWRENCE C. KUO, DAVID B. OLSEN, AND STEVEN S. CARROLL VOLUME 276. Macromolecular Crystallography (Part A) Edited by CHARLES W. CARTER, JR., AND ROBERT M. SWEET VOLUME 277. Macromolecular Crystallography (Part B) Edited by CHARLES W. CARTER, JR., AND ROBERT M. SWEET VOLUME 278. Fluorescence Spectroscopy Edited by LUDWIG BRAND AND MICHAEL L. JOHNSON VOLUME 279. Vitamins and Coenzymes (Part I) Edited by DONALD B. MCCORMICK, JOHN W. SUTTIE, AND CONRAD WAGNER VOLUME 280. Vitamins and Coenzymes (Part J) Edited by DONALD B. MCCORMICK, JOHN W. SUTTIE, AND CONRAD WAGNER VOLUME 281. Vitamins and Coenzymes (Part K) Edited by DONALD B. MCCORMICK, JOHN W. SUTTIE, AND CONRAD WAGNER VOLUME 282. Vitamins and Coenzymes (Part L) Edited by DONALD B. MCCORMICK, JOHN W. SUTTIE, AND CONRAD WAGNER VOLUME 283. Cell Cycle Control Edited by WILLIAM G. DUNPHY VOLUME 284. Lipases (Part A: Biotechnology) Edited by BYRON RUBIN AND EDWARD A. DENNIS VOLUME 285. Cumulative Subject Index Volumes 263, 264, 266–284, 286–289 VOLUME 286. Lipases (Part B: Enzyme Characterization and Utilization) Edited by BYRON RUBIN AND EDWARD A. DENNIS VOLUME 287. Chemokines Edited by RICHARD HORUK VOLUME 288. Chemokine Receptors Edited by RICHARD HORUK VOLUME 289. Solid Phase Peptide Synthesis Edited by GREGG B. FIELDS VOLUME 290. Molecular Chaperones Edited by GEORGE H. LORIMER AND THOMAS BALDWIN VOLUME 291. Caged Compounds Edited by GERARD MARRIOTT VOLUME 292. ABC Transporters: Biochemical, Cellular, and Molecular Aspects Edited by SURESH V. AMBUDKAR AND MICHAEL M. GOTTESMAN VOLUME 293. Ion Channels (Part B) Edited by P. MICHAEL CONN
xlvi
Methods in Enzymology
VOLUME 294. Ion Channels (Part C) Edited by P. MICHAEL CONN VOLUME 295. Energetics of Biological Macromolecules (Part B) Edited by GARY K. ACKERS AND MICHAEL L. JOHNSON VOLUME 296. Neurotransmitter Transporters Edited by SUSAN G. AMARA VOLUME 297. Photosynthesis: Molecular Biology of Energy Capture Edited by LEE MCINTOSH VOLUME 298. Molecular Motors and the Cytoskeleton (Part B) Edited by RICHARD B. VALLEE VOLUME 299. Oxidants and Antioxidants (Part A) Edited by LESTER PACKER VOLUME 300. Oxidants and Antioxidants (Part B) Edited by LESTER PACKER VOLUME 301. Nitric Oxide: Biological and Antioxidant Activities (Part C) Edited by LESTER PACKER VOLUME 302. Green Fluorescent Protein Edited by P. MICHAEL CONN VOLUME 303. cDNA Preparation and Display Edited by SHERMAN M. WEISSMAN VOLUME 304. Chromatin Edited by PAUL M. WASSARMAN AND ALAN P. WOLFFE VOLUME 305. Bioluminescence and Chemiluminescence (Part C) Edited by THOMAS O. BALDWIN AND MIRIAM M. ZIEGLER VOLUME 306. Expression of Recombinant Genes in Eukaryotic Systems Edited by JOSEPH C. GLORIOSO AND MARTIN C. SCHMIDT VOLUME 307. Confocal Microscopy Edited by P. MICHAEL CONN VOLUME 308. Enzyme Kinetics and Mechanism (Part E: Energetics of Enzyme Catalysis) Edited by DANIEL L. PURICH AND VERN L. SCHRAMM VOLUME 309. Amyloid, Prions, and Other Protein Aggregates Edited by RONALD WETZEL VOLUME 310. Biofilms Edited by RON J. DOYLE VOLUME 311. Sphingolipid Metabolism and Cell Signaling (Part A) Edited by ALFRED H. MERRILL, JR., AND YUSUF A. HANNUN
Methods in Enzymology
xlvii
VOLUME 312. Sphingolipid Metabolism and Cell Signaling (Part B) Edited by ALFRED H. MERRILL, JR., AND YUSUF A. HANNUN VOLUME 313. Antisense Technology (Part A: General Methods, Methods of Delivery, and RNA Studies) Edited by M. IAN PHILLIPS VOLUME 314. Antisense Technology (Part B: Applications) Edited by M. IAN PHILLIPS VOLUME 315. Vertebrate Phototransduction and the Visual Cycle (Part A) Edited by KRZYSZTOF PALCZEWSKI VOLUME 316. Vertebrate Phototransduction and the Visual Cycle (Part B) Edited by KRZYSZTOF PALCZEWSKI VOLUME 317. RNA–Ligand Interactions (Part A: Structural Biology Methods) Edited by DANIEL W. CELANDER AND JOHN N. ABELSON VOLUME 318. RNA–Ligand Interactions (Part B: Molecular Biology Methods) Edited by DANIEL W. CELANDER AND JOHN N. ABELSON VOLUME 319. Singlet Oxygen, UV-A, and Ozone Edited by LESTER PACKER AND HELMUT SIES VOLUME 320. Cumulative Subject Index Volumes 290–319 VOLUME 321. Numerical Computer Methods (Part C) Edited by MICHAEL L. JOHNSON AND LUDWIG BRAND VOLUME 322. Apoptosis Edited by JOHN C. REED VOLUME 323. Energetics of Biological Macromolecules (Part C) Edited by MICHAEL L. JOHNSON AND GARY K. ACKERS VOLUME 324. Branched-Chain Amino Acids (Part B) Edited by ROBERT A. HARRIS AND JOHN R. SOKATCH VOLUME 325. Regulators and Effectors of Small GTPases (Part D: Rho Family) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 326. Applications of Chimeric Genes and Hybrid Proteins (Part A: Gene Expression and Protein Purification) Edited by JEREMY THORNER, SCOTT D. EMR, AND JOHN N. ABELSON VOLUME 327. Applications of Chimeric Genes and Hybrid Proteins (Part B: Cell Biology and Physiology) Edited by JEREMY THORNER, SCOTT D. EMR, AND JOHN N. ABELSON VOLUME 328. Applications of Chimeric Genes and Hybrid Proteins (Part C: Protein–Protein Interactions and Genomics) Edited by JEREMY THORNER, SCOTT D. EMR, AND JOHN N. ABELSON
xlviii
Methods in Enzymology
VOLUME 329. Regulators and Effectors of Small GTPases (Part E: GTPases Involved in Vesicular Traffic) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 330. Hyperthermophilic Enzymes (Part A) Edited by MICHAEL W. W. ADAMS AND ROBERT M. KELLY VOLUME 331. Hyperthermophilic Enzymes (Part B) Edited by MICHAEL W. W. ADAMS AND ROBERT M. KELLY VOLUME 332. Regulators and Effectors of Small GTPases (Part F: Ras Family I) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 333. Regulators and Effectors of Small GTPases (Part G: Ras Family II) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 334. Hyperthermophilic Enzymes (Part C) Edited by MICHAEL W. W. ADAMS AND ROBERT M. KELLY VOLUME 335. Flavonoids and Other Polyphenols Edited by LESTER PACKER VOLUME 336. Microbial Growth in Biofilms (Part A: Developmental and Molecular Biological Aspects) Edited by RON J. DOYLE VOLUME 337. Microbial Growth in Biofilms (Part B: Special Environments and Physicochemical Aspects) Edited by RON J. DOYLE VOLUME 338. Nuclear Magnetic Resonance of Biological Macromolecules (Part A) Edited by THOMAS L. JAMES, VOLKER DO¨TSCH, AND ULI SCHMITZ VOLUME 339. Nuclear Magnetic Resonance of Biological Macromolecules (Part B) Edited by THOMAS L. JAMES, VOLKER DO¨TSCH, AND ULI SCHMITZ VOLUME 340. Drug–Nucleic Acid Interactions Edited by JONATHAN B. CHAIRES AND MICHAEL J. WARING VOLUME 341. Ribonucleases (Part A) Edited by ALLEN W. NICHOLSON VOLUME 342. Ribonucleases (Part B) Edited by ALLEN W. NICHOLSON VOLUME 343. G Protein Pathways (Part A: Receptors) Edited by RAVI IYENGAR AND JOHN D. HILDEBRANDT VOLUME 344. G Protein Pathways (Part B: G Proteins and Their Regulators) Edited by RAVI IYENGAR AND JOHN D. HILDEBRANDT VOLUME 345. G Protein Pathways (Part C: Effector Mechanisms) Edited by RAVI IYENGAR AND JOHN D. HILDEBRANDT
Methods in Enzymology
xlix
VOLUME 346. Gene Therapy Methods Edited by M. IAN PHILLIPS VOLUME 347. Protein Sensors and Reactive Oxygen Species (Part A: Selenoproteins and Thioredoxin) Edited by HELMUT SIES AND LESTER PACKER VOLUME 348. Protein Sensors and Reactive Oxygen Species (Part B: Thiol Enzymes and Proteins) Edited by HELMUT SIES AND LESTER PACKER VOLUME 349. Superoxide Dismutase Edited by LESTER PACKER VOLUME 350. Guide to Yeast Genetics and Molecular and Cell Biology (Part B) Edited by CHRISTINE GUTHRIE AND GERALD R. FINK VOLUME 351. Guide to Yeast Genetics and Molecular and Cell Biology (Part C) Edited by CHRISTINE GUTHRIE AND GERALD R. FINK VOLUME 352. Redox Cell Biology and Genetics (Part A) Edited by CHANDAN K. SEN AND LESTER PACKER VOLUME 353. Redox Cell Biology and Genetics (Part B) Edited by CHANDAN K. SEN AND LESTER PACKER VOLUME 354. Enzyme Kinetics and Mechanisms (Part F: Detection and Characterization of Enzyme Reaction Intermediates) Edited by DANIEL L. PURICH VOLUME 355. Cumulative Subject Index Volumes 321–354 VOLUME 356. Laser Capture Microscopy and Microdissection Edited by P. MICHAEL CONN VOLUME 357. Cytochrome P450, Part C Edited by ERIC F. JOHNSON AND MICHAEL R. WATERMAN VOLUME 358. Bacterial Pathogenesis (Part C: Identification, Regulation, and Function of Virulence Factors) Edited by VIRGINIA L. CLARK AND PATRIK M. BAVOIL VOLUME 359. Nitric Oxide (Part D) Edited by ENRIQUE CADENAS AND LESTER PACKER VOLUME 360. Biophotonics (Part A) Edited by GERARD MARRIOTT AND IAN PARKER VOLUME 361. Biophotonics (Part B) Edited by GERARD MARRIOTT AND IAN PARKER VOLUME 362. Recognition of Carbohydrates in Biological Systems (Part A) Edited by YUAN C. LEE AND REIKO T. LEE
l
Methods in Enzymology
VOLUME 363. Recognition of Carbohydrates in Biological Systems (Part B) Edited by YUAN C. LEE AND REIKO T. LEE VOLUME 364. Nuclear Receptors Edited by DAVID W. RUSSELL AND DAVID J. MANGELSDORF VOLUME 365. Differentiation of Embryonic Stem Cells Edited by PAUL M. WASSAUMAN AND GORDON M. KELLER VOLUME 366. Protein Phosphatases Edited by SUSANNE KLUMPP AND JOSEF KRIEGLSTEIN VOLUME 367. Liposomes (Part A) Edited by NEJAT DU¨ZGU¨NES VOLUME 368. Macromolecular Crystallography (Part C) Edited by CHARLES W. CARTER, JR., AND ROBERT M. SWEET VOLUME 369. Combinational Chemistry (Part B) Edited by GUILLERMO A. MORALES AND BARRY A. BUNIN VOLUME 370. RNA Polymerases and Associated Factors (Part C) Edited by SANKAR L. ADHYA AND SUSAN GARGES VOLUME 371. RNA Polymerases and Associated Factors (Part D) Edited by SANKAR L. ADHYA AND SUSAN GARGES VOLUME 372. Liposomes (Part B) Edited by NEJAT DU¨ZGU¨NES VOLUME 373. Liposomes (Part C) Edited by NEJAT DU¨ZGU¨NES VOLUME 374. Macromolecular Crystallography (Part D) Edited by CHARLES W. CARTER, JR., AND ROBERT W. SWEET VOLUME 375. Chromatin and Chromatin Remodeling Enzymes (Part A) Edited by C. DAVID ALLIS AND CARL WU VOLUME 376. Chromatin and Chromatin Remodeling Enzymes (Part B) Edited by C. DAVID ALLIS AND CARL WU VOLUME 377. Chromatin and Chromatin Remodeling Enzymes (Part C) Edited by C. DAVID ALLIS AND CARL WU VOLUME 378. Quinones and Quinone Enzymes (Part A) Edited by HELMUT SIES AND LESTER PACKER VOLUME 379. Energetics of Biological Macromolecules (Part D) Edited by JO M. HOLT, MICHAEL L. JOHNSON, AND GARY K. ACKERS VOLUME 380. Energetics of Biological Macromolecules (Part E) Edited by JO M. HOLT, MICHAEL L. JOHNSON, AND GARY K. ACKERS VOLUME 381. Oxygen Sensing Edited by CHANDAN K. SEN AND GREGG L. SEMENZA
Methods in Enzymology
li
VOLUME 382. Quinones and Quinone Enzymes (Part B) Edited by HELMUT SIES AND LESTER PACKER VOLUME 383. Numerical Computer Methods (Part D) Edited by LUDWIG BRAND AND MICHAEL L. JOHNSON VOLUME 384. Numerical Computer Methods (Part E) Edited by LUDWIG BRAND AND MICHAEL L. JOHNSON VOLUME 385. Imaging in Biological Research (Part A) Edited by P. MICHAEL CONN VOLUME 386. Imaging in Biological Research (Part B) Edited by P. MICHAEL CONN VOLUME 387. Liposomes (Part D) Edited by NEJAT DU¨ZGU¨NES VOLUME 388. Protein Engineering Edited by DAN E. ROBERTSON AND JOSEPH P. NOEL VOLUME 389. Regulators of G-Protein Signaling (Part A) Edited by DAVID P. SIDEROVSKI VOLUME 390. Regulators of G-Protein Signaling (Part B) Edited by DAVID P. SIDEROVSKI VOLUME 391. Liposomes (Part E) Edited by NEJAT DU¨ZGU¨NES VOLUME 392. RNA Interference Edited by ENGELKE ROSSI VOLUME 393. Circadian Rhythms Edited by MICHAEL W. YOUNG VOLUME 394. Nuclear Magnetic Resonance of Biological Macromolecules (Part C) Edited by THOMAS L. JAMES VOLUME 395. Producing the Biochemical Data (Part B) Edited by ELIZABETH A. ZIMMER AND ERIC H. ROALSON VOLUME 396. Nitric Oxide (Part E) Edited by LESTER PACKER AND ENRIQUE CADENAS VOLUME 397. Environmental Microbiology Edited by JARED R. LEADBETTER VOLUME 398. Ubiquitin and Protein Degradation (Part A) Edited by RAYMOND J. DESHAIES VOLUME 399. Ubiquitin and Protein Degradation (Part B) Edited by RAYMOND J. DESHAIES VOLUME 400. Phase II Conjugation Enzymes and Transport Systems Edited by HELMUT SIES AND LESTER PACKER
lii
Methods in Enzymology
VOLUME 401. Glutathione Transferases and Gamma Glutamyl Transpeptidases Edited by HELMUT SIES AND LESTER PACKER VOLUME 402. Biological Mass Spectrometry Edited by A. L. BURLINGAME VOLUME 403. GTPases Regulating Membrane Targeting and Fusion Edited by WILLIAM E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 404. GTPases Regulating Membrane Dynamics Edited by WILLIAM E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 405. Mass Spectrometry: Modified Proteins and Glycoconjugates Edited by A. L. BURLINGAME VOLUME 406. Regulators and Effectors of Small GTPases: Rho Family Edited by WILLIAM E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 407. Regulators and Effectors of Small GTPases: Ras Family Edited by WILLIAM E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 408. DNA Repair (Part A) Edited by JUDITH L. CAMPBELL AND PAUL MODRICH VOLUME 409. DNA Repair (Part B) Edited by JUDITH L. CAMPBELL AND PAUL MODRICH VOLUME 410. DNA Microarrays (Part A: Array Platforms and Web-Bench Protocols) Edited by ALAN KIMMEL AND BRIAN OLIVER VOLUME 411. DNA Microarrays (Part B: Databases and Statistics) Edited by ALAN KIMMEL AND BRIAN OLIVER VOLUME 412. Amyloid, Prions, and Other Protein Aggregates (Part B) Edited by INDU KHETERPAL AND RONALD WETZEL VOLUME 413. Amyloid, Prions, and Other Protein Aggregates (Part C) Edited by INDU KHETERPAL AND RONALD WETZEL VOLUME 414. Measuring Biological Responses with Automated Microscopy Edited by JAMES INGLESE VOLUME 415. Glycobiology Edited by MINORU FUKUDA VOLUME 416. Glycomics Edited by MINORU FUKUDA VOLUME 417. Functional Glycomics Edited by MINORU FUKUDA VOLUME 418. Embryonic Stem Cells Edited by IRINA KLIMANSKAYA AND ROBERT LANZA
Methods in Enzymology
liii
VOLUME 419. Adult Stem Cells Edited by IRINA KLIMANSKAYA AND ROBERT LANZA VOLUME 420. Stem Cell Tools and Other Experimental Protocols Edited by IRINA KLIMANSKAYA AND ROBERT LANZA VOLUME 421. Advanced Bacterial Genetics: Use of Transposons and Phage for Genomic Engineering Edited by KELLY T. HUGHES VOLUME 422. Two-Component Signaling Systems, Part A Edited by MELVIN I. SIMON, BRIAN R. CRANE, AND ALEXANDRINE CRANE VOLUME 423. Two-Component Signaling Systems, Part B Edited by MELVIN I. SIMON, BRIAN R. CRANE, AND ALEXANDRINE CRANE VOLUME 424. RNA Editing Edited by JONATHA M. GOTT VOLUME 425. RNA Modification Edited by JONATHA M. GOTT VOLUME 426. Integrins Edited by DAVID CHERESH VOLUME 427. MicroRNA Methods Edited by JOHN J. ROSSI VOLUME 428. Osmosensing and Osmosignaling Edited by HELMUT SIES AND DIETER HAUSSINGER VOLUME 429. Translation Initiation: Extract Systems and Molecular Genetics Edited by JON LORSCH VOLUME 430. Translation Initiation: Reconstituted Systems and Biophysical Methods Edited by JON LORSCH VOLUME 431. Translation Initiation: Cell Biology, High-Throughput and Chemical-Based Approaches Edited by JON LORSCH VOLUME 432. Lipidomics and Bioactive Lipids: Mass-Spectrometry–Based Lipid Analysis Edited by H. ALEX BROWN VOLUME 433. Lipidomics and Bioactive Lipids: Specialized Analytical Methods and Lipids in Disease Edited by H. ALEX BROWN VOLUME 434. Lipidomics and Bioactive Lipids: Lipids and Cell Signaling Edited by H. ALEX BROWN
liv
Methods in Enzymology
VOLUME 435. Oxygen Biology and Hypoxia Edited by HELMUT SIES AND BERNHARD BRU¨NE VOLUME 436. Globins and Other Nitric Oxide-Reactive Protiens (Part A) Edited by ROBERT K. POOLE VOLUME 437. Globins and Other Nitric Oxide-Reactive Protiens (Part B) Edited by ROBERT K. POOLE VOLUME 438. Small GTPases in Disease, Part A (in preparation) Edited by WILLIAM E. BALCH, CHANNING J. DER, AND ALAN HALL
S E C T I O N
O N E
NITRIC OXIDE-METABOLISING AND DETOXIFYING ENZYMES
This page intentionally left blank
C H A P T E R
O N E
Structural Studies on Flavodiiron Proteins Joa˜o B. Vicente, Maria Arme´nia Carrondo, Miguel Teixeira, and Carlos Fraza˜o Contents 4 4
1. Introduction 2. Crystallization of Flavodiiron Proteins 3. Diffraction Data Collection, Structure Determination, and Refinement 4. Overall Description of Structures 4.1. Metallo-b-lactamase-like domain of flavodiiron proteins 4.2. Flavodoxin domain of flavodiiron proteins 4.3. Features of the nonheme diiron center 4.4. Features of the flavin mononucleotide moiety 5. Conclusion References
7 8 9 11 11 14 16 17
Abstract Crystallographic studies on flavodiiron proteins (FDPs) have revealed that the common sequence core (400 residues) that defines this protein family comprises two structural domains. The N-terminal domain (of approximately 250 residues) displays a metallo-b-lactamase-like-fold, being indeed structurally homologous to b-lactamases and glyoxalases, despite the poor sequence similarity. Whereas b-lactamases have mono- or dizinc sites and glyoxalases a mixed iron–zinc site, the lactamase domain of FDPs harbors a nonheme diiron center with carboxylate and histidine residues as ligands, assigned as the active site of NO and/or O2 reduction. The C-terminal domain of FDPs is characterized by a flavodoxin-like fold, homologous to short-chain flavodoxins, and harbors a flavin mononucleotide moiety, stabilized by van der Waals interactions and a number of hydrogen bonds. Structures of FDPs obtained in different conditions and oxidation states display some heterogeneities, mostly at the diiron site, but still fail to provide unequivocal evidence for some pending questions regarding the substrate activation mechanism of FDPs, namely the preference for either Instituto de Tecnologia Quı´mica e Biolo´gica, Universidade Nova de Lisboa, Oeiras, Portugal Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37001-8
#
2008 Elsevier Inc. All rights reserved.
3
4
Joa˜o B. Vicente et al.
substrate (NO or oxygen) observed in different members of this protein family. More structural studies are therefore required to achieve a deeper understanding on these matters.
1. Introduction Flavodiiron proteins (FDPs) are widespread in Archaea and Bacteria (Wasserfallen et al., 1998), mostly anaerobic organisms, extending also to the Eukarya, namely in a few pathogenic protozoa (Andersson et al., 2003, 2006; Sarti et al., 2004). Currently, accumulated data suggest alternative or complementary roles in the detoxification of NO and/or oxygen (toxic for anaerobes). Initial studies on Desulfovibrio gigas rubredoxin:oxygen oxidoreductase (Dg_ROO, the designation given to the FDP of this organism) indicated a role in the protection against oxygen toxicity to this strict anaerobe (Chen et al., 1993), as D. gigas was considered at the time. Subsequent reports focusing on Escherichia coli FDP (named flavorubredoxin, FlRd, NorV) have demonstrated a function of this protein in the anaerobic protection against NO-derived stress (Gardner et al., 2002; Gomes et al., 2002b; Justino et al., 2005). Several members of this protein family were later shown to be endowed with both NO and oxygen reductase activities, with different substrate specificities, as judged by the comparative affinities (Rodrigues et al., 2006; Silaghi-Dumitrescu et al., 2003, 2005b). More recent reports on FDPs from methanogenic archaea suggest an exclusive role in oxygen reduction (Seedorf et al., 2004, 2007), accepting electrons directly from the organic cofactor F420H2 (ubiquitous in methanogens). Although preliminary studies on several FDPs revealed important features common to the protein family, it was not until the first crystallographic structure of an FDP was solved that the nature of the active site was unraveled (Fraza˜o et al., 2000). Indeed, the structure of Dg_ROO revealed several aspects (described in detail later), which shed some light on the functional properties of FDPs. Although the subsequent FDP structures confirmed many of the Dg_ROO structural features, some heterogeneities at the active site are noteworthy. Several structural studies also aimed at understanding the structural conditionals that confer FDPs a selectivity for either substrate (NO or oxygen).
2. Crystallization of Flavodiiron Proteins The flavodiiron proteins for which X-ray crystallographic structures have been solved were isolated as described in Vicente et al. (2007), except the FDP from Thermotoga maritima (Tm_FDP). Whereas D. gigas rubredoxin:oxygen oxidoreductase (Dg_ROO) was isolated from its source
Structural Studies on Flavodiiron Proteins
5
organism (Chen et al., 1993), the remainder of FDPs were overexpressed heterologously in E. coli (Seedorf et al., 2007; Silaghi-Dumitrescu et al., 2005a). Tm_FDP was overexpressed in E. coli and purified by a highthroughput automated method (DiDonato et al., 2004). Flavodiiron proteins were crystallized essentially by two methods: (i) the batch method, using Zinc acetate as precipitant, or (ii) the vapor diffusion method, using alcohol type precipitants (PEGs or MPD). When crystal cryostabilization was necessary for data collection, it was performed by incremental additions of the stabilizing agent. Dg_ROO crystals, obtained with PEG 1 to 6K as precipitating agents in a wide range of pH (5–9) (Fraza˜o et al., 1999), appeared within 1 day at room temperature in large numbers, amidst a gelatinous precipitate. To prevent the formation of this gel and to control the number of crystals, crystallization trials were performed at 277K, and also adding the detergent SB12 (N-dodecyl-N,N-dimethyl-3-ammonio-1-propansulfonate). Crystals obtained at both temperatures had similar parallelepiped shapes, although their diffraction quality changed significantly. Whereas the diffraction of crystals obtained at 277K exhibited twinning statistics, those obtained at room temperature developed as single crystals. Crystals that allowed solving the Dg_ROO structure were typically obtained as follows. A sitting drop composed by 3 ml of 10 mg/ml protein solution and an equal volume of precipitant solution [PEG 6K 10% (v/v), 100 mM Tris-maleic acid, pH 6.0] was equilibrated against 500 ml of the same solution. Orange-brown parallelepiped-shaped crystals grew within approximately 3 weeks up to 0.2 to 0.4 mm in length and were cryostabilized by adding to 20-ml drops of precipitant solution containing the crystals, small amounts of precipitant solution (initially 0.2 ml each) complemented with glycerol (25%, v/v), in the cold room, up to a final glycerol concentration of 25% (v/v). After stabilization, crystals were flash frozen in a nitrogen stream and diffraction data collected. Crystals belonged to space group P21212 (cell dimensions ˚ , b ¼ 101.2 A ˚ , and c ¼ 90.8 A˚), diffracted to 2.5 A ˚ resolution a ¼ 98.2 A and contained two molecules (one dimer) in the asymmetric unit. The flavodiiron protein from Moorella thermoacetica (Mt_FDP) was crystallized (Silaghi-Dumitrescu et al., 2005a) by the batch method in melting point capillaries (at room temperature) by mixing 10 ml of 1 mM (42 mg/ml) oxidized (as-isolated) (Mt_FDP_ox) solution to an equal volume of precipitant solution. Two slightly different precipitant solutions yielded crystals, 200 mM zinc acetate, 5% 2-propanol, 50 mM sodium cacodylate, pH 6.5, and 200 mM zinc acetate, 10% PEG 3000, 100 mM sodium acetate, pH 4.5. Relatively large crystals (0.2 0.2 0.5 mm) formed within 7 to 10 days, endowed with diffraction quality, were cryostabilized by soaking the crystals in mother liquor [1:1 (v/v) 25 mM MOPS (pH 7.3):precipitant] containing step increments of 5% ethylene glycol up to 20%, with each step taking 20 min. Crystals of
6
Joa˜o B. Vicente et al.
reduced Mt_FDP (Mt_FDP_red) were obtained in an anaerobic glove box by the addition of sodium dithionite powder to as-isolated Mt_FDP crystals [in 200 ml of mother liquor containing 20% (v/v) ethylene glycol and the pH 6.5 precipitant], resulting in crystals changing from orange-brown to colorless within a few minutes. Crystals of Mt_FDP reoxidized with NO (Mt_FDP_NO) were obtained starting from the reduced crystals prepared as described earlier. These crystals were back-soaked into the same mother liquor (though lacking sodium dithionite) and further incubated with the NO-releasing compound DEA-NONOate (in powder), leading to a fast color change back to the original orange-brown, attesting the Mt_FDP reoxidation by NO. The crystals obtained in three states (oxidized, reduced, and reduced þ NO reacted) were flash frozen in liquid nitrogen; all belonged to space group P43212. The cell dimensions and resolution of ˚ , c ¼ 276.7 A ˚ , and 3.0 A ˚ resolution for diffraction data were a ¼ b ¼ 159.6 A ˚ ˚ crystals of Mt_FDP_ox; a ¼ b ¼ 159.6 A, c ¼ 278.1 A, and 2.8 A˚ resolution ˚ , c ¼ 279.1 A ˚ , and 2.8 A˚ for Mt_FDP_red crystals; and a ¼ b ¼ 159.6 A resolution for Mt_FDP_NO crystals. Moreover, all the crystals were shown to contain four molecules (an assembly of two homodimers) in the asymmetric unit. Crystals of T. maritima FDP were obtained as part of a high-throughput automated method for protein production, crystallization, and structure determination, focusing on the proteome of this organism (DiDonato et al., 2004). Because the structure of Tm_FDP was deposited in the PDB but not published, information on the crystallization procedure is still scarce (PDB entry 1VME). Crystals obtained at 277 K in sitting nanodrops, by the vapor diffusion method, using 35.0% MPD, 0.1 M acetate (pH 4.5), belonged to space group P21 with cell dimensions a ¼ 55.24 A˚, b ¼ ˚ , c ¼ 90.13, and b ¼ 95.43 , diffracted to 1.8 A˚ and contained 95.83 A two molecules (one dimer) in the asymmetric unit. Contrary to other FDPs for which the crystallographic structure was solved, the FprA from Methanothermobacter marburgensis (Mm_FprA) was purified and crystallized only in anaerobic conditions (in an anaerobic glove box) (Seedorf et al., 2007). Different crystallization conditions yielded three crystal forms, grown in drops up to 20 ml using 20 mg/ml Mm_FprA solution and precipitant solution, all three of them obtained by the hanging drop vapor diffusion method, at 283 K, and in the presence of 1 mM dithiothreitol. Crystals obtained with 0.2 M ammonium sulfate, 0.1 M MES/KOH, pH 6.5, and 16–22% PEG MME 5000 displayed two different monoclinic P21 crystal forms, one containing eight molecules in the asym˚ , b ¼ 123.1 A˚, c ¼ 135.9 A ˚ , and metric unit, with cell dimensions a ¼ 97.8 A ˚ b¼ 103.4 , diffracting to 2.25 A resolution; and the second P21 form containing four molecules in the asymmetric unit, with cell dimensions ˚ , b ¼ 120.9 A ˚ , c ¼ 92.7 A ˚ , and b¼ 110.4 , diffracting to 1.7 A ˚ a ¼ 73.7 A resolution. The third crystal form resulted from a lower concentration of
Structural Studies on Flavodiiron Proteins
7
PEG MME 5000 (8–16%), compensated with 15% glycerol, yielding tetragonal crystals in space group P43212, displaying four molecules in the ˚ , c ¼ 450.4 A ˚ , that asymmetric unit, with cell dimensions a ¼ b ¼ 88.7 A ˚ diffracted to 2.25 A resolution. Differences in data sets also resulted from the distinct ways the crystals were manipulated. The first crystal form (P21 with eight molecules in the a.u.) was measured after flash cooling the crystals inside the anaerobic chamber (grown in the presence of F420H2, the electron donor for Mm_FprA) in liquid nitrogen, corresponding to reduced Mm_FprA. The second and third crystal forms (P21 with four molecules in the a.u., and P43212) were flash frozen in a nitrogen gas stream after being exposed to air for minutes at 191 K, yielding presumably oxidized forms of Mm_Fpra.
3. Diffraction Data Collection, Structure Determination, and Refinement Two diffraction data sets of Dg_ROO were collected at cryogenic temperatures, a four-wavelength MAD data set around the Fe absorption ˚ resolution at beam line BM14 of ESRF (Fraza˜o et al., 2000) edge to 2.7 A ˚ resolution at beam line X11 of DESY, and a single wavelength set to 2.5 A EMBL Hamburg outstation (Fraza˜o et al., 1999). Diffraction intensities were measured with DENZO and scaled together with SCALEPACK of the HKL suite (Otwinowski and Minor, 1997). MADSYS (Hendrickson and Ogata, 1997) extracted F A values from MAD data which were used in SHELXS-97 (Sheldrick, 1990) to locate three out of the four Fe sites. The sites were further refined with SHARP (De La Fortelle and Bricogne, 1997), which found the fourth site from residual maps and included it in the phase improvement process using SOLOMON (Solomon et al., 2000), to ˚ . The polypeptide chain was modeled in the an overall f.o.m. of 0.95 at 2.7 A experimental maps followed by residues assignment using XTALVIEW (McRee, 1992). The refinement to 2.5 A˚ of the two molecules in the a.u. proceeded using SHELXL (Sheldrick and Schneider, 1997), restraining to their common values homologous 1–4 dihedral distances and atomic displacement parameters (a.d.p.s) of homologous atoms, and of waters with equivalent H-bonds to each monomer. The diiron centers and their ligating residues were restrained to a common geometry without target values. A residual lobe of electron density found close to the diiron site in different Fourier maps was initially assigned to a water molecule, but its refinement led to too low a.d.p.s. Taking into account the crystallization conditions it was then tentatively assigned to a dioxygen molecule. Tm_FDP diffraction data were obtained at 100 K at SSRL on beam line 9–2. Intensities were measured with MOSFLM (Leslie, 2006) and scaled together using SCALA (Evans, 2006) of the CCP4 suite (Collaborative
8
Joa˜o B. Vicente et al.
Computational Project Number 4, 1994). The phase problem was solved by MAD using a selenium-derivatized protein crystal. The structure was refined using the maximum likelihood target in REFMAC with six TLS groups ( Winn et al., 2003). Mt_FDP diffraction data were acquired at 98 K at the Advanced Light Source, Berkeley, on beam line 5.0.2. The molecular replacement solution by CNS (Brunger et al., 1998), using the Dg_ROO structure truncated to poly-Ala as the search model, was corroborated by the SAD solution of the iron sites using highly redundant (760 degrees) diffraction data, collected at the University of Georgia on a Rigaku Ru-200 rotating anode equipped with Osmic focusing mirrors and a R-axisIIc imaging plate detector. The graphical software O ( Jones et al., 1991) was used for model edition and completion and CNS (Brunger et al., 1998) for structure refinement, using the four-fold noncrystallographic symmetry operators to improve the quality of the electron density maps. Mm_FprA diffraction data were collected at cryo-temperature using beam line X10SA of the Swiss-Light-Source. Diffraction data were processed and scaled using the HKL (Otwinowski and Minor, 1997) and XDS (Kabsch, 1993) packages. The structure of active oxidized Mm_FprA was solved by molecular replacement with EPMR (Kissinger et al., 1999) using the Mt_FDP structure as the search model (the Dg_ROO model gave less reliable results). Phases for the two other crystal forms of Mm_FprA were obtained by molecular replacement using the active oxidized model. Refinement and model edition were performed with CNS (Brunger et al., 1998) and O ( Jones et al., 1991), making use of the noncrystallographic symmetry relationships. Refinement was completed with REFMAC (Murshudov et al., 1997) using the TLS option (each monomer treated as a separate TLS group), maximum likelihood minimization, and isotropic a.d.p.s refinement.
4. Overall Description of Structures Flavodiiron proteins have been isolated as functional homodimers and homotetramers (Vicente et al., 2007), which have been confirmed by the available X-ray structures (Fraza˜o et al., 2000; Seedorf et al., 2007; SilaghiDumitrescu et al., 2005a). The homodimeric structure (Fig. 1.1) results from a ‘‘head-to-tail’’ arrangement of two monomers where approximately 12% of the possible solvent-accessible surface becomes occluded (Fraza˜o et al., 2000; Seedorf et al., 2007). The tetrameric arrangement is composed by a loose dimer of two such dimers, where a further 9.5% of otherwise solvent-accessible surface becomes occluded (Seedorf et al., 2007). The available structures show similar three-dimensional arrangements with
Structural Studies on Flavodiiron Proteins
9
Figure 1.1 Three-dimensional structure of Desulfovibrio gigas rubredoxin:oxygen oxidoreductase. Solvent-accessible transparent surface and ribbon representation of the ‘‘head-to-tail’’ homodimeric arrangement of D. gigas rubredoxin:oxygen oxidoreductase (Dg__ROO, PDB code 1E5D), the first flavodiiron protein whose crystallographic structure was solved. Each monomer is composed of two structural domains: the N-terminal domain with a metallo-b-lactamase-like fold is represented in lighter colors, whereas the C-terminal domain with a flavodoxin-like fold is represented in darker colors. The head-to-tail homodimeric structure places the nonheme diiron center active site (golden spheres) of each monomer in close proximity with the FMN moiety (yellow sticks) of the other monomer.
˚ upon superposition of the overall monomer Ca r.m.s.d.s within 1.1 to 1.8 A structures from four different species. The FDP monomers are composed of two different structural domains: an N-terminal domain with a metallo-blactamase-like fold and a C-terminal domain showing a flavodoxin-like fold.
4.1. Metallo-b-lactamase-like domain of flavodiiron proteins The lactamase N-terminal domain (up to residue 249, using the Dg_ROO numbering) folds in a abba sandwich, with the two inner b sheets being surrounded by two sets of three solvent-exposed a helices. Despite the low sequence identity between the lactamase domain of FDPs and other lactamase-like proteins (Gomes et al., 2002a), its overall fold (Fig. 1.2) is structurally very similar to the folds found in both class B Zn-b-lactamases (Carfi et al., 1995) (that hydrolyze penicillin) and human glyoxalase (Cameron et al., 1999) (that convert 2-oxoaldehydes into the corresponding 2-hydroxycarboxylic acids). Another striking difference between the lactamase domain of FDPs and other lactamase-like proteins concerns the nature of the metal active site. FDPs were found to harbor a nonheme diiron center (with histidine and carboxylate residues as ligands), whereas lactamases contain a mono- or a dizinc center and glyoxalases bind a mixed iron–zinc
10
Joa˜o B. Vicente et al.
A N
C B N
C C
N
C
Figure 1.2 Topology diagrams of the N-terminal domain of flavodiiron proteins and other lactamase-like folds.Topology diagrams of the structurally homologous lactamase domains found in (A) glyoxalases (first domain of human glyoxalase II, PDB entry 1QH5) (Cameron et al., 1999); (B) Zn-binding metallo-b-lactamases (Stenotrophomonas maltophilia, PDB entry 1SML) (Ullah et al., 1998); and (C) the N-terminal domain of Desulfovibrio gigas rubredoxin:oxygen oxidoreductase, PDB entry 1E5D) (Fraza˜o et al., 2000). Circles depict helices, triangles depict b chains, and N and C represent the peptide chain direction. The simplest common secondary structure, found in glyoxalase, is represented in dark gray, light gray displays additional secondary structure elements present in metallo-b-lactamases, and white depicts those unique to FDPs.
binuclear center (Cameron et al., 1999; Zang et al., 2001). In FDPs the diiron site is located within a shallow groove in a cavity at the interface of the two sheets and is surrounded by ab loops and the C-terminal domain from the other monomer (which is made possible by the ‘‘head-to-tail’’ dimeric arrangement). Whereas the substrates of lactamases and glyoxalase are bulky and require large cavities near the binuclear metal centers, the FDPs substrates (NO and/or O2) are much smaller and can reach the diiron site, despite being covered by an additional two-stranded b sheet.
Structural Studies on Flavodiiron Proteins
11
4.2. Flavodoxin domain of flavodiiron proteins The C-terminal domain of FDPs (from residue 250, Dg_ROO numbering) displays the typical aba flavodoxin fold, where a flavin mononucleotide (FMN) moiety is embedded that allegedly acts as the electron donor to the diiron site, where reduction of NO and/or oxygen takes place. Within a ˚ away from the diiron center, too far single monomer, the flavin is 35 A to be considered for effective electron transfer between the two redox sites. This observation highlights the relevance of the quaternary structure common to FDPs, in which the minimal operative unit is thought to be the ‘‘head-to-tail’’ homodimer. This arrangement places the diiron ˚ ) with the FMN moiety center of each monomer in close contact (6 A from the opposing monomer, yielding two independent catalytic sites per homodimer. Despite the significant structural similarity between the flavodoxin domain of FDPs and short-chain flavodoxins [comparable to the homology among flavodoxins themselves (Fraza˜o et al., 2000)], an important difference is that the FMN reduction potentials in FDPs are higher than those in flavodoxins (see e.g., Gomes et al., 1997; Ludwig et al., 1997; Silaghi-Dumitrescu et al., 2003; Vicente and Teixeira, 2005), an observation that has been assigned to the relative excess of basic versus acidic residues surrounding the FMN isoalloxazine ring in FDPs.
4.3. Features of the nonheme diiron center Since the nonheme diiron center has been assigned as the active site of NO/O2 reduction in flavodiiron proteins (Gomes et al., 2002b; Silaghi-Dumitrescu et al., 2003), special emphasis has been given to analysis of the structural and functional features of this metal center. One obvious aspect is the pursuit of an understanding of the molecular basis for the selectivity of each studied FDP for either substrate (NO and/or O2). Although a definitive answer is far from being attained, some heterogeneities at the FDPs diiron centers (whose overall structure is globally conserved) are noteworthy. Whereas in Dg_ROO, Mt_FDP, and Tm_FDP the diiron center geometry is retained within the independent molecules in the a.u.s for the various available oxidation states, in Mm_FprA the diiron center differs significantly not only among the different oxidation states, but also between independent molecules of the same asymmetric units (by alternative conformations within a monomer). As a common feature in most structures, each iron is coordinated by two imidazole nitrogens, two oxygens from carboxylate residues, and a bridging solvent molecule, assigned as a putative m-oxo-, m-hydroxyl-, or m-aquo-ligand, since resolutions of the structures do not allow determination of its protonation state. However, in Mm_FprA, only 1 out of 16 independent structures deposited reports a bridging m-oxygen atom in the structure at the highest resolution. Interestingly, the bridging moiety appears to be redox insensitive, as the
12
Joa˜o B. Vicente et al.
reduced structure of Mt_FDP still holds it, unlike other nonheme diiron centers that lose and regain the bridge throughout the oxygen-activating reaction cycle (Solomon et al., 2000). Considering the proximal iron the one closer to FMN (denoting the other as distal iron) and using Dg_ROO numbering (also throughout the text), the proximal iron is coordinated by residues His79, His146, and Glu81, and the distal iron is coordinated by His84, His226, and Asp83, with Asp165 acting as a bridging ligand for both irons. The ligands herein described are almost strictly conserved in this protein family, with few exceptions observed in FDPs encoded in the genomes of some cyanobacteria, which encode other copies of FDPs within the same genome, that retain the conserved ligands. A fifth coordination position remains open in each iron (somewhat trans to His79 and His84, parallel to each other), where a substrate is prone to bind. The exception to the general coordination observed in most FDPs is precisely found in the structure of Dg_ROO (Fraza˜o et al., 2000), where the His84 ligand is displaced from the center, being replaced as a ligand by a solvent water molecule. This is because of the unique side chain conformation of His84, which makes a trans chi-1 angle, instead of the gauche(-) conformation found in other FDPs, in lactamases (Ullah et al., 1998), and in human glyoxalase (Cameron et al., 1999). A closer look at the structures allowed finding a structural feature that accounts for this heterogeneity: one H-bond between the His84 imidazole ND1 atom and the neighboring Asp225 carboxylate. The equivalent position to Asp225 in other FDP sequences (survey of 150 sequences) is not conserved, but varies among Ser (50%), Gly (34%), Asp (8%), and Ala (6%). From the available crystal structures, only Dg_ROO has an Asp in this position, which can account for its unique coordination, contrasting with the other FDPs, where the replacing residues are not able to make the same H-bond that ‘‘pulls’’ His84 away from the center in Dg_ROO. A Gly residue is in the equivalent position in Tm_FDP and a Ser in Mt_FDP, whose OG atom is too far to make the H-bond. Although these observations present a structural justification for the ligand heterogeneity between Dg_ROO and the other FDPs, there are still no clues for the structural properties that determine the specificity of FDPs for either substrate. To infer functional properties from the structural studies, it is important to attempt trapping reaction intermediates; therefore, structures of FDPs were obtained in different oxidation states, except Dg_ROO and Tm_FDP, which were reported only in the as-isolated oxidized state. Whereas Mt_FDP was crystallized in the oxidized form and then the crystals were reduced and reoxidized with an NO releaser, the Mm_FprA was crystallized anaerobically in the reduced form and oxidized upon O2 exposure before the crystal flash-cooling procedure.
13
Structural Studies on Flavodiiron Proteins
FMN
E81
FMN
E81
D83
D83 H226
D228 H146
H146 H79
H84 D165
D225
H79
H84
D225
D165
Figure 1.3 Structural features of the nonheme diiron active site of flavodiiron proteins. Stereo representation (cross-eyes) of the superposition of the diiron center from Desulfovibrio gigas rubredoxin:oxygen oxidoreductase (Dg__ROO) is shown in black (with residues numbering), whereas that of Thermotoga maritima flavodiiron protein (Tm_FDP) is represented in light gray.The two selected structures depict the heterogeneity of the active sites in the structures of FDPs, related with a conserved histidine residue (His84 in Dg__ROO), which is a ligand of the distal iron in most FDPs, but in Dg__ROO is replaced by a water molecule and appears displaced from the center, making a H-bond with Asp225 (dark gray). InTm_FDP, position 225 is a Gly residue (white symbol).
One evident observation (and possibly the most significant one) upon comparing the oxidized and reduced structures of Mt_FDP is that the latter has a larger active site cavity, being even able to harbor one ethylene glycol molecule ‘‘above’’ the diiron center, where an oxygen molecule is found in the oxidized structure, and one solvent water molecule in the NO-reoxidized structure. The aforementioned molecules located above the diiron center were refined at noncovalent bound distances to the iron atoms, on average at 2.8 A˚ among the four independent structures, to be compared with the typical 1.84 A˚ for Fe-O2 distances in high-resolution structures in the Cambridge Structural Database (Fraza˜o et al., 2000). Contrasting with the low degree of structural variations in the different oxidation states of Mt_FDP, Mm_FprA structures display large differences among the reported crystal forms. Upon oxidation by oxygen exposure of the crystals obtained in anaerobic conditions, the diiron center becomes disrupted. Whereas the distal iron maintains its coordination sphere, the proximal iron is significantly displaced and its former ligands are displaced from the original conformation. Particularly, Glu81 is rotated away from the proximal position and forms a new remote metal-binding site, along ˚ from the distal with His26 and His267 (Mm_FprA numbering), at 7 A iron. Moreover, a third putative metal-binding site at the protein surface is reported, coordinated by His151, Asp320 (Mm_FprA residues numbering), and a water molecule. This displacement of the proximal iron is accompanied by a large conformational arrangement of the loop between residues 148–151, designated as ‘‘switch loop,’’ increasing the accessibility of
14
Joa˜o B. Vicente et al.
the redox centers. The resulting structure was therefore called the ‘‘inactive oxidized state.’’ As an intermediate situation between the as-isolated ‘‘active reduced’’ Mm_FprA structure and the ‘‘inactive oxidized state,’’ oxidation of the monoclinic crystals yielded an integral diiron center, although displaying a mixed conformation of the switch loop: 0.6 occupancy of the closed conformation found in the ‘‘reduced state’’ (and all the other FDP structures) and 0.4 occupancy of the open conformation observed in the ‘‘inactive oxidized state.’’ This ‘‘intermediate’’ structure was designated by the ‘‘active oxidized state.’’ Another observation regarding the oxidized structures is that the inactive open loop conformation swaps into trans the otherwise cis peptide bond Leu145-His146 (a nonproline cis bond also present in Dg_ROO and Tm_FDP, but not reported in Mt_FDP), which is necessary to project the imidazolyl ring toward the proximal iron. The four independent molecules in the asymmetric unit of the active oxidized state display a heterogeneously populated catalytic cavity, harboring either a diatomic molecule or a sulfate together with a monoatomic m-ligand to the diiron. The three different redox and conformational states of Mm_FprA may represent reaction intermediates with mechanistic implications. In the absence of reducing equivalents to feed the catalytic cycle of oxygen reduction by Mm_FprA, it appears that the enzyme may become inactivated upon prolonged oxygen exposure, still being unclear whether this inactivation is reversible. Notably, although the proximal iron is displaced away from its original location, it may still be bound to the protein in the alternative putative metal-binding sites aforementioned. Because the other structures were obtained in aerobic conditions with integer diiron centers, it is presently uncertain whether the observations regarding Mm_FprA are relevant to other FDPs.
4.4. Features of the flavin mononucleotide moiety Despite the poor sequence similarity between the C-terminal domain of flavodiiron proteins and short-chain flavodoxins, their structural homology is considerable, and both have been shown to bind a flavin mononucleotide moiety. The notable exception is Tm_FDP, whose deposited structure lacks the FMN moiety, an observation that can be accounted for by at least two nonmutually exclusive facts. On the one hand, as described by Vicente and co-workers (2007), the flavin content of heterologously overexpressed FDPs was improved by growing the expression cells and performing the purifications under specifically controlled conditions, which contrast with the experimental approach undertaken by high-throughput protocols. On the other hand, the structural motifs that bind the FMN in Dg_ROO, Mt_FDP, and Mm_FprA are conserved regions in the FDP family, but are less conserved in Tm_FDP. Therefore, the absence of FMN from the Tm_FDP structure can result from accumulation of the inherent
Structural Studies on Flavodiiron Proteins
15
lower ability of Tm_FDP to stabilize the FMN (inferred from the sequence) and the automated high-throughput overexpression protocol. The absence of FMN in Tm_FDP is accompanied by two alternative conformations of the main chain segment composed by residues 312–320: one conformer overlaps with the protein backbone of the remaining FDPs, while the other partially occupies the region otherwise occupied by the missing FMN moiety. The crystallographic structures of FDPs provide structural clues for some of the spectroscopic and functional properties determined for studied FDPs. As mentioned earlier, the prevalence of basic over acidic residues in the flavin pocket of FDPs accounts for the higher potential of the second reduction step of FMN, with respect to canonical flavodoxins. The nature of the flavin semiquinone (one-electron reduced) that is stabilized also differs between the two protein families, which are related with the accessibility for protonation of the N5 position in the flavin isoalloxazine ring system. The FDP polypeptide establishes interactions with the isoalloxazine ring of FMN by both van der Waals contacts and H-bonds between main-chain amino groups and ring N atoms (at positions 1 and 5), or ring carbonyl O atoms (at positions 2 and 4). The H-bond with N5 precludes the protonation of this site, in contrast to what happens with canonical flavodoxins where N5 is accessible for protonation. Therefore, in the one-electron reduced state, the FMN in FDPs is deprotonated at N5 and stabilizes the red anionic semiquinone, whereas the FMN in canonical flavodoxins is protonated in the same position, therefore stabilizing the neutral blue form. The designation of ‘‘red’’ and ‘‘blue’’ for the anionic and neutral semiquinones (respectively) is associated with the visible absorption spectral features that identify either form (Ghisla and Edmondson, 2001). The structure of the FMN moiety also accounts for the ability of Mm_FprA to accept electrons directly from an organic cofactor (F420H2), contrasting with other FDPs that accept electrons from other redox proteins (typically rubredoxins). The FMN is located on the edge of the flavodoxindomain internal b sheet and, although its dimethyl benzene edge protrudes from the domain surface, it becomes occluded from solvent exposure upon dimerization. Comparing the FMN moiety among structures containing flavin, it is observed that the indole group of a tryptophan residue in Dg_ROO and Mt_FDP (Trp347 in Dg_ROO) is coplanar with the isoalloxazine plane, stacked at 3.5 A˚ distance in a ‘‘sandwich’’-like conformation, shielding the isoalloxazine Si face from the solvent. This residue is almost strictly conserved in the FDP family, being replaced by a glycine in some sequences, such as in Mm_FprA, where no homologous aromatic residue is found at this position. Moreover, the presence of this aromatic ‘‘sandwich’’ above the flavin ring may be related with the visible absorption spectral heterogeneity of the flavin moiety in FDPs (Vicente et al., 2007). The absence of this Trp residue is particularly relevant in the oxidized
16
Joa˜o B. Vicente et al.
structure of Mm_FprA, where the open conformation of the ‘‘switch loop’’ leaves enough space to accommodate parallel docking of the deazaisoalloxazine ring of F420H2 in front of the Si face of FMN (Seedorf et al., 2007), as required to closely approach the C5 of F420H2 and N5 of FMN to allow electron transfer, as observed in various other systems (Pejchal et al., 2005; Warkentin et al., 2001). The presence of the Trp residue in the majority of the FDP sequences (and possibly the same localization above the flavin ring system) precludes most FDPs from accepting electrons directly from organic cofactors (as F420H2). The absence of this Trp residue and its replacement by a Gly residue is found exclusively in methanogens (where F420H2 is an abundant cytoplasmic cofactor) and FDPs from cyanobacteria, which have an extra C-terminal flavin-binding domain, homologous to NAD(P)H: flavin oxidoreductases. It may be envisaged that the flavin from the C-terminal domain in these diflavin FDPs may have the same structural accessibility to dock in front of the flavodoxin-domain FMN prior to electron transfer. Currently, it is not known how this ‘‘sandwich’’ Trp residue affects the mechanism by which most FDPs accept electrons at the flavin moiety from the respective redox protein partners.
5. Conclusion Structural studies on flavodiiron proteins were remarkably important to deepen the knowledge of this protein family. Altogether, the solved crystallographic structures indicated a ‘‘head-to-tail’’ homodimeric structure as the minimal operative unit of this protein family. Each monomer is composed of an N-terminal metallo-b-lactamase-like domain and a C-terminal flavodoxin-like fold. Moreover, crystallographic structures provided the first and unequivocal evidence for the active site of NO/O2 reduction, a nonheme diiron center with histidine and carboxylate residues, embedded in the lactamase-like domain. The flavodoxin-like domain in turn binds a flavin mononucleotide moiety, which is located within one ˚ away from the diiron center. This observation attests the monomer at 35 A functional relevance of the ‘‘head-to-tail’’ homodimeric arrangement, as it brings the FMN cofactor of one monomer in close contact with the diiron active site from the opposing monomer. This quaternary structure thus allows FDPs to accept electrons at the flavin moiety and transfer them rapidly to the diiron center, where NO and/or O2 reduction occurs. Despite some heterogeneities found at the diiron coordination sphere among the solved structures, the accumulated body of structural data failed thus far to provide a definite answer on what determines the specificity of the various FDPs for either substrate. Therefore, more available structures are still required, ideally involving also FDPs with extra C-terminal redox-active domains.
Structural Studies on Flavodiiron Proteins
17
REFERENCES Andersson, J. O., Hirt, R. P., Foster, P. G., and Roger, A. J. (2006). Evolution of four gene families with patchy phylogenetic distributions: Influx of genes into protist genomes. BMC Evol. Biol. 6, 27. Andersson, J. O., Sjogren, A. M., Davis, L. A., Embley, T. M., and Roger, A. J. (2003). Phylogenetic analyses of diplomonad genes reveal frequent lateral gene transfers affecting eukaryotes. Curr. Biol. 13, 94–104. Brunger, A. T., Adams, P. D., Clore, G. M., DeLano, W. L., Gros, P., GrosseKunstleve, R. W., Jiang, J. S., Kuszewski, J., Nilges, M., Pannu, N. S., Read, R. J., Rice, L. M., et al. (1998). Crystallography and NMR system: A new software suite for macromolecular structure determination. Acta Crystallogr. D Biol. Crystallogr. 54, 905–921. Cameron, A. D., Ridderstrom, M., Olin, B., and Mannervik, B. (1999). Crystal structure of human glyoxalase II and its complex with a glutathione thiolester substrate analogue. Structure 7, 1067–1078. Carfi, A., Pares, S., Duee, E., Galleni, M., Duez, C., Frere, J. M., and Dideberg, O. (1995). The 3-D structure of a zinc metallo-beta-lactamase from Bacillus cereus reveals a new type of protein fold. EMBO J. 14, 4914–4921. Chen, L., Liu, M.-Y., LeGall, J., Fareleira, P., Santos, H., and Xavier, A. V. (1993). Rubredoxin oxidase, a new flavo-hemo-protein, is the site of oxygen reduction to water by the ‘‘strict anaerobe’’ Desulfovibrio gigas. Biochem. Biophys. Res. Commun. 193, 100–105. Collaborative Computational Project Number 4 (1994). The CCP4 suite: Programs for protein crystallography. Acta Cryst. D 50, 760–763. De La Fortelle, E., and Bricogne, G. (1997). Maximum-likelihood heavy-atom parameter refinement for multiple isomorphous replacement and multiwavelength anomalous diffraction methods. Methods Enzymol. 276, 472–494. DiDonato, M., Deacon, A. M., Klock, H. E., McMullan, D., and Lesley, S. A. (2004). A scaleable and integrated crystallization pipeline applied to mining the Thermotoga maritima proteome. J. Struct. Funct. Genomics 5, 133–146. Evans, P. (2006). Scaling and assessment of data quality. Acta Crystallogr. D Biol. Crystallogr. 62, 72–82. Fraza˜o, C., Sieker, L., Coelho, R., Morais, J., Pacheco, I., Chen, L., LeGall, J., Dauter, Z., Wilson, K., and Carrondo, M. A. (1999). Crystallization and preliminary diffraction data analysis of both single and pseudo-merohedrally twinned crystals of rubredoxin oxygen oxidoreductase from Desulfovibrio gigas. Acta Crystallogr. D Biol. Crystallogr. 55, 1465–1467. Fraza˜o, C., Silva, G., Gomes, C. M., Matias, P., Coelho, R., Sieker, L., Macedo, S., Liu, M.-Y., Oliveira, S., Teixeira, M., Xavier, A. V., Rodrigues-Pousada, C., et al. (2000). Structure of a dioxygen reduction enzyme from Desulfovibrio gigas. Nat. Struct. Biol. 7, 1041–1045. Gardner, A. M., Helmick, R. A., and Gardner, P. R. (2002). Flavorubredoxin, an inducible catalyst for nitric oxide reduction and detoxification in Escherichia coli. J. Biol. Chem. 277, 8172–8177. Ghisla, S., and Edmondson, D. A. (2001). Flavin coenzymes. In ‘‘Encyclopedia of Life Sciences.’’ Nature Publishing Group. Gomes, C. M., Fraza˜o, C., Xavier, A. V., Legall, J., and Teixeira, M. (2002a). Functional control of the binuclear metal site in the metallo-beta-lactamase-like fold by subtle amino acid replacements. Protein Sci. 11, 707–712. Gomes, C. M., Giuffre, A., Forte, E., Vicente, J. B., Saraiva, L. M., Brunori, M., and Teixeira, M. (2002b). A novel type of nitric-oxide reductase: Escherichia coli flavorubredoxin. J. Biol. Chem. 277, 25273–25276.
18
Joa˜o B. Vicente et al.
Gomes, C. M., Silva, G., Oliveira, R., LeGall, J., Liu, M.-Y., Xavier, A. V., RodriguesPousada, C., and Teixeira, M. (1997). Studies on the redox centers of the terminal oxidase from Desulfovobrio gigas and evidence for its interaction with rubredoxin. J. Biol. Chem. 272, 22502–22508. Hendrickson, W. A., and Ogata, C. M. (1997). Phase determination from multiwavelength anomalous diffraction measurements. Methods Enzymol. 276, 494–523. Jones, T. A., Zou, J. Y., Cowan, S. W., and Kjeldgaard, M. (1991). Improved methods for building protein models in electron density maps and the location of errors in these models. Acta Crystallogr. A47, 110–119. Justino, M. C., Vicente, J. B., Teixeira, M., and Saraiva, L. M. (2005). New genes implicated in the protection of anaerobically grown Escherichia coli against nitric oxide. J. Biol. Chem. 280, 2636–2643. Kabsch, W. (1993). Automatic processing of rotation diffraction data from crystals of initially unknown symmetry and cell constants. J. Appl. Crystallogr. 26, 795–800. Kissinger, C. R., Gehlhaar, D. K., and Fogel, D. B. (1999). Rapid automated molecular replacement by evolutionary search. Acta Crystallogr. D Biol. Crystallogr. 55, 484–491. Leslie, A. G. (2006). The integration of macromolecular diffraction data. Acta Crystallogr. D Biol. Crystallogr. 62, 48–57. Ludwig, M. L., Pattridge, K. A., Metzger, A. L., Dixon, M. M., Eren, M., Feng, Y., and Swenson, R. P. (1997). Control of oxidation-reduction potentials in flavodoxin from Clostridium beijerinckii: The role of conformation changes. Biochemistry 36, 1259–1280. McRee, D. E. (1992). A visual protein crystallographic software system for X11/XView. J. Mol. Graphics 10, 44–46. Murshudov, G. N., Vagin, A. A., and Dodson, E. J. (1997). Refinement of macromolecular structures by the maximum-likelihood method. Acta Crystallogr. D Biol. Crystallogr. 53, 240–255. Otwinowski, Z., and Minor, W. (1997). Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol. 276, 307–326. Pejchal, R., Sargeant, R., and Ludwig, M. L. (2005). Structures of NADH and CH3-H4 folate complexes of Escherichia coli methylenetetrahydrofolate reductase reveal a spartan strategy for a ping-pong reaction. Biochemistry 44, 11447–11457. Rodrigues, R., Vicente, J. B., Felix, R., Oliveira, S., Teixeira, M., and RodriguesPousada, C. (2006). Desulfovibrio gigas flavodiiron protein affords protection against nitrosative stress in vivo. J. Bacteriol. 188, 2745–2751. Sarti, P., Fiori, P. L., Forte, E., Rappelli, P., Teixeira, M., Mastronicola, D., Sanciu, G., Giuffre, A., and Brunori, M. (2004). Trichomonas vaginalis degrades nitric oxide and expresses a flavorubredoxin-type protein: A new pathogenic mechanism? Cell. Mol. Life Sci. 61, 618–623. Seedorf, H., Dreisbach, A., Hedderich, R., Shima, S., and Thauer, R. K. (2004). F420H2 oxidase (FprA) from Methanobrevibacter arboriphilus, a coenzyme F420-dependent enzyme involved in O2 detoxification. Arch. Microbiol. 182, 126–137. Seedorf, H., Hagemeier, C. H., Shima, S., Thauer, R. K., Warketin, E., and Ermler, U. (2007). Structure of coenzyme F420H2 oxidase (FprA): A di-iron flavoprotein from methanogenic Archaea catalyzing the reduction of O2 to H2O. FEBS J. 274, 1588–1599. Sheldrick, G. M. (1990). Phase annealing in SHELX-90: Direct methods for larger structures. Acta Crystallogr. A46, 467–473. Sheldrick, G. M., and Schneider, T. R. (1997). SHELXL: High-resolution refinement. Methods Enzymol. 277, 319–343. Silaghi-Dumitrescu, R., Coulter, E. D., Das, A., Ljungdahl, L. G., Jameson, G. N., Huynh, B. H., and Kurtz, D. M., Jr. (2003). A flavodiiron protein and high molecular weight rubredoxin from Moorella thermoacetica with nitric oxide reductase activity. Biochemistry 42, 2806–2815.
Structural Studies on Flavodiiron Proteins
19
Silaghi-Dumitrescu, R., Kurtz, D. M., Jr., Ljungdahl, L. G., and Lanzilotta, W. N. (2005a). X-ray crystal structures of Moorella thermoacetica FprA: Novel diiron site structure and mechanistic insights into a scavenging nitric oxide reductase. Biochemistry 44, 6492–6501. Silaghi-Dumitrescu, R., Ng, K. Y., Viswanathan, R., and Kurtz, D. M., Jr. (2005b). A flavodiiron protein from Desulfovibrio vulgaris with oxidase and nitric oxide reductase activities: Evidence for an in vivo nitric oxide scavenging function. Biochemistry 44, 3572–3579. Solomon, E. I., Brunold, T. C., Davis, M. I., Kemsley, J. N., Lee, S.-K., Lehnert, N., Neese, F., Skulan, A. J., Yang, Y.-S., and Zhou, J. (2000). Geometric and electronic structure/function correlations in non-heme iron enzymes. Chem. Rev. 100, 235–350. Ullah, J. H., Walsh, T. R., Taylor, I. A., Emery, D. C., Verma, C. S., Gamblin, S. J., and Spencer, J. (1998). The crystal structure of the L1 metallo-beta-lactamase from Stenotrophomonas maltophilia at 1.7 A resolution. J. Mol. Biol. 284, 125–136. Vicente, J. B., Justino, M. C., Gonc¸alves, V. L., Saraiva, L. M., and Teixeira, M. (2007). Biochemical, spectroscopic, and thermodynamic properties of flavodiiron proteins. Methods Enzymol. 437(this volume). Vicente, J. B., and Teixeira, M. (2005). Redox and spectroscopic properties of the Escherichia coli nitric oxide-detoxifying system involving flavorubredoxin and its NADH-oxidizing redox partner. J. Biol. Chem. 280, 34599–34608. Warkentin, E., Mamat, B., Sordel-Klippert, M., Wicke, M., Thauer, R. K., Iwata, M., Iwata, S., Ermler, U., and Shima, S. (2001). Structures of F420H2:NADPþ oxidoreductase with and without its substrates bound. EMBO J. 20, 6561–6569. Wasserfallen, A., Ragettli, S., Jouanneau, Y., and Leisinger, T. (1998). A family of flavoproteins in the domains Archaea and Bacteria. Eur. J. Biochem. 254, 325–332. Winn, M. D., Murshudov, G. N., and Papiz, M. Z. (2003). Macromolecular TLS refinement in REFMAC at moderate resolutions. Methods Enzymol. 374, 300–321. Zang, T. M., Hollman, D. A., Crawford, P. A., Crowder, M. W., and Makaroff, C. A. (2001). Arabidopsis glyoxalase II contains a zinc/iron binuclear metal center that is essential for substrate binding and catalysis. J. Biol. Chem. 276, 4788–4795.
This page intentionally left blank
C H A P T E R
T W O
Biochemical, Spectroscopic, and Thermodynamic Properties of Flavodiiron Proteins Joa˜o B. Vicente, Marta C. Justino, Vera L. Gonc¸alves, Lı´gia M. Saraiva, and Miguel Teixeira Contents 1. Introduction 2. Cloning of Genes Encoding Flavodiiron Proteins and Their Truncated Domains 3. Production and Purification of Recombinant Flavodiiron Proteins 4. Biochemical Characterization of Flavodiiron Proteins 5. Spectroscopic Properties 6. Redox Properties 7. Conclusions 7.1. Functional properties Acknowledgments References
22 24 25 26 29 32 37 39 42 42
Abstract The flavodiiron proteins (FDPs), present in Archaea, Bacteria, and some protozoan pathogens (mostly anaerobes or microaerophiles), have been proposed to afford protection to microbes against nitric oxide and/or oxygen (toxic for anaerobes). The structural prototype of this protein family is a homodimer assembled in a ‘‘head-to-tail’’ configuration, with each monomer being composed of two domains: an N-terminal metallo-b-lactamase module harboring a nonheme diiron center (active site of NO/O2 reduction) and a C-terminal flavodoxin module, where a flavin mononucleotide moiety is embedded. Several FDPs bear C-terminal extra domains, which influence the composition of the respective electron transfer chains that couple NAD(P)H oxidation to NO/O2 reduction. Herein are described methodologies employed to successfully produce, isolate, and characterize fully operative recombinant flavodiiron proteins. Spectroscopic techniques, namely absorption (visible and near-ultraviolet) and Instituto de Tecnologia Quı´mica e Biolo´gica, Universidade Nova de Lisboa, Oeiras, Portugal Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37002-X
#
2008 Elsevier Inc. All rights reserved.
21
22
Joa˜o B. Vicente et al.
electron paramagnetic resonance spectroscopies, allowed redox-sensitive spectral fingerprints to be obtained, used further in the functional characterization of isolated flavodiiron proteins. Altogether, these studies on pure proteins contribute to understanding the molecular determinants that govern the in vivo function of the FDPs.
1. Introduction The first report on a flavodiiron protein (FDP) focused on Desulfovibrio gigas rubredoxin:oxygen oxidoreductase (Dg_ROO), the terminal component of a soluble electron transfer chain proposed to be involved in oxygen detoxification, affording this (then considered) ‘‘strict’’ anaerobe protection from an otherwise toxic dioxygen (Chen et al., 1993b). Dg_ROO, a flavinbinding homodimer of 43-kDa monomers, was proposed to fully reduce oxygen to water, using electron equivalents from NADH, shuttled by rubredoxin and a NADH:rubredoxin oxidoreductase (Chen et al., 1993a; Gomes et al., 1997). The flurry of complete genome sequences led to the discovery of several Dg_ROO homologues widespread in Bacteria and Archaea and to establishment of the family of A-type flavoproteins (the former designation of flavodiiron proteins) (Wasserfallen et al., 1998). It was proposed that there is a common sequence core of about 400 amino acids, where a putative flavodoxin-like domain could be identified at the C terminus, and it was noted that some members of the protein family had extra C-terminal extensions. These extensions were identified as possible redox active domains, namely a rubredoxin domain in the Escherichia coli protein and a NAD(P)H:flavin oxidoreductase domain in the Synechocystis one. It was not until the crystallographic structure of Dg_ROO was solved that further insights into its functional properties were attained (Fraza˜o et al., 2000). This structure elucidated that the previously proposed core of this protein family is indeed composed of an N-terminal b-lactamase-like domain fused to the flavin mononucleotide (FMN)-binding flavodoxin domain and revealed the active site of oxygen reduction: a nonheme diiron center in the lactamase fold, with carboxylate and histidine residues in its coordination sphere. It is worth noting that the structure revealed that a ‘‘head-to-tail’’ homodimeric quaternary arrangement is required to place the FMN cofactor of one monomer in close contact with the diiron center from the other monomer, allowing otherwise impaired electron transfer (the two cofactors are 25 A˚ apart in each monomer)(Vicente et al., 2007a). A survey of available FDP sequences suggested four structural classes for this protein family [adding one class to a previous classification (Saraiva et al., 2004)], accounting for the C-terminal extensions (Fig. 2.1), whose nature reflects itself in the composition of the electron transfer chains that couple
23
Properties of Flavodiiron Proteins
Class A
FMN Fe-Fe
Class B
Class C
Class D
FAD
Fe-S
FMN
Flv
FMN Fe-Fe
Fe-S
FMN Fe-Fe
Fe-Fe
Figure 2.1 Modular arrangements in the flavodiiron protein family and corresponding structural classes. FMN, flavodoxin-like module, binding FMN; Fe-Fe, metallo-b-lactamase module, harboring the nonheme diiron active site; Fe-S, rubredoxin-like module, harboring a Fe-Cys4 center; Flv, NAD(P)H:flavin oxidoreductase module, binding FAD or FMN; FAD, predicted NAD(P)H:rubredoxin oxidoreductase module.
NAD(P)H or F420H2 oxidation to nitric oxide (NO) or O2 reduction. Class A FDPs are the simplest, consisting solely of the bidomain structural core (400 residues), and represent the majority of the found sequences. Class B FDPs (480 residues) bear a C-terminal rubredoxin domain and are restricted, so far, to enterobacteria. Class C FDPs (600 residues), also found so far only in cyanobacteria, have a NAD(P)H:flavin oxidoreductase C-terminal domain, and often there are multiple genes encoding these FDPs within the same organism. The newly proposed class D (900 residues) comprises FDPs where two extra C-terminal domains are fused: a rubredoxin domain and a NADH:rubredoxin oxidoreductase domain (homologous to the cognate reductase of class B FDPs). Class D FDPs were found in the genome sequences of some Clostridiales and of the protozoan pathogen Trichomonas vaginalis. Phylogenetic analyses revealed two interesting observations: (i) FDPs bearing C-terminal extensions cluster together according to their class (Saraiva et al., 2004) and (ii) genes encoding FDPs are prone to be transferred via lateral gene transfer among coexisting organisms (Andersson et al., 2003). This observation accounts for the finding of FDP-encoding genes in pathogenic protozoa, so far the only known eukaryotic FDPs. It is envisaged that the complexity of the modular arrangement of FDPs contrasts with the number of components of the corresponding electron transfer chains, i.e., class C and D FDPs should accomplish coupling of NAD(P)H oxidation to substrate reduction within the same polypeptide chain. Class B FDPs require one extra redox protein, and class A may require as many as two more redox partners to accomplish the same. This idea has been challenged only recently by a class A FDP (from a methanogenic source) that oxidizes F420H2 directly (Seedorf et al., 2004), an abundant redox cofactor in methanogenic organisms, and thus dispenses the involvement of other redox proteins.
24
Joa˜o B. Vicente et al.
The idea of flavodiiron proteins as oxygen reductases came to a halt with molecular genetics studies on the E. coli flavodiiron protein (a class B FDP named flavorubredoxin because of its C-terminal rubredoxin domain) that proposed a role for this protein in NO detoxification. It was demonstrated that expression of the norV gene—encoding flavorubredoxin—was induced by NO and that an E. coli norV mutant strain was more sensitive to NO than the wild-type strain (Gardner et al., 2002; Justino et al., 2005b), with deleterious effects to NO-sensitive metabolic enzymes and affecting cell survival (Gardner et al., 2002). A role in anaerobic NO detoxification was thus proposed for flavorubredoxin, acting as an NO reductase, an activity that was confirmed further in vitro (Gomes et al., 2002). Flavodiiron proteins are presently considered a prominent family of NO-detoxifying enzymes, in the line of flavohemoglobin, although some members of the protein family retain a preference for oxygen as their substrate (Rodrigues et al., 2006; Seedorf et al., 2007). Research efforts have been employed to clarify the ambiguity of the possible roles for flavodiiron proteins. A thorough biochemical characterization of each studied member (and the corresponding electron transfer chains) is essential to understanding the molecular basis for the substrate preference (NO vs O2). In parallel, molecular genetics studies provide clues to understanding the function and relative role of each FDP in (i) NO detoxification (e.g., as a subversive mechanism of pathogens to counteract the host immune response) and/or (ii) O2 detoxification in anaerobic organisms (to allow survival of transient exposure to toxic environments).
2. Cloning of Genes Encoding Flavodiiron Proteins and Their Truncated Domains The production of recombinant flavodiiron proteins from various microbial sources has been successfully achieved by overexpression in E. coli, with T7 promoter-based expression vectors (Gomes et al., 2000, 2002; Rodrigues et al., 2006; Seedorf et al., 2004; Silaghi-Dumitrescu et al., 2003, 2005; Vicente et al., 2002; Wasserfallen et al., 1998). In our laboratory, flavodiiron proteins from classes B and C have been isolated as recombinant proteins overexpressed in E. coli: the E. coli flavorubredoxin (FlRd) and its truncated rubredoxin and flavodiiron domains, the Synechocystis sp. PCC6803 SsATF573 (the original designation for the 573 amino acid FDP from this organism, encoded by gene sll0550), and its C-terminal domain (Gomes et al., 2000, 2002; Vicente et al., 2002). The coding regions were amplified by polymerase chain reaction from genomic DNA, using primers containing restriction sites that allow cloning into the T7 expression vectors pET24aþ or pT7–7 (for the rubredoxin
Properties of Flavodiiron Proteins
25
domain, Rd domain). Cloning of DNA fragments encoding the C-terminal domains required the introduction of a NdeI site in the sense primers that changed to initiation codon (ATG) the codons of residue 422 of E. coli FlRd (Gomes et al., 2002) and residue 402 of Synechocystis SsATF573. Sequencing of the recombinant plasmids confirmed the correct nucleotide sequences (Vicente et al., 2002).
3. Production and Purification of Recombinant Flavodiiron Proteins Overexpression of E. coli flavorubredoxin (and its truncated domains) (Gomes et al., 2000) and Synechocystis SsATF573 (and the C-terminal domain) (Vicente et al., 2002) is performed in BL21-Gold(DE3) cells (Stratagene) under conditions that have been progressively optimized. Initially, Luria-Bertani (LB) broth (supplemented with 10 mM ferrous sulfate) was used to attempt the overexpression of FDPs. However, an improvement of iron and flavin incorporation was achieved by decreasing the rate of protein synthesis. This was attained by changing the growth medium to minimal medium M9 (Gomes et al., 2002; Silaghi-Dumitrescu et al., 2003), reducing the air chamber, and decreasing the growth temperatures from 37 to 28 C. Under optimized conditions, freshly transformed cells are grown in M9 minimal medium with 10 mM glucose (Ausubel et al., 1995) supplemented with 10 mM FeSO47H2O, in flasks filled to 70% of the volume, at 28 C and 130 rpm. Induction of expression is made with 100 mM isopropyl-1-thio-b-D-galactopyranoside (IPTG) when the cultures reach OD600 ¼ 0.3–0.4, and the cells are harvested after 7 h by centrifugation (11,000g, 10 min, 4 C). Cells resuspended in 10 mM Tris-HCl, pH 7.6, are disrupted in a French press cell at 130 MPa, followed by a 2-h ultracentrifugation (100,000g, 4 C) to remove cell debris. The soluble extracts are dialyzed against 10 mM Tris-HCl, pH 7.6, containing 18% (v/v) glycerol (buffer A). Complementing the buffers with glycerol increases the stability of the enzymes, preventing the loss of flavin moieties throughout the purification. In the purifications of intact flavorubredoxin, 500 mM of the protease inhibitor phenylmethanesulfonyl fluoride is added to all buffers to prevent peptidic breakage between the structural modules of FlRd (Vicente and Teixeira, 2005). All purification steps are done at 4 C. Dialyzed soluble extracts are applied onto a Q-Sepharose column (Amersham) equilibrated previously with buffer A and, by applying a gradient up to 1 M NaCl, proteins are eluted at 400–450 mM NaCl (Gomes et al., 2000, 2002; Vicente et al., 2002). After desalting, fractions are introduced into a Fractogel EMD TMAE column (Merck), eluted at 250 mM NaCl, concentrated, and further applied onto a gel filtration column (Superdex S-75 or S-200, both from Amersham)
26
Joa˜o B. Vicente et al.
equilibrated with buffer A containing 150 mM NaCl. Regarding purification of the C-terminal domain of Synechocystis SsATF573, the protein is purified from the soluble extract, in two steps: a Q-Sepharose fast flow column (Amersham) equilibrated in 20 mM KP buffer at pH 6 (buffer B) and a SP-Sepharose column (Amersham). The truncated form of SsATF573 is eluted with 200 mM KCl (Vicente et al., 2002). Protein purity is evaluated throughout the purification steps by SDS-PAGE (Garfin, 1990). The purifications of recombinant FDPs from Moorella thermoacetica (Silaghi-Dumitrescu et al., 2003), Desulfovibrio vulgaris (Silaghi-Dumitrescu et al., 2005), and D. gigas (Rodrigues et al., 2006) are conducted in similar ways, although with minor differences on the expression conditions. The production and purification of Methanobrevibacter marburgensis FprA are significantly different from other FDPs, namely the fact that the protein is only successfully isolated under anaerobic conditions. M. marburgensis FprA is overexpressed in E. coli Rosetta(DE3)pRare cells are induced at OD600 0.8 by the addition of 1 mM IPTG. Harvested cells are disrupted by ultrasonication and heated at 60 for 20 min. FprA is isolated from the soluble extract (obtained after a 150,000g ultracentrifugation) in one purification step, using a DEAE-Sepharose fast flow column equilibrated with 50 mM Tris-HCl, pH 7.6. The protein is recovered in the 400 mM NaCl fraction (Seedorf et al., 2004).
4. Biochemical Characterization of Flavodiiron Proteins Flavodiiron proteins from three (of the aforementioned four) classes have been studied, the majority of which belong to class A (properties summarized in Table 2.1). The monomeric molecular masses are determined by SDS-PAGE (Garfin, 1990), and the measured molecular masses are in accordance with the expected values inferred from the peptide sequences, namely 43–48 kDa per monomer for class A FDPs (400 residues), 54 kDa for class B FDPs (480 residues), and 70 kDa for class C FDPs (600 residues). The quaternary structure of isolated FDPs is measured by analytical gel permeation chromatography using the appropriate molecular mass standards. FDPs alternate between homodimers and homotetramers, satisfying the prerequisite of a dimer as the minimal functional unit, to allow proximity between the FMN from one monomer and the diiron center from the other monomer (Fraza˜o et al., 2000; Seedorf et al., 2007; Silaghi-Dumitrescu et al., 2005). The purified proteins have been assayed for their cofactor content, namely in terms of flavin and iron incorporation. For the FDPs (and
Table 2.1
Physical-chemical properties and cofactor content of flavodiiron proteins Monomer molecular mass*
Quaternary structure
Cofactor content{ per monomer
43 kDa (44.8)
Homodimer
2 Fe (XRC) 1 FMN (XRC)
399
45 kDa (44.3)
Homodimer
410
n. d. (47.1) 45 kDa (45.1)
Homodimer
1.9 0.5 Fe (PAEA) 0.85 0.1 FMN 2 Fe (XRC)
Protein
Microorganism
a.a length
Class A Rubredoxin:oxygen oxidoreductase (ROO)
Desulfovibrio gigas
402
Flavodiiron protein
Moorella thermoacetica
Flavoprotein (Tm0755) Flavodiiron protein
Thermotoga maritima Desulfovibrio vulgaris
Flavoprotein A (FprA) Flavoprotein A (FprA) Flavoprotein A (FprA)
Methanobrevibacter arboriphilus Methanothermobacter marburgensis Rhodobacter capsulatus
402
— 404 420
45 kDa (46.1) 43 kDa (45.3) 48 kDa (46.2)
n. d.
1.8 0.1 Fe (PAEA) 0.8 0.1 FMN 2 Fe 1 FMN
Homotetramer
2 Fe 1 FMN
Homodimer
0.9 FMN (AE-HPLC)
Homodimer
Ref.
(Chen et al., 1993; Frazao et al., 2000) (SilaghiDumitrescu et al., 2003)
(SilaghiDumitrescu et al., 2005) (Seedorf et al., 2004) (Seedorf et al., 2004) (Jouanneau et al., 2000; Wasserfallen et al., 1998) (continued)
Table 2.1 (continued) Monomer molecular mass*
Quaternary structure
Cofactor content{ per monomer
1.3 FMN (AE-HPLC) 1 mol Fe/mol FMN (N/C) 0.7 FMN (AE-HPLC)
(Nolling et al., 1995)
Ref.
Protein
Microorganism
a.a length
Flavoprotein A (FprA)
Methanobacterium thermoautotrophicum strain DH
409
45 kDa (46.0)
Homodimer
Flavoprotein A (FprA)
Methanobacterium thermoautotrophicum Marburg
404
43 kDa (45.7)
Homotetramer
Escherichia coli
479
54 kDa (54.2)
Homotetramer
2.9 0.5 Fe (TPTZ) 0.8 0.2 FMN (AE-UVS)
(Frazao et al., 2000; Wasserfallen et al., 1998)
Synechocystis
573
70 kDa (63.5)
Homodimer
1.9 Fe (TPTZ) 0.8 FMN (AE-HPLC)
(Gomes et al., 2002; Wasserfallen et al., 1998)
Class B Flavorubredoxin (FlRd)
Class C SsATF573
(Wasserfallen et al., 1995)
* between brackets, molecular mass estimated from the aminoacid (a.a.) sequence. { between brackets, experimental methodology by which the cofactor was identified and quantified: XRC, X-ray crystallography; PAEA, plasma atomic emission analysis; AE-UVS, acid extraction plus visible spectroscopy; AE-HPLC, acid extraction followed by HPLC analysis; N/C – colorimetric method, using neocuproine or ferrozine. n.d. not determined.
Properties of Flavodiiron Proteins
29
truncated domains) of E. coli and Synechocystis, the protein concentrations are measured by the 2-bicinchoninic acid protein assay (Pierce) (Walker, 1994). The iron content is determined by the 2,4,6-tripyridyl-1,3,5-triazine method (Fischer and Price, 1964). Flavin quantification (adapted from Susin et al., 1993) is performed by acid extraction with TCA (10%) followed by centrifugation and supernatant neutralization with 1 M NH4CH3COO, pH 7. The nature of the flavin cofactors (FMN or FAD) is determined, when needed (Wasserfallen et al., 1998), by reversed-phase chromatography using a Nucleosil 100–5 C18 column equilibrated with 10 mM ammonium formate, pH 6.4 (containing 12% methanol), and performing a three-step gradient of increasing methanol concentration. The appropriate commercial flavin standards (FAD and FMN from Fluka) are treated and measured identically to flavins extracted from the protein samples. The extracted flavins are quantified spectrophotometrically using the following molar absorption coefficients: EFMN (l445) ¼ 12,200 M1 cm1; EFAD (l450) ¼ 11,300 M1 cm1; and EFMN þ FAD (l447) ¼ 11,750 M1 cm1 (Sober and Harte, 1968). The cofactor content of studied FDPs (Table 2.1) yields 1–2 Fe and 0.7–1 FMN per monomer. E. coli flavorubredoxin binds instead three iron ions per monomer, one in the rubredoxin domain and two in the diiron site. By replacing Fe2þ with Zn2þ in the growth medium of E. coli cells overexpressing M. thermoacetica FDP (Silaghi-Dumitrescu et al., 2003), the isolated FDP comes with a binuclear zinc site in place of the diiron site. This promiscuity is explained by equivalently high affinities of the center for Fe and Zn (Schilling et al., 2005).
5. Spectroscopic Properties To probe the functional properties of isolated flavodiiron proteins, spectroscopic methods proved to be essential, namely in characterization of the redox-active cofactors. Whereas visible spectroscopy was used mainly to characterize the flavins, electron paramagnetic resonance (EPR) spectroscopy allowed characterization of the diiron center. Visible and near-ultraviolet absorption spectra of as-isolated flavodiiron proteins are mostly dominated by the contribution of their flavin moieties. Nonheme diiron centers (Solomon et al., 2000) have much lower extinction coefficients than flavins (free or protein bound) (Ghisla and Edmonson, 2001), and therefore spectra of class A (Silaghi-Dumitrescu et al., 2003) and class C (Vicente et al., 2002) FDPs (which have only flavin and diiron cofactors) have features almost solely attributable to the flavin moieties (Fig. 2.2A and B). It is noteworthy that visible spectra of FDPs are slightly heterogeneous among different members of the protein family, in the sense that the band centered at 450 nm is broad and smooth in some cases
30
Joa˜o B. Vicente et al.
A FMN Fe-Fe
Class A
B Flv
FMN Fe-Fe
Classes A and C
C Fe-S
FMN Fe-Fe
Class B
300
400
500 600 Wavelength (nm)
700
Figure 2.2 Visible spectra of flavodiiron proteins. (A) Flavodiiron domain of Escherichia coli flavorubredoxin (i.e., with the rubredoxin domain truncated); (B) flavodiiron protein from Synechocystis sp. PCC6803, named SsATF573; and (C) E. coli flavorubredoxin. All spectra in 20 mM Tris-HCl,18% glycerol, pH 7.6, at 25.
( Jouanneau et al., 2000; Silaghi-Dumitrescu et al., 2003; Wasserfallen et al., 1998) and has two shoulders in others (Nolling et al., 1995; Vicente et al., 2002; Wasserfallen et al., 1995). This band is commonly assigned to charge transfer transitions within the isoalloxazine core, from the xylene ring to the pyrimidine ring. To understand this spectral heterogeneity, an inspection of the flavin pocket in the Dg_ROO structure was undertaken and compared with structural models generated with that structure as the template. In the structure of Dg_ROO, a tryptophan residue is coplanar with the FMN isoalloxazine ring (Trp347 in Dg_ROO). This Trp residue is conserved in FDPs where the same broad spectrum (Fig. 2.2A) is observed and appears in the same position in the modeled structures of other FDPs (Saraiva et al., 2004). However, in FDPs where this Trp is lacking, the spectral band centered at 450 nm has two shoulders (Fig. 2.2B). Therefore, it has been
Properties of Flavodiiron Proteins
31
proposed that this Trp residue may account for the spectral heterogeneity (Saraiva et al., 2004) by interacting with the FMN moiety (Vicente et al., 2008a). On top of the flavin absorption spectrum, the E. coli flavorubredoxin (class B FDP) has the contribution of the [Fe-Cys4] center from the rubredoxin domain (Fig. 2.2C) (Gomes et al., 2000; Vicente and Teixeira, 2005). Although the spectrum of FlRd overlaps in almost the entire visible region, above 560 nm, the observed broad band is almost exclusively because of the rubredoxin domain. This observation is of great value, namely in deconvoluting the functional behavior of the different cofactors. Electron paramagnetic resonance spectroscopy has proved to be a valuable tool in characterizing the cofactors of flavodiiron proteins, providing in fact the first direct spectroscopic evidence for the diiron center for a member of this protein family. Initially, EPR was used to characterize the flavin cofactor in D. gigas ROO (Gomes et al., 1997), where a signal at g 2.0 obtained under reductive conditions was attributed to the one electron-reduced semiquinone state of the flavin, proposed to correspond to the red anionic radical, based on the 1.6-mT line width (which concurred with visible spectroscopy data) (Gomes et al., 1997). Electron paramagnetic resonance spectroscopy is essential in studying the diiron site, which has very low absorptivity in the visible region. In oxidized states, only FDPs containing a rubredoxin core are EPR active, with the characteristic g 4.3 resonance typical of high-spin (S ¼ 5/2) ferric iron (Gomes et al., 2000; Vicente and Teixeira, 2005). Upon reduction, because of the spin change to S ¼ 2, the rubredoxin resonance vanishes. The diiron center is EPR silent in the oxidized state, as the two ferric ions are coupled antiferromagnetically [as confirmed by Mo¨ssbauer spectroscopy for the M. thermoacetica FDP (Silaghi-Dumitrescu et al., 2003)]. For this reason, the diiron center is only clearly detected by EPR spectroscopy in its one electron-reduced, mixed-valence (FeIII-FeII) state, displaying a rhombic signal with g values at g < 2.0 (Fig. 2.3), which has its maximal intensity at 7K (Vicente and Teixeira, 2005). An interesting observation is that obtained spectra differed in their shape and g values according to the way by which the mixed-valence state was obtained, i.e., in the presence (line 1 in Fig. 2.3) or absence (line 2 in Fig. 2.3) of redox mediators (in both cases, reduction was achieved by the addition of sodium dithionite). Nevertheless, the relaxation properties do not appear to be affected by the different shape, as their corresponding temperature dependences are practically identical (not shown). Full reduction of the diiron center to the FeII-FeII state results in the disappearance of this signal, leaving as sole EPR evidence for the diferrous state a g 11 signal in parallel-mode EPR, indicating an S ¼ 4 spin state. Hence, the achievement of a spectroscopic signature for the diiron center in E. coli FlRd allowed characterization of its thermodynamic
32
Joa˜o B. Vicente et al.
1.95
Fe-Fe
1.80
Intensity (A.U.)
1.74
1
1.93 1.88 2
1.82
300
350
400
450
Magnetic field (mT)
Figure 2.3 EPR spectra of the mixed-valence nonheme diiron center in flavorubredoxin. EPR spectra of the flavodiiron structural core (FDP-domain) of Escherichia coli flavorubredoxin, obtained in the course of a redox titration (line1) and by mild chemical reduction (line 2) with sodium dithionite. Spectra focus on the g <2 region, where mixed-valence nonheme diiron centers have known EPR signatures.The FDP-domain (250 mM) was titrated at 25 C, in 50 mM Tris-HCl,18% glycerol, pH 7.5. Arrows indicate g values assigned to each signal. Spectra collected at 7K; microwave power: 2.4 mW; microwave frequency: 9.64 GHz; modulation amplitude:1 mT.
properties (see later), and it is envisaged that the same methodology can be applied to study the diiron centers in other proteins of this family.
6. Redox Properties The thermodynamic properties of FDPs have been determined essentially for the FMN moiety and only for a few cases. The redox titration of D. gigas ROO, followed by visible spectroscopy, yielded reduction potentials of 0 15 mV for the FMNox/FMNsq step and –130 15 mV for the FMNsq/ FMNred step (Gomes et al., 1997). For M. thermoacetica FprA (Mt_FDP), the determined reduction potentials were –117 mV for the FMNox/FMNsq step and –220 mV for the FMNsq/FMNred step (Silaghi-Dumitrescu et al., 2003). In both cases, the observed semiquinone was of the red anionic type, as judged by the corresponding spectral features (Gomes et al., 1997; Silaghi-Dumitrescu et al., 2003). Concerning E. coli flavorubredoxin, the visible absorption spectrum (Fig. 2.2C) comprises features of both FMN and rubredoxin cofactors (Gomes et al., 2000), with overlapping features, which hamper the deconvolution of their individual redox properties. Herein is described a methodology that allows identification of the reduction potentials of each
Properties of Flavodiiron Proteins
33
cofactor in E. coli FlRd. The experimental approach combined potentiometric methods with visible and EPR spectroscopies and relied significantly on the use of truncated proteins for the separate characterization of each domain, thus allowing the deconvolution of superimposing spectral features. This methodology was further employed to analyze modulation of the redox properties of FlRd that occurs upon interaction with its cognate reductase partner (FlRd-reductase) (Vicente et al., 2007b,c). Redox titrations are performed anaerobically at 25 C, with the isolated E. coli FlRd whole protein or truncated domains (Rd domain and FDP domain) and with a stoichiometric (1:1) mixture of FlRd (or its Rd domain) with the FlRd-reductase partner. Reduction is attained by the stepwise addition of sodium dithionite (250 mM Tris-HCl, pH 8) or NADH, the physiological electron donor of FlRd-reductase. Anaerobic conditions were maintained by continuously degassing on surface the titration buffer (50 mM Tris-HCl pH 7.6, 18% glycerol) with argon and by the addition of oxygen scavengers (glucose, glucose oxidase and catalase). The redox mediators used are methylene blue (Eo0 ¼ 11 mV), indigo tetrasulfonate (Eo0 ¼ –30 mV), indigo trisulfonate (Eo0 ¼ –70 mV), indigo disulfate (Eo0 ¼ –82 mV), indigo disulfate anthraquinone 2,7-disulfonate (Eo0 ¼ –182 mV), safranine (Eo0 ¼ – 280 mV), neutral red (Eo0 ¼ –325 mV), benzyl viologen (Eo0 ¼ –359 mV), and methyl viologen (Eo0 ¼ –446 mV). The concentration of redox mediators ranges between 30 and 80 mM in the EPR-monitored titrations and between 0.25 and 0.5 mM in the visible monitored ones. A silver/silver chloride electrode is used, calibrated with a saturated quinhydrone solution at pH 7, and the reduction potentials are quoted against the standard hydrogen electrode (Gomes et al., 1997; Vicente and Teixeira, 2005). Spectral deconvolution and experimental data analysis is performed using MATLAB (Mathworks, South Natick, MA) for Windows. Analysis of titration of intact flavorubredoxin (followed by visible spectroscopy) is initiated, taking into account that the absorbance changes at 560 nm are almost exclusively attributable to the Fe-Cys4 center in the rubredoxin module (Vicente and Teixeira, 2005). Plotting the absorbance changes at 560 nm as a function of the reduction potential and fitting data with a Nernst equation for a one-electron transition yield a reduction potential of –123 15 mV. Since the redox titration of the truncated Rddomain (following by visible spectroscopy the bleaching at 484 nm) revealed an identical reduction potential to the Fe-Cys4 centre in the intact protein, it was possible to subtract the spectra of the Rd-domain titration (Fig. 2.4C) to the ones of the intact FlRd titration, where the corresponding experimental redox values match. The subtraction procedure yields a matrix comprising solely spectral features of the flavodiiron core of FlRd, which is clearly dominated by the FMN moiety (see Fig. 2.4D). Spectra in Fig. 2.4D show formation of the red semiquinone upon one-electron reduction of the FMN, characterized by the decrease in the absorbance at 450 nm accompanied by
34
Joa˜o B. Vicente et al.
B 0.5
0.8
Fe-S FMN Fe-Fe
0.6
Absorbance
Absorbance
A 1.0
FAD
0.4 0.2
0.3 FAD
0.2 0.1
0.0 400
500 600 l (nm)
0.0
700
400
C 0.3
500 l (nm)
600
700
D 0.3 Fe-S
Absorbance
Absorbance
0.4
0.2
0.1 0.0 400
500 l (nm)
600
700
FMN Fe-Fe 0.2 0.1 0.0 300
400
500
600
l (nm)
Figure 2.4 Redox properties of Escherichia coli flavorubredoxin and its partner NADH: flavorubredoxin oxidoreductase. Flavorubredoxin and its cognate reductase titrated in a stoichiometric mixture to probe a possible modulation of the redox properties upon interaction of the two partner proteins. An elaborate spectral deconvolution (described in the text) allowed us to isolate the redox behavior of each cofactor (in both FlRd and the reductase). (A) Absolute absorption changes in visible spectra of a stoichiometric mixture of FlRd and FlRd-Red (both at 20 mM, in 50 mM Tris-HCl, 18% glycerol, pH 7.5) titrated with NADH, at 25 C; (B) matrix comprising the optical contribution of the FlRd-reductase to the titration of the mixture (A); (C) matrix of the spectral contribution of the Fe-Cys4 center in the rubredoxin domain; (D) spectral progression of the flavodiiron domain in FlRd, obtained by subtraction of the optical contributions of the other cofactors (B and C) to the overall changes (A); block arrows depict progression of the absorbance changes, i.e., upon reduction absorbance at 450 nm decreases, whereas absorbance at 390 nm initially increases (with the formation of the one electronreduced flavin) and then decreases upon full reduction of FMN.
an increase at 390 nm, and its further disappearance resulting from full reduction to hydroquinone, i.e., the two-electron-reduced flavin. This observation is confirmed by the bell-shaped curve of data corresponding to the difference in absorption at 390 and 458 nm as a function of the redox potential, which was fitted with reduction potentials of –40 15 mV for the FMNox/FMNsq step and –130 15 mV for the FMNsq/FMNred step. These values are similar to the aforementioned reduction potentials measured for D. gigas ROO (Gomes et al., 1997) and are each approximately 80 mV higher than those determined for M. thermoacetica FDP (Silaghi-Dumitrescu et al., 2003). As stated earlier, EPR spectroscopy proved to be essential to probe the
Properties of Flavodiiron Proteins
35
redox properties of the diiron center: data obtained by monitoring this EPRactive species (the mixed valence FeIII-FeII form) yielded bell-shaped curves, corresponding to its appearance (one electron reduction of fully oxidized FeIII-FeIII to FeIII-FeII) and subsequent disappearance (one electron reduction of FeIII-FeII to the fully reduced FeII-FeII state). Data were fitted as intermediate species of two consecutive one-electron step processes with potentials for isolated FlRd of –20 20 mV for the FeIII-FeIII/FeIII-FeII step and –90 20 mV for the FeIII-FeII/FeII-FeII step. Equivalent data for the truncated form of FlRd consisting of its flavodiiron core (FDP-domain) yielded slightly higher redox potentials (0 20 and –50 20 mV) (Vicente and Teixeira, 2005). With the knowledge of the reduction potentials for all cofactors in isolated FlRd, studies were undertaken to evaluate the possible changes in its redox properties upon formation of an electron transfer (eT) complex with its partner, FlRd-reductase. This was accomplished by titrating both proteins, combined in stoichiometric amounts, and using NADH as the reducing agent. FlRd (or the truncated Rd domain) does not accept electrons directly from NADH, so its reduction is exclusively accomplished by reduced FlRd-reductase. In the titration followed by visible spectroscopy, data comprise absorption features from three almost overlapping redox centers (see Fig. 2.4A): the FAD from FlRd-reductase, the FMN from the flavodoxin module, and the [Fe-Cys4] center from the Rd module. Because the reduction potentials of FAD in FlRd-reductase are significantly lower than those of the FlRd centers (depicted in Fig. 2.5) (Vicente and Teixeira, 2005), deconvolution of FlRd-reductase data is immediate. The reduction potentials used to fit the data with two consecutive one-electron Nernst curves (FADox/FADsq: –250 15 mV and FADsq/FADred: –220 15 mV) reflect that reduction of FlRd-reductase proceeds ‘‘macroscopically’’ as a two-electron process. Using the reduction potentials for FlRdreductase and spectra of its oxidized and reduced forms, it was possible to create a matrix of the FlRd-reductase spectral contribution (see Fig. 2.4B) to be subtracted from the titration of the mixture in the following manner: for each experimental potential, a fraction of oxidized and reduced FlRdRed was assigned and concomitantly a spectrum with the contribution of FlRd-reductase to the overall spectrum of the mixture titration. After subtracting the FlRd-reductase matrix, a matrix was obtained consisting solely of FlRd spectra in the course of the titration, which were treated in the same manner as the data obtained for the titration of isolated FlRd (described earlier). Data for the Rd domain were fitted with a one-electron Nernst curve with a reduction potential of –65 15 mV (see Fig. 2.5), which is upshifted with respect to the isolated FlRd titration. Because this reduction potential was identical to that measured in another titration of the truncated Rd domain in a stoichiometric mixture with FlRd-reductase, the analysis proceeded to deconvolution of the reduction potentials for the
36
Joa˜o B. Vicente et al.
1.0
Normalized Δ Abs
0.8 FMN
0.6 FAD
Fe-Fe
0.4
Fe-S
0.2 0.0 −300
−200
−100 E0 (mV)
0
100
Figure 2.5 Titration curves for the redox cofactors of Escherichia coli flavorubredoxin and its reductase partner. The best fits to experimental data (described throughout the text) are represented for each of the redox cofactors in the two interacting partners. FAD, flavin cofactor in FlRd-reductase, ^220 mV for the FADox/FADsq step and ^260 mV for the FADsq/FADred; FMN, flavin mononucleotide bound to the flavodoxin module, ^ 40 mV for the FMNox/FMNsq step and ^260 mV for the FMNsq/FMNred; Fe-S, Fe-Cys4 center in the rubredoxin module, ^65 mV; Fe-Fe, nonheme diiron center in the metallob-lactamase domain (measured by combining potentiometry with EPR spectroscopy), þ 20 mVfor the FeIII-FeIII/FeIII-FeII step and ^50 mVfor the FeIII-FeII/FeII-FeII step.
FMN in FlRd-reductase/FlRd titration. This was achieved by subtracting a matrix with the contribution of the Rd domain (see Fig. 2.5C) from the one containing FlRd data (i.e., obtained after subtraction of the FlRd-reductase contribution). The isolated Rd-domain matrix was in turn obtained by subtracting the FlRd-reductase contribution from data obtained for titration of the truncated Rd domain in the presence of FlRd-reductase, as described earlier for the FlRd-reductase/FlRd titration. By subtracting the Rddomain matrix from the FlRd matrix (both obtained after subtracting the FlRd-reductase component out of their mixed titrations), a matrix was obtained comprising solely the flavodiiron core spectral features dominated by the FMN moiety (see Fig. 2.4D), identical to the one observed for the isolated FlRd titration. It was then possible to fit potentials for the FMN one-electron reduction to the semiquinone state (–40 15 mV) and full reduction to flavin hydroquinone (–130 15 mV) that are identical to those obtained earlier for the isolated FlRd. To probe the influence of the presence of FlRd-reductase on the reduction potentials of the nonheme diiron site of FlRd, the 1:1 FlRd/FlRd-reductase titration was repeated and followed by EPR spectroscopy. Resulting data were treated in the same manner as the titration of the isolated FlRd and revealed an upshift of 40 mV for each transition. The resulting modulation of the reduction
Properties of Flavodiiron Proteins
37
potentials of FlRd by its cognate reductase led to the proposal of an electron transfer mechanism, which is discussed in the next section.
7. Conclusions The work summarized herein described several experimental methodologies that altogether contributed to a successful characterization of flavodiiron proteins from different source organisms. Studies undertaken at a molecular level are essential to complement the functional in vivo studies, with each approach reciprocally contributing to a deeper knowledge on the structure–function relationship of a novel family of proteins. Most of the studies on FDPs were performed with recombinant proteins overexpressed heterologously in E. coli. The quality of recombinant proteins in terms of cofactor incorporation and sample homogeneity is a recurrent challenge. Through a permanent search for improving expression conditions of flavodiiron proteins, it has been observed that flavin and iron incorporations are favored by lower temperature and decreased aeration of growing cultures, with the latter being achieved by reducing the gas headspace and stirring speed. Flavodiiron proteins from different sources were purified successfully in a small number of chromatographic steps, which include anion-exchange and gel filtration columns. A development in the quality of purified FDPs was the observation that cofactor integrity (namely FMN) throughout the purification steps benefited from the inclusion of glycerol (18%) in all the buffers. The redox-active cofactors in FDPs were readily extractable by standard procedures (typically acid extraction) and easily quantified by spectrophotometric (flavin) and colorimetric (iron) methodologies. As summarized in Table 2.1, the cofactor content of isolated FDPs (0.7–1 FMN and 1–2 Fe per monomer of flavodiiron core) corresponds to what can be inferred from the peptide sequences, where each monomer comprises one FMN-binding flavodoxin module and a b-lactamase module where a nonheme diiron center is embedded. Consistently, FDPs with extra C-terminal structural modules have a higher cofactor content. For instance, the E. coli flavorubredoxin contains three Fe ions per monomer, one from the [Fe-Cys4] center in the rubredoxin domain and the diiron center in the lactamase module. The characterization of flavodiiron proteins by spectroscopic techniques allowed redox-sensitive spectral fingerprints assigned to each cofactor to be established. Visible spectroscopy has been used to characterize essentially the flavin moieties (and also the rubredoxin domain in flavorubredoxin) and EPR and Mo¨ssbauer spectroscopies were employed to study the nonheme diiron center.
38
Joa˜o B. Vicente et al.
The thermodynamic properties of flavodiiron proteins, in terms of reduction potentials of their cofactors, have only been determined for a few cases. Whereas known potentials for M. thermoacetica FDP (SilaghiDumitrescu et al., 2003) and D. gigas ROO (Gomes et al., 1997) regard solely their flavin moieties, a thorough redox characterization has been undertaken with E. coli flavorubredoxin (Vicente and Teixeira, 2005). The determined reduction potentials of FDP-bound FMN moieties range between 0 ! –117 mV for the FMNox/FMNsq pair and –130 ! –220 mV for the FMNsq/FMNred pair. These potentials differ from the established redox behavior of canonical flavodoxins. Flavodoxins greatly stabilize the semiquinone state because of the large difference of reduction potentials between the FMNox/FMNsq pair (þ121 ! –229 mV) and the FMNsq/ FMNred pair (–372 ! –522 mV), an effect attributed to conformational rearrangements and a series of hydrogen bonds (Hoover et al., 1999; Kasim and Swenson, 2000; O’Farrell et al., 1998; Paulsen et al., 1990). Moreover, flavodoxins stabilize the blue neutral semiquinone radical, contrasting with the red anionic semiquinone observed in FDP-bound flavodoxin modules. Formation of a red anionic semiquinone in FDPs may be related to a prevalence of basic over acidic residues in the FMN pocket (Fraza˜o et al., 2000), which could contribute to lower the pKa of 8.3 for the equilibrium between the red and the blue semiquinone forms of free FMN (Ghisla and Edmonson, 2001). The extensive redox characterization focusing on E. coli flavorubredoxin and its cognate reductase combined potentiometric and spectroscopic methods, as described in detail earlier. It should be emphasized that complete understanding of the thermodynamic properties of this system was significantly supported by studying the truncated modules of flavorubredoxin in parallel with the whole enzyme. Results are summarized in Fig. 2.6, depicting the reduction potentials of each cofactor and the proposed electron transfer mechanism inferred on a strictly thermodynamic basis. By titrating flavorubredoxin in a stoichiometric mixture with the reductase, it was observed that upon interaction of the two redox partners, the reduction potentials of the iron cofactors are upshifted with respect to the values obtained for the isolated FlRd, whereas those of the flavins in FlRd (FMN) and FlRd-reductase (FAD) remain essentially unaltered. Modulation of the redox properties of FlRd by its reductase poses the possibility of intramolecular eT steps to be inferred differently, with respect to the reduction potentials of the isolated FlRd. In the isolated protein the more favorable intramolecular eT mechanism involves full reduction of FMN, to further allow two-electron reduction of the diiron site. However, the observed redox shifts resulting from the interaction with FlRd-reductase change the situation regarding possible eT mechanisms. On the one hand, the upshift in the Rd potential creates a thermodynamic barrier for full reduction (two electrons) of FMN. On the other hand, the upshift observed in the reduction potentials of the diiron center allows the flavin
39
Properties of Flavodiiron Proteins
Reduction Potential (mV)
Stoichiometric mixture FeIII FeIII
FMNox −40 −65
Fe-S
+20 FeIII
FeII
−50 FMNsq
FeII FeII
−130 FMNred −238
FAD
Figure 2.6 Electron transfer mechanism of Escherichia coli flavorubredoxin and its reductase on a strictly thermodynamic base. The scheme depicts reduction potentials upon interaction of the two redox partners. Curved full arrows indicate probable electron transfer steps inferred on a pure thermodynamic basis. Dotted arrow depicts the (unlikely) formation of the flavin two-electron-reduced hydroquinone.
semiquinone to act as a one-electron shuttle to the diiron center, without the need to reach the hydroquinone state. Assuming this eT mechanism on a pure thermodynamic basis, the ‘‘fully’’ reduced FlRd under ‘‘normal’’ operative conditions would have a total of four electrons available for reductive chemistry, sufficient to catalyze the reduction of four NO to two N2O or full reduction of oxygen to water. This mechanism was supported by a thorough kinetic study (Vicente et al., 2007b,c).The complexity of the redox behavior of the eT chain composed by flavorubredoxin and its reductase suggests that a redox characterization of other flavodiiron proteins, taking into account all the eT components, may provide clues for the efficiency and functionality of the corresponding eT chains, whereby substrate (NO and/or oxygen) reduction is coupled to NAD(P)H oxidation.
7.1. Functional properties Since the establishment of the family of flavodiiron proteins, their functional properties have been assessed through a combination of parallel in vitro and in vivo studies. Whereas studies on the isolated proteins focused on the structure–function relationship and the corresponding in vitro NO and/or oxygen reductase activities, in vivo studies attempted to provide clues for the relative role of each FDP in its source organism. The first function assigned to a flavodiiron protein concerned the oxygen reductase activity of D. gigas rubredoxin:oxygen oxidoreductase,
40
Joa˜o B. Vicente et al.
proposed to provide a protective mechanism against oxygen toxicity for this anaerobic organism (Chen et al., 1993b; Gomes et al., 1997). Later on, it was demonstrated that the E. coli orthologous enzyme flavorubredoxin had considerable NO reductase activity (Gomes et al., 2002), in the order of respiratory heme b3:nonheme iron NO reductases. More reports on other flavodiiron proteins revealed that their substrate selectivity is different, despite the structural similarity of the studied FDPs. Whereas E. coli flavorubredoxin has a clear preference for NO, FDPs from M. thermoacetica, D. gigas, and D. vulgaris have comparable NO and oxygen reductase activities (refer to Vicente et al., 2007b). At the other extreme is the FDP from M. marburgensis, which reduces oxygen to water exclusively, displaying no activity toward NO (Seedorf et al., 2004, 2007). The role of E. coli flavorubredoxin in in vivo NO detoxification was first proposed by Gardner and colleagues (2002) based on observations that a deletion of norV, encoding flavorubredoxin, results in a strain with higher sensitivity to nitric oxide releasing compounds (Gardner et al., 2002; Hutchings et al., 2002; Justino et al., 2005b). More recently, it has been demonstrated that a mutant strain of D. gigas where the Dg_ROO encoding gene has been silenced is also more sensitive to nitrosative stress (both NO and GSNO) than the wild-type strain (Rodrigues et al., 2006). Moreover, it has been shown by complementation studies in an E. coli norV mutant that D. gigas ROO (Rodrigues et al., 2006 ), as well as FDPs from M. thermoacetica (Silaghi-Dumitrescu et al., 2003) and D. vulgaris (SilaghiDumitrescu et al., 2005), can protect in vivo E. coli from NO toxicity. In E. coli, the norV gene is cotranscribed in a di-cistronic unit with norW, which encodes its redox partner, the NADPH:flavorubredoxin oxidoreductase (da Costa et al., 2003). The norW mutant does not show such pronounced phenotypes as the norV mutant, and indeed this strain still retains some NO reductase activity, which is completely abolished in the norV strain, suggesting an ancillary role for NorW that may be accomplished by other reductases (Gardner et al., 2002). Transcriptional regulation of the norVW operon of E. coli has been studied extensively. The norV promoter is activated by reactive nitrogen species, both aerobically and anaerobically (da Costa et al., 2003; Gardner et al., 2002; Hutchings et al., 2002), and also during nitrate/nitrite respiration when traces of NO may be formed (da Costa et al., 2003; Hutchings et al., 2002). The regulation of norVW is controlled by the oxygen-sensitive transcription factor FNR and the nitrate/nitrite responsive regulators NarL/NarP (Constantinidou et al., 2006; da Costa et al., 2003). Studies showing that deletion of the divergently transcribed gene norR caused similar phenotype as the deletion of norV (Gardner et al., 2002; Hutchings et al., 2002; Justino et al., 2005b) and completely abolished the nitrosative induction of norVW (Gardner et al., 2003; Hutchings et al., 2002;
Properties of Flavodiiron Proteins
41
Justino et al., 2005a) have established unequivocally NorR as an NOdependent regulator of the flavorubredoxin operon. Great interest has been raised in the bacterial response to NO and nitrosative stress because of the important role that the production of NO, by the host immune system, has in controlling infection. Several reports have been published regarding the transcriptional response of E. coli to nitrosative stress. These microarray studies differentiate themselves on the basis of several factors: the source and extent of the stress (e.g., pure NO, GSNO, or acidified nitrite), growth media (either rich or defined fermentative/nonfermentative media), or even the type of growth conditions (e.g., batch/ chemostat continuous cultures, aerobic/anaerobic cultures) (Flatley et al., 2005; Justino et al., 2005b; Mukhopadhyay et al., 2004; Pullan et al., 2007). Interestingly, results obtained show little in common. Only a very small group of genes had their expression significantly induced in most of the studies—hmp, norVW, nrdH, and ytfE genes—and indeed only norV gene expression was induced under all the conditions examined. While hmp and norV encode two inducible NO detoxification systems of E. coli, respectively, flavohemoglobin and flavorubredoxin, the ytfE gene has been assigned a function in the repair of damaged iron–sulfur clusters (Justino et al., 2006, 2007). The role of nrdH, encoding a glutaredoxin-like protein, remains elusive. The relevance of these results regarding an actual interaction with the host immune system is strongly supported by the fact that, apart from nrdH, these genes were also induced in Salmonella enterica after internalization into activated macrophages (Eriksson et al., 2003). Flavohemoglobin (Hmp) is a member of the family of NO-metabolizing cytosolic globins and was the first inducible NO detoxification system identified in E. coli (Poole et al., 1996). In vitro Hmp exhibits weak NO reductase activity, suggesting a primary function as an O2-dependent NO dioxygenase or as an NO denitrosylase (reviewed in Gonc¸alves et al., 2006; Poole, 2005). Initially, it was assumed that Hmp would be the main participant in NO detoxification aerobically, and FlRd anaerobically, and that under microaerobic conditions both enzymes would contribute similarly a double mutant strain showed a very strong growth inhibition upon NO exposure (Gardner et al., 2002). Indeed, in aerobic cultures, a norVW mutant strain does not display increased sensitivity to nitroprusside when compared with the parent strain (Hutchings et al., 2002), while Hmp does not confer anaerobic protection to aconitase NO-derived damage (Gardner and Gardner, 2002; Gardner et al., 2002). However, it was later shown that the growth inhibition caused by NO to anaerobic cultures growing in minimal medium is equal in norV or hmp single mutant strains and that, as observed for microaerobic conditions, it is much more severe in a strain with both genes deleted ( Justino et al., 2005b). This fact clearly demonstrated a role for Hmp in anaerobic NO detoxification in E. coli.
42
Joa˜o B. Vicente et al.
These two systems differ also in their protein expression rates, with FlRd showing the faster response. Under anaerobic conditions, NO induced maximal expression of the FlRd protein within 5–15 min, whereas Hmp maximal level requires longer times, being reached only after 45 min, suggesting that the faster response is achieved by the enzyme that in vitro has the higher NO reductase activity, i.e., flavorubredoxin ( Justino et al., 2005b). Interestingly, the aerobic transcriptional response of hmp and norV to a constant nitrosative stress showed oscillatory behaviors with distinct features. The genes showed two peaks of induction, one after 5 min and the other after 90 min, but while the norV gene showed the highest induction in the first peak, the induction of hmp was more pronounced in its second peak (Mukhopadhyay et al., 2004). These results further show that flavorubredoxin appears to have a more efficient initial response. Thus, it has been demonstrated that in E. coli, which possesses two inducible NO-detoxifying systems, Hmp is an active participant in a broad range of O2 concentrations, whereas FlRd seems to contribute essentially when oxygen is limited. It should be noted that this does not entirely diminish the role of FlRd in protection from NO toxicity in in vivo situations, as pathogen colonization occurs close to anaerobic environments. So far, only flavohemoglobin appears to have a role in protecting S. enterica (Stevanin et al., 2002) or E. coli from macrophage NO-dependent killing, as an E. coli norV mutant strain showed similar survival ability as the parent strain (Pullan et al., 2007). Furthermore, Hmp, but not FlRd, is required for Salmonella virulence in mice (Bang et al., 2006) under the tested conditions. However, the functional characterization of E. coli flavorubredoxin (both in vivo and in vitro) clearly shows its involvement in NOderived stress response, revealing that more studies need to be performed on flavodiiron proteins to fully elucidate their physiological role.
ACKNOWLEDGMENTS This work was partially supported by FCT Projects POCTI/1999/BME/36558, POCTI/ 2002/BME/44597, and POCI/SAU-IMI/56088/2004. JBV, MCJ, and VLG benefited from FCT Ph.D. grants, respectively, SFRH/BD/9136/2002, SFRH/BD/13756/2003, and SFRH/BD/29428/2006.
REFERENCES Andersson, J. O., Sjogren, A. M., Davis, L. A., Embley, T. M., and Roger, A. J. (2003). Phylogenetic analyses of diplomonad genes reveal frequent lateral gene transfers affecting eukaryotes. Curr. Biol. 13, 94–104. Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A., and Struhl, K. (1995). ‘‘Current Protocols in Molecular Biology.’’ Greene Publishing Associates and Wiley Interscience, New York.
Properties of Flavodiiron Proteins
43
Bang, I. S., Liu, L., Vazquez-Torres, A., Crouch, M. L., Stamler, J. S., and Fang, F. C. (2006). Maintenance of nitric oxide and redox homeostasis by the Salmonella flavohemoglobin hmp. J. Biol. Chem. 281, 28039–28047. Chen, L., Liu, M. Y., Legall, J., Fareleira, P., Santos, H., and Xavier, A. V. (1993a). Purification and characterization of an NADH-rubredoxin oxidoreductase involved in the utilization of oxygen by Desulfovibrio gigas. Eur. J. Biochem. 216, 443–448. Chen, L., Liu, M. Y., LeGall, J., Fareleira, P., Santos, H., and Xavier, A. V. (1993b). Rubredoxin oxidase, a new flavo-hemo-protein, is the site of oxygen reduction to water by the ‘‘strict anaerobe’’ Desulfovibrio gigas. Biochem. Biophys. Res. Commun. 193, 100–105. Constantinidou, C., Hobman, J. L., Griffiths, L., Patel, M. D., Penn, C. W., Cole, J. A., and W., O. (2006). A reassessment of the FNR regulon and transcriptomic analysis of the effects of nitrate, nitrite, NarXL, and NarQP as Escherichia coli K12 adapts from aerobic to anaerobic growth. J. Biol. Chem. 281, 4802–4815. da Costa, P. N., Teixeira, M., and Saraiva, L. M. (2003). Regulation of the flavorubredoxin nitric oxide reductase gene in Escherichia coli: Nitrate repression, nitrite induction, and possible post-transcription control. FEMS Microbiol. Lett. 218, 385–393. Eriksson, S., Lucchini, S., Thompson, A., Rhen, M., and Hinton, J. C. D. (2003). Unravelling the biology of macrophage infection by gene expression profiling of intracellular Salmonella enterica. Mol. Microbiol. 47, 103–118. Fischer, D. S., and Price, D. C. (1964). A simple serum iron method using the new sensitive chromogen tripyridyl-s-triazine. Clin. Chem. 10, 21–31. Flatley, J., Barrett, J., Pullan, S. T., Hughes, M. N., Green, J., and Poole, R. K. (2005). Transcriptional responses of Escherichia coli to S-nitrosoglutathione under defined chemostat conditions reveal major changes in methionine biosynthesis. J. Biol. Chem. 280, 10065–10072. Frazao, C., Silva, G., Gomes, C. M., Matias, P., Coelho, R., Sieker, L., Macedo, S., Liu, M. Y., Oliveira, S., Teixeira, M., Xavier, A. V., Rodrigues-Pousada, C., et al. (2000). Structure of a dioxygen reduction enzyme from Desulfovibrio gigas. Nat. Struct. Biol. 7, 1041–1045. Gardner, A. M., and Gardner, P. R. (2002). Flavohemoglobin detoxifies nitric oxide in aerobic, but not anaerobic, Escherichia coli: Evidence for a novel inducible anaerobic nitric oxide-scavenging activity. J. Biol. Chem. 277, 8166–8171. Gardner, A. M., Gessner, C. R., and Gardner, P. R. (2003). Regulation of the nitric oxide reduction operon (norRVW ) in Escherichia coli: Role of NorR and sigma54 in the nitric oxide stress response. J. Biol. Chem. 278, 10081–10086. Gardner, A. M., Helmick, R. A., and Gardner, P. R. (2002). Flavorubredoxin, an inducible catalyst for nitric oxide reduction and detoxification in Escherichia coli. J. Biol. Chem. 277, 8172–8177. Garfin, D. E. (1990). One-dimensional gel electrophoresis. Methods Enzymol. 182, 425–441. Ghisla, S., and Edmonson, D. (2001). Flavin coenzymes. In ‘‘Encyclopedia of Life Sciences.’’ Gomes, C. M., Giuffre, A., Forte, E., Vicente, J. B., Saraiva, L. M., Brunori, M., and Teixeira, M. (2002). A novel type of nitric-oxide reductase: Escherichia coli flavorubredoxin. J. Biol. Chem. 277, 25273–25276. Gomes, C. M., Silva, G., Oliveira, S., LeGall, J., Liu, M. Y., Xavier, A. V., RodriguesPousada, C., and Teixeira, M. (1997). Studies on the redox centers of the terminal oxidase from Desulfovibrio gigas and evidence for its interaction with rubredoxin. J. Biol. Chem. 272, 22502–22508. Gomes, C. M., Vicente, J. B., Wasserfallen, A., and Teixeira, M. (2000). Spectroscopic studies and characterization of a novel electron-transfer chain from Escherichia coli involving a flavorubredoxin and its flavoprotein reductase partner. Biochem. 39, 16230–16237. Gonc¸alves, V. L., Nobre, L. S., Vicente, J. B., Teixeira, M., and M., S. L. (2006). Flavohemoglobin requires microaerophylic conditions for nitrosative protection of Staphylococcus aureus. FEBS Lett. 580, 1817–1821.
44
Joa˜o B. Vicente et al.
Hoover, D. M., Drennan, C. L., Metzger, A. L., Osborne, C., Weber, C. H., Pattridge, K. A., and Ludwig, M. L. (1999). Comparisons of wild-type and mutant flavodoxins from Anacystis nidulans: Structural determinants of the redox potentials. J. Mol. Biol. 294, 725–743. Hutchings, M. I., Mandhana, N., and Spiro, S. (2002). The NorR protein of Escherichia coli activates expression of the flavorubredoxin gene norV in response to reactive nitrogen species. J. Bacteriol. 184, 4640–4643. Jouanneau, Y., Meyer, C., Asso, M., Guigliarelli, B., and Willison, J. C. (2000). Characterization of a nif-regulated flavoprotein (FprA) from Rhodobacter capsulatus: Redox properties and molecular interaction with a [2Fe-2S] ferredoxin. Eur. J. Biochem. 267, 780–787. Justino, M. C., Almeida, C. C., Gonc¸alves, V. L., Teixeira, M., and Saraiva, L. M. (2006). Escherichia coli YtfE is a di-iron protein with an important function in assembly of ironsulphur clusters. FEMS Microbiol. Lett. 257, 278–284. Justino, M. C., Almeida, C. C., Teixeira, M., and Saraiva, L. M. (2007). Escherichia coli di-iron YtfE protein is necessary for the repair of stress damaged iron-sulfur clusters. J. Biol. Chem. 282, 10352–10359. Justino, M. C., Gonc¸alves, V. M., and Saraiva, L. M. (2005a). Binding of NorR to three DNA sites is essential for promoter activation of the flavorubredoxin gene, the nitric oxide reductase of Escherichia coli. Biochem. Biophys. Res. Commun. 328, 540–544. Justino, M. C., Vicente, J. B., Teixeira, M., and Saraiva, L. M. (2005b). New genes implicated in the protection of anaerobically grown Escherichia coli against nitric oxide. J. Biol. Chem. 280, 2636–2643. Kasim, M., and Swenson, R. P. (2000). Conformational energetics of a reverse turn in the Clostridium beijerinckii flavodoxin is directly coupled to the modulation of its oxidationreduction potentials. Biochemistry 39, 15322–15332. Mukhopadhyay, P., Zheng, M., Bedzyk, L. A., LaRossa, R. A., and Storz, G. (2004). Prominent roles of the NorR and Fur regulators in the Escherichia coli transcriptional response to reactive nitrogen species. Proc. Natl. Acad. Sci. USA 101, 745–750. Nolling, J., Ishii, M., Koch, J., Pihl, T. D., Reeve, J. N., Thauer, R. K., and Hedderich, R. (1995). Characterization of a 45-kDa flavoprotein and evidence for a rubredoxin, two proteins that could participate in electron transport from H2 to CO2 in methanogenesis in Methanobacterium thermoautotrophicum. Eur. J. Biochem. 231, 628–638. O’Farrell, P. A., Walsh, M. A., McCarthy, A. A., Higgins, T. M., Voordouw, G., and Mayhew, S. G. (1998). Modulation of the redox potentials of FMN in Desulfovibrio vulgaris flavodoxin: Thermodynamic properties and crystal structures of glycine-61 mutants. Biochemistry 37, 8405–8416. Paulsen, K. E., Stankovich, M. T., Stockman, B. J., and Markley, J. L. (1990). Redox and spectral properties of flavodoxin from Anabaena 7120. Arch. Biochem. Biophys. 280, 68–73. Poole, R. K. (2005). Nitric oxide and nitrosative stress tolerance in bacteria. Biochem. Soc. Trans. 33, 176–180. Poole, R. K., Anjum, M. F., Membrillo-Hernandez, J., Kim, S., Hughes, M. N., and Stewart, V. (1996). Nitric oxide, nitrite, and Fnr regulation of hmp (flavohemoglobin) gene expression in Escherichia coli K-12. J. Bacteriol. 178, 5487–5492. Pullan, S. T., Gidley, M. D., Jones, R. A., Barrett, J., Stevanin, T. M., Read, R. C., Green, J., and Poole, R. K. (2007). Nitric oxide in chemostat-cultured Escherichia coli is sensed by Fnr and other global regulators: Unaltered methionine biosynthesis indicates lack of S nitrosation. J. Bacteriol. 189, 1845–1855. Rodrigues, R., Vicente, J. B., Fe´lix, R., Oliveira, S., Teixeira, M., and RodriguesPousada, C. (2006). Desulfovibrio gigas flavodiiron protein affords protection against nitrosative stress in vivo. J. Bacteriol. 188, 2745–2751. Saraiva, L. M., Vicente, J. B., and Teixeira, M. (2004). The role of the flavodiiron proteins in microbial nitric oxide detoxification. Adv. Microb. Physiol. 49, 77–129.
Properties of Flavodiiron Proteins
45
Schilling, O., Vogel, A., Kostelecky, B., Natal da Luz, H., Spemann, D., Spath, B., Marchfelder, A., Troger, W., and Meyer-Klaucke, W. (2005). Zinc- and iron-dependent cytosolic metallo-beta-lactamase domain proteins exhibit similar zinc-binding affinities, independent of an atypical glutamate at the metal-binding site. Biochem. J. 385, 145–153. Seedorf, H., Dreisbach, A., Hedderich, R., Shima, S., and Thauer, R. K. (2004). F420H2 oxidase (FprA) from Methanobrevibacter arboriphilus, a coenzyme F420-dependent enzyme involved in O2 detoxification. Arch. Microbiol. 182, 126–137. Seedorf, H., Hagemeier, C. H., Shima, S., Thauer, R. K., Warkentin, E., and Ermler, U. (2007). Structure of coenzyme F420H2 oxidase (FprA), a di-iron flavoprotein from methanogenic Archaea catalyzing the reduction of O2 to H2O. FEBS J. 274, 1588–1599. Silaghi-Dumitrescu, R., Coulter, E. D., Das, A., Ljungdahl, L. G., Jameson, G. N., Huynh, B. H., and Kurtz, D. M., Jr. (2003). A flavodiiron protein and high molecular weight rubredoxin from Moorella thermoacetica with nitric oxide reductase activity. Biochem. 42, 2806–2815. Silaghi-Dumitrescu, R., Kurtz, D. M., Jr., Ljungdahl, L. G., and Lanzilotta, W. N. (2005). X-ray crystal structures of Moorella thermoacetica FprA: Novel diiron site structure and mechanistic insights into a scavenging nitric oxide reductase. Biochem. 44, 6492–6501. Sober, H. A., and Harte, R. A. (1968). ‘‘Handbook of Biochemistry: Selected Data for Molecular Biology.’’ CRC Press, Boca Raton, FL. Solomon, E. I., Brunold, T. C., Davis, M. I., Kemsley, J. N., Lee, S. K., Lehnert, N., Neese, F., Skulan, A. J., Yang, Y. S., and Zhou, J. (2000). Geometric and electronic structure/function correlations in non-heme iron enzymes. Chem. Rev. 100, 235–350. Stevanin, T. M., Poole, R. K., Demoncheaux, E. A. G., and Read, R. C. (2002). Flavohemoglobin Hmp protects Salmonella enterica serovar typhimurium from nitric oxide-related killing by human macrophages. Infect. Immun. 70, 4399–4405. Susin, S., Abian, J., Sanchez-Baeza, F., Peleato, M. L., Abadia, A., Gelpi, E., and Abadia, J. (1993). Riboflavin 30 - and 50 -sulfate, two novel flavins accumulating in the roots of irondeficient sugar beet (Beta vulgaris). J. Biol. Chem. 268, 20958–20965. Vicente, J. B., Carrondo, M. A., Teixeira, M., and Fraza˜o, C. (2007a). Structural studies on flavodiiron proteins. Methods Enzymol. 437[1] (this volume). Vicente, J. B., Gomes, C. M., Wasserfallen, A., and Teixeira, M. (2002). Module fusion in an A-type flavoprotein from the cyanobacterium Synechocystis condenses a multiplecomponent pathway in a single polypeptide chain. Biochem. Biophys. Res. Commun. 294, 82–87. Vicente, J. B., Scandurra, F. M., Forte, E., Brunori, M., Sarti, P., Teixeira, M., and Alessandro Giuffre`, A. (2007b). Kinetic characterization of the Escherichia coli nitric oxide reductase flavorubredoxin. Methods Enzymol. 437[3] (this volume). Vicente, J. B., Scandurra, F. M., Rodrigues, J. V., Brunori, M., Sarti, P., Teixeira, M., and Giuffre`, A. (2007c). Kinetics of electron transfer from NADH to the Escherichia coli nitric oxide reductase flavorubredoxin. FEBS J. 274, 677–686. Vicente, J. B., and Teixeira, M. (2005). Redox and spetroscopic properties of the Escherichia coli nitric oxide-detoxifying system involving flavorubredoxin and its NADH-oxidizing redox partner. J. Biol. Chem. 280, 34599–34608. Walker, J. M. (1994). The bicinchoninic acid (BCA) assay for protein quantitation. Methods Mol. Biol. 32, 5–8. Wasserfallen, A., Huber, K., and Leisinger, T. (1995). Purification and structural characterization of a flavoprotein induced by iron limitation in Methanobacterium thermoautotrophicum Marburg. J. Bacteriol. 177, 2436–2441. Wasserfallen, A., Ragettli, S., Jouanneau, Y., and Leisinger, T. (1998). A family of flavoproteins in the domains Archaea and Bacteria. Eur. J. Biochem. 254, 325–332.
This page intentionally left blank
C H A P T E R
T H R E E
Kinetic Characterization of the Escherichia coli Nitric Oxide Reductase Flavorubredoxin Joa˜o B. Vicente,* Francesca M. Scandurra,† Elena Forte,† Maurizio Brunori,† Paolo Sarti,† Miguel Teixeira,* and Alessandro Giuffre`† Contents 1. Introduction 2. Amperometric Measurements 3. Spectroscopic Measurements 3.1. Reduction of flavorubredoxin reductase by NADH 3.2. Reduction of the rubredoxin domain of FlRd by FlRd-reductase 3.3. Reduction of FlRd by FlRd-reductase 3.4. The effect of ionic strength and pH 4. Conclusions Acknowledgments References
48 49 51 53 54 55 59 61 61 61
Abstract In strict or facultative anaerobic microorganisms, the flavodiiron proteins (FDP) have been recognized to take part in the response mechanism to both nitrosative and oxidative stress. Their function consists of the reduction of nitric oxide and/or oxygen at the diiron center, and specificity for one substrate or the other appears to be characteristic of the corresponding microorganism, possibly depending on the properties of the catalytic site. Particularly focused on the flavorubredoxin, i.e., the Escherichia coli FDP, herein the amperometric and time-resolved spectroscopic approaches are presented, giving access to the study of in vitro reactivity of a complex multi-redox center enzyme.
* {
Instituto de Tecnologia Quı´mica e Biolo´gica, Universidade Nova de Lisboa, Oeiras, Portugal Department of Biochemical Sciences, CNR Institute of Molecular Biology and Pathology and Istituto Pasteur–Fondazione Cenci Bolognetti Sapienza, University of Rome, Rome, Italy
Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37003-1
#
2008 Elsevier Inc. All rights reserved.
47
48
Joa˜o B. Vicente et al.
1. Introduction Flavodiiron proteins [FPD, originally named A-type flavoproteins (Wasserfallen et al., 1998)] are microbial enzymes, expressed in Bacteria, Archaea, and some protozoan pathogens, that are involved in the response to nitrosative and/or oxidative stress (Saraiva et al., 2004). Initially, FDPs were proposed to be O2-scavenging enzymes, allowing anaerobic microorganisms to survive transient exposure to O2 (Chen et al., 1993; Frazao et al., 2000; Gomes et al., 1997). Later on, a general role in nitric oxide (NO) detoxification was also proposed for these enzymes (Gardner et al., 2002; Gomes et al., 2002), working on the FDP from Escherichia coli, named flavorubredoxin (FlRd) (Gomes et al., 2000). The role of FlRd in NO detoxification was put forward by Gardner and co-workers (2002), who discovered that NO induces the expression of FlRd and that deletion of the FlRd gene leads to an enhanced sensitivity to NO of anaerobically grown E. coli. FlRd may protect E. coli against NO, catalyzing the anaerobic reduction of NO to N2O, according to the reaction:
2e þ 2Hþ þ 2NO ! N2 O þ H2 O: The protective function of FlRd acting as an anaerobic NO reductase is thus complementary to the aerobic NO-detoxifying role of flavohemoglobin, which oxidatively degrades NO into nitrate. Flavorubredoxin (whose basic properties are described in Gomes et al., 2000) and its redox partner protein NADH:flavorubredoxin oxidoreductase (FlRd-reductase) constitute an eT chain, with NADH acting as the primary electron donor (Fig. 3.1). FlRd and FlRd-reductase are encoded in a dicistronic unit, norVW, which is under the regulation of NorR, the product of the adjacent but divergently transcribed norR gene (da Costa et al., 2003). FlRd-reductase, encoded by norW, is a 43-kDa monomeric protein, containing one FAD moiety. FlRd, the product of norV, is a
Rd-domain
Flavodiiron-Domain
NADH
NO FAD
NAD+
e−
FIRd-reductase
Fe-S
FMN
Fe-Fe N2O
Flavorubredoxin
Figure 3.1 Schematic representation of E. coli electron transfer chain coupling NADH oxidation to NO reduction by flavorubredoxin.
Kinetic Properties of E. coli Flavorubredoxin
49
homotetramer of 54 kDa monomers, each composed of three structural domains: the flavodiiron structural core, comprising an N-terminal b-lactamase-like module (harboring the non-heme diiron active site); and an FMNbinding flavodoxin-like module; and a C-terminal rubredoxin module. The NO reductase activity of FlRd can be assessed by performing amperometric measurements with NO-specific electrodes on the purified recombinant protein (Gomes et al., 2002). The same approach was easily extended to other FDPs also endowed with NO reductase activity (Rodrigues et al., 2006; Silaghi-Dumitrescu et al., 2003, 2005a). This chapter reports on the methodological approaches undertaken to characterize the NO reductase activity of FlRd and the kinetics of electron transfer along the FlRd eT chain, coupling NADH oxidation to NO reduction in E. coli.
2. Amperometric Measurements The NO reductase activity of E. coli FlRd, purified as described previously (Gomes et al., 2000), is measured under anaerobic conditions using a NO-selective Clark-type electrode (Gomes et al., 2002). The basic equipment includes a commercially available gas-tight chamber allocating an electrode sensitive to NO (ISO-NO MarkII, World Precision Instruments, Sarasota, FL); the amperometer detects the current generated at the tip of the electrode where NO is oxidized (NO ! NOþ þ e) according to the applied voltage (0.9 V). The instrument is interfaced with a computer where digitized data are transferred for analysis. The reaction chamber has a volume of 1 ml and is capped with a stopper with two holes, one for the NO electrode and the other one for sample addition through Hamilton syringes. A typical amperometric assay, carried out at room temperature, is shown in Fig. 3.2. The empty chamber is first flushed with nitrogen to minimize oxygen contamination and is then filled with deoxygenated buffer. In these measurements, 20 mM EDTA is used to chelate possible contaminant metal ions in the buffer, which are known to affect the stability of NO in solution. The lid is then fitted, preventing formation of gas bubbles that can interfere with the measurements. Contaminant O2 in the measuring chamber is further scavenged by the addition of glucose (2 mM ), glucose oxidase (17 units/ml), and catalase (130 units/ml). Afterward, the redox partner protein of FlRd (i.e., FlRd-reductase) is added to efficiently shuttle electrons between NADH and FlRd (see Fig. 3.1). When a stable baseline is set, the NO electrode is calibrated by recording the signal produced by sequential additions of aliquots of a NO stock solution. The latter solution is prepared by equilibrating degassed water with NO gas at 1 atm and room temperature; the procedure yields 2 mM NO in solution. The NO
50
Joa˜o B. Vicente et al.
Fe-S FMN Fe-Fe 1mM NADH 1.4 mM NO
22.5 nM FIRd
Fe-S 22.5 nM RdT
2.7 mM NO
2.6 mM 2.7 mM FIRd-red NO
3 min
10 s
Figure 3.2 NO reductase activity of E. coli FlRd.To the chamber containing degassed buffer and 2.6 mM FlRd-reductase, three aliquots of a NO solution are added in sequence to yield 6.8 mM NO in solution, followed by the addition of 1 mM NADH. A fast NO consumption, following zero-order kinetics, is then observed after the addition of 22.5 nM FlRd, but not after the addition of the same amount of RdT. Buffer: 50 mM Tris-HCl, pH 7.6, 20 mM EDTA.
concentration in stock solutions is checked by spectrophotometric titrations using reduced bovine cytochrome c oxidase, which binds one NO molecule per enzyme unit (Stubauer et al., 1998). As shown in Fig. 3.2, the addition of NO to the deoxygenated buffer causes an upward signal proportional to the amount of NO added, thus allowing calibration of the electrode current. The onset kinetics of the signal after NO addition provides information on the response time of the electrode, which depends on experimental conditions (particularly temperature), as well as on the quality of the electrode. The addition to the reaction chamber of 1 mM NADH, the electron donor of FlRd-reductase, triggers an unspecific consumption of NO, independent of the enzyme (FlRd). However, upon addition of FlRd (22.5 nM ), a much faster consumption of NO is observed (see Fig. 3.2). The reaction follows zero-order kinetics, and the rate of NO consumption, estimated from the slope of the NO decay corrected for unspecific NO consumption, allows the turnover number of FlRd to be determined. Enzyme integrity is requested since, while FlRd addition causes significant NO consumption, no effect is observed if the sole rubredoxin domain of FlRd (RdT) is added (see Fig. 3.2); such a domain, obtained by truncation of the FlRd gene and purification (Gomes et al., 2000), proved to be particularly useful also in the spectroscopic experiments that follow.
Kinetic Properties of E. coli Flavorubredoxin
51
The finding that RdT does not catalyze NO degradation is consistent with information that NO chemistry occurs at the Fe-Fe site of FDPs, as demonstrated by replacing iron with zinc in the Moorella thermoacetica enzyme (Silaghi-Dumitrescu et al., 2003). Overall, data suggest that E. coli FlRd is endowed with a considerable NOdegrading activity (14.9 6.7 mol NOmol FlRd1s1) comparable to that determined for the heme b3/FeB-containing nitric oxide reductases (NOR) expressed in denitrifying bacteria [e.g., 10–50 mol NOmol enzyme1s1 for the Paracoccus denitrificans enzyme (Zumft, 1997)]. By measuring the NO consumption rate at different NO concentrations, it was shown that FlRd has a high affinity for NO (KM 1.2 mM, unpublished results), in line with the hypothesis that the physiological role of this enzyme is the NO detoxification of E. coli under anaerobic conditions (Gardner et al., 2002; Gomes et al., 2002; Justino et al., 2005). Interestingly, E. coli FlRd displays a very low O2 reductase activity (<1 s1, unpublished results), overestimated in a previous study (Gomes et al., 2002) as a consequence of the direct reaction of FlRd-reductase with O2. Several other FDP proteins from different bacterial sources have been characterized in terms of both NO and O2 reductase activity, displaying different substrate selectivity (Table 3.1). The enzyme from E. coli displays a clear preference for NO (Gomes et al., 2002), whereas the one from the methanogenic archaeon Methanothermobacter marburgensis appears to be specific for O2, displaying no activity with NO (Seedorf et al., 2004, 2007). FDPs from other bacteria display comparable NO and O2 reductase activities, although with different affinities (Rodrigues et al., 2006; Silaghi-Dumitrescu et al., 2003, 2005a). The different substrate selectivity among FDPs has not been rationalized as yet. The available crystallographic structures (Fraza˜o et al., 2000; Seedorf et al., 2007; Silaghi-Dumitrescu et al., 2005b), although limited, indeed display considerable structural similarities, which have so far prevented the identification of structural features as the basis for the different specificity for O2 or NO.
3. Spectroscopic Measurements These measurements are carried out using a stopped-flow thermostated instrument (DX.17MV, Applied Photophysics, Leatherhead, UK) equipped with either a monochromator (single wavelength mode) or a diode-array detector (multiwavelength mode). The apparatus enables the rapid (1–2 ms) mixing of two solutions into a reaction chamber illuminated by either monochromatic or polychromatic light. In the latter case, a diode array allows rapid collection of absorption spectra as a function of time and over a wide wavelength range (190–730 nm). The Applied Photophysics instrument in the photodiode array mode operates with a minimum
Table 3.1 Nitric oxide and oxygen reductase activities of flavodiiron proteins
Organism
Protein
NO reductase activity (s1)
KM NO (mM)
O2 reductase activity (s1)
KM O2 (mM)
Ref.s
D. gigas
ROO
15
n. d.
> 50
n. d.
(Rodrigues et al., 2006)
Moo. thermoacetica
FDP
48
5
50
26
(Silaghi-Dumitrescu et al., 2003)
Met. marburgensis
FprA
0
0
180
2
(Seedorf et al., 2007)
E. coli
FlRd
15
1.2a
< 1a
n. d.
(Gomes et al., 2002)
D. vulgaris
ROO
12
19
17
24
(Silaghi-Dumitrescu et al., 2005a)
D. – Desulfovibrio; Moo. – Moorella; Met. – Methanothermobacter; E. – Escherichia; a – our own unpublished results.
53
Kinetic Properties of E. coli Flavorubredoxin
acquisition time of 2.56 ms per spectrum and a wavelength resolution of 2.1 nm. Typically, after mixing the two solutions, a large number of spectra (up to 400) are collected according to either a linear or a logarithmic function of time. A limit of the multiwavelength optical measurements is represented by the use of an incident, focused white-light beam, potentially leading to photochemical artifacts. After acquisition, data are transferred to a computer to be analyzed by global analysis algorithms using the software MATLAB (MathWorks, Natick, MA).
3.1. Reduction of flavorubredoxin reductase by NADH The reduction of FAD bound to FlRd-reductase by NADH is associated with an absorption decrease in the 400- to 500-nm range (Fig. 3.3); the reaction involves the transfer of two electrons and, according to our data, proceeds as a single kinetic step with no evidence for flavin radical accumulation. This is possibly consistent with the information that the FADox!FADsq and the FADsq!FADred steps have very close reduction potentials [–220 15 and –260 15 mV, respectively (Vicente and Teixeira, 2005)]. These experiments require strict anaerobic conditions 0.08
Δ A (455 nm)
0.06 Δ Absorbance
0.06
455 nm
0.04
0.04 0.02
2 0.00
0
1 25
50
75
100
Time (ms)
0.02 0.00
−0.02
400
500 l (nm)
600
700
Figure 3.3 Reduction of FlRd-reductase by NADH. Difference spectrum obtained upon reduction of 7.5 mM FlRd-reductase by 16.5 mM NADH. (Inset) Reduction time course of FlRd-reductase by NADH, followed at 455 nm after mixing anaerobically the oxidized protein (15.2 mM) with NADH 20 mM (trace1) or 200 mM (trace 2). At the lower NADH concentration (trace 1), pseudo-first-order conditions are not attained and the experimental time course, clearly deviating from a single exponential, was fitted to a k scheme of the type A þ B ! C, yielding k ¼ 5.5 2.2 106 M1 s1. At the higher NADH concentration (trace 2), as expected, the reaction follows a single exponential time course (pseudo-first-order conditions), proceeding at k0 ¼ 255 17 s1. Buffer: 50 mM Tris-HCl,18% glycerol, pH 8.0.T ¼ 5 C.
54
Joa˜o B. Vicente et al.
and are therefore carried out in the presence of glucose and glucose oxidase and catalase to prevent the reaction of O2 with the FlRd-reductase after reduction by NADH. To measure the second-order rate constant, this reaction was probed by varying the NADH concentration (from 10 to 250 mM) and keeping the concentration of FlRd-reductase constant at 7.5 mM. At [NADH] ¼ 50 mM, for example, at a [NADH]/[FlRd-reductase] ratio <10, the pseudo-first order conditions are not attained; the reaction thus follows a time course clearly deviating from a single exponential (inset to Fig. 3.3, trace 1). Because the NADþ/NADH couple has a redox potential (E ¼ –320 mV) significantly lower than those measured for FlRd-reductase [E1 ¼ –220 15 mV and E2 ¼ –260 15 mV (Vicente and Teixeira, 2005)], the reaction can be modeled according to a bimolecular irreversible scheme of the type k A þ B! C(where A and B denote NADH and oxidized FlRd-reductase and C denotes reduced FlRd-reductase). This analysis allows the secondorder rate constant k ¼ 5.5 2.2 106 M1 s1 to be estimated. At [NADH] 100 mM, i.e., under pseudo-first-order conditions ([NADH]/ [FlRd-reductase] >10), the accumulation of reduced FlRd-reductase within experimental error follows a single exponential time course (inset to Fig. 3.3, trace 2), proceeding at k0 ¼ 255 17 s1 independently of [NADH] (Vicente et al., 2007). The latter observation provides evidence for a limiting rate for intramolecular eT within the NADH/FlRd-reductase complex.
3.2. Reduction of the rubredoxin domain of FlRd by FlRd-reductase As depicted in Fig. 3.1 and according to previous studies (Gomes et al., 2000), FlRd-reductase efficiently shuttles electrons between NADH and the rubredoxin domain of FlRd. To further test this hypothesis, stoppedflow experiments are designed to follow the kinetics of eT between reduced FlRd-reductase and RdT. The latter protein can be suitably investigated by absorption spectroscopy, as it binds an Fe-Cys4 center that in the oxidized state is characterized by a typical absorption spectrum (lmax ¼ 484 nm), completely bleached upon reduction. The rationale of a typical experiment described later is therefore to mix in the stopped-flow instrument oxidized RdT with reduced FlRd-reductase at increasing concentrations and monitor the reduction of RdT by following the absorption decrease at 484 nm. Given the information that FlRd-reductase (but not RdT) is reduced efficiently by NADH (see earlier discussion), conditions in these experiments are chosen so as to keep FlRd-reductase constantly reduced over the time window of the reaction with RdT. Namely, FlRd-reductase (from 0.76 to 52 mM) is prereduced by the addition of a large excess of NADH (750 mM) prior to mixing with oxidized
Kinetic Properties of E. coli Flavorubredoxin
55
RdT (15.4 mM). The chosen NADH concentration is largely saturating and accounts for a fast reduction of FlRd-reductase (k0 ¼ 255 17 s1). Under these conditions, the FlRd-reductase transiently oxidized by RdT after mixing is expected to be quickly rereduced by NADH to keep the concentration of reduced FlRd-reductase essentially constant over the progress of the reaction. This experiment was designed to attain pseudo-first-order conditions, and the reaction is thus expected to follow a single exponential time course, independent of [FlRd-reductase]. Data confirm these expectations in so far as the absorption changes detected at 484 nm followed singleexponential time courses at the different FlRd-reductase concentrations tested (Vicente et al., 2007). The corresponding pseudo-first-order rate constants show a hyperbolic dependence on the concentration of FlRdreductase (Fig. 3.4, top). This behavior allows a kinetic scheme to be proposed (Fig. 3.4, bottom), whereby the following sequence of events occurs: (i) the reduced FlRd-reductase (A*) and oxidized RdT (B) form a complex (A* B) according to a reversible bimolecular process (k1, k1); (ii) within such a complex eT occurs (k2), thus yielding oxidized FlRd-reductase and reduced RdT (A B*); (iii) the two redox partners split apart quickly (k3 >> k2); and (iv) oxidized FlRd-reductase is rereduced by NADH (k4 ¼ 255 s1). As an oversimplification, the intramolecular eT within the complex (A* B ! A B*) was assumed to be an irreversible process, based on the information that the reduction of RdT by FlRd-reductase is largely favored thermodynamically, according to the redox potentials determined by Vicente and Teixeira (2005). Simulations of the model just described were carried out using the software FACSIMILE (AEA Technology, United Kingdom). Using this approach, experimental rates were suitably fitted when assuming the following set of rate constants: k1 ¼ 1.3 107 M1 s1, k1
13 s1, k2 ¼ 300 s1, and k3 5000 s1 (see Fig. 3.4, bottom).
3.3. Reduction of FlRd by FlRd-reductase Using stopped-flow spectroscopy, these experiments aim at measuring the kinetics of eT from FlRd-reductase to the whole, intact FlRd enzyme. These measurements are quite challenging because absorption spectra of the Fe-Cys4 and FMN centers bound to FlRd largely overlap in the UV/visible range. The optical contribution of the sole Fe-Cys4 center of FlRd can be dissected using the truncated Rd domain of the enzyme (RdT); by its subtraction from the overall spectrum of the native enzyme, the spectrum of the flavodiiron domain, largely dominated by the absorption of FMN, can be obtained (Fig. 3.5A). The validity of this experimental approach can be confirmed by expressing and purifying the isolated flavodiiron domain of FlRd, this time genetically truncated from the Rd domain, that indeed displays an absorption spectrum (not shown) very similar to that in Fig. 3.5A
56
Joa˜o B. Vicente et al.
400
k (s−1)
300
200
100
0
0
5
A* + B
10 15 20 [FIRd-reductase] (mM)
k1 k−1
A*− B
k2
25
30
A− B*
k3 k4
A + B* C
Figure 3.4 Reduction of the Rd domain of FlRd by FlRd-reductase. (Top) Rate constants observed upon mixing RdT in the oxidized state with FlRd-reductase at increasing concentrations, prereduced by NADH. Concentrations after mixing: RdT ¼ 7.7 mM; NADH ¼ 375 mM. Buffer: 50 mM Tris-HCl, 18% glycerol, pH 8.0. T ¼ 5 C. (Bottom) Model used to fit data in the top panel. A ¼ FlRd-reductase; B ¼ RdT; C ¼ NADH. The asterisk denotes the reduced state of the proteins. Oxidized RdT and reduced FlRd-reductase form a reversible complex (A*^ B; k1, k1). Afterward, intracomplex eToccurs (k2), followed by dissociation of the partner proteins (k3 k2) and rereduction of oxidized FlRd-reductase by NADH at k4 ¼ 255 s1, as determined independently. Data in the top panel are fitted by assuming k1 1.3 107 M1 s1; k1 ¼13 s1; k2 ¼ 300 s1; k3 5000 s1.
and labeled as ‘‘FDP.’’ Thus, by taking advantage of the expression of the isolated Rd and flavodiiron domains of FlRd, information is acquired on the relative optical contributions of the Fe-Cys4 and the FMN centers in FlRd, information that, as detailed later, proved to be crucial to the investigation of the eT properties of FlRd.
57
Kinetic Properties of E. coli Flavorubredoxin
A
Absorbance
0.12
FIRd
Fe-S
FMN
RdT
0.08
Fe-Fe
Fe-S
FMN
FDP
Fe-Fe
0.04
0.00
B
400
500 l (nm)
600
700
0.04
0.02 Fe-Fe
0.00 60 −0.02
k (s−1)
Δ Absorbance
FMN
−0.04
40 20 0
0
−0.06 400
500
5 10 15 20 25 [FIRd-reductase] (mM) 600
700
l (nm)
Figure 3.5 Flavorubredoxin reduction by FlRd-reductase. (A) Absorption spectrum of 10 mM oxidized flavorubredoxin (FlRd) and its individual domains (RdT and FDP). The spectrum of the flavodiiron (FDP) domain was obtained by subtracting the spectrum of RdT from the spectrum of the whole FlRd protein. (B) Absorption changes assigned to the FMN moiety of FlRd, after mixing the protein with FlRd-reductase prereduced by excess NADH. Concentrations after mixing: [FlRd] ¼ 10 mM; [FlRdreductase] ¼ 2.25 mM; [NADH] ¼ 375 mM. Buffer: 50 mM Tris-HCl,18% glycerol, pH 8. T ¼ 5 C. The optical contribution of the FMN center was obtained as detailed in the text. (Inset) Rate constant at which Fe-Cys4 (circles) and FMN (squares) centers of FlRd are reduced at increasing concentrations of FlRd-reductase.
The kinetics of eT from FlRd-reductase to oxidized FlRd was probed under pseudo-first-order conditions by mixing oxidized FlRd (20 mM) with FlRd-reductase (4.5 mM) prereduced by 750 mM NADH in the
58
Joa˜o B. Vicente et al.
stopped-flow instrument. Under these conditions, both the Fe-Cys4 and the FMN center in FlRd are reduced quickly by FlRd-reductase (<1.5 s), leading to characteristic absorption changes (decrease) in the 400- to 650-nm range. To separate the optical contribution of each cofactor (FeCys4 or FMN), information that at l ¼ 560 nm essentially only the Fe-Cys4 center absorbs is used (see Fig. 3.5A). The reduction time course of the FeCys4 center is therefore deduced from the absorption changes at 560 nm. As expected, this reaction proceeds as a single exponential decay, yielding the pseudo-first-order rate constant k0 ¼ 3.9 s1. From the latter time course, the fraction of oxidized Fe-Cys4 at every time point is estimated, which allows reconstruction of the optical contribution of the latter redox center over the entire wavelength range at every time point. By subtraction, the time-resolved absorption changes of the FMN cofactor can be then estimated, as shown in Fig. 3.5B. The latter data set provides evidence for formation of a red flavin semiquinone (FMNsq: see the absorption increase at 390 nm and the absorption decrease at 450 nm), synchronous to the reduction of the Fe-Cys4 center. If measurements are repeated at higher FlRd-reductase concentrations, the reaction rate increases linearly with [FlRd-reducase], and the reduction of FeCys4 and the formation of FMNsq proceed at identical rates (inset to Fig. 3.5B). The apparent second-order rate constant is estimated by regression analysis of these data, yielding k ¼ 2.4 106 M1 s1. The observation that FMNsq is populated synchronously with the reduction of Fe-Cys4, even at the highest concentration of FlRd-reductase, indicates that Fe-Cys4 and FMN are in rapid redox equilibrium. Although the Fe-Fe site lacks significant UV-visible spectral features, kinetic data collected indirectly allow the suggestion that this site is in rapid redox equilibrium with Fe-Cys4 via FMNsq. Indeed, if this was not the case, only two electrons would quickly equilibrate intramolecularly within FlRd, one at the Fe-Cys4 center and the other at FMNsq. Because both the [Fe3þCys4]/[Fe2þ-Cys4] and the FMNox/FMNsq redox couples have similar redox potentials (Vicente and Teixeira, 2005), in the absence of other effects the observed apparent second-order rate constant for the reduction of RdT (accepting only 1 electron) should be approximately twofold greater than that measured for the reduction of Fe-Cys4 and FMN in the whole enzyme (involving two electrons). Consistent with the hypothesis that electrons entering FlRd quickly also equilibrate with the Fe-Fe site, the second-order rate constants for eT to RdT (1.3 107 M1 s1) are four- to fivefold greater than those measured with FlRd (2.4 106 M1 s1). This section shows how useful it can be to identify, with the aid of molecular genetic tools, the optical fingerprints of different cofactors in a multi-redox center protein (such as FlRd), whose eT properties have to be kinetically characterized by time-resolved spectroscopy.
59
Kinetic Properties of E. coli Flavorubredoxin
3.4. The effect of ionic strength and pH The effect of ionic strength on eT properties of FlRd-reductase, FlRd, or RdT is assessed using the purified proteins desalted previously by gel chromatography and equilibrated with 5 mM Tris-HCl, 18% glycerol, pH 7.6. The FlRd-reductase is used at a fairly low concentration (1 or 4 mM), prereduced by incubation with a large excess of NADH (750 mM). Afterward, this solution is mixed in the stopped flow with either RdT or the intact FlRd enzyme in the oxidized state, and the ionic strength is adjusted by the addition of KCl. Because of the low FlRd-reductase concentration, it is unnecessary to lower the temperature. At 20 C, the reduction of RdT is monitored at 484 nm and that of the intact FlRd enzyme at 474 nm, where the synchronous reduction of Fe-Cys4 and FMN leads to maximal absorption changes (see Fig. 3.5A). As shown in Fig. 3.6, the reaction of FlRd-reductase with both RdT and FlRd is equally affected by the ionic strength, following a bell-shaped dependence. The reaction proceeds rather slowly at very low ionic strength (<5 mM ) and becomes faster on increasing the ionic strength up to m ¼ 40– 50 mM, where maximal rates are observed. A further increase of ionic strength begins to negatively affect the reaction, and eT slows down.
Fe-S 6 k (s−1)
6.8
Fe-S FMN Fe-Fe log k
8
6.6 6.4 6.2 0.0 0.1 0.2 0.3 0.4 0.5 m1/2(M1/2)
4
2
0 0.0
0.1
0.2 m (M)
0.3
0.4
Figure 3.6 Effect of ionic strength on reaction kinetics. Effect of the ionic strength on reduction kinetics of FlRd (circles) or RdT (squares) by FlRd-reductase, prereduced by NADH. Concentrations after mixing: 8.5 mM FlRd, 2 mM FlRd-reductase, 375 mM NADH (circles) or 10.5 mM RdT, 0.5 mM FlRd-reductase, 375 mM NADH (squares). In these experiments, proteins are desalted previously by gel permeation chromatography and the ionic strength is adjusted by the addition of KCl to the buffer (5 mM Tris-HCl,18% glycerol, pH 7.6).T ¼ 20 C. (Inset) Ionic strength dependence of the second-order rate constant observed upon reduction of 10 mM FlRd-reductase by 25 mM NADH.T ¼ 5 C.
60
Joa˜o B. Vicente et al.
The bell-shaped dependence of the apparent rate constants on ionic strength is a common feature for intermolecular eT between proteins that are able to form tight electrostatic complexes (McLendon and Hake, 1992). Data as those in Fig. 3.6 allow one to conclude that the binding of FlRd-reductase to the Rd domain, either isolated or integrated into the whole protein, involves charged residues. In contrast, the effect of ionic strength on the reaction of NADH with oxidized FlRd-reductase, assessed at 5 C, is not bell shaped and is much less pronounced (see inset to Fig. 3.6); this is expected, as NADH is charged only modestly. The rate of reduction of FlRd-reductase by NADH appears to constantly decrease on increasing m (inset to Fig. 3.6). These data allow a more quantitative analysis based on the Broensted–Bjerrum equation:
log k ¼ log k0 þ 2A ZA ZB m1=2
ð3:1Þ
where k is the second-order rate constant for the reaction between the two charged molecules measured at different ionic strengths, k0 is the value of k extrapolated at m ¼ 0, A is a constant approximately equal to 0.49 at 5 C, and ZA and ZB are the charges involved in the reaction. Data in the inset to Fig. 3.6 were thus fitted to Eq. (1), which yielded ZAZB –1.3, pointing to the involvement of just about one elementary charge in the reaction of NADH with FlRd-reductase. The FlRd eT chain in E. coli is located in the cytosol at pH 7.5. Thus, the effect of pH on the kinetics of eT from NADH to FlRd-reductase and from FlRd-reductase to FlRd (or RdT) has to be investigated to assess the physiological relevance of this parameter. To perform the experiments, the purified recombinant proteins are equilibrated by gel chromatography with a solution of low buffering capacity (5 mM Tris-HCl, 18% glycerol, pH 7.6) and mixed in the stopped-flow apparatus with a concentrated buffer (100 mM ) at the desired pH value (from 6.0 to 8.0); the set pH is reached after mixing within the dead time of the instrument. In these experiments the ionic strength is kept constant and equal to 145 mM (after mixing). The rate of the reaction between NADH and FlRd-reductase is independent of pH (between 6.0 and 8.0, not shown). In contrast, the kinetics of eT between FlRd-reductase and FlRd or RdT (monitored at 474 and 484 nm, respectively) is affected equally by pH (Vicente et al., 2007), further supporting the conclusion that the rubredoxin domain of FlRd, when expressed as a truncated domain, retains its structural properties relevant to the reaction with FlRd-reductase. Interestingly, FlRd (likewise RdT) is reduced by FlRd-reductase more rapidly at alkaline pH, yielding an apparent pKa 7.3. This finding leads to the conclusion that, in E. coli, at physiological pH 7.5, eT along the NADH ! FlRd-reductase ! FlRd
Kinetic Properties of E. coli Flavorubredoxin
61
chain proceeds at fairly high rates, with FlRd acting as an efficient NOdetoxifying enzyme.
4. Conclusions Flavodiiron proteins are a suitable system to be investigated by amperometric and time-resolved spectrophotometric techniques. By combining these methodologies with molecular genetic tools, it was possible to acquire knowledge on the role played by these enzymes in the context of the microbial response to nitrosative and oxidative stress. Namely, studies focusing on E. coli flavorubredoxin led to the conclusion that this enzyme has been designed to ensure that the electrons donated by NADH and efficiently shuttled by FlRd-reductase are rapidly transferred intramolecularly to the Fe-Fe active site of FlRd, where NO is promptly reduced to N2O.
ACKNOWLEDGMENTS Work was partially supported by Fundac¸a˜o para a Cieˆncia e Tecnologia of Portugal (Project Grant POCTI/2002/BME/44597 to M.T. and Ph.D. Grant SFRH/BD/9136/2002 to J.B.V.), by MiUR of Italy (PRIN ‘‘Meccanismi molecolari e aspetti fisiopatologici dei sistemi bioenergeticidi membrana’’ to P.S.), and by Consiglio Nazionale delle Ricerche of Italy and Gabinete de Relac¸o˜es Internacionais da Cieˆncia e do Ensino Superior of Portugal (to A.G. and M.T.).
REFERENCES Chen, L., Liu, M. Y., Legall, J., Fareleira, P., Santos, H., and Xavier, A. V. (1993). Purification and characterization of an NADH-rubredoxin oxidoreductase involved in the utilization of oxygen by Desulfovibrio gigas. Eur. J. Biochem. 216, 443–448. da Costa, P. N., Teixeira, M., and Saraiva, L. M. (2003). Regulation of the flavorubredoxin nitric oxide reductase gene in Escherichia coli: Nitrate repression, nitrite induction, and possible post-transcription control. FEMS Microbiol. Lett. 218, 385–393. Frazao, C., Silva, G., Gomes, C. M., Matias, P., Coelho, R., Sieker, L., Macedo, S., Liu, M. Y., Oliveira, S., Teixeira, M., Xavier, A. V., Rodrigues-Pousada, C., et al. (2000). Structure of a dioxygen reduction enzyme from Desulfovibrio gigas. Nat. Struct. Biol. 7, 1041–1045. Gardner, A. M., Helmick, R. A., and Gardner, P. R. (2002). Flavorubredoxin, an inducible catalyst for nitric oxide reduction and detoxification in Escherichia coli. J. Biol. Chem. 277, 8172–8177. Gomes, C. M., Giuffre`, A., Forte, E., Vicente, J. B., Saraiva, L. M., Brunori, M., and Teixeira, M. (2002). A novel type of nitric-oxide reductase: Escherichia coli flavorubredoxin. J. Biol. Chem. 277, 25273–25276.
62
Joa˜o B. Vicente et al.
Gomes, C. M., Silva, G., Oliveira, S., LeGall, J., Liu, M. Y., Xavier, A. V., RodriguesPousada, C., and Teixeira, M. (1997). Studies on the redox centers of the terminal oxidase from Desulfovibrio gigas and evidence for its interaction with rubredoxin. J. Biol. Chem. 272, 22502–22508. Gomes, C. M., Vicente, J. B., Wasserfallen, A., and Teixeira, M. (2000). Spectroscopic studies and characterization of a novel electron-transfer chain from Escherichia coli involving a flavorubredoxin and its flavoprotein reductase partner. Biochem. 39, 16230–16237. Justino, M. C., Vicente, J. B., Teixeira, M., and Saraiva, L. M. (2005). New genes implicated in the protection of anaerobically grown Escherichia coli against nitric oxide. J. Biol. Chem. 280, 2636–2643. McLendon, G., and Hake, R. (1992). Interprotein electron transfer. Chem. Rev. 92, 481–490. Rodrigues, R., Vicente, J. B., Felix, R., Oliveira, S., Teixeira, M., and RodriguesPousada, C. (2006). Desulfovibrio gigas flavodiiron protein affords protection against nitrosative stress in vivo. J. Bacteriol. 188, 2745–2751. Saraiva, L. M., Vicente, J. B., and Teixeira, M. (2004). The role of the flavodiiron proteins in microbial nitric oxide detoxification. Adv. Microb. Physiol. 49, 77–129. Seedorf, H., Dreisbach, A., Hedderich, R., Shima, S., and Thauer, R. K. (2004). F420H2 oxidase (FprA) from Methanobrevibacter arboriphilus, a coenzyme F420-dependent enzyme involved in O2 detoxification. Arch. Microbiol. 182, 126–137. Seedorf, H., Hagemeie, C. H., Shima, S., Thauer, R. K., Warkentin, E., and Ermler, U. (2007). Structure of coenzyme F420H2 oxidase (FprA), a di-iron flavoprotein from methanogenic Archaea catalyzing the reduction of O2 to H2O. FEBS J. 274, 1588–1599. Silaghi-Dumitrescu, R., Coulter, E. D., Das, A., Ljungdahl, L. G., Jameson, G. N., Huynh, B. H., and Kurtz, D. M., Jr. (2003). A flavodiiron protein and high molecular weight rubredoxin from Moorella thermoacetica with nitric oxide reductase activity. Biochemistry 42, 2806–2815. Silaghi-Dumitrescu, R., Ng, K. Y., Viswanathan, R., and Kurtz, D. M., Jr. (2005a). A flavodiiron protein from Desulfovibrio vulgaris with oxidase and nitric oxide reductase activities: Evidence for an in vivo nitric oxide scavenging function. Biochemistry 44, 3572–3579. Silaghi-Dumitrescu, R., Kurtz, D. M., Jr., Ljungdahl, L. G., and Lanzilotta, W. N. (2005b). X-ray crystal structures of Moorella thermoacetica FprA. Novel diiron site structure and mechanistic insights into a scavenging nitric oxide reductase. Biochemistry 44, 6492–501. Stubauer, G., Giuffre`, A., Brunori, M., and Sarti, P. (1998). Cytochrome c oxidase does not catalyze the anaerobic reduction of NO. Biochem. Biophys. Res. Commun. 245, 459–465. Vicente, J. B., Scandurra, F. M., Rodrigues, J. V., Brunori, M., Sarti, P., Teixeira, M., and Giuffre, A. (2007). Kinetics of electron transfer from NADH to the Escherichia coli nitric oxide reductase flavorubredoxin. FEBS J. 274, 677–686. Vicente, J. B., and Teixeira, M. (2005). Redox and spectroscopic properties of the Escherichia coli nitric oxide-detoxifying system involving flavorubredoxin and its NADH-oxidizing redox partner. J. Biol. Chem. 280, 34599–34608. Wasserfallen, A., Ragettli, S., Jouanneau, Y., and Leisinger, T. (1998). A family of flavoproteins in the domains Archaea and Bacteria. Eur. J. Biochem. 254, 325–332. Zumft, W. G. (1997). Cell biology and molecular basis of denitrification. Microbiol. Mol. Biol. Rev. 61, 533–616.
C H A P T E R
F O U R
Escherichia coli Cytochrome c Nitrite Reductase NrfA Thomas A. Clarke,* Paul C. Mills,* Susie R. Poock,* Julea N. Butt,*,† Myles R. Cheesman,† Jeffrey A. Cole,‡ Jay C. D. Hinton,§ Andrew ¨derberg,* M. Hemmings,*,† Gemma Kemp,† Christopher A. G. So k † Stephen Spiro, Jessica Van Wonderen, and David J. Richardson* Contents 1. Introduction 2. Measurement of Cytochrome c Nitrite Reductase-Dependent Consumption of Nitric Oxide in Whole Cells 3. Growth of E. coli Optimized for Cytochrome c Nitrite Reductase Production for Use in Enzyme Purification 4. Purification of Cytochrome c Nitrite Reductase 5. Assaying the Cytochrome c Nitrite Reductase 6. Crystallization of E. coli Cytochrome c Nitrite Reductase 7. Concluding Remarks Acknowledgments References
64 66 66 68 69 73 74 76 76
Abstract The periplasmic cytochrome c nitrite reductase (Nrf ) system of Escherichia coli utilizes nitrite as a respiratory electron acceptor by reducing it to ammonium. Nitric oxide (NO) is a proposed intermediate in this six-electron reduction and NrfA can use exogenous NO as a substrate. This chapter describes the method used to assay Nrf-catalyzed NO reduction in whole cells of E. coli and the procedures for preparing highly purified NrfA suitable for use in kinetic, spectroscopic, voltammetric, and crystallization studies.
* { { } k
Centre for Metalloprotein Spectroscopy and Biology, School of Biological Sciences, University of East Anglia, Norwich, United Kingdom School of Biological Sciences, University of East Anglia, Norwich, United Kingdom School of Biosciences, University of Birmingham, Edgbaston, Birmingham Institute of Food Research, Norwich, United Kingdom Department of Molecular and Cell Biology, University of Texas at Dallas, Richardson, Texas
Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37004-3
#
2008 Elsevier Inc. All rights reserved.
63
64
Thomas A. Clarke et al.
1. Introduction Enteric bacteria such as Escherichia coli and Salmonella typhimurium have evolved to survive in electron acceptor-limited anaerobic conditions. Under anoxic and microoxic conditions in the presence of low levels of nitrate, the periplasmic nitrate reductase (Nap) system and the periplasmic nitrite reductase (Nrf) system are expressed (Potter et al., 2001); these conditions are similar to those found in the mammalian host environment, such as the gut and bloodstream. The NapA enzyme is responsible for the two electron reduction of nitrate to nitrite, while the NrfA enzyme reduces nitrite to ammonium through a six-electron reduction (Eq. [4.1]) proposed to involve bound intermediates of nitric oxide (NO) and NH2OH (Einsle et al., 2002). In E. coli, the reduction of nitrate to ammonia can be coupled to energy-conserving electron transport pathways with formate as an electron donor (Potter et al., 2001). NO−3
NapA
NO−2
NrfA
+
NH4
2e− + 2H+ H2O 6e− + 8H+ 2H2O
ð4:1Þ
The NrfA enzyme is a deca-heme homodimeric molecule with eight bis-His-coordinated hemes (hemes 2 to 5) and two active site hemes (heme 1) coordinated by lysine residues on the proximal side and by a water (hydroxide) molecule or substrate on the distal side (Figs. 4.1A and 4.1B). The proposed electron input site into NrfA is through heme 2 (see Fig. 4.1A), and there are two potential routes for electrons to move to an active site: electrons can either move directly from heme 2 to the nearest active site via heme 3 or across the NrfA dimer interface via heme 5 to the remote active site (Fig. 4.1A). The reduction of nitrite to ammonia requires six electrons, indicating that an additional electron is required by the NrfA monomers, either through a second electron-donating step or from the other NrfA monomer in the dimer. Crystal structures of NrfA are currently available from bacteria Wolinella succinogenes (Einsle et al., 1999), Sulfurospirillum deleyianum (Einsle et al., 2000), E. coli (Bamford et al., 2002 ), Desulfovibrio desufuricans (Cunha et al., 2003), and a NrfAH complex from Desulfovibrio vulgaris (Rodrigues et al., 2006). The active site in all NrfA structures contains five highly conserved residues providing a positive environment around the active site heme and acting as potential proton donors (see Fig. 4.1B). The ability of purified E. coli NrfA to act as an NO reductase was first identified in 1990 (Costa et al., 1990), and its ability to catalyze this reaction in whole cells and confer resistance to NO was demonstrated in 2002 through comparative studies on wild-type and nrf mutant strains of E. coli (Poock et al., 2002).
65
Cytochrome c Nitrite Reductase
A
2 5 3
4
1
B
Ca 2+ Gln263 His264
Arg106 Tyr216
Lys126
Figure 4.1 Molecular structure of cytochrome c nitrite reductase NrfA. (A) The dimeric structure showing the10 bound hemes. Hemes of one of the NrfA monomers are numbered to correlate with the description in the text.The active site heme of the other NrfA monomer is circled. (B) Detail of the active site showing the lysine coordinated to the heme. Figures were prepared using Pymol (Delano Scientific).
66
Thomas A. Clarke et al.
2. Measurement of Cytochrome c Nitrite Reductase-Dependent Consumption of Nitric Oxide in Whole Cells For a rapid assessment of NrfA NO reductase activity in intact cells, 100 ml of E. coli is grown anaerobically in minimal media containing glycerol as the carbon and electron source and nitrate and fumarate as electron acceptors. The cultures are grown in a completely filled bottle until they enter the late exponential phase of growth. The bacterial cells are then centrifuged at 6000g for 10 min, and the cell pellet is resuspended in the growth medium in which they were grown, but without any electron acceptors present. An oxygen- and NO-sensitive amperometric Clarke electrode (Hansatech Instruments) is polarized to detect NO (Field et al., 2007), and 3 ml of concentrated cell culture is loaded into the electrode chamber. The cells are incubated at room temperature until the oxygen present in the chamber is consumed, and saturated NO solution is then added to the chamber from a stock solution (Field et al., 2007) to give a final NO concentration of 100 mM. The rate of NO consumption is then measured from the change in signal over time. NO consumption is measured in nmol NO consumed min1 mg cells1 and the specific activity is highly dependent on the anaerobic growth procedures. For example, it is some 10-fold lower following anaerobic growth in minimal media containing glucose compared to anaerobic growth in the presence of glycerol with nitrate and fumarate (GNF) present as electron acceptors (Fig. 4.2). This pattern mirrors the pattern of nrfA expression in these different growth media (Potter and Cole, 1999) and reflects how the choice of growth medium is critical in assessing the contribution that NrfA makes to NO reduction. Because E. coli has other NO-consuming systems, such as cytoplasmic flavorubredoxin (NorV) and flavohemoglobin (HmpA), it is important to estimate the proportion of activity measured that arises from NrfA. This can be done by conducting identical experiments on a nrfA mutant. Data in Fig. 4.2 are taken from such an experiment in which the nrfA mutant carried a deletion of the nrfA gene.
3. Growth of E. coli Optimized for Cytochrome c Nitrite Reductase Production for Use in Enzyme Purification Production of NrfA in large quantities suitable for crystallographic and spectropotentiometric analysis is complicated by the requirement for multiple cofactors to be inserted into the apoprotein to form the active enzyme.
67
NO consumption (µmol NO min−1 mg cells−1)
Cytochrome c Nitrite Reductase
6
WT E. coli E. coli nrfA::Kan
5 4 3 2 1 0
Bacteria grown on GNF medium Bacteria grown on glucose medium
Figure 4.2 Nitric oxide consumption by WT and nrfA::kan mutant strains of E. coli. Bacteria were grown under anaerobic respiratory conditions using glycerol as the electron donor and carbon source with nitrate and fumarate present as electron acceptors (GNF medium) or under anaerobic fermentative conditions with glucose as the carbon and energy source.
Four of the NrfA hemes are bis-His-coordinated c-type hemes and are incorporated though the standard E. coli cytochrome c biosynthesis pathway, which uses the CXXCH amino acid sequence motif for covalent heme incorporation. However, the active site c-type heme is bound by a unique CXXCK motif, which requires heme lyases encoded by the nrfEFG genes, for heme attachment. For production of active NrfA, the heme lyases from two separate cytochrome systems must therefore be encoded. To date, the most effective method of preparing NrfA protein is from E. coli strain LCB2048. This kanamycin-resistant strain contains lesions in the structural genes for both nitrate reductase A and Z (narA and narZ) and results in optimized expression of the periplasmic nitrate and nitrite reductase system genes of the nap and nrf operons (Potter and Cole, 1999). For optimal production, cells are grown anaerobically at 37 in minimal salts media containing 5% LB, 40 mM fumarate, 0.4% glycerol, and 20 mM nitrate (Potter and Cole, 1999). Initially a 5-ml culture is inoculated with a colony of E. coli strain LCB2048 and grown overnight at 37 anaerobically. This is then transferred to a 200-ml culture and grown overnight at 37 . This 200-ml culture is then used to inoculate a 1.5-liter culture, which in turn is used to inoculate a 100-liter New Brunswick 5000 fermenter, operated for 16 h with an agitation of 200 rpm. Cells are harvested using a CEPA continuous flow centrifuge and stored at –80 until required. Typical cell yields are approximately 150 g. NrfA can also be produced as a recombinant protein by a procedure that has been optimized with E. coli strain JCB4083a (DnarZ::o DnarL::Tn10
68
Thomas A. Clarke et al.
DnapGH DnrfAB DnirBDC::KanR). The JCB4083a strain carries two plasmids: plasmid pEC86 expresses the cytochrome c maturation system (chloramphenicol resistant) and pJG1.9a contains the entire nrf operon (ampicillin resistant). Briefly, E. coli strain JCB4083a is freshly transformed with pEC86, colony purified on chloramphenicol plates, and subjected to a second transformation to introduce plasmid pJG1.9a followed by growth on plates containing ampicillin and chloramphenicol. Colonies of the resulting JCB4083a (pEC86, pJG1.9a) strain are used to inoculate 5 ml Terrific Broth containing ampicillin (100 mg ml1) and chloramphenicol (30 mg ml1) and are grown aerobically at 37 for 8 h. The 5-ml culture is used to inoculate 1 liter of Terrific Broth containing 100 mg ml1 ampicillin and 30 mg ml1 chloramphenicol and is grown aerobically at 37 C overnight. The 1-liter culture is then used to inoculate 90 liters of Terrific Broth in a 100-liter New Brunswick 5000 fermenter, with a controlled oxygen level of 60%. Cells are harvested using a CEPA continuous flow centrifuge and stored at –80 until required. The typical weight of cells harvested using this method is approximately 2 kg.
4. Purification of Cytochrome c Nitrite Reductase The first stage of NrfA purification is to resuspend cells in spheroplasting buffer (0.5 M sucrose, 100 mM Tris-HCl, pH 8.0, 1 mM EDTA) with 100 mg liter1 DNase 1 and 1 g liter1 lysozyme to disrupt the cell wall and release the periplasmic contents. Cells should be stirred for 30 min at 30 to ensure complete cell lysis; the periplasm is then separated from the suspension by centrifugation at 9000g for 1 h, and the periplasm-containing supernatant is retained. Ammonium sulfate is slowly added to the supernatant to give a final concentration of 65% (w/v) and left stirring for 3 h at 4 C. The supernatant is then centrifuged at 30,000g for 1 h to pellet the precipitated protein. The supernatant is discarded and the pellet is resuspended in 50 mM Tris-HCl, pH 8.0, and incubated for 1 h with stirring at 4 . The solubilized NrfA-containing fraction is centrifuged at 30,000g for 1 h at 4 to remove precipitated protein and is then dialyzed extensively with 50 mM Tris-HCl, pH 8.0, at 4 until the conductivity is less than 4 mS. The dialyzed, solubilized, NrfA-containing fraction is loaded onto a 25 2-cm Q-Sepharose column (Amersham) equilibrated with 50 mM Tris, pH 8.0, and washed with 200 ml of the same buffer until the baseline absorbance has stabilized. Proteins are then eluted using a gradient of 0–150 mM NaCl over 200 ml, followed by 150–1000 mM NaCl over 150 ml. At this stage it is also possible to separate NrfB, the electron donor to NrfA (Clarke et al., 2004, 2007), and the periplasmic nitrate reductase
Cytochrome c Nitrite Reductase
69
proteins NapA and NapB for further purification (Jepson et al., 2007). Fractions containing NrfA can be identified using heme-stained gels. Briefly, fractions are loaded onto a 12 % SDS-PAGE gel, and the gel is run as normal. The gel is washed in water and transferred to 70 ml 50 mM Na-acetate buffer at pH 5.0 and left for 15 min in the dark. Methanol (30 ml) containing 5-tetramethylbenzidine (1 mg ml1) is added, and the gel is incubated in the dark for a further 15 min. A 30% (v/v) H2O2 solution (300 ml) is added to initiate the reaction, and the gel is incubated in the dark until bands appear. Fractions containing NrfA can be identified by a band at 53 kDa. Fractions containing NrfA are pooled and concentrated using a 30-kDa Mw cutoff Amicon filter (Millipore). The concentrated NrfA protein is then passed through a Superdex 200 26/60 column (Amersham) equilibrated with 50 mM Tris-HCl, pH 7.0, 50 mM NaCl and fractions containing NrfA (identified using conventional Coomassie-stained SDS-PAGE gels). As a final purification step, NrfA fractions are pooled, dialyzed into 50 mM Tris-HCl, pH 7.0, and the dialyzed NrfA is loaded onto a Dionex anionexchange column equilibrated with 50 mM Tris, pH 7.0. The column is washed with 30 ml buffer, and NrfA is eluted from the column using a 50-ml gradient of 0–200 mM NaCl, 50 mM Tris-HCl, pH 7.0. NrfA fractions that appear as a single band on a Coomassie-stained SDS-PAGE gel are considered to be pure. These fractions are pooled and concentrated before dialysis into 50 mM HEPES, pH 7.0. The purified protein is characterized, in the oxidized state, by an A410/280 ratio of 3.1 (Fig. 4.3A) and has an extinction coefficient of 497 mM1 cm1 at 410 nm (Bamford et al., 2002). The X-band EPR spectrum (see Fig. 4.3B) in the oxidized state has characteristic features at g ¼ 10.5 and 3.6 that arise from the magnetically coupled active site S ¼ 5/2 heme 1 and the nearby S ¼ ½ heme 3 and a rhombic trio at g z,y,x ¼ 2.92, 2.3, 1.52 that arises from the S ¼ ½ heme 2 (see Fig. 4.3B). The g ¼ 3.6 feature has a shoulder at high field (g 3.5) that is a ‘‘Large gmax’’ feature that arises from either (or both) heme 4 or heme 5 (Bamford et al., 2002). Evidence for the integrity of the protein comes from assessing the high-spin heme signal at g ¼ 6. This reflects damaged enzyme and so the intensity of the signal should be low (see Fig. 4.3B). NrfA can be stored in 50 mM HEPES, pH 7.0, in 100-ml aliquots at –80 without a noticeable loss of activity.
5. Assaying the Cytochrome c Nitrite Reductase The conventional method used to measure reduction of the nitrite substrate is to determine the rate of methyl viologen oxidation. Methyl viologen has a redox midpoint potential of –446 mV at 25 (Mayhew, 1978), and the reduced form has a characteristic blue color that has an
70
Thomas A. Clarke et al.
A 1.4 1.2
Absorbance
1.0 0.8 0.6 0.4 0.2 0.0 250
300
350
400
nm
450
500
550
600
B
3.6 2.92
g ~10.8
0
1000
2.3
2000 3000 4000 Magnetic field (gauss)
1.52
5000
6000
Figure 4.3 UV-visible and X-band EPR spectroscopic properties of purified cytochrome c nitrite reductase. (A) UV-visible spectrum of a 2.2 mM NrfA solution (solid line, oxidized; dashed line, reduced). (B) X-band EPR spectrum of 50 mM NrfA collected at 10 K, 2 mW power, and 10 mTmodulation amplitude.
Amax at 600 nm with an extinction coefficient of 13,700 M1cm1. Methyl viologen assays are performed in 1-cm glass cuvettes containing 2 ml of 50 mM HEPES, pH 7.0, 2 mM CaCl2, 1–2 nM NrfA protein, and 0.8 mM methyl viologen. The cuvettes are stoppered using rubber septa and sparged with air-free nitrogen gas for 10 min. Absorbance measurements are performed using a Varian 3100 dual-beam absorbance spectrophotometer at 600 nm. After sparging, the absorbance of the cuvette at 600 nm is used as a baseline, and 1–3 ml of a degassed 10 mM sodium dithionite solution is
71
Cytochrome c Nitrite Reductase
injected through the septa to increase the absorbance of the solution in the cuvette to between 1 and 1.5 AU at 600 nm. The absorbance is recorded for approximately 30 s, and then an aliquot of a 1 M stock NaNO2 solution sparged with nitrogen is added to give the desired NO2 concentration. The rate of the decrease in absorbance is used to determine the rate of oxidation of methyl viologen by NrfA (Fig. 4.4). The activity of a highly purified and fully active preparation of NrfA toward nitrite measured with this assay should be approximately 625 NO2 s1 at 25 at saturating nitrite, and the Km should be in the region of 38 mM. These values are subject to slight variation due to the interference of dithionite and/or dithionite degradation products but are reproducible for a given stock of reagents. NrfA is also expressed by bacteria such as Desulfovibrio desulfuricans in the presence of sulfate, and the expressed NrfA enzymes are known to reduce sulfite (Clarke et al., 2006); sulfite is a product of dithionite oxidation and so, to determine the maximal rate of NrfA-catalyzed nitrite reduction, it is necessary to use an alternative method to produce reduced methyl viologen such as zinc. Granules of zinc must be washed in N2-sparged 1 M HCl for 30 s to remove the external layer of zinc oxide. The granules are then blotted dry before one is added to 100 mM methyl viologen in a glass vessel that is sealed and then made anaerobic by sparging with nitrogen. A color change is immediately apparent. After 5 min, the granule of zinc is removed, and the reduced methyl viologen is sparged with nitrogen to remove any oxygen. Ten microliters of a zinc-reduced methyl viologen solution is added to 2 ml of 50 mM HEPES, pH 7.0, 2 mM CaCl2, 1–2 nM NrfA, and 5 mM EDTA in a N2-sparged sealed cuvette. This produces an absorbance of 2 AU at 600 nm and, on addition of nitrite, a decrease in the absorbance at 600 nm is observed. Using this protocol, the Vmax of NrfA for
Absorbance at 600 nm
A 2
1 mM NO−2
1
0 0
20
40
60
80
100
Time (s)
Figure 4.4 Methyl viologen-based assay for the cytochrome c nitrite reductase NrfA. Methyl viologen oxidation assay: change in absorbance at 600 nm of a 2-ml solution containing 50 mM HEPES, pH 7.0, 2 mM CaCl2,1.6 nM NrfA, and 0.8 mM Na2S2O4 reduced methyl viologen. At the indicated time,1 mM NO2 is injected into the cuvette to initiate NrfA-catalyzed oxidation of methyl viologen.
72
Thomas A. Clarke et al.
nitrite was determined as 769 NO2 s1 and the Km was 22 mM. EDTA is present in this assay because, in control experiments, Zn2þ is found to be an inhibitor of NrfA activity (Fig. 4.5A) but the enzyme is protected by the chelator EDTA (see Fig. 4.5B). Because the EDTA protects NrfA from zinc inhibition, it suggests that, in the concentration range used, it is not disrupting the active site through chelation of the active site calcium ion (see Fig. 4.1B), although this possibility is something to be aware of when adding this chelator. An alternative method of assessing NrfA activity utilizes protein film voltammetry (PFV) in which NrfA is immobilized onto an electrode surface
NrfA activity (e s−1)
A
NrfA reduction (e s−1)
B
4500 4000 3500 3000 2500 2000 1500 1000 500 0 0
0.2
0.4 0.6 ZnCl2 (mM)
0.8
0
1
2 3 EDTA (mM)
4
1
4000 3500 3000 2500 2000 1500 1000 500 0 5
Figure 4.5 Zinc ion inhibition of the cytochrome c nitrite reductase NrfA. (A) NrfA activity measured using dithionite-reduced methyl viologen assays in the presence of increasing concentrations of zinc chloride. Zinc chloride was added to cuvettes immediately prior to the addition of NrfA. The concentration of NrfA in these assays was 1.6 nM, and 1 mM NO 2 was used to initiate the reaction. (B) Activity of 1.6 nM NrfA in the presence of 1 mM zinc chloride and 0^5 mM EDTA as indicated. Reagents were incubated for 5 min before addition of dithionite to reduce methyl viologen. Activity assays were initiated by the addition of 1 mM NO 2 . In all experiments, 0.8 mM methyl viologen was used in 50 mM HEPES, pH 7.0, 2 mM CaCl2.
Cytochrome c Nitrite Reductase
73
that can then serve to donate electrons to the enzyme. This method removes the need for chemical reductants that might inhibit the enzyme or react directly with a substrate such as nitric oxide. PFV is performed in a three-electrode cell configuration with the sample chamber equilibrated at the required temperature (Angove et al., 2002). The cell is placed inside a Faraday cage housed in a N2-filled chamber with atmospheric O2 <2 ppm. Pyrolytic graphite edge working electrodes of 3 mm diameter must be polished immediately prior to use with an aqueous slurry of 0.3 mm Al2O3, sonicated, rinsed, and dried with a tissue. The freshly polished electrodes are then taken into the anaerobic chamber, together with an aliquot of frozen NrfA (routinely around 0.5 to 1 mM). Immediately after the NrfA sample has thawed, 3–5 ml is placed on the electrode surface and after approximately 15–20 s excess solution is removed from the electrode, taking care not to dry the electrode surface. The electrode is then placed in the electrochemical cell and voltammetry commences. Voltammetry is performed with an Autolab electrochemical analyzer under the control of GPES software, and electrode rotation is driven with an EG&G Model 636 electrode rotator. A typical catalytic current collected under substratelimiting conditions with nitrite at pH 7 is shown in Fig. 4.6. The upper part of Fig. 4.6 shows the current measured as a function of potential. The Y axis can be replotted as a derivative of the current with respect to potential (bottom part of Fig. 4.6) to more clearly resolve features in the catalytic waveshape. The positive peak at –105 mV approximates to the midpoint potential of the active site heme (see Fig. 4.1B) measured from the dependence of the g ¼ 3.5 and 10.6 EPR signals on electrochemical potential (Bamford et al., 2002). The negative peak at –300 mV reflects attenuation in NrfA activity at a potential that approximates to the midpoint potential of the species giving rise to the g 3.5 Large gmax signal (see Fig. 4.3B). The magnitude of the catalytic current is dependent on substrate concentration and can be used to determine the Km. In the case of nitrite, this is close to that determined in solution state studies (Angove et al., 2002). The catalytic waveshape is dependent on many factors, including substrate concentration, pH, electrode rotation rate, and scan rate, which must be carefully assessed in any study of the interaction between a substrate or inhibitor and cytochrome c nitrite reductase (Gwyer et al., 2004, 2005, 2006).
6. Crystallization of E. coli Cytochrome c Nitrite Reductase For effective crystallization, NrfA must be concentrated to 10 mg ml1. The crystal screens 1, 2, and 3 (described by Jancarik and Kim, 1991) can be used to identify suitable crystallization conditions for NrfA.
74
Thomas A. Clarke et al.
0 i (mA)
−1
−105 mV 0.5 di/dE (mA V−1) −325 mV 0.0 −0.6 −0.4 −0.2 0.0 0.2 Potential vs SHE (V)
Figure 4.6 Protein filmvoltammetry-based assay for the cytochrome c nitrite reductase NrfA.The assay is performed in the presence of 1.7 mM nitrite, 2 mM CaCl2, and 50 mM HEPES, pH 7.0, with data collected at 20 at a scan rate of 20 mV s1 and an electrode rotation rate of 3000 rpm. (Top) Current measured as a function of potential; (bottom) Y axis is replotted as a derivative of the current with respect to potential.
Hanging drop vapor diffusion methods have yielded red crystals (for an example, see Fig. 4.7) in five conditions that resulted in several different crystal packing motifs (Table 4.1). The red color of the crystal arises from the five hemes bound by each NrfA molecule. The original structure of ˚ through crystal condition type I (Bamford et al., NrfA was obtained at 2.5 A 2002). However, higher-resolution structures to 1.7 A˚ have been obtained using crystal condition type II.
7. Concluding Remarks The cytochrome c nitrite reductase or NrfA from E. coli is a fascinating enzyme because of the complex six-electron reaction it can catalyze and the number of substrates, including nitric oxide, that it can use. The physiological relevance of the NO reduction reaction and the detailed mechanism of
75
Cytochrome c Nitrite Reductase
Figure 4.7 Cytochrome c nitrite reductase crystals. Crystals of NrfA were formed using the vapor diffusion method. NrfA (1 ml) at a concentration of 10 mg ml1 was mixed with 1 ml mother liquor containing 100 mM HEPES, pH 7.5, 20% PEG 10K and incubated in hanging drop trays with a 1-ml well solution at 4. Crystals typically developed in 5^14 days.
Table 4.1 Crystallographic parameters of NrfA crystals
Type Shape
Space group
˚) Cell (A
I
Rectangular plates
P212121
81 90 294
II
Square plates
P21
89 80 142
III
Trapezoid
C222
83 91 362
IV
Cubes
P213
188 188 188
V
Rectangular plates
I222
158 178 264
Conditions
10% isopropanol, 20% PEG 4K, 100 mM HEPES, pH 7.5 20% PEG 10K, 100 mM HEPES, pH 7.5 100 mM MES, pH 6.5, 25% PEG 8K, 0.8 M Na,K tartrate 2 M (NH4)2SO4, 100 mM NH4Ac, pH 4.0 50% K phosphate, pH 7.0
76
Thomas A. Clarke et al.
the enzyme as an NO reductase remain to be determined. However, the assay, purification, and crystallization procedures presented here provide a strong foundation from which to explore studies on the NO reductase activity of native and engineered enzymes in whole cells and as purified enzymes.
ACKNOWLEDGMENTS This work was supported by BBSRC Grant B18695 to DJR, JNB, MRC AMH, and JAC, by the BBSRC core strategic grant to JH, BBSRC studentships to PCM, GK, and SP, and a JIF award (062178). We are grateful to Hayley Angove, Vicki Bamford, Benedicte Burlat, James Gwyer, and Harriet Seward for their past contributions to work on the cytochrome c nitrite reductase
REFERENCES Angove, H., Cole, J. A., Richardson, D. J., and Butt, J. N. (2002). Protein film voltammetry reveals distinctive fingerprints of nitrite and hydroxylamine reduction by a cytochrome c nitrite reductase. J. Biol. Chem. 277, 23374–23381. Bamford, V., Angove, H., Seward, H., Butt, J. N., Cole, J. A., Thomson, A. J., Hemmings, A. M., and Richardson, D. J. (2002). The structure and spectroscopy of the cytochrome c nitrite reductase of Escherichia coli. Biochemistry 41, 2921–2931. Clarke, T. A., Dennison, V., Seward, H. E., Burlat, B., Cole, J. A., Hemmings, A. M., and Richardson, D. J. (2004). Purification and spectropotentiometric characterization of Escherichia coli NrfB, a decaheme homodimer that transfers electrons to the decaheme periplasmic nitrite reductase complex. J. Biol. Chem. 279, 41333–41339. Clarke, T. A., Hemmings, A. M., Burlat, B., Butt, J. N., Cole, J. A., and Richardson, D. J. (2006). Comparison of the structural and kinetic properties of the cytochrome c nitrite reductases from Escherichia coli, Wolinella succinogenes, Sulfurospirillum deleyianum and Desulfovibrio desulfuricans. Biochem. Soc. Trans. 34, 143–145. Clarke, T. C., Cole, J. A., Richardson, D. J., and Hemmings, A. M. (2007). The structure of the pentaheme c-type cytochrome NrfB and characterisation of its solution-state interaction with the pentaheme nitrite reductase NrfA. Biochem. J. (in press). Costa, C., Moura, J. J., Moura, I., Liu, M. Y., Peck, H. D., Jr., LeGall, J., Wang, Y. N., and Huynh, B. H. (1990). Hexaheme nitrite reductase from Desulfovibrio desulfuricans: Mossbauer and EPR characterization of the heme groups. J. Biol. Chem. 265, 14382–14388. Cunha, C. A., Macieira, S., Dias, J. M., Almeida, G., Goncalves, L. L., Costa, L. L., Lampreia, J., Huber, R., Moura, J. J. G., Moura, I., and Romao, M. J. (2003). Cytochrome c nitrite reductase from Desulfovibrio desulfuricans ATCC 27774: The relevance of the two calcium sites in the structure of the catalytic subunit (NrfA). J. Biol. Chem. 278, 17455–17465. Einsle, O., Messerschmidt, A., Huber, R., Kroneck, P. M., and Neese, F. (2002). Mechanism of the six-electron reduction of nitrite to ammonia by cytochrome c nitrite reductase. J. Am. Chem. Soc. 124, 11737–11745. Einsle, O., Messerschmidt, A., Stach, P., Bourenkov, G. P., Bartunik, H. D., Huber, R., and Kroneck, P. M. (1999). Structure of cytochrome c nitrite reductase. Nature 400, 476–480.
Cytochrome c Nitrite Reductase
77
Einsle, O., Stach, P., Messerschmidt, A., Simon, J., Kroger, A., Huber, R., and Kroneck, P. M. (2000). Cytochrome c nitrite reductase from Wolinella succinogenes: Structure at 1.6 A resolution, inhibitor binding, and heme-packing motifs. J. Biol. Chem. 275, 39608–39616. Field, S. J., Thorndycroft, F. H., Matorin, A. D., Richardson, D. J., and Watmough, N. J. (2007). The respiratory nitric oxide reductase (NorBC) from Paracoccus denitrificans. Methods Enzymol. 437[5] (this volume). Gwyer, J., Angove, H., Richardson, D. J., and Butt, J. N. (2004). Redox triggered events in cytochrome c nitrite reductase. Bioelectrochemistry 63, 43–47. Gwyer, J., Richardson, D. J., and Butt, J. N. (2006). Inhibiting Escherichia coli cytochrome c nitrite reductase: Voltammetry reveals an enzyme equipped for action despite the chemical challenges it may face. In vivo. Biochem. Soc. Trans. 34, 133–135. Gwyer, J. D., Richardson, D. J., and Butt, J. N. (2005). Diode or tunnel-diode characteristics? Resolving the catalytic consequences of proton coupled electron transfer in a multi-centered oxidoreductase. J. Am. Chem. Soc. 127, 14964–14965. Jancarik, K., and Kim, S. H. (1991). Sparse matrix sampling: A screening method for crystallization of proteins. J. Appl. Crystallogr. 24, 409–491. Jepson, B., Mohan, S., Clarke, T. A., Gates, A. J., Cole, J. A., Butt, J. N., Hemmings, A. M., and Richardson, D. J. (2007). Spectropotentionmetric and structural characterisation of the periplasmic nitrate reductases of Escherichia coli. J. Biol. Chem. 282, 6425–6437. Mayhew, S. G. (1978). The redox potential of dithionite and SO2 from equilibrium reactions with flavodoxins, methyl viologen and hydrogen plus hydrogenase. Eur. J. Biochem. 85, 535–547. Poock, S.R, Leach, E. R., Moir, J. W. B., Cole, J. A., and Richardson, D. J. (2002). The respiratory detoxification of nitric oxide by Escherichia coli. J. Biol. Chem. 277, 23664–23669. Potter, L., Angove, H., Richardson, D. J., and Cole, J. A. (2001). Nitrate reduction in the periplasm of Gram negative bacteria. Adv. Microbial. Physiol. 45, 52–102. Potter, L. C., and Cole, J. A. (1999). Essential roles for the products of the napABCD genes, but not napFGH, in periplasmic nitrate reduction by Escherichia coli K-12. Biochem. J. 344, 69–76.
This page intentionally left blank
C H A P T E R
F I V E
The Respiratory Nitric Oxide Reductase (NorBC) from Paracoccus denitrificans Sarah J. Field, Faye H. Thorndycroft, Andrey D. Matorin, David J. Richardson, and Nicholas J. Watmough Contents 80 82 85 86 88 91 91 91 94 96 98 99 99
1. 2. 3. 4. 5. 6.
Introduction Purification of Native NorBC from Paracoccus denitrificans Purification of Recombinant NorBC Amperometric Assays of NO Consumption Pseudoazurin as an Electron Donor in Assays of NorBC Preparation of NOR for Spectroscopic Investigation 6.1. Fully oxidized NorBC 6.2. Fully reduced NorBC 6.3. Partially reduced forms of NorBC 7. Electron Paramagnetic Resonance Spectroscopy 8. Concluding Remarks Acknowledgments References
Abstract The two subunit cytochrome bc complex (NorBC) isolated from membranes of the model denitrifying soil bacterium Paracoccus denitrificans is the best characterized example of the bacterial respiratory nitric oxide reductases. These are members of the superfamily of heme-copper oxidases and are characterized by the elemental composition of their active site, which contains nonheme iron rather than copper, at which the reductive coupling of two molecules of nitric oxide to form nitrous oxide is catalyzed. This chapter describes methods for the purification and characterization of both native nitric oxide reductase from P. denitrificans and a recombinant form of the enzyme expressed in Escherichia coli, which enables site-directed mutagenesis of the catalytic subunit NorB. Center for Metalloprotein Spectroscopy and Biology, School of Biological Sciences, University of East Anglia, Norwich, United Kingdom Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37005-5
#
2008 Elsevier Inc. All rights reserved.
79
80
Sarah J. Field et al.
Examples are given of electronic absorption and electron paramagnetic resonance spectra that characterize the enzyme in a number of redox states, along with a method for the routine assay of the complex using its natural electron donor pseudoazurin.
1. Introduction Nitric oxide (NO) is generated as a free intermediate during denitrification, a process in which nitrate is reduced to dinitrogen gas by denitrifying bacteria (Richardson and Watmough, 1999):
NO3- ! NO2- ! NO ! N2 O ! N2 The free NO generated during denitrification is reduced to N2O by the enzyme nitric oxide reductase (NOR) as shown in the following scheme (Richardson and Watmough, 1999; Watmough et al., 1999).
2NO þ 2e þ 2Hþ ! N2 O þ H2 O The product of this reaction, nitrous oxide (N2O), is a greenhouse gas that, along with carbon dioxide and methane, makes a significant contribution to the process of global warming. Although the amounts of N2O emitted into the atmosphere are considerably lower than those of carbon dioxide, the contribution of N2O per unit volume to global warming is 300 times greater because of its persistence in the atmosphere. Worldwide, the largest source of anthropogenic N2O emissions is agriculture (>70%), probably because of the increased application of nitrate-based fertilizers in the past 100 years, which in turn has lead to increased denitrification activity in the subsoil. This is a relatively aerobic environment, which, combined with the sensitivity of the bacterial N2O reductase to oxygen, causes many bacteria to act as partial denitrifiers and form N2O rather than dinitrogen (Takaya et al., 2003). Moreover, fungal members of the soil denitrifying community lack a N2O reductase and can only reduce nitrate to N2O. Genes encoding NOR have been identified in the genomes of many pathogenic bacteria, for example, Brucella suis (Paulsen et al., 2002), Burkholderia mallei (Nierman et al., 2004), Staphylococcus aureus (Gill et al., 2005), and Neisseria meningitides (Tunbridge et al., 2006). This may be related to the host immune response that causes expression of the inducible NO synthetase of macrophages to be increased, which causes a rapid rise in the concentration of NO. Those bacterial species that express a respiratory
Bacterial NorBC
81
NOR can remove this NO from their immediate environment and link this process to respiration. Those species, including enteric bacteria that lack a respiratory NOR, have developed a number of alternative strategies that enable them to evade macrophage NO release. These include the expression of a number of other enzymes, which are discussed in detail elsewhere in this volume and include flavohemoglobin (Stevanin et al., 2002), flavorubedoxin (NorVW) (Gardner et al., 2002), and the multiheme nitrite reductase (NrfA) (Poock et al., 2002). Three types of respiratory NORs have been reported in Eubacteria (Busch et al., 2002; Hendriks et al., 1998a; Suharti et al., 2001; Watmough et al., 1999), but most detailed biochemical and spectroscopic information about NOR have been obtained using the enzyme purified from Paracoccus denitrificans (Girsch and de Vries, 1997; Gro¨nberg et al., 1999; Hendriks et al., 1998b). 0 The reduction of NO to N2O (DEm ¼ þ1.12 V) is energetically more 0 favorable than the reduction of oxygen to water (DEm ¼ þ0.8 V); however, unlike heme-copper oxidases (HCuOs), NOR is not electrogenic and instead of pumping protons across the cytoplasmic membrane, NOR derives its substrate protons from the periplasm (Bell et al., 1992). Nevertheless, NO reduction by NOR can be coupled to energy conservation via NADH dehydrogenase and the bc1 complex from which the electrons required for catalysis are derived. Those NORs that derive their electrons from the bc1 complex via soluble periplasmic electron carriers, usually a cupredoxin or a c-type cytochrome, are usually known as cytochrome c-dependent nitric oxide reductases (cNORs). cNORs represent one of three classes of NOR reported in Eubacteria (Busch et al., 2002; Hendriks et al., 1998a; Suharti et al., 2001; Watmough et al., 1999). Representative cNORs have been purified from several bacterial species, including Pseudomonas stuzeri (Kastrau et al., 1994), Paracoccus halodenitrificans (Sakurai and Sakurai, 1997), and Paracoccus denitrificans (Girsch and de Vries, 1997; Gro¨nberg et al., 1999; Hendriks et al., 1998b). For an up-to-date and comprehensive review of bacterial NORs, the reader is referred to the review by Zumft (2005). Biochemically and spectroscopically, the best characterized cNOR is the NorBC enzyme from P. denitrificans, which is isolated as a two-subunit complex. The catalytic subunit NorB is a 56-kDa integral membrane protein structurally related to the catalytic subunit (subunit I) of HCuOs, with which it shares an overall architecture of 12 transmembrane helices (van der Oost et al., 1994). The active site of NOR is a dinuclear center that comprises high-spin heme b (heme b3) coupled magnetically to a nearby nonheme iron (FeB) center rather than of the heme/CuB center found in the canonical HCuOs (e.g., the mitochondrial cytochrome aa3 type oxidase) (Hendriks et al., 1998b). NorB also contains a magnetically isolated low-spin heme b, which is thought to pass electrons to the active site during catalysis. The six histidine residues responsible for ligating the metal centers in subunit I of HCuOs are completely conserved in NorB (van der Oost et al., 1994).
82
Sarah J. Field et al.
NorC is a 17-kDa protein, which has a periplasmic facing globular region and a single helical transmembrane anchor; it contains a single low-spin heme c, which is the immediate acceptor of electrons from periplasmic electron donors, such as cytochrome c550 or pseudoazurin to the low-spin heme b in NorB (Thorndycroft et al., 2007). Genes encoding NorB and NorC in P. denitrificans are part of the norCBQEDF operon (de Boer et al., 1996). Both norE, which shows some homology to the gene encoding HCuO subunit III, and norF are predicted to code for membrane proteins with five and three membrane spans, respectively. However, neither of these proteins has been detected in NorBC preparations. The norD gene is predicted to code for a 69-kDa cytoplasmic protein, which has a C-terminal van Willebrand factor A metal-binding domain. The product of norQ is homogeneous with CbbQ and contains a putative ATP-binding site (de Boer et al., 1996). The functions of these last two proteins remain uncertain. Representatives of the other two classes of respiratory NOR have been purified. Quinol oxidizing NORs have been isolated from Wautersia eutropha (Cramm et al., 1999) and the archaeon Pyrobaculum aerophilim (de Vries et al., 2003). Both are single subunit enzymes, lacking NorC, and with an N-terminal extension to the catalytic subunit (NorZ), which is believed to harbor the quinol-binding site. A third class of respiratory membrane-bound NOR has been purified from the Gram-positive bacterium Bacillus azotoformans; this protein is of the NorBC form, but has a CuA site instead of a heme in NorC. Interestingly, it also has an N-terminal extension to NorB, which is believed to form a quinol-binding site so that the enzyme can derive electrons from two different sources (Suharti et al., 2001). This chapter deals with the methods developed in our laboratory for purifying and handling the respiratory cNOR found in the model soil denitrifying bacterium P. denitrificans. Native NorBC can be prepared as a very pure two-subunit complex from P. denitrificans cytoplasmic membranes. In addition the P. denitrificans enzyme can be expressed in E. coli, which allows for the rapid screening and subsequent purification of sitedirected mutants. Finally the P. denitrificans NorBC is amenable to the manipulation of its redox state, which facilitates detailed spectroscopic and mechanistic investigations.
2. Purification of Native NorBC from Paracoccus denitrificans NorBC is routinely purified from P. denitrificans strain 93.11 (DctaDI, DctaCII qoxB::kanR). The strain is grown in batch culture under anaerobic denitrifying conditions in a Bioflow 5000 bioreactor (New Brunswick
Bacterial NorBC
83
Scientific) in 90 liters of minimal medium containing 29 mM Na2HPO4, 11 mM KH2PO4, 10 mM NH4Cl, 0.4 mM MgSO47H2O, and 10 mM NaNO3 supplemented with Vishniac and Sante’s trace element solution (2 ml liter1) and kanamycin (25 mg ml1). Cells are harvested with continuous flow centrifugation and stored at –80 until required. A typical culture yields a wet weight of approximately 500 g of cells. The purification of native NorBC from batches of 50 g of cells is based on a method published by Hendriks and colleagues (1998b). Cells are homogenized on ice in 50 mM Tris-HCl, pH 7.6, supplemented with 2.5 mM MgCl2 and 5 mg ml1 DNase to a final volume of 150–200 ml and broken by two passes through a French press pressure cell at 1,100 psi. In all subsequent operations, broken cells are kept on ice. Broken cells are diluted to 360 ml in 50 mM Tris-HCl, pH 7.6, supplemented with 5 mM EDTA and the protease inhibitors 1 mM 4-(2-aminoethyl) benzenesulfonyl fluoride (AEBSF), 1 mM pepstatin, and 1 mM leupeptin. Fractions enriched in cytoplasmic membranes are isolated by ultracentrifugation (180,000 g, for 2 h at 4 ). After ultracentrifugation, tubes contain a pellet, which is composed of three layers: a hard black layer, a very narrow dark red layer, and a bright red soft membranous layer. Chalky layers present in the pellet indicate limited denitrification during growth and yields of NOR will be poor. The supernatant is decanted carefully to prevent soft red membranes containing NorBC from being lost. Typically the red membranes are resuspended in 20–30 ml 20 mM Tris-HCl, pH 7.6, and stored at –80 until required. For purification of NorBC, the frozen membranes are thawed and diluted to 360 ml in 20 mM Tris-HCl, 500 mM NaCl, and 0.5 mM EDTA, pH 7.6. To this suspension, 1 ml of a 10% sodium deoxycholate (DOC) solution is added drop-wise with stirring at 4 . The suspension is left to stir at 4 for a further 5 min and the membranes are sedimented by ultracentrifugation (180,000 g, for 2 h at 4 ). This time the pellet contains two red layers with a very small black layer at the bottom. Both red layers are collected and resuspended in 360 ml 20 mM Tris-HCl and 1 mM EDTA, pH 7.6, to remove excess NaCl and DOC. The membranes are again harvested by ultracentrifugation (180,000 g, for 2 h at 4 ) and resuspended to 20 mg/ml total membrane protein, as determined by the bicinchoninic acid method with bovine serum albumin as a standard, in 20 mM Tris-HCl, 50 mM NaCl, 1 mM EDTA, and 0.5 mM AEBSF, pH 7.6. Solubilization of membrane proteins is achieved by stirring overnight at 4 with dodecyl-b-D-maltoside (DDM) at a ratio of 1:1 DDM:total membrane protein. Unsolubilized material is removed by ultracentrifugation (180,000 g, for 2 h at 4 ) and the red supernatant is diluted twofold with 20 mM Tris-HCl, pH 7.6, and loaded onto a 100-ml Q-Sepharose column preequilibrated in 20 mM Tris-HCl, 0.1 mM EDTA, and 0.02% DDM, pH 7.6 (buffer A).
84
Sarah J. Field et al.
The column is washed with 1.5 column volumes of buffer A. Protein is eluted with the following gradient: 0–20% 20 mM Tris-HCl, 1 M NaCl, and 0.02% DDM, pH 7.6 (buffer B), over 100 ml, 20–50% buffer B over 1,200 ml. A flow rate of 5 ml min1 is used throughout, and 10-ml fractions are collected. NorBC elutes at approximately 450 mM NaCl. Fractions containing significant amounts of NorBC are identified by their characteristic ultraviolet (UV)-visible spectrum and pooled. A distinctive feature at 595 nm, together with bands at 550, 520, and 410 nm indicative of heme proteins, is diagnostic of NorBC (Fig. 5.1). The pooled fractions are loaded onto a 5-ml HiTrap chelating Sepharose column charged with 50 mM copper sulfate and preequilibrated in 20 mM Tris-HCl, 50 mM NaCl, 0.1 mM imidazole, and 0.02% DDM, pH 7.6 (buffer C). NorBC is eluted with the following gradient: 0–5% 20 mM Tris-HCl, 50 mM NaCl, 50 mM imidazole, and 0.02% DDM, pH 7.6 (buffer D), over 25 ml, 5% buffer D over 50 ml, 5–100% buffer D over 10 ml, 100% buffer D over 10 ml. A flow rate of 1 ml min1 is used throughout, and 2-ml fractions are collected. NorBC elutes in a tight peak at approximately 40 mM imidazole. Fractions containing NorBC are pooled and exchanged into 20 mM Bis-Tris propane (BTP) and 50 mM NaCl, pH 7.6. At this stage the A410:A280 ratio is typically between 0.95 and 1.05, and the concentration of NorBC can be determined using a molar extinction coefficient at 410 nm of e410 ¼ 3.11 105 M1 cm1.
300
e mM −1 cm−1
250 200 150 100 50 0 300
350
400 450 500 Wavelength (nm)
550
600
650
Figure 5.1 The electronic absorption spectrum of purified native NorBC from P. denitrificans.The spectrum of fully oxidized as-isolated NOR (70 mM) in 20 mM TrisHCl buffer, 0.1 mM EDTA, and 0.02% DDM. pH 7.6 was recorded in a 1-mm path length quartz cuvette. The ratioA410:A280 ¼1.19 indicates that the enzyme is >95% pure.
85
Bacterial NorBC
If further purification of NorBC is required, then the sample can be exchanged into buffer A and applied to a Q-Sepharose column (0.98 ml) equilibrated in the same buffer. The column is first washed with 5 ml of buffer A before any weakly bound proteins are removed with a gradient of 0–20% buffer B over 5 column volumes. Finally NorBC is eluted by applying a gradient of 20–50% buffer B over 30 column volumes. The flow rate is maintained at 0.3 ml min1 throughout, and 0.5-ml fractions are collected. Fractions containing NorBC with an A410:A280 ratio of >1.15 are pooled and concentrated.
3. Purification of Recombinant NorBC Recombinant P. denitrificans NorBC is purified from cultures of E. coli JM109 that have been cotransformed with the pEC86 plasmid, which allows constitutive expression of the cytochrome c maturation genes required for NorC assembly (Arslan et al., 1998; Tho¨ny-Meyer et al., 1996) and the pNOREX plasmid (Butland et al., 2001). Methods of cell culture and purification are essentially as described by Butland et al. (2001) with some minor modifications. For large-scale preparations of recombinant NorBC, up to 20 cultures (800 ml of Terrific Broth in a 2-liter unbaffled conical flask) are used. Each culture is supplemented with the required antibiotics (100 mg ml1 ampicillin and 36 mg ml1 chloramphenicol) and inoculated with 15 ml of a 50-ml starter grown overnight in LB. After inoculation, the 800-ml cultures are grown at 37 on a rotary shaker (200 rpm) until A600 ¼ 0.4, at which point isopropyl-b-D-thiogalactoside is added to a final concentration of 1 mM in order to induce expression of the nor genes. After induction the cultures are grown according to the schedule outlined in Table 5.1. Cells are subsequently harvested (6900 g for 15 min at 4 ), frozen in liquid nitrogen, and stored at –80 until needed. Cells are thawed and resuspended in 100 mM Tris-HCl, 50 mM NaCl, and 1 mM EDTA, pH 7.6, and broken by two passages through a French pressure cell at 11,000 psi. Protease inhibitors (5 mM EDTA, 1 mM AEBSF 1 mM pepstatin, and 1 mM leupeptin) are added and membranes
Table 5.1 Batch culture growth conditions for E. coli expressing recombinant NorBC Time after induction (hours)
Shaking at 180 rpm
Temperature ( )
0–5 5–10 10–18
Yes No Yes
30 18 30
86
Sarah J. Field et al.
collected by ultracentrifugation (180,000 g, 1 h, 4 ). Membranes are resuspended in the same resuspension buffer and reharvested by ultracentrifugation (180,000 g, for 60 min at 4 ). The washed membranes are again resuspended in the same buffer, frozen rapidly in liquid nitrogen, and stored at –80 until needed. Membranes are thawed and solubilized using the same protocol described earlier for the solubilization of P. denitrificans membranes. Unsolubilized material is removed by ultracentrifugation (180,000 g, for 60 min at 4 ). The supernatant is diluted threefold in 50 mM Tris-HCl and 0.05% DDM, pH 7.6 (buffer A), and loaded onto a 100-ml Q-Sepharose column preequilibrated in buffer A. The column is then washed with buffer A until all unbound protein has washed through the column as judged by A280 of the eluent returning to zero. Other unwanted proteins are removed by exposing the column to the following gradient: 0–10% 50 mM Tris-HCl, 1 M NaCl, and 0.05% DDM (buffer B) over 200 ml. The column is then washed with 10% buffer B until the A280 of the eluent returns to zero. Finally, NorBC is eluted with a gradient of 10–55% buffer B over 1000 ml. Recombinant NorBC elutes at approximately 450 mM NaCl. A flow rate of 5 ml min1 is used throughout, and 10-ml fractions are collected routinely. This method also provides the basis for the study of wild-type and mutant forms of NorBC that have been engineered in a proton-conducting E-channel that leads from the periplasm to the active site of the enzyme (Butland et al., 2001; Flock et al., 2005; Thorndycroft et al., 2007).
4. Amperometric Assays of NO Consumption The consumption of NO by both membrane vesicles and purified NorBC can be measured under anaerobic conditions using a Clark-type oxygen electrode (Oxytherm, Hansatech Instruments, Kings Lynn, UK) that has been modified to allow the platinum cathode to be polarized at –0.8 V with respect to the Ag:AgCl reference electrode, which increases its sensitivity toward NO. The electrolyte is 1 M KCl, and a polytetrafluoroethylene membrane (12.5 mM thickness) is used as a gas-permeable barrier to protect electrodes from the reactants in the chamber. Stock solutions of NO are prepared by the addition of 8 ml NO gas to 3 ml, nitrogen-sparged, water, pH 3.0. The concentration of NO in the stock solution can be checked readily by titrating it into a solution of reduced myoglobin of known concentration and monitoring the absorption changes in the Soret region (e421 ¼ 114 mM1 cm1). The reaction chamber of the electrode is filled with 2 ml of 20 mM phosphate buffer, pH 7.6, warmed to 30 while stirring, and allowed to equilibrate for 5 min. Oxygen is removed from the chamber by the glucose
87
Bacterial NorBC
(16 mM )/glucose oxidase (4 U/ml)/catalase (20 U/ml) system (Fig. 5.2). Once anaerobic, 25 ml of a saturated NO solution (45 nmol) is added to the chamber. The electrode is allowed to equilibrate and three further aliquots are added. The response of the electrode should be directly related to the amount of NO, and this calibration should be carried out each time the electrode is used. If the response is too small, then the NO solution should be discarded and a fresh solution made. This calibration routine is especially important with the Oxytherm electrode, which is interfaced to a software program that converts the electrode current into the amount of oxygen present in solution. The present version of the software does not allow any other output. Consequently, when the electrode responds to the addition of NO under anaerobic conditions, the output given is in ‘‘nmol O2.’’ The calibration routine allows derivation of a conversion factor so that the amount of NO in solution can be calculated readily from the apparent amount of oxygen reported by the software.
75 65
DM Glucose Glucose oxidase
Paz
Electrode response (arbitary units)
55
25ul saturated NO
NOR
45 Background rate
35
Total rate
25 15 5 −5
Catalase Ascorbic Acid
−15 −25 0.0
6
12
Time (min)
18
21
Min 30.
Figure 5.2 NO reduction by purified NorBC. The assay was carried out in 20 mM phosphate buffer, 50 mM NaCl, and 0.02% DDM, pH 7.0, in a stirred reaction chamber thermostated at 30.The vessel was made anaerobic by the addition of glucose (16 mM ), glucose oxidase (4 U ml1), and catalase (20 U ml1). Subsequent additions to the reaction mixture were as follow: ascorbic acid (5 mM), saturated NO solution (4 45 nmol), and pseudoazurin (20 mM). The reaction was initiated by the addition of purified NorBC (10 nM ). Rates of NO consumption were determined by the time taken for the NO signal at the start of the enzyme-catalyzed reaction to reduce by 20%.
88
Sarah J. Field et al.
5. Pseudoazurin as an Electron Donor in Assays of NorBC One problem in assaying NOR is that although NO reductase activity can be measured easily using an amperometric steady-state assay, conditions reported in the literature have used different mediators either singly or in combination with the chemical electron donor ascorbate. For example, phenazine methosulfate (PMS) (Dermastia et al., 1991), N,N,N’,N’-tetramethyl-p-phenylenediamine (TMPD) (Hendriks et al., 1998b), and a combination of horse heart cytochrome c plus TMPD (Matsuda et al., 2002) or 2,3,5,6-tetramethylphenylenediamine (Carr and Ferguson, 1990) have all been used to mediate electron transfer to NOR. No single definitive method for assaying NOR using a physiologically relevant electron donor has been reported, which led us to devise a reliable steady-state assay of NOR activity using a physiologically relevant species, pseudoazurin, as the immediate electron donor. Figure 5.3 shows a representative trace of the consumption of NO by resuspended P. denitrificans membrane vesicles. The vesicles (100 ml containing
30 20
NO
40
NO
50
Succinate
Arbitary units
60
NO
70
Pseudoazurin
P.denitrificans membranes
80
10 0 0
2
4
6 8 Time (min)
10
12
Figure 5.3 Succinate-dependent NO consumption by P. denitrificans membrane vesicles. The reaction was carried out using 100 ml of membrane vesicles in nitrate-free succinate medium, succinate (30 mM ), Na2HPO4 (55 mM ), KH2PO4 (11 mM ), NH4Cl (6 mM ), MgSO4 (0.4 mM ), and Vishniac and Sante’s trace element solution (2 ml l1), pH 7.6, in a stirred reaction chamber thermostated to 30.The vessel was made anaerobic by the addition of glucose (16 mM ), glucose oxidase (4 U ml1), and catalase (20 U ml1). The reaction was initiated by the addition of saturated NO solution (90 mM) in the absence or presence of additional pseudoazurin (60 mM ).
Bacterial NorBC
89
approximately 0.2 mg total protein) are added to 2 ml of nitrate-free succinate medium, succinate (30 mM ), Na2HPO4 (55 mM ), KH2PO4 (11 mM ), NH4Cl (6 mM ), MgSO4 (0.4 mM ), and Vishniac and Sante’s trace element solution (2 ml liter1), pH 7.6, in a stirred reaction chamber thermostated at 30 . The lid is secured, and oxygen is removed from the chamber by the glucose (16 mM)/glucose oxidase (4 U ml1)/catalase (20 U ml1) system. Once the chamber becomes anaerobic, 90 mM NO is added. There is rapid consumption of NO because residual pseudoazurin and cytochrome c550 remain associated with the membrane after cell breakage and are able to shuttle electrons, initially donated to the electron transfer chain by the oxidation of succinate, from the cytochrome bc1 complex to NorC. The addition of nitrogen-sparged pseudoazurin (60 mM) increases the rate of NO reduction significantly, implying both that the amount of electron carriers remaining after membrane preparation are not sufficient to maintain optimal rates for NO-dependent respiration and that pseudoazurin has a physiological role in donating electrons to NOR. These data are consistent with other experiments showing that the ability of NOR to reduce nitric oxide is attenuated in strains unable to express either pseudoazurin of cytochrome c550 (Pearson et al., 2003; Thorndycroft et al., 2007). Recombinant pseudoazurin is used routinely in our laboratory as the immediate electron donor in assays of both native and recombinant NorBC in membrane fractions and after purification. To obtain recombinant pseudoazurin for these experiments, the gene encoding pseudoazurin in Paracoccus pantotrophus, pazS, was amplified and cloned into the pET-24d expression vector exactly as described by Pauleta and colleagues (2004). The resulting vector (pET-psaz) is used to transform E. coli BL21(DE3), which is grown aerobically at 37 in LB medium supplemented with kanamycin (30 mg ml1) and CuSO4 (0.5 mM ), for 24 h, without induction. Cells are harvested by sedimentation (5500 g for 20 minutes) and used immediately for the purification of pseudoazurin using a previously published method (Leung et al., 1997). Cells from 12 1-liter cultures cells are resuspended in a volume of 100 mM Tris-HCl, pH 7.3, equivalent to eight times their wet weight before being broken by five cycles of freezing, in a dry-ice ethanol bath, and thawing. Broken cells are sedimented by centrifugation (10,000 g, 30 min, at 4 ) and the supernatant, which contains the periplasmic proteins, including pseudoazurin, is recovered. A few crystals of potassium ferricyanide are added to the periplasmic fraction to oxidize the sample before it is applied directly to a 750-ml DEAE-Sepharose fast-flow column equilibrated with 50 mM Tris-HCl, pH 7.3. The column is developed with a gradient of 0–400 mM NaCl over 5 column volumes. The blue fractions that contain pseudoazurin are pooled, and ammonium sulfate is added to 80% saturation (561 g liter1). Precipitated proteins are removed by centrifugation, and the blue supernatant is applied to a 25-ml 15PHE-Sepharose
90
Sarah J. Field et al.
column equilibrated with 80% NH4SO4 in 50 mM Tris-HCl, pH 8.0. The column is first washed with 3 column volumes of the same buffer before the column is developed using a gradient of 80–0% NH4SO4 over 10 column volumes at a flow rate of 2.0 ml min1. Fractions containing pseudoazurin (which elutes at about 30% NH4SO4) are pooled and exchanged to 20 mM potassium phosphate, pH 8.0, and the protein is concentrated. The concentration of the pseudoazurin solution can be determined using an E ¼ 1300 M1 cm1 at 590 nm. Pseudoazurin is also used as an electron mediator in the amperometric assay of purified NorBC, which is carried out in anaerobic potassium phosphate buffer supplemented with 0.02% DDM in a stirred reaction chamber thermostated at 30 . After the chamber has been made anaerobic with glucose/glucose oxidase/catalase as described earlier, ascorbic acid is added to a final concentration of 5 mM (see Fig. 5.2). Ascorbic acid is able to reduce pseudoazurin rapidly and hence acts as a source of electrons for the reaction. The trace is allowed to stabilize before the NO is added to a final concentration of 90 mM. Once the trace becomes stable, pseudoazurin is added and a small reduction in the NO signal is seen due to minute quantities of oxygen present in the pseudoazurin solution reacting with the NO (see Fig. 5.2). Once a background rate of NO consumption is established, the reaction is initiated by the addition of purified NorBC. The representative progress curve describing the NorBC catalyzed consumption of NO shown in Fig. 5.2 shows how the rate of NO consumption increases toward the end of the curve as a consequence of the relief of substrate inhibition (Girsch and de Vries, 1997). To compare the rates of reaction between samples of NOR at a constant concentration of NO, the rate of reaction is calculated by the time taken for 20% of the initial NO present in the system to be consumed. Using this approach, we have determined that for native NorBC the optimum pH for the reaction is pH 6 with an apparent Vmax of 904 min1 and an apparent Km for pseudoazurin of 20 mM (Thorndycroft et al., 2007). This assay also allows us to measure the activity of recombinant NorBC expressed in membranes of E. coli, which allows site-directed mutants to be screened rapidly and reliably for changes in catalytic activity (Thorndycroft et al., 2007). This is possible because E. coli lacks a cytochrome bc1 complex so that there is no input of electrons to NorBC from the E. coli respiratory chain and the rate of NO consumption measured depends solely on the exogenous electron donors. This system has been used to show that mutations in two conserved glutamate residues, E122 and E125, which lie in a periplasmic loop between helix III and helix IV of NorB, lead to changes in steady-state activity that are consistent with them forming the entrance to a channel that might couple the movement of protons from the periplasm to the active site with internal electron transfer during turnover (Thorndycroft et al., 2007).
Bacterial NorBC
91
6. Preparation of NOR for Spectroscopic Investigation NorBC has four distinct redox centers and, in principle, can be studied at five different levels of reduction. In the fully oxidized state, the UVvisible spectrum of NorBC, as isolated, is characterized by a ligand-to-metal charge transfer (LMCT) absorption band at 595 nm. This distinct feature is thought to arise from the active site ferric heme b3, which is linked to FeB by a m-oxo bridge. At present, it is not known to which redox state of the enzyme nitric oxide binds during the catalytic cycle. In order to address this problem, we have developed methods for preparing NorBC in different redox states, which are described next.
6.1. Fully oxidized NorBC NorBC has been studied in three different redox states by both UV/visible and electron paramagnetic resonance (EPR) spectroscopies. The spectrum of the fully oxidized, as prepared form of the native enzyme from P. denitrificans, is shown in Fig. 5.1. The spectrum is typical of a hemoprotein and is characterized by expected bands in the a,b (550 and 523 nm) and Soret (410 nm) regions of the spectrum. In addition, there is a weak LMCT band at 595 nm, which is diagnostic of the m-oxo bridge that forms between heme b3 and FeB in the active site (Field et al., 2002; Gro¨nberg et al., 1999). The form of the UV/visible spectrum of fully oxidized NorBC as isolated is independent of pH, and the presence of the m-oxo bridge prevents the binding of small anionic ligands such as cyanide and fluoride ions to the ferric heme b3 (Gro¨nberg et al., 2004).
6.2. Fully reduced NorBC Figures 5.4A and 5.4B show the visible spectrum of samples of fully reduced NorBC prepared for EPR spectroscopy. This technique requires samples in the concentration range of 80–150 mM and consequently only the a,b region of the spectrum lies within the dynamic range of a typical UV/visible spectrophotometer. Because of the inherent oxidase activity of NorBC, fully and partially reduced samples must be prepared under highly anaerobic conditions, which are achieved by placing the enzyme solution, cuvettes, EPR tubes, and syringes in an anaerobic glove box ( Belle Technologies) for at least 30 min before beginning an experiment. Reduced and partially reduced samples are made routinely by the titration of NOR with europium dichloride (EuCl2). EuCl2 has a midpoint potential of approximately –400 mV depending on conditions; when complexed 1:1 with EGTA,
92
Sarah J. Field et al.
A
B 65
65
60
60 OH2
55
Fe
e mM−1 cm−1
50
FeB
Fe
FeB
50
45
45
40
40
35
35
30
30
25
25
20
20
15
15
10
10
5
5
0 500
OH
55
0 550
600
650
700
500
550
600
650
700
Wavelength (nm)
Figure 5.4 Visible region electronic absorption spectra of fully and partially reduced forms of native NorBC from P. denitrificans. Electronic absorption spectra in the 500- to 700-nm region were recorded for NorBC in fully oxidized (solid line),‘‘three electronreduced’’ (dashed line), and fully reduced (dotted line) states. Samples were prepared in 20 mM Bis-Tris propane, 50 mM NaCl, and 0.02% DDM, pH 6 (A), and pH 8.5 (B). Levels of reduction indicated were achieved by slow titration of EuCl2:EGTA. (Inset) The proximal ligand to the active site heme b3 at each pH.
this drops to approx –1 V, allowing stochiometric reduction of proteins ( Vincent et al., 2003). EuCl2 is extremely oxygen sensitive and stock 100 mM solutions are prepared in anaerobic water inside an anaerobic glove box. Solutions (1 ml) of EuCl2 are prepared at the desired concentration by mixing with anaerobic EGTA (450 mM stock in 1 M NaOH) and dilution. Fresh EuCl2/EGTA solutions must be made every few hours. Protein samples for EPR are prepared in 1-mm path length cuvettes at a concentration of 80–150 mM. The EuCl2/EGTA is added to the enzyme solution in sealed cuvettes inside the glove box, and the extent of reduction of NorBC is monitored by UV/visible spectroscopy until the required partially or fully reduced state is reached. The UV/visible spectrum of fully reduced NorBC is essentially the same irrespective of pH and exhibits intense features at 525 and 550 nm that are characteristic of the a,b bands of reduced heme c in NorC with a shoulder at 560 nm arising from the reduced low-spin heme b in NorB.
93
Bacterial NorBC
Carbon monoxide (CO) has been used extensively in the study of HCuOs because of the photolability of the bond between ferrous active site heme and CO. The recombination of photolyzed CO to the heme iron is influenced by a number of factors; therefore, this technique can give information about the effective pKa of the proximal ligand to the heme, the polarity of the heme distal pocket, and steric factors influencing binding of the substrate. Fully reduced NorBC is prepared by the addition of sodium ascorbate (5 mM ) and phenazine methosulfate (1 mM) (ascorbate/PMS) to yield a spectrum similar to that seen in Fig. 5.4. The effect of adding 1 mM CO to NorBC prepared in this way is shown in Fig. 5.5. While there is no change in wavelength maximum of the Soret band (420 nm) in response to the addition of CO, it does intensify and become narrower: e420 ffi 480 mM1 cm1 for the CO-bound form compared to e420 ffi 370 mM1 cm1 for the ascorbate/PMS-reduced form. This form of the enzyme has been used successfully to investigate the dynamics of CO recombination to the active site (Hendriks et al., 2001) and also as the starting point for ‘‘flow-flash’’ experiments to examine proton-coupled electron transfer monitored either electrochemically (Hendriks et al., 2002) or optically (Flock et al., 2005).
500 450 400 e mM−1 cm−1
350 300 250 200 150 100 50 0 400
500 600 Wavelength (nm)
700
Figure 5.5 CO binding to fully reduced native P. denitrificans NorBC. The electronic absorption spectrum (400^700 nm) was recorded of the fully oxidized form NorBC as prepared (solid line). Reduction of NorBC was achieved by the addition of sodium ascorbate (5 mM ) and phenazine methosulfate (1 mM) (dashed line). Sufficient saturated CO solution (1 mM ) was added anaerobically to the sample of reduced NorBC to give a final concentration of 0.1 mM CO in the cuvette (dotted line). All samples were prepared anaerobically in 20 mM BTP, 50 mM NaCl, and 0.02% DDM, pH 7.6.
94
Sarah J. Field et al.
6.3. Partially reduced forms of NorBC Visible region electronic absorption spectra of the partially reduced form of NorBC referred to previously as ‘‘three electron-reduced’’ species are shown in Fig. 5.4. It can be seen that the low-spin hemes b and c are reduced substantially, as indicated by the increased intensity of the bands at 550 and 560 nm compared to the fully oxidized enzyme. In addition, the LMCT band seen at 595 nm in the fully oxidized enzyme shifts, although the position and the intensity of the shifted band are critically dependent on the pH of the buffer (Field et al., 2002). At pH 8.5, the distal ligand to heme b3 is a hydroxide, which gives rise to a LMCT band at 605 nm (see Fig. 5.4B). Decreasing the pH leads to protonation of the hydroxide and a shift in the LMCT band to 630 nm, as shown in Fig. 5.4A. The initial shift in the LMCT band from 595 nm associated with partial reduction is because of the breaking of the m-oxo bridge at the active site, which opens it to the binding of exogenous ligands to heme b3 (Gro¨nberg et al., 2004). The trigger for this breaking of the m-oxo bridge was thought to be because of the reduction of the nonheme iron FeB (Gro¨nberg et al., 1999). However, more recent work in our laboratory suggests that the trigger for breaking the m-oxo bridge may, in fact, be reduction of the low-spin heme b (S. J. Field, unpublished data). Visible region electronic absorption spectra also suggest that there is an increase in the amount of reduced heme in the fully reduced form of the enzyme compared to the ‘‘three electron-reduced’’ samples (see Fig. 5.4). A trivial explanation might be that when NorBC is poised in a partially reduced form, some of the low-spin heme b remains oxidized. However, this would also have to be true for the low-spin heme c, as the intensity of the 550-nm band also increases as the protein becomes fully reduced (see Fig. 5.4). Because the increase in absorbance associated with complete reduction affects both 550- and 560-nm bands, it is most likely that it is due to an underlying and poorly resolved feature, which is likely to be the broad transition associated with reduction of the low-potential heme b3 (see Figs. 5.4A and 5.4B). These characteristic spectroscopic changes have been exploited to determine the redox potential of the metal centers in NorBC in mediated potentiometic titration using sodium dithionite as a reductant (Gro¨nberg et al., 1999). As reductant is introduced into the system, the low-spin b- and c-type hemes start to reduce, as indicated by the increase in absorbance at 550 and 560 nm. There is also a shift in the position of the CT band from 595 nm described earlier (Fig. 5.6). Reduction potentials of low-spin heme c and low-spin heme b have been estimated to be 310 and 345 mV, respectively (Gro¨nberg et al., 1999). A plot of change of absorbance at 595 nm against potential gives rise to a curve that can fitted to two independent 0 0 n ¼ 1 oxidation/reduction events (Em ¼ þ320 mV and Em ¼ þ60 mV).
95
Bacterial NorBC
0.55 0.50 0.45
Absorbance
0.40 0.35 0.30 0.25 0.20 0.15 0.10 0.05 0.00 500
600 650 550 Wavelength (nm)
700
Figure 5.6 Representative spectra recorded during a mediated potentiometric titration of native P. denitrificans NorBC. NOR (10 mM) in 20 mM Tris-HCl, 0.02% DDM, 340 mM NaCl, and 0.5 mM EDTA, pH 7.6, was reduced by additions of dithionite in the presence of 20 mM of each of the following mediators: phenazine methosulfate (þ80 mV), phenazine ethosulfate (þ55 mV), diaminodurene (þ250 mV), juglone (þ30 mV), 5-anthroquinone-2-sulfonate (^225 mV), 6-anthroquinone-2,6-disulfonate (^185 mV), and benzyl viologen (^311 mV). Spectra were recorded at the following potentials: E0m¼ þ400 mV (solid trace), þ360 mV (dotted trace), þ300 mV (dashed trace), and þ140 mV (dot^dash trace). 0
The second phase of reduction (Em ¼ þ60 mV) is associated with disappearance of the red-shifted CT band and can confidently be assigned to the reduction of heme b3 (Gro¨nberg et al., 1999). The relatively low potential of the catalytic heme in NorBC is somewhat surprising and has led to the proposal that the fully reduced form of the enzyme is not the catalytically relevant species. The apparent separation of the redox potential of heme b3 and the redox potentials of heme c and heme b can be exploited to poise NorBC, either chemically (as described earlier) or electrochemically (Field et al., 2002; Gro¨nberg et al., 1999). These partially reduced forms of the enzyme are excellent subjects for further interrogation of the environments of the heme centers by advanced spectroscopic methods. This approach to the study of NOR is illustrated by the EPR experiments described in the next section.
96
Sarah J. Field et al.
7. Electron Paramagnetic Resonance Spectroscopy Originally the distribution of electrons in the ‘‘three electron-reduced’’ from of NorBC seemed clear; both low-spin heme c and low heme b are fully reduced and the third electron resides on the active site FeB (Field et al., 2002; Gro¨nberg et al., 1999). However, the electronic absorption spectra shown in Fig. 5.4 suggest that the situation is rather more complicated, as the molar extinction coefficient of the ‘‘three electron-reduced’’ form appears to be dependent on the pH of the sample. The extent to which both low-spin hemes are reduced is less at pH 8.5 (see Fig. 5.4B) than at pH 6.0 (see Fig. 5.4A). This may be accounted for by a different distribution of the added electrons among the four redox centers, which is also indicated by low-temperature EPR spectra of the same samples (Fig. 5.7). Figure 5.7A shows X-band EPR spectra of NorBC in three different oxidation states recorded at pH 6.0. The spectrum of the oxidized form of the enzyme is dominated by features originating from the low-spin hemes. A rhombic trio with g values of 3, 2.25, and 1.45 has been assigned to the B
A
Fully oxidised
Fully oxidised
3 e− reduced 3 e− reduced
Fully reduced
1000 2000 3000 4000 5000 6000 Field (gauss)
Fully reduced
1000 2000 3000 4000 5000 6000 Field (gauss)
Figure 5.7 X-band EPR spectra of NorBC. The fully oxidized, ‘‘three electronreduced’’and fully reduced states buffered in 20 mM BTP, 50 mM NaCl, and 0.02% DM, pH 6.0 (A), or 20 mM BTP, 50 mM NaCl, and 0.02% DM, pH 8.5 (B). All spectra were recorded at 10 K, with a modulation amplitude of 1 mT and a microwave power of 2 mW.
Bacterial NorBC
97
ferric low-spin heme b in NorB (Cheesman et al., 1998; Gro¨nberg et al., 1999). The ramp-shaped signal at g ¼ 3.5 is a ‘‘high gmax signal’’ and represents the gz component of a rhombic trio in which values of the gy and gx components are less than 2. This signal is assigned to the ferric lowspin heme c in NorC (Cheesman et al., 1998; Gro¨nberg et al., 1999). Although the ferric high-spin heme b3 and ferric FeB both contain unpaired electrons and so in principle should be visible by EPR spectroscopy, no signals arising from the active site are present in the spectrum of the oxidized enzyme. This is because the high-spin heme b3 is coupled magnetically to the FeB to form an integer spin and hence EPR silent species (Cheesman et al., 1998). The signals seen at g ¼ 4.3 and g ¼ 6 may be because of small amounts of uncoupled active site components that represent less than 2% of the sample. The spectrum of oxidized NorBC is essentially the same at pH 8.5 (see Fig. 5.7B). Electron paramagnetic resonance spectra of the ‘‘three electronreduced’’ form of NorBC recorded at pH 6.0 (see Fig. 5.7A) and pH 8.5 (see Fig. 5.7B) are somewhat different. In both spectra the high gmax signal is no longer present, indicating that the low-spin heme c is fully reduced. Both spectra apparently show a small amount of the gz component (g ¼ 3.01) of the rhombic trio seen in oxidized spectra. The most obvious explanation is that this is a consequence of some residual oxidized low-spin heme b, although this is not consistent with the reported reduction potentials of heme b and heme c (Gro¨nberg et al., 1999). Careful inspection of the EPR signal at g ¼ 3.01 in the ‘‘three electron-reduced’’ EPR spectrum reveals it to have a slightly different line shape to the feature seen in the oxidized spectrum. This leads us to believe that it could be because of a small amount of uncoupled low-spin heme b3 at the active site, which becomes visible in the EPR experiment as the nonheme FeB becomes reduced. At pH 8.5, a signal at g ¼ 4.3 appears in the EPR spectrum of the ‘‘three electron-reduced’’ sample (see Fig. 5.7B) that is attributed to ferric nonheme FeB. Electrochemical poising experiments monitored by EPR spectroscopy have suggested that at this pH the reduction potentials of the centers at the active site are much closer than those determined from spectropotentiometric titrations recorded at pH 7.6 (S. J. Field, unpublished data). Earlier analyses of spectropotentiometric titrations placed the reduction potential of the nonheme FeB at around 320 mV (Gro¨nberg et al., 1999). If this is correct, then it would be expected that, in the ‘‘three electron-reduced’’ form, the nonheme FeB, along with both low-spin hemes, would be reduced and that the EPR spectrum would show a significant signal at g ¼ 6 arising from uncoupled ferric high-spin heme b3 (E0m ¼ þ60 mV). Figure 5.7 shows this not to be the case; instead, only small increases in both the g ¼ 6 (ferric heme b3) and the g ¼ 4.3 (ferric nonheme FeB) signals are seen together with the g ¼ 3.01 feature mentioned previously.
98
Sarah J. Field et al.
This observation, together with the pH dependence of the electronic absorption (see Fig. 5.4) and EPR spectra (see Fig. 5.7) of the ‘‘three electron-reduced’’ form of NorBC, leads us to believe that not only is the reduction potential of the nonheme FeB lower than we originally suggested, but also that it is pH and possibly temperature dependent. If the reduction potentials of heme b3 and nonheme FeB are sufficiently close when the protein is poised in the ‘‘three electron-reduced’’ state, the sample will in reality be a mixture of two, three, and four electron-reduced forms of NorBC. In the two electron-reduced form, the dinuclear center will be fully oxidized and hence EPR silent. The four electron-reduced (fully reduced) enzyme will also be EPR silent as both active site metal centers are reduced. In the remaining fraction of the enzyme that genuinely contains three electrons, the single electron in the active site can reside on either heme b3 or the nonheme FeB, giving a mixed population of EPR active species. In one of these subpopulations, heme b3 is oxidized and the nonheme FeB is reduced. This accounts for the small g ¼ 6 EPR signal from ferric highspin heme b3 and possibly the new rhombic signal from ferric low-spin b3 described earlier. The second subpopulation of the enzyme containing three electrons has a dinuclear center in which the ferrous heme b3 is EPR silent and ferric nonheme FeB gives rise to the g ¼ 4.3 feature, which is more noticeable at pH 8.5 (see Fig. 5.7B). Further redox titrations in this pH range, which are monitored by both optical and EPR spectroscopies, will be required to resolve this issue completely.
8. Concluding Remarks NorBC is a complex integral membrane protein containing four distinct redox centers, two of which form a dinuclear active site, which can be poised in a number of different redox states. A robust spectropotentiometric and kinetic description of the native enzyme has depended on development of the purification and assay procedures described in this chapter. These methods have in turn enabled the regular production of a highly purified enzyme that has allowed us to record a series of benchmark spectra that report on its structural integrity. Engineered forms of NorBC will be important for future structure–function studies on the enzyme, but it is important to remember that E. coli does not have its own norCB genes nor does not have any enzyme that has a nonheme iron center such as that present in the active site of NOR. Consequently, the successful synthesis of recombinant holo P. denitrificans NorCB in E. coli was a major challenge, and our success has opened the way for the production of characterization of a range of mutant forms of the enzyme deficient in key residues important in
Bacterial NorBC
99
proton transfer and metal ion binding. This will underpin future structural biology programs that will provide molecular structures to interface with currently available spectropotentiometric and kinetic data.
ACKNOWLEDGMENTS We acknowledge the contributions to the NOR work presented in this chapter made by a number of colleagues, including Dr. Gareth Butland, Dr. Karin Gro¨nberg, Dr. Louise Prior, and Dr. Lola Roldan. Dr. Myles Cheesman and Dr. Julea Butt provided much helpful advice on the application of EPR spectroscopy and electrochemical methods to the study of NOR. FHT was funded by a U.K. Medical Research Council Priority Area Studentship. Recent work in our laboratory was supported by grants from the John and Pamela Salter Trust to NJW and the U.K. Biotechnology and Biological Sciences Research Council to DJR (B19851) and NJW/DJR (BBC0077191).
REFERENCES Arslan, E., Schulz, H., Zufferey, R., Kunzler, P., and Tho¨ny-Meyer, L. (1998). Overproduction of the Bradyrhizobium japonicum c-type cytochrome subunits of the cbb3 oxidase in E. coli. Biochem. Biophys. Res. Comm. 251, 744–747. Bell, L. C., Richardson, D. J., and Ferguson, S. J. (1992). Identification of nitric oxide reductase activity in Rhodobacter capsulatus: The electron transport pathway can either use or bypass both cytochrome c2 and the cytochrome bc1 complex. J. Gen. Microbiol. 138, 437–443. Busch, A., Friedrich, B., and Cramm, R. (2002). Characterization of the norB gene, encoding nitric oxide reductase, in the nondenitrifying cyanobacterium Synechocystis sp. strain PCC6803. Appl. Environ. Microbiol. 68, 668–672. Butland, G., Spiro, S., Watmough, N. J., and Richardson, D. J. (2001). Two conserved glutamates in the bacterial nitric oxide reductase are essential for activity but not assembly of the enzyme. J. Bacteriol. 183, 189–199. Carr, G. J., and Ferguson, S. J. (1990). The nitric oxide reductase of P. denitrificans. Biochem. J. 269, 423–429. Cheesman, M. R., Zumft, W. G., and Thomson, A. J. (1998). The MCD and EPR of the heme centers of nitric oxide reductase from Pseudomonas stutzeri: Evidence that the enzyme is structurally related to the heme-copper oxidases. Biochem. 37, 3994–4000. Cramm, R., Pohlmann, A., and Friedrich, B. (1999). Purification and characterization of the single-component nitric oxide reductase from Ralstonia eutropha H16. FEBS Lett. 460, 6–10. de Boer, A. P., van der Oost, J., Reijnders, W. N., Westerhoff, H. V., Stouthamer, A. H., and van Spanning, R. J. (1996). Mutational analysis of the nor gene cluster which encodes nitric-oxide reductase from P. denitrificans. Eur. J. Biochem. 242, 592–600. de Vries, S., Strampraad, M. J., Lu, S., Moenne-Loccoz, P., and Schroder, I. (2003). Purification and characterization of the MQH2:NO oxidoreductase from the hyperthermophilic archaeon Pyrobaculum aerophilum. J. Biol. Chem. 278, 35861–35868. Dermastia, M., Turk, T., and Hollocher, T. C. (1991). Nitric oxide reductase: Purification from P. denitrificans with use of a single column and some characteristics. J. Biol. Chem. 266, 10899–10905.
100
Sarah J. Field et al.
Field, S. J., Prior, L., Roldan, M. D., Cheesman, M. R., Thomson, A. J., Spiro, S., Butt, J. N., Watmough, N. J., and Richardson, D. J. (2002). Spectral properties of bacterial nitric-oxide reductase: Resolution of pH-dependent forms of the active site heme b3. J. Biol. Chem. 277, 20146–20150. Flock, U., Watmough, N. J., and A¨delroth, P. (2005). Electron/proton coupling in bacterial nitric oxide reductase during reduction of oxygen. Biochemistry 44, 10711–10719. Gardner, A. M., Helmick, R. A., and Gardner, P. R. (2002). Flavorubredoxin, an inducible catalyst for nitric oxide reduction and detoxification in E. coli. J. Biol. Chem. 277, 8172–8177. Gill, S. R., Fouts, D. E., Archer, G. L., Mongodin, E. F., Deboy, R. T., Ravel, J., Paulsen, I. T., Kolonay, J. F., Brinkac, L., Beanan, M., Dodson, R. J., Daugherty, S. C., et al. (2005). Insights on evolution of virulence and resistance from the complete genome analysis of an early methicillin-resistant S. aureus strain and a biofilm-producing methicillinresistant S. epidermidis strain. J. Bacteriol. 187, 2426–2438. Girsch, P., and de Vries, S. (1997). Purification and initial kinetic and spectroscopic characterization of NO reductase from Paracoccus denitrificans. Biochim. Biophys. Acta 1318, 202–216. Gro¨nberg, K. L., Watmough, N. J., Thomson, A. J., Richardson, D. J., and Field, S. J. (2004). Redox-dependent open and closed forms of the active site of the bacterial respiratory nitric oxide reductase revealed by cyanide binding studies. J. Biol. Chem. 279, 17120–17125. Gro¨nberg, K. L. C., Rolda´n, M. D., Prior, L., Butland, G., Cheesman, M. R., Richardson, D. J., Spiro, S., Thomson, A. J., and Watmough, N. J. (1999). A lowredox potential heme in the dinuclear center of bacterial nitric oxide reductase: Implications for the evolution of energy-conserving heme-copper oxidases. Biochem. 38, 13780–13786. Hendriks, J., Gohlke, U., and Saraste, M. (1998a). From NO to OO: Nitric oxide and dioxygen in bacterial respiration. J. Bioenerg. Biomembr. 30, 15–24. Hendriks, J., Warne, A., Gohlke, U., Haltia, T., Ludovici, C., Lubben, M., and Saraste, M. (1998b). The active site of the bacterial nitric oxide reductase is a dinuclear iron center. Biochemistry 37, 13102–13109. Hendriks, J. H., Jasaitis, A., Saraste, M., and Verkhovsky, M. I. (2002). Proton and electron pathways in the bacterial nitric oxide reductase. Biochemistry 41, 2331–2340. Hendriks, J. H., Prior, L., Baker, A. R., Thomson, A. J., Saraste, M., and Watmough, N. J. (2001). Reaction of carbon monoxide with the reduced active site of bacterial nitric oxide reductase. Biochemistry 40, 13361–13369. Kastrau, D. H. W., Heiss, B., Kroneck, P. M. H., and Zumft, W. G. (1994). Nitric oxide reductase from Pseudomonas stutzeri: Phospholipid requirement, electron paramagnetic resonance and redox properties. Eur. J. Biochem. 222, 293–303. Leung, Y. C., Chan, C., Reader, J. S., Willis, A. C., van Spanning, R. J., Ferguson, S. J., and Radford, S. E. (1997). The pseudoazurin gene from Thiosphaera pantotropha: Analysis of upstream putative regulatory sequences and overexpression in E. coli. Biochem. J. 321, 699–705. Matsuda, Y., Inamori, K., Osaki, T., Eguchi, A., Watanabe, A., Kawabata, S., Iba, K., and Arata, H. (2002). Nitric oxide-reductase homologue that contains a copper atom and has cytochrome c-oxidase activity from an aerobic phototrophic bacterium Roseobacter denitrificans. J. Biochem. 131, 791–800. Nierman, W. C., DeShazer, D., Kim, H. S., Tettelin, H., Nelson, K. E., Feldblyum, T., Ulrich, R. L., Ronning, C. M., Brinkac, L. M., Daugherty, S. C., Davidsen, T. D., Deboy, R. T., et al. (2004). Structural flexibility in the Burkholderia mallei genome. Proc. Natl. Acad. Sci. USA 101, 14246–14251.
Bacterial NorBC
101
Pauleta, S. R., Guerlesquin, F., Goodhew, C. F., Devreese, B., Van Beeumen, J., Pereira, A. S., Moura, I., and Pettigrew, G. W. (2004). Paracoccus pantotrophus pseudoazurin is an electron donor to cytochrome c peroxidase. Biochemistry 43, 11214–11225. Paulsen, I. T., Seshadri, R., Nelson, K. E., Eisen, J. A., Heidelberg, J. F., Read, T. D., Dodson, R. J., Umayam, L., Brinkac, L. M., Beanan, M. J., Daugherty, S. C., Deboy, R. T., et al. (2002). The Brucella suis genome reveals fundamental similarities between animal and plant pathogens and symbionts. Proc. Natl. Acad. Sci. USA 99, 13148–13153. Pearson, I. V., Page, M. D., van Spanning, R. J., and Ferguson, S. J. (2003). A mutant of Paracoccus denitrificans with disrupted genes coding for cytochrome c550 and pseudoazurin establishes these two proteins as the in vivo electron donors to cytochrome cd1 nitrite reductase. J. Bacteriol. 185, 6308–6315. Poock, S. R., Leach, E. R., Moir, J. W., Cole, J. A., and Richardson, D. J. (2002). Respiratory detoxification of nitric oxide by the cytochrome c nitrite reductase of E. coli. J. Biol. Chem. 277, 23664–23669. Richardson, D. J., and Watmough, N. J. (1999). Inorganic nitrogen metabolism in bacteria. Curr. Opin. Chem. Biol. 3, 207–219. Sakurai, N., and Sakurai, T. (1997). Isolation and characterization of nitric oxide reductase from Paracoccus halodenitrificans. Biochemistry 36, 13809–13815. Stevanin, T. M., Poole, R. K., Demoncheaux, E. A., and Read, R. C. (2002). Flavohemoglobin Hmp protects Salmonella enterica serovar typhimurium from nitric oxide-related killing by human macrophages. Infect. Immun. 70, 4399–4405. Suharti, R. C., Strampraad, M. J., Schroder, I., and de Vries, S. (2001). A novel copper A containing menaquinol NO reductase from Bacillus azotoformans. Biochem. 40, 2632–2639. Takaya, N., Catalan-Sakairi, M. A., Sakaguchi, Y., Kato, I., Zhou, Z., and Shoun, H. (2003). Aerobic denitrifying bacteria that produce low levels of nitrous oxide. Appl. Environ. Microbiol. 69, 3152–3157. Tho¨ny-Meyer, L., Kunzler, P., and Hennecke, H. (1996). Requirements for maturation of Bradyrhizobium japonicum cytochrome c550 in E. coli. Eur. J. Biochemistry. 235, 754–761. Thorndycroft, F. H., Butland, G., Richardson, D. J., and Watmough, N. J. (2007). A new assay for nitric oxide reductase reveals two conserved glutamate residues form the entrance to a proton-conducting channel in the bacterial enzyme. Biochem. J. 401, 111–119. Tunbridge, A. J., Stevanin, T. M., Lee, M., Marriott, H. M., Moir, J. W., Read, R. C., and Dockrell, D. H. (2006). Inhibition of macrophage apoptosis by Neisseria meningitidis requires nitric oxide detoxification mechanisms. Infect. Immun. 74, 729–733. van der Oost, J., deBoer, A. P. N., deGier, J.-W. L., Zumft, W. G., Stouthamer, A. H., and van Spanning, R. J. M. (1994). The heme-copper oxidase family consists of three distinct types of terminal oxidases and is related to nitric oxide reductase. FEMS Microbiol. Lett. 121, 1–9. Vincent, K. A., Tilley, G. J., Quammie, N. C., Streeter, I., Burgess, B. K., Cheesman, M. R., and Armstrong, F. A. (2003). Instantaneous, stoichiometric generation of powerfully reducing states of protein active sites using Eu(II) and polyaminocarboxylate ligands. Chem. Commun. 259, 0–2591. Watmough, N. J., Butland, G., Cheesman, M. R., Moir, J. W., Richardson, D. J., and Spiro, S. (1999). Nitric oxide in bacteria: Synthesis and consumption. Biochem. Biophys. Acta 1411, 456–474. Zumft, W. G. (2005). Nitric oxide reductases of prokaryotes with emphasis on the respiratory, heme-copper oxidase type. J. Inorg. Biochem. 99, 194–215.
This page intentionally left blank
C H A P T E R
S I X
Redox-Controlled Dinitrosyl Formation at the Diiron-Oxo Center of NorA Rainer Cramm and Katja Strube Contents 1. Introduction 2. Genetic Context and Expression of the NorA Gene in R. eutropha 3. Purification of NorA 3.1. Overproduction 3.2. Isolation procedures 4. Disulfide Bridges in NorA 5. Iron Analysis and Preparation of Apo-NorA 6. Interconversion of Redox Forms of NorA 6.1. Preparation of diferrous NorA 6.2. Preparation of diferric NorA and oxyNorA 7. Generation of NorA-DNIC In Vitro 7.1. Procedure using NO-saturated buffer 7.2. Procedure using nitrite 8. Preparation of NorA-DNIC Formed In Vivo 9. Quantification of NO from NorA-DNIC 10. Outlook References
104 105 106 106 106 107 108 109 110 110 111 111 112 113 113 114 114
Abstract In the denitrifying bacterium Ralstonia eutropha H16, the NorA protein is coproduced with the respiratory nitric oxide (NO) reductase. NorA contains a diiron-oxo center, which can form stable adducts with dioxygen and NO. In contrast to other diiron proteins, the formation of NorA-NO requires both fully reduced protein and additional electrons. A minor fraction of in vitro NorANO represents a paramagnetic dinitrosyl iron complex (DNIC), while the major fraction is attributed to a DNIC of the structure {Fe(NO)2},10 which shows no electron paramagnetic resonance. The NorA-DNIC may be formed either upon Institut fu¨r Biologie/Mikrobiologie, Humboldt-Universita¨t zu Berlin, Berlin, Germany Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37006-7
#
2008 Elsevier Inc. All rights reserved.
103
104
Rainer Cramm and Katja Strube
direct reaction of the protein with NO or upon incubation with nitrite due to an intrinsic nitrite reduction activity of NorA that liberates NO. NorA can be purified rapidly as a six histidine-tagged derivative from overproducing cells of Escherichia coli. This chapter describes procedures for the preparation of different redox forms of NorA for the formation of NorA adducts with NO, dioxygen, and azide, as well as for the quantification of NorA-bound NO.
1. Introduction Binding of nitric oxide (NO) to iron protein plays an important role in physiological systems. In addition to nitrosation of heme proteins, e.g., hemoglobin and soluble guanylate cyclase (Henry and Guissani, 1999), NO has also been shown to interact with many nonheme iron sites to form stable Fe-NO complexes (Butler and Megson, 2002). Of these, dinitrosyl iron complexes (DNICs) in which two NO molecules are bound to one iron atom are of particular interest, as they are supposed to represent a transport form of NO (Chiang and Darensbourg, 2006; Keese et al., 1997; Ueno and Yoshimura, 2000; Vanin, 1998) and to act as potent NO donors (Vanin et al., 1996). DNICs are generally described as {Fe(NO)2}n (with n being the sum of d-type electrons from the metal and unpaired electrons from the NO) system according to Enemark–Feltham formalism (Enemark and Feltham, 1974). Note that DNICs differ from diiron-dinitrosyls of the structure [FeNO]2 that have been described for some diiron proteins, such as R2 protein of aerobic ribonucleotide reductase and methane monooxygenase (Coufal et al., 1999; Haskin et al., 1995). In bacteria, protein-bound DNICs have been implicated in regulatory mechanisms. In Escherichia coli, the [4Fe-4S] cluster of the oxygen sensor Fnr (Cruz-Ramos et al., 2002) and the [2Fe-2S] cluster of the superoxide sensor SoxR (Ding and Demple, 2000) are converted to monomeric and dimeric thiolate-ligated DNICs. Furthermore, the iron uptake regulator Fur, which contains a monoiron center, can react with NO to a major paramagnetic {Fe(NO)2}9 DNIC, proposed to originate from a diamagnetic {Fe(NO)2}8 DNIC intermediate (D’Autreaux et al., 2004). A paramagnetic {Fe(NO)2}9 DNIC has also been detected upon binding of NO to the diiron protein NorA from the denitrifying bacterium Ralstonia eutropha H16 (Strube et al., 2007). However, this species represented a minor fraction (up to 20%), while the major species was an electron paramagnetic resonance (EPR)-silent dinitrosyl complex, most probably of the form {Fe(NO)2}10. Notably, and in contrast to Fur, DNIC formation at NorA is dependent on the supply of external electrons. It is unknown, however, whether the activity of NorA in vivo is controlled by a specialized redox system or if NorA is reduced by an unspecific electron donor.
105
Dinitrosyl Formation at a Diiron-Oxo Center
2. Genetic Context and Expression of the NorA Gene in R. eutropha In R. eutropha, the norA structural gene is coexpressed with norB encoding the respiratory nitric oxide reductase (Bu¨sch et al., 2004; Pohlmann et al., 2000). The transcriptional activator NorR is encoded upstream of norA and is transcribed in the opposite direction (Fig. 6.1). In fact, the wild-type H16 contains a second set of the norRAB genes termed norR2A2B2 (Pohlmann et al., 2006). Products of the latter genes have not yet been characterized. Note that all physiological experiments related to NorR, NorA, or NorB were performed using a deletion mutant that lacks the norR2A2B2 set. NorR belongs to a subgroup of the NifA/NtrC family of regulators, which interact directly with their signal molecules (Studholme and Dixon, 2003). NorR orthologs are encoded in several bacterial genomes, including E. coli (Rodionov et al., 2005). E. coli NorR has been shown to bind NO by a monoiron center, which is coordinated by the N-terminal domain of the protein (D’Autreaux et al., 2005). Putative ligands of the iron have been identified for R. eutropha NorR by site-directed mutagenesis (Klink et al., 2007). In its active state, NorR activates expression of the norAB operon (Bu¨sch et al., 2004). The subsequent formation of nitric oxide reductase NorB is essential for denitrifying cells, probably because NO accumulates in
norR
− + norA
NO ON NO NorR
norB
N2O
NorA
NorB 2 NO
NO
Figure 6.1 Genetic context of the gene for NorA in R. eutropha. The regulator NorR activates expression of the norAB operon in response to NO. The norR gene is autoregulated negatively.The respiratory nitric oxide reductase NorB is instrumental in denitrification of R. eutropha.
106
Rainer Cramm and Katja Strube
the cell to toxic concentrations if NorB is lacking (Cramm et al., 1997). In contrast, cells lacking NorA did not show a growth defect during denitrification (Pohlmann et al., 2000). Nevertheless, formation of the NorA-NO complex appears to be of physiological relevance. The reaction triggers a feedback loop, as NO-dependent activation of NorR (and hence expression of the norAB operon) is decreased when NO is scavenged by NorA (Strube et al., 2007).
3. Purification of NorA 3.1. Overproduction To allow isolation of NorA by a rapid one-step method (e.g., to purify the NO-bound form of NorA from cell cultures), NorA has been modified genetically to contain six extra histidine residues at its C terminus (His tag). R. eutropha NorA can be overproduced to high yields in E. coli BL21(DE3) or Rosetta(DE3) cells containing a pET22b-based overexpression plasmid. Fresh transformants should be used for each preparation. About 10 mg NorA can be easily prepared from a 1-liter culture grown in LB medium. Ten milliliters of an overnight culture is used to inoculate 1 liter of LB medium, freshly prepared in a baffled 5-liter Erlenmeyer flask. The culture is incubated at 30 with shaking (140 rpm) for 3 h to reach an optical density at 600 nm of 0.6 (3 h). Overexpression of norA is induced by the addition of isopropyl-b-D-thiogalactopyranoside to a final concentration of 1 mM, and incubation is continued for another 3 to 4 h at the aforementioned conditions. The culture is then chilled on ice, and cells are harvested by centrifugation at 5000 g for 15 min.
3.2. Isolation procedures Although NorA is formed during denitrification (and thus under conditions of reduced oxygen tension), the protein is fairly stable in the presence of air. Thus the isolation procedure may be carried out under aerobic conditions. Note, however, that during aerobic isolation, different redox states of the diiron center may form (see later) and an increased fraction of dimeric NorA will appear due to the formation of disulfide bridges. Therefore, it is advisable to perform the isolation of NorA in an anaerobic chamber. The pellet of a 1-liter culture is resuspended in 5 ml 50 mM sodium phosphate buffer, pH 8.0, 300 mM NaCl, and 20 mM imidazole, and cells are immediately broken by sonication on ice. Debris and unbroken cells are then removed by centrifugation at 12,000 g for 15 min, and the supernatant
Dinitrosyl Formation at a Diiron-Oxo Center
107
(soluble extract) is transferred to a clean tube. His-tagged NorA is purified from the soluble extract by affinity chromatography using commercial nickel-chelating resin (nitrilotriacetic acid) and standard buffers (50 mM sodium phosphate buffer, pH 8.0, 300 mM NaCl) containing the following concentrations of imidazole: equilibration, 20 mM; washing, 50 mM; and elution, 250 mM. Eluted fractions are combined and concentrated using a microfiltrator (Millipore, Microcon centrifugal filter device, cutoff size 10 kDa). Typically, this procedure will yield about 1 ml containing 300 to 400 nmol hexahistidine-tagged NorA. Such preparations are 95% pure, as judged by SDS-PAGE and subsequent staining with Coomassie brilliant blue. If necessary, further purification steps may be added: the sample is desalted using PD-10 columns (GE Healthcare) and applied to an anion-exchange column (ResourceQ, GE Healthcare) preequilibrated with 50 mM sodium phosphate buffer, pH 8.0, and 10 mM NaCl. NorA elutes at approximately 150 mM NaCl in a linear gradient ranging from 10 to 500 mM NaCl in 50 mM sodium phosphate buffer (pH 8.0). Purified NorA is used immediately for experiments. For storage, glycerol is added to a final concentration of 20% (v/v). Aliquots are then frozen in liquid nitrogen and stored at –80 .
4. Disulfide Bridges in NorA If isolated under aerobic conditions, preparations contain aggregated NorA species that are readily detected in nonreducing SDS-PAGE. The amount of a dimeric NorA species increases upon prolonged exposure of such preparations to air (Fig. 6.2). After overnight incubation at 4 , up to 20% of NorA is detectable as a homodimer. NorA contains four cysteine residues, three of which are conserved according to a multiple alignment of NorA orthologs from different bacteria (Strube et al., 2007). Two of these are part of the conspicuous CCGG sequence within the N-terminal domain of NorA. After anaerobic preparation of NorA from aerobically grown cells of E. coli, all of the four cysteines possess free thiol groups, according to a titration with 5,50 -dithiobis(2-nitrobenzoic acid) (DTNB) of the native protein. Therefore, all four cysteines are solvent accessible, and NorA does not form disulfide-bridged homodimers in the cell. Interestingly, only two free thiols can be detected by DTNB titration after prolonged incubation of NorA in air. Because only a monomeric and a dimeric form are resolved by SDS-PAGE, this result has to be explained by an additional intramolecular cysteine bridge, which, however, does not lead to an altered mobility of the protein in standard gels.
108
Rainer Cramm and Katja Strube
A
S
1
2
B
S
1
2
83 62 47.5
32.5
Figure 6.2 Disulfide bridging in NorA. NorA (10 mg) aerobically purified (A) and after aerobic incubation overnight at 4 (B) in the absence (lane 1) or in the presence (lane 2) of 10 mM DTT, separated in a12% SDS-PAGE gel. Molecular weights of protein standard (lanes S) are shown on the left side.
5. Iron Analysis and Preparation of Apo-NorA Selection of an appropriate method for chemical iron determination is crucial in avoiding underestimation of the iron content of NorA. For example, the chelator ferene S turned out to be less suitable even if the incubation periods were extended considerably. Good results are obtained with a rapid and simple method using the iron (II) chelator 2,20 -bipyridyl. In an anaerobized rubber-sealed reaction tube, 100 ml NorA (0.01–0.3 mM ) is reduced with excess sodium dithionite for 5 min. Subsequently, 0.1 ml of a 20 mM 2,20 -bipyridyl stock solution is added and the sample is vortexed vigorously for 5 min. During this time, the red-pink bipyridyl-iron complex is formed. Protein is removed prior to iron quantification using a microfiltrator (Millipore, Microcon centrifugal filter device, cutoff size 10 kDa). Absorption of the filtrate is measured at 520 nm, and iron is quantified using an absorption coefficient of e ¼ 8000 M1 cm1. A similar method is used to prepare apo-NorA: 0.5 ml of 0.3 mM purified NorA is mixed with 10 mM ascorbate and 10 mM phenazine methosulfate (PMS) in a reaction tube placed in an anaerobic chamber. Iron is chelated by the addition of 0.1 ml of a 20 mM 2,20 -bipyridyl stock solution. After incubation for 1 h with gentle shaking, the solution is loaded onto a PD-10 desalting column (GE Healthcare) and eluted with 50 mM sodium phosphate buffer, pH 8.0, 300 mM NaCl. The eluate contains iron-free apo-NorA with no detectable absorption above 300 nm.
109
Dinitrosyl Formation at a Diiron-Oxo Center
6. Interconversion of Redox Forms of NorA Binuclear iron clusters may adopt three redox states: (i) fully oxidized, (ii) fully reduced, or (iii) mixed valent. A fully reduced diiron center is a prerequisite for the formation of dioxygen or nitric oxide adducts of NorA. Solutions of diferrous NorA are colorless, and the protein shows only weak (if any) signals in both EPR spectra and optical spectra above 300 nm (Fig. 6.3). As expected for a diferric diiron-(hydr)oxo center, fully oxidized NorA is EPR silent because of antiferromagnetic coupling. Solutions of diferric NorA are pale yellow. In the optical spectrum the protein shows a prominent signal at 353 nm (e 4500 M1 cm1) and a shoulder at 430 nm (see Fig. 6.3, inset trace a). Diferric NorA combines readily with azide, yielding a light orange product with characteristic signals at 327 and 453 and a shoulder at 490 nm (see Fig. 6.3, inset trace b). Frequently, in aged preparations of NorA, a minor fraction gives rise to an S ¼ 1/2 EPR signal with gz¼ 1.9665, gy¼ 1.9230, gx ¼ 1.8732 (below 30 K), which is because of a mixed-valence center (Strube et al., 2007). Mixed-valent NorA is not detectable in samples prepared freshly under anaerobic conditions.
0.4 Absorbtion (arbitrary units)
Absorbtion (arbitrary units)
0.8
0.6 b 0.4
0.2
0.3 0.2
b
0.1 a 0.0 300
400
500 600 Wavelength (nm)
700
800
a
c 0.0 300
400
500 Wavelength (nm)
600
Figure 6.3 Absorption spectrum of anaerobically purified NorA.Trace a:150 mM NorA isolated anaerobically; trace b: sample from trace a exposed to air for 60 min; trace c: addition of a few crystals of dithionite to sample of trace a. (Inset) Trace a, 70 mM air-exposed NorA incubated overnight; trace b, sample from trace a incubated with excess azide (10 mM ) for 1 min.
110
Rainer Cramm and Katja Strube
6.1. Preparation of diferrous NorA Purification under anaerobic conditions yields mainly diferrous NorA. However, depending on the concentration of residual oxygen during the isolation procedure, such NorA preparations may also contain traces of oxy- and diferric NorA. A homogeneous sample of diferrous NorA is prepared by the addition of reductant (ascorbate/PMS or dithionite, cf. Fig. 6.3, trace c and Fig. 6.4, trace a) and subsequent recovery of NorA by dialysis or gel filtration. A convenient procedure is as follows: later. In an anaerobized cuvette sealed with a rubber stopper, 300 mM NorA is reduced by the addition of 10 mM ascorbate and 10 mM PMS. Bleaching of NorA absorbance may be monitored by optical spectroscopy (keep in mind the intrinsic absorbance of the reducing compounds). Subsequently, the solution is applied to a PD-10 column (GE Healthcare) placed in an anaerobic chamber. Note that after removal of the reductant, traces of oxygen present in the environment will immediately lead to formation of some oxyNorA.
6.2. Preparation of diferric NorA and oxyNorA Diferric NorA can be prepared easily from oxyNorA by a simple incubation step. To prepare oxyNorA, a sample of diferrous NorA (devoid of artificial reductant) is exposed to air at room temperature for 30 to 60 min.
Absorbtion (arbitrary units)
0.14
Absorbtion (arbitrary units)
0.5
0.4
0.3 b
0.2
0.12 0.10 0.08 0.06 0.04 0.02 0.00 400
500
600 700 Wavelength (nm)
800
0.1 a 0.0 400
500
600
700
800
900
Wavelength (nm)
Figure 6.4 Absorption spectrum of NorA-NO. NorA (150 mM) was incubated in a cuvette closed with a rubber septum and reduced with 10 mM ascorbate and 10 mM PMS (trace a). Subsequently, NO-saturated buffer (400 mM NO) was added. (Inset) NorA-NO (110 mM) as isolated from E. coli cell cultures treated with 10 mM NO.
Dinitrosyl Formation at a Diiron-Oxo Center
111
Only minimal amounts of dioxygen are required; formation of oxyNorA was observed even in an anaerobic cuvette closed with a Teflon stopper due to the presence of dioxygen adhered to the Teflon material. OxyNorA is yellow-red in solution and is characterized by a 520-nm band (e ¼ 1153 M1 cm1) in the optical spectrum (see Fig. 6.3, trace b). Because the region between 400 and 550 nm contains bands for ligand-to-iron charge transfer transitions with noninnocent ligands such as peroxide (Reem et al., 1989), it is feasible that oxygen is bound as a hydroperoxo-Fe3þ adduct, similar to oxygen binding by hemerythrin (Stenkamp, 1994). Upon prolonged aerobic incubation on ice, oxyNorA autooxidizes probably by a mechanism similar to the formation of diferric hemerythrin from oxygenized hemerythrin by loss of the peroxide ligand (Stenkamp, 1994). After an overnight incubation at 4 , NorA is recovered quantitatively in the diferric state. Diferric NorA is readily distinguished from diferrous NorA by the addition of azide, as only the diferric form reacts to an orange adduct (see Fig. 6.3, inset, trace b).
7. Generation of NorA-DNIC In Vitro 7.1. Procedure using NO-saturated buffer In general, the addition of NO-saturated buffer to stirred NorA solutions in rubber-sealed cuvettes is superior to the addition of NO gas. The latter procedure leads to high local concentrations of NO, which may result in the precipitation of protein. Aqueous solutions saturated with nitric oxide (1.9 mM ) can be prepared by bubbling nitric oxide gas through pure water under anaerobic conditions. An alkaline trap should be connected upstream of the water vial to remove traces of certain other N-oxides. If the system is not anaerobized properly, the NO solution will finally contain considerable amounts of nitrite. For experiments with NorA, this would be highly undesirable, as the protein also reacts with nitrite (see later). It is advisable to check NO solutions by optical spectroscopy. Pure NO solutions will show no absorption above 300 nm, while contaminations (nitrite or other oxidation products of NO) give additional absorption between 300 and 400 nm. For the reaction of NorA with nitric oxide, 200 ml of 150 mM diferrous NorA is transferred to an anaerobic microcuvette that contains a small stirrer bar and is sealed with a rubber septum. This step should be carried out in an anaerobic chamber to prevent oxidation of both reactants. Subsequently, 1 mM of ascorbate and catalytic amounts of PMS (1 mM) are added with a gas-tight syringe and the cuvette is placed on a stirring device, ideally integrated into the cuvette holder of a spectrophotometer. After an
112
Rainer Cramm and Katja Strube
incubation period of 1 min at room temperature, 20 ml of a saturated NO solution is injected with a gas-tight syringe. This procedure is repeated three times to provide an excess of NO. The resulting NorA-DNIC is greenish and is characterized by the formation of two specific signals at 420 and 750 nm in the optical spectrum (Fig. 6.4, trace b). The absorption coefficient of the 750-nm band is E ¼ 126 M1 cm1.
7.2. Procedure using nitrite
NorB
NO (nmol)
30
NO (nmol)
In the presence of reductant, NorA is able to catalyze a limited number of nitrite reduction cycles, yielding NO as the end product. After the reaction is finished, the optical spectrum of the sample is indistinguishable from NorA-DNIC as prepared with NO-saturated buffer. Thus the catalytic turnover of nitrite is probably product inhibited as a consequence of NorA-DNIC formation. Nevertheless, excess nitric oxide is formed during this reaction, which can be monitored by amperometric methods using NO-sensitive electrodes (Fig. 6.5). Nitrite reduction by NorA is slow, and high amounts of nitrite are required to follow this reaction as the Km of NorA for nitrite is around 7 mM. The reaction is best followed in phosphate buffer, pH 8.0. Lower pH values increase the chemical formation of NO from nitrous acid, which may cover the NorA-derived reaction at some point. Higher pH values decrease the activity of NorA.
30 15 0 0
20
4 8 12 Time (min)
10 0 0
2
4 6 8 Time (min)
10
Figure 6.5 Nitrite reduction to NO. NO formation by NorA (3 mM) was followed in a Clark electrode. The reaction buffer was 50 mM sodium phosphate buffer (pH 7.0), 20 mM D-glucose,10 units of glucose oxidase, 250 units of catalase,10 mM ascorbate, and 10 mM PMS.The reaction volume was 2 ml. Bold face arrows denote addition of 2.5 mM nitrite. After incubation for 5 min, 250 nM purified NO reductase NorB of R. eutropha was added (indicated by a thin arrow) to verify that the electrode signal was caused by NO. (Inset) An experiment without adding NO reductase.
Dinitrosyl Formation at a Diiron-Oxo Center
113
8. Preparation of NorA-DNIC Formed In Vivo NorA-DNIC is formed in vivo (see Fig. 6.4, inset) and can be isolated from NO-exposed cultures of E. coli. NorA is overproduced as described earlier except that cells are transferred to a sealed bottle 1.5 h after induction. NO-saturated buffer (1.9 mM NO) is added to the culture to a final concentration of 10 mM. Cells are exposed to NO for 10 min. NorA is isolated in an anaerobic chamber by a rapid procedure (10 min) using commercial nickel-chelate spin columns (Qiagen) equilibrated with phosphate buffer, pH 8.0, 300 mM NaCl, 20 mM imidazole. Spin steps are carried out at a low centrifugal force (2000 rpm) for 1 min. Soluble extracts are passed three times through the spin columns to increase the total yield. After extensive washing (phosphate buffer, pH 8.0, 300 mM NaCl, 20 mM imidazole), the columns are eluted with phosphate buffer, pH 8.0, 300 mM NaCl, 250 mM imidazole. To save time, further purification steps are omitted and the protein is transferred to a rubber-sealed cuvette. In the absence of reductant, NorA-DNIC is sensitive toward exposure to oxygen. Assuming that the initial amount of NorA-DNIC in the cell is 100%, a half-time of about 20 min is calculated for the decay of the DNIC in air.
9. Quantification of NO from NorA-DNIC The following procedure has to be carried out in an anaerobic chamber to avoid loss of NO by air/oxidation. NorA (50 to 200 mM) is incubated with three equivalents of NO, 1 mM ascorbate, and 10 mM PMS in a sealed vial without headspace for 15 min. Alternatively, a sealed cuvette may be used to monitor formation of the DNIC by optical spectroscopy. Using a gas-tight syringe, the solution is then transferred to a sealed microfiltrator (Millipore, Microcon centrifugal filter device, cutoff size 10 kDa) and centrifuged at 12,000g for 1 min to obtain some protein-free filtrate. NO is quantified in a modified Griess assay. Aliquots from the retenate and the filtrate are injected into rubber-sealed reaction tubes that contain 300 ml Griess reagent under an atmosphere of dioxygen. Under these conditions, NO is oxidized to nitrite, which is quantified in the Griess reaction by measuring the absorption at 550 nm. Note that the retenate (containing NorA-DNIC) has to be incubated for about 5 h for quantitative analysis, perhaps as a consequence of slow conversion of the DNIC into nitrite (Boese et al., 1995). In agreement with the initial NorA/NO ratio of 1/3 and DNIC formation by NorA, a ratio of 1.8 (0.15) molecules NO per diiron center versus 0.8 (0.1) molecules free NO is determined by this procedure.
114
Rainer Cramm and Katja Strube
As a control, NO-exposed apo-NorA may be used in this assay to exclude contamination with unspecifically bound NO or nitrite.
10. Outlook Orthologs of NorA are encoded in many proteobacteria and firmicutes and have been termed YtfE, DnrN, ScdA, or NipC. In most cases where it has been studied, these proteins are formed in response to nitrosative stress ( Justino et al., 2005; Kim et al., 2003; Pohlmann et al., 2000; Sebbane et al., 2006; Vollack and Zumft, 2001). The physiological role of NorA and its orthologs is still enigmatic. In E. coli, YtfE has been shown to be involved in the response toward oxidative and nitrosative stress ( Justino et al., 2005, 2006). In particular, the activity of the [Fe-S] containing nitrate reductase NAR was decreased greatly in the YtfE mutant. It was demonstrated that YtfE facilitates the repair of damaged [4Fe-4S] centers, and a role of YtfE in iron insertion was suggested ( Justino et al., 2007). However, no molecular basis for this activity is yet available. In contrast, the growth of denitrifying (and thus NO-exposed) R. eutropha NorA mutant cells was not impaired significantly (Pohlmann et al., 2000; Strube et al., 2007), showing that the activity of the NAR enzyme in R. eutropha does not depend on NorA. Therefore, it has to be considered that, albeit both proteins are 50% identical in their amino acid sequence, NorA and YtfE serve different physiological functions.
REFERENCES Boese, M., Mordvintcev, P. I., Vanin, A. F., Busse, R., and Mulsch, A. (1995). S-nitrosation of serum albumin by dinitrosyl-iron complex. J. Biol. Chem. 270, 29244–29249. Bu¨sch, A., Pohlmann, A., Friedrich, B., and Cramm, R. (2004). A DNA region recognized by the nitric oxide-responsive transcriptional activator NorR is conserved in beta- and gamma-proteobacteria. J. Bacteriol. 186, 7980–7987. Butler, A. R., and Megson, I. L. (2002). Non-heme iron nitrosyls in biology. Chem. Rev. 102, 1155–1166. Chiang, C. Y., and Darensbourg, M. Y. (2006). Iron nitrosyl complexes as models for biological nitric oxide transfer reagents. J. Biol. Inorg. Chem. 11, 359–370. Coufal, D. E., Tavares, P., Pereira, A. S., Hyunh, B. H., and Lippard, S. J. (1999). Reactions of nitric oxide with the reduced non-heme diiron center of the soluble methane monooxygenase hydroxylase. Biochemistry 38, 4504–4513. Cramm, R., Siddiqui, R. A., and Friedrich, B. (1997). Two isofunctional nitric oxide reductases in Alcaligenes eutrophus H16. J. Bacteriol. 179, 6769–6777.
Dinitrosyl Formation at a Diiron-Oxo Center
115
Cruz-Ramos, H., Crack, J., Wu, G., Hughes, M. N., Scott, C., Thomson, A. J., Green, J., and Poole, R. K. (2002). NO sensing by FNR: Regulation of the E. coli NO-detoxifying flavohaemoglobin, Hmp. EMBO J. 21, 3235–3244. D’Autreaux, B., Horner, O., Oddou, J. L., Jeandey, C., Gambarelli, S., Berthomieu, C., Latour, J. M., and Michaud-Soret, I. (2004). Spectroscopic description of the two nitrosyl-iron complexes responsible for fur inhibition by nitric oxide. J. Am. Chem. Soc. 126, 6005–6016. D’Autreaux, B., Tucker, N. P., Dixon, R., and Spiro, S. (2005). A non-haem iron centre in the transcription factor NorR senses nitric oxide. Nature 437, 769–772. Ding, H., and Demple, B. (2000). Direct nitric oxide signal transduction via nitrosylation of iron-sulfur centers in the SoxR transcription activator. Proc. Natl. Acad. Sci. USA 97, 5146–5150. Enemark, J. H., and Feltham, R. D. (1974). Principles of structure, bonding and reactivity for metal nitrosyl complexes. Coord. Chem. Rev. 13, 339–406. Haskin, C. J., Ravi, N., Lynch, J. B., Munck, E., and Que, L., Jr. (1995). Reaction of NO with the reduced R2 protein of ribonucleotide reductase from E. coli. Biochemistry 34, 11090–11098. Henry, Y., and Guissani, A. (1999). Interactions of nitric oxide with hemoproteins: Roles of nitric oxide in mitochondria. Cell. Mol. Life Sci. 55, 1003–1014. Justino, M. C., Almeida, C. C., Goncalves, V. L., Teixeira, M., and Saraiva, L. M. (2006). Escherichia coli YtfE is a di-iron protein with an important function in assembly of ironsulphur clusters. FEMS Microbiol. Lett. 257, 278–284. Justino, M. C., Almeida, C. C., Teixeira, M., and Saraiva, L. M. (2007). Escherichia coli di-iron YtfE protein is necessary for the repair of stress-damaged iron-sulfur clusters. J. Biol. Chem. 282, 10352–10359. Justino, M. C., Vicente, J. B., Teixeira, M., and Saraiva, L. M. (2005). New genes implicated in the protection of anaerobically grown E. coli against nitric oxide. J. Biol. Chem. 280, 2636–2643. Keese, M. A., Bose, M., Mulsch, A., Schirmer, R. H., and Becker, K. (1997). Dinitrosyldithiol-iron complexes, nitric oxide (NO) carriers in vivo, as potent inhibitors of human glutathione reductase and glutathione-S-transferase. Biochem. Pharmacol. 54, 1307–1313. Kim, C. C., Monack, D., and Falkow, S. (2003). Modulation of virulence by two acidified nitrite-responsive loci of Salmonella enterica serovar typhimurium. Infect. Immun. 71, 3196–3205. Klink, A., Elsner, B., Strube, K., and Cramm, R. (2007). Characterization of the signaling domain of the NO-responsive regulator NorR from R. eutropha H16 by site-directed mutagenesis. J. Bacteriol. 189, 2743–2749. Pohlmann, A., Cramm, R., Schmelz, K., and Friedrich, B. (2000). A novel NO-responding regulator controls the reduction of nitric oxide in R. eutropha. Mol. Microbiol. 38, 626–638. Pohlmann, A., Fricke, W. F., Reinecke, F., Kusian, B., Liesegang, H., Cramm, R., Eitinger, T., Ewering, C., Potter, M., Schwartz, E., Strittmatter, A., Voss, I., et al. (2006). Genome sequence of the bioplastic-producing ‘‘Knallgas’’ bacterium R. eutropha H16. Nat. Biotechnol. 24, 1257–1262. Reem, R. C., McCormick, J. M., Richardson, D. E., Devlin, F. J., Stephens, P. J., Musselman, R. L., and Solomon, E. I. (1989). Spectroscopic studies of the coupled binuclear ferric active site in methemerythrins and oxyhemerythrin: The electronic structure of each iron center and the iron-oxo and iron-peroxide bonds. J. Am. Chem. Soc. 111, 4688–4704. Rodionov, D. A., Dubchak, I. L., Arkin, A. P., Alm, E. J., and Gelfand, M. S. (2005). Dissimilatory metabolism of nitrogen oxides in bacteria: Comparative reconstruction of transcriptional networks. PLoS Comput. Biol. 1, e55.
116
Rainer Cramm and Katja Strube
Sebbane, F., Lemaitre, N., Sturdevant, D. E., Rebeil, R., Virtaneva, K., Porcella, S. F., and Hinnebusch, B. J. (2006). Adaptive response of Yersinia pestis to extracellular effectors of innate immunity during bubonic plague. Proc. Natl. Acad. Sci. USA 103, 11766–11771. Stenkamp, R. E. (1994). Dioxygen and hemerythrin. Chem. Rev. 94, 715–726. Strube, K., de Vries, S., and Cramm, R. (2007). Formation of a dinitrosyl iron complex by NorA, a nitric oxide binding di-iron protein from Ralstonia eutropha H16. J. Biol. Chem. 282, 20292–20300. Studholme, D. J., and Dixon, R. (2003). Domain architectures of s 54-dependent transcriptional activators. J. Bacteriol. 185, 1757–1767. Ueno, T., and Yoshimura, T. (2000). The physiological activity and in vivo distribution of dinitrosyl dithiolato iron complex. Jpn. J. Pharmacol. 82, 95–101. Vanin, A. F. (1998). Dinitrosyl iron complexes and S-nitrosothiols are two possible forms for stabilization and transport of nitric oxide in biological systems. Biochemistry (Mosc.) 63, 782–793. Vanin, A. F., Stukan, R. A., and Manukhina, E. B. (1996). Physical properties of dinitrosyl iron complexes with thiol-containing ligands in relation with their vasodilator activity. Biochem. Biophys. Acta 1295, 5–12. Vollack, K. U., and Zumft, W. G. (2001). Nitric oxide signaling and transcriptional control of denitrification genes in Pseudomonas stutzeri. J. Bacteriol. 183, 2516–2526.
C H A P T E R
S E V E N
Purification and Functional Analysis of Fungal Nitric Oxide Reductase Cytochrome P450nor Li Zhang* and Hirofumi Shoun† Contents 118 119 119 120 121 122 123 124 125 126 127 128 130 131 131 131 131
1. Introduction 2. Screening of P450nor Activity 3. Gas Analysis 4. Purification of P450nor 5. Nitric Oxide Reductase Activity Assay 6. Protein Sequencing 7. Isolation of cDNA 8. Subcellular Fractionation of T. cutaneum 9. Site-Directed Mutagenesis 10. Expression of Recombinant Proteins 11. Purification of Recombinant Proteins 12. Titration of NAD Analogs 13. Stopped-Flow Rapid Scan Analysis 14. Other Analysis 15. Conclusion Acknowledgments References
Abstract Cytochrome P450nor (P450nor) is a nitric oxide (NO) reductase involved in fungal denitrification. Denitrification is a biological process in which nitrate or nitrite is reduced to gaseous nitrogen, the reverse reaction of nitrogen fixation. It therefore plays an important role in maintaining global environmental homeostasis. The involvement of P450nor in fungal denitrification indicates that denitrification not only occurs in prokaryotic bacteria, but also in eukaryotic fungi. In addition, the reduction of NO to nitrous oxide catalyzed by P450nor has added new insight into the function of cytochrome P450s, which are usually
* {
Department of Biology, University of Kentucky, Lexington, Kentucky Department of Biotechnology, University of Tokyo, Bunkyo-ku, Tokyo, Japan
Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37007-9
#
2008 Elsevier Inc. All rights reserved.
117
118
Li Zhang and Hirofumi Shoun
monooxygenases. Currently, five isozymes of P450nor have been isolated from the subdivisions of eumycota, and studies on the function and structure of P450nor have provided important information for both molecular mechanisms of P450 reactions and wastewater treatment. This chapter describes the screening of NO reductase activities, cloning, purification, and functional analysis of P450nor.
1. Introduction Fungal nitric oxide reductase cytochrome P450nor (P450nor) is a unique heme-thiolate protein involved in fungal denitrification by reducing nitric oxide (NO) to nitrous oxide (N2O) (for reviews, see Daiber et al., 2005; Shoun, 2006; Zhang et al., 2002a). Unlike other cytochrome P450s, P450nor is a reductase without monooxygenase activity, although it belongs to the P450 super family (CYP55A). It receives electrons directly from the distal pocket from NADH and/or NADPH without any help from a flavoprotein such as P450 reductase (Nakahara et al., 1993). The overall reaction is described in the following equation:
2NO þ NADðPÞH þ Hþ ! N2 O þ NADðPÞþ þH2 O P450nor was originally isolated from the filamentous fungus Fusarium oxysporum (Shoun and Tanimoto, 1991), indicating for the first time that denitrification is not restricted in prokaryotes. Denitrification is a reverse procedure of nitrogen fixation, which plays crucial roles in the nitrogen cycle of the ecosystem. As part of the whole denitrification procedure, fungal denitrification contains reduction steps of nitrogen sources from nitrate to nitrous oxide. P450nor plays a central role in the last step by catalyzing the reduction of NO to N2O. Knockout of P450nor from F. oxysporum results in the defect of denitrification activity, indicating that P450nor is necessary for fungal denitrification (Takaya and Shoun, 2000). We have isolated and cloned five isoforms of P450nor, which all belong to the CYP55A family: Fnor (CYP55A1) from F. oxysporum (Kizawa et al., 1991; Nakahara and Shoun, 1996), Cnor1 (CYP55A2) and Cnor2 (CYP55A3) from Cylindrocarpon tonkinense (Kudo et al., 1996; Usuda et al., 1995) (both fungi are the anamorph generation of ascomycetous fungi), Tnor (CYP55A4) from basidiomycetous yeast Trichosporon cutaneum (Zhang et al., 2001), and Anor (CYP55A5) from another ascomycetous fungus Aspergillus oryzae (Kaya et al., 2004). Therefore, P450nor activity is widely distributed in the subdivisions of eumycota.
Fungal Nitric Oxide Reductase
119
The protocols described herein have been used successfully to clone and purify fungal P450nor from yeast. This chapter also includes protocols for the functional analysis of this novel P450.
2. Screening of P450nor Activity Denitrification activities were screened successfully from soil fungi and yeasts (Shoun et al., 1992; Tsuruta et al., 1998). Yeast strains are obtained from the Japanese Collection of Microorganisms ( JCM, RIKEN, Japan). Potato dextrose (PD) broth (Difco; pH 7.4) is used in preculture. Each strain is inoculated into 120 ml of PD broth in a 500-ml Erlenmeyer flask sealed with a cotton plug for 3 days on a rotary shaker (120 rpm). Glycerol-peptone medium is used for detecting the denitrifying activity by anaerobic incubation at 30 or 25 (Candida sp. is incubated at 15 because of its isolation from Antarctica) for 4 days: 3% glycerol, 0.2% peptone, 10 mM sodium nitrite (or nitrate), a trace element solution (1 ml/liter), which contains the following chemicals (in 1 liter of distilled water): 0.2 g CoCl2H2O, 0.2 g FeSO47H2O, 1.0 g FeCl26H2O, 4.0 g CuSO45H2O, 0.2 g CaCl2, 8.6 mg Na2MoO42H2O, and 0.2 mg Na2SeO3. In brief, every 40 ml of preculture is inoculated into 120 ml of the aforementioned medium in a 500-ml Erlenmeyer flask with two side arms, which is sealed with a rubber stopper after replacing the headspace air with argon, and then incubated on a rotary shaker (120 rpm).
3. Gas Analysis An aliquot (usually 0.5 ml) of the upper-space gas in each flask is collected with a syringe through the side arm after incubation. The amount of gas (N2O) formed is determined by gas chromatography with a Shimadzu gas chromatograph GC 12A equipped with a Porapack Q column (3 2 mm, inner diameter) and by isotope mass spectrometry with a Finnigan Delta plus isotope mass spectrometer when necessary. Peaks are identified and quantified by comparison with standard gases (GL Sciences, Tokyo, Japan). [15N]nitrite (99 atom %) can be obtained from Shoko-Tsusho (Tokyo). A typical result is shown in the work of Tsuruta et al. (1998). Different from bacterial denitrification, which usually contains the whole process from nitrate to nitrogen (NO3 ! N2), most of the yeast strains that exhibit denitrifying activity use only the minimal process of denitrification from nitrite to nitrous oxide (NO2 ! N2O).
120
Li Zhang and Hirofumi Shoun
4. Purification of P450nor Because P450nor is typically detectable in yeast T. cutaneum, it was used to purify P450nor (Zhang et al., 2001). Preculture is prepared by seeding the yeast in 300 ml of YPD medium (10 g/liter glucose, 10 g/liter peptone, and 5 g/liter yeast extract, pH 7.4) in a 500-ml flask and incubating at 30 for 24 h at 100–120 rpm. The whole preculture is then transferred to 3 liters of YPD medium in a 5-liter conical flask and incubated in the same conditions as the preculture. Wet cells of yeast are filter collected and kept at –80 until use. In contrast to Fnor in F. oxysporum, Tnor does not require any induction by nitrogen sources. Therefore, nitrite or nitrate can be omitted from the whole incubation. Wet cells are disrupted at 4 by grinding with aluminum oxide in 0.1 M potassium phosphate buffer (pH 7.15) containing 10% (v/v) glycerol, 0.1 mM EDTA, 0.1 mM dithiothreitol (DTT), and each of 0.25 mM protease inhibitors, phenylmethylsulfonyl fluoride (PMSF), leupeptin, and tosylphenylalanyl chloromethyl ketone (TPCK). The cell homogenate is subjected to sequential centrifugation at 1500g for 15 min, at 10,000g for 20 min, and at 105,000g for 1 h. The resulting supernatant (soluble fraction) is dialyzed overnight against buffer A (10 mM potassium phosphate buffer, pH 7.3, containing 10% glycerol, 0.1 mM EDTA, and 0.1 mM DTT), and then the dialyzed fraction is again subjected to 105,000g centrifugation for 1 h. The whole supernatant is loaded onto a DEAEcellulose (Whatman DE-52, England) open column equilibrated with the same buffer. After being washed with the same buffer, proteins are eluted with a linear gradient of 0–0.5 M KCl in the same buffer, and fractions containing P450nor are collected, concentrated, and dialyzed overnight against buffer B (50 mM potassium phosphate containing 10% glycerol, 0.1 mM EDTA, and 0.1 mM DTT, pH 7.5). The sample is then mixed with the same volume of 4 M ammonium sulfate in buffer B and precipitates formed are removed by centrifugation at 10,000g for 10 min. The resulting supernatant is applied to a phenyl-Superose HR5/5 column (Pharmacia) equilibrated with 2 M ammonium sulfate in buffer B. The column is then eluted with a linear gradient of ammonium sulfate in buffer B from 2 to 0 M. Fractions containing P450nor are collected and dialyzed overnight against buffer A and loaded onto a Mono Q HR5/5 column (Pharmacia) equilibrated with the same buffer. Protein is eluted with a linear gradient of 0–0.5 M KCl in buffer A. Fractions containing purified P450nor are dialyzed against buffer A and stored at –80 until use. An aliquot is always taken at each step for Nor activity assay and SDS-PAGE. All purification manipulations should be conducted below 4 or on ice with the exception of phenyl-Superose and Mono Q column chromatographies, which are
121
Fungal Nitric Oxide Reductase
conducted at room temperature. Tnor is very abundant in yeast T. cutaneum; 2 mg Tnor is purified from only 63 g wet cells using the aforementioned method. The recovery of Tnor is half of the total soluble P450 proteins. The molecular mass is around 43 kDa, which is similar to Fnor and Cnor (42– 46 kDa) (Fig. 7.1). In addition, absorption spectra of the purified protein also indicate the characteristic of cytochrome P450 (Zhang et al., 2001). Absorption spectra of the purified protein in the resting state are in an oxidized form (ferric state) with a mixture of high-spin (390 nm) and low-spin (412 nm) states.
5. Nitric Oxide Reductase Activity Assay Nor activity can be assayed by measuring N2O production (Nakahara et al., 1993; Usuda et al., 1995) or NADH consumption (Kaya et al., 2004; Shiro et al., 1995). Typically, Tnor activity is assayed anaerobically in a 20-ml tube sealed with a double butyl rubber stopper. The reaction mixture
M
Sup DE52 (NH4)2SO4
P
Q10 Q11
kDa 94 67
Tnor 43
30
Figure 7.1 SDS-PAGE analysis of purified P450nor fromT. cutaneum. Lanes from left to right were, respectively, performed using 1, marker proteins containing carbonic anhydrase (30 kDa), ovalbumin (43 kDa), bovine serum albumin (67 kDa), and phosphorylase b (94 kDa); 2, soluble fraction (222 mg); 3, after DEAE52-cellulose column (180 mg); 4, after precipitation with ammonium sulfate (227 mg); 5, after phenyl-Superose column (53 mg); 6 and 7, fractions 10 and 11 after Mono Q column (15.6 and 6.2 mg, respectively). The Tnor band is cut out for trypsin treatment, HPLC, N-terminal amino acid sequencing.
122
Li Zhang and Hirofumi Shoun
(5 ml) contains 100 mM potassium phosphate buffer (pH 7.2), 10% glycerol, 0.1 mM EDTA, 0.1 mM DTT, 5 mM NADH or NADPH, and 4 nM purified Tnor. After replacing the gas phase with helium, the reaction mixture is prewarmed at 30 for 2 min. The reaction is initiated by adding 1.0 ml NO gas with a syringe. The mixture is vortexed frequently, and every 0.5-ml gas phase sample of the reaction is collected by a syringe at each time point for GC analysis. Figure 7.2 shows typical P450nor activity from purified Tnor using NADH as the electron donor. Turnover rates against NO (nmol NO/min/nmol P450nor) of purified Tnor using NADH and NADPH as electron donors are 1.27 and 1.05 104, respectively, which are almost the same as that of purified recombinant Fnor (1.28 104), Cnor1 (1.05 104), and Cnor2 (1.13 104) using NADH as the electron donor at 30 .
6. Protein Sequencing Purified protein (60 mg) is applied to SDS-PAGE and blotted onto a nitrocellulose membrane. The blot is stained with Ponceau S (Sigma), and the protein band is excised and digested partially with trypsin (Aebersold et al., 1987). The digested peptides are separated by HPLC on a reversedphase C18 column, and their N-terminal amino acid sequences are determined using a Perkin Elmer Precise 492 automated protein sequencer. Sequences of peptides show a very high identity to other P450nors, for example, to the Cnor1 amino acid sequence (Fig. 7.3).
N2O (nmol)
1200
800
400
0 0
5 10 Reaction time (min)
15
Figure 7.2 Nitric oxide reductase activity of P450nor fromT. cutaneum. Incubation was conducted anaerobically in a total volume of 5 ml with 5 mM NADH and 4 nM purified Tnor, and the total amounts of N2O were determined by GC with Porapak Q. , with NADH; ○, without NADH.
Fungal Nitric Oxide Reductase
123
STRAAPKFPFARASGMEPPA (N-terminus) MASEPPSFPFQRASGMEPPAEFARLRATDPVSKVKLFDGSLAWLVTKYKDVTFVATDERLS AAAQAQPTFVDMDAPDHMK (P8) KVRTRPGFPELNAGGKQAAKAKPTFVDMDAPDHMNQRGMVESLFTLEHVKKLQPYIQKTVD DLLAAMKKKGCANGPVDLVKEFALPVPSYIIYTILGVPFNDLDHLTNQNAIRTNGSSTARE ASAANQELLDYLASLVEKRLEEPKDDLISKLCTEQVKPGNIEKADAVQIAFLLLVAGNATM VNMIALGVVTLFQHPEQLAQLKANPSLAPQFVEELCRYHTASALAIKRTAKVDLEIGGKHI TFDKDPLAFGQGPHR (P13) VELTAVF KANEGIIASNQSANRDADIFENPDEFNMNRKWPAEDPLGYGFGPHRCIAEHLAKAELTTVF ETLYK (P21) DVGIVELPVT (P20) ATLFKEFPDLNIAVPFEKINFTPLGGDVGVVDLPVTF
Figure 7.3 Comparison of amino acid sequences of Tnor tryptic peptides with C. tonkinense Cnor1.
7. Isolation of cDNA Total DNA is prepared from T. cutaneum by the standard method (also see Tomura et al., 1994). In brief, about 5–10 g of yeast wet cells is freeze dried and disrupted by grinding in 8 ml of TE buffer (10 mM Tris-HCl, 1 mM EDTA, pH 8.0). Then 4 ml lysis buffer (50 mM Tris-HCl, pH 8.0, 100 mM EDTA, 100 mM NaCl, 1% SDS) is added and incubated at 65 for 30 min with gentle mixing. After adding 1/10th volume of 5 M (pH 5.2) potassium acetate, an equal volume of phenol-chloroform (1:1) is added and the mixture is incubated for another 30 min. After centrifugation at 11,000 rpm for 20–30 min, the aqueous phase is added with two-fold 100% ethanol and put in an –80 freezer for 30 min before centrifugation for another 30 min at 11,000 rpm. The pellet is washed with 70% ethanol, dissolved in 4 ml of TE with RNase, and incubated at room temperature overnight. DNA is extracted with phenol-chloroform another two to three times to completely get rid of proteins. The DNA is precipitated by ethanol, washed with 70% ethanol, and finally dissolved in 4 ml of TE buffer. Four mixed primers are designed according to the amino acid sequences of the purified P450nor and the trypsin digests (see Fig. 7.3, underlined): P1, 50 -CCNAARTTYCCNTTYGC-30 P2, 50 -TCNACDATNCCNACRTC-30 P3, 50 -GAYATGGAYGCNCCNGAYCAYATG-30 P4, 50 -GGRTCYTTRTCRAANGT-30 The first polymerase chain reaction (PCR) is conducted using 400 pmol of P1 and P2 primers each and 0.5 mg total DNA as the template. Amplification is performed by the process of denaturation at 94 for 1 min followed
124
Li Zhang and Hirofumi Shoun
by 30 cycles of incubations at 94 for 20 s, 50 for 1 min, and 72 for 1.5 min, and extension at 72 for 10 min. Second-round PCR is performed under the same conditions using the product of the first PCR as the template and P3 and P4 as the primers. The nucleotide sequence of the second PCR product is determined by the dideoxy chain-termination method (Sanger et al., 1977) using the automated DNA sequencer LONG READIR Model 4200 (LI-COR). The entire cDNA is cloned by the rapid amplification of cDNA ends (RACE) using the SMART RACE cDNA amplification kit (Clontech). Poly(A)þ mRNA is prepared from T. cutaneum cells using an mRNA purification kit (Pharmacia) according to the supplier’s instructions. Primers are designed based on the nucleotide sequence of the PCR product given earlier. The nucleotide sequences of the primers are GSP1, 50 -GGGTCCTTGTCAAACGTGCGGCGAATGT-30 NGSP1, 50 -CGCTGTAGTTACTCGCGATGATGCCCTC-30 (for 50 RACE) GSP2, 50 -CACATGAAGCAGCGCGGCCTCGTCGAGG-30 NGSP2, 50 -TCCAGAGCGTCATCGACGAGGCGCTCGA-30 (for 30 RACE) The RACE products are ligated to pGEM-T vectors (Promega) and their nucleotide sequences are determined. The deduced amino acid sequence from the sequence of cDNA contains the same sequences as those of the N terminus of native Tnor and its tryptic fragments (see Fig. 7.3). It contains 416 amino acids with a molecular mass of 45.2 kDa. The sequence shows overall similarities to those of other isoforms, with identities of 65.1, 65.3, and 64.8% to Fnor, Cnor1, and Cnor2, respectively.
8. Subcellular Fractionation of T. cutaneum By comparing the amino acid sequences of P450nor isoforms, it is interesting to note that the sequences of the N-terminal residues are not conserved among these proteins. The N-terminal portion of Fnor and Cnor1 contains hydrophobic and positively charged amino acid residues, which acts as the targeting signal for transportation to mitochondria; this portion is cleaved before the mature protein is formed. The purified Tnor protein also lacks the first 18 amino acid residues and therefore is considered to be a mitochondrial protein as well. To confirm its localization within mitochondria, yeast cells cultured under the denitrifying conditions described earlier are disrupted below 4 by grinding with a 2.5-fold (w/w) excess of quartz sand in 0.8 M sucrose buffer (isotonic conditions), pH 7.2, containing 0.1 mM EDTA, 10 mM
Fungal Nitric Oxide Reductase
125
Tris-HCl, and the protease inhibitors described previously. The quartz sand and undisrupted cells are removed by centrifugation at 1500g for 30 min. The cell homogenate is then subjected to centrifugation at 10,000g for 60 min. The resulting supernatant is separated into soluble (cytoplasm) and microsomal fractions by further centrifugation at 105,000g for 90 min. The precipitate resulting from the centrifugation at 10,000g described earlier (large particles) is resuspended in sucrose buffer and subjected to centrifugation at 105,000g for 180 min on a discontinuous sucrose density gradient (1.0, 1.26, and 1.46 M ). The resulting interface fractions are collected, and their activities of nitric oxide reductase and the mitochondrial marker enzyme (cytochrome c oxidase, COX) are measured. Most of the NADH- and NADPH-dependent Nor activities can be recovered together with COX activity, indicating that Tnor is localized in the mitochondria (Zhang et al., 2001).
9. Site-Directed Mutagenesis Of the five cloned CYP55A genes, CYP55A1 (Fnor) from F. oxysporum has been the most studied. Recombinant proteins of Fnor have been made in order to study the structure and function of P450nor (Kudo et al., 2001; Obayashi et al., 2000; Okamoto et al., 1997; Shimizu et al., 2000; Su et al., 2004; Umemura et al., 2004; Zhang et al., 2002b) (Table 7.1, primers used for mutagenesis can be found from the listed references). A study on the crystal structure of the P450nor double mutant GG (Gly73/Gly75) with substrate confirmed the importance of these amino acids in the P450nor reaction (Oshima et al., 2004). The P450nor isozymes isolated from fungi can be divided into two types, depending on their specificity for electron donors, NADH and NADPH. Fnor and Cnor1 are specific for NADH, whereas Cnor2, Tnor, and Anor can utilize both NADH and NADPH. Cnor2, Tnor, and Anor contain multiple Ala and Gly residues in the region 73SA(P)GGKAAA80. Using site-directed mutagenesis to modify the structure of Fnor, studies indicate that Ser73 and Ser75 in the B’-helix (74ASGKQA79) determine the specificity for NADH and NADPH (Zhang et al., 2002b). Site-directed mutagenesis is achieved by PCR using template pfp(450)20, which has the Fnor cDNA cloned in the pUC18 vector (Kizawa et al., 1991). Primers M13–47 and M13-RV (Takara, Otsu, Japan) are specific for pUC18. The following primers are designed to study the function of B’helix and its neighboring region (73SASGKQAA80 in Fnor) (mutated sites are underlined): 3A (74ASGKQA79 ! AAA): GAGCTTAGCGCCGCTGCAGCCAA GGCAAA
126
Li Zhang and Hirofumi Shoun
Table 7.1.
Functional amino acids in P450nor from F. oxysporum
Amino acid position
Thr
243
Ser286, Asp393 Arg64, Arg174 Ser73, Ser75 Asp88 Glu71
Function
References
Binding to nicotinic acid ring
Okamoto et al. (1997) Obayashi et al. (2000) Shimizu et al. (2000) Kudo et al. (2001)
Proton delivery Interact with pyrophospate moiety Access 2’-phosphate group of NADPH Release of NADþ Help Arg64 for NADH binding
Zhang et al. (2002b) Umemura et al. (2004) Su et al. (2004)
SG (Ser73/Gly75): CTTAGCGCCGGTGGAAAGCAAGCA GG (Gly73/Gly75): TTCCCTGAGCTTGGCGCCGGTGGAAAG The first PCR (30 cycles) using M13–47 and mutant forward primers or M13-RV and mutant reverse primers is performed as follows: 1 min for 94 , 2 min for 55 , and 2 min for 72 . The PCR products are purified from agarose gel and mixed in PCR buffer and H2O and are subjected to the following treatment: 10 min for 95 , 30 min with a gradient decrease of temperature from 95 to 37 , and 30 min for 37 . After the addition of Taq polymerase and dNTP, the mixture is kept for 10 min at 72 , after which M13–47 and M13-RV primers are added into the reaction mixture and PCR (30 cycles) is performed again as described earlier. The mutated DNA products are digested with restriction nucleases BssHII and BglII, and the 1.1-kb fragments are purified and inserted into the expression vector for P450nor [pT7nor, which comprises pRSET C (Invitrogen) with the T7 promoter in which the N-terminal His tag is replaced with the full-length P450nor cDNA] in lieu of the corresponding wild-type restriction fragments. The presence of the desired mutation is confirmed by DNA sequencing, and segments corresponding to the inserted restriction fragments are then sequenced completely to exclude the acquisition of unwanted mutations during amplification or cloning.
10. Expression of Recombinant Proteins Using the aforementioned strategy, the DNAs of Fnor, its derivatives, and Tnor in pT7nor vector are introduced individually into Escherichia coli JM109(DE3). The transformed cells are precultured overnight at 37 in
Fungal Nitric Oxide Reductase
127
20 ml of LA broth (1% tryptone, 0.5% yeast extract, 0.5% NaCl, 25 mg/ml ampicillin) supplemented with 0.5% glucose. Transformant precultures are transferred to 2 liters of LA broth in 5-liter Erlenmeyer flasks, shaken for 5–6 h, and further shaken at 120 rpm overnight at 30 in the presence of 1 mM isopropyl-1-thio-b-D-galactoside.
11. Purification of Recombinant Proteins The transformed cells are harvested, suspended in Tris buffer [10 mM Tris-HCl, pH 8.0, 10% (v/v) glycerol, 0.1 mM DTT, 0.1 mM EDTA, 0.1 mM PMSF, 0.1 mM TPCK] and disrupted twice using a French pressure cell press (SimAminco, France) at 20,000 psi. The suspension is centrifuged at 1500g for 30 min to remove cell debris and then centrifuged at 20,000g for 1 h. The resulting supernatant is dialyzed overnight against Tris buffer. A Whatman DEAE-cellulose column is equilibrated with Tris buffer, and then the sample is charged and eluted with a 0 to 0.4 M KCl gradient. Fractions containing P450nor are collected and dialyzed against Tris buffer. The P450nor fraction is purified further by chromatography through either Mono Q HR 5/5 or Resource Q columns (Pharmacia Biotech) in the same manner as described previously for DEAE. Using this method, the absorbance ratio of mutant proteins is always over 1.6 at 413 nm compared with that at 280 nm. Recombinant proteins of Fnor, its mutants, and Tnor have almost the same high-spin ratio between 54 and 59% in 50 mM N-Tris (hydroxymethyl) methyl-2-aminoethanesulfonic acid (TES) buffer (pH 7.2), suggesting that they have similar structures and the mutations do not affect overall structural conformation. The content of the high-spin form is used as an approximate value in comparing P450 with its mutants. The percentage of high-spin heme of P450nor at the resting state is estimated from the ratio of the absorbance of the high-spin form (390 nm) to that of the lowspin form (414 nm) as described previously (Imai and Komori, 1992; Imai et al., 1997). From data provided by Dr. Yoshio Imai through personal communication, we simply calculate spin ratios (%) using YH ¼ (101XH – 39.9)/(43.6 þ 56.7XH) and YL ¼ (83.5XL – 44.3)/(56.7 þ 43.6XL) in which YH and YL represent high-spin (%) and low-spin (%) ratios, respectively, while XH ¼ A390/A414 and XL ¼ A414/A390, respectively. For example, the purified Tnor at ferric state gives absorbance of 0.454 and 0.47 at 390 and 414 nm, respectively (Zhang et al., 2001); therefore, XH ¼ 0.966. From the formula just given, the high-spin ratio YH ¼ 58.6%. In order to investigate the possible role of B’-helix in the access of NAD(P)H, mutant 3A in which the 74ASGKQA79 is replaced by three alanines to destroy the B’-helix can be utilized. NADH-dependent
128
Li Zhang and Hirofumi Shoun
Nor activity of mutant 3A decreases to the same level as the NADPHdependent activity of wild-type Fnor, indicating its crucial role in conferring full Nor activity (Zhang et al., 2002b). Mutations at Ser73 and Ser75 improve NADPH-dependent Nor activity significantly without affecting NADH-dependent activity (Fig. 7.4), indicating that these amino acids are crucial for determining the specificity of P450nor for NADH and NADPH.
12. Titration of NAD Analogs Although the spectral perturbation of P450nor-bound heme by some ligands can be observed (Okamoto et al., 1998), it is difficult to detect spectral perturbation upon mixing P450nor and NADH, possibly because of the powerful absorbance of NADH that overwhelms the spectral perturbation of bound heme. Because SG and GG mutant proteins can enhance the rate of reduction of NO with NADPH, it is possible to screen NAD(P)H analogs to detect spectral perturbation using such mutants. NAD(P)H and NAD(P) analogs (all from Sigma) are used to investigate their binding to P450nor. The screening of NAD(P) analogs b-NADþ, b-NADPþ, and 500 WT_NADH WT_NADPH SG_NADPH GG_NADPH
N2O (nmol)
400
300
200
100
0 0
1
2
3
Min
Figure 7.4 B’-helix of Fnor modulate the interaction with NADPH. Nor activity of WT (Ser73/Ser75) with NADPH as electron donor is only 13% of that with NADH as electron donor. SG (Ser73/Gly75) and GG (Gly73/Gly75) mutant proteins with NADPH as electron donor exhibit 63 and 79% activities of WT with NADH. NAD(P)H, 1 mM; P450nor,4 nM.
129
Fungal Nitric Oxide Reductase
70
70
60
60
50
50
40
1/ΔA
1/ΔA
3-pyridinealdehyde adenine dinucleotide (PAAD) are found to cause a detectable spectral change in WT and SG mutant proteins. Therefore, they can be analyzed further by spectrophotometric titration using a Beckman DU 7500 spectrophotometer. The dissociation constant (Kd) is calculated from the difference in absorbance (DA) at 413 nm from that at 395 nm. Recombinant P450nor (5 mM) in 50 mM TES buffer at pH 7.2 is mixed with an equal volume (100 ml) of each substrate analog (dissolved in the same buffer) and the spectrum of the mixture (200 ml volume) is recorded. 3-Pyridinealdehyde adenine dinucleotide and b-NADþ are found to give a type I spectral change (the increase in absorbance around 390 nm because of the high-spin state heme with a concomitant decrease in absorbance at 413 nm because of the low-spin heme) in Fnor, SG, GG, and Tnor (Oshima et al., 2004; Zhang et al., 2002b). It is interesting to note that while b-NADPþ induces a reverse type I spectral change in Fnor and its mutants, it causes a type I spectral change in Tnor, which can utilize both NADH and NADPH as electron donors. Thus, the reverse type I spectral change appears to be caused by binding in an unfavorable state. It is obvious that these NAD(P)H analogs can bind to P450nor in two alternative states within the bound form that cause either a type I or a reverse type I spectral change. It is also evident from the effects of these mutations in the B’-helix region that these analogs bind to P450nor from the distal side. The spectral perturbation caused by NADH analogs shows saturation against the ligand concentration, suggesting a specific binding of each ligand to Fnor and its mutants. The dissociation constant (Kd) of each Fnor-ligand complex is determined by spectrophotometric titration (Fig. 7.5). Maximum spectral changes of Fnor WT and SG proteins by PAAD are the same (0.0627 and 0.0628, respectively), whereas Kd values are 6.11 and 1.65 mM,
WT Kd = 6.11 mM
30
SG Kd = 1.65 mM
30 20
20 y = 15.947 + 97.385x R2 = 0.985
10 0
40
0.0
y = 15.913 + 26.239x R2 = 0.991
10 0
0.1
0.2
0.3 0.4 1/PAAD
0.5
0.6
0
1
2
3
1/PAAD
Figure 7.5 Double-reciprocal plot of PAAD concentration and the absorbance change at 395 and 413 nm of FnorWTand mutant SG.
130
Li Zhang and Hirofumi Shoun
respectively, showing much higher affinity of NADH analogs to mutants, which led to the success in obtaining the crystal structure for the PAAD and GG complex (Oshima et al., 2004).
13. Stopped-Flow Rapid Scan Analysis The overall reaction can be divided into three partial reactions (Schemes 1–3) (Shiro et al., 1995): Fe3þ þ NO ! Fe3þ-NO (Scheme 1) Fe3þ-NO þ NAD(P)H þ Hþ ! I þ NAD(P)þ (Scheme 2) I þ NO ! Fe3þ þ N2O þ H2O (Scheme 3) Except for the last step (Scheme 3), these partial reactions can be observed as an isolated reaction. The chemical nature of the specific intermediate (I ) with a Soret absorption peak at 444 nm, which is formed upon reduction of the ferric-NO complex (Fe3þ-NO) with NAD(P)H (Scheme 2), is still under debate (Daiber et al., 2002, 2005; de Groot et al., 2005; Lehnert et al., 2006; Silaghi-Dumitrescu, 2003). Because the mutations at Ser73 and/or Ser75 in Fnor modulate the interaction with NADPH as well as overall Nor activity, the intermediate (I ) can be observed as an isolated reaction by rapid scan analysis as reported (Shiro et al., 1995). The P450nor reducing half-reaction is analyzed by following the appearance of the intermediate (I ) at 444 nm upon reduction of the Fe3þ-NO complex with NAD(P)H at 10 using a Unisoku rapid scan analyzer (Osaka, Japan) where 50 mM TES buffer (pH 7.2) is used, instead of phosphate buffer or Tris-HCl buffer, to eliminate the effects of anions on Nor activity (Kudo et al., 2001). The P450nor enzyme (final, 5 mM) in Fe3þ-NO is mixed with an equal volume of NADH or NADPH (final concentration, 20 mM ) anaerobically and the spectral changes are recorded. The gate time is set to 1 ms and the rate of I formation (kobs) is calculated. Under controlled conditions, accumulation of the intermediate (I ) is detectable upon reduction of Fe3þ-NO with NADH in WT Fnor (Scheme 2), whereas it is not available with NADPH (Zhang et al., 2002b). This is possibly because reduction (I formation) is slower than its decomposition. However, the accumulation of I upon reduction with NADPH as well as with NADH can be seen in both SG and GG mutants, showing that the rate of the reduction with NADPH was enhanced by the mutations. The apparent rate constant for the reduction at a fixed NADH (or NADPH) concentration (kobs) is obtained with each mutant from the time-dependent spectral change during the process of I formation (Zhang et al., 2002b), which clearly demonstrates that NADHor NADPH-dependent kobs for each enzyme species is parallel with the overall activity (see Fig. 7.4), indicating that the enhanced overall NADPH-dependent Nor activities in SG or GG arise from the reduction step.
Fungal Nitric Oxide Reductase
131
14. Other Analysis Protein concentration is determined by the standard method. Cytochrome c oxidase activity is measured by the method of Orii and Okunuki (1965). The P450nor concentration is determined spectrophotometrically from the CO difference spectrum as reported (Omura and Sato, 1964) using an extinction coefficient of 86.3 mM1cm1 for the difference in absorbance between 448 and 490 nm (Nakahara et al., 1993). NADH is determined from the absorbance at 340 nm using a molar absorption coefficient of 6.22 mM1cm1.
15. Conclusion The P450nor product is available commercially (Gekkeikan, Japan), with the goal of lowering NO levels within the human body. The importance of NO biology on human health has been well documented. Additionally, plants use NO as a weapon against invading pathogens. This chapter described some basic protocols on screening, purification, cloning, and functional analysis of cytochrome P450nor, providing some information and methods that may be helpful in further NO research.
ACKNOWLEDGMENTS Thanks are given to Professor Yoshio Imai for his kind discussion and suggestion of spin ratio calculations. We also thank R.W. Cameron Dingle and Georgia Zeigler, Department of Molecular and Biomedical Pharmacology, University of Kentucky, for reading and suggestions on the manuscript.
REFERENCES Aebersold, R. H., Leavitt, J., Saavedra, R. A., Hood, L. E., and Kent, S. B. H. (1987). Internal amino acid sequence analysis of proteins separated by one- or two-dimensional gel electrophoresis after in situ protease digestion on nitrocellulose. Proc. Natl. Acad. Sci. USA 84, 6970–6974. Daiber, A., Nauser, T., Takaya, N., Kudo, T., Weber, P., Hultschig, C., Shoun, H., and Ullrich, V. (2002). Isotope effects and intermediates in the reduction of NO by P450NOR. J. Inorg. Biochem. 88, 343–352. Daiber, A., Shoun, H., and Ullrich, V. (2005). Nitric oxide reductase (P450nor) from Fusarium oxysporum. J. Inorg. Biochem. 99, 185–193. de Groot, M. T., Merkx, M., Wonders, A. H., and Koper, M. T. (2005). Electrochemical reduction of NO by hemin adsorbed at pyrolitic graphite. J. Am. Chem. Soc. 127, 7579–7586.
132
Li Zhang and Hirofumi Shoun
Imai, Y., and Komori, M. (1992). In ‘‘Shin Seikagaku Jikken Koza,’’ Vol. 5, pp. 205–221. Japanese Biochemical Society, Tokyo-kagakudojin, Tokyo. Imai, Y., Okamoto, N., Nakahara, K., and Shoun, H. (1997). Absorption spectral studies on heme ligand interactions of P-450nor. Biochem. Biophys. Acta 1337, 66–74. Kaya, M., Matsumura, K., Higashida, K., Hata, Y., Kawato, A., Abe, Y., Akita, O., Takaya, N., and Shoun, H. (2004). Cloning and enhanced expression of the cytochrome P450nor gene (nicA; CYP55A5) encoding nitric oxide reductase from Aspergillus oryzae. Biosci. Biotechnol. Biochem. 68, 2040–2049. Kizawa, H., Tomura, D., Oda, M., Fukamizu, A., Hoshino, T., Gotoh, O., Yasui, T., and Shoun, H. (1991). Nucleotide sequence of the unique nitrate/nitrite-inducible cytochrome P-450 cDNA from F. oxysporum. J. Biol. Chem. 266, 10632–10637. Kudo, T., Takaya, N., Park, S.-Y., Shiro, Y., and Shoun, H. (2001). A positively charged cluster formed in the heme-distal pocket of cytochrome P450nor is essential for interaction with NADH. J. Biol. Chem. 276, 5020–5026. Kudo, T., Tomura, D., Liu, D., Dai, X., and Shoun, H. (1996). Two isozymes of P450nor of Cylindrocarpon tonkinense: Molecular cloning of the cDNAs and genes, expressions in the yeast, and the putative NAD(P)H-binding site. Biochimie 78, 792–799. Lehnert, N., Praneeth, V. K. K., and Paulat, F. (2006). Electronic structure of iron(II)– porphyrin nitroxyl complexes: Molecular mechanism of fungal nitric oxide reductase (P450nor). J. Comput. Chem. 27, 1338–1351. Nakahara, K., and Shoun, H. (1996). N-terminal processing and amino acid sequence of two isoforms of nitric oxide reductase cytochrome P450nor from F. oxysporum. J. Biochem. (Tokyo) 120, 1082–1087. Nakahara, K., Tanimoto, T., Hatano, K., Usuda, K., and Shoun, H. (1993). Cytochrome P-450 55A1 (P-450dNIR) acts as nitric oxide reductase employing NADH as the direct electron donor. J. Biol. Chem. 268, 8350–8355. Obayashi, E., Shimizu, H., Park, S.-Y., Shoun, H., and Shiro, Y. (2000). Mutation effects of a conserved threonine (Thr243) of cytochrome P450nor on its structure and function. J. Inorg. Biochem. 82, 103–111. Okamoto, N., Imai, Y., Shoun, H., and Shiro, Y. (1998). Site-directed mutagenesis of the conserved threonine (Thr243) of the distal helix of fungal cytochrome P450nor. Biochemistry 37, 8839–8847. Okamoto, N., Tsuruta, K., Imai, Y., Tomura, D., and Shoun, H. (1997). Fungal P450nor: Expression in Escherichia coli and site-directed mutagenesis at the putative distal region. Arch. Biochem. Biophys. 337, 338–344. Omura, T., and Sato, R. (1964). The carbon monoxide-binding pigment of liver microsomes. II. Solubilization, purification and properties. J. Biol. Chem. 239, 2379–2385. Orii, Y., and Okunuki, K. (1965). Studies on cytochrome a. XV. Cytochrome oxidase activity of the Okunuki preparation and its activation by heat, alkali and detergent treatments. J. Biochem. (Tokyo) 58, 561–568. Oshima, R., Fushinobu, S., Su, F., Zhang, L., Takaya, N., and Shoun, H. (2004). Structural evidence for direct hydride transfer from NADH to cytochrome P450nor. J. Mol. Biol. 342, 207–217. Sanger, F., Nicklen, S., and Coulson, A. R. (1977). DNA sequencing with chainterminating inhibitors. Proc. Natl. Acad. Sci. USA 74, 5463–5467. Shimizu, H., Obayashi, E., Gomi, Y., Arakawa, H., Park, S.-Y., Nakamura, H., Adachi, S., Shoun, H., and Shiro, Y. (2000). Proton delivery in NO reduction by fungal nitric-oxide reductase: Cryogenic crystallography, spectroscopy, and kinetics of ferric-NO complexes of wild-type and mutant enzymes. J. Biol. Chem. 275, 4816–4826. Shiro, Y., Fujii, M., Iizuka, T., Adachi, S., Tsukamoto, K., Nakahara, K., and Shoun, H. (1995). Spectroscopic and kinetic studies on reaction of cytochrome P450nor with nitric
Fungal Nitric Oxide Reductase
133
oxide: Implication for its nitric oxide reduction mechanism. J. Biol. Chem. 270, 1617–1623. Shoun, H. (2006). Denitrification and anaerobic energy producing mechanisms by fungi. Tanpakushitsu Kakusan Koso. 51, 419–429. Shoun, H., Kim, D. H., Uchiyama, H., and Sugiyama, J. (1992). Denitrification by fungi. FEMS Microbiol. Lett. 73, 277–281. Shoun, H., and Tanimoto, T. (1991). Denitrification by the fungus Fusarium oxysporum and involvement of cytochrome P-450 in the respiratory nitrite reduction. J. Biol. Chem. 266, 11078–11082. Silaghi-Dumitrescu, R. (2003). Nitric oxide reduction by heme-thiolate enzymes (P450nor): A reevaluation of the mechanism. Eur. J. Inorg. Chem. 2003, 1048–1052. Su, F., Fushinobu, S., Takaya, N., and Shoun, H. (2004). Involvement of a Glu71-Arg64 couple in the access channel for NADH in cytochrome p450nor. Biosci. Biotechnol. Biochem. 68, 1156–1159. Takaya, N., and Shoun, H. (2000). Nitric oxide reduction, the last step in denitrification by F. oxysporum, is obligatorily mediated by cytochrome P450nor. Mol. Gen. Genet. 263, 342–348. Tomura, D., Obika, K., Fukamizu, A., and Shoun, H. (1994). Nitric oxide reductase cytochrome P-450 gene, CYP55, of the fungus Fusarium oxysporum containing a potential binding-site for FNR, the transcription factor involved in the regulation of anaerobic growth of E. coli. J. Biochem. (Tokyo) 116, 88–94. Tsuruta, S., Takaya, N., Zhang, L., Shoun, H., Kimura, K., Hamamoto, M., and Nakase, T. (1998). Denitrification by yeasts and occurrence of cytochrome P450nor in Trichosporon cutaneum. FEMS Microbiol. Lett. 168, 105–110. Umemura, M., Su, F., Takaya, N., Shiro, Y., and Shoun, H. (2004). D88A mutant of cytochrome P450nor provides kinetic evidence for direct complex formation with electron donor NADH. Eur. J. Biochem. 271, 2887–2894. Usuda, K., Toritsuka, N., Matsuo, Y., Kim, D.-H., and Shoun, H. (1995). Denitrification by the fungus Cylindrocarpon tonkinense: Anaerobic cell growth and two isozyme forms of cytochrome P-450nor. Appl. Environ. Microbiol. 61, 883–889. Zhang, L., Kudo, T., Takaya, N., and Shoun, H. (2002a). Distribution, structure and function of fungal nitric oxide reductase P450nor: Recent advances. Int. Congr. Ser. 1233, 197–202. Zhang, L., Kudo, T., Takaya, N., and Shoun, H. (2002b). The B’ helix determines cytochrome P450nor specificity for the electron donors NADH and NADPH. J. Biol. Chem. 277, 33842–33847. Zhang, L., Takaya, N., Kitazume, T., Kondo, T., and Shoun, H. (2001). Purification and cDNA cloning of nitric oxide reductase cytochrome P450nor (CYP55A4) from Trichosporon cutaneum. Eur. J. Biochem. 268, 3198–3204.
This page intentionally left blank
C H A P T E R
E I G H T
A Quantitative Approach to Nitric Oxide Inhibition of Terminal Oxidases of the Respiratory Chain Maria G. Mason,* Rebecca S. Holladay,* Peter Nicholls,* Mark Shepherd,† and Chris E. Cooper* Contents 1. Introduction 2. Evaluation of Current Techniques for Measuring pNO, pO2, and KM (O2) 3. Nitric Oxide Donor Compounds 4. Nitric Oxide Kinetics 4.1. Measurement of IC50 (NO) 4.2. Factors that may influence IC50 values 4.3. Measurement of KD (NO) 4.4. Pitfalls of nitric oxide kinetic analysis 4.5. Comparison of dynamic and steady state IC50 NO measurements 4.6. Measurement of NOoff rates and estimation of NOon rates 5. Oxygen Kinetics 6. Optical Detection of Enzyme Intermediates in the Presence of Oxygen and NO 6.1. Spectro-electrode system 6.2. Identification of nitric oxide-bound intermediates Appendices Appendix A. Correction for electrode response time Appendix B. Instrumentation for measuring nitric oxide and oxygen kinetics Appendix C. Equipment suppliers Acknowledgments References
* {
136 137 138 139 139 141 141 144 146 146 149 151 152 153 153 153 153 155 156 156
Department of Biological Sciences, University of Essex, Colchester, United Kingdom University of Sheffield, Western Bank, Sheffield, United Kingdom
Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37008-0
#
2008 Elsevier Inc. All rights reserved.
135
136
Maria G. Mason et al.
Abstract Inhibition of terminal respiratory oxidases by nitric oxide (NO) plays important physiological roles in signaling and host defense. Using a bacterial quinol oxidase and mitochondrial cytochrome c oxidase, this chapter describes simple polarographic methods to quantify the kinetic characteristics of inhibition by NO. This chapter points out the inherent pitfalls of both experimental design and data analysis and compares alternative methods. Additionally, it describes a system designed to acquire polarographic and spectral data simultaneously to permit identification of spectral intermediates under defined conditions.
1. Introduction Nitric oxide (NO) inhibition of bacterial and mitochondrial terminal respiratory chain oxidases is increasingly being seen as playing a key physiological role in host defense (Fang, 1997; Pieters and Ploegh, 2003) and signaling (Carreras and Poderoso, 2007). NO can reversibly bind, irreversibly damage, or be metabolized by oxidases (Cooper, 2002). In mitochondria, nitric oxide inhibition can perturb electron transfer chain redox states at sites distinct from the oxidase without directly compromising energy metabolism (Quintero et al., 2006); thus it can play a role in cell signaling, possibly via modulating the production of reactive oxygen species (Palacios-Callender et al., 2004; Sarkela et al., 2001). Quantitative measurements are crucial to the interpretation of the importance of these interactions. At sufficiently high levels there are few biomolecules that do not react with NO. Therefore, understanding at what level NO may interact with an oxidase and measuring NO levels in vivo are both necessary. In addition to the importance for cell physiology (Cooper and Giulivi, 2007; Giulivi et al., 2006), analyzing the mechanism of inhibition can also inform an understanding of the chemistry of the interactions (Mason et al., 2006). However, quantitative measurements in this area are difficult. NO is an unstable compound that can react at multiple sites in cells (Cooper, 1999); indeed even individual oxidases can have multiple, and distinct, interactions with NO (Cooper, 2002). The purpose of this chapter is to describe methods for the measurement of the kinetic constants associated with nitric oxide inhibition of oxygen consumption by respiratory chain oxidases. It focuses on numbers derived primarily from steady state enzyme kinetics. It largely ignores the wealth of spectroscopic data describing NO/oxidase interactions, although it illustrates a system for measurement of the nature of optically detectable intermediates at defined levels of nitric oxide and oxygen. Because it is impossible to analyze nitric oxide inhibition of oxygen consumption without a parallel discussion of the methods used to measure
Nitric Oxide Inhibition of Oxidases
137
the oxygen kinetics of respiration, optimum methods are described for measuring low levels of both substrate (O2) and inhibitor (NO). We use examples from the interaction of NO with a bacterial quinol oxidase from Escherichia coli and mitochondrial cytochrome c oxidase. However, the analysis methods are appropriate for any enzyme that consumes oxygen and is inhibited reversibly by nitric oxide, particularly if the ligands share the same site of interaction. In the case of quinol and cytochrome c oxidase, the oxygen reactive site is a high-spin ferrous heme center.
2. Evaluation of Current Techniques for Measuring pNO, pO2, and KM (O2) For a real time non-invasive measure of steady state nitric oxide partial pressure and/or concentration, polarography is currently the only viable technique. A number of sensors have been developed, including optical (Barker et al., 1999a,b), electrochemical (Wang et al., 2005), and dualmode amperometric/voltammetric (Malinski and Taha, 1992). However, these sensors are not yet available commercially. Optical (Larfars and Gyllenhammar, 1995) and paramagnetic ( James and Swartz, 2002) measurements can integrate NO accumulation over time, but generally do so by removing nitric oxide, thus perturbing the system in the process. However, for steady state oxygen measurements, a range of techniques are available (Swartz et al., 1997). Some of these directly measure pO2, e.g., most Clark-type polarographic electrodes and electron paramagnetic resonance (EPR) oximetry based on solid particles; others measure parameters that respond to the product of the diffusion rate and the concentration of oxygen, e.g., probes of absorbance (D’Mello et al., 1995), phosphorescence (Lo et al., 1996; Wilson et al., 1988), fluorescence ( Ji et al., 2002), and EPR probes based on soluble materials (Swartz et al., 1997). The distinction between pO2 and [O2] is important for in vivo experiments where the oxygen solubility and diffusion rate might be unknown, but is less critical for the in vitro measures of the type described here, where the measurements can be readily interconverted. The systems we study are always likely to produce biologically active molecules, e.g., nitrogen oxides, superoxide, and peroxide. Any measurement requires a range of controls to ensure that the probe is not being perturbed or causing a perturbation. Therefore, in general, we prefer using nonbiologically active systems such as oxygen electrodes or the current range of artificial phosphorescent and fluorescent probes rather than heme proteins such as myoglobin or hemoglobin. Most terminal oxidases are membrane proteins. Therefore, oxygen reacts with these oxidases in an active site separated from the bulk phase by the protein interior and a biological (cell membrane) or artificial
138
Maria G. Mason et al.
(detergent micelle) membrane phase. As oxygen is measured in the external medium, it could be argued that oxygen diffusion and solubility in the membrane phase distort the oxygen kinetics of terminal membrane-bound oxidases. However, measurements of the interactions of hydrophobic gases (CO, O2, or NO) with oxidases seem, fortunately, insensitive to the nature of the aqueous/hydrophobic interface (Meunier and Rich, 1998). Extreme caution is required when measuring oxygen in the presence of NO, as it can perturb the measurement in two ways: either NO can act directly on the probe or the probe can affect the NO concentration. In the case of hemoglobin or myoglobin as an oxygen probe, NO can both alter the oxygen p50 ( Jia et al., 1996; Yonetani et al., 1998) and be scavenged by the oxygenated heme protein itself (Larfars and Gyllenhammar, 1995). Scavenging is likely to be less of an issue for nonbiological phosphorescent and fluorescent probes; although the probe signal depends on physical quenching by oxygen, it is unlikely that the intrinsic signal will be unduly affected by the presence of NO, as its concentration is likely to be at least 100 times less than that of oxygen in the types of experiments described here. However, potential artifacts should always be assessed by the use of suitable controls. Under certain conditions, similar problems can arise with polarographic electrodes, as cross talk can occur between dissolved oxygen and the nitric oxide probe, and vice versa; however, these conditions are unlikely to occur during steady state measurements of NO inhibition. Fluorescent or phosphorescent oxygen probes rely on oxygen perturbing a strong intrinsic signal. Therefore, they are most sensitive at low oxygen concentrations and run the risk of saturation at high pO2. In contrast, the sensitivity of polarographic electrodes is weakest at low oxygen tensions as the probe directly measures pO2; therefore, specialized, expensive highresolution oxygen electrode systems are needed for the low micromolar range (Gnaiger et al., 1995). The optimal choice of pO2 probe may therefore in some situations be dependent on the range of oxygen concentrations being studied.
3. Nitric Oxide Donor Compounds We commonly use nitric oxide donor compounds as an alternative to solutions saturated with NO gas. The principal advantages of using NO donors are as follow: stock solutions are relatively stable at high pH with sensible handling (i.e., use at 5 and store at –20 under N2); the concentration and any degradation of the donor compound can be checked readily by UV spectroscopy; freshly made stock solutions do not contain nitrate/ nitrite as a contaminant (which is difficult to remove completely from solutions of NO gas); and NO donor solutions can be prepared at much
139
Nitric Oxide Inhibition of Oxidases
higher concentrations than those of the gas. A suitable donor for nitric oxide kinetics is the fast-releasing NO donor Proli NONOate (t1/2 ¼ 1.8 s at 37 in 0.1 M KPi at pH 7.4) (from Alexis Biochemicals, http://www.axxora.com).
4. Nitric Oxide Kinetics 4.1. Measurement of IC50 (NO) The most common method for determining the IC50 for NO is to measure the recovering oxygen respiration rate simultaneously with declining nitric oxide concentration, after inhibition by NO (Borutaite and Brown, 1996; Brown and Cooper, 1994; Griffiths and Garthwaite, 2001). An example of this type of experiment is shown in Fig. 8.1. Respiration by the oxidase is initiated by the addition of substrate, and the oxygen concentration begins to decrease. Once respiration reaches a steady rate, an aliquot of nitric oxide (or nitric oxide donor) solution is added to the chamber and respiration becomes inhibited. Substrate
3.0
250
B
NO donor
2.5
0.8
200
150 1.5 100 1.0 50
0.5
[O2] mM
[NO] mM
2.0
Respiration rate (mM O2 s−1)
A
0.6
0.4
0.2
0.0
0.0 0
100
200 300 Time (s)
400
0 500
0
0.5
1
[NO] mM
Figure 8.1 Measurement of IC50 (NO). (A) Time-dependent traces of oxygen (black line) and nitric oxide (grey triangles) concentration. Oxygen decreases as E. coli membranes respire on NADH until nitric oxide is added at approximately 100 s and respiration is inhibited. Respiration recovers as nitric oxide decays. Data shown within the boxed area are transformed into nitric oxide concentration-dependent respiration rates; (B) data points (þ) are fitted to the Hill equation ( ) to obtain values for IC50 and h (also see text). Membranes from E. coli strain RKP4544 (Stevanin et al., 2000) were grown (Gibson et al., 1977) and prepared (Poole and Haddock, 1974) as described previously. Experiments were performed using combined NO/O2 polarography (see Appendix B) in a light-impenetrable chamber at 35 in 0.1 M KPi, pH 7.4.
140
Maria G. Mason et al.
Nitric oxide does not persist in aerobic solutions as it reacts with oxygen (Ford et al., 1993) (Eq. (8.1)) and other materials such as membrane lipids (O’Donnell et al., 1997) and reduced cytochrome c (Sharpe and Cooper, 1997) used commonly in experiments of this type.
4NO þ O2 þ 2H2 O ! 4NO2- þ 4Hþ
ð8:1Þ
Inhibition of cytochrome c oxidase by NO is generally reversible (Brown and Cooper, 1994) and continues only while NO remains in solution. Data obtained during the recovery of respiration (boxed area in Fig. 8.1) are transformed to give respiration rates as a function of [NO], shown plotted in Fig. 8.1B, together with the fit to an adapted form of the Hill equation (Eq. (8.2)). This fitting procedure gives values for IC50 and h. Traditionally, the Hill equation is used to determine the number of molecules bound; however, in this type of analysis, the value for h should not be interpreted in this way, as values greater than 1 are obtained from this procedure for cytochrome c oxidase, which requires only one molecule of NO to inhibit, and the value of h relates to the turnover number and the degree of buffering of inhibition (Mason et al., 2006). During partial inhibition the reductive pressure on the uninhibited population increases if the turnover is less than maximal (Chance et al., 1970) and consequently inhibition is buffered more at low turnover and h has a greater value.
v ¼V
V ½ih Kh0:5 þ ½ih
ð8:2Þ
where v is velocity, V is limiting velocity, [i] is inhibitor concentration, K0.5 is the inhibitor concentration ([i]) giving 50% V, and h is the Hill coefficient. Certain substrates, especially of a nonphysiological type, react directly with oxygen. This nonenzymatic autooxidation is evident when an enzyme cannot be completely inhibited. Equation (8.2) can be modified to account for this inherent autooxidation (see Eq. (8.3)). The oxygen concentrationdependent autooxidation rate is determined either by adding increasing amounts of NO until further decreases in respiration are not observed, (the residual rate is then the autooxidation rate) or by performing separate control experiments, in the absence of oxidase, at a number of oxygen concentrations (autooxidation is negligible below 40 mM O2 with ascorbate/ N,N,N 0 ,N 0 -tetramethyl-p-phenylenediamine).
141
Nitric Oxide Inhibition of Oxidases
v ¼ Va
V ½ih Kh0:5 þ ½ih
ð8:3Þ
where Va is the sum of the autooxidation rate and the limiting velocity.
4.2. Factors that may influence IC50 values An important consideration when undertaking these types of experiments is to ensure that all factors are carefully controlled, such as temperature, incident light, and substrate concentration. This is because the IC50 values obtained depend on all these factors. Temperature is important because NO binding and, more importantly, dissociation rates are temperature-dependent (Sawicki and Gibson, 1977). Additionally, experiments are performed in a lightimpenetrable chamber because the bonds between nitric oxide and ferrous heme a3 and, to a lesser extent, cupric CuB are photosensitive (Borutaite et al., 2000; Sarti et al., 2000; Wever et al., 1985). Finally, if the experimental results are to be interpreted in a meaningful way, the enzyme turnover number must be known. This is for two reasons: (i) the predominance of the different pathways (competitive and noncompetitive with oxygen) of inhibition is determined by the electron flux and therefore turnover number of the enzyme and (ii) the competitive interaction is turnover number-dependent because the apparent affinity for oxygen (KM0 O2) depends on the turnover number. Consequently, the apparent IC50 for NO is also dependent on the turnover number (Antunes et al., 2004; Mason et al., 2006). To illustrate the effect that experimental conditions can have, literature values of IC50 (NO) for cytochrome c oxidase are shown in Fig. 8.2, plotted as a function of oxygen concentration. The data are so scattered is because values were obtained under different experimental conditions of enzyme turnover, temperature, and light.
4.3. Measurement of KD (NO) An example of experimental results obtained where the temperature and enzyme turnover were carefully controlled is shown in Fig. 8.3. Values for slope and intercept were obtained by regression analysis. Because of the relatively small error on the value of the slope (i.e., IC50 dependence on oxygen concentration), we can be confident that it is quite accurate. However, the large error on the value of the intercept, KD (NO), means that it is rather inaccurate and therefore not a reliable way of determining the KD. A more accurate way to determine the KD for a competitive
142
Maria G. Mason et al.
1500
Apparent IC50 (NO) nM
1200
900
600
300
0 0
50
100 [O2] mM
150
200
Figure 8.2 Literature values of IC50 (NO) for cytochrome c oxidase. Oxygen concentration-dependent IC50 values for NO are shown from several literature sources (Bellamy et al., 2002; Borutaite and Brown,1996; Brookes et al., 2003; Brown and Cooper, 1994; Hollis et al., 2003; Koivisto et al., 1997; Lizasoain et al., 1996; Mason et al., 2006; Palacios-Callender et al., 2004; Torres et al., 1995). The distribution of values is scattered because it depends not only on oxygen concentration (Brown and Cooper, 1994), but also on other factors, such as enzyme turnover number (Antunes et al., 2004; Mason et al., 2006) and light (Borutaite et al., 2000; Sarti et al., 2000).Values are from experiments using purified enzyme, mitochondria, and whole cells.
inhibitor is from the relationship described in Eq. (8.4), between the apparent IC50 (NO), the substrate (oxygen) concentration, and the KM for oxygen (see later); it is essential that the KM (O2) is measured under the same experimental conditions as those used to obtain IC50 values.
IC50 0 ¼ KD þ KD ½S=KM
ð8:4Þ
(Cornish-Bowden, 2004). Once the KM (O2) has been determined, mean IC50 values can be plotted as a function of O2/KM (O2). Both the slope and the intercept are now dependent on KD, and the analysis is more robust. Figure 8.4 shows mean data from Fig. 8.3 replotted and fitted to Eq. (8.4). Although the error on the intercept is still large, the error on the slope is small and the value of KD is determined with greater precision. It is worth emphasizing that this type of analysis is for a purely competitive inhibitor and therefore if the interaction is not purely competitive the
143
Nitric Oxide Inhibition of Oxidases
450 400
apparent IC50 (nM NO)
350 300 250 200 150 100 50 0 0
20
40
60
80 [O2] mM
100
120
140
160
Figure 8.3 Oxygen concentration-dependent IC50 (NO) values in a controlled oxidase system. IC50 (NO) values for E. coli membranes, as a function of oxygen concentration, were obtained as described in Fig. 8.1 and using Eq. (8.1). Turnover number (244 e s16 SE) and temperature (35) were controlled carefully, and the reaction was performed in a light-impenetrable chamber. The solid line shows the fit from regression analysis, which gave values of 2.0 ( 0.24 SE) for the slope and 2.6 ( 25 SE) for the intercept. For other details, see legend to Fig. 8.1.
apparent IC50 (nM NO)
400
300
200
100
0 0
200
400 [O2] / KM(O2)
600
800
Figure 8.4 Measurement of KD (NO). IC50 (NO) values from Fig. 8.3 were binned into three different oxygen tension groups and the mean values of IC50 are plotted as a function of [O2]/KM (O2). Data were fitted to Eq. (8.4) before averaging.The solid line shows the fit from regression analysis, which gave values of 0.54 ( 0.06 SE) for the slope and 2.6 ( 25 SE) for the intercept. For other details, see legend to Fig. 8.3.
144
Maria G. Mason et al.
fit may be poor. Mixed inhibition by nitric oxide has been demonstrated in mitochondrial cytochrome c oxidase (Mason et al., 2006); in addition to binding at the ferrous heme a3 of the binuclear center, in competition with oxygen (Brown and Cooper, 1994), there is also a noncompetitive inhibitory interaction that predominates when electron flux, and therefore enzyme turnover, is low and a high proportion of the enzyme is in the oxidized form (Mason et al., 2006; Sarti et al., 2000). Although mixed inhibition may also occur in other heme-copper oxidases, it has not yet been verified.
4.4. Pitfalls of nitric oxide kinetic analysis A number of pitfalls are associated with this type of analysis; an important consideration is the response time of the electrodes. Although oxygen electrode response times are relatively fast (t1–3 s) and are not usually corrected for in this type of experiment, this is not the case with nitric oxide electrodes. There are a number of commercially available NO electrodes with different characteristics, but generally the faster responding electrodes have greater noise and the slower responding electrodes have less noise. Figure 8.5 shows representative traces from similar inhibition experiments using NO electrodes with different response times; the NO trace in the lower panel is from a fast-responding electrode (t 4 s) with maximal baseline noise equivalent to 50 nM NO, whereas that in the upper panel is from a slow-responding NO electrode with no discernible noise (i.e., 1 nM NO resolution). In both traces the respiration rate returns to its maximal value just before 400 s (boxed areas). In the lower trace the recovery to maximal respiration corresponds to disappearance of NO, as expected, whereas in the upper trace (slow electrode response time) at the point of recovery to maximal respiration, the NO electrode records residual NO of approximately 800 nM. This is clearly an artifact of the response time of the electrode. The current challenge for nitric oxide electrode design and manufacture is for a fast-responding nitric oxide electrode with subnanomolar resolution, which is not perturbed by commonly used reductants such as ascorbate. Slow electrode response times can be corrected for (see Appendix A). However, the benefit of using a noise-free electrode is then lost by the noise introduced from the differential used in the correction process. Additionally, it is important to confirm that the response time of the electrode is the same upon a sudden increase in NO concentration as it is for a sudden decrease in NO concentration, as slow-response time electrodes do not always have this characteristic. The response time should also be NO concentration-independent.
145
Nitric Oxide Inhibition of Oxidases
NO
1000
0.6
Substrate
[NO] nM
0.2 0
0
100
200
300
400
500
600
0.0 700 0.6
1000
0.4 500
Respiration rate ( mM O2 s−1)
0.4 500
0.2 0
0
100
200
300 400 Time (s)
500
600
0.0 700
Figure 8.5 Nitric oxide electrode response time is critical for accurate IC50 measurements. Experimental traces showing respiration rates (grey lines) of mitochondrial cytochrome c oxidase together with nitric oxide concentrations (black lines).The upper and lower panels compare results obtained using different NO electrodes with slow (t 14 s) and fast (t 4 s) response times.Traces obtained from the electrode with a fast response time show simultaneous disappearance of NO and recovery of full respiration rate (boxed region, lower panel); conversely, traces obtained from the electrode with a slow response time show that relief of inhibition (boxed region, upper panel) is not simultaneous with disappearance of NO.This apparent residual NO is an artifact of the electrode response time. Bovine cytochrome c oxidase, purified as described previously (Mason et al., 2006; Yonetani, 1960), was respiring on ascorbate plus cytochrome c (TN 10 e s1 of aa31) in 0.1 M KPi/0.1% lauryl maltoside, pH 7.4, at 35 in a light-impenetrable chamber using combined NO/O2 polarography (see Appendix B).
The only way to check if the electrode response time needs correcting or not is to correct and compare; if the portion of the trace required for analysis is different in the corrected and uncorrected traces, then correction is required; if they overlay, correction is not required. Although not recommended for this type of dynamic experiment, lownoise, slow-responding electrodes are useful for measurement of stable steady state NO concentrations. The creation of true NO steady states, and measurement of the respective enzyme inhibition, has only been attempted by the group of Garthwaite (Bellamy et al., 2002; Griffiths and Garthwaite, 2001; Griffiths et al., 2003). It has the advantage that it creates a situation analogous to ‘‘normal’’ enzyme kinetic measurements (i.e., with a stable inhibitor concentration). It is not a trivial process, requiring a balance of NO production by a
146
Maria G. Mason et al.
suitable slow-release NO donor and NO removal via oxygen or an added scavenger. The oxygen concentration-dependence of NO removal, both directly (Lewis and Deen, 1994) and via cells and cell membranes (Shiva et al., 2001), further complicates the measurement. In contrast to the dynamic measurement, each experimental run creates only a single data point; multiple measurements are required to generate an IC50 value. However, a priori one would expect this method to generate the ‘‘true’’ IC50.
4.5. Comparison of dynamic and steady state IC50 NO measurements The dynamic method will only approximate the steady state method if the enzyme samples multiple steady states during the relief of inhibition. Given the rather slow NO dissociation rate from some terminal oxidases, this is not necessarily true. To our knowledge there has been no direct test of dynamic and steady state methods. The inhibition constants derived from both methods are broadly similar, but as already pointed out, a meaningful comparison would have to be done under identical conditions because the NO inhibition constants depend on so many variables. We have developed a simple dynamic model of NO inhibition of an oxidase, competitive with oxygen (Cooper and Giulivi, 2007). We then calculated the true steady state NO inhibition constants analytically and compared them to those generated from the model, following a bolus addition of nitric oxide. Figure 8.6 shows that provided the NOoff rate is at least 0.1 s1, dynamic and steady state methods produce identical results. However, at slower NOoff rates the dynamic method may measure artificially low IC50 values, especially at higher oxygen concentrations.
4.6. Measurement of NOoff rates and estimation of NOon rates Ligand dissociation rates can be measured as the return of respiratory activity after addition of a reagent that scavenges the inhibitory ligand. For accuracy, this type of experiment is best performed using a high-resolution respirometer. Traces from an experiment of this type, using the Oxygraph 2K instrument, are shown in Fig. 8.7. Respiration of E. coli membranes is initiated by the addition of NADH and then an aliquot of the fast-release NO donor Proli NONOate is added to inhibit respiration. Immediately after addition of NO donor, an increase in the respiration rate is observed; this is because of the reaction of nitric oxide with oxygen (Eq. (8.1)). Once respiration is fully inhibited, an excess of the NO scavenger 2-(4-carboxyphenyl)-4,4,5,5-tetramethylimidazoline-1-oxyl-3-oxide (cPTIO) is added, causing a rapid and complete depletion of nitric oxide and simultaneous recovery of the
147
Nitric Oxide Inhibition of Oxidases
Oxygen consumption
300
Competitive inhibition
k3 [O2]
250
k1 E
k4 [NO]
k2 ES
ER
ERNO k−4
Oxygen-dependent NO removal
IC50 (NO) nM
200 NO + O2
k5 [O2] [NO]
NO−2
NO off rate 0.1 s−1
150
NO off rate 0.02 s−1
100
50
0 0
50
100 [O2] mM
150
200
Figure 8.6 Use of a dynamic model to compare different methods of measuring NO IC50. A dynamic model of nitric oxide inhibition of mitochondrial respiration was used (inset) to add a bolus of NO (0.5^2 mM) at different oxygen concentrations to the enzyme in turnover. Modeled time traces and derived plots of the type shown in Fig. 8.1 were then created.These were used to create a graph of IC50 versus [O2] for two different NOoff rates (note that the NOon rate was also changed so that the KD remained unaltered). Modeled data (points) were then compared to analytically derived values (straight line) created via solving the appropriate steady state rate equation (via King^ Altman methods). Values used in the model were k1 ¼ 103 s1; k2 ¼ 325 s1; k3 ¼ 400 mM1 s1; k4 ¼ 100 mM1 s1; k-4 ¼ 0.02 s1; k5 ¼ 0.0001. The enzyme concentration was 3 nM. For the faster NOoff rate: k4 ¼500 mM1 s1; k-4 ¼ 0.1 s1. Note that the mechanism and values for removal of NO via oxygen were chosen solely to create traces that approximated original data. In reality, NO removal occurs via multiple processes, including a direct reaction that is second order in [NO](see Eq. (8.1)).
respiration rate. Data acquired during the recovery of respiration are then corrected for electrode response time, as described in Appendix A. The corrected trace, together with an exponential fit, is shown in Fig. 8.7B. The mean functional NO dissociation rate of 0.163 s1 obtained by this method is in close agreement with that of 0.133 s1 reported by Borisov et al. (2007) for the same oxidase, and the relatively small difference can be accounted for by the difference in experimental temperature.
148
Maria G. Mason et al.
A 9
B
300
NO donor Substrate
0.20 200 [O2] mM
[NO] mM
100
3
0
0
500 Time (s)
1000
Respiration rate (mM O2 s−1)
NO scavenger
6
0.15
0.10
0.05
0.00
0
0 20 40 Time after NO removal (s)
Figure 8.7 Measurement of NO dissociation rate. (A) Uncorrected traces of oxygen (black line) and nitric oxide (grey triangles) concentrations during an experiment in which NO is scavenged so that the functional NO dissociation rate can be measured. Respiration is initiated by the addition of substrate (NADH), and then an aliquot of the fast NO-releasing donor Proli NONOate is added to fully inhibit respiration; finally the NO scavenger cPTIO is added and inhibition is relieved.The rate of recovery of respiration after scavenging NO is shown (B), after correction for electrode response time and changes in oxygen background, together with exponential fit for the functional NO dissociation rate.The NO dissociation rate (0.163 s1) for this enzyme is approaching the limit of resolution for this technique and therefore a number of replicates are necessary. Experiments were performed using combined NO/high resolution O2 polarography (see Appendix B). For other details, see legend to Fig. 8.1.
In these experiments we used cPTIO as a nitric oxide scavenger. However, cPTIO needs to be used with caution as it can react directly with some reductants, such as ascorbate. In these circumstances, oxyhemoglobin can be used as an alternative NO scavenger, although it should be used above its p50 (O2) to avoid release of oxygen that interferes with the oxygen respiration trace, making it difficult to measure the recovery rate. Oxymyoglobin is not a suitable alternative as it has a much lower p50 than oxyhemoglobin ( Wittenberg, 1974). Once the NO dissociation rate and KD have been measured, the NO association rate can be estimated from the relationship.
kon ¼
koff KD
ð8:5Þ
Nitric Oxide Inhibition of Oxidases
149
5. Oxygen Kinetics In order to analyze optimally the nitric oxide inhibition kinetics, it is necessary to have a measurement of the oxygen KM under identical conditions to those of the NO inhibition kinetics. As this measurement is made in the absence of NO, the earlier critique of O2 measurement systems that are NO sensitive is not relevant. As a result, we have the full panoply of systems to choose from. A full analysis and critique of the literature surrounding the measurement of the oxygen KM of terminal oxidases, both isolated and in cells, are beyond the scope of this chapter. Gnaiger and colleagues provided a useful comparison of the different methods available in 1995; the methods available, and their associated pitfalls, have not changed noticeably since then. Perhaps the key difference is between open and closed systems (i.e., whether a gas phase is present in the study). Removal of all effects of the gas phase requires careful system design to reduce back diffusion and oxygen stores; even the choice of an oxygen-rich Teflon-coated stirrer bar can make a big difference to measurements at low oxygen. Although these effects are not present in open systems (Brookes et al., 2003; Cole et al., 1982; Petersen et al., 1976), oxygen transfer between gaseous and aqueous phases can give rise to problems with unstirred boundary layers and oxygen gradients between the sensor and the sample. For this reason we favor a sealed system with no gas phase. In this case, similar values are obtained whether using high-resolution polarographic or optical techniques (Gnaiger et al., 1995). There is a major caveat in using polarography, however. It should be noted that a standard Clark-type electrode is unable to measure an oxygen KM in the low micromolar range (Schindler, 1967). This is because of a low intrinsic sensitivity at low [O2], resulting in few data points in the critical region. To obtain more data points, the oxygen consumption rate must be decreased into a regime where oxygen back diffusion becomes a major artifact. We have seen many supposedly sealed oxygen electrode systems where even an enzyme with a very low KM cannot decrease the [O2] to <1 mM; the system enters a steady state where oxygen back diffusion balances enzymatic oxygen consumption. If oxygen consumption is inhibited by NO, for example, the oxygen trace shows an apparent evolution of oxygen by the respiratory oxidase! Fortunately, the advent of high-resolution respirometers has made such extreme traces a much less common occurrence in the literature. However, we illustrate that without high-resolution respirometry, the measurement of KM (O2) is rather inaccurate. The traces in Fig. 8.8 are from identical experiments, except that the upper trace was obtained from a highresolution respirometer and the lower trace was from a standard Clark-type oxygen electrode. The upper trace has been corrected for electrode response time and oxygen background (as described earlier) and fits a rectangular
150
Maria G. Mason et al.
High resolution oxygen electrode, corrected Respiration rate ( mM O2 s−1)
0.3
0.2 Standard oxygen electrode, uncorrected
0.1
0.0
0
5
10
15
20
25
30
[O2] mM
Figure 8.8 Measurement of KM (O2).Traces show respiration rates as a function of oxygen concentration.The upper trace was measured using a high-resolution respirometer (see Appendix B) after correction of data for electrode response time and background changes, including diffusion of oxygen into the chamber that occurs at low oxygen concentrations (Gnaiger, 2001).The lower trace was measured using a standard oxygen electrode and has not been corrected for electrode response time or background oxygen changes. Solid lines show the best fit to the Michaelis^Menten equation with KM (O2) values of 0.27 mM (upper trace) and 1.1 mM (lower trace). Both experiments were performed under identical conditions; for other details, see legend to Fig. 8.1.
hyperbola with a KM of 0.27 mM O2, whereas the fit for the lower uncorrected trace is four times greater with a KM (O2) of 1.1 mM. In the absence of high-resolution respirometry, the measurement of KM (O2) may be made in the presence of a competitive inhibitor, for which the KD is known or can be measured accurately. The apparent KM (O2) is measured at a number of inhibitor concentrations and is plotted as a function of [I], and data are then fitted to the equation for a simple competitive inhibitor (Eq. (8.6)); by forcing the fit line to give intercept/ slope ¼ KD, the value for KM in the absence of an inhibitor can be solved. It should be noted that this method will have limited use, depending on the properties of the instrumental system and the oxygen affinity of the enzyme. Whichever method is used, it is of paramount importance that the conditions, especially enzyme turnover, which affects KM (O2), are identical to those used to determine the IC50 (NO) values. apparent
KM
(Cornish-Bowden, 2004).
¼ KM þ
KM ½I KD
ð8:6Þ
Nitric Oxide Inhibition of Oxidases
151
6. Optical Detection of Enzyme Intermediates in the Presence of Oxygen and NO In order to understand the mechanism, as opposed to the kinetics, of nitric oxide inhibition it is useful to be able to detect the enzyme intermediates simultaneously with the O2 and NO concentrations. There have been a number of multipurpose cuvette systems that allow simultaneous optical and electrochemical detection. These have been particularly developed by the photosynthetic research community (Kraayenhof et al., 1982). Given the current design of electrodes, it is also possible to sample in a standard cuvette, sealed or unsealed (Rogers et al., 1995). This has the advantage that the spectral measurements are identical to other studies in a laboratory. However, a flexible, dedicated system can be put together at a relatively inexpensive cost using current spectrometers and fiber optics. A key component is the cuvette system itself and dedicated software to integrate electrode data with optical spectra. Although commercial systems can be adapted, there is always some requirement for self-build and customwritten software. Two fiber-optic systems have been developed and integrated with the polarographic measurements of nitric oxide and oxygen: a visible light system (VLS) described in Hollis et al. (2003) and our own spectro-electrode system (SES) designed here (see later). Apart from the fact that the VLS system uses reflectance and the SES system uses transmittance, both systems use similar principles. To ensure synchronization with the electrode system, optical detection is optimally multiwavelength (otherwise, in a dynamic system, spectra measured at one end of the wavelength range will report on different oxygen and NO concentrations from that at the other end) and electrode response times must be corrected for. There are a number of detector options to choose from. An inexpensive, portable charge-coupled device (CCD) (e.g., Ocean Optics, Avantes, Newport) is perfectly acceptable for most uses. If light is limiting, a back-illuminated and/or cooled CCD can be preferable. For the ultimate in sensitivity (and cost!), a cooled twodimensional CCD array is probably best. However, if the sample is optically transparent (low light scattering), then a diode array detector coupled to a strong light source is likely to match this in signal:noise at lower expense. In our earlier descriptions of methods to quantify nitric oxide kinetic and inhibition constants, we stressed the need to exclude light when undertaking these experiments, yet here we are using light to detect the inhibitory species. We should therefore emphasize that using high levels of light to detect an inhibitory species that is photodissociable (Borutaite et al., 2000; Sarti et al., 2000; Wever et al., 1985) can itself perturb the oxygen kinetics, the extent of which can only be evaluated by comparing results with those obtained under conditions of darkness.
152
Maria G. Mason et al.
6.1. Spectro-electrode system The SES (see Fig. 8.9) is a custom-built (in-house) instrument that permits simultaneous optical spectroscopy and measurements of dissolved oxygen and nitric oxide. It consists of a light-impenetrable housing chamber (Hansatech DW2) containing a 2-ml borosilicate reaction vessel, connected to a water circulating bath, for temperature control. The housing chamber has five acrylic optical ports. In this setup, two of the ports are used and those unused are blanked off to prevent external light interference. Fiber-optic cables (Ocean Optics QP600–2-UV-BX) are used to direct light from the deuterium halogen light source (Mikropak DH2000 BAL) to the sample and to collect light transmitted through the sample to a fiber-optic CCD or diode array photometer (Ocean Optics). The light guides are fitted with collimating lenses (Edmund Optics) to focus the light through the sample in the reaction chamber. The collimator/fiber-optic assembly is held at the
Fibre optic cables Light source Port Diode array/ CCD detector
Cuvette
Nitric oxide electrode
A/D card
Oxygen electrode Collimating lenses
Figure 8.9 The spectro-electrode system measures the optical spectrum simultaneously with nitric oxide and oxygen concentrations. The cylindrical cuvette, which acts as the reaction chamber, has an oxygen electrode at its base, whereas the nitric oxide electrode is inserted through a port in the chamber stopper. The chamber housing has ports for fiber-optic cables to guide light from the lamp through the sample and to the photometer. Electrode and optical data are digitized and displayed simultaneously on a single screen via custom-written software.The sample is maintained at a constant temperature by water from a recirculating thermostated bath (not shown). For full details of instrumentation, see the text.
Nitric Oxide Inhibition of Oxidases
153
optical port by a cylindrical adapter fitted with a nylon thumb screw. The Hansatech Clark-type oxygen electrode sits at the base of the reaction chamber and is connected to an electrode control unit (Hansatech). The chamber stopper, machined from Macor, accommodates the WPI ISO-NO nitric oxide electrode and has a sample injection port. Analogue signals from the oxygen electrode and WPI ISO-NO Mark II nitric oxide meter are converted to a digital signal by an A/D converter (National Instruments). Custom-written acquisition software allows simultaneous online display of nitric oxide and oxygen traces, and spectral data.
6.2. Identification of nitric oxide-bound intermediates The spectro-electrode system quantifies the amount of cytochrome c oxidase ferrous heme a3-nitrosyl complex formed under specific conditions and allows the user to relate this to the redox status of cytochrome c. Results from an experiment of this type are shown in Fig. 8.10 (see legend for details). With this instrumental setup it will also be possible to investigate the extent to which nitric oxide perturbs the steady state levels of spectral intermediates (Torres et al., 1998) in the cytochrome oxidase catalytic cycle (Wikstrom and Morgan, 1992).
Appendices Appendix A. Correction for electrode response time The response time of the electrode is corrected using Eq. (8.7):
½A ¼ ½Araw þðt d½Araw =dtÞ
ð8:7Þ
where [A] is the concentration of A, after correcting for electrode response time; [A]raw is the concentration of A, prior correction for electrode response time; t is the time taken for the electrode to respond to 63% of maximal change after a rapid change in [A], and d[A]raw/dt is the uncorrected rate of change of [A].
Appendix B. Instrumentation for measuring nitric oxide and oxygen kinetics Combined O2 and NO polarography Simultaneous measurements of dissolved oxygen and nitric oxide are made in a 2.5-ml glass chamber with a glass base, housing the Clark-type oxygen electrode, which is connected to a Digital Model 10 controller (all from
154
Maria G. Mason et al.
NO Cytochrome oxidase Anaerobiosis NO cyt c
0.06
550−540
Δ Abs
A
0.02
0.00
0
200
400
600
800
1000
1200
[O2] mM
200
1400 3 2
100 1 0 0
200
400
600 800 Time (s)
1000
1200
0 1400
Δ Abs (B minus A)
598−612
B
[NO] mM
0.04
605 598
0.08
550
0.03 510
560 610 Wavelength (nm)
660
Figure 8.10 Spectroscopic investigation into nitric oxide-bound species in the cytochrome oxidase catalytic cycle. Using the spectro-electrode system, oxygen and nitric oxide concentrations were monitored together with spectral changes. In the lower panel, oxygen (black line) and nitric oxide (grey line) concentrations are shown for a typical experiment, during which respiration is inhibited by the addition of nitric oxide. The upper panel shows simultaneous optical changes for (reduced minus oxidized) cytochrome c (grey line) monitored at 550^540 nm and for the heme a3^NO complex of cytochrome c oxidase (black line) monitored at 598^612 nm. Upon addition of cytochrome c, an increase in amplitude of 550^540 nm Dabsorbance is observed, as cytochrome c is reduced by ascorbate; however, when cytochrome c oxidase is added, a decrease of similar magnitude is observed, indicating that cytochrome c is mainly oxidized during the steady state in this experiment. After the addition of nitric oxide, both Dabsorbance traces increase in amplitude because of formation of the inhibitory heme a32þ^NO complex, and an accumulation of reduced cytochrome c. As NO decays in solution, the concentration of both these species decreases. Once anaerobiosis has been reached and cytochrome c has become fully reduced, a further aliquot of NO is added to reveal the maximal heme a3^NO complex of cytochrome c oxidase.The difference spectrum, B minus A (points indicated on the 598- to 612-nm Dabsorbance trace), shown in the RHS panel has peaks at 605, 598, and 550 nm.The 605- and 598-nm peaks are diagnostic of reduced heme a and the heme a32þ -NO complex of cytochrome oxidase, respectively, whereas the 550-nm peak is from reduced cytochrome c. Other experimental conditions are the same as those described in Fig. 8.5, except that cytochrome c oxidase TN 0.5 e s1 aa31.These spectra were acquired on a low-cost CCD; the signal:noise could be improved upon by using a diode array or two-dimensional CCD.
Rank Brothers Ltd.). The chamber is connected to a water bath, to enable temperature control of the sample, and is shielded from ambient light with tin foil. The chamber stopper, either machined in-house from Macor rods, an oxygen-impermeable, machinable glass ceramic (from Corning), or custom made by Rank Bros., has an injection port and an aperture to
Nitric Oxide Inhibition of Oxidases
155
accommodate a nitric oxide electrode. The nitric oxide electrode (ISONOP) is connected to an ISO-NO Mark II meter (both from World Precision Instruments). Data from both oxygen and nitric oxide systems are collected by a data acquisition system (MacLab 8E Powerlab) and analyzed with commercially available software (Chart) (both from ADInstruments). The software allows instantaneous data analysis and fitting to user-defined equations. Oxygraph 2K; high-resolution O2 and NO polarography The Oroboros Oxygraph 2K (from Oroboros instruments) is a highresolution respirometer that comes with analysis software that corrects for two important factors when measuring oxygen kinetics: the response time of the electrode and the oxygen background, caused by instrumental and chemical effects. Instrumental effects are due to oxygen consumption by the electrode at high oxygen tension and back diffusion of oxygen into the chamber at low oxygen tension and are instrument-dependent, whereas chemical artifacts are usually caused by oxygen concentration-dependent oxidation of nonphysiological substrates, such as ascorbate, and depend on experimental conditions. The instrument has polarographic oxygen sensors housed beneath the twin chambers and also has two BNC connectors so that an additional polarographic sensor can be inserted into the modified stoppers of each chamber. When used to acquire simultaneous NO and O2 electrode data, the ISO-NO Mark II meter is connected to one of the BNC ports, and the NO electrode is inserted through the modified chamber stopper. The signal from the NO electrode is amplified by a factor of 50, and the range, as shipped, is 10 V. In practical terms, this means that a maximum of 5 mM NO can be measured before going off scale, although the potentiometers on the underside of the instrument allow adjustment of the gain and offset.
Appendix C. Equipment suppliers ADInstruments Pty Ltd, Australia, http://www.adinstruments.com Avantes B.V., The Netherlands, http://www.avantes.com Corning, USA, http://www.corning.com/ Edmund Optics Ltd., York, UK, http://www.edmundoptics.com Hansatech, King’s Lynn, UK, http://www.hansatech-instruments.com Mikropak GmbH, Ostfildern, Germany, http://www.mikropack.de National Instruments Corp., Texas, http://www.ni.com Newport Corporation, California, http://www.newport.com Ocean Optics, Florida, http://www.oceanoptics.com Oroboros Instruments, Innsbruck, Austria, http://www.oroboros.at Rank Brothers Ltd., Cambridge, UK, http://www.rankbrothers.co.uk
156
Maria G. Mason et al.
World Precision Instruments Ltd., Stevenage, UK, http://www. wpi-europe.com
ACKNOWLEDGMENTS Thanks to Kevin Oxborough for SES software, to Tony Jordan, Richard Ranson, and Julie Double for technical support, and to The Wellcome Trust and the British Biotechnology and Biological Sciences Research Council (BBSRC) for financial support.
REFERENCES Antunes, F., Boveris, A., and Cadenas, E. (2004). On the mechanism and biology of cytochrome oxidase inhibition by nitric oxide. Proc. Natl. Acad. Sci. USA. 101, 16774–16779. Barker, S. L., Clark, H. A., Swallen, S. F., Kopelman, R., Tsang, A. W., and Swanson, J. A. (1999a). Ratiometric and fluorescence-lifetime-based biosensors incorporating cytochrome c0 and the detection of extra- and intracellular macrophage nitric oxide. Anal. Chem. 71, 1767–1772. Barker, S. L., Zhao, Y., Marletta, M. A., and Kopelman, R. (1999b). Cellular applications of a sensitive and selective fiber-optic nitric oxide biosensor based on a dye-labeled heme domain of soluble guanylate cyclase. Anal. Chem. 71, 2071–2075. Bellamy, T. C., Griffiths, C., and Garthwaite, J. (2002). Differential sensitivity of guanylyl cyclase and mitochondrial respiration to nitric oxide measured using clamped concentrations. J. Biol. Chem. 277, 31801–31807. Borisov, V. B., Forte, E., Sarti, P., Brunori, M., Konstantinov, A. A., and Giuffre, A. (2007). Redox control of fast ligand dissociation from Escherichia coli cytochrome bd. Biochem. Biophys. Res. Commun. 355, 97–102. Borutaite, V., and Brown, G. C. (1996). Rapid reduction of nitric oxide by mitochondria, and reversible inhibition of mitochondrial respiration by nitric oxide. Biochem. J. 315 (Pt 1), 295–299. Borutaite, V., Budriunaite, A., and Brown, G. C. (2000). Reversal of nitric oxide-, peroxynitrite- and S-nitrosothiol-induced inhibition of mitochondrial respiration or complex I activity by light and thiols. Biochem. Biophys. Acta. 1459, 405–412. Brookes, P. S., Kraus, D. W., Shiva, S., Doeller, J. E., Barone, M. C., Patel, R. P., Lancaster, J. R., Jr., and Darley-Usmar, V. (2003). Control of mitochondrial respiration by NO*, effects of low oxygen and respiratory state. J. Biol. Chem. 278, 31603–31609. Brown, G. C., and Cooper, C. E. (1994). Nanomolar concentrations of nitric oxide reversibly inhibit synaptosomal respiration by competing with oxygen at cytochrome oxidase. FEBS Lett. 356, 295–298. Carreras, M. C., and Poderoso, J. J. (2007). Mitochondrial nitric oxide in the signaling of cell integrated responses. Am. J. Physiol. Cell Physiol. 292, C1569–C1580. Chance, B., Erecinska, M., and Wagner, M. (1970). Mitochondrial responses to carbon monoxide toxicity. Ann. N. Y. Acad. Sci. 174, 193–204. Cole, R. P., Sukanek, P. C., Wittenberg, J. B., and Wittenberg, B. A. (1982). Mitochondrial function in the presence of myoglobin. J. Appl. Physiol. 53, 1116–1124. Cooper, C. E. (1999). Nitric oxide and iron proteins. Biochem. Biophys. Acta 1411, 290–309.
Nitric Oxide Inhibition of Oxidases
157
Cooper, C. E. (2002). Nitric oxide and cytochrome oxidase: Substrate, inhibitor or effector? Trends Biochem. Sci. 27, 33–39. Cooper, C. E., and Giulivi, C. (2007). Nitric oxide regulation of mitochondrial oxygen consumption. II. Molecular mechanism and tissue physiology. Am. J. Physiol. Cell Physiol. 292, C1993–C2003. Cornish-Bowden, A. (2004). ‘‘Fundamentals of Enzyme Kinetics.’’ Portland Press Ltd., London. D’Mello, R., Hill, S., and Poole, R. K. (1995). The oxygen affinity of cytochrome bo0 in Escherichia coli determined by the deoxygenation of oxyleghemoglobin and oxymyoglobin: Km values for oxygen are in the submicromolar range. J. Bacteriol. 177, 867–870. Fang, F. C. (1997). Perspectives series: Host/pathogen interactions. Mechanisms of nitric oxide-related antimicrobial activity. J. Clin. Invest. 99, 2818–2825. Ford, P. C., Wink, D. A., and Stanbury, D. M. (1993). Autoxidation kinetics of aqueous nitric oxide. FEBS Lett. 326, 1–3. Gibson, F., Cox, G. B., Downie, J. A., and Radik, J. (1977). A mutation affecting a second component of the F0 portion of the magnesium ion-stimulated adenosine triphosphatase of E. coli K12: The uncC424 allele. Biochem. J. 164, 193–198. Giulivi, C., Kato, K., and Cooper, C. E. (2006). Nitric oxide regulation of mitochondrial oxygen consumption. I. Cellular physiology. Am. J. Physiol. Cell Physiol 291, C1225–C1231. Gnaiger, E. (2001). Bioenergetics at low oxygen: Dependence of respiration and phosphorylation on oxygen and adenosine diphosphate supply. Respir. Physiol. 128, 277–297. Gnaiger, E., Steinlechner-Maran, R., Mendez, G., Eberl, T., and Margreiter, R. (1995). Control of mitochondrial and cellular respiration by oxygen. J. Bioenerg. Biomembr. 27, 583–596. Griffiths, C., and Garthwaite, J. (2001). The shaping of nitric oxide signals by a cellular sink. J. Physiol. 536, 855–862. Griffiths, C., Wykes, V., Bellamy, T. C., and Garthwaite, J. (2003). A new and simple method for delivering clamped nitric oxide concentrations in the physiological range: Application to activation of guanylyl cyclase-coupled nitric oxide receptors. Mol. Pharmacol. 64, 1349–1356. Hollis, V. S., Palacios-Callender, M., Springett, R. J., Delpy, D. T., and Moncada, S. (2003). Monitoring cytochrome redox changes in the mitochondria of intact cells using multiwavelength visible light spectroscopy. Biochem. Biophys. Acta 1607, 191–202. James, P. E., and Swartz, H. M. (2002). Simultaneous detection of pO2 and NO by electron paramagnetic resonance. Methods Enzymol. 359, 52–66. Ji, J., Rosenzweig, N., Jones, I., and Rosenzweig, Z. (2002). Novel fluorescent oxygen indicator for intracellular oxygen measurements. J. Biomed. Opt 7, 404–409. Jia, L., Bonaventura, C., Bonaventura, J., and Stamler, J. S. (1996). S-nitrosohaemoglobin: A dynamic activity of blood involved in vascular control. Nature 380, 221–226. Koivisto, A., Matthias, A., Bronnikov, G., and Nedergaard, J. (1997). Kinetics of the inhibition of mitochondrial respiration by NO. FEBS Lett. 417, 75–80. Kraayenhof, R., Schuurmans, J. J., Valkier, L. J., Veen, J. P., Van Marum, D., and Jasper, C. G. (1982). A thermoelectrically regulated multipurpose cuvette for simultaneous time-dependent measurements. Anal. Biochem. 127, 93–99. Larfars, G., and Gyllenhammar, H. (1995). Measurement of methemoglobin formation from oxyhemoglobin: A real-time, continuous assay of nitric oxide release by human polymorphonuclear leukocytes. J. Immunol. Methods 184, 53–62. Lewis, R. S., and Deen, W. M. (1994). Kinetics of the reaction of nitric oxide with oxygen in aqueous solutions. Chem. Res. Toxicol. 7, 568–574.
158
Maria G. Mason et al.
Lizasoain, I., Moro, M. A., Knowles, R. G., Darley-Usmar, V., and Moncada, S. (1996). Nitric oxide and peroxynitrite exert distinct effects on mitochondrial respiration which are differentially blocked by glutathione or glucose. Biochem. J. 314(Pt 3), 877–880. Lo, L. W., Koch, C. J., and Wilson, D. F. (1996). Calibration of oxygen-dependent quenching of the phosphorescence of Pd-meso-tetra (4-carboxyphenyl) porphine: A phosphor with general application for measuring oxygen concentration in biological systems. Anal. Biochem. 236, 153–160. Malinski, T., and Taha, Z. (1992). Nitric oxide release from a single cell measured in situ by a porphyrinic-based microsensor. Nature 358, 676–678. Mason, M. G., Nicholls, P., Wilson, M. T., and Cooper, C. E. (2006). Nitric oxide inhibition of respiration involves both competitive (heme) and noncompetitive (copper) binding to cytochrome c oxidase. Proc. Natl. Acad. Sci. USA 103, 708–713. Meunier, B., and Rich, P. R. (1998). Quantitation and characterization of cytochrome c oxidase in complex systems. Anal. Biochem. 260, 237–243. O’Donnell, V. B., Chumley, P. H., Hogg, N., Bloodsworth, A., Darley-Usmar, V. M., and Freeman, B. A. (1997). Nitric oxide inhibition of lipid peroxidation: Kinetics of reaction with lipid peroxyl radicals and comparison with alpha-tocopherol. Biochemistry 36, 15216–15223. Palacios-Callender, M., Quintero, M., Hollis, V. S., Springett, R. J., and Moncada, S. (2004). Endogenous NO regulates superoxide production at low oxygen concentrations by modifying the redox state of cytochrome c oxidase. Proc. Natl. Acad. Sci. USA 101, 7630–7635. Petersen, L. C., Nicholls, P., and Degn, H. (1976). The effect of oxygen concentration on the steady-state kinetics of the solubilized cytochrome c oxidase. Biochem. Biophys. Acta 452, 59–65. Pieters, J., and Ploegh, H. (2003). Microbiology: Chemical warfare and mycobacterial defense. Science 302, 1900–1902. Poole, R. K., and Haddock, B. A. (1974). Energy-linked reduction of nicotinamide– adenine dinucleotide in membranes derived from normal and various respiratorydeficient mutant strains of Escherichia coli K12. Biochem. J. 144, 77–85. Quintero, M., Colombo, S. L., Godfrey, A., and Moncada, S. (2006). Mitochondria as signaling organelles in the vascular endothelium. Proc. Natl. Acad. Sci. USA 103, 5379–5384. Rogers, M. S., Patel, R. P., Reeder, B. J., Sarti, P., Wilson, M. T., and Alayash, A. I. (1995). Pro-oxidant effects of cross-linked haemoglobins explored using liposome and cytochrome c oxidase vesicle model membranes. Biochem. J. 310(Pt 3), 827–833. Sarkela, T. M., Berthiaume, J., Elfering, S., Gybina, A. A., and Giulivi, C. (2001). The modulation of oxygen radical production by nitric oxide in mitochondria. J. Biol. Chem. 276, 6945–6949. Sarti, P., Giuffre, A., Forte, E., Mastronicola, D., Barone, M. C., and Brunori, M. (2000). Nitric oxide and cytochrome c oxidase: Mechanisms of inhibition and NO degradation. Biochem. Biophys. Res. Commun 274, 183–187. Sawicki, C. A., and Gibson, Q. H. (1977). Quaternary conformational changes in human oxyhemoglobin studied by laser photolysis. J. Biol. Chem. 252, 5783–5788. Schindler, F. J. (1967). Determination of oxygen affinity. In ‘‘Oxidation and Phosphorylation’’ ( R. Estabrook and M. Pullman, eds.), Vol. 10, pp. 629–634. Elsevier. Sharpe, M. A., and Cooper, C. E. (1997). Nitric oxide reacts with mitochondrial cytochrome c. Biochem. Soc. Trans. 25, 407S. Shiva, S., Brookes, P. S., Patel, R. P., Anderson, P. G., and Darley-Usmar, V. M. (2001). Nitric oxide partitioning into mitochondrial membranes and the control of respiration at cytochrome c oxidase. Proc. Natl. Acad. Sci. USA 98, 7212–7217.
Nitric Oxide Inhibition of Oxidases
159
Stevanin, T. M., Ioannidis, N., Mills, C. E., Kim, S. O., Hughes, M. N., and Poole, R. K. (2000). Flavohemoglobin Hmp affords inducible protection for Escherichia coli respiration, catalyzed by cytochromes bo0 or bd, from nitric oxide. J. Biol. Chem. 275, 35868–35875. Swartz, H. M., Dunn, J., Grinberg, O., O’Hara, J., and Walczak, T. (1997). What does EPR oximetry with solid particles measure—and how does this relate to other measures of PO2? Adv. Exp. Med. Biol. 428, 663–670. Torres, J., Cooper, C. E., and Wilson, M. T. (1998). A common mechanism for the interaction of nitric oxide with the oxidized binuclear centre and oxygen intermediates of cytochrome c oxidase. J. Biol. Chem. 273, 8756–8766. Torres, J., Darley-Usmar, V., and Wilson, M. T. (1995). Inhibition of cytochrome c oxidase in turnover by nitric oxide: Mechanism and implications for control of respiration. Biochem. J. 312(Pt 1), 169–173. Wang, C. Y., Liu, C. A., Piantadosi, C. A., and Allen, B.w. (2005). A novel electrochemical nitric oxide sensor: Aligned RuO2 nanowires deposited on Pt filament. ‘‘Nanotech 2005,’’ Vol. 1, pp. 434–437. Taylor & Francis, CA. Wever, R., Boelens, R., De Boer, E., Van Gelder, B. F., Gorren, A. C., and Rademaker, H. (1985). The photoreactivity of the copper-NO complexes in cytochrome c oxidase and in other copper-containing proteins. J. Inorg. Biochem. 23, 227–232. Wikstrom, M., and Morgan, J. E. (1992). The dioxygen cycle: Spectral, kinetic, and thermodynamic characteristics of ferryl and peroxy intermediates observed by reversal of the cytochrome oxidase reaction. J. Biol. Chem. 267, 10266–10273. Wilson, D. F., Rumsey, W. L., Green, T. J., and Vanderkooi, J. M. (1988). The oxygen dependence of mitochondrial oxidative phosphorylation measured by a new optical method for measuring oxygen concentration. J. Biol. Chem. 263, 2712–2718. Wittenberg, J. B. (1974). Facilitated oxygen diffusion: The role of leghemoglobin in nitrogen fixation by bacteroids isolated from soybean root nodules. J. Biol. Chem. 249, 4057–4066. Yonetani, T. (1960). Studies on cytochrome oxidase. I. Absolute and difference absorption spectra. J. Biol. Chem. 235, 845–852. Yonetani, T., Tsuneshige, A., Zhou, Y., and Chen, X. (1998). Electron paramagnetic resonance and oxygen binding studies of alpha-nitrosyl hemoglobin: A novel oxygen carrier having no-assisted allosteric functions. J. Biol. Chem. 273, 20323–20333.
This page intentionally left blank
S E C T I O N
T W O
SENSOR PROTEINS
This page intentionally left blank
C H A P T E R
N I N E
Cloning, Expression, and Purification of the N-terminal Heme-Binding Domain of Globin-Coupled Sensors Jennifer A. Saito,* Tracey Allen K. Freitas,* and Maqsudul Alam*,† Contents 164 164 165 166 166 167 168 170 170 171 171
1. Introduction 2. Bioinformatic Search of Globin-Coupled Sensors 2.1. N-terminal sensor globin 2.2. C-terminal transmitter 3. Functional Analysis of Globin-Coupled Sensors 3.1. Cloning and expression 3.2. Protein purification 3.3. SDS-PAGE and spectroscopy 3.4. Minimum heme-binding domain Acknowledgments References
Abstract Globin-coupled sensors (GCSs) are multidomain proteins, consisting of an N-terminal globin domain fused to a variety of C-terminal transmitter domains. Functional classification of GCSs is based on the transmitter domain(s) they possess, broadly falling under either aerotaxis or gene regulation. This chapter describes methods and strategies for cloning, expression, and purification of GCSs for spectroscopic analysis and determination of the minimum heme-binding domain, together with bioinformatic approaches for database searching and examination of domain architectures.
* {
Department of Microbiology, University of Hawaii, Honolulu, Hawaii Advanced Studies in Genomics, Proteomics, and Bioinformatics, College of Natural Sciences, University of Hawaii, Honolulu, Hawaii
Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37009-2
#
2008 Elsevier Inc. All rights reserved.
163
164
Jennifer A. Saito et al.
1. Introduction Globin-coupled sensors (GCSs) are a family of sensor transducers that have been found only in Archaea and Bacteria (Freitas et al., 2005; Hou et al., 2001; Vinogradov et al., 2006) and, in particular, Proteobacteria. This feature may, however, simply be an emulation of the taxonomic bias of the genomes chosen to be sequenced. GCSs contain a globin domain—usually at the amino (N) terminus—that binds (‘‘senses’’) compatible ligands and at least one transmitter domain—usually at the carboxy (C) terminus—that relays (‘‘transduces’’) this ligand-bound state in a form that is compatible with the host cell. GCSs are currently classified as either heme-based aerotaxis transducers (HemATs) (Hou et al., 2000) or gene regulators (Freitas et al., 2003, 2005), but this classification may expand in the future. Presently, only molecular oxygen has been shown to act as a physiological ligand for these proteins. Sensor and transducer domains tend to reside in the same contiguous polypeptide in the majority of GCSs, although a few proteins have been identified that consist only of an isolated globin domain. These globins possess significant similarity to the sensor globin domain of the GCSs and are therefore referred to simply as sensor globins. Although partner proteins equivalent to the C-terminal transducer domains in complete GCSs may reside at some other position within the genome, these types of proteins are not discussed further. Flavohemoglobins (FHbs) are the only other class of multidomain globins known; however, they are excluded from the family of GCSs based on the sensor-transducer prerequisite. This chapter discusses how to detect GCSs within the current protein sequence databases and, once identified, how to clone, express, and purify one of them for in vitro analysis.
2. Bioinformatic Search of Globin-Coupled Sensors Identification of a GCS starts with identification of the globin domain using certain criteria (discussed later), usually using BLAST (http://www.ncbi. nlm.nih.gov/BLAST) or hmmpfam of the HMMER package (http:// hmmer.janelia.org/). Once the globin domain has been identified, the entire protein is examined by a protein domain database such as SMART (http:// smart.embl-heidelberg.de; Letunic et al., 2006), Pfam (http://www.sanger.ac. uk/Software/Pfam; Finn et al., 2006), or CDD (http://www.ncbi.nlm.nih. gov/Structure/cdd/cdd.shtml; Marchler-Bauer et al., 2007) to uncover the domain topologies present. Currently, only CDD from NCBI recognizes
N-terminal Heme-Binding Domain of GCSs
165
the GCS globin domain as a unique domain (sensor_globin, cd01068). After examining the domain topologies, GCSs are organized according to the C-terminal domains present, the number and order of which vary considerably. The GCS from the b-Proteobacterium Azoarcus sp. EbN1 contains, in addition to the sensor globin domain, four additional recognizable domains (PAS-PAS-GGDEF-EAL) that is exceeded only by the g-Proteobacterium Reinekea sp. MED297 with six additional domains.
2.1. N-terminal sensor globin As stated previously, the sensor globin domain has been found at the amino (N) terminus, whereas all other domains tend to occur at the C terminus. When searching for GCSs, the sensor globin domain should be used, unless GCSs with homologous C-terminal transmitter domains are desired. It may also be beneficial to limit the globin domain template to a small motif of approximately 64 amino acids (Fig. 9.1). The globin domain is identified by a motif of primarily two regions: (1) the region around the B10 tyrosine and (2) the region around the proximal histidine. The B9 phenylalanine tends to stabilize the heme in the pocket by overlapping its p orbitals with those of the aromatic heme. The B10 tyrosine is likely conserved because of crucial interactions with the bound ligand in the distal pocket (Couture et al., 1999; Kloek et al., 1994; Ouellet et al., 2003). The proximal histidine provides the only known covalent bond to the heme prosthetic group, and the absolute conservation of this histidine is essential for our classification as a globin. Proteins with significant similarity to the sensor globin domain have been identified that substitute polar residues such as glutamate for the proximal histidine (unpublished results). The heme-binding capacity of such proteins has not been examined yet, and therefore these proteins are not included in the family of GCSs. If the goal is to uncover many distantly related GCSs, the PSI-BLAST program (http://www.ncbi.nlm.nih.gov/BLAST) is best. Searching in
|--#1--| |---#2---| FYRIVRIDPHAEEFLSNEQVERQLKSAMERWIINVLSAQVDDVERLIQIQHTVAEVHARIGIPV Region #1: The phenylalanine-tyrosine pair is common throughout the GCSs, although occasionally a single variant is found among the two. Region #2: The proximal histidine is considered an absolute requirement for heme binding to globins.
Figure 9.1 Consensus sequence for approximately 20 GCSs, emphasizing the two relevant regions for globin identification and their positional reference to each other. Shaded regions are absolutely conserved in the source alignment.
166
Jennifer A. Saito et al.
this fashion will undoubtedly retrieve FHb hits and, when these proteins are consistently retrieved over new GCSs, it is safe to halt additional search cycles. The PSI-BLAST program is, however, limited to the completely sequenced genomes for which a protein table has been generated. A standard BLAST search (blastp or tblastn) is available on Microbial Genome BLAST page at NCBI (http://www.ncbi.nlm.nih.gov/sutils/genom_table.cgi). An alternative is to visit individual sequencing institutions for BLASTing services such as JGI (http://img.jgi.doe.gov/cgi-bin/pub/main.cgi), TIGR (http://tigrblast.tigr.org/ufmg), and the Sanger Institute (http://www. sanger.ac.uk/DataSearch/blast.shtml). A final option is to subscribe to the data sets from each institution, download the updates as they become available, and create and format your own databases for either local BLAST (ftp://ftp.ncbi.nih.gov/blast) or local hmmpfam searches. When screening many potential GCSs, it is helpful to first identify the proximal histidine in region #2 and then scrutinize region #1.
2.2. C-terminal transmitter Once potential sensor globin domains have been identified, FHbs and solo sensor globin domains should be removed. No concrete guidelines for GCS classification have been instituted other than those based on their general functions of aerotaxis and gene regulation (Freitas et al. 2003, 2005). There is one pitfall, however, when assigning such functions that is not limited to only GCSs. Although the C-terminal transmitter regions share rough similarities to other stretches of sequence commonly found in domain X or domain Y, they may be missing the key residues found in the active sites of such domains. Caution should therefore be employed before assigning a function to the GCS based solely on the gross similarity of the amino acid sequence with a known domain.
3. Functional Analysis of Globin-Coupled Sensors Once putative GCSs are found, the goal is to characterize their functions. This involves examination of the globin and transmitter domains, how they interact with each other, and how the protein as a whole affects the cell. This section describes experimental methods for characterizing the GCSs, focusing mainly on the globin sensor domain. We use HemAT-Hs (Hou et al., 2000) as an example, but these methods can easily be adapted for other GCSs.
167
N-terminal Heme-Binding Domain of GCSs
3.1. Cloning and expression A two-step polymerase chain reaction (PCR) method is used to add the sequence encoding a C-terminal 6xHis tag, as well as add NdeI and BamHI restriction sites to facilitate cloning into the expression vector (Fig. 9.2A). The following primers are used for the PCR reactions: 1. Hs_NdeI top: 50 -CGCATATGAGCAACGATAATGAC-30 2. Hs_6His bot: 50 -GTGGTGGTGGTGGTGGTGGCTGAGCTTGCCGACCGTC-30 3. BamHI bot: 50 -GCGGATCCTTAGTGGTGGTGGTGGTGGTG-30 The hemAT-Hs gene is first amplified with PfuTurbo DNA polymerase (Stratagene), using the Hs_NdeI top and Hs_6His bot primers. The PCR product is used as template for a second PCR reaction with the Hs_NdeI top
A
B Genomic DNA
20 mL overnight culture
PCR using primers 1 & 2 NdeI
6His PCR using primers 1 & 3
NdeI
6His-stop-BamHI
1 L fresh medium
Shake at 37º until OD600 = 0.5 – 0.6
Add IPTG, FeSO4, & ALA
TOPO cloning Shake at 37º for ~3h Harvest cells by centrifugation Subclone into pET-3a
Store at −20º until needed
Sequence insert
Figure 9.2 Construction and expression of a C-terminal 6xHis-tagged GCS. (A) Cloning of HemAT-Hs. Primers 1, 2, and 3 refer to those listed in the text. (B) Expression of HemAT-Hs in E. coli. IPTG, isopropyl-b-D-thiogalactopyranoside; ALA, d-aminolevulinic acid hydrochloride.
168
Jennifer A. Saito et al.
and BamHI bot primers. The final PCR product is cloned into the pCR4Blunt-TOPO vector (Invitrogen) and is then subcloned into the pET-3a expression vector (Novagen) using restriction enzymes NdeI and BamHI. Positive clones are confirmed by sequencing the insert. The hemAT-Hs/pET-3a plasmid is transformed into BL21(DE3)pLysS Escherichia coli cells (Novagen). A 20-ml starter culture is grown overnight in LB medium containing 100 mg/ml ampicillin and 34 mg/ml chloramphenicol and is used to inoculate 1 liter of fresh medium (Fig. 9.2B). The culture is grown to OD600 ¼ 0.5–0.6 at 37 with shaking (250 rpm). Protein expression is induced with 0.6 mM isopropyl-b-D-thiogalactopyranoside (IPTG). At this point, we also add 100 mg/liter FeSO47H2O and 17 mg/liter d-aminolevulinic acid hydrochloride to enhance heme biosynthesis. The culture is further incubated at 37 for 2–3 hours (shaking at 250 rpm). Cells are harvested by centrifugation and stored at –20 prior to purification. Depending on which GCS is being expressed, solubility of the protein may become an issue with the aforementioned culturing conditions. In those cases, we found that the solubility improves by using E. coli strain Rosetta 2(DE3)pLysS (Novagen). Also, reducing the IPTG concentration to 20 mM and incubating at 30 for 8 h (shaking at 150 rpm) or at room temperature overnight will help increase the yield of soluble protein.
3.2. Protein purification This method of fast-flow perfusion chromatography (Fig. 9.3) was initially developed for the purification of HemAT-Hs (Piatibratov et al., 2000), but it has worked quite well for several other GCSs, without any modification of the protocol. A cell pellet from 2 liters of culture is resuspended in 20 ml of buffer I (200 mM NaCl, 50 mM Na2HPO4, pH 8.0) containing 1 mM phenylmethylsulfonyl fluoride and is sonicated for 8 min (24 pulses of 20 s with 20-s pauses, Sonic Dismembrator 550, Fisher Scientific). The cell lysate is centrifuged to remove insoluble cell debris, and the supernatant containing HemAT-Hs is filtered (0.22 mm pore size, Corning) and kept on ice until it is loaded onto the purification column. The BioCAD SPRINT perfusion system (Applied Biosystems, now discontinued) is used for metal-affinity chromatography, but any affinity column chromatography system can be used. Other groups have used similar methods to purify HemAT-Bs as a His-tagged protein (Zhang and Phillips, 2003; Pinakoulaki et al., 2006). The column (POROS MC/20, 10 100 mm, column volume 7.9 ml) of immobilized Co2þ resin is prepared by first washing with five column volumes (CV) of 50 mM EDTA, 1 M NaCl (pH 8.0) and rinsing with an equal volume of distilled water. The metal is loaded with 5 CV of 100 mM
169
N-terminal Heme-Binding Domain of GCSs
Sample preparation
Column preparation 1. 5 CV EDTA 2. 5 CV distilled H2O
1. Resuspend cell pellet
3. 5 CV CoCl2
2. Lyse cells (sonication)
4. 5 CV distilled H2O
3. Centrifuge
5. 5 CV NaCl
4. Filter supernatant
6. 5 CV distilled H2O 7. 5 CV 0.5 mM imidazole
Load and wash 1. Protein sample 2. 6−8 CV Na2HPO4 buffer
Elution 1. 8 CV 0.5 mM imidazole 2. 0.5 mM –250 mM imidazole gradient 3. Collect eluate
Figure 9.3 Schematic flowchart for protein purification.
CoCl26H2O, washed with 5 CV of water, and followed by 5 CV of 3 M NaCl to remove residual metal ions. A flow rate of 15 ml/min is used for all the aforementioned washes. Prior to sample loading, the column is washed with 5 CV of buffer I containing 0.5 mM imidazole (flow rate 8.0 ml/min). The sample is loaded onto the column at a flow rate of 0.5–1.0 ml/min, followed by 6–8 CV of buffer I at 2.0 ml/min. The column is then washed with 8 CV of buffer I containing 0.5 mM imidazole (5.0 ml/min) to remove nonspecifically bound proteins. The remaining bound proteins are eluted by a linear gradient of imidazole from 0.5 to 250 mM in buffer I (5.0 ml/min flow rate, 7 CV total) and collected in 1.5-ml fractions. Fractions containing HemAT-Hs are combined and concentrated using an Amicon Ultra-15 centrifugal filter (30 k MWCO, Millipore). The protein concentration is determined using the Coomassie Plus protein assay reagent kit (Pierce, IL) according to the manufacturer’s protocol.
170
Jennifer A. Saito et al.
3.3. SDS-PAGE and spectroscopy Protein purification is confirmed by running SDS-PAGE (Fig. 9.4A). Absorption spectra are measured in buffer I using a Cary 1E UV-visible spectrophotometer (Varian). Figure 9.4B shows a typical absorbance spectrum for HemAT-Hs.
3.4. Minimum heme-binding domain Depending on the application (e.g., ligand binding, crystallization), it may be more useful to express and characterize only the globin domain of the GCSs. This can be done with any of the GCSs, but in cases where the full-length protein is membrane bound, difficult to express, or unstable in solution, this approach would be particularly beneficial. The ligand-binding properties of the isolated globin domain will likely differ only slightly, if at all, from that of the full-length protein, so the absence of the transmitter domain(s) should not pose any problems. Zhang et al. (2005) showed that the ligand-binding properties of the HemAT-Bs globin domain are independent of the C-terminal domain. Identification of the minimum heme-binding domain is important for determining the functional requirements of binding heme and the ligand, as well as pinpointing the position of any linker regions that may exist. The globin domain length based on CDD can serve as a good starting point, but it may not be the actual minimum heme-binding domain. For example, A kDa
B M
1
0.4
200
116 97.4
66
Absorbance
0.3
0.2
0.1 45 0.0 300
350
400 450 500 Wavelength (nm)
550
600
650
Figure 9.4 Analysis of purified HemAT-Hs. (A) SDS-PAGE. M, molecular weight marker;1, HemAT-Hs. (B) Absorption spectrum.
N-terminal Heme-Binding Domain of GCSs
171
whereas the experimentally determined minimum heme-binding domain of HemAT-Hs extends to residue 195 (Hou et al., 2001), the globin domain detected by CDD only goes up to residue 187. Therefore, various lengths of the globin domain should be examined and compared. Because each GCS is different, the minimum heme-binding domain will have to be determined on a case-by-case basis. Truncated versions of the GCSs are cloned, expressed, and purified as described previously in this chapter. Variants that are insoluble (i.e., form inclusion bodies) are disregarded from further analysis. Protein concentrations of the remaining polypeptides are normalized, and absorption spectra are measured. Examination of Soret, a, and b absorption peaks is used as an indication of heme-binding efficiency. Fragments shorter than the minimum heme-binding domain should exhibit a significant decrease in absorption.
ACKNOWLEDGMENTS This work was supported by National Science Foundation Grant MCB0446431 and U.S. Army Award TATRC #W81XWH-05–2-0013.
REFERENCES Couture, M., Yeh, S. R., Wittenberg, B. A., Wittenberg, J. B., Ouellet, Y., Rousseau, D. L., and Guertin, M. (1999). A cooperative oxygen-binding hemoglobin from Mycobacterium tuberculosis. Proc. Natl. Acad. Sci. USA 96, 11223–11228. Finn, R. D., Mistry, J., Schuster-Bockler, B., Griffiths-Jones, S., Hollich, V., Lassmann, T., Moxon, S., Marshall, M., Khanna, A., Durbin, R., Eddy, S. R., Sonnhammer, E. L., and Bateman, A. (2006). Pfam: Clans, web tools and services. Nucleic Acids Res. 34, D247–D251. Freitas, T. A. K., Hou, S., and Alam, M. (2003). The diversity of globin-coupled sensors. FEBS Lett. 552, 99–104. Freitas, T. A. K., Saito, J. A., Hou, S., and Alam, M. (2005). Globin-coupled sensors, protoglobins, and the last universal common ancestor. J. Inorg. Biochem. 99, 23–33. Hou, S., Freitas, T., Larsen, R. W., Piatibratov, M., Sivozhelezov, V., Yamamoto, A., Meleshkevitch, E. A., Zimmer, M., Ordal, G. W., and Alam, M. (2001). Globin-coupled sensors: A class of heme-containing sensors in Archaea and Bacteria. Proc. Natl. Acad. Sci. USA 98, 9353–9358. Hou, S., Larsen, R. W., Boudko, D., Riley, C. W., Karatan, E., Zimmer, M., Ordal, G. W., and Alam, M. (2000). Myoglobin-like aerotaxis transducers in Archaea and Bacteria. Nature 403, 540–544. Kloek, A. P., Yang, J., Mathews, F. S., Frieden, C., and Goldberg, D. E. (1994). The tyrosine B10 hydroxyl is crucial for oxygen avidity of Ascaris hemoglobin. J. Biol. Chem. 269, 2377–2379. Letunic, I., Copley, R. R., Pils, B., Pinkert, S., Schultz, J., and Bork, P. (2006). SMART 5: Domains in the context of genomes and networks. Nucleic Acids Res. 34, D257–D260. Marchler-Bauer, A., Anderson, J. B., Derbyshire, M. K., DeWeese-Scott, C., Gonzales, N. R., Gwadz, M., Hao, L., He, S., Hurwitz, D. I., Jackson, J. D., Ke, Z.,
172
Jennifer A. Saito et al.
Krylov, D., et al. (2007). CDD: A conserved domain database for interactive domain family analysis. Nucleic Acids Res. 35, D237–D240. Ouellet, H., Juszczak, L., Dantsker, D., Samuni, U., Ouellet, Y. H., Savard, P. Y., Wittenberg, J. B., Wittenberg, B. A., Friedman, J. M., and Guertin, M. (2003). Reactions of Mycobacterium tuberculosis truncated hemoglobin O with ligands reveal a novel ligand-inclusive hydrogen bond network. Biochemistry 42, 5764–5774. Piatibratov, M., Hou, S., Brooun, A., Yang, J., Chen, H., and Alam, M. (2000). Expression and fast-flow purification of a polyhistidine-tagged myoglobin-like aerotaxis transducer. Biochim. Biophys. Acta 1524, 149–154. Pinakoulaki, E., Yoshimura, H., Daskalakis, V., Yoshioka, S., Aono, S., and Varotsis, C. (2006). Two ligand-binding sites in the O2-sensing signal transducer HemAT: Implications for ligand recognition/discrimination and signaling. Proc. Natl. Acad. Sci USA 103, 14796–14801. Vinogradov, S. N., Hoogewijs, D., Bailly, X., Arredondo-Peter, R., Gough, J., Dewilde, S., Moens, L., and Vanfleteren, J. R. (2006). A phylogenomic profile of globins. BMC Evol. Biol. 6, 31. Zhang, W., Olson, J. S., and Phillips, G. N., Jr. (2005). Biophysical and kinetic characterization of HemAT, an aerotaxis receptor from Bacillus subtilis. Biophys. J. 88, 2801–2814. Zhang, W., and Phillips, G. N., Jr. (2003). Crystallization and X-ray diffraction analysis of the sensor domain of the HemAT aerotactic receptor. Acta Crystallogr. D Biol Crystallogr. 59, 749–751.
C H A P T E R
T E N
Oxygen-Sensing Histidine-Protein Kinases: Assays of Ligand Binding and Turnover of Response-Regulator Substrates Marie-Alda Gilles-Gonzalez, Gonzalo Gonzalez, Eduardo Henrique Silva Sousa, and Jason Tuckerman Contents 174 175 175 177 181 185 186 187 187
1. Introduction 2. Assays 2.1. General considerations 2.2. Protein purifications 2.3. Measuring the Kd for binding of ligand 2.4. Determination of turnover rates, kcat 2.5. Novel heme-containing histidine-protein kinases Acknowledgments References
Abstract Heme-based sensors are a recently discovered functional class of heme proteins that serve to detect physiological fluctuations in oxygen (O2), carbon monoxide (CO), or nitric oxide (NO). Many of these modular sensors detect heme ligands by coupling a histidine-protein kinase to a heme-binding domain. They typically bind O2, CO, and NO but respond only to one of these ligands. Usually, they are active in the ferrous unliganded state but are switched off by saturation with O2. The heme-binding domains of these kinases are quite varied. They may feature a PAS fold, as in the Bradyrhizobium japonicum and Sinorhizobium melitoti FixL proteins, or a GAF fold, as in the Mycobacterium tuberculosis DevS and DosT proteins. Alternative folds, such as HNOB (also H-NOX), have also been noted for such signal-transducing kinases, although these classes are less well studied. Histidine-protein kinases function in partnership with cognate response-regulator substrate(s): usually transcription factors that they activate Department of Biochemistry, University of Texas Southwestern Medical Center, Dallas, Texas Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37010-9
#
2008 Elsevier Inc. All rights reserved.
173
174
Marie-Alda Gilles-Gonzalez et al.
by phosphorylation. For example, FixL proteins specifically phosphorylate their FixJ partners, and DevS and DosT proteins phosphorylate DevR in response to hypoxia. We present methods for purifying these sensors and their protein substrates, verifying the quality of the preparations, determining the Kd values for binding of ligand and preparing sensors of known saturation, and measuring the rates of turnover (kcat) of the protein substrate by sensors of known heme status.
1. Introduction The FixL-FixJ two-component regulatory system is a paradigm for O2 signal transduction (Gilles-Gonzalez, 2001; Gilles-Gonzalez and Gonzalez, 2005). This simple and classical system consists of only the sensory kinase, FixL, and its substrate and partner, FixJ, each of which is modular (David et al., 1988; Gilles-Gonzalez et al., 1991, 1994; Monson et al., 1992). In FixL, a heme-binding domain couples to a histidine-protein kinase such that the unliganded (deoxy, FeII) form is the ‘‘on-state’’ active kinase, and the oxygen-bound (Kd 50–140 mM) form is the ‘‘off state’’ (Gilles-Gonzalez and Gonzalez, 1993; Lois et al., 1993; Mukai et al., 2000). The FixL-catalyzed phosphorylation of a receiver domain in FixJ promotes this response regulator to a dimeric form that activates transcription of target genes (Agron et al., 1993; Da Re et al., 1999; Galinier et al., 1994; Reyrat et al., 1993). The best-studied FixL–FixJ systems are those that control nitrogen fixation in Bradyrhizobium japonicum and Sinorhizobium meliloti (formerly Rhizobium meliloti) (Dixon and Kahn, 2004; Fischer, 1994; Gilles-Gonzalez and Gonzalez, 2005). Although all ligands of ferrous heme bind to BjFixL and RmFixL, O2 abolishes their kinase activity (>200-fold inhibition), but carbon monoxide (CO) and nitric oxide (NO) do not (less than threefold inhibition) (Dunham et al., 2003; Gilles-Gonzalez et al., 1994; Sousa et al., 2007a). The unliganded form of ferric BjFixL is fully active, and cyanide inhibits this species analogously to the O2 inhibition of the ferrous form (Gilles-Gonzalez et al., 2006). The electronic resemblance between the cyanomet species (FeIIICN-) and the substantively polarized O2-bound state (FeIIO2FeIIIO2-), together with crystal structures of liganded forms of the heme-binding domain, suggest that regulatory switching of the kinase is triggered by changed polar interactions of the heme with residues of the heme pocket (Gong et al., 1998; Hao et al., 2002; Olson and Phillips, 1997). As a rule, microbial O2 sensors serve to initiate substantive lifestyle changes. For example, hypoxia is a key determinant of the rhizobial switch from a vegetative to a nonreplicative symbiotic state (David et al., 1988; Ditta et al., 1987; Sciotti et al., 2003; Soupene et al., 1995; Virts et al., 1988). In low O2, rhizobial FixL–FixJ systems induce a cascade of gene (nif, fix) expression that produces the nitrogen-fixation enzymes and their accessory proteins,
Heme-Ligand Sensing Histidine-Protein Kinases
175
the key regulators of denitrification, and one or more high-affinity alternative terminal oxidases for respiration in low O2. The latter function, which enables respiration during hypoxia, is also reported for homologs of FixL and FixJ in microorganisms that do not fix nitrogen (Crosson et al., 2005). In addition to FixLs, several classes of histidine-protein kinases employ alternative heme-binding domains to sense gaseous ligands (GillesGonzalez and Gonzalez, 2005). For example, the Mycobacterium tuberculosis DevS and DosT hypoxia sensors couple their kinase to a heme-binding GAF domain instead of the heme-binding PAS domain found in FixLs (Ioanoviciu et al., 2007; Sardiwal et al., 2005; Sousa et al., 2007b). Like FixL, these M. tuberculosis sensors phosphorylate a response-regulating transcription factor (DevR), and the resulting broad changes in gene expression trigger a state of nonreplicative persistence: in this case a latent infection of a human host (Roberts et al., 2004; Saini et al., 2004a,b). Clearly, the readily accessible ligand-binding and enzymatic parameters of heme-based O2 sensors make them ideal subjects for studies of signal transduction, and their involvement in relevant lifestyle changes of bacteria lends additional significance to their study.
2. Assays 2.1. General considerations 2.1.1. Autophosphorylation If a histidine-protein kinase is supplied with its ATP but not its protein substrate, it slowly (10–40 min) converts itself to a phosphorylated species (Hess et al., 1991; Stock et al., 1989). For example, the addition of ATP to deoxy-FixL yields a phospho-FixL species (Gilles-Gonzalez and Gonzalez, 1993; Tuckerman et al., 2001). Autophosphorylation FixL2 þ 2 ATP ⇆ P-FixL2 þ 2 ADP The so-called autophosphorylation is really a trans-phosphorylation between the subunits of these requisitely dimeric kinases (Ninfa et al., 1993). For sensors such as FixL, this reaction will reliably report qualitative information on
Ligands that regulate the kinase Preferred divalent cation for the enzyme Contamination of the enzyme preparation with a phosphatase
176
Marie-Alda Gilles-Gonzalez et al.
However, this reaction cannot give a quantitative measure of the enzymatic activity or its regulation because it is not a valid, rate-limiting, half-reaction. Contrary to the ping-pong bi–bi mechanism often presumed for many histidine-protein kinases, FixL does not process its nucleotide and protein substrates independently (Tuckerman et al., 2001, 2002). FixJ clearly enhances the initial FixL phosphorylation. For example, phosphorylation of RmFixL is accelerated about eightfold by the inclusion of an unphosphorylatable FixJ (the D54N RmFixJ variant) in the reaction (Sousa et al., 2005). More importantly, even in a large excess of ATP the reaction of FixL with only ATP stops when only about 20% of FixL is phosphorylated. In contrast, formation of the phosphorylated FixL intermediate proceeds essentially to completion if the true intermediate is trapped by including the unphosphorylatable FixJ substrate in the reaction (Sousa et al., 2005). Finally, when FixJ is added to phospho-FixL ‘‘preformed’’ by autophosphorylation, very little phosphoryl transfer occurs; instead, most of the phosphorylated protein is hydrolyzed to free phosphate. These observations imply that the ‘‘phospho-FixL’’ produced by these two methods are not kinetically equivalent (Tuckerman et al., 2001). This may also be the case for many other sensory kinases presumed to carry out their phosphoryl transfers sequentially and reported to show highly inefficient phosphoryl transfer to their protein substrate under those conditions. 2.1.2. Turnover Since FixL-catalyzed phosphoryl transfers clearly take place with FixL, FixJ, and ATP present, the turnover rate kcat provides the best measure of the FixL reaction kinetics and the effects of regulatory ligands on those kinetics (Sousa et al., 2007a; Tuckerman et al., 2002).
Turnover: 2 FixJ þ 2 ATP
FixL2 ⇆ P-FixJ2 þ 2 ADP
This rate represents the number of molecules of a specific substrate that one molecule of a specified form of an enzyme will phosphorylate per minute while it is saturated with all of its substrates. Consequently, an accurate kcat measurement requires that the kinase be kept at levels sufficiently high to ensure preservation of its dimeric state and yet sufficiently low to guarantee that the level of enzyme–substrate complex will not change during the measurement, i.e., the time for the enzyme to reach steady state and complete at least 10 turnovers. For the FixL–FixJ system, this means about 1 mM FixL, >20 mM FixJ, and 1.0 mM ATP. For kcat determinations, it is also essential to collect complete time courses so that the portion the time course may be found where product accumulates at a
Heme-Ligand Sensing Histidine-Protein Kinases
177
constant rate (i.e., when the rate of phosphorylation of the protein substrate matches the rate of replenishment of the phosphorylated enzyme intermediate). For inhibited states of the enzyme (e.g., FixL partially saturated with O2), there can be a lag while the enzyme intermediates build up. For highly active states (e.g., deoxy-FixL), there can be an early deceleration of the reaction rate due to rapid depletion of substrate. An informative indicator of the efficacy of a heme ligand is its inhibition factor, defined as the ratio of turnover rates of the unliganded and the fully liganded forms for the same oxidation state. For lowaffinity native and mutant sensors, saturation with O2 may not be practical. In these cases, great caution should be used in estimating the activity of a hypothetical fully oxygenated sensor, as the activity of partially saturated mixtures is not necessarily the sum of the activities of the liganded and unliganded species weighted by their relative abundance (Sousa et al., 2007a). In addition to the aforementioned considerations about steady state, determinations of inhibition factor require verifying that the heme status (saturation, oxidation state) remains unchanged throughout the measurements of reaction kinetics.
2.2. Protein purifications 2.2.1. Strategy for purifying heme-containing histidine-protein kinases Sensory kinases such as FixL and DosT are easily monitored from their intense red color. They readily yield to traditional methods of protein fractionation, with an anion-exchange step giving the most significant purification because of their unusually low isoelectric points (pI 5–6) (Scopes, 1994). Moderate overexpression of the corresponding genes at about 5% of total cell proteins from an inducible Escherichia coli promoter (e.g., tac) usually gives the highest yield of soluble heme protein. While it is possible to obtain much higher expression with some vectors, the resulting protein is usually of low quality, with weak enzymatic activity and a tendency to aggregate. It is far easier to grow more cells than to try to recover activity from misfolded or aggregated protein. Typically, a 4-liter culture of an E. coli lacIq strain (e.g., TG1) harboring the gene on a plasmid is grown overnight in a fermentor (Bioflow 3000) at 37 , 200–500 rpm, and 20% of atmospheric O2. When the culture reaches an OD600 of about 0.5, expression of the heme protein is induced with isopropyl-b-D-thiogalactoside (1 mM ). When growth of the cells begins to slow (cell density about 30 g/liter), they are cooled to 4 , harvested, and lysed by sonication in 2 volumes of lysis buffer [20 mM Tris-HCl, pH 8.0, 100 mM NaCl, 3 mM KCl, 1 mM EDTA, 10 mM b-mercaptoethanol, 0.04 mg/ml lysozyme, 0.17 mg/ml phenylmethylsulfonyl fluoride (added at room temperature from a 40-mg/ml solution in acetone)]. The lysate is cleared by centrifugation at 70,000 rpm (Ti 70 rotor, 30 min, 4 ).
178
Marie-Alda Gilles-Gonzalez et al.
The clear-red lysate, kept at about 4 , is slowly brought to 1.2 M ammonium sulfate with stirring and dropwise addition of 4.0 M ammonium sulfate, 2 mM EDTA, pH 7.5, and centrifuged at 12,000 rpm (SS43 rotor, 4 ). Assuming the pellet density to be about 1 g/ml, the red precipitate is diluted threefold with 50 mM Tris-HCl, 50 mM NaCl, 5% (v/v) glycerol, and 10 mM b-mercaptoethanol, pH 7.5, and desalted on a size-exclusion column (Sephadex G-25, GE Healthcare) preequilibrated with 50 mM Tris-HCl, 50 mM NaCl, 5% (v/v) glycerol, and 10 mM b-mercaptoethanol, pH 7.5, at 4 . Subsequent tracking of the heme protein is done automatically from its 415-nm absorption (Bio-Rad QuadTec UV/vis detector). The protein mixture is chromatographed on an anion-exchange column (DEAE-Sephacel, Amersham), with thorough washing in 100 mM NaCl, and elution from a gradient of 100–300 mM NaCl in 50 mM Tris-HCl, 5% (v/v) glycerol, and 10 mM b-mercaptoethanol, pH 7.5, at 4 . The heme protein-containing fractions are further purified (to about 95% purity) by gel filtration (Superdex S-200, GE Healthcare) on a column preequilibrated with 50 mM Tris-HCl, 50 mM NaCl, and 5% (v/v) glycerol, pH 8.0, at 4 . Depending on the stability to oxidation of the purified protein, it will be a mixture of FeIII and FeIIO2 states, or entirely in the FeIII state (Gonzalez et al., 1998). Concentrate the protein to about 100 mM in a filtration unit (Amicon, 10-kDa membrane cutoff ) and store in aliquots of about 0.5 ml each at –70 . 2.2.2. Quality control Heme content The heme content of the purified proteins may be quantified by a pyridine hemochromogen assay, with hemin as the standard (Appleby, 1980). When bacterial heme-containing histidine-protein kinases are overproduced in E. coli, they are typically recovered with their full complement of heme (one heme per monomer). For comparison to the heme content, the protein concentration may be measured by the BCA protein assay (Pierce Biotechnology Inc.) with bovine serum albumin as the standard. Preparation of deoxy protein Incubate the purified protein (100 mM) for about 15 min with an anaerobic solution of 10 mM dithiothreitol in 50 mM Tris-HCl, 50 mM KCl, 5.0% (v/v) ethylene glycol, and 1 mM MgCl2, pH 8.0, inside of an anaerobic chamber (Coy Laboratory Products, Inc.). This procedure works well for heme-based sensors of relatively low redox potential (e.g., FixL, EcDos, DosT) and converts them fully to the deoxy state. If this approach does not yield the deoxy state within 15 min, reduce the protein with 1 equivalent of the stronger reducing agent sodium dithionite inside the anaerobic chamber and immediately remove this chemical and its by-products by gel filtration on a Sephadex-G25 column (about 3 ml) equilibrated with the same buffer as described earlier. If the protein is to be used in the deoxy state or mixed with CO, keep it in the aforementioned buffer. If the protein is to be mixed with O2, dilute
Heme-Ligand Sensing Histidine-Protein Kinases
179
the solution to less than 1 mM dithiothreitol; if it is to be mixed with NO, remove all of the reducing agent. Verify the quality of the preparation by recording the 350- to 700-nm absorption spectrum. Exploiting autophosphorylation to verify phosphatase contamination Transfer the deoxy-protein solution to a sealable cuvette and measure its 350- to 700-nm absorption. Assuming the 434-nm extinction (e434) of the deoxy state to be about 130 mM-1cm-1, adjust the protein concentration to 4–5 mM in 50 mM Tris-HCl, 50 mM KCl, 5.0% (v/v) ethylene glycol, and 1 mM MgCl2, pH 8.0, and equilibrate it at 23 . Separately prepare a series of Eppendorf tubes labeled for each time point and each containing 3.3 ml of ‘‘stop buffer’’ [10 mM EDTA, 2% (w/v) sodium dodecyl sulfate, 0.40 M Tris-HCl, 50% (v/v) glycerol, and 2% (v/v) b-mercaptoethanol, pH 6.8]. Begin the reactions by introducing ATP to a final concentration of 1.0 mM (unlabeled ATP from Sigma and [g-32P]ATP from Amersham Pharmacia Biotech, specific activity 0.21 Ci/mmol). At timed intervals (e.g., 1.5, 3.0, 6.0, 12, 24, and 48 min), withdraw a reaction aliquot from the cuvette (10 ml) and mix it rapidly with the stop buffer (3.3 ml) from the appropriate tube. The stopped reactions may now be transferred to air. Electrophorese 10 ml from each time point on an 11% (w/v) SDS polyacrylamide gel (Fig. 10.1A) (Laemmli, 1970). Spot 1 ml from the later time points (e.g., 12, 24, 48 min) onto a polyethyleneimine-cellulose thin-layer chromatographic (TLC) plate, about 2 cm from the bottom and 1 cm from the next time point; include a control on the plate containing only the radiolabeled ATP (Fig. 10.1B). After air drying the plate, develop it in a sealed TLC container with a 1-cm layer of 0.75 M NaH2PO4, pH 3.5. Quantify the levels of phosphorylated protein in the dried gels and of low-molecular weight species on the TLC plates with a phosphorimager (Bio-Rad Personal Molecular Imager FX). As standards for the quantification, spot 1-ml 10-fold dilutions of the stock ATP solution onto a strip of cellulose for development alongside the gel and TLC. If the protein is free of phosphatase, the TLC will show no significant generation of free phosphate (Pi) in 45 min, and the autoradiograph will show a continuous increase or a leveling of the autophosphorylation over the same period (Fig. 10.1). If the protein is contaminated with a phosphatase, free phosphate will be obvious from the TLC, and the level of protein phosphorylation will decline after reaching an apparent peak. For such preparations, further purification is advised. Verification of kinase activity The absence of contaminating phosphatase activity is necessary but not sufficient to ensure that the entire enzyme in a preparation is active. Neither is the initial rate of autophosphorylation a reliable indicator of enzyme quality. A healthy kinase preparation should yield about 20% phosphorylated kinase at equilibrium. If a modified protein
180
Marie-Alda Gilles-Gonzalez et al.
A 0.5
1
1.5
3
6
12
24 48 min
DevS
B Pi
ATP
3 12 48 min ATP
DevS
Figure 10.1 Quality of the M. tuberculosis O2 sensor DevS is verified readily by autophosphorylation. The increasing autophosphorylation of deoxy-DevS even after 24 min from the autoradiograph (A) and the insignificant production of free phosphate even in 48 min from the polyethyleneimine TLC plates (B) show this preparation to be free of phosphatase contamination. From Sousa et al. (2007b).
substrate unable to accept phosphate is used, phosphorylation of the kinase should proceed nearly to completion. Any enzyme that does not meet these standards should be discarded if one wishes to have measurements that are both accurate and reproducible from batch to batch. 2.2.3. Strategy for purifying response-regulator substrates These proteins are best isolated by affinity purification. We have found that an epitope-tagged 6-His (Invitrogen) introduced at the N-terminal end of the gene by recombinant DNA methods works well for purifying these response regulators. The protein is purified as recommended by the manufacturer of the tag and affinity column (nickel-charged affinity resin) and assayed based on Western blotting and antibody recognition of the tag
Heme-Ligand Sensing Histidine-Protein Kinases
181
(Invitrogen). For quality control, examine phosphorylation of the protein in an extended turnover assay, with time points ranging from 1 to 30 min (see later). Quantitative phosphorylation of response regulators (i,e., FixJ) entails a phosphorylation in one subunit of a dimer (Da Re et al., 1999). Acceptable preparations of kinase and protein substrate should result in about 50% phosphorylation of the protein substrate at equilibrium (Sousa et al., 2007a).
2.3. Measuring the Kd for binding of ligand 2.3.1. Direct titration with O2 Preparation of O2-saturated buffer Prepare a solution of 1.3 mM O2 in 50 mM Tris-HCl, 50 mM KCl, and 5% (v/v) ethylene glycol at pH 8.0 by bubbling pure O2 for about 30 min through this buffer at room temperature (23 ) in a septum-sealed glass vial with a needle and an escape. Move the sealed vial to an anaerobic chamber. Transfer aliquots of the O2 solution from the vial to anaerobic buffer or the deoxy protein, as necessary, with a gas-tight Hamilton syringe. Basis spectra and exploratory titration Before starting a ligand titration, it is essential to record absorption spectra of the protein at 0 and 100% of saturation with the ligand. These spectra will serve as basis spectra for data analysis and will make clear the boundaries for titration. Prepare deoxy protein (4 mM ) as described earlier; place the protein solution in a septumsealed cuvette and record its 350- to 700-nm absorption spectrum. Add a concentrated aliquot of the deoxy heme protein (final concentration 4 mM ) to a septum-sealed solution of O2-saturated buffer (1.3 mM O2) with a gas-tight syringe and record the 350- to 700-nm absorption. Record both spectra in an Excel spreadsheet. An optimum linear combination of these spectra may be used to fit the spectrum resulting from each experimental titration (acceptable overall fit should be >0.98) and generate a percentage of saturation. If the spectrum cannot be closely fit to a linear combination of liganded and unliganded spectra, this means that some process other than ligand association is occurring and that some other species is contributing to spectra collected during titration. Most commonly, this means that the heme iron is oxidizing during titration, and the error spectrum is that of the ferric state. The reason that whole spectra should always be fitted, rather than a few wavelengths, is to detect the generation of extra species during the titration. One should not attempt to correct for oxidized protein or any third species; instead, find a way to control the interfering process and repeat the titration. Heme-based sensor variants with Kd values for O2 ranging from millimolar to submicromolar may be encountered; usually, there is no previous information on the expected affinity for O2. Therefore, a quick exploratory titration is recommended. Add five O2 concentrations, ranging
182
Marie-Alda Gilles-Gonzalez et al.
from about 1 to 600 mM O2, sequentially to the protein within the same cuvette, minimizing the headspace as much as possible, and record the absorption spectrum resulting from each O2 concentration. Estimate the point of half saturation and design an experiment for exploring the most informative range of O2 concentrations (O2 required to give 10–100% saturation). Full titration A full titration experiment might require about 10 O2 concentrations to cover the range for 10–100% saturation and should be repeated at least twice (Fig. 10.2). To minimize escape of gases or concurrent reactions (e.g., oxidation), O2 titration points should be prepared individually and their absorption recorded immediately. For example, prepare a solution of a known O2 concentration (e.g., 2.0 mM O2); combine it with deoxy protein (1–4 mM final) in a 1-cm path length cuvette, taking care to minimize the headspace, and immediately seal the cuvette. Record A Fraction Oxy-DosT
1.00 0.75 Kd = 26 mM n = 1.0
0.50 0.25 0.00
0
100
200 300 O2 (mM)
400
500
Fraction Oxy-DevS
B 1.00 0.75 Kd = 3 mM n = 1.0
0.50 0.25 0.00 0
25
50 O2 (mM)
75 150 250
Figure 10.2 The Kd for binding of O2 can be directly determined by titration with ligand. (A) Ferrous M. tuberculosis DosT (2 mM) was titrated with 1^1200 mM O2 at pH 8.0 at 25. Saturation data were fit to a nonlinear Hill equation from which the Kd and Hill coefficient n were determined. (B) The Kd for binding of O2 to the higher-affinity M. tuberculosis DevS protein was determined similarly by titrating the deoxy form (2 mM) with 0.8^256 mM O2 under the same conditions as in A. From Sousa et al. (2007b).
Heme-Ligand Sensing Histidine-Protein Kinases
183
the 350- to 700-nm absorption of this mixture. Repeat this procedure for all the O2 concentrations covered by titration. If the protein is prone to oxidation, the most challenging aspect of the titration is to make certain that the spectrum for each concentration is recorded after thorough mixing but before any oxidation. Reducing agents to control oxidation should be avoided, especially at low O2 concentrations, as these will consume O2. If it is necessary to use reducing agents, one must verify that the O2 saturation does not change for that specific O2 concentration during the time needed to collect the spectrum. Saturation can be determined by comparing actual spectra to linear combinations of basis deoxy and oxy spectra. Treated data are plotted and fitted to a nonlinear Hill plot equation and to a quadratic single-binding equation using one’s favorite curve-fitting software (e.g., Microsoft Excel or GraphPad Prism). An alternative to titration in a sealed cuvette is to use a system equilibrated with a large headspace of gas (e.g., a tonometer) at a controlled gas concentration. This is how most of the early titrations of hemoglobin and myoglobin were done. This approach requires designing and fabricating custom glassware, but it may be easier than the sealed approach if one has fairly large volumes of protein available. 2.3.2. Direct titration with CO Determination of the Kd value for binding CO can be done by a procedure similar to the aforementioned determination of O2 affinity for most sensors because their Kd values for binding of CO typically fall in the 0.5 to 10 mM range. The CO-saturated buffer (1.0 mM) is prepared in a fume hood by bubbling pure CO for about 30 min through an anaerobic solution of 50 mM Tris-HCl, 50 mM KCl, and 5% (v/v) ethylene glycol at pH 8.0 and 25 in a gas-washing bottle or any septum-sealed glass vial with a needle and an escape. For basis spectra, record the 350- to 700-nm absorption of the deoxy heme protein (1 mM) and the identical concentration of ferrous protein in a 1.0 mM solution of CO. For titrations, transfer aliquots of the CO solution from a septum-sealed vial to a sealable cuvette with the deoxy protein via a gastight Hamilton syringe. 2.3.3. Competition titration with NO The much higher NO affinity of heme proteins forbids their direct titration with NO. Instead, carbon monoxy protein is prepared at a known CO concentration and then titrated with NO, which displaces the CO (Fig. 10.3). The Kd value for binding of NO to the heme protein can be calculated from the ‘‘apparent’’ Kd in the presence of CO and the precisely known Kd value for binding of CO. Data analysis is the same as that of an ordinary titration, except that the basis spectrum for the unliganded state is replaced by the spectrum of carbon monoxy protein. To prepare the solution of NO-saturated heme protein, first prepare anaerobic buffer [50 mM
184
Marie-Alda Gilles-Gonzalez et al.
Fraction NO-DosT
A 1.00 0.75 0.50
Kd (NO) ~ 5 nM
0.25 0.00 0
2 4 6 8 NO (mM), in 240 mM CO
10
Fraction NO-DevS
B 1.00 0.75 0.50 Kd (NO) ~ 10 nM 0.25 0.00
0
6 12 18 24 NO (mM), in 10 mM CO
30
Figure 10.3 The Kd for binding of NO can be determined by competition against CO. (A) Ferrous M. tuberculosis DosT in 240 mM CO (Kd(CO) ¼ 0.94 mM, n ¼ 1.0) was titrated competitively with 0.50^9.0 mM NO at pH 8.0 at 25. (B) Ferrous M. tuberculosis DevS in 10 mM CO (Kd(CO) ¼ 0.036 mM, n ¼ 1.0) was titrated competitively with 0.50^28 mM NO under the same conditions as in A. From Sousa et al. (2007b).
Tris-HCl, 50 mM KCl, and 5% (v/v) ethylene glycol at pH 8.0 and 23 ] inside of an anaerobic chamber in a vessel fitted with a three-way valve. Maximize the surface area of the solution and do not fill more than one quarter of the vessel with liquid. Seal the vessel, transfer it to a fume hood, and connect it to nitrogen and NO tanks via the three-way valve. Briefly flow nitrogen through the line and reservoir and then switch to the NO tank and flow a gentle stream of NO through the reservoir for 30 min. Transfer the NOsaturated buffer (2.0 mM NO, 1 ml) from the reservoir to the anaerobic chamber in a gas-tight Hamilton syringe. Prepare a solution of 1 mM ferrous protein in 20–40 mM NO; the heme protein should be fully saturated with NO. The very popular NO-generating reagents are not suitable for quantitative analytical work because they do not allow concentrations of NO to be precisely controlled and determined and they generate reactive species other than NO. The importance of handling NO and its solutions properly cannot be overemphasized. The pure gas is extremely reactive and rapidly destroys most metals and plastics. Nitric oxide vessels and tanks should be kept in a fume
185
Heme-Ligand Sensing Histidine-Protein Kinases
hood. Only polytetrafluoroethylene, silicone rubber, stainless steel, and glass should be used in apparatus that contacts NO. All traces of O2 and other substances that will react with NO must be scrupulously excluded from all solutions. Otherwise, the very remarkable phenomena that you will certainly observe will be only indirectly because of NO. A serious leak will result in nitric acid generation, which can be easily verified by checking the sample pH. One may also generate highly oxidizing species such as peroxynitrite. A sure sign that there is a leak somewhere in the NO-handling apparatus, or that something in the buffer reacts with NO, is that the heme protein absorption spectrum irreversibly changes or the protein denatures. One should also be aware that the pure NO in the supply tank, even in the absence of contaminants, may disproportionate into N2O5 and N2 over time. In the event that a novel heme protein displays highly unusual behavior toward NO, one should always test a protein control such as myoglobin, whose reactions with NO are thoroughly known, before announcing one’s discovery in a press release.
2.4. Determination of turnover rates, kcat 2.4.1. General procedure For heme-controlled histidine-protein kinases such as FixL, assays of kcat measure the rates at which a sensor of known heme status phosphorylates its protein substrate (Fig. 10.4). Such studies are essential for quantifying the
Phospho-FixJ (pmol)
100 80 60 40
1.0 2.5 5.0 10 20 30 min
20 0 0
10
20 Time (min)
30
40
Figure 10.4 The best measure of kinase activity for B. japonicum FixL is its rate of turnover (kcat) of FixJ to phospho-FixJ. This reaction was catalyzed by deoxy-BjFixL (1 mM) for 25 mM BjFixJ and 1 mM ATP at pH 8.0 at 23. The kcat is 1.5 min-1 (gray line). Quantitative phosphorylation of FixJ occurred at equilibrium, verifying the excellent quality of the enzyme and substrates, with one-half of the BjFixJ (200 pmol) being phosphorylated within 30 min. From Sousa et al. (2007a).
186
Marie-Alda Gilles-Gonzalez et al.
influence of the heme on the kinase. For example, a typical experiment might compare phosphorylation of a response regulator (e.g., FixJ or DevR) by the deoxy, oxy, or carbon monoxy states of its kinase partner (e.g., FixL or DevS). For the turnover of FixJ to phospho-FixJ, reaction mixtures typically contain 1 mM FixL and 25 mM FixJ in 50 mM Tris-HCl, 50 mM KCl, 50 mM MnCl2, 1 mM MgCl2, and 5% (v/v) ethylene glycol, pH 8.0. Reactions are started by introducing the ATP (i.e., unlabeled ATP from Sigma and [g-32P]ATP of specific activity 0.42 Ci/mmol from Amersham Pharmacia Biotech) (see Fig. 10.4). They are stopped at 1.0, 2.5, 5.0, 10, 20, and 30 min by mixing 10-ml aliquots of the reaction mixtures with 3.3 ml of ‘‘stop buffer’’ [10 mM EDTA, 4% (w/v) sodium dodecyl sulfate, 0.50 M Tris-HCl, 0.20 M NaCl, 50% (v/v) glycerol, and 2% (v/v) b-mercaptoethanol, pH 6.8]. The products are electrophoresed on 15% (w/v) polyacrylamide gels (Laemmli, 1970). The levels of phosphorylated FixJ protein in the dried gels are quantified with a phosphorimager (Bio-Rad Personal Molecular Imager FX). A good preparation of kinase, in the on state, should quantitatively phosphorylate its protein substrate at equilibrium (see Fig. 10.4). Interpretation of kinetic data such as these critically relies on the assumption that the heme state remains the same throughout every reaction time course. It is therefore essential to verify 350- to 700-nm absorption spectra before and after every time course. It is also important to sample a sufficient number of time points so that each turnover rate may be computed at steady state, i.e., when the rate of FixJ phosphorylation matches the rate of replenishment of the phospho-FixL intermediate. One must additionally ensure that the concentration of all substrates is constant or else well above Km throughout the period of the reaction time course on which rate measurements are based, as extraction of fundamental rate constants from the reaction time course will otherwise be prohibitively complex and error prone. 2.4.2. Sensors with partially saturated heme Prepare the liganded protein as described earlier. Calculate saturation based on linear regression using whole spectra. Again, verify that the heme state remains unchanged throughout every time course, make sure that the enzyme is saturated with its substrates during the entire reaction, and collect enough data points to ensure coverage of the turnover rates at steady state.
2.5. Novel heme-containing histidine-protein kinases The vast majority of proteins predicted to be heme-containing histidine-protein kinases, or even identified as FixLs from their genetic context or possession of heme, have not been examined for their enzymatic activity and regulation of this activity by heme ligands (Freitas et al., 2003; Gilles-Gonzalez and Gonzalez, 2005; Iyer et al., 2003; Sardiwal et al., 2005;
Heme-Ligand Sensing Histidine-Protein Kinases
187
Taylor and Zhulin, 1999). We anticipate that many of these sensors will soon be studied for signal transduction as the complete enzymes. For initial study of a novel heme-containing kinase, additional considerations should be divalent-cation and heme-state requirements for activity. Although most histidine-protein kinases work well with MgII, others display strong preferences for specific divalent cations (Hess et al., 1991). For example, S. meliloti FixL performs best in assay mixtures that include MnII (50 mM) (Gilles-Gonzalez and Gonzalez, 1993). Likewise, although all hemecontaining kinases examined so far work best in the deoxy state, it is quite possible that this represents the ‘‘off ’’ state of some sensors, and inclusion of a ligand in the assays is necessary to manifest the ‘‘on’’ state.
ACKNOWLEDGMENTS This work was supported by NSF Grant 620531, the National Research Initiative of the USDA Cooperative State Research, Education and Extension Service, Grant 2002–35318– 14039, and Welch Foundation Grant I-1575.
REFERENCES Agron, P. G., Ditta, G. S., and Helinski, D. R. (1993). Oxygen regulation of nifA transcription in vitro. Proc. Natl. Acad. Sci. USA 90, 3506–3510. Appleby, C. A. (1980). In ‘‘Methods for Evaluating Biological Nitrogen Fixation’’ (F. J. Bergersen, ed.), pp. 315–335. Wiley, New York. Crosson, S., McGrath, P. T., Stephens, C., McAdams, H. H., and Shapiro, L. (2005). Conserved modular design of an oxygen sensory/signaling network with species-specific output. Proc. Natl. Acad. Sci. USA 102, 8018–8023. Da Re, S., Schumacher, J., Rousseau, P., Fourment, J., Ebel, C., and Kahn, D. (1999). Phosphorylation-induced dimerization of the FixJ receiver domain. Mol. Microbiol. 34, 504–511. David, M., Daveran, M.-L., Batut, J., Dedieu, A., Domergue, O., Ghai, J., Hertig, C., Boistard, P., and Kahn, D. (1988). Cascade regulation of nif gene expression in Rhizobium meliloti. Cell 54, 671–683. Ditta, G., Virts, E., Palomares, A., and Kim, C.-H. (1987). The nifA gene of Rhizobium meliloti is oxygen regulated. J. Bacteriol. 169, 3217–3223. Dixon, R., and Kahn, D. (2004). Genetic regulation of biological nitrogen fixation. Nat. Rev. Microbiol. 2, 621–631. Dunham, C. M., Dioum, E. M., Tuckerman, J. R., Gonzalez, G., Scott, W. G., and GillesGonzalez, M. A. (2003). A distal arginine in oxygen-sensing heme-PAS domains is essential to ligand binding, signal transduction, and structure. Biochemistry 42, 7701–7708. Fischer, H. M. (1994). Genetic regulation of nitrogen fixation in rhizobia. Microbiol. Rev. 58, 352–386. Freitas, T. A., Hou, S., and Alam, M. (2003). The diversity of globin-coupled sensors. FEBS Lett. 552, 99–104. Galinier, A., Garnerone, A.-M., Reyrat, J.-M., Kahn, D., Batut, J., and Boistard, J. (1994). Phosphorylation of the Rhizobium meliloti FixJ protein induces its binding to a compound regulatory region at the fixK promoter. J. Biol. Chem. 269, 23784–23789.
188
Marie-Alda Gilles-Gonzalez et al.
Gilles-Gonzalez, M. A. (2001). Oxygen signal transduction. IUBMB Life 51, 165–173. Gilles-Gonzalez, M. A., Caceres, A. I., Sousa, E. H., Tomchick, D. R., Brautigam, C., Gonzalez, C., and Machius, M. (2006). A proximal arginine R206 participates in switching of the Bradyrhizobium japonicum FixL oxygen sensor. J. Mol. Biol. 360, 80–89. Gilles-Gonzalez, M. A., Ditta, G. S., and Helinski, D. R. (1991). A haemoprotein with kinase activity encoded by the oxygen sensor of Rhizobium meliloti. Nature 350, 170–172. Gilles-Gonzalez, M. A., and Gonzalez, G. (1993). Regulation of the kinase activity of heme protein FixL from the two-component system FixL/FixJ of Rhizobium meliloti. J. Biol. Chem. 268, 16293–16297. Gilles-Gonzalez, M. A., and Gonzalez, G. (2005). Heme-based sensors: Defining characteristics, recent developments, and regulatory hypotheses. J. Inorg. Biochem. 99, 1–22. Gilles-Gonzalez, M. A., Gonzalez, G., Perutz, M. F., Kiger, L., Marden, M. C., and Poyart, C. (1994). Heme-based sensors, exemplified by the kinase FixL, are a new class of heme protein with distinctive ligand binding and autoxidation. Biochemistry 33, 8067–8073. Gong, W., Hao, B., Mansy, S. S., Gonzalez, G., Gilles-Gonzalez, M. A., and Chan, M. K. (1998). Structure of a biological oxygen sensor: A new mechanism for heme-driven signal transduction. Proc. Natl. Acad. Sci. USA 95, 15177–15182. Gonzalez, G., Gilles-Gonzalez, M. A., Rybak-Akimova, E. V., Buchalova, M., and Busch, D. H. (1998). Mechanisms of autoxidation of the oxygen sensor FixL and Aplysia myoglobin: Implications for oxygen-binding heme proteins. Biochemistry 37, 10188–10194. Hao, B., Isaza, C., Arndt, J., Soltis, M., and Chan, M. K. (2002). Structure-based mechanism of O2 sensing and ligand discrimination by the FixL heme domain of Bradyrhizobium japonicum. Biochemistry 41, 12952–12958. Hess, J. F., Bourret, R. B., and Simon, M. I. (1991). Phosphorylation assays for proteins of the two-component regulatory system controlling chemotaxis in Escherichia coli. Methods Enzymol. 200, 188–204. Ioanoviciu, A., Yukl, E. T., Moenne-Loccoz, P., and Montellano, P. R. (2007). DevS, a heme-containing two-component oxygen sensor of Mycobacterium tuberculosis. Biochemistry 46, 4250–4260. Iyer, L. M., Anantharaman, V., and Aravind, L. (2003). Ancient conserved domains shared by animal soluble guanylyl cyclases and bacterial signaling proteins. BMC Genomics 4, 5–12. Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685. Lois, A. F., Weinstein, M., Ditta, G. S., and Helinski, D. R. (1993). Autophosphorylation and phosphatase activities of the oxygen-sensing protein FixL of Rhizobium meliloti are coordinately regulated by oxygen. J. Biol. Chem. 268, 4370–4375. Monson, E. K., Weinstein, M., Ditta, G. S., and Helinski, D. R. (1992). The FixL protein of Rhizobium meliloti can be separated into a heme-binding oxygen-sensing domain and a functional C-terminal kinase domain. Proc. Natl. Acad. Sci. USA 89, 4280–4284. Mukai, M., Nakamura, K., Nakamura, H., Iizuka, T., and Shiro, Y. (2000). Roles of Ile209 and Ile210 on the heme pocket structure and regulation of histidine kinase activity of oxygen sensor FixL from Rhizobium meliloti. Biochemistry 39, 13810–13816. Ninfa, E. G., Atkinson, M. R., Kamberov, E. S., and Ninfa, A. J. (1993). Mechanism of autophosphorylation of Escherichia coli nitrogen regulator II (NRII or NtrB): Trans-phosphorylation between subunits. J. Bacteriol. 175, 7024–7032. Olson, J. S., and Phillips, G. N. (1997). Myoglobin discriminates between O2, NO, and CO by electrostatic interactions with the bound ligand. J. Biol. Inorg. Chem. 2, 544–552.
Heme-Ligand Sensing Histidine-Protein Kinases
189
Reyrat, J. M., David, M., Blonski, C., Boistard, P., and Batut, J. (1993). Oxygen-regulated in vitro transcription of Rhizobium meliloti nifA and fixK genes. J. Bacteriol. 175, 6867–6872. Roberts, D. M., Liao, R. P., Wisedchaisri, G., Hol, W. G., and Sherman, D. R. (2004). Two sensor kinases contribute to the hypoxic response of Mycobacterium tuberculosis. J. Biol. Chem. 279, 23082–23087. Saini, D. K., Malhotra, V., Dey, D., Pant, N., Das, T. K., and Tyagi, J. S. (2004a). DevRDevS is a bona fide two-component system of Mycobacterium tuberculosis that is hypoxiaresponsive in the absence of the DNA-binding domain of DevR. Microbiology 150, 865–875. Saini, D. K., Malhotra, V., and Tyagi, J. S. (2004b). Cross talk between DevS sensor kinase homologue, Rv2027c, and DevR response regulator of Mycobacterium tuberculosis. FEBS Lett. 565, 75–80. Sardiwal, S., Kendall, S. L., Movahedzadeh, F., Rison, S. C., Stoker, N. G., and Djordjevic, S. (2005). A GAF domain in the hypoxia/NO-inducible Mycobacterium tuberculosis DosS protein binds haem. J. Mol. Biol. 353, 929–936. Sciotti, M. A., Chanfon, A., Hennecke, H., and Fischer, H. M. (2003). Disparate oxygen responsiveness of two regulatory cascades that control expression of symbiotic genes in Bradyrhizobium japonicum. J. Bacteriol. 185, 5639–5642. Scopes, R. K. (1994). ‘‘Protein Purification Principles and Practice,’’ 3rd ed. SpringerVerlag, New York. Soupene, E., Foussard, M., Boistard, P., Truchet, G., and Batut, J. (1995). Oxygen as a key developmental regulator of Rhizobium meliloti N2 fixation gene expression within the alfalfa root nodule. Proc. Natl. Acad. Sci. USA 92, 3759–3763. Sousa, E. H. S., Gonzalez, G., and Gilles-Gonzalez, M.-A. (2005). Oxygen blocks reaction of the FixL/FixJ complex with ATP but does not influence binding of FixJ or ATP to FixL. Biochemistry 44, 15359–15365. Sousa, E. H. S., Tuckerman, J. R., Gonzalo, G., and Gilles-Gonzalez, M.-A. (2007a). A memory of oxygen binding explains the dose response of the heme-based sensor FixL. Biochemistry 46, 6249–6257. Sousa, E. H. S., Tuckerman, J. R., Gonzalo, G., and Gilles-Gonzalez, M.-A. (2007b). DosT and DevS are oxygen-switched kinases in Mycobacterium tuberculosis. Protein Sci. 16, 1708–1719. Stock, J. B., Ninfa, A. J., and Stock, A. M. (1989). Protein phosphorylation and regulation of adaptive responses in bacteria. Microbiol. Rev. 53, 450–490. Taylor, B. L., and Zhulin, I. B. (1999). PAS domains: Internal sensors of oxygen, redox potential, and light. Microbiol. Mol. Biol. Rev. 63, 479–506. Tuckerman, J. R., Gonzalez, G., Dioum, E. M., and Gilles-Gonzalez, M. A. (2002). Ligand and oxidation-state specific regulation of the heme-based oxygen sensor FixL from Sinorhizobium meliloti. Biochemistry 41, 6170–6177. Tuckerman, J. R., Gonzalez, G., and Gilles-Gonzalez, M. A. (2001). Complexation precedes phosphorylation for two-component regulatory system FixL/FixJ of Sinorhizobium meliloti. J. Mol. Biol. 308, 449–455. Virts, E. L., Stanfield, S. W., Helinski, D. R., and Ditta, G. S. (1988). Common regulatory elements control symbiotic and microaerobic induction of nifA in Rhizobium meliloti. Proc. Natl. Acad. Sci. USA 85, 3062–3065.
This page intentionally left blank
C H A P T E R
E L E V E N
Reactions of Nitric Oxide and Oxygen with the Regulator of Fumarate and Nitrate Reduction, a Global Transcriptional Regulator, during Anaerobic Growth of Escherichia coli Jason C. Crack,* Nick E. Le Brun,* Andrew J. Thomson,* Jeffrey Green,† and Adrian J. Jervis† Contents 1. Introduction 2. Production of 4Fe-FNR Protein 2.1. Purification of native 4Fe-FNR 2.2. Purification of reconstituted 4Fe-FNR 2.3. Cleaning and concentration of 4Fe-FNR 3. Determination of Iron and Acid-Labile Sulfide Content of FNR 3.1. Determination of iron content 3.2. Determination of sulfide content 4. UV-Visible Absorbance Spectra of FNR 5. Cluster Reaction with Nitric Oxide and Oxygen 5.1. Reaction of 4Fe-FNR with nitric oxide 5.2. Reaction of 4Fe-FNR with oxygen 6. Purification of 2Fe-FNR 7. Detection of Other Reaction Products 7.1. Detection of sulfide 7.2. Detection of iron 7.3. Detection of superoxide and peroxide 8. Conclusions References
* {
192 194 194 195 197 197 197 198 198 198 199 201 204 204 204 205 205 206 207
Centre for Metalloprotein Spectroscopy and Biology, School of Chemical Sciences and Pharmacy, University of East Anglia, Norwich, United Kingdom Department of Molecular Biology and Biotechnology, University of Sheffield, Sheffield, United Kingdom
Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37011-0
#
2008 Elsevier Inc. All rights reserved.
191
192
Jason C. Crack et al.
Abstract The Escherichia coli fumarate and nitrate reductase (FNR) regulator protein is an important transcriptional regulator that controls the expression of a large regulon of more than 100 genes in response to changes in oxygen availability. FNR is active when it acquires a [4Fe-4S]2þ cluster under anaerobic conditions. The presence of the [4Fe-4S]2þ cluster promotes protein dimerization and sitespecific DNA binding, facilitating activation or repression of target promoters. Oxygen is sensed by the controlled disassembly of the [4Fe-4S]2þ cluster, ultimately resulting in inactive, monomeric, apo-FNR. The FNR [4Fe-4S]2þ cluster is also sensitive to nitric oxide, such that under anaerobic conditions the protein is inactivated by nitrosylation of the iron-sulfur cluster, yielding a mixture of monomeric and dimeric dinitrosyl-iron cysteine species. This chapter describes some of the methods used to produce active [4Fe-4S] FNR protein and investigates the reaction of the [4Fe-4S]2þ cluster with nitric oxide and oxygen in vitro.
1. Introduction Fumarate and nitrate reductase (FNR) regulator proteins are members of the cAMP receptor protein (CRP)/FNR superfamily of transcription regulators (Korner et al., 2003). All members of the family are structurally related to the founder member (Schultz et al., 1991). Thus, CRP/FNR family members share a basic two-domain structure consisting of an N-terminal sensory domain and a C-terminal DNA-binding domain. The versatility of this framework is illustrated by the range of metabolic and environmental signals perceived by different members of the family and by the variety of physiological responses that they control (Green et al., 2001). The best-characterized FNR protein is that of Escherichia coli. In E. coli FNR acts as a direct oxygen sensor and is the primary transcriptional regulator of the switch between aerobic and anaerobic growth (Green et al., 2001; Guest, 1992, 1995; Sawers, 1999). Thus, under anaerobic conditions FNR is in its active state and is able to bind specific palindromic sequences of DNA (Eiglmeier et al., 1989; Green et al., 1996b; Lazazzera et al., 1993, 1996). Once bound to DNA, FNR activates target gene expression by recruiting RNA polymerase (RNAP) or represses transcription by preventing the formation of productive RNAP:DNA complexes (Barnard et al., 2003; Wing et al., 1995). Generally, FNR activates genes encoding products involved in anaerobic metabolism, such as the nar operon (nitrate reductase), the dms operon (dimethyl sulfoxide reductase), and the frd operon (fumarate reductase), and represses genes encoding products involved in aerobic
Reaction of FNR with Nitric Oxide and Oxygen
193
metabolism, such as the sdh operon (succinate dehydrogenase) and ndh (NADH dehydrogenase II). FNR is activated under anaerobic conditions by the acquisition of one [4Fe-4S]2þ cluster per monomer (designated 4Fe-FNR). Each [4Fe-4S]2þ cluster is ligated by four cysteine residues (Cys20, Cys23, Cys29, and Cys122), and the presence of the cluster promotes FNR dimerization, increasing its capacity to bind specifically to DNA (Green et al., 1996a; Khoroshilova et al., 1997; Kiley and Beinert, 1999; Lazazzera et al., 1993; 1996; Melville and Gunsalus, 1990; Sharrocks et al., 1990). When exposed to oxygen the [4Fe-4S]2þ cluster is oxidized to a [2Fe-2S]2þ cluster, via a transient [3Fe-4S]1þ form, resulting in the formation of 2Fe-FNR (Crack et al., 2004, 2007; Jordan et al., 1997; Khoroshilova et al., 1995; Lazazzera et al., 1996; Popescu et al., 1998; Sutton et al., 2004a). If aerobic conditions persist, the [2Fe-2S]2þ cluster is further degraded, yielding apo-FNR (Achebach et al., 2005; Green et al., 1991; Sutton et al., 2004b; Unden and Schirowski, 1997). In contrast to 4Fe-FNR, apo-FNR and 2Fe-FNR are monomeric and are unable to bind efficiently to DNA and are therefore inactive (Green et al., 1991; Lazazzera et al., 1996; Sutton et al., 2004a,b). If anaerobic conditions return, 4Fe-FNR is generated either by recycling apo-FNR or 2Fe-FNR or by de novo synthesis of 4Fe-FNR (Dibden and Green, 2005; Khoroshilova et al., 1997; Mettert and Kiley, 2005). In addition to its reaction with oxygen, the FNR [4Fe-4S]2þ cluster is also sensitive to nitric oxide (NO). Upon exposure to NO the [4Fe-4S]2þ cluster becomes nitrosylated, forming a combination of monomeric and dimeric dinitrosyl-iron-cysteine (DNIC) complexes, again abolishing its ability to bind DNA (Cruz-Ramos et al., 2002). Among the genes repressed by FNR under anaerobic conditions is hmp, which encodes the flavohemoglobin that is one of the major mechanisms for detoxifying NO in E. coli (Poole, 2005; Poole et al., 1996). When cultures of E. coli are exposed to NO, FNR repression of hmp is relieved, suggesting that the reaction between FNR and NO is physiologically significant (Cruz-Ramos et al., 2002). This suggestion is supported by transcript profiling experiments that reveal that the abundances of many FNR-activated genes are lower, and many FNR-repressed genes are greater, when E. coli is exposed to NO (Justino et al., 2005; Pullan et al., 2007). Thus, 4Fe-FNR is a sensor of both oxygen and NO; consequently, purification and manipulation in vitro must be carried out under strictly anoxic conditions. This chapter provides information on how to isolate 4Fe-FNR and study its reaction with oxygen and NO. It is hoped that some of the techniques will be transferable, such that they are of value not only for those interested in FNR proteins, but also for those working with proteins possessing labile iron-sulfur clusters.
194
Jason C. Crack et al.
2. Production of 4Fe-FNR Protein Two current methods are used to obtain functional, anaerobic 4Fe-FNR. The first is to overproduce and purify native 4Fe-FNR under anaerobic conditions (Crack et al., 2006; Sutton et al., 2003). The second is to produce apo-FNR under aerobic conditions and then reconstitute the purified protein with the [4Fe-4S]2þ cluster under anaerobic conditions in vitro (Crack et al., 2004; Green et al., 1996a). There has been some debate centered around whether reconstituted 4Fe-FNR behaves in the same way as 4Fe-FNR that has been isolated under anaerobic conditions, although evidence indicates that there is essentially no difference in their reaction with oxygen (Crack et al., 2006, 2007). Regardless of the method used to produce anaerobic 4Fe-FNR, the protein is very oxygen sensitive, and hence certain stages of production, purification, and storage must be carried out under strictly anaerobic conditions. To achieve this, anaerobic cabinets (Belle Technology/Don Whitley) typically operating at 2.0 ppm oxygen are required. To help maintain anaerobic conditions in the cabinets, all buffers are sparged with oxygen-free nitrogen gas for a minimum of 2 h before introduction into the anaerobic cabinet. In addition, plastic items are equilibrated in the anaerobic cabinet for a minimum of 24 h before being used. Furthermore, it is beneficial if the anaerobic cabinets are also equipped with a protein purification system, a fridge-freezer for anaerobic sample storage (Belle Technology), and a liquid nitrogen access port.
2.1. Purification of native 4Fe-FNR The fnr gene was polymerase chain reaction amplified using oligonucleotide primers that introduced a unique upstream NcoI site and included the native downstream BamHI site. The amplified DNA was then ligated into the commercially available vector pET11d (Novagen) to create pGS1859 (Crack et al., 2006). This expression plasmid has a copy of fnr under the control of the inducible T7 promoter. Native FNR is produced from cultures of E. coli BL21lDE3 containing pGS1859 as described by Sutton and Kiley (2003) with the following modifications. Cultures that have been induced for FNR production are subsequently incubated on ice for 5 to 10 min and supplemented with 200 mM ferric ammonium citrate, 50 mM L-methionine, and 0.001% (v/v) antifoam 204 (Sigma) prior to sparging with oxygen-free nitrogen gas at 4 overnight. Unless otherwise stated, all subsequent work and manipulations are performed under strictly anaerobic conditions. The sparged cultures are transferred to O-ring-sealed centrifuge tubes (Beckman), and bacteria are collected by centrifugation outside the anaerobic cabinet at 6500 g for 10 min at 4 .
Reaction of FNR with Nitric Oxide and Oxygen
195
From a total culture volume of 5 liters the bacterial cell pellets are resuspended in 105 ml of buffer A [10 mM potassium phosphate, 100 mM KCl, 10% (v/v) glycerol, pH 6.8]. Then 12 ml of CelLytic lysis buffer (CelLytic B 10; Sigma), lysozyme (21.3 mg ml-1), DNase I (0.5 mg ml-1), phenylmethylsulfonyl fluoride (PMSF, 0.1 mM), and benzamidine (1 mM ) are added (total volume 120 ml). The suspension is mixed gently and incubated at room temperature for up to 20 min. The cell lysate is then transferred to O-ring-sealed centrifuge tubes (Nalgene) and centrifuged outside the anaerobic cabinet at 40,000 g for 45 min at 2 . The resulting cell-free extract is made 2% (v/v) with buffer B [10 mM potassium phosphate, 800 mM KCl, 10% (v/v) glycerol, pH 6.8] and is applied to a 10-ml HiTrap SP FF cation-exchange column (GE Healthcare) equilibrated previously with buffer A. After application of the extract, the column is washed with 15% (v/v) buffer B (in buffer A), and bound proteins are eluted using a linear gradient (60 ml) of 15 to 100% (v/v) buffer B. Fractions containing FNR are pooled, diluted threefold with buffer C [10 mM potassium phosphate, 10% (v/v) glycerol, pH 6.8], and loaded onto a 2-ml HiTrap heparin column (GE Healthcare) equilibrated previously with 5% (v/v) buffer B. The column is washed with 10% (v/v) buffer B, and bound protein is eluted using a linear gradient (50 ml) from 10 to 100% (v/v) buffer B. Fractions containing FNR are pooled, concentrated if necessary (see later), and stored in glass vials in an anaerobic fridge-freezer until needed.
2.2. Purification of reconstituted 4Fe-FNR In this method, FNR is produced aerobically with an N-terminal GST tag and then purified under anaerobic conditions by affinity purification using a glutathione-Sepharose column (GE Healthcare). On column cleavage of the GST-FNR fusion protein results in apo-FNR, which contains no cluster. To obtain 4Fe-FNR, apo-FNR is reconstituted with a [4Fe-4S]2þ cluster by in vitro incubation with L-cysteine and NifS [a cysteine desulfurase purified as described by Zheng and co-workers (1993)] as a source of sulfur and ammonium ferrous sulfate as a source of iron. To create the GST-FNR fusion protein, the same fnr-containing DNA fragment used for native FNR production (see earlier discussion) is ligated onto vector pGEX-KG (Guan and Dixon, 1991) to create plasmid pGS572 (Green et al., 1996b). E. coli BL21lDE3 pGS572 cultures are grown in 2.5-liter conical flasks containing 500 ml L-broth (Sambrook and Russell, 2001) medium supplemented with ampicillin (100 mg liter-1) at 37 with shaking at 250 rpm. When cultures reach A600 0.6, the production of GST-FNR is induced by the addition of 1 mM isopropyl-b-thiogalactopyranoside and incubation is continued for 3–4 h. The cultures are then transferred to O-ring-sealed centrifuge tubes (Beckman), and bacteria are
196
Jason C. Crack et al.
pelleted at 18,000 g for 10 min. The supernatant is discarded, and the cell pellet is either used immediately or stored at –80 until needed. Cell pellets are resuspended in 35 ml of buffer D (25 mM HEPES, 2.5 mM CaCl2, 100 mM NaCl, 100 mM NaNO3, pH 7.5) per liter of culture, and 1 mM benzamidine and 0.1 mM PMSF are added before cell lysis by passage through a French pressure cell three times at 20,000 psi or by sonication in the presence of lysozyme (21.3 mg ml-1). Unbroken cells are removed by centrifugation at 40,000 g for 20 min, and the resulting supernatant is then filtered through a 0.45-mm pore syringe filter. To isolate FNR from extracts containing GST-FNR, all steps are carried out in an anaerobic cabinet. Aliquots of the filtered supernatant from 1 liter of culture are applied to three columns containing 3 ml glutathioneSepharose (GE Healthcare) preequilibrated with 60 ml buffer D. Once the extract has passed through the column, the matrix is washed with 60 ml of buffer D. Once washed, the column is capped and 10 units of thrombin (Sigma) in 0.5 ml buffer D is added and mixed into the matrix bed. The column is then incubated at an ambient temperature (21 ) overnight to allow release of FNR from the bound GST-FNR fusion, by specific cleavage of the linker sequence joining GST to FNR. The anaerobic apo-FNR protein is eluted by the addition of 4 ml buffer D in 1-ml aliquots, which are collected as 1-ml fractions. Fractions are then pooled, giving a total of 12 ml eluent from 1 liter of culture. A 200-ml sample of the pooled fractions is taken for protein assays (Bio-Rad) to determine its concentration (usually close to 80 mM) using a correction factor of 0.83 (Green et al., 1996a). Purity is usually assessed by SDS-PAGE analysis. To reconstitute the iron-sulfur cluster into apo-FNR solutions, FeS-1 (50 mM L-cysteine, 125 mM dithiothreitol, 25 mM HEPES, 2.5 mM CaCl2, 100 mM NaCl, 100 mM NaNO3, pH 7.5) and FeS-2 [20 mM (NH4)2Fe(SO4)2, 25 mM HEPES, 2.5 mM CaCl2, 100 mM NaCl, 100 mM NaNO3, pH 7.5] are freshly prepared under anaerobic conditions using anaerobic buffer D to dissolve the solutes. Reconstitution of the ironsulfur cluster is initiated by the addition of an aliquot of FeS-1 (1 mM L-cysteine, 2.5 mM dithiothreitol, final concentrations), an appropriate amount of FeS-2, such that there is a 7 M excess of Fe2þ ions per FNR monomer, and an aliquot of NifS (225 nM final concentration). Small-scale (¼3 ml) reconstitutions are carried out in anaerobic cuvettes sealed with a screw cap containing a septa made from rubber and coated with a layer of Teflon (Hellma UK, Ltd.). Once sealed, the cuvettes may be removed from the anaerobic cabinet and transferred to a spectrophotometer fitted with a thermostatic cell holder to follow the progress of the reconstitution reaction. Larger reconstitutions are carried out inside the anaerobic cabinet using a small conical flask fitted with a water jacket. The reaction is sealed using a rubber stopper.
Reaction of FNR with Nitric Oxide and Oxygen
197
Reconstitutions, via either method, are stirred (using a magnetic follower) throughout and typically are complete after 4.5 h at 37 . Reconstituted FNR protein is then cleaned and concentrated using a 1-ml HiTrap heparin column (GE Healthcare) as described later.
2.3. Cleaning and concentration of 4Fe-FNR After reconstitution, 4Fe-FNR is separated from low molecular weight components using a 1-ml HiTrap heparin column (GE Healthcare). This step can also be used to concentrate dilute preparations of 4Fe-FNR from either purification method. The 4Fe-FNR binds to the columns at low KCl concentrations (100 mM ), but is eluted at higher concentrations (>400 mM KCl). All stages are carried out under strictly anaerobic conditions. The heparin column is equilibrated with 10 ml of the appropriate buffer (i.e., buffer A for native preparations and buffer D for reconstitutions, see earlier discussion). The pooled FNR fractions are then passed through a 0.22-mm filter onto the column and washed with 10–20 ml of the equilibration buffer to remove low molecular weight species. The column is then inverted, and FNR is eluted from the end of the column to which it was applied, using buffer (typically buffer A or D) containing 500 mM KCl. The eluent is collected in microcentrifuge tubes, four drops at a time to obtain the highest concentration. At low concentrations 4Fe-FNR has a straw brown color but is green/black at high concentrations. The most concentrated fractions are pooled together. The concentration of 4Fe-FNR present in the final preparation is determined by assaying for protein (described earlier), iron, acid-labile sulfide (see later), and spectrophotometrically assuming e405 nm of 16,200 M-1 cm-1 per [4Fe-4S]2þ cluster (Crack et al., 2006).
3. Determination of Iron and Acid-Labile Sulfide Content of FNR 3.1. Determination of iron content The iron content of FNR is determined as follows: 0.1 ml of 21.7% HNO3 is added to the same volume of protein and incubated at 95 for 30 min. Cooled samples are centrifuged to remove any precipitate, treated with 0.6 ml of 7.5% (w/v) ammonium acetate, 0.1 ml of 12.5% (w/v) ascorbic acid, and 0.1 ml of 10 mM ferene (5,50 (3-(2-pyridyl)-1,2,4-triazine5,6-diyl)-bis-2-furansulfonate), mixed, and incubated at room temperature for 30 min. The concentration of the ferene-iron complex is calculated from measurements of its absorbance at 593 nm. Iron concentrations are
198
Jason C. Crack et al.
determined by reference to a calibration curve generated from Fe(III) solutions in the range of 0–200 mM, prepared from SpectrosoL standard iron solution (BDH), and treated as described earlier.
3.2. Determination of sulfide content Acid-labile sulfide is determined, via the formation of methylene blue, according to the method of Beinert (1983). Briefly, 0.2 ml of appropriately diluted sample is treated with 0.6 ml of freshly prepared 1% (w/v) zinc acetate and 50 ml of 12% (w/v) NaOH. The sample tube is capped, inverted, and incubated at room temperature for 15 min. After incubation the sample is centrifuged at 3000 rpm for 60 s, treated with 0.15 ml 0.1% (w/v) N,N-dimethyl-p-phenylenediamine dihydrochloride dissolved in 5 M HCl and 0.15 ml 10 mM FeCl3 dissolved in 1 M HCl. The sample is then vortexed for 30 s and incubated at room temperature for a further 30 min. The concentration of the acid-labile sulfide is calculated from the formation of methylene blue measured through its absorbance at 670 nm. Sulfide concentrations are determined by reference to a calibration curve generated from anaerobic Na2S solutions containing 10 mM NaOH in the range of 0–300 mM, according to the method of Beinert (1983), and treated as described earlier.
4. UV-Visible Absorbance Spectra of FNR The state of FNR can be followed using UV-visible spectroscopy as the [4Fe-4S]2þ cluster has characteristic spectral features. Apo-FNR typically exhibits only protein-associated absorbance at 280 nm, whereas 4Fe-FNR shows absorbance maxima at 320 nm (e ¼ 20,075 350 M-1 cm-1) and 405 nm (e ¼ 16,200 135 M-1 cm-1) with a characteristic broad shoulder around 420 nm. Examples of apo-FNR and of [4Fe-4S] FNR spectra are shown in Fig. 11.1.
5. Cluster Reaction with Nitric Oxide and Oxygen FNR acts as both a NO and an oxygen sensor via its [4Fe-4S]2þ cluster (see references given earlier). Much research has been performed to investigate the reaction of the cluster with molecular oxygen, but less on its reaction with NO. The two main types of experiment aim to determine the rates and stoichiometries of the reactions.
199
Reaction of FNR with Nitric Oxide and Oxygen
35,000 1.4 30,000 Absorbance
1.2
e M −1 cm−1
30,000 20,000 15,000
0.8 0.6 0.4 0.2 0 280
10,000 5000 0 260
1.0
330
400
470
350
420 490 560 Wavelength (nm)
610 680 540 Wavelength (nm)
750
630
820
890
Figure 11.1 UV-visible spectrum of isolated 4Fe-FNR. (Inset) Changes in the UV-visible spectrum of apo-FNRduring reconstitution of 4Fe-FNR.
When 4Fe-FNR is reacted with either NO or oxygen the electronic spectrum changes, most notably at 420 nm, because of alteration of the [4Fe4S]2þ cluster. To follow these changes, reactions are generally carried out in anaerobic cuvettes sealed with a screw cap containing a septum made from rubber and coated with a layer of Teflon (as used in the reconstitution reaction). The seal allows the injection of buffer containing known concentrations of the appropriate reactant (NO or oxygen) after which the electronic spectrum of the protein can be obtained, or samples can be removed for electron paramagnetic resonance (EPR) spectroscopy. Reactions of FNR with NO or oxygen are fast, typically being complete within 5 min at 20 .
5.1. Reaction of 4Fe-FNR with nitric oxide Generally the fast NO-releasing compound proline NONOate is used as a source of NO, although NO solutions can be used (Cruz-Ramos et al., 2002). Each NONOate molecule releases 1.9 molecules of NO, with a half-life of 13 s in the working buffer at 25 (Cruz-Ramos et al., 2002). An anoxic 4.17 mM stock solution of proline NONOate is made just before use in 25 mM Tris, pH 10.0, 100 mM NaCl and stored at –1 until needed. Typically 1 ml of 4Fe-FNR with a cluster concentration of 30 mM is transferred to a cuvette in an anaerobic cabinet. Then fixed concentrations of proline NONOate are introduced into the cuvette by injection through
200
Jason C. Crack et al.
the septum, and the solution is mixed. The reaction of NO with 4Fe-FNR is very rapid and is essentially dependent on the release of NO from proline NONOate at 25 . Reaction with NO (Fig. 11.2A) results in the appearance A
1.0
Absorbance
0.9
0.8
0.8
0.7
0.7
0.6 0.5
0.6
0.4
0.5
0.3
0.4
0.2 300 325 350 375 400 425 450 475 500
0.3 0.2 0.1 0 280
380
480 580 Wavelength (nm)
680
780
B
2.08
2.06
2.04
2.02
2.00
1.98 NO (mM) 5.0 22.4 60.0 140.0
g value
Figure 11.2 Spectral changes upon reaction of 4Fe-FNR with NO. (A) UV-visible spectra of anaerobic 4Fe-FNR (27 mM ) before (thin line) and after (thick line) treatment with NO (140 mM ). (B) EPR spectra at 77 K of anaerobic 4Fe-FNR treated with NO as indicated. The microwave power was 2.000 mW and the frequency was 9.669 GHz. Modulation amplitude, frequency, and receiver gain were 10 gauss, 100 kHz, and 1 105. For both experiments the buffer used was 25 mM HEPES, 2.5 mM CaCl2, 100 mM NaCl, and 100 mM NaNO3, pH 7.5.
Reaction of FNR with Nitric Oxide and Oxygen
201
of a shoulder at 360 nm in the UV-visible spectrum together with loss of the broad 420-nm shoulder. An isosbestic point is observed at 480 nm (CruzRamos et al., 2002). The final spectrum corresponds to a mix of monomeric and dimeric DNIC species in an approximate 20:80 ratio, giving a relatively intense yellow color to the solution. Because the product of reaction of 4Fe-FNR with NO is stable for several hours, the stoichiometry of the reaction can be determined by adding limiting amounts of NONOate, allowing it to react, and then recording spectra. By plotting the dA360 against NO:[4Fe-4S]2þ it was found that the reaction of NO and FNR is complete at a ratio of 3:1 (Cruz-Ramos et al., 2002). Determining the rate of reaction when using proline NONOate is difficult, as the NO is not immediately available for reaction when introduced. To avoid this, aqueous solutions of dissolved NO gas could be used. However, some protein precipitation has been observed, resulting in baseline changes, while using aqueous NO solutions. Another useful tool for investigating the reaction of 4Fe-FNR with NO is EPR spectroscopy. The EPR system used consists of an X-band Bruker EMX spectrometer equipped with an ESR-900 helium flow cryostat and a TE-102 microwave cavity (Oxford Instruments). The spin intensities of paramagnetic samples can be quantified by the integration of EPR spectra using 1 mM Cu(II), 10 mM EDTA as the standard. Neither apo-FNR nor 4Fe-FNR exhibits an EPR signal; however, samples of 4Fe-FNR reacted with NO, as described earlier ([NO]:[4Fe-4S]2þ of 5.0), that have been frozen rapidly in EPR tubes to 77 K, display a signal with an axial g tensor centered on g 2.03, reminiscent of EPR signals observed for monomeric cysteine DNIC species (see Fig. 11.2B) (Cruz-Ramos et al., 2002).
5.2. Reaction of 4Fe-FNR with oxygen Similar experiments to those described earlier can be carried out to investigate the FNR cluster reaction with oxygen following the addition of air-saturated buffer (buffer A or D, see earlier discussion) to anaerobic samples of FNR. The concentration of dissolved oxygen present in the aerobic buffers at the reaction temperature is determined chemically (Vogel, 1989). Briefly, aliquots (100 ml) of the buffer are equilibrated to the reaction temperature in a water bath and then sealed. Dissolved oxygen is fixed by injecting Mn(II)SO4 H2O (2.1 M, 1 ml) below the surface of the sample, followed by 1 ml of a solution of NaOH (40%), NaI (20%), and NaN3 (0.5%). The headspace of the volumetric flask is then purged briefly with oxygen-free nitrogen and the contents are mixed by inversion. After incubation for 15 min, 2 ml of phosphoric acid (85%) is added and the liberated iodine is titrated with a solution of Na2S2O3 5H2O (49.9 mM )
202
Jason C. Crack et al.
and a starch indicator (1%). Air-saturated buffers typically contain 220 mM oxygen at 20 . Figure 11.3A shows electronic spectra for 4Fe-FNR before and after oxidation with oxygen. The reaction with oxygen changes the electronic spectrum. The 420-nm shoulder is replaced with a broad shoulder at 430 nm, the absorbance maxima at 310 nm (e ¼ 12,089 M-1 cm-1) and 420 nm (e ¼ 8690 M-1 cm-1) decrease in intensity, and there is an increase in absorbance between 500 and 600 nm (Crack et al., 2004; Jordan et al., 1997; Khoroshilova et al., 1997). This new spectrum indicates the presence of 2Fe-FNR and is characterized by a red/brown color. Stoichiometric experiments carried out as described earlier (and plotting O2:[4Fe-4S]2þ against dA420) show the reaction of 4Fe-FNR with oxygen to be 87% complete at an O2:[4Fe-4S]2þ of 2.0, with the addition of further oxygen only having a minimal effect (Crack et al., 2004). To investigate the rate of reaction of oxygen with 4Fe-FNR, typically a total volume of 2 ml is used. Aerobic and anaerobic buffers are mixed in the cuvette to give a known oxygen concentration, and the cuvette is transferred to a spectrophotometer. The reaction is started by the addition of an aliquot of 4Fe-FNR to the desired concentration. Changes in absorbance spectra during the reactions are monitored at a single wavelength, typically 420 nm, with data points being recorded every 0.5 s. Hence, the conversion of 4Fe-FNR to 2Fe-FNR can be followed. The concentration of 4Fe-FNR for these experiments is variable depending on the ratio of O2: [4Fe-4S]2þ required, but typically it is in the range of 5–30 mM cluster. To calculate the kinetics of the reaction, the decrease in absorbance at 420 nm is plotted. As mentioned earlier, 4Fe-FNR, 2Fe-FNR, and apo-FNR are devoid of an EPR signal. However, when 4Fe-FNR is reacted with stoichiometric amounts of oxygen and frozen rapidly in EPR tubes to 77 K, samples display a signal centered at g 2.024, which behaves in a manner characteristic of [3Fe-4S]1þ clusters (Crack et al., 2004, 2007). Time-resolved appearance of the [3Fe-4S]1þ cluster is observed by reacting 4Fe-FNR with excess oxygen (e.g., a O2:[4Fe-4S]2þ of 10) and rapidly freezing aliquots in EPR tubes throughout the time course (see Fig. 11.3A). Initially, the [3Fe-4S]1þ signal grows in intensity, before decreasing; this behavior is typical of an intermediate, transient species (Crack et al., 2004, 2007). Another useful tool in following the reaction of 4Fe-FNR with oxygen is circular dichroism (CD) spectroscopy. 4Fe-FNR samples are prepared as described previously and reacted with oxygenated buffer, and spectral changes are recorded using Jasco J-810 spectropolarimeter scanning at 200 nm min-1. CD spectroscopy of 4Fe-FNR reveals weak, but reproducible, bands in the 280- to 800-nm region with three positive features (lmax 330, 380, and 420 nm) that, upon oxygen titration, are replaced by a broader spectrum with two positive features (lmax 320 and 440 nm) (see Fig. 11.3C).
203
Reaction of FNR with Nitric Oxide and Oxygen
A
35,000 30,000
e M −1 cm−1
25,000 20,000 15,000 10,000 5000 0 260
360
460 560 660 Wavelength (nm)
760
860
g = 2.024 g = 2.031 g = 2.019
B 0s 16 s
22 s 45 s 95 s 125 s
3330
3380
3430
3480
3530
Field (gauss) C6 5
4Fe-FNR 4.4 mM O2 6.5 mM O2 10.7 mM O2 21.3 mM O2
4 3 2 1 0 280
330
380
430
480
530
580
630
680
−1 −2
Wavelength (nm)
Figure 11.3 Spectral changes upon reaction of 4Fe-FNRwith oxygen. (A) UV-visible spectra of anaerobic 4Fe-FNR (110 mM ) before treatment with oxygen (thick line). Isolated 2Fe-FNR (80 mM ) generated after treatment of 4Fe-FNRwith oxygen (thin line). (B) EPR spectra of anaerobic 4Fe-FNR (20 mM ) at different times during treatment with oxygen (220 mM ). (C) CD spectra of anaerobic 4Fe-FNR (30.8 mM ) titrated with oxygen (22 mM final). In all cases the buffer used was 25 mM HEPES, 2.5 mM CaCl2,100 mM NaCl, and 100 mM NaNO3, pH 7.5.
204
Jason C. Crack et al.
6. Purification of 2Fe-FNR Once produced, 2Fe-FNR is stable for several hours even in the presence of oxygen and can be purified after exposure of 4Fe-FNR to oxygen, as described previously. To obtain sufficient yields, typically a 1-ml solution of 4Fe-FNR in the region of 120 mM [4Fe-4S]2þ is transferred to a glass vial and removed from the anaerobic cabinet. The lid of the vial is removed and 500 ml aerobic buffer is added. The solution is exposed to the air while being agitated gently for a maximum of 90 s. The lid is then replaced and the vial is transferred back to the anaerobic cabinet. The 2Fe-FNR is isolated from cluster breakdown products using a preequilibrated Sephadex G25 PD10 column (GE Healthcare) according to the manufacturer’s instructions, collecting the colored fractions in microcentrifuge tubes.
7. Detection of Other Reaction Products Techniques have been developed or modified to allow the detection of potential products of the reaction of 4Fe-FNR with oxygen, including ferrous ions, sulfide, hydrogen peroxide, and superoxide. Precise identification of these products is important because the nature of the reaction between [4Fe-4S]2þ clusters and oxygen (metal or sulfide based) can be determined by such analyses.
7.1. Detection of sulfide If the oxidation of 4Fe-FNR is sulfide based, then sulfur, rather than sulfide, would most likely be the product of the reaction with oxygen. To test for the production of sulfide during the reaction of 4Fe-FNR with oxygen, a modified version of the protocol reported by Nashef and co-workers (1977) was developed (Crack et al., 2006). Briefly, aliquots of 4Fe-FNR in anaerobic cuvettes (final concentration 5–10 mM) are treated with 200 mM 5,50 dithiobis-(2-nitrobenzoic acid) (DTNB) under anaerobic conditions at room temperature. After 2 min the absorbance is measured at 412 nm. This measurement allows calculation of the amount of reactive thiol in the sample. Oxygen-saturated buffer is then injected to a final oxygen concentration of 40 mM and left to react for 12 min at which point the reaction is complete. The change in absorbance at 412 nm is measured, allowing calculation of the amount of sulfide released in the conversion of 4Fe-FNR to 2Fe-FNR, after correcting the value for the change at 412 nm because of cluster
Reaction of FNR with Nitric Oxide and Oxygen
205
conversion (obtained by reacting 4Fe-FNR with oxygen in the absence of DTNB). The reactions are calibrated using a standard sodium sulfide solution, prepared as described by Beinert (1983). The molar extinction coefficients used are: e412 nm ¼ 14,151 M-1 cm-1 in phosphate buffer (buffer A) and e412 nm ¼ 12,344 M-1 cm-1 in HEPES buffer (buffer D).
7.2. Detection of iron The oxidation state of the iron atoms released during the 4Fe-FNR to 2FeFNR conversion is of considerable current interest. Ferene, an Fe(II)specific chelator, has been used to detect iron release from 4Fe-FNR (Crack et al., 2007; Sutton et al., 2004a). It is important to establish that the iron-sulfur cluster under consideration is stable in the presence of ferene. This is the case for FNR in the absence of oxygen. In the presence of Fe(II), the [Fe(II)(ferene)3]4þ complex is formed, which can be quantified using its intense absorbance at 593 nm (e593 nm for [Fe(II)(ferene)3]4þ in phosphatebased buffers is 39,600 M-1 cm-1; e593 nm for [Fe(II)(ferene)3]4þ in HEPESnitrate-based buffers is 32,243 M-1 cm-1). The amount and oxidation state of iron released upon 4Fe-FNR cluster oxidation may be determined by introducing 4Fe-FNR (2 mM final concentration) into phosphate buffer (50 mM potassium phosphate, 400 mM KCl, 10% glycerol, pH 6.8) containing different amounts of oxygen and ferene (100 mM ). Under these conditions, 1Fe(II) is detected for each [4Fe-4S]2þ cluster oxidized (Crack et al., 2007). The total amount of iron released can then be determined by repeating the reaction in the presence of the reductant, sodium ascorbate (4 mM ). This converts any Fe(III) released to Fe(II), which is again detected as a ferene complex. For the oxidation of FNR by oxygen, these experiments suggest that 1Fe(II) and 1Fe(III) are released in the 4Fe-FNR to 2Fe-FNR conversion (Crack et al., 2007). A similar mixture of Fe(II) and Fe(III) has been observed using Mo¨ssbauer spectroscopy in vivo and in vitro (Khoroshilova et al., 1997; Popescu et al., 1998).
7.3. Detection of superoxide and peroxide Two methods have been developed to detect the oxygen species generated during the reaction of 4Fe-FNR with oxygen. To detect superoxide, the method described by McCord and Fridovich (1968) was modified. This method is based on the spectrophotometric detection of cytochrome c reduction by superoxide. Thus, 4Fe-FNR (10 mM ) is injected into a cuvette containing cytochrome c (74 mM, Sigma) in aerobic buffer A (220 mM oxygen, at 21 ). After mixing, the reaction is incubated at room
206
Jason C. Crack et al.
temperature. By monitoring changes in absorbance at 550 nm the reduction of cytochrome c is monitored (De550 nm ¼ 21,000 M-1 cm-1). To show that the reduction of cytochrome c is caused by superoxide, control reactions containing superoxide dismutase (16 units) are required (Crack et al., 2007). Superoxide can dismutate spontaneously to form hydrogen peroxide, which will reoxidize reduced cytochrome c. Therefore, reactions containing catalase (234 units) are also informative. For FNR oxidation, the presence of catalase increased the amount of superoxide detected, indicating that some hydrogen peroxide is generated during the oxidation of 4FeFNR (Crack et al., 2007). Indeed, hydrogen peroxide can be detected in a coupled reaction using the fluorescent probe Amplex Red (Molecular Probes). A solution of Amplex Red (400 mM ) containing 2 units of horseradish peroxidase is prepared in FNR purification buffer. One milliliter of 4Fe-FNR (60 mM ) is added to 1 ml of Amplex Red solution in a standard fluorescence cuvette under anaerobic conditions. The sealed cuvette is transferred to a Perkin Elmer LS55 spectrofluorimeter and titrated with oxygen by injection of aliquots of aerobic purification buffer. Fluorescence spectra are recorded using an excitation wavelength of 545 nm (5-nm slit width) and an emission slit width of 2 nm. Hydrogen peroxide is detected by an emission maximum at 587 nm and can be quantified by calibrating the reactions by the addition of an aliquot of a standard hydrogen peroxide solution into the reaction mixture at the end of titration (Crack et al., 2004).
8. Conclusions Using the methods described in this chapter, the reactions of FNR with oxygen and NO have been investigated. The application of these techniques has provided new insight into FNR [4Fe-4S] cluster chemistry, suggesting the following reactions with NO [Eq. (11.1)] and oxygen [Eq. (11.2)]:
½4Fe 4S2þ þ 4NO ! 2FeðNOÞ2 þ 2Fe3þ þ 4S2 ½4Fe 4S2þ þ O2 ! ½2Fe 2S2þ þ Fe2þ þ Fe3þ þ 2S2 þ O 2
ð11:1Þ ð11:2Þ
It should be appreciated that quantitative experiments to investigate the reactions of iron-sulfur clusters with oxidants are technically demanding and chemically complex, and despite the progress that has been made there are many outstanding questions, which will require the development of new methodologies before a deeper understanding of these processes is achieved.
Reaction of FNR with Nitric Oxide and Oxygen
207
REFERENCES Achebach, S., Selmer, T., and Unden, G. (2005). Properties and significance of apo-FNR as a second form of air-inactivated [4Fe-4S] FNR of Escherichia coli. FEBS J. 272, 4260–4269. Barnard, A. M., Green, J., and Busby, S. J. (2003). Transcription regulation by tandembound FNR at Escherichia coli promoters. J. Bacteriol. 185, 5993–6004. Beinert, H. (1983). Semi-micro methods for analysis of labile sulfide, and of labile sulfide plus sulfane sulfur in unusually stable iron-sulfur proteins. Anal. Biochem. 131, 373–378. Crack, J., Green, J., and Thomson, A. J. (2004). Mechanism of oxygen sensing by the bacterial transcription factor fumarate-nitrate reduction (FNR). J. Biol. Chem. 279, 9278–9286. Crack, J. C., Green, J., Cheesman, M. R., Le Brun, N. E., and Thomson, A. J. (2007). Superoxide-mediated amplification of the oxygen-induced switch from [4Fe-4S] to [2Fe-2S] clusters in the transcriptional regulator FNR. Proc. Natl. Acad Sci. 104, 2092–2097. Crack, J. C., Green, J., Le Brun, N. E., and Thomson, A. J. (2006). Detection of sulfide release from the oxygen-sensing [4Fe-4S] cluster of FNR. J. Biol. Chem. 281, 18909–18913. Cruz-Ramos, H., Crack, J., Wu, G., Hughes, M. N., Scott, C., Thomson, A. J., Green, J., and Poole, R. K. (2002). NO sensing by FNR: Regulation of the Escherichia coli NOdetoxifying flavohaemoglobin, Hmp. EMBO J. 21, 3235–3244. Dibden, D., and Green, J. (2005). In vivo cycling of the Escherichia coli transcription factor FNR between active and inactive states. Microbiology 151, 4063–4070. Eiglmeier, K., Honore, N., Iuchi, S., Lin, E. C. C., and Cole, S. T. (1989). Molecular genetic analysis of FNR-dependent promoters. Mol. Microbiol. 3, 869–878. Green, J., Bennett, B., Jordan, P., Ralph, E. T., Thomson, A. J., and Guest, J. R. (1996a). Reconstitution of the [4Fe-4S] cluster in FNR and demonstration of the aerobicanaerobic transcription switch in vitro. Biochem. J. 316, 887–892. Green, J., Irvine, A. S., Meng, W., and Guest, J. R. (1996b). FNR-DNA interactions at natural and semi-synthetic promoters. Mol. Microbiol. 19, 125–137. Green, J., Scott, C., and Guest, J. R. (2001). Functional versatility in the CRP-FNR superfamily of transcription factors: FNR and FLP. Adv. Microb. Phys. 44, 1–34. Green, J., Trageser, M., Six, S., Unden, G., and Guest, J. R. (1991). Characterization of the FNR protein of Escherichia coli, an iron-binding transcriptional regulator. Proc. Roy. Soc. Lond. B 244, 137–144. Guan, K. L., and Dixon, J. E. (1991). Eukaryotic proteins expressed in Escherichia coli: An improved thrombin cleavage and purification procedure of fusion proteins with glutathione S-transferase. Anal. Biochem. 192, 262–267. Guest, J. R. (1992). Oxygen-regulated gene expression in Escherichia coli: The 1992 Marjory Stephenson Prize Lecture. J. Gen. Microbiol. 138, 2253–2263. Guest, J. R. (1995). The Leeuwenhoek Lecture, 1995: Adaptation to Life without Oxygen. Philos. Trans. R. Soc. Lond. B. Biol. Sci. 350, 189–202. Jordan, P. A., Thomson, A., Ralph, E. T. J., Guest, J. R., and Green, J. (1997). FNR is a direct oxygen sensor having a biphasic response curve. FEBS Lett. 416, 349–352. Justino, M. C., Vicente, J. B., Teixeira, M., and Saraiva, L. M. (2005). New genes implicated in the protection of anaerobically grown Escherichia coli against nitric oxide. J. Biol. Chem. 280, 2636–2643. Khoroshilova, N., Beinert, H., and Kiley, P. J. (1995). Association of a polynuclear ironsulfur center with a mutant FNR protein enhances DNA binding. Proc. Natl. Acad. Sci. USA 92, 2499–2503.
208
Jason C. Crack et al.
Khoroshilova, N., Popescu, C., Munck, E., Beinert, H., and Kiley, P. J. (1997). Iron-sulfur cluster disassembly in the FNR protein of Escherichia coli by O2: [4Fe-4S] to [2Fe-2S] conversion with loss of biological activity. Proc. Natl. Acad. Sci. USA 94, 6087–6092. Kiley, P. J., and Beinert, H. (1999). Oxygen sensing by the global regulator, FNR: The role of the iron-sulfur cluster. FEMS Microbiol. Rev. 22, 341–352. Korner, H., Sofia, H. J., and Zumft, W. G. (2003). Phylogeny of the bacterial superfamily of Crp-Fnr transcription regulators: Exploiting the metabolic spectrum by controlling alternative gene programs. FEMS Microbiol. Rev. 27, 559–592. Lazazzera, B. A., Bates, D., and Kiley, P. J. (1993). The activity of the Escherichia coli transcription factor FNR is regulated by a change in oligomeric state. Genes Dev. 7, 1993–2005. Lazazzera, B. A., Beinert, H., Khoroshilova, N., Kennedy, M. C., and Kiley, P. J. (1996). DNA binding and dimerization of the Fe-S-containing FNR protein from Escherichia coli are regulated by oxygen. J. Biol. Chem. 271, 2762–2768. McCord, J. M., and Fridovich, I. (1968). The reduction of cytochrome c by milk xanthine oxidase. J. Biol. Chem. 243, 5753–5760. Melville, S. B., and Gunsalus, R. P. (1990). Mutations in fnr that alter anaerobic regulation of electron transport-associated genes in Escherichia coli. J. Biol. Chem. 265, 18733–18736. Mettert, E. L., and Kiley, P. J. (2005). ClpXP-dependent proteolysis of FNR upon loss of its O2-sensing [4Fe-4S] cluster. J. Mol. Biol. 354, 220–232. Nashef, A. S., Osuga, D. T., and Fenney, R. E. (1977). Determination of hydrogen sulfide with 5,50 -dithiobis-(2-nitrobenzoic acid), N-ethylmaleimide, and parachloromercuribenzoate. Anal. Biochem. 79, 394–405. Poole, R. K. (2005). Nitric oxide and nitrosative stress tolerance in bacteria. Biochem. Soc. Trans. 33, 176–180. Poole, R. K., Anjum, M. F., Membrillo-Hernandez, J., Kim, S. O., Hughes, M. N., and Stewart, V. (1996). Nitric oxide, nitrite, and Fnr regulation of hmp (flavohemoglobin) gene expression in Escherichia coli K-12. J. Bacteriol. 178, 5487–5492. Popescu, C. V., Bates, D. M., Beinert, H., Munck, E., and Kiley, P. J. (1998). Mossbauer spectroscopy as a tool for the study of activation/inactivation of the transcription regulator FNR in whole cells of Escherichia coli. Proc. Natl. Acad. Sci. USA 95, 13432–13435. Pullan, S. T., Gidley, M. D., Jones, R. A., Barrett, J., Stevanin, T. M., Read, R. C., Green, J., and Poole, R. K. (2007). Nitric oxide in chemostat-cultured Escherichia coli is sensed by Fnr and other global regulators: Unaltered methionine biosynthesis indicates lack of S-nitrosation. J. Bacteriol. 189, 1845–1855. Sambrook, J., and Russell, D. W. (2001). ‘‘Molecular Cloning: A Laboratory Manual,’’ 3rd ed. Cold Spring Harbor LaboratoryCold Spring Harbor, NY. Sawers, G. (1999). The aerobic/anaerobic interface. Curr. Opin. Microbiol. 2, 181–187. Schultz, S. C., Shields, G. C., and Steitz, T. A. (1991). Crystal structure of a CAP-DNA complex: The DNA is bent by 90 . Science 253, 1001–1007. Sharrocks, A., Green, J., and Guest, J. R. (1990). In vivo and in vitro mutants of FNR, the anaerobic transcription factor of Escherichia coli. FEBS Lett. 270, 119–122. Sutton, V. R., and Kiley, P. J. (2003). Techniques for studying the oxygen-sensitive transcription factor FNR from Escherichia coli. Methods Enzymol. 370, 300–312. Sutton, V. R., Mettert, E. L., Beinert, H., and Kiley, P. J. (2004a). Kinetic analysis of the oxidative conversion of the [4Fe-4S]2þ cluster of FNR to a [2Fe-2S]2þ cluster. J. Bacteriol. 186, 8018–8025. Sutton, V. R., Stubna, A., Patschkowski, T., Munck, E., Beinert, H., and Kiley, P. J. (2004b). Superoxide destroys the [2Fe-2S]2þ cluster of FNR from Escherichia coli. Biochemistry 43, 791–798. Unden, G., and Schirawski, J. (1997). The oxygen-responsive transcriptional regulator FNR of Escherichia coli: The search for signals and reactions. Mol. Microbiol. 25, 205–210.
Reaction of FNR with Nitric Oxide and Oxygen
209
Vogel, A. I. (1989). In ‘‘Textbook of Quantitative Chemical Analysis’’ ( J. Mendham, G. H. Jeffery, J. Bassett, and R. C. Denney, eds.) 5th ed. pp. 395–399. Longman Scientific & Technical, Harlow UK. Wing, H. J., Williams, S. M., and Busby, S. J. (1995). Spacing requirements for transcription activation by Escherichia coli FNR protein. J. Bacteriol. 177, 6704–6710. Zheng, L., White, R. H., Cash, V. L., White, R. F., and Dean, D. R. (1993). Cysteine desulfurase activity indicates a role for NifS in metallocluster biosynthesis. Proc. Natl. Acad. Sci. USA 90, 2754–2758.
This page intentionally left blank
C H A P T E R
T W E LV E
Genome-Wide Identification of Binding Sites for the Nitric Oxide-Sensitive Transcriptional Regulator NsrR Sam Efromovich,* David Grainger,† Diane Bodenmiller,‡ and Stephen Spiro§ Contents 1. Introduction 2. Strain Construction 3. Reference and Control Samples 4. Culture Conditions 5. Immunoprecipitation of DNA Targets Associated with NsrR 6. DNA Labeling, Microarray Hybridization, and Processing 7. Visualization and Analysis of DNA Microarray Data 8. A New Statistical Methodology for Treatment of ChIP-on-Chip Data 9. Conclusions Acknowledgments References
212 214 216 217 218 219 220 222 231 231 231
Abstract NsrR is a nitric oxide-sensitive regulator of transcription. In Escherichia coli, NsrR is a repressor of the hmp gene encoding the flavohemoglobin that detoxifies nitric oxide. Three other transcription units ( ytfE, ygbA, and hcp-hcr) are known to be subject to regulation by NsrR. This chapter describes experimental and statistical protocols used to identify NsrR-binding sites in the E. coli chromosome using chromatin immunoprecipitation and microarray analysis. The methods are applicable, with suitable modifications, to any regulatory protein and any organism.
* Department of Mathematical Sciences, University of Texas at Dallas, Richardson, Texas School of Biosciences, University of Birmingham, Edgbaston, Birmingham, United Kingdom School of Biology, Georgia Institute of Technology, Atlanta, Georgia } Department of Molecular and Cell Biology, University of Texas at Dallas, Richardson, Texas { {
Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37012-2
#
2008 Elsevier Inc. All rights reserved.
211
212
Sam Efromovich et al.
1. Introduction Escherichia coli uses nitrate and nitrite as terminal electron acceptors for anaerobic respiration, reducing nitrate to nitrite, and nitrite to ammonia. There is good evidence to indicate that low concentrations of nitric oxide (NO) are made as a by-product of this respiratory metabolism (Corker and Poole, 2003; Ji and Hollocher, 1988; Van Doorslaer et al., 2003). This endogenous formation of NO provides a physiological rationale for the expression of enzymes that reduce or oxidize NO to less toxic species, and regulatory proteins that sense NO and mediate transcriptional responses to NO exposure. In E. coli, three enzymes have established roles in detoxifying NO: the flavohemoglobin (Hmp) oxidizes NO to nitrate (in the presence of oxygen) and reduces NO to nitrous oxide (Poole and Hughes, 2000); the flavorubredoxin (FlRd) reduces NO to nitrous oxide (Gardner et al., 2002; Gomes et al., 2002); and the respiratory nitrite reductase (Nrf ) reduces NO to ammonia (Poock et al., 2002). The same enzymes may allow pathogenic strains of E. coli and related enteric bacteria to detoxify the NO made by host cells (Bang et al., 2006; Sebbane et al., 2006; Stevanin et al., 2002). A number of E. coli regulatory proteins have been reported to be sensitive to NO and to control gene expression in response to NO, including SoxR, OxyR, FNR, Fur, and MetR (Spiro, 2007). However, the most important regulators (in terms of providing physiological regulation of expression of genes encoding enzymes that detoxify NO) are, probably, NsrR and NorR (Spiro, 2007). NorR activates expression of the genes encoding the flavorubredoxin and its reductase in response to NO (D’Autre´aux et al., 2005; Gardner et al., 2003). NsrR is a recently described repressor protein, which is sensitive to NO (or possibly another N compound) and controls expression of, among others, genes encoding Hmp and Nrf (Bodenmiller and Spiro, 2006; Spiro, 2007). This chapter describes methods used to identify NsrR-binding sites in the E. coli genome. A recent study made use of comparative genomics to reconstruct the transcriptional networks that respond to nitric oxide (and other nitrogen oxides and oxyanions) in a wide range of prokaryotic species (Rodionov et al., 2005). In the case of E. coli, this study predicted that the product of the nsrR (formerly yjeB) gene is a transcriptional regulator of four transcription units: hmp, ytfE, ygbA, and hcp-hcr (Rodionov et al., 2005). Of these, the products of the hmp and ytfE genes have established roles in protecting against NO ( Justino et al., 2005, 2006; Poole, 2005). It has been confirmed experimentally that NsrR is a repressor of hmp, ytfE, and ygbA and that repression is relieved in the presence of physiological or chemical sources of NO (Bodenmiller and Spiro, 2006). We have subsequently also confirmed
The NsrR Regulon of Escherichia coli
213
(Filenko et al., 2007) that NsrR is a repressor of the hcp-hcr genes (encoding the hybrid cluster protein and its redox partner) and the nrf genes encoding the respiratory periplasmic nitrite reductase, which also has NO reductase activity (Poock et al., 2002). Thus, it seems that NsrR mediates widespread changes in the E. coli transcriptome in response to NO. It is of considerable interest to identify the full extent of the NsrR regulon, in other words, to identify all genes subject to regulation by NsrR. To this end, we have employed chromatin immunoprecipitation with microarray analysis (ChIPon-chip) to identify binding sites for NsrR in the E. coli genome. This chapter describes the techniques involved, which are applicable, with some modifications, to any regulatory protein (given a suitable antibody) and to any organism for which a high-density microarray is available. Chromatin immunoprecipitation is an in vivo technique used to monitor interactions between DNA-binding proteins and their targets. Proteins are cross-linked to DNA in vivo (typically by treatment with formaldehyde); after cell disruption the cross-linked chromatin is sheared randomly by sonication or enzymatic digestion. Protein–DNA complexes are selectively immunoprecipitated with a suitable antibody, the cross-linking is reversed, and the precipitated DNA is purified. The polymerase chain reaction (PCR) can then be used with specific primers to test whether candidate target sequences are enriched in the precipitated DNA population. In ChIP-on-chip, the precipitated DNA is (in some protocols) amplified and is suitably labeled for hybridization to a microarray. Since the precipitated DNA is typically in the range of 500–1000 bp, a conventional array with one probe per gene can be used. In this case the outcome is that probes within 500–1000 bp of the protein-binding site give a positive signal, allowing the experimenter to associate the site with one or two genes with a reasonable level of confidence (Molle et al., 2003). A modification of this approach involves the use of high-density microarrays, with short gaps between probes, which correspond to both coding and noncoding regions. This improves the resolution of data and allows signals to be associated unambiguously with a single transcription unit or with two divergently oriented transcription units. General considerations for the design and interpretation of ChIP-on-chip experiments have been reviewed elsewhere (Buck and Lieb, 2004). Chromatin immunoprecipitation and ChIP-on-chip methods were developed in eukaryotic systems, but have more recently been applied successfully in prokaryotes, for example, Bacillus subtilis (Molle et al., 2003; Rokop et al., 2004), Salmonella enterica (Shin and Groisman, 2005), and E. coli (Grainger et al., 2004, 2005). ChIP-on-chip has several advantages and disadvantages compared to other methods used to define regulons, for example, the use of microarrays to compare the transcriptomes of wild-type and mutant strains. Genes that
214
Sam Efromovich et al.
are subject to indirect regulation by NsrR will show NsrR-dependent responses to NO in a microarray experiment, but be negative in a ChIPon-chip experiment. Thus, the latter approach identifies candidate genes whose transcription is directly regulated by NsrR. This is both a strength and a weakness of the approach, depending on the investigator’s priorities. ChIP-on-chip requires no prior knowledge about other regulatory mechanisms that may impinge upon genes controlled by NsrR. For example, if expression of an NsrR-regulated gene is also dependent on another factor, then that gene may not be identified in a transcriptomics experiment, unless the appropriate growth conditions are used. ChIP-on-chip does identify binding sites associated with genes expressed at a low level, and where the regulatory protein may have little effect (Grainger et al., 2006). This is a potential disadvantage of the approach if binding at such sites has no biological significance. A major limitation of ChIP-on-chip is that some binding sites known from other approaches are not identified, in other words there is a significant false-negative rate (Grainger et al., 2006). Importantly, ChIP-on-chip only identifies factor-binding sites, demonstrating regulation of transcription occurring at those sites requires confirmation by an independent technique. Further, where binding sites are located in noncoding regions between divergently transcribed genes, ChIP-on-chip data cannot distinguish which gene (if either) is subject to regulation. An overview of the procedure is illustrated in Fig. 12.1 and a detailed protocol is presented here.
2. Strain Construction A requirement for ChIP-on-chip is the availability of an antibody for the selective precipitation of cross-linked protein–DNA complexes. In the case of NsrR, specific anti-NsrR antibodies are not yet available. Instead, we chose to epitope tag NsrR by introduction of the 22 amino acid 3 Flag tag at the C terminus of the protein. Introduction of this construct onto the chromosome avoids any artifacts associated with the overexpression that may result from the use of plasmid-borne constructs. The epitope tag is introduced by a PCR-based method with plasmid pSUB11 as the template (Uzzau et al., 2001). The tagged construct is then introduced onto the chromosome of E. coli strain BW25113 using the lRed recombinase method (Datsenko and Wanner, 2000). At each stage, DNA and strain constructions are confirmed by PCR and/or sequencing. This approach results in the introduction of a kanamycin resistance cassette in the chromosome downstream of the tagged gene. The cassette can be removed by recombination (Datsenko and Wanner, 2000), although this was not done
215
The NsrR Regulon of Escherichia coli
− Nitrate
+ Nitrate
Step 1
Step 2
Step 3
Cy5
Cy5
Step 4
Cy3
Cy5
Cy5 Cy5
Cy5
Cy5
Cy3 Cy3
Cy5 Cy3 Cy3
Cy5
Step 5
Cy5/Cy3 ratio
Cy3
Genomic location
NsrR 3xFLAG
Other DNA binding proteins
Anti-FLAG
DNA
Figure 12.1 Schematic of the ChIP-on-chip protocol. Two anaerobic cultures are grown of a strain expressing NsrR with a C-terminal epitope tag. Because growth in the presence of nitrate causes derepression of NsrR targets (Bodenmiller and Spiro, 2006), the occupancy of NsrR binding sites is lower under these conditions. Step1: nucleoprotein is cross-linked by treatment with formaldehyde. Step 2: chromatin is extracted and randomly sheared by sonication to produce fragments in the 500- to 1000-bp range. Step 3: NsrR:DNA complexes are selectively immunoprecipitated with an antibody raised against the epitope tag. Step 4: cross-linking is reversed and DNA is purified and fluorescently labeled using the Klenow fragment of DNA polymerase and random hexanucleotide primers. Step 5: labeled DNA samples are hybridized to a high-density microarray.
for the experiments described later. The cassette also provides a selectable marker allowing the epitope-tagged gene to be transferred to other strains by phage P1 transduction. It is important to confirm that the epitope tag
216
Sam Efromovich et al.
does not interfere with the normal activity of the regulatory protein. To do this, we transfer the construct into strain JOEY19, which contains a ytfElacZ reporter fusion (Bodenmiller and Spiro, 2006), and confirm that the epitope-tagged NsrR exerts NO-sensitive repression of the ytfE promoter in a manner indistinguishable from the wild-type protein. The tagged nsrR gene is also transferred into the wild-type strain MG1655 by transduction to generate strain JOEY135, which is used for the ChIP-on-chip experiments. The nsrR gene is cotranscribed with rnr, encoding ribonuclease R (Cairrao et al., 2003). The epitope-tagging procedure described earlier results in the introduction of a kanamycin resistance cassette between nsrR and rnr, which may perturb the level of expression of rnr by polarity. While this would be a concern for a transcriptomics experiment, we believe that it is not an issue for the ChIP-on-chip experiment described here because (a) an altered activity of RNase R is unlikely to have any consequence for the ability of NsrR to bind to its target sequences and (b) the test and control strains used in the protocol described later are genetically identical; they both contain the same construct used to epitope tag NsrR.
3. Reference and Control Samples For ChIP-on-chip experiments, we employed a dual-labeling protocol (Grainger et al., 2005), in which the immunoprecipitated DNA (later referred to as the experimental sample) is labeled with Cy5 (or Cy3) and a reference DNA is labeled with Cy3 (or Cy5). The two samples are then hybridized to the array simultaneously, and the fluorescence ratio at each probe is measured. Thus, each ChIP-on-chip experiment requires a hybridization reference DNA. Ideally, this should be the same for each experiment to allow comparisons to be made between experiments (Buck and Lieb, 2004). For example, a sample of genomic DNA taken after the sonication step and prior to immunoprecipitation can be used as the reference (Molle et al., 2003). The control sample is designed to detect nonbiological sources of variation. Possible sources of the control DNA are (1) DNA from a ‘‘dummy’’ immunoprecipitation reaction in which no antibody is present, (2) DNA immunoprecipitated from a control strain in which the regulatory protein is not epitope tagged, and (3) DNA immunoprecipitated from the epitope-tagged strain grown under conditions where the regulatory protein does not bind to DNA. In practice, criteria for an ideal experiment may not be met for technical reasons. For example, we have found that chromatin immunoprecipitation from strain MG1655 (in which NsrR is not epitope tagged) with the commercially available anti-Flag monoclonal antibody yielded insufficient DNA to use as a control in the ChIP-on-chip protocol.
The NsrR Regulon of Escherichia coli
217
Therefore, we prepared the control DNA from a culture of strain JOEY135 grown under conditions leading to derepression of NsrR targets, in other words, conditions where NsrR-binding sites are not fully occupied. This DNA was used as the reference in array hybridizations in which the experimental DNA came from a culture grown in the absence of nitrate. It has been argued that the control sample should not be used as the hybridization reference, as the ‘‘perfect’’ control experiment would not contain any DNA (Buck and Lieb, 2004). However, the approach described here has the advantage of increasing the information content of a single experiment in that where sites are identified we can be certain that binding at those sites is regulated (in this case by nitrate). Indeed, it is typical in published experiments for a control DNA to be used as the hybridization reference (Grainger et al., 2006), which is the approach recommended in protocols provided by Oxford Gene Technology (OGT)(www.ogt.co.uk).
4. Culture Conditions It is likely that the effector for NsrR is NO, either supplied exogenously or made endogenously as a by-product of the respiratory metabolism of nitrate and nitrite (Bodenmiller and Spiro, 2006; Spiro, 2007). Thus, preparation of the control DNA requires growth of a culture in the presence of NO. Experimentally, NO can be provided as a bolus of aqueous solution, or in the gas phase, but this makes the concentration difficult to control over time, as cultures of E. coli consume NO rapidly, and NO is susceptible to autooxidation, especially in the presence of oxygen. Compounds that release NO with known kinetics are also useful experimental tools. For the experiments described here, we instead chose to use a physiological source of NO, as it is known that E. coli makes NO endogenously as a by-product of the respiration of nitrate and nitrite (Corker and Poole, 2003; Van Doorslaer et al., 2003) and the NsrR regulon is derepressed in anaerobic cultures respiring nitrate or nitrite (Bodenmiller and Spiro, 2006). Because nitrite is toxic, for studies of gene expression we typically add it to cultures at a low concentration (5 mM ) shortly before samples are taken. Nitrate is not toxic and can therefore be present in cultures throughout growth in excess at high millimolar concentrations. For the purposes of this experiment, it is not necessary to know the concentration of NO made during nitrate and nitrite respiration. We assume that NO is made continuously during the logarithmic phase of growth in the presence of excess nitrate and know from studies of gene expression that both NorR and NsrR are sufficiently sensitive to respond to the NO made under these conditions. For the ChIP-on-chip experiment, the control culture of JOEY135 is grown anaerobically (in filled bottles) in 200 ml of L-broth supplemented with 0.5% (w/v) glucose and 50 mM potassium nitrate.
218
Sam Efromovich et al.
The experimental culture is grown under identical conditions, except that nitrate is omitted from the medium. In late logarithmic phase (OD650 nm 0.75–1.0), formaldehyde is added to cultures (final concentration 1%), which are then incubated for another 20 min. To quench the cross-linking, glycine is added to a final concentration of 500 mM, and cultures are incubated for a further 5 min.
5. Immunoprecipitation of DNA Targets Associated with NsrR The following protocols for chromatin immunoprecipitation and DNA labeling and hybridization are based on published methods (Grainger et al., 2005) with some minor modifications (e.g., volumes are scaled up to account for the lower yield of anaerobic cultures). 1. Harvest 200 ml culture by centrifugation and wash twice in 10 ml of Tris-buffered saline (TBS), pH 7.5. After the second wash, resuspend cells in 0.5 ml of lysis buffer (10 mM Tris-HCl, pH 8.0, 20% sucrose, 50 mM NaCl, 10 mM EDTA, 10 mg/ml lysozyme) and incubate for 30 min at 37 . A stock of lysis buffer without lysozyme can be prepared in advance and supplemented with lysozyme immediately prior to use. 2. After treatment with lysozome, cell lysis is completed by the addition of 2 ml of immunoprecipitation buffer (50 mM HEPES-KOH, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 0.1% sodium deoxycholate, 0.1% SDS). To minimize protein degradation, phenylmethylsulfonyl fluoride is added to a final concentration of 1 mM and the sample is chilled on ice in preparation for sonication. 3. Sonication is used to fragment randomly the nucleoprotein content of the sample, with the goal being to obtain DNA fragments in the 500 to 1000 bp range. The sonication protocol depends on the type of sonicator and the amount of biomass used to generate the sample and must be determined empirically by electrophoresis of samples taken at various intervals. For 200 ml late log phase anaerobic cultures prepared as described earlier, we found that eight 15 s treatments, with 30 s intervals for chilling, with a Branson S150D sonicator on the 7.5 power setting was sufficient to generate DNA fragments 500–1000 bp in length. 4. Remove cell debris from the sample by centrifugation. 5. Wash 25 ml of Ultralink protein A/G beads (Pierce) three times with TBS prior to use. After the final wash, remove as much of the liquid phase as possible. To immunoprecipitate cross-linked NsrR-3FlagDNA complexes, add an 800 ml aliquot of the sonicated chromatin to the washed protein A/G beads and then 5 ml of monoclonal anti-Flag antibody (Sigma). Mix overnight on a rotating wheel at 4 .
The NsrR Regulon of Escherichia coli
219
6. After immunoprecipitation, the protein A/G beads can be collected and washed in Spin-X columns supplied by Corning Life Sciences (Grainger et al., 2005). We have found a convenient and effective alternative to be harvesting and washing by centrifugation in a plastic microcentrifuge tube, and resuspension of the pellet in the wash buffers. In this way, the beads are washed twice in 1 ml immunoprecipitation buffer, once in 1 ml of immunoprecipitation buffer supplemented with 500 mM NaCl, once with 1 ml wash buffer (10 mM Tris-HCl, pH 8.0, 250 mM LiCl, 1 mM EDTA, 0.5% Nonidet P-40, 0.5% Na deoxycholate), and once with 1 ml Tris-EDTA, pH 7.5. 7. Transfer the beads in TE to a 2 ml screw cap tube, centrifuge, and resuspend in 50 ml of elution buffer (50 mM Tris-HCl, pH 7.5, 10 mM EDTA, 1% SDS) and incubate the sample at 65 for 10 min and then at 42 for 2 h, without agitation. 8. Add 0.8 mg pronase (made up in water or TBS) per milliliter of sample and incubate overnight at 65 . Pronase is added in sufficient water to dilute the sample twofold, so this step is done in 0.5 elution buffer. Centrifuge the sample and transfer the supernatant to a clean tube. 9. Purify the DNA fragments using a Qiagen PCR purification kit according to the manufacturer’s protocol. Elute the DNA from the Qiagen column in 30 ml of water and determine the DNA concentration (a Nanodrop is useful for this measurement). A single immunoprecipitation usually yields about 100 ng of DNA.
6. DNA Labeling, Microarray Hybridization, and Processing The high-density E. coli MG1655 microarrays fabricated and supplied by Oxford Gene Technology have been described previously (Grainger et al., 2005). Briefly, the arrays comprise 21,321 60-base-long oligonucleotides, with an average spacing between them of 160 bp. The probes are not in genome order on the array, which provides a control against position effects. For our analysis, we labeled DNA immunoprecipitated from each culture with Cy5 and Cy3 and compared the two samples by simultaneous hybridization to the array. Note that there is no amplification of the precipitated DNA in this protocol, which avoids possible artifacts because of amplification bias. 1. Before beginning, DNA samples should be in a volume of 20 ml and at a minimum concentration of 5 mg/ml. Mix the DNA with 20 ml of 2.5 random primer (Bioprime kit, Invitrogen) and 0.25 ml of sterile water and denature by heating to 94 for 3 min.
220
Sam Efromovich et al.
2. After denaturation, add 5 ml of dNTP mix (2 mM dATP, 2 mM dGTP, 2 mM dTTP, 0.5 mM dCTP), 3.75 ml of Cy5- or Cy3-labeled dCTP (1 mM; GE Healthcare), and 1 ml of the Klenow fragment of DNA polymerase (Bioprime kit), mix gently, and incubate at 37 for 2 h. 3. Add a further 0.5 ml of DNA polymerase to each tube and incubate at 37 for a further 2 h. 4. Purify DNA fragments from the labeling reaction using QIAquick PCR purification columns (Qiagen), according to the manufacturer’s instructions. The labeled DNA should be eluted from the column using 50 ml of Qiagen elution buffer. 5. The labeled DNA samples can now be mixed and hybridized to the array according to the manufacturer’s instructions (freely available at www. ogt.co.uk). Oxford Gene Technology arrays are manufactured on an Agilent platform, and hybridizations are best performed in an Agilent SureHyb apparatus.
7. Visualization and Analysis of DNA Microarray Data After scanning the microarray, a list of Cy5 and Cy3 signal intensities is generated. The fluorescence intensity ratio is then calculated for each probe on the array and plotted against the genomic coordinate of the probe, generating a genome-wide DNA-binding profile for NsrR. This analysis can be done in a Microsoft Excel spreadsheet, and binding sites are localized approximately by comparison of peak heights with a suitable E. coli database. Alternatively, the profile can be scrutinized using a ‘‘genome browser,’’ in which the fluorescence ratios are superimposed on the genetic map of E. coli. Oxford Gene Technology’s ChIP Browser software (www.ogt.co.uk) facilitates this type of analysis. Our ChIP-on-chip data for NsrR binding to the ytfE promoter from three repeat experiments are shown as an example in Fig. 12.2A. In typical ChIP-on-chip data, several genetically adjacent probes register a fluorescence ratio that is significantly above the background (see Fig. 12.2A). This feature in data is usually referred to as a ‘‘peak’’ (although, strictly speaking, the peak is the single probe with the highest fluorescence ratio). In what follows, we use ‘‘peak’’ in the conventional manner to refer to a series of probes with an above-background fluorescence ratio. The peak should be centered over the genome location of the protein-binding site, and its width and shape depend on the size of DNA fragments after sonication and labeling (Buck et al., 2005). In many cases, these peaks can simply be identified by visual inspection of data. If required, a cutoff (which is necessarily arbitrary) in the fluorescence ratio can be
221
The NsrR Regulon of Escherichia coli
A 24 ytfE
22
ytfF
20 Fluorescence ratio
18 16 14 12 10 8 6 4 2 0 4428500 B
4429000
4429500 Coordinate (bp)
4430000
4430500
6 5
ygbA
mutS
Fluorescence ratio
4 3 2 1 0 −1 −2 2854000
2854500
2855000
2855500
2856000
Coordinate (bp)
Figure 12.2 Strong and weak signals in raw ChIP-on-chip data. (A) ChIP-on-chip data for a 2-kb window around the NsrR-regulated ytfE promoter (Bodenmiller and Spiro, 2006). The fluorescence ratios are from immunoprecipitated cross-linked DNA extracted from cells grown in the absence and presence of nitrate. Raw data from three experiments are shown. Each individual dataset was centered on a ratio of 1 (by subtraction of the mean fluorescence ratio for that dataset þ 1); this is a necessary prerequisite to subsequent analysis with ChIPOTle (Buck et al., 2005). The probe giving the highest fluorescence ratio is at coordinates 4429599^4429659; the ytfE start codon is at 4429561, and the predicted NsrR-binding site (Bodenmiller and Spiro, 2006) is centered on 4429598.5. The exact coincidence of the peak maximum with the NsrR-binding site in this particular case is a consequence of the fortuitous fact that the probe on the array overlaps the binding site. More typically, peaks in raw data may be up to 100 bp away
222
Sam Efromovich et al.
applied, and peaks assigned if two or more adjacent probes exceed the cutoff. This can be a laborious process in large data sets with many protein-binding sites, and some data sets are more challenging, especially if there are peaks that represent real binding sites, but do not clearly rise above the background noise. This issue can be illustrated by our ChIP-onchip data for the ygbA promoter (see Fig. 12.2B). This promoter is known to be regulated by NsrR and contains an NsrR-binding site (Bodenmiller and Spiro, 2006), but gives a weak signal in data that may, or may not, be discarded if judged by purely subjective criteria (see Fig. 12.2B). The identification of peaks can be addressed computationally, for example, by using chromatin immunoprecipitation on tiled arrays (ChIPOTle), a freely available and easily applied Microsoft Excel macro (Buck et al., 2005). ChIPOTle analyzes ChIP-on-chip data using a sliding window and reports the locations of statistically significant peaks. The user sets parameters for the window and step sizes and for the P value cutoff. This approach allows rapid identification of all significant peaks in a large data set and assigns a quantitative confidence measure (a P value) to each peak (Buck et al., 2005). This valuable approach allows rapid identification of peaks in large data sets. However, some real binding sites may fail to exceed the statistical cutoff and therefore fail to be reported. Relaxation of search criteria may help resolve this problem, but will also increase the rate of false positives. In the case of ygbA, ChIPOTle finds a significant peak (P ¼ 0.0001) in two of the three data sets. The ygbA promoter would not be reported as positive if an arbitrary cutoff were employed of two or more adjacent probes showing greater than twofold enrichment in at least two data sets (see Fig. 12.2B).
8. A New Statistical Methodology for Treatment of ChIP-on-Chip Data In ChIP-on-chip experiments, each probe on the array measures the abundance of a population of DNA fragments that differ in length as a consequence of sonication and labeling by random priming. As a result, from the binding site. Note that the coordinates assigned to probes by Oxford GeneTechnology refer to a non-current annotation of the E. coli genome (http://genolist.pasteur. fr/Colibri/). A gray line is plotted at the two-fold enrichment level. (B) ChIP-on-chip data for a 2-kb window around the NsrR-regulated ygbA promoter (Bodenmiller and Spiro, 2006). Data are averaged from the same three experiments, shown in a, and were manipulated in the same way.The peak in the noncoding region between ygbA and mutS is statistically significant (P ¼ 0.0001) in two datasets, when data are analyzed with ChIPOTle (Buck et al., 2005). Note that this result would not be reported as a positive if, as the cutoff, two or more adjacent probes are required to show greater than twofold enrichment in the immunoprecipitated population in two of the three experiments.
The NsrR Regulon of Escherichia coli
223
several probes close to the protein-binding site will give a relatively large positive signal. Depending on the number and location of probes near a binding site, and the distribution of labeled fragment lengths, a specific profile of intensities of signals for genetically adjacent probes is created. In general (and always for equally spaced probes) there should also be a maximum enrichment for the probe closest to the binding site. A ‘‘peak’’ in data is therefore characterized by a relatively large signal for several adjacent probes (see Fig. 12.2A). The second feature of ChIP-on-chip data is that the measurements are a mixture of two distributions, with the larger one being a background noise and the smaller being of larger fluorescence ratios. These larger ratios cannot be explained by background distribution and thus are positive outliers from background distribution. In a traditional transcriptomics experiment, the distribution of signals is two tailed and mostly symmetric; both low- and high-intensity signals are potentially of interest. In ChIP-on-chip data, only the larger signals are of interest because they correspond to the population of genomic fragments specifically enriched by the ChIP. Because of such an asymmetric distribution of signals with an extremely heavy right tail, it is often recommended to analyze a logarithmically transformed signal. Typically a log with base 2 is used and applied to the ratio of experimental to reference signals (Buck et al., 2005). The main statistical issue frequently considered in ChIP-on-chip literature is how to find peaks in (log-transformed) fluorescence ratios. To this end, several different methods and software have been published. In the aforementioned ChIPOTle, moving windows are used to smooth data and then to find peaks as local points of maximum (Buck et al., 2005). Then, an algorithm calculates P values for proposed peaks based on the standard normal error function; a conservative Bonferroni correction may complement the procedure. Another approach is to use a percentile rank analysis (Bieda et al., 2006). For a more precise estimation of binding-site locations, a ‘‘triangle’’ peak-finding model has been proposed (Kim et al., 2005). Given the 100 bp resolution of the array, the authors (with some assumptions) derived a model that states that the log ratio should decrease linearly with distance from the true binding site. Then model fitting is done by traditional linear regression. This method of peak finding is interesting because it seeks to estimate the binding-site locations underlying each peak. However, the method has been criticized on the basis that profiles of log ratios near a peak varied greatly in their waveform, amplitude, and size, thus making the rigid triangle profile not feasible (Bieda et al., 2006). Instead, an approach for peak finding that makes minimal assumptions about the shape and amplitude of the log-ratios profile is recommended (Bieda et al., 2006). To this end, the authors suggested the use of a set of percentile thresholds (specifically the 95th and 98th percentiles), and then analysis of a peak is conducted via a
224
Sam Efromovich et al.
pattern of indicator functions that a log ratio exceeds the threshold (Bieda et al., 2006). The underlying probability theory of such an approach is based on the theory of random runs. Here, we propose a novel approach to the estimation of binding site locations from ChIP-on-chip data, and the following considerations underlie the proposed methodology. An important feature of OGT microarrays is the relatively large (on average 160 bp) and highly irregular distances between the probes. The probe that gives the highest fluorescence ratio may thus be 100 bp (or, in some cases, considerably more) distant from the protein binding site. We have developed a statistical approach to estimate the locations of protein binding sites from ChIP-on-chip data obtained from OGT (and similar) microarrays, where probes are relatively sparse and irregularly spaced. To locate a binding site, a probabilistic model for prediction of the shape of fluorescence ratio data between experimental and reference signals for a particular configuration of neighboring probes is developed. This model assumes (i) a particular probability distribution of DNA fragment lengths after sonication and labeling by random priming; (ii) that for a given immunoprecipitated DNA fragment, the probability of a binding site occurring anywhere within that fragment is equal and independent of the fragment length; (iii) DNA fragments will hybridize to probes if the two overlap for at least half the length of the probe (30 bp); (iv) a given DNA fragment has equal chances of hybridizing to those probes that it overlaps; (v) the probability that a DNA fragment will hybridize to a particular probe is the same for reference and experimental samples, but may differ from probe to probe; and (vi) each hybridized DNA fragment adds a fluorescence signal intensity that is proportional to its length, and the coefficient of proportionality is the same for all fragments and probes. The model predicts a profile of fluorescence intensity ratios for probes surrounding a conjectured binding site. The conjectured site is then moved along the chromosome in the region of a large positive fluorescence signal, and for each conjectured site the statistical performance of the model is evaluated via empirical standard deviations of residuals (differences between observed signals and those predicted by the model). Because the standard deviation of residuals depends on the location of the binding site, the estimate of the true site location is that which minimizes the standard deviation of residuals. If several experiments are conducted, then binding sites are calculated for each experiment and then reported together with corresponding standard deviations; the reasoning behind this approach is explained later. The method of statistical visualization and analysis we have developed consists of three major steps. Step 1 is a preliminary step, which allows one to visualize the experimental and reference signals and their ratio for each experiment (in our case 3).
225
The NsrR Regulon of Escherichia coli
The procedure is illustrated for a 65.5 kb window around the hcp gene in Fig. 12.3 The hcp promoter contains an NsrR binding site (Rodionov et al., 2005), and we have confirmed that NsrR is a repressor of hcp transcription (Filenko et al., 2007). A pronounced peak is visible in denoised fluorescence data for the reference sample (see Fig. 12.3B). All other potential peaks in this region result from noise and thus the procedure shrinks them to zero. The procedure used is based on data-driven multiwavelet Efromovich block-shrinkage filtering (Efromovich, 2001; Efromovich et al., 2004). Raw and denoised data for the experimental sample are shown in Figs. 12.3C and 12.3D. The denoised experimental signal exhibits two peaks: one corresponds to the one observed in the denoised reference signal (which is therefore a candidate for being technical artifact) and the second peak localizes to the hcp promoter region. In denoised fluorescence ratio data (see Fig. 12.3F ) the only prominent signal results from NsrR binding to
B 5 0 −5
A
Observed reference signal
⫻ 104
Denoised reference signal
8.9
D 5 0 −5
9.0
9.1
9.2
9.3
9.4
C
Observed experimental signal
⫻ 104
Denoised experimental signal
8.9
9.0 E
9.1
9.2
9.3
9.4
9.5
9.6
9.7 ⫻ 105
9.5
9.6
9.7 ⫻ 105
9.6
9.7 ⫻ 105
Observed ratio of experimental/reference signals
F
Denoised ratio of experimental/reference signals
10 5 0
8.9
9.0
9.1
9.2 9.3 9.4 Chromosome location
9.5
Figure 12.3 Observed and denoised signals for the first of three ChIP-on-chip experiments. A 65.5-kb region of the E. coli chromosome is shown. Raw (A, C, and E) and denoised (B, D, and F) fluorescence data are shown for reference (A and B) and experimental samples (C and D), and for the ratios (E and F ).The peak at 9.14 105 bp in fluorescence ratio data is in the promoter region of the hcp gene.
226
Sam Efromovich et al.
a site in the vicinity of the hcp promoter. The two weaker signals in the fluorescence ratios should be explored with extra statistical precaution, and by visual inspection, because they are candidates for false positives, or may represent true binding sites. The developed method of a multiwavelet signal denoising allows one to visualize peaks in the ratio signal and to identify possible false positives. Thus, in the first step of the process, it is useful to examine reference and experimental fluorescence data separately. Step 2 is devoted to another preliminary analysis of data, which allows one to obtain a preliminary estimate of the possible number of binding sites, as well as to check the assumption that the observed background ratios (because of pure noise or technical artifacts) have a normal distribution. This is done by visualization of the histogram of z-scored ratios (Fig. 12.4). Obtaining z-scored ratios for a particular experiment is based on the following procedure. First, a sample median is calculated. Then a robust standard deviation estimator is calculated as 1.48 times the median of absolute deviations from the median in a procedure called mad (Efromovich, 1999). Finally, a classical z scoring is performed. Histograms in the left panels of Fig. 12.4 show z-scored ratios that do not exceed a threshold T ¼ 4.89. This threshold is chosen to satisfy the relation P(max(Z1,. . .,Zn) >T) <0.01 (here n ¼ 21321, the number of probes in the microarray) and depends on both the sample size n and the chosen familywise error rate (here 1%). The underlying idea of such a thresholding is similar to the universal thresholding procedure popular in the wavelet literature (Efromovich, 1999). Then histograms are overlaid by solid lines, which are standard normal curves, showing that the observed ratios fit closely to a standard normal distribution. Histograms shown in the right panels of Fig. 12.4 are based on ratios that exceed the threshold and are not explained by a standard normal distribution; they are outliers (or peaks following the traditional terminology in the ChIP-on-chip literature) under the null hypothesis that the distribution of ratios is standard normal. Chromosomal locations corresponding to these peaks are the primary candidates for protein-binding sites. It is important to count the number of such outliers: we have 116, 128, and 128 outliers for experiments 1–3, respectively. Two important conclusions come out of this analysis. First, each peak in the data can be visualized and statistically analyzed. Second, the z-scoring procedure used (based on the median-mad z scoring) is justified. Otherwise, if the number of outliers was much larger (for a more promiscuous DNAbinding protein), then more specialized robust methods of z scoring could be used (such as a positively truncated mean/median or an L estimator). It is also of interest to know the number of ratios that are larger than the threshold T’ ¼ 3, often mentioned in the microarray literature. Here, for the three experiments the numbers are 136, 177, and 198, respectively.
227
The NsrR Regulon of Escherichia coli
A
Experiment 1
B
0.4
0.12
Density
0.3 Density
Experiment 1
0.2 0.1
0.08 0.04 0.00
0.0 −2 0 2 4 Number of ratios 21205 C
Experiment 2
0
20
D
0.4
40 60 80 100 Number of ratios 116
120
Experiment 2 0.12
Density
Density
0.3 0.2 0.1
0.08 0.04 0.00
0.0 0 2 4 −2 Number of ratios 21193 E
Experiment 3
20 40 60 80 Number of ratios 128 F 0.25
0.4
0.20 Density
0.3 Density
Experiment 3
0.2 0.1
0.15 0.10 0.05
0.0
0.00 −4
−2 0 2 4 Number of ratios 21193
10
20 30 40 50 60 Number of ratios 128
70
Figure 12.4 Histograms of z-scored ratios for the three experiments. (A, C, and E) Histograms based on ratios lower than the thresholdT ¼ 4.89. Histograms are overlaid by the standard normal density (the solid line). (B, D, and F) Histograms of ratios larger than the threshold. The number of ratios below or above the threshold is shown in each case (the total number of probes in the microarray is n ¼ 21,321).
As shown, there are not too many ratios in the middle region between threshold T ’ ¼ 3 and T ¼ 4.89. Of course, it may be of interest to look at smaller ratios as well, but in this case very rigorous statistical procedures
228
Sam Efromovich et al.
should be used to keep the family-wise error rate (the number of false positives) under control. To do this, a w2 test is used, and then a Holm procedure (in place of a traditionally recommended in the microarray literature Bonferroni procedure) is implemented. Step 3 deals with estimating the location of the protein-binding site for a particular peak. This is done separately for each experiment (the underlying motivation of this approach is explained later). To illustrate this step, we consider data for a newly discovered NsrR binding site in the promoter of the nrf operon that encodes a respiratory nitrite reductase (Fig. 12.5). NsrR regulation of nrf expression has been confirmed by microarray analysis and with a nrf-lacZ reporter fusion (Filenko et al., 2007). Bioinformatic analysis suggests the presence of an NsrR binding site in the nrf promoter at coordinate 4285199. Z-scored ratios for all three experiments are shown in Fig. 12.5A. Here the horizontal solid lines (of length 60 bp) exhibit the physical location of probes in the chromosome. To assess the level of exhibited ratios, the height of these horizontal lines is equal to the threshold T (recall that any z-scored ratio of experimental to reference signals beyond this threshold is highly unlikely under the null hypothesis that the distribution of these ratios is standard normal). As a result, in each experiment, at least three z-scored ratios cannot be explained by standard normal distribution (see Fig. 12.5A). An interesting feature of this location (and the peak profile) is the highly irregular placement of the probes. There is a large gap between the two probes with the highest ratios. Another interesting feature is that the two highest ratios are about the same for the second experiment. Then the mathematical model for a prediction of shapes of (z-scored) ratios is used. The model depends on a conjectured location of the binding site (this is its input) and calculates fitted peak profiles and corresponding standard deviations of errors for a set of potential locations of the binding site. For each potential binding location, the obtained empirical standard deviations are shown in Fig. 12.5B (numbered 1, 2, and 3, corresponding to the three experiments). A larger set of potential binding locations is shown for purely illustrative purposes. The true binding site location should provide the best fit to the shape of observed peak profiles (and the lowest standard deviations), and these locations (for each experiment) are shown in the subtitle as BN for the Nth experiment. The best estimate (B1) places the NsrR binding site <10 bp away from the site predicted by bioinformatics. The bottom panel of Fig. 12.6 exhibits how well the model fits the observed ratios. The minimal empirical standard deviations (which correspond to the estimated binding locations shown in the subtitle of the middle panel) are also shown. These characteristics of the goodness of fit can be used in assigning weights to the estimated binding sites found for the different experiments. A similar analysis for the NsrR-binding site in the hmp promoter is shown in Fig. 12.6. Here the mean estimate of the binding site from
229
The NsrR Regulon of Escherichia coli
Ratios for 3 experiments
A
2 1
Ratio
30 20
21 3
10 0
1 32 32 1
2 1 3
2 13 213
32 1
2683000
2683500
2684000 Chromosome location
23 1
2684500
Search for a binding site
B 8 6 Std
3
4 2
11 22 22 21 21 11 12 21 21 12 21 21 1 1 2 22 112 2 1122 22 11 2 111 22 11222 1 11 2 11 22 11 22 11 2 11 22 3 33 3 3333 11 22 33333 11 222 11 22 1 33333 2 1 3 1 3 2 1 11 22 33 3 33 2 3 33 11 11 22 3333 2 2 3 3 3 33 33 1 1 1 1 333 333 2 2 33 22 3 3 333 33 3 3 1 1 1 3 3 31 31 11 2 2 2 2 2 2 2 3 3 3 33 33 2 3 3 1 1 333 333 33 2 2 2 2222 333 1 1 22 2 2 1 11 11 11 1111 11 222
222
2683700
2683750 2683800 2683850 2683900 2683950 2684000 Bknown = 2683818, B1 = 2683808, B2 = 2683838, B3 = 2683789 Observed and fitted ratios
C
43
12
Ratio
30 20
123 54 6
10 0
1462 53 6543 12
2683000
65
4 3 12 65
3 4 125 6 41653 2
45 36
12
2683500 2684000 Std1 = 0.72, Std2 = 1.74, Std3 = 1.89
43 6 5 12
2684500
Figure 12.5 Performance of the adaptive estimator of a binding site location for a particular peak. (A) Fluorescence ratios are shown for three experiments at chromosome locations strictly above the middle of a probe. Data points are numbered to correspond to the three experiments. Locations of 60mer probes are shown as solid horizontal lines at a height equal to the threshold T ¼ 4.89. (B) Empirical standard deviations for model fitting when a notional binding point is the x coordinate of a shown point. Estimated binding sites for the three experiments (B1, B2, and B3) are shown, along with the location predicted bybioinformatics (Bknown). (C) Empirical ratios and those predicted by the model. Here, numbers 1 and 2 indicate empirical and predicted ratios for the first experiment 3 and 4 for the second experiment, and 5 and 6 for the third experiment.
230
Sam Efromovich et al.
Ratios for 3 experiments
A 15
3
Ratio
3
5 0
3 2
3 1
1
2 2 3 1
213
1
2 1
4284500
31 2
21 3
4285000 4285500 Chromosome location
4286000
Search for a binding site
B 5 4 Std
32
2 1
10
3 2 1
3333 333 33333 33 333 22 2 2 2 23 3 3 3 3333 3 3 3 2 2 2 23 3 333 2 23 23 33 23 23 2 22 33 3 23 23 2 2 2 2 2 11 11 1112 21111 3 3 23 2 33 1 11 1 112 2 2 2 2 3 23 2 33 111 11111 22 2 3 3 1 1 2 11 3 3322 22 111 11 222 3322 33 111 11 222 3 111 11 332 22 1 11 2 2 2 332 111 331 22 11111 311 322 31 22 111 332 3 2333 13 13 1 11 22 11 2 21 1 111 22 1 11 1 1 1 2 2 2 222 2
4285000
4285100
4285200
4285300
4285400
4285500
Bknown = 4285199, B1 = 4285205, B2 = 4285274, B3 = 4285236 Observed and fitted ratios
Ratio
C 15
6 5
10
4321
5 0
5
462 351
3 46 52 1
4284500
63 421
5 6 1 42 3
465 3
21
5 3 64 2 1 641 32 5
4285000
4285500
561 432
4286000
Std1 = 0.87, Std2 = 0.69, Std3 = 1.43
Figure 12.6 NsrR binding site in the hmp promoter. All details are as for Fig.12.5.
ChIP-on-chip data (2683811) is <20 bp distant from the site predicted by sequence examination (Bodenmiller and Spiro, 2006). Thus, we believe that these statistical procedures will be useful for estimating the locations of protein-binding sites in promoters where sites cannot be predicted by sequence examination and for which no other information is available.
The NsrR Regulon of Escherichia coli
231
9. Conclusions ChIP-on-chip is a very useful addition to the arsenal of tools that can be used to identify the genes that are potentially regulated by a particular protein, such as NsrR. However, this approach apparently does not give a definitive readout of regulon members, in particular because of false negatives (Grainger et al., 2006) and signals that may be present in data but discarded as statistically insignificant (see Fig. 12.2). For NsrR and other regulators, there are also binding sites within coding regions and in the intergenic regions between convergently transcribed genes. There is currently no reason to believe that such sites have regulatory function (except perhaps sites close to the 30 ends of genes, which may control expression of the downstream gene). Interpretation of ChIP-on-chip data is probably best done by a combination of computational and manual approaches. The relatively small size of microbial genomes makes visual inspection of data a manageable proposition. Our suggested new statistical methodology based on wavelet denoising and probabilistic modeling of peak profiles allows relatively accurate predictions of binding site locations to be made. As a tool to enumerate regulon members, ChIP-on-chip is best employed alongside other highthroughput methods, one-by-one analysis of individual promoters, and bioinformatics (Molle et al., 2003).
ACKNOWLEDGMENTS We are grateful to Mike Humphrys for doing the array hybridizations and data collection and for his helpful advice. We thank Zoltan Szarka, Douglas Hurd, and their colleagues at Oxford Gene Technology for advice concerning the use of high-density microarrays. This work was supported by Award MCB-0702858 from the National Science Foundation (to SS). SE was supported by Awards DMS-0243606 and DMS-0604558 from the National Science Foundation and by Award MSPF-06G-014 from the National Security Agency. DCG was supported by a Wellcome Trust program grant awarded to SJ Busby.
REFERENCES Bang, I.-S., Liu, L., Vazquez-Torres, A., Crouch, M.-L., Stamler, J. S., and Fang, F. C. (2006). Maintenance of nitric oxide and redox homeostasis by the Salmonella flavohemoglobin Hmp. J. Biol. Chem. 281, 28039–28047. Bieda, M., Xu, X., Singer, M. A., Green, R., and Farnham, P. J. (2006). Unbiased location analysis of E2F1-binding sites suggests a widespread role for E2F1 in the human genome. Genome Res. 16, 595–605. Bodenmiller, D. M., and Spiro, S. (2006). The yjeB (nsrR) gene of Escherichia coli encodes a nitric oxide sensitive transcriptional regulator. J. Bacteriol. 188, 874–881.
232
Sam Efromovich et al.
Buck, M. J., and Lieb, J. D. (2004). ChIP-chip: Considerations for the design, analysis, and application of genome-wide chromatin immunoprecipitation experiments. Genomics 83, 349–360. Buck, M. J., Nobel, A. B., and Lieb, J. D. (2005). ChIPOTle: A user-friendly tool for the analysis of ChIP-chip data. Genome Biol. 6, R97. Cairrao, F., Cruz, A., Mori, H., and Arraiano, C. M. (2003). Cold shock induction of RNase R and its role in the maturation of the quality control mediator SsrA/tmRNA. Mol. Microbiol. 50, 1349–1360. Corker, H., and Poole, R. K. (2003). Nitric oxide formation by Escherichia coli: Dependence on nitrite reductase, the NO-sensing regulator Fnr, and flavohemoglobin Hmp. J. Biol. Chem. 278, 31584–31592. Datsenko, K. A., and Wanner, B. L. (2000). One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc. Natl. Acad. Sci. USA 97, 6640–6645. D’Autre´aux, B., Tucker, N. P., Dixon, R., and Spiro, S. (2005). A non-haem iron centre in the transcription factor NorR senses nitric oxide. Nature 437, 769–772. Efromovich, S. (1999). ‘‘Nonparametric Curve Estimation: Methods, Theory and Applications.’’ Springer, New York. Efromovich, S. (2001). Multiwavelets and signal denoising. Sankhya 63, 367–393. Efromovich, S., Lakey, J., Pereyra, M. C., and Tymes, N. (2004). Data-driven and optimal denoising of a signal and recovery of its derivatives using multiwavelets. IEEE Trans. Signal Process. 52, 628–635. Filenko, N., Spiro, S., Browning, D. F., Squire, D., Overton, T. W., Cole, J., and Constantinidou, C. (2007). The NsrR regular of Escherichia coli K12 includes genes encoding the hybrid cluster protein and the periplasmic, respiratory nitrite reductase. J. Bacteriol. 189, 4410–4417. Gardner, A. M., Gessner, C. R., and Gardner, P. R. (2003). Regulation of the nitric oxide reduction operon (norRVW) in Escherichia coli: Role of NorR and s54 in the nitric oxide stress response. J. Biol. Chem. 278, 10081–10086. Gardner, A. M., Helmick, R. A., and Gardner, P. R. (2002). Flavorubredoxin, an inducible catalyst for nitric oxide reduction and detoxification in Escherichia coli. J. Biol. Chem. 277, 8172–8177. Gomes, C. M., Giuffre, A., Forte, E., Vicente, J. B., Saraiva, L. M., Brunori, M., and Teixeira, M. (2002). A novel type of nitric-oxide reductase. Escherichia coli flavorubredoxin. J. Biol. Chem. 277, 25273–25276. Grainger, D. C., Aiba, H., Hurd, D., Browning, D. F., and Busby, S. J. W. (2006). Transcription factor distribution in Escherichia coli: Studies with FNR protein. Nucleic Acids Res. 35, 269–278. Grainger, D. C., Hurd, D., Harrison, M., Holdstock, J., and Busby, S. J. W. (2005). Studies of the distribution of Escherichia coli cAMP-receptor protein and RNA polymerase along the E. coli chromosome. Proc. Natl. Acad. Sci. USA 102, 17693–17698. Grainger, D. C., Overton, T. W., Reppas, N., Wade, J. T., Tamai, E., Hobman, J. L., Constantinidou, C., Struhl, K., Church, G., and Busby, S. J. W. (2004). Genomic studies with Escherichia coli MelR protein: Applications of chromatin immunoprecipitation and microarrays. J. Bacteriol. 186, 6938–6943. Ji, X. B., and Hollocher, T. C. (1988). Reduction of nitrite to nitric oxide by enteric bacteria. Biochem. Biophys. Res. Commun. 157, 106–108. Justino, M. C., Almeida, C. C., Goncalves, V. L., Teixeira, M., and Saraiva, L. M. (2006). Escherichia coli YtfE is a di-iron protein with an important function in assembly of iron-sulphur clusters. FEMS Microbiol. Lett. 257, 278–284. Justino, M. C., Vicente, J. B., Teixeira, M., and Saraiva, L. M. (2005). New genes implicated in the protection of anaerobically grown Escherichia coli against nitric oxide. J. Biol. Chem. 280, 2636–2643.
The NsrR Regulon of Escherichia coli
233
Kim, T. H., Barrera, L. O., Zheng, M., Qu, C., Singer, M. A., Richmond, T. A., Wu, Y., Green, R. D., and Ren, B. (2005). A high-resolution map of active promoters in the human genome. Nature 436, 876–880. Molle, V., Fujita, M., Jensen, S. T., Eichenberger, P., Gonzalez-Pastor, J. E., Liu, J. S., and Losick, R. (2003). The Spo0A regulon of Bacillus subtilis. Mol. Microbiol. 50, 1683–1701. Poock, S. R., Leach, E. R., Moir, J. W., Cole, J. A., and Richardson, D. J. (2002). Respiratory detoxification of nitric oxide by the cytochrome c nitrite reductase of Escherichia coli. J. Biol. Chem. 277, 23664–23669. Poole, R. K. (2005). Nitric oxide and nitrosative stress tolerance in bacteria. Biochem. Soc. Trans. 33, 176–180. Poole, R. K., and Hughes, M. N. (2000). New functions for the ancient globin family: Bacterial responses to nitric oxide and nitrosative stress. Mol. Microbiol. 36, 775–783. Rodionov, D. A., Dubchak, I. L., Arkin, A. P., Alm, E. J., and Gelfand, M. S. (2005). Dissimilatory metabolim of nitrogen oxides in Bacteria: Comparative reconstruction of transcriptional networks. PLOS Comput. Biol. 1e55. Rokop, M. E., Auchtung, J. M., and Grossman, A. D. (2004). Control of DNA replication initiation by recruitment of an essential initiation protein to the membrane of Bacillus subtilis. Mol. Microbiol. 52, 1757–1767. Sebbane, F., Lemaitre, N., Sturdevant, D. E., Rebeil, R., Virtaneva, K., Porcella, S. F., and Hinnebusch, B. J. (2006). Adaptive response of Yersinia pestis to extracellular effectors of innate immunity during bubonic plague. Proc. Natl. Acad. Sci. USA 103, 11766–11771. Shin, D., and Groisman, E. A. (2005). Signal-dependent binding of the response regulators PhoP and PmrA to their target promoters in vivo. J. Biol. Chem. 280, 4089–4094. Spiro, S. (2007). Regulators of bacterial responses to nitric oxide. FEMS Microbiol. Rev. 31, 193–211. Stevanin, T. M., Poole, R. K., Demoncheaux, E. A., and Read, R. C. (2002). Flavohemoglobin Hmp protects Salmonella enterica serovar typhimurium from nitric oxide-related killing by human macrophages. Infect. Immun. 70, 4399–4405. Uzzau, S., Figueroa-Bossi, N., Rubino, S., and Bossi, L. (2001). Epitope tagging of chromosomal genes in Salmonella. Proc. Natl. Acad. Sci. USA 98, 15264–15269. Van Doorslaer, S., Dewilde, S., Kiger, L., Nistor, S. V., Goovaerts, E., Marden, M. C., and Moens, L. (2003). Nitric oxide binding properties of neuroglobin: A characterization by EPR and flash photolysis. J. Biol. Chem. 278, 4919–4925.
This page intentionally left blank
C H A P T E R
T H I R T E E N
Characterization of the Nitric Oxide-Reactive Transcriptional Activator NorR Benoıˆt D’Autre´aux,* Nick Tucker,† Stephen Spiro,‡ and Ray Dixon† Contents 1. 2. 3. 4.
Introduction Measurement of NorR Activity In Vivo Measurement of Transcriptional Activation by NorR In Vitro Detection of the Ferrous-Nitrosyl Form of NorR by In Vivo Electron Paramagnetic Resonance (EPR) 5. In Vitro Reconstitution of the Iron Center in NorR 6. Measurement of NO Affinity 7. Standardization of the NO Electrode 8. Determination of NorRFe(NO) Kd 9. Conclusions Acknowledgment References
236 237 238 240 242 243 246 247 248 248 249
Abstract The prokaryotic transcriptional regulator NorR is unusual in that it utilizes a mononuclear ferrous iron center rather than a heme moiety as a means of sensing nitric oxide (NO). Binding of NO to the nonheme iron center in the amino-terminal GAF domain of NorR results in formation of a mononitrosyl iron complex and relieves intramolecular repression within NorR, allowing this regulatory protein, a member of the s54-dependent family of enhancer-binding proteins, to activate expression of genes required for NO detoxification. This chapter describes detailed protocols for measuring transcriptional activation by Escherichia coli NorR in vivo and in vitro. It also details spectroscopic methods for analysis of the interaction of NO with the nonheme iron center and determination of the NO-binding affinity constant.
* { {
Laboratoire Stress Oxydant et Cancer, Service de Biologie Inte´grative et Ge´ne´tique Mole´culaire, Institut de Biologie et de Technologies de Saclay, CEA-Saclay, Gif-sur-Yvette Cedex, France; John Innes Centre, Colney, Norwich, United Kingdom; Department of Molecular and Cell Biology, University of Texas at Dallas, Richardson, Texas
Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37013-4
#
2008 Elsevier Inc. All rights reserved.
235
236
Benoıˆt D’Autre´aux et al.
1. Introduction Several bacterial transcriptional regulators that sense nitric oxide (NO) and control the expression of genes encoding enzymes that detoxify NO have been characterized (Spiro, 2007). Among these is the transcriptional activator NorR, which, in response to NO, switches on the expression of genes encoding a respiratory NO reductase in Ralstonia eutropha, an NO-reducing flavorubredoxin in Escherichia coli and a flavohemoglobin in Pseudomonas aeruginosa (Arai et al., 2005; Gardner et al., 2003; Hutchings et al., 2002; Pohlmann et al., 2000). The norR gene is typically adjacent to its regulatory target(s), which can be transcribed either in the same direction or divergently (Rodionov et al., 2005). Upstream DNA sequence elements that act as enhancer-binding sites for transcriptional activation by NorR have been identified, and two or more copies of these sequences are found in the promoter regions of NorR-regulated genes (Bu¨sch et al., 2004; Rodionov et al., 2005; Tucker et al., 2004). Because the norR gene itself is subject to negative autoregulation, NorR-binding sites are frequently associated with the norR promoter region (Rodionov et al., 2005; Tucker et al., 2004). As far as is currently known, in E. coli the NorR regulon is restricted to norVW genes and the norR gene itself. Mechanistic data for NO signaling mechanisms in bacteria are currently rather limited. The few mechanisms that have been described all involve metal–NO complexes (Cruz-Ramos et al., 2002; D’Autre´aux et al., 2002, 2005; Ding and Demple, 2000; Zhao et al., 1999). Direct modification of cysteine and tyrosine residues by the oxidation products of NO or by nitrosating agents provides a distinct type of signaling mechanism (Hausladen et al., 1998) that is not considered here. The reaction of NO with a metal ion center is nondestructive for NO, allowing reversibility, as the interaction is noncovalent. NO–metal complexes are usually unstable (potentially redox active and easily dissociated when NO is removed), which presents challenges for structural characterization using conventional biochemical techniques. Furthermore, metal ion centers in proteins can themselves be labile, adding further complications for in vitro characterization. In the case of NorR, the regulatory N-terminal GAF domain contains a mononuclear nonheme iron center, which is the binding site for NO (D’Autre´aux et al., 2005). NO binding results in the formation of a ferrous mononitrosyl complex, which is a necessary prerequisite for the productive interaction of NorR with the transcriptional apparatus (D’Autre´aux et al., 2005). Current evidence indicates that, in the resting state of NorR, the GAF domain inhibits the catalytic activity of the central AAAþ domain of the protein by intramolecular repression (D’Autre´aux et al., 2005; Gardner et al., 2003; Pohlmann et al., 2000). Binding of NO to the iron center within the sensory GAF domain activates the ATPase activity of the AAAþ
Analysis of the Nonheme Iron NO Sensor, NorR
237
domain, enabling interaction with the s54 subunit of RNA polymerase to drive the formation of open promoter complexes, and thus enable transcription initiation from the target promoter (D’Autre´aux et al., 2005). A combination of site-directed mutagenesis, spectroscopic techniques, and structural modeling is beginning to provide an insight into the identities of the protein ligands to the iron center of NorR from R. eutropha (Klink et al., 2007) and E. coli (Tucker et al., submitted for publication). This chapter provides protocols for methods used to study NorR and its mutant derivatives in vivo and in vitro. While these methods were developed for E. coli NorR, they may be applicable, with some modifications, to other organisms, especially those for which a genetic system is available.
2. Measurement of NorR Activity In Vivo In order to measure NorR activity in vivo, we have constructed a chromosomal lacZ transcriptional fusion to the norV promoter. The fusion is on a l phage derivative that can be integrated into the l attachment site of any E. coli strain that is l sensitive. Because the phage also has a kanamycin resistance gene, it can also be moved between strains by P1 transduction. To assay the effects of site-directed mutations that result in amino acid substitutions in NorR, we used a strain in which the chromosomal norR gene is deleted and then introduced mutant norR alleles on a plasmid derived from pET21a. In strains devoid of T7 RNA polymerase, there is a low-level expression of genes cloned in pET vectors (Moore and Kiley, 2001). Expression of norR in this system is evidently at a higher level than in a wild-type strain, as we recorded higher levels of norV-lacZ expression when norR is expressed from a pET vector, even in the absence of T7 RNA polymerase. Evidently, NorR abundance limits norV transcription in a wild-type strain, which may in part be because of negative autoregulation of the norR promoter. This is a potential drawback of our expression system, as small changes in NorR specific activity may be masked or amplified as a consequence of even modest overexpression. Nevertheless, this system has proved very useful as a means of quantifying large changes in NorR activity and can be complemented by a variety of other in vivo and in vitro techniques (see later). NorR activates transcription in response to NO, so a technical issue is what source of NO should be used for in vivo and in vitro measurements of NorR activity. NorR responds in vivo to pure NO provided in the gas or aqueous phase (Gardner et al., 2003; Justino et al., 2005) and to a variety of treatments that impose nitrosative stress, including exposure to acidified nitrite, S-nitrosglutathione, and nitroprusside (Flatley et al., 2005; Hutchings et al., 2002; Mukhopadhyay et al., 2004; Pohlmann et al., 2000). The respiratory reduction of nitrite during anaerobic growth at neutral pH is accompanied by the
238
Benoıˆt D’Autre´aux et al.
formation of traces of NO (Corker and Poole, 2003; Ji and Hollocher, 1988; Van Doorslaer et al., 2003). This endogenously generated NO is sufficient to activate NorR (Constantinidou et al., 2006), implying that the NO affinity of NorR is poised to be sensitive to the NO generated endogenously from nitrite respiration. Thus, one physiological function of the NorR response might be to protect E. coli against the harmful effects of nitrite respiration (although this idea has never been tested rigorously). For these reasons, we have chosen to use NO generated endogenously from nitrite as a physiological source of NO for measurements of NorR activity. For measurements of norV-lacZ reporter activity, we have developed the following growth protocol. Plasmids expressing norR and its mutant alleles are transformed into NPT1003, a derivative of E. coli MC1000 (araD139 D[ara-leu] D[codB-lacI ] galK16 galE15 relA1 rpsL spoT1) with a norR::cat mutation and a lacZ reporter fusion to the norVW promoter inserted at the phage l attachment site. Cultures are grown with shaking in 50 ml of L-broth (tryptone 10 g.liter1; yeast extract 5 g.liter1; NaCl 5 g.liter1) at 37 until the optical density (at 650 nm) is approximately 0.3. At this point, glucose is added to the culture to a final concentration of 1% (w/v). Cultures are split into 8-ml Bijou bottles and are grown anaerobically overnight at 37 , either in the presence or in the absence of potassium nitrite (4 mM ). The following morning, b-galactosidase activity is assayed using a standard procedure (Miller, 1992).
3. Measurement of Transcriptional Activation by NorR In Vitro The ability of NorR to bind to its target DNA recognition sequences has been measured by gel retardation, DNase I protection, and methylation protection experiments (Tucker et al., 2004). However, this activity does not appear to be influenced by the binding of NO to the ferrous iron center in the GAF domain of NorR (D’Autre´aux et al., 2005) and therefore NorR can apparently bind enhancer sites in both induced and noninduced states. To determine the ability of NorR to activate transcription, we measured the formation of open promoter complexes in which the activator causes RNA polymerase to undergo the transition from the closed promoter state to the open state in which the promoter DNA is locally melted to facilitate transcription initiation. In the case of s54-dependent systems, the AAAþ domain of enhancer-binding proteins is required for the coupling of energy yielded from nucleotide hydrolysis to the remodeling of the s54 subunit of RNA polymerase (Rappas et al., 2005; Zhang et al., 2002). Open complex formation is typically characterized by the presence of a heparin-resistant species in gel retardation assays containing all the components required for
Analysis of the Nonheme Iron NO Sensor, NorR
239
transcription initiation (Eydmann et al., 1995). Heparin is used to distinguish between open promoter complexes and other protein–DNA interactions because open promoter complexes are insensitive to the competitive binding of heparin. Interestingly, NorR binding at enhancer-binding sites upstream of the norV promoter is heparin resistant, so open complexes in this case are distinguished by the formation of a supershifted species (D’Autre´aux et al., 2005). This observation suggests that NorR forms highly stable complexes with DNA that are difficult to disrupt competitively, although this has not yet been fully investigated. The open complex assay for NorR was initially established using a constitutively active form of the protein that lacks the GAF domain (NorRDGAF) to allow assays to be carried out under aerobic conditions in the absence of NO. These studies have demonstrated that open complex formation at the norV promoter on linear DNA templates requires NorRDGAF, core RNA polymerase (Epicentre), s54, integration host factor (IHF), ATP, and CTP (D’Autreaux et al., 2005). Binding of IHF to the DNA fragment between the s54 promoter and the NorR enhancer sites is presumably important for the bending of the DNA to allow contact between upstream-bound NorR and s54 RNA polymerase (Hoover et al., 1990; Perez-Martin and de Lorenzo, 1996). CTP is important for the stability of open complexes at the norV promoter because it is the initiating nucleotide for the norV transcript, while ATP is required to provide the energy needed to remodel s54 RNA polymerase. Stabilization of open promoter complexes by the presence of the initiating nucleotide has been described previously (Eydmann et al., 1995). The template DNA fragment used to assay NorR-dependent open complex formation is derived from the plasmid pNPTprom (D’Autre´aux et al., 2005; Tucker et al., 2004). pNPTprom was generated by cloning the entire norR/norVW intergenic region, as well as the first 86 and 62 bp of the norR and norV coding regions, respectively, into the SmaI site of pUC19. Restriction digestion of pNPTprom with EcoRI and BamHI yields a 362-bp fragment, which is then 30 end labeled with [g32P]dGTP using the Klenow fragment of DNA polymerase I (Invitrogen). IHF and s54 are purified as described previously (Soderback et al., 1998), except that EDTA is omitted from buffers to prevent the loss of iron from NorR in the final reactions. Open complex assays carried out using iron-reconstituted NorR are done in an anaerobic chamber (<1 ppm oxygen, Belle Technologies) to prevent oxidation of the ferrous iron center. The open complex formation is assayed in TM buffer (50 mM Tris-HCl, pH 8.0, 8 mM MgCl2, 25 mM NaCl, 3.5% polyethylene glycol 6000) containing 1 nM template DNA, 200 nM core RNA polymerase (Epicentre), 200 nM s54, 130 nM IHF, 5 mM ATP, and 0.5 mM CTP. Reaction components are incubated in TM buffer for 10 min at 30 before the addition of ferrous NorR to a final concentration of 500 nM. A solution of NO is prepared by decomposing
240
Benoıˆt D’Autre´aux et al.
MAHMA NONOate in TM buffer (t1/2 ¼ 11 min, 30 , at pH 8.0) for 30 min before adding it to the reaction mixtures to obtain a final concentration of 20 mM NO. After 30 min of incubation, reaction mixtures are mixed with OPC Dye (50% glycerol, 0.05% bromphenol blue, 0.1% xylene cyanol, 0.7 mg/ml heparin) to a final concentration of 15% (w/v) and immediately loaded onto a 4% (w/v) polyacrylamide gel (acrylamide:bisacrylamide ratio, 80:1) made in M buffer (25 mM Tris-HCl, pH 8.6, 400 mM glycine). Gels are prerun for 1 h at 180 V in M buffer prior to loading and at 150 V after the samples are applied to the gel. Gels are then dried down onto filter paper for analysis. A Fuji film FLA-7000 phosphorimager is used to visualize the image and to quantify the percentage of radioactivity in open promoter complexes. These experiments clearly establish that interaction of NO with the ferrous ion center in NorR is necessary for transcriptional activation.
4. Detection of the Ferrous-Nitrosyl Form of NorR by In Vivo Electron Paramagnetic Resonance (EPR) In the case of transient NO complexes with metal centers, EPR spectroscopy is a powerful tool that can be adapted for in vivo detection and characterization of NO-bound species. NO is a radical molecule (with one unpaired electron) and therefore has paramagnetic properties. Paramagnetic molecules are usually EPR active, whereas diamagnetic molecules are EPR silent, thus allowing detection and characterization of NO complexes in biological samples (that contain an abundance of EPR silent diamagnetic compounds). EPR allows detection of NO complexes within cells and can provide information about the chemical composition as well as the geometry of NO complexes. In this respect, NO serves as a structural probe of metal ion centers. We have used EPR spectroscopy to detect NO-bound NorR species and to infer information about their properties in vivo (D’Autre´aux et al., 2005). Escherichia coli cells are treated with a pure source of NO under anaerobic conditions to avoid oxidation and thus production of other reactive N species that may decrease the NO concentration and/or interfere with NO signaling. Because the sensitivity of EPR is in the submicromolar range, detection of EPR active species requires overexpression of the candidate protein. In addition, controlled overexpression allows identification of the EPR signal specifically associated with the protein of interest. Thus, for these experiments we cloned the norR gene downstream of the T7 promoter in the pET28 vector and transformed the expression clone into a strain (BL21) that expresses T7 RNA polymerase. Freshly transformed
241
Analysis of the Nonheme Iron NO Sensor, NorR
bacteria are grown in 25 ml L-broth containing 1% glucose (w/v). Cultures are grown at 30 under anaerobic conditions, achieved by filling the culture bottles. Expression of norR is induced when the culture density (at 600 nm) reaches approximately 0.6 by the addition of 50 mM isopropyl-b-D-thiogalactoside (IPTG). Three hours after induction, 50 ml of 100 mM MAHMA NONOate is added. The alkaline stock NONOate solution is stable, but the compound degrades at pH 7.5 (t1/2 of 3 min at 30 ) releasing pure NO. The cells are harvested at 4000 rpm for 10 min, and the pellets are resuspended in 0.5 ml of 100 mM Tris-HCl (pH 7.5), 25 mM NaCl, and 10% glycerol (v/v). Cell suspensions are transferred to an EPR tube and immediately frozen in liquid nitrogen. Similar experiments can be performed with clones expressing domains of NorR or proteins with single amino acid substitutions. Controls are provided by cells that do not overexpress NorR. Typical results for cells overexpressing NorR, and control experiments, are shown in Fig. 13.1. In contrast to untreated cells (spectrum 1), cells treated with NO exhibit a signal in the g ¼ 2 region (spectrum 2), which is
(1)
(2)
(3) g = 4.20 (4)
0
g = 3.80
1000 2000 3000 Magnetic field (G)
4000
Figure 13.1 In vivo EPR of NorR overexpressing E. coli cells. EPR spectra were recorded on intact E. coli BL21(DE3) cells expressing NorR. Cultures were given the following treatments: (1) none, (2) treatment with NO, (3) induced for protein expression with 50 mM IPTG, and (4) induced for protein expression with 50 mM IPTG and then treated with NO. X-band EPR spectra were recorded on a Bruker ER 200 D-SRC instrument with the following conditions: frequency, 9.477 GHz; power, 2 mW; modulation amplitude,10 G; modulation frequency,100 kHz; and temperature 8 K.
242
Benoıˆt D’Autre´aux et al.
probably because of the interaction of NO with cells or medium components. In the absence of NO, cells induced for NorR expression (spectrum 3) have similar spectral features to uninduced cells. EPR spectra of cells expressing NorR and treated with NO (spectrum 4) display an EPR signal in the g ¼ 4 region (H 1650 G) that is not apparent in the controls, therefore suggesting that it is because of the interaction of NO with NorR. We have shown previously that this signal arises from a complex formed between NO and NorR and is not because of any other protein that might be expressed only when NorR is active (D’Autre´aux et al., 2005). The g values associated with the EPR spectrum (g ¼ 3.80 and g ¼ 4.20) are consistent with a complex associated with an S ¼ 3/2 ground state. Similar NO complexes have been documented in the literature (Arciero et al., 1983; Clay et al., 2003; Hauser et al., 2000; Ray et al., 1999) as arising from the interaction of reduced nonheme iron containing enzymes (S ¼ 2) and NO (S ¼ 1/2), leading to [Fe(NO)] complexes (S ¼ 3/2). These EPR experiments probe the structure of the NorR sensing domain and demonstrate that, in vivo, NorR contains a nonheme iron center that binds NO.
5. In Vitro Reconstitution of the Iron Center in NorR The mononuclear nonheme iron center of NorR requires reconstitution, as the purified protein contains only 0.3 Fe atom/monomer when purified under anaerobic conditions (D’Autre´aux et al., 2005). This is typical behavior for nonheme iron proteins, where the metal center is often labile, as in the case of superoxide dismutase (Hartman et al., 1986; Ken et al., 2005) and superoxide reductase ( Jovanovic et al., 2000). The GAF domain of NorR (called GAFNorR), which contains the iron-binding site, is completely devoid of iron when purified, but can be reconstituted in vitro, with a high yield (0.7 Fe atom/GAFNorR monomer), by incubating the protein with a slight excess of Fe2þ under anaerobic conditions. Surprisingly, the same procedure is totally ineffective for reconstitution of the full-length NorR protein. The addition of reducing agents (dithiothreitol, dithionite), MgATP, or the use of increased temperatures does not improve the reconstitution reaction. However, full-length NorR can be reconstituted by the addition of Fe2þ directly to a crude extract prepared from E. coli cells overexpressing NorR. The protein subsequently purified from this extract is reconstituted with high yield (0.9 Fe atom/NorR monomer). Our inability to reconstitute NorR postpurification suggests that there may be a factor in cell extracts required for efficient incorporation of iron into NorR. Reconstitution of the iron center in the isolated GAF domain does
Analysis of the Nonheme Iron NO Sensor, NorR
243
not require this unknown factor, suggesting that the central and the DNA-binding domains in the full-length NorR inhibit iron incorporation. Purified GAFNorR protein (D’Autre´aux et al., 2005) is reconstituted anaerobically in a glove box. A solution of GAFNorR (1 mM ) is incubated with sodium dithionite (1 mM ) for 10 min. The dithionite is removed by gel filtration through Sephadex G25 (NAP5, Amersham Biosciences), and the GAFNorR is then incubated for 30 min at 30 with stirring with two equivalents of Fe2þ (2 mM ) from a solution of Fe(NH4)2(SO4)2 prepared in distilled water. Excess Fe2þ is removed by gel filtration through Sephadex G25. Reconstitution typically yields 0.7 Fe atom/GAFNorR monomer. For reconstitution of full-length NorR, all the cell extraction steps are performed under anaerobic conditions. Cells pellets are resuspended in 100 mM Tris-HCl, pH 8.5, 25 mM NaCl, and 5% glycerol and are made anaerobic by equilibration in a glove box (<1 ppm oxygen). Cell disruption is performed under constant nitrogen flushing. The membrane fraction and insoluble material are removed by centrifugation at 45,000 rpm for 1 h. Iron reconstitution and affinity chromatography are performed in the glove box. The crude extract (30 ml) is incubated with Fe2þ (1 mM ) from an aqueous solution of Fe(NH4)2(SO4)2 and MgATP (1 mM ). NorR is chromatographed on a heparin column using the following loading buffer: 100 mM Tris-HCl, pH 8.5, 10 mM NaCl, 1 mM MgATP, 5% glycerol and elution buffer: 100 mM Tris-HCl, pH 8.5, 1 M NaCl, 1 mM MgATP, 5% glycerol (v/v). Fractions containing NorR are pooled, and the protein is further purified on a Superdex 200 16/60 column (Amersham Biosciences) at a flow rate of 1.5 ml/min in 100 mM Tris-HCl, pH 8.5, 100 mM NaCl, and 5% glycerol. Iron-containing NorR elutes as a trimer, as does the apo form of the protein. The reconstitution yields 0.9 Fe atom/NorR monomer. The protein is stored in liquid nitrogen with 50% glycerol.
6. Measurement of NO Affinity The dissociation constant, Kd, of the NorRFe(NO) complex [Eqs. (13.1) and (13.2)] indicates the lowest range of NO concentration NorR is able to detect in the cell. Determination of this parameter is thus crucial to the understanding of the role of NorR as a NO sensor.
NorRFeðNOÞ ¼ NorRFe þ NO Kd ¼
½NorRFe½NO ½NorRFeðNOÞ
ð13:1Þ ð13:2Þ
244
Benoıˆt D’Autre´aux et al.
However, measuring the NO-binding affinity of NorR has proven a challenging task because of a lack of sensitive spectroscopic features for monitoring formation of the NorRFe(NO) complex. Sensitivity and uncertainty are two critical criteria that will drive the choice of the most appropriate technique. UV-visible spectroscopy has been used extensively with heme-based proteins (Butler et al., 1997). Similarly, the NorRFe(NO) complex can be monitored in the UV-visible range (lmax ¼ 435 nm), but not free NorRFe or NO. The UV-visible features of the NorRFe(NO) complex are rather weak compared to heme proteins, raising the question of accuracy in Kd determinations. The relative uncertainty of Kd (DKd/Kd) is the sum of single relative uncertainties of each concentration at equilibrium [Eq. (13.3)], where E, L, and EL represent NorRFe, NO, and NorRFe (NO), respectively,
DKd D½E D½L D½EL þ þ ¼ Kd ½E ½L ½EL
ð13:3Þ
The Kd value derives from the measurement of concentrations at equilibrium. The absolute uncertainty of EL, D[EL], is because of the absorbance reading accuracy, which is 0.001 absorbance units according to the manufacturer (Perkin Elmer Lambda 35). The uncertainty of [E] and [L] is linked to D[EL], [E]0 (initial concentration in E) and [L]0 (initial concentration in L) by the law of conservation of matter [Eqs. (13.4) and (13.5)]:
D½E ¼ D½E0 D½EL
ð13:4Þ
D½L ¼ D½L0 D½EL
ð13:5Þ
We have computed the variation of relative uncertainty of Kd (DKd/Kd) upon saturation of the enzyme with L for several values of [E]0 (Fig. 13.2). These data show that the relative uncertainty decreases as [E]0 increases and reaches a limit at high E0. As [E]0 increases, the range of absorbance values increases along with the precision of the concentration determination. Nevertheless, as [E]0 increases, the concentration of free ligand at equilibrium decreases and therefore the advantage of higher absorbance values will be lost as a consequence of the significant drop in free ligand concentration. In addition, the Beer–Lambert law is only valid for absorbances below approximately 1.0; therefore, with E435 nm ¼ 3500 M1cm1, [E]0 should not exceed 300 mM. We have estimated the relative uncertainty of Kd (DKd/Kd) for several values of Kd (Fig. 13.3). These data indicate that, in the case of NorR, the UV-visible method allows only determination of Kd
245
Analysis of the Nonheme Iron NO Sensor, NorR
100
80 ΔKd/Kd (%)
1 60 2 40
3 4 5
6 7
20
8 0 0.0
0.5
1.0
1.5
2.0
[L]o / [E]o
Figure 13.2 Relative uncertainty of Kd as a function of [E]0.The relative uncertainty of Kd, DKd/Kd, has been computed as a function of [L]0/[E]0 for several values of [E]0: 1, 2, 5, 10, 20, 50, 100, and 200 mM, corresponding to the curves labeled 1 to 8, respectively, using a Kd value of 10 mM. For convenience in graphical viewing, the values of DKd/Kd have been plotted as a function of [L]0/[E]0. The absolute uncertainty D[EL] ¼ D[E] ¼ D[L] has been set to 0.001/e, where e is the absorption coefficient of the NorRFe(NO) complex at lmax ¼ 435 nm, e435 nm ¼ 3500 M1cm1; uncertainty of E0 and L0 has not been considered.Values of [EL] derived from the expression of Kd, [EL] ¼ , and [L] and [E] are obtained using the laws of conservation of matter.
100
5 4 3
ΔKd/Kd (%)
80
6
2
7 60
8
40
1
9
20 0 0
50
100
150 200 250 [L]o (mM)
300
350
Figure 13.3 Relative uncertainty of Kd as a function of Kd.The relative uncertainty of Kd (DKd/Kd) has been computed as a function of [L]0 for several values of Kd at [E]0 ¼ 300 mM using an absolute uncertainty D[EL] ¼ D[E] ¼ D[L] set to 0.001/E, where E is the absorption coefficient of the NorRFe(NO) complex at lmax ¼ 435 nm, E435 nm ¼ 3500 M1cm1.Values of DKd/Kd are plotted for Kd ¼ 1 nM, 10 nM, 50 nM, 100 nM, 200 nM, 500 nM,1 mM, 2 mM, and 10 mM, corresponding to the curves labeled 1 to 9.Values of [EL] derived from the expression of Kd, [EL] ¼ , and [L] and [E] are obtained using the laws of conservation of matter.
246
Benoıˆt D’Autre´aux et al.
above 10 mM with a precision below 10%. In comparison, determination of NO-binding affinity with myoglobin (e420 nm ¼ 150,000 M1cm1) using the same method is expected to allow determination of Kd above 100 nM. We have therefore used a more appropriate technique for our system allowing direct detection of free NO. NO electrodes are now designed to measure NO specifically, with sensitivity approaching 1 nM and allow direct measurements in solution. However, as NO equilibrates with the gas phase and is highly reactive with oxygen, these experiments require anaerobic conditions and no headspace above the reaction mixture.
7. Standardization of the NO Electrode All the reactions are performed under anaerobic conditions. We use the ISO-NOP 200-mm electrode connected to an ISO-NO Mark II nitric oxide sensor (World Precision Instruments) for measurement of NO concentrations. The electrode is calibrated over the range of utilization with standard NO solutions. Prior to any measurements, the NO electrode is soaked in buffer A (100 mM Tris-HCl, pH 8.5, 200 mM NaCl, 5% glycerol) and left to equilibrate until the current stabilizes (usually overnight). A standard stock solution is prepared by bubbling pure NO gas through buffer A and is titrated using conversion of horse heart deoxymyoglobin, MbFe, to nitrosyl myoglobin, MbFe(NO). The deoxy form of myoglobin [MbFe(II)] is prepared by incubating horse heart myoglobin (50 mg, Sigma) in 100 mM Bis-Tris propane (pH 7.0) with sodium dithionite (2 mg) under anaerobic conditions for 10 min under stirring. The dithionite is removed by purification of MbFe using Sephadex G25 (NAP5, Amersham Biosciences). The absence of met myoglobin [MbFe (III)] and oxymyoglobin [MbFe(II)-O2] is confirmed by the absence of the corresponding UV-visible features: 417, 542, and 580 nm and 408, 502, and 630 nm, respectively. The concentration of MbFe is determined spectrophotometrically at 434 nm (e434 nm ¼ 114 mM1cm1) (Herold et al., 2001). The formation of MbFe(NO) is monitored at 420 nm upon saturation of the enzyme with NO from the stock solution (Fig. 13.4). The isosbestic point at 426 nm indicates that the MbFe is converted exclusively and quantitatively to MbFe(NO) and not to met- or oxymyoglobin. The concentration of MbFe(NO) is determined from the absorbance at 420 nm according to Eq. (13.6):
½MbFeðNOÞ ¼
A420 nm e420 nm ðMbFeÞ ½MbFe0 ðe420 nm ðMbFeNOÞ e420 nm ðMbFeÞ
ð13:6Þ
247
Analysis of the Nonheme Iron NO Sensor, NorR
1.0 MbFeNO MbFe
Absorbance
0.8 0.6 0.4
0.2 0.0 380
400
420 440 l nm
460
Figure 13.4 Titration of NO solution using myoglobin. Electronic absorption spectra, in the UV-visible domain, of horse heart deoxy MbFe at a concentration of 5.73 mM (1 ml), upon saturation with NO by the addition of 1, 2, 4, 5, 7, and 8 ml from a saturated solution of NO.
The absorption coefficients e420 nm(MbFe) and e420 nm(MbFeNO) are determined from the spectrum of pure MbFe and pure MbFeNO obtained by the addition of excess NO to MbFe. We have determined the absorption coefficients e420 nm (MbFe) ¼ 87 mM1cm1 and e420 nm (MbFeNO) ¼ 148 mM1cm1 and a concentration of NO, [NO] ¼ 955 mM, for our stock saturated solution using this procedure. For calibration of the NO electrode, the stock solution is diluted into buffer A. The diluted solutions are prepared in a septum-sealed bottle with a limited free headspace to avoid exchange of NO with the atmosphere, with the NO electrode passing through the septum. A series of NO solutions is prepared by successive addition of a fixed volume of the stock NO solution into the septum-sealed bottle using a gas-tight Hamilton syringe. The current (mA) is measured after each addition and is plotted as a function of the concentration of NO.
8. Determination of NorRFe(NO) Kd A solution of the NorRFe protein (3 mM) in buffer A is progressively saturated by the addition of NO, from the standard NO solution, using a gas-tight Hamilton syringe. The concentration of free NO is measured upon saturation of NorRFe (Fig. 13.5).
248
Benoıˆt D’Autre´aux et al.
1.2
[NO]eq (mM)
1.0 0.8 0.6 0.4 0.2 0.0 0
1
2 3 [NO]0 (mM)
4
Figure 13.5 Determination of NorRFe(NO) dissociation constant. NorRFe (3 mM) was incubated with various amounts of NO from a stock NO solution. The concentration of free NO at equilibrium, [NO]eq, was plotted versus the concentration of NO initially added to the solution, [NO]0. Two different experiments are represented as assay 1 and assay 2. The thin line represents the best fit obtained with the theoretical model described in the text.
½NO ¼
½NorRFe0 Kd þ ½NO0 þ
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ð½NorRFe0 þ Kd ½NO0 Þ2 þ 4:Kd :½NO0 2
ð13:7Þ
Data are analyzed by nonlinear regression analysis using Eq. (13.7), derived from the equation of Kd [Eq. (13.2)], for a 1:1 binding model in two independent experiments. We have determined a Kd for the NorRFe (NO) complex of 50 nM with a relative uncertainty of 5%.
9. Conclusions This chapter described a range of molecular, biochemical, and spectroscopic tools for characterization of the nonheme iron-based NO sensor NorR. Application of these tools is providing us and other investigators with new insights into the mechanism of NO sensing by NorR.
ACKNOWLEDGMENT This work was funded by grants from the Biotechnology and Biological Sciences Research Council (BB/D009588/1 to RD) and the National Science Foundation (MCB-0702858 to SS).
Analysis of the Nonheme Iron NO Sensor, NorR
249
REFERENCES Arai, H., Hayashi, M., Kuroi, A., Ishii, M., and Igarashi, Y. (2005). Transcriptional regulation of the flavohemoglobin gene for aerobic nitric oxide detoxification by the second nitric oxide-responsive regulator of Pseudomonas aeruginosa. J. Bacteriol. 187, 3960–3968. Arciero, D. M., Lipscomb, J. D., Huynh, B. H., Kent, T. A., and Munck, E. (1983). EPR and Mossbauer studies of protocatechuate 4,5-dioxygenase: Characterization of a new Fe2þ environment. J. Biol. Chem. 258, 14981–14991. Bu¨sch, A., Pohlmann, A., Friedrich, B., and Cramm, R. (2004). A DNA region recognized by the nitric oxide-responsive transcriptional activator NorR is conserved in b- and gproteobacteria. J. Bacteriol. 186, 7980–7987. Butler, C. S., Seward, H. E., Greenwood, C., and Thomson, A. J. (1997). Fast cytochrome bo from Escherichia coli binds two molecules of nitric oxide at CuB. Biochemistry 36, 16259–16266. Clay, M. D., Cosper, C. A., Jenney, F. E., Jr., Adams, M. W., and Johnson, M. K. (2003). Nitric oxide binding at the mononuclear active site of reduced Pyrococcus furiosus superoxide reductase. Proc. Natl. Acad. Sci. USA 100, 3796–3801. Constantinidou, C., Hobman, J. L., Griffiths, L., Patel, M. D., Penn, C. W., Cole, J. A., and Overton, T. W. (2006). A Reassessment of the FNR regulon and transcriptomic analysis of the effects of nitrate, nitrite, NarXL, and NarQP as Escherichia coli K12 adapts from aerobic to anaerobic growth. J. Biol. Chem. 281, 4802–4815. Corker, H., and Poole, R. K. (2003). Nitric oxide formation by Escherichia coli: Dependence on nitrite reductase, the NO-sensing regulator Fnr, and flavohemoglobin Hmp. J. Biol. Chem. 278, 31584–31592. Cruz-Ramos, H., Crack, J., Wu, G., Hughes, M. N., Scott, C., Thomson, A. J., Green, J., and Poole, R. K. (2002). NO sensing by FNR: Regulation of the Escherichia coli NO-detoxifying flavohaemoglobin, Hmp. EMBO J. 21, 3235–3244. D’Autre´aux, B., Touati, D., Bersch, B., Latour, J. M., and Michaud-Soret, I. (2002). Direct inhibition by nitric oxide of the transcriptional ferric uptake regulation protein via nitrosylation of the iron. Proc. Natl. Acad. Sci. USA 99, 16619–16624. D’Autre´aux, B., Tucker, N. P., Dixon, R., and Spiro, S. (2005). A non-haem iron centre in the transcription factor NorR senses nitric oxide. Nature 437, 769–772. Ding, H., and Demple, B. (2000). Direct nitric oxide signal transduction via nitrosylation of iron-sulfur centers in the SoxR transcription activator. Proc. Natl. Acad. Sci. USA 97, 5146–5150. Eydmann, T., Soderback, E., Jones, T., Hill, S., Austin, S., and Dixon, R. (1995). Transcriptional activation of the nitrogenase promoter in vitro: Adenosine nucleotides are required for inhibition of NIFA activity by NIFL. J. Bacteriol. 177, 1186–1195. Flatley, J., Barrett, J., Pullan, S. T., Hughes, M. N., Green, J., and Poole, R. K. (2005). Transcriptional responses of Escherichia coli to S-nitrosoglutathione under defined chemostat conditions reveal major changes in methionine biosynthesis. J. Biol. Chem. 280, 10065–10072. Gardner, A. M., Gessner, C. R., and Gardner, P. R. (2003). Regulation of the nitric oxide reduction operon (norRVW) in Escherichia coli: Role of NorR and s54 in the nitric oxide stress response. J. Biol. Chem. 278, 10081–10086. Hartman, J. R., Geller, T., Yavin, Z., Bartfeld, D., Kanner, D., Aviv, H., and Gorecki, M. (1986). High-level expression of enzymatically active human Cu/Zn superoxide dismutase in Escherichia coli. Proc. Natl. Acad. Sci. USA 83, 7142–7146. Hauser, C., Glaser, T., Bill, E., Weyhermuller, T., and Wieghardt, K. (2000). The electronic structures of an isostructural series of octahedral nitrosyliron complexes {Fe-NO}(6,7,8) elucidated by Mossbauer spectroscopy. J. Am. Chem. Soc. 122, 4352–4365.
250
Benoıˆt D’Autre´aux et al.
Hausladen, A., Gow, A. J., and Stamler, J. S. (1998). Nitrosative stress: Metabolic pathway involving the flavohemoglobin. Proc. Natl. Acad. Sci. USA 95, 14100–14105. Herold, S., Exner, M., and Nauser, T. (2001). Kinetic and mechanistic studies of the NO*mediated oxidation of oxymyoglobin and oxyhemoglobin. Biochemistry 40, 3385–3395. Hoover, T. R., Santero, E., Porter, S., and Kustu, S. (1990). The integration host factor stimulates interaction of RNA polymerase with NIFA, the transcriptional activator for nitrogen fixation operons. Cell 63, 11–22. Hutchings, M. I., Mandhana, N., and Spiro, S. (2002). The NorR protein of Escherichia coli activates expression of the flavorubredoxin gene norV in response to reactive nitrogen species. J. Bacteriol. 184, 4640–4643. Ji, X. B., and Hollocher, T. C. (1988). Reduction of nitrite to nitric oxide by enteric bacteria. Biochem. Biophys. Res. Commun. 157, 106–108. Jovanovic, T., Ascenso, C., Hazlett, K. R., Sikkink, R., Krebs, C., Litwiller, R., Benson, L. M., Moura, I., Moura, J. J., Radolf, J. D., Huynh, B. H., Naylor, S., et al. (2000). Neelaredoxin, an iron-binding protein from the syphilis spirochete, Treponema pallidum, is a superoxide reductase. J. Biol. Chem. 275, 28439–28448. Justino, M. C., Vicente, J. B., Teixeira, M., and Saraiva, L. M. (2005). New genes implicated in the protection of anaerobically grown Escherichia coli against nitric oxide. J. Biol. Chem. 280, 2636–2643. Ken, C. F., Hsiung, T. M., Huang, Z. X., Juang, R. H., and Lin, C. T. (2005). Characterization of Fe/Mn-superoxide dismutase from diatom Thallassiosira weissflogii: Cloning, expression, and property. J. Agric. Food Chem. 53, 1470–1474. Klink, A., Elsner, B., Strube, K., and Cramm, R. (2007). Characterization of the signaling domain of the NO-responsive regulator NorR from Ralstonia eutropha H16 by sitedirected mutagenesis. J. Bacteriol. 189, 2743–2749. Miller, J. H. (1992). ‘‘A Short Course in Bacterial Genetics.’’ Cold Spring Harbor Laboratory Press. Cold Spring Harbor, NY. Moore, L. J., and Kiley, P. J. (2001). Characterization of the dimerization domain in the FNR transcription factor. J. Biol. Chem. 276, 45744–45750. Mukhopadhyay, P., Zheng, M., Bedzyk, L. A., LaRossa, R. A., and Storz, G. (2004). Prominent roles of the NorR and Fur regulators in the Escherichia coli transcriptional response to reactive nitrogen species. Proc. Natl. Acad. Sci. USA 101, 745–750. Perez-Martin, J., and de Lorenzo, V. (1996). In vitro activities of an N-terminal truncated form of XylR, a sigma 54-dependent transcriptional activator of Pseudomonas putida. J. Mol. Biol. 258, 575–587. Pohlmann, A., Cramm, R., Schmelz, K., and Friedrich, B. (2000). A novel NO-responding regulator controls the reduction of nitric oxide in Ralstonia eutropha. Mol. Microbiol. 38, 626–638. Rappas, M., Schumacher, J., Beuron, F., Niwa, H., Bordes, P., Wigneshweraraj, S., Keetch, C. A., Robinson, C. V., Buck, M., and Zhang, X. (2005). Structural insights into the activity of enhancer-binding proteins. Science 307, 1972–1975. Ray, M., Golombek, A. P., Hendrich, M. P., Yap, G. P. A., Liable-Sands, L. M., Rheingold, A. L., and Borovik, A. S. (1999). Structure and magnetic properties of trigonal bipyramidal iron nitrosyl complexes. Inorg. Chem. 38, 3110–3115. Rodionov, D. A., Dubchak, I. L., Arkin, A. P., Alm, E. J., and Gelfand, M. S. (2005). Dissimilatory metabolim of nitrogen oxides in bacteria: Comparative reconstruction of transcriptional networks. PLOS Comput. Biol. 1, e55. Soderback, E., Reyes-Ramirez, F., Eydmann, T., Austin, S., Hill, S., and Dixon, R. (1998). The redox- and fixed nitrogen-responsive regulatory protein NIFL from Azotobacter vinelandii comprises discrete flavin and nucleotide-binding domains. Mol. Microbiol. 28, 179–192.
Analysis of the Nonheme Iron NO Sensor, NorR
251
Spiro, S. (2007). Regulators of bacterial responses to nitric oxide. FEMS Microbiol. Rev. 31, 193–211. Tucker, N. P., D’Autre´aux, B., Studholme, D. J., Spiro, S., and Dixon, R. (2004). DNA binding activity of the Escherichia coli nitric oxide sensor NorR suggests a conserved target sequence in diverse proteobacteria. J. Bacteriol. 186, 6656–6660. Van Doorslaer, S., Dewilde, S., Kiger, L., Nistor, S. V., Goovaerts, E., Marden, M. C., and Moens, L. (2003). Nitric oxide binding properties of neuroglobin: A characterization by EPR and flash photolysis. J. Biol. Chem. 278, 4919–4925. Zhang, X., Chaney, M., Wigneshweraraj, S. R., Schumacher, J., Bordes, P., Cannon, W., and Buck, M. (2002). Mechanochemical ATPases and transcriptional activation. Mol. Microbiol. 45, 895–903. Zhao, Y., Brandish, P. E., Ballou, D. P., and Marletta, M. A. (1999). A molecular basis for nitric oxide sensing by soluble guanylate cyclase. Proc. Natl. Acad. Sci. USA 96, 14753–14758.
This page intentionally left blank
S E C T I O N
T H R E E
ADVANCE SPECTROSCOPIC METHODS
This page intentionally left blank
C H A P T E R
F O U R T E E N
Hemoglobins from Mycobacterium tuberculosis and Campylobacter jejuni: A Comparative Study with Resonance Raman Spectroscopy Changyuan Lu,* Tsuyoshi Egawa,* Masahiro Mukai,† Robert K. Poole,‡ and Syun-Ru Yeh* Contents 1. Hemoglobin Superfamily: An Overview 2. Microbial Hemoglobins 3. Resonance Raman Spectroscopy: Applications in Hemeproteins 3.1. Porphyrin core vibrational modes 3.2. The proximal iron-histidine stretching mode 3.3. Distal axial ligand vibrational modes 3.4. Potential pitfalls 4. Structures and Functions of Microbial Hemoglobins 4.1. The TrHb-I from M. tuberculosis (TrHbN) 4.2. The TrHb-II from the M. tuberculosis (TrHbO) 4.3. The TrHb-III from the C. jejuni (TrCtb) 4.4. The sdHb from C. jejuni (Cgb) 5. Closing Remarks Acknowledgments References
256 257 258 261 262 263 266 266 267 275 277 279 281 282 282
Abstract Three groups of hemoglobins (Hbs) have been identified in unicellular organisms: (1) the truncated Hbs (trHb) that display a novel two-over-two a-helical structure, (2) the flavohemoglobins (FHb) that comprise a Hb domain with a classical three-over-three a-helical structure and a covalently attached flavincontaining reductase domain, and (3) the single-domain Hbs (sdHb) that exhibit high sequence homology and structural similarity to the Hb domain of FHb.
* { {
Department of Physiology and Biophysics, Albert Einstein College of Medicine, Bronx, New York Mitsubishi Kagaku Institute of Life Sciences, Minamiooya, Machida, Tokyo, Japan Department of Molecular Biology and Biotechnology, University of Sheffield, Sheffield, United Kingdom
Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37014-6
#
2008 Elsevier Inc. All rights reserved.
255
256
Changyuan Lu et al.
On the basis of phylogenetic analysis, the trHbs can be further divided into three subgroups: TrHb-I, TrHb-II, and TrHb-III. The various classes of Hbs may coexist in the same organism, suggesting distinct functions for each class of Hb. This chapter reviews the structural and functional properties of a TrHb-I (trHbN) and a TrHb-II (trHbO) from Mycobacterium tuberculosis, as well as a TrHb-III (trCtb) and a sdHb (Cgb) from Campylobacter jejuni on the basis of resonance Raman spectroscopic studies.
1. Hemoglobin Superfamily: An Overview Hemoglobins (Hbs) have been discovered in organisms from virtually all kingdoms (Egawa and Yeh, 2005). The individual subunit of all Hbs discovered to date consists of a polypeptide chain with six to eight a-helical segments that fold around a heme group (Fig. 14.1). The helices building up the globin fold are conventionally labeled A to H according to the linear sequence order; in addition, the various topological positions within each helix are numbered sequentially. The proximal heme ligand is always a histidine residue at the F8 position. The distal ligand-binding pocket is typically constructed from the B, E, and part of the G helices. The identity of the B10 residue on the B helix and the E7, E10, and E11 residues on one A
D
B B A
B G
E
E
C
G H
H
trHbN
F8 F swMb C trHbN E11 B10 E7 E10
Figure 14.1 Crystal structures of (A) sperm whale oxymyoglobin (PDB:1MBO) and (B and C) oxy-trHbN from M. tuberculosis (PDB:1IDR). (A) Nomenclatures of the eight helices are labeled A to H as indicated.
257
Resonance Raman Studies of Microbial Hemoglobins
unique topological surface of the E helix (see Fig. 14.1C) are important for the regulation of ligand binding and discrimination. In mammalian globins, the E7 position is almost invariably occupied by a histidine. The E7His is believed to stabilize the heme-bound dioxygen by H-bonding to it; however, the B10, E10, and E11 positions are typically occupied by hydrophobic residues to ensure that the chemical integrity of the heme-bound dioxygen is preserved. In contrast, the distal residues in the hemeproteins performing oxygen chemistry, such as peroxidases and oxidases, are much more polar. The polar nature of the distal heme environment in these proteins plays an important role in activating the O-O bond of hemebound peroxide molecules. Intriguingly, the distal residues of microbial Hbs are also much more polar than that of mammalian globins, suggesting that the structures of these Hbs are tailored to perform functions other than oxygen transport.
2. Microbial Hemoglobins Since the first recognition of Hb in microorganisms in the 1930s by Warburg, three groups of Hbs have been characterized in unicellular organisms (Fig. 14.2) (Frey and Kallio, 2003; Moens et al., 1996; Vinogradov et al., 1992; Wajcman and Kiger, 2002; Wu et al., 2003). The first group, termed truncated Hbs (trHb), consists of proteins with 110–140 amino acid residues and a novel two-over-two a-helical sandwich motif Microbial Hb trHb
trHbI trHbII (trHbN) (trHbO)
sdHb (Cgb) trHbIII (trCtb)
E7 E10 M. tb C. jejuni
FHb (Hmp)
E11 B10 CD1 G8
trHbN Leu Lys Gln Tyr Phe Val trHbO Ala Arg Leu Tyr trCtb His Lys Ile
Tyr Trp
Tyr Phe Trp
Gln Ala
Leu Tyr Phe Val
E. coli
Hmp Gln Ala
Leu Tyr Phe Val
Mb
swMb His Thr Val
Cgb
Leu Phe
Ile
Figure 14.2 Classification of microbial Hbs as discussed in the text, the representative Hb in each group, and a list of the critical residues associated with each globin listed.
258
Changyuan Lu et al.
(see Fig. 14.1B). The second group consists of flavohemoglobins (FHb) from bacteria and fungi, which comprise a Hb domain with a classical threeover-three a-helical sandwich motif and a covalently attached flavincontaining reductase domain. The third group is made up of single-domain Hbs (sdHb) that exhibit high sequence homology and structural similarity to the Hb domain of FHb. On the basis of phylogenetic analysis, trHbs can be further divided into three subgroups: TrHb-I, TrHb-II, and TrHb-III (Wittenberg et al., 2002). Intriguingly, the various classes of microbial Hbs may coexist in the same organism. For example, Mycobacterium tuberculosis contains a TrHb-I (trHbN) and a TrHb-II (trHbO), Campylobacter jejuni contains a TrHb-III (trCtb) and a sdHb (Cgb), and Mycobacterium avium contains three TrHbs, one from each subgroup. These findings suggest distinct functions for each class of Hb. Despite their structural and functional diversity, all microbial Hbs contain a proximal histidine at the F8 position that coordinates to the heme iron and a highly conserved tyrosine residue at the B10 position in the distal pocket. The distal histidine at the E7 position, which is important in stabilizing heme-bound oxygen in mammalian globins, may be replaced by a variety of different polar or nonpolar residues (see Fig. 14.2). The crystal structures of at least one member of each group/subgroup of microbial Hbs are available in the protein data bank. The two-over-two a-helical structure of trHbs is made up of B, E, G, and H helices (see Fig. 14.1B). The F helix, hosting the F8 histidine, is mostly replaced by an extended loop; in addition, the pre-EF loop is shorter, as compared to the classical globin structure. The exogenous ligand-binding site, like mammalian globins, is formed by the B, E, and part of the G helices. The CD-D region linking the B and E helices is very short, forcing the E helix to be very close to the heme distal face. The functional diversity of trHbs is dictated by a sophisticated H-bonding network linking the highly conserved B10Tyr and the neighboring E7, E11, CD1, and/or G8 residues, as discussed in this chapter. Likewise, the ligand-binding properties and chemical reactivities of FHb and sdHb are modulated by a H-bonding interaction between B10Tyr and/or E7Gln residues, although their tertiary fold is similar to the three-over-three a-helical structure of mammalian globins.
3. Resonance Raman Spectroscopy: Applications in Hemeproteins The quantum-mechanical basis for Raman scattering has been reviewed extensively (Ferraro et al., 2002; Long, 2001). The general principle of Raman scattering is depicted briefly in Fig. 14.3. When a molecule is subjected to radiation with a frequency n0, most of the photons are scattered
259
Resonance Raman Studies of Microbial Hemoglobins
A
Laser n = n0
B
Raman scattering n = n0 – ni (Stokes) Molecule n = n0 + ni (anti-stokes)
Excited state
v = 1 v = 0
hnex Rayleigh scattering n = n0
hni Ground state
v=1 v=0 Stokes Anti- Resonance stokes
Figure 14.3 Schematic illustration of vibrational Raman effect as discussed in the text.
A
B
Stokes
Soret band
Peptide bond Rayleigh Trp Tyr Phe
Anti-stokes
→
(+)
0 → Raman shift (Δ cm−1)
(–)
200
Q band
300 400 500 Wavelength (nm)
600
Figure 14.4 (A) Schematic illustration of a typical scattering spectrum containing Rayleigh, Stokes, and anti-Stokes bands. (B) A typical optical absorption spectrum of a heme protein. The Soret and Q bands originate from the p to p* electronic transitions of the heme prosthetic group. Those in the 280- and 200-nm regions are assigned to the electronic transitions of the aromatic side chain groups (Trp,Tyr, and Phe) and the peptide backbone, respectively.
from the molecule without a change in frequency (n0 ¼ n0). This is called elastic or Rayleigh scattering. A small fraction of the photons (1 in 107 photons) is scattered by losing or gaining a quantum of vibrational energy (n0 ¼ n0 ni), termed Stokes and anti-Stokes Raman scattering, respectively, as illustrated in Fig. 14.3B. The Raman shift ni,, reflects the energy of an internal vibrational mode of the molecule. The scattering event occurs in 1014 seconds or less. In the spectrum of the scattered light, frequencies of the Stokes and anti-Stokes bands are equally displaced with respect to the frequency of the incident light (Fig. 14.4A). The intensities of the antiStokes lines are typically much weaker than those of the Stokes lines because the anti-Stokes scattering originates from molecules in excited vibrational states (e.g., v ¼ 1 in Fig. 14.3B), which is typically poorly populated at room
260
Changyuan Lu et al.
temperature based on the Boltzmann distribution theory. Because of the weak nature of the anti-Stokes lines, only Stokes lines are considered in most applications. When the incident photon energy (hno) coincides with the electronic transition energy of the molecule (hnex), the Raman intensity is strongly enhanced, typically by a factor of 102 to 106. This is termed the resonance Raman (RR) effect. Metalloporphyrins present in heme proteins have strong electronic transitions in the Soret (390–450 nm) and visible (500–600 nm) regions as shown in Fig. 14.4B (Gouterman, 1978). When excited with a laser that has an output coincident with these electronic transitions, the Raman spectrum of the heme moiety is resonantly enhanced whereas that of the surrounding protein matrix is not. This allows the probe of the heme active center without spectral interference from the surrounding protein matrix. Resonance Raman spectroscopy is a structure-rich probe. Like infrared absorption spectroscopy, it provides information regarding bond strength, angle, and geometry of the functional groups of the molecule of interest. Despite their similar information content, there are several advantages of using RR over infrared absorption spectroscopy for biological applications. 1. The water background in the RR spectrum is very weak, making it an outstanding technique for studying biological molecules in aqueous environments. 2. The ability of RR to selectively enhance the vibrational modes of an individual chromophore in proteins makes it very convenient for studying proteins with multiple chromophores. 3. RR with Soret excitation offers an exquisitely sensitive probe for the structural properties of heme proteins, with which the interactions between an endogenous or an exogenous ligand bound to the heme and the protein matrix surrounding it can be explored in depth, and, with a pulsed laser system, dynamic properties of a heme protein can be further unraveled. 4. Because of its sensitivity, only a minute amount of protein sample is required for each measurement (50–100 ml of 10–50 mM protein). A typical Raman instrument is depicted in Fig. 14.5 (McCreery, 2000; Smith, 2004; Spiro and Czernuszewicz, 2000; Wang et al., 1996). In this setup, the monochromatic output from a laser first passes through a bandpass filter to remove plasma lines from the laser tube and a neutral density filter to reduce the laser intensity, before it is focused on a sample cell. For steady-state measurements, the sample cell [e.g., a cylindrical cuvette or a nuclear magnetic resonance (NMR) tube] is kept spinning to avoid laserinduced photo damage to the sample. The scattered light, typically at right angles to the excitation, is focused into the entrance slit of a polychromator through an achromatic camera lens. A notch filter placed between the
261
Resonance Raman Studies of Microbial Hemoglobins
Mirror
Focus lens
Laser Neutral density filter
Band-pass filter Polychromator
Sample cell Camera lens
Computer
Notch filter CCD detector
Figure 14.5 Schematic illustration of a typical Raman instrument as described in the text.
camera lens and the polychromator is used to filter out the incident laser light (the Rayleigh light). The dispersed light is detected by a liquid nitrogencooled charge-coupled device attached to the exit port of the polychromator and analyzed by a computer attached to it. In a Raman spectrum, the intensity of the scattered light is plotted as a function of the Raman shift (Dcm1), the relative energy of the scattered photon with respect to the incident photon. Several other types of sampling arrangements, such as a cryostat for low-temperature measurements or a flow cell for continuous-flow applications, may be used depending on the special needs of the experiments (Spiro and Czernuszewicz, 2000). The application of RR for the studies of structure, function, folding, and dynamics of hemeproteins has been proven exceptionally fruitful in the past several decades. A number of excellent reviews have been published previously on this subject (Asher, 1981; Carey, 1982; Desbois, 1994; Friedman, 1994; Hildebrandt, 1995; Kincaid, 1999; Kitagawa and Ozaki, 1987; Loehr, 1998; Rousseau and Ondrias, 1983; Schweitzer-Stenner, 1989; Shelnutt et al., 1998; Smith, 1993; Smulevich, 1993; Spiro, 1988; Varotsis and Babcock, 1993; Wang et al., 1996). This section briefly introduces the information content of RR of hemeproteins.
3.1. Porphyrin core vibrational modes The prosthetic heme group in heme proteins is composed of a porphyrin macrocycle with conjugated double bonds and an iron atom that is coordinated by four nitrogen atoms of the porphyrin. Under typical equilibrium conditions, the iron can be in a ferrous (Fe2þ) and a ferric (Fe3þ) state. With excitation in the Soret region, the in-plane ring breathing modes of the porphyrin macrocycle are enhanced because the normal coordinates of
262
Changyuan Lu et al.
these modes coincide with the direction that the porphyrin ring expands during the p-p* electronic transition (the so-called Franck–Condon enhancement). Extensive RR studies of heme proteins and metalloporphyrin model compounds reveal that the totally symmetric in-plane ring breathing modes of the porphyrin macrocycle characterized in the 1300- to 1700-cm1 region are sensitive to the oxidation state, spin state, and the nature of axially coordinated ligands of the heme iron (Hu et al., 1996; Kitagawa and Ozaki, 1987; Spiro and Li, 1988). In particular, the n2 mode, in the region between 1550 and 1600 cm1, is sensitive to the spin state of the heme iron. The n3 mode in the 1475- to 1520-cm1 region is sensitive to both the spin state of the iron and the axial ligand coordination state. The strong n4 mode in the 1350- and 1400-cm1 region is sensitive to the oxidation state of the iron, as well as the p electron density of the porphyrin macrocycle. The frequency and intensity of these Raman lines are in general modulated by the protein environment surrounding the heme and hence provide useful structural information for heme proteins.
3.2. The proximal iron-histidine stretching mode The RR band associated with the iron-histidine stretching mode (nFe-His) of heme proteins was first reported by Kitagawa et al. (1979) for deoxymyoglobin. It was later recognized that this mode is only observable in the fivecoordinate deoxy derivative and is very sensitive to the protein matrix surrounding the histidine (Friedman, 1994; Gilch et al., 1993; Kitagawa, 1988; Rousseau and Ondrias, 1983; Rousseau and Rousseau, 1992; Wang and Spiro, 1998). In monomeric sperm whale myoglobin (swMb), the frequency of nFe-His is 220 cm1. In the tetrameric human Hb (HbA), the iron-histidine bond is constrained by the F helix because of T-state quaternary interactions, hence it exhibits a low nFe-His frequency at 214 cm1 (Kitagawa, 1988). When a ligand, e.g., CO, binds to the heme iron, it pulls the iron into the heme plane, thereby releasing the tension on the iron-histidine bond, generating the so-called relaxed R state. The release of the tension can be indirectly characterized by photolyzing the CO from the CO-bound derivative with a nanosecond laser pulse. In the nanosecond photoproduct of the CO derivative, the iron atom moves out of plane in a direction toward the proximal histidine, leading to a shorter and stronger iron-histidine bond, as reflected by the high frequency of the nFe-His mode at 229 cm1 (15 cm1 higher than that in the equilibrium T state) (Friedman, 1994; Friedman et al., 1982; Rousseau and Ondrias, 1983; Scott and Friedman, 1984). In horseradish peroxidase (HRP), the nFe-His mode shifts up to 244 cm1, because the proximal histidine forms a strong H-bond with a nearby aspartate residue, which withdraws the Ne proton away from the histidine, giving it an imidazolate character (Kitagawa, 1988).
Resonance Raman Studies of Microbial Hemoglobins
263
3.3. Distal axial ligand vibrational modes Distal axial ligand-related modes in heme proteins are usually observable in RR spectra with Soret excitation due to the electronic coupling between the ligand and the heme. Since the first identification of the iron-dioxygen stretching mode (nFe-O2) in the RR spectrum of HbA in erythrocytes by Brunner in 1974, the RR technique has developed into a powerful tool for understanding ligand–protein interactions in heme proteins. The most frequently studied ferrous heme ligands are oxygen-containing diatomic molecules, O2, CO, and NO. The frequencies of specific iron-ligandrelated modes, including Fe-XO stretching (nFe-XO) and Fe-X-O bending (dFe-X-O), are typically located in the low-frequency region (200 to 800 cm1) of the RR spectrum (Yu and Kerr, 1988). Here X represents O, C, and N for O2, CO, and NO, respectively. The X-O stretching mode (nX-O), however, is usually found at much higher frequencies (>1000 cm1) (Yu and Kerr, 1988). The nC-O frequency of gaseous CO is 2143 cm1 (Nakamoto, 1970). When CO is coordinated to the ferrous heme iron, the nC-O frequency shifts to the 1900- to 1970-cm1 range. In general, the s bond formed by donation of two electrons from CO to the heme iron greatly increases the electron density on the heme iron; to stabilize the Fe-C-O moiety, excess electron density on the heme iron is donated from its dp orbital back to the p* orbital of the CO. As a result of this so-called ‘‘p back-bonding’’ effect, the Fe-C-O moiety can be presented by the following two extreme structures:
Fe C Oþ $Fe¼C¼O ðIÞ ðIIÞ
ð14:1Þ
The degree of p back bonding is modulated by the protein matrix surrounding the CO moiety. As a general rule, a positive polar environment destabilizes form I and facilitates the p back-bonding interaction. Consequently, the bond order of the Fe-CO is increased, which is concurrent with a decrease in the bond order of the C-O. On this basis, a well-known inverse correlation between nFe-CO and nC-O frequencies is established, as illustrated in Fig. 14.6 (Feis et al., 1998; Li et al., 1994; Vogel et al., 2000; Yu and Kerr, 1988). In the wild-type swMb, the nFe-CO and nC-O modes are at 512 and 1944 cm1, respectively (Tsubaki et al., 1982). When the distal histidine is mutated to various nonpolar residues, such as Leu or Phe, the nFe-CO frequency decreases and the nC-O frequency increases (Vogel et al., 2000). Accordingly, the data points move toward the lower right corner of the correlation line. However, the data point associated with HRP lies in the upper left corner of the correlation line because the heme-bound CO in this protein is stabilized by H-bonds donated from two positive polar
264
Changyuan Lu et al.
560
L-Fe– −C
≡ O+ ↔ L-Fe = C = O
ν (Fe-CO) (cm−1)
540 520
5C
500 L = His
480
L = Cys
460 1900
1920
1940 1960 ν (C-O) (cm−1)
1980
2000
Figure 14.6 The nFe-CO versus nC-O inverse correlation diagram for a variety of heme proteins and porphyrin derivatives. The bottom two lines are for complexes in which the proximal ligand is histidine (L ¼ His) and thiolate (L ¼ Cys).The top line is for fivecoordinate (5C) CO adducts.The solid circle, square, and up triangle are obtained from horseradish peroxidase, cytochrome c peroxidase, and sperm whale myoglobin, respectively.The corresponding open symbols are obtained from their mutant proteins.
residues, His-42 and Arg-38, in the distal pocket (Feis et al., 1998). The data point for another peroxidase, cytochrome c peroxidase (CcP), which has a similar distal environment for the heme-bound CO, is also located in the upper left corner of the correlation line. When the distal polar residues are mutated to nonpolar residues, the data points of these two peroxidases shift along the inverse correlation line to the lower right corner, like that in swMb (see Fig. 14.6). A similar nFe-CO and nC-O inverse correlation is operative in the cytochrome P450 family of proteins, which contains a thiolate proximal ligand rather than a histidine. The distinct electronic properties of the thiolate modulate the electron density distribution on the Fe-C-O moiety, thereby offsetting the inverse correlation line from the histidine line (see Fig. 14.6). Likewise, five-coordinate CO adducts of ferrous hemes fall on a distinct inverse correlation line that sit above the histidine line. Because of its sensitivity to the identity of the proximal heme ligand, as well as the distal environment, CO has been recognized as a useful structural probe for heme proteins. Analogous p back-bonding effects are also found in NO- and O2-bound ferrous heme complexes. When NO coordinates to the heme iron of swMb, the nN-O frequency shifts down from 1876 cm1 (associated with the gaseous NO) to 1612 cm1 (Nakamoto, 1970; Tomita et al., 1999; Zhao et al., 1994). Similarly, when O2 binds to the heme iron of swMb, the
Resonance Raman Studies of Microbial Hemoglobins
265
nO-O frequency shifts down from 1556 cm1 (associated with the gaseous O2) to 1130 cm1 (Herzberg, 1950; Maxwell et al., 1974). The nO-O frequency of the heme-bound dioxygen is similar to that of superoxide 1 anion (O 2 ) in free solution (1100–1150 cm , depending on solution conditions) (Drago and Corden, 1980), indicating a single bond character of the O-O bond. It is important to note that the nO-O of oxy derivatives of heme proteins with histidine as the proximal ligand were not observed in the RR spectrum until recently (Das et al., 2001; Kincaid, 1999; Lu et al., 2007; Mukai et al., 2002). In contrast to the CO derivative, the correlation of nFe-XO and nX-O in NO and O2 derivatives is ambiguous because of the presence of the lone pair electrons on the X atom, which forces the Fe-X-O moiety to adopt a bent geometry, in contrast to the linear nature of the Fe-C-O moiety. In NO-bound hemeproteins, the nN-O frequency is determined by the electrostatic field surrounding the ligand, as in the CO derivatives, but its inverse correlation with the nFe-NO frequency is modulated by the tilting and the bending angles of the Fe-N-O moiety. As a result, the nFe-NO and nN-O frequencies in six-coordinate NO derivatives cannot be correlated by a simple linear function as that established for the CO derivatives. Likewise, there is no clear correlation between nFe-O2 and nO-O frequencies for sixcoordinate O2-bound hemeproteins (Yu and Kerr, 1988). In addition, both nFe-O2 and nO-O frequencies are insensitive to the electrostatic field surrounding the ligand, plausibly because of the fact that the heme-bound O2 is highly polarized by the heme iron. The ferric complexes of globins, analogous to their ferrous derivatives, bind a variety of small molecules, such as H2O, OH, NO, CN, and N 3. Water is the most common ligand for ferric hemeproteins. In the resting states of swMb or HbA, the ferric heme iron is axially coordinated by an exogenous water ligand in the distal-binding site. The electronic configuration of the heme iron in these aquo-met derivatives is a mixture of sixcoordinate high-spin and low-spin at room temperature. The nFe-H2O mode has never been detected in the RR spectrum of any heme protein. In swMb, when the distal water is deprotonated at elevated pH, the low-spin contribution increases (due to the fact that hydroxide is a stronger field ligand than water for the heme iron), and two nFe-OH modes assigned to the six-coordinate high-spin and low-spin species were identified at 490 and 550 cm1, respectively (Feis et al., 1994). In the high-spin species, the heme-bound hydroxide accepts an H-bond from the distal His-64, which is not present in the low-spin species. In contrast to swMb, the heme-bound hydroxide in HRP donates an H-bond to its distal His-42, and only the low-spin species with the nFe-OH mode at 503 cm1 is present at room temperature (Howes et al., 1997).
266
Changyuan Lu et al.
3.4. Potential pitfalls The application of RR spectroscopy in studying heme proteins, although informative, could be complicated by a number of potential pitfalls. The most commonly encountered obstacles, along with their possible resolutions, are listed next. 1. The fluorescence background resulting from either intrinsic chromophors or impurities because of imperfect purification may mask RR bands, making spectral interpretation a nontrivial task. The fluorescence background from intrinsic chromophors may be avoided by careful selection of the excitation wavelength, whereas that resulting from impurities may be eliminated by additional purification procedures. 2. The combination of RR spectroscopy and site-specific mutagenesis has been shown to be important in revealing the structural and functional role of a specific amino acid residue in a given heme protein. However, in some cases, site-directed mutagenesis may perturb the structural regions in the vicinity of the mutation site or remote from it, hence additional measurements may be required to confirm the integrity of the overall protein structure of the protein of interest. 3. Ligand-associated vibrational modes may overlap with similar modes associated with alternative protein conformations or vibrational modes of the heme, which may complicate the spectral analysis. To resolve these issues, additional isotope substitution experiments may be used to clarify the identity of the vibrational modes. 4. Photo-induced artifacts, such as photodissociation of CO in the CO derivatives or photoreduction of the ferric derivatives, may present a potential problem, leading to erroneous interpretations. Laser powerdependent studies may be carried out to avoid these problems. 5. For most heme proteins, O2 is the most important physiological ligand. It can be readily prepared by the addition of O2 to the dithionite-reduced deoxy protein; however, the O2 complex thus produced is often not stable because of its fast autooxidation rate, especially in the presence of a high concentration of the oxidation products of dithionite. A desalting column can be used to separate the oxidation by-products of dithionite from the protein following the oxygen-binding reaction to minimize autooxidation of the O2 complexes prior to RR measurements.
4. Structures and Functions of Microbial Hemoglobins Although crystallographic studies play an indispensable role in advancing our knowledge of the structure–function relationships of microbial Hbs, it is sometimes a concern as to whether a crystal structure reflects the true
Resonance Raman Studies of Microbial Hemoglobins
267
solution structure, not a metastable structure trapped in the crystal lattice. In addition, it remains a nontrivial task to obtain the crystal structures of a Hb of interest in all desired oxidation and coordination states, despite the fact that the crystallographic techniques have been substantially improved over the years. In this regard, it is important to complement crystallographic studies with other spectroscopic approaches, such as RR, ultraviolet-visible spectroscopy, circular dichroism, Fourier transform infrared spectroscopy, electron paramagnetic resonance, or NMR, that are applicable for samples of aqueous nature. Among these techniques, RR spectroscopy is especially versatile and informative. The following section reviews the structural and functional studies of the microbial Hbs from M. tuberculosis and C. jejuni with RR spectroscopy.
4.1. The TrHb-I from M. tuberculosis (TrHbN) Mycobacterium tuberculosis is a Gram-positive obligate aerobic bacterium that grows most successfully in tissues with high oxygen content, such as the lungs. In most healthy individuals, the initial infection by the tubercule bacilli is contained by the immune system, forcing the bacteria to enter latency for decades with possible reactivation later in life. M. tuberculosis possesses two Hbs, trHbN and trHbO, which belong to the trHb-I and trHb-II family of proteins, respectively (see Fig. 14.2) (Couture et al., 1999; Mukai et al., 2002; Yeh et al., 2000). The extent of amino acid identity between these two proteins is only 18%. In aerobic cultures of M. bovis BCG, the nonpathogenic model for M. tuberculosis, a steady level of trHbO is detected throughout the growth phase, whereas trHbN is only detected after cells have reached the stationary phase (Fig. 14.7) (Couture et al., 1999; Mukai et al., 2002). Moreover, when expressed in E. coli, trHbO is mainly associated with membranes, while trHbN remains mostly cytoplasmic (Pathania et al., 2002). The distinctive temporal and spatial distributions of these two Hbs suggest that they perform different functions. TrHbN is a homodimer with a molecular mass of 14.4 kDa per monomer (Couture et al., 1999). It contains 136 amino acids. The B10, E7, E10, and E11 positions in trHbN are occupied by Tyr-33, Leu-54, Lys-57, and Gln-58, respectively (Milani et al., 2001). In the crystal structure of the O2-bound trHbN (Fig. 14.8A), the heme-bound O2 accepts H-bonds from the B10Tyr, which in turn accepts a H-bond from the E11Gln. The ferrous ligand-free derivative of trHbN has a five-coordinate highspin configuration as judged by the n3 and n4 at 1471 and 1356 cm1, respectively (Fig. 14.9) ( Yeh et al., 2000). The nFe-His mode was found at 226 cm1. It is significantly higher than that of HbA (214 cm1) and swMb (220 cm1), as listed in Table 14.1, indicating a staggered orientation of the imidazole ring of the proximal histidine with respect to the pyrrole nitrogen atoms, in contrast to the eclipsed orientation in HbA and swMb
268
Changyuan Lu et al.
10
A580
1 0.1 trHbO 0.01
trHbN
0.001 0
50
100 Time (h)
150
200
Figure 14.7 Growth curve of M. bovis BCG cells and associated temporal expression patterns of trHbN and trHbO. Data are adapted from Mukai et al. (2002).
A
B B10
G8
E11
trHbN
B10
CD1
trHbO
C
D B10
B10
E7
G8 E7
trCtb
Cgb
Figure 14.8 Distal heme pockets of the oxy derivative of trHbN (A) and ferric cyanide derivatives of trHbO (B), trCtb (C), and Cgb (D). PDB codes for trHbN, trHbO, and trCtb are 1IDR,1NGK, and 2IG3, respectively.The dotted line represents H-bonds.The possible rotation of the His side chain in trCtb as discussed in the text is indicated by the arrow.
(Samuni et al., 2003). In the CO derivative, two nFe-CO modes are identified at 500 and 535 cm1 based on the 12C16O-13C18O isotope difference spectrum shown in Fig. 14.10A ( Yeh et al., 2000). These two modes are associated with two nC-O modes at 1960 and 1916 cm1, respectively. The two pairs of nFe-CO/nC-O modes at 500/1960 and 535/1916 cm1 are
269
Resonance Raman Studies of Microbial Hemoglobins
ν4
TrHbN
1356
ν3 ν2 1471
Deoxy pH 7.5
1561 1501 1579 1479
Ferric pH 7.5
pH 10.5
900
1622 1564
x
1100
1300
1500
1700
Raman shift (cm−1)
Figure 14.9 High-frequency resonance Raman spectra of the ferrous deoxy derivative (pH 7.5) and ferric derivative (pH 7.5 and 10.5) of wild-type trHbN as indicated. The peak indicated as‘‘x’’ is a laser plasma line. Data adapted fromYeh et al. (2000).
assigned to ‘‘open’’ and ‘‘closed’’ conformations, respectively. The presence of the two alternative conformations in trHbN reflects its structural plasticity that is not observed in mammalian globins. In the nC-O-nFe-CO inverse correlation line, the two data points are located at the lower right corner and upper left corner, respectively (see Figs. 14.10B). Mutation of the B10Tyr to Phe converts trHbN to a single open structure, as indicated in Fig. 14.11A, suggesting that the B10Tyr forms a H-bond with the CO in the wild-type protein. Intriguingly, mutation of the E11Gln to Leu also converts trHbN to a single open conformation (unpublished data), indicting that without the E11Gln, the B10Tyr by itself cannot form an H-bond with the heme-bound CO. Data hence suggest that the B10Tyr in wild-type trHbN is positioned by E11Gln in a proper stereo-orientation to stabilize the heme-bound ligand via an H-bond. In the oxy derivative, the nFe-O2 mode was found at 560 cm1, based on the 16O2-18O2 isotope difference spectrum shown in Fig. 14.12 (Couture et al., 1999). Although rarely seen in RR spectra of heme proteins with a proximal histidine ligand, the nO-O mode was identified at 1139 cm1, indicating superoxide character of the heme-bound dioxygen (unpublished data). When the B10Tyr is mutated to Leu, the nFe-O2 frequency shifts to 570 cm1 (see Fig. 14.12), a frequency similar to that of swMb (569 cm1) (Yeh et al., 2000). In swMb, the heme-bound dioxygen is stabilized by the
Table 14.1 Coordination and spin states of various derivatives of differing Hbs and their ligand-associated vibrational modes
trHb I II III sdHb FHb swMbb HRPb a b c d e
trHbNb trHbOb trCtbd Cgbc Hmpb
Ferrica
Ligandsa
Deoxya
nFe-His
nFe-O2
nO-O
nFe-CO
6CHSþLS 6CHSþLS 6CHSþLS 5CHS 5CHS 6CHSþLS 5CHS
H-H2O H-H2O H-H2O H H H-H2O H
5CHS 5CHS 5CHS 5CHS 5CHS 5CHS 5CHS
226 226 226 251 244 220 240
562 559 542 554 — 569 562
1139c 1140 1133 nde — nd nd
500, 535 525 515 492, 529 494, 535 512 531, 541
6C, 5C, HS, and LS stand for six-coordinate, five-coordinate, high-spin, and low-spin heme iron, respectively; H stands for histidine. See references in Egawa and Yeh (2005). To be published. From Wainwright et al. (2006). Not detected.
271
Resonance Raman Studies of Microbial Hemoglobins
A
B 550
wt B10 Y→L
540 530 520 510 500
wt 12C16O-13C18O
489 518
200
560
trHbN-CO νFe-CO(cm−1)
νFe-CO 500 534
490
trHbNC HmpC CgbC trHbO swMb trCtb trHbNO HmpO CgbO 1890 1905 1920 1935 1950 1965 νC-O (cm−1)
B10 Y→L 12C16O-13C18O
500 800 Raman Shift (cm−1)
Fe−–C≡O+ ↔ Fe=C=O
1100
Figure 14.10 Low-frequency resonance Raman spectra of the CO derivative of the wild type (wt) and B10Tyr!Leu mutant of trHbN at pH 7.5 and corresponding 12 16 C O-13C18O difference spectra (A), and the nFe-CO versus nC-O inverse correlation line for the various microbial Hbs and sperm whale myoglobin (B). Subscripts ‘‘O’’and ‘‘C’’ stand for open and closed conformations, respectively. (a) Data adapted from Yeh et al. (2000).
E7His, mostly via an H-bond with its terminal oxygen atom; in addition, the O-O bond of the heme-bound O2 is highly polarized by the heme iron. Accordingly, the nFe-O2 frequency is relatively insensitive to the E7His mutations (Hirota et al., 1996). The 10-cm1 shift in the nFe-O2 frequency in the B10Tyr!Leu mutant of trHbN suggests that the B10Tyr interacts with the proximal oxygen atom (instead of the terminal oxygen atom) of the heme-bound O2 in the wild-type protein (Yeh et al., 2000). This proposal is supported by the observation that mutation of the B10Tyr to Phe increases the O2 off rate by a factor of 200. Mutation of the E11Gln to Leu also increases the off rate, but by a factor of only 20 (unpublished data). These observations are consistent with the scenario that the E11Gln does not interact directly with the heme-bound ligand, instead it positions the B10Tyr in a more favorable stereo-orientation for forming H-bond(s) with the ligand, as that proposed for the CO derivative. The H-bonding network in trHbN may account for its unusually high O2 affinity as listed in Table 14.2. At neutral pH, the ferric derivative exhibits a mixture of six-coordinate high-spin and six-coordinate low-spin configurations as indicated by the n2 modes at 1561 and 1579 cm1, respectively (see Fig. 14.9), signifying a water-bound ferric heme species (Yeh et al., 2000). When the pH is raised to pH 10.5, the contribution from the six-coordinate low-spin species
272
Changyuan Lu et al.
νFe-CO (cm−1)
A
560 WTC
540 520
WTC B10
500
trHbN
swMb B10 B10/E7 E7
Cgb
480 B
560
WTO E11
WTO
E7
νFe-CO (cm−1)
trCtb 540 520 500
B10 B10/E7 E7 WT CD1 G8 CD1/G8 trHbO WT B10
G8
480
1900
1920
1940 1960 1980
νC-O (cm−1)
Figure 14.11 The nFe-CO versus nC-O inverse correlation line for the wild type (WT) and mutants of trHbN (squares)/Cgb (circles) (A) and trCtb (squares)/trHbO (circles) (B). The shaded up triangle is a reference data point from sperm whale myoglobin. The corresponding solid and open symbols are from wild-type proteins and mutants, respectively. Arrows indicate the direction along which data shift in response to mutations in the distal residues from polar to apolar substituents. B10, E7, E11, CD1, and G8 represent B10Tyr!Phe, E7His!Leu, E11Gln!Leu, CD1Tyr!Phe, and G8Trp!Phe mutants, respectively.
increases because of the deprotonation of the heme-bound water. On the basis of H216O-D218O isotope substitution experiments, two nFe-OH modes are identified at 454 and 561 cm1 at pH 10.5 (Yeh et al., 2000). When the B10Tyr is mutated to Phe, the line at 454 cm1 totally disappears, leaving only a high-frequency line at 552 cm1. Accordingly, the 454-cm1 mode was assigned to a closed conformation in which the hydroxide is stabilized by an H-bond donated from the B10Tyr, whereas the 561-cm1 mode was identified as an open conformation in which the H-bond is absent (Yeh et al., 2000). The presence of two alternative conformations, reminiscent of the CO derivative, reflects the plasticity of the distal heme pocket of trHbN. Because the expression of trHbN is enhanced greatly during the stationary phase in aerobic cultures and because the environment that bacteria experience in the stationary phase in vitro is similar to that in the latent phase found in vivo, we proposed that the physiological role of trHbN is to protect the bacilli against NO produced by the host macrophage during latency
273
Resonance Raman Studies of Microbial Hemoglobins
νFe-O2
trHbN-O2
560 wt 570
B10 Y→L 542
wt 16O -18O 2 2
545
B10 Y→L 16O -18O 2 2
200
500
800
1100
Raman shift (cm−1)
Figure 14.12 Low-frequency resonance Raman spectra of the O2 derivative of the wild type (wt) and the B10Tyr!Leu mutant of trHbN at pH 7.5 and corresponding 16 O2 -18O2 difference spectra. Data adapted fromYeh et al. (2000) and Couture et al. (1999).
(Couture et al., 1999), possibly via the NO dioxygenase (NOD) reaction as illustrated (Gardner, 2005):
Fe2þ -O-Oþ N¼O ! ½Fe3þ-OONO ! Fe3þ þ NO 3
ð14:2Þ
The NOD activity of trHbN is confirmed by in vitro experiments, which show that titration of oxygenated trHbN with NO results in stoichiometric oxidation of the protein along with nitrate formation; in addition, the second-order rate constant between the oxy derivative and NO is 745 mM1s1, 20-fold faster than swMb (Ouellet et al., 2002). These in vitro data are supported by in vivo results, which show that disruption of the glbN gene encoding trHbN in M. bovis BCG causes a dramatic reduction in the NO-consuming activity of the stationary phase cells; this activity could be fully restored by complementing the knockout cells with wild-type glbN, but not the B10 mutant of glbN (Ouellet et al., 2002). Taken together, data suggest that the H-bonding network involving the heme-bound O2, B10Tyr and E11Gln may promote the O-O bond cleavage of the heme-bound peroxynitrite intermediate, Fe3þ-OONO ( Yeh et al., 2000). In addition, the plasticity of the distal heme pocket may further accelerate the turnover of the reaction by facilitating the release of the product nitrate.
Table 14.2 O2 and CO on- and off-rate constants of various Hbs, as well as associated equilibrium constants (KO2 and KCO) calculated from rate constants
b c d e f g
koff(O2) (s1)
KO2 (mM1)
kon(CO) (mM1s1)
koff(CO) (s1)
KCO (mM1)
KO2/ KCO
444 93 22 222 172
5.5a 0.01a
0.005b 0.0015c 0.0041c 0.0048a 0.024a
1100 6.7 2.5 13.3 2208 27.5 386 77.8 26.8
0.4 14 8.8 17 0.08
trHb I II
trHbN trHbO
71a 0.13a
III sdHb
trCtb Cgb
0.91d 150e
0.16a 0.0014c 0.0058c 0.0041d 0.87a
FHb
Hmp
38f
0.44f
86
14g
12g
1.2
swMb a
kon(O2) (mM1s1)
To be published. From Couture et al. (1999). From Ouellet et al. (2003). From Lu et al. (2007). From Farres et al. (2005). From Gardner et al. (2000). From Quillin et al. (1995) and Springer et al. (1989).
0.064a 53a 0.66a 22f 1.4f 0.51g
0.057f 0.018f 0.019g
0.22 1.1 0.04
Resonance Raman Studies of Microbial Hemoglobins
275
4.2. The TrHb-II from the M. tuberculosis (TrHbO) The TrHb-II from M. tuberculosis, trHbO, has 128 amino acid residues and a molecular mass of 14.9 kDa (Mukai et al., 2002). Under solution conditions, trHbO exists as a mixture of monomers and dimers (Ouellet et al., 2003). The monomer and dimer equilibrium is sensitive to the ionic strength of the solution. In the presence of a high concentration of salt, the protein is mostly in a monomeric state, suggesting that the dimer interface is stabilized by salt bridges. Under the solution conditions used in the RR studies presented here, trHbO exists primarily as a dimer (Ouellet et al., 2003). In contrast, in the crystalline state, trHbO displays as a dodecamer with six pairs of asymmetric dimeric units (Milani et al., 2003). The B10 and E10 positions in trHbO are occupied by Tyr-23 and Arg47, respectively. The E7 and E11 positions are both occupied by apolar amino acid residues, Ala-44 and Leu-48, respectively, in sharp contrast to the TrHb-I family of proteins, precluding ligand stabilization by these two residues. The most intriguing feature of trHbO is that the CD1 residue is a Tyr instead of Phe, which is highly conserved in most other Hbs discovered to date. Furthermore, a covalent bond between the phenyl oxygen of the B10Tyr and the CE2 of the CD1Tyr was observed in six subunits of the dodecamer (see Fig. 14.8B), but not in the other six subunits, although in the latter the aromatic side chain groups of the B10Tyr and the CD1Tyr are in very close contact and in a similar orthogonal orientation (Milani et al., 2003). A Trp residue at the G8 position, which is highly conserved in the trHb-II and trHb-III family of proteins, also plays a critical role in ligand stabilization. In the crystal structure of the ferric cyanide-bound derivative of trHbO, the heme-bound cyanide accepts H-bonds from the CD1Tyr as well as the G8Trp, but not the B10Tyr (see Fig. 14.8B). The RR spectrum of the ferric derivative of trHbO shows a mixture of six-coordinate high-spin and low-spin configurations, indicating a typical aquo-met heme species (Mukai et al., 2002). The ferrous derivative of trHbO is a five-coordinate high spin, with the nFe-His mode at 226 cm1, indicating a staggered geometry of the proximal histidine with respect to the pyrrole nitrogen atoms of the porphyrin ring (Samuni et al., 2004). The CO derivative of trHbO exhibits only one nFe-CO mode at 525 cm1 (see Table 14.1), which is assigned to a closed conformation (Mukai et al., 2002). It represents the only trHb, which exhibits a single conformation locked in the closed state (see Fig. 14.10b). Mutation of the B10Tyr to Phe does not affect the position of the data point (see Fig. 14.11B), indicating that the B10Tyr does not play any significant role in ligand stabilization. However, mutation of CD1Tyr or G8Trp to Phe causes the data point to shift down to the middle of the correlation line, whereas the CD1/G8 double mutation causes it to further shift to the right lower corner of the correlation line, indicating that in the wild-type protein the heme ligand is
276
Changyuan Lu et al.
stabilized by the CD1Tyr and G8Trp in a synergetic fashion (unpublished results). It is important to note that the G8Trp!Phe mutant exhibits an additional open conformation (see Fig. 14.11B), suggesting that the H-bond between the G8Trp and the heme-bound ligand helps position the ligand to accept an additional H-bond from the CD1Tyr. The nFe-O2 frequency of the oxy derivative of trHbO was identified at 559 cm1 and the typically RR silent nO-O mode was detected at 1140 cm1 (Mukai et al., 2002). Mutation of the B10Tyr to Phe does not affect these two modes (Mukai et al., 2002; Ouellet et al., 2003). In contrast, mutation of the CD1Tyr to Phe causes a slight downshift in the nFe-O2 frequency to 556 cm1, which is accompanied by disappearance of the nO-O mode from the RR spectrum (Mukai et al., 2002). Data are consistent with the scenario that the heme-bound ligand in trHbO is stabilized by H-bonds donated by both the CD1Tyr and the G8Trp as that suggested for CO derivatives (see Fig. 14.11B). This unique distal interaction results in an extremely slow O2 off rate (0.0014 s1), which is 10,000-fold slower than that of swMb, making it one of the slowest oxygen-releasing Hbs reported to date (see Table 14.2). On rates for O2 and CO are also very slow (Ouellet et al., 2003), 100- and 50-fold slower than those of swMb, respectively (see Table 14.2). When the CD1Tyr is mutated to Phe, on rates increase by a factor of 25 and 77 for O2 and CO, respectively (Ouellet et al., 2003), demonstrating the important role of the CD1Tyr in controlling ligand entry into the heme distal site. TrHbO and trCtb (see later) represent the only two Hbs discovered to date that preferentially bind O2 over CO. The intrinsic affinity of the heme prosthetic group for O2 in free solution is roughly 20,000-fold weaker than that for CO (Springer et al., 1994). In swMb, this ratio is reduced to 25 because the electrostatic environment created by the distal E7His encourages the binding of O2 (Spiro and Kozlowski, 2001). In trHbO, the affinity for O2 is 14 times stronger than that of CO, presumably because of its novel distal environment. The tight distal pocket of trHbO congested with the aromatic side chain groups of the CD1Tyr, G8Trp, B10Tyr, and B9Phe may disfavor the linear Fe-C-O moiety. In contrast, the hemebound O2 is stabilized by two H-bonds provided by the CD1Tyr and the G8Trp in an optimized geometry. The structurally confined distal heme pocket may also account for the high geminate CO recombination yield observed in nanosecond flash photolysis studies (Ouellet et al., 2003). The physiological function of trHbO remains to be determined. In vivo work shows that overexpression of trHbO in E. coli recombinant cells stimulates cellular respiration and oxygen uptake in wild-type cells, but not in terminal oxidase-deficient mutant cells, suggesting a direct interaction between trHbO and terminal oxidases (Pathania et al., 2002). TrHbO hence may function as an oxygen sequester in M. tuberculosis to sustain
Resonance Raman Studies of Microbial Hemoglobins
277
aerobic metabolism. Although trHbO has a relatively low oxygen affinity and slow O2 on and off rates (see Table 14.2), trHbO does have the advantage of its localization in the cell membrane. In addition, membrane binding may modulate the ligand-binding properties of trHbO, making it a better oxygen sequester. The oxygen sequestering function of trHbO may also be used to facilitate the NOD reaction carried out by trHbN by providing O2 required for the reaction.
4.3. The TrHb-III from the C. jejuni (TrCtb) Campylobacter jejuni is a Gram-negative bacterium present in the gut of many food-supply animals and birds. It is an obligate microaerophile, meaning oxygen is necessary for growth yet is also toxic when at atmospheric concentrations. Like M. tuberculosis, C. jejuni comprises two Hbs, trCtb and Cgb, which belong to the TrHb-III and sdHb groups of Hbs, respectively (Elvers et al., 2004; Wainwright et al., 2005, 2006). It has been demonstrated that these two Hbs are not required for the survival of the bacterium in air (Wainwright et al., 2005, 2006). In addition, the expression of Cgb was found to be strongly and specifically induced by nitrosative stress (Elvers et al., 2004). Along the same lines, a Cgb knockout mutant of C. jejuni was shown to be hypersensitive to reactive nitrogen species (Elvers et al., 2004). As such, Cgb, like trHbN in M. tuberculosis, has been proposed to protect the bacterium from the toxic effects of NO by means of an NOD reaction. Although the trCtb knockout mutant of C. jejuni did not display any sensitivity to nitrosative stress, the expression of trCtb can be induced by NO donors (Wainwright et al., 2005). More importantly, the O2 consumption rate of trCtb knockout mutant cells showed a 50% reduction as compared to wild-type cells, suggesting its involvement in regulating the flux of O2 into and within the cell (Elvers et al., 2005). TrCtb is a monomeric protein with 127 amino acid residues and a molecular mass of 14.1 kDa. The B10, E7, E10, and E11 positions of trCtb are occupied by Tyr-19, His-46, Lys-49, and Ilu-50, respectively. Like trHbO, the G8 position is occupied by a Trp residue. In the crystal structure of the cyanide-bound ferric derivative, the cyanide accepts H-bonds donated from the B10Tyr and the G8Trp (see Fig. 14.8C) (Nardini et al., 2006). The ferric state of trCtb is in an aquo-met form, with the heme in a mixture of six-coordinate high- and low-spin configurations (Wainwright et al., 2006). The ferrous derivative has a five-coordinate high-spin configuration with a single histidine as the axial ligand. The nFe-His mode was found at 226 cm1 (Wainwright et al., 2006), which is similar to that of most other trHbs (see Table 14.1), indicating a staggered orientation of the proximal histidine with respect to the pyrrole nitrogen atoms of the
278
Changyuan Lu et al.
porphyrin ring. In the CO derivative, nFe-CO and nC-O modes are located at 515 and 1936 cm1, respectively (Wainwright et al., 2006). The data point sits in the middle of the correlation line as shown in Fig. 14.10b. In trHbN and trHbO, mutation in the key distal polar residues causes the data point to shift along the correlation line to the lower right corner (see Fig. 14.11) as a consequence of the reduction of the electrostatic potential surrounding the heme-bound CO. In contrast, mutation of either the B10Tyr to Phe or the E7His to Leu in trCtb causes data points to shift toward the upper left corner of the inverse correlation line (see Fig. 14.11B), indicating a higher electrostatic potential of the protein environment surrounding the CO (Lu et al., 2007). How can the removal of a positive distal residue result in an increase in the electrostatic potential? We hypothesize that in the wild-type trCtb, the E7His forms an H-bond with the B10Tyr, which prevents either residue from forming an H-bond with the CO ligand. Mutation in one of the two residues releases the structural constraint on the other, allowing it to form an H-bond with the CO. In addition to the closed form with the extremely high nFe-CO, the E7His mutant can also adopt a wild-type-like conformation, suggesting that the B10Tyr may exist in two alternative conformations in the absence of the H-bond donated from the E7His. It is noticeable that the imidazole side chain of the E7 residue in the crystallographic structure of the ferric cyanide-bound derivative is not in a correct orientation for forming an H-bond with the B10 Tyr (see Fig. 14.8C), suggesting that reduction of the heme iron may induce the rotation and repositioning of the His side chain. The nFe-O2 and nO-O modes of the oxy derivative of trCtb were found at 542 and 1133 cm1, respectively (Lu et al., 2007). Like other trHbs, the appearance of the nO-O mode in the RR spectrum is attributed to an H-bonding network between the heme-bound O2 and the distal residues in its close proximity (Lu et al., 2007). This unique H-bonding interaction leads to a very slow O2 off rate (0.0041 s1) (Lu et al., 2007). Mutation of the B10Tyr to Phe causes the off rate to increase to 0.0088 s1, whereas the single mutation of E7His to Leu or double mutation of the E7His/B10Tyr leads to a decrease in the off rate to 0.0003 and 0.0028 s1, respectively. Data are consistent with the proposal that the heme-bound dioxygen is stabilized by accepting H-bonds from the G8Trp and B10Tyr, and the additional H-bond between E7His and B10Tyr plays an important role in preventing overstabilization of the heme-bound dioxygen. Like trHbO, the distal heme pocket of trCtb is congested with bulky aromatic residues, including the G8Trp, B9Phe, B10Tyr, CD1Phe, and E7His, which may account for the slow on rate of O2 and CO, as well as the preferential binding of O2 versus CO (see Table 14.2). The extremely high oxygen affinity of trCtb, mostly resulting from the unusually slow off rate, makes it unlikely to function as an oxygen transporter; however, the distal heme environment of trCtb is surprisingly similar to that of cytochrome c
279
Resonance Raman Studies of Microbial Hemoglobins
peroxidase, suggesting a role of trCtb in performing a peroxidase- or P450type of oxygen chemistry (Lu et al., 2007).
4.4. The sdHb from C. jejuni (Cgb) The sdHb from C. jejuni, Cgb, is a monomer with 140 amino acid residues and a molecular mass of 16.08 kDa (Elvers et al., 2004). The B10, E7, E10, and E11 positions of Cgb are occupied by Tyr-28, Gln-52, Ala-55 and Leu56, respectively. In the crystal structure of the cyanide-bound ferric derivative of Cgb (to be published), the heme-bound cyanide is stabilized by an H-bond donated by the B10Tyr, which in turn accepts an H-bond from the E7Gln (Fig. 14.13). At neutral pH, the RR spectrum of the ferric derivative of Cgb exhibits a five-coordinate high-spin configuration, indicating that the distal heme ligand-binding site is empty, in contrast to the water-bound heme in most of the other Hbs discovered to date. The ferrous derivative of Cgb also displays a five-coordinate high-spin configuration, with a single histidine ligand coordinated to the heme iron. The Fe-His mode is located at 251 cm1 (see Table 14.1), which is much higher than that of swMb (220 cm1), but similar to those of heme peroxidases, e.g., 244 cm1 for HRP and 248 cm1 for CcP (Dasgupta et al., 1989; Smulevich et al., 1996). The high frequency of the nFe-His mode in peroxidases has been attributed to the imidazolate character of the proximal histidine due to the presence of a strong H-bond between the histidine and a nearby negatively charged amino acid side chain (Smulevich et al., 1996). Intriguingly, the ferric derivative of CcP, like Cgb, has a five-coordinate high-spin configuration (Anni and Yonetani, 1988;
B10
E7
Cgb
Figure 14.13 Crystal structures of Cgb. B and E helices are labeled in green and yellow, respectively. Dotted lines indicate H-bonding interactions.
280
Changyuan Lu et al.
Smulevich, 1998; Yonetani and Anni, 1987). It is believed that the strong proximal iron-histidine bond in CcP pulls the iron out of the heme plane, and the repulsive force exerted by the pyrrole nitrogen atoms of the porphyrin ring on the water prevents it from binding to the heme iron (Anni and Yonetani, 1988; Smulevich, 1998; Yonetani and Anni, 1987). Accordingly, we attribute the five-coordinate high-spin nature of the ferric heme in Cgb to the same origin. Along this line, an extended H-bonding network, involving His-Glu-Tyr, was recognized in the crystallographic structure of Cgb (see Fig. 14.13), which pulls a proton from the proximal histidine, leaving it an imidazolate character. This H-bonding network is highly conserved in sdHbs and FHbs, hinting that it may play important structural and functional roles in these Hbs. In the O2 derivative of Cgb, the nFe-O2 was found at 554 cm1, which is low compared to swMb, but is comparable to those of other microbial Hbs (see Table 14.1), suggesting a possible H-bonding interaction between the heme-bound O2 and the B10Tyr and/or E7Gln residues. In the CO derivative, two pairs of nFe-CO/nC-O modes assigned to open and closed conformations were found at 492/1963 and 529/1914 cm1, respectively (see Table 14.1). The presence of the two alternative conformations in Cgb demonstrates the plasticity of its ligand-binding pocket. When the B10Try is mutated to Phe, the two conformations convert to one associated with the nFe-CO/nC-O at 511/1931 cm1, which locates in the middle of the inverse correlation line (see Fig. 14.11a), suggesting that the heme-bound CO in the wild-type protein is stabilized by an H-bond donated from the B10Tyr. When the E7Gln is mutated to Leu, only an open conformation was observed, indicating that the B10Tyr by itself cannot form an H-bond with CO and that the E7Gln plays an important role in positioning the B10Tyr for ligand stabilization in the wild-type protein. This proposal is consistent with the crystallographic structure shown in Fig. 14.13 and is in line with the observation that only open conformation was observed when both B10Tyr and E7Gln are mutated (see Fig. 14.11A). The ability of Cgb to adopt a peroxidases-like structure led us to propose that the protein architecture is designed to perform oxygen chemistry rather than oxygen transport. Cgb has been proposed to protect C. jejuni from NO attack, plausibly by carrying out the NOD reaction. We believe that the proximal imidazolate ligand offers an electronic push and that the distal B10Tyr and the E7Gln provide an electronic pull to activate the O-O bond of the heme-bound peroxynitrite intermediate, thereby facilitating the NOD reaction. It is important to note that all the Hbs implicated in NOD chemistry so far, including trHbN, Cgb, and Hmp (a FHb from E. coli ), exhibit two conformations in their CO derivatives, implying that the plasticity of the distal pockets may be important for the NOD reaction.
281
Resonance Raman Studies of Microbial Hemoglobins
5. Closing Remarks Microbial Hbs exhibit a wide spectrum of structural features and ligand-binding properties. In mammalian globins, the oxygen transport function is believed to be regulated by the motion of the E7His, the so-called ‘‘histidine gate,’’ which swings in and out of the distal pocket in response to ligand loading and release. In contrast, oxygen binding and release in the microbial Hbs rely on the intricate distal H-bonding interactions involving the B10Tyr, E7Gln/His, E11Gln, CD1Tyr, and/or G8Trp. In trHbN (a trHbI), the B10Tyr is positioned by the E11Gln in an appropriate stereo-orientation for stabilization of the heme-bound ligand via an H-bond. Likewise, the B10Tyr in Cgb (a sdHb) is positioned by the E7Gln for proper ligand stabilization. In contrast, in trHb-II and trHb-III groups of Hbs, ligand stabilization is controlled primarily by the H-bond donated from the G8Trp to the ligand. In trHbO (a trHb-II), ligand stabilization is further enhanced by an additional H-bond donated from the CD1Tyr to the ligand, whereas in trCtb (a trHb-III), an additional H-bond between B10Tyr and E7His prevents overstabilization of the heme ligands by the B10Tyr. Intriguingly, as shown in Fig. 14.14, the electrostatic potential of the heme ligand-binding pocket of microbial Hbs (as indicated by the nFe-CO) appears to correlate well with their O2- and CO-binding rate constants. It is noticeable that all the Hbs implicated in NOD functionality, including trHbN, Cgb, and Hmp, exhibit low electrostatic potential (here only the open conformation is considered, assuming that the open and closed conformations are interconverting and only the open conformation can uptake ligands) and fast on rates, whereas trHbO and trCtb, which
8.0 log (kon)
7.0
kon(O2)
Cgb Hmp
6.0 5.0
trHbN
kon (CO)
4.0
Mb trCtb trHbO
3.0 490
500
510
520
530
νFe-CO (cm−1)
Figure 14.14 A plot of log(kon) of the various microbial globins versus their associated nFe-CO, where kon is the O2 or CO on rates, as indicated. For trHbN, Cgb, and Hmp, only the nFe-CO of the corresponding open conformations are plotted.
282
Changyuan Lu et al.
comprise a rigid and congested distal pocket, exhibit high electrostatic potential and slow on rates. Although the physiological functions of these microbial Hbs remain to be investigated further, it is clear that the unique structural features of the microbial Hbs point to functions other than oxygen transport.
ACKNOWLEDGMENTS This work was supported by National Institute of Health Research Grant HL65465 to S.-R.Y. We thank Dr. Denis L. Rousseau for many invaluable discussions.
REFERENCES Anni, H., and Yonetani, T. (1988). The coordination and spin states of yeast cytochrome c peroxidase and their implication to peroxidase mechanism. Prog. Clin. Biol. Res. 274, 437–449. Asher, S. A. (1981). Resonance Raman spectroscopy of hemoglobin. Methods Enzymol. 1, 371–413. Brunner, H. (1974). Naturwissenschaften 61, 129–130. Carey, P. R. (1982). In ‘‘Biochemical applications of Raman and resonance Raman spectroscopies.’’ New York: Academic Press, New York. Couture, M., Yeh, S. R., Wittenberg, B. A., Wittenberg, J. B., Ouellet, Y., Rousseau, D. L., and Guertin, M. (1999). A cooperative oxygen-binding hemoglobin from Mycobacterium tuberculosis. Proc. Natl. Acad. Sci. USA 96, 11223–11228. Das, T. K., Couture, M., Ouellet, Y., Guertin, M., and Rousseau, D. L. (2001). Simultaneous observation of the O–O and Fe–O2 stretching modes in oxyhemoglobins. Proc. Natl. Acad. Sci. USA 98, 479–484. Dasgupta, S., Rousseau, D. L., Anni, H., and Yonetani, T. (1989). Structural characterization of cytochrome c peroxidase by resonance Raman scattering. J. Biol. Chem. 264, 654–662. Desbois, A. (1994). Resonance Raman spectroscopy of c-type cytochromes. Biochimie 76, 693–707. Drago, R. S., and Corden, B. B. (1980). Spin-pairing model of dioxygen binding and its application to various transition-metal systems as well as hemoglobin cooperativity. Acc. Chem. Res. 13, 353–360. Egawa, T., and Yeh, S. R. (2005). Structural and functional properties of hemoglobins from unicellular organisms as revealed by resonance Raman spectroscopy. J. Inorg. Biochem. 99, 72–96. Elvers, K. T., Turner, S. M., Wainwright, L. M., Marsden, G., Hinds, J., Cole, J. A., Poole, R. K., Penn, C. W., and Park, S. F. (2005). NssR, a member of the Crp-Fnr superfamily from Campylobacter jejuni, regulates a nitrosative stress-responsive regulon that includes both a single-domain and a truncated haemoglobin. Mol. Microbiol. 57, 735–750. Elvers, K. T., Wu, G., Gilberthorpe, N. J., Poole, R. K., and Park, S. F. (2004). Role of an inducible single-domain hemoglobin in mediating resistance to nitric oxide and nitrosative stress in Campylobacter jejuni and Campylobacter coli. J. Bacteriol. 186, 5332–5341. Farres, J., Rechsteiner, M. P., Herold, S., Frey, A. D., and Kallio, P. T. (2005). Ligand binding properties of bacterial hemoglobins and flavohemoglobins. Biochemistry 44, 4125–4134.
Resonance Raman Studies of Microbial Hemoglobins
283
Feis, A., Marzocchi, M. P., Paoli, M., and Smulevich, G. (1994). Spin state and axial ligand bonding in the hydroxide complexes of metmyoglobin, methemoglobin, and horseradish peroxidase at room and low temperatures. Biochemistry 33, 4577–4583. Feis, A., Rodriguez-Lopez, J. N., Thorneley, R. N. F., and Smulevich, G. (1998). The distal cavity structure of carbonyl horseradish peroxidase as probed by the resonance Raman spectra of His 42 Leu and Arg 38 Leu mutants. Biochemistry 37, 13575–13581. Ferraro, J. R., Nakamoto, K., and Brown, C. W. (2002). In ‘‘Introductory Raman Spectroscopy.’’ Academic Press, San Diego. Frey, A. D., and Kallio, P. T. (2003). Bacterial hemoglobins and flavohemoglobins: Versatile proteins and their impact on microbiology and biotechnology. FEMS Microbiol. Rev. 27, 525–545. Friedman, J. M. (1994). Timeresolved resonance Raman spectroscopy as probe for structure, dynamics, and reactivity in hemoglobin. Methods Enzymol. 232C, 205–246. Friedman, J. M., Rousseau, D. L., Ondrias, M. R., and Stepnoski, R. A. (1982). Transient Raman study of hemoglobin: Structural dependence of the iron-histidine linkage. Science 218, 1244–1246. Gardner, A. M., Martin, L. A., Gardner, P. R., Dou, Y., and Olson, J. S. (2000). Steady-state and transient kinetics of Escherichia coli nitric-oxide dioxygenase (flavohemoglobin): The B10 tyrosine hydroxyl is essential for dioxygen binding and catalysis. J. Biol. Chem. 275, 12581–12587. Gardner, P. R. (2005). Nitric oxide dioxygenase function and mechanism of flavohemoglobin, hemoglobin, myoglobin and their associated reductases. J. Inorg. Biochem. 99, 247–266. Gilch, H., Schweitzer-Stenner, R., and Dreybrodt, W. (1993). Structural heterogeneity of the Fe(2þ)-N epsilon (HisF8) bond in various hemoglobin and myoglobin derivatives probed by the Raman-active iron histidine stretching mode. Biophys. J. 65, 1470–1485. Gouterman, M. (1978). In ‘‘The Porphyrins’’ (D. Dolphin, ed.), Vol. 3, pp. 1–165. Academic Press, New York. Herzberg, G. (1950). In ‘‘Spectra of Diatomic Molecules, Molecular Spectra and Molecular Structure.’’ Van Nostrand, New York. Hildebrandt, P. (1995). Resonance Raman spectroscopy of cytochrome c. In ‘‘Cytochrome C: A Multidisciplinary Approach’’ (R. A. Scott and A. G. Mauk, eds.), pp. 285–314. University Science Books, Sausalito, CA. Hirota, S. T. L., Phillips, G. N., Olson, J. S., Mukai, M., and Kitagawa, T. (1996). Perturbation of the Fe-O2 bond by nearby residues in heme pocket: Observation of Fe-O2 Raman bands for oxymyoglobin mutants J. Am. Chem. Soc. 118, 7845–7846. Howes, B. D., Rodriguez-Lopez, J. N., Smith, A. T., and Smulevich, G. (1997). Mutation of distal residues of horseradish peroxidase: Influence on substrate binding and cavity properties. Biochemistry 36, 1532–1543. Hu, S., Smith, K. M., Frei, H., and Spiro, T. G. (1996). Assignment of protoheme resonance Raman spectrum by heme labeling in myoglobin. J. Am. Chem. Soc. 118, 12638–12646. Kincaid, J. R. (1999). Resonance Raman spectroscopy of heme proteins and model compounds. In ‘‘The Porphyrin Handbook’’ (K. M. Kadish, K. M. Smith, and R. Guillard, eds.), 7, pp. 225–291. Academic Press, New York. Kitagawa, T. (1988). The heme protein structure and the iron histidine stretching mode. In ‘‘Biological Applications of Raman Spectroscopy: Resonance Raman Spectra of Hemes and Metalloproteins’’ (T. G. Spiro, ed.), Vol. 3, pp. 97–131. Wiley, New York. Kitagawa, T., Nagai, K., and Tsubaki, M. (1979). Assignment of the Fe-Ne (His F8) stretching band in the resonance Raman spectra of deoxy myoglobin. FEBS Lett. 104, 376–378. Kitagawa, T., and Ozaki, Y. (1987). Infrared and Raman spectra of metalloporphyrins. Struct. Bond. 64, 71–114. Li, T., Quillin, M. L., Phillips, G. N., Jr., and Olson, J. S. (1994). Structural determinants of the stretching frequency of CO bound to myoglobin. Biochemistry 33, 1433–1446.
284
Changyuan Lu et al.
Loehr, T. M. (1998). Recent advances in resonance Raman spectroscopy. In ‘‘Spectroscopic Methods in Bioinorganic Chemistry, ACS Symp. Ser. 692,’’ pp. 136–153. American Chemical Society, Washington, DC. Long, D. A. (2001). In ‘‘The Raman effect: A Unified Treatment of the Theory of Raman Scattering by Molecules.’’ New York: Wiley, New York. Lu, C., Egawa, T., Wainwright, L. M., Poole, R. K., and Yeh, S. R. (2007). Structural and functional properties of a truncated hemoglobin from a food-borne pathogen Campylobacter jejuni. J. Biol. Chem. 282, 13627–13636. Maxwell, J. C., Volpe, J. A., Barlow, C. H., and Caughey, W. S. (1974). Infrared evidence for the mode of binding of oxygen to iron of myoglobin from heart muscle. Biochem. Biophys. Res. Commun. 58, 166–171. McCreery, R. L. (2000). In ‘‘Raman Spectroscopy for Chemical Analysis.’’ Wiley, New York. Milani, M., Pesce, A., Ouellet, Y., Ascenzi, P., Guertin, M., and Bolognesi, M. (2001). Mycobacterium tuberculosis hemoglobin N displays a protein tunnel suited for O2 diffusion to the heme. EMBO J. 20, 3902–3909. Milani, M., Savard, P. Y., Ouellet, H., Ascenzi, P., Guertin, M., and Bolognesi, M. (2003). A TyrCD1/TrpG8 hydrogen bond network and a TyrB10TyrCD1 covalent link shape the heme distal site of Mycobacterium tuberculosis hemoglobin O. Proc. Natl. Acad. Sci. USA 100, 5766–5771. Moens, L., Vanfleteren, J., Van de Peer, Y., Peeters, K., Kapp, O., Czeluzniak, J., Goodman, M., Blaxter, M., and Vinogradov, S. (1996). Globins in nonvertebrate species: Dispersal by horizontal gene transfer and evolution of the structure-function relationships. Mol. Biol. Evol. 13, 324–333. Mukai, M., Savard, P. Y., Ouellet, H., Guertin, M., and Yeh, S. R. (2002). Unique ligandprotein interactions in a new truncated hemoglobin from Mycobacterium tuberculosis. Biochemistry 41, 3897–3905. Nakamoto, K. (1970). Infrared spectra of inorganic and coordination compounds. In ‘‘Infrared Spectra of Inorganic and Coordination Compounds,’’ p. 78. Wiley-Interscience, New York. Nardini, M., Pesce, A., Labarre, M., Richard, C., Bolli, A., Ascenzi, P., Guertin, M., and Bolognesi, M. (2006). Structural determinants in the group III truncated hemoglobin from Campylobacter jejuni. J. Biol. Chem. 281, 37803–37812. Ouellet, H., Juszczak, L., Dantsker, D., Samuni, U., Ouellet, Y. H., Savard, P. Y., Wittenberg, J. B., Wittenberg, B. A., Friedman, J. M., and Guertin, M. (2003). Reactions of Mycobacterium tuberculosis truncated hemoglobin O with ligands reveal a novel ligand-inclusive hydrogen bond network. Biochemistry 42, 5764–5774. Ouellet, H., Ouellet, Y., Richard, C., Labarre, M., Wittenberg, B., Wittenberg, J., and Guertin, M. (2002). Truncated hemoglobin HbN protects Mycobacterium bovis from nitric oxide. Proc. Natl. Acad. Sci. USA 99, 5902–5907. Pathania, R., Navani, N. K., Rajamohan, G., and Dikshit, K. L. (2002). Mycobacterium tuberculosis hemoglobin HbO associates with membranes and stimulates cellular respiration of recombinant Escherichia coli. J. Biol. Chem. 277, 15293–15302. Quillin, M. L., Li, T., Olson, J. S., Phillips, G. N., Jr., Dou, Y., Ikeda-Saito, M., Regan, R., Carlson, M., Gibson, Q. H., Li, H., et al. (1995). Structural and functional effects of apolar mutations of the distal valine in myoglobin. J. Mol. Biol. 245, 416–436. Rousseau, D. G., and Rousseau, D. L. (1992). Hydrogen bonding of iron-coordinated histidine in heme proteins. J. Struct. Biol. 109, 13–17. Rousseau, D. L., and Ondrias, M. R. (1983). Resonance Raman scattering studies of the quaternary structure transition in hemoglobin. Annu. Rev. Biophys. Bioeng. 12, 357–389. Samuni, U., Dantsker, D., Ray, A., Wittenberg, J. B., Wittenberg, B. A., Dewilde, S., Moens, L., Ouellet, Y., Guertin, M., and Friedman, J. M. (2003). Kinetic modulation in carbonmonoxy derivatives of truncated hemoglobins: The role of distal heme pocket residues and extended apolar tunnel. J. Biol. Chem. 278, 27241–27250.
Resonance Raman Studies of Microbial Hemoglobins
285
Samuni, U., Ouellet, Y., Guertin, M., Friedman, J. M., and Yeh, S. R. (2004). The absence of proximal strain in the truncated hemoglobins from Mycobacterium tuberculosis. J. Am. Chem. Soc. 126, 2682–2683. Schweitzer-Stenner, R. (1989). Allosteric linkage-induced distortions of the prosthetic group in haem proteins as derived by the theoretical interpretation of the depolarization ratio in resonance Raman scattering. Q. Rev. Biophys. 22, 381–479. Scott, T. W., and Friedman, J. M. (1984). Tertiary-Structure relaxation in hemoglobin: A transient Raman study. J. Am. Chem. Soc. 106, 5677–5687. Shelnutt, J. A., Song, X.-Z., Ma, J.-G., Jia, S.-L., Jentzen, W., and Medforth, C. J. (1998). Nonplanar porphyrins and their significance in proteins. Chem. Soc. Rev. 27, 31–41. Smith, W. E. (1993). Surface enhanced resonance Raman scattering. Methods Enzymol. 226, 482–495. Smith, W. E. (2004). In ‘‘Modern Raman Spectroscopy.’’ Wiley, New York. Smulevich, G. (1993). Structure-function relationship of peroxidases via resonance Raman spectroscopy and site directed mutagenesis: Cytochrome c peroxidase. In ‘‘Advances in Spectroscopy’’ (R. J. H. Clark and R. E. Hester, eds.), Vol. 20, pp. 163–193. Wiley, West Sussex. Smulevich, G. (1998). Understanding heme cavity structure of peroxidases: Comparison of electronic absorption and resonance Raman spectra with crystallographic results. Biospectroscopy 4, S3–S17. Smulevich, G., Hu, S., Rodgers, K. R., Goodin, D. B., Smith, K. M., and Spiro, T. G. (1996). Heme-protein interactions in cytochrome c peroxidase revealed by site-directed mutagenesis and resonance Raman spectra of isotopically labeled hemes. Biospectroscopy 2, 365–376. Spiro, T. G. (1988). ‘‘BiologicalApplications of Raman Spectroscopy: Resonance Raman Spectra of Hemes and Metalloproteins,’’ Vol. 3. Wiley, New York. Spiro, T. G., and Czernuszewicz, R. S. (2000). Resonance Raman spectroscopy. In ‘‘Physical Method in Bioinorganic Chemistry: Spectroscopy and Magnetism’’ ( J. L. Que, ed.), pp. 59–119. University Science Book, Sausalito, CA. Spiro, T. G., and Kozlowski, P. M. (2001). Is the CO adduct of myoglobin bent, and does it matter? Acc. Chem. Res. 34, 137–144. Spiro, T. G., and Li, X.-Y. (1988). Resonance Raman spectra of metalloproteins. In ‘‘Biological Applications of Raman Spectroscopy: Resonance Raman Spectra of Hemes and Metalloproteins’’ (T. G. Spiro, ed.), Vol. 3, pp. 1–37. Wiley, New York. Springer, B. A., Egeberg, K. D., Sligar, S. G., Rohlfs, R. J., Mathews, A. J., and Olson, J. S. (1989). Discrimination between oxygen and carbon monoxide and inhibition of autooxidation by myoglobin: Site-directed mutagenesis of the distal histidine. J. Biol. Chem. 264, 3057–3060. Springer, B. A., Sligar, S. G., Olson, J. S., and Phillips, G. N., Jr. (1994). Mechanisms of ligand recognition in myoglobin. Chem. Rev. 94, 699–714. Tomita, T., Hirota, S., Ogura, T., Olson, J. S., and Kitagawa, T. (1999). Resonance Raman investigation of Fe-N-O structure of nitrosylheme in myoglobin and its mutants. J. Phys. Chem. B 103, 7044–7054. Tsubaki, M., Srivastava, R. B., and Yu, N. T. (1982). Resonance Raman investigation of carbon monoxide bonding in (carbon monoxy)hemoglobin and -myoglobin: Detection of Fe-CO stretching and Fe-C-O bending vibrations and influence of the quaternary structure change. Biochemistry 21, 1132–1140. Varotsis, C., and Babcock, G. T. (1993). Nanosecond time resolved resonance Raman spectroscopy. Methods Enzymol. 226, 409–431. Vinogradov, S. N., Walz, D. A., and Pohajdak, B. (1992). Organization of non-vertebrate globin genes. Comp. Biochem. Physiol. B 103, 759–773.
286
Changyuan Lu et al.
Vogel, K. M., Kozlowski, P. M., Zgierski, M. Z., and Spiro, T. G. (2000). Role of the axial ligand in heme-CO backbonding; DFT analysis of vibrational data. Inorg. Chim. Acta 297, 11–17. Wainwright, L. M., Elvers, K. T., Park, S. F., and Poole, R. K. (2005). A truncated haemoglobin implicated in oxygen metabolism by the microaerophilic food-borne pathogen Campylobacter jejuni. Microbiology 151, 4079–4091. Wainwright, L. M., Wang, Y., Park, S. F., Yeh, S. R., and Poole, R. K. (2006). Purification and spectroscopic characterization of Ctb, a group III truncated hemoglobin implicated in oxygen metabolism in the food-borne pathogen Campylobacter jejuni. Biochemistry 45, 6003–6011. Wajcman, H., and Kiger, L. (2002). Hemoglobin, from microorganisms to man: A single structural motif, multiple functions. C. R. Biol. 325, 1159–1174. Wang, D., and Spiro, T. G. (1998). Structure changes in hemoglobin upon deletion of C-terminal residues, monitored by resonance Raman spectroscopy. Biochemistry 37, 9940–9951. Wang, J., Caughey, W. S., and Rousseau, D. L. (1996). ‘‘Resonance Raman Scattering: A Probe of Heme Protein-bound Nitric Oxide.’’ Wiley, New York. Wittenberg, J. B., Bolognesi, M., Wittenberg, B. A., and Guertin, M. (2002). Truncated hemoglobins: A new family of hemoglobins widely distributed in bacteria, unicellular eukaryotes, and plants. J. Biol. Chem. 277, 871–874. Wu, G., Wainwright, L. M., and Poole, R. K. (2003). Microbial globins. Adv. Microb. Physiol. 47, 255–310. Yeh, S. R., Couture, M., Ouellet, Y., Guertin, M., and Rousseau, D. L. (2000). A cooperative oxygen binding hemoglobin from Mycobacterium tuberculosis: Stabilization of heme ligands by a distal tyrosine residue. J. Biol. Chem. 275, 1679–1684. Yonetani, T., and Anni, H. (1987). Yeast cytochrome c peroxidase: Coordination and spin states of heme prosthetic group. J. Biol. Chem. 262, 9547–9554. Yu, N.-T., and Kerr, E.A (1988). Vibrational modes of coordinated CO, CN-, O2, and NO. In ‘‘Biological Applications of Raman Spectroscopy: Resonance Raman Spectra of Hemes and Metalloproteins’’ (T. G. Spiro, ed.), Vol. 3, pp. 39–95. Wiley, New York. Zhao, X.-J., Sampath, V., and Caughey, W. S. (1994). Infrared characterization of nitric oxide bonding to bovine heart cytochrome c oxidase and myoglobin. Biochem. Biophys. Res. Commun. 204, 537–543.
C H A P T E R
F I F T E E N
The Power of Using Continuous-Wave and Pulsed Electron Paramagnetic Resonance Methods for the Structure Analysis of Ferric Forms and Nitric Oxide-Ligated Ferrous Forms of Globins Sabine Van Doorslaer and Filip Desmet Contents 288 289 295 301 304 305 305 306
1. Introduction 2. Electron Paramagnetic Resonance in a Nutshell 3. EPR Studies of NO-Ligated Globins 4. EPR Studies of Ferric globins 5. Spin-Labeling Heme Proteins 6. Future Challenges and Possibilities Acknowledgments References
Abstract For several decades now, electron paramagnetic resonance (EPR) has been a valuable spectroscopic tool for the characterization of globin proteins. In the early years, the majority of EPR studies were performed using standard continuous-wave EPR techniques at conventional microwave frequencies. In the last years, the field of EPR has known tremendous technological developments, including the introduction of advanced pulsed EPR and high-frequency EPR techniques. After a short overview of the basics of EPR and recent advances in the field, we will illustrate how these different EPR methods can provide information about the dynamics and geometric and electronic structures of heme proteins. Although the main focus of this chapter lies on the EPR analysis
University of Antwerp, Department of Physics, SIBAC Laboratory, Antwerp, Belgium Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37015-8
#
2008 Elsevier Inc. All rights reserved.
287
288
Sabine Van Doorslaer and Filip Desmet
of nitric oxide-ligated ferrous heme proteins and ferric heme systems, we also briefly outline the possibility of site-directed spin labeling of heme proteins. The last section highlights the future potential and challenges in using this magnetic resonance technique in globin research.
1. Introduction Although the function and structure of mammalian myoglobins and hemoglobins have been studied for decades now, these heme proteins have not ceased to surprise different generations of scientists (Giardina et al., 1995; Perutz, 1979; Wittenberg and Wittenberg, 1990). This was recently again demonstrated by the exciting discovery that myoglobin is involved in the removal of toxic nitric oxide (NO) in cardiac and striated muscle with formation of the ferric or ‘‘met’’ form of the globin (Brunori, 2001). Moreover, in the past years, several new vertebrate globins (Pesce et al., 2002), nonvertebrate globins (Weber and Vinogradov, 2001), bacterial globins (Frey and Kallio, 2003), plant globins (Kundu et al., 2003), and even ancestor ‘‘protoglobins’’ (Freitas et al., 2005) have been identified, whereby the function of many of these globins is as yet unknown. Therefore, there is still a large interest in biophysical techniques able to provide insight into the structure and structure–function relations of globins. For decades, electron paramagnetic resonance (EPR), also sometimes referred to as electron spin resonance or electron magnetic resonance, has been one of the methods of choice in investigating the geometric and electronic structures of paramagnetic transition metal-containing proteins. It therefore does not come as a surprise that the EPR technique played an important role in the early biophysical research of heme proteins (Dickinson and Symons, 1983; Smith and Pilbrow, 1980). The paramagnetic states of heme proteins comprise ferric and NO-ligated ferrous forms. The earlier mentioned discovery that myoglobin plays a role in NO metabolism (Brunori, 2001) has strongly renewed interest in these globin forms. The majority of early EPR studies on globins were performed using continuous-wave (cw) EPR at the conventional X-band microwave (mw) frequency (9.5 GHz). However, since the late 1980s, the field of EPR has been revolutionized by the introduction of different pulsed EPR and highfield EPR techniques (Mo¨bius, et al., 2005; Schweiger and Jeschke, 2001). This chapter first gives the non-EPR specialist reader a short overview of the basics of EPR and recent technological developments in this field. It then emphasizes how the different EPR techniques can be used advantageously in the study of globins. The chapter ends with highlighting the future challenges and possibilities of using EPR in heme-protein research.
EPR Methods to Study Ferric and NO-Ligated Globins
289
2. Electron Paramagnetic Resonance in a Nutshell The basis of EPR lies in the electron Zeeman effect, which states that when an electron spin is placed in a magnetic field, B0, the magnetic moment of the electron spin is aligned parallel or opposed to the field. The energy difference between these two states is proportional to the magnetic field. The state of the electron spin can be changed by mw irradiation with an energy hnmw, matching the energy difference between the two levels. The mw absorption condition thus is
hnmw ¼ ge be B0 ;
ð15:1Þ
where be is the Bohr magneton and h is Planck’s constant. In a continuouswave EPR experiment, the mw frequency, nmw, is kept constant and the mw absorption is monitored in function of the external magnetic field (Fig. 15.1A) (Weil et al., 1994). For technical reasons, the first derivative of the EPR spectrum is detected (modulation-amplitude detection). The proportionality factor g is characteristic of the local environment of the unpaired electron and is, in its most general case, represented by a 3 3 matrix, usually referred to as the g tensor. This matrix is the mathematical translation of the fact that the g value varies with the orientation of the magnetic field versus the molecular axes (the unpaired electron ‘‘senses’’ the spatial differences in its direct surrounding). The resulting g tensor is fully characterized by the three principal g values ( gx,gy,gz) and the orientation of the principal g axes in the molecular frame. The g tensor reveals direct information about the local symmetry of the paramagnetic site and about the electronic structure of the paramagnetic center (e.g., oxidation state of the transition metal) (Weil et al., 1994). For a disordered system, such as a frozen solution of proteins, the EPR spectrum is the sum of EPR spectra corresponding to the individual molecular orientations versus the magnetic field (see Fig. 15.1B). Conversely, each magnetic field position in the spectrum of the disordered system then corresponds to a selection of the orientations characterized by the same g value. The situation given in Eq. (15.1) and in Fig. 15.1 holds only for paramagnetic systems with one unpaired electron (S ¼ 1/2). For high-spin systems (S > 1/2), an additional interaction, the zero-field splitting, needs to be taken into account. This interaction is field independent and reflects the dipole–dipole coupling between the electron spins. It can reveal additional information on the spin system. Furthermore, the situation described by Eq. (15.1) holds only for unpaired electrons embedded in an environment without nuclei with a nuclear spin I 1/2. This is of course not the case for (heme) proteins, where the
290
Sabine Van Doorslaer and Filip Desmet
A
B0
B ms = 1/2
B0
B0
hνmw
B0
ms = −1/2
B0
Σ gx
cw-EPR signal
gz 280
300
gy 320 B0 /mT
340
Figure 15.1 (A) Schematic drawing of the basic principle of continuous-wave EPR. The electron Zeeman effect causes a splitting of the energy electronic energy levels proportional to the magnetic field, B0. The absorption of microwaves with fixed frequency, nmw, is monitored in function of the magnetic field. Because of technical reasons, the first derivative of the signal is detected. (B) Drawing explaining the form of the cw EPR spectrum of a disordered system, such as a frozen solution. The full spectrum is the sum of EPR spectra corresponding to the different orientations of the molecule versus the magnetic field.The ellipse represents the molecule.
unpaired electron(s) will be surrounded by different magnetic nuclei, such as 1H (I ¼ 1/2), 14N (I ¼ 1), and 13C (I ¼ 1/2, natural abundance 1.11%). In this case, hyperfine coupling between the electron and nuclear spins will cause a further splitting of the energy levels in Fig. 15.1A. Similar to the situation discussed for the g tensor, the hyperfine coupling depends on the orientation of the magnetic field, B0, in the molecular frame. The hyperfine tensor, A, is fully characterized by its three principal values and axes, whereby the hyperfine principal axes are not necessarily coinciding with the g principal axes. The hyperfine tensor gives valuable information about the spindensity distribution in the molecule, about distances between the unpaired electron center and the nucleus (up to 0.7 nm), and about the type of molecular bonds that are formed (s-,p-bond, . . .). In principle, the hyperfine tensors of the surrounding nuclei can be determined from a cw EPR experiment (analysis of the line splittings). However, this information can seldom be resolved in practice because for hyperfine interactions with n nuclei, with nuclear spin In, each EPR line will split into Pð2In þ 1Þ lines. n
EPR Methods to Study Ferric and NO-Ligated Globins
291
This usually results in a broad line with no resolved substructure. Only in the cases of strong hyperfine couplings can this splitting be observed in a classic cw EPR experiment. This is, for instance, the case for NO-ligated forms of ferrous heme proteins, as discussed in the next section. Detailed information about the smaller hyperfine couplings can, in turn, only be obtained from more advanced experiments, such as pulsed EPR and electron nuclear double resonance (ENDOR) analyses. These experiments also lead to identification of the type of nuclei under study (through the nuclear Zeeman interaction) and provide information about the nuclear quadrupole couplings of high-spin nuclei (I > 1/2). The latter interaction is directly linked to the electric field gradient at the nucleus. By puzzling together all the information obtained from the different interactions, one can form a clear picture of the local geometric and electronic structures of the paramagnetic center. For metallo proteins, this is usually the transition metal center, which is also the prime center of interest. As mentioned before, different (pulsed) techniques need to be used to determine the full set of interactions. One class of these methods consists of electron spin echo modulation (ESEEM) techniques (Schweiger, 1991; Schweiger and Jeschke, 2001). Figure 15.2 outlines the steps in an ESEEM experiment schematically. First, a sequence of mw pulses that generate an electron spin echo (ESE) is applied at a fixed magnetic field, B0. The ESE intensity is then monitored in function of a variation of one or more time intervals between the mw pulses. In this way a one- or two-dimensional modulated time domain signal is obtained. After Fourier transformation, this results in a frequency domain spectrum, which can be linked directly to the nuclear frequencies at the observer position. The experiment is then repeated for different magnetic field settings and hence different selections of spatial orientations. A combination of all these measurements gives the orientation dependence of the nuclear frequencies and can be translated in the hyperfine and nuclear quadrupole tensors. This information can then be linked to the molecular structure. The physical origin of the modulation is complex and pulse sequence dependent. We refer the interested reader to specialized literature on this topic (Schweiger and Jeschke, 2001). Several ESEEM experiments are known, with different advantages and disadvantages. The two-dimensional hyperfine sublevel correlation (HYSCORE) technique, a four-pulse ESEEM experiment, is one of the favored techniques because it provides direct correlations among the different nuclear frequencies and thus facilitates spectral interpretations (Ho¨fer et al., 1986). In the last decade, a number of new ESEEM techniques have been developed, such as matched ESEEM ( Jeschke et al., 1998; Liesum and Schweiger, 2001) and decoupling experiments (Van Doorslaer and Schweiger, 1999), all targeted at enhancing the amount of information that can be obtained from these experiments. For a detailed description of these methods, we refer again to specialized literature (Schweiger and Jeschke, 2001).
292
Sabine Van Doorslaer and Filip Desmet
p/2
p t
t
0
500
1000 t/ns
1500
2000
FT
NZ A Q
0
5
20 10 15 nESEEM/MHz
25
30
Figure 15.2 Schematic representation of an ESEEM experiment. A sequence of mw pulses is applied to generate an electron spin echo (ESE).The time evolution of the ESE intensity is monitored. Fourier transformation (FT) of this signal gives nuclear frequencies that can be translated into nuclear Zeeman (NZ), hyperfine (A), and nuclear quadrupole information. These interactions provide insight into the geometric and electronic structures of the center under study.
Electron nuclear double resonance forms the second major class of techniques used to determine the hyperfine and nuclear quadrupole couplings of magnetic nuclei surrounding the unpaired electron(s). In this type of method, the sample is irradiated with a combination of radio frequency (rf ) and microwaves. The continuous-wave ENDOR technique was introduced by Feher in 1959. The basic principle is quite simple, whereby the cw EPR signal at a certain magnetic field setting is saturated by an increase of the mw power and the recovery of the signal is monitored in function of the frequency of continuously irradiated rf waves. Each time the radio frequency matches a nuclear transition, the populations of the different energy levels will be changed and a net mw absorption can be detected. In this way, nuclear frequencies can be measured in detail. Already at an early stage, Mims (1965) and Davies (1974) demonstrated that ENDOR experiments can also be performed using a combination of mw and rf pulses, whereby a much larger efficiency can be reached and whereby unwanted relaxation effects, known to occur in cw ENDOR, can be excluded. However, it was only when commercial pulsed EPR spectrometers appeared on the market
EPR Methods to Study Ferric and NO-Ligated Globins
293
that Mims and Davies ENDOR started to be applied at length. Since then, different new ENDOR schemes have been developed (Goldfarb, 2006; Jeschke and Schweiger, 1995). A third major group of pulsed EPR techniques consists of so-called electron–electron double resonance (ELDOR) techniques. The majority of ELDOR techniques are targeted at determining interelectron spin distances and have become, in combination with spin-labeling techniques, important tools in distance measurements in biomolecules in the 1.5- to 8-nm range (Freed, 2000; Hubbell et al., 2000; Jeschke, 2002). However, it has also been shown that the ELDOR-detected nuclear magnetic resonance (NMR) technique can provide a valuable alternative for ENDOR, especially at high mw frequencies ( Jeschke and Spiess, 1998). In the last decade, much effort has been put into the development of high-field EPR techniques (mw frequencies of 95 GHz and higher) (Bennati and Prisner, 2005; Mo¨bius et al., 2005). This trend parallels the earlier mentioned advances in pulsed EPR and ENDOR methodology. Because the electron Zeeman interaction scales with the external magnetic field [Eq. (15.1)], a significant gain in spectral resolution can be obtained using high-field EPR, as illustrated in Fig. 15.3A. Because the nuclear Zeeman interaction also depends linearly on B0, the spectral contributions of different nuclei can be disentangled more easily using high-field ENDOR, as exemplified in Fig. 15.3B (Goldfarb, 2006). Furthermore, the amount of sample needed for the experiment scales down as the mw frequency increases. Where one typically needs about 100 ml of a 0.2–2 mM protein solution to perform pulsed EPR experiments of metallo proteins at X-band frequencies, only 1 ml of the solution is needed for W-band EPR (95 GHz). This is of course an important factor in biological applications. Given the clear advantages of the pulsed EPR/ENDOR techniques and the high-frequency experiments, the reader may be surprised that these techniques have only recently started to be applied at larger scales. The answer to this lies in the demanding technological aspects of the techniques. Although the basic ideas about pulsed EPR and ENDOR were already developed in the 1960s (Mims, 1964, 1965), the fast electronics needed to generate short mw pulses and detect transient signals in the nanosecond time scale became only available in the 1980s. Furthermore, in order to perform high-field EPR, one not only needs to have magnets that can be swept in a controlled manner over several Tesla, state-of-the-art mw technology is also needed to construct the mw bridge and cavities (Block et al., 2004; Mo¨bius et al., 2005). This implies that the pulsed and/or high-field EPR and ENDOR techniques are still specialized tools. This applies of course also to other characterization techniques, such as X-ray diffraction, neutron diffraction, or NMR. One of the main concerns in the field of EPR is undoubtedly the optimal translation of the measured EPR parameters into valuable
294
Sabine Van Doorslaer and Filip Desmet
A
B
gz
13C
gx
gy
15N
320
330
340
350
0
1
2 3 4 nENDOR/MHz
B0 /mT
5
6
13C
gy
gz
15N
3200
3300
3400 B0 /mT
3500
0
10
20 30 nENDOR/MHz
40
Figure 15.3 (A) Spectral simulation showing the difference in resolution between X-band cw EPR (top) and W-band cw EPR (bottom). EPR parameters of the rhombic form of NO-ligated human neuroglobin were taken for the simulation (see Table 15.2). In the X-band EPR spectrum, a large degree of spectral overlap is seen that is lifted in the high-field EPR spectrum due to field dependence of the electron Zeeman interaction. Note that some of the additional splittings because of the hyperfine interaction may get lost in the high-field EPR spectrum and that a multifrequency approach is preferred at all times.(B) Spectral simulation showing the difference in resolution between X-band ENDOR (top) andW-band ENDOR (bottom). Simulation parameters: S¼1/2, g¼ [2.252.252.25], 15N, AN¼ [1,1,7] MHz, 13C, AC¼ [1,1,2] MHz. A large spectral overlap is seen in the X-band ENDOR spectrum, whereby contributions of the two nuclei are clearly separated in the W-band ENDOR because of the larger nuclear Zeeman interaction.
electronic and geometric information about the paramagnetic center. Several approaches were developed in the past decades with varying levels of complexity and accuracy (Weil et al., 1994). The recent boost in the available detailed EPR data and the increase in modern-day computational power have led to the development of new quantum chemical approaches to compute the EPR parameters (Kaupp et al., 2004; Neese and Solomon, 2003). Although these developments are still ongoing, it is already clear that many EPR studies in the future will consist of combined experimental and quantum chemical analyses so that experimental data can be optimally
EPR Methods to Study Ferric and NO-Ligated Globins
295
converted in valuable structural and electronic information. In turn, the computational models can help facilitate complex spectral analyses. The following sections show how these different EPR and ENDOR techniques can be used to study the paramagnetic forms of heme proteins in detail.
3. EPR Studies of NO-Ligated Globins The binding of nitric oxide to heme proteins is different from that of other gaseous diatomic ligands such as dioxygen and carbon monoxide. First, NO can bind reversibly to both ferrous and ferric iron, whereas O2 and CO will only bind to the ferrous form of heme proteins (Cooper, 1999). Second, the NO-ligated ferrous heme complex is paramagnetic (S ¼ 1/2) in contrast to the EPR-silent, diamagnetic O2- and CO-ligated ferrous heme protein. Because of this and the electronic similarities between NO and O2, NO was initially mainly used as a magnetic probe to mimic the ligation of dioxygen to globins. Since the physiological importance of NO ligation to ferrous globins recently became apparent (Brunori, 2001; Packer, 1996), the interest in these earlier studies has been renewed for obvious reasons. Despite the fact that NO was used so often as the model to study the binding of diatomic gases, such as CO or O2, some important differences exist between these molecules. Ferrous heme proteins have a very large affinity for NO binding that exceeds considerably the corresponding affinities for O2 and CO binding. Another marked difference between NO and the two other physiologically important diatomic gases is that nitric oxide can bind more tightly to the ferrous heme iron if no proximal ligand is present, which is not the case for CO or O2 (Cooper, 1999). In fact, when NO binds to a heme protein, the His-Fe bond may weaken as much as a 1000-fold and can even cleave (Yonetani et al., 1998). Conversely, the binding of a trans ligand may induce NO dissociation (Cooper, 1999). The five-coordinated NO-ligated ferrous heme iron has, for instance, been observed in human hemoglobin (Yonetani et al., 1998) and in cytochrome c 0 (Lawson et al., 2003). Electron paramagnetic resonance provides an excellent tool to study the paramagnetic five- and six-coordinated NO-ligated ferrous heme complexes. The five-coordinated form is characterized by an X-band cw EPR spectrum that shows clear splitting in the high-field range due to the hyperfine interaction with the 14N of NO (Fig. 15.4A) (Morse and Chan, 1980). 14N has a nuclear spin I ¼ 1 and thus the EPR line is split in three lines (2I þ 1). Because 15N has a lower nuclear spin (I ¼ 1/2), the origin of
296
Sabine Van Doorslaer and Filip Desmet
O N A
FeII O N
B
FeII His N
C
O
FeII His 320
330 B0 /mT
340
350
Figure 15.4 (A) Simulated X-band cw EPR spectrum of a five-coordinated NOligated ferrous heme protein. Simulation is done using the principal g and nitrogen hyperfine values reported for cytochrome c 0 (Table 15.1, Usov et al., 2006). (B) Experimental X-band cw EPR spectrum of the ferrous nitrosyl form of the recombinant E7Leu mutant of human neuroglobin recorded at 175 K.This spectrum is dominated by the contribution of the ‘‘axial’’ six-coordinated species. (C) Experimental X-band cw EPR spectrum of the ferrous nitrosyl form of the recombinant E7Leu mutant of sperm whale myoglobin recorded at 13 K. This spectrum is dominated by the contribution of the ‘‘rhombic’’ six-coordinated species. Species responsible for EPR spectra are drawn schematically on the right.
splitting in the EPR spectrum could be traced easily by replacement of 14NO with 15NO (Usov et al., 2006). A more detailed insight into the electronic structure of five- and sixcoordinated nitrosyl–heme complexes can only be obtained using a combination of different EPR and ENDOR techniques. In their study of frozen solutions of NO-ligated cytochrome c 0 , a heme protein exhibiting a five coordination of the heme iron, Charles Scholes and co-workers showed that Q-band cw ENDOR provides an excellent probe of the unpaired electron density on the nitric oxide nitrogen (Table 15.1,14N hyperfine coupling) (Usov et al., 2006). The hyperfine information revealed that, in contrast to the expectations, the electron spin on the NO is not found in a p* orbital with only 2p character, but resides in an orbital with both 2s and 2p character (1:2 ratio). Furthermore, Q-band proton ENDOR spectra revealed essential information on the heme pocket structure. The Fe-NO plane was found to bisect the heme N-Fe-N. The nearest proton of the Phe14 ring is located 0.31 nm away from the heme iron, whereby Phe14 is positioned to obstruct NO binding. Furthermore, exchangeable protons
Table 15.1
Principal g and 14N(NO) hyperfine values of five- and six-coordinated NO-ligated ferrous heme proteinsa T/ K
gz
gy
Six-coordinated NO-ligated ferrous heme proteins NGBb,c(rhombic 10 2.079 2.004 species) 190 2.039 2.039 NGBc(axial species) HB (a unit)d 200 2.062 2.010 HB (a unit)e 85 2.0604 1.9995 HB (b unit)d 200 2.032 2.032 Five-coordinated NO-ligated ferrous heme proteins Cytochrome c 0 f 2 2.116 2.017 a b c d e f
gx
A1/ MHz
A2/ MHz
A3/ MHz
Ref.
1.974
32
64
47
Trandafir et al.(2004)
1.993 1.989 1.9653
n.d. n.d. 29.6
n.d. n.d. 63.6
n.d. n.d. 32.9
1.987
n.d.
n.d.
n.d.
Trandafir et al.(2004) Flores et al.(1997) Utterback et al. (1983) Flores et al. (1997)
2.008
35.7
36
43.7
Usov et al. (2006)
The temperature at which data were determined are given. n.d., not determined; NGB, human neuroglobin; HB, human hemoglobin. Hyperfine values of the 14N nucleus of the proximal histidine were estimated at A1 ¼ 25 MHz, A2 ¼ 19 MHz, and A3 ¼ 30 MHz. X-band cw EPR of frozen globin solutions. Q-band cw EPR of frozen globin solutions. X-band cw EPR of globin single crystals. Q-band cw EPR and ENDOR of frozen protein solutions.
298
Sabine Van Doorslaer and Filip Desmet
of Arg127 could be identified that may H-bond to the NO ligand. Finally, small features near 6 MHz could be detected in the Q-band ENDOR spectra. These signals stem from the heme nitrogens, but the ENDOR spectra did not allow for a detailed analysis of the couplings. In an earlier model study on five-coordinated nitrosyl iron(II) tetraphenylporphyrin, Gilbert and Doetschman (2001) showed that HYSCORE provides an excellent tool to determine the hyperfine and nuclear quadrupole couplings of the heme nitrogens. The X-band cw EPR spectrum of the five-coordinated form differs clearly from the ones of six-coordinated NO-ligated heme proteins, for which two forms can be identified (see Figs. 15.4B and 15.4C). At high temperatures (>150 K), EPR spectra are dominated by species characterized by an axial g tensor (see Table 15.1, Fig. 15.4B). At lower temperatures, a second form, characterized by a rhombic g tensor, contributes predominantly to the cw EPR spectrum (Flores et al., 1997; Schmidt et al., 2001) (see Fig. 15.4C). The central line is split into a triplet of triplets as a consequence of the hyperfine interaction with the 14N nucleus of the nitric oxide and the directly coordinating nitrogen of the proximal histidine. When 15NO is bound to the heme, a doublet of triplets is observed, confirming the origin of the dominant hyperfine interaction (Yonetani et al., 1972). In order to gain further insight into the origin of the different forms, ENDOR and ESEEM techniques need to be applied. Based on a combination of X-band cw ENDOR, three-pulse ESEEM, and HYSCORE spectroscopy, the form characterized by the rhombic g tensor could be identified as a form in which the Fe-NNO bond does not coincide with the heme normal (see Fig. 15.4C) (Flores et al., 2000; Ho¨hn et al., 1983; Hu¨tterman et al., 1994; Kappl and Hu¨tterman, 1989; LoBrutto et al., 1983; Tyryshkin et al., 1999). This tilt was also confirmed in a combined X-ray, EPR, and Mo¨ssbauer study of nitrosyl iron(II) porphyrinates (Wyllie et al., 2003). The tilt is also found by density functional theory (DFT) computations (Patchkovskii and Ziegler, 2000). The latter study also indicates that the Fe-NO plane is quasi-bisecting the Npyrrole-Fe-Npyrrole bonds. The geometric structure of the form characterized by the axial g tensor is more debated. Most authors ascribe the form to a species in which the FeNNO bond is coinciding with the normal (Flores et al., 2000; Hu¨tterman et al., 1994; LoBrutto et al., 1983; Schmidt et al., 2001; Tyryshkin et al., 1999). DFT computations predicted a bent end-on orientation of the NO with the Fe-NO plane eclipsing one of the Fe-Npyrrole bonds or a partial dissociation of the NO whereby the NO is freely rotating as possible structures (Patchkovskii and Ziegler, 2000). The latter structure seems unlikely to occur at lower temperatures. The X-band proton cw ENDOR analysis of nitrosyl-ligated horse heart myoglobin showed that the E7His and E11Val amino acid residues stabilize
EPR Methods to Study Ferric and NO-Ligated Globins
299
both NO forms but with differing strength (Flores et al., 2000). The ESEEM and HYSCORE studies of NO-ligated myo- and hemoglobin identified an interaction of the unpaired electron with the Ne of the distal E7His in axial case, which was not observed in the rhombic species (Tyryshkin et al., 1999). Hyperfine parameters for a model of Mb-15NO, including distal HisE7 and ValE11, have been determined using density functional calculations. Calculations were carried out for different values of the Fe-N-O angle and the Fe-NNO distance (Zhi et al., 2004). It was found that the 15N NO hyperfine values are very sensitive to the value of the angle and the distance. Although there was some uncertainty in matching experimental data with the calculations, a value of 140–150 for the Fe-N-O angle and a ˚ provide the best match, in line with X-ray Fe-NNO distance of 1.74 A absorption fine structure data (Rich et al., 1998). A better match between experiment and theory was found for the 14Ne of HisF8. In this case there was no strong dependence of the hyperfine values on the values of the angle and the distance and there is a preference for larger angles (>130 ). These results are in line with those of another density functional theory investigation in which the hyperfine coupling components were calculated for a sixcoordinated NO-ligated ferrous porphyrin model complex (Zhang et al., 2003). DFT calculations for a number of different Fe-N-O angle and FeNNO bond length combinations were compared with experimental data of HbNO. It was found that the most likely Fe-N-O angle and Fe-NNO bond length at 100 K are 136–137 and 1.79 A˚, respectively. Analysis of the temperature dependence of the cw EPR spectra of sixcoordinated NO-ligated ferrous heme proteins reveals interesting information on the local environment of the NO ligand (Flores et al., 1997; Schmidt et al., 2001; Trandafir et al., 2004). If a simple two-state model would be valid whereby the rhombic form (R) is converted in the axial form (A) in function of temperature, then the natural logarithm of the ratio of A species versus R species should depend linearly on 1/T (T ¼ temperature) (Flores et al., 1997). This was not observed for mammalian myo-, hemo-, and neuroglobin, suggesting that a continuous change occurs in the conformations as the temperature increases (Flores et al., 1997; Trandafir et al., 2004). Furthermore, the temperature dependence of the percentage of the rhombic form contributing to the cw EPR spectrum is also clearly different for the three mammalian globins (Flores et al., 1997; Trandafir et al., 2004). This is related to the specific position and interaction of the E7His with NO, as could be determined from experiments on different E7 mutants of neuroglobin (Trandafir et al., 2004). As would be expected, the spectral difference between the cw EPR spectra of the five- and six-coordinated NO-forms has been used extensively to identify the type of NO ligation in different heme proteins (Bemski, 1997; Yonetani et al., 1998). It is interesting to mention that all aforementioned types of NO ligation have been observed in human
300
Sabine Van Doorslaer and Filip Desmet
hemoglobin. Hemoglobin consists of four polypeptide chains (two a and two b chains). The a hemes have more freedom of motion than the b hemes. In the fully oxygenated state, the structure is referred to as the relaxed structure, whereby the tense (T) quaternary structure is formed in the absence of oxygen. Where all hemes are six coordinated for the relaxed structure of the NO-ligated ferrous hemoglobin, the a hemes are five coordinated and the b hemes are six coordinated in the T conformation (Bemski, 1997). EPR has played an important role in the elucidation of this NO-binding behavior (Bemski, 1997). Furthermore, X-band cw EPR allows for a quantification of the amount of NO-ligated globins, as was demonstrated Piknova et al. (2005), who developed an accurate assay to detect the level of nitrosyl hemoglobin in human blood. Electron paramagnetic resonance has also been combined with cryogenic photolysis experiments in order to study NO-binding kinetics to heme proteins. LoBrutto et al. (1984) studied the recovery kinetics to heme for the nitrosyl ferrocytochrome a3 center in comparison to nitrosyl myoglobin. The EPR spectrum of the NO-ligated ferrous form of the two proteins decreased significantly upon photolysis at low temperature. The NO rebinding to myoglobin started already at 15 K, in a temperature region where quantum tunneling could occur. In contrast, the NO rebinding to the heme a3 center started only near a temperature of 50 K. This shows that the barrier to NO recombination is much higher and/or wider in the oxidase than in myoglobin. An ENDOR analysis proved that a unique proton is associated with the cytochrome a3 center that is not found in other heme NO systems and that may be part of a nearby protein side chain capable of perturbing a distal ligand such as NO (LoBrutto et al., 1983). It may be that this nearby proton influences the activation energy barrier and helps steer the incoming ligand to the correct position in the cytochrome. As mentioned in the beginning, not only do ferrous heme proteins bind NO, their related ferric form can also bind NO to form a diamagnetic center. Upon photolytic dissociation of ferric NO-ligated globins, EPR active ferric high-spin heme and free nitric oxide are generated. Upon cryogenic photolysis, the photo-induced intermediates are trapped in the heme cavity and can be detected by EPR. The complex cw EPR spectra then reflect the interaction between the NO molecule (S ¼ 1/2) and the ferric heme species (S ¼ 5/2). Ikeda-Saito and co-workers proved in this way that the photo-dissociated nitric oxide cannot leave the protein matrix of myoglobin at temperatures below 100 K (Hori et al., 2000). The stabilizing influence of the different amino acids in the heme pocket could be probed by combining the photolytic EPR experiments with selective mutagenesis in the heme pocket. The mobility appeared to be governed by the amino residue at position E11, a result that relates also to the earliermentioned ENDOR work of Flores et al. (2000), who identified the E11Val residue as one of the important stabilizing amino acids in the heme pocket of NO-ligated ferrous myoglobin.
EPR Methods to Study Ferric and NO-Ligated Globins
301
4. EPR Studies of Ferric globins The spin state of the ferric globins is found to vary from a low-spin (S ¼ 1/2) to a high-spin (S ¼ 3/2 or 5/2) state. The difference between the states is governed by the strength of the distal ligand: binding of a strong distal ligand, such as an endogenous histidine, induces a low-spin state, whereby weak distal ligands, such as exogenous water, will lead to a highspin state. Different EPR strategies have to be followed to study the lowand high-spin ferric forms of globins. X-band cw EPR experiments can already reveal valuable information about the direct heme environment of low-spin ferric heme proteins (Blumberg and Peisach, 1971). The three principal g values can be directly linked to the tetragonal splitting D and the rhombic splitting parameter V as derived in the ligand-field analysis of Taylor (1977)
gy V gx þ ; ¼ l ðgz þ gy Þ ðgz gx Þ
ð15:2Þ
D gx gz V þ ; ¼ l ðgz þ gy Þ ðgy gx Þ 2l
ð15:3Þ
where l is the spin-orbit coupling constant. Blumberg and Peisach (1971) found that the position in the (jD/lj, jV/Dj) graph (the so-called ‘‘truth table’’) gives immediate information on the type of axial ligands. Table 15.2 gives some examples of typical g and ligand-field parameters for different ferric forms of globins. The ligand-field parameters of ferric OH-ligated horse myoglobin fall within the ‘‘type O’’ area in the truth tables (O indicating the direct axial coordination of an oxygen), whereby the parameters of ferric tomato hemoglobin, mouse neuroglobin, and human cytoglobin are typical for bis-histidine-coordinated globins. Individual differences in the ligand-field parameters of the latter three globins relate directly to the orientation of the relative histidine planes. It is found that V/l reaches a value of 2 when the two imidazole planes of the E7 and F8 histidines align parallel to each other (Walker et al., 1986). V/l will become smaller and gz larger when the two base planes are perpendicular to each other. Furthermore, the g values depend strongly on the relative orientation of the imidazole planes versus the Fe-Npyrrole bonds. gz increases and V/l and V/D decrease when going from a situation where the planes are eclipsing the bonds to a bisecting conformation (Quinn et al., 1987). For a quasi-parallel orientation of the two axial bases, V/D lies in the region of 0.65–0.75 for a quasi-eclipse configuration and is only 0.4–0.5 for a quasibisecting orientation (Quinn et al., 1987). Table 15.2 shows that V/l
302
Sabine Van Doorslaer and Filip Desmet
Table 15.2 Typical principal g and corresponding ligand-field parameters of ferric forms of globins gx
gy
gz
V/l
D/l
V/D
Ref.
Ioanitescu et al. (2005) Vinck et al. (2004) Vinck et al. (2004) Dickinson and Symons (1983) Svistunenko et al. (2000)
Ferric tomato hemoglobin Ferric mouse neuroglobin Ferric human cytoglobin CN-ligated myoglobin
1.44 2.23
2.98
1.72
3.19
0.54
1.29 2.15
3.12
1.42
3.16
0.45
1.20 2.08
3.20
1.27
3.23
0.39
0.93 1.89
3.45
0.92
3.31
0.28
Ferric horse MbOH
1.84 2.16
2.59
3.27
6.85
0.48
deviates increasingly more from 2 in the series of ferric tomato hemoglobin, mouse neuroglobin, and human cytoglobin. This implies that the two histidine planes deviate increasingly more from the parallel orientation characterized by V/l ¼ 2. Furthermore, V/D values indicate that the histidine planes are bisecting the Fe-Npyrrole bonds for ferric neuroglobin and cytoglobin. For ferric tomato hemoglobin, an intermediate situation between full eclipse and bisection is expected (V/D ¼ 0.54). For ferric mouse neuroglobin and human cytoglobin, the crude predictions based on the g values are confirmed by X-ray data (Pesce et al., 2004). Finally, the gz value of cyanide-ligated myoglobin is typical for a cyanideligated heme protein whereby the axial histidine has a neutral character (see Table 15.2). In case the axial histidine takes on an imidazolate character (e.g., in cyanide-ligated ferri-horseradish peroxidase), the gz value decreases significantly ( gz ¼ 3.05) (Blumberg et al., 1968). Low-spin ferric centers identified by a maximal gz value larger than 3.3 are usually referred to as highly anisotropic low-spin species ( Walker et al., 1986). In many cases the corresponding X-band cw EPR spectrum is characterized by a single feature at the maximal g value and can only be observed at very low temperatures (4–20 K). Although much of the information about low-spin ferric heme proteins can be obtained from simple X-band cw EPR experiments, hyperfine and nuclear quadrupole interactions of the surrounding nuclei can only be mapped out using different ENDOR and ESEEM techniques. Pioneering work was performed by Scholes and co-workers (1986), who determined for the first time the heme and imidazole nitrogen hyperfine couplings and proton hyperfine couplings for imidazole-ligated ferric myoglobin and
EPR Methods to Study Ferric and NO-Ligated Globins
303
related model systems using X-band cw ENDOR in combination with isotope labeling. Despite the fact that their study revealed interesting information about the spin density distribution in these low-spin ferric complexes that could only be confirmed recently by quantum chemical computations ( Johansson et al., 2002), they were not able to determine the full sets of hyperfine and nuclear quadrupole data because of the large spectral overlap of the contributions of the different nuclei. This type of limitation was also encountered in X-band Davies ENDOR and threepulse ESEEM analyses of heme proteins (Fahnenschmidt et al., 2000; Peisach et al., 1979). In their structural analyses of chloroperoxidase, Hoffman and co-workers showed that these problems can partially be circumvented by the use of cw and pulsed ENDOR techniques at Q-band mw frequencies (35 GHz) (Fann et al., 1994; Lee et al., 1997). Surprisingly, their approach has not yet been implemented in other EPR studies on heme proteins. Similarly, two-dimensional ESEEM techniques have been used only rarely to study heme proteins. The first HYSCORE experiments on ferric heme proteins were performed by Garcı´a-Rubio and co-workers (2003) in their analysis of ferric cytochrome b559. In a model study on a ferric bis (imidazole)-coordinated porphyrin ( Vinck and Van Doorslaer, 2004), we showed how HYSCORE spectroscopy can be combined with the earlier proposed proton combination frequency analyses (Astashkin et al., 1999; Raitsimring et al., 1996) to determine the relative orientation of the imidazole planes in the molecular frame. The EPR strategy was tested for the case of ferric mouse neuroglobin (Vinck et al., 2006) and quite good agreements with earlier X-ray data could be obtained. This opens the way to obtain detailed information on the heme pocket structure of ferric globins and other heme proteins for which no X-ray structure is (yet) available, as was shown in our study of tomato hemoglobin (Ioanitescu et al., 2005). Where the X-band cw EPR studies of low-spin ferric globins could still provide interesting information about the directly coordinating axial ligands, X-band cw EPR spectra of all high-spin ferric heme proteins look disappointingly similar, revealing little or no information on the heme pocket structure (Ikeda-Saito et al., 1992). The zero-field parameters can only be determined from small shifts in cw EPR spectra taken at high mw frequencies, but insufficient data are currently available to directly couple these parameters to the heme iron environment (van Kan et al., 1998). Furthermore, this approach demands EPR spectrometers working at very high mw frequencies of 285 GHz or higher. Only a few prototypes of these spectrometers exist worldwide. More information about nuclei surrounding the unpaired electron can, in principle, be derived from ENDOR and ESEEM experiments. After their initial cw ENDOR work on protohemin chloride and protohemin bromide (Van Camp et al., 1976), Scholes and co-workers (1982)
304
Sabine Van Doorslaer and Filip Desmet
performed a seminal X-band cw ENDOR study of single crystals of aquometmyoglobin where they unraveled the hyperfine and nuclear quadrupole tensors of the heme and histidine nitrogens in detail. Surprisingly, this study did not mark the start of an extensive use of ENDOR and ESEEM techniques to analyze high-spin ferric heme proteins. In fact, when no single crystals of the protein are available, the amount of information that can be derived from an X-band cw ENDOR study decreases extremely because of the large spectral overlap of the signals. It was again Hoffman and co-workers who demonstrated that some of these problems can be partially solved by combining X- and Q-band ENDOR techniques (Fann et al., 1995). In their analysis of fluorometmyoglobin, these authors proved that the fluoride ligand is hydrogen bonded to the NeH of the E7 histidine over a wide pH range. Furthermore, they showed that the nuclear frequencies can be used to determine zero-field splitting without the need to go to very high mw frequencies. Despite the successes of their multifrequency approach, Q-band ENDOR was only used in one later study on a high-spin ferric heme center in cytochrome bo3 (Veselov et al., 2000). Furthermore, the amount of ESEEM studies of high-spin ferric heme proteins is even smaller than the number of ENDOR studies (Aissaoui et al., 1998). As outlined elsewhere (Trandafir, 2007), this can be traced back to inherent properties of the spin system and to method-specific limitations of standard ESEEM experiments. In this work, we showed how these limitations can be overcome by the use of different matched HYSCORE schemes. In this way, the nitrogen hyperfine couplings of the heme and F8His nitrogens of the E7Q mutant of ferric mouse neuroglobin could be determined. Via a combination of a standard HYSCORE experiment with a H2 O$2 H2 O isotope exchange we revealed that the ferric E7Q mutant lacks a distal water. Earlier studies had already shown that similar information on the water ligation of high-spin ferric heme proteins can also be obtained from ENDOR experiments in combination with H2 O$2 H2 O and H2 O$H17 2 O exchange experiments (Aissaoui et al., 1998; Veselov et al., 2000).
5. Spin-Labeling Heme Proteins The previous sections discussed in detail the EPR analysis of intrinsically paramagnetic forms of heme proteins. In this small section, we would like to point out that the diamagnetic heme proteins can be turned into EPR-active proteins by site-directed spin labeling. In this technique, a paramagnetic spin label (usually a nitroxide label) is attached to the proteins via a reaction of its functional group with the thiol group of the cysteine(s) of a protein (Hubbell et al., 2000). Native cysteines can be used for this
EPR Methods to Study Ferric and NO-Ligated Globins
305
procedure or the cysteines can be introduced via site-directed mutagenesis to achieve labeling at selected positions. Protein dynamics can be studied via analysis of room-temperature cw EPR spectra of mono-labeled proteins. In doubly labeled proteins, the interspin distances can be determined giving a direct insight in the protein structure. EPR studies of mono-labeled ferrous yeast iso-1-cytochrome c 0 revealed important information on proteinfolding and -unfolding mechanisms (De Weerd et al., 2001; Qu et al., 1997). In spin-labeled ferric heme proteins, the iron-label distance can be determined from a series of cw and pulsed EPR measurements (Eaton and Eaton, 2000). This distance information can then be translated in direct information about the protein structure. We believe that there lies a big potential in using spin-label EPR to gain a deeper insight in the structure and protein (folding) dynamics of globins.
6. Future Challenges and Possibilities From the aforementioned sections it will have become clear that EPR offers a valuable and diverse biophysical tool to derive a manifold of information on the geometric and electronic structures and ligand-binding kinetics of globins. However, it is our strong belief that the full potential of EPR and ENDOR has not been exploited in heme research. There are less than 10 HYSCORE studies of paramagnetic heme proteins reported, despite the fact that this technique has proven its value in many other metallo-protein studies. Many advanced pulsed EPR and ENDOR sequences developed in the last decade (Schweiger and Jeschke, 2001) have not been applied to heme research or globin studies. There is a marked lack of the use of W-band EPR spectroscopy (95 GHz) in heme protein research, despite the fact that these high-frequency EPR spectrometers are available commercially and have been introduced in many laboratories throughout the world in the last years. Future challenges in the EPR analysis of heme proteins include therefore the use of more advanced multidimensional pulsed EPR and ENDOR at different mw frequencies, the development of photolysis experiments at high frequencies, the development and optimization of DFT computations to interpret the EPR parameters of paramagnetic heme systems, and an intensified use of spin-label EPR. In short, state-of-the-art EPR should be used more intensively in future studies of globins.
ACKNOWLEDGMENTS This work was supported by the Fund for Scientific Research-Flanders (Research Grant G.0468.03). Filip Desmet acknowledges the University of Antwerp–BOF Fund for Ph.D. funding.
306
Sabine Van Doorslaer and Filip Desmet
REFERENCES Aissaoui, H., Bachmann, R., Schweiger, A., and Woggon, W. D. (1998). On the origin of the low-spin character of cytochrome P450(cam) in the resting state: Investigations of enzyme models with pulse EPR and ENDOR spectroscopy. Angew. Chem. Int. Ed. 37, 2998–3002. Astashkin, A. V., Raitsimring, A. M., and Walker, F. A. (1999). Two- and four-pulse ESEEM studies of the heme binding center of a low-spin ferriheme protein: The importance of a multi-frequency approach. Chem. Phys. Lett. 306, 9–17. Bemski, G. (1997). Contribution of electron paramagnetic resonance to the studies of hemoglobin: The nitrosylhemoglobin system. Mol. Biol. Rep. 24, 263–269. Bennati, M., and Prisner, T. F. (2005). New developments in high field electron paramagnetic resonance with applications in structural biology. Rep. Prog. Phys. 68, 411–448. Block, H., Disselhorst, J. A. H. M., Orlinskii, S. B., and Schmidt, J. (2004). A continuouswave and pulsed electron spin resonance spectrometer operating at 275 GHz. J. Magn. Reson. 166, 92–99. Blumberg, W. E., and Peisach, J. (1971). A unified theory for low spin forms of all ferric heme proteins as studied by EPR. In ‘‘Probes of Structure and Function of Macromolecules and Membranes’’ (B. Chance, T. Yonetani, and A. S. Mildvan, eds.), Vol. 2, p. 215. Academic Press, New York. Blumberg, W. E., Peisach, J., Wittenberg, B. A., and Wittenberg, J. B. (1968). The electronic structure of protoheme proteins. I. An electron paramagnetic resonance and optical study of horseradish peroxidase and its derivatives. J. Biol. Chem. 243, 1854–1862. Brunori, M. (2001). Nitric oxide moves myoglobin centre stage. Trends Biochem. Sci. 26, 209–210. Cooper, C. E. (1999). Nitric oxide and iron proteins. Biochim. Biophys. Acta Bioenerg. 1411, 290–309. Davies, E. R. (1974). A new pulse ENDOR technique. Phys. Lett. 74A, 1–2. De Weerd, K., Gigoryants, V., Sun, Y., Fetrow, J. S., and Scholes, C. P. (2001). EPRdetected folding kinetics of externally located cysteine-directed spin-labeled mutants of iso-1-cytochrome c. Biochemistry 40, 15846–15855. Dickinson, L. C., and Symons, M. C. R. (1983). Electron spin resonance of haemoglobin and myoglobin. Chem. Soc. Rev. 12, 387–414. Eaton, S. S., and Eaton, G. R. (2000). Determination of distances based on T1 and Tm effects. In ‘‘Biological Magnetic Resonance. 19. Distance Measurements in Biological Systems by EPR’’ (L. J. Berliner, S. S. Eaton, and G. R. Eaton, eds.), p. 347. Kluwer Academic/Plenum, New York. Fahnenschmidt, M., Bittl, R., Rau, H. K., Haehnel, W., and Lubitz, W. (2000). Electron paramagnetic resonance and electron nuclear double resonance spectroscopy of a heme protein maquette. Chem. Phys. Lett. 323, 329–339. Fann, Y. C., Gerber, N. C., Osmulski, P. A., Hager, L. P., Sligar, S. G., and Hoffman, B. M. (1994). ENDOR determination of heme ligation in chloroperoxidase and comparison with cytochrome P450cam. J. Am. Chem. Soc. 116, 5989–5990. Fann, Y., Ong, J., Nocek, J. M., and Hoffman, B. M. (1995). 19F and 1,2H ENDOR study of distal-pocket N(E)-H. . .F hydrogen bonding in fluorometmyoglobin. J. Am. Chem. Soc. 117, 6109–6116. Feher, G. (1959). Observation of nuclear magnetic resonances via the electron spin resonance line. Phys. Rev. 103, 834–835. Flores, M., Wajnberg, E., and Bemski, G. (1997). Temperature dependence of Q-band electron paramagnetic resonance spectra of nitrosyl heme proteins. Biophys. J. 73, 3225–3229.
EPR Methods to Study Ferric and NO-Ligated Globins
307
Flores, M., Wajnberg, E., and Bemski, G. (2000). Proton electron nuclear double resonance from nitrosyl horse heart myoglobin: The role of His-E7 and Val-E11. Biophys. J. 78, 2107–2115. Freed, J. H. (2000). New technologies in electron spin resonance. Annu. Rev. Phys. Chem. 51, 655–689. Freitas, T. A. K., Saito, J. A., Hou, S., and Alam, M. (2005). Globin-coupled sensors, protoglobins, and the last universal common ancestor. J. Inorg. Biochem. 99, 23–33. Frey, A. D., and Kallio, P. T. (2003). Bacterial hemoglobins and flavohemoglobins: Versatile proteins and their impact on microbiology and biotechnology. FEMS Microbiol. Rev. 27, 525–545. Garcı´a-Rubio, I., Martı´nez, J. I., Picorel, R., Yruela, I., and Alonso, P. J. (2003). HYSCORE spectroscopy in the cytochrome b559 of the photosystem II reaction center. J. Am. Chem. Soc. 125, 15846–15854. Giardina, B., Messana, I., Scatena, R., and Castagnola, M. (1995). The multiple functions of hemoglobin. Crit. Rev. Biochem. Mol. Biol. 30, 165–196. Gilbert, D. C., and Doetschman, D. C. (2001). Five-coordinate nitrosyl iron(II) tetraphenylporphyrin exhibits porphyrin ring 14N symmetry about the Fe-NO plane: A hyperfine sublevel correlation spectroscopy study. Chem. Phys. 269, 125–135. Goldfarb, D. (2006). High field ENDOR as a characterization tool for functional sites in microporous materials. Phys. Chem. Chem. Phys. 8, 2325–2343. Ho¨fer, P., Grupp, A., Nebenfu¨hr, H., and Mehring, M. (1986). Hyperfine sublevel correlation (HYSCORE) spectroscopy: A 2D ESR investigation of the squaric acid radical. Chem. Phys. Lett. 132, 279–282. Ho¨hn, M., Hu¨ttermann, J., Chien, J. C. W., and Dickinson, L. C. (1983). 14N and 1H ENDOR of Nitrosylhemoglobin. J. Am. Chem. Soc. 105, 109–115. Hori, H., Masuya, F., Dou, Y., and Ikeda-Saito, M. (2000). EPR studies on the photoinduced intermediates of NO complexes in recombinant ferric Mb trapped at low temperature. J. Inorg. Biochem. 82, 181–187. Hubbell, W. L., Cafiso, D. S., and Altenbach, C. (2000). Identifying conformational changes with site-directed spin labeling. Nat. Struct. Mol. Biol. 7, 735–739. Hu¨ttermann, J., Burgard, C., and Kappl, R. (1994). Proton ENDOR from randomly oriented NO-ligated hemoglobin: Approaching the structural basis for the R-T transition. J. Chem. Soc. Faraday Trans. 90, 3077–3087. Ikeda-Saito, M., Hori, H., Andersson, L. A., Prince, R. C., Pickering, I. J., George, G. N., Sanders, C. R., II, Lutz, R. S., McKelvey, E. J., and Mattera, R. (1992). Coordination structure of the ferric heme iron in engineered distal histidine myoglobin mutants. J. Biol. Chem. 267, 22843–22852. Ioanitescu, A. I., Dewilde, S., Kiger, L., Marden, M. C., Moens, L., and Van Doorslaer, S. (2005). Characterization of nonsymbiotic tomato hemoglobin. Biophys. J. 89, 2628–2639. Jeschke, G. (2002). Distance measurements in the nanometer range by pulse EPR. Chem. Phys. Chem. 3, 927–932. Jeschke, G., Rakhmatullin, R., and Schweiger, A. (1998). Sensitivity enhancement by matched microwave pulses in one- and two-dimensional electron spin echo envelope modulation spectroscopy. J. Magn. Reson. 131, 261–271. Jeschke, G., and Schweiger, A. (1995). Hyperfine-correlated electron nuclear double resonance spectroscopy. Chem. Phys. Lett. 246, 431–438. Jeschke, G., and Spiess, H. W. (1998). NMR-correlated high-field electron paramagnetic resonance spectroscopy. Chem. Phys. Lett. 293, 9–18. Johansson, M. P., Sundholm, D., Gerfen, G., and Wilkstro¨m, M. (2002). The spin distribution in low-spin iron porphyrins. J. Am. Chem. Soc. 124, 11771–11780.
308
Sabine Van Doorslaer and Filip Desmet
Kappl, R., and Hu¨tterman, J. (1989). An ENDOR study of nitrosyl myoglobin single crystals. Isr. J. Chem. 29, 73–84. Kaupp, M., Bu¨hl, M., and Malkin, V. G. (eds.), (2004). In ‘‘Calculations of NMR and EPR parameters.’’ Wiley-VCH, Weinheim. Kundu, S., Trent, J. T., III, and Hargrove, M. S. (2003). Plants, humans and hemoglobins. Trends Plant Sci. 8, 387–393. Lawson, D. M., Stevenson, C. E. M., Andrew, C. R., George, S. J., and Eady, R. R. (2003). A two-faced molecule offers NO explanation: The proximal binding of nitric oxide to haem. Biochem. Soc. Trans. 31, 553–557. Lee, H. I., Dexter, A. F., Fann, Y. C., Lakner, F. J., Hager, L. P., and Hoffman, B. M. (1997). Structure of the modified heme in allylbenzene-inactivated chloroperoxidase determined by Q-band CW and pulsed ENDOR. J. Am. Chem. Soc. 119, 4059–4069. Liesum, L., and Schweiger, A. (2001). Multiple quantum coherence in HYSCORE spectra. J. Chem. Phys. 114, 9478–9488. LoBrutto, R., Wei, Y. H., Mascarenhas, R., Scholes, C. P., and King, T. E. (1983). Electron nuclear double resonance and electron paramagnetic resonance study on the structure of the NO-ligated heme a3 in cytochrome c oxidase. J. Biol. Chem. 258, 7437–7448. LoBrutto, R., Wei, Y. H., Yoshida, S., Van Camp, H. L., Scholes, C. P., and King, T. E. (1984). Electron paramagnetic resonance- (EPR-) resolved kinetics of cryogenic nitric oxide recombination to cytochrome c oxidase and myoglobin. Biophys. J. 45, 473–479. Mims, W. B. (1964). Electric field effects in spin echoes. Phys. Rev. 133, A835–A840. Mims, W. B. (1965). Pulsed ENDOR experiments. Proc. R. Soc. Lond. 283, 452–457. Mo¨bius, K., Savitsky, A., Schnegg, A., Plato, M., and Fuchs, M. (2005). High-field EPR spectroscopy applied to biological systems: Characterization of molecular switches for electron and ion transfer. Phys. Chem. Chem. Phys. 7, 19–42. Morse, R. H., and Chan, S. I. (1980). Electron paramagnetic resonance studies of nitrosyl ferrous heme complexes: Determination of an equilibrium between two conformations. J. Biol. Chem. 255, 7876–7882. Neese, F., and Solomon, E. I. (2003). Interpretation and calculation of spin-Hamiltonian parameters in transition metal complexes. In ‘‘Magnetism: Molecules to Materials IV’’ ( J. S. Miller and M. Drillon, eds.), pp. 345–466. Wiley-VCH, Weinheim. Packer, L. (ed.), (1996). In ‘‘Methods in Enzymology,’’ Vol. 269. Academic Press, San Diego. Patchkovskii, S., and Ziegler, T. (2000). Structural origin of two paramagnetic species in sixcoordinated nitrosoiron(II) porphyrins revealed by density functional theory Analysis of the g tensors. Inorg. Chem. 39, 5354–5364. Peisach, J., Mims, W. B., and Davis, J. L. (1979). Studies of the electron-nuclear coupling between Fe(III) and 14N in cytochrome P450 and in a series of low spin heme compounds. J. Biol. Chem. 254, 12379–12389. Perutz, M. (1979). Regulation of oxygen affinity of hemoglobin: Influence of structure of the globin on the heme iron. Annu. Rev. Biochem. 48, 327–386. Pesce, A., Bolognesi, M., Bocedi, A., Ascenzi, P., Dewilde, S., Moens, L., Hankeln, T., and Burmester, T. (2002). Neuroglobin and cytoglobin: Fresh blood for the vertebrate globin family. EMBO Rep. 3, 1146–1151. Pesce, A., de Sanctis, D., Nardini, M., Dewilde, S., Moens, L., Hankeln, T., Burmester, T., Ascenzi, P., and Bolognesi, M. (2004). Reversible hexa- to penta-coordination of the heme Fe atom modulates ligand binding properties of neuroglobin and cytoglobin. IUBMB Life 56, 657–664. Piknova, B., Gladwin, M. T., Schechter, A. N., and Hogg, N. (2005). Electron paramagnetic resonance analysis of nitrosylhemoglobin in humans during NO inhalation. J. Biol. Chem. 280, 40583–40588. Qu, K., Vaughn, L., Sienkiewicz, A., Scholes, C. P., and Fetrow, J. S. (1997). Kinetics and motional dynamics of spin-labeled yeast iso-1-cytochrome c. 1. Stopped-flow electron
EPR Methods to Study Ferric and NO-Ligated Globins
309
paramagnetic resonance as a probe for protein folding/unfolding of the C-terminal helix spin-labeled at cysteine 102. Biochemistry 36, 2884–2897. Quinn, R., Valentine, J. S., Byrn, M. P., and Strouse, C. E. (1987). Electronic structure of low-spin ferric porphyrins: A single-crystal EPR and structural investigation of the influence of axial ligand orientation and the effects of pseudo-Jahn-Teller distortion. J. Am. Chem. Soc. 109, 3301–3308. Raitsimring, A. M., Borbat, P., Shokhireva, T. K., and Walker, F. A. (1996). Magnetic field (g-value) dependence of proton hyperfine couplings obtained from ESEEM measurements: Determination of the orientation of the magnetic axes of model heme complexes in glassy media. J. Am. Chem. Soc. 100, 5235–5244. Rich, A. M., Armstrong, R. S., Ellis, P. J., and Lay, P. A. (1998). Determination of the Feligand bond lengths and Fe-N-O bond angles in horse heart ferric and ferrous nitrosylmyoglobin using multiple-scattering XAFS analyses. J. Am. Chem. Soc. 120, 10827–10836. Schmidt, P. P., Kappl, R., and Hu¨ttermann, J. (2001). On the Mode of hexacoordinated NO-Binding to myo- and hemoglobin: Variable-temperature EPR studies at multiple microwave frequencies. Appl. Magn. Reson. 21, 423–440. Scholes, C. P., Falkowski, K. M., Chen, S., and Bank, J. (1986). Electron nuclear double resonance (ENDOR) of bis(imidazole)-ligated low-spin ferric heme systems. J. Am. Chem. Soc. 108, 1660–1671. Scholes, C. P., Lapidot, A., Mascarenhas, R., Inubushi, T., Isaacson, R. A., and Feher, G. (1982). Electron nuclear double resonance (ENDOR) from heme and histidine nitrogens in single crystals of aquometmyoglobin. J. Am. Chem. Soc. 104, 2724–2735. Schweiger, A. (1991). Pulsed electron spin resonance spectroscopy: Basic principles, techniques, and examples of applications. Angew. Chem. Int. Ed. Engl. 30, 265–292. Schweiger, A., and Jeschke, G. (2001). In ‘‘Principles of Pulse Electron Paramagnetic Resonance.’’ Oxford: Oxford University Press, Oxford. Smith, D. T., and Pilbrow, J. R. (1980). ESR of iron proteins. In ‘‘Biological Magnetic Resonance’’ (L. J. Berliner and J. Reuben, eds.), Vol. 2, p. 85. Plenum, New York. Svistunenko, D. A., Sharpe, M. A., Nicholls, P., Blenkinsop, C., Davies, N. A., Dunne, J., Wilson, M. T., and Cooper, C. E. (2000). The pH dependence of naturally occurring low-spin forms of methaemoglobin and metmyoglobin: An EPR study. Biochem. J. 351, 595–605. Taylor, C. P. S. (1977). The EPR of low spin heme complexes: Relation of the t2g hole model to the directional properties of the g tensor, and a new method for calculating the ligand field parameters. Biochim. Biophys. Acta 491, 139–148. Trandafir, F., ter Heerdt, P., Fittipaldi, M., Vinck, E., Dewilde, S., Moens, L., and Van Doorslaer, S. (2007). Studying high-spin ferric heme proteins using pulsed EPR spectroscopy: Analysis of the ferric form of the E7Q mutant of human neuroglobin. Appl. Magn. Reson. 31, 553–572. Trandafir, F., Van Doorslaer, S., Dewilde, S., and Moens, L. (2004). Temperature dependence of NO binding modes in human neuroglobin. Biochim. Biophys. Acta 1702, 153–161. Tyryshkin, A. M., Dikanov, S. A., Reijerse, E. J., Burgard, C., and Hu¨ttermann, J. (1999). Characterization of bimodal coordination structure in nitrosyl heme complexes through hyperfine couplings with pyrrole and protein nitrogens. J. Am. Chem. Soc. 121, 3396–3406. Usov, O. M., Choi, P. S. T., Shapleigh, J. P., and Scholes, C. P. (2006). ENDOR of NOligated cytochrome c0 . J. Am. Chem. Soc. 128, 5021–5032. Utterback, S. G., Doetschman, D. C., Szumowski, J., and Rizos, A. K. (1983). EPR study of the structure and spin distribution at the binding site in human nitrosylhemoglobin single crystals. J. Chem. Phys. 78, 5874–5880.
310
Sabine Van Doorslaer and Filip Desmet
Van Camp, H. L., Scholes, C. P., and Mulks, C. F. (1976). Electron nuclear double resonance on heme compounds: ENDOR from the iron ligands in protohemin chloride and protohemin bromide. J. Am. Chem. Soc. 98, 4094–4098. Van Doorslaer, S., and Schweiger, A. (1999). New hyperfine-decoupling schemes in electron paramagnetic resonance spectroscopy. Chem. Phys. Lett. 281, 297–305. van Kan, P. J. M., van der Horst, E., Reijerse, E. J., van Bentum, P. J. M., and Hagen, W. R. (1998). Multi-frequency EPR spectroscopy of myoglobin: Spectral effects for high-spin iron(III) ion at high magnetic fields. J. Chem. Soc. Faraday Trans. 94, 2975–2978. Veselov, A. V., Osborne, J. P., Gennis, R. B., and Scholes, C. P. (2000). Q-band ENDOR (electron nuclear double resonance) of the heme o3 liganding environment at the binuclear center in cytochrome bo3 from. Escherichia coli. J. Am. Chem. Soc. 122, 8712–8716. Vinck, E., and Van Doorslaer, S. (2004). Analysing low-spin ferric complexes using pulse EPR techniques: A structure determination of bis(4-methylimidazole)(tetraphenylporphyrinato)iron(III). Phys. Chem. Chem. Phys. 6, 5324–5330. Vinck, E., Van Doorslaer, S., Dewilde, S., Mitrikas, G., Schweiger, A., and Moens, L. (2006). Analyzing heme proteins using EPR techniques: The heme-pocket structure of ferric mouse neuroglobin. J. Biol. Inorg. Chem. 11, 467–475. Vinck, E., Van Doorslaer, S., Dewilde, S., and Moens, L. (2004). Structural change of the heme pocket due to disulfide bridge formation is significantly larger for neuroglobin than for cytoglobin. J. Am. Chem. Soc. 126, 4516–4517. Walker, F. A., Huynh, B. H., Scheidt, W. R., and Osvath, S. R. (1986). Models of the cytochromes. b. Effect of axial ligand plane orientation on the EPR and Moessbauer spectra of low-spin ferrihemes. J. Am. Chem. Soc. 108, 5288–5297. Weber, R. E., and Vinogradov, S. N. (2001). Nonvertebrate hemoglobins: Functions and molecular adaptations. Physiol. Rev. 81, 569–628. Weil, J. A., Bolton, R., and Wertz, J. E. (1994). ‘‘Electron Paramagnetic Resonance.’’ Clarendon, Oxford. Wittenberg, J. B., and Wittenberg, B. A. (1990). Mechanisms of cytoplasmic hemoglobin and myoglobin function. Annu. Rev. Biophys. Chem. 19, 217–241. Wyllie, G. R. A., Schulz, C. E., and Scheidt, W. R. (2003). Five- to six-coordination in (nitrosyl)iron (II) porphyrinates: Effects of binding the sixth ligand. Inorg. Chem. 42, 5722–5734. Yonetani, T., Tsuneshige, A., Zhou, Y., and Chen, X. (1998). Electron paramagnetic resonance and oxygen binding studies of alpha-nitrosyl hemoglobin: A novel oxygen carrier having NO-assisted allosteric functions. J. Biol. Chem. 273, 20323–20333. Yonetani, T., Yamamoto, H., Erman, J. E., Leigh, J. S., Jr., and Reed, G. H. (1972). Electromagnetic properties of hemoproteins. V. Optical and electron paramagnetic resonance characteristics of nitric oxide derivatives of metalloporphyrin-apohemoprotein complexes. J. Biol. Chem. 247, 2447–2455. Zhang, Y., Gossman, W., and Oldfield, E. (2003). A density functional theory investigation of Fe-N-O bonding in heme proteins and model systems. J. Am. Chem. Soc. 125, 16387–16396. Zhi, Z., Guenzburger, D., and Ellis, D. E. (2004). Magnetic and hyperfine properties of deoxymyoglobin and nitrosyl-myoglobin. J. Mol. Struct. THEOCHEM 678, 145–156.
C H A P T E R
S I X T E E N
Oxygen Binding to Heme Proteins in Solution, Encapsulated in Silica Gels, and in the Crystalline State Luca Ronda, Stefano Bruno, Serena Faggiano, Stefano Bettati, and Andrea Mozzarelli Contents 1. Oxygen-Binding Curves to Heme Proteins 2. Determination of OBCs for Hemoglobin in Solution 3. Determination of K1 for Hemoglobin in Solution in the Absence of Allosteric Effectors 4. Determination of OBCs for T State Hemoglobin Gels in the Absence and Presence of Allosteric Effectors 4.1. Protocol A 4.2. Protocol B 5. Determination of OBCs for T State Hemoglobin Crystals 6. Determination of OBCs for Hemocyanin in Solution and in Silica Gels Acknowledgments References
313 316 318 318 319 320 320 323 325 325
Abstract The determination of accurate oxygen-binding curves for heme-containing proteins is a demanding task. In fact, great care is required in the (i) preparation of accurate gas mixtures at defined oxygen partial pressures, (ii) precise measurement of changes in protein absorbance, (iii) calculation of the fraction of oxygen-containing sites, and (iv) analysis of the dependence of fractional saturation on oxygen pressure using phenomenological or model-dependent equations. Over the years, methods have been developed for the determination of oxygen-binding curves based either on discrete steps in oxygen partial pressure (‘‘static’’ method) or on continuous variations (‘‘dynamic’’ method). This work presents a novel, versatile setup that allows one to determine oxygen-binding curves for heme and nonheme proteins in solution, Department of Biochemistry and Molecular Biology, University of Parma, Parma, Italy Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37016-X
#
2008 Elsevier Inc. All rights reserved.
311
312
Luca Ronda et al.
encapsulated in wet, nanoporous silica gels, in the crystalline state, and for hemoglobin within single red blood cells. The apparatus is composed of a tandem of high-precision gas mixture generators and either an equilibration chamber coupled to a spectrophotometer cuvette or a gas-tight flow cell, placed on the stage of a microspectrophotometer, for immobilized samples down to a few micrometers in size.
Oxygen plays several roles in biological systems, being the substrate for several enzyme reactions and the final electron acceptor in oxidative phosphorylation. The transport of oxygen within the body is obtained in vertebrates by the interplay of hemoglobin (Hb) and myoglobin (Mb), with the former loading oxygen in the lungs and delivering it to tissues and, eventually, to the latter, storing it in the muscle. Many physiological functions depend on the efficiency of oxygen transport, on the relative concentration of oxygen with respect to carbon dioxide and protons, and, ultimately, on the oxygen affinity of the heme-binding proteins and their response to allosteric effectors. Therefore, several methods have been developed for the precise evaluation of the saturation degree of Hb and Mb as a function of oxygen pressure. The same methods can be adapted to the determination of oxygen-binding curves (OBCs) for other heme and nonheme proteins, as well as to the determination of binding curves for other gaseous ligands, such as carbon monoxide and nitric oxide. Two basic requirements must be fulfilled: (i) preparation of gas mixtures at variable concentrations that equilibrate fast with the sample and (ii) evaluation of the oxygen fractional saturation, i.e., the ratio between the concentration of oxygenated and total sites. The oxygen partial pressure is usually expressed in torr, and the oxygen affinity is expressed as p50, i.e., the oxygen pressure at 50% saturation, or its reciprocal value, K, in torr1. The p50 for Hb under human physiological conditions, pH 7.4, at 37 , is 26 1 torr, which corresponds to a K of 0.0385 torr1. Given the relevance of a precise measurement of the p50 for the evaluation of Hb function and regulation and for the detection of pathological conditions, methods for the determination of OBCs have been developed since the beginning of the 20th century (Van Slyke and Neill, 1924). Two distinct modes have been used for OBC determination: ‘‘dynamic’’ and ‘‘static.’’ In ‘‘static’’ methods, OBCs are determined by producing stepwise changes in the oxygen pressure to which the sample is exposed and by measuring the corresponding fractional saturation, usually recording the spectral changes of the heme protein upon ligand binding either at a single wavelength or within a defined spectral range. In ‘‘dynamic’’ methods, OBCs are determined by producing a continuous, slow change in oxygen pressure in the atmosphere surrounding a dilute sample solution that is usually present in a very thin layer to obtain fast equilibration. The simultaneous recording of absorbance changes at defined wavelengths and partial oxygen pressures by an oxygen
Oxygen-Binding Curves for Heme Proteins
313
electrode allows one to determine the OBC. There are advantages and disadvantages with both approaches, which are discussed thoroughly by Imai (1982). Other pros and cons on ‘‘dynamic’’ and ‘‘static’’ methods have been reported by Giardina and Amiconi (1981), Gill (1981), Imai (1981), and Lapennas et al. (1981). It is generally assumed that ‘‘static’’ methods are more precise but more time-consuming, whereas ‘‘dynamic’’ methods are faster and minimize oxidation, but are less precise. An improved ‘‘dynamic’’ system has been developed by Yonetani and co-workers (2002) to determine OBCs for Hb over a wide range of experimental conditions. Instrumentation has been developed for the determination of OBCs on Hb crystals (Mozzarelli et al., 1991; Rivetti et al., 1993a) and Hb encapsulated in silica gels (Bettati and Mozzarelli, 1997). In this case, because of the limitation imposed by the slow diffusion of oxygen within either crystals or gels, only a ‘‘static’’ method can be applied.
1. Oxygen-Binding Curves to Heme Proteins In order to develop a more efficient and versatile apparatus that allows one to obtain OBCs for heme proteins in solution or immobilized either in the crystalline state or in polymeric matrices, a novel system was designed (Fig. 16.1). The preparation of gas mixtures is obtained by using two gas mixture generators, Environics 200 and 4000, based on mass-flow controllers. The gas mixers can work either separately, delivering different gas mixtures via dedicated lines to the equilibration chamber or to a flow cell, or in series (a gas mixture prepared by one mixer can be fed into the other), allowing sequential dilution or addition of other components. The Environics 4000 gas mixer has the capability of controlling the relative humidity from 0 to 99% through a rear panel-mounted device, consisting of a humidifier and a 250-ml gravity water reservoir. The humidifier is a Nafion waterÒ vapor-permeable tube, generating a gas mixture at a defined degree of humidity within 1 min, thus allowing the investigation of oxygen binding as a function of hydration. Gas humidification can also be obtained by bubbling the gas mixture in a thermostated solution, with the same composition of gel or crystal suspending medium. The humidifier bottle is equipped with a GL45 polypropylene cap adapted with gas-tight fittings. The precise control of the temperature and the composition of the solution in the humidifier avoids any change in the concentration of the equilibrating sample. This system is necessary for prolonged oxygen-binding measurements on crystals, where the precipitant [polyethylene glycol, (NH4)2SO4, phosphate] concentration is critical for crystal stability. Up to six gas cylinders are attached to the gas mixture generators, thus providing great flexibility in the preparation
Thermostated humidifier
Nafion® humidifier
Environics 4000 Port 1
Output
Port 3
Port 2
Environics 200
Output
DIL
Port 1
Port 2
Port 3
Port 4
A
Spectrophotometer
B
He
CO2
O2 1%
O2 5%
O2 20%
O2 100%
Microspectrophotometer
Thermostated humidifier
Figure 16.1 Scheme of the apparatus for measuring OBCs on heme proteins either in solution or immobinilized in wet silica gels or single crystals. (A) The equilibration chamber, placed in the sample compartment of a spectrophotometer. (B) The gas-tight flow cell, placed on the microspectrophotometer stage.
Oxygen-Binding Curves for Heme Proteins
315
of gas mixtures. The gas-mixing apparatus can produce a gas flow between 5 and 250 ml/min, but usually is set at 50 ml/min. Tests using an oxygen electrode (FOXY sensor, Ocean Optics) indicated that the precision and reproducibility of the prepared gas mixtures are so high that an online oxygen electrode for the determination of oxygen partial pressures is not normally required. Moreover, the use of two mixers in series allows one to dilute oxygen down to a hundredth of a torr, making the determination of OBCs accessible to heme proteins with very low p50. The humidified gas mixture flows, via Swagelok stainless steel 316 tubing, valves, and fittings and NO-OX (Kontes) tubing, to the sample chamber. For solution experiments, the sample chamber consists of a 2-mm optical path length cuvette fused to a 25-ml open-top threaded reservoir, where a screw cap with fittings for inlet and outlet gas lines can be mounted. Unlike classic tonometers, where the reservoir is bigger (500 ml) and the oxygen is added with a syringe in a closed volume, this sample chamber has a continuous gas flow passing through, maintaining the defined oxygen partial pressure, with no pressure changes within the chamber. During sample equilibration, the chamber is thermostated in a shaking bath, and the cuvette is thermostated when inserted in the sample compartment of a CARY 400 spectrophotometer (Varian) to measure absorption spectra. For the determination of OBCs on immobilized samples, such as Hb crystals, Hb gels, or single red blood cells, the gas mixtures are flown into a gastight Dvorak–Stotler flow cell (Dvorak and Stotler, 1971), where the sample is covered with a gas-permeable membrane (MEM13, General Electric) (Gill, 1981). This membrane does not polarize light and, therefore, is suitable for recording polarized absorption spectra. The cell is placed on the thermostated stage of a polarized absorption microspectrophotometer (Zeiss MPM03) (Pearson et al., 2004; Rivetti et al., 1993a). A microspectrophotometer consists of a high-quality microscope, usually equipped with quartz optics, and a single-beam spectrophotometer. A peculiar feature of a microspectrophotometer is the possibility of determining OBCs on samples of the order of a few micrometers in diameter. This is particularly relevant for precious biological samples or small heme protein crystals. Indeed, OBCs were determined on single red blood cells from the turtle Trachemys scripta (Frische et al., 2001). Spectral measurements on single red cells using a microspectrophotometer were carried out previously (Coletta et al., 1987; Mozzarelli et al., 1987). With a microspectrophotometer, high-quality spectra are obtained by double masking the transmitted light beam in order to select the measuring field and to reject scattered light (Rivetti et al., 1993a). Spectra of protein crystals are recorded with linearly polarized light parallel to the crystal optical directions (Eaton and Hofrichter, 1981; Rivetti et al., 1993a). Spectra for both soluble and solid samples are usually recorded over the spectral range 450–700 nm. Fractional saturations with oxygen and
316
Luca Ronda et al.
fraction of oxidized hemes are determined by fitting observed spectra to a linear combination of reference spectra (Bettati and Mozzarelli, 1997; Rivetti et al., 1993a), i.e., spectra of pure deoxygenated Hb, oxygenated Hb, and oxidized Hb, plus a baseline, and, in the case of gels, a slope to take into account the nonperfect optical quality of the gel surface. It should be noted that a set of metHb reference spectra at different pH values is required because metHb spectra change as a function of pH. This procedure leads to a very precise determination of the fractional saturation because, using the absorbance changes at 251 different wavelengths, it is possible to discriminate the effect of oxygenation, oxidation, and background drifts on absorbance changes and to take into account small changes in protein concentration or crystal and gel thickness. Furthermore, in the case of Hb crystals, because the a and b hemes contribute differently to the absorption intensity along the two polarization directions, the separate OBCs for each subunit can be calculated, provided that the heme orientation within the crystal lattice is known (Mozzarelli et al., 1997; Rivetti et al., 1993a). A typical procedure for the determination of OBCs for five distinct systems is reported next. i. OBCs for Hb in solution. ii. Evaluation of K1 for Hb in solution, in the absence of allosteric effectors. iii. OBCs for T state Hb gels, in the absence and presence of allosteric effectors. iv. OBCs for T state Hb crystals. v. OBCs for hemocyanin in solution and in silica gels.
2. Determination of OBCs for Hemoglobin in Solution A solution with a volume of 400–500 ml, containing 180 mM Hb, 100 mM HEPES, 1 mM EDTA, pH 7, is placed in the sample chamber. The Hayashi enzymatic reducing system (Hayashi et al., 1973) is added to prevent metHb accumulation. The solution is equilibrated for 120 min with a gas mixture composed of 95% helium and 5% CO2 at 15 . The spectrum of deoxyHb is recorded (Fig. 16.2A). In the subsequent seven to eight steps, humidified oxygen mixtures, ranging between 0 and 760 torr, are flown into the sample chamber for 30 min and then spectra are recorded (see Fig. 16.2A). The fitting of observed spectra to deoxyHb, oxyHb, and metHb reference spectra, plus a baseline, allows the determination of the amount of metHb and of the fractional saturation with oxygen of the reduced hemes. Reference spectra and a representative analysis of a spectrum recorded at 2.86 torr are reported in Fig. 16.2B. The OBC, showing
317
Oxygen-Binding Curves for Heme Proteins
A
B 0.6
0.4
0.4
760 torr
0.3
0 torr
Absorbance
Absorbance
0.5 0.3
0.2
0.2 0.1 0.1 0.0 450
500
550 600 650 Wavelength (nm)
700
0.0 450
500
550 600 650 Wavelength (nm)
700
Fractional saturation
C 1.0 0.8 0.6 0.4 0.2 0.0
0
2
4 6 pO2 (torr)
8
10
Figure 16.2 OBC for Hb in solution. (A) Absorption spectra of a solution containing 180 mM Hb, 100 mM HEPES, 1 mM EDTA, 5% CO2, pH 7.0, 15, at increasing oxygen pressures. (B) Fitting of the spectrum, recorded at 2.86 torr, to a linear combination of deoxyHb (short dashed line), oxyHb (long dashed line), metHb (dash-dot-dotted line), spectra, and baseline (dash-dotted line), essentially undetectable. Calculated (dotted line) and observed (solid line) spectra are almost indistinguishable. The calculated fractional saturation with oxygen is 0.267, and the fraction of oxidized hemes is 0.059. (C) The OBC was analyzed according to Eq. (16.1). The calculated p50 is 4.09 0.03 torr and the Hill coefficient is 2.70 0.05.
the typical sigmoidal shape of Hb in solution (see Fig. 16.2C), is analyzed using Eq. (16.1):
Y¼
pn pn þ p50n
ð16:1Þ
where Y is the fractional saturation, p is the oxygen partial pressure in torr, and n is the Hill coefficient.
318
Luca Ronda et al.
The calculated p50 and the Hill coefficient are 4.09 0.03 torr and 2.70 0.05, respectively. Full reversibility of oxygen binding can be demonstrated by exposing the Hb solution backward to the same oxygen pressures.
3. Determination of K1 for Hemoglobin in Solution in the Absence of Allosteric Effectors The determination of Hb affinity for the first binding oxygen is a critical measurement for evaluation of the affinity of the T quaternary state because it requires the determination of very low fractional saturations, at low oxygen pressures, in a small range of spectral changes. As described earlier, the Hb solution, containing 100 mM HEPES, 1 mM EDTA, pH 7.0, 15 , is placed in the chamber and equilibrated thoroughly with a helium atmosphere in the presence of the enzymatic reducing system (Hayashi et al., 1973) to remove any metHb trace. The recorded absorption spectrum (Fig. 16.3A) indicates a fully deoxygenated Hb. Oxygen gas mixtures with partial pressures between 0 and 0.17 torr are prepared and fluxed into the chamber. Absorption spectra, recorded at each oxygen pressure (see Fig. 16.3A), are analyzed using reference spectra. The initial part of the OBC, reported in Fig. 16.3B, leads to an estimated K1 of 0.068 0.007 torr1.
4. Determination of OBCs for T State Hemoglobin Gels in the Absence and Presence of Allosteric Effectors The encapsulation of proteins in wet, nanoporous silica gels is a convenient method to select conformational states that are metastable or poorly populated in solution, to stabilize proteins against thermal and chemical denaturation, and to decrease the rate of conformational transitions (Avnir et al., 1994; Bettati et al., 2004; Ellerby et al., 1992; Gill and Ballesteros, 2000; Mozzarelli and Bettati, 2001). In the case of Hb, gel encapsulation was exploited to trap either the T or the R quaternary state and to characterize their functional and structural properties (Abbruzzetti et al., 2001; Bettati and Mozzarelli, 1997; Bruno et al., 2001; Juszczak and Friedman, 1999; Khan et al., 2000; Ronda et al., 2006; Shibayama and Saigo, 1995, 1999, 2001; Sottini et al., 2004, 2005; Viappiani et al., 2004). DeoxyHb was encapsulated in silica gel either in the presence (protocol A) or in the absence of allosteric effectors (protocol B) (Ronda et al., 2006).
319
Oxygen-Binding Curves for Heme Proteins
A
0 torr
Absorbance
0.6
Absorbance
0.72
0.8
0.68
0.64
540
0.4
550 560 570 Wavelength (nm)
0.17 torr
0.2
0.0 450
500
550 600 650 Wavelength (nm)
700
0.16
0.20
Fractional saturation
B 0.016
0.012
0.008
0.004
0.000 0.00
0.04
0.08 0.12 pO2 (torr)
Figure 16.3 Determination of K1 for Hb in solution in the absence of allosteric effectors. (A) Representative absorption spectra of a solution containing 280 mM Hb, 100 mM HEPES, 1 mM EDTA, pH 7.0, 15, recorded at increasing oxygen pressures between 0 and 0.17 torr. (Inset) Close-up view of all recorded spectra. (B) Analysis of the dependence of fractional saturation on oxygen pressure, according to Eq. (16.1), leads to a calculated value of 0.068 0.007 torr1 for K1.
4.1. Protocol A A solution containing tetramethylorthosilicate, water, and hydrochloric acid is sonicated for 20 min. Then, an equal volume of a buffer solution containing 10 mM potassium phosphate, 1 mM EDTA at pH 6.0 is added. The mixture is deoxygenated by bubbling nitrogen for 40 min. Finally, 1.5 volumes of a solution containing 10 g/l deoxyHb in 50 mM potassium phosphate, 10 mM inositol hexaphosphate (IHP), 2 mM bezafibrate, 1 mM EDTA, 10 mM sodium dithionite, pH 7.2, is added anaerobically to the mixture.
320
Luca Ronda et al.
Gelification occurs in a few minutes, at 4 . A solution containing 100 mM HEPES, 10 mM IHP, 2 mM bezafibrate, 200 mM Cl, 1 mM EDTA, 30 mM sodium dithionite, pH 7.0, is layered on the gel for storage.
4.2. Protocol B A solution containing 10 mM HEPES, 1 mM EDTA at pH 6.2 is added to an equal volume of tetramethylorthosilicate and vortexed for 2 min. The mixture is deoxygenated by bubbling humidified helium for 90 min at 4 . A solution containing 10 g/l deoxyHb in 10 mM HEPES, 1 mM EDTA, 10 mM sodium dithionite, pH 6.2, is finally added. The solution mixture is left at 4 for 30 min. Gelification occurs in about 10 min after the mixture is brought to room temperature. A solution containing 100 mM HEPES, 1 mM EDTA, 30 mM sodium dithionite at pH 7.0 is layered on the gel for storage. All samples of encapsulated Hb are made into 1-mm thin layers by anaerobically pouring 200 ml of the sol mixture onto 9 35-mm panes of quartz before gelification. The pane of quartz fits in standard 1-cm cuvettes. For oxygen-binding measurements, a small, thin fragment of Hb gel is loaded anaerobically in the Dvorak–Stotler flow cell (Dvorak and Stotler, 1971) of the microspectrophotometric apparatus. Oxygen-binding measurements are carried out by equilibrating with distinct oxygen pressures the T state Hb gels prepared with either protocol A (presence of allosteric effectors) or protocol B (absence of allosteric effectors). The equilibration of the Hb gel with the gas phase is followed with the collection of absorption spectra every 2 min (Fig. 16.4A). The equilibration is complete after 20 min (see Fig. 16.4A, inset). T state Hb gels in the presence of allosteric effectors exhibit a p50 of 137.2 2.3 torr, whereas those in the absence of allosteric effectors exhibit a p50 of 25.9 1.0 torr (see Fig. 16.4B). Remarkably, the Hill coefficient is 0.91 0.02 and 0.95 0.03, respectively, indicating that oxygen binding is noncooperative (see Fig. 16.4B). This finding is fully in keeping with the postulate of the Monod, Wyman, and Changeux (MWC) model (Monod et al., 1965) that ligand binding is not cooperative within a quaternary state and, with its recently proposed extension, the tertiary two-state allosteric model, accounting for noncooperative tertiary changes within a quaternary state (Eaton et al., 2007; Henry et al., 2002; Viappiani et al., 2004).
5. Determination of OBCs for T State Hemoglobin Crystals The functional characterization of a protein in the crystal state is dictated by the necessity of obtaining information on function and structure in the same physical state, thus allowing to draw straightforward
321
Oxygen-Binding Curves for Heme Proteins
A
0.6
Absorbance
Fractional saturation
0.6
0.4
0.4 0.2 0.0
0
20 Time (min)
40
0.2
0.0 450
500
550 600 650 Wavelength (nm)
700
Fractional saturation
B 1.0 0.8 0.6 0.4 0.2 0.0
0
200
400 pO2 (torr)
600
Figure 16.4 OBCs for T state Hb silica gels in the absence and presence of allosteric effectors. (A) Absorption spectra of Hb encapsulated in silica gels using protocol A suspended in a solution containing 100 mM HEPES, 10 mM IHP, 2 mM bezafibrate, 200 mM Cl, 1 mM EDTA, pH 7.0, 15, at 145.3 torr, as a function of equilibration time. (Inset) Time course of oxygen fractional saturation calculated by fitting spectra recorded every 2 min to a linear combination of reference spectra. (B) OBCs for Hb gels prepared in the presence (open circles) (protocol A) and absence (closed circles) (protocol B) of allosteric effectors. OBCs, analyzed according to Eq. (16.1), yield a p50 of 137.2 2.3 torr and a Hill coefficient of 0.91 0.02, and a p50 of 25.9 1.0 torr and a Hill coefficient of 0.95 0.03, respectively.
structure–function relationships (Mozzarelli and Rossi, 1996; Pearson et al., 2004). Orthorhombic crystals of T state HbA, grown from polyethylene glycol solutions (Brzozowski et al., 1984; Rivetti et al., 1993a), are mounted in the gas-tight flow cell, covered with an optically isotropic, gas-permeable silicon copolymer membrane. Hb crystals, suspended in a solution containing 10 mM potassium phosphate, 54% (w/v) PEG 8000, 2 mM IHP, 1 mM EDTA, 90 mg/ml catalase, pH 7.0, 15 , are exposed to oxygen–helium
322
Luca Ronda et al.
mixtures between 0 and 760 torr of oxygen. Polarized absorption spectra (Fig. 16.5A) are recorded on oriented crystals using linearly polarized light parallel to the crystal extinction directions, which are also parallel to the a and c crystal axis (see Fig. 16.5A, inset). Fractional saturation with oxygen and the fraction of oxidized hemes are determined by fitting observed spectra to a linear combination of polarized spectra of the pure species, deoxyHb, oxyHb, metHb, and a baseline. Because of the low oxygen affinity of T state Hb, the spectrum of the pure oxygenated species cannot be obtained even upon exposure to 1 atm of oxygen. Therefore, the oxyHb reference
A b2
1.2 c
Absorbance
a1
a a2
1.0 Ea
0.8
b1
0.6 0.4 0.2
Ec
0.0 450
500
550 600 650 Wavelength (nm)
700
Fractional saturation
B 1.0 0.8 0.6 0.4 0.2 0.0
0
200
400 pO2 (torr)
600
Figure 16.5 OBC for T state Hb crystals. (A) Polarized absorption spectra were recorded on crystals of Hb, suspended in 10 mM potassium phosphate, 54% (w/v) PEG 8000, 2 mM IHP,1 mM EDTA, 90 mg/ml catalase, pH 7.0,15, at oxygen pressures between 0 and 760 torr. (Inset) Orientation of a and b hemes in the crystal asymmetric unit with respect to the a or c crystal axis. (B) Calculated OBC for a (closed circles) and b (open circles) hemes. Fractional saturations were determined according to Rivetti et al. (1993a), and calculated p50s for a and b hemes were 94 6 and 235 10 torr, respectively.
Oxygen-Binding Curves for Heme Proteins
323
spectrum is obtained by recording spectra at different oxygen pressures between 160 and 760 torr, at 5 to lower the p50, and by extrapolating the series of spectra at infinite oxygen pressure (Rivetti et al., 1993a). The corresponding OBCs exhibit a p50 of 139 5 torr, along the a axis, and 132 6 torr, along the c axis. The higher affinity recorded along the c axis is related to the higher projection of a hemes along this axis (see Fig. 16.5A, inset) (Rivetti et al., 1993a). On the basis of a structural–functional analysis, it was calculated (see Fig. 16.5B) that a hemes bind oxygen with an affinity twofold higher than b hemes (Mozzarelli et al., 1997). This finding was also confirmed by OBCs measured on metal hybrids (Bettati et al., 1996; Bruno et al., 2000). The calculated Hill coefficient is very close to unity, 0.94 0.01, both along the a and c axis. This value, as well as those measured on crystals of Hb mutants and metal hybrids (Bettati et al., 1996, 1997, 1998; Bruno et al., 2000; Kavanaugh et al., 1995, 2001; Noble et al., 2001, 2002; Rivetti et al., 1993b), is consistent with the MWC model (Eaton et al., 1999; Monod et al., 1965). The procedure was also applied to the measurement of OBCs for dimeric Scapharca inequivalvis HbI crystals (Mozzarelli et al., 1996) and Vitreoscilla Hb crystals (Bolognesi et al., 1999). In the former case, it was found that the crystal lattice does not hamper cooperativity in oxygen binding, whereas, in the latter case, data suggest that the ligand-free homodimer, observed in the crystalline state, is constrained in a low-affinity conformation whose ligand-binding properties closely resemble those of the monoligated species in solution.
6. Determination of OBCs for Hemocyanin in Solution and in Silica Gels As an example of the potentiality of this novel apparatus, the measurement of OBCs for the nonheme-containing protein hemocyanin from the tarantula Eurypelma californicum (Decker et al., 1988), both in solution and encapsulated in silica gels, is described. In order to isolate the quaternary conformations of hemocyanin, the 24-meric giant protein was encapsulated in wet, nanoporous silica gels, either in the absence or in the presence of oxygen. Spectra of hemocyanin in solution (Fig. 16.6A) and in gels (not shown) were collected at oxygen pressures between 0 and 760 torr, pH 7.0, at 4 . Because the visible spectrum of deoxygenated hemocyanin is featureless (see Figs. 16.6A and 16.6B), fractional saturations were calculated from the intensity at 340 nm of the Cu2-O2 absorption band. OBCs measured on the deoxy- and oxy-encapsulated protein (see Fig. 16.6B) exhibit a Kavg of 0.09 0.02 and 0.40 0.16 torr1, respectively, in close agreement with the affinity calculated for hemocyanin in solution at low and high oxygen
324
Luca Ronda et al.
A
0.12
Absorbance
760 torr
0.08 0 torr
0.04
0.00 300
320
340 360 380 Wavelength (nm)
400
1
10 100 pO2 (torr)
1000
Fractional saturation
B 1.0 0.8 0.6 0.4 0.2 0.0
Figure 16.6 OBCs for E. californicum hemocyanin in solution and in silica gel. (A) Spectra of hemocyanin in solution, in 50 mM Bis-Tris propane, 50 mM Tris, 5 mM MgCl2, 5 mM CaCl2, pH 7.0, at 4, recorded at increasing oxygen pressures between 0 and 760 torr. (B) The OBC for hemocyanin in solution (open circles) was analyzed according to the MWC model (Monod et al., 1965), leading to a KT of 0.10 0.01 torr1, KR of 0.75 0.36 torr1 OBCs for hemocyanin encapsulated in deoxy (filled squares) and oxy (filled circles) conformations, analyzed assuming two independent binding sites, gave an averaged K of 0.09 0.02 and 0.40 0.16 torr1, respectively.
pressures. The observed Hill coefficients, 0.72 0.11 and 0.82 0.05 for deoxy- and oxyhemocyanin gels, respectively, are significantly lower than unity, indicating a conformational heterogeneity within each locked conformational state, a finding in agreement with the assumption that at least four conformational states are required to explain the oxygen-binding properties of E. californicum hemocyanin in solution (Ronda et al., 2007). Overall, key features of the setup described are the (i) accuracy in the determination of OBCs using either a conventional spectrophotometer or a microspectrophotometer; (ii) easy generation of complex gas mixtures,
Oxygen-Binding Curves for Heme Proteins
325
containing up to four components, with also very dilute components; (iii) flexibility of recording OBCs on both soluble and immobilized samples, with measuring fields of a few micrometers; (iv) precise multiwavelength evaluation of the fractional saturation, taking into account sample damaging due, for example, to oxidation. We are currently developing a fully automatized system for the determination of OBCs for heme and nonheme proteins operating in either a ‘‘static’’ or a ‘‘dynamic’’ mode. In this case, a thin layer of a sample solution is placed in a modified sample holder within the gas-tight flow cell on the microspectrophotometer stage. By controlling the gas mixture preparation, the timing of the delivery, and the spectra recording simultaneously, OBCs can be determined under easily adjustable conditions.
ACKNOWLEDGMENTS We thank Drs. William A. Eaton, Eric R. Henry, and Robert W. Noble for valuable advice and discussions during development of this instrumentation. This work was supported in part by a grant from the European Union (LSHB.CT-2004–503023 to A.M.).
REFERENCES Abbruzzetti, S., Viappiani, C., Bruno, S., Bettati, S., Bonaccio, M., and Mozzarelli, A. (2001). Functional characterization of heme proteins encapsulated in wet nanoporous silica gels. J. Nanosci. Nanotechnol. 1, 407–415. Avnir, D., Braun, S., Lev, O., and Ottolenghi, M. (1994). Enzymes and other proteins entrapped in sol-gel materials. Chem. Mater. 6, 1605–1614. Bettati, S., Kwiatkowski, L. D., Kavanaugh, J. S., Mozzarelli, A., Arnone, A., Rossi, G. L., and Noble, R. W. (1997). Structure and oxygen affinity of crystalline des-his-146beta human hemoglobin in the T state. J. Biol. Chem. 272, 33077–33084. Bettati, S., and Mozzarelli, A. (1997). T state hemoglobin binds oxygen noncooperatively with allosteric effects of protons, inositol hexaphosphate, and chloride. J. Biol. Chem. 272, 32050–32055. Bettati, S., Mozzarelli, A., and Perutz, M. F. (1998). Allosteric mechanism of haemoglobin: Rupture of salt-bridges raises the oxygen affinity of the T-structure. J. Mol. Biol. 281, 581–585. Bettati, S., Mozzarelli, A., Rossi, G. L., Tsuneshige, A., Yonetani, T., Eaton, W. A., and Henry, E. R. (1996). Oxygen binding by single crystals of hemoglobin: The problem of cooperativity and inequivalence of alpha and beta subunits. Proteins 25, 425–437. Bettati, S., Pioselli, B., Campanini, B., Viappiani, C., and Mozzarelli, A. (2004). Proteindoped nanoporous silica gels. In ‘‘Encyclopedia of Nanoscience and Nanotechnology,’’ Vol. IX, pp. 81–103. American Scientific Publishers. Bolognesi, M., Boffi, A., Coletta, M., Mozzarelli, A., Pesce, A., Tarricone, C., and Ascenzi, P. (1999). Anticooperative ligand binding properties of recombinant ferric Vitreoscilla homodimeric hemoglobin: A thermodynamic, kinetic and X-ray crystallographic study. J. Mol. Biol. 291, 637–650.
326
Luca Ronda et al.
Bruno, S., Bettati, S., Manfredini, M., Mozzarelli, A., Bolognesi, M., Deriu, D., Rosano, C., Tsuneshige, A., Yonetani, T., and Henry, E. R. (2000). Oxygen binding by alpha(Fe2þ)2beta(Ni2þ)2 hemoglobin crystals. Protein Sci. 9, 683–692. Bruno, S., Bonaccio, M., Bettati, S., Rivetti, C., Viappiani, C., Abbruzzetti, S., and Mozzarelli, A. (2001). High and low oxygen affinity conformations of T state hemoglobin. Protein Sci. 10, 2401–2407. Brzozowski, A., Derewenda, Z., Dodson, E., Dodosn, G., Grabowski, M., Liddington, R., Skarzynski, T., and Vallely, D. (1984). Bonding of molecular oxygen to T state human haemoglobin. Nature 307, 74–76. Coletta, M., Brumen, M., Giardina, B., Benedetti, P. A., and Brunori, M. (1987). A microspectroscopic analysis of ligand binding in single erythrocyte. Biomed. Biochim. Acta 46, S108–S112. Decker, H., Connelly, P. R., Robert, C. H., and Gill, S. J. (1988). Nested allosteric interaction in tarantula hemocyanin revealed through the binding of oxygen and carbon monoxide. Biochemistry 27, 6901–6908. Dvorak, J. A., and Stotler, W. F. (1971). A controlled-environment culture system for high resolution light microscopy. Exp. Cell Res. 68, 144–148. Eaton, W. A., Henry, E. R., Hofrichter, J., Bettati, S., Viappiani, C., and Mozzarelli, A. (2007). Evolution of allosteric models for hemoglobin. IUMB Life 59, 586–599. Eaton, W. A., Henry, E. R., Hofrichter, J., and Mozzarelli, A. (1999). Is cooperative oxygen binding by hemoglobin really understood? Nat. Struct. Biol. 6, 351–358. Eaton, W. A., and Hofrichter, J. (1981). Polarized absorption and linear dichroism spectroscopy of hemoglobin. Methods Enzymol. 76, 175–261. Ellerby, L. M., Nishida, C. R., Nishida, F., Yamanaka, S. A., Dunn, B., Valenitine, J. S., and Zink, J. I. (1992). Encapsulation of proteins in transparent porous silicate glasses prepared by the sol-gel method. Science 255, 1113–1115. Frische, S., Bruno, S., Fago, A., Weber, R. E., and Mozzarelli, A. (2001). Oxygen binding by single red blood cells from the red-eared turtle Trachemys scripta. J. Appl. Physiol. 90, 1679–1684. Giardina, B., and Amiconi, G. (1981). Measurement of binding of gaseous and nongaseous ligands to hemoglobins by conventional spectrophotometric procedures. Methods Enzymol. 76, 417–427. Gill, I., and Ballesteros, A. (2000). Bioencapsulation within synthetic polymers. 1. Sol-gel encapsulated biologicals. Trends Biotechnol. 18, 282–296. Gill, S. J. (1981). Measurement of oxygen binding by means of a thin-layer optical cell. Methods Enzymol. 76, 427–438. Hayashi, A., Suzuki, T., and Shin, M. (1973). An enzymic reduction system for metmyoglobin and methemoglobin, and its application to functional studies of oxygen carriers. Biochim. Biophys. Acta. 310, 309–316. Henry, E. R., Bettati, S., Hofrichter, J., and Eaton, W. (2002). A tertiary two-state allosteric model for hemoglobin. Biophys. Chem. 98, 149–164. Imai, K. (1981). Measurement of accurate oxygen equilibrium curves by an automatic oxygenation apparatus. Methods Enzymol. 76, 438–449. Imai, K. (1982). ‘‘Allosteric Effects in Haemoglobin.’’ Cambridge: Cambridge University Press, Cambridge. Juszczak, L. J., and Friedman, J. M. (1999). UV resonance Raman spectra of ligand binding intermediates of sol-gel encapsulated hemoglobin. J. Biol. Chem. 274, 30357–30360. Kavanaugh, J. S., Chafin, D. R., Arnone, A., Mozzarelli, A., Rivetti, C., Rossi, G. L., Kwiatkowski, L. D., and Noble, R. W. (1995). Structure and oxygen affinity of crystalline desArg141 alpha human hemoglobin A in the T state. J. Mol. Biol. 248, 136–150. Kavanaugh, J. S., Weydert, J. A., Rogers, P. H., Arnone, A., Hui, H. L., Wierzba, A. M., Kwiatkowski, L. D., Paily, P., Noble, R. W., Bruno, S., and Mozzarelli, A. (2001).
Oxygen-Binding Curves for Heme Proteins
327
Site-directed mutations of human hemoglobin at residue 35beta: A residue at the intersection of the alpha1beta1, alpha1beta2, and alpha1alpha2 interfaces. Protein Sci. 10, 1847–1855. Khan, I., Shannon, C. F., Dantsker, D., Friedman, A. J., Perez-Gonzalez-de-Apodaca, J., and Friedman, J. M. (2000). Sol-gel trapping of functional intermediates of hemoglobin: Geminate and bimolecular recombination studies. Biochemistry 39, 16099–16109. Lapennas, G. N., Colacino, J. M., and Bonaventura, J. (1981). Thin-layer methods for determination of oxygen binding curves of hemoglobin solutions and blood. Methods Enzymol. 76, 449–470. Monod, J., Wyman, J., and Changeux, J.-P. (1965). On the nature of allosteric transitions: A plausible model. J. Mol. Biol. 12, 88–118. Mozzarelli, A., and Bettati, S. (2001). Functional properties of immobilized proteins. In ‘‘Advanced Functional Molecules and Polymers,’’ pp. 55–97. Overseas Publishers Association, Tokyo. Mozzarelli, A., Bettati, S., Rivetti, C., Rossi, G. L., Colotti, G., and Chiancone, E. (1996). Cooperative oxygen binding to Scapharca inaequivalvis hemoglobin in the crystal. J. Biol. Chem. 271, 3627–3632. Mozzarelli, A., Hofrichter, J., and Eaton, W. A. (1987). Delay time of hemoglobin S polymerization prevents most cells from sickling in vivo. Science 237, 500–506. Mozzarelli, A., Rivetti, C., Rossi, G. L., Eaton, W. A., and Henry, E. R. (1997). Allosteric effectors do not alter the oxygen affinity of hemoglobin crystals. Protein Sci. 6, 484–489. Mozzarelli, A., Rivetti, C., Rossi, G. L., Henry, E. R., and Eaton, W. A. (1991). Crystals of haemoglobin with the T quaternary structure bind oxygen non-cooperatively with no Bohr effect. Nature 351, 416–418. Mozzarelli, A., and Rossi, G. L. (1996). Protein function in the crystal. Annu. Rev. Biophys. Biomol. Struct. 25, 343–365. Noble, R. W., Hui, H. L., Kwiatkowski, L. D., Paily, P., DeYoung, A., Wierzba, A., Colby, J. E., Bruno, S., and Mozzarelli, A. (2001). Mutational effects at the subunit interfaces of human hemoglobin: Evidence for a unique sensitivity of the T quaternary state to changes in the hinge region of the alpha 1 beta 2 interface. Biochemistry 40, 12357–12368. Noble, R. W., Kwiatkowski, L. D., Hui, H. L., Bruno, S., Bettati, S., and Mozzarelli, A. (2002). Correlation of protein functional properties in the crystal and in solution: The case study of T-state hemoglobin. Protein Sci. 11, 1845–1849. Pearson, A. R., Mozzarelli, A., and Rossi, G. L. (2004). Microspectrophotometry for structural enzymology. Curr. Opin. Struct. Biol. 14, 656–662. Rivetti, C., Mozzarelli, A., Rossi, G. L., Henry, E. R., and Eaton, W. A. (1993a). Oxygen binding by single crystals of hemoglobin. Biochemistry 32, 2888–2906. Rivetti, C., Mozzarelli, A., Rossi, G. L., Kwiatkowski, L. D., Wierzba, A. M., and Noble, R. W. (1993b). Effect of chloride on oxygen binding to crystals of hemoglobin Rothschild (beta 37 Trp!Arg) in the T quaternary structure. Biochemistry 32, 6411–6418. Ronda, L., Bruno, S., Viappiani, C., Abbruzzetti, S., Mozzarelli, A., Lowe, K. C., and Bettati, S. (2006). Circular dichroism spectroscopy of tertiary and quaternary conformations of human hemoglobin entrapped in wet silica gels. Protein Sci. 15, 1961–1967. Ronda, R., Faggiano, S., Bettati, S., Hellmann, N., Decker, H., Weidenbach, T., and Mozzarelli, A. (2007). Hemocyanin from E. californicum encapsulated in silica gels: Oxygen binding and conformational states. Gene 398, 202–207. Shibayama, N., and Saigo, S. (1995). Fixation of the quaternary structures of human adult haemoglobin by encapsulation in transparent porous silica gels. J. Mol. Biol. 251, 203–209. Shibayama, N., and Saigo, S. (1999). Kinetics of the allosteric transition in hemoglobin within silica sol-gels. J. Am. Chem. Soc. 121, 444–445.
328
Luca Ronda et al.
Shibayama, N., and Saigo, S. (2001). Direct observation of two distinct affinity conformations in the T state human deoxyhemoglobin. FEBS Lett. 492, 50–53. Sottini, S., Abbruzzetti, S., Viappiani, C., Bettati, S., Ronda, L., and Mozzarelli, A. (2005). Evidence for two geminate rebinding states following laser photolysis of R state hemoglobin encapsulated in wet silica gels. J. Phys. Chem. B. 109, 11411–11413. Sottini, S., Viappiani, C., Ronda, L., Bettati, S., and Mozzarelli, A. (2004). CO rebinding kinetics to myoglobin- and R state hemoglobin-doped silica gels in the presence of glycerol. J. Phys. Chem B 108, 8475–8484. Van Slyke, D. D., and Neill, J. M. (1924). The determination of gases in blood and other solutions by vacuum extraction and manometric measurement. I. J. Biol. Chem. 61, 523–573. Viappiani, C., Bettati, S., Bruno, S., Ronda, L., Abbruzzetti, S., Mozzarelli, A., and Eaton, W. A. (2004). New insights into allosteric mechanisms from trapping unstable protein conformations in silica gels. Proc. Natl. Acad. Sci. USA 101, 14414–14419. Yonetani, T., Park, S., Tsuneshige, A., Imai, K., and Kanaori, K. (2002). Global allostery model of hemoglobin: Modulation of O2 affinity, cooperativity, and Bohr effect by heterotropic allosteric effectors. J. Biol. Chem. 277, 34508–34520.
C H A P T E R
S E V E N T E E N
Characterization of Ligand Migration Mechanisms inside Hemoglobins from the Analysis of Geminate Rebinding Kinetics Stefania Abbruzzetti,*,‡ Stefano Bruno,† Serena Faggiano,† Luca Ronda,† Elena Grandi,* Andrea Mozzarelli,† and Cristiano Viappiani*,‡ Contents 330 330 331 335
1. 2. 3. 4. 5.
Introduction Principles of Nanosecond Laser Flash Photolysis Basic Experimental Layouts Encapsulation of Hbs in Silica Gels Enhancement of Geminate Rebinding and Advantages of Gel Encapsulation 6. Extraction of Kinetic Information Acknowledgments References
336 337 342 342
Abstract The presence of internal hydrophobic cavities and packing defects has been demonstrated for several small globular proteins, including hemoglobins. The reduced thermodynamic stability appears to be compensated for by the capability of controlling ligand diffusion through the protein matrix to the active site, possibly by stocking more than one reactant molecule in selected sites. Photolysis of carbon monoxide complexes of hemoglobins encapsulated in silica gels leads to multiphasic geminate rebinding kinetics at room temperature, reflecting rebinding also from different temporary docking sites inside the protein matrix. A careful analysis of the ligand rebinding kinetics allows the determination of the microscopic rates for the underlying reactions, including those governing the migration to and from the docking sites. This chapter
* { {
Dipartimento di Fisica Universita` degli Studi di Parma, Parma, Italy Dipartimento di Biochimica e Biologia Molecolare, Universita` degli Studi di Parma, Parma, Italy NEST CNR-INFM, Pisa, Italy
Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37017-1
#
2008 Elsevier Inc. All rights reserved.
329
330
Stefania Abbruzzetti et al.
describes the experimental approach used to characterize the ligand rebinding kinetics for heme proteins in silica gels after nanosecond laser flash photolysis and the computational methods necessary to retrieve the kinetic parameters.
1. Introduction Ligand binding kinetics following nanosecond laser photolysis reflects the ligand migration inside the protein matrix and the reactivity toward molecules diffusing from the bulk solution. Although O2 is the physiological ligand of myoglobin (Mb) and hemoglobin (Hb), it is often difficult to avoid irreversible oxidation or side reactions while studying these very reactive species in vitro. In model studies, carbon monoxide (CO) is often preferred to oxygen as a ligand, as it does not further react chemically with the heme after binding to the Fe atom. CO generally binds even more strongly than O2 to the heme and its complexes are extremely stable in the long term. The Fe-CO bond is photolabile, and the use of short (nanosecond) light pulses allows a large population of reactive deoxy states to be generated transiently and the rebinding of CO over time to be monitored spectroscopically. The time span of this process is generally extended over several orders of magnitude in time. Immediately after the end of the laser pulse, the ligand undergoes competitive reactions in which it can either move through the protein matrix and temporarily dock to hydrophobic pockets or react with the heme to form the carboxy adduct. The presence of kinetic traps results in nonexponential relaxations, and analysis of the time course in geminate recombination can yield useful mechanistic information on the ligand migration pathways. The amplitude of this kinetic phase is extremely variable, depending on how easy it is for the photodissociated ligand to escape to the bulk solution. To overcome the problem of small geminate yields, proteins can be encapsulated in silica gels. Under these conditions, geminate recombination is often enhanced, particularly for monomeric proteins, and the sensitivity to migration processes is accordingly increased. For multimeric proteins, such as the tetrameric human hemoglobin A, substantial enhancement of the geminate phase can be observed when encapsulated in silica gels under increased viscosity, for instance, in the presence of glycerol.
2. Principles of Nanosecond Laser Flash Photolysis The general principles of the method are covered elsewhere (Bonneau et al., 1997; Cosa and Scaiano, 2004), some of which put special emphasis on the applications to proteins (Abbruzzetti et al., 2006; Chen et al., 1997; Tetreau and Lavalette, 2005).
331
Geminate Recombination and Ligand Migration
Nanosecond time-resolved absorption techniques rely on the use of a probe beam that interrogates the sample before and after excitation with a nanosecond laser (laser flash photolysis). The probe beam is most generally a broadband [from the ultraviolet (UV) to the near infrared], continuous wave (cw) light source. The near-UV spectrum is of special interest for flash photolysis studies of heme proteins, as the so-called Soret band of the heme is located in this region. Interest stems from the remarkable sensitivity of this absorption band to the ligation state of the heme Fe, as well as to changes in the surrounding structure. The a and b bands, located in the visible spectral region, are also interesting as they are sensitive to ligation of the heme Fe and reflect other features, such as ligation of water molecules or internal ligands. However, the much lower extinction coefficient requires the use of higher protein concentrations. The weak charge transfer band located in the near-infrared region (band III, 760 nm) has also proved useful in a number of investigations, as it is particularly sensitive to protein dynamics near the heme group (Iizuka et al., 1974; Nienhaus et al., 1992, 1994; Steinbach, 1991). The experimental parameter of interest is a change in absorbance (DA), reflecting changes in transmitted light from the probe beam before, I(t < t0), and after, I(t), laser excitation, according to Eq. (17.1):
DAðtÞ ¼ −log
IðtÞ Iðt < t0 Þ
ð17:1Þ
Typical instrumental outlines are summarized in Fig. 17.1, where pump/ probe geometries are presented. Because of the geometrical configuration of the gel-embedded samples, which are thin slices (0.1–1 mm) deposited on an optical quality silica surface, with an absorbance of 1 in the Soret region, it is important to work with (almost) collinear pump and probe beams to optimize the overlap between the two beams and to obtain a homogeneous excitation of the probed volume. This geometry is essential in order to keep the gels inside a bathing solution with the proper concentration of buffering salts and, other factors, such as allosteric effectors.
3. Basic Experimental Layouts The two basic configurations outlined in Fig. 17.1 allow detection of single wavelength transient absorbance traces or time-resolved spectra with laser-limited time resolution. A convenient photolysis source for photolysis of CO complexes of heme proteins is the second harmonic (532 nm) of a Q-switched Nd:YAG laser. In our setup, we use either a Surelite II-10 (Continuum) or a Handy YAG
332
Stefania Abbruzzetti et al.
A From laser Filter
Monochromator 2
Xe arc lamp Monochromator 1
Shutter
PMT
Beam dump
B From laser Filter Xe arc lamp
Monochromator/ spectrograph
Shutter
PMT, APD
Beam dump iced
Figure 17.1 Experimental layout for the collinear nanosecond laser flash photolysis setup with (A) monochromatic and (B) white light detection beams.
HYL-101 (Quanta System). Linear polarization is converted to circular by means of a quarter wave plate in order to minimize photoselection effects. Near-full photolysis is normally obtained with the setups described later, with laser pulses of 20–40 mJ, depending on the sample. This can be checked by performing photolysis experiments as a function of pump laser pulse energy and plotting the observed absorbance change at the end of the laser pulse as a function of pulse energy to verify that the absorbance change saturates to a limiting value. The need for pursuing full photolysis conditions is motivated by the necessity of working with known concentrations of reactants (deoxyheme and CO) for quantitative analyses. In the setup described in Fig. 17.1A, absorbance changes are monitored using a monochromatic cw output of a 150-W Xe arc lamp coupled to a 0.25-m monochromator (AMKO GmbH). The transient absorbance traces are measured through a second, 0.125-m monochromator (77250, LOTOriel) with a five-stage photomultiplier (PMT) (Applied Photophysics) (Bonneau et al., 1997). It is important to limit exposure of the PMT to the intense monitoring light to a few hundred milliseconds in order to avoid fatigue of the detector (Fenster et al., 1973). The output current can be typically 2–4 mA and results in 100- to 200-mV signals when fed into the 50-O load of a digital oscilloscope, without the need for intervening
Geminate Recombination and Ligand Migration
333
amplification, thus preserving maximum signal bandwidth. Timescales longer than a few hundred milliseconds can be accessed by limiting the PMT currents (e.g., using lower light intensity) and using higher loads (Bonneau et al., 1997). The voltage signal is digitized by a digital oscilloscope (LeCroy LT374, 500 MHz, 4 GS/s; LeCroy 9370, 1 GHz, 1 GS/s). A custom dichroic filter (Omega optical) is positioned between the exit slit of the monochromator and the PMT to remove residual stray light from the pump laser at 532 nm. A fast shutter ( Vincent Associates, Uniblitz VS35) is positioned between the output of the first monochromator and the sample holder. The opening of the shutter is controlled by a dedicated microprocessor ( Banderini et al., 2004), while exposure time is set by the shutter driver ( Vincent Associates, Uniblitz VMM-T1). The sample holder is accurately temperature controlled with a Peltier element, allowing a temperature stability of better than 0.1 C. A second setup is outlined in Fig. 17.1B (Abbruzzetti et al., 2005), where the cw output of a 75-W Xe arc lamp is focused onto the sample holder, then collimated, and finally entered into the entrance slits of a MS257 (LOT-Oriel) monochromator/spectrograph. A fast shutter ( Vincent Associates, Uniblitz VS35) is positioned between the output of the Xe lamp and the sample holder. It is convenient to shutter the monitoring beam to limit the amount of white light shining on the sample, thus preventing photoinduced damage. The off-axis port of the MS257 is used to monitor single wavelength kinetics. Monochromatic light is collected from the exit slit by dedicated optics after being passed through a custom dichroic filter (Omega optical), which is necessary to reject the stray light at 532 nm, and focused onto a Si APD (Hamamatsu S2382) coupled with a transimpedance amplifier (Avtech AV149, 600 MHz). The light intensity available in single monochromator applications is high enough to allow the use of photodiode detectors. Also for this setup, the voltage signal is digitized by a digital oscilloscope (LeCroy LT374, 500 MHz, 4 GS/s; LeCroy 9370, 1 GHz, 1 GS/s). Repetition rate is a relevant parameter of the experiment as the sample must fully recover between laser flashes. Thus, depending on the sample, repetition rates as low as 0.1 Hz could be necessary. Synchronization of the overall experiment (laser firing, shutter opening, and iCCD triggering) is achieved by means of dedicated hardware (Banderini et al., 2004). The finding that ligand migration signatures in absorbance change on the nanosecond timescale after photolysis of immobilized systems (Cordone et al., 2005; Dantsker et al., 2005; Sottini et al., 2005a) calls for laserlimited time resolution in any experimental setup intended for ligand binding studies. The time resolution is, in principle, set by the laser pulse width, which, for modern solid state lasers, is normally around 5 ns. However, several parameters and experimental details contribute to the overall bandwidth of the detection systems and affect access to the short timescale after laser excitation. A critical issue in determining time
334
Stefania Abbruzzetti et al.
resolution is the intense light scattering from the pump source, which is generated when the light pulse hits the gel slab. For this reason, it is necessary to work with a counterpropagating pump and probe beams (in colinear pump probe) or orient the sample toward the light source in the right angle setup. No matter how careful the alignment is, a relevant portion of the pump light is scattered toward the detector. Reduction of the scattered light is obtained with proper positioning of the optical components, and the insertion of one or more beam dumps to block reflections from the optics, including the cuvette walls. Use of a monochromator and suitable dichroic mirrors in front of the detector is essential to reduce the scattered pump light to reasonable levels. The residual stray light coming from the pump laser can be subtracted from the transmitted light signal by measuring a baseline, i.e., a signal for which the detection light is blocked. This must then be subtracted from the signal measured with the monitoring beam passing through the sample. In this way, artifacts on the short nanosecond timescale are generally removed from the resulting absorbance change signal, except for highly scattering samples. In these cases, the time resolution may be impaired. Provided laser scatter is efficiently rejected, time resolution is then determined by the overall electronic bandwidth of the detection system. Each component of the detection chain should have at least a 200-MHz bandwidth, with the limiting stage being normally the preamplifier used for conditioning the signal of the light detector. The signals we are interested in are generally extended in time over several orders of magnitude. Thus, the voltage must be sampled at a high sampling rate for the first few microseconds following the nanosecond excitation in order to reconstruct the fast kinetics with the appropriate resolution. Acquisition of the signal at several time bases is necessary to cover the whole time course, with settings on the scope being determined by the actual time span of the kinetics. An alternative option is the use of exponentially increasing delays in the data acquisition system (Banderini et al., 2004; Kriegl et al., 2003), but this requires dedicated electronics. Signals shown in Fig. 17.2 are normally acquired using a digital oscilloscope. Modern digital oscilloscopes are generally capable of acquiring data at sampling rates exceeding 1 GS/s, with an analog bandwidth of 500 MHz. The vertical resolution of the ADC is 8 bit in the single acquisition mode, requiring amplification of the signal and application of an offset. Both functions are available on a digital oscilloscope. In order to improve the signal-to-noise ratio, the kinetic traces are typically averaged over 100 laser shots. In this process, random noise is stronglypreduced (the standard ffiffiffi deviation of the noise decreases proportionally to n , with n being the number of averaged signals) and the vertical resolution is effectively increased. This is especially useful for the long time tail of the kinetics, which can account for less than 1% of the signal.
335
Geminate Recombination and Ligand Migration
0.10
0.10
0.01
0.01
0.6
0.6 g(log(t))
g(log(t))
N(t)
1.00
N(t)
1.00
0.4 0.2 0.0 10−8
10−6
10−4 Time (s)
10−2
0.4 0.2 0.0 10−8
10−6 10−4 Time (s)
10−2
Figure 17.2 CO rebinding kinetics to R state Hb gel bathed in a buffered solution containing 80% glycerol at 1 atm CO (filled circles) and 0.1 atm CO (open circles) at T ¼10 C (left) andT ¼ 40 C (right).The rebinding kinetics was measured by monitoring the absorbance changes at 436 nm. (Top) Fits obtained with the maximum entropy method (solid line,1 atm CO; dotted line, 0.1 atm CO). (Bottom) Lifetime distributions associated with the rebinding kinetics (solid line,1 atm CO; dotted line, 0.1 atm CO).
Signals acquired at different time bases are usually merged into a single signal covering the whole time range. When oversampling of the signal is performed (i.e., when the kinetics is sampled at a rate largely exceeding the rate at which the signal changes), a smoothing of data can be applied, which decreases the overall detection bandwidth but allows the noise to be reduced substantially, especially on the long timescale. Data are finally downsampled logarithmically to reduce the number of experimental points. Provided that the signal does not have a steep slope in the early nanoseconds, the absorbance change is normalized to 1 at the maximum absorbance change, and thus is plotted eventually as a fraction of deoxyhemes that have survived at time t [survival probability, N(t)].
4. Encapsulation of Hbs in Silica Gels Hemoglobin gels are prepared using a protocol that is a slight modification of the method proposed by Shibayama and Saigo (1995; Bruno et al., 2001). A solution containing 10 mM HEPES, 1 mM EDTA at pH 6 is added to an equal volume of tetramethyl orthosilicate and vortexed for
336
Stefania Abbruzzetti et al.
2 min at 4 C. The mixture is then deoxygenated by bubbling helium for 90 min. An equal volume of a solution containing 1% (w/v) Hb, 10 mM HEPES, 1 mM EDTA, 30 mM sodium dithionite at pH 6 is added. Deoxy and carboxy species can be selectively entrapped in silica gel. In the former case, the Hb solution is equilibrated with helium or nitrogen, whereas in the latter case it is equilibrated with CO at 1 atm. Gelation normally occurs in 10–20 min at room temperature. When the gel is formed, a solution containing 100 mM HEPES, 1 mM EDTA, 30 mM sodium dithionite at pH 7, saturated with either helium or CO, is layered on it. The sample thickness can be adjusted between 0.1 and 1 mm. The gels are transparent in the UV-visible range and their optical quality allows spectroscopic measurements. In order to modulate viscosity, soaking solutions containing glycerol at concentrations spanning from 10 to 100% (by weight) can be employed. In this case, it is important to store the samples after they are bathed in the water/glycerol mixtures at 5 C for at least 3 days before performing the kinetic experiments to allow for diffusion of glycerol within the gel pores and to avoid gradients. We have found that after this period, kinetic traces are normally independent of aging. Samples are kept at all stages in gas-tight vials to prevent O2 leakage into the sample compartment. After an appropriate aging period, gels are washed anaerobically in a solution of the desired composition and pH and loaded into a gas-tight cuvette provided with a reservoir. The final soaking solution had been equilibrated previously with CO at the desired partial pressure, usually ranging from 0.1 to 1 atm. A gas mixture of the same composition in CO is then fluxed briefly in the cuvette reservoir in order to maintain a constant partial pressure throughout the experiment. For deoxyHb gels, in order to monitor all possible conformational rearrangements associated with the change in ligation state, particular care is taken to carry out the flash photolysis experiment within a short time from exposure of the sample to CO. For the same reason, all samples are kept on ice until the measurement is carried out, as low temperatures were shown to dramatically slow down conformational transitions in silica gel.
5. Enhancement of Geminate Rebinding and Advantages of Gel Encapsulation Ligand migration affects geminate recombination in the transient absorbance following laser flash photolysis. A number of investigations have shown that this kinetic phase can be exploited to gain useful information about the mechanism of ligand migration (Nienhaus and Nienhaus, 2002; Nienhaus et al., 2003a,b, 2005; Schmidt et al., 2005; Tetreau et al.,
Geminate Recombination and Ligand Migration
337
2004; Vallone et al., 2004). Several studies have taken advantage of the enhancement of the geminate phase observed for proteins immobilized in sugar glasses or in silica gels in the presence of glycerol (Abbruzzetti et al., 2006; Dantsker et al., 2002; Khan et al., 2000; Milani et al., 2004; Samuni et al., 2002, 2003, 2004, 2006; Sottini et al., 2004, 2005a,b,c). The reduced mobility of the polypeptide chain reduces the probability of escaping to the surrounding solvent for the photodissociated ligand, while favoring competitive diffusion to internal sites (Dantsker et al., 2002; Samuni et al., 2003). A tremendous advantage of the gel approach is that ligand migration and structural relaxation following photodissociation can be detected as typical signatures in the ligand rebinding kinetics even at near room temperature, appearing as multiple phases in the geminate recombination. When ligand recombination is investigated at cryogenic temperatures, the observed kinetics indicate the existence of a frozen distribution of functionally distinct conformational substates (Austin et al., 1973, 1975). As the temperature is raised, thermal averaging of the conformational substates cancels the kinetic hole burning phenomena (Huang et al., 1997) observed at lower temperatures (Agmon, 1988; Campbell et al., 1987) and new kinetic phases are observed (Agmon et al., 1994; Austin et al., 1973, 1975; Doster et al., 1993; Kleinert et al., 1998; Steinbach, 1991), which lead to the apparent inverse temperature effect of the observed kinetics.
6. Extraction of Kinetic Information Kinetic information contained in the ligand rebinding curves can be retrieved in several ways, all of which require a proper kinetic model to obtain microscopic rates. Escape of the photodissociated ligand from the protein matrix to the surrounding solvent is generally observed when the protein is encapsulated in silica gels, although this occurs to a greatly reduced extent, especially when the gels are soaked in pure glycerol. The first step is then the identification of geminate rebinding versus solution rebinding, a task that can be accomplished with the study of the kinetics as a function of ligand concentration. CO partial pressures should be changed over at least one order of magnitude (typically 0.1–1 atm). This allows identification of the portion of the rebinding kinetics involving ligands that have not escaped, for which the kinetics is unaffected by CO partial pressure. Figure 17.2 shows the CO rebinding kinetics measured after photolysis of an R state human Hb A gel soaked in a buffered solution containing 80% glycerol at 10 C. The transient absorbance signal at 436 nm is displayed as a fraction of deoxy species as a function of time. The absorbance change was monitored at this wavelength because previous studies showed that this signal represents pure CO rebinding kinetics (Jones et al., 1992). This may
338
Stefania Abbruzzetti et al.
not be true at other wavelengths, as the absorbance changes may partly reflect the protein structural relaxation. Under these experimental conditions, CO rebinding is observed to the pure R quaternary state, as switching to the T quaternary state is completely inhibited by the gel matrix. The first qualitative analysis recommended is the estimate of the number of kinetic steps in geminate rebinding, a task that can be accomplished by a fit with a multiple exponential function (stretched exponentials may be required to account for overlapping structural relaxations and/or kinetic heterogeneity in the sample) (Abbruzzetti et al., 2001a,b; Sottini et al., 2004; Viappiani et al., 2004). The rebinding kinetics shown in Fig. 17.2 can be fitted with a triple stretched exponential decay (Sottini et al., 2004), with the two faster decays characterized by the same rates, independent of CO concentration. The slowest process is characterized by a CO concentrationdependent lifetime. This allows the attribution of the two faster phases to the geminate recombination and the slowest to the second order rebinding from the solution. However, this qualitative description is not satisfactory, as it does not provide proper microscopic rates nor allow realistic lifetime distributions to be obtained. An estimate of the effective lifetime distributions associated with the rebinding kinetics takes advantage of a model-independent analysis based on a maximum entropy method (Steinbach, 2002; Steinbach et al., 2002). Convenient software, capable of determining model-independent lifetime distributions associated with the measured rebinding kinetics and developed by P. J. Steinbach, is available on the internet at http://cmm.cit.nih.gov/ memexp/. The bottom panels of Fig. 17.2 show lifetime distributions associated with the rebinding kinetics in the top panels. The lifetime distributions confirm the qualitative picture emerging by the simple visual inspection of the rebinding kinetics. The lifetime distribution corresponding to the fast phase is independent of CO concentration, showing the unimolecular nature of the event (geminate recombination). On longer timescales, the bimolecular nature of the process is demonstrated by the CO concentration dependence on the position of the associated band, which moves to shorter values as the CO concentration is increased. The lifetime distribution associated with the geminate recombination at 40 C is a broad band peaked at 100 ns, with a shoulder at 500 ns. When the temperature is lowered, the band splits in two, with maxima at 60 and 600 ns at 10 C (see Fig. 17.2, bottom left). The appearance of several distinct peaks in geminate recombination is an indication of rebinding from different sites located inside the protein matrix. The number of these sites can be reasonably estimated from the lifetime distribution. The peak observed at the shortest times is due to rebinding from primary docking sites for the photodissociated CO, which are located in the distal heme pocket.
339
Geminate Recombination and Ligand Migration
A(HbCO)
hv kCA
C(Hb : CO)
kCS
S(Hb + CO)
kSC kBC
kCB
B(Hb :: CO)
Scheme 1 Simplified four state kinetic model.
Additional peaks are indicative of additional docking sites. In the case reported in Fig. 17.2, a single secondary site appears to contribute to the rebinding kinetics (Sottini et al., 2005a,b). Once the number of steps has been defined, a suitable kinetic model should be sketched to extract microscopic rates from the apparent ones. This can be accomplished by solving the differential equations associated with the kinetic scheme having the proper number of intermediates, suggested by the previous analysis. The analytical solution must then be used to write the apparent rates in terms of the microscopic rates. This may become impractical for multistep reactions, in which the analytical solution is not easily rearranged to obtain a useful relationship between microscopic and apparent rates. In the case of the kinetics depicted in Fig. 17.2, the minimal model requires a four state equilibrium, sketched in Scheme 1. In the previous example, the differential equations corresponding to Scheme 1 can be solved analytically and microscopic rates can be obtained, as reported later (Sottini et al., 2005b). The overall rebinding rate kCG for the geminate phase is
kCG ¼ kCB þ kCA þ kCS ¼
Ag1 k2g1 þ Ag2 k2g2 kCA
ð17:2Þ
where
kCA ¼ Ag1 kg1 þ Ag2 kg2
ð17:3Þ
kBC ¼ kg1 þ kg2 kCG
ð17:4Þ
kCB ¼ kCG kCA kCS
ð17:5Þ
kCS ¼
Ab kCA 1 Ab
ð17:6Þ
Ag1, Ag2, and Ab represent the normalized areas corresponding to the two geminate peaks, and Ab is the normalized area corresponding to the bimolecular rebinding. A convenient way to estimate these values from data in
340
Stefania Abbruzzetti et al.
the bottom panels in Fig. 17.2 is by fitting the lifetime distributions to a sum of Gaussian shapes. kg1 and kg2 are the average rates corresponding to the two geminate peaks, whereas kb is the average rate for the bimolecular distribution. It has been assumed that the contribution to Ag1 and Ag2 resulting from CO molecules that come into the heme pocket from the solution can be neglected. The rebinding rate from solution can be estimated in the steady-state approximation (Henry et al., 1983) as
kSC ½CO ¼
kb kBC ðkCG − kCB Þ − k2b kCG − ðkCG − kCS Þkb þ kBC kCA
ð17:7Þ
In the steady-state approximation the concentration of protein in state C is assumed to be independent of time. At [CO] ¼ 1 mM, the relative values of the rate constants justify the steady-state approximation and show that during ligand rebinding there is a pre-equilibrium with respect to the CO entering and exiting a noncovalent binding site in the protein (Henry et al., 1983). In the limit of vanishing kBC and kCB, Eqs. (17.2) through (17.7) reduce to those already reported for the three state model (Sottini et al., 2004). A more direct approach to the definition of microscopic rates involves fitting of the numerical solution of the set of coupled differential equations associated with the kinetic scheme defined by the previous qualitative analysis. To this purpose, the overall concentration of deoxy species as a function of time, f(t), is fitted to the progress curve, N(t), determined experimentally. Parameters to be optimized normally include the rate constants and the initial concentrations of reactants. The latter optimization is necessary for compensating the limited precision by which the CO concentration is set, the lack of real full photolysis, and the difficulty in evaluating the concentration of Hb in the gel. However, it is important to reach near-full photolysis conditions to make sure that the initial concentrations of the reactants are known as precisely as possible. A major limitation of this approach is that it neglects any kinetic heterogeneity that may arise from proteins being trapped in slightly different environments or from structural relaxation following laser photolysis. Figure 17.3 reports a fit to the experimental curves at 10 C shown in Fig. 17.2. The overall rebinding curve is reproduced fairly well, and only some minor mismatch is observed between the experimental curve and the model. Several checks need to be performed to make sure that the set of rates determined with this numerical analysis is realistic. In the absence of conformational switching associated with the ligation state, the set of rates must be independent of the initial reactant concentrations. This is normally tested by varying the CO concentration. The rates must follow a regular trend, normally an Arrhenius-like behavior, when determined as a function of temperature.
341
[deoxy Hb] (mM)
[deoxy Hb] (mM)
Geminate Recombination and Ligand Migration
10−5
10−6
10−7
10−5
10−6
10−7
10−8 10−7 10−6 10−5 10−4 10−3 10−2 Time (s)
10−8 10−7 10−6 10−5 10−4 10−3 10−2 Time (s)
Figure 17.3 Fitting of CO rebinding kinetics to R state Hb gel bathed in a buffered solution containing 80% glycerol at 0.1 atm CO and T ¼ 10 (left) and T ¼ 40 (right). Circles are experimental data; the solid line overlapping experimental data is the fitted curve. Other curves represent the time courses of C (solid), B (dashed), and S (dotted).
The effect of viscosity is also relevant, which can be tested by increasing the glycerol concentration from 0 to 100%. This allows bulk solvent viscosity to be modulated from 1 to 1000 cP when experiments are conducted at near room temperature. Rates in general do not strictly follow the Kramers relation but are rather described by a generalized expression in which an internal viscosity is taken into account. When protein conformational changes are expected to contribute to the observed kinetics, an additional source of friction, s, has been envisioned as a consequence of hindered intrachain motion within the protein. This friction adds to the friction from solvent molecules, , at the protein surface in resisting protein relaxation after a perturbation as, e.g., heme ligand photodissociation (Ansari et al., 1992). Under these conditions, the Kramers equation becomes
kðT ; Þ ¼
Ca − Ea e RT sþ
ð17:8Þ
where Ea is an activation energy and Ca is a proportionality factor. The viscosity of the bulk solution is temperature dependent, and it is important to note that Eq. (17.8) requires fitting experimental data with a function of two variables (T and ). In order to obtain a good range of values of , it is useful to investigate the temperature dependence of the kinetics over a range of glycerol concentrations (from pure water to pure glycerol). A very effective way of controlling the regular behavior of rate constants is to perform global analysis by fitting simultaneously several rebinding curves taken under different experimental conditions and imposing constraints on the parameters to be optimized. The simplest approach is to fit data taken at different CO concentrations, with the other experimental conditions being unchanged (same temperature, same bathing solution, etc.).
342
Stefania Abbruzzetti et al.
The rate constants are kept as shared (global) parameters. In single curve fitting the function to be optimized (minimized) is normally the sum of the square of the residuals:
2 n Σ ¼ ∑ N ðti Þ − f ðti Þ i¼1
ð17:9Þ
where n is the number of data points for the experimental curve and f is the function used to describe the observed kinetics. Although desirable, a w2 test is normally unpractical, as the variance for each data point is not known a priori and should be determined experimentally, a procedure that substantially complicates data acquisition and is generally not performed. In multiple curve optimizations, a convenient test for convergence is to minimize the sum of the test functions for each experimental curve:
2 m m n Σ ¼ ∑ Sj ¼ ∑ ∑ Nj ðti Þ − fj ðti Þ j¼1
j¼1 i¼1
ð17:10Þ
Shared and unshared parameters can be optimized using standard minimization algorithms such as the Levenberg–Marquardt or the Simplex, normally available in several commercial analysis softwares. A detailed description of the optimization algorithms and methods is beyond the scope of this chapter and can be found in several publications (Bevington, 1969; Brand and Johnson, 1992).
ACKNOWLEDGMENTS The authors acknowledge MIUR (PRIN2004) for financial support. P. J. Steinbach is kindly acknowledged for the use of MemExp.
REFERENCES Abbruzzetti, S., Bruno, S., Faggiano, S., Grandi, E., Mozzarelli, A., and Viappiani, C. (2006). Monitoring heme proteins at work with nanosecond laser flash photolysis. Photochem. Photobiol. Sci. 5, 1109–1120. Abbruzzetti, S., Sottini, S., Viappiani, C., and Corrie, J. E. T. (2005). Kinetics of proton release after flash photolysis of 1-(2-nitrophenyl)ethyl sulfate (caged sulfate) in aqueous solutions. J. Am. Chem. Soc. 127, 9865–9874. Abbruzzetti, S., Viappiani, C., Bruno, S., Bettati, S., Bonaccio, M., and Mozzarelli, A. (2001a). Functional characterization of heme proteins encapsulated in wet nanoporous silica gels. J. Nanosci. Nanotech. 1, 407–415.
Geminate Recombination and Ligand Migration
343
Abbruzzetti, S., Viappiani, C., Bruno, S., and Mozzarelli, A. (2001b). Enhanced geminate ligand rebinding upon photo-dissociation of silica gel-embedded myoglobin-CO. Chem. Phys. Lett. 346, 430–436. Agmon, N. (1988). Reactive line-shape narrowing in low-temperature inhomogeneous geminate recombination of CO to myoglobin. Biochemistry 27, 3507–3511. Agmon, N., Doster, W., and Post, F. (1994). The transition from inhomogeneous to homogeneous kinetics in CO binding to myoglobin. Biophys. J. 66, 1612–1622. Ansari, A., Jones, C. M., Henry, E. R., Hofrichter, J., and Eaton, W. (1992). The role of solvent viscosity in the dynamics of protein conformational changes. Science 256, 1796–1798. Austin, R. H., Beeson, K., Eisenstein, L., Frauenfelder, H., Gunsalus, I. C., and Marshall, V. P. (1973). Dynamics of carbon monoxide binding by heme proteins. Science 181, 541–543. Austin, R. H., Beeson, K. W., Eisenstein, L., Frauenfelder, H., and Gunsalus, I. C. (1975). Dynamics of ligand binding to myoglobin. Biochemistry 14, 5355–5373. Banderini, A., Sottini, S., and Viappiani, C. (2004). Method for detecting extended realtime kinetics in nanosecond laser flash photolysis experiments. Rev. Sci. Instrum. 75, 2257–2261. Bevington, P. R. (1969). In ‘‘Data Reduction and Error Analysis for the Physical Sciences.’’ New York: McGraw-Hill, New York. Bonneau, R., Wirz, J., and Zuberbuehler, A. D. (1997). Methods for the analysis of transient absorbance data. Pure Appl. Chem. 69, 979–992. Brand, L., and Johnson, M. L. (1992). Numerical computer methods. Methods Enzymol. 210. Bruno, S., Bonaccio, M., Bettati, S., Rivetti, C., Viappiani, C., Abbruzzetti, S., and Mozzarelli, A. (2001). High and low oxygen affinity conformations of T state hemoglobin. Protein Sci. 10, 2401–2407. Campbell, B. F., Chance, M. R., and Friedman, J. M. (1987). Linkage of functional and structural heterogeneity in proteins: Dynamic hole burning in carboxymyoglobin. Science 238, 373–376. Chen, E., Goldberg, R. A., and Kliger, D. S. (1997). Nanosecond time-resolved spectroscopy of biomolecular processes. Annu. Rev. Biophys. Biomol. Struct. 26, 327–355. Cordone, L., Cottone, G., Giuffrida, S., Palazzo, G., Venturoli, G., and Viappiani, C. (2005). Internal dynamics and protein–matrix coupling in trehalose-coated proteins. Biochim. Biophys. Acta Proteins Proteom. 1749, 252–281. Cosa, G., and Scaiano, J. C. (2004). Laser techniques in the study of drug photochemistry. Photochem. Photobiol. 80, 159–174. Dantsker, D., Samuni, U., Friedman, A. J., Yang, M., Ray, A., and Friedman, J. M. (2002). Geminate rebinding in trehalose-glass embedded myoglobins reveals residue-specific control of intramolecular trajectories. J. Mol. Biol. 315, 239–251. Dantsker, D., Samuni, U., Friedman, J. M., and Agmon, N. (2005). A hierarchy of functionally important relaxations within myoglobin based on solvent effects, mutations and kinetic model. Biochim. Biophys. Acta 1749, 234–251. Doster, W., Kleinert, T., Post, F., and Settles, M. (1993). Effect of solvent on protein internal dynamics: The kinetics of ligand binding to myoglobin. In ‘‘Protein-Solvent Interactions’’ (R. B. Gregory, ed.), p. 375. Dekker, New York. Fenster, A., LeBlanc, J. C., Taylor, W. B., and Jones, H. E. (1973). Linearity and fatigue in photomultipliers. Rev. Sci. Instrum. 44, 689–694. Henry, E. R., Sommer, J. H., Hofrichter, J., and Eaton, W. A. (1983). Geminate recombination of carbon monoxide to myoglobin. J. Mol. Biol. 166, 443–451. Huang, J., Ridsdale, A., Wang, J., and Friedman, J. M. (1997). Kinetic hole burning, hole filling, and conformational relaxation in heme proteins: Direct evidence for the
344
Stefania Abbruzzetti et al.
functional significance of a hierarchy of dynamical processes. Biochemistry 36, 14353–14365. Iizuka, T., Yamamoto, H., Kotani, M., and Yonetani, T. (1974). Low temperature photodissociation of hemoproteins: Carbon monoxide complex of myoglobin and hemoglobin. Biochim. Biophys. Acta 371, 126–139. Jones, C. M., Ansari, A., Henry, E. R., Christoph, G. W., Hofrichter, J., and Eaton, W. A. (1992). Speed of intersubunit communication in proteins. Biochemistry 31, 6692–6702. Khan, I., Shannon, C. F., Dantsker, D., Friedman, A. J., Perez-Gonzales-de-Apodaca, J., and Friedman, J. M. (2000). Sol-gel trapping of functional intermediates of hemoglobin: Geminate and bimolecular recombination studies. Biochemistry 39, 16099–16109. Kleinert, T., Doster, W., Leyser, H., Petry, W., Schwarz, V., and Settles, M. (1998). Solvent composition and viscosity effects on the kinetics of CO binding to horse myoglobin. Biochemistry 37, 717–733. Kriegl, J. M., Forster, F. K., and Nienhaus, G. U. (2003). Charge recombination and protein dynamics in bacterial photosynthetic reaction centers entrapped in a sol-gel matrix. Biophys. J. 85, 1851–1870. Milani, M., Pesce, A., Ouellet, Y., Dewilde, S., Friedman, J., Ascenzi, P., Guertin, M., and Bolognesi, M. (2004). Heme-ligand tunneling in group I truncated hemoglobins. J. Biol. Chem. 279, 21520–21525. Nienhaus, G. U., Mourant, J. R., and Frauenfelder, H. (1992). Spectroscopic evidence for conformational relaxation in myoglobin. Proc. Natl. Acad. Sci. USA 89, 2902–2906. Nienhaus, G. U., Mowant, J. R., Chu, K., and Frauenfelder, H. (1994). Ligand binding to heme proteins: The effect of light on ligand binding in myoglobin. Biochemistry 33, 13413–13430. Nienhaus, G. U., and Nienhaus, K. (2002). Infrared study of carbon monoxide migration among internal cavities of myoglobin mutant L29W. J. Biol. Phys. 28, 163–172. Nienhaus, K., Deng, P., Kriegl, J. M., and Nienhaus, G. U. (2003a). Structural dynamics of myoglobin: Effect of internal cavities on ligand migration and binding. Biochemistry 42, 9647–9658. Nienhaus, K., Deng, P., Kriegl, J. M., and Nienhaus, G. U. (2003b). Structural dynamics of myoglobin: Spectroscopic and structural characterization of ligand docking sites in myoglobin mutant L29W. Biochemistry 42, 9633–9646. Nienhaus, K., Ostermann, A., Nienhaus, G. U., Parak, F. G., and Schmidt, M. (2005). Ligand migration and protein fluctuations in myoglobin mutant L29W. Biochemistry 44, 5095–5105. Samuni, U., Dantsker, D., Juszczak, L. J., Bettati, S., Ronda, L., Mozzarelli, A., and Friedman, J. M. (2004). Spectroscopic and functional characterization of T state hemoglobin conformations encapsulated in silica gels. Biochemistry 43, 13674–13682. Samuni, U., Dantsker, D., Khan, I., Friedman, A. J., Peterson, E., and Friedman, J. M. (2002). Spectroscopically and kinetically distinct conformational populations of sol-gelencapsulated carbonmonoxy myoglobin. J. Biol. Chem. 277, 25783–25790. Samuni, U., Dantsker, D., Ray, A., Wittenberg, J. B., Wittenberg, B. A., Dewilde, S., Moens, L., Ouellet, Y., Guertin, M., and Friedman, J. M. (2003). Kinetic modulation in carbonmonoxy derivatives of truncated hemoglobins: The role of distal heme pocket residues and extended apolar tunnel. J. Biol. Chem. 278, 27241–27250. Samuni, U., Roche, C. J., Dantsker, D., Juszczak, L. J., and Friedman, J. M. (2006). Modulation of reactivity and conformation within the T-quaternary state of human hemoglobin: The combined use of mutagenesis and sol-gel encapsulation. Biochemistry 45, 2820–2835. Schmidt, M., Nienhaus, K., Pahl, R., Krasselt, A., Anderson, S., Parak, F., Nienhaus, G. U., and Srajer, V. (2005). Ligand migration pathway and protein dynamics in myoglobin:
Geminate Recombination and Ligand Migration
345
A time-resolved crystallographic study on L29W MbCO. Proc. Natl. Acad. Sci. USA 102, 11704–11709. Shibayama, N., and Saigo, S. (1995). Fixation of the quaternary structures of human adult hemoglobin by encapsulation in transparent porous silica gels. J. Mol. Biol. 251, 203–209. Sottini, S., Abbruzzetti, S., Spyrakis, F., Bettati, S., Ronda, L., Mozzarelli, A., and Viappiani, C. (2005a). Geminate rebinding in R state hemoglobin: Kinetic and computational evidence for multiple hydrophobic pockets. J. Am. Chem. Soc. 127, 17427–17432. Sottini, S., Abbruzzetti, S., Viappiani, C., Bettati, S., Ronda, L., and Mozzarelli, A. (2005b). Evidence for two geminate rebinding states following laser photolysis of R state hemoglobin encapsulated in wet silica gels. J. Phys. Chem. B 109, 11411–11413. Sottini, S., Abbruzzetti, S., Viappiani, C., Ronda, L., and Mozzarelli, A. (2005c). Determination of microscopic rate constants for CO binding and migration in myoglobin encapsulated in silica gels. J. Phys. Chem. B 109, 19523–19528. Sottini, S., Viappiani, C., Ronda, L., Bettati, S., and Mozzarelli, A. (2004). CO rebinding kinetics to myoglobin- and R state hemoglobin-doped silica gels in the presence of glycerol. J. Phys. Chem. B 108, 8475–8484. Steinbach, P. J. (1991). Ligand binding to heme proteins: Connection between dynamics and function. Biochemistry 30, 3988–4001. Steinbach, P. J. (2002). Inferring lifetime distributions from kinetics by maximizing entropy using a bootstrapped model. J. Chem. Inf. Comput. Sci. 42, 1476–1478. Steinbach, P. J., Ionescu, R., and Matthews, C. R. (2002). Analysis of kinetics using a hybrid maximum-entropy/nonlinear-least-squares method: Application to protein folding. Biophys. J. 82, 2244–2255. Tetreau, C., Blouquit, Y., Novikov, E., Quiniou, E., and Lavalette, D. (2004). Competition with xenon elicits ligand migration and escape pathways in myoglobin. Biophys. J. 86, 435–447. Tetreau, C., and Lavalette, D. (2005). Dominant features of protein reaction dynamics: Conformational relaxation and ligand migration. Biochim. Biophys. Acta 1724, 411–424. Vallone, B., Nienhaus, K., Matthes, K., Brunori, M., and Nienhaus, G. U. (2004). The structure of murine neuroglobin: Novel pathways for ligand migration and binding. Proteins 56, 85. Viappiani, C., Bettati, S., Bruno, S., Ronda, L., Abbruzzetti, S., Mozzarelli, A., and Eaton, A. W. (2004). New insights into allosteric mechanisms from trapping unstable protein conformations. Proc. Natl. Acad. Sci. USA 101, 14414–14419.
This page intentionally left blank
C H A P T E R
E I G H T E E N
Ligand Dynamics in Heme Proteins Observed by Fourier Transform Infrared Spectroscopy at Cryogenic Temperatures Karin Nienhaus* and G. Ulrich Nienhaus*,† Contents 1. Introduction 2. Materials 2.1. Carbon monoxide and nitric oxide handling 2.2. Sample preparation 2.3. Fourier transform infrared cryospectroscopy equipment 3. Fourier Transform Infrared Cryospectroscopy 3.1. Environmental effects on ligand IR bands 3.2. Photolysis difference spectroscopy 3.3. FTIR-temperature derivative spectroscopy (FTIR-TDS) 4. Low-Temperature FTIR Spectroscopy on NO-Ligated Heme Proteins 4.1. Nitric oxide-ligated myoglobin 4.2. Nitric oxide-ligated nitrophorin 4 (NP4) 5. Concluding Remarks Acknowledgments References
348 349 349 350 351 353 353 355 359 365 365 370 373 374 374
Abstract Fourier transform infrared spectroscopy is a powerful tool for the investigation of protein–ligand interactions in heme proteins. From the variety of ligands that bind to the heme iron, nitric oxide and carbon monoxide are particularly attractive, as their bond-stretching vibrations give rise to strong mid-infrared absorption bands that can be measured with exquisite sensitivity and precision using photolysis difference spectroscopy at cryogenic temperatures. These stretching bands are fine-tuned by electrostatic interactions with the environment and, therefore, the ligands can be utilized as local probes of structure and dynamics. Bound to the heme iron, the ligand-stretching bands are susceptible to changes * {
Institute of Biophysics, University of Ulm, Ulm, Germany Department of Physics, University of Illinois at Urbana–Champaign, Urbana, Illinois
Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37018-3
#
2008 Elsevier Inc. All rights reserved.
347
348
Karin Nienhaus and G. Ulrich Nienhaus
in the iron-ligand bond and the electric field at the active site. Upon photolysis, the vibrational bands reveal changes due to ligand relocation to docking sites within the protein, rotational motions of the ligand in these sites, and protein conformational changes. Photolysis difference spectra taken over a wide temperature range (3-300 K) using specific temperature protocols for sample photodissociation thus can provide detailed insights into both protein and ligand dynamics. Moreover, temperature-derivative spectroscopy has proven to be a particularly powerful technique to study protein–ligand interactions. This technique has been extensively applied to studies of carbon monoxide binding to heme proteins, whereas measurements with nitric oxide are still scarce. This chapter describes infrared cryospectroscopy techniques and presents examples that demonstrate their applicability to nitric oxide binding to heme proteins.
1. Introduction The heme prosthetic group confers diverse functionality to proteins, including transfer of electrons, enzymatic activity and storage, transport, and sensing of small diatomic ligands such as dioxygen (O2), carbon monoxide (CO), and nitric oxide (nitrogen monoxide, NO) (Anderson and Chapman, 2005; Chapman et al., 1997). For ligand-binding heme proteins, efficient mechanisms have evolved through which the protein is able to discriminate between these ligands to perform its specific physiological role (GillesGonzalez and Gonzalez, 2005). Recent years have witnessed an enormous surge in studies of the interaction of NO with heme proteins. NO is a highly reactive molecule due to its free-radical nature, yet it participates in a multitude of physiological functions (Cooper, 1999; Lipton et al., 1994; Stamler and Meissner, 2001; Stamler et al., 1992). At nanomolar concentration, NO acts as a messenger involved in the relaxation of blood vessels. Its lifetime of 100 ms in tissue is governed by its reaction with hemoglobin (Hb) or myoglobin (Mb). NO is also involved in the prevention of platelet aggregation and neurotransmission (Pacher et al., 2007; Packer, 1996). In the immune response, activated macrophages produce NO and superoxide at concentrations in the micromolar range, which react to form peroxynitrite (Huie and Padmaja, 1993). This potent oxidant kills intruders but can also create damage to healthy tissues. Indeed, peroxynitrite is involved in the pathophysiology of diseases as diverse as acute stroke, Alzheimer’s disease, or chronic ischemic heart disease (Halliwell, 2006). Infrared (IR) spectroscopy has long been recognized as a powerful method for the study of ligand binding and functional processes in heme proteins (Alben et al., 1982; Dong and Caughey, 1994). Infrared spectra consist of a vast number of overlapping bands, the most prominent of which are the amide I (1650 cm1) and amide II (1550 cm1) bands of the peptide backbone. IR-active bond-stretching vibrations have been investigated
NO-Ligated Heme Proteins
349
for many heme-bound ligands (Dong and Caughey, 1994). Among these, CO and NO are particularly attractive because they have strong absorption bands, and furthermore, their IR spectra can be studied selectively by using photolysis-induced difference spectroscopy, which involves measurement of IR spectra before and after photolysis. The difference of the two spectra contains only absorption features due to photodissociation of the ligand from the heme iron. This technique yields IR spectra of heme-bound and photodissociated ligands with exquisite sensitivity and precision. The stretching bands are fine-tuned by interactions of the ligand with its environment and are a rich source of information on active-site conformations, effects of mutation, and ligand-docking sites. Moreover, the response of the protein structure to effector molecules, pH, temperature, and pressure can be examined via the ligand IR bands. At ambient temperature, however, the difference spectrum decays rapidly after photolyzing illumination due to ligands that rebind to the heme iron. Therefore, time-resolved techniques are required, including rapid-scan or step-scan Fourier transform infrared (FTIR) spectroscopy (Gerwert, 1993; Ma¨ntele, 1993), or transient or pump-probe laser spectroscopy (Bredenbeck et al., 2007; Johnson et al., 1996; Lim et al., 1997). Alternatively, one can employ FTIR spectroscopy at cryogenic temperatures, which takes advantage of the fact that rate processes in proteins are thermally activated. Ligand rebinding in heme proteins becomes sufficiently slow near 0 K that IR spectra of photodissociated species can be measured on a steady-state spectrometer. One could argue that processes observed at low temperature may not be relevant for the functioning of a protein at physiological temperature. However, evidence has accumulated over the years that this objection is unfounded. The use of cryogenic temperatures has certain advantages: Reaction intermediates can be enhanced and studied in detail by using elaborate illumination protocols at varying temperatures (Chu et al., 1995; Nienhaus et al., 1992, 1994), and the temperature dependence of the IR bands themselves provides additional information on protein dynamics as well as ligand dynamics and binding (Alben et al., 1982; Ansari et al., 1987; Kriegl et al., 2003; Nienhaus et al., 1998). As yet, these techniques have been applied extensively to CO-ligated heme proteins, but as shown here, they can also be brought to good use in the study of NO binding to heme proteins.
2. Materials 2.1. Carbon monoxide and nitric oxide handling High-purity CO and NO gas cylinders can be purchased from commercial suppliers. For reasons of their toxicity, preparations involving these gases should be performed exclusively under a fume hood. To avoid O2
350
Karin Nienhaus and G. Ulrich Nienhaus
contamination in the preparation of CO-ligated samples and higher nitrogen oxides when working with NO, gas supply lines are evacuated thoroughly using an oil-sealed mechanical pump, and the air above liquid samples is exchanged with nitrogen or argon. Small quantities of NO gas can also be produced conveniently in the laboratory by exposing copper filings to nitric acid, Cu þ HNO3 $ CuO þ NO. To remove traces of NO2, NO gas is bubbled through 1 M KOH solution before exposure to deaerated protein solutions. Anaerobic NO gas handling for heme ligation has been described elsewhere (Lim et al., 2005). Alternatively, NO can be generated by using ‘‘caged compounds’’ such as potassium pentachloronitrosylruthenate [K2Ru(NO)Cl5]. These stable and water-soluble compounds can be made to release NO by photolysis with near-UV light around 350 nm, which has the advantage that the NO concentration can be changed in situ via the extent of UV light exposure. Details can be found in a special volume of this series (Marriott, 1998).
2.2. Sample preparation Water is an extremely strong IR absorber that can obscure IR absorption by dissolved proteins. Therefore, one aims to maximize the ratio of protein to water molecules by using highly concentrated protein solutions, which should, however, still be tolerable from the viewpoint of protein stability and proper functioning. Typical protein concentrations are in the range of 10-300 mg/ml, and sample thicknesses are between 1 and 100 mm, depending on the region of the IR spectrum that is of interest. Absorption by H2O is particularly strong from 3100 to 3700 cm1 (OH stretching) and 1600 to 1700 cm1 (H2O bending). Because these windows are shifted to lower frequencies in D2O due to the isotope effect, replacing H2O by D2O can be beneficial when studying NO-ligated ferrous heme proteins, which display the NO-stretching vibration in the H2O-bending region. For free NO and CO, NO-ligated ferric heme and CO-ligated ferrous heme, the ligandstretching bands fortunately appear in a region of low water absorption. For low-temperature IR spectroscopy, protein samples are prepared in a cryoprotective solvent that avoids formation of ice. Small ice crystals cause strong scattering of the IR light, which leads to a poor quality of IR spectra. Moreover, ice formation may also denature the protein molecules. In contrast, cryoprotecting solvents such as mixtures of alcohols or sugars and water are viscous liquids that form optically transparent glasses at cryogenic temperatures (Carpenter and Crowe, 1988; Douzou, 1977; Paiva and Panek, 1996). Our default cryosolvent for infrared spectroscopy is glycerol/1 M potassium phosphate at a ratio of 3 to 1 (by volume). For the preparation of highly concentrated IR samples, it is advantageous to start with lyophilized protein material. Freeze-drying is tolerated by many small proteins, but one needs to carefully check that the protein
NO-Ligated Heme Proteins
351
properties are indeed unaffected by this procedure. Before freeze-drying heme proteins, one should analyze if the protein material contains ferric or ferrous, possibly CO-ligated hemes, as is sometimes the case with heterologous protein expression. In this case, a pure NO-ligated sample can only be prepared after complete CO removal, which is conveniently achieved by heme oxidation with potassium ferricyanide (Antonini, 1971) and stirring under an N2 atmosphere. To prepare an MbCO sample, freeze-dried protein powder is slowly dissolved in cryosolvent to a final concentration of up to 20 mM. The solution is subsequently stirred under 1 atm of CO for 1 h to exchange dissolved O2 for CO. A twofold molar excess of anaerobically prepared sodium dithionite solution is added with a gastight syringe to reduce the heme iron (Nienhaus and Nienhaus, 2005). Ferrous MbNO samples are prepared by dissolving freeze-dried protein powder in (deuterated) cryosolvent and stirring under 1 atm of N2 for 1 h so as to remove dissolved O2. The heme is reduced by a twofold molar excess of anaerobically prepared sodium dithionite solution (in D2O), followed by addition of a twofold molar excess of anaerobically prepared sodium nitrite solution (also in D2O). This preparation is convenient for the preparation of isotope samples with shifted IR bands [see Eq. (18.1) later], e.g., by using Na15N16O2 (Isotec, Miamisburg, OH). Alternatively, one may expose samples to NO gas after reduction. Many heme proteins, including Mb and Hb, undergo reductive nitrosylation of the heme iron in the presence of excess NO (Hoshino et al., 1996), which complicates the preparation of ferric MbNO samples. We again start by preparing a concentrated, deaerated ferric protein solution. Subsequently, the N2 atmosphere above the sample solution is replaced by NO gas. The progress of NO binding to the ferric heme iron is monitored by repeatedly taking UV/visible spectra of small aliquots diluted in deaerated buffer. After nitrosylation of a significant fraction of the hemes, the NO atmosphere above the sample is replaced by N2 gas to prevent reductive nitrosylation. NP4 samples are prepared in the same way, except that monitoring progress of the ligation reaction is not required because autocatalytic reduction does not occur (Roberts et al., 2001).
2.3. Fourier transform infrared cryospectroscopy equipment Fourier transform infrared spectroscopy is an established technique that is well documented in the literature (Gerwert, 1993; Ma¨ntele, 1993). Here we only give a brief overview followed by specific details concerning cryospectroscopy. FTIR spectrometers are available commercially from various companies. In our laboratory, we are equipped with an IFS 66v/S FTIR spectrometer (Bruker, Karlsruhe, Germany). This instrument can measure
352
Karin Nienhaus and G. Ulrich Nienhaus
steady-state and time-resolved spectra using either rapid-scan or step-scan data acquisition schemes. We only cover steady-state experiments here; for time-resolved experiments, we refer to Ko¨tting and Gerwert (2005). To efficiently suppress water vapor and CO2 infrared lines, the spectrometer bench is evacuated and the sample compartment is purged with dry air from an Ecodry dry-air purge gas generator (Zander, Essen, Germany). FTIR transmission spectra (single beam spectra) are collected at a resolution of 2 cm1 using liquid nitrogen-cooled detectors. For ferrous NO samples, spectra are taken with a mercury-cadmium-telluride detector from 1000 to 4000 cm1. The more sensitive indium antimonide (InSb) detector is preferred for recording spectra of CO and ferric NO ligated samples between 1700 and 2300 cm1. Typically 1000 mirror scans are averaged for each spectrum to achieve a high signal-to-noise ratio. For low-temperature data collection, the sample is kept in a cryostat. When cooling samples for extended periods of time and/or performing multiple heat–cool cycles, closed-cycle refrigerator systems are preferable to liquid helium-cooled cryostats as they do not require costly liquid helium coolant. In our laboratory, we use a SRDK-205AW refrigerator cryostat (Sumitomo, Tokyo, Japan), which allows us to cool samples to 3 K. Attached to the cold finger of the refrigerator is a copper block (made from oxygen-free high-conductivity copper) in which the protein sample assembly is encapsulated. The latter consists of a few microliters of protein solution, sandwiched between two CaF2 windows separated by a mylar spacer of 10 mm (ferrous NO samples) and 75 mm (ferric NO samples) thickness, respectively. The temperature at the sample is measured by a silicon diode temperature sensor and regulated with a 50-O resistive heater attached to a digital temperature controller (Model 330, Lake Shore Cryotronics, Westerville, OH). For thermal insulation, the cold finger is enclosed by a vacuum shroud. Its internal volume is filled with N2 gas at room temperature. Below 180 K, when the sample solution is solid, this volume is evacuated by an oil-sealed vacuum pump to a residual pressure of 103 mbar to minimize heat flow to the cold finger. An assembly of 50-mm CaF2 windows (diver’s helmet) around the sample admits both photolysis and infrared monitoring light to the sample. Heme protein samples are photolyzed with a continuous-wave, frequency-doubled Nd:YAG laser (Forte 530-300, Laser Quantum, Manchester, UK) delivering up to 300 mW output at 532 nm. The laser beam is divided by using a beam splitter and focused by lenses on the sample from both sides, yielding a photolysis rate coefficient kL 20 s1 at full laser power. A mask with a pinhole of 1 mm diameter made from adhesive copper or aluminum tape is attached to one of the windows of the sample assembly, which ensures that the laser beam irradiates the same sample area as the infrared beam of the FTIR spectrometer.
353
NO-Ligated Heme Proteins
3. Fourier Transform Infrared Cryospectroscopy 3.1. Environmental effects on ligand IR bands Bond-stretching vibrations are excited in an isolated diatomic molecule with a permanent dipole moment by resonant absorption of infrared light. In a simple classical picture, the two atoms of the molecule are represented by masses m1 and m2 connected by a spring with harmonic force constant k that quantifies the strength of the chemical bond. The vibrational frequency of the system is given by
1 n¼ 2p
sffiffiffi k m
ð18:1Þ
with reduced mass m ¼ m1m2/(m1 þ m2). This equation is often used to estimate isotopic shifts of IR bands. The use of isotopes is helpful for the assignment of IR bands and, moreover, for shifting the absorption to a region where there is less background absorption (or temperature dependence thereof). This point is of relevance for ferrous NO samples, in which the IR bands of bound NO overlay with water bands and the strong amide I band of the peptide carbonyls. Within a heme protein, the force constant k of the ligand is sensitive to chemical bonding, as seen from the fact that heme-bound ligands exhibit bond-stretching frequencies significantly displaced from the gas-phase values of the ligands. For free CO, the stretching frequency n ¼ 2143 cm1, whereas it is 1950 cm1 for heme-bound CO in MbCO. Likewise, n ¼ 1876 cm1 for NO gas, as compared to 1930 and 1610 cm1 for heme-bound NO in MbIIINO and MbIINO, respectively. The local electrostatic field also causes a shift of the IR-stretching vibrations. This vibrational Stark effect is described by the relation
1 Dn ¼ jDmjjEjcos g; h
ð18:2Þ
where Dn represents the band shift from its value without an external electric field, |Dm| denotes the magnitude of the Stark tuning rate (or difference dipole moment), |E| is the magnitude of the electric field at the ligand location, and g denotes the angle between vectors Dm and E. In fact, all band parameters, integrated absorbance (band area), frequency, and bandwidth, are sensitive to the environment, which makes a small ligand an excellent local probe of protein structure and dynamics. In heme proteins, ligand IR spectra frequently consist of multiple bands that
354
Karin Nienhaus and G. Ulrich Nienhaus
are associated with discrete active-site conformations (Alben et al., 1982; Frauenfelder et al., 1991a; Potter et al., 1990), which are difficult to study by other techniques. Even at high spatial resolution, X-ray crystallography is usually unable to resolve these subconformations. Each of the discrete stretching bands displays a pronounced broadening due to further structural heterogeneity, which is sometimes directly visible via an effect called ‘‘kinetic hole burning,’’ in which a ligand-stretching band shifts during ligand rebinding at low temperature (Huang et al., 1997; Nienhaus et al., 2002; Ormos et al., 1998). This behavior indicates that the associated rebinding energy barrier and the band position are governed by a common structural coordinate. Interactions between the ligand and its environment are not static, but subject to changes as a consequence of conformational dynamics and thermal equilibrium motions of both ligand and protein. The equilibrium effects are clearly visible in Figs. 18.1A and 18.1B, in which IR spectra of CO and NO bound to the heme iron of Mb are plotted for select temperatures between 3 and 300 K. These difference spectra were calculated from transmission spectra, I(n), of MbCO and MbIIINO and the aquomet derivative of the same protein preparation,
AðnÞ ¼ log ½IMbmet ðnÞ=IMbL ðnÞ:
B
Absorption
A
ð18:3Þ
1925
Area
C
1950 1925 Wavenumber (cm−1)
1950
1.0 0.5 0.0
0
100 200 Temperature (K)
300
Figure 18.1 Infrared absorption spectra of heme-bound ligands in (A) MbCO at (top to bottom in the dominant peak) 3, 50, 100, 150, 200, 250, and 290 K and (B) MbIIINO at the same temperatures. (C) Integrated IR absorption (band area) of the ligands for MbCO (open symbols) and MbIIINO (closed symbols) plotted as a function of temperature.
NO-Ligated Heme Proteins
355
As is evident from Fig. 18.1, temperature strongly influences both intensity and frequency of the bands. From 3 to 300 K, the integrated absorbances of bound CO and NO decrease by 22 and 74%, respectively (see Fig. 18.1C). One explanation of these temperature dependencies involves the dielectric properties of the ligand environment. In the protein, the ligand molecule is embedded in a polarizable medium to which its charge oscillations are coupled. This effect causes an enhancement of the overall oscillator strength, which decreases with increasing temperature. However, both ligands reside in a similar environment, and therefore, the pronounced difference in their temperature dependencies suggests that dielectric effects play only a minor role. The observed integrated absorbance also decreases if the ligand performs fast librational motions that change the transition dipole orientation on subpicosecond timescales. Indeed, the substantial loss of absorbance observed for NO bound to ferric NP4 has been ascribed to substantial ligand-bending vibrations (Nienhaus et al., 2004). The temperature-dependent loss of band intensity due to reorientational dynamics is especially pronounced for photodissociated ligands trapped in internal cavities of the protein, where they are not covalently bound and thus can perform large-amplitude oscillations (Kriegl et al., 2003; Lehle et al., 2005). In addition to intensity changes, temperature-dependent band shifts may also arise from librational motions because ligand and polar side chains may change their average mutual distance. A comparison of the CO and NO spectra in Figs. 18.1A and 18.1B reveals a pronounced temperature shift of the dominant band for NO but not for CO, which indicates a larger change in the coupling of the NO ligand to a nearby residue due to thermal motions. This observation is in line with the larger mobility of the bound NO ligand that was already inferred from the temperature dependence of the band intensity.
3.2. Photolysis difference spectroscopy Mb is probably the best-studied heme protein with respect to its ligand-binding reaction (Antonini, 1971; Frauenfelder et al., 1991a; Nienhaus and Young, 1996; Olson and Phillips, 1996). Most IR studies on Mb have focused on the vibrational absorption bands of CO and not the physiologically more relevant ligands O2 and NO, mainly for two technical reasons:(1) The absorbance of CO bound to the heme iron is exceptionally strong and in a region of the spectrum devoid of other protein bands and (2) CO can be photodissociated with near-unity quantum yield. As a consequence, IR photolysis difference spectra can be collected on MbCO with uncompromised precision. The similar sizes of the diatomic ligands CO, NO, and O2 suggest that at least part of the observations made with CO may also be of relevance for the other two ligands.
356
Karin Nienhaus and G. Ulrich Nienhaus
Figure 18.2 shows absorbance difference spectra of MbCO at pH 7.0 in the spectral regions of heme-bound and photodissociated CO. The absorbance spectrum, A(n), represented by the solid line, was calculated from transmission spectra, Idark(n) and Ilight(n), taken at 3 K before and after laser illumination for 1 s, respectively,
AðnÞ ¼ log ½Idark ðnÞ=Ilight ðnÞ:
ð18:4Þ
In the spectral region of the heme-bound CO-stretching absorption, the IR spectrum displays three discrete bands at 1966, 1945, and 1927 cm1 (Alben et al., 1982; Ansari et al., 1987; Fuchsman and Appleby, 1979; Johnson et al., 1996; Li et al., 1994; Mourant et al., 1993; Shimada and Caughey, 1982). They are denoted as A0, A1, and A3, respectively, and arise from three discrete bound-state conformations denoted as ‘‘taxonomic’’ or ‘‘A’’ substates (Frauenfelder et al., 1991a,b). Experiments with many distal heme pocket mutants have revealed that electrostatic interactions between the CO dipole and the imidazole side chain of the distal histidine, H64, which can assume different conformations, are responsible for the dispersion of the CO-stretching absorption into three separate A substate bands (Braunstein et al., 1993; Kushkuley and Stavrov, 1996, 1997; Li et al., 1994; Mu¨ller et al., 1999). At neutral pH, A1 and A3 dominate the spectrum; the A0 band is only weakly populated. At low pH, the A0 population grows at the expense of A1 and A3 with a pK of 4.5, which has been assigned to protonation of the imidazole side chain of H64 by mutational analysis (Mu¨ller et al., 1999).
Absorption (mOD)
150
MbCO
B2
100 50 0 −50
B1 B0
Heme-bound CO A3
A0
−100 −150
A1 1920
Photolyzed co 20 A1
1950 2120 2140 Wave number (cm−1)
Figure 18.2 Photolysis difference spectra of MbCO at 3 K in the spectral regions of heme-bound and photolyzed CO obtained after 1 s of illumination at 3 K (solid line) and slow cooling from 160 to 3 K under constant illumination (dotted line). For photodissociated CO, the absorbance was multiplied by 20.
NO-Ligated Heme Proteins
357
X-ray structure analysis has revealed that the imidazolium moiety does not stay inside the hydrophobic distal heme pocket but rotates toward the solvent to screen its charge (Yang and Phillips, 1996). Consequently, in the A0 conformation, the CO interacts only weakly with the remaining atoms in the distal pocket and displays a frequency similar to those observed for mutants in which H64 is replaced by small aliphatic amino acids (Li et al., 1994; Quillin et al., 1993). The shifts to lower frequencies observed for A1 and A3 result from a positive partial charge on Ne-H of the H64 imidazole side chain, with a stronger charge coupling for the A3 ˚ X-ray substate. In 1999, Vojtechovsky and colleagues refined their 1.2-A structure of MbCO with multiple occupancies of H64 and identified—in addition to 20% occupancy for A0—two slightly different positions of the imidazole side chain inside the pocket, occupied to 60 and 20% (Vojtechovsky, 1999). In agreement with the interpretations based on the CO-stretching bands, they interpreted these two substates as A1 and A3, with the distal histidine positioned deeper in the distal pocket and closer to the CO in A3. This assignment is also supported by kinetic studies of A substate interconversions: exchange between A1 and A3 is orders of magnitude faster than the A1/A3 exchange with A0, which suggests that the former involves smaller structural changes than the latter ( Johnson et al., 1996). Detailed pH-dependent kinetic studies suggested the presence of three active-site conformations also for MbO2 (Tetreau et al., 2002). Likewise, the spectra of MbNO exhibit similar features (see later). At 3 K, the integrated absorbance of the stretching bands of the photodissociated CO is 25 times smaller than the one of the A bands. In the difference spectrum, three B bands were originally identified and labeled B0 (2149 cm1), B1 (2131 cm1), and B2 (2119 cm1) (Alben et al., 1982). They appear near the gas-phase frequency of 2143 cm1, indicating that the photodissociated CO experiences only weak interactions with the protein. The observation of quantum mechanical tunneling of ligands in heme proteins below 50 K (Alben et al., 1980; Lamb et al., 1998) implies that the CO resides close to the heme iron after photolysis at these temperatures. Its location on top of the heme macrocycle was determined by X-ray cryocrystallography on MbCO crystals under photolyzing illumination (Hartmann et al., 1996; Schlichting et al., 1994; Teng et al., 1994). Figure 18.3 shows the active site of MbCO, with CO in the bound-state (A) and photodissociated-state (B) positions. Multiple B bands arise because the three active-site conformations persist after photolysis at low temperatures and interact differently with photodissociated CO. Moreover, CO has been observed to adopt opposite orientations in docking site B with different electrostatic coupling to its surroundings, causing a Stark splitting of the absorption band (Lim et al., 1997; Nienhaus et al., 2003c, 2005). Therefore, two bands, B1(A1) and B2(A1), are present
358
Karin Nienhaus and G. Ulrich Nienhaus
for the majority substate A1. The CO in Fig. 18.3 is depicted in the B2 orientation (Meuwly, 2006; Nienhaus et al., 2005). We have been able to show that B0 is actually the corresponding ‘‘B1’’ band of A3, and B2(A3) strongly overlaps with B2(A1) in the IR spectrum (Bredenbeck et al., 2007). For A0, which is only weakly populated at neutral pH, B bands have been observed at 2118, 2128, and 2138 cm1 (Mourant et al., 1993). Figure 18.2 displays another spectrum in the region of photodissociated CO (dotted line), which was obtained after slow cooling of the sample under laser illumination from 160 to 3 K instead of 1-s photolysis at 3 K. Apparently, photoproduct spectra depend on the temperature at which the sample was illuminated. Even more pronounced changes have been observed for a variety of MbCO mutants (Lamb et al., 2002; Nienhaus and Nienhaus, 2002; Nienhaus et al., 2003a,b,c). They reflect CO migration from the primary docking site B to secondary docking sites C and D (see Fig. 18.3). In the secondary sites, which coincide with two of four hydrophobic xenon-binding cavities, Xe4 and Xe1, respectively (Nienhaus et al., 2003a,b; Ostermann et al., 2000; Tilton et al., 1984), the CO experiences different local electric fields that are responsible for the spectral changes. These ligand migration processes, however, cannot be unraveled by photolysis difference spectroscopy at a single low temperature, but require temperature-dependent experiments. In our research, we have extensively employed a technique called temperature derivative spectroscopy (TDS) to explore ligand migration and dynamics at low temperatures. Gly25 Leu29
Ile28 Leu32
His64
C Ile111
B Ile107
A Heme
Leu104
His93 D
Figure 18.3 Active-site structure of MbCO, with the CO molecule depicted as bound to the heme iron (A), in the primary docking site (B), and in secondary sites (C and D).
359
NO-Ligated Heme Proteins
3.3. FTIR-temperature derivative spectroscopy (FTIR-TDS) 3.3.1. Theoretical background This temperature ramp protocol is a convenient method for the investigation of thermally activated rate processes that are characterized by distributed enthalpy barriers (Berendzen and Braunstein, 1990; Mourant et al., 1993; Nienhaus et al., 1994). The technique has found widespread application in the study of CO migration and binding in heme proteins but has only recently been introduced to investigations of NO ligands (Nienhaus et al., 2004). It is one of a variety of two-dimensional spectroscopic techniques that disperses the information not only over frequency space, but also over the temperature, T, at which rate processes become thermally activated. Temperature derivative spectroscopy is a relaxation method in which a nonequilibrium state is created by perturbation of a sample at a temperature at which approach to equilibrium is slow. In our heme-protein studies, the sample is perturbed by ligand photolysis, after which its temperature, T, is ramped up linearly in time, t, in the dark at a constant rate b (usually 5 mK/s), starting at an initial low temperature Ti,
T ¼ Ti þ bt:
ð18:5Þ
During the temperature ramp, IR spectra are taken continuously (1 spectrum/ K). They reveal the response of the sample after the perturbation, including ligand rebinding to the heme iron. At cryogenic temperatures (T < 180 K in our cryosolvent), rebinding is characterized by a static distribution of rebinding enthalpy barriers due to the absence of large-scale conformational motions (Frauenfelder et al., 1991a; Parak et al., 1988). Protein molecules with small barriers rebind their ligand first, and the continuous increase in temperature causes molecules to rebind sequentially with respect to their barrier heights. TDS is a so-called ‘‘rate window method,’’ which activates all processes to occur on a characteristic timescale, which is 100 s at the heating rate used in our experiments. Figure 18.4A shows plotted integrated intensities from a TDS experiment on MbCO, i.e., band areas missing in the A1 and A3 substates at a particular temperature due to photolysis at 3 K. The decay over a temperature interval of several 10 K is because of rebinding. In the following, we take these data to represent the fractions, N(T ), of molecules that are still photodissociated from heme-bound substates A1 and A3. For the simple case of a unimolecular rate process with a distribution of activation enthalpy barriers, data can be analyzed quantitatively. We only briefly sketch the relevant equations here; details can be found in Berendzen and Braunstein (1990).
360
Karin Nienhaus and G. Ulrich Nienhaus
A N(T)
A1 A3
B −dN/dT
A1
A3 20
g(H) (mol/kJ)
C
40 60 80 Temperature (K)
100
A = 108.9 s−1 HP = 10.8 kJ/mol s = 2.7 kJ/mol A = 1010.4 s−1 HP = 19.5 kJ/mol s = 3.7 kJ/mol 0
5
10
15 H (kJ/mol)
20
25
Figure 18.4 (A) Integrated intensities (band areas) missing in the A1 and A3 substate bands at a particular temperature during aTDS experiment after 1 s of photolysis at 3 K, calculated from experimental data on MbCO. (B) Successive differences (numerical derivatives) of the population decay data shown in (A). Solid lines are fits of data with Eq. (18.10), using Gaussian functions, Eq. (18.11), to model the enthalpy barrier distributions. (C) Enthalpy barrier distributions of A1 and A3 substates obtained from these fits; fit parameters are included in the figure.
Geminate rebinding of ligands after photodissociation is a unimolecular process described by a first-order rate equation,
dN ¼ kN ; dt
ð18:6Þ
where k is the rate coefficient given by the Arrhenius relation,
T exp ðH=RT Þ; k¼A T0
ð18:7Þ
361
NO-Ligated Heme Proteins
with pre-exponential A, temperature T, and reference temperature T0 ¼ 100 K. Because of the linear temperature ramp, T 1 t, we can solve Eq. (18.6) and calculate the derivative of N with respect to T,
Z T dN Ni k 0 Ni k exp k expðyÞ: ¼ dT ¼ b b dT Ti b
ð18:8Þ
The integral abbreviated by y describes the temperature (and time) dependence of k according to Eqs. (18.5) and (18.7); its evaluation yields
2 # T 2 H Ti H E3 E3 T0 T0 RT RTi Z 1 xt e with E3 ðxÞ ¼ dt: t3 1 AT0 y¼ b
"
ð18:9Þ
Here, E3 denotes the exponential integral of order 3. Finally, we introduce the distribution (probability density) of activation enthalpy barriers, g(H ), into Eq. (18.8), which quantifies the probability with which a certain enthalpy barrier height occurs in the protein ensemble,
dN Ni ¼ b dT
Z
1
k expðyÞgðHÞdH:
ð18:10Þ
0
According to Eq. (18.10), the (negative) temperature derivative of N, –dN/ dT, is equivalent to g(H ) broadened by a (temperature-dependent) resolution function, k exp(–y). In the quantitative TDS analysis, g(H ) is extracted from –dN/dT, which is calculated numerically simply as –DN/DT by taking differences between IR spectra at successive temperatures so that DT ¼ 1 K. For the integrated absorbances of A1 and A3 in Fig. 18.4A, the resulting derivatives are shown as points in Fig. 18.4B. The solid lines are fits of data with enthalpy barrier distributions represented by Gaussian model functions,
gðHÞ ¼ ð2ps2 Þ1=2 exp½ðH HP Þ2 =2s2 :
ð18:11Þ
Here, HP and s are the peak and the standard deviation of the distribution, respectively. Fig. 18.4C shows the enthalpy barrier distributions resulting from fits in which HP and s were varied, the pre-exponential A was fixed at the value determined by time-resolved IR spectroscopy (Steinbach et al., 1991). The fit parameters are included in Fig. 18.4C. The excellent agreement of data and fitted curve suggests that the method is able to precisely determine enthalpy barrier distributions. However, we note that
362
Karin Nienhaus and G. Ulrich Nienhaus
a quantitative analysis requires that the observed rate processes are simple first-order reactions with a temperature dependence given by the Arrhenius relation. At very low temperature, however, tunneling becomes significant and causes systematic errors, as is evident in Fig. 18.4B below 30 K. Moreover, the Arrhenius pre-exponential A needs to be known from kinetic experiments. In principle, A can also be determined by multiple TDS experiments with ramp rates differing by orders of magnitude, but that approach appears hardly feasible in practice. We also remark that the analysis is based on spectroscopic data, which may contain temperature-dependent intensity changes and other spectral changes that do not originate from population changes due to reactions. However, even without a quantitative analysis, TDS has proven valuable as a qualitative tool because of its ability to separate processes according to their temperature of activation. 3.3.2. FTIR-TDS on MbCO as an example Temperature derivative spectroscopy is a differential method in which relatively small effects are dispersed over 100 or more spectra. Consequently, an excellent signal-to-noise ratio is required in the raw data. Figures 18.5A and 18.5B show results from a TDS experiment on an MbCO sample, pH 7.0, started immediately after photodissociation for 1 s at 3 K. One spectrum was collected every 200 s, so that altogether 118 transmission spectra were obtained during the temperature ramp from 3 to 120 K with a heating rate of 5 mK/s. Data are presented as two-dimensional contour plots, in
B
Wave number (cm−1)
A
2000 1980
C 2000 1980
1960
1960
1940
1940
1920 B1(A1) A1 B1(A3) A3 1900 D 2160 B (A ) B1(A3) 2150 2 3 2140 2130 2120 2110 B2(A1) B1(A1) 2100 20 40 60 80 100 120 0 Temperature (K)
1920 1900 2160 2150 2140 2130 2120 2110 2100 0
B
A0
D A3 B A1 B A3
C
A1 D A1
50 100 Temperature (K)
150
Figure 18.5 TDS contour maps of MbCO (A, B) after 1 s of illumination at 3 K and (C, D) after slow cooling from160 to 3 K under continuous illumination. (A, C) Absorption changes in bands of heme-bound CO; (B, D) absorption changes in photoproduct bands. Contours are spaced logarithmically, solid and dotted lines represent an absorption increase and decrease, respectively, and solid and dotted arrows indicate rebinding processes and ligand reorientation in docking site B, respectively.
NO-Ligated Heme Proteins
363
which the absorbance changes between successive spectra are plotted versus temperature and wave number in the spectral regions of the heme-bound (see Fig. 18.5A) and photodissociated (see Fig. 18.5B) ligands. The contour spacing is usually done logarithmically, which emphasizes weak features, and solid and dashed lines depict gain and loss of absorption with temperature, respectively. These plots are a convenient means to visualize a set of absorbance difference spectra calculated from transmission spectra at successive temperatures,
! I n; T þ 12 K : DAðn; T Þ ¼ log I n; T 12 K
ð18:12Þ
In addition to absorbance changes because of ligand recombination, spectra also contain temperature-dependent changes in the absorption background onto which the ligand-stretching bands are superimposed. Therefore, a quadratic (or cubic) baseline is fitted to flat regions of each absorption spectrum to the left and the right of the absorption bands and is subtracted from the spectrum. If the IR bands show strong intrinsic temperature dependence, it is advantageous to take additional background spectra as a function of temperature before calculating the numerical derivative, to which TDS data are referenced. For example, to produce excellent TDS maps of heme-bound CO in MbCO, a TDS run without photolysis on the identical sample provides a reference set of transmission spectra, Iref (n, T ), with which we calculate derivative spectra as
! ! I n;T þ 12 K I n;T 12 K log : ð18:13Þ DAðn;T Þ ¼ log Iref n;T þ 12 K Iref n;T 12 K
To further decrease the noise and improve the readability of the contour plots, data may be averaged along the temperature and wave number axes. In the A substate map in Fig. 18.5A, the dominant A1 substate is seen to rebind between 3 and 100 K, with a peak at 49 K. The temperaturedependent shifts of the A1 feature along the wave number axis reflect kinetic hole burning. Because of its higher enthalpy barriers, the A3 substate peaks at a higher temperature of 78 K. Even the weak A0 band is still resolvable, as is the weak replica of the A1 band of the 13C16O isotope at 1900 cm1 due to the natural abundance of 13C of 1.1%. Absorbance increase in the A bands as a consequence of recombination is accompanied by a loss of photoproduct species and thus features of the opposite sign in Fig. 18.5B. However, for temperatures below 25 K, positive and negative features are observed within the photoproduct region at the same temperature, which reveal population exchange between photoproduct species and not rebinding. These changes are due to part of the CO population reorienting within the B states before they rebind at higher temperature.
364
Karin Nienhaus and G. Ulrich Nienhaus
Figures 18.5C and 18.5D show TDS contour maps of the identical sample. Instead of 1-s illumination at 3 K, the sample was cooled from 160 to 3 K at 5 mK/s during illumination, i.e., over several hours. This procedure enables CO ligands to explore secondary docking sites C and D (see Fig. 18.3), which they cannot reach upon photolysis at 3 K. By slowly cooling under illumination, we ensure that a significant fraction of the CO molecules becomes trapped in these sites. At 3 K, the photolysis light is switched off and the TDS measurement is started. The trapped molecules rebind once the temperature is again high enough to overcome the barriers against bond formation at the heme iron. For A1, there are three discrete populations in Fig. 18.5C, with rebinding peaks at 50, 80, and 120 K. By using MbCO mutants in which the xenon cavities are selectively blocked by tryptophan side chains, we were able to assign these features to rebinding from the primary docking site B and secondary docking sites C (Xe4) and D (Xe1) depicted in Fig. 18.3 (Nienhaus et al., 2003a,b). In Fig. 18.6A, we have plotted the band areas of the A1 substate from TDS experiments on wild-type MbCO and two mutants after slow-cool illumination as a function of temperature. These data are obtained from a set of TDS spectra, e.g., those of wild-type (wt) MbCO in Fig. 18.5C, by integration along the wave number axis. Peak D at 120 K is completely suppressed in L104W MbCO, A L104W
wt
B
Absorption
I28W
CO missing from bound state
Photodiss. CO
0
50
100
150
Temperature (K)
Figure 18.6 (A) TDS data integrated over the A1 substate band on wild-type MbCO (see Fig. 18.5C) and mutants I28Wand L104Wafter slow-cool illumination as a function of temperature. (B) Integrated absorption missing from bands of heme-bound CO (solid line) and present in photoproduct bands (dotted line) of mutant MbCO L29WS108L during aTDS ramp after slow-cool illumination (160 to 3 K).
NO-Ligated Heme Proteins
365
which has the Xe1 site occluded by the indole side chain of W104. Peak C is strongly suppressed in the I28W MbCO sample, in which the Xe4 site is efficiently blocked. In MbCO, these docking sites play important roles in physiological ligand binding by enhancing the probability of ligands to escape from the protein after bond dissociation (Nienhaus et al., 2003b; Scott and Gibson, 1997; Scott et al., 2001). The slow-cool experiments presented here are convenient for obtaining a quick overview of various processes that can occur in the protein upon photolysis. Other more specific illumination procedures can be devised to focus on particular features (Kriegl et al., 2003; Lehle et al., 2005). For example, ligands can be selectively deposited in docking site C or D and investigated individually. In order to analyze the relative fractions of boundstate or photoproduct-state species, one has to be careful to account for temperature-dependent intensity changes. While this is not an important issue for heme-bound CO, it is a serious concern for ferric heme-bound NO, as already seen in Fig. 18.1. Moreover, photodissociated ligands in protein cavities perform substantial reorientational motions, leading to enormous losses in band intensity. Figure 18.6B illustrates this point with IR data on mutant MbCO L29W-S108L after slow-cool illumination TDS. This plot does not show derivatives, but the entire (missing) integrated absorption in the heme-bound bands and the absorption in the photoproduct bands as a function of temperature. Both curves should coincide if the band areas would represent fractional populations of ligands, but they are significantly different. Whereas a pronounced plateau between 40 and 90 K is visible for heme-bound CO as a consequence of efficient ligand trapping in docking site C before rebinding at 120 K, absorbance in the photoproduct bands decays in an essentially continuous fashion. This behavior arises from thermal activation of CO librations in docking site C, which affects both band intensities and band shifts. These effects can be exploited to study electric fields and ligand dynamics in cavities (Kriegl et al., 2003; Lehle et al., 2005). Data in Fig. 18.6B show that one should be careful when taking integrated absorbances to represent fractional ligand populations.
4. Low-Temperature FTIR Spectroscopy on NO-Ligated Heme Proteins 4.1. Nitric oxide-ligated myoglobin 4.1.1. Photolysis difference spectra In our experimental setup, illumination for 1 s is entirely sufficient to completely photodissociate MbCO. In contrast, there is only very little photoproduct generated after 1 s of irradiation of MbNO. Therefore, we illuminate NO-ligated samples for 20 s at 3 K to obtain a photolyzed
366
Karin Nienhaus and G. Ulrich Nienhaus
fraction of more than 50%, which is sufficient for collection of high-quality photolysis difference spectra. Essentially complete photolysis is achievable but takes on the order of hours. This behavior indicates that both ferrous and ferric hemes are extremely reactive toward NO, so that many ligand trajectories again terminate in the bound state right after photodissociation. Apparently, it takes multiple attempts for NO ligands to settle in the primary and secondary docking sites at 3 K. Figure 18.7A shows a photolysis difference spectrum of MbIINO (pH 6.9) after 20 s of illumination at 3 K. Two absorption bands are visible at 1607 and 1614 cm1. These frequencies are much lower than for gas phase NO (n ¼ 1876 cm1), indicating that the NO bond is substantially weakened upon bond formation with the ferrous heme iron. Based on extensive mutant studies by Coyle et al. (2003), we suggest that the two absorption bands should be assigned to the A3 (n ¼ 1607 cm1) and A
60 40
MbIINO
20 0
Heme-bound NO Photodiss. NO 10
B
Absorption (mOD)
−20 −40 1600 1620
1840
1860
20 MbIIINO 10 Heme-bound NO
0 −10
Photodiss. NO 10
−20 1840
1860 1920 Wave number (cm−1)
1940
Figure 18.7 (A) Photolysis difference spectra of (a) MbIINO and (B) MbIIINO in spectral regions of heme-bound and photolyzed NO obtained after a 20-s illumination at 3 K (solid line) and slow cooling from 160 to 3 K under constant illumination (dotted line). Photoproduct bands are scaled up by a factor of 10; the dashed-dotted line represents the 20-s photoproduct band of MbIIINO scaled to the same height as the one obtained after slow cooling under illumination.
NO-Ligated Heme Proteins
367
A1 (n ¼ 1614 cm1) conformations. The photoproduct bands appear at 1856.6 and 1867.4 cm1, and there is also a weak shoulder at 1852.3 cm1. The overall doublet shape suggests an assignment of the two major bands to B1 and B2 rotamers of NO in the primary docking site, as for CO. The integrated intensity of the photoproduct bands is 15 smaller than for the bands of heme-bound NO. In Fig. 18.7A, a second photoproduct spectrum is represented by the dotted line. This spectrum appears after illumination during slow cooling from 160 to 3 K. In the spectral region of heme-bound NO, it merely indicates that more protein molecules can be photodissociated by this procedure than with a 20-s illumination at 3 K. Apparently, NO rebinds much more slowly in a fraction of Mb molecules after this photolysis procedure. In the spectral region of photodissociated NO, the IR spectrum is completely different from the one measured after photolysis at 3 K and consists of two narrow bands at 1864.8 and 1868.6 cm1; their overall integrated absorbance is 17-fold smaller than the one of the corresponding A bands and is thus similar to the ratio of 15 determined for the 3-K photoproduct after 20 s of illumination. These findings suggest that NO has migrated to a secondary docking site in which it can assume two different orientations. Meuwly and co-workers performed ab initio calculations of the potential energy surface for NO and imidazole-bound ferrous heme and found a minimum associated with reverse binding of NO (Fe2þ–ON), which should give rise to photoproduct bands 20 cm1 to the blue of the bound-state IR bands (Nutt et al., 2005). Although we should expect to be able to populate this state at 3 K, we failed to detect photoproduct bands in the region of 1600 50 cm1 that could be assigned to the Fe2þ–ON species. In MbIIINO (pH 7.3), the stretching bands of heme-bound NO appear between 1900 and 1950 cm1, i.e., in the same frequency range as the bands of the CO-bound complex and at higher frequency than gas-phase NO. The NO bond order apparently increases upon binding to the heme iron. This effect arises from a net transfer of the unpaired electron in the antibonding NO p* orbital to the heme iron, creating essentially an Fe2þ–NOþ unit isoelectronic with Fe2þ–CO. The band at 1927 cm1 dominates the spectrum, and there are two minor bands peaking at 1914 and 1942 cm1. The same number of ligand-stretching bands as in MbCO suggests that similar taxonomic substates exist. The dominant band would thus be assigned as A1. pH-dependent studies reveal that the low-frequency band at 1908 cm1 grows with decreasing pH, which implies that A0 and A3 are exchanged in their frequencies with respect to A1. This effect was noticed earlier by Miller and Chance and was associated with the presence of the His64 tautomer pointing the lone-pair orbital of the unprotonated Ne toward the NO (Miller et al., 1997). Its negative partial charge should create a local electric field that is opposite to the Ne-protonated species present in MbCO and thus revert the frequency ordering of the A substates.
368
Karin Nienhaus and G. Ulrich Nienhaus
The photoproduct spectrum of MbIIINO obtained after 20 s of illumination displays a single, weak, and Gaussian-shaped band at 1869.0 cm1, which Miller and Chance were not able to observe (Miller et al., 1997). Accordingly, the ratio of integrated absorbances between the heme-bound and photodissociated NO at 3 K is rather large, 160. We also note that the absence of a photoproduct band at 1858 cm1 indicates that the sample is not contaminated with MbIINO. As with MbIINO, significantly more protein molecules can be photodissociated upon slow cooling under light, as shown by the dotted spectrum in the bound-state bands in Fig. 18.7B, so that the yield of photoproduct increases to 90% of the total population. The integrated absorbance of the photoproduct bands is much larger after slow cooling, so that the ratio of the integrated absorbances decreases to 24. A fit with two Gaussians reveals band positions at 1864.5 0.3 and 1868.5 0.3 cm1, which are identical within the experimental error to those of the photoproduct bands of MbIINO after slow cooling under light. Therefore, it appears that the photoproduct after slow cooling is identical for MbIINO and MbIIINO. 4.1.2. Temperature derivative spectroscopy experiments In the contour plots of the heme-bound and photodissociated NO of MbIINO in Figs. 18.8A and 18.8B, respectively, we observe significant rebinding already at 3 K and complete rebinding at 25 K. This temperature dependence suggests that after a 20-s photodissociation at 3 K, NO ligands reside close to the heme iron, from where they rebind by overcoming very low enthalpy barriers. Presumably, the NO docks in primary docking site B, as is suggested by the doublet structure reminiscent of the B states in MbCO. However, interconversion processes as observed for CO are absent, and therefore, the current structural assignment is only tentative. To clarify this issue, further data are needed, either from cryocrystallography or from IR spectroscopy of Mb mutants known to have modified B state spectra in the CO-ligated form. It appears that NO ligands rebind as soon as their thermal energy is sufficient to overcome the weak interactions in the primary docking site. In contrast, CO can reorient in site B without rebinding because it experiences a significant enthalpy barrier against bond formation (see Fig. 18.5B). For MbIIINO, the TDS map obtained after a 20-s illumination at 3 K also shows predominant NO rebinding below 25 K (see Figs. 18.8C and 18.8D). The weak contour at 40 K may indicate that rebinding in A3 requires more thermal activation due to the H64 imidazole in close vicinity of the active site. For A1, a weak tail in the contours extends to 65 K. The population rebinding at higher temperature is significantly enhanced under slow-cool illumination, yielding a huge peak at 72 K (see Fig. 18.8E). This finding suggests that a large fraction of NO migrates to a secondary site under slow-cool illumination. Even after 20 s of light at 3 K, a small fraction already escapes to this state, as indicated by the
369
NO-Ligated Heme Proteins
A
1660
B 1880
1640
1870
1620
1860
C
E
Wave number (cm−1)
1600 1580 1960
1850 Mb(II)NO, 20 s illum.
1840 D1880
1940
1870
1920
1860
1900 1880 1960
1850 Mb(III)NO, 20 s illum. F
1940 1920
1870 1860
1900 1880 0
1840 1880
1850 Mb(III)NO, slow cool illum. . 20
40 60 80 100 120 Temperature (K)
1840
0
20
40 60 80 100 120 Temperature (K)
Figure 18.8 TDS contour maps of MbNO. The experiment was started after (A, B) 20 s of illumination of MbIINO at 3 K, (C, D) 20 s of illumination of MbIIINO at 3 K, and (E, F) slow-cool illumination of MbIIINO (160 to 3 K). Left column: absorption changes in bands of heme-bound NO. Right column: absorption changes in photoproduct bands. Contours are spaced logarithmically; solid/dotted lines indicate an absorption increase/decrease.
tail in Fig. 18.8C. For rebinding from this site, the rate-determining barrier is not located at the heme iron but rather between the primary and the secondary docking sites. By using the same strategy for the identification of docking sites as for MbCO, namely incorporating bulky side chains that block access to cavities, we have identified docking site C, the Xe4 cavity, as the one that is occupied by NO (data not shown). From identical photoproduct difference spectra after slow-cool illumination, we expect that the same ligand migration process is also present in MbIINO. Indeed, the ferrous protein has essentially identical TDS maps for NO rebinding from this secondary site (data not shown). Therefore, we conclude that NO migrates to the Xe4 cavity in both MbIINO and MbIIINO. Despite their similar sizes, there are distinct differences in ligand migration in Mb between CO and NO. In wild-type MbCO, CO cannot migrate to site C at 3 K, whereas NO does that readily in both MbIINO and MbIIINO. CO moves to site D (Xe1 cavity) above 100 K, but we were
370
Karin Nienhaus and G. Ulrich Nienhaus
unable to detect any sign of NO migration into site D. As already observed from photolysis difference spectra, NO in site C is characterized by a doublet of bands at 1864.5 and 1868.5 cm1. In analogy to the experiments with CO (Kriegl et al., 2003; Lehle et al., 2005), we assign these two bands to two discrete orientations of the NO in the Xe4 cavity. We emphasize that the contours in the photoproduct map of these two photoproduct bands (see Fig. 18.8F) are not only due to rebinding, but partially reflect the decrease of band area with temperature due to librational motions of NO in the trap site. The weak, single photoproduct state of MbIIINO raises questions as to its structural properties. The large integrated intensity ratio of 160 between bound state and photoproduct could indicate that additional photoproduct states exist outside of the spectral range covered in Fig. 18.7B. However, careful inspection has not revealed any additional photoproduct bands. For the secondary photoproduct, the doublet of NO-stretching bands close to the gas-phase value indicates weak perturbations of the electronic system of the ligand as a consequence of the local electric field in the Xe4 cavity. However, if NO is also weakly interacting with its surroundings in the primary photoproduct site as inferred from its band at 1869.0 cm1, how can its vibrational transition dipole be so much reduced in comparison to its value in the secondary site? A possible explanation is provided by the work of Nutt and Meuwly (2007), who have argued that a metastable Fe–ON species should also exist for ferric heme, characterized by a red shift of 50 cm1 from the Fe–NO-bound state, which is similar to the red shift of the primary photoproduct in our data by 58 cm1. Considering the partial charges on the N and O atoms reported by Nutt and Meuwly (2007), the Fe–ON species could indeed have a weak stretching absorption. However, clarification of this issue requires further investigations, for example, identification of a Fe3þ–ON-stretching vibration of the primary photoproduct by Raman spectroscopy.
4.2. Nitric oxide-ligated nitrophorin 4 (NP4) Nitrophorin 4 is a heme protein of the blood-sucking insect Rhodnius prolixus, which is structurally completely different from the globins (Andersen et al., 1998; Montfort et al., 2000). Its polypeptide chain has a b-barrel lipocalin fold, with a highly nonplanar, ruffled heme inserted into one end of the barrel. NP4 functions as a transporter of NO from the salivary glands of the insect to its victim’s tissue. NO release results in vasodilation and reduced blood coagulation, therefore facilitating blood uptake by the insect. As NO-ligated ferrous heme is extremely stable, NP4 relies on a ferric heme for reversible NO binding. In Fig. 18.9, the 3-K photolysis difference spectra of NP4NO, pH 7.5, measured after 20 s of illumination at 3 K (solid line) and slow cool
371
NO-Ligated Heme Proteins
Absorption (mOD)
30 Heme-bound NO
0 −30
Photolyzed NO 20
−60 NP4NO −90 1860
1870 1900 1920 Wavenumber (cm−1)
1940
Figure 18.9 Photolysis difference spectra of ferric NP4NO at 3 K in spectral regions of heme-bound and photolyzed NO collected after a 20-s illumination at 3 K (solid line) and slow cooling from 160 to 3 K under continuous illumination (dotted line). Overall absorption because of heme-bound NO is represented by the dashed line.The photolysis difference spectrum at 140 K, measured after 2 h of illumination at 185 K and subsequent slow cooling under continuous illumination to 140 K, is plotted with spheres.
illumination (160 to 3 K, dotted line), are shown together with a difference spectrum referenced against an aquomet sample (dashed line). Apparently, a significant fraction of ligands cannot be kept photodissociated during the 200 s that it takes to acquire the spectrum, which indicates very low rebinding barriers. Spectra display two absorption bands of bound NO at 3 K, a large band at 1908 cm1 and a smaller one at 1922 cm1 (see Fig. 18.9). Our lowtemperature IR studies revealed that these bands are associated with two discrete conformations of the protein. In the ‘‘closed’’ A1908 species, the AB and GH loops pack around the heme-bound ligand, whereas the active site of the ‘‘open’’ A1922 conformation is more solvent accessible. The population ratio is pH dependent; at higher pH, the band at 1922 cm1 dominates the spectrum (Nienhaus et al., 2004). We already mentioned that the A1908 band is strongly temperature dependent, implying a more flexible Fe-N-O unit as compared to Fe-C-O. This conclusion is supported by the ultrahighresolution structure of NP4NO, in which the smeared-out electron density of the ligand oxygen indicates that the NO bond angle is easily deformable in the plane of the proximal histidine (Roberts et al., 2001). Interestingly, the A1922 band area does not show this temperature effect, possibly because water molecules in the open pocket hinder NO fluctuations. Stretching bands of photodissociated NO in the protein are observed at 1858, 1864, and 1868 cm1 (see Fig. 18.9). After slow cooling under illumination from 160 to 3 K (dotted line), the photoproduct band B1864 has markedly gained intensity, while the fraction of NO ligands at photoproduct site B1858 has decreased slightly. In general, the three photoproduct bands of NP4NO could arise from different orientations of NO within one
372
Karin Nienhaus and G. Ulrich Nienhaus
site, from different sites within a single conformation, or from different docking sites in different protein conformations. From the FTIR spectrum collected after illumination at higher temperature (see Fig. 18.9, symbols, see later), we can assign B1864 to a photoproduct state of NP4 molecules in the A1922 conformation. B1858 and B1868 are spectrally and kinetically distinguishable intermediate states of the more hydrophobic NP4 conformation A1908. The ratio of integrated absorbances of NO in the bound and photolyzed states is 66, which is in between the values of 150 and 24 for the primary and secondary photoproduct states of MbIIINO, respectively. Temperature derivative spectroscopy experiments with a 20-s photolysis at 3 K reveal significant recombination of NO already at the lowest temperature; a local maximum appears at 13 K (Fig. 18.10A). Evidently, ligands rebind from the vicinity of the active site by overcoming a very low enthalpy barrier against recombination. The X-ray structure of NP4 at 100 K in the presence of xenon revealed two xenon-binding cavities (Nienhaus et al., 2004). To explore if these cavities function as transient ligand storage sites, we performed TDS experiments with slow cooling under light from 160 K. With this treatment, we were able to observe rebinding up to 70 K. However, the fraction of NO ligands that rebind at higher temperatures after extended illumination is extremely small, as can be inferred from the integrated absorbance differences plotted in Fig. 18.10C. 1950
A
B
Wavenumber (cm−1)
1930 1910 1890 1870 1880 1870 1860 1850
Absorption
1840 C
0.06 0.04 0.02 0.00
0
10
20 30 40 50 Temperature (K)
60
Figure 18.10 TDS contour maps of NP4NO after a 20-s illumination at 3 K in the spectral region of (A) heme-bound and (B) photolyzed NO. Solid/dotted lines indicate an absorption increase/decrease. (C) Integrated absorption changes of A substate bands of NP4NO after a 20-s illumination at 3 K (closed symbols) and slow cooling from 160 to 3 K (open symbols).
NO-Ligated Heme Proteins
373
Therefore, we conclude from these FTIR-TDS experiments that the Xe cavities are not significantly accessed by photodissociated ligands at low temperature (Nienhaus et al., 2004). Rather remarkable, however, is the finding that illumination at 185 K results in essentially complete photolysis of the NO ligands from the open A1922 conformation, whereas minimal photolysis is observed for the closed A1908 conformation (see Fig. 18.9, open symbols). Incidentally, for CO, the photolysis yield is also rather small (10%). The temperature of 185 K is above the glass transition of the glycerol-water cryosolvent so that ligands can migrate into the solvent, and the protein can change its conformation in response to ligand exit. After dissociation from the hydrophilic, open A1922 conformation, the NO ligands escape into the solvent, and the ferric heme is subsequently blocked by a water ligand, which prevents an immediate reassociation. In contrast, in the hydrophobic, closed A1908 conformation, the AB and GH loops are tightly wrapped around the ligand so that it is efficiently recaptured after photodissociation (Andersen et al., 2000). Our FTIR-TDS experiments support the following dynamic model, which explains the function of NP4 by pH-controlled switching between the two conformations. Under physiological conditions, NO dissociates thermally. At pH 5, the approximate pH of the insect saliva, the ‘‘closed’’ conformation is stabilized. Ligand escape from the active site is prohibited by an extremely reactive heme iron and the absence of additional ligand-docking sites and escape routes. At pH 7, the approximate pH of the victim’s tissue, the loops open up and facilitate NO escape into the blood, assisted by the concurrent blockage of the active site by water and by histamine, thereby suppressing an inflammatory response of the damaged tissue.
5. Concluding Remarks Fourier transform infrared spectroscopy is a powerful technique for the exploration of protein structure and dynamics. Photolysis difference spectroscopy studies at cryogenic temperatures in combination with sophisticated illumination and data acquisition temperature protocols can provide precise quantitative data on protein–ligand interactions, which will be invaluable for theorists developing quantum-chemical descriptions of protein–ligand interactions (Strickland and Harvey, 2007). Detailed insights have been obtained with a variety of CO-ligated heme proteins. With Mb and NP4 as examples, this chapter has shown that these techniques can also be applied to problems of NO binding. NO is clearly the physiologically more important ligand, and future studies along these lines are likely to contribute to a better understanding of functional processes in which this ligand is involved.
374
Karin Nienhaus and G. Ulrich Nienhaus
ACKNOWLEDGMENTS It is a pleasure to acknowledge enjoyable collaborations with J. S. Olson (Rice University, Houston, TX) and W. R. Montfort (University of Arizona, Tucson, AZ). We thank Pasquale Palladino for technical assistance. This work was supported by the Deutsche Forschungsgemeinschaft (Grant Ni291/3) and the Fonds der Chemischen Industrie.
REFERENCES Alben, J. O., Beece, D., Bowne, S. F., Doster, W., Eisenstein, L., Frauenfelder, H., Good, D., McDonald, J. D., Marden, M. C., Moh, P. P., Reinisch, L., Reynolds, A. H., Shyamsunder, E., and Yue, K. T. (1982). Infrared spectroscopy of photodissociated carboxymyoglobin at low temperatures. Proc. Natl. Acad. Sci. USA 79, 3744–3748. Alben, J. O., Beece, D., Bowne, S. F., Eisenstein, L., Frauenfelder, H., Good, D., Marden, M., Moh, P. P., Reinisch, L., Reynolds, A. H., and Yue, K. T. (1980). Isotope effect in molecular tunneling. Phys. Rev. Lett. 44, 1157–1160. Andersen, J. F., Ding, X. D., Balfour, C., Shokhireva, T. K., Champagne, D. E., Walker, F. A., and Montfort, W. R. (2000). Kinetics and equilibria in ligand binding by nitrophorins 1-4: Evidence for stabilization of a nitric oxide-ferriheme complex through a ligand-induced conformational trap. Biochemistry 39, 10118–10131. Andersen, J. F., Weichsel, A., Balfour, C. A., Champagne, D. E., and Montfort, W. R. (1998). The crystal structure of nitrophorin 4 at 1.5 A˚ resolution: Transport of nitric oxide by a lipocalin-based heme protein. Structure 6, 1315–1327. Anderson, J. L., and Chapman, S. K. (2005). Ligand probes for heme proteins. Dalton Trans. 3–24. Ansari, A., Berendzen, J., Braunstein, D., Cowen, B. R., Frauenfelder, H., Hong, M. K., Iben, I. E., Johnson, J. B., Ormos, P., Sauke, T. B., et al. (1987). Rebinding and relaxation in the myoglobin pocket. Biophys. Chem. 26, 337–355. Antonini, E., and Brunori, M. (1970). ‘‘Hemoglobin and Myoglobin in Their Reactions with Ligands.’’ Amsterdam: North-Holland, Amsterdam. Berendzen, J., and Braunstein, D. (1990). Temperature-derivative spectroscopy: A tool for protein dynamics. Proc. Natl. Acad. Sci. USA 87, 1–5. Braunstein, D. P., Chu, K., Egeberg, K. D., Frauenfelder, H., Mourant, J. R., Nienhaus, G. U., Ormos, P., Sligar, S. G., Springer, B. A., and Young, R. D. (1993). Ligand binding to heme proteins. III. FTIR studies of His-E7 and Val-E11 mutants of carbonmonoxymyoglobin. Biophys. J. 65, 2447–2454. Bredenbeck, J., Helbing, J., Nienhaus, K., Nienhaus, G. U., and Hamm, P. (2007). Protein ligand migration mapped by nonequilibrium 2D-IR exchange spectroscopy. Proc. Natl. Acad. Sci. USA 104, 14243–14248. Carpenter, J. F., and Crowe, J. H. (1988). The mechanism of cryoprotection of proteins by solutes. Cryobiology 25, 244–255. Chapman, S. K., Daff, S., and Munro, A. W. (1997). Heme: The most versatile redox centre in biology? In ‘‘Structure and Bonding’’ (P. J. Sadler, ed.), Vol. 88, pp. 39–70. SpringerVerlag, Berlin. Chu, K., Ernst, R. M., Frauenfelder, H., Mourant, J. R., Nienhaus, G. U., and Philipp, R. (1995). Light-induced and thermal relaxation in a protein. Phys. Rev. Lett. 74, 2607–2610. Cooper, C. E. (1999). Nitric oxide and iron proteins. Biochim. Biophys. Acta 1411, 290–309.
NO-Ligated Heme Proteins
375
Coyle, C. M., Vogel, K. M., Rush, T. S., 3rd, Kozlowski, P. M., Williams, R., Spiro, T. G., Dou, Y., Ikeda-Saito, M., Olson, J. S., and Zgierski, M. Z. (2003). FeNO structure in distal pocket mutants of myoglobin based on resonance raman spectroscopy. Biochemistry 42, 4896–4903. Dong, A., and Caughey, W. S. (1994). Infrared methods for study of hemoglobin reactions and structures. Methods Enzymol. 232, 139–175. Douzou, P. (1977). ‘‘Cryobiochemistry.’’ New York: Academic Press, New York. Frauenfelder, H., Nienhaus, G. U., and Johnson, J. B. (1991a). Rate processes in proteins. Ber. Bunsenges. Phys. Chem. 95, 272–278. Frauenfelder, H., Sligar, S. G., and Wolynes, P. G. (1991b). The energy landscapes and motions of proteins. Science 254, 1598–1603. Fuchsman, W. H., and Appleby, C. A. (1979). CO and O2 complexes of soybean leghemoglobins: pH effects upon infrared and visible spectra. Comparisons with CO and O2 complexes of myoglobin and hemoglobin. Biochemistry 18, 1309–1321. Gerwert, K. (1993). Molecular recognition mechanisms of proteins as monitored by timeresolved FTIR spectroscopy. Curr. Opin. Struct. Biol. 3, 769–773. Gilles-Gonzalez, M. A., and Gonzalez, G. (2005). Heme-based sensors: Defining characteristics, recent developments, and regulatory hypotheses. J. Inorg. Biochem. 99, 1–22. Halliwell, B. (2006). Oxidative stress and neurodegeneration:Where are we now? J. Neurochem. 97, 1634–1658. Hartmann, H., Zinser, S., Komninos, P., Schneider, R. T., Nienhaus, G. U., and Parak, F. (1996). X-ray structure determination of a metastable state of carbonmonoxy myoglobin after photodissociation. Proc. Natl. Acad. Sci. USA 93, 7013–7016. Hoshino, M., Maeda, M., Konishi, R., Seki, H., and Ford, P. C. (1996). Studies on the reaction mechanism for reductive nitrosylation of ferrihemoproteins in buffer solutions. J. Am. Chem. Soc. 118, 5702–5707. Huang, J., Ridsdale, A., Wang, J., and Friedman, J. M. (1997). Kinetic hole burning, hole filling, and conformational relaxation in heme proteins: Direct evidence for the functional significance of a hierarchy of dynamical processes. Biochemistry 36, 14353–14365. Huie, R. E., and Padmaja, S. (1993). The reaction of NO with superoxide. Free Radic. Res. Commun. 18, 195–199. Johnson, J. B., Lamb, D. C., Frauenfelder, H., Mu¨ller, J. D., McMahon, B., Nienhaus, G. U., and Young, R. D. (1996). Ligand binding to heme proteins. VI. Interconversion of taxonomic substates in carbonmonoxymyoglobin. Biophys. J. 71, 1563–1573. Ko¨tting, C., and Gerwert, K. (2005). Monitoring protein-ligand interactions by time-resolved ftir difference spectroscopy. In ‘‘Methods in Molecular Biology’’ (G. U. Nienhaus, ed.), Vol. 305, pp. 261–286. Humana Press, Totowa, NJ. Kriegl, J. M., Nienhaus, K., Deng, P., Fuchs, J., and Nienhaus, G. U. (2003). Ligand dynamics in a protein internal cavity. Proc. Natl. Acad. Sci. USA 100, 7069–7074. Kushkuley, B., and Stavrov, S. S. (1996). Theoretical study of the distal-side steric and electrostatic effects on the vibrational characteristics of the FeCO unit of the carbonylheme proteins and their models. Biophys. J. 70, 1214–1229. Kushkuley, B., and Stavrov, S. S. (1997). Theoretical study of the electrostatic and steric effects on the spectroscopic characteristics of the metal-ligand unit of heme proteins. 2. C-O vibrational frequencies, 17O isotropic chemical shifts, and nuclear quadrupole coupling constants. Biophys. J. 72, 899–912. Lamb, D. C., Nienhaus, K., Arcovito, A., Draghi, F., Miele, A. E., Brunori, M., and Nienhaus, G. U. (2002). Structural dynamics of myoglobin: Ligand migration among protein cavities studied by fourier transform infrared/temperature derivative spectroscopy. J. Biol. Chem. 277, 11636–11644.
376
Karin Nienhaus and G. Ulrich Nienhaus
Lamb, D. C., Prusakov, V., Engler, N., Ostermann, A., Schellenberg, P., Parak, F. G., and Nienhaus, G. U. (1998). Photodissociation and rebinding of H2O to ferrous sperm whale myoglobin. J. Am. Chem. Soc. 120, 2981–2982. Lehle, H., Kriegl, J. M., Nienhaus, K., Deng, P., Fengler, S., and Nienhaus, G. U. (2005). Probing electric fields in protein cavities by using the vibrational Stark effect of carbon monoxide. Biophys. J. 88, 1978–1990. Li, T., Quillin, M. L., Phillips, G. N., Jr., and Olson, J. S. (1994). Structural determinants of the stretching frequency of CO bound to myoglobin. Biochemistry 33, 1433–1446. Lim, M., Jackson, T. A., and Anfinrud, P. A. (1997). Ultrafast rotation and trapping of carbon monoxide dissociated from myoglobin. Nat. Struct. Biol. 4, 209–214. Lim, M. D., Lorkovic, I. M., and Ford, P. C. (2005). The preparation of anaerobic nitric oxide solutions for the study of heme model systems in aqueous and nonaqueous media: Some consequences of NO x impurities. Methods Enzymol. 396, 3–17. Lipton, S. A., Singel, D. J., and Stamler, J. S. (1994). Nitric oxide in the central nervous system. Prog. Brain Res. 103, 359–364. Ma¨ntele, W. (1993). Reaction-induced infrared difference spectroscopy for the study of protein function and reaction mechanisms. Trends Biochem. Sci. 18, 197–202. Marriott, G. (ed.). (1998). Caged compounds. In ‘‘Methods Enzymol.’’ 291, 1–529 Acedemic Press, San Diego. Meuwly, M. (2006). On the influence of the local environment on the CO stretching frequencies in native myoglobin: Assignment of the B-states in MbCO. Chemphyschem 7, 2061–2063. Miller, L. M., Pedraza, A. J., and Chance, M. R. (1997). Identification of conformational substates involved in nitric oxide binding to ferric and ferrous myoglobin through difference fourier transform infrared spectroscopy (FTIR). Biochemistry 36, 12199–12207. Montfort, W. R., Weichsel, A., and Andersen, J. F. (2000). Nitrophorins and related antihemostatic lipocalins from Rhodnius prolixus and other blood-sucking arthropods. Biochim. Biophys. Acta 1482, 110–118. Mourant, J. R., Braunstein, D. P., Chu, K., Frauenfelder, H., Nienhaus, G. U., Ormos, P., and Young, R. D. (1993). Ligand binding to heme proteins. II. Transitions in the heme pocket of myoglobin. Biophys. J. 65, 1496–1507. Mu¨ller, J. D., McMahon, B. H., Chien, E. Y., Sligar, S. G., and Nienhaus, G. U. (1999). Connection between the taxonomic substates and protonation of histidines 64 and 97 in carbonmonoxy myoglobin. Biophys. J. 77, 1036–1051. Nienhaus, G. U., Chu, K., and Jesse, K. (1998). Structural heterogeneity and ligand binding in carbonmonoxy myoglobin crystals at cryogenic temperatures. Biochemistry 37, 6819–6823. Nienhaus, G. U., Mourant, J. R., Chu, K., and Frauenfelder, H. (1994). Ligand binding to heme proteins: The effect of light on ligand binding in myoglobin. Biochemistry 33, 13413–13430. Nienhaus, G. U., Mourant, J. R., and Frauenfelder, H. (1992). Spectroscopic evidence for conformational relaxation in myoglobin. Proc. Natl. Acad. Sci. USA 89, 2902–2906. Nienhaus, G. U., and Nienhaus, K. (2002). Infrared study of carbon monoxide migration among internal cavities of myoglobin mutant L29W. J. Biol. Phys. 28, 163–172. Nienhaus, G. U., and Young, R. D. (1996). Protein dynamics. In ‘‘Encyclopedia of Applied Physics’’ (G. L. Trigg, ed.), pp. 163–184. VCH Publishers, New York. Nienhaus, K., Deng, P., Kriegl, J. M., and Nienhaus, G. U. (2003a). Structural dynamics of myoglobin: Spectroscopic and structural characterization of ligand docking sites in myoglobin mutant L29W. Biochemistry 42, 9633–9646. Nienhaus, K., Deng, P., Kriegl, J. M., and Nienhaus, G. U. (2003b). Structural dynamics of myoglobin: The effect of internal cavities on ligand migration and binding. Biochemistry 42, 9647–9658.
NO-Ligated Heme Proteins
377
Nienhaus, K., Deng, P., Olson, J. S., Warren, J. J., and Nienhaus, G. U. (2003c). Structural dynamics of myoglobin: Ligand migration and binding in valine 68 mutants. J. Biol. Chem. 278, 42532–42544. Nienhaus, K., Lamb, D. C., Deng, P., and Nienhaus, G. U. (2002). The effect of ligand dynamics on heme electronic transition band III in myoglobin. Biophys. J. 82, 1059–1067. Nienhaus, K., Maes, E. M., Weichsel, A., Montfort, W. R., and Nienhaus, G. U. (2004). Structural dynamics controls nitric oxide affinity in nitrophorin 4. J. Biol. Chem. 279, 39401–39407. Nienhaus, K., and Nienhaus, G. U. (2005). Probing heme protein-ligand interactions by UV/visible absorption spectroscopy. In ‘‘Methods Molecular Biology’’ (G. U. Nienhaus, ed.), Vol. 305, pp. 215–242. Nienhaus, K., Olson, J. S., Franzen, S., and Nienhaus, G. U. (2005). The origin of Stark splitting in the initial photoproduct state of MbCO. J. Am. Chem. Soc. 127, 40–41. Nutt, D. R., Karplus, M., and Meuwly, M. (2005). Potential energy surface and molecular dynamics of MbNO: Existence of an unsuspected FeON minimum. J. Phys. Chem. B 109, 21118–21125. Nutt, D. R., and Meuwly, M. (2007). Ferric and ferrous iron in nitroso-myoglobin: Computer simulations of stable and metastable states and their infrared spectra. Chemphyschem 8, 527–536. Olson, J. S., and Phillips, G. N., Jr. (1996). Kinetic pathways and barriers for ligand binding to myoglobin. J. Biol. Chem. 271, 17593–17596. Ormos, P., Szaraz, S., Cupane, A., and Nienhaus, G. U. (1998). Structural factors controlling ligand binding to myoglobin: A kinetic hole-burning study. Proc. Natl. Acad. Sci. USA 95, 6762–6767. Ostermann, A., Waschipky, R., Parak, F. G., and Nienhaus, G. U. (2000). Ligand binding and conformational motions in myoglobin. Nature 404, 205–208. Pacher, P., Beckman, J. S., and Liaudet, L. (2007). Nitric oxide and peroxynitrite in health and disease. Physiol. Rev. 87, 315–424. Packer, L. (ed.). (1996). Nitric oxide, part B, physiological and pathological processes. In ‘‘Methods in enzymology’’, 268, pp. 1–555. Acedemic Press, San Diego. Paiva, C. L., and Panek, A. D. (1996). Biotechnological applications of the disaccharide trehalose. Biotechnol. Annu. Rev. 2, 293–314. Parak, F., Heidemeier, J., and Nienhaus, G. U. (1988). Protein structural dynamics as determined by Mo¨ssbauer-spectroscopy. Hyperfine Interact. 40, 147–157. Potter, W. T., Hazzard, J. H., Choc, M. G., Tucker, M. P., and Caughey, W. S. (1990). Infrared spectra of carbonyl hemoglobins: Characterization of dynamic heme pocket conformers. Biochemistry 29, 6283–6295. Quillin, M. L., Arduini, R. M., Olson, J. S., and Phillips, G. N., Jr. (1993). High-resolution crystal structures of distal histidine mutants of sperm whale myoglobin. J. Mol. Biol. 234, 140–155. Roberts, S. A., Weichsel, A., Qiu, Y., Shelnutt, J. A., Walker, F. A., and Montfort, W. R. (2001). Ligand-induced heme ruffling and bent NO geometry in ultra-high-resolution structures of nitrophorin 4. Biochemistry 40, 11327–11337. Schlichting, I., Berendzen, J., Phillips, G. N., Jr., and Sweet, R. M. (1994). Crystal structure of photolysed carbonmonoxy-myoglobin. Nature 371, 808–812. Scott, E. E., and Gibson, Q. H. (1997). Ligand migration in sperm whale myoglobin. Biochemistry 36, 11909–11917. Scott, E. E., Gibson, Q. H., and Olson, J. S. (2001). Mapping the pathways for O2 entry into and exit from myoglobin. J. Biol. Chem. 276, 5177–5188.
378
Karin Nienhaus and G. Ulrich Nienhaus
Shimada, H., and Caughey, W. S. (1982). Dynamic protein structures: Effects of pH on conformer stabilities at the ligand-binding site of bovine heart myoglobin carbonyl. J. Biol. Chem. 257, 11893–11900. Stamler, J. S., and Meissner, G. (2001). Physiology of nitric oxide in skeletal muscle. Physiol. Rev. 81, 209–237. Stamler, J. S., Singel, D. J., and Loscalzo, J. (1992). Biochemistry of nitric oxide and its redox-activated forms. Science 258, 1898–1902. Steinbach, P. J., Ansari, A., Berendzen, J., Braunstein, D., Chu, K., Cowen, B. R., Ehrenstein, D., Frauenfelder, H., Johnson, J. B., Lamb, D. C., Luck, S., Mourant, J. R., Nienhaus, G. U., Ormos, P., Philipp, R., Xie, A., and Young, R. D. (1991). Ligand binding to heme proteins: Connection between dynamics and function. Biochemistry 30, 3988–4001. Strickland, N., and Harvey, J. N. (2007). Spin-forbidden ligand binding to the ferrous-heme group: Ab initio and DFT studies. J. Phys. Chem. B 111, 841–852. Teng, T. Y., Srajer, V., and Moffat, K. (1994). Photolysis-induced structural changes in single crystals of carbonmonoxy myoglobin at 40 K. Nat. Struct. Biol. 1, 701–705. Tetreau, C., Novikov, E., Tourbez, M., and Lavalette, D. (2002). Kinetic evidence for three photolyzable taxonomic conformational substates in oxymyoglobin. Biophys. J. 82, 2148–2155. Tilton, R. F., Jr., Kuntz, I. D., Jr., and Petsko, G. A. (1984). Cavities in proteins: Structure of a metmyoglobin-xenon complex solved to 1.9 A˚. Biochemistry 23, 2849–2857. Vojtechovsky, J., Chu, K., Berendzen, J., Sweet, R. M., and Schlichting, I. (1999). Crystal structures of myoglobin-ligand complexes at near-atomic resolution. Biophys. J. 77, 2153–2174. Yang, F., and Phillips, G. N., Jr. (1996). Crystal structures of CO-, deoxy- and metmyoglobins at various pH values. J. Mol. Biol. 256, 762–774.
C H A P T E R
N I N E T E E N
Time-Resolved X-Ray Crystallography of Heme Proteins Vukica Sˇrajer* and William E. Royer, Jr.† Contents 379 381 385 388 391 393 393
1. Introduction 2. Experiment 3. Data Processing and Analysis 4. A Case Study: Scapharca Dimeric Hemoglobin 5. Conclusions Acknowledgments References
Abstract Heme proteins, with their natural photosensitivity, are excellent systems for the application of time-resolved crystallographic methods. Ligand dissociation can be readily initiated by a short laser pulse with global structural changes probed at the atomic level by X-rays in real time. Third-generation synchrotrons provide 100-ps X-ray pulses of sufficient intensity for monitoring very fast processes. Successful application of such time-resolved crystallographic experiments requires that the structural changes being monitored are compatible with the crystal lattice. These techniques have recently permitted observing for the first time allosteric transitions in real time for a cooperative dimeric hemoglobin.
1. Introduction Time-resolved X-ray diffraction is a unique tool for investigation of real-time structural changes that molecules undergo while performing their function. While static structures providing atomic models, now available for many macromolecules, greatly aid our understanding of molecular function, the detailed mechanism by which macromolecules function still often * {
Center for Advanced Radiation Sources, The University of Chicago, Chicago, Illinois Department of Biochemistry and Molecular Pharmacology, University of Massachusetts Medical School, Worcester, Massachusetts
Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37019-5
#
2008 Elsevier Inc. All rights reserved.
379
380
Vukica Sˇrajer and William E. Royer
cannot be discerned. Time-resolved crystallography can elucidate the mechanism by providing snapshots of molecules in action. Short and intense X-ray pulses at third-generation synchrotron sources are used to probe, in real time, reactions that are initiated synchronously and rapidly in molecules within the crystal. Such time-resolved studies aim to identify reaction intermediates that are often very short-lived, determine their structures, and describe the complete reaction mechanism, including reaction rates that govern the concentrations of intermediates during the reaction (Schmidt et al., 2005a). As an alternative to time-resolved crystallography, various physical or chemical trapping methods are often applied to extend the lifetimes of reaction intermediates sufficiently so that they can be studied by the more conventional and less technically challenging static crystallography. Examples of such methods are lowering temperature, trapping of intermediates by freezing, pH, or solvent modifications, and chemical modifications of the macromolecule (including mutations), substrate, or cofactor. Although trapping methods provide valuable insight into structures of intermediates, they at the same time perturb the reaction, requiring consequences of the perturbation to be evaluated. In addition to time-resolved crystallography, real-time information on structural changes for macromolecules containing a chromophore, such as heme proteins, is also provided by time-resolved spectroscopic studies, including absorption, resonance Raman, and infrared spectroscopy. Such techniques offer great sensitivity to small structural changes, but most are limited to the chromophore environment. In addition, structural changes are not directly observed and the measured spectral changes must be interpreted in terms of underlying structural change. Time-resolved crystallography therefore has a distinct advantage as it provides direct and global structural information, for the entire molecule, in atomic detail as a function of time. As such, it is ideally suited for following the propagation of structural changes throughout the protein from the active site (e.g., heme in heme proteins), for exploring the events involved in an allosteric mechanism, and for tracking the ligands on their migration pathways throughout a protein. As in any other type of time-resolved measurement, successful identification of an intermediate requires that the fraction of molecules in the intermediate accumulates, at some time following the start of the reaction, to a level significant enough to be detected by a measurement with an appropriate time resolution. The magnitude of structural change between the known initial state and the intermediate also has to be large enough to be detectable, but small enough to be accommodated by the crystal lattice. As many macromolecules are active and perform their function in the crystalline state, the structural changes involved in these cases are clearly compatible with the crystal lattice. However, a key question is if the changes are large enough to be reliably detected. Another important issue for time-resolved studies is that at any given time during the reaction several intermediates
Time-Resolved X-Ray Crystallography
381
are likely to be present in the crystal, unless intermediates are very well separated in time by vastly different lifetimes. In other words, there may not be a time window where only one intermediate is predominant. It is therefore very important to have a method of data analysis that will facilitate extracting the structures of individual intermediates from the measured mixture of states as a function of time. During the last two decades, essential advancements have been made in the development of both high-intensity third-generation synchrotron X-ray sources and necessary time-resolved diffraction instrumentation, as well as in methodology and software for data processing and analysis (Bourgeois et al., 1996; Rajagopal et al., 2004b; Ren et al., 1999; Schmidt et al., 2003). The technique is in its mature phase today, where detection of structural changes as small as 0.2–0.3 A˚ with a time resolution of 100 ps is possible, even when the reaction is initiated in only 10–20% of molecules in the crystal (Anderson et al., 2004; Bourgeois et al., 2006, 2003; Ren et al., 2001; Schmidt et al., 2005b; Schotte et al., 2003, 2004; Sˇrajer et al., 1996, 2001). In addition, extraction of time-independent structural intermediates from measured time-dependent structure factor amplitudes has been demonstrated successfully (Ihee et al., 2005; Rajagopal et al., 2004b, 2005; Schmidt et al., 2003, 2004). This chapter discusses some general aspects of time-resolved crystallography applicable to all systems, but focuses on the technique as applied to heme proteins. The natural photosensitivity of the heme-CO bond, which was first described over a hundred years ago (Haldane and Smith, 1896), permits rapid triggering of ligand release in hemoglobins and myoglobins, rendering them particularly amenable to time-resolved crystallographic experiments. As another contribution in this volume will discuss methods as applied to myoglobin, we restrict our coverage to hemoglobins here.
2. Experiment Structural changes in macromolecules at room temperature span many orders of magnitude in time, from femtoseconds to seconds and longer. To examine short-lived intermediates in real time, it is critical to initiate the reaction in all molecules very rapidly, in a time period that is significantly shorter than the lifetime of such intermediates. The fastest method of reaction initiation by far is the use of ultrashort laser pulses. This method is readily applicable to molecules that are inherently photosensitive and undergo a reversible reaction, such as ligand photodissociation and rebinding in heme proteins. Alternatively, other molecules can be rendered light sensitive by chemically attaching photosensitive groups to substrates, cofactors, or important protein residues. The goal of such ‘‘caging’’ is to make the
382
Vukica Sˇrajer and William E. Royer
molecule inert until light is absorbed by the photosensitive group, which in turn triggers the reaction. In this case the reaction will be irreversible. Also, reaction initiation in such systems is typically relatively slow (microseconds to milliseconds) as it is governed by the release of the caged group and often involves diffusion of the released group into the active site. Time-resolved X-ray diffraction experiments are of a pump-probe type: laser pulses are used as ‘‘pump’’ pulses, to trigger the reaction rapidly, whereas X-ray pulses are used to ‘‘probe’’ the reaction at various time delays following the pump pulse. Time resolution of the experiment is determined by the duration of the pump or probe pulses, whichever is longer. A critical requirement for the X-ray source in such experiments is a very high X-ray flux in short pulses, typically available at third-generation synchrotron sources, such as Advanced Photon Source (Argonne National Laboratory), European Synchrotron Radiation Facility (Grenoble, France), and SPring 8 ( Japan). As synchrotron radiation consists of a continuous 100-ps pulse train, for the best time resolution one needs to isolate a single 100-ps pulse from such a train and synchronize its arrival at the sample with the laser pump pulse. For slower reactions with a less demanding requirement on time resolution, an X-ray pulse train of a longer total duration can be used as a probe pulse to increase the probe pulse intensity and therefore provide an improved signal-to-noise ratio for recorded diffraction images. Mechanical crystal rotations used in standard oscillation crystallography are much too slow to probe fast, subsecond reactions. Instead, Laue diffraction of stationary crystals is used. Laue diffraction requires significantly wider energy bandwidth than the standard monochromatic diffraction. It has been shown that undulators are the best X-ray sources for timeresolved Laue diffraction experiments (Bourgeois et al., 2000; Ren et al., 1999; Sˇrajer et al., 2000). The typical band pass of undulators used today for time-resolved experiments is 3–5% at 12–15 keV. Softer, lower energy X-rays increase radiation damage, whereas harder, higher energy X-rays are diffracted and detected less efficiently. Undulators, as compared to other synchrotron sources with wider energy bandwidth such as bending magnets or wigglers, have numerous advantages. They have higher peak intensity and, because of the narrow bandwidth, result in lower polychromatic background as well as reduced spatial and harmonic overlap in typically crowded Laue diffraction patterns. Overall data quality obtained from undulator sources is superior as judged by all important indicators, such as Rmerge, data completeness, and map quality. When a single 100-ps X-ray pulse is used to record a diffraction image, a flux greater than 1010 photons/pulse is needed, focused to match the crystal size. If such high-flux and well-focused X-ray pulses are not available, the pump-probe sequence has to be repeated numerous times and diffraction data accumulated at the detector prior to the readout of a diffraction image. Such measurements with a repeated pump-probe sequence are clearly
383
Time-Resolved X-Ray Crystallography
possible only for fully reversible systems. However, even for such systems, it helps to reduce the number of pump (laser) pulses as they are potentially damaging to the crystal. Single X-ray pulse data acquisition is essential for irreversible systems, where each image requires a single pump-probe sequence and a new crystal. A critical part of time-resolved experiments is an X-ray shutter train that isolates single X-ray pulses or longer pulse trains from the continuous stream of synchrotron pulses (Bourgeois et al., 1996). Typically, a fast rotating chopper is used in series with a slower, single opening shutter that isolates a single opening of the chopper. For example, at the BioCARS facility, located at Sector 14 at the Advanced Photon Source (APS) at Argonne National Laboratory, an ultrafast chopper (Forschungszentrum Ju¨lich, Germany) is capable of isolating X-ray pulses that are 153.4 ns apart during the standard, 24-bunch mode of the APS storage ring (Fig. 19.1). The ability to utilize this standard rather than a special operating mode of the storage ring is highly beneficial as significantly more beam time becomes available for conducting these technically challenging experiments. Crystals for time-resolved crystallography are typically mounted in thinwalled glass capillaries as data collection is conducted at or near room temperature. Protein crystals contain a very large number of molecules, on the order of 1013–1014, and therefore typically have high optical density (OD)
X-ray pulses 24-bunch mode APS
153.4 ns
Ultra-fast X-ray Chopper ~1 KHz
< 200 ns
ms X-ray shutter
< 2 ms
Laser-X-ray pulse delay: Δt
100 ps X-ray pulse duration
~1 ms
Δt
Figure 19.1 Schematic diagram of timing for time-resolved experiments at BioCARS beam line 14-ID, APS. An ultrafast chopper with an opening shorter than 200 ns, along with a slower millisecond shutter, permits isolation of a single100-ps X-ray pulse during the standard, 24-bunch, mode of the APS storage ring. Time-resolved experiments are carried out using a laser pulse synchronized with that X-ray pulse, with an adjustable time delay, △t, between the pulses.
384
Vukica Sˇrajer and William E. Royer
in regions of the absorption spectrum where the chromophore absorbs significantly. In order to photo-initiate the reaction by laser pulses throughout the crystal, smaller (thinner) crystals have to be used and the laser wavelength tuned to a spectrum region where OD << 1. For nanosecond time-resolved experiments, where high laser pulse energies are typically available, one can afford 1–2 mJ/pulse, defocused to a size larger than the crystals and at the wavelength where the crystal OD is 0.1–0.2. Both low OD and large laser beam size enhance the uniform photo-initiation throughout the crystal. For the same reason, it is also advantageous to illuminate the crystal by laser light from the opposite sides (Knapp et al., 2006). With shorter, subnanosecond laser pulses, typical pulse energies are 10–100 mJ/pulse. With the number of available photons just barely matching the number of molecules in smaller crystals in this case, all photons need to be absorbed. The laser beam has to be focused to match the size of the crystal. Typically a small X-ray beam is used to probe only the laserilluminated surface layer of the crystal (Schotte et al., 2004). This way only the volume of the crystal that is penetrated by the laser light is probed by X-rays. When crystals are exposed to laser pulses, both increased crystal mosaicity and crystal motion are often observed, as evident by elongation of the diffraction spots. The crystal motion is believed to be heating related. To reduce the motion, a sufficient wait time between pump-probe cycles and the minimal number of such cycles for each diffraction image should be considered. Also, immobilizing the crystal inside the capillary is highly beneficial, as accomplished successfully for Hb crystals using polyvinyl or epoxy (Knapp et al., 2004). A complete time-resolved data set spans four dimensions: three traditional reciprocal space dimensions and time. Similar to a standard monochromatic data set, a polychromatic Laue data set at each time delay contains images collected at many different crystal orientations to completely sample the reciprocal space. The angular step in crystal orientation for Laue data depends on the band pass of the X-ray source and is typically 2–3 for undulator sources. The number of time points required to characterize a reaction that is being investigated and how the time points are sampled ultimately depend on the number of intermediates and their lifetimes. A good starting point is to collect five points per decade in time, equally spaced in logarithmic time. The time interval between consecutive pump-probe cycles depends both on lifetimes of intermediate states and time needed for dissipation of heat deposited by the laser pulses. It is clearly necessary to allow sufficient time for the recovery of the initial state. A typical repetition rate used for heme proteins is 1–3 Hz. The most straightforward way of collecting time-resolved data is to collect complete laser-on and laser-off data sets for a given time delay between laser and X-ray pulses before moving to the next time delay.
Time-Resolved X-Ray Crystallography
385
Since the signal of interest is the difference in diffraction before and after photo-activation, it is best to collect both data sets on the same crystal and to interleave the laser-on and laser-off images for each crystal orientation. As crystal damage by the X-ray and laser pulses limits the maximum number of images that can be collected from one crystal, typically not all desirable time delays can be collected on the same crystal. As a result, crystal-to-crystal variations, as well as laser intensity fluctuations and drifts, introduce potential systematic errors in the time domain as a consequence of variations in reaction initiation in this scheme of data collection. These effects can have a detrimental influence on the determination of time constants when the entire series of time delays is examined. To minimize such errors across the time domain, more recent experiments employ time delay as the fast variable for time-resolved data collection (Schmidt et al., 2005a). In this procedure, the time delay is scanned for a given crystal orientation, starting with the laser-off image and collecting a series of laser-on images that span the desired time domain. The time scan is then repeated for a new crystal orientation. Because of radiation damage issues, several crystals are normally required for a complete space–time data set. On each crystal, a data set that is partial in reciprocal space but which covers all time delays is collected. Because the goal is to determine accurately the differences between the structure factor amplitudes of the initial dark state, Fdark, and the state at a particular time delay following the laser pulse, Flight(t), such differences DF(t) ¼ Flight(t) – Fdark should be determined first, before they are merged across all crystals to obtain a time series of complete difference data sets. Overall, this data collection strategy constitutes the best method for accurately measuring the differences in structure factor amplitudes as a function of time, DF(t), as they change as a consequence of the formation and decay of intermediate states.
3. Data Processing and Analysis Processing of time-resolved data can be roughly divided into three steps: (1) Laue data processing to derive time-dependent structure factor amplitudes from recorded diffraction images; (2) calculation of timedependent difference electron density maps and analysis of such maps; and (3) determination of structures of intermediate states. Software for Laue data processing has played a critical role in the success of time-resolved crystallography. Several problems specific to the Laue diffraction method had to be addressed and resolved: spatial overlap of diffraction spots in typically crowded Laue diffraction patterns, wavelength normalization, and resolving the harmonic overlaps. The so-called wavelength normalization is needed due to the fact that the intensity of the
386
Vukica Sˇrajer and William E. Royer
polychromatic radiation, the scattering power of the crystal, and detector sensitivity are all wavelength dependent. The same reflection or its symmetry mate may be stimulated by a different wavelength depending on the crystal orientation. In order to merge data, the intensities therefore need to be brought to a common scale. The result of wavelength normalization is a l curve that combines all wavelength-dependent effects in diffraction intensities as recorded at the detector. Harmonic or energy overlap results from reflections that lie on a radial line (starting at the origin) in the reciprocal space. Such reflections are stimulated by different energies but scatter in exactly the same direction and therefore overlap exactly at the detector. These characteristic features of Laue diffraction data, traditionally considered ‘‘problems,’’ have been addressed very successfully by the Laue processing software developed in the mid-1990s. Several excellent Laue processing packages exist today: Precognition (Renz Research Inc), LaueView (Ren and Moffat, 1995), PrOW (Bourgeois, 1999), and Daresbury Laboratory Laue Processing Suite (Arzt et al., 1999; Campbell, 1995). Currently, the main ongoing software efforts center on automation of data processing with minimum user input and intervention to facilitate fast, online Laue data processing. The quality of Laue data today and electron density maps derived from Laue data are comparable to standard monochromatic data. In the next step of time-resolved data analysis, time-dependent difference electron density maps Dr(t) are calculated in the following way. The structure factor (SF) amplitudes of the initial, dark state jFD(hkl)j and a corresponding set of time-dependent structure factor amplitudes jF(hkl,t)j are used to calculate time-dependent difference amplitudes, DF(hkl,t) ¼ jF(hkl,t)j – jFD(hkl)j, for each time point t. Phases, used together with difference amplitudes DF(hkl,t) for calculating difference maps Dr(t), are obtained from the known dark state structural model. In order to reduce experimental and data processing errors, several weighting schemes for difference amplitudes have been proposed when calculating difference maps (Ren et al., 2001; Schmidt et al., 2003; Sˇrajer et al., 2001; Ursby and Bourgeois, 1997). In such maps the features that appear above 3s level are considered significant, where the s value is the root mean square deviation of the difference density from the mean value in the asymmetric unit. Negative electron density features in such maps represent the loss of electrons and therefore indicate areas where atoms have moved from their positions in the dark state following photolysis, whereas positive features represent the gain of electrons and indicate areas where atoms have moved to. When SF amplitudes are given on the absolute scale, difference electron density in selected regions can be integrated to provide information about the total number of electrons displaced from or into a particular volume in space as a function of time. Program Promsk (Schmidt et al., 2005b), for example, can be used to integrate difference density within a specified mask, generated by supplying atomic coordinates and a radius of integration
Time-Resolved X-Ray Crystallography
387
around the coordinates for the structural region of interest. It is sometime beneficial to integrate jDr(t)j rather than Dr(t) values in order to obtain the sum of both positive and negative signals in a particular region. In this case a similar integration of jDr(t)j is needed for a difference map free of signal to estimate the noise contribution and subtract it from the integrated jDr((t))j values for the signal. Time courses of Dr(t) or jDr(t)j provide important information on the formation and decay of structural intermediates, as illustrated in the next section, on the example of time-resolved studies of dimeric Scapharca hemoglobin. Singular value decomposition (SVD) provides a substantially more complete method to address questions of the number of global intermediates, time-independent structures of such intermediates, rates of formation and decay of intermediates, and the overall kinetic mechanism for the process studied by time-resolved crystallography (Ihee et al., 2005; Rajagopal et al., 2004b, 2005; Schmidt et al., 2003, 2004). Both SVD and post-SVD analyses are needed and applied to time-dependent difference electron density maps to obtain information about intermediates and the kinetic mechanism. A comprehensive description of these methods can be found in Schmidt et al. (2005a) In short, SVD is a method of global analysis applied to a data matrix composed of N time-dependent difference electron density maps, Dr(t). The SVD procedure decomposes such a matrix into N time-independent Dr maps or left singular vectors (lSV) and corresponding N time courses or right singular vectors (rSV). Each vector pair has a singular value associated with it, which weights the contribution of the lSVs to the experimental difference maps. The actual difference signal is typically contained in only a few significant vectors, associated with singular values of large amplitude. The remaining singular vectors contain only noise. The SVD analysis therefore provides an effective noise filter, as the input series of maps can be approximated by S/N improved maps Dr0 (t), reconstituted from significant singular vectors only. The number of significant singular values and vectors is related to the number of intermediates in the reaction. As rSV time courses are linear combinations of the timedependent concentrations of intermediates, they provide information on the number of relaxation processes and the associated relaxation rates associated with the time-dependent concentrations. From a global fit of all significant rSVs by a sum of exponential relaxations, with rates common to all rSVs, the number of relaxation processes and associated rates in the reaction are determined. The number of relaxation rates is the lower bound on the number of intermediates. With the number of relaxations and associated rates determined from the rSVs, post-SVD analysis provides a method for determination of timeindependent difference maps, Dri, corresponding to intermediates (Rajagopal et al., 2004b, 2005; Schmidt et al., 2003, 2005a). At this stage of data analysis, assumption of a candidate reaction mechanism is required
388
Vukica Sˇrajer and William E. Royer
for the number of intermediates determined by SVD analysis. For the assumed mechanism, the time-independent difference maps of intermediates are synthesized as a linear combination of significant lSV maps (Schmidt et al., 2005a). The mixture of intermediates in the experimental time-dependent difference maps has therefore been separated into the difference maps corresponding to pure intermediates. In an iterative procedure, plausible reaction mechanisms are examined and the best mechanism is determined by comparing the measured time-dependent difference maps, Dr(t), and difference maps, Drcalc(t), calculated for each candidate mechanism by using corresponding time-independent difference maps, as well as time-dependent concentrations of pure intermediates (Schmidt et al., 2005b). Finally, the structures of reaction intermediates are refined. For this purpose, conventional electron density maps are calculated for each intermediate from the so-called extrapolated SFs (Schmidt et al., 2005a). Extrapolated SFs are obtained by vector summation of calculated SFs for the dark state and difference SFs for a pure intermediate, where difference factors result from the Fourier transform of the difference map of the intermediate. Such a conventional electron density map therefore represents just the intermediate, with no contribution from the dark state. The structures of intermediates are then modeled and refined using these conventional maps.
4. A Case Study: Scapharca Dimeric Hemoglobin The dimeric hemoglobin (HbI) from the blood clam Scapharca inaequivalvis is very well suited for the investigation of allosteric structural transitions by time-resolved crystallography. The allosteric reaction from the high-affinity R state to the low-affinity T state can be triggered by a laser flash, which causes the release of bound CO as in other heme proteins. Importantly, allosteric transition in HbI involves more localized structural changes than some other allosteric proteins; as a result, cooperative oxygen binding has been observed in the crystalline state (Mozzarelli et al., 1996), and full ligand-linked transitions, including the small (3 ) subunit rotation, can occur within crystals grown in the CO-liganded form (Knapp and Royer, 2003). Despite the rather limited structural changes, these transitions have large functional ramifications, with the R state estimated to bind oxygen about 300-fold more tightly than the T state (Royer et al., 1996). Conventional crystallographic analyses complemented by site-directed mutagenesis and ligand-binding experiments have revealed three key structural transitions that contribute to the functional differences between the R and the T states of HbI: movement of Phe 97 (corresponding to position F4 of myoglobin) from the subunit interface to the proximal pocket upon
Time-Resolved X-Ray Crystallography
389
ligand loss (Knapp et al., 2005; Pardanani et al., 1997), redistribution of interface water molecules (Pardanani et al., 1998; Royer et al., 1996), and movement of the heme groups toward the interface upon ligand loss (Knapp et al., 2001). We carried out time-resolved crystallographic experiments of this molecule to elucidate the time-dependent interplay among these functionally important structural transitions (Knapp et al., 2006). Key to the success of these experiments is the ability of HbI crystals to tolerate laser-induced ligand release and reversible ligand rebinding for many (thousands) cycles. Crystals are mounted in thin-walled glass capillaries and immobilized under a poly(vinyl) film (Knapp et al., 2004) in order to minimize crystal movements that had been observed upon laser exposure in preliminary experiments. Because of rapid binding of oxygen compared with CO and oxidation of the heme iron in the presence of oxygen, it is important to remove oxygen from the capillaries; this is accomplished by purging the capillaries with CO before adding crystal stabilizing solution that has been saturated with sodium dithionite. Photodissociation of the bound CO molecule is induced by 7-ns laser pulses (FWHM) from a Nd:YAG pumped dye laser (Continuum). Crystals are simultaneously illuminated from two opposite sides, at 615 nm. Three different time series, each using time delay as the fast variable (see earlier discussion), are employed to elucidate the structural transitions that occur between 5 ns and 80 ms following photolysis. For the first time series, data are collected with time delays between laser and X-ray pulses of 5 ns, 25 ns, 75 ns, 200 ns, 700 ns, and 3 ms. For each of four different crystals, 21 orientations, with a 9 increment between angular settings of the crystal, are used to collect both dark images (no laser illumination) and the six time delays after laser illumination. The two other time series cover time delays between 2 and 80 ms. Results produce less noisy time courses for structural changes, demonstrating that this procedure, in which data are collected at all time delays from the same set of crystals, minimizes systematic errors between different time delays. Difference Fourier maps for the M37V mutant of HbI used for these studies revealed the time-dependent structural changes that follow the photolytic release of ligand. Successful photolysis is evident in the earliest time delay (5 ns) by strong negative density (14 and 17s for subunits A and B, respectively) at the ligand position (Fig. 19.2). Integration of this density indicates roughly 40% photolysis, which decays to about 10% by 1 ms. Results from early time delays (5–200 ns) suggest that structural changes are largely, but not entirely, limited to the ligand-binding site, whereas difference maps at later time points in the microsecond time range show propagation of the structural changes into the subunit interface (see Fig. 19.2A). Therefore, the ligand-linked transition can be usefully divided into two phases: an early intermediate phase during the nanosecond time period and an allosteric phase.
390
Vukica Sˇrajer and William E. Royer
B
A
C E
F
CD
5 ns
F4
CO*
CD1 E7
CO
CD3 F
Fe
F8
F4
F7
F
E CD
F
CD1
F4
E7
CD3
60 ms F
Fe
F8
F4
F7
F Subunit A
Figure 19.2 Difference Fourier maps between Scapharca HbI* (photoproduct) and HbI-CO at time delays of 5 ns and 60 ms. (A) Ribbon diagram of HbI-CO dimer (gray) with side chains for His F8 (cyan), Phe F4 (yellow), and two key interface water molecules (small cyan spheres) are shown along with the difference Fourier map contoured at þ3.5s (blue) and 3.5s (red). Density at 5 ns suggests that early structural changes are largely concentrated at the ligand-binding site, with major interface allosteric structural transitions occurring by 60 ms. An exception is the density for the pair of R-state water molecules shown in cyan (arrows), which show clear negative density by 5 ns. (B) An a-carbon trace (gray) for the CD region and E and F helices along with the heme group (salmon) and key side chains from subunit A. [Map contoured at þ3.5s (blue) and 3.5s (red).] Photolysis is clearly evident by the strong negative density observed at the CO-binding site for both subunits A (14s) and B (17s) at 5 ns. (C) Difference density for the region around Phe F4, with map contoured at þ2.5s (blue) and 2.5s (red). The R-to-T transition of Phe F4 has not started by 5 ns, but is complete by 60 ms. From Knapp etal. (2006) # National Academyof Sciences, USA.
Evident in the difference Fourier maps is the formation of a tertiary intermediate within 5 ns as each R-state subunit responds to the presence of an unliganded heme group. This intermediate is characterized by a buckled ˚ as heme group, with the iron displaced from the heme plane by about 0.4 A is apparent from strong positive peaks in the electron density maps and confirmed by difference refinement (Terwilliger and Berendzen, 1995).
Time-Resolved X-Ray Crystallography
391
Rapid structural changes are also observed in the F helix, particularly around the heme-linked proximal histidine (see Fig. 19.2B). Density features, and difference refinement, indicate that the F helix moves in the direction of the dimeric interface, which results in a disruption of water molecules unique to the R state. Most clearly affected are two R-state water molecules (see Fig. 19.2A) located near the F4 and F7 side chains and also hydrogen bond to heme propionates. It appears that their disordering is needed to facilitate the R-to-T movement of the heme groups toward the interface and thus may lay the foundation for the allosteric transition. Key structural transitions characteristic of the T state commence after 1 ms. These include movement of the heme groups toward the interface, highlighted by the density peaks for the iron atom, movement of the Phe F4 side chain from the interface into the proximal pocket, and accumulation of water molecules at T-state-specific locations in the subunit interface. Integration of the difference Fourier density for the T-state position of Phe F4 at its maximum accumulation (10–30 ms) suggests 8% population of the T-state Phe F4. This matches the deliganded heme population in the same time interval. Phe F4 has therefore switched from the R to T state in all deliganded subunits, suggesting a nearly complete R-to-T transition in those subunits without ligand 10–30 ms following the laser flash. Comparison of the time courses of the structural changes associated with the R-to-T transitions indicates that all three central movements (heme, PheF4, water molecules) follow very similar time courses (Fig. 19.3). Thus, these movements are tightly coupled, suggesting a rather concerted R-to-T transition. The time-resolved crystallographic experiments on Scapharca dimeric hemoglobin reveal an early intermediate whose structural properties suggest that it facilitates the transition between R and T states. The structural transitions to form the T-state molecule are observed to occur in the microsecond time range, with the functionally key structural changes following very similar time courses indicative of a highly coupled transition (Knapp et al., 2006). These experiments demonstrate feasibility of following functionally relevant structural transitions in heme proteins by time-resolved crystallography.
5. Conclusions With significant technical and software developments over the last decade, the time-resolved macromolecular crystallography technique has reached a mature phase, with the demonstrated ability to detect relatively small structural changes even at levels of reaction initiation of only 10–40%. The important development of essential methods for global analysis of timeresolved data, such as SVD and more recently developed cluster analysis
392 4
Electrons
2
CD region E helix F helix G helix H helix
0
1.5 1
−2
0.5 CO CO*
−4
0 D
Phe F4 R Phe F4 T Phe 127 R
0.8 0.6 0.4 0.2 0 −0.2 −0.4 −0.6 −0.8
Core R allosteric T allosteric
Electrons
4 C 4 3 2 1 0 −1 −2 −3
10−8
10−7 10−6 Time (s)
10−5
10−4 10−8
10−7 10−6 Time (s)
10−5
|Electrons|/residue
B 2
Electrons/water molecule
A
Vukica Sˇrajer and William E. Royer
10−4
Figure 19.3 Time-dependent integrated difference electron density values [Dr(t)] following photolysis of Scapharca HbI for several structural regions, with values averaged over subunits A and B. (A) Integrated Dr(t) values for the loss of bound CO (red circles) and the photodissociated CO at the distal pocket docking site (CO*, blue circles). Both features exhibit similar biphasic behavior that suggests an exponential geminate ligand rebinding phase and a bimolecular rebinding phase. (B) Integrated absolute values of difference electron densities, jDr(t)j, for helices E^H and the CD loop region.The F helix and the CD loop region show nanosecond changes corresponding to formation of an early intermediate. Both the E and F helices as well as the CD loop exhibit an increase in signal in the microsecond time range similar to signals associated with the rearrangements of Phe F4 (C) and allosteric water molecules (D). Simultaneous fits of data in B^D by a common exponential function in the microsecond time region are shown as solid lines. The very similar time courses observed suggest a concerted allosteric transition. From Knapp et al. (2006) # National Academy of Sciences, USA.
(Rajagopal et al., 2004a), is also well under way. The technique provides an important tool for insight into functionally important structural relaxation processes, ligand migration, and allosteric action on the atomic level, with 100-ps time resolution at the most intense present-day X-ray sources such as 3rd generation synchrotrons. One of the main challenges for time-resolved crystallography is the application of the technique to a wider range of molecules of biological interest. Most such macromolecules are not inherently photosensitive, and system-specific efforts are needed to determine a suitable reaction initiation method, including methods such as caging that make triggering of the reaction by light possible. Other challenges for the technique involve studies of irreversible processes, further improvements in time resolution to sub-100 ps with new intense X-ray source such as XFELs, and combining experimental results from time-resolved crystallography
Time-Resolved X-Ray Crystallography
393
with computational and theoretical approaches to provide a complete description of reaction pathways, including the transition states.
ACKNOWLEDGMENTS We thank our colleagues, particularly James Knapp and Reinhard Pahl, for their many contributions to our time-resolved work. This work, supported by NIH Grant GM66756 and the BioCARS facility, is supported by NIH Grant RR07707.
REFERENCES Anderson, S., Sˇrajer, V., Pahl, R., Rajagopal, S., Schotte, F., Anfinrud, P. A., Wulff, M., and Moffat, K. (2004). Chromophore conformation and the evolution of tertiary structural changes in photoactive yellow protein. Structure 12, 1039–1045. Arzt, S., Campbell, J. W., Harding, M. M., Hao, Q., and Helliwell, J. R. (1999). LSCALE: The new normalization, scaling and absorption correction program in the Daresbury Laue software suite. J. Appl. Cryst. 32, 554–562. Bourgeois, D. (1999). New processing tools for weak and/or spatially overlapped macromolecular diffraction patterns. Acta Crystallogr. D Biol. Crystallogr. 55, 1733–1741. Bourgeois, D., Ursby, T., Wulff, M., Pradervand, C., Legrand, A., Schildkamp, W., Laboure, S., Srajer, V., Teng, T., Roth, M., and Moffat, K. (1996). Feasibility and realization of single-pulse Laue diffraction on macromolecular crystals at ESRF. J. Synch. Radiat. 3, 65–74. Bourgeois, D., Vallone, B., Arcovito, A., Sciara, G., Schotte, F., Anfinrud, P. A., and Brunori, M. (2006). Extended subnanosecond structural dynamics of myoglobin revealed by Laue crystallography. Proc. Natl. Acad. Sci. USA 103, 4924–4929. Bourgeois, D., Vallone, B., Schotte, F., Arcovito, A., Miele, A. E., Sciara, G., Wulff, M., Anfinrud, P., and Brunori, M. (2003). Complex landscape of protein structural dynamics unveiled by nanosecond Laue crystallography. Proc. Natl. Acad. Sci. USA 100, 8704–8709. Bourgeois, D., Wagner, U., and Wulff, M. (2000). Towards automated Laue data processing: Application to the choice of optimal X-ray spectrum. Acta Crystallogr. D Biol. Crystallogr. 56, 973–985. Campbell, J. W. (1995). LAUEGEN, an X-windows-based program for the processing of Laue diffraction data. J. Appl. Cryst. 28, 228–236. Haldane, J., and Smith, J. L. (1896). The oxygen tension of arterial blood. J. Physiol. 20, 497–520. Ihee, H., Rajagopal, S., Srajer, V., Pahl, R., Anderson, S., Schmidt, M., Schotte, F., Anfinrud, P. A., Wulff, M., and Moffat, K. (2005). Visualizing reaction pathways in photoactive yellow protein from nanoseconds to seconds. Proc. Natl. Acad. Sci. USA 102, 7145–7150. Knapp, J. E., Bonham, M. A., Gibson, Q. H., Nichols, J. C., and Royer, W. E., Jr. (2005). Residue F4 plays a key role in modulating oxygen affinity and cooperativity in Scapharca dimeric hemoglobin. Biochemistry 44, 14419–14430. Knapp, J. E., Gibson, Q. H., Cushing, L., and Royer, W. E., Jr. (2001). Restricting the ligand-linked heme movement in Scapharca dimeric hemoglobin reveals tight coupling between distal and proximal contributions to cooperativity. Biochemistry 40, 14795–14805.
394
Vukica Sˇrajer and William E. Royer
Knapp, J. E., Pahl, R., Srajer, V., and Royer, W. E., Jr. (2006). Allosteric action in real time: Time-resolved crystallographic studies of a cooperative dimeric hemoglobin. Proc. Natl. Acad. Sci. USA 103, 7649–7654. Knapp, J. E., and Royer, W. E., Jr. (2003). Ligand-linked structural transitions in crystals of a cooperative dimeric hemoglobin. Biochemistry 42, 4640–4647. Knapp, J. E., Sˇrajer, V., Pahl, R., and Royer, W. E., Jr. (2004). Immobilization of Scapharca HbI crystals improves data quality in time-resolved crystallographic experiments. Micron 35, 107–108. Mozzarelli, A., Bettati, S., Rivetti, C., Rossi, G. L., Colotti, G., and Chiancone, E. (1996). Cooperative oxygen binding to Scapharca inaequivalvis hemoglobin in the crystal. J. Biol. Chem. 271, 3627–3632. Pardanani, A., Gambacurta, A., Ascoli, F., and Royer, W. E., Jr. (1998). Mutational destabilization of the critical interface water cluster in Scapharca dimeric hemoglobin: Structural basis for altered allosteric activity. J. Mol. Biol. 284, 729–739. Pardanani, A., Gibson, Q. H., Colotti, G., and Royer, W. E., Jr. (1997). Mutation of residue Phe97 to Leu disrupts the central allosteric pathway in Scapharca dimeric hemoglobin. J. Biol. Chem. 272, 13171–13179. Rajagopal, S., Anderson, S., Srajer, V., Schmidt, M., Pahl, R., and Moffat, K. (2005). A structural pathway for signaling in the E46Q mutant of photoactive yellow protein. Structure (Camb.) 13, 55–63. Rajagopal, S., Kostov, K. S., and Moffat, K. (2004a). Analytical trapping: Extraction of timeindependent structures from time-dependent crystallographic data. J. Struct. Biol. 147, 211–222. Rajagopal, S., Schmidt, M., Anderson, S., Ihee, H., and Moffat, K. (2004b). Analysis of experimental time-resolved crystallographic data by singular value decomposition. Acta Crystallogr. D Biol. Crystallogr. 60, 860–871. Ren, Z., Bourgeois, D., Helliwell, J. R., Moffat, K., Sˇrajer, V., and Soddard, B. L. (1999). Laue crystallography: Coming of age. J. Synch. Radiat. 6, 891–917. Ren, Z., and Moffat, K. (1995). Quantitative analysis of synchrotron Laue diffraction patterns in macromolecular crystallography. J. Appl. Cryst. 28, 461–481. Ren, Z., Perman, B., Sˇrajer, V., Teng, T. Y., Pradervand, C., Bourgeois, D., Schotte, F., Ursby, T., Kort, R., Wulff, M., and Moffat, K. (2001). A molecular movie at 1.8 A˚ resolution displays the photocycle of photoactive yellow protein, a eubacterial blue-light receptor, from nanoseconds to seconds. Biochemistry 40, 13788–13801. Royer, W. E., Jr., Pardanani, A., Gibson, Q. H., Peterson, E. S., and Friedman, J. M. (1996). Ordered water molecules as key allosteric mediators in a cooperative dimeric hemoglobin. Proc. Natl. Acad. Sci. USA 93, 14526–14531. Schmidt, M., Ihee, H., Pahl, R., and Sˇrajer, V. (2005a). Protein-ligand interaction probed by time-resolved crystallography. In ‘‘Methods in Molecular Biology: Protein-Ligand Interactions. Methods and Applications.’’ (N. G. Ulrich, ed.), Vol. 35, pp. 115–154. Humana Press, Totawa, NJ. Schmidt, M., Nienhaus, K., Pahl, R., Krasselt, A., Anderson, S., Parak, F., Nienhaus, G. U., and Srajer, V. (2005b). Ligand migration pathway and protein dynamics in myoglobin: A time-resolved crystallographic study on L29W MbCO. Proc. Natl. Acad. Sci. USA 102, 11704–11709. Schmidt, M., Pahl, R., Sˇrajer, V., Anderson, S., Ren, Z., Ihee, H., Rajagopal, S., and Moffat, K. (2004). Protein kinetics: Structures of intermediates and reaction mechanism from time-resolved x-ray data. Proc. Natl. Acad. Sci. USA 101, 4799–4804. Schmidt, M., Rajagopal, S., Ren, Z., and Moffat, K. (2003). Application of singular value decomposition to the analysis of time-resolved macromolecular x-ray data. Biophys. J 84, 2112–2129.
Time-Resolved X-Ray Crystallography
395
Schotte, F., Lim, M., Jackson, T. A., Smirnov, A. V., Soman, J., Olson, J. S., Phillips, G. N., Jr., Wulff, M., and Anfinrud, P. A. (2003). Watching a protein as it functions with 150-ps time-resolved x-ray crystallography. Science 300, 1944–1947. Schotte, F., Soman, J., Olson, J. S., Wulff, M., and Anfinrud, P. A. (2004). Picosecond timeresolved X-ray crystallography: Probing protein function in real time. J. Struct. Biol. 147, 235–246. Sˇrajer, V., Crosson, S., Schmidt, M., Key, J., Schotte, F., Anderson, S., Perman, B., Ren, Z., Teng, T. Y., Bourgeois, D., Wulff, M., and Moffat, K. (2000). Extraction of accurate structure-factor amplitudes from Laue data: Wavelength normalization with wiggler and undulator X-ray sources. J. Synchr. Radiat. 7, 236–244. Sˇrajer, V., Ren, Z., Teng, T. Y., Schmidt, M., Ursby, T., Bourgeois, D., Pradervand, C., Schildkamp, W., Wulff, M., and Moffat, K. (2001). Protein conformational relaxation and ligand migration in myoglobin: A nanosecond to millisecond molecular movie from time-resolved Laue X-ray diffraction. Biochemistry 40, 13802–13815. Sˇrajer, V., Teng, T., Ursby, T., Pradervand, C., Ren, Z., Adachi, S., Schildkamp, W., Bourgeois, D., Wulff, M., and Moffat, K. (1996). Photolysis of the carbon monoxide complex of myoglobin: nanosecond time-resolved crystallography. Science 274, 1726–1729. Terwilliger, T. C., and Berendzen, J. (1995). Difference refinement: Obtaining differences between two related structures. Acta Crystallogr. D Biol. Crystallogr. 51, 609–618. Ursby, T., and Bourgeois, D. (1997). Improved estimation of structure-factor difference amplitutes from poorly accurate data. Acta Crystallogr. A 53, 564–575.
This page intentionally left blank
C H A P T E R
T W E N T Y
Structural Dynamics of Myoglobin M. Brunori,* D. Bourgeois,† and B. Vallone* Contents 398 399 400 400 402 408 413 413
1. Background 2. Crystallographic Studies of Myoglobin States 3. Experimental Approaches 3.1. Crystallization and sample handling 3.2. Pump and probe picosecond Laue diffraction 3.3. Myoglobin relaxations: A synopsis Acknowledgments References
Abstract Protein structure is endowed with a complex dynamic nature, which rules function and controls activity. The experimental investigations that yield information on protein dynamics are carried out in solution; however, in most cases, the determination of protein structure is carried out by crystallography that relies on the diffraction properties of a large number of molecules, in approximately the same conformation, arranged in a three-dimensional lattice. Myoglobin, maybe the most thoroughly characterized protein, has allowed the formulation of general principles in the field of protein structure–function correlation and, since the late 1990s, it has been possible to obtain directly some insight into the complex dynamic behavior of myoglobin and other proteins by Laue diffraction. This chapter describes some of the technological features involved in obtaining reliable data by time-resolved Laue crystallography, with subnanosecond time resolution. A synopsis of the more significant findings obtained by laser photolysis of myoglobin-CO crystals is also presented, emphasizing the more general aspects of dynamics relevant to the complex energy landscape of a protein.
* {
Dipartimento di Scienze Biochimiche ‘‘A. Rossi Fanelli,’’ Universita` di Roma ‘‘La Sapienza,’’ Roma, Italy Institut de Biologie Structurale Jean-Pierre Ebel, CEA, CNRS, Universite´ Joseph Fourier and European Synchrotron Radiation Facility, Grenoble Cedex, France
Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37020-1
#
2008 Elsevier Inc. All rights reserved.
397
398
M. Brunori et al.
1. Background The essential role of dynamics in controlling protein activity is an accepted fact, based on a large number of experimental data indicating that the primary function, as well as the control of reactivity, has a dynamic component. These concepts were formulated, by and large, based on the interpretation of time-resolved spectroscopy and nuclear magnetic resonance. In this respect, myoglobin (Mb) acquired, over the years, the undisputed role of a paradigmatic case for studying the correlations between structure and dynamics; despite its small size and simple structure, Mb displays a dynamic behavior of considerable complexity and hence has been referred to as the ‘‘paradigm of complexity’’ (Frauenfelder et al., 2003). The discovery of geminate rebinding after laser photolysis of the oxygen-heme iron bond and the correlation of protein relaxation with overall rates of binding and dissociation were essential to set the stage. The combined use of time-resolved spectroscopy and site-directed mutagenesis yielded an overall kinetic picture of the binding of many photodissociable ligands (O2, CO, NO, and others) to ferrous Mb and an assessment of the role of the inner and outer barriers in controlling entry and escape of the ligand in the protein matrix (Scott et al., 2001). In addition, studies on the geminate recombination provided evidence for an unexpected role of the internal structure of the protein in controlling function. Far from being a compact homogeneous matrix, the protein interior contains packing defects and cavities, which were already identified in the mid-1970s (Richards, 1974). In the case of Mb, direct evidence for the presence of these cavities was obtained by crystallography under high pressures of xenon; Tilton et al. (1984) showed that met-Mb can bind four atoms of xenon in small preexisting cavities numbered from Xe1 to Xe4. The role of these cavities in Mb function has been assessed either by filling them with xenon or by mutagenesis (Brunori and Gibson, 2001; Scott et al., 2001). Many investigations contributed to our understanding of the general significance of protein dynamics and paved the way to the concept of a complex energy landscape of a protein. Frauenfelder et al. (1988, 1991) and Ansari et al. (1992) extensively analyzed the problem and highlighted the principal correlations between the function of a protein and its dynamics. Using time-resolved X-ray crystallographic methods to directly measure the relaxation of Mb over the whole range of times explored by the protein matrix is the focus of this chapter.
Structural Dynamics of Myoglobin
399
2. Crystallographic Studies of Myoglobin States The notion that crystallography provides a static picture of the protein structure is widespread. Generally, understanding a mechanism demands analysis of the three-dimensional structure of a protein at least in the initial and final states of the reaction; seldom it proved possible to crystallize reaction intermediates and analogues. The dynamics of motions along the pathway from one state to another is not amenable to classical X-ray crystallography. Attempts to extract dynamic information from diffraction experiments have been published (Frauenfelder et al., 1979; Rasmussen et al., 1992). Frauenfelder et al. (1979) analyzed the asymmetric distribution of thermal B factors from diffraction data of a met-Mb crystal at different temperatures and attempted to correlate this distribution with the ligandlinked conformational changes. A more challenging and informative crystallographic approach to unveil the structure of reaction intermediates was based on using trapping techniques (Bourgeois and Royant, 2005). Mb proved especially suitable because of the well-known photosensitivity of the ligand-iron bond; thus, crystals of MbCO can be photodissociated by stationary light at ultralow temperatures (10–20 K) so that intermediate states may accumulate in a steady-state regime. Analysis of the photolytic intermediate at ultralow temperatures (Schlichting et al., 1994; Teng et al., 1997) demonstrated, for the first time, that photodissociated CO resides in the distal heme pocket at a distance of 3.6 A˚ from the iron (reviewed by Schlichting and Chu, 2000). Using the same approach a few years later, Brunori et al. (2000) showed that the photodissociated ligand in the triple Mb mutant (called YQR) migrated away even at extremely low temperatures to populate the Xe4 cavity on the distal side. Informative as they are, these experiments failed to detect the larger motions of the protein moiety (stuck at ultralow temperatures) unless the crystal was progressively warmed up to the so-called dynamical transition temperature where ligand diffusion further away becomes detectable (Ostermann et al., 2000). These results showed how crucial it is to obtain time-resolved crystallographic data at room temperature (Bourgeois and Royant, 2005). This missing information was tackled by the pioneering work of Moffat and collaborators (Srajer et al., 1996); since the late 1990s, Laue diffraction results with nanosecond and subnanosecond time resolutions have provided considerable insight into the complex dynamic behavior of Mb and other proteins (Bourgeois et al., 2003; Genick et al., 1997; Ihee et al., 2005; Knapp et al., 2006; Schotte et al., 2003; Srajer et al., 2001). This chapter presents a short description of the methodological improvements that made it possible to crack this challenging technological tour de
400
M. Brunori et al.
force, and presents a short synopsis of the essential results obtained by following the structural dynamics of Mb from picoseconds to milliseconds and the general conclusions emerging from this approach.
3. Experimental Approaches 3.1. Crystallization and sample handling The level of detail provided by recent time-resolved studies carried out on Mb-YQR (see later) is the result of both careful tuning of sample preparation and drastic improvements in the experimental strategy employed for the collection of pump-probe Laue data on specialized synchrotron beam lines. The quality of the samples is a mandatory prerequisite for the success of time-resolved experiments and can, to a certain extent, be assessed before data collection is attempted. Myoglobin crystals are normally grown in the oxidized ferric form (met-Mb) and are subsequently soaked with the appropriate ligand in order to analyze the bound state. In the case of carbon monoxide, the state competent for ligation is the ferrous one, and pretreatment with a reducing agent, usually sodium dithionite, is necessary in order to obtain the species competent for CO (and dioxygen) binding. This procedure is normally mild enough to yield samples that can still be subjected to data collection for standard structural studies. Nevertheless, in the case of Mb-YQR, reduction with dithionite and subsequent soaking with a CO-saturated mother liquor result in a substantial increase in mosaicity (from 0.3 to 0.7 and above) and in a loss of crystal order that hamper the detection of fine structural details and prevent the unambiguous detection of CO docking within protein cavities. This problem can be overcome by growing Mb-YQR crystals in batches starting from the ferrous CO-bound derivative; in this way, not only is crystal degradation upon treatment avoided but, at the same time, the contribution from residual ferric unbound protein, decreasing overall photolytic yield in the crystal, is essentially eliminated. A protocol describing the conditions for batch crystallization starting from a vapor diffusion setup and leading to growth of Mb-YQR-CO crystals suitable for timeresolved studies is provided in the following section. 3.1.1. Seed preparation Crystals are first grown using the vapor diffusion method (Phillips et al., 1990) in 2.7 M ammonium sulfate, 20 mM Tris-Cl, pH 8.7, 1 mM EDTA, and 20 mg/ml met-Mb-YQR. A single crystal with no apparent defect is crushed within 1 ml mother liquor, mixed with a vortex mixer for a few seconds, and serially diluted in mother liquor (from 1 to 10,000 times).
Structural Dynamics of Myoglobin
401
These ‘‘seed stocks’’ have to be stored carefully sealed at 20 , and are effective for seeding purposes for a few months. 3.1.2. Search for optimal batch crystallization growth conditions First, determine the optimal seed dilution by setting up five batch crystallization trials in the following way: equilibrate distilled water, 4.1 M ammonium sulfate solution (A/S), 1.5 M Tris-Cl pH 8.7 (stock buffer), 0.1 M EDTA, and mineral oil with CO at 1 atm by bubbling under a gas hood; the CO concentration after equilibration in water at 20 is 1 mM (some deviations from this concentration are expected for the other solutions). Prepare anaerobically a solution of 15 mM sodium dithionite using the CO-equilibrated stock buffer. First lay 30 ml of mineral oil in each tube, add 10 ml of ferric Mb-YQR under the oil layer with a gas-tight microsyringe, and add 2 ml of dithionite 15 mM in CO-equilibrated stock buffer. This ensures that the protein is reduced and bound with CO in an anaerobic environment. Then add 0.2 ml of EDTA and water (if necessary) and the ammonium sulfate-saturated solution. Finally, mix the solution with a gastight microsyringe; some precipitation may appear. As a last step, add 0.5 ml of the five seed stock solutions at different dilutions, after mixing them thoroughly with a vortex mixer. The volumes of the different components to mix to obtain 0.1 M Tris-Cl, 1 mM EDTA, and 10 mg/ml Mb solutions of a final concentration of 2.8 M ammonium sulphate, are as follow: 5 ml protein at 40 mg/ml, 13.5 ml A/S, 1.3 ml of stock buffer plus 15 mM sodium dithionite, 0.2 ml of 0.1 M EDTA, and 0.5 ml of each seed stock solution. The Tris-Cl concentration is increased with respect to the vapor diffusion protocol given the presence of 10 mM dithionite; the protein concentration is decreased to reduce the amount necessary for setting up the experiment and, as a consequence, the precipitant (ammonium sulfate) concentration is increased slightly. Immediately after preparation, the tubes have to be stored at 20 in a glass vessel purged with CO. The seed dilution that is found to yield the optimal number of crystals (5–10 per vial) is 1:10. With that particular protein and seed batch using 2.8 M A/S and 5 mg/ml protein concentration, we obtain growth of crystals within 2–3 days, completed within 1 week with 5–10 single crystals in each tube of a size ranging from 100 to 400 mm in the largest dimension. Some precipitate is often observed in the, batches and crystals tend to grow at the oil–solution interface. Figure 20.1 shows an example of a microbatch vial after crystal growth. ˚ resolution in timeCrystals grown using this protocol diffract up to 1.55 A resolved experiments, with a mosaicity ranging between 0.25 and 0.30 (mosaicity is assessed independently by static X-ray data collection). They show 100% CO ligation and undetectable ferric Mb contamination, as assessed by microspectrophotometric measurements.
402
M. Brunori et al.
Mineral oil
Gas tight wax Crystals
CO atmosphere
Oil/solution interface
Mineral oil Mineral oil
Mb crystal
Protein precipitate
Figure 20.1 (Left) Microbatch-grown crystals of Mb-YQR-CO. Crystals grew at the oil/solution interface; some precipitate is present at the bottom of the vial. (Right) Crystal mounting of Mb-YQR-CO for time-resolved experiments.
Crystals are mounted in 1-mm quartz capillaries, using mother liquor equilibrated with CO and containing 1 mM sodium dithionite under 1 atm of CO. The presence of a reducing agent and gaseous CO ensures that Mb is kept reduced and bound to CO throughout the measurement. In order to reduce the effects of heat delivered onto the sample by laser illumination, mild cooling to 10 C is ensured by means of a nitrogen cryostream. The combination of laser heating and cryostream cooling tends to induce temperature gradients throughout the capillary, resulting in solvent distillation to/from the crystal. The problem can be overcome by mounting the crystal in the capillary between two plugs of mineral oil placed very close to the sample (see Fig. 20.1).
3.2. Pump and probe picosecond Laue diffraction Time-resolved Laue diffraction with subnanosecond resolution employs the ‘‘pump and probe’’ strategy: a laser pulse initiates the reaction within the crystal and is followed by an X-ray pulse providing diffraction data from the excited molecules (Schotte et al., 2004; Wulff et al., 2002). The X-ray pulse originates from a single electron bucket circulating around the synchrotron storage ring. The sequence needs to be repeated many times to (i) build up a sufficient signal-to-noise ratio for each individual diffraction pattern, (ii) collect diffraction patterns at several crystal orientations so as to sample the reciprocal space entirely, and (iii) obtain structural information at different pump-probe delays to scan the time dimension. Therefore, the
Structural Dynamics of Myoglobin
403
reaction under study needs to relax back to the dark resting state in seconds or less so that many cycles can be performed on the same sample within a reasonable experimental time. The crystal must also withstand repeated laser and X-ray pulses without significant loss of diffraction quality. This is why the method is best suited for light-sensitive proteins that undergo reversible photo cycles and yield excellent quality and radiationhard crystals. Investigations of single-turnover reactions based, for example, on photolysis of caged compounds (Bourgeois et al., 2005) have been sometimes successful in the past (Schlichting et al.,1990; Stoddard et al., 1998), but seem poorly compatible with the now consensual ‘‘pink’’ Laue data collection strategy employed on most time-resolved beam lines (see later). 3.2.1. Crystal photolysis Reaction initiation by laser light within a protein crystal is one of the most challenging issues of time-resolved Laue experiments and is a critical determinant for success or failure. To provide efficient and homogeneous excitation throughout the sample, one must deal with the generally very high optical density of colored protein crystals. The laser wavelength, pulse energy, and pulse duration must also be chosen to minimize heating and temperature gradients across the crystal and to avoid spurious oxidation or ionization events. Laser pulses approaching the femtosecond regime may generate nonlinear or multiphotons effects that may pose insurmountable problems (P. Anfinrud, personal communication). Figure 20.2 shows a basic simulation of how light penetrates into a Mb-CO crystal at three different wavelengths. It is striking to see that photons at wavelengths near the Q absorption bands of the heme moiety penetrate very little into the crystal, whereas photons in the Soret band do not penetrate at all. Photons at remote wavelengths penetrate deeper but require much more laser power to achieve photolysis. In practice, Fig. 20.2 shows that a rather good compromise is obtained at 500 nm, provided that thin crystals (less than 50 mm thick) are considered. As Mb crystals are typically much thicker, either the X-ray focal spot must be restricted to match the illuminated layer of the crystal (solution adopted at beam line ID09-B of the ESRF) or the crystal must be illuminated from several sides to increase the excited volume (solution adopted at beam line 14-ID, BioCARS, Advanced Radiation Source at Argonne National Laboratory). With Mb-YQR crystals on beam line ID09-B of the ESRF, the X-ray beam (50 mm) is typically centered 25 mm below the illuminated top edge of the crystal. The simulations used in Fig. 20.2 predict an overall photolysis yield of 74% at 505 nm excitation, yet such a yield is never achieved. Many pitfalls combine to reduce the effective photolysis yield obtainable in crystals. Among these are imperfect alignment of the laser beam, X-ray beam, and crystal; loss of photolysis light as a consequence of Fresnel reflections on
404
N photons absorbed per heme-CO
M. Brunori et al.
10 30000
582
8
20000 10000
6
Mb-CO Mb-deoxy Mb-met
0 450
4
500 550 600 650 Wavelength (nm)
700
505 nm 2 630 nm 0 0
10
20 30 Depth (mm)
40
50
Figure 20.2 Theoretical penetration of visible laser light into a myoglobin crystal.The crystal is assumed to be composed of 80% Mb-CO and 20% of oxidized Mb-met. A laser pulse energy of 40 mJ is assumed that impinges on a crystal of dimensions 200 200 50 mm and space group P61. The calculation neglects effects discussed in the text, such as misalignments, polarization effects, Fresnel reflections, or transient change of absorption in the excited states.The calculation makes use of experimentally measured extinction coefficients of myoglobin in the CO-bound and oxidized met states, which are shown in the inset, together with those of the deoxy state. It is seen that whereas light at 582 nm (Q-band maximum) creates a large absorption gradient in the crystal, light at 630 nm is almost not absorbed. Light at 500 nm represents a good compromise, although only a maximum crystal thickness of 50 mm can be photolyzed successfully if only one laser beam is used.
capillary and crystal surfaces; and effect of nonrandom orientation of the absorbing groups within the crystal unit cell (potentially implying a dependence of photolysis yield on crystal orientation). In Mb, the presence of a fraction of oxidized met-Mb in the crystal, as well as incomplete CO saturation, will also alter the observed yield. In addition, other more fundamental aspects may play a role: the amount of very fast geminate recombination (on the subnanosecond timescale) might be enhanced in the crystalline state because of reduced conformational freedom or because of a different thermodynamic balance between conformational substates present in the crystal and in solution. At a given laser pulse energy, the photolysis yield might also depend on the pulse duration. When the pulse duration is shortened to the femtosecond regime, it may approach the lifetime of early excited states. Therefore, even if several photons are available per molecule, only the first one absorbed may lead to photolysis, as the following ones will then be dumped onto an already excited molecule that had no time to relax back to the ground state. Furthermore, the absorption spectrum of early excited states may strongly absorb at the excitation wavelength and transiently shield the crystal (P. Anfinrud, personal communication). On the contrary, longer laser pulses allow individual molecules to
Structural Dynamics of Myoglobin
405
relax back to the ground state between successive absorption events, offering multiple chances for successful photolysis. In general, long pulse durations achieve better photolysis yield, while minimizing multiphoton excitations, and are therefore easier to handle. With pulses in the 100-ps to nanosecond regime, typically photolysis yields between 10 and 40% are obtained. On beam line ID09-B of the ESRF, laser pulses of 40 mJ energy, 100 ps duration, and with wavelengths spanning the 480- to 560-nm range have been used for studies on Mb. Pulses of 100 fs duration are produced by an optical parametric amplifier pumped by a Ti:sapphire regenerative amplifier. Temporal pulse stretching from 100 fs to 100 ps is realized by the combined use of a Brewster cut fused silica glass block and of an optical fiber transporting laser light to the focusing optics near the sample. The optical fiber also achieves spectral broadening so that a 40-nm band-pass spectrum impinges on the sample (Schotte et al., 2004). Under these conditions, in the case of Mb-YQR, which provides rather large crystals, it is usually possible to utilize a single crystal for collecting more than one set of time points, taking advantage of different regions of the crystal. The rather low laser energy employed also reduces the problem of crystal slippage, which was evident in early experiments when a hot laser spot delivering 1.0–1.5 mJ at softer wavelengths was used. In order to avoid movements of the crystal induced by laser flashes, a sticky polyvinyl coating can be adopted (Knapp et al., 2004). 3.2.2. X-ray data collection The collection of Laue data (Clifton et al., 1997; Cruickshank et al., 1991; Ren and Moffat, 1995a; Ren et al., 1999) with subnanosecond time resolution (Bourgeois et al., 1996; Wulff et al., 2002) has been reviewed in detail (Bourgeois et al., 2007) and is briefly reported. Time-resolved Laue data collection is carried out under specific configurations of the synchrotron ring: the so-called ‘‘single-bunch’’ or ‘‘few-bunches’’ modes are used, where one or few electron bunches generate very intense 100-ps X-ray pulses when traversing an insertion device. These X-ray pulses are now produced with narrow band-pass monoharmonic undulators, not with broad bandpass wigglers. Monoharmonic undulators deliver a hot, so-called ‘‘pink’’ beam, with most photons emitting in a single tooth-shaped line with a sharp energy cutoff. Development of the pink Laue concept followed from the realization that the broad X-ray band pass generated by wiggler insertion devices, despite large coverage of reciprocal space in a single shot, delivers data of rather poor quality (Bourgeois et al., 2000). Undulator beams with a few percent band pass improve the signal-to-noise ratio greatly because, while the intensity of each Laue diffraction spot is preserved, the X-ray noise at every pixel of the detector is reduced in proportion to the band pass.
406
M. Brunori et al.
This improvement in the signal-to-noise ratio is at the expense of a reduced coverage of reciprocal space per frame, but this is a tractable problem for the study of cyclic reactions in crystals able to sustain repeated excitations. The sparse reciprocal lattice points stimulated by the pink beam also result in simplifications of the diffraction images, reducing the number of harmonic and spatial overlaps drastically (Ren et al., 1999), thus providing better raw data (Bourgeois et al., 2000). Enhanced evenness of reciprocal space coverage also squeezes the ‘‘low-resolution hole’’ typical of wiggler Laue diffraction data and known to cause discontinuities in electron density maps. Advantages in Laue geometry provided by monoharmonic undulators are further strengthened by a decrease in beam line heat load, resulting in a more stable beam. To allow for the selection of properly timed X-ray pulses on the crystal, high-speed choppers with opening times in the microsecond range have been designed. These devices rotate at about a kilohertz and are phase locked to the storage ring clock, ensuring synchronization to the X-ray beam with nanosecond mechanical jitter. Opening times down to 100 ns have been achieved. Synchronization between laser and X-ray pulses is realized by sophisticated electronics providing picosecond accuracy (Schotte et al., 2004). The collection of time-resolved Laue data occurs in a four-dimensional space and therefore should be planned carefully. To avoid systematic errors that may bias the observed structural changes along the time axis, e.g., because of crystal photo damage, time is used as the ‘‘fast variable.’’ At a given crystal orientation, all the diffraction images at the investigated pumpprobe delays (sampled equidistantly on the logarithmic timescale) are collected and then the crystal is rotated and the process is repeated (Ihee et al., 2005). A ‘‘laser off ’’ image is also recorded to provide a ‘‘dark state’’ reference structure of the sample. The calculation of difference electron density maps between different time points then mostly subtracts out irreversible structural changes caused by X-ray or laser light. It should be noted that such a data collection scheme spreads out errors along the time axis, but does not eliminate them. The quality of the diffraction images should therefore be checked throughout the experiment by visual inspection or preferably by online data processing. On-the-fly assessment of data quality during the course of the experiment is now possible thanks to recent progress in software and fast computers. In the last few years, methodology for reducing Laue data has improved considerably, providing accuracy and automation in addition to speed (F. Schotte, unpublished results; Bourgeois et al., 2000; Ren, 2006). Highly robust indexing algorithms have been developed, which are based on the recognition of entire conics and are therefore able to cope with the limited number of nodal spots present in ‘‘pink’’ Laue patterns (Ren, 2006 E. R. Henry, unpublished results). Wavelength-dependent prediction of
Structural Dynamics of Myoglobin
407
the crystal resolution limit has been implemented and has brought about substantial improvement in data quality (Bourgeois et al., 2000). Spatial and harmonic overlaps are now deconvoluted computationally on a routine basis (Arzt et al., 1999; Bourgeois et al., 1998; Ren and Moffat, 1995b). However, other challenges remain in the processing of Laue data from narrow band-pass sources, as expected in the future from, for example, X-ray-free electron lasers. Proper wavelength normalization of ‘‘mosaic’’ or ‘‘partial’’ reflections is still lacking and will become of increasing importance as the X-ray band pass decreases. Structural changes at the investigated time delays are classically revealed by computing difference electron density Fourier maps preferably enhanced by Bayesian weighting of the (small) difference structure factor amplitudes (Ursby et al., 1997). Qualitative assessment of data may benefit from maps extrapolated to full photolysis to eliminate the contribution of the nonexcited fraction of the crystal, which can then be rendered with special coloring schemes so as to guide the eye in following the subtle motions of individual atoms (Schotte et al., 2004). Quantitative evaluation of data is often based on integrating density features in difference maps at the locations of interest so as to assess the evolution of their electron content over time (Bourgeois et al., 2003; Schmidt et al., 2005). The time evolution of these features may then be fitted by kinetic schemes (Knapp et al., 2006; Schmidt et al., 2005). However, the method suffers from a fundamental limitation, as it is the rule rather than the exception that several conformational states coexist in the crystal at any given time. Hence, density features generally do not represent a single species, but rather a combination of several species. This is why refinement of structural models from data at a single time point should be attempted and interpreted only with great care (Aranda et al., 2006; Bourgeois et al., 2003), preferably using the method of difference refinement (Terwilliger and Berendzen, 1995). To extract the structures of time-independent transient species, as well as their rates of interconversion, the technique of singular value decomposition (SVD) has been elegantly generalized to time-resolved crystallography (Rajagopal et al., 2004; Schmidt et al., 2003). This method works well when the amount of time points is sufficient (approximately three per decade) and when data are devoid of systematic errors along the time axis. In addition, SVD analysis provides an efficient noise filter and therefore enhances the signal-to-noise ratio in reconstituted electron density difference maps. However, kinetic models describing the time evolution of the investigated sample may only be shown to be consistent with data, i.e., they may not be proven to govern the reaction scheme. They may, though, be compared to the SVD outcome of spectroscopic data, which brings links between crystal and solution states and adds considerable weight to the validity of structural data (Yeremenko et al., 2005, 2006).
408
M. Brunori et al.
3.3. Myoglobin relaxations: A synopsis The pioneer in this field is undoubtedly K. Moffat, who worked for two decades to obtain the first reliable time-resolved diffraction data presenting difference electron density maps between dark and photolyzed states of Mb (Srajer et al., 1996). The quality of results obtained at the time bears no comparison with current data due to progressive improvements in biochemistry, synchrotron technology and analytical methods, as outlined briefly earlier. Indeed, when looking at the paper by Srajer et al. (1996), it is clear that quantitative information was very limited; nevertheless this was a tremendous step forward, as it proved that the experiment was feasible and could potentially yield novel information. An intrinsic difficulty of the approach that will be hard to overcome is the yield of photodissociated Mb obtained by the laser flash, as well as the time window that can be explored before ligand recombination leads back to the equilibrium state. Over and above the problem related to laser penetration in the optically very dense crystal (see earlier discussion), it is obvious that ligand recombination events, either geminate or bimolecular, progressively reduce the yield of deoxy Mb photoproduct that undergoes relaxation. In this respect, site-directed mutagenesis proved very valuable; among others, the triple Mb mutant called YQR (Fig. 20.3) (Brunori et al., 1999) was found to be an excellent model system because mutated residues in the distal pocket were found to reduce geminate recombination to zero and to slow down (by 10-fold) the bimolecular recombination, extending
Tyr29 Gln64
Phe46
Haem
Bound CO Photolyzed CO (in Xe1 site) His93
Figure 20.3 Overview of the modeled structure of YQR-MbCO (dark gray) and YQR-Mb* (light gray). The heme and several key residues (Tyr29B10, Phe46CD4, Gln64E7, His93F8) are rendered as balls and sticks, the CO is rendered as space filling, and the rest of the protein is rendered as a ribbon. (From Bourgeois et al., 2003, modified).
Structural Dynamics of Myoglobin
409
considerably the time window available for observation of the relaxation phenomena. Thus, it is very likely that engineering the globin in the future may produce additional variants that will allow the range of useful data acquisition to be extended. Wild-type sperm whale Mb and three mutants have been investigated in great detail (Bourgeois et al., 2003, 2006; Schmidt et al., 2005; Schotte et al., 2003; Srajer et al., 2001). In all cases, a fundamental feature of the time course after pholysis is the nonexponential extended relaxation of deoxy Mb, ranging from subnanoseconds up to milliseconds; in this respect, all the Mb variants tested share a common behavior. This first-order conclusion is very rewarding because it provides generality to the main observation. At the same time, subtle differences in dynamic behavior between mutants and the wild-type protein had to be interpreted and correlated with the extensive kinetics obtained in solution. Figure 20.3 and 20.4 depict a synopsis of data reported for Mb-YQR (Bourgeois et al., 2006). The outstanding result extracted from a quantitative analysis of the electron density difference Fourier maps is as follows: the relaxation profile of the heme and the globin moiety cover a time range from 150 ps (the earliest available time frame) to milliseconds when recombination to the dark state occurs. The latest data obtained on this triple mutant show that heme relaxation is at least biphasic, with some change occurring synchronously with the photolysis laser pulse, and additional shifts in heme geometry within nanoseconds. This relaxation behavior may also be present in the wild-type protein; however, in the specific case of Mb-YQR the heterogeneity was obvious and was attributed (Bourgeois et al., 2006) to a strain on heme pyrrole C exerted by the E helix via the CD turn. Most prominent from Fig. 20.4 is that the globin conformational change appears with a lag of several hundred nanoseconds, as may be seen by looking at the relaxation of the distal helix E and several residues of the CD turn. The extended relaxation behavior represents strong support to the idea (championed by Frauenfelder and colleagues) that photolysis sets into motion a protein quake; changes in the immediate environment of the heme, including distal side chains, occur rapidly, and the perturbation diffuses away with rearrangements of the globin at later times following the focal quake. The main results, which, despite some differences, are shared by essentially all the different variants, are of significance in the light of the energy landscape concept of a protein (Frauenfelder et al., 1991), which envisages an extended relaxation as the protein skates along a complex set of conformational coordinates. Moreover, the extended relaxation is the structural counterpart of numerous and detailed kinetic data obtained over the years by time-resolved spectroscopy (Ansari et al., 1994; Scott et al., 2001); as an example, we recall the photolytic experiments reported by Anfinrud and co-workers (Lim et al., 1997), which showed that the relaxation of
410 A
Positive density [normalized electron]
M. Brunori et al.
0.6 0.5 Fe 0.4 0.3 Heme
0.2 0.1 0 0.1
1
Positive density [normalized electron]
B
100
1000
0.10 0.09 0.08 0.07 0.06
Phe46
0.05 0.04 0.03 0.02 0.01 0 0.1
Phe43 1
C Positive density [normalized electron]
10 Time (ns)
10 Time (ns)
1000
100
0.050 0.450 0.040 0.350
HelixE Xe4
0.030
Xe1
0.250 0.020 0.015 0.010 0.005 0 0.1
1
10 Time (ns)
100
1000
Figure 20.4 Time dependence after photolysis of difference electron densities for key structural features. Numerical values reflect the integral of the positive electron density beyond 3.0s and are corrected for variations in photolysis yield.They are normalized so that the negative bound-CO feature is assigned a value of 1. Average values weighted by photolysis yield, over four independent data sets are shown. (A) Key features that appear
Structural Dynamics of Myoglobin
411
photolyzed deoxy Mb covers a range from 10 ps up to many hundred microseconds. One more fact that is common and well established concerns the pathway of internal migration of the photolyzed CO, which remains trapped momentarily in the protein matrix, and the differences in the dynamics of these events among the different variants. The notion that CO migrates from the distal pocket of the heme through the protein matrix after photolysis was well anticipated by analysis of the geminate recombination profile of different ligands (CO, NO and O2) and by the effect of specific mutations and xenon at high pressures on the rebinding mechanism (Brunori and Gibson, 2001; Scott and Gibson, 1997; Scott et al., 2001). It was remarkable to confirm these hypotheses using Laue diffraction by watching directly the route followed by CO inside the protein. The permanence of CO in the distal pocket depends on the type of variant, being longer in the wild-type protein and considerably shorter with Mb-YQR (Bourgeois et al., 2006; Schotte et al., 2003); in the latter case, the photolyzed CO diffuses into the Xe4 cavity on the distal side almost immediately (<100 ps). This observation is in complete agreement with the lack of geminate recombination yield for Mb-YQR, and it was interpreted in terms of the different role played by the Gln-Tyr couple on the distal site, as compared to the wild-type His-Leu couple. The subsequent migration of CO along the web of cavities occurs at longer times (see Fig. 20.4) and displays essentially the same time course in the different variants. Thus the migration of CO from distal pocket to Xe4 and then to Xe2 and Xe1 on the proximal side takes around 100–300 ns, suggesting a correlation between ligand migration and large-scale structural relaxation of the globin; this point was analyzed carefully by Srajer et al. (2001) and compared with the recombination time course. This feature, which is well established, is not a peculiarity of Mb because it is known that internal packing defects identified in the structure of many proteins (especially from mesophiles) may represent temporary stations for small ligands or substrates entrapped inside the protein. Several authors have addressed the point of the possible significance of migration in the different cavities and functional control, as discussed in some review articles (Brunori et al., 2004; Lavalette et al., 2006; Nienhaus and Nienhaus, 2004; Teeter, 2004; Vallone and Brunori, 2004); some envisaged that the protein and its web of cavities may be viewed as a molecular reactor in more complex oxidative processes.
promptly. (B) Residues involved in the strain of the CD turn, lagging behind. (C) Population of CO in Xe1 and Xe4 sites and conformational changes of the E helix (where the average integrated density per residue is shown). (From Bourgeois et al., 2006, modified).
412
M. Brunori et al.
A final important consideration afforded by recent data is the direct comparison between experimental results and molecular dynamics simulations. This has become possible because the time resolution of Laue diffraction data extended down to the hundreds of picoseconds while computational power allowed an increase of the length of the simulation up to 100 ns or above, providing an extended time overlap between the two techniques. A direct comparison of experimental data on protein relaxation and molecular dynamics simulations had not been possible before and it represents a test case of utmost importance. It is comforting that the two papers reporting a detailed comparison between experimental data and simulations (Bossa et al., 2005; Hummer et al., 2004) actually demonstrated an excellent agreement between the two approaches (Fig. 20.5); the opportunity to combine the two techniques is of tremendous significance for the future of the field (see Fig. 20.5). First of all, simulations may allow a clearer identification of essential dynamic features that may have escaped when DP
Xe4
Ph1
Ph2
Xe3 Xe2
Distance Å
Xe1
DP
6.0
Xe4
Ph1
Ph2
Xe3
Xe2
Xe1
4.0 2.0 0.0
0
10
20
70
80
90
Time (ns)
Figure 20.5 Crystal structure of the wild-type CO-bound sperm whale Mb is shown in the center. Cavities involved in the migration of the ligand, detected by the package SURFNET (Laskowski, 1995), are shown in dark gray and are indicated by legends; DP is the distal pocket. CO is not shown. (Bottom) The trajectory of the distance between the CO center of mass and the (arbitrarily chosen) center of each cavity is depicted. The different shades of gray and labels represent the locations of CO corresponding to cavities in the top panel. A few backward and forward transitions between adjacent cavities can be observed (DP-Xe4 at t ¼ 10 ns; Xe4-Ph1 at t ¼ 15 ns; Xe1-Xe2 in the 70- to 90-ns range); in the 25- to 65-ns time range (not shown), CO remains in the Xe1 cavity. It may be noticed that transitions in between cavities are very fast, showing that the ligand is hopping rapidly through channels connecting the cavities. (From Bossa et al., 2005, modified).
Structural Dynamics of Myoglobin
413
looking at the ensemble averaged difference maps at different times. Moreover, the number of experimental time points along the much extended time course of relaxation of Mb is necessarily limited, contrary to the molecular dynamic trajectories. Last but not least, comparison to experimental data represents a control of validity of the procedures used in molecular dynamic simulations and thereby allows envisaging that these results may pave the way to a much more widespread application of this approach to describe the conformational relaxation of many proteins not amenable to Laue diffraction. After all, the approach by time-resolved Laue crystallography (informative as it is) is limited to heme proteins and a handful of other systems, given the absolute requirement to trigger a fully reversible process quickly and synchronously in the macromolecular crystal.
ACKNOWLEDGMENTS Work was partially supported by the Ministero Italiano dell’Universita` e della Ricerca (grants to M.B: FIRB 2003 RBLA03B3KC and PRIN 2005 ‘‘Dinamica strutturale di metalloproteine’’).
REFERENCES Ansari, A., Jones, C. M., Henry, E. R., Hofrichter, J., and Eaton, W. A. (1992). The role of solvent viscosity in the dynamics of protein conformational changes. Science 256, 1796–1798. Ansari, A., Jones, C. M., Henry, E. R., Hofrichter, J., and Eaton, W. A. (1994). Conformational relaxation and ligand binding in myoglobin. Biochemistry 33, 5128–5145. Aranda, R. T., Levin, E. J., Schotte, F., Anfinrud, P. A., and Phillips, G. N., Jr. (2006). Time-dependent atomic coordinates for the dissociation of carbon monoxide from myoglobin. Acta Crystallogr. D Biol. Crystallogr. 62, 776–783. Arzt, S., Campbell, J. W., Harding, M. M., Hao, Q., and Helliwell, J. R. (1999). LSCALE: The new normalization, scaling and absorption correction program in the Daresbury Laue software suite. J. Appl. Cryst. 32, 554–562. Bossa, C., Amadei, A., Daidone, I., Anselmi, M., Vallone, B., Brunori, M., and Di Nola, A. (2005). Molecular dynamics simulation of sperm whale myoglobin: Effects of mutations and trapped CO on the structure and dynamics of cavities. Biophys. J. 89, 465–474. Bourgeois, D., Nurizzo, D., Kahn, R., and Cambillau, C. (1998). An integration routine based on profile fitting with optimized fitting area for the evaluation of weak and/or overlapped 2-dimensional Laue or monochromatic patterns. J. Appl. Cryst. 31, 22–35. Bourgeois, D., and Royant, A. (2005). Advances in kinetic protein crystallography. Curr. Opin. Struct. Biol. 15, 538–547. Bourgeois, D., Schotte, F., Brunori, M., and Vallone, B. (2007). Time-resolved Laue crystallography as a tool to investigate photo-activated protein dynamics. Photochem. Photobiol. Sci. 6, 1047–1056. Bourgeois, D., Ursby, T., Wulff, M., Pradervand, C., Legrand, A., Schildkamp, W., Laboure´, S., Srajer, V., Teng, T. Y., Roth, M., and Moffat, K. (1996). Feasibility and realization of single pulse Laue diffraction on macromolecular crystals at ESRF. J. Synchr. Radiat. 3, 65–74.
414
M. Brunori et al.
Bourgeois, D., Vallone, B., Arcovito, A., Sciara, G., Schotte, F., Anfinrud, P. A., and Brunori, M. (2006). Extended subnanosecond structural dynamics of myoglobin revealed by Laue crystallography. Proc. Natl. Acad. Sci. USA 103, 4924–4929. Bourgeois, D., Vallone, B., Schotte, F., Arcovito, A., Miele, A. E., Sciara, G., Wulff, M., Anfinrud, P., and Brunori, M. (2003). Complex landscape of protein structural dynamics unveiled by nanosecond Laue crystallography. Proc. Natl. Acad. Sci. USA 100, 8704–8709. Bourgeois, D., Wagner, U., and Wulff, M. (2000). Towards automated Laue data processing: Application to the choice of optimal X-ray spectrum. Acta Crystallogr. D Biol. Crystallogr. 56, 973–985. Bourgeois, D., and Weik, M. (2005). In ‘‘Dynamic Studies in Biology’’ (M Goeldner and R Givens, eds.), pp. 410–434. Wiley-vch, Weinham. Brunori, M., Bourgeois, D., and Vallone, B. (2004). The structural dynamics of myoglobin. J. Struct. Biol. 147, 223–234. Brunori, M., Cutruzzola, F., Savino, C., Travaglini-Allocatelli, C., Vallone, B., and Gibson, Q. H. (1999). Structural dynamics of ligand diffusion in the protein matrix: A study on a new myoglobin mutant Y(B10) Q(E7) R(E10). Biophys. J. 76, 1259–1269. Brunori, M., and Gibson, Q. H. (2001). Cavities and packing defects in the structural dynamics of myoglobin. EMBO Rep. 2, 674–679. Brunori, M., Vallone, B., Cutruzzola, F., Travaglini-Allocatelli, C., Berendzen, J., Chu, K., Sweet, R. M., and Schlichting, I. (2000). The role of cavities in protein dynamics: Crystal structure of a photolytic intermediate of a mutant myoglobin. Proc. Natl. Acad. Sci. USA 97, 2058–2063. Clifton, I. J., Duke, E. M. H., Wakatsuki, S., and Ren, Z. (1997). Evaluation of Laue diffraction patterns. Methods Enzymol. 277, 448–467. Cruickshank, D. W. J., Helliwell, J. R., and Moffat, K. (1991). Angular distribution of reflections in Laue diffraction. Acta Crystallogr. A 47, 352–373. Frauenfelder, H., McMahon, B. H., and Fenimore, P. W. (2003). Myoglobin: The hydrogen atom of biology and a paradigm of complexity. Proc. Natl. Acad. Sci. USA 100, 8615–8617. Frauenfelder, H., Parak, F., and Young, R. D. (1988). Conformational substates in proteins. Annu. Rev. Biophys. Chem. 17, 451–479. Frauenfelder, H., Petsko, G. A., and Tsernoglou, D. (1979). Temperature-dependent X-ray diffraction as a probe of protein structural dynamics. Nature 280, 558–563. Frauenfelder, H., Sligar, S. G., and Wolynes, P. G. (1991). The energy landscapes and motions of proteins. Science 254, 1598–1603. Genick, U. K., Borgstahl, G. E., Ng, K., Ren, Z., Pradervand, C., Burke, P. M., Srajer, V., Teng, T. Y., Schildkamp, W., McRee, D. E., Moffat, K., and Getzoff, E. D. (1997). Structure of a protein photocycle intermediate by millisecond time-resolved crystallography. Science 275, 1471–1475. Hummer, G., Schotte, F., and Anfinrud, P. A. (2004). Unveiling functional protein motions with picosecond x-ray crystallography and molecular dynamics simulations. Proc. Natl. Acad. Sci. USA 101, 15330–15334. Ihee, H., Rajagopal, S., Srajer, V., Pahl, R., Anderson, S., Schmidt, M., Schotte, F., Anfinrud, P. A., Wulff, M., and Moffat, K. (2005). Visualizing reaction pathways in photoactive yellow protein from nanoseconds to seconds. Proc. Natl. Acad. Sci. USA 102, 7145–7150. Knapp, J. E., Pahl, R., Srajer, V., and Royer, W. E., Jr. (2006). Allosteric action in real time: Time-resolved crystallographic studies of a cooperative dimeric hemoglobin. Proc. Natl. Acad. Sci. USA 103, 7649–7654. Knapp, J. E., Srajer, V., Pahl, R., and Royer, W. E., Jr. (2004). Immobilization of Scapharca HbI crystals improves data quality in time-resolved crystallographic experiments. Micron 35, 107–108.
Structural Dynamics of Myoglobin
415
Laskowski, R. A. (1995). SURFNET: A program for visualizing molecular surfaces, cavities, and intermolecular interactions. J. Mol. Graph. 13, 323–330, 307–308. Lavalette, D., Tetreau, C., and Mouawad, L. (2006). Ligand migration and escape pathways in haem proteins. Biochem. Soc. Trans. 34, 975–978. Lim, M., Jackson, T. A., and Anfinrud, P. A. (1997). Ultrafast rotation and trapping of carbon monoxide dissociated from myoglobin. Nat. Struct. Biol. 4, 209–214. Nienhaus, K., and Nienhaus, G. U. (2004). The effect of protein internal cavities on ligand migration and binding in myoglobin. Micron 35, 67–69. Ostermann, A., Waschipky, R., Parak, F. G., and Nienhaus, G. U. (2000). Ligand binding and conformational motions in myoglobin. Nature 404, 205–208. Phillips, G. N., Jr, Arduini, R. M., Springer, B. A., and Sligar, S. G. (1990). Crystal structure of myoglobin from a synthetic gene. Proteins 7, 358–365. Rajagopal, S., Schmidt, M., Anderson, S., Ihee, H., and Moffat, K. (2004). Analysis of experimental time-resolved crystallographic data by singular value decomposition. Acta Crystallogr. D Biol. Crystallogr. 60, 860–871. Rasmussen, B. F., Stock, A. M., Ringe, D., and Petsko, G. A. (1992). Crystalline ribonuclease A loses function below the dynamical transition at 220 K. Nature 357, 423–424. Ren, Z. (2006). Precognition user guide with reference and tutorials. http://renzresearch. com/Precognition. Ren, Z., Bourgeois, D., Helliwell, J. R., Moffat, K., Srajer, V., and Stoddard, B. L. (1999). Laue crystallography: Coming of age. J. Synch. Radiat. 6, 891–917. Ren, Z., and Moffat, K. (1995a). Quantitative-analysis of synchrotron Laue diffraction patterns in macromolecular crystallography. J. Appl. Cryst. 28, 461–481. Ren, Z., and Moffat, K. (1995b). Deconvolution of energy overlaps in Laue diffraction. J. Appl. Cryst. 28, 482–493. Richards, F. M. (1974). The interpretation of protein structures: Total volume, group volume distributions and packing density. J. Mol. Biol. 82, 1–14. Schlichting, I., Almo, S. C., Rapp, G., Wilson, K., Petratos, K., Lentfer, A., Wittinghofer, A., Kabsch, W., Pai, E. F., Petsko, G. A., et al. (1990). Time-resolved X-ray crystallographic study of the conformational change in Ha-Ras p21 protein on GTP hydrolysis. Nature 345, 309–315. Schlichting, I., Berendzen, J., Phillips, G. N., Jr., and Sweet, R. M. (1994). Crystal structure of photolysed carbonmonoxy-myoglobin. Nature 371, 808–812. Schlichting, I., and Chu, K. (2000). Trapping intermediates in the crystal: Ligand binding to myoglobin. Curr. Opin. Struct. Biol. 10, 744–752. Schmidt, M., Nienhaus, K., Pahl, R., Krasselt, A., Anderson, S., Parak, F., Nienhaus, G. U., and Srajer, V. (2005). Ligand migration pathway and protein dynamics in myoglobin: A time-resolved crystallographic study on L29W MbCO. Proc. Natl. Acad. Sci. USA 102, 11704–11709. Schmidt, M., Rajagopal, S., Ren, Z., and Moffat, K. (2003). Application of singular value decomposition to the analysis of time-resolved macromolecular x-ray data. Biophys. J. 84, 2112–2129. Schotte, F., Lim, M., Jackson, T. A., Smirnov, A. V., Soman, J., Olson, J. S., Phillips, G. N., Jr., Wulff, M., and Anfinrud, P. A. (2003). Watching a protein as it functions with 150-ps time-resolved x-ray crystallography. Science 300, 1944–1947. Schotte, F., Soman, J., Olson, J. S., Wulff, M., and Anfinrud, P. A. (2004). Picosecond timeresolved X-ray crystallography: Probing protein function in real time. J. Struct. Biol. 147, 235–246. Scott, E. E., and Gibson, Q. H. (1997). Ligand migration in sperm whale myoglobin. Biochemistry 36, 11909–11917. Scott, E. E., Gibson, Q. H., and Olson, J. S. (2001). Mapping the pathways for O2 entry into and exit from myoglobin. J. Biol. Chem. 276, 5177–5188.
416
M. Brunori et al.
Srajer, V., Ren, Z., Teng, T. Y., Schmidt, M., Ursby, T., Bourgeois, D., Pradervand, C., Schildkamp, W., Wulff, M., and Moffa, K. (2001). Protein conformational relaxation and ligand migration in myoglobin: A nanosecond to millisecond molecular movie from time-resolved Laue X-ray diffraction. Biochemistry 401, 3802–3815. Srajer, V., Teng, T., Ursby, T., Pradervand, C., Ren, Z., Adachi, S., Schildkamp, W., Bourgeois, D., Wulff, M., and Moffat, K. (1996). Photolysis of the carbon monoxide complex of myoglobin: Nanosecond time-resolved crystallography. Science 274, 1726–1729. Stoddard, B. L., Cohen, B. E., Brubaker, M., Mesecar, A. D., and Koshland, D. E., Jr. (1998). Millisecond Laue structures of an enzyme-product complex using photocaged substrate analogs. Nat. Struct. Biol. 5, 891–897. Teeter, M. M. (2004). Myoglobin cavities provide interior ligand pathway. Protein Sci. 13, 313–318. Teng, T. Y., Srajer, V., and Moffat, K. (1997). Initial trajectory of carbon monoxide after photodissociation from myoglobin at cryogenic temperatures. Biochemistry 36, 12087–12100. Terwilliger, T. C., and Berendzen, J. (1995). Difference refinement: Obtaining differences between two related structures. Acta Crystallogr. D Biol. Crystallogr. 51, 609–618. Tilton, R. F., Jr., Kuntz, I. D., Jr., and Petsko, G. A. (1984). Cavities in proteins: Structure of a metmyoglobin-xenon complex solved to 1.9 A. Biochemistry 23, 2849–2857. Vallone, B., and Brunori, M. (2004). Roles for holes: Are cavities in proteins mere packing defects? Ital. J. Biochem. 53, 46–52. Ursby, T., and Bourgeois, D. (1997). Improved estimation of structure factor difference amplitudes from poorly accurate data. Acta Crystallogr. A 53, 564–575. Wulff, M., Plech, A., Eybert, L., Randler, R., Schotte, F., and Anfinrud, P. A. (2002). The realization of sub-nanosecond pump and probe experiments at the ESRF. Faraday Disc. 122, 13–26. Yeremenko, S., and Hellingwerf, K. J. (2005). Resolving protein structure dynamically. Structure (Camb.) 13, 4–6. Yeremenko, S., Van Stokkum, I. H., Moffat, K., and Hellingwerf, K. J. (2006). Influence of the crystalline state on photoinduced dynamics of photoactive yellow protein studied by ultraviolet-visible transient absorption spectroscopy. Biophys. J. 90, 4224–4235.
C H A P T E R
T W E N T Y- O N E
Use of the Conjugate Peak Refinement Algorithm for Identification of Ligand-Binding Pathways in Globins Stephen D. Golden and Kenneth W. Olsen Contents 418 418 419 420 421 422 425 425 425 426 428 428 429 429 429 430 431 431 432 433
1. 2. 3. 4. 5.
Introduction Exploration of Oxygen-Binding Pathways in Myoglobin Theoretical Models Potential Energy Function Transition Pathways 5.1. Conjugate peak refinement 6. Methods 6.1. Structure preparation 6.2. Energy calculations and minimization 6.3. Ligand coordinate generation 6.4. Running the CPR procedure 6.5. Fully refined pathways 6.6. Classification of ligand-binding pathways 7. Results 7.1. Myoglobin 7.2. Mt-trHbN 7.3. Pc-trHb 7.4. Ce-trHb 8. Conclusions References
Abstract Determination of the three-dimensional structures of the globins led to the problem of determining how the ligands bound to the heme. In many of these structures there was no clear path from the solvent to the ligand-binding site. Even in those structures that appeared to have one or more tunnels from the exterior to the heme, it was not clear that these were the only paths for ligand access. Department of Chemistry, Loyola University Chicago, Chicago, Illinois Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37021-3
#
2008 Elsevier Inc. All rights reserved.
417
418
Stephen D. Golden and Kenneth W. Olsen
Conjugate peak refinement (CPR) is a computational technique that can provide a minimum energy path between two conformations of a protein. By defining one conformation as the unbound structure with an external ligand and the other as the ligand-bound structure, CPR can be used to determine a pathway for ligand binding.
1. Introduction This chapter reviews the use of conjugate peak refinement (CPR), a computational method for the study of ligand binding in globins. The globins contain ligand-binding sites that are buried in the interior of the protein. Some also appear to have tunnels that could potentially provide access to these binding sites. Determining the paths that ligands take to heme is a significant problem. The conjugate peak refinement method for determining multidimensional reaction pathways has been used to investigate the paths for ligand binding to sperm whale myoglobin and group I truncated hemoglobins from the eubacteria Mycobacterium tuberculosis, the ciliated protozoa Paramecium caudatum, and the unicellular alga Chlamydomonas eugametos. The method is general and could be applied to other globins.
2. Exploration of Oxygen-Binding Pathways in Myoglobin Myoglobin serves as a model for studying the dynamics of ligand binding in hemoproteins, with information pertaining to the structural, spectroscopic, and kinetic characteristics of the protein being readily available. It is also utilized as a model system for molecular engineering studies to investigate the structure–function relationships of hemoenzymes (Ozaki et al., 2000, 2001). It was evident after the discovery of the three-dimensional structure (Kendrew et al., 1960) that there is no visible opening into the hydrophobic-binding pocket large enough for a diatomic ligand to pass. X-ray diffraction and spectroscopic methods (Bourgeois et al., 2003; Brunori et al., 2000; Chu et al., 2000; Hartmann et al., 1996; Nienhaus et al., 2003a,c, 2005; Ostermann et al., 2000; Schmidt et al., 2005; Schotte et al., 2003, 2004; Sˇrajer et al., 2001; Teng et al., 1997; Tetreau et al., 2004; Vojtechovsky et al., 1999), ligand-binding kinetics (Austin et al., 1975; Dantsker et al., 2004; Gibson et al., 1986, 1992; Huang and Boxer, 1994; Ishikawa et al., 2001; Olson and Phillips, 1996; Scott and Gibson, 1997; Scott et al., 2001), and molecular dynamics calculations (Banushkina et al., 2005; Bossa et al., 2004, 2005; Carlson et al., 1994, 1996; Case and Karplus, 1979; Cohen et al., 2006; Elber and Karplus, 1990; Gibson et al., 1992;
Ligand-Binding Pathways in Globins
419
Hummer et al., 2004; Ma et al., 1997; Meller and Elber, 1998; Nutt et al., 2004; Quillin et al., 1995) have had much success in revealing passageways for ligands such as O2, CO, and NO to diffuse to and from the heme of wild-type and mutant myoglobins. Packing defects within the myoglobin structure play important roles in ligand binding and are well characterized (Brunori, 2000; Brunori and Gibson, 2001; Frauenfelder et al., 2001; Nienhaus et al., 2003a,b,c, 2005; Schotte et al., 2004; Tilton et al., 1984). Theoretical models have been validated as useful tools for studying the dynamic nature of ligand diffusion through the myoglobin. Molecular dynamics simulations and packing defect studies have suggested various ligand-binding pathways for myoglobin. These studies have examined ligand diffusion starting from within the binding cavity traveling into the solvent. Truncated hemoglobins are small heme proteins found in bacteria, unicellular eukaryotes, and higher plants, forming a group within the hemoglobin superfamily (Watts et al., 2001). Three specific truncated hemoglobins from the eubacteria M. tuberculosis (Mt-trHbN), the ciliated protozoa P. caudatum (Pc-trHb), and the unicellular alga C. eugametos (Ce-trHb) have been studied using CPR. trHbs have less than 15% sequence identity when compared with vertebrate and nonvertebrate hemoglobins (Couture et al., 1994; Moens et al., 1996). Many trHbs display amino acid sequences 20–40 residues shorter than nonvertebrate hemoglobins to which they are scarcely related by sequence similarity (Wittenberg et al., 2002). When compared to myoglobin, sequence alignments reveal residue deletions at N or C termini and in the CD-D region of the nonvertebrate globin fold. Ligand binding can also be studied using an algorithm capable of determining the lowest energy pathway from reactants to products. The algorithm is the conjugate peak refinement method (Fischer and Karplus, 1992), as implemented in the molecular mechanics program CHARMM (Brooks et al., 1983). CPR has been used to calculate the binding pathways for O2 traveling into the distal pocket of myoglobin. A sphere of 73 O2 molecules was positioned, in solvent, around the deoxymyoglobin starting structure and binding trajectories were calculated for each of these starting positions. Thirty-nine plausible pathways were categorized into nine families of paths. Further, less extensive CPR studies of ligand-binding paths for three group I truncated hemoglobins have indicated at least four common pathways.
3. Theoretical Models Theoretical calculations date back to the 1940s with the first attempts at molecular mechanical calculations performed in 1946 by Frank Westheimer with calculation of the relative racemization rates of biphenyl derivatives (Schlick, 2000). There has been great progress since then due to faster
420
Stephen D. Golden and Kenneth W. Olsen
computers and significantly improved algorithms. With these improvements comes the ability to explore larger molecular structures, more specifically biomolecules. Biomolecular modeling started in the 1960s with the development of molecular mechanical force-field methodologies. Computational techniques cover a wide range, including ab initio, semiempirical, and molecular mechanical calculations applied to molecular dynamics (MD), Monte Carlo, free energy and solvation methods, structure/activity relationships, and X-ray and nuclear magnetic resonance (NMR) structure refinement. The appropriate model must therefore be selected based on the question being asked. Although recent improvements to algorithms allow the use of quantum mechanical methods to study key structural features of large biomolecular systems, it would not be feasible to select a purely quantum mechanical model to study a system with a very large number of degrees of freedom. It would also not be practical to select molecular mechanics to investigate nucleophilic additions to cyclohexanone, when quantum mechanical models would produce more detailed information about the reaction.
4. Potential Energy Function The current study uses molecular mechanics for the basis of the calculations. Molecular mechanics uses the potential energy function to calculate the potential energy of a chemical system. A typical potential energy function is as follows:
X X ðVÞ ¼ bonds Kb ðb b0 Þ2 þ K ðY Y0 Þ2 angles y X X þ jK j K cosðn Þ þ K ðo o0 Þ2 f f dihedral f improper o " ! !# X X Aij Bij qi qj þ þ 6 12 VDW Electrostatic eR Rij Rij ij ð21:1Þ For the bonded interactions, the first term represents the covalent bond energy with an empirically determined force constant Kb, the current bond length b, and the equilibrium bond length b0. The second term models the bond angle energy with Ky being the empirically determined constant, y the current bond angle, and y0 the equilibrium bond angle. The next two terms account for proper and improper torsions. Torsions are created by three bonds. K’ and Ko are both empirically determined constants, n is the periodicity, w and o are the torsional angles, and o0 is the equilibrium angle.
421
Ligand-Binding Pathways in Globins
The last two terms in the potential energy function represent nonbonded interactions. The fifth term accounts for the van der Waals interactions with both repulsive and attractive forces. This is normally modeled using the Lennard–Jones 6–12 function, as seen in Eq. (21.1). The final term accounts for electrostatic interactions using Coulomb’s law: qi and qj are the charges of the atoms, e is the effective dielectric constant, and Rij is the distance between the two atoms. More terms can be added to the function depending on the potential energy function used (MacKerell et al., 1998). Summing all the energy terms over all interaction pairs equals the potential energy of the molecular system in a given conformation. The calculated potential energy of a molecular system will be utilized in studying the ligand binding to heme proteins. The CHARMM potential energy function was used for the CPR studies.
5. Transition Pathways Computationally, the problem of finding a likely transition pathway between a reactant and a product corresponds to identifying a low-energy pathway between them on the modeled potential energy surface of the protein. The path can be represented by a series of structures connecting the reactant and product end states, which are known from X-ray crystallographic or NMR studies. The intermediate points along the path have to be found. After obtaining an initial guessed path, the intermediate points can be optimized locally using method-dependent criteria to obtain a plausible reaction pathway (Noe´ et al., 2003). Penalty function and heuristic methods can be utilized to explore these transition pathways. Information obtained from these methods gives insight on processes that occur on timescales too long for standard MD simulations (Noe´ et al., 2003). One broad class of methods used for studying transition pathways are penalty functions. Penalty functions assign a cost or penalty for infeasibility and force the solution to feasibility. There have been a variety of proposed definitions of the cost function: the self-penalty walk method, the nudged elastic band method, and the MaxFlux algorithm (Czerminski and Elber, 1990; Elber and Karplus, 1987; Huo and Straub, 1999; Mills and Jonsson, 1994). These methods define a path cost function C:
C¼
M 1 X
cðrk Þjrkþ1 rk j
ð21:2Þ
k¼0
where cðrk Þjrkþ1 rk j assigns a penalty for displacement from position by an increment of jrkþ1 rk j. Using techniques such as steepest descent, conjugate gradients, or simulated annealing, the initial guess path can be
422
Stephen D. Golden and Kenneth W. Olsen
improved by minimizing its cost function in the space of all possible paths (Noe´ et al., 2003). A number of constraints and restraints are also needed in the cost function to create a meaningful tool to explore transition pathways, for example, constraints to remove the relative rigid-body translations and rotations of the structures along the path and self-avoidance terms to prevent the path from folding back upon itself (Becker et al., 2001; Noe´ et al., 2003). The second class of methods used for studying transition pathways is heuristic methods. Heuristic methods follow a specific set of rules to improve the initial path. These methods have the advantage over penalty methods because they can be made more efficient by being able to spend more optimization effort on the saddle points than on other points. The heuristic method of choice for this study is the conjugate peak refinement algorithm (Fischer and Karplus, 1992).
5.1. Conjugate peak refinement In order to study ligand binding in myoglobin and truncated hemoglobins, a robust method efficient in very large biomolecular systems is needed. CPR, as implemented in the TRAVEL module in the CHARMM program (Brooks et al., 1983), is a heuristic method for determining multidimensional reaction coordinates between two known conformations, reactant and product, from an initial path. The low-energy reactant and product can be thought of as existing at the bottom of two wells on the surface of an energy landscape. The path connecting these two wells travels along valleys and over passes, saddle points, creating a minimum energy path (MEP). The saddle points on the energy surface correspond to the transition states. No assumption is required concerning the pathway because the outcome of the simulation is the MEP between reactant and product. CPR has a simple set of rules to add, remove, or refine a path point on each cycle to eventually produce a MEP. CPR identifies those points along the path where the energy is highest and moves those points closer to the MEP by a controlled conjugate-gradient minimization. The algorithm does not evaluate second derivatives but uses only the energy and its gradient. The refinement ends when all energy maxima along the path are identified as saddle points. This algorithm requires a one-to-one correspondence of atoms in the reactant and product. For this reason, no explicit water can be included in CPR calculations. If you try to place the protein in a water box, as is usually done for MD, then you must know which water molecule in the product water box corresponds to each water molecule in the reactant water box. This is impossible to determine. If you superimpose the two water boxes and assume that the closest ones represent the same water molecule, then the protein is effectively in a solid and its movements are restricted.
Ligand-Binding Pathways in Globins
423
Thus, implicit models must be used for CPR. The most common method has been the use of a distance-dependent dielectric constant combined with nonuniform scaling of the partial atomic charges (Blondel et al., 1999; Dutzler et al., 2002; Santos et al., 2000) but more sophisticated models can also be used (Gruia et al., 2005; Khavrutskii et al., 2006). The conjugate peak refinement algorithm has been utilized for both small molecule and macromolecular systems. The small molecule studies have included isomerizations of prolyl dipeptides (Fischer et al., 1994), interconversions of calyxarenes (Fischer et al., 1995), and rotating the aromatic ring of methotrexate bound to dihydrofolate dehydrogenase (Verma et al., 1996). The macromolecular studies, which are more relevant to this chapter, include studies on the FK506-binding protein (Caflisch et al., 1997; Fischer et al., 1993), the allosteric transition of hemoglobin (Olsen et al., 2000), the conformational changes of annexin (Santos et al., 2000), retinoic acid binding to its receptor (Blondel et al., 1999), transport of maltodextrin through maltoporin (Dutzler et al., 2002), and the pathway for chloride ion movement in halorhodopsin (Gruia et al., 2005). The following is a brief summary of the CPR algorithm. More mathematical and detailed descriptions can be found elsewhere (Fischer and Karplus, 1992; Noe´ et al., 2003, 2006). CPR requires three-dimensional structures for the reactant and the product. These must have been determined by X-ray crystallography or NMR spectroscopy and can be found in the Protein Data Bank (PDB) (Berman et al., 2000; www.rcsb.org). It also requires an energy function, such as the molecular mechanics one given in Eq. (21.1), that can be differentiated to provide a gradient. In the absence of any structural information concerning intermediates, the procedure starts with a linear path across the energy landscape connecting the reactant and the product, which are in energy minima (Fig. 21.1A). During refinement, points corresponding to intermediate structures are added, removed, or improved so that the resulting path avoids high-energy conformations. At the beginning of a simulation, the initial path is searched for its energy maximum (see point C in Fig. 21.1B). The conjugate direction is then searched locally for the lowest energy structure (see point D in Fig. 21.1B). Unless the original maximum is a saddle point, there must be a lower energy structure along this direction in conformational space. In the early stages of a path refinement, the initial path is a very poor representation of the actual path and the energies of the structures can be very large as a consequence of the van der Waals repulsions of overlapping atoms. The energy gradient generated in this case would result in a significant distortion of the molecule. To avoid this situation, atomic movements are limited to a reasonable ˚ ) on any cycle of refinement. The new point (see D distance (usually 1 A in Fig. 21.1B) is added to the path, which is now A-D-B. The procedure is repeated along the intermediate segments (A-D and D-B) to generate additional points along the path. In a simple case, such as that of Fig. 21.1,
424
Stephen D. Golden and Kenneth W. Olsen
Figure 21.1 Explanation of CPR. (A) This energy landscape has two minima (A and B), representing the globin with unbound and bound ligand, respectively.The initial path is a straight line across the energy landscape connecting the two minima. (B) Point C represents the maximum along the path AB. A search for an energy minimum is done in the conjugate direction, which gives point D. The new path is ADB. (C) The process is repeated on line segments AD and DB to obtain additional points on the path. (D) The final path is obtained by repeated use of this algorithm and by elimination and refinement of some of the path points as described in the text.
a reasonable representation of the minimum energy path is obtained easily (see Figs. 21.1C and 21.1D). In reality, the process is more complex and requires additional steps. On the high dimensional energy surface of a protein, a newly added path point is rarely at the energy minimum after a single cycle of CPR. In later cycles, the energy of intermediate structures can be lower by a process of local refinement. If the point is a local energy peak along the path, then tangent to the path at that point is searched for a local energy maximum. Such a maximum would be present if that portion of the path either crosses a saddle region or goes up and down a convex portion of the energy surface. If no maximum is found, then the point is eliminated from the path. If one is found, then the conjugate direction is searched for an energy minimum.
Ligand-Binding Pathways in Globins
425
This new energy minimum point replaces the original one in the path. Local refinement of a globally defined reaction path is a major advantage of the CPR method and leads to a minimum energy pathway on which all of the energy maxima represent saddle (transition-state) points.
6. Methods 6.1. Structure preparation The atomic coordinates of all structures used in the simulations are obtained from the PDB (Berman et al., 2000; www.rcsb.org). The Mb deoxy structure (PDB: 1A6N) used is obtained from X-ray diffraction with a 1.15-A˚ resolution and the oxy structure (PDB: 1A6M) at a 1.0-A˚ resolution (Vojtechovsky et al., 1999), the Mt-trHbN structure at a 1.9-A˚ resolution (Milani et al., 2001) (PDB: 1idr), the Pc-trHb structure at a 1.54-A˚ resolu˚ tion (Pesce et al., 2000) (PDB: 1dlw), and the Ce-trHb structure at a 1.80-A resolution (Pesce et al., 2000) (PDB: 1dly). Because CPR requires a one-toone correspondence of atoms in the reactant and product, the PDB files are edited to remove the water molecules and sulfate ions from the structures. The PDB files for bound and unbound conformations are checked in order to ensure the consistency of atom naming within side chain residues. Any naming inconsistencies could cause unrealistic motions such as rotations. For example, a phenylalanine residue has delta carbons labeled CD1, CD2 and epsilon carbons CE1, CE2 in the PDB file. If the atom labels are not the same in each conformation, CPR will try to rotate the phenylalanine to alleviate the naming problem. In order to calculate the minimum energy pathway using CPR, structures of both the reactant and the product are needed. These structures correspond to the unbound and ligand bound forms of Mb, Mt-trHbN, Pc-trHb, and Ce-trHb. Because the unbound structures for Mt-trHbN, Pc-trHb, and Ce-trHb have not yet been solved, coordinates for the unbound forms are generated by moving the ligand to various positions outside the protein surface. The B subunit of Mt-trHbN is also removed and simulations are only performed on the A subunit. Missing residues in MttrHbN are built back into the structure using the SwissPDB viewer (Guex and Peitsch, 1997). The missing residues are Met 1, Gly 129, Glu 130, Ser 131, Thr 132, Thr 133, Ala 134, Pro 135, and Val 136.
6.2. Energy calculations and minimization All calculations are performed using Version 27 of the program CHARMM (Brooks et al., 1983). Parameters (PARAM 19, polar hydrogen set) are used for the protein because nonpolar hydrogens tend to behave poorly in CPR simulations. All charged residues are modeled in their respective charged
426
Stephen D. Golden and Kenneth W. Olsen
state with the exception of histidine, which is modeled as neutral. Charges on the charged residues are reduced by a factor of 5 to approximate the linearized Poisson–Boltzmann model (Santos et al., 2000). After performing an HBUILD to add polar hydrogens to the structures, the energy of both structures is minimized using 200 steps of the steepest descent algorithm followed by 2000 steps of the conjugate gradient algorithm. A distancedependent dielectric is used, and both electrostatic and van der Waals interactions are truncated with a shifting function at 8.5 A˚ over a distance ˚ . A saddle gradient of 0.001 kcal/(molA ˚ ) and a step size of 0.002 A ˚ are of 1.0 A used. The refinement cycle is continued until all trajectories are fully refined.
6.3. Ligand coordinate generation The position assigned to the ligand in the unbound structure will greatly affect the results of the CPR simulation. Because the ligand could be approaching the globin from any direction, the best approach is to try a large number of possible starting positions. A sphere of dioxygens surrounding Mb is created to examine the entire protein surface for possible binding paths. An initial dioxygen coordinate is generated by placing an O2 molecule ˚ , in the z direction, away from the O2 of the energyat a distance 35 A minimized oxymyoglobin structure. Thirty-five angstroms is chosen for the radius of the sphere, as it allows all dioxygens to be at a distance where they do not come in contact with the outer surface of the protein. This starting coordinate is then used to calculate the remaining 72 O2 positions. Cartesian coordinate positions are converted to spherical coordinates, where f ¼ 0 to ˚ . The initial oxygen position, position 1, is 180 , y ¼ 0 to 330 , and r ¼ 35 A ˚ . Moving y in 30 increments up to 330 set as f ¼ 0 , y ¼ 0 , and r ¼ 35 A before increasing f by 30 generates a sphere of O2 molecules around the protein (Fig. 21.2). Each dioxygen position requires the creation of separate coordinate files, which are then used in 73 CPR calculations. Seven different unbound ligand positions are generated for each Mt-trHbN, Pc-trHb, and Ce-trHb to evaluate specific potential pathways for ligand binding. For Mt-trHbN, two of the dioxygen starting positions are placed just outside the entrance of the two channels proposed by Milani ˚ away et al. (2001). One dioxygen molecule is placed at a distance 19 A from the position of the dioxygen of the minimized oxy Mt-trHbN structure. This position is at a distance 10 A˚ outside of the G helix/H helix ˚ away from the tunnel. The second dioxygen molecule is placed 32 A position of the dioxygen of the minimized oxy Mt-trHbN structure and ˚ outside of the AB/GH turn tunnel. The remaining five dioxygen set 12 A starting positions are placed at points around the protein that appear to have clear or semiclear pathways through the protein into the binding pocket. Distances from the deoxy state at which the ligands are placed outside the
Ligand-Binding Pathways in Globins
427
Figure 21.2 Sphere of oxygen molecules around myoglobin. Each of the 73 O2 molecules in the sphere can serve a separate starting pointing for generation of a CPRbinding pathway.
˚ . Distances from the starting position to protein range from 19 to 35 A ˚. the protein surface range from 4 to 25 A In the X-ray structure of Pc-trHb the ligand is water. One of the water molecule starting positions is placed just outside the entrance of the channel proposed by Pesce et al. (2000). This water molecule is placed at a distance ˚ away from the position of the bound water molecule of the mini40 A mized Pc-trHb structure. This position is at a distance 18 A˚ outside the tunnel. The remaining six ligand starting positions are placed at points around the protein that appear to have a clear or semiclear pathway through the protein into the binding pocket. Distances from the bound state at ˚. which the ligands are placed outside the protein range from 23 to 44 A Distances from the starting position to the protein surface range from ˚. 12 to 23 A For Ce-trHb, one of the cyanide anion starting positions is placed just outside the entrance of the channel proposed by Pesce et al. (2000). This ˚ away from the position of the cyanide anion is placed at a distance 41 A bound cyanide anion of the minimized Ce-trHb structure. This position is at a distance 20 A˚ outside the tunnel. The remaining six ligand starting positions are placed at points around the protein that appear to have a clear or semiclear pathway through the protein into the binding pocket. Distances from the bound state at which the ligands are placed outside the ˚ . Distances from the starting position to protein range from 27 to 47 A ˚. the protein surface range from 15 to 30 A
428
Stephen D. Golden and Kenneth W. Olsen
6.4. Running the CPR procedure The CHARMM program has been described elsewhere (Brooks et al., 1983). An example script to run a CPR calculation for oxygen binding to myoglobin is shown. The mb_gene_o2.str file contains the amino acid sequence for the protein, the patch for attaching the heme to the proximal histidine, the definition of the oxygen molecule, and the file names for the parameter and topology files needed by the program. An energy determination is done before the TRAVEL (CPR) routine is called to properly set up the parameters for the energy determinations during CPR. The two coordinate (CRD) files represent the beginning and ending structures on the path. For subsequent runs, these are replaced by the trajectory (DCD) file from the previous run. The REFINE statement can be repeated numerous times. It is a good idea to save intermediate trajectories, as shown. A complete description of the input parameters is given in the CHARMM documentation (www.charmm.org). * TRAVEL.STR * STRE mb_gene_o2.str open unit 2 form read name 1A6N_O2_POS_66.CRD read coor card unit 2 close unit 2 ENER SHIFT VSHIFT CUTNB 8.5 CTOFNB 7.5 CTONNB 4.3 – RDIEL EPS 1.0 E14FAC 0.4 ATOM VATOM WMIN 0.6 TRAVEL MAXP 900 TRAJ READ 1A6N_O2_POS_66.CRD MIN_1A6M_CHARMM.CRD DONE ! TRAJ READ NAME mb_pos_66_10050_c66.out !1 REFINE NCYCLE 50 STEP 0.002 LOOP 0 SADGRA 0.001 SADCYC 50 TRAJ WRITE NAME mb_temp.dcd TRAJ ANAL
6.5. Fully refined pathways In order to monitor the progress of the simulations, the reaction pathway is plotted. The energy and root mean-squared distance (RMSD) from the reactant structure are plotted against the step number or the percentage of the total path length with the determined saddle points indicated. Figure 21.3 is an example of one of the fully refined reaction pathways.
429
Ligand-Binding Pathways in Globins
% Complete 10% 22% 34% 45% 60% 72% 79% 89%
−2950 −3000 −3050 −3100
1.20 1.00 0.80 0.60 0.40
RMS (Å)
Energy (kcal/mol)
0% −2900
0.20 0.00
Figure 21.3 Example of myoglobin fully refined pathway. This is a plot of the fully refined pathway for the binding of O2 to myoglobin.The blue line shows energy versus percent completeness of the pathway with saddle points indicated by red diamonds. The green line shows the RMSD comparing the initial structure to the current structure in the path.
6.6. Classification of ligand-binding pathways The pathways are placed into families or groups of paths by the helices through which the ligand travels to bind to the heme. A contact analysis is conducted to define the binding pathways. The contact ‘‘collision’’ is based on the distance between ligand and protein heavy (nonhydrogen) atoms of ˚ . The collision radius is a mean value based on van der Waals less than 4 A radii of the heavy atoms (Elber and Karplus, 1990). The contact analysis script shown here is used within VMD (Humphrey et al., 1996; www.ks. uiuc.edu/Research/vmd/www.ks.uiuc.edu/Research/vmd/) to identify the residue contacts. proc contact set sel [atomselect top ‘‘protein and same residue as (within 4 of resname O2)’’] set n [molinfo top get numframes] for { set i 0 } { $i < $n } { incr i } { $sel frame $i $sel update $sel writepdb pos15_$i.pdb
7. Results 7.1. Myoglobin Analysis of trajectories resulted in identification of 39 dioxygen positions out of the 73 positions examined that have a plausible pathway. The dioxygen-binding pathway can be defined by the helices and turns
430
Stephen D. Golden and Kenneth W. Olsen
that it moves past when it enters the myoglobin molecule. Nine families of pathways were found within the 39 energetically reasonable paths. The families are as follows, with the number of O2 starting positions following in parentheses. Pathways containing more than two ligands traveling the path are the CD loop (10), the histidine gate (9), the C helix/FG turn (6), and the B helix/G helix (5). Paths with two or less ligands traveling the path are the A helix/GH turn (2), the GH turn (2), the B helix/E helix (2), the A helix/E helix (2), and the E helix/F helix (1). Figure 21.4 shows members of the nine families of paths.
7.2. Mt-trHbN Results for the two trajectories for positions placed outside of the AB turn/ GH turn and the G helix/H helix tunnel branches reveal two pathways in which the dioxygen diffuses to the heme through either tunnel with minimal side chain fluctuations guiding the ligand to the heme-binding pocket. Three other tunnel branches were identified: E helix/H helix, B helix/E helix, and C helix/G helix (Fig. 21.5). Phe 32, Tyr 33, and Phe 62 have been suggested as possible gating residues that determine which tunnels are used (Crespo et al., 2005; Dantsker et al., 2004). CPR studies confirm this conclusion. For example, when Phe 62 moves so that it closes the AB turn/GH turn tunnel, it simultaneously opens the E helix/H helix tunnel branch.
Figure 21.4 Nine families of oxygen-binding paths for myoglobin. One example is shown for each of the nine families of oxygen-binding pathways found for myoglobin using CPR.
Ligand-Binding Pathways in Globins
431
Figure 21.5 Five oxygen-binding paths for Mt trHbN found by CPR.The five binding pathways are AB turn/GH turn (red), G helix/H helix (blue), E helix/H helix (green), C helix/G helix (purple), and B helix/E helix (orange, shown from three different starting points). The key residues that gate these pathways are Tyr 33 (green), Gln 58 (orange), Phe 32 (purple), and Phe 62 (red).
7.3. Pc-trHb Analysis of the resulting trajectory for the apolar tunnel branch confirmed a pathway in which the water molecule diffuses to the heme through the tunnel with minimal side chain fluctuations guiding the ligand into the heme-binding pocket. The first pathway explored using CPR was the apolar tunnel proposed by Pesce et al. (2000). The tunnel branch entrance is located at the AB turn/GH turn. Five new pathways were identified: CD turn/G helix, E helix/heme, E helix/H helix, AB turn, and B helix/E helix.
7.4. Ce-trHb Analysis of the resulting trajectory for the apolar tunnel branch confirmed a pathway in which the cyanide anion diffuses to the heme through the tunnel with minimal side chain fluctuations guiding the ligand into the heme-binding pocket. Five new pathways were identified: the BC turn/G
432
Stephen D. Golden and Kenneth W. Olsen
helix path, the G helix/H helix path, the EF turn/E helix path, the CD loop pathway, and the B helix/E helix path.
8. Conclusions Comparison of CPR results for ligand binding to the four small globins is shown in Table 21.1. At least three paths (AB turn/GH turn, B helix/E helix, and C helix/G helix) are found in all four proteins, whereas several other paths (G helix/H helix, CD loop, and E helix/F helix) are found in more than one of them. Most of these paths are also predicted by implicit ligand sampling and locally enhanced sampling molecular dynamics (see the next two chapters). The fact that similar paths for ligand binding are found by these very different computational algorithms suggests that the level of confidence in these results is high. The major advantage of the CPR method is that it can provide a detailed pathway for ligand binding to proteins. The path determined should have only low-energy barriers. Those paths that do not fit this criterion can be rejected. By generating a minimum energy path, CPR provides the transition states of the binding process. There are two major problems with the application of CPR to ligand binding. First, because the unbound position of the ligand is not known, many potential starting positions must be Table 21.1 Comparison of CPR results for four globins Mb
A helix/GH turn GH turn
Mt-trHbN
AB turn/GH turna G helix/H helixa E helix/H helixa B helix/E helix B helix/E helixa C helix/FG C helix/G helix turn CD loop A helix/EF turn E helix/F helix Histidine gate B helix/G helix
Pc-trHb
Ce-trHb
AB turn/GH turna
AB turn/GH turna G helix/H helixa E helix/H helixb B helix/E helix BC turn/G helix
E helix/H helixa B helix/E helix CD turn/G helixa
CD loop AB turn EF turn/E helix
E helix/hemea a b
Pathways are observed in LESMD simulations as well as CPR. Pathway was only observed in LESMD simulation.
Ligand-Binding Pathways in Globins
433
simulated to provide a complete understanding of the possible binding paths. Second, using a linear extrapolation between reactant and product structures on a complex energy surface as the initial guess for the path often generates transition state energies that are unrealistically high (Noe´ et al., 2006). Finally, the CPR path does not represent the actual trajectory that would be taken by a ligand during binding due to inertial and frictional effects. It is rather a representation of a family of trajectories that has been determined in the absence of those thermal motions not essential to the binding process.
REFERENCES Austin, R. H., Beeson, K. W., Eisenstein, L., and Frauenfelder, H. (1975). Dynamics of ligand binding to myoglobin. Biochemistry 14, 5355–5373. Banushkina, P., and Meuwly, M. (2005). Free-energy barriers in MbCO rebinding. J. Phys. Chem. B 109, 16911–16917. Becker, O. M., MacKerell, A. D., Jr., Roux, B., and Watanabe, M. (2001). ‘‘Computational Biochemistry and Biophysics.’’ New York: Dekker, New York. Berman, H. M., Westbrook, J., Feng, Z., Gilliland, G., Bhat, T. N., Weissig, H., Shindyalov, I. N., and Bourne, P. E. (2000). The protein databank. Nucleic Acids Res. 28, 235–242. Blondel, A., Renaud, J. P., Fischer, S., Moras, D., and Karplus, M. (1999). Retinoic acid receptor: A simulation analysis of retinoic acid binding and the resulting conformational changes. J. Mol. Biol. 291, 101–115. Bossa, C., Amadei, A., Daidone, I., Anselmi, M., Vallone, B., Brunori, M., and Di Nola, A. (2005). Molecular dynamics simulation of sperm whale myoglobin: Effects of mutations and trapped CO on the structure and dynamics of cavities. Biophys. J. 89, 465–474. Bossa, C., Anselmi, M., Roccantano, D., Amadei, A., Vallone, B., Brunori, M., and Di Nola, A. (2004). Extended molecular dynamics simulation of the carbon monoxide migration in sperm whale myoglobin. Biophys. J. 86, 3855–3862. Bourgeois, D., Vallone, B., Schotte, F., Arcovito, A., Miele, A. E., Sciara, G., Wulff, M., Anfinrud, P., and Brunori, M. (2003). Complex landscape of protein structural dynamics unveiled by nanosecond Laue crystallography. Proc. Natl. Acad. Sci. USA 100, 8704–8709. Brooks, B. R., Bruccoleri, R. E., Olafson, B. D., States, D. J., Swaminathan, S., and Karplus, M. (1983). CHARMM: A program for macromolecular energy, minimization, and dynamics calculations. J. Comput. Chem. 4, 187–217. Brunori, M. (2000). Structural dynamics of myoglobin. Biophys. Chem. 86, 221–230. Brunori, M., and Gibson, Q. (2001). Cavities and packing defects in the structural dynamics of myoglobin. EMBO Rep. 2, 674–679. Caflisch, A., Fischer, S., and Karplus, M. (1997). Docking by Monte Carlo minimization with a solvation correction: Application to an FKBP-substrate complex. J. Comput. Chem. 18, 723–743. Carlson, M. L., Regan, R., Elber, R., Li, H., Phillips, G. N., Jr., Olson, J. S., and Gibson, Q. H. (1994). Nitric oxide recombination to double mutants of myoglobin: Role of ligand diffusion in a fluctuating heme pocket. Biochemistry 33, 10597–10606. Carlson, M. L., Regan, R. M., and Gibson, Q. H. (1996). Distal cavity fluctuations in myoglobin: Protein motion and ligand diffusion. Biochemistry 35, 1125–1136.
434
Stephen D. Golden and Kenneth W. Olsen
Case, D. A., and Karplus, M. (1979). Dynamics of ligand binding to heme proteins. J. Mol. Biol. 132, 343–368. Chu, K., Vojtchovsky, J., McMahon, B. H., Sweet, R. M., Berendzen, J., and Schlichting, I. (2000). Structure of a ligand-binding intermediate in wild-type carbonmonoxy myoglobin. Nature 403, 921–923. Cohen, J., Arkhipov, A., Braun, R., and Schulten, K. (2006). Imaging the migration pathways for O2, CO, NO, and Xe inside myoglobin. Biophys. J. 91, 1844–1857. Couture, M., Chamberland, H., St.Pierre, B., Lafontaine, J., and Guertin, M. (1994). Nuclear genes encoding chloroplast hemoglobins in the unicellular green alga Chlamydomonas eugametos. Mol. Gen. Genet. 243, 185–197. Crespo, A., Marti’, M. A., Kalko, S. G., Morreale, A., Orozco, M., Gelpi, J. L., Luque, F. J., and Estrin, D. A. (2005). Theoretical study of the truncated hemoglobin HbN: Exploring the molecular basis of the NO detoxification mechanism. J. Am. Chem. Soc. 127, 4433–4444. Czerminski, R., and Elber, R. (1990). Reaction path study of conformational transitions in flexible systems: Applications to peptides. J. Chem. Phys. 92, 5580–5601. Dantsker, D., Samuni, U., Ouellet, Y., Wittenberg, B. A., Wittenberg, J. B., Milani, M., Bolognesi, M., Guertin, M., and Friedman, J. M. (2004). Viscosity-dependent relaxation significantly modulates the kinetics of CO recombination in the truncated hemoglobin TrHbN from mycobacterium tuberculosis. J. Biol. Chem. 279, 38844–38853. Dutzler, R., Schirmer, T., Karplus, M., and Fischer, S. (2002). Translocation mechanism of long sugar chains across the maltoporin membrane channel. Structure 10, 1273–1284. Elber, R., and Karplus, M. (1987). A method for determining reaction paths in large molecules: Application to myoglobin. Chem. Phys. Lett. 139, 375–380. Elber, R., and Karplus, M. (1990). Enhanced sampling in molecular dynamics: Use of the time-dependent hartree approximation for a simulation of carbon monoxide diffusion through myoglobin. J. Am. Chem. Soc. 112, 9161–9175. Fischer, S., Dunbrack, R., and Karplus, M. (1994). Cis-Trans imide isomerization of the proline dipeptide. J. Am. Chem. Soc. 116, 11931–11937. Fischer, S., Grootenhuis, P. D. J., Groenen, L. C., vanHoorn, W. P., vanVeggel, F., Reinhoudt, D., and Karplus, M. (1995). Pathways for conformational interconversion of calyx[4]arenes. J. Am. Chem. Soc. 117, 1611–1620. Fischer, S., and Karplus, M. (1992). Conjugate peak refinement: An algorithm for finding reaction paths and accurate transition states in systems with many degrees of freedom. Chem. Phys. Lett. 194, 252–261. Fischer, S., Michnick, S., and Karplus, M. (1993). A mechanism for rotamase catalysis by the FK506 binding protein (FKBP). Biochemistry 32, 13830–13837. Frauenfelder, H., McMahon, B. H., Austin, R. H., Chu, K., and Groves, J. T. (2001). The role of structure, energy landscape, dynamics, and allostery in the enzymatic function of myoglobin. Proc. Natl. Acad. Sci. USA 98, 2370–2374. Gibson, Q. H., Olson, J. S., McKinnie, R. E., and Rohlfs, R. J. (1986). A kinetic description of ligand binding to sperm whale myoglobin. J. Biol. Chem. 261, 10228–10239. Gibson, Q. H., Regan, R., Elber, R., Olson, J. S., and Carver, T. E. (1992). Distal pocket residues affect picosecond ligand recombination in myoglobin. J. Biol. Chem. 267, 22022–22034. Gruia, A. D., Bondar, A.-N., Smith, J. C., and Fischer, S. (2005). Mechanism of a molecular valve in the halorhodopsin chloride pump. Structure 13, 617–627. Guex, N., and Peitsch, M. C. (1997). SWISS-MODEL and the Swiss-pdbviewer: An environment for comparative protein modeling. Electrophoresis 18, 2714–2723. Hartmann, H., Zinser, S., Komninos, P., Schneider, R. T., Nienhaus, G. U., and Parak, F. (1996). X-ray structure determination of a metastable state of carbonmonoxy myoglobin after photodissociation. Proc. Natl. Acad. Sci. USA 93, 7013–7016.
Ligand-Binding Pathways in Globins
435
Huang, X., and Boxer, S. G. (1994). Discovery of new ligand binding pathways in myoglobin by random mutagenesis. Nat. Struct. Biol. 1, 226–229. Hummer, G., Schotte, F., and Anfinrud, P. A. (2004). Unveiling functional protein motions with picosecond x-ray crystallography and molecular dynamics simulations. Proc. Natl. Acad. Sci. USA 101, 15330–15334. Humphrey, W., Dalke, A., and Schulten, K. (1996). VMD: Visual Molecular Dynamics. J. Mol. Graph. 14, 33–38. Huo, S., and Straub, J. E. (1999). Direct computation of long time processes in peptides and proteins: Reaction path study of the coil-to-helix transition in polyalanine. Proteins 36, 249–261. Ishikawa, H., Uchida, T., Takahashi, S., Ishimori, K., and Morishima, I. (2001). Ligand migration in human myoglobin: Steric effects of isoleucine 107(G8) on O2 and CO binding. Biophys. J. 80, 1507–1517. Kendrew, J. C., Strandberg, R. E., Hart, B. E., Davies, R. G., Phillips, D. R., and Shore, V. C. (1960). Structure of myoglobin: A three-dimensional Fourier synthesis at 2 A˚ resolution. Nature 185, 422–427. Khavrutski, I. V., Byrd, R. H., and Brooks, C. L., III (2006). A line integral reaction path approximation for large systems via nonlinear constrained optimization: Application to alanine dipeptide and the b hairpin of protein G. J. Chem. Phys. 124, 194903-1–194903-14. Ma, J., Huo, S., and Straub, J. E. (1997). Molecular dynamics simulation study of the B-states of solvated carbon monoxymyoglobin. J. Am. Chem. Soc. 119, 2541–2551. MacKerell, A. D., Jr., Bashford, D., Bellott, M., Dunbrack, R. L., Jr., Evanseck, J. D., Field, M. J., Fischer, S., Gao, J., Guo, H., Ha, S., Joseph-McCarthy, D., Kuchnir, L., et al. (1998). All-atom empirical potential for molecular modeling and dynamics studies of proteins. J. Phys. Chem. B 102, 3586–3616. Meller, J., and Elber, R. (1998). Computer simulations of carbon monoxide photodissociation in myoglobin: Structural interpretation of the B states. Biophys. J. 74, 789–802. Milani, M., Pesce, A., Ouellet, Y., Ascenzi, P., Guertin, M., and Bolognesi, M. (2001). Mycobacterium tuberculosis hemoglobin N displays a protein tunnel suited for O2 diffusion to the heme. EMBO J. 20, 3902–3909. Mills, G., and Jonsson, H. (1994). Quantum and thermal effects in H2 dissociative adsorption: Evaluation of free energy barriers in multidimensional quantum systems. Phys. Rev. Lett. 72, 1124–1127. Moens, L., Vanfleteren, J., Van de Peer, Y., Peeters, K., Kapp, O., Czeluzniak, J., Goodman, M., Blaxter, M., and Vinogradov, S. (1996). Globins in nonvertebrate species: Dispersal by horizontal gene transfer and evolution of the structure-function relationship. Mol. Biol. Evol. 13, 324–333. Nienhaus, K., Deng, P., Kriegl, J. M., and Nienhaus, G. U. (2003a). Structural dynamics of myoglobin: Spectroscopic and structural characterization of ligand docking sites in myoglobin mutant L29W. Biochemistry 42, 9633–9646. Nienhaus, K., Deng, P., Kriegl, J. M., and Nienhaus, G. U. (2003b). Structural dynamics of myoglobin: Effect of internal cavities on ligand migration and binding. Biochemistry 42, 9647–9658. Nienhaus, K., Deng, P., Olson, J. S., Warren, J. J., and Nienhaus, G. U. (2003c). Structural dynamics of myoglobin. J. Biol. Chem. 278, 42532–42544. Nienhaus, K., Ostermann, A., Nienhaus, G. U., Parak, F. G., and Schmidt, M. (2005). Ligand migration and protein fluctuations in myoglobin mutant L29W. Biochemistry 44, 5095–5105. Noe´, F., Schwarzl, S. M., Fischer, S., and Smith, J. C. (2003). Computational tools for analyzing structural changes in proteins in solution. Appl. Bioinformat. 2(3 Suppl.), S11–S17.
436
Stephen D. Golden and Kenneth W. Olsen
Noe´, F., Krachtus, D., Smith, J. C., and Fischer, S. (2006). Transition networks for the comprehensive characterization of complex conformational change in proteins. J. Chem. Theory Comp. 2, 840–857. Nutt, D. R., and Meuwly, M. (2004). CO migration in native and mutant myoglobin: Atomistic simulations for the understanding of protein function. Proc. Natl. Acad. Sci. USA 101, 5998–6002. Olsen, K., Fischer, S., and Karplus, M. (2000). A continuous path for the T to R allosteric transition of hemoglobin. Biophys. J. 78, 394A. Olson, J. S., and Phillips, G. N., Jr. (1996). Kinetic pathways and barriers for ligand binding to myoglobin. J. Biol. Chem. 271, 17593–17596. Ostermann, A., Waschipky, R., Parak, F. G., and Nienhaus, G. U. (2000). Ligand binding and conformational motions in myoglobin. Nature 404, 205–208. Ozaki, S., Hara, I., Matsui, T., and Watanabe, Y. (2001). Molecular engineering of myoglobin: The improvement of oxidation activity by replacing phe-43 with tryptophan. Biochemistry 40, 1044–1052. Ozaki, S., Matsui, T., Roach, M. P., and Watanabe, Y. (2000). Rational molecular design of a catalytic site: Engineering of catalytic functions to the myoglobin active site framework. Coord. Chem. Rev. 198, 39–59. Pesce, A., Couture, M., Dewilde, S., Guertin, M., Yamauchi, K., Ascenzi, P., Moens, L., and Bolognesi, M. (2000). A novel two-over-two a-helical sandwich fold is characteristic of the truncated hemoglobin family. EMBO J. 19, 2424–2434. Quillin, M. L., Li, T., Olson, J. S., Phillips, G. N., Jr., Dou, Y., Ikeda-Saito, M., Regan, R., Carlson, M., Gibson, Q. H., Li, H., and Elber, R. (1995). Structural and functional effects of apolar mutations of the distal valine in myoglobin. J. Mol. Biol. 245, 416–436. Santos, S.-de O., Fischer, S., Guilbert, C., Lewitt-Bentley, A., and Smith, J. C. (2000). Pathway for large-scale conformational change in annexin V. Biochemistry 39, 14065–14074. Schlick, T. (2000). ‘‘Molecular Modeling and Simulation: An Interdisciplinary Guide.’’ New York: Springer, New York. Schmidt, M., Nienhaus, K., Pahl, R., Krasselt, A., Anderson, S., Parak, F., Nienhaus, G. U., and Srajer, V. (2005). Ligand migration pathway and protein dynamics in myoglobin: A time-resolved crystallographic study on L29W MbCO. Proc. Natl. Acad. Sci. USA 102, 11704–11709. Schotte, F., Lim, M., Jackson, T. A., Smirnov, A. V., Soman, J., Olson, J. S., Phillips, G. N., Jr., Wulff, M., and Anfinrud, P. A. (2003). Watching a protein as it functions with 150-ps time-resolved X-ray crystallography. Science 300, 1944–1947. Schotte, F., Soman, J., Olson, J. S., Wulff, M., and Anfinrud, P. A. (2004). Picosecond timeresolved X-ray crystallography: Probing protein function in real time. J. Struct. Biol. 147, 235–246. Scott, E. E., and Gibson, Q. H. (1997). Ligand migration in sperm whale myoglobin. Biochemistry 36, 11909–11917. Scott, E. E., Gibson, Q. H., and Olsen, J. S. (2001). Mapping the pathways for O2 entry into and exit from myoglobin. J. Biol. Chem. 276, 5177–5188. Sˇrajer, V., Ren, Z., Teng, T., Schmidt, M., Ursby, T., Bourgeois, D., Pradervand, C., Schildkamp, W., Wulff, M., and Moffat, K. (2001). Protein conformational relaxation and ligand migration in myoglobin: A nanosecond to millisecond molecular movie from time-resolved Laue X-ray diffraction. Biochemistry 40, 13802–13815. Teng, T.-Y., Sˇrajer, V., and Moffat, K. (1997). Initial trajectory of carbon monoxide after photodissociation from myoglobin at cryogenic temperatures. Biochemistry 36, 12087–12100. Tetreau, C., Blouquit, Y., Novikov, E., Quiniou, E., and Lavalette, D. (2004). Competition with xenon elicits ligand migration and escape pathways in myoglobin. Biophys. J. 86, 435–447.
Ligand-Binding Pathways in Globins
437
Tilton, R. F., Kuntz, I. D., and Petsko, G. A. (1984). Cavities in proteins: Structure of a metmyoglobin-xenon complex solved to 1.9 A˚. Biochemistry 23, 2849–2857. Verma, C. S., Fischer, S., Caves, L., Roberts, G. C. K., and Hubbard, R. E. (1996). Calculation of the reaction pathway of the aromatic ring flip in methotrexate complexed to dihydrofolate reductase. J. Phys. Chem. 100, 2510–2518. Vojtechovsky, J., Chu, K., Berendzen, J., Sweet, R. M., and Schlichting, I. (1999). Crystal structures of myoglobin-ligand complexes at near-atomic resolution. Biophys. J. 77, 2153–2174. Watts, R. A., Hunt, P. W., Hvitved, A. N., Hargrove, M. S., Peacock, W. J., and Dennis, E. S. (2001). A hemoglobin from plants homologous to truncated hemoglobins of microorganisms. Proc. Natl. Acad. Sci. USA 98, 10119–10124. Wittenberg, J., Bolognesi, M., Wittenberg, B., and Guertin, M. (2002). Truncated hemoglobins: A new family of hemoglobins widely distributed in bacteria, unicellular eukaryotes, and plants. J. Biol. Chem. 277, 871–874.
This page intentionally left blank
C H A P T E R
T W E N T Y- T W O
Finding Gas Migration Pathways in Proteins Using Implicit Ligand Sampling Jordi Cohen,* Kenneth W. Olsen,† and Klaus Schulten* Contents 440 442 442 443 444 445
1. Introduction 2. Methods 2.1. Required software 2.2. Procedure outline 2.3. Automated VMD script 2.4. Implicit ligand sampling configurable parameters 3. Example Calculation: Truncated Hemoglobin (trHb) from Paramecium caudatum 4. Discussion 4.1. Interpretation 4.2. Effect of water 4.3. Error bars 4.4. PMF projections 4.5. Computing probabilities and occupancies 4.6. Limitations Acknowledgments References
446 449 449 450 451 453 454 455 455 456
Abstract Implicit ligand sampling is a practical, efficient, and accurate method for finding the gas migration pathways for small hydrophobic gas molecules, such as oxygen, inside proteins. The method infers the gas migration pathways by calculating the potential of mean force for the gas molecule everywhere inside the protein by means of a molecular dynamics simulation of the protein in the absence of the gas molecule. Pathways can be constructed by connecting the areas of the protein that are favorable to the presence of gas. This method has the advantage of providing a comprehensive overview of all possible gas
* {
Beckman Institute, University of Illinois, Urbana, Illinois Department of Chemistry, Loyola University Chicago, Chicago, Illinois
Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37022-5
#
2008 Elsevier Inc. All rights reserved.
439
440
Jordi Cohen et al.
migration pathways and barriers in a given protein from a single simulation run. Implicit ligand sampling has been applied to a large number of hemoproteins. The example of the truncated hemoglobin from Paramecium caudatum is given to illustrate the method.
1. Introduction Many proteins interact with gas ligands such as O2 to perform their function. For such proteins, knowing how O2 and other gas molecules tunnel their way into, and through, the protein matrix is an important step toward understanding their function. Finding gas pathways inside proteins is not a trivial task, however, even when the atomic structure of the protein is well characterized. The reason for this is that gas molecules are small and hydrophobic and, as such, may enter the protein through many pathways that are not readily apparent from simply looking at the three-dimensional structure of the protein. In order to find gas migration pathways inside proteins, thermal fluctuations of the protein must be taken into account. A range of methods, both experimental and computational, has been developed to locate and characterize gas pathways inside proteins. If a gas molecule can bind reversibly to a prosthetic group inside the protein (e.g., such as the heme, in the case of hemoproteins), then the gas pathways for that protein can be probed experimentally. By photolysis of the bond between the prosthetic group and the gas molecule, the kinetic rates for the geminate recombination process can be measured, and by effectuating mutations in the protein, specific residues and their effect on recombination rates can be probed (Austin et al., 1975; Olson and Phillips, 1996). In addition, photolyzable bonds can also be used to probe gas migration pathways using time-resolved X-ray crystallography (Sˇrajer et al., 2001; Schotte et al., 2003). By then looking at the time evolution of the gas density in a crystal after gas ligand photolysis, important clues as to where the gas pathways are located can be inferred. Finally, the gas diffusion process inside the protein can be studied using crystallography under high xenon pressure (Tilton et al., 1984), as well as computer simulation. Simulation techniques that have often been associated with the search for gas migration pathways are locally enhanced sampling (Elber and Karplus, 1990), which allows for the use of many simultaneous gas molecules, and cavity searching (Amara et al., 2001; Cohen et al., 2005). More recently, the conjugate peak refinement method (Fischer and Karplus, 1992) has been applied to this problem (see Chapter 21). All these methods have their relative strengths and deficiencies, but all fail in one important respect: providing a reliable and complete picture of the gas pathways inside a protein.
441
Finding Gas Migration Pathways in Proteins
To specifically address the problem of finding gas migration pathways inside proteins, a method called implicit ligand sampling has been developed and applied to a growing number of protein families (Cohen and Schulten, 2007; Cohen et al., 2006; Johnson et al., 2007; Wang et al., 2007). The idea behind this method is based on the fact that gas migration occurs not along permanent channels, but rather along specific pathways that are defined by regions of the protein in which transient cavities can be formed due to thermal fluctuations. Such cavities can connect over time to form a continuous network of pathways inside the protein, enabling the passage of gas molecules across the protein matrix (Cohen et al., 2005). Because gas molecules interact weakly with the protein, the cavities in question can form regardless of whether or not a gas molecule is present. This is advantageous because it implies that, by running an equilibrium simulation of the thermal motion of a protein in the absence of gas, the gas migration pathways can be inferred. The potential of mean force (PMF)—the free energy of placing a gas molecule anywhere inside the protein—can thus be estimated by treating the placement of gas as a perturbation to the equilibrium dynamics of the protein, using the following formula derived in Cohen et al. (2006):
Gimplicit ðrÞ ¼ kB T ln
N X C DEm;k ðrÞ=kB T X e n¼1 k¼1
NC
;
ð22:1Þ
where Gimplicit(r) is the free energy of placing the ligand at a position r, kBT is the thermal energy, N is the number of simulation frames used in the analysis, C is the number of probed internal conformations/rotations of the implicit ligand, and DEm,k(r) is the position-dependent interaction energy of adding the ligand with a given rotation to a given simulation frame. The PMF obtained from Eq. (22.1) is equivalent to the free energy of placing a gas molecule anywhere in the protein keeping all other conditions identical. This free energy is related to the relative probability of finding a gas molecule at a given location. In practice, the PMF [Eq. (22.1)] is evaluated on a finely spaced three-dimensional grid, effectively creating a map of the PMF in space. From this map, iso-energy surfaces can be computed and displayed by most modern molecular visualization software. Figure 22.1 shows, as an example, the iso-energy surfaces of the O2 PMF computed for sperm whale myoglobin. By highlighting the regions of the protein that are favorable (or at the very least contain no high energy barriers) to a specific gas molecule using iso-energy surfaces (i.e., surfaces that contain all the points inside the protein for which the PMF of placing the gas molecule is below a certain value), one can thus map the network of likely pathways taken by the gas as it migrates inside the protein.
442
Jordi Cohen et al.
Figure 22.1 Iso-energy surfaces of the implicit ligand O2 PMF computed using Eq. (22.1) applied to a 10-ns molecular dynamics simulation of sperm whale myoglobin. Shown are the (A) 0 kT, (B) 3.3 kT, which corresponds to the solvation energy of O2 in water, (C) 6.7 kT, and (D) 10 kT iso-energy surfaces. Low-energy iso-surfaces delimit regions of high probability of finding a gas ligand. Common to all images are the myoglobin heme (as licorice) and the –3 kT iso-energy surface (in a darker color).
2. Methods The implicit ligand sampling method has been implemented in freely available open source software. The following section describes the steps necessary for performing the implicit ligand sampling analysis on an arbitrary protein or molecular assembly.
2.1. Required software 1. Simulation software capable of performing all-atom molecular dynamics simulations of the protein of interest, such as NAMD (download at http://www.ks.uiuc.edu/Research/namd) (Phillips et al., 2005), CHARMM (Brooks et al., 1983), or AMBER (Perlman et al., 1995), is required to create an equilibrium simulation of the protein of interest.
Finding Gas Migration Pathways in Proteins
443
All the results presented in this chapter used NAMD. The prewritten analysis program described here assumes that one uses the CHARMM molecular force field (MacKerell et al., 1995) used by NAMD and CHARMM. 2. The VMD (download at http://www.ks.uiuc.edu/Research/vmd) molecular visualization and analysis software (Humphrey et al., 1996), in which the implicit ligand sampling method is implemented. 3. The volutil (download at http://www.ks.uiuc.edu/Development/ MDTools/volutil) accessory software, which contains necessary routines and example scripts for applying the implicit ligand sampling method described in this chapter, as well as for manipulating the resulting three-dimensional PMF maps.
2.2. Procedure outline The procedure for creating an implicit ligand sampling PMF map is outlined here. A Tcl-language script, which performs these steps and which can be run in the VMD software environment, is detailed in the next section. 1. A 5- to 10-ns equilibrium simulation of the protein of interest, solvated in a water box, must be generated. To compute the Gibbs free energy of placing a gas molecule, simulation must be done using a constant pressure and temperature ensemble (NPT). A constant volume simulation would provide the Helmholtz free energy instead. Longer simulation times will produce more accurate results and lower error bars on the energy. An exponential increase in simulation time will produce a linear decrease in the error (Cohen et al., 2006). 2. The simulation trajectory then needs to be divided into segments that will fit into computer memory, and these segments should be loaded into VMD one at a time (ideally, instances of VMD will be run as a background text-based process for the analysis of each segment). Simulation trajectory frames should be spaced 1 to 2 ps apart; shorter times result in very similar frames with very little gain in sampling, whereas longer times will result in less frames overall and consequently suboptimal statistics. 3. A PDB reference frame should be created and saved to disk, which will be used as an alignment template for all subsequent trajectories to be analyzed and turned into PMF maps. 4. For each trajectory segment, all frames of the trajectory should be aligned to the PDB reference frame to compensate for translations and rotations of the protein that occur during simulation. Typically, the frames are aligned using a best fit of the alpha carbon positions across all frames. For each trajectory segment, the CHARMM force-field van der Waals (vdW) parameters must be specified for each atom of the
444
Jordi Cohen et al.
system so that these can be accessed by the implicit ligand sampling routine. In VMD, each atom possesses a set of associated fields that can store numerical values. The ‘‘beta’’ field of VMD must be set to contain, for each atom independently, the CHARMM vdW radius parameter, and the ‘‘occupancy’’ field must contain the CHARMM vdW energy parameter. The volutil package contains utility functions to perform the automatic loading of these parameters (see later). The parameters, as used by the VMD analysis routines, must follow the CHARMM Lennard–Jones formulation for the vdW energy:
"
Rmin=2i þ Rmin=2j 12 rij # Rmin=2i þ Rmin=2j 6 2 ; ei ; ej < 0; rij
pffiffiffiffiffiffi VvdW ðrÞ ¼ ei ej
ð22:2Þ
where the i and j indices represent two interacting atoms. e (negative value) is the vdW minimum well energy, and Rmin/2 is half of the distance at which this well occurs. If using a different force field than CHARMM, one must ensure that the Lennard–Jones parameters are converted so that they follow the convention of Eq. (22.2). 5. Perform the implicit ligand sampling analysis in VMD on each trajectory segment and specify sampling parameters as well as a description of the desired gas ligand (see description and options later). This can take a day to a week on contemporary single-processor computers; however, each trajectory segment can be run on a separate computer or processor to accelerate the computation. The final result is a three-dimensional map of the PMF of placing the specified gas molecule at every point on a grid inside the solvated protein system for the specified trajectory segment. 6. The various PMF maps for each trajectory segment must then be combined into a single map using the volutil command-line utility (part of the volutil package): volutil –combinepmf pmf 1.dx pmf 2.dx [pmf 3.dx . . .] –o combo_pmf.dx
2.3. Automated VMD script The following script can be used to run implicit ligand sampling in the background using VMD. It can be run on a UNIX-type command line terminal, using
445
Finding Gas Migration Pathways in Proteins
vmd -dispdev text -e example-ils.tcl,
where example-ils.tcl (part of the volutil package) contains sample code to run the implicit ligand analysis, which can be adapted easily for the analysis of any protein or simulation.
2.4. Implicit ligand sampling configurable parameters The implementation of implicit ligand sampling contained in VMD allows the user to adjust a few parameters.
Resolution: the spacing between grid points can be adjusted. Optimally, ˚ per grid point or better. this should be set to 1 A Temperature: the computation of the PMF depends on the temperature of the simulation, which can be specified. Ligand description: the gas molecule can be defined as being either monoatomic or diatomic, and the Lennard–Jones parameters for each atom of the ligand, as well as the bond length, for diatomic ligands, can be specified. Recommended parameters for commonly studied gas species that yield correct water solvation energies are listed in Table 22.1. Number of rotamers: when using a diatomic ligand (such as O2, CO), sampling can be made more accurate by averaging the PMFs over a random sampling of the internal degrees of freedom of the ligand. VMD allows the sampling of various random orientations of the ligand. It is recommended that one averages over about 50 orientations of the ligand; adding more states has only a negligible effect. When neglecting electrostatic effects, the computational cost for adding extra orientational states is low, as only short-range interactions are recom˚ ) being puted for each orientation [with long-range interactions (>5 A relatively unchanged in practice]. VMD allows the user to set the cutoff radius, which defines what constitutes ‘‘short-range’’ interactions and also contains a mode that does not perform this approximation to the detriment of computation speed. Table 22.1 Recommended CHARMM parameters of common gas molecules for use as implicit ligand sampling probes Gas ligand
e (kcal/mol)
Rmin/2 ˚) (A
Bond length ˚) (A
O2 NO CO Xe
O: –0.12 N: –0.20 C: –0.11 –0.49
O: 1.70 N: 1.85 C: 2.10 2.24
1.12 1.15 1.13 N/A
446
Jordi Cohen et al.
Subsampling per grid point: the PMF of placing the gas ligand is computed for every point on a regular grid. While the easiest way to increase the precision of the PMF is by computing more simulation frames (Fig. 22.2A), it is also possible to do so by increasing the amount of samples per grid point. This is achieved by translating the probe gas molecule by many fractional amounts within each grid point and averaging over the PMFs at each of these locations instead of just placing the center of mass of the molecule at the center of the grid points. The average over rotamers is still performed for each position. This additional sampling is referred to as ‘‘subsampling.’’ VMD allows the specification of N, where the total number of positions sampled per grid is N3 (subsampling locations are evenly distributed on a N N N subgrid within each grid point). Increasing N results in a more accurate free energy (lower error) per unit volume. Figure 22.2b displays distribution of the PMF for placing O2 at each grid point in a (40 A˚)3 water box (averaged over 4000 1-ps frames of simulation). With perfect sam˚ )3 grid point, but for pling, the PMF should be identical for each (1 A finite trajectories, using subsampling reduces the error on the PMF at each grid point. It can be seen that increasing the per grid point subsampling narrows the distribution of measured PMF values while keeping the same average PMF (i.e., it reduces the error). A larger N results in better data but a much slower performance; in practice, using subdivisions smaller than half of an Angstrom (e.g., N ¼ 2 when using a 1-A˚ resolution) does not yield significantly improved results (see Figure 22.2B).
3. Example Calculation: Truncated Hemoglobin (trHb) from Paramecium caudatum Truncated hemoglobins are small heme proteins found in microbes and plants that are distantly related to animal hemoglobins. Their tertiary structure consists of a two-on-two helical sandwich in contrast to the threeon-three helical sandwich found in animal globin such as myoglobin (Wittenberg et al., 2002). trHbs are normally 20 to 40 residues shorter than myoglobin. There are over 100 known sequences for trHbs distributed in three groups. Within each group, the degree of sequence identity is very high but it is low between groups ( Vuletich and Lecomte, 2006). Analysis of phylogenic trees suggests that the group II trHbO gene was the origin of the family ( Vuletich and Lecomte, 2006). As an example of the application of the implicit ligand sampling method to oxygen binding in globins, we have studied the pathways found in the truncated hemoglobin from a ciliated protozoon. The X-ray structure of the trHb from P. caudatum (Pc-trHb) has been determined
Finding Gas Migration Pathways in Proteins
447
Figure 22.2 Effect of the (A) number of trajectory frames and (B) subsampling per grid point used in implicit ligand sampling analysis on the accuracy of the measured PMF. In all cases, the PMF for placing O2 was computed for every (1 —)3 grid point of the PMF map computed from a 4-ns NPT simulation of a (40 —)3 water box (using NAMD and TIP3 water). Unless otherwise specified, 4000 frames and a subsampling of 23 were used. Distribution of the measured PMFs at every point is plotted, and the overall computed PMF of solvation for each case is indicated as vertical lines. When sufficiently averaged over time, the PMF in water ideally should be spatially uniform, which would correspond to a narrow distribution of the PMF about its averaged value.
448
Jordi Cohen et al.
(Pesce et al., 2000). It is a member of the group I family of trHbs ( Vuletich and Lecomte, 2006). Helicies B, E, G, and H form the two-on-two helical sandwich. The tertiary structure lacks substantial portions of the A helix, the CD-D region, and the EF-F region when compared to myoglobin (Pesce et al., 2000). The heme is held in a hydrophobic pocket between the E helix and the shortened F helix by a coordinate covalent bond from the proximal His 62 and an electrostatic interaction between Lys 44 and one of the propionate groups. The water ligand in the X-ray structure is stabilized by a network of hydrogen bonds involving Tyr 20, Gln 41, and Thr 45. This ligand is removed during the molecular dynamics simulation used for implicit ligand sampling. The classical histidine gate path (Perutz, 1989) is blocked by the position of the E helix in Pc-trHb. As in other trHbs, this globin has a long hydrophobic tunnel leading from the binding site through the AB turn/GH turn region to the solvent (Pesce et al., 2000). Because this tunnel is conserved in trHbs, it may be functionally important (Wittenberg et al., 2002). For the application of implicit ligand sampling to Pc-trHb, the coordinates (PDB: 1DLV; Pesce et al., 2000) are obtained from the Protein Data Bank (Berman et al., 2000;www.rcsb.org/pdb). The VMD molecular graphic program (Humphrey et al., 1996) is used to prepare the structure for the molecular dynamics simulation. The charge states of the histidine residues are determined by visual inspection of their surroundings. The protein is solvated and ions are added to the system zero net charge. Coordinate (.pdb) and protein structure (.psf) files are created using CHARMM 27 topology and parameter files (MacKerell et al., 1998). The NAMD molecular dynamics program (Phillips et al., 2005) is used to generate the needed simulation. The energy of the water in the system is first minimized with 1000 steps of a conjugate gradient routine to remove any bad contacts. The system is then equilibrated at 1 atm and 310 K for 1 ns using Langevin dynamics. A 2-fs time step is used with the rigid bond parameter. All systems are simulated with periodic boundary conditions using the particle-mesh Ewald method for accurate treatment of long-range electrostatic interactions. Then, 10 ns of dynamics is simulated to be used for the implicit ligand-sampling calculation. The PMF map is generated using VMD and the volutil package mentioned earlier and is interpreted visually using VMD to determine the potential ligand-binding pathways. The PMF map (.dx) file is loaded into the VMD program along with the coordinate (.pdb) file for the protein. The PMF map is displayed using the Isosurface drawing method in VMD. The analysis of implicit ligand sampling for Pc-trHb indicates that there are five potential pathways for ligand access to the heme in this protein. One path exits between the B and E helices after passing over the C helix. A second path exits close by the first between the B and the E helices. The third one loops around the heme to leave between the E and the H
Finding Gas Migration Pathways in Proteins
449
Figure 22.3 Stereo diagram of potential oxygen-binding tunnels in Pc-trHb. Portions of helices B (pink), C (orange), E (yellow), G (silver), and H (green), as well as the heme (gray, behind helix E), can be seen. Portions of four tunnels are seen in this view. The largest one exits to the left between helices E and H. The second tunnel exits to the right between helices B and G. The third can be seen below the second one, also exiting between the B and the G helices. The fourth tunnel extends backwards toward the upper right-hand corner between helices B and E and over the heme and helix C, but the exit of this tunnel cannot be seen in this view.
helices. The fourth path goes past the A helix to exit between the B and the H helices. The final path leaves the protein between the BC loop and the G helix. Figure 22.3 shows a cross section of the PMF map and the protein. Several of the paths can be seen in Fig. 22.3.
4. Discussion 4.1. Interpretation Implicit ligand sampling PMF maps represent the free energy of placing a gas ligand at any location inside a protein. As such, this free energy takes into account energetic effects relating to how favorably the gas molecule interacts with its surroundings, as well as entropic effects, which take into account both the rotational states of the gas molecule and how many states of the protein are compatible with having the gas molecule at a given location. Through the formulation of Eq. (22.1), the implicit ligand sampling PMF is implicitly normalized such that a PMF of 0 kT is equivalent to the gas molecule being in an isolated vacuum (i.e., in an ideal gas phase). All areas of the protein for which the gas PMF is less than 0 kT are thus areas
450
Jordi Cohen et al.
in which it is more likely to find a gas molecule than in a gas phase of identical volume.
4.2. Effect of water One definite advantage of the implicit ligand method over cavity-searching methods is that it deals with water molecules correctly. First, implicit ligand sampling makes definite predictions as to the effect of water molecules trapped inside the protein. For example, the cavity-searching method detailed in Cohen et al. (2005) always treated internal water molecules as impassable to O2 because O2-sized bubbles were rarely observed near them. However, in the explicit gas simulations of Cohen et al. (2005), it was found by implicit ligand sampling that in some cases, gas could pass through regions inside the protein that were solvated by water molecules, whereas other water-containing regions appeared impassable to gas. This behavior matched exactly that observed by simulations of explicit gas molecules in the same protein. A second advantage of the implicit ligand approach is its correct treatment of external water molecules. Cavity searching is intrinsically hard in water because the environment is heterogeneous at the length scales of interest to gas diffusion. For this reason, the external water solution is simply omitted in cavity-based approaches. This is unfortunate because the interplay between the external water solution and the residues on the surface of the protein creates regions that can be favorable or unfavorable to gas molecules, which in turn may influence the uptake of gas by the protein. Implicit ligand sampling has been able to predict which locations containing water in hydrogenase were passable and which were not according to simulation (Cohen et al., 2005), as well as being able to predict water-containing pockets in myoglobin that were xenon-binding sites (Cohen et al., 2006). In practice, water molecules that are strongly interacting and tightly coordinated by the protein (such as is often the case in a water file) more often than not will not allow for the passage of small gas ligands. However, disordered water pockets inside proteins offer an ideally favorable environment for gas molecules. Because disordered water molecules, especially those inside partially hydrophobic regions of the protein, do not pack well, they create ample space for gas molecules to fit. Also, because of the favorable interaction energies between water and gas molecules (even if electrostatic effects are ignored), such regions are often even more favorable for gas than even the gas phase (vacuum). Implicit ligand sampling treats the water molecules correctly as the measured PMF near water will depend on the thermal behavior of the water molecules.
Finding Gas Migration Pathways in Proteins
451
4.3. Error bars In order to interpret the computed PMF, one needs to estimate how accurate it is. Estimating the error on a PMF inferred from molecular dynamics is always hard because the PMF is based on a minuscule sampling of all the possible states of the system. It is usually assumed that the states sampled in the MD simulation are representative. However, the PMF is frequently dominated by rare events, in this case, by protein conformations of the lone protein ensemble that are very favorable to the placement of the gas ligand. One way to approximate the error on the PMF is to estimate the average maximum effect of rare events. Because energy is the only quantity measured during the calculation of the PMF, we can define ‘‘rare events’’ to mean those events for which the ligand–protein interaction energy would have been very favorable (which would be able to alter the PMF even if only one such event were added). Because gas molecules interact weakly with the protein, the effect of such ‘‘rare events’’ is bounded. For example, when favorably surrounded by atoms, such as is the case in water, the energy of interaction of the system with O2, is only of about –3.2 kT. For the case of O2, we would only then need to consider the possible effect on the PMF of ‘‘maximally favorable’’ states, for which the implicit O2 interaction energy would be –3.2 kT (states even more favorable to O2 are highly unlikely and, in many cases, not possible, and less favorable states would have a lesser impact on the PMF and thus their effect would be contained in the estimated error bars). Finally, in order to compute an ‘‘average’’ maximum error, one can argue that if a maximally favorable state has not occurred over the course of N samples, then, on average, such states happen at most every N sample. Armed with both a lower bound on the interaction energy and a lower bound on the frequency of occurrence of rare events, one can then compute, on average, the maximum impact that such states would have on the measured PMF. This can be done by computing the PMF as is normally done, except that one would then add a fabricated ‘‘N þ 1’’th state, which would be maximally favorable. This will give us a lower bound on the PMF (undersampling, i.e., omitting rare states, will always result in an overestimated PMF), derived in Cohen et al. (2006): DGimplicit ðrÞ
eb½Gimplicit ðrÞDEmin ; ¼ kB T ln 1 þ N
ð22:3Þ
where DG(r) is the lower error on the computed PMF, G(r), N is the number of independent frames used in the computation, and DEmin is an energy scale representing the interaction energy of the gas ligand with its environment in the most favorable conditions. If the PMF is close to this energy, then the PMF is guaranteed to be well-sampled, and the sampling
452
Jordi Cohen et al.
error will be small. If the PMF is much larger than this energy, the sampling cannot be guaranteed to be accurate, and the error will be correspondingly large. In reality, the quantity DEmin is a function of the location of the ligand inside the protein. In practice, DEmin is approximated to be a constant and should be measured using simulation of the explicit ligand in a region of maximally favorable interaction energy with its environment. For small apolar gas ligands, the interaction energy of the solvated ligand with water is a suitable way of evaluating DEmin. Table 22.2 contains values for DEmin for common gas ligands, computed from their interaction with water. Equation (22.3) can also be used to determine whether a ligand is appropriate for use with implicit ligand sampling. Immediately from this equation, one can see that large ligands will always have large or divergent measured PMFs at each point (since they will never favorably fit into the lone protein), resulting in huge computed error bars. Similarly, small charged ligands such as ions will have very low values of DEmin, which will never be attained by the computed PMF, also resulting in gigantic error bars. Implicit ligand sampling therefore cannot be applied to such ligands. In addition to the error caused by overestimation of the PMF caused by the insufficient sampling of favorable rare events, there is also the usual sampling error. A good estimate of this error can be obtained by examining the variation in the measured PMF of the gas molecule at different points in a water box (see Figure 22.2B). Because water is uniform, the spatial variation is representative of the temporal variation. This variation, for O2, obeys the empirical rule.
60kB T þ= DGimplicit pffiffiffiffiffi N
ð22:4Þ
where N is the number of simulation frames used in the analysis and would result in a sampling error of about 0.6 kT when analyzing every 1 ps of a 10-ns simulation. Table 22.2 Recommended values of DEmin for common gas molecules for use in implicit ligand samplinga
a
Ligand
DEmin (kcal/mol)
O2 NO CO Xe
–3.2 –4.1 –3.7 –5.6
Details on how these values were obtained are found in Cohen et al. (2006).
Finding Gas Migration Pathways in Proteins
453
4.4. PMF projections The implicit ligand sampling method computes the PMF per unit volume at every point inside a grid. In many cases, the quantity of interest is the PMF along a given reaction coordinate rather than along a very specific path. A straightforward example is characterization of the free energy barriers of gas permeation across a transmembrane protein containing gas channels, where one is interested in the overall barrier of crossing the channels at a function of position along the channel (which we will call z). Such an example is gas permeation in AQP1 aquaporin, detailed in Wang et al. (2007). The projection G(z) of the PMF onto the z axis can be computed using the formula
GðzÞ ¼
X X kB T ln eGðx;y;zÞ=kB T ; Ao ¼ ð1Þ; Ao ðx;yÞ2S ðx;yÞ2S
ð22:5Þ
where S defines the xy cross-sectional surface over which the projection is performed and Ao is the cross-sectional area of S. When making PMF projections, a few things must be kept in mind. First, the cross-sectional area along the reaction coordinate must be kept constant during the calculation, and it must be realized that one can only compare different PMF projections that use the same cross-sectional area. For the case of AQP1, there were three different possible types of gas channels across the protein with different cross-sectional areas. In that case, to assess the relative barriers of each channel for gas permeation processes, one must compute the PMF projection using a common crosssectional area Ao (taken to be the cross section of an entire AQP1 tetramer) for all channels, setting G(x,y,z) to be infinite in all areas outside the specific channel along which the projection is being computed. An important barrier encountered by any molecule going through a channel is the interface between the water solution and the channel. A commonly made mistake is to ignore the effect of the spatial concentration of channels in the membrane on the PMF along z. In fact, for the case of AQP1, the free energy profile of gas crossing AQP1 is very dependent on the AQP1 concentration in the membrane. When computing the PMF profile, one must consider that the PMF projection is really interpreted over an area A1, which corresponds to the membrane area per aquaporin (i.e., A1 is the total membrane area divided by the number of aquaporins in that membrane area). To compare PMF computed with cross-sectional area Ao to the desired PMF at cross-sectional area A1, the PMF projection must be shifted inside the channel as such:
454
Jordi Cohen et al.
G1 ðzÞ ¼
in water Go ðzÞ; : Go ðzÞ þ kB T lnðA1 =Ao Þ; in membrane
ð22:6Þ
This operation is necessary to compare energy barriers between simulations or experiments performed at different AQP1 concentrations (there is no such thing as an ‘‘absolute’’ concentration-independent free energy barrier when looking at PMF profiles along a path). Go(z) in water should be constant and corresponds to the solvation energy of the gas in water (and is independent of cross-sectional area as water is translationally invariant). If the solvation energy is known experimentally, then the profile G1(z) should be shifted to match it in the water solution and to thus provide an absolute free energy profile.
4.5. Computing probabilities and occupancies The relative probability, p(r), of finding a gas molecule inside any grid point inside the protein is related to the PMF through the equation.
pðrÞ ¼ po e½GðrÞGo =kB T ;
ð22:7Þ
where p and G are the probabilities of finding a gas ligand and associated PMF of placing the ligand in a given unit volume, respectively. In practice, one usually knows the external partial pressure of the gas ligand. If we use the gas phase as our point of reference (where Go ¼ 0 kT by definition), then po, the probability of finding a gas molecule in a unit volume of the gas/vacuum phase, can be estimated from the partial pressure of the gas, Po, using the ideal gas law:
po ¼
Po : kB T
ð22:8Þ
Using Eq. (22.8), the probability of finding the gas molecule everywhere inside the protein/solvent system can be computed, provided that the external gas pressure is known. This calculation only holds for low gas concentrations, where ligand–ligand interactions and site saturation effects are negligible. Because the computed probabilities are specified per unit volume, to find the probability of finding a gas molecule present in a given region of the protein, or to find the occupancy of gas at a given binding site, one must sum over the probabilities of finding gas molecules in all the grid points located inside the region of interest.
Finding Gas Migration Pathways in Proteins
455
4.6. Limitations
Rare events and conformational changes. With infinite sampling, PMF maps would be 100% exact. However, because the sampling in molecular dynamics is confined to a narrow range of states, the effect of slow conformational and allosteric changes will not be observed during the course of the simulation. Therefore, there is no guarantee that all biologically relevant pathways will be discovered through simulation. It must be realized that gas molecules migrating across proteins can take advantage of thermal motion (which is sampled adequately in the nanosecond timescale), as well as slow conformational changes, such as the opening of a channel gate (which are usually not sampled at simulation timescales). Only pathways defined by the thermal motion of the protein are described by implicit ligand sampling. Pathways affected by slower dynamics must be assessed separately, for example, by repeating the implicit ligand sampling analysis on separate equilibrium trajectories obtained using different starting points that correspond to different protein conformational or allosteric states imposed a priori. Electrostatic effects. The implementation of implicit ligand sampling described here allows for specifying partial charges. However, the approximation of computing long-range interactions only once for all the rotational states of the gas molecule does not work when the ligand is charged, and the significantly slower ‘‘slowligand’’ implementation of implicit ligand sampling must be used. When the ligand is weakly dipolar (e.g., partial charges <0.02e), the effect of the charge of the ligand on the final PMF is negligible, such that, in practice, the computation of electrostatics is usually omitted. Also, if the partial electric charges of the ligand are significant, it may no longer be weakly interacting with the protein and the approximations implied by the implicit ligand sampling analysis will no longer hold. Finally, in regions of the protein that are charged, electrostatic effects arising from electric quadrupole moment or polarizability of the gas molecule may become significant; these are not accounted for by the aforementioned implicit ligand sampling analysis without modification. Quantum and bonded effects. In many proteins, including globins, an important determinant of gas migration rates is the binding energy of the gas ligand to a reactive site inside the protein (e.g., the globin heme). The free energy of placing a bonded ligand is very different from that of placing a free one, and such energies cannot be computed from the implementation of implicit ligand sampling described here.
ACKNOWLEDGMENTS The authors acknowledge grants from the National Institutes of Health PHS-5-P41RR05969 ( JC, KS), the National Science Foundation SCI04–38712 ( JC, KS), and the Department of Energy ( JC, KS).
456
Jordi Cohen et al.
REFERENCES Amara, P., Andreoletti, P., Jouve, H. M., and Field, M. J. (2001). Ligand diffusion in the catalase from Proteus mirabilis: A molecular dynamics study. Protein Sci. 10, 1927–1935. Austin, R. H., Beeson, K. W., Eisenstein, L., Frauenfelder, H., and Gunsalus, I. C. (1975). Dynamics of ligand binding to myoglobin. Biochemistry 14, 5355–5373. Berman, H. M., Westbrook, J., Feng, Z., Gilliland, G., Bhat, T. N., Weissig, H., Shindyalov, I. N., and Bourne, P. E. (2000). The protein databank. Nucleic Acids Res. 28, 235–242. Brooks, B. R., Bruccoleri, R. E., Olafson, B. D., States, D. J., Swaminathan, S., and Karplus, M. (1983). CHARMM: A program for macromolecular energy, minimization, and dynamics calculations. J. Comput. Chem. 4, 187–217. Cohen, J., Arkhipov, A., Braun, R., and Schulten, K. (2006). Imaging the migration pathways for O2, CO, NO, and Xe inside myoglobin. Biophys. J. 91, 1844–1857. Cohen, J., Kim, K., King, P., Seibert, M., and Schulten, K. (2005). Finding gas diffusion pathways in proteins: Application to O2 and H2 transport in CpI [FeFe]-hydrogenase and the role of packing defects. Structure 13, 1321–1329. Cohen, J., and Schulten, K. (2007). O2 migration pathways in monomeric globins are determined by residue composition, not tertiary structure. Submitted for publication. Elber, R., and Karplus, M. (1990). Enhanced sampling in molecular dynamics: Use of the time-dependent Hartree approximation for a simulation of carbon monoxide diffusion through myoglobin. J. Am. Chem. Soc. 112, 9161–9175. Fischer, S., and Karplus, M. (1992). Conjugate peak refinement: An algorithm for finding reaction paths and accurate transition states in systems with many degrees of freedom. Chem. Phys. Lett. 194, 252–261. Humphrey, W., Dalke, A., and Schulten, K. (1996). VMD: Visual Molecular Dynamics. J. Mol. Graph. 14, 33–38. Johnson, B. J., Cohen, J., Welford, R. W., Pearson, A. R., Schulten, K., Klinman, J. P., and Wilmot, C. M. (2007). Exploring molecular oxygen pathways in Hansenula polymorpha copper-containing amine oxidase. Submitted for publication. MacKerell, A. D., Jr., Bashford, D., Bellott, M., Dunbrack, R. L., Jr., Evanseck, J., Field, M. J., Fischer, S., Gao, J., Guo, H., Ha, S., Joseph, D., and Kuchnir, L., (1998)., et al. All-atom empirical potential for molecular modeling and dynamics studies of proteins J. Phys. Chem. B 102, 3586–3616. Olson, J. S., and Phillips, G. N., Jr. (1996). Kinetic pathways and barriers for ligand binding to myoglobin. J. Biol. Chem. 271, 17593–17596. Perlman, D. A., Case, D. A., Caldwell, J. W., Ross, W. S., Cheatham, T. E., III, Debolt, S., Ferguson, D., Seibel, G., and Kollman, P. (1995). AMBER, a package of computer programs for applying molecular mechanics, normal mode analysis, molecular dynamics and free energy calculations to simulate the structural and energetic properties of molecules. Comp. Phys. Comm. 91, 1–41. Perutz, M. F. (1989). Myoglobin and haemoglobin role of distal residues in reactions with haem ligands. Trends Biochem. Sci. 14, 42–44. Pesce, A., Couture, M., Dewilde, S., Guertin, M., Yamauchi, K., Ascenzi, P., Moens, L., and Bolognesi, M. (2000). A novel two-over-two a-helical sandwich fold is characteristic of the truncated hemoglobin family. EMBO J. 19, 2424–2434. Phillips, J. C., Braun, R., Wang, W., Gumbart, J., Tajkhorshid, E., Villa, E., Chipot, C., Skeel, R. D., Kale, L., and Schulten, K. (2005). Scalable molecular dynamics with NAMD. J. Comp. Chem. 26, 1781–1802. Schotte, F., Lim, M., Jackson, T. A., Smirnov, A. V., Soman, J., Olson, J. S., Phillips, G. N., Jr., Wulff, M., and Anfinrud, P. A. (2003). Watching a protein as it functions with 150-ps time-resolved X-ray crystallography. Science 300, 1944–1947.
Finding Gas Migration Pathways in Proteins
457
Sˇrajer, V., Ren, Z., Teng, T. Y., Schmidt, M., Ursby, T., Bourgeois, D., Pradervand, C., Schildkamp, W., Wulff, M., and Moffat, K. (2001). Protein conformational relaxation and ligand migration in myoglobin: A nanosecond to millisecond molecular movie from time-resolved Laue X-ray diffraction. Biochemistry 40, 13802–13815. Tilton, R. F., Kuntz, I. D., and Petsko, G. A. (1984). Cavities in proteins: Structure of a metmyoglobin-xenon complex solved to 1.9 A˚. Biochemistry 23, 2849–2857. Vuletich, D. A., and Lecomte, J. T. J. (2006). A phylogenetic and structural analysis of truncated hemoglobins. J. Mol. Evol. 63, 196–210. Wang, Y., Cohen, J., Boron, W. F., Schulten, K., and Tajkhorshid, E. (2007). Exploring gas permeability of cellular membranes and membrane channels with molecular dynamics. J. Struct. Biol. 157, 534–544. Wittenberg, J., Bolognesi, M., Wittenberg, B., and Guertin, M. (2002). Truncated hemoglobins: A new family of hemoglobins widely distributed in bacteria, unicellular eukaryotes, and plants. J. Biol. Chem. 277, 871–874.
This page intentionally left blank
C H A P T E R
T W E N T Y- T H R E E
Identification of Ligand-Binding Pathways in Truncated Hemoglobins Using Locally Enhanced Sampling Molecular Dynamics Stephen D. Golden and Kenneth W. Olsen Contents 1. Introduction 1.1. Investigation of ligand-binding pathways in group I truncated hemoglobins 1.2. LESMD studies of other globins and available programs 2. Molecular Dynamics 3. Locally Enhanced Sampling Molecular Dynamics 4. Methods 4.1. Structure preparation 4.2. Minimization and equilibration 4.3. Classification of ligand-binding pathways 5. Results 5.1. Mt-trHbN 5.2. Pc-trHb 5.3. Ce-trHb 6. Conclusions References
460 460 461 462 465 466 466 467 468 468 468 469 471 471 472
Abstract This chapter reviews the use of a locally enhanced sampling molecular dynamics (LESMD) for the study of ligand binding in truncated hemoglobins. The method, however, can be applied to any protein–ligand system. Truncated hemoglobins appear to have a tunnel(s) potentially used by the ligand to bind. These structural features give some indication of how the ligand moves through the protein to bind but do not give the complete picture. The LESMD
Department of Chemistry, Loyola University Chicago, Chicago, Illinois Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37023-7
#
2008 Elsevier Inc. All rights reserved.
459
460
Stephen D. Golden and Kenneth W. Olsen
method has been used to investigate the pathways of ligand binding to group I truncated hemoglobins from the eubacteria Mycobacterium tuberculosis, the ciliated protozoa Paramecium caudatum, and the unicellular alga Chlamydomonas eugametos.
1. Introduction 1.1. Investigation of ligand-binding pathways in group I truncated hemoglobins Truncated hemoglobins (trHbs) are small heme proteins found in bacteria, unicellular eukaryotes, and higher plants, forming a group within the hemoglobin superfamily (Watts et al., 2001). TrHbs have less than 15% sequence identity when compared with vertebrate and nonvertebrate hemoglobins (Couture et al., 1994; Moens et al., 1996). Many trHbs display amino acid sequences 20–40 residues shorter than nonvertebrate hemoglobins to which they are scarcely related by sequence similarity (Wittenberg et al., 2002). Truncated hemoglobins bind various ligands with their functions still being a mystery. Questions have been raised as to the function of truncated hemoglobin from Mycobacterium tuberculosis (Mt-trHb) with oxygen transportation not being the main role (Das et al., 2000; Hvitved et al., 2001; Weber et al., 2001). The protein may be a defense mechanism to protect the bacilli against reactive nitrogen species produced by the host (Pathania et al., 2002; Yeh et al., 2000). Functions of the truncated hemoglobins from Paramecium caudatum (Pc-trHb) and Chlamydomonas eugametos (Ce-trHb) are still unknown. Each of these proteins appears to have a tunnel(s) potentially used by the ligand to bind. In attempts to understand the function of truncated hemoglobins, resonance Raman, X-ray diffraction, kinetic, molecular dynamics simulations, and mutation studies have been used to characterize the process of ligand binding (Crespo et al., 2005; Dantsker et al., 2005; Milani et al., 2001, 2004; Mouawad et al., 2005; Mukai et al., 2004; Pesce et al., 2000; Samuni et al., 2003; Yeh, 2004). This chapter examines ligand binding traveling from the distal binding cavity using a mean field approach enhancing the sampling of the conformational space due to the presence of multiple copies. This methodology is locally enhanced sampling molecular dynamics (LESMD) (Elber and Karplus, 1990). We have used LESMD, as implemented in the molecular dynamics program NAMD (Phillips et al., 2005), to calculate the binding pathways for ligands traveling from the distal binding pocket of truncated hemoglobins. These studies have indicated that each of the three truncated hemoglobins contains new branches of the existing tunnel system.
Locally Enhanced Sampling Pathways in Globins
461
1.2. LESMD studies of other globins and available programs Three programs have been used for LESMD studies of globins. The first investigation, which introduced the LESMD method and gave the mathematical basis of the procedure, was done by Elber and Karplus (1990), who used CHARMM (Brooks et al., 1983; www.charmm.org) to examine CO diffusion through myoglobin. This study demonstrated that molecular fluctuations of the protein are required for its function. CHARMM was also used (Gibson et al., 1992) to examine the effects of mutation of the distal pocket Leu 29 of myoglobin, confirming that the dynamics of the ligand movement can explain the photodissociation experiments. Both CHARMM and MOIL were used in LESMD studies on dimeric clam hemoglobin (Chiancone et al., 1993). The two programs gave similar results even though they use different molecular mechanics parameters. MOIL has been developed by Professor Ron Elber and co-workers (1995; cbsu.tc.cornell.edu/software/moil/moil.html), who have explained the mathematical details of the method in several of their papers (Roitberg and Elber, 1991; Utilsky and Elber, 1993, 1994). The effects of distal pocket mutations of myoglobin at Leu 29 on nitric oxide (NO) recombination were reexamined using MOIL (Li et al., 1993). Results were consistent with the previous study (Gibson et al., 1992), showing that the recombination rate is determined by cavities in the protein and that the ligand goes first to a site in the back of the heme pocket after dissociation. Utilsky and Elber (1994) examined argon (Ar) diffusion, as well as NO recombination, in myoglobin using a binary collision method (cLES) as a correction to LESMD to give better estimates of the actual escape times. The cLES procedure produced somewhat different escape routes for Ar because the ligand is at a lower temperature and, thus, may only be finding the most probable pathways. Several additional studies (Carlson et al., 1994; Quillin et al., 1995) have examined the effects of mutations on ligand diffusion in myoglobin, showing that the distal residues have a significant effect on NO recombination, in agreement with experimental results. The LESMD routine of MOIL was also used to show that the rigid shift of the C helix in leghemoglobin provides a ligand exit path (Czerminski and Elber, 1991) not observed in myoglobin (Elber and Karplus, 1990). More recently, the NAMD program (Phillips et al., 2005; www.ks.uiuc. edu/Research/namd/) has been used to determine the ligand escape paths of several globins. This program offers the advantage of relative ease of use compared to other molecular dynamics packages. Orlowski and Nowak (2007a) examined oxygen paths in the minihemoglobin from the worm Cerebratulus lacteus. Phe 10 and Tyr 48 formed gates for the ligand to leave the distal pocket and access the large channel seen in the X-ray crystal structure (Pesce et al., 2002). LESMD results suggest at least six pathways for
462
Stephen D. Golden and Kenneth W. Olsen
ligand escape. Only two of these paths use the large channel. Multiple oxygen escape routes were also found in human cytoglobin using LESMD as implemented in NAMD (Orlowski and Nowak, 2007b). Five cavities and four distinct ligand exit paths were found in this study. Four of the cavities correspond to the Xe-binding sites found previously by X-ray crystallography (De Sanctis et al., 2004).
2. Molecular Dynamics Locally enhanced sampling molecular dynamics uses molecular mechanics for the basis of calculations. A discussion of the molecular mechanics energy function can be found in the chapter on conjugate peak refinement (CPR). The molecular mechanics potential function is used to calculate energies during molecular dynamics simulation. In the broadest sense, molecular dynamics (MD) is concerned with molecular motion. Biomolecules exhibit a wide range of timescales over which specific processes occur. Local motions, such as atomic fluctuations, side chain motions, and loop motions, occur within 1015 to 101 s with magnitudes of 0.01 to 5 A˚. Rigid body motions, such as helix motions, domain movements, and subunit motions, occur in a range of 109 to 1 s with magnitudes of 1 to 10 A˚. Large-scale movements, such as helix-coil transitions, dissociation/association, and folding and ˚ . MD unfolding, occur within 107 to 104 s with magnitudes of >5 A is a method for simulating the thermodynamic behavior of molecules using their forces, velocities, and positions. Newtonian molecular dynamics, which is used in the current study, uses the laws of classical mechanics, Newton’s equations of motion, to study the movement of molecules:
Fi ¼ m i a i ; where Fi is the force exerted on a particle i, mi is the mass of the particle I, and ai is its acceleration. Knowing the force on each atom, it is possible to determine the acceleration of each atom in the system. Integration of the equations of motion then yields a trajectory that describes the positions, velocities, and accelerations of the particles as they vary with time. From this trajectory, the average values of properties can be determined. MD is deterministic; once the positions and velocities of each atom are known, the state of the system can be predicted at any time in the future or the past. The force on an atom can be calculated from the change in energy between its current position and its position a small distance away. This can be recognized as the derivative of the energy with respect to the change in the position of the atom:
463
Locally Enhanced Sampling Pathways in Globins
@E ¼ Fi ; @ri
where E is the potential energy of the system and r is the position of particle i. Newton’s equation of motion can then relate the derivative of the potential energy to the changes in position as a function of time. Energies are calculated using the CHARMM potential energy function. Calculating the atomic forces and knowing the atomic masses can then be used to calculate the positions of each atom along a series of extremely small time steps, on the order of femtoseconds. The use of this method to compute trajectories is clearer when Newton’s equation is expressed in the following form:
@E @ 2 ri ¼ mi 2 : @t @ri
In practice, trajectories are not directly obtained from Newton’s equation due to lack of an analytical solution. First, the accelerations are computed from the forces and masses. Next the velocities are calculated from the accelerations based on the following relationship:
ai ¼
@vi : @t
Taking the simple case where the acceleration is constant, an expression for the velocity after integration is obtained:
v ¼ at þ v0 and since
vi ¼
@ri ; @t
we can once again integrate to obtain
r ¼ vt þ r0 : Combining this equation with the expression for the velocity, the following relationship is obtained giving the value of r at time t as a function of the acceleration, a, the initial position, r0 , and the initial velocity, v0:
r ¼ at 2 þ v0 t þ r0 : The acceleration is given as the derivative of the potential energy with respect to the position, r,
464
Stephen D. Golden and Kenneth W. Olsen
a¼
1 @E : mi @ri
Therefore, to calculate a trajectory, the only values that are needed are the initial positions of the atoms, an initial distribution of velocities, and the acceleration, which is determined by the gradient of the potential energy function. The equations of motion are deterministic; the positions and the velocities at time zero determine the positions and velocities at all other times, t. The initial positions can be obtained from experimental structures, such as X-ray crystal structures or solution structures determined by nuclear magnetic resonance (NMR) spectroscopy. The initial distribution of velocities is usually determined from a random distribution with the magnitudes conforming to the required temperature and corrected so there is no overall momentum:
P¼
N X
mi vi ¼ 0:
i¼1
The velocities, vi, are often chosen randomly from a Maxwell–Boltzmann or Gaussian distribution at a given temperature, which gives the probability that atom i has a velocity vx in the x direction at a temperature T:
pðvix Þ ¼
mi 2pkB T
12
1 mi vix2 : exp 2 kB T
The temperature can be calculated from the velocities using the relation N 1 X jpi j T¼ ; ð3N Þ i¼1 2mi
where N is the number of atoms in the system. The potential energy is a function of the atomic positions (3N ) of all the atoms in the system. Because of the complicated nature of this function, there is no analytical solution to the equations of motion. They must be solved numerically. Numerous numerical algorithms have been developed for integrating the equations of motion, such as the Verlet algorithm and the leap-frog algorithm (Verlet, 1967). MD generates information at the microscopic level, including atomic positions and velocities. The conversion of this microscopic information to macroscopic observables, such as pressure, energy, and heat capacities, requires statistical mechanics.
Locally Enhanced Sampling Pathways in Globins
465
A typical molecular dynamics simulation involves the following basic steps. The preliminary preparation requires a molecular structure with all Cartesian coordinates defined. This structure is usually obtained experimentally from X-ray diffraction or NMR studies. After determining the internal coordinate values of the molecule, total energy as a function of Cartesian coordinates is computed by evaluating the individual terms of the potential energy function. Energy minimization of this structure is the next step in an MD study. Because the simulation begins with an initial structure that may be derived from experimental data, energy minimization is performed on structures prior to dynamics to relax the conformation and remove steric overlap that produces bad contacts. An energy-minimized structure represents the molecule at a temperature close to absolute zero. Heating, which is the next step in an MD simulation, is accomplished by initially assigning random velocities according to a Gaussian distribution appropriate for that low temperature and then running dynamics. The temperature is increased gradually by assigning greater random velocities to each atom at predetermined time intervals. The next step is equilibration. Equilibration is achieved by allowing the system to evolve spontaneously for a period of time and integrating the equations of motion until the average temperature and structure remain stable. This is facilitated by periodically reassigning velocities appropriate to the desired temperature. Generally, the procedure is continued until various statistical properties of the system become independent of time. The final step is the production run. The equilibrated structure is its starting point. In a typical simulation, the trajectory traces the motions of the molecule through a period of at least 100 ps but longer runs are now very common. An optional sixth step in a molecular dynamics simulation is quenching. The opposite of heating, this step takes the molecule from the equilibrated temperature to zero. Quenching is a form of minimization, utilizing molecular dynamics to slowly remove all kinetic energy from the system.
3. Locally Enhanced Sampling Molecular Dynamics At room temperature, normal nanosecond-length molecular dynamics simulations have difficulty overcoming barriers to conformational transitions and may only sample conformations in the neighborhood of the initial structure. Among the various techniques used to enhance sampling during a simulation is LESMD (Elber and Karplus, 1990). Locally enhanced sampling (LES) is a mean field approach that enhances the sampling of the conformation space as a consequence of the presence of multiple copies and their higher (with respect to ‘‘cold’’ protein) effective temperature (Roitberg and
466
Stephen D. Golden and Kenneth W. Olsen
Elber, 1991). LES allows multiple copies of ligands to be placed into the simulation. These copies are generated in a way in which they do not interact with each other and interact with the rest of the system in an average way. During the simulation, the copies are free to move apart and explore different regions of conformational space, thereby increasing the statistical sampling. The barriers to conformational transitions in a LES system are reduced as compared to the original system, resulting in more frequent conformational changes (Roitberg and Elber, 1991). For example, the dioxygen ligand in Mt-trHbN has been replaced with 15 copies. Steric conflicts or other unfavorable interactions may have created high barriers. However, the rest of the system sees each of these 15 copies with a scaling factor of 1/15. If one copy is in an unfavorable conformation, the others may not be, and the effective barriers with a distribution of copies are less than with the single copy. A key feature is that the energy function is modified such that the energy is identical to that of the original system when all LES copies have the same coordinates. During the simulation, the copies are free to move apart and explore different regions of conformational space, thereby increasing the statistical sampling. This means that multiple trajectories can be obtained for the region of interest while carrying out only a single simulation. If the LES region is a small part of the system (such as a diatomic ligand in a protein), then the additional computational effort from the added LES molecules will be a small percentage of the total number of atoms, and the multiple trajectories will be obtained with a small additional computational effort. Another key feature of the LES system is that the global energy minimum occurs when all copies occupy the position of the global energy minimum in the original system (Roitberg and Elber, 1991). This means that optimization of the LES system directly provides information about the original system without complicated mapping procedures. Another major advantage of LES over alternate methods to reduce barriers or improve sampling is that it is compatible (Simmerling and Elber, 1994; Simmerling et al., 1998) with simulation techniques such as molecular dynamics in explicit aqueous solvation and the particle-mesh Ewald technique for accurate treatment of long-range electrostatic interactions.
4. Methods 4.1. Structure preparation Coordinates for the structures are obtained from the Protein Data Bank (Berman et al., 2000; www.rcsb.org/pdb). The VMD molecular graphic program (Humphrey et al., 1996) is used to prepare the structure for LESMD simulation. The charge states of the histidine residues are determined by visual inspection of their surroundings. In the examples given
Locally Enhanced Sampling Pathways in Globins
467
here, two structures for Mt-trHbN (Milani et al., 2001) (PDB: 1idr) were built. The first structure includes only subunit A, while the second structure includes both subunits of the dimer. The Mt-trHbN and Pc-trHb (Pesce et al., 2000) (PDB: 1dlw) models are embedded in a TIP3P water box of 2115 waters, and Ce-trHb (Pesce et al., 2000) (PDB: 1dly) is embedded in ˚ . Missing a water box of 2284 waters, with dimension 52 54 57 A residues are not built back into the LESMD structures. Fifteen copies of the ligand are inserted using the multiply command in VMD. It is important to check that each atom of the multiple copies is flagged with the same number in the B column of the output coordinate (pdb) file. Both coordinate (pdb) and protein structure (psf) files, which are needed for the simulation, can be generated by VMD using either menus or appropriate tcl scripts.
4.2. Minimization and equilibration Locally enhanced sampling molecular dynamics calculations are performed using NAMD (Phillips et al., 2005) and the CHARMM force field (MacKerell et al., 1998). All systems are minimized with the backbone fixed for 1000 steps and then all atoms are minimized for another 1000 steps using the conjugate gradient algorithm. Occasionally, longer minimizations are needed. Usually, the step size is set at 2 fs with the rigid bonds parameter activated, although a 1-fs time step can be used without the rigid bonds parameter. The temperature of the simulation is then increased from 10 to 310 K in 10-K increments over 30,000 steps. The volume is then equilibrated for 10,000 steps by turning the Langevin piston on. Finally the entire system is equilibrated at constant temperature (310 K) and pressure (1 atm) for at least 30,000 steps. Then the production dynamics, which are used in the LESMD analysis, are run. The length of the production run depends on how rapidly the ligands escape the protein. Generally, several nanoseconds of simulation are needed so that most of the ligands escape. Occasionally a ligand will rebind to the protein during the simulation and often the simulation is stopped before all the ligands have escaped. All systems are simulated with periodic boundary conditions using the particlemesh Ewald method. The original LESMD method can give an incorrect partitioning of energy between the protein and the enhanced ligand (Karplus and Straub, 1991). In MOIL, this is avoided using the cLES procedure (Utilsky and Elber, 1994). In NAMD, the temperature should be monitored carefully and kept constant by a Langevin dynamics protocol (Orlowski and Nowak, 2007b). All runs are done at constant volume and temperature. A complete description of input parameters is given in the NAMD documentation (www.ks.uiuc.edu/Research/namd/).
468
Stephen D. Golden and Kenneth W. Olsen
4.3. Classification of ligand-binding pathways Pathways are placed into families or groups of paths by the helices through which the ligand traveled to escape the binding pocket. A contact analysis is conducted to define the binding pathways. The contact ‘‘collision’’ is based on the distance between the ligand and the protein heavy atom that is less than 4 A˚ where the collision radius is a mean value based on van der Waals radii of the heavy atoms (Elber and Karplus, 1990). An example of the contact analysis script can be found in the chapter on CPR.
5. Results 5.1. Mt-trHbN 5.1.1. Subunit A For the entirety of the simulation, the ligands explored, exited, and reentered the AB turn/GH turn and G helix/H helix tunnel cavities. After 960 ps, all dioxygens diffused out through the protein interior using the AB turn/GH turn and the G helix/H helix tunnel branches. Two dioxygens remained in the AB turn/GH turn G helix/H helix tunnel cavities at the end of the simulation. Fourteen of the 15 ligands exited out of the shorter G helix/H helix tunnel branch into the solvent. Only 1 ligand traveled the AB turn/GH turn tunnel branch to exit the protein. During the trajectory, 6 of the ligands had exited the AB turn/GH turn tunnel branch only to reenter the protein and travel back down the AB turn/GH turn tunnel branch to finally exit the G helix/H helix tunnel. Root mean square deviations in reference to the minimized starting structure were calculated for binding pocket residues for each step in the trajectory. Binding pocket residues Tyr 33, Gln 58, and Phe 32 exhibited maximum motions of 1.6, 1.2, and 1.5 ˚ , respectively. Phe 62 showed a maximum movement of 2.3 A˚. A 5.1.2. Subunits A and B Throughout the trajectory, ligands from both subunits explored all their respective tunnel systems prior to exiting into the solvent. After 1.4 ns, a total of six ligands remained in the interior of the protein, four in subunit A and two in subunit B. Subunit A had three ligands exiting the AB turn/GH turn tunnel, four exiting the G helix/H helix tunnel, three out of the B helix/E helix tunnel, and one ligand exiting the E helix/H helix tunnel. Subunit B had six ligands exiting the AB turn/GH turn tunnel, five exiting the G helix/H helix tunnel, one out of the B helix/E helix tunnel, and one ligand exiting the E helix/H helix tunnel (Fig. 23.1). In the first 90 ps of the trajectory, three ligands exited subunit A, one from B helix/E helix, one from G helix/H helix, and one from the AB turn/GH turn tunnel, and no
Locally Enhanced Sampling Pathways in Globins
469
Figure 23.1 Results of LESMD simulation of the dimer displaying the extended tunnel system in sidebys stereo. AB turn/GH turn path (red), G helix/H helix path (blue), E helix/H helix path (green), and B helix/E helix (orange). Helices are represented as subunit A (ice blue) and subunit B (red). The heme is represented in licorice. Key residues are displayed as licorice as follows:Tyr 33 (green), Gln 58 (orange), Phe 32 (purple), and Phe 62 (purple). Multiple copy dioxygens areVDWspheres.
ligands had escaped subunit B. During one portion of the trajectory, one ligand from subunit B exited the G helix/H helix tunnel into the solvent and traveled through the solvent to enter into subunit A through the B helix/E helix tunnel. Subunit A binding pocket residues Tyr 33, Gln 58, ˚ , respecand Phe 32 exhibited maximum motions of 1.8, 1.0, and 1.7 A tively. Subunit B binding pocket residues Tyr 33, Gln 58, and Phe 32 exhibited maximum motions of 1.3, 1.4, and 1.5 A˚, respectively. Phe 62 ˚ in subunit A and 3.5 A ˚ in showed maximum movements of 2.2 A subunit B.
5.2. Pc-trHb Throughout the LESMD trajectory, the ligand water molecules utilized four pathway tunnels. These tunnels are defined by the helices through which the ligand travels to escape from the interior of the protein and travel out into the solvent. These pathway tunnels are the CD turn/G helix, AB
470
Stephen D. Golden and Kenneth W. Olsen
turn/GH turn, E helix/H helix, and the E helix/heme (Fig. 23.2). After 407 ps, a total of 14 water molecules had exited the protein into the solvent. One single water molecule attempted to exit the CD turn/G helix tunnel but remained in the interior of the protein at the end of the simulation. Seven ligands exited the CD turn/G helix, five via the AB turn/GH turn tunnel, two exited the E helix/H helix tunnel, and one exited out of the E helix/ heme tunnel. In the initial 56 ps of the simulation, two ligands exited the AB turn/GH turn tunnel. Upon completion of the first 77 ps of the simulation, one of the previously released ligands reentered the AB turn/GH turn tunnel. The next 82 ps of simulation contained five ligands exiting the protein using all four tunnels. Within the next 86 ps of simulation, six ligands exited either the AB turn/GH turn tunnel or the CD turn/G helix tunnel. This portion of the simulation also contained two ligands, which reentered the protein after being in the solvent through the AB turn/GH turn tunnel. During this same portion of the trajectory, a different water molecule exited out of the E helix/H helix tunnel, reentered the E helix/H helix tunnel, and then traveled through the AB turn/GH turn tunnel to exit into the solvent. Yet another ligand explored three of the four tunnels prior to exiting the protein, the AB turn/GH turn tunnel, the E helix/H helix tunnel, and the E helix/heme tunnel, and then finally exited the E helix/H helix tunnel. The next 50 ps contained three ligands exiting the E helix/H helix and CD turn/G helix tunnels. The remaining 100 ps consisted of a single water molecule exploring the binding pocket cavity and the CD turn/G helix tunnel but never fully exiting the protein. Binding pocket residues Tyr 20, Phe 33, Gln 41, and Thr 45 exhibited maximal motions of 1.0, 1.2, 1.0, and 1.2 A˚, respectively. No solvent water molecules were observed entering the tunnel cavities.
Figure 23.2 Results of LESMD simulation displaying the extended tunnel system in side-by-side stereo. AB turn/GH turn path (red), CD turn/G helix path (blue), E helix/ heme path (green), and E helix/H helix (red). Helices are represented as ribbons. The heme is represented as licorice. Multiple copy water molecules are represented as VDW spheres (red, green, blue).
Locally Enhanced Sampling Pathways in Globins
471
Figure 23.3 Results of LESMD simulation displaying the extended tunnel system in side-by-side stereo. AB turn/GH turn path (red), G helix/H helix path (green), and EF turn/H helix path (orange). Helices are represented as ribbons.The heme is represented as licorice. Multiple copy cyanide molecules are represented as multicoloredVDWspheres.
5.3. Ce-trHb Throughout the LESMD trajectory the cyanide anions utilized three pathway tunnels. These tunnels are defined by the helices through which the ligand travels to escape from the interior of the protein and travel out into the solvent. These pathway tunnels are the AB turn/GH turn, the EF turn/H helix, and the G helix/H helix (Fig. 23.3). After 905 ps, all 15 cyanide anions had exited the protein into the solvent. Nine ligands exited the EF turn/H helix, five via the AB turn/GH turn tunnel, and one via the G helix/H helix tunnel. The first ligand was not released until 455 ps into the simulation. During this portion of the simulation, one ligand exited out of the EF turn/H helix tunnel and three out of the AB turn/GH turn tunnel. The next 401 ps displayed eight ligands exiting the EF turn/H helix tunnel, two via the AB turn/GH turn tunnel, and one exiting the G helix/H helix tunnel. One of the ligands within this portion of the trajectory exited the EF turn/H helix tunnel into the solvent and then reentered the EF turn/H helix tunnel only to exit out the AB turn/GH turn tunnel later in the simulation. During the duration of the simulation, ligands explored the binding pocket as well as the entire tunnel system. Binding pocket residues Tyr 20, Phe 33, Gln 41, and Gln 45 ˚ , respectively. exhibited maximal motions of 2.0, 1.3, 1.7, and 2.6 A
6. Conclusions The comparison of LESMD results for ligand binding to the three truncated hemoglobins is shown in Table 23.1. At least two paths (AB turn/ GH turn and E helix/H helix) are found in all three proteins. Most of these
472
Stephen D. Golden and Kenneth W. Olsen
Table 23.1 Comparison of LESMD results for three truncated hemoglobins Mt-trHbN
Pc-trHb
Ce-trHb
AB turn/GH turn G helix/H helix E helix/H helix B helix/E helix
AB turn/GH turn
AB turn/GH turn G helix/H helix E helix/H helix
E helix/H helix CD turn/G helix E helix/heme
paths are also predicted by implicit ligand sampling and/or CPR algorithms (see the previous two chapters). The fact that similar paths for ligand binding are found by these very different computational algorithms suggests that the level of confidence in these results is high. The three theoretical techniques outlined in this and the previous two chapters are quite different in concept. Each has specific uses where it is the preferred method. CPR (Fischer and Karplus, 1992) gives a minimum energy path for ligand binding. This is not the actual path that any specific ligand takes into the binding site. However, the CPR path does identify the essential movements that must occur along a specific path. This can be very useful in designing mutational experiments. The major disadvantages of the method are that the external starting positions for the ligand are unknown and that there is no explicit solvent in the calculation. The first problem requires many simulations with different starting positions to obtain a reasonable picture of the possible binding pathways. This is intensive in computer resources. Alternatively, CPR can be used to determine the minimum energy path along a route suggested by either implicit ligand sampling or LESMD. The second problem with CPR can be solved by using an implicit solvent model in the calculation. Implicit ligand sampling (Cohen et al., 2006) has the advantage of giving all possible pathways for ligand escape in a single simulation. The disadvantage is that the method is limited to small, hydrophobic ligands, such as oxygen. LESMD shows the actual paths taken by individual ligand molecules. This provides information not only on the escape route, but also on the relative amount of time spent in internal cavities of the protein. Thus, the three computational techniques are complementary and, taken together, can provide a more complete picture of ligand-binding pathways in globins.
REFERENCES Berman, H. M., Westbrook, J., Feng, Z., Gilliland, G., Bhat, T. N., Weissig, H., Shindyalor, I. N., and Bourne, P. E. (2000). The Protein databank. Nucleic Acids Res. 28, 235–242.
Locally Enhanced Sampling Pathways in Globins
473
Brooks, B. R., Bruccoleri, R. E., Olafson, B. D., States, D. J., Swaminathan, S., and Karplus, M. (1983). CHARMM: A program for macromolecular energy, minimization, and dynamics calculations. J. Comput. Chem. 4, 187–217. Carlson, M. L., Regan, R., Elber, R., Li, H., Phillips, G. N., Jr., Olson, J. S., and Gibson, Q. H. (1994). Nitric oxide recombination to double mutants of myoglobin: Role of ligand diffusion in a fluctuating heme pocket. Biochemistry 33, 10597–10606. Chiancone, E., Elber, R., Royer, W. E., Jr., Regan, R., and Gibson, Q. H. (1993). Ligand binding and conformation change in the dimeric hemoglobin of the clam Scapharca inaequivalvis. J. Biol. Chem. 268, 5711–5718. Cohen, J., Arkhipov, A., Braun, R., and Schulten, K. (2006). Imaging the migration pathways for O2, CO, NO, and Xe inside myoglobin. Biophys. J. 91, 1844–1857. Couture, M., Chamberland, H., St. Pierre, B., Lafontaine, J., and Guertin, M. (1994). Nuclear genes encoding chloroplast hemoglobins in the unicellular green alga Chlamydomonas eugametos. Mol. Gen. Genet. 243, 185–197. Crespo, A., Martı´, M. A., Kalko, S. G., Morreale, A., Orozco, M., Gelpi, J. L., Luque, F. J., and Estrin, D. A. (2005). Theoretical study of the truncated hemoglobin HbN: Exploring the molecular basis of the NO detoxification mechanism. J. Am. Chem. Soc. 127, 4433–4444. Czerminski, R., and Elber, R. (1991). Computational studies of ligand diffusion in globins. I. Leghemoglobin. Proteins 10, 70–80. Dantsker, D., Samuni, U., Friedman, J. M., and Agmon, N. (2005). A hierarchy of functionally important relaxations within myoglobin based on solvent effects, mutations and kinetic model. Biochim. Biophys. Acta 1749, 234–251. Das, T. K., Weber, R. E., Dewilde, S., Wittenberg, J. B., Wittenberg, B. A., Yamauchi, K., Van Hauwaert, M. L., Moens, L., and Rousseau, D. L. (2000). Ligand binding in the ferric and ferrous states of paramecium hemoglobin. Biochemistry 39, 14330–14340. De Sanctis, D., Dewilde, S., Pesce, A., Moens, L., Ascenzi, P., Hankeln, T., Burmester, T., and Bolognesi, M. (2004). Crystal structure of cytoglobin: The fourth globin type discovered in man displays heme hexa-coordination. J. Mol. Biol. 336, 917–927. Elber, R., and Karplus, M. (1990). Enhanced sampling in molecular dynamics: Use of the time-dependent hartree approximation for a simulation of carbon monoxide diffusion through myoglobin. J. Am. Chem. Soc 112, 9161–9175. Elber, R., Roitberg, A., Simmerling, C., Goldstein, R., Li, H., Verkhivker, G., Keasar, C., Zhang, J., and Utilsky, A. (1995). MOIL: A program for simulations of macromolecules. Comput. Phys. Comm. 91, 159–189. Fischer, S., and Karplus, M. (1992). Conjugate peak refinement: An algorithm for finding reaction paths and accurate transition states in systems with many degrees of freedom. Chem. Phys. Lett. 194, 252–261. Gibson, Q. H., Regan, R., Elber, R., Olson, J. S., and Carver, T. E. (1992). Distal pocket residues affect picosecond ligand recombination in myoglobin. J. Biol. Chem. 267, 22022–22034. Humphrey, W., Dalke, A., and Schulten, K. (1996). VMD: Visual Molecular Dynamics. J. Mol. Graph. 14, 33–38. Hvitved, A. N., Trent, J. T., 3rd, Premer, S. A., and Hargrove, M. S. (2001). Ligand binding and hexacoordination in synechocystis hemoglobin. J. Biol. Chem. 276, 34714–34721. Li, H., Elber, R., and Straub, J. E. (1993). Molecular dynamics simulation of NO recombination to myoglobin mutants. J. Biol. Chem. 268, 17908–17916. MacKerell, A. D., Jr., Bashford, D., Bellott, M., Dunbrack, R. L., Jr., Evanseck, J., Field, M. J., Fischer, S., Gao, J., Guo, H., Ha, S., Joseph, D., Kuchnir, L., et al. (1998). All-atom empirical potential for molecular modeling and dynamics studies of proteins. J. Phys. Chem. B 102, 3586–3616.
474
Stephen D. Golden and Kenneth W. Olsen
Milani, M., Pesce, A., Ouellet, Y., Ascenzi, P., Guertin, M., and Bolognesi, M. (2001). Mycobacterium tuberculosis hemoglobin N displays a protein tunnel suited for O2 diffusion to the heme. EMBO J. 20, 3902–3909. Milani, M., Pesce, A., Ouellet, Y., Dewilde, S., Friedman, J., Ascenzi, P., Guertin, M., and Bolognesi, M. (2004). Ligand tunneling in truncated hemoglobins. J. Biol. Chem. 279, 21520–21525. Moens, L., Vanfleteren, J., Van de Peer, Y., Peeters, K., Kapp, O., Czeluzniak, J., Goodman, M., Blaxter, M., and Vinogradov, S. (1996). Globins in nonvertebrate species: Dispersal by horizontal gene transfer and evolution of the structure-function relationship. Mol. Biol. Evol. 13, 324–333. Mouawad, L., Marechal, J., and Perahia, D. (2005). Internal cavities and ligand passageways in human hemoglobin characterized by molecular dynamics simulations. Biochim. Biophys. Acta 1724, 385–393. Mukai, M., Ouellet, Y., Ouellet, H., Guertin, M., and Yeh, S. (2004). No binding induced confirmational changes in a truncated hemoglobin from Mycobacterium tuberculosis. Biochemistry 43, 2764–2770. Orlowski, S., and Nowak, W. (2007a). Oxygen diffusion in minihemoglobin from Cerebratulus lacteus: A locally enhanced sampling study. Theor. Chem. Acc. 117, 253–258. Orlowski, S., and Nowak, W. (2007b). Locally enhanced sampling molecular dynamics study of the dioxygen transport in human cytoglobin. J. Mol. Model. 13, 715–723. Pathania, R., Navani, N. K., Gardner, A. M., Gardner, P. R., and Dikshit, K. L. (2002). Nitric oxide scavenging and detoxification by the Mycobacterium tuberculosis haemoglobin, HbN in Escherichia coli. Mol. Microbiol. 45, 1303–1314. Pesce, A., Couture, M., Dewilde, S., Guertin, M., Yamauchi, K., Ascenzi, P., Moens, L., and Bolognesi, M. (2000). A novel two-over-two a-helical sandwich fold is characteristic of the truncated hemoglobin family. EMBO J. 19, 2424–2434. Pesce, A., Nardini, M., Dewilde, S., Geusens, E., Yamauchi, K., Ascenzi, P., Riggs, A., Moens, L., and Bolognesi, M. (2002). The 109 residue nerve tissue minihemoglobin from Cerebratulus lacteus highlights striking structural plasticity of the a-helical globin fold. Structure 10, 725–735. Phillips, J. C., Braun, R., Wang, W., Gumbart, J., Tajkhorshid, E., Villa, E., Chipot, C., Skeel, R. D., Kale, L., and Schulten, K. (2005). Scalable molecular dynamics with NAMD. J. Comp. Chem. 26, 1781–1802. Quillin, M. L., Li, T., Olson, J. S., Phillips, G. N., Jr., Dou, Y., Ikeda-Saito, M., Regan, R., Carlson, M., Gibson, Q. H., Li, H., and Elber, R. (1995). Structural and functional effects of apolar mutations of the distal valine in myoglobin. J. Mol. Biol. 245, 416–436. Roitberg, A., and Elber, R. (1991). Modelling side chains in peptides and proteins: Application of the locally enhanced sampling and simulated annealing methods to find minimum energy conformations. J. Chem. Phys. 95, 9277–9287. Samuni, U., Dantsker, D., Ray, A., Wittenberg, J. B., Wittenberg, B. A., Dewilde, S., Moens, L., Ouellet, Y., Guertin, M., and Friedman, J. M. (2003). Kinetic modulation in carbonmonoxy derivatives of truncated hemoglobins. J. Biol. Chem. 278, 27241–27250. Simmerling, C., and Elber, R. (1994). Hydrophobic ‘‘collapse’’ in a cyclic hexapeptide: Computer simulations of CHDLFC and CAAAAC in water. J. Am. Chem. Soc. 116, 2534–2547. Simmerling, C., Fox, T., and Kollman, P. A. (1998). Use of locally enhanced sampling in free energy calculations: Testing and application to the a!b anomerization of glucose. J. Am. Chem. Soc. 120, 5771–5782. Straub, J., and Karplus, M. (1991). Energy equipartioning in the classical time-dependent Harfree approximation. J. Chem. Phys. 94, 6737–6739.
Locally Enhanced Sampling Pathways in Globins
475
Utilsky, A., and Elber, R. (1993). The thermal equilibrium aspects of the time dependent Hartree and the locally enhanced sampling approximation: Formal properties, a correction, and computational examples for rare gas clusters. J. Chem. Phys. 88, 3380–3388. Utilsky, A., and Elber, R. (1994). Application of the locally enhanced sampling (LES) and a mean field with a binary collision correction (cLES) to the simulation of Ar diffusion and NO recombination in myoglobin. J. Chem. Phys. 98, 1034–1043. Verlet, L. (1967). Computer ‘‘experiments’’ on classical fluids. I. Thermodynamical properties of Lennard-Jones molecules. Phys. Rev. 159, 98–103. Watts, R. A., Hunt, P. W., Hvitved, A. N., Hargrove, M. S., Peacock, W. J., and Dennis, E. S. (2001). A hemoglobin from plants homologous to truncated hemoglobins of microorganisms. Proc. Natl. Acad. Sci. USA 98, 10119–10124. Weber, R. E., and Vinogradov, S. N. (2001). Nonvertebrate hemoglobins: Functions and molecular adaptations. Physiol. Rev. 81, 569–628. Wittenberg, J., Bolognesi, M., Wittenberg, B., and Guertin, M. (2002). Truncated hemoglobins: A new family of hemoglobins widely distributed in bacteria, unicellular eukaryotes, and plants. J. Biol. Chem. 277, 871–874. Yeh, S. R. (2004). A novel intersubunit communication mechanism in a truncated hemoglobin from Mycobacterium tuberculosis. J. Phys. Chem. B 108, 1478–1484. Yeh, S. R., Couture, M., Ouellet, Y., Guertin, M., and Rousseau, D. L. (2000). A cooperative oxygen binding hemoglobin from Mycobacterium tuberculosis: Stabilization of heme ligands by a distal tyrosine residue. J. Biol. Chem. 275, 1679–1684.
This page intentionally left blank
C H A P T E R
T W E N T Y- F O U R
Nitric Oxide Reactivity with Globins as Investigated Through Computer Simulation Marcelo A. Marti,* Luciana Capece,* Axel Bidon-Chanal,† Alejandro Crespo,* Victor Guallar,‡ F. Javier Luque,† and Dario A. Estrin* Contents 1. Introduction 2. Molecular Dynamics (MD) Methods 2.1. Setup of the system 2.2. Equilibration 2.3. Essential dynamics 2.4. Free energy profile calculations 2.5. Protein energy landscape exploring 2.6. Heme group parameters 3. Quantum Mechanical-Molecular Mechanical Methods 3.1. Selection of the QM subsystem 3.2. Quantum mechanical-molecular mechanical optimizations 3.3. Binding energy calculations 3.4. Reaction pathway search 4. Illustrative Examples 4.1. Structural flexibility of globins as studied by MD simulations: Mycobacterium tuberculosis truncated hemoglobin N 5. Ligand Migration Profiles from MSMD and PELE Simulations: Exploring Ligand Entry Pathways in M. tuberculosis trHbN 5.1. Oxygen affinity of wild-type and mutant M. tuberculosis trHbN as compared to myoglobin 5.2. Chemical reactions between NO and globins: NO detoxification in M. tuberculosis trHbN
* { {
478 479 480 481 481 482 484 485 485 486 486 487 487 488 488 490 492 493
Departamento de Quı´mica Inorga´nica, Analı´tica y Quı´mica Fı´sica/INQUIMAE-CONICET, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Ciudad Universitaria, Buenos Aires, Argentina Departament de Fisicoquı´mica, Facultat de Farma`cia, Universitat de Barcelona, Barcelona, Spain Catalan Institute for Research and Advanced Studies (ICREA), Computational Biology Program, Barcelona Supercomputing Center, Barcelona, Spain
Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37024-9
#
2008 Elsevier Inc. All rights reserved.
477
478
Marcelo A. Marti et al.
6. Conclusions Acknowledgments References
494 495 495
Abstract This chapter reviews the application of classical and quantum-mechanical atomistic simulation tools used in the investigation of several relevant issues in nitric oxide reactivity with globins and presents different simulation strategies based on classical force fields: standard molecular dynamics, essential dynamics, umbrella sampling, multiple steering molecular dynamics, and a novel technique for exploring the protein energy landscape. It also presents hybrid quantum-classical schemes as a tool to obtain relevant information regarding binding energies and chemical reactivity of globins. As illustrative examples, investigations of the structural flexibility, ligand migration profiles, oxygen affinity, and reactivity toward nitric oxide of truncated hemoglobin N of Mycobacterium tuberculosis are presented.
1. Introduction Computational techniques for modeling large biological molecules have emerged during the last decades as important tools to complement experimental information. The in silico-generated models and the information obtained by these studies are essential to complement the structural, energetic, and kinetic data of biological systems obtained through experimental methods and to shed light onto the relationships between structure and function. Computer simulations also afford a systematic and economical tool to analyze the dependence of the property of interest not only on the static structure (e.g., amino acid sequence), but also on dynamic behavior. Moreover, because of the increase in computer power and the accuracy of the models, in silico experiments are valuable to propose new hypothesis and to draw biologically relevant conclusions about the molecular mechanisms that operate in biological systems (Karplus and Petsko, 1990; Leach, 2001). Modeling of biological processes that do not involve formation and/or breaking of chemical bonds can be achieved by employing classical force fields. Among the most widely used biomolecular force fields are AMBER (Perlman et al., 1995) and CHARMM (MacKerell et al., 1998). Although the timescales accessible in atomistic simulations with present hardware technology are still limited to the nanosecond/microsecond range, the predictive power of simulation schemes applied to biomolecules has increased by the development of enhanced sampling techniques. However, reactive processes are a key ingredient in understanding nitric oxide (NO) physiology. In these cases, one has to resort to quantum
479
Computer Simulation of Globin Reactivity
mechanical (QM)-based schemes. There are two main strategies for the investigation of reactive chemical processes in biomolecules. The first one consists of performing QM electronic structure calculations on adequate model systems, which are chosen to represent the main features of the active site and eventually the most relevant region of the surrounding environment. The second strategy is to employ hybrid quantum mechanical/molecular mechanical (QM-MM) schemes (Capece et al., 2006; Crespo et al., 2003; Elola et al., 1999; Friesner and Guallar, 2005; Guallar and Friesner, 2004; Warshel and Levitt, 1976), which represent the chemically relevant part of the system at the QM level using different strategies (valence bond theory, semiempirical, or Hartree–Fock methods and density functional theory), whereas the rest of the system is treated at the less expensive MM level. This work describes the implementation of several computational methods partly developed by our groups, based on the schemes mentioned earlier, to investigate the molecular basis of NO reactivity with heme proteins. Selected examples are given to illustrate the capabilities of these methods. Finally, a critical analysis of the described computational schemes is presented.
2. Molecular Dynamics (MD) Methods Molecular dynamics simulations have become a powerful tool in computational biology and are widely used to obtain information about processes involving conformational changes. In MD simulations, atoms are treated as charged spheres connected by springs and the whole system is described by a potential energy function, that is, the so-called force field, composed of various terms, each of which represents a portion of the interactions present in the system. The basic terms describe bond distances, angles, dihedrals, electrostatic (Coulombic), and other nonbonded interactions between atoms, and there might be also additional terms used to describe specific interactions such as hydrogen bonds or nuclear magnetic resonance restraints. The AMBER package (Perlman et al., 1995), for example, uses a simple but very efficient potential energy function, given by
UðRÞ ¼
X bonds
Kr ðr req Þ2 þ
X
Ky ðy yeq Þ2
angles
X Vn ð1þcos½nf gÞ þ 2 dihedrals ! atoms atoms X X Aij Bij qi qj þ þ 6 12 Rij Rij eRij i<j i<j
ð24:1Þ
480
Marcelo A. Marti et al.
In Eq. (24.1), the first three terms describe the so-called bonded interactions, namely bond stretching, angle bending, and dihedral torsions, respectively. The last two terms correspond to van der Waals and electrostatic potentials, the so-called nonbonded interactions. The parameters Kr, Ky, and Vn represent the force constants for stretching and bending of bonds, and the barrier for torsional angles, respectively. The subindex eq specifies the values of bond length and angle at the equilibrium geometry, whereas n and g denote the periodicity and phase angle of torsions. Finally, A and B represent van der Waals parameters, q denotes the atomic partial charges, E is the permittivity, and Rij stands for interatomic distances. Particles in a given system must be defined with fixed parameters to account for the complete set of bonded and nonbonded interactions. Given a set of coordinates for the particles of the system, the dynamic behavior can be examined from the trajectory obtained by solving Newton’s equations of motion numerically with finite difference methods (Leach, 2001).
2.1. Setup of the system When starting the study of a given protein through MD simulations, one has to choose the starting structure and the force field. The starting structure is typically taken from the crystal structure of the protein, when available. When several structures are available, that solved at the highest resolution is usually the best choice. However, care must be taken in selecting the structure that most likely represents the desired simulation conditions. For example, myoglobin (Mb) is found crystallized in several oxidation and coordination states (FeIII, FeII, FeII-O2, FeII-CO, FeII-NO, among others) (Brunori et al., 2004, Chu et al., 2000, Copeland et al., 2003, 2006), and the best choice for simulation of the nitrosyl protein would be the crystallized Mb-NO structure (Copeland et al., 2003). If the precise desired state is unavailable, the most similar structure should be taken as the starting point, and the desired state is built through in silico modification. If the crystal structure of the system is not available, homology modeling is an option (Blundell, 1987). For a biochemical study, however, the level of identity/ similarity in the active site needs to be of the order of 70–80%. Once the initial structure and the force field have been chosen, the next step consists in setting up the system. As X-ray techniques usually do not provide information about the hydrogen atoms, positions, they have to be added to the experimental structure. Most MD packages have an automated H-adding function, although some user specifications are needed. Of particular relevance in globins is the histidine case, as this residue has three representations in most force fields. Two of them are neutral, one protonated in the NE and the other protonated in the Nd, whereas the third form is a positively charged doubly protonated histidine. Many globins such as
Computer Simulation of Globin Reactivity
481
Mb or hemoglobin (Hb) have two conserved histidines: the proximal histidine F8 is nonprotonated, as NE is coordinated to the iron of the heme group, and the distal histidine E7 is a protonated residue with the NEH group capable of forming a hydrogen bond with ligands. Clearly, this suffices to remark that a close inspection of the possible interactions is essential to describe the system properly. The last step in the setup of the system is the choice of the solvent, typically water, and the addition of ions to mimic a given ionic atmosphere. MD packages are able to perform MD simulations in different solvation conditions, such as implicit water with the generalized Born approximation (Onufriev et al., 2004), a cluster of explicit water models, and the insertion of the protein in a box containing water molecules, employing periodic boundary conditions to represent the bulk solvation.
2.2. Equilibration The equilibration protocol is critical to allow for initial relaxation and thermalization yielding an equilibrated structure at the desired simulation temperature. Most equilibration protocols perform an initial optimization of the system, followed by a slow heating up to the desired temperature using, for instance, the Berendsen thermostat (Berendsen et al., 1984). Similarly, the pressure can be kept fixed at a given value by coupling the system to a barostat (Berendsen et al., 1984). In our simulations, heating is generally performed in about 100 to 200 ps at constant volume. Once the system is equilibrated, the different MD runs are performed at constant temperature and pressure (NPT ensemble), as these conditions resemble the typical experimental situation.
2.3. Essential dynamics This technique determines the essential motions of the simulated system that explain most of the structural variance detected during the trajectory (Amadei et al., 1993). Technically, this implies diagonalization of the covariance matrix of the positional deviations given in Eq. (24.2), which affords a set of 3N (N ¼ number of atoms in the system) eigenvectors and their associated eigenvalues:
C ¼ covðxÞ ¼ hðx hxiÞðx hxiÞT i;
ð24:2Þ
where x denotes a 3N-dimensional vector of all atomic coordinates, and <>stands for an average over time. The eigenvectors can be considered to be 3N-dimensional vectors representing the nature of the essential motions. Each eigenvalue represents
482
Marcelo A. Marti et al.
the percentage of structural variance explained by each motion. By using harmonic approximations, eigenvalues can be translated into frequencies, which indicate the softness of a given essential motion. The eigenvectors of one trajectory can be compared with those of another trajectory, deriving quantitative measures of the similarity between the essential motions of two independent trajectories [Eq. (24.3)]:
gAB ¼
n X n 1X ðnA nB Þ2 ; n j¼1 i¼1 i j
ð24:3Þ
where n stands for the minimum number of eigenvectors, which explain more than a given threshold of the variance. Essential dynamics calculations are very useful in identifying structural fluctuations in MD simulations. However, caution is needed in the analysis, as the technique is very sensitive to the extension of the trajectory and to numerical errors in the calculations and can neglect important local distortions in favor of general but perhaps less relevant movements. Additional sources of error exist in the definition of a common reference system for the trajectories and on the elimination of translational and rotational degrees of freedom of the molecule. It is very important to keep in mind that essential motions are detected only if they occur in the MD trajectory, but slow motions might be difficult to detect in current ‘‘state-of-the-art’’ simulations (5–10 ns). Caution and common sense are then necessary for a reasonable use of the very powerful essential dynamics tool.
2.4. Free energy profile calculations In the study of proteins through MD, relevant information about a given process is obtained from the mean force potential (PMF) associated with the process (Leach, 2001). Examples of the potential applications of this tool are ligand migration through the protein matrix (Bidon-Chanal et al., 2006) or the conformational change of the side chain movement of a residue between two different stable conformations (Marti et al., 2006a). The PMF can be compared directly related to the experimental results, as it takes into account thermal and entropic effects. If the free energy barriers are of the same magnitude as the thermal fluctuations, it is feasible to obtain the free energy profiles associated with a given process directly from classical MD simulations. In these cases, an adequate sampling of the relevant configurations may be achieved in accessible simulation times, and the free energy profile can be obtained by computing the probability distribution along the selected reaction coordinate, P(x):
Computer Simulation of Globin Reactivity
bAðxÞ ln½pðxÞ;
483
ð24:4Þ
where b1 ¼ kBT is the Boltzmann constant times the temperature and A is the free energy. To have an appropriate sampling in 5- to 10-ns simulation times, the barriers should be 2 kcal/mol. In cases where barriers are suspected to be higher, biased sampling is required to obtain the PMF. Two different biased sampling methods are now presented: umbrella sampling and steered molecular dynamics. 2.4.1. Umbrella sampling This method (Torrie and Valleau, 1977) attempts to overcome the sampling problem by modifying the potential function so that the unfavorable states are sampled sufficiently. The potential function is modified by adding a weighting function that usually takes a harmonic form according to
E0 ðrÞ ¼ EðrÞ þ kðx xoÞ2 ;
ð24:5Þ
where E(r) is the potential energy of the protein for a given configuration, denoted by r, and x denotes a specific reaction coordinate. For configurations that are far from the equilibrium state, the weighting function assumes large values and so the simulation using the modified energy function E0 (r) is biased along the reaction coordinate. An umbrella sampling calculation involves a series of stages (called simulation windows), each characterized by a particular value of the reaction coordinate. The PMF is then obtained by superposing the results obtained for all the series of windows. To obtain the PMF in the unbiased simulation, the PMF obtained from the simulation is corrected by subtracting the contribution due to the weighting function. Although the method itself seems to be implemented easily in multiprocessor parallel computers, several technical details must be taken into account: (i) two consecutive windows must overlap in order to correctly superimpose them, (ii) the force constant for each stage has to be chosen carefully in order to make an efficient sampling of the potential energy surface, and (iii) the initial thermalization simulation must be performed for all the windows. 2.4.2. Multiple Steering Molecular Dynamics (MSMD) The multiple steering molecular dynamics approach, originally proposed by Jarzynski (1997) is based on the following relation between nonequilibrium dynamics and equilibrium properties,
484
Marcelo A. Marti et al.
exp½DAðxÞ=kB T ¼ hexp½W ðxÞ=kB T i;
ð24:6Þ
where W ðxÞ is the external work performed on the system as it evolves from the initial to the final state along the reaction coordinate x. In MSMD, the original potential is modified by adding a timedependent external potential, usually harmonic, to the potential energy that moves the system along the reaction coordinate by varying the potential well according to
E0 ðrÞ ¼ EðrÞ þ k½x ðxo þ vDtÞ2 ;
ð24:7Þ
where v is the pulling speed that moves the system along the reaction coordinate. The PMF is obtained by performing several MSMD runs, collecting the work done at each time step, and then averaging it properly, according to Jarzynski’s equation [Eq. (6)]. Usually, the pulling speed is chosen so that the system moves smoothly, but faster than in a true reversible simulation (Crespo et al., 2005a; Hummer and Szabo, 2001; Park and Schulten, 2004; Xiong et al., 2006).
2.5. Protein energy landscape exploring Although MD simulations have been widely used to study biomolecular systems, the modeling of long time dynamics still remains a challenge. In the last decade, there has been a significant effort toward the development of theoretical methods for protein structure prediction based on the use of rotamer libraries (Dwyer et al., 2004). Jacobson and colleagues (2002) have developed a program for protein modeling using specialized sampling algorithms for side chain prediction. The sampling algorithms include the use of highly detailed rotamer state libraries for side chain conformational searching, hierarchical screening methods based on steric overlap and approximate electrostatics to rapidly eliminate obviously incorrect conformations, and a multiscale minimization algorithm, one to two orders of magnitude more efficient than conventional approaches. Using these technological advances in protein structure prediction methods, a new approach to study the protein energy landscape exploration (PELE) associated with long time dynamics events has been developed (Borrelli et al., 2005). The heuristic algorithm for the PELE method involves consecutive iteration of three main moves: localized perturbation, side chain sampling, and minimization.
Computer Simulation of Globin Reactivity
485
2.5.1. Local perturbation The procedure begins with the generation of a local perturbation. When studying ligand diffusion in a protein, this first step involves a ligand perturbation. Initially, the ligand is treated as a rigid body. Hundreds of perturbations are generated within seconds and the one with the best energy is selected. 2.5.2. Side chain sampling The algorithm proceeds by placing all side chains local to the atoms perturbed in step 1 using the algorithms described earlier ( Jacobson et al., 2002). 2.5.3. Minimization The last step in every move involves minimization of a region including, at least, all residues local to the atoms involved in steps 1 and 2. These three steps compose a move that is accepted (defining a new minima) or rejected based on a Metropolis criterion for a given temperature. The collection of accepted steps forms a stochastic trajectory. The changes are propagated to the nonlocal environment by means of diffusion of the local perturbed area and by intercalated longer range (larger local area) steps. Many trajectories are typically run in parallel with a collective task. Processors ahead in the collective task will spawn the coordinates to those left behind. The task protocol allows for the simulation of reaction coordinates or to focus any search in a given trajectory.
2.6. Heme group parameters A critical point in MD simulations of heme proteins is the parameters associated with the heme group. Typically, all bonded and Lennard–Jones parameters can be obtained from the force field. However, it is crucial to select the atomic charges very carefully because they are essential to obtain accurate results, taking into account the different states of the heme group. Typically, partial charges for the heme moiety in the different coordination/ oxidation states are obtained according to the standard protocol, which consists in performing Hartree–Fock or density functional theory (DFT) calculations for the isolated heme group and subsequently deriving electrostatic potential-fitted charges (Wang et al., 2000). Other electronic structure methods can also be used to determine the electrostatic potential.
3. Quantum Mechanical-Molecular Mechanical Methods Several electronic structure computations can be used to study reactive processes in proteins. This chapter shows results obtained with a QM-MM scheme at the DFT level with the SIESTA code. DFT is, to our knowledge,
486
Marcelo A. Marti et al.
the method with the best cost-benefit ratio, as it provides quite accurate results at a reasonable computational cost (Soler et al., 2002). The SIESTA code has shown excellent performance for medium and large systems, including biomolecules, and heme models in particular (Capece et al., 2006; Crespo et al., 2003, 2005b; Fernandez et al., 2005; Marti et al., 2004, 2005, 2006a,b). Basis functions consist of localized (numerical) pseudoatomic orbitals, projected on a real space grid to compute the matrix elements. In the examples presented in this chapter, basis sets of double zeta plus polarization quality were employed for all atoms. Calculations were performed using the generalized gradient approximation to the exchangecorrelation energy proposed by Perdew and colleagues (1996). Such a combination of exchange-correlation functional, basis sets, and grid parameters has been widely validated for heme models (Capece et al., 2006; Crespo et al., 2005b; Fernandez et al., 2005; Marti et al., 2004, 2005, 2006a,b).
3.1. Selection of the QM subsystem Selection of the QM subsystem in simulations of NO reactivity with globins is quite intuitive. In order to take into account the chemical reactivity of the heme group, the iron porphyrinate and the bound ligands, e.g., the proximal histidine and the distal O2, NO, etc., must be selected as the quantum subsystem. In cases where other residues play a relevant role in the chemical reaction, they should also be included in the QM subsystem. The rest of the system is treated at the MM level, and the interface between QM and MM portions is treated using several formalisms, such as the scaled position link atom method (Eichinger et al., 1999), adapted to the SIESTA QM-MM code. This method completes the valence of the QM system with hydrogen atoms placed along the QM carbon atom–MM carbon atom bond.
3.2. Quantum mechanical-molecular mechanical optimizations The computational cost of QM-MM calculations limits the feasibility to explore the dynamics of the system. Instead, geometry optimizations of the whole system are less expensive and afford the energy of the potential energy minimum. To obtain an initial structure representative of the protein conformation in physiological conditions for QM-MM optimization, a conformational trapping protocol is performed. This is done by starting from an equilibrated snapshot of the protein obtained in a classical MD run and cooling the system down slowly to 0 K. As shown later, this method must be used starting from different snapshots to obtain representative structures of the protein. The cooling process is followed by QM-MM geometry optimizations using a conjugate gradient algorithm. Because we
Computer Simulation of Globin Reactivity
487
are only interested in the changes that take part in the active site, only residues located less than 10 A˚ apart from the heme reactive center are allowed to move freely in QM-MM runs. This protocol yields structures of the protein–ligand complex that can be compared with experimental data.
3.3. Binding energy calculations A relevant feature of a globin is its ligand-binding properties, which can be examined by the ligand-binding energy (DEL) determined from Eq. (24.8). Using Eq. (24.8), no entropy contribution is included in calculations. However, for comparative analysis, it can be omitted in a first approximation, and the QM-MM interaction energies are then expected to be useful.
DEL ¼ EprotL ðEprot þ EL Þ;
ð24:8Þ
where Eprot-L is the energy of the ligand–protein complex obtained from a QM-MM optimization, Eprot is the QM-MM energy of the free protein, and EL is the QM energy of the isolated ligand. It is worth mentioning that the DFT at the generalized gradient approximation level exhibits a bias in the description of the spin-state energetics of iron-porphyrins, in general favoring low-spin configurations. In particular, DFT energies for the free heme, whose electronic structure corresponds to a high-spin state, are somehow overestimated with respect to the ligandbound form, which presents a low-spin ground state (Deeth and Fey, 2004, Franzen, 2002). However, even though the estimates for DE may be sometimes above the experimental values, the predicted trends are in agreement with the experimental results.
3.4. Reaction pathway search As noted earlier, QM-MM calculations permit identifying the minimum energy point on the potential energy surface through geometry optimizations. However, much chemical interest lies in the free energy barrier between stable minima or for a given chemical reaction between reactants and products. The energy barrier is defined as the maximum energy along the minimum energy path (i.e., the reaction path) that connects the two minima. The conversion of one minimum energy (reactants) into another (products) may sometimes occur primarily along just one or two coordinates, denoted reaction coordinates. In such cases, an approximation to the reaction pathway, called restrained minimization, can be obtained by changing these coordinates gradually, allowing the system to relax at each stage while keeping the chosen coordinates fixed. The point of higher
488
Marcelo A. Marti et al.
energy on the path is an approximation to the transition state, and structures generated during the course of the calculation can be considered to represent a sequence of points on the interconversion pathway (Leach, 2001). To carry out restrained minimizations, an additional term is added to the potential energy, as noted in Eq. (24.9):
1 VR ¼ kðx x0 Þ2 ; 2
ð24:9Þ
in which k is an adjustable force constant, x is the value of the reaction coordinate, and x0 is the value of the reaction coordinate for a particular configuration. The path is constructed by (i) unrestrained minimization of reactant or product to generate an initial configuration for the reaction path, (ii) adding VR to the potential energy, varying x0 , and finally (iii) performing a series of energy minimizations at each step. The actual energy of each configuration along the reaction path is obtained by subtracting the restraint term from the total energy.
4. Illustrative Examples 4.1. Structural flexibility of globins as studied by MD simulations: Mycobacterium tuberculosis truncated hemoglobin N Flexibility of proteins can play an important role in modulating ligand diffusion and binding to the active site. This is the case of the truncated hemoglobin N (trHbN) from M. tuberculosis, which is a small globin that presents NO-dioxygenase activity in its oxygenated form (Ouellet et al., 2002). Thus, the protein contributes to the detoxification of NO through the conversion to nitrate anion [Eq. (24.10)]:
FeðIIÞO2 þ NO ! FeðIIIÞ þ NO 3:
ð24:10Þ
To carry out the reaction, the protein must ensure the presence of oxygen in the active site before entrance of NO. This is achieved by means of a dual-path ligand-induced regulation mechanism underlying diffusion of both O2 and NO through two distinct branches of the ligand migration tunnel. It has been proposed that this mechanism might control diatomic ligand migration to/from the heme as the rate-limiting step in NO conversion to nitrate (Bidon-Chanal et al., 2006). The short branch of ˚ ) is mainly defined by residues pertaining to helices the tunnel (around 8 A
Computer Simulation of Globin Reactivity
489
Figure 24.1 Long and short tunnel branches in M. tuberculosis trHbN. Helix names are indicated in capital letters.
G and H, and the large branch (around 20 A˚) is mainly formed by residues in helices B and E (Milani et al., 2001)(Fig. 24.1). Binding of O2, which accesses the heme cavity through the short branch, modulates the specific pattern of hydrogen-bonded contacts between residues TyrB10 and GlnE11, which in turn alters the global dynamics of the protein, favoring the relative displacement of helices B and E. The combined effect of both local conformational changes in the TyrB10–GlnE11 pair and global structural changes in the protein facilitates the conformational transition. This transition then leads to opening of the tunnel long branch, which consequently allows the diffusion of small ligands (NO) to the O2-bound heme. Inspection of the essential dynamics of the oxygenated and deoxygenated forms of trHbN shows clear differences that can be related to the distinct migration pathways of O2 and NO (Crespo et al., 2005b). As shown in Fig. 24.2, the major motions affect helixes C, G, and H in deoxy-trHbN. In contrast, the major motions in oxy-trHbN involve the relative displacement of helices B and E, which largely contribute to delineate the shape of the tunnel long branch. Finally, the difference in the main structural fluctuations of deoxy-trHbN and oxy-trHbN can be measured by means of a similarity index, which takes into account not only the nature of the essential movements, but also their contribution to the structural variance of the protein. Thus, for the 10 most relevant essential motions of the backbone skeleton in the core region of trHbN, which account for around 70% of the structural variance, the similarity index between deoxy and oxy forms of trHbN only amounts to 0.17. Overall, these findings reveal the different nature of the motions in the oxygenated and deoxygenated forms of the protein (Crespo et al., 2005b).
490
Marcelo A. Marti et al.
Figure 24.2 Representation of structural distortions in the peptide backbone of trHbN associated with the first essential motion in (left) oxygenated and (right) deoxygenated forms of the protein.
5. Ligand Migration Profiles from MSMD and PELE Simulations: Exploring Ligand Entry Pathways in M. tuberculosis trHbN From the plain MD simulations mentioned earlier, we observed two states for the long tunnel branch, where PheE15, which acts as a gate, adopts two conformational states corresponding to open and closed forms of the tunnel. Interestingly, the open state was only found in the oxygenated protein. Using MSMD, we computed the free energy profile for ligand migration along the two branches of the tunnel in the oxy and deoxy forms of trHbN (Bidon-Chanal et al., 2006). As an illustrative example, results for the open and closed states of the tunnel long branch in the oxy-trHbN are shown later (see Fig. 24.3).
Free energy (kcal/mol)
4 PheE15 closed state 2
0 −2 PheE15 open state −4
6
8
10
12 14 16 18 Fe--N(NO) distance
20
22
24
Figure 24.3 Free energy profiles for ligand diffusion along the tunnel long branch in open and closed states defined by the orientation of PheE15 in oxy-trHbN.
Computer Simulation of Globin Reactivity
491
In the closed state, access of the ligand to the heme cavity is impeded by PheE15, whose side chain protrudes into the channel, and the ligand remains preferably in the secondary docking sites. However, in the open state, where the benzene ring of PheE15 lies parallel to the axis of the tunnel, the ligand is directed toward the active site, as the highest barrier is around 1.5 kcal/mol. The opening of the tunnel long branch in oxy-trHbN allowed us to explain why the NO detoxification rate, determined by NO migration into the oxy protein, is around 100 times faster than O2 binding to the deoxy protein, as observed experimentally (Bidon-Chanal et al., 2006). As already mentioned in the computational methods section, an alternative strategy for the study of migration processes is to use the PELE algorithm. Figure 24.4 displays results for the carbon monoxide migration in deoxy trHbN. The CO migration was studied with 10 different runs. Each run produced a complete escape of the CO from the active site to the solvent and took about 120–192 CPU hours. Results show that the ligand first spends a large amount of cycles in the active site until Phe32 allows it to move into a cavity defined by Phe32 and Phe62. From this point, the CO
Figure 24.4 CO migration in M. tuberculosis trHbN as determined by the PELE algorithm.The heme group and relevant Phe residues are shown as sticks. Each CO position accepted by the algorithm along the run is shown as a dot. The arrows indicate the two migration paths.
492
Marcelo A. Marti et al.
ligand can escape through different pathways, shown in Fig. 24.4 with arrows and numbers 1 and 2. These exit pathways are in good agreement with the crystallographic and MSMD results mentioned previously.
5.1. Oxygen affinity of wild-type and mutant M. tuberculosis trHbN as compared to myoglobin Quantum mechanical-molecular mechanical calculations provide useful information about structural features. The comparison of structures for different proteins, or between wild-type and single residue mutants, is useful in gaining insight into the relationship between structure and function. A key issue in globin function is ligand affinity, which is modulated by the association (kon) and dissociation (koff) rate constants (Olson and Phillips, 1997; Scott et al., 2001). The theoretical investigation of the ligand association process is rather complex. However, ligand release is mainly controlled by the Fe-ligand bond breaking and is therefore intimately related to the calculated DE of ligand release (Marti et al., 2006b). As an example of how to study the structure and ligand affinity characteristics of several globins, Fig. 24.5 and Table 24.1 show the structural features and the binding energy of the oxy complexes of wild-type and mutant trHbN, as compared with those of Mb. We selected the TyrB10!Ala mutant in trHbN and the HisE7!Gly mutant in Mb, as wild-type residues are the main residues affecting the complex structure and oxygen affinity. Thus, Fig. 24.5 clearly shows that TyrB10 in trHbN and HisE7 in Mb are the main hydrogen bond donors to the oxygen ligand. Mutation of these two residues reduces the oxygen
Figure 24.5 Optimized structures of the active site of (right) wild-type trHbN and (left) myoglobin.
493
Computer Simulation of Globin Reactivity
Table 24.1 Geometrical parameters (angstroms and degrees), charge received by the O2 (DqO2) and donated by the proximal HisF8 Dqprox (in e), O2 binding energy DE (kcal/mol), and oxygen dissociation rate constant koff (s1)
d Fe-O d O-O dFe-NeHisF8 dFe-NeHisF8 (free protein) DqO2 Dqprox DE koff a b
trHb N
trHb N TyrB10!Ala
Wild-type Mb
HisE7!Gly Mb
1.84 1.31 2.06 2.13
1.86 1.31 2.09 2.10
1.84 1.30 2.18 2.194
1.76 1.29 2.18 2.190
–0.360 0.160 37.2 0.2a
–0.35 0.160 29.8 45a
–0.214 0.146 27.7 12b
–0.219 0.142 18.2 1600b
From Ouellet et al. (2002) From Scott et al. (2001)
affinity (see Table 24.1), which agrees with experimental data (Scott et al., 2001). However, in trHbN TyrB10!Ala the affinity is still similar to that of wild-type Mb because of the presence of GlnE11, which now interacts strongly with the bound ligand. The presence of two hydrogen bond donors in trHbN (TyrB10 and GlnE11) also explains the higher O2 affinity as compared to Mb. Another important finding is that hydrogen bond interactions and p-backbonding effects produce a net increase in the oxygen negative charge, DqO2. The p-backbonding effect is also different in both proteins, being more important in trHbN, probably because of a closer proximal histidine (HisF8) (Capece et al., 2006).
5.2. Chemical reactions between NO and globins: NO detoxification in M. tuberculosis trHbN The toxic effects of NO can be reduced or even eliminated by the development of resistance mechanisms, which consist of the oxidation of nitric oxide with heme-bound O2 to yield the harmless nitrate ion. NO scavenging functions have been observed in red blood cell hemoglobin, muscle Mb (Blomberg et al., 2004; Eich et al., 1996), neuroglobin within neuronal cells, and leghemoglobin, as well as in flavohemoglobins and truncated hemoglobins (Gardner et al., 2000, 2006). The study of the chemical reaction itself in the protein environment requires a hybrid QM-MM treatment. This section shows results obtained through this methodology for the hemecontrolled oxidation of NO in M. tuberculosis trHbN (Crespo et al., 2005a).
494
Marcelo A. Marti et al.
We start by performing QM-MM optimizations of the oxy-trHbN with NO located in the active site. The resulting structure shows that NO is close to O2 and interacts with Tyr33 and Gln58. Starting from this structure, we performed reaction pathway searches for the three reactions steps involved in the detoxification mechanism. In the first step, a peroxynitrite ion is formed from the attack of NO to the coordinated O2, as noted in Eq. (24.11). Subsequently, the metal center catalyzes the two-step isomerization of peroxynitrite to nitrate through an iron-oxo intermediate [Eqs. (24.12) and (24.13)]:
FeðIIÞ O2 þ NO ! FeðIIIÞ½ OONO
FeðIIIÞ½ OONO ! FeðIVÞ ¼ O þ NO2 FeðIVÞ ¼ O þ NO2 !
FeðIIIÞ½NO 3
ð24:11Þ ð24:12Þ ð24:13Þ
The selected reaction coordinates for describing reactions (11), (12), and (13) were the O2–NNO distance, the O1–O2 distance, and the O1–NNO2 distances, respectively (O1 is the Fe-bound O2 oxygen atom, and O2 is the distal O2 oxygen atom). For comparative purposes, calculations were performed for the isolated model system (vacuum), for the same model system solvated in aqueous solution (by using a cluster of 1061 water molecules), in the wild-type protein, and in the Tyr33!Phe mutant. Results indicate that the protein catalyzes the chemical reactions, leading to the formation of nitrate with no significant contributions of the protein environment, as the potential energy profiles were almost barrierless in vacuum, in the aqueous solution, and in the protein (Crespo et al., 2005b). This suggests that the rate-limiting process is ligand diffusion to the heme, as can be seen from the experimental bimolecular rate constant (745 mM1s1) (Ouellet et al., 2002) typical of an almost diffusioncontrolled process.
6. Conclusions The present results make apparent the contribution of computational classical, quantum mechanical, and hybrid quantum-classical techniques to the investigation of NO reactivity with globins. In particular, it has been shown how structural and conformational features and ligand migration paths can be investigated with classical MD simulations in combination with advanced sampling tools to yield information about free energy barriers and possible secondary docking sites. However, it was shown that QM-MM schemes are specially suited to investigate ligand binding and other chemical processes. By covering different time and space scales, these two
Computer Simulation of Globin Reactivity
495
frameworks, classical MD and QM-MM, complement each other in providing a complete picture of the protein function. The agreement with experimental data, when available, constitutes a stringent test to the reliability of these approaches. Computational simulation is a valuable tool even if it is often used to validate and contrast experimental observations because it offers a microscopic view, sometimes resolved in real time, that is very difficult to attain with any other method. With refinement of the methodology and available computers, classical and QM-MM simulations will be more and more at the center of the stage in this field, likely to become independent tools with the same hierarchy as any experimental technique.
ACKNOWLEDGMENTS This work was partially supported by the University of Buenos Aires, Agencia Nacional de Promocio´n Cientı´fica y Tecnolo´gica (Project PICT 25667), CONICET (PIP 02508), and the Spanish Ministerio de Educacio´n y Ciencia (Grant CTQ2005-08797-C02-01/BQU). The Barcelona Supercomputer Center is kindly acknowledged for providing access to the Marenostrum supercomputer.
REFERENCES Amadei, A., Linssen, A. B. M., and Berendsen, H. J. C. (1993). Essential dynamics of proteins. Proteins 17, 412–425. Berendsen, H. J. C., Postma, J. P. M., Van Gunsteren, W. F., DiNola, A., and Haak, J. R. (1984). Molecular dynamics with coupling to an external bath. J. Chem. Phys. 81, 3684–3690. Bidon-Chanal, A., Marti, M. A., Crespo, A., Milani, M., Orozco, M., Bolognesi, M., Luque, F. J., and Estrin, D. A. (2006). Ligand-induced dynamical regulation of NO conversion in Mycobacterium tuberculosis truncated hemoglobin-N. Proteins 64, 457–464. Blomberg, L. M., Blomberg, M. R., and Siegbahn, P. E. (2004). A theoretical study of myoglobin working as a nitric oxide scavenger. J. Biol. Inorg. Chem. 9, 923–935. Blundell, T. L., Sibanda, B. L., Sternberg, M. J. E., and Thornton, J. M. (1987). Knowledgebased prediction of protein structures and the design of novel molecules. Nature 326, 347–352. Borrelli, K. W., Vitalis, A., Alcantara, R., and Guallar, V. (2005). PELE: Protein Energy Landscape Exploration. A novel Monte Carlo based technique. J. Chem. Theory Comput. 1, 1304–1311. Brunori, M., Bourgeois, D., and Vallone, B. (2004). The structural dynamics of myoglobin. J. Struct. Biol. 147, 223–234. Capece, L., Marti, M. A., Crespo, A., Doctorovich, F., and Estrin, D. A. (2006). Heme protein oxygen affinity regulation exerted by proximal effects. J. Am. Chem. Soc. 128, 12455–12461. Chu, K., Vojtchovsky, J., McMahon, B. H., Sweet, R. M., Berendzen, J., and Schlichting, I. (2000). Structure of a ligand-binding intermediate in wild-type carbonmonoxy myoglobin. Nature 403, 921–923.
496
Marcelo A. Marti et al.
Copeland, D. M., Soares, A. S., West, A. H., and Richter-Addo, G. B. (2006). Crystal structures of the nitrite and nitric oxide complexes of horse heart myoglobin. J. Inorg. Biochem. 100, 1413–1425. Copeland, D. M., West, A. H., and Richter-Addo, G. B. (2003). Crystal structures of ferrous horse heart myoglobin complexed with nitric oxide and nitrosoethane. Proteins 53, 182–192. Crespo, A., Marti, M. A., Estrin, D. A., and Roitberg, A. E. (2005a). Multiple-steering QMMM calculation of the free energy profile in chorismate mutase. J. Am. Chem. Soc. 127, 6940–6941. Crespo, A., Marti, M. A., Kalko, S. G., Morreale, A., Orozco, M., Gelpi, J. L., Luque, F. J., and Estrin, D. A. (2005b). Theoretical study of the truncated hemoglobin HbN: Exploring the molecular basis of the NO detoxification mechanism. J. Am. Chem. Soc. 127, 4433–4444. Crespo, A., Scherlis, D. A., Marti, M. A., Ordejo´n, P., Roitberg, A. E., and Estrin, D. A. (2003). A DFT-based QM-MM approach designed for the treatment of large molecular systems: Application to chorismate mutase. J. Phys. Chem. B 107, 13728–13736. Deeth, R. J., and Fey, N. (2004). The performance of nonhybrid density functionals for calculating the structures and spin states of Fe(II) and Fe(III) complexes. J. Comput. Chem. 25, 1840–1848. Dwyer, M. A., Looger, L. L., and Hellinga, H. W. (2004). Computational design of a biologically active enzyme. Science 304, 1967–1971. Eich, R. F., Li, T., Lemon, D. D., Doherty, D. H., Curry, S. R., Aitken, J. F., Mathews, A. J., Johnson, K. A., Smith, R. D., Phillips, G. N., Jr., and Olson, J. S. (1996). Mechanism of NO-induced oxidation of myoglobin and hemoglobin. Biochemistry 35, 6976–6983. Eichinger, M., Tavan, P., Hutter, J., and Parrinello, M. (1999). A hybrid method for solutes in complex solvents: Density functional theory combined with empirical force fields. J. Chem. Phys. 110, 10452–10467. Elola, M. D., Estrin, D. A., and Laria, D. (1999). Hybrid quantum classical molecular dynamics simulation of the proton-transfer reaction of HO- with HBr in aqueous clusters. J. Phys. Chem. A 103, 5105–5112. Fernandez, M. L., Marti, M. A., Crespo, A., and Estrin, D. A. (2005). Proximal effects in the modulation of nitric oxide synthase reactivity: A QM-MM study. J. Biol. Inorg. Chem. 10, 595–604. Franzen, S. (2002). Spin-dependent mechanism for diatomic ligand binding to heme. Proc. Natl. Acad. Sci. USA 99, 16754–16759. Friesner, R. A., and Guallar, V. (2005). Ab initio quantum chemical and mixed quantum mechanics/molecular mechanics (QM/MM) methods for studying enzymatic catalysis. Annu. Rev. Phys. Chem. 56, 389–427. Gardner, A. M., Martin, L. A., Gardner, P. R., Dou, Y., and Olson, J. S. (2000). Steady-state and transient kinetics of Escherichia coli nitric-oxide dioxygenase (flavohemoglobin): The B10 tyrosine hydroxyl is essential for dioxygen binding and catalysis. J. Biol. Chem. 275, 12581–12589. Gardner, P. R., Gardner, A. M., Brashear, W. T., Suzuki, T., Hvitved, A. N., Setchell, K. D., and Olson, J. S. (2006). Hemoglobins dioxygenate nitric oxide with high fidelity. J. Inorg. Biochem. 100, 542–550. Guallar, V., and Friesner, R. A. (2004). Cytochrome P450CAM enzymatic catalysis cycle: A quantum mechanics/molecular mechanics study. J. Am. Chem. Soc. 126, 8501–8508. Hummer, G., and Szabo, A. (2001). Free energy reconstruction from nonequilibrium singlemolecule pulling experiments. Proc. Natl. Acad. Sci. USA 98, 3658–3661. Jacobson, M. P., Friesner, R. A., Xiang, Z., and Honig, B. (2002). On the role of the crystal environment in determining protein side-chain conformations. J. Mol. Biol. 320, 597–608.
Computer Simulation of Globin Reactivity
497
Jarzynski, C. (1997). Nonequilibrium equality for free energy differences. Phys. Rev. Lett. 78, 2690–2693. Karplus, M., and Petsko, G. A. (1990). Molecular dynamics simulations in biology. Nature 347, 631–639. Leach, A. R. (2001). ‘‘Molecular Modelling: Principles and Applications.’’ : Pearson Education EMA. MacKerell, A. D., Jr., Bashford, D., Bellott, M., Dunbrack, R. L., Jr., Evanseck, J. D., Field, M. J., Fischer, S., Gao, J., Guo, H., Ha, S., Joseph-McCarthy, D., Kuchnir, L., Kuczera, K., Lau, F. T. K., Mattos, C., Michnick, S., Ngo, T., Nguyen, D. T., Prodhom, B., Reiher, W. E., III, Roux, B., Schlenkrich, M., Smith, J. C., Stote, R., Straub, J., Watanabe, M., Wio´rkiewicz-Kuczera, J., Yin, D., and Karplus, M. (1998). All-atom empirical potential for molecular modeling and dynamics studies of proteins. J. Phys. Chem. B 102, 3586–3616. Marti, M. A., Bikiel, D. E., Crespo, A., Nardini, M., Bolognesi, M., and Estrin, D. A. (2006a). Two distinct heme distal site states define Cerebratulus lacteus mini-hemoglobin oxygen affinity. Proteins 62, 641–648. Marti, M. A., Capece, L., Crespo, A., Doctorovich, F., and Estrin, D. A. (2005). Nitric oxide interaction with cytochrome c0 and its relevance to guanylate cyclase: Why does the iron histidine bond break? J. Am. Chem. Soc. 127, 7721–7728. Marti, M. A., Crespo, A., Bari, S. E., Doctorovich, F. A., and Estrin, D. A. (2004). QM-MM study of nitrite reduction by nitrite reductase of Pseudomonas aeruginosa. J. Phys. Chem. B 108, 18073–18080. Marti, M. A., Crespo, A., Capece, L., Boechi, L., Bikiel, D. E., Scherlis, D. A., and Estrin, D. A. (2006b). Dioxygen affinity in heme proteins investigated by computer simulation. J. Inorg. Biochem. 100, 761–770. Milani, M., Pesce, A., Ouellet, Y., Ascenzi, P., Guertin, M., and Bolognesi, M. (2001). Mycobacterium tuberculosis hemoglobin N displays a protein tunnel suited for O2 diffusion to the heme. EMBO J. 20, 3902–3909. Olson, J. S., and Phillips, G. N., Jr. (1997). Myoglobin discriminates between O2, NO, and CO by electrostatic interactions with the bound ligand. J. Biol. Inorg. Chem. 2, 544–552. Onufriev, A., Bashford, D., and Case, D. A. (2004). Exploring protein native states and large-scale conformational changes with a modified generalized born model. Proteins 55, 383–394. Ouellet, H., Ouellet, Y., Richard, C., Labarre, M., Wittenberg, B., Wittenberg, J., and Guertin, M. (2002). Truncated hemoglobin HbN protects Mycobacterium bovis from nitric oxide. Proc. Natl. Acad. Sci. USA 99, 5902–5907. Park, S., and Schulten, K. (2004). Calculating potentials of mean force from steered molecular dynamics simulations. J. Chem. Phys. 120, 5946–5961. Perlman, D. A., Case, D. A., Caldwell, J. W., Ross, W. S., Cheatham, T. E., III, DeBolt, S., Ferguson, D., Seibel, G., and Kollman, P. (1995). AMBER, a package of computer programs for applying molecular mechanics, normal mode analysis, molecular dynamics and free energy calculations to simulate the structural and energetic properties of molecules. Comput. Phys. Commun. 91, 1–41. Perdew, J. P., Burke, K., and Ernzerhof, M. (1996). Generalized gradient approximation made simple. Phys. Rev. Lett. 77, 3865–3868. Scott, E. E., Gibson, Q. H., and Olson, J. S. (2001). Mapping the pathways for O2 entry into and exit from myoglobin. J. Biol. Chem. 276, 5177–5188. Soler, J. M., Artacho, E., Gale, J. D., Garcı´a, A., Junquera, J., Ordejo´n, P., and Sa´nchez-Portal, D. (2002). The SIESTA method for ab initio order-N materials simulation. J. Phys. Cond. Matt. 14, 2745–2779. Torrie, G. M., and Valleau, J. P. (1977). Nonphysical sampling distributions in Monte Carlo free-energy estimation: Umbrella sampling. J. Comput. Phys. 23, 187–199.
498
Marcelo A. Marti et al.
Wang, J., Cieplak, P., and Kollman, P. A. (2000). How well does a restrained electrostatic potential (RESP) model perform in calculating conformational energies of organic and biological molecules? J. Comput. Chem. 21, 1049–1074. Warshel, A., and Levitt, M. (1976). Theoretical studies of enzymatic reactions: Dielectric, electrostatic and steric stabilization of the carbonium ion in the reaction of lysozyme. J. Mol. Biol. 103, 227–249. Xiong, H., Crespo, A., Marti, M., Estrin, D., and Roitberg, A. E. (2006). Free energy calculations with non-equilibrium methods: Applications of the Jarzynski relationship. Theor. Chem. Acc. 116, 338–346.
C H A P T E R
T W E N T Y- F I V E
Microbial Responses to Nitric Oxide and Nitrosative Stress: Growth, ‘‘Omic,’’ and Physiological Methods Steven T. Pullan,1 Claire E. Monk,2 Lucy Lee,3 and Robert K. Poole* Contents 500
1. Introduction 1.1. Overview of bacterial exposure to nitric oxide (NO) and induction of microbial stress responses 1.2. Methods for analyzing global stress responses: Transcriptomics and proteomics 1.3. Batch versus continuous culture methods 2. Methods 2.1. Escherichia coli 2.2. Campylobacter jejuni 3. Nitric Oxide, NO-Releasing Agents, and Nitrosating Agents 3.1. Methods for assaying microbial sensitivity to NO 4. Illustrative Results from Applications of These Methods References
500 502 503 504 505 511 512 513 514 516
Abstract The study of bacterial responses to nitric oxide (NO), nitrosating agents, and other agents of nitrosative stress has a short history but has rapidly produced important insights into the interactions of these agents with model microbial systems as well as pathogenic species. Several methodological problems arise in attempting to define the global responses to these agents, whether in simply measuring growth or performing ‘‘omic’’ experiments in which the objective is to determine the genome-wide (transcriptomic) or proteome-wide responses. The first problem is the relatively long timescale over which the experiments are conducted—minutes, hours, or days in the case of slow-growing cultures. * 1 2
3
Department of Molecular Biology and Biotechnology, University of Sheffield, Sheffield, United Kingdom Current address: John Innes Centre, Norwich Research Park, Colney, Norwich NR4 7UH, UK Current address: MRC Protein Phosphorylation Unit, School of Life Sciences, University of Dundee, Dundee, DD1 5EH, UK Current address: School of Medicine and Biomedical Sciences, Beech Hill Road, Sheffield, S10 2RX, UK
Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37025-0
#
2008 Elsevier Inc. All rights reserved.
499
500
Steven T. Pullan et al.
The second problem is not unique to NO and its congeners but concerns the difficulties encountered when sensitive and comprehensive analytical techniques (such as transcriptomics) are applied to cultures whose growth and physiology are perturbed by an inhibitor. In essence, the problem is ‘‘seeing the wood for the trees.’’ This chapter reviews briefly the state of knowledge of NO responses and mechanisms in bacteria, particularly Escherichia coli and Campylobacter jejuni. Continuous culture has several advantages for investigating the consequences of NO exposure, and this approach is outlined with examples of recent results and conclusions. The major advantage of the chemostat is establishment of a reproducible quasi-steady state in growth, in which the growth rate can be controlled and maintained. Contrary to common belief, neither the concept nor the apparatus is difficult. Commercially available and homemade systems are described with practical advice. Establishing continuous cultures paves the way for other ‘‘omic’’ approaches, particularly proteomics and metabolomics, which are not covered here, as their application to the field of NO biology is in its infancy. A key to the literature describing methods suitable for assessing toxicity to microbes of NO and reactive nitrogen species is given.
1. Introduction 1.1. Overview of bacterial exposure to nitric oxide (NO) and induction of microbial stress responses There has been an explosion of interest over the past decade in the mechanisms deployed by living organisms to resist and detoxify NO and related agents that cause nitrosative stress. Much of the exciting information emerging (showcased in these two volumes) on hemoglobins and other proteins (NO reductases, flavorubredoxins, nitrite reductases, NO sensors) that react with nitric oxide is derived from microbial examples and an appreciation of the underpinning microbiology is warranted. While it may be argued that biochemical and biophysical methods applied in vitro to such proteins need not pay homage to the microbial provenance of these proteins, the same cannot be said of attempts to understand their functions in vivo. Indeed, the NO reactivity of some of the proteins now under intensive study was discovered using the methods of microbial genetics, genomics, and physiology. For example, prior to the sequencing of the Escherichia coli genome, the hmp gene was discovered by serendipity (Vasudevan et al., 1991) as a result of attempts to clone the gene for dihydropteridine reductase. The first clue to the real function of hmp came not from the extensive characterization in vitro of its oxygen-binding and electron transfer functions (Poole et al., 1994, 1997), but rather from a molecular genetic approach in which transcription of hmp was shown to be markedly stimulated by the addition to cultures of nitrite or NO solutions (Poole et al., 1996). The hmp gene was ‘‘rediscovered’’ by Gardner and colleagues during a
Physiological Methods for Studying Microbial NO Responses
501
search among N-methyl-N 0 -nitro-N-nitrosoguanidine-generated mutants for isolates able to grow anaerobically in the presence of NO concentrations (on agar plates supplemented with phenazine methosulfate) toxic to wildtype cells. The demonstration in one such NO-resistant mutant of high levels of NO consumption and the subsequent N-terminal sequencing of a protein associated with that activity demonstrated elegantly and convincingly (Gardner et al., 1998b) that the Hmp protein has robust NO dioxygenase or denitrosylase activity (Gardner et al., 1998b; Hausladen et al., 1998; Mills et al., 2001). The power of a cellular approach to understanding NO metabolism could be illustrated by numerous other examples. Equally, the dangers of drawing conclusions from poorly designed microbiological experiments could be demonstrated vividly. The purpose of this chapter is to offer advice and guidance on methods for investigating in vivo the roles of globins and other NO-reactive and oxygen-reactive proteins in microbes. Particular emphasis is placed on growth regimes suitable for the increasingly used ‘‘omic’’ methods, in which the aim is to identify those proteins and genes implicated in NO-related functions, understand their relationships, and untangle the global responses to NO and nitrosative stresses. Nitric oxide is a key component of the host immune response and is encountered by pathogenic bacteria during their lifestyles outside and within the host. In particular, phagocytic cells of the host produce the antimicrobial radical NO in micromolar quantities through the activity of inducible NO synthase (iNOS) (Fang, 2004). Enteric bacteria, such as E. coli and Salmonella enterica serovar Typhimurium, use two major mechanisms to detoxify NO (Poole, 2005): the flavohemoglobin Hmp and the flavorubredoxin NorV. The former catalyzes an O2-dependent denitrosylase (‘‘dioxygenase’’) reaction converting NO to nitrate ion using in addition an electron from NAD(P)H (Gardner et al., 1998b; Hausladen et al., 1998; Mills et al., 2001) or an anoxic reductive reaction forming NO (Hausladen et al., 1998; Kim et al., 1999). Flavorubredoxin NorV and its cognate reductase, NorW, however, catalyze only the reductive detoxification of NO under microaerobic or anaerobic conditions (Gardner et al., 2002). The synthesis of NorV and NorW is positively regulated at the transcriptional level by the NorR NO-sensing transcription factor (Hutchings et al., 2002). The regulation of Hmp synthesis is more complex. First, transcription of the hmp gene is repressed anaerobically by the oxygen-responsive regulator Fnr but, in the presence of NO, the DNA-binding activity of Fnr is diminished by the formation of dinitrosyl-iron complexes in the reaction with NO of the iron-sulfur cluster (Cruz-Ramos et al., 2002; Crack et al., 2007) so that NO derepresses hmp transcription. Second, MetR activates hmp transcription; nitrosation (see earlier discussion) of homocysteine (Hcy) forms S-nitrosoHcy and withdraws Hcy, a key intermediate from the biosynthetic pathway leading to methionine. In the absence
502
Steven T. Pullan et al.
of its cofactor Hcy, MetR binds to the hmp promoter and activates transcription (Membrillo-Herna´ndez et al., 1998). Third, hmp transcription is repressed by NsrR, an effect reversed by nitrite or NO (Bodenmiller and Spiro, 2006). Weaker transcriptional control may additionally be exerted by Fur and RamA (Herna´ndez-Urzua et al., 2007). The mechanisms employed by enterobacteria (see earlier discussion) are not universal, however. The food-borne bacterial pathogen Campylobacter jejuni, for example, contains neither flavohemoglobin nor flavorubredoxin, but a compact regulon responds to NO and nitrosative stress (Elvers et al., 2005). The most prominent NO-regulated protein, in the sense that a function in protection from these stresses has been unambiguously identified by mutation of the gene, is the hemoglobin Cgb (Elvers et al., 2004). A truncated globin (Ctb) is also encoded by a gene of that regulon (Elvers et al., 2005) and may have an auxiliary role in NO detoxification (Pickford et al., 2007; Wainwright et al., 2005, 2006).
1.2. Methods for analyzing global stress responses: Transcriptomics and proteomics A powerful approach to investigating bacterial responses to nitrosative stress is to measure the global changes in gene expression that occur upon exposure to such stress. This has been used to study the nitrosative stress responses of several bacteria, such as Bacillus subtilis (Moore et al., 2004), Pseudomonas aeruginosa (Firoved et al., 2004), and Mycobacterium tuberculosis (Ohno et al., 2003). The global transcriptional response of E. coli to nitrosative stress has been reported elsewhere (Flatley et al., 2005; Justino et al., 2005; Mukhopadhyay et al., 2004) with sometimes conflicting results. Indeed, in all these E. coli nitrosative stress microarray studies reported to date, only three transcriptional units were common, namely norVW, hmp, and nrdH, the last encoding a glutaredoxinlike protein. A probable explanation for these apparent discrepancies is in methodology, emphasizing the need for a careful assessment of experimental design and execution. Thus, various reactive nitrogen species (RNS) have been used as mediators of the stress, namely S-nitrosoglutathione (GSNO), acidified sodium nitrite, and NO gas, and the contributions of individual stress agents to the patterns of gene expression observed have not previously been deconvoluted. As described elsewhere (Aga and Hughes, 2007), nitric oxide and related RNS (NOþ, NO, NO) have unique chemistries (Hughes, 1999) reflecting the presence of nitrogen in oxidation states N(III), N(II), and N(I). Thus, NO is not per se a nitrosating agent, but nitrosation reactions are promoted by the presence of a metal ion or oxygen. Nitrosating agents (such as GSNO) and NO are often considered to be interchangeable, despite clear evidence that they elicit quite different effects in biological processes as diverse as caspase activation (Borutaite et al., 2000) or the respiratory oscillations of Saccharomyces cerevisiae in a chemostat (Murray et al., 1998).
Physiological Methods for Studying Microbial NO Responses
503
A further shortcoming of most microarray experiments is the difficulty of distinguishing between the nitrosative stress per se and the unintentional perturbation of culture behavior arising from the stress. Chemostat continuous culture offers major benefits in postgenomic global studies, such as proteomics, transcriptomics, and metabolomics (Hoskisson and Hobbs, 2005). The greater biological homogeneity of continuous cultures and control of all relevant growth conditions, such as oxygen levels, pH, and especially growth rate, avoids the masking effects of secondary stresses and growth rate changes, allowing a more precise delineation of the response to an individual stress. In the case of transcriptomics, it has been demonstrated that the reproducibility of analyses between different laboratories is superior when chemostat cultures are used, compared to identical analyses performed in batch culture (Piper et al., 2002). Methods for chemostat culture, which need not be complex, are presented here, following an explanation of the fundamental differences between the conventional batch culture and the chemostat.
1.3. Batch versus continuous culture methods The vast majority of experiments on growing microbial cultures are conducted on batch cultures, i.e., those in which a batch of growth medium is inoculated and the grown culture is subjected to analysis. This is a closed system in the sense that no further additions of medium are made: as the medium components are utilized during growth and as metabolic products form, the cell population responds by changing physiology. The most obvious manifestations, described in all elementary microbiology texts, are the transition of the inoculum through a lag phase followed by a period of growth, which may briefly be exponential (see later), and a ‘‘stationary phase’’ in which the population is prevented from a further increase in biomass accumulation by limiting nutrients (including oxygen) or other changes in the environment, such as adverse pH shifts. During exponential growth, the increase in population density (biomass, often measured as culture turbidity or mass of dried biomass samples) is constrained only by provision of nonlimiting nutrient levels. During this phase, sometimes termed ‘‘balanced growth’’ [in which intracellular metabolites are in a quasi-steady state (Provost and Bastin, 2004)], cultures are frequently harvested. In contrast, in continuous chemostat culture, fresh growth medium is pumped into a culture vessel with a constant volume. Essentially, the provision of medium controls growth, as the population density attained is limited by one or more nutrients in the feed. In defined or synthetic media, it is possible to manipulate growth yield by the concentration of a single nutrient, frequently the carbon source, nitrogen source, or other component: such a nutrient-limited continuous culture is termed a chemostat. Although less widely used now (certainly in the context of the field covered by these volumes), other forms of continuous culture have been
504
Steven T. Pullan et al.
devised (Pirt, 1985); for example, in the turbidostat, growth continues until a preset turbidity or culture density is reached when the medium flow is restarted. These variants of continuous culture are not covered here nor are the many ingenious procedures in which growth vessels are linked in series or which permit analysis of growth in response to nutrient gradients (Wimpenny et al., 1992). The chemostat was devised in the 1950s, based on formal descriptions of bacterial growth kinetics and substrate affinities (Ks). It is now clear that the concept of constants such as Ks is poorly founded. The ‘‘constants’’ are affected by bacterial adaptations and are a complex function of nutrient concentration (Ferenci, 1999). Nevertheless, despite such a shaking of some of the foundations of classical chemostat theory, there has been a resurgence of interest among microbiologists and molecular biologists in chemostats (Hoskisson and Hobbs, 2005). As discussed by Ferenci (2006, 2007), this is in part attributable to increased use of the technique for comparing gene expression under stable physiological conditions and also because of the increasing interest in experimental evolution in selective environments. The practical consequences of such adaptations occurring are covered briefly here. Finally, a greater understanding of the chemostat environment has been achieved since the pioneering, early concept-driven formulation of the chemostat, which is now being exploited to better explain gene expression studies such as those covered in this chapter. An important criticism of the use of continuous culture has been the potential selective pressure placed upon loss-of-function rpoS mutations, particularly at slow growth rates, leading to mutant bacteria overtaking the culture (Notley-McRobb et al., 2002). However, a closer examination of the literature reveals that this phenomenon is not observed under anaerobic conditions (King and Ferenci, 2005) or in the wild-type MG1655 strain used in many array studies (King et al., 2004). Importantly, while such adaptation or evolution might adversely affect the usefulness of chemostats in providing a more stable physiological state than comparable batch cultures, it is virtually certain that undesirable, uncontrolled, and perhaps unrecognized changes in growth rate will affect the outcome and interpretation of batch growths when growth-inhibitory reagents such as NO are employed.
2. Methods Our current procedures for defining the global consequences of bacterial exposure to NO and other RNS by transcriptomic analysis are given here. These protocols should be readily adaptable to individual circumstances.
Physiological Methods for Studying Microbial NO Responses
505
2.1. Escherichia coli The wild-type E. coli strain used in our laboratory is MG1655. This strain is well characterized, the subject of intensive mutational and physiological analyses (Blattner et al., 1997; Kolisnychenko et al., 2002), and appears to have the advantage of being resistant to mutational ‘‘sweeps’’ in rpoS (King et al., 2004). Mutant strains are generally MG1655 derivatives with Tn5 insertions in desired genes and are purchased from the E. coli Genome Project (University of Wisconsin, Madison) or constructed using standard methods (Datsenko and Wanner, 2000). 2.1.1. Chemostat design and use We have used two basic patterns of chemostat vessel in recent years. The first is a commercial apparatus, comprising a steel and glass culture vessel (New Brunswick Scientific Bioflow III Biofermentor). The second was constructed in the laboratory and is described later under ‘‘Campylobacter jejuni.’’ In the New Brunswick fermenter, and numerous other commercial systems, stirring is by impellers with vertical vanes (Pirt, 1985), which are mounted on a drive shaft operated from above by a motor. The vessel provides ports for nutrient feed, pH adjustment, inoculation, venting of waste gas, sampling, and other essential manipulations. A weir is fitted in the side wall and acts as an overflow to maintain constant volume. We have used a working volume of 1 liter and a dilution rate D of 0.2 h1. The dilution rate is defined as D ¼ F/V, where F is flow rate (ml h1) and V is the working volume (ml); it is thus the flow rate per unit volume and, at steady state—ignoring here the validity of this concept for evolving microbial cultures (Ferenci, 2007)—it will theoretically equal the specific growth rate of the culture, m (h1). Note that growth rate has a major influence on physiology: the concentrations of s factors, cAMP and ppGpp, as well as cell size, are all highly dependent on growth rate so that the pattern of gene expression is expected to be highly growth rate-dependent. Indeed, this concern was a major driving force in our decision to adopt chemostats for almost all gene expression studies, including our interests in NO-related metabolism, since 2002. A powerful feature of the chemostat is that the dilution rate controls growth rate even in the presence of compounds (such as NO) that lower the maximal growth rate in batch cultures. Thus, the experiment can be designed so that the selected dilution (growth) rate will always be lower (slower) than the growth rate after adding NO or other experimental stressor (Flatley et al., 2005). We have used the same principle to compare gene expression in cultures of E. coli grown without, and in the continual presence of, inhibitory concentrations of zinc (Lee et al., 2005). For aerobic growth, a useful guide to air flow rates is 1 liter air min1 per liter culture, which has proved adequate for maintaining dissolved oxygen tension at 40% air saturation in the 1-liter New Brunswick fermenter.
506
Steven T. Pullan et al.
In this apparatus, dissolved oxygen tension in the culture is measured using a Broadley James D140 OxyProbe electrode and maintained at a preset value, within narrow limits, by employing an automated adjustment of stirring rate; some other systems use mass flow valves to control rates of gas entry or gas composition at constant stirring rates. Studying nitrosative stress under different regimes of oxygen provision may prove to be a rewarding area for future work. Elegant methods have been developed for quantifying O2 bioavailability (Alexeeva et al., 2002) but have not yet been used in this field. In order to establish anaerobic growth, nitrogen is sparged through the chemostat medium prior to inoculation and throughout the experiment at a rate of 0.2 liter min1. No dissolved oxygen is detectable using the OxyProbe. Sodium fumarate is added to anaerobic medium at a final concentration of 50 mM to act as a terminal electron acceptor (Haddock et al., 1976). In other published work, we have used a different chemostat apparatus for purely pragmatic reasons during the course of studies on respiratory adaptation in E. coli (Partridge et al., 2006) and recently purchased new Infors vessels. The general principles outlined here are broadly applicable to a number of situations. 2.1.2. Choice and design of growth medium Where possible, we advocate the use of a defined medium in which the concentrations of all medium components are known and can be adjusted. Such a medium simplifies the identification and manipulation of the growth-limiting nutrient (as required in a chemostat) and can also help to ensure optimal bioavailability of all micronutrients (Hughes and Poole, 1991). Metal ions in particular can be tightly bound by the components (such as yeast extract) of complex media typified by Luria and TY broths (Hughes and Poole, 1991). Using the defined medium described later, we have obtained evidence from transcriptomic profiling, and fur-regulated gene expression in particular, that complex broths as used in a study of the GSNO regulon (Mukhopadhyay et al., 2004) appear to result in ironlimited growth. For aerobic growth, our medium contains glycerol (as the sole and limiting source of energy and carbon), 23 mM K2HPO4, 7.3 mM KH2PO4, 18.7 mM NH4Cl, 69 mM CaCl2, 15 mM K2SO4, 1 mM MgCl2, and the following trace elements in 134 mM EDTA: 31 mM FeCl36H2O, 6.15 mM ZnO, 570 nM CuCl22H2O, 340 nM CoNO36H2O, and 1.6 mM H3BO3 (all final concentrations) in distilled water. An important feature of this medium is the use of EDTA to chelate trace metal ions, which may otherwise be not biologically available (Pirt, 1985). In practice, we prepare a stock solution containing the EDTA and all trace element salts at 1000-fold the final concentrations shown earlier. The pH of the medium is maintained automatically at 7.0 by the addition of sterile 1 M NaOH or HCl and monitored using a Broadley James Fermprobe.
Physiological Methods for Studying Microbial NO Responses
507
To test for glycerol limitation in culture samples, supernatants are assayed using an enzymatic method (Garland and Randle, 1962). When other growth-limiting nutrients are used (see later), thought should be given to the availability of a convenient assay. In earlier work, we used the same glycerol concentration (54 mM) under both aerobic and anaerobic conditions (Flatley et al., 2005), which resulted in higher cell yields aerobically. More recently we have reduced, under aerobic conditions, the concentration of glycerol in the medium from 54 to 8 mM so that the aerobic growth yield, measured as OD600, was equal to that of the anaerobic culture (Pullan et al., 2007). Although limitation can also be achieved with phosphate, sulfate, nitrogen sources, and other nutrients, cultures with quite different physiologies will result. The concentration of the limiting nutrient will, of course, influence the biomass density of the chemostat. Note that high cell densities influence quorum-sensing aspects of bacterial physiology and, importantly, make it difficult to avoid a secondary limitation of growth by oxygen tension. Ferenci (2007) advocates adjusting cell densities to around 108 cells ml1 for simple chemostats in which aeration is accomplished by sparging and magnetic stirring. For our commercial vessels with multiple internal baffles, engineered impellors, and forced gassing, we have routinely achieved cell densities at least 10-fold higher than this without oxygen limitation. As in all recommendations here, preliminary testing would be needed to confirm the efficacy of particular combinations of strain, medium, and growth environment. 2.1.3. Monitoring and sampling chemostat cultures A potential criticism of long-term chemostat cultures (i.e., longer than the few hours generally needed to follow a batch culture from inoculation to the stationary phase of growth) is that the evolution and adaptation occurring in the culture in response to nutrient deprivation render the culture heterogeneous and unstable. A particular problem arises from the accumulation in bacterial cultures of rpoS mutations; to ensure that prolonged growth in the chemostat does not affect the transcriptome as a result, we routinely sequence the rpoS gene region in all chemostat samples prior to transcriptomic analysis. The fact that microbial cultures evolve on prolonged cultivation also raises questions about the optimal time for cell harvesting for physiological or gene expression studies. There is a widespread but incorrect view that chemostat cultures reach a true steady state. Indeed, the sweep of spontaneous mutations through bacterial cultures is now well established. Several studies have shown that the concentration of limiting nutrient continues to fall for many generations even after a ‘‘steady state’’ in microbial biomass density has been attained. Ferenci (2007) argues that at D ¼ 0.1 h1 (0.1 culture volumes per hour), the passage of 5 culture volumes equates
508
Steven T. Pullan et al.
to 50 h, which is sufficient to produce an almost complete replacement of the original population by mutants. Thus, to avoid gross mutational changes, it is probably preferable to harvest and analyze cells from chemostats soon after the initial rapid depletion of the limiting nutrient and the attainment of a constant cell density, which can be determined by experiment. It should be noted that, in experiments designed along the lines shown in Fig. 25.1 where a stressing agent (NO) is added for only minutes before harvest, we may take some comfort from the fact that a transcriptomic comparison, for example, will be of two culture samples separated by a very short interval
GSNO or NOCs
Reservoir
Reservoir
⫻ 5 min
“RNA protect” Control RNA
Experimental RNA
Array slide 1
Array slide 2
Cy3/Cy5
Dye swap
Bio discovery software
Figure 25.1 Experimental design for study of the bacterial transcriptome in response to GSNO or NOCs. A chemostat culture grown under reproducible and defined cultures is sampled for RNA quenching and preparation. To the culture and simultaneously to the medium feed (or a small sample of it) is added the stressing agent (GSNO, NOC, or other RNS). After exposure (typically 5 min), the culture is again sampled for RNA preparation. Each RNA sample is used to synthesize cDNA incorporating the fluorescent dye Cy5 or Cy3. A pair of samples (e.g., Cy5-labeled cDNA from the control culture and Cy3-labeled cDNA from the treated culture) is hybridized to Ocimum E. coli gene arrays. In a separate hybridization, the labeling of samples is dye swapped, i.e., switched (Cy3-labeled cDNA from the control culture and Cy5-labeled cDNA from the treated culture) to avoid the problem of unequal dye incorporation. The protocol used in published work on E. coli is shown (Flatley et al., 2005; Pullan et al., 2007) but fundamentally similar principles apply to our experiments with C. jejuni. Note that the procedure is illustrative of that used in this laboratory but that numerous alternative designs are found in the transcriptome literature.
Physiological Methods for Studying Microbial NO Responses
509
and that mutational changes that occurred prior to the establishment of a virtual ‘‘steady state’’ will be present in both control and test samples. Biomass density is determined most easily by culture turbidity or ‘‘apparent absorbance,’’ paying attention to the usual precautions (Fewson et al., 1984). Of special importance is assessing the linearity of response of the spectrophotometer or turbidimeter at high cell densities and quoting the details of the instrument and measurement conditions used. When spectrophotometers designed for clear solutions (i.e., virtually all commonly used instruments, apart from Klett meters) are used for measuring turbidity, alarming differences in apparent absorbance (Fewson et al., 1984) or optical densities (OD) are obtained as a result of the differential optical configurations in use (Koch, 1970). Although the collection of a sample for OD measurement from larger chemostat cultures poses no problems and may be achieved by the various siphon devices often fitted, the effluent culture from small-scale cultures might require alternative methods of biomass determination (cell counts, protein assay) or microcuvettes. 2.1.4. Microarray analysis We present here briefly our standard protocol for transcriptomic analysis of E. coli cultures exposed to NO and GSNO, but alternative methods are given elsewhere (Firoved et al., 2004; Flatley et al., 2005; Hyduke et al., 2007; Justino et al., 2005; Moore et al., 2004; Mukhopadhyay et al., 2004; Ohno et al., 2003). Cells are harvested directly into RNA Protect (Qiagen) and total RNA is purified using Qiagen’s RNeasy minikit (see Fig. 25.1). Equal quantities of RNA from control and GSNO-supplemented cultures are labeled using nucleotide analogues of dCTP containing either Cy3 or Cy5 fluorescent dyes. For each microarray slide, one sample is labeled with Cy3-dCTP, whereas the other incorporates Cy5-dCTP. Dye-swap experiments are performed for each pair to compensate for different efficiencies of incorporation of the labeled nucleotides. The slides used are E. coli K12 PAN arrays purchased from Ocimum Biosolutions (The Magdalen Centre, Oxford Science Park, Oxford, OX4 4GA, UK; previously marketed by MWG Biotech). These slides contain 4288 gene-specific oligonucleotide probes representing the complete E. coli (K12) genome. cDNA synthesis is carried out using 12 mg of RNA, primed with 9 mg pd(N)6 random hexamers (Amersham Biosciences). Reaction mixes (20 ml) containing 0.5 mM dATP, dTTP, and dGTP, 0.2 mM dCTP, and 0.11 mM Cy3/ Cy5-dCTP are incubated overnight at 37 with 200 units Superscript II RNase H reverse transcriptase (Invitrogen). cDNA is purified using a PCR purification kit (Qiagen), and equal volumes of cDNAs are combined and evaporated for approximately 45 min in a Thermo Savant SPD121P Speed Vac. cDNA is resuspended in salt-based hybridization buffer, heated to 95 for 3 min, and applied to the slides, which are hybridized for 16–24 h
510
Steven T. Pullan et al.
in a shaking water bath at 42 . Slides are washed in decreasing salt concentrations, dried by centrifugation, and scanned on an Affymetrix 428 scanner. Average signal intensity and local background correction are obtained using BioDiscovery Inc. software (Imagene, Version 4.0 and GeneSight, Version 3.5). Mean values from each channel are log2 transformed and normalized using the LOWESS method to remove intensity-dependent effects in the log2 (ratios) values. Cy3/Cy5 fluorescent ratios are calculated from normalized values. Biological experiments, i.e., chemostat growths, are carried out at least twice, and dye-swap analysis is performed on each experiment, providing a minimum of four technical repeats. Data from independent experiments are combined. Genes differentially regulated 2-fold and displaying a p value of 0.05 (using a t test) are defined here as being statistically, differentially transcribed (Fig. 25.2).
Log2 (anaerobic fold regulation)
6 5 4 3 2
−3
nor
−2
hmp
1
2 3 4 Log2 (aerobic fold regulation)
−2 −3
Figure 25.2 Changes in E. coli transcript levels in response to GSNO. Data are those published before (Flatley et al., 2005) but not illustrated in this form.Vertical (anaerobic data) and horizontal (aerobic data) axes show log2 fold-regulation values, relative to an untreated culture, after exposure of a chemostat culture to GSNO. Genes upregulated both aerobically and anaerobically are in the upper right quadrant and include genes encoding the major NO detoxification mechanisms, i.e., norVW encoding flavorubredoxin and hmp encoding flavohemoglobin. Other prominent upregulated genes are the met genes involved in methionine biosynthesis (triangles). It is proposed that GSNO nitrosates homocysteine, withdrawing this intermediate from the methionine biosynthetic pathway and, in thisdefined mediumwithoutadded amino acids, upregulates metgene expression.
511
Physiological Methods for Studying Microbial NO Responses
2.2. Campylobacter jejuni Many of the principles and methods detailed earlier have also been adopted for studies of the transcriptome of the food-borne pathogen C. jejuni in response to NO and other nitrosative stresses (C. Monk, B. Pearson, F. Mulholland, and R. K. Poole, in preparation). However, a convenient defined medium is not available for this organism and laboratory growth is generally conducted in rich broth media in which the limiting nutrient is not known. For these reasons, and also considering the pathogenic nature of the organism, we have adopted a much simpler culture system of small capacity (Lee et al., 2005) but modified for microaerophilic growth. The growth vessel is a Sigma Proculture Dynalift spinner flask, with 125-ml capacity, modified to include an overflow weir as an outlet for waste and spent media (Fig. 25.3). These modifications are carried out by a custom glass blower, taking care to set the height of the overflow exactly the same for the two vessels. Silicon tubing with a flow-back trap attached is connected to the overflow weir to allow waste to run into a flask. Two holes are drilled into the plastic screw lid of each culture vessel, into which 17-mm suba seals (rubber septa that fit snugly into tubes and can be punctured with a hypodermic needle; Fisher) are fitted. In one, a needle Gas vent
Inoculation port Stirrer bar housing Medium input
125 mL
Teflon stirrer
Overflow weir
Figure 25.3 A simple homemade chemostat apparatus.The growth vessel is a modified Sigma Proculture Dynalift spinner flask, 125-ml capacity, with an overflow weir as an outlet for waste and spent media. A tube connected to the overflow weir allows waste to run into a flask. Two holes are drilled into the plastic screw lid of each flask in which 17-mm suba seals (Fisher) are fitted. In one, a needle attached to an air filter is pushed through the seal, which acts as a vent to allow waste gas to be expelled. The other seal is used as an inoculation port. More in-depth details of its use for E. coli have been published (Lee et al., 2005), and the apparatus has been modified for growth of the microaerophilic bacterium C. jejuni (C. E. Monk and R. K. Poole, in preparation)
512
Steven T. Pullan et al.
attached to a Hepa-Vent filter (Whatman) is pushed through the seal, which acts as an air vent to allow waste gas to be expelled. The other seal is used as an inoculation port for the addition of 1% inoculum using a needle and 5-ml syringe. Culture medium is pumped from a reservoir (1-liter Duran vessel) using a small peristaltic pump [Minipuls 2 (Gilson) or Watson-Marlow]. Glass tubing is fed through a septum in the lid of the reservoir and attached to silicon tubing. The tubing is connected, via a flow-back trap, into the sidearm of the vessel through glass tubing inserted through a silicon bung. The septum also allows for a second piece of tubing to be connected to a Hepa-Vent filter for gas exchange. With a medium inflow rate of 12.5 ml h1, the dilution rate for a 125-ml continuous culture is 0.1 h1. A gas mix of 10% O2, 10% CO2, and 80% N2 is passed into the headspace of the culture vessel at a rate of 0.25 liter min1 to obtain a microaerobic atmosphere, and water jackets constructed from silicon tubing (attached to the outside of a modified plastic 250-ml beaker, which houses the spinner flask) maintain the temperature at 42 . The beakers with water jackets and the spinner flasks are mounted on KMO 2 ‘‘Basic’’ IKA-Werke stirrers to give effective transfer of O2 from the gaseous atmosphere to the culture, without sparging, by virtue of a stable vortex (Pirt, 1985). A fresh overflow sample is used to check the pH (which is consistently between 6.6 and 7) at the time of harvest. The small scale of the apparatus precludes pH control using an electrode and automated acid/alkali additions. To check for contamination, 20-ml samples are plated onto (a) nutrient agar plates incubated at 37 at atmospheric oxygen and (b) Mueller–Hinton agar plates incubated at 42 in a microaerophilic work station (Don Whitley Scientific).
3. Nitric Oxide, NO-Releasing Agents, and Nitrosating Agents Our chemostat experiments have been performed with both NO and S-nitrosoglutathione. Other agents could be used with similar protocols; examples of NO-releasing molecules and their use and properties are given elsewhere (Aga and Hughes, 2007). For GSNO, E. coli cells are grown as described earlier to steady state, defined as constant optical density (OD600), after approximately 5 culture volumes have passed through the vessel. At steady state, GSNO (synthesized according to the method of Hart (1985)) is added to the chemostat culture and to the nutrient feed at a final concentration of 200 mM (Flatley et al., 2005). This ensures that the culture receives GSNO as a bolus but that incoming medium from that point also contains the same GSNO concentration. To avoid adding GSNO to a large volume of medium in the reservoir, we switch the feed line from the usual reservoir
Physiological Methods for Studying Microbial NO Responses
513
to a much smaller reservoir just sufficient to maintain chemostat operation during the GSNO treatment. Samples are taken immediately prior to the addition of GSNO and after a period of 5 min exposure to GSNO for subsequent analysis using microarrays. The most convenient way to treat cultures with NO is to use the NO-releasing compounds 3-[2-hydroxy-1-(1-methylethyl)-2-nitrosohydrazino]-1-propanamine (NOC-5) and 3-(2-hydroxy-1-methyl-2-nitrosohydrazino)-N-methyl-1-propanamine (NOC-7), which have respective half-lives of NO release of 93 and 10 min, respectively (at pH 7.4, 22 ). These were purchased from Calbiochem. Stock solutions (100 mM ) of each are made up in 0.1 M NaOH (in which they are stable) and these are added simultaneously to give final concentrations of 10 mM each (Pullan et al., 2007). These NO–amine complexes spontaneously release two equivalents of NO under physiological conditions. As described before (Cruz-Ramos et al., 2002), use of both compounds ensures continuous release of NO over several tens of minutes. Release of NO after adding a solution containing both NOC compounds to defined medium may be measured using the apparatus described before (Mills et al., 2001). NOC-5 and NOC-7 are added simultaneously to chemostat cultures 5 min prior to sampling. Other NO-releasing compounds (Hyduke et al., 2007) may substitute.
3.1. Methods for assaying microbial sensitivity to NO Measuring sensitivity of growth to NO and related species is not a trivial task, because NO per se is stable for only minutes under physiological conditions. NO-releasing NOCs and NONOates can be added to growing cultures but their cost may make this prohibitive. However, NO gas solutions, prepared in the laboratory (Poole et al., 1996), may be added to growing cultures. GSNO, however, can be easily added as a powder or fresh solution and growth assessed by conventional means (turbidity, cell counts). In our experience, measuring growth in disk diffusion assays (where an inhibitory agent is applied to a sterile filter disk overlaid on a lawn of cells growing in agar) is problematic, not only for NO (because of its short halflife) but also for GSNO where millimolar or even molar stock solutions need to be used (although in small volumes). Examples of methods for the measurements of growth in the presence of NO or nitrosating agents are given in the following papers. The toxicity of Na nitroprusside and GSNO to E. coli has been measured using both viability assays and turbidity (Herna´ndez-Urzua et al., 2003; MembrilloHerna´ndez et al., 1999). Examples of disk diffusion assays applied to Rhodobacter capsulatus and an unusual method of determining sensitivity to NO gas are shown in Cross et al. (2000). In the latter method, bacteria suspended in soft agar (0.3% agar) are incubated in sealed tubes capped with
514
Steven T. Pullan et al.
suba seals and NO gas is injected into the headspace above the agar. The depth in the agar above which growth does not occur provides a semiquantitative assay for NO sensitivity. The Gardner group has also used growth on plates (Gardner et al., 1998b) or in liquid media (Gardner et al., 1998a, 2002) in atmospheres containing 960 ppm NO containing 10% air (balanced with nitrogen) and with constant replenishment of the NO. To avoid the problem of NO depletion in growing cultures, Demple’s group performed growth experiments in bottles fitted with two needles for attachment of various lengths of silastic membrane tubing. A mixture of 10% NO/90% Ar was then passed through the membrane tubing and entered the culture via diffusion through the tubing wall. Dose rates of NO delivery are proportional to the length of the tubing immersed in the culture (Nunoshiba et al., 1995). Thus, ingenious investigators will find solutions to the problems of NO lifetimes in bacterial cultures.
4. Illustrative Results from Applications of These Methods In earlier work (Flatley et al., 2005), glycerol-limited chemostat cultures of E. coli MG1655 in the chemically defined medium described in this chapter were used to provide bacteria in defined physiological states before applying nitrosative stress by the addition of S-nitrosoglutathione. Exposure to 200 mM GSNO for 5 min was sufficient to elicit an adaptive response as judged by the development of NO-insensitive respiration. Transcriptome profiling experiments, conducted as in Fig. 25.1 were used to investigate the transcriptional basis of the observed adaptation to the presence of GSNO (Flatley et al., 2005). In aerobic cultures, only 17 genes were significantly upregulated, including genes known to be involved in NO tolerance, particularly hmp (encoding the NO-consuming flavohemoglobin Hmp) and norV (encoding flavorubredoxin). Significantly, none of the upregulated genes was a member of the Fur regulon. Six genes involved in methionine biosynthesis or regulation were significantly upregulated (see Fig. 25.2); metN, metI, and metR were shown to be important for GSNO tolerance, as mutants in these genes exhibited GSNO growth sensitivity. Furthermore, exogenous methionine abrogated the toxicity of GSNO, supporting the hypothesis that GSNO nitrosates homocysteine, thereby withdrawing this intermediate from the methionine biosynthetic pathway. In an anaerobic chemostat, 10 genes showed significant upregulation, of which norV, hcp, metR, and metB were also upregulated aerobically. Thus, these data revealed new genes important for nitrosative stress tolerance and demonstrated that methionine biosynthesis is a casualty of nitrosative stress. It also revealed important changes in expression that had not been observed in previous batch
Physiological Methods for Studying Microbial NO Responses
515
culture-based studies in complex media (Mukhopadhyay et al., 2004), particularly those relating to the methionine biosynthetic pathway. The aims of more recent work were threefold. First, we set out to explore systematically the effects of NO per se delivered by wellcharacterized NO-releasing compounds in chemically defined media using the earlier study with GSNO as a reference point (Flatley et al., 2005). Of particular interest was the possibility that, under aerobic conditions, NO might exert nitrosating activities, and so, in the same experimental system used to study the GSNO stimulon, NO responses under both aerobic and anaerobic conditions were investigated. Previously, such comparisons had been made only in batch culture and separately under aerobic and anoxic conditions (Justino et al., 2005; Mukhopadhyay et al., 2004). Second, because Fnr has been shown previously to react with NO per se (Crack et al., 2007; Cruz-Ramos et al., 2002 ), evidence for global regulation of Fnrresponsive genes under anaerobic conditions was sought. Finally, additional components of the NorR regulon were sought under anaerobic conditions and we assessed the potential protective role of this regulon during internalization in murine macrophages. The resulting study (Pullan et al., 2007) was the first to examine the transcriptome of NO-exposed E. coli in a chemostat. Under identical conditions, the GSNO stimulon was compared with that of NO released from two NOC compounds simultaneously, and marked differences in the transcriptional responses to these distinct nitrosative stresses were demonstrated. Exposure to NO did not induce met genes, suggesting that, unlike GSNO, NO does not elicit homocysteine S-nitrosation and compensatory increases in methionine biosynthesis. Exogenous methionine, on entry into cells, afforded protection from GSNO- but not NO-mediated killing. Anaerobic exposure to NO led to upregulation of multiple Fnr-repressed genes and downregulation of Fnr-activated genes, including nrfA, which encodes cytochrome c nitrite reductase, providing strong evidence for NO inactivation of Fnr. Other global regulators apparently affected by NO were IscR, Fur, SoxR, NsrR, and NorR. Components of the NorR regulon were sought by microarray comparison of NO-exposed wild-type and norR mutant strains: only norVW, encoding the NO-detoxifying flavorubredoxin and its cognate reductase, were unambiguously identified. Mutation of norV or norR was without effect on E. coli survival in mouse macrophages. Thus, GSNO (a nitrosating agent) and NO exhibit distinct cellular effects: NO interacts more effectively with global regulators that mediate adaptive responses to nitrosative stress, but does not affect methionine requirements arising from homocysteine nitrosation. The study of microbial responses to NO and related nitrosative stresses is a very active and exciting area, but investigators must pay due attention to the design and execution of experiments if clear interpretations and physiological relevance are to be achieved.
516
Steven T. Pullan et al.
REFERENCES Alexeeva, S., Hellingwerf, K. J., and deMattos, M. J. T. (2002). Quantitative assessment of oxygen availability: Perceived aerobiosis and its effect on flux distribution in the respiratory chain of Escherichia coli. J. Bacteriol. 184, 1402–1406. Aga, R. G., and Hughes, M. N. (2007). The preparation and purification of NO gas and the use of NO releasers. The application of NO donors and other agents of nitrosative stress in biological systems. Meth. Enzymol. 436. Blattner, F. R., Plunkett, G., Bloch, C. A., Perna, N. T., Burland, V., Riley, M., ColladoVides, J., Glasner, J. D., Rode, C. K., Mayhew, G. F., Gregor, J., Davis, N. W., Kirkpatrick, H. A., Goeden, M. A., Rose, D. J., Mau, B., and Shao, Y. (1997). The complete genome sequence of Escherichia coli K-12. Science 277, 1453–1462. Bodenmiller, D. M., and Spiro, S. (2006). The yjeB (nsrR) gene of Escherichia coli encodes a nitric oxide-sensitive transcriptional regulator. J. Bacteriol. 188, 874–881. Borutaite, V., Morkuniene, R., and Brown, G. C. (2000). Nitric oxide donors, nitrosothiols and mitochondrial respiration inhibitors induce caspase activation by different mechanisms. FEBS. Lett. 467, 155–159. Crack, J. C., Le Brun, N. E., Thomson, A. J., Green, J., and Jervis, A. J. (2007). Reactions of nitric oxide and oxygen with the regulator of fumarate and nitrate reduction, a global transcriptional regulator, during anaerobic growth of Escherichia coli. Methods Enzymol. 437 [11] (this volume). Cross, R., Aish, J., Paston, S. J., Poole, R. K., and Moir, J. W. B. (2000). Cytochrome c0 from Rhodobacter capsulatus confers increased resistance to nitric oxide. J. Bacteriol. 182, 1442–1447. Cruz-Ramos, H., Crack, J., Wu, G., Hughes, M. N., Scott, C., Thomson, A. J., Green, J., and Poole, R. K. (2002). NO sensing by FNR: Regulation of the Escherichia coli NOdetoxifying flavohaemoglobin, Hmp. EMBO J. 21, 3235–3244. Datsenko, K. A., and Wanner, B. L. (2000). One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc. Natl. Acad. Sci. USA 97, 6640–6645. Elvers, K. T., Turner, S. M., Wainwright, L. M., Marsden, G., Hinds, J., Cole, J. A., Poole, R. K., Penn, C. W., and Park, S. F. (2005). NssR, a member of the Crp-Fnr superfamily from Campylobacter jejuni, regulates a nitrosative stress-responsive regulon that includes both a single-domain and a truncated haemoglobin. Mol. Microbiol. 57, 735–750. Elvers, K. T., Wu, G., Gilberthorpe, N. J., Poole, R. K., and Park, S. F. (2004). Role of an inducible single-domain hemoglobin in mediating resistance to nitric oxide and nitrosative stress in Campylobacter jejuni and Campylobacter coli. J. Bacteriol. 186, 5332–5341. Fang, F. C. (2004). Antimicrobial reactive oxygen and nitrogen species: Concepts and controversies. Nat. Rev. Microbiol. 2, 820–832. Ferenci, T. (1999). ‘‘Growth of bacterial cultures’’ 50 years on: Towards an uncertainty principle instead of constants in bacterial growth kinetics. Res. Microbiol. 150, 431–438. Ferenci, T. (2006). A cultural divide on the use of chemostats. Microbiology 152, 1247–1248. Ferenci, T. (2007). Bacterial physiology, regulation and mutational adaptation in a chemostat environment. In ‘‘Advances in Microbial Physiology’’ (R. K. Poole, ed.), vol 53. pp. 169–229. Elsevier, San Diego. Fewson, C. A., Poole, R. K., and Thurston, C. F. (1984). Spectrophotometry in microbiology: Symbols and terminology, scattered thoughts on opaque problems. Soc. Gen. Microbiol. Q. 11, 87–89. Firoved, A. M., Wood, S. R., Ornatowski, W., Deretic, V., and Timmins, G. S. (2004). Microarray analysis and functional characterization of the nitrosative stress response in nonmucoid and mucoid Pseudomonas aeruginosa. J. Bacteriol. 186, 4046–4050.
Physiological Methods for Studying Microbial NO Responses
517
Flatley, J., Barrett, J., Pullan, S. T., Hughes, M. N., Green, J., and Poole, R. K. (2005). Transcriptional responses of Escherichia coli to S-nitrosoglutathione under defined chemostat conditions reveal major changes in methionine biosynthesis. J. Biol. Chem. 280, 10065–10072. Gardner, A. M., Helmick, R. A., and Gardner, P. R. (2002). Flavorubredoxin, an inducible catalyst for nitric oxide reduction and detoxification in Escherichia coli. J. Biol. Chem. 277, 8172–8177. Gardner, P. R., Costantino, G., and Salzman, A. L. (1998a). Constitutive and adaptive detoxification of nitric oxide in Escherichia coli: Role of nitric-oxide dioxygenase in the protection of aconitase. J. Biol. Chem. 273, 26528–26533. Gardner, P. R., Gardner, A. M., Martin, L. A., and Salzman, A. L. (1998b). Nitric oxide dioxygenase: An enzymic function for flavohemoglobin. Proc. Natl. Acad. Sci. USA 95, 10378–10383. Garland, P. B., and Randle, P. J. (1962). A rapid enzymatic assay for glycerol. Nature 196, 987–988. Haddock, B. A., Downie, J. A., and Garland, P. B. (1976). Kinetic characterization of the membrane-bound cytochromes of Escherichia coli grown under a variety of conditions by using a stopped-flow dual-wavelength spectrophotometer. Biochem. J. 154, 285–294. Hart, T. W. (1985). Some observations concerning the S-nitroso and S-phenylsulphonyl derivatives of L-cysteine and glutathione. Tetrahedron Lett. 26, 2013–2016. Hausladen, A., Gow, A. J., and Stamler, J. S. (1998). Nitrosative stress: Metabolic pathway involving the flavohemoglobin. Proc. Natl. Acad. Sci. USA 95, 14100–14105. Herna´ndez-Urzua, E., Mills, C. E., White, G. P., Contreras-Zentella, M. L., Escamilla, E., Vasudevan, S. G., Membrillo-Herna´ndez, J., and Poole, R. K. (2003). Flavohemoglobin Hmp, but not its individual domains, confers protection from respiratory inhibition by nitric oxide in Escherichia coli. J. Biol. Chem. 278, 34975–34982. Herna´ndez-Urzua, E., Zamorano-Sanchez, D. S., Ponce-Coria, J., Morett, E., Grogan, S., Poole, R. K., and Membrillo-Herna´ndez, J. (2007). Multiple regulators of the flavohaemoglobin (hmp) gene of Salmonella enterica serovar Typhimurium include RamA, a transcriptional regulator conferring the multidrug resistance phenotype. Arch. Microbiol. 187, 67–77. Hoskisson, P. A., and Hobbs, G. (2005). Continuous culture: Making a comeback? Microbiology 151, 3153–3159. Hughes, M. N. (1999). Relationships between nitric oxide, nitroxyl ion, nitrosonium cation and peroxynitrite. Biochim. Biophys Acta Bioenerg. 1411, 263–272. Hughes, M. N., and Poole, R. K. (1991). Metal speciation and microbial growth: The hard (and soft) facts. J. Gen. Microbiol. 137, 725–734. Hutchings, M. I., Mandhana, N., and Spiro, S. (2002). The NorR protein of Escherichia coli activates expression of the flavorubredoxin gene norV in response to reactive nitrogen species. J. Bacteriol. 184, 4640–4643. Hyduke, D. R., Jarboe, L. R., Tran, L. M., Chou, K. J. Y., and Liao, J. C. (2007). Integrated network analysis identifies nitric oxide response networks and dihydroxyacid dehydratase as a crucial target in Escherichia coli. Proc. Natl. Acad. Sci. USA 104, 8484–8489. Justino, M. C., Vicente, J. B., Teixeira, M., and Saraiva, L. M. (2005). New genes implicated in the protection of anaerobically grown Escherichia coli against nitric oxide. J. Biol. Chem. 280, 2636–2643. Kim, S. O., Orii, Y., Lloyd, D., Hughes, M. N., and Poole, R. K. (1999). Anoxic function for the Escherichia coli flavohaemoglobin (Hmp): Reversible binding of nitric oxide and reduction to nitrous oxide. FEBS Lett. 445, 389–394. King, T., and Ferenci, T. (2005). Divergent roles of RpoS in Escherichia coli under aerobic and anaerobic conditions. FEMS. Microbiol. Lett. 244, 323–327.
518
Steven T. Pullan et al.
King, T., Ishihama, A., Kori, A., and Ferenci, T. (2004). A regulatory trade-off as a source of strain variation in the species Escherichia coli. J. Bacteriol. 186, 5614–5620. Koch, A. L. (1970). Turbidity measurements of bacterial cultures in some available commercial instruments. Anal. Biochem. 38, 252–259. Kolisnychenko, V., Plunkett, G., Herring, C. D., Feher, T., Posfai, J., Blattner, F. R., and Posfai, G. (2002). Engineering a reduced Escherichia coli genome. Genome Res. 12, 640–647. Lee, L. J., Barrett, J. A., and Poole, R. K. (2005). Genome-wide transcriptional response of chemostat-cultured Escherichia coli to zinc. J. Bacteriol. 187, 1124–1134. Membrillo-Herna´ndez, J., Coopamah, M. D., Anjum, M. F., Stevanin, T. M., Kelly, A., Hughes, M. N., and Poole, R. K. (1999). The flavohemoglobin of Escherichia coli confers resistance to a nitrosating agent, a ‘‘nitric oxide releaser,’’ and paraquat and is essential for transcriptional responses to oxidative stress. J. Biol. Chem. 274, 748–754. Membrillo-Herna´ndez, J., Coopamah, M. D., Channa, A., Hughes, M. N., and Poole, R. K. (1998). A novel mechanism for upregulation of the Escherichia coli K-12 hmp (flavohaemoglobin) gene by the ‘NO releaser’, S-nitrosoglutathione: Nitrosation of homocysteine and modulation of MetR binding to the glyA-hmp intergenic region. Mol. Microbiol. 29, 1101–1112. Mills, C. E., Sedelnikova, S., Sballe, B., Hughes, M. N., and Poole, R. K. (2001). Escherichia coli flavohaemoglobin (Hmp) with equistoichiometric FAD and haem contents has a low affinity for dioxygen in the absence or presence of nitric oxide. Biochem. J. 353, 207–213. Moore, C. M., Nakano, M. M., Wang, T., Ye, R. W., and Helmann, J. D. (2004). Response of Bacillus subtilis to nitric oxide and the nitrosating agent sodium nitroprusside. J. Bacteriol. 186, 4655–4664. Mukhopadhyay, P., Zheng, M., Bedzyk, L. A., LaRossa, R. A., and Storz, G. (2004). Prominent roles of the NorR and Fur regulators in the Escherichia coli transcriptional response to reactive nitrogen species. Proc. Natl. Acad. Sci. USA 101, 745–750. Murray, D. B., Engelen, F. A. A., Keulers, M., Kuriyama, H., and Lloyd, D. (1998). NOþ, but not NO center dot, inhibits respiratory oscillations in ethanol-grown chemostat cultures of Saccharomyces cerevisiae. FEBS Lett. 431, 297–299. Notley-McRobb, L., King, T., and Ferenci, T. (2002). rpoS mutations and loss of general stress resistance in Escherichia coli populations as a consequence of conflict between competing stress responses. J. Bacteriol. 184, 806–811. Nunoshiba, T., Derojaswalker, T., Tannenbaum, S. R., and Demple, B. (1995). Roles of nitric oxide in inducible resistance of Escherichia coli to activated murine macrophages. Infect. Immun. 63, 794–798. Ohno, H., Zhu, G. F., Mohan, V. P., Chu, D., Kohno, S., Jacobs, W. R., and Chan, J. (2003). The effects of reactive nitrogen intermediates on gene expression in Mycobacterium tuberculosis. Cell. Microbiol. 5, 637–648. Partridge, J. D., Scott, C., Tang, Y., Poole, R. K., and Green, J. (2006). Escherichia coli transcriptome dynamics during the transition from anaerobic to aerobic conditions. J. Biol. Chem. 281, 27806–27815. Pickford, J. L., Wainwright, L., Wu, G., and Poole, R. K. (2007). Expression and Purification of Cgb and Ctb, NO-Inducible Globins of the Foodborne Bacterial Pathogen Campylobacter jejuni. Meth. Enzymol. 436. Piper, M. D. W., Daran-Lapujade, P., Bro, C., Regenberg, B., Knudsen, S., Nielsen, J., and Pronk, J. T. (2002). Reproducibility of oligonucleotide microarray transcriptome analyses: An interlaboratory comparison using chemostat cultures of Saccharomyces cerevisiae. J. Biol. Chem. 277, 37001–37008. Pirt, S. J. (1985). ‘‘Principles of Microbe and Cell Cultivation.’’ Oxford: Blackwell Scientific Publications, Oxford.
Physiological Methods for Studying Microbial NO Responses
519
Poole, R. K. (2005). Nitric oxide and nitrosative stress tolerance in bacteria. Biochem. Soc. Trans. 33, 176–180. Poole, R. K., Anjum, M. F., Membrillo-Herna´ndez, J., Kim, S. O., Hughes, M. N., and Stewart, V. (1996). Nitric oxide, nitrite, and Fnr regulation of hmp (flavohemoglobin) gene expression in Escherichia coli K-12. J. Bacteriol. 178, 5487–5492. Poole, R. K., Ioannidis, N., and Orii, Y. (1994). Reactions of the Escherichia coli flavohaemoglobin (Hmp) with oxygen and reduced nicotinamide adenine dinucleotide: Evidence for oxygen switching of flavin oxidoreduction and a mechanism for oxygen sensing. Proc. R. Soc. Lond. Ser. B Biol. Sci. 255, 251–258. Poole, R. K., Rogers, N. J., Hughes, M. N., D’mello, R. A. M., and Orii, Y. (1997). Escherichia coli flavohaemoglobin (Hmp) reduces cytochrome c and Fe(III)-hydroxamate K by electron transfer from NADH via FAD: Sensitivity of oxidoreductase activity to haem-bound dioxygen. Microbiology 143, 1557–1565. Provost, A., and Bastin, G. (2004). Dynamic modelling under the balanced growth condition. J. Proc. Contr. 14, 717–728. Pullan, S. T., Gidley, M. D., Jones, R. A., Barrett, J., Stevanin, T. A., Read, R. C., Green, J., and Poole, R. K. (2007). Nitric oxide in chemostat-cultured Escherichia coli is sensed by Fnr and other global regulators: Unaltered methionine biosynthesis indicates lack of S nitrosation. J. Bacteriol. 189, 1845–1855. Vasudevan, S. G., Armarego, W. L. F., Shaw, D. C., Lilley, P. E., Dixon, N. E., and Poole, R. K. (1991). Isolation and nucleotide sequence of the hmp gene that encodes a haemoglobin-like protein in Escherichia coli K-12. Mol. Gen. Genet. 226, 49–58. Wainwright, L. M., Elvers, K. T., Park, S. F., and Poole, R. K. (2005). A truncated haemoglobin implicated in oxygen metabolism by the microaerophilic food-borne pathogen Campylobacter jejuni. Microbiology 151, 4079–4091. Wainwright, L. M., Wang, Y. H., Park, S. F., Yeh, S. R., and Poole, R. K. (2006). Purification and spectroscopic characterization of ctb, a group III truncated hemoglobin implicated in oxygen metabolism in the food-borne pathogen Campylobacter jejuni. Biochemistry 45, 6003–6011. Wimpenny, J. W. T., Earnshaw, R. G., Gest, H., Hayes, J. M., and Favinger, J. L. (1992). A novel directly coupled gradostat. J. Microbiol. Methods 16, 157–167.
This page intentionally left blank
C H A P T E R
T W E N T Y- S I X
Analysis of Nitric Oxide-Dependent Antimicrobial Actions in Macrophages and Mice Andre´s Vazquez-Torres,* Tania Stevanin,† Jessica Jones-Carson,* Margaret Castor,‡ Robert C. Read,† and Ferric C. Fang‡ Contents 1. NO -Dependent Antimicrobial Actions of Murine Macrophages 1.1. Equipment 1.2. Reagents 1.3. Protocol 2. NO -Dependent Antimicrobial Actions of Human Macrophages 2.1. Equipment 2.2. Reagents 2.3. Protocol 3. NO -dependent Antimicrobial Actions in Laboratory Mice 3.1. Equipment 3.2. Reagents 3.3. Protocol References
522 523 523 524 528 529 529 529 532 533 533 534 536
Abstract Nitric oxide (NO ) is a critical component of mammalian host defense that is produced in macrophages and other cells comprising the innate immune system. Isolated mammalian macrophages have been utilized to measure the kinetics of NO production and to demonstrate NO-related antimicrobial actions. Some microorganisms possess enzymes to detoxify nitrogen oxides, and mutant strains lacking these enzymes can be used to demonstrate the importance of these mechanisms for intracellular bacterialsurvival. This chapter describes techniques with which to analyze the antimicrobial actions of nitric oxide in murine and human macrophages and in laboratory mice. * { {
University of Colorado Health Sciences Center, Aurora, Colorado University of Sheffield, Sheffield, United Kingdom University of Washington School of Medicine, Seattle, Washington
Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37026-2
#
2008 Elsevier Inc. All rights reserved.
521
522
Andre´s Vazquez-Torres et al.
Nitric oxide (NO ) produced by host phagocytic cells plays a major role in innate immunity, in large part because of the ability of NO to inhibit or kill a broad range of microorganisms (Fang, 2004). By targeting protein thiols and metal centers (Stamler et al., 2001), NO can block essential microbial physiological processes, including respiration (Stevanin et al., 2000) and DNA replication (Schapiro et al., 2003). This chapter discusses methods for the examination of NO -dependent antimicrobial actions using murine or human macrophages and laboratory mice. The protocols described have specifically been employed to investigate the actions of host cell-derived NO to inhibit the facultative intracellular pathogen Salmonella enterica serovar Typhimurium. However, these methods should be generally applicable with minor modifications to investigate the role of NO in infections caused by other microbes as well.
1. NO -Dependent Antimicrobial Actions of Murine Macrophages The contribution of NO to the anti-Salmonella activity of interferon (IFN)g-activated macrophages is well accepted (McCollister et al., 2005; Vazquez-Torres et al., 2000, 2004; Webb et al., 2001). However, the importance of NO production by nonactivated macrophages is less certain (Bang et al., 2006; Chakravortty et al., 2002; Ekman et al., 1999; Saito et al., 1991; Shiloh et al., 1997, 1999; Vazquez-Torres et al., 2000, 2004). This largely reflects the greater NO production by activated macrophages (Ding et al., 1988; Vazquez-Torres et al., 2004). An important determinant of the outcome of such assays is the ability of the microbe to detoxify NO . For Salmonella, the major enzyme responsible for NO detoxification is the flavohemoglobin Hmp (Bang et al., 2006). Hmp increases the survival of Salmonella in unstimulated macrophages, suggesting that this flavohemoprotein can detoxify the quantity of NO synthesized by these cells. A comparison of the anti-Salmonella activity of unstimulated macrophages isolated from C3H/HeN and congenic Tlr4-deficient C3H/HeJ mice has revealed that the innate ability of macrophages to produce NO in response to Salmonella requires LPS/Tlr4 signaling (Vazquez-Torres et al., 2004). Increased killing of hmp mutant Salmonella is abrogated in iNOS-deficient macrophages, demonstrating that Hmp-mediated NO detoxification increases Salmonella intracellular survival. However, the Hmp flavohemoprotein is unable to counteract the fourfold higher NO fluxes1 produced 1
The production of NO was estimated by measuring accumulation of the stable NO oxidation product NO 2 . Total NO2 concentrations were measured spectrophotometrically at 550 nm after mixing sample supernatants with an equal volume of Griess reagent (0.5% sulfanilamide and 0.05% N-1-naphthylethylenediamide hydrochloride in 2.5% phosphoric acid). Standard curves were prepared with NaNO2.
Analysis of NO -Dependent Antimicrobial Actions
523
by IFNg-activated macrophages (Bang et al., 2006; McCollister and Vazquez-Torres, unpublished results). The use of NO -resistant and -susceptible bacteria in combination with cells producing various quantities of NO can elucidate both NO -dependent antimicrobial actions and mechanisms of microbial resistance. The following section describes a protocol used to examine the role of iNOS in the anti-Salmonella activity of macrophages and the role of the Hmp flavohemoglobin in antagonizing this activity.
1.1. Equipment Centrifuge equipped with holders for microtiter plates CO2 incubator Inverted and upright microscopes Hemocytometer Flow cytometer (optional)
1.2. Reagents Frozen bacterial stock cultures Microbial culture medium (liquid and solid) Antibiotics as indicated Phosphate-buffered saline (PBS) Dulbecco’s phosphate-buffered saline (DPBS) without Ca2þ or Mg2þ Hanks’ balance salt solution (HBSS) RPMI 1640 tissue culture medium Sodium m-periodate Fetal bovine serum Normal mouse serum HEPES L-Glutamine Sodium pyruvate Trypan blue Penicillin/streptomycin Dimethyl sulfoxide (DMSO) Recombinant murine IFNg Antibodies to macrophage surface markers Deoxycholic acid Collagenases I and IV EDTA Glucose dehydrogenase NADPH Glucose-6-phosphate Topro-3
524
Andre´s Vazquez-Torres et al.
5-ml polystyrene round-bottom tubes 15-ml polypropylene tubes Syringes, 10 ml Needles, 18 and 25 gauge Microcentrifuge tubes (0.65 and 1.5 ml) Flat-bottom 96-well microtiter plates Surgical scissors and forceps
1.3. Protocol 1.3.1. Isolation of murine macrophages 1. C57BL/6 mice and congenic iNOS/ controls2 (MacMicking et al., 1995) are inoculated intraperitoneally with 1 ml of 5 mM sodium m-periodate (Sigma-Aldrich, St. Louis, MO) prepared in PBS. The solution is drawn into the peritoneal cavity of hand-held mice using a 23-gauge needle mounted in a 3-ml syringe. 2. Inflammatory cells can be harvested 4 days after sodium m-periodate is injected. The peritoneum of mice euthanized in a CO2 chamber is exposed following sterile technique. Ten milliliters of RPMI 1640 medium supplemented with 10% heat-inactivated fetal bovine serum (BioWhittaker, Walkersville, MD), 15 mM HEPES, 2 mM L-glutamine, and 1 mM sodium pyruvate (Sigma-Aldrich, St. Louis, MO) (RPMIþ medium) is injected using an 18-gauge needle mounted in a 10-cc syringe. The medium is immediately drawn back with the same syringe. 3. Peritoneal exudate cells (PEC) are concentrated following a 5-min centrifugation at 200 g in a Centra CL3R centrifuge (Thermo Fisher Scientific Inc., Waltham, MA). The cell pellet is resuspended in 1 ml of RPMIþ medium. 4. Cells present in the suspension are counted in a hemocytometer after 1:50 dilution in 0.5% Trypan blue prepared in PBS (w/v). Cells are adjusted to 2 106 PCE/ml in RPMIþ medium containing 100 U ml1/ 100 mg ml1 of penicillin/streptomycin (Mediatech Inc. Herndon, 5 VA). For macrophage killing assays and NO 2 determination, 2 10 PCE/100-ml well are seeded into tissue culture grade, flat bottom 96-well plates (Becton Dickinson Labware, Franklin Lakes, NJ). Selected groups of macrophages are treated with 200 U/ml of IFNg (Life Technologies, St. Paul, MN) during the last 20 h of culture prior to Salmonella infection.
2
The contribution of NO to innate Tlr4-dependent host defenses of macrophages can be best shown in the presence of a functional Nramp1 (Slc11a1) cation transporter. The antimicrobial activity of peritoneal macrophages isolated from wild-type C3H/HeN can therefore be compared to controls isolated from congenic Tlr4-deficient C3H/HeJ mice.
Analysis of NO -Dependent Antimicrobial Actions
525
5. Macrophages contained in the PEC are selected after 48 h of culture at 37 in a 5% CO2 incubator by gently washing out nonadherent cells with 2 volumes of prewarmed RPMIþ medium.3 1.3.2. Phenotypic analysis of murine macrophages (optional) To characterize the population of macrophages isolated using this protocol, adherent macrophages are selected as described earlier in 24-well plates. Five hundred microliters of cells is added per well of 24-well plates after PEC are adjusted to 2 106 cells/ml in RPMIþ medium containing 100 U ml1/100 mg ml1 of penicillin/streptomycin as described previously. After adjusting for differences in volume, the procedure is as described for the selection of macrophages in 96-well microtiter plates.
6. Adherent cells are detached upon replacement of RPMIþ medium with 500 ml of prewarmed DPBS without Ca2þ or Mg2þ (Sigma-Aldrich) containing 0.2% collagenases I and IV (Sigma-Aldrich) and 2% heatinactivated fetal calf serum (FCS). Cells are incubated for 5 min in a humidified, 5% CO2 incubator at 37 . Culture plates are placed on ice for 15 min after the addition of 500 ml of 10 mM EDTA diluted in DPBS containing 2% heat-inactivated FCS. 7. Detached macrophages are recovered by pipetting, and the wells are further washed with 500 ml of DPBS supplemented with 2% FCS. EDTA is removed by centrifugation, and cells are resuspended in icecold HBSS with 5% FCS. 8. After nonspecific labeling is blocked using purified antibodies to FcRgII and FcRgIII, cells are stained with the macrophage phenotypic markers FITC-labeled F4/80 (FL1) (Caltag, Burlingame, CA) and PE-labeled CD11b Mac-1 (FL2) (eBioscience, San Diego, CA). Dead cells are excluded from the analysis after labeling with 1 nM Topro-3, a fluorescent vital stain (Invitrogen, Molecular Probes, Eugene, OR). 9. Cells are analyzed on a FACSCalibur flow cytometer. The population of adherent PEC is highly enriched for mononuclear phagocytes, as indicated by the fact that over 95% of the cells are F4/80þ CD11bþ macrophages (Fig. 26.1). 1.3.3. Bacterial cultures 10. Salmonella enterica serovar Typhimurium strain ATCC 14028s and isogenic bacterial strains of interest [e.g., hmp mutant strains IB3 and 3
Typically, about 105 macrophages are seeded per well, as quantified in a hemocytometer after adherent cells are released by collagenase treatment. Macrophage-like cell lines such as J774.1 or RAW264.7 cells can be very useful in the characterization of NO -related antimicrobial activity. As for primary macrophages, macrophage-like cells are seeded at densities of 105 cells/well of 96-well tissue culture plates and cultured in the presence or absence of 200 U/ml IFNg for 16–20 h prior to infection.
526
CD11b (FL4)
Andre´s Vazquez-Torres et al.
F4/80 (FL1)
Figure 26.1 Phenotypic analysis of C57BL/6 murine macrophages by flow cytometry. More than 95% of the adherent peritoneal exudate cells isolated by this protocol are F4/80þ CD11bþ.
AV0468 (Bang et al., 2006; McCollister et al., 2005)] are used for infection of macrophages. Bacterial cultures are grown from a –80 frozen stock prepared in DMSO. A crystal scraped from bacterial frozen stocks is grown in Luria-Bertani (LB) broth to stationary phase for 16 h at 37 in a shaker incubator set at 315 rpm.4 1.3.4. Macrophage killing assays 11. Macrophages are challenged at a multiplicity of infection (MOI) of 2 bacteria:1 macrophage with Salmonella that have been opsonized previously for 20 min in RPMIþ medium containing 10% normal mouse serum5 as described (McCollister et al., 2005). Briefly, 15 ml of overnight bacterial cultures diluted 1:5 (v/v) in PBS is added to 435 ml RPMIþ medium and 50 ml of normal mouse serum.6 Bacteria are then cultured in a 5% CO2 incubator at 37 for 20 min. 12. Macrophages are challenged with 50 ml opsonized bacteria that have been diluted with 3 volumes of prewarmed RPMIþ medium. Plates are 4
5
6
The majority of the iNOS-dependent anti-Salmonella activity of IFNg-activated macrophages is associated with repression of SPI2 transcription (8). The bacterial culture conditions described herein have been selected because in our experience they consistently induce expression of a functional SPI2 phenotype as indicated by the increased intracellular survival of wild-type Salmonella compared to an DspiC::FRT isogenic control. Various MOI may be used in macrophage killing assays. An MOI of 2:1 has been chosen in this case to diminish nonspecific killing associated with gentamicin included in the culture medium. At this low MOI, a functional NADPH phagocyte oxidase enzymatic complex accounts for all early Salmonella killing by control and IFNg-primed macrophages (B. D. McCollister and A. Vazquez-Torres, unpublished observations). Normal mouse serum (Sigma-Aldrich) is stored in 100-ml aliquots at –80 .
527
Analysis of NO -Dependent Antimicrobial Actions
centrifuged at 3200 g for 1 min, and cells are cultured at 37 in a 5% CO2 incubator. 13. Extracellular bacteria are removed 25 min after challenge by washing the monolayers with 2 volumes of prewarmed RPMIþ medium containing 6 mg/ml gentamicin (Sigma). To estimate the number of intracellular bacteria at time zero (t0), selected groups of macrophages are lysed 25 min after challenge with 0.5% deoxycholic acid prepared in PBS (w/v). The role of NO in the antimicrobial activity of macrophage populations expressing iNOS can be studied by adding inhibitors such as monomethyl L-arginine (250 mM) at time zero of infection. To demonstrate NO synthesis or its pharmacological inhibition, accumulation of its oxidation product NO 2 (nitrite) in the supernatant can be quantified with the Griess reagent after 8–20 h of infection. If quantitation of NO 3 is also desired, the Griess reaction can be used to determine total NOx after enzymatic reduction of NO3 to NO2 in pH 7 sodium phosphate buffer containing 1.6 U/ml Aspergillus nitrate reductase, 16 U/ml glucose dehydrogenase, 10 mM NADPH, and 10 mM glucose-6-phosphate. NO3 can be calculated as NOx – NO2 (Fig. 26.2). 14. Samples are serially diluted and spotted in LB agar plates. The rest of the Salmonella-infected macrophages are lysed at various time points after
20 C57BL/6 iNOS-/NOx (mM/h/105 mf)
15
10
5
0 0
5 Time (h)
10
Figure 26.2 NOx (NO2 þ NO3 ) production by IFNg-activated murine peritoneal macrophages over time. Macrophages were infected with Salmonella at t ¼ 0. Each data point represents NOx production over a 1-h sampling period. Reprinted with permission fromVazquez-Torres et al. (2000).
528
Andre´s Vazquez-Torres et al.
WT
hmp
10 % Survival (20 h)
Control
IFNg
1
0.1 C57/BL6
iNOS
C57/BL6
iNOS
Figure 26.3 Salmonella survival after 20 h in wild-type and iNOS-deficient C57BL/6 murine peritoneal macrophages. Reduced survival of the hmp mutant is observed in NO -producing macrophages. Interferon-g enhances NO -dependent antimicrobial activity but has a minimal effect on NO -independent antimicrobial activity. Reprinted with permission from Bang et al. (2006).
challenge (tn), and surviving bacteria are enumerated as described earlier by serial culture on LB agar plates. Results are expressed as % survival ¼ (tn/t0)100 (Fig. 26.3).
2. NO -Dependent Antimicrobial Actions of Human Macrophages Although NO production by murine macrophages can be demonstrated readily in response to inflammatory stimuli, the production of NO by human monocyte-derived macrophages is much more difficult to elicit under experimental conditions. Numerous publications have shown that macrophages from patients with infections or other inflammatory diseases exhibit expression of inducible NO synthase and production of NO (Weinberg, 1998), but the quantities of NO produced in vitro by human macrophages from healthy subjects are modest compared to those produced by murine cells. This increases considerably the challenge of demonstrating NO -dependent antimicrobial actions. Evidence that NO -related killing occurs in human macrophages includes the observation that NO detoxification mechanisms of both Neisseria meningitidis (Stevanin et al., 2005) and Salmonella (Stevanin et al., 2002) protect these organisms from killing by 10- to 12-day cultured human monocytes-derived macrophages, an effect that is eliminated by the presence of iNOS inhibitors.
Analysis of NO -Dependent Antimicrobial Actions
529
2.1. Equipment Category 2 laminar flow hood CO2 incubator Inverted and upright microscopes Hemocytometer Orbital shaker centrifuge Spectrophotometer Flow cytometer (optional) Sievers NO analyzer NOA 280i (optional)
2.2. Reagents Frozen bacterial stock cultures Microbial culture medium (liquid and solid) Antibiotics as indicated PBS RPMI 1640 tissue culture medium Fetal bovine serum L-Glutamine CPDA-anticoagulant Trypan blue DMSO L-NG-monomethyl arginine (L-NMMA) Saponin (1%) Antibodies to macrophage surface markers 5-ml polystyrene round-bottom tubes 15-ml polypropylene tubes Microcentrifuge tubes (0.65 and 1.5 ml) 24-well flat-bottom plates Sterile glass beads (1 mm diameter) Cell scraper (25 cm) (optional) Hamilton syringes, 50 and 100 ml (optional)
2.3. Protocol 2.3.1. Isolation of human monocyte-derived macrophages The GE-Healthcare Ficoll-Paque Plus offers a quick and reliable away of isolating primary human mononuclear cells from whole blood. 1. Peripheral blood is collected from healthy volunteers after informed consent is obtained. Universal precautions to protect the worker from blood-borne virus infection must be implemented. The worker must be vaccinated against hepatitis B.
530
Andre´s Vazquez-Torres et al.
2. The blood specimen is collected into a blood transfusion bag containing 63 ml of CPDA-anticoagulant. The amount of blood needed will vary from experiment to experiment, but 1 108 MDM can usually be derived from one unit (450 ml) of whole blood. Mononuclear cells are then isolated by density centrifugation by Ficoll–Paque and subsequently purified on Percoll density gradients to separate monocytes from lymphocytes (as per manufacturer’s instructions). Because of differential density, monocytes aggregate at the interface between the plasma and the Ficoll–Paque. 3. The layer containing monocytes is gently removed to a fresh tube and washed twice with PBS. After the second wash, the pellet is resuspended in RPMI 1640 medium containing 2 mM L-glutamine and 10% heat-inactivated fetal bovine serum. Monocytes are plated at a density of 1 106 cell/well in flat-bottom 24-well plates. Plates are then incubated at 37 , 95% air, 5% CO2. 4. After 24 h incubation, nonadherent cells are gently washed out, and adherent cells are cultured in RPMI 1640 with fresh medium and allowed to differentiate into macrophages over a 12-day period, with media being aspirated and replenished at 48-h intervals. 2.3.2. Bacterial cultures 5. Human MDM have been inoculated with a wide variety of bacteria, including category 2 and category 3 (biosafety level 3) organisms. An important consideration in planning experiments is the MOI, which will vary with the organism under examination and the nature of the experiment. The following description refers to our experience with S. enterica serovar Typhimurium and its isogenic hmp derivative. Strain ATCC 14028s and an isogenic strain carrying an antibody resistance cassette inserted in the hmp gene (14028 hmp or MCS2A) (Crawford and Goldberg, 1998) are used for infection of macrophages. Bacterial cultures are grown from a –80 frozen stock prepared in DMSO onto nutrient agar (NA) plates and incubated overnight at 37 . Transfer three to four colonies from the NA plate to 10 ml LB broth in a shaker set at 200 rpm at 37 until it reaches an OD600 of 0.025 nm (midexponential phase). This was identified in viability studies as the optimum OD600 to obtain an MOI of approximately 10 (10 bacteria/MDM) in a 250-ml suspension. 6. Bacteria are harvested and washed three times in PBS. The pellet should be resuspended in RPMI 1640 in the presence of two or three 1-mm glass beads to break up clumps of bacteria. 2.3.3. Macrophage killing assays 7. MDM are challenged with 250 ml/well suspensions of nonopsonized bacteria. There is no need to centrifuge plates at this stage. At the
Analysis of NO -Dependent Antimicrobial Actions
8.
9.
10.
11.
531
recommended MOI, nonopsonized Salmonella binds at approximately four bacterial cells/MDM (Stevanin et al., 2002). Plates are incubated at 4 for 60 min to allow binding of bacteria to MDM. (This is an optional method that can be used to prevent phagocytosis by cells if there is a possibility that the strains being tested are differentially internalized. MDM will only internalize bacteria at a temperature >17 .) Plates should be transferred to an incubator at 37 and incubated for a total of 4 h, taking samples every 30 min for determination of viability. After the first 30 min at 37 , the medium is removed and all wells are washed twice with PBS. To some wells, 250 ml of saponin (1%) will be added to disrupt the MDM, which will subsequently be incubated for 12 min. This control is important when comparing small differences in MDM killing between different strains and allows one to confirm that the numbers of bacteria associated with MDM have remained similar for all strains used in the experiment. This can be considered as time zero relative to gentamicin treatment. To the remaining wells, a minimum of 1 ml fresh RPMI 1640 containing gentamicin is added to kill any bacteria that have not been internalized by MDM. It is important to add enough antibiotic-containing medium to completely cover the wells, as residual viable extracellular bacteria may confound the results. The minimum bactericidal concentration for gentamicin must be established for each test strain or when varying experimental conditions such as inoculum size. The plates should be returned to the incubator for an additional 30 min. To confirm that noninternalized bacteria have been killed, the supernatant is collected into a centrifuge tube prior to the addition of saponin to the wells. The wells are to be rinsed twice with PBS to remove the antibiotic before plating samples for viable counts. The remaining wells are rinsed twice with PBS before replacement with fresh medium. After 12 min with saponin, the supernatant is collected into a centrifuge tube. The wells are washed with 500 ml PBS by pipetting up and down a few times to ensure that as many bacteria as possible are collected. Bacteria are harvested with resuspension of the pellet in 150 ml PBS. Fifty to 100 ml of 1 100, 1 101, and 1 102 dilutions of the bacterial suspension is spread onto agar plates for enumeration of viable counts.
Generally speaking, it is difficult to study iNOS activation in human macrophages in vitro, but it is nevertheless possible to show the importance of NO for the antimicrobial activity of human macrophages by inhibiting iNOS using L-NMMA. 12. One millimolar of L-NMMA may be added to the MDM 48 h prior to challenging with bacteria, with subsequent protocol steps as described earlier. 13. Supernatants from MDM that have or have not been treated with L-NMMA are collected to measure nitrite accumulation in MDM
532
Andre´s Vazquez-Torres et al.
supernatants and to confirm a decrease in NO production by L-NMMA-treated MDM. The Griess reagent used to measure NO 2 generation by murine macrophages is insufficiently sensitive to detect variations in NO production by human macrophages. A more sensitive chemiluminescence-based method should therefore be employed (see Chapter 7 in volume 436). To study MDM killing of Salmonella for longer than 4 h, the medium should be replaced with RPMI containing 20 mg/ml gentamicin to suppress extracellular bacterial growth.
3. NO -dependent Antimicrobial Actions in Laboratory Mice As for studies with murine macrophages, the basic approach to examining NO -dependent antimicrobial actions in laboratory mice involves a comparison between wild-type and isogenic iNOS knockout mice or between mice with and without administration of an NO synthase inhibitor. The availability of iNOS/ mice has greatly facilitated analysis of the role of NO in infection and other conditions (MacMicking et al., 1995; Mastroeni et al., 2000). However, studies using iNOS-/- mice are generally performed in a C57BL/6 background. C57BL/6 mice, like a number of inbred mouse strains, carry a mutation at the Slc11a1 (Nramp1 G169D) locus that enhances susceptibility to certain pathogens, including Mycobacterium bovis BCG (bacillus Calmette-Guerin), S. enterica serovar Typhimurium, and Leishmania donovani. The Slc11a1 locus encodes a cation transporter that appears to limit concentrations of manganese and iron within the phagosome ( Jabado et al., 2000; Zaharik et al., 2004). The mutant Slc11a1 locus of C57BL/6 mice may mask the ability to detect NO dependent antimicrobial actions in vivo (Bang et al., 2006). Thus, it may also be informative to perform infections of inbred mouse strains carrying a functional Slc11a1 locus (e.g., C3H/HeN, 129Sv), in which case an NO synthase inhibitor must be administered to abrogate NO production. A nonselective NO synthase inhibitor cannot be used for this purpose, as the inhibition of eNOS can have deleterious effects on tissue perfusion and cardiac output (Cobb et al., 1995). Because of its relative selectivity for iNOS and low cost, aminoguanidine has been employed in a number of studies. However, adverse effects of aminoguanidine during infection of iNOS/ mice (Zhou et al., 2002) indicate that nonspecific actions of aminoguanidine may complicate the interpretation of experimental results. Therefore, the use of more specific and costly iNOS-selective inhibitors (e.g., N-imino-ethyl L-lysine or L-NIL) is preferred (Stenger et al., 1996).
Analysis of NO -Dependent Antimicrobial Actions
533
Several complementary end points may be utilized to determine the effects of host NO production during infection. Historically, virulence was measured by the determination of mortality following different inocula and calculating the 50% lethal dose (LD50). This is no longer regarded as ethically acceptable under most circumstances, but one may still monitor infected animals and euthanize moribund animals humanely, using the number of days until euthanasia as a surrogate indicator of mortality. Alternatively, animals may be sacrificed after a designated time interval, and the organism burden in specific organs may be determined by dilution and plating (Mastroeni et al., 2000). A variation on this method involves the coinoculation of equal quantities of wild-type and mutant bacteria to determine whether the mutant is at a competitive disadvantage in the host. This can be useful in evaluating whether a specific genetic locus is involved in resistance to host-derived NO . The competitive disadvantage of a strain carrying a mutation in the locus of interest could be measured in the presence or absence of host NO production. The following protocol was developed for S. enterica serovar Typhimurium but may be adapted for use with other organisms.
3.1. Equipment Spectrophotometer Tissue homogenizer with disposable probes
3.2. Reagents Frozen bacterial stock culture Microbial culture medium (liquid and solid) Antibiotics as indicated PBS 70% ethanol Sterile wooden stick applicators Sterile toothpicks Cuvettes for spectrophotometer 5-ml polystyrene round-bottom tubes 15-ml polypropylene tubes Syringes Needles, 18 and 25 gauge Microcentrifuge tubes (0.65 and 1.5 ml) 96-well microtiter plates Ice bucket Scissors Forceps Dissecting board
534
Andre´s Vazquez-Torres et al.
3.3. Protocol 1. The relationship between OD600 (optical density at 600 nm) and colony-forming units (cfu) can be determined by growing an overnight culture, preparing 1:10 serial dilutions, measuring the OD600 on a spectrophotometer, and plating dilutions onto solid medium. 2. Bacteria are inoculated into culture medium, with antibiotic selection as appropriate, the night prior to infecting the mice (or longer as required for slowly growing organisms). 3. The OD600 of a 1:10 dilution of cultures is measured on a spectrophotometer. Based on the calculated colony-forming units per milliliter, the culture can be diluted into PBS so that the precise desired inoculum will be contained in 500 ml of the diluted culture. Aliquots of the diluted culture are plated onto solid medium for quantitation of the actual inoculum. 4. The inoculum is administered intraperitoneally to each mouse using a syringe and 25-gauge needle. Typically, at least 5–10 mice are included in each experimental group, with a minimum of two experimental replicates. A control group inoculated only with PBS should be included for each mouse strain to be tested. Gloves should be worn when handling mice. For intraperitoneal inoculation, mice can be restrained by grasping the scruff of the neck between the thumb and fingers to control the head; the tail and right hind leg may then be restrained using the fifth finger of the same hand. With the ventral side of the mouse facing the experimenter, intraperitoneal injection should be made into the right lower quadrant of the abdomen, lifting the needle against the abdominal wall after penetration of the musculature to minimize the risk of visceral perforation. Alternative routes of infection that might be appropriate for selected organisms include oral, intragastric, intravenous, subcutaneous, intradermal, or intramuscular. 5. Mice must be monitored twice daily for the duration of the experiment (typically 7 to 28 days, depending on the virulence of the bacterium and intrinsic resistance of the mouse strain). Moribund animals (i.e., animals not eating or drinking, lacking spontaneous movement, or exhibiting labored respiration) or animals that have lost 10% or more of their body weight should be euthanized humanely. A Kaplan–Meier plot of survival for each experimental group provides a useful representation of the contribution of NO to host resistance (Fig. 26.4). 3.3.1. Quantitation of bacterial burden 6. For enumeration of bacteria in the liver and spleen, mice must be euthanized by an approved method, such as a CO2 chamber (AVMA Panel on Euthanasia, 2001). Mice should be removed from the chamber and secured to a dissecting board. The ventral surface should be
535
Analysis of NO -Dependent Antimicrobial Actions
WT
% Survival
100
hmp
80
WT L-NIL
60
hmp L-NIL
40 20 0
0
5
10
15
Days
Figure 26.4 Survival of C3H/HeN mice after intraperitoneal challenge with wildtype or hmp mutant Salmonella. Addition of the iNOS inhibitor L-NIL to the drinking water restores virulence to the hmp mutant strain, suggesting that the Hmp flavohemoglobin promotes virulence by detoxifying host-derived NO. Reprinted with permission from Bang et al. (2006).
disinfected with 70% ethanol prior to cutting the skin along the midline to expose abdominal viscera. The spleen is removed by retracting the organ with forceps and cutting the associated vessels and membrane with scissors. Subsequently the esophagus can be severed to allow retraction of the stomach and intestines, thereby facilitating removal of the liver. As with the spleen, the liver is removed by cutting associated membranes with the scissors. Organs are placed in tubes containing 1 ml PBS on ice prior to homogenization with a disposable probe. Homogenates are serially diluted in a microtiter plate, followed by the plating of samples for quantitation of cfu. 3.3.2. Competition assays 7. For competition assays, mixed infections are performed using two strains distinguishable by antibiotic markers. Bacteria should be recovered from the initial inoculum and from the liver and spleen as in step 6. One hundred colonies from the initial inoculum and from each organ are patched onto nutrient agar plates containing antibiotics as indicated to distinguish the infecting strains. The competitive index (CI) can be calculated as CI ¼ strain #1 ðinputÞ=strain #2 ðinputÞ strain #1 ðorganÞ=strain #2 ðorganÞ If strain #1 is wild type and strain #2 is a mutant strain, then a CI < 1 would indicate that the mutant strain has attenuated virulence. Restoration of a CI 1 by the abrogation of NO production would be consistent with a role of the mutated locus in providing resistance to NO .
536
Andre´s Vazquez-Torres et al.
3.3.3. Inhibition of inducible NO synthase 8. If L-NIL is to be employed as an iNOS inhibitor, L-NIL should be provided in drinking water at a final concentration of 4 mM prior to and for the duration of the experiment. 3.3.4. Measurement of host NO production (optional) 9. The excretion of NOx (NO 2 þ NO3 ) as a measure of host NO production may be measured by collecting 24-h urine samples on isopropanol, centrifuging at 6000 g for 10 min to remove debris, and diluting 1:10 in HEPES (0.2 M, pH 7.4) (Boockvar et al., 1994). Samples may then be treated with nitrate reductase and assayed for nitrite using the Griess reagent as described in the first procedure.
REFERENCES AVMA Panel on Euthanasia (2001). 2000 Report of the AVMA panel on euthanasia. J. Am. Vet. Med Assoc. 218, 669–696. Bang, I. S., Liu, L., Vazquez-Torres, A., Crouch, M. L., Stamler, J. S., and Fang, F. C. (2006). Maintenance of nitric oxide and redox homeostasis by the SALMONELLA flavohemoglobin hmp. J. Biol Chem. 281, 28039–28047. Boockvar, K. S., Granger, D. L., Poston, R. M., Maybodi, M., Washington, M. K., Hibbs, J. B., Jr., and Kurlander, R. L. (1994). Nitric oxide produced during murine listeriosis is protective. Infect. Immun. 62, 1089–1100. Chakravortty, D., Hansen-Wester, I., and Hensel, M. (2002). SALMONELLA pathogenicity island 2 mediates protection of intracellular SALMONELLA from reactive nitrogen intermediates. J. Exp Med. 195, 1155–1166. Cobb, J. P., Natanson, C., Quezado, Z. M., Hoffman, W. D., Koev, C. A., Banks, S., Correa, R., Levi, R., Elin, R. J., Hosseini, J. M., et al. (1995). Differential hemodynamic effects of L-NMMA in endotoxemic and normal dogs. Am. J. Physiol. 268, H1634–H1642. Crawford, M. J., and Goldberg, D. E. (1998). Role for the SALMONELLA flavohemoglobin in protection from nitric oxide. J. Biol Chem. 273, 12543–12547. Ding, A. H., Nathan, C. F., and Stuehr, D. J. (1988). Release of reactive nitrogen intermediates and reactive oxygen intermediates from mouse peritoneal macrophages: Comparison of activating cytokines and evidence for independent production. J. Immunol. 141, 2407–2412. Ekman, P., Saarinen, M., He, Q., Virtala, M., Salmi, M., and Granfors, K. (1999). Human monocytic U937 cells kill SALMONELLA in vitro by NO-independent mechanisms. Infect. Immun. 67, 3670–3673. Fang, F. C. (2004). Antimicrobial reactive oxygen and nitrogen species: Concepts and controversies. Nat. Rev. Microbiol. 2, 820–832. Jabado, N., Jankowski, A., Dougaparsad, S., Picard, V., Grinstein, S., and Gros, P. (2000). Natural resistance to intracellular infections: Natural resistance-associated macrophage protein 1 (Nramp1) functions as a pH-dependent manganese transporter at the phagosomal membrane. J. Exp. Med. 192, 1237–1248.
Analysis of NO -Dependent Antimicrobial Actions
537
MacMicking, J. D., Nathan, C., Hom, G., Chartrain, N., Fletcher, D. S., Trumbauer, M., Stevens, K., Xie, Q. W., Sokol, K., Hutchinson, N., et al. (1995). Altered responses to bacterial infection and endotoxic shock in mice lacking inducible nitric oxide synthase. Cell 81, 641–650. Mastroeni, P., Vazquez-Torres, A., Fang, F. C., Xu, Y., Khan, S., Hormaeche, C. E., and Dougan, G. (2000). Antimicrobial actions of the NADPH phagocyte oxidase and inducible nitric oxide synthase in experimental salmonellosis. II. Effects on microbial proliferation and host survival in vivo. J. Exp. Med. 192, 237–248. McCollister, B. D., Bourret, T. J., Gill, R., Jones-Carson, J., and Vazquez-Torres, A. (2005). Repression of SPI2 transcription by nitric oxide-producing, IFNgammaactivated macrophages promotes maturation of SALMONELLA phagosomes. J. Exp. Med. 202, 625–635. Reed, L. J., and Muench, H. (1938). A simple method of estimating 50 per cent end-points. Am. J. Hygiene 27, 493–497. Saito, S., Onozuka, K., Shinomiya, H., and Nakano, M. (1991). Sensitivity of bacteria to NaNO2 and to L-arginine-dependent system in murine macrophages. Microbiol. Immunol. 35, 325–329. Schapiro, J. M., Libby, S. J., and Fang, F. C. (2003). Inhibition of bacterial DNA replication by zinc mobilization during nitrosative stress. Proc. Natl. Acad. Sci USA 100, 8496–8501. Shiloh, M. U., MacMicking, J. D., Nicholson, S., Brause, J. E., Potter, S., Marino, M., Fang, F., Dinauer, M., and Nathan, C. (1999). Phenotype of mice and macrophages deficient in both phagocyte oxidase and inducible nitric oxide synthase. Immunity 10, 29–38. Shiloh, M. U., Ruan, J., and Nathan, C. (1997). Evaluation of bacterial survival and phagocyte function with a fluorescence-based microplate assay. Infect. Immun. 65, 3193–3198. Stamler, J. S., Lamas, S., and Fang, F. C. (2001). Nitrosylation, the prototypic redox-based signaling mechanism. Cell 106, 675–683. Stenger, S., Donhauser, N., Thuring, H., Rollinghoff, M., and Bogdan, C. (1996). Reactivation of latent leishmaniasis by inhibition of inducible nitric oxide synthase. J. Exp. Med. 183, 1501–1514. Stevanin, T. M., Ioannidis, N., Mills, C. E., Kim, S. O., Hughes, M. N., and Poole, R. K. (2000). Flavohemoglobin Hmp affords inducible protection for Escherichia coli respiration, catalyzed by cytochromes bo’ or bd, from nitric oxide. J. Biol. Chem. 275, 35868–35875. Stevanin, T. M., Moir, J. W., and Read, R. C. (2005). Nitric oxide detoxification systems enhance survival of Neisseria meningitidis in human macrophages and in nasopharyngeal mucosa. Infect. Immun. 73, 3322–3329. Stevanin, T. M., Poole, R. K., Demoncheaux, E. A., and Read, R. C. (2002). Flavohemoglobin Hmp protects Salmonella enterica serovar typhimurium from nitric oxide-related killing by human macrophages. Infect. Immun. 70, 4399–4405. Vazquez-Torres, A., Jones-Carson, J., Mastroeni, P., Ischiropoulos, H., and Fang, F. C. (2000). Antimicrobial actions of the NADPH phagocyte oxidase and inducible nitric oxide synthase in experimental salmonellosis. I. Effects on microbial killing by activated peritoneal macrophages in vitro. J. Exp. Med. 192, 227–236. Vazquez-Torres, A., Vallance, B. A., Bergman, M. A., Finlay, B. B., Cookson, B. T., JonesCarson, J., and Fang, F. C. (2004). Toll-like receptor 4 dependence of innate and adaptive immunity to SALMONELLA: Importance of the Kupffer cell network. J. Immunol. 172, 6202–6208. Webb, J. L., Harvey, M. W., Holden, D. W., and Evans, T. J. (2001). Macrophage nitric oxide synthase associates with cortical actin but is not recruited to phagosomes. Infect. Immun. 69, 6391–6400.
538
Andre´s Vazquez-Torres et al.
Weinberg, J. B. (1998). Nitric oxide production and nitric oxide synthase type 2 expression by human mononuclear phagocytes: A review. Mol. Med. 4, 557–591. Zaharik, M. L., Cullen, V. L., Fung, A. M., Libby, S. J., Kujat Choy, S. L., Coburn, B., Kehres, D. G., Maguire, M. E., Fang, F. C., and Finlay, B. B. (2004). The SALMONELLA enterica serovar typhimurium divalent cation transport systems MntH and SitABCD are essential for virulence in an Nramp1G169 murine typhoid model. Infect. Immun. 72, 5522–5525. Zhou, X., Potoka, D. A., Boyle, P., Nadler, E. P., McGinnis, K., and Ford, H. R. (2002). Aminoguanidine renders inducible nitric oxide synthase knockout mice more susceptible to SALMONELLA typhimurium infection. FEMS Microbiol. Lett. 206, 93–97.
C H A P T E R
T W E N T Y- S E V E N
Measuring Nitric Oxide Metabolism in the Pathogen Neisseria meningitidis Melanie J. Thomson,* Tania M. Stevanin,† and James W. B. Moir* Contents 1. Introduction 1.1. Neisseria phylogeny and pathogenicity 2. Safety Aspects of Handling N. meningitidis in the Laboratory 3. Metabolism of Neisseria sp. 4. Experimental Approaches to Analyzing Nitrogen Metabolism Relevant to NO 4.1. Growth of cultures 5. Simultaneous Measurement of Oxygen and NO during Pure Culture of N. meningitidis 5.1. Action of amperometric electrodes for oxygen and NO measurement 5.2. Measuring oxygen and NO simultaneously in live N. meningitidis culture 5.3. Use of NO donors as inducers of NO-dependent gene expression 6. Measurement of NO Production/Disappearance in Tissue Culture Using Human Monocyte-Derived Macrophages 6.1. Overview 6.2. Macrophage culture and infection with N. meningitidis 6.3. Monocyte-derived macrophage infection with N. meningitidis 6.4. Measuring NO (and nitrite and nitrate) using chemiluminescence 6.5. How chemiluminescence measurement works 6.6. Advantages and disadvantages of the chemiluminescence method 6.7. Summary References
* {
540 540 541 541 544 544 547 547 551 553 555 555 555 556 556 557 558 558 558
Department of Biology, University of York, Heslington, York School of Medicine and Biomedical Science, University of Sheffield, Sheffield, United Kingdom
Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37027-4
#
2008 Elsevier Inc. All rights reserved.
539
540
Melanie J. Thomson et al.
Abstract This chapter illustrates some of the considerations that need to be made when analyzing nitric oxide (NO) metabolism of the pathogen Neisseria meningitidis. These considerations are pertinent to other bacteria and, in particular, other pathogens. First, because of the small culture volumes that can generally be managed safely, culture experiments are maintained in small volumes. We show a method for simultaneously measuring oxygen and nitric oxide during growth of N. meningitidis in a bioreactor/electrode chamber adapted from commercially available equipment. NO and NO-generating compounds can be used to investigate the impact of NO on N. meningitidis metabolism and gene expression in pure culture. Also, methods are described for analyzing the role of NO during the interaction between N. meningitidis and human macrophage cells that generate NO.
1. Introduction 1.1. Neisseria phylogeny and pathogenicity The human-restricted pathogen Neisseria meningitidis is a species of Betaproteobacteria from the family Neisseriaceae and is closely related to the sexually transmitted pathogen Neisseria gonorrhoeae with at least 95% homology based on nucleotide identity (Kawai et al., 2006). N. meningitidis is a capsulated, Gram-negative diplococcus, which resides as part of the normal flora of the nasopharynx of approximately 10% of the human population (Claus et al., 2005). However, asymptomatic carriage of this species of bacteria can lead to invasive disease via mechanisms, which are poorly understood. Other human pathogens in the Betaproteobacteria genus include Bordetella pertussis and Burkholderia mallei. Meningococcal disease is a life-threatening illness that can manifest itself either as bacterial septicemia and systemic shock or by bacteria crossing the blood–brain barrier to inflame the membranes called meninges surrounding the brain. Both outcomes can lead to serious morbidity (such as brain damage and limb necrosis) and, in some cases, to rapid onset of death. Treatment is high-dose antibiotic therapy, which is relatively successful if administered in the early stages disease—even so, up to 10–15% of patients treated will die (Fraser et al., 2006). Recent advances in vaccine technology have seen the commercial release of a conjugate serotype C vaccine for infants and quadrivalent polysaccharide subunit vaccine (MENCEVAX, SmithKline Beecham) to serotypes A, C, Y, and W135. This vaccine is based on the capsular polysaccharide antigens of these four strains. MENCEVAX is now given routinely to adolescents in Britain and pilgrims to the Hajj in Saudi Arabia and has reduced the carriage and infection rates of these strains of N. meningitidis in adults. The most prevalent serotypes in Europe, America, and Australia/New Zealand are
Measuring NO in N. meningitidis
541
serotypes B and C. The capsule of serotype B mimics the polysaccharide structure on the surface of human cells and hence cannot be used as a vaccine target in case of autoimmune reactions in the recipients. Current trials for a serotype B vaccine seek to exploit the natural ‘‘blebbing’’ of the outer membrane of Neisseria sp. into vesicles to provoke an immune response. These advances in N. meningitidis vaccinology are explained in more detail in an excellent review by Girard and colleagues (2006).
2. Safety Aspects of Handling N. meningitidis in the Laboratory Because of its status as a human pathogen, certain special measures need to be considered when working with nitric oxide (NO) metabolism in N. meningitidis in order to ensure the protection of laboratory workers (and other staff ). Laboratory strains of serotype B N. meningitidis, such as MC58, are classified in the United Kingdom as biohazard category 2 organisms, as no vaccine is currently available but early treatment with high-dose antibiotics may be successful. Adequate provision of personnel training is necessary to reduce the risks, and awareness of the symptoms of infection may aid in early medical intervention. Because the meningococcus can be transferred by airborne droplets, most handling procedures must minimize the formation of aerosols. All manipulations of open cultures should be carried out in a class II laminar flow safety cabinet to ensure the containment of aerosols at all times. Cultures should be in air-tight, disposable plastic containers [e.g., universal bottles (Sterilin) with gasketed lids], and volumes of liquid cultures are to be kept small (between 5 and 25 ml). Another method for minimizing the risks to laboratory workers is to consider using nonpathogenic strains of Neisseria, such as a noncapsulated strain or N. lactamica, a benign commensal strain found to inhabit the throats of infants. However, when conducting physiological studies, such strains may prove to be inappropriate as a consequence of different growth rates and culturing requirements. Experimental design is another aspect where safety plays a part—metabolic studies using chemostats or fermenters are hindered in a laboratory setting because of the requirement to have small volumes that can be manipulated easily inside a safety cabinet.
3. Metabolism of Neisseria sp. Neisseria are typically considered as aerobic organisms but have since been shown in both N. meningitidis and N. gonorrhoeae to be capable of using molecules other than oxygen as respiratory electron acceptors as seen,
542
Melanie J. Thomson et al.
for example, in studies by Lissenden and colleagues (2000) and Anjum and co-workers (2002). N. meningitidis has a strict requirement for oxygen, and in low oxygen (microaerobic) conditions can supplement growth using a truncated denitrification pathway, using nitrogen compounds as electron acceptors (Rock et al., 2005) (Fig. 27.1). Neisseria sp. have the ability to reduce nitrite (via nitric oxide) to nitrous oxide. The enzyme that catalyzes the nitrite reduction to nitric oxide is a two-domain, coppercontaining nitrite reductase, AniA, which is predicted to be in the periplasm and attached covalently to the outer membrane by a palmityl residue via a lipoprotein anchor. The product of this reaction is then reduced further to nitrous oxide by the membrane-bound, nitric oxide reductase enzyme, NorB. The aniA and norB genes are divergently transcribed from a shared promoter region and their regulation is far from completely elucidated. It is known that aniA is controlled positively by the global anaerobic transcriptional activator FNR [originally named fumarate and nitrate reduction regulator (Lambden and Guest, 1976)] (Householder et al., 1999) and negatively by the nitric oxide responsive repressor NsrR [originally named nitrite sensitive response regulator (Beaumont et al., 2004)] (Overton et al., 2006; Rock et al., 2007). Binding sites for further regulators NarP/Q
O.M.
e− Cyt c
Periplasm
NorB
2NO
N2O
QH2 Q
NADH Dehydrogenase
NADH NAD+
e−
e−
QH2 Q
bc1
Cyt c
AniA
NO−2 NO
e−
cbb3 Oxidase
½ O2
I.M.
H2O
Cytoplasm
Figure 27.1 Principal respiratory pathways in N. meningitidis. Electrons pass from NADH to oxygen via NADH dehydrogenase, the cytochrome bc1 complex, and the cytochrome cbb3 oxidase under aerobic conditions. Aerobically expressed proteins are shown in white. Gray boxes show the anaerobically inducible proteins AniA (nitrite reductase) and NorB (nitric oxide reductase). I.M., inner membrane; O.M., outer membrane. Not drawn to scale.
Measuring NO in N. meningitidis
543
(nitrite responsive) and Fur (iron responsive) are also present in the aniA promoter region (Householder et al., 1999). The regulation of norB is less complicated as it appears only to be repressed by NsrR, which is removed rapidly in the presence of NO (Rock et al., 2007). Possible reasons for this difference in levels of control may be because of the toxic nature of NO. Expression of a gene product that produces NO as a by-product (AniA) must be tightly regulated to prevent ‘‘suicide,’’ whereas the gene product that detoxifies NO to a harmless substance (NorB) is controlled minimally so that it may respond quickly to either endogenous or exogenous (host) NO production. These denitrification enzymes are found in all five completed Neisseria genomes now sequenced. This system was originally thought to be present only as a detoxification system for protecting bacteria from nitrosative bursts via inducible nitric oxide synthase (iNOS) from host immune cells, such as macrophages. Indeed, such detoxification systems have been shown to enhance the survival of N. meningitidis within human macrophages and in a nasopharyngeal mucosa organ culture model (Stevanin et al., 2005). However, as the microenvironments within the host have variable levels of oxygen, it may be advantageous for bacteria to utilize the nitrite in host tissues. Nitrite is produced from the oxygenation of NO—a common signaling molecule in the human host as well as a toxin produced by the host innate immune system. Nitrite is also produced in the nasopharynx by commensal facultative nitrate reducers from ingested nitrates in food—levels of 0.01–5 mM are considered normal in saliva (Lundberg et al., 2004). Nitric oxide levels in the nasopharynx are relatively high (250 ppb 12.5 nM ) and are produced by sinus cells, which constitutively express nitric oxide synthase (NOS) (Lundberg and Weitzberg, 1999). Denitrification has been shown in laboratory cultures to supplement growth under oxygen-limited conditions (Rock and Moir, 2005), leading to the suggestion that such flexibility in respiratory lifestyles may add to the success of the pathogen in the human host. Evidence to support the role for denitrifying enzymes in pathogenesis of Neisseria sp. is that the AniA protein can be detected by sera from patients suffering from gonococcal infection (Clark et al., 1988) and confers resistance to serum killing by the human complement system (Cardinale and Clark, 2000). Two main types of nitric oxide reductase enzymes have been identified in bacteria. The cNOR type found in proteobacteria such as Pseudomonas stutzeri (Heiss et al., 1989) and Paracoccus denitrificans (Carr and Ferguson, 1990) consists of a two-component complex made up of the membraneanchored c-type cytochrome (NorC), which donates electrons to the transmembrane NorB subunit, which contains two b-type hemes. The other type, first identified in Ralstonia eutropha, is the single subunit qNOR, which is thought to accept electrons from quinols, using a quinol-oxidizing domain located on the extended N terminus of the protein instead of from c-type cytochromes. N. meningitidis NorB is closely related to the qNOR from the Ralstonia sp. (Hendriks et al., 2000)
544
Melanie J. Thomson et al.
4. Experimental Approaches to Analyzing Nitrogen Metabolism Relevant to NO 4.1. Growth of cultures Neisseria meningitidis is fastidious and should be grown overnight on horse blood agar from glycerol stocks for a maximum of 16 h before use to ensure minimal mutation rates or downregulation of virulence factors. This may result from the phase variability of some Neisseria genes. N. meningitidis can be cultured successfully in a low-carbohydrate medium, such as Mueller– Hinton broth (MHB) both aerobically and microaerobically. Aerobic growth is typically performed in small volumes (e.g., 5 ml) in a container with large air space with rapid agitation (i.e., 200 rpm). This is to ensure adequate gas exchange to saturate the growth medium with air and to prevent the expression of gene products controlled by FNR in response to low oxygen conditions. The growth rate should be rapid and the resulting curve should have a short lag phase, followed by an exponential phase and ending in a stationary phase as seen as the solid black lines of both graphs of Fig. 27.2. Microaerobic conditions are achieved by increasing the volume to 25 ml, allowing only minimal air space and reducing the shaking speed to 90 rpm in an effort to reduce the air exchange with the liquid medium. This is to produce an oxygen-limited environment, which in turn enables FNR to activate the expression of gene products such as AniA. The growth rate is much slower and final cell density is lower (see Fig. 27.2). Neisseria sp. are obligate human pathogens and require carbon dioxide (CO2) for growth. Therefore, 10 mM fresh sodium bicarbonate is added to cultures as a CO2 donor. To create denitrifying conditions, 5 mM nitrite is added to the growth medium. The amount of inoculum is also important, and optimum for these growth experiments was found to be 0.05–0.1 OD600 nm. Before any physiological studies can be performed, basic growth experiments of various strains must be performed. This knowledge identifies the optimum phases in which to harvest bacterial cells for further analysis and hence aids experimental design. This is particularly relevant to the comparison of bacteria from aerobic and microaerobic cultures, as an ‘‘aerobic’’ culture may inadvertently become oxygen limited if allowed to grow beyond the midexponential phase. This lesson was learned when aerobic cultures of MC58 were supplemented with 5 mM nitrite as a control. The growth levels þ/ nitrite were similar until cultures reached the midexponential phase (OD600 nm 0.6), whereupon the nitritecontaining sample had a rapid drop in OD600 nm (see Fig. 27.2A). This phenomenon was seen in other laboratory and clinical strains of N. meningitidis that contained operational denitrification genes. The clinical strains were a gift from Professor Robert Read and appear in a paper by Townsend
545
Measuring NO in N. meningitidis
A 1.4 1.2
OD600 nm
1.0 0.8 0.6 0.4 0.2 0.0 0
1
2
3
4
5
6
7
8
9
0
1
2
3
4 5 Time (h)
6
7
8
9
B 0.8 0.7
OD600 nm
0.6 0.5 0.4 0.3 0.2 0.1 0.0
Figure 27.2 Growth curves of MC58 (A) and Z4662 aniA (B) strains of N. meningitidis. Conditions are labeled as aerobic (black line), aerobic þ 5 mM nitrite (gray line), and denitrifying (black dashed line). A rapid drop in OD600 nm was seen at OD 0.95 in the MC58 aerobic þ nitrite cultures as opposed to no difference observed between the aerobic cultures of the Z4662 aniA strain. Growth in MC58 denitrifying cultures continues until nitrite is depleted at 5^6 h, after which the cultures enter stationary phase. The inability of Z4662 aniA to use nitrite as an electron acceptor restricts growth under denitrifying conditions and hence little increase in OD600 nm is seen.
and co-workers (2002). Intriguingly, it was not observed in supplemented aerobic cultures of the mutant strain MC58 aniA or two natural aniA mutant clinical strains (Table 27.1)—the growth profiles of one clinical strain (Z4662 serotype B) are shown in Fig. 27.2B. This anecdotal evidence suggests that an increase in cell density leads to oxygen becoming limited in liquid media, and hence the expression of AniA is triggered as a consequence of FNR activation in response to low oxygen tension. This in turn would produce a rapid and toxic buildup of NO, which diffuses rapidly into
546
Melanie J. Thomson et al.
Table 27.1 Comparison of metabolic and denitrifying factors in laboratory and clinical strains of N. meningitidisa
Strain
Laboratory MC58 MC58 aniA C11 Clinical Z3524 Z3842 Z4242 Z4421 Z4662 Z4673 a
Aerobic growth
Aerobic growth þ NO 2
AniA expression
Nitrite reduction
NOR activity
þ þ
þ
þ
þ
þ
þ
þ
þ
þ
þ þ þ þ þ þ
þ þ
þ þ þ þ
þ þ þ þ
þ þ þ þ
Aerobic growth was carried out in 7.5 ml MHB in a 50ml tube, 200 rpm þ/ 5 mM NO 2 . AniA expression and nitrite usage were tested by Western blotting of whole cell lysates and nitrite assays from microaerobic cultures. NO reductase activity was assessed after exogenous addition of 10–50 mM NONOate to aerobic cultures. Strains that expressed functional AniA had a rapid drop in OD600 nm in cultures containing nitrite compared to healthy growth seen in the cultures of aniA strains (see Fig. 27.2). This decrease may be because of a toxic accumulation of NO resulting from AniA activity. NO reductase activity could not be induced in strains lacking in AniA activity even though the level of exogenous NONOate required to induce NorB expression (at least 10 mM ) was not toxic to the cells.
the culture medium. Accumulation of toxic NO is so rapid that death occurs before sufficient NorB activity has been expressed. This of course would only happen in strains that could express viable AniA protein and not those that lack the enzyme. Indeed, nitrite sensitivity or tolerance of some N. meningitidis strains has been described in the past (Bovre, 1984). Initial investigations into the existence of denitrification enzymes in N. meningitidis were performed by Hoehn and Clark (1990). That report described the levels of Pan1 (former name of AniA) in meningococcus as much lower (or absent) than those seen in gonococcus. These low levels in the meningococcus may be related to the fact that the workers used ‘‘nitrite-tolerant’’ strains of N. meningitidis. Initially one might assume that nitrite tolerance would be associated with the expression of a nitrite detoxification system, such as the nitrite reductase AniA. As it turns out, strains expressing AniA are, in fact, more sensitive to nitrite under many culture conditions. This was in contrast to strains that did not express a functional AniA protein that showed no difference in aerobic growth patterns, regardless of the addition or absence of nitrite (see Table 27.1).
547
Measuring NO in N. meningitidis
5. Simultaneous Measurement of Oxygen and NO during Pure Culture of N. meningitidis 5.1. Action of amperometric electrodes for oxygen and NO measurement 5.1.1. Overview As explained earlier, traditional procedures for physiological studies in bacteria involving the use of chemostats are not viable for these dropletborne pathogens. This is because the maximum volume allowed for cultures is lower than that of the smallest chemostat currently available commercially. Chemostats are replaced by smaller scale equipment, which can be set up easily inside a class II safety cabinet. The physical setup of the chamber consists of a standard Rank Brothers oxygen electrode (Rank Brothers, Cambridge, UK) that has been adapted to allow insertion of the ISO-NOP 2-mm nitric oxide electrode, supplied by World Precision Instruments (WPI, Stevenage, UK) (Fig. 27.3). Both oxygen and NO levels are monitored using amperometric detection, which is an electrochemical technique where the electrode potential is held constant and the current is the measured variable. Most commercially available systems are based on a Clark-type electrode connected to a potentiostat and data collection system. The Clark electrode model consists of an electrode compartment isolated from the reaction chamber by a thin Teflon membrane, which is permeable to gases allowed to reach the cathode, where they are reduced electrolytically. The reduction allows a current to flow, which is recorded on a data recorder. The trace is thus a measure of the levels of a specific gas in the reaction mixture. The oxidation of the counter (or reference) electrode completes the opposite reaction.
5.1.2. The oxygen electrode The oxygen electrode used is supplied by Rank Brothers, and the following information is adapted with permission from the electrode instruction manual (Peter Rank, Rank Brothers). The system consists of a platinum electrode to detect oxygen and a reference electrode made from silver. When the platinum electrode is polarized at –600 mV with respect to the silver electrode, oxygen molecules that reach its surface from the test medium are reduced to hydroxide ions through the following part reaction:
O2 þ 2H2 O þ 4e ! 4OH :
ð27:1Þ
548
Melanie J. Thomson et al.
Lead to ISO-NO-mark II potentiostat
Hamilton syringe for injection/sampling
Tape to stop NO electrode falling to base of chamber Two extra holes drilled through wall of plunger
ISO-NOP 2mm nitric oxide electrode
Rubber seal
Water bath IN
Water bath OUT NO O2 Magnetic stirrer bar
Dissolved oxygen electrode
Lead to digital 20 controller
Figure 27.3 Oxygen and NO electrodes for simultaneous measurement of O2 and NO.
For every reduction reaction there must be oxidation, which occurs at the silver electrode as follows:
4Ag þ 4Cl 4e ! AgCl:
ð27:2Þ
Thus, the overall electrochemical process that occurs in an oxygen electrode is as follows:
4Ag þ O2 þ 2H2 O þ 4Cl ! 4AgCl þ 4OH :
ð27:3Þ
Measuring NO in N. meningitidis
549
The current flowing is proportional to the activity of oxygen, provided the solution is stirred constantly (by a mechanical flea) to minimize formation of an unstirred layer next to the membrane. A digital controller mechanism converts the current to a voltage output, which represents the percentage air saturation of the solution. The Rank Brothers electrode setup has been designed to measure the uptake or production of oxygen by cell suspensions, subcellular particles, or enzyme systems. As supplied it comes in two main components.
The electrode base (pedestal), which houses the central platinum working electrode and the surrounding silver/silver chloride reference electrode. Conduction between these electrodes is by saturated potassium chloride with a semipermeable Teflon membrane used to separate the electrode from the sample. The incubation chamber mounted on the base has a built-in water jacket. This ensures that measurements are conducted at a constant temperature, which is essential for accurate readings. A plunger fits into the top of the chamber to seal the sample from the atmosphere apart from a very small hole used to inject materials with a syringe. This plunger can be adapted by having two holes (2.5 mm diameter) drilled in the walls, enabling insertion of the ISO-NOP 2-mm nitric oxide electrode (WPI) in one hole and injection/ sampling from the second hole with a Hamilton syringe. Incubation chambers come in a variety of sizes and can be made from Perspex or glass. The smaller Perspex chamber is more robust but has a maximum volume of 5–7 ml, whereas the larger glass chamber can hold up to 20 ml and has the added advantage of being autoclavable. However, gas exchange in the larger, taller electrode is poor compared to that of the smaller, shorter electrode when carrying out aerobic studies but is ideal when microaerobic conditions are required. The oxygen electrode is easy to use and maintain as all components are separate and hence easily assembled, disassembled and stored. The silver reference electrode will become oxidized when in contact with air and the manufacturers recommend leaving a thin, uniform layer of this oxidation in place to stabilize the electrode readings. When this layer becomes too thick or a higher sensitivity is required, the silver disc can be polished with alumina powder. 5.1.3. The nitric oxide electrode The nitric oxide electrode is supplied by WPI as a 2-mm probe consisting of an inner combination working/reference electrode and a Faraday shielded stainless steel sleeve, which contains the electrolyte. The tip of the sleeve is covered with a proprietary NO selective membrane able to exclude other anionic molecules such as ascorbic acid and nitrite. The working electrode consists of a platinum disc polarized at 860 mV with respect to the Ag/AgCl
550
Melanie J. Thomson et al.
reference electrode. When this positive potential is applied via the ISO-NO Mark II potentiostat, NO is oxidized at the working electrode surface, producing a redox current. The reaction is as follows:
NO e ! NOþ :
ð27:4Þ
This forms the cation nitrosonium, a strong Lewis acid that is converted to nitrite in the presence of OH- via the reaction
NOþ þ OH ! HNO2 ! Hþ þ NO 2:
ð27:5Þ
Nitrite can then be oxidized further to nitrate (Zhang, 2004). The sensitivity of this NO detection system is said to be as low as 0.1 pA as a consequence of the low noise circuits of the ISO-NO Mark II and the large size of the working electrode. The system rapidly detects NO in real time, with a small delay of approximately 0.5 s. The NO electrode is relatively easy to use as long as special care is taken not to disrupt the electrode, which can happen during rough handling, particularly insertion into the side wall of the oxygen electrode chamber. It also requires at least 16 h connected to the ISO NO Mark II potentiostat before use to allow the electrode to stabilize at a baseline—this process is known as electrode conditioning. It should always be suspended in fluid (stored in distilled H2O) and not allowed to touch any hard surfaces or be allowed to hit the bottom/stirrer bar in the oxygen electrode chamber. Following such an incident, the electrode should be allowed sufficient time to stabilize to baseline before further use. If the user suspects the membrane on the outer sleeve to be ruptured, immersion in 1 M NaCl results in an ‘‘overscale’’ reading and the sleeve and electrolyte must be replaced.
5.1.4. Calibration of oxygen and NO electrodes Both the oxygen and the NO electrodes are temperature sensitive and hence all calibrations must be carried out at the temperature required for experiments. The oxygen electrode can be calibrated to 100% air by using an airsaturated solution of the medium to be used in the experiment using the protocol supplied by the manufacturer. The NO electrode can be calibrated in many ways for fluid measurement of NO. These involve chemical generation of NO from acidified nitrite, decomposition of NO donors [such as S-nitrosothiol, S-nitroso-Nacetylpenicillamine (SNAP)], or the use of aqueous standards prepared from a saturated NO solution. The chemical method is the most convenient and cost effective for the ISO-NOP 2-mm sensors, which exploits the
551
Measuring NO in N. meningitidis
stoichiometry of nitrite reduction to calculate the amount of NO released as per the following equation:
2NaNO2 þ 2KI þ 2H2 SO4 ! 2NO þ I2
ð27:6Þ
þ 2H2 O þ 2Na2 SO4 :
Potassium iodide (KI) and sulfuric acid (H2SO4) are used in excess in the calibration procedure and hence the limiting reagent is sodium nitrite (NaNO2). The calibration is conducted according to the manufacturer’s instructions, but using a 5-ml volume in the chamber as opposed to the 10 ml recommended by WPI. The concentrations of NO for the standard curve are then adjusted to account for the dilution factor. Typical calibration and standard curve data are shown in Figure 27.4.
5.2. Measuring oxygen and NO simultaneously in live N. meningitidis culture 5.2.1. Overview The combination system described earlier was used by the authors of Rock and co-workers (2005) to study the physiology of microaerobic growth of MC58 and the switch from aerobic respiration to denitrification.
Current (pA)
Current (pA)
5000
y = 2.15x + 63.2 R2 = 0.9937
4000 3000 2000 1000 0
0
500
1000
1500
2000
[NO] (nM) Time (s) 25 ml
50 ml
100 ml
200 ml
Volume added to chamber of 50 mM sodium nitrite
Figure 27.4 Typical NOcalibration outputdata and standard curve (inset).
552
Melanie J. Thomson et al.
70 60 50 40 30 20 10 0
1.0 0.8 0.6 0.4
[NO] ( mM)
[O2] (% air saturation)
This chapter states that while N. meningitidis fails to grow under strict anaerobic conditions, under low oxygen conditions it can supplement growth using a truncated denitrification pathway. In these experiments, a 25-ml glass electrode chamber was used with no cap. These conditions, combined with slow stirring, mimicked the microaerobic conditions in sealed universal bottles used in complementary growth, nitrite, and b-galactosidase assays. The inherent oxygen respiratory activity in the culture led to oxygen depletion from the medium over the course of an hour (Fig. 27.5). One of the biggest obstacles for this type of work with slow-growing bacteria is the threat of contamination by other faster-growing species. Some of the test strains contained chromosomally located antibioticresistant genes and hence this can be exploited to keep other contaminants at bay. Careful study of the metabolism of these bacteria with and without these antibiotics is required first to ensure that the presence of antibiotics in the culture medium does not have inhibitory growth effects. If an effect is seen, the antibiotics should be excluded for the liquid culture experiments. While the glass chamber could be autoclaved, the Perspex oxygen electrode base and the components required to set up the oxygen electrode could not. Disinfection of the base and components with 70% ethanol and then washing with copious amounts of dH2O before immediately setting up inside the sterile class II cabinet were precautions taken to avoid contamination by other laboratory strains (e.g., Escherichia coli). Contamination by commensal skin bacteria can also be reduced by using a good aseptic technique. Each culture was tested at the end of the incubation period by Gram staining and light microscopy for contaminants.
0.2 0
50 100 150 200 250 300 350 400 450 Time (min)
0
Figure 27.5 Typical O2 and NO output data from live cultures of N. meningitidis MC58 grown under denitrifying conditions in a 25-ml glass chamber. This figure is adapted from Rock et al. (2005). The O2 (gray line) is slowly depleted by bacterial respiration, which in turn induces AniA expression and a rapid accumulation of NO (black line) is seen. After approximately 1 h, NorB activity then removes NO from the medium. During the period of NO accumulation, oxygen increases due to inhibition of the oxygen reductase activity of cytochrome cbb3 oxidase.
Measuring NO in N. meningitidis
553
Figure 27.5 shows typical oxygen/NO output data of a microaerobic culture of wild-type MC58 N. meningitidis supplemented with 5 mM nitrite. Over the first 1–2 h, oxygen (see Fig. 27.5, gray line) is depleted from the medium. NO (black line) then starts to accumulate in the medium as a result of AniA nitrite reductase activity. When [NO] reaches 100–200 nM, oxygen begins to accumulate as the rate of oxygen respiration drops (NO is a powerful inhibitor of the oxidase). NO continues to accumulate, reaching a peak at [NO] 1–5 mM. No growth is seen during period of high [NO]. About 1 h after the NO starts to accumulate, [NO] depletion from the medium accelerates at a rapid rate, indicating that NorB nitric oxide reductase is being expressed in response to increased [NO]. 5.2.2. Protocol for simultaneous measurement of oxygen and NO in N. meningitidis culture 1. Sterilize (autoclave) the glass electrode chamber and set up on an oxygen electrode base inside a biohazard type II cabinet. 2. Heat the recirculating water bath to 37 and connect to a water jacket of an electrode chamber. 3. Calibrate the oxygen electrode (as described earlier) with air-saturated MHB. 4. Add 30 ml MHB supplemented with 5 mM nitrite and allow adjustment to temperature. 5. Suspend the NO electrode into the middle of the chamber and allow the current to stabilize. 6. Adjust the stirring speed until the surface of the medium has minimal perturbation. 7. Inoculate the chamber from a midlog phase aerobic culture to a starting OD600 nm of 0.05–0.1. 8. Monitor oxygen and NO levels simultaneously over several hours of growth. 9. Take samples for growth, nitrite, and b-galactosidase assays. Check for culture purity by Gram stain.
5.3. Use of NO donors as inducers of NO-dependent gene expression 5.3.1. Overview Nitric oxide donors are chemicals that generate NO through mechanisms that are independent of the enzymatic action of NOS (see review by Wang et al., 2002). Commonly used agents are organic nitrates (e.g., glyceryl trinitrate, isosorbide dinitrate), sodium nitroprusside, sydnonimines (e.g., molsidomine, SIN-1), S-nitrosothiols (e.g., s-nitrosoglutathione, SNAP), and NONOates (e.g., spermine NONOate, DETA-NONOate).
554
Melanie J. Thomson et al.
Each type of NONOate dissociates into a free amine group and NO in a pH-dependent manner according to first-order kinetics. Each different NONOate has a specific half-life at pH 7.4 and 37 . For example, spermine NONOate has a half-life of 39 min at pH 7.4, 37 , whereas DEANONOate has a shorter half-life of 2 min and DETA-NONOate a longer half-life of 20 h. NONOate compounds are highly hygroscopic and must be stored at –20 or below in an air-tight container with moisture-absorbing granules. Vials are to be discarded if the white powder turns yellow upon long-term storage. Experimental design will dictate which NONOate is to be used. For experiments to induce NorB expression in N. meningitidis (Rock et al. 2007), spermine NONOate was chosen as it donates NO for the amount of time deemed to be sufficient (t½ ¼ 39 min) for the induction of gene expression but then dissipates before harvesting of the bacteria (1–2 h postpulse) and subsequent NO challenge in the oxygen/NO electrode system. The quantity of exogenous NO (via NONOate) to which a culture is exposed is also important, as in low doses it is ineffective at inducing gene expression and in high doses it is toxic. Titration of N. meningitidis strains with spermine NONOate showed that the lowest dose to induce detectable NorB expression was 10 mM and the highest dose tolerated by MC58 wild type was 50 mM. Intriguingly, strains that lacked a functional nitrite reductase protein (AniA) were more sensitive to the higher dose of NONOate, and hence the lower dose was required to prevent cell death (see Table 27.1). When the ability of these AniA- strains to reduce NO was compared to that of MC58 wild type and other AniAþ clinical strains treated with the same low dose (10 mM), no induction of NorB expression was detected on subsequent challenge with NO in the oxygen/NO electrode system (see Table 27.1). Controls are also necessary to rule out the role of nitrite in the system, as some of the NONOate is oxidized to nitrite. Negative controls such as cells harvested from cultures not pulsed with NONOate are also to be performed on the same day. Results of this functional assay can be supplemented by promoter studies using lacZ–fusion constructs and relative quantitation of mRNA of target genes using a real-time polymerase chain reaction. 5.3.2. Protocol for assessing NO reductase activity via NO uptake assay 1. Culture bacteria aerobically (200 rpm) in 7.5 ml MHB until the OD600 nm is in early midexponential phase (0.3–0.6). 2. Pulse cultures by adding 50 mM spermine NONOate (1/1000 dilution from a 50 mM stock in 0.1 M NaOH) and culture for a further 2 h before harvest. 3. During bacterial culture time, set up oxygen electrode/water bath as before using a 5-ml Perspex chamber. Calibrate. 4. Insert NO electrode through hole in cap.
Measuring NO in N. meningitidis
5. 6. 7. 8.
9. 10. 11. 12. 13.
555
Calibrate NO electrode at 37 . Remove calibration solutions and rinse chamber twice with dH2O. Add 5 ml MHB medium and allow to adjust to 37 . Assess [NO] and normal breakdown rate of the gas supplied by a saturated NO solution (Cole et al., 2007) by removing oxygen from medium using a glucose/glucose oxidase plus catalase system and then adding a ‘‘spike’’ (10–50 ml) of the solution to the anoxic chamber. Remove this solution, rinse chamber twice with dH2O, and replace with fresh MHB. Harvest a maximum of 2 mg whole cells at 8000 rpm for 5 min in a microfuge. Resuspend the cell pellet in 50 ml MHB and inject into electrode chamber using a clean Hamilton syringe via the second hole in the wall of the cap. Allow bacterial cellular respiration to remove all oxygen from the medium before adding an aliquot of a known [NO] to the chamber. Record rate of NO uptake. Compare rates between pulsed and nonpulsed cultures of the same strain on the same day. Data can then be converted using the NO calibration equation, the amount of protein, and the slope of uptake data to give a rate of NO reductase activity in nmol min-1 mg of protein-1.
6. Measurement of NO Production/ Disappearance in Tissue Culture Using Human Monocyte-Derived Macrophages 6.1. Overview Experimental infection of macrophages and cell lines with intact bacteria, lipopolysaccharide (LPS), or killed bacteria, including N. meningitidis (Marriott et al., 2004), leads to activation of iNOS. However, unlike murine macrophages, NO concentrations within human macrophages remain low even following activation by proinflammatory agonists such as LPS (Weinberg et al., 1995), indicating that the NO concentration is tightly regulated in human cells. In aerated tissues, NO is oxidized rapidly to nitrite and nitrate, which, as stable products, can be used as measures of NO production/disappearance in biological systems.
6.2. Macrophage culture and infection with N. meningitidis A rapid and efficient way to isolate primary human peripheral monocytederived macrophages (MDM) is to use the Ficoll–Paque Plus (GE-Healthcare) monocyte purification kit. It consists of adding whole blood to a
556
Melanie J. Thomson et al.
Ficoll–Paque solution and centrifuging the suspension for a short period of time. Differential migration between different cell types ensures aggregation of lymphocytes in the interface between the upper granulocyte layer and the lower erythrocyte–Ficoll aggregate layer. The mononuclear cells are then gently transferred to a clean centrifuge tube and washed three times with a sterile saline solution, usually phosphate-buffered saline (PBS). For most experiments, mononuclear cells are seeded at a density of 1 106 cells/well in flat-bottom 24-well plates in 1 ml RPMI 1640 medium (Gibco BRL) supplemented with 2 mM L-glutamine and 10% heat-inactivated fetal calf serum. Plates containing mononuclear cells are incubated for 24 h at 37 in 95% air, 5% CO2, after which nonadherent mononuclear cells can be washed away. Monocytes are then allowed to differentiate into macrophages over a period of 12–14 days (Callahan et al., 2003). The medium should be changed every 3–4 days. Nutrient-depleted macrophages change their round shape, assuming a dendritic appearance.
6.3. Monocyte-derived macrophage infection with N. meningitidis When using a CO2 incubator, and grown in small volumes (5–10 ml in MHB), N. meningitidis can be grown without the addition of sodium bicarbonate for up to 3 h, which is usually enough to allow cultures to reach the optical density required in these experiments. Six medium-sized colonies are transferred from an overnight plate into MHB and incubated until it reaches an OD600 of 0.25–0.3. Bacteria are then harvested, and pellets are washed twice with PBS. N. meningitidis does not usually bind very efficiently to MDM. Therefore, to increase the numbers of bacteria associated to MDM, bacteria are opsonized in 10% human serum for 15 min at 37 with rotation. Opsonized bacteria are then harvested and washed three times in PBS. Pellets are resuspended in RPMI 1640 to obtain an inoculum of approximately 3 107 CFU in 250 ml of RPMI 1640. Wells containing MDM are infected with 250–500 ml bacterial suspensions and incubated at 37 . It takes 3–4 h for any variations in nitrite concentrations in the supernatant to be observed and 4–8 h for most of the nitrite to disappear from suspensions containing the wild-type N. meningitidis strain (MC58).
6.4. Measuring NO (and nitrite and nitrate) using chemiluminescence The standard method for measurement of NO production is to measure nitrite plus nitrate colorimetrically in the Griess reaction, which has a limit of detection 2.5 mM (Grand et al., 2001). However, measurements of NO
Measuring NO in N. meningitidis
557
produced by human macrophages showed that only about 0.5 mM could be detected in those cells (Stevanin et al., 2002), thus making the sensitivity of the Griess reaction too low to study NO in MDM. Chemiluminescence resulting from the reaction of NO with ozone, however, has a limit of detection of nM and is, therefore, a sensitive enough method to be used in studies of human macrophages. Nitrite accumulated in MDM supernatants are rereduced to NO and concentration determined using the Sievers nitric oxide analyzer (Sievers Instruments Inc.) by assaying the nitrite/nitrate accumulation in the supernate of cultured cells over a 24-h incubation at 37 . Nitrite is quantified by reduction to NO gas in a mixture of acetic acid and KI, at room temperature, whereas nitrate can be measured using vanadium(III) chloride in hydrochloric acid at 95 . At 95 , vanadium converts nitrate as well as nitrite to NO, and the difference between measurements using both solutions will give the amount of each in a given sample. Using a series of standards, calibration curves are obtained for both nitrite and nitrate, and the NO concentration is calculated from the integral of the detected signal over time in relation to the standards. In general, as little as 50 ml of sample is enough for each measurement; however, it is often useful to make duplicate injections, as handling such small volumes is difficult and any variations in the amount of sample that reach the reaction mixture will produce large variations in the final concentration detected.
6.5. How chemiluminescence measurement works The ‘‘gold-standard’’ method for the detection of NO metabolites being used at present is the ozone-based chemiluminescent assay. This assay is based on a gas-phase chemiluminescence reaction between NO and ozone:
NO þ O3 ! NO 2 þ O2 ! NO2 þ hn:
ð27:7Þ
Nitrite and/or nitrate in samples is reduced to NO under acidic conditions. The NO gas is purged out of solution by bubbling inert gas such as nitrogen through the reaction mixture. A vacuum pump is used to force the purged NO from the reaction chamber into the photomultiplier chamber inside the NO analyzer, where it reacts with ozone that has been produced by an ozone generator, connected to an oxygen cylinder. The reaction of NO with ozone generates excited state nitrogen dioxide, which emits light. Emission is in the red and near-infrared regions of the spectrum and is detected by a thermoelectrically cooled, red-sensitive photomultiplier. The photomultiplier tube is kept at subzero temperatures to maximize sensitivity.
558
Melanie J. Thomson et al.
6.6. Advantages and disadvantages of the chemiluminescence method The most obvious advantage of the method described earlier is the enhanced sensitivity. Measuring NO metabolism in biological samples can be a challenging experience because of the very low levels of detectable NO usually produced by human cells. Ozone-based chemiluminescence assays help address the sensitivity issue, as it allow detection of as little as 1–5 nM NO. Other advantages are reproducibility and the very small volumes of samples usually needed, allowing for multiple measurements, thus increasing reliability and the possibility of measuring the concentration of several metabolites in the same sample. One of the main limitations of this system, however, is the impossibility of doing real-time measurements of NO in situ. A reliable measurement of nitrate is also difficult as background levels of this ion in buffers, tissue culture medium, and so on are often greater than nitrate generated from NO in NO-producing systems.
6.7. Summary Using this sensitive measurement technique, it has been possible to show that resting MDM produce low level concentrations of NO, which is subsequently oxidized to nitrite (Stevanin et al., 2002). Is has also been possible to demonstrate that, unlike bacteria such as Streptococcus pneumoniae (Marriott et al., 2004), which induce increases in nitrite accumulation during cocultivation of MDM, in MDM infected with N. meningitidis the steady-state nitrite actually decreases, even in the presence of SNAP, demonstrating that N. meningitidis utilizes NO produced by MDM (Stevanin et al., 2005). The ability of N. meningitidis to metabolize NO is an important aspect of meningococcal survival in human macrophages, as demonstrated by the observation that an N. meningitidis knockout mutant defective in the NO reductase gene (norB) survives poorly once phagocytosed, compared to the wild-type strain (Stevanin et al., 2005). Using the iNOS inhibitor L-N G-monomethyl arginine (L-NMMA), it was possible to confirm that this survival advantage is a consequence of the resistance of the wild-type to intracellular NO-related molecule and, interestingly, there was a small but significant impairment of survival of wildtype bacteria in MDM pretreated with L-NMMA, suggesting that cell synthesis of NO is a requirement for optimal survival of wild-type N. meningitidis in the intracellular environment.
REFERENCES Anjum, M. F., Stevanin, T. M., Read, R. C., and Moir, J. W. (2002). Nitric oxide metabolism in Neisseria meningitidis. J Bacteriol. 184, 2987–2993.
Measuring NO in N. meningitidis
559
Beaumont, H. J., Lens, S. I., Reijnders, W. N., Westerhoff, H. V., and van Spanning, R. J. (2004). Expression of nitrite reductase in Nitrosomonas europaea involves NsrR, a novel nitrite-sensitive transcription repressor. Mol. Microbiol. 54, 148–158. Bovre, K. (1984). Family VIII. Neisseriaceae. In ‘‘Bergey’s Manual of Systematic Bacteriology’’ (N. R. Krieg and J. G. Holt, eds.), pp. 288–296. Williams & Wilkins, Baltimore, MD. Callahan, M. K., Halleck, M. S., Krahling, S., Henderson, A. J., Williamson, P., and Schlegel, R. A. (2003). Phosphatidylserine expression and phagocytosis of apoptotic thymocytes during differentiation of monocytic cells. J. Leukocyte Biol. 74, 846–856. Cardinale, J. A., and Clark, V. L. (2000). Expression of AniA, the major anaerobically induced outer membrane protein of Neisseria gonorrhoeae, provides protection against killing by normal human sera. Infect. Immun. 68, 4368–4369. Carr, G. J., and Ferguson, S. J. (1990). The nitric oxide reductase of Paracoccus denitrificans. Biochem. J. 269, 423–429. Clark, V. L., Knapp, J. S., Thompson, S., and Klimpel, K. W. (1988). Presence of antibodies to the major anaerobically induced gonococcal outer membrane protein in sera from patients with gonococcal infections. Microb. Pathog. 5, 381–390. Claus, H., Maiden, M. C., Wilson, D. J., McCarthy, N. D., Jolley, K. A., Urwin, R., Hessler, F., Frosch, M., and Vogel, U. (2005). Genetic analysis of meningococci carried by children and young adults. J. Infect. Dis. 191, 1263–1271. Cole, L. J., Huston, W. M., and Moir, J. W. B. (2007). Delivery of nitric oxide for analysis of the function of cytochrome c0 . Methods Enzymol. 436, chapter 2 (in press). Fraser, A., Gafter-Gvili, A., Paul, M., and Leibovici, L. (2006). Antibiotics for preventing meningococcal infections. Cochrane Database Syst. Rev. CD004785. Girard, M. P., Preziosi, M. P., Aguado, M. T., and Kieny, M. P. (2006). A review of vaccine research and development: Meningococcal disease. Vaccine 24, 4692–4700. Grand, F., Guitton, J., and Goudable, J. (2001). Optimisation of the measurement of nitrite and nitrate in serum by the Griess reaction. Ann. Biol. Clin. (Paris) 59, 559–565. Heiss, B., Frunzke, K., and Zumft, W. G. (1989). Formation of the N-N bond from nitric oxide by a membrane-bound cytochrome bc complex of nitrate-respiring (denitrifying) Pseudomonas stutzeri. J. Bacteriol. 171, 3288–3297. Hendriks, J., Oubrie, A., Castresana, J., Urbani, A., Gemeinhardt, S., and Saraste, M. (2000). Nitric oxide reductases in bacteria. Biochim. Biophys. Acta 1459, 266–273. Hoehn, G. T., and Clark, V. L. (1990). Distribution of a protein antigenically related to the major anaerobically induced gonococcal outer membrane protein among other Neisseria species. Infect. Immun. 58, 3929–3933. Householder, T. C., Belli, W. A., Lissenden, S., Cole, J. A., and Clark, V. L. (1999). cis- and trans-acting elements involved in regulation of aniA, the gene encoding the major anaerobically induced outer membrane protein in Neisseria gonorrhoeae. J. Bacteriol. 181, 541–551. Kawai, M., Nakao, K., Uchiyama, I., and Kobayashi, I. (2006). How genomes rearrange: Genome comparison within bacteria Neisseria suggests roles for mobile elements in formation of complex genome polymorphisms. Gene 383, 52–63. Lambden, P. R., and Guest, J. R. (1976). Mutants of Escherichia coli K12 unable to use fumarate as an anaerobic electron acceptor. J. Gen. Microbiol. 97, 145–160. Lissenden, S., Mohan, S., Overton, T., Regan, T., Crooke, H., Cardinale, J. A., Householder, T. C., Adams, P., O’Conner, C. D., Clark, V. L., Smith, H., and Cole, J. A. (2000). Identification of transcription activators that regulate gonococcal adaptation from aerobic to anaerobic or oxygen-limited growth. Mol. Microbiol. 37, 839–855. Lundberg, J. O., and Weitzberg, E. (1999). Nasal nitric oxide in man. Thorax 54, 947–952. Lundberg, J. O., Weitzberg, E., Cole, J. A., and Benjamin, N. (2004). Nitrate, bacteria and human health. Nat. Rev. Microbiol. 2, 593–602.
560
Melanie J. Thomson et al.
Marriott, H. M., Ali, F., Read, R. C., Mitchell, T. J., Whyte, M. K., and Dockrell, D. H. (2004). Nitric oxide levels regulate macrophage commitment to apoptosis or necrosis during pneumococcal infection. FASEB J. 18, 1126–1128. Overton, T. W., Whitehead, R., Li, Y., Snyder, L. A., Saunders, N. J., Smith, H., and Cole, J. A. (2006). Coordinated regulation of the Neisseria gonorrhoeae-truncated denitrification pathway by the nitric oxide-sensitive repressor, NsrR, and nitrite-insensitive NarQ-NarP. J. Biol. Chem. 281, 33115–33126. Rock, J. D., Mahnane, M. R., Anjum, M. F., Shaw, J. G., Read, R. C., and Moir, J. W. (2005). The pathogen Neisseria meningitidis requires oxygen, but supplements growth by denitrification: Nitrite, nitric oxide and oxygen control respiratory flux at genetic and metabolic levels. Mol. Microbiol. 58, 800–809. Rock, J. D., Thomson, M. J., Read, R. C., and Moir, J. W. (2007). Regulation of denitrification genes in Neisseria meningitidis by nitric oxide and the repressor NsrR. J. Bacteriol. 189, 1138–1144. Stevanin, T. M., Moir, J. W., and Read, R. C. (2005). Nitric oxide detoxification systems enhance survival of Neisseria meningitidis in human macrophages and in nasopharyngeal mucosa. Infect. Immun. 73, 3322–3329. Stevanin, T. M., Poole, R. K., Demoncheaux, E. A., and Read, R. C. (2002). Flavohemoglobin Hmp protects Salmonella enterica serovar typhimurium from nitric oxide-related killing by human macrophages. Infect. Immun. 70, 4399–4405. Townsend, R., Goodwin, L., Stevanin, T. M., Silcocks, P. B., Parker, A., Maiden, M. C., and Read, R. C. (2002). Invasion by Neisseria meningitidis varies widely between clones and among nasopharyngeal mucosae derived from adult human hosts. Microbiology 148, 1467–1474. Wang, P. G., Xian, M., Tang, X., Wu, X., Wen, Z., Cai, T., and Janczuk, A. J. (2002). Nitric oxide donors: Chemical activities and biological applications. Chem. Rev. 102, 1091–1134. Weinberg, J. B., Misukonis, M. A., Shami, P. J., Mason, S. N., Sauls, D. L., Dittman, W. A., Wood, E. R., Smith, G. K., McDonald, B., Bachus, K. E., Hanley, A. F., and Granager, D. L. (1995). Human mononuclear phagocyte inducible nitric oxide synthase (iNOS): Analysis of iNOS mRNA, iNOS protein, biopterin, and nitric oxide production by blood monocytes and peritoneal macrophages. Blood 86, 1184–1195.
C H A P T E R
T W E N T Y- E I G H T
Localization of S-Nitrosothiols and Assay of Nitric Oxide Synthase and S-Nitrosoglutathione Reductase Activity in Plants Francisco J. Corpas,*,1 Alfonso Carreras,† Francisco J. Esteban,† Mounira Chaki,† Raquel Valderrama,† Luis A. del Rı´o,* and Juan B. Barroso† Contents 1. Introduction 2. Determination of L-Arginine-Dependent NOS Activity by Ozone Chemiluminescence in Plant Tissues 2.1. Principles and general considerations 2.2. Methods 3. Assay of GSNOR Activity 3.1. Spectrophotometric assay of GSNOR 3.2. Nondenaturing gel electrophoresis and staining for GSNOR activity 4. Localization of S-Nitrosothiols and S-Nitrosoglutathione in Plant Tissues by Confocal Laser-Scanning Microscopy (CLSM) 4.1. Overview 4.2. Detection of RSNOs with a fluorescent indicator and visualization by CLSM 4.3. Procedure 4.4. Immunolocalization of S-nitrosoglutathione in leaves by CLSM 4.5. Procedure 5. Conclusion Acknowledgments References
562 563 563 564 566 566 567
567 567 568 569 569 571 571 572 572
Departamento de Bioquimica, Biologı´a Celular y Molecular de plantas, Estacio´n Experimental del Zaidı´n (EEZ) CSIC, Apartado 419, E-18080 Grannada, Spain { Grupo de Sen ˜ alizacio´n Molecular y Sistemas Antioxidantes en Plantas, Unidad Asociada al CSIC (EEZ), ´ rea de Bioquı´mica y Biologı´a Molecular, Universidad de Jae´n, E-23071 Jae´n, Spain A 1 Corresponding author; E-mail: [email protected] *
Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37028-6
#
2008 Elsevier Inc. All rights reserved.
561
562
Francisco J. Corpas et al.
Abstract The study of the metabolism of nitric oxide and other reactive nitrogen species (RNS) in plants has been the subject of intensive work in the last decade due to the relevance of these molecules in the physiology and biochemistry of plants. However, there are still many methodological limitations in the specific and accurate determination and localization of RNS. This chapter describes several biochemical and cellular methods demonstrated to be useful for this purpose in different plant tissues. These methods are the determination of L-argininedependent nitric oxide synthase and S-nitrosoglutathione reductase activities, as well as cellular localization by confocal laser-scanning microscopy of S-nitrosothiols, particularly S-nitrosoglutathione. These approaches can help advance the knowledge of the function of RNS in plant cells.
1. Introduction The free radical nitric oxide (NO) has been recognized as a signal molecule in many important physiological processes in higher plants (del Rı´o et al., 2004; Lamattina et al., 2003; Shapiro 2005). NO has a role in plant growth and development (Corpas et al., 2006; Ribeiro et al., 1999), mainly including seed germination (Bethke et al., 2006), primary and lateral root growth (Correa-Aragunde et al., 2004), flowering and pollen tube growth regulation (Prado et al., 2004), fruit ripening (Leshem et al., 1998), senescence (Corpas et al 2004), and defense response and abiotic stress (Durner et al., 1998; Valderrama et al., 2007), and is also a key signaling molecule in different intracellular processes (Neil et al., 2003). In plant biology, knowledge of the physiological functions of NO and other reactive nitrogen species (RNS) has experienced a significant advance in recent years, mainly as a consequence of the use of pharmacological approaches. However, there is still little information on basic questions such as the enzymatic source of NO in a determined physiological process, and also the metabolism of different RNS, such as S-nitrosoglutathione (GSNO), under natural and stress conditions. As enzymatic sources of NO in plants, nitric oxide synthase (NOS), nitrate reductase, and other proteins have been proposed (Corpas et al., 2006; del Rı´o et al., 2004). However, the glutathione-dependent enzyme formaldehyde dehydrogenase (FALDH; EC 1.2.1.1), also known as class III alcohol dehydrogenase (ADH3), has been demonstrated to have GSNO reductase activity in bacteria, yeast, and mammals (Liu et al., 2001) and to be involved in the mechanism of protein S-nitrosation in mammalian cells (Haqqani et al., 2003). The enzyme GSNO reductase (GSNOR) catalyzes the NADHdependent reduction of GSNO to GSSG and NH3, and its presence has been demonstrated in several plant species ( Barroso et al., 2006; Dı´az et al.,
563
Localization of S-Nitrosothiols
2003; Sakamoto et al., 2002; for review, see Corpas et al., 2007). Therefore, GSNOR is able to regulate the level of GSNO and indirectly the NO content and its availability. This chapter describes methods for the determination in plant tissues of enzymatic activities of NOS and GSNOR, as well as the cellular localization of S-nitrosothiols and GSNO. These methods can help improve our knowledge of the function of NO and other RNS in the physiology and biochemistry of plants.
2. Determination of L-Arginine-Dependent NOS Activity by Ozone Chemiluminescence in Plant Tissues In animal cells, the enzyme NOS (EC 1.14.13.39) is well characterized. NOS catalyzes the oxygen- and NADPH-dependent oxidation of L-arginine to NO and citrulline in a reaction requiring flavin adenosine diphosphate (FAD), flavin mononucleotide (FMN), tetrahydrobiopterin (BH4), Ca2þ, and calmodulin (CaM) (Alderton et al., 2001). In plants, the enzymatic generation of NO using L-arginine as substrate and all the animal NOS cofactors has also been described, although the gene of the plant NOS has not been identified yet (Barroso et al., 1999; Corpas et al., 2004, 2006; del Rı´o et al., 2004). However, it must be indicated that NO in plants can also be produced by nonenzymatic systems and by other enzymes apart from NOS (del Rı´o et al., 2004).
2.1. Principles and general considerations Nitric oxide has a relatively short half-life (Ignarro, 1990) and, in water solutions, reacts spontaneously with O2 with a stoichiometry of 4:1 [Eq. (28.1)] and a rate constant of 6.3 106 M-2 s1 (Ford et al., 1993): þ 4NO þ O2 þ 2H2 O ! 4NO 2 þ 4H
ð28:1Þ
On the basis of this reaction, it is possible to detect the enzymatic generation of NO by ozone chemiluminescence. First, the NO-derived nitrite generated by the samples is reduced back to NO by vanadium trichloride (VCl3) in 1 M HCl at 95 C and a nitrogen stream bubbling through the purge vessel carries the resulting NO gas to the measurement cell. This NO in the gas phase reacts with ozone, according to Eqs. (28.2) and (28.3),
NO þ O3 ! NO2 þ O2
ð28:2Þ
564
Francisco J. Corpas et al.
Table 28.1 Activities of L-arginine-dependent nitric oxide synthases (NOSs) from animal and plant tissues determined by ozone chemiluminescence
Species/tissue or cell type
Specific activity (pmolmin-1mg-1 protein)
Animals Mus musculus (mouse) Purified iNOS from 395 103 macrophage Rattus norvegicus (rat) Purified nNOS 200 103 Aristichthys nobilis (carp) Brain 133 Plants Pisum sativum (pea) Leaf peroxisomes 5.0 103 Root 240 Stem 630 Leaf 120 Helianthus annuus (sunflower) Hypocotyls 294 Olea europaea (olive) Leaf 280
NO2 ! NO2 þ hn
References
Maurer and Fung, (2000)
Corpas et al. (2006) Wong et al. 1998
Corpas et al. (2004) Corpas et al. (2006)
Chaki et al. (2007) Valderrama et al. (2007)
ð28:3Þ
where NO2* denotes the NO2 molecule in the excited state and hn represents an emitted photon. Chemiluminescence resulting from these reactions is detected by a photomultiplier tube. This method has a sensitivity of 1 pmol (1 nM for a 1-ml injection) and can be used to quantify the NO generated in the course of the L-argininedependent NOS activity of different plant samples (Table 28.1).
2.2. Methods Fresh plant tissue (1 g fresh weight) is frozen in liquid N2 and ground to a powder in a mortar with a pestle. The powder is suspended in 3 ml of homogenizing medium containing 40 mM HEPES buffer, pH 7.2, 0.2 mM CHAPS, 10 mM FAD, 10 mM FMN, 10 mM BH4, 10 mg/ml CaM, and 1.25 mM CaCl2 (1:4; w/v). It must be mentioned that this buffer can be replaced by other buffers according to the requirement of plant samples (Valderrama et al., 2007). Homogenates are filtered through one layer of
Localization of S-Nitrosothiols
565
Miracloth (Calbiochem) and centrifuged at 36,000 g for 20 min. Then, the supernatants are passed through Sephadex G-25 desalting columns (NP-10 from Amersham) to remove possible nitrites and nitrates present in the samples. The columns are equilibrated and eluted with 40 mM HEPES buffer, pH 7.2, containing 0.2 mM CHAPS, 15 mM dithiothreitol, 1.25 mM CaCl2, 1 mM b-NADPH, 5 mg/ml calmodulin, 10 mM FAD, 10 mM FMN, and 10 mM BH4. The NOS activity is performed in duplicate, for each sample, in a reaction medium containing 40 mM HEPES buffer, pH 7.2, 0.2 mM CHAPS, 10 mg/ml calmodulin, 1.25 mM CaCl2, 1 mM b-NADPH, 10 mM FAD, 10 mM FMN, 10 mM BH4, and 1 mM L-Arg in a final volume of 1 ml. The reaction mixture is incubated at 37 C for 30 min. The NO produced in the enzymatic reaction is quickly oxidized to its stable end product nitrite. At 0 time and 30 min in, 200 ml of the reaction mixture is deproteinated by the addition of 100 ml 0.8 N NaOH and 100 ml 16% ZnSO4, and the mixture is shaken vigorously for 30 s and centrifuged at 16,000g for 10 min. Aliquots of supernatants (40 ml) are injected into the purge chamber of a nitric oxide analyzer (NOA; Model 280i from Sievers Instruments, Boulder, CO) containing 5 ml of 50 mM vanadium trichloride (VCl3) in 1 M HCl at 95 C under a nitrogen stream. Under these conditions, the NO-derived nitrite is reduced back to gaseous NO, which is mixed with ozone, generated in the NOA, and the light emitted from the chemiluminescence reaction between NO and ozone is quantified by a photomultiplier tube. The amount of NO generated is computed by referring the magnitude of these signals to those generated by a nitrite standard curve between 1 nM and 100 mM prepared from a standard stock solution of 100 mM nitrite. Figure 28.1 shows typical peaks obtained after the injection of several plant deproteinated samples at 0 time and 30 min. Thus, the production of NO in the 30-min reaction is calculated by subtracting the blank value (zero time), which represents the nonenzymatic NO production, and the activity is expressed as pmol NO mg-1 protein min-1. The NOS activity determined in different plant and animal tissues using this method is shown in Table 28.1. As negative controls, aliquots of the samples are preheated at 95 C for 10 min prior to the enzymatic reaction. As a positive control of this method of determination of L-arginine-dependent NOS activity in plant samples, a commercial rat neuronal NOS from Calbiochem (2.9 U) is assayed separately. The NOS activity can also be determined using a concentration of 100 mM L-Arg and in the presence of 1 mM aminoguanidine, which is an irreversible inhibitor of both constitutive and inducible NOS activity in animal cells (Laszlo et al., 1995). Use of the NOS inhibitors analogous to L-arginine, N G-nitro-L-arginine methyl ester and N G-monomethyl-L-arginine, is not recommended because they cause interference with the method.
566
Francisco J. Corpas et al.
mV 200.0 190.0 180.0 170.0 160.0 150.0 140.0 130.0 120.0 110.0 100.0 90.0 80.0 70.0 60.0 50.0 40.0 30.0 20.0 10.0
0
30
0
30
Reaction time (min)
Figure 28.1 Output signals (mV) obtained in the analysis of NO with the nitric oxide analyzer. Every peak corresponds to independent injections at 0 time and at 30 min after the NOS reaction.The difference between both peaks (30 and 0 min) corresponds to the NO generated in the reaction time (30 min) by the L^arginine^dependent NOS activity of samples.
It must be mentioned that, to our knowledge, there are no reports on the presence of BH4 in plants. This cofactor promotes and/or stabilizes the active dimeric form of the three mammal NOS isoforms (Alderton et al., 2001). In plants, this function perhaps could be carried out by tetrahydrofolate (FH4), whose biosynthesis and distribution are well known in higher plants (Sahr et al., 2005). At present, the function of FH4 as a cofactor of the L-arginine-dependent NOS activity in plants is under study.
3. Assay of GSNOR Activity 3.1. Spectrophotometric assay of GSNOR GSNOR activity is assayed spectrophotometrically at 25 C by monitoring the oxidation of NADH at 340 nm (Barroso et al., 2006; Sakamoto et al., 2002). Table 28.2 shows components of the reaction mixture. Plant samples are passed through Sephadex G-25 desalting columns (NP-10 from Amersham) to remove possible endogenous GSNO. Then, samples are incubated in an assay mixture containing 20 mM Tris-HCl, pH
567
Localization of S-Nitrosothiols
Table 28.2 Components of the reaction mixture Component
Volume ( ml)
20 mM Tris-HCl buffer, pH 8.0, 0.5 mM EDTA Plant sample 0.2 mM NADH 0.4 mM GSNO
650 to 750
Final volume
1000 ml
50 to 150 100 100
8.0, 0.2 mM NADH, and 0.5 mM EDTA, and the reaction is started by adding GSNO (Calbiochem) at a final concentration of 0.4 mM. Fresh solutions of GSNO are prepared before use (they must have a pink color) and are kept protected from light. The activity is expressed as nmol NADH consumed per min and mg protein (e340 ¼ 6.22 mM-1cm-1)
3.2. Nondenaturing gel electrophoresis and staining for GSNOR activity Native polyacrylamide gel electrophoresis is performed using 6% acrylamide gels in Tris-boric-EDTA (8.9 mM Tris base, 8.9 mM boric acid, 0.2 mM Na2EDTA) buffer, pH 8, as described by Laemmli (1970). Staining for GSNOR activity is carried out using a modification of the method reported by Seymour and Lazarus (1989) and Ferna´ndez et al (2003). Gels are soaked in 0.1 M sodium phosphate, pH 7.4, containing 2 mM NADH, for 15 min, in an ice bath. Excess buffer is drained, and gels are covered with filter paper strips soaked in freshly prepared 3 mM GSNO. After 10-15 min, the filter paper is removed, and bands can be visualized when gels are exposed to ultraviolet light. The presence of GSNOR activity is shown by positive bands corresponding to the disappearance of NADH fluorescence. Quantification of the bands can be performed using a UV Gel Documentation system coupled with a highly sensitive charge-coupled device camera. An example of the analysis of GSNOR activity by native polyacrylamide gel electrophoresis in leaf extracts from pea plants is shown in Figure 28.2
4. Localization of S-Nitrosothiols and S-Nitrosoglutathione in Plant Tissues by Confocal Laser-Scanning Microscopy (CLSM) 4.1. Overview S-Nitrosothiols (RSNOs) in general and S-nitrosoglutathione in particular play important roles in the physiological chemistry of NO. In animal cells, RSNOs have been suggested to participate in the transport, storage, and
568
Francisco J. Corpas et al.
Native-PAGE
C
Cd
Figure 28.2 Native polyacrylamide gel electrophoresis of pea leaf extracts and staining for GSNOR activity. Pea plants were grown under greenhouse conditions without cadmium (C) and with 50 mM cadmium chloride (Cd). Leaf extracts (200 mg protein) were subjected to electrophoresis on 6% acrylamide gels, and then gels were stained for GSNOR activity. Taken from Barroso et al. (2006).
delivery of NO and in posttranslational modifications in cell signaling under physiological and pathological conditions (Foster et al., 2003). However, much less is known about RSNOs in plants, although some evidence indicates that they may play a similar function in plant cells (Barroso et al., 2006; Feechan et al., 2005; Valderrama et al., 2007). The different procedures used to quantify the content of S-nitrosothiols have been reviewed elsewhere (Jourd’heuil et al. 2005), and a complementary technique to localize the presence and distribution of RSNOs in plant tissues is presented here. R1
Hg+ + RSNO
Hg-link phenylmercury
R1
HgSR Thiolate
ð28:4Þ
4.2. Detection of RSNOs with a fluorescent indicator and visualization by CLSM Alexa Fluor Hg-Link dye is a new generation of fluorescent probes that exhibit some advantages over other fluorescent dyes, such as being highly fluorescent over a broad pH range (pH 4 to 10), good water solubility, and higher photostability, allowing more time for image capture. Specifically, the Alexa Fluor 488 Hg-Link phenyl mercury (Molecular Probes) can be
Localization of S-Nitrosothiols
569
used for direct RSNO detection in tissues when samples are preincubated with N-ethylmaleimide (NEM) to block free sulfhydryl groups. Under these conditions, the fluorescent Hg-Link reagent labels RSNO-modified proteins and peptides according to Eq. (28.4) [technical brochure of Molecular Probes]:
4.3. Procedure Leaf segments of approximately 25 mm2 are incubated at 25 C for 2 h, in darkness, with 10 mM NEM prepared in ethanol, which blocks free sulfhydryl groups. Then, they are washed three times in 10 mM Tris-HCl buffer, pH 7.4, for 15 min each and with 10 mM b-mercaptoethanol for 30 min, followed by three washes with the same buffer for 15 min each. After that, samples are incubated with 10 mM Alexa Fluor 488 Hg-Link phenyl mercury for 1 h at 25 , in darkness. After washing three times in the previous buffer, leaf sections are embedded in order to obtain appropriate sections to be observed under confocal microscopy (Germroth et al., 1995). Samples are soaked in 15% acrylamide [in 0.01 M phosphate-buffered saline (PBS) including 0.3% TEMED] for 4 h and are then polymerized in inclusion containers (1.5 0.9 0.5 mm; Sorvall Instruments) using 0.5 ml of a fresh 15% acrylamide stock solution and 50 ml of 2% persulfate ammonium (PSA). Samples must be quickly oriented after PSA addition and, to improve polymerization, the containers are covered with Parafilm. Sections 80 to 100 mm thick are cut, as indicated by the Vibratome scale, under 10 mM PBS. Sections are then soaked in glycerol:PBS (containing azide) (1:1, v/v) and mounted in the same medium for examination with a CLSM system (Leica TCS SL) using standard filters for Alexa Fluor 488 green fluorescence (excitation 495 nm; emission 519 nm). Usually, three background staining controls are used: (i) sections incubated with NEM plus b-mercaptoethanol without Alexa Fluor 488 Hg-Link phenyl mercury; (ii) sections with b-mercaptoethanol plus Alexa Fluor 488 without NEM; and (iii) sections with only b-mercaptoethanol. All procedures must be performed under a red safety light. Figure 28.3A shows the cellular localization of RSNOs in an olive leaf section where green fluorescence corresponds to the presence of RSNOs. Figure 28.3B shows the appearance of a negative control where the olive leaf section was incubated without both NEM and Alexa Fluor 488.
4.4. Immunolocalization of S-nitrosoglutathione in leaves by CLSM The thiol tripeptide g-glutamyl-cysteinyl-glycine (glutathione, GSH) is a major low molecular weight soluble antioxidant of plant cells. GSH is involved in the antioxidative ascorbate-glutathione cycle and it also acts as
570
Francisco J. Corpas et al.
A
B
RSNO
X
P
Pa 200 µm
200 µm C
GSNO
D
GSNO
X Pa Co
100 µm
Ec
80 µm
Figure 28.3 Representative images illustrating the cellular localization of S^nitrosothiols (RSNOs) and S^nitrosoglutathione (GSNO) in leaves of higher plants by fluorescence microscopy and CLSM. (A) Detection of RSNOs by fluorescence microscopy using the fluorescent Alexa Fluor 488 Hg^Link reagent in olive leaf sections. The bright green fluorescence corresponds to RSNOs. The orange^yellow color is due to the chlorophyll autofluorescence. Each picture was prepared from 30 to 40 cross sections of olive leaves analyzed by CLSM. (B) Appearance of an olive leaf section incubated without both 10 mM N^ethylmaleimide (NEM) and Alexa Fluor (negative control for RSNOs). (C) Immunolocalization of GSNO in a pea leaf section by fluorescence microscopy using anti^GSNO and Cy2^conjugated antibody. Intense green anti^GSNO immunofluorescence is localized mainly in collenchyma cells. (D) Immunolocalization of GSNO in an olive leaf section by CLSM. The intense green immunofluorescence, attributable to anti^GSNO, is localized mainly in vascular tissue and spongy mesophyll cells. Xylem (X), phloem (P), collenchyma (Co), parenchyma (Pa), and epidermal cuticle (Ec). A, B, and D are fromValderrama etal. (2007) and C is from Barroso et al. (2006).
an independent redox-signaling molecule (Foyer, 2001; Foyer and Noctor, 2005). In the presence of O2, GSH can react with NO to form GSNO. This nitrosothiol is quite stable in the dark and in water solutions in the presence of a metal ion chelator (Smith and Dasgupta, 2000). At 37 C and pH 7.4, GSNO decays with a second-order rate constant of 3 10-4 M-1 s-1 (Park et al., 1993). In plants, it has been proposed that GSNO can be a natural mobile reservoir of NO bioactivity (Wang et al., 2006) and is involved in protein S-nitrosylation (Lindemayr et al., 2005).
Localization of S-Nitrosothiols
571
4.5. Procedure Leaf sections are cut into 4- to 5-mm pieces and fixed in 4% (w/v) paraformaldehyde in 0.1 M phosphate buffer, pH 7.4 (PB), for 3 h at room temperature. They are then cryoprotected by immersion in 30% (w/v) sucrose in PB overnight at 4 C. Serial sections, 60 mm thick, are obtained by means of a cryostat (2800 Frigocut E, Reichert-Jung, Vienna, Austria). Free-floating sections are incubated at room temperature overnight with a commercial rat antibody against S-nitrosoglutathione (Calbiochem or A.G. Scientific, Inc.) diluted 1:2500 in 5 mM Tris buffer, pH 7.2, containing 0.9% (w/v) NaCl, 0.05% (w/v) sodium azide, 0.1% (w/v) bovine serum albumin (BSA), and 0.1% (v/v) Triton X-100 (TBSABSAT). After several washes with TBSA-BSAT, sections are incubated with biotinylated goat antirat IgG (Amersham, Buckinghamshire, UK), diluted 1:1000 in TBSA-BSAT, for 1 h at room temperature. Sections are then washed again and incubated with Cy2-streptavidin (Amersham) and diluted 1:1000 in TBSA-BSAT at room temperature for 1.5 h. Sections are mounted on glass slides with PBS-glycerol (v:v; 1:1), covered with a coverslip, and observed using either a fluorescence microscope or a confocal laser-scanning microscope with standard filters for Cy2 (excitation, 495 nm; emission, 515 nm). Controls for background staining, usually negligible, are performed by replacing the corresponding primary antiserum by preimmune serum in adjacent sections or just without primary antiserum. The immunolocalization of GSNO in a pea leaf section, observed in a fluorescence microscope, is shown in Figure 28.3C. A strong green GSNOderived immunofluorescence is detected mainly in collenchyma cells (Co) of pea leaves. In this case, yellowish autofluorescence appears in xylem, parenchymal, and epidermal cuticle cells. However, in olive leaves observed by CLSM, the green fluorescence attributed to GSNO is localized in vascular tissues and spongy mesophyll cells (see Figure 28.3D), and the orange-yellowish color corresponds to autofluorescence.
5. Conclusion This chapter presented several procedures, the assay of L-argininedependent NOS and GSNOR activities and the cellular distribution of RSNO and GSNO, that have been applied successfully to different organs, including roots, stems, leaves, and hypocotyls of several plant species. The methods outlined here can be useful in obtaining deeper insights into the metabolism of NO in higher plants but can also be applied to other organisms.
572
Francisco J. Corpas et al.
ACKNOWLEDGMENTS This work was supported by grants from the Ministry of Education and Science (BIO200614949-C02-01 and BIO2006-14949-C02-02) and Junta de Andalucı´a (groups CVI 0192, CVI 0286, and CVI 302), Spain.
REFERENCES Alderton, W. K., Cooper, C. E., and Knowles, R. G. (2001). Nitric oxide synthases: Structure, function and inhibition. Biochem. J. 357, 593–615. Barroso, J. B., Corpas, F. J., Carreras, A., Rodrı´guez-Serrano, M., Esteban, F. J., Ferna´ndezOcan˜a, A., Chaki, M., Romero-Puertas, M. C., Valderrama, R., Sandalio, L. M., and del Rı´o, L. A. (2006). Localization of S-nitrosoglutathione and expression of S-nitrosoglutathione reductase in pea plants under cadmium stress. J. Exp. Bot. 57, 1785–1793. Barroso, J. B., Corpas, F. J., Carreras, A., Sandalio, L. M., Valderrama, R., Palma, J. M., Lupia´n˜ez, J. A., and del Rı´o, L. A. (1999). Localization of nitric-oxide synthase in plant peroxisomes. J. Biol. Chem. 274, 36729–36733. Bethke, P. C., Libourel, I. G., and Jones, R. L. (2006). Nitric oxide reduces seed dormancy in Arabidopsis. J. Exp. Bot. 57, 517–526. Chaki, M. (2007). In ‘‘Function of Reactive Nitrogen Species in Sunflower (Helianthus annuus) in Response to Abiotic and Biotic Stresses.’’ Spain: Ph.D. thesis University of Jae´n, Spain. Corpas, F. J., Barroso, J. B., Carreras, A., Quiro´s, M., Leo´n, A. M., Romero-Puertas, M. C., Esteban, F. J., Valderrama, R., Palma, J. M., Sandalio, L. M., Go´mez, M., and del Rı´o, L. A. (2004). Cellular and subcellular localization of endogenous nitric oxide in young and senescent pea plants. Plant Physiol. 136, 2722–2733. Corpas, F. J., Barroso, J. B., Carreras, A., Valderrama, R., Palma, J. M., and del Rı´o, L. A. (2007). Nitrosative stress in plants: A new approach to understand the role of NO in abiotic stress. In ‘‘Nitric Oxide in Plant Growth, Development and Stress Physiology’’ (L. Lamattina and J. C. Polacco, eds.), pp. 187–205. Springer, New York. Corpas, F. J., Barroso, J. B., Carreras, A., Valderrama, R., Palma, J. M., Leo´n, A. M., Sandalio, L. M., and del Rı´o, L. A. (2006). Constitutive arginine-dependent nitric oxide synthase activity in different organs of pea seedlings during plant development. Planta 224, 246–254. Correa-Aragunde, N., Graziano, M., and Lamattina, L. (2004). Nitric oxide plays a central role in determining lateral root development in tomato. Planta 218, 900–905. del Rı´o, L. A., Corpas, F. J., and Barroso, J. B. (2004). Nitric oxide and nitric oxide synthase activity in plants. Phytochemistry 65, 783–792. Dı´az, M, Achkor, H., Titarenko, E., and Martı´nez, M. C. (2003). The gene encoding glutathione-dependent formaldehyde dehydrogenase/GSNO reductase is responsive to wounding, jasmonic acid and salicylic acid. FEBS Lett. 543, 136–139. Durner, J., Wendehenne, D., and Klessing, D. F. (1998). Defense gene induction in tobacco by nitric oxide, cyclic GMP, and cyclic ADP-ridobse. Proc. Natl. Acad. Sci. USA 95, 10328–10333. Feechan, A., Kwon, E., Yun, B. W., Wang, Y., Pallas, J. A., and Loake, G. J. (2005). A central role for S-nitrosothiols in plant disease resistance. Proc. Nat. Acad. Sci. USA 102, 8054–8059. Ferna´ndez, M. R., Biosca, J. A., and Pares, X. (2003). S-nitrosoglutathione reductase activity of human and yeast glutathione-dependent formaldehyde dehydrogenase and its nuclear and cytoplasmic localisation. Cell. Mol. Life Sci. 60, 1013–1018.
Localization of S-Nitrosothiols
573
Ford, P. C., Wink, D. A., and Stanbury, D. M. (1993). Autoxidation kinetics of aqueous nitric oxide. FEBS Lett. 326, 1–3. Foster, M. W., McMahon, T. J., and Stamler, J. S. (2003). S-nitrosylation in health and disease. Trends Mol. Med. 9, 160–168. Foyer, C. H. (2001). Prospects for enhancement of the soluble antioxidants, ascorbate and glutathione. Biofactors 15, 75–78. Foyer, C. H., and Noctor, G. (2005). Oxidant and antioxidant signalling in plants: A re-evaluation of the concept of oxidative stress in a physiological context. Plant Cell Environ. 28, 1056–1071. Germroth, P. G., Gourdie, R. G., and Thompson, R. P. (1995). Confocal microscopy of thick sections from acrylamide gel embedded embryos. Microsc. Res. Tech. 30, 513–520. Haqqani, A. S., Do, S. K., and Birnboim, H. C. (2003). The role of a formaldehyde dehydrogenase-glutathione pathway in protein S-nitrosation in mammalian cells. Nitric Oxide 9, 172–181. Ignarro, L. J. (1990). Biosynthesis and metabolism of endothelium-derived nitric oxide. Annu. Rev. Pharmacol. Toxicol. 30, 535–560. Jourd’heuil, D., Jourd’heuil, F. L., Lowery, A. M., Hughes, J., and Grisham, M. B. (2005). Detection of nitrosothiols and other nitroso species in vitro and in cells. Methods Enzymol. 396, 118–131. Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685. Lamattina, L., Garcı´a-Mata, C., Graziano, M., and Pagnussat, G. (2003). Nitric oxide: The versatility of an extensive signal molecule. Annu. Rev. Plant Biol. 54, 109–136. Laszlo, F., Evans, S. M., and Whittle, B. J. (1995). Aminoguanidine inhibits both constitutive and inducible nitric oxide synthase isoforms in rat intestinal microvasculature in vivo. Eur. J. Pharmacol. 272, 169–175. Leshem, Y. Y., Wills, R. B. H., and Veng-Va, K. V. (1998). Evidence for the function of the free radical gas-nitric oxide (NO) as an endogenous maturation and senescence regulating factor in higher plants. Plant Physiol. Biochem. 36, 825–833. Lindermayr, C., Saalbach, G., and Durner, J. (2005). Proteomic identification of S–nitrosylated proteins in Arabidopsis. Plant Physiol. 137, 921–930. Liu, L., Hausladen, A., Zeng, M., Que, L., Heitman, J., and Stamler, J. S. (2001). A metabolic enzyme for S–nitrosothiol conserved from bacteria to humans. Nature 410, 490–494. Maurer, T. S., and Fung, H. L. (2000). Evaluation of nitric oxide synthase activity and inhibition kinetics by chemiluminescence. Nitric Oxide 4, 372–378. Neill, S. J., Desikan, R., Clarke, A., and Hancock, J. T. (2003). Nitric oxide signalling in plants. New Phytol. 159, 11–35. Park, J. W., Billman, G. E., and Means, G. E. (1993). Transnitrosation as a predominant mechanism in the hypotensive effect of S–nitrosoglutathione. Biochem. Mol. Biol. Int. 30, 885–889. Prado, A. M., Porterfield, D. M., and Feijo, J. A. (2004). Nitric oxide is involved in growth regulation and re–orientation of pollen tubes. Development 131, 2707–2714. Ribeiro, E. A., Cunha, F. Q., Tamashiro, W. M., and Martins, I. S. (1999). Growth phase–dependent subcellular localization of nitric oxide synthase in maize cells. FEBS Lett. 445, 283–286. Sahr, T., Ravanel, S., and Rebeille, F. (2005). Tetrahydrofolate biosynthesis and distribution in higher plants. Biochem. Soc. Trans. 33, 758–762. Sakamoto, A., Ueda, M., and Morikawa, H. (2002). Arabidopsis glutathione–dependent formaldehyde dehydrogenase is an S–nitrosoglutathione reductase. FEBS Lett. 515, 20–24.
574
Francisco J. Corpas et al.
Seymour, J. L., and Lazarus, R. A. (1989). Native gel activity stain and preparative electrophoretic method for the detection and purification of pyridine nucleotide–linked dehydrogenases. Anal. Biochem. 178, 243–247. Shapiro, A. D. (2005). Nitric oxide signaling in plants. Vitam. Horm. 72, 339–398. Smith, J. N., and Dasgupta, T. P. (2000). Kinetics and mechanism of the decomposition of S–nitrosoglutathione by L–ascorbic acid and copper ions in aqueous solution to produce nitric oxide. Nitric Oxide 4, 57–66. Valderrama, R., Corpas, F. J., Carreras, A., Luque, F., Ferna´ndez–Ocan˜a, A., Chaki, M., Go´mez–Rodrı´guez, M. V., del Rı´o, L. A., and Barroso, J. B. (2007). Nitrosative stress in plants. FEBS Lett. 581, 453–461. Wang, Y., Yun, B. W., Kwon, E., Hong, J. K., Yoon, J., and Loake, G. J. (2006). S–nitrosylation: An emerging redox–based post–translational modification in plants. J. Exp. Bot. 57, 1777–1784. Wong, H. Y., Fung, L. Y., Kwok, F., and Lo, S. C. L. (1998). Constitutive nitric oxide synthase (NOS) activities in big–head carp (Aristichthys nobilis). Fish Physiol. Biochem. 19, 171–179.
C H A P T E R
T W E N T Y- N I N E
Methods for Nitric Oxide Detection during Plant–Pathogen Interactions E. Vandelle and M. Delledonne Contents 1. Introduction 2. Nitric Oxide Detection by Mass Spectrometry 2.1. Procedure 2.2. Results 2.3. Comments 3. Nitric Oxide Detection by Laser Photoacoustic Spectroscopy 3.1. Procedure 3.2. Results 3.3. Comments 4. Nitric Oxide Detection by Chemiluminescence 4.1. Procedure 4.2. Results 4.3. Comments 5. Nitric Oxide Detection by Hemoglobin Conversion 5.1. Procedure 5.2. Results 5.3. Comments 6. Nitric Oxide Detection by Electron Paramagnetic Resonance (EPR) Spin Trap 6.1. Procedure 6.2. Results 6.3. Comments 7. Nitric Oxide Detection Using Diaminofluoresceins 7.1. Procedure 7.2. Results 7.3. Comments 8. Conclusion References
576 577 577 578 579 579 579 581 581 582 582 583 583 583 583 584 584 585 585 586 587 587 588 589 589 590 591
Dipartimento Scientifico e Tecnologico, Universita` degli Studi di Verona, Verona, Italy Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37029-8
#
2008 Elsevier Inc. All rights reserved.
575
576
E. Vandelle and M. Delledonne
Abstract Nitric oxide (NO) is involved in the transduction of numerous signals in living organisms, and its biological effects are often influenced by its concentration. Therefore, the ability to reliably detect and quantify NO is crucial to understanding its role in cellular processes. Many techniques are available to detect and quantify NO, but depending on the material and the aim of the analysis, specific adaptations are often required because its high chemical reactivity leads to the formation of numerous reactive nitrogen species that make the accurate determination of NO levels difficult. Moreover, the pathogen-induced hypersensitive response leads to high rates of reactive oxygen species production that react with NO and lead to the formation of its oxidized derivates. The aim of this chapter is to provide an overview of the methods that have so far been employed to detect and measure NO in plants during the hypersensitive disease resistance response.
1. Introduction Nitric oxide (NO) is a bioactive molecule that regulates an ever-growing list of biological processes in phylogenetically distant species (Beligni and Lamattina, 2001). In addition to acting as a potent endogenous vasodilator and having a role in inflammation, thrombosis, immunity, and neurotransmission in animals, NO is also involved in diverse physiological processes in plants under normal growth conditions (germination, leaf senescence, root growth) and under stress situations, such as pathogen attack (Wendehenne et al., 2004). As a modulator of plant disease resistance, it plays a central role in hypersensitive response (HR) establishment, characterized by rapid and localized cell death (Delledonne et al., 2001). The discovery of NO with these important physiological functions has led to the development of various analytical methods for its detection and quantification. However, the accurate measurement of NO production is somewhat tedious. Indeed, NO is a gaseous free radical that is extremely labile and its short half-life (6–10 s) reflects its highly reactive nature because of the presence of an unshared electron. Its broad chemistry involves an interplay among three species differing in their physical properties and chemical reactivity: the nitrosium cation (NOþ), the radical (NO), and the nitroxyl anion (NO) (Neill et al., 2003). Moreover, NO interacts rapidly with O2 to yield a variety of nitrogen oxides: NO2, N2O3, NO2, and NO 3 , collectively termed reactive nitrogen species (RNS). In addition, the amount of NO available is highly dependent on the redox state of the cell, which makes the detection of NO, and in particular in plants challenged by pathogens, difficult. Indeed, an early event in the HR is the generation of superoxide (O2) and accumulation of hydrogen peroxide
Nitric Oxide Detection Methods
577
(H2O2), referred to as the oxidative burst, which is necessary to trigger host cell death in cooperation with NO (Delledonne et al., 2001). During this process, the cell undergoes significant changes in its redox state that contribute to modification of the reactivity of NO. The diverse assays available to measure NO are based on its particular physical and chemical properties and have been first optimized on animal systems before adapting them for plant study. Some assays detect NO gas emitted from cells, as NO has a high capacity of diffusion across membranes, whereas others measure NO derivatives, such as N2O3, taking into account its high oxidative metabolism. Most methods have been already used to detect NO during plant–pathogen interactions. They are described herein and compared in terms of their accuracy and reliability in measuring NO in this context.
2. Nitric Oxide Detection by Mass Spectrometry This technique is based on the diffusion of dissolved gases through a capillary and their identification with a benchtop mass spectrometer according to their different mass/charge ratios (m/z). Changes in NO levels are evaluated by changes in the abundance of mass 30, corresponding to NO (m/z ¼ 30). This method, referred to as membrane inlet mass spectrometry (MIMS), was developed in animals using mammalian cell cultures, but has been subsequently adapted to study NO production in plant cell suspensions and combined with restriction capillary inlet mass spectrometry (RIMS) to detect NO released from intact plant leaves or small plants (Conrath et al., 2004). The instrument setup is presented in Fig. 29.1.
2.1. Procedure In MIMS, 10 ml of cell suspension is transferred into an aquatic chamber (10 ml volume) and maintained under continuous agitation using a magnetic stirrer. The dissolved gases released by cells diffuse through a semipermeable Teflon membrane (50 mm), which then evaporate into the ionization chamber of a benchtop mass spectrometer (Conrath et al., 2004). In RIMS, intact plant leaves or small plants are introduced in a leaf/plant cuvette placed in a translucent chamber, in which a metal bellows pump ensures rapid and efficient mixture of the 120-ml gas phase. The dissolved gases produced by the plant diffuse from the gas phase through a thin restriction capillary (inner diameter: 0.1 mm, length: 2 m) directly into a benchtop mass spectrometer (Conrath et al., 2004)
578
E. Vandelle and M. Delledonne
Laser gases 7
8 6
5
Air flow
4
9
Air
* 3
1
2
Figure 29.1 Experimental setup for CO LPAD of NO. Up to three samples [typically, a bacterially challenged leaf, a mock (water)-injected leaf, and an instrumental control (baseline)] were sealed in three glass cuvettes (1), which could be sampled by alternating electronic valves (3). Airflow (1.5 liter h^1) through the cuvettes was regulated by mass flow controllers (2).When not being sampled, the gas flow was vented from the cuvette to the atmosphere.Water vapor in the gas flow was removed using a Peltier cooling element (^5) and a cold trap (^80; 4), prior to passage into the photoacoustic cell (5).The photoacoustic cell was inserted into a laser cavity to improve laser power and, thus, detection sensitivity. This laser cavity consisted of a gas discharge tube (6), a rotatable grating (7; for laser line selection), and a 100% reflecting mirror (8).To generate a photoacoustic signal, the laser light was modulated by a chopper (9; modulation frequency 1000 Hz). An additional water scrubber made of CaCl2 (*) had no effect on the NO signal. Reproduced with permission from Mur et al. (2005).
2.2. Results The specificity of the NO signal obtained by MIMS has been estimated using NO donors (180 mM ), S-nitroso-N-acetyl-DL-penicillamine and S-nitroso-L-glutathione, and validated by the addition of the NO scavenger carboxy-2-phenyl-4,4,5,5-tetramethyl-imidazoline-1-oxyl-3-oxide (150 mM ). The NO signal can be calibrated using NO-saturated water. The aqueous sample chamber is filled with 10 ml of water and is then flushed with N2 for up to 5 min to purge the system of O2. At different time points, 5 ml of NO-saturated water (corresponding to a final NO concentration of 0.95 mM ) is added to the chamber. Under these conditions, Conrath et al. (2004) have estimated that 1 abundance unit at m/z ¼ 30 corresponds to
Nitric Oxide Detection Methods
579
10 to 13 pmol of NO. NO for signal quantification can also be generated by quantitatively reducing KNO2 with KI (Berkels et al., 2001). Using MIMS, Conrath et al. (2004) have evaluated the amount of NO released by tobacco cells elicited by avirulent Pseudomonas syringae pv. tomato. They detected two NO bursts: the first at 1 h reaches 0.2 nmol of NO and the second more intense, occurring 4–8 h postinfection, presents a maximum around 0.5 nmol of NO (Conrath et al., 2004).
2.3. Comments The MIMS/RIMS method allows direct, fast, specific, and noninvasive online detection of NO and other gases in cell suspensions, as well as in entire leaves or plants. Moreover, this technique can be combined with an isotope tracing experiment, allowing source identification of gaseous NO by plant cells, as the isotopomers 14NO and 15NO can be distinguished. However, this method is less sensitive than other techniques such as photoacoustic laser spectroscopy (see later) and only allows the detection of extracellular NO. A similar technique based on mass spectrometry has been used by Bethke et al. (2004) to analyze apoplastic NO production by plant tissues via the nonenzymatic reduction of nitrite. These authors describe a continuous sampling method where NO (and other gases) diffuses through a polyethylene membrane directly to the vacuum of the mass spectrometer.
3. Nitric Oxide Detection by Laser Photoacoustic Spectroscopy Laser photoacoustic detection (LPAD) is another direct-trace gas, noninvasive online sampling technique for measuring NO. This method is based on the photoacoustic effect, i.e., acoustic wave generation as a consequence of light absorption. Discontinuous laser illumination of a gas leads to temperature variations accompanied by pressure variations that create an audible sound detectable by a sensitive microphone (Mur et al., 2005). The instrument setup is presented in Fig. 29.2.
3.1. Procedure Samples are sealed in distinct glass cuvettes, which can be alternately sampled using electronic valves. Gases emitted from each sample are transported to the photoacoustic detection cell by an airflow (1.5 literh1) applied to sampling cells. To remove excess water vapor as a result of a temperature increase, gas samples pass into a Peltier cooling element (–5 ) and a cold trap (–80 ) prior to passage into the photoacoustic cell where
580
E. Vandelle and M. Delledonne
Chamber 1 (aquatic sample) Injection tube
Mass spectrometer 14
NO = 30
Plug Coolant
Tissue culture sample Stirring bar
Out 15
NO = 31
MIMS
Coolant in O2 = 32 Capillary Ionization chamber
3-way valve
Teflon membrane
Magnetic stirrer
Restriction capillary Valve
Leaf/plant cuvette
Injection opening Valve Pump
RIMS Chamber 2 (gaseous sample)
Figure 29.2 Schematic diagram of the experimental setup for MIMS/RIMS-based NO measurements. In MIMS, a cell suspension in an 8- to 10-ml reaction chamber was circulated over a thin (50 mm) Teflon membrane by a magnetic stirrer. Dissolved gases, such as NO, diffused through the membrane and evaporated into the ionization chamber of a mass spectrometer. In RIMS, a metal bellows pump ensured rapid and efficient mixture of the 120-ml gas phase, which included the volume of a leaf cuvette (880.4 cm). Before entering the mass spectrometer, NO and other gases passed a restriction capillary (inner diameter: 0.1 mm; length: 2 m). A three-way valve served to switch between the two sample chambers. Reproduced with permission from Conrath et al. (2004).
they are excited with a chopped carbon monoxide (CO) laser as a source of infrared light. To improve laser power, the photoacoustic cell is inserted into a laser cavity, consisting of a gas discharge tube, a rotatable grating (for laser line selection), and a 100% reflecting mirror. Five laser lines are used to measure NO concentration. Among them, the strongest one for NO measurement is at 1900.0426 cm1, and one is used to determine the water concentration in the sample (at 1790.6576 cm1). Because there is a mixture of gases in the detection cell, and each gas has different absorption strength on every laser line, the mixed absorption
Nitric Oxide Detection Methods
581
strength pattern is unraveled using a multicomponent matrix calculation algorithm. Results are presented as moles of NOh1g1 of fresh weight (if the amount of plant material has been determined at the beginning of the experiment) (Mur et al., 2005).
3.2. Results The sensitivity of the method has been assessed by applying known concentrations of NO to the airflow. In this way, a near 1:1 predicted/ measured relationship between approximately 28 and 4.5 nmolh1 was observed (Mur et al., 2005). The sensitivity and the reliability of the method in planta have been evaluated by infiltrating different concentrations of the NO donor sodium nitroprusside (SNP; 0.1 to 10 mM ) in tobacco leaves. During the first hour postinfiltration, NO emission by plant was proportional to added SNP. Finally, the specificity of the NO signal has been confirmed by adding ozone to the air mixture, as NO readily forms NO2 in the presence of O3: the addition of 0.2 literh1 NO to 1.3 literh1 airflow gives a photoacoustic signal that is completely abolished following the addition of 0.2 literh1 O3 to the mixture (with a concomitant reduction in air to 1.1 literh1 to maintain a constant gas flow). In the same way, the NO signal due to SNP is abolished if O3 is added to the gas flow after passing through the cuvette. Using this method, Mur et al. (2005) have observed a rapid increase in NO production in tobacco leaves challenged with avirulent P. syringae pv. phaseolicola. This production, which occurs 40 min after infection, reaches about 17 nmol NOh2g1 fresh weight after 1 h.
3.3. Comments A similar technique has been employed by Leshem and Pinchasov (2000) to determine relative endogenous NO content during the ripening of strawberries. In this work, the authors used a CO2 laser to excite samples placed in a transparent sample cuvette at room temperature (22 C) in fluorescent light at 150 mMs1m2 intensity. To prevent NO2 formation that may occur in air, the sample cuvette was filled with N2 prior to NO measurement. Laser photoacoustic detection displays a high sensitivity, with approximately 21.3 pmolh-1 that could be measured accurately and which is equivalent to thresholds for chemiluminescent detection (Archer, 1993); greater than 10-fold increased sensitivity compared to RIMS is seen (Conrath et al., 2004). However, because LPAD detects NO in a gas mixture emitted from plants, it does not allow intracellular NO measurements.
582
E. Vandelle and M. Delledonne
4. Nitric Oxide Detection by Chemiluminescence The concentration of NO can be determined using a simple chemiluminescent reaction involving ozone to produce oxygen and nitrogen dioxide:
NO þ O3 ! NO2 þ O2 ; NO2 ! NO2 þ hu This reaction produces light (chemiluminescence) that can be measured using a photodetector. The amount of light produced is proportional to the amount of NO in the sample. The assay takes advantage of the low solubility of NO in aqueous solution by measuring NO in the gas phase. This technique has been described in detail elsewhere (Planchet and Kaiser, 2006; Planchet et al., 2005, 2006), as follows.
4.1. Procedure For experiments with detached leaves, leaves are cut off from the plant and immediately placed in nutrient solution (adapted to studied plants), where the petiole is cut off a second time below the solution surface. The leaves (petiole in nutrient solution) are placed in a transparent lid chamber with 2 or 4 liters air volume, depending on the leaf size and number. A constant flow of measuring gas (purified air or nitrogen) of 1.5 litermin1 is pulled through the chamber and subsequently through the chemiluminescence detector [detection limit 20 parts per trillion (ppt); 20-s time resolution] by a vacuum pump connected to an ozone destroyer. The ozone generator of the chemiluminescence detector is supplied with dry oxygen (99%). The measuring gas (air or nitrogen) is made NO free by conducting it through a custom-made charcoal column (1 m long, 3 cm internal diameter, particle size 2 mm). Calibration is routinely carried out with NO free air (0 ppt NO) and with various concentrations of NO (1–35 ppb) adjusted by mixing the calibration NO gas (500 ppb in nitrogen) with NO-free air. Light is provided by a 400-W Hqi lamp above the cuvette. Quantum flux density is adjusted within limits (150–400 mmolm2s1 photon flux density) by changing the distance between lamp and cuvette. Air temperature in the cuvette is monitored continuously and is usually about 20 in the dark and 23–25 C in the light. For measurement of NO production from cell suspensions (10 ml), solutions are placed in small glass beakers of suitable size, located in a transparent lid chamber (1 liter gas volume) mounted on a shaker. NO is then measured by chemiluminescence detection as described earlier.
Nitric Oxide Detection Methods
583
4.2. Results Gaseous NO or aqueous solutions containing various amounts of NO can be used to calibrate the experiment. For example, based on the solubility of pure NO in water (1.9 mM at atmospheric pressure and 22 C), the equilibrium solution of a buffered solution (100 mM HEPES, pH 7.5) flushed for 15 min with 100 ppm NO gas contains 190 pmol NOml1 (Planchet and Kaiser, 2006; Planchet et al., 2006). Aliquots of this solution are rapidly injected under a gas stream of air or nitrogen into a small beaker with buffer solution mounted in the headspace cuvette on a magnetic stirrer, and the NO released is detected by chemiluminescence. It has been reported that the integrated amount of NO detected by this method is practically identical to the theoretical NO content (Planchet and Kaiser, 2006). This technique does not detect NO production in cryptogein-treated leaves, although it has been shown to reveal NO emission from tobacco cells treated with the elicitor cryptogein (Planchet et al., 2006). In the latter case, however, NO production was detected only after a 3- to 6-h treatment and was not affected by NOS inhibitors, contrary to previous results obtained with 4,5-diaminofluorescein diacetate (DAF-2 DA) staining (Lamotte et al., 2004; see Section 7).
4.3. Comments The chemical properties of NO enhance the specificity of the assay. This method offers high sensitivity, around 20 pmol NO (Archer, 1993), and is quantitative at NO concentrations in the picomolar range. However, it does not allow for intracellular detection of NO, and results obtained with this method compared to other techniques are debatable.
5. Nitric Oxide Detection by Hemoglobin Conversion This method is based on hemoglobin absorbance changes as a result of its conversion from oxyhemoglobin (HbO2) to methemoglobin (metHb) in the presence of NO:
Hb FeðIIÞ O2 þ NO ! Hb FeðIIIÞ þ NO3 5.1. Procedure For experiments with detached leaves (Orozco-Cardenas and Ryan, 2002), 200 mg of frozen leaves (harvested after treatment) is ground and homogenized in 1 ml of cooled buffer [0.1 M sodium acetate, 1 M NaCl, and
584
E. Vandelle and M. Delledonne
1% (w/v) ascorbic acid, pH 6.0]. The homogenates are centrifuged at 10,000 g for 20 min at 4 C, and the supernatants are clarified by passing through a 0.8 4-cm column in 1-X8 resin. For experiments with cell suspensions (Clarke et al., 2000; Delledonne et al., 1998), cells are washed twice and resuspended at 0.1 gml1 in a minimum buffer (e.g., 50 mM MES, 75 mM sucrose, 1 mM CaCl2, 1 mM K2SO4, pH 5.5; Clarke et al., 2000). Following treatments, 1-ml aliquots of cells are removed at various time points. In each case, 1 ml of sample (clear leaf homogenate or cell suspension) is incubated for 5 min with 100 U catalase and 100 U superoxide to remove reactive oxygen intermediates that could interfere with the assay (Delledonne et al., 1998). It is important to note that the buffer should not contain any compounds that could absorb in the range of 390–430 or 560–610 nm. The HbO2 solution stock is then added to samples to a final concentration between 5 and 10 mM. After 2–5 min of incubation (depending on the rate of NO production in the system under study), the rate of HbO2 to metHb conversion is evaluated spectrophotometrically. If NO production is analyzed in cell suspensions, cells are pelleted by centrifugation at 10,000 g for 30 s prior to measuring the absorbance.
5.2. Results The absorbance peak for HbO2 at 415 nm shifts toward 406 nm for metHb. If a dual-wavelength spectrophotometer is available, measurements of the conversion are obtained by the difference absorbance of 401–411 nm (De ¼ 38 mM1cm1), 421–411 nm (De ¼ 39 mM1cm1), or 401–421 nm (De ¼ 77 mM1cm1). Using the last pair of wavelengths, the theoretical detection limit for NO is 1.3 nM. If only single wavelength readings are possible, 401 or 421 nm will give reliable data (Murphy and Noack, 1994). Nitric oxide emission by plant cells infected with pathogens, i.e., a rapid and relatively weak NO burst (around 0.5 mM ) after 30 min followed by a second one severalfold greater (2 mM ), specifically induced in cells inoculated with the avirulent bacterial strain (Delledonne et al., 1998), was first demonstrated using this method.
5.3. Comments This assay has few technical requirements and has a NO detection threshold around 1 nM. However, it is prone to interference by ROS. The addition of superoxide dismutase and catalase as sample pretreatments can minimize this problem, but does not completely ensure the specificity of the method.
Nitric Oxide Detection Methods
585
6. Nitric Oxide Detection by Electron Paramagnetic Resonance (EPR) Spin Trap Electron paramagnetic resonance detection is based on the fact that at a discrete amount of energy (microwave frequency) and magnetic field strength, unpaired electrons are promoted to higher energy levels; following this, the relaxation from this state produces a characteristic spectrum. Although NO is a paramagnetic molecule with an unpaired electron, it cannot be studied by simple EPR, as the relaxation time of the stimulated electron to the ground state is too rapid to be detected (Maples et al., 1991). Therefore, EPR detection has been combined with spin trapping to stabilize the labile free radical and allow NO measurement. Different types of spin traps have been used to detect NO by EPR, but only those that have been used to measure NO emission during plant–pathogen interactions are described here. The spin-trapping agents used are diethyldithiocarbamate (DETC) and N-methyl-D-glucamine dithiocarbamate (MGD). Binding NO with hydrophobic Fe2þ–dithiocarbamate complexes results in the formation of paramagnetic mononitrosyl iron complexes with dithiocarbamate that can be detected by EPR spectroscopy at 77 K and ambient temperature.
6.1. Procedure 6.1.1. Use of DETC To overcome the problem of water solubility of the Fe2þ(DETC)2 complex, albumin should be added to the reaction mixture (Tsuchiya et al., 1996). Moreover, the trapping reaction should be carried out in presence of Na2S2O4 in order to avoid oxidation of iron and NO in the Fe2þ(DETC)2NO complex (Tsuchiya et al., 1996). For experiments performed with detached leaves (Huang et al., 2004), about 0.6-g leaves frozen in liquid nitrogen are crushed with a mortar and pestle and incubated in 1.2 ml of buffered solution [50 mM HEPES, 1 mM dithiothreitol (DTT), 1 mM MgCl2, pH 7.6] for 2 min. For experiments performed with cell suspensions (Zeidler et al., 2004), 500 ml of cells are harvested at different time points after treatment and incubated in 0.6 ml of 50 mM HEPES, pH 7.6, 1 mM DTT, 1 mM MgCl2, at 37 C for 2 min. In each case, the mixture is centrifuged at 13,000 g for 2 min. The supernatant is then added to 300 ml of freshly made Fe2þ(DETC)2 solution (2 M Na2S2O4, 3.3 mM DETC, 3.3 mM FeSO4, 33 mgml-1 bovine serum albumin), incubated for 2 min at room temperature, and frozen in liquid nitrogen. Electron paramagnetic resonance measurements are performed on a Bruker ESP300 X-band spectrometer under the following conditions:
586
E. Vandelle and M. Delledonne
room temperature; microwave power, 20 mW; modulation amplitude, 3 G; scan rate, approximately 2.5 Gs1; time constant, 164 ms (Huang et al., 2004; Zeidler et al., 2004). 6.1.2. Use of MGD N-Methyl-D-glucamine dithiocarbamate is a derivative of DETC (Komarov and Lai, 1995). It is readily soluble in water and forms a water-soluble Fe2þ(MGD)2NO complex. Extracts of leaves are obtained by homogenizing 170 mg of leaf tissue in 200 ml of phosphate buffer (100 mM, pH 7.2) using a Polytron mixture system (Kinematica AG; Modolo et al., 2005). After centrifugation at 10,000 g for 10 min, the supernatant is incubated for 1 h at room temperature in an equal volume of 100 mM phosphate buffer containing Fe2þ(MGD)2 at 1 mM Fe2þ [stock solutions of Fe2þ(MGD)2 are prepared by dissolving MGD sodium salt and FeSO4 in deionized water, with a molar ratio of 5:1, respectively]. Care should be taken to use this trap in anaerobic conditions to avoid oxidation of Fe2þ to Fe3þ, which results in loss of trap as well as subsequent production of superoxide that would interfere with NO measurement. Samples are then frozen, stored in liquid nitrogen, and thawed immediately before EPR analysis. Electron paramagnetic resonance measurements are carried out with a Bruker EMX instrument under the following conditions: room temperature; microwave power, 20 mW; modulation amplitude, 2.5 G; scan rate, 2.4 Gs1; time constant, 81.92 ms; gain, 2.0 105 (Modolo et al., 2005).
6.2. Results The NO-Fe2þ(DETC)2 complex gives a three-line EPR spectrum at room temperature (giso ¼ 2.04; aN ¼ 2.7 G; see Fig. 29.2). A calibration curve can be obtained using NaNO2 as an NO source (reaction mixture: 3.3 mM Fe2þ, 3.3 mM DETC, 33 mgml1 albumin, about 2 M NaS2O4, 0–10 mM NaNO2). The intensity of the signal increases linearly with the NO concentration, with a correlation coefficient of 0.998 (Tsuchiya et al., 1996). The NO donor SNP can also be used as a standard in the presence of excess Fe2þ(DETC)2 complex (Huang et al., 2004; Tsuchiya et al., 1996; Zeidler et al., 2004). The concentration can be determined using peak intensity or by double integration. Using EPR analysis with DETC as a spin-trapping agent, NO has been detected early in Arabidopsis thaliana cells elicited with lipopolysaccharides after 10 min (Zeidler et al., 2004). These authors have observed the characteristic three-line spectrum, without quantifying the amount of NO produced.
Nitric Oxide Detection Methods
587
The Fe2þ(MGD)2NO complex has an EPR spectrum very similar to that of Fe2þ(DETC)2NO, i.e., a three-line spectrum with giso ¼ 2.04, aN ¼ 12.9 G in aqueous solution. Quantification of the spin adduct can be performed using an aqueous solution of TEMPOL (4-hydroxy-2,2,6,6-tetramethyl piperidine N-oxyl) as a standard ( Jasid et al., 2006). TEMPOL solutions are standardized spectrophotometrically at 429 nm using e ¼ 13.4 M1cm1. Next, the concentration of the Fe2þ(MGD)2NO adduct is obtained by double integration of the three lines and cross-checked with TEMPOL spectra. Using this method with MGD as a spin-trapping agent, Modolo et al. (2005) have demonstrated NO emission in A. thaliana leaves in response to avirulent P. syringae pv. maculicola 6 h after infection and have shown the involvement of nitrite as a major source of NO during this process. In this study, however, the amount of NO produced was not determined.
6.3. Comments Care must be taken in terms of the chemicals used to analyze NO production using EPR spectrometry. For example, L-NG-nitroarginine methylester (L-NAME) is not a suitable inhibitor of NO synthase in measurement of NO by the Fe2þ(DETC)2 complex method, as under strong reductive conditions it shows the same spectrum of the Fe2þ(DETC)2NO complex (Tsuchiya et al., 1996). Also, the potential interference from nitrite must be considered. For instance, the use of reducing agents that convert NO2 back to NO results in overestimating the amount of free NO in aqueous solutions (Venkataraman et al., 2002). Thus, the use of DETC as a spin trap that requires the addition of Na2S2O3 in excess is not recommended and MGD is preferred. Finally, EPR does not allow easy continuous NO detection in planta.
7. Nitric Oxide Detection Using Diaminofluoresceins This method is based on the reaction of aromatic amines with NO in the presence of dioxygen to produce the corresponding triazenes; the corresponding triazole ring compounds are generated spontaneously from aromatic vicinal diamines under neutral conditions (Nagano et al., 1995). The most common probe is 4,5-diaminefluorescein, a fluorescein derivate (Kojima et al., 1998). DAF-2 does not react directly with NO, but with the
588
E. Vandelle and M. Delledonne
N2O3 formed during the course of NO oxidation, according to the following reactions (Ignarro et al., 1993):
2NO þ O2 ! 2NO2
ð29:1Þ
2NO þ 2NO $ 2N2 O3
ð29:2Þ
The reaction between DAF-2 and the oxidation product of NO leads to formation of the highly fluorescent triazolofluorescein DAF-2T (Kojima et al., 1998). For intracellular NO detection, the cell-permeable derivate of DAF-2, namely DAF-2 diacetate (DAF-2 DA), can permeate readily into cells where it is hydrolyzed by intracellular esterases to generate DAF-2 (Kojima et al., 1998). The fluorescence intensities of the triazole derivates of DAFs are dependent on pH (Kojima et al., 1999). 3-Amino-4-(N-methylamino)20 ,70 -difluofluorescein (DAF-FM) is an improved DAF analogue, which, after reaction with NO, results in the triazole DAF-FM T that shows stable and intense fluorescence in a wide range of pH values (Kojima et al., 1999).
7.1. Procedure Leaves are infiltrated with a 10 mM DAF-2 DA-containing solution (10 mM MES-Tris, 10 mM KCl, 0.1 mM CaCl2, pH 7.6), and 1 h after infiltration, the treated leaf areas are analyzed by confocal microscopy (Qu et al., 2006). Alternatively, leaf segments (2 2 mm) are mounted in MES buffer (10 mM MES, 50 mM KCl, pH 6.15) on a glass slide with a coverslip. Samples are immersed in 10 mM DAF-2 DA in MES buffer for 10 min in the dark at room temperature, rinsed in pure MES buffer to remove excess dye for another 10 min, and analyzed by microscopy (Prats et al., 2005). Experiments can also be carried out with leaf epidermal sections produced from the abaxial or the adaxial surface (Foissner et al., 2000; Gould et al., 2003; Zeidler et al., 2004). Leaf peels are incubated in a 10 mM DAF-2 DA- or 5 mM DAFFM DA-containing solution (10 mM Tris, pH 7.0–7.2) for 10 to 30 min in the dark at room temperature. Sections are then removed and transferred to a dish of fresh loading buffer (without probe) to wash excess fluorophore for 10–20 min. Next, the samples are mounted on microscope slides, where they are still immersed in fresh loading buffer, and examined immediately by microscopy. Microscopic analyses on whole leaves or epidermal peels are usually carried out with a confocal laser microscope (488 nm excitation, emission spectrum comprised between 500 and 550 nm), but have also been performed using an epifluorescence microscope equipped with an FITC filter set (excitation, 490 nm; beam splitter, 510 nm; emission, 525 nm). To detect NO in plant cell suspensions (Lamotte et al., 2004; Planchet et al., 2005, 2006; Vandelle et al., 2006; Wang and Wu, 2004; 2005;
Nitric Oxide Detection Methods
589
Yamamoto et al., 2004), cells are washed in fresh culture medium or usual assay buffer and incubated in the same buffer containing 10 to 20 mM of DAF-2 DA for 15–60 min in the dark at room temperature under constant agitation. Next, cells are washed (once or twice) with fresh suspension buffer to wash off excess fluorophore. Cell treatments can be performed directly in the flask containing DAF-2 DA-loaded cells. In this case, aliquots (1 or 2 ml) of cells are taken at different times after treatment and analyzed for DAF-2T fluorescence (Planchet et al., 2005, 2006; Wang and Wu, 2004, 2005; Yamamoto et al., 2004). Alternatively, cell suspensions are transferred into a 24-well plate (1 ml per well) and treated in the dark. NO production is then measured using a plate reader fluorometer (Lamotte et al., 2004; Vandelle et al., 2006). NO production in cell suspensions is measured using a luminescence spectrophotomer or a spectrofluorometer (485 or 495 nm excitation, 510 or 515 nm emission).
7.2. Results The formation of DAF-2T produces a green fluorescence attributable to the presence of NO in cells. DAF-2T fluorescence is expressed in relative units. As a positive control to test the efficiency of the staining, some samples are incubated in a solution containing an NO donor, in addition to the buffered DAF-2 DA. In contrast, as negative controls, in all types of experiments, samples are incubated in assay buffer lacking DAF-2 DA. Using DAF-2 DA, Foissner et al. (2000) have reported on the real-time imaging of NO production in cryptogein-treated epidermal tobacco cells and have shown that NO accumulation and/or production by confocal microscopy analysis occurs early in chloroplasts, in the nucleus and along the plasma membrane, and in distinct cellular compartments in the vicinity of chloroplasts, most likely peroxisomes. A monophasic NO burst that occurs within a few minutes after elicitor treatment has been detected in tobacco cell suspensions elicited with cryptogein by measuring the fluorescence increase of DAF-2T using a plate reader fluorometer (Lamotte et al., 2004). 3-Amino-4-(N-methylamino)-20 ,70 -difluofluorescein staining, combined with confocal laser-scanning microscopy, has been used to analyze the time course of NO production in A. thaliana cell suspensions elicited with lipopolysaccharides (Zeidler et al., 2004). In this study, an increase in DAF-FM T fluorescence was observed as early as 2 min after elicitation.
7.3. Comments 4,5-Diaminefluorescein cross-reacts with dehydroascorbic acid (DHA) to produce fluorescent compounds, termed DAF-2-DHAs, while ascorbic acid considerably attenuates the formation of DAF-2T, probably by affecting the formation of N2O3 (Zhang et al., 2002). In addition,
590
E. Vandelle and M. Delledonne
catecholamines, superoxide radical, dithiothreitol, 2-mercaptoethanol, glutathione, and divalent cations such as Ca2þ or Mg2þ can interfere with DAF-2 during NO detection (Broillet et al., 2001; Nagata et al., 1999). In order to avoid false-positive results, it is also necessary to check for any eventual autofluorescence detectable around 515 nm for all substances employed in the assay. Decreased fluorescence of samples not protected from light even for a few minutes has been reported (Ra¨thel et al., 2003) and it is, therefore, strictly necessary to work in the dark when handling DAF-2 or DAF-2T samples. 4,5-Diaminefluorescein does not react in neutral solutions with other oxidized forms of NO, such as NO 2 and NO3 , or with other ROS, such as O2 , H2O2, and ONOO , providing specificity for NO detection (Kojima et al., 1998). However, because DAF-2 does not react with the NO-free radical but rather with N2O3, the fluorescence intensity depends on the rate of NO oxidation, which requires oxygen, making the use of DAFs under anoxia difficult. By calibrating NO standard curves, it has been shown that the detection limit of NO by DAF-2 is 3–5 nM (Itoh et al., 2000; Kojima et al., 1998).
8. Conclusion Ideally, methods for determination of NO should exhibit a high degree of sensitivity and specificity and should allow assessment of intra- and extracellular levels of NO from gas or liquid phases (Mur et al., 2005). Even so, no assay described herein possesses all these ideal characteristics. Some techniques allow direct measurement of NO concentration, such as EPR spin trap, or indirect NO detection, for example, using DAF-2DA/ DAF-FM staining. Other methods are suitable for NO detection in the gas phase, such as laser photoacoustic spectroscopy, chemiluminescence, or mass spectrometry. Another method consists of electrochemical detection of NO using a Clark-type NO electrode (Yamasaki et al., 1999). This method has not been described in this chapter because it has never been used to study plant defense responses. All the other methods cited have, however, been used to study NO during plant–pathogen interactions, often leading to large discrepancies in the results, even for the same plant–pathogen system. For example, laser photoacoustic spectroscopy (Mur et al., 2005), as well as DAF2DA/FM staining (Foissner et al., 2000; Lamotte et al., 2004; Zeidler et al., 2004) and chemiluminescence (Planchet et al., 2006), revealed monophasic NO production in tobacco leaves or cells after elicitation. In contrast, when analyzed by MIMS, the same tobacco cells showed two NO bursts after avirulent P. syringae infection (Conrath et al., 2004), as also observed in
Nitric Oxide Detection Methods
591
elicited soybean cells by MIMS and by the oxyhemoglobin assay (Conrath et al., 2004; Delledonne et al., 1998). Moreover, even if both DAF-2DA staining (Lamotte et al., 2004) and chemiluminescence (Planchet et al., 2006) detected a monophasic pattern of NO production in cryptogein-treated tobacco cells, the former method detects these changes very early, after a few minutes of treatment, whereas the latter detects differences after a 6-h treatment. These discrepancies may be because of differences in sensitivity, specificity, and localization of NO detection. Taking into account the most important features that accurate methods for NO detection should possess, DAF-2DA/FM staining, which consists of measuring NO indirectly by detecting N2O3 (Kojima et al., 1998), appears to be an easy and suitable technique. It is conceptually similar to the indirect measurement of NO by measuring nitrite levels using the Griess reaction or the aromatic diamino compound 2,3-diaminonaphthalene, which are commonly used to measure NO in animal systems (Nagano, 1999). Moreover, given the importance of the spatiotemporal aspects of NO production, the main asset of DAF-2DA/FM dyes is their capacity to detect NO at the intracellular level in real-time conditions. However, many questions still remain that make the study of NO during the HR problematic: how and where is NO produced in this process, is it trapped in cells or does it mainly diffuse across membranes, and what are its targets? It is still not known how NO reacts in plant cells where redox balance is highly affected because of the massive concomitant oxidative burst required for triggering hypersensitive cell death. For these reasons, it seems clear that more than one method is needed to accurately quantify the NO produced in plants when challenged by a pathogen. Both intracellular and extracellular NO content should be measured in addition to detecting different forms of NO (gas, radical, or oxidative metabolites).
REFERENCES Archer, S. (1993). Measurement of nitric oxide in biological models. FASEB J. 7, 349–360. Beligni, M. V., and Lamattina, L. (2001). Nitric oxide: A non-traditional regulator of plant growth. Trends Plant Sci. 6, 508–509. Berkels, R., Purol-Schnabel, S., and Roesen, R. (2001). A new method to measure nitrate/ nitrite with a NO-sensitive electrode. J. Appl. Physiol. 90, 317–320. Bethke, P. C., Badger, M. R., and Jones, R. L. (2004). Apoplastic synthesis of nitric oxide by plant tissues. Plant Cell 16, 332–341. Broillet, M., Randin, O., and Chatton, J. (2001). Photoactivation and calcium sensitivity of the fluorescent NO indicator 4,5-diaminofluorescein (DAF-2): Implications for cellular NO imaging. FEBS Lett. 491, 227–232. Clarke, A., Desikan, R., Hurst, R. D., Hancock, J. T., and Neill, S. J. (2000). NO way back: Nitric oxide and programmed cell death in Arabidopsis thaliana suspension cultures. Plant J. 24, 667–677.
592
E. Vandelle and M. Delledonne
Conrath, U., Amoroso, G., Ko¨hle, H., and Su¨ltemeyer, D. (2004). Non-invasive online detection of nitric oxide from plants and some other organisms. Plant J. 38, 1015–1022. Davies, M. J. (2002). Recent developments in EPR spin-trapping. Electron Paramagn. Reson. 18, 47–73. Delledonne, M. (2005). NO news is good news for plants. Curr. Opin. Plant Biol. 8, 390–396. Delledonne, M., Xia, Y., Dixon, R. A., and Lamb, C. (1998). Nitric oxide functions as a signal in plant disease resistance. Nature 394, 585–588. Delledonne, M., Zeier, J., Marocco, A., and Lamb, C. (2001). Signal interactions between nitric oxide and reactive oxygen intermediates in the plant hypersensitive disease resistance response. Proc. Natl. Acad. Sci. USA 98, 13454–13459. de Pinto, M. C., Tommasi, F., and De Gara, L. (2002). Changes in the antioxidant systems as part of the signaling pathway responsible for the programmed cell death activated by nitric oxide and reactive oxygen species in tobacco Bright-Yellow 2 cells. Plant Physiol. 130, 698–708. Foissner, I., Wendehenne, D., Langebartels, C., and Durner, J. (2000). In vivo imaging of an elicitor-induced nitric oxide burst in tobacco. Plant J. 23, 817–824. Gould, K. S., Lamotte, O., Klinguer, A., Pugin, A., and Wendehenne, D. (2003). Nitric oxide production in tobacco leaf cells: A generalized stress response? Plant Cell Environ. 26, 1851–1862. Huang, X., Stettmaier, K., Michel, C., Hutzler, P., Mueller, M. J., and Duner, J. (2004). Nitric oxide is induced by wounding and influences jasmonic acid signalling in Arabidopsis thaliana. Planta 218, 938–946. Ignarro, L. J., Fukuto, J. M., Griscavage, J. M., Rogers, N. E., and Byrns, R. E. (1993). Oxidation of nitric oxide in aqueous solution to nitrite but not nitrate: Comparison with enzymatically formed nitric oxide from L-arginine. Proc. Natl. Acad. Sci. USA 90, 8103–8107. Itoh, Y., Ma, F. H., Hoshi, H., Oki, M., Noda, K., Ukai, Y., Kojima, H., Negano, T., and Toda, N. (2000). Determination and bioimaging method for nitric oxide in biological specimens by diaminofluorescein fluorometry. Anal. Biochem. 287, 203–209. Jasid, S., Simontacchi, M., Bartoli, C. G., and Puntarulo, S. (2006). Chloroplasts as nitric oxide cellular source: Effect of reactive nitrogen species on chloroplastic lipids and proteins. Plant Physiol. 142, 1246–1255. Kojima, H., Nakatsubo, N., Kikuchi, K., Kawahara, S., Kirino, Y., Nagoshi, H., Hirata, Y., and Nagano, T. (1998). Detection and imaging of nitric oxide with novel fluorescent indicators: Diaminofluoresceins. Anal. Chem. 70, 2446–2453. Kojima, H., Urano, Y., Kikuchi, K., Higuchi, T., Hirata, Y., and Nagano, T. (1999). Fluorescent indicators for imaging nitric oxide production. Angew. Chem. Int. Ed. Engl. 38, 3209–3212. Komarov, A. M., and Lai, C.-S. (1995). Detection of nitric oxide production in mice by spin-trapping electron paramagnetic resonance. Biochim. Biophys. Acta 1272, 29–36. Lai, C.-S., and Komarov, A. M. (1994). Spin trapping of nitric oxide produced in vivo in septic-shock mice. FEBS Lett. 345, 120–124. Lamotte, O., Gould, K., Lecourieux, D., Sequeira-Legrand, A., Lebrun-Garcia, A., Durner, J., Pugin, A., and Wendehenne, D. (2004). Analysis of nitric oxide signaling functions in tobacco cells challenged by the elicitor cryptogein. Plant Physiol. 135, 516–529. Leshem, Y. Y., and Pinchasov, Y. (2000). Non-invasive photoacoustic spectroscopic determination of relative endogenous nitric oxide and ethylene content stoichiometry during the ripening of strawberries Fragaria anannasa (Duh.) and avocados Persea Americana (Mill.). J. Exp. Bot. 51, 1471–1473.
Nitric Oxide Detection Methods
593
Maples, K. R., Sandstrom, T., Su, Y. F., and Henderson, R. F. (1991). The nitric oxide/ heme protein complex as a biologic marker of exposure to nitrogen dioxide in humans, rats, and in vitro models. Am. J. Respir. Cell Mol. Biol. 4, 538–543. Modolo, L. V., Augusto, O., Almeida, I. M., Magalhaes, J. R., and Salgano, I. (2005). Nitrite as the major source of nitric oxide production by Arabidopsis thaliana in response to Pseudomonas syringae. FEBS Lett. 579, 3814–3820. Mur, L. A. J., Santosa, I. E., Laarhoven, L. J. J., Holton, N. J., Harren, F. J. M., and Smith, A. R. (2005). Laser photoacoustic detection allows in planta detection of nitric oxide in tobacco following challenge with avirulent and virulent Pseudomonas syringae pathovars. Plant Physiol. 138, 1247–1258. Murphy, M. E., and Noack, E. (1994). Nitric oxide assay using hemoglobin method. Methods Enzymol. 233, 240–250. Nagano, T. (1999). Practical methods for detection of nitric oxide. Luminescence 14, 283–290. Nagano, T., Takizawa, H., and Hirobe, M. (1995). Reaction of nitric oxide with amines in the presence of dioxygen. Tetrahedron Lett. 36, 8239–8242. Nagata, N., Momose, K., and Ishida, Y. (1999). Inhibitory effects of catecholamines and anti-oxidants on the fluorescence reaction of 4,5-diaminofluorescein, DAF-2, a novel indicator of nitric oxide. J. Biochem. (Tokyo) 125, 658–661. Neill, S. J., Desikan, R., and Hancock, J. T. (2003). Nitric oxide signalling in plants. New Phytol. 159, 11–35. Orozco-Ca´rdenas, M. L., and Ryan, C. A. (2002). Nitric oxide negatively modulates wound signalling in tomato plants. Plant Physiol. 130, 487–493. Planchet, E., Jagadis Gupta, K., Sonoda, M., and Kaiser, W. M. (2005). Nitric oxide emission from tobacco leaves and cell suspensions: Rate limiting factors and evidence for the involvement of mitochondrial electron transport. Plant J. 41, 732–743. Planchet, E., and Kaiser, W. M. (2006). Nitric oxide (NO) detection by DAF fluorescence and chemiluminescence: A comparison using abiotic and biotic NO sources. J. Exp. Bot. 57, 3043–3055. Planchet, E., Sonoda, M., Zeier, J., and Kaiser, W. M. (2006). Nitric oxide (NO) as an intermediate in the cryptogein-induced hypersensitive response: A critical re-evaluation. Plant Cell Environ. 29, 59–69. Prats, E., Mur, L. A. J., Sanderson, R., and Carver, T. L. W. (2005). Nitric oxide contributes both papilla-based resistance and the hypersensitive response in barley attacked by Blumeria graminis f.sp. hordei. Mol. Plant Pathol. 6, 65–78. Qu, Z.-L., Zhong, N.-Q., Wang, H.-Y., Chen, A.-P., Jian, G. -L., and Xia, G.-X. (2006). Ectopic expression of the cotton non-symbiotic haemoglobin gene GhHb1 triggers defense responses and increases disease tolerance in Arabidopsis. Plant Cell Physiol. 47, 1058–1068. Ra¨thel, T. R., Leikert, J. F., Vollmar, A. M., and Dirsch, V. M. (2003). Application of 4,5-diaminofluorescein to reliably measure nitric oxide released from endothelia cells in vitro. Biol. Proced. Online 5, 136–142. Rockel, P., Strube, F., Rockel, A., Wildt, J., and Kaiser, W. M. (2002). Regulation of nitric oxide (NO) production by plant nitrate reductase in vivo and in vitro. J. Exp. Bot. 53, 103–110. Tsuchiya, K., Takasugi, M., Minakuchi, K., and Fukuzawa, K. (1996). Sensitive quantitation of nitric oxide by EPR spectroscopy. Free Radic. Biol. Med. 21, 733–737. Vandelle, E., Poinssot, B., Wendehenne, D., Bentejac, M., and Alain, P. (2006). Integrated signaling network involving calcium, nitric oxide, and active oxygen species but not mitogen-activated protein kinases in BcPG1-elicited grapevine defenses. Mol. Plant Microbe Interact. 19, 429–440.
594
E. Vandelle and M. Delledonne
Venkataraman, S., Martin, S. M., and Buettner, G. R. (2002). Electron paramagnetic resonance for quantitation of nitric oxide in aqueous solutions. Methods Enzymol. 359, 3–18. Wang, J. W., and Wu, J. Y. (2004). Involvement of nitric oxide in elicitor-induced defense responses and secondary metabolism of Taxus chinensis cells. Nitric Oxide 11, 298–306. Wang, J. W., and Wu, J. Y. (2005). Nitric oxide is involved in methyljasmonate-induced defense responses and secondary metabolism activities of Taxus cells. Plant Cell Physiol. 46, 923–930. Wendehenne, D., Durner, J., and Klessig, D. F. (2004). Nitric oxide: A new player in plant signalling and defence responses. Curr. Opin. Plant Biol. 7, 449–455. Yamamoto, A., Katou, S., Yoshioka, H., Doke, N., and Kawakita, K. (2004). Involvement of nitric oxide generation in hypersensitive cell death induced by elicitin in tobacco cell suspension culture. J. Gen. Plant Pathol. 70, 85–92. Yamasaki, H., Sakihama, Y., and Takahashi, S. (1999). An alternative pathway for nitric oxide production in plants: New features of an old enzyme. Trends Plant Sci. 4, 128–129. Zafiriou, O. C., and McFarland, M. (1980). Determination of trace levels of nitric oxide in aqueous solution. Anal. Chem. 52, 1662–1667. Zaninotto, F., La Camera, S., Polverari, A., and Delledonne, M. (2006). Cross talk between reactive nitrogen and oxygen species during the hypersensitive disease resistance response. Plant Physiol. 141, 379–383. Zeidler, D., Za¨hringer, U., Gerber, I., Dubery, I., Hartung, T., Bors, W., Hutzler, P., and Durner, J. (2004). Innate immunity in Arabidopsis thaliana: Lipopolysaccharides activate nitric oxide synthase (NOS) and induce defense genes. Proc. Natl. Acad. Sci. USA 101, 15811–15816. Zhang, X., Kim, W. S., Hatcher, N., Potgieter, K., Moroz, L. L., Gillette, R., and Sweedler, J. V. (2002). Interfering with nitric oxide measurements: 4,5-Diaminofluorescein reacts with dehydroascorbic acid and ascorbic acid. J. Biol. Chem. 277, 48472–48478.
C H A P T E R
T H I R T Y
Bioimaging Techniques for Subcellular Localization of Plant Hemoglobins and Measurement of Hemoglobin-Dependent Nitric Oxide Scavenging In Planta Kim H. Hebelstrup,* Erik stergaard-Jensen,† and Robert D. Hill* Contents 1. Introduction 2. Measuring Hemoglobin-Dependent NO Scavenging 3. Techniques for Determination of Subcellular Localization of Plant Hemoglobins 4. Imaging of Hemoglobin-Dependent NO Scavenging in Arabidopsis Plants 5. Engineering of GLB1-GFP/GLB2-GFP Constructs and Microscopic Analysis of A. thaliana Plants Expressing GFP-Tagged Hemoglobin References
596 596 597 598 600 603
Abstract Plant hemoglobins are ubiquitous in all plant families. They are expressed at low levels in specific tissues. Several studies have established that plant hemoglobins are scavengers of nitric oxide (NO) and that varying the endogenous level of hemoglobin in plant cells negatively modulates bioactivity of NO generated under hypoxic conditions or during cellular signaling. Earlier methods for determination of hemoglobin-dependent scavenging in planta were based on measuring activity in whole plants or organs. Plant hemoglobins do not contain specific organelle localization signals; however, earlier reports on plant hemoglobin have demonstrated either cytosolic or nuclear localization, depending on the method or cell type investigated. We have developed two bioimaging techniques: one for visualization of hemoglobin-catalyzed scavenging of NO in specific cells and another for visualization of subcellular * {
Department of Plant Science, University of Manitoba, Winnipeg, Manitoba, Canada Department of Molecular Biology, University of Aarhus, Aarhus, Denmark
Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37030-4
#
2008 Elsevier Inc. All rights reserved.
595
596
Kim H. Hebelstrup et al.
localization of green fluorescent protein-tagged plant hemoglobins in transformed Arabidopsis thaliana plants.
1. Introduction Hemoglobin genes are ubiquitous among plants, and various experiments have indicated that their primary role is to scavenge nitric oxide (NO) (Dordas et al., 2003; Perazzolli et al., 2004). This function has been suggested to play an important role during hypoxic situations, particularly in roots (Igamberdiev and Hill, 2004), but it is also involved with modulating NO signaling during development in normoxic shoots (Hebelstrup et al., 2006). Because hemoglobin is an effective NO-scavenging agent in plants, cellular localization of hemoglobin is of interest. The high reactivity and the short lifetime of NO in a cell environment limit the distribution of NO from its source. Tissue and even intracellular levels may, therefore, be expected to vary considerably. Initial experiments used to determine cellular localization of plant hemoglobins have presented contradictory results depending on the method and model plant used. This chapter describes two bioimaging techniques used for the localization of hemoglobins and NO in plant tissue: (A) detection of tissue- and/or cell-specific NO levels altered by overexpression or silencing of hemoglobin expression in Arabidopsis thaliana and (B) imaging of subcellular localization of plant hemoglobins using transgenic A. thaliana plants expressing green fluorescent protein (GFP)-tagged plant hemoglobin. Arabidopsis thaliana contains three classes of hemoglobin genes (Trevaskis et al., 1997; Watts et al., 2001). Hemoglobin-dependent scavenging of nitric oxide was studied by comparing NO levels in wild-type plants with lines having overexpression of either class 1 hemoglobin Glb1 or class 2 hemoglobin Glb2 (35S:Glb1 or 35S:Glb2) or with silencing (Hg:Glb1) or knockout (Glb2dSpm) of the genes. The engineering of lines has been described elsewhere (Hebelstrup et al., 2006). Engineering of constructs for A. thaliana lines with expression of GFP-tagged Glb1 or Glb2 is described here.
2. Measuring Hemoglobin-Dependent NO Scavenging Electron paramagnetic resonance spectroscopy (Dordas et al., 2003) and chemiluminescent detection of NO in emission gases from leaves (Perazzolli et al., 2004) have been used previously for the measurement of hemoglobin-dependent NO scavenging in planta. These methods are useful for analytical determination of hemoglobin-dependent scavenging of NO
Bioimaging Techniques for Subcellular Localization
597
in a whole organ or even a whole plant. Under some circumstances, such as the involvement of NO in cell signaling, a method that is tissue or cell specific may be preferred. The fluorescent probe 4,5-diaminofluorescein diacetate (DAF-2 DA) has been used to visualize and determine the tissueand cell-specific presence of nitric oxide in plant tissues in several studies. NO levels were determined, for example, in stomatal guard cells loaded with DAF-2 DA to show that NO is a central component in the signal transduction of abscisic acid-induced stomatal closure (Desikan et al., 2002; Garcia-Mata and Lamattina, 2002). DAF-2 DA has also been used to visualize NO during the gravitropic response of horizontally placed roots of Pisum sativum (Hu et al., 2005), showing that NO accumulates nonsymmetrically in a similar fashion to the hormone auxin. DAF-2 DA was used to confirm low NO levels in root and stomatal guard cells of plants with a genetically based impairment of NO synthesis (Guo et al., 2003). We provide here a technique to image cellular and subcellular hemoglobindependent nitric oxide scavenging in specific tissues using the NO-specific fluorescent probe DAF-2 DA. We observed that overexpression of plant hemoglobin effectively scavenges nitric oxide, even when present at high levels. We, therefore, suggest that modulating cellular levels of hemoglobin can be used together with DAF-2 DA to confirm when NO is formed in specific cells and is involved in physiological processes.
3. Techniques for Determination of Subcellular Localization of Plant Hemoglobins No plant hemoglobin genes have been reported to contain nuclear localization signals and, therefore, subcellular localization of plant hemoglobins to the cytosol would be expected. In line with this, hemoglobin is present in the cytosolic fraction of cell extracts of alfalfa root cells (Igamberdiev et al., 2004). However, when using different localization techniques and/or different plant species, nuclear localization has been observed. Electron microscopy of immunogold-labeled hemoglobin in cultured alfalfa root cells indicated a strong tendency to nuclear localization (Seregelyes et al., 2000). Similarly, it has been reported that cotton hemoglobin, tagged with GFP, shows a tendency to nuclear localization when expressed transgenically in onion epidermal cells (Qu et al., 2005). This discrepancy in reported subcellular localization of plant hemoglobins indicates that localization is dynamic and depends on cell type or condition of the cell studied. Since the first report on genetic GFP tagging of specific proteins and its use in studying expression and localization (Chalfie et al., 1994), GFP and derived fluorescent proteins have been used widely in many
598
Kim H. Hebelstrup et al.
cell types and organisms, including plants (Chudakov et al., 2005). The technique reported here uses GFP-tagged hemoglobin in a whole organism, using a nontransgenic strategy, offering a method for imaging native, tagged hemoglobins in live A. thaliana cells under various conditions.
4. Imaging of Hemoglobin-Dependent NO Scavenging in Arabidopsis Plants We use the NO-activated fluorescent probe DAF-2 DA for detection of NO in A. thaliana cells. Plant organs are loaded with DAF-2 DA and prepared for microscopy as described later. The incubation time used gives satisfactory fluorescent levels with A. thaliana leaves and inflorescences. Wounding activates the formation of NO in plant tissue, and strong NO levels are detected around wounds such as the petiole or stem where the organ has been cut from the plant. It is therefore not recommended to cut organs close to the cells of interest. Fluorescence is detected well with either a mercury-lamp fluorescent microscope or a laser-scanning confocal microscope. Fluorescence microscopy is carried out with a Zeiss Axioplan 2 microscope. Filter settings are as follows: excitation window, 460–480 nm; emission window, 505–530 nm. When using these filter settings, autofluorescence is nearly undetectable in untreated A. thaliana epidermal and root cells. The following protocol is used for loading DAF: 1. The plant organ is incubated for 2 h in DAF loading buffer:10 mM MES-Tris (pH 5.6), 0.1 mM CaCl2, and 10 mM KCl containing either (a) 200 mM cPTIO (Calbiochem, Merck KGaA, Darmstadt, Germany), a specific scavenger of NO to provide reference samples without NO, or (b) 0.4 mM sodium nitroprusside (SNP; Sigma-Aldrich Denmark A/S, Denmark) or an equivalent NO donor molecule. 2. DAF-2 DA is added to a final concentration of 10 mM in the DAF loading buffer and incubation is continued for 1 h. DAF-2 DA will react specifically with intracellular NO because DAF-2 DA is only activated for reaction with NO after modification by intracellular esterases. 3. Organs are washed briefly in DAF loading buffer and are then ready for microscopy. Figure 30.1 shows NO-dependent fluorescence of DAF-2 DA in floral buds and young flowers visualized by fluorescence microscopy. Very little NO is detected in wild-type A. thaliana young flowers (see Figs. 30.1C and 30.1D). However, in young flowers and floral buds from plants with silencing of Glb1 (Hg:Glb1), a higher level of nitric oxide is detected (see Figs. 30.1A and 30.1B). Negative controls of Glb1-silenced plants with the
Bioimaging Techniques for Subcellular Localization
599
Figure 30.1 Detection of NO-activated fluorescence of DAF-2 DA in A. thaliana flowers with fluorescence microscopy.When Glb1is silenced, a high endogenous nitric oxide concentration is observed in various organs, including floral buds and young flowers. (A) Transmission white light image of young flowers from plant with silencing of Glb1. (B) Fluorescence image of same flowers as in A. (C) Transmission white light image of a young wild-typeA. thaliana flower. (D) Fluorescence image of the same flower as in C.
addition of cPTIO show a fluorescence level close to wild-type lines, confirming that NO is the component responsible for activation of DAF2 DA fluorescence. Positive controls of wild-type plants with the addition of 1.2 mM SNP have a very high fluorescence level. Interestingly, overexpression of either Glb1 or Glb2 in 35S:Glb1 or 35S:Glb2 lines prevents accumulation of intracellular NO generated from 1.2 mM SNP, demonstrating that ectopic expression of plant hemoglobins can scavenge high levels of intracellular nitric oxide. Hb overexpression may, therefore, be useful in examining situations in which NO effects are difficult to detect. When hemoglobin is overexpressed, NO is scavenged and its effect should disappear. This method has the advantage that NO scavenging can be directed to specific tissues, cells, or situations by choosing a promoter that will be active only in specific cells or under certain conditions. We also used laser-scanning confocal microscopy for detection of nitric oxide-dependent
600
Kim H. Hebelstrup et al.
activation of DAF-2 DA with a Zeiss LSM 510 META microscope. The samples are excited by an argon laser at 488 nm. NO-dependent DAF-2 DA fluorescence is detected in a 505- to 530-nm window. Chlorophyll autofluorescence is detected in a 670- to 690-nm window. Figure 30.2 shows a three-dimensional confocal image of how NO accumulates in the hydathode of a Glb1-silenced (Hg:Glb1) A. thaliana leaf. Gene expression studies have shown that hemoglobin is expressed specifically at hydathodes in A. thaliana leaves (Hebelstrup et al., 2006) and, in line with this, no NO accumulation is detected in the hydathodes of wild-type A. thaliana leaves.
5. Engineering of GLB1-GFP/GLB2-GFP Constructs and Microscopic Analysis of A. thaliana Plants Expressing GFP-Tagged Hemoglobin Constructs for GFP tagging of A. thaliana hemoglobin gene classes 1 and 2 (Glb1 and Glb2) are prepared by first purifying total RNA from seedlings of A. thaliana (Col-0) with a commercial kit (RNeasy Plant Mini Kit, Qiagen) according to the manufacturer’s instructions. cDNA is constructed with M-MLV reverse transcriptase (Promega, WI) using oligo(dT) primers (Qiagen). Glb1 and Glb2 open reading frames are amplified by polymerase chain reaction (PCR) from the cDNA using the following primers: 50 -TCTAGAGGTTGTGAAATATTATGGAG-30 and 50 -TCTA GAGTTGGAAAGATTCATTTCAG-30 for Glb1 and 50 -TCTAGATGG GAGAGATTGGGTTTACA-30 and 50 -TCTAGACTCTTCTTGTTTC ATCTCGG-30 for Glb2. PCR fragments are ligated into the vector pCRTOPO and cloned in Escherichia coli using a commercial kit (TOPO TA cloning kit, Invitrogen) according to the manufacturer’s description. pCRTOPO plasmids containing cloned Glb1 or Glb2 are digested with XbaI, and fragments are ligated into the XbaI site in the T-DNA region of the modified Ti plasmid pPZP211–35S-GFP-pANOS (Anderssen et al., 2005) to generate the two constructs 35S::GLB1-GFP and 35S::GLB2-GFP. These constructs contain the open reading frame of either Glb1 or Glb2 followed by a GFPcoding sequence, giving rise to C-terminal-tagged hemoglobin when expressed in a plant host. Constructs also contain a selection marker for kanamycin resistance and should be useful for transformation methods, including direct DNA bombardment and transformation of plant tissue with Agrobacterium tumefaciens. We tested the constructs in A. thaliana by first cloning the constructs in E. coli (DH10) and then subsequently in A. tumefaciens for eventual transformation of A. thaliana plants by the floral dip method (Clough and Bent, 1998). Positive first-generation transformants (T1) are selected by growth in medium containing 50 mg/ml kanamycin.
Bioimaging Techniques for Subcellular Localization
601
Figure 30.2 Three-dimensional projection of confocal scanning microscopic imaging of an enlarged hydathode from a Glb1-silenced plant. An elevated NO level is found in a spherical region consisting of chlorophyll-less cells at the tip of the hydathode. Chlorophyll (A) and NO (B) were detected as described in the text. (C) A merged image is of A and B.
602
Kim H. Hebelstrup et al.
Figure 30.3 Confocal image of roots cells of transgenic A. thaliana plants with ectopic expression of GFP-tagged hemoglobins Glb2-GFP (A) by the 35S:Glb1-GFP or Glb2GFP (B) by the 35S:Glb2-GFP constructs. Both types of hemoglobin localize throughout the cytosol and nucleoplasm of the cells. However, a higher concentration may be seen in the nucleus.
Bioimaging Techniques for Subcellular Localization
603
Expression of GFP-tagged hemoglobin in T2 seedlings is confirmed by examination with both fluorescence and confocal microscopy. In epidermal cells, strong fluorescence is detected in stomatal guard cells. In roots, fluorescence is found in all cell types. GLB1-GFP and GLB2-GFP show similar localization patterns. Fluorescence is observed throughout the cytosol and nucleus; however, it is absent in the nucleolus (Fig. 30.3). There appears to be a higher concentration in the nucleus than in the cytosol. Lines with various levels of fluorescence are examined, and the subcellular distribution pattern is similar in all lines. In summary, this chapter presented two bioimaging techniques:one showing that hemoglobin-catalyzed scavenging of NO can be altered in planta by the modulation of intracellular hemoglobin levels and another demonstrating that GFP-tagged hemoglobins can be used for the visualization of subcellular localization of plant hemoglobins. These techniques present resources for studying the effect of NO metabolism in plant cells and for studying further the function of plant hemoglobins.
REFERENCES Anderssen, S. U., Cvitanich, C., Gronlund, M., Busk, H., Jensen, D. B., and Jensen, E. O. (2005). Vectors for reverse genetics and expression analysis. In ‘‘Lotus Japonicus Handbook’’ (A. J. Marquez, ed.). Kluwer Academic, New York. Chalfie, M., Tu, Y., Euskirchen, G., Ward, W. W., and Prasher, D. C. (1994). Green fluorescent protein as a marker for gene expression. Science 263, 802–805. Chudakov, D. M., Lukyanov, S., and Lukyanov, K. A. (2005). Fluorescent proteins as a toolkit for in vivo imaging. Trends Biotechnol. 12, 605–613. Clough, S. J., and Bent, A. F. (1998). Floral dip: A simplified method for Agrobacteriummediated transformation of Arabidopsis thaliana. Plant J. 16, 735–743. Desikan, R., Griffiths, R., Hancock, J., and Neill, S. (2002). A new role for an old enzyme: Nitrate reductase-mediated nitric oxide generation is required for abscisic acid-induced stomatal closure in Arabidopsis thaliana. Proc. Natl. Acad. Sci. USA 99, 16314–16318. Dordas, C., Hasinoff, B. B., Igamberdiev, A. U., Manac’h, N., Rivoal, J., and Hill, R. D. (2003). Expression of a stress-induced hemoglobin affects NO levels produced by alfalfa root cultures under hypoxic stress. Plant J. 35, 763–770. Garcia-Mata, C., and Lamattina, L. (2002). Nitric oxide and abscisic acid cross talk in guard cells. Plant Physiol. 128, 790–792. Guo, F. Q., Okamoto, M., and Crawford, N. M. (2003). Identification of a plant nitric oxide synthase gene involved in hormonal signaling. Science 302, 100–103. Hebelstrup, K. H., Hunt, P., Dennis, E., Jensen, S. B., and Jensen, E. O. (2006). Hemoglobin is essential for normal growth of Arabidopsis organs. Physiol. Plant. 127, 157–166. Hu, X., Neill, S. J., Tang, Z., and Cai, W. (2005). Nitric oxide mediates gravitropic bending in soybean roots. Plant Physiol. 137, 663–670. Igamberdiev, A. U., and Hill, R. D. (2004). Nitrate, NO and haemoglobin in plant adaptation to hypoxia:An alternative to classic fermentation pathways. J. Exp. Bot. 55, 2473–2482.
604
Kim H. Hebelstrup et al.
Igamberdiev, A. U., Seregelyes, C., and Manac’h, N., and Hill, R. D. (2004). NADHdependent metabolism of nitric oxide in alfalfa root cultures expressing barley hemoglobin. Planta 219, 95–102. Perazzolli, M., Dominici, P., Romero-Puertas, M. C., Zago, E., Zeier, J., Sonoda, M., Lamb, C., and Delledonne, M. (2004). Arabidopsis nonsymbiotic hemoglobin AHb1 modulates nitric oxide bioactivity. Plant Cell 16, 2785–2794. Qu, Z. L., Wang, H. Y., and Xia, G. X. (2005). GhHb1:A nonsymbiotic hemoglobin gene of cotton responsive to infection by Verticillium dahliae. Biochim. Biophys. Acta Gene Struct. Express. 1730, 103–113. Seregelyes, C., Mustardy, L., Ayaydin, F., Sass, L., Kovacs, L., Endre, G., Lukacs, N., Kovacs, I., Vass, I., Kiss, G. B., Horvath, G. V., and Dudits, D. (2000). Nuclear localization of a hypoxia-inducible novel non-symbiotic hemoglobin in cultured alfalfa cells. FEBS Lett. 482, 125–130. Trevaskis, B., Watts, R. A., Andersson, C. R., Llewellyn, D. J., Hargrove, M. S., Olson, J. S., Dennis, E. S., and Peacock, W. J. (1997). Two hemoglobin genes in Arabidopsis thaliana:The evolutionary origins of leghemoglobins. Proc. Natl. Acad. Sci. USA 94, 12230–12234. Watts, R. A., Hunt, P. W., Hvitved, A. N., Hargrove, M. S., Peacock, W. J., and Dennis, E. S. (2001). A hemoglobin from plants homologous to truncated hemoglobins of microorganisms. Proc. Natl. Acad. Sci. USA 98, 10119–10124.
C H A P T E R
T H I R T Y- O N E
Use of Recombinant Iron-Superoxide Dismutase as A Marker of Nitrative Stress Estı´baliz Larrainzar,* Estı´baliz Urarte,† In˜igo Auzmendi,* Idoia Ariz,† Cesar Arrese-Igor,* Esther M. Gonza´lez,* and Jose F. Moran† Contents 1. Introduction 2. Immunodetection of Nitrated Proteins: Metal-Mediated Tyrosine Nitration of BSA 3. Tyrosine Nitration of Purified Recombinant Vu_FeSOD Affects its Enzymatic Activity 4. Tyrosine Nitration in Vu_FeSOD can be Estimated Using Antibodies Against 3-Nitrotyrosine 5. SIN-1-Dependent Vu_FeSOD Nitration can be Detected by the Loss of Enzymatic Activity Acknowledgments References
606 608 610 612 612 616 616
Abstract Superoxide dismutases (SODs; EC 1.15.1.1) are a group of metalloenzymes which are essential to protect cells under aerobic conditions. In biological systems, it has been reported that SODs and other proteins are susceptible to be attacked by peroxynitrite (ONOO) which can be originated from the reaction of nitric oxide with superoxide radical. ONOO is a strong oxidant molecule capable of nitrating peptides and proteins at the phenyl side chain of the tyrosine residues. In the present work, bovine serum albumin (BSA) and recombinant ironsuperoxide dismutase from the plant cowpea (Vu_FeSOD) are used as target molecules to estimate ONOO production. The method employs the compound
* {
Departamento de Ciencias del Medio Natural, Universidad Pu´blica de Navarra, Campus de Arrosadı´a, E-31006 Pamplona, Navarre, Spain Instituto de Agrobiotecnologı´a, Universidad Pu´blica de Navarra-CSIC-Gobierno de Navarra, Campus de Arrosadı´a, E-31006 Pamplona, Navarre, Spain
Methods in Enzymology, Volume 437 ISSN 0076-6879, DOI: 10.1016/S0076-6879(07)37031-6
#
2008 Elsevier Inc. All rights reserved.
605
606
Estı´baliz Larrainzar et al.
SIN-1, which simultaneously generates NO and O2 in aerobic aqueous solutions. First, assay conditions were optimized incubating BSA with different concentrations of SIN-1, and at a later stage, the effect on the tyrosine nitration and catalytic activity of Vu_FeSOD was examined by in-gel activity and spectrophotometric assays. Both BSA and Vu_FeSOD are nitrated in a dose-dependent manner, and, at least in BSA nitration, the reaction seems to be metal catalyzed.
1. Introduction Nitric oxide (NO) has emerged as a key signaling molecule in plants during the past decade.NO has been shown to be involved in important plant physiological processes such as plant defence (Delledonne et al., 1998; Durner et al., 1998), stomatal closure (Garcı´a-Mata and Lamattina, 2001), root development (Pagnussat et al., 2002), and iron homeostasis (Murgia et al., 2002) among many others. In aerobic aqueous solutions,NO can react with molecular oxygen (O2) to form nitrogen dioxide (NO2), which is further degraded to nitrite and nitrate (NO2/NO3). In the presence of transition metals (Fe, Cu, Zn, and Mn), NO can form metal–nitrosyl complexes (Met-NO). NO and its cation NOþ can also react with cysteine residues, resulting in S-nitrosylation (Stamler et al., 2001). Additionally, the diffusion-controlled reaction between NO and superoxide (O2) can produce peroxynitrite (ONOO) (Feelisch et al., 1989). Peroxynitrite is a strong oxidant, able to react with a variety of biomolecules, and it is capable of nitrating peptides and proteins at the phenyl side chain of the tyrosine residues (Beckman et al., 1990; Ischiropoulos et al., 1992a; Reiter et al., 2000). In biomedical research, protein tyrosine nitration has been extensively studied as it can alter protein function and it is associated to acute and chronic disease states (for a review, Radi, 2004). In contrast, despite some initial works (Saito et al., 2006; Valderrama et al., 2007), protein nitration in plants is a largely unexplored field. Bovine serum albumin (BSA) has been widely used as a model for in vitro protein nitration (Kamisaki et al., 1998; Khan et al., 1998; Thomas et al., 2002, among others). The presence of nitrated residues in a protein can be analyzed by well-established mass spectrometry-based methods, such as high performance liquid chromatography (HPLC) (Crow and Ischiropoulos, 1996; Shigenaga et al., 1997), liquid chromatography/mass spectrometry (Yi et al., 2000), gas chromatography/mass spectrometry (Sarver et al., 2001; Schwedhelm et al., 1999), and ELISA (Khan et al., 1998). The limitation of these approaches is the requirement of expensive and sophisticated
Fe-Superoxide Dismutase as a Marker of Nitrative Stress
607
instrumentation. One of the most commonly applied method is the immunodetection of nitrated proteins using specific antibodies against 3-nitrotyrosine (NO2-Tyr) (MacMillan-Crow and Thompson, 1999; Viera et al., 1999). In animal systems, superoxide dismutases (SODs; EC 1.15.1.1) have been described as target of tyrosine nitration both in vitro and in vivo. Pioneered work by Ischiropoulos et al. (1992a) reported the nitration of bovine Cu,ZnSOD by peroxynitrite without significant loss of enzymatic activity. However, peroxynitrite was reported to inactivate human MnSOD based on single nitration at Tyr34 (MacMillan-Crow et al., 1996; Yamakura et al., 1998). Regarding FeSOD, which is so far only described in plant, protists, and prokaryotic organisms, contradictory results have been reported. Ischiropoulos et al. (1992a) observed an inactivation of E. coli FeSOD in a peroxynitrite dose-dependent manner, whereas for Soulere et al. (2001) its enzymatic activity remained unaffected after bolus addition of peroxynitrite up to 7 mM. Symbiotic legume nodules can be considered ideal model systems for the analysis of protein nitration due to its high metal and metalloprotein content, and high production rate of reactive oxygen and nitrogen species (ROS and RNS) (Becana and Klucas, 1992; Dalton, 1995; Pauly et al., 2006; Puppo et al., 1981). Legume nodules possess several types of SODs, metalloenzymes whose primary activity is the dismutation of the superoxide radical anion into hydrogen peroxide and molecular oxygen (Fridovich, 1995). Among them, a FeSOD is present in the cytosol of infected nodule cells of cowpea (Vigna unguiculata), which is an unusual localization for a plant FeSOD (Moran et al., 2003). This FeSOD isoenzyme, designated Vu_FeSOD, has been expressed in bacteria, purified to homogeneity, crystallized, and the 3-D structure has been established (Moran et al., 2003; Mun˜oz et al., 2005). Nowadays, the study of RNS is having a pivotal role in biological systems. The use of proteins as markers for peroxynitrite-dependent oxidation is a useful method that may be applicable both in vitro and in vivo. SODs have been shown to be nitrated in vivo in several model systems, including rat alveolar activated macrophage (Ischiropoulos et al., 1992b) and human renal allografts (MacMillan-Crow et al., 1996). The use of an in vitro model could be useful for the study of proteins which interact in the NO homeostasis. In this work, both BSA and recombinant Vu_FeSOD are employed as targets for protein nitration. SIN-1 (3-morpholinosydnonimine-N-ethylcarbamide), which simultaneously generates NO and O2 in aerobic aqueous solutions (Feelisch et al., 1989; Hogg et al., 1992), is used as nitrating agent. The influence of in vitro tyrosine nitration on the enzymatic activity of a plant FeSOD and the effect of metal chelating on the nitration process have been analyzed.
608
Estı´baliz Larrainzar et al.
2. Immunodetection of Nitrated Proteins: Metal-Mediated Tyrosine Nitration of BSA Assay conditions were first optimized using commercial BSA (SigmaAldrich, St. Louis, Mo, USA). Protein nitration elicited by the reaction between NO and O2 was examined using SIN-1, at concentrations ranging from 100 mM to 1 mM. At neutral pH, the release of NO and O2 from SIN-1 is a relatively slow process (t1/2 ¼ 40 min; Lomonosova et al., 1998). Under these conditions, it has been estimated that within approximately two hours SIN-1 is degraded and equimolar amounts of ONOO are generated (Martin-Romero et al., 2004). BSA protein aliquots (0.3 mg ml1) were incubated in reaction buffer (50 mM sodium phosphate, 1 mM EDTA, pH 7.4) containing different SIN-1 concentrations at 37 C for 2 h. SIN-1 (1 mg aliquots; Invitrogen, Carlsbad, CA, USA) was freshly dissolved and maintained light-protected on ice before use. After incubation, proteins (1 mg per lane) were separated on 10% (w/v) SDS-PAGE and electroblotted onto polyvinylidene difluoride membranes. Membranes were blocked with 5% (w/v) BSA in Tris-buffered saline overnight at 4 C. Protein tyrosine nitration was immunochemically detected using polyclonal anti-3-nitrotyrosine antibody (1 mg protein mL1) purchased from Upstate Biotechnology (Lake Placid, NY). As secondary antibody, goat anti-rabbit IgG alkaline phosphatase conjugate (1:5000, v/v; SigmaAldrich) was used. Cross-reacting protein bands were visualized using nitroblue tetrazolium and 5-bromo-4-chloro-3-indolyl phosphate (BCIP/ NBT, Sigma-Aldrich) as substrates. To confirm equal protein loading, replica gels were run in parallel and stained using Gel-Code Blue Stain reagent (Pierce Biotechnology, Inc., Rockford, USA). Consistent with previous studies (Beckman et al., 1994; Thomas et al., 2002), a concentration-dependent NO2-Tyr signal was detected following BSA exposure to SIN-1 (Fig. 31.1). The lower limit of detection with this system was 0.3 mM, with a maximum intensity at a concentration of 1 mM SIN-1. In the absence of SIN-1, no signal was detectable, indicating the absence of endogenous BSA nitration. NO2-Tyr immunodetection was further validated using dithionite, which reduces nitrotyrosine residues to aminotyrosine, consequently, preventing antibody binding (Fig. 31.2A, lane 3). Additionally, the influence of free metal ions in the solution on the nitration reaction was addressed. Aliquots were initially incubated with desferrioxamine, a soluble iron-chelating agent that binds iron in a noncatalytic form. Interestingly, when desferrioxamine was present in the reaction buffer, nitration was completely inhibited (Fig. 31.2A, lane 4), suggesting that SIN-1-derived BSA tyrosine nitration is catalyzed by trace amounts of free iron present in the reaction buffer or pre-adsorbed to the protein.
609
Fe-Superoxide Dismutase as a Marker of Nitrative Stress
1
2
3
4
5
6
7
A
B
Figure 31.1 Immunodetection of tyrosine nitrated bovine serum albumin (BSA). (A) BSA was incubated with SIN-1 at different concentrations: lane 1, control, BSA incubated without SIN-1; from lane 2 to lane 7, BSA incubated with increasing concentrations of SIN-1: 0.1, 0.2, 0.3, 0.4, 0.5, and 1 mM. (B) Replica gel stained with Coomassie blue brilliant.1 mg protein was loaded into each lane.
A
1
2
3
4
B
Figure 31.2 Dose-dependent effect of SIN-1 and desferrioxamine on bovine serum albumin (BSA) tyrosine nitration. (A) Result of the immunoblot against 3-nitrotyrosine. Lane 1: BSA; lane 2: nitrated BSA (1 mM SIN-1); lane 3: nitrated BSA (1 mM SIN-1) and subsequently reduced with dithionite; lane 4: nitrated BSA (1 mM SIN-1) incubated with 10 mM desferrioxamine. (B) Replica gel stained with Coomassie blue brilliant.1 mg of protein was loaded into each lane.
610
Estı´baliz Larrainzar et al.
3. Tyrosine Nitration of Purified Recombinant Vu_FeSOD Affects its Enzymatic Activity Once the experimental conditions for in vitro nitration and nitrotyrosine immunodetection were established, the influence of tyrosine nitration on the catalytic activity of Vu_FeSOD was analyzed. Target recombinant Vu_FeSOD was obtained as described in Moran et al. (2003) with modifications. Transformed E. coli BL21(DE3) cells, containing the pET-28a(þ)::Vu_FeSOD construct, were grown at 37 C in Luria-Bertani broth supplemented with kanamycin (100 mg ml1) until A600 reached 0.5. After induction by Isopropyl-b-D-thiogalacto-pyramoside (IPTG), cells were pelleted and total protein was extracted on a buffer containing 20 mM sodium phosphate, pH 7.5, 500 mM NaCl, 0.1% (v/v) Triton X-100, DNase I (25 mg ml1), RNase (50 mg ml1), pepstatin (1.25 mg ml1), and leupeptin (1.25 mg ml1). The cell suspension was incubated with lysozyme (2 mg ml1) at 4 C for 15 min with gentle stirring, sonicated with three 30-second pulses, and cleared by centrifugation at 48,000g at 4 C for 20 min. The supernatant was loaded onto a 5-ml Ni-nitrilotriacetic acid resin-based column (His-Trap Chelating, GE Healthcare, Uppsala, Sweden) pre-equilibrated with 20 mM sodium phosphate, pH 7.5. A linear gradient from 50 to 500 mM imidazole was applied, and recombinant Vu_FeSOD eluted at 160 mM imidazole. Collected fractions were sterilized through a 0.20-m m filter syringe. Subsequently, they were overnight dialyzed at 4 C against 10 mM sodium phosphate buffer (pH 7.0) containing 1 mg thrombin (SigmaAldrich) in order to remove the (His)6 tag. Protein purification was assessed by gel electrophoresis and accordingly, fractions were pooled. As a second purification step, pooled fractions were loaded onto a 5-ml Hi-Trap DEAE FF cellulose column (GE Healthcare), equilibrated with sodium phosphate 10 mM (pH 7.0). A 10-column-volume linear sodium phosphate gradient (10–100 mM) was applied, and Vu_FeSOD eluted at 50 mM. Protein concentration was determined by a dye-binding assay (Bio-Rad Laboratories, Hercules, CA) using BSA as a standard. Vu_FeSOD was subjected to in vitro nitration, SDS-PAGE electrophoresis and immunodetection of NO2-Tyr, in a similar manner to BSA (Fig. 31.3. A and B). While recombinant Vu_FeSOD showed no detectable levels of endogenous nitration (Fig. 31.3A, lane 1), after incubation with 1 mM SIN1, an intense NO2-Tyr signal was detectable (Fig. 31.3A, lane 2). As a negative control for the immunodetection, aliquots of nitrated Vu_FeSOD were reduced with a trace of sodium dithionite (Fig. 31.3A, lane 3). Dithionite reduction clearly diminished the signal, validating the detection system. Due to the intense nitration process, a faint band could be yet observed. To test whether tyrosine nitration influenced the enzymatic activity of Vu_FeSOD, in-gel SOD activity assays were performed as described in
611
Fe-Superoxide Dismutase as a Marker of Nitrative Stress
1
2
3
A
B
C
Figure 31.3 Immunochemical detection of tyrosine nitration onVu__ FeSOD and activity assay. (A) Immunoblot against 3-nitrotyrosine. Lane 1: Control Vu__ FeSOD without SIN-1; lane 2: nitrated Vu__ FeSOD (1 mM SIN-1); lane 3: nitrated Vu__ FeSOD (1 mM SIN-1) and subsequently reduced with dithionite. (B) Replica gel stained with Coomassie blue brilliant. (C) Replica gel stained for SOD activity assay. 1 mg of protein was loaded into each lane.
Beauchamp and Fridovich (1971) (Fig. 31.3C). These assays are based on SOD inhibition of the reduction of NBT by photochemically generated superoxide radicals. Treated recombinant Vu_FeSOD samples (1 mg per lane) were run in 15% (w/v) native-PAGE at 200 V. The electrophoresis was run for 2 h, 1 h after the bromophenol blue reaches the end of the gel. This process allowed obtaining an improved mobility of SODs into the gels. SIN-1 (1 mM) treatment induced the nitration of Vu_FeSOD, as detected by Western blot using anti-nitrotyrosine antibody (Fig. 31.3A, lane 2). After electrophoresis, gels were first incubated in reaction buffer (50 mM sodium phosphate buffer, pH 7.8) for 30 min. Second, they were transferred into reaction buffer containing 0.5 mM NBT and further incubated for 20 min. Lastly, they were incubated in reaction buffer supplemented with 0.03 mM riboflavin and 0.2% (v/v)
612
Estı´baliz Larrainzar et al.
N,N,N’,N’,-Tetramethyl-ethylenediamine (TEMED) for 20 min. All incubations were performed under darkness. For visualization of SOD activity bands, gels were exposed to white light for 2–5 min. SIN-1 treatment did not seem to degrade Vu_FeSOD, as seen in the Coomassie stained gel (Fig. 31.3B). However, SIN-1 caused a clear reduction of Vu_FeSOD activity (Fig 31.3C, lane 2). Dithionite reduction suppressed the binding of the antibody, but did not reverse the inactivation of Vu_FeSOD activity. Dithionite reduction slightly modified the ratio mass/charge of the protein (Fig. 31.3C, lane 3). To better visualize the slight differences in mobility among the control Vu_FeSOD, its nitrated form and the nitrated-reduced form, this gel was run for further 30 min (Fig. 31.3C).
4. Tyrosine Nitration in Vu_FeSOD can be Estimated Using Antibodies Against 3-Nitrotyrosine To further investigate the influence of SIN-1 concentration on the level of protein nitration, Vu_FeSOD was incubated with increasing amounts of SIN-1 (0.2, 0.4, 0.6, 0.8, and 1 mM ) as described above for BSA. Western blot using anti-3-nitrotyrosine antibodies indicates that Vu_FeSOD is nitrated in a SIN-1 dose-dependent manner (Fig. 31.4). Thus, nitration starts to be immunochemically detectable at a concentration of 0.4 mM. The signal increases proportionally to the concentration of the nitrating agent, reaching saturation at 0.8 mM of SIN-1. Thus, incubation with SIN-1 can be considered a useful system for in vitro Vu_FeSOD nitration. The slow release of NO and O2 represents an alternative to direct addition of synthetic peroxynitrite, as the latter has been shown to not accurately reflect the chemistry of this species under biological conditions (Espey et al., 2002; Thomas et al., 2002). Unlike BSA, Vu_FeSOD nitration is not fully suppressed by addition of desferrioxamine (Fig. 31.4, lane 7). This could be related to the fact that Vu_FeSOD keeps Fe at its catalytic center during the incubation with desferrioxamine. Our experiments incubating mixtures of Vu_FeSOD and desferrioxamine indicated that the Fe atom is not released from the active center (data not shown), and SOD in-gel activity remains stable.
5. SIN-1-Dependent Vu_FeSOD Nitration can be Detected by the Loss of Enzymatic Activity The in vitro nitrated Vu_FeSOD samples were further subjected to SOD activity assays. In-gel SOD activity was visualized using in-gel staining after separation of protein in 15% (w/v) native-PAGE as described above.
613
Fe-Superoxide Dismutase as a Marker of Nitrative Stress
1
2
3
4
5
6
7
8
A
B
Figure 31.4 Dose-dependent effect of SIN-1 on Vu__ FeSOD tyrosine nitration. (A) Immunoblot against 3-nitrotyrosine. (B) Replica gel stained with Coomassie blue brilliant. Lane1: ControlVu__ FeSOD without SIN-1; lane 2: nitrated Vu__ FeSOD (0.2 mM SIN-1); lane 3: nitrated Vu__ FeSOD (0.4 mM SIN-1); lane 4: nitrated Vu__ FeSOD (0.6 mM SIN-1); lane 5: nitrated Vu__ FeSOD (0.8 mM SIN-1); lane 6: nitrated Vu__ FeSOD (1 mMSIN-1); lane 7: nitratedVu__ FeSOD (1 mM SIN) incubated with10 mMdesferrioxamine; lane 8: nitrated Vu__ FeSOD (1 mM SIN-1) and subsequently reduced with dithionite.1 mg of protein was loaded into each lane.
Inactivation of the enzyme occurred in a dose-dependent manner as seen by the gradual loss of SOD activity in the gel (Fig. 31.5A). This observation indicates that the loss of enzymatic activity can be used as an estimation of the level of Vu_FeSOD protein nitration. The loss of enzymatic activity was more evident the higher the SIN-1 concentrations applied. To estimate the loss of enzymatic activity in the in-gel assay, densitometry analyses were performed using the Quant 1 software in GelDoc 2000 (Bio-Rad). Control samples incubated in the absence of the nitrating agent were considered as 100% of the activity. Densitometry analysis showed that after incubation with 1 mM SIN-1, an average of 57% of the SOD activity was lost (n ¼ 3). Alternatively, total SOD activity was assayed using the spectrophotometrical method based on the ability of SODs to inhibit the reduction of ferric cytochrome c by the xanthine–xanthine oxidase system (McCord and Fridovich, 1969). Briefly, reduction of ferricytochrome c was monitored at 25 C by following the increase in absorbance at 550 nm for 2 min.
614
Estı´baliz Larrainzar et al.
1
2
3
4
5
6
7
8
9
A
B
Figure 31.5 Dose-dependenteffectof SIN-1onin-gelVu_ FeSODactivity.(A)SODactivity assay. Lane 1: Control Vu_ FeSOD without SIN-1; lane 2: nitrated Vu_ FeSOD (0.2 mM SIN-1); lane 3: nitrated Vu_ FeSOD (0.4 mM SIN-1); lane 4: nitrated Vu_ FeSOD (0.6 mM SIN-1); lane5: nitratedVu_ FeSOD (0.8 mMSIN-1); lane 6: nitratedVu_ FeSOD (1 mMSIN-1); lane7:controlVu_ FeSODwithoutSIN-1;lane8:nitratedVu__ FeSOD(1 mMSIN-1)incubated with 10 mM desferrioxamine; lane 9: nitrated Vu_ FeSOD (1 mM SIN-1) and subsequently reduced with dithionite. (B) Replica gel stained with Coomassie blue brilliant. 1 mg of proteinwasloaded into each lane.
The reaction mixture contained reaction buffer (50 mM phosphate buffer, 0.1 mM Na2EDTA, pH 7.8), 1 mM xanthine, and 1 mM horse heart cytochrome c (Sigma-Aldrich). Samples were preincubated at 25 C and kept in darkness. Reaction was started by the addition of 20 ml diluted xanthine oxidase. Commercial milk xanthine oxidase (Sigma-Aldrich) was 10-fold diluted in reaction buffer in order to reach 0.1 units of absorbance variation within the blank mix. Sample volumes for control and SIN-1treated Vu_FeSOD proteins were adjusted to obtain an 50% reduction on the activity rate, since higher rates of SOD activity could lead to a nonlinear kinetic or saturation of the enzymatic activity. Hence, 0.4 mg of Vu_FeSOD were finally employed. The spectrophotometric analysis confirmed the gradual inactivation of the enzyme in a dose-dependent manner (Fig. 31.6). The levels of Vu_FeSOD inactivation ranged from 17.8% (for a concentration of 0.2 mM SIN-1) to 61.6% (for 1 mM SIN-1), in agreement with the results obtained in the in-gel SOD assay (Fig. 31.5A). The presence of desferrioxamine partially suppressed the inactivation of the Vu_FeSOD, both in the in-gel SOD assay (Fig. 31.5, lane 8) and in the spectrophotometric analysis (Fig. 31.6). This could be related to its inability to avoid the metal-catalyzed nitrative attack at the active center of the protein. Regarding the molecular mechanism of peroxynitrite-mediated nitration of SODs, Ischiropoulos et al. (1992a) suggested that peroxynitrite in the transconfiguration could access the active center of Cu,ZnSOD, transferring one electron to the copper atom. These authors suggested that the metal at the catalytic center may catalyze the nitration and inactivation of the Fe- and
615
Fe-Superoxide Dismutase as a Marker of Nitrative Stress
SOD activity (%)
100 75 50 25 0
0
0.2
0.4
0.6 0.8 [SIN-1] (mM)
1
1 + Desf 1 + Dit
Figure 31.6 Effect of SIN-1 onVu__ FeSOD activity assayed by the spectrophotometric method. Equivalent concentrations of SIN-1 to Fig. 31.5 were used. Desferrioxamine (10 mM) and a trace of dithionite were used where stated. Activity of control samples were taken as 100% corresponded to 1.5 units of SOD activity. One unit of SOD activity was the amount of the enzyme which inhibited the O2 -dependent reduction of ferricytochrome c by 50%.
MnSOD (Ischiropoulos et al., 1992a). Multiple Vu_FeSOD residues may be susceptible to nitration. Vu_FeSOD contains seven tyrosine residues, one of them in particular, Tyr51, located only at 5.69 A˚ from the catalytic iron center, but not bound to it (Mun˜oz et al., 2005). Nitration often occurs at selective tyrosyl residues possibly biased by proximity to glycyl turns and charged residues (Souza et al., 1999). It is tentative to think that the addition of a nitro group to a Tyr residue near the catalytic center causes enzymatic inhibition by steric interference, weakening the hydrogen-bond network as it has been proposed for nitrated human MnSOD. A possible electrostatic effect of the nitro group in the nitro-Tyr51 on the redox potential to achieve the catalysis in the active center cannot be discarded (Quint et al., 2006). Previous reports determined the number of nitrotyrosine residues within E. coli FeSOD (Soulere et al., 2001) proportionally to the amount of peroxynitrite applied. At 1 mM peroxynitrite concentration, 2 nitrated Tyr residues per E. coli FeSOD subunit were reported (Soulere et al., 2001). FeSODs are highly conserved at the amino acid level and the structures of E. coli and Vu_FeSOD are very similar, both containing seven Tyr residues. Based on these estimations, we would expect one or two nitrated Tyr residues in Vu_FeSOD under the conditions analyzed. Future mass spectrometric analysis would help to determine whether Tyr51 is the main target of tyrosine nitration and responsible for the loss of activity observed. In conclusion, a method for in vitro nitration of FeSOD using SIN-1 is described. Target Vu_FeSOD can be easily produced and purified to homogeneity. Using Vu_FeSOD protein nitration can be readily estimated
616
Estı´baliz Larrainzar et al.
by Western blot analysis, in-gel and spectrophotometric SOD activity assays. Given the increasing interest on NO and peroxynitrite chemistry and its implication in biological processes, this assay may have future biological applications.
ACKNOWLEDGMENTS E. Larrainzar and JFM ( J. F. Moran) are the recipients of a ‘‘Formacio´n de Profesorado Universitario’’ fellowship and a contract within the Ramo´n y Cajal program, respectively, from the Spanish Ministry of Education and Science (MEC). E Urarte and I Auzmendi thank the Dept. of Industry of the Government of Navarre for the fellowship within the Programa de formacio´n de te´cnicos especializados. I. Ariz is the recipient of a PhD fellowship from the Public University of Navarre. This work was supported by grant from the Government of Navarre, Spain (Res. 57/2007), and from DGI-MEC, Spain (grant AGL 2007-64432/AGR).
REFERENCES Beauchamp, C. O., and Fridovich, I. (1971). Superoxide dismutase: Improved assays and an assay applicable to acrylamide gels. Anal. Biochem. 44, 276–287. Becana, M., and Klucas, R. V. (1992). Transition metals in legume root nodules: Irondependent free radical production increases during nodule senescence. Proc. Natl. Acad. Sci. USA 89, 8958–8962. Beckman, J. S., Beckman, T. W., Chen, J., Marshall, P. A., and Freeman, B. A. (1990). Apparent hydroxyl radical production by peroxynitrite: Implications for endothelial injury from nitric oxide and superoxide. Proc. Natl. Acad. Sci. USA 87, 1620–1624. Beckman, J. S., Chen, J., Ischiropoulos, H., and Crow, J. P. (1994). Oxidative chemistry of peroxynitrite. Methods Enzymol. 233, 229–240. Crow, J. P., and Ischiropoulos, H. (1996). Detection and quantitation of nitrotyrosine residues in proteins: In vivo marker of peroxynitrite. Methods Enzymol. 269, 185–194. Dalton, D. A. (1995). Antioxidant defences of plants and fungi. In ‘‘Oxidative Stress and Antioxidants Defences in Biology’’ (S. Ahmad, ed.), pp. 298–355. Chapman and Hall, New York. Delledonne, M., Xia, Y., Dixon, R. A., and Lamb, C. (1998). Nitric oxide functions as a signal in plant disease resistance. Nature 394, 585–588. Durner, J., Wendehenne, D., and Klessig, D. F. (1998). Defence gene induction in tobacco by nitric oxide, cyclic GMP, and cyclic ADP-ribose. Proc. Natl. Acad. Sci. USA 95, 10328–10333. Espey, M. G., Xavier, S., Thomas, D. D., Miranda, K. M., and Wink, D. A. (2002). Direct real-time evaluation of nitration with green fluorescent protein in solution and within human cells reveals the impact of nitrogen dioxide vs. peroxynitrite mechanisms. Proc. Natl. Acad. Sci. USA 99, 3481–3486. Feelisch, M., Ostrowski, J., and Noack, E. (1989). On the mechanism of NO release from sydnonimines. J. Cardiovasc. Pharmacol. 14(Suppl. 11), S13–S22. Fridovich, I. (1995). Superoxide radical and superoxide dismutases. Annu. Rev. Biochem. 64, 97–112. Garcı´a-Mata, C., and Lamattina, L. (2001). Nitric oxide induces stomatal closure and enhances the adaptive plant responses against drought stress. Plant Physiol. 126, 1196–1204.
Fe-Superoxide Dismutase as a Marker of Nitrative Stress
617
Hogg, N., Darley-Usmar, V. M., Wilson, M. T., and Moncada, S. (1992). Production of hydroxyl radicals from the simultaneous generation of superoxide and nitric oxide. Biochem. J. 281, 419–424. Ischiropoulos, H., Zhu, L., Chen, J., Tsai, M., Martin, J. C., Smith, C. D., and Beckman, J. S. (1992a). Peroxynitrite-mediated tyrosine nitration catalyzed by superoxide dismutase. Arch. Biochem. Biophys. 298, 431–437. Ischiropoulos, H., Zhu, L., and Beckman, J. S. (1992b). Peroxynitrite formation from macrophage-derived nitric oxide. Arch. Biochem. Biophys. 298, 446–451. Kamisaki, Y., Wada, K., Bian, K., Balabanli, B., Davis, K., Martin, E., Behbod, F., Lee, Y. C., and Murad, F. (1998). An activity in rat tissues that modifies nitrotyrosinecontaining proteins. Proc. Natl. Acad. Sci. USA 95, 11584–11589. Khan, J., Brennand, D. M., Bradley, N., Gao, B., Bruckdorfer, R., and Jacobs, M. (1998). 3-Nitrotyrosine in the proteins of human plasma determined by an ELISA method. Biochem. J. 330, 795–801. Lomonosova, L. L., Kirsch, M., Rauen, U., and de Groot, H. (1998). The critical role of Hepes in SIN-1 cytotoxicity, peroxynitrite versus hydrogen peroxide. Free Radic. Biol. Med. 24, 522–528. MacMillan-Crow, L. A., Crow, J. P., Kerby, J. D., Beckman, J. S., and Thompson, J. A. (1996). Nitration and inactivation of manganese superoxide dismutase in chronic rejection of human renal allografts. Proc. Natl. Acad. Sci. USA 93, 11853–11858. MacMillan-Crow, L. A., and Thompson, J. A. (1999). Immunoprecipitation of nitrotyrosine-containing proteins. Methods Enzymol. 301, 135–145. Martin-Romero, F. J., Gutierrez-Martin, Y., Henao, F., and Gutierrez-Merino, C. (2004). Fluorescence measurements of steady-state peroxynitrite production upon SIN-1 decomposition: NADH versus dihydrodichlorofluorescein and dihydrorhodamine 123. J. Fluoresc. 14, 17–23. McCord, J. M., and Fridovich, I. (1969). Superoxide dismutase: An enzymic function for erythrocuprein (hemocuprein). J. Biol. Chem. 244, 6049–6055. Moran, J. F., James, E. K., Rubio, M. C., Sarath, G., Klucas, R. V., and Becana, M. (2003). Functional characterization and expression of a cytosolic iron-superoxide dismutase from cowpea root nodules. Plant Physiol. 133, 773–782. Mun˜oz, I. G., Moran, J. F., Becana, M., and Montoya, G. (2005). The crystal structure of an eukaryotic iron superoxide dismutase suggests intersubunit cooperation during catalysis. Protein Sci. 14, 387–394. Murgia, I., Delledonne, M., and Soave, C. (2002). Nitric oxide mediates iron-induced ferritin accumulation in Arabidopsis. Plant J. 30, 521–528. Pagnussat, G. C., Simontacchi, M., Puntarulo, S., and Lamattina, L. (2002). Nitric oxide is required for root organogenesis. Plant Physiol. 129, 954–956. Pauly, N., Pucciariello, C., Mandon, K., Innocenti, G., Jamet, A., Baudouin, E., Herouart, D., Frendo, P., and Puppo, A. (2006). Reactive oxygen and nitrogen species and glutathione: Key players in the legume-Rhizobium symbiosis. J. Exp. Bot. 57, 1769–1776. Puppo, A., Rigaud, J., and Job, D. (1981). Role of superoxide anion in leghemoglobin autoxidation. Plant Sci. Lett. 22, 353–360. Quint, P., Reutzel, R., Mikulski, R., McKenna, R., and Silverman, D. N. (2006). Crystal structure of nitrated human manganese superoxide dismutase: Mechanism of inactivation. Free Radic. Biol. Med. 40, 453–458. Radi, R. (2004). Nitric oxide, oxidants, and protein tyrosine nitration. Proc. Natl. Acad. Sci. USA 101, 4003–4008. Reiter, C. D., Teng, R.-J., and Beckman, J. S. (2000). Superoxide reacts with nitric oxide to nitrate tyrosine at physiological pH via peroxynitrite. J. Biol. Chem. 275, 32460–32466.
618
Estı´baliz Larrainzar et al.
Saito, S., Yamamoto-Katou, A., Yoshioka, H., Doke, N., and Kawakita, K. (2006). Peroxynitrite generation and tyrosine nitration in defence responses in tobacco BY-2 cells. Plant Cell Physiol. 47, 689–697. Sarver, A., Scheffler, N. K., Shetlar, M. D., and Gibson, B. W. (2001). Analysis of peptides and proteins containing nitrotyrosine by matrix-assisted laser desorption/ionization mass spectrometry. J. Am. Soc. Mass. Spectrom. 12, 439–448. Schwedhelm, E., Tsikas, D., Gutzki, F. M., and Fro¨lich, J. C. (1999). Gas chromatographictandem mass spectrometric quantification of free 3-nitrotyrosine in human plasma at the basal state. Anal. Biochem. 276, 195–203. Shigenaga, M. K., Lee, H. H., Blount, B. C., Christen, S., Shinego, E. T., Yip, H., and Ames, B. N. (1997). Inflammation and NOx-induced nitration: Assay for 3-nitrotyrosine by HPLC with electrochemical detection. Proc. Natl. Acad. Sci. USA 94, 3211–3216. Souza, J. M., Daikhin, E., Yudkoff, M., Raman, C. S., and Ischiropoulos, H. (1999). Factors determining the selectivity of protein tyrosine nitration. Arch. Biochem. Biophys. 371, 169–178. Stamler, J. S., Lamas, S., and Fang, F. C. (2001). Nitrosylation: The prototypic redox-based signaling mechanism. Cell 106, 675–683. Soulere, L., Claparols, C., Perie, J., and Hoffmann, P. (2001). Peroxynitrite-induced nitration of tyrosine-34 does not inhibit Escherichia coli iron superoxide dismutase. Biochem. J. 360, 563–567. Thomas, D. D., Espey, M. G., Vitek, M. P., Miranda, K. M., and Wink, D. A. (2002). Protein nitration is mediated by heme and free metals through Fenton-type chemistry: An alternative to the NO/O2 reaction. Proc. Natl. Acad. Sci. USA 99, 12691–12696. Valderrama, R., Corpas, F., Carreras, A., Ferna´ndez-Ocan˜a, A., Chaki, M., Luque, F., Go´mez-Rodrı´guez, M., Colmenero-Varea, P., del Rı´o, L., and Barroso, J. (2007). Nitrosative stress in plants. FEBS Lett. 581, 453–461. Viera, L., Ye, Y. Z., Estevez, A. G., and Beckman, J. S. (1999). Immunohistochemical methods to detect nitrotyrosine. Methods Enzymol. 301, 373–381. Yamakura, F., Taka, H., Fujimura, T., and Murayama, K. (1998). Inactivation of human manganese-superoxide dismutase by peroxynitrite is caused by exclusive nitration of tyrosine 34 to 3-nitrotyrosine. J. Biol. Chem. 273, 14085–14089. Yi, D., Ingelse, B. A., Duncan, M. W., and Smythe, G. A. (2000). Quantification of 3-nitrotyrosine in biological tissues and fluids: Generating valid results by eliminating artifactual formation. J. Am. Soc. Mass. Spectrom. 11, 578–586.
Author Index
A Abadia, A., 29 Abadia, J., 29 Abbruzzetti, S., 318, 320, 329, 330, 333, 335, 337, 338, 339 Abe, Y., 118, 121 Abian, J., 29 Achebach, S., 193 Achkor, H., 562 Adachi, S., 121, 125, 129, 130, 381 Adams, M. W., 242 Adams, P., 542 Adams, P. D., 8 A¨delroth, P., 86, 93 Aebersold, R., 122 Agmon, N., 333, 337, 460 Agron, P. G., 174 Aguado, M. T., 541 Aiba, H., 214, 217, 231 Aish, J., 513 Aissaoui, H., 304 Aitken, J. F., 493 Akita, O., 118, 121 Alain, P., 588, 589 Alam, M., 163, 164, 166, 168, 171, 186, 288 Alayash, A. I., 151 Alben, J. O., 354, 356, 357, 358, 359 Alcantara, R., 484 Alderton, W. K., 563, 565 Alessandro Giuffre`, A., 33, 39, 40 Alexeeva, S., 506 Ali, F., 555, 558 Allen, B. W., 137 Alm, E. J., 105, 212, 225, 236 Almeida, C. C., 41, 124, 212 Almeida, G., 64 Almeida, I. M., 586, 587 Almo, S. C., 403 Alonso, P. J., 303 Altenbach, C., 293, 304 Amadei, A., 412, 418, 481 Amara, P., 440 Amiconi, G., 313 Amoroso, G., 577, 578, 579, 580, 581, 590, 591 Anantharaman, V., 186 Andersen, J. F., 370, 373 Anderson, J. B., 164
Anderson, J. L., 348 Anderson, J. O., 4 Anderson, P. G., 146 Anderson, S., 336, 381, 382, 386, 387, 388, 399, 406, 407, 409, 418 Anderssen, S. U., 600 Andersson, C. R., 596 Andersson, J. O., 4, 23 Andersson, L. A., 303 Andreoletti, P., 440 Andrew, C. R., 295 Anfinrud, P. A., 349, 357, 381, 384, 387, 399, 402, 405, 406, 407, 409, 411, 412, 418, 419, 430 Angove, H., 64, 69, 73, 74 Anjum, M. F., 41, 193, 500, 513, 542, 551, 552, 558 Anni, H., 279, 280 Ansari, A., 337, 341, 349, 356, 361, 398, 409 Anselmi, M., 412, 418 Antonini, E., 351, 355 Antunes, F., 141, 142 Aono, S., 168 Appleby, C. A., 178, 356 Arai, H., 236 Arakawa, H., 125 Aranda, R. T., 407 Arata, H., 88 Aravind, L., 186 Archer, G. L., 80 Archer, M., 64 Archer, S., 581, 583 Arciero, D. M., 242 Arcovito, A., 381, 399, 407, 409, 411, 418 Arduini, R. M., 357 Ariz, I., 605 Arkhipov, A., 418, 441, 443, 450, 451, 452, 472 Arkin, A. P., 105, 212, 225, 236 Armarego, W. L. F., 500 Armstrong, F. A., 92 Armstrong, R. S., 299 Arndt, J., 174 Arnone, A., 323 Arraiano, C. M., 216 Arredondo-Peter, R., 164 Arslan, E., 85 Artacho, E., 486
619
620
Author Index
Arzt, S., 386, 407 Ascenso, C., 242 Ascenzi, P., 267, 275, 277, 288, 302, 323, 337, 425, 426, 427, 431, 448, 460, 461, 462, 467, 489 Ascoli, F., 389 Asher, S. A., 261 Asso, M., 30 Astashkin, A. V., 303 Atkinson, M. R., 175 Auchtung, J. M., 213 Augusto, O., 586, 587 Austin, R. H., 337, 418, 419, 440 Austin, S., 239 Ausubel, F. M., 25 Auzmendi, I., 605 Aviv, H., 242 AVMA Panel on Euthansia, 534 Avnir, D., 318 Ayaydin, F., 597 B Babcock, G. T., 261 Bachmann, R., 304 Bachus, K. E., 555 Badger, M. R., 579 Bailly, X., 164 Baker, A. R., 93 Balfour, C. A., 370, 373 Ballesteros, A., 318 Ballou, D. P., 236 Bamford, V., 69, 73, 74 Banderini, S., 333, 334 Bang, I. S., 42, 212, 522, 523, 526, 528, 532, 535 Bank, J., 302 Banks, S., 532 Banushkina, P., 418 Bari, S. E., 486 Barker, S. L., 137 Barlow, C. H., 265 Barnard, A. M., 192 Barone, M. C., 141, 142, 144, 149, 151 Barrera, L. O., 223 Barrett, J., 41, 42, 193, 237, 502, 505, 507, 508, 509, 510, 512, 513, 514, 515 Barroso, J. B., 561, 562, 563, 564, 566, 568, 570 Bartfeld, D., 242 Bartoli, C. G., 587 Bartunik, H. D., 64 Bashford, D., 421, 443, 448, 467, 478, 481 Bastin, G., 503 Bateman, A., 164 Bates, D. M., 192, 193, 205 Batut, J., 174 Beanan, M. J., 80 Beauchamp, C. O., 611 Beaumont, H. J., 542
Becana, M., 607, 610, 615, 616, 617 Becker, K., 104 Becker, O. M., 422 Beckman, J. S., 348, 606 Bedriunaite, A., 142, 151 Bedzyk, L. A., 41, 42, 237, 502, 506, 509, 515 Beece, D., 354, 356, 357, 358, 359 Beeson, K. W., 337, 418, 440 Beinert, H., 192, 193, 198, 202, 205 Beligni, M. V., 576 Bell, L. C., 81 Bellamy, T. C., 142, 145 Belli, W. A., 542, 543 Bellott, M., 421, 443, 448, 467, 478 Bemski, G., 297, 298, 299, 300 Benedetti, P. A., 315 Benjamin, N., 543 Bennati, M., 293 Bennett, B., 193, 194, 196 Benson, L. M., 242 Bent, A. F., 600 Bentejac, M., 588, 589 Berendsen, H. J. C., 481 Berendzen, J., 349, 356, 357, 359, 361, 390, 398, 399, 407, 411, 418, 425, 480 Bergman, M. A., 522 Berkels, R., 579 Berman, H. M., 423, 425, 448, 466 Bersch, B., 236 Berthiaume, J., 136 Berthomieu, C., 104 Bethke, P. C., 562, 568, 579 Bettati, S., 311, 313, 316, 318, 320, 323, 324, 333, 335, 337, 338, 339, 340, 388 Beuron, F., 238 Bevington, P. R., 342 Bhat, T. N., 423, 425, 448 Bidon-Chanal, A., 477, 482, 488, 490, 491 Bieda, M., 223, 224 Bikiel, D. E., 486, 492 Bill, E., 242 Biosca, J. A., 567 Birnboim, H. C., 562 Bittl, R., 303 Blattner, F. R., 505 Blaxter, M., 257, 419, 460 Blenkinsop, C., 302 Bloch, C. A., 505 Block, H., 293 Blomberg, L. M., 493 Blomberg, M. R., 493 Blondel, A., 423 Blonski, C., 174 Bloodsworth, A., 140 Blouquit, Y., 336, 418 Blumberg, W. E., 301, 302 Blundell, T. L., 480
621
Author Index
Bocedi, A., 288 Bodenmiller, D. M., 211, 212, 215, 216, 217, 221, 222, 230, 502 Boechi, L., 486, 492 Boelens, R., 141, 151 Boese, M., 113 Boffi, A., 323 Bogdan, C., 532 Boistard, J., 174 Boistard, P., 174 Bolli, A., 277 Bolognesi, M., 258, 267, 275, 277, 288, 302, 323, 337, 418, 419, 425, 426, 427, 430, 431, 446, 448, 460, 461, 462, 467, 482, 486, 488, 489, 490, 491 Bolton, R., 289, 294 Bonaccio, M., 318, 335, 338 Bonaventura, C., 138 Bonaventura, J., 138, 313 Bondar, A.-N., 423 Bonham, M. A., 389 Bonneau, R., 330, 332, 333 Borbat, P., 303 Bordes, P., 238 Borgstahl, G. E., 399 Borisov, V. B., 147 Bork, P., 164 Boron, W. F., 441, 453 Borovik, A. S., 242 Borrelli, K. W., 480, 484 Bors, W., 585, 586, 588, 589, 590 Borutaite, V., 139, 141, 142, 151, 502 Bose, M., 104 Bossa, C., 412, 418 Bossi, L., 214 Boudko, D., 164, 166 Bourenkov, G. P., 64 Bourgeois, D., 381, 382, 383, 386, 397, 399, 403, 405, 406, 407, 408, 409, 411, 418, 440, 480 Bourne, P. E., 423, 425, 448 Bourret, R. B., 175, 187, 522, 523, 526 Boveris, A., 141, 142 Bovre, K., 546 Bowne, S. F., 354, 356, 357, 358, 359 Boxer, S. G., 418 Boyle, P., 532 Brand, L., 342 Brandish, P. E., 236 Brashear, W. T., 493 Braun, R., 418, 441, 442, 443, 448, 450, 451, 452, 460, 461, 467, 472 Braun, S., 318 Braunstein, D., 349, 356, 358, 359, 361 Brause, J. E., 522 Brautigam, C., 174 Bredenbeck, J., 349, 358 Brent, R., 25
Bricogne, G., 7 Brinkac, L., 80 Brinkac, L. M., 80 Bro, C., 503 Broillet, M., 590 Bronnikov, G., 142 Brookes, P. S., 142, 146, 149 Brooks, B. R., 419, 422, 425, 428, 442, 461 Brooks, C. L. III, 423 Brooun, A., 168 Brown, C. W., 258 Brown, G. C., 139, 140, 141, 142, 144, 151, 502 Browning, D. F., 214, 217, 231 Brubaker, M., 403 Bruccoleri, R. E., 419, 422, 425, 428, 442, 461 Brumen, M., 315 Brunger, A. T., 8 Bruno, S., 311, 315, 318, 320, 323, 329, 330, 335, 337, 338 Brunold, T. C., 7, 12, 29 Brunori, M., 4, 11, 24, 25, 33, 39, 40, 47, 49, 50, 51, 52, 54, 55, 60, 141, 142, 144, 147, 151, 212, 288, 295, 315, 337, 351, 355, 381, 397, 398, 399, 405, 407, 408, 409, 411, 412, 418, 419, 480 Brzozowski, A., 321 Buchalova, M., 178 Buck, M., 238 Buck, M. J., 213, 216, 217, 220, 221, 222, 223 Buettner, G. R., 587 Bu¨hl, M., 294 Burgard, C., 298, 299 Burgess, B. K., 92 Burke, K., 486 Burke, P. M., 399 Burland, V., 505 Burlat, B., 68, 71 Burmester, T., 288, 302, 462 Busby, S. J., 192, 201, 213, 214, 216, 217, 218, 219, 231 Busch, A., 81, 105, 236 Busch, D. H., 178 Busk, H., 600 Busse, R., 113 Butland, G., 80, 81, 82, 85, 86, 89, 90, 91, 94, 95, 96, 97 Butler, A. R., 104 Butler, C. S., 244 Butt, J. N., 63, 69, 71, 73, 74, 91, 94, 95, 96 Byrd, R. H., 423 Byrn, M. P., 301 Byrns, R. E., 592 C Caceres, A. I., 174 Cadenas, E., 141, 142
622 Cafiso, D. S., 293, 304 Caflisch, A., 423 Cai, T., 553 Cai, W., 597 Cairrao, F., 216 Caldwell, J. W., 442, 478, 479 Callahan, M. K., 546 Cambillau, C., 407 Cameron, A. D., 9, 10, 12 Campanini, B., 318 Campbell, B. F., 337 Campbell, J. W., 386, 407 Cannon, W., 238 Capece, L., 477, 479, 482, 486, 492, 493 Cardinale, J. A., 542, 543 Carey, P. R., 261 Carfi, A., 9 Carlson, M., 274, 419, 461 Carlson, M. L., 418, 461 Carpenter, J. F., 350 Carr, G. J., 88, 543 Carreras, A., 561, 562, 563, 564, 566, 568, 570 Carreras, M. C., 136 Carrondo, M. A., 3, 5, 7, 22, 31 Carver, T. E., 418, 461 Carver, T. L. W., 588 Case, D. A., 418, 442, 478, 479, 481 Cash, V. L., 195 Castagnola, M., 288 Castor, M., 521 Castresana, J., 543 Catalan-Sakairi, M. A., 80 Caughey, W. S., 260, 261, 264, 265, 348, 349, 354, 356 Caves, L., 423 Chafin, D. R., 323 Chaki, M., 561, 562, 564, 566, 568, 570 Chakravortty, D., 522 Chalfie, M., 597 Chamberland, H., 419, 460 Champagne, D. E., 370, 373 Chan, C., 89 Chan, J., 502, 509 Chan, M. K., 174 Chan, S. I., 295 Chance, B., 140 Chance, M. R., 337, 367, 368 Chaney, M., 238 Chanfon, A., 174 Changeux, J.-P., 320, 323 Channa, A., 502 Chapleigh, J. P., 296, 297 Chapman, S. K., 348 Chartrain, N., 524, 532 Chatton, J., 590 Cheatham, T. E., III, 442, 478, 479 Cheesman, M. R., 63, 80, 81, 91, 92, 94, 95, 96, 97, 193, 194, 202, 205, 206
Author Index
Chen, A.-P., 588 Chen, E., 330 Chen, H., 168 Chen, L., 4, 5, 7, 22, 40, 48 Chen, S., 302 Chen, X., 138, 295, 299 Chiancone, E., 323, 388, 461 Chiang, C. Y., 104 Chien, E. Y., 356 Chien, J. C. W., 301 Chipot, C., 442, 448, 460, 461, 467 Choc, M. G., 354 Choi, P. S. T., 296, 297 Chou, K. J. Y., 509, 513 Christoph, G. W., 337 Chu, D., 502, 509 Chu, K., 331, 349, 356, 357, 358, 359, 361, 398, 399, 411, 418, 419, 425, 480 Chudakov, D. M., 598 Chumley, P. H., 140 Church, G., 213 Cieplak, P., 485 Clark, H. A., 137 Clark, V. L., 542, 543, 546 Clarke, A., 562, 584 Clarke, T. A., 63, 68, 69, 71 Clarke, T. C., 68 Claus, H., 540 Clay, M. D., 242 Clifton, I. J., 405 Clore, G. M., 8 Clough, S. J., 600 Cobb, J. P., 532 Coburn, B., 532 Coelho, R., 4, 5, 7, 8, 10, 11, 12, 13, 22, 26, 38, 48, 51 Cohen, B. E., 403 Cohen, J., 418, 439, 440, 441, 443, 450, 451, 452, 453, 472 Colacino, J. M., 313 Colby, J. E., 323 Cole, J. A., 40, 63, 64, 66, 67, 68, 69, 71, 73, 74, 81, 212, 213, 238, 277, 502, 542, 543 Cole, L. J., 555 Cole, R. P., 149 Cole, S. T., 192 Coletta, M., 315, 323 Collaborative Computational Project Number 4, 7 ColladoVides, J., 505 Colombo, S. L., 136 Colotti, G., 323, 388, 389 Connelly, P. R., 323 Conrath, U., 577, 578, 579, 580, 581, 590, 591 Constantinidou, C., 40, 213, 238 Contreras-Zentella, M. L., 513 Cookson, B. T., 522 Coopamah, M. D., 502, 513
623
Author Index
Cooper, C. E., 135, 136, 139, 140, 141, 142, 144, 145, 146, 153, 295, 302, 348, 563, 565 Copeland, D. M., 480, 486 Copley, R. R., 164 Corden, B. B., 265 Cordone, L., 333 Corker, H., 212, 217, 238 Cornish-Bowden, A., 142, 150 Corpas, F. J., 561, 562, 563, 564, 566, 568, 570 Correa, R., 532 Correa-Aragunde, N., 562 Corrie, J. E. T., 333 Cosa, G., 330 Cosper, C. A., 242 Costa, C., 64 Costa, L. L., 64 Costantino, G., 514 Cottone, G., 333 Coufal, D. E., 104 Coulson, A. R., 124 Coulter, E. D., 3, 4, 11, 24, 25, 26, 29, 30, 31, 32, 34, 40, 49, 51, 52 Couture, M., 165, 265, 267, 268, 269, 271, 272, 273, 274, 419, 425, 427, 431, 448, 460 Cowan, S. W., 8 Cowen, B. R., 349, 356, 361 Cox, G. B., 139 Coyle, C. M., 366 Crack, J., 104, 191, 193, 194, 197, 199, 201, 202, 204, 205, 206, 236, 501, 513, 515 Cramm, R., 81, 82, 103, 104, 105, 106, 107, 109, 114, 236, 237 Crawford, M. J., 530 Crawford, N. M., 597 Crawford, P. A., 10 Crespo, A., 430, 460, 477, 479, 482, 484, 486, 488, 489, 490, 491, 492, 493, 494 Crooke, H., 542 Cross, R., 513 Crosson, S., 175, 382 Crouch, M. L., 42, 212, 522, 523, 526, 528, 532, 535 Crow, J. P., 606 Crowder, M. W., 10 Crowe, J. H., 350 Cruickshank, D. W. J., 405 Cruz, A., 216 Cruz-Ramos, H., 104, 193, 199, 201, 236, 501, 513, 515 Cullen, V. L., 532 Cunha, C. A., 64 Cunha, F. Q., 562 Cupane, A., 354 Curry, S. R., 493 Cushing, L., 389 Cutruzzola, F., 398, 399, 408, 411 Cvitanich, C., 600 Czeluzniak, J., 257, 419, 460
Czerminski, R., 421, 461 Czernuszewicz, R. S., 260, 261 D D’Autre´aux, B., 104, 105, 212, 217, 236, 237, 238, 239, 240, 242, 243 D’Mello, R., 137 da Costa, P. N., 40, 48 Da Re, S., 174, 181 Daff, S., 348 Dai, X., 118 Daiber, A., 118, 130 Daidone, I., 412, 418 Dalke, A., 429, 443, 448, 466 Dalton, D. A., 607 Dantsker, D., 165, 268, 274, 275, 276, 318, 333, 337, 418, 430, 460 Daran-Lapujade, P., 503 Darensbourg, M. Y., 104 Darley-Usmar, V. M., 140, 142, 146, 149 Das, A., 3, 4, 11, 24, 25, 26, 29, 30, 31, 32, 34, 40, 49, 51, 52 Das, T. K., 175, 265, 460 Dasgupta, S., 279 Daskalakis, V., 168 Datsenko, K. A., 214, 505 Daugherty, S. C., 80 Dauter, Z., 5, 7 Daveran, M.-L., 174 David, M., 174 Davidsen, T. D., 80 Davies, E. R., 292 Davies, N. A., 302 Davies, R. G., 418 Davis, J. L., 303 Davis, L. A., 4, 23 Davis, M. I., 7, 12, 29 Davis, N. W., 505 de Boer, A. P., 82 De Boer, E., 141, 151 De Gara, L., 592 de Groot, M. T., 130 De La Fortelle, E., 7 de Lorenzo, V., 239 de Pinto, M. C., 592 de Sanctis, D., 302, 462 de Vries, S., 81, 82, 90, 104, 106, 107, 109, 114 De Weerd, K., 305 Deacon, A. M., 5, 6 Dean, D. R., 195 deBoer, A. P. N., 81 DeBolt, S., 442, 478, 479 Deboy, R. T., 80 Decker, H., 323, 324 Dedieu, A., 174 Deen, W. M., 146
624 Deeth, R. J., 487 deGier, J.-W. L., 81 Degn, H., 149 del Rı´o, L. A., 561, 562, 563, 564, 566, 568, 570 DeLano, W. L., 8 Delledonne, M., 575, 576, 577, 584, 591, 594, 596, 606, 616, 617 Delpy, D. T., 142, 151 deMattos, M. J. T., 506 Demoncheaux, E. A., 42, 81, 212, 528, 531, 557, 558 Demple, B., 104, 236, 514 Deng, P., 336, 349, 354, 355, 357, 358, 364, 365, 370, 418, 419 Dennis, E. S., 419, 460, 596, 600 Dennison, V., 68 Derbyshire, M. K., 164 Deretic, V., 502, 509 Deriu, D., 323 Dermastia, M., 88 Derojaswalker, T., 514 Derrewenda, Z., 321 Desbois, A., 261 DeShazer, D., 80 Desikan, R., 562, 576, 584, 597 Desmet, F., 287 Devlin, F. J., 111 Devreese, B., 89 DeWeese-Scott, C., 164 Dewilde, S., 164, 212, 217, 238, 268, 288, 297, 299, 302, 303, 304, 337, 425, 427, 431, 448, 460, 461, 462, 467 Dexter, A. F., 303 Dey, D., 175 DeYoung, A., 323 Di Nola, A., 412, 418 Dias, J. M., 64 Dı´az, M., 562 Dibden, D., 193 Dickinson, L. C., 288, 301, 302 Dideberg, O., 9 DiDonato, M., 5, 6 Dikanov, S. A., 298, 299 Dikshit, K. L., 267, 276, 460 Dinauer, M., 522 Ding, A. H., 522 Ding, H., 104, 236 Ding, X. D., 373 DiNola, A., 481 Dioum, E. M., 174, 176 Dirsch, V. M., 590 Disselhorst, J. A. H. M., 293 Ditta, G. S., 174 Dittman, W. A., 555 Dixon, J. E., 195 Dixon, M. M., 7 Dixon, N. E., 500
Author Index
Dixon, R., 105, 174, 212, 217, 236, 237, 238, 239, 240, 242, 243 Djordjevic, S., 175, 186 Do, S. K., 562 Dockrell, D. H., 80, 555, 558 Doctorovich, F., 479, 482, 486, 493 Dodson, E., 321 Dodson, E. J., 8 Dodson, G., 321 Dodson, R. J., 80 Doeller, J. E., 142, 149 Doetschman, D. C., 297, 298 Doherty, D. H., 493 Doke, N., 589 Domergue, O., 174 Dominici, P., 596 Dong, A., 348, 349 Donhauser, N., 532 Dordas, C., 596 Doster, W., 337, 354, 356, 357, 358, 359 Dou, Y., 274, 366, 419, 461, 493 Dougan, G., 532, 533 Dougaparsad, S., 532 Douzou, P., 350 Downie, J. A., 139, 506 Drago, R. S., 265 Dreisbach, A., 4, 23, 24, 26, 40, 51 Drennan, C. L., 38 Dreybrodt, W., 262 Dubchak, I. L., 105, 212, 225, 236 Dubery, I., 585, 586, 588, 589, 590 Dudits, D., 597 Duee, E., 9 Duez, C., 9 Duke, E. M. H., 405 Dunbrack, R., 423 Dunbrack, R. L., Jr., 421, 443, 448, 467, 478 Duner, J., 585, 586 Dunham, C. M., 174 Dunn, B., 318 Dunn, J., 137 Dunne, J., 302 Durbin, R., 164 Durner, J., 562, 570, 576, 583, 585, 586, 588, 589, 590, 591, 606 Dutzler, R., 423 Dvorak, J. A., 315 Dwyer, M. A., 484 E Eady, R. R., 295 Earnshaw, R. G.4, 504 Eaton, A. W., 338 Eaton, G. R., 305 Eaton, S. S., 305 Eaton, W., 320, 341
625
Author Index
Eaton, W. A., 313, 315, 316, 318, 320, 321, 322, 323, 337, 340, 398, 409 Ebel, C., 174, 181 Eberl, T., 138, 149 Eddy, S. R., 164 Edmondson, D. A., 15, 29, 38 Efromovich, S., 211, 225, 226 Egawa, T., 255, 256, 265, 274, 278, 279 Egeberg, K. D., 274, 356 Eguchi, A., 88 Ehrenstein, D., 361 Eich, R. F., 493 Eichenberger, P., 213, 216, 231 Eichinger, M., 486 Eiglmeier, K., 192 Einsle, O., 64 Eisen, J. A., 80 Eisenstein, L., 337, 354, 356, 357, 358, 359, 418, 440 Eitinger, T., 105 Ekman, P., 522 Elber, R., 418, 419, 421, 429, 440, 460, 461, 465, 466, 467, 468 Elfering, S., 136 Elin, R. J., 532 Ellerby, L. M., 318 Ellis, D. E., 299 Ellis, P. J., 299 Elola, M. D., 479 Elsner, B., 105, 237 Elvers, K. T., 277, 279, 502 Embley, T. M., 4, 23 Emery, D. C., 10, 12 Endre, G., 597 Enemark, J. H., 104 Engelen, F. A. A., 502 Engler, N., 357 Erecinska, M., 140 Eren, M., 7 Eriksson, S., 41 Erman, J. E., 298 Ermler, U., 4, 5, 6, 8, 16, 23, 24, 26, 40, 51, 52 Ernst, R. M., 349 Ernzerhof, M., 486 Escamilla, E., 513 Esteban, F. J., 561, 562, 563, 564, 566, 568, 570 Estrin, D. A., 430, 460, 477, 479, 482, 484, 486, 488, 489, 490, 491, 492, 493, 494 Euskirchen, G., 597 Evans, P., 7 Evans, S. M., 565 Evans, T. J., 522 Evanseck, J. D., 421, 443, 448, 467, 478 Ewering, C., 105 Exner, M., 246 Eybert, L., 402, 405 Eydmann, T., 239
F Faggiano, S., 311, 324, 329, 330, 337 Fago, A., 315 Fahnenschmidt, M., 303 Falkow, S., 114 Falkowski, K. M., 302 Fang, F., 522 Fang, F. C., 42, 136, 212, 501, 521, 522, 523, 526, 527, 528, 532, 533, 535 Fann, Y., 303, 304 Fareleira, P., 4, 5, 22, 40, 48 Farnham, P. J., 223, 224 Farres, J., 274 Favinger, J. L., 504 Feechan, A., 568 Feelisch, M., 606–607 Feher, G., 303 Feher, T., 505 Feijo, J. A., 562 Feis, A., 263, 264, 265 Feldblyum, T., 80 Felix, R., 4, 24, 26, 40, 49, 51, 52 Feltham, R. D., 104 Feng, Y., 7 Feng, Z., 423, 425, 448 Fengler, S., 355, 365, 370 Fenimore, P. W., 398 Fenney, R. E., 204 Fenster, A., 332 Ferenci, T., 504, 505, 507 Ferguson, D., 442, 478, 479 Ferguson, S. J., 81, 88, 89, 543 Fernandez, M. L., 486 Ferna´ndez, M. R., 567, 568 Ferna´ndez-Ocan˜a, A., 562, 564, 566, 568, 570 Ferraro, J. R., 258 Fetrow, J. S., 305 Fewson, C. A., 509 Fey, N., 487 Field, M. J., 421, 440, 443, 448, 467, 478 Field, S. J., 66, 79, 91, 94, 95, 96 Figueroa-Bossi, N., 214 Finlay, B. B., 522, 532 Finn, R. D., 164 Fiori, P. L., 4 Fiossner, I., 588, 589, 590 Firoved, A. M., 502, 509 Fischer, D. S., 29 Fischer, H. M., 174 Fischer, S., 419, 421, 422, 423, 426, 440, 443, 448, 467, 472, 478 Fittipaldi, M., 304 Fiuffre, A., 212 Flatley, J., 41, 237, 502, 505, 507, 508, 509, 510, 512, 514, 515 Fletcher, D. S., 524, 532 Flock, U., 86, 93
626
Author Index
Flores, M., 297, 298, 299, 300 Fogel, D. B., 8 Ford, H. R., 532 Ford, P. C., 140, 350, 351, 563 Forster, F. K., 334 Forte, E., 4, 11, 24, 25, 33, 39, 40, 47, 49, 51, 52, 141, 142, 144, 147, 151, 212 Foster, M. W., 568 Foster, P. G., 4 Fourment, J., 174, 181 Foussard, M., 174 Fouts, D. E., 80 Foyer, C. H., 569 Franzen, S., 357, 358, 487 Fraser, A., 540 Frauenfelder, H., 331, 337, 349, 354, 355, 356, 357, 358, 359, 361, 398, 409, 418, 419, 440 Fraza˜o, C., 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 22, 26, 31, 38, 48, 51 Freed, J. H., 293 Freeman, B. A., 140 Frei, H., 262 Freitas, T. A., 163, 164, 166, 171, 186, 288 Frenzke, K., 543 Frere, J. M., 9 Frey, A. D., 257, 274, 288 Fricke, W. F., 105 Fridovich, I., 205, 607, 611, 613, 616, 617 Frieden, C., 165 Friedman, A. J., 318, 337 Friedman, J. M., 165, 261, 262, 268, 274, 275, 276, 318, 333, 337, 354, 388, 389, 418, 430, 460 Friedrich, B., 81, 82, 105, 106, 114, 236, 237 Friesner, R. A., 479, 484, 485 Frische, S., 315 Frosch, M., 540 Fuchs, J., 349, 355, 365, 370 Fuchs, M., 288, 293 Fuchsman, W. H., 356 Fujii, M., 121, 129, 130 Fujita, M., 213, 216, 231 Fukamizu, A., 118, 123, 125 Fukuto, J. M., 592 Fukuzawa, K., 585, 586, 587 Fung, A. M., 532 Fung, H. L., 564 Fung, L. Y., 564 Fushinobu, S., 125, 128, 129 G Gafter-Gvili, A., 540 Gale, J. D., 486 Galinier, A., 174 Galleni, M., 9 Gambacurta, A., 389 Gambarelli, S., 104
Gamblin, S. J., 10, 12 Gao, J., 421, 443, 448, 467, 478 Garcı´a, A., 486 Garcı´a-Mata, C., 562, 597, 606 Garcı´a-Rubio, I., 303 Gardner, A. M., 4, 24, 40, 41, 48, 51, 81, 212, 236, 237, 460, 493 Gardner, P. R., 4, 24, 40, 41, 48, 51, 81, 212, 236, 237, 273, 274, 460, 493, 501, 514 Garfin, D. E., 26 Garland, P. B., 506, 507 Garnerone, A.-M., 174 Garthwaite, J., 139, 142, 145 Gates, A. J., 69 Gehlhaar, D. K., 8 Gelfand, M. S., 105, 212, 225, 236 Geller, T., 242 Gelpi, E., 29 Gelpi, J. L., 430, 460, 486, 489, 494 Gemeinhardt, S., 543 Genick, U. K., 399 Gennis, R. B., 304 George, G. N., 303 George, S. J., 295 Gerber, I., 585, 586, 588, 589, 590 Gerber, N. C., 303 Gerfen, G., 303 Germroth, P. G., 569 Gerwert, K., 349, 351, 352 Gessner, C. R., 40, 212, 236, 237 Gest, H., 504 Getzoff, E. D., 399 Geusens, E., 461 Ghai, J., 174 Ghisla, S., 15, 29, 38 Giardina, B., 288, 313, 315 Gibson, F., 139 Gibson, Q. H., 141, 274, 365, 388, 389, 398, 408, 409, 411, 418, 419, 461, 492, 493 Gidley, M. D., 41, 42, 193, 507, 508, 513, 515 Gigoryants, V., 305 Gilbert, D. C., 298 Gilberthorpe, N. J., 277, 279, 502 Gilch, H., 262 Gill, I., 318 Gill, R., 522, 523, 526 Gill, S. J., 315, 318, 323 Gill, S. R., 80 Gilles-Gonzalez, M.-A., 168, 174, 175, 176, 177, 178, 180, 181, 182, 184, 185, 186, 187, 348 Gillette, R., 589 Gilliland, G., 423, 425, 448 Girard, M. P., 541 Girsch, P., 81, 90 Giuffre`, A., 4, 11, 24, 25, 33, 39, 40, 47, 49, 50, 51, 52, 54, 55, 60, 141, 142, 144, 147, 151 Giuffrida, S., 333
627
Author Index
Giulivi, C., 136, 146 Gladwin, M. T., 300 Glaser, T., 242 Glasner, J. D., 505 Gnaiger, E., 138, 149, 150 Godfrey, A., 136 Goeden, M. A., 505 Gohlke, U., 81, 83, 88 Goldberg, D. E., 165, 530 Goldberg, R. A., 330 Golden, S. D., 417, 459 Goldfarb, D., 293 Golombek, A. P., 242 Gomes, C. M., 4, 7, 8, 9, 10, 11, 12, 13, 22, 24, 25, 26, 29, 30, 31, 32, 33, 34, 38, 40, 48, 49, 50, 51, 52, 54, 212 Go´mez, M., 562, 563, 564 Go´mez-Rodrı´guez, M. V., 562, 564, 568, 570 Gomi, Y., 125 Goncalves, L. L., 64 Gonc¸alves, V. L., 4, 8, 14, 15, 21, 41, 124, 212 Gonc¸alves, V. M. M., 41 Gong, W., 174 Gonzales, N. R., 164 Gonzalez, C., 174 Gonzalez, G., 168, 174, 175, 176, 178, 186, 187, 348 Gonzalez-Pastor, J. E., 213, 216, 231 Gonzalo, G., 174, 175, 176, 177, 180, 181, 182, 184, 185 Good, D., 354, 356, 357, 358, 359 Goodhew, C. F., 89 Goodin, D. B., 279 Goodman, M., 257, 419, 460 Goodwin, L., 545 Goovaerts, E., 212, 217, 238 Gorecki, M., 242 Gorren, A. C., 141, 151 Gossman, W., 299 Gotoh, O., 118, 125 Goudable, J., 556 Gough, J., 164 Gould, K., 583, 588, 589, 590, 591 Gourdie, R. G., 569 Gouterman, M., 260 Gow, A. J., 236, 501 Grabowski, M., 321 Grainger, D. C., 211, 213, 214, 216, 217, 218, 219, 231 Grand, F., 556 Grandi, E., 329, 330, 337 Granfors, K., 522 Graziano, M., 562 Green, J., 41, 42, 104, 191, 192, 193, 194, 195, 196, 197, 199, 201, 202, 204, 205, 206, 236, 237, 501, 502, 505, 506, 507, 508, 509, 510, 512, 513, 514, 515 Green, R., 223, 224
Green, R. D., 223 Green, T. J., 137 Greenwood, C., 244 Gregor, J., 505 Griffiths, C., 139, 142, 145 Griffiths, L., 40, 238 Griffiths, R., 597 Griffiths-Jones, S., 164 Grinberg, O., 137 Grinstein, S., 532 Griscavage, J. M., 592 Grisham, M. B., 568 Groenen, L. C., 423 Grogan, S., 502 Groisman, E. A., 213 Gro¨nberg, K. L., 81, 91, 94, 95, 96, 97 Gronlund, M., 600 Grootenhuis, P. D. J., 423 Gros, P., 8, 532 Grosse-Kunstleve, R. W., 8 Grossman, A. D., 213 Groves, J. T., 419 Gruia, A. D., 423 Grupp, A., 291 Guallar, V., 477, 479, 484 Guan, K. L., 195 Guenzburger, D., 299 Guerlesquin, F., 89 Guertin, M., 165, 258, 265, 267, 268, 269, 271, 272, 273, 274, 275, 276, 277, 337, 418, 419, 425, 426, 427, 430, 431, 446, 448, 460, 467, 488, 489, 493, 494 Guest, J. R., 192, 193, 194, 195, 196, 202, 542 Guex, N., 425 Guigliarelli, B., 30 Guilbert, C., 423, 426 Guissani, A., 104 Guitton, J., 556 Gumbart, J., 442, 448, 460, 461, 467 Gunsalus, I. C., 337, 440 Gunsalus, R. P., 193 Guo, G. Q., 597 Guo, H., 421, 443, 448, 467, 478 Gwadz, M., 164 Gwyer, J., 73 Gybina, A. A., 136 Gyllenhammar, H., 137, 138 H Ha, S., 421, 443, 448, 467, 478 Haak, J. R., 481 Haddock, B. A., 139, 506 Haehnel, W., 303 Hagemeier, C. H., 4, 5, 6, 8, 16, 23, 24, 26, 40, 51, 52 Hagen, W. R., 303
628 Hager, L. P., 303 Hake, R., 50 Haldane, J., 381 Halleck, M. S., 546 Halliwell, B., 348 Haltia, T., 81, 83, 88 Hamamoto, M., 119 Hamm, P., 349, 358 Hancock, J. T., 562, 576, 584, 597 Hankeln, T., 288, 302, 462 Hansen-Wester, I., 522 Hao, B., 174 Hao, L., 164 Hao, Q., 386, 407 Haqqani, A. S., 562 Hara, I., 418 Harding, M. M., 386, 407 Hargrove, M. S., 288, 419, 460, 596 Harren, F. J. M., 578, 579, 581, 590 Harrison, M., 213, 216, 218, 219 Hart, B. E., 418 Hart, T. W., 512 Harte, R. A., 29 Hartman, J. R., 242 Hartmann, H., 357, 418 Hartung, T., 585, 586, 588, 589, 590 Harvey, J. N., 373 Harvey, M. W., 522 Hasinoff, B. B., 596 Haskin, C. J., 104 Hata, Y., 118, 121 Hatano, K., 118, 121, 130 Hatcher, N., 589 Hauser, C., 242 Hausladen, A., 236, 501, 562 Hayashi, A., 316, 318 Hayashi, M., 236 Hayes, J. M., 504 Hazlett, K. R., 242 Hazzard, J. H., 354 He, Q., 522 He, S., 164 Hebelstrup, K. H., 589, 596, 600 Hedderich, R., 4, 23, 24, 26, 30, 40 Heidelberg, J. F., 80 Heidemeier, J., 359 Heiss, B., 81, 543 Heitman, J., 562 Helbing, J., 349, 358 Helinski, D. R., 174 Hellinga, H. W., 484 Hellingwerf, K. J., 407, 506 Helliwell, J. R., 381, 382, 386, 405, 407 Hellmann, N., 324 Helmann, J. D., 502, 509 Helmick, R. A., 4, 24, 41, 48, 51, 81, 212, 501, 514 Hemmings, A. M., 63, 68, 69, 71, 73, 74
Author Index
Henderson, A. J., 546 Henderson, R. F., 585 Hendrich, M. P., 242 Hendrickson, W. A., 7 Hendriks, J., 81, 83, 88, 543 Hendriks, J. H., 93 Hennecke, H., 85, 174 Henry, E. R., 313, 315, 316, 320, 321, 322, 323, 337, 340, 341, 398, 409 Henry, Y., 104 Hensel, M., 522 Herna´ndez-Urzua, E., 502, 513 Herold, S., 246, 274 Herring, C. D., 505 Hertig, C., 174 Herzberg, G., 265 Hess, J. F., 175, 187 Hessler, F., 540 Higashida, K., 118, 121 Higgins, T. M., 38 Higuchi, T., 588, 590 Hildebrandt, P., 261 Hill, R. D., 589, 596, 597 Hill, S., 137, 239 Hinds, J., 277, 502 Hinnebusch, B. J., 114, 212 Hinton, J. C. D., 41, 63 Hirata, Y., 587, 588, 590, 591 Hirota, S. T. L., 264, 271 Hirt, R. P., 4 Hobbs, G., 503, 504 Hobman, J. L., 40, 213, 238 Hoehn, G. T., 546 Ho¨fer, P., 291 Hoffman, B. M., 303, 304 Hoffman, W. D., 532 Hofrichter, J., 315, 320, 323, 337, 340, 341, 398, 409 Hogg, N., 140, 300, 607 Ho¨hn, M., 301 Hol, W. G., 175 Holden, D. W., 522 Holdstock, J., 213, 216, 218, 219 Holladay, R. S., 135 Hollich, V., 164 Hollis, V. S., 136, 142, 151 Hollman, D. A., 10 Hollocher, T. C., 88, 212, 238 Holton, N. J., 578, 579, 581, 590 Hom, G., 524, 532 Hong, J. K., 570 Hong, M. K., 349, 356 Honig, B., 484, 485 Honore, N., 192 Hood, L. E., 122 Hoogewijs, D., 164 Hoover, D. M., 38 Hoover, T. R., 239
629
Author Index
Hori, H., 300, 303 Hormaeche, C. E., 532, 533 Horner, O., 104 Horvath, G. V., 597 Hoshi, H., 590 Hoshino, M., 351 Hoshino, T., 118, 125 Hoskisson, P. A., 503, 504 Hosseini, J. M., 532 Hou, S., 164, 166, 168, 171, 186, 288 Householder, T. C., 542, 543 Howes, B. D., 265 Hsiung, T. M., 242 Hu, S., 262, 279 Hu, X., 597 Huang, J., 337, 354 Huang, X., 418, 585, 586 Huang, Z. X., 242 Hubbard, R. E., 423 Hubbell, W. L., 293, 304 Huber, K., 30 Huber, R., 64 Hughes, J., 568 Hughes, M. N., 41, 104, 139, 193, 199, 201, 212, 236, 237, 500, 501, 502, 505, 506, 507, 508, 509, 510, 512, 513, 514, 515, 522 Hui, H. L., 323 Huie, R. E., 348 Hultschig, C., 130 Hummer, G., 412, 418, 484 Humphrey, W., 429, 443, 448, 466 Hunt, P. W., 419, 460, 596, 600 Huo, S., 419, 421 Hurd, D., 213, 214, 216, 217, 218, 219, 231 Hurst, R. D., 584 Hurwitz, D. L., 164 Huston, W. M., 555 Hutchings, M. I., 40, 41, 236, 237, 501 Hutchinson, N., 524, 532 Hutter, J., 486 Hu¨ttermann, J., 298, 299, 301 Hutzler, P., 585, 586, 588, 589, 590 Huynh, B. H., 3, 4, 11, 24, 25, 26, 29, 30, 31, 32, 34, 40, 49, 51, 52, 64, 242, 301, 302 Hvitved, A. N., 419, 460, 493, 596 Hyduke, D. R., 509, 513 Hyunh, B. H., 104 I Iba, K., 88 Iben, I. E., 349, 356 Igamberdiev, A. U., 596, 597 Igarashi, Y., 236 Ignarro, L. J., 563, 592 Ihee, H., 380, 381, 385, 387, 399, 406 Iizuka, T., 121, 129, 130, 174, 331
Ikeda-Saito, M., 274, 300, 303, 366, 419, 461 Imai, K., 313 Imai, Y., 125, 127 Inamori, K., 88 Inubushi, T., 303 Ioanitescu, A. I., 302, 303 Ioannidis, N., 139, 500, 522 Ioanoviciu, A., 175 Ionescu, R., 338 Irvine, A. S., 192, 195 Isaacson, R. A., 303 Isaza, C., 174 Ischiropoulos, H., 522, 527, 606–607, 614–615 Ishida, Y., 590 Ishihama, A., 504, 505 Ishii, M., 30, 236 Ishikawa, H., 418 Ishimori, K., 418 Itoh, Y., 590 Iuchi, S., 192 Iwata, M., 16 Iwata, S., 16 Iyer, L. M., 186 J Jabado, N., 532 Jackson, J. D., 164 Jackson, T. A., 349, 357, 381, 399, 409, 411, 418, 430 Jacobs, W. R., 502, 509 Jacobson, M. P., 484, 485 Jagadis Gupta, K., 582, 588, 589 James, E. K., 607, 610 James, P. E., 137 Jameson, G. N., 3, 4, 11, 24, 25, 26, 29, 30, 31, 32, 34, 40, 49, 51, 52 Jancarik, K., 73 Janczuk, A. J., 553 Jankowski, A., 532 Jarboe, L. R., 509, 513 Jarvis, S., 501 Jarzynski, C., 483 Jasaitis, A., 93 Jasid, S., 587 Jasper, C. G., 151 Jeandey, C., 104 Jenney, F. E., Jr., 242 Jensen, D. B., 600 Jensen, E. O., 596, 600 Jensen, S. B., 596, 600 Jensen, S. T., 213, 216, 231 Jentzen, W., 261 Jepson, B., 69 Jervis, A. J., 191, 515 Jeschke, G., 288, 291, 293, 305 Jesse, K., 349
630
Author Index
Ji, J., 137 Ji, X. B., 212, 238 Jia, L., 138 Jia, S.-L., 261 Jian, G.-L., 588 Jiang, J. S., 8 Job, D., 607 Johansson, M. P., 303 Johnson, B. J., 441 Johnson, J. B., 349, 354, 355, 356, 357, 359, 361 Johnson, K. A., 493 Johnson, M. K., 242 Johnson, M. L., 342 Jolley, K. A., 540 Jones, C. M., 337, 341, 398, 409 Jones, H. E., 332 Jones, I., 137 Jones, R. A., 41, 42, 193, 507, 508, 513, 515 Jones, R. L., 562, 568, 579 Jones, T., 239 Jones, T. A., 8 Jones-Carson, J., 521, 522, 523, 526, 527 Jonsson, H., 421 Jordan, P., 193, 194, 196, 202 Joseph, D., 443, 448, 467 Joseph-McCarthy, D., 421, 478 Jouanneau, Y., 4, 22, 24, 29, 30, 48 Jourd’heuil, D., 568 Jourd’heuil, F. L., 568 Jouve, H. M., 440 Jovanovic, T., 242 Juang, R. H., 242 Junquera, J., 486 Justino, M. C., 4, 8, 14, 15, 21, 24, 40, 41, 42, 51, 124, 193, 212, 237, 502, 509, 515 Juszczak, L., 165, 274, 275, 276, 318, 337 K Kabsch, W., 8, 403 Kahn, D., 174, 181 Kahn, R., 407 Kaiser, W. M., 582, 588, 589, 590, 591, 593 Kale, L., 442, 448, 460, 461, 467 Kalko, S. G., 430, 460, 486, 489, 494 Kallio, P. T., 257, 274, 288 Kamberov, E. S., 175 Kamisaki, Y., 606 Kanaori, K., 313 Kanner, D., 242 Kapp, O., 257, 419, 460 Kappl, R., 298, 299 Karatan, E., 164, 166 Karplus, M., 367, 418, 419, 421, 422, 423, 425, 428, 429, 440, 442, 460, 461, 465, 467, 468, 472, 478
Kasim, M., 38 Kastrau, D. H. W., 81 Kato, I., 80 Kato, K., 136 Katou, S., 589 Kaupp, M., 294 Kavanaugh, J. S., 323 Kawabata, S., 88 Kawahara, S., 587, 588, 591 Kawai, M., 540 Kawakita, K., 589 Kawato, A., 118, 121 Kaya, M., 118, 121 Ke, Z., 164 Keese, M. A., 104 Keetch, C. A., 238 Kehres, D. G., 532 Kelly, A., 513 Kemp, G., 63 Kemsley, J. N., 7, 12, 29 Ken, C. F., 242 Kendall, S. L., 175, 186 Kendrew, J. C., 418 Kennedy, M. C., 192, 193 Kent, S. B. H., 122 Kent, T. A., 242 Kerr, E. A., 263, 265 Keulers, M., 502 Key, J., 382 Khan, I., 318, 337 Khan, S., 532, 533 Khan, J., 606 Khanna, A., 164 Khoroshilova, N., 192, 193, 202, 205 Kieny, M. P., 541 Kiger, L., 174, 212, 217, 238, 257, 302, 303 Kikuchi, K., 587, 588, 590, 591 Kiley, P. J., 192, 193, 194, 202, 205, 237 Kim, C. C., 114 Kim, C.-H., 174 Kim, D.-H., 118, 119, 121 Kim, H. S., 80 Kim, K., 440, 441, 450 Kim, S., 41 Kim, S. H., 73 Kim, S. O., 139, 193, 500, 501, 513, 522 Kim, T. H., 223 Kim, W. S., 589 Kimura, K., 119 Kincaid, J. R., 261, 265 King, P., 440, 441, 450 King, T., 504, 505 King, T. E., 298, 300 Kingston, R. E., 25 Kirino, Y., 587, 588, 591 Kirkpatrick, H. A., 505 Kiss, G. B., 597 Kissinger, C. R., 8
631
Author Index
Kitagawa, T., 261, 262, 264, 271 Kitazume, T., 118, 120, 121, 125, 127 Kizawa, H., 118, 125 Kjeldgaard, M., 8 Kleinert, T., 337 Klessig, D. F., 576 Kliger, D. S., 330 Klimpel, K. W., 543 Klinguer, A., 588 Klink, A., 105, 237 Klinman, J. P., 441 Klock, H. E., 5, 6 Kloek, A. P., 165 Klucas, R. V., 607, 610, 616, 617 Knapp, J. E., 384, 388, 389, 390, 391, 392, 399, 405, 407 Knapp, J. S., 543 Knowles, R. G., 142, 563, 565 Knudsen, S., 503 Kobayashi, I., 540 Koch, A. L., 509 Koch, C. J., 137 Koch, J., 30 Koev, C. A., 532 Ko¨hle, H., 577, 578, 579, 580, 581, 590, 591 Kohno, S., 502, 509 Koivisto, A., 142 Kojima, H., 587, 588, 590, 591 Kolisnychenko, V., 505 Kollman, P., 442, 466, 478, 479, 485 Kolonay, J. F., 80 Komarov, A. M., 586, 592 Komninos, P., 357, 418 Komori, M., 127 Kondo, T., 118, 120, 121, 125, 127 Konishi, R., 351 Konstantinov, A. A., 147 Kopelman, R., 137 Koper, M. T., 130 Kori, A., 504, 505 Korner, H., 192 Kort, R., 381, 386 Koshland, D. E., Jr., 403 Kostelecky, B., 29 Kostov, K. S., 392 Kotani, M., 331 Ko¨tting, C., 352 Kovacs, I., 597 Kovacs, L., 597 Kozlowski, P. M., 263, 276, 366 Kraayenhof, R., 151 Krachtus, D., 423 Krahling, S., 546 Krasselt, A., 336, 381, 386, 387, 388, 407, 409, 418 Kraus, D. W., 142, 149 Krebs, C., 242
Kriegl, J. M., 334, 336, 349, 355, 358, 364, 365, 370, 418, 419 Kroger, A., 64 Kroneck, P. M., 64, 81 Krylov, D., 164 Kuchnir, L., 421, 443, 448, 467, 478 Kuczera, K., 478 Kudo, T., 118, 125, 127, 128, 130 Kujat Choy, S. L., 532 Kundu, S., 288 Kuntz, I. D., Jr., 358, 398, 419, 440 Kunzler, P., 85 Kuriyama, H., 502 Kuroi, A., 236 Kurtz, D. M., Jr., 3, 4, 5, 8, 11, 24, 25, 26, 29, 30, 31, 32, 34, 40, 49, 51, 52 Kushkuley, B., 356 Kusian, B., 105 Kustu, S., 239 Kuszewski, J., 8 Kwiatkowski, L. D., 323 Kwok, F., 564 Kwon, E., 568, 570 L La Camera, S., 594 Laarhoven, L. J. J., 578, 579, 581, 590 Labarre, M., 273, 277, 488, 493, 494 Laboure, S., 381, 383, 405 Laemmli, U. K., 179 Lafontaine, J., 419, 460 Lai, C.-S., 586, 592 Lakey, J., 225 Lakner, F. J., 303 Lamas, S., 522 Lamattina, L., 562, 576, 597, 606 Lamb, C., 576, 577, 596, 606 Lamb, D. C., 349, 354, 356, 357, 361 Lambden, P. R., 542 Lamotte, O., 583, 588, 589, 590, 591 Lampreia, J., 64 Lancaster, J. R., Jr., 142, 149 Langebartels, C., 588, 589, 590 Lanzilotta, W. N., 5, 8, 24, 26, 40 Lapennas, G. N., 313 Lapidot, A., 303 Larfars, G., 137, 138 Laria, D., 479 LaRossa, R. A., 41, 42, 237, 502, 506, 509, 515 Larrinello, M., 486 Larsen, R. W., 164, 166, 171 Laskowski, R. A., 412 Lassmann, T., 164 Laszlo, F., 565 Latour, J. M., 104, 236 Lau, F. T. K., 478
632 Lavalette, D., 330, 336, 357, 418 Lawson, D. M., 295 Lay, P. A., 299 Lazarus, R. A., 567 Lazazzera, B. A., 192, 193 Le Brun, N. E., 191, 193, 194, 197, 202, 204, 205, 206, 515 Leach, A. R., 478 Leach, E. R., 64, 81, 212, 213 Leavitt, J., 122 LeBlanc, J. C., 332 Lebrun-Garcia, A., 583, 588, 589, 590, 591 Lecomte, J. T. J., 446, 448 Lecourieux, D., 583, 588, 589, 590, 591 Lee, H. I., 303 Lee, K. K., 606 Lee, L., 499 Lee, L. J., 505, 511 Lee, M., 80 Lee, S.-K., 7, 12, 29 LeGall, J., 4, 5, 7, 9, 11, 22, 31, 32, 33, 34, 38, 40, 48, 64 Legrand, A., 381, 383, 405 Lehle, H. M. K., 355, 365, 370 Lehnert, N., 7, 12, 29, 130 Leibovici, L., 540 Leigh, J. S., Jr., 298 Leikert, J. F., 590 Leisinger, T., 4, 22, 24, 29, 30, 48 Lemaitre, N., 114, 212 Lemon, D. D., 493 Lens, S. I., 542 Lentfer, A., 403 Leo´n, A. M., 562, 563, 564 Leshem, Y. Y., 562, 581 Lesley, S. A., 5, 6 Leslie, A. G., 7 Letunic, I., 164 Leung, Y. C., 89 Lev, O., 318 Levi, R., 532 Levin, E. J., 407 Levitt, M., 479 Lewis, R. S., 146 Lewitt-Bentley, A., 423, 426 Leyser, H., 337 Li, H., 274, 418, 419, 461 Li, T., 263, 274, 356, 357, 419, 461, 493 Li, X.-Y., 262 Li, Y., 542 Liable-Sands, L. M., 242 Liao, J. C., 509, 513 Liao, R. P., 175 Liaudet, L., 348 Libby, S. J., 522, 532 Libourel, I. G., 562, 568 Liddington, R., 321
Author Index
Lieb, J. D., 213, 216, 217, 220, 221, 222, 223 Liesegang, H., 105 Liesum, L., 291 Lilley, P. E., 500 Lim, M., 349, 350, 357, 381, 399, 409, 411, 418, 430 Lin, C. T., 242 Lin, E. C. C., 192 Lindermayr, C., 570 Linssen, A. B. M., 481 Lippard, S. J., 104 Lipscomb, J. D., 242 Lipton, S. A., 348 Lissenden, S., 542, 543 Litwiller, R., 242 Liu, C. A., 137 Liu, D., 118 Liu, J. S., 213, 216, 231 Liu, L., 42, 212, 522, 523, 526, 528, 532, 535, 562 Liu, M.-Y., 4, 5, 7, 8, 10, 11, 12, 13, 22, 26, 31, 32, 33, 34, 38, 40, 48, 51, 64 Lizasoain, I., 142 Ljungdahl, L. G., 3, 4, 5, 8, 11, 24, 25, 26, 29, 30, 31, 32, 34, 40, 49, 51, 52 Llewellyn, D. J., 596 Lloyd, D., 501, 502 Lo, L. W., 137 Lo, S. C. L., 564 Loake, G. J., 568, 570 LoBrutto, R., 298, 300 Loehr, T. M., 261 Lois, A. F., 174 Lomonosova, L. L., 608 Long, D. A., 258 Looger, L. L., 484 Lorkovic, I. M., 350 Loscalzo, J., 348 Losick, R., 213, 216, 231 Lowe, K. C., 318 Lowery, A. M., 568 Lu, C., 255, 265, 274, 278, 279 Lu, S., 82 Lubben, M., 81, 83, 88 Lubitz, W., 303 Lucchini, S., 41 Luck, S., 361 Ludovici, C., 81, 83, 88 Ludwig, M. L., 7, 16, 38 Lukacs, N., 597 Lukyanov, K. A., 598 Lukyanov, S., 598 Lundberg, J. P., 543 Lupia´n˜ez, J. A., 563 Luque, F. J., 430, 460, 477, 482, 486, 488, 489, 490, 491, 494, 562, 564, 568, 570 Lutz, R. S., 303 Lynch, J. B., 104
633
Author Index M Ma, F. H., 590 Ma, J., 419 Ma, J.-G., 261 Macedo, S., 4, 7, 8, 10, 11, 12, 13, 22, 26, 38, 48, 51 Machius, M., 174 Macieira, S., 64 MacKerell, A. D., Jr., 421, 422, 443, 448, 467, 478 MacMicking, J. D., 522, 524, 532 Maeda, M., 351 Maes, E. M., 355, 359, 371, 372, 373 MacMillan-Crow, L. A., 607 Magalhaes, J. R., 586, 587 Maguire, M. E., 532 Mahnane, M. R., 542, 551, 552 Maiden, M. C., 540, 545 Makaroff, C. A., 10 Malhotra, V., 175 Malinski, T., 137 Malkin, V. G., 294 Mamat, B., 16 Manac’h, N., 596, 597 Mandhana, N., 40, 41, 236, 237, 501 Manfredini, M., 323 Mannervik, B., 9, 10, 12 Mansy, S. S., 174 Ma¨ntele, W., 349, 351 Manukhina, E. B., 104 Maples, K. R., 585 Marchfelder, A., 29 Marchler-Bauer, A., 164 Marden, M. C., 174, 212, 217, 238, 302, 303, 354, 356, 357, 358, 359 Marechal, J., 460 Margreiter, R., 138, 149 Marino, M., 522 Markley, J. L., 38 Marletta, M. A., 137, 236 Marocco, A., 576, 577 Marriott, G., 350 Marriott, H. M., 80, 555, 558 Marsden, G., 277, 502 Marshall, M., 164 Marshall, V. P., 337 Marti, M. A., 430, 460, 477, 479, 482, 484, 486, 488, 489, 490, 491, 492, 493, 494 Martin, L. A., 274, 493, 501, 514 Martin, S. M., 587 Martin-Romero, F. J., 608 Martı´nez, J. I., 303 Martı´nez, M. C., 562 Martins, I. S., 562 Mascarenhas, R., 298, 300, 303 Mason, M. G., 135, 140, 141, 142, 144, 145 Mason, S. N., 555
Mastroeni, P., 522, 527, 532, 533 Mastronicola, D., 4, 141, 142, 144, 151 Masuya, F., 300 Mathews, A. J., 274, 493 Mathews, F. S., 165 Matias, P., 4, 7, 8, 10, 11, 12, 13, 22, 26, 38, 48, 51 Matorin, A. D., 66, 79 Matsuda, Y., 88 Matsui, T., 418 Matsumura, K., 118, 121 Matsuo, Y., 118, 121 Mattera, R., 303 Matthes, K., 337 Matthews, C. R., 338 Matthias, A., 142 Mattos, C., 478 Mau, B., 505 Maurer, T. S., 564 Maxwell, J. C., 265 Mayhew, G. F., 505 Mayhew, S. G., 38, 69 McAdams, H. H., 175 McCarthy, A. A., 38 McCarthy, N. D., 540 McCollister, B. D., 522, 523, 526 McCord, J. M., 205, 613 McCormick, J. M., 111 McCreery, R. L., 260 McDonald, B., 555 McDonald, J. D., 354, 356, 357, 358, 359 McFarland, M., 594 McGinnis, K., 532 McGrath, P. T., 175 McKelvey, E. J., 303 McKinnie, R. E., 418 McLendon, G., 50 McMahon, B., 349, 356, 357 McMahon, B. H., 356, 398, 418, 419, 480 McMahon, T. J., 568 McMullan, D., 5, 6 McRee, D. E., 7, 399 Medforth, C. J., 261 Megson, I. L., 104 Mehring, M., 291 Meissner, G., 348 Meleshkevitch, E. A., 164, 171 Meller, J., 419 Melville, S. B., 193 Membrillo-Herna´ndez, J., 41, 193, 500, 502, 513 Mendez, G., 138, 149 Meng, W., 192, 195 Merkx, M., 130 Mesecar, A. D., 403 Messana, I, 288 Messerschmidt, A., 64 Mettert, E. L., 193, 205
634 Metzger, A. L., 7, 38 Meunier, B., 138 Meuwly, M., 358, 367, 370, 418, 419 Meyer, C., 30 Meyer-Klaucke, W., 29 Michaud-Soret, I., 104, 236 Michel, C., 585, 586 Michnick, S., 423, 478 Miele, A. E., 381, 418 Milani, M., 267, 275, 337, 418, 425, 426, 430, 460, 467, 482, 488, 489, 490, 491 Miller, J. H., 238 Miller, J. L., 466 Miller, L. M., 367, 368 Mills, C. E., 139, 501, 513, 522 Mills, G., 421 Mills, P. C., 63 Mims, W. B., 300, 303 Minakuchi, K., 585, 586, 587 Minor, W., 7, 8 Mistry, J., 164 Misukonis, M. A., 555 Mitchell, T. J., 555, 558 Mitrikas, G., 303 Mo¨bius, K., 288, 293 Modolo, L. V., 586, 587 Moenne-Loccoz, P., 82, 175 Moens, L., 164, 212, 217, 238, 257, 268, 288, 297, 299, 302, 303, 304, 337, 419, 425, 427, 431, 448, 460, 461, 462, 467 Moffat, K., 357, 381, 382, 383, 386, 387, 392, 399, 405, 406, 407, 408, 409, 411, 418, 440 Moh, P. P., 354, 356, 357, 358, 359 Mohan, S., 69, 542 Mohan, V. P., 502, 509 Moir, J. W., 64, 80, 81, 212, 213, 513, 528, 539, 542, 543, 551, 552, 554, 555, 558 Molle, V., 213, 216, 231 Momose, K., 590 Monack, D., 114 Moncada, S., 136, 142, 151 Mongodin, E. F., 80 Monk, C. E., 499 Monod, J., 320, 323 Monson, E. K., 174 Montellano, P. R., 175 Montfort, W. R., 351, 355, 359, 370, 371, 372, 373 Moore, C. M., 502, 509 Moore, D. D., 25 Moore, L. J., 237 Morais, J., 5, 7 Moran, J. F., 605, 607, 610, 615, 616, 617 Moras, D., 423 Mordvintcev, P. I., 113 Morett, E., 502 Morgan, J. E., 153
Author Index
Mori, H., 216 Morikawa, H., 562, 566 Morishima, I., 418 Morkuniene, R., 502 Moro, M. A., 142 Moroz, L. L., 589 Morreale, A., 430, 460, 486, 489, 494 Moran, J. F., 607, 610 Morse, R. H., 295 Mouawad, L., 460 Moura, I., 64, 242 Moura, J. J., 64, 242 Mourant, J. R., 331, 349, 356, 358, 359, 361 Movahedzadeh, F., 175, 186 Mowant, J. R., 331 Moxon, S., 164 Mozzarelli, A., 311, 313, 315, 316, 318, 320, 321, 322, 323, 324, 329, 330, 333, 335, 337, 338, 339, 340, 388 Mueller, M. J., 585, 586 Mukai, M., 174, 255, 265, 267, 268, 271, 275, 276, 460 Mukhopadhyay, P., 41, 42, 237, 502, 506, 509, 515 Mulks, C. F., 303 Mu¨ller, J. D., 349, 356, 357 Mulsch, A., 104, 113 Munck, E., 104, 193, 202, 205, 242 Mun˜oz, I. G., 607, 615 Munro, A. W., 348 Mur, L. A., 578, 579, 581, 588, 590 Murphy, M. E., 578, 579, 581, 590 Murray, D. B., 502 Murgia, I., 606 Murshudov, G. N., 8 Musselman, R. L., 111 Mustardy, L., 597 N Nadler, E. P., 532 Nagai, K., 262 Nagano, T., 587, 588, 590, 591 Nagata, N., 590 Nagoshi, H., 587, 588, 591 Nakahara, K., 118, 121, 127, 129, 130 Nakamoto, K., 258, 263, 264 Nakamura, H., 125, 174 Nakamura, K., 174 Nakano, M., 522 Nakano, M. M., 502, 509 Nakao, K., 540 Nakase, T., 119 Nakatsubo, N., 587, 588, 591 Nardini, M., 277, 302, 461, 486 Nashef, A. S., 204 Natal da Luz, H., 29 Natanson, C., 532 Nathan, C., 522
635
Author Index
Nathan, C. F., 522, 524, 532 Nauser, T., 130, 246 Navani, N. K., 267, 276, 460 Naylor, S., 242 Nebenfu¨hr, H., 291 Nedergaard, J., 142 Neese, F., 7, 12, 29, 64, 294 Negano, T., 590 Neill, J. M., 312 Neill, S. J., 562, 576, 584, 597 Nelson, K. E., 80 Ng, K., 399 Ng, K. Y., 4, 49, 51, 52 Ngo, T., 478 Nguyen, D. T., 478 Nicholls, P., 135, 140, 141, 142, 144, 145, 149, 302 Nichols, J. C., 389 Nicholson, S., 522 Nicklen, S., 124 Nielsen, J., 503 Nienhaus, G. U., 331, 334, 336, 337, 347, 349, 351, 354, 355, 356, 357, 358, 359, 361, 364, 365, 370, 371, 372, 373, 381, 386, 387, 388, 407, 409, 418, 419 Nienhaus, K., 336, 337, 347, 349, 351, 354, 355, 357, 358, 359, 364, 365, 370, 371, 372, 373, 381, 386, 387, 388, 407, 409, 418, 419 Nierman, W. C., 80 Nilges, M., 8 Ninfa, A. J., 175 Ninfa, E. G., 175 Nishida, C. R., 318 Nishida, F., 318 Nistor, S. V., 212, 217, 238 Niwa, H., 238 Noack, E., 578, 579, 581, 590 Nobel, A. B., 220, 221, 222, 223 Noble, R. W., 323 Nobre, L. S., 41 Nocek, J. M., 304 Noctor, G., 569 Noda, K., 590 Noe´, F., 421, 422, 423 Nolling, J., 30 Notley-McRobb, L., 504 Novikov, E., 336, 357, 418 Nowak, W., 461, 462, 467 Nunoshiba, T., 514 Nurizzo, D., 407 Nutt, D. R., 367, 370, 419 O O’Conner, C. D., 542 O’Donnell, V. B., 140 O’Farrell, P. A., 38 O’Hara, J., 137
Obayashi, E., 125 Obika, K., 123 Oda, M., 118, 125 Oddou, J. L., 104 Ogata, C. M., 7 Ogura, T., 264 Ohno, H., 502, 509 Okamoto, M., 597 Okamoto, N., 125, 127 Oki, M., 590 Olafson, B. D., 419, 422, 425, 428, 442, 461 Oldfield, E., 299 Olin, B., 9, 10, 12 Oliveira, R., 11, 48, 51 Oliveira, S., 4, 7, 8, 10, 11, 12, 13, 22, 24, 26, 31, 32, 33, 34, 38, 40, 48, 49, 51, 52 Oliveira, T. F., 64 Olsen, J. S., 357, 358, 418 Olsen, K., 423 Olsen, K. W., 417, 439, 459 Olson, J. S., 170, 174, 263, 264, 271, 274, 276, 355, 356, 357, 358, 365, 366, 381, 384, 398, 399, 402, 405, 406, 407, 409, 411, 418, 419, 430, 440, 461, 492, 493, 596 Omura, T., 130 Ondrias, M. R., 261, 262 Ong, J., 304 Onozuka, K., 522 Onufriev, A., 481 Ordal, G. W., 164, 166, 171 Ordejo´n, P., 479, 486 Orii, Y., 500, 501 Orlinskii, S. B., 293 Orlowski, S., 461, 462, 467 Ormos, P., 349, 354, 356, 358, 359, 361 Ornatowski, W., 502, 509 Orozco, M., 430, 460, 482, 486, 488, 489, 490, 491, 494 Orozco-Ca´rdenas, M. L., 593 Osaki, T., 88 Osborne, C., 38 Osborne, J. P., 304 Oshima, R., 125, 128, 129 Osmulski, P. A., 303 stergaard-Jensen, E., 589 Ostermann, A., 336, 357, 358, 418 Osuga, D. T., 204 Osvath, S. R., 301, 302 Ottolenghi, M., 318 Otwinowski, Z., 7, 8 Oubrie, A., 543 Ouellet, H., 165, 265, 267, 268, 273, 274, 275, 276, 488, 493, 494 Ouellet, Y., 165, 265, 267, 268, 269, 271, 272, 273, 274, 275, 337, 418, 425, 426, 430, 460, 467, 488, 489, 493, 494 Ouellet, Y. H., 165, 274, 275, 276 Overton, T. W., 40, 213, 238, 542
636
Author Index
Ozaki, S., 418 Ozaki, Y., 261, 262 P Pacheco, I., 5, 7 Pacher, P., 348 Packer, L., 295, 348 Padmaja, S., 348 Page, M. D., 89 Pagnussat, G., 562 Pagnussat, G. C., 606 Pahl, R., 336, 380, 381, 384, 385, 386, 387, 388, 389, 390, 391, 392, 399, 405, 406, 407, 409, 418 Pai, E. F., 403 Paily, P., 323 Paiva, C. L., 350 Palacios-Callender, M., 136, 142, 151 Palazzo, G., 333 Pallas, J. A., 568 Palma, J. M., 562, 563, 564 Palomares, A., 174 Panek, A. D., 350 Pannu, N. S., 8 Pant, N., 175 Paoli, M., 265 Papiz, M. Z., 8 Parak, F., 336, 357, 358, 359, 381, 386, 387, 388, 398, 407, 409, 418 Pardanani, A., 388, 389 Pares, S., 9 Pares, X., 567 Park, S., 313, 484 Park, S. F., 270, 277, 278, 279, 502 Park, S.-Y., 125, 130 Parker, A., 545 Partridge, J. D., 506 Paston, S. J., 513 Patchkovskii, S., 298 Patel, M. D., 40, 238 Patel, R. P., 142, 146, 149, 151 Pathania, R., 267, 276, 460 Patschkowski, T., 193 Pattridge, K. A., 7, 38 Paul, M., 540 Paulat, F., 130 Pauleta, S. R., 89 Paulsen, I. T., 80 Paulsen, K. E., 38 Pauly, A., 607 Peacock, W. J., 419, 460, 596 Pearson, A. R., 315, 321, 441 Pearson, I. V., 89 Peck, H. D., Jr., 64 Pedraza, A. J., 367, 368 Peeters, K., 257, 419, 460 Peisach, J., 301, 302, 303
Peitsch, M. C., 425 Pejchal, R., 16 Peleato, M. L., 29 Penn, C. W., 40, 238, 277, 502 Perahia, D., 460 Perazzoli, M., 596 Perdew, J. P., 486 Pereira, A. S., 104 Pereira, I. A., 64 Pereyra, M. C., 225 Perez-Gonzalez-de-Apodaca, J., 318, 337 Perez-Martin, J., 239 Perlman, D. A., 442, 478, 479 Perman, B., 381, 382, 386 Perna, N. T., 505 Peruta, M., 288 Perutz, M. F., 174, 323 Pesce, A., 267, 277, 288, 302, 323, 337, 425, 426, 427, 431, 448, 460, 461, 462, 467, 489 Petersen, L. C., 149 Peterson, E., 337 Peterson, E. S., 388, 389 Petratos, K., 403 Petry, W., 337 Petsko, G. A., 358, 398, 399, 403, 419, 440, 478 Philipp, R., 349, 361 Phillips, D. R., 418 Phillips, G. N., Jr., 168, 170, 174, 263, 271, 274, 276, 355, 356, 357, 381, 399, 407, 409, 411, 418, 419, 430, 440, 461, 492, 493 Phillips, J. C., 442, 448, 460, 461, 467 Piantadosi, C. A., 137 Piatibratov, M., 164, 168, 171 Picard, V., 532 Pickering, I. J., 303 Pickford, A., 502 Picorel, R., 303 Pieters, J., 136 Pihl, T. D., 30 Piknova, B., 300 Pilbrow, J. R., 288 Pils, B., 164 Pinchasov, Y., 581 Pinkert, S., 164 Pioselli, B., 318 Piper, M. D. W., 503 Pirt, S. J., 504, 505, 506, 512 Planchet, E., 582, 588, 589, 590, 591 Plato, M., 288, 293 Plech, A., 402, 405 Ploegh, H., 136 Plunkett, G., 505 Poderoso, J. J., 136 Pohajdak, B., 257 Pohlmann, A., 82, 105, 106, 114, 236, 237 Poinssot, B., 588, 589 Polverari, A., 594 Ponce-Coria, J., 502
637
Author Index
Poock, S. R., 63, 64, 81, 212, 213 Poole, R. K., 41, 42, 81, 104, 137, 139, 193, 199, 201, 212, 217, 236, 237, 238, 255, 257, 265, 270, 274, 277, 278, 279, 499, 500, 501, 502, 505, 506, 507, 508, 509, 510, 511, 512, 513, 514, 515, 522, 528, 531, 557, 558 Popescu, C., 193, 202, 205 Porcella, S. F., 114, 212 Porter, S., 239 Porterfield, D. M., 562 Posfai, G., 505 Posfai, J., 505 Post, F., 337 Postma, J. P. M., 481 Potgieter, K., 589 Potoka, D. A., 532 Potter, L. C., 64, 66, 67 Potter, M., 105 Potter, S., 522 Potter, W. T., 354 Poyart, C., 174 Pradervand, C., 381, 383, 386, 399, 405, 408, 409, 411, 418, 440 Prado, A. M., 562 Praneeth, V. K. K., 130 Prasher, D. C., 597 Prats, E., 588 Premer, S. A., 460 Preziosi, M. P., 541 Price, D. C., 29 Prince, R. C., 303 Prior, L., 81, 91, 93, 94, 95, 96, 97 Prisner, T. F., 293 Prodhom, B., 478 Pronk, J. T., 503 Provost, A., 503 Prusakov, V., 357 Pugin, A., 583, 588, 589, 590, 591 Pullan, S. T., 41, 42, 193, 237, 499, 502, 505, 507, 508, 509, 510, 512, 513, 514, 515 Puntarulo, S., 587 Puppo, A., 607, 617 Purol-Schnabel, S., 579 Q Qiu, Y., 351, 371 Qu, C., 223 Qu, K., 305 Qu, Z.-L., 588, 597 Quammie, N. C., 92 Que, L., 562 Que, L., Jr., 104 Quezado, Z. M., 532 Quillin, M. L., 263, 274, 356, 357, 419, 461 Quiniou, E., 336, 418 Quinn, R., 301 Quint, P., 615
Quintero, M., 136, 142 Quiro´s, M., 562, 563, 564 R Rademaker, H., 141, 151 Radford, S. E., 89 Radi, R., 606 Radik, J., 139 Radolf, J. D., 242 Ragettli, S., 4, 22, 24, 29, 30, 48 Raitsimring, A. M., 303 Rajagopal, S., 381, 387, 392, 399, 406, 407 Rajamohan, G., 267, 276 Rakhmatullin, R., 291 Ralph, E. T., 193, 194, 196 Ralph, E. T. J., 193, 202 Randin, O., 590 Randle, P. J., 507 Randler, R., 402, 405 Rapp, G., 403 Rappas, M., 238 Rappelli, P., 4 Ra¨thel, T. R., 590 Rau, H. K., 303 Ravanel, S., 566 Ravel, J., 80 Ravi, N., 104 Ray, A., 268, 337, 460 Ray, M., 242 Read, R. C., 41, 42, 80, 81, 193, 212, 507, 508, 513, 515, 521, 528, 531, 542, 543, 545, 551, 552, 554, 555, 557, 558 Read, R. J., 8 Read, T. D., 80 Reader, J. S., 89 Rebeil, R., 114, 212 Rebeille, F., 566 Rechsteiner, M. P., 274 Reed, G. H., 298 Reeder, B. J., 151 Reem, R. C., 111 Reeve, J. N., 30 Regan, R., 274, 418, 419, 461 Regan, T., 542 Regenberg, B., 503 Reiher, W. E. III, 478 Reijerse, E. J., 298, 299, 303 Reijnders, W. N., 82, 542 Reinecke, F., 105 Reinhoudt, D., 423 Reinisch, L., 354, 356, 357, 358, 359 Reiter, C. D., 606 Ren, B., 223 Ren, Z., 381, 382, 386, 387, 399, 405, 407, 408, 409, 411, 418, 440 Renaud, J. P., 423 Reppas, N., 213
638 Reyes-Ramirez, F., 239 Reynolds, A. H., 354, 356, 357, 358, 359 Reyrat, J.-M., 174 Rheingold, A. L., 242 Rhen, M., 41 Ribeiro, E. A., 562 Rice, L. M., 8 Rich, A. M., 299 Rich, P. R., 138 Richard, C., 273, 277, 488, 493, 494 Richards, F. M., 398 Richardson, D. E., 111 Richardson, D. J., 63, 64, 66, 68, 69, 71, 73, 74, 79, 80, 81, 82, 85, 86, 89, 90, 91, 94, 95, 96, 97, 212, 213 Richmond, T. A., 223 Richter-Addo, G. B., 480, 486 Ridderstrom, M., 9, 10, 12 Ridsdale, A., 337, 354 Rigaud, J., 607, 617 Riggs, A., 461 Riley, C. W., 164, 166 Riley, M., 505 Rison, S. C., 175, 186 Rivetti, C., 313, 315, 316, 318, 321, 322, 323, 335, 388 Rivoal, J., 596 Rizos, A. K., 297 Roach, M. P., 418 Robert, C. H., 323 Roberts, D. M., 175 Roberts, G. C. K., 423 Roberts, S. A., 351, 371 Robinson, C. V., 238 Roccantano, D., 418 Roche, C. J., 337 Rock, J. D., 542, 543, 551, 552, 554 Rockel, A., 593 Rockel, P., 593 Rode, C. K., 505 Rodgers, K. R., 279 Rodionov, D. A., 105, 212, 225, 236 Rodrigues, J. V., 33, 39, 54, 55, 60 Rodrigues, M. L., 64 Rodrigues, R., 4, 24, 26, 40, 49, 51, 52 Rodrigues-Pousada, C., 4, 7, 8, 10, 11, 12, 13, 22, 24, 26, 31, 32, 33, 34, 38, 40, 48, 49, 51, 52 Rodrı´gues-Serrano, M., 562, 566, 568, 570 Rodriguez-Lopez, J. N., 263, 264, 265 Roesen, R., 579 Roger, A. J., 4, 23 Rogers, M. S., 151 Rogers, N. E., 592 Rogers, P. H., 323 Rohlfs, R. J., 274, 418 Roitberg, A., 461, 465, 466, 479, 484, 489, 493 Rokop, M. E., 213 Roldan, M. D., 81, 91, 94, 95, 96, 97
Author Index
Rollinghoff, M., 532 Romao, M. J., 64 Romero-Puertas, M. C., 562, 563, 564, 566, 568, 570, 596 Ronda, L., 311, 318, 320, 329, 333, 337, 338, 339, 340 Ronda, R., 324 Ronning, C. M., 80 Rosano, C., 323 Rose, D. J., 505 Rosenzweig, N., 137 Rosenzweig, Z., 137 Ross, W. S., 442, 478, 479 Rossi, G. L., 313, 315, 316, 321, 322, 323, 388 Roth, M., 381, 383, 405 Rousseau, D. L., 165, 260, 261, 262, 265, 267, 268, 269, 271, 272, 273, 274, 279, 460 Rousseau, P., 174, 181 Roux, B., 422, 478 Royant, A., 399, 403 Royer, W. E., Jr., 379, 384, 388, 389, 390, 391, 392, 399, 405, 407, 461 Ruan, J., 522 Rubino, S., 214 Rumsey, W. L., 137 Rush, T. S. III, 366 Russell, D. W., 195 Ryan, C. A., 593 Rybak-Akimova, E. V., 178 S Saalbach, G., 570 Saarinen, M., 522 Saavedra, R. A., 122 Sahr, T., 566 Saigo, S., 318, 335 Saini, D. K., 175 Saito, J. A., 163, 164, 166, 288 Saito, S., 522, 606 Sakaguchi, Y., 80 Sakamoto, A., 562, 566 Sakihama, Y., 590 Sakurai, N., 81 Sakurai, T., 81 Salgano, I., 586, 587 Salmi, M., 522 Salzman, A. L., 501, 514 Sambrook, J., 195 Sampath, V., 264 Samuni, U., 165, 268, 274, 275, 276, 333, 337, 418, 430, 460 Sanchez-Baeza, F., 29 Sa´nchez-Portal, D., 486 Sanciu, G., 4 Sandalio, L. M., 562, 563, 564, 566, 568, 570 Sanders, C. R. II, 303 Sanderson, R., 588
Author Index
Sandstrom, T., 585 Sanger, F., 124 Santero, E., 239 Santos, H., 4, 5, 22, 40, 48 Santosa, I. E., 578, 579, 581, 590 Saraiva, L. M., 4, 8, 11, 14, 15, 21, 22, 23, 24, 25, 30, 31, 40, 41, 42, 48, 49, 51, 52, 124, 193, 212, 237, 502, 509, 515 Saraste, M., 81, 83, 88, 93, 543 Sarath, G., 607, 610, 617 Sardiwal, S., 175, 186 Sargeant, R., 16 Sarkela, T. M., 136 Sarti, P., 4, 33, 39, 40, 47, 50, 54, 55, 60, 141, 142, 144, 147, 151 Sarver, A., 606 Sass, L., 597 Sato, R., 130 Sauke, T. B., 349, 356 Sauls, D. L., 555 Saunders, N. J., 542 Savard, P. Y., 165, 265, 267, 268, 274, 275, 276 Savino, C., 408 Savitsky, A., 288, 293 Sawers, G., 192 Sawicki, C. A., 141 Scaiano, J. C., 330 Scandurra, F. M., 33, 39, 40, 47, 54, 55, 60 Scatena, R., 288 Schapiro, J. M., 522 Schechter, A. N., 300 Scheidt, W. R., 298, 301, 302 Schellenberg, P., 357 Scherlis, D. A., 479, 486, 492 Schildkamp, W., 381, 383, 386, 399, 405, 408, 409, 411, 418, 440 Schilling, O., 29 Schimdt, J., 293 Schimdt, M., 381, 386, 399, 406 Schindler, F. J., 149 Schirawski, J., 193 Schirmer, R. H., 104 Schirmer, T., 423 Schlegel, R. A., 546 Schlenkrich, M., 478 Schlichting, I., 357, 398, 399, 403, 411, 418, 425, 480 Schlick, T., 419 Schmelz, K., 105, 106, 114, 236, 237 Schmidt, M., 336, 380, 381, 382, 385, 386, 387, 388, 399, 407, 409, 411, 418, 440 Schmidt, P. P., 298, 299 Schnegg, A., 288, 293 Schneider, R. T., 357, 418 Schneider, T. R., 7 Scholes, C. P., 296, 297, 298, 300, 302, 303, 304, 305
639 Schotte, F., 381, 382, 384, 386, 387, 399, 402, 405, 406, 407, 409, 411, 412, 418, 419, 430 Schroder, I., 81, 82 Schulten, K., 418, 429, 439, 440, 441, 442, 443, 448, 450, 451, 452, 453, 460, 461, 466, 467, 472, 484 Schultz, J., 164 Schultz, S. C., 192 Schulz, C. E., 298 Schulz, H., 85 Schumacher, J., 174, 181, 238 Schuster-Bockler, B., 164 Schuurmans, J. J., 151 Schwartz, E., 105 Schwarz, W., 337 Schwarzl, S. M., 421, 422, 423 Schweiger, A., 288, 291, 293, 303, 304, 305 Schweitzer-Stenner, R., 261, 262 Schwedhelm, E., 606 Sciara, G., 381, 399, 407, 409, 411, 418 Sciotti, M. A., 174 Scopes, R. K., 177 Scott, C., 104, 192, 193, 199, 201, 236, 501, 506, 513, 515 Scott, E. E., 365, 398, 409, 411, 418, 492, 493 Scott, T. W., 262 Scott, W. G., 174 Sebbane, F., 114, 212 Sedelnikova, S., 501, 513 Seedorf, H., 4, 5, 6, 8, 16, 23, 24, 26, 40, 51, 52 Seibel, G., 442, 478, 479 Seibert, M., 440, 441, 450 Seidman, J. G., 25 Seki, H., 351 Selmer, T., 193 Sequeira-Legrand, A., 583, 588, 589, 590, 591 Seregelyes, C., 597 Seshadri, R., 80 Setchell, K. D., 493 Settles, M., 337 Seward, H. E., 68, 69, 73, 74, 244 Seymour, J. L., 567 Shami, P. J., 555 Shannon, C. F., 318, 337 Shao, Y., 505 Shapiro, A. D., 562 Shapiro, L., 175 Sharpe, M. A., 140, 302 Sharrocks, A., 193 Shaw, D. C., 500 Shaw, J. G., 542, 551, 552 Sheldrick, G. M., 7 Shelnutt, J. A., 261, 351, 371 Shepherd, M., 135 Sherman, D. R., 175
640 Shibayama, N., 318, 335 Shields, G. C., 192 Shigenaga, M. K., 606 Shiloh, M. U., 522 Shima, S., 4, 5, 6, 8, 16, 23, 24, 26, 40, 51, 52 Shimada, H., 356 Shimizu, H., 125 Shin, D., 213 Shin, M., 316, 318 Shindyalov, I. N., 423, 425, 448 Shinomiya, H., 522 Shiro, Y., 121, 125, 127, 129, 130, 174 Shiva, S., 142, 146, 149 Shokhireva, T. K., 303, 373 Shore, V. C., 418 Shoun, H., 80, 117, 118, 119, 120, 121, 123, 125, 127, 128, 129, 130 Shyamsunder, E., 354, 356, 357, 358, 359 Sibanda, B. L., 480 Siddiqui, R. A., 106 Siegbahn, P. E., 493 Sieker, L., 4, 5, 7, 8, 10, 11, 12, 13, 22, 26, 38, 48, 51 Sienkiewicz, A., 305 Sikkink, R., 242 Silaghi-Dumitrescu, R., 3, 4, 5, 8, 11, 24, 25, 26, 29, 30, 31, 32, 34, 40, 49, 51, 52, 130 Silcocks, P. B., 545 Silva, G., 4, 7, 8, 10, 11, 12, 13, 22, 26, 31, 32, 33, 34, 38, 40, 48, 51 Simmerling, C., 466 Simon, J., 64 Simon, M. I., 175, 187 Simontacchi, M., 587 Singel, D. J., 348 Singer, M. A., 223, 224 Sitton, V. R., 193, 205 Sivozhelezov, V., 164, 171 Six, S., 192, 193 Sjogren, A. M., 4, 23 Skarzynski, T., 321 Skavrutski, I. V., 423 Skeel, R. D., 442, 448, 460, 461, 467 Skulan, A. J., 7, 12, 29 Sligar, S. G., 274, 276, 303, 356, 398, 409 Smirnov, A. V., 381, 399, 409, 411, 418, 430 Smith, A. R., 578, 579, 581, 590 Smith, A. T., 265 Smith, D. T., 288 Smith, G. K., 555 Smith, H., 542 Smith, J. A., 25 Smith, J. C., 421, 422, 423, 426, 478 Smith, J. L., 381 Smith, K. M., 262, 279 Smith, R. D., 493 Smith, W. E., 260, 261
Author Index
Smulevich, G., 261, 263, 264, 265, 279, 280 Snyder, L. A., 542 Soares, A. S., 480 Soballe, B., 501, 513 Sober, H. A., 29 Soddard, B. L., 381, 382 Soderback, E., 239 So¨derberg, C. A. G., 63 Sofia, H. J., 192 Sokol, K., 524, 532 Soler, J. M., 486 Solomon, E. I., 7, 12, 29, 111, 294 Soltis, M., 174 Soman, J., 381, 384, 399, 402, 405, 406, 407, 409, 411, 418, 419, 430 Sommer, J. H., 340 Song, X.-Z., 261 Sonnhammer, E. L., 164 Sonoda, M., 582, 588, 589, 590, 591, 596 Sopkova-De Oliveira Santos, J., 423, 426 Sordel-Klippert, M., 16 Sottini, S., 318, 333, 334, 337, 338, 339, 340 Soupene, E., 174 Sousa, E. H. S., 168, 174, 175, 176, 177, 180, 181, 182, 184, 185 Souza, J. M., 615 Spath, B., 29 Spemann, D., 29 Spencer, J., 10, 12 Spiess, H. W., 293 Spiro, S., 40, 41, 63, 80, 81, 85, 86, 91, 94, 95, 96, 97, 105, 211, 212, 215, 216, 217, 221, 222, 230, 236, 237, 238, 239, 240, 242, 243, 501, 502 Spiro, T. G., 260, 261, 262, 263, 276, 279, 366 Springer, B. A., 274, 276, 356 Springett, R. J., 136, 142, 151 Spyrakis, F., 333, 337, 339 Sˇrajer, V., 336, 357, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 399, 405, 406, 407, 408, 409, 411, 418, 440 Srivastava, R. B., 263 St. Pierre, B., 419, 460 Stach, P., 64 Stamler, J. S., 42, 138, 212, 236, 348, 501, 522, 523, 526, 528, 532, 535, 562, 568, 606 Stanbury, D. M., 140, 563 Stanfield, S. W., 174 Stankovich, M. T., 38 States, D. J., 419, 422, 425, 428, 442, 461 Stavrov, S. S., 356 Steinbach, P. J., 328, 331, 337, 361 Steinlechner-Maran, R., 138, 149 Steitz, T. A., 192 Stenger, S., 532 Stenkamp, R. E., 111
641
Author Index
Stephens, C., 175 Stephens, P. J., 111 Stepnoski, R. A., 262 Sternberg, M. J. E., 480 Stettmaier, K., 585, 586 Stevanin, T. M., 41, 42, 80, 81, 139, 193, 212, 507, 508, 513, 515, 521, 522, 528, 531, 539, 543, 545, 557, 558 Stevens, K., 524, 532 Stevenson, C. E. M., 295 Stewart, V., 41, 193, 500, 513 Stock, A. M., 175 Stock, J. B., 175 Stockman, B. J., 38 Stoddard, B. L., 403 Stoker, N. G., 175, 186 Storz, G., 41, 42, 237, 502, 506, 509, 515 Stote, R., 478 Stotler, W. F., 315 Stouthamer, A. H., 81, 82 Strampraad, M. J., 81, 82 Strandberg, R. E., 418 Straub, J. E., 419, 421, 461, 467, 478 Streeter, I., 92 Strickland, N., 373 Strittmatter, A., 105 Strouse, C. E., 301 Strube, F., 593 Strube, K., 103, 104, 105, 106, 107, 109, 114, 237 Struhl, K., 25, 213 Stubauer, G., 50 Stubna, A., 193 Studholme, D. J., 105, 237, 238, 239 Stuehr, D. J., 522 Stukan, R. A., 104 Sturdevant, D. E., 114, 212 Su, F., 125, 128, 129 Su, Y. F., 585 Sugiyama, J., 119 Suharti, R. C., 81, 82 Sukanek, P. C., 149 Su¨ltemeyer, D., 577, 578, 579, 580, 581, 590, 591 Sun, Y., 305 Sundholm, D., 303 Susin, S., 29 Sutton, V. R., 193, 194 Suzuki, T., 316, 318, 493 Svistunenko, D. A., 302 Swallen, S. F., 137 Swaminathan, S., 419, 422, 425, 428, 442, 461 Swanson, J. A., 137 Swartz, H. M., 137 Sweedler, J. V., 589 Sweet, R. M., 357, 398, 399, 411, 418, 425, 480 Swenson, R. P., 7, 38 Symons, M. C. R., 288, 302
Szabo, A., 484 Szaraz, S., 354 Szumowski, J., 297 T Taha, Z., 137 Tajkhorshid, E., 441, 442, 448, 453, 460, 461, 467 Takahashi, S., 418, 590 Takasugi, M., 585, 586, 587 Takaya, N., 80, 118, 119, 120, 121, 125, 127, 128, 129, 130 Tamai, E., 213 Tamamoto, H., 298 Tamashiro, W. M., 562 Tang, X., 553 Tang, Y., 506 Tang, Z., 597 Tanimot, T., 118, 121, 130 Tanimoto, T., 118 Tannenbaum, S. R., 514 Tarricone, C., 323 Tavan, P., 486 Tavares, P., 104 Taylor, B. L., 187 Taylor, C. P. S., 301 Taylor, I. A., 10, 12 Taylor, W. B., 332 Teixeira, M., 3, 4, 7, 8, 9, 10, 11, 12, 13, 14, 15, 21, 22, 23, 24, 25, 26, 29, 30, 31, 32, 33, 34, 35, 38, 39, 40, 41, 42, 47, 48, 49, 50, 51, 52, 53, 54, 55, 58, 60, 124, 193, 212, 237, 502, 509, 515 Teng, T., 381, 383, 399, 408, 409, 411, 418 Teng, T. Y., 357, 381, 382, 386, 399, 405, 418, 440 ter Heerdt, P., 304 Terwilliger, T. C., 390, 407 Teter, M. M., 411 Tetreau, C., 330, 336, 357, 418 Tettelin, H., 80 Thauer, R. K., 4, 5, 6, 8, 16, 23, 24, 26, 30, 40, 51, 52 Thomas, D. D., 606, 608, 612 Thompson, A., 41 Thompson, R. P., 569 Thompson, S., 543 Thomson, A. J., 69, 73, 74, 81, 91, 93, 94, 95, 96, 97, 104, 191, 193, 194, 196, 197, 199, 201, 202, 204, 205, 206, 236, 244, 501, 513, 515 Thomson, M. J., 539, 542, 543, 554 Tho¨ny-Meyer, L., 85 Thorndycroft, F. H., 66, 79, 82, 86, 89, 90 Thorneley, R. N. F., 263, 264 Thornton, J. M., 480 Thuring, H., 532 Thurston, C. F., 509
642
Author Index
Tilley, G. J., 92 Tilton, R. F., Jr., 358, 398, 419, 440 Timmins, G. S., 502, 509 Titarenko, E., 562 Toda, N., 590 Tomchick, D. R., 174 Tomita, T., 264 Tommasi, F., 592 Tomura, D., 118, 123, 125 Toritsuka, N., 118, 121 Torres, J., 142, 153 Torrie, G. M., 483 Touati, D., 236 Tourbez, M., 357 Townsend, R., 545 Trageser, M., 192, 193 Tran, L. M., 509, 513 Trandafir, F., 297, 299, 304 Travaglini-Allocatelli, C., 398, 399, 408, 411 Trent, J. T. III, 288, 460 Trevaskis, B., 596 Troger, W., 29 Truchet, G., 174 Trumbauer, M., 524, 532 Tsang, A. W., 137 Tsernoglou, D., 399 Tsubaki, M., 262, 263 Tsuchiya, K., 585, 586, 587 Tsukamoto, K., 121, 129, 130 Tsuneshige, A., 138, 295, 299, 313, 323 Tsuruta, K., 125 Tsuruta, S., 119 Tu, Y., 597 Tucker, M. P., 354 Tucker, N. P., 105, 212, 217, 236, 237, 238, 239, 240, 242, 243 Tuckerman, J., 168 Tuckerman, J. R., 174, 175, 176, 177, 180, 181, 182, 184, 185 Tunbridge, A. J., 80 Turk, T., 88 Turner, S. M., 277, 502 Tyagi, J. S., 175 Tymes, N., 225 Tyryshkin, A. M., 298, 299 U Uchida, T., 418 Uchiyama, H., 119 Uchiyama, I., 540 Ueda, M., 562, 566 Ueno, T., 104 Ukai, Y., 590 Ullah, J. H., 10, 12 Ullrich, V., 118, 130 Ulrich, R. L., 80 Umayam, L., 80
Umemura, M., 125 Unden, G., 192, 193 Urano, Y., 588, 590 Urarte, E., 605 Urbani, A., 543 Ursby, T., 381, 383, 386, 399, 405, 407, 408, 409, 411, 418, 440 Urwin, R., 540 Usov, O., 296, 297 Usuda, K., 118, 121, 130 Utilsky, A., 461, 467 Utterback, S. G., 297 Uzzau, S., 214 V Vagin, A. A., 8 Valderrama, R., 561, 562, 563, 564, 566, 568, 570 Valentine, J. S., 301, 318 Valkier, L. J., 151 Vallance, B. A., 522 Valleau, J. P., 483 Vallely, D., 321 Vallone, B., 337, 381, 397, 398, 399, 405, 407, 408, 409, 411, 412, 418, 480 Van Beeumen, J., 89 van Bentum, P. J. M., 303 Van Camp, H. L., 300, 303 Van de Peer, Y., 257, 419, 460 van der Horst, E., 303 van der Oost, J., 81, 82 Van Doorslaer, S., 212, 217, 238, 287, 291, 297, 299, 302, 303, 304 Van Gelder, B. F., 141, 151 Van Gunsteren, W. F., 481 Van Hauwaert, M. L., 460 van Kan, P. J. M., 303 Van Marum, D., 151 Van Slyke, D. D., 312 van Spanning, R. J., 81, 82, 89, 542 Van Stokkum, I. H., 407 Van Wonderen, J., 63 Vandelle, E., 575, 588, 589 Vanderkooi, J. M., 137 Vanfleteren, J. R., 164, 257, 419, 460 vanHoorn, W. P., 423 Vanin, A. F., 104, 113 vanVeggel, F., 423 Varotsis, C., 168, 261 Vass, I., 597 Vasudevan, S. G., 500, 513 Vaughn, L., 305 Vazquez-Torres, A., 42, 212, 521, 522, 523, 526, 527, 528, 532, 533, 535 Veen, J. P., 151 Veng-Va, K. V., 562 Venkataraman, S., 587 Venturoli, G., 333
643
Author Index
Verkhovsky, M. I., 93 Verlet, L., 464 Verma, C. S., 10, 12, 423 Veselov, A. V., 304 Viappiani, C., 318, 320, 329, 330, 333, 334, 335, 337, 338, 339, 340 Vicente, J. B., 3, 4, 8, 11, 14, 15, 21, 22, 23, 24, 25, 26, 29, 30, 31, 32, 33, 35, 38, 39, 40, 41, 42, 47, 48, 49, 50, 51, 52, 53, 54, 55, 58, 60, 124, 193, 212, 237, 502, 509, 515 Viera, L., 607 Villa, E., 442, 448, 460, 461, 467 Vincent, K. A., 92 Vinck, E., 302, 303, 304 Vinogradov, S. N., 164, 257, 288, 419, 460 Virtala, M., 522 Virtaneva, K., 114, 212 Virts, E., 174 Viswanathan, R., 4, 49, 51, 52 Vitalis, A., 484 Vogel, A., 29 Vogel, A. I., 201 Vogel, K. M., 263, 366 Vogel, U., 540 Vojtchovsky, J., 357, 418, 425, 480 Vollack, K. U., 114 Vollmar, A. M., 590 Volpe, J. A., 265 Voordouw, G., 38 Voss, I., 105 Vuletich, D. A., 446, 448 W Wade, J. T., 213 Wagner, M., 140 Wagner, U., 382, 405, 406, 407 Wainwright, L. M., 257, 265, 270, 274, 277, 278, 279, 502 Wajcman, H., 257 Wajnberg, E., 297, 298, 299, 300 Wakatsuki, S., 405 Walczak, T., 137 Walker, F. A., 301, 302, 303, 351, 371, 373 Walker, J. M., 29 Walsh, M. A., 38 Walsh, T. R., 10, 12 Walz, D. A., 257 Wang, C. Y., 137 Wang, D., 262 Wang, H. Y., 588, 597 Wang, J., 260, 261, 337, 354, 485 Wang, J. W., 588, 589 Wang, P. G., 553 Wang, T., 502, 509 Wang, W., 442, 448, 460, 461, 467 Wang, Y., 270, 277, 278, 441, 453, 568, 570 Wang, Y. H., 502
Wang, Y. N., 64 Wanner, B. L., 214, 505 Ward, W. W., 597 Warkentin, E., 4, 5, 6, 8, 16, 23, 24, 26, 40, 51, 52 Warne, A., 81, 83, 88 Warren, J. J., 357, 358, 418, 419 Warshel, A., 479 Waschipky, R., 358, 418 Wasserfallen, A., 4, 22, 24, 25, 26, 29, 30, 31, 32, 48, 49, 50, 54 Watanabe, A., 88 Watanabe, M., 422, 478 Watanabe, Y., 418 Watmough, N. J., 66, 79, 80, 81, 82, 85, 86, 89, 90, 91, 93, 94, 95, 96, 97 Watts, R. A., 419, 460, 596 Webb, J. L., 522 Weber, C. H., 38 Weber, P., 130 Weber, R. E., 288, 315, 460 Wei, Y. H., 298, 300 Weichsel, A., 351, 355, 359, 370, 371, 372, 373 Weidenbach, T., 324 Weik, M., 403 Weil, J. A., 289, 294 Weinberg, J. B., 528, 555 Weinstein, M., 174 Weissig, H., 423, 425, 448 Weitzberg, E., 543 Welford, R. W., 441 Wen, Z., 553 Wendehenne, D., 576, 583, 588, 589, 590, 591 Wertz, J. E., 289, 294 West, A. H., 480, 486 Westbrook, J., 423, 425, 448 Westerhoff, H. V., 82, 542 Wever, R., 141, 151 Weydert, J. A., 323 Weyhermuller, T., 242 White, G. P., 513 White, R. F., 195 White, R. H., 195 Whitehead, R., 542 Whittle, B. J., 565 Whyte, M. K., 555, 558 Wicke, M., 16 Wieghardt, K., 242 Wierzba, A. M., 323 Wigneshweraraj, S. R., 238 Wikstro¨m, M., 153 Wildt, J., 593 Wilkstro¨m, M., 303 Williams, R., 366 Williams, S. M., 201 Williamson, P., 546 Willis, A. C., 89 Willison, J. C., 30 Wills, R. B. H., 562
644
Author Index
Wilmot, C. M., 441 Wilson, D. F., 137 Wilson, D. J., 540 Wilson, K., 5, 7, 403 Wilson, K. E., 607, 617 Wilson, M. T., 140, 141, 142, 144, 145, 151, 153, 302 Wimpenny, J. W. T., 504 Wing, H. J., 201 Wink, D. A., 140, 563 Winn, M. D., 8 Wio´rkiewicz-Kuczera, J., 478 Wirz, J., 330, 332, 333 Wisedchaisri, G., 175 Wittenberg, B. A., 149, 165, 258, 267, 268, 269, 273, 274, 275, 276, 288, 302, 337, 418, 419, 430, 446, 448, 460, 488, 493, 494 Wittenberg, J. B., 148, 149, 165, 258, 267, 268, 269, 273, 274, 275, 276, 288, 302, 337, 418, 419, 430, 446, 448, 460, 488, 493, 494 Wittinghofer, A., 403 Woggon, W. D., 304 Wolynes, P. G., 356, 398, 409 Wonders, A. H., 130 Wong, H. Y., 564 Wood, E. R., 555 Wood, S. R., 502, 509 Wu, G., 104, 193, 199, 201, 236, 257, 277, 279, 501, 502, 513, 515 Wu, J. Y., 588, 589 Wu, X., 553 Wu, Y., 223 Wulff, M., 381, 382, 383, 384, 386, 387, 399, 402, 405, 406, 407, 408, 409, 411, 418, 419, 430, 440 Wykes, V., 145 Wyllie, G. R. A., 298 Wyman, J., 320, 323 X Xavier, A. V., 4, 5, 7, 8, 9, 10, 11, 12, 13, 22, 26, 31, 32, 33, 34, 38, 40, 48, 51 Xia, G.-X., 588, 597 Xian, M., 553 Xiang, Z., 484, 485 Xie, A., 361 Xie, Q. W., 524, 532 Xiong, H., 484 Xu, X., 223, 224 Xu, Y., 532, 533 Y Yamakura, F., 607 Yamamoto, A., 164, 171, 589 Yamamoto, H., 331 Yamanaka, S. A., 318
Yamasaki, H., 590 Yamauchi, K., 425, 427, 431, 448, 460, 461, 467 Yang, F., 357 Yang, J., 165, 168 Yang, M., 337 Yang, Y.-S., 7, 12, 29 Yap, G. P. A., 242 Yasui, T., 118, 125 Yavin, Z., 242 Ye, R. W., 502, 509 Yeh, S. R., 165, 255, 256, 265, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 460, 502 Yeremenko, S., 407 Yi, D., 606 Yin, D., 478 Yonetani, T., 138, 145, 279, 280, 295, 298, 299, 313, 323, 331 Yoon, J., 570 Yoshida, S., 300 Yoshimura, H., 168 Yoshimura, T., 104 Yoshioka, H., 589 Yoshioka, S., 168 Young, R. D., 349, 355, 356, 357, 358, 359, 361, 398 Yruela, I., 303 Yu, N.-T., 263, 265 Yue, K. T., 354, 356, 357, 358, 359 Yukl, E. T., 175 Yun, B. W., 568, 570 Z Zafiriou, O. C., 594 Zago, E., 596 Zaharik, M. L., 532 Za¨hringer, U., 585, 586, 588, 589, 590 Zamorano-Sanchez, D. S., 502 Zang, T. M., 10 Zaninotto, F., 594 Zeidler, D., 585, 586, 588, 589, 590 Zeier, J., 576, 577, 582, 588, 589, 590, 591, 596 Zeng, M., 562 Zgierski, M. Z., 263, 366 Zhang, 550 Zhang, L., 117, 118, 119, 120, 121, 125, 127, 128, 129, 130 Zhang, W., 168, 170 Zhang, X., 238, 589 Zhang, Y., 299 Zhao, X.-J., 264 Zhao, Y., 137, 236 Zheng, L., 195 Zheng, M., 41, 42, 223, 237, 502, 506, 509, 515 Zhi, Z., 299 Zhong, N-Q., 588
645
Author Index
Zhou, J., 7, 12, 29 Zhou, X., 532 Zhou, Y., 138, 295, 299 Zhou, Z., 80 Zhu, G. F., 502, 509 Zhulin, I. B., 187 Ziegler, T., 298
Zimmer, M., 164, 166, 171 Zink, J. I., 318 Zinser, S., 357, 418 Zou, J. Y., 8 Zuberbuehler, A. D., 330, 332, 333 Zufferey, R., 85 Zumft, W. G., 51, 81, 97, 114, 192, 543
This page intentionally left blank
Subject Index
A AMBER program implicit ligand sampling, 442 potential energy function, 479–480 Amperometry Neisseria meningitidis nitric oxide metabolism studies, see Neisseria meningitidis nitric oxide metabolism nitric oxide reductase assays fluvorubredoxin, Escherichia coli, 49–51 NorBC, Paracoccus denitrificans, 86–87 Aquaporin, implicit ligand sampling for AQP1, 453–454 B Bioinformatics, see Globin-coupled sensors Bovine serum albumin (BSA) for in vitro protein nitration, 606–607 C Carbon monoxide, globin ligand dynamics, see Flash photolysis; Fourier transform infrared spectroscopy; Locally enhanced sampling molecular dynamics; X-ray crystallography Ce-trHb, see Chladymonas eugametos truncated hemoglobin Cgb, see Hemoglobin CHARMM program conjugate peak refinement, 419, 421–422, 425, 428 implicit ligand sampling, 442–444 locally enhanced sampling molecular dynamics, 461, 463, 467 Chemiluminescence nitrate/nitrite assay in Neisseria meningitidis infection studies, 556–558 nitric oxide detection in plant–pathogen interactions, 582–583 nitric oxide synthase ozone chemiluminescence assay in plant tissues data collection, 564–566 principles, 563–564 ChIP-on-chip, NsrR binding site identification advantages, 213–214 cell culture, 217–218 control sample, 216–217 DNA microarray
labeling, hybridization, and processing, 219–220 visualization and analysis, 220, 222, 231 immunoprecipitation, 218–219 principles, 213 reference DNA sample, 216 statistical analysis, 222–230 strain construction, 214–216 Chlamydomonas eugametos truncated hemoglobin conjugate peak refinement of ligand-binding pathways, 431–432 locally enhanced sampling molecular dynamics of ligand-binding pathways, 471 Chromatin immunoprecipitation, see ChIPon-chip Confocal laser-scanning microscopy, S-nitrosoglutathione localization in plant tissues, 567–571 Conjugate peak refinement, ligand-binding pathways in globins disadvantages, 472 energy calculations and minimization, 425–426 findings Chlamydomonas eugametos truncated hemoglobin, 431–432 comparisons of globins, 432–433 Mycobacterium tuberculosis truncated hemoglobin N, 430 myoglobin, 429 Paramecium caudatum truncated hemoglobin, 431 fully refined pathway, 428 ligand-binding pathway classification, 429 ligand coordinate generation, 426–427 potential energy function, 420–421 principles of algorithm, 422–425 running, 428 structure preparation, 425 theoretical models, 419–420 CPR, see Conjugate peak refinement CYP55A, see Cytochrome P450nor Cytochrome c nitrite reductase, see Nrf, Escherichia coli Cytochrome P450nor complementary DNA isolation, 123–124 function, 118 gas analysis, 119 isoforms, 118 NAD analog titration, 127–129
647
648
Subject Index
Cytochrome P450nor (cont.) nitric oxide reductase assay, 121–122 purification of native proteins, 120–121 purification of recombinant proteins chromatography, 126–127 expression, 126 quantification, 130 screening of activity, 119 sequencing of protein, 122–123 site-directed mutagenesis, 125–126 stopped-flow rapid scan analysis, 129–130 subcellular fractionation of Trichosporon cutaneum, 124–125
D DAF-2, see 4,5-Diaminofluorescein diacetate 4,5-Diaminofluorescein diacetate nitric oxide detection in plant–pathogen interactions, 587–590 nitric oxide scavenging assay in plants, 597–600 Dinitrosyl iron complex bacterial regulation, 104 fumarate and nitrate reduction regulatory protein, 193 nitric oxide donor, 104 NorA from Ralstonia eutrophia nitric oxide quantification, 113–114 preparation of complexes in vitro, 111–112 in vivo, 113 DNA microarray, see ChIP-on-chip; Nitric oxide DNIC, see Dinitrosyl iron complex E Electron paramagnetic resonance ELDOR techniques, 293 electron spin echo modulation, 291, 298, 302–304 ENDOR techniques, 292–293, 296, 298, 300, 302–304 ferric globin studies, 301–304 flavodiiron proteins, 31–32 HYSCORE technique, 291, 298–299, 303–304 nitric oxide detection in plant–pathogen interactions, 585–587 nitric oxide-gated globin studies, 295–300 NorBC from Paracoccus denitrificans, 96–98 NorR ferrous-nitrosyl form, 240–242 paramagnetic states of heme proteins, 288 principles, 289–295 prospects for globin studies, 305 spin-labeling of heme proteins, 304–305 EPR, see Electron paramagnetic resonance
F FixL autophosphorylation assay, 175–176 deoxy protein preparation, 178–179 dissociation constant determinations carbon monoxide titration, 183 nitric oxide competition titration, 183–185 oxygen titration basis spectra and exploratory titration, 181–182 full titration, 182–183 oxygen-saturated buffer preparation, 181 FixL–FixJ system and function, 174–175 heme content analysis, 178 novel heme histidine-protein kinase discovery, 186–187 phosphatase contamination detection and kinase activity verification, 179–180 purification from recombinant Escherichia coli, 177–178 response regulator substrate purification, 180–181 turnover rate determination, 176–177, 185–186 Flash photolysis hemoglobin ligand migration mechanism characterization geminate rebinding enhancement, 336–337 hemoglobin encapsulation in silica gels, 335–336 instrumentation, 331–335 kinetic information extraction, 337–342 principles, 330–331 Flavodiiron proteins, see also specific entries classes, 22–23 cloning, 24–25 cofactor content analysis, 26–29 functional properties, 4, 39–42 purification of recombinant proteins, 25–26 quaternary structure analysis, 26 redox titration studies, 32–37 spectroscopic characterization absorbance spectroscopy, 29–31 electron paramagnetic resonance, 31–32 structure, see X-ray crystallography thermodynamic properties, 38–39 FLbR, see Ferric leghemoglobin reductase Flow cytometry, macrophage phenotype analysis for nitric oxide-dependent antimicrobial actions, 525 FlRd, see Fluvorubredoxin, Escherichia coli Fluvorubredoxin, Escherichia coli electron transfer chain, 48 nitric oxide detoxification, 48 nitric oxide reductase amperometric assay, 49–51 recombinant protein expression and purification, 24–26
649
Subject Index
spectroscopic characterization ionic strength and pH effects, 59–61 reduction by FlRd reductase, 54–58 reduction by NADH, 53–54 stopped-flow spectroscopy, 51, 53 FNR, see Fumarate and nitrate reduction regulatory protein Fourier transform infrared spectroscopy, heme protein ligand dynamics studies with cryospectroscopy environmental effects on ligand infrared bands, 353–355 materials carbon monoxide and nitric oxide handling, 349–350 instrumentation, 351–352 sample preparation, 350–351 nitric oxide-ligated heme proteins nitrophorin 4, 370–373 photolysis difference spectroscopy of myoglobin, 365–368 temperature derivative spectroscopy of myoglobin, 368–370 overview, 348–349 photolysis difference spectroscopy, 355–360 temperature derivative spectroscopy myoglobin-carbon monoxide example, 362–365 theory, 359–362 FprA, Methanothermobacter marburgensis crystallization, 6–7 nitric oxide reductase activity, 52 recombinant protein expression and purification, 26 FTIR, see Fourier transform infrared spectroscopy Fumarate and nitrate reduction regulatory protein absorbance spectroscopy, 198 dinitrosyl iron complex formation, 193 [4Fe-4S]2+ cluster nitric oxide reaction, 198–201 oxygen reaction, 193, 201–203 function, 192–193 iron content determination, 197–198 oxygen reactivity, 193 purification of 2Fe-FNR, 204 purification of 4Fe-FNR cleaning and concentration, 197 native protein, 194–195 overview, 194 reconstituted protein, 195–197 reaction product detection hydrogen peroxide, 206 iron, 205 sulfide, 204–205 superoxide, 205–206 structure, 192 sulfide content determination, 198
G Gas chromatography, cytochrome P450nor gas analysis, 119 Gas migration pathways, see Flash photolysis; Implicit ligand sampling; Multiple steering molecular dynamics; X-ray crystallography GCSs, see Globin-coupled sensors GFP, see Green fluorescent protein Globin-coupled sensors absorbance spectroscopy, 170 bioinformatic search C-terminal transmitter, 166 N-terminal sensor globin, 165–166 software, 164–165 function, 164 gel electrophoresis, 170 purification of recombinant protein chromatography, 168–169 cloning and expression, 167–168 minimum heme-binding domain, 170–171 species distribution, 164 Goat anti-rabbit IgG alkaline phosphatase conjugate, 608 Green fluorescent protein, hemoglobin fusion protein and subcellular localization in plants, 600–603 GSNO, see S-Nitrosoglutathione H Hemocyanin, oxygen-binding curve determination, 323–325 Hemoglobin conjugate peak refinement of ligand-binding pathways Chlamydomonas eugametos truncated hemoglobin, 431–432 comparisons of globins, 432–433 Mycobacterium tuberculosis truncated hemoglobin N, 430 Paramecium caudatum truncated hemoglobin, 431 Fourier transform infrared cryospectroscopy of ligand dynamics environmental effects on ligand infrared bands, 353–355 materials carbon monoxide and nitric oxide handling, 349–350 instrumentation, 351–352 sample preparation, 350–351 overview, 348–349 photolysis difference spectroscopy, 355–360 temperature derivative spectroscopy, 359 implicit ligand sampling for Paramecium caudatum truncated hemoglobin, 446, 448–449
650 Hemoglobin (cont.) ligand migration mechanism characterization with flash photolysis geminate rebinding enhancement, 336–337 hemoglobin encapsulation in silica gels, 335–336 instrumentation, 331–335 kinetic information extraction, 337–342 principles, 330–331 locally enhanced sampling molecular dynamics of truncated hemoglobin ligand-binding pathways Chlamydomonas eugametos truncated hemoglobin, 471 comparison of results, 471–472 Mycobacterium tuberculosis truncated hemoglobin N subunit A, 468 subunits A and B, 468–469 overview, 460 Paramecium caudatum truncated hemoglobin, 469–470 microbial hemoglobin features, 257–258 oxygen-binding curves, see Oxygen-binding curves, heme proteins plants genes, 596 nitric oxide detection in plant–pathogen interactions with conversion assay, 583–585 nitric oxide scavenging assay, 596–600 subcellular localization green fluorescent protein fusion protein, 600–603 overview, 597–598 quantum mechanical/molecular mechanical dynamics of Mycobacterium tuberculosis truncated hemoglobin N nitric oxide detoxification, 493–494 oxygen affinity, 492–493 resonance Raman spectroscopy of microbial hemoglobins applications, 260 Campylobacter jejuni hemoglobins Cgb, 279–280 trCtb, 277–279 distal axial ligand vibrational modes, 263–265 instrumentation, 260–261 Mycobacterium tuberculosis hemoglobins trHbN, 267–274 trHbO, 275–277 pitfalls, 266 porphyrin core vibrational modes, 261–262 principles, 258–260 prospects for study, 281–282 proximal iron-histidine stretching mode, 262
Subject Index
structural overview, 256–257 time-resolved X-ray crystallography analysis of Scapharca inaequivalvis hemoglobin ligand dynamics, 388–391 Heuristic methods, transition pathway analysis, 421–422 Histidine-protein kinase, see FixL Hydrogen peroxide, fumarate and nitrate reduction regulatory protein product detection, 206 HYSCORE, see Electron paramagnetic resonance I Implicit ligand sampling aquaporin calculations, 453–454 automated VMD script, 444–445 configurable parameters, 445–446 error bars, 451–452 gas migration pathway elucidation overview, 440–441 limitations, 455 Paramecium caudatum truncated hemoglobin calculations, 446, 448–449 potential of mean force equation, 441 map creation, 443–444 map interpretation, 449–450 projections, 453 probabilities and occupancies, 454 software, 442–443 water modeling, 450 Isopropyl-b-D-thiogalacto-pyramoside (IPTG), 610 L Laser photoacoustic detection, nitric oxide detection in plant–pathogen interactions, 579–582 Laue diffraction, see X-ray crystallography LESMD, see Locally enhanced sampling molecular dynamics Locally enhanced sampling molecular dynamics advantages and limitations, 472 ligand-binding pathway classification, 468 minimization and equalization, 467 molecular dynamics overview, 462–465 principles, 465–466 software, 461–462 structure preparation, 466–467 truncated hemoglobin ligand-binding pathway investigation Chlamydomonas eugametos truncated hemoglobin, 471 comparison of results, 471–472 Mycobacterium tuberculosis truncated hemoglobin N
651
Subject Index
subunit A, 468 subunits A and B, 468–469 overview, 460 Paramecium caudatum truncated hemoglobin, 469–470 LPAD, see Laser photoacoustic detection M Macrophage, nitric oxide-dependent antimicrobial actions human macrophage studies bacterial cultures, 530 isolation of macrophages, 529–530 killing assays, 530–532 materials, 529 overview, 528 murine macrophage studies bacterial cultures, 525–526 isolation of macrophages, 524–525 killing assays, 526–528 materials, 523–524 overview, 522–523 phenotypic analysis, 525 Neisseria meningitidis infection studies chemiluminescence assay of nitrate/nitrite, 556–558 culture and infection, 555–556 monocyte-derived macrophage infection, 556 Mass spectrometry, nitric oxide detection in plant–pathogen interactions, 577–579 MOIL program, locally enhanced sampling molecular dynamics, 461, 467 Molecular dynamics equilibration, 481 essential dynamics, 481–482 force field methods, 478 free energy profile calculations, 482–484 heme group parameters, 485 implicit ligand sampling, 442 locally enhanced sampling molecular dynamics multiple steering molecular dynamics Mycobacterium tuberculosis truncated hemoglobin N structural flexibility simulations, 488–490 potential energy function, 479–480 potential of mean force, 482–483 protein energy landscape exploring, see Protein energy landscape exploring quantum mechanical/molecular mechanical schemes, see Quantum mechanical/ molecular mechanical dynamics setup of system, 480–481 umbrella sampling, 483 Moorella thermoacetica flavodiiron protein crystallization, 5–6
nitric oxide reductase activity, 51–52 3-Morpholinosydnonimine-N-ethylcarbamide (SIN-1) nitrating agent, 607 MSMD, see Multiple steering molecular dynamics Mt-trHbN, see Mycobacterium tuberculosis truncated hemoglobin N Multiple steering molecular dynamics overview, 483–484 prospects, 494–495 ligand migration profiles of Mycobacterium tuberculosis truncated hemoglobin N, 490–492 Mycobacterium tuberculosis truncated hemoglobin N conjugate peak refinement of ligand-binding pathways, 430 locally enhanced sampling molecular dynamics of ligand-binding pathways subunits A and B, 468–469 molecular dynamics structural flexibility simulations, 488–490 multiple steering molecular dynamics of ligand migration profiles, 490–492 quantum mechanical/molecular mechanical dynamics nitric oxide detoxification, 493–494 oxygen affinity, 492–493 resonance Raman spectroscopy, 277–279 Myoglobin conjugate peak refinement of ligand-binding pathways, 429 Fourier transform infrared cryospectroscopy of ligand dynamics environmental effects on ligand infrared bands, 353–355 materials carbon monoxide and nitric oxide handling, 349–350 instrumentation, 351–352 sample preparation, 350–351 nitric oxide-ligated myoglobin photolysis difference spectroscopy, 365–368 temperature derivative spectroscopy, 368–370 overview, 348–349 photolysis difference spectroscopy, 355–360 temperature derivative spectroscopy myoglobin-carbon monoxide example, 362–365 theory, 359–362 oxygen-binding pathway exploration, 418–419 time-resolved X-ray crystallography of structural dynamics batch crystallization optimization, 401–402 crystal photolysis, 403–405 crystallization, 400 data collection, 405–407
652
Subject Index
Myoglobin (cont.) myoglobin states, 399–400 overview, 399 pump and probe picosecond Laue diffraction, 402–403 seed preparation, 400–401 synopsis of relaxations, 408–413 N NAMD program implicit ligand sampling, 442, 448 locally enhanced sampling molecular dynamics, 460–462, 467 NapA, function, 64 Neisseria meningitidis nitric oxide metabolism amperometric electrodes calibration, 550–551 nitric oxide electrode, 549–550 overview, 547 oxygen electrode, 547–549 simultaneous measurement of oxygen and nitric oxide, 551–553 culture conditions for analysis, 544–546 macrophage infection studies chemiluminescence assay of nitrate/nitrite, 556–558 culture and infection, 555–556 monocyte-derived macrophage infection, 556 nitric oxide donor induction of gene expression, 553–554 nitric oxide reductase activity, 542–543 assay, 554–555 nitrite reduction, 542–543 phylogeny and pathogenicity, 540–541 respiratory pathways, 541–542 safety aspects, 541 Nitric oxide bacteria response studies Campylobacter jejuni studies, 511–512 culture, batch versus continuous, 503–504 Escherichia coli studies chemostat design and use, 505–506 gene expression analysis with microarrays and reverse transcriptasepolymerase chain reaction, 509–510, 514–515 medium, 506–507 monitoring and sampling of cultures, 507–509 strain, 505 exposure and stress responses, 500–502 global stress analysis proteomics, 502–503 transcriptomics, 502–503 nitric oxide donors, 513
S-nitrosoglutathione studies, 512–513 sensitivity assays, 513–514 detoxification in Escherichia coli, 212 globin ligand dynamics, see Flash photolysis; Fourier transform infrared spectroscopy; Quantum mechanical/molecular mechanical dynamics immune response to bacteria macrophages humans, 529–532 mice, 522–528 mouse studies bacterial burden quantification, 534–535 competition assays, 535 inducible nitric oxide synthase inhibition, 536 inoculation and monitoring, 534 materials, 533 nitric oxide production assay, 536 overview, 532 Neisseria meningitidis metabolism, see Neisseria meningitidis nitric oxide metabolism plant functions, 562 plant–pathogen interactions and nitric oxide detection chemiluminescence assay, 582–583 4,5-diaminofluorescein diacetate, 587–590 electron paramagnetic resonance, 585–587 hemoglobin conversion assay, 583–585 laser photoacoustic detection, 579–582 mass spectrometry, 577–579 overview, 577–577, 590–591 terminal respiratory chain oxidase inhibition combined oxygen and nitric oxide polarography, 153–155 comparison of analytical techniques, 137–138 nitric oxide donors, 138–139 nitric oxide kinetic analysis dissociation constant, 141–144 dynamic versus steady-state IC50 measurements, 146 IC50, 139–141 off-rates and on-rates, 146–148 pitfalls, 144–146, 153 optical detection of intermediates nitric oxide-bound intermediates, 153 overview, 151 spectro-electrode system, 152–153 overview, 136–138 oxygen kinetic analysis, 149–150 as signaling molecule in plants, 606 Nitric oxide reductase, see Cytochrome P450nor; Fluvorubredoxin, Escherichia coli; Neisseria meningitidis nitric oxide metabolism; NorA, Ralstonia eutrophia; NorBC, Paracoccus denitrificans; Nrf, Escherichia coli
653
Subject Index
Nitric oxide synthase inducible nitric oxide synthase inhibition, 536 ozone chemiluminescence assay in plant tissues data collection, 564–566 principles, 563–564 Nitrophorin 4, Fourier transform infrared cryospectroscopy of nitric oxide ligand dynamics, 370–373 S-Nitrosoglutathione bacteria nitric oxide response studies, 512–513 confocal laser-scanning microscopy localization in plant tissues, 567–571 S-Nitrosoglutathione reductase assays in plant tissues native gel electrophoresis and staining, 567 spectrophotometric assay, 566–567 plant functions, 562 NO, see Nitric oxide NorA, Ralstonia eutrophia apoenzyme preparation, 108 dinitrosyl iron complex preparation in vitro nitric oxide-saturated buffer technique, 111–112 nitrite technique, 112 in vivo, 113 disulfide bridges, 107 gene expression regulation, 105–106 iron analysis, 108 nitric oxide quantification from dinitrosyl iron complex, 113–114 orthologs, 114 purification from recombinant Escherichia coli expression, 106 lysis, 106 nickel affinity chromatography, 107 spectroscopic characterization diferric enzyme preparation, 110 diferrous enzyme preparation, 110 oxyNorA preparation, 110–111 redox form interconversion, 109–111 NorBC, Paracoccus denitrificans assays amperometric assay, 86–87 pseudoazurin as electron donor, 88–90 function, 80–81 prospects for study, 98–99 purification native protein, 82–85 recombinant protein from Escherichia coli, 85–86 spectroscopic characterization electron paramagnetic resonance, 96–98 enzyme preparation fully oxidized, 91 fully reduced, 91–93 partially reduced, 94–95 structure, 81–82
NorR electron paramagnetic resonance of ferrousnitrosyl form, 240–242 function, 236 iron center reconstitution, 242–243 nitric oxide interactions binding, 236–237 dissociation constant determination, 243–248 nitric oxide electrode standardization, 246–247 targets of regulation, 236 transcriptional activation assays in vitro, 238–240 in vivo, 237–238 NOS, see Nitric oxide synthase Nrf, Escherichia coli crystallization, 73–75 cytochrome c nitrite reductase assay, 69–73 function, 64 nitric oxide consumption assay, 66 prospects for study, 74, 76 purification of recombinant protein chromatography, 68–69 expression, 66–68 lysis, 68 structure, 64–65 NsrR binding site identification on Escherichia coli chromosome using ChIP-on-chip advantages, 213–214 cell culture, 217–218 control sample, 216–217 DNA microarray labeling, hybridization, and processing, 219–220 visualization and analysis, 220, 222, 231 immunoprecipitation, 218–219 principles, 213 reference DNA sample, 216 statistical analysis, 222–230 strain construction, 214–216 function, 212 targets of regulation, 212 O OBCs, see Oxygen-binding curves, heme proteins Oxygen-binding curves, heme proteins apparatus for determination, 313–316 hemocyanin, 323–325 hemoglobin K1 determination in absence of allosteric effectors, 318 solution determination, 316–318 T state hemoglobin crystals, 320–323
654
Subject Index
Oxygen-binding curves, heme proteins (cont.) T state hemoglobin gels allosteric effector absence, 320 allosteric effector presence, 319–320 overview, 318 static versus dynamic techniques, 312–313 Oxygen-sensing histidine-protein kinase, see FixL Oxygraph 2K, combined oxygen and nitric oxide polarography, 155 P P450nor, see Cytochrome P450nor Paramecium caudatum truncated hemoglobin conjugate peak refinement of ligand-binding pathways, 431 implicit ligand sampling, 446, 448–449 locally enhanced sampling molecular dynamics of ligand-binding pathways, 469–470 Pc-trHb, see Paramecium caudatum truncated hemoglobin PELE, see Protein energy landscape exploring Penalty function, transition pathway analysis, 421–422 Peroxynitrite-mediated nitration of SODs, molecular mechanism of, 614 Polarography, combined oxygen and nitric oxide polarography, 153–155 Polyclonal anti-3-nitrotyrosine antibody and protein tyrosine nitration, 608 Potential energy function, conjugate peak refinement, 420–421 Potential of mean force, see Implicit ligand sampling; Molecular dynamics Protein energy landscape exploring ligand migration profiles of Mycobacterium tuberculosis truncated hemoglobin N, 491 local perturbation, 485 minimization, 485 overview, 484 side chain sampling, 485 Protein nitration BSA and recombinant Vu_FeSOD as targets for, 607 SIN-1 by, 608 and symbiotic legume nodules, 607 Proteomics, global nitric oxide stress response studies in bacteria, 502–503 Pseudoazurin, electron donor in nitric oxide reductase assays, 88–90 Q QM-MM, see Quantum mechanical/molecular mechanical dynamics Quant 1 software in GelDoc 2000 (Bio-Rad), 613 Quantum mechanical/molecular mechanical dynamics
Mycobacterium tuberculosis truncated hemoglobin N studies nitric oxide detoxification, 493–494 oxygen affinity, 492–493 nitric oxide–globin dynamics studies binding energy calculations, 487 QM subsystem selection, 486 QM-MM optimizations, 486–487 reaction pathway search, 487–488 overview, 479 prospects, 494–495 SIESTA code, 485–486 R Recombinant Vu_FeSOD catalytic activity and tyrosine nitration, 610–612 Resonance Raman spectroscopy, microbial hemoglobins applications, 260 Campylobacter jejuni hemoglobins Cgb, 279–280 trCtb, 277–279 distal axial ligand vibrational modes, 263–265 instrumentation, 260–261 Mycobacterium tuberculosis hemoglobins trHbN, 267–274 trHbO, 275–277 pitfalls, 266 porphyrin core vibrational modes, 261–262 principles, 258–260 prospects for study, 281–282 proximal iron-histidine stretching mode, 262 Rubredoxin:oxygen oxidoreductase, Desulfovibrio gigas function, 4 nitric oxide reductase activity, 52 X-ray crystallography crystallization, 4–5 data collection and structure determination, 7–8 flavin mononucleotide moiety features, 14–16 flavodoxin domain, 11 metallo-b-lactamase-like domain, 9–10 nonheme diiron center features, 11–14 S SES, see Spectro-electrode system SIN-1 and desferrioxamine dose-dependent effect on bovine serum albumin (BSA) tyrosine nitration, 609 dependent Vu_FeSOD tyrosine nitration and enzymatic activity, 612–616 effect on Vu_FeSOD activity assayed by spectrophotometric method, 615
655
Subject Index
Singular value decomposition, time-resolved X-ray crystallography analysis of heme ligand dynamics, 387–388 Spectro-electrode system nitric oxide-bound intermediate detection in terminal respiratory chain oxidases, 153 overview, 151–153 Stopped-flow spectroscopy fluvorubredoxin from Escherichia coli, 51, 53 rapid scan analysis of cytochrome P450nor, 129–130 Sulfide, fumarate and nitrate reduction regulatory protein product detection, 204–205 Superoxide dismutases (SODs), role in animal systems, 607 Superoxide, fumarate and nitrate reduction regulatory protein product detection, 205–206 SVD, see Singular value decomposition T Temperature derivative spectroscopy, see Fourier transform infrared spectroscopy N,N,N0 ,N0 ,-Tetramethyl-ethylenediamine (TEMED), 612 Thermotoga maritima flavodiiron protein crystallization, 6 nitric oxide reductase activity, 52 Transcriptomics, global nitric oxide stress response studies in bacteria, 502–503, 509–510, 514–515 Truncated hemoglobins, see Hemoglobin Tyrosine nitrated bovine serum albumin (BSA), immunodetection of, 608–609 Tyrosine nitration on Vu_FeSOD and activity assay, immunochemical detection of, 611 in Vu_FeSOD by anti-3-nitrotyrosine antibodies, 612 V VMD program contact analysis in ligand-binding pathways, 429
implicit ligand sampling, 443–446, 448 Vu_FeSOD, FeSOD isoenzyme, 607 X Xanthine-xanthine oxidase system, 613 X-ray crystallography flavodiiron proteins crystallization FprA from Methanothermobacter marburgensis, 6–7 Moorella thermoacetica protein, 5–6 Nrf from Escherichia coli, 73–75 rubredoxin:oxygen oxidoreductase from Desulfovibrio gigas, 4–5 Thermotoga maritima protein, 6 data collection, 7–8 flavin mononucleotide moiety features, 14–16 flavodoxin domain, 11 metallo-b-lactamase-like domain, 9–10 nonheme diiron center features, 11–14 structure determination, 8 time-resolved studies heme ligand dynamics data collection, 381–385 data processing and analysis, 385–388 hemoglobin of Scapharca inaequivalvis, 388–391 principles, 379–381 prospects, 391–393 myoglobin structural dynamics batch crystallization optimization, 401–402 crystallization, 400 crystal photolysis, 403–405 data collection, 405–407 myoglobin states, 399–400 overview, 399 pump and probe picosecond Laue diffraction, 402–403 seed preparation, 400–401 synopsis of relaxations, 408–413