Matrix Metalloproteinase Protocols, 2nd Edition (Methods in Molecular Biology, 622)

ME T H O D S

IN

MO L E C U L A R BI O L O G Y

Series Editor John M. Walker School of Life Sciences University of Hertfordshire Hatfield, Hertfordshire, AL10 9AB, UK

For other titles published in this series, go to www.springer.com/series/7651

TM

Matrix Metalloproteinase Protocols Second Edition

Edited by

Ian M. Clark University of East Anglia, Norwich, UK Consulting Editors

David A. Young Newcastle University, Newcastle upon Tyne, UK

Andrew D. Rowan Newcastle University, Newcastle upon Tyne, UK

Editor Ian M. Clark University of East Anglia School of Biological Sciences Norwich United Kingdom NR4 7TJ [email protected] Consulting Editors David A. Young Newcastle University Institute of Cellular Medicine Musculoskeletal Research Group Framlington Place Newcastle upon Tyne 4th Floor, Catherine Cookson Bldg. United Kingdom NE2 4HH [email protected]

Andrew D. Rowan Newcastle University Institute of Cellular Medicine Musculoskeletal Research Group Framlington Place Newcastle upon Tyne 4th Floor, Catherine Cookson Bldg. United Kingdom NE2 4HH [email protected]

ISSN 1064-3745 e-ISSN 1940-6029 ISBN 978-1-60327-298-8 e-ISBN 978-1-60327-299-5 DOI 10.1007/978-1-60327-299-5 Springer New York Dordrecht Heidelberg London Library of Congress Control Number: 2009943297 © Springer Science+Business Media, LLC 2001, 2010 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Humana Press, c/o Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. Cover illustration: is microcarrier beads invasion assay using HeLa cells, courtesy of Ralf Palmisano and Yoshifumi Itoh. Cover design: Nancy Fallatt Printed on acid-free paper Humana Press is part of Springer Science+Business Media (www.springer.com)

Preface Research in the matrix metalloproteinase field began with the demonstration by Gross and Lapi`ere, in 1962, that resorbing tadpole tail expressed an enzyme which could degrade collagen gels. These humble beginnings have led us to more than 20 distinct vertebrate MMPs, along with a variety of homologues from diverse organisms such as sea urchins, plants, insects, nematode worm, and bacteria. Related enzymes, the ADAMs and ADAMTSs, as well as the inhibitors, TIMPs, create a complex picture. Section I of this book provides the reader with a brief overview of the MMP arena from how these enzymes fit into the larger degradome to what happens when you modulate their expression and function in the mouse. Hopefully this complements the methodology that comes later. Section II presents the reader with a diverse set of methods for the expression and purification of MMPs and TIMPs, bringing together the long and often hard-earned experience of a number of researchers. Section III enables the reader to detect MMPs and TIMPs at both the protein and the mRNA level and Section IV provides the ability to assay MMP and TIMP activities in a wide variety of circumstances. For folks who are new to MMP research, we hope that this book will enable you to “hit the ground running”, acquiring both some background and many useful laboratory techniques in one easy volume. For the seasoned MMPer, we hope that it will add some new methods to your old favourites or help anyone new in your lab to get going quickly. The first edition of this book was published in 2001 but was assembled in the 2 years before. This second edition contains several updates of the original chapters, as well as new technologies and methodologies that were not available in the dim and the distant past of a decade ago! We would like to thank all of the authors for the time and the effort they put in. The demands of writing a book chapter are not trivial, especially against the background of heavy research and teaching commitments, and we have again been overwhelmed with the number of eminent and senior people in MMP research who have contributed. Ian M. Clark with David A. Young Andrew D. Rowan

v

Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

v

Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

ix

TIMPS: AN OVERVIEW . . . . . . . . . . . . . . . . . .

1

1.

Metalloproteases and the Degradome . . . . . . . . . . . . . . . . . . . . . . . Alejandro P. Ugalde, Gonzalo R. Ord´on ˜ ez, Pedro M. Quir´os, Xose S. Puente, and Carlos L´opez-Ot´ın

3

2.

Mouse Models of MMP and TIMP Function . . . . . . . . . . . . . . . . . . . Sean E. Gill, Sean Y. Kassim, Timothy P. Birkland, and William C. Parks

31

TIMPS . . . . . .

53

3.

Expression of Recombinant MMP-28 in Mammalian Cells . . . . . . . . . . . . . Ursula R. Rodgers and Ian M. Clark

55

4.

Expression of Recombinant Matrix Metalloproteinases in Escherichia coli . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . L. Jack Windsor and Darin L. Steele

SECTION I

SECTION II

MMPS

AND

EXPRESSION

AND

PURIFICATION

OF

MMPS

AND

67

5.

Expression of Recombinant ADAMTS in Insect Cells . . . . . . . . . . . . . . . Gavin C. Jones, Mireille N. Vankemmelbeke, and David J. Buttle

83

6.

Expression and Purification of Membrane-Type MMPs . . . . . . . . . . . . . . Jing Nie and Duanqing Pei

99

7.

Refolding of TIMP-2 from Escherichia coli Inclusion Bodies . . . . . . . . . . . . 111 Richard A. Williamson

8.

Purification of MMPs and TIMPs . . . . . . . . . . . . . . . . . . . . . . . . . 123 Ken-ichi Shimokawa and Hideaki Nagase

S ECTION III 9.

DETECTION

OF

MMPS

AND

TIMPS

. . . . . . . . . . . . . . . . . 157

Real-Time PCR Expression Profiling of MMPs and TIMPs . . . . . . . . . . . . 159 Caroline J. Pennington and Dylan R. Edwards

10. Analysis of the Degradome with the CLIP-CHIPTM Microarray . . . . . . . . . . 175 Reinhild Kappelhoff, Ulrich auf dem Keller, and Christopher M. Overall 11. In Situ Hybridization for Metalloproteinases and Their Inhibitors . . . . . . . . . 195 Tiina L. Hurskainen and Suneel S. Apte 12. Immunohistochemistry of MMPs and TIMPs . . . . . . . . . . . . . . . . . . . 211 Yasunori Okada

vii

viii

Contents

13. Single Nucleotide Polymorphism Genotyping in MMP Genes: The 5 Nuclease Assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221 Ross Laxton and Shu Ye S ECTION IV

ASSAY

OF

MMP AND TIMP A CTIVITIES . . . . . . . . . . . . . . . 231

14. Methods for Studying Activation of Matrix Metalloproteinases Vera Kn¨ auper and Gillian Murphy

. . . . . . . . . . 233

15. Assay of Matrix Metalloproteinases Against Matrix Substrates . . . . . . . . . . . 245 Tim E. Cawston, Rachel L. Lakey, and Andrew D. Rowan 16. Zymography and Reverse Zymography for Detecting MMPs and TIMPs . . . . . 257 Susan P. Hawkes, Hongxia Li, and Gary T. Taniguchi With contributions from Marc A. Lafleur 17. In Situ Zymography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271 Sarah J. George and Jason L. Johnson 18. Near-Infrared Optical Proteolytic Beacons for In Vivo Imaging of Matrix Metalloproteinase Activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279 J. Oliver McIntyre, Randy L. Scherer, and Lynn M. Matrisian 19. Neoepitope Antibodies Against MMP-Cleaved and Aggrecanase-Cleaved Aggrecan . . . . . . . . . . . . . . . . . . . . . . . . 305 Amanda J. Fosang, Karena Last, Heather Stanton, Suzanne B. Golub, Christopher B. Little, Lorena Brown, and David C. Jackson 20. In Vitro Model of Cartilage Degradation Wang Hui and Tim E. Cawston

. . . . . . . . . . . . . . . . . . . . . 341

21. Immunoassays for Collagenase-Mediated Cleavage of Type I and II Collagens . . 349 R. Clark Billinghurst, Mirela Ionescu, and A. Robin Poole 22. Collagen Degradation Assays . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367 Anthony P. Hollander 23. Analysis of MMP-Dependent Cell Migration and Invasion . . . . . . . . . . . . . 379 Ralf Palmisano and Yoshifumi Itoh 24. Using Fluorogenic Peptide Substrates to Assay Matrix Metalloproteinases . . . . . 393 Gregg B. Fields 25. Kinetic Analysis of the Inhibition of Matrix Metalloproteinases: Lessons from the Study of Tissue Inhibitors of Metalloproteinases . . . . . . . . . . . . . 435 Frances Willenbrock, Daniel A. Thomas, and Augustin Amour 26. Identification of Cellular MMP Substrates Using Quantitative Proteomics: Isotope-Coded Affinity Tags (ICAT) and Isobaric Tags for Relative and Absolute Quantification (iTRAQ) . . . . . . . . . . . . . . . . . . . . . . . . . 451 Georgina S. Butler, Richard A. Dean, Charlotte J. Morrison, and Christopher M. Overall 27. Mechanism-Based Profiling of MMPs . . . . . . . . . . . . . . . . . . . . . . . 471 Jed F. Fisher and Shahriar Mobashery Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 489

Contributors AUGUSTIN AMOUR • GlaxoSmithKline, Stevenage, Hertfordshire, UK SUNEEL S. APTE • Department of Biomedical Engineering, Lerner Research Institute, Cleveland Clinic Foundation, Cleveland, OH, USA R. CLARK BILLINGHURST • School of Applied Sciences, St. Lawrence College, Kingston, ON, Canada TIMOTHY P. BIRKLAND • Center for Lung Biology, University of Washington School of Medicine, Seattle, WA, USA LORENA BROWN • Department of Microbiology, University of Melbourne, Melbourne, VIC, Australia GEORGINA S. BUTLER • Centre for Blood Research, University of British Columbia, Vancouver, BC, Canada DAVID J. BUTTLE • School of Medicine and Biomedical Sciences, University of Sheffield, Sheffield, UK TIM E. CAWSTON • Musculoskeletal Research Group, Institute of Cellular Medicine, Newcastle University, Newcastle upon Tyne, UK IAN M. CLARK • School of Biological Sciences, University of East Anglia, Norwich, UK RICHARD A. DEAN • Centre for Blood Research, University of British Columbia, Vancouver, BC, Canada DYLAN R. EDWARDS • School of Biological Sciences, University of East Anglia, Norwich, UK GREGG B. FIELDS • Department of Chemistry and Biochemistry, Florida Atlantic University, Boca Raton, FL, USA JED F. FISHER • Department of Chemistry and Biochemistry, Walther Cancer Research Center, University of Notre Dame, Notre Dame, IN, USA AMANDA J. FOSANG • Department of Paediatrics and Murdoch Children’s Research Institute, Royal Children’s Hospital, University of Melbourne, Melbourne, VIC, Australia SARAH J. GEORGE • Bristol Heart Institute, Bristol Royal Infirmary, University of Bristol, Bristol, UK SEAN E. GILL • Center for Lung Biology, University of Washington School of Medicine, Seattle, WA, USA SUZANNE B. GOLUB • Department of Paediatrics and Murdoch Children’s Research Institute, Royal Children’s Hospital, University of Melbourne, Melbourne, Australia SUSAN P. HAWKES • Departments of Biopharmaceutical Sciences and Pharmaceutical Chemistry, University of California, San Francisco, CA, USA

ix

x

Contributors

ANTHONY P. HOLLANDER • Department of Cellular and Molecular Medicine, University of Bristol, Bristol, UK WANG HUI • Musculoskeletal Research Group, Institute of Cellular Medicine, Newcastle University, Newcastle upon Tyne, UK TIINA L. HURSKAINEN • Clinical Research Center, Oulu University Hospital, Oulu, Finland MIRELA IONESCU • Departments of Immunology and Laboratory Sciences, Charles River, Montreal, QC, Canada YOSHIFUMI I TOH • Imperial College London, Kennedy Institute of Rheumatology, London, UK DAVID C. JACKSON • Department of Microbiology, University of Melbourne, Melbourne, VIC, Australia JASON L. JOHNSON • Bristol Heart Institute, Bristol Royal Infirmary, University of Bristol, Bristol, UK GAVIN C. JONES • School of Biological Sciences, University of East Anglia, Norwich, UK REINHILD KAPPELHOFF • Departments of Oral Biological and Medical Sciences, Centre for Blood Research, University of British Columbia, Vancouver, BC, Canada SEAN Y. KASSIM • Center for Lung Biology, University of Washington School of Medicine, Seattle, WA, USA ULRICH AUF DEM KELLER • Departments of Oral Biological and Medical Sciences, Centre for Blood Research, University of British Columbia, Vancouver, BC, Canada VERA KN A¨ UPER • Dental School, Cardiff University, Cardiff, UK MARC A. LAFLEUR • Amgen, Thousand Oaks, CA, USA RACHEL L. LAKEY • Musculoskeletal Research Group, Institute of Cellular Medicine, Newcastle University, Newcastle upon Tyne, UK KARENA LAST • Department of Paediatrics and Murdoch Children’s Research Institute, Royal Children’s Hospital, University of Melbourne, Melbourne, VIC, Australia ROSS LAXTON • Barts and the London School of Medicine and Dentistry, William Harvey Research Institute, London, UK HONGXIA LI • Departments of Biopharmaceutical Sciences and Pharmaceutical Chemistry, University of California, San Francisco, CA, USA CHRISTOPHER B. LITTLE • Raymond Purves Bone and Joint Research Laboratories, Royal North Shore Hospital, University of Sydney, Sydney, Australia ´ CARLOS LOPEZ -OT ´I N • Departamento de Bioqu´ımica y Biolog´ıa Molecular, Universidad de Oviedo, Oviedo, Spain L YNN M. MATRISIAN • Department of Cancer Biology, Vanderbilt University, Nashville, TN, USA

Contributors

xi

J. OLIVER MCINTYRE • Department of Cancer Biology, Vanderbilt University, Nashville, TN, USA SHAHRIAR MOBASHERY • Department of Chemistry and Biochemistry, Walther Cancer Research Center, University of Notre Dame, Notre Dame, IN, USA CHARLOTTE J. M ORRISON • Centre for Blood Research, University of British Columbia, Vancouver, BC, Canada GILLIAN MURPHY • Department of Oncology, University of Cambridge, Cambridge, UK HIDEAKI NAGASE • Imperial College London, Kennedy Institute of Rheumatology, London, UK JING NIE • Hope Heart Program, Benaroya Research Institute at Virginia Mason, Seattle, WA, USA YASUNORI OKADA • Department of Pathology, School of Medicine, Keio University, Tokyo, Japan ˜ EZ • Departamento de Bioqu´ımica y Biolog´ıa Molecular, GONZALO R. O RD O´ N Universidad de Oviedo, Oviedo, Spain CHRISTOPHER M. OVERALL • Departments of Oral Biological and Medical Sciences, Biochemistry and Molecular Biology, Centre for Blood Research, University of British Columbia, Vancouver, BC, Canada RALF PALMISANO • Imperial College London, Kennedy Institute of Rheumatology, London, UK WILLIAM C. PARKS • Center for Lung Biology, University of Washington School of Medicine, Seattle, WA, USA DUANQING PEI • Chinese Academy of Sciences, Guangzhou Institute of Biomedicine and Health, Guangzhou, China CAROLINE J. PENNINGTON • School of Biological Sciences, University of East Anglia, Norwich, UK A. ROBIN POOLE • Department of Surgery, McGill University, Montreal, QC, Canada XOSE S. PUENTE • Departamento de Bioqu´ımica y Biolog´ıa Molecular, Universidad de Oviedo, Oviedo, Spain ´ • Departamento de Bioqu´ımica y Biolog´ıa Molecular, Universidad PEDRO M. QUIR OS de Oviedo, Oviedo, Spain URSULA R. RODGERS • School of Biological Sciences, University of East Anglia, Norwich, UK ANDREW D. ROWAN • Musculoskeletal Research Group, Institute of Cellular Medicine, Newcastle University, Newcastle upon Tyne, UK RANDY L. SCHERER • Department of Cancer Biology, Vanderbilt University, Nashville, TN, USA

xii

Contributors

KEN-ICHI SHIMOKAWA • Department of Physical Pharmacy, Meiji Pharmaceutical University, Tokyo, Japan HEATHER STANTON • Department of Paediatrics and Murdoch Children’s Research Institute, University of Melbourne, Royal Children’s Hospital, Melbourne, Australia DARIN L. STEELE • Department of Biochemistry and Molecular Genetics, University of Alabama, Birmingham, AL, USA GARY T. TANIGUCHI • Departments of Biopharmaceutical Sciences and Pharmaceutical Chemistry, University of California, San Francisco, CA, USA DANIEL A. THOMAS • GlaxoSmithKline, Harlow, Essex, UK ALEJANDRO P. UGALDE • Departamento de Bioqu´ımica y Biolog´ıa Molecular, Universidad de Oviedo, Oviedo, Spain MIREILLE N. VANKEMMELBEKE • School of Molecular Medical Sciences, University of Nottingham, Nottingham, UK FRANCES WILLENBROCK • Orpington, Kent, UK RICHARD A. WILLIAMSON • Department of Biosciences, University of Kent, Canterbury, UK L. JACK WINDSOR • Department of Oral Biology, Indiana University School of Dentistry, Indianapolis, IN, USA SHU YE • Barts and the London School of Medicine and Dentistry, William Harvey Research Institute, London, UK

Section I MMPs and TIMPs: An Overview

Chapter 1 Metalloproteases and the Degradome ˜ ´ Alejandro P. Ugalde, Gonzalo R. Ordo´ nez, Pedro M. Quiros, ´ Xose S. Puente, and Carlos Lopez-Ot´ ın Abstract Metalloproteases comprise a heterogeneous group of proteolytic enzymes whose main characteristic is the utilization of a metal ion to polarize a water molecule and perform hydrolytic reactions. These enzymes represent the most densely populated catalytic class of proteases in many organisms and play essential roles in multiple biological processes. In this chapter, we will first present a general description of the complexity of metalloproteases in the context of the degradome, which is defined as the complete set of protease genes encoded by the genome of a certain organism. We will also discuss the functional relevance of these enzymes in a large variety of biological and pathological conditions. Finally, we will analyze in more detail three families of metalloproteases: ADAMs (a disintegrin and metalloproteinase), ADAMTSs (ADAMs with thrombospondin domains), and MMPs (matrix metalloproteinases) which have a growing relevance in a number of human pathologies including cancer, arthritis, neurodegenerative disorders, and cardiovascular diseases. Key words: Enzyme, proteolysis, metzincin, cancer, metastasis, arthritis.

1. Introduction Since their initial discovery, concepts on proteases have evolved from the consideration of these enzymes as proteins merely implicated in the non-specific degradation of dietary proteins to their wide recognition as members of complex systems which modulate multiple biological processes through cleavage of specific substrates (1). By performing these proteolytic processing reactions, proteases regulate a wide range of cellular processes such as DNA replication, cell-cycle progression, cell proliferation, differentiation and migration, apoptosis, senescence, and autophagy. In metazoans, proteolytic systems are also involved I.M. Clark (ed.), D.A. Young, A.D. Rowan (Consulting editors), Matrix Metalloproteinase Protocols, Methods in Molecular Biology 622, DOI 10.1007/978-1-60327-299-5 1, © Springer Science+Business Media, LLC 2001, 2010

3

4

Ugalde et al.

in the maintenance of tissue homeostasis and in the regulation of different physiological processes such as fertilization and fecundation, embryonic development, immune response, wound healing, tissue remodeling, and angiogenesis. The importance of proteolysis is underscored by the fact that proteases can be found in all kingdoms of life, from archaea and eubacteria to plants and animals, as well as in numerous viruses Additionally, alterations in the structure or regulation of proteolytic systems can have dramatic consequences on the whole organism and underlie many human pathologies such as neurodegenerative diseases, progeroid syndromes, or cancer (2). To understand the increasing complexity of proteolytic enzymes from a global perspective, the term degradome has been recently coined to define the entire complement of protease genes encoded by the genome of one organism (1). The recent availability of genome sequences from several organisms has allowed the annotation and comparison of their respective degradomes. Thus, a highly curated protease database (http://www.uniovi.es/degradome), which does not incorporate protease pseudogenes or retrovirus-derived protease-like sequences, currently annotates 569 human proteases and homologues classified in 68 families. The chimpanzee degradome is very similar to the human degradome despite notable variations in immune defense proteases like caspase-12 which is absent or non-functional in humans but is present in chimpanzee and all other mammals (3). Interestingly, mouse and rat degradomes are much larger (644 and 629 protease genes, respectively) than the human degradome despite their genomes are smaller. These differences mainly derive from the specific expansion in rodents or the specific inactivation in humans of members of protease families – such as placental cathepsins and kallikreins – involved in reproductive and immunological functions (2). The recent analysis of the degradome of other mammals such as the duck-billed platypus has revealed not only the presence of more than 500 protease genes but also a lack of all genes encoding gastric pepsins, which are highly conserved digestive proteases present in the rest of mammals (4). Birds, amphibians, and fish also contain large numbers of protease genes (more than 300 in every analyzed genome), although the protease annotation work in these species has not been as detailed as in mammals. Remarkably, analysis of the degradome of invertebrates such as Drosophila melanogaster – a model organism with a gene content considerably lower than vertebrates – has shown the presence of more than 600 protease genes (5). This relatively large number of insect proteases mainly derives from the genomic expansion of a family of serine proteases genes implicated in embryonic development and immune defense. Comparative genomic analysis has also facilitated the annotation of the protease repertoire of model plants such as Arabidopsis thaliana which contains more than 700 pro-

Metalloproteases and the Degradome

5

tease genes (6). Finally, there is a growing interest in the careful analysis of the degradomes of bacteria, viruses, fungi, and parasites to define novel therapeutic targets. In this regard, the annotation in the MEROPS database (http://merops.sanger.ac.uk) of more than 100 protease genes in the genome of bacteria such as Yersinia pestis and in the malaria parasite Plasmodium falciparum, which cause important human diseases, is noteworthy. In summary, these recent genomic studies have provided a first global view of the complexity of proteolytic systems in all living organisms. Although all these proteases catalyze the same biochemical reaction – the hydrolysis of a peptide bond – different mechanistic solutions have been developed during evolution to efficiently perform this type of reaction. Attending to their catalytic mechanism, proteolytic enzymes can be grouped into six different classes: aspartic, glutamic, serine, cysteine, threonine, and metalloproteases. The general mechanism for all of them consists in the nucleophilic attack to the carbonyl carbon of the peptide bond. In aspartic, glutamic, and metalloproteases, a polarized water molecule located in the active center acts directly as a nucleophile. By contrast, in the other three classes the reactive element consists of a hydroxyl (serine and threonine) or sulfydryl (cysteine) group from the corresponding side chain of these residues at the catalytic core. In this chapter we will focus on the analysis of metalloproteases, which represent the most densely populated catalytic class of proteases in many organisms including mammals (Fig. 1.1). We will first present a general description of the complexity of this large group of enzymes. We will also discuss different functions of metalloproteases in both normal and pathological conditions. Finally, we will analyze in more detail the ADAM (a disintegrin and metalloproteinase domain), ADAMTS (ADAMs with thrombospondin domains), and MMP (matrix metalloproteinases) families of enzymes which are of special relevance in the context of this book.

Fig. 1.1. Distribution of proteases in the human degradome.

6

Ugalde et al.

2. Metalloprotease Classification

Metalloproteases comprise a heterogeneous group of proteases that can be classified at various levels, according to their catalytic mechanism, their substrates and products, or their structural homology (Fig. 1.2). Similar to other protease classes, a first classification scheme can be established attending to the cleavage position in the substrate. Following this criterion, metalloproteases can be classified as endopeptidases or exopeptidases. Endopeptidases comprise those enzymes with ability to cleave inner peptide bonds of proteins, whereas exopeptidases cleave a peptide bond located not further than three amino acids from the N-terminal (aminopeptidases) or C-terminal end (carboxypeptidases) of the protein. The main characteristic of metalloproteases is the utilization of a metal ion – usually Zn – to polarize a water molecule and perform the hydrolysis reaction. The problem of coordinating a metal ion and accommodate the substrate in a polypeptide scaffold has found different solutions throughout evolution. As

Fig. 1.2. Classification of metalloproteases. The figure shows an overview of the different catalytic mechanisms employed by metalloproteases.

Metalloproteases and the Degradome

7

a result, metalloproteases can be classified in different groups, termed clans or families, based on structural and sequence homology. The last MEROPS database release (8.00) contains 32,275 sequences of metalloproteases, which can be grouped into at least 14 clans and 54 different families. Figure 1.2 shows a simple classification of metalloproteases based on the nature and sequence of the active site as well as other structural determinants. The metal ion present in the active center of metalloproteases is usually coordinated with the polypeptide backbone through the binding to two histidines and a third residue which can be another histidine or an acidic residue (Asp, Glu). In addition to these three metalbinding residues, a catalytic residue is absolutely necessary for the enzymatic activity of the protease. This residue acts as a general base, first accepting a proton from water and then transferring it to the scissile bond. The positions of these key four residues are usually conserved among members of a clan and can be deduced from the alignment of primary sequences or, more accurately, through the study of 3D structures. Although the three zincbinding residues and the general base are crucial for the activity of metalloproteases, other residues can influence enzyme efficiency by participating in different aspects of catalysis such as carboxyanion stability, substrate accommodation, and protein fold.

3. The HExxH Motif and Its Variants

The most conserved zinc-binding motif in metalloproteases is the HExxH sequence. In this short motif, the two histidines coordinate the zinc ion and the glutamate acts as a general base in the catalytic reaction. This structural motif characterizes the MA clan, the largest clan of metalloproteases, which contains two subclans and 40 families of peptidases, including MMPs, ADAMs, and ADAMTSs which will be further discussed in this chapter. In this group of enzymes, the third coordination position of the zinc ion is occupied by another residue situated at a distal position from the core sequence (Fig. 1.2). This residue can be a glutamate, an aspartate, or another histidine. Additionally, the HExxH motif can also be found in the MM clan of membraneembedded metalloproteases, although they have been placed in a separate clan because of their transmembrane-spanning segments that carry the zinc-binding residues (7). These peculiar metalloproteases are able to cleave the transmembrane-spanning helical regions of membrane proteins within the lipid bilayer, thus participating in a signal transduction system termed regulated intramembrane proteolysis. A representative member of this metalloprotease clan is the site-2 protease (MBTPS2) which regulates

8

Ugalde et al.

the level of cholesterol in cell membranes through the processing of the sterol regulatory element-binding protein (8). Therefore, the HExxH sequence motif constitutes the most successful solution to the zinc coordination problem and has been employed by more than 40 different families of metalloproteases. A simple variation of the HExxH motif is present in the ME clan of peptidases, whose members display an HxxEH motif, an inversion of the zinc-binding sequence presented above. Additionally, an aspartate residue acts as the third ligand for the metal ion. Due to the inversion of the structural motif, this group of enzymes has been named inverzincins. There are four functional inverzincins encoded in the human genome: the insulin-degrading enzyme insulysin, the convertase nardilysin, and the mitochondrial presequence peptidases eupitrilysin and MPP (mitochondrial processing peptidase beta-subunit) (9). In addition, there are three clans (MC, MP, and MK) of nonHExxH metalloproteases distributed in humans. The MC clan comprises a unique family of carboxypeptidases (M14), with a consensus motif HxxE, where glutamate and histidine act as zinc-binding residues, together with a third His located 103– 143 residues toward the C terminus. In humans, there are 19 members of this family involved in the digestion of dietary proteins in the intestine, protein degradation by mast cells, or proteolytic processing of different bioactive peptides (10, 11). The MP clan includes the family M67 containing enzymes that chelate the zinc ion through two histidines and an aspartic acid situated within the motif HSHP(x)9 D. There are seven human MP members which participate in the detachment of the ubiquitin tag from proteins marked for proteasome degradation (12). The MK clan has a unique family of endopeptidases (M22) which cleaves O-sialoglycosylated proteins and contains two members in the human genome: O-sialoglycosylated endopeptidase (OSGEP) and OSGEP-2 with an HxExH motif. Finally, there are two additional clans of metalloproteases (MD and MO) without the HExxH motif. Both of them are broadly distributed in bacteria and display major roles in bacterial wall synthesis and homeostasis. MD clan contains metalloproteases with an alternative consensus sequence HxxxxxxD and a distal histidine (13). MO clan proteases also employ two histidines and an aspartic acid as metal ligands, but they occur within the motifs HxxxD and HxH (13).

4. Co-catalytic Metalloproteases All of the above-described metalloproteases contain a unique zinc ion in their active center and are known as zincins, but there is a second group of metalloproteases termed dizincins or cocatalytic zincins which are characterized by an active center that

Metalloproteases and the Degradome

9

coordinates two zinc ions. As in zincins, each metal ion is coordinated with three residues of the protein, but in dizincins, at least one of them acts as a bridge between the two zinc ions. Cocatalytic proteases can be grouped into two main classes. The first class contains clan MF aminopeptidases involved in processes such as MHC-I antigen presentation or lens crystallin protein turnover (14). The two zinc ions are coordinated through six residues, and an arginine in the active center interacts with a bicarbonate ion that functions as a general base. The second group of dizincins includes aminopeptidases from clans MH and MN whose zincbinding structure is composed of five residues and a glutamate or a histidine residue acting as a general base. Peptidases from clan MN are D-amino acid-specific aminopeptidases distributed only in bacteria and archaea and implicated in peptidoglycan synthesis (15). The MH clan is composed of four families (M18, M20, M28, and M42) of amino- and carboxypeptidases present in animals, with the exception of family M42, which is restricted to archaea and bacteria. Human members of this family include aspartyl aminopeptidase, the only member of family M18 in mammals, carnosine dipeptidases I and II from family M20, and several glutamate carboxypeptidases from M28 family. Finally, a third group of co-catalytic metalloproteases that require other bivalent ions like Mn(II), Co(II), or Fe(II) for their activity is formed by clans MQ and MG. Proteases from the MQ clan are only distributed in plants, bacteria, and archaea, but several members of the MG clan can be found in mammals (family M24). Human members include three methionyl aminopeptidases, which release the N-terminal methionine of nascent proteins and X-Pro dipeptidases involved in the turnover of collagen degradation products (16).

5. The MA Clan of Metalloproteases As mentioned above, most metalloproteases belong to the MA clan. Peptidases from this clan contain an active center fold related to that of thermolysin and are classically divided into three groups: gluzincins, aspzincins, and metzincins (17–19). The first two groups were named according to the third zinc-binding amino acid, glutamate and aspartate, respectively, whereas the metzincin group was defined attending to a conserved methionine residue in a structure called Met-turn (20). Metzincins usually employ a histidine as the third zinc-binding residue, although an aspartate can be found at this position in some members. Metzincins and aspzincins are grouped together in the same MA(M) subclan, while gluzincins are classified into subclan MA(E).

10

Ugalde et al.

5.1. Gluzincins

The gluzincin superfamily contains 18 families of both exopeptidases and endopeptidases and is widely distributed throughout all kingdoms of life. Among them, only six families (M1, -2, -3, -13, -41, and -48) are present in animals. Members of the M1 family are aminopeptidases which participate in the processing of different bioactive peptides and proteins, thus regulating important aspects of the organism physiology such as blood pressure, embryonic implantation, placental homeostasis, inflammation, or angiogenesis (21). On the other hand, members of the M2 family of metalloproteases show carboxypeptidase activity with only two representatives in humans: angiotensin-converting enzymes-1 and -2 (ACE and ACE2). ACE has gained enormous interest due to its role in the regulation of blood pressure by processing angiotensin I to angiotensin II (22). The importance of this enzyme in the control of blood pressure is highlighted by the fact that inhibitors targeting ACE are widely used for the treatment of hypertension and congestive heart failure. An additional characteristic feature of ACE is the presence, in the same polypeptide chain, of two different gluzincin domains. This structure is extremely unusual among metalloproteases, because only carboxypeptidase D contains two different metalloprotease domains within the same polypeptide chain. The four other families of gluzincins present in humans are endopeptidases. Members of the M3 and M13 families are oligopeptidases involved in the processing of bioactive peptides. The M3 family only contains three members: thimet oligopeptidase, which is implicated in the regulation of the MHCI system of antigen presentation; neurolysin, which degrades intracellular bioactive peptides as neurotensin; and the mitochondrial intermediate peptidase that participates in the maturation of several nuclear-encoded mitochondrial proteins (23, 24). The M13 family is formed by seven members: neprilysins -1 and -2, endothelin-converting enzymes -1 and -2, Kell blood group antigen, and DINE and PHEX peptidases, all of them membrane peptidases involved in the metabolism of a great variety of regulatory peptides of nervous, cardiovascular, or immune systems (25). The M41 family contains nuclear-encoded mitochondrial metalloproteases that play important roles in mitochondrial homeostasis (26). By contrast, the M48 family is composed of two transmembrane endopeptidases: ZMPSTE24/FACE-1 which participates in the processing of farnesylated proteins (27) and the mitochondrial protease OMA1. This family of gluzincins has gained considerable interest after the finding that mutations in the genes encoding FACE-1 or its substrate (lamin A) are responsible for different premature ageing syndromes including Hutchinson–Gilford progeria or mandibuloacral dysplasia (28–30). In addition to these families of gluzincins widely distributed in animals, there are several metalloproteases belonging to

Metalloproteases and the Degradome

11

families absent in metazoans but of notable relevance because they behave as virulence factors for a large variety of pathogenic bacteria and fungi. A good example is the M4 thermolysin family, which is composed of secreted proteases produced by different bacterial strains. Some members of this family such as Pseudomonas aeruginosa pseudolysin, Legionella sp. Msp peptidase or Vibrio cholerae vibriolysin are important virulence factors (31). The M27 family of tetanus and botulism toxins is another example of a group of microbial metalloproteases that participate in the pathogenicity mechanism, as well as the anthrax lethal factor from Bacillus anthracis (family M34) (19). Bacterial collagenases from Vibrio and Clostridium degrade extracellular components of host interstitial tissue causing gas gangrene and septicemia and are frequently used in molecular biotechnology (32). Fungi metalloproteases belong to the M36 family and include fungalysin, an elastinolytic enzyme from Aspergillus fumigates, and several keratinases from fungi dermatophytes (33, 34). 5.2. Aspzincins

Aspzincins comprise a unique family of zincins (M35) constituted by deuterolysins from molds and bacteria. Most of them are neutral proteinases that participate in organism nutrition using as substrates, basic nuclear proteins like histones and protamines (17). Deuterolysins from A. oryzae and A. sojae have special interest since they are secreted by these molds during soybean fermentation. Sequence alignment of members of this family reveals an HExxH zinc-binding motif while the third zinc ligand is an aspartate residue situated 10 residues C-terminal to the second histidine (17). Although some metzincins from families M6 and M7 also contain an aspartate as third ligand, aspzincins are considered a different group due to the lack of a Met-turn and the absence of significant sequence homology with metzincins. This distinction is less clear in the M64 family of IgA peptidases from Clostridium (35). These enzymes also have an aspartate as the third zinc-binding residue and lack the consensus methionine present in metzincins, but contain a highly conserved glycine residue within the sequence HExxHxxxG which resembles the consensus sequence of metzincins. Therefore, it will be necessary to solve the 3D structure of these metalloproteases in order to determine the importance of each position and define with which group each is more akin.

5.3. Metzincins

The superfamily of metzincins is composed of metalloproteases with the HExxHxxxGxxH/D consensus zinc-binding sequence in their active center, where the first two histidines coordinate the zinc ion, and the glutamate acts as a general base. The conserved glycine serves as a hinge for a turn that brings the third metal ligand (His or Asp) near the zinc ion. An N-terminal segment of variable length connects the third ligand residue with the

12

Ugalde et al.

invariant methionine which forms part of the 1,4-␤-turn, generating a hydrophobic pillow for the zinc ion (18, 20, 36). The catalytic domain of metzincins consists of two lobes leaving a central groove which contains the zinc ion at its bottom, resulting in an optimum configuration for accommodating elongated substrates. All metzincins are metalloendopeptidases, with the possible exception of archaeometzincins, a family of aminopeptidases from the MA(M) clan widely distributed in archaea and vertebrates but whose structures remain undetermined (37). Metzincin families with reported 3D structures include astacins, serralysins, adamalysins/reprolysins, matrixins/MMPs, snapalysins, and leishmanolysins. The astacin subfamily (M12A) receives its name from the crayfish digestive enzyme astacin, the first discovered member of this group of enzymes. Astacins are broadly distributed among the animal kingdom as well as in eubacteria, fungi, or protozoans and perform diverse functions different from digestion. These additional astacins include a group of metalloproteases present in a variety of animal species and termed hatching enzymes, due to their function during the breakdown of the egg envelope (38, 39). Another important group of astacins includes BMP1/tolloid-like peptidases from Xenopus, Drosophila, and mammals. These enzymes are extracellular proteases able to target various extracellular matrix (ECM) components, growth factors, and enzymes (40). Accordingly, BMPs are implicated in processes such as bone formation, collagen assembly, embryonic development, differentiation, and TGF-␤ signaling (39, 41). Meprins constitute another group of astacins, widely distributed among chordates and displaying a membrane-bound localization in the brush border of the small intestine and kidney, as well as in leukocytes. Most of them are glycoproteins involved in the processing of bioactive peptides, like bradykinin, substance P, bombesin, and gastrin, as well as different basement membrane components such as collagen IV, nidogen-1, and fibronectin (38, 39). Interestingly, most astacins contain additional C-terminal modules such as EGF (epidermal growth factor)-like, CUB (complement subcomponents C1r/C1 s, embryonic sea urchin protein Uegf, BMP-1), and MATH (meprin and TRAF homology) (39). Serralysins (M10B) and snapalysins (M7) are metzincins restricted to bacterial species. The former receives its name from the Serratia marcescens serralysin, which, like the rest of members of this family, acts as a virulence factor. Other examples are mirabilysin from Proteus mirabilis, aeruginolysin from Pseudomonas aeruginosa, and peptidases from Erwinia chrysanthemi or Yersinia pestis. These microorganisms are the causal agents of several human pathologies such as meningitis, endocarditis, otitis, pneumonia, conjunctivitis, and urinary infections that mainly affect immunocompromised hosts. Serralysins play

Metalloproteases and the Degradome

13

important roles in host invasion and virulence by degrading host defense mediators, such as interleukins, immunoglobulins, coagulation factors, and antimicrobial peptides (42–45). Snapalysins are secreted peptidases from various species of Streptomyces characterized by having an aspartic acid as third zinc ligand, a feature shared with peptidases from the M6 family. This family is formed by immune inhibitors from different Bacillus species that infect insects, including the immune inhibitor A from Bacillus thuringensis, which cleaves antibacterial humoral proteins produced by the immune system of insects (20, 46). The leishmanolysin family of metzincins (M8) is a group of peptidases identified as the major surface glycoproteins (MSPs) of different protozoan parasites, mainly from genders Leishmania and Trypanosoma. These organisms are responsible for several anthroponotic diseases, such as the sleeping sickness, Chagas’ disease, and various leishmaniosis (47). Leishmanolysins are expressed in the amastigote and promastigote stage of the parasite and contribute to different aspects of protozoan survival and host invasion. MSPs are extracellular membrane-anchored peptidases that bind complement component C3, or cleave CD4 molecules, immunoglobulins, and other components of the host immune system, favoring resistance to complement-mediated lysis, attachment, entrance, and survival into the macrophages (48, 49). Recently, several mammalian and insect homologues have been discovered, such as invadolysin, which plays important roles in cell division and development in Drosophila (50). Mammalian homologues of invadolysin have been named leishmanolysin-2 and -3 and their functions remain unknown. The reprolysins/adamalysins form the second group of M12 peptidases (M12B), which comprises snake venom metalloproteases (SVMPs), ADAMs, and ADAMTSs. The name reprolysin reflects the fact that the first family members were identified in reptiles and in mammalian reproductive tissues. Currently, there are more than 140 reprolysin entries in MEROPS database, broadly distributed throughout the animal kingdom and expressed in a wide variety of tissues. The minimum structure of an adamalysin is found in some snake venom proteases which only contain the peptidase unit. However, most members of this group of metalloproteases contain additional domains which contribute to various aspects of their specificity, localization, or activation. SVMPs are toxic proteins present in snake venom which cause some of the symptoms associated with snake venom intoxication, such as hemorrhage, necrosis, edema, and inflammation. These metalloproteases exert their functions through degradation of proteins of the endothelial basement membrane, such as fibronectin, collagen, nidogen, or laminin, as well as by processing of plasma coagulation factors like fibrinogen and von Willebrand factor (20, 51). In addition to SVMPS, the reprolysins

14

Ugalde et al.

also include ADAMs and ADAMTSs, two groups of metalloproteases of growing interest in multiple physiological and pathological conditions. Accordingly, both subfamilies will be discussed in detail in the following sections. 5.3.1. ADAMs (A Disintegrin and Metalloprotease)

The ADAM subfamily of reprolysins is composed of a variety of widely distributed enzymes that share a structural domain organization, constituted by a peptidase unit followed by a disintegrin domain, an EGF-like module, a transmembrane region, and a cytoplasmic tail (Fig. 1.3). The peptidase unit also contains an N-terminal signal peptide and a pro-domain which keeps ADAMs inactive through interaction with the active center via a conserved cysteine residue, in a mechanism known as cysteine switch. Interestingly, the pro-domain may also have a secondary function as a chaperone, favoring the proper fold of these metalloproteases (52). The signal peptide directs ADAMs to the endoplasmic reticulum where the pro-domain is cleaved off by furin-like pro-protein convertases. Some ADAMs are located at the Golgi but most show a cellular surface localization or are

Fig. 1.3. Human and mouse ADAMs and ADAMTSs. The figure shows the domain organization of both groups of proteolytic enzymes; (np) indicates non-protease homologues; genes, and pseudogenes (ps) absent in one species are shadowed in gray and green, respectively.

Metalloproteases and the Degradome

15

secreted. Nevertheless, several family members, such as ADAM12, undergo alternative splicing events giving rise to a large cellsurface protein and a short secreted isoform (53). In humans there are 21 ADAMs out of the 38 members identified in mammals, whereas a total of 37 family members have been identified in the mouse genome (Fig. 1.3). It is remarkable that several mouse ADAMs, such as ADAM-1, -3, -4, and -25, are pseudogenes in the human genome (2). This situation also occurs in rat, whose genome lacks three mouse ADAM orthologs and presents three pseudogenes that are functional in mice (54). Moreover, ADAM20 is a human-specific gene which is absent in rodent genomes. The main subgroup of ADAMs expanded in the mouse genome is that of testases, whose name derives from their predominant expression in testis and their functions in reproduction. In mice there are nine testase genes, all of them being absent in humans with the exception of testase-2 which in humans is a pseudogene. Although ADAMs are expressed in various somatic tissues, most of them show expression in reproductive tissues. Other ADAMs are detected predominantly in male reproductive organs, such as ADAM-2, -3, -5, -18, -24, -25, -26, or -32, whereas some family members, like ADAM-11, -22, or -23, are mainly expressed in the central nervous system (55). The main function of several ADAMs is the ectodomain shedding from the cell surface of a variety of transmembrane proteins, such as cytokines, growth factors and their receptors, and adhesion proteins. A well-known example is ADAM-17, also known as TNF-␣ converting enzyme (TACE). TNF-␣ is a key regulator of the immune response that exists in two alternative forms, a small secreted isoform and a large membraneanchored protein. ADAM-17, as well as ADAM-9 and -10, can release the ectodomain of the membrane-bound TNF-␣, leading to its activation (56). However, the fact that only Adam17deficient mice display a significantly inhibited TNF-␣ shedding has led to the proposal that ADAM-17 is the major TNF-␣ sheddase (57). The EGF family of growth factors is also susceptible to ADAM-mediated shedding. EGF-like growth factors are synthesized as membrane-bound precursors that, upon proteolytic cleavage, release a soluble mitogenic form that binds to its receptor at the cell surface. ADAM-17 can shed various EGFlike growth factors, such as amphiregulin, epiregulin, heparinbinding EGF, neuregulin, and TGF-␣. Similarly, other ADAMs such as ADAM-9, -12, and -19 can cleave some of these substrates. This functional overlapping is supported by the fact that Adam8, -9, and -15 knockout mice develop normally, whereas Adam12-deficient mice display minor development defects (58). Furthermore, mice simultaneously lacking functional ADAM-9, -12, and -15 are viable and fertile, with no apparent pathology (59). ADAMs also regulate growth factor and cytokine signaling

16

Ugalde et al.

pathways at the receptor level. For example, shedding of IL-1, IL6, and TNF-␣ receptors is impaired in Adam17-null cells. Growth factor receptors such as HER4JM-a isoform, hepatocyte growth factor receptor Met, and nerve growth factor receptors are also susceptible to shedding by ADAM-17 (56). Another function associated with ADAMs is the ␣-secretase activity against amyloid precursor protein (APP), which generates a soluble ␣-APP with neurotrophic effects, instead of the harmful effects derived from ␤-amyloid production. It has been reported that ADAM-9, ADAM-10, and ADAM-17 are able to generate ␣-APP in physiological conditions. However, only Adam17-null cells are unable to produce soluble ␣-APP, indicating that this ADAM may be the main ␣-secretase. Other important substrates of ADAMs are insulin-like growth factor binding proteins (IGFBPs) and Notch receptor. ADAM12 cleaves IGFBP-3 and -5 and may be responsible for the increased levels of active IGF in plasma during pregnancy (60). Finally, Adam10-deficient mice die at day 9.5 of embryogenesis due to multiple defects in nervous and cardiovascular systems development associated with a downregulation of Notch signaling (61). As a consequence of their ability to target this variety of substrates, ADAMs influence cell behavior at several different levels, including regulation of the balance between survival and apoptotic signals, promotion of cell migration, or induction of angiogenesis (62). For example, ADAM-15 disintegrin domain constitutes a potent anti-angiogenic factor with capacity of disrupting the association between integrin ␣v ␤3 and vitronectin (63). In this regard, it is necessary to emphasize that the proteolytic activity is not an absolute requirement for certain functions of ADAMs. This is reflected by the fact that many ADAMs have changes in their consensus catalytic sequence that impair their catalytic properties but do not affect other essential functions of these enzymes. This is the case of ADAM-2, a non-protease ADAM whose deficiency in mice causes infertility, resembling the phenotype of the ADAM-3 knockout. Recent studies have proved that ADAM-2 is necessary for the correct stability and location of ADAM-3 (64). Another interesting example is ADAM-23, which promotes cell adhesion via interaction of its disintegrin domain with ␣v ␤3 integrin (65). One remarkable feature of ADAM activity is that, in most cases, ADAMs require an activation signal, like that provided by phorbol esters, to exert their effects. This phenomenon is related to the ability of ADAMs to mediate the transactivation of EGF receptors. Angiotensin II, endothelin-1, lysophosphatidic acid, or carbachol are examples of molecules that can activate EGF receptors without direct interaction with them (66). These ligands bind to G protein-coupled receptors (GPCRs) and initiate a signal transduction pathway that leads to ADAM activation. Activation

Metalloproteases and the Degradome

17

of ADAM is thought to be due to protein–protein interactions and phosphorylation at the cytoplasmic tail of these enzymes. The cytoplasmic tails of several ADAMs contain binding sites for SH3 domain proteins as well as phosphorylation sites for protein kinases (56). The activation mechanisms are not well defined, but various second messengers such as c-Src, PKC, Ca2+ , and ROS have been found to be implicated in the signaling of shedding processes (66). The two major pathways associated with activation of ADAM-mediated shedding are the Erk and p38 mitogenactivated protein kinases (MAPK). ADAM activity is also subjected to regulation by inhibitor molecules. Tissue inhibitors of metalloproteases (TIMPs) are endogenous molecules classically linked to MMP inhibition, but various works support a crossreaction with ADAMs. For example, ADAM-17 is inhibited by TIMP-3, whereas ADAM-10 is susceptible to TIMP-1 and -3 blockade. By contrast other ADAMs such as ADAM-8 and -9 are not controlled by TIMPs (67). Finally, it is remarkable that these structural and functional studies on ADAMs have an additional dimension after the description of ADAMTSs, a related group of reprolysins that share some features with ADAMs and also exhibit distinctive properties which will be discussed in the next section. 5.3.2. ADAMTSs (A Disintegrin and Metalloprotease with Thrombospondin Motifs)

This group of secreted reprolysins share some ADAM domains but lack the transmembrane and cytoplasmic tail and contain additional domains, including a central thrombospondin (TS) type-1 motif and a series of C-terminal TS repeats, ranging from zero in ADAMTS-4 to 14 in the case of ADAMTS-20 (68, 69). ADAMTS-7 and -12 have a mucin domain between the third and fourth of their seven C-terminal TS repeats. ADAMTS-20 and ADAMTS-9 have a GON domain, and ADAMTS-13 has two CUB domains. A PLAC (protease and lacunin) domain is also present in several ADAMTSs (Fig. 1.3) (70). One feature that distinguishes ADAMTSs from ADAMs is their ability to bind to the ECM through their central and C-terminal TS domains or through their spacer regions (71). Nascent ADAMTSs undergo N-terminal processing first in the endoplasmic reticulum by a signal peptidase and second in the trans-Golgi network by a proprotein convertase. Latency of pro-ADAMTSs may not be due to a cysteine-switch mechanism like in ADAM or MMPs, because only six ADAMTSs contain a conserved cysteine residue in their pro-domains. In addition, ADAMTSs can be processed at their C-terminal end, mainly within the spacer region or in the mucin domain in the case of ADAMTS-12. This processing affects ECM binding and substrate specificity of these enzymes. Thus, fulllength ADAMTS-4 undergoes C-terminal processing generating two short-length isoforms that are not bound to the ECM. The full-length isoform cleaves aggrecan at a different position than

18

Ugalde et al.

the short isoforms and is inhibited by addition of fibronectin, indicating that the spacer region is responsible for ECM binding and substrate specificity (71). Another interesting example is ADAMTS-1, which shows a potent anti-angiogenic activity which is absent in shorter isoforms lacking the C-terminal TS repeats and part of the spacer region (71). This C-terminal processing of ADAMTSs is carried out by members of the MMP family, such as MMP-2, -8, -15, and -17 (70). ADAMTSs are expressed in a wide range of adult tissues, being more restricted in fetal tissues. ADAMTS-1, -4, -5, -8, -9, and -15 constitute a subgroup of enzymes with ability to degrade aggrecan, the major cartilage proteoglycan. Accordingly, these ADAMTSs are known as aggrecanases, although they can also degrade brevican, mainly expressed in the CNS, and versican, a blood vessel-specific proteoglycan. Aggrecanases contribute to the development of osteoarthritis and other pathologies not related to cartilage metabolism, such as growth retardation, impaired adipogenesis, and decreased fertility (72). ADAMTS-2, -3, and -14 are procollagen N-proteinases implicated in the releasing of the N-terminal propeptides of procollagen. ADAMTS-2 acts on procollagen I, II, and III, but ADAMTS-3 and ADAMTS14 are restricted to procollagen II and I, respectively. ADAMTS-2 mutations cause Ehlers–Danlos syndrome in humans, characterized by skin fragility and joint laxity (73). Likewise, ADAMTS-10 mutations are responsible of human Weill–Marchesani syndrome, another pathology associated with altered ECM homeostasis and characterized by short stature, brachydactyly, joint stiffness, eye anomalies, and heart defects (74). ADAMTS-9 and -20 contain a GON domain and show structural similarity with the Caenorhabditis elegans gon-1, an ADAMTS associated with gonadal development in this organism. Adamts20-mutant mice are viable but display defects in melanocyte development (75). To date, it is unknown whether ADAMTS-9 controls gonadal formation in mammals, although a high-level expression in various embryonic tissues has been reported (76). The only ADAMTS-containing CUB domain is ADAMTS-13, also known as von Willebrand factor cleaving protease (vWFCP), whose deficiency is responsible for thrombotic thrombocytopenic purpura, characterized by the formation of microvascular thrombi and the development of anemia, renal failure, and neurological dysfunction (77). ADAMTS1 and -8 stand out by their anti-angiogenic properties, as both can inhibit angiogenesis through VEGF binding and sequestration by their TS motifs (78). Another recently reported antiangiogenic ADAMTS is ADAMTS-12, which is able to abolish tubule formation when exogenously added to epithelial cells (79). Interestingly, ADAMTS-12 also has anti-tumorigenic properties, decreasing tumor growth by preventing neovascularization (79).

Metalloproteases and the Degradome

19

Likewise, ADAMTS-1, -8, -9 -15, and -18 have also been proposed to play antitumor roles (80). In addition to ADAMs and ADAMTSs, there is a third group of metzincins called MMPs that have been subjected to an exhaustive research due to their relevance in multiple normal and pathological conditions, especially in cancer and arthritis. The next section will review the nature, functions, and implications in disease of this group of metalloproteases. 5.3.3. MMPs (Matrix Metalloproteinases)

MMPs or matrixins are a group of secreted or membraneanchored enzymes with at least 25 members distributed among vertebrates (Fig. 1.4) (81). Similar to reprolysins, MMPs have a HExxHxxGxxH peptidase motif in their proteolytic domain, an N-terminal signal peptide, and a pro-domain. This minimum domain structure is characteristic of matrilysins (MMP7 and MMP-26). However, the archetypal domain organization of MMP contains an additional C-terminal hinge region followed by several hemopexin domains. This organization

Fig. 1.4. The MMP family of matrix metalloproteases. Schematic domain organization and human and mouse family members.

20

Ugalde et al.

is present in stromelysin-1 and -2 (MMP-3 and -10), collagenases (MMP-1, -8, and -13), metalloelastase (MMP-12), enamelysin (MMP-20), and MMP-27. A similar domain structure is present in stromelysin-3 (MMP-11), epilysin (MMP28), and MMP-21, but these enzymes have a furin-like target sequence inserted in their pro-domains, and constitute the group of secreted convertase-activatable MMPs. Other convertaseactivatable MMPs are membrane-anchored by a glycosylphosphatidylinositol (GPI), or by type I or type II transmembrane segments, constituting a group called membrane-type MMPs (MT-MMPs). MMP-23A and 23 B also belong to this group, but contain an N-terminal type II transmembrane region and a cysteine array and immunoglobulin domains in their Cterminal end. Finally, MMP-2 and MMP-9 are secreted MMPs with gelatinolytic activity and characterized by the presence of fibronectin type II modules inserted into the peptidase domain (Fig. 1.4) (81). MMP activity regulation has been widely studied with the finding of three main levels of endogenous control: gene transcription, proenzyme activation, and activity inhibition. Transcription regulation of MMPs is a complex mechanism in which no single factor is responsible for induction of MMP expression. However, cytokines such as TNF-␣ or IL-1 and growth factors such as TGF-␤ have been widely associated with the regulation of MMP expression. Likewise, several signal transduction pathways are implicated in the control of MMP transcription, like the p38 MAPK, which displays a dual role, enhancing or repressing MMP expression depending on the cell type (82). From the large variety of stimuli influencing MMP expression, several transcription factors have been found to mediate transcriptional changes in these metalloprotease genes. For example, most promoters of MMP genes contain functional binding sites for AP1, a transcription factor composed of dimers of members of the Fos and Jun family of oncoproteins. Other transcription factors associated with regulation of MMP expression are nuclear factor ␬B (NF-␬B), T-cell factor 4 (TCF4), p53, and members of the Ets family of oncoproteins (81–83). The second general level of regulation of MMP activity is proenzyme activation. MMPs are synthesized as zymogens where a conserved cysteine residue of the pro-domain occupies the active center abolishing its activity. The release of the pro-domain is a complex process that usually involves additional proteolytic enzymes. Thus, there are several MMPs that contain a pro-protein convertase sequence in their pro-domain and are activated in the secretory pathway by furin-like peptidases (84). However, most MMPs are activated at the pericellular space by other MMPs, such as MT-MMPs and MMP-3, or by serine proteases like plasmin, chymase, and tryptase. In most cases, this activation requires two

Metalloproteases and the Degradome

21

cleavages: a first one that exposes cryptic parts of the pro-domain and a second one in the exposed pro-domain that finally releases this structure (82). Substrate binding may also be responsible for pro-domain disengagement, triggering protease activation (85). Finally, other non-protease molecules can also be required for MMP activation, as in the case of MMP-2 activation by MT1MMP, in which TIMP-2 acts as a scaffold protein to locate both MMPs at an adjacent position (82, 85). The third main point of MMP regulation is the inhibition of the enzymatic activity of these metalloproteases. MMP activity can be blocked by general inhibitors present in plasma or tissue fluids, like ␣2-macroglobulins, and also by TIMPs which are more specific inhibitors. There are four TIMPs in humans, which differ in their expression pattern and MMP specificity. In addition, other molecules contain domains with structural similarity to TIMPs which allow them to act as MMP inhibitors. These include the C-terminal fragment of the procollagen C-terminal proteinase enhancer protein (PCPE), the NC1 domain of collagen type IV, the reversion-inducing cysteine-rich protein with Kazal motifs (RECK), and the tissue factor pathway inhibitor-2 (TFPI2) (82, 86–88). Over many years, it has been considered that the main ability of MMPs is the degradation of ECM substrates, such as fibrillar and non-fibrillar collagens, gelatin, aggrecan, fibronectin, laminin, and elastin. However, a large number of non-matrix substrates have been reported for MMPs including cytokines, growth factors, adhesion molecules, other proteases, peptidase inhibitors, clotting factors, and several receptors (89). MMPs play an important role in tissue resorption in several physiological processes, such as endometrial cycling, embryonic development, bone remodeling, postpartum uterine involution, post-lactation mammary involution, and wound healing (81). Nevertheless, many of their effects are not due to tissue destruction since MMPs are implicated in the regulation of a large number of processes via specific processing of a wide range of ECM and non-ECM substrates. These MMP-regulated processes include proliferation, migration, apoptosis, angiogenesis, and inflammation. Several MMPs participate in the regulation of cytokine and chemokine release and activation, which are key steps in the immune response. For example, MMP-1, -2, -3, -7, -9, and -12 are able to process proTNF-␣ into soluble active TNF-␣. MMP-2, -3, and -9 also have the ability to cleave IL-1␤, generating a more active form (90). Moreover, MMP-9 controls IL-2-dependent proliferation of T lymphocytes through degradation of IL-2 receptor type-␣, and MMP-8, -13, and -14 can cleave IL-8 generating truncated forms with increased activity (91, 92). Similar to the case of ADAMs, MMPs are also involved in the regulation of cell proliferation by modulating the bioavailability

22

Ugalde et al.

of active growth factors. Thus, MMP-1 and -3 cleave perlecan and release FGF, whereas MMP-2, -3, and -7 process decorin liberating ECM-anchored TFG-␤ (93, 94). A good example of nonECM substrate processed by MMPs that influences the bioavailability of growth factors is the family of IGFBPs, soluble proteins that bind free IGF, and block its activity. All IGFBPs are susceptible to degradation by MMPs: MMP-1 cleaves IGFBP-2 and -3, MMP-2 processes IGFBP-3 and -5, and MMP-7 degrades all of them (89, 95). On the other hand, MMPs can regulate apoptosis via shedding of some death ligands (96) and modulate angiogenesis through the proteolytic exposure of cryptic anti-angiogenic signals present in ECM proteins. For example, arrestin, canstatin, and tumbastin are angioinhibitory peptides generated after MMP cleavage of collagen IV ␣-1, ␣-2, and ␣-3 chain, respectively. Moreover, MMP-7 mediates the release of endostatin from collagen XVIII and MMP-9 liberates tumstatin from collagen IV, both of which are anti-angiogenic molecules. Epithelial migration is a necessary step for new vessel formation and MMPs also participate in this process. MT1-MMP and MMP-2 cleave laminin5 and generate a cryptic fragment that promotes cell motility. Migration-stimulating factor liberated through MMP-mediated processing of fibronectin also displays cell migration activity. Likewise, ECM degradation mediated by MMPs causes the release of angiogenic factors like VEGF, which is a potent pro-angiogenic molecule and osteoclast chemoattractant. In summary, MMPs perform a broad spectrum of actions influencing important cell processes such as apoptosis, cell migration, and cell proliferation, which ultimately control tissue homeostasis. Due to the variety of processes in which MMPs are implicated, alterations in the structure or regulation of their genes are frequently associated with the development of several human pathologies, including cancer. Enamelysin (MMP-20) mutations cause amelogenesis imperfecta, an inherited tooth malformation due to defects in enamel maturation (97). Likewise, mutations in MMP-13 are responsible for human spondyloepimetaphyseal dysplasia, an autosomal dominant disorder characterized by defective growth and modeling of vertebrae and long bones (98). Moreover, MMP-2 is mutated in patients with multicentric osteolysis with arthritis, another inherited human skeletal disorder that displays altered bone resorption (99). Furthermore, specific MMP-1 and MMP-3 alleles have been associated with an increased susceptibility to different diseases including cancer (81). On the other hand, and as a direct consequence of their important roles of apoptosis, proliferation, and cell migration, deregulation of MMPs has been associated with tumorigenesis in a great variety of tissues. Growth factors released by MMPs generate proliferative and survival signals required by tumor cells. For example, Mmp11-deficient mice develop less

Metalloproteases and the Degradome

23

dimethylbenzanthracene-induced carcinomas whereas MMP-11 increases tumorigenicity in a syngenic model of tumorigenesis, both effects attributed to the ability of this MMP to release IGFs (100, 101). Similarly, MMP-7 inhibits cancer cell apoptosis by releasing HB-EGF or cleaving Fas-L. Angiogenesis is another key step in tumor growth which is regulated by MMPs at several levels (102). Thus, MMP-9 increases the bioavailability of VEGF whereas MMP-14 promotes the endothelial migration by degradation of the vessels matrix (103, 104). However, the most classical roles attributed to MMPs in cancer are those associated with tumor invasion and metastasis. Various MMPs have been implicated in cancer metastasis, like MMP-9, whose downregulation in cancer cells reduces the number of metastasis in mice, or MMP-14, which participates in the degradation of laminin-5 during cancer cell migration (105, 106). Furthermore, MMP-1 and MMP-2 are important functional mediators in the generation of pulmonary metastasis from breast carcinomas (107). MMPs may also contribute to cancer cell evasion from immune system. This is the case of MMP-9, which suppresses proliferation of T lymphocytes via degradation of their IL-2 receptors. Other MMPs reduce the neutrophil efflux by chemokine degradation or inhibit the T-lymphocyte response by TGF-␤ production (81, 82, 108). On the other hand, and contrary to classical ideas in this field, recent works have demonstrated that MMPs are able to inhibit cancer development, thereby acting as tumor suppressors. MMP-8 was the first family member associated with this function after the finding that Mmp8-deficient mice develop more skin tumors in a carcinogenesis assay (109). The mechanism by which this protease exerts its effects has been related to an abnormal inflammatory response observed in these mice subjected to chemical carcinogenesis. Likewise, MMP-26 has been associated with favorable clinical outcome when upregulated in hormonedependent carcinomas (110). Other MMPs, such as MMP-3, -9, -11, -12, and -19, have dual functions in cancer progression, being associated with both favorable and bad prognosis (80).

6. Conclusions In summary, metalloproteases comprise a diverse and complex group of hydrolytic enzymes that participate in almost every biochemical process of the cell and in the tissue homeostasis and physiology of pluricellular organisms. In particular, three groups of metalloproteases, MMPs, ADAMs, and ADAMTSs, have been subjected to intensive research as a consequence of their wide presence in mammals and their important roles in normal

24

Ugalde et al.

physiology and in pathogenic processes, including cancer and arthritis. Accumulative discoveries have revealed the enormous complexity of these three families, which makes it necessary to adopt a global perspective for their study, including the analysis of their inhibitors and regulatory elements. Therefore, it will be necessary to combine classical molecular tools with high-throughput degradomic approaches to shed new light in the functional roles of these proteolytic enzymes in physiology and disease.

Acknowledgments We thank all members of our group for helpful comments and advice. Our work is supported by grants from Ministerio de Edu´ y Ciencia-Spain, Fundacion ´ Lilly, Fundacion ´ La Caixa; cacion ´ “M. Bot´ın”; and the European Union (FP7). The Fundacion Instituto Universitario de Oncolog´ıa is supported by Obra Social Cajastur-Asturias, Spain. References 1. Lopez-Otin, C. and Overall, C. M. (2002) Protease degradomics: a new challenge for proteomics. Nat Rev Mol Cell Biol 3, 509–519. 2. Puente, X. S., Sanchez, L. M., Overall, C. M., and Lopez-Otin, C. (2003) Human and mouse proteases: a comparative genomic approach. Nat Rev Genet 4, 544–558. 3. Puente, X. S., Gutierrez-Fernandez, A., Ordonez, G. R., Hillier, L. W., and LopezOtin, C. (2005) Comparative genomic analysis of human and chimpanzee proteases. Genomics 86, 638–647. 4. Ordo˜ nez, G. R., Hillier, L. W., Warren, W. C., Gr¨utzner, F., Lopez-Otin, C., and Puente, X. S. (2008) Loss of genes implicated in gastric function during platypus evolution. Genome Biol 9, R81. 5. Shah, P. K., Tripathi, L. P., Jensen, L. J., Gahnim, M., Mason, C., Furlong, E. E., Rodrigues, V., White, K. P., Bork, P., and Sowdhamini, R. (2008) Enhanced function annotations for Drosophila serine proteases: a case study for systematic annotation of multimember gene families. Gene 407, 199–215. 6. Garcia-Lorenzo, M., Sjodin, A., Jansson, S., and Funk, C. (2006) Protease gene families in Populus and Arabidopsis. BMC Plant Biol 6, 30. 7. Kinch, L. N., Ginalski, K., and Grishin, N. V. (2006) Site-2 protease regulated intramem-

8.

9. 10.

11.

12.

13.

14.

brane proteolysis: sequence homologs suggest an ancient signaling cascade. Protein Sci 15, 84–93. Brown, M. S. and Goldstein, J. L. (1999) A proteolytic pathway that controls the cholesterol content of membranes, cells, and blood. Proc Natl Acad Sci U S A 96, 11041–11048. Rawlings, N. D. and Barrett, A. J. (1995) Evolutionary families of metallopeptidases. Methods Enzymol 248, 183–228. Reznik, S. E. and Fricker, L. D. (2001) Carboxypeptidases from A to z: implications in embryonic development and Wnt binding. Cell Mol Life Sci 58, 1790–1804. Arolas, J. L., Vendrell, J., Aviles, F. X., and Fricker, L. D. (2007) Metallocarboxypeptidases: emerging drug targets in biomedicine. Curr Pharm Des 13, 349–366. Ambroggio, X. I., Rees, D. C., and Deshaies, R. J. (2004) JAMM: a metalloprotease-like zinc site in the proteasome and signalosome. PLoS Biol 2, E2. Bochtler, M., Odintsov, S. G., Marcyjaniak, M., and Sabala, I. (2004) Similar active sites in lysostaphins and D-Ala-D-Ala metallopeptidases. Protein Sci 13, 854–861. Matsui, M., Fowler, J. H., and Walling, L. L. (2006) Leucine aminopeptidases: diversity in structure and function. Biol Chem 387, 1535–1544.

Metalloproteases and the Degradome 15. Holz, R. C., Bzymek, K. P., and Swierczek, S. I. (2003) Co-catalytic metallopeptidases as pharmaceutical targets. Curr Opin Chem Biol 7, 197–206. 16. Leopoldini, M., Russo, N., and Toscano, M. (2007) Which one among Zn(II), Co(II), Mn(II), and Fe(II) is the most efficient ion for the methionine aminopeptidase catalyzed reaction? J Am Chem Soc 129, 7776–7784. 17. Fushimi, N., Ee, C. E., Nakajima, T., and Ichishima, E. (1999) Aspzincin, a family of metalloendopeptidases with a new zinc-binding motif. Identification of new zinc-binding sites (His(128), His(132), and Asp(164)) and three catalytically crucial residues (Glu(129), Asp(143), and Tyr(106)) of deuterolysin from Aspergillus oryzae by site-directed mutagenesis. J Biol Chem 274, 24195–24201. 18. Bode, W., Gomis-Ruth, F. X., and Stockler, W. (1993) Astacins, serralysins, snake venom and matrix metalloproteinases exhibit identical zinc-binding environments (HEXXHXXGXXH and Met-turn) and topologies and should be grouped into a common family, the ‘metzincins’. FEBS Lett 331, 134–140. 19. Hooper, N. M. (1994) Families of zinc metalloproteases. FEBS Lett 354, 1–6. 20. Gomis-Ruth, F. X. (2003) Structural aspects of the metzincin clan of metalloendopeptidases. Mol Biotechnol 24, 157–202. 21. Albiston, A. L., Ye, S., and Chai, S. Y. (2004) Membrane bound members of the M1 family: more than aminopeptidases. Protein Pept Lett 11, 491–500. 22. Danilczyk, U., Eriksson, U., Crackower, M. A., and Penninger, J. M. (2003) A story of two ACEs. J Mol Med 81, 227–234. 23. Lim, E. J., Sampath, S., Coll-Rodriguez, J., Schmidt, J., Ray, K., and Rodgers, D. W. (2007) Swapping the substrate specificities of the neuropeptidases neurolysin and thimet oligopeptidase. J Biol Chem 282, 9722–9732. 24. Chew, A., Buck, E. A., Peretz, S., Sirugo, G., Rinaldo, P., and Isaya, G. (1997) Cloning, expression, and chromosomal assignment of the human mitochondrial intermediate peptidase gene (MIPEP). Genomics 40, 493–496. 25. Turner, A. J., Isaac, R. E., and Coates, D. (2001) The neprilysin (NEP) family of zinc metalloendopeptidases: genomics and function. Bioessays 23, 261–269. 26. Koppen, M. and Langer, T. (2007) Protein degradation within mitochondria: versatile activities of AAA proteases and other peptidases. Crit Rev Biochem Mol Biol 42, 221–242.

25

27. Pendas, A. M., Zhou, Z., Cadinanos, J., Freije, J. M., Wang, J., Hultenby, K., Astudillo, A., Wernerson, A., Rodriguez, F., Tryggvason, K., and Lopez-Otin, C. (2002) Defective prelamin A processing and muscular and adipocyte alterations in Zmpste24 metalloproteinase-deficient mice. Nat Genet 31, 94–99. 28. Navarro, C. L., Cadinanos, J., De SandreGiovannoli, A., Bernard, R., Courrier, S., Boccaccio, I., Boyer, A., Kleijer, W. J., Wagner, A., Giuliano, F., Beemer, F. A., Freije, J. M., Cau, P., Hennekam, R. C., LopezOtin, C., Badens, C., and Levy, N. (2005) Loss of ZMPSTE24 (FACE-1) causes autosomal recessive restrictive dermopathy and accumulation of Lamin A precursors. Hum Mol Genet 14, 1503–1513. 29. Varela, I., Cadinanos, J., Pendas, A. M., Gutierrez-Fernandez, A., Folgueras, A. R., Sanchez, L. M., Zhou, Z., Rodriguez, F. J., Stewart, C. L., Vega, J. A., Tryggvason, K., Freije, J. M., and Lopez-Otin, C. (2005) Accelerated ageing in mice deficient in Zmpste24 protease is linked to p53 signalling activation. Nature 437, 564–568. 30. Ramirez, C. L., Cadinanos, J., Varela, I., Freije, J. M., and Lopez-Otin, C. (2007) Human progeroid syndromes, aging and cancer: new genetic and epigenetic insights into old questions. Cell Mol Life Sci 64, 155–170. 31. Hase, C. C. and Finkelstein, R. A. (1993) Bacterial extracellular zinc-containing metalloproteases. Microbiol Rev 57, 823–837. 32. Harrington, D. J. (1996) Bacterial collagenases and collagen-degrading enzymes and their potential role in human disease. Infect Immun 64, 1885–1891. 33. Markaryan, A., Morozova, I., Yu, H., and Kolattukudy, P. E. (1994) Purification and characterization of an elastinolytic metalloprotease from Aspergillus fumigatus and immunoelectron microscopic evidence of secretion of this enzyme by the fungus invading the murine lung. Infect Immun 62, 2149–2157. 34. Jousson, O., Lechenne, B., Bontems, O., Capoccia, S., Mignon, B., Barblan, J., Quadroni, M., and Monod, M. (2004) Multiplication of an ancestral gene encoding secreted fungalysin preceded species differentiation in the dermatophytes Trichophyton and Microsporum. Microbiology 150, 301–310. 35. Kosowska, K., Reinholdt, J., Rasmussen, L. K., Sabat, A., Potempa, J., Kilian, M., and Poulsen, K. (2002) The Clostridium ramosum IgA proteinase represents a novel type

26

36.

37.

38.

39. 40.

41.

42.

43.

44.

45.

46.

Ugalde et al. of metalloendopeptidase. J Biol Chem 277, 11987–11994. Stocker, W., Grams, F., Baumann, U., Reinemer, P., Gomis-Ruth, F. X., McKay, D. B., and Bode, W. (1995) The metzincins – topological and sequential relations between the astacins, adamalysins, serralysins, and matrixins (collagenases) define a superfamily of zinc-peptidases. Protein Sci 4, 823–840. Diaz-Perales, A., Quesada, V., Peinado, J. R., Ugalde, A. P., Alvarez, J., Suarez, M. F., Gomis-Ruth, F. X., and Lopez-Otin, C. (2005) Identification and characterization of human archaemetzincin-1 and -2, two novel members of a family of metalloproteases widely distributed in Archaea. J Biol Chem 280, 30367–30375. Quesada, V., Sanchez, L. M., Alvarez, J., and Lopez-Otin, C. (2004) Identification and characterization of human and mouse ovastacin: a novel metalloproteinase similar to hatching enzymes from arthropods, birds, amphibians, and fish. J Biol Chem 279, 26627–26634. Bond, J. S. and Beynon, R. J. (1995) The astacin family of metalloendopeptidases. Protein Sci 4, 1247–1261. Hopkins, D. R., Keles, S., and Greenspan, D. S. (2007) The bone morphogenetic protein 1/Tolloid-like metalloproteinases. Matrix Biol 26, 508–523. Angerer, L., Hussain, S., Wei, Z., and Livingston, B. T. (2006) Sea urchin metalloproteases: a genomic survey of the BMP1/tolloid-like, MMP and ADAM families. Dev Biol 300, 267–281. McCoy, A. J., Liu, H., Falla, T. J., and Gunn, J. S. (2001) Identification of Proteus mirabilis mutants with increased sensitivity to antimicrobial peptides. Antimicrob Agents Chemother 45, 2030–2037. Kida, Y., Inoue, H., Shimizu, T., and Kuwano, K. (2007) Serratia marcescens serralysin induces inflammatory responses through protease-activated receptor 2. Infect Immun 75, 164–174. Driscoll, J. A., Brody, S. L., and Kollef, M. H. (2007) The epidemiology, pathogenesis and treatment of Pseudomonas aeruginosa infections. Drugs 67, 351–368. Matheson, N. R., Potempa, J., and Travis, J. (2006) Interaction of a novel form of Pseudomonas aeruginosa alkaline protease (aeruginolysin) with interleukin6 and interleukin-8. Biol Chem 387, 911–915. Grandvalet, C., Gominet, M., and Lereclus, D. (2001) Identification of genes involved in the activation of the Bacillus thuringien-

47. 48.

49.

50.

51.

52.

53.

54. 55.

56.

57.

sis inhA metalloprotease gene at the onset of sporulation. Microbiology 147, 1805–1813. Gelb, M. H. and Hol, W. G. (2002) Parasitology. Drugs to combat tropical protozoan parasites. Science 297, 343–344. Grandgenett, P. M., Otsu, K., Wilson, H. R., Wilson, M. E., and Donelson, J. E. (2007) A function for a specific zinc metalloprotease of African trypanosomes. PLoS Pathog 3, 1432–1445. Yao, C., Donelson, J. E., and Wilson, M. E. (2003) The major surface protease (MSP or GP63) of Leishmania sp. Biosynthesis, regulation of expression, and function. Mol Biochem Parasitol 132, 1–16. McHugh, B., Krause, S. A., Yu, B., Deans, A. M., Heasman, S., McLaughlin, P., and Heck, M. M. (2004) Invadolysin: a novel, conserved metalloprotease links mitotic structural rearrangements with cell migration. J Cell Biol 167, 673–686. Ramos, O. H. and Selistre-de-Araujo, H. S. (2006) Snake venom metalloproteases – structure and function of catalytic and disintegrin domains. Comp Biochem Physiol C Toxicol Pharmacol 142, 328–346. Milla, M. E., Leesnitzer, M. A., Moss, M. L., Clay, W. C., Carter, H. L., Miller, A. B., Su, J. L., Lambert, M. H., Willard, D. H., Sheeley, D. M., Kost, T. A., Burkhart, W., Moyer, M., Blackburn, R. K., Pahel, G. L., Mitchell, J. L., Hoffman, C. R., and Becherer, J. D. (1999) Specific sequence elements are required for the expression of functional tumor necrosis factor-alpha-converting enzyme (TACE). J Biol Chem 274, 30563–30570. Hougaard, S., Loechel, F., Xu, X., Tajima, R., Albrechtsen, R., and Wewer, U. M. (2000) Trafficking of human ADAM 12-L: retention in the trans-Golgi network. Biochem Biophys Res Commun 275, 261–267. Puente, X. S. and Lopez-Otin, C. (2004) A genomic analysis of rat proteases and protease inhibitors. Genome Res 14, 609–622. Sagane, K., Yamazaki, K., Mizui, Y., and Tanaka, I. (1999) Cloning and chromosomal mapping of mouse ADAM11, ADAM22 and ADAM23. Gene 236, 79–86. Seals, D. F. and Courtneidge, S. A. (2003) The ADAMs family of metalloproteases: multidomain proteins with multiple functions. Genes Dev 17, 7–30. Zheng, Y., Saftig, P., Hartmann, D., and Blobel, C. (2004) Evaluation of the contribution of different ADAMs to tumor necrosis factor alpha (TNFalpha) shedding and of the function of the TNFalpha ectodomain in ensuring selective stimulated shedding by the

Metalloproteases and the Degradome

58.

59.

60.

61.

62.

63.

64.

65.

66.

TNFalpha convertase (TACE/ADAM17). J Biol Chem 279, 42898–42906. Kurisaki, T., Masuda, A., Sudo, K., Sakagami, J., Higashiyama, S., Matsuda, Y., Nagabukuro, A., Tsuji, A., Nabeshima, Y., Asano, M., Iwakura, Y., and Sehara-Fujisawa, A. (2003) Phenotypic analysis of Meltrin alpha (ADAM12)-deficient mice: involvement of Meltrin alpha in adipogenesis and myogenesis. Mol Cell Biol 23, 55–61. Huovila, A. P., Turner, A. J., Pelto-Huikko, M., Karkkainen, I., and Ortiz, R. M. (2005) Shedding light on ADAM metalloproteinases. Trends Biochem Sci 30, 413–422. Loechel, F., Fox, J. W., Murphy, G., Albrechtsen, R., and Wewer, U. M. (2000) ADAM 12-S cleaves IGFBP-3 and IGFBP-5 and is inhibited by TIMP-3. Biochem Biophys Res Commun 278, 511–515. Hartmann, D., de Strooper, B., Serneels, L., Craessaerts, K., Herreman, A., Annaert, W., Umans, L., Lubke, T., Lena Illert, A., von Figura, K., and Saftig, P. (2002) The disintegrin/metalloprotease ADAM 10 is essential for Notch signalling but not for alphasecretase activity in fibroblasts. Hum Mol Genet 11, 2615–2624. Maretzky, T., Reiss, K., Ludwig, A., Buchholz, J., Scholz, F., Proksch, E., de Strooper, B., Hartmann, D., and Saftig, P. (2005) ADAM10 mediates E-cadherin shedding and regulates epithelial cell-cell adhesion, migration, and beta-catenin translocation. Proc Natl Acad Sci U S A 102, 9182–9187. Beck, V., Herold, H., Benge, A., Luber, B., Hutzler, P., Tschesche, H., Kessler, H., Schmitt, M., Geppert, H. G., and Reuning, U. (2005) ADAM15 decreases integrin alphavbeta3/vitronectin-mediated ovarian cancer cell adhesion and motility in an RGD-dependent fashion. Int J Biochem Cell Biol 37, 590–603. Nishimura, H., Myles, D. G., and Primakoff, P. (2007) Identification of an ADAM2ADAM3 complex on the surface of mouse testicular germ cells and cauda epididymal sperm. J Biol Chem 282, 17900–17907. Cal, S., Freije, J. M., Lopez, J. M., Takada, Y., and Lopez-Otin, C. (2000) ADAM 23/MDC3, a human disintegrin that promotes cell adhesion via interaction with the alphavbeta3 integrin through an RGDindependent mechanism. Mol Biol Cell 11, 1457–1469. Ohtsu, H., Dempsey, P. J., and Eguchi, S. (2006) ADAMs as mediators of EGF receptor transactivation by G protein-coupled receptors. Am J Physiol Cell Physiol 291, C1–C10.

27

67. Rocks, N., Paulissen, G., El Hour, M., Quesada, F., Crahay, C., Gueders, M., Foidart, J. M., Noel, A., and Cataldo, D. (2007) Emerging roles of ADAM and ADAMTS metalloproteinases in cancer. Biochimie 90(2), 369–379. 68. Llamazares, M., Cal, S., Quesada, V., and Lopez-Otin, C. (2003) Identification and characterization of ADAMTS-20 defines a novel subfamily of metalloproteinasesdisintegrins with multiple thrombospondin1 repeats and a unique GON domain. J Biol Chem 278, 13382–13389. 69. Cal, S., Obaya, A. J., Llamazares, M., Garabaya, C., Quesada, V., and LopezOtin, C. (2002) Cloning, expression analysis, and structural characterization of seven novel human ADAMTSs, a family of metalloproteinases with disintegrin and thrombospondin-1 domains. Gene 283, 49–62. 70. Porter, S., Clark, I. M., Kevorkian, L., and Edwards, D. R. (2005) The ADAMTS metalloproteinases. Biochem J 386, 15–27. 71. Hashimoto, G., Shimoda, M., and Okada, Y. (2004) ADAMTS4 (aggrecanase-1) interaction with the C-terminal domain of fibronectin inhibits proteolysis of aggrecan. J Biol Chem 279, 32483–32491. 72. Little, C. B., Meeker, C. T., Golub, S. B., Lawlor, K. E., Farmer, P. J., Smith, S. M., and Fosang, A. J. (2007) Blocking aggrecanase cleavage in the aggrecan interglobular domain abrogates cartilage erosion and promotes cartilage repair. J Clin Invest 117, 1627–1636. 73. Colige, A., Nuytinck, L., Hausser, I., van Essen, A. J., Thiry, M., Herens, C., Ades, L. C., Malfait, F., Paepe, A. D., Franck, P., Wolff, G., Oosterwijk, J. C., Smitt, J. H., Lapiere, C. M., and Nusgens, B. V. (2004) Novel types of mutation responsible for the dermatosparactic type of Ehlers-Danlos syndrome (Type VIIC) and common polymorphisms in the ADAMTS2 gene. J Invest Dermatol 123, 656–663. 74. Dagoneau, N., Benoist-Lasselin, C., Huber, C., Faivre, L., Megarbane, A., Alswaid, A., Dollfus, H., Alembik, Y., Munnich, A., Legeai-Mallet, L., and Cormier-Daire, V. (2004) ADAMTS10 mutations in autosomal recessive Weill-Marchesani syndrome. Am J Hum Genet 75, 801–806. 75. Rao, C., Foernzler, D., Loftus, S. K., Liu, S., McPherson, J. D., Jungers, K. A., Apte, S. S., Pavan, W. J., and Beier, D. R. (2003) A defect in a novel ADAMTS family member is the cause of the belted white-spotting mutation. Development 130, 4665–4672.

28

Ugalde et al.

76. Jungers, K. A., Le Goff, C., Somerville, R. P., and Apte, S. S. (2005) Adamts9 is widely expressed during mouse embryo development. Gene Expr Patterns 5, 609–617. 77. Levy, G. G., Nichols, W. C., Lian, E. C., Foroud, T., McClintick, J. N., McGee, B. M., Yang, A. Y., Siemieniak, D. R., Stark, K. R., Gruppo, R., Sarode, R., Shurin, S. B., Chandrasekaran, V., Stabler, S. P., Sabio, H., Bouhassira, E. E., Upshaw, J. D., Jr., Ginsburg, D., and Tsai, H. M. (2001) Mutations in a member of the ADAMTS gene family cause thrombotic thrombocytopenic purpura. Nature 413, 488–494. 78. Luque, A., Carpizo, D. R., and Iruela-Arispe, M. L. (2003) ADAMTS1/METH1 inhibits endothelial cell proliferation by direct binding and sequestration of VEGF165. J Biol Chem 278, 23656–23665. 79. Llamazares, M., Obaya, A. J., MoncadaPazos, A., Heljasvaara, R., Espada, J., Lopez-Otin, C., and Cal, S. (2007) The ADAMTS12 metalloproteinase exhibits antitumorigenic properties through modulation of the Ras-dependent ERK signalling pathway. J Cell Sci 120, 3544–3552. 80. Lopez-Otin, C. and Matrisian, L. M. (2007) Emerging roles of proteases in tumour suppression. Nat Rev Cancer 7, 800–808. 81. Folgueras, A. R., Pendas, A. M., Sanchez, L. M., and Lopez-Otin, C. (2004) Matrix metalloproteinases in cancer: from new functions to improved inhibition strategies. Int J Dev Biol 48, 411–424. 82. Overall, C. M. and Lopez-Otin, C. (2002) Strategies for MMP inhibition in cancer: innovations for the post-trial era. Nat Rev Cancer 2, 657–672. 83. Clark, I. M., Swingler, T. E., Sampieri, C. L., and Edwards, D. R. (2007) The regulation of matrix metalloproteinases and their inhibitors. Int J Biochem Cell Biol 40(6–7), 1362–1378. 84. Pei, D. and Weiss, S. J. (1995) Furindependent intracellular activation of the human stromelysin-3 zymogen. Nature 375, 244–247. 85. Kinoshita, T., Sato, H., Takino, T., Itoh, M., Akizawa, T., and Seiki, M. (1996) Processing of a precursor of 72-kilodalton type IV collagenase/gelatinase A by a recombinant membrane-type 1 matrix metalloproteinase. Cancer Res 56, 2535–2538. 86. Oh, J., Takahashi, R., Kondo, S., Mizoguchi, A., Adachi, E., Sasahara, R. M., Nishimura, S., Imamura, Y., Kitayama, H., Alexander, D. B., Ide, C., Horan, T. P., Arakawa, T., Yoshida, H., Nishikawa, S., Itoh, Y., Seiki, M., Itohara, S., Takahashi, C., and Noda,

87.

88.

89.

90.

91.

92.

93.

94.

95.

M. (2001) The membrane-anchored MMP inhibitor RECK is a key regulator of extracellular matrix integrity and angiogenesis. Cell 107, 789–800. Herman, M. P., Sukhova, G. K., Kisiel, W., Foster, D., Kehry, M. R., Libby, P., and Schonbeck, U. (2001) Tissue factor pathway inhibitor-2 is a novel inhibitor of matrix metalloproteinases with implications for atherosclerosis. J Clin Invest 107, 1117–1126. Mott, J. D., Thomas, C. L., Rosenbach, M. T., Takahara, K., Greenspan, D. S., and Banda, M. J. (2000) Post-translational proteolytic processing of procollagen Cterminal proteinase enhancer releases a metalloproteinase inhibitor. J Biol Chem 275, 1384–1390. McCawley, L. J. and Matrisian, L. M. (2001) Matrix metalloproteinases: they’re not just for matrix anymore! Curr Opin Cell Biol 13, 534–540. Schonbeck, U., Mach, F., and Libby, P. (1998) Generation of biologically active IL1 beta by matrix metalloproteinases: a novel caspase-1-independent pathway of IL-1 beta processing. J Immunol 161, 3340–3346. Van Lint, P. and Libert, C. (2007) Chemokine and cytokine processing by matrix metalloproteinases and its effect on leukocyte migration and inflammation. J Leukoc Biol 82, 1375–1381. Sheu, B. C., Hsu, S. M., Ho, H. N., Lien, H. C., Huang, S. C., and Lin, R. H. (2001) A novel role of metalloproteinase in cancermediated immunosuppression. Cancer Res 61, 237–242. Imai, K., Hiramatsu, A., Fukushima, D., Pierschbacher, M. D., and Okada, Y. (1997) Degradation of decorin by matrix metalloproteinases: identification of the cleavage sites, kinetic analyses and transforming growth factor-beta1 release. Biochem J 322 (Pt 3), 809–814. Whitelock, J. M., Murdoch, A. D., Iozzo, R. V., and Underwood, P. A. (1996) The degradation of human endothelial cell-derived perlecan and release of bound basic fibroblast growth factor by stromelysin, collagenase, plasmin, and heparanases. J Biol Chem 271, 10079–10086. Nakamura, M., Miyamoto, S., Maeda, H., Ishii, G., Hasebe, T., Chiba, T., Asaka, M., and Ochiai, A. (2005) Matrix metalloproteinase-7 degrades all insulin-like growth factor binding proteins and facilitates insulin-like growth factor bioavailability. Biochem Biophys Res Commun 333, 1011–1016.

Metalloproteases and the Degradome 96. Ii, M., Yamamoto, H., Adachi, Y., Maruyama, Y., and Shinomura, Y. (2006) Role of matrix metalloproteinase-7 (matrilysin) in human cancer invasion, apoptosis, growth, and angiogenesis. Exp Biol Med (Maywood) 231, 20–27. 97. Papagerakis, P., Lin, H. K., Lee, K. Y., Hu, Y., Simmer, J. P., Bartlett, J. D., and Hu, J. C. (2008) Premature stop codon in MMP20 causing amelogenesis imperfecta. J Dent Res 87, 56–59. 98. Kennedy, A. M., Inada, M., Krane, S. M., Christie, P. T., Harding, B., Lopez-Otin, C., Sanchez, L. M., Pannett, A. A., Dearlove, A., Hartley, C., Byrne, M. H., Reed, A. A., Nesbit, M. A., Whyte, M. P., and Thakker, R. V. (2005) MMP13 mutation causes spondyloepimetaphyseal dysplasia, Missouri type (SEMD(MO)). J Clin Invest 115, 2832–2842. 99. Martignetti, J. A., Aqeel, A. A., Sewairi, W. A., Boumah, C. E., Kambouris, M., Mayouf, S. A., Sheth, K. V., Eid, W. A., Dowling, O., Harris, J., Glucksman, M. J., Bahabri, S., Meyer, B. F., and Desnick, R. J. (2001) Mutation of the matrix metalloproteinase 2 gene (MMP2) causes a multicentric osteolysis and arthritis syndrome. Nat Genet 28, 261–265. 100. Wu, E., Mari, B. P., Wang, F., Anderson, I. C., Sunday, M. E., and Shipp, M. A. (2001) Stromelysin-3 suppresses tumor cell apoptosis in a murine model. J Cell Biochem 82, 549–555. 101. Boulay, A., Masson, R., Chenard, M. P., El Fahime, M., Cassard, L., Bellocq, J. P., Sautes-Fridman, C., Basset, P., and Rio, M. C. (2001) High cancer cell death in syngeneic tumors developed in host mice deficient for the stromelysin-3 matrix metalloproteinase. Cancer Res 61, 2189–2193. 102. Noel, A., Jost, M., and Maquoi, E. (2008) Matrix metalloproteinases at cancer tumorhost interface. Semin Cell Dev Biol 19, 52–60.

29

103. Bergers, G., Brekken, R., McMahon, G., Vu, T. H., Itoh, T., Tamaki, K., Tanzawa, K., Thorpe, P., Itohara, S., Werb, Z., and Hanahan, D. (2000) Matrix metalloproteinase-9 triggers the angiogenic switch during carcinogenesis. Nat Cell Biol 2, 737–744. 104. Hiraoka, N., Allen, E., Apel, I. J., Gyetko, M. R., and Weiss, S. J. (1998) Matrix metalloproteinases regulate neovascularization by acting as pericellular fibrinolysins. Cell 95, 365–377. 105. Koshikawa, N., Giannelli, G., Cirulli, V., Miyazaki, K., and Quaranta, V. (2000) Role of cell surface metalloprotease MT1-MMP in epithelial cell migration over laminin-5. J Cell Biol 148, 615–624. 106. Itoh, T., Tanioka, M., Matsuda, H., Nishimoto, H., Yoshioka, T., Suzuki, R., and Uehira, M. (1999) Experimental metastasis is suppressed in MMP-9-deficient mice. Clin Exp Metastasis 17, 177–181. 107. Gupta, G. P., Nguyen, D. X., Chiang, A. C., Bos, P. D., Kim, J. Y., Nadal, C., Gomis, R. R., Manova-Todorova, K., and Massague, J. (2007) Mediators of vascular remodelling coopted for sequential steps in lung metastasis. Nature 446, 765–770. 108. Egeblad, M. and Werb, Z. (2002) New functions for the matrix metalloproteinases in cancer progression. Nat Rev Cancer 2, 161–174. 109. Balbin, M., Fueyo, A., Tester, A. M., Pendas, A. M., Pitiot, A. S., Astudillo, A., Overall, C. M., Shapiro, S. D., and Lopez-Otin, C. (2003) Loss of collagenase-2 confers increased skin tumor susceptibility to male mice. Nat Genet 35, 252–257. 110. Savinov, A. Y., Remacle, A. G., Golubkov, V. S., Krajewska, M., Kennedy, S., Duffy, M. J., Rozanov, D. V., Krajewski, S., and Strongin, A. Y. (2006) Matrix metalloproteinase 26 proteolysis of the NH2-terminal domain of the estrogen receptor beta correlates with the survival of breast cancer patients. Cancer Res 66, 2716–2724.

Chapter 2 Mouse Models of MMP and TIMP Function Sean E. Gill, Sean Y. Kassim, Timothy P. Birkland, and William C. Parks Abstract As their name implies, matrix metalloproteinases (MMPs) are thought to be responsible for the turnover of connective tissue proteins, a function that is indeed performed by some family members. However, matrix degradation is possibly not the predominant function of these enzymes. Several studies have demonstrated that MMPs also act on a variety of non-matrix extracellular proteins, such as cytokines, chemokines, receptors, junctional proteins, and antimicrobial peptides, to mediate a wide range of biological processes, such as repair, immunity, and angiogenesis. Our understanding of the many, diverse and, at times, unexpected functions of MMPs largely arose from the use of gene-targeted mice. In this chapter, we discuss the phenotypes of some MMP-deficient and TIMP-null mice and strategies and pitfalls in targeted mutagenesis. Key words: Knockout, knockdown, substrate identification, conditional, inhibitors, tissue specific, compensation, redundancy.

1. Introduction Proteolysis is one of the several post-translational mechanisms that regulate protein activity, and it is the principal means of ending a protein’s life and recycling its amino acids for reuse. Evolution has provided us with six families of proteinases, defined by the amino acid or the cofactor that catalyzes the nucleophilic attack on the peptide backbone of substrate proteins. Proteinases function both inside and outside of the cell, with serine and metalloproteinases being the most abundant of extracellular proteinases (1). Of the metalloproteinases, matrix metalloproteinases (MMPs) – the subject of this volume – have been long thought I.M. Clark (ed.), D.A. Young, A.D. Rowan (Consulting editors), Matrix Metalloproteinase Protocols, Methods in Molecular Biology 622, DOI 10.1007/978-1-60327-299-5 2, © Springer Science+Business Media, LLC 2001, 2010

31

32

Gill et al.

to be responsible for turnover of connective tissue proteins, such as the collagens, elastin, and basement membrane components, functions which some specific MMPs – often to a limited extent – do confer. However, matrix turnover is neither the sole nor the predominant function of MMPs. Findings from many groups over the past decade or so have shown that MMPs function in a variety of processes, such as immunity, epithelial repair, leukocyte influx and activation, and more (2–6). Over a tenth of our genome codes for proteins with a signal sequence, thus placing them (as well as secreted proteins whose genes do not code for a signal peptide) within the reach of MMPs. That is a lot of potential substrates. Because proteolysis is a common mechanism used to control the activity of extracellular proteins (e.g., the coagulation and complement cascades, activation of latent cytokines, prohormones, neuropeptides, and digestive enzymes, processing of matrix precursors, and more), it is not at all surprising that a large family of proteinases, like the MMPs, shape a variety of physiological processes. The realization that, as a family, MMPs have evolved pleiotropic functions (in contrast, individual MMPs are limited in their scope of activities and, hence, substrates) came about with the widespread use of reverse genetics. Understanding what an MMP actually does (as opposed to what it can do, i.e., what it can cleave or degrade in a defined, in vitro setting) was not achievable until the consequence of enzyme depletion could be studied in genetically modified mice. In this chapter, we discuss how targeted mutagenesis in mice can be used to understand MMP function.

2. Matrix Metalloproteinases (MMPs)

MMPs are a family of zinc-dependant endopeptidases composed of 25 known members, 24 of which are in mammals (3). As stated, MMPs cleave a variety of extracellular substrates, resulting in the release and activation of growth factors from the cell membrane or the extracellular matrix, the shedding of receptors and cell adhesion proteins from the cell surface, the breakdown or modification of connective tissue proteins, and potentially the activation of the zymogen form of other MMPs, among several other functions (5, 7–9). As a consequence of this ability to act on effector proteins, MMP-mediated proteolysis controls a wide range of cell behaviors and responses to environmental insults. The structural features of MMPs have been thoroughly discussed in several reviews (6, 10–12). The defining features of MMPs – the pro-, catalytic, and hemopexin-like domains – are quite similar among members (10), and the inclusion of other

MMP and TIMP Models

33

motifs or the omission of the hemopexin-like domain (as in MMP7 and 26) provides the individual enzymes with their unique catalytic properties. MMPs are often subdivided into subgroups based on structural similarities or substrate preference, the latter classification being largely misguided. There are, however, two motifs that we agree can be used to sub-classify MMPs in a meaningful way. One is the furin-recognition site found at the junction of the pro and catalytic domains in about one-third of MMPs. Pro-MMPs that contain this motif are activated within the secretion pathway. In contrast, we really know little of how the other two-thirds are activated. Identifying the mechanisms controlling how pro-MMPs are activated in vivo is an area ripe for investigation. We have recently discussed these concepts in another review (9). The other clear division among MMPs, and one which impacts function, is between those enzymes with a transmembrane domain (the MT-MMPs) and those without. The MMPs without a transmembrane domain are often called “soluble MMPs”. This term is not used to imply that the MT-MMPs are insoluble, but rather that the non-MT-MMPs are secreted. But what does soluble mean? That the enzymes are floating around aimlessly in search of a substrate? Unlikely. Modeling and biochemical studies of granular serine proteinases released by neutrophils demonstrate that proteinases rapidly loose effective catalytic ability as they diffuse from the cell surface (14). In contrast, at the cell surface, enzymes (and other proteins) can be oligomerized into locally high concentrations. We propose that all “soluble” MMPs are anchored to something (integrins, proteoglycans, lipids, etc.) at or on the cell surface and it is in this compartmentalized state that the proteinases act on their target substrates (3, 9). The MT-MMPs have a built-in means to be compartmentalized at the cell surface. If this anchored-to-the-cell concept is indeed true, then the substrates must also be nearby. Indeed, many of the confirmed physiological substrates of MMPs are membrane proteins themselves.

3. Tissue Inhibitors of Metalloproteinases (TIMPs)

Once activated, MMP catalysis needs to be eventually silenced, and being bound by one of the four tissue inhibitors of metalloproteinases (TIMPs) is considered the principal mechanism of enzyme inactivation (13, 15, 16). However, other inhibitors and inhibitory mechanisms of metalloproteinases have been identified. These include ␣2 -macroglobulin, a potent inhibitor of proteinases in tissue fluids, and reversion-inducing cysteine-rich protein with Kazal motifs (RECK), the only known transmem-

34

Gill et al.

brane MMP inhibitor (7, 15, 17). Reactive oxygen species and endocytosis may function in silencing MMPs in vivo (18). Despite the identification of these various MMP inhibitors, there are many gaps in our knowledge of how MMP catalysis is silenced in vivo. TIMPs inhibit MMPs in a 1:1 inhibitor-to-enzyme ratio (13, 15, 19). This inhibition occurs through interaction of the Nterminal domain of the TIMP molecule, specifically the first four amino acids, with the active site of the MMP (19). TIMPs coordinate the catalytic site Zn2+ and bind to the active site in a similar fashion to an MMP substrate (19). Generally, all TIMPs are capable of inhibiting all known MMPs; however, the efficacy of MMP inhibition varies with each TIMP (15, 19). TIMP1, 2, and 4 are secreted (15, 19) (and likely indeed function in a soluble state), while TIMP3 is bound to sulfated glycosaminoglycans in the extracellular matrix (20). Interestingly, TIMP2 is required for MMP2 activation in vivo (21, 22), and other TIMPs have demonstrated functions independent of blocking MMP activity (23, 24).

4. MMP Function: Identifying Substrates

To understand the function of an individual MMP expressed by a specific cell type within a physiological setting – such as organogenesis, repair, inflammation, or tumor progression – we need to determine both the protein substrate upon which the proteinase acts and the consequence of that proteolysis, be it a gain– or lossof-function processing. Thus, an important goal of current MMP research is the identification of physiological substrates and an understanding of how this proteolysis affects a specific function. Because MMPs do not act on consensus cleavage sites, candidate substrates cannot be selected in silico. Identifying MMP substrates has been accomplished using various strategies. Possibly the most common approach has been to incubate an active MMP with a suspected substrate under optimal, defined conditions and assess if the target protein is cleaved or degraded (see, e.g., Chapters 15, 16, 22, and 24). However, this approach tells us only what an MMP can do, not what it does do. In a test tube, most MMPs are non-specific and – as for many proteinases – can cleave peptide bonds in proteins they may never see in real life. (Papain is an illustrative example of this concept. This enzyme is routinely used to cleave all sorts of animal proteins in a lab setting, although it is unlikely that papain evolved in papaya to perform such processing events in the living plant.) Although in vitro proteolysis assays are easy and an essential tool for verification (see below), most of the proteins identified as MMP substrates only by this approach probably are not.

MMP and TIMP Models

35

Some groups have successfully used exosite screening and other affinity-based systems to find potential substrates. This approach takes advantage of the ability of the non-catalytic domains of MMPs, typically the hemopexin domain, to bind potential substrates. For example, using exosite screening in a yeast two-hybrid system, Overall and coworkers identified CCL7/MCP3 as a substrate of MMP2 (25), providing among the earliest evidence supporting the concept that MMPs function – either directly or indirectly – in controlling chemokine activity. Over the past decade, proteomics has emerged as the way to find substrates when used to study proteinases, this approach is sometimes referred to as degradomics (26). The basic strategy is to use mass spectrometry to compare the proteome of a control sample to that of sample with altered expression (either loss or gain of function) of a specific MMP. Tryptic peptides that are underrepresented in the presence of an over-expressed MMP or over-represented in a sample from knockout or knockdown tissues or cells would be candidate substrates (see e.g. Chapter 26). Such an approach was recently use to identify CD18 as a substrate shed by MMP9 from the surface of macrophages (89). The use of knockout and transgenic mice, however, dramatically improves the odds of uncovering specific MMP-substrate relationships, and as alluded to above, control over MMP expression is key to proteomics approaches. Genetic manipulation of MMP expression coupled with deductive experimentation has been used to successfully identify substrates in the pre-proteomics era. Close examination of the phenotypes of knockout mice does point to candidates, such as an excess of type I collagen deposition in MMP14 null mice (27) and reduced apoptosis in the prostate glands of castrated MMP7 deficient mice (28). By careful observation of phenotypes, substrates can be found and confirmed.

5. MMP and TIMP-Null and Transgenic Mice

The 2007 Nobel Prize in Medicine was awarded to Mario R. Capecchi, Sir Martin J. Evans, and Oliver Smithies “for their discoveries of principles for introducing specific gene modifications in mice by the use of embryonic stem cells” (nobelprize.org). The ability to selectively knock out, knock down, mutate, or over-express specific genes has proved to be a valuable tool in modern biology, and several studies have used mice deficient in specific MMPs to identify in vivo substrates. Several mouse strains have been engineered for many of the MMP and TIMP families using traditional gene-targeting technology (Tables 2.1, 2.2, and 2.3). These mouse lines have resulted in identification and verification of biological processes requiring MMPs and TIMPs. The accompanying tables

36

Gill et al.

Table 2.1 Developmental Phenotypes in MMP and TIMP Null Mice Development/Homeostasis Lethal Mouse

Embryonic

Tissue Defect Postnatal

Transient

Permanent

Growth Metabolism

Mmp2 -/Mmp3 -/Mmp7 -/Mmp8 -/Mmp9 -/Mmp2/9 -/Mmp10

-/-

Mmp11

-/-

Mmp12

-/-

Mmp13

-/-

Mmp14

-/-

Mmp2/14 -/Mmp16 -/Mmp14/16 -/2

Mmp17 -/-

Mmp19 -/Mmp20 -/Mmp24 2

-/-

Mmp25 -/-

Mmp28 -/Mmp7/28 Timp1

-/-

Timp2

-/-

-/-

Timp3 -/Timp4

-/-

Developmental Phenotypes1 None

Mild

Severe

1. Scoring is based on phenotypes reported. 2. Generated by companies and included in NIH Mouse Genome Informatics Database; however, little information on these mouse lines are provided.

summarize – albeit somewhat subjectively and incompletely – the phenotypes observed for the knockout mice models generated to date. These tables were generated from data of many papers, too many to list here. For detailed information on the phenotypes reported in specific MMP and TIMP knockout mice, we recommend that the original reports be read, and a good starting point are some reviews from the past several years focused on more detailed information on phenotypes (3, 4, 6, 7, 13, 29–33). The most evident conclusion from this collection of information is that MMPs and TIMPs have a minor, if any, role in development and homeostasis in the unchallenged mouse. A key exception to this generalization is MMP14. Mice lacking this proteinase suffer from extensive bony defects and die some weeks after birth. Although mice deficient in either MMP2, 9, 12, 13, 14, or 20 have permanent tissue defects, most are not severe and

MMP and TIMP Models

37

Table 2.2 Predicted Functions of MMPs and TIMPs in Repair and Immunity1 Normal Response to Injury & Infection Immunity Inflammation

Mouse Mmp2

Angiogenesis

Tissue Repair

-/-

Mmp3 -/Mmp7 -/-

NF

Mmp8 -/Mmp9 -/-

2

Mmp2/9 -/Mmp10 -/Mmp11 -/Mmp12 -/Mmp13 -/-

2

Mmp14 -/Mmp2/14

-/-

Mmp16 -/Mmp14/16 -/Mmp17 -/Mmp19 -/Mmp20 -/Mmp24 -/Mmp25 -/Mmp28 -/Mmp7/28 -/Timp1 -/Timp2 -/Timp3 -/Timp4 -/-

MMP Function in Tissue Responses Promotes Inhibits None Reported NF

None Found

1. The functions assigned are based on observed phenotypes in null mice and are those predicted that the MMP or TIMP serves when present (i.e., in wildtype mice). Thus, the functions suggested in this table are the opposite of those seen null mice. 2. Different phenotypes observed with different tissue-specific injury models or knock-out lines.

do not grossly affect fertility, growth, or lifespan. It is interesting to note that three of the TIMP knockout animals have permanent tissue defects, such as spontaneous emphysema in Timp3−/− mice. The broader role of TIMPs rather than MMPs in development may be related to there being fewer TIMP family members or, more likely, to the multiple potential targets of TIMP inhibition (i.e., if TIMP1 inhibits several MMPs, its deletion could result in the disruption of several MMP-dependent responses). Overall, most MMPs appear not to be necessary for the development and maintenance of an independently living animal. The lack of a phenotype in an MMP-null mouse could be attributed to compensation by another proteinase or redundancy by coexpressed backup systems. However, in vivo evidence for compensation among MMPs is lacking (3, 34). But what is meant by compensation? For MMPs, compensation is the activity of one MMP making up for the loss of another. Currently, there is little evidence for increased or compensating MMP activity in MMP knockout mice. Although increased expression of stromelysin-

38

Gill et al.

Table 2.3 Predicted Function in Disease1 Disease Joint & Bone3

Mouse

Cancer

Vascular3

Mmp2 -/Mmp3 -/Mmp7 -/-

NF

Mmp8 -/Mmp9 -/Mmp2/9 -/Mmp10 -/Mmp11 -/Mmp12 -/Mmp13 -/Mmp14 -/Mmp2/14 -/Mmp16 -/Mmp14/16 -/Mmp17 -/Mmp19 -/-

2

Mmp20 -/Mmp24 -/Mmp25 -/Mmp28 -/Mmp7/28 -/Timp1 -/-

2

Timp2 -/Timp3 -/Timp4 -/-

MMP Function in Disease Promotes Inhibits None Reported NF

None Found

1. The functions assigned are based on observed phenotypes in null mice and are the functions predicted that the MMP or TIMP serves when present (i.e., in wildtype mice). Thus, the functions suggested in this table are the opposite of those seen null mice. 2. Different phenotypes observed with different injury models or knock-out lines. 3. A role in these categories refers to predicted functions in acquired, spontaneous, or induced disease models. Congenital defects, which are listed in Table 2.1, are not included here.

1 (MMP3) and stromelysin-2 (MMP10) is seen in the uteri of some Mmp7−/− mice (35), the stromelysins are expressed in compartments distinct from where matrilysin (MMP7) is produced. Because MMP7 functions in mucosal immunity (3), the increased expression of MMP3 and MMP10 may represent an altered host response and not compensation. Instead of compensation, redundancy in essential biological processes may explain the overall lack of development phenotypes among MMP-null mice. This possibility is suggested by the examination of MMP-substrate interactions in vitro, where multiple MMPs have been demonstrated to cleave the same substrate (29). For example, many MMPs, including MMP2, 3, 7, 11, 12, 14, 19, 25, and 26, can degrade fibronectin in vitro

MMP and TIMP Models

39

(29). The efficacy, however, with which these MMPs cleave fibronectin in vitro differs among enzymes. If the principal MMP that cleaves fibronectin in vivo is deleted (and it is not yet clear if fibronectin turnover is the responsibility of an MMP), it is possible that another MMP may process (or degrade) fibronectin but at a slower rate. In a wild-type animal, this lower affinity MMP would likely not interact with fibronectin, as fibronectin would have already been bound and processed by the higher affinity MMP; however, in the knockout mouse, the substrate is suddenly made available and, as observed in vitro, the lower affinity MMP subsequently cleaves fibronectin providing redundancy for the system and leading to a lack of identifiable phenotypes. However, reduced processing of a substrate may not be manifest by an overt phenotype. Demonstrating compensation or redundancy in developmental models would require generating mice with multiple MMP gene deletions. Triple ADAM (a disintegrin and metalloproteinase domain) null mice, lacking ADAMs 9, 12, and 15, have been generating, yet these animals have no overt phenotype or alterations in litter size or expected genotype ratios (36). Either the individual role of each ADAM is subtle or these enzymes do not function in developmental processes. However, assigning relevance to an enzyme (or any protein, for that matter) by its requirement for development or homeostasis is, of course, limiting. After all, nature did not allow MMPs and TIMPs to expand through evolution just be vestigial genetic baggage. The generalized lack of developmental phenotypes among most MMP knockout mice is not surprising. Typically, MMPs are not expressed in normal, healthy tissues, or with notable exceptions, their production and activity are at nearly undetectable levels. In contrast, some level of MMP expression is seen in any repair or remodeling process and in any diseased or inflamed tissue. Although the qualitative patterns and quantitative levels of MMPs vary among tissues, diseases, tumors, inflammatory conditions, and cell lines, a reasonably safe generalization is that activated cells, whether in tissues or in culture dish, express MMPs. As seen in Tables 2.2 and 2.3, phenotypes are revealed under challenged states (and due to space considerations, we highlight only a few, broad processes). Importantly, phenotypes are seen and are mechanistically distinct among MMP-null mice, and the fact that many roles for specific MMPs and TIMPs have been reported in challenged mice argues against compensation or redundancy within the family. Even though two or more MMPs may be able to cleave or degrade the same proteins in vitro, this does not mean that they do so in vivo. Overall, it appears that the MMP family expanded and evolved to function in the host response to environmental stress.

40

Gill et al.

As seen in Tables 2.2 and 2.3, MMPs can either promote or restrain disease or repair processes and likely do so by affecting multiple and apparently opposing processes by the same cell at the same time (3, 7, 37). For example, in our studies in tissue repair and inflammation, we have determined that MMP7 is required for wound closure and generation of antimicrobial activity (38–40), clearly beneficial functions, but this MMP also promotes neutrophil influx and activation, which can be damaging and lethal (41, 42). Cancer provides another illustrative example of the good and bad effects of MMPs. Numerous studies have shown that most MMPs are expressed in some form of cancer and that MMPs are produced by all cells (cancer cells, interstitial cells, endothelial cells, leukocytes) in the tumor environment (7). However, these studies have been largely descriptive and have not shed much light on function. Thus, it is not clear if an individual MMP made by a given cell type contributes to cancer progression and growth or if it functions as part of a host defense mechanism. Several studies have shown that specific MMPs function in cancer progression. For example, when crossed with min/+ (ApcMin ) mice, which spontaneously develop intestinal tumors, Mmp7−/− mice – the first MMP null generated – had a significantly lower tumor load and the tumors that arose were significantly smaller than those in Mmp7+/+ min/+ mice (38). Gelatinase-B (MMP9) is seen in essentially all forms of cancer, where it is primarily produced by neutrophils and macrophages in the inflammatory reaction surrounding the tumor. Although these leukocytes have generally been considered to be part of host-mediated anti-tumor immunity (and hence a beneficial response), Coussens and coworkers, using a multi-stage model of viral-induced tumorigenesis, demonstrated that macrophagederived MMP9 contributes to the progression of more advanced and aggressive adenocarcinomas (43). Furthermore, studies by Weiss and coworkers established that MT1-MMP (MMP14) is essential for tumor cell survival and growth within an interstitial matrix environment (44, 45). On the other hand, reduced tumor burden is seen in mice lacking MMP3 (46) or MMP8 (47). These studies – and others (7) – demonstrate that specific MMPs function in different aspects of tumorigenesis.

6. Basic Strategies for Targeted Mutagenesis

Over the past 12 years, mice carrying mutations causing loss of function for many of the MMP and TIMP families have been generated using standard targeted mutagenesis strategies and techniques. A vector designed to mutate or delete regions of the gene that code for functional elements is transfected

MMP and TIMP Models

41

into embryonic stem (ES) cells. In other words, these vectors lack the sequences to be targeted. For the MMPs, most knockouts were made by deleting the catalytic domain or by targeting upstream exons, resulting in premature stop codons and/or frame-shift mutations. TIMP-null mice were generated by deletion of known or suspected function motifs. One or more exons are usually targeted, and the vector is designed such that a frame shift is inserted into the remaining exons causing generation of new, premature stop codons. With somewhat broad probability (about 1/100 or lower), the vector is incorporated into the targeted allele by homologous recombination. ES cells with the incorporated vector are identified and amplified by positive selection, typically by including a neomycin phosphotransferase cDNA driven by a constitutive promoter, such as the phosphoglycerate kinase (PGK) promoter. The PGK-neomycin cassette is often used to replace the targeted gene elements and is flanked by arms of homologous gene sequence. The homologous arms can be between 1 and 6 kb, and in general, the longer the arms, the greater the probability of homologous recombination. Constructs containing a short and long arm, rather than two long arms, may lead to a slightly decreased incidence of homologous recombination but have the advantage of allowing a PCR genotyping strategy running from the PGK-neomycin cassette to gene sequence outside of the targeting construct. To control for insertion at a nonhomologous locus, targeting vectors can also include an expression cassette for a poison, such as diphtheria toxin, as we used in generating Mmp28−/− mice (90). Such negative selection cassettes are placed at either end of the targeting vector, thereby ensuring that they are deleted during homologous (but not nonhomologous) recombination. Selected ES cells are then injected in blastocysts, which requires a skilled technician. Typically, mouse ES cells used to generate MMP and TIMP knockout mice typically originated from 129/SJ strain or close variants and the blastocysts were from pseudo-pregnant C57Bl/6 mice. Chimeric mice are identified by mixed coat color and are interbred to generate germline heterozygotes, which are then bred to produce homozygous nulls. If no homozygous nulls are produced, then silencing the specific gene likely leads to a lethal embryonic defect. If wild-type, heterozygous nulls and homozygous nulls are born in the expected Mendelian ratio (i.e., 1:2:1), which has been the case for all MMP and TIMP knockouts, then one may conclude that the targeted gene is not essential for development. However, as is seen with Mmp14−/− and Mmp20−/− mice, prominent defects may be seen in specific tissues. Compared to other congenic strains, the 129/SJ line has marked genetic heterogeneity that may confound interpretation of phenotypes when animals are studied in a mixed background.

42

Gill et al.

In these instances, the inclusion of the right control animals is critical to minimize strain effects. Interbreeding heterozygotic null (+/−) mice should yield litters with Mendelian ratios of genotypes, that is, 25% wild-type (+/+), 50% heterozygous (+/−), and 25% homozygous null (−/−). These littermate wild-type and null offspring are the most appropriate choice for experiments to control for unwanted or unexpected strain effects. The preferred alternative to mixed-strain animals is to backcross heterozygotic null (or viable homozygous nulls) mice upward of ten generations to established congenic mouse lines such as C57Bl/6, BALB/c, and others. Various optional features can be added to enhance your knockout experience. Targeting vectors can include a reporter gene, most typically ␤-galactosidase, whose expression is driven by the target gene’s promoter. An important advantage of a reporter is that it can be used to assess where and when the target gene is being expressed. A LacZ expression cassette was inserted in Mmp14−/− line generated by Seiki and coworkers (48) and the Mmp16−/− and Mmp17−/− made by Deltagen, Inc. (available through the Mouse Genome Informatics database managed by the Jackson Laboratories; http://www.informatics.jax.org/). A truly powerful optional strategy is the ability to generate conditional knockouts, in which gene deletion is confined to specific cell types, to specific investigator-defined times, or both. Conditional nulls are made by incorporating either or both the Cre/loxP or the Flp/FRT recombination systems. With conditional knockouts, the gene exon sequence is left unaltered but the PGK-Neomycin resistance cassette is flanked by FRT sites and two loxP sites (a 34-bp palindromic sequence with two 13-bp inverted repeats separated by a central 8-bp that defines loxP orientation) are placed in intronic sequence on either side of the exon-intron run to be deleted. With this somewhat complex approach, PCR screening strategies are needed to track each stage of the knockout process. Usually, once the complete targeting construct has been incorporated into the germline to generate the conditional knockout mice, the selectable marker is removed by FLP recombinase activity introduced by breeding with a commercially available germline-specific Flp transgenic mouse. This step can also be done at the level of ES by transfection of a recombinase expression construct. Generation of the null animal is accomplished by Cre recombinase activity in vivo that can be introduced by breeding to transgenic germline-specific Cre recombinase animals or animals expressing Cre under the control of a tissue– or cell-specific promoter. Many Cre-expressing mouse lines are available (e.g., from the Jackson Laboratories, http://jaxmice.jax.org/). An advantage of removing the resistance cassette is the reduction of potential “neighborhood” effects on nearby genes caused by the active PGK promoter. Inserting an exogenous strong

MMP and TIMP Models

43

promoter, such as PGK, within a targeted locus combined with alternations to native chromatin structure, due to deletion of endogenous sequences and insertion of new DNA, can potentially lead to confounding artifacts. Targeted mutagenesis can influence the expression of neighboring genes, especially if the targeted gene is a member of a cluster of genes, such as the downregulation of Cyp2a5 in Cyp2g1 knockout animals (49), or if a locus control site (50) or other regulatory element is affected. In most reported instances of neighborhood effects, the downstream gene has attenuated expression, although it is not known whether the influence on expression is due to the presence of the PGK-neomycin resistance promoter, the removal of gene regulatory sequence, or changes in chromatin structure surrounding the mutated site. One way to minimize targeting vector promoter effects is to remove the PGK-neomycin resistance cassette after generation of targeted ES cells or in the whole animal by the recombinase strategies discussed here. Neighborhood artifacts are a real issue, and investigators should take the steps to check for and ideally correct this potential problem. Controlling when the targeted allele is mutated can also be done using inducible Cre systems. For example, Hayashi et al. used a fusion protein of Cre and a mutated form of the ligandbinding domain of the estrogen receptor (Cre-ER) to create tamoxifen-inducible Cre transgenic mice (51). The transgenics are then bred with mice containing the conditional or “floxed” knockout allele, thereby generating animals that express the CreER transgene and that are homozygous for the conditional knockout allele. Alternatively, heterozygous nulls can be bred with floxed mice to generate mice with one functional, yet floxed allele of the gene being targeted. Not only does this approach increase the efficiency of the subsequent recombination step, but it is also particularly useful if one has already generated conventional homologous nulls that turn out to be embryonic lethal or otherwise highly impaired. These mice are treated with tamoxifen to induce Cre expression and generate the null allele. Analysis of Cre recombinase efficiency can be tracked using the Rosa reporter allele R26R (52), and past studies have emphasized the need to verify Cre-mediated recombination to generate the null allele. Some recombination was detected in the absence of tamoxifen treatment, attributed to tamoxifen-independent, leaky recombination of Cre. Additionally, while most organs showed efficient tamoxifen-induced Cre activity, the liver resulted in a mosaic pattern, pointing out the need to evaluate the effectiveness of the gene activity manipulation in all organs under study. A similar strategy can be used with the tetracycline (TET) response elements, resulting in Cre under the control of this element. The TET system relies on two components, which include a tetracycline-controlled transactivator (tTA or rtTA) and

44

Gill et al.

a tTA/rtTA-dependent promoter that controls the expression of the downstream Cre in a tetracycline-dependent manner. Similar problems can potentially occur with this control system as well. Leaky Cre expression or incomplete recombination penetrance in cells of a given tissue or a whole animal must be evaluated, emphasizing the need for a reporter system like the Rosa allele for determination of recombination efficiency. Some good news is that obtaining new knockout lines is becoming easier. For example, with the initiation of the knockout mouse project and a collaboration of three major funding agencies (the NIH, the European Commission, and Genome Canada and its partners), gene targeting and trapping will be used to complete a library of mutated ES cells or mice. Distribution centers and web-based data dissemination will be established for the biomedical community and all mouse genes may be knocked out and available to researchers in the next 5 years (53). 6.1. Verification

Evidence that a given MMP can cleave a substrate, even coupled with the absence of that cleavage in a knockout animal, is suggestive but not sufficient to prove a direct MMP/substrate relationship. To be more confident of such a conclusion, various verification studies should be done. Ideally, an MMP-substrate relationship would be supported by the following observations: (1) the substrate is cleaved in vivo in wild-type but not in MMPnull mice (under the appropriate conditions); (2) in vivo, the substrate and the MMP co-localize; (3) the cleavage site (or sites) produced in vivo is identical to that produced in vitro; and (4) over-expression or add-back of the MMP increases substrate cleavage in wild-type mice or cells and restores cleavage in null models. A further control is to mutate the substrate cleavage site, rendering the protein resistant to MMP proteolysis. In addition, generating knockout or knock-in mutant mice is not the only way to ablate a specific MMP. RNA interference, dominant negative proteins, or blocking antibodies provide other approaches to inhibit the activity of a specific enzyme. Regarding colocalization, an MMP and a substrate must be in the same microenvironment during periods of proteolysis. Compartmentalization, that is, where and how in the pericellular environment an MMP is released and held, may be the most important step in regulating the specificity of proteolysis (9). Colocalization can be assessed by immunostaining, fractionation, co-immunoprecipitation, or pull-down, among other means. A caveat with immunoprecipitation/pull-down approaches is that a proteinase may only transiently interact with its substrate. However, a catalytically dead proteinase should theoretically stably bind substrate. For MMPs, activity is typically blocked by mutating the glutamate in the catalytic domain (HEXXHXXGXXH) to an alanine.

MMP and TIMP Models

45

Gain-of-function approaches include either the seemingly straightforward add-back of exogenous (ideally, activated) proteinase or a variety of over-expression strategies. Addition of exogenous proteinase to restore function has some important caveats. In vivo, the activity of an MMP is constrained by compartmentalization. Endogenous enzymes are often anchored to membrane proteins and accessory factors confining their proteolytic potential to specific substrates within specific pericellular environments (9). Thus, adding excess amounts of a truly soluble enzyme may lead to indiscriminate proteolysis of other proteins, resulting in confounding effects. Still, the ease of this approach – particularly in cell-based models – makes it an attractive option. There are several ways to over-express an MMP (or any gene), and the use of viruses is well established as a powerful tool to restore function in knockout mice. In such restoration studies, one should include an inactive form of the MMP to control for confounding results due to the infection or the expression system itself. Interestingly, some MMPs may contribute to the dissemination of viral vectors as was observed when an MMP8-expressing virus and an oncolytic virus were co-infected into a tumor, resulting in the spread of the virus throughout the tumor (54). Transgenic mice are a powerful tool for verifying and potentially identifying MMP substrates. For example, creating a mouse that conditionally expresses an MMP in specific cell types or tissues on the same MMP-null background provides a highly convincing approach for evaluating a cleavage target. It is important to remember that rescue of a substrate cleavage in either of the in vivo models described above only indicates an MMP’s role in the cleavage of the potential target, since substrate cleavage may be the result of a series of required steps. To confidently conclude a direct role for an MMP in cleaving its substrate, the sequence at the site of cleavage in a controlled system should be recapitulated in the cleavage site observed in vivo. Nevertheless, the use of only a few of the techniques described herein will often provide sufficient evidence to reliably predict an MMPsubstrate relationship.

7. Insight from Knockout Mouse Studies: MMP-TIMP Interactions

As stated, mice lacking MMP7, generated by Carole Wilson, Brigit Hogan, and Lynn Matrisian, were the first line of MMP knockout mice and were also the first used to uncover a novel physiological substrate: pro-␣-defensins, a family of antimicrobial peptides (55). Key to this discovery was the observation that

46

Gill et al.

MMP7 is co-packaged with pro-␣-defensins in the secretory granules of mouse Paneth cells and the knowledge that these factors had to be cleaved to gain activity. Since then, Mmp7−/− mice have been used to identify or at least highly implicate FasL (28), syndecan-1 (41), E-cadherin (40), RANKL (56), latent TNF-␣ (57), elastin (58), and notch-1 (59) as substrates of MMP7. From the perspective of identifying physiological substrates – via the lack of proteolysis observed in null mice – MMP7 may be the best understood member of the MMP family. MMPs are regulated at multiple levels including expression, activation, compartmentalization, inhibition, and degradation. Many describe TIMPs as the primary inhibitors of MMPs; however, much of this is based on in vitro data with little information regarding TIMP function in vivo. Despite the fact that TIMPs are effective inhibitors of MMPs in vitro (e.g., the Ki for TIMP1:MMP1 interaction is about 10−11 M), no group has – until recently (see below) – demonstrated an interaction between a TIMP and an active MMP in vivo. In fact, in one study of wound fluids, which contain both active MMP1 and TIMP1, all recovered MMP1 was complexed to ␣2 -macroglobulin (60). However, phenotypes observed in mice either over-expressing or lacking a TIMP family member do provide support – albeit indirectly – for the in vitro data, at least for TIMP1 and 3. Data in models of lung injury provided compelling evidence that a specific TIMP can block the activity of a specific MMP in vivo. Re-epithelialization and neutrophil influx are impaired in the injured lungs of Mmp7−/− mice (41), whereas these repair processes are accelerated or enhanced in Timp1−/− mice (61, 62). These opposite phenotypes suggest that MMP7 may be silenced by TIMP1, and indeed Chen et al. have recovered complexes of active MMP7 bound to TIMP1 in airway wound fluids (62). Furthermore, mice lacking TIMP1 have increased metallogelatinolytic activity in their livers and when these mice are bred into ApoE−/− mice, increased metallo-collagenolytic activity is observed in atherosclerotic plaques (63–65). In addition, transgenic mice over-expressing TIMP1 under the control of the MMP9 promoter have reduced gelatinolytic activity during cutaneous wound healing (66). In the absence of TIMP3, increased gelatinolytic activity is seen in a number of tissues, including lung and kidney, and, at least in the developing lung, this increased activity is possibly the result of increased activation and activity of MMP2 (67– 70). TIMP3-deficient mice develop a spontaneous enlargement of their alveolar space (67), and such emphysematous-like changes are often attributed to aberrant MMP activity. Indeed, Gill et al. demonstrated increased gelatinase activity in Timp3−/− lungs (68) and partially rescued the developmental lung phenotype with a synthetic MMP inhibitor (71). Similarly, Khokha and coworkers

MMP and TIMP Models

47

have reported that other phenotypes in Timp3−/− mice, such as increased apoptosis during mammary gland involution (72) and spontaneous cardiomyopathy (73), are associated with increased MMP levels and are partially reversed by global non-specific inhibition of metalloproteinase activity. Furthermore, TIMP3 seems to govern TNF-␣ levels by moderating metalloproteinase activity. TNF-␣ is primarily released from its cell-bound latent form by ADAM17/TACE (74). Constitutive levels of TNF-␣ are elevated in Timp3−/− mice, as well as the basal activity of ADAM17 (75), providing strong evidence that TIMP3 does function to govern the activity of this metalloproteinase in vivo. However, direct evidence that TIMP3 silences MMPs in vivo remains lacking. Roles for TIMP2 and TIMP4 in regulating MMP activity, however, are not so clear. In cell-based and defined in vitro models, activation of pro-MMP2 is mediated by the activity of MMP14 and requires the presence of the right amount of TIMP2 (76, 77). Although the role of TIMP2 in pro-MMP2 activation is supported by a lack of zymogen activation in TIMP2 null mice (21, 22, 78), a similar defect in MMP14 knockout mice has not been reported and activation of pro-MMP2 in fibroblasts is not markedly, if at all, affected by deficiency of MMP14 (79). The role of MMP14 in activation of MMP2 is further questioned by a lack of phenocopy between null mouse lines. Whereas MMP2 null mice reveal mild phenotypes, mostly related to neovascularization and inflammation (80–84), MMP14 knockout mice have severe defects in skeletal development and turnover of type I collagen (85–88). Overall, the in vivo data indicate that MMP14 is not required for activation of MMP2 but that TIMP2 is.

8. Summary The evaluation of MMP and TIMP function in vivo is essential to understanding the role of these complex enzyme families. Knockout technology provides enormous contributions to biology in general and has notably furthered the understanding of how MMPs and TIMPs function in vivo. The existing mouse models are an important first step in discovering how these proteins relate to physiology and pathophysiology. Future work must be focused on the generation of conditional knockout and inducible transgenic models. Additionally, future experiments using these models to identify MMP substrates or TIMP interactions must include the controls discussed above to confidently conclude novel MMP and TIMP biology.

48

Gill et al.

References 1. Lopez-Otin, C. and Overall, C. M. (2002) Protease degradomics: a new challenge for proteomics. Nat Rev Mol Cell Biol 3, 509–519. 2. Mott, J. D. and Werb, Z. (2004) Regulation of matrix biology by matrix metalloproteinases. Curr Opin Cell Biol 16, 558–564. 3. Parks, W. C., Wilson, C. L., and LopezBoado, Y. S. (2004) Matrix metalloproteinases as modulators of inflammation and innate immunity. Nat Rev Immunol 4, 617–629. 4. Agrawal, S. M., Lau, L., and Yong, V. W. (2008) MMPs in the central nervous system: where the good guys go bad. Semin Cell Dev Biol 19, 42–51. 5. Manicone, A. M. and McGuire, J. K. (2008) Matrix metalloproteinases as modulators of inflammation. Semin Cell Dev Biol 19, 34–41. 6. Page-McCaw, A., Ewald, A. J., and Werb, Z. (2007) Matrix metalloproteinases and the regulation of tissue remodeling. Nat Rev Mol Cell Biol 8, 221-233. 7. Egeblad, M. and Werb, Z. (2002) New functions for the matrix metalloproteinases in cancer progression. Nat Rev Cancer 2, 161–174. 8. Van Lint, P. and Libert, C. (2007) Chemokine and cytokine processing by matrix metalloproteinases and its effect on leukocyte migration and inflammation. J Leukoc Biol 82, 1375–1381. 9. Ra, H. J. and Parks, W. C. (2007) Control of matrix metalloproteinase catalytic activity. Matrix Biol 26, 587–596. 10. Massova, I., Kotra, L. P., Fridman, R., and Mobashery, S. (1998) Matrix metalloproteinases: structures, evolution, and diversification. FASEB J 12, 1075–1095. 11. Nagase, H., Visse, R., and Murphy, G. (2006) Structure and function of matrix metalloproteinases and TIMPs. Cardiovasc Res 69, 562–573. 12. Maskos, K. and Bode, W. (2003) Structural basis of matrix metalloproteinases and tissue inhibitors of metalloproteinases. Mol Biotechnol 25, 241–266. 13. Gill, S. E. and Parks, W. C. (2008) Metalloproteinases and their inhibitors: Regulators of wound healing. Int J Biochem Cell Biol 40, 1334–1347. 14. Campbell, E. J., Campbell, M. A., Boukedes, S. S., and Owen, C. A. (1999) Quantum proteolysis by neutrophils: implications for pulmonary emphysema in alpha 1-antitrypsin deficiency. J Clin Invest 104, 337–344.

15. Baker, A. H., Edwards, D. R., and Murphy, G. (2002) Metalloproteinase inhibitors: biological actions and therapeutic opportunities. J Cell Sci 115, 3719–3727. 16. Leco, K. J., Apte, S. S., Taniguchi, G. T., Hawkes, S. P., Khokha, R., Schultz, G. A., and Edwards, D. R. (1997) Murine tissue inhibitor of metalloproteinases-4 (Timp-4): cDNA isolation and expression in adult mouse tissues. FEBS Lett 401, 213–217. 17. Oh, J., Takahashi, R., Kondo, S., Mizoguchi, A., Adachi, E., Sasahara, R. M., Nishimura, S., Imamura, Y., Kitayama, H., Alexander, D. B., Ide, C., Horan, T. P., Arakawa, T., Yoshida, H., Nishikawa, S., Itoh, Y., Seiki, M., Itohara, S., Takahashi, C., and Noda, M. (2001) The membrane-anchored MMP inhibitor RECK is a key regulator of extracellular matrix integrity and angiogenesis. Cell 107, 789–800. 18. Kassim, S. Y., Fu, X., Liles, W. C., Shapiro, S. D., Parks, W. C., and Heinecke, J. W. (2005) NADPH oxidase restrains the matrix metalloproteinase activity of macrophages. J Biol Chem 280, 30201–30205. 19. Brew, K., Dinakarpandian, D., and Nagase, H. (2000) Tissue inhibitors of metalloproteinases: evolution, structure and function. Biochim Biophys Acta 1477, 267–283. 20. Yu, W. H., Yu, S., Meng, Q., Brew, K., and Woessner, J. F., Jr. (2000) TIMP-3 binds to sulfated glycosaminoglycans of the extracellular matrix. J Biol Chem 275, 31226–31232. 21. Caterina, J. J., Yamada, S., Caterina, N. C., Longenecker, G., Holmback, K., Shi, J., Yermovsky, A. E., Engler, J. A., and Birkedal-Hansen, H. (2000) Inactivating mutation of the mouse tissue inhibitor of metalloproteinases-2(Timp-2) gene alters proMMP-2 activation. J Biol Chem 275, 26416–26422. 22. Wang, Z., Juttermann, R., and Soloway, P. D. (2000) TIMP-2 is required for efficient activation of proMMP-2 in vivo. J Biol Chem 275, 26411–26415. 23. Seo, D. W., Li, H., Guedez, L., Wingfield, P. T., Diaz, T., Salloum, R., Wei, B. Y., and Stetler-Stevenson, W. G. (2003) TIMP2 mediated inhibition of angiogenesis: an MMP-independent mechanism. Cell 114, 171–180. 24. Qi, J. H., Ebrahem, Q., Moore, N., Murphy, G., Claesson-Welsh, L., Bond, M., Baker, A., and Anand-Apte, B. (2003) A novel function for tissue inhibitor of metalloproteinases-3 (TIMP3): inhibition of

MMP and TIMP Models

25.

26. 27.

28.

29.

30.

31.

32.

33.

34.

35.

angiogenesis by blockage of VEGF binding to VEGF receptor-2. Nat Med 9, 407–415. McQuibban, G. A., Gong, J. H., Tam, E. M., McCulloch, C. A., Clark-Lewis, I., and Overall, C. M. (2000) Inflammation dampened by gelatinase A cleavage of monocyte chemoattractant protein-3. Science 289, 1202–1206. auf dem Keller, U., Doucet, A., and Overall, C. M. (2007) Protease research in the era of systems biology. Biol Chem 388, 1159–1162. Holmbeck, K., Bianco, P., Caterina, J., Yamada, S., Kromer, M., Kuznetsov, S. A., Mankani, M., Robey, P. G., Poole, A. R., Pidoux, I., Ward, J. M., and BirkedalHansen, H. (1999) MT1-MMP-deficient mice develop dwarfism, osteopenia, arthritis, and connective tissue disease due to inadequate collagen turnover. Cell 99, 81–92. Powell, W. C., Fingleton, B., Wilson, C. L., Boothby, M., and Matrisian, L. M. (1999) The metalloproteinase matrilysin proteolytically generates active soluble Fas ligand and potentiates epithelial cell apoptosis. Curr Biol 9, 1441–1447. McCawley, L. J. and Matrisian, L. M. (2001) Matrix metalloproteinases: they’re not just for matrix anymore! Curr Opin Cell Biol 13, 534–540. Folgueras, A. R., Pendas, A. M., Sanchez, L. M., and Lopez-Otin, C. (2004) Matrix metalloproteinases in cancer: from new functions to improved inhibition strategies. Int J Dev Biol 48, 411–424. Janssens, S. and Lijnen, H. R. (2006) What has been learned about the cardiovascular effects of matrix metalloproteinases from mouse models? Cardiovasc Res 69, 585–594. Hu, J., Van den Steen, P. E., Sang, Q. X., and Opdenakker, G. (2007) Matrix metalloproteinase inhibitors as therapy for inflammatory and vascular diseases. Nat Rev Drug Discov 6, 480–498. Greenlee, K. J., Werb, Z., and Kheradmand, F. (2007) Matrix metalloproteinases in lung: multiple, multifarious, and multifaceted. Physiol Rev 87, 69–98. Zeisberg, M., Khurana, M., Rao, V. H., Cosgrove, D., Rougier, J. P., Werner, M. C., Shield, C. F., 3rd, Werb, Z., and Kalluri, R. (2006) Stage-specific action of matrix metalloproteinases influences progressive hereditary kidney disease. PLoS Med 3, e100. Rudolph-Owen, L. A., Hulboy, D. L., Wilson, C. L., Mudgett, J., and Matrisian, L. M. (1997) Coordinate expression of matrix metalloproteinase family members in the uterus of normal, matrilysin-deficient, and

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

49

stromelysin-1-deficient mice. Endocrinology 138, 4902–4911. Sahin, U., Weskamp, G., Kelly, K., Zhou, H. M., Higashiyama, S., Peschon, J., Hartmann, D., Saftig, P., and Blobel, C. P. (2004) Distinct roles for ADAM10 and ADAM17 in ectodomain shedding of six EGFR ligands. J Cell Biol 164, 769–779. Coussens, L. M., Fingleton, B., and Matrisian, L. M. (2002) Matrix metalloproteinase inhibitors and cancer: trials and tribulations. Science 295, 2387–2392. Wilson, C. L., Heppner, K. J., Labosky, P. A., Hogan, B. L. M., and Matrisian, L. M. (1997) Intestinal tumorigenesis is suppressed in mice lacking the metalloproteinase matrilysin. Proc Natl Acad Sci U S A 94, 1402–1407. Dunsmore, S. E., Saarialho-Kere, U. K., Roby, J. D., Wilson, C. L., Matrisian, L. M., Welgus, H. G., and Parks, W. C. (1998) Matrilysin expression and function in airway epithelium. J Clin Invest 102, 1321–1331. McGuire, J. K., Li, Q., and Parks, W. C. (2003) Matrilysin (matrix metalloproteinase-7) mediates E-cadherin ectodomain shedding in injured lung epithelium. Am J Pathol 162, 1831–1843. Li, Q., Park, P. W., Wilson, C. L., and Parks, W. C. (2002) Matrilysin shedding of syndecan-1 regulates chemokine mobilization and transepithelial efflux of neutrophils in acute lung injury. Cell 111, 635–646. Swee, M., Wilson, C. L., Wang, Y., McGuire, J. K., and Parks, W. C. (2008) Matrilysin (MMP7) controls neutrophil egress and activation through generation of chemokine gradients. J Leuko Biol 83, 1402–1412. Coussens, L. M., Tinkle, C. L., Hanahan, D., and Werb, Z. (2000) MMP-9 supplied by bone marrow-derived cells contributes to skin carcinogenesis. Cell 103, 481–490. Hotary, K., Allen, E., Punturieri, A., Yana, I., and Weiss, S. J. (2000) Regulation of cell invasion and morphogenesis in a three-dimensional type I collagen matrix by membrane-type matrix metalloproteinases 1, 2, and 3. J Cell Biol 149, 1309–1323. Hotary, K. B., Allen, E. D., Brooks, P. C., Datta, N. S., Long, M. W., and Weiss, S. J. (2003) Membrane type I matrix metalloproteinase usurps tumor growth control imposed by the three-dimensional extracellular matrix. Cell 114, 33–45. McCawley, L. J., Crawford, H. C., King, L. E., Jr., Mudgett, J., and Matrisian, L. M. (2004) A protective role for matrix

50

47.

48.

49.

50.

51.

52. 53. 54.

55.

56.

Gill et al. metalloproteinase-3 in squamous cell carcinoma. Cancer Res 64, 6965–6972. Balbin, M., Fueyo, A., Tester, A. M., Pendas, A. M., Pitiot, A. S., Astudillo, A., Overall, C. M., Shapiro, S. D., and Lopez-Otin, C. (2003) Loss of collagenase-2 confers increased skin tumor susceptibility to male mice. Nat Genet 35, 252–257. Yana, I., Sagara, H., Takaki, S., Takatsu, K., Nakamura, K., Nakao, K., Katsuki, M., Taniguchi, S., Aoki, T., Sato, H., Weiss, S. J., and Seiki, M. (2007) Crosstalk between neovessels and mural cells directs the site-specific expression of MT1-MMP to endothelial tip cells. J Cell Sci 120, 1607–1614. Zhou, H. E., Zhang, X., and Nothnick, W. B. (2004) Disruption of the TIMP-1 gene product is associated with accelerated endometrial gland formation during early postnatal uterine development. Biol Reprod 71, 534–539. Olson, E. N., Arnold, H.-H., Rigby, P. W. J., and Wold, B. J. (1996) Know your neighbors. Three phenotypes in null mutants of the myogenic bHLH gene MRF4. Cell 85, 1–4. Hayashi, S., Lewis, P., Pevny, L., and McMahon, A. P. (2002) Efficient gene modulation in mouse epiblast using a Sox2Cre transgenic mouse strain. Mech Dev 119(Suppl 1), S97–S101. Soriano, P. (1999) Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat Genet 21, 70–71. Collins, F. S., Rossant, J., and Wurst, W. (2007) A mouse for all reasons. Cell 128, 9–13. Cheng, J., Sauthoff, H., Huang, Y., Kutler, D. I., Bajwa, S., Rom, W. N., and Hay, J. G. (2007) Human matrix metalloproteinase8 gene delivery increases the oncolytic activity of a replicating adenovirus. Mol Ther 15, 1982–1990. Wilson, C. L., Ouellette, A. J., Satchell, D. P., Ayabe, T., Lopez-Boado, Y. S., Stratman, J. L., Hultgren, S. J., Matrisian, L. M., and Parks, W. C. (1999) Regulation of intestinal alpha-defensin activation by the metalloproteinase matrilysin in innate host defense. Science 286, 113–117. Lynch, C. C., Hikosaka, A., Acuff, H. B., Martin, M. D., Kawai, N., Singh, R. K., Vargo-Gogola, T. C., Begtrup, J. L., Peterson, T. E., Fingleton, B., Shirai, T., Matrisian, L. M., and Futakuchi, M. (2005) MMP-7 promotes prostate cancer-induced osteolysis via the solubilization of RANKL. Cancer Cell 7, 485–496.

57. Haro, H., Crawford, H. C., Fingleton, B., Shinomiya, K., Spengler, D. M., and Matrisian, L. M. (2000) Matrix metalloproteinase-7-dependent release of tumor necrosis factor-alpha in a model of herniated disc resorption. J Clin Invest 105, 143–150. 58. Filippov, S., Caras, I., Murray, R., Matrisian, L. M., Chapman, H. A., Jr., Shapiro, S., and Weiss, S. J. (2003) Matrilysin-dependent elastolysis by human macrophages. J Exp Med 198, 925–935. 59. Sawey, E. T., Johnson, J. A., and Crawford, H. C. (2007) Matrix metalloproteinase 7 controls pancreatic acinar cell transdifferentiation by activating the Notch signaling pathway. Proc Natl Acad Sci U S A 104, 19327–19332. 60. Grinnell, F., Zhu, M., and Parks, W. C. (1997) Collagenase-1 complexes with a2macroglobulin in the acute and chronic wound environments. J Invest Dermatol 110, 771-776. 61. Chen, P., Farivar, A. S., Mulligan, M. S., and Madtes, D. K. (2006) Tissue inhibitor of metalloproteinase-1 deficiency abrogates obliterative airway disease after heterotopic tracheal transplantation. Am J Respir Cell Mol Biol 34, 464–472. 62. Chen, P., McGuire, J., Kim, K.-H., Hackman, R., Chen, A., Parks, W., and Madtes, D. (2008) Tissue inhibitor of metalloproteinase1 moderates airway re-epithelialization by regulating matrilysin activity. Am J Pathol 172, 1256–1270. 63. Mohammed, F. F., Pennington, C. J., Kassiri, Z., Rubin, J. S., Soloway, P. D., Ruther, U., Edwards, D. R., and Khokha, R. (2005) Metalloproteinase inhibitor TIMP-1 affects hepatocyte cell cycle via HGF activation in murine liver regeneration. Hepatology 41, 857–867. 64. Lijnen, H. R., Van Hoef, B., Vanlinthout, I., Verstreken, M., Rio, M. C., and Collen, D. (1999) Accelerated neointima formation after vascular injury in mice with stromelysin3 (MMP-11) gene inactivation. Arterioscler Thromb Vasc Biol 19, 2863–2870. 65. Silence, J., Collen, D., and Lijnen, H. R. (2002) Reduced atherosclerotic plaque but enhanced aneurysm formation in mice with inactivation of the tissue inhibitor of metalloproteinase-1 (TIMP-1) gene. Circ Res 90, 897–903. 66. Salonurmi, T., Parikka, M., Kontusaari, S., Pirila, E., Munaut, C., Salo, T., and Tryggvason, K. (2004) Overexpression of TIMP1 under the MMP-9 promoter interferes with wound healing in transgenic mice. Cell Tissue Res 315, 27–37.

MMP and TIMP Models 67. Leco, K. J., Waterhouse, P., Sanchez, O. H., Gowing, K. L., Poole, A. R., Wakeham, A., Mak, T. W., and Khokha, R. (2001) Spontaneous air space enlargement in the lungs of mice lacking tissue inhibitor of metalloproteinases-3 (TIMP-3). J Clin Invest 108, 817–829. 68. Gill, S. E., Pape, M. C., Khokha, R., Watson, A. J., and Leco, K. J. (2003) A null mutation for tissue inhibitor of metalloproteinases3 (Timp-3) impairs murine bronchiole branching morphogenesis. Dev Biol 261, 313–323. 69. Martin, E. L., Moyer, B. Z., Pape, M. C., Starcher, B., Leco, K. J., and Veldhuizen, R. A. (2003) Negative impact of tissue inhibitor of metalloproteinase-3 null mutation on lung structure and function in response to sepsis. Am J Physiol Lung Cell Mol Physiol 285, L1222–L1232. 70. Kawamoto, H., Yasuda, O., Suzuki, T., Ozaki, T., Yotsui, T., Higuchi, M., Rakugi, H., Fukuo, K., Ogihara, T., and Maeda, N. (2006) Tissue inhibitor of metalloproteinase3 plays important roles in the kidney following unilateral ureteral obstruction. Hypertens Res 29, 285–294. 71. Gill, S. E., Pape, M. C., and Leco, K. J. (2006) Tissue inhibitor of metalloproteinases 3 regulates extracellular matrix – cell signaling during bronchiole branching morphogenesis. Dev Biol 298, 540–554. 72. Fata, J. E., Leco, K. J., Voura, E. B., Yu, H. Y., Waterhouse, P., Murphy, G., Moorehead, R. A., and Khokha, R. (2001) Accelerated apoptosis in the Timp-3-deficient mammary gland. J Clin Invest 108, 831–841. 73. Fedak, P. W., Smookler, D. S., Kassiri, Z., Ohno, N., Leco, K. J., Verma, S., Mickle, D. A., Watson, K. L., Hojilla, C. V., Cruz, W., Weisel, R. D., Li, R. K., and Khokha, R. (2004) TIMP-3 deficiency leads to dilated cardiomyopathy. Circulation 110, 2401–2409. 74. Mohan, M. J., Seaton, T., Mitchell, J., Howe, A., Blackburn, K., Burkhart, W., Moyer, M., Patel, I., Waitt, G. M., Becherer, J. D., Moss, M. L., and Milla, M. E. (2002) The tumor necrosis factor-alpha converting enzyme (TACE): a unique metalloproteinase with highly defined substrate selectivity. Biochemistry 41, 9462–9469. 75. Mohammed, F. F., Smookler, D. S., Taylor, S. E., Fingleton, B., Kassiri, Z., Sanchez, O. H., English, J. L., Matrisian, L. M., Au, B., Yeh, W. C., and Khokha, R. (2004) Abnormal TNF activity in Timp3−/− mice leads to chronic hepatic inflammation and failure of liver regeneration. Nat Genet 36, 969–977.

51

76. English, J. L., Kassiri, Z., Koskivirta, I., Atkinson, S. J., Di Grappa, M., Soloway, P. D., Nagase, H., Vuorio, E., Murphy, G., and Khokha, R. (2006) Individual Timp deficiencies differentially impact pro-MMP-2 activation. J Biol Chem 281, 10337–10346. 77. Murphy, G., Knauper, V., Cowell, S., Hembry, R., Stanton, H., Butler, G., Freije, J., Pendas, A. M., and Lopez-Otin, C. (1999) Evaluation of some newer matrix metalloproteinases. Ann N Y Acad Sci 878, 25–39. 78. Aoki, T., Kataoka, H., Moriwaki, T., Nozaki, K., and Hashimoto, N. (2007) Role of TIMP-1 and TIMP-2 in the progression of cerebral aneurysms. Stroke 38, 2337–2345. 79. Ruangpanit, N., Price, J. T., Holmbeck, K., Birkedal-Hansen, H., Guenzler, V., Huang, X., Chan, D., Bateman, J. F., and Thompson, E. W. (2002) MT1-MMP-dependent and independent regulation of gelatinase a activation in long-term, ascorbate-treated fibroblast cultures: regulation by fibrillar collagen. Exp Cell Res 272, 109–118. 80. Corry, D. B., Rishi, K., Kanellis, J., Kiss, A., Song, L. Z., Xu, J., Feng, L., Werb, Z., and Kheradmand, F. (2002) Decreased allergic lung inflammatory cell egression and increased susceptibility to asphyxiation in MMP2-deficiency. Nat Immunol 3, 347–353. 81. Itoh, T., Matsuda, H., Tanioka, M., Kuwabara, K., Itohara, S., and Suzuki, R. (2002) The role of matrix metalloproteinase2 and matrix metalloproteinase-9 in antibody-induced arthritis. J Immunol 169, 2643–2647. 82. Kato, T., Kure, T., Chang, J. H., Gabison, E. E., Itoh, T., Itohara, S., and Azar, D. T. (2001) Diminished corneal angiogenesis in gelatinase A-deficient mice. FEBS Lett 508, 187–190. 83. Ohno-Matsui, K., Uetama, T., Yoshida, T., Hayano, M., Itoh, T., Morita, I., and Mochizuki, M. (2003) Reduced retinal angiogenesis in MMP-2-deficient mice. Invest Ophthalmol Vis Sci 44, 5370–5375. 84. Berglin, L., Sarman, S., van der Ploeg, I., Steen, B., Ming, Y., Itohara, S., Seregard, S., and Kvanta, A. (2003) Reduced choroidal neovascular membrane formation in matrix metalloproteinase-2-deficient mice. Invest Ophthalmol Vis Sci 44, 403–408. 85. Holmbeck, K., Bianco, P., Caterina, J., Yamada, S., Kromer, M., Kuznetsov, S. A., Mankani, M., Robey, P. G., Poole, A. R., Pidoux, I., Ward, J. M., and BirkedalHansen, H. (1999) MT1-MMP-deficient

52

Gill et al.

mice develop dwarfism, osteopenia, arthritis, and connective tissue disease due to inadequate collagen turnover. Cell 99, 81–92. 86. Holmbeck, K., Bianco, P., Chrysovergis, K., Yamada, S., and Birkedal-Hansen, H. (2003) MT1-MMP-dependent, apoptotic remodeling of unmineralized cartilage: a critical process in skeletal growth. J Cell Biol 163, 661–671. 87. Holmbeck, K., Bianco, P., Pidoux, I., Inoue, S., Billinghurst, R. C., Wu, W., Chrysovergis, K., Yamada, S., Birkedal-Hansen, H., and Poole, A. R. (2005) The metalloproteinase MT1-MMP is required for normal development and maintenance of osteocyte processes in bone. J Cell Sci 118, 147–156. 88. Zhou, Z., Apte, S. S., Soininen, R., Cao, R., Baaklini, G. Y., Rauser, R. W., Wang, J.,

Cao, Y., and Tryggvason, K. (2000) Impaired endochondral ossification and angiogenesis in mice deficient in membrane-type matrix metalloproteinase I. Proc Natl Acad Sci U S A 97, 4052–4057. 89. Vaisar, T., Kassim, S. Y., Gomez, I. G., Green, P. S., Hargarten, S., Gough, P. J., Parks, W. C., Wilson, C. L., Raines, E. W., and Heinecke, J. W. 2009 MMP9 sheds the ␤2 integrin subunit (CD18) from macrophages. Mol Cell Proteomics 8, 1044–1060. 90. Manicone, A. M., Birkland, T. P., Yang, Y., Betsuyaku, T., Lohi, J., Skerrett, S. J., and Parks, W. C. 2009 Epilysin (MMP-28) restrains early macrophage recruitment in Pseudomonas aeruginosa pneumonia. J Immunol 182, 3866–3876.

Section II Expression and Purification of MMPs and TIMPs

Chapter 3 Expression of Recombinant MMP-28 in Mammalian Cells Ursula R. Rodgers and Ian M. Clark Abstract The expression of a recombinant MMP in a mammalian cell line can be useful, e.g., for purification of the enzyme, to characterize function of the enzyme, or to uncover its substrates. In this chapter, we have therefore documented our experience with the recently discovered MMP-28. Key words: Recombinant, transfection, transient expression, stable expression, antibiotic selection, cell line.

1. Introduction Matrix metalloproteinases (MMPs) are often expressed by cells at low levels. In order to assay function and activity, or purify recombinant enzyme, forced overexpression in an appropriate mammalian cell line is often necessary. This chapter will, therefore, document our experience in overexpressing MMP-28 in a chondrosarcoma cell line, including subcloning the gene into an appropriate expression vector, producing an inactive mutant, transfecting, and selecting stable cell lines, analyzing expression. Matrix metalloproteinase-28 (MMP-28, epilysin) is a recently cloned member of the matrix metalloproteinase family (1, 2). It is highly expressed in the skin by keratinocytes, the developing and regenerating nervous system, and a number of other normal human tissues (1–4). In epithelial cells, overexpression of MMP-28 mediates irreversible epithelial to mesenchymal transition concomitant with loss of E-cadherin from the cell surface I.M. Clark (ed.), D.A. Young, A.D. Rowan (Consulting editors), Matrix Metalloproteinase Protocols, Methods in Molecular Biology 622, DOI 10.1007/978-1-60327-299-5 3, © Springer Science+Business Media, LLC 2001, 2010

55

56

Rodgers and Clark

and an increase in active transforming growth factor beta (5). We recently reported the expression of MMP-28 in both cartilage and synovium where expression is increased in patients with osteoarthritis (6, 7). Normal primary human articular chondrocytes or human chondrosarcoma cells express undetectable levels of MMP-28 and hence we sought to overexpress the protein for functional analyses.

2. Materials 2.1. Subcloning cDNA

1. Oligonucleotide primers (Sigma-Genosys). 2. 25 mM dNTP mix (Bioline). 3. AccuTaq LA polymerase (Sigma-Aldrich). 4. cDNA from cells expressing MMP-28 (see Note 1). 5. Phenol:chloroform:isoamyl alcohol (25:24:1 v/v) saturated with 10 mM Tris-HCl, pH 8.0. 6. 3 M sodium acetate. 7. Ethanol, 100% (v/v) and 70% (v/v). 8. 10 mM Tris-HCl, pH 8.0. 9. Restriction enzymes (Roche). 10. Mammalian expression vector (e.g., pcDNA4/His) (see Note 2). 11. T4 DNA ligase. 12. QIAquick gel extraction kit (QIAGEN).

2.2. Transfection of Cells

1. SW1353 cells. 2. Tissue culture medium: Dulbecco’s modified Eagle’s medium (DMEM) containing 10% fetal calf serum, 100 IU/mL penicillin, and 100 ␮g/mL streptomycin. 3. FuGENE HD (Roche) transfection reagent. 4. Phosphate-buffered saline (PBS). 5. Zeocin: Dissolve at 20 mg/mL in PBS and filter sterilize. Dilute 100–fold into culture medium to achieve a 200 ␮g/mL final concentration.

2.3. Western Blot

1. Trichloroacetic acid, 10% (w/v) 2. Acetone 3. SDS-PAGE sample buffer: 0.058 M Tris-HCl, pH 6.8; 5% (v/v) glycerol; 1.7% (w/v) SDS, 0.002% (w/v) bromophenol blue, 50 mM DTT, and 6 M urea. Store at –20◦ C.

Expression of Recombinant MMP-28 in Mammalian Cells

57

4. 0.5 mM EDTA. Prepare a 10 mM stock of EDTA, pH 8.0, then dilute in 1 × PBS. 5. 10 × Tris-buffered saline (TBS): 24.2 g Tris and 80 g NaCl, pH 7.6, in a final volume of 1 L. 6. Tris-buffered saline with Tween (TBS/T): 1 × TBS with 0.1% (v/v) Tween 20. 7. Blocking buffer: 5% (w/v) non-fat milk powder in TBS/T. R M2 IgG monoclonal anti8. Primary antibody: anti-FLAG body (Sigma-Aldrich).

9. Secondary antibody: anti-mouse Igs conjugated to horse radish peroxidase (Dako). 2.4. qRT-PCR

1. TRIzol (Invitrogen) 2. Chloroform 3. Isopropanol, 100% (v/v) 4. Random hexamers, 3 ␮g/␮L (Invitrogen) 5. Superscript II reverse transcriptase (Invitrogen) 6. RNasin (40,000 units/mL, Promega) 7. TaqMan primers and probe (Applied Biosystems) 8. TaqMan master mix (Applied Biosystems)

2.5. Immunocytochemistry

1. Paraformaldehyde, 4% (w/v) in 1 × phosphate-buffered saline. 2. Triton X-100, 0.1% (v/v) in 1 × phosphate-buffered saline. 3. Blocking buffer: 3% (w/v) bovine serum albumin in phosphate-buffered saline. 4. Wash buffer: 1 × PBS. R M2 IgG monoclonal anti5. Primary antibody: anti-FLAG body (Sigma-Aldrich).

6. Secondary antibody: Alexa Fluor 488 goat anti-mouse IgG1 (␥1 ) (Invitrogen).

3. Methods 3.1. Cloning the MMP-28 cDNA

1. Design oligonucleotide primers to amplify the MMP-28 cDNA. These should include restriction sites for subcloning into the expression vector of choice (underlined below, BamHI and XhoI for forward and reverse, respectively), additional three or four base extensions for efficient restriction digestion of PCR product

58

Rodgers and Clark

(italics below) and the forward primer should contain a Kozak sequence for efficient translation in mammalian cells (bold below). Hence, our forward primer was 5 CGCGGATCCGCCGCCGCCATGGTCGCGCGCGTCGGCCTC-3 and the reverse primer was ACGTCTCGAG CCGAACAGGGCGCTCCCCGAGTTG-3 . 2. Perform PCR using 100 ng cDNA template (see Note 1), 0.4 ␮M each primer, 500 ␮M dNTPs, and 0.5 units of AccuTaqTM LA polymerase (Sigma-Aldrich) in 50 ␮L of the manufacturer’s buffer. Cycling conditions should be optimized for each primer pair (see Note 3), though we used the following program successfully: 98◦ C for 30 s; 20–30 cycles of 94◦ C for 20 s, 55◦ C for 20 s, 68◦ C for 1 min; 68◦ C for 10 min. 3. Run 5 ␮L PCR product on a 1% agarose gel. If there is a PCR product of the correct size, run the remaining product on a 1% agarose gel using high-grade agarose (see Notes 4 and 5). Cut out the band of the expected size with a clean scalpel (see Note 6) and extract using the QIAquick gel extraction kit according to the manufacturer’s instructions. For the final step elute into 30 ␮L nuclease-free water. Quantify the purified DNA using absorbance at 260 nm (see Note 7). 4. Digest 10 ␮g PCR product with appropriate enzymes (chosen in Step 1) in the manufacturer’s recommended buffer. We recommend using at least 50 units of enzyme for 2 h at 37◦ C. 5. Clean up the digested PCR product by extracting in an equal volume of phenol:chloroform:isoamyl alcohol (25:24:1) saturated with 10 mM Tris-HCl, pH 8.0. Vortex the sample for 30 s, followed by centrifugation at 13,000 × g for 5 min. Re-extract the aqueous phase with an equal volume of chloroform. 6. Precipitate the DNA by adding 0.1 volumes of 3 M sodium acetate and 3 volumes 100% (v/v) ethanol and incubating at –80◦ C overnight. 7. Pellet the DNA by centrifugation at >13,000 × g for 30 min at 4◦ C. Remove supernate and wash with 0.5 mL of ice-cold 70% (v/v) ethanol. Spin again at >13,000 × g for 5 min at 4◦ C. 8. Remove supernate and air-dry pellet for 15 min at room temperature. 9. Resuspend pellet in 20 ␮L 10 mM Tris-HCl, pH 8.0. Read absorbance at 260 nm and calculate concentration (see Note 6).

Expression of Recombinant MMP-28 in Mammalian Cells

59

10. Digest 10 ␮g expression vector as in Step 4. 11. Run the products on a 0.8% agarose gel (high quality, see Note 4) and cut out and purify the linear plasmid (see Step 3 above). 12. Ligate the PCR product into the vector using approximately 50 ng of cut vector and a 1:1 and 1:3 molar ratio of vector to insert. Use 1 unit of T4 DNA ligase in a final volume of 15 ␮L and ligate overnight at 16◦ C. Run vector-only controls both without and with ligase to ascertain amount of circular plasmid remaining (from the former) and amount of single-cut plasmid (from the latter) (see Note 8). 13. Transform 5 ␮L ligation products into 100 ␮L competent Escherichia coli. 14. Pick colonies for mini-prep and sequence inserts to verify. 3.2. Creating an Inactive Mutant

We use extension overlap PCR to mutate the active site glutamate (E) residue to alanine (A), creating a catalytically inactive protease (see Note 9) 1. Use the wild-type MMP-28 cDNA sub-cloned above as a template for mutagenesis. 2. Perform two separate PCR reactions, using 500 ng cDNA per reaction. The first reaction contains the forward primer in Section 3.1, Step 1 and an internal reverse primer across the region to be mutated (5 -CAAGCGTGTGACCGATCGCGTGCGCCAGCAC-3 , mutation in bold). The second reaction contains the reverse primer in Section 3.1, Step 1 and an internal forward primer containing the desired mutation (5 -GTGCTGGCGCAC GCGATCGGTCACACGCTTG-3 , mutation in bold). 3. Perform a third PCR reaction using the two PCR products from Step 2 as template (approximately 1 ␮L from the PCR reactions in Step 2 above) and the original forward and reverse primers in Section 3.1, Step 1. Increase the extension step of the PCR from 1 to 1.5 min. 4. Follow Section 3.1, Steps 3 through 13 to ligate this product into the vector and verify the mutation by sequencing.

3.3. Transient Transfection of Chondrosarcoma Cells

1. Culture SW1353 cells in DMEM containing 10% FCS, 100 IU/mL penicillin, and 100 ␮g/mL streptomycin. 2. Plate cells at 100,000 cells/well in a 6-well plate. Incubate at 37◦ C, 5% CO2 . 3. After 24 h transfect 1 ␮g of expression plasmid (from Section 3.1 or 3.2 above) into cells using FuGENE HD (see Note 10). Per well, mix 3 ␮L FuGENE HD with

60

Rodgers and Clark

97 ␮L serum-free medium and incubate for 5 min at room temperature. Add 1 ␮g plasmid DNA and incubate for a further 15 min. Mix gently and add 100 ␮L dropwise per well. Incubate at 37◦ C, 5% CO2 . 4. After 24 h, wash cells ×2 in PBS and add 3 mL per well of serum-free medium. Incubate at 37◦ C, 5% CO2 . 5. After 48 h analyze cell fractions by Western blot (see Section 3.5.1). 3.4. Stable Transfection of a Chondrosarcoma Cell Line

1. Culture SW1353 cells in DMEM containing 10% FCS, 100 IU/mL penicillin, and 100 ␮g/mL streptomycin. 2. Plate cells at 100,000 cells/well in a 6-well plate. Incubate at 37◦ C, 5% CO2 for 24 h. 3. Linearize the expression plasmid to be transfected using a restriction enzyme which cuts the vector backbone at a site distinct from the antibiotic resistance gene that will be used to select the transfected cells and which does not cut the cDNA to be expressed (in our case BglII). Clean up the linearized expression plasmid according to Section 3.1, Steps 5–9. 4. Transfect 1 ␮g of linearized expression plasmid (from Section 3.1 or 3.2) into cells using FuGENE HD (see Note 10). Per well, mix 3 ␮L FuGENE HD with 97 ␮L serum-free medium and incubate for 5 min at room temperature. Add 1 ␮g plasmid DNA and incubate for a further 15 min. Mix gently and add 100 ␮L dropwise per well. Incubate at 37◦ C, 5% CO2 . 5. After 48 h, switch cells into culture medium containing 200 ␮g/mL zeocin or other appropriate selection antibiotic (see Note 11). Incubate at 37◦ C, 5% CO2 . Change the medium after 3–4 days. 6. After 7 days (see Note 12), trypsinize surviving cells and plate into 96-well plates at dilutions of 3, 1, and 0.3 cells per well (i.e., make up dilutions of 30, 10, and 3 cells per mL and add 100 ␮L per well) maintaining zeocin selection. 7. Monitor plates and note wells in which cells begin to grow. Continue to change the medium every 3–4 days. 8. When confluent, trypsinize these wells with 30 ␮L Trypsin–EDTA and re-plate at 0.3 cells per well, again maintaining zeocin selection. 9. When confluent, trypsinize cells and expand through 12-well plates, 6-well plates, and 25 cm2 flasks maintaining zeocin selection. 10. Cryopreserve cells as soon as possible, maintaining a growing culture.

Expression of Recombinant MMP-28 in Mammalian Cells

3.5. Analysis of Cells

3.5.1. Western Blot

61

A number of techniques can be used to assess expression of the transgene. We routinely use Western blotting, quantitative RT-PCR, and immunocytochemistry. 1. Plate cells at 200,000 cells/well in a 6-well plate in serumcontaining medium. Incubate at 37◦ C, 5% CO2 . 2. After 24 h, wash cells ×2 in PBS and add 3 mL per well of serum-free medium. Incubate at 37◦ C, 5% CO2 . 3. After 48 h, collect conditioned medium. Add 0.5 volumes of 10% (w/v) trichloroacetic acid to 1 mL medium and precipitate on ice for 1 h. Centrifuge at >13,000 × g for 15 min at 4◦ C and aspirate supernate. Wash pellets with 0.5 mL ice-cold acetone, centrifuge again, and aspirate supernate. Air-dry pellet for 15 min at room temperature and resuspend in 60 ␮L 1× SDS-PAGE sample buffer containing 50 mM DTT and 6 M urea. 4. Incubate the cell layer with 1 mL per well 0.5 mM EDTA for 1 h at 37◦ C to detach cells. Pellet cells at 13,000 × g for 5 min at 4◦ C. Remove supernate and resuspend cell pellet in 100 ␮L 1× SDS-PAGE sample buffer containing 50 mM DTT and 6 M urea. Sonicate briefly. 5. Rinse wells once with PBS, then add 60 ␮L per well of 1× SDS-PAGE sample buffer containing 50 mM DTT and 6 M urea and scrape extracellular matrix directly into the buffer. 6. Run 25 ␮L of each sample on a 10 or 12.5% SDS-PAGE and transfer to a PVDF or nitrocellulose membrane using electroblotting apparatus. 7. Block membrane for 1 h at room temperature with TBS/T containing 5% (w/v) milk powder on a rocking platform. 8. Wash membrane ×1 with TBS/T for 5 min. 9. Add primary antibody at appropriate dilution (e.g., antiFLAG antibody at 1/5,000 dilution) in TBS/T containing 5% (w/v) milk powder. Incubate overnight at 4◦ C on a rocking platform (see Note 13). 10. Wash membrane ×5 with TBS/T at room temperature for 5 min each wash. 11. Add secondary antibody at appropriate dilution (e.g., horseradish peroxidise-linked anti-mouse Igs at 1/1,000 dilution) in TBS/T containing 5% (w/v) milk powder. Incubate at room temperature for 2 h on a rocking platform. 12. Wash membrane ×5 with TBS/T at room temperature for 5 min each wash and develop using chemiluminescent substrate (according to manufacturer’s instructions).

62

Rodgers and Clark

3.5.2. Quantitative RT-PCR

1. Plate cells at 200,000 cells/well in a 6-well plate in serumcontaining medium. Incubate at 37◦ C, 5% CO2 . 2. After 24 h, wash cells ×2 in PBS. 3. Harvest cells into 0.5 mL of TRIzol reagent or equivalent. 4. Add 100 ␮L chloroform, vortex, centrifuge at >13,000 × g for 15 min at 4◦ C. 5. Collect upper, aqueous phase into a new tube and add an equal volume of 100% isopropanol. Incubate at room temperature for 10 min. 6. Pellet RNA by centrifugation at >13,000 × g for 30 min at 4◦ C. 7. Wash pellet with 0.2 mL ice-cold 70% (v/v) ethanol and centrifuge at >13,000 × g for 10 min at 4◦ C. Carefully remove ethanol and air-dry pellets for 15 min. Resuspend pellets in 20 ␮L nuclease-free water. Samples may be frozen at –80◦ C at this step. 8. Quantify RNA using a spectrophotometer (see Note 7) 9. Add 1 ␮g RNA to 200 ng random hexamers in a total volume of 11 ␮L. 10. Incubate at 70◦ C for 10 min. 11. Incubate briefly on ice then add 4 ␮L 5× first-strand cDNA synthesis buffer, 2 ␮L 0.1 M DTT, 1 ␮L 25 mM dNTP mix, 1 ␮L RNase inhibitor (40 units per ␮L), 1 ␮L (200 units) Superscript II reverse transcriptase (see Note 14). 12. Incubate at 42◦ C for 1 h, then inactivate the reaction at 70◦ C for 15 min. Store cDNA at –20◦ C if necessary. 13. For TaqMan PCR reactions, use 5 ng of reverse transcribed RNA (1 ng for housekeeping genes), 50% TaqMan 2× Master Mix, 100 nM of each primer, and 200 nM of probe in a total volume of 25 ␮L. 14. Cycle at 2 min at 50◦ C, 10 min at 95◦ C, then 40 cycles each consisting of 15 s at 95C and 1 min at 60◦ C. 15. Normalize the data to the housekeeping gene by calculating log10 2−CT where CT is CT (target gene)– CT (housekeeping gene). See Note 15.

3.5.3. Immunocytochemistry

1. Plate cells into chamber slides at 15,000 cells per well in serum-containing medium. Incubate at 37◦ C, 5% CO2 . 2. After 24 h, wash cells ×2 in PBS and replace medium with serum-free medium. Incubate at 37◦ C, 5% CO2 . 3. After 24 h, aspirate medium and fix cells in 4% (w/v) paraformaldehyde in PBS for 10 min at room temperature. Rinse gently ×3 with PBS.

Expression of Recombinant MMP-28 in Mammalian Cells

63

4. Permeabilize cells (if required, see Notes) by incubating with 0.1% (v/v) Triton X-100 in PBS for 10 min at room temperature. 5. Block cells with 3% (w/v) BSA in PBS for 20 min. 6. Wash cells with PBS ×1. 7. Add primary antibody at appropriate dilution in 3% (w/v) BSA in PBS (e.g., anti-FLAG M2 IgG at 1/1,000 dilution) for 60 min at room temperature. 8. Rinse ×3 in PBS for 5 min each rinse. 9. Block again with 3% (w/v) BSA in PBS for 20 min. 10. Add secondary antibody at appropriate dilution in 3% (w/v) BSA in PBS (e.g., Alexa Fluor 488 at 1/1,000 dilution) for 30 min at room temperature. 11. Rinse ×3 in PBS for 5 min each rinse. 12. Stain cell nuclei with 2.5 ␮g/mL DAPI (in 3% (w/v) BSA in PBS) for 5 min. 13. Rinse ×3 PBS for 5 min each rinse. 14. Mount the slide with hydromount prior to adding a coverslip.

4. Notes 1. cDNA should be made from cells or tissues expressing high levels of the gene. This can be ascertained using quantitative RT-PCR as described in Section 3.5.2 where a CT of below 25 is ideal. 2. Any mammalian expression plasmid could be used, but we used pcDNA4/His modified to include a C-terminal FLAG sequence between the end of the coding sequence and the 6×His tag. This gives the option of detecting and/or purifying via either tag. This plasmid also encodes zeocin resistance. 3. PCR conditions should be optimized for each primer pair, particularly across a range of annealing temperatures. The extension time in the 20–30 cycles of amplification will be dependent on the polymerase used and the length of the amplicon, but approximately 1 min for each 1 kb to be amplified should be sufficient. A high-fidelity polymerase like AccuTaq minimizes PCR errors. 4. The quality of agarose used when gel purifying DNA for subcloning can impact on downstream ligation efficiency. We generally use Seakem GTG agarose (Lonza).

64

Rodgers and Clark

5. If the PCR product is predominantly a single band, there is no need to gel purify. Clean up can be by phenol:chloroform extraction (as Section 3.1, Steps 5–9) or via PCR clean-up spin columns. 6. When cutting out DNA from an ethidium bromide-stained agarose gel, keep exposure of the DNA to UV light to a minimum. If it is possible, turn down the power on the UV light box. If not, excise the band as rapidly as possible. 7. We quantify DNA and RNA using a NanoDrop spectrophotometer (NanoDrop Technologies) which requires just 1 ␮L sample. 8. We generally keep the DNA concentration low (using below 100 ng DNA in the reaction) and ligate overnight at 16◦ C. By cutting the vector with an excess of restriction enzyme and then gel purifying the linearized vector, we do not find the need to dephosphorylate either vector or insert. If the vector only control (with ligase) gives a high number of colonies, this may be required. If the vector only control (without ligase) gives colonies, then there is circular plasmid in the prep and therefore this requires further gel purification or cut more vector and increase the amount of restriction enzymes used. 9. Other methods for mutagenesis can be used. For example, we have also used the Stratagene QuikChange methodology. 10. For transfection into mammalian cells, it is essential that the plasmid DNA is of high quality. We generally use QIAGEN plasmid purification kits for this purpose. Dependent on the cell type, a number of transfection reagents or methodologies can be employed, and this will have to be worked out empirically. We have found that FuGENE HD tranfects many different cell types with high efficiency. 11. The concentration of zeocin required for selection will vary between cell lines but is usually between 50 and 1,000 ␮g/mL. The sensitivity of each cell line must be determined empirically prior to making the stable cell lines. Sensitivity may alter depending on confluence (i.e., rate of cell division) and we have generally re-plated cells at 25% confluence at this point where necessary (i.e., where they are more dense than this at 48-h post-transfection). 12. Again, this time point will vary according to the sensitivity of the cells to zeocin or other selection antibiotic. 13. We generally incubate primary antibodies overnight at 4◦ C which we find gives the best sensitivity, but it is often possible to use 2 h at room temperature.

Expression of Recombinant MMP-28 in Mammalian Cells

65

14. If multiple samples are being processed, then making a “master mix” of all additions and then adding an appropriate volume to all samples is often both quicker and more accurate. 15. There are several possible housekeeping genes, but we generally use 18S rRNA for cell culture experiments. This method of data analysis makes the assumption that all PCR reactions occur with similar amplification efficiency. For increased accuracy, it is possible to establish a standard curve by serial dilution of a sample with high expression of the target gene and use this to transform the CT . References 1. Lohi, J., Wilson, C. L., Roby, J. D., and Parks, W. C. (2001) Epilysin, a novel human matrix metalloproteinase (MMP-28) expressed in testis and keratinocytes and in response to injury. J Biol Chem 276, 10134–10144. 2. Marchenko, G. N., and Strongin, A. Y. (2001) MMP-28, a new human matrix metalloproteinase with an unusual cysteineswitch sequence is widely expressed in tumors. Gene 265, 87–93. 3. Nuttall, R. K., Pennington, C. J., Taplin, J., Wheal, A., Yong, V. W., Forsyth, P. A., and Edwards, D. R. (2003) Elevated membranetype matrix metalloproteinases in gliomas revealed by profiling proteases and inhibitors in human cancer cells. Mol Cancer Res 1, 333–345. 4. Werner, S. R., Mescher, A. L., Neff, A. W., King, M. W., Chaturvedi, S., Duffin, K. L., Harty, M. W., and Smith, R. C. (2007)

Neural MMP-28 expression precedes myelination during development and peripheral nerve repair. Dev Dyn 236, 2852–2864. 5. Illman, S. A., Lehti, K., Keski-Oja, J., and Lohi, J. (2006) Epilysin (MMP-28) induces TGF-beta mediated epithelial to mesenchymal transition in lung carcinoma cells. J Cell Sci 119, 3856–3865. 6. Davidson, R. K., Waters, J. G., Kevorkian, L., Darrah, C., Cooper, A., Donell, S. T., and Clark, I. M. (2006) Expression profiling of metalloproteinases and their inhibitors in synovium and cartilage. Arthritis Res Ther 8, R124. 7. Kevorkian, L., Young, D. A., Darrah, C., Donell, S. T., Shepstone, L., Porter, S., Brockbank, S. M., Edwards, D. R., Parker, A. E., and Clark, I. M. (2004) Expression profiling of metalloproteinases and their inhibitors in cartilage. Arthritis Rheum 50, 131–141.

Chapter 4 Expression of Recombinant Matrix Metalloproteinases in Escherichia coli L. Jack Windsor and Darin L. Steele Abstract Matrix metalloproteinases (MMPs) are a group of zinc-dependent endopeptidases that are capable of cleaving all of the components of the extracellular matrix (ECM). The role that the MMPs play in normal and pathological conditions has long been of interest. The mechanisms by which the MMPs cleave the different components of the ECM have been examined extensively. Some of these studies have been made possible, in part, by the ability to express recombinant MMPs. These recombinant MMPs have been utilized in both structural and functional studies. In addition, future studies can benefit from the availability of recombinant MMPs. Recombinant MMPs have been expressed in mammalian and bacterial recombinant expression systems. The most common bacterial expression system employed for this has been the utilization of expression plasmids in Escherichia coli. This has resulted in the production of a large amount of protein in a short period of time. The expression of a recombinant truncated form of human stromelysin-1 (MMP-3) will be used to illustrate the methods utilized for the expression of a MMP in E. coli. This will include discussions about the expression vector, the cloning of the MMP cDNA into the expression vector, protein induction, protein extraction, protein refolding and purification, and protein characterization. Key words: Recombinant proteins, matrix metalloproteinases, Escherichia coli, stromelysin, expression plasmid.

1. Introduction With the advent of recombinant DNA technology, numerous systems have been utilized for the overexpression of proteins. Recombinant protein expression in Escherichia coli (E. coli) typically provides large quantities of the protein of interest in a I.M. Clark (ed.), D.A. Young, A.D. Rowan (Consulting editors), Matrix Metalloproteinase Protocols, Methods in Molecular Biology 622, DOI 10.1007/978-1-60327-299-5 4, © Springer Science+Business Media, LLC 2001, 2010

67

68

Windsor and Steele

relatively short period of time. The expression can result in the accumulation of the recombinant protein to levels approaching 30–50% of the total E. coli protein. Expression of matrix metalloproteinases (MMPs) in E. coli has proven to be very useful in generating proteins for structural and functional studies, including X-ray crystallography. Numerous MMPs, including altered forms, have been successfully expressed in and purified from E. coli. These include forms of stromelysin-1, stromelysin-2, stromelysin-3, matrilysin, elastase, collagenase-1, collagenase-3, neutrophil collagenase, and membrane type-1 MMP (1–20). The expression of a truncated form of human stromelysin-1 (SL-1) will be used to illustrate the methods utilized for the expression of a matrix metalloproteinase in E. coli and for its extraction, refolding, and purification.

2. Materials 1. pGEMEX-1 expression system (Promega; Madison, WI). 2. Matrix Metalloproteinase cDNA. 3. E. coli strains CJ236, DH5␣, and BL21 (DE3). 4. Helper phage VCS-M13 (Stratagene; La-Jolla, CA). 5. Oligonucleotide primers. 6. Restriction enzymes, T7 DNA polymerase, and T4 DNA ligase. 7. Agarose and DNA sequencing gel equipment. 8. DYT (double yeast tryptone) media. 9. Ampicillin. 10. IPTG (isopropyl-␤-D-thio-galactopyranoside). 11. French Press. 12. PMSF (phenylmethylsulfonyl fluoride). 13. Standard (or wash) buffer: 50 mM Tris-HCl, pH 7.5, 0.2 M NaCl, 5 mM CaCl2 , and 1 ␮M ZnCl2 at 4◦ C. 14. Extraction buffer: 6 M urea in 50 mM Tris-HCl, pH 7.5, 0.2 M NaCl, 5 mM CaCl2 , and 1 ␮M ZnCl2 at 4◦ C. 15. SDS-PAGE (sodium dodecyl sulfate-polyacrylamide gel electrophoresis) equipment. 16. Whatman filter paper #1. 17. Chromatography equipment. 18. Sephacryl S-200 HR (Pharmacia; Piscataway, NJ). 19. Anti-MMP antibody.

Expression of Recombinant Matrix Metalloproteinases in Escherichia coli

69

20. CNBr-activated Sepharose 4B (Pharmacia). 21. APMA (aminophenyl-mercuric acetate). 22. Mca peptide (Bachem; King of Prussia, PA). 23. Casein.

3. Methods The methods described below outline (1) the construction of the expression plasmid, (2) the induction of protein expression, (3) the extraction of the protein from E. coli, (4) the refolding and purification of the protein, and (5) the characterization of the protein. 3.1. Expression Plasmid

The construction of the expression plasmid for a truncated form of stromelysin-1 (SL-1) is described in Subheading 3.1.1–3.1.5 This includes (a) the description of the expression vector, (b) the description of the SL-1 cDNA, (c) cloning, (d) the deletion of the coding regions for the T7 gene 10 protein in the pGEMEX-1 vector and for the signal peptide of SL-1, and (e) the deletion of the coding region for the hemopexin-like domain of SL-1.

3.1.1. pGEMEX-1 Expression Vector

The pGEMEX-1 (Fig. 4.1A) expression system (Promega) is based on the T7 expression system developed by Studier (21). Sequences inserted into the multiple cloning site can be expressed as T7 gene 10 fusion proteins that are transcriptionally initiated from the pGEMEX-1 T7 promoter when transformed into E. coli DE3 cells, which contain an IPTG (isopropyl-␤- D-thiogalac-topyranoside) inducible T7 RNA polymerase gene. Expression of the recombinant protein alone can be accomplished by removing the coding region for the T7 gene 10 protein. The pGEMEX-1 expression vector contains a phage f1 origin of replication that provides a mechanism for the production of singlestranded DNA, which can then be used in site-directed mutagenesis in order to construct expression plasmids and/or mutants in the recombinant protein. The pGEMEX-1 expression vector also contains an ampicillin resistance gene for selectivity. These and other characteristics make pGEMEX-1 a suitable expression system to use for the expression of truncated SL-1 and other proteins (see Note 1).

3.1.2. cDNA

A plasmid (pBS’SL) containing the cDNA for human SL-1 was kindly donated by Dr. Goldberg (22). It contained the initiation methionine codon followed by 1431 nucleotides coding for the

70

Windsor and Steele

Fig. 4.1. (A) Schematic drawing of pGEMEX-1 expression plasmid adapted from Promega (Madison, WI). (B) Schematic drawing of expression plasmid pGXmini-SL-1 for the expression of truncated SL-1.

Expression of Recombinant Matrix Metalloproteinases in Escherichia coli

71

477 amino acid preprostromelysin-1 protein along with a portion of the 3 untranslated region. The coding region for SL-1 was removed from pBS’SL by digestion with EcoRI and BamHI by standard molecular biology techniques (23) (see Note 2). 3.1.3. Cloning

DNA manipulations were performed by standard recombinant DNA methods (23) to construct the expression plasmid and are not described here in detail due to space limitations. The expression plasmid for truncated SL-1 (mini-SL-1) was constructed by adding BamHI linkers to the EcoRI-BamHI fragment from pBS’SL. After digestion with BamHI and ligation into the BamHI restriction site of the pGEMEX-1 vector, the DNA was transformed into E. coli DH5␣ cells by standard methods (23). The E. coli DH5␣ cells were then plated on DYT (double yeast and tryptone) plates containing ampicillin (50–100 ␮g/mL) and incubated overnight at 37o C. Single colonies were selected and grown overnight in DYT with ampicillin. The plasmid DNA was then isolated (23) and checked for the presence of the insert and for the correct orientation using restriction enzyme digestions and DNA sequencing (24,25).

3.1.4. Removal of the Coding Regions for the SL-1 Signal Peptide and for the T7 Gene 10 Protein

The first 18 amino acids of human SL-1 constitute the signal peptide that is removed upon secretion into the extracellular environment of eukaryotic cells and is not necessary for protein expression in prokaryotic cells such as E. coli. The signal peptide sequence of the SL-1 insert and the coding sequence for the T7 gene 10 protein in pGEMEX-1 were removed by site-directed mutagenesis (26) so that only the SL-1 protein would be produced. Single-stranded DNA was produced from E. coli CJ236 cells (previously transformed with pGEMEX-1 containing the cDNA for SL-1) infected with helper phage VCS-M13 (Stratagene, La-Jolla, CA) as described by the manufacturer. Site-directed mutagenesis was performed using this single-stranded DNA, the oligonucleotide primer 5 GGAGA TATACATATGTATCCGCTGGATGGAGCTGC 3 , T7 DNA polymerase, and T4 DNA ligase. This oligonucleotide primer was also specifically designed to place the initiation codon ATG in frame with the coding region of the SL-1 cDNA. This expression plasmid, pGSL-1 (9), containing the full-length cDNA for SL-1 minus the signal peptide was verified using restriction enzyme digestions and DNA sequencing.

3.1.5. Deletion of the Coding Region for the Hemopexin-Like Domain (Carboxyl Terminus) of SL-1

Single-stranded DNA was produced from E. coli CJ236 cells transformed with the pGSL-1 plasmid using helper phage VCSM13 as described by the manufacturer (Stratagene). The pGSL-1 single-stranded DNA and the primer 5 TATGGACCTCCCCCGGATCCCTGATAGAGAGATATGTAGAAG 3 were used to remove the coding sequence for the hemopexin-like domain of

72

Windsor and Steele

SL-1 by introducing the stop codons TGA and TAG after the codon for amino acid 253 of SL-1. This resulted in expression plasmid pGXmini-SL-1 (Fig. 4.1B), which contained the coding region of SL-1 minus the signal peptide and the hemopexin-like domain (mini-SL-1). Before attempting protein expression, the expression construct was verified by restriction enzyme digestions and DNA sequencing. 3.2. Protein Induction

3.2.1. E. coli Transformation and Plasmid Selection

The next steps in this process involve the transformation of E. coli with the expression plasmid for mini-SL-1 followed by the induction with IPTG to initiate protein expression (see Note 3). 1. Transform DH5␣ E. coli cells with plasmid pGXmini-SL-1 DNA using standard molecular biology methods (23). 2. Plate cells on DYT plates containing ampicillin and incubate overnight at 37◦ C. 3. Select single colonies and grow overnight at 37◦ C in DYT media containing ampicillin. Take care selecting the single colonies from the plates to avoid cross contamination from the other colonies. 4. Freeze samples of the overnight cultures at –70◦ C in DYT media containing 25% glycerol. 5. Purify the plasmid DNA from the remaining overnight cultures by standard methods (23).

3.2.2. Induction

1. Transform E. coli BL21 (DE3) cells with the pGXminiSL-1 plasmid DNA using standard methods (23) and grow overnight at 37◦ C in DYT media with ampicillin. 2. Freeze samples of the overnight cultures at –70◦ C in DYT media containing 25% glycerol. 3. Inoculate individual liters of DYT containing ampicillin with aliquots (2–4 mL) of the overnight cultures and allow to grow to a density of 108 cells/mL. 4. Induce the cells with IPTG (1 mM) for 2–3 h. It should be noted that the induction time can be extended to increase expression, however for mini-SL-1 this usually results in more heterogeneity in the protein.

3.2.3. Harvest

1. After 2–3 h of induction, harvest the cells by centrifugation (30,000g; 30 min) and resuspend the cell pellet in standard buffer. 2. After resuspension of the pellet, the cells can either be directly extracted or frozen at –20◦ C for as long as a month before extraction without significant breakdown of the miniSL-1 protein.

Expression of Recombinant Matrix Metalloproteinases in Escherichia coli

3.3. Protein Extraction

3.3.1. Cell Disruption

73

The steps described in Subheadings 3.3.1 and 3.3.2 outline the procedure for extraction of the protein from the E. coli cells in a manner that yields significant quantities of native-like protein after refolding (see Note 4). 1. Add PMSF (phenylmethylsulfonyl fluoride) to a final concentration of 1 mM to freshly resuspended cells or fully thawed frozen cells. 2. Pass the cells through a French press at 10,000–15,000 psi, then repeat to ensure complete cell lysis.

3.3.2. Isolation of Inclusion Bodies and Extraction

1. Centrifuge the cell lysate (30,000g; 20 min) and retain both supernatant (to check for presence of the mini-SL-1 protein) and pellet (see Fig. 4.2).

Fig. 4.2. Extraction of mini-SL-1 from E. coli. The supernatant following passage through a French press (FP sup), the wash supernatant of the pellet after passage through the French press (FP wash), and the 6 M urea extract of the pellet containing the inclusion bodies (6 M urea extract) were resolved in a 10% SDS-PAGE gel and stained with Coomassie blue.

2. Wash the pellet (which contains the protein laden inclusion bodies) with standard buffer by vigorously pipeting to a homogeneous suspension, then re-pellet by centrifugation as in step 1. 3. Solubilize the pellet from step 2 in extraction buffer and extract at 4◦ C on an orbital mixer for 2–3 h or overnight. Centrifuge as in step 1 and retain the supernatant (extract). 4. The extract can be frozen at –20◦ C for up to 2 wk before purification.

74

Windsor and Steele

3.4. Protein Refolding and Purification

3.4.1. Gel Filtration

Described below are the steps that can be utilized in the separation of the mini-SL-1 protein from the E. coli proteins and its progressive purification to a homogeneous sample (see Note 5). 1. Apply the urea extract onto a gel filtration column (2.5 cm × 88 cm) containing Sephacryl S-200 HR (Pharmacia) equilibrated with extraction buffer and elute with the same buffer. 2. Identify fractions of interest by Coomassie blue staining of a SDS-PAGE gel (see Fig. 4.3) and verify by a Western blot probed with anti-SL-1 antibody IID4 or equivalent (27).

Fig. 4.3. Gel filtration column. Fractions collected from a Sephacryl S-200 HR column were resolved in a 10% SDS-PAGE gel and stained with Coomassie blue.

3.4.2. Refolding

1. Pool the fractions of interest and dilute in the extraction buffer to a protein concentration less than 1 mg/mL. This dilution reduces steric hindrance by providing more space for refolding to occur, thus allowing for more efficient protein refolding and decreased protein precipitation. 2. Dialyze at 4◦ C against at least a 100-fold volume of standard buffer to remove the urea. 3. Alternatively, dilute the pooled fractions more than 100-fold with standard buffer to lower the urea and protein concentrations and incubate at 4◦ C overnight. 4. After refolding, concentrate the sample and store at –20◦ C until use. If the sample is to be further purified, it can be frozen at –20◦ C for a few days before proceeding to the next step.

3.4.3. Affinity Chromatography after Gel Filtration

The purity of the samples from the gel filtration varies, but greater than 80% pure protein can be obtained from this step alone (Fig. 4.4A, pooled fractions). A small smount of contaminating sub-28,000 kDa proteins can be seen in the dialyzed pooled fractions from the gel filtration column (Fig. 4.4A, pooled fractions). To remove these proteins, the samples can be further purified by affinity chromatography using anti-SL-1 monoclonal

Expression of Recombinant Matrix Metalloproteinases in Escherichia coli

75

Fig. 4.4. Affinity purification of fractions from the gel filtration column. Selected fractions from the Sephacryl S-200 HR column were pooled, diluted and/or dialyzed to remove the urea, and passed over the anti-SL-1 antibody IID4 affinity column. The bound material was eluted with 6 M urea (peak) and resolved in 12.5% SDS-PAGE gels. Companion gels were stained with Coomassie blue (A) or transferred to nitrocellulose for Western blot analysis using 5 ␮g/mL of antihuman-SL-1 monoclonal antibody IID4 (B) as described in ref. (27).

antibody IID4 or equivalent, coupled to CNBr-activated Sepharose 4B (27). 1. Apply the diluted or dialyzed sample from gel filtration above to the anti-SL-1 monoclonal antibody-Sepharose affinity column pre-equilibrated with standard buffer. 2. Wash the column with 3–5 column volumes of standard buffer or until the absorbance at 280 nm returns to that of the buffer. 3. Elute the bound material with extraction buffer (this results in mini-SL-1 which is greater than 98% pure, Fig. 4.4B) 4. Dialyze the sample against standard buffer to remove the urea. An estimated 10–20 mgs per liter of mini-SL-1 can be obtained in this manner by growing individual liters of E. coli at 37◦ C. It is possible to increase the yield to 20– 40 mgs per liter by optimizing the growth conditions in a fermentor. 3.4.4. Affinity Chromatography without Gel Filtration

Affinity chromatography can also be performed using the 6 M urea extract of the inclusion body pellet without prior passage over the gel filtration column. 1. Dilute the urea extract dropwise into greater than 100 vol of standard buffer (to minimize protein precipitation).

76

Windsor and Steele

Fig. 4.5. Affinity purification of 6 M urea extract. The 6 M urea extract was diluted more than 100-fold before passage over the anti-SL-1 antibody IID4 affinity column without prior passage over the gel filtration column. The supernatant following passage through a French press (FP sup), the wash supernatant of the pellet after passage through the French press (FP wash), the 6 M urea extract of the pellet containing the inclusion bodies (6 M urea extract), and the peak eluted from the anti-SL-1 antibody IID4 affinity column with 6 M urea (6 M urea peak) were resolved in a 12.5% SDS-PAGE gel and stained with Coomassie blue.

2. Filter the diluted extract through #1 Whatman filter paper 3. Apply to the affinity column, wash, and elute as above. This method also yields greater than 98% pure mini-SL-1 (see Fig. 4.5, 6 M urea peak). However, this method results in a greater loss of the protein due to precipitation during the dilution step when compared to the samples passed over the gel filtration column first. 3.5. Characterization

3.5.1. APMA-Induced Activation

The purified mini-SL-1 can be assayed for catalytic competency by assessing its ability to undergo APMA (aminophenyl-mercuric acetate)-induced activation, its ability to cleave casein in a zymogram, and its ability to cleave a synthetic peptide. 1. Incubate samples of mini-SL-1 in the presence and absence of 1 mM APMA for 12 h at 37◦ C. 2. Resolve the samples in a 12.5% SDS-PAGE gel. 3. Transfer to nitrocellulose and analyze by Western blot utilizing anti-SL-1 monoclonal antibody IID4 or equivalent (27). The APMA induces the mini-SL-1 protein (Mr 28,000) to autocatalytically convert (see Fig. 4.6) to the lower molecular weight active form (Mr 18,000). This conversion demonstrates that the mini-SL-1 is catalytically competent and able to process its own propeptide during APMA-induced self-activation.

Expression of Recombinant Matrix Metalloproteinases in Escherichia coli

77

Fig. 4.6. APMA-induced activation. Samples of the affinity purified mini-SL-1 were incubated with and without 1 mM APMA at 37◦ C for 12 h. The samples were resolved in an 12.5% SDS-PAGE gel and transferred to nitrocellulose for Western blot staining with anti-SL-1 monoclonal antibody IID4 (5 ␮g/mL) as previously described (27).

3.5.2. Casein Zymography

1. Resolve a sample of the mini-SL-1 in a 10% SDS-PAGE gel co-polymerized with 1 mg/mL casein. 2. After electrophoresis, wash the gel vigorously in three changes of 1% Triton X-100 and incubate in standard buffer for 16 h at 37◦ C (9).

Fig. 4.7. Casein zymography of mini-SL-1. A sample of the affinity purified mini-SL-1 was resolved in a 10% SDS-PAGE gel and stained with Coomassie blue. A companion sample was resolved in a 10% SDS-PAGE gel copolymerized with 1 mg/mL of casein (zymogram) and incubated at 37◦ C for 16 h and stained with Coomassie blue to visualize the lytic band.

78

Windsor and Steele

3. Stain with Coomassie blue to visualise bands of lysis. The correctly folded mini-SL-1 is capable of cleaving the casein incorporated in the gel as visualized by the zone of clearing (see Fig. 4.7, zymogram). 3.5.3. Cleavage of the Mca-Peptide

1. Incubate the coumarinyl peptide derivative Mca-P-L-G-LDpa-A-R (2.5 ␮M) (28) with 3 nM mini-SL-1 (activated with APMA) in standard buffer at 23◦ C. 2. Monitor emerging fluorescence to calculate kcat /Km . A kcat /Km of 33.3 ␮M–1 h–1 is comparable to that previously reported for mini-SL-1 and indicates that the protein is correctly refolded (9).

4. Notes 1. The pGEMEX-1 expression vector has been utilized very effectively in producing recombinant MMPs (3,9,10). Other expression vectors have also proven to be very effective in producing recombinant MMPs in E. coli. Some of these include (1) pET plasmids (4–6,11,15–17), (2) pRSET plasmids (1,12,14), and (3) pBluescript plasmids (8,30). All of these are based on the T7 expression system and take advantage of the same strategies for protein expression. The selection of an appropriate expression vector depends on the objectives set forth in the investigation. 2. With the rapid advancement in molecular biology techniques, multiple methods can be utilized in the expression of proteins and the construction of expression plasmids. Methods such as cloning by PCR (polymerase chain reaction) have been very effective in the construction and reconstruction of the cDNAs, as well as the expression plasmids for the MMPs. 3. Several modifications to the described methods can be performed to optimize protein expression. Fermentation facilities, which optimize the growth conditions for the bacteria, can be used to produce larger volumes of the E. coli. This not only results in more protein overall, but more protein per liter of E. coli. A problem that has been observed is that the expression plasmid in E. coli BL21 (DE3) cells can be unstable. Therefore, instead of expressing the protein from a frozen stock of transformed E. coli BL21 (DE3) cells, transformation of the E. coli BL21 (DE3) cells may have to be performed just prior to scaling up for expression. Also, it has been observed that the expression of more protein can be

Expression of Recombinant Matrix Metalloproteinases in Escherichia coli

79

obtained if the overnight culture used to inoculate the liters is in log phase. Another modification includes changing the temperature at which the cells are grown and induced to 30◦ C instead of 37◦ C to optimize protein expression and to minimize breakdown of the protein. 4. Alternative methods to the use of the French press for E. coli cell lysis include treatment with lysozyme, freezethawing, and sonication or some combination of these. The utilization of a protease inhibitor (for example PMSF) or a cocktail of protease inhibitors is contingent on the amount of breakdown of the recombinant MMP that occurs throughout the extraction process. This chapter describes the use of urea to extract the inclusion bodies, but 4–6 M guanidine hydrochloride can also be used. For full-length Collagenase-1 extracted with 6 M guanidine hydrochloride, a stepwise dilution to approx 2 M guanidine hydrochloride with 20% glycerol, 2.5 mM glutathione (reduced and oxidized), and 5 mM CaCl2 followed by buffer exchange to remove denaturant and thiol reagents was determined to yield the best results (7). The refolding steps, especially for the full-length MMPs, have been the greatest challenge in attempting to maximize yield of native-like protein. Determination of which extraction and refolding methods to be utilized appears to vary from protein to protein. 5. Most purification schemes of MMPs from E. coli inclusion bodies involve extraction with a denaturant followed by a gel filtration step to separate the recombinant protein from the E. coli proteins. In this report, an antibody affinity column has been utilized to purify the protein further. Since an affinity purification step may not always be available, other purification steps have been utilized that include an anion and/or cation exchange column (Q-Sepharose or S-Sepharose, respectively) (11).

Acknowledgments The authors thank Dr. G. I. Goldberg for providing the cDNA for human stromelysin-1. We also thank Drs. Susan B. LeBlanc and Grazyna Galazka for construction of plasmids pGSL-1 and pGXmini-SL-1. This work was supported by USPHS grants DE08228 (to S. M. Michalek), DE10631 (to J. A. Engler), DE/CA 11910 (to J. A. Engler), and AR44701 (to L. J. Windsor).

80

Windsor and Steele

References 1. Rosenfeld, S. A., Ross, O. H., Corman, J. I., Pratta, M. A., Blessington, D. L., Fesser, W. S., and Freimark, B. D. (1994). Production of human matrix metalloproteinase 3 (stromelysin) in Escherichia coli. Gene 139, 281–286. 2. Salowe, S. P., Marcy, A. I., Cuca, G. C., Smith, C. K., Kopka, I. E., Hagmann, W. K., and Hermes, J. D. (1992). Characterization of zinc-binding sites in human stromelysin1: stoichiometry of the catalytic domain and identification of a cysteine ligand in the proenzyme. Biochemistry 31, 4535–4540. 3. Ye, Q.-Z., Johnson, L. L., Hupe, D. J., and Baragi, V. (1992). Purification and characterization of the human stromelysin catalytic domain expressed in Escherichia coli. Biochemistry 31, 11,231–11,235. 4. Gronski, Jr., T. J., Martin, R. L., Kobayashi, D. K., Walsh, B. C., Holman, M. C., Huber, M., Van Wart, H. E., and Shapiro, S. D. (1997). Hydrolysis of a broad spectrum of extracellular matrix proteins by human macrophage elastase. J. Biol. Chem. 272, 121,89–121,94. 5. Ho, T. F., Qoronfleh, M. W., Wahl, R. C., Pulvino, T. A., Vavra, K. J., Falvo, J., Banks, T. M., Brake, P. G., and Ciccarelli, R. B. (1994). Gene expression, purification and characterization of recombinant human neutrophil collagenase. Gene 146, 297–301. 6. Lemaitre, V., Jungbluth, A., and Eeckhout, Y. (1997). The recombinant catalytic domain of mouse collagenase-3 depolymerizes type I collagen by cleaving its aminotelopeptides. Biochem. Biophys. Res. Commun. 230, 202– 205. 7. Zhang, Y. and Gray, R. D. (1996). Characterization of folded, intermediate, and unfolded states of recombinant human interstitial collagenase. J. Bio. Chem. 271, 8015–8021. 8. Windsor, L. J., Birkedal-Hansen, H., Birkedal-Hansen, B., and Engler, J. A. (1991). An internal cysteine plays a role in the maintenance of the latency of human fibroblast collagenase. Biochemistry 30, 641– 647. 9. Windsor, L. J., Steele, D. L., LeBlanc, S. B., and Taylor, K. B. (1997). Catalytic domain comparisons of human fibroblast-type collagenase, stromelysin-1, and matrilysin. Biochim. Biophys. Acta. 1334, 261–272. 10. Ye, Q.-Z., Johnson, L. L., and Baragi, V. (1992). Gene synthesis and expression in E. colifor pump, a human matrix metalloproteinase. Biochem. Biophys. Res. Commun. 186, 143–149.

11. Marcy, A. I., Eiberger, L. L., Harrison, R., Chan, H. K., Hutchinson, N. I., Hagmann, W. K., Cameron, P. M., Boulton, D. A., and Hermes, J. D. (1991). Human fibroblast stromelysin catalytic domain: expression, purification, and characterization of a C-terminally truncated form. Biochem. 30, 6476–6483. 12. Pendas, A. M., Knauper, V., Puente, X. S., Llano, E., Mattei, M.-G., Apte, S., Murphy, ´ G., and Lopez-Otin, C. (1997). Identification and characterization of a novel human matrix metalloproteinase with unique structural characteristics, chromosomal location, and tissue distribution. J. Biol. Chem. 272, 4281–4286. 13. Kinoshita, T., Sato, H., Takino, T., Itoh, M., Akizawa, T., and Seiki, M. (1996). Processing of a precursor of 72-kilodalton type IV collagenase/gelatinase A by a recombinant membrane-type I matrix metalloproteinase. Cancer Res. 56, 2535– 2538. 14. Freimark, B. D., Feeser, W. S., and Rosenfeld, S. A. (1994) Multiple sites of the propeptide region of human stromelysin1 are required for maintaining a latent form of the enzyme. J. Biol. Chem. 269, 26,982–26,987. 15. Murphy, G., Segain, J.-P., O’Shea, M., Cockett, M., Ioannou, C., Lefebre, O., Chambon, P., and Basset, P. (1993). The 28-kDa N-terminal domain of mouse stromelysin-3 has the general properties of a weak metalloproteinase. J. Biol. Chem. 268, 15,435–15,441. 16. Shapiro, S. D., Griffin, G. L., Gilbert, D. J., Jenkins, N. A., Copeland, N. G., Welgus, H. G., Senior, R. M., and Ley, T. (1992). Molecular cloning, chromosomal localization, and bacterial expression of a murine macrophage metalloelastase. J. Biol. Chem. 267, 4664–4671. 17. Lichte, A., Kolkenbrock, H., and Tschesche, H. (1996). The recombinant catalytic domain of membrane-type matrix metalloproteinase-1 (MTI-MMP) induces activation of progelatinase A and progelatinase A complexed with TIMP-2. FEBS Lett. 397, 277–282. 18. Sato, H., Kinoshita, T., Takino, T., Nakayama, K., and Seiki, M. (1996). Activation of a recombinant membrane type 1matrix metalloproteinase (MT1-MMP) by furin and its interaction with tissue inhibitor of metalloproteinases (TIMP-2). FEBS Lett. 393, 101–104.

Expression of Recombinant Matrix Metalloproteinases in Escherichia coli 19. Itoh, M., Masuda, K., Ito, Y., Akizawa, T., Yoshioka, M., Imai, K., Okada, Y., Sato, H., and Seiki, M. (1996). Purification and refolding of recombinant human proMMP7 (pro-matrilysin) expressed in Escherichia coli and its characterization. J. Biochem. 119, 667–673. 20. Welch, A. R., Holman, C. M., Browner, M. F., Gehring, M. R., Kan, C.-C. and Van Wart, H. E. (1995). Purification of human matrilysin produced in Escherichia coli and characterization using a new optimized fluorogenic peptide substrate. Archives of Biochemistry and Biophysics 324, 59–64. 21. Studier, F. W. and Moffatt, B. A. (1986). Use of Bacteriophage T7 RNA Polymerase to Direct Selective High-level Expression of Cloned Genes. J. Mol. Biol. 189, 113–130. 22. Wilhelm, S. M., Collier, I. E., Kronberger, A., Eisen, A. Z., Marmer, B. L., Grant, G. A., Bauer, E. A., and Goldberg, G. I. (1987). Human skin fibroblast stromelysin:Structure, glycosylation, substrate specificity, and differential expression in normal and tumorigenic cells. Proc. Natl. Acad. Sci. USA 84, 6725–6729. 23. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989). Molecular Cloning, A Laboratory Manual, Second ed. Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press. 24. Sanger, F., Nicklen, S., and Coulson, A. R. (1977). DNA sequencing with chainterminating inhibitors. Proc. Natl. Acad. Sci. USA 74, 5463–5467.

81

25. Biggin, M. D. , Gibson, T. J., and Hong, G. F. (1983). Buffer gradient gels and 35 S label as an aid to rapid DNA sequence determination. Proc. Natl. Acad. Sci. USA 80, 3963– 3965. 26. Zoller, M. J. and Smith, M. (1983). Oligonucleotide-directed mutagenesis of DNA fragments cloned into M13 vectors. Methods Enzymol. 100, 468–500. 27. Windsor, L. J., Grenett, H., BirkedalHansen, B., Bodden, M. K., Engler, J. A., and Birkedal-Hansen, H. (1993). Cell-typespecific regulation of SL-1 and SL-2 genes. Induction of SL-2, but not SL-1, in human keratinocytes in response to cytokines and phorbolesters. J. Biol. Chem. 266, 13,064– 13,069. 28. Knight, C. G., Willenbrock, F., and Murphy, G. (1992). A novel coumarin-labelled peptide for sensitive continuous assays of the matrix metalloproteinases. FEBS Lett. 296, 263–266. 29. Lovejoy, B., Cleasby, A., Hassell, A. M., Longley, K., Luther, M. A., Weigl, D., McGeehan, G., McElroy, A. B., Drewry, D., Lambert, M. H., and Jordan, S. R. (1994). Structure of the catalytic domain of fibroblast collagenase complexed with an inhibitor. Science 263, 375–377. 30. Windsor, L. J., Bodden, M. K., BirkedalHansen, B., Engler, J. A., and BirkedalHansen, H. (1994). Mutational analysis of residues in and around the active site of human fibroblast-type collagenase. J. Biol. Chem. 269, 26,201–26,207.

Chapter 5 Expression of Recombinant ADAMTS in Insect Cells Gavin C. Jones, Mireille N. Vankemmelbeke, and David J. Buttle Abstract The “a disintegrin and metalloproteinase with thrombospondin motifs” (ADAMTS) enzymes are secreted proteinases involved in development, blood clotting and the turnover of extracellular matrix. Manufacturing recombinant enzyme presents quite a challenge due to the presence of disulphide bridges, the large size and modular structure. A sub-group of these enzymes are known as “aggrecanases” and it is likely that they are involved in a number of pathologies related to increased turnover of the extracellular matrix, particularly in tissues where the concentration of proteoglycans is high, such as cartilage and the central nervous system. We have expressed three of these enzymes, ADAMTS-1, -4 and -5, in insect cells using plasmid-based systems. Key words: ADAMTS, expression, purification, insect cell.

1. Introduction The “a disintegrin and metalloproteinase with thrombospondin motifs” (ADAMTS) group of enzymes encompasses 19 genes in the human genome, as well as representatives in most if not all other vertebrates as well as invertebrates as primitive as the nematode Caenorhabditis elegans (1). These enzymes are closely related to the “a disintegrin and metalloproteinase”(ADAM) reprolysins and phylogenetically classified together with them in subfamily M12B in the MEROPS database (http://merops.sanger.ac.uk/) (2). Some of the ADAMTS enzymes have no known function while others fall into three distinct functional classifications: (1) ADAMTS-2, -3 and -14 are procollagen N-peptidases; (2) ADAMTS-13 is a von Willebrand factor-cleaving protease; and I.M. Clark (ed.), D.A. Young, A.D. Rowan (Consulting editors), Matrix Metalloproteinase Protocols, Methods in Molecular Biology 622, DOI 10.1007/978-1-60327-299-5 5, © Springer Science+Business Media, LLC 2001, 2010

83

84

Jones, Vankemmelbeke, and Buttle

(3) ADAMTS-1, -4, -5, -8, -9 and -15 are potential aggrecanases. This latter group all have the capacity to cleave large aggregating proteoglycans such as aggrecan and may therefore be implicated in diseases where breakdown of the extracellular matrix occurs, such as the arthritides, tumour growth and metastasis, and certain diseases of the central nervous system such as stroke and multiple sclerosis. As secreted proteins, the ADAMTS possess many disulphide bridges, and most enzymes are glycosylated. They are modular proteins containing signal- and pro-peptides, a metalloproteinase catalytic domain, a disintegrin-like domain, cysteine-rich and “spacer” regions and a variable number of C-terminal domains and modules (3). The presence of a metalloproteinase catalytic unit may present the ADAMTS enzymes with the capacity to inter-molecularly cleave or process themselves, and the ancillary domains interact with sulphated glycosaminoglycans and mediate binding to extracellular matrix (4, 5). Making recombinant forms of these enzymes therefore presents a considerable challenge. Recombinant ADAMTS-1, -4 and -5 are available commercially, but in all cases these are truncated forms which nonetheless possess catalytic activity (though potentially with differing substrate specificity to the full-length enzyme). We have expressed these three enzymes in insect cells using plasmid-based systems (6) and will describe in detail how this was achieved.

2. Materials 2.1. Construction of Expression Vectors

1. Custom primers (Invitrogen, Paisley, UK) dissolved in deionized H2 O to stock solutions of 100 ␮M and working dilutions of 10 ␮M. 2. Tli polymerase, 10× polymerase buffer [500 mM KCl, 100 mM Tris-HCl, 1% (v/v) Triton X-100, pH 9.0] and 25 mM MgSO4 (Promega, Southampton, UK). 3. 10 mM dNTP mix (2.5 mM dATP, 2.5 mM dCTP, 2.5 mM dGTP, 2.5 mM dTTP) (Promega). 4. 50× TAE buffer: 2 M Tris-HCl, 50 mM EDTA, 1 M acetic acid, pH 7.6–7.8. Dilute to 1× prior to use. 5. 10 mg/mL ethidium bromide (EtBr) (Sigma-Aldrich, Dorset, UK). 6. QIAquick Gel Sussex, UK).

Extraction

Kit

(QIAGEN,

West

7. Taq polymerase, 5× polymerase buffer (containing 7.5 ␮M MgCl2 ) (Promega).

Expression of Recombinant ADAMTS in Insect Cells

85

8. 10 mM dATP (Promega). R vector (Invitrogen). 9. pIB/V5-His-TOPO

10. Chemically competent TOP10 Escherichia coli (Invitrogen). 11. SOC medium: 2% (w/v) tryptone (pancreatic digest of casein), 0.5% w/v yeast extract, 8.6 mM NaCl, 2.5 mM KCl, 20 mM MgSO4 , 20 mM glucose (Invitrogen). 12. LB agar and LB broth (Sigma-Aldrich). 13. Ampicillin (Sigma-Aldrich). 14. 10× Hogness buffer: 40% (v/v) glycerol, 36 mM K2 HPO4 , 13 mM KH2 PO4 , 20 mM trisodium citrate, 10 mM MgSO4 , pH 7.0. 15. QIAprep Spin Miniprep Kit (QIAGEN). 2.2. Cell Culture

1. Lysis buffer: 150 mM glucose, 25 mM Tris-HCl, 10 mM EDTA, pH 8.0. 2. NaOH/SDS: 0.2 M NaOH, 1% (w/v) SDS. Prepare from separate NaOH and SDS stocks immediately before use. 3. 3 M sodium acetate, pH 4.8. 4. STE buffer: 150 mM NaCl, 5 mM EDTA, 50 mM TrisHCl, pH 8.0. 5. 1% (v/v) Triton X-100 (BDH, Dorset, UK). 6. CsCl (Sigma-Aldrich). 7. 3.9 mL polyallomer tubes [Beckman Coulter (UK) Limited – Biomedical Research, High Wycombe, UK]. 8. TLN-100 rotor and TL-100 ultracentrifuge [Beckman Coulter (UK) Limited – Biomedical Research, High Wycombe, UK]. 9. H2 O-saturated butan-1-ol. 10. 0.3 M sodium acetate, pH 7.0. 11. High-FiveTM cells (Invitrogen). 12. Foetal Bovine Serum (FBS) (Invitrogen). 13. HyQ-SFX serum-free insect culture medium (Perbio Science UK Ltd, Cheshire, UK). 14. Gentamycin (Sigma-Aldrich). R reagent (Invitrogen). 15. Cellfectin

16. Blasticidin (Invitrogen). Dissolved in de-ionized H2 O to a concentration of 100 mg/mL. 17. 10× assay buffer: 1 M Tris-HCl, 100 mM CaCl2 , 1% (w/v) CHAPS, pH 8.0. Dilute to 1× immediately prior to use.

86

Jones, Vankemmelbeke, and Buttle

2.3. Purification of Recombinant ADAMTS Enzymes

1. 3,4-Dichloroisocoumarin (DCI) dissolved in dimethylsulphoxide (DMSO) to a stock concentration of 10 mM (Sigma-Aldrich). 2. L -trans-Epoxysuccinyl-leucylamido-(4-guanidino)butane (E64) dissolved in H2 O to a stock concentration of 10 mM (Sigma-Aldrich). ¨ KTA FPLC system (GE Healthcare, Chalfont St Giles, 3. A Bucks, UK). 4. Sephadex G-25 desalting column (GE Healthcare). 5. Heparin-Sepharose Fast-Flow Column (GE Healthcare). 6. Mono Q HR 5/5 column (GE Healthcare). 7. Column buffer A1: 50 mM Tris-HCl, 0.15 M NaCl, 0.1% (w/v) CHAPS, pH 7.0. Filter through a 0.45 ␮m membrane and degas. Store at ambient temperature and use within 2 weeks. 8. Column buffer A2: 50 mM Tris-HCl, 1 M NaCl, 0.1% (w/v) CHAPS, pH 7.0. Filter through a 0.45 ␮m membrane and degas. Store at ambient temperature and use within 2 weeks. 9. Column buffer B1: 50 mM Tris-HCl, 0.1% (w/v) CHAPS, pH 7.5. Filter through a 0.45 ␮m membrane and degas. Store at ambient temperature and use within 2 weeks. 10. Column buffer B2: 50 mM Tris-HCl, 1 M NaCl, 0.1% (w/v) CHAPS, pH 7.5. Filter through a 0.45 ␮m membrane and degas. Store at ambient temperature and use within 2 weeks.

3. Methods The expression of recombinant human ADAMTS enzymes requires the construction of expression vectors and hence a source of cDNA. While the production of custom cDNA libraries from which to isolate such sequences is quite feasible, the completion of the human genome sequencing project has allowed a much improved availability of cDNA. We obtained ADAMTS-1 and -4 clones from the HUGE (human unidentified gene-encoded) large protein database and an ADAMTS-5 clone from Neurocrine Biosciences (San Diego, CA, USA), but many of the ADAMTS sequences are now available as IMAGE clones. A wide variety of expression vectors are commercially available and should be selected depending on their suitability for the desired expression system (host cell type, constitutive/inducible expression, addition of protein tags, etc.), which itself should be selected based on the

Expression of Recombinant ADAMTS in Insect Cells

87

characteristics of the protein to be expressed. Native ADAMTS enzymes are disulphide-bonded, multi-domain proteins, which undergo post-translational modifications, processing and secretion. The expression of full-length ADAMTS in prokaryotic cells is therefore likely to be problematic, although the expression of individual functional domains is feasible (7). Based on previous experience we have used an insect cell-based expression system for these enzymes in combination with the insect expression vecR . This vector possesses the Opie-2 protor pIB/V5-His-TOPO moter, which drives constitutive expression of recombinant protein; an ampicillin resistance gene for selection within E. coli; a blasticidin resistance gene, enabling direct selection when genR cloning method, erating stable transfectants; and the TOPO which enables the direct ligation of DNA possessing single 3 adenine overhangs. We have used Trichoplusia ni High-FiveTM cells as the host because they are reported to express high levels of recombinant protein (8–10) (though this may depend on culture conditions (11)) and are cultured in serum-free medium, which is advantageous for subsequent purification of secreted proteins such as the ADAMTS. 3.1. Construction of Expression Vectors 3.1.1. Amplification of cDNA

1. The target-coding sequence is amplified from the template DNA by PCR. Specific primers flanking the coding sequence should be designed so that they possess an annealing temperature of 50–60◦ C and a GC content of <80%. For each reaction, PCR primers with similar annealing temperatures should be used. Inter-(dimer-forming) and intra(hairpin-forming) molecular sequence complementarity of primers should be avoided. Computer software programs such as Primer ExpressTM (Applied Biosystems) can aid primer design. The primer pairs used to amplify ADAMTS1, -4 and -5 are shown in Table 5.1 2. PCR reagents should be thawed, mixed thoroughly by vortexing and then placed on ice. 3. On ice, prepare a high-fidelity PCR reaction containing 1× Tli PCR buffer, 1 mM MgCl2 , 500 ␮M dNTP mix, 100 pmol forward primer, 100 pmol reverse primer, 1–5 ng template DNA, 2.5U Tli polymerase and sterile dH2 O to a final volume of 100 ␮L (Notes 1 and 2). 4. Incubate reactions at 95◦ C for 5 min in order to denature the template DNA then incubate for 30–40 cycles at 94◦ C for 1 min, the desired annealing temperature for 1 min, then 72◦ C for the desired extension time (allow 1 min per 1,000 bases of sequence to be amplified) to amplify DNA. The annealing temperatures and extension times used in the amplification of ADAMTS-1, -4 and -5 are indicated in Table 5.1.

88

Jones, Vankemmelbeke, and Buttle

Table 5.1 Primers used in the cloning of ADAMTS-1, -4 and -5 Primer

Stranda

5 -3 sequence

Sequence positionb

ADAMTS-1 forward

F

GACATGGGGAACGCGGAG

291–308

ADAMTS-1 reverse

R

CTTAACTGCAATCTGCCATTG

3147–3127

ADAMTS-4 forward

F

GCCATGTCCCAGACAGG

398–414

ADAMTS-4 reverse

R

GGTTATTTCCTGCCCGC

2916–2900

ADAMTS-5 forward

F

GAAGATCTGACTACAAGGACGACGATGAC

N/Ac

ADAMTS-5 reverse

R

CCTCTAGATTACTAACATTTCTTCAACAAGCATTG

2915–2892d





F = forward (sense) strand, R = reverse (anti-sense) strand. Position on ADAMTS-1 (EMBL:AF170084), ADAMTS-4 (EMBL:AF148213) or ADAMTS-5 (EMBL:AF142099) sequence. c This primer is complementary to the FLAGTM sequence and includes a BglII restriction site (AGATCT). d The first 11 bases of the primer are not complementary to the ADAMTS-5 sequence and includes a XbaI restriction site (TCTAGA). a

b

3.1.2. Purification of Amplified DNA by Agarose Gel Electrophoresis

1. These instructions are for use with horizontal gel tank systems. It is important that the comb and plate are clean and free from gel prior to starting. The agarose concentration of each gel should be chosen to maximize the separation of DNA at the predicted size of the target product. A rough guideline is that DNA of greater than 2 kbp should be separated on a 0.8% (w/v) gel, of 1–2 kbp on a 1% (w/v) gel, of 200 bp–1 kbp on a 1.5% (w/v) gel and of DNA <200 bp on a 2% (w/v) gel. 2. Prepare the gel by mixing agarose with 1× TAE buffer to the desired agarose concentration and dissolve by placing in a microwave at full power for 20 s. Mix carefully by swirling gently, then return to the microwave at full power for a further 10 s. If the agarose has dissolved then allow the solution to cool to approximately 50–60◦ C, otherwise continue to microwave the mixture for 10 s periods. 3. Add EtBr to 0.002% (w/v), mix by swirling then pour the gel. Insert the comb. Allow the gel to set fully at ambient temperature for 30 min. 4. Once the gel has set, remove the comb and submerge the gel in 1× TAE buffer.

Expression of Recombinant ADAMTS in Insect Cells

89

5. Add 5 ␮L of DNA loading dye to the PCR reaction. Load the PCR reaction into the gel, taking care not to overfill the well. Include one well of DNA molecular weight markers that span the expected DNA product size. 6. Complete the assembly of the gel unit and connect to a power supply. Run the gel at 30 mA until the loading dye has moved at least 5 cm across the gel. 7. Visualize the DNA within the gel by placing on a UV transilluminator. Always use UV protective eye-ware. 8. Using a scalpel, excise the DNA band of expected size, taking care to minimize the extent of the surrounding gel. Gel purify the DNA using the QIAquick Gel Extraction Kit (Promega), following the manufacturer’s instructions (Note 3). 3.1.3. Addition of 3  Adenine Overhangs to DNA

1. On ice, prepare a 3 A reaction mix containing 30 ␮L of gelpurified PCR product, 100 ␮M dATP, 1× Taq PCR buffer and 2.5U Taq polymerase into thin-walled PCR reaction tubes. 2. Incubate reactions at 72◦ C for 15 min.

3.1.4. Ligation of pIB/V5-His-TOPOR Vector and cDNA Insert

1. To 4 ␮L of the 3 A PCR product add 1 ␮L of salt solution (1.2 M NaCl, 60 ␮M MgCl2 ) and 1 ␮L of pIB/V5-HisTOPO vector. 2. Incubate at ambient temperature for 20 min, then place on ice.

3.1.5. Transformation of Chemically Competent E. coli (Note 4)

1. Prepare LB agar plates containing 50 ␮g/mL ampicillin prior to use. Plates can be stored at 4◦ C for a number of weeks. 2. Remove chemically competent TOP10 cells from –80◦ C storage and place on ice for approximately 10 min to allow gentle thawing. 3. Thaw SOC medium, then place on ice. Place LB agar plates containing 50 ␮g/mL ampicillin at 37◦ C, so that they are pre-warmed when required. 4. Add 2 ␮L of ligation reaction to thawed cells. Mix by gently tapping. Do not vortex or mix by pipetting as this will damage the cells. Incubate on ice for 20 min. 5. Transfer cells to a 42◦ C water bath for 30 s, then place on ice. 6. Add 250 ␮L of SOC to each aliquot of cells. Incubate cells at 37◦ C for 1 h in an end-over-end mixer or held horizontally in an orbital shaker (200 rpm).

90

Jones, Vankemmelbeke, and Buttle

7. Spread cultures onto pre-warmed LB agar plates and incubate at 37◦ C for 16–18 h. To ensure a good spacing of colonies spread a number of different culture volumes (10–100 ␮L should be sufficient) onto parallel plates. 3.1.6. Screening of Colonies for the Presence of Suitable Constructs

1. If ligation and transformation have been successful then colonies ought to be visible on the LB agar plate. Individual colonies should be picked and tested for the presence and correct orientation of the insert. In our experience screening of five colonies is usually sufficient to identify a desired clone although more colonies may have to be tested in specific cases. 2. Prepare LB broth containing 50 ␮g/mL ampicillin and aliquot 3 mL into each of the required number of 20 mL universal containers. Using a sterile 200 ␮L pipette tip carefully touch an individual colony, then transfer the tip to an aliquot of LB broth. Repeat for other colonies as required. Incubate cultures at 37◦ C for 16–18 h with shaking at 200 rpm. 3. Screen cultures for the presence and orientation of inserts by Taq polymerase PCR using the vector-specific Opie-2 forward primer and a target-specific reverse primer. Primers utilized for ADAMTS-1, -4 and -5 are indicated in Table 5.2. PCR reactions should be prepared so that each reaction contains 1× PCR buffer, 100 ␮M dNTP mix, 10 pmol forward primer, 10 pmol reverse primer, 1.5 mM MgCl2 , 1U Taq polymerase and 1 ␮L of E. coli culture. Reactions should be performed as described in Section 3.1.1 (Step 4). 4. Separate the PCR products by agarose gel electrophoresis as described in Section 3.1.2 and visualize the PCR products on a UV transilluminator. Cultures containing plasmid with an insert in the correct orientation will give bands of expected size. Prepare stocks of positive cultures by adding 450 ␮L of culture to a 1.5 mL tube containing 50 ␮L of 10× Hogness buffer. Store culture stocks at –80◦ C. Once positive cultures have been identified, the corresponding plasmid inserts should be sequenced to ensure that no errors have been introduced during the PCR amplification. It is advantageous to sequence both the sense and anti-sense strands as this increases the likelihood of useful sequence data within any one region. Primers for sequencing should be designed to span the entire insert utilizing one primer approximately every 500 bases. A primer within the vector sequence, immediately 5 to the insert should also be included to ascertain the 5 sequence and frame of the insert. The sequencing primers used for ADAMTS-1, -4 and -5 are indicated

Expression of Recombinant ADAMTS in Insect Cells

91

Table 5.2 Primers used in the sequencing of ADAMTS-1, -4 and -5 constructs Stranda

5 -3 sequence

Sequence positionb

ATS1Fseq1

F

GAGGCGTATTTCATCCAG

702–719

ATS1Fseq2

F

CAGTTAGCCTGGTGGTGG

1144–1161

ATS1Fseq3

F

CTGGATAATGGTCATGGGG

1605–1623

ATS1Fseq4

F

CAAACGAGTGCGCTACAG

2042–2059

ATS1Fseq5

F

CATCGAAGTGAAACAGCG

2492–2509

ATS1Fseq6

F

ACCAGACCTTGTGCAGAC

2943–2960

ATS1Rseq1

R

CACCACTAGCTCCTGGTC

428–411

ATS1Rseq2

R

CTTCGTCCTTCGTCTTCG

897–881

ATS1Rseq3

R

CATCAGCCATCCCAAGAG

1365–1348

ATS1Rseq4

R

GTTTGACACACCAGCACC

1810–1793

ATS1Rseq5

R

GGCTGCAAAACGAAGAAG

2266–2249

ATS1Rseq6

R

GGATGGTCAAGGGCTC

2712–2697

ATS4Fseq1

F

GGCGTGTTACAATATCGG

851–868

ATS4Fseq2

F

CGGACCCTGACCACTTTG

1314–1331

ATS4Fseq3

F

CACGCCATTGTCCACAG

1770–1786

ATS4Fseq4

F

CTGGGTTCCTCGCTACAC

2215–2232

ATS4Fseq5

F

CAGTCAGCTTGCGCTACAG

2667–2685

ATS4Rseq1

R

GCCGTTGAGCTTCTCTG

607–591

ATS4Rseq2

R

GCGGCCATCTTGTCATC

1095–1079

ATS4Rseq3

R

GCGAGAGGTGCTCAAAG

1564–1548

Primer ADAMTS-1

ADAMTS-4

ATS4Rseq4

R

CAGGTCCGAGAGCAGTC

2007–1991

ATS4Rseq5

R

CTTGCTGCAACCAGAACC

2461–2450

ATS5Fseq1

F

GGAGCCACTGCTTCTATCG

502–520

ATS5Fseq2

F

CAGGTGGAGCTGCTTCT

924–940

ATS5Fseq3

F

AGCACCAACACAACCAGC

1156–1173

ATS5Fseq4

F

CACGAAAGCAGATCCTGG

1549–1566

ATS5Fseq5

F

AGAGGGCCATCTACCGC

1948–1964

ATS5Fseq6

F

TACACTGACGTGGTGAGG

2353–2372

ATS5Fseq7

F

TGGGATCACACACTTCGC

2731–2748

ATS5Rseq1

R

GGAGTAGAGTTGGTCGATG

385–350

ATS5Rseq2

R

TAGACGTGCAGGATCCGT

700–683

ATS5Rseq3

R

GGCGGATGTGGTTCTCG

1059–1043

ATS5Rseq4

R

GATGGAAGACATTAAGCGC

1448–1430

ADAMTS-5

(continued)

92

Jones, Vankemmelbeke, and Buttle

Table 5.2 (continued) Primer

Stranda

5 -3 sequence

Sequence positionb

ATS5Rseq5

R

TGGCCCCAGGATCCCC

1855–1840

ATS5Rseq6

R

TCATACTGCAGCTTTGAGC

2278–2260

ATS5Rseq7

R

GTTTAGTGGGGTCTGTTG

2646–2628

OPIE2F

F

CCGCAACGATCTGGTAAACA

N/Ac

OPIE2R

R

ACATGACAATACAAACTAAG

N/Ac

pIB/V5-HisR VECTOR TOPO

F = forward (sense) strand, R = reverse (anti-sense) strand. Position on ADAMTS-1 (EMBL:AF170084), ADAMTS-4 (EMBL:AF148213) or ADAMTS-5 (EMBL:AF142099) sequence. R c This sequence is complementary to the pIB/V5-His-TOPO vector sequence. a

b

in Table 5.2. To generate template for the sequencing reactions purify plasmid from positive cultures using an appropriate method. We have found that the QIAGEN’s QIAprep Spin Miniprep Kit, used according to the manufacturer’s instructions, is a quick and reliable method of purifying plasmid. Sequence the purified plasmid DNA using the appropriate primers. 5. The sequence data should be analysed and compared to the expected sequence. To aid this analysis a number of programs available as part of the EMBOSS suite may be beneficial: returned sequence data can be assembled into a single sequence using the program “megamerger” (http://emboss.bioinformatics.nl/cgi-bin/emboss/megamerger) and aligned against the target sequence using the global alignment program “stretcher” (http://emboss. bioinformatics.nl/cgi-bin/emboss/stretcher) (12). 3.2. Transfection of High-FiveTM Cells with Expression Constructs 3.2.1. Preparation of Verified Expression Plasmids by CsCl Density Gradient Centrifugation (Note 5)

All HyQ-SFX medium should be supplemented with gentamycin to a concentration of 10 ␮g/mL. 1. Remove from –80◦ C storage the frozen culture stock transformed with the verified expression plasmid and thaw partially on ice. Add 10 ␮L of the stock to 10 mL of LB broth containing 50 ␮g/mL ampicillin. Incubate at 37◦ C with shaking at 200 rpm for 6 h (13). 2. Transfer the culture to a large (2–3 L) conical flask containing 500 mL of LB broth and 50 ␮g/mL ampicillin. Continue to incubate at 37◦ C with shaking for 12–18 h.

Expression of Recombinant ADAMTS in Insect Cells

93

3. Centrifuge the overnight culture at 3,800 × g and 4◦ C for 10 min. 4. Discard the supernatant and resuspend the pelleted culture in a total volume of 10 mL of lysis buffer. 5. Add 20 mL of 0.2 M NaOH, 1% (w/v) SDS solution, mix by inversion and incubate on ice for 5 min. 6. Add 15 mL of 3 M sodium acetate, pH 4.8, mix by inversion then incubate on ice for a further 10 min. 7. Centrifuge at 15,300 × g and 4◦ C for 10 min. 8. Transfer supernatant to a clean centrifuge tube and centrifuge at 15,300 × g and 4◦ C for 10 min. 9. Transfer supernatant to another clean centrifuge tube and add 120 mL of ethanol. Incubate at –80◦ C for 1.5 h. 10. Centrifuge at 15,300 × g and 4◦ C for 10 min. 11. Discard the supernatant and resuspend the pellet in 6 mL of STE buffer. To this add 80 ␮L of 1% (v/v) Triton X-100, 200 ␮L of 10 mg/mL EtBr and CsCl to a final density of 1.57–1.61 g/mL. 12. Transfer mixture to 3.9 mL polyallomer tubes and seal. 13. Centrifuge at 450,000 × g and 27◦ C for 4 h. 14. Carefully extract the DNA band from each tube by inserting a needle through the wall of each centrifuge tube and transfer to a 15 mL tube. 15. In the fume hood remove EtBr using H2 O-saturated butan-1-ol. Add 1 mL of butanol to the EtBr/DNA mixture, shake, then remove the top layer containing the butanol and dissolved EtBr. Repeat this process until the bottom layer (containing the DNA) becomes clear. 16. Transfer the DNA fraction to a clean tube and add three volumes of H2 O and eight volumes of ethanol. Mix by inversion then place at –80◦ C for 1 h. 17. Centrifuge at 15,000 × g and 4◦ C for 30 min. 18. Discard supernatant and resuspend the pellet in 400 ␮L of 0.3 M sodium acetate, pH 7.0. 19. Transfer to a 1.5 mL microfuge tube and add 1 mL of ethanol. Mix by inversion then incubate at ambient temperature until a precipitate develops. 20. Centrifuge at 18,000 × g and ambient temperature for 20 min. 21. Discard the supernatant and wash the pellet with 500 ␮L of 70% (v/v) ethanol.

94

Jones, Vankemmelbeke, and Buttle

22. Centrifuge at 18,000 × g and ambient temperature for 15 s. Remove the supernatant by aspiration. 23. Dry the pellet by centrifuging at 18,000 × g and ambient temperature for 3 min with the tube lid open. 24. Resuspend the DNA pellet in 250 ␮L of dH2 O by incubating at ambient temperature for 1 h. Store the DNA at –20◦ C. 3.2.2. Culture of High-FiveTM Insect Cells

1. To initiate a culture from frozen stocks thaw a 1 mL vial of frozen cells in a 37◦ C water bath. Just before the cells have completely thawed, transfer them into a T25 flask containing 4 mL of HyQ-SFX culture medium at ambient temperature. Incubate the new culture at ambient temperature for 15–30 min to allow the cells to attach. Decant the medium from the attached cells and discard. Add 5 mL of fresh medium to the cells and incubate as normal. 2. Culture High-FiveTM cells at 27◦ C in a refrigerating incubator. The cells will require passaging every 3–4 days as they approach confluence. To passage, detach the cells by firmly tapping the culture flask, then add the cell suspension to fresh HyQ-SFX culture medium at a ratio of 1:3.

3.2.3. Transfection of High-FiveTM Insect Cells with ADAMTS Expression Plasmid

1. Resuspend High-FiveTM cells by firmly tapping the culture flask and determine the cell density using a haemocytometer. 2. Seed the cells into 12-well culture plates at a density of 1.5 × 105 cells/well and add fresh HyQ-SFX medium to a final volume of 1.1 mL. 3. Incubate the cells at 27◦ C for 20 min. 4. Prepare a transfection mixture containing 1 ␮g/mL purified plasmid and 2% (v/v) CellfectinTM reagent in fresh Hyq-SFX culture medium. Mix by inversion for 20 s, then incubate at ambient temperature for 15 min. 5. Carefully remove and discard the medium from plated cells by pipetting. Carefully add 120 ␮L of transfection mixture per well in a dropwise manner across the entire well surface. 6. Incubate the transfected cell cultures at ambient temperature for 2 h on a rocking table at 1 tilt/min. 7. Add 500 ␮L of fresh culture medium to each transfection well and incubate the cultures at 27◦ C. 8. Expression of the recombinant enzyme by transfected cells can be tested at the mRNA level by RT-PCR or at the protein level by Western blotting using appropriate antibodies or by using an appropriate activity assay. For instance, aggrecanase activity within the culture supernatant can be assayed using a polyacrylamide-entrapped aggrecan particle

Expression of Recombinant ADAMTS in Insect Cells

95

assay (14). We found that there was little or no activity when assaying culture supernatants containing ADAMTS-4 and -5 but that activity could be recovered after dialysis of the supernatants against assay buffer. 3.2.4. Generation of Stable pIB/V5-His Transfectants

1. Four to five days post-transfection remove medium from cells and replace with 750 ␮L of culture medium containing 50 ␮g/mL blasticidin per well. Incubate at 27◦ C. 2. Replace the medium with fresh selection medium containing 50 ␮g/mL blasticidin every 3–4 days until discrete colonies begin to grow out. 3. To select individual transfected colonies aliquot 100 ␮L of fresh blasticidin-free culture medium per well of a 96-well culture plate (Note 6). Using a sterile pipette tip, pick a single colony and inoculate a culture well with it. Repeat for other colonies. Incubate the cells at ambient temperature for approximately 20 min to allow them to attach. Remove and discard the medium and replace with 100 ␮L of culture medium containing 50 ␮g/mL blasticidin. 4. Continue to culture transfectants as before in medium containing 50 ␮g/mL blasticidin, sub-cloning individual clones into larger culture vessels. Once transfectants are growing at a similar rate to non-transfected cells, the blasticidin concentration within the selection medium can be reduced to 10 ␮g/mL.

3.2.5. Generation of Frozen Stocks of Stable pIB/V5-His Transfectants

3.3. Purification of Recombinant ADAMTS Enzymes

1. Resuspend cells by firmly tapping the culture flask. Estimate the cell density using a haemocytometer. 2. Centrifuge cells at 1,000 × g for 5 min. Decant and retain the supernatant. Prepare a freezing solution of 42.5% (v/v) supernatant, 42.5% (v/v) fresh medium, 10% (v/v) DMSO and 5% (v/v) FBS. Resuspend the cells in freezing solution to a density of 3 × 106 /mL. Prepare 1 mL aliquots into cryovials. Place the aliquots into an isopropanol freezing bath and place at –80◦ C for 24 h. Transfer the frozen aliquots to liquid N2 for long-term storage. We have purified ADAMTS-1, -4 and -5 from transfected HighFiveTM cells using a common purification procedure which is described in (6). In order to simplify this procedure purification was undertaken on the conditioned medium of transfected cell lines. Protein was present within the non-soluble (cell and ECM) fractions but the proportion of recombinant protein within these has not been determined. Other ADAMTS enzymes may require modification of this purification scheme. A method of identifying the ADAMTS of interest within samples is required, such as

96

Jones, Vankemmelbeke, and Buttle

by Western blotting or through assay of activity. In the identification of the aggrecanolytic ADAMTS enzymes, we have found the polyacrylamide-entrapped aggrecan activity assay a convenient method (14). 1. Collect culture supernatant from transfected cells during passaging and store at –80◦ C until required. 2. Equilibrate a preparative Sephadex G-25 column with buffer A1 by running 5 column volumes (CV) of buffer over the column using a flow rate of 100–150 cm/h (Note 7). 3. Thaw culture supernatant and add 3,4-DCI and E-64 to final concentrations of 50 and 10 ␮M to inhibit endogenous serine and cysteine proteinase activities, respectively. Pass the medium through a 0.45 ␮m filter and degas by keeping the filtered medium under a partial vacuum for 30 min. 4. Using a flow rate of 100–150 cm/h apply the filtered medium to the G25 column and perform an isocratic elution with buffer A1. Collect the protein fraction which elutes in the excluded volume of the column. Proteincontaining fractions can be identified by measuring the absorbance of the eluate at 280 nm, using either an online or post-elution spectrophotometer. To ensure separation of the protein and small solute components of the medium apply to the column a sample volume of less than 20% of the column volume. 5. Equilibrate a heparin-sepharose fast-flow column with buffer A1 by running 5 CV of buffer over the column using a flow rate of 100 cm/h. 6. Apply the protein fraction to the heparin column. Elute the bound components with a stepwise NaCl gradient of 0.15, 0.5 and 1.0 M NaCl by passing over the column 85:15, 50:50 and 0:100 mixtures of buffers A1 and A2, respectively. Use 5 CV of buffer per elution step. Identify fractions containing protein (by Western blotting) or activity (activity assay) as required. Fractions containing ADAMTS protein/activity should be pooled. In our experience ADAMTS-1, -4 and -5 elute in the 0.5–1.0 M NaCl step. 7. Equilibrate a preparative Sephadex G-25 column with buffer B1 as described in point 2. 8. Apply the pooled ADAMTS fraction from point 6 to the G-25 column and perform an isocratic elution using buffer B1 as described in point 4. Collect the protein fraction from the column.

Expression of Recombinant ADAMTS in Insect Cells

97

9. Equilibrate a Mono Q column with buffer B1 by passing 5 CV of buffer over the column using a flow rate of 30 cm/h. 10. Apply the protein fraction from point 8 to the Mono Q column and elute unbound protein with buffer B1. Elute bound protein using a linear NaCl gradient of 50 mM NaCl/CV between 0 and 1 M NaCl, starting with buffer B1 and using buffer B2 to make the gradient. The presence of ADAMTS within fractions should be assessed and ADAMTS-containing fractions pooled. In our experience ADAMTS-1, -4 and -5 elute between 0.2 and 0.45 M NaCl. 11. In our experience this method results in a purification of approximately two orders of magnitude, determined by measurement of specific activity.

4. Notes 1. Alternative high-fidelity polymerases such as Pfu are available and may be substituted. 2. High-fidelity “proof-reading” polymerases such as Tli and Pfu possess a 3 –5 exonuclease activity capable of digesting primers within the PCR reaction. The PCR reactions should, therefore, be assembled on ice with the polymerases added last. Alternatively, the polymerase can be added after the initial 95◦ C denaturation step. 3. Any suitable method of purifying DNA from agarose gels may be used. 4. Any suitable chemically competent cell strain may be used for transformation. 5. A number of commercial kits are available that generate plasmid preparations of suitable scale and purity for subsequent transfection reactions, and often have advantages of speed and lower g centrifugation requirements. 6. Blasticidin interferes with the attachment of High-FiveTM cells to the culture plastic. Passaging should, therefore, be performed in blasticidin-free conditions. Once attachment has occurred (usually after 15 min) blasticidin can be added to the culture medium. 7. Flow rates indicated are within the manufacturer’s recommended rates for the respective media and prevent significant compression of the constituent beads. Rates are defined as flow over the column height (cm) per hour (h) and

98

Jones, Vankemmelbeke, and Buttle

therefore the diameter of the column will determine the rate in mL/min. For example, to achieve a flow rate of 300 cm/h using a TricornTM Mono Q 5/50 GL (GE Healthcare Life Sciences: bed volume = 1 mL; diameter = 5 mm; height = 50 mm), a flow rate of 1 mL/min is required (1 mL = 1 CV, 50 mm column height. Therefore, 1 mL/min = 50 mm/min = 300 cm/h). References 1. Huxley-Jones, J., Apte, S. S., Robertson, D. L., and Boot-Handford, R. P. (2005) The characterisation of six ADAMTS proteases in the basal chordate Ciona intestinalis provides new insights into the vertebrate ADAMTS family. Int J Biochem Cell Biol 37, 1838–1845. 2. Rawlings, N. D., Morton, F. R. and Barrett, A. J. (2006) MEROPS: the peptidase database. Nucleic Acids Res 34, D270–D272. 3. Jones, G. C. and Riley, G. P. (2005) ADAMTS proteinases: a multi-domain, multi-functional family with roles in extracellular matrix turnover and arthritis. Arthritis Res Ther 7, 160–169. 4. Flannery, C. R., Zeng, W., Corcoran, C., Collins-Racie, L. A., Chockalingam, P. S., Hebert, T., Mackie, S. A., McDonagh, T., Crawford, T. K., Tomkinson, K. N., LaVallie, E. R., and Morris, E. A. (2002) Autocatalytic cleavage of ADAMTS-4 (Aggrecanase-1) reveals multiple glycosaminoglycan-binding sites. J Biol Chem 277, 42775–42780. 5. Kuno, K. and Matsushima, K. (1998) ADAMTS-1 protein anchors at the extracellular matrix through the thrombospondin type I motifs and its spacing region. J Biol Chem 273, 13912–13917. 6. Vankemmelbeke, M. N., Jones, G. C., Fowles, C., Ilic, M. Z., Handley, C. J., Day, A. J., Knight, C. G., Mort, J. S., and Buttle, D. J. (2003) Selective inhibition of ADAMTS-1, -4 and -5 by catechin gallate esters. Eur J Biochem 270, 2394–2403. 7. Liu, C. J., Kong, W., Xu, K., Luan, Y., Ilalov, K., Sehgal, B., Yu, S., Howell, R. D., and Di Cesare, P. E. (2006) ADAMTS-12 associates

8.

9.

10.

11.

12. 13.

14.

with and degrades cartilage oligomeric matrix protein. J Biol Chem 281, 15800–15808. Davis, T. R., Wickham, T. J., McKenna, K. A., Granados, R. R., Shuler, M. L., and Wood, H. A. (1993) Comparative recombinant protein production of eight insect cell lines. In Vitro Cell Dev Biol Anim 29A, 388–390. Wickham, T. J., Davis, T., Granados, R. R., Shuler, M. L., and Wood, H. A. (1992) Screening of insect cell lines for the production of recombinant proteins and infectious virus in the baculovirus expression system. Biotechnol Prog 8, 391–396. Wickham, T. J. and Nemerow, G. R. (1993) Optimization of growth methods and recombinant protein production in BTI-Tn-5B1-4 insect cells using the baculovirus expression system. Biotechnol Prog 9, 25–30. Rhiel, M., Mitchell-Logean, C. M., and Murhammer, D. W. (1997) Comparison of Trichoplusia ni BTI-Tn-5B1-4 (high fivetrade mark) and Spodoptera frugiperda Sf-9 insect cell line metabolism in suspension cultures. Biotechnol Bioeng 55, 909–920. Myers, E. W. and Miller, W. (1988) Optimal alignments in linear space. Comput Appl Biosci 4, 11–17. Sambrook, J. and Russell, D. W. (2001). Purification of closed circular DNA by equilibrium centrifugation in CsCl-ethidium bromide gradients: continuous gradients. In Sambrook, J. and Russell, D. W. (eds.), Molecular cloning: a laboratory manual. New York: Cold Spring Harbour Laboratory Press, pp. 1.65–1.69. Handley, C. J. and Buttle, D. J. (1995) Assay of proteoglycan degradation. Methods Enzymol 248, 47–58.

Chapter 6 Expression and Purification of Membrane-Type MMPs Jing Nie and Duanqing Pei Abstract Matrix metalloproteinases (MMPs) modulate cell–cell and cell–matrix interactions and the membranetype MMPs (MT-MMPs) are especially important for pericellular proteolysis. Understanding the activity, regulation, and function of MT-MMPs would provide important insight for ongoing research in many diseases including cancer, fibrosis, and autoimmune diseases. In this chapter, we introduce the methods to express and purify recombinant MT-MMPs using Escherichia coli or mammalian cells, as well as how to detect the activity of MT-MMPs in proMMP-2 activation using gelatin zymography. Key words: Membrane-type MMPs, Ni-NTA, anti-FLAG column, calcium phosphate precipitation, zymograms.

1. Introduction MT-MMPs regulate extracellular matrix turnover and signaling pathways in multiple physiological and pathological processes such as adipogenesis, bone remodeling, tumor progression, wound healing, and fibrosis (1, 2). For the characterization of enzyme activity and substrate specificity of MT-MMPs in vitro, it is important to produce recombinant enzyme in large quantities. We have successfully expressed the catalytic domains of multiple His-tagged MTMMPs in Escherichia coli. We chose the pET15b vector as the expression vector because it provides a 6xHis-tag at the N terminus of the recombinant protein and allows a simple purification with a Ni-NTA column. The cDNAs were subcloned into a modified pET15b vector followed by transformation of the plasmid I.M. Clark (ed.), D.A. Young, A.D. Rowan (Consulting editors), Matrix Metalloproteinase Protocols, Methods in Molecular Biology 622, DOI 10.1007/978-1-60327-299-5 6, © Springer Science+Business Media, LLC 2001, 2010

99

100

Nie and Pei

DNA into the BL21 host for the induction of protein production. After solubilization, refolding, and purification, these enzymes exhibited catalytic activity against both small peptide and protein substrates (3, 4). These recombinant proteins have been used for kinetic analyses of catalytic activity (3, 4) and also for polyclonal antibody production in hamsters or rabbits. To investigate the role of MT-MMPs in cell function, e.g., migration and invasion, we generated stable expression cell lines (4–6). These cell lines also allow the study of subcellular localization and trafficking of MT-MMPs (7, 8). To produce recombinant protein with correct folding and posttranslational modifications, we purified a FLAG-tagged soluble form of MT-MMP protein from these stable lines using a M2 column (9). The method is described below. These proteins provide an opportunity to verify the protein interaction and the substrate specificity in vitro. Transient expression of MT-MMPs is a quick and simple method for the investigation of enzyme function ex vivo (10). There are an increasing number of transfection methods available including calcium phosphate precipitation, liposomal reagents, and electroporation. Described below is a protocol for the transient transfection into HEK293 cells using calcium phosphate precipitation. One of the important functions of MT-MMPs is to activate proMMP-2 (11). Detection of proMMP-2 activation by gelatin zymography provides a simple and straightforward way to assess the activity of MT-MMPs in the cells. Briefly, conditioned medium is electrophoresed in an SDS-polyacrylamide gel containing gelatin followed by protease activation and proteolysis. To detect the activity of MT1-MMP, MT3-MMP, and MT5MMP, recombinant proMMP-2 or 5% fetal bovine serum may be the source of proMMP-2. For MT-MMPs of lower activity against proMMP-2, it is preferable to express proMMP-2 in cells (4).

2. Materials 2.1. E. coli Expression of the Catalytic Domain of MT-MMPs

1. pET15b expression vector (Novagen, Gibbstown, NJ), store at –20◦ C. 2. pfu polymerase (Stratagene, La Jolla, CA), store at –20◦ C. 3. E. coli DH5␣ for DNA cloning and analysis, store at –80◦ C. 4. E. coli BL21 (DE3) pLysS (Novagen) for expression of protein, store at –80◦ C. 5. 2 × YT medium: 16 g/L tryptone, 10 g/L yeast extract, 5 g/L NaCl, pH 7.4, autoclaved, and store at 4◦ C.

Expression and Purification of Membrane-Type MMPs

101

6. Ampicillin: 100 mg/mL (stock solution) in water, store in aliquots at –80◦ C. 7. 0.1 M isopropyl ␤-D-1-thiogalactopyranoside (IPTG), aliquot and store at –20◦ C. 8. LB (Lysogeny broth) medium, store at 4◦ C. 9. SDS-polyacrylamide gel. 10. 50 mL conical tubes. 11. Sonicator. 12. Inclusion body wash buffer: 50 mM Tris-HCl, pH 7.6, 5 mM CaCl2 , 1 ␮M ZnCl2 , 0.1% Triton X-100, store at 4◦ C. 13. 8 M urea 14. Binding buffer: 5 mM imidazole, 0.5 M NaCl, 20 mM Tris-HCl, pH 7.9. 15. Washing buffer: 60 mM imidazole, 0.5 M NaCl, 20 mM Tris-HCl, pH 7.9. 16. Elution buffer: 1 M imidazole, 0.5 M NaCl, 20 mM TrisHCl, pH 7.9. 17. Strip buffer: 100 mM EDTA, 0.5 M NaCl, 20 mM TrisHCl, pH 7.9. 18. Charging buffer: 50 mM NiSO4 . 19. 2 mL Ni-NTA His-bind resins, store at 4◦ C. 20. TBS: 150 mM NaCl, 50 mM Tris-HCl, 5 mM CaCl2 , pH 7.5. 21. 5 mM CaCl2 , 25 mM Tris-HCl, 150 mM NaCl, pH 7.5. 22. 0.03% (w/v) NaN3 in TBS. 2.2. Expression and Purification of MT-MMPs in Mammalian Cells 2.2.1. Production of Stable Cell Lines Expressing MT-MMP

1. pfu polymerase (Stratagene, La Jolla, CA), store at –20◦ C. 2. pCR3.1uni expression vector (Invitrogen), store at –20◦ C. 3. E. coli DH5␣ for DNA cloning and analysis, store at –80◦ C. 4. LB medium, store at 4◦ C. 5. Cell line MDCK, store in liquid N2 . 6. Dulbecco’s modified Eagle’s medium (DMEM), store at 4ºC. 7. Fetal bovine serum (FBS), heat inactivated 56ºC, 30 min. Aliquot and store at –20ºC. 8. Penicillin/streptomycin: final concentrations, 100 IU/mL penicillin, 50 mg/mL streptomycin (Sigma), store at 4◦ C. 9. Growth medium: 10% FBS, DMEM with glutamine (2 mM), penicillin (100 IU/mL), streptomycin (50 mg/mL), store at 4◦ C.

102

Nie and Pei

10. Trypsin-EDTA in Hank’s Balanced Salt Solution: 0.05% (w/v) trypsin, 0.53 mM EDTA, without Ca2+ , store at 4ºC. 11. 6- and 24-well culture plates, 10 cm cell culture dishes. 12. Lipofectamine, Lipofectamine 2000, or other liposomal transfection reagents, store at 4ºC. 13. G418 (Geneticin), aliquot and store at –20◦ C. 14. Anti-FLAG M2 antibody (Sigma), aliquot and store at –20◦ C. 2.2.2. Purification of MT-MMP Using M2 Column

1. PBS: 10 mM sodium phosphate, 2.7 mM potassium chloride, 137 mM sodium chloride, pH 7.4. 2. 150 mm cell culture dishes. 3. Q-Sepharose resin. 4. Peristaltic pump, fraction collector, and gradient maker. 5. Starting buffer: 0.3 M NaCl in TBS. 6. Final buffer: 1.2 M NaCl in TBS. 7. Anti-FLAGM2 affinity resin (Sigma, St. Louis, MO) in buffered aqueous glycerol solution, store at −20◦ C freezer. 8. 0.1 M glycine-HCl, pH 3.5. 9. TBS: 150 mM NaCl, 50 mM Tris-HCl, 5 mM CaCl2 , pH 7.5. 10. TBS/Brij35: 150 mM NaCl, 50 mM Tris-HCl, 5 mM CaCl2 , 0.1% (v/v) Brij35, pH 7.5. 11. 100 mg/mL FLAG peptide (Sigma) in TBS. 12. Storage buffer: 50% glycerol, 150 mM NaCl, 50 mM TrisHCl, 0.02% (w/v) NaN3 .

2.3. Transient Expression in HEK293 Cells 2.3.1. Subculture of HEK293 Cells for Transfection

1. Cell line HEK293, store in liquid N2 . 2. DMEM, FBS, Section 2.2).

growth

medium,

and

trypsin

(see

3. 0.001% (w/v) poly-L -lysine, filter sterilized, and store at 4◦ C. 4. 24-well culture plates

2.3.2. Transfection

1. Plasmid DNA: MT-MMP in pCR3.1uni expression vector, pCR3.1uni vector only. Store at –20◦ C. 2. 2 × HBS (2 × Hepes-buffered saline): 8.0 g NaC1, 0.37 g KCl, 0.2 g Na2 HPO4 ·7H2 O, 5.0 g HEPES; dissolve in 400 mL distilled H2 O, adjust pH to 7.05 using NaOH, make up volume to 500 mL. Filter sterilize, aliquot into sterile tubes, and store at –20◦ C. Keep one working tube at 4◦ C (see Note 1).

Expression and Purification of Membrane-Type MMPs

103

3. 2 M CaC12 . Filter sterile and store at 4◦ C. 4. Filter-sterilized distilled H2 O. 2.4. Detection of MT-MMP Activity in ProMMP-2 Activation by Gelatin Zymography 2.4.1. ProMMP-2 Activation with ProMMP-2 Cotransfection

1. ProMMP-2 expression vector, store at –20◦ C. 2. DMEM (see Section 2.2). 3. Gelatin stock solution: 10 mg/mL in H2 O, aliquot and store at –20◦ C. 4. Zymogram gel: 7.5–8% SDS-polyacrylamide gel with 1 mg/mL gelatin incorporated (see Note 2). 5. Laemmli sample buffer. 6. 2.5% (v/v) Triton X-100. 7. Zymogram developing buffer: 50 mM Tris-HCl, pH 7.6, 5 mM CaCl2 , 1 ␮M ZnCl2 , 0.1% (v/v) Triton X-100. 8. Zymogram staining buffer: 0.1% Coomassie blue, 50% methanol, 10% acetic acid. 9. 10% (v/v) methanol, 10% (v/v) acetic acid.

2.4.2. ProMMP-2 Activation with ProMMP-2 Cotransfection

1. Recombinant human proMMP-2, aliquot and store at –20◦ C. 2. 5% FBS and DMEM: 5% FBS, DMEM with glutamine (2 mM), penicillin (100 IU/mL), streptomycin (50 mg/mL), store at 4◦ C.

3. Methods 3.1. E. coli Expression of the Catalytic Domain of MT-MMPs

1. We modified the pET15b vector by inserting a palindromic sequence, GCCCGGGC, into the XhoI site repaired with Klenow polymerase. The resulting vector, pET15bS, contains a SmaI site in which a blunt-end fragment can be inserted in frame with an NH2 -terminal His-tag. 2. Amplify the catalytic domain or the ectodomains of the desired MT-MMP using polymerase chain reaction (PCR) with pfu polymerase and insert into pET15bS. The resulting clones are screened by restriction analysis and amplified using E. coli DH5␣ host cells. 3. To test the expression of clones: Transform 3–5 error-free plasmid DNAs into E. coli BL21 (DE3) pLysS for expression. Most of the His-tagged protein will accumulate in inclusion bodies. 4. Pick a single colony into 10 mL of 2 × YT medium containing ampicillin. Incubate at 37◦ C with shaking at 250 rpm until A600 is about 0.8 (0.6–0.9) (see Note 3).

104

Nie and Pei

5. Add IPTG to 0.4 mM and incubate at 37◦ C with shaking at 250 rpm for 4–5 h (see Note 3). 6. Remove 100 ␮L of culture and centrifuge for 1 min in a microcentrifuge at full speed. 7. Resuspend the cell pellet in 10 ␮L of 1 × SDS sample buffer and analyze by SDS-polyacrylamide gel. Protein can be visualized by Coomassie blue staining and the recombinant protein may be confirmed by Western blotting using His-tag antibody. 8. Larger scale expression: Start with a fresh transformed colony and inoculate into 5 mL 2 × YT medium containing ampicillin. Incubate at 37◦ C until slightly turbid. 9. Transfer the cells to 100 mL fresh pre-warmed 2 × YT medium, incubate at 37◦ C with shaking at 250 rpm for 2–4 h or until A600 is about 0.8 (0.6–0.9). Add IPTG for a final concentration of 0.4 mM and continue to incubate at 37◦ C for 5–6 h (see Note 3). 10. Transfer the culture to 50 mL conical tubes and harvest the cells by centrifugation at approximately 5,000 × g for 10 min at 4◦ C. 11. Add 1 mL inclusion body wash buffer for each 50 mL original culture solution. Resuspend pellet by sonication for 0.5–1 min with 50% pulse per round on ice. The resuspended mixture can be aliquoted and stored at –80◦ C for several months (see Note 4). 12. Purification of inclusion body: Centrifuge the mixture at approximately 30,000 × g for 5 min at 4◦ C, remove supernatants, and then add the same volume of inclusion body wash buffer to the pellet. Sonicate to resuspend the pellet. Repeat about 2–4 times until the supernatant is clear and the pellet is white. 13. Solubilization of inclusion body: Dissolve the pellet in 8 M urea. Then centrifuge the mixture at approximately 30,000 × g for 5 min at 4◦ C (see Note 5). 14. Protein refolding: Dilute the supernatant in binding buffer at 1:10 ratio, mix well. Centrifuge the mixture at approximately 30,000 × g for 10 min to clear debris. 15. Affinity purification using Ni-NTA column (2 mL resin): • Wash column: H2 O, 8 mL; charging buffer, 8 mL; binding buffer, 8 mL • Load sample to the column • Wash column: binding buffer, 8 mL; washing buffer, 8 mL • Elution: elute stepwise with 1 mL elution buffer, repeat 4–7 times

Expression and Purification of Membrane-Type MMPs

105

• Strip column: striping buffer, 8 mL • Store column in 0.03% NaN3 solution[L1] 16. Gel monitoring: Analyze 1–5 ␮L of each elution fraction by SDS-PAGE. The His-tagged protein is usually in fractions 2–3. 17. Dialyze protein positive fraction twice against 5 mM CaCl2 , 150 mM NaCl, 25 mM Tris-HCl pH 7.5 at 4◦ C to remove the imidazole. 18. The purified enzyme is very concentrated and active. Since MMPs tend to auto-degrade, purified enzyme needs to be aliquoted and stored at –80◦ C. 3.2. Expression and Purification of MT-MMPs from Mammalian Stable Line Cells

1. Amplify the sequence encoding the extracellular domains of the MT-MMP by high-fidelity PCR with pfu polymerase. To facilitate detection and purification with FLAG epitope and M2 column, the 3 primer contains the FLAG sequence (see Note 6).

3.2.1. Production of Stable Cell Lines Expressing MT-MMP

2. Clone the PCR fragment into pCR3.1uni expression vector. Identify transformed colonies of E. coli DH5␣ containing the subcloned cDNA in the appropriate orientation by restriction enzyme analysis and purify plasmid DNA. 3. Grow MDCK cells to approximately 75% confluency in 5–10% FBS in DMEM. 4. Transfect the correct expression vector into MDCK cells by liposome-mediated transfection and incubate cells at 37◦ C overnight. 5. Trypsinize the cells and resuspend in growth medium supplemented with G418. Plate cells at selection density (105 per 10 cm dish). Change medium every 4–5 days to provide enough nutrient and remove dead cells. Maintain the G418 selection in the medium (0.4–1.0 mg/mL) to ensure expression of the transfected gene during establishment of the stable cell lines. 6. After 2–3 weeks, resistant clones emerge. Collect clones of cells by trypsinization and seed into 24-well plates. Expand the individual clones and screen for MMP production by Western blot using anti-FLAG antibody or zymography (see Section 3.4.2). Positive clones are then frozen at –80◦ C and used for further characterization.

3.2.2. Purification of MT-MMP Using M2 Column

1. Expand stable clones that have a relatively high level of expression for MT-MMP. Grow cells with full medium supplemented with 0.4–1.0 mg/mL G418 in 15 cm dishes to 90% confluency (see Note 7).

106

Nie and Pei

2. Wash cells with PBS three times and supplement with serum-free DMEM medium with or without G418. 3. Forty-eight hours later, harvest conditioned medium and centrifuge at 5,000×g for 30 min at 2–8◦ C to clear debris (see Note 6). 4. Load conditioned medium onto a Q-Sepharose column (2 × 20 cm) at 4◦ C. Connect one end of the column to a fraction collector via a peristaltic pump and the other end to a gradient maker. Use equal volume of starting buffer and final buffer to achieve the elution gradient of 0.15– 1.0 M NaCl. Elute protein at 4◦ C. 5. Analyze elution fractions and identify positive fractions by Western blotting, pool together, and store at –80◦ C. 6. Pack 2 mL of anti-FLAG M2 affinity resin in an empty column with TBS. Allow the gel bed to drain and wash with TBS several times. 7. Wash the gel by loading three sequential column volumes of 0.1 M glycine-HCl, pH 3.5. Avoid disturbing the gel bed while loading. Let each aliquot drain completely before adding the next (see Note 8). 8. Wash the resin with five column volumes of TBS to equilibrate the resin for use. Do not let the bed run dry. Allow a small amount of buffer to remain on the top of the column. 9. Load the Q-Sepharose fractions onto the M2 column twice to maximize the binding at 4◦ C. 10. Wash the column with 10–20 column volumes of TBS/Brij35 to remove unbound materials. Allow the column to drain completely. 11. Elute MT-MMP-Flag by competitive elution with 5 one column volumes of 100 mg/mL FLAG peptide in TBS. 12. Regenerate the column by washing with three column volumes of 0.1 M glycine-HCI, pH 3.5. Equilibrate the column in TBS until the effluent is at neutral pH. Wash the column with ten column volumes of storage buffer and store at 2–8 or –20◦ C without draining (see Note 8). 13. Analyze the elution fractions by SDS-PAGE, Western blotting, or zymography (see Section 3.4.2). 3.3. Transient Expression of MT-MMPs in HEK293 Cells 3.3.1. Subculture of HEK293 Cells for Transfection

1. Revive HEK293 cells from liquid N2 and grow to 60–80% confluency in growth medium. 2. Day 1: Coat 24-well plates with 0.001% poly-L-lysine for 10–15 min. Wash cells once in PBS. Add pre-warmed trypsin-EDTA (1 mL) to one confluent 10 cm cell culture dish of cells and incubate 1–3 min at 37◦ C until the cells

Expression and Purification of Membrane-Type MMPs

107

lift off. Add 4 mL of growth medium and spin down the cells for 5 min at 500 × g at 4◦ C. Resuspend the cell pellet in growth medium and count the cells in a hemocytometer. Remove 0.001% poly-L-lysine and seed the cells at 105 cells per well for 24-well plates. Incubate at 37◦ C, 5% CO2 . 3. Day 2: Change to fresh growth medium (500 ␮L/well) 2–4 h before transfection. The cells are now ready for transfection. 3.3.2. Transfection

The following protocol is for one well in a 24-well plate: 1. Prepare the DNA–calcium phosphate precipitates: Mix 1 ␮g plasmid DNA with 3.1 ␮L 2 M CaCl2 and sterile water so that the final volume is 25 ␮L, mix well by pipetting. 2. Add an equal volume (25 ␮L) of 2 × HBS dropwise to the DNA/CaCl2 mix and gently agitate the mixture (see Note 9). 3. Within 5 min, add the DNA mixture dropwise onto the cells in the growth medium; make sure to evenly sprinkle the droplet over the entire area (50 ␮L per well) (see Note 10). 4. Gently return the cells to the incubator and leave undisturbed at 37◦ C, 5% CO2 for 5–6 h. 5. Wash the cells briefly with DMEM and gently add 0.5 mL of fresh growth medium. Incubate for 24 h at 37◦ C, 5% CO2 . The cells are now ready for testing the expression of MTMMP.

3.4. Detection of MT-MMP Activity in ProMMP-2 Activation by Gelatin Zymogram 3.4.1. ProMMP-2 Activation with ProMMP-2 Cotransfection

1. Follow the transfection protocol in Section 3.2.2. For cotransfection, include 20 ng proMMP-2 expression vector with each 1 ␮g of MT-MMP expression vector. The ratio of proMMP-2 expression vector to MT-MMP expression vector should be ∼1:50. 2. Remove growth medium and wash the cells three times carefully with DMEM. Add 300 ␮l DMEM per well (24well plates). Incubate the cells for 16–24 h at 37◦ C, 5% CO2 . 3. Harvest the conditioned medium and centrifuge at 1,200 × g for 10 min at 4◦ C to clear debris. Samples may be stored at –20◦ C. 4. Prepare 7.5–8% SDS-polyacrylamide gels: Add gelatin stock solution to reach the concentration of 1 mg/mL. 5. Prepare 5–12 ␮L samples of cell supernatants with Laemmli sample buffer. 6. Electrophorese under non-reducing conditions at 4◦ C. 7. Wash gel in 2.5% Triton X-100 for 2 × 15 min to remove SDS.

108

Nie and Pei

8. Change buffer to zymogram developing buffer and incubate gel at 37◦ C for 4–24 h (see Note 9). 9. Stain with zymogram staining buffer for 30 min. 10. Destain in 10% (v/v) methanol, 10% (v/v) acetic acid using several changes over 1–2 h. 11. Data analysis: Coomassie blue stains gelatin in the zymogram gel. Gelatin degradation by proteases allows the diffusion of small peptides out of the gel and leaves the area unstained. Thus, the clear bands in the blue-stained gel indicate the size of the proteases. SDS partially unfolds MMPs and relieves the inhibitory effect of the prodomain on the catalytic domain. This is why MMP-2 may be detected in the pro, intermediate, and active forms at 66, 62, and 59 kDa, respectively. 3.4.2. ProMMP-2 Activation Using Recombinant ProMMP-2

1. Grow HEK293cells to 80–90% confluency. 2. Follow the transfection protocol in Section 3.2.2. 3. Remove growth medium and carefully wash the cells two to three times with DMEM. For each well in 24-well plates, add 300 ␮L 5% FBS in DMEM or DMEM-containing 4 nM recombinant proMMP-2. Incubate the cells for 16–24 h at 37◦ C, 5% CO2 . (see Note 11) 4. Continue from Step 3 of Section 3.4.1.

4. Notes 1. The pH of the HBS should be between 7.05 and 7.12. The pH of the solution may change after long storage periods. Always use HBS solution that has been properly stored. 2. Incorporate gelatin into gel: Use a certain volume of gelatin stock solution to replace the same volume of water when preparing the SDS-polyacrylamide gel. 3. 30◦ C can be used. Prolonged incubation (overnight) for induction does not increase yield and can lead to degradation of the protein. 4. Because the target protein is very active enzyme and may easily auto-degrade, it is important to keep samples on ice and not let them overheat during sonication or in the following procedures. 5. The pellet is rather difficult to dissolve. Make sure the pellet is completely dissolved; otherwise, the yield will be low.

Expression and Purification of Membrane-Type MMPs

109

6. Full-length MT-MMPs can be purified using the cell lysate. Do not use SDS, urea, or reducing agents which denature the M2 antibody on the resin and destroy the binding of FLAG-tagged protein to the resin. 7. MT-MMP can be purified from transiently transfected cells too. For example, grow HEK293 cells with full medium in 15 cm dishes to 80–90% confluency and transfect the MTMMP into cells using the calcium phosphate precipitation technique (see Section 3.3). 8. Do not leave the column in glycine-HCl for longer than 20 min. 9. Formation of the appropriate size of DNA–calcium phosphate precipitates is critical to achieve a high transfection efficiency. The precipitates should be very fine, dust-like precipitates. Over agitating, vortexing, or pipetting usually dramatically increases the size of the precipitate. Add about 1/8 vol of 2 × HBS to DNA/CaCl2 mixture each time. Adding too much 2 × HBS to DNA/CaCl2 mixture each time can also lead to large precipitates which can decrease the transfection efficiency and increase cytotoxicity. 10. A large quantity of very fine, dust-like precipitate should be visible 1 h after transfection under the microscope. 11. The incubation time depends on the volume of conditioned medium used for analysis and the activity of the MTMMP in activating proMMP-2. References 1. Nagase, H. and Woessner, J. F., Jr. (1999) Matrix metalloproteinases. J Biol Chem 274, 21491–21494. 2. Zucker, S., Pei, D., Cao, J., and Lopez-Otin, C. (2003) Membrane type-matrix metalloproteinases (MT-MMP). Curr Top Dev Biol 54, 1–74. 3. Nie, J. and Pei, D. (2004) Rapid inactivation of alpha-1-proteinase inhibitor by neutrophil specific leukolysin/membrane-type matrix metalloproteinase 6. Exp Cell Res 296, 145–150. 4. Nie, J. and Pei, D. (2003) Direct activation of pro-matrix metalloproteinase-2 by leukolysin/membrane-type 6 matrix metalloproteinase/matrix metalloproteinase 25 at the asn(109)-Tyr bond. Cancer Res 63, 6758–6762. 5. Jiang, A. and Pei, D. (2003) Distinct roles of catalytic and pexin-like domains in membrane-type matrix metalloproteinase (MMP)-mediated pro-MMP-2 activation

6.

7.

8.

9.

and collagenolysis. J Biol Chem 278, 38765–38771. Kang, T., Yi, J., Yang, W., Wang, X., Jiang, A., and Pei, D. (2000) Functional characterization of MT3-MMP in transfected MDCK cells: progelatinase a activation and tubulogenesis in 3-D collagen lattice. FASEB J 14, 2559–2568. Wang, P., Wang, X., and Pei, D. (2004) Mint-3 regulates the retrieval of the internalized membrane-type matrix metalloproteinase, MT5-MMP, to the plasma membrane by binding to its carboxyl end motif EWV. J Biol Chem 279, 20461–20470. Jiang, A., Lehti, K., Wang, X., Weiss, S. J., Keski-Oja, J., and Pei, D. (2001) Regulation of membrane-type matrix metalloproteinase 1 activity by dynamin-mediated endocytosis. Proc Natl Acad Sci U S A 98, 13693–13698. Wang, X., Yi, J., Lei, J., and Pei, D. (1999) Expression, purification and char-

110

Nie and Pei

acterization of recombinant mouse MT5MMP protein products. FEBS Lett 462, 261–266. 10. Wang, P., Nie, J., and Pei, D. (2004) The hemopexin domain of membrane-type matrix metalloproteinase-1 (MT1-MMP) Is not required for its activation of proMMP2 on cell surface but is essential for MT1-MMPmediated invasion in three-dimensional

type I collagen. J Biol Chem 279, 51148–51155. 11. Sato, H., Takino, T., Kinoshita, T., Imai, K., Okada, Y., Stetler Stevenson, W. G., and Seiki, M. (1996) Cell surface binding and activation of gelatinase A induced by expression of membrane-type-1-matrix metalloproteinase (MT1-MMP). FEBS Lett 385, 238–240.

Chapter 7 Refolding of TIMP-2 from Escherichia coli Inclusion Bodies Richard A. Williamson Abstract The TIMP proteins contain six intramolecular disulfide bonds and form unfolded insoluble aggregates when expressed in E. coli. Eukaryotic expression systems provide the necessary post-translational modification apparatus to produce authentic TIMP but are comparatively slow and more expensive. This chapter describes the production of native TIMP-2 (both full-length and the N-terminal domain) from E. coli by in vitro refolding. The technique allows high-level intracellular expression and efficient isolation of the recombinant product without the use of fusion tags or partners. Protein purity after ion exchange and gel filtration chromatography was judged to be greater than 95% with yields of 15 mg/L from LB medium and 10 mg/L from minimal medium. Key words: Refolding, renaturation, inclusion bodies, TIMP-2, purification.

1. Introduction E. coli is a convenient host in which to express recombinant proteins. The technology is available for most laboratories, as it is relatively inexpensive and does not require extensive expertise. The major drawback of E. coli as an expression host is the inability of the organism to carry out many post-translational modifications, including glycosylation and disulfide bond formation. High-level intracellular expression of many mammalian proteins in E. coli results in the formation of large insoluble aggregates, known as inclusion bodies (1). These dense bodies consist predominantly of the misfolded recombinant product, together with components of the transcription/translation machinery (i.e., RNA polymerase, ribosomal RNA, and plasmid DNA). The TIMPs are invariably I.M. Clark (ed.), D.A. Young, A.D. Rowan (Consulting editors), Matrix Metalloproteinase Protocols, Methods in Molecular Biology 622, DOI 10.1007/978-1-60327-299-5 7, © Springer Science+Business Media, LLC 2001, 2010

111

112

Williamson

insoluble when expressed in E. coli, their folding requirement for the formation of six disulfide bonds being incompatible with the reducing environment of the E. coli cell. Fortunately, active, correctly folded recombinant protein can often be recovered from insoluble inclusion bodies by a process of solubilization and in vitro refolding (2, 3). Indeed, inclusion body formation has the advantage that the recombinant product often accumulates to high levels in the cell (up to 30% of total cell protein) and allows easy isolation of that protein in a relatively pure and stable form (inclusion bodies are typically >50% recombinant protein). The refolding protocol presented here was devised for the active N-terminal domain of TIMP-2 (N-TIMP-2) (4). The development of a refolding protocol is often a lengthy empirical task where each parameter at the in vitro refolding stage (e.g., redox potential, residual denaturant concentration, protein concentration, temperature, pH, time) must be optimized to achieve maximal yields. For N-TIMP-2, the refolding procedure was developed by following protein refolding in real time using intrinsic fluorescence of the unique tryptophan residue (5). This approach enabled rapid screening of the refolding parameters and is a generally applicable method if a source of native material is available to characterize the fluorescence properties of the fully folded state. The N-TIMP-2 refolding procedure can be broken down into several steps. 1. Cell lysis and inclusion body isolation (see Fig. 7.1). 2. Inclusion body washing (adapted from Marston et al. (6)). 3. N-TIMP-2 reduction and solubilization in 5 M guanidinium chloride (GdmCl). 4. In vitro refolding by dilution into refolding buffer (pH 8.75) containing a redox couple of reduced and oxidized glutathione. 5. Protein concentration by ultrafiltration and purification of the refolded form. The procedure can usually be completed in 6 days starting from E. coli cell pellets. The refolding protocol has proved to be very robust in our laboratory with over 20 preparations successfully completed. Typical yields of N-TIMP-2 after refolding and purification are 30% of that present in the washed inclusion bodies. Although the refolding procedure presented here was devised for N-TIMP-2, subsequent work to develop a protocol for fulllength TIMP-2 found maximal recoveries under very similar refolding conditions, except for the protein concentration which should be reduced to 10 ␮g/mL (see Note 4). Final yields for refolded full-length TIMP-2 are, however, significantly lower than those for the N-terminal domain (typically 16%). To achieve higher yields from the same quantity of inclusion bodies it is

Refolding of TIMP-2 from Escherichia coli Inclusion Bodies

113

Fig. 7.1. Expression of pET23d-N-TIMP-2 in E. coli strain BL21 (DE3). Lane 1, molecular weight marker proteins including N-TIMP-2 (T2). Lane 2, E. coli before induction with IPTG. Lane 3, E. coli after induction with IPTG for 3 h. Lane 4, lysis pellet from induced cells (unwashed inclusion bodies). Lane 5, lysis supernatant from induced cells. Samples were run reduced on a 10–20% SDS-PAGE gradient gel.

possible to recycle the misfolded forms back through the protocol (see Note 11). This has proved very successful with full-length TIMP-2 where 50% of the starting material was collected as misfolded forms and reprocessed with a refolding and purification yield of 18%. The procedure starts from E. coli cell pellets containing N-TIMP-2 (or full-length TIMP-2) inclusion bodies. In our work, both these molecules are expressed from pET23 vectors (Novagen) in E. coli strain BL21 (DE3), where the insert is under the transcriptional control of the T7 promoter. The proteins are expressed as the mature sequence with the addition of an N-terminal methionine residue for initiation of translation. This methionine residue is not present in the final refolded product, however, as it is removed during expression (5). Typical expression yields in Luria-Bertani (LB) medium are 40–60 mg/L for NTIMP-2 and 20–30 mg/L for the full-length molecule. Minimal medium has also been used extensively in our laboratory for the production of uniform 15 N- and 13 C-labeled TIMP-2 proteins for heteronuclear nuclear magnetic resonance (NMR) studies (7– 9). Expression yields of 25–35 mg/L for the N-terminal domain and 12–15 mg/L for full-length TIMP-2 have been obtained in Spizizen’s minimal medium using 13 C-labeled glucose and 15 Nlabeled ammonium sulfate as sole carbon and nitrogen sources. The refolding procedure presented here should not be dependent

114

Williamson

on the E. coli expression system used if the protein is produced in a similar form. The effect of fusion partners or expression tags (e.g., glutathione S-transferase or the His-tag) on TIMP-2 refolding has not been evaluated but we have successfully used the procedure to produce active site-directed mutants of TIMP-2, TIMP-4, and TIMP-2/TIMP-4 chimeras (10–12).

2. Materials 2.1. Inclusion Body Recovery and Refolding

Except where explicitly stated, all buffers and reagents should be stored at 4◦ C and are stable for up to 3 months. Those buffers and reagents containing free thiol groups (i.e., dithiothreitol (DTT) and reduced glutathione) should only be prepared as required. 1. Lysis buffer: 50 mM Tris–HCl, pH 8.0, 2 mM EDTA. 2. Lysozyme: 10 mg/mL in lysis buffer. Store at –20◦ C. 3. Triton X-100: 5% (v/v) in lysis buffer. Store at 4◦ C. 4. Wash buffer: 50 mM Tris–HCl, pH 8.0, 10 mM EDTA, 0.5% Triton X-100. 5. Solubilization buffer: 50 mM Tris–HCl, pH 8.75, 5 M GdmCl. 6. DTT: 0.5 M in solubilization buffer (prepare immediately before use). 7. Refolding buffer: 50 mM Tris–HCl, pH 8.75, 0.45 M GdmCl, 0.78 mM reduced glutathione, 0.44 mM oxidized glutathione (see Note 14) (prepare immediately before use). 8. Dialysis buffer: 50 mM NaH2 PO4 –NaOH, pH 6.0.

2.2. Purification of Refolded N-TIMP-2

1. Ion exchange buffer A: 50 mM NaH2 PO4 –NaOH, pH 6.0 (use within 7 days). 2. Ion exchange buffer B: 50 mM NaH2 PO4 –NaOH pH 6.0, 0.5 M NaCl (use within 7 days). 3. Gel filtration buffer: 25 mM NaH2 PO4 –NaOH pH 6.0, 100 mM NaCl, 0.03% (w/v) NaN3 .

3. Methods 3.1. Inclusion Body Recovery and Refolding

The procedure given below is for cell pellets obtained from 400 mL cultures (final A600 = 1.2–1.5). After step 2, lysates may be pooled and the procedure scaled-up accordingly.

Refolding of TIMP-2 from Escherichia coli Inclusion Bodies

115

1. Add 10 mL of lysis buffer to the E. coli cell pellet and thoroughly resuspend by pumping in and out of a 10 mL pipette. Transfer the E. coli suspension to a 50 mL benchcentrifuge tube and lyse the cells by adding 100 ␮L of lysozyme and 200 ␮L of Triton X-100 stock solutions. Mix thoroughly after each addition and incubate at room temperature for 30 min. The solution becomes highly viscous as DNA is released. Cool the lysate on ice. 2. Sonicate the lysis mixture using an appropriate ultrasound generator and probe. Prevent the lysis mixture from warming by placing the tube in an ice-cooled water bath. The viscosity of the mixture should return close to that of water as the DNA is sheared by the effect of the sonication and the mixture becomes whiter in color as the cellular material is solubilized (see Note 1). 3. Sediment the released N-TIMP-2 inclusion bodies by centrifugation at 12,000g for 15 min at 4◦ C. Decant off the supernatant. 4. Break up the pellet in 2.5 mL of inclusion body wash buffer and transfer to a Teflon or ground glass homogenizer where it can be thoroughly resuspended (see Note 2). Recover the inclusion body fraction by centrifugation at 12,000g for 10 min at 4◦ C. 5. Repeat wash step 4. 6. Repeat wash step 4 using 2.5 mL of water in place of the wash buffer. 7. Estimate the total amount of N-TIMP-2 in the inclusion preparation by comparing the intensity of Coomassie blue-stained bands on reducing SDS-PAGE. Use either NTIMP-2 standards of known amount for comparison or a different protein of similar molecular weight. Inclusion body samples for SDS-PAGE are conveniently taken after resuspension of the pellet in step 6. 8. Add solubilization buffer to the inclusion body pellet to give an N-TIMP-2 concentration of 2 mg/mL. The pellet becomes “glassy” on contact with the 5 M GdmCl solution and solubilization is best achieved using the Teflon or ground glass homogenizer (see Note 3). Add DTT from the 0.5 M stock to 11 mM, mix thoroughly, and leave at room temperature for 30 min to ensure complete reduction of the recombinant N-TIMP-2. 9. Remove any non-solubilized material by centrifugation at 12,000g for 5 min at 4◦ C. 10. Dilute the solubilized inclusion bodies into refolding buffer (1 in 100) to give a final N-TIMP-2 concentration of

116

Williamson

20 ␮g/mL. Add the solubilized inclusion bodies drop-wise and slowly from a pipette (10 mL over 2–3 min) while stirring the refolding buffer to ensure thorough mixing and continue to stir at room temperature for 2 h. Store the refolding mixture at 4◦ C overnight without stirring. 11. Concentrate the refolding mixture to 1/50–1/100 of its original volume by ultrafiltration using a 3 or 5 kDa molecular weight cutoff membrane (see Note 6). Either a stirred cell or pumped tangential-flow system may be used. 12. Dialyze the concentrated refolding mixture against 2.5 L of dialysis buffer for 16 h at 4◦ C. During dialysis a precipitate is formed which consists primarily of misfolded N-TIMP-2. 13. Transfer the contents of the dialysis bag to a centrifuge tube and incubate at 35◦ C for 45 min. This incubation step causes further precipitation of misfolded N-TIMP-2 (see Note 7). Remove the precipitate by centrifugation at 1,500g for 15 min at room temperature and then filter the supernatant through a 0.2 ␮m filter to ensure complete removal. 3.2. Purification of Refolded N-TIMP-2

The following method is routinely used to purify refolded NTIMP-2 from misfolded forms and contaminating E. coli proteins. This protocol generates N-TIMP-2 judged to be better than 95% pure as determined by image analysis of Coomassie blue-stained SDS-PAGE gels. The Resource-S column is run on an AKTA or FPLC chromatography system (see Note 9) and the Sephacryl S100 column is packed following GE Healthcare’s guidelines. Fulllength TIMP-2 behaves very similar to N-TIMP-2 and can be purified in an identical manner. Non-reducing SDS-PAGE analysis of samples taken during the purification of N-TIMP-2 is shown in Fig. 7.3. 1. Load 10 mL aliquots of the refolded N-TIMP-2 preparation onto a 1 mL Resource-S column (GE Healthcare) equilibrated at room temperature in ion exchange buffer A (see Note 8). After loading, wash the column with 10 mL (10 column volumes) of buffer A and then elute with a linear gradient of 0–50% buffer B over 15 mL at a flow rate of 1 mL/min. Refolded N-TIMP-2 elutes in fractions 7–9 (1 mL fractions collected from the start of the gradient elution). A misfolded form of N-TIMP-2 runs as a shoulder on the leading edge of the N-TIMP-2 peak and fractions across the N-TIMP-2 peak should be analyzed on reducing/nonreducing SDS-PAGE in order to omit this material from the N-TIMP-2 pool (see Note 5). A typical trace for Resource-S purification of refolded N-TIMP-2 is shown in Fig. 7.2.

Refolding of TIMP-2 from Escherichia coli Inclusion Bodies

117

Fig. 7.2. A280 profile from Resource-S chromatography of N-TIMP-2. The flow-through fraction (I) and the shoulder on the major peak (II) contain misfolded forms of N-TIMP-2. Fractions 7, 8, and 9 (III) are combined to give the refolded N-TIMP-2 pool. The NaCl gradient used to develop the column is shown by the dashed line and right-hand axis.

2. Load 5–10 mL aliquots of the N-TIMP-2 pool onto a Sephacryl S-100 column (5.3 cm2 × 38 cm; 180 mL bed volume) equilibrated in gel filtration buffer at 4◦ C. Elute the column with gel filtration buffer at a flow rate of 1.4 mL/min. N-TIMP-2 elutes at 120–140 mL from the start of elution (see Note 10).

4. Notes 1. As an alternative to sonication, the viscosity of the lysis mixture can be reduced by treatment with DNase I; to use this method, add MgCl2 to 10 mM and DNase I to 20 ␮g/mL. Mix well and incubate at room temperature for a further 30 min (or until the viscosity is reduced). Even if DNase I is used, it is still advisable to sonicate the lysis mixture to ensure complete cell breakage and so prevent unnecessary contamination of the inclusion body pellet with E. coli proteins. 2. Improved washing of the N-TIMP-2 inclusion body pellet can be achieved by sonication of the resuspended inclusion bodies in wash buffer in step 4. This additional step also

118

Williamson

helps to shear any remaining DNA. N-TIMP-2 inclusion bodies are typically 60–70% pure at the end of the washing stage (step 6). 3. If difficulties are experienced solubilizing the inclusion bodies in solubilization buffer (5 M GdmCl), then an alternative strategy is to first resuspend the pellet in 1/6 of the required final volume with water, and then to make up to the final volume with solubilization buffer at 1.2× concentration (i.e., 60 mM Tris–HCl, 6 M GdmCl, pH 8.75). The inclusion body suspension in water dissolves rapidly in the solubilization buffer, whereas the inclusion body pellet takes on a “glassy” nature and can be difficult to solubilize quickly. 4. Full-length TIMP-2 should be refolded using the same conditions as N-TIMP-2, except the protein concentration should be reduced to 10 ␮g/mL (solubilize inclusion bodies at 1 mg/mL in solubilization buffer and then dilute 1:100 into refolding buffer as for N-TIMP-2). Refolding yields for both molecules drop significantly if higher protein concentrations are used at the refolding stage. 5. Full-length and N-TIMP-2 refolding is monitored by comparing mobility on SDS-PAGE gels when samples are run with and without reduction. The size of the mobility difference on disulfide bond cleavage is approx. 3 kDa. Non-reduced SDS-PAGE is an effective method for determining the degree of contamination of the refolded TIMP2 with misfolded forms. The misfolded material is not correctly disulfide bonded and migrates as a smear above the native molecule (see Fig. 7.3). 6. A 10 kDa molecular weight cutoff filter can be used for NTIMP-2 (14.1 kDa) without suffering substantial losses of material. A 10 kDa molecular weight cutoff filter is recommended for full-length TIMP-2 (21.8 kDa) as flow rates are approximately twice that of a 3 kDa membrane. 7. The 35◦ C incubation step is essential to prevent problems with protein precipitation when loading the sample on the Resource-S column at room temperature. 8. 50 mM MES buffer (pH 6.0) may be used as an alternative to the phosphate buffer for the ion exchange chromatography. The advantage of MES is that this buffer is compatible with Ca2+ ions if the TIMP-2 is to be assayed for MMP inhibitory activity (the MMPs require Ca2+ for stability). A wide range of buffers can be used for the subsequent gel filtration step (step 2). 9. S-Sepharose FF and mono-S (Pharmacia) can both be used as alternative cation exchange matrices for N-TIMP-

Refolding of TIMP-2 from Escherichia coli Inclusion Bodies

119

Fig. 7.3. Non-reducing SDS-PAGE analysis of fractions taken during the N-TIMP-2 refolding and purification procedure. Lane 1, the protein precipitate formed during dialysis against phosphate buffer and the 35◦ C heat treatment step. Lane 2, soluble-refolded N-TIMP-2 prior to loading onto Resource-S. Lane 3, flow-through fraction containing misfolded forms of N-TIMP-2. Lane 4, the refolded N-TIMP-2 pool collected from the Resource-S column. Lane 5, the refolded N-TIMP-2 pool collected from the Sephacryl S-100 column. Samples were run on a 10–20% SDS-PAGE gradient gel. All loadings were normalized to the same proportion of the total, except for lane 4, where fourfold more material was loaded to show the contaminants at 30 kDa which were subsequently removed by gel filtration. Misfolded forms of N-TIMP-2 (labeled MF-T2) are incorrectly disulfide bonded and run above the correctly folded material (labeled T2).

2 purification. On both matrices N-TIMP-2 was found to elute slightly later in the NaCl gradient. 10. Gel filtration chromatography is important for the removal of several protein species around 30 kDa. Only one of these bands is seen on reducing SDS-PAGE suggesting that the others are misfolded disulfide-linked N-TIMP-2 dimers. The band seen on reducing SDS-PAGE is thought to be ␤-lactamase as it is also upregulated on IPTG induction. 11. The folding efficiencies for the N-terminal domain and fulllength TIMP-2 are typically 30% and 15–18%, respectively. A large proportion of the misfolded full-length TIMP-2 precipitates during dialysis into phosphate buffer and the subsequent 35◦ C incubation step (steps 12 and 13). This material, together with the flow-through fraction and leading edge fractions from the ion exchange separation, is easily collected and can be recycled back through the refolding protocol starting at step 7. The precipitate can be solubilized directly in 1.2× solubilization buffer (see Note 3); the ion exchange fractions are concentrated by ultrafiltration and then added to the solubilized pellet such that the final GdmCl concentration is 5 M. We have used this recycling approach to obtain good yields of 15 N/13 C-labeled

120

Williamson

TIMP-2 where the high cost of the isotopes necessitate that every effort is made to obtain maximal recoveries. The yields from the reprocessing of TIMP-2 are typically 50– 60% of the initial inclusion body refolding. 12. Estimates of the A280 extinction coefficients for the Nterminal domain and full-length TIMP-2 were calculated from their tryptophan, tyrosine, and cysteine content as described by Gill et al. (13). These values in terms of A280 absorbance for a 1 mg/mL solution (0.88 and 1.49 for NTIMP-2 and full-length TIMP-2, respectively) were used to determine the concentration of refolded TIMP-2 at the end of the purification procedure. 13. The results from biochemical and conformational analysis of refolded N-TIMP-2 can be found in (5). 14. The addition of 0.5 M L-arginine to the refolding buffer was found to increase the refolding yields of some sitedirected mutants of TIMP-2 and N-TIMP-4 (12).

Acknowledgments The author would like to thank Dessy Natalia, John Sergent, Michael Rae, Mark Carr, Magdalini Rapti, and Robert Freedman for their contributions to this work. The DNA for full-length TIMP-2 was kindly provided by Alan Bradbury at the National Institute of Medical Research, Mill Hill, London, UK. This work was supported by the Arthritis Research Campaign, UK. References 1. Thatcher, D. R. and Hitchcock, A. (1994) Protein folding in biotechnology. In Pain, R. H. (ed.), Mechanisms of protein folding. Oxford, UK: Frontiers in Molecular Biology Series, IRL Press, pp. 229–261. 2. Fischer, B., Sumner, I., and Goodenough, P. (1993) Isolation, renaturation and formation of disulphide bonds of eukaryotic proteins expressed in Escherichia coli as inclusion bodies. Biotechnol Bioengineering 41, 3–13. 3. Wetzel, R. (1992) Principles of protein stability. Part 2 - enhanced folding and stabilization of proteins by suppression of aggregation in vitro and in vivo. In Rees, A. R., Sternberg, M. J. E., and Wetzel, R. (eds.), Protein engineering – a practical approach. Oxford, UK: IRL Press, pp. 191–219. 4. Murphy, G., Houbrechts, A., Cockett, M. I., Williamson, R. A., O’Shea, M., and Docherty, A. J. P. (1991) The N-terminal

domain of tissue inhibitor of metalloproteinases retains metalloproteinase inhibitory activity. Biochem 30, 8097–8102. 5. Williamson, R. A., Natalia, D., Gee, C. K., Murphy, G., Carr, M. D., and Freedman, R. B. (1996). Chemically and conformationally authentic active domain of human tissue inhibitor of metalloproteinases2 refolded from bacterial inclusion bodies. Eur J Biochem 241, 476–483. 6. Marston, F. A. O., Lowe, P. A., Doel, M. T., Schoemaker, J. M., White, S., and Angal, S. (1984) Purification of calf prochymosin (prorennin) synthesized in Escherichia coli. Bio/Technol 2, 800–807. 7. Williamson, R. A., Carr, M. D., Frenkiel, T. A., Feeney, J., and Freedman, R. B. (1997). Mapping the binding site for matrix metalloproteinase on the N-terminal domain of the tissue inhibitor of metalloproteinases-2

Refolding of TIMP-2 from Escherichia coli Inclusion Bodies by NMR chemical shift perturbation. Biochemistry 36, 13882–13889. 8. Muskett, F. W., Frenkiel, T. A., Feeney, J., Freedman, R. B., Carr, M. D., and Williamson, R. A. (1998). High resolution structure of the N-terminal domain of tissue inhibitor of metalloproteinases-2 and characterisation of its interaction site with matrix metalloproteinase-3. J Biol Chem 273, 21736–21743. 9. Williamson, R. A., Muskett, F. W., Howard, M. J., Freedman, R. B., and Carr, M. D. (1999). The effect of matrix metalloproteinase complex formation on the conformational mobility of tissue inhibitor of metalloproteinases-2 (TIMP-2). J Biol Chem 274, 37226–37232. 10. Bulter, G. S., Hutton, M., Wattam, B. A., Williamson, R.A., Knauper, V., Willenbrock,

121

F., and Murphy, G. (1999) The specificity of TIMP-2 for matrix metalloproteinases can be modified by single amino acid mutations. J Biol Chem 274, 20391–20396. 11. Williamson, R. A., Hutton, M., Vogt, G., Rapti, M., Knauper, V., Carr, M. D, andMurphy, G. (2001) Tyrosine 36 plays a critical role in the interaction of the AB loop of TIMP-2 with matrix metalloproteinase-14. J Biol Chem 276, 32966–32970. 12. Rapti, M., Knauper, V., Murphy, G., and Williamson, R. A. (2006) Characterization of the AB loop of TIMP-2. Involvement in pro-MMP-2 activation. J Biol Chem 281, 23386–23394. 13. Gill, S. C. and Von Hippel, P. H. (1989) Calculation of molar extinction coefficients from amino acid sequence data. Anal Biochem 182, 319–326.

Chapter 8 Purification of MMPs and TIMPs Ken-ichi Shimokawa and Hideaki Nagase Abstract A number of matrix metalloproteinases (MMPs) and their endogenous inhibitors called tissue inhibitors of metalloproteinases (TIMPs) from natural sources have been identified. This chapter describes the purification methods of MMPs-1, -2, -3, -7, -8, -9, -10, -12, and -13 and TIMPs-1 and -2. The sources of the proteins and assay methods to detect their activities are also described. Key words: MMPs, TIMPs, purification, chromatography, enzyme assays, MMP inhibition assays, fibroblasts.

1. Introduction Currently more than 20 different matrix metalloproteinases (MMPs) and 4 tissue inhibitors of metalloproteinases (TIMPs) have been identified (1, 2) in vertebrates. This chapter describes methods to purify collagenases (MMP-1, MMP-8, and MMP-13), gelatinases (MMP-2 and MMP-9), stromelysins (MMP-3 and MMP-10), matrilysin 1 (MMP-7), macrophage elastase (MMP-12), and TIMP-1 and TIMP-2 from natural sources. While a number of MMPs and TIMPs are available in recombinant forms, the preparations from natural sources provide standards of enzymatic and inhibitory activities.

I.M. Clark (ed.), D.A. Young, A.D. Rowan (Consulting editors), Matrix Metalloproteinase Protocols, Methods in Molecular Biology 622, DOI 10.1007/978-1-60327-299-5 8, © Springer Science+Business Media, LLC 2001, 2010

123

124

Shimokawa and Nagase

2. Materials 2.1. Purification of proMMP-1

1. Sources: Serum-free conditioned medium of human fibroblasts (rheumatoid synovium, skin, or gingiva) cultured in the presence of phorbol myristate acetate (PMA) (50 ng/mL) and/or interleukin 1 (IL-1) (10 ng/mL) for 3–4 days. 2. Substrates: 14 C-labeled collagen or Mca-Pro-Gly-LeuDpa-Ala-Arg-NH2 from Peptides International (Louisville, KY), Peptide Institute Inc. (Osaka, Japan), or Bachem AG (Bubendorf, Switzerland). 3. DEAE-cellulose (DE-52) from Whatman, UK. 4. Green A Dyematrex gel, Diaflo apparatus, and YM-10 membrane from Millipore (Billevica, MA). 5. Affi-Gel 10 from Bio-Rad (Hercules, CA). 6. Sephacryl S-200, concanavalin A (Con A)-Sepharose, and gelatin-Sepharose 4B from GE Healthcare Life Sciences. 7. 0.2 M 4-aminophenylmercuric acetate (APMA) solution: APMA from Merck Chemicals Nottingham, UK is dissolved in 0.2 M NaOH and titrated with concentrated HCl to pH 11.0. Below pH 11.0, APMA precipitates at that concentration. The solution is stable at room temperature for at least 4 weeks. 8. TNC buffer: 50 mM Tris-HCl (pH 7.5), 0.15 M NaCl, 10 mM CaCl2 , 0.02% NaN3 , 0.05% Brij 35. 9. TC buffer: 50 mM Tris-HCl (pH 7.5), 10 mM CaCl2 , 0.02% NaN3 , 0.05% Brij 35. 10. Anti-(human MMP-3) IgG-Affi-Gel 10 (see below for preparation).

2.2. Purification of Human proMMP-2

1. Sources: Conditioned medium of cultured fibroblasts. Fibroblasts from various species in culture constitutively produce proMMP-2. Other cell types such as epithelial cells, neoplastic cell lines, macrophages in culture also produce this zymogen, but the level may be variable. 2. Substrates: 14 C-labeled gelatin (heat-denatured collagen) or Mca-Pro-Leu-Gly-Leu-Dpa-Ala-Arg-NH2 from Peptides International, Peptide Institute Inc., or Bachem AG. 3. Gelatin-Sepharose 4B and Sephacryl S-200 from GE Healthcare Life Sciences. 4. Diaflo apparatus and YM-10 membrane from Millipore. 5. TNC buffer: see Section 2.1.

Purification of MMPs and TIMPs

2.3. Purification of Human proMMP-3

125

1. Materials required for purification are essentially the same as for human proMMP-1 (see Section 2.1). 2. Anti-(human MMP-1) IgG-Affi-Gel 10 and CM-cellulose (CM-52; Whatman). 3. Protein substrates: reduced, [3 H]carboxymethylated transferrin (14), casein, or Azocoll (Merck Chemicals). 4. Synthetic substrates: Mca-Pro-Leu-Gly-Leu-Dpa-Ala-ArgNH2 or Mca-Arg-Pro-Lys-Pro-Val-Glu-Nva-Trp-Arg-Lys (Dnp)-NH2 from Peptides International, Peptide Institute, Inc., or Bachem AG.

2.4. Purification of MMP-7

1. Sources: Conditioned medium of CaR-1 human rectal carcinoma cells (Japanese Cancer Research Cell Bank) (3).

2.4.1. Purification from the medium of CaR-1 cells

2. Substrates: Azocoll from Merck Chemicals or Mca-Pro-LeuGly-Leu-Dpa-Ala-Arg-NH2 from Peptides International, Peptide Institute, Inc., or Bachem AG. 3. DEAE-cellulose from Whatman. 4. Diaflo apparatus, YM-10 membrane, and Green A Dyematrex gel from Millipore. 5. Sephacryl S-200 from GE Healthcare Life Sciences. 6. TNC buffer: see Section 2.1. 7. Buffer 7.1: 10 mM Tris-HCl (pH 8.0), 0.5 mM CaCl2 , 0.02% NaN3 , 0.05% Brij 35. 8. Buffer 7.2: 25 mM sodium borate (pH 8.0), 0.15 M NaCl, 1 mM CaCl2 , 0.02% NaN3 , 0.05% Brij 35. 9. Buffer 7.3: 25 mM sodium cacodylate (pH 6.5), 1 mM CaCl2 , 0.02% NaN3 , 0.05% Brij 35.

2.4.2. Purification from Postpartum Rat Uterus

1. Sources: Rat uterus at 1 day postpartum (4). 2. Substrate: Same as in Section 2.4.1. 3. Ultrogel AcA54 from Sigma. 4. Blue-Sepharose and chelating Sepharose Fast Flow resin from GE Healthcare Life Sciences. 5. Buffer 7.4: 50 mM Tris-HCl (pH 7.5), 0.15 M NaCl, 0.1 M CaCl2 . 6. Buffer 7.5: 50 mM Tris-HCl (pH 7.5), 0.2 M NaCl, 10 mM CaCl2 , 0.02% NaN3 , 0.05% Brij 35. 7. Buffer 7.6: 25 mM boric acid–NaOH (pH 8.0), 1 mM CaCl2 , 0.05% Brij 35, 0.03% toluene. 8. Buffer 7.7: 25 mM sodium cacodylate (pH 6.5).

126

Shimokawa and Nagase

2.5. Purification of Human proMMP-8 and MMP-8

1. Source: human blood buffy coat (neutrophils) 2. Plasmatonin solution: 6% hydroxyethylstarch in 0.9% NaCl 3. Substrates: same as in Section 2.1. 4. Chelating Sepharose Fast Flow, activated CH-Sepharose 4B, Q-Sepharose Fast Flow, Orange-Sepharose CL-6B, and Sephacryl S-300 from GE Healthcare Life Sciences. 5. Buffer 8.1: 20 mM Tris-HCl (pH 9.0), 5 mM CaCl2 , 0.5 mM ZnCl2. 6. Buffer 8.2: 20 mM Tris-HCl (pH 8.5), 5 mM CaCl2 , 0.5 mM ZnCl2 , 0.02% NaN3 . 7. Buffer 8.3: 20 mM Tris-HCl (pH 7.5), 5 mM CaCl2 , 0.5 mM ZnCl2 , 0.1 M NaCl. 8. Buffer 8.4: 50 mM Tris-HCl (pH 7.5), 5 mM CaCl2 , 0.5 M NaCl, 0.5 mM ZnCl2 , 0.02% NaN3 . 9. Buffer 8.5: 20 mM Tris-HCl (pH 7.5), 5 mM CaCl2 , 0.5 M NaCl. 10. Buffer 8.6: 100 mM Tris-HCl (pH 9.0), 20 mM CaCl2 0.5 M NaCl. 11. Buffer 8.7: 600 mM Tris-HCl (pH 7.5), 20 mM CaCl2 , 0.5 M NaCl. 12. Benzamidine, diisopropyl fluorophosphate (DFP), and phenylmethyl sulfonylfluoride (PMSF) and Pro-Leu-Glyhydroxamate HCl from Sigma (St. Louis, MO).

2.6. Purification of proMMP-9

1. Sources: The serum-free conditioned medium of U937 monocytic leukemia cells or HT1080 fibrosarcoma cells stimulated with PMA (50 ng/mL) (5). 2. Substrates: same as in Section 2.2. 3. Gelatin-Sepharose, Con A-Sepharose, and Sephacryl S-200 from GE Healthcare Life Sciences. 4. TNC buffer: see Section 2.1.

2.7. Purification of proMMP-10

1. Sources: the conditioned medium of human keratinocytes in culture stimulated with PMA, transforming growth factor ␣, epidermal growth factor, or TNF (6); the conditioned medium of OSC-20 human oral squamous carcinoma cells stimulated with PMA (7). 2. Substrates: same as in Section 2.3. 3. DEAE-cellulose (DE52) from Whatman. 4. Green A Dyematrex gel and YM-10 membrane from Millipore.

Purification of MMPs and TIMPs

127

5. Affi-Gel 10 from Bio-Rad. 6. Sephacryl S-200 from GE Healthcare Life Sciences. 7. Anti-(human MMP-1) antibody (clone 78-12G8) and anti(human MMP-3) antibody (clone 55-2A4) are from Fuji Chemical Industries, Inc. (Takaoka, Japan). 8. 0.2 M APMA: see Section 2.1. 9. TNC buffer: see Section 2.1. 10. Buffer 10.1: 50 mM Tris-HCl (pH 8.0), 5 mM CaCl2 , 0.05% Brij 35, 0.02% NaN3 . 2.8. Purification of MMP-12

1. Sources: the conditioned culture medium of mouse peritoneal macrophages treated with 2 ␮M colchicine for 48 h. 2. Brewer’s thioglycollate medium from BD Diagnostic Systems (Heidelberg Germany). 3. DME medium: Dulbecco’s modified Eagle’s medium, high glucose formulation from Invitrogen. 4. DME-LH medium: DME medium containing 0.2% lactalbumin hydrolysate from Invitrogen. 5. Substrates: 3 H-labeled elastin or Mca-Pro-Leu-Gly-LeuDpa-Ala-Arg-NH2 from Peptides International, Peptide Institute Inc., or Bachem AG. 6. DEAE-Sephadex A-25 from GE Healthcare Life Sciences. 7. Ultrogel AcA54 from LKB. 8. Buffer 12.1: 20 mM Tris-HCl (pH 7.6). 9. Buffer 12.2: 50 mM Tris-HCl (pH 7.6), 0.15 M NaCl, 50 mM CaCl2 , 0.02% NaN3 .

2.9. Purification of TIMP-1

1. Sources: the conditioned medium of fibroblasts (skin, tendon, gingiva, etc); human plasma. 2. MMP-1 or MMP-3. 3. Substrate for MMP-1 or MMP-3 (see Sections 2.1 and 2.3). 4. Heparin-Sepharose CL-6B and DEAE-Sepharose from GE Healthcare Life Sciences. 5. C4 reverse-phase HPLC from Vydac (Hesperia, CA). 6. Anti-(human TIMP-1) IgG-Affi-Gel 10. 7. TCO buffer: 50 mM Tris-HCl (pH 7.5), 10 mM CaCl2 , 0.01% ␤-octylglucoside. 8. Buffer T1-1: 25 mM sodium cacodylate (pH 7.6), 10 mM CaCl2 , 0.5 M NaCl, 0.05% Brij 35, 0.02% NaN3 . 9. Buffer T1-2: 50 mM glycine–HCl (pH 3.0), 0.5 M NaCl.

128

Shimokawa and Nagase

10. Buffer T1-3: 25 mM sodium cacodylate (pH 7.2), 1 M NaCl, 0.05% Brij 35, 0.02% NaN3 . 2.10. Purification of TIMP-2

1. Sources: the serum-free conditioned medium of fibroblasts (human uterine cervix, human gingiva).

2.10.1. Method 1

2. MMP-1 or MMP-3. 3. Substrates for MMP-1 or MMP-3 (see Sections 2.1 and 2.3). 4. Gelatin-Sepharose and Sephacryl S-200 from GE Healthcare Life Sciences. 5. TNC buffer: see Section 2.1.

2.10.2. Method 2

1. Source: the serum-free conditioned medium of human hepatoma HLE cells from Japanese Cancer Research Resource Bank. 2. MMP-1 or MMP-3. 3. Substrates for MMP-1 or MMP-3 (see Sections 2.1 and 2.3). 4. Cellulofine GCL-2000-m column from Chisso (Tokyo, Japan). 5. Reactive Red-agarose from Sigma. 6. SynChropak RP-4 column from SynChrom (Lafayette, IN). 7. Buffer T2-1: 10 mM Tris-HCl (pH 7.5), 0.5 M NaCl, 0.1% Brij 35. 8. Buffer T2-2: 20 mM Tris-HCl (pH 7.5), 0.01% Brij 35.

3. Methods See Notes 1, 2, 4, and 51 for general points. 3.1. Purification of Human proMMP-1 (see Note 3) 3.1.1. Assays

3.1.2. Activation of proMMP-1 (see Notes 5 and 6)

MMP-1 activity is measured, after activation of proMMP-1 with APMA, against synthetic substrates such as Mca-Pro-Leu-GlyLeu-Dpa-Ala-Arg-NH2 (8) (see the details of assay procedures in Chapter 24) and 14 C-labeled collagen diffuse fibril assay described by Cawston and Barrett (9) (see Chapter 15). 1. Activation by APMA. Add 0.2 M APMA stock solution to proMMP-1 (1–100 ␮g/mL) to a final concentration of 1 mM and incubate the mixture at 37◦ C for 2 h. SDS-PAGE analysis shows 43-kDa and 41-kDa MMP-1. The 43-kDa form has Met68 and the 41-kDa has Val82 and Leu83 for

Purification of MMPs and TIMPs

129

their N termini, respectively (10). These forms have about 25% of the full activity of MMP-1. 2. Activation by APMA in the presence of MMP-3 (full activation of proMMP-1). Add an equimolar amount of MMP-3 and 1/200 vol of 0.2 M APMA stock solution to proMMP-1 in TNC buffer, and incubate the mixture at 37◦ C for 4 h. ProMMP-1 is activated to the 41-kDa form with Phe81 for its N terminus under these conditions, and it exhibits the full collagenolytic activity. Remove MMP-3 by anti-(MMP-3) IgG-Affi-Gel 10, if necessary. 3.1.3. Preparation of Anti-(human MMP-3) IgG-Affi-Gel 10

First prepare the IgG from antiserum: 1. Add 0.5 vol of 60% (w/v) polyethylene glycol 8000 in 50 mM Tris-HCl (pH 7.8) to the specific antiserum raised against MMP-3. 2. Stand overnight at 4◦ C. 3. Centrifuge at 10,000g for 15 min, dissolve the pellet with 0.1 M Tris-HCl (pH 8.0), and dialyze the sample against the same buffer. 4. Apply the sample to a column of DEAE-cellulose equilibrated with 0.1 M Tris-HCl (pH 8.0) and wash the column until A280 nm reaches the baseline. 5. Combine unbound fractions (IgG fraction) and concentrate to about 10–15 mg/mL. 6. Dialyze the IgG fraction against 0.1 M NaHCO3 (pH 8.5). Then, couple the anti-(human MMP-3) IgG to AffiGel 10: 7. Shake the Affi-Gel 10 vial and make a uniform suspension. 8. Transfer 2 mL of slurry to a glass-fritted funnel and wash with 6 mL of cold deionized water. 9. Add 3 mL of IgG solution immediately to the gel cake of Affi-Gel 10. 10. Transfer the gel suspension to a plastic tube and gently mix the gel slurry on a rocker, shaker, or wheel for 4 h at room temperature. 11. Transfer the gel slurry to a glass-fritted funnel and wash with 6 mL of deionized water. 12. Suspend the gel cake into 6 mL of 0.1 M glycine–NaOH, pH 9.0, and gently mix for 2 h at room temperature. 13. Transfer the gel to a column and equilibrate with the desired buffer.

3.1.4. Purification Procedures

Step 1: DEAE-Cellulose Chromatography (see Note 7) 1. Concentrate the crude serum-free conditioned medium from cultured fibroblasts (500 mL) about 20-fold using

130

Shimokawa and Nagase

Diaflo apparatus with an Amicon YM-10 membrane and apply to a column of DEAE-cellulose (25 mL packed resin) equilibrated with TNC buffer without Brij 35. 2. Wash the column with the same buffer until A280 nm reaches the baseline and collect unbound protein fractions, which contain proMMP-1, proMMP-2, and proMMP-3. Step 2: Green A Dyematrex Gel Chromatography (see Note 8) 3. Apply the DEAE unbound fraction to a column of Green A Dyematrex gel (2.5 × 5 cm) equilibrated with TNC buffer without Brij 35. Most proteins in the medium do not bind to the column, whereas three proMMPs (proMMPs-1, -2, and -3) bind. 4. Wash the column with TNC buffer, and elute proMMPs bound to the column in a stepwise manner with TC buffer containing 0.3 M NaCl and then 0.5 M NaCl. 5. Monitor protein concentrations in fractions by reading at A280 nm and assay for MMP-1 activity. 6. Pool 0.5 M NaCl fractions which should contain proMMP-1. Step 3: Gel Filtration on Sephacryl S-200 7. Add 10% (w/v) Brij 35 to the 0.5 M NaCl fraction containing proMMP-1 (40 mL) to the final concentration of 0.05%; concentrate it to 1.5 mL with an Amicon YM-10 membrane, and then apply to a Sephacryl S-200 column (1.5 × 120 cm) equilibrated with TNC buffer. 8. Monitor MMP-1 activity and pool fractions containing proMMP-1. The yield of proMMP-1 from 500 mL of the conditioned medium is about 150–200 ␮g with recovery of about 45%. Step 4: Gelatin-Sepharose and Anti-MMP-3 Affinity Chromatography (see Note 9) 9. Concentrate the Sephacryl S-200 fraction by Amicon YM10 membrane to 2–3 mL and apply to a column of gelatinSepharose (5 mL) and a column of anti-(MMP-3) IgGAffi-Gel 10 successively. 10. Collect unbound fractions of each chromatography. The final product exhibits a major 52-kDa form (unglycosylated) and a minor 56 kDa (10–20%; glycosylated) of proMMP-1 on SDS-PAGE (see Note 11). Step 5: Separation of Glycosylated and Unglycosylated proMMP-1 11. Apply proMMP-1 preparation containing both glycosylated and unglycosylated forms to a column of Con A-Sepharose equilibrated with TC buffer containing 0.4 M NaCl.

Purification of MMPs and TIMPs

131

12. Collect the unbound fraction (containing unglycosylated proMMP-1) and apply it to a gel filtration column of Sephacryl S-200 (1.5 × 110 cm) equilibrated with TC buffer containing 0.4 M NaCl (see Note 10). 13. Elute the glycosylated 56-kDa form from the Con A-Sepharose column with 1 M ␣-methyl-D-mannoside in TC buffer containing 0.4 M NaCl. 14. Monitor the protein peak by reading A280 nm . 15. Concentrate the fraction eluted with ␣-methyl-mannoside and apply it to gel filtration chromatography on Sephacryl S-200 as above. 3.1.5. Separation of 41-kDa MMP-1, 22-kDa Catalytic Domain, and 25-kDa C-Terminal Domain of MMP-1

When proMMP-1 is incubated with 1 mM APMA at 37◦ C for >2 h, proMMP-1 converts to the 41-kDa active form (major), 25-kDa C-terminal domain, and 22-kDa catalytic domain (minor). The 22-kDa MMP-1 has peptidase activity but lacks collagenolytic activity. The 22-kDa catalytic domain of MMP-1 can be separated from the mixture by Green A Dyematrex gel chromatography. 1. Apply the mixture of the 41-kDa and 22-kDa MMP-1 in TNC buffer to a column of Green A Dyematrex gel (1 × 5 cm) equilibrated with TNC buffer. The 22-kDa catalytic domain of MMP-1 does not bind to the resin, whereas the 41-kDa active form and the 25-kDa C-terminal domain does. 2. Elute the bound protein with 0.5 M NaCl in TNC buffer. This fraction contains the 41-kDa MMP-1 and the 25-kDa C-terminal domain. 3. Concentrate the 0.5 M NaCl fraction with an Amicon YM-10 membrane and apply it to a gel filtration column of Sephacryl S-200 (1.5 × 110 cm) equilibrated with TNC buffer. This step separates the 41-kDa MMP-1 and the 25-kDa C-terminal domain.

3.2. Purification of Human proMMP-2 (see Note 3) 3.2.1. Assays

3.2.2. Activation of proMMP-2 (see Note 12)

MMP-2 activity is measured after activation with APMA using gelatin (denatured type I collagen) as a substrate (see Chapter 15) or synthetic substrates such as Mca-Pro-Leu-Gly-Leu-Dpa-AlaArg-NH2 (see Chapter 24). Gelatin zymography (see Chapter 16) is also useful to detect both proMMP-2 and active form of MMP-2. 1. Dilute 0.2 M APMA solution to 2 mM in TNC buffer. 2. Add an equal volume (10–50 ␮L) of 2 mM APMA to proMMP-2 preparation (1–5 ␮g/mL), incubate at 37◦ C for 45 min, and assay immediately for MMP-2 activity.

132

Shimokawa and Nagase

3.2.3. Purification Procedures

This method is based on Itoh et al. (11) and Ward et al. (12). Step 1. Gelatin-Sepharose Chromatography (see Notes 13 and 14) 1. Apply the conditioned culture medium (250 mL) of fibroblasts to a column of gelatin-Sepharose (1 × 8 cm) equilibrated with TNC buffer. 2.

Wash the column with the same buffer, and then TNC buffer containing 1% (v/v) dimethyl sulfoxide (DMSO).

3.

Elute the bound proMMP-2 with 5% DMSO in TNC buffer.

4.

Read A280 nm , combine the protein peak, and dialyze against TNC buffer for 18 h at 4◦ C.

Step 2: Gel Filtration on Sephacryl S-200 5. Concentrate the dialyzed material to about 1.5 mL by Diaflo apparatus with an Amicon YM-10 membrane and apply to a gel filtration column of Sephacryl S-200 (1.5 × 110 cm) equilibrated with TNC buffer.

3.3. Purification of proMMP-3 (see Note 3) 3.3.1. Assays

3.3.2. Activation of proMMP-3

3.3.2.1. Activation by APMA (see Note 15) 3.3.2.2. Activation by Chymotrypsin (Generation of the Full-Length [45 kDa] Mmp-3) (see Note 16)

6.

Analyze fractions for proMMP-2 by SDS-PAGE (10% acrylamide) with reduction. Molecular mass of proMMP-2 is about 72 kDa. Some fractions contain proMMP-2 and TIMP-2 (21 kDa). Reduced fibronectin exhibits a protein band of 220,000, which is eluted close to the void volume of the column.

7.

Pool fractions containing only proMMP-2. The proMMP2-TIMP-2 complex is eluted between fibronectin and proMMP-2.

MMP-3 activity is measured after activation of proMMP3 with APMA using Azocoll (13), casein, reduced [3 H]carboxymethylated transferrin, [3 H]Cm-Tf) as protein substrates (14), or synthetic substrates such as Mca-Pro-Leu-Gly-Leu-DpaAla-Arg-NH2 (8) and Mca-Arg-Pro-Lys-Pro-Val-Nva-Glu-TrpArg-Lys(Dnp)-NH2 (15). 1. Add 0.2 M APMA stock solution to proMMP-3 solution (1–100 ␮g/mL) to a final concentration of 1.5 mM and incubate at 37◦ C for 22 h. 1. Incubate proMMP-3 (100 ␮g/mL) and chymotrypsin (10 ␮g/mL, final concentration) at 37◦ C for 2 h. 2. Stop the reaction by adding 200 mM PMSF in dry isopropanol to a final concentration 1 mM.

Purification of MMPs and TIMPs

3.3.3. Purification Procedures 3.3.3.1. Method 1

133

Step 1: DEAE-Cellulose Chromatography (see Note 17) 1. Concentrate the crude conditioned culture medium of fibroblasts stimulated by PMA and/or IL-1 (500 mL) about 20-fold using Diaflo apparatus with an Amicon YM-10 membrane and apply it to a column of DEAEcellulose (25 mL) equilibrated with TNC buffer (pH 8.0) without Brij 35. 2.

Collect unbound fractions. All proMMPs in the medium are recovered in the unbound fraction.

Step 2: Green A Dyematrex Gel Chromatography 3. Apply the unbound fraction of DEAE-cellulose to a column of Green A Dyematrex gel (2.5 × 5 cm) equilibrated with TNC buffer. 4.

Wash the column with TNC buffer extensively, and elute proMMP-3 with TC buffer containing 0.3 M NaCl and collect the eluted protein peak. (see Note 18).

Step 3: Gelatin-Sepharose Column Chromatography 5. Apply the 0.3 M NaCl fraction to a column of gelatinSepharose column (5 mL) equilibrated with TC buffer containing 0.3 M NaCl. 6.

Wash the column with the same buffer and collect the unbound fraction.

Step 4: Gel Filtration on Sephacryl S-200 7. Concentrate the proMMP-3 fraction to 1.5 mL using a YM-10 membrane and apply to a column of gel filtration on Sephacryl S-200 equilibrated with TC buffer containing 0.4 M NaCl. 8.

Collect the main protein peak eluted in fractions corresponding to 50–60 kDa. ProMMP-3 exhibits a doublet of 57-kDa (major) and 59-kDa (minor) bands on SDS-PAGE.

Step 5: Further Purification of ProMMP-3 (see Note 19) 9. Apply the proMMP-3 fraction to a column of anti-(human MMP-1) IgG-Affi-Gel 10 (2 mL) equilibrated with TNC buffer. 10. Wash the column with the same buffer and collect the unbound protein peak. The final product exhibits a doublet of proMMP-3 (57 and 59 kDa). 3.3.3.2. Method 2: Immunoaffinity Chromatography

This procedure has only two chromatographic steps, but requires antibody against MMP-3 (16). 1. Apply the crude conditioned medium (500 mL) or tissue extracts to a column of Affi-Gel 10 coupled with polyclonal anti-(human MMP-3) (1.5 × 5 cm) equilibrated with TNC buffer without Brij 35.

134

Shimokawa and Nagase

2. Wash the column with the same buffer until A280 nm reaches the baseline. 3. Elute proMMP-3 with TNC buffer containing 6 M urea without Brij 35 (see Note 20). 4. Pool fractions with a protein peak and dialyze it against 200 vol of TNC buffer without Brij 35 three times. 5. Concentrate the sample using Diaflo apparatus with an Amicon YM-10 membrane. 6. Apply it to a column of Sephacryl S-200 (1.5 × 10 cm) equilibrated with TC buffer containing 0.4 M NaCl. 7. Analyze fractions with the major protein peak by SDSPAGE. Those fractions contain essentially homogeneous proMMP-3 of 57 and 59 kDa (see Note 22). 3.3.4. Separation of Glycosylated and Unglycosylated Forms of proMMP-3 (see Note 21)

1. Apply purified proMMP-3 consisting of unglycosylated 57-kDa and glycosylated 59-kDa forms to a column of Con A-Sepharose (1.5 ×5 cm) equilibrated with TC buffer containing 0.4 M NaCl. 2. Wash the column with the same buffer until A280 nm reaches the baseline. 3. Collect the unbound fraction as unglycosylated proMMP-3. 4. Elute the bound glycosylated proMMP-3 with 1 M ␣-methyl-D -mannoside in the same buffer and collect the protein peak. 5. Concentrate the unglycosylated proMMP-3 (unbound fraction) and the glycosylated proMMP-3 (bound fraction) separately; and apply each fraction to a gel filtration column of Sephacryl S-200 (1.5 × 110 cm) equilibrated with TC buffer containing 0.4 M NaCl.

3.3.5. Separation of Active MMP-3 and proMMP-3

1. Apply the mixture of proMMP-3 and active form of MMP-3 in TNC buffer to a column of Green A Dyematrex gel equilibrated with TNC buffer. 2. Wash the column with the same buffer until A280 nm reaches the baseline. 3. Collect the unbound protein peak as active MMP-3 (45 and 28 kDa). 4. Wash the column with TNC buffer, then with TC buffer containing 0.3 M NaCl. Collect the protein peak as proMMP-3.

3.4. Purification of proMMP-7 (see Notes 23–25) 3.4.1. Assays

MMP-7 activity is measured after activation with 1 mM APMA using Azo-coll or synthetic peptides (e.g., Mca-Pro-Leu-Gly-LeuDpa-Ala-Arg-NH2 ).

Purification of MMPs and TIMPs

3.4.2. Activation of proMMP-7

3.4.3. Purification from the Medium of CaR-1 Cells

135

1. Dilute 0.2 M APMA solution to 2 mM in TNC buffer 2. Add an equal volume (10–50 ␮L) of 2 mM APMA to proMMP-7 preparation and incubate the mixture at 37◦ C for 1 h. This method is based on Imai et al. (3) Step 1: DEAE-Cellulose Chromatography 1. Concentrate the serum-free conditioned culture medium of CaR-1 cells (500 mL) about 20-fold by Diaflo apparatus with an Amicon YM-10 membrane. 2.

Apply the sample to a column of DEAE-cellulose (2.5 × 8.5 cm) equilibrated with TNC buffer.

3.

Wash the column with the same buffer until A280 nm reaches the baseline.

4.

Pool unbound protein fractions.

Step 2: Green A Dyematrex Gel Chromatography 5. Apply the unbound fraction of DEAE-cellulose to a column of Green A Dyematrex gel (2.5 × 8 cm) equilibrated with TNC buffer. 6.

Wash the column with TNC buffer until A280 nm reaches to the baseline.

7.

Elute the bound proMMP-7 with a linear gradient of NaCl (0.15 M–2.0 M) in the same buffer.

8.

Monitor proMMP-7 activity against a synthetic substrate or Azocoll after activation of a 50–100 ␮L portion of each fraction with APMA. Alternatively, analyze the eluted fractions by SDS-PAGE (12.5% acrylamide). ProMMP-7 exhibits a protein band around 28 kDa.

9.

Combine fractions containing proMMP-7, and dialyze the sample against buffer 7.1.

Step 3: DEAE-Cellulose Chromatography 10. Apply the dialyzed sample to a column of DEAE-cellulose (1 × 18.5 cm) equilibrated in buffer 7.1 and wash the column with the same buffer until A280 nm reaches the baseline. 11. Combine unbound protein fractions and dialyze the pool against buffer 7.2. Step 4: Zn2+ -Chelating Sepharose Fast Flow Chromatography 12. Make a Zn2+ -chelating Sepharose Fast Flow column (0.7 × 8 cm) by passing through 10 mL of 0.2 M ZnCl2 . 13. Wash the column with 20 mL of buffer 7.2. 14. Apply the dialyzed sample to the Zn2+ -chelating Sepharose Fast Flow column and wash the column with buffer 7.2 until A280 nm reaches to the baseline.

136

Shimokawa and Nagase

15. Elute the bound active MMP-7 with buffer 7.3 containing 0.15 M NaCl. 16. Wash the column with the same buffer until A280 nm reaches to the baseline. 17. Elute proMMP-7 with buffer 7.3 containing 1 M NaCl. 18. Identify fractions containing proMMP-7 by assaying MMP-7 activity after activation of a 50–100 ␮L portion with APMA or by SDS-PAGE. ProMMP-7 exhibits a band of 28 kDa on SDS-PAGE. 19. Pool fractions containing proMMP-7 and concentrate it to 1.5 mL with an Amicon YM-10 membrane. Step 5: Gel Filtration on Sephacryl S-200 20. Apply the sample to a gel filtration column of Sephacryl S-200 (1.5 × 15 cm) equilibrated with TNC buffer containing 0.4 M NaCl. 21. Analyze fractions for proMMP-7 (28 kDa) by SDS-PAGE (12.5% acrylamide). 22. Collect fractions containing proMMP-7, and store at 4◦ C for a short term (up to 4 weeks), or at –20◦ C for a longer term. 3.4.4. Purification from Postpartum Rat Uterus (see Note 26)

This method is based on Woessner and Taplin (4). All purification steps should be carried out at 4◦ C unless otherwise mentioned. Step 1: Extraction 1. Dissect rat uteri (12 g, 9 uterine tissues) at 1 day postpartum free of mesentery and cervix. 2.

Rinse the tissue with saline, and mince and homogenize in 10 vol of 10 mM CaCl2 containing 0.25% (w/v) Triton X-100 in a glass homogenizer.

3.

Centrifuge the homogenate at 600g for 30 min.

4.

Re-homogenize the precipitate in 10 vol of buffer 7.4; place this suspension in metal centrifuge tubes, heat at 60◦ C for 4 min, and then centrifuge at 17,500g for 30 min at 4◦ C. Save the supernatant for proMMP-7 purification.

Step 2: (NH4 )2 SO4 Fractionation 5. Add solid (NH4 )2 SO4 to the uterine extract to 25% saturation; stand the solution for 30 min on ice and remove the precipitate by centrifugation at 20,000g for 40 min. 6.

Add solid (NH4 )2 SO4 to the supernatant to 75% saturation; stand the solution for 30 min on ice.

7.

Collect the precipitate by centrifugation at 20,000g for 40 min and dissolve the pellet in 40 mL of buffer 7.5.

Purification of MMPs and TIMPs

137

Step 3: Gel Filtration on Ultrogel AcA54 8. Apply the ammonium sulfate fraction to a column of Ultrogel AcA54 (5 × 50 cm) equilibrated with buffer 7.5. The latent enzyme emerges at about 680 mL, the volume between 630 and 730 mL. 9.

Assay for MMP-7 activity with Azocoll or a peptide substrate after APMA activation; pool the fractions with proMMP-7. This position is close to molecular weight marker of 25 kDa.

Step 4: Blue-Sepharose Chromatography 10. Apply the sample from Ultrogel AcA54 to a column of Blue-Sepharose (1.5 mL) equilibrated with buffer 7.4 and washed with the same buffer until A280 nm reaches to the baseline. 11. Elute the enzyme with a linear NaCl gradient (0.2–2.0 M) in buffer 7.5. The enzyme emerges at about 0.8 M NaCl. At this stage the enzyme is only about 5% pure. 12. Dialyze the sample from the Blue-Sepharose column against buffer 7.6. Step 5: Zn2+ -Chelating Sepharose Fast Flow Chromatography 13. Make Zn2+ -chelating Fast Flow column (1.5 mL) by passing through 10 mL of 0.2 M ZnCl2 . 14. Wash the column with 20 mL of buffer 7.6. 15. Apply the dialyzed sample to a column of Zn2+ – Sepharose Fast Flow and wash the column with buffer 7.7. 16. Elute the enzyme with a linear NaCl gradient (0–0.5 M) in buffer 7.7. The enzyme is eluted at about 50 mM NaCl. 17. Dialyze the eluted peak against 25 mM Tris-HCl (pH 7.5), 5 mM CaCl2 , 0.02% NaN3 , 0.05% Brij 35. Step 6: Gel Filtration on Ultrogel AcA54 (see Note 27) 18. Apply the dialyzed sample to a column of Ultrogel AcA54 (1.6 × 80 cm) equilibrated with buffer 7.5. 19. Store the sample as in Section 3.4.3. 3.5. Purification of proMMP-8 and MMP-8

The procedures are based on Engelbrecht et al. (17) and Kn¨auper et al. (18).

3.5.1. Assays

Same as MMP-1 (see Section 3.1.1).

3.5.2. Activation of proMMP-8

Same as MMP-1 (see Section 3.1.2) except activation by MMP-3 does not require APMA (19).

3.5.3. Preparation of Hydroxamate-Sepharose

1. Allow to swell 15g of activated CH-Sepharose 4B in 1 mM HCl and wash the swollen gel with 3 L of 1 mM Tris-HCl

138

Shimokawa and Nagase

on a glass sintered funnel. This gives 45 mL of gel with a capacity of 5–7 ␮mol/mL. 2. Dissolve 180 mg of Pro-Leu-Gly-NHOH in 45 mL of 10 mM sodium bicarbonate (pH 8.0) and mix with the gel for 60 min at room temperature. 3. Wash the coupled gel with 0.1 M Tris-HCl (pH 8.0), 0.5 M NaCl, and shake the gel gently in this buffer for 1 h at room temperature. 4. Wash the coupled gel three times, alternating with 0.1 M sodium acetate (pH 4.0), 0.5 M NaCl and with 0.1 M TrisHCl (pH 8.0), 0.5 M NaCl. 5. Store the coupled gel at 4◦ C in 50 mM Tris-HCl (pH 7.5), 0.5 M NaCl, 10 mM CaCl2 , 0.02% NaN3 . 3.5.4. Purification of proMMP-8

Step 1: Preparation of Buffy Coat Extract 1. Add plasmatonin solution to buffy coat (2 L from 20 L of human blood) in a 1:4 (v/v) ratio. 2.

Stand buffy coat for 75 min and aspirate leukocytes in the supernatant.

3.

Centrifuge the collected supernatant to pellet leukocytes at 150g for 20 min.

4.

Suspend the pellet in 0.45% NaCl, add 6 vol of water, and shake for 30 s for hypotonic hemolysis. Immediately after this process, add 2.4 vol of 3.5% NaCl to make the solution isotonic (0.9% NaCl).

5.

Centrifuge the suspension at 150g for 10 min and wash the pellet with 0.9% NaCl and pellet the cells again by centrifugation at 150g for 10 min.

6.

Suspend the cells in an equal volume of buffer 8.1 containing 2 mM benzamidine, 1 mM PMSF, and 1 mM DFP and homogenize in a Virtis homogenizer.

7.

Freeze-thaw the homogenate five times in dry ice/acetone and in a 35◦ C water bath.

8.

Dilute the homogenate to approximately 300 mL with buffer 8.1 containing 2 mM benzamidine, 1 mM PMSF, and 1 mM DFP.

9.

Centrifuge the cell homogenate at 48,000g for 30 min and save the supernatant as crude buffy coat extract.

Step 2: Zn2+ -Chelating Sepharose Fast Flow Chromatography 10. Make a column of Zn2+ -chelating Sepharose Fast Flow column (3.4 ×19 cm) by slowly passing through 200 mL of 0.2 M ZnCl2 and wash the column with buffer 8.2. 11. Apply the crude buffy coat extract to the column of Zn2+ chelating Sepharose Fast Flow column.

Purification of MMPs and TIMPs

139

12. Wash the column with buffer 8.2 until the A280 nm reaches the baseline. Collect the unbound protein peak. Step 3: Q-Sepharose Fast Flow Chromatography (see Note 28) 13. Apply the flow-through fraction from the Zn2+ -chelating column to a column of Q-Sepharose Fast Flow (3.5 × 25 cm) equilibrated with buffer 8.2. 14. Wash the column until A280 nm reaches the baseline and elute the bound proMMP-8 with a linear NaCl gradient (0–0.15 M; 2 × 600 mL) in buffer 8.2. 15. Collect the initial protein peak up to a conductivity of 4.5 mS/cm (low conductivity fraction) and fractions with conductivity 4.5–7.0 mS/cm (high conductivity fraction). Step 4: Orange-Sepharose CL-6B Chromatography 16. Dialyze the low conductivity fractions from the QSepharose column against buffer 8.3 and apply to a column of Orange-Sepharose CL-6B (4 ×25 cm) equilibrated in buffer 8.3. 17. Wash the column with buffer 8.3 until A280 nm reaches the baseline and elute the bound proMMP-8 with nonlinear NaCl gradient [0.1 M (600 mL)–2 M (250 mL)] in buffer 8.3. 18. Pool fractions with MMP-8 activity after activation with APMA. ProMMP-8 elutes in fractions with conductivity around 35–50 mS/cm. Step 5: Gel Filtration on Sephacryl S-300 19. Concentrate the sample from the Orange-Sepharose column to 1.5 mL using Diaflo apparatus with an Amicon XM-50 membrane, and apply it to a gel filtration column of Sephacryl S-300 (1.8 × 80 cm) equilibrated with buffer 8.4. 20. Analyze fractions for proMMP-8 by assaying MMP-8 activity after activation with APMA and by SDS-PAGE. The main protein peak contains proMMP-8 of 85 kDa on SDSPAGE (see Notes 29–31). 3.5.5. Purification of the Active Form of MMP-8

The method is based on Moore and Spilburg (20). 1. Use the high conductivity fraction containing proMMP-8 from the Q-Sepharose column. Step 1: Orange-Sepharose CL-6B Chromatography 2. The procedure is essentially identical to that of proMMP-8 (see Section 3.5.4). Pool proMMP-8 fractions detected by peptidolytic activity after activation with APMA. 3. Concentrate the pool from the Orange-Sepharose CL-6B column to 100 mL by Diaflo apparatus with a Millipore

140

Shimokawa and Nagase

XM50 membrane, and activate proMMP-8 in the sample by incubating with 1 mM HgCl2 at 37◦ C for 1 h. Step 2: Hydroxamate-Sepharose Affinity Chromatography 4. Dialyze the activated MMP-8 fraction against buffer 8.5 overnight at 4◦ C, and apply the sample to a column of hydroxamate (Pro-Leu-Gly-NHOH)-Sepharose (1.2 × 15 cm) equilibrated with the same buffer. 5. Wash the column with buffer 8.5 to remove all unbound proteins. 6. Elute MMP-8 with buffer 8.6. 7. Collect 6 mL fractions into tubes containing 2 mL of buffer 8.7. Monitor the MMP-8 peak by measuring peptidolytic activity, pool active fractions, and dialyze against buffer 8.5. Step 3: Gel Filtration on Sephacryl S-300 8. Concentrate the sample using Diaflo apparatus with a Millipore XM50 membrane to 1.5 mL and apply it to a column of Sephacryl S-300 (1.8 × 80 cm) equilibrated with buffer 8.5. The main protein peak contains active form of MMP-8 that exhibits a protein band of 64 kDa on SDS-PAGE (see Notes 29–31). 3.6. Purification of proMMP-9 3.6.1. Preparation of Conditioned Medium

1. Grow U937 (or HT1080) cells in Dulbecco’s modified Eagle’s medium (DMEM) containing 10% fetal calf serum. 2. After confluency, wash the cells with phosphatebuffered saline (PBS) three times and treat with PMA (50 ng/mL) in serum-free DMEM supplemented with 0.2% (w/v) ␣-lactalbumin hydrolyzate (LH). 3. Harvest the conditioned medium after 2–3 days and store at –20◦ C after removal of the cells and cell debris by centrifugation at 400g for 5 min.

3.6.2. Assays

3.6.3. Activation of proMMP-9

MMP-9 activity is measured after activation with APMA using 3 H-gelatin (see Chapter 15), a synthetic substrate (e.g., Mca-ProLeu-Gly-Leu-Dpa-Ala-Arg-NH2 ) (see Chapter 24) or directly applying to gelatin zymography (see Chapter 16). 1. Dilute 0.2 M APMA stock solution to 2 mM with TNC buffer. 2. Add an equal volume (50–100 ␮L) of 2 mM APMA to proMMP-9 prep (0.1–10 ␮g/mL) and incubate at 37◦ C for 24 h.

Purification of MMPs and TIMPs

3.6.4. Purification Procedures

141

The method is based on Morodomi et al. (5). Step 1: Gelatin-Sepharose Chromatography 1. Apply the serum-free conditioned medium of U937 cells or HT1080 cells (500 mL) to a column of gelatin-Sepharose (1.5 × 15 cm) equilibrated with TNC buffer. 2. Wash the column with TNC buffer until A280 nm reaches the baseline. 3. Elute the bound proMMP-9 with 5% (v/v) DMSO in TNC buffer. 4. Analyze the eluted fractions for proMMP-9 by direct application to gelatin zymography or by detecting gelatinase activity after treating the sample with 1 mM APMA for 24 h at 37◦ C. ProMMP-9 is detected as Mr around 92,000 on gelatin zymography (see Note 32). 5. Pool the fractions containing proMMP-9 and dialyze against TNC buffer. Step 2: Con A-Sepharose Affinity Chromatography 6. Apply the dialyzed sample to a column of Con A-Sepharose (3 mL) equilibrated with TNC buffer. 7.

Wash the column with the same buffer until the A280 nm reaches the baseline.

8.

Elute the bound proMMP-9 with 1 M ␣-methyl-Dmannoside in TNC buffer (see Note 33).

9.

Collect the protein peak eluted from the column and concentrate it to 1.5 mL using Diaflo apparatus with a Millipore YM-10 membrane.

Step 3: Gel Filtration Chromatography on Sephacryl S-200 10. Apply the concentrated proMMP-9 sample to a column of gel filtration chromatography on Sephacryl S-200 (1.5 × 110 cm) equilibrated with TNC buffer. 11. Analyze the fractions by SDS-PAGE and by gelatin zymography. The earlier fractions from this column contain the proMMP-9-TIMP-1 complex (92 and 30 kDa bands) and the later fractions contain proMMP-9 (92 kDa) free of TIMP-1 (29 kDa) (see Note 34). 3.7. Purification of proMMP-10

Same as MMP-3 (see Section 3.3.1).

3.7.1. Assays 3.7.2. Activation of proMMP-10

Same as for proMMP-3 (see Section 3.3.2).

3.7.3. Purification Procedures

The method is based on Nakamura et al. (7). Step 1: DEAE-Cellulose Chromatography 1. Concentrate the conditioned media of OSC-20 cells 10– 20-fold; apply the sample to a column of DEAE-cellulose

142

Shimokawa and Nagase

(2.5 × 8.5 cm) equilibrated with TNC buffer without Brij 35. 2.

Wash the column with the same buffer until A280 nm reaches the baseline; pool unbound fractions.

Step 2: Green A Dyematrex Gel Chromatography (see Note 35) 3. Apply the unbound sample to a column of Green A Dyematrex gel (2.5 × 8 cm) equilibrated with TNC buffer without Brij 35, and wash the column with TNC buffer. 4.

Elute the bound proteins with a linear gradient of NaCl (0.15–2.0 M) in TC buffer.

5.

Monitor the MMP-10 activity of the eluted fractions after activation with APMA.

6.

Pool fractions with MMP-10 activity.

Step 3: Gelatin-Sepharose Chromatography (see Note 36) 7. Apply the sample from the Green A Dyematrex gel to a column of gelatin-Sepharose (2.5 × 4 cm) equilibrated with TNC buffer. 8.

Wash the column with TNC buffer until A280 nm reaches the baseline and collect unbound fractions.

Step 4: Immunoadsorbent Chromatography on Anti-(human MMP-1) IgG-Sepharose and on Anti-(MMP-3) IgG-Sepharose (see Notes 37 and 38) 9. Apply the sample from the gelatin-Sepharose column to a column of Affi-Gel 10 coupled with anti-(human MMP-1) IgG (clone 78-12G8, Fuji Chemical Industries, Ltd.) (2 mL) and a column of Affi-Gel 10 coupled with anti-(human MMP-3) IgG (clone 55-2A4, Fuji Chemical Industries, Ltd.) (2 mL) equilibrated with TNC buffer. Collect flow-through fractions from each chromatography. Step 5: DEAE-Cellulose Chromatography 10. Combine the flow-through fraction of the immunoadsorbent columns and dialyze it against buffer 10.1. 11. Apply the dialyzed sample to a column of DEAE-cellulose (1.5 × 12 cm) equilibrated with the same buffer. 12. Wash the column with the same buffer until A280 nm reaches to the baseline; elute the bound proMMP-10 by a linear gradient of NaCl (0–0.5 M) in the same buffer. 13. Monitor MMP-10 activity after activation with APMA and combine active fractions. Step 6: Gel Filtration on Sephacryl S-200 14. Concentrate the sample from the DEAE-cellulose column to 1.5 mL using Diaflo apparatus with a Millipore YM-10 membrane and apply it to a gel filtration column of Sephacryl S-200 (1 × 95 cm) equilibrated with TC buffer containing 0.4 NaCl.

Purification of MMPs and TIMPs

143

15. Monitor A280 nm and MMP-10 activity after activation with APMA and analyze for proMMP-10 by SDS-PAGE. ProMMP-10 exhibits a protein band of 56 kDa on SDSPAGE (see Notes 39 and 40). 3.8. Purification of MMP-12 3.8.1. Preparation of Mouse MacrophageConditioned Medium

The method is based on Banda and Werb (21). MMP-12 is purified as an active enzyme. 1. Inject 1.0 mL of 3% Brewer’s thioglycollate medium into mice intraperitoneally (see Note 41). 2. After 4 days, obtain peritoneal macrophage from peritoneal lavarge with PBS containing 100 U heparin/mL. 3. Wash cells with PBS and suspend in DME medium containing 10% (v/v) fetal calf serum. 4. Culture at a density of 1.5 × 107 –2 × 107 cells per 75 cm2 flask and allow cells to adhere for 2–4 h. 5. Wash cells with Hank’s balanced salt solution twice. 6. Add 10 mL of DME-LH medium containing 2 ␮M colchicine to each flask. 7. After 48 h culture, harvest the conditioned medium, which is centrifuged at 8,000g for 10 min, and store at –20◦ C. 8. Add fresh DME-LH medium containing 2 ␮M colchicine and culture for 48 h. Repeat this process to obtain about 500–600 mL of conditioned medium.

3.8.2. Assays

MMP-12 activity is measured by measuring the amount of radioactivity released after incubation of the enzyme preparation with [3 H]-labeled elastin. MMP-12 hydrolyzes a synthetic substrate Mca-Pro-Leu-Gly-Leu-Dpa-Ala-Arg-NH2 (8), but this substrate is hydrolyzed by other MMPs. Therefore, at the initial stage of purification, the assay using [3 H]-labeled elastin is recommended.

3.8.3. Purification Procedures

All procedures should be carried out at 4◦ C. Step 1: Dialysis and Centration of Conditioned Medium 1. Dialyze the pooled conditioned medium against three changes of 20 vol of 10 mM NH4 HCO3 and freeze-dry the sample. 2.

Dissolve the freeze-dried material with 4 mL of cold water and dialyze it against 4 liters of buffer 12.1.

Step 2: DEAE-Sephadex A-25 3. Apply the dialyzed condition medium to a column of DEAE-Sephadex A-25 (0.9 × 27 cm) equilibrated with buffer 12.1 and wash the column with the same buffer until A280 nm reaches the baseline.

144

Shimokawa and Nagase

4. Elute the bound protein with a linear gradient (0–0.7 M; 80 mL total volume) in buffer 12.1 and collect 2-mL fractions. 5. Assay for protein (A280 nm ) and enzyme activity. [Note: Two major peaks of elastase are detected; one eluted in the flow-through fraction and a second peak eluted with conductivity 2–15 mS/cm.] 6. Pool two peaks of elastase activity separately (pool I and pool II, respectively). 7. Collect fractions with conductivity ranging from 15 to 30 mS/cm as pool III. 8. Freeze-dry pools I, II, and III separately; dissolve each pool into 1.8 mL of water and add 0.2 mL of glycerol. 9. Apply each pool of MMP-12 (pools I–III) to a gel filtration column of Ultrogel AcA54 (1.5 × 87 cm) equilibrated with buffer 12.2. Three peaks of elastase activity can be resolved by this gel filtration (peak A, Kav 0.13; peak B, Kav 0.33; peak C, Kav 0.61). Pool I gives peaks A and B in a ratio of 1:4; pool II gives peaks A, B, and C in a ratio of 1:15:4; and pool III gives four peaks of elastase activity. The peak B from pool II gives the highest specific activity with the main protein of 21 kDa (see Note 42). 10. Freeze active fractions in aliquots at –70◦ C for storage. 3.9. Purification of TIMP-1

3.9.1. Assay for TIMP Activity (see Notes 43–45)

TIMP-1 is a glycoprotein of 29–30 kDa. Methods discussed below are based on Sudbeck et al. (22) and Cawston et al. (23). The latter method is rapid but requires anti-(human TIMP-1) IgG-Sepharose. 1. Mix 45 ␮L of MMP-3 (1 ␮g/mL) in TNC buffer and various amounts (5–45 ␮L) of the inhibitor sample; make the mixture to a final volume of 90 ␮L by adding TNC buffer. 2. As positive control, mix 45 ␮L of MMP-3 (1 ␮g/mL) in TNC buffer and 45 ␮L of TNC buffer in duplicate. This gives 100% MMP-3 activity. 3. As negative control, place 90 ␮L of TNC buffer in an assay tube in duplicate. 4. Stand mixtures at 37◦ C for at least 30 min. 5. Add 10 ␮L of 10 ␮M Mca-Pro-Leu-Gly-Leu-Dpa-Ala-ArgNH2 . 6. Incubate 30 min at 37◦ C. 7. Stop the reaction by adding 900 ␮L of 3% acetic acid. 8. Read fluorescence with excitation at 325 nm and emission at 393 nm.

Purification of MMPs and TIMPs

3.9.2. Purification of TIMP-1 from the Conditioned Medium of Bovine Skin Fibroblasts

145

This method is based on Sudbeck et al. (22). Step 1: Heparin-Sepharose Chromatography. 1. Apply serum-free conditioned medium (3 L) of bovine fetal skin fibroblasts to a column of heparin-Sepharose (1.6 × 12 cm) equilibrated with TCO buffer. 2.

Wash the column with TCO buffer until A280 nm reaches the baseline and elute the bound TIMP-1 with a 500 mL linear gradient (0–2 mg/mL) of dextran sulfate in TCO buffer (5 mL/fraction).

3.

Monitor TIMP-1 activity in each fraction and pool fractions containing MMP inhibitory activity.

Step 2: DEAE-Sepharose Column 4. Apply the pooled fractions from the heparin-Sepharose column to a column of DEAE-Sepharose (0.5 mL) equilibrated with TCO buffer. 5.

Wash the column with the same buffer until A280 nm reaches to the baseline. Collect the unbound protein peak.

Step 3: Concentration by Heparin-Sepharose Chromatography 6. Apply the flow-through fraction of the DEAE-Sepharose column to a column of heparin-Sepharose (0.5 mL) equilibrated with TCO buffer. 7.

Wash the column with the same buffer until A280 nm reaches the baseline; elute TIMP-1 with TCO buffer containing 2 M NaCl.

8.

Monitor the protein peak by reading A280 nm .

9.

Collect the protein peak and dialyze it against 0.1% trifluoroacetic acid (TFA) in water.

Step 4: C4 Reverse-Phase HPLC 10. Apply the dialyzed sample to a C4 reverse-phase column (Vydac, Hesperia, CA) equilibrated in 0.1% TFA; elute TIMP-1 with a linear gradient of 30–50% acetonitrile containing 0.1% TFA over 50 min. 11. Analyze components of the major peaks for TIMP-1 by SDS-PAGE and pool the fractions containing TIMP-1. TIMP-1 exhibits a band around 29-30 kDa. Typically, about 200 ␮g of TIMP-1 is recovered from 3 L of the conditioned medium (see Note 46 and 47). 3.9.3. Purification of TIMP-1 from Human Plasma

This method is based on Cawston et al. (23). 1. Dialyze human plasma (400 mL) against 20–40 vol of buffer T1-1 at 4◦ C overnight twice. 2. Remove precipitates formed by centrifugation at 40,000g for 1 h and pass the supernatant through a column of Sepharose

146

Shimokawa and Nagase

6B (1.6 × 30 cm) equilibrated with buffer T1-1 to remove any plasma protein that binds to Sepharose nonspecifically. 3. Wash the column with buffer T1-1 and collect the flowthrough fraction. Anti-(Human TIMP-1) Affinity Chromatography 4. Apply the flow-through fraction to a column of anti-(human TIMP-1) IgG-Affi-Gel 10 (5 × 0.6 cm). 5. Wash the column extensively, elute the bound TIMP-1 with buffer T1-2, and collect 2-mL fractions; pool the fractions with a protein peak. 6. Concentrate the sample eluted from the affinity chromatography to 1.5 mL by Amicon Diaflo with a YM-10 membrane and apply the concentrated sample to a gel filtration column on Ultrogel AcA44 (1.6 × 88 cm) equilibrated with buffer T1-3. A single peak of MMP inhibitory activity is eluted at a position corresponding to Mr 30,500. About 160 ␮g of TIMP-1 is purified from 400 mL plasma (see Note 46 and 47). 3.10. Purification of TIMP-2 3.10.1. Method 1

This method is based on the isolation of the proMMP-2TIMP-2 complex from the conditioned medium of fibroblasts and the dissociation of TIMP-2 from the complex (11, 12) (see Note 48). Step 1: Gelatin-Sepharose Chromatography 1. Apply the conditioned culture medium of human fibroblasts (2 L) to a column of gelatin-Sepharose (2.5 × 5 cm) equilibrated with TNC buffer without Brij 35. 2.

Wash the column with TNC buffer containing 1% DMSO until A280 nm reaches the baseline.

3.

Elute the proMMP-2-TIMP-2 complex with 5% DMSO in TNC buffer without Brij 35.

4.

Monitor A280 nm and pool the protein peak.

Step 2: Gel Filtration on Sephacryl S-200 5. Dialyze the sample against 200 vol of TNC buffer and concentrate by Diaflo apparatus with an Amicon YM-10 membrane. 6.

Apply the concentrated samples to a gel filtration column of Sephacryl S-200 (1.5 × 110 cm) equilibrated with TNC buffer.

7.

Monitor A280 nm for protein and analyze the proMMP-2TIMP-2 complex by SDS-PAGE (10% acrylamide).

8.

Pool the fraction containing the proMMP-2-TIMP-2 complex.

Purification of MMPs and TIMPs

147

Step 3: Dissociation of TIMP-2 from the proMMP-2-TIMP-2 Complex and Gel Filtration on Sephacryl S-200 9. Concentrate the pool from Sephacryl S-200 to 1 mL and add 1 mL of 0.4 M glycine-HCl (pH 2.8), 50 mM EDTA. 10. Heat the solution at 45◦ C for 30 min. 11. Apply the sample to a column of gel filtration on Sephacryl S-200 (1.6 ×110 cm) equilibrated with TNC buffer containing 25 mM EDTA. 12. Monitor A280 nm for protein and the presence of TIMP-2 by SDS-PAGE (10% acrylamide). 13. Pool the fractions containing TIMP-2 and store the sample at –70◦ C in aliquots (see Notes 49 and 50).

3.10.2. Method 2

This is based on the method by Umenishi et al. (24) using the conditioned medium of HLE human hepatoma cells. Step 1: (NH4 )2 SO4 Precipitation and Gel Filtration on Cellulofine GCL-2000-m 1. Add solid (NH4 )2 SO4 to pooled serum-free conditioned medium of HLE cells (about 4 L) to a final saturation of 80% at 4◦ C and allow the mixture to stand overnight at 4◦ C. 2.

Collect the precipitate by centrifugation at 20,000g at 40 min.

3.

Dissolve the pellet with 10 mL of buffer 1; centrifuge the solution at 20,000g for 5 min and use the supernatant for purification.

Step 2: Gel Filtration on Cellulofine GCL-2000-m 4. Apply the supernatant to a column of Cellulofine GCL2,000-m (2.6 × 98 cm) equilibrated with buffer T2-1. 5.

Monitor A280 nm and MMP inhibition activity.

6.

Pool fractions containing MMP inhibitory activity and dialyze against 200 vol of buffer T2-2.

Step 3: Reactive Red-Agarose Chromatography 7. Apply the dialyzed sample to a column of Reactive Redagarose (1.5 × 7 cm) equilibrated with buffer T2-2. 8. Wash the column until A280 nm reaches the baseline and elute the bound materials with a linear gradient of NaCl (0–2 M; total 200 mL). 9. Monitor A280 nm and MMP inhibitory activity, and pool fractions with the inhibitor. TIMP-2 is eluted at NaCl concentration of 0.9–1.3 M. 10. Dialyze the sample against 0.05% TFA.

148

Shimokawa and Nagase

Step 4: Reverse-Phase HPLC 11. Apply the sample in 0.05% TFA to reverse-phase HPLC on a SynChropak RP-4 column (0.41 × 25 cm) equilibrated with 0.05% TFA. 12. Elute TIMP-2 with a linear gradient of acetonitrile (0– 60%) in 30 mL of 0.05% TFA. 13. Monitor A280 nm and MMP inhibitory activity. TIMP-2 is eluted at an acetonitrile concentration of about 40%. 14. Analyze fractions containing TIMP-2 by SDS-PAGE (12.5% acrylamide). Later fractions of the inhibitory peak contains a small amount of TIMP-1 (see Notes 49 and 50).

4. Notes 1. Purification procedures can be carried out either at room temperature when the starting material is a conditioned culture medium. However, when a tissue extract is used, all purification steps should be at 4◦ C. 2. Many MMPs are stable as proenzyme. When activated, they often lose their C-terminal hemopexin domains. Therefore, it is generally recommended to purify MMPs as proenzyme and activate them after purification of proMMPs. 3. Many connective tissue cells in culture stimulated with IL-1 or PMA produce proMMP-1, proMMP-2, and proMMP-3. These three zymogens can be purified from the same conditioned medium. The recommended purification procedures of proMMP-1, proMMP-2, and proMMP-3 from the conditioned medium of IL-1-treated fibroblasts are as follows: a. Concentrate the medium about 20-fold and apply to a column of DEAE-cellulose as in Section 3.1.4. The flow-through fraction contains proMMPs-1, -2, and -3. b. Apply the flow-through fraction to a column of gelatinSepharose as in Section 3.2.3. The flow-through fraction contains proMMP-1 and proMMP-3. The bound proMMP-2 is eluted from the column with 5% DMSO in TNC buffer and further purified as in Section 3.2.3. c. Apply the flow-through fraction from the gelatinSepharose column to a column of Green A Dyematrex gel as in Section 3.1.4. Elute proMMP-3 and proMMP-1 with 0.3 M NaCl and 0.5 M NaCl in TC buffer, respectively.

Purification of MMPs and TIMPs

149

d. Further purify proMMP-3 and proMMP-1 by gel filtration Sephacryl S-200. ProMMP-1 and proMMP-3 fractions may contain small amounts of proMMP-3 and proMMP-1. Because they are close in molecular mass, it is recommended to check contamination by Western blotting analysis. The contaminating proMMPs may be removed using immunoadsorbent chromatography using appropriate antibodies. 4. Assays with synthetic substrates are convenient to measure MMP activities but most substrates are hydrolyzed by many MMPs. To purify a specific MMP, assays methods with protein substrates, Western blotting are often required. 5. Human proMMP-1 is activated by treatment with APMA in the presence or absence of MMP-3 (10). On activation, the proMMP-1 is processed to the active forms (41 kDa) which may undergo autolytic cleavage into the C-terminal domain (25 kDa) and catalytic domain (22 kDa). In the case of MMP-1 activated with APMA and MMP-3, it may be necessary to remove MMP-3 depending on the study. However, MMP-3 does not participate in interstitial collagen degradation (10). 6. When proMMP-1 is activated with APMA and MMP-3, the primary product is the 41-kDa MMP-1. 7. Step 1: DEAE-cellulose chromatography removes glycosaminoglycans and phenol red. This step is essential for reproducible chromatography of Green A Dyematrex gel in the next step. 8. Step 2: Green A Dyematrex gel chromatography. The majority of proMMP-2 and proMMP-3 are recovered in the 0.3 M NaCl fraction, whereas about 75% of proMMP-1 is recovered in the 0.5 M NaCl fraction. 9. The purified proMMP-1 fraction often contains a small amount of proMMP-2 and proMMP-3. ProMMP-2 may be removed by chromatography on a gelatin-Sepharose (Pharmacia) and ProMMP-3 on an anti-(MMP-3) IgGAffi-Gel 10 affinity chromatography. 10. Step 5: Con A-Sepharose chromatography. The sample passed through or eluted from this column is contaminated with a small amount of Con A from the resin. Thus, it is essential to purify it further by gel filtration as indicated. 11. Store proMMP-1 at 4◦ C for short storage (up to 4 weeks), but at –70◦ C for longer storage. 12. ProMMP-2 is rapidly activated by APMA, but it is also gradually inactivated by autodegradation. Therefore, when the enzyme and substrate are incubated for a long time

150

Shimokawa and Nagase

(>2 h), it is recommended to incubate proMMP-2, the substrate to be tested and 1 mM APMA together. 13. Gelatin-Sepharose chromatography adsorbs proMMP2, the proMMP-2-TIMP-2 complex, and fibronectin. These three components (molecular mass of fibronectin [440,000], proMMP-2-TIMP-2 complex [95,000], and proMMP-2 [72,000]) are reasonably well separated by gel filtration on Sephacryl S-200. 14. When the conditioned medium of cytokine-stimulated cells is used it may also contain proMMP-9 and the complex of proMMP-9 and TIMP-1. These two components can be removed by Con A-Sepharose chromatography. 15. Activation of proMMP-3 by APMA results in formation of both 45-kDa and 28-kDa forms. The latter is the catalytic domain lacking the C-terminal hemopexin-like domain. Both are equally active against various substrates including extracellular matrix macromolecules (25). 16. Activation of proMMP-3 by chymotrypsin generates primarily the 45-kDa form. 17. Step 1: DEAE-cellulose chromatography removes glycosaminoglycans and phenol red. This step is essential for reproducible chromatography of Green A Dyematrex gel in the next step. 18. About 50% of proMMP-3 is recovered in 0.3 M NaCl fraction of Green A Dyematrex gel chromatography. This fraction contains proMMP-2, which is effectively removed by the next gelatin-Sepharose step. About 75% proMMP-1 is recovered in 0.5 M NaCl. This fraction can be used for proMMP-1 purification (see Section 3.1.4). 19. Step 5 removes a small amount of proMMP-1 contamination in the proMMP-3 preparation. When an anti-(MMP1) affinity column is not available, dialyze the sample against 20 mM Tris-HCl (pH 7.5), 10 mM CaCl2 and apply to a column of CM-cellulose (1.5 × 5 cm) equilibrated with the same buffer and collect the unbound fraction (26). ProMMP-3 is recovered in the unbound fraction. The CM-cellulose step may be used after step 1, DEAE-cellulose chromatography. 20. ProMMP-3 is eluted from an immunoaffinity column with 6 M urea. Under these conditions proMMP-3 is stable and the majority (>90%) is recovered as a proenzyme. 21. Con A-Sepharose chromatography separates glycosylated and unglycosylated forms of proMMP-3, but each fraction after a Con A-Sepharose column contains a small amount

Purification of MMPs and TIMPs

151

of Con A from the column. This is removed by gel filtration on Sephacryl S-200. 22. ProMMP-3 is stable at 4◦ C at least for 4 weeks. For a longer storage, it is recommended to store at –20◦ C. 23. ProMMP-7 stored at 4◦ C is stable for at least 4 weeks. 24. ProMMP-7 stored at –20◦ C is activated to a 19-kDa form when it is thawed. 25. APMA-activated MMP-7 is stable at 23◦ C for at least 4 weeks. 26. Rat uterine extract contains several MMPs. These enzymes also hydrolyze Azocoll and synthetic substrates. However, the molecule mass of proMMP-7 (28 kDa) is smaller than that of other proMMPs (>55 kDa). For example, gel filtration in step 3 separates gelatinases, which are eluted in the void volume. 27. Rat proMMP-7 in 25 mM Tris-HCl (pH 7.5), 5 mM CaCl2 , 0.02% NaN3 , 0.05% Brij 35 interacts with Ultrogel AcA54 (step 6) and emerges later than normal (Kav = 0.65). The product after step 6 is homogeneous and shows a single band of 28 kDa on SDS-PAGE after staining with silver. However, a minor band of 19 kDa may appear sometimes due to partial activation of proMMP-7. 28. ProMMP-8 from the Q-Sepharose anion exchange column (step 4) is found at conductivity up to 7.0 mS/cm, but only fractions with the lower conductivity (up to 4.5 mS/cm) can be purified as proMMP-8. Fractions with higher conductivity contain proteins that cannot be removed in the subsequent purification steps. Those fractions can be used separately for another purification or for purification of the active form of MMP-8 by hydroxamate-Sepharose affinity chromatography (20) after separate purification on Orange-Sepharose CL-6B. 29. Immediate freezing and storage at –70◦ C in aliquots is required to prevent spontaneous activation of proMMP-8 or autolytic degradation of both active and latent MMP-8. 30. Storage at –10 to –20◦ C results in a slow loss in activity. 31. Frequent freeze-thawing of enzyme samples results in loss of activity. 32. The major components eluted from the gelatin-Sepharose column are proMMP9, proMMP-9-TIMP-1 complex, and possibly a small amount of proMMP-2. 33. The sample eluted from the Con A-Sepharose column contains both proMMP-9 and the proMMP-9-TIMP-1 com-

152

Shimokawa and Nagase

plex. The proMMP-9-TIMP-1 complex and proMMP-9 can be separated by gel filtration on Sephacryl S-200. Alternatively the proMMP-9-TIMP-1 complex may be removed by anti-(TIMP-1) IgG-Sepharose. In this case antiTIMP-1 antibody must be checked to see whether it recognizes TIMP-1 in the complex. Polyclonal antibodies raised against reduced human TIMP-1 (e.g., TIMP-1 eluted after SDS-PAGE under reducing conditions) do not recognize the native TIMP-1. 34. The proMMP-9-TIMP-1 complex may be isolated by collecting earlier fractions of gel filtration (step 3). The treatment of the complex with APMA does not exhibit gelatinolytic activity against 14 C-labeled gelatin in solution (27). Any gelatinolytic activity detected with the complex is indicative of the presence of free proMMP-9. 35. Green A Dyematrex gel chromatography (step 2). The majority of proteins do not bind to the column. The bound proMMP-10 may also be eluted from the column with 0.3 M NaCl in TC buffer. 36. Gelatin-Sepharose chromatography (step 3) removes proMMP-2 and proMMP-9. ProMMP-10 is recovered in the unbound fraction. 37. Anti-(human MMP-1) IgG-Sepharose and anti-(human MMP-3) IgG-Sepharose remove proMMP-1 and proMMP-3, respectively. Polyclonal antibodies against those proteins can also be used. 38. MMP-3 and MMP-10 are about 73% identical in sequence, but sheep anti-(human MMP-3) IgG does not cross-react with proMMP-10. 39. The final product has a major species of 56 kDa but it may contain minor protein bands of 47, 24, and 22 kDa. Both 47 and 24 kDa species are activated forms of MMP-10. 40. About 400 ␮g of proMMP-10 may be obtained from 600 mL of the conditioned medium of OSC-20 human oral squamous carcinoma cells. 41. About 1.5 × 107 exudate cells are obtained from each mouse injected with 3% Brewer’s thioglycollate medium. About 3 × 108 cells are required for a complete enzyme purification. 42. The zymogen of MMP-12 is estimated to be about 56 kDa (28), but proMMP-12 is readily activated during purification. Only the active form of 21 kDa is purified by this method. 43. MMP-1 and MMP-3, which may be obtained relatively easily, are recommended to use for TIMP inhibition assay,

Purification of MMPs and TIMPs

153

although any MMP, except MT1-MMP (MMP-14), may be used as a target enzyme. MMP-2 is not a good enzyme for routine detection of TIMP as it autodegrades after activation. 44. The concentration of target enzyme should be greater than 20 nM when the quantity of TIMP-1 is determined. 45. To determine the amount of TIMP, inhibition of MMP has to be 25–75%. When inhibition is >90%, dilute the inhibitor sample severalfold and reassay. 46. The A1% 280 nm,cm value of human TIMP-1 is 9.14. 47. TIMP-1 is very stable; >95% activity is recovered after heating the solution of 10 ␮g/mL at 95◦ C for 15 min. 48. TIMP-2 in the conditioned medium of fibroblasts exists primarily as a complex with proMMP-2. However, the medium usually contains free proMMP-2. The material eluted from the gelatin-Sepharose column is a mixture of the proMMP-2-TIMP-2 complex and proMMP-2, which are separated by gel filtration on Sephacryl S-200. 49. The A1% 280 nm,cm value of human TIMP-2 is 18.0. 50. Like TIMP-1, TIMP-2 is very stable. 51. After activation and during storage MMPs often undergo autodegradation. It is therefore important to titrate the active site of an activated MMP. This is accomplished by using TIMP-1, TIMP-2, or ␣2 -macroglobulin (␣2 M) which form a 1:1 complex with the enzyme. When ␣2 M is used, a large protein substrate must be used. The following procedure is recommended. a. Mix 10 ␮L of the MMP solution (about 100 nM) with 10 ␮L of 0, 20, 40, 60, . . ., 200 nM TIMP-1, TIMP-2, or ␣2 M. b. Incubate at 37◦ C for 1 h. c. Mix a 10-␮L portion of the mixture with 10 ␮L of an appropriate substrate for the MMP and measure the residual enzyme activity. d. Plot the enzyme activity against the inhibitor concentration. e. Draw a linear line through the plots and determine, or extrapolate, the inhibitor molarity that gives zero activity of the enzyme. This molarity of inhibitor is equal to that of the enzyme solution.

154

Shimokawa and Nagase

Acknowledgments We thank Linda Chung for useful comments. This work is supported by NIH grants AR39189 and AR40994. References 1. Massova, I., Kotra, L. P., Fridman, R., and Mobashery, S. (1998) Matrix metalloproteinases – structures, evolution, and diversification. FASEB J 12, 1075–1095. 2. Gomez, D. E., Alonso, D. F., Yoshiji, H., and Thorgeirsson, U. P. (1997) Tissue inhibitors of metalloproteinases—structure, regulation and biological functions. Eur J Cell Biol 74, 111–122. 3. Imai, K., Yokohama, Y., Nakanishi, I., Ohuchi, E., Fujii, Y., Nakai, N., and Okada, Y. (1995) Matrix metalloproteinase 7 (matrilysin) from human rectal carcinoma cells. Activation of the precursor, interaction with other matrix metalloproteinases and enzymic properties. J Biol Chem 270, 6691–6697. 4. Woessner, J. F., Jr. and Taplin, C. J. (1988) Purification and properties of a small latent matrix metalloproteinase of the rat uterus. J Biol Chem 263, 16918–16925. 5. Morodomi, T., Ogata, Y., Sasaguri, Y., Morimatsu, M., and Nagase, H. (1992) Purification and characterization of matrix metalloproteinase 9 from U937 monocytic leukemia and HT1080 fibrosarcoma cells. Biochem J 285, 603–611. 6. Windsor, L. J., Grenett, H., BirkedalHansen, B., Bodden, M. K., Engler, J. A., and Birkedal-Hansen, H. (1993) Cell typespecific regulation of SL-1 and SL-2 genes. Induction of the SL-2 gene but not the SL-1 gene by human keratinocytes in response to cytokines and phorbolesters. J Biol Chem 268, 17341–17347. 7. Nakamura, H., Fujii, Y., Ohuchi, E., Yamamoto, E., and Okada, Y. (1998) Activation of the precursor of human stromelysin 2 and its interactions with other matrix metalloproteinases. Eur J Biochem 253, 67–75. 8. Knight, C. G., Willenbrock, F., and Murphy, G. (1992) A novel coumarin-labelled peptide for sensitive continuous assays of the matrix metalloproteinases. FEBS Lett 296, 263–266. 9. Cawston, T. E. and Barrett, A. J. (1979) A rapid and reproducible assay for collagenase using [1-14 C]acetylated collagen. Anal Biochem 99, 340–345.

10. Suzuki, K., Enghild, J. J., Morodomi, T., Salvesen, G., and Nagase, H. (1990) Mechanisms of activation of tissue procollagenase by matrix metalloproteinase 3 (stromelysin). Biochemistry 29, 10261–10270. 11. Itoh, Y., Binner, S., and Nagase, H. (1995) Steps involved in activation of the complex of pro-matrix metalloproteinase 2 (progelatinase A) and tissue inhibitor of metalloproteinases (TIMP)-2 by 4-aminophenylmercuric acetate. Biochem J 308, 645–651. 12. Ward, R. V., Hembry, R. M., Reynolds, J. J., and Murphy, G. (1991) The purification of tissue inhibitor of metalloproteinases2 from its 72 kDa progelatinase complex. Demonstration of the biochemical similarities of tissue inhibitor of metalloproteinases-2 and tissue inhibitor of metalloproteinases-1. Biochem J, 278, 179–187. 13. Gunja-Smith, Z., Nagase, H., and Woessner, J. F., Jr. (1989) Purification of the neutral proteoglycan-degrading metalloproteinase from human articular cartilage tissue and its identification as stromelysin matrix metalloproteinase-3. Biochem J 258, 115–119. 14. Nagase, H. (1995) Human stromelysins 1 and 2. Methods Enzymol 248, 449–470. 15. Nagase, H., Fields, C. G., and Fields, G. B. (1994) Design and characterization of a fluorogenic substrate selectively hydrolyzed by stromelysin 1 (matrix metalloproteinase- 3). J Biol Chem 269, 20952–20957. 16. Ito, A. and Nagase, H. (1988) Evidence that human rheumatoid synovial matrix metalloproteinase 3 is an endogenous activator of procollagenase. Arch Biochem Biophys 267, 211–216. 17. Engelbrecht, S., Pieper, E., Macartney, H. W., Rautenberg, W., Wenzel, H. R., and Tschesche, H. (1982) Separation of the human leucocyte enzymes alanine aminopeptidase, cathepsin G, collagenase, elastase and myeloperoxidase. Hoppe Seylers Z Physiol Chem 363, 305–315. 18. Kn¨auper, V., Kramer, S., Reinke, H., and Tschesche, H. (1990) Characterization and activation of procollagenase from human

Purification of MMPs and TIMPs

19.

20.

21.

22.

23.

polymorphonuclear leucocytes. N-terminal sequence determination of the proenzyme and various proteolytically activated forms. Eur J Biochem 189, 295–300. Kn¨auper, V., Wilhelm, S. M., Seperack, P. K., DeClerck, Y. A., Langley, K. E., Osthues, A., and Tschesche, H. (1993) Direct activation of human neutrophil procollagenase by recombinant stromelysin. Biochem J 295, 581–586. Moore, W. M. and Spilburg, C. A. (1986) Purification of human collagenases with a hydroxamic acid affinity column. Biochemistry 25, 5189–5195. Banda, M. J. and Werb, Z. (1981) Mouse macrophage elastase. Purification and characterization as a metalloproteinase. Biochem J 193, 589–605. Sudbeck, B. D., Jeffrey, J. J., Welgus, H. G., Mecham, R. P., McCourt, D., and Parks, W. C. (1992) Purification and characterization of bovine interstitial collagenase and tissue inhibitor of metalloproteinases. Arch Biochem Biophys 293, 370–376. Cawston, T. E., Noble, D. N., Murphy, G., Smith, A. J., Woodley, C., and Hazleman, B. (1986) Rapid purification of tissue inhibitor of metalloproteinases from human plasma and identification as a gamma-serum protein. Biochem J 238, 677–682.

155

24. Umenishi, F., Umeda, M., and Miyazaki, K. (1991) Efficient purification of TIMP-2 from culture medium conditioned by human hepatoma cell line, and its inhibitory effects on metalloproteinases and in vitro tumor invasion. J Biochem (Tokyo) 110, 189–195. 25. Okada, Y., Nagase, H., and Harris, E. D., Jr. (1986) A metalloproteinase from human rheumatoid synovial fibroblasts that digests connective tissue matrix components. Purification and characterization. J Biol Chem 261, 14245–14255. 26. Lark, M. W., Walakovits, L. A., Shah, T. K., Vanmiddlesworth, J., Cameron, P. M., and Lin, T. Y. (1990) Production and purification of prostromelysin and procollagenase from IL-1 ␤-stimulated human gingival fibroblasts. Connect Tissue Res 25, 49–65. 27. Ogata, Y., Itoh, Y., and Nagase, H. (1995) Steps involved in activation of the promatrix metalloproteinase 9 (progelatinase B)-tissue inhibitor of metallopro-teinases1 complex by 4-aminophenylmercuric acetate and proteinases. J Biol Chem 270, 18506–18511. 28. Shapiro, S. D., Kobayashi, D. K., and Ley, T. J. (1993): Cloning and characterization of a unique elastolytic metalloproteinase produced by human alveolar macrophages. J Biol Chem 268, 23824–23829

Section III Detection of MMPs and TIMPs

Chapter 9 Real-Time PCR Expression Profiling of MMPs and TIMPs Caroline J. Pennington and Dylan R. Edwards Abstract Quantitative reverse transcriptase polymerase chain reaction enables the accurate quantification of gene R expression in cultured cells or small tissue samples. In this chapter, we describe the use of Taqman technology to measure expression of matrix metalloproteinases and related genes. Key words: RT-PCR, mRNA, gene expression, CT .

1. Introduction Quantitative reverse transcriptase polymerase chain reaction (qRT-PCR) is described as one of the enabling tools of the genomic age. The method overcomes some of the main challenges associated with characterization and accurate quantification of protease gene expression in tissues and cell lines (1). Unlike earlier methods of quantifying gene expression, such as northern blotting which requires 5–30 ␮g of RNA, qRT-PCR can accurately detect as little as 100 copies of target sequence in a 5 ng pool of reverse-transcribed cDNA, equivalent to about 1 copy per cell (2). The technique can therefore be adapted when clinical tissue or cell samples are limited. As qRT-PCR simultaneously detects and quantifies the presence of a specific region of DNA during the early, efficient phase of PCR, results obtained are more accurate and reproducible than semi-quantitative PCR techniques. The relative ease of methodology has resulted in high-throughput profiling studies of large gene families in extensive collections of samples (3–6).

I.M. Clark (ed.), D.A. Young, A.D. Rowan (Consulting editors), Matrix Metalloproteinase Protocols, Methods in Molecular Biology 622, DOI 10.1007/978-1-60327-299-5 9, © Springer Science+Business Media, LLC 2001, 2010

159

160

Pennington and Edwards

2. Materials 2.1. RNA Isolation from Tissue Samples and Cells

1. RNA-BeeTM (contains poison, phenol, and an irritant, guanidine thiocyanate, so handle with care). 2. Chloroform. 3. 95% (v/v) ethanol 4. Promega SV total RNA isolation (Promega, Southampton, UK). 5. Nanodrop (Nanodrop Technologies, Wilmington, DE, USA) or similar instrument for accurate RNA quantification.

2.2. Reverse Transcription

1. Random hexamers (pdN6 ) resuspended in 1325 ␮L water to 1 ␮g/␮L (GE Healthcare, Little Chalfont, UK, product no. 27-2166-01). 2. PCR nucleotide mix, 10 mM each dNTP (Roche Applied Science, Burgess Hill, UK). 3. SuperscriptTM ll 200 U/␮L supplied with 5X buffer and 0.1 M dithiothreitol (DTT) (Invitrogen, Paisley, UK). R 4. RNasin RNAse inhibitor 40 U/␮L (Promega).

2.3. Real-Time PCR

R 1. Taqman 2X Universal Mastermix (Applied Biosystems, Warrington, UK. Product no. 4326614).

2. Gene-specific PCR oligonucleotide primers and a flurogenic probe labelled with FAM (6-carboxyfluorescein, a fluorescent reporter) at the 5 end and with TAMARA (6carboxytetramethylrhodamine, a fluorescent quencher) at the 3 end. 3. Optical 96-well fast thermal cycling plate (Applied Biosystems, product no. 4346906). 4. ABITM prism optical adhesive covers (Applied Biosystems, product no. 4311971).

3. Methods 3.1. RNA Isolation from Tissue Samples

To preserve RNA integrity, tissue samples should be snap frozen in liquid nitrogen and then stored at –80◦ C or stored in R RNAlater (Ambion, Warrington, UK) immediately upon collection. When ready to extract RNA, add 1 mL RNA-BeeTM per 50 mg tissue and homogenize using a hand-held homogenizer or a tissue lyser with steel beads (Qiagen, Crawley, UK). Tissues

Real-Time PCR Expression Profiling of MMPs and TIMPs

161

should be homogenized quickly and thoroughly to ensure rapid penetrance of RNA-Bee into the tissues. At this point samples can be stored at –80◦ C or extraction can continue immediately by adding 0.1 mL of chloroform per 1 mL of homogenate. Mix samples thoroughly by inverting the tubes and then incubate on ice for 5 min. Centrifuge samples at 12,000g for 15 min at 4◦ C. Transfer the upper aqueous phase to a fresh tube and to this add 200 ␮L of 95% ethanol. Transfer this mix to a spin column assembly from the Promega SV total RNA isolation kit and then centrifuge at 12,000g for 1 min. Continue with the protocol detailed in the Promega catalogue from step 3 in the RNA purification by centrifugation section. 3.2. RNA Isolation from Cells

To isolate RNA from cells grown in suspension, sediment cells, resuspend in PBS, and sediment again to remove all traces of PBS. Lyse the cells by addition of 0.2 mL of RNA-BeeTM per 106 cells and pipette well to mix. For adherent cells, first wash the cells with PBS then remove as much PBS as possible. Add 1 mL of RNA-BeeTM per 10 cm2 monolayer of cells. Solubilize by pipetting several times and transfer to a fresh tube using a cell scraper to help collect all of the solute. Continue with the RNA isolation protocol detailed as above for tissue samples.

3.3. Reverse Transcription

Once RNA has been accurately quantified, add 1 ␮g of RNA to 2 ␮g of random hexamers and bring the total volume to 11 ␮L with PCR-grade water. Incubate the samples at 70◦ C for 10 min then keep on ice. Add 4 ␮L, 5X buffer, 2 ␮L DTT, 1 ␮L dNTP R , and 1 ␮L Superscript ll to bring the total mix, 1 ␮L RNasin reaction volume to 20 ␮L. Incubate this reaction mix at 42◦ C for 1 h. The cDNA from this reaction is relatively stable and can be stored at 4◦ C for short-term storage (weeks) or 20◦ C for a longer term. If the concentration of starting RNA is low it is possible to reverse transcribe as little as 0.1 ␮g of RNA and adjust the quantities in the following protocol to ensure the correct amount of cDNA is used in subsequent reactions.

3.4. Primer and Probe Design

As with standard end-point PCR, primer design is vital to the success of qRT-PCR. The default settings in Primer Express 3.0 (Applied Biosystems) ensure the strict criteria for qRT-PCR R probes amplicon design are selected, for example, Taqman should have • no more than 2 Gs or Cs in the last five bases; • a G/C content of 30–80%; • an amplicon size range of 50–150 bp; • a maximum amplicon melting temperature of 85◦ C; • a primer length of 9–40 bp;

162

Pennington and Edwards

• a primer melting temperature of 58–60◦ C with a difference of 2◦ C between primers; • a probe melting temperature that is 10◦ C higher than the primers; • more Cs than Gs in the probes sequence; • no G on the 5 end of the probe; • no self-binding complementarity. It is important that at least one primer, but preferably the probe, crosses an exon junction. The primers thus created would amplify mRNA (or cDNA made from it), but not genomic DNA. This is important since DNase treatment of RNA is rarely 100% efficient. Primer and probe sets should be checked for sequence specificity by BLAST analysis and ideally by sequencing of the PCR products. Mouse and human primer and probe sequences are given in Tables 9.1 and 9.2.

Table 9.1 Primer and probe sequences for human MMP and related genes TaqMan primers and probes (spp. Homo) Forward primer: Reverse primer: Probe: Forward primer: Reverse primer: Probe:

5 -GACGGCCTTCTGCAATTCC-3 5 -GTATAAGGTGGTCTGGTTGACTTCTG-3 5 -FAM-ACCTCGTCATCAGGGCCAAGTTCGT-TAMRA-3 5 -GAGCCTGAACCACAGGTACCA-3 5 -AGGAGATGTAGCACGGGATCA-3 5 -FAM-CTGCGAGTGCAAGATCACGCGC-TAMRA-3

Forward primer: Reverse primer: Probe: Forward primer: Reverse primer: Probe:

5 -CCAGGACGCCTTCTGCAA-3 5 -CCCCTCCTTTACCAGCTTCTTC-3 5 -FAM-CGACATCGTGATCCGGGCCA-TAMRA-3 5 -CACCCTCAGCAGCACATCTG-3 5 -GGCCGGAACTACCTTCTCACT-3 5 -FAM-CACTCGGCACTTGTGATTCGGGC-TAMRA-3

MMP1

Forward primer: Reverse primer: Probe:

MMP2

Forward primer: Reverse primer: Probe:

5 -AAGATGAAAGGTGGACCAACAATT-3 5 -CCAAGAGAATGGCCGAGTTC-3 5 -FAM-CAGAGAGTACAACTTACATCGTGTTGCGGCTCTAMRA-3  5 -AACTACGATGACGACCGCAAGT-3 5 -AGGTGTAAATGGGTGCCATCA-3 5 -FAM-CTTCTGCCCTGACCAAGGGTACAGCC-TAMRA-3

MMP3

Forward primer: Reverse primer: Probe: Forward primer: Reverse primer: Probe:

5 -TTCCGCCTGTCTCAAGATGATAT-3 5 -AAAGGACAAAGCAGGATCACAGTT-3 5 -FAM-TCAGTCCCTCTATGGACCTCCCCCTGAC-TAMRA-3 5 -CTTTGCGCGAGGAGCTCA-3 5 -CAGGCGCAAAGGCATGA-3 5 -FAM-CCATTTGATGGGCCAGGAAACACG-TAMRA-3

TIMP1

TIMP2

TIMP3

TIMP4

MMP7

(continued)

Real-Time PCR Expression Profiling of MMPs and TIMPs

163

Table 9.1 (continued) TaqMan primers and probes (spp. Homo) Forward primer: Reverse primer: Probe: Forward primer: Reverse primer: Probe:

5 -CACTCCCTCAAGATGACATCGA-3 5 -ACGGAGTGTGGTGATAGCATCA-3 5 -FAM-CAAGCAACCCTATCCAACCTACTGGACCAA-TAMRA-3 5 -AGGCGCTCATGTACCCTATGTAC-3 5 -GCCGTGGCTCAGGTTCA-3 5 -FAM-CATCCGGCACCTCTATGGTCCTCG-TAMRA-3

Forward primer: Reverse primer: Probe: Forward primer: Reverse primer: Probe:

5 -GGACCTGGGCTTTATGGAGATAT-3 5 -CCCAGGGAGTGGCCAAGT-3 5 -FAM-CATCAGGCACCAATTTATTCCTCGTTGCT-TAMRA-3 5 -GGGTGCCCTCTGAGTCGA-3 5 -TCACAGGGTCAAACTTCCAGTAGA-3 5 -FAM-ATGCTGATGGCTATGCCTACTTCCTGCG-TAMRA-3

Forward primer: Reverse primer: Probe: Forward primer: Reverse primer: Probe:

5 -CGCCTCTCTGCTGATGACATAC-3 5 -GGTAGTGACAGCATCAAAACTCAAA-3 5 -FAM-TCCCTGTATGGAGACCCAAAAGAGAACCA-TAMRA-3 5 -AAATTATGGAGGAGATGCCCATT-3 5 -TCCTTGGAGTGGTCAAGACCTAA-3 5 -FAM-CTACAACTTGTTTCTTGTTGCTGCGCATGA-TAMRA-3

MMP14

Forward primer: Reverse primer: Probe:

MMP15

Forward primer: Reverse primer: Probe:

5 -AAGGCCAATGTTCGAAGGAA-3 5 -GGCCTCGTATGTGGCATACTC-3 5 -FAM-CAACATAATGAAATCACTTTCTGCATCCAGAATTACATAMRA-3  5 -GCGCTTCAACGAGGAGACA-3 5 -TCCAGTATTTGGTGCCCTTGT-3 5 -FAM-CCTTCCTGAGCAATGACGCAGCCTAC-TAMRA-3

MMP16

Forward primer: Reverse primer: Probe:

5 -ATGATTTACAGGGCATCCAGAAA-3 5 -TGGAGGCCGAGGAGGTTT-3 5 -FAM-CAAGATTCCTCCACCTACAAGACCTCTACCGACTAMRA-3

MMP17

Forward primer: Reverse primer: Probe:

5 -FAM-GCGGGTATCCTTCCTCTACGT-3 5 -CAGCGACCACAAGATCGTCTT-3 5 -FAM-ATTGTCCTTGAACACCCAGTACCTGTCTCCTTTAATAMRA-3

MMP19

Forward primer: Reverse primer: Probe: Forward primer: Reverse primer: Probe:

5 -GGCTACCGGCCCCACTT-3 5 -TGTCTCTTCTTCTTCCTCATCCCTTA-3 5 -FAM-ATCCAGGCTCTCTATGGCAAGAAGAGTCCA-TAMRA-3 5 -CCTTTGACGCTGTGACAATGC-3 5 -TCCTGTCCGCAAGTGAACCT-3 5 -FAM-TCCTGCTCTTCAAGGACCGGATTTTCTG-TAMRA-3

Forward primer: Reverse primer: Probe: Forward primer: Reverse primer: Probe:

5 -CCAGTGACACGGGCATCA-3 5 -ATTATGGATCCCGTCCTGTAGGT-3 5 -FAM-CCTTCTCAAGGTGGCCGTCCATGA-TAMRA-3 5 -TCCACAAGAAAGGGAAAGTGTACTG-3 5 -ACGGCGTTGGCGATGAT-3 5 -FAM-TTCTCCTACCCCGGCTACCTGGCC-3

MMP8

MMP9

MMP10

MMP11

MMP12

MMP13

MMP20

MMP21

MMP23

(continued)

164

Pennington and Edwards

Table 9.1 (continued) TaqMan primers and probes (spp. Homo) Forward primer: Reverse primer: Probe: Forward primer: Reverse primer: Probe:

5 -CCAGTACATGGAGACGCACAA-3 5 -TGCGGACGGGGAGTGT-3 5 -FAM-CAGGGCATCCAGAAGATCTATGGACCC-TAMRA-3 5 -GACGATGAGGAGACCTGGACTTT-3 5 -CCTGGTAGAAGGGCCTCATAATG-3 5 -FAM-CCGACCTGTTTGCCGTGGCTGTC-3

MMP26

Forward primer: Reverse primer: Probe:

5 -AATTCTGGAAATCCTGGAGTTGTC-3 5 -CAAAGAATGCCCAATCTCATGA-3 5 -FAM-CTGGTCAGCTTCAGACACTGGATATAATCTGTTCCTAMRA-3

MMP27

Forward primer: Reverse primer: Probe:

5 -GTTTAGAAGTGTGGAGCAAAGTCACT-3 5 -ATAGCGAGGACACCGACCAT-3 5 -FAM-CATCATGATTGCCTTTAGGACTCGAGTCCATAMRA-3

MMP28

Forward primer: Reverse primer: Probe:

EMMPRIN

Forward primer: Reverse primer: Probe:

5 -TTTGAGACCTGGGACTCCTACAG-3 5 -CCCAGAAATGGCTCCCTTTA-3 5 -FAM-ACTCTTCCTTCGATGCCATCACTGTAGACAGTAMRA-3  5 -TGCTGGTCACCATCATCTTCAT-3 5 -CCGGCGTCGTCATCATC-3 5 -FAM-ACGAGAAGCGCCGGAAGCCC-TAMRA-3

RECK

Forward primer: Reverse primer: Probe:

5 -GCCGTCGGAGTCCTCTCA-3 5 -AGAGCGAAGGACACTTGACAGAA-3 5 -FAM-AGCACAGCTCCGTCGCCGAGTG-TAMRA-3

ADAM8

Forward primer: Reverse primer: Probe:

5 -AAGCAGCCGTGCGTCATC-3 5 -AACCTGTCCTGACTATTCCAAATCTC-3 5 -FAM-AATCACGTGGACAAGCTATATCAGAAACTCAACTTCC-TAMRA-3

ADAM9

Forward primer: Reverse primer: Probe:

5 -GGAAACTGCCTTCTTAATATTCCAAA-3 5 -CCCAGCGTCCACCAACTTAT-3 5 -FAM-CCTGATGAAGCCTATAGTGCTCCCTCCTGTTAMRA-3

ADAM10

Forward primer: Reverse primer: Probe:

5 -AGCGGCCCCGAGAGAGT-3 5 -AGGAAGAACCAAGGCAAAAGC-3 5 -FAM-ATCAAATGGGACACATGAGACGCTAACTGCTAMRA-3

ADAM12

Forward primer: Reverse primer: Probe:

ADAM15

Forward primer: Reverse primer: Probe:

5 -AGCTATGTCTTAGAACCAATGAAAAGTG-3 5 -CCCCGGACGCTTTTCAG-3 5 -FAM-ACCAACAGATACAAACTCTTCCCAGCGAAGATAMRA-3  5 -CCAGCTGTCACCCTCGAAA-3 5 -GGCAATCGAGGCAGCAAA-3 5 -FAM-TTCCTCCACTGGCGCAGGGC-TAMRA-3

ADAM17

Forward primer: Reverse primer: Probe:

5 -GAAGTGCCAGGAGGCGATTA-3 5 -CGGGCACTCACTGCTATTACC-3 5 -FAM-TGCTACTTGCAAAGGCGTGTCCTACTGC-TAMRA-3

MMP24

MMP25

(continued)

Real-Time PCR Expression Profiling of MMPs and TIMPs

165

Table 9.1 (continued) TaqMan primers and probes (spp. Homo) ADAM19

ADAM28

ADAM33

ADAMTS1

Forward primer: Reverse primer: Probe: Forward primer: Reverse primer: Probe:

5 -TGACAGCAAGGGCCAACAC-3 5 -AGGTCGCTTCTTGGTCTGTTGT-3 5 -FAM-TCGAGCACTCCAAGCCCACCACC-TAMRA-3 5 -GGGCCCACGATTTGCA-3 5 -TGAACCTTCCTGTCTTTCAATTTTACT-3 5 -FAM-AGAACATTGCCCTACCTGCCACCAAAC-TAMRA-3

Forward primer: Reverse primer: Probe: Forward primer Reverse primer Probe

5 -CAATAGCAACCATAACTGCCACTGT-3 5 -CCACCAAAGCCTGGCTTGT-3 5 -FAM-CTCCAGGCTGGGCTCCACCCTTC-TAMRA-3 5 -GGACAGGTGCAAGCTCATCTG-3 5 -TCTACAACCTTGGGCTGCAAA-3 5 -FAM-CAAGCCAAAGGCATTGGCTACTTCTTCG-TAMRA-3

ADAMTS2

Forward primer 5 -CTGGCAAGCATTGTTTTAAAGGA-3 Reverse primer 5 -GGAGCCAAACGGACTCCAA-3 5 -FAM-ATCTGGCTGACACCTGACATCCTCAAACG-TAMRA-3 Probe N.B. will not detect shortest splice variant

ADAMTS3

Forward primer 5 -GCAGCATTCCATCGTTACCA-3 Reverse primer 5 -CCATAGAATAATTGATTCCAGGAAGTT-3 5 -FAM-CCATTCCTATGACTGTCTCCTTGATGACCCProbe TAMRA-3

ADAMTS4

Forward primer Reverse primer Probe Forward primer Reverse primer Probe

ADAMTS5

ADAMTS6

ADAMTS7

ADAMTS8

ADAMTS9

5 -CAAGGTCCCATGTGCAACGT-3 5 -CATCTGCCACCACCAGTGTCT-3 5 -FAM-CCGAAGAGCCAAGCGCTTTGCTTC-TAMRA-3 5 -TGTCCTGCCAGCGGATGT-3 5 -ACGGAATTACTGTACGGCCTACA-3 5 -FAM-TTCTCCAAAGGTGACCGATGGCACTG-TAMRA-3

Forward primer 5 -GGCTGAATGACACATCCACTGTT-3 Reverse primer 5 -CAAACCGTTCAATGCTCACTGA-3 5 -FAM-CACTACCAATTAACAACACACATATCCACCACAGAProbe CAG-TAMRA-3  Forward primer 5 -CAGCCTACGCCCAAATACAAA-3 Reverse primer 5 -CCCTTGTAGAGCATAGCGTCAAA-3 5 -FAM-AAGCGCTTCCGCCTCTGCAACC-TAMRA-3 Probe Forward primer 5 -CCGCCACCCAGAGCACTA-3 Reverse primer 5 -TCGATCACGGAGCAGCTTTT-3 5 -FAM-CCATCCTGCTCACCAGACAGAACTTCTGTGProbe TAMRA-3  Forward primer 5 -TTAGTGAAGATAGTGGATTGAGTACAGCTT-3 Reverse primer 5 -TGTTGGAGCCATGACATGCT-3 5 -FAM-ATCGCCCATGAGCTGGGCCA-TAMRA-3 Probe

ADAMTS10 Forward primer Reverse primer Probe ADAMTS12 Forward primer Reverse primer Probe

5 -AGAGAACGGTGTGGCTAACCA-3 5 -TCTCTCGCGCTCACACATTC-3 5 -FAM-CAGTGCTCATCACACGCTATGACATCTGC-TAMRA-3 5 -CACGACGTGGCTGTCCTTCT-3 5 -CCGAATCTTCATTGATGTTACAACTG-3 5 -FAM-AGGACATCTGTGCTGGTTTCAATCGCC-TAMRA-3 (continued)

166

Pennington and Edwards

Table 9.1 (continued) TaqMan primers and probes (spp. Homo) ADAMTS13

Forward primer Reverse primer Probe

ADAMTS14

Forward primer Reverse primer Probe

5 -CAGAGCGAGAGAATATGTCACATTTC-3 5 -ACCGCCAAGTGTGTGAAGAGA-3 5 -FAM-CCAACCTGACCAGTGTCTACATTGCCAACTAMRA-3  5 -CGCTGGATGGGACTGAGTGT-3 5 -CGCGAACATGACCCAAACTT-3 5 -FAM-CCCGGCAAGTGGTGCTTCAAAGGT-TAMRA-3

ADAMTS15

Forward primer Reverse primer Probe Forward primer Reverse primer Probe

5 -ATGTGCTGGCACCCAAGGT-3 5 -CAGCCAGCCTTGATGCACTT-3 5 -FAM-CCTGACTCCACCTCCGTCTGTGTCCA-TAMRA-3 5 -GCCCATGAGTCTGGACACAA-3 5 -GCAGGGTGACCAGGAGAAGA-3 5 -FAM-TGCAAAAAGTCCGAGGGCAACATCAT-TAMRA-3

Forward primer Reverse primer Probe Forward primer Reverse primer Probe

5 -GGTCTCAATTTGGCCTTTACCAT-3 5 -GACCTGCCAGCGCAAGAT-3 5 -FAM-CCACAACTTGGGCATGAACCACGA-TAMRA-3 5 -CTCATTGGAAAGAATGGCAAGAG-3 5 -GGTACAACTTCGGTACTTAGAGCACAT-3 5 -FAM-TGTGACACTCTAGGGTTTGCCCCCAC-TAMRA-3

Forward primer Reverse primer Probe Forward primer Reverse primer Probe

5 -GGTGTAAGGCTGGAGAATGTACCA-3 5 -TGCGCTCTCGACTGCTGAT-3 5 -FAM-CCTCAGCACCTGAACATCTGGCCG-TAMRA-3 5 -ACTGTCCCGAGTGACGAGAGA-3 5 -AACAAGGCACTCGCTCCATT-3 5 -FAM-AATTTTCCTGTCCCAGTTGGGCTGCTA-TAMRA-3

ADAMTS16

ADAMTS17

ADAMTS18

ADAMTS19

ADAMTS20

Table 9.2 Primer and probe sequences for murine MMP and related genes TaqMan primers and probes (spp. Mus) Forward primer: Reverse primer: Probe: Forward primer: Reverse primer: Probe:

5 -CATGGAAAGCCTCTGTGGATATG-3 5 -AAGCTGCAGGCACTGATGTG-3 5 -FAM-CTCATCACGGGCCGCCTAAGGAAC-TAMRA-3 5 -CCAGAAGAAGAGCCTGAACCA-3 5 -GTCCATCCAGAGGCACTCATC-3 5 -FAM-ACTCGCTGTCCCATGATCCCTTGC-TAMRA-3

Timp3

Forward primer: Reverse primer: Probe:

Timp4

Forward primer: Reverse primer: Probe:

5 -GGCCTCAATTACCGCTACCA-3 5 -CTGATAGCCAGGGTACCCAAAA-3 5 -FAM-TGCTACTACTTGCCTTGTTTTGTGACCTCCATAMRA-3  5 -TGCAGAGGGAGAGCCTGAA-3 5 -GGTACATGGCACTGCATAGCA-3 5 -FAM-CCACCAGAACTGTGGCTGCCAAATC-TAMRA-3

Timp1

Timp2

(continued)

Real-Time PCR Expression Profiling of MMPs and TIMPs

167

Table 9.2 (continued) TaqMan primers and probes (spp. Mus) Mmp1a

Forward primer: Reverse primer: Probe:

5 -CGTGGACCAACAGCAGTGAA-3 5 -GAGTGAGCCCAAGGGAGTGA-3 5 -FAM-TCAACTTGTTCTATGTTACGGCTCATGAACTGGTAMRA-3

Mmp1b

Forward primer: Reverse primer: Probe:

5 -TGGACCGACAACAATGAGGAT-3 5 -TGGGAGAGTCCAAGGGAGTG-3 5 -FAM-TCAACTTGTTCTATGTTACGGCTCATGAACTGGTAMRA-3

Mmp2

Forward primer: Reverse primer: Probe:

5 -AACTACGATGATGACCGGAAGTG-3 5 -TGGCATGGCCGAACTCA-3 5 -FAM-TCTGTCCTGACCAAGGATATAGCCTATTCCTCGTAMRA-3

Mmp3

Forward primer: Reverse primer: Probe:

5 -GGAAATCAGTTCTGGGCTATACGA-3 5 -TAGAAATGGCAGCATCGATCTTC-3 5 -FAM-AGGTTATCCTAAAAGCATTCACACCCTGGGTCTTAMRA-3

Mmp7

Forward primer: Reverse primer: Probe:

5 -GCAGAATACTCACTAATGCCAAACA-3 5 -CCGAGGTAAGTCTGAAGTATAGGATACA-3 5 -FAM-CCAAAATGGCATTCCAGAATTGTCACCTAC-3

Mmp8

Forward primer: Reverse primer: Probe:

5 -GATTCAGAAGAAACGTGGACTCAA-3 5 -CATCAAGGCACCAGGATCAGT-3 5 -FAM-CATGAATTTGGACATTCTTTGGGACTCTCTCACTAMRA-3

Mmp9

Forward primer: Reverse primer: Probe:

5 -CGAACTTCGACACTGACAAGAAGT-3 5 -GCACGCTGGAATGATCTAAGC-3 5 -FAM-TCTGTCCAGACCAAGGGTACAGCCTGTTCTAMRA-3

Mmp10

Forward primer: Reverse primer: Probe:

5 -CCTGCTTTGTCCTTTGATTCAGT-3 5 -CGGGAT TCCAATGGGATCT-3 5 -FAM-TCCTATTCTTTAAAGACAGGTACTTCTGGCGCATAMRA-3

Mmp11

Forward primer: Reverse primer: Probe:

5 -ATTGATGCTGCCTTCCAGGAT-3 5 -GGGCGAGGAAAGCCTTCTAG-3 5 -FAM-TCCTTCGTGGCCATCTCTACTGGAAGTTTGTAMRA-3

Mmp12

Forward primer: Reverse primer: Probe:

5 -GAAACCCCCATCCTTGACAA-3 5 -TTCCACCAGAAGAACCAGTCTTTAA-3 5 -FAM-AGTCCACCATCAACTTTCTGTCACCAAAGCTAMRA-3

Mmp13

Forward primer: Reverse primer: Probe:

Mmp14

Forward primer: Reverse primer: Probe:

5 -GGGCTCTGAATGGTTATGACATTC-3 5 -AGCGCTCAGTCTCTTCACCTCTT-3 5 -FAM-AAGGTTATCCCAGAAAAATATCTGACCTGGGATTCTAMRA-3  5 -AGGAGACAGAGGTGATCATCATTG-3 5 -GTCCCATGGCGTCTGAAGA-3 5 -FAM-CCTGCCGGTACTACTGCTGCTCCTG-TAMRA-3 (continued)

168

Pennington and Edwards

Table 9.2 (continued) TaqMan primers and probes (spp. Mus) Mmp15

Forward primer: Reverse primer: Probe:

5 -ATCCCCTATGACCGCATTGAC-3 5 -CCCCTGCCAGACACTGATG-3 5 -FAM-ACACAGCATGGAGACCCTGGCTACCC-TAMRA-3

Mmp16

Forward primer: Reverse primer: Probe:

5 -GGCTACCTTCCACCGACTGA-3 5 -CTTCATCCAGTCGATTGTGTTTCT-3 5 -FAM-CTGCAGAGACCATGCAGTCAGCTCTAGCTTAMRA-3

Mmp17

Forward primer: Reverse primer: Probe: Forward primer: Reverse primer: Probe:

5 -GGCAGTATGTTCCTGCACTTCA-3 5 -GCTAGCACTGCCCTCAGGAT-3 5 -FAM-CCTGTGGACCTCAGTCTCTGCCAAGG-TAMRA-3 5 -GCCCATTTCCGGTCAGATG-3 5 -AGGGATCCTCCAGACCACAAC-3 5 -FAM-CCACAAGGGCCCGTATGAAGCAGC-TAMRA-3

Mmp20

Forward primer: Reverse primer: Probe:

5 -GATCAGGAGGATTAAGGAGCTACAAA-3 5 -GGCGGTAGTTAGCCACATCAG-3 5 -FAM-CCAGAATACAATGAATGTGATCAAGAAGCCTCGTAMRA-3

Mmp21

Forward primer: Reverse primer: Probe:

5 -TCCAAAGAAGATGAGCCAAGTG-3 5 -ACGCTGAATCGAGGTTTCTG-3 5 -FAM-TTCCAGCAATAATGCCTCAAAACCACCCTAMRA-3

Mmp23

Forward primer: Reverse primer: Probe: Forward primer: Reverse primer: Probe:

5 -CAGACTGTTGACCATGTCGGTAA-3 5 -GAAGGAAAGAACTCTGTATGTGAGGTT-3 5 -FAM-CCGCTACACGCTGACACCGGC-TAMRA-3 5 -TATCATGGCTCCCTTCTACCAATAC-3 5 -CTGCGGACCGGGAGTGT-3 5 -FAM-CCAGCTGAGCCCTCTGGAGCCA-TAMRA-3

Mmp25

Forward primer: Reverse primer: Probe:

Mmp27

Forward primer: Reverse primer: Probe:

5 -TGGCTGTCTGGGCTACTGAA-3 5 -GGTAGGCCCGAGCAAAGTG-3 5 -FAM-AATTCTCAGTACCAGGAGCCTGACATCATTATCCTAMRA-3  5 -AGGATAATAAAGTGCTTCCCAGGA-3 5 -AAGAAATAGAGGAATCCATTATGTTGG-3 5 -FAM-TCGCCTCCGTGTGGATGCTGTC-TAMRA-3

Mmp28

Forward primer: Reverse primer: Probe:

5 -CCACTTGGACAGAGAGGATCAGT-3 5 -AAGCGTTTCTTACGCCTCATTT-3 5 -FAM-CTGCTTGCTGGACACCGAGCCAA-TAMRA-3

Mmp19

Mmp24

Designing primers and probes for a gene of interest is relatively quick and offers more control of the regions targeted so that specific splice variants can be targeted. However, this stage of preparation can be reduced by purchasing ready-made, fully validated primer/probe sets from companies specializing in the manufacture of these, e.g., Qiagen and Applied Biosystems both have a genome-wide stock of primers and probes. It should be noted,

Real-Time PCR Expression Profiling of MMPs and TIMPs

169

however, that “off the shelf” ready-made primer/probe sets are more expensive than custom-designed primers/probes. An alternative approach is to purchase a Universal Probe Library from Roche Applied Science; this library of short, 8–9 nucleotide (nt) probes offers transcriptome-wide coverage for specific organisms. Locked nucleic acid (LNA) chemistry means these probes are highly specific for 8–9 nt complementary strands. Added specificity is achieved by designing flanking 5 and 3 primers. A full set of 165 universal probe library probes will effectively provide a probe for every gene in a number of specified organisms (including human, mouse, rat). Although initially expensive, the Universal Probe Library allows a researcher to design and buy only relatively cheap primers to match an existing “in-house” probe. This makes preliminary studies into potential target genes a much more financially feasible option. 3.5. Quantitative Real-Time PCR

To determine the relative quantity of your gene of interest, use one sample to prepare a standard curve in the first six wells of a 96-well plate. Make twofold dilutions in the range of 20– 0.625 ng (for 18S rRNA reactions use a range of 1–0.03125 ng) presuming that 1 ng of input RNA in the RT reaction yields 1 ng cDNA. For absolute quantification, see notes in Section 4.3. Volumes should not exceed 10 ␮L but avoid pipetting less than 2 ␮L as precise pipetting is vital to the success of qPCR. Load 5 ng of cDNA in subsequent wells (1 ng for 18S rRNA). Samples should be loaded in triplicate and care should be taken to ensure all the sample is dispensed to the bottom of each well. To R 2X Universal Mastermix, each reaction add 8.33 ␮L of Taqman 200 nM of each gene-specific primer, and 100 nM of fluorogenic probe, and adjust the total volume of the reaction to 25 ␮L with PCR-grade water. Cycling conditions for Applied Biosystems Real-Time PCR 7,500 and 7,900 instruments should be programmed to 2 min at 50◦ C, 10 min at 95◦ C, and then 40 cycles, each consisting of 15 s at 95◦ C, and 1 min at 60◦ C. On completion of the reaction the cycle threshold (C T ) values should be adjusted so that measurement is taken in the early stage of the exponential phase of the PCR. On Applied Biosystems machines this is achieved by manually adjusting the threshold bar to the early exponential phase of the amplification plot but above any “noise”; for light cyclers and other instruments follow the manufacturer’s instructions. qRTPCR data can be analysed using an absolute or relative standard curves method or a comparative C T method. The relative standard curve method of post-run analysis uses the standard curve chart values to determine gene expression levels in arbitrary units. Calculate the log of the input amount of each sample:

170

Pennington and Edwards

Log of the input amount = (C T value − b)/m (where b is the intercept of the standard curve and m is the slope of the standard curve). Next calculate input amount Input amount = 10∧ (log of the input amount) Lastly, divide the input amount of your gene of interest with the input amount of your endogenous control. The final amount will be in arbitrary units, giving relative differences between samples. This method cannot be used to compare expression between genes. See the notes in Section 4 regarding absolute quantification of gene expression using this method. The comparative C T method compares the C T values of the samples with a control such as a non-treated sample or normal tissue RNA. The C T values of both the control or calibrator sample and the samples of interest are normalized to an appropriate endogenous control gene. This method is also known as the 2−CT method, where first C T is calculated, i.e. the C T for a gene of interest – endogenous control C T . Then C T is calculated by taking the C T of a sample – the C T of a control or a calibrator sample. Finally 2−CT will give the fold changes between the gene of interest in a sample versus a control sample. This method is relatively easy and negates the need to include standard curves on every plate; however, before using the comparative C T method, reactions should be optimized so that the slope of the standard curves for each reaction varies by <0.1. If the plot of log of the cDNA dilution versus C T is close to zero, PCR efficiencies of the target genes and house-keeping genes are very similar and the comparative method can be used. If amplification efficiencies vary, and the slope of the standard curve is > 0.1, the standard curve method must be used. Full details of both methods and examples are found in User Bulletin #2 produced by Applied Biosystems. Whichever method of post-run analysis is used, it is important that C T values between triplicates should vary by <1 C T (or at least close to this) and standard curve slope should be ≤−3.2.

4. Notes 4.1. RNA Quality and Quantity

One of the prime benefits of using qRT-PCR to quantify gene expression is the sensitivity of the technique when using very small amounts of precious clinical specimens or cells, laser-captured tissue, or even formalin-fixed paraffin-embedded archived material. Each of these strategies imposes challenges on the extraction

Real-Time PCR Expression Profiling of MMPs and TIMPs

171

of high-quality, intact RNA that is free of genomic DNA, nucleases, and other PCR contaminants. It is easy to argue that the quality of RNA is perhaps the most important determinant of the reproducibility and biological relevance of qRT-PCR data, so it is important to recognize and remove samples that have degraded to such an extent that their use could result in a misinterpretation of results. The initial method of homogenization may require some optimization with different tissues. Whereas the tissue lyser (Qiagen) or a hand-held homogenizer may be adequate for the fast breakdown of most tissues, others may require manual grinding in a pestle and mortar in liquid nitrogen. Whichever method is chosen, the fast breakdown and penetrance of lysis buffers are important. Alternative methods of RNA extraction to that detailed above are available and may prove to be more suitable in some circumstances, e.g. when extracting from blood or when the expected yield is particularly low. Ambion and Qiagen both offer RNA extraction kits for particular applications. The quality and quantity of RNA is initially assessed by measuring UV absorbance in a spectrophotometer. Nucleic acids absorb light at a wavelength of 260 nm, whereas organic compounds like phenol/trizol/RNA-BeeTM , sugars, and alcohols absorb light at 230 nm and proteins at 280 nm. Absorbance ratios for 260/280 and 260/230 >1.8 are generally indicative of goodR NDquality, uncontaminated RNA. We use the NanoDrop 1000 (Nanodrop Technologies, Wilmington, DE, USA), which is highly accurate and requires the use of just 1 ␮L of RNA that can be used without dilution or the need for cuvettes. A 260/280 ratio of less than 1.8 can indicate protein contamination or possibly overdried RNA pellets that are not fully in solution. Freezing overnight at –80◦ C or briefly heating the RNA to 65◦ C can help rectify the latter problem. Contamination of RNA with residual extraction substances is more of a problem and can have wide ranging consequences for qRT-PCR results. A significant reduction in the sensitivity and kinetics of PCR assays is caused by inhibitory components; also different reactions may not be inhibited to the same degree and the effects can be compounded in absolute quantification where an external calibration curve is used to calculate the number of transcripts in the test samples. A further problem that can be encountered when handling RNA for qRT-PCR is RNA degradation. Degraded RNA can reduce the yield of reverse-transcribed cDNA and subsequent qPCR resulting in inaccurate representations of gene expression. Amplicons for real-time qRT-PCR are typically short (70–150 bp), so some degradation of the RNA can be tolerated; however, it is important to recognize and remove samples that

172

Pennington and Edwards

have degraded to avoid misinterpretation of results. In addition to the endogenous control chosen for a particular analysis we use 18S ribosomal RNA as a quality control reaction for RNA/cDNA and exclude any samples that are more than 1 C T different from the median 18S C T of the sample set. This is because the ribosomal RNA (rRNA) content of a cell is between 80 and 90% of the total RNA, thus a spectrophotometer reading is a good reflection of the quantity of rRNA present in a sample. Since 1 ␮g of the quantified total RNA was added to the reverse transcription reaction we would have expected that all samples would subsequently have had very similar 18S rRNA C T values. 4.2. Absolute Quantification

qPCR can be used for relative or absolute quantification of gene expression. While the relative quantification detailed above is adequate for most studies it is sometimes necessary to determine the absolute copy number of transcripts. This is achieved by subcloning the amplicon or full-length cDNA of the gene of interest behind a T7 or SP6 RNA polymerase promoter in a plasmid vector. We have used pBluescript KS(–) vector (Stratagene, Amsterdam, The Netherlands). Plasmids containing the correct inserts are linearized with an appropriate endonuclease, and sense RNA for each gene in vitro transcribed using the appropriate polymerase (T7 or T3, Roche Applied Science, East Sussex, UK). The plasmid DNA is then digested with RNase-free DNase I (Roche), the RNA precipitated and resuspended in RNase-free water, and then quantified accurately. Using the sequence identity and the length of the synthesized RNA, the molecular weight of each RNA can be calculated, and the number of molecules synthesized determined. One microgram of in vitro transcribed RNA is then reversed transcribed (as described above), and 10-fold serial dilutions of cDNA prepared covering concentrations ranging from the equivalent of 1,010 copies of RNA to the equivalent of 101 copies of RNA (2).

4.3. Normalizing and Endogenous Controls

Ideally an internal control used to normalize between samples should be constitutively expressed in all cell types at similar levels to the target gene and should remain constant, independent of disease status or experimental conditions. The consensus regarding subsequent normalization of qRTPCR data has been reached following years of publications advocating the use of one or another reference gene followed by contradictions and further propositions. The recent introduction of GeNormTM (7) provides an Excel-based program for determining the most stable reference genes from a panel of potential endogenous control genes. Vandesompele et al. (7) show that the common practice of using a single normalizing gene can lead to erroneous normalization. They suggest that an ideal, universal control gene probably does not exist and that normalizing

Real-Time PCR Expression Profiling of MMPs and TIMPs

173

to more than one endogenous control may be a more robust strategy. To determine the most appropriate normalizing genes, GeNormTM analyses a panel of 6 or 12 genes (ACTB, B2M, GAPD, HMBS, HPRT1, RPL13A, RPL32, RPS18, SDHA, TBP, UBC, and YWHAZ) to find the optimal genes to use as endogenous controls in each unique experimental system. PrimerDesign (Southampton, UK) provides a panel of primers for use with R SYBR primer or fluorescently labelled probe sets which can be used in conjunction with GeNormTM although other potential normalizers can be added to this panel if appropriate. GeNormTM has become the gold standard for determining the ideal number and identity of the most stable normalizing genes for qRT-PCR analyses and overcomes many of the uncertainties that existed prior to its introduction. References 1. Bustin, S. A., Benes, V., Nolan, T., and Pfaffl, M. W. (2005). Quantitative real-time RT-PCR–a perspective. J Mol Endocrinol 34, 597–601. 2. Morimoto, A. M., Tan, N., West, K., McArthur, G., Toner, G. C., Manning, W. C., Smolich, B. D., and Cherrington, J. M. (2004). Gene expression profiling of human colon xenograft tumors following treatment with SU11248, a multitargeted tyrosine kinase inhibitor. Oncogene 23, 1618–1626. 3. Nuttall, R. K., Pennington, C. J., Taplin, J., Wheal, A., Yong, V. W., Forsyth, P. A., and Edwards, D. R. (2003). Elevated membranetype matrix metalloproteinases in gliomas revealed by profiling proteases and inhibitors in human cancer cells. Mol Cancer Res 1, 333–345. 4. Overbergh, L., Giulietti, A., Valckx, D., Decallonne, R., Bouillon, R., and Mathieu,

C. (2003) The use of real-time reverse transcriptase PCR for the quantification of cytokine gene expression. J Biomol Tech 14, 33–43. 5. Porter, S., Scott, S. D., Sassoon, E. M., Williams, M. R., Jones, J. L., Girling, A. C., Ball, R. Y., and Edwards, D. R. (2004) Dysregulated expression of adamalysinthrombospondin genes in human breast carcinoma. Clin Cancer Res 10, 2429–2440. 6. Vandesompele, J., De Preter, K., Pattyn, F., Poppe, B., Van Roy, N., De Paepe, A., and Speleman, F. (2002) Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol 3 RESEARCH0034. 7. Wall, S. J. and Edwards, D. R. (2002) Quantitative reverse transcription-polymerase chain reaction (RT-PCR): a comparison of primer-dropping, competitive, and real-time RT-PCRs. Anal Biochem 300, 269–273.

Chapter 10 Analysis of the Degradome with the CLIP-CHIPTM Microarray Reinhild Kappelhoff, Ulrich auf dem Keller, and Christopher M. Overall Abstract The degradome microarray – CLIP-CHIPTM – is a dedicated and focused array that allows the analysis of all proteases, non-proteolytic homologs, and protease inhibitor gene transcripts in the human and murine genomes at the mRNA transcript level. Based on unique 70-mer oligonucleotides, designed to match parts of the sequence of known or predicted protease and inhibitor mRNAs in both species and printed on a glass-matrix surface, the CLIPCHIPTM microarray can be used to analyze differentially expressed protease and inhibitor gene products and give expression profiles for any analyzed sample. Key words: CLIP-CHIPTM microarray, linear aRNA amplification, gene expression profiling, protease, inhibitor, degradome, non-proteolytic homolog, differentially expressed genes.

1. Introduction Microarrays are designed to analyze mRNA transcripts in experimental samples. This can rank from 1,000 to 100,000 genes if using whole-genome arrays and providing expression profiles or differentially expressed genes for thousands of genes simultaneously. Since this amount of data can be distracting, the CLIPCHIPTM microarray was designed (1) as a specific customized array with its main focus on the analysis of the degradome. The degradome comprises proteases and non-proteolytic homologs and is divided into the five classes of aspartic, cysteine, metallo, serine, and threonine proteases (2) and their natural inhibitors I.M. Clark (ed.), D.A. Young, A.D. Rowan (Consulting editors), Matrix Metalloproteinase Protocols, Methods in Molecular Biology 622, DOI 10.1007/978-1-60327-299-5 10, © Springer Science+Business Media, LLC 2001, 2010

175

176

Kappelhoff, auf dem Keller, and Overall

Table 10.1 Human and murine proteases and inhibitors available on the CLIP-CHIPTM for the degradome analysis. The number of non-proteolytic and non-inhibitor homologs are in brackets Aspartic Cysteine Metallo

Serine

Threonine Inhibitor

Human 21 (0)

151 (16) 187 (40) 178 (26) 28 (12)

156 (17)

Murine 27 (4)

162 (14) 205 (43) 226 (28) 26 (12)

197 (17)

(see Table 10.1). Based on known and predicted protease and inhibitor sequences (3) unique 70-mer oligonucleotides representing all proteases, non-proteolytic homologs, and inhibitors gene transcripts from human (721) and mouse (843) were designed and printed as duplicate arrays onto aminosilane-coated glass slides. The CLIP-CHIPTM is printed in three versions: The human CLIP-CHIPTM contains two copies of the human oligonucleotide set; the mouse CLIP-CHIPTM contains two copies of the murine oligonucleotide set; and a hybrid version contains one set of each. Besides the fact that the CLIPCHIPTM is a focused microarray, another advantage is the flexibility of the oligonucleotide-based printing platform. If proteases and inhibitors are newly annotated or identified in the database, unique oligonucleotides can be designed and added to the CLIPCHIPTM at the next batch print.

2. Materials 2.1. RNA Extraction and Purification

1. Tissue homogenizer. 2. RNeasy Mini Kit (Qiagen, cat# 74104). 3. ␤-Mercaptoethanol. 4. RNase-free DNase I Set (Qiagen, cat# 79254). 5. QiaShredder (Qiagen, cat# 79654) or 20 G needle with syringe. 6. TRIzol (Invitrogen, cat# 15596-026) or TriReagent (Sigma, cat# T9424). 7. Nuclease-free or DEPC-treated water (add 1 mL DEPC (diethylpyrocarbonate); Sigma, #D5758) to 1 L dH2 O, shake vigorously to dissolve the “oily” DEPC, incubate for 12 h at 37ºC, and inactive DEPC by autoclaving for 15 min (see Note 1).

Analysis of the Degradome with the CLIP-CHIPTM Microarray

177

8. 100% Ethanol (ASC grade) and 70 % ethanol (in nucleasefree or DEPC-treated water). 9. RNAlater (Sigma, cat# R0901). 10. 50x TAE buffer: 242 g Tris–HCl, 100 mL 0.5 M EDTA, pH 8.0, 57.1 mL glacial acetic acid, adjust pH to 7.6– 7.8, and fill up to 1,000 mL with nuclease-free or DEPCtreated water. 11. 1.2% Agarose in 1x TAE buffer. 12. RNase Zap (Ambion) or RNase away (Molecular BioProducts); RNase-free tips and tubes. 13. UV–VIS spectrophotometer or NanoDrop ND-1000A. 14. Optional: 2100 Bioanalyzer (Agilent). 2.2. aRNA Amplification

1. MessageAmp II aRNA amplification kit II (Ambion, cat# AM1751). For all master mix calculations, the Message Amp Master Mix calculator ( www.ambion.com ) can be used (see Note 2). 2. Oligo (dT)-T7 primer. 3. Reverse Transcription Master Mix (4 ␮L per sample; assemble at room temperature): 1 ␮L 10x first-strand buffer, 2 ␮L dNTP mix (2.5 mM of each dNTP), 0.5 ␮L 10 U/␮L RNase inhibitor, and 0.5 ␮L 200 U/␮L ArrayScript reverse transcriptase. 4. Second-Strand Master Mix (40 ␮L per sample; assemble on ice): 31.5 ␮L nuclease-free H2 O, 5 ␮L 10x second-strand buffer, 2 ␮L dNTP mix (2.5 mM of each dNTP), 1 ␮L 5 U/␮L DNA polymerase, and 0.5 ␮L 10 U/␮L RNase H. 5. In Vitro Transcription Master Mix (12 ␮L per sample; assemble at room temperature): 2 ␮L of each 10x T7 reaction buffer and 75 mM ATP, 75 mM CTP, 75 mM GTP, 75 mM UTP, and T7 Polymerase enzyme mix. 6. RNase-free 5 M ammonium acetate.

2.3. Fluorescent Labeling of aRNA and Fragmentation

1. ULS-aRNA labeling kit with Cy3- and Cy5-ULS (Kreatech, cat# EA-006). 2. Fragmentation reagent (Ambion, cat# AM8740). 3. SpeedVac (Thermo Savant).

2.4. Pre-hybridization of the CLIP-CHIPTM Microarray

1. Slide holder vials (Starplex, cat# V302-SH). 2. 10% BSA in nuclease-free or DEPC-treated water and filtered through a 0.2 ␮m filter. 3. 10% SDS in nuclease-free or DEPC-treated water and filtered through a 0.2 ␮m filter.

178

Kappelhoff, auf dem Keller, and Overall

4. 20x SSC (Ambion, cat# AM9763), nuclease-free. 5. Pre-hybridization buffer: 5x SSC, 0.2% BSA, and 0.1% SDS (see Note 3). 6. Isopropanol. 2.5. Hybridization of Labeled aRNA to the CLIP-CHIPTM Microarray

1. KreaBlock (component in ULS labeling kit). 2. Formamide (Sigma, cat# F9037). 3. 2x Hybridization buffer: 100 ␮L 100% formamide, 100 ␮L 20x SSC, pre-heat to 42◦ C, then add 20 ␮L 10% SDS, and keep at 42◦ C (see Note 3). 4. LifterSlips 22×60 mm (Erie Scientific, cat#22x60I-2-4861, to cover both subarrays) or 22×25 mm (cat#22 × 25I-24635, to cover subarrays individually) with Teflon bars providing a 0.75-mm gap between microarray and LifterSlip (see Note 4). 5. 3x SSC in nuclease-free or DEPC-treated water. 6. Hybridization chamber (Corning, cat# 2551).

2.6. Washing of the CLIP-CHIPTM Microarray

1. Wash buffer 1: 1x SSC+0.2% SDS at 42◦ C (see Note 3). 2. Wash buffer 2: 0.1x SSC+0.2% SDS at 42◦ C (see Note 3). 3. Wash buffer 3: 0.1x SSC at 42◦ C. 4. Nuclease-free water. 5. Black storage boxes for microarray glass slides.

2.7. Microarray Scanning, Image, and Data Analysis

1. Microarray scanner (e.g., ScanArray, Perkin-Elmer; arrayWoRx, Applied Precision; GenePix, Axon Instruments; or 428 Array scanner, MWG). 2. Image analysis software: either freeware like Spotfinder (TIGR, TM4) (4) or ScanAlyze (Eisen lab) or commercial software like ImaGene (Biodiscovery) or GenePix (Molecular Devices). 3. Data analysis software: either freeware like CARMAWeb (5), TM4; Microarray Software Suite (4) and ArrayPipe (6) or commercial products like GeneSpring (Agilent).

3. Methods 3.1. RNA Extraction and Purification

For the CLIP-CHIPTM microarray analysis, the extraction and purification of DNA-free total RNA (see Note 5) is necessary since DNA can cross-hybridize with the microarray and it can also interfere with the correct concentration determination of the

Analysis of the Degradome with the CLIP-CHIPTM Microarray

179

RNA (see Notes 6 and 7). Therefore, total RNA has to be DNase I-treated. Total RNA from test and control cells or tissue can be isolated according to the following steps: 1a. Tissue culture cell lines: grow appropriate amount of cells in tissue culture dishes, wash 3x with PBS to remove dead cells, cell debris, and serum components. If using 6-well tissue culture dishes apply 700 ␮L RLT buffer (RNeasy Mini Kit, Qiagen) to the cells. This buffer contains guanidine thiocyanate and ␤-mercaptoethanol that allows a complete lysis of the cells, denaturation of proteins, and inhibition of RNases. If the RNA is not isolated immediately, cells can be stored in RLT buffer at –70◦ C. 1b. Whole tissues: for some tissues (especially spleen tissue, which has high RNase activity) it is favorable to store the tissue in a stabilizing reagent like RNAlater (Sigma) overnight before homogenizing in Trizol (Invitrogen) or TriReagent (Sigma). (Use a tissue homogenizer and follow the manufacturer’s protocol.) The RNA that is extracted from Trizol or TriReagent contains phenol and chloroform carryovers that results in a low A 260 /A 230 ratio. These carryovers will interfere with downstream reactions (e.g., reverse transcription and in vitro transcription), lowering their efficiency. Therefore the RNA should be purified again using the RNeasy Mini Kit. Precipitated RNA should therefore be dissolved in 700 ␮L RLT buffer (RNeasy Mini Kit) for a second purification round. 2. Homogenize the lysate further by pipetting it onto a QIAshredder spin column (Qiagen) and microcentrifuge for 2 min at 13,000g at room temperature or by passing the lysate five times through a 20 G needle using a syringe to shear the DNA. 3. Precipitate RNA by adding 700 ␮L of 70% ethanol to the lysate and mix gently. 4. Apply lysate two to three times to a RNeasy Mini column (Qiagen) and microcentrifuge 15 s at 8,000g, each time discarding the flow-through. Exceeding 8,000g can easily shear and damage the RNA bound to the column. 5. Wash RNA by adding 350 ␮L of RW1 buffer (from RNeasy Mini Kit) onto the RNeasy Mini column and microcentrifuge for 15 s at 8,000g. 6. Digest DNA by adding 10 ␮L RNase-free DNase I in 70 ␮L RDD buffer (RNase-free DNase I kit, Qiagen) onto the column. Mix DNase I gently with the buffer as it is extremely sensitive to physical denaturation. Incubate for 15 min at room temperature, not exceeding 30◦ C.

180

Kappelhoff, auf dem Keller, and Overall

7. Wash RNA by adding 350 ␮L of RW1 buffer (from RNeasy Mini Kit) onto the RNeasy Mini column and microcentrifuge for 15 s at 8,000g. 8. Transfer the column to a new collection tube and wash the column by adding 500 ␮L of RPE buffer (from RNeasy Mini kit) and microcentrifuge 15 s at 8,000g. Discard the flow-through and repeat the wash. Discard the flowthrough and microcentrifuge for an additional 2 min at 8,000g to dry the RNeasy silica gel membrane. 9. Place column in a new collection tube and elute the RNA by pipetting 50 ␮L of RNase-free water directly onto the silica gel membrane and microcentrifuge for 1 min at 8,000g. 10. Determine the RNA quantity and quality using UV– Vis spectrophotometer (Nanodrop ND-1000A preferable). The A260 /A280 ratio and the A260 /A230 ratio should be >1.8. Lower ratios show impurities and the RNA extraction should be repeated. 11. Determine RNA integrity and quality by electrophoresis using 1 ␮g of total RNA on a 1.2% native agarose gel and thereby examining the 28 S and 18 S ribosomal RNA bands (Fig. 10.1A). The bands should be intact and no significant smearing should be visible below the individual bands (see Note 8). 12. Store RNA samples at –70◦ C until required. 3.2. aRNA Amplification

The origin for the development of the RNA amplification method arose because in some cases only very small amounts of RNA were present and insufficient for hybridization to microarray slides. Many groups showed that the bias that is introduced during RNA amplification into the material is minimal (7–9). Nowadays RNA amplification has become a standard method in preparing RNA samples for microarray analysis as amplification brings the benefits of obtaining more reproducible expression profiles from a wide range of RNAs (10, 11). RNA amplification actually improves the reliability of array results regardless of whether it is needed for sample expansion (8, 9). The amplified RNA is also more stable. For the aRNA amplification, the MessageAmp II aRNA amplification kit is used following the Eberwine method (12). It is based on a reverse transcription with an oligo(dT) primer that contains the T7 promoter region for first-strand cDNA synthesis. The cDNA undergoes a second-strand dsDNA synthesis, which then becomes the template for the in vitro transcription using T7 RNA polymerase to generate multiple aRNA copies of each mRNA molecule that was in the sample. The linear amplified aRNA, also called antisense RNA or cRNA, is not modified

Analysis of the Degradome with the CLIP-CHIPTM Microarray

181

Fig. 10.1. Overview of the workflow for the CLIP-CHIPTM analysis. A 1.2% agarose gel of total RNA after extraction and purification. B 1.2% agarose gel of amplified aRNA. C Spectrogram of Cy3- and Cy5-labeled aRNA samples; as an example Universal aRNA-Cy3 and Brain aRNA-Cy5 are shown. D The CLIP-CHIPTM microarray: oligonucleotides can be seen as dark salt spots, the two subarrays are covered with LifterSlips and the glass slide is barcoded. E Microarray image analysis with ImaGene. Circles are overlain over the signal spots and are adjusted, bad spots are flagged, signal intensity for both channels is measured and saved as raw data in separate files. F Possible diagrams of differentially expressed genes and a gene cluster analysis of an experiment after data analysis using CARMAweb (5) for normalization and statistical analysis and presentation with MeV (MultiExperimentViewer, TM4.org) (4).

during the in vitro transcription. This means no modified nucleotides (e.g., aminoallyl-UTP, biotinylated-UTP, or cyanine 3/cyanine 5-CTP and UTP) are introduced into the RNA during the synthesis. These bulky modifications sterically interfere with

182

Kappelhoff, auf dem Keller, and Overall

the T7 polymerase during the in vitro transcription resulting in shorter aRNA fragments in comparison to longer RNA fragments from amplifications using unmodified nucleotides. 1. Synthesize first-strand cDNA by mixing 1 ␮g total RNA (see Note 9) in 5.5 ␮L of nuclease-free water with 0.5 ␮L oligo(dT)24-primer containing the region for the T7 polymerase promoter. 2. Let the primer anneal to the RNA by incubating the sample for 10 min at 70◦ C; then snap cool on ice. 3. Add 4 ␮L Reverse Transcription Master Mix and mix gently by pipetting up and down and flicking the tube, and collect the sample by a quick spin in a centrifuge. Do not vortex as this may shear RNA. 4. Incubate reaction in a hybridization oven (or thermal cycler) to prevent condensation for 2 h at 42◦ C; then place on ice. 5. Synthesize second-strand dsDNA by adding 40 ␮L secondstrand Master Mix and mix gently by pipetting up and down and flicking the tube and collect the sample by a quick spin in a centrifuge. 6. Place the tube into a pre-chilled 16◦ C heat block or thermal cycler and incubate for 2 h; then place on ice. Collect the reaction by a quick spin. 7. Purify dsDNA using the DNAclear kit (or MessageAmp II kit). Pre-heat nuclease-free water to 55◦ C and equilibrate DNAclear filter cartridges by adding 30 ␮L cDNA binding buffer onto the filter and incubate for 5 min. 8. Bring sample volume to 100 ␮L and add 250 ␮L cDNA binding buffer and mix thoroughly by pipetting and gently flicking the tube. Collect the reaction by a quick spin. 9. Pipette the mixture into an equilibrated cDNA filter cartridge and microcentrifuge 1 min at 10,000g at room temperature and discard the flow-through. 10. Wash cDNA filter cartridge by applying 500 ␮L cDNA wash buffer and microcentrifuge 1 min at 10,000g at room temperature. Discard the flow-through and spin the cDNA filter cartridge for an additional minute at 10,000g to remove all traces of ethanol. 11. Transfer the cDNA filter cartridge to a new tube and elute the dsDNA by applying 12 ␮L of 55◦ C nuclease-free water. Incubate for 2 min at room temperature, then microcentrifuge for 2 min at 10,000g, room temperature. Repeat the elution by reapplying the eluate onto the cDNA filter cartridge and approximately 8 ␮L dsDNA will be eluted. Store the dsDNA overnight at –20◦ C if necessary.

Analysis of the Degradome with the CLIP-CHIPTM Microarray

183

12. Synthesize aRNA by in vitro transcription. Add 12 ␮L of the in vitro Transcription Master Mix to the dsDNA. Mix thoroughly by flicking the tube three times, then microcentrifuge 5 s to collect the reaction mixture at the bottom of the tube. 13. Incubate the reaction for >14 h at 37◦ C in a hybridization oven. 14. Bring the volume of the reaction to 100 ␮L with nucleasefree water to stop the reaction and mix thoroughly by gentle vortexing (see Note 6). 15. Pre-heat nuclease-free water to 60◦ C and purify the synthesized aRNA with the MegaClear kit (or Message Amp II) by adding 350 ␮L aRNA binding buffer to the reaction and mix gently by pipetting. 16. Precipitate aRNA by adding 250 ␮L 100% ethanol and mix gently by pipetting. 17. Pipette the mixture onto an aRNA filter cartridge and microcentrifuge 1 min at 10,000g, room temperature, and discard the flow-through. 18. Wash aRNA filter cartridge by applying 650 ␮L aRNA wash buffer and microcentrifuge 1 min at 10,000g at room temperature. Discard the flow-through and spin the aRNA filter cartridge for an additional minute at 10,000g to remove all traces of ethanol. 19. Transfer aRNA filter cartridge to a new tube and elute the aRNA by adding 100 ␮L of 60◦ C nuclease-free water, incubate for 2 min at room temperature, then microcentrifuge 2 min at 10,000g. 20. Determine the RNA quantity and quality using UV–Vis spectrophotometer (or NanoDrop ND-1000A). 21. Precipitate aRNA by adding 10 ␮L of 5 M ammonium acetate (1:10 volume) and 275 ␮L of 100% ethanol (2.5 volumes) to the elute, mix well, and incubate at –20◦ C for 30 min. 22. Microcentrifuge at 13,000g for 15 min at 4◦ C and remove and discard the supernatant carefully. 23. Wash aRNA pellet by adding 500 ␮L 70% cold ethanol and microcentrifuge for an additional 5 min at 4◦ C. 24. Remove any traces of ethanol by using a very fine tipped pipette, air-dry the pellet for 10 min and resuspend the pellet in nuclease-free water in a desired volume (end concentration <0.2 ␮g/␮L of aRNA). 25. Determine the aRNA concentration. The A 260 /A 280 ratio should be >1.8. If it is lower, sample quality is probably too

184

Kappelhoff, auf dem Keller, and Overall

poor to use for the microarray experiment, and the sample should be discarded. 26. For quality control, electrophorese an aliquot on a 1.2% agarose gel (under RNase-free conditions) to show amplified aRNA as a smear from 250 to 5,000 nucleotides. The average size of unmodified aRNA should be ∼1,200 nucleotides (Fig. 10.1B). 27. Split aRNA into 5–20 ␮g aliquots for microarray labeling and hybridization to minimize multiple freeze/thaw cycles and freeze them at –70◦ C. If possible, use low-binding nuclease-free microfuge tubes. 3.3. Fluorescent Labeling of aRNA and aRNA Fragmentation

For the fluorescent labeling of the unmodified linear amplified aRNA, the universal linkage system (ULS, Kreatech) can be used. ULS is a chemical compound with a reactive platinum group on one side that can form a stable complex with the aRNA by forming a coordinate bond to the N7 position in guanine. On the other side, Cy3- or Cy5-fluorophor groups are linked to the ULS for detection (see Note 10). The big advantage of the ULS labeling system in comparison with others on the market is the easy cleanup between labeled aRNA and free ULS-dye molecules with the KreaPure column matrix which captures free ULS molecules. The yield of recovered labeled aRNA is nearly 100% as it does not bind to the KreaPure column matrix and can be used immediately for hybridization without a second precipitation or cleanup step. 1. If aRNA was frozen, then thaw the samples and determine its concentration again. 2. For each labeling reaction, pipette 2 ␮g aRNA into nuclease-free (low-binding) microfuge tubes and adjust reactions to 16 ␮L with nuclease-free water. 3. Add 2 ␮L of 10x ULS labeling buffer. 4. Add 2 ␮L of either Cy3- or Cy5-ULS to each reaction, mix well, and collect the reaction with a quick spin. Keep the reactions in the dark when possible as this reduces photobleaching of the fluorescent dyes. 5. Incubate samples in the dark for 30 min at 85◦ C. 6. Snap cool samples on ice and collect the reaction at the bottom of the tube by a quick spin. 7. During the incubation period prepare KREApure spin column material for the following purification of unbound dye from the dyed aRNA. Vortex the column material, loosen the cap of the column by a quarter turn, and snap off the bottom closure. 8. Transfer column into a 2 mL collection tube and spin at room temperature for 2 min at 20,000g or at the highest

Analysis of the Degradome with the CLIP-CHIPTM Microarray

185

speed in a microcentrifuge (generally between 13,000 and 18,000g) and discard the flow-through. 9. Wash column material by adding 300 ␮L nuclease-free water on top and centrifuge for an additional 2 min at 20,000g (or high speed). 10. Discard tube and flow-through and transfer column into a nuclease-free 1.5 mL collection tube. 11. Apply ULS-labeled aRNA onto the prepared KREApure spin column material and spin at room temperature for 2 min at 20,000g (or high speed). The flow-through contains the purified labeled aRNA. 12. Determine the concentration and the degree of labeling (DoL) with a spectrophotometer (e.g., NanoDrop) by measuring A260 and A550 for Cy3-ULS-labeled aRNA and A260 and A650 for Cy5-ULS-labeled aRNA (Fig. 10.1C). The degree of labeling can be either calculated by using the DoL calculator on the www.kreatech.com webpage or by using the following formulae: aRNA concentration (ng/␮ L) = A260 × dilution factor × 40/cuvette length in cm Dye concentration (pmol/␮ L) = A550 or A650 × dilution factor/cuvette length in cm × edye × 10−6 Degree of labeling (%) = 340 × pmoldye × 100%/ngaRNA × 1,000(see Note 11) edye Cy3 reagent [1/cm/M] = 150,000 edye Cy5 reagent [1/cm/M] = 150,000 Incorporation rates vary between samples and dyes and should be between 60 and 210 pmol per 2 ␮g aRNA, giving a DoL of 1–3.6 dyes per 100 nucleotides. For background issues and good hybridization, the DoL should not exceed 3.6 otherwise less sample has to be hybridized to the microarray. 13. Pool Cy3-ULS-aRNA and Cy5-ULS-aRNA together and add 4 ␮L 10x fragmentation buffer that contains a high zinc solution. Incubate for 15 min at 70◦ C and stop the fragmentation reaction by adding 1 ␮L 200 mM EDTA, pH 8.0 (stop solution) and place sample on ice. The average size of the resulting fragments will be

186

Kappelhoff, auf dem Keller, and Overall

60–200 nucleotides. The fragmentation step improves hybridization kinetics and can lead to an enhanced signal. 14. Spin the labeled aRNA and vacuum dry the sample by using a SpeedVac and cover the lid to protect the samples from light (see Note 12). 3.4. Pre-hybridization of the CLIP-CHIPTM Microarray

The 70-mer oligonucleotides on the CLIP-CHIPTM microarray are printed in a concentration of 30 ␮M onto the aminosilane glass surface, cross-linked via UV irradiation and stored at room temperature in a desiccated environment. Before use, the microarray slides need to be washed to remove unbound material and the oligonucleotides need rehydration. Denaturation with isopropanol dissolves secondary structure and makes the entire oligonucleotide accessible to the labeled aRNA fragments during hybridization. 1. Place microarray slides into slide holder vials (barcode facing downward) and fill the vial up carefully with pre-heated prehybridization buffer (see Note 13) and incubate for 45 min at 48◦ C. 2. Rinse microarray slides under agitation in distilled water for 30 s at room temperature. Exchange water and repeat rinsing four times. 3. Denature the oligonucleotides on the slides by placing them into isopropanol for 15–30 s at room temperature. 4. Dry slides immediately by transferring them into 50 mL conical tubes and centrifuge for 5 min at 500g at room temperature. Hydrated slides should be used within an hour as hybridization efficiency decreases rapidly if slides are allowed to dehydrate.

3.5. Hybridization of Labeled aRNA to the CLIP-CHIPTM Microarray

1. Dissolve dried aRNA samples in 10 ␮L nuclease-free water. 2. Add 10 ␮L KREAblock to the reaction, mix well, and boil for 3 min at 95◦ C. Briefly spin the samples and place them in the dark. 3. Add 20 ␮L of 2x hybridization buffer (pre-heated to 42◦ C) to the reaction. 4. Pipette 10 ␮L 3x SSC into the two humidifying wells of a microarray hybridization chamber. 5. Place microarray glass slide into hybridization chamber and place a 22×60 mm LifterSlip (see Note 4) over the whole printed area of the microarray or use two 22×25 mm LifterSlips to cover the duplicate CLIP-CHIPTM subarrays individually. 6. Pipette 40 ␮L of reaction mixture closely beside the LifterSlip (20 ␮L for each LifterSlip if using both subarrays).

Analysis of the Degradome with the CLIP-CHIPTM Microarray

187

Surface tension drives the liquid underneath the Teflon bar-produced space between the CLIP-CHIPTM microarray slide and the LifterSlip (Fig. 10.1D). 7. Close the hybridization chamber with a cover and snap the metal closer on. 8. Ensure a constant humidity by wrapping the chamber in wet towels and double zip-lock bags. 9. Place bag in a hybridization oven and incubate for at least 18 h at 42◦ C. 3.6. Washing of the CLIP-CHIPTM Microarray

After hybridization the microarray slides are washed under stringent conditions to ensure that non-specifically bound and nonbinding fragments are washed away (see Note 14). 1. Remove the LifterSlips from the Microarray by dipping and gentle agitation of the slides in 42◦ C pre-heated wash buffer 1. 2. Transfer the slides into a new vial containing wash buffer 1 and incubate the slides for 5 min at 42◦ C with gentle agitation or rotation in a hybridization oven in the dark. Exchange the buffer and incubate for additional 5 min. 3. Exchange with wash buffer 2 and incubate for 5 min at 42◦ C. 4. Exchange with wash buffer 3 and incubate for 5 min at 42◦ C. 5. Dip the microarray slides into RNase-free water and transfer them into 50 mL conical tubes and spin them dry for 5 min at 500g. 6. Place microarray slide in clean dust-free glass slide storage boxes and keep them in the dark.

3.7. Microarray Scanning, Image, and Data Analysis

A large variety of both commercial and non-commercial software suites are available for the analysis of microarray-derived data sets. Priority should be given to packages providing robust and highly accepted quality control, normalization, and statistical analysis algorithms. The analysis pipeline described in this section has been proven to be both user-friendly and reliable for analyzing data obtained from CLIP-CHIPTM microarray experiments. However, it should be mentioned that similar results might also be derived using other software products. 1. Scan slides in a microarray scanner capable of reading spotted arrays (e.g., MWG 428 microarray scanner). Depending on the absorption wavelengths of the dyes used, scan the microarray slides at 550 nm (Cy3) and 650 nm (Cy5) at a resolution of 10 ␮m. Use different settings for the photomultiplier tube (PMT) and laser intensities until the best

188

Kappelhoff, auf dem Keller, and Overall

signal intensity is reached and none of the spots are saturated (65,635 is the highest pixel intensity, which gives a white, total saturated signal; 0 is the lowest pixel intensity and gives a black signal). Signal intensities will fade with time; therefore, scan microarray slides within 3–4 days (Fig. 10.1E). 2. Each microarray slide scan produces a 16-bit grayscale tiff image file for each dye channel. Save both tiff files with unique names that identify the slide number, the sample name, the scanner settings, and the dye used. 3. Open saved tiff files corresponding to one specific microarray slide in appropriate image analysis software like ImaGene (BioDiscovery). Place a grid over the image that contains all information about the numbers and locations of the spots and their gene identity (appropriate templates for the CLIP-CHIPTM are available from the authors upon request or online at www.clip.ubc.ca ). Adjust spot size automatically and process spot intensity data. Some software is capable of flagging empty and bad spots. Check flagged spots and flag spots manually if necessary. Save raw intensity data from each dye channel for normalization and further statistical analysis. 4. Use the web-based CARMAweb application from the Tyrolean Cancer Research Institute in Graz/Austria (https://carmaweb.genome.tugraz.at/carma/) for data normalization and statistical analysis of the raw intensity data (see Note 15). 5. Create a free account and log in to use CARMAweb. All your data are always accessible for you under your account name and will be deleted after 18 days from the server in Austria. All analyzed data can be downloaded onto your own computer at anytime. 6. Go to data directory to upload single or multiple raw intensity data files and start a new analysis. 7. Choose the type of analysis you want to perform, which is the two-color microarray preprocessing and analysis. 8. Choose the data files that should be analyzed in the order they pair together, starting with the Cy3 signal and followed by the Cy5 signal. 9. Define the array type and the used scanning software and check if the settings for the chosen software are correct (see Note 16) and proceed to the two-color array normalization. 10. For the normalization of the microarray data choose the background correction method (the normexp (13) method

Analysis of the Degradome with the CLIP-CHIPTM Microarray

189

has been proven to be reliable for CLIP-CHIPTM derived data), the within-array normalization method (print-tip loess 14), and the between-array normalization method (quantile 15), and check the boxes of those plots that should be drawn by the program during the normalization. 11. If you have replicate arrays proceed to replicate handling and define the number of arrays and files that should be merged in the experiment and proceed to the data analysis window. 12. To determine statistically differentially regulated genes choose the test statistics and the signal that should be used to calculate and determine the differentially expressed genes (either on expression values or on relative expression (M) values). 13. Assign the signals to the groups and choose the test statistic method (permutation, t-test, Wilcoxon, and moderated or paired-moderated t-test, using linear models for microarray data “limma”) or SAM (significance analysis of microarray) (16). 14. Select the models to correct between multiple tests (Bonferroni, Holm, Hochberg, SidakSS and SidakSD, Benjamini & Hochberg, or Benjamini & Yekutieli) and decide which tables and plots should be saved. If a second statistical analysis is preferred it can be added. 15. Start the normalization process and data analysis and wait until the data can be retrieved from the server. Besides text files containing lists of differentially expressed genes, it contains all plots in pdf-file format chosen to be drawn during the microarray analysis. 16. Further analysis to find gene ontologies or gene clusters can be done with the Genesis software (https://carmaweb.genome.tugraz.at/genesis/), integrated in the CARMAweb software, or any other software package capable of drawing heat maps and creating cluster analysis like the MultiExperimentViewer MeV (Fig. 10.1F).

4. Notes 1. Caution: diethylpyrocarbonate (DEPC) is a suspected carcinogen. It breaks down to CO2 and ethanol but may break down to urethane, a possible carcinogen in the presence of ammonia. Use it in a fume hood and wear gloves. 2. To save costs, half of the reactions can be used per sample. Additional purification kits for purifying cDNA (DNAclear,

190

Kappelhoff, auf dem Keller, and Overall

Ambion, cat# AM1756) and aRNA (MEGAclear, Ambion, cat# AM1908) can be purchased separately. 3. Always dilute solutions containing 20x SSC with water or other liquids before adding 10% SDS which precipitates in high salt solution. Pre-heating of buffers helps to prevent this precipitation. 4. Dirty LifterSlips can give a high background. If LifterSlip appears dirty, clean them in the following order with soap, plenty of water, 70% ethanol, 100% ethanol, and acetone. 5. The quality of RNA is very important for any microarray experiment and only intact RNA will give good results. Always work under RNase-free conditions. Spraying and wiping of bench and pipettes with RNaseZap, changing gloves regularly, using nuclease-free filter pipette tips and nuclease-free tubes, and preparing solutions with nucleasefree or DEPC-treated water are mandatory for the success of each microarray experiment. 6. For crucial samples, which have not been DNase I-treated during the RNA extraction, this is still possible after the 14h in vitro transcription step. Just add 70 ␮L RDD Buffer and 10 ␮L DNase I (Qiagen) to the sample and incubate for 20 min at room temperature. Then follow the aRNA purification protocol. 7. RNA isolated from tumor cells normally has higher amounts of DNA due to abnormal gene duplications. Therefore, DNase I treatment of the isolated RNA is very important in regards of cross-hybridization and RNA concentration determination. 8. Native agarose gel electrophoresis may be sufficient to judge the overall integrity and quality of the total RNA preparation by inspecting the 28S and 18S rRNA bands, but the secondary structure alters the migration pattern of RNA in a native gel and will not show the true size of the RNA since the bands are not as sharp as in a denaturing formaldehyde gel. Also the ratio between 28S and 18S being 2:1 cannot be assessed as clearly (Fig. 10.1A). 9. If the aRNA amount is between 0.1 and 100 ng (e.g., RNA from laser capture microdissected samples), two rounds of amplification are possible to obtain enough material for microarray hybridization (17). 10. Besides Cyanine 3 (Cy3)/Cyanine 5 (Cy5) also DY547/ DY647, or Biotin labeling are possible. 11. Formula can be shortened to DoL=34×pmoldye /ngaRNA . 12. During the process of vacuum drying the samples, prehybridization of the CLIP-CHIPTM microarray can be carried out at the same time (Section 3.4).

Analysis of the Degradome with the CLIP-CHIPTM Microarray

191

13. Barcode sticker contains glue that can be dissolved during the isopropanol step. Having the barcode facing down prevents glue from running over the microarray during centrifugation. 14. Wash buffers can be prepared ahead of time. If the buffers show salt formation, filter the solutions before use. Dust is also an enemy of each microarray experiment. Dust on an illuminated microarray slide interferes with the scanning process and can give false-positive spot intensities. 15. CARMAweb (5) (comprehensive R-based microarray analysis web service) is a web-based front end to the R-based Bioconductor package (http://www.bioconductor.org/) designed for the analysis of microarray data. It performs data preprocessing (background correction, quality control, and normalization), detection of differentially expressed genes, cluster analysis, dimension reduction and visualization, classification, and gene ontology-term analysis. 16. Scanning software compatible with the CARMAweb software are Agilent Feature extraction, ArrayVision, BlueFuse, GenePix, ImaGene, QuantArray, Stanford Microarray Database files, and SPOT. 4.1. Experimental Design

There are some important issues that have to be thought through before starting a microarray experiment – the experimental design: 1. RNA samples: the test and control RNA should be isolated and prepared from the same source or donor if possible and in the same way to give an equal quality. Best matching pairs for tumor studies are if tumor sample and control sample are from the same donor (e.g., normal and breast tumor tissue from one patient). If using a tissue culture cell model for a microarray experiment ensure that treated and untreated cells are grown under the same conditions at the same time to eliminate variability. 2. Replicates: for a good statistical analysis, at least three replicates of control and test RNA should be hybridized to the CLIP-CHIPTM (the more samples the better). Biological replicates should always be preferred over technical replicates. For example, in the case of a mouse experiment at least three healthy mice and three disease-induced mice should be sacrificed. In a tissue culture model at least three different dishes mimicking the same conditions should be used. In tumor studies, three tumors, preferably at the same stage of the disease, should be compared to three control tissues. If a control tissue is not available, comparisons can be made using a universal reference RNA. The universal RNA can also

192

Kappelhoff, auf dem Keller, and Overall

be used as an internal control in complicated experiments that involve multiple pair-wise comparison between several samples and to compare any sample to any other sample. Therefore the universal reference RNA is always labeled with the same dye. 3. Slide-to-slide variations: each microarray slide is printed slightly different. These variations are even bigger when different printing batches of microarray chips are used in an experiment and this could affect the results. Therefore, ensure that enough microarray slides from one printing batch are available for the whole experiment or that at least slides from one batch are available for the replicate experiments.

Acknowledgements Christopher M. Overall is supported by a Canada Research Chair in Metalloproteinase Proteomics and Systems Biology, by research grants from the National Cancer Institute of Canada (with funds raised by the Canadian Cancer Association) and from the Canadian Breast Cancer Research Alliance Special Program Grant on Metastasis, and by a Centre Grant from the Michael Smith Research Foundation. Ulrich auf dem Keller is supported by a research fellowship of the Deutsche Forschungsgemeinschaft Germany. References 1. Overall, C. M., Tam, E. M., Kappelhoff, R., Connor, A., Ewart, T., Morrison, C. J., Puente, X., Lopez-Otin, C., and Seth, A. (2004) Protease degradomics: mass spectrometry discovery of protease substrates and the CLIP-CHIP, a dedicated DNA microarray of all human proteases and inhibitors. Biol Chem 385 (6), 493–504. 2. Barrat, A. J., Rawlings, N. D., and Woessner, J. F. (eds.) (1998) Handbook of proteolytic enzymes. Academic Press, New York. 3. Puente, X. S., Sanchez, L. M., Overall, C. M., and Lopez-Otin, C. (2003) Human and mouse proteases: a comparative genomic approach. Nat Rev Genet 4(7), 544–558. 4. Saeed, A. I., Sharov, V., White, J., Li, J., Liang, W., Bhagabati, N., Braisted, J., Klapa, M., Currier, T., Thiagarajan, M., Sturn, A., Snuffin, M., Rezantsev, A., Popov, D., Ryltsov, A., Kostukovich, E., Borisovsky, I.,

Liu, Z., Vinsavich, A., Trush, V., and Quackenbush, J. (2003) TM4: a free, open-source system for microarray data management and analysis. Biotechniques 34(2), 374–378. 5. Rainer, J., Sanchez-Cabo, F., Stocker, G., Sturn, A., and Trajanoski, Z. (2006) CARMAweb: comprehensive R- and bioconductor-based web service for microarray data analysis. Nucleic Acids Res 34, W498–W503. 6. Hokamp, K., Roche, F. M., Acab, M., Rousseau, M. E., Kuo, B., Goode, D., Aeschliman, D., Bryan, J., Babiuk, L. A., Hancock, R. E., and Brinkman, F. S. (2004) ArrayPipe: a flexible processing pipeline for microarray data. Nucleic Acids Res 32, W457–W459. 7. Li, Y., Li, T., Liu, S., Qiu, M., Han, Z., Jiang, Z., Li, R., Ying, K., Xie, Y., and Mao, Y. (2004) Systematic comparison of the fidelity of aRNA, mRNA and T-RNA on gene

Analysis of the Degradome with the CLIP-CHIPTM Microarray

8.

9.

10.

11.

12.

expression profiling using cDNA microarray. J Biotechnology 107(1), 19–28. Feldman, A. J., Costouros, N. G., Wang, E., Qian, M., Marincola, F. M., Alexander, H. R., and Libutti, S. K. (2002) Advantages of mRNA amplification for microarray analysis. Biotechniques 33(4), 906–914. Polacek, D. C., Passerini, A. G., Shi, C., Francesco, N. M., Manduchi, E., Grant, G. R., Powell, S., Bischof, H., Winkler, H., Stoeckert, C. J. Jr., and Davies, P. F. (2003) Fidelity of enhanced sensitivity of differential transcription profiles following linear amplification of nanogram amounts of endothelial mRNA. Physiol Genomics 13, 147–156. Kacharmina, J. E., Crino, P. B., and Eberwine, J. (1999) Preparation of cDNA from single cells and subcellular regions. Methods Enzymol 300, 3–18. Pabon, C., Modrusan, Z., Ruvolo, M. V., Coleman, I. M., Daniel, S., Yue, H., and Arnold, L. J. Jr. (2001) Optimized T7 amplification system for microarray analysis. Biotechniques 31(4), 874–879. Van Gelder, R. N., von Xastrow, M. E., Yool, A., Dement, D. C., Barchas, J. D., and Eberwine, J. H. (1990) Amplified RNA synthesized from limited quantities of heteroge-

13.

14. 15.

16.

17.

193

neous cDNA. Proc Natl Acad Sci U S A 87, 1663–1667. Ritchie, M. E., Silver, J., Oshlack, A., Holmes, M., Diyagama, D., Holloway, A., and Smyth, G. K. (2007) A comparison of background correction methods for twocolour microarrays. Bioinformatics 23(20), 2700–2707. Smyth, G. K. and Speed, T. (2003) Normalization of cDNA microarray data. Methods 31(4), 265–273. Yang, Y. H. and Thorne, N. P. (2003) Normalization for two-color cDNA microarray data. In Goldstein D. R. (ed.), Science and Statistics: A Festschrift for Terry Speed, IMS Lecture Notes – Monograph Series, 40, 403–418. Tusher, V. G., Tibshirani, R., and Chu, G. (2001) Significance analysis of microarrays applied to the ionizing radiation response. Proc Natl Acad Sci U S A 98(9), 5116–5121. Luo, L., Salunga, R. C., Guo, H., Bittner, A., Joy, K. C., Galindo, J. E., Xiao, H., Rogers, K. E., Wan, J. S., Jackson, M. R., and Erlander, M. G. (1999) Gene expression profiles of laser-captured adjacent neuronal subtypes. Nature Med 5(1), 117–122.

Chapter 11 In Situ Hybridization for Metalloproteinases and Their Inhibitors Tiina L. Hurskainen and Suneel S. Apte Abstract In situ hybridization (ISH) is an invaluable tool in understanding tissue-specific gene expression and gene regulation within a spatial context and at a resolution that is not possible by any other method. In this chapter, we provide ISH methodology that has successfully been applied to the detection of metalloproteinases and their inhibitors. Key words: In situ hybridization, localization, tissue section, cRNA, DIG.

1. Introduction 1.1. The Role of In Situ Hybridization in Metalloprotease and TIMP Biology

The completion of the human and mouse genome projects has revealed all metalloproteases and metalloprotease inhibitors such as those belonging to the MMP, TIMP, ADAM and ADAMTS (1) families. Considerable effort will subsequently be required to undertake a systematic examination of the expression, function, and regulation of these genes. In situ hybridization (ISH) is invaluable in understanding tissue-specific gene expression and gene regulation within a spatial context and at a resolution that is not possible by any other method. While northern analysis and RT-PCR are amenable to quantitation, they are bulk methods, and cannot pinpoint the precise cell type which is expressing a gene. This is desirable information in organs which contain a heterogeneous cell pop-

I.M. Clark (ed.), D.A. Young, A.D. Rowan (Consulting editors), Matrix Metalloproteinase Protocols, Methods in Molecular Biology 622, DOI 10.1007/978-1-60327-299-5 11, © Springer Science+Business Media, LLC 2001, 2010

195

196

Hurskainen and Apte

ulation, such as the brain, bone marrow, lung, and so on, but less so in relatively monotypic tissues such as cartilage and muscle. Localization of mRNA expression has been pivotal in developing a hypothesis of function and examining this in the context of a knock-out mouse or a naturally occurring gene mutation. For example, Mmp9 is highly expressed by cells of the monocytemacrophage lineage, particularly osteoclasts, in primary metaphyseal trabeculae, suggesting a role in cartilage resorption during endochondral ossification (2). Targeted inactivation of Mmp9 has demonstrated that this is indeed the case, since MMP-9 deficient mice have delayed resorption of hypertrophic cartilage (3). Mmp14 (encoding mouse MT1-MMP or MMP-14) is expressed in osteoclasts, osteoblasts, and dense connective tissues such a joint capsules, ligaments, and tendons (4, 5). Consequently Mmp14 inactivation in transgenic mice (6, 7) leads to severe anomalies of skeletal development. Elucidation of the expression pattern has thus been helpful in understanding the mechanism underlying the phenotype of these two knock-out mice, the only ones among MMP or TIMP gene knockouts to show a developmental abnormality. ISH may be performed on cells, tissue sections, or whole mounts of embryos or specific organs. Whole mount studies are useful for studying genes with a role in pattern formation and are not generally used in studies of metalloproteinases and their inhibitors. In this chapter we provide a simple annotated nonisotopic method which is optimized for the use of digoxigenin (dig)-labeled cRNA probes on tissue sections, but may easily be adapted for other material. cRNA probes are preferred over double-stranded cDNA probes because of their greater specificity; in this method, antisense cRNA provides experimental information and the sense cRNA acts as a negative control. Double-stranded cDNA probes may self-anneal and less probe is then available for hybridization to target mRNA. The greater strength of cRNA-mRNA associations allows hybridization to be performed at much higher stringency and specificity. Sense and antisense oligonucleotide probes may also be used in a similar fashion (8). 1.2. Caveats in the Use of ISH

The presence of mRNA implies but does not always predict that the respective protein will be present. Many MMPs are secreted as proenzymes and thus the presence of mRNA or even protein does not necessarily imply activity. In some instances, mRNA may be present in one cell type and the protein in another (9). This is an apparent discrepancy; it may be explained by the uptake of secreted protein by another cell type via cellular receptors. In the case of transmembrane metalloproteases there may be alternatively spliced soluble forms or sheddable extracellular domains that may show a different immunohistological localization from

In Situ Hybridization for Metalloproteinases and Their Inhibitors

197

the mRNA. Although attempts have been made to derive quantitative data from ISH, such data is meaningless at worst and semiquantitative at best, and detracts from the real value of ISH as a descriptive tool for gene regulation analyses. 1.3. Isotopic ISH (IISH) or Nonisotopic ISH (NISH)?

Proponents of IISH argue that NISH is less sensitive and is not useful for detecting low levels of expression. On the other hand, NISH does offer a fast and simple technique that relieves the investigator of two major drawbacks of IISH i.e., the use of radioisotope and the need for autoradiography. Another practical advantage of NISH is that dual dark field/bright field images are not required for visualization of the signal and tissue morphology. Bright field images of NISH make for more informative and elegant illustrations and presentations. NISH is rapid; the NISH protocol we describe here may be completed in as little as two whole days, more usually three. IISH may require a period of days to weeks depending on which isotope is used and how high the levels of probe labeling and gene expression are. NISH is cheaper, since autoradiographic emulsion and radioisotope used for ISH are more expensive than the labeling and detection reagents used in NISH and have a shorter shelf life. Additionally, NISH has the advantage of providing accurate subcellular localization of mRNA which may not be provided by IISH because of the scatter of radiation. Subcellular localization of mRNA is particularly useful in studies of polarized cells as we have shown recently for MMP-19 (Hurskainen, T. and Apte, S.S., unpublished data).

2. Materials Unless specified, reagents are from Sigma or Fisher. Solution volumes may be scaled up or down as required. Reagents with a long shelf life such as DEPC-water and PBS may be stored at ambient temperatures. Where required, solutions can be filtered with a 0.2 ␮m filter. Refer to manufacturer’s material safety data sheets (MSDS) for detailed information on hazards of individual chemicals. 1. 8% Paraformaldehyde (PFA) stock (Caution: PFA is toxic and must be handled in a fume hood and disposed of correctly): add 8 g of PFA powder to 80 mL of heated distilled water (approx 80◦ C), add one or more drops of 10 N NaOH to dissolve the paraformaldehyde and equilibrate pH to about 7.4, stir until dissolved, make up to 100 mL, and store at 4◦ C. Good for 2–3 mo.

198

Hurskainen and Apte

2. Fixative (4% PFA in PBS): mix equal vol of 8% PFA and 2X PBS. Store at 4◦ C. Good for one month. 3. 10X Phosphate buffered saline (PBS): 80 g NaCl, 2 g KCl, 11.5 g Na2 HPO4 , 2 g KH2 PO4 per liter. Store at 4◦ C. Dilute to 1X PBS with distilled water, and sterilize by autoclaving. 4. Diethylpyrocarbonate (DEPC)-treated water (Caution: DEPC is toxic and must be handled in a fume hood): add 1 mL of DEPC to 1 L of water in a glass bottle, shake well and let stand overnight in fume hood. The next morning, autoclave and store at room temperature. We re-use the same bottles for successive DEPC treatments without washing. Overnight incubation with the DEPC is essential for inactivation of ribonucleases; autoclaving subsequently destroys the DEPC. 5. 0.2 N HCl: add 0.5 mL concentrated HCl (approx 12 N) to 30 mL distilled water. Use a plugged pipet tip to transfer conc. HCl. Make just before use at ambient temperature. Do not add water to concentrated acid. 6. Acetylation mix (Caution: Acetic anhydride is toxic and must be handled in a fume hood). To make 50 mL of acetylation mix, add 660 ␮L of triethanolamine and 125 mL of acetic anhydride to 50 ␮L of distilled water. Make immediately prior to use (acetic anhydride is unstable) and incubate slides in this solution with constant stirring or rocking at room temperature. 7. 20X SSC: 3 M NaCl, 300 mM sodium citrate, pH adjusted to 7.0 with 1 M citric acid. 8. Hybridization solution (Caution: Formamide is toxic and must be handled in a fume hood and disposed of correctly; SDS powder is a respiratory irritant if inhaled). 50% formamide (Fluka, molecular biology grade), 10 mM Tris-HCl, pH 7.6, 200 mg/mL tRNA (Nuclease free, Sigma), 1X Denhardt’s solution, 10% dextran sulfate, 600 mM NaCl, 0.25% sodium lauryl sulphate (SDS, enzyme grade), 1 mM EDTA (tetrasodium salt). Store at –20◦ C in 1 mL aliquots. 9. Denhardt’s solution: 0.02% (w/v) Ficoll, 0.02% (w/v) polyvinylpyrrolidone and 10 mg/mL RNAse free bovine serum albumin. Make as a 50X solution, aliquot, and store at –20◦ C. 10. TNE: 10 mM Tris-HCl, pH 7.5, 0.5 M NaCl, 1 mM EDTA. Filter and store at room temperature. 11. Substrate mix: Nitro blue tetrazolium chloride (NBT) stock is 75 mg/mL (in 70% dimethylformamide) and

In Situ Hybridization for Metalloproteinases and Their Inhibitors

199

5-bromo-4-chloro-3-indolyl phosphate (BCIP) stock is 50 mg/mL (in dimethylformamide). Store at –20◦ C. (NBT and BCIP are purchased from Roche Biosciences and are also available as a premixed stock). Levamisole (Sigma) stock is 1M in water and is stored at –20◦ C. Levamisole is an inhibitor of many isoforms of alkaline phosphatase, other than that which is tagged to the antibody used here. It is used as an inhibitor of endogenous alkaline phosphatase. Working substrate solution is prepared fresh from the above stocks by mixing 4.5 ␮L of NBT and 3.5 ␮L of BCIP with 1.0 ␮L of levamisole in 1.0 mL of DIG3 buffer. 12. DIG 1 solution: 100 mM Tris-HCl, pH 7.5, 150 mM NaCl. Filter after preparation. 13. DIG 3 solution: 100 mM Tris-HCl, pH 9.5, 100 mM NaCl, 50 mM MgCl2 . Make the Tris plus NaCl solution first, adjust pH to 9.5, then add MgCl2 , otherwise the MgCl2 will precipitate out. Filter after preparation. This solution is stable for 6 months at ambient temperature. Note that pH 9.5 is critical for optimal activity of alkaline phosphatase. 14. Plastic peel away molds (Polysciences). 15. Superfrost Plus slides (Fisher). 16. Tissue embedding medium, e.g., OCT (Miles Scientific). 17. PAP pen or grease pencil. 18. Methyl green (Sigma). 19. Aqua Polymount cover slip mounting medium (Polysciences Inc). 20. Proteinase K (nuclease free, Roche Biosciences): 10 ␮g/mL in 10 mM Tris-HCl, pH 7.4, 1 mM calcium chloride. This solution should be prepared fresh and warmed to 37◦ C before use. 21. RNAse A (Roche Biosciences). 22. Digoxigenin-labeled UTP (as Dig-labeling mix, Roche Biosciences). 23. RNA polymerases and restriction enzymes (Promega, New England Biolabs or Roche Biosciences).

3. Methods 3.1. Tissue Fixation

1. Fix as required in 4% paraformaldehyde in PBS at 4◦ C without agitation (see Note 1).

200

Hurskainen and Apte

3.2. Sectioning

1. Cryosections or paraffin sections of 5–10 ␮m thickness may be used (see Note 2).

3.3. Preparation of cDNA Template (see Note 3)

1. Cut 5–10 ␮g of plasmid DNA with a suitable restriction enzyme to generate a blunt or 3 recessed end. 2. Electrophorese the entire digest on a preparative gel to confirm completeness of the digest and cut out the band. It is very important to ensure that there is no circular DNA remaining at the end of the digestion. If there is difficulty in getting complete digests, cut less DNA in a larger final volume and add 1–10 U more enzyme after 2 h of digestion. Digest for another hour. 3. Elute the DNA from the gel slice using the Gene Clean kit (BIO 101). The protocol recommended by the manufacturer is modified so that DNA is eluted with 500 ␮L sodium iodide and 15 ␮L of glass milk. DNA is eluted into 10 ␮L DEPCtreated water. Assuming at least 75% recovery, this will be sufficient for 5–10 transcription reactions.

3.4. In Vitro Transcription

1. Mix the following in a microcentrifuge tube (1.5 mL) in a total vol of 20 ␮L: 2 ␮L each of 10X transcription buffer and Dig-labeling mix, and 1 ␮g of linearized DNA template. 2. Make up the volume with DEPC-treated water and add 1 ␮L of the appropriate RNA polymerase (T7, T3 or SP6, purchased from Promega or Roche Biosciences). 3. Incubate for 1.5 h at 37◦ C. 4. Electrophorese 1 ␮L of the reaction added to 1 ␮L of 10X loading buffer (RNAse free) and 8 ␮L of DEPCtreated water. Use a 1% agarose minigel containing ethidium bromide to visualize the template and transcript (see Note 4). 5. Add 2 ␮L of DNAse I and incubate for 15 min at 37◦ C to destroy the template.

3.5. Ethanol Precipitation of Probe

1. To the transcription reaction, add 2 ␮L of 3M sodium acetate (pH 5.2, made in DEPC-treated water), 2 ␮L of yeast tRNA (RNAse free, 20 mg/mL in DEPCtreated water, used as RNA co-precipitant) and 200 ␮L of ice-cold absolute ethanol (dedicated for RNA work). 2. Vortex lightly, and keep at –70◦ C to –80◦ C for 20 min. 3. Centrifuge at 14,000g (in a microfuge) for 15 min at 4◦ C. 4. Discard the ethanol by inverting the tube and add 200 ␮L of 70% ethanol (diluted with DEPC-treated water). 5. Centrifuge again for 5 min.

In Situ Hybridization for Metalloproteinases and Their Inhibitors

201

6. Discard the supernatant by inverting, allow the pellet to dry and suspend in 50 ␮L of DEPC-water. Store at –70◦ C. 3.6. Hybridization (see Note 5)

Slides are prepared for hybridization as follows: 1. Immerse slides in 0.2 N hydrochloric acid for 15 min at ambient temperature. Incubate (wash) slides twice in PBS (5 min each). 2. Treat slides with freshly prepared proteinase K (10 ␮g/mL) warmed to 37◦ C prior to use. Incubate (wash) slides twice in PBS (5 min each) (see Note 6). 3. Postfixation: Immerse slides in 4% PFA for 5 min at room temperature (see Note 7). 4. Acetylation: Immerse slides in acetylation mix for 15 min at ambient temperature. Incubate (wash) slides twice in PBS (5 min washes each) (see Note 8). Note: steps 3 and 4 should be carried out in an externally vented, fume hood and the user should use appropriate personal protection such as a lab coat, gloves and safety glasses. 5. Dehydrate the sections successively in 70%, 90%, and two changes of 100% ethanol (2 min in each change). Allow the slides to dry completely, sections up. Use a PAP pen or grease pencil to draw circles around each section and create wells. 6. Hybridization (see Note 9). Preincubate aliquots of the hybridization fluid at 80◦ C for 10 min. Add probe to the desired volume of hybridization fluid (usually 150 ␮L/section). Vortex vigorously to disperse the probe, heat for 3 min more, and add to preheated slides (80◦ C). Hybridize overnight at 50–55◦ C. We use 52◦ C for most probes. The temperature of hybridization may be titrated as indicated for different probes.

3.7. Stringency Washes and Ribonuclease Treatment

1. Pour off the hybridization fluid and rinse briefly in 5X SSC at 50–60◦ C. 2. Place the slides in 2X SSC/50% formamide for 30 min at 56◦ C. (Caution: Use formamide in a fume hood, particularly when it is heated, and collect the waste for proper disposal). 3. Immerse the slides in TNE at 37◦ C for 10 min, then in TNE with RNAse A (10 ␮g/mL) at 37◦ C for 30 min (see Note 11). 4. Place the slides in TNE at 37◦ C for 10 min to remove excess RNAse A. 5. Carry out stringency washes next, as follows: In 2X SSC at 50–60◦ C for 20 min (one wash) and 0.2X SSC at 50–60◦ C for 20 min (two washes).

202

Hurskainen and Apte

3.8. Immunohistochemical Detection of DIG and Counterstaining

1. Place slides in DIG1 buffer at room temperature for 5 min and then incubate with 1.5% (w/v) blocking reagent (Roche Biosciences) in DIG1 buffer for 30 min (see Note 12). 2. Following this incubation, dip the slides once or twice in another change of DIG1 buffer and place 1:500 to 1:1000 anti-Dig antibody (sheep anti-Dig antibody, Fab fragment, alkaline phosphatase conjugated, Roche BioSciences) diluted in DIG1 buffer in each well. 3. Return the slides to the humid box and incubate for an hour at room temperature (note that this incubation may also be carried out overnight, but at 4◦ C). 4. Wash the antibody off with two rinses in DIG1 buffer (15 min each) followed by immersion in DIG3 buffer (5 min) to allow equilibration of sections for substrate deposition. 5. Prepare working substrate solution and add 100–300 ␮L of substrate/slide. Monitor continuously for development of magenta color. We find that genes with high level, specific expression (such as the col2a1 gene expressed specifically in cartilage) develop a color product within one hour. Most other genes require incubation periods ranging from 3–18 h. 6. Rinse the slides in Tris-EDTA buffer (pH 7.5) or in distilled water when the desired signal/noise ratio is reached, then rinse in distilled water. 7. Counterstain sections in 0.3% (w/v) methyl green in distilled water (15–60 s), then rinse in cold running tap water to remove excess stain or until desired stain intensity is achieved. 8. Blot the slides dry with a paper towel and mount coverslips with warmed (to 45◦ C) Aqua Polymount (Polysciences). Remove air-bubbles. Coverslips may be held in place with nail-varnish.

4. Notes 1. Procuring tissue sections: Although in situ hybridization has been successfully performed on decades old archival tissue, it is best to procure and process tissue specifically for this purpose and to fix it immediately to be assured of mRNA preservation. In general, mRNA preservation is inconsistent in whole mouse embryos older than 15.5 d or in pieces of tissue greater than 0.5 cm thick in their smallest dimension. It is recommended that older embryos have multiple

In Situ Hybridization for Metalloproteinases and Their Inhibitors

203

punctures or slits made in the skin to allow penetration of fixative. Fresh 4% paraformaldehyde in PBS should be made just prior to fixation and tissue immersed in at least 10 vol of fixative at 4◦ C. Agitation is not necessary; vigorous agitation may damage delicate tissues. The duration of fixation should be adjusted as required to optimize mRNA preservation and to permit probe access to the mRNA. Underfixation may result in loss of tissue architecture during the ISH protocol whereas over-fixation may diminish the signal intensity. As a rule of thumb, overnight fixation (18 h) for tissue not less than 3 cubic mm but not larger than 1 cubic cm is effective. Very small pieces of tissue may require short fixation times. 2. Tissue embedding and sectioning: Following fixation, the tissue may be processed for cryosections or embedded in paraffin. Paraffin embedding is most reproducible in an automated tissue processor, allowing well controlled processing and vacuum infiltration prior to embedding in paraffin. For cryosections, following fixation, the tissue is rinsed briefly in PBS, then immersed in 20% w/v sucrose in PBS (autoclaved, stored at 4◦ C) until it sinks (usually 24–48 h). This provides the tissue with a buoyant density similar to most commercially available freezing media and facilitates sectioning. Freezing is done in plastic peelaway molds. Tissue is placed in the desired orientation, freezing medium (e.g., OCT compound) poured around it (avoiding formation of air bubbles) and frozen rapidly by immersion in liquid nitrogen or in a slurry of dry ice and ethanol. We have frozen tissue simply by placing the blocks on the shelf of a –80◦ C freezer without detriment of mRNA or appearance of freeze artefact. These blocks of frozen tissue are now ready for sectioning or may be stored at–80◦ C in an airtight container such as sealable freezer bags to prevent dehydration. We have successfully used tissue stored for up to one year. Note that tissues such as muscle and liver are prone to freeze artefact and should be frozen rapidly. Cryosections should be taken on Superfrost Plus slides (Fisher) and air-dried for up to 3 h. Dried sections may be stored at –70◦ C in an airtight box for up to 3 wk. We find that signal decreases after this time. Prior to use, slides are allowed to warm to room temperature and dry again. A PAP pen or grease pencil may be used to encircle the section thus allowing retention of fluids used for each step. Paraffin sections should be taken on Superfrost Plus slides and stored in a clean box at room temperature. They will

204

Hurskainen and Apte

last indefinitely. If sections are taken on the lower half of each slide, smaller volumes of all reagents and washes are required during in situ hybridization. 3. Generating cloned cDNA templates for ISH probes: cRNA probes are generated by in vitro transcription using prokaryotic RNA polymerases and cloned cDNA fragments. We describe at length in this section the requirements for and procurement of cloned DNA fragments. An alternative is to use PCR generated templates containing RNA polymerase promoter sequences. New genes: the first step is to clone cDNA fragments of a suitable size into a vector suitable for generation of labeled riboprobes. Inserts ranging from 200 bp to 900 bp work well. Shorter fragments may not be labeled sufficiently and longer fragments may not permeate into cells, although this problem may be circumvented by hydrolysis of long probes. The vectors used for transcription should have a versatile multiple cloning site (MCS), which should be immediately flanked by RNA polymerase binding sites (T3,