Reporter Genes. A Practical Guide

Reporter Genes A Practical Guide

M E T H O D S I N M O L E C U L A R B I O L O G Y™

John M. Walker, SERIES EDITOR 436. Avian Influenza Virus, edited by Erica Spackman, 2008 435. Chromosomal Mutagenesis, edited by Greg Davis and Kevin J. Kayser, 2008 434. Gene Therapy Protocols: Vol. 2: Design and Characterization of Gene Transfer Vectors, edited by Joseph M. LeDoux, 2008 433. Gene Therapy Protocols: Vol. 1: Production and In Vivo Applications of Gene Transfer Vectors, edited by Joseph M. LeDoux, 2008 432. Organelle Proteomics, edited by Delphine Pflieger and Jean Rossier, 2008 431. Bacterial Pathogenesis: Methods and Protocols, edited by Frank DeLeo and Michael Otto, 2008 430. Hematopoietic Stem Cell Protocols, edited by Kevin D. Bunting, 2008 429. Molecular Beacons: Signalling Nucleic Acid Probes, Methods and Protocols, edited by Andreas Marx and Oliver Seitz, 2008 428. Clinical Proteomics: Methods and Protocols, edited by Antonio Vlahou, 2008 427. Plant Embryogenesis, edited by Maria Fernanda Suarez and Peter Bozhkov, 2008 426. Structural Proteomics: High-Throughput Methods, edited by Bostjan Kobe, Mitchell Guss, and Huber Thomas, 2008 425. 2D PAGE: Vol. 2: Applications and Protocols, edited by Anton Posch, 2008 424. 2D PAGE: Vol. 1: Sample Preparation and Pre-Fractionation, edited by Anton Posch, 2008 423. Electroporation Protocols, edited by Shulin Li, 2008 422. Phylogenomics, edited by William J. Murphy, 2008 421. Affinity Chromatography: Methods and Protocols, Second Edition, edited by Michael Zachariou, 2008 420. Drosophila: Methods and Protocols, edited by Christian Dahmann, 2008 419. Post-Transcriptional Gene Regulation, edited by Jeffrey Wilusz, 2008 418. Avidin–Biotin Interactions: Methods and Applications, edited by Robert J. McMahon, 2008 417. Tissue Engineering, Second Edition, edited by Hannsjörg Hauser and Martin Fussenegger, 2007 416. Gene Essentiality: Protocols and Bioinformatics, edited by Svetlana Gerdes and Andrei L. Osterman, 2008 415. Innate Immunity, edited by Jonathan Ewbank and Eric Vivier, 2007 414. Apoptosis in Cancer: Methods and Protocols, edited by Gil Mor and Ayesha Alvero, 2008

413. Protein Structure Prediction, Second Edition, edited by Mohammed Zaki and Chris Bystroff, 2008 412. Neutrophil Methods and Protocols, edited by Mark T. Quinn, Frank R. DeLeo, and Gary M. Bokoch, 2007 411. Reporter Genes: A Practical Guide, edited by Don Anson, 2007 410. Environmental Genomics, edited by Cristofre C. Martin, 2007 409. Immunoinformatics: Predicting Immunogenicity In Silico, edited by Darren R. Flower, 2007 408. Gene Function Analysis, edited by Michael Ochs, 2007 407. Stem Cell Assays, edited by Vemuri C. Mohan, 2007 406. Plant Bioinformatics: Methods and Protocols, edited by David Edwards, 2007 405. Telomerase Inhibition: Strategies and Protocols, edited by Lucy Andrews and Trygve O. Tollefsbol, 2007 404. Topics in Biostatistics, edited by Walter T. Ambrosius, 2007 403. Patch-Clamp Methods and Protocols, edited by Peter Molnar and James J. Hickman, 2007 402. PCR Primer Design, edited by Anton Yuryev, 2007 401. Neuroinformatics, edited by Chiquito J. Crasto, 2007 400. Methods in Membrane Lipids, edited by Alex Dopico, 2007 399. Neuroprotection Methods and Protocols, edited by Tiziana Borsello, 2007 398. Lipid Rafts, edited by Thomas J. McIntosh, 2007 397. Hedgehog Signaling Protocols, edited by Jamila I. Horabin, 2007 396. Comparative Genomics, Vol. 2, edited by Nicholas H. Bergman, 2007 395. Comparative Genomics, Vol. 1, edited by Nicholas H. Bergman, 2007 394. Salmonella: Methods and Protocols, edited by Heide Schatten and Abraham Eisenstark, 2007 393. Plant Secondary Metabolites, edited by Harinder P. S. Makkar, P. Siddhuraju, and Klaus Becker, 2007 392. Molecular Motors: Methods and Protocols, edited by Ann O. Sperry, 2007 391. Methicillin-Resistant Staphylococcus aureus (MRSA) Protocols, edited by Yinduo Ji, 2007 390. Protein Targeting Protocols, Second Edition, edited by Mark van der Giezen, 2007 389. Pichia Protocols, Second Edition, edited by James M. Cregg, 2007

M E T H O D S I N M O L E C U L A R B I O L O G Y™

Reporter Genes A Practical Guide

Edited by

Donald S. Anson Department of Genetic Medicine, Women’s and Children’s Hospital, Adelaide, Australia

HUMANA PRESS

TOTOWA, NEW JERSEY

© 2007 Humana Press Inc. 999 Riverview Drive, Suite 208 Totowa, New Jersey 07512 www.humanapress.com All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording, or otherwise without written permission from the Publisher. Methods in Molecular BiologyTM is a trademark of The Humana Press Inc. All papers, comments, opinions, conclusions, or recommendations are those of the author(s), and do not necessarily reflect the views of the publisher. This publication is printed on acid-free paper. ∞ ANSI Z39.48-1984 (American Standards Institute) Permanence of Paper for Printed Library Materials. Cover illustration: Figure 1, Chapter 11, “Green Fluorescent Protein as a Tracer in Chimeric Tissues,” by Harold Jockusch and Daniel Eberhard. Fig. 1A from Chapter 11, "Human Solid Tumors," by Pietro Production Editor: Michele Seugling Cover design by Karen Schulz For additional copies, pricing for bulk purchases, and/or information about other Humana titles, contact Humana at the above address or at any of the following numbers: Tel.: 973-256-1699; Fax: 973-256-8341; E-mail: [email protected]; or visit our Website: www.humanapress.com Photocopy Authorization Policy: Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients, is granted by Humana Press Inc., provided that the base fee of US $30.00 per copy is paid directly to the Copyright Clearance Center at 222 Rosewood Drive, Danvers, MA 01923. For those organizations that have been granted a photocopy license from the CCC, a separate system of payment has been arranged and is acceptable to Humana Press Inc. The fee code for users of the Transactional Reporting Service is: [978-1-58829-739-6/07 $30.00]. Printed in the United States of America. 10 9 8 7 6 5 4 3 2 1 e-ISBN: 978-1-60327-072-4 Library of Congress Cataloging-in-Publication Data Reporter genes : a practical guide / edited by Donald S. Anson. p. ; cm. — (Methods in molecular biology ; 411) Includes bibliographical references and index. ISBN-13: 978-1-58829-739-6 (alk. paper) 1. Reporter genes. 2. Mammals—Cytology. 3. Mammals— Genetics. I. Anson, Donald. II. Series: Methods in molecular biology (Clifton, N.J.); v.411. [DNLM: 1. Genes, Reporter. 2. Cytological Techniques. 3. Mammals—genetics. W1 ME9616J v.411 2007 / QU 470 R425 2007] QH447.8.R47R47 2007 572.8'19—dc22

2007004637

Preface Reporter genes have played, and continue to play, a vital role in many areas of biological research by providing a ready means for qualitative and quantitative assessment of the activity of genes and location of gene products in different environments. For example, reporter genes have played a major role in defining the activity of different genetic elements that control transcription of genes, both in vitro and in vivo (1,2), in the study of cell lineages (2,3) and in determining the effectiveness of different gene transfer technologies (4,5). While early reporter genes required fixation of cells for visualization, or the preparation of cell extracts for quantitative assays, there has been a move towards reporter genes that can be assayed quantitatively and/or qualitatively in live cells and animals. The two widely used examples of such markers are the fluorescent proteins (6–8) and luciferase (9,10). However, despite the development of these new reporter genes, one of the earliest, E. coli β-galactosidase (LacZ), is still widely used as a marker and offers significant advantages, especially for histological analysis (11). Most reporter genes originate from non-mammalian species, and most have subsequently been modified to enhance their expression in mammalian cells and/or to modify their characteristics, vastly increasing their usefulness. For example, codon optimisation for expression in mammalian cells has been applied to the fluorescent proteins (12), luciferase (13) and β-galactosidase (14). The characteristics of fluorescent proteins have been extensively modified, with variants of many different colours and characteristics now available (15,16 and see chapter by Patterson in this book). LacZ (14,17,18) and fluorescent proteins (18,19) can also be targeted to the nucleus of cells by addition of suitable trafficking signals. Fluorescent proteins are also widely used for making fusion proteins to allow analysis of sub-cellular trafficking and compartmentalisation of proteins of interest (2,20,21). Reporter genes have provided powerful tools for analysis of gene expression, either by localisation and/or by quantitative analysis. Examples of the former include the use of LacZ, where staining of microscopic sections gives information on gene expression at cellular resolution, and fluorescent proteins, where direct visualisation of tissue can give similar information. The use of luciferase as a marker gene in vivo also results in information regarding the localisation of gene expression in the living animal, although at a much lower resolution. In this system in vivo gene expression can be quantified by photonic imaging (see chapter by Ray and Gambhir). v

vi

Preface

Most marker genes can be used to provide quantitative data regarding gene expression. However, in some cases this requires the preparation of cell lysates. For example, β-galactosidase (LacZ) can be quantified by ELISA or enzyme activity and kits for this are available from a number of manufacturers, as are kits for quantitative analysis of several other enzymatic reporter genes such as alkaline phosphatase and luciferase. Reporter genes that can be quantitatively assayed without the preparation of a cell lysate include luciferase (via photonic imaging), secreted alkaline phosphatase (enzyme assay) and fluorescent proteins (via FACS analysis or microscopy). New marker/reporter gene systems are currently being developed. For example, systems based on the use of positron-emission tomography (PET) offer a means for non-invasive imaging of all tissues (22,23). While these are not covered in this book, the development of microPET machines suitable for imaging rodents means that this technology is likely to be rapidly developed over the next few years. As indicated above, molecular engineering is constantly being used to improve existing reporter systems. This book will describe practical protocols for experimentation with the most useful reporter genes for mammalian systems that are available and will concentrate on those marker genes that are currently most commonly used. Donald S. Anson

References 1. Mayer-Kuckuk, P., Menon, L. G., Blasberg, R. G., Bertino, J. R., and Banerjee, D. (2004) Role of reporter gene imaging in molecular and cellular biology. Biol. Chem. 385, 353–361. 2. Yu, Y. A., Szalay, A. A., Wang, G., and Oberg, K. (2003) Visualization of molecular and cellular events with green fluorescent proteins in developing embryos: a review. Luminescence 18, 1–18. 3. Trainor, P. A., Zhou, S. X., Parameswaran, M., et al. (1999) Application of lacZ transgenic mice to cell lineage studies. Methods Mol. Biol. 97, 183–200. 4. Bogdanov, A. Jr. (2003) In vivo imaging in the development of gene therapy vectors. Curr. Opin. Mol. Ther. 5, 594–602. 5. McCaffrey, A., Kay, M. A., and Contag, C. H. (2003) Advancing molecular therapies through in vivo bioluminescent imaging. Mol. Imaging 2, 75–86. 6. Hoffman, R. M. (2005) Advantages of multi-color fluorescent proteins for wholebody and in vivo cellular imaging. J. Biomed. Opt. 10, 41202-1–41202-10. 7. Passamaneck, Y. J., Di Gregorio, A., Papaioannou, V. E., and Hadjantonakis, A. K. (2006) Live imaging of fluorescent proteins in chordate embryos: from ascidians to mice. Microsc. Res. Tech. 69, 160–167.

Preface

vii

8. Wiedenmann, J. and Nienhaus, G. U. (2006) Live-cell imaging with EosFP and other photoactivatable marker proteins of the GFP family. Expert Rev. Proteomics 3, 361–374. 9. Sadikot, R. T. and Blackwell, T. S. (2005) Bioluminescence imaging. Proc. Am. Thorac. Soc. 2, 537–540. 10. Welsh, D. K. and Kay, S. A. (2005) Bioluminescence imaging in living organisms. Curr. Opin. Biotechnol. 16, 73–78. 11. Franco, D., de Boer, P. A., de Gier-de Vries, C., Lamersm W. H., and Moorman, A. F. (2001) Methods on in situ hybridization, immunohistochemistry and betagalactosidase reporter gene detection. Eur. J. Morphol. 39, 169–191. 12. Yang, T. T., Cheng, L., and Kain, S. R. (1996) Optimized codon usage and chromophore mutations provide enhanced sensitivity with the green fluorescent protein. Nucleic Acids Res. 24, 4592–4593. 13. Promega Inc sells synthetic luc2 (Photinus pyralis, see http://www.promega.com/ pnotes/89/12416_07/12416_07.pdf) and hRluc (Renilla reniformis, see http://www. promega.com/pnotes/79/9492_06/9492_06.pdf) genes that are codon-optimised for expression in mammalian cells. 14. Anson, D. S. and Limberis, M. (2004) An improved beta-galactosidase reporter gene. J. Biotechnol. 108, 17–30. 15. Chudakov, D. M., Lukyanov, S., and Lukyanov, K. A. (2005) Fluorescent proteins as a toolkit for in vivo imaging. Trends Biotechnol. 23, 605–613. 16. Miyawaki, A., Nagai, T., and Mizunom H. (2005) Engineering fluorescent proteins. Adv. Biochem. Eng. Biotechnol. 95, 1–15. 17. Bonnerot, C., Rocancourt, D., Briand, P., Grimber, G., and Nicolas, J. F. (1987) A beta-galactosidase hybrid protein targeted to nuclei as a marker for developmental studies. Proc. Natl. Acad. Sci. USA 84, 6795–6799. 18. Sorg, G. and Stamminger, T. (1999) Mapping of nuclear localization signals by simultaneous fusion to green fluorescent protein and to beta-galactosidase. Biotechniques 26, 858–862. 19. Kanda, T., Sullivan, K. F., and Wahl, G. M. (1998) Histone-GFP fusion protein enables sensitive analysis of chromosome dynamics in living mammalian cells. Curr. Biol. 8, 377–385. 20. Miyawaki, A., Nagai, T., and Mizuno, H. (2005) Engineering fluorescent proteins. Adv. Biochem. Eng. Biotechnol. 95, 1–15. 21. van Roessel, P. and Brand, A. H. (2002) Imaging into the future: visualizing gene expression and protein interactions with fluorescent proteins. Nat. Cell. Biol. 4, E15–20. 22. Serganova, I. and Blasberg, R. (2005) Reporter gene imaging: potential impact on therapy. Nucl. Med. Biol. 32, 763–780. 23. Herschman, H. R. (2004) PET reporter genes for noninvasive imaging of gene therapy, cell tracking and transgenic analysis. Crit. Rev. Oncol. Hematol. 51, 191–204.

Contents Preface ......................................................................................................... 11v Contributors .................................................................................................. .1ix 1 Methodologies for Staining and Visualisation of β-Galactosidase in Mouse Embryos and Tissues Siobhan Loughna and Deborah Henderson ...................................... 111 2 Immunohistochemical Detection of β-Galactosidase or Green Fluorescent Protein on Tissue Sections Philip A. Seymour and Maike Sander ............................................... 113 3 Detection of Reporter Gene Expression in Murine Airways Maria Limberis, Peter Bell, and James M. Wilson ............................ 125 4 Three-Dimensional Analysis of Molecular Signals with Episcopic Imaging Techniques Wolfgang J. Weninger and Timothy J. Mohun .................................. 135 5 Fluorescent Proteins for Cell Biology George H. Patterson ......................................................................... 147 6 Detection of GFP During Nervous System Development in Drosophila melanogaster Karin Edoff, James S. Dods, and Andrea H. Brand ............................. 81 7 Autofluorescent Proteins for Flow Cytometry Charles G. Bailey and John E. J. Rasko ............................................... 99 8 Fluorescent Protein Reporter Systems for Single-Cell Measurements Steven K. Dower, Eva E. Qwarnstrom, and Endre Kiss-Toth ............ 111 9 Subcellular Imaging of Cancer Cells in Live Mice Robert M. Hoffman .......................................................................... 121 10 Noninvasive Imaging of Molecular Events with Bioluminescent Reporter Genes in Living Subjects Pritha Ray and Sanjiv Sam Gambhir ................................................ 131 11 Green Fluorescent Protein as a Tracer in Chimeric Tissues: The Power of Vapor Fixation Harald Jockusch and Daniel Eberhard ............................................. 145 Index ........................................................................................................... 155

viii

Contributors DONALD S. ANSON • Gene Technology Unit, Department of Genetic Medicine, Children, Youth and Women’s Health Service, Adelaide, Australia CHARLES G. BAILEY • Gene and Stem Cell Therapy Program, Centenary Institute of Cancer Medicine and Cell Biology, Australia PETER BELL • Gene Therapy Program, Department of Pathology and Laboratory Medicine, University of Pennsylvania, USA ANDREA H. BRAND • Wellcome Trust/Cancer Research UK Gurdon Institute and Department of Physiology, Development and Neuroscience, University of Cambridge, UK JAMES S. DODS • Wellcome Trust/Cancer Research UK Gurdon Institute and Department of Physiology, Development and Neuroscience, University of Cambridge, UK STEVEN K. DOWER • Section of Functional Genomics, School of Medicine and Biomedical Sciences, University of Sheffield, UK DANIEL EBERHARD • Department of Biology and Biochemistry, Centre for Regenerative Medicine, Bath University, UK KARIN EDOFF • Wellcome Trust/Cancer Research UK Gurdon Institute and Department of Physiology, Development and Neuroscience, University of Cambridge, UK SANJIV SAM GAMBHIR • Molecular Imaging Program at Stanford (MIPS), Departments of Radiology and Bioengineering, Bio-X Program, Stanford University, USA ROBERT M. HOFFMAN • AntiCancer, Inc., San Diego, and Department of Surgery, University of California at San Diego, USA HARALD JOCKUSCH • Developmental Biology and Molecular Pathology, Bielefeld University, Germany DR DEBORAH HENDERSON • Institute of Human Genetics, University of Newcastle Upon Tyne, UK ENDRE KISS-TOTH • Cardiovascular Research Unit, School of Medicine and Biomedical Sciences, University of Sheffield, UK MARIA LIMBERIS • Gene Therapy Program, Department of Pathology and Laboratory Medicine, University of Pennsylvania, USA DR SIOBHAN LOUGHNA • School of Biomedical Sciences, University of Nottingham, UK ix

x

Contributors

TIMOTHY J MOHUN • Developmental Biology Division, MRC National Institute for Medical Research, Mill Hill, UK GEORGE H. PATTERSON • Cell Biology and Metabolism Branch, National Institute of Child Health and Human Development, National Institutes of Health, USA EVA E. QWARNSTROM • Section of Cell Biology, School of Medicine and Biomedical Sciences, University of Sheffield, UK JOHN E.J. RASKO • Gene and Stem Cell Therapy Program, Centenary Institute of Cancer Medicine and Cell Biology, Australia and Cell and Molecular Therapies, Sydney Cancer Centre, Royal Prince Alfred Hospital, Australia PRITHA RAY • Molecular Imaging Program at Stanford (MIPS), Departments of Radiology and Bioengineering, Bio-X Program, Stanford University, USA MAIKE SANDER • Department of Developmental and Cell Biology, University of California, Irvine, USA PHILIP A. SEYMOUR • Department of Developmental and Cell Biology, University of California, Irvine, USA WOLFGANG J WENINGER • Integrative Morphology Group, Medical University of Vienna, Austria JAMES M. WILSON • Gene Therapy Program, Department of Pathology and Laboratory Medicine, University of Pennsylvania, USA

Detection of β-Galactosidase in Mouse Embryos

1

1 Methodologies for Staining and Visualisation of β -Galactosidase in Mouse Embryos and Tissues Siobhan Loughna and Deborah Henderson Summary This chapter provides information on β-galactosidase staining of whole mouse embryos, organs, tissue sections, and cultured cells, as well as double staining with horseradish peroxidase and use as a tool for genotyping. Using these protocols, localization of β-galactosidase can be visualized throughout development and in adult tissues. β-Galactosidase staining may be used purely as a marker of gene expression and also as a tracer in cell lineage studies. Key Words: β-Galactosidase; embryo; tissues; cultured cells; immunohistochemistry.

1. Introduction The E. coli lacZ gene is the first gene of the lac operon. The LacZ gene, encoding β-galactosidase (β-gal), is the classical histochemical reporter gene (1). β-Gal is a stable enzyme that may be expressed in cultured cells, in the fruit fly Drosophila, and in transgenic animals, with no apparent side effects. β-Gal hydrolyzes (2) the disaccharide lactose into two monosaccharide sugar groups, glucose and galactose. Enzyme activity can be detected using a variety of chromogenic substrates, such as the indole derivative 5-bromo-4-chloro-3indolyl-β-D-galactoside (X-gal). X-gal is cleaved by β-gal into an insoluble, stable, bright blue precipitate (3,4). The dimerization and oxidation reactions require transfer of an electron by the electron acceptors provided by the ferric and ferrous ions included in the LacZ staining solution (5,6). Protocols have arisen that allow gene expression to be monitored by linking up the promoter of a gene of interest to a marker gene, typically LacZ. An alternative strategy is to fuse a marker gene, in frame, to the coding sequence of a gene of interest. Both of these strategies allow the expression of a gene to be readily monitored, with the normal temporal pattern of expression maintained. From: Methods in Molecular Biology, vol. 411: Reporter Genes: A Practical Guide Edited by: D. Anson © Humana Press Inc., Totowa, NJ

1

2

Loughna and Henderson

This means that expression of LacZ will decrease when the gene would normally be downregulated, which allows a faithful pattern of gene expression to be obtained. However, for lineage tracing of specific cell types, another strategy is adopted. The ideal marker for lineage tracing of specific cell types in a developing embryo would completely and exclusively mark that cell and all daughter cells, with the minimum of background, while maintaining expression throughout the lifetime of the cell. This has been achieved over recent years in mammalian embryos by utilizing a two-component system, based on Cre/lox recombination, to mark specific cell lineages indelibly. The first component is a transgene expressing the enzyme Cre recombinase under the control of a cell type- or tissue-specific promoter. This allows Cre recombinase to be expressed only in those cells in which the promoter would normally drive expression. The second component is a conditional reporter gene that expresses a histological marker only upon Cre-mediated recombination. The most commonly used of these is the R26R gene, which expresses β-gal from the ROSA26 locus only upon Cre-mediated recombination (7). The R26R system has been shown to be ideal for these purposes, as ROSA26 is ubiquitously and uniformly expressed at all developmental and postnatal times, with no apparent sensitivity to genetic or environmental manipulation. In the absence of Cre recombinase, the ROSA26 gene is not expressed, but following recombination, a functional β-gal protein is produced. In this system, once the recombination event has occurred, it will be transferred to progeny of the original cells, even though the initial transgene is no longer expressed. Figure 1 shows LacZ staining when the ROSA26 β-gal gene is activated by expression of Cre from the Wnt1 promoter, which is specifically active in neural crest cells and their progeny (8,9). This chapter describes protocols for detecting β-gal expression in wholemount embryos, organs, or tissues, cryostat tissue sections, and cultured cells and for double staining for β-gal and horseradish peroxidase. Owing to endogenous β-gal activity, background staining can be a problem. However, this chapter will provide some useful tips on reducing background while maintaining high levels of positive staining. All protocols follow the same principles in terms of the initial fixation, washing, and staining, as outlined in the flowchart (Fig. 2). 2. Materials 2.1. Fixation 1. Phosphate-buffered saline (PBS), calcium free (Oxoid). Make up in deionized water and autoclave before use (see Note 1). 2. 25% glutaraldehyde (Sigma). May be freeze/thawed, but best to work from small aliquots. Caution: toxic.

Detection of β-Galactosidase in Mouse Embryos

3

Fig. 1. β-Galactosidase staining of an embryonic day 10.5 Wnt1CreRosa26R transgenic mouse embryo. Whole mount embryo (A) and tissue section (B) stained for β-galactosidase expression (the dotted line in A denotes the approximate line of cut for tissue section B). (A) LacZ staining can be seen in the craniofacial region, branchial arches, dorsal root ganglia (drg), and less intense staining can be seen in the outflow tract of the heart (oft). (B) Sectioning the same embryo confirms LacZ staining in the dorsal part of the neural tube (nt), surrounding the branchial arch arteries, and in the outflow tract of the heart (oft). a, aortic sac; l, limb; acv, anterior cardinal vein. See accompanying CD for color version.

3. Paraformaldehyde (PFA) comes as a powder (Sigma). We generally make up fresh each time, although it can be made up in aliquots and stored at −20°C. Avoid freeze/ thawing. Caution: toxic. a. Make up 4% PFA in water, and dissolve by adding 1 to 2 drops of 5 M NaOH and heating at about 60°C with gentle stirring. b. Add 10X PBS to make a 1X PBS solution. c. Cool and check pH with pH paper—must be pH 7.0. 4. LacZ fix: 1% PFA, 0.2% glutaraldehyde, 2 mM MgCl2, 5 mM EGTA (pH 8.0), 0.2% Nonidet-P40 (NP40) (stock: 10%, made in H2O stored at 4°C. We make fresh every few weeks). Make fresh before use. To make up the 0.5 M EGTA stock solution, add the required amount of EGTA to water, and dissolve by adding small amounts of 5 M NaOH and mixing until the EGTA has fully dissolved (see Note 2).

2.2. Washing and Staining 1. LacZ wash solution in 1X PBS: 2 mM MgCl2, 0.01% sodium deoxycholate (stock: 1%, made in H2O stored at 4°C), 0.02% NP40.

4

Loughna and Henderson

Fig. 2. Flowchart of the general procedure for LacZ staining.

2. LacZ stock substrate: 10% X-gal (Sigma or Melford Laboratories) dissolved in dimethylformamide. Caution: toxic. Store in the dark at 4°C. Solid X-gal should be stored at −20°C. Before opening bottle to remove any X-gal, allow to warm up to room temperature first to prevent moisture uptake by the solid, which will lead to gradual deterioration. 3. LacZ staining solution: a working staining solution is 5 mM potassium ferrocyanide, 5 mM potassium ferricyanide (Caution: these two are toxic), 2 mM MgCl2, 0.01% sodium deoxycholate (stock: 1%, made in H2O stored at 4°C), 0.02% NP40 (stock: 10%, made in H2O stored at 4°C), 0.1% X-gal (from 10% stock solution). Store in the dark at 4°C; make fresh before use (see Note 3).

2.3. Tissue Embedding and Coating Slides 1. Tissue-Tek OCT compound (Sakura, available from Raymond Lamb). 2. TESPA (or 3-aminopropyltriethoxysilane [APES]; Sigma) coated slides are used (can be performed in house (see Note 4) or bought commercially (Marienfeld). 3. Paraffin wax (Fisher Scientific), and Histoclear (National Diagnostics).

2.4. β -Galactosidase and Horseradish Peroxidase Double Staining 1. Tris-buffered saline (TBS): 0.1 M Tris-HCl, pH 7.5, 0.15 M NaCl, in deionized water. Autoclave before use. 2. TBS-Tx: TBS with 0.5% Triton X-100 added. Do not autoclave after the addition of Triton X-100.

Detection of β-Galactosidase in Mouse Embryos

5

Table 1 Suggested Fixation and Staining Times for Embryonic and Fetal Mouse Specimens Agea E9.5 E10.5 E11.5 E12.5 E13.5–E14.5 E15.5–E18.5 aE

Whole or dissected embryo

Fixation time b (min)

Staining time

Whole Whole Whole Dissected Dissected Dissected

5 5 15 15 20 30–90

6h 6h Overnight Overnight Overnight Overnight

is embryonic day, where E0.5 is the morning of finding the copulation plug. time in wash buffer is equal to the time in fixing solution for each age.

b Total

3. Fetal calf serum (FCS): heat inactivate at 56°C for 30 min and then store as aliquots at −20°C. 4. AB Complex conjugated to horseradish peroxidase (HRP) (Dako) and made up according to the manufacturer’s instructions. 5. Diaminobenzidene (DAB) tablets (obtained from Sigma): Stored at −20°C. Caution: toxic. Make up according to the manufacturer’s instructions and filter through a 0.45-µm filter before use. 6. 100 mM Sodium citrate buffer, pH 6.0, is a 10X stock solution. Adjust the pH with NaOH. Autoclave before use. 7. Histoclear (National Diagnostics).

3. Methods 3.1. β -Galactosidase Staining of Whole Embryos (or Tissue Pieces/Whole Organs) β-Gal staining can be performed on whole embryos up to about mouse embryonic day 11.5, after which it is advisable to dissect the embryo into smaller pieces, isolate whole organs, or perform cryostat sectioning (see Subheading 3.2.) owing to decreased penetration of fix and staining reagents in large specimens. All steps take place at room temperature unless otherwise stated. 1. Wash embryos in PBS three times at room temperature to remove any media or serum. At this stage, larger embryos may need to be dissected to allow penetration of the staining solutions (see Note 5). 2. Fix embryos in cold LacZ fixative at 4°C for 5 to 90 min (see Note 2 and Table 1). 3. Wash embryos in LacZ wash solution three times for 20 min each at room temperature. 4. Stain the embryos in the LacZ staining solution for the required period at 37°C. This stage will need optimizing, but see Table 1 for guidelines. By the end of the

6

Loughna and Henderson

Table 2 Suggested Times for Fixation and Embedding of LacZ-Stained Embryos Embryo stage Treatment

E9.5

4% PFA postfixation O/N 1X PBS × 3 10 min Distilled H2O × 2 10 min 50% Ethanol 30 min 70% Ethanol × 2 30 min 95% Ethanol 30 min 100% Ethanol × 2 30 min Histoclear × 3 5 min Histoclear/wax 10 min Wax × 3 5–10 min

5. 6. 7.

8.

E10.5

E11.5

E12.5 E13.5–14.5 E15.5–18.5

O/N 10 min 10 min 30 min 30 min 30 min 30 min 10 min 20 min 20 min

O/N 2d 10 min 20 min 10 min 20 min 60 min 2h 60 min 2h 60 min 2h 60 min 2 h-O/N 10 min 15 min 20 min 30 min 20 min 30 min

2d 20 min 20 min 2h 2h 2h 2 h-O/N 15 min 60 min 60 min

3d 30 min 30 min 2h 2h 2h 2 h-O/N 20 min 90 min 90+ min

staining period the embryos/tissues should be stained dark blue at sites of β-gal expression, but with no staining elsewhere (i.e., minimal background; Fig. 1). Negative controls should always be included (see Note 6). Wash the embryos three times in PBS at room temperature. Postfix specimens in 4% PFA overnight or longer (Table 2). Embryos can be analyzed as whole mounts (Fig. 1A) or processed and embedded as for normal histology, although solvents can partially dissolve the LacZ precipitate. Therefore, stained tissues should be paraffin embedded and sectioned using minimum necessary times in solvents (Table 2). Sections (8–10 µm) should be cut for optimum visualization of β-gal staining (Fig. 1B). Tissue sections can then be counterstained (see Note 7).

3.2. β -Galactosidase Staining of Tissue Sections Perfusion fixation and cryostat sectioning is optimal for large pieces of tissue, as penetration of fix and staining reagents will be reduced in large specimens and paraffin embedding destroys β-gal activity. All steps take place at room temperature unless otherwise stated. 1. Wash embryos in PBS (see Note 1) three times at room temperature to remove any media or serum. 2. Fix embryos in cold LacZ fixative (see Note 2) at 4°C for 5 to 90 min (Table 1). 3. Wash embryos in PBS, three times for 20 min each at room temperature. 4. Equilibrate in 15% sucrose/PBS for 10 min to 1 h at 4°C. 5. Equilibrate in 30% sucrose/PBS until the tissue drops at 4°C (may take several hours). 6. Equilibrate further in Tissue-Tek (OCT) for 1 h at 4°C. 7. Embed in OCT (see Note 8). 8. Store blocks at −80°C in an air-tight container.

Detection of β-Galactosidase in Mouse Embryos

7

9. When ready to section, place block in cryostat chamber for approx 20 min. 10. Section cryostat blocks at 5 to 10 µm onto poly-lysine- or TESPA-coated slides (see Note 4). 11. Air-dry slides. 12. Store slides at −80°C in an air-tight slide box with desiccant. 13. To stain sections, take slide box out of freezer, and equilibrate to room temperature before opening box. Unused slides may be placed back in the freezer in an air-tight slide box with desiccant, although avoid freeze-thawing if possible. 14. Fix sections to slides in 0.2% glutaraldehyde in PBS for 5 to 10 min at 4°C. 15. Wash slides in LacZ wash solution, three times, for 5 min each at room temperature. 16. Stain the slides in the LacZ staining solution for the required period at the required temperature in the dark. This stage will need optimizing. Control (LacZ-negative) tissue should be used each time to control for endogenous β-gal (see Note 6). 17. Wash the slides in PBS, three times at room temperature. 18. Postfix in 4% PFA for 10 min, and wash with 1X PBS. Counterstaining may now be performed (see Note 7).

3.3. β -Galactosidase and Horseradish Peroxidase Double Staining This protocol can be used to carry out immunohistochemical staining for specific antigens on embryos that have already been stained for β-gal and then have been postfixed, embedded in paraffin wax, and sectioned onto slides at 8 to 10 µm (as in Subheading 3.1.). The indigo product of X-gal absorbs in the wavelengths emitted by the standard fluorescent-conjugated antibodies and is so dark that it can obscure the products from either horseradish peroxidase or alkaline phosphatase-conjugated antibodies. However, by carefully controlling the reaction time in LacZ staining solution so that only a small amount of indigo is produced (10), by having β-gal localized to the nucleus when the cellular antigen is nonnuclear (11), or by using antibodies to detect both β-gal and the cellular antigen, one can overcome these problems. All steps take place at room temperature unless otherwise stated. 1. 2. 3. 4. 5. 6. 7. 8. 9.

Place the slides in Histoclear for 5 min, twice, to dewax the sections. Remove the Histoclear with two rinses in 100% ethanol for 2 min each. Dehydrate through an alcohol series (90%, 70%, 50% ethanol) for 2 min each. Rinse in TBS for 5 min. Inactivate endogenous peroxidase in the tissue with 3% H2O2 in deionized water, for 5 min. Rinse in deionized water. If antigen retrieval is required for the unmasking of specific epitopes (see Note 9). Wash slides three times in TBS-Tx for 5 min each time. Block nonspecific antigens in the tissue sections in 10% FCS in TBS-Tx for 30 min. Place slides in a humid chamber. Add the primary antibody diluted in TBS-Tx containing 2% FCS. Add 80 to 100 µL per slide (to cover all the tissue sections), and cover with a Parafilm cover slip. Leave overnight at 4°C on a flat surface.

8

Loughna and Henderson

10. Equilibrate the slides back to room temperature for 15 min, and then wash three times in TBS-Tx for 5 min each. 11. Dilute the appropriate secondary antibody in TBS-Tx containing 2% FCS. Add 80 to 100 µL per slide, cover with a Parafilm cover slip, and leave at room temperature for 1 to 2 h. 12. Make AB Complex 30 min before use (according to the manufacturer’s instructions) and store at room temperature. 13. Wash slides three times in TBS-Tx for 5 min each. 14. Add AB Complex and leave for 30 min. Do not cover slides. 15. Rinse slides three times in TBS-Tx for 5 min each. Make up DAB solution according to the manufacturer’s instructions. 16. Add DAB solution and leave for up to 30 min until turns brown (see Note 10). 17. Stop reaction by rinsing well in PBS. 18. Counterstain as necessary. Choosing the counterstain can be problematic, as after LacZ and DAB staining, red in addition to blue stains can be confusing. We have had the best success with a light aqueous eosin stain in which the pale pink contrasts with the brick red/brown of the DAB staining (see Note 7).

3.4. β -Galactosidase Staining of Cultured Cells This protocol can be used on cell lines expressing LacZ or on primary cell lines isolated from LacZ-expressing tissues. For best results, cells should be grown on coated circular cover slips in 24-well plates. The cover slips may be coated using a variety of substances, such as fibronectin and collagen; the choice is cell type dependent. All steps are carried out in the 24-well plates using the same solutions as used for staining of cryosections and whole embryos. The cover slips are removed from the plates for photography. All steps take place at room temperature unless otherwise stated. 1. 2. 3. 4.

Remove the growth media from the cells and wash well with PBS at 4°C. Fix the cultured cells with in LacZ fixing solution for 2 min at 4°C. Wash the cells three times with wash buffer, for 2 min each. Replace the wash buffer with staining solution and wrap the plate in aluminum foil. Incubate at 37°C. The time of incubation varies greatly and should be optimized (see Note 6). 5. Once LacZ staining has developed, remove the staining solution, wash the cells twice with wash buffer, and carefully remove the cover slips from the 24-well plate using fine forceps or a hooked needle. 6. Mount the cover slip and photograph the staining using phase contrast microscopy.

3.5. Genotyping by β -Galactosidase Staining This is a quick and easy protocol for genotyping embryos that express LacZ. However, care must be taken to stain a piece of embryonic or adult tissue (for example, the yolk sac, an ear clip, or a tail tip) that expresses the LacZ product.

Detection of β-Galactosidase in Mouse Embryos

9

This can be a problem when tissue-specific promoters are used to drive LacZ expression. 1. 2. 3. 4. 5.

Place tissue piece in PBS in a 96-well plate. Fix for 15 min (0.2% glutaraldehyde in PBS) at room temperature. Wash three times in PBS for 5 min each. Stain with LacZ staining solution at 37°C in the dark (as in Subheading 3.1.). After 5 to 30 min, staining should be visible.

4. Notes 1. PBS must be calcium-free, pH 7.5. PBS tablets can be used. pH 7.5 is optimal, as endogenous β-gal (mammalian) is optimal at acidic pH, whereas the E. coli β-gal is optimal at neutral to slightly alkaline pH. 2. The authors have used two different types of fixatives. As well as the fixative described in Subheading 2.1., the same ingredients can also be made up in phosphate buffer to aid stability. Phosphate buffer: (0.1 M phosphate buffer, pH 7.5: 115 mL 0.1 M NaH2PO4 + 385 mL 0.1 M Na2HPO4). A mixture of glutaraldehyde/ PFA (Caution: both toxic) is often used for fixation, or 0.2% glutaraldehyde alone. Fixation with glutaraldehyde gives superior staining compared with PFA fixation alone. However, glutaraldehyde may also preserve endogenous enzyme activity, leading to nonspecific staining. 3. Although a stock solution can be made of potassium ferrocyanide and potassium ferricyanide and kept in the dark at 4°C, these solutions do go off with time (turn grayish yellow and grayish orange, respectively). The concentration used can vary, but usually 5 to 10 mM is used. Caution: these salts are also harmful if inhaled. The ferricyanide and ferrocyanide can form a blue precipitate (Prussian blue) upon reaction with free ferric ion. Therefore, do not use metal forceps to manipulate the tissue while it is in the LacZ staining solution. If a purple precipitate is preferred, nitroblue tetrazolium (NBT) salt can be added to the X-gal reaction in place of iron. This may be a faster and more sensitive reaction. A stock solution of NBT can be prepared by adding 50 mg NBT to 1 mL of 70% dimethylformamide and stored at −20°C. The final working concentrations of NBT are 0.25 to 1.0 mg/mL. Phenazine methosulfate (PMS) can also be added in conjunction with NBT to increase the reaction rate further. As PMS is very unstable, prepare a 100X stock of 2 mg/mL in H2O and use immediately. However, the authors have no personal experience with this variation to the X-gal reaction mix. 4. To coat slides with TESPA (also called APES), rack the slides and place overnight in 2% Decon (a detergent) to clean the slides. Rinse under hot tap water for 1 h, and then briefly wash in distilled water. Dry the slides at 60°C (overnight if convenient). Next day, place the slides in troughs for 5 min each with 2% TESPA in 100% EtOH, followed by two washes of 100% EtOH and one wash of 100% acetone. Once the slides are air-dried, they are ready to use. 5. Larger specimens may need dissection to allow full penetration of staining solutions. This may involve removal of tissues that are not of interest, or isolation of organs of interest. In older fetuses, removal of the skin will aid penetration.

10

Loughna and Henderson

6. The LacZ staining step can vary depending on the degree of expression. With embryos, low levels of expression may need an overnight step at 37°C. However, strong levels of expression may be seen in a few hours. Background staining can also be reduced if staining is performed at lower temperatures (20–30°C). With slides, staining can usually be seen after 3 to 6 h. The gut and kidney usually have particularly high levels of endogenous β-gal. However, to a degree all tissues have endogenous, lysosomal β-gal. The optimal pH for lysosomal β-gal is very low (acidic), and thus it is not very active at pH 7.5. However, some tissues also have a cytosolic form of β-gal, which may be active enough to give confounding results. Background may be reduced by varying the fixative type, length of time in fixative, pH of the buffer, and amount of time in the staining solution (12). If, after varying these conditions, background is still a problem, the LacZ enzyme can be detected by immunohistochemistry using commercially available antibodies. 7. Avoid counterstains that stain tissue blue, such as toluidine blue. Hematoxylin is also best avoided, as some tissues stain blueish. Counterstaining can be performed with stains such as 1% aqueous eosin (from seconds to minutes, depending on intensity required), which stains practically all cytoplasmic and intercellular substances a reddish pink color but does not stain the nucleus. Staining times may also vary depending on the tissue density and type. 8. There are several methods for embedding tissue samples in OCT. For larger specimens for which correct orientation is crucial, we usually place the sample in OCT in a plastic mold, orient the tissue/embryo with needles, and then freeze on dry ice (with the bucket lid on). For small samples or when orientation is not important, melting isopentane can be used. This involves a shallow dish with isopentane in the bottom (enough to surround a mold, but not to submerge). Place level on dry ice. Allow to freeze, and then remove from dry ice. As it starts to melt, place mold with tissue in OCT in isopentane, until OCT freezes. 9. For antigen retrieval: place slides into a plastic rack, place in a large beaker, and cover with 10 mM sodium citrate buffer (ix), pH 6.0. Cover the beaker with cling film and pierce for vent. Microwave on high for 10 min (check level of buffer during this step and top up if required). Allow to cool for 5 min and then wash in cool tap water for 10 min. Proceed to step 7 of Subheading 3.3. Other methods for antigen retrieval, such as trypsin treatment, may also be possible, although these have not been verified by the authors. 10. DAB can be precipitated very quickly in some cases (less than 1 min) but in other cases may take as long as 20 min. There is little advantage to leaving slides longer than this, as the enzymatic reaction is almost complete after this period. Staining of larger specimens can be seen with the naked eye. For smaller specimens, or when staining is expected in small areas or isolated cells, develop under a microscope.

References 1. Beckwith, J. R. (2005) Lac: The genetic system, in The Operon (Miller, J. H. and Reznikoff, W. S., eds.), Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.

Detection of β-Galactosidase in Mouse Embryos

11

2. Loughna, S., Bennett, P., Gau, G., Nicolaides, K., Blunt, S., and Moore, G. (1993) Overexpression of esterase D in kidney from trisomy 13 fetuses. Am. J. Hum. Genet. 53, 810–816. 3. Horwitz, J. P., Chua, J., Curby, R. J., et al. (1964) Substrates for cytochemical demonstration of enzyme activity. I. Some substituted 3-indolyl-beta-D-glycopyranosides. J. Med. Chem. 53, 574–575. 4. Davies, J. and Jacob, F. (1968) Genetic mapping of the regulator and operator genes of the lac operon. J. Mol. Biol. 36, 413–417. 5. Cotson, S. and Holt, S. J. (1958) Studies in enzyme cytochemistry. IV. Kinetics of aerial oxidation of indoxyl and some of its halogen derivatives. Proc. R. Soc. Lond B Biol. Sci. 148, 506–519. 6. Lojda, Z. (1970) Indigogenic methods for glycosidases. I. An improved method for beta-D-glucosidase and its application to localization studies on intestinal and renal enzymes. Histochemie 22, 347–361. 7. Soriano, P. (1999) Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat. Genet. 21, 70–71. 8. Danielian, P. S., Muccino, D., Rowitch, D. H., Michael, S. K., and McMahon, A. P. (1998) Modification of gene activity in mouse embryos in utero by a tamoxifeninducible form of Cre recombinase. Curr. Biol. 8, 1323–1326. 9. Jiang, X., Rowitch, D. H., Soriano, P., McMahon, A. P., and Sucov, H. M. (2000) Fate of the mammalian cardiac neural crest. Development 127, 1607–1616. 10. Vaysse, P. J. and Goldman, J. E. (1990) A clonal analysis of glial lineages in neonatal forebrain development in vitro. Neuron 5, 227–235. 11. Bonnerot, C., Rocancourt, D., Briand, P., Grimber, G., and Nicolas, J. F. (1987) A beta-galactosidase hybrid protein targeted to nuclei as a marker for developmental studies. Proc. Natl. Acad. Sci. USA 84, 6795–6799. 12. Rosenberg, W. S., Breakefield, X. O., deAntonio, C., and Isacson, O. (1992) Authentic and artifactual detection of the E. coli lacZ gene product in the rat brain by histochemical methods. Brain Res. Mol. Brain Res. 16, 311–315.

Immunofluorescent Staining Method for GFP and β-Gal

13

2 Immunohistochemical Detection of β -Galactosidase or Green Fluorescent Protein on Tissue Sections Philip A. Seymour and Maike Sander Summary With the recent advances in mouse genetics, it is now possible to mark specific cell types genetically in vivo and to study the fate of cells during development and adulthood. Cells are labeled and followed in vivo through the stable expression of reporter genes in particular cell types. The two most commonly used reporter genes are LacZ, which encodes the enzyme β-galactosidase (β-gal), and green fluorescent protein (GFP). β-Gal expression can be detected enzymatically, using 5-bromo-4-chloro-3-indolyl-β-Dgalactopyranoside (X-gal) as a substrate, and GFP can be directly visualized by fluorescence microscopy. However, with single detection of β-gal or GFP, it is often impossible to determine whether expression of the reporter protein is restricted to a particular cell type. To ascertain the identity of individual cells within a multicellular tissue, β-gal or GFP proteins must be visualized in conjunction with additional cellular markers. For such experiments, specific antibodies raised against β-gal or GFP can be used in coimmunofluorescence analyses. Such double-staining analyses on tissue sections are a powerful tool to study transgene expression or to trace cells in multicellular tissues. Key Words: Reporter gene; mouse; green fluorescent protein; LacZ; β-galactosidase; immunofluorescence; immunohistochemistry; fluorochrome; antibody.

1. Introduction Through the insertion of reporter genes into the genome, individual cells can be visualized in a given tissue. In mice, reporter genes are commonly used to monitor the expression of transgenes or to study the expression pattern of individual genes during development and in adulthood. In recent years, binary genetic systems, which are largely based on Cre/loxP-mediated recombination, have been developed to mark and trace cells genetically in vivo by activating expression of a reporter gene (1–5). Reporter genes that are commonly used to label cells genetically include the LacZ gene, alkaline phosphatase, luciferase, and From: Methods in Molecular Biology, vol. 411: Reporter Genes: A Practical Guide Edited by: D. Anson © Humana Press Inc., Totowa, NJ

13

14

Seymour and Sander

fluorochromes (6). The optimal choice of reporter gene depends on the intended application. The LacZ gene, which encodes β-galactosidase (β-gal), allows for detection of expressing cells through a simple tissue stain with 5-bromo-4chloro-3-indolyl-β-D-galactopyranoside (X-gal) as a substrate. X-Gal staining is highly sensitive and robust and can be performed either on whole-mount specimens or on tissue sections (7). The unique feature of fluorochromes is their ability to fluoresce when excited under a fluorescence microscope. Because of this feature, fluorochromes are widely used to visualize gene expression in living cells (8). A limitation to the use of fluorochromes in direct cell imaging is that their fluorescence is diminished by tissue fixation. To allow for the visualization of fluorochromes on fixed tissue, specific antibodies against green fluorescent protein (GFP) have been developed that can be used in indirect immunofluorescence or immunoperoxidase staining on tissue sections. A limitation of immunoperoxidase staining is that it does not allow for simultaneous detection of two antigens within one cell; therefore colocalization studies are precluded. Since staining for βgal or GFP in conjunction with additional antigens is commonly used to determine which cell type expresses the reporter gene, we have focused this chapter on immunofluorescence staining. We describe a double-immunofluorescence protocol that allows for simultaneous detection of β-gal or GFP with additional cellular markers on tissue sections. Additional information about the described protocols can be found in Current Protocols in Molecular Biology (9). 2. Materials 2.1. Fixation of Tissue 1. 1X phosphate-buffered saline (PBS): 137 mM NaCl, 10 mM Na2HPO4, 1.8 mM KH2PO4, 2.7 mM KCl. Adjust to pH 7.4 with HCl. Store at room temperature. Can be prepared as a 10X stock. 2. 4% Paraformaldehyde in PBS: a. Dissolve paraformaldehyde (Fisher) in 1X PBS in a Pyrex container with a stir bar (4 g to 100 mL for 4% solution). b. Add a few drops of NaOH and heat in a hood (keep bottle cap loose) at 60°C to dissolve. c. Cool to room temperature on ice and adjust pH to 7.4. d. Filter to remove any undissolved powder. e. Make fresh prior to use; aliquots can be stored at −20°C.

2.2. Preparation of Cryoprotected Tissue Sections 1. Sucrose (Fisher) in PBS: prepare 10%, 15%, and 30% (w/v) solution in 1X PBS. Make fresh prior to use. Can be stored at 4°C for up to 1 d. 2. Tissue-Tek® O.C.T. (Optimal Cutting Temperature) Compound (Sakura Finetek, Torrance, CA).

Immunofluorescent Staining Method for GFP and β-Gal 3. 4. 5. 6.

15

Peel-A-Way® embedding molds (Polysciences, Warrington, PA). SuperFrost/Plus glass slides (Microm, Walldorf, Germany). Humidity Sponge™ silica desiccant sachets (Fisher). Polypropylene microscope slide box (Fisher).

2.3. Immunofluorescence Detection of GFP or β -Gal on Tissue Sections 1. Wheaton glass 20-slide staining dish with removable rack (Fisher). 2. Shandon Coverplate™ and Shandon Sequenza® Slide Rack (Thermo Electron, Waltham, MA). 3. Phosphate-buffered saline/Tween (PBST): prepare a 0.1% (v/v) solution of Tween20 (Sigma) in 1X PBS. Store at room temperature. 4. Blocking solution: 1% (v/v) normal goat serum (NGS) in PBST. Store 1 mL stock aliquots at −20°C; 1% NGS in PBST can be stored at 4°C for up to 1 wk (see Note 1). 5. Primary antibodies: rabbit anti-β-gal (MP Biomedicals, Irvine, CA; formerly ICN, cat. no. 55976, lot 03660) and rabbit anti-GFP (Invitrogen [Molecular Probes, cat. no. A6455], Carlsbad, CA) (see Note 2). Store small aliquots at −80°C. Once thawed, antibodies should be stored at 4°C. 6. Secondary antibody: cyanine (Cy3)-conjugated goat anti-rabbit IgG (H+L) (Jackson ImmunoResearch, West Grove, PA) (see Note 3). Store small (5–10 µL) aliquots in a 1:1 dilution with glycerol at −80°C. Once thawed, antibodies should be stored at 4°C. Avoid exposure of fluorophore-conjugated antibodies to direct overhead light, as this will induce photobleaching. 7. Nuclear stain: 300 nM DAPI (4.6-diamidino-2-phenylindole) in 1X PBS. 8. Mounting medium: Vectashield® (Vector, Burlingame, CA) or Aqua Poly/Mount (Polysciences). 9. Fisherbrand microscope coverslips (Fisher). 10. Nail polish. 11. Microscope slide folder (Fisher).

3. Methods Both β-gal and GFP can be readily detected by indirect immunofluorescence on frozen tissue sections. In our hands, detection on paraffin-embedded tissue is far less reliable. Unless directed to the nucleus through the addition of a nuclear localization signal, β-gal and GFP are normally localized to the cytoplasm. When localized to the cytoplasm, immunofluorescence detection of β-gal or GFP on frozen sections does not require additional procedures to unmask the epitopes. However, for nuclear detection of β-gal or GFP, epitope unmasking procedures often improve the intensity of the signal (see Note 4; 10–12). It should be recognized that any antigen unmasking procedure may introduce artifactual falsepositive staining. In all staining experiments, a negative control slide should be included, on which preimmune serum is applied instead of the primary antibody. If preimmune serum is not available, blocking solution can be used instead. It is also advisable to include a positive control side.

16

Seymour and Sander

In fluorescence microscopy of tissue sections, non-fluorophore-labeled cells are not readily visible under the microscope. A commonly used method to visualize all nuclei in a tissue section is to counterstain with DAPI, which is excited at 358 nm (blue emission). With the appropriate filters, DAPI can be observed on the same tissue section together with fluorochromes that emit in the red (e.g., Cy3) or green spectrum (e.g., fluorescein). Secondary antibodies with different emission spectra are commonly used to detect two different antigens simultaneously (see Notes 5 and 6). Double-labeling experiments are helpful for determining which cell populations express GFP or β-gal. The two different antigens are best distinguished by fluorochromes that emit in the green and red spectrum, respectively. Importantly, because of its more intense fluorescent emission, the Cy3-conjugated antibody should be used to stain the antigen that is more difficult to detect (see Note 7). For the same reason, Cy3 is the preferred fluorochrome for detection of the sole target antigen in single-labeling experiments. 3.1. Fixation of Tissue 1. Chill freshly prepared 4% paraformaldehyde in PBS on ice and transfer to appropriate sealable containers (e.g., screw-top glass or plastic tubes). 2. After the animal is sacrificed, transfer the tissue quickly into the fixative. Dissections of tissue under the microscope should be performed in ice-cold PBS prior to fixation. To ensure proper fixation, add at least 10 times the volume of fixative to the tissue sample. Fix the tissue under gentle agitation at 4°C. Fixation time depends on the size of the tissue (see Note 8).

3.2. Preparation of Cryoprotected Tissue Sections 1. After fixation, discard paraformaldehyde into a hazardous waste container, and wash the tissue three times for 15 min each in 1X PBS at 4°C, again with gentle agitation throughout. 2. Transfer samples into prechilled 10% sucrose in PBS and leave at 4°C until the tissue has sunk to the bottom of the tube. Larger samples can also be left in 10% sucrose in PBS at 4°C overnight. Subsequently transfer the tissue to ice-cold 15% sucrose in PBS and again leave at 4°C until the tissue has sunk to the bottom of the tube. Repeat the same steps in 30% sucrose in PBS. 3. Mix Tissue-Tek O.C.T. compound 1:1 (v/v) with 30% sucrose in PBS and leave tissue in the O.C.T./sucrose mixture for 30 min at room temperature under gentle agitation. 4. Transfer tissue to Tissue-Tek O.C.T. compound and gently agitate for 30 min at room temperature. 5. Place the embedding molds in an ethanol/dry ice bath, or place on a slab of dry ice dampened with ethanol to promote evaporation and further cooling. Fill the molds with Tissue-Tek O.C.T. compound and transfer tissue into the mold. Ensure that

Immunofluorescent Staining Method for GFP and β-Gal

6.

7. 8.

9.

10.

11.

17

no residual sucrose solution is surrounding the tissue. Leave the samples in the ethanol/dry ice bath until the O.C.T. compound is completely frozen. Then, transfer samples to the −80°C freezer for storage. Cut sections from the cryoblocks on a cryostat (e.g., Leica CM3050 S). Precool the chamber of the cryostat to the desired temperature for the tissue to be sectioned, and place the chuck (onto which the cryoblock will be mounted) inside. Cutting temperature is dependent on the fat content of the tissue sample; for most tissues, the temperature should be around −18°C to −20°C. Transfer the cryoblock(s) to the cryostat chamber on dry ice and allow 30 min to equilibrate to the chamber temperature. Cover the chuck with O.C.T. and place inside the chamber. Remove the cryoblock from the mold by splitting the mold at its corners, and orient it so that the desired section plane will be obtained. When the clear O.C.T. begins to turn opaque white in colour, press the cryoblock onto the chuck and apply pressure for 10 to 20 s to freeze the cryoblock securely onto the chuck. Allow a further 30 min for equilibration of the sample to the chamber temperature. Clamp the chuck into the specimen holder. Use the “trim” function of the cryostat to advance the block until the specimen can be seen through the O.C.T. With a new, precooled razor blade, trim the cryoblock face to give a 5- to 10-mm border around the specimen. Set the section thickness: 10-µm sections are typical, although thinner sections will give better resolution of cell layers when viewed microscopically. a. If a particular tissue needs to be identified within the sectioning block (e.g., within an embryo), view cut sections periodically under a binocular dissecting microscope until the desired tissue appears. b. Cut serial sections and mount on glass slides in “sets” of 5 or 10 slides as required for later staining. This method allows adjacent serial sections to be easily stained for different antigens by staining individual whole slides with distinct antisera. c. To prepare sets of five slides, for example with three sections per slide, hold the first slide at the frosted end and mount the first section at the “top,” or opposite end of the slide. d. Mount the successive four sections at the top ends of the next four slides in the set. e. Mount the next five sections centrally on the same five slides, followed by the successive five sections on the “bottom” position of each of those five slides. f. Repeat this process for further sets of five slides from the same specimen until the tissue is exhausted. g. A border of 5 mm should be left around each edge of the slide when mounting to allow for an unstained area owing to the interface with the staining coverplates. Leave freshly cut sections for between 1 and 6 h to air-dry at room temperature before storage. Slides can either be stained immediately or stored frozen. For storage, care should be taken to minimize water ingress, as this will lead to tissue damage. Place slides into a labeled polypropylene microscope slide box containing a

18

Seymour and Sander Humidity Sponge™ silica desiccant sachet (Fisher) and seal the box inside a Ziploc® bag for storage at −80°C.

3.3. Immunofluorescence Detection of GFP or β -Gal on Tissue Sections 1. When you are retrieving slides from the −80°C freezer for staining, temporarily place boxes on dry ice to avoid freeze-thaw damage to sections. Then, dry the removed slides at room temperature for approx 30 min. Subsequently, place slides in a slide holder and transfer to a glass staining dish with 1X PBS. Put a stir bar on the bottom of the glass dish and wash for 5 min at room temperature to dissolve the O.C.T. compound. 2. Apply 500 µL of 1X PBS to each Shandon Coverplate and slowly lower the slide, tissue-side down, onto the coverplate. Avoid the generation of air bubbles between the tissue and the coverplate. Transfer the coverplate and slide into the cassette slots in the Shandon Sequenza slide rack. The slides remain in the slide rack until they are mounted and coverslipped (see Note 9). Pipet 200 µL of fresh, prechilled blocking solution (1% NGS in PBST) into the cavity between the slide and the coverplate, and leave the samples at room temperature for a minimum of 30 min. 3. Dilute the primary antibody (anti-GFP 1:2000; anti-β-gal 1:500) in blocking solution. 4. Pipet 200 µL of primary antibody in 1% NGS/PBST into the coverplate cavity. Incubate overnight at 4°C. 5. Wash slides by applying 2 mL of 1X PBS into the coverplate cavity. Leave at room temperature for 5 min. Repeat the washing step twice for a total of three washes in 1X PBS. 6. Dilute the secondary antibody to a final dilution of 1:2000 in blocking solution (1% NGS in PBST). The secondary antibody is freshly prepared for each experiment. When exposed to light, photobleaching decreases the staining intensity of fluorochrome-conjugated antibodies. Therefore, samples should be protected from light exposure by placing the cassettes or slides in the dark during all following steps. Samples can be easily protected by covering the containers with aluminum foil. All subsequent steps should also be performed in the absence of direct overhead illumination. 7. Pipet 200 µL of diluted secondary antibody in 1% NGS/PBST into the coverplate cavity. Incubate for 1 h at room temperature. 8. Wash slides three times for 5 min each by applying 2 mL of 1X PBS as in step 5. 9. Once the final wash is completed, take each slide and its corresponding coverplate out of the slide rack together. To remove the coverplate, place the slide and coverplate gently into a 500-mL beaker containing 1X PBS, and allow the coverplate to float away from the slide. Forceful removal of the coverplate will damage the sections. 10. In preparation for reuse, first wash the coverplates under running double-distilled water and then immerse briefly in 70% ethanol before leaving them to air-dry. 11. The slides are then ready to be mounted in aqueous mounting medium. Although these can be prepared in the lab, it is often more efficient to purchase commercially available products (e.g. Vectashield or Aqua Poly/Mount) owing to the time-expense of preparation.

Immunofluorescent Staining Method for GFP and β-Gal

19

a. Apply a few drops of aqueous mounting medium onto the tissue sections and carefully invert the coverslip into the mounting medium. b. Remove air bubbles by gently pressing down on the coverslip. If a large number of air bubbles obscure any of the sections, then float the coverslip off the slide by immersing the newly mounted slide vertically in the beaker of 1X PBS and gently shaking until the coverslip slides off the end of the slide. The slide can then be remounted usually with no detriment to the sections. c. To seal the coverslip onto the slide, carefully apply nail polish to the corners of the coverslip. Mounting, as with step 6 and onward, should be performed in an area without direct overhead illumination. The samples can be viewed immediately after the nail polish has dried. d. For storage, lay the slides flat in a microscope slide folder or slide box and protect them from light. When slides are stored at 4°C, fluorescence is stable for several weeks. For longer term storage, slides should be placed at −20°C. 12. View the slides under a fluorescence microscope. Excitation at 543 nm induces Cy3 fluorescence, while excitation at 358 nm induces DAPI fluorescence. To gain better resolution or to analyze coexpression of two proteins in the same cell, imaging under a confocal microscope is superior to a fluorescence microscope. Images for each fluorochrome are taken separately and can be overlaid with the appropriate software (e.g., Photoshop, Adobe, San Jose, CA). Examples of immunofluorescence staining for β-gal and GFP are shown in Fig. 1.

4. Notes 1. For detection of GFP and β-gal with secondary antibodies raised in goat, blocking with 1% NGS in PBST produces high intensity staining with minimal background. In double-labeling experiments, detection of the second antigen sometimes requires use of an alternative blocking solution (e.g., when the primary antibody is produced in goat). A blocking solution that works very well with many antibodies is 1% IgGfree bovine serum albumin (BSA; lyophilized powder; Sigma) in PBST. When 1% IgG-free BSA in PBST is used for blocking, the primary and secondary antibodies should also be diluted in this solution. 2. We have found that these two antibodies work best in immunofluorescence on tissue sections. We tested numerous competitive reagents from other commercial sources but did not obtain comparable results. 3. In double-labeling experiments, the two antigens are commonly detected by use of a Cy3-conjugated secondary antibody in conjunction with a secondary antibody that emits in the green spectrum, such as Alexa Fluor 488-conjugated antibodies (Invitrogen [Molecular Probes], Carlsbad, CA). Like fluorescein, Alexa Fluor 488 fluorescent dye is excited at 495 nm but provides superior photostability. When you are choosing secondary antibodies for multiple labeling, both secondary antibodies should be derived from the same host species so they do not recognize one another. One should also ensure that the secondary antibodies do not cross-react with immunoglobulins from other species possibly present in the assay system, or with endogenous immunoglobulins possibly present in the tissues under investigation.

20

Seymour and Sander

Fig. 1. Immunofluorescence detection of β-galactosidase (β-gal) and green fluorescent protein (GFP) on pancreatic tissue sections from transgenic mice. (A–C) Transgene expression in pancreas from Pdx1-Nkx6.1-IRESLacZ transgenic embryos at embryonic day (E) 14.5 is identified by coimmunofluorescence staining with a rabbit anti-β-gal and a guinea pig anti-Pdx1 antibody. The anti-β-gal and anti-Pdx1 antibodies are detected with an Alexa 488-conjugated goat anti-rabbit IgG and a Cy3-conjugated goat anti-guinea pig IgG antibody, respectively. Since the LacZ transgene carries a nuclear localization signal, β-gal colocalizes with the transcription factor Pdx1 in the nucleus. (D–F) Transgene expression in pancreas from Ngn3-Nkx6.1-IRESeGFP transgenic embryos at E16.5 is identified by coimmunofluorescence staining with a rabbit anti-GFP and a guinea pig anti-Ngn3 antibody. The anti-GFP and anti-Ngn3 antibodies are detected with an Alexa 488-conjugated goat anti-rabbit IgG and a Cy3-conjugated goat anti-guinea pig IgG antibody, respectively. The transcription factor Ngn3 is localized to the nucleus, whereas GFP is detected in the cytoplasm. In coexpressing cells, a red nucleus is surrounded by green cytoplasm. See accompanying CD for color version. 4. Because antigens are frequently masked by the fixation methods used to prepare tissues for staining, detection of some antigens, in particular nuclear antigens, requires antigen retrieval or unmasking methods. Such methods usually rely on the use of slightly acidic solutions and heat treatments in various combinations to catalyze the hydrolysis of formaldehyde-induced crosslinks between neighboring

Immunofluorescent Staining Method for GFP and β-Gal

21

proteins. It should be noted that although the advantages of antigen unmasking on paraffin sections are well accepted, the efficacy of unmasking on frozen sections proves to be a matter of some contention. However, we find that antigen unmasking on frozen sections is beneficial to the staining of many nuclear antigens. Since each antibody can produce highly variable staining results contingent on the antigen retrieval procedure used, it is advisable to perform a pilot staining experiment using more than one method to optimize the protocol. Antigen retrieval is performed after the initial washing step with 1X PBS before application of the blocking solution. a. Leave slides in the glass dish, and incubate in citrate buffer (to make the 200 mL required to immerse the slides fully in the staining dish, add 98 mL doubledistilled water and 83 mL 0.1 M sodium citrate to 19 mL 0.1 M citric acid) for 1 h at 37°C. b. Seal the dish and lid with Parafilm to prevent evaporation. c. Wash in 1X PBS three times for 5 min in the glass dish. d. An alternative, harsher method is to microwave the slides in the glass dish containing 10 mM sodium citrate, pH 6.0, for 7 min at 30% power so the liquid is gently boiling. The dish should be completely filled with liquid to prevent the sections from drying out owing to evaporation. e. Slides should be left to cool at room temperature for 45 min before being washed in 1X PBS as described. f. Antigen unmasking can be used in conjunction with an additional permeabilization step to facilitate antibody access to the cell interior. g. To do this, transfer slides to Shandon Sequenza slide rack and add 1 mL of 0.15% Triton X-100 in 1X PBS to each coverplate. h. Incubate for 1 h at room temperature. i. Proceed directly with the blocking step. 5. In double-staining experiments, both primary and secondary antibodies are normally applied at the same time. In some instances, however, simultaneous application of both primary antibodies impairs the results of the immunostaining. In such cases, staining for both antigens should be performed sequentially, beginning with the antigen that is more difficult to detect. The Shandon Sequenza slide rack should be kept at 4°C following the first staining round until application of the second primary antibody. Usually, it is unnecessary to perform a second blocking step prior to the second staining round. 6. When mouse primary monoclonal antibodies are used to stain mouse tissue, high background staining often obscures the specific staining. This is because of the inability of the anti-mouse secondary antibody to distinguish between the mouse primary antibody and endogenous mouse immunoglobulins in the tissue. This background problem can be significantly reduced by using the Vector M.O.M. immunodetection kit. 7. For antigens that are difficult to detect, tyramide signal amplification™ (Invitrogen [Molecular Probes], Carlsbad, CA) can significantly enhance the signal. However, signal amplification often also increases background.

22

Seymour and Sander

8. Since antigens are sensitive to fixation, overfixation tends to impair immunodetection. Fixation times need to be optimized for each tissue, ranging from 45 min to overnight. 9. An alternative to working with the Shandon Coverplates and cassette system is the use of a hydrophobic PAP-Pen (DAKO, Carpinteria, CA) in conjunction with a humid chamber. Individual tissue sections are circled with the PAP-Pen, and a drop of 100 to 200 µL solution is applied to each section. The Shandon system has the advantages that the tissue quality is consistently maintained over the course of the immunostaining procedure and that less antibody is needed.

Acknowledgments The authors thank Dr. Shelley B. Nelson for contributing the images. We would also like to thank Dr. Christopher V.E. Wright for providing the anti-Pdx1 antibody. This work was supported by NIH/NIDDK-1R01-DK68471-01, NIH/ NIDDK-1U19-DK072495-01, the American Diabetes Association and the Juvenile Diabetes Research Foundation. References 1. Branda, C. S. and Dymecki, S. M. (2004) Talking about a revolution: the impact of site-specific recombinases on genetic analyses in mice. Dev. Cell 6, 7–28. 2. Soriano, P. (1999) Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat. Genet. 21, 70–71. 3. Lobe, C. G., Koop, K. E., Kreppner, W., Lomeli, H., Gertsenstein, M., and Nagy, A. (1999) Z/AP, a double reporter for cre-mediated recombination. Dev. Biol. 208, 281-292. 4. Novak, A., Guo, C., Yang, W., Nagy, A., and Lobe, C. G. (2000) Z/EG, a double reporter mouse line that expresses enhanced green fluorescent protein upon Cremediated excision. Genesis 28, 147–155. 5. Mao, X., Fujiwara, Y., Chapdelaine, A., Yang, H., and Orkin, S. H. (2001) Activation of EGFP expression by Cre-mediated excision in a new ROSA26 reporter mouse strain. Blood 97, 324–326. 6. Hadjantonakis, A. K., Dickinson, M. E., Fraser, S. E., and Papaioannou, V. E. (2003) Technicolour transgenics: imaging tools for functional genomics in the mouse. Nat. Rev. Genet. 4, 613–625. 7. Mombaerts, P., Wang, F., Dulac, C., et al. (1996) Visualizing an olfactory sensory map. Cell 87, 675–686. 8. Giepmans, B. N., Adams, S. R., Ellisman, M. H., and Tsien, R. Y. (2006) The fluorescent toolbox for assessing protein location and function. Science 312, 217–224. 9. Ausubel, F., Brent, R., Kingston, R. E., et al., eds. (1987) Current Protocols in Molecular Biology, vol. 3, John Wiley & Sons, New York, 14.1.1–14.6.8. 10. Brown, R. W. and Chirala, R. (1995) Utility of microwave-citrate antigen retrieval in diagnostic immunohistochemistry. Mod. Pathol. 8, 515–520.

Immunofluorescent Staining Method for GFP and β-Gal

23

11. Shi, S. R., Cote, R. J., and Taylor, C. R. (1997) Antigen retrieval immunohistochemistry: past, present, and future. J. Histochem. Cytochem. 45, 327–343. 12. Shi, S. R., Chaiwun, B., Young, L., Cote, R. J., and Taylor, C. R. (1993) Antigen retrieval technique utilizing citrate buffer or urea solution for immunohistochemical demonstration of androgen receptor in formalin-fixed paraffin sections. J. Histochem. Cytochem. 41, 1599–1604.

Detection of Reporter Gene Expression In Vivo

25

3 Detection of Reporter Gene Expression in Murine Airways Maria Limberis, Peter Bell, and James M. Wilson Summary We have shown that to overcome the low levels of expression from gene transfer vector-mediated β-galactosidase expression in lung, it is essential to replace the cytoplasmic β-galactosidase gene with a nuclear targeted β-galactosidase gene. We found that lung should be sectioned and fixed prior to staining for β-galactosidase expression and that en bloc staining of intact lung is inefficient at staining positively transduced cells located deeper in the lung spaces. For GFP fluorescence, it is important to inflate the lungs with fixative prior to freezing and subsequent sectioning. For processing of nasal tissues for β-galactosidase expression, we expand on a protocol used in previously reported gene transfer studies. Key Words: LacZ, β-galactosidase; GFP; placental alkaline phosphatase; airway; epithelium; nose; lung; gene expression.

1. Introduction Airway-directed gene transfer has emerged as a promising curative approach for the treatment of genetic lung diseases such as cystic fibrosis and α1-antitrypsin deficiency (1). In theory, delivery of a normal copy of the therapeutic gene to the defective airway epithelium will restore normal airway function. Gene transfer vectors for the treatment of lung diseases are initially evaluated in the airways of small animal models including mice. Therefore, it becomes important that detection of reporter transgene expression not be compromised by applying universal detection protocols that are suitable to other tissues. We found that sectioning of cryopreserved tissues followed by fixation with 0.5% glutaraldehyde/DPBS offers sensitive detection of β-galactosidase (βgal) gene expression (2). The detection efficiency of β-galactosidase-express-

From: Methods in Molecular Biology, vol. 411: Reporter Genes: A Practical Guide Edited by: D. Anson © Humana Press Inc., Totowa, NJ

25

26

Limberis, Bell, and Wilson

ing cells was greatly improved by using nuclear targeted β-gal as a transgene and maintaining a pH of 7.0 for the working 5-bromo-4-chloro-3-indoyl-β-Dgalactoside (X-gal) solution (2). En bloc staining of intact lung was inefficient at staining positively transduced cells located deeper in the lung (2), presumably owing to the poor diffusion of the staining solution throughout the tissue. To preserve green fluorescence protein (GFP) expression from the positively transduced cells the tissue had to be inflated with fixative followed by cryosectioning. Staining of nasal airway tissues for β-gal expression was performed based on previously reported protocols (3,4). 2. Materials 2.1. Collection of Airway Tissues 1. Dulbecco’s phosphate-buffered saline (without calcium and magnesium) (DPBS; Cellgro, Mediatech, Herndon, VA). Store at room temperature. 2. Optimal cutting temperature (OCT) freezing compound Tissue-Tek® (Fisher). Store at room temperature. 3. Inflation solution: prepare by shaking well equal parts of DPBS with OCT. Allow the solution to settle; the inflation solution should be free of air bubbles prior to use and kept at room temperature. 4. 2% (v/v) Paraformaldehyde (PFA) (16% [w/v] stock; Electron Microscopy Sciences, Hatfield, PA)/0.5% (v/v) glutaraldehyde (25% [w/v] stock; Electron Microscopy Sciences) in DPBS. Make fresh as required. 5. Formalin: 10% neutral buffered formalin (NBF) (Fisher). Store at room temperature. 6. Isopentane (methylbutane) (Fisher). Extremely volatile. 7. Cryomolds (Fisher). 8. Monoject Blunt needles (Sherwood Medical, St. Louis, MO). 9. Syringes (Fisher).

2.2. Sectioning Lung Tissues 1. Fisherbrand Superfrost Plus disposable microscope slides (Fisher). 2. Fisherbrand cover glasses (square). 3. Cryostat: used to section the cryorpreserved lung tissues.

2.3. Sectioning Nasal Tissues 1. Demineralizing solution (1X): 7% (v/v) HCl in 1.5% (w/v) EDTA. Carefully dissolve EDTA in H2O on a stirring hot plate in a fume hood (do not allow solution to exceed 60°C). When EDTA is completely dissolved, the solution should appear cloudy. Carefully add conc. HCl and continue to stir until solution turns clear. Caution: The demineralizing solution is highly toxic and should only be opened in a fume hood (see Note 1). Stable at room temperature for 1 mo. 2. Carnoy’s fixative: 3:1 of 95% ethanol to glacial acetic acid. Store at room temperature and use only in a fume hood. 3. Microtome: used to section paraffin-embedded tissues.

Detection of Reporter Gene Expression In Vivo

27

4. Fisherbrand Unisette Biopsy processing/embedding cassette with Lid (Fisher).

2.4. Histochemical Staining for Reporter Gene Expression in Lung 2.4.1. X-Gal Staining for β-Gal Expression 1. 1 mM MgCl2 in DPBS. Store at room temperature. 2. 0.5% (v/v) Glutaraldehyde in DPBS. Make fresh as required. 3. Pre X-gal solution: 5 mM potassium ferrocyanide (K4Fe(CN)6; Sigma, St. Louis, MO), 5 mM potassium ferricyanide (K3Fe(CN)6; Sigma) and 1 mM MgCl2 (Sigma) in 1X DPBS. Solution is light sensitive and is stable at room temperature for 1 mo. 4. X-gal (Fisher). Dissolve 40 mg/mL X-gal in dimethylformamide. Aliquot in 500-µL lots, and store at −20°C for 3 mo. Solution is light sensitive. 5. Working X-gal solution: make fresh as required by adding X-gal stock solution (1:40) to Pre-X-gal and mixing well. If necessary, adjust pH (with HCl) to 7.0 (see Note 2). Working X-gal solution is light sensitive and should be used within 1 h. 6. Glass staining dishes (Fisher).

2.4.2. NBT/BCIP Staining for Placental Alkaline Phosphatase Expression 1. 1 mM MgCl2 in DPBS. Store at room temperature. 2. 0.5% (v/v) glutaraldehyde in DPBS. Stable at 4°C for 1 mo. 3. NBT/BCIP (nitroblue tetrazolium chloride/5-bromo-4-chloro-3-indolyl-phosphate, toluidine salt) working solution: dissolve 1 ready-to-use tablet (Roche, Indianapolis, IN) in 10 mL of distilled H2O. Solution is light sensitive and should be made fresh as required. A dark blue color precipitate indicates sites of alkaline phosphatase enzyme activity. Incubation times vary; for highest sensitivity, the sections should be stained overnight. 4. Glass staining dishes (Fisher).

2.5. Counterstaining Tissues 1. 2. 3. 4. 5. 6. 7. 8. 9.

70% Ethanol (Fisher). Store at room temperature. 95% Ethanol (Fisher). Store at room temperature. 100% Ethanol (Fisher). Store at room temperature. Xylene (Fisher). Store at room temperature and handle in fume hood. Nuclear Fast Red (NFR pre-made solution, Vector). For long-term storage, keep at 4°C. NFR can be reused. Harris Hematoxylin #7211 (Richard Allen Scientific). Clarifier (Clarifier 2, Richard Allen Scientific). Eosin Y (Richard Allen Scientific). Permount mounting medium (Fisher).

2.6. Analysis of GFP 1. Vectashield mounting medium with DAPI (Vector).

28

Limberis, Bell, and Wilson

3. Methods 3.1. Collection of Airway Tissues 3.1.1. Collection of Lungs 1. Anesthetize mice weighing approx 25 g by an intraperitoneal injection of a mixture of ketamine/xylazine in DPBS (70/7 mg/kg). 2. Bleed the heart transdiaphragmatically to minimize blood spillage into the airway lumen, as blood interferes with the histochemical staining (5). 3. Expose the ventrum of the thorax and neck. a. Cut along the diaphragm and then cut along both sides of the ribs and remove the rib cage. b. Dissect the cervical musculature to expose the trachea. c. Using a 27-g needle, make a small hole between the cartilage rings and insert a blunt needle attached to a 5-mL syringe that contains inflation solution. d. Inflate the lung while it is still inside the rib cage to avoid overinflation. Lungs are adequately inflated when the inflation solution reaches the margins of all lung lobes and the lungs fill the chest cavity. e. Immediately remove the needle, and grasp the trachea with forceps to keep the lumen closed. f. Cut the trachea proximal to the forceps using scissors. g. Holding the trachea, remove the entire pluck by cutting the esophagus and aorta at the diaphragm. h. Using fine-tip forceps and dissection scissors remove the entire heart. i. Place the lungs into a cryomold in the preferred orientation, and cover completely in OCT. 4. Use two containers, of which the outer one holds liquid nitrogen, and the inner one, preferably a metal beaker, is filled with isopentane. Place the container with isopentane into the liquid nitrogen until the isopentane starts to freeze on the wall of the inner container. 5. With a pair of long forceps, carefully drop the whole cryomold with the OCTembedded lungs into the cold isopentane. Leave for approx 10 to 30 s (until completely frozen), and then remove the sample and place on dry ice until all tissues are collected. 6. Store cryomolds at −80°C until sectioned.

3.1.2. Collection of Heads 1. Proceed with steps 1 and 2 in Subheading 3.1.1. 2. Separate the head from the carcass and remove the skin and eyes using fine-tip forceps and scissors. 3. Snip the soft portion of the nose tip with fine scissors and flush the head with 2% (v/v) PFA/0.5% (v/v) glutaraldehyde in DPBS via the tracheal remnant. Keep on ice for 2 h.

Detection of Reporter Gene Expression In Vivo

29

3.2. Sectioning Lung Tissues 1. Remove lung tissues from −80°C and equilibrate to −25°C. 2. Using the cryostat, section lung tissues at 8–10 µm. 3. Store slides at −80°C until time for staining. If slides are stored in a box, remove box from −80°C and do not open until box has reached room temperature.

3.3. Sectioning Nasal Tissues 1. Decalcify the heads in demineralizing solution for 22 h in a fume hood. Do not tighten caps, as the decalcification process releases toxic fumes. 2. Decant the demineralizing solution in the fume hood, and wash softened heads under running H2O for 30 min (see Note 3). 3. Place heads in 70% ethanol for long-term storage or until paraffin embedding. 4. Remove the jaw and tongue. 5. Mark the right side with a strip of black ink, which is made permanent by dabbing the marked head region in Carnoy’s fixative solution. 6. The softened heads are sectioned with disposable microtome blades in three standard sections (4). These sections are taken at level 6 (immediately posterior to the dorsal incisor), level 16 (where the two nasal airways coalesce into the nasopharyngeal duct), and level 24 (at the rear of the head) (6). 7. Place the three cross-sections in sectioning cassettes, anterior face down, and embed in paraffin.

3.4. Histochemical Staining for Reporter Gene Expression in Lung 3.4.1. X-Gal Staining for β-Gal Expression 3.4.1.1. LUNG 1. Slides should be at room temperature. 2. Fix lung sections in a glass staining dish that contains 0.5% (v/v) glutaraldehyde/ DPBS (see Note 4) (cooled to 4°C) for 10 min. 3. Remove slide and immediately blot excess fixative from the sides of the slide. (Do not allow slide to dry.) 4. Wash slide(s) in a glass staining dish containing 1 mM MgCl2 in DPBS for 15 min. 5. Repeat step 4. 6. Remove slide and immediately blot excess wash solution from the sides of the slide. (Do not allow slide to dry.) 7. Stain tissues for β-gal expression by placing them in a glass staining dish containing working X-gal solution (see Note 5). Protect from light and incubate at 37°C (see Note 6) from 30 min to 16 h. A blue color precipitate indicates sites of enzyme activity. Incubation times may vary; for highest sensitivity, the sections should be stained overnight (approx 16 h). 3.4.1.2. NOSE 1. Flush the head with 1 mM MgCl2/DPBS through the tracheal remnant and place in 1 mM MgCl2/DPBS. Incubate on ice for 15 min.

30

Limberis, Bell, and Wilson

2. Repeat step 1. 3. Stain heads for β-gal expression by placing them in polystyrene containers containing working X-gal solution (approx 8 mL/head). Protect from light and incubate at 37°C for 8 to 16 h with intermittent flushing of the nasal cavity though the tracheal remnant. (Solution should come out of the nose.) 4. Wash the heads in 0.9% (w/v) NaCl/H2O for 30 min at room temperature. 5. Postfix the heads in 10% (v/v) NBF for 24 h at room temperature. (Keep in fume hood.)

3.4.2. NBT/BCIP Staining for Placental Alkaline Phosphatase Expression 1. Warm a glass staining dish containing 1 mM MgCl2/DPBS to 65°C (see Note 7). 2. Incubate the glass slide with the sectioned lung tissue in the heated 1 mM MgCl2/ DPBS solution for 30 min. 3. Remove the slide and immediately blot excess solution from the sides of the slide. (Do not allow slide to dry.) 4. Stain tissues for placental alkaline phosphatase gene expression by placing them in a glass staining dish containing working NBT/BCIP solution. Protect from light and incubate at room temperature. Incubation times may vary; for highest sensitivity, the sections should be stained overnight (approx 16 h).

3.5. Counterstaining Tissues 3.5.1. Counterstaining Cryopreserved Tissues with NFR 1. Remove the slide holder and immediately blot excess stain solution from the sides of the slide holder. (Do not allow slide to dry.) 2. Rinse slide holder in PBS (at room temperature). 3. Rinse slide holder in H2O (at room temperature). 4. Blot excess H2O from the sides of the slide holder, place in a glass staining dish that contains NFR (at room temperature), and incubate for 3 to 5 min (see Note 8). 5. Rinse slide holder in H2O (at room temperature). 6. Dehydrate lung tissues by placing the slide holder through a graded ethanol series starting once at 70% ethanol, twice at 95% ethanol, twice at 100% ethanol, and three times with xylene for 1 min each. (Blot excess solution from slide holder in between steps.) 7. Cover slip with Permount mounting medium and allow to dry overnight.

3.5.2. Counterstaining Cryopreserved Tissues with Hematoxylin and Eosin 1. Proceed with steps 1 to 3 of Subheading 3.5.1. 2. Blot excess H2O from the sides of the slide holder and place in a glass staining dish that contains hematoxylin (at room temperature) and incubate for 1 min. 3. Rinse slide holder in warm H2O (37°C) for 1 min. 4. Blot excess H2O from the sides of the slide holder and place in a glass staining dish that contains Clarifier solution for 10 s.

Detection of Reporter Gene Expression In Vivo

31

5. Rinse slide holder in H2O (at room temperature) for 10 s. 6. Blot excess H2O from the sides of the slide holder and place in a glass staining dish that contains Blueing Reagent for 10 s. 7. Rinse slide holder in H2O for 10 s. 8. Blot excess H2O from the sides of the slide holder and place in a glass staining dish that contains Eosin Y for 10 s. 9. Rinse slide holder in H2O for 3 s. 10. Dehydrate tissues by placing the slide holder through a graded ethanol series starting twice at 70% ethanol, twice at 95% ethanol, twice at 100% ethanol, and three times with xylene for 1 min each. 11. Cover slip with Permount mounting medium and allow to dry overnight.

3.5.3. Counterstaining Paraffin-Embedded Tissues with NFR 1. Place slide holder with tissue in xylene for 5 min at room temperature (in a fume hood). 2. Repeat step 1 with fresh xylene solution. 3. Remove slide holder and immediately blot excess xylene solution from the sides of the slide holder (do not allow slide to dry) and add to xylene:ethanol (1:1 [v/v]) solution for 1 min at room temperature. 4. Place the slide holder through a graded ethanol series starting twice at 100% ethanol, once at 95% ethanol, and once at 70% ethanol for 1 min each. 5. Rinse slide holder in H2O (at room temperature). 6. Repeat step 5. 7. Blot excess H2O from the sides of the slide holder and place in a glass staining dish that contains NFR (at room temperature) and incubate for 3 to 5 min. 8. Rinse slide holder in H2O for 10 s. 9. Proceed to steps 12 and 13 of Subheading 3.5.2. 10. Cover slip with Permount mounting medium and allow to dry overnight.

3.5.4. Counterstaining Paraffin-Embedded Tissue with Hematoxylin and Eosin 1. Place slide holder with tissue in xylene for 5 min at room temperature (in a fume hood). 2. Repeat step 1 with fresh xylene solution. 3. Remove slide holder and immediately blot excess xylene solution from the sides of the slide holder (do not allow slide to dry) and add to xylene:ethanol (1:1 [v/v]) solution for 1 min at room temperature. 4. Place the slide holder through a graded ethanol series starting twice at 100% ethanol, once at 95% ethanol, and once at 70% ethanol for 1 min each. 5. Rinse slide holder in H2O (at room temperature). 6. Repeat step 5. 7. Blot excess H2O from the sides of the slide holder, place in a glass staining dish that contains hematoxylin (at room temperature), and incubate for 2 min.

32

Limberis, Bell, and Wilson

8. Rinse slide holder in warm H2O (37°C) for 1 min. 9. Blot excess H2O from the sides of the slide holder and place in a glass staining dish that contains Clarifier solution for 1 min. 10. Rinse slide holder in H2O for 10 s. 11. Blot excess H2O from the sides of the slide holder and place in a glass staining dish that contains Blueing Reagent for 1 min. 12. Rinse slide holder in H2O for 10 s. 13. Blot excess H2O from the sides of the slide holder and place in a glass staining dish that contains Eosin Y for 10 s. 14. Rinse slide holder in H2O for 3 s. 15. Proceed to steps 12 and 13 of Subheading 3.5.2. 16. Cover slip with Permount mounting medium and allow to dry overnight.

3.6. Cryopreserving Green Fluorescence Protein-Expressing Lung 1. Follow steps 1 and 2 of Subheading 3.1.1. 2. Expose the ventrum of thorax and neck. a. Cut along the diaphragm and then cut along both sides of the ribs and remove the rib cage. b. Dissect the cervical musculature to expose the trachea. c. Using a 27-g needle, make a small hole between the cartilage rings and insert a blunt needle attached to a 5-mL syringe that contains 10% NBF (see Note 9). d. Inflate the lung while it is still inside the rib cage to avoid over inflation. Lungs are adequately inflated when the 10% NBF solution reaches the margins of all lung lobes and the lungs fill the chest cavity. e. Immediately remove needle, and suture close the trachea. f. Cut the trachea proximal to the suture using scissors. g. Place the lungs into a polystyrene container filled with 10% NBF. h. Protect from light and incubate overnight on a shaker on a low setting. 3. Wash lung with DPBS for 2 h at room temperature. Protect from light. 4. Repeat step 3 twice. 5. Blot the lungs dry using 3MM Whatman paper (Fisher) and place into a cryomold. 6. Cover lung with OCT and freeze as described in Subheading 3.1.1., steps 4 and 5. 7. Prepare cryosections at 8–10 µm and collect on slides. 8. Cover slip immediately with Vectrashield (containing DAPI to visualize cell nuclei).

4. Notes 1. Crystals may form in the demineralizing solution. The solution can be filtered occasionally, although this step is not necessary. 2. The β-gal signal increased when the pH was lowered from 7.5 to 6.5 and decreased slightly at pH 6.0. However, from pH 6.0 to 6.5, nonspecific staining was obtained that was not present at (and higher than) pH 7.0. Therefore, a pH of 7.0 appears to provide a good tradeoff between staining sensitivity and the elimination of falsepositive signals.

Detection of Reporter Gene Expression In Vivo

33

3. Tape uncovered specimen pots with autoclave tape in a cross formation so the tissues stay in their correct pots. 4. Other aldehyde-based fixatives such as 10% NBF or PFA resulted in slightly weaker but still acceptable staining reactions with X-gal. However, Streck tissue fixative (S.T.F.) inactivated the enzyme activity completely. Solvents such as methanol or acetone, used at −20°C for 10 min, weakened β-gal activity, resulting in relatively weak and diffuse staining. 5. An alternative substrate for β-gal staining is Bluo-gal, which has been reported to have a lower diffusion rate and to provide a “crisper” color precipitate than X-gal (7). Owing to its higher electron density, Bluo-gal has also been used for the detection of β-gal activity by electron microscopy (e.g., refs. 8–10). We found that this substrate results in the generation of small speck-like precipitates that are nonspecific. 6. Avoid cell culture incubators, because the high CO2 concentration can cause a drop in pH. 7. This step eliminates endogenous alkaline phosphatase activities but does not affect the heat-resistant human placental alkaline phosphatase. 8. Counterstaining with NFR or hematoxylin and eosin should be monitored since the aim is to stain the lung tissue only lightly to allow the quantitation of β-gal-expressing cells. 9. It is important to fix the whole lungs before freezing and sectioning in order to preserve the GFP fluorescence. Sections from unfixed tissues, even when fixed after sectioning, show a highly diffuse fluorescence, which makes it almost impossible to localize GFP-expressing cells.

References 1. Crystal, R. G. (1992) Gene therapy strategies for pulmonary disease. Am. J. Med. 92, 44S–52S. 2. Bell, P., Limberis, M., Gao, G., et al. (2005) An optimized protocol for detection of E. coli beta-galactosidase in lung tissue following gene transfer. Histochem. Cell. Biol. 124, 77–85. 3. Parsons, D. W., Grubb, B. R., Johnson, L. G., and Boucher, R. C. (1998) Enhanced in vivo airway gene transfer via transient modification of host barrier properties with a surface-active agent. Hum. Gene Ther. 9, 2661–2672. 4. Limberis, M., Anson, D. S., Fuller, M., and Parsons, D. W. (2002) Recovery of airway cystic fibrosis transmembrane conductance regulator function in mice with cystic fibrosis after single-dose lentivirus-mediated gene transfer. Hum. Gene Ther. 13, 1961–1970. 5. Johnson, L. G., Olsen, J. C., Naldini, L., and Boucher, R. C. (2000) Pseudotyped human lentiviral vector-mediated gene transfer to airway epithelia in vivo. Gene Ther. 7, 568–574. 6. Mery, S., Gross, E. A., Joyner, D. R., Godo, M., and Morgan, K. T. (1994) Nasal diagrams: a tool for recording the distribution of nasal lesions in rats and mice. Toxicol. Pathol. 22, 353–372.

34

Limberis, Bell, and Wilson

7. Weis, J., Fine, S. M., David, C., Savarirayan, S., and Sanes, J. R. (1991) Integration site-dependent expression of a transgene reveals specialized features of cells associated with neuromuscular junctions. J. Cell Biol. 113, 1385–1397. 8. Loewy, A. D., Bridgman, P. C., and Mettenleiter, T. C. (1991) Beta-galactosidase expressing recombinant pseudorabies virus for light and electron microscopic study of transneuronally labeled CNS neurons. Brain Res. 555, 346–352. 9. Blanks, J. C., Spee, C., Barron, E., Rich, K. A., and Schmidt, S. (1997) Lineage study of degenerating photoreceptor cells in the rd mouse retina. Curr. Eye Res. 16, 733–737. 10. Sekerkova, G., Katarova, Z., Joo, F., Wolff, J. R., Prodan, S., and Szabo, G. (1997) Visualization of beta-galactosidase by enzyme and immunohistochemistry in the olfactory bulb of transgenic mice carrying the LacZ transgene. J. Histochem. Cytochem. 45, 1147–1155.

3D Analysis of Molecular Signals

35

4 Three-Dimensional Analysis of Molecular Signals with Episcopic Imaging Techniques Wolfgang J. Weninger and Timothy J. Mohun Summary This chapter describes two episcopic imaging methods, episcopic fluorescence image capturing (EFIC) and high-resolution episcopic microscopy (HREM). These allow analysis of molecular signals in a wide variety of biological samples such as tissues or embryos, in their precise anatomical and histological context. Both methods are designed to work with histologically prepared and whole-mount stained material, and both provide highresolution data sets that lend themselves to 3D visualization and modeling. Specimens are embedded in wax (EFIC) or resin (HREM) and sectioned on a microtome. During the sectioning process, a series of digital images of each freshly cut block surface is captured, using a microscope and CCD camera aligned with the position at which the microtome block holder comes to rest after each cutting cycle. The resulting stacks of serial images retain virtually exact alignment and are readily converted to volume data sets. The two methods differ in how tissue architecture is visualized and hence how specific molecular signals are detected. EFIC uses endogenous, broad-range, tissue autofluorescence to reveal specimen structure. Addition of dyes to the wax embedding medium suppresses detection of any signal except that originating from the block surface. EFIC can be used to detect specific signals (such as LacZ) by virtue of their ability to suppress such fluorescence. In contrast, the plastic embedding medium used in HREM is strongly fluorescent, and tissue architecture is detected at the surface because of the ability of cellular and subcellular structures to suppress this signal. Specific signals generated as a result of chromogenic reactions can be visualized using band-pass filters that suppress the appearance of morphological data. In both methods, the digital volume data show high contrast; for HREM, such data achieve true cellular resolution. Their intrinsic alignment greatly facilitates their use for 3D analysis of transgene activity that can be visualized in the context of complex cellular and tissue morphology. Both methods are relatively simple and can be set up using common laboratory apparatuses. Together, they provide powerful tools for analyzing gene function in embryogenesis or tissue remodeling and for investigating developmental malformations. Key Words: Gene expression; gene activity; RNA pattern; imaging; 3D analysis; embryogenesis; development; 3D reconstruction; episcopic microscopy; remodeling. From: Methods in Molecular Biology, vol. 411: Reporter Genes: A Practical Guide Edited by: D. Anson © Humana Press Inc., Totowa, NJ

35

36

Weninger and Mohun

1. Introduction The shaping of an embryo is orchestrated by complex interactions of thousands of genes and their products that regulate changing patterns of cell type differentiation, cellular morphology, and tissue interactions throughout development. By unraveling this complex 4D developmental program in experimental models such as the mouse, we will better understand the genesis of human birth defects, improve prenatal diagnosis, and lay the basis for the rational development of therapies. A prerequisite is the acquisition of detailed knowledge about the function of proteins and the regulation of genes that encode them. Three main approaches are commonly taken for studying gene function in normal and abnormal embryogenesis. First, the roles played by individual genes can be analyzed by examining the consequences of gene mutation or ablation on normal embryo morphology. Such mutants are commonly obtained using genetic engineering to target the gene of interest but can also be obtained as individual genetic lines obtained from random mutagenesis screens. A broad range of methods can be used to analyze the phenotypes generated in such mutants, such as computed tomography (CT), magnetic resonance imaging (MRI), ultrasound (US), optical projection tomography (OPT), electron microscopy, confocal imaging, histological sections, and episcopic imaging (1–19). A second approach is to study the expression of individual genes and gene products in the developing embryo and correlate this spatiotemporal pattern with the changing morphologies of tissues and organs. As long as the gene transcript or protein product can be labeled in a specific manner, its distribution can be examined using a variety of imaging techniques (e.g., RNA and immunohistochemistry, OPT, confocal imaging, or episcopic imaging methods). In both of these approaches, gene function or expression is directly correlated with embryo morphology, allowing inferences to be made about the role of individual genes on tissue or organ development. A third and more systematic approach is to analyze changing patterns of gene activity using cDNA microarray techniques, which facilitate the simultaneous study of many thousands of genes (20–24). This approach cannot provide the spatial resolution possible with the imaging techniques available with the first two methods, since it depends on extraction of RNA from isolated cells or pieces of tissue. It cannot, for example, support the analysis of gene function or expression in individual cell lineages, nor can it examine the expression of gene products rather than gene transcripts themselves. In this chapter we will discuss two episcopic imaging methods, episcopic fluorescence image capturing (EFIC) and high-resolution episcopic microscopy (HREM). These methods provide an imaging procedure useful for comparing normal and abnormal embryo morphology as well as for studying the topology

3D Analysis of Molecular Signals

37

of individual gene expression. Both provide highly detailed images of tissue and organ architecture upon which can be superimposed a similarly precise visualization of molecular signals reflecting gene expression. By providing precisely aligned, serial images captured during sample sectioning, the data readily provide spatial visualization of gene expression in its accurate histological and anatomical context (16,17). Of course, these methods are not restricted to embryological studies and can equally be used to study growth, maintenance, or remodeling in postnatal organs or adult tissues. EFIC and HREM are both destructive imaging techniques, utilizing fixed tissue specimens progressively sectioned on a microtome. Sample fixation is therefore common. In each method, successive episcopic images (i.e., images of the block surface) are captured using similar data acquisition procedures. However, the two methods differ in the embedding medium used, necessitating individual protocols for each, and the characteristics of the images they yield are radically different. As a result, each is best suited for a distinct method of data processing. EFIC produces “negative images” since it detects tissue autofluorescence, and background fluorescence from the embedding medium is suppressed. Its ability to detect molecular signals is also a negative contrast technique, since it relies on the extinction of autofluorescence by the LacZ chromogenic precipitate. In contrast, HREM provides “positive images,” since tissue is visualized by its ability to reduce the high level of fluorescence from the plastic embedding medium. Its ability to detect specific signals rests on the use of filters to reduce the signal:background ratio for tissue without a comparable reduction in the ratio obtained with a chromogenic reaction product. HREM is therefore a positive contrast technique, which provides information on molecular signals by detecting color reactions. Both techniques can be used with a broad range of biological specimens, ranging from embryos to adult tissues to biopsy material in a variety of species. In principle, imaging can be carried out in any resolution, this being limited only by the optics used for image capture and the resolution of the CCD camera used. In practice, the different methods by which tissue is visualized affect the effective resolution and the size of the specimen that is appropriate. In particular, since EFIC depends on endogenous autofluorescence of tissue, its ability to resolve different tissues and reveal architecture within tissues or organs depends on intrinsic variations in autofluorescence between cell types. For example, late-gestation mouse embryos show a wide range of autofluorescence levels in different tissues and yield complex images revealing considerable morphology. Early gestation embryos, however, show far more uniform and much lower levels of autofluorescence, presumably reflecting the much lower degree of

38

Weninger and Mohun

tissue specialization and tissue density in early embryos. Some individual cells can readily be detected by EFIC (for example, red blood cells, which are highly fluorescent); however, in general, the dispersion of fluorescent signal precludes effective imaging at cellular resolution within most tissues. Despite these limitations, the negative image character and the broad range of autofluorescence signals detected in EFIC data sets make them ideally suited for 3D volume rendering. EFIC therefore provides the simplest way to obtain 3D models of morphology. Unlike EFIC, HREM is capable of consistently providing cellular resolution, and indeed subcellular structures such as nuclei are the most prominent feature of HREM images. HREM imaging shows none of the signal dispersion found with EFIC at higher magnifications, and useful data can readily be obtained over a range from 1- to 200-fold magnification. Paradoxically, the wealth of structural information available from HREM limits the ease with which data sets can be used for 3D modeling, Volume rendering is possible but generally less effective, both because of the narrower range of grayscale values used to represent tissue structure and also because of the inherent complexity of cellular resolution data. However, HREM data are ideal for the more laborious methods of isosurface rendering after manual or semiautomatic segmentation. In each imaging method, embryo or tissue specimens are first used for labeling in a whole mount (RNA in situ hybridization or immunohistochemistry) and then embedded in an embedding medium of wax (EFIC) or resin (HREM); the resulting blocks are mounted and sectioned on a microtome. In order to capture perfectly aligned images of the block face, microscope optics and digital camera are positioned perpendicular to the block face at the stopping position of the block holder (Fig. 1). Appropriate emission and detection filter sets can be introduced into the light path, and sequential cycles of block cutting, filter changing, and image acquisition can be performed manually or automatically depending on the precise equipment. In either case, stacks comprising hundreds of precisely aligned digital images of freshly cut block surface can be generated within a few hours. Figure 2 outlines the steps and their order during episcopic data generation. Depending on the optical magnification and the CCD camera chip size, digital images with pixel sizes between 0.4 and 50 µm can be achieved in the image data. Because of their inherent alignment, captured image stacks can immediately be converted into volume data sets and used for precise 2D and 3D analysis. With minimal voxel sizes as small as 0.4 × 0.4 × 1.0 µm, cellular features, tissue architecture, and organ morphology can be visualized extremely precisely along with equally precise representations of specific gene or transgene expression (see, for example, http://www.univie.ac.at/efic and http://www.meduniwien. ac.at/3D-Rekonstr/HREM).

3D Analysis of Molecular Signals

39

Fig. 1. Data capturing apparatuses. (A) Based on a sliding microtome. (B) Based on a rotary microtome.

2. Materials 2.1. Specimen Preparation 2.1.1. EFIC 1. 4% Paraformaldehyde in phosphate-buffered saline (PBS) or 10% phosphate-buffered formalin (Sigma Aldrich) (see Note 1). 2. Ethanols: 30%, 50%, 70%, and 100% (see Note 2).

40

Weninger and Mohun

Fig. 2. Steps in data generation and data analysis. 3. Histoclear. 4. Embedding wax comprising 20% Vybar, 76.5% paraffin wax, 3.5% stearic acid, and 0.35% Sudan IV aniline dye (Sigma Aldrich). The mixture is melted at 65°C and left overnight. Undissolved wax is then removed by filtration through paper at 65°C overnight. 5. Disposable embedding molds (Leica Microsystems) (see Note 3).

2.1.2. HREM 1. 2. 3. 4. 5. 6.

4% Paraformaldehyde in PBS. PBS, 0.1% (v/v) Tween 20: methanol mixes, 25% (v/v), 50% (v/v), and 75% (v/v). 100% Methanol. In situ hybridization solutions (e.g., see refs. 25 and 26). Ethanols: 30%, 50%, and 70%. Infiltration solution: 100 mL JB-4 Solution A (Polysciences, www.polysciences. com), mixed with 1.25 g Benzoyl Peroxide Plasticized (Catalyst) and 0.4 g eosin (Eosin, spritlöslich, Waldeck), stirred at 4°C. 7. Embedding molds and block holders (Leica Microsystems). 8. Embedding solution: 25 mL infiltration solution mixed with 1 mL of JB-4 Solution B.

2.2. Data Capturing 2.2.1. EFIC 1. Sliding microtome (e.g., Leica SM2500), equipped with a photo-stop position (see Note 4).

3D Analysis of Molecular Signals

41

2. Dissecting microscope (MZ16 F/FA, Leica Microsystems), equipped with a GFP1 barrier filter set (peak excitation 425 nm, emission barrier 480 nm), mounted on an x-y micrometer stage above the photo-position of the microtome (see Note 5). 3. Cooled, grayscale CCD camera capable of detecting low-intensity fluorescence (e.g., Hamamatsu Orca ER). 4. Image acquisition software with appropriate camera driver and computer (e.g., Power Mac G5 and OpenLab [Improvision] or IPLab [Scanalytics] software) (see Note 6).

2.2.2. HREM 1. Rotary microtome (CUT 4060 E, microTec), specially adapted by the company to increase the accuracy of the block holder resting position. The microtome is placed on a motor-driven x-y stage (Walter Uhl, technische Mikroskopie). 2. Optics of a compound microscope (DM LM Head, Leica Microsystems), equipped with band-pass GFP (excitation 470/40 nm, emission 525/50 nm) and TX2 (excitation 560/40 nm, emission 645/75 nm) filter cubes. The optics are placed on a solid, z-axis adjustable stage (components from Stahlbau Reumüller and MikroscopieService). 3. Digital color video camera (Leica DFC 480) with a target size of 2560 × 1920 pixels. 4. PC, equipped with camera driver and data capturing routine (software assembling by M. Donoser, TU-Graz, A).

2.3. Data Processing and Visualization Any 32- or 64-bit computer with image manipulation and rendering software, for example, Power Mac G5 or Windows PC with Openlab (Improvision), Photoshop (Adobe Systems), and “Volocity” (Improvision) packages, or alternatively, a Linux-based workstation with Amira (Mercury Computer Systems) software. Data stacks are large, and 2 to 4 GB of RAM are essential. 3. Methods As with all histological methods, the optimum processing time depends on specimen size and tissue density. The protocols below are appropriate for “average” specimens (for EFIC: midgestation mouse embryos; for HREM specimens of 1 to 2 mm). Up to 10 specimens can be comfortably processed simultaneously. Small specimens such as young embryos can be processed more quickly; larger or denser specimens will require considerable increase in preparation and processing times. As a guide, embedding a single sample requires approximately 10 mL of wax (EFIC) or 2 mL of resin (HREM). 3.1. Specimen Preparation 1. Harvest specimens, transfer them into PBS, if necessary dissect them, and fix them overnight in 4% formaldehyde, buffered with PBS. Wash at 4°C in PBS, 0.1% (v/v) Tween 20 twice for 5 min.

42

Weninger and Mohun

2. Perform whole-mount staining. If working with specimens carrying a LacZ transgene, use X-gal staining for detection of transgene expression patterns. Use digoxigenin (DIG)-labeled RNA probes and the NBT/BCIP detection system for visualizing specific RNA expression patterns following a standard protocol appropriate for the sample (e.g., refs. 25 and 26). Stained specimens can be stored in PBS, 0.1% (v/v) Tween 20 for several months at 4°C. 3. Dehydrate specimens through a graded series of alcohols, e.g., 30%, 50%, and 70% ethanol using wash times appropriate for sample size and density. (One to 2 h each is generally adequate.) Samples are now ready for infiltration with embedding medium. They can be stored in 70% ethanol at 4°C prior to embedding, although prolonged storage prior to EFIC is not recommended.

3.2. Infiltration and Embedding 3.2.1. EFIC 1. Complete the dehydration of samples using sequential washes with 80%, 85%, 90%, 95%, and 100% ethanol (20–30 min each). a. Wash with Histoclear/ethanol (1:1 mix) briefly (5–20 min depending on specimen size) followed by Histoclear for a similar period and immediately transfer into wax dye mix maintained at 65°C. b. Wash twice for a minimum of 15 min with fresh wax and then infiltrate with wax for a minimum of 60 min (twice). c. For larger samples, infiltration times may need to be extended several fold, although vacuum embedding can increase infiltration rates. 2. Transfer samples into fresh wax mix in a disposable mold, and carefully alter specimen orientation at room temperature as the wax begins to set. Samples can be photographed at this stage to record their precise location within the block (see Note 3).

3.2.2. HREM 1. Use infiltration solution at 4°C under continuous rocking. Infiltrate the specimen for 1 h, change, and infiltrate overnight; change and infiltrate for an additional 1 to 2 h (see Note 7). 2. Prepare fresh embedding solution and embed the specimens (see Notes 8–10). a. Put the specimens into the embedding molds, fill them with embedding solution, and orient the specimens as the medium becomes “sticky”. b. Immediately put the block holder into the mold and fill the rest of the mold with embedding solution. c. Cover the mold tightly and store at 18°C overnight. d. Remove the blocks from the molds and keep them for 4 to 24 h at room temperature prior to data capture.

3.3. Data Capturing 1. Prepare the block for sectioning. For wax blocks (EFIC), trim using the grid lines on the image captured during embedding to ensure that the specimen is centered

3D Analysis of Molecular Signals

2. 3. 4. 5. 6. 7.

8.

43

in the remaining wax. For plastic blocks (HREM), illuminate from the side and draw in the future field of view by projecting the position of the specimen inside the block onto the block surface (see Notes 12 and 13). Mount the blocks on the microtome and move it near the knife (see Notes 15 and 16). Choose an objective and the x-y position of the optics according to the position and the size of the desired field of view. Section until the entire block surface is freshly cut. Focus the optics on the block surface using the camera images. Select appropriate filter (for EFIC: GFP3; for HREM: GFP or TX2) (see Note 11). Select appropriate section thickness, adequate exposure times, number of images to be captured from each freshly cut block surface, and total number of images to be captured in this session, and commence image capturing. For HREM, switch between GFP and TX2 filter sets during each image capture cycle.

3.4. Data Processing and Visualization With either imaging procedure, a single data set obtained with the GFP filter captures morphology. For EFIC, this also contains the specific signal data since this is represented by localized regions of signal suppression. To assist in identifying such regions, it is helpful to compare data with an equivalent, unstained sample. For HREM, the specific signal is obtained from a second data set, captured with the TX2 filters. Further processing and analysis of data sets depends on the choice of software used and the visualization method required. 1. Commercial software tools (e.g., Amira or Volocity) or public domain software (e.g., ImageJ or EMAP software) permit rapid review of image stacks either in the original section plane or as virtual, resectioned image stacks. 2. For 3D modeling by volume rendering, HREM data sets require inversion of the gray LUT to transform images from “positive” to “negative”. Grayscale mapping within each data set can then be adjusted in order to obtain suitable contrast and background black levels in volume-rendered models. Visualization “inside” regions of the model or through appropriately positioned “windows” is easily achieved either by modeling of partial stacks or by modeling entire stacks in which an area of data has been removed on successive images. 3. Image editing software can be used to extract structures of interest, generate surfaced-rendered virtual 3D models, and visualize them simultaneously. This requires segmentation of data. The method chosen to do this depends on both the structure under analysis and the type of data set. Where the structure boundary is accurately defined by a clear transition in grayscale value, segmentation can be achieved using automated thresholding algorithms. This approach can be used with HREM data sets of molecular signals captured with the TX2 filter, since morphology data are suppressed under these conditions. Most morphological structures, however,

44

Weninger and Mohun must be extracted from GFP data sets set using semiautomatic thresholding or purely interactive (manual) contour finding.

4. Notes 1. The choice of fixative can affect the level of autofluorescent signal and the range of contrast between different tissues. Formaldehyde-based fixatives yield the strongest signal with the greatest contrast. They also show the brightest signal from blood cells. Bouin’s fixative will yield much less intense signals from blood, but the overall range of contrast between tissues is significantly reduced. 2. There is some evidence that prolonged storage of samples in ethanol prior to analysis leads to a loss of autofluorescent signal. Samples are probably better stored in fixative. 3. In order to locate samples within the wax block and to align the optics appropriately, it can be helpful to photograph the sample in the embedding mold while the wax remains molten. This is possible using transmitted light and a reference grid placed either below or within the base of the mold. 4. An economical alternative to the motorized SM2500 microtome is a manual sliding blade microtome (e.g., Leica SM2000R), since the position of the block face remains fixed in all three dimensions. Data acquisition is more laborious, since sectioning is manual. 5. For the motorized SM2500 microtome, the optics and x-y base stage must be mounted directly onto the blade holder assembly to ensure that the block face remains in focus after each section cycle. 6. Complete automation of sectioning and image capture can be achieved using the camera trigger signal from the SM2500 microtome to initiate image capture within the software package. 7. Because of eosin, the infiltration solution soon becomes viscous and hardens within a few days, even without resin solution B. It is therefore essential to keep the eosinstained solution A in the fridge and use freshly mixed solutions for infiltration and embedding. Penetration of the specimen by the infiltration solution is often inadequate if old solutions are used. 8. Cool the embedding solution outside the fridge during embedding. 9. Tightly cover the filled embedding molds with paraffin. This is essential for adequate hardening of the embedding solution. 10. Before sectioning, store the blocks for a few hours in the room in which it is sectioned. The embedding medium can then adapt to the humidity of the room, which prevents the block from breaking during sectioning. 11. As an alternative to the GFP filter set, the Leica YFP filter set (excitation filter 500 nm, emission filter 535 nm) can be used. This reduces excitation times. 12. Keep the light in the sectioning room dark and at constant intensity. This enhances the quality of the single images and ensures homogeneity of the image series. 13. Keep the temperature in the sectioning room low and constant. This avoids artifacts caused by heating of the block.

3D Analysis of Molecular Signals

45

14. Avoid vibrations near the data capturing apparatus. 15. Test the orientation of the digital images. Some optical arrangements generate mirrored images, which must be flipped prior to reconstruction. 16. Use D-faceted tungsten knifes for sectioning. This minimizes scratching artifacts.

References 1. Schneider, J. E., Bamforth, S. D., Farthing, C. R., Clarke, K., Neubauer, S., and Bhattacharya, S. (2003) Rapid identification and 3D reconstruction of complex cardiac malformations in transgenic mouse embryos using fast gradient echo sequence magnetic resonance imaging. J. Mol. Cell. Cardiol. 35, 217–222. 2. Sharpe, J., Ahlgren, U., Perry, P., et al. (2002) Optical projection tomography as a tool for 3D microscopy and gene expression studies. Science 296, 541–545. 3. Denk, W. and Horstmann, H. (2004) Serial block-face scanning electron microscopy to reconstruct three-dimensional tissue nanostructure. PLoS Biol. 2, e329. 4. Effmann, E. L., Johnson, G. A., Smith, B. R., Talbott, G. A., and Cofer, G. (1988) Magnetic resonance microscopy of chick embryos in ovo. Teratology 38, 59–65. 5. Yu, Q., Shen, Y., Chatterjee, B., et al. (2004) ENU induced mutations causing congenital cardiovascular anomalies. Development 131, 6211–6223. 6. Shen, Y., Leatherbury, L., Rosenthal, J., et al. (2005) Cardiovascular phenotyping of fetal mice by noninvasive high-frequency ultrasound facilitates recovery of ENUinduced mutations causing congenital cardiac and extracardiac defects. Physiol. Genomics 24, 23–36. 7. Streicher, J., Weninger, W. J., and Muller, G. B. (1997) External marker-based automatic congruencing: a new method of 3D reconstruction from serial sections. Anat. Rec. 248, 583–602. 8. Schierlitz, L., Dumanli, H., Robinson, J. N., et al. (2001) Three-dimensional magnetic resonance imaging of fetal brains. Lancet 357, 1177–1178. 9. Ruijter, J. M., Soufan, A. T., Hagoort, J., and Moorman, A. F. (2004) Molecular imaging of the embryonic heart: fables and facts on 3D imaging of gene expression patterns. Birth Defects Res. C Embryo Today 72, 224–240. 10. Soufan, A. T., Ruijter, J. M., van den Hoff, M. J., de Boer, P. A., Hagoort, J., and Moorman, A. F. (2003) Three-dimensional reconstruction of gene expression patterns during cardiac development. Physiol. Genomics 13, 187–195. 11. Weninger, W. J., Streicher, J., and Müller, G. B. (1996) [3-Dimensional reconstruction of histological serial sections using a computer]. Wien Klin. Wochenschr. 108, 515–520. 12. Kaufman, M. H. and Richardson, L. (2005) 3D reconstruction of the vessels that enter the right atrium of the mouse heart at Theiler stage 20. Clin. Anat. 18, 27–38. 13. Louie, A. Y., Huber, M. M., Ahrens, E. T., et al. (2000) In vivo visualization of gene expression using magnetic resonance imaging. Nat. Biotechnol. 18, 321–325. 14. Ewald, A. J., McBride, H., Reddington, M., Fraser, S. E., and Kerschmann, R. (2002) Surface imaging microscopy, an automated method for visualizing whole embryo samples in three dimensions at high resolution. Dev. Dyn. 225, 369–375.

46

Weninger and Mohun

15. Weninger, W. J., Meng, S., Streicher, J., and Müller, G. B. (1998) A new episcopic method for rapid 3-D reconstruction: applications in anatomy and embryology. Anat. Embryol. (Berl.) 197, 341–348. 16. Weninger, W. J. and Mohun, T. (2002) Phenotyping transgenic embryos: a rapid 3-D screening method based on episcopic fluorescence image capturing. Nat. Genet. 30, 59–65. 17. Weninger, W. J., Geyer, S. H., Mohun, T. J., et al. (2006) High-resolution episcopic microscopy: a rapid technique for high detailed 3D analysis of gene activity in the context of tissue architecture and morphology. Anat. Embryol. (Berl.) 211, 213–221. 18. Kerwin, J., Scott, M., Sharpe, J., et al. (2004) 3-Dimensional modelling of early human brain development using optical projection tomography. BMC Neurosci. 5, 27. 19. Rosenthal, J., Mangal, V., Walker, D., Bennett, M., Mohun, T. J., and Lo, C. W. (2004) Rapid high resolution three dimensional reconstruction of embryos with episcopic fluorescence image capture. Birth Defects Res. C Embryo Today 72, 213–223. 20. Chen, H. W., Yu, S. L., Chen, W. J., et al. (2004) Dynamic changes of gene expression profiles during postnatal development of the heart in mice. Heart 90, 927– 934. 21. Kaynak, B., von Heydebreck, A., Mebus, S., et al. (2003) Genome-wide array analysis of normal and malformed human hearts. Circulation 107, 2467–2474. 22. Balza, R. O. Jr. and Misra, R. P. (2005) The role of serum response factor in regulating contractile apparatus gene expression and sarcomeric integrity in cardiomyocytes. J. Biol. Chem. 281, 6498–6510. 23. Napoli, C., Lerman, L. O., Sica, V., Lerman, A., Tajana, G., and de Nigris, F. (2003) Microarray analysis: a novel research tool for cardiovascular scientists and physicians. Heart 89, 597–604. 24. Alvarez, E., Zhou, W., Witta, S. E., and Freed, C. R. (2005) Characterization of the Bex gene family in humans, mice, and rats. Gene 357, 18–28. 25 Wilkinson, D. G. (1998) In Situ Hybridisation: A Practical Approach. Oxford University Press, Oxford. 26. Streit, A. and Stern, C.D. (2001) Combined whole-mount in situ hybridization and immunohistochemistry in avian embryos. Methods 23, 339–344.

Fluorescent Proteins for Cell Biology

47

5 Fluorescent Proteins for Cell Biology George H. Patterson Summary Through the use of exogenous labels, such as antibodies and synthetic fluorophores, experimenters have been able to readily observe the localization of proteins and organelles within a cell by fluorescence microscopy. The discovery and application of fluorescent proteins spanning a large wavelength range have revolutionized these studies. This chapter attempts to introduce the vast array of these molecules, discuss their characteristics, and assess the advantages and disadvantages that each displays for use in imaging. Key Words: GFP; DsRed; photoactivatation; imaging.

1. Why Use a Fluorescent Protein? With the number of tools available to the biologist for the study of his/her favorite gene or protein, why choose to work with a fluorescent protein? First, fluorescent proteins can be specifically fused to a protein of interest without the nonspecific labeling associated with many other techniques. This is made possible with the straightforward methods of genetically tagging proteins of interest and the more than 50 variations of fluorescent protein coding sequences contributed to GENBANK. Second, their fluorescence forms without the requirements of factors from native organisms or any additional exogenous agents other than molecular oxygen, which allows expression in a diverse range of organisms, tissues, and cell types. Finally, the fluorescence from these molecules can be observed directly without the requirement of fixation and addition of a secondary label or reactant, which makes observations possible in the living specimen. 2. Aequorea victoria GFP Advances in fluorescent protein technology in imaging began with the cloning of the Aequorea victoria green fluorescent protein (GFP) gene (1) and its From: Methods in Molecular Biology, vol. 411: Reporter Genes: A Practical Guide Edited by: D. Anson © Humana Press Inc., Totowa, NJ

47

48

Patterson

functional expression in heterologous systems (2). The impact of these two studies on cell biology should not be underestimated since they promoted GFP from an interesting protein isolated from a marine organism to a cell biology tool cited in almost 20,000 publications (PubMed search for ‘GFP’; June 2006). Prior to 1994, extensive biochemical and spectroscopic data were collected on the bioluminescent protein extract isolated from the jellyfish (3–7). Evidence from these early investigations demonstrated that the protein held promise as a useful tool for monitoring cell and organism processes if its fluorescence properties could be transferred into other species. The nuclear magnetic resonance (NMR) structure of proteolytic peptides containing the GFP chromophore indicated that residues serine 65, tyrosine 66, and glycine 67 undergo an uncommon cyclization reaction (8). Since such a structure had been previously unknown in protein structural biology, the novel chromophore suggested a requirement for endogenous jellyfish factor(s). Nevertheless, upon expression of the cDNA in heterologous systems, it was evident that the formation of this structure occurs in the absence of specific jellyfish factors (2,9). It was later shown that oxygen is required for production of the functional chromophore (10); however, no other exogenous factors have been reported as being necessary for the formation of the fluorescent structure. 2.1. Aequorea victoria Fluorescent Protein Structure Two independent crystal structure determinations showed that GFP is an 11-strand β-barrel containing a short segment of α-helix within the interior of the barrel (11,12) (Fig. 1). The light-producing p-hydroxybenzylidene-imidozolidinone chromophore consists of a cyclized tripeptide, composed of residues serine 65, tyrosine 66, and glycine 67, located in the central portion of the αhelix. The formation of a fluorescing protein requires proper folding into the barrel structure, followed by cyclization of the three amino acids making up the chromophore and oxidation/dehydration reactions to produce a functional fluorescent molecule (13). 2.2. Aequorea victoria Fluorescent Protein Folding Fluorescent proteins provide a good self-indicator for their folding, since the unfolded proteins generally do not fluoresce. However, for a newly synthesized fluorescent protein, it is often difficult to separate the kinetics of folding into the barrel structure from the formation of the chromophore. A study addressing these two aspects of GFP folding kinetics was performed on the S65T mutant (discussed later) (14). After denaturation in 8 M urea, refolded S65T protein recovers fluorescence with biexponential kinetics (fast phase t1/2 approx 28 s and slow phase t1/2 approx 284 s), which are much faster than the formation of a fluorescent protein from misfolded S65T protein purified from inclusion bodies

Fluorescent Proteins for Cell Biology

49

Fig. 1. The structure of the green fluorescent protein is an 11-strand β-barrel (peptide backbone rendered as a tube) with an α-helix located in the center containing the chromophore (rendered in ball and stick model). This figure was produced using Cn3D 4.1 (National Center for Biotechnology Information, National Institutes of Health, Bethesda, MD) using the protein data bank coordinates 1GFL submitted by the authors of ref. 11. See accompanying CD for color version.

(14). In the latter experiments, the folding into the barrel is best described by a t1/2 of approx 10 min, whereas the cyclization step (Fig. 2) is faster (t1/2 approx 180 s); the final oxidized and fluorescent chromophore (Fig. 2) requires much longer (t1/2 approx 76 min) to form. These slow time constants are not thought to be owing to the starting material, since proteolysis resistance indicates that folding is similar between the insoluble nonfluorescent proteins purified from bacterial inclusion bodies and soluble fluorescent proteins purified from bacteria (14). 2.3. Aequorea victoria Fluorescent Chromophore Formation The proposed mechanism for formation of the GFP chromophore (Fig. 2) was based largely on chemical logic (13,15) using the chromophore structure derived from an NMR structure of a proteolytic peptide (8) and eventually the crystal structures of the wild-type (11) and S65T (12) Aequorea victoria (avGFP) proteins. The first step in the chromophore formation after protein folding is a cyclization step in which the peptide amido nitrogen at amino acid position G67 undergoes a nucleophilic attack on the carbonyl of the serine at position S65 to produce a five-membered imidazolinone ring (13) (Fig. 2). In recent years,

50

Patterson

Fig. 2. Formation of the Aequorea victoria GFP chromophore. The chromophore of GFP consists of S65, Y66, and G67. These residues undergo a cyclization reaction between S65 and G67 to form a five-membered ring. Oxidation and dehydration reactions follow to extend the π-bonding system of the aromatic Y66 and produce a fluorescent molecule.

molecular modeling and crystal structures of GFP mutants trapped in intermediate stages of the chromophore formation have confirmed the basic mechanism, but the role of the surrounding secondary amino acid residues and the sequence of the oxidation and dehydration steps remain unresolved (16–19). 2.4. Aequorea victoria Fluorescent Chromophore Requirements Whereas the secondary residues involved in chromophore formation, as well as their roles, are still under debate, more is known about the residues within the main chromophore. Position 65 in avGFP can be altered to a number of amino acids without deterring the development of a fluorescent molecule. Furthermore, the equivalent position in other fluorescent proteins can be any of several amino acids (Tables 1–3). The amino acid in position Y66 in the cyclized peptide is usually a tyrosine, and it can also be altered but generally requires a phenylalanine, histidine, or tryptophan to produce a protein that fluoresces (13). However, the glycine at position 67 appears to be required in most, if not all, of the GFP-like molecules discovered to date. 2.5. Improvements in Aequorea victoria GFP Although the expression of avGFP in the bacterial and nematode systems (2) demonstrated the potential of this marker, the initial transition to other cell systems was problematic.

Fluorescent Proteins for Cell Biology

51

Table 1 Selected Fluorescent Protein Variants Developed from Aequorea victoria GFP Wavelengths (nm) a Protein

Amino acid substitutions

wtGFP Sapphire (H9-40) BFP CFP Cerulean

CGFP GFP Emerald YFP (10C) Topaz Citrine Venus

T203I F64L, Y66H, Y145F F64L, S65T, Y66W, N146I, M153T, V163A F64L, S65T, Y66W, S72A, N146I, Y145A, H148D, M153T, V163A, A206K F64L, S65T, Y66W, N146I, M153T, V163A, T203Y F64L, S65T S65T, S72A, N149K, M153T, I167T S65G, V68L, S72A, T203Y S65G, S72A, K79R, T203Y S65G, V68L, Q69M, S72A, T203Y F46L, F64L, M153T, V163A, S175G, T203Y

λ ex

λ em

Reference

393–400 (473–475) 399

504–509

2,3,5

511

13,35

380–383 434 (452)

440–447 476 (505)

13,87 13,87

433

475

39

463

506

88

488–489 487

507–509 509

514 514 516

527 527 529

12,34 34 41

515

528

42

10,32,87 34

a Wavelengths in parentheses represent minor peaks. Where various peaks have been reported,

the ranges of reported wavelengths are indicated.

The first obstacle was that wild-type avGFP has a tendency to misfold at approx 37°C. Several mutations have been reported to improve folding and/or chromophore formation (20–22) and are helpful when the protein is expressed at 37°C. In addition, folding reporter and superfolder GFPs, which develop fluorescence when tagged to poorly folded proteins, were produced (23). Some such mutations, such as F64L, V68L, S72A, Y145F, and I167T, are located in close proximity to the chromophore, and it is generally assumed that these assist in positioning residues for the peptide cyclization or the chromophore oxidation. However, others, such as F99S, M153T, V163A, and S175G, are located many Angstroms from the chromophore within the exterior barrel structure, and we have no knowledge of the mechanism by which these mutations improve folding (22).

52

Patterson

Table 2 Selected Fluorescent Protein Variants Developed from Discosoma drFP583 Wavelengths (nm) Protein DsRed (drFP583) DsRed1 DsRed2

Amino acid substitutions

R2A,K5E, K9T, V105A, I167T, S197A DsRed1.T1 R2A, K5E, N6D, T21S, H41T, N42Q, V44A, C117S, T217A dimer2 R2A, K5E, N6D, T21S, H41T, N42Q, V44A, V71A, C117T, F118L, I125R, V127T, S131P, K163Q, S179T, S197T, T217S mRFP1 R2A, K5E, N6D, T21S, H41T, N42Q, V44A, V71A, K83L, C117E, F124L, I125R, V127T, L150M, R153E, V156A, H162K, K163M, A164R, L174D, V175A, F177V, S179T, I180T, Y192A, Y194K, V195T, S197I, T217A, H222S, L223T, F224G, L225A mHoneydew Q66M, Y67W, T147Sa mBanana b Q66C, T147S, Q213L, I197Ea b mOrange V7I, T41F, L83F, Q66T, T147S, M182K, T195V dTomato b Q66Ma mTangerine Q66C, T147S, Q213La mStrawberry b V7I, S62T, Q64N, Q66T, T147S, M182K, T195V, Q213La b mCherry V7I, Q66M, T147S, M163Q, T195Va mRaspberry F65C, A71G, I161Ma mPlum V16E, R17H, K45R, F65I, L124V, I161M, K166Ra a Mutations

λ ex

λ em

Reference

558

583

46

558 561

583 587

554

586

54

552

579

51

584

607

51

487 (504) 537 (562) 540 553 548 562

55 55 55

554 568 574

581 585 596

55 55 55

587

610

55

598 590

625 649

56 56

in mRFP1 coding sequence. the first seven N-terminal amino acids of mRFP1 replaced with the N-terminal amino acids MVSKGEE of avGFP and has the addition of the C-terminal seven amino acids GMDELYK of EGFP. b Has

53

asulGFP (asFP499) asFP522 asCP562 mcGFP mmGFP MiCy Kusabira-Orange 522

511 562 432 398 472 (380) 548 477 505 495 561

597 620 612 612 600 499

572 583 590 590 592 480

595

572 A148S S165V S165A S165C S165T

645 538

598 528

HcRed zoanYFP (zFP538) asCP (asculCP, asFP595)

509

λem

498

λex

Wavelengths (nm)

Renilla GFP

Protein

Amino acid substitutions

M65, Y66, G67 Q62, Y63, G64 T60, Y61, G62 Q69, Y70, G71 C64, Y65, G66

M65, Y66, G67 M65, Y66, G67 M65, Y66, G67 M65, Y66, G67 M65, Y66, G67 Q65, Y66, G67

M65, Y66, G67

E64, Y65, G66 K66, Y67, G68

S66, Y67, G68

Chromophore peptide

Anemonia sulcata Anemonia sulcata Montastrea cavernosa Meandrina meandrites Acropara sp. Fungia concinna

Anemonia sulcata Anemonia sulcata Anemonia sulcata Anemonia sulcata Anemonia sulcata Anemonia sulcata

Anemonia sulcata

Heteractis crispa Zoanthus sp.

Renilla reniformis

Organism

Table 3 Selected Fluorescent Protein Variants Discovered and Developed from Other Marine Organisms

(continued)

91 91 92 92 60 60

90 90 90 90 90 91

80

5 Patent: US 6232107-B 15-MAY-2001 89 46

Reference

Fluorescent Proteins for Cell Biology 53

54 S3P, K6T, K7E, V19I, Y101S, C143S, M146I, S158A, N168D

V123T, Y188A, F190K

F102S, A104S, V123T, C151S, F162Y, F193Y, G195S, K11R, V25I, K32R, S55A, T62V, Q96E, E117Y, V133I, S139V, T150A, A166E, Q190G, F13Y, C115T, C217S

eqFP611 zoanGFP(zFP506) amajGFP(amFP486) dstrGFP(dsFP483)

gtCP aeCP597 AQ143

Azami-Green (AG) mAG

mKO

Protein

Amino acid substitutions

Table 3 (Continued)

559 496 458 456

580 abs 597 abs 595

492

492

548

λex

611 506 486 484

655

505

505

559

λem

Wavelengths (nm)

M63, Y64, G65 N66, Y67, G68 K68, Y69, G70 Q66, Y67, G68

Q65, Y66, G67 M63, Y64, G65 M63, Y64, G65

Q62, Y63, G64

Q62, Y63, G64

C64, Y65, G66

Chromophore peptide

Entacmaea quadricolor Zoanthus sp. Anemonia majano Discosoma striata

Goniopora tenuidens Actinia equina Actinia equina

Galaxeidae (stony coral)

Galaxeidae (stony coral)

Fungia concinna

Organism

93 46 46 46

59 57 57

59

59

60

Reference

54 Patterson

dis3GFP dendGFP mcavGFP rfloGFP scubGFP1 scubGFP2 dis2RFP (dsFP593) zoan2RFP mcavRFP rfloRFP anm1GFP1 anm1GFP2 phiYFP anm2CP ppluGFP1 ppluGFP2 laesGFP pmeaGFP1

rmueGFP

clavGFP(cFP484) cgigGFP hcriGFP ptilGFP

510

483 496 500 508

503 512 494 508 506 516 508 518 497 506 497 506 573 593 552 576 508 (572) 520 (580) 506 (566) 517 (574) 475 495 490 504 525 537 572 597 480 500 482 502 491 506 489 504

498

443 399 (482) 405 (481) 500

Q62, Y63, G64 H62, Y63, G64 D69, Y70, G71 Q62, Y63, G64 Q66, Y67, G68 Q68, Y69, G70 Q66, Y67, G68 D66, Y67, G68 H62, Y63, G64 H62, Y63, G64 S99, Y100, G101 S62, Y63, G64 T65, Y66, G67 Q65, Y66, G67 G57, Y58, G59 G57, Y58, G59 G57, Y58, G59 G57, Y58, G59

Q69, Y70, G71

Q104, Y105, G106 Q63, Y64, G65 R63, Y64, G65 Q69, Y70, G71

55

(continued)

46 45 45 Patent: US 6232107-B 15 MAY 2001 Renilla muelleri Patent: US 6232107-B 15 MAY 2001 Discosoma sp. 3 45 Dendronephthya sp. 45 Montastracea cavernosa 45 Ricordea florida 45 Scolymia cubensis 45 Scolymia cubensis 45 Discosoma sp. 2 94 Zoanthus sp. 2 45 Montastracea cavernosa 45 Ricordea florida 45 Anthomedusae sp. SL-2003 58 Anthomedusae sp. SL-2003 58 Phialidium sp. 58 Anthomedusae sp. SL-2003 58 Pontellina plumata 58 Pontellina plumata 58 Labidocera aestiva 58 Pontella meadi 58

Clavularia sp. Condylactis gigantea Heteractis crispa Ptilosarcus sp.

Fluorescent Proteins for Cell Biology 55

56

a Proteins

487 491 V11I, F64L, K101E, 480 T206A, E222G 578 abs 571 abs 571 abs 580 abs

pmeaGFP2 pdae1GFP acGFPL (aceGFP) hcriCP a (hcCP) cgigCP a (cgCP) cpasCP a (cpCP) gtenCP a (gtCP)

502 511 505

λem

Pontella meadi Pontellidae sp. SL-2003 Aequorea coerulescens Heteractis crispa Condylactis gigantea Condylactis passiflora Goniopora tenuidens

E63, Y64, G65 A63, Y64, G65 A63, Y64, G65 Q62, Y63, G64

Organism

G57, Y58, G59 G57, Y58, G59 S65, Y66, G67

Chromophore peptide

with “CP” in their name are referred to as chromoproteins and are not fluorescent.

λ ex

Protein

Wavelengths (nm)

Amino acid substitutions

Table 3 (Continued)

89 89 89 89

58 58 78

Reference

56 Patterson

Fluorescent Proteins for Cell Biology

57

Second, the Aequorea victoria specific codon usage of wtGFP rendered protein expression in other systems inefficient. Mutagenesis improved these properties by converting the coding sequence to versions more efficiently used by plant (24,25), yeast (26), and mammalian systems (27). Although these mutations may have resulted in small changes in the intrinsic brightness of the protein, their major influence was to produce more observable molecules. Because the fluorescent protein requires synthesis by the cell, a time delay is inevitable between the formation of the translated protein and the maturation into a fluorescing protein. Work on the avGFP protein revealed that the wildtype GFP formed fluorescence with a time constant of approx 2 h (10). In contrast, the S65T mutant was observed to develop fluorescence with a time constant of approx 0.45 h (10). Experiments with fluorescent protein fusion proteins can be complicated by any influence that the fluorescent protein has over the protein of interest, such as self-association or oligomerization of the fluorescent protein. Although the wtGFP and its derivatives dimerize at high concentrations (Kd approx 0.11 mM) (22), this potential artifact was not generally considered a problem until a study of plasma membrane raft molecules was found to be complicated by significant oligomerization (28). In this study, localization of molecules in lipid rafts was analyzed using a technique called Förster resonance energy transfer (FRET), and an artifactual FRET signal was attributed to the interaction of the fluorescent proteins used (28). This could be disrupted by mutation of one of three hydrophobic residues on the exterior of the GFP barrel to charged residues (A206K, L221K, or F223R) (28). At low expression levels, self-association may show negligible interference with the proper localization and dynamics of many chimeras, but the use of fluorescent proteins with little affinity for each other is generally encouraged to avoid this complication (29). 2.6. Variations in Aequorea victoria GFP Spectra Further refinements of the Aequorea victoria GFP concentrated on its spectral properties. One of the first improvements dealt with the major (approx 400 nm) and the minor (approx 475 nm) excitation peaks (Fig. 3A). Excitation of either peak results in green emission at approx 508 nm, but this is not optimal for imaging with common laser lines and/or filter sets. Second, the two excitation peaks of wtGFP exhibit a photo-induced phenomenon, referred to as photoconversion or photoisomerization, in which the peaks interconvert during excitation (2,30,31). A single substitution of the serine at the 65 position with either a threonine, alanine, glycine, cysteine, or leucine collapses the major and minor peaks to a single peak at approx 489 nm while maintaining green fluorescence (10,32) (Fig. 3B), thus alleviating the effects of photoconversion. In addition, the excitation peak at 489 nm makes GFP brighter under excitation at

58

Patterson

Fig. 3. Excitation spectra, emission spectra, and chromophores of selected Aequorea victoria variants. The excitation spectra (solid lines) and emission spectra (dashed lines) for (A) wtGFP, (B) EGFP, (C) EBFP, (D) ECFP, (E) T203I, and (F) EYFP are shown with drawings of their respective chromophores and surrounding residues that result in their phenotypic spectra.

488 nm (reported extinction coefficients between 52,000 and 58,000) (22). Codon-optimized versions with this mutation, EGFP (33) and Emerald (34), are the common choices for imaging in mammalian systems. The next major advance in GFP imaging came with the ability to image more than one tagged molecule within the same cell. For this type of multicolor experiment, fluorescent proteins with differing spectra properties were required. The necessary spectra changes were accomplished by alterations of amino acids within and around the GFP chromophore and were discovered through random

Fluorescent Proteins for Cell Biology

59

mutagenesis experiments of GFP (10,13,35). Substitution of the tyrosine at position Y66 with a histidine shifts the absorbance spectrum to a peak at approx 380 nm with emission at approx 445 nm to produce the blue fluorescent proteins (BFPs) (13,22) (Fig. 3C). The BFPs (Y66H mutants) and the GFPs (S65 mutants) were breakthroughs in allowing the simultaneous imaging of multiple fluorescent proteins. The excitation and emission spectra of BFPs differ enough from these of GFPs to allow separation of the two signals for dualcolor imaging, even though the BFP emission overlaps sufficiently with the GFP absorbance spectrum to allow energy transfer (36). Researchers were able to take advantage of these properties in several studies, although the application of BFPs is severely limited by modest fluorescence (31,36), near ultraviolet excitation (13), and a tendency to photobleach readily (31,37). Substitution with a tryptophan in the 66 position produced W7, one of the first cyan fluorescent proteins (CFPs), which was identified in the same mutagenesis screen (13). The CFPs feature moderate photostability (37,38) and have peaks in the 435- to 450-nm (absorbance) and 475- to 505-nm (emission) regions (13,36), which places them between the BFP mutants and the green S65 mutants in the fluorescence spectrum (Fig. 3D). However, Y66W mutants have multiple fluorescent lifetimes, which complicates their use in a specialized imaging technique known as fluorescence lifetime imaging (FLIM). Structural evidence indicated that CFP molecules probably existed as a mixed population, in which the histidine at 148 and tyrosine at 145 were in alternate orientations, and their effect on the chromophore led to the multiple fluorescent lifetimes. Since these fluorescent proteins have side chains within a few Angstroms of the chromophore, Rizzo et. al. (39) hypothesized that converting the H148 into a hydrophilic amino acid (aspartic acid) would stabilize one position. The H148D mutation Cerulean did indeed result in behavior better suited for fluorescence lifetime work, but more generally, Cerulean has increased brightness by approx 2.5-fold and increased photostability by approx 30% compared with ECFP (39). An additional mutant of note from the initial mutagenesis screen of Tsien and colleagues (13), which was also independently isolated by Prendergast and co-workers (35), is referred to as Sapphire (34). Similar to wild-type GFP, Sapphire has a tyrosine in the 66 position but has an isoleucine substituted for the threonine in the 203 position. The T203I mutant has a single major peak at approx 400 nm and a green emission peak at 511 nm (Fig. 3E). This mutant has been used as a near-UV excitable fluorophore. However, Sapphire probably lacks the stability necessary for any long-term imaging and has a characteristic similar to wild-type GFP in that it undergoes a photoconversion under approximately 400 nm excitation, in which the 400-nm absorbance peak decreases, and a peak at approximately 504 nm increases (40).

60

Patterson

The T203 position was later exploited in mutants rationally designed from the crystal structure of the S65T mutant of GFP (12). These are generally referred to as yellow fluorescent proteins (YFPs) and were developed by coupling an S65G mutation with a substitution at the neighboring T203 position with tyrosine, phenylalanine, tryptophan, or histidine (12). The rationale behind targeting this residue was to place histidine or one of the aromatic amino acids in a position such that a π-orbital stacking system was produced. This was proposed to lower the energy required for excitation of the molecule and thus shift the excitation and emission wavelengths toward the red part of the spectrum. These mutations indeed led to spectra that are red-shifted by approx 20 nm. The most common version of YFP contains S65G and T203Y substitutions, which result in excitation and emission peaks at 513 and 527 nm, respectively (Fig. 3F). The shifted excitation spectrum of YFP allows excitation with the 514-nm line of an argon ion laser, and it is easily distinguished from the CFPs. The YFPs have also undergone significant improvement. Two notable versions are Citrine (41) and Venus (42). The original versions of YFP displayed pH sensitivity (pKa approx 6.9), sensitivity to chloride ions (43,44), and decreased maturation to the fluorescent form at 37°C compared with GFP (41). YFPs are also generally thought to have lower photostability compared with GFP (41), although single-molecule work addressing this issue found that YFP was more stable (38). Nevertheless, Citrine and Venus have each improved on several YFP sensitivities. Citrine resulted from a single mutation, Q69M, and was developed in the context of a Ca2+ indicator construct, camgaroo-2. The Q66M mutation largely alleviates both the halide (principally chloride) sensitivity and the pH dependence, producing a molecule with a pKa of approx 5.7. In addition, the photostability of Citrine is increased by about twofold compared with the older YFP variants and appears brighter when expressed at 37°C (41). Venus, on the other hand, relies on a mutation, F46L, which increases the efficiency of development of a fluorescent molecule at 37°C by 20- to 30-fold. This mutation, coupled with other mutations (Table 1), produces a molecule with decreased sensitivity to pH (pKa approx 6.0) and decreased chloride sensitivity (42). Venus matures more rapidly and is brighter when expressed in cells compared with Citrine (29). However, it lacks the photostability of Citrine, since it photobleaches with similar kinetics as the previous YFPs. This property is not improved by the Q69M mutation (42). 3. DsRed Fluorescent Protein Up to this point, the discussion has centered on fluorescent proteins developed from the original Aequorea victoria GFP (avGFP). Discoveries of similar proteins from other marine organisms have extended the bank of fluorescent

Fluorescent Proteins for Cell Biology

61

Fig. 4. Structure of the DsRed fluorescent protein. DsRed is also a β-barrel (peptide backbone rendered as a tube) with an α-helix located in the center containing the chromophore (rendered in ball and stick model). Wild-type DsRed is normally an (A) tetramer made up of (B) monomer subunits that are structurally similar to avGFP. This figure was produced using Cn3D 4.1 (National Center for Biotechnology Information, National Institutes of Health, Bethesda, MD, USA) using the protein data bank coordinates 1G7K submitted by the authors of ref. 48. See accompanying CD for color version.

proteins for imaging to the red wavelength range (45). The first and best characterized of the red fluorescent proteins is the DsRed protein cloned from the Discosoma coral (Fig. 4). The wild-type DsRed (drFP583) has a major absor-

62

Patterson

bance peak at 558 nm and minor peaks at 530 and 487 nm, with an emission peak at 583 nm (46), and has led to the production of a plethora of new variants with diverse characteristics. 3.1. DsRed Fluorescent Protein Structure and Chromophore Formation The overall protein structure of DsRed is similar to that of avGFP in that it consists of a β-barrel with an α-helix in the center containing the chromophore (47,48). The formation of the DsRed fluorescent chromophore is proposed to proceed similarly to that of the avGFPs by a mechanism whereby a chromophore that produces green fluorescence is initially formed but matures via an additional oxidation step into the red species (49). The cyclized tripeptide is made of Q66, Y67, and G68, with the π-bonding system of the chromophore extended along the Cα–N bond of the glutamine at position Q66 (47,48) resulting in a shift of the excitation/emission spectra to longer wavelengths (49,50). 3.2. DsRed Improvements As with the wild-type avGFP, the native DsRed protein has many undesirable characteristics, including several that rendered it almost unusable in many fusion constructs. The reaction for forming the mature DsRed chromophore requires molecular oxygen and up to 48 h (50,51). This time lag between protein expression and development of the red fluorescence made use of DsRed as a secondary red marker difficult because of cross-talk with green fluorescent signals. A mutant referred to as E5 (fluorescent timer), which contains V105A and S197T substitutions, changes DsRed fluorescence from a predominantly green to a red signal with a t1/2 of approx 10 h (52). The addition of an I161T substitution to E5, referred to as E57, further reduces the t1/2 for formation of the mature red chromophore to 3 to 4 h (53). Finally, a mutagenesis screen searching for mutants that rapidly form red fluorescence eventually reduced maturation to less than 1 h in a variant named DsRed1.T1 (54). Many of the DsRed1.T1 mutations listed in Table 2 are scattered about the DsRed barrel structure and provide no clear indication of their influence on protein folding and/or chromophore formation, illustrating the fact that there is much left to learn about the folding mechanism of these proteins. Notably, many of the currently used DsRed variants listed in Table 2 contain mutations similar to DsRed1.T1. Another major problem with DsRed is that it forms obligatory tetramers, even at low concentrations (50). Therefore, any intracellular trafficking studies have to contend with potential artifacts from the self-interactions of the DsRed attached to the protein of interest, particularly with fusions that naturally form oligomers or biopolymers. However, combinations of mutations were found that result in the dimeric red fluorescent protein dimer2 and the monomeric red fluorescent protein mRFP1 (Fig. 5) (51). Unfortunately, mRFP1 has decreased

Fluorescent Proteins for Cell Biology

63

Fig. 5. Excitation spectrum (solid lines), emission spectrum (dashed lines), and chromophore (drawing) of mRFP1.

brightness and reduced photostability relative to DsRed, both of which limit its utility. Rather than undergo the effort to monomerize other fluorescent molecules, oligomerization problems were avoided by using mRFP1 as starting material to improve on and further develop new spectral variants (55). In addition to random mutagenesis experiments, the similarity between the overall structure of avGFP (11,12) and the DsRed structure (47,48) led to sitedirected mutagenesis of mRFP1 to produce molecules with the desired characteristics. For example, mRFP1 has decreased fluorescence when fused to the C terminus of a protein of interest (55). Since the avGFP does not seem to be similarly affected, the N terminus of avGFP was used to replace that of mRFP1, and the C terminus of avGFP was added to the C terminus of mRFP1. The result was a molecule that more efficiently develops red fluorescence when tagged to a polymerizing molecule, such as tubulin (55). In addition, because it was established that mutations at the S65 position in avGFP improve its maturation into the fluorescent form (10), the equivalent residue (Q66) was targeted within the mRFP1 chromophore. This approach produced molecules containing a Q66M mutation that exhibited more efficient chromophore maturation and provided the starting material for subsequent rounds of random and directed mutagenesis (55). One of the final results of this approach was the discovery of mCherry, one of the most advanced red fluorescent proteins yet developed. mCherry is only slightly brighter and slightly red-shifted compared with mRFP1, but it is 10-fold more photostable under imaging. 3.3. Spectral Variations in mRFP1 The knowledge and experience gained with avGFP proved most useful for altering the spectra of mRFP1. Residue Y67 of mRFP1 (equivalent to Y66 in

64

Patterson

avGFP) is the aromatic amino acid that is one of the key residues in defining a fluorescent molecule’s spectral characteristics. Changing this amino acid to histidine or tryptophan was known to blue-shift the GFP spectra to produce the BFP and CFP variants, respectively (13). A similar substitution in mRFP1, Y67W, produced the mHoneydew variant, a molecule with the expected blueshifted spectra (55) (Table 2). Other blue-shifted variants were produced by targeting another key amino acid within the chromophore, residue 66. In this case, the M66 mutant produced as a precursor to mCherry was converted into a cysteine or threonine and further mutated to eventually produce the mTangerine and mBanana or the mOrange and mStrawberry derivatives, respectively (55) (Table 2). The other residues modified in this process are predicted to surround the chromophore and change its local environment. The mRFP1 family was red-shifted even more toward the infrared region of the spectrum with the introduction of mPlum and mRaspberry (56). Rather than relying on random mutagenesis and screening bacteria colonies, the coding region for the mRFP1.2 variant was expressed in B lymphocytes, which use somatic hypermutation to alter and improve immunoglobins. The mRFP1 gene introduced into these cells was mutated by somatic hypermutation with over 23 rounds of cell selection employed to achieve far-red-shifted fluorescence. The mRaspberry mutant has excitation at 598 nm and emission at 625 nm and a maturation half-time of approx 55 min. mPlum has an excitation peak at 590 nm and emission at 649 nm, which is the most red-shifted emission peak reported for a fluorescent protein with the exception of the tetrameric AQ143 at 655 nm (57) (Table 3). mPlum matures into a fluorescent molecule with a half-time of approx 100 min and is 30-fold more photostable than mRFP1 (56). Based entirely on the extinction coefficients and quantum yields, the mHoneydew, mBanana, mTangerine, and mPlum variants are less bright than mRFP1. The mOrange derivative has a quantum efficiency and extinction coefficient comparable to those of the avGFP fluorescent proteins that are commonly used, such as EGFP, EYFP, Citrine (41), and Venus (42). However, mOrange displays a moderate sensitivity to pH (pKa approx 6.5) and matures slowly (t1/2 approx 2.5 h). The rest of the fruit-named molecules mature within a reasonable period (t1/2 ≤ 1 h) but display varying degrees of photostability. Of the monomeric forms, mCherry (approx 10-fold better than mRFP1) and mPlum (approx 30-fold better than mRFP1.2) are the most photostable. 4. Fluorescent Proteins from Other Organisms The remarkably diverse array of proteins discussed above has been derived from just two sources, the Aequorea victoria jellyfish and the Discosoma coral; they represent only the beginning of fluorescent protein technology. The past 5 years have witnessed a marvelous expansion of new fluorescent proteins derived

Fluorescent Proteins for Cell Biology

65

from an array of marine species representing three different classes, Copepoda, Hydrozoa, and Anthozoa. The evolution and diversity of fluorescent proteins have been previously reviewed (45,58). A listing of molecules from more than 20 species is included in Table 3, and these span almost the entire visible spectral region, with emission wavelengths ranging from 477 to 655 nm. Unfortunately, published accounts of their use are limited, but given the broad diversity of spectral characteristics, these molecules offer much potential in cell and developmental biology experiments. Readers should be aware that many of these fluorescent proteins have very low fluorescence (quantum yields < 0.1) and/or tend to oligomerize. Both characteristics decrease the utility of these molecules for trafficking and/or localization experiments. On the other hand, several reported monomeric molecules such as mAG (59), phiYFP (58), and mKO (60), provide alternatives for green, yellow, and red fluorescent proteins. 5. Photoactivatable Fluorescent Proteins In most cases, the battery of fluorescent proteins so far discovered has proved quite suitable for studying gene expression, for observing localization of proteins, and for monitoring the dynamics of organelles and cells; however, the dynamics of localized proteins within a population may remain unresolved (61). For this, the steady-state fluorescence profile must be altered such that a subpopulation of the fluorescent protein chimeras is highlighted. Photobleaching (excitation-induced photodestruction) is generally used to highlight a population of molecules indirectly, but advances and discoveries have been made for directly highlighting a pool of molecules using “photoactivatable” or “optical highlighter” fluorescent proteins (62) (Table 4). Molecules placed in the photoactivatable category have the general characteristic that they are initially dark at the activated fluorescence wavelength but upon activation display an increase in fluorescence and are “highlighted” over a darker background. At the writing of this chapter, proteins derived from nine species have been reported (Table 4). The optical highlighters have been subclassed based on the spectral characteristics of their activated and nonactivated states (green vs red fluorescence) and on the reversibility of the activated state (irreversible vs reversible). An overview of these molecules is given below, and interested readers are directed to a review concentrating on this category of fluorescent proteins for further information (62). 5.1. Aequorea victoria GFP One early approach to develop a photoactivatable fluorescent protein relied on the fact that several Aequorea victoria variants, wtGFP, GFPmut1, -2, and -3 (20), S65T (10), I167T (13), and GFPuv (63), convert into red fluorescent

66 V123T/T158H

mEosFP

L40A, L61V, Y95F, N121K, M123T, Y188A, S199G, G213A

T158H

d2EosFP

DRONPA Dendra

V123T

d1EosFP

Kikume Green-Red (KikGR) PAmRFP1-1 PAmRFP1-2 PAmRFP1-3 EosFP S146H, I161V, I197H S146H, I161C, I197H S146H, I161S, I197H

V163A, T203H

PA-GFP

Kaede

Amino acid substitutions

Protein 400 (Pre) 504 (Post) 508 (Pre) 572 (Post) 507 (Pre) 583 (Post) 578 (Post) 578 (Post) 578 (Post) 506 (Pre) 571 (Post) 505 (Pre) 571 (Post) 506 (Pre) 569 (Post) 505 (Pre) 569 (Post) 503 (Post) 486 (Pre) 558 (Post)

λ ex 515 517 518 582 517 593 605 605 605 516 581 516 581 516 581 516 581 518 505 575

λ em

Wavelengths (nm) a

Table 4 Selected Optical Highlighter Proteins (Photoactivatable Fluorescent Proteins)

Pectiniidae spp. Dendronephthya sp.

Lobophyllia hemprichii

Lobophyllia hemprichii

Lobophyllia hemprichii

Discosoma sp. Discosoma sp. Discosoma sp. Lobophyllia hemprichii

Favia favus

Trachyphyllia geoffroyi

Aequorea victoria

Organism

83 76

75

75

75

72 72 72 75

74

73

40

Reference

66 Patterson

a(Pre)

A148G G222E T62A, N121S, H148T, K158R, I167V, E172K, F221L, G222E, K238Q T62A, S108T, N121S, H148T, M153V, T154A, K158R, I167V, E172K, F221L, G222E, K238Q

L40A, L61V, Y95F, N121K, M123T, Y188A, S199G, G213A, A224V

400 (Pre) 490 (Post)

580 (Post) 480 402 (Pre) 490 (Post)

486 (Pre) 558 (Post)

470 511

600 505 468 511

505 575

Aequorea coerulescens

Anemonia sulcata Anemonia sulcata Aequorea coerulescens Aequorea coerulescens

Dendronephthya sp.

represents the major peak prior to photoactivation, and (Post) represents the major peak after photoactivation.

PS-CFP2

asFP595 KFP1 aceGFP PS-CFP

Dendra2

80 82 78 77

Fluorescent Proteins for Cell Biology

67

67

68

Patterson

species upon irradiation with 488-nm light (64,65). This “photoactivation” produced high contrast with background fluorescence and worked well for monitoring protein diffusion in bacteria (64). However, the widespread use of this phenomenon in cell biology has been limited by the low oxygen conditions required. An approach suitable for aerobic conditions exploited the photo-induced conversion (photoconversion) that occurs within the wild-type avGFP (66). The wtGFP chromophore population is thought to exist as a mixture of neutral phenols (Y66 is protonated) and anionic phenolates (Y66 is deprotonated), giving rise to a major absorbance peak at approx 397 nm and a minor absorbance peak at 475 nm, respectively. Upon irradiation, the chromophore undergoes a proton transfer and photoconverts into the anionic form (22,30,67). The threonine at the 203 position rotates, and the hydroxyl group could help stabilize the anionic chromophore (68–70). In addition, the glutamic acid at position 222 (E222), which is in close proximity to the chromophore, is decarboxylated (70). Either or both of these structural changes may play key roles in stabilizing the shift of the chromophore population from a predominantly neutral species to an anionic species. The resulting absorbance increase at the minor peak leads to an increase in the fluorescence intensity when excited at this wavelength. Unfortunately, owing to the high initial background, photoconversion of the wild-type avGFP results in only a modest (about threefold) increase in fluorescence (22,40,66). The T203I mutant of avGFP produces a variant that has predominantly neutral phenol chromophore, which reduces the minor absorbance peak while maintaining the major peak (13,35). Further mutagenesis of the T203 position uncovered several other substitutions that also reduce the minor peak and retain the major peak at approx 400 nm (40). Surprisingly, many of these mutants, including T203I, maintain the ability to undergo photoconversion. However, the T203H mutant (denoted PA-GFP) was found to be exceptional, with the minor absorbance peak drastically decreased, giving more than 60- and more than 100-fold fluorescence increases after photoactivation in cells and in vitro, respectively (40). 5.2. DsRed Fluorescent Protein The DsRed protein has been used as a photoactivatable fluorescent protein in two different approaches. In the first case, the molecules of the DsRed tetramer exist as mixed populations of immature green fluorescent molecules and mature red fluorescent molecules. Since the protomers are in such close proximity and the emission spectrum of the green species overlaps with that of the red species, the energy of the excited green molecules can transfer nonradiatively to the red molecules. Selective photobleaching of the red species dequenches

Fluorescent Proteins for Cell Biology

69

the green fluorescence and leads to an approx 2.4-fold increase in green fluorescence emission (71). The second approach involves conversion of the monomeric version of DsRed, mRFP1, into a series of photoactivatable fluorescent proteins, PAmRFP1-1, PAmRFP1-2, and PAmRFP1-3 (72). The brightest of these, PAmRFP1-1, has a quantum yield of only 0.08, but produces an approx 70-fold increase in red fluorescence upon UV light excitation (72). Although this molecule is significantly less bright than other red photoactivatable fluorescent proteins, its derivation from mRFP1 suggests it may be used as a less perturbative protein marker. In addition, the lack of a green fluorescent component before or after activation renders it more amenable for use in multifluorescent protein experiments. 5.3. Green-to-Red Photoconversions The stony coral, Trachyphyllia geoffroyi, produces the protein Kaede, which can be photoactivated by irradiation at approx 400 nm (73). Prior to photoactivation, Kaede has a major absorbance peak at 508 nm and emission at 518 nm. After photoactivation, Kaede exhibits a new red-shifted absorbance peak at 572 nm, which upon excitation fluoresces with a new emission peak at 582 nm. This shift in both the excitation and emission peaks results in a more than 2000 fold increase in the red-to-green fluorescence ratio. Unfortunately, Kaede forms tetramers (73), which limit its usefulness as a protein trafficking tool. Nevertheless, as with the DsRed protein, the self-association properties can perhaps be suppressed (51), and Kaede’s large contrast with background after photoactivation gives it enormous potential as a photoactivatable fluorescent protein marker. Another coral fluorescent protein, KikG from Favia favus (74), was developed as a photoactivatable marker. Although the original KikG from did not exhibit photoactivatable properties, using information learned from their structural characterizations of Kaede, Miyawaki and colleagues engineered KikG into the KikGR variant, which produced a more than 2000-fold increase in fluorescence contrast during ratio imaging of the red and green components after photoactivation. The authors found that the purified KikGR protein lacked the brightness of Kaede in vitro but exhibited several fold more fluorescence than Kaede when expressed in cells. It was unclear whether this difference reflected protein expression levels or protein folding efficiency. As with Kaede, KikGR was also found to be a tetramer (74). The fluorescent protein EosFP from the stony coral Lobophyllia hemprichii also exhibits a green-to-red fluorescence photoconversion upon UV or near UV light irradiation (75). Similar to Kaede, EosFP has a preactivated excitation maximum at 506 nm with emission at 516 nm. Upon activation at 405 nm, the photoactivated excitation peak is located at 571 nm with emission at 581 nm. Initially

70

Patterson

determined to be a tetramer, EosFP was engineered into two dimeric forms, d1EosFP and d2EosFP. The combination of these mutations produced a monomeric molecule with a Kd of approx 0.1 mM, which has been named mEosFP. The emission maxima of these mutants remain constant whereas the excitation maxima and brightness change slightly (Table 4). Thus it would seem obvious that the mEosFP, owing to its decreased self-association, is the protein of choice in this subclass. However, mEosFP inefficiently forms a fluorescent molecule when expressed at 37°C (75). A recent addition to this category of optical highlighters is Dendra, which exhibits up to 4500-fold photoconversion from its green-to-red fluorescent forms (76). The wild-type form, isolated from Dendronephthya sp. and named dendGFP (45), was engineered into the monomeric Dendra with its photoactivation properties. Dendra can be activated with approx 400 nm light, which is the wavelength required by most of the other photoactivatable fluorescent proteins, but allows the option of activation with high levels (>0.5 W/cm2) of potentially less phototoxic wavelengths (approx 488 nm). It was noted that excitation of the unactivated green state of Dendra with lower levels of 488-nm light (<50 mW/cm2) did not produce photoconversion to the red state (76). Although Dendra is reported to develop fluorescence efficiently when expressed at 37°C, a mutated version, Dendra2, commercially available from Evrogen, represents an improvement in this characteristic. 5.4. Cyan-to-Green Photoconversion Similar to the green-to-red converting molecules discussed above, photoswitchable (PS)-CFP displays a spectral shift of both absorbance and emission, but this shift involves a change from a cyan-to-green fluorescent protein (77). Derived from a nonfluorescent variant, acGFPL (78), PS-CFP initially displays excitation at 402 nm and emission at 468 nm (77). Upon activation with approx 400-nm laser light, the protein displays new peaks at 490 and 511 nm, respectively, leading to an approx 1500-fold increase in the green-to-cyan fluorescence ratio. PS-CFP is reported to be a monomeric protein, and the diverse possibilities for its use have been demonstrated by imaging PS-CFP-tagged actin and dopamine transporter (77). An improved version, PS-CFP2, which develops fluorescence more efficiently and is thus brighter, is now available (Table 4). 5.5. Reversible This category displays the fascinating capability of being switched “on” and switched “off” by excitation at different wavelengths. The on-off switching has been observed in single molecules of avGFP T203 mutants (79), but reversible photoactivatable fluorescent proteins exhibit this behavior in a bulk popu-

Fluorescent Proteins for Cell Biology

71

lation. In the first observation of this phenomenon, the red fluorescence (lmax approx 595 nm) of asFP595, a protein isolated from the sea anemone, Anemonia sulcata, was found to be enhanced by exposure to green light and quenched by exposure to blue light (80). Initially, asFP595 found little practical use as a marker since it matures slowly, exists as a tetramer, and has low quantum yields. Recently, however, the reversible capabilities of asFP595 have been utilized in a new fluorescence microscopy technique to break the diffraction barrier (81). Therefore, it is anticipated that there will be renewed interest in this molecule despite its deficiencies as a protein localization and trafficking marker. Another variant of asFP595, KFP1, has been introduced as a photoactivatable fluorescent protein that has the capability to be reversibly or irreversibly photoactivated (82). KFP1 initially exhibits little fluorescence when excited at approx 560 nm, but upon activation with 532-nm laser light increases its red fluorescence by approx 30-fold. The reversion of KFP1 depends on the photoactivation excitation intensity. Lower level excitation (approx 1 W/cm2 for 2 min) produced red KFP1 fluorescence that relaxed to the nonfluorescent state with a half-time of approx 50 s, whereas 200 times this excitation (approx 20 W/cm2 for 20 min) irreversibly photoactivated approx 50% of the population. With a quantum yield of 0.07, the photoactivated KFP1 lacks the brightness of many of the other photoactivatable fluorescent proteins. In addition, its tetrameric oligomerization state and slow maturation time (t1/2 approx 5 h) will limit its use as a marker. However, the use of the potentially less phototoxic green activation light (532 nm), instead of the approx 400-nm light required by most other photoactivatable fluorescent proteins, makes KFP1 attractive for use in live cell imaging experiments (82). Perhaps the most intriguing of the reversible PA-FPs is a protein derived from Pectimiidae named Dronpa (83). This molecule initially displays green fluorescence with an excitation maximum at 503 nm and emission maximum of 518 nm. Upon intense irradiation at 490 nm (0.4 W/cm2), the absorbance at 503 nm as well as the green fluorescence emission is rapidly lost. However, upon irradiation at 400 nm (0.14 W/cm2), the 503-nm absorbance as well as the green fluorescence emission is immediately restored. Remarkably, the authors found that Dronpa could undergo the on-off cycling 100 times with a loss of only 25% of the original fluorescence. The molecule is monomeric and has an extinction coefficient (95,000 M−1/cm−1) and a quantum yield (0.85) that make it one of the brightest fluorescent proteins available. Its photoactivation capabilities should make Dronpa useful as a marker for any photoactivation experiment, but the reversible nature of the fluorescence should allow the same photoactivation experiment to be repeated multiple times within the same region of interest (83).

72

Patterson

5.6. Why Use a Photoactivatable Fluorescent Protein? The common characteristic of all photoactivatable fluorescent proteins is that they are observed as fluorescent signals over darker backgrounds. Thus, subpopulations of proteins, organelles, or cells can be highlighted and their dynamics monitored temporally and spatially within the entire population. However, a major advantage with photoactivation is the circumvention of new protein synthesis. Unlike photobleaching techniques, the fluorescence of newly synthesized proteins will not contaminate the photoactivated protein signal. Not only does this characteristic simplify interpretation of trafficking experiments previously described, it introduces a new approach to monitor protein turnover. In addition to monitoring degradation of a protein by radioactive pulse labeling, brief activations of a tagged protein of interest can be employed as a fluorescence pulse label. Although it should only be considered complementary to biochemical pulse labeling, such a technique has several desirable properties for the study of protein degradation. The temporal resolution of fluorescence pulse labeling is essentially limited by the instrument parameters, has subcellular spatial resolution, and lacks the expense associated with radioactivity. 6. Considerations for Using Fluorescent Proteins The vast array of fluorescent proteins currently available offers tools for numerous applications from simply localizing proteins of interest or monitoring gene expression to observing specific interactions between molecules within defined regions of a living cell. Researchers are now left with the daunting task of choosing between more than 100 different varieties of fluorescent proteins, since testing each is generally not feasible. Optimally, researchers could rely on the fluorescent protein experts to categorize the variants, determine which has the most favorable characteristics, and simply tell us which one to use. Although this does work to a limited degree for preliminary work, experimenters eventually wish to push the boundaries of their experiments to image fewer molecules with higher signal-to-background ratios for longer periods in every conceivable region of a cell and/or organism. Unfortunately, a single fluorescent protein that is optimal for all these functions has not been reported. Therefore, the investigator must tailor each fluorescent protein to meet his/her needs. Below are some general fluorescent protein characteristics discussed throughout this chapter of which researchers should be aware when choosing a fluorescent protein for their work; readers are also directed to a review of this topic by Tsien and colleagues (84). First to be considered is how low do you want to go? The amount of fluorescence is dependent on the intrinsic brightness of the molecule and the number of those molecules available for observation. An estimate of the amount of wild-

Fluorescent Proteins for Cell Biology

73

type avGFP required for observation over the autofluorescence of a cell was determined to be approximately 1 µM. Significant improvements in the GFP fluorescence have pushed this level even lower and will perhaps reach the singlemolecule level. Red fluorescent proteins, although generally less bright than their green counterparts, may enable lower amounts of fluorescence to be detected, since autofluorescence is generally lower toward the red end of the visible spectrum. Second, where is the protein to be localized? For genetically encoded fluorescent proteins, the brightness can be subject to dramatic changes. For instance, the cellular environment (acidic vs basic; oxidizing vs reducing, and so on) can profoundly affect fluorescence. Targeting fluorescent proteins to acidic or degradative environments can diminish fluorescence immensely. For this, molecules that are less sensitive to the environment will be critical. Third, how quickly do you need to observe the molecule after synthesis? Unlike organic dyes, fluorescent proteins are synthesized by the cell and, as a consequence, must develop into observable fluorescent molecules and do so with varying kinetics and efficiency. In general, this may be of little consequence, but if the protein of interest is short-lived, the observation window may be very narrow. In this case, a fluorescent protein that matures quickly will be necessary. How forgiving is your molecule in allowing addition of genetic tags? Generally, protein fusions that result in altered protein behavior are not reported, and investigators simply have to test for themselves. Nevertheless, useful constructions can be produced under more favorable conditions by relying on knowledge gained with other tagged proteins. For instance, tagging the N terminus vs the C terminus for proteins of the same family may produce dramatically different results. Fusions of transmembrane proteins and cytosolic proteins may not work well with the same fluorescent protein. In addition, the oligomeric state of the protein of interest can be a deciding factor as well. For instance, if the fluorescent protein self-associates, this can alter degradation, localization, and/or trafficking of the protein of interest, as well as potentially creating a toxic situation for the cell. Finally, the stability of the molecule under imaging can limit the length and sensitivity of experiments but is often overlooked as a critical characteristic. Photostability has been reported in different manners, including photobleaching quantum yields and average number of photons emitted before photodestruction. However, these quantifications usually mean little to many investigators except when compared with other variants of fluorescent proteins. A convenient comparison between molecules is determination of the photobleaching halflives under normalized excitation, and these are reported in many papers referenced in this chapter.

74

Patterson

7. Concluding Remarks and Future Advances A potential long-term goal in imaging is to monitor the behavior of individual molecules of interest within the living cells of an organism. Most of this chapter is devoted to recounting advances in the development of fluorescent proteins, and these are likely to play a key role in reaching this goal. On the other hand, numerous developments in instrumentation, such as stimulated emission depletion microscopy (STED) (85), have advanced the resolution of light microscopy with existing fluorophores beyond the diffraction limit. In trying to predict the advancement of any scientific technique, it is helpful to note the significant milestones in reaching a particular goal. For instance, the fluorescent protein technology discussed earlier is less than less than 15 yr old, and single-fluorescent molecule imaging is less than 15 yr old (86). Thus, if advances on the sample side (i.e., fluorescent proteins, labels, and so on) as well as the instrument side continue at the same rate, molecular fluorescence imaging in vivo is no longer just a microscopist’s fanciful dream but an actual goal that may eventually come to realization. References 1. Prasher, D. C., Eckenrode, V. K., Ward, W. W., Prendergast, F. G., and Cormier, M. J. (1992) Primary structure of the Aequorea victoria green-fluorescent protein. Gene 111, 229–233. 2. Chalfie, M., Tu, Y., Euskirchen, G., Ward, W. W., and Prasher, D. C. (1994) Green fluorescent protein as a marker for gene expression. Science 263, 802–805. 3. Morise, H., Shimomura, O., Johnson, F. H., and Winant, J. (1974) Intermolecular energy transfer in the bioluminescent system of Aequorea. Biochemistry 13, 2656– 2662. 4. Bokman, S. H. and Ward, W. W. (1981) Renaturation of Aequorea green-fluorescent protein. Biochem. Biophys. Res. Commun. 101, 1372–1380. 5. Ward, W. W., Cody, C. W., Hart, R. C., and Cormier, M. J. (1980) Spectrophotometric identity of the energy transfer chromophores in Renilla and Aequorea greenfluorescent proteins. Photochem. Photobiol. 31, 611–615. 6. Ward, W. W. and Bokman, S. H. (1982) Reversible denaturation of Aequorea green-fluorescent protein: physical separation and characterization of the renatured protein. Biochemistry 21, 4535–4540. 7. Ward, W. W., Prentice, H. J., Roth, A. F., Cody, C. W., and Reeves, S. C. (1982) Spectral perturbations of the Aequorea green-fluorescent protein. Photochem. Photobiol. 35, 803–808. 8. Cody, C. W., Prasher, D. C., Westler, W. M., Prendergast, F. G., and Ward, W. W. (1993) Chemical structure of the hexapeptide chromophore of the Aequorea greenfluorescent protein. Biochemistry 32, 1212–1218.

Fluorescent Proteins for Cell Biology

75

9. Inouye, S. and Tsuji, F. I. (1994) Aequorea green fluorescent protein. Expression of the gene and fluorescence characteristics of the recombinant protein. FEBS Lett. 341, 277–280. 10. Heim, R., Cubitt, A. B., and Tsien, R. Y. (1995) Improved green fluorescence. Nature 373, 663–664. 11. Yang, F., Moss, L. G., and Phillips, G. N. Jr. (1996) The molecular structure of green fluorescent protein. Nat. Biotechnol. 14, 1246–1251. 12. Ormo, M., Cubitt, A. B., Kallio, K., Gross, L. A., Tsien, R. Y., and Remington, S. J. (1996) Crystal structure of the Aequorea victoria green fluorescent protein. Science 273, 1392–1395. 13. Heim, R., Prasher, D. C., and Tsien, R. Y. (1994) Wavelength mutations and posttranslational autoxidation of green fluorescent protein. Proc. Natl. Acad. Sci. USA 91, 12,501–12,504. 14. Reid, B. G. and Flynn, G. C. (1997) Chromophore formation in green fluorescent protein. Biochemistry 36, 6786–6791. 15. Prendergast, F. G. (1999) Biophysics of the green fluorescent protein, in Green Fluorescent Proteins, vol. 58, (Sullivan, K. F. and Kay, S. A., eds.), Academic Press, San Diego, pp. 1–18. 16. Barondeau, D. P., Putnam, C. D., Kassmann, C. J., Tainer, J. A., and Getzoff, E. D. (2003) Mechanism and energetics of green fluorescent protein chromophore synthesis revealed by trapped intermediate structures. Proc. Natl. Acad. Sci. USA 100, 12,111–12,116. 17. Branchini, B. R., Nemser, A. R., and Zimmer, M. (1998) A computational analysis of the unique protein-induced tight turn that results in posttranslational chromophore formation in green fluorescent protein. J. Am. Chem. Soc. 120, 1–6. 18. Barondeau, D. P., Kassmann, C. J., Tainer, J. A., and Getzoff, E. D. (2005) Understanding GFP chromophore biosynthesis: controlling backbone cyclization and modifying post-translational chemistry. Biochemistry 44, 1960–1970. 19. Rosenow, M. A., Huffman, H. A., Phail, M. E., and Wachter, R. M. (2004) The crystal structure of the Y66L variant of green fluorescent protein supports a cyclization-oxidation-dehydration mechanism for chromophore maturation. Biochemistry 43, 4464–4472. 20. Cormack, B. P., Valdivia, R. H., and Falkow, S. (1996) FACS-optimized mutants of the green fluorescent protein (GFP). Gene 173, 33–38. 21. Siemering, K. R., Golbik, R., Sever, R., and Haseloff, J. (1996) Mutations that suppress the thermosensitivity of green fluorescent protein. Curr. Biol. 6, 1653–1663. 22. Tsien, R. Y. (1998) The green fluorescent protein. Annu. Rev. Biochem. 67, 509– 544. 23. Pedelacq, J. D., Cabantous, S., Tran, T., Terwilliger, T. C., and Waldo, G. S. (2006) Engineering and characterization of a superfolder green fluorescent protein. Nat. Biotechnol. 24, 79–88. 24. Chiu, W., Niwa, Y., Zeng, W., Hirano, T., Kobayashi, H., and Sheen, J. (1996) Engineered GFP as a vital reporter in plants. Curr. Biol. 6, 325–330.

76

Patterson

25. Haseloff, J., Siemering, K. R., Prasher, D. C., and Hodge, S. (1997) Removal of a cryptic intron and subcellular localization of green fluorescent protein are required to mark transgenic Arabidopsis plants brightly. Proc. Natl. Acad. Sci. USA 94, 2122–2127. 26. Cormack, B. P., Bertram, G., Egerton, M., Gow, N. A., Falkow, S., and Brown, A. J. (1997) Yeast-enhanced green fluorescent protein (yEGFP)a reporter of gene expression in Candida albicans. Microbiology 143(Pt 2), 303–311. 27. Haas, J., Park, E. C., and Seed, B. (1996) Codon usage limitation in the expression of HIV-1 envelope glycoprotein. Curr. Biol. 6, 315–324. 28. Zacharias, D. A., Violin, J. D., Newton, A. C., and Tsien, R. Y. (2002) Partitioning of lipid-modified monomeric GFPs into membrane microdomains of live cells. Science 296, 913–916. 29. Zhang, J., Campbell, R. E., Ting, A. Y., and Tsien, R. Y. (2002) Creating new fluorescent probes for cell biology. Nat. Rev. Mol. Cell. Biol. 3, 906–918. 30. Chattoraj, M., King, B. A., Bublitz, G. U., and Boxer, S. G. (1996) Ultra-fast excited state dynamics in green fluorescent protein: multiple states and proton transfer. Proc. Natl. Acad. Sci. USA 93, 8362–8367. 31. Patterson, G. H., Knobel, S. M., Sharif, W. D., Kain, S. R., and Piston, D. W. (1997) Use of the green fluorescent protein and its mutants in quantitative fluorescence microscopy. Biophys. J. 73, 2782–2790. 32. Delagrave, S., Hawtin, R. E., Silva, C. M., Yang, M. M., and Youvan, D. C. (1995) Red-shifted excitation mutants of the green fluorescent protein. Biotechnology (N.Y.) 13, 151–154. 33. Yang, T. T., Cheng, L., and Kain, S. R. (1996) Optimized codon usage and chromophore mutations provide enhanced sensitivity with the green fluorescent protein. Nucleic Acids Res. 24, 4592–4593. 34. Cubitt, A. B., Woollenweber, L. A., and Heim, R. (1997) Understanding structure-function relationships in the Aequorea victoria green fluorescent protein, in Green Fluorescent Proteins, vol. 58, (Sullivan, K. F. and Kay, S. A., eds.), Academic Press, San Diego, pp. 19–30. 35. Ehrig, T., O’Kane, D. J., and Prendergast, F. G. (1995) Green-fluorescent protein mutants with altered fluorescence excitation spectra. FEBS Lett. 367, 163–166. 36. Heim, R. and Tsien, R. Y. (1996) Engineering green fluorescent protein for improved brightness, longer wavelengths and fluorescence resonance energy transfer. Curr. Biol. 6, 178–182. 37. Ellenberg, J., Lippincott-Schwartz, J., and Presley, J. F. (1998) Two-color green fluorescent protein time-lapse imaging. Biotechniques 25, 838–842, 844–846. 38. Harms, G. S., Cognet, L., Lommerse, P. H., Blab, G. A., and Schmidt, T. (2001) Autofluorescent proteins in single-molecule research: applications to live cell imaging microscopy. Biophys. J. 80, 2396–2408. 39. Rizzo, M. A., Springer, G. H., Granada, B., and Piston, D. W. (2004) An improved cyan fluorescent protein variant useful for FRET. Nat. Biotechnol. 22, 445–449.

Fluorescent Proteins for Cell Biology

77

40. Patterson, G. H. and Lippincott-Schwartz, J. (2002) A photoactivatable GFP for selective photolabeling of proteins and cells. Science 297, 1873–1877. 41. Griesbeck, O., Baird, G. S., Campbell, R. E., Zacharias, D. A., and Tsien, R. Y. (2001) Reducing the environmental sensitivity of yellow fluorescent protein. Mechanism and applications. J. Biol. Chem. 276, 29,188–29,194. 42. Nagai, T., Ibata, K., Park, E. S., Kubota, M., Mikoshiba, K., and Miyawaki, A. (2002) A variant of yellow fluorescent protein with fast and efficient maturation for cell-biological applications. Nat. Biotechnol. 20, 87–90. 43. Wachter, R. M. and Remington, S. J. (1999) Sensitivity of the yellow variant of green fluorescent protein to halides and nitrate. Curr. Biol. 9, R628–R629. 44. Wachter, R. M., Yarbrough, D., Kallio, K., and Remington, S. J. (2000) Crystallographic and energetic analysis of binding of selected anions to the yellow variants of green fluorescent protein. J. Mol. Biol. 301, 157–171. 45. Labas, Y. A., Gurskaya, N. G., Yanushevich, Y. G., et al. (2002) Diversity and evolution of the green fluorescent protein family. Proc. Natl. Acad. Sci. USA 99, 4256–4261. 46. Matz, M. V., Fradkov, A. F., Labas, Y. A., et al. (1999) Fluorescent proteins from nonbioluminescent Anthozoa species. Nat. Biotechnol. 17, 969–973. 47. Wall, M. A., Socolich, M., and Ranganathan, R. (2000) The structural basis for red fluorescence in the tetrameric GFP homolog DsRed. Nat. Struct. Biol. 7, 1133– 1138. 48. Yarbrough, D., Wachter, R. M., Kallio, K., Matz, M. V., and Remington, S. J. (2001) Refined crystal structure of DsRed, a red fluorescent protein from coral, at 2.0-A resolution. Proc. Natl. Acad. Sci. USA 98, 462–467. 49. Gross, L. A., Baird, G. S., Hoffman, R. C., Baldridge, K. K., and Tsien, R. Y. (2000) The structure of the chromophore within DsRed, a red fluorescent protein from coral. Proc. Natl. Acad. Sci. USA 97, 11,990–11,995. 50. Baird, G. S., Zacharias, D. A., and Tsien, R. Y. (2000) Biochemistry, mutagenesis, and oligomerization of DsRed, a red fluorescent protein from coral. Proc. Natl. Acad. Sci. USA 97, 11,984–11,989. 51. Campbell, R. E., Tour, O., Palmer, A. E., et al. (2002) A monomeric red fluorescent protein. Proc. Natl. Acad. Sci. USA 99, 7877–7882. 52. Terskikh, A., Fradkov, A., Ermakova, G., et al. (2000) “Fluorescent timer”: protein that changes color with time. Science 290, 1585–1588. 53. Terskikh, A. V., Fradkov, A. F., Zaraisky, A. G., Kajava, A. V., and Angres, B. (2002) Analysis of DsRed mutants. Space around the fluorophore accelerates fluorescence development. J. Biol. Chem. 277, 7633–7636. 54. Bevis, B. J. and Glick, B. S. (2002) Rapidly maturing variants of the Discosoma red fluorescent protein (DsRed). Nat. Biotechnol. 20, 83–87. 55. Shaner, N. C., Campbell, R. E., Steinbach, P. A., Giepmans, B. N., Palmer, A. E., and Tsien, R. Y. (2004) Improved monomeric red, orange and yellow fluorescent proteins derived from Discosoma sp. red fluorescent protein. Nat. Biotechnol. 22, 1567–1572.

78

Patterson

56. Wang, L., Jackson, W. C., Steinbach, P. A., and Tsien, R. Y. (2004) Evolution of new nonantibody proteins via iterative somatic hypermutation. Proc. Natl. Acad. Sci. USA 101, 16,745–16,749. 57. Shkrob, M. A., Yanushevich, Y. G., Chudakov, D. M., et al. (2005) Far-red fluorescent proteins evolved from a blue chromoprotein from Actinia equina. Biochem. J. 392, 649–654. 58. Shagin, D. A., Barsova, E. V., Yanushevich, Y. G., et al. (2004) GFP-like proteins as ubiquitous metazoan superfamily: evolution of functional features and structural complexity. Mol. Biol. Evol. 21, 841–850. 59. Karasawa, S., Araki, T., Yamamoto-Hino, M., and Miyawaki, A. (2003) A greenemitting fluorescent protein from Galaxeidae coral and its monomeric version for use in fluorescent labeling. J. Biol. Chem. 278, 34,167–34,171. 60. Karasawa, S., Araki, T., Nagai, T., Mizuno, H., and Miyawaki, A. (2004) Cyanemitting and orange-emitting fluorescent proteins as a donor/acceptor pair for fluorescence resonance energy transfer. Biochem. J. 381, 307–312. 61. Lippincott-Schwartz, J., Altan-Bonnet, N., and Patterson, G. H. (2003) Photobleaching and photoactivation: following protein dynamics in living cells. Nat. Cell. Biol. Suppl, S7–S14. 62. Lukyanov, K. A., Chudakov, D. M., Lukyanov, S., and Verkhusha, V. V. (2005) Innovation: photoactivatable fluorescent proteins. Nat. Rev. Mol. Cell. Biol. 6, 885–891. 63. Crameri, A., Whitehorn, E. A., Tate, E., and Stemmer, W. P. (1996) Improved green fluorescent protein by molecular evolution using DNA shuffling. Nat. Biotechnol. 14, 315–319. 64. Elowitz, M. B., Surette, M. G., Wolf, P. E., Stock, J., and Leibler, S. (1997) Photoactivation turns green fluorescent protein red. Curr. Biol. 7, 809–812. 65. Sawin, K. E. and Nurse, P. (1997) Photoactivation of green fluorescent protein. Curr. Biol. 7, R606–R607. 66. Yokoe, H. and Meyer, T. (1996) Spatial dynamics of GFP-tagged proteins investigated by local fluorescence enhancement. Nat. Biotechnol. 14, 1252–1256. 67. Creemers, T. M., Lock, A. J., Subramaniam, V., Jovin, T. M., and Volker, S. (1999) Three photoconvertible forms of green fluorescent protein identified by spectral hole-burning. Nat. Struct. Biol. 6, 557–560. 68. Brejc, K., Sixma, T. K., Kitts, P. A., et al. (1997) Structural basis for dual excitation and photoisomerization of the Aequorea victoria green fluorescent protein. Proc. Natl. Acad. Sci. USA 94, 2306–2311. 69. Palm, G. J., Zdanov, A., Gaitanaris, G. A., Stauber, R., Pavlakis, G. N., and Wlodawer, A. (1997) The structural basis for spectral variations in green fluorescent protein. Nat. Struct. Biol. 4, 361–365. 70. van Thor, J. J., Gensch, T., Hellingwerf, K. J., and Johnson, L. N. (2002) Phototransformation of green fluorescent protein with UV and visible light leads to decarboxylation of glutamate 222. Nat. Struct. Biol. 9, 37–41. 71. Marchant, J. S., Stutzmann, G. E., Leissring, M. A., LaFerla, F. M., and Parker, I. (2001) Multiphoton-evoked color change of DsRed as an optical highlighter for cellular and subcellular labeling. Nat. Biotechnol. 19, 645–649.

Fluorescent Proteins for Cell Biology

79

72. Verkhusha, V. V. and Sorkin, A. (2005) Conversion of the monomeric red fluorescent protein into a photoactivatable probe. Chem. Biol. 12, 279–285. 73. Ando, R., Hama, H., Yamamoto-Hino, M., Mizuno, H., and Miyawaki, A. (2002) An optical marker based on the UV-induced green-to-red photoconversion of a fluorescent protein. Proc. Natl. Acad. Sci. USA 99, 12,651–12,656. 74. Tsutsui, H., Karasawa, S., Shimizu, H., Nukina, N., and Miyawaki, A. (2005) Semi-rational engineering of a coral fluorescent protein into an efficient highlighter. EMBO Rep. 6, 233–238. 75. Wiedenmann, J., Ivanchenko, S., Oswald, F., et al. (2004) EosFP, a fluorescent marker protein with UV-inducible green-to-red fluorescence conversion. Proc. Natl. Acad. Sci. USA 101, 15,905–15,910. 76. Gurskaya, N. G., Verkhusha, V. V., Shcheglov, A. S., et al. (2006) Engineering of a monomeric green-to-red photoactivatable fluorescent protein induced by blue light. Nat. Biotechnol. 24, 461–465. 77. Chudakov, D. M., Verkhusha, V. V., Staroverov, D. B., Souslova, E. A., Lukyanov, S., and Lukyanov, K. A. (2004) Photoswitchable cyan fluorescent protein for protein tracking. Nat. Biotechnol. 22, 1435–1439. 78. Gurskaya, N. G., Fradkov, A. F., Pounkova, N. I., et al. (2003) A colourless green fluorescent protein homologue from the non-fluorescent hydromedusa Aequorea coerulescens and its fluorescent mutants. Biochem. J. 373, 403–408. 79. Dickson, R. M., Cubitt, A. B., Tsien, R. Y., and Moerner, W. E. (1997) On/off blinking and switching behaviour of single molecules of green fluorescent protein. Nature 388, 355–358. 80. Lukyanov, K. A., Fradkov, A. F., Gurskaya, N. G., et al. (2000) Natural animal coloration can be determined by a nonfluorescent green fluorescent protein homolog. J. Biol. Chem. 275, 25,879–25,882. 81. Hofmann, M., Eggeling, C., Jakobs, S., and Hell, S. W. (2005) Breaking the diffraction barrier in fluorescence microscopy at low light intensities by using reversibly photoswitchable proteins. Proc. Natl. Acad. Sci. USA 102, 17,565– 17,569. 82. Chudakov, D. M., Belousov, V. V., Zaraisky, A. G., et al. (2003) Kindling fluorescent proteins for precise in vivo photolabeling. Nat. Biotechnol. 21, 191– 194. 83. Ando, R., Mizuno, H., and Miyawaki, A. (2004) Regulated fast nucleocytoplasmic shuttling observed by reversible protein highlighting. Science 306, 1370–1373. 84. Shaner, N. C., Steinbach, P. A., and Tsien, R. Y. (2005) A guide to choosing fluorescent proteins. Nat. Methods 2, 905–909. 85. Dyba, M. and Hell, S. W. (2002) Focal spots of size lambda/23 open up far-field fluorescence microscopy at 33 nm axial resolution. Phys. Rev. Lett. 88, 163901. 86. Betzig, E. and Chichester, R. J. (1993) Single molecules observed by near-field scanning optical microscopy. Science 262, 1422–1425. 87. Cubitt, A. B., Heim, R., Adams, S. R., Boyd, A. E., Gross, L. A., and Tsien, R. Y. (1995) Understanding, improving and using green fluorescent proteins. Trends Biochem. Sci. 20, 448–455.

80

Patterson

88. Sawano, A. and Miyawaki, A. (2000) Directed evolution of green fluorescent protein by a new versatile PCR strategy for site-directed and semi-random mutagenesis. Nucleic Acids Res. 28, E78. 89. Gurskaya, N. G., Fradkov, A. F., Terskikh, A., et al. (2001) GFP-like chromoproteins as a source of far-red fluorescent proteins. FEBS Lett. 507, 16–20. 90. Bulina, M. E., Chudakov, D. M., Mudrik, N. N., and Lukyanov, K. A. (2002) Interconversion of Anthozoa GFP-like fluorescent and non-fluorescent proteins by mutagenesis. BMC Biochem. 3, 7. 91. Wiedenmann, J., Elke, C., Spindler, K. D., and Funke, W. (2000) Cracks in the beta-can: fluorescent proteins from Anemonia sulcata (Anthozoa, Actinaria). Proc. Natl. Acad. Sci. USA 97, 14,091–14,096. 92. Sun, Y., Castner, E. W. Jr., Lawson, C. L., and Falkowski, P. G. (2004) Biophysical characterization of natural and mutant fluorescent proteins cloned from zooxanthellate corals. FEBS Lett. 570, 175–183. 93. Wiedenmann, J., Schenk, A., Rocker, C., Girod, A., Spindler, K. D., and Nienhaus, G. U. (2002) A far-red fluorescent protein with fast maturation and reduced oligomerization tendency from Entacmaea quadricolor (Anthozoa, Actinaria). Proc. Natl. Acad. Sci. USA 99, 11,646–11,651. 94. Fradkov, A. F., Chen, Y., Ding, L., Barsova, E. V., Matz, M. V., and Lukyanov, S. A. (2000) Novel fluorescent protein from Discosoma coral and its mutants possesses a unique far-red fluorescence. FEBS Lett. 479, 127–130.

Detection of GFP in D. melanogaster

81

6 Detection of GFP During Nervous System Development in Drosophila melanogaster Karin Edoff, James S. Dods, and Andrea H. Brand Summary Using the vital marker GFP and its spectral variants, it is possible to visualize multiple proteins in individual cells and thereby monitor embryonic development on a cellular and molecular level. In the following chapter we describe how to prepare Drosophila embryos or larvae for live imaging or immunohistochemical staining and provide some guidelines for optimal GFP detection. Key Words: Green fluorescent protein; fluorescent proteins; nervous system development; confocal microscopy.

1. Introduction The purification of green fluorescent protein (GFP) from Aequorea victoria (1) and its subsequent cloning three decades later (2) has revolutionized cell and molecular biology. The ability to label specific proteins fluorescently in individual cells has allowed numerous studies investigating cell and molecular behaviour in vivo. Engineered point mutations in the GFP coding sequence and the continuous identification of novel fluorescent proteins from different organisms have resulted in the availability of a broad repertoire of spectral variants of fluorescent probes. These, in combination with recent imaging innovations, allow the synchronous observation of several fluorescent markers (3). Moreover, the cloning and engineering of fluorescent proteins that can be photoactivated (4–9) offers further possibilities. The decision as to which fluorescent protein is “best” for your experiment will depend on the study to be undertaken and the imaging equipment available. Parameters such as expression levels, maturation rate, brightness, photostability, oligomerization, protein

From: Methods in Molecular Biology, vol. 411: Reporter Genes: A Practical Guide Edited by: D. Anson © Humana Press Inc., Totowa, NJ

81

82

Edoff, Dods, and Brand

stability, sensitivity to pH, and separation of excitation and emission spectra should be taken into account when you are choosing a fluorescent protein (reviewed in ref. 10). This chapter focuses on the expression and detection of GFP and its variants in embryos and larvae of Drosophila melanogaster. Our lab is interested in the processes governing neuronal development in Drosophila melanogaster. Using live cell imaging or fluorescent immunohistochemical staining, it is possible to investigate the generation of diversity within the developing central nervous system (CNS) and to track cell lineages throughout development. Fluorescent proteins are used in two broad approaches: as a tracer, to follow cell lineage and behavior, and as a fusion protein to elucidate protein localization and function within a cell. Taking advantage of the GAL4 system (11–14), expression of fluorescent proteins can be targeted directly to specific neuronal lineages. This allows the expression of GFP or GFP tagged fusion proteins in spatially and temporally restricted patterns in the CNS and in specific subcellular compartments. Once GFP is expressed, its chromophore must first undergo cyclization and autooxidation before becoming actively fluorescent. Mutations have been introduced into wild-type GFP to improve maturation times, increase brightness, and shift excitation and emission maxima (15–19). Groups are now reporting the isolation and characterization of GFP-like proteins from other Cnidaria classes (Hydrozoa and Anthoza [6], and, more recently from more evolutionarily distant marine organisms of the Arthropoda class [20]), offering an astounding choice for the fluorescent microscopist (reviewed in ref. 21). Table 1 lists some examples of GFP variants and other fluorescent proteins and their respective excitation and emission wavelengths (adapted from ref. 10). GFP expression in Drosophila embryos is most often detected using laser scanning and spinning disc confocal microscopy owing to the higher resolution this offers over conventional epifluorescent techniques. Confocal microscopy allows optical sectioning of a 3D sample as well as the removal of out-offocus light. By combining a series of optical sections (a z stack), it is possible to create a 3D reconstruction of the sample that can be rotated or tilted. Most imaging software will also allow imaging over time of a single focal plane and 4D imaging, whereby z stacks are taken at regular intervals over time. These data can then be combined to make movies that represent 3D conformational changes over time, allowing direct visualization of complex cell or tissue behavior. Deconvolution algorithms are increasingly being used to complement and enhance confocal data. Deconvolution software uses a computational approach to remove out-of-focus light from each optical slice to enhance image resolution. The choice between laser point scanning and spinning disc confocal microscopy depends on the biological phenomenon being investigated. Laser scan-

Detection of GFP in D. melanogaster

83

Table 1 Examples of GFP Variants and Other Fluorescent Proteins and Their Respective Excitation and Emission Wavelengths Fluorescent protein

Excitation max (nm)

Blue fluorescent proteins wt GFP T-Sapphire Cyan fluorescent proteins ECFP mCFP Cerulean CyPet Green fluorescent proteins EGFP GFP-S65T Emerald Azami Green Yellow fluorescent proteins EYFP Venus mCitrine YPet Orange and red fluorescent proteins Kusabira Orange mOrange dTomato-Tandem DsRed-Monomer mStrawberry mRFP1 mCherry mPlum

Emission max (nm)

In vivo structure

395 399

509 511

Monomer Monomer

439 433 433 435

476 475 475 477

Monomer Monomer Weak dimer Weak dimer

484 489 487 492

507 509 509 505

Monomer Monomer Monomer Monomer

514 515 516 517

527 528 529 530

Monomer Weak dimer Weak dimer Weak dimer

548 548 554 556 574 584 587 590

559 562 581 586 596 607 610 649

Monomer Monomer Tandem dimer Monomer Monomer Monomer Monomer Monomer

Adapted from ref. 10.

ning confocal microscopy employs a pair of oscillating mirrors to direct a light beam across a sample, generating a 2D image of an optical section of the sample 1 pixel at a time. A spinning disc confocal employs numerous pinholes to illuminate many pixels simultaneously. This results in an increased frame rate, albeit at the expense of sensitivity and confocality. The setup and advantages/ disadvantages of these systems are reviewed elsewhere (22).

84

Edoff, Dods, and Brand

2. Materials 2.1. GAL4 Driver Lines There is a vast array of GAL4 driver lines expressing the yeast transcription factor GAL4 in stereotyped patterns (11–13). Table 2 provides examples of GAL4 lines and their expression pattern, which can be useful in studying nervous system development. Many GAL4 and UAS lines are freely available from the fly stock centers: Bloomington Drosophila Stock Center (http://flystocks. bio.indiana.edu/), The Drosophila Genetic Resource Center (http://www.dgrc. kit.ac.jp/en/index.html), Drosophila Stocks of Ehime University (http://218. 44.182.89/~ehime/cgi-bin/index.cgi), Szeged Drosophila Stock Center (http:// expbio.bio.u-szeged.hu/fly/index.php), and the Tucson Drosophila Stock Center (http://stockcenter.arl.arizona.edu/). 2.2. UAS Reporter Lines Numerous UAS reporter lines have been generated that incorporate GFP or one of its variants. It is also possible to clone the coding sequence of a protein of interest fused to a fluorescent protein into the polylinker of the pUAST vector (11–13,23). Commonly used fluorescent UAS reporter lines and fusion proteins can be found in Table 3. It is also possible to fuse a promoter of interest directly to a fluorescent reporter to investigate expression patterns or mark cells or tissues. 2.3. Embryos 2.3.1. Staged Embryo Collection 1. Fly cages. 2. Apple juice plates with wet yeast.

2.3.2. Live Drosophila Embryos 1. 2. 3. 4. 5. 6. 7. 8. 9.

Small plastic paint brush. Sieve, 20 to 40 µm (BD Biosciences). Bleach, 50% Clorox. Water. Air-permeable Teflon Biofoil25 membrane (Heraeus). Perspex frame (made in house). 10S Voltalef oil (Sigma-Aldrich). 18 × 18-mm Cover slips (no. 1.5). 40 × 22-mm Cover slips (no. 1 or 1.5 depending on microscope lenses to be used [see Note 1]).

2.3.3. Fixed Drosophila Embryos 1. Heptane (Sigma-Aldrich).

V37

α-tubulingal4VP16

85

Ventral neuroectoderm CNS neuroblasts

en-gal4

Segmental expression

ase-gal4

sca-gal4 wor-gal4

wg-gal4

twi-gal4 how-gal4 drm-gal4

e16E

24B

V2H

α-tubulingal4VP16

nos-gal4

Short name

GAL4 driver line

Mesoderm All muscles Endoderm

Blastoderm (ubiquitous)

Tissue/cell type

Comments

8 kb upstream sequence of the worniu gene

Expresses GAL4 in embryonic proventriculus, anterior midgut, posterior midgut, Malpighian tubules, and small intestine

Expressed under the control of the alphaTub67C promoter maternally and loaded into eggs Expressed under the control of the alphaTub67C promoter maternally and loaded into eggs 5.1 kb insert includes the nanos promoter and 5' and 3' UTR’s.

Table 2 Examples of GAL4 Lines and Their Expression Pattern

34

A. Brand (unpublished) and 29 J. Pradel (unpublished) and 30,31 32 33

27 11 28

D. St. Johnston (unpublished) and 25 D. St. Johnston (unpublished) and 25,26

References

— (continued)

— —

4918

6356

914 2703 7098

4937

7063

7062

BL stock no.

Detection of GFP in D. melanogaster 85

Motor neurons Cholinergic neurons Dopaminergic neurons Dopaminergic and seretonergic neurons Seretonergic neurons Peptidergic neurons Midline Glia

PNS pI SOP and progeny Neurons

Tissue/cell type

Table 2 (Continued)

86 eg-gal4 crc-gal4 sim-gal4 repo-gal4 gcm-gal4

neur-gal4 elav-gal4 nrv-gal4 — cha-gal4 TH-gal4 Ddc-gal4

GAL4 driver line

mz360 c929

D42

Short name Most if not all neurons Most if not all neurons

Comments

42 43 44 45 46

35 36 37 38 39 40 41

References

8758 — 9150 7415 —

6393 458 — — 6798 — 7010

BL stock no.

86 Edoff, Dods, and Brand

Detection of GFP in D. melanogaster

87

Table 3 Examples of Fluorescent UAS Reporter Lines and Fusion Proteins Structure/cellular localization

Reporter line

Cell membrane Extracellular

G454-eGFP (viking) General mCD8-eGFP Apical in epithelia and neuroblasts insc-GFP Apical in epithelia and neuroblasts baz-GFP Basal in epithelia and neuroblasts pon eGFP Basal in epithelia and neuroblasts mira GFP Zonula adherens junctions ubi-DE-cad-GFP

Cytoskeleton MT MT Actin Myosin Centrosomes Chromatin

ER Axons Synapses Synapses

2. 3. 4. 5. 6. 7. 8. 9. 10.

tau-mGFP6 G147-eGFP GFP-actin sqh-GFP eGFP-D-TACC eYGP-D-TACC Histone H2B-YFP Histone H2B-mRFP PDI-GFP tau-mGFP6 syt-eGFP eGFP-dlg

Comments

Reference

Protein trap line trap line UAS line UAS line UAS line UAS line UAS line Driven off the ubiquitin promoter in all cells

47

UAS line Protein trap line

UAS line

54 47 55 56 57 57 35

UAS line

58

Protein trap line UAS line UAS line UAS line

47 54 59 60

48 49 50 51 52 53

4% Formaldehyde (BDH Chemicals) in PBS (make up fresh). Methanol (Fisher Scientific International). PBT: phosphate-buffered saline (PBS), 0.1% Triton X-100 (Sigma-Aldrich). Primary antibodies. Secondary antibodies. 70% Glycerol/Vectashield (Vector). Glass slides. 18 × 18-mm Cover slips (no. 1.5). 40 × 22-mm Cover slips (see Note 1).

88

Edoff, Dods, and Brand

2.4. Larvae 2.4.1. Staged Larval Collections 1. Fly cages. 2. Fly food plates.

2.4.2. Live Drosophila Larval Tissue 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.

40 × 22-mm Cover slips (see Note 1). 18 × 18-mm Cover slips (no. 1.5). Voltalef oil with 15% chloroform (Sigma-Aldrich). Sharp and fine forceps (Dumont no. 5, Fine Science Tools). Microscissors (Fine Science Tools). Cold PBS. Lids from plastic cell culture Petri dishes (e.g., Nunc). D-22 insect medium (US Biological) supplemented with 7.5% fetal bovine serum (FBS; Sigma-Aldrich) and 0.5 mM ascorbic acid (Sigma-Aldrich). Fat bodies from wild-type third instar larvae. 18 × 18-mm poly-L-lysine-coated cover slips (no. 1.5). Air-permeable teflon Biofoil25 membrane (Heraeus). Perspex frame (made in house).

2.4.3. Fixed Drosophila Larval Tissue 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13.

Ca2+-free PBS (Sigma-Aldrich). Dissection dish filled with 5-mm depth of polymerized Sylgard (BDH Chemicals). Dissecting pins (Fine Science Tools). For larval flat preps: 4% formaldehyde (BDH) in PBS (make up fresh). For dissected larval CNS: 2% paraformaldehyde in PEMs buffer (100 mM PIPES, 2 mM EGTA, 1 mM MgSO4, pH 7.4). PBT (Sigma-Aldrich). 5% Normal goat serum (NGS; Jackson Laboratories) in PBT. Primary antibodies. Secondary antibodies. 70% Glycerol/Vectashield (Vector). Glass slides. Vaseline. 40 × 22-mm Cover slips (see Note 1).

3. Methods The following protocols can be used to examine GFP fluorescence in live or fixed whole-mount samples. There is also a protocol for staining embryos or larvae with or without simultaneous costaining for other markers.

Detection of GFP in D. melanogaster

89

3.1. Live Drosophila Embryos 1. Collect embryos on apple juice agar plates for 1 to 2 h and age them until the appropriate stage of development. 2. When the embryos reach the correct stage, add water to the apple juice plate and gently stroke with a soft paint brush. Pour the water with the embryos into the sieve. 3. Dechorionate by placing the sieve in 50% Clorox for 3 min, and then wash thoroughly with water. 4. Transfer the embryos with a paintbrush into a drop of Voltalef oil placed in the middle of an air-permeable Teflon membrane stretched over a Perspex frame. Vary the amount of Voltalef oil in case the embryos float around/move. (For an alternative way of mounting, see Note 2.) 5. To prevent the embryos from being squashed, place 18 × 18-mm cover slips as spacers on either side of the embryos, resting partly on the Perspex frame, and then cover with a 22 × 40-mm cover slip (see Note 3).

3.2. Fixed Drosophila Embryos GFP remains fluorescent after fixation and can then be compared with other endogenous markers through antibody costaining (see Note 4). 1. Transfer dechorionated embryos to a small bottle or tube (<3 mL) filled with one part heptane and one part 4% formaldehyde in PBS. The embryos should float at the interface between the heptane (top) and formaldehyde (bottom). Fix for 20 min at room temperature on a roller. 2. Remove all the formaldehyde (the bottom layer) and replace it with methanol. Crack open the vitelline membranes of the embryos by shaking or vortexing. Let the two phases separate and the embryos sink through the methanol to the bottom. 3. Transfer the embryos at the bottom to Eppendorf tubes and rinse in fresh methanol. Do not take the embryos that remain at the interface. For direct visualization of GFP, proceed with the rehydration. For most antibody stainings, the fixed embryos can be stored away in methanol at −20°C for up to a few months. 4. To continue, rehydrate the embryos by rinsing in 50% methanol/50% PBT and then twice in 100% PBT. 5. For direct visualization, replace the PBT with 70% glycerol and let the embryos sink. Mount the embryos in a drop of glycerol/Vectrashield on a slide. Place 18 × 18-mm spacer cover slips on the sides and then cover with a 22 × 40-mm cover slip. Do not seal the cover slips, as it is often useful to roll the embryos (by gently moving the 22 × 40-mm cover slip) to obtain dorsal, ventral, or lateral views. 6. To proceed with antibody staining, incubate embryos in primary antibodies diluted in PBT at 4°C overnight or according to antibody specifications. The next day, recover the antibody solution, which may often be reused several times. 7. Rinse in PBT at room temperature 3 × 5 min and then 2 × 10 min. Apply the secondary antibodies diluted in PBT and incubate for more than 2 h at room temperature.

90

Edoff, Dods, and Brand

8. Rinse in PBT 5 × 5 min. Replace the last rinse with 70% glycerol and let the embryos sink to the bottom of the Eppendorf tube. 9. Mount embryos as described for the direct visualization above.

3.3. Live Drosophila Larvae 1. Third instar larvae can be collected from the walls of bottles or vials gently with tweezers. First or second instar larvae are obtained from staged collections on Petri dishes filled with fly food. 2. Transfer the larvae to an empty Petri dish containing some tissue paper soaked in water. Leave the larvae to crawl around for a while to remove yeast and fly food. For whole larva: 3a. Put 18 × 18-mm cover slips at each end of a 40 × 22-mm cover slip and place the larva in the middle. 4a. Add a drop of Voltalef oil with 15% chloroform on the larva and then place a second 40 × 22-mm cover slip on top and squeeze gently until the larva is flattened. This preparation may be imaged for 30 to 45 min. For dissection of larva: 3b. Place the larva in a drop of cold PBS. 4b. Tear the larva in half in the middle. Take the anterior half, hold the mouth hooks with one pair of forceps, and then, with a second pair, push the body wall of the larva back over the mouth to turn it inside out—as you would invert a sock. Then, carefully identify the CNS and release it by cutting the nerves and the esophagus/ proventriculus (see Note 5 for an alternative dissection method). 5b. Transfer the dissected brain to an Eppendorf tube with cold PBS using the forceps or a pipet (see Note 6). 6b. Continue dissecting additional material. 7b. Transfer the dissected CNS to a poly-L-lysine-coated cover slip with a drop of D22 insect medium supplemented with 7.5% FBS, 0.5 mM ascorbic acid, and fat bodies from wild-type third instar larvae. 8b. Under a dissection microscope, orient the CNS and gently push it down to adhere to the coated surface. 9b. Place the cover slip with the dissected CNS upside down on an air-permeable Teflon membrane mounted on a Perspex frame (see Note 7).

3.4. Fixed Drosophila Larva 1. Place the larva in a drop of Ca2+-free PBS in a small Sylgaard-filled Petri dish. The Ca2+-free PBS is used to reduce muscle activity during the dissection. 2. With the dorsal side up, gently hold the larva with a pair of forceps, wait until it extends its mouthparts, and then immobilize it by pinning the head down. 3. Stretch the larva and put a second pin in the posterior end. Ensure that the dorsal side is still up.

Detection of GFP in D. melanogaster

91

4. Cut open the larva along the dorsal midline from the posterior to the anterior with the microscissors. 5. Use forceps to stretch out the posterior and anterior ends of the larva gently, and pin them down. Carefully remove the gut with the forceps. 6. Wash with Ca2+-free PBS to remove debris. 7. Fix the larva with 4% formaldehyde (Fisher Scientific) for 30 to 45 min at room temperature. 8. Wash three times with PBT. Unpin the larva and transfer to a 1.5-mL Eppendorf tube using a pipet. 9. Rinse three times in PBT. Wash 2 × 30 min in PBT on a shaker. 10. For direct visualization of GFP, proceed with mounting the flat preps in Vectrashield (see step 14 below). 11. To proceed with antibody staining, preincubate (block) with 5% NGS in PBT for 30 min at room temperature. 12. Replace the NGS with primary antibodies diluted in PBT and incubate at 4°C overnight or according to antibody specifications. Next day, recover the antibody solution. This can often be reused several times. 13. Rinse three times with PBT. Wash twice for 30 min with PBT on a shaker. 14. Mount the flat preps by transferring them to a drop of Vectrashield on a slide. Under a microdissection microscope, arrange the flat prep with some fine tweezers. Squeeze a thin rectangle of Vaseline around the drop with the specimen. Load a cover slip on top, and gently push this down.

3.5. Confocal Microscopy: Optimization for Detection of GFP Depending on the properties of the process or staining, optimal image acquisition will be achieved with different types of microscopes. Laser scanning microscopes collect images with very high spatial resolution, whereas spinning disc confocal microscopes are designed for high temporal resolution. Below are some general points/notes on GFP detection. 1. Most GFP variants are excited by the 488-nm laser line. 2. The standard setup for imaging fluorescein isothiocyanate (FITC) can be used for GFP. It uses a 522/32 emission filter (i.e., wavelengths between 506 and 538 nm are transmitted). The Chroma HQ500LP emission filter, which transmits all light of wavelengths greater than 500 nm, gives a bright GFP signal (see Note 8). 3. To distinguish the GFP signal from autofluorescence generated by yolk when looking through the eyepieces, it is useful to have a long-pass emission filter. Autofluorescence will look yellowish, and GFP fluorescence will look green (see Note 9). 4. There are five parameters that can be varied to influence the image collection (Fig. 1): Laser power High: stronger signal but risk of fading/bleaching, saturation (see Note 10). Low: insufficient signal.

92

Edoff, Dods, and Brand

Fig. 1. Confocal parameter adjustment. Examples of adjustment of confocal parameters on a 2D confocal image of neurons expressing eGFP. All conditions are kept constant throughout except the parameter in question.

Detection of GFP in D. melanogaster

93

PMT gain High: stronger signal but grainy/noisy image, saturation. Low: insufficient signal. Pinhole High: stronger signal but loss of confocality. Low: insufficient signal. Scan speed High: grainy/noisy image. Low: fading/bleaching (see Note 10). Off set High: saturation. Low: under exposure. 5. To reduce noise and sharpen the image, Kalman average when possible (see Note 10).

3.6. Multichannel Image Collection 1. To detect colocalization of proteins or to distinguish between different cellular compartments, it is useful to employ fluorescent markers with different spectral properties. For long-term in vivo imaging experiments, it is best to avoid fluorescent proteins that require excitation with UV or blue laser light, as this is particularly damaging to living cells. 2. We regularly image embryos that express Histone 2B fused with mRFP to visualize chromatin and the CD8 receptor fused to EGFP to visualize cell membranes. In this background, we can then investigate the localization and behavior of other proteins of interest fused to YFP or CFP. 3. When imaging multiple spectra, it is often best to perform sequential image acquisitions. This increases the acquisition time but reduces the risk of cross-talk between the channels. 4. Alternatively, for live imaging applications, for which rapid scanning is essential, such as when following cell divisions, microscope manufacturers now offer hardware and software solutions that allow separation/unmixing of complex fluorescent signals after the image is captured (see Note 11).

3.7. Time Lapse Microscopy 1. Recording a time lapse series requires a relatively bright fluorescent signal that will withstand the bleaching that results from repeated scanning. Laser power should be minimized to reduce phototoxicity. 2. To achieve noise reduction when making time lapse series, perform “line averaging” instead of “frame averaging” during image acquisition to avoid the risk of blurring if the specimen moves. Alternatively, it may be better to use a low scan speed without averaging. 3. When imaging rapid or dynamic cellular processes, such as cell division or axonal transport, laser scanning confocal microscopy may lack the temporal resolution to capture the biological phenomenon. In such cases, we employ a spinning disc con-

94

Edoff, Dods, and Brand focal, which has the benefit of a greater acquisition rate, albeit at the expense of confocality and sensitivity compared with a conventional laser scanning confocal microscope.

4. Notes 1. Ensure that slides and cover slips are clean and dust free. We use the brand Menzel and store opened packages in larger boxes with lids. Imaging with high-magnification lenses (40× and up) requires the use of cover slips with the thickness for which the microscope objectives are designed and corrected. Most close working distance objectives are designed for no. 1.5 cover slips. There are also objectives that have an adjustable correction ring allowing the use of cover glasses with different thicknesses. Standard cover slip thickness No. 0 0.08–0.12 mm No. 1 0.13–0.17 mm No. 1.5 0.16–0.19 mm No. 2 0.19–0.23 mm 2. For live imaging and time-course studies, it can be useful to orient all embryos to a desired position (i.e., ventral side up) on apple juice plates and then to immobilize the embryos in position on a cover slip by the use of a glue. If this is necessary, we use a nontoxic glue such as Scotch tape dissolved in heptane. 3. The 18 × 18 cover slips prevent the embryos from being squashed. Do not seal the cover slips. It is often useful to roll the embryos (by gently moving the 22 × 40mm cover slip) to obtain straight dorsal, ventral, or lateral views (not possible or required if the embryos are glued as in Note 2). Do not use nail polish, which will inhibit the fluorescence of live GFP. 4. GFP fluorescence may be somewhat diminished after formaldehyde fixation and is further quenched by prolonged storage in methanol (10). 5. An alternative way of dissecting larval CNS in third instar larva is to grab hold of the mouth hooks with one pair of tweezers. With a second pair of tweezers, take a firm hold of the trunk of the larva (about 2/5 along the anterior-posterior axis of the larva) and then pull so that the larva splits open. The CNS, along with the eye and antennal discs, should be attached to the mouth hooks and can then be dissected free. This method is very fast. However, the ventral nerve cord is sometimes torn off. 6. To avoid the prep sticking to the pipet tip, pretreat the tip by pipeting some serum or a drop of saliva a few times before transferring your larval specimen. 7. If the medium evaporates, it is possible to add some more at the edges of the cover slip. 8. The filters given here are for GFP excitation and fluorescence but will change according to the fluorescent protein being detected. Excitation and emission wavelengths for some GFP variants are given in Table 1. 9. The long-pass filter should contain the usual GFP excitation filter, combined with an emission filter that lets all wavelengths pass from 500 or 520 nm. For a comprehensive review on autofluorescence, see ref. 24.

Detection of GFP in D. melanogaster

95

10. Although GFP is fairly resistant to bleaching, excessive laser power or a slow scan speed can lead to a loss of signal. This will vary with the expression level and stability of the fusion protein. To limit this loss and reduce photo-damage to the sample, the number of sections within a z series and the number of Kalman averages should be kept to the minimum. With some samples we can collect images in a single focal plane every 15 s for 2 h without significant loss in signal intensity. When collecting 4D images, we see no significant loss in signal in z series taken every 5 min for up to 5 h. The vast majority of our imaging is done on whole-mount embryos, through the vitelline membrane. In our experience, dissected samples (for example, “flat preps” and single-cell preparations) tend to bleach more readily. 11. The Zeiss LSM 510 META microscope contains a scanning module with a META detector that covers the entire fluorescence spectrum. This allows emission fingerprinting and unmixing of signals based on the shape of the whole emission spectrum of each individual fluorescent marker. Other software solutions do not require particular hardware and employ statistical algorithms that filter away overlapping emission signals.

References 1. Shimomura, O., Johnson, F. H., and Saiga, Y. (1962) Extraction, purification and properties of aequorin, a bioluminescent protein from the luminous hydromedusan, Aequorea. J. Cell Comp. Physiol. 59, 223–239. 2. Prasher, D. C., Eckenrode, V. K., Ward, W. W., Prendergast, F. G., and Cormier, M. J. (1992) Primary structure of the Aequorea victoria green-fluorescent protein. Gene 111, 229–233. 3. Yuste, R. (2005) Fluorescence microscopy today. Nat. Methods 2, 902–904. 4. Ando, R., Mizuno, H., and Miyawaki, A. (2004) Regulated fast nucleocytoplasmic shuttling observed by reversible protein highlighting. Science 306, 1370–1373. 5. Ando, R., Hama, H., Yamamoto-Hino, M., Mizuno, H., and Miyawaki, A. (2002) An optical marker based on the UV-induced green-to-red photoconversion of a fluorescent protein. Proc. Natl. Acad. Sci. USA 99, 12,651–12,656. 6. Shaner, N. C., Campbell, R. E., Steinbach, P. A., Giepmans, B. N., Palmer, A. E., and Tsien, R. Y. (2004) Improved monomeric red, orange and yellow fluorescent proteins derived from Discosoma sp. red fluorescent protein. Nat. Biotechnol. 22, 1567–1572. 7. Miyawaki, A. (2005) Innovations in the imaging of brain functions using fluorescent proteins. Neuron 48, 189–199. 8. Matz, M. V., Fradkov, A. F., Labas, Y. A., et al. (1999) Fluorescent proteins from nonbioluminescent Anthozoa species. Nat. Biotechnol. 17, 969–973. 9. Gurskaya, N. G., Verkhusha, V. V., Shcheglov, A. S., et al. (2006) Engineering of a monomeric green-to-red photoactivatable fluorescent protein induced by blue light. Nat. Biotechnol. 24, 461–465. 10. Shaner, N. C., Steinbach, P. A., and Tsien, R. Y. (2005) A guide to choosing fluorescent proteins. Nat. Methods 2, 905–909.

96

Edoff, Dods, and Brand

11. Brand, A. H. and Perrimon, N. (1993) Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118, 401–415. 12. Brand, A. H. and Dormand, E. L. (1995) The GAL4 system as a tool for unravelling the mysteries of the Drosophila nervous system. Curr. Opin. Neurobiol. 5, 572–578. 13. Brand, A. H., Manoukian, A. S., and Perrimon, N. (1994) Ectopic expression in Drosophila, in Drosophila melanogaster: Practical Uses in Cell and Molecular Biology, Academic Press, San Diego, pp. 635–652. 14. Phelps, C. B. and Brand, A. H. (1998) Ectopic gene expression in Drosophila using GAL4 system. Methods 14, 367–379. 15. Heim, R., Prasher, D. C., and Tsien, R. Y. (1994) Wavelength mutations and posttranslational autoxidation of green fluorescent protein. Proc. Natl. Acad. Sci. USA 91, 12,501–12,504. 16. Cubitt, A. B., Heim, R., Adams, S. R., Boyd, A. E., Gross, L. A., and Tsien, R. Y. (1995) Understanding, improving and using green fluorescent proteins. Trends Biochem. Sci. 20, 448–455. 17. Heim, R., Cubitt, A. B., and Tsien, R. Y. (1995) Improved green fluorescence. Nature 373, 663–664. 18. Zernicka-Goetz, M., Pines, J., Ryan, K., et al. (1996) An indelible lineage marker for Xenopus using a mutated green fluorescent protein. Development 122, 3719–3724. 19. Siemering, K. R., Golbik, R., Sever, R., and Haseloff, J. (1996) Mutations that suppress the thermosensitivity of green fluorescent protein. Curr. Biol. 6, 1653–1663. 20. Shagin, D. A., Barsova, E. V., Yanushevich, Y. G., et al. (2004) GFP-like proteins as ubiquitous metazoan superfamily: evolution of functional features and structural complexity. Mol. Biol. Evol. 21, 841–850. 21. Chudakov, D. M., Lukyanov, S., and Lukyanov, K. A. (2005) Fluorescent proteins as a toolkit for in vivo imaging. Trends Biotechnol. 23, 605–613. 22. Conchello, J. A. and Lichtman, J. W. (2005) Optical sectioning microscopy. Nat. Methods 2, 920–931. 23. Haseloff, J., Dormand, E. L., and Brand, A. H. (1999) Live imaging with green fluorescent protein. Methods Mol. Biol. 122, 241–259. 24. Billinton, N. and Knight, A. W. (2001) Seeing the wood through the trees: a review of techniques for distinguishing green fluorescent protein from endogenous autofluorescence. Anal. Biochem. 291, 175–197. 25. Hacker, U. and Perrimon, N. (1998) DRhoGEF2 encodes a member of the Dbl family of oncogenes and controls cell shape changes during gastrulation in Drosophila. Genes Dev. 12, 274–284. 26. Van Doren, M., Williamson, A. L., and Lehmann, R. (1998) Regulation of zygotic gene expression in Drosophila primordial germ cells. Curr. Biol. 8, 243–246. 27. Greig, S. and Akam, M. (1993) Homeotic genes autonomously specify one aspect of pattern in the Drosophila mesoderm. Nature 362, 630–632. 28. Green, R. B., Hatini, V., Johansen, K. A., Liu, X. J., and Lengyel, J. A. (2002) Drumstick is a zinc finger protein that antagonizes Lines to control patterning and morphogenesis of the Drosophila hindgut. Development 129, 3645–3656.

Detection of GFP in D. melanogaster

97

29. Fietz, M. J., Jacinto, A., Taylor, A. M., Alexandre, C., and Ingham, P. W. (1995) Secretion of the amino-terminal fragment of the hedgehog protein is necessary and sufficient for hedgehog signalling in Drosophila. Curr. Biol. 5, 643–650. 30. Payre, F., Vincent, A., and Carreno, S. (1999) ovo/svb integrates Wingless and DER pathways to control epidermis differentiation. Nature 400, 271–275. 31. Sanson, B., Alexandre, C., Fascetti, N., and Vincent, J. P. (1999) Engrailed and hedgehog make the range of Wingless asymmetric in Drosophila embryos. Cell 98, 207–216. 32. Klaes, A., Menne, T., Stollewerk, A., Scholz, H., and Klambt, C. (1994) The Ets transcription factors encoded by the Drosophila gene pointed direct glial cell differentiation in the embryonic CNS. Cell 78, 149–160. 33. Albertson, R., Chabu, C., Sheehan, A., and Doe, C. Q. (2004) Scribble protein domain mapping reveals a multistep localization mechanism and domains necessary for establishing cortical polarity. J. Cell Sci. 117, 6061–6070. 34. zur Lage, P. and Jarman, A. P. (1999) Antagonism of EGFR and notch signalling in the reiterative recruitment of Drosophila adult chordotonal sense organ precursors. Development 126, 3149–3157. 35. Bellaiche, Y., Gho, M., Kaltschmidt, J. A., Brand, A. H., and Schweisguth, F. (2001) Frizzled regulates localization of cell-fate determinants and mitotic spindle rotation during asymmetric cell division. Nat. Cell Biol. 3, 50–57. 36. Lin, D. M. and Goodman, C. S. (1994) Ectopic and increased expression of Fasciclin II alters motoneuron growth cone guidance. Neuron 13, 507–523. 37. Sun, B., Xu, P., and Salvaterra, P. M. (1999) Dynamic visualization of nervous system in live Drosophila. Proc. Natl. Acad. Sci. USA 96, 10,438–10,443. 38. Yeh, E., Gustafson, K., and Boulianne, G. L. (1995) Green fluorescent protein as a vital marker and reporter of gene expression in Drosophila. Proc. Natl. Acad. Sci. USA 92, 7036–7040. 39. Salvaterra, P. M. and Kitamoto, T. (2001) Drosophila cholinergic neurons and processes visualized with Gal4/UAS-GFP. Brain Res. Gene Expr. Patterns 1, 73–82. 40. Friggi-Grelin, F., Iche, M., and Birman, S. (2003) Tissue-specific developmental requirements of Drosophila tyrosine hydroxylase isoforms. Genesis 35, 175–184. 41. Li, H., Chaney, S., Roberts, I. J., Forte, M., and Hirsh, J. (2000) Ectopic G-protein expression in dopamine and serotonin neurons blocks cocaine sensitization in Drosophila melanogaster. Curr. Biol. 10, 211–214. 42. Ito, K., Urban, J., and Technau, G. M. (1995) Distribution, classification, and development of Drosophila glial cells in the late embryonic and early larval ventral nerve cord. Rouxs Arch. Dev. Biol. 204, 284–307. 43. Hewes, R. S., Schaefer, A. M., and Taghert, P. H. (2000) The cryptocephal gene (ATF4) encodes multiple basic-leucine zipper proteins controlling molting and metamorphosis in Drosophila. Genetics 155, 1711–1723. 44. Golembo, M., Raz, E., and Shilo, B. Z. (1996) The Drosophila embryonic midline is the site of Spitz processing, and induces activation of the EGF receptor in the ventral ectoderm. Development 122, 3363–3370.

98

Edoff, Dods, and Brand

45. Sepp, K. J., Schulte, J., and Auld, V. J. (2001) Peripheral glia direct axon guidance across the CNS/PNS transition zone. Dev. Biol. 238, 47–63. 46. Booth, G. E., Kinrade, E. F., and Hidalgo, A. (2000) Glia maintain follower neuron survival during Drosophila CNS development. Development 127, 237–244. 47. Morin, X., Daneman, R., Zavortink, M., and Chia, W. (2001) A protein trap strategy to detect GFP-tagged proteins expressed from their endogenous loci in Drosophila. Proc. Natl. Acad. Sci. USA 98, 15,050–15,055. 48. Lee, T. and Luo, L. (1999) Mosaic analysis with a repressible cell marker for studies of gene function in neuronal morphogenesis. Neuron 22, 451–461. 49. Akiyama-Oda, Y., Hotta, Y., Tsukita, S., and Oda, H. (2000) Mechanism of glianeuron cell-fate switch in the Drosophila thoracic neuroblast 6-4 lineage. Development 127, 3513–3522. 50. Benton, R., and St Johnston, D. (2003) A conserved oligomerization domain in Drosophila Bazooka/PAR-3 is important for apical localization and epithelial polarity. Curr. Biol. 13, 1330–1334. 51. Lu, B., Ackerman, L., Jan, L. Y., and Jan, Y. N. (1999) Modes of protein movement that lead to the asymmetric localization of partner of Numb during Drosophila neuroblast division. Mol. Cell 4, 883–891. 52. Ohshiro, T., Yagami, T., Zhang, C., and Matsuzaki, F. (2000) Role of cortical tumour-suppressor proteins in asymmetric division of Drosophila neuroblast. Nature 408, 593–596. 53. Oda, H. and Tsukita, S. (2001) Real-time imaging of cell-cell adherens junctions reveals that Drosophila mesoderm invagination begins with two phases of apical constriction of cells. J. Cell Sci. 114, 493–501. 54. Kaltschmidt, J. A., Davidson, C. M., Brown, N. H., and Brand, A. H. (2000) Rotation and asymmetry of the mitotic spindle direct asymmetric cell division in the developing central nervous system. Nat. Cell Biol. 2, 7–12. 55. Roper, K., Mao, Y., and Brown, N. H. (2005) Contribution of sequence variation in Drosophila actins to their incorporation into actin-based structures in vivo. J. Cell Sci. 118, 3937–3948. 56. Sisson, J. C., Field, C., Ventura, R., Royou, A., and Sullivan, W. (2000) Lava lamp, a novel peripheral golgi protein, is required for Drosophila melanogaster cellularization. J. Cell Biol. 151, 905–918. 57. Gergely, F., Karlsson, C., Still, I., Cowell, J., Kilmartin, J., and Raff, J. W. (2000) The TACC domain identifies a family of centrosomal proteins that can interact with microtubules. Proc. Natl. Acad. Sci. USA 97, 14,352–14,357. 58. Langevin, J., Le Borgne, R., Rosenfeld, F., Gho, M., Schweisguth, F., and Bellaiche, Y. (2005) Lethal giant larvae controls the localization of notch-signaling regulators numb, neuralized, and Sanpodo in Drosophila sensory-organ precursor cells. Curr. Biol. 15, 955–962. 59. Zhang, Y. Q., Rodesch, C. K., and Broadie, K. (2002) Living synaptic vesicle marker: synaptotagmin-GFP. Genesis 34, 142–145. 60. Koh, Y. H., Popova, E., Thomas, U., Griffith, L. C., and Budnik, V. (1999) Regulation of DLG localization at synapses by CaMKII-dependent phosphorylation. Cell 98, 353–363.

Autofluorescent Proteins for Flow Cytometry

99

7 Autofluorescent Proteins for Flow Cytometry Charles G. Bailey and John E. J. Rasko Summary The unique spectral properties and versatility of autofluorescent proteins have facilitated their widespread use in flow cytometric applications. The ability to analyze heterologous fluorescent protein expression conveniently and noninvasively by individually interrogating cells has facilitated increasingly more sophisticated experimental designs to address important biological questions. Improved multilaser flow cytometers have allowed the fluorescent protein field to flourish by permitting high-speed, multiparametric analysis of biological samples. Fluorescent proteins are well suited for either transient or stable expression analysis. Therefore, achieving efficient gene transfer and expression in cells by transfection or viral transduction is paramount to the optimal use of fluorescent proteins in flow cytometry. The archetypal autofluorescent protein, enhanced green fluorescent protein (eGFP), can be used successfully in combination with other fluorescent protein variants. Two such variants, Cerianthus sp. orange fluorescent protein (cOFP) and a fast maturing variant of Discosoma sp. red protein (DsREDExpress), are well suited for flow cytometric applications in combination with eGFP and do not require special filters for optimal excitation and detection. Key Words: Fluorescent protein; flow cytometry; eGFP; cOFP; DsREDExpress; transfection; transduction.

1. Introduction The application of fluorescent proteins in biology has revolutionized experimental design and expanded the range of tools available to the researcher. Green fluorescent protein (GFP) and its derivatives have become arguably the most widely used reporters of heterologous gene expression since their first use (1). Fluorescent proteins were subsequently engineered to improve their expression, spectral properties, and chromophore maturation (2,3). The red-shifted variant, enhanced (e)GFP, was modified through human-codon optimization and chromophore substitution to increase brightness and expression in mammalian cells From: Methods in Molecular Biology, vol. 411: Reporter Genes: A Practical Guide Edited by: D. Anson © Humana Press Inc., Totowa, NJ

99

100

Bailey and Rasko

(4,5). Now, a burgeoning family of engineered fluorescent proteins including eGFP have found increased use in flow cytometric applications (6); our aim in this chapter is to provide insights into the application of some newer variants. A whole toolbox of fluorescent proteins now exists, and their successful application in a given experiment will depend on such properties as brightness, photostability, oligomerization, toxicity, and pH sensitivity (6). As many such variants are now commercially available, the choices made in combining fluorescent proteins used for multiparametric analysis will be determined significantly by the capabilities of the available flow cytometer. Analysis of four fluorescent proteins simultaneously is possible using a dual-laser machine (7), but this requires meticulous preparation, operator proficiency, and precise filter combinations. We have shown that two recently developed fluorescent protein variants, the coral Discosoma sp. red fluorescent protein (DsREDExpress) (8) and the tube anemone Cerianthus sp. orange fluorescent protein (cOFP) (9), can be effectively used in combination with eGFP. Analysis of these three autofluorescent proteins can be performed using excitation by one laser (488 nm), thereby permitting the simultaneous detection of fluorochromes excited by separate laser lines. A common application of fluorescent proteins, particularly eGFP and its variants, is as a rapid and convenient positive control for gene transfer experiments. When one is performing transfection experiments, eGFP has proved particularly useful for quantifying transfection efficiency. Initially DNA uptake can be determined qualitatively by direct visualization under fluorescence microscopy with the added advantage that adherent cells can remain in situ in a sterile plastic dish. Flow cytometry provides added advantages by allowing accurate and precise determination of transfection efficiency. Both transient and long-term stable expression of autofluorescent protein-expressing cells can be achieved using viral vector systems. In the following method, the investigator is given the option either to transfect cells with plasmid DNA or to transduce cells with retroviral or lentiviral vectors to achieve autofluorescent protein expression. 2. Materials 2.1. Cell Culture 1. Dulbecco’s modified Eagle’s medium (DMEM)/10% fetal calf serum (FCS): DMEM (Invitrogen, Carlsbad, CA) supplemented with 3.7 g/L NaHCO3, 25 mM HEPES, 100 U/mL penicillin, 100 µg/mL streptomycin, and 10% (v/v) FCS (JRH Biosciences, Lenexa, KS). 2. Trypsin/EDTA solution containing 0.25% porcine trypsin (Sigma, St. Louis, MO), 0.2 g/L EDTA, phenol red (Sigma). 3. Phosphate-buffered saline (PBS): prepare as an 8.0 g/L NaCl, 0.2 g/L KCl, 1.44 g/L Na2HPO4, and 0.24 g/L KH2PO4 solution; adjust pH to 7.4 and autoclave.

Autofluorescent Proteins for Flow Cytometry

101

2.2. Transfection Option 1. This adapted calcium phosphate transfection protocol (10) uses four solutions that should be warmed to room temperature before use. 2. 300 mM Phosphate buffer: 195 mM Na2HPO4, 105 mM NaH2PO4, pH 7.05; autoclave and store at 4°C (see Note 1). 3. 2X HEPES buffer: 140 mM NaCl, 50 mM HEPES, pH 7.05; 0.22 µm filter-sterilize and store at 4°C (see Note 2). 4. 2.5 M CaCl2: autoclave and store at 4°C. 5. 1/10 concentration TE buffer: 1 mM Tris-HCl, 0.1 mM EDTA, pH 7.6; autoclave and store at 4°C (see Note 3). 6. 25 mM Chloroquine: prepare as single-use aliquots in water, 0.22 µm filter-sterilize, and store at −20°C protected from light. 7. Plasmid DNA: prepare as a 1 µg/µL stock in 1/10 TE. Precipitate a known amount of DNA from a miniprep or maxiprep with 1/10 final volume 3 M NaAc, pH 5.2, and 2 volumes of 95% ethanol. Pellet DNA for 5 min at 12,000g, wash pellet in sterile 70% ethanol in a laminar flow hood, and resuspend in 1/10 TE buffer. 8. Denatured, fractionated herring sperm DNA (Roche, Indianapolis, IN): prepare as a 1 µg/µL stock in 1/10 TE, and heat at 100°C for 5 min, before snap-freezing in small aliquots and storing at −20°C.

2.3. Transduction Option 1. Polybrene® (hexadimethrine bromide, Sigma): prepare as a 4 mg/mL stock in water, 0.22 µm filter-sterilize, and store at 4°C. 2. 3% (w/v) Lactose/PBS: prepare 3% (w/v) lactose (Sigma) in PBS, 0.22 µm filtersterilize, and store at 4°C. 3. Non-replication-competent viral supernatant: prepare from HEK293T packaging cells by transfection or from stable packaging cell-lines (see ref. 11 for examples). In this chapter, retroviral supernatants were prepared from: derivatives of the pLN vector series (12) (also purchasable from Clontech, Mountain View, CA) combined with the HIV-1 tat-dependent retroviral vector packaging system (13); lentiviral supernatants were prepared from third-generation self-inactivating lentiviral vectors (14). Both retroviral and lentiviral vectors were pseudotyped with the vesicular stomatitis virus glycoprotein (VSV-G) envelope. All transductions described used 0.45-µm-filtered snap-frozen aliquots of virus-containing supernatant or those that had been concentrated 100-fold-by-volume by centrifugation for 5 h at 29,000g in a FiberLite F13S-14x50cy carbon fiber rotor (Piramoon Technologies, Santa Clara, CA) in a Sorvall 5C series centrifuge and resuspended overnight at 4°C in 3% (w/v) lactose/PBS.

2.4. Preparation of Cells for Flow Cytometry 1. Fluorescence-activated cell sorting (FACS) tubes (uncapped, capped, or filter-capped depending on the need to maintain sterility; BD Biosciences, Bedford, MA). 2. FACS wash buffer: prepare 2% FCS, 2 mM EDTA in PBS, 0.22 µm filter-sterilize, and store at 4°C up to 3 mo.

102

Bailey and Rasko

3. Propidium iodide (PI): prepare as a 5 mg/mL stock and store in small aliquots at −20°C. 4. DNase I solution (Roche): prepare as a 10 mg/mL stock (approx 30,000 U/mL), and store at −20°C. 5. Paraformaldehyde 2% (v/v) (Sigma): prepare fresh (see Note 4).

2.5. Flow Cytometer Setup and Analysis 1. Most modern multilaser flow cytometers include a blue 488-nm argon ion laser line and a red helium-neon 633-nm laser line (e.g., FACSCalibur, FACSCanto, BD Biosciences; MoFlo, Dako-Cytomation, Ft Collins, CO; Cytomics FC500, Beckman Coulter, Fullerton, CA). Newer multilaser machines such as the LSRII (BD Biosciences) offer up to four lasers, which provide a spectral coverage for ultraviolet at 351 nm, violet at 407 nm, blue at 488 nm, and red at 633 nm. Such devices are capable of detecting multiple fluorescent proteins used with a combination of different fluorochromes. In this chapter all flow cytometric data acquisitions were performed on a FACSCalibur (model no.163-A202) equipped with a blue 488-nm and 635-nm red diode laser. 2. A 15-mW blue laser line with a narrow 10-nm bandwith filter adequately excites many autofluorescent proteins, including eGFP, optimally at 488 nm. A 505-nm long-pass filter blocks all shorter wavelength fluorescence emitted into the first photomultiplier tube (PMT) detector preceded by a 530 ± 30-nm bandpass filter. Other more red-shifted autofluorescent proteins such as cOFP (Stratagene, La Jolla, CA; Fig. 1) and dsREDExpress (Clontech; Fig. 2) can be excited at 488 nm and are detected by the second PMT. This detector has a 556-nm long-pass filter to block lower wavelength fluorescence and a 585 ± 42-nm bandpass filter (all filters from Chroma Technology, Rockingham, VT). 3. Virkon, which is an effective agent for surface decontamination of viruses, bacteria, fungi, and mycobacteria, is used to flush out and decontaminate the fluidics system. Prepare as a 1 or 2% (w/v) solution in water, and store at room temperature for 1 wk or until pink color fades.

3. Methods 3.1. Cell Culture 1. Cells that are being transfected should be of a low passage number. Inoculate cells into the culture flask or dish 18 to 24 h prior to transduction to ensure that they have entered log-phase division. To achieve optimal transfection efficiency, cells should be at approx 80% confluency at the time of transfection. 2. For transfection of cells in 6-well plates, inoculate 5 × 105 HEK293T cells per well containing 2 mL DMEM/10% FCS. 3. For transduction of cells, inoculate 1 × 105 HT1080 cells per well (12-well plate) or 4 × 105 cells per well (6-well plate), and disperse evenly. Several hours after adding adherent cells, view wells under low magnification to determine that the cells are distributed uniformly (see Note 5).

Autofluorescent Proteins for Flow Cytometry

103

Fig. 1. Cotransduction with lentiviral vectors containing eGFP and cOFP fluorescent proteins. Lentiviral vectors, pHIV-1SDm, were derived from a Tat-dependent, third-generation, self-inactivating vector system (14) and contain eGFP or cOFP, a tube anemone Cerianthus sp. fluorescent protein (Stratagene). VSV-G-pseudotypted lentiviral particles were produced by calcium phosphate transfection of HEK293T cells and then concentrated by high-speed centrifugation and titered onto HT1080 cells. Equivalent transducing units were used to transduce HT1080 cells separately and in combination for 48 h before analysis. Contour plots of cells with cOFP in FL2 vs eGFP in FL1 are shown. (A) Negative control cells transduced with Polybrene-containing medium only. (B) eGFP-containing lentivirus only. (C) cOFP-containing lentivirus only. (D) Equal mixture of cOFP and eGFP cells transduced separately. (E) Simultaneous cotransduction with cOFP-and eGFP-containing lentivirus. PI-positive cells in this case were detected in FL3 and excluded by electronically gating them in an FL3:PI vs FL2:cOFP scatter plot (not shown). Although the autofluorescent protein cOFP is suboptimally excited by the 488-nm laser, it still can be distinctly separated from eGFP fluorescence in a bivariate analysis. Cells transduced separately by the two vectors and that are then combined equally for analysis do not exhibit cOFP- and eGFP-coexpressing cells (D, upper right quadrant). In contrast, simultaneously cOFP- and eGFP-expressing cells are detected when virus transduction occurs in the same dish (E, upper right quadrant).

3.2. Transfection Option 1. One hour before transfection, add fresh 1.5 mL DMEM/10% FCS medium containing 25 µM chloroquine (see Note 6).

104

Bailey and Rasko

Fig. 2. Cotransduction with retroviral vectors containing eGFP and DsREDExpress fluorescent proteins. Retroviral constructs were based on the pLN retroviral vectors (12) and contained eGFP or DsREDExpress (Clontech), a fast-maturing DsRED variant suitable for flow cytometric applications. VSV-G-pseudotyped retroviral particles were produced by cotransfection with a Tat-dependent packaging system (13) in HEK293T cells and titered onto HT1080 fibroblasts. Equivalent amounts of retroviral vector were used to transduce HT1080 cells separately (B and C) and in combination (D) for 48 h before analysis. Contour plots with DsREDExpress in FL2 vs eGFP in FL1 are shown. (A) Negative control cells transduced with Polybrene-containing medium only. (B) Cells transduced with LNCG supernatant alone. (C) Cells transduced with LNC (DsREDExpress) supernatant alone. (D) Cells simultaneously transduced with LNC (DsREDExpress) and LNCG supernatants. PI-positive cells in this case were detected in FL3 and excluded by electronically gating them in an FL3:PI vs FL2:DsREDExpress scatter plot (not shown). DsREDExpress has little spectral overlap with eGFP, so it can

Autofluorescent Proteins for Flow Cytometry

105

2. Prepare solution A containing 0.9 µL phosphate buffer and 182 µL 2X HEPES buffer. In a separate tube prepare solution B containing 18 µL 2.5 M CaCl2 and 163 µL 1/10 concentration TE buffer. Heat plasmid DNA for 10 min at 65°C and then add 4.5 µg DNA to solution B (see Note 7). 3. Add solution B containing DNA dropwise to solution A while vortexing; a precipitate should form as a slightly milky solution. Within 1 min, gently add the precipitate to the cells by dripping them around the periphery of the well. HEK293T cells can be inadvertently dislodged if the addition of precipitate is too vigorous. 4. Within 12 to 18 h of transfection, it is possible to obtain a preliminary estimate of transfection efficiency by using UV fluorescence stereomicroscopy (such as MZ FZLIII, Leica, Bannockburn, IL) while maintaining sterility in cell culture (see Note 8).

3.3. Transduction Option 1. View the cells using phase contrast light microscopy to assess their confluency. Ideally the cells should be 50 to 80% confluent to achieve optimum transduction efficiency. 2. Remove the culture medium and replace with fresh medium containing Polybrene (4–8 µg/mL). To make a stock of transduction medium containing 8 µg/mL final concentration of Polybrene that accommodates the addition of a volume of viruscontaining supernatant, add 2.5 µL Polybrene per mL of DMEM/10% FCS medium. For 24-well, 12-well, 6-well, and 10-cm plates, add 400 µL, 800 µL, 1.6 mL, and 8 mL, respectively, of Polybrene-containing medium (see Note 9). 3. Thaw viral supernatant rapidly in a 37°C water bath, and prepare dilutions in DMEM/10% FCS depending on the initial viral titer and whether it was concentrated by high-speed centrifugation. Prepare serial dilutions of virus, and add 100 µL, 200 µL, 400 µL, and 2 mL for transducing 24-well, 12-well, 6-well, and 10-cm plates, respectively. All concentrated virus stocks should be diluted into this larger volume to allow an even distribution of virus on the target cells. 4. If cells are transduced in multiwell plates, these may be spun or “spinoculated” (15) at 450g for 1 h at room temperature to increase transduction efficiency before returning to the 37°C CO2 incubator. After exposing cells to virus-containing supernatant for 4 h, remove the medium and add an equal volume of fresh medium. Cells can be analyzed after 24 h for fluorescent protein expression, but ideally higher transgene expression is achievable after 48 h. Cells can be monitored by fluorescence microscopy to determine transduction efficiency qualitatively.

Fig. 2. (continued) be successfully used in bivariate analyses with eGFP. In the case of extremely bright eGFP fluorescence, electronic gating should be applied using polygons rather than quadrants, as compensation cannot remove all the spectral overlap of eGFP into FL2 (see B, upper right quadrant containing 4.15% eGFP-positive cells).

106

Bailey and Rasko

3.4. Preparation of Cells for Flow Cytometry 1. Harvest the cells from the multiwell plates or flasks using trypsin/EDTA at 37°C. Inactivate the trypsin by collecting the cells in at least four volumes of DMEM/ 10% FCS. Titurate the cells to break up clumps and then transfer to capped FACS tubes containing 2 mL FACS wash buffer. 2. Centrifuge the cells at 400g for 5 min, and remove the supernatant carefully by aspiration (see Note 10). Resuspend the cells in 100 to 500 µL FACS wash buffer containing 5 µg/mL PI to a maximum density of 5 × 106 cells/mL (see Note 11). 3. If cells have been transduced with viral supernatant capable of transducing human cells, then they should be fixed before flow cytometric analysis for safety and to avoid contamination. After preparing the cells according to step 1, resuspend them in 100 to 500 µL FACS wash and then dropwise add an equal volume of freshly prepared 2% paraformaldehyde while vortexing. PI should not be used if cells are fixed, as all cells will be positive and live/dead cell discrimination cannot be performed. 4. Store cells on ice and protect from light until analysis. If necessary, fresh samples can be maintained at 4°C overnight before analysis. Paraformaldehyde-fixed samples are stable for up to 1 wk if stored at 4°C provided they are uniformly fixed and the fixative is removed by thorough washing (see Note 12).

3.5. Flow Cytometer Setup and Analysis 1. The flow cytometer should be switched on at least 20 min before sample analysis to achieve laser stability. Check the fluidics system including the waste fluid reservoir and the sheath fluid levels. Run water through the system to flush out any previous users’ samples. 2. Using standard software (e.g., FlowJo, Treestar, Ashland, OR; CellQuest, BD; Summit 3.0, Dako-Cytomation, Ft. Collins, CO), open templates to detect all necessary parameters in real time. These templates should include (1) a forward scatter vs side scatter plot to exclude debris, doublets, and outliers; (2) a histogram for live/ dead cell discrimination using PI; and (3) dot plots or histograms to allow fluorescent protein detection and quantitation (Fig. 3). 3. Forward vs side light scatter analysis allows cells of uniform size including healthy and viable cells to be analyzed (Fig. 3i[A–D]) and distinguishes them from cellular debris and clumps. Once a population is selected based on light scatter, a second plot to allow electronic exclusion of dead cells by PI staining can be applied. Normally, a PI (FL2) vs forward scatter plot is sufficient for dead cell discrimination. In the case of exceedingly bright eGFP fluorescence, it is preferable to apply electronic gating on FL2 vs FL1 (eGFP) (Fig. 3ii[A–D]) to exclude PI-positive cells (see Note 13). 4. Once dead cells have been electronically excluded, a new scattergram of FL2 vs FL1 will allow the quantitation of eGFP-positive cells. The analysis can be performed using a density plot or a histogram. Gating is established to exclude known control eGFP-negative cells to allow a quantitative assessment of the percentage of eGFP-positive cells. These eGFP-negative cells are typically cells transfected

Autofluorescent Proteins for Flow Cytometry

107

Fig. 3. Quantitative promoter analysis using eGFP as a reporter gene. HEK293T cells were transfected with a negative control or one of three eGFP reporter constructs: (A) control herring sperm DNA; (B) peGFP (a promoter-less eGFP construct; Clontech); (C) pUbiCeGFP (containing a ubiquitin C promoter); and (D) peGFP-C1 (Clontech). Plasmid DNA (0.5 µg) was combined with 4 µg control herring sperm DNA and transfected by calcium phosphate precipitation. The cells were analyzed by flow cytometry 48 h later. Data analysis was performed with FlowJo software: (i) side scatter (SSC-H) vs forward scatter (FSC-H) plots to select a uniformly sized population; (ii) FL2-H:PI (propidium iodide) vs FL1-H:eGFP contour plots for live/dead cell discrimination; and (iii) FL2-H:PI vs FL1-H:eGFP contour plots to quantify eGFP expression with the percentage of eGFP-expressing viable cells shown at the top right of the contour plots. Flow cytometry is an excellent means of quantifying gene expression from different promoters. It should be noted, however, that very high-level eGFP expression from transfected cells can result in increased autofluorescence in the negative population owing to secreted eGFP protein.

with a plasmid not containing eGFP or that have been exposed to supernatant containing a viral vector that does not express eGFP. All populations with greater than 0.01% of eGFP-positive events (Fig. 3iii[A–D]) are considered positive.

108

Bailey and Rasko

5. Adequate analysis of the population can be determined by acquiring 10,000-gated events. Acquisition rates of greater than 4000 events per second should be avoided to ensure the accuracy of the analysis. 6. After acquiring data for all samples, Virkon should be run through the sample line for 10 min to decontaminate and clean it for the next user. Virkon should be flushed out with water for 5 min.

4. Notes 1. Unless stated otherwise, prepare all solutions in triple-distilled water that has a resistivity of 18.2 MΩ/cm. The phosphate buffer may form crystals with prolonged storage at 4°C. 2. 2X HEPES buffer requires careful adjustment of pH with 0.1 M NaOH. If the pH is increased beyond 7.05, do not readjust the pH with HCl, but remake the buffer with smaller additions of NaOH. 3. The EDTA concentration is minimized as the standard concentration (1 mM) will result in chelation of divalent Ca2+ ions and thus reduce precipitate formation. 4. Caution: paraformaldehyde is toxic and may be carcinogenic and therefore requires care in handling. a. Wearing safety glasses and gloves; add 1.0 g paraformaldehyde (Sigma) to 45 mL water. b. Add 4 µL of 10 N NaOH, and warm to 65°C. c. When all solids have dissolved, add 5 mL 10X PBS, and then adjust pH to 7.4. d. Finally, 0.22 µm filter-sterilize to remove debris, and store at 4°C protected from light, for several days maximum. 5. Immortalized cell lines such as HT1080, NIH3T3, and HEK293 cells (ATCC numbers CCL-121, CRL-1658, and CRL-1573, respectively) are ideally suited for transduction experiments. As a general rule, 1 cm2 of a confluent cell monolayer is equivalent to 2 × 105 cells. 6. Chloroquine reduces lysosomal degradation of transfected DNA within the cell and thus increases transfection efficiency (16). For transfection of higher cell numbers, the volumes of reagents are increased proportionally to the increase in surface area. 7. If there are insufficient amounts of plasmid DNA for transfection, sterile fractionated herring sperm DNA can be substituted. Autofluorescent protein expression can be readily detected by flow cytometry with as little as 0.5 to 1.0 µg plasmid DNA. 8. Different excitation and emission filter sets facilitate optimal microscopic viewing of combinations of fluorescent proteins: eGFP (480 ± 40 nM with 510 nM long pass); eCFP (436 ± 20 nM with 480 ± 40 nM long pass); eYFP (510 ± 20 nM with 560 ± 40 nM bandpass); and DsRED variants and cOFP (546 ± 12 nM with 560 nM long pass) (all filters from Chroma Technology, Rockingham, VT). 9. For some applications such as transduction of CD34+ hemopoietic progenitors, 4 µg/mL protamine sulfate may be more suitable as it is less toxic. Coating the culture dish surface with 20 µg/mL Retronectin® (Takara, Shiga, Japan) or fibronectin fragments and then preloading with viral particles should also increase transduction efficiency (17).

Autofluorescent Proteins for Flow Cytometry

109

10. If multiple samples are being prepared, a sterile yellow micropipet tip fitted over the end of a glass Pasteur pipet allows quick aspiration of supernatant and reduces the risk of accidentally aspirating the cell pellet. 11. If cells have a tendency to clump, add FACS wash buffer containing DNase I. Strain the cells through a nylon mesh filter to remove larger clumps and pass through a fine-gauge needle (23 or 26G) before running the sample on a flow cytometer. 12. Paraformaldehyde fixation increases the intrinsic fluorescence of cells. After fixation, cellular autofluorescence will increase over time and potentially compromise the detection of heterologous fluorescent protein reporters (18). 13. In cases of exceedingly bright eGFP-positive cells that cannot be electronically compensated (using % FL2-FL1) and that consequently appear in the FL2 channel (see Figs. 2 and 3), a neutral density filter would reduce the signal detected in FL1. However, this has the effect of reducing fluorescence of the negative population and may make quantitation more troublesome. Figure 3iii(C,D) illustrates typical results observed in a cell population that is expressing eGFP at very high levels. The cross channel “spillover” of eGFP into the FL2 (PI) detector cannot be eliminated by compensation (% FL2-FL1; see Fig. 2B, top left quadrant). Also, afterward, the negative untransfected or untransduced cells can increase in autofluorescence owing to “painting” (or pseudotransduction), which results from high levels of eGFP protein that was secreted into the cell culture supernatant along with viral particles (19).

Acknowledgments The authors thank Cynthia Ng for technical assistance. This work was supported by the NHMRC and the Leukaemia Foundation of Australia. References 1. Chalfie, M., Tu, Y., Euskirchen, G., Ward, W. W., and Prasher, D. C. (1994) Green fluorescent protein as a marker for gene expression. Science 263, 802–805. 2. Heim, R., Prasher, D. C., and Tsien, R. Y. (1994) Wavelength mutations and posttranslational autoxidation of green fluorescent protein. Proc. Natl. Acad. Sci. USA 91, 12,501–12,504. 3. Heim, R. and Tsien, R. Y. (1996) Engineering green fluorescent protein for improved brightness, longer wavelengths and fluorescence resonance energy transfer. Curr. Biol. 6, 178–182. 4. Cormack, B. P., Valdivia, R. H., and Falkow, S. (1996) FACS-optimized mutants of the green fluorescent protein (GFP). Gene 173, 33–38. 5. Yang, T. T., Cheng, L., and Kain, S. R. (1996) Optimized codon usage and chromophore mutations provide enhanced sensitivity with the green fluorescent protein. Nucleic Acids Res. 24, 4592–4593. 6. Shaner, N. C., Steinbach, P. A., and Tsien, R. Y. (2005) A guide to choosing fluorescent proteins. Nat. Methods 2, 905–909.

110

Bailey and Rasko

7. Hawley, T. S., Telford, W. G., Ramezani, A., and Hawley, R. G. (2001) Fourcolor flow cytometric detection of retrovirally expressed red, yellow, green, and cyan fluorescent proteins. Biotechniques 30, 1028–1034. 8. Bevis, B. J. and Glick, B. S. (2002) Rapidly maturing variants of the Discosoma red fluorescent protein (DsRed). Nat. Biotechnol. 20, 83–87. 9. Ip, D., Chan, S.-H., Allen, M., Bycroft, M., Wan, D., and Wong, K.-B. (2004) Crystallization and preliminary crystallographic analysis of a novel orange fluorescent protein from the Cnidaria tube anemone Cerianthus sp. Acta Crystallograph. D60, 340–341. 10. Jordan, M., Schallhorn, A., and Wurm, F. M. (1996) Transfecting mammalian cells: optimization of critical parameters affecting calcium-phosphate precipitate formation. Nucleic Acids Res. 24, 596–601. 11. Pear, W. (1996) Transient transfection methods for high-titre retroviral supernatants, in Current Protocols in Molecular Biology (Ausubel, F. M., Brent, R., Kingston, R. E., et al., eds.), John Wiley & Sons, New York, pp. 9.11.11–9.11.18. 12. Miller, A. D. and Rosman, G. J. (1989) Improved retroviral vectors for gene transfer and expression. Biotechniques 7, 980–982, 984–986, 989–990. 13. Bartz, S. R. and Vodicka, M. A. (1997) Production of high-titer human immunodeficiency virus type 1 pseudotyped with vesicular stomatitis virus glycoprotein. Methods 12, 337–342. 14. Koldej, R., Cmielewski, P., Stocker, A., Parsons, D. W., and Anson, D. S. (2005) Optimisation of a multipartite human immunodeficiency virus based vector system; control of virus infectivity and large-scale production. J. Gene Med. 7, 1390– 1399. 15. Forestell, S. P., Dando, J. S., Bohnlein, E., and Rigg, R. J. (1996) Improved detection of replication-competent retrovirus. J. Virol. Methods 60, 171–178. 16. Luthman, H. and Magnusson, G. (1983) High efficiency polyoma DNA transfection of chloroquine treated cells. Nucleic Acids Res. 11, 1295–1308. 17. Hanenberg, H., Xiao, X. L., Dilloo, D., Hashino, K., Kato, I., and Williams, D. A. (1996) Colocalization of retrovirus and target cells on specific fibronectin fragments increases genetic transduction of mammalian cells. Nat. Methods 2, 876– 882. 18. Rasko, J. E. J. (1999) Reporters of gene expression: autofluorescent proteins, in Current Protocols in Cytometry (Robinson, J. P., Zbigniew, P. N., Dressler, L. G., et al., eds.), John Wiley & Sons, Inc, New York, pp. 9.12.11–19.12.16. 19. Alexander, I. E., Russell, D. W., and Miller, A. D. (1997) Transfer of contaminants in adeno-associated virus vector stocks can mimic transduction and lead to artifactual results. Hum. Gene Ther. 8, 1911–1920.

Reporter Systems for Single-Cell Measurements

111

8 Fluorescent Protein Reporter Systems for Single-Cell Measurements Steven K. Dower, Eva E. Qwarnstrom, and Endre Kiss-Toth Summary The unraveling of the complex dynamic networks that underlie cellular (and, by extension, tissue, organ, and organism) function requires sophisticated mathematical models and, in order to test those models, rich data sets. In addition, even in clonal populations of cells, there is a wide range of variability in cellular function at any given time, even in simple parameters such as the concentration of critical signaling components such as receptors or transcription factors. It remains a matter of conjecture as to whether this is noise, to which the system is inherently robust, or whether the cellular control network can exist in multiple discrete internal states, with indistinguishable input/output characteristics. Fluorescent protein-based methods have two features useful for addressing these issues. First, they can be used to retrieve data from individual cells. Second, in combination with confocal fluorescence microscopy, they can be used nondestructively and can thus follow one or more individual cells in culture or in an intact organism over time. Key Words: Fluorescent proteins; confocal microscopy; reporters; signal transduction; transcription; cytokines.

1. Introduction The discovery that the green fluorescent protein (GFP) from the Aequorea victoria jellyfish autocatalytically converts endogenous amino acid side chains when becoming fluorescent has had a significant technological impact on cell biology, as it allows reporters to be introduced into cells, tissues, and organisms by using fluorescent protein coding sequences, either as plasmids or in genomic DNA. Subsequently fluorescent proteins with a wide range of spectral properties have been found both by searching for the molecular basis of fluorescence in other organisms and by mutagenesis of known fluorescent proteins. There are now approx 50 different fluorescent proteins in use in the life sciences. As an example, a widely used variant of GFP, enhanced EGFP, was From: Methods in Molecular Biology, vol. 411: Reporter Genes: A Practical Guide Edited by: D. Anson © Humana Press Inc., Totowa, NJ

111

112

Dower, Qwarnstrom, and Kiss-Toth

selected for spectral properties that closely map to those of fluorescein, a dye that has been a standard fluorescent label in the life sciences for decades, and for which therefore standard filter sets exist for all instruments. In addition, the excitation spectrum of fluorescein is maximal close to the 488-nm line of the widely available gas lasers Ar or Kr/Ar. EGFP also contains additional mutations that decrease the autocatalytic conversion time and increase stability at 37°C. It has also been codon optimized for mammalian systems. Fluorescent proteins are used in two broad categories of assays, both of which are illustrated in this chapter. In the first they are used in the same way as other transcription reporters (such as luciferase) to monitor expression of genes or activity of promoters in a variety of settings (1). Using the same approach, fluorescent proteins can track lineages of cells in living organisms if they are introduced into the genome at a gene known to be expressed in that lineage. In the second type of assay, they are used in the form of fusion proteins, in which the cDNA for the fluorescent protein is fused in frame to that encoding a protein of interest, and the expressed product is used to monitor any of an increasingly wide range of cellular processes involving that protein, for example, location and trafficking, posttranslational modification, interactions with other proteins, turnover, and degradation (2–6). Again, all of these can be applied in whole organisms such as zebrafish. Although descriptions of instrumentation are beyond the scope of this chapter, a brief summary is important, as different instruments have limitations and advantages in monitoring fluorescent protein expression and properties. Fluorescent proteins, like other labels, can be quantified in a wide range of fluorimeters. For example, in in vitro gene expression assays on cultured cells, fluorescent proteins have the advantages over luciferase that no additional reagents are needed, the cultures can be assayed directly for fluorescence, and, in addition, since no manipulations of the culture are required, repeated measurements can be taken from the same sample. The main drawback of such fluorescent proteinbased reporter assays is that measurements on a fluorimeter/luminometer are much less sensitive than, for example, luciferase assays, because of the lack of a signal amplification step. Fluorescent proteins can easily be measured quantitatively by flow cytometry. Although this is more sensitive than fluorimetry and returns more information (since it serially measures individual cells), primarily population distribution of the label, it clearly has the drawback that the sample is at a minimum stressed by shear forces during measurement and of course is usually discarded. The most information-rich method for retrieving data from fluorescent protein-labeled samples is fluorescence microscopy; with certain caveats, it can also yield quantitative information if the instrument is confocal. It is also nondestructive if the stage is set up to provide the appropriate environment for the

Reporter Systems for Single-Cell Measurements

113

sample and the excitation light is attenuated to minimize damage: a compromise with delivering useful output signal. Unlike the other methods, fluorescence microscopy provides subcellular data. If measurements are taken on a confocal instrument, the image can be used to extract concentration information; the critical parameter here is to ensure that the optical section depth is less than the depth of the sample, otherwise the fluorescence intensity will be proportional to the fraction of the optical section occupied by the sample. This is achieved by a suitable combination of confocal aperture and objective lens numerical aperture (NA). As guide— with an NA of more than 1 and a confocal aperture of 50 µm— at approx 500-nm fluorescence emission wavelength, the optical section will be approx 0.5 µm; this will be less than the thickness near the center of, for example, a human fibroblast adhered to a cover slip. 2. Materials 2.1. EGFP Transcription Reporter Assays 1. The 5' noncoding region of the human interleukin 8 (IL-8) gene (accession no.: M28130) was cloned by polymerase chain reaction (PCR) amplification. The PCR primers were: sense from 1308 to 1325 and antisense from 1647 to 1629 of the IL8 gene genomic sequence. 2. A 220-bp BglII–HindIII fragment of this PCR product, containing the 59-bp noncoding region and the transcription start, was subcloned into pEGFP1 and pd2EGFP1 vectors (Clontech, Palo Alto, CA), yielding plasmids pIL8/EGFP1 and pIL8/d2 EGFP1. 3. HeLa cells (the European Collection of Cell Cultures [ECACC], cat. no. 85060701). 4. Nunc Chambered Coverglasses or Chambered Slides (1-, 2-, 4-, or 8-well) (Fisher Scientific, Loughborough, Leics, UK). 5. Nunc 96-well tissue culture plates (Fisher Scientific). 6. Cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM; Life Technologies), containing 10% fetal calf serum (Hyclone, Ogden, UT). 7. Cells were transfected using SuperFect (Qiagen, Crawley, UK).

2.2. EGFPRelA Fusion Protein Assays 1. The plasmid pEGFPRELA containing the p65-RELA cDNA fused in frame to the carboxyl terminus of EGFP downstream of the cytomegalovirus (CMV) promoter was constructed by subcloning a HindIII to BamHI fragment from pBluescriptRELA (4) into pEGFP.C1, followed by infilling of the XhoI site using Klenow. 2. Human gingival fibroblasts were a kind gift of Dr. R. Page (University of Washington). 3. Cells were cultured in DMEM (Life Technologies), containing 10% fetal calf serum (Hyclone). 4. Nunc Chambered Coverglasses (1-, 2-, 4-, or 8-well) (Fisher Scientific).

114

Dower, Qwarnstrom, and Kiss-Toth

3. Methods 3.1. EGFP Transcription Reporter Assay The assay system was developed using Nunc 8-well chambered cover glasses in combination with a Nikon Diaphot inverted microscope interfaced with a Molecular Dynamics CSLM 2010 Kr/Ar laser confocal microscope and incorporating a 37°C stage incubator and a programmable stage drive for repeated imaging of multiple individual live cells. Cover glasses were sealed to avoid the need for repeated gassing, and thus could be maintained in the stage incubator for 4 h without the need to return to the laboratory incubator. Use of the chambered cover glasses allowed use of the high-NA objectives 60X oil, and 100X oil and hence quantitative data acquisition. With these objectives, it is important to use a blown air enclosure stage (cage) incubator rather than a closed stage mounted incubator and to turn the air on at least 30 min before beginning an experiment, to give the objectives time to warm up to 37°C. If this is not done, then heat transfer through the oil causes variable expansion of the objective, and it is impossible to stabilize the focal plane. Subsequently we compared the results from this method, which measured EGFP concentration in a number of individual cells, with imaging in 96-well tissue culture plates with a low NA (10X air) objective— here measuring total EGFP in the field of view since the cell monolayer would be contained totally within the optical section. The results were comparable, and we switched to the latter method, for cost and convenience reasons. The reporter assay was validated using a luciferase version of the IL-8 promoter reporter plasmid. The fold induction of luciferase was compared with the fold induction of secreted IL-8 protein (expressed from the endogenous Il8 gene) in HeLa cells cotransfected with reporter and expression plasmids for human Toll-like receptor-1 and -2 and exposed to 10 µg/mL S. minnesota Re595 lipopolysaccharide (5). 1. HeLa cells (1.5 × 104 per well) were seeded into 96-well tissue culture plates 24 h prior to transfection. This provides adherent cultures that are 70% confluent at time of use (see Note 1). 2. Transfections were performed using SuperFect (Qiagen). a. A total of 1 µg of DNA was used, consisting of 0.5 µg of IL-8 promoter reporter construct, up to 0.5 µg of plasmid-encoding proteins that transactivate the IL-8 promoter, and an appropriate amount of empty pcDNA3.1 to keep the DNA concentration constant. b. DNA was diluted in a total of 30 µL of serum-free DMEM, 2.5 µL of SuperFect were added, and the sample was vigorously mixed. c. The mixture was incubated for 7 to 10 min at room temperature, supplemented with 150 µL DMEM containing 10% FBS and added to the cells (see Note 2).

Reporter Systems for Single-Cell Measurements

115

3. Cultures were incubated at 37°C for 3 h, and the medium was replaced by 150 µL DMEM containing 10% FBS. 4. If the cultures are used for reporter assays for soluble mediators, and were therefore transfected with reporter plasmid and empty vector only (see Note 2), then at this point 50 µL of an appropriate dilution of agonist is added and the cultures are incubated at 37°C for 6 h. In experiments using 1 nM recombinant human interleukin-1β as the agonist and pIL8/d2EGFP1 as the reporter, this is a time that we have established as being sufficient for fluorescence to reach a steady state. 5. Twenty-four hours after transfection, cultures were washed twice with DMEM without phenol red and cultured with this medium containing 10% FBS during the course of the experiments. Using the CLSM 2010 confocal microscope, three confocal images were taken from each well. 6. The digital images obtained were from a sample section composed of 1024 × 1024 voxels of dimensions 1.3(x) × 1.3(y) × 5.3(z) µm. 7. Analysis was done on a Macintosh computer with software written in Microsoft Foxpro, Excel, and the NIH Image macro language. 8. To examine individual cells, the Analyse Particles algorithm of NIH Image v1.62 was used. A particle was defined as more than 50 pixels in area. The algorithm returns the mean pixel intensity of the particle (B), which represents the brightness of the cell, and the particle area (A). The product of these values is approximately proportional to the total EGFP content of the cell (see Notes 3 and 5). 9. A convenient way to represent the data is to plot the area on the x-axis and the brightness on the y-axis, each individual cell (particle) being a point on the scatter plot. This allows for a qualitative assessment of cellular viability, as cells that are small and bright lie along the y-axis, and were found to be annexin V positive (i.e., apoptotic) in most instances. Typical results are shown in Fig. 1.

3.2. NF-κ B Translocation Assay To illustrate the use of fluorescent protein fusions for monitoring cellular processes, the application of an EGFP-RelA fusion to analysis of cellular signaling processes is described. This is one of the most well-characterized eukaryotic transcription factor systems. Briefly, the classic nuclear factor-κB (NF-κB) protein is a heterodimer of RelA (p65) and NF-κB p50. This heterodimer binds to a characteristic and approximately palindromic, 10-bp DNA sequence and recruits RNA polymerase via the p65 component. In its inactive form the heterodimer is retained in the cytoplasm in a complex with an inhibitor from the inhibitor κB (IκB) family. A critical step in the activation process occurs when the IκB is phosphorylated by IκB kinase and subsequently becomes ubiquitinated and degraded via a classical proteasome-mediated process. This unmasks a nuclear localization signal on NF-κB, which is then actively transported into the nucleus, where it binds to cognate recognition sites in genomic DNA and initiates transcription. The correlation of the biological function with movement

116

Dower, Qwarnstrom, and Kiss-Toth

Fig. 1. Reproducibility of the pIL8-d2EGFP response to FADD and TRAF-6 cotransfection. Hela cells were transfected with (1) pIL8-d2EGFP (500 ng) and empty vector (500 ng; control); (2) pIL8-d2EGFP (500 ng), FADD (2 ng), and empty vector (500 ng); and (3) d2EGFP (500 ng), TRAF6 (2 ng), and empty vector (500 ng). Images were taken from five separate wells for each condition (three fields per well) and postprocessed. The data are plotted as described in the text. (From ref. 1.)

from cytoplasm to nucleus makes this system an ideal one for use of fluorescent protein tags and for input of the data from such studies into quantitative mathematical models. There have been a number of studies of this aspect of the NFκB system in the last few years, including those from our own laboratory. 1. Human gingival fibroblasts (three lines) were maintained in DMEM containing 10% fetal calf serum. 2. Cells were transfected by calcium phosphate coprecipitation with glycerol shock (60 s, 20% glycerol in PBS) 4 h after transfection, typically yielding 10 to 20% transfection efficiency. Cells were plated on 8-chamber coverslips (Nunc) at 10,000

Reporter Systems for Single-Cell Measurements

3.

4.

5. 6. 7.

117

cells/well in 0.5 mL of medium 24 h prior to transfection. For analysis of kinetics, 0.5 µg of pEGFPRELA was transfected per chamber in a total volume of 50 µL. For cotransfection experiments, a total of 1.0 µg of plasmid was transfected using empty vector (pcDNA3.1) as required. For confocal microscopy, cells were washed twice with DMEM without phenol red and incubated with the same medium plus 10% FBS during the course of the experiments. EGFP fusion protein was visualized using a confocal laser scanning microscope with a 37°C stage incubator (see Subheading 3.1.). Standard settings for analysis of EGFP were laser power 5 to 10 mW, excitation line selection 488 nm, and PMT voltage 667, with the attenuation varied to maintain pixel density below 200 (8-bit grayscale), which is within the linear range of the instrument. Time course data were collected with a 60X oil immersion objective (NA 1.4) and 50-µm aperture, generating an optical section of 0.54 µm. For quantitation of nuclear and cytoplasmic protein levels, horizontal sections were scanned through the nucleus of transfected cells. Images were exported and analyzed using NIH image. Relative fluorescence was calculated by measuring the mean intensity of representative areas of nucleus and cytoplasm and dividing by the attenuation setting. When multiple images of the same cell were analyzed, the readings were corrected for fade by comparison with measurements made using unstimulated cells. (For a discussion of quantitation, see Notes 4 and 5.) Typical results are shown in Fig. 2.

4. Notes 1. It is important that the cultures be subconfluent, or the transfection efficiency and viability fall significantly. 2. Typical transfection efficiencies are between 30 and 60%, if these are too high, the analyze particles macro cannot distinguish individual cells. 3. Quantitation of data: for fluorescence-based quantitation of protein concentrations, the fluorescence of a series of dilutions of purified EGFP spotted on a cover slip was determined. A relative fluorescence of 1 corresponds to an EGFP concentration of 67 µg/mL (in other units: 2.47 µM or 1500 molecules/µm3). The average volume of a fibroblast is 2000 µm3, so if the purified EGFP has similar fluorescent properties to that expressed in the fibroblasts, a relative cytoplasmic fluorescence of 1 corresponds to approx 3 × 106 molecules of fusion protein per cell. To obtain an estimate based on immunoblot analysis, cells were transfected with pCMV-EGFP, and the mean fluorescence was determined by confocal microscopy to be 13, suggesting an average EGFP expression level per cell of 3.9 × 106 molecules. However, the same cells analyzed by immunoblot against EGFP standards indicated a mean expression level of 0.7 × 106 molecules per cell, i.e., approx sixfold lower than by fluorescence. The most likely cause of this discrepancy is nonoptimal fluorescence of the purified EGFP, so the lower estimate was used in our work. 4. Comparison of endogenous RelA levels with transfected EGFPRelA levels in individual cells.

118

Dower, Qwarnstrom, and Kiss-Toth

Fig. 2. Analysis of EGFPRelA redistribution in single cells. (A) Visualization of relocalization in an individual cell after induction with 1 nM IL-1β. Images correspond to 0, 5, 10, 20, 30, and 60 min after stimulation, as indicated. (B) Quantitation of nuclear and cytoplasmic levels in representative cells induced with 1 nM IL-1β demonstrating cell-to-cell variability of response. Diamond, cytoplasmic fluorescence; circle, nuclear fluorescence; square, nuclear/cytoplasmic fluorescence ratio. The top left graph corresponds with the cell shown. (From ref. 3.) See accompanying CD for color version.

a. Fibroblasts were transfected as above, and images of EGFP fluorescence were recorded at 24 h. b. These cells were then fixed/permeabilized in methanol (5 min, −10°C) and then blocked overnight at 4°C in PBS + 5% normal goat serum.

Reporter Systems for Single-Cell Measurements

119

c. Immunostaining was performed with primary antiserum (2 mg/mL). d. The cells were washed three times in blocking buffer, incubated with biotinylated secondary antiserum (sc-2040, Santa Cruz Biotechnology), washed three times in blocking buffer, incubated with streptavidin/Texas Red (Molecular Probes), washed three times in blocking buffer and twice in PBS, and then mounted using Prolong Anti-Fade Reagent (Molecular Probes). e. Previously scanned cells were identified and rescanned as for EGFP fluorescence except using excitation at 568 nm and a photomultiplier tube at 750 V. There was a linear relationship between EGFP fluorescence and red immunofluorescence, giving an eightfold overexpression of EGFPRelA compared with endogenous RelA at relative fluorescence 1. Calibrating the system as above indicates that relative fluorescence 1 corresponds to expression of approx 300,000 molecules per cell. Thus the endogenous level of RelA expression is approx 40,000 molecules per cell. 5. The values given above are for guidance only and to illustrate procedures. Clearly, the conversion of fluorescence into concentration will be instrument, objective, and culture device dependent, at any given laser power/attenuation/PMT voltage settings, and each instrument and experimental system must be independently calibrated.

Acknowledgments This work was supported by the Medical Research Council, The Wellcome Trust, the Arthritis Research campaign, and the British Heart Foundation. References 1. Kiss-Toth, E., Guesdon, F. M. J., Wyllie, D., Qwarnstrom, E. E., and Dower, S. K. (2000) A novel mammalian expression screen exploiting green fluorescent proteinbased transcription detection in single cells. J. Immunol. Methods 239, 125–135. 2. Maamra, M., Finidori, J., Von Laue, S., et al. (1999) Real time regulation of full length and truncated growth hormone receptor trafficking by its ligand and antagonist. J. Biol. Chem. 274, 14,791–14,798. 3. Carlotti, F., Chapman, R., Dower, S. K., and Qwarnstrom, E. E. (1999) Interleukin1 induced nuclear uptake of NF-κB in single living cells: concentration dependence of EGFPrelA redistribution and anti-apoptotic function. J. Biol. Chem. 274, 37,941– 37,949. 4. Carlotti, F., Dower, S. K., and Qwarnstrom, E. E. (2000) Dynamic shuttling of nuclear factor kappa B between the nucleus and cytoplasm as a consequence of inhibitor dissociation. J. Biol. Chem. 274, 41,028–41,034. 5. Wyllie, D. H., Smith, S. C., Visintin, A., et al. (2000) Evidence for an accessory protein function for TLR-1 in anti-bacterial responses. J. Immunol. 165, 7125–7132. 6. Yang, L., Ross, K., and Qwarnstrom, E. E. (2003) RelA regulation of IκB phosphorylation— A positive feedback-loop for high affinity NF κB complexes. J. Biol. Chem. 278, 30,881–30,888.

Subcellular Imaging of Cancer Cells

121

9 Subcellular Imaging of Cancer Cells in Live Mice Robert M. Hoffman Summary Dual-color fluorescent cells, with one color in the nucleus and the other in the cytoplasm, enable real-time nuclear-cytoplasmic dynamics to be visualized in living cells in vivo as well as in vitro. To obtain the dual-color cells, red fluorescent protein (RFP) is expressed in the cytoplasm of cancer cells, and green fluorescent protein (GFP) linked to histone H2B is expressed in the nucleus. Nuclear GFP expression allows visualization of nuclear dynamics, whereas simultaneous cytoplasmic RFP expression allows visualization of nuclear-cytoplasmic ratios as well as simultaneous cell and nuclear shape changes. This methodology has allowed us to show that the cells and nuclei of cancer cells in the capillaries elongate to fit the width of these vessels. The average length of the major axis of the cancer cells in the capillaries increased to approximately four times their normal length. The nuclei increased their length 1.6 times. Cancer cells in capillaries over 8 µm in diameter were shown to migrate at up to 48.3 µm/h. With the use of dual-color fluorescent cells and the Olympus OV100, a highly sensitive whole-mouse imaging system with both macrooptics and microoptics, it is possible to achieve subcellular real-time imaging of cancer cell trafficking in live mice. Extravasation can also be imaged in real time. Dual-color imaging showed that cytoplasmic processes of cancer cells exited the vessels first, with nuclei following along the cytoplasmic projections. Dual-color in vivo cellular imaging was also used to visualize trafficking, nuclear-cytoplasmic dynamics, and the viability of cancer cells after their injection into the portal vein of mice. Key Words: Green fluorescent protein; red fluorescent protein; in vivo imaging; nuclear-cytoplasmic dynamic imaging; cancer cell trafficking; metastasis.

1. Introduction Visualization of microscopic cancer is essential for the understanding and control of cancer dormancy, growth, and colonization of distant sites (1–3). Several approaches involving tumor-cell labeling have been developed for visualizing tumor cells in vivo. The Escherichia coli β-galactosidase (lacZ) gene has been used to detect micrometastases (4). However, lacZ detection requires extensive histological preparation and sacrifice of the tissue or animal. ThereFrom: Methods in Molecular Biology, vol. 411: Reporter Genes: A Practical Guide Edited by: D. Anson © Humana Press Inc., Totowa, NJ

121

122

Hoffman

fore, other techniques are required for real-time imaging and study of tumor cells in viable fresh tissue or living animals (5). The use of skin-fold chambers, exteriorization of organs, and subcutaneous windows inserted with semitransparent material (6–8) has yielded insights into microscopic tumor behavior. However, these techniques are only suitable for ectopic models (6) or for investigations with short periods of observation (7,8). The difficulties of maintaining windows or other devices that are made with heavy, stiff, or other types of foreign materials limit the length of time that these tools can be used in vivo (6,7). Cutaneous windows made with polyvinyl chloride film become opaque or detached after time, which precludes their use in long-term studies (5,7). To image noninvasively and follow the natural course or impediment of tumor progression and metastasis, high specificity and sensitivity, a strong signal, and high resolution are necessary. The green fluorescent protein (GFP) gene, cloned from the bioluminescent jellyfish Aequorea victoria (9), was chosen to satisfy these conditions because it has great potential for use as a cellular marker (10,11). GFP cDNA encodes a 283-amino acid monomeric polypeptide with Mr 27,000 (12,13) that requires no other A. victoria proteins, substrates, or cofactors to fluoresce (14). Gain-of-function bright mutants of the GFP gene have been generated by various techniques (15–17) and have been humanized for high expression and signal (18). Red fluorescent proteins (RFPs) from the Discosoma coral have similar features as well as the advantage of longer wavelength emission (5,19–21). Initial studies of tumor biology that used stable GFP expression focused on static images and examination of metastases (22,23). The first use of GFP to image cancer cells in vivo was by Chishima et al. in our laboratory (24). GFP was subsequently used to observe motility and shape changes of carcinoma cells in vivo (25,26). Subsequently, Chishima et al. (24) and Huang et al. (27) showed that GFPtransduced cancer cells allowed the imaging of tumor cells in blood vessels. To examine cell behavior during intravasation, Naumov et al. (8) used GFP imaging to visualize fine cellular details such as pseudopodial projections, even after extended periods of in vivo growth. Wyckoff et al. (28) have used GFP imaging to view these cells in time-lapse images within a single optical section using a confocal microscope. Mook et al. (29) visualized GFP colon cancer cells in sinusoids of the liver. Al-Mehdi et al. (30) observed the steps in early hematogenous metastasis of tumor cells expressing GFP in subpleural microvessels in intact, perfused mouse and rat lungs. Using multiphoton microscopy and GFP labeling, Wang et al. (31) examined differences in carcinoma cell behavior between the nonmetastatic and meta-

Subcellular Imaging of Cancer Cells

123

static primary breast tumors. Goswami et al. (32) have shown that macrophages promote the invasion of GFP-labeled breast cancer cells. However, nuclearcytoplasmic dynamics could not be visualized in the trafficking cells because the cancer cells were entirely labeled with GFP (1). The visualization of nuclear cytoplasmic dynamics in living cells could enable the real-time study of the normal and malignant cell cycle and apoptotic behavior in vivo as well as in vitro. Chambers et al. (33) and Naumov et al. (8) used GFP-tagged tumor cells and intravital imaging to visualize individual cells. These studies visualized the shape of metastatic tumor cells in vivo but could not visualize nuclearcytoplasmic dynamics, or nuclear shape changes, because the nucleus and cytoplasm could not be distinguished (5). A fusion protein of GFP and yeast histone H2B was shown to localize in yeast nuclei (34). Subsequently, a fusion protein of GFP and human histone H2B (H2B-GFP) was shown to be incorporated into nucleosome core particles of HeLa cells without perturbing cell cycle progression (35). H2B-GFP allowed high-resolution imaging of both mitotic chromosomes and interphase chromatin in live cells (5,36). With only H2B-GFP labeling of cells, an overlay with differential interference contrast images along with the GFP fluorescence images was necessary to visualize nuclear-cytoplasmic morphology. We have used H2B-GFP and RFP to differentially label the nucleus and cytoplasm of human HT-1080 human fibrosarcoma cells. This strategy allows the visualization of the cell cycle, apoptosis, and nuclear deformability in live cells in real time. This dual-color tagging strategy also allows real-time observation of nuclear-cytoplasmic dynamics in vivo as well as in vitro (5). Using the dual-colored cancer cells and a highly sensitive small-animal macroimaging/microimaging system, the Olympus OV100, real-time dynamic subcellular imaging of cancer cell trafficking in live mice is possible. We described how this in vivo subcellular imaging technology can be used to visualize the cytoplasmic and nuclear dynamics of intravascular tumor cell migration and extravasation in live mice (1). This chapter assumes that the investigator has expertise in microsurgery in the mouse. A useful reference for those learning this field is Experimental Microsurgery (36a). 2. Materials 2.1. Transduction of Cancer Cells 1. PT67 LHCX2-H2B-GFP producer cells (5) (see Note 1). 2. PT67 LNCX2-DsRed2 producer cells (5) (see Note 1). 3. Dulbecco’s modified Eagle’s medium (DMEM; Irvine Scientific, Santa Ana, CA).

124 4. 5. 6. 7.

Hoffman

Heat-inactivated fetal bovine serum (Gemini Bio-products, Calabasas, CA). G418 (Life Technologies). Hygromycin (Life Technologies). Cloning cylinders (Bel-Art Products, Pequannock, NJ).

2.2. Mice 1. Athymic nu/nu nude mice. Mice should be kept in a barrier facility under HEPA filtration and fed with autoclaved rodent food. 2. Tecklad LM-485 autoclaved laboratory rodent diet (Western Research Products, Orange, CA).

2.2. Delivery of Tumor Cells 1. Ketamine anesthetic: 10 µL ketamine HCl, 7.6 µL xylazine, 2.4 µL acepromazine maleate, and 10 µL H2O. 2. Dissecting instruments: a. Blunt-end hook (Fine Science Tools, Foster City, CA). b. 33-gage Needle (Fine Science Tools). c. 30-gage Needle (Fine Science Tools). d. Dissecting microscope, (MZ6; Leica, Deerfield, IL). e. 6-0 Suture (Ethicon, Somerville, NJ). f. 1 mL Latex-free syringe (Becton Dickinson).

2.3. Imaging of Tumor Cells 1. Phosphate-buffered saline (PBS; Irvine Scientific). 2. Olympus OV100 small animal imaging system (see Note 2).

3. Methods 3.1. RFP and Histone H2B-GFP Gene Transduction of Cancer Cells For RFP and H2B-GFP gene transduction, 70% confluent human cancer cells were used. To establish dual-color cells, clones of cancer cells expressing RFP in the cytoplasm were initially established. 1. Incubate cancer cells with a 1:1 mixture of retroviral supernatant from PT67-RFP cells and normal growth medium for 72 h (see Note 1). 2. Aspirate media and replace with normal growth medium. 3. At 72 h posttransduction, harvest cells and subculture cells at a ratio of 1:15 in growth medium containing 200 to 800 µg/mL G418. 4. Isolate clones of transduced cancer cells, identified by fluorescence microscopy, with cloning cylinders and amplify by conventional culture methods.

For establishing dual-color cells, the process is then repeated on a clone of cancer cells expressing high levels of RFP using retroviral supernatants from PT67 H2B-GFP cells and selecting with 200 to 400 µg/mL of hygromycin. The

Subcellular Imaging of Cancer Cells

125

resulting clones should be checked for stably expressed GFP in the nucleus and RFP in the cytoplasm using fluorescence microscopy. 3.2. Real-Time Visualization of Dual-Labeled Cells In Vivo Examples are given for the analysis of cells in the brain, ear, and capillaries. Cells should be harvested by trypsinization and washed twice with serum-free medium before injection. Cells should be used within 40 min of harvesting. 3.2.1. Imaging of Cells in the Brain of Live Mice To visualize cell dynamics in the brain of living mice, cells were injected in the common carotid artery. All procedures of the operation described are performed with a x7 dissection microscope 1. Anesthetize nude mice with ketamine mixture via s.c. injection. 2. Make a longitudinal skin incision on the neck. a. After exposing the submandibular gland, cut it in the middle and retract to each side. b. Separate the right sternohyoid muscle, right sternomastoideus muscle, and connective tissue with a blunt instrument. c. After isolation of the right common carotid artery, gently release the artery from surrounding connective tissue. d. Place light tension on the proximal site of the artery with a blunt-end hook. 3. Inject a total of 200 µL of medium containing 2 × 105 dual-color cells into the artery using a 33-gage needle. Immediately after injection, press the injected site with a swab to prevent bleeding or leakage of injected tumor cells. Close the skin with a 6-0 suture.

Visualize (see Note 2) tumor cells in the brain through the skull via a skinflap window. 1. Anesthetize the animals with the ketamine mixture via s.c. injection. 2. Make an arc-shaped incision in the scalp, and separate the s.c. connective tissue to free the skin flap. 3. The skin flap can be opened repeatedly to image (see Note 2) tumor cells in the brain through the nearly transparent mouse skull and can be simply closed with a 6-0 suture. This procedure greatly reduced the scatter of fluorescent photons (5).

For examples of the sort of results that can be obtained with this technique, see ref. 5. 3.2.2. Noninvasive Imaging of Cells in the Ear 1. Anesthetize nude mice with ketamine mixture via s.c. injection. 2. Inject dual-color cells intradermally into the ear of the mouse using a 27-gage needle.

126

Hoffman

3. Observe the surface of the ear of the intact animal under fluorescence microscopy (5) (see Note 2).

For examples of the sort of results that can be obtained with this technique, see ref. 5. 3.2.3. Imaging of Cells in Blood Vessels To visualize cell dynamics in vessels in live mice, cells are injected into the heart. During the period of measurement, the animal is kept under anesthesia and kept warm. To determine migration velocities, measurements are taken at the initial time and 2 h later. Blocking off the epigastric cranialis vein traps the cells in the vein and forces them into the surrounding capillaries at the observation site (abdominal skin flap). 1. Anesthetize nude mice with ketamine mixture via s.c. injection. 2. Expose the ribs using a 1-cm midline incision. a. Using a 30-gage 1/2 needle and a 1-mL latex-free syringe, inject a total of 100 to 200 µL of medium containing 1 to 5 × 106 dual-color cells into the heart ventricle. b. This is done by inserting the needle into the second intercostal space 2 mm left of the sternum and aimed centrally. The spontaneous, pulsatile entrance of bright red oxygenated blood into the needle hub indicates proper positioning. c. The cells should be injected over 1 to 2 min. 3. Immediately after injection, make an arc-shaped incision in the abdominal skin, and then separate the s.c. connective tissue to free the skin flap without injuring the epigastric cranialis artery and vein. 4. Expose the epigastric cranialis vein of the mouse and block with a 6-0 suture. 5. Spread the skin flap and fix on a flat stand. 6. Visualize cells (see Notes 2 and 3). 7. During the intervals between imaging, occasionally spray PBS on the inside of the skin flap to keep the surface wet (see Notes 4–6).

For examples of the sort of results that can be obtained with this technique, see refs. 1 and 26. 4. Notes 1. Any vector that expresses high levels of the relevant fluorescent proteins can be used. A selective marker on each vector simplifies isolation of transduced cells, although this could also be achieved by fluorescence-activated cell sorting (FACS) or cloning/FACS analysis. 2. The Olympus OV100 Small Animal Imaging System (Olympus Corp., Tokyo, Japan), containing an MT-20 light source (Olympus Biosystems, Planegg, Germany) and DP70 CCD camera (Olympus), was used for subcellular imaging in live mice. The optics of the OV100 fluorescence imaging system have been specially developed for macroimaging as well as microimaging with high light-gath-

Subcellular Imaging of Cancer Cells

3. 4. 5. 6. 7. 8. 9.

127

ering capacity. The instrument incorporates a unique combination of high numerical aperture and long working distance. Five individually optimized objective lenses, parcentered and parfocal, provide a 105-fold magnification range for seamless imaging of the entire body down to the subcellular level without disturbing the animal. The OV100 has the lenses mounted on an automated turret with a high magnification range of 1.6 to 16 and a field of view ranging from 6.9 to 0.69 mm. The optics and antireflective coatings ensure optimal imaging of multiplexed fluorescent reporters in small animals. High-resolution images were captured directly on a PC (Fujitsu Siemens, Munich, Germany). Images were processed for contrast and brightness and analyzed with the use of PaintShopPro 8 and Cell® (Olympus Biosystems) (1). The images should include the cell in its blood vessel, as well as the surrounding vessels, which can be used as a map to relocate the cell in its vessel at later time points. The skin flap can be completely reversed (26). For motility analysis, the epigastric cranialis vein was not wired. For motility analysis, cells were reimaged after 2 h. For imaging cancer cell trafficking in blood vessels, images were acquired in real time. For analysis of extravasation, images were acquired every hour after injection with the skin flap open, or every 12 h by opening and closing the skin flap. Using the skin flap for observation of cells in capillaries has important advantages. The skin can be spread stably on a stand, such that motion from the mouse’s heartbeat or breathing has no influence on imaging. Disturbance of the blood supply for the skin does not occur during the skin flap procedure, because the epigastric cranialis artery is not injured during the procedure. In addition to these advantages, the skin flap could be completely reversed such that the mice need not be sacrificed. In our study, the skin flap was reversed after 24 h (26).

Acknowledgments These studies were supported in part by National Cancer Institute grants CA103563 and CA099258. References 1. Yamauchi, K., Yang, M., Jiang, P., et al. (2006) Development of real-time subcellular dynamic multicolor imaging of cancer-cell trafficking in live mice with a variable-magnification whole-mouse imaging system. Cancer Res. 66, 4208–4214. 2. Tsuji, K., Yamauchi, K., Yang, M., et al. (2006) Dual-color imaging of nuclearcytoplasmic dynamics, viability, and proliferation of cancer cells in the portal vein area. Cancer Res. 66, 303–306. 3. Chambers, A. F., Groom, A. C., and MacDonald, I. C. (2002) Dissemination and growth of cancer cells in metastatic sites. Nat. Rev. Cancer 2, 563–572. 4. Lin, W. C., Pretlow, T. P., Pretlow, T. G. II, and Culp, L. A. (1990) Bacterial lacZ gene as a highly sensitive marker to detect micrometastasis formation during tumor progression. Cancer Res. 50, 2808–2817.

128

Hoffman

5. Yamamoto, N., Jiang, P., Yang, M., et al. (2004) Cellular dynamics visualized in live cells in vitro and in vivo by differential dual-color nuclear-cytoplasmic fluorescent-protein expression. Cancer Res. 64, 4251–4256. 6. Brown, E. B., Campbell, R. B., Tsuzuki, Y., et al. (2001) In vivo measurement of gene expression, angiogenesis and physiological function in tumors using multiphoton laser scanning microscopy. Nat. Med. 7, 864–868. 7. Ciancio, S. J., Coburn, M., and Hornsby, P. J. (2000) Cutaneous window for in vivo observations of organs and angiogenesis. J. Surg. Res. 92, 228–232. 8. Naumov, G. N., Wilson, S. M., MacDonald, I. C., et al. (1999) Cellular expression of green fluorescent protein, coupled with high-resolution in vivo video-microscopy, to monitor steps in tumor metastasis. J. Cell Sci. 112, 1835– 1842. 9. Prasher, D. C., Eckenrode, V. K., Ward, W. W., et al. (1992) Primary structure of the Aequorea victoria green-fluorescent protein. Gene 111, 229–233. 10. Chalfie, M., Tu, Y., Euskirchen, G., et al. (1994) Green fluorescent protein as a marker for gene expression. Science 263, 802–805. 11. Cheng, L., Fu, J., Tsukamoto, A., and Hawley, R. G. (1996) Use of green fluorescent protein variants to monitor gene transfer and expression in mammalian cells. Nat. Biotechnol. 14, 606–609. 12. Cody, C. W., Prasher, D. C., Westler, W. M., et al. (1993) Chemical structure of the hexapeptide chromophore of the Aequorea green fluorescent protein. Biochemistry 32, 1212–1218. 13. Yang, F., Moss, L. G., and Phillips, G. N. Jr. (1996) The molecular structure of green fluorescent protein. Nat. Biotechnol. 14, 1246–1251. 14. Morin, J. and Hastings, J. (1971) Energy transfer in a bioluminescent system. J. Cell Physiol. 77, 313–318. 15. Cormack, B., Valdivia, R., and Falkow, S. (1996) FACS-optimized mutants of the green fluorescent protein (GFP). Gene 173, 33–38. 16. Crameri, A., Whitehorn, E. A., Tate, E., and Stemmer, W. P. (1996) Improved green fluorescent protein by molecular evolution using DNA shuffling. Nat. Biotechnol. 14, 315–319. 17. Delagrave, S., Hawtin, R. E., Silva, C. M., et al. (1995) Red-shifted excitation mutants of the green fluorescent protein. Biotechnology 13, 151–154. 18. Heim, R., Cubitt, A. B., and Tsien, R. Y. (1995) Improved green fluorescence. Nature 373, 663–664. 19. Zolotukhin, S., Potter, M., Hauswirth, W. W., et al. (1996) A ‘humanized’ green fluorescent protein cDNA adapted for high-level expression in mammalian cells. J. Virol. 70, 4646–4654. 20. Gross, L. A., Baird, G. S., Hoffman, R. C., et al. (2000) The structure of the chromophore within DsRed, a red fluorescent protein from coral. Proc. Natl. Acad. Sci. USA 97, 11,990–11,995. 21. Fradkov, A. F., Chen, Y., Ding, L., et al. (2000) Novel fluorescent protein from Discosoma coral and its mutants possesses a unique far-red fluorescence. FEBS Lett. 479, 127–130.

Subcellular Imaging of Cancer Cells

129

22. Hoffman, R. M. (2002) Green fluorescent protein imaging of tumour growth, metastasis, and angiogenesis in mouse models. Lancet Oncol. 3, 546–556. 23. Condeelis, J. and Segall, J.E. (2003) Intravital imaging of cell movement in tumours. Nat. Rev. Cancer 3, 921–930. 24. Chishima, T., Miyagi, Y., Wang, X., et al. (1997) Cancer invasion and micrometastasis visualized in live tissue by green fluorescent protein expression. Cancer Res. 57, 2042–2047. 25. Farina, K. L., Wyckoff, J. B., Rivera, J., et al. (1998) Cell motility of tumor cells visualized in living intact primary tumors using green fluorescent protein. Cancer Res. 58, 2528–2532. 26. Yamauchi, K., Yang, M., Jiang, P., et al. (2005) Real-time in vivo dual-color imaging of intracapillary cancer cell and nucleus deformation and migration. Cancer Res. 65, 4246–4252. 27. Huang, M. S., Wang, T. J., Liang, C. L., et al. (2002) Establishment of fluorescent lung carcinoma metastasis model and its real-time microscopic detection in SCID mice. Clin. Exp. Metastasis 19, 359–368. 28. Wyckoff, J. B., Jones, J. G., Condeelis, J. S., and Segall, J. E. (2000) A critical step in metastasis: in vivo analysis of intravasation at the primary tumor. Cancer Res. 60, 2504–2511. 29. Mook, O. R. F., Marle, J. V., and Vreeling-Sindelarova, H. (2003) Visualization of early events in tumor formation of eGFP-transfected rat colon cancer cells in liver. Hepatology 38, 295–304. 30. Al-Mehdi, A. B., Tozawa, K., Fisher, A. B., Shientag, L., Lee, A., and Muschel, R. J. (2000) Intravascular origin of metastasis from the proliferation of endothelium-attached tumor cells: a new model for metastasis. Nat. Med. 6, 100–102. 31. Wang, W., Wyckoff, J. B., Frohlich, V. C., et al. (2002) Single cell behavior in metastatic primary mammary tumors correlated with gene expression patterns revealed by molecular profiling. Cancer Res. 62, 6278–6288. 32. Goswami, S., Sahai, E., Wyckoff, J. B., et al. (2005) Macrophages promote the invasion of breast carcinoma cells via a colony-stimulating factor-1/epidermal growth factor paracrine loop. Cancer Res. 65, 5278–5283. 33. Chambers, A. F., Schmidt, E. E., MacDonald, I. C., Morris, V. L., and Groom, A. C. (1992) Early steps in hematogenous metastasis of B16F1 melanoma cells in chick embryos studied by high-resolution intravital videomicroscopy. J. Natl. Cancer Inst. 84, 797–803. 34. Flach, J., Bossie, M., Vogel, J., et al. (1994) A yeast RNA-binding protein shuttles between the nucleus and the cytoplasm. Mol. Cell. Biol. 14, 8399–8407. 35. Kanda, T., Sullivan, K. F., and Wahl, G. M. (1998) Histone-GFP fusion protein enables sensitive analysis of chromosome dynamics in living mammalian cells. Curr. Biol. 8, 377–385. 36. Manders, E. M., Visser, A. E., Koppen, A., et al. (2003) Four-dimensional imaging of chromatin dynamics during the assembly of the interphase nucleus. Chromosome Res. 11, 537–547. 36a. Lee. S. (1987) Experimental microsurgery. Igaka-Shoin, New York.

Noninvasive Bioluminescent Imaging of Living Subjects

131

10 Noninvasive Imaging of Molecular Events with Bioluminescent Reporter Genes in Living Subjects Pritha Ray and Sanjiv Sam Gambhir Summary Bioluminescence imaging has become a very popular tool for noninvasive monitoring of fundamental biological and molecular processes in small living subjects. Luciferases are light-emitting enzymes that can generate light (known as bioluminescence) after reacting with specific substrates. The emitted light is used as a detection system for luciferase activity, which acts as a “reporter” for the activity of any regulatory elements that control its expression. These enzymes are isolated from various organisms, conveniently modified for expression in mammalian cells, and are extensively used in molecular biology and cell culture experiments. Recent advances in optical technology have opened a new dimension for in vivo application of luciferase enzymes in biomedical research. The most commonly utilized luciferases for in vivo bioluminescence are isolated from two very different sources: firefly luciferase (or beetle luciferase) and renilla luciferase (isolated from sea pansy). Although both these luciferases can produce light following interaction with the substrates, structurally and biochemically they are very different. Here we describe the methods and applications of firefly and renilla luciferases in molecular imaging using small animals. Key Words: Firefly luciferase; Renilla luciferase; CCD; in vivo imaging.

1. Introduction Luciferases are light-emitting enzymes that emit photons at different wavelengths in the visible range of light (400–615 nm) after interaction with a specific substrate. Luciferases are particularly useful as reporters in living cells and organisms. These enzymes are naturally occurring catalases that are found in a diverse group of organisms (e.g., bacteria, corals, and firefly) and that have very little structural homology (1). The chemical nature of the substrates also varies significantly. Some luciferin requires the presence of cofactor(s), such as FMNH2+, Ca2+, or ATP, to undergo oxidation. Complexes that contain a luciferase and a luciferin and that generally require O2 are also called photoproteins. From: Methods in Molecular Biology, vol. 411: Reporter Genes: A Practical Guide Edited by: D. Anson © Humana Press Inc., Totowa, NJ

131

132

Ray and Gambhir

Firefly luciferase (fl) is one of the most widely used reporter genes in modern molecular biology and in molecular imaging. (Note that fl refers to the gene and FL to the enzyme.) The 62-kDa luciferase enzyme was originally isolated from the North American firefly (Photinus pyralis), which emits light in the yellow-green region through oxidation of D-Luciferin in a reaction that also requires ATP, magnesium, and oxygen. The enzyme has been well characterized, and crystal structures are available (2,3). The fast rate of firefly luciferase enzyme turnover (t1/2 approx 3 h) in the presence of the substrate D-Luciferin allows for real-time measurements because the enzyme does not accumulate intracellularly to the extent of other reporters. The relationship between the enzyme concentration and the peak height of emitted light in vitro is linear up to 7 to 8 orders of magnitude. Therefore, these properties potentially allow for sensitive, noninvasive imaging of firefly luciferase reporter gene expression in living subjects. In recent years, considerable work with noninvasive imaging of firefly luciferase has been carried out (4–8). Additionally, many thermostable fl mutants with more stabilized light output have been created and validated by in vivo bioluminescence imaging using small animals (9,10). Renilla luciferase (rl), purified from the sea pansy (Renilla reniformis), is another widely used bioluminescent reporter gene in molecular biology. (Note that rl refers to the gene and RL to the enzyme.) RL catalyzes oxidation of coelenterazine, leading to bioluminescence. Coelenterazine consists of an imidazolopyrazine structure [2-( p-hydroxybenzyl)-6-( p-hydroxyphenyl)-8-benzylimidazo (1, 2-a) pyrazin-3-(7H)-on] that releases blue light across a broad range, peaking at 480 nm, upon oxidation by RL in vitro (1). Among the several coelenterazine analogs that are substrates for RL (e.g., coelenterazine-cp, coelenterazine-n, bisdeoxycoelenterazine [DeepBlueC]), coelenterazine produces the highest light output while reacting with RL (11). This substrateenzyme reaction does not require ATP as a cofactor, as do most of the beetle luciferases. Originally isolated from a marine organism, rl has some inherent limitations when used in mammalian cells. Approximately 10% of its codons have low corresponding tRNA expression in mammalian cells, thus limiting expression efficiency. The presence of a large number of potential transcription factor binding sites also causes anomalous transcriptional behavior in mammalian cells. We have utilized a synthetic Renilla luciferase reporter gene (hrl) from Promega: the native rl gene has been redesigned and the codon usage optimized to improve mammalian expression of the reporter, although the resulting primary sequence of the protein remains unchanged (Promega technical manual no. 237, p. 3). The resulting reporter gene has higher transcriptional efficiency, which enhances the detection of the reporter enzyme in cell culture and living animals (6). The monomeric structure of Renilla luciferase confers

Noninvasive Bioluminescent Imaging of Living Subjects

133

additional advantages when it is used to construct fusion proteins with another reporter or with a native protein (12,13) (see Note 1). This is particularly true when the fusion partner is a cell surface protein, as such proteins have generally very little or no ATP in the extracellular environment (13,14). Fusion of a native protein to hrl or its mutants can build novel imaging probes for detection of endogenous cellular events and thus are more useful tools for optical imaging of cancer (11,13). Currently we are developing red-shifted and thermostable mutants of the hrl gene that will allow better in vivo imaging (11,15). Tannous et al. (2005) have evaluated a third luciferase as a new bioluminescence reporter gene for in vitro and in vivo imaging in mammalian cells and small animals that also uses coelenerazine as substrate but is secretory in nature (16). Gaussia luciferase (gl), isolated from the copepod marine organism (Gaussia princeps), is the smallest luciferase known (19.9 kDa) and emits light at a peak of 480 nm after reacting with coelenterazine. Like RL, it does not need ATP as a cofactor (17). Expression of the humanized version of gl in mammalian cells/animals can produce 100-fold higher light intensity compared with hrl or fl; however, its secretory nature is not useful for many applications, and this system needs more standardization and validation. The gl reporter is now commercially available from Clontech (Mountain View, CA). 2. Materials 2.1. Vector Construction and Chemicals 1. Both cytomegalovirus (CMV) promoter-driven fl and hrl plasmids are available from Promega (Madison, WI). The genes can be cloned under different promoters and in different vector backbones by appropriate cloning. 2. D-Luciferin potassium salt, the substrate for FL, can be purchased from Biosynth (Naperville, IL). Make a 15 mg/mL stock in PBS and filter through a 0.22-µm filter. For lower gene expression, or for imaging fewer cells, make a stock concentration of 45 mg/mL in phosphate-buffered saline (PBS) and filter (see Note 2). 3. Coelenterazine, the substrate for RL, is available from Prolume (Pinetop, AZ). Prepare a stock of 2 to 5 mg/mL in methanol, and store at −20°C. A working stock should be made by diluting at a 1:10 ratio in PBS, as direct addition/injection of methanol is toxic to cells and animals.

2.2. Cell Lines, Culture Condition, and Transfection Procedure 1. High-glucose Dulbecco’s minimum Eagle’s medium (DMEM) and minimum essential medium (MEM) (Gibco, Grand Island, NY) supplemented with 10% fetal bovine serum (FBS) (Invitrogen), 1% penicillin (100 µg/mL), and streptomycin (292 µg/mL; Invitrogen, Grand Island, NY) are used for culturing C6 (rat glioma) and 293T (human embryonic kidney) cells. We have also used other cell lines

134

Ray and Gambhir

such as A375M (human melanoma), H9C2 (human cardiac myoblasts), and N2a (rat neuroblasts) as needed for various applications. 2. Geneticin (G418 sulfate; Gibco) is dissolved directly in cell culture media at a concentration of 500 µg/mL and filtered through a 0.22-µm filter. 3. All transient transfections are carried out using the Superfect transfection reagent (Qiagen, Valencia, CA) following the protocol recommended by the manufacturer.

2.3. Luminometer Measurement 1. 2. 3. 4. 5.

5X Lysis buffer (Promega). LAR II (for FL assay; Promega) or coelenterazine (for hRL assay) (Prolume). TD 20/20 luminometer (Turner Designs, Sunnyvale, CA). Bio-Rad protein assay reagent. DU-50 Spectrophotometer (Beckman Instruments).

2.4. Imaging and Quantification of Bioluminescence Data 1. Imaging system (IVIS, Xenogen, Almeda, CA) consisting of a cooled CCD camera mounted on a light-tight specimen chamber (dark box), a camera controller, a camera cooling system, and a Windows computer system. 2. The gray scale photographic images and bioluminescence color images are superimposed using the LIVINGIMAGE V.2.11 software overlay (Xenogen). 3. Image analysis is done using the IGOR image analysis software (V.4.02A, Wave Metrics, Lake Oswego, OR) (Fig. 1).

2.4. Imaging RL and FL Bioluminescence in Mice 1. Mice: (4–6-wk-old nude/nude mice (Charles River). 2. Anesthesia: isoflurane or 40 µL of a 4:1 mixture of ketamine and xylazine. 3. D-Luciferin or coelenterazine substrates as Subheading 2.1.

3. Methods The use of luciferase enzymes in modern molecular biology to monitor gene expression, promoter activation, transcriptional activation, and many other cellular and biochemical processes has been a routine procedure for long time. Both the firefly and Renilla luciferases are commercially available from Promega. The completely different kinetics of light output and substrate specification of these enzymes allows the simultaneous measurement of both the FL and RL from the same set of samples. Since there is no crossreactivity between the substrates, the two different molecular processes can easily be monitored from the same set of cells or animals. Bioluminescence generates a very small numbers of photons and therefore is difficult to image by conventional optical microscopes at the cellular level. However, optimization of a thermoelectrically cooled CCD camera with cooling of the instrument at −120oC can detect weak luminescent sources within a light-tight “black-box” chamber from cells as well as from small animals in a

Noninvasive Bioluminescent Imaging of Living Subjects

135

Fig. 1. Bioluminescence in vivo imaging system. The Xenogen In Vivo Imaging System (IVIS) consists of a cooled CCD camera mounted on a light-tight imaging chamber, a cryogenic refrigeration unit, a camera controller, and a computer system for data analysis. See accompanying CD for color version.

semiquantitative manner. The method of imaging bioluminescence sources in living subjects with a CCD camera is relatively straightforward: 1. The anesthetized animal is injected with substrate (either intravenously or intraperitoneally) and placed in the light-tight chamber for a few seconds to minutes. 2. A standard light photographic image of the animal is obtained, prior to the bioluminescence image captured by the cooled CCD camera positioned above the subject within the confines of the dark chamber. 3. A computer subsequently superimposes the two images on one another, and the relative location of luciferase activity is inferred from the composite image. 4. An adjacent pseudocolor scale indicates the relative or absolute number of photons detected. This scale does not reflect the color (wavelength) of the emitted photons but only the number of such photons, measured in relative light units per minute per area (RLU/cm2/min).

The intensity of signal varies with the depth of tissues since light is absorbed and attenuated while traveling through tissues. Light in the near infrared (NIR) region (700–900 nm) has the lowest tissue absorption rate. Therefore, intensity of signal also depends on the emission wavelength. hRL can produce a ten times higher number of photons than FL, but owing to the longer emission range

136

Ray and Gambhir

of FL compared with RL (540 nm vs 480 nm), FL is preferred for in vivo imaging from deeper tissues. In addition, the half-life of each enzyme is an important factor in choosing between RL and FL systems. RL is preferred over FL for a timedependent study since FL has a long half-life (approx 3 h) and the luminescence signal generated lasts for 7 to 8 h, whereas the RL signal lasts for only few minutes (9,18). RL is also preferred over FL for imaging fusion proteins, particularly cell surface receptor-RL fusion proteins, as it does not require ATP to react with coelenterazine (13). 3.1. Vector Development 1. The firefly luciferase (pGL3) and Renilla luciferase reporter vectors (phRL) are obtained from Promega and then cloned into the pcDNA3.1 (Clontech) expression vector backbone using appropriate restriction sites. 2. pcDNA3.1 vector carries the neomycin-resistant gene, and thus it is easy to generate stable clones of mammalian cells expressing the specific reporter using antibiotic (G418 sulfate) selection.

3.2. Transient and Stable Transfection of CMV-fl and CMV-hrl Luciferases in Different Cells 1. The day before transfection, plate 0.5 to 4 × 105 of 293T cells/well in a 12-well dish and incubate them at 37°C. a. The next day add 1.5 µg reporter plasmid DNA (for each well) to 5 µL of Superfect in 75 µL of serum and antibiotic-free media for 10 min. b. Wash the cells in 1X PBS, and then add 400 µL of the complete media to the DNA/Superfect mixture; add to the cells. c. After 3 to 4 h of incubation at 37°C, remove the media, wash the cell monolayer with PBS once, and then add 1.5 mL of complete media. d. Incubate the cells for another 24 or 48 h at 37°C before assaying the reporter activity. e. Higher numbers of cells can also be used with larger volume tissue culture dishes. 2. To generate stably expressing clones of mammalian cells, follow the transfection procedure as described in Subheading 3.2.1. and then split the cells at a 1:5 ratio after 48 h. a. Culture the cells in G418-containing media (500–600 µg/mL) for 14 to 20 d. Cell death will be seen after 3 to 4 d (see Note 3). b. Remove media and wash the plates two to three times in 1X PBS and again maintain in G418-containing media. c. At 18 to 20 d, when most of the cells are dead, small clones will start to appear. When the clones are big enough to visualize, add 1 mg of D-Luciferin to 5 mL of 1X PBS, add 1 mL to each of the plates, and image in the IVIS imaging system.

Noninvasive Bioluminescent Imaging of Living Subjects

137

d. After 1-min acquisition, remove the PBS and add media to the cells, remembering to mark the clones. e. Pick the clones with high expression level, culturing them first in 96-well plates, and then gradually transfer them to bigger vessels as they expand. f. Finally, measure the relative light unit per microgram of protein (RLU/µg P) for a few selected clones, and use the best expresser for both in vitro and in vivo studies.

3.3. Measurement of Bioluminescence from Cell Extracts 1. Remove media from the cells transiently transfected with either CMV-fl or CMV-hrl plasmid, or cells stably expressing either CMV-fl or CMV-hrl plasmid, and wash with 1X PBS. Add 100 µL of 1X lysis buffer (5X stock buffer from Promega diluted with PBS) to each well and shake on a rotating platform for 10 min at room temperature. 2. Transfer the cell lysates to 1.5-mL centrifuge tubes, spin at 6,500g for 2 min, and then transfer the supernatant to a fresh tube. Store the tube on ice (see Note 4). 3. Add 100 µL of LAR II (substrate for FL enzyme) or 100 µL of coelenterazine solution (see Note 5) (0.5 µL coelenterazine from a stock of 1 µg/µL diluted to 100 µL in PBS; for RL assay) to 5 µL of the lysate. 4. Measure the reaction (RLU) in a TD 20/20 luminometer for 10 s. 5. Add 5 µL of the lysate to 1 mL of Bradford reagent, mix thoroughly, and keep at room temperature for 5 min. Measure the absorbance at A595 in a Beckman spectrophotometer, and calculate the protein concentration by using standards made using the same lot of reagent. 6. Calculate the luminescent results as RLU/µg P.

3.4. Measurement of Bioluminescence from Living Cells 1. Plate cells expressing CMV-fl or CMV-hrl in 12-well dishes, and incubate at 37°C. The next day remove the media and wash the cells with 1X PBS. 2. Add 500 µL of PBS containing either 50 µg of D-Luciferin in cells expressing CMVfl reporter gene, or 1 µg of coelenterazine (see Notes 2, 5, and 6) in cells expressing the CMV-hrl reporter gene, and place the cells in the CCD imaging system. a. Acquire images for 1 min. b. Draw regions of interest (ROIs) on each well, and calculate the signal intensity using the IGOR image analysis software. c. Bioluminescence is expressed as maximum photons/second/centimeter2/steradian (p/s/cm2/sr). Signals can also be measured as average photons/second/cm2/ steradian. 3. Add 100 µL of 1X lysis buffer, lyse the cells and measure the protein concentration of each well as described above (see Subheading 3.3.5.). 4. Final results are to be reported as bioluminescence normalized to protein content of each well.

138

Ray and Gambhir

3.5. Imaging of FL and RL Bioluminescence in Various Tissues from Living Mice 1. All animal handling needs to be performed in accordance with the guidelines of the respective university. 2. Anesthetize 4- to 6-wk-old nude/nude mice either by using intravenous injection of 40 µL of ketamine and xylazine (4:1) or by keeping the animals in a chamber with isoflurane (2% isoflurane with 100% oxygen at a flow rate of 1 L/min). 3. To image bioluminescence from subcutaneous regions, implant 1 × 106 cells expressing either CMV-fl or CMV-hrl plasmids on the shoulder of each mouse using a 26-gage needle. Before injection, suspend the cells in 100 µL of cold PBS, and keep the cells on ice. 4. After 1 or 2 h of cell implantation, inject 100 µL of 15 mg/mL of D-Luciferin intraperitoneally in each of the anesthetized mice for imaging FL activity. For imaging hrl gene expression, intravenously inject 20 to 40 µg of coelenterazine (see Note 7), dissolved in 100 to 150 µL PBS per mouse. 5. Place the mice (the IVIS system can image five mice simultaneously) in the black light tight chamber in supine position, and image the mice for 1 min. The acquisition time will vary depending on the level of expression. The RL shows a very high signal from 1 min of injection of the substrate that rapidly declines with time. The signal for FL peaks from 5 min of injection of the substrate and remains stable for 1 h. The grayscale photographic images and bioluminescence color images can be superimposed using the LIVINGIMAGE v 2.5 software overlay (Xenogen). Draw ROIs over the signals and calculate the intensity as described above in Subheading 3.4. 6. To image bioluminescence from lungs, inject 5 × 105 cells expressing either CMVfl or CMV-hrl plasmid/mouse via the tail vein. The cells travel through the bloodstream and get trapped inside the lungs. After 2 h, image the mice following the protocol described step 5. 7. Implant 1 × 106 cells expressing either CMV-fl or CMV-hrl plasmid into the peritoneal cavity of nude mice. After 2 h, image the mice as in step 5 (Fig. 2) (18).

3.6. Simultaneous Imaging of FL and RL Bioluminescence in Living Mice 1. Implant 1 × 106 cells expressing CMV-fl plasmid on one shoulder of a nude mouse. Next implant a similar number of cells expressing CMV-hrl plasmid into the other shoulder. 2. After 1 h, intravenously inject 20 to 40 µg of coelenterazine, dissolved in 100 to 150 µL PBS, per mouse and immediately image the mouse in the IVIS camera for 1 min. Only cells expressing the CMV-hrl plasmid will show a bioluminescence signal. 3. After 1 h of RL imaging, inject 100 µL of 15 mg/mL of D-Luciferin intraperitoneally, and image for FL activity after 5 min. Again, only cells expressing the CMVfl plasmid will show a signal (Fig. 3) (6).

Noninvasive Bioluminescent Imaging of Living Subjects

139

Fig. 2. RL bioluminescence from C6-rl cells present in various tissues in living mice. (A) The C6-rl cells (1.0 × 106) were injected via the tail vein, and coelenterazine was tail-vein-injected 90 min later. The bioluminescence seen represents the thorax region of the mouse where C6-rl cells are trapped in the lungs. (B) C6-rl cells (1.0 × 106) were implanted in the peritoneum of a different mouse, and coelenterazine was tailvein-injected immediately afterward. Bioluminescence is seen only from the i.p. region. R and L represent the right and left side of the mouse resting in the supine position. (Reproduced from ref. 18.) See accompanying CD for color version.

3.7. Repetitive Monitoring of Tumor Xenograft, Cardiac Cell Implantation, and Metastasis by Bioluminescence Imaging in Living Subjects 1. Implant 5 × 105 tumor cells that stably express the CMV-fl reporter gene and control cells (without any reporter gene) in two shoulders of a nude mouse. Repeatedly image the mouse over time with the CCD camera (Fig. 4). 2. To monitor tumor metastases by bioluminescence imaging, inject 0.5 to 1 × 105 of A375M cells expressing the reporter gene via the tail vein. Scan the mouse repeatedly over time. At d 0, significant bioluminescence signal will be present in both the lungs. The signal will decrease over time and then reappear from d 35 onward in lungs and other sites of metastasis (Fig. 4). 3. Inject 5 × 105 embryonic cardiac myoblasts carrying the bioluminescence reporter gene into the heart of a rat and image over time by injecting the appropriate substrate in a cooled CCD camera (Fig. 5) (19). 4. Inject purified fusion protein consisting of RL and a ligand for/antibody to a cell surface protein into nude mice bearing tumors expressing the cell surface protein

140

Ray and Gambhir

Fig. 3. Kinetics of light production from mice carrying s.c. C6-fl and C6-rl cells after simultaneous tail-vein injection of both D-Luciferin and coelenterazine. A mouse was injected s.c. at points A (with C6-fl), B (C6-rl), and C (C6 control cells) on the right forearm, left forearm, and right thigh regions, respectively. Simultaneous injection of both coelenterazine and D-Luciferin mixture via the tail vein shows bioluminescence from both the sites simultaneously, but with distinct kinetics. A series of images at 2min intervals is shown from the same mouse. Each image represents a scan time of 1 min. The signal from C6-rl cells (point B) peaks early and is nearly extinguished within 10 min. Bioluminescence from C6-fl cells (point A) shows a relatively strong signal beyond 10 min. The region of control cells does not show any significant bioluminescence. R and L represent the right and left side of the mouse resting in the supine position. (Reproduced from ref. 18.) See accompanying CD for color version.

Fig. 4. Imaging serial increase in rl gene expression over time in tumors stably expressing the tk20rl fusion protein. A total of 2 × 106 of N2a cells stably expressing the tk20rl fusion gene and control N2a cells were implanted on the left and right shoulders, respectively, of a single nude mouse and imaged daily using the optical CCD camera after injection of coelenterazine. A gradual increase in bioluminescence was observed in the tumor expressing tk20rl fusion over time but not in the control tumor. (Reproduced from ref. 12.) See accompanying CD for color version.

Noninvasive Bioluminescent Imaging of Living Subjects

141

Fig. 5. Molecular imaging of cardiac cell transplantation in living animals. Optical imaging shows a representative rat transplanted with embryonic cardiomyoblasts expressing the CMV-fl reporter gene that emits significant cardiac bioluminescence activity at d 1, 2, 4, 8, 12, and 16 (p < 0.05 vs control). Control rat has background signal only. (Reproduced from ref. 19.) See accompanying CD for color version. and image serially over time after injection of coelenterazine. The same mice can also be imaged in a microPET device using a radiolabeled probe designed specifically for the cell surface protein (Fig. 6).

4. Notes 1. D-Luciferin should be dissolved in sterile PBS and filtered before injecting into animals. The stock solution can be stored at −20°C. 2. All vectors with the reporter gene must be sequenced before starting the experiments. 3. Use puramycin or hygromycin selection to create stable clones of 293T cells. 293T cells are already resistant to G418 selection due to presence of the SV40-T antigen expressing plasmid. 4. The cell lysates must be stored on ice. 5. The stock solution of coelenterazine should be prepared in ethanol/methanol since it does not dissolve in aqueous solution. This stock solution must be aliquoted and stored at −80°C in a freezer in colored centrifuge tubes. (Coelenterazine is light

142

Ray and Gambhir

142

Noninvasive Bioluminescent Imaging of Living Subjects

143

sensitive.) Immediately before injection into animals, this stock solution needs to be diluted in PBS at a 1:10 ratio and mixed thoroughly by vortexing. Prolonged storage of this diluted stock results in precipitation. 6. While you are assaying FL or RL activity in the luminometer, mix the cell lysate and substrate thoroughly by vortexing before measurement. Care should be taken while pipeting coelenterazine, as it starts dripping owing to the presence of the volatile organic solvents. 7. Tail vein injection of substrate in living mouse is a delicate procedure. Injection of a lesser amount of substrate often results in a lower signal. Researchers must practice this procedure systematically.

Acknowledgments This work is supported in part by an NIH In vivo Cellular and Molecular Imaging Center grant (ICMIC grant P50CA114747). We would also like to acknowledge funding from the NCI Small Animal Imaging Resource Program (SAIRP grant R24CA92862). References 1. Hastings, T. W. (1998) Bioluminescence. Annu. Rev. Cell Dev. Biol. 14, 197–230. 2. Branchini, B. R., Southworth T. L., Wilkinson, S. R., Hattak, N. F. K., Rosenberg, J. C., and Zimmer, M. (2005) Mutagenesis evidence that the partial reactions of firefly bioluminescence are catalyzed by different conformations of the luciferase C-terminal domain. Biochemistry 44, 1385–1393. 3. Conti, E., Franks, N. P., and Brick, P. (1996) Crystal structure of firefly luciferase throws light on a superfamily of adenylate-forming enzymes. Structure 15, 287– 298. Fig. 6. (Opposite page) Bioluminescence and microPET imaging in living animals using Db (Diabody)-18-RLuc8 fusion protein. (A) Schematic diagram of an anti-CEA (carcinoembryonic antigen) diabdoy-luciferase protein. The dimer of RLuc8 protein (blue ribbons) is fused with the variable region of heavy and light chains (green ribbons) of anti-CEA diabody joined by a flexible spacer (white ribbons). (B) Optical imaging. Athymic mouse bearing CEA-positive LS174T xenograft (thick arrow) on the left shoulder and CEA-negative C6 xenograft (thin arrow) on the right shoulder. Images were obtained using a CCD camera 2, 4, 6, 8, and 24 h after tail vein injection of the Db-18-RLuc8 fusion protein. Color scale represents photons/s/cm2/steradian. Quantitation by ROI analysis of the mouse shown gave maximum signals of 6.95 × 104, 6.60 × 103, and 7.10 × 103 photons/s/cm2/steradian in LS174T, C6 and background tissues, respectively, at 6 h. (C) MicroPET imaging. MicroPET images obtained at 4 (left panel) and 21 h (middle panel) after injection of 124I-Db-18-RLuc8. Right panel shows the microPET image superimposed on the corresponding microCT image to provide anatomical localization. (Reproduced from ref. 13.) See accompanying CD for color version.

144

Ray and Gambhir

4. Massoud, T. and Gambhir, S. (2003) Molecular imaging in living subjects: seeing fundamental biological processes in a new light. Genes Dev. 17, 545–580. 5. Wu J. C., Iyer, M., and Gambhir, S. S. (2001) Noninvasive optical imaging of firefly luciferase reporter gene expression in skeletal muscles of living. Mol. Ther. 4, 297–306. 6. Bhaumik, S., Lewis, X. Z., and Gambhir, S. S. (2003) Optical imaging of Renilla luciferase, synthetic Renilla luciferase, and firefly luciferase reporter gene expression in living mice. J. Biomed. Optics 9, 578–586. 7. Contag, P. R., Olomu, I. N., Stevenson, D. K., and Contag, C. H. (1998) Bioluminescent indicators in living mammals. Nat. Med. 4, 245–247. 8. Contag, C. H., Spilman, S. D., Contag, P. R., et al. (1997) Visualizing gene expression in living mammals using a bioluminescent reporter. Photochem. Photobiol. 66, 523–531. 9. Braggett, B. Roy, R., Morgan, S., Tisi, L., Morse, D., and Gillies, R. J. (2004) Thermostability of firefly luciferase affects efficiency by in vivo bioluminescence. Mol. Imaging 3, 324–332. 10. Ray, P. and Gambhir, S. S. (2005) Construction and validation of improved triple fusion reporter gene vectors. Mol. Imaging Biol. 7, 153P. 11. Loening, A., Fenn, T. D., Wu, A. M., and Gambhir, S. S. (2006) Consensus guided mutagenesis of Renilla luciferase yields enhanced stability and light output. Protein Eng. Des. Sel. 19, 391–400. 12. Ray, P., Wu, A. M., and Gambhir, S. S. (2003) Optical bioluminescence and positron emission tomography imaging of a novel fusion reporter gene in tumor xenografts of living mice. Cancer Res. 63, 1160–1165. 13. Venisnik, K. M., Loening, A. M., Iyerm M., Gambhir, S. S., and Wu, A. M, (2006) Bifunctional antibody-Renilla luciferase fusion protein for in vivo optical detection of tumors. Protein Eng. Des. Sel. 19, 453–460. 14. Ray, P., De, A., Min, J. J., Tsien, R., and Gambhir, S. S. (2004) Imaging tri-fusion multimodality reporter gene expression in living subjects. Cancer Res. 64, 1323– 1320. 15. Loening, A. M. and Gambhir, S. S. (2006) Improved mutants of Renilla luciferase for imaging applications in living subjects. Mol. Imaging 5, 412. 16. Tannous, B. A., Fernandez, J. L., Weissleder, R., and Breakefield, X. O. (2005) Codon-optimized Gaussia luciferase cDNA for mammalian gene expression in culture and in vivo. Mol. Ther. 11, 435–443. 17. Verhaegent, M. (2002) Recombinant Gaussia luciferase. Overexpression, purification, and analytical application of a bioluminescent reporter for DNA hybridization. Anal. Chem. 74, 4378–4385. 18. Bhaumik, S. and Gambhir, S. S. (2002) Optical imaging of Renilla luciferase reporter gene expression in living mice. Proc. Natl. Acad. Sci. USA 99, 377–382. 19. Wu, J. C., Sundaresan, G., Min, J. J., et al. (2003) Molecular imaging of cardiac cell transplantation in living animals using optical bioluminescence and positron emission tomography. Circulation 108, 1302–1305.

Vapor Fixation of GFP

145

11 Green Fluorescent Protein as a Tracer in Chimeric Tissues The Power of Vapor Fixation Harald Jockusch and Daniel Eberhard Summary Green fluorescent protein (GFP) and its variants, small, highly soluble proteins, are routinely used as reporters for patterns of gene expression and the origin of cells in transplantation experiments. When not linked as fusion proteins to other polypeptides, they distribute rapidly in the cytoplasm of a given cell, thus allowing real-time observations on living material. For histological analysis, previous bath fixation of whole organs or tissues seemed obligatory, because, during drop fixation of sections, GFP rapidly leaks from cells whose membrane has been damaged by freezing and/or sectioning. The fluorescence of GFP and its derivatives is retained upon fixation, but most enzyme and antigenic activities of interest will be lost in the whole sample as a consequence of formaldehyde (FA) fixation. We have therefore developed an alternative method to fix GFP in frozen tissue sections by FA vapor. This method prevents leakage and redistribution of GFP and allows any cytochemical method to be applied to unfixed adjacent serial sections. Key Words: GFP; eGFP; green fluorescent protein; formaldehyde; vapor; fixation; histology; solubility; leakage; diffusion.

1. Introduction In experimental cell biology, embryology, and pathology, it is often necessary to distinguish cells of different origin, e.g., donor from host in transplantation experiments. Green fluorescent protein (GFP) and its variants like enhanced (e)GFP, eCFP, eYFP, and eRFP (green, cyan, yellow, and red fluorescent proteins, respectively) are commonly used for this purpose.

From: Methods in Molecular Biology, vol. 411: Reporter Genes: A Practical Guide Edited by: D. Anson © Humana Press Inc., Totowa, NJ

145

146

Jockusch and Eberhard

The cDNA coding for GFP was originally cloned from the pacific jellyfish Aequorea victoria (1). Natural GFP consists of 238 amino acids (M = 27 kDa) and can be directly visualized under UV in real time without adding external substrates. Modifications of the amino acid sequence by in vitro mutagenesis led to variants with improved stability and fluorescence efficiency, such as eGFP (2), and with varied emission spectra (YFP, RFP, and others) (3). In small translucent biological objects like nematodes (for which GFP labeling was originally developed [4]), fish (5) or, amphibian larvae (6), GFP introduced as a stable transgene or by transfection can be used for real-time observations. In the living environment, it is confined to the cells of interest, in which it diffuses to even the finest cytoplasmic extensions like growth cones of nerve fibers. However, larger, opaque objects can only be analyzed in detail by histological sectioning, in order to characterize GFP-positive cells in their tissue environment. Such analysis involves staining for protein markers like enzyme activities or antigens. GFP is best retained in cells by fixing the complete tissue in formaldehyde prior to sectioning (7–9). However, formaldehyde (FA) fixation will destroy most antigen epitopes and inactivate most enzymes. Ideally, then, it would be best first to section frozen unfixed tissue and then selectively fix only the sections used to analyze GFP fluorescence. In tissues with large cells, particularly cardiac and skeletal muscles, the extreme solubility of GFP prevents the use of even small volumes of aqueous fixative on sections during “drop fixation”: the leakage of GFP is much faster than its fixation within the surrounding cytoplasm (“When you are adhering the section to the slide glass, all the GFP will disperse somewhere. GFP is not fixed on the section by ethanol nor acetone,” Okabe, http://133.1.15.131/tg/observation.cfm; cf. ref. 10). In 1990 we characterized muscle fiber types in mutant muscles by using classical standard fiber type markers like the enzyme activities of succinate dehydrogenase (SDH) (11) or myosin ATPase (12,13) on the one hand and immunocytochemistry of myosin heavy chain isoforms (MyHCs) and the calcium binding protein parvalbumin (PV) on the other (14). Like GFP, PV is a highly soluble protein. In contrast to those of the MyHCs, the important antigenic epitopes of PV are resistant to FA fixation, i.e., the fixation requirements for MyHCs and PV are incompatible, so that adjacent sections of the same muscle have to be treated differently. Drop fixation of sections did not work for PV, because the antigen diffused out before it was fixed. In this situation, the classical Falck Hillarp hot FA vapor method to visualize sympathetic nerve fibers (15) gave a hint as to how to solve the problem. This histochemical method produces the fluorogen by covalent condensation of the aromatic transmitters adrenalin and noradrenalin with hot FA vapor and is thus different from a

Vapor Fixation of GFP

147

simple fixation, but on its basis we developed “cold FA vapor”fixation for PV (14), and later for GFP (10), as described in this chapter. Using FA vapor fixation, we obtained excellent retention of GFP in skeletal, heart, and smooth muscle tissue (e.g., stomach wall) and epithelial tissues such as pancreas and kidney (10,16) in sections of a transgenic mouse expressing eGFP downstream of a chicken β-actin promoter (7). Although this transgenic line (C57BL/6 Cr Slc TgN-(act-EGFP)OsbC15-001-FJ001, with the insertion of the transgene on chromosome 15) has been reported to express the eGFP transgene “ubiquitously,”eGFP concentrations varied by orders of magnitude from tissue to tissue. In our breeding stock, they were extremely high in all the muscle types (smooth, cardiac, and skeletal), as well as in the pancreas, high in kidney (10), but very low in neurons and germ cells. This pattern is not expected for the β-actin promoter and may be owing to a position effect of the locus of transgene insertion. In this chapter we have chosen skeletal muscle as an example for the application of FA vapor fixation because of its high GFP content and highly ordered tissue structure. FA vapor fixation was successfully used in cell labeling experiments to fix GFP in grafted or host skeletal muscle (Fig.1) (16) or to trace GFP-labeled dendritic cells in lymph nodes, spleen, and thymus (17), as well as in kidney (18). Examples of the application of FA vapor fixation of highly soluble proteins other than GFP are the immunohistochemical quantification of myoglobin in human skeletal and heart muscle cells (19) and of cytosolic cytochrome c in failing myocardium (20). In these cases, as with PV, the vapor-fixed proteins retained the antigenic determinants necessary for antibody staining. Thus, vapor fixation of sections (as opposed to bath fixation of organs or tissue blocs) serves two purposes: to avoid leakage of highly soluble proteins of interest from the section and to spare the rest of a tissue sample from fixation. In all likelihood, FA vapor fixation may also be applied to other GFP variants like YFP, CFP, and RFP as well as to other highly soluble proteins. 2. Materials 1. A hard-boiled hen’s egg. 2. 37% Formaldehyde solution (e.g., Sigma, cat. no. F-1268). The ready-to-use, concentrated FA solutions usually contain 15% methanol as an antioxidant. This prevents the solutions from freezing at −20°C. It is not necessary for vapor fixation to prepare FA freshly from para-FA. 3. Adhesive microscope slides (e.g., SuperFrostPlus, Menzel-Gläser, Braunschweig, Germany). Adhesion of (semi)dry sections is the most critical part of the method. 4. Calcium- and magnesium-free phophate-buffered saline (CMF-PBS). The absence of Ca2+ and Mg2+ ions is not critical; complete PBS will also do.

148

Jockusch and Eberhard

Vapor Fixation of GFP

149

Fig. 1. Examples for the application of the formaldehyde vapor fixation of GFP in skeletal muscle, in conjunction with histochemistry of adjacent unfixed frozen sections. Shown are specimens that contain GFP-positive and GFP-negative mouse tissues side by side. Wherever GFP fluorescence is shown (A, A', B, C— one component of overlay; D) a frozen section was FA vapor-fixed; in all other cases, histochemistry was done on unfixed sections. (A, A') Composite block GFP vs non-GFP; gastrocnemius muscle. (A) Overlay of GFP (green) and Hoechst (blue) fluorescence. (A') Same overlay as in A with additional overlay of rhodamine-labeled wheat germ agglutinin (WGA) to stain the extracellular matrix. Different fibers display different levels of eGFP. In the upper half of the border between the two muscles, a slight overspill of GFP can be seen. (B, B') Muscle graft: GFP-positive anterior tibial muscle (TA) grafted into TA bed of GFP-negative nude mouse host, after 70 d of regeneration (cf. ref. 16); the border (dashed line) between graft (g) and host (h) muscle tissue is shown. (B) The strongly heterogeneous GFP fluorescence in the donor muscle indicates participation of host muscle cells during regeneration of the graft. Nuclei are stained with Hoechst. (B') Adjacent section stained for succinate dehydrogenase (SDH) and viewed with Nomarski optics. Whereas the host muscle is predominantly glycolytic (light blue), the regenerative muscle is entirely oxidative (dark blue). (C, C', C'') Cell transplantation experiment. A suspension of C2C12 myogenic cells (GFP negative) was injected under the kidney capsule of a GFP-positive nude mouse where, 20 d p.o., they had differentiated into myosin heavy chain (MyHC)-positive myotubes. (C) Overlay of GFP image from FA vapor-fixed section, showing host kidney tissue with cotton-like distribution of GFP (not owing to diffusion of GFP!), and from adjacent unfixed section, myotubes stained with monoclonal antibody MF20 (specific for sarcomeric MyHC) (21) and Cy3-labeled second antibody (red fluorescence). The lack of green fluorescence in the region of the myotubes indicates that they are exclusively of donor origin. (C'') Combination of SDH and antilaminin (for basal lamina) staining on unfixed parallel section; kidney tissue is strongly SDH positive. Myotubes, in contrast to mature muscle fibers, stain only weakly for SDH. (C'') Conventional hematoxylin and eosin staining to show the distribution of nuclei and cytoplasm; this staining can be applied to unfixed as well as to FA-fixed specimens. (D, D') Cell transplantation experiment as in C. (D) GFP fluorescence of FA vapor-fixed section. (D') MyHC staining of adjacent unfixed section (MF20). (Inset in D') Overlay of region indicated by frames in D and D'. In this case, regenerating muscle of the abdominal wall (upper right), had attached to the kidney capsule and resulted in a mixing of host and graft cells, with cofusion (GFP-positive myotubes, yellowish in inset). Inducible myogenic cells (22) were tested in a similar experiment; host contribution (later demonstrated by GFP labeling of the host) had led to the illusion that these cells produced mature muscle fibers at a graft site. Scale bars in A– D' = 100 µm. See accompanying CD for electronic version.

150

Jockusch and Eberhard

5. Glass jar with closely fitting lid (e.g., as used for staining microscopic slides) or plastic dish. 6. Nonautofluorescent embedding medium (e.g., the aqueous medium Elvanol [10 g Mowiol 4-88], Hoechst, Frankfurt, Germany in 40 mL CMF-PBS and 20 mL glycerol).

3. Methods The following protocol refers to solid tissues or organs up to the size of a mouse brain hemisphere. 1. Mount with embedding medium on labeled cork/filter paper disk; if necessary, support tissue with a wedge of hard-boiled egg white (15) (see Note 1). 2. Shock-freeze fresh tissue in propane cooled in liquid nitrogen (−190°C). The frozen tissue may be kept at −70°C in a small sealed vial for many months. 3. Cut on cryostat (e.g., Frigocut 2800, Reichert-Jung) in sections 5 to 8 µm in thickness, and transfer directly onto adhesive microscope slides; let dry for about 5 min at room temperature (see Note 2). Collect sections for GFP histology (e.g., every fifth section in a rotational scheme) on slides separate from those to be used for conventional histochemistry (see Note 3). 4. For GFP histology, place slides with sections into a precooled (−20°C), tightly closing container (glass jar or plastic dish), the bottom of which is covered with three layers of filter paper soaked with 37% FA. The container should be sealed with parafilm and then kept for 6 h to overnight at −20°C. During handling, the sections should not get in contact with the FA solution. As the air in the jar will become saturated with FA vapor, after fixation the jar should be opened under the hood. 5. Rinse slide briefly and gently in PBS to remove the FA. 6. Recommended: counterstain nuclei for 10 min in Hoechst, DAPI, or propidium iodide in CMF-PBS (1–2 µg/mL). This allows for visualization of the whole tissue section including GFP-negative areas (see Note 4). 7. Embed sections in Elvanol and keep in the dark at room temperature. 8. Observe and photograph in the epifluorescence mode of a UV microscope, and check GFP fixation with UV microscope. Vapor-fixed sections retain their quality for several days. Upon prolonged storage, sections will appear blurred owing to slow leakage of GFP. 9. Immunohistochemistry for stable antigens, especially with polyclonal antibodies, of FA vapor-fixed sections can be performed according to standard protocols (14, 18). For fixation-sensitive antigens or enzymes, use adjacent serial sections on a separate slide (see Note 5). In serial cross-sections of skeletal muscle, individual fibers can easily be followed from section to section. 10. In the last few years, several reports on the successful application of FA vapor fixation have been published. Table 1 lists those for GFP fixation, as well as papers dealing with other highly soluble proteins, intrinsic to the tissue investigated. However, the FA vapor fixation method is open to improvement. Critical issues are the adhesion of the sections to the slide and the long-term storage of embedded sections. It will be useful to compare different brands of adhesive slides or work

151

Skeletal muscle Skeletal and heart muscle Heart muscle Lymph nodes, spleen, and thymus Kidney

eGFP

Myoglobin

Cytochrome c

eGFP

eGFP

b Jockusch

et al., 1991 (14) cited. et al., 2003 (10) cited.

a Fü chtbauer

Skeletal muscle

Tissue type

Parvalbumin (PV)

Protein

Migration of host myogenic cells into muscle grafts Quantification of myoglobin and correlation with SDH Determination of concentration at single cell level Tracing of eGFP-labeled dendritic cells after systemic injection in the mouse Analysis of expression of the core 2GnT gene

Distribution of PV among muscle fiber types

Subject of investigation

Table 1 Examples of Successful Applications of FA Vapor Fixation

Vectabond, hot FA vapor (70°C) Aceton postfixation and subsequent anti-GFP immunostaining Megalin antibody stain

Myosin ATPase, myosin immunohistochemistry, SDH activity Lac transgene (β-galactosidase), SDH activity SDH

Remarks/parallel stains

17b

16

19a.b

18

15

13

Reference

Vapor Fixation of GFP 151

152

Jockusch and Eberhard

out adhesive coatings for standard slides. A nonautofluorescent, nonaqueous embedding medium, if available, might improve the long-term localization of GFP in FA vapor-fixed sections (see Note 6).

4. Notes 1. Boiled rat or mouse liver in conjunction with embedding medium (e.g., TissueTek O.C.T Compound, Sakura, Heppenheim, Germany) is often used as a support for tissue blocs. However, if one uses a DNA stain for the nuclei, the support tissue will also be stained, which will confuse the picture. We therefore use an acellular material, boiled eggwhite, as a support (10,16). 2. The sections can also be mounted on precooled slides (cryostat cutting temperature) with the help of a brush and directly transferred to the precooled FA-containing container. This minimizes the GFP overspill occurring by drying the sections for 5 min at room temperature. However, adherence of the cryosection to precooled slides may be critical. 3. Dry sections (prior to embedding) can be viewed in epifluorescence at low magnification to check for GFP-positive tissue. 4. Most confocal microscopes are not equipped with filter combinations for the blue fluorescence of the Hoechst DNA stain; therefore, propidium iodide (with red fluorescence) can be a useful alternative to Hoechst stain. 5. According to Schimmelpfennig et al. (17), it is possible to postfix with acetone (7 min and dry for 2 min) after FA vapor fixation and proceed with immunostaining of GFP. Postfixation with 3.7% FA is also possible and might slow down diffusion of GFP into the embedding medium. 6. We have tried alternatives to the vapor fixation method, e.g., using cold liquid 37% formaldehyde (instead of the usual 3.7%) in drop fixation as well as immersing the slide with frozen sections into a formaldehyde solution containing glycerol as an antifreeze, at −20°C. In these cases, GFP leaked out of the section as with standard drop fixation using 3.7% FA at 4°C.

Acknowledgments We thank Professor Masaru Okabe (Osaka University) for providing “green mice,”Sylvana Voigt (Bielefeld University) for histology and documentation, and Prof. Gillian Butler-Browne (UniversitéParis 7) for suggesting to H. J. to publish the FA vapor fixation method in the first place. We also thank Dr. P. Heimann (Bielefeld University) for help. Work performed at Bielefeld University was supported by the Deutsche Forschungsgemeinschaft (SFB 549) and Fonds der Chemischen Industrie (FCI). References 1. Prasher, D. C., Eckenrode, V. K., Ward, W. W., Prendergast, F. G., and Cormier, M. J. (1992) Primary structure of the Aequorea victoria green-fluorescent protein. Gene 111, 229–233.

Vapor Fixation of GFP

153

2. Cormack, B. P., Valdivia, R. H., and Falkow, S. (1996) FACS-optimized mutants of the green fluorescent protein (GFP). Gene 173, 33–38. 3. Hadjantonakis, A. K. and Nagy, A. (2001) The color of mice: in the light of GFPvariant reporters. Histochem. Cell. Biol. 115, 49–58. 4. Chalfie, M., Tu, Y., Euskirchen, G., Ward, W. W., and Prasher, D. C. (1994) Green fluorescent protein as a marker for gene expression. Science 263, 802–805. 5. Amsterdam, A., Lin, S., Moss, L. G., and Hopkins, N. (1996) Requirements for green fluorescent protein detection in transgenic zebrafish embryos. Gene 173, 99–103. 6. Tannahill, D., Bray, S., and Harris, W. A. (1995) A Drosophila E(spl) gene is “neurogenic”in Xenopus: a green fluorescent protein study. Dev. Biol. 168, 694– 697. 7. Okabe, M., Ikawa, M., Kominami, K., Nakanishi, T., and Nishimune, Y. (1997) ‘Green mice’as a source of ubiquitous green cells. FEBS Lett. 407, 313–319. 8. Eberhard, D. and Jockusch, H. (2005) Patterns of myocardial histogenesis as revealed by mouse chimeras. Dev. Biol. 278, 336–346. 9. Eberhard, D. and Jockusch, H. (2004) Intermingling versus clonal coherence during skeletal muscle development: mosaicism in eGFP/nLacZ-labeled mouse chimeras. Dev. Dyn. 230, 69–78. 10. Jockusch, H., Voigt, S., and Eberhard, D. (2003) Localization of GFP in frozen sections from unfixed mouse tissues: immobilization of a highly soluble marker protein by formaldehyde vapor. J. Histochem. Cytochem. 51, 401–404. 11. Nachlas, M. M., Tsou, K. C., De Souza, E., Cheng, C. S., and Seligman, A. M. (1957) Cytochemical demonstration of succinic dehydrogenase by the use of a new p-nitrophenyl substituted ditetrazole. J. Histochem. Cytochem. 5, 420–436. 12. Green, H. J., Reichmann, H., and Pette, D. (1982) A comparison of two ATPase based schemes for histochemical muscle fibre typing in various mammals. Histochemistry 76, 21–31. 13. Brooke, M. H. and Kaiser, K. K. (1970) Muscle fiber types: how many and what kind? Arch. Neurol. 23, 369–379. 14. Füchtbauer, E. M., Rowlerson, A. M., Gotz, K., et al. (1991) Direct correlation of parvalbumin levels with myosin isoforms and succinate dehydrogenase activity on frozen sections of rodent muscle. J. Histochem. Cytochem. 39, 355–361. 15. Falck, B., Hillarp, N., Thieme, G., and Torp, A. (1962) Fluorescence of catechol amines and related compounds condensed with formaldehyde. J. Histochem. Cytochem. 10, 348. 16. Jockusch, H. and Voigt, S. (2003) Migration of adult myogenic precursor cells as revealed by GFP/nLacZ labelling of mouse transplantation chimeras. J. Cell Sci. 116, 1611–1616. 17. Schimmelpfennig, C. H., Schulz, S., Arber, C., et al. (2005) Ex vivo expanded dendritic cells home to T-cell zones of lymphoid organs and survive in vivo after allogeneic bone marrow transplantation. Am. J. Pathol. 167, 1321–1331. 18. Sekine, M., Taya, C., Shitara, H., et al. (2006) The cis-regulatory element Gsl5 is indispensable for proximal straight tubule cell-specific transcription of core 2

154

19.

20.

21.

22.

Jockusch and Eberhard

beta-1,6-N-acetylglucosaminyltransferase in the mouse kidney. J. Biol. Chem. 281, 1008–1015. van Beek-Harmsen, B. J., Bekedam, M. A., Feenstra, H. M., Visser, F. C., and van der Laarse, W. J. (2004) Determination of myoglobin concentration and oxidative capacity in cryostat sections of human and rat skeletal muscle fibres and rat cardiomyocytes. Histochem. Cell. Biol. 121, 335–342. van Beek-Harmsen, B. J. and van der Laarse, W. J. (2005) Immunohistochemical determination of cytosolic cytochrome C concentration in cardiomyocytes. J. Histochem. Cytochem. 53, 803–807. Bader, D., Masaki, T., and Fischman, D. A. (1982) Immunochemical analysis of myosin heavy chain during avian myogenesis in vivo and in vitro. J. Cell Biol. 95, 763–770. Bartsch, J. W., Jäckel, M., Perz, A., and Jockusch, H. (2000) Steroid RU 486 inducible myogenesis by 10T1/2 fibroblastic mouse cells. FEBS Lett. 467, 123–127.

Index

155

Index A Airway-directed gene transfer, see β-Galactosidase Alkaline phosphatase, see Placental alkaline phosphatase B Bioluminescence imaging, see Luciferase C Cancer cell trafficking bioluminescence imaging, see Luciferase fluorescent protein dual labeling studies blood vessel cell imaging, 126, 127 brain imaging in live mice, 125, 126 ear cell noninvasive imaging, 125, 126 gene transduction of DsRed and H2B-GFP, 124–126 materials, 123, 124, 126 principles, 122, 123 intravital microscopy, 122 overview, 121–123 Confocal microscopy Drosophila nervous system development studies of green fluorescent protein multichannel image collection, 93, 95 optimization, 91, 93, 95 overview, 82, 83 time lapse microscopy, 93, 94 signal transduction studies of green fluorescent protein reporter in single cells materials, 113 nuclear factor-κB translocation assay, 115–119 overview, 111–113 transcription reporter assay, 114, 115, 117, 119 D Dendra, green-to-red photoconversion, 70 Drosophila nervous system development, green fluorescent protein studies

confocal microscopy multichannel image collection, 93, 95 optimization, 91, 93, 95 overview, 82, 83 time lapse microscopy, 93, 94 embryo expression, 82, 85, 86 embryo preparation fixed , 89, 90, 94 live, 89, 94 larva preparation fixed, 90, 91 live, 90, 94 materials, 83–88, 94 DsRed metastasis studies, see Cancer cell trafficking monomeric red fluorescent protein properties, 62–64 photoactivatable fluorescent protein, 68, 69 structure and fluorescent chromophore formation, 62 variants and fluorescence properties, 52, 62, 63, 100 E EFIC, see Episcopic fluorescence image capturing Embryos, see Drosophila nervous system development; Episcopic fluorescence image capturing; β-Galactosidase; High-resolution episcopic microscopy EosFP, green-to-red photoconversion, 69, 70 Episcopic fluorescence image capturing (EFIC) embryo analysis data capture, 42–43 data processing and visualization, 43, 44 infiltration and embedding, 42, 44 materials, 39–41, 44 specimen preparation, 41, 42

155

156 overview, 36, 37 principles, 37, 38 resolution, 38 Episcopic microscopy, see Episcopic fluorescence image capturing; Highresolution episcopic microscopy F Firefly luciferase, see Luciferase Flow cytometry, transfection efficiency analysis with fluorescent proteins cell culture, 102, 108 instrument setup and analysis, 106–109 materials, 100–102, 108 sample preparation, 106 sensitivity, 112 transduction, 105, 108 transfection, 103, 105, 108 Fluorescence resonance energy transfer (FRET), fluorescent proteins, 57 Fluorescent proteins, see also DsRed; Green fluorescent protein advantages, 47 experimental considerations, 72, 73 flow cytometry, see Flow cytometry photoactivatable fluorescent proteins cyan-to-green photoconversion, 70 DsRed, 68, 69 green fluorescent protein, 65, 68 green-to-red photoconversion Dendra, 70 EosFP, 69, 70 Kaede, 69 KikG, 69 overview, 65–67 rationale for use, 72 reversible proteins, 70, 71 prospects, 74 types in marine organisms, 53–56, 64, 65 Formaldehyde vapor fixation, green fluorescent protein samples applications, 150, 151 cryosectioning, 150, 152 fixation, 150, 152 formaldehyde fixation limitations, 146 immunohistochemistry, 150, 152 materials, 147, 150 overview, 146, 147 FRET, see Fluorescence resonance energy transfer

Index G β-Galactosidase airway-directed gene transfer and detection clinical prospects, 25 counterstaining hematoxylin and eosin on cryosections, 30 hematoxylin and eosin on paraffin-embedded tissue, 31, 32 Nuclear Fast Red on paraffinembedded tissue, 31 materials, 26, 27, 32 overview, 25, 26 sectioning, 29, 33 tissue collection heads, 28 lungs, 28 X-Gal staining, 29, 30, 33 chromogenic substrates, 1 immunohistochemistry on tissue sections antibody incubation and washing, 18, 22 cryosection preparation, 16–18 fixation, 16, 22 imaging, 19 materials, 14, 15, 19, 20 mounting, 18, 19 overview, 13–16 micrometastasis detection, 121 transgenic mouse embryo detection cultured cell staining, 8 double staining with horseradish peroxidase, 7, 8, 10 genotyping, 8, 9 materials, 2–5, 9 overview, 1, 2 tissue sections, 6, 7, 9, 10 whole embryo staining, 5, 6, 9, 10 Genotyping, β-galactosidase staining, 8, 9 GFP, see Green fluorescent protein Green fluorescent protein (GFP) advantages, 47 Aequorea victoria protein fluorescent chromophore formation, 49, 50 folding, 48, 49 overview, 47, 48 structure, 48, 122, 146 variants and fluorescence properties, 50, 51, 57–60, 83, 99, 100, 145, 146

Index airway section imaging, 32 confocal microscopy signal transduction studies, see Confocal microscopy Drosophila nervous system development studies confocal microscopy multichannel image collection, 93, 95 optimization, 91, 93, 95 overview, 82, 83 time lapse microscopy, 93, 94 embryo expression, 82, 85, 86 embryo preparation fixed, 89, 90, 94 live, 89, 94 larva preparation fixed, 90, 91 live, 90, 94 materials, 83–88, 94 formaldehyde vapor fixation applications, 150, 151 cryosectioning, 150, 152 fixation, 150, 152 formaldehyde fixation limitations, 146 immunohistochemistry, 150, 152 materials, 147, 150 overview, 146, 147 immunohistochemistry on tissue sections antibody incubation and washing, 18, 22 cryosection preparation, 16–18 fixation, 16, 22 imaging, 19 materials, 14, 15, 19, 20 mounting, 18, 19 overview, 13–16 metastasis studies, see Cancer cell trafficking photoactivatable fluorescent protein, 65, 68 H Hematoxylin and eosin counterstaining cryosections, 30 paraffin-embedded tissue, 31, 32 High-resolution episcopic microscopy (HREC) embryo analysis data capture, 42–43 data processing and visualization, 43, 44

157 infiltration and embedding, 42, 44 materials, 39–41, 44 specimen preparation, 41, 42 overview, 36, 37 principles, 37, 38 resolution, 38 Horseradish peroxidase, double staining with β-galactosidase, 7, 8, 10 HREC, see High-resolution episcopic microscopy I Immunohistochemistry airway-directed gene transfer and detection clinical prospects, 25 counterstaining hematoxylin and eosin on cryosections, 30 hematoxylin and eosin on paraffin-embedded tissue, 31, 32 Nuclear Fast Red on paraffinembedded tissue, 31 materials, 26, 27, 32 overview, 25, 26 sectioning, 29, 33 tissue collection heads, 28 lungs, 28 X-Gal staining, 29, 30, 33 β-galactosidase on tissue sections antibody incubation and washing, 18, 22 cryosection preparation, 16–18 fixation, 16, 22 imaging, 19 materials, 14, 15, 19, 20 mounting, 18, 19 overview, 13–16 green fluorescent protein antibody incubation and washing, 18, 22 cryosection preparation, 16–18 fixation, 16, 22 formaldehyde vapor fixation of sections, see Formaldehyde vapor fixation imaging, 19 materials, 14, 15, 19, 20 mounting, 18, 19 overview, 13–16

158 K Kaede, green-to-red photoconversion, 69 KFP1, photoactivation, 71 KikG, green-to-red photoconversion, 69 L LacZ, see β-Galactosidase Luciferase bioluminescent imaging cardiac cell implantation studies, 141, 143 cell extract measurements, 137, 152 charge-coupled device camera, 134–136 dual imaging of firefly and Renilla enzymes, 138 emission wavelength, 135, 136 living cell measurements, 137, 152 materials, 133, 134, 141, 143 metastasis studies, 139, 141 mouse imaging in live subjects, 138 transfection, 136, 137, 141 tumor xenograft studies, 139, 141 vectors, 136 cofactors, 131 firefly enzyme features, 132 Gaussia enzyme features, 133 Renilla enzyme features, 132 reporter constructs, 132, 133 M Metastasis, see Cancer cell trafficking; Luciferase Monomeric red fluorescent protein, see DsRed N NF-κB, see Nuclear factor-κB NFR, see Nuclear Fast Red

Index Nuclear factor-κB (NF-κB), translocation assay with green fluorescent protein, 115–119 Nuclear Fast Red (NFR), counterstaining on paraffin-embedded tissue, 31 P Placental alkaline phosphatase, airway section staining, 30, 33, 30 R Renilla luciferase, see Luciferase S Signal transduction, confocal microscopy studies of green fluorescent protein reporter in single cells materials, 113 nuclear factor-κB translocation assay, 115–119 overview, 111–113 transcription reporter assay, 114, 115, 117, 119 T Transfection efficiency, see Flow cytometry Transgenic mouse, see β-Galactosidase Tumors, see Cancer cell trafficking; Luciferase V Vapor fixation, see Formaldehyde vapor fixation W Whole embryo, β-galactosidase staining, 5, 6, 9, 10 X X-Gal, see β-Galactosidase