Green Fluorescent Protein: Properties, Applications and Protocols [2 ed.] 0471736821

Citation preview

GREEN FLUORESCENT PROTEIN

METHODS OF BIOCHEMICAL ANALYSIS

Volume 47

GREEN FLUORESCENT PROTEIN Properties, Applications, and Protocols SECOND EDITION

Edited by

Martin Chalfie Steven R. Kain

A JOHN WILEY & SONS, INC., PUBLICATION

Copyright © 2006 by John Wiley & Sons, Inc. All rights reserved Published by John Wiley & Sons, Inc., Hoboken, New Jersey Published simultaneously in Canada No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, (978) 750-8400, fax (978) 750-4470, or on the web at www.copyright.com. Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, (201) 748-6011, fax (201) 748-6008, or online at http://www.wiley.com/go/permission. Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives or written sales materials. The advice and strategies contained herein may not be suitable for your situation. You should consult with a professional where appropriate. Neither the publisher nor author shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages. For general information on our other products and services or for technical support, please contact our Customer Care Department within the United States at (800) 762-2974, outside the United States at (317) 572-3993 or fax (317) 572-4002. Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic formats. For more information about Wiley products, visit our web site at www.wiley.com. Library of Congress Cataloging-in-Publication Data: Green fluorescent protein : properties, applications, and protocols / edited by Martin Chalfie and Steven R. Kain.—2nd ed. p. ; cm.—(Methods of biochemical analysis) Includes bibliographical references and index. ISBN-13 978-0-471-73682-0 (pbk.) ISBN-10 0-471-73682-1 (pbk.) 1. Green fluorescent protein—Laboratory manuals. [DNLM: 1. Green Fluorescent Proteins—analysis—Laboratory Manuals. 2. Green Fluorescent Proteins—biosynthesis—Laboratory Manuals. 3. Green Fluorescent Proteins—diagnostic use—Laboratory Manuals. 4. Luminescent Agents—analysis—Laboratory Manuals. QU 25 G795 2005] I. Chalfie, Martin. II. Kain, Steven. III. Series. QP552.G73G467 572¢.6—dc22

2005 2004029639

Printed in the United States of America 10 9 8 7 6 5 4 3 2 1

CONTENTS

First Edition Preface

vii

Preface

xi

Contributors

xiii

1.

DISCOVERY OF GREEN FLUORESCENT PROTEIN Osamu Shimomura

2.

PHOTONS FOR REPORTING MOLECULAR EVENTS: GREEN FLUORESCENT PROTEIN AND FOUR LUCIFERASE SYSTEMS J. Woodland Hastings and James G. Morin

15

BIOCHEMICAL AND PHYSICAL PROPERTIES OF GREEN FLUORESCENT PROTEIN William W. Ward

39

THE THREE-DIMENSIONAL STRUCTURE OF GREEN FLUORESCENT PROTEIN AND ITS IMPLICATIONS FOR FUNCTION AND DESIGN George N. Phillips, Jr.

67

MOLECULAR BIOLOGY AND MUTATION OF GREEN FLUORESCENT PROTEIN David A. Zacharias and Roger Y. Tsien

83

3. 4. 5. 6.

DISCOVERY AND PROPERTIES OF GFP-LIKE PROTEINS FROM NONBIOLUMINESCENT ANTHOZOA Konstantin A. Lukyanov, Dmitry M. Chudakov, Arkady F. Fradkov, Yulii A. Labas, Mikhail V. Matz, and Sergey Lukyanov

1

121

7.

EVOLUTION OF FUNCTION AND COLOR IN GFP-LIKE PROTEINS Mikhail V. Matz, Yulii A. Labas, and Juan Ugalde

139

8.

THE USES OF GREEN FLUORESCENT PROTEIN IN PROKARYOTES Raphael H. Valdivia, Brendan P. Cormack, and Stanley Falkow

163

9.

THE USES OF GREEN FLUORESCENT PROTEIN IN YEASTS Amy L. Hitchcock, Jason A. Kahana, and Pamela A. Silver

179 v

vi

CONTENTS

10.

USES OF GFP IN CAENORHABDITIS ELEGANS Oliver Hobert and Paula Loria

203

11.

GREEN FLUORESCENT PROTEIN APPLICATIONS IN DROSOPHILA Tulle Hazelrigg and Jennifer H. Mansfield

227

12.

THE USES OF GREEN FLUORESCENT PROTEIN IN PLANTS Jim Haseloff and Kriby R. Siemering

259

13.

USES OF GFP IN TRANSGENIC VERTEBRATES Sean Magason, Adam Amsterdam, Nancy Hopkins, and Shuo Lin

285

14.

THE USES OF GREEN FLUORESCENT PROTEIN IN MAMMALIAN CELLS Theresa H. Ward and Jennifer Lippincott-Schwartz

15.

PRACTICAL CONSIDERATIONS FOR USE OF REEF CORAL FLUORESCENT PROTEINS IN MAMMALIAN CELLS: APPLICATIONS IN FLUORESCENCE MICROSCOPY AND FLOW CYTOMETRY Yu Fang, Olivier Déry, Michael Haugwitz, Pierre Turpin, and Steven R. Kain

16.

PHARMACEUTICAL APPLICATIONS OF GFP AND RCFP Nicola Bevan and Stephen Rees

17.

REASSEMBLED GFP: DETECTING PROTEIN–PROTEIN INTERACTIONS AND PROTEIN EXPRESSION PATTERNS Thomas J. Magliery and Lynne Regan

305

339

361

391

Methods and Protocols Steven R. Kain

407

Index

423

FIRST EDITION PREFACE Now it is such a bizarrely improbable coincidence that anything so mind-bogglingly useful could have evolved purely by chance that some thinkers have chosen to see it as a final and clinching proof of the nonexistence of God. Douglas Adams, Hitchhikers Guide to the Galaxy

In 1955, Davenport and Nicol reported that the light-producing cells of the jellyfish Aequorea victoria fluoresced green when animals were irradiated with long-wave ultraviolet. Five years later, Shimomura et al. (1962) described a protein extract from this jellyfish that could produce this fluorescence. Independently, Morin and Hastings (1971) found the same protein a few years later. This protein, now called the Green Fluorescent Protein (GFP), was studied for many years in virtual obscurity. However, with the cloning and expression of A. victoria GFP (Prasher et al., 1992; Chalfie et al., 1994), interest in this protein has grown enormously. To steal a phrase from a recent movie, GFP has gone from “zero to hero.” As of January, 1998 at least 500 scientific publications have been published with the term “GFP” in their titles or abstracts. In the last 3 years hundreds of people have used GFP to mark proteins, cells, and organisms in a wide range of prokaryotic and eukaryotic species. They have used GFP to investigate fundamental questions in cell biology, developmental biology, neurobiology, and ecology. The interest in GFP goes beyond its utility as a biological marker. The protein is intrinsically intriguing, and investigators have sought to understand its structure, fluorescent properties, and biochemistry. This increased interest in GFP, serves as an important reminder of the usefulness of studying the biology of organisms that are not among the chosen “model” systems. The usefulness of GFP as a biological marker derives from the finding that the protein’s fluorescence requires no other cofactor: The fluorophore forms from the cyclization of the peptide backbone. This feature makes the molecule a virtually unobtrusive indicator of protein position in cells. Indeed, use of GFP as a tag suggests that the protein does not alter the normal function or localization of the fusion partner. Because permeabilization for substrate entry and fixation are not needed to localize GFP, proteins, organelles, and cells marked with this protein can be examined in living tissue. This ability to examine processes in living cells has permitted biologists to study the dynamics of cellular and developmental processes in intact tissues. In addition to the broad impact of GFP technology on basic research, several companies have also incorporated this important reporter into more applied efforts such as high throughput drug screening, evaluation of viral vectors for human gene therapy, biological pest control, and monitoring genetically altered microbes in the environment. Most notable on this list are applications for GFP in drug discovery, here the potential for real time kinetics, ease of use, and cost savings provided by this reporter are leading to the replacement of other markers such as firefly luciferase and b-galactosidase. As the development of GFP technology continues to expand, the instrument companies are introducing new and better instruments for detecting GFP fluorescent. Finally, two U.S. patents have issued (as of July, 1997) on GFP and its variants, with many more certain to appear in the next few years. vii

viii

FIRST EDITION PREFACE

As editors we find ourselves in the exciting, yet frustrating, position of producing a book that, in some aspects, will be out of date as it is published. The excitement comes from seeing the wealth of information being discovered about GFP and the many uses that people are finding for this molecule. The frustration results from the same source: New applications and information about GFP are published weekly, and no book on this subject can remain current. For example, as we write this preface, two papers have appeared on single molecule fluorescence of GFP (Dickson et al., 1997; Pierce et al., 1997), three on modifying GFP to measure calcium levels (Miyawaki et al., 1997; Persechini et al., 1997; Romoser et al., 1997), two on conditions that make GFP fluoresce red (Elowitz et al., 1997; Sawin and Nurse, 1997), and one on converting GFP to a voltage indicator (Siegel and Isacoff, 1997). We feel, however, that the contents of this volume serve as an important foundation for strategies that utilize GFP, and should guide the reader in using the marker in his or her system. We are in a period of rapid development of GFP as a tool for the biological sciences as people adapt the molecule for use in different organisms, generate variants with altered properties, and discover new ways that the protein can be used. Despite the intrinsic incompleteness of this enterprise, we have asked our colleagues to summarize the state of GFP research and they have done an admirable job. We are grateful that so many of the initial investigators that pioneered the study and use of GFP consented to write chapters for this volume. We have organized this book into four sections. We start with two introductory chapters by Osamu Shimomura on the discovery of GFP and by Woody Hastings and James Morin on bioluminescence and biofluorescence in nature. The second section describes the biochemistry and molecular biology of GFP. Bill Ward has written a very useful description of the biochemistry of GFP, pointing out both the gaps in our knowledge and the importance of physical chemical criteria for evaluating new variants of GFP. George Phillips then discusses the structure of GFP and implications of this structure for its function as a fluorescent molecule. In the last chapter in this section, Roger Tsien and Douglas Prasher describe many of these variants, their uses, and how they were derived. The third section documents various biological applications of GFP. The people we asked to contribute these chapters are the major developers of GFP in the various organisms described. As described above these chapter are incomplete in that new information and application are developing at a very rapid rate. Nonetheless, each of these chapters provides insights into how GFP is being applied to particular species. We urge readers not to look only at the organism they love best, since approaches used for one organism may prove important when applied to others. For example, the use of species-specific codon usage, presumably by allowing greater production of protein, has been very important for GFP expression in mammalian cells. Also Andy Fire has found that GFP (and bgalactosidase) expression is elevated in the nematode Caenorhabditis elegans when artificial introns are interspersed in the cDNA sequence. Both of these observations may be important for those considering optimizing GFP expression in their organisms. Finally, we asked the contributors in the third section to provide protocols on the purification of GFP and its application in various organisms and Bill Ward to contribute information on purifying GFP. Sharyn Endow and David Piston have admirably taken on the formidable task of collecting, editing, and adding to this material for the fourth section of this book. In particular, they have provided outstanding protocols for visualizing and recording GFP fluorescence. We are just beginning to learn about and use GFP, and, as always, many questions remain. Much still needs to be learned about the chemistry of fluorophore formation and

ix

FIRST EDITION PREFACE

the role of the protein structure in this formation. Additional variants are needed. In particular, variants with spectra that do not significantly overlap with those of existing variants would be very useful. Such variants could be used in multiple labeling experiments, but they may have an even greater potential. Specifically, the use of fluorescence resonance energy transfer between two fluorescent proteins would enable the generation of a system analogous to the yeast two-hybrid system (Fields and Song, 1989) to look at protein: protein interactions. The advantage of such a system is that it would not require transcription as a readout of the interaction, and could therefore be used anywhere in the cell (e.g., cytosol, plasma membrane, mitochondria). Morever, suitably marked molecules would allow the testing of protein interactions in situ in a variety of organisms. Finally, as we learn more about the properties of these protein, we need to take advantage of this information to optimize GFP fluorescence intensity, excitation and emission spectra, and protein and message stability for different uses. In the next few years, we will undoubtedly see many more uses for this protein. The future does look bright for GFP. The editors of a book have, perhaps, the easiest jobs; everyone contributes to an effort that they get the credit for. As this was the first volume that either of us had edited, we are particularly grateful for all the help that we have been given. Foremost we wish to thank the contributors who graciously consented to write chapters and then put up with our requests for rewrites and for “just a little more information” with great good humor. We are indebted to David Ades and Kaaren Janssen for starting us on this endeavor. We will get even. Finally, we are most obligated to Colette Bean, our editor at John Wiley, for showing us the ropes, keeping us on schedule, and getting us over the anxieties of producing this volume. Martin Chalfie Steven Kain

REFERENCES 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. Davenport, D., and Nicol, J. A. C. Luminescence in Hydromedusae. Proc. R. Soc. London Ser. B 144:399–411. Dickson, R. M., Cubitt, A. B., Tsien, R. Y., and Moener, W. E. (1997). On/off blinking and switching behavior of single molecules of green fluorescent protein. Nature (London) 388:355–358. 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. Fields, S., and Song, O. K. (1989). A novel genetic system to detect protein-protein interactions. Nature (London) 340:245–246. Miyawaki, A., Llopis, J., Heim, R., McCaffery, J. M., Adams, J. A., Ikura, M., and Tsien, R. Y. (1997). Fluorescent indicators for Ca2+ based on green fluorescent proteins and calmodulin. Nature 388:882–887. Morin, J. G., and Hastings, J. W. (1997). Biochemistry of the bioluminescence of colonial hydroids and other coelenterates. J. Cell. Physiol. 77:305–312. Persechini, A., Lynch, J. A., and Romoser, V. A. (1997). Novel fluorescent indicator proteins for monitoring free intracellular Ca2+. Cell Calcium 22:209–216. Pierce, D. W., Hom-Booher, N., and Vale, R. D. (1997). Imaging individual green fluorescent proteins. Nature (London) 388:338.

x

FIRST EDITION PREFACE

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. Romoser, V. A., Hinkle, P. M., and Persechini, A. (1997). Detection in living cells of Ca2+dependent changes in the fluorescence emission of an indicator composed of two green fluorescent protein variants linked by a calmodulin-binding sequence. A new class of fluorescent indicators J. Biol. Chem. 272:13270–13274. Sawin, K. E., and Nurse, P. (1997). Photoactivation of green fluorescent protein. Curr. Biol. 7:R606–R607. 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.

PREFACE In the preface to the first edition of this book seven years ago, we predicted that we were just beginning to see the usefulness of GFP. Although we expected many new uses for this molecule, we are amazed at the extent to which GFP, it derivatives, and similar fluorescent proteins have been used in biology today. A very approximate estimate of the general usefulness of these proteins can be seen in the number of journal articles that have citations to them. In 2004 roughly 50%, 35%, 60%, and 20% of the articles in Cell, Development, Journal of Cell Biology, and Neuron, respectively, mentioned or used these proteins (values were obtained by searching journal web site for articles with the words GFP, CFP, YFP or dsRed and then estimating the total number of articles published in the year). The fluorescent proteins are not only a general tool in basic biological research; they have also been used extensively by industry. As one unusual example, a group of entrepreneurs in the San Francisco Bay Area are pursuing a venture called Canary, Inc. which uses GFP as the basis for detecting landmines and other unexploded remnants of war. Another indicator of the growing use and importance of GFP is that this year saw the publication of the first book about GFP for the general public (Zimmer, 2005). The last seven years have also seen the introduction of many new fluorescent proteins and derivatives. Perhaps the most striking change in the field has been the discovery of the coral fluorescent proteins (Matz et al., 1999). These proteins not only provide a wealth of new colors, but also demonstrate that these types of proteins exist in a wide range of organisms. People continue to modify the fluorescent proteins and discover interesting new properties and uses. As the first edition was coming to press, we noted the GFP-based calcium sensors had just been developed. Now many more derivatives have been produced. One intriguing discovery was made by Ghosh et al., 2000. They split GFP into two separate polypeptioles. Coexpression of these proteins alone did not yield any fluorescence. Remarkably however, fluorescence could be reconstituted when covalently linked interacting protein domains brought the two parts of GFP together. This discovery has already led to an alternative to fluorescence resonance energy transfer (FRET) as a way of examining protein-protein interactions and a combinatorial method of labeling cells. These and other advances rendered the first edition of this book considerably out-ofdate. In the hope of bringing these more recent discoveries to the attention of a general audience, we were persuaded to edit a second edition of this book. As we noted in the preface to the first edition, trying to evaluate the field of fluorescent proteins is very difficult, because it is a moving target. The field is changing all the time. This dynamic feature of the field reflects its strength, but also means that reviews will always be incomplete. For that reason, we are delighted that so many of our previous authors were gracious enough to consider updating and rewriting their contributions. In addition, we are delighted to include as new authors, researchers who have done so much to move the field in new directions. As for the first editions, our editors at Wiley have been particularly supportive and diligent. These people include Luna Han, who first convinced us that this enterprise was xi

xii

PREFACE

worthwhile, and Darla Henderson and Danielle Lacourciere, who saw this second edition to completion. We are grateful for their help. Finally, we once again look forward to being astonished by even newer uses for this remarkable collection of proteins. Judging by the past, the future continues to look bright. Martin Chalfie Steven Kain

REFERENCES Matz, M. V., Fradkov, A. F., Labas, Y. A., Savitsky, A. P., Zaraisky, A. G., Markelov, M. L., and Lukyanov, S. A. (1999) Fluorescent proteins from nonbioluminescent Anthozoa species. Nat. Biotechnol. 17:969–973. Zimmer, M. (2005) Glowing Genes: A Revolution in Biotechnology, Prometheus Press, 250 pp.

CONTRIBUTORS Adam Amsterdam Center for Cancer Research, Massachusetts Institute of Technology, Cambridge, MA 02139 Nicola Bevan Screening Development and Compound Profiling, GlaxoSmithKline, Stevenage, Herts, SG1 2NY, United Kingdom Dmitry M. Chudakov Shemyakin and Ovchinnikov Institute of Bioorganic Chemistry RAS, Moscow 117997, Russia Brendan P. Cormack Department of Microbiology and Immunology, Stanford University School of Medicine, Stanford, CA 94305 Olivier Déry BD Biosciences Clontech, Palo Alto, CA 94303 Stanley Falkow Rocky Mountain Laboratories, National Institute of Allergy and Infections Diseases, Hamilton, MT; and Department of Microbiology and Immunology, Stanford University School of Medicine, Stanford, CA 94305 Yu Fang BD Biosciences Clontech, Palo Alto, CA 94303 Arkady F. Fradkov Shemyakin and Ovchinnikov Institute of Bioorganic Chemistry RAS, Moscow 117997, Russia Jim Haseloff Division of Cell Biology, MRC Laboratory of Molecular Biology, CB2 2QH Cambridge, United Kingdom. Present address: Department of Plant Sciences, University of Cambridge, Cambridge CB2 3EA. United Kingdom J. Woodland Hastings Department of Molecular and Cellular Biology, Harvard University, Cambridge, MA 02138 Michael Haugwitz BD Biosciences Clontech, Palo Alto, CA 94303 Tulle Hazelrigg Department of Biological Sciences, Columbia University, New York, NY 10027 Amy L. Hitchcock Department of Molecular and Celluar Biology, Harvard University, Cambridge, MA Oliver Hobert Department of Biochemistry and Molecular Biophysics, Columbia University, College of Physicians and Surgeons, New York, NY 10032 Nancy Hopkins Center for Cancer Research, Massachusetts Institute of Technology, Cambridge, MA 02139 Jason A. Kahana Department of Alzheimer’s, Research, Merck Research Laboratories, West Point, PA xiii

xiv

CONTRIBUTORS

Steven R. Kain Agilent Technologies, Inc. 3500 Deer Creek Road Palo Alto, CA 94304 Yulii A. Labas Institute of Biochemistry RAS, 117071 Moscow, Russia Shuo Lin Department of Molecular, Cell, and Developmental Biology, UCLA, Los Angeles, CA 90095 Jennifer Lippincott-Schwartz Department of Cell Biology and Metabolism, NICHD, NIH, Bethesda, MD 20892 Paula Loria Department of Biochemistry and Molecular Biophysics, Columbia University, College of Physics and Surgeons, New York, NY 10032 Konstantin A. Lukyanov Shemyakin and Ovchinnikov Institute of Bioorganic Chemistry RAS, Moscow 117997, Russia Sergey Lukyanov Shemyakin and Ovchinnikov Institute of Bioorganic Chemistry RAS, Moscow, 117997, Russia Thomas J. Magliery Department of Molecular Biophysics & Biochemistry, Yale University, New Haven, CT 06520. Present address: Department of Chemistry and Department of Biochemistry, The Ohio State University, Columbus, OH Jennifer H. Mansfield Department of Genetics, Harvard Medical School, Boston, MA 02115. Present address: Department of Biological Sciences, Columbia University, New York, NY 10027 Mikhail V. Matz Whitney Laboratory, University of Florida, St. Augustine, FL 32080 Sean Megason Beckman Institute of Biological Imaging, California Institute of Technology, Pasadena, CA 91125 James G. Morin Section of Ecological Systematics, Cornell University, Ithaca, NY 14850 George N. Phillips Jr. Department of Biochemistry, University of Wisconsin—Madison, Madison, Wi 53706 Stephen Rees Screening and Compound Profiling GlaxoSmithKline, Stevenage, Herts, SG1 2NY, United Kingdom Lynne Regan Department of Molecular Biophysics & Biochemistry and Department of Chemistry, Yale University, New Haven, CT 06520 Osamu Shimomura The Photoprotein Laboratory, Falmouth, MA 02540 Kirby Siemering Division of Cell Biology, MRC Laboratory of Molecular Biology, Cambridge, United Kingdom, CB2 2QH Pamela Silver Department of Systems Biology, Harvard Medical School, Boston, MA; Department of Cancer Biology, Dana-Farber Cancer Institute, Boston, MA 02115 Roger Y. Tsien Department of Pharmacology, University of California, San Diego, La Jolla, CA 92093 Pierre Turpin BD Biosciences Clontech, Palo Alto, CA 94303

CONTRIBUTORS

Juan Ugalde Laboratory of Bioinformatics and Gene Expression, University of Chile, Santiago, Chile Raphael Valdivia Department of Microbiology and Immunology, Stanford University School of Medicine, Stanford, CA. 94305 Present address: Department of Molecular Genetics and Microbiology, Duke University, Durham , NC 27710 Theresa H. Ward Immunology Unit, Department of Infectious and Tropical Diseases, London School of Hygiene and Tropical Medicine, London WC1E 7HT, United Kingdom William W. Ward Department of Biochemistry and Microbiology, Rutgers University, Cook College, New Brunswick, NJ 08901 David A. Zacharias The Whitney Laboratory for Marine Bioscience, University of Florida, Department of Neuroscience, St. Augustine, FL

xv

Figure 1.1. Mid-summer specimens of Aequorea aequorea photographed in natural environment (top) and in seawater supplemented with KCl in darkroom (bottom), both at the University of Washington’s Friday Harbor Laboratories.

Green Fluorescent Protein: Properties, Applications, and Protocols, Second Edition Edited by Martin Chalfie and Steven R. Kain Copyright © 2006 John Wiley & Sons, Inc.

Figure 2.7. Streaks of luminescent bacteria photographed by their own light, showing two strains of Photobacterium fischeri, one of which emits yellow light by virtue of having YFP (yellow fluorescent protein). The other lacks YFP, emitting only blue light.

Figure 2.10. Transgenic tobacco plant carrying the firefly luciferase gene photographed by its own light. The continuous luminescence occurs following the uptake of luciferin by the roots. [From Ow et al. (1986).]

Figure 2.11. Bacterial colonies carrying four different beetle luciferase genes cloned from the ventral organ, distinguished by their different luminescence colors: green, yellow-green, yellow and orange (Wood et al., 1989).

3

2 5 1 6 4

9

11

8 7

10

10 A Figure 4.1. End-on (top left) and side (top right) views of the cylindrical b-can structure of GFP. Eleven strands of b-sheet form an antiparallel barrel with short helices forming lids on each end. The fluorophore is inside the can, as a part of a distorted a-helix, which runs along the axis of the cylinder. The GFP usually forms dimers in the crystal, aligned largely along the sides of the cylinders. Drawing by Ribbons (Carson, 1987), coordinates (Protein Data Bank entry 1GFL). [From Yang et al. (1996.) Reprinted with permission from Nature Biotechnology.]

Phe165

Phe165 Val150

Gln183 Arg96 Gln69 Thr62

Ile167 His148 Thr203 Tyr145

Gln94 Gly67

Tyr66

Ser205

Gln183 Arg96 Gln69 Thr62

Ile167 His148 Thr203

Val61 Glu222 Ser65

Val150

Tyr145 Val68

Gln94 Gly67

Tyr66 Val61 Glu222 Ser65

Val68

Ser205

Figure 4.3. Stereoview of the fluorophore and its environment. His148, Gln94, Arg96, and Glu222 can be seen on opposite ends of the fluorophore and probably stabilize anionic resonant forms. Water molecules, charged, polar, and nonpolar side chains all contact the fluorophore in various ways.

Figure 5.3. Visual appearance of E. coli expressing four differently colored mutants of GFP. Clockwise from upper right: Blue mutant P4-3 (= Y66H, Y145F) (Heim and Tsien, 1996); cyan mutant W7 (Y66W, N146I, M153T, V163A, N212K) (Heim and Tsien, 1996); green mutant S65T (Heim et al., 1995); yellow mutant 10C (= S65G, V68L, S72A, T203Y) (Ormo et al., 1996). In each of these lists of mutants, the mutation most responsible for the special alterations is underlined, while the other substitutions improve folding or brightness. The bacteria were streaked onto nitrocellulose, illuminated with a Spectraline B-100 mercury lamp (Spectronics Corp., Westbury, NY) emitting mainly at 365 nm, and photographed with Ektachrome 400 slide film through a low-fluorescence 400 nm and a 455-nm colored glass long-pass filter in series. The relative brightness of the bacteria in this image is not a good guide to the true brightness of the GFP mutants. Expression levels are not normalized, and the 365 nm excites the blue and cyan mutants much more efficiently than the green and yellow mutants, but the blue emission is significantly filtered by the 455-nm filter required to block violet haze.

Figure 5.4. The crystal structure of dimeric GFP (1GFL) (Yang et al., 1996a). The residues A206 (red), L221 (orange), and F223 (lavender) are shown as ball-and-stick representations. Replacing any of these residues with the positively charged residues lysine or arginine effectively monomerizes the protein.

Figure 5.5. GFP biosensors. (A) GFP can be engineered to be directly sensitive to a small molecule of interest. (B) Insertion of a conformationally dynamic domain into GFP can result in a chimera in which the fluorescence properties of GFP are modulated by a change in conformation of the domain.

Figure 5.5. (continued) (C) Similarly, proteins or peptides with dyanmic, associative properties can be fused to the N and C termini of circularly permuted GFPs, thereby reporting on the changes in the association in response to a stimulus.

Figure 5.6. A backbone representation of the three-dimensional structure of GFP (1EMG) (Elsliger et al., 1998). The residues where circular permutations are permitted while retaining fluorescence are color highlighted. E142, hot pink; Y143, gray; Y145, dark blue; H148, fuchsia; D155, yellow; H169, red; E172, light blue; D173, orange; A227, Cyan; I229, light purple. These residues represent sites where the main chain can be interrupted. In most cases, resumption of GFP sequence can occur one to four residues following the initial interruption.

Figure 7.2. (This also appears in color insert.) Copepoda species that yielded fluorescent GFPlike proteins (from Shagin et al., 2004). Images were taken by fluorescent microscope using combined illumination with white light and standard FITC filter set. (a) Pontellina plumata. Inset magnifies the head. (b) Labidocera aestiva. (c) Compare Pontella meadi.

Figure 8.1. Fluorescence images of sporulating B. subtilis cells expressing transcriptional and translational GFP fusions. Two sporangia are shown per panel. (A) Forespore-specific expression of a sF-dependent SspE2G-GFP fusion. (B) Mother cell-specific expression of a sE-dependent cotEgfp fusion. (C) Localization of a SpoIVFB-GFP translational fusion (note localized fluorescence seen as a shell at one end of each sporangium). Courtesy of O. Resnekov and C. Webb, Harvard University.

Figure 8.4. Laser scanning confocal images of R. meloliti infection threads in plant root hairs. The R. meloliti bearing a plasmid with a trp-gfp fusion was used to infect alfalfa plants. Infection threads can be seen within individual root hair as they extend toward the main root body (stained red with propidium iodide).

Figure 8.5. Visualization of S. typhimurium intracellular-specific gene expression by fluorescence microscopy. S. typhimurium bearing a pagA::gfp fusion shows gene induction inside an infected mammalian cell but not in the extracellular medium. The corresponding DIC images show the relative topology of bacteria with respect to the infected cell.

䉴 Figure 10.1. Examples of subcellular structures visualized with GFP. (A) Presynaptic specializations: Transgenic animals expressing a synaptobrevin-GFP fusion construct reveal localization of GFP to synaptic sites in all neurons. Here, punctate fluorescence can be seen in the SAB motor neurons. (Reprinted from Nonet, M., Visualization of synaptic specializations in live C. elegans with synaptic vesicle protein-GFP fusions, J. Neurosc. Methods, 89:33–40. Copyright © 1999, with permission from Elsevier.) (B) Splicing speckles: Live transgenic animals expressing rescuing unc75::GFP show GFP localization in subnuclear puncta predicted to be splicing speckles (Loria et al., 2003). Here, multiple puncta can be seen in the nucleus of a ventral cord motorneuron (arrows). The corresponding Nomarski-DIC image is on the right. (C) Transcription factor target sites: elt1::GFP binds to its own promoter and leads to discrete fluorescent foci in nuclei. The embryonic gut nuclei of eight-cell embryos homozygous for a transgenic array containing fully rescuing elt2::GFP are shown. Many of the nuclei show two striking and intense foci of fluorescence (arrows). The image represents a stack of serially collected 400-nm optical sections projected without further manipulation. [Courtesy of Fukushige et al. (1999).] (D) Dense bodies: Transgenic animals expressing a rescuing unc-97::GFP fusion construct show localization to discrete spots and lines that correspond to dense bodies (DB) and M lines (M) of the body wall muscle. The expression of the unc-97::GFP reporter gene can be monitored in live or fixed animals. Some subcellular structures appear more crisp in formaldehyde-fixed animals (shown here), although they are also distinctly visible in live animals. (Reproduced from The Journal of Cell Biology, 1999, 144:53 by copyright permission of the Rockefeller University Press.) (E) Nuclear spindles: Multiphoton image series of GFP::b-tubulin in a live wild-type embryo from metaphase through telophase reveals centrosome dynamics. Elapsed time from the first frame is shown in minutes:seconds. (Reprinted from Molecular Biology of the Cell, 2001, 12:1751–1764) with permission by the American Society for Cell Biology.)

A

C

E

B

D

A

B

C

Figure 10.2. Examples of axon anatomy visualized with GFP. (A,B) Growth cone: Confocal micrograph of live L2 larvae at 17 hours post-hatching showing GFP driven under the unc-47 promoter that expresses in the ventral nerve cord, the DD and VD cell bodies (open arrows), the DD commissures (arrowheads), and the VD growth cones (solid arrows). (B) High magnification confocal micrograph of VD growth cones shown above (arrows). Filopodia extend from the round central mass of the growth cone on left. The middle growth cone is anvil-shaped. The right growth cone is extending a single finger toward the dorsal nerve cord. Existing embryonic DD commissures are marked with arrowheads. [Reprinted from Knobel et al. (1999), Development 126:4489–4498 with permission from The Company of Biologists Ltd.] (C) Axon co-labeling: Double labeling of axons in the ventral nerve cord (schematic). 3D image stacks of the ventral cord of double-labeled animals were recorded with a confocal microscope and subjected to a deconvolution algorithm to improve spatial resolution. Image shows an interneuron labeled with CFP (glr-1::GFP) and motorneuron axons labeled with GFP (unc-4::GFP). The image on the right is a cross-section through the ventral nerve cord at the position marked by the arrowhead in the left image. The orientation of images is depicted in the schematic. (Reprinted by permission of Wiley-Liss, Inc., a subsidiary of John Wiley and Sons, Inc. from Hutter, H., New ways to look at axons in Caenorhabditis elegans, Microscopy Research and Technique, copyright © 2000.)

D

E * * * *

*

Figure 10.2. (continued) (D) PVT axon morphology: The pioneer neuron PVT has a large cell body (large arrow), situated in the pre-anal ganglion that sends out an anteriorly-directed process in the ventral nerve cord (small arrows). Original EM reconstructions suggested that the axon of PVT terminated in the posterior body. A zig-2::GFP reporter shows strong expression in PVT. This analysis shows that the PVT axon in fact extends the entire length of the nerve cord and terminates within the nerve ring (arrowhead). Asterisk denotes gut autofluorescence. (E) PVD axon morphology: Previously, the processes of the PVD interneuron were not completely reconstructed by EM. Analysis of a GFP reporter for PVD shows that the axon displays an elaborate branching pattern not previously appreciated (arrow indicates position of cell body, and asterisk denotes gut autofluorescence).

A

B wild type

mec-3

Figure 10.4. Examples of the use of GFP as a tool to identify cells. (A) Identifying cells for electrophysiological recording: The gcy-5::GFP reporter was used to identify the chemosensory neuron ASER for in situ patch-clamp recording. (Top) Nomarski-DIC micrograph showing exposed neuronal cell bodies and a recording pipette sealed to ASER. Scale bar is 10 mm, anterior is left. (Bottom) Fluorescent micrograph of the same field above showing the GFP label in ASER, which allowed the unambiguous identification of the neuron. [Courtesy of Goodman et al. (1998).] (B) Analyzing of mutant cells in vitro: GFP-positive touch neurons from wild type (left) and mec-3 mutant animals (right) were enriched by fluorescence-activated cell sorting, cultured in vitro, and used to isolate RNA for DNA microarray analysis. This technique allowed the identification of mec-3-dependent genes and demonstrated that, unlike whole-worm RNA analysis, genes expressed in only a few cells can be identified systematically. [Reprinted from Zhang et al. (2002) by copyright permission of the Nature Publishing Group.]

Figure 11.4. GFP-tagged Gag proteins of two non-LTR retrotransposons, HeT-A and TART, shown in interphase Drosophila tissue culture cells. When transfected singly, HeT-A GFP-Gag is targeted to telomeres, but TART GFP-Gag is not. A. HeT-A GFP-Gag. B. TART GFP-Gag. Left panels: GFPGag; middle panels—anti-HOAP, which labels the telomeres; right panels—superimposed images, with DAPI-stained chromosomes. When co-transfected, Het-A Gag recruits TART-Gag to telomeres (not shown here). (From Rashkova et al., 2002. Courtesy of Mary-Lou Pardue and Svetlana Rashkova.)

Figure 11.7. GFP-a-Tubulin (Tub) in dividing germ cells in the ovary. During oogenesis, a germ stem cell gives rise to a daughter cystoblast, which subsequently undergoes 4 incomplete mitotic divisions to produce a cyst of 16 interconnected germ cells, one of which is destined to become an oocyte. The germ cells of developing cysts are connected by the spectrin-rich fusome, which plays an important role in orienting these cells and in oocyte specification. GFP-Tub was expressed in the germ cells by the Gal4-UAS system. In this mitotic cyst producing 8 germ cells, one end of each spindle is associated with the fusome (red). Later, after 16 cells are formed, the fusome is required to polarize the interphase microtubule network, an event that accompanies oocyte specification. (From Grieder et al., 2000. Courtesy of Allan Spradling.)

Figure 11.11. GFP-labeled nos mRNA in living (A, D, E) and fixed (B, C, F) Drosophila egg chambers and embryos. (A) GFP-labeled nos RNA is visible in the oocyte of early egg chambers. Excess MCP-GFP fusion protein that is not bound to nos RNA enters the nurse cell nuclei. In these egg chambers, MCP-GFP fusion protein alone is also expressed in the follicle cells. (B) During midoogenesis, GFP-labeled nos RNA is transiently localized to the anterior margin of the oocyte. Nurse cell and follicle nuclei appear yellow/orange due to the overlap of Hoescht (red) and unbound MCP-GFP (green). (C) Z-series projection of the posterior of a stage 13 oocyte showing particles of GFP-labeled nos RNA at the cortex. (D) GFP-labeled nos RNA is also detected in particles at the posterior cortex of the early embryo. (E) GFP-labeled nos RNA in pole cells during gastrulation. (F) GFPlabeled nos RNA overlaps Vasa protein (detected by anti-Vasa immunostaining in red) in newly formed pole cells. (From Forrest and Gavis, 2003. Figure and legend courtesy of Elizabeth Gavis.)

Figure 11.14. A GFP reporter to study innate immunity. The drosomycin (dros) promoter was used to drive expression of GFP in transgenic flies. A. dros-GFP is induced strongly in flies that are immunized (challenged by microbial infection), but only at low levels in unchallenged flies (compare the top and bottom flies). B. dros-GFP is expressed in the fat body of immunized larvae (top) but not in control larvae (bottom). C. Dissected fat body of an immunized adult. Higher magnification image of the immunized larva shown in B. (From Ferrandon et al., 1998. Courtesy of Dominique Ferrandon.)

䉳 Figure 11.13. GFP as a reporter for transcriptional silencing. Reporter constructs were designed to measure the effects of heterochromatic gene silencing on two adjacent genes, mini-white and UASGFP. With this system, de-repressive effects of Gal4-induction of UASGFP in the context of flanking heterochromatin could be determined. Shown are eyes of flies carrying two different reporter insertions, x21 (A) and x18.4.1 (B), and two different Gal4 drivers (A5CGAL in A, and GMRGAL in B). The left panels are light microscopy images of the eyes, and the middle and right panels are fluorescent images showing the red pigments (middle) or both the red pigments and GFP (right). In some cells, uncoupling of UASGFP and mini-white expression occurs: the white lines indicate areas where GFP is expressed, but mini-white is silenced. (From Ahmad and Henikoff, 2001. Coutesy of Steve Henikoff and Kami Ahmad.)

Figure 11.15. Two-color GFP shows that RNAi is cell-autonomous in Drosophila embryos. In the control panel (left), all segments of the embryo express a GFP fusion protein from a transgene driven by the poly-ubiquitin promoter. The posterior domain of each segment (two segments are shown) also expresses a GFP fusion protein from a UAS-regulated transgene in combination with en-GAL4. The overlap of GFP and CFP appears yellow. In the right panel, both transgenes are expressed in addition to an en-Gal4-induced ds RNA that targets the RNA encoding GFP. Expression of the GFP fusion is specifically repressed in the en domain, indicating that RNAi is cell autonomous and cannot spread to the anterior compartments. (From van Roessel et al., 2003.

A

B

C

D

E

F

Figure 11.16. GFP balancer chromosomes. Balancer chromosomes were constructed that carry UAS-GFP and Kruppel (Kr)-GAL4. Flies bearing these balancers express GFP in the Kr pattern. In embryos, zygotic expression of GFP commences during germ band extension and can be detected at all subsequent stages. A. Stages 4–5, cellularization. The yellow signal is yolk. B. Stages 8, early germ band extension. C. Stages 9–12, germ band extension. GFP is first detected. D. Stages 13–14, germ band retraction. E. Stages 16. F. Stages 17, late embryo. a = amnioseroasa, bo = bolwig’s organ, cd = central domain, s = spiracles, y = yolk. (From Casso et al., 2000. Courtesy of Tom Kornberg and Dave Casso.)

Figure 13.1. Stable transgenic zebrafish expressing GFP in specific tissues. (A, B) GATA-1 GFP expression in hematopoietic cells (Long et al., 1997). (C, D) GATA-2 BAC GFP expression in neuronal cells (Shuo Lin, unpublished). (E, F) Rag-1 GFP expression in olfactory sensory neurons (Jessen et al., 1999). (G, H) Rag-1 BAC GFP expression in thymus (Jessen et al., 1999). (I) Insulin GFP expression in pancreatic beta cells (Huang et al., 2001). (J) POMC GFP expression in pituitary cells (Liu et al., 2003). (K) FLK GFP expression in vascular cells (Cross et al., 2003).

Figure 13.2. GFP expression in transgenic mouse and chick. (A, B) GFP expression in neural tube and neural crest following electroporation of chick with a GFP encoding plasmid (green) and anti-HNK1 immunostaining (red) to mark neural crest. (A) Lateral view of whole mount (Maria Elena de Bellard and Marianne Bronner-Fraser, unpublished). (B) Cross section through neural tube with DAPI staining (blue) to mark nuclei (Ed Coles and Marianne Bronner-Fraser, unpublished observations). (C) Yolk sac of an E9.5 transgenic mouse showing e–globin GFP expression in red blood cells (Dyer et al., 2001; Elizabeth Jones, unpublished observations). (D) Section through cerebellum of Calbindin BAC GFP transgenic mouse showing expression in Purkinje cells (Xiangdong William Yang and Nat Heintz, unpublished observations).

A

ER

ssKDEL-GFP

ER exit sites

Sec13-GFP

Golgi

GalT-GFP

PM + Golgi

GPI-GFP

B

Figure 14.1. Examples of GFP chimeras and their subcellular localization. (A) Seady-state distribution of several proteins. (B) Confocal images of a cell expressing the secretory cargo protein VSVG-GFP imaged by time lapse as the protein leaves the Golgi apparatus. Eight images at 10-s intervals were overlaid. (Boxed areas) The route of a single post-Golgi carrier to the cell periphery. [Courtesy of Hirschberg et al. (1998).] A

B

Control

Staurosporin

Figure 15.4. Dual-color analysis for monitoring Bid activation with DsRed2. HeLa cells were transiently cotransfected with plasmids encoding the fusion protein Bid-DsRed2 and a mitochondriatargeted ZsGreen1 (ZsGreen1-Mito). (A) Before induction of apoptosis, Bid-DsRed2 is localized in the cytosol and ZsGreen1-Mito labels the mitochondria. (B) After induction of apoptosis with 1 mM staurosporine for 3 h, the relocalization of Bid-DsRed2 to mitochondria as revealed by the colocalization with the mitochondria marker ZsGreen1-Mito. Images were taken with a 100¥ objective using Chroma filter sets hq460/40x, 490dclp, and hq515/30m for ZsGreen1 and using hq545/50x, 580dcxr, and hq630/60m for DsRed2.

Figure 15.5. Detection of three fluorescent proteins by fluorescent microscopy. HeLa cells were separately transfected with plasmids pAmCyan1-N1, pZsYellow1-N1, and pHcRed1-N1, mixed, and observed by microscopy using Chroma Technology Corp. filter sets d440/40x, 470dcxr, and d500/40m for AmCyan1, using hq500/40, 530dclp, and hq550/40m for ZsYellow1, and using hq575/50x, 610dclp, and hq640/50m for HcRed1.

GABA R2 cyan

GABA RI yellow

Overlay

FRET

Figure 16.3. Confocal visualization of GABA-B receptor heterodimerization. HEK 293T cells were transfected with the fusion proteins, GABA-BR2/cyanRCFP and GABA-BR1/yellowRCFP. From left to right in the figure, the images show cellular expression of GABA-BR2/cyanRCFP (excitation at 433 nM, emission at 475 nM), GABA-BR1/yellowRCFP (excitation at 488 nM, emission at 525 nM; the overlay of the first two images demonstrate that both proteins are expressed at the same site), and the FRET signal (excitation at 433 nM; emission at 525 nM). The FRET event demonstrates that the proteins are in close proximity.

Figure 17.1. Schematic of GFP dissection. (A) The original system used by Ghosh et al. (2000) split GFP at 157–158. The reassembled GFP, fused to antiparallel leucine zipper peptides (blue), is depicted with the N- and C-terminal fragments are colored green and red, respectively. (B) The dissection points discussed in the text are highlighted. Those in bold have been the most generally successful. Created with PyMOL (http://www.pymol.org) from PDB entries 1EMA and 1SER.

Figure 17.2. Multicolor reassembly of fluorescent proteins. Reassembly of CFP(155–238) with (A) YFP(1–172), (B) GFP(1–172), (C) BFP(1–172), and (D) CFP(1–172) results in yellow, green, blue, and cyan cells. [Adapted from Hu and Kerppola (2003) with permission.]

1 DISCOVERY OF GREEN FLUORESCENT PROTEIN Osamu Shimomura The Photoprotein Laboratory, Falmouth, MA

1.1

DISCOVERY OF GFP

It was early July in 1961. Dr. Frank Johnson and I were studying the bioluminescence of the jellyfish Aequorea aequorea (see Section 1.4 concerning the species name) at the Friday Harbor Laboratories of the University of Washington, located on a small island near Victoria, British Columbia, Canada. Since early morning of that day, we were trying to develop a practical method to isolate the light-emitting matter of the jellyfish, a substance later named “aequorin” (cf. Shimomura et al., 1962; Shimomura, 1995a), of which we had found the basic principle of solubilization and extraction the day before. In the course of our experiments, however, I became deeply annoyed and also puzzled when I realized that the light emitted from the extract was clearly blue, contrary to our expectation of green light identical to the luminescence of live specimens. A mature specimen of A. aequorea looks like a transparent, hemispherical umbrella, with its mouth at the underside of the body (Fig. 1.1, top). Average mature specimens measure 7–10 cm in diameter. Due to the high transparency of the body, the jellyfish can function as a magnifier lens when the mouth is fully open. The light organs, consisting of about 200 tiny granules, are distributed evenly along the edge of the umbrella, making a full circle. Soaking a specimen of the jellyfish in a dilute potassium chloride (KCl) solution in a darkroom causes the light organs to luminesce, exhibiting a ring of bright green light in the darkness (Fig. 1.1, bottom). If a specimen is soaked in distilled water, a green ring is first observed, which gradually changes into blue with the progress of the cytolysis of cells. Under an ultraviolet light, a specimen of fresh jellyfish exhibits a ring of brilliant green fluorescence, similar to the luminescence caused by KCl. Green Fluorescent Protein: Properties, Applications, and Protocols, Second Edition Edited by Martin Chalfie and Steven R. Kain Copyright © 2006 John Wiley & Sons, Inc.

1

2

DISCOVERY OF GREEN FLUORESCENT PROTEIN

Figure 1.1. Mid-summer specimens of Aequorea aequorea photographed in natural environment (top) and in seawater supplemented with KCl in darkroom (bottom), both at the University of Washington’s Friday Harbor Laboratories. See color insert.

The margin of the umbrella containing the light organs can be cut off with a pair of scissors, yielding a 2- to 3-mm-wide strip called the “ring.” When the rings obtained from 20–30 jellyfish were squeezed through a rayon gauze, a dimly luminescent, turbid liquid called the “squeezate” is obtained. The granules of light organs in the squeezate can be collected by filtration or centrifugation. When the granules are mixed with dilute neutral

DISCOVERY OF GFP

buffer solutions, they are cytolyzed and emit light. When mixed with a pH 4.0 buffer, however, the granules are cytolyzed without light emission, preserving the light-emitting activity in the solution. After the removal of cell debris by centrifugation, the pH 4.0 cellfree solution can be luminesced by the addition of a neutral buffer solution containing Ca2+. These are the outline of the procedure we were doing on that day in July 1961, and I saw that the luminescence of the neutralized solution was blue, contrary to our expectation. I doubled, then tripled, the number of the jellyfish used in each experiment in order to make the final luminescence stronger and clearer, but these efforts only helped to confirm my observation. My question concerning the seeming discrepancy remained in my consciousness, until we found an explanation more than 10 years later. After returning to Princeton University with the jellyfish extracts, we purified the light-emitting substance. The substance obtained was a protein capable of emitting light in the presence of Ca2+; the protein was named aequorin. During the purification of aequorin, we noticed the existence of a green fluorescent protein in the jellyfish extract. Upon column chromatography, a green fluorescent band moved closely together with the band of aequorin on a Sephadex G-100 column and moved ahead of the aequorin band on a DEAE-cellulose column. Although the presence of a green fluorescent substance in the light organs was previously known (Davenport and Nicol, 1955), it was the first time that the substance was isolated and recognized to be a protein. Our observation was mentioned in our first full article on the purification and characterization of aequorin (Shimomura et al., 1962), in a footnote, as follows: A protein giving solutions that look slightly greenish in sunlight though only yellowish under tungsten lights, and exhibiting a very bright, greenish fluorescence in the ultraviolet of a Mineralite, has also been isolated from the squeezates. No indications of a luminescent reaction of this substance could be detected.

The first measurements of the luminescence spectrum of aequorin and the fluorescence spectrum of the green protein were reported quickly (Johnson et al., 1962). The luminescence spectrum of aequorin was broad, with a peak at 460 nm. The fluorescence spectrum of the green protein was sharp, with a peak at 508 nm. Apparently, the light organs of the jellyfish contain these two proteins, aequorin and the green protein, of which the former emits blue light in the presence of Ca2+ and the latter emits green fluorescence when excited. The green protein was later called green fluorescent protein, GFP (Hastings and Morin, 1969). One average-sized specimen contains 20–30 mg of aequorin (Shimomura and Johnson, 1979), and each of its about 200 light organs contains approximately 0.1 mg of aequorin and 0.02 mg of GFP (Morise et al., 1974; Cutler, 1995). How can a protein, aequorin, luminesce just by the addition of Ca2+, even in the absence of oxygen? Why is the luminescence of a live jellyfish green, while aequorin emits blue light? Regarding the first question, it seems clear that the luminescence is produced by an intramolecular chemical reaction of aequorin triggered by Ca2+. Thus, we would need to understand the mechanism of this intramolecular reaction, which was a formidable task to accomplish at the time. To answer the second question, it would be necessary to consider two possibilities: (1) a filtering effect by the green protein or something else that shifts the emission maximum of aequorin luminescence to longer wavelength and (2) an energy transfer from aequorin molecules to the green protein by a certain mechanism. Considering that the fluorescent protein was created by nature presumably under some selective pressure, the possibility of an energy transfer would be more likely. In those days, however, we were not concerned with the details of the energy-transfer mechanism;

3

4

DISCOVERY OF GREEN FLUORESCENT PROTEIN

we merely assumed that the green protein absorbed the blue light of aequorin, and then reemitted the absorbed energy as green light (i.e., an energy transfer by the trivial mechanism). We deferred the studies of these subjects for the next 5 years, because of various reasons. Aequorin is an unusual protein that contains an energy-producing source for light emission inside the molecule, resembling a luciferin in this respect. However, it seemed inappropriate to designate aequorin a luciferin because of its heat-labile and nondiffusible nature. In 1965, we discovered the second example of a bioluminescent protein that contains the energy source of luminescence in the molecule from the marine tubeworm Chaetopterus, and we proposed to use the general term “photoprotein” to refer to this type of protein (Shimomura and Johnson, 1966). Thus, a photoprotein is a naturally occurring bioluminescent protein that is capable of emitting light in proportion to the amount of the protein (Shimomura, 1984). The term is now widely used, and many different kinds of photoprotein are presently known—for example, Ca2+-sensitive photoproteins from coelenterates (aequorin, obelin, mnemiopsin) and protozoa (thalassicolin); superoxide-activated photoproteins from scaleworm (polynoidin) and the clam Pholas (pholasin); and an ATP-activated photoprotein from a Sequoia millipede Luminodesmus (Shimomura, 1984).

1.2 ISOLATION AND PROPERTIES OF THE GREEN FLUORESCENT PROTEIN Ridgway and Ashley (1967) reported the first successful application of aequorin bioluminescence. They microinjected aequorin into barnacle muscle single fibers, and they monitored the concentration changes of Ca2+ that occur during muscle contraction. The study clearly demonstrated the usefulness and importance of aequorin in the studies of intracellular calcium, causing a rush of requests for this photoprotein. For the efficient and productive use of aequorin, detailed knowledge on the properties of aequorin and the mechanism of light emission became necessary. Thus, we decided to try to solve the chemical mechanism of aequorin luminescence, an intramolecular reaction. It seemed to be an exceedingly difficult, almost unachievable undertaking at the time. After several years of strenuous efforts, however, we had the luck to be able to uncover a large part of the intramolecular chemistry involved in the Ca2+ triggered luminescence of aequorin, including the chemical structure of the functional moiety “coelenterazine” in the protein and also the means to regenerate spent aequorin into the original, active aequorin (Shimomura and Johnson, 1969, 1972, 1973, 1975). During the same period, green fluorescent proteins similar to Aequorea GFP were found in a number of other bioluminescent coelenterates (Hastings and Morin, 1969; Morin and Hastings, 1971a,b; Wampler et al., 1971, 1973; Cormier et al., 1973, 1974; Morin, 1974); those green fluorescent proteins apparently function as the light emitter of in vivo bioluminescence, as in the case of Aequorea. Green fluorescent protein was not found in the jellyfish of Scyphozoa (such as Pelagia and Periphylla) and Ctenophora (such as Mnemiopsis and Beroë). The following genera of bioluminescent coelenterates contain GFP: Class Hydrozoa The jellyfish Aequorea The jellyfish Mitrocoma (synonym Halistaura) The hydroid Obelia The jellyfish Phialidium (hydroid Clytia)

ISOLATION AND PROPERTIES OF THE GREEN FLUORESCENT PROTEIN

Class Anthozoa Acanthoptilum The sea cactus Cavernularia The sea pansy Renilla The sea pen Ptilosarcus and Pennatula Stylatula Concerning the mechanism of energy transfer from the excited state of photoprotein molecule to GFP molecule, Morin and Hastings (1971b) suggested for the first time that the mechanism of coelenterate bioluminescence possibly involves the Förster-type radiationless energy transfer. To clarify the mechanism of energy transfer involved in the emission of green light from the jellyfish Aequorea, we isolated and purified the green fluorescent protein from the jellyfish, and then we studied its properties in detail (Morise et al., 1974). The purified Aequorea GFP was easily crystallized by decreasing the ionic strength of the solvent (Fig. 1.2). We investigated the energy transfer from aequorin molecule to GFP molecule during the Ca2+-triggered luminescence reaction of aequorin, under two sets of conditions: one with high concentrations of GFP (1.7–5.5 mg/ml) and the other with relatively low concentrations of GFP (0.15–1.1 mg/ml). In the presence of the high concentrations of GFP, apparently an energy transfer by the trivial (radiative) mechanism takes place, at least to some extent. Namely, the light emitted from aequorin (emission lmax 460 nm) is absorbed by GFP (lmax 400 and 480 nm),

Figure 1.2. The fluorescence photomicrograph of the crystals of Aequorea GFP formed in a low ionic strength aqueous solution, by dialysis against pure water. The fluorescence of GFP crystal is strongly anisotropic (Inoué et al., 2002). The view field shown is about 0.5 mm wide. Photograph by Dr. Shinya Inoué.

5

6

DISCOVERY OF GREEN FLUORESCENT PROTEIN

followed by reemission of the absorbed energy from GFP as fluorescence (emission lmax 509 nm). In this mechanism, the extent of energy transfer and the spectral shape of emitted light are dependent on the GFP concentration. It is clear, however, that GFP cannot absorb all the light emitted from aequorin, because the luminescence emission of aequorin extends to about 600 nm on the red side of wavelength whereas GFP can absorb light up to only about 510 nm. Therefore, a complete energy transfer by the trivial mechanism is clearly impossible. In any event, a very high concentration of GFP (with a very high level of absorbance) is required to obtain a significant extent of energy transfer by the trivial mechanism. Under such a condition, the self-absorption of GFP would strongly affect the spectral shape of the fluorescence emitted from GFP, in two ways: (1) a very steep decrease in the light intensities below 510 nm and (2) a red shift of the fluorescence peak position. The actual luminescence spectrum should be the sum of the aequorin luminescence unabsorbed by GFP and the GFP fluorescence distorted by self-absorption; it would be unthinkable that such a spectrum coincides with the true, undistorted spectrum of GFP fluorescence or with the luminescence from the live Aequorea. When aequorin was luminesced with Ca2+ in a low ionic strength buffer (10 mM sodium phosphate) in the presence of relatively low concentrations of GFP (about 0.15 mg/ml), the emission spectrum of aequorin was little affected by GFP. However, when a small amount of fine particles of diethylaminoethyl (DEAE) cellulose or DEAE Sephadex (anion exchangers) was mixed in advance to the same solution, the Ca2+triggered luminescence of the clouded mixture became spectrally identical with the in vivo bioluminescence of Aequorea, indicating the occurrence of an efficient energy transfer from the aequorin light emitter to GFP. It should be pointed out that the amounts of aequorin and GFP, as well as the volume used, were kept equal in the aforementioned experiments (i.e., the overall concentrations and the absorbance values were unchanged); the only difference was the DEAE material added in the latter experiment. The interpretation of the above-mentioned finding is as follows. Under the conditions used, the DEAE cellulose particles had co-adsorbed GFP and aequorin by anion exchange mechanism, greatly increasing the local concentrations of the two proteins around the particles. The co-adsorption perhaps made the distance between the GFP molecules and the aequorin molecules sufficiently short (roughly 30 Å) to make the Förster-type (radiationless) energy transfer workable. Thus, the result observed was the green light that spectrally matches with the in vivo luminescence and the fluorescence emission of GFP. Because the radiationless process is not significantly influenced by the concentration of GFP and does not require a very high concentration of GFP, the energy transfer can take place without being significantly affected by the absorbance of GFP. On the basis of the above experiments and discussion, the energy transfer involved in the emission of green light from live Aequorea is considered to be mostly, if not entirely, a radiationless process. The quantum yield of the Ca2+-triggered aequorin luminescence is approximately 0.16 at 23–24°C (Shimomura and Johnson, 1970; Shimomura, 1986), and that of aequorin coadsorbed with GFP is the same as that of aequorin alone (Morise et al., 1974). In a live specimen of Aequorea, each light organ (0.4 ¥ 0.2 ¥ 0.1 mm) is packed with photogenic cells (average size 10 mm), and each photogenic cell is again densely packed with fine particles (diameter 0.5 mm), according to Davenport and Nicol (1955). It is believed that these particles contain high concentrations of aequorin and GFP. In the particles, aequorin molecules and GFP molecules must be very closely and tightly arranged, if they are not directly bound to each other, to allow an efficient energy transfer by a radiationless process. In fact, the concentration of aequorin and GFP in the photogenic cells are previously estimated to be 5% each, or 10% altogether, of the weight of the cells

ISOLATION AND PROPERTIES OF THE GREEN FLUORESCENT PROTEIN

(Morise et al., 1974). In a more recent estimate, the concentration of GFP was estimated at 2.5% (Cutler, 1995). Another kind of green fluorescent protein, the GFP of the sea pansy Renilla, was purified and physicochemically characterized (Ward and Cormier, 1979). There are substantial differences between the bioluminescence systems of Aequorea and Renilla, though in both systems the light energy is provided by the oxidation of coelenterazine. The in vivo bioluminescence reaction of Renilla requires coelenterazine (the luciferin), Renilla luciferase, Renilla GFP, and molecular oxygen, whereas that of Aequorea requires only aequorin, Ca2+, and Aequorea GFP. The fluorescence emission peak of Renilla GFP (509 nm) is identical to that of Aequorea GFP, but its absorption spectrum (lmax 498 nm) is markedly different from that of Aequorea GFP (lmax 400 nm and 480 nm). Addition of coelenterazine to a solution containing Renilla luciferase results in the emission of blue light. However, when Renilla GFP has been added to the luciferase solution before the addition of coelenterazine, green luminescence is emitted with a threefold increase in the quantum yield, clearly indicating the occurrence of radiationless energy transfer (Ward and Cormier, 1979). Thus, in the case of Renilla, there must be a sufficiently strong binding affinity between the molecules of luciferase and GFP, to make the distance between the chromophores sufficiently short for the energy transfer by radiationless process. It appears that the affinity between Renilla luciferase and Renilla GFP is much greater than that between aequorin and Aequorea GFP. The fluorescence quantum yields of Aequorea GFP and Renilla GFP are nearly equal in a range of 0.7–0.8 (Morise et al., 1974; Kurian et al., 1994; Chapter 4, this volume). However, Renilla GFP significantly increases the quantum yield of bioluminescence, but Aequorea GFP does not, as noted earlier. The difference must come mainly from the difference in the fluorescence quantum yields of coelenterazine light-emitters in the two systems, on the basis of the following discussion. The quantum yield of bioluminescence, Qbl, can be expressed, in a practical way, as the product of the yield of the excited state generated, E, and the fluorescence quantum yield of the light emitter, Qf. Thus, Qbl = EQf, where the values of Qbl, E, and Qf cannot exceed 1. In the Ca2+-triggered light emission of aequorin, the photoprotein is decomposed into apoaequorin, coelenteramide, and carbon dioxide, wherein apoaequorin, coelenteramide, and calcium ions form a complex called the blue fluorescent protein “BFP” (Shimomura and Johnson, 1970). The aequorin luminescence is emitted from BFP or, more precisely, from the amide anion of coelenteramide in excited state (i.e., coelenterazine light emitter) bound to apoaequorin in BFP (Hori et al., 1973; Shimomura, 1995b). The quantum yield Qbl of aequorin luminescence is 0.16 as already noted. The fluorescence quantum yield Qf of BFP measured a few seconds after the light emission was 0.12, but this value is considered to be inaccurate on the basis that BFP is a dissociable equilibrium complex and also that the conformation of apoaequorin changes after the light-emitting reaction (Morise et al., 1974). At present, there seems to be no way to measure an accurate value of Qf at the moment of light emission in the case of aequorin luminescence. However, because the Qbl of aequorin luminescence and the Qbl of the luminescence of aequorin coadsorbed with GFP are equal (0.16), the Qf of BFP at the moment of light emission should be equal to that of GFP (0.7–0.8) when the energy transfer from BFP to GFP is 100%. In the luciferase-catalyzed Renilla bioluminescence, quantum yield Qbl is increased three times by the addition of Renilla GFP, and the fluorescence quantum yield Qf of Renilla GFP is 0.7–0.8 as already noted. The bioluminescence quantum yield Qbl in the absence of GFP was reported at 0.055 (Matthews et al., 1977) and 0.1 (Inouye and

7

8

DISCOVERY OF GREEN FLUORESCENT PROTEIN

Shimomura, 1997). Assuming the energy transfer at 100%, the fluorescence quantum yield Qf of the coelenterazine light-emitter of the Renilla system is calculated to be approximately 0.15–0.27, which is significantly lower than the Qf value for the coelenterazine light emitter in BFP (0.7–0.8). Regarding the nature of the chromophore, it is believed that the GFPs of Aequorea, Renilla, and many other coelenterates contain an identical chromophore (Ward and Cormier, 1978; Ward et al., 1980); the only exception presently known is the GFP of the jellyfish Phialidium that shows a blue-shifted fluorescence emission peak at 497 nm (Levine and Ward, 1982).

1.3

DISCOVERY OF THE STRUCTURE OF GFP CHROMOPHORE

In 1979, I was interested in the chemical structure of the chromophore of Aequorea GFP, which had never been studied before. From a papain digest of heat-denatured GFP, I isolated a small peptide containing the chromophore. I synthesized a model compound of the chromophore. Based on the resemblance between this model compound and the chromophore of the peptide, I was able to deduce the structure of the GFP chromophore to be the structure D in Fig. 1.3 (Shimomura, 1979). It might look as though I were very lucky in my guesswork, because the data obtained from the peptide were clearly insufficient to elucidate the structure of the chromophore. In fact, several people questioned me as to how I could guess the imidazolone structure. The truth is that I was certainly lucky, but not only in my guesswork. CH3 CH3

Rb Ra

Figure 1.3. (A) A tentative structure of Cypridina luciferin proposed in 1959. (B) One of the model compounds synthesized to test the feasibility of the structure A. (C) A model compound of GFP chromophore synthesized. (D) The chromophore of GFP proposed in 1979. Both Ra and Rb are peptide residues.

A NOTE ON THE SPECIES NAME OF THE JELLYFISH FROM WHICH AEQUORIN AND GFP WERE ISOLATED

In the period of the late 1950s, I was studying the structure of the luciferin of the ostracod Cypridina at the laboratory of Professor Y. Hirata, Nagoya University. The techniques for structure determination available at the time were not as sophisticated as at present. The modern techniques that would produce clear-cut information, such as nuclear magnetic resonance (NMR), high-resolution mass spectroscopy, and high-performance liquid chromatography (HPLC), were not available. In an early stage of our study on Cypridina luciferin, we arrived at a tentative structure that contained an imidazolone ring, A (Hirata et al., 1959). To test the absorption spectrum of this tentative structure, we synthesized various imidazolone compounds that contained one double bond conjugated with the imidazolone ring (Shimomura and Eguchi, 1960), although the results eventually showed that structure A was incorrect. Compound B was one of the imidazolones synthesized at that time. When I obtained the chromophore-bearing peptide from Aequorea GFP in 1979, I immediately noticed a close resemblance in spectroscopic and other properties between the chromophore-bearing peptide obtained from GFP and the imidazolone compound B, which was synthesized some 20 years before. A small difference found in the wavelength of the absorption peak was thought to be the effect of a phenolic OH, based on the evidence that acid hydrolysis of the peptide yielded p-hydroxybenzaldehyde. I synthesized a new model compound C. The spectroscopic properties of compound C were in satisfactory agreement with those of the peptide. Thus, structure D was proposed as the chromophore of GFP (Shimomura, 1979). The chromophore structure was confirmed later to be correct, although the side chains were different (Cody et al., 1993). I learned in 1979 that W. W. Ward of Rutgers University, the pioneer of the isolation of the photosensitive ctenophore photoproteins (Ward and Seliger, 1974a,b), had been working on Aequorea GFP in addition to Renilla GFP. I thought my role was over and decided to discontinue my work on GFP. Since then, the work on Aequorea GFP by Ward and others has steadily progressed, finally developing into the successful cloning of GFP (Prasher et al., 1992), a memorable event that established the basis of using GFP. The cloning was soon followed by the expression of GFP in living organisms by Chalfie et al. (1994) that triggered the explosive popularity of GFP and made the foundation of the present volume.

1.4 A NOTE ON THE SPECIES NAME OF THE JELLYFISH FROM WHICH AEQUORIN AND GFP WERE ISOLATED This brief discussion concerning the names of Aequorea species is included here in consideration of the problems and confusions induced by the recent common use of the species name Aequorea victoria in place of Aequorea aequorea (and Aequorea forskalea). The species names A. aequorea (Forskal, 1775) and A. forskalea (Peron and Lesueur, 1809) are synonymous, and both names have been commonly used; the decision of priority between them appears to be a matter of opinion. The species A. aequorea is highly variable in both form and color (Mayer, 1910) and is distributed very widely—Mediterranean; Atlantic coasts, from Norway to South Africa and Cape Cod to Florida; northeastern Pacific; east coast of Australia; and Iranian Gulf (Kramp, 1968). According to Mayer (1910), A. victoria (Murbach and Shearer, 1902) from the northeastern Pacific is probably a variety of A. aequorea. He stated “I cannot distinguish this medusa from Aequorea forskalea of the Atlantic and Mediterranean. Were it described from the Atlantic I would not hesitate to designate it A. forskalea.” Mayer’s opinion has been overwhelmingly accepted until recently (Russell, 1953; Kramp, 1965, 1968); thus the jellyfish we collected

9

10

DISCOVERY OF GREEN FLUORESCENT PROTEIN

in the Friday Harbor area have been called A. aequorea. The situation changed, however, after Arai and Brinckmann-Voss (1980) reported their conclusion to separate A. victoria from the species A. aequorea, based on their study of about 40 specimens collected from more than 10 different areas around Vancouver Island (Friday Harbor included). Their reasons were that A. victoria has much more regularly serrated mouth lobes and a much larger (thus, taller), almost hemispherical lens in the stomach region, when compared with A. aequorea from the Mediterranean. It is not clear in the Arai and Brinckmann-Voss article why the conclusion to separate A. victoria from A. aequorea was made on the basis of the comparison between the former (from British Columbia and Puget Sound) and the latter from only the Mediterranean; their use of only a few specimens per study area, collected probably on a single occasion, brings about another problem. It has been well documented that a wide intraspecific variation of A. aequorea by geography exists (Mayer, 1910; Russell, 1953; Kramp, 1959, 1965). The Mediterranean form of A. aequorea is only one of many variations of this species. Therefore, the difference between A. victoria and A. aequorea cannot be fully determined by the comparison between the former and the latter from only the Mediterranean; to determine the difference, A. victoria should be compared with various other forms of A. aequorea. It seems particularly intriguing to compare A. victoria with the varieties of A. aequorea from the northern and western Atlantic. In our record, we collected a large number of A. aequorea at Woods Hole, MA, in the summer of 1987, when there was a strong easterly wind; those medusae appeared to be indistinguishable from the average specimens of Aequorea obtained at Friday Harbor in both form and the composition of aequorin isoforms. Most of the several million specimens of Aequorea used for biochemical research had been collected around Friday Harbor, where the specimens were extremely abundant at least until 1988 (since then, they virtually disappeared from the area for unknown reason). If all those medusae were a single species of A. victoria, as implied by Arai and Brinckmann-Voss (1980), it seems that A. victoria must have a very wide variation, like A. aequorea. We have collected over 1 million specimens of Aequorea in the vicinity of Friday Harbor in 17 summers between 1961 and 1988. All the specimens were mature (>7 cm in diameter), and they were collected, handled, and excised individually. More than several times during our operation, we observed pronounced changes in the form of the jellyfish collected. The jellyfish can drift far and widely by current, tide, and wind in groups, and the changes that we observed usually lasted for only a few days but occasionally continued for several weeks. The bell height of the medusae were sometimes markedly higher than usual relative to the diameter (thus taller than hemispherical), and sometimes much flatter and saucer-like. In one of these occasions, we thought that the jellyfish we had collected were a wrong species because they were too flat; we suspended our operation until we had an assurance by a jellyfish expert in the lab that they were indeed a variety of A. aequorea (as known then). If all those medusae at Friday Harbor are the variations of A. victoria, the situation would be very confusing. Because both species have very wide variations without any clear difference between them, distinction of the two species would be extremely difficult. A detailed discussion on the species name of Aequorea is given by Claudia E. Mills (2003). Despite the high intraspecific variability that causes confusions, the species names A. aequorea and A. forskalea have been accepted and used by the majority of researchers for the period of at least 60 years until 1980. To avoid further confusion, and with reference to discussion in other chapters of this volume, the name A. victoria should be considered as a synonym, not as the name of a separate species, until the difference between A.

REFERENCES

victoria and A. aequorea (or A. forskalea) is firmly established on genetic basis. Until that time, the species name A. aequorea (or A. forskalea) should have priority.

ACKNOWLEDGMENTS Our work on the jellyfish Aequorea was initiated by the late Professor Frank H. Johnson, whose contribution to the project was enormous and immeasurable. I sincerely thank all the people who contributed directly or indirectly to this project. The work was made possible by the excellent facilities of the Friday Harbor Laboratories, University of Washington, and research grants from the National Science Foundation and National Institutes of Health.

REFERENCES Arai, M. N., and Brinckman-Voss, A. (1980). Hydromedusae of British Columbia and Puget Sound. Can. Bull. Fish Aquat. Sci. Bull. 204:1–181. 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. 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 green-fluorescent protein. Biochemistry 32:1212–1218. Cormier, M. J., Hori, K., Karkhanis, Y. D., Anderson, J. M., Wampler, J. E., Morin, J. G., and Hastings, J. W. (1973). Evidence for similar biochemical requirements for bioluminescence among the coelenterates. J. Cell. Physiol. 81:291–298. Cormier, M. J., Hori, K., and Anderson, J. M. (1974). Bioluminescence in coelenterates. Biochim. Biophys. Acta 346:137–164. Cutler, M. W. (1995). Characterization and energy transfer mechanism of green fluorescent protein from Aequorea victoria. Ph.D. dissertation, Rutgers University, New Brunswick, NJ. Davenport, D., and Nicol, J. A. C. (1955). Luminescence of hydromedusae. Proc. R. Soc. London Ser. B 144:399–411. Forskal, P. (1775). Descriptiones animalium avium, amphibiorum, piscium, insectorum, vermium: Quae in itinere orientali observavit Petrus Forskal. Post mortem auctoris edidit Carsten Niebuhr. 164 pages. Ex Officina Moller Hauniae (Copenhagen). Hastings, J. W., and Morin, J. G. (1969). Comparative biochemistry of calcium-activated photoproteins from the ctenophore, Mnemiopsis and the coelenterates. Aequorea, Obelia, Pelagia and Renilla. Biol. Bull. 137:402. Hirata, Y., Shimomura, O., and Eguchi, S. (1959). The structure of Cypridina luciferin. Tetrahedron Lett. 5:4–9. Hori, K., Wampler, J. E., and Cormier, M. J. (1973). Chemiluminescence of Renilla (sea pansy) luciferin and its analogues. Chem. Commun. 492–493. Inouye, S., and Shimomura, O. (1997). The use of Renilla luciferase, Oplophorus luciferase, and apoaequorin as bioluminescent reporter protein in the presence of coelenterazine analogues as substrate. Biochem. Biophys. Res. Commun. 233:349–353. Inoué, S., Shimomura, O., Goda, M., Shribak, M., and Tran, P. T. (2002). Fluorescence polarization of green fluorescence protein. Proc. Natl. Acad. Sci. USA 99:4272–4277. Johnson, F. H., Shimomura, O., Saiga, Y., Gershman, L. C., Reynolds, G. T., and Waters, J. R. (1962). Quantum efficiency of Cypridina luminescence, with a note on that of Aequorea. J. Cell. Comp. Physiol. 60:85–104.

11

12

DISCOVERY OF GREEN FLUORESCENT PROTEIN

Kramp, P. L. (1959). The hydromedusae of the Atlantic ocean and adjacent waters. Dana Report No. 46. Carlsberg Foundation, Copenhagen, Denmark. Kramp, P. L. (1965). The hydromedusae of the Pacific and Indian Oceans. Dana Report No. 63. Carlsberg Foundation, Copenhagen, Denmark. Kramp, P. L. (1968). The hydromedusae of the Pacific and Indian oceans, sections II and III. Dana Report No. 72. Carlsberg Foundation, Copenhagen, Denmark. Levine, L. D., and Ward, W. W. (1982). Isolation and characterization of a photoprotein, “phialidin,” and a spectrally unique green-fluorescent protein from the bioluminescent jellyfish Phialidium gregarium. Comp. Biochem. Physiol. 72B:77–85. Matthews, J. C., Hori, K., and Cormier, M. J. (1977). Purification and properties of Renilla reniformis luciferase. Biochemistry 16:85–91. Mayer, A. G. (1910). Medusae of the world, Vol. II (Hydromedusae). Carnegie Institute Washington Publication, Washington, D.C., pp. 231–498. Mills, C. E. (2003). http://faculty.washington.edu/cemills/Aequorea.html. Morin, J. G. (1974). Coelenterate bioluminescence. In Coelenterate Biology. Reviews and Perspectives, Muscatine, L., and Lenhoff, H. M., Eds., Academic, New York, pp. 397–438. Morin, J. G., and Hastings, J. W. (1971a). Biochemistry of the bioluminescence of colonial hydroids and other coelenterates. J. Cell. Physiol. 77:305–311. Morin, J. G., and Hastings, J. W. (1971b). Energy transfer in a bioluminescent system. J. Cell. Physiol. 77:313–318. Morise, H., Shimomura, O., Johnson, F. H., and Winant, J. (1974). Intermolecular energy transfer in the bioluminescent system of Aequorea. Biochemistry 13:2656–2662. Murbach, L., and Shearer, C. (1902). Preliminary report on a collection of medusae from the coast of British Columbia and Alaska. Ann. Mag. Nat. Hist. Ser. 7 9:71–73. Peron, F., and Lesueur, C. A. (1809). Tableau des caracteres generiques et specifiques de toutes les especes de Meduses connues jusqu’a ce jour. Ann. Mus. Hist. Nat. Paris 14:325–366. 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. Ridgway, E. B., and Ashley, C. C. (1967). Calcium transients in single muscle fibers. Biochem. Biophys. Res. Commun. 29:229–234. Russell, F. S. (1953). The Medusae of the British Isles, Vol. I: Anthomedusae, Leptomedusae, Limnomedusae, Trachymedusae and Narcomedusae, Cambridge University Press, London, 530 pages. Shimomura, O. (1979). Structure of the chromophore of Aequorea green fluorescent protein. FEBS Lett. 104:220–222. Shimomura, O. (1984). Bioluminescence in the sea: photoprotein systems. Symp. Soc. Exp. Biol. 39:351–372. Shimomura, O. (1995a). A short story of aequorin. Biol. Bull. 189:1–5. Shimomura, O. (1995b). Cause of spectral variation in the luminescence of semisynthetic aequorins. Biochem. J. 306:537–543. Shimomura, O., and Eguchi, S. (1960). Studies on 5-imidazolone. I–II. Nippon Kagaku Zasshi 81:1434–1439. Shimomura, O., and Johnson, F. H. (1966). Partial purification and properties of the Chaetopterus luminescence system. In Bioluminescence in Progress, Johnson, F. H., and Haneda, Y., Eds., Princeton University Press, Princeton, NJ, pp. 495–521. Shimomura, O., and Johnson, F. H. (1969). Properties of the bioluminescent protein aequorin. Biochemistry 8:3991–3997. Shimomura, O., and Johnson, F. H. (1970). Calcium binding, quantum yield, and emitting molecule in aequorin bioluminescence. Nature 227:1356–1357.

REFERENCES

Shimomura, O., and Johnson, F. H. (1972). Structure of the light-emitting moiety of aequorin. Biochemistry 11:1602–1608. Shimomura, O., and Johnson, F. H. (1973). Chemical nature of the light emitter in bioluminescence of aequorin. Tetrahedron Lett. 31:2963–2966. Shimomura, O., and Johnson, F. H. (1975). Regeneration of the photoprotein aequorin. Nature 256:236–238. Shimomura, O., and Johnson, F. H. (1979). Comparison of the amounts of key components in the bioluminescence systems of coelenterates. Comp. Biochem. Physiol. 64B:105–107. 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. Wampler, J. E., Hori, K., Lee, J., and Cormier, M. J. (1971). Structured bioluminescence. Two emitters during both the in vitro and the in vivo bioluminescence of the sea pansy, Renilla. Biochemistry 10:2903–2909. Wampler, J. E., Karkhanis, Y. D., Morin, J. G., and Cormier, M. J. (1973). Similarities in the bioluminescence from the Pennatulacea. Biochim. Biophys. Acta 314:104–109. Ward, W. W., and Cormier, M. J. (1978). Energy transfer via protein–protein interaction in Renilla bioluminescence. Photochem. Photobiol. 27:389–396. Ward, W. W., and Cormier, M. J. (1979). An energy transfer protein in coelenterate bioluminescence. Characterization of the Renilla green fluorescent protein. J. Biol. Chem. 254:781–788. Ward, W. W., and Seliger, H. H. (1974a). Extraction and purification of calcium-activated photoproteins from ctenophores. Biochemistry 13:1491–1499. Ward, W. W., and Seliger, H. H. (1974b). Properties of mnemiopsin and berovin, calcium-activated photoproteins. Biochemistry 13:1500–1510. 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 green fluorescent proteins. Photochem. Photobiol. 31:611–615.

13

2 PHOTONS FOR REPORTING MOLECULAR EVENTS: GREEN FLUORESCENT PROTEIN AND FOUR LUCIFERASE SYSTEMS J. Woodland Hastings Department of Molecular and Cellular Biology, Harvard University, Cambridge, MA

James G. Morin Section of Ecological Systematics, Cornell University, Ithaca, NY

2.1

INTRODUCTION

During the course of evolution, bioluminescence has repeatedly appeared where it serves biological functions important to the organism. Functions may differ among organisms and a given organism may utilize luminescence in more than one way (Morin, 1983; Hastings, 1983; Hastings and Morin, 1991). The different specific recognized functions may be classed under three major rubrics: defensive (to help deter predators), offensive (to aid in predation), and communication (e.g., for courtship or mating). Within each category a number of different specific strategies are recognized; for example, luminescence may be used defensively as a decoy to divert, as a flash to frighten, or as ventral luminescence to camouflage the silhouette. In terms of the total number of different species, the emission of bioluminescence is rather rare, but it occurs in many phylogenetically different groups (Table 2.1; Harvey, 1952; Herring, 1978). In those groups that do emit light, the biochemical and physiological mechanisms responsible for it are often very different, as are its several functional roles. Indeed, luciferase (the enzyme) and luciferin (the substrate) are generic terms, and quite different in the different groups. Thus, the organism from which they are obtained must be specified.

Green Fluorescent Protein: Properties, Applications, and Protocols, Second Edition Edited by Martin Chalfie and Steven R. Kain Copyright © 2006 John Wiley & Sons, Inc.

15

Photobacterium Vibrio Xenorhabdus Panus, Armillaria Pleurotus Gonyaulax Pyrocystis Noctiluca

Bacteria

Vargula; Cypridina

Crustacea Ostracod

Meganyctiphanes Gaussia

Aldehyde Clam luciferin, structure?, Cu2+ Bacterial symbionts (490)

Latia Pholas Heteroteuthis

Shrimp Copepods and others

N-isovalyeryl-3 amino propanal Unknown (465) Unknown (510) Unknown (530)

Diplocardia Chaetopterus Odontosyllis Acholoë

Linear tetrapyrrole (470) Unknown

Imadazopyrazine nucleus (465)

Ca2+, coelenterazine/aequorin Imidazo pyrazine nucleus (460–510), GFP as accessory emitter Ca2+, coelenterazine (460)

Aequorea Obelia Renilla Mnemiopsis; Beroë

Reduced flavin and long chain aldehyde (475–540) YFP and LUMP as accessory emitters Unknown (535) Linear tetrapyrrole Cell organelles (scintillons) (470)

Luciferins and Other Factors (Emission Max. (nm))

Cnidaria Jellyfish Hydroid Sea Pansy Ctenophores Annelids Earthworms Chaetopterid worm Syllid fireworm Scale worm Molluscs Limpet Clam Squid

Dinoflagellates

Mushrooms

Representative Genera

Type of Organism

TABLE 2.1. Representatives of the Major Bioluminescent Organisms

Squirts enzyme and substrate Diversion, decoy, courtship Photophores; camouflage Deter predators

Exuded luminescence in all three. Photophores and symbiotic bacteria in some squid. Functions: diversion, decoy, camouflage, probably others

Cellular exudates or intracellular flashes sometimes very bright. To divert or deter; courtship

Bright flashes; frighten or deter

Bright flash or train of flashes To frighten or deter

pH change causes short (0.1 s) bright flashes To frighten or deter

Steady bright glow after autoinduction of luciferase Symbiosis Steady dim glow; to attract insects

Displays Features and Functions

16 PHOTONS FOR REPORTING MOLECULAR EVENTS

Symbiotic luminous bacteria (~490) Self-luminous, Vargula-type luciferin, Nutritionally obtained (485) Self-luminous, biochemistry unknown Self-luminous, biochemistry unknown Self-luminous, biochemistry unknown

Cyclothone

Neoscopelus Tarletonbeania

Midwater fishes

Symbiotic luminous bacteria (~490) Symbiotic luminous bacteria (~490)

Leiognathus Photoblepharon

Cryptopsaras Porichthys

Unknown

Cell organelles evolved from bacteria (480–500)

Pyrosoma

Isistius

Trains of rapid flashes; frighten, divert

Biochemistry unknown, Ca2+

Ophiopsila

Camouflage, courtship, deterrence, capture prey Many photophores, ventral and lateral Photophores: lateral, on tongue Sexual dimorphism; males have dorsal (police car) photophores

Ventral luminescence; camouflage, courtship

Ventral luminescence and flashes; camouflage, attract and capture prey, courtship, deter predators, communication

Ventral glow; camouflage

Brilliant trains of flashes; function unknown Stimulated by light and other factors

Lure to attract prey

Flashes, specific kinetic patterns Deter predators; courtship, mating

Benzothiazole, ATP, Mg2+ Similar chemistry in all coleoptera (most 550–580) ATP in Arachnocampa (460 –480)

Photinus, Photuris Pyrophorus Phengodes, Phrixothrix Arachnocampa, Orfelia

Angler fish Midshipman

Fishes Cartilaginous fishes Bony fishes Ponyfish Flashlight fish

Insects Coleopterids (beetles) firefly click beetles railroad worm Diptera (flies) Echinoderms Brittle stars Chordates Tunicates

INTRODUCTION

17

18

PHOTONS FOR REPORTING MOLECULAR EVENTS

Bioluminescence is thus not an evolutionarily conserved function; in the different groups of organisms the genes and proteins involved are mostly unrelated, and evidently originated and evolved independently. How many times this may have occurred is difficult to say, but it has been estimated that present day luminous organisms come from as many as 30 different evolutionarily distinct origins (Hastings, 1983; Hastings and Morin, 1991). Cnidarian luminescence, with green fluorescent protein (GFP) present as an accessory emitter in some but not all species, is thus only one of a wide array of luminescent systems and biochemistries. The genes and proteins of several of the other systems (notably bacterial and firefly) have been used for many different analytical and reporter purposes (DeLuca, 1978; Hastings et al., 1997). Their diversity allows for many different possible applications in which photons are the reporter, but none of the other proteins or systems possesses the unique features of GFP. The biochemistries of only four of the different luminous systems are known in detail (Table 2.1), namely, bacteria, dinoflagellates, cnidarians, and fireflies (Hastings and Morin, 1991; Wilsman and Hastings, 1998). Although some information is known for another half-dozen or so, we confine our review to the four best described three dimensional crystal structures of all four have been reported. While these systems differ in the structures of the luciferins (Fig. 2.1) and luciferases, all systems have some features in common at the chemical level. All known luciferases are oxygenases that utilize molecular oxygen to

Figure 2.1. Structures of four different luciferins, oxygen-containing intermediates, and postulated emitters (see Table 2.1).

CNIDARIANS, CTENOPHORES, AND GFP

oxidize the associated luciferin, giving an intermediate enzyme-bound peroxide, whose breakdown then results in the production of an intermediate or product directly in its excited singlet state. In most systems, emission occurs from the luciferase-bound substrate-derived excited molecule (see Fig. 2.1), but an accessory secondary emitter occurs in certain cnidarians and some bacteria. With GFP in cnidarians, the mechanism responsible was postulated and later confirmed to involved Förster-type energy transfer (Morin and Hastings, 1971b; Morise et al., 1974; Ward et al., 1980). GFP is unusual in that its chromophore is a part of the (modified) primary structure of the protein, thus not subject to dissociation (Cody et al., 1993; Heim et al., 1994). Its use as a transgene reporter, pioneered by Chalfie et al. (1994), relies on this feature and its fluorescence alone.

2.2

CNIDARIANS, CTENOPHORES, AND GFP

Luminescence is common and widely distributed in these groups (Morin, 1974; Cormier, 1981; Herring, 1978). In the ctenophores (comb jellies), they comprise over one-half of all genera, whereas in the cnidarians it is about 6%. These organisms are mostly sessile, sedentary, or planktonic, and upon stimulation they emit light as flashes. Hydroids such as Obelia occur as plant-like growths, typically adhering to rocks and kelp below low-tide level in many of the world’s oceans. Upon stimulation, a conducted scintillating emission emanates as a wave along the colony from individual photocytes (cells specialized for light emission); repetitive waves may occur from a single stimulus. Aequorea, a hydromedusan that is very abundant in the San Juan Islands region of the northwest United States, has been extensively used for biochemical studies (Shimomura and Johnson, 1975; Cormier et al., 1989). The biochemistry of the sea pansy, Renilla, which occurs near shore on sandy bottoms, has also been elucidated (Cormier, 1981). Early observations, attributable to what we now know as GFP, were reported by several investigators, including an emission spectrum of the bioluminescence of the sea pen Pennatula phosphorea showing a narrow bandwidth emission in the green (Nicol, 1958), which is now known to be characteristic of GFP. Shimomura et al. (1962) later noted that the bioluminescence of Aequorea extracts was blue while that of the intact organism was green, attributing this to a protein that fluoresced green in extracts (Johnson et al., 1962). More suggestive of a relationship between luminescence and fluorescence was the observation of green fluorescence in cells located in the vicinity of bioluminescence activity by Titschack (1964) in the pennatulacean Veretillum cynomorium. In none of these studies, however, was the relationship of the green fluorescence to the bioluminescence clearly established. We discovered GFP quite independently while we were examining both the mechanisms controlling luminescent flashes and the biochemical underpinnings of the luminescence in the colonial hydroid Obelia geniculata. Our biochemical studies quickly expanded to studying calcium activated photoproteins in a variety of coelenterates, including several species of hydrozoans (Obelia, Aequorea, Clytia), pennatulaceans (the sea pens Renilla and Ptilosarcus), the scyphozoan jellyfish Pelagia, and the ctenophore Mnemiopsis (Morin et al., 1968; Hastings and Morin 1969a,b). Measurements of emission spectra of reactions in extracts gave wide bandwidth curves peaking in the blue, but somewhat different in different species; the blue emissions of the luciferase systems of Aequorea, Obelia, and Renilla exhibit maxima at 460, 472, and 486 nm, respectively (Fig. 2.2). In vivo luminescent spectra from all three, however, were narrow, peaking in the green at

19

20

PHOTONS FOR REPORTING MOLECULAR EVENTS

Figure 2.2. Emission spectra for bioluminescence in vivo for Obelia, Aequorea, and Renilla compared with spectra from emission of in vitro reactions isolated from these same organisms. [From Morin and Hastings (1971b).]

Figure 2.3. Luminescent flash (L) and luminescent potential (E) bursts recorded concurrently from an O. geniculata photocyte and hydranth, respectively. Responses are to a train of stimuli applied about once every 2 s, indicated by solid triangles beneath the lower trace (horizontal bar = 1 s); lower (E) ordinate, vertical bar = 1 mV; upper (L) ordinate, light intensity in relative units. [From Morin and Cooke (1971a).]

about 508 nm, matching fluorescence emission spectrum of GFP (Morin and Hastings, 1971a). So while the underlying biochemistry may differ somewhat, emissions are all in the green peaking at about 508 nm so long as the luciferases interact appropriately with GFP in order to transfer excitation energy. GFP is thus an accessory emitter protein of the cnidarian luminescent system, deriving its excitation by nonradiative energy transfer in association with the luciferase reaction, which in the absence of GFP emits blue light (Morin and Hastings, 1971b). Many of the properties of GFP, such as its thermal stability and remarkable resistance to proteolysis, derive from its barrel structure, dubbed a b-can, with 11 b strands arranged protectively around a central chromophore (Ormö et al., 1996; Yange et al., 1996). Solvent access to the inside cavity in these “lanterns” is blocked on top and bottom by short segments of a-helices, although some water molecules are immobilized inside; there would clearly be no room for an enzyme to catalyze chromophore formation. The biochemical studies provided critical information for our physiological and morphological studies on the colonial hydroid O. geniculata, where action potentials propagate through electrically excitable epithelial cells. These action potentials spread incrementally through a colony and repetitively elicit flashes from individual photocytes (Fig. 2.3), which are located along the length of the stems and pedicels but not in the polyps themselves (see color Fig. 2.4; Morin et al., 1968; Morin and Reynolds, 1969;

21

CNIDARIANS, CTENOPHORES, AND GFP

(a)

(b)

Figure 2.4. Fluorescence micrograph of a living colony of Obelia species showing photocytes visualized by GFP. Height of field shown is about 3 mm in (a) and 1.5 mm in (b). (a) Dispersed photocytes (bright green spots) in an upright of O. geniculata (two polyps also shown at lower right and upper left). (b) Concentrated photocytes at the tip of a pedicel below the base of a hydranth of O. bidentata (= bicuspidata).

Morin and Cooke, 1971a,b). Indeed, it was the green fluorescence of the protein that allowed us, in conjunction with image intensification, to identify the photocytes by the colocalization of the fluorescence and bioluminescence of the photocytes and to establish irrefutably the association of GFP with the luminescent system (Fig. 2.5; Morin and Reynolds, 1969, 1970, 1974). In different species, GFP is confined to discrete photocytes (~10–20 mm in diameter), which are either dispersed [Fig. 2.4a] or clumped [Fig. 2.4B] in specific locations within the gastrodermis of the colonies (Morin and Reynolds, 1969, 1970, 1974; Morin, 1974). In measuring in vivo bioluminescence from single Obelia photocytes, GFP allowed us to identify their location prior to stimulation, so as to record photometrically via a fine-tipped (0.5 mm) light guide (Fig. 2.3). Trains of action potentials, termed luminescent potentials, initiated by single stimuli, propagate via electrically excitable epithelial cells (rather than neurons). These action potentials spread incrementally through a colony and repetitively excite the photocytes, and they also couple to other neuroid conducting systems such as those governing polyp contraction. Based on the number of photocytes, as determined with the aid of GFP, we were able to calculate from photometric measurements that each cell could emit about 1–2 ¥ 108 quanta/cell (Shorey and Morin, 1974). Studies of the GFP in the hydroid Obelia were also instrumental in providing the first demonstration that gap junctions can pass chemical signals in excitable tissues (Dunlap

22

PHOTONS FOR REPORTING MOLECULAR EVENTS

Figure 2.5. Bioluminescence (A), fluorescence (B), and rear illumination (C) pictures of an Obelia geniculata upright. Note that the luminescent and fluorescent spots (six in each) directly superimpose. The scale bar indicates 200 mm. [From Morin and Reynolds (1974).]

et al., 1987; Brehm et al., 1989). They showed that calcium actually enters neighboring nonluminescent, but electrically excitable, epithelial cells via voltage-dependent calcium channels and then into the photocytes, which are nonexcitable, via secondary calcium diffusion through gap junctions. This calcium then triggers the luminescence from the calcium activated photoprotein (obelin) with subsequent energy transfer to and emission from the GFP. Finally, by using GFP fluorescence as a reporter for the spatial distribution of luminescent cells in pennatulaceans (sea pens; see color Fig. 2.6) and photometry to measure the temporal aspects of the light emission, we have been able to infer that luminescence in sea pens and probably all cnidarians functions as an aposematic signal to deter damage to the colonies by potential aggressors or predators such as fishes and crustaceans (Morin, 1976, 1983). This inference has been experimentally verified for both pennatulaceans and brittle stars by Grober (1988a,b). At the biochemical level, the luciferin (coelenterazine) is the same in different cnidarian luminescent systems. Coelenterazine possesses an imidazopyrazine skeleton (Fig. 2.1) and is notable for its widespread phylogenetic distribution (Thomson et al., 1997), but whether the reason is nutritional or genetic (hence, possible evolutionary relatedness) has not yet been elucidated. But there are differences between the cnidarian anthozoan and hydrozoan systems with regard to the site of calcium action. In the anthozoan, coelenterazine is sequestered by a Ca2+-sensitive binding protein, and Ca2+ causes its release, thus triggering the in vivo flash. The Renilla luciferase reaction (EC 1.13.12.5) does not itself require calcium (Lorenz et al., 1991). In the hydrozoan Aequorea, calcium reacts instead at the luciferase stage, namely with aequorin, a luciferase-bound hydroperoxy coelenterazine intermediate, poised for the completion of the reaction. Aequorin was isolated by Shimomura et al. (1962) from the jellyfish Aequorea [in the presence of ethylenediaminetetraacetate (EDTA) to chelate calcium] and shown to emit light simply upon the addition of Ca2+, which is presumably the trigger in vivo (Hastings and Morin 1971; Blinks et al., 1982; Cormier et al., 1989). It was postulated (Hastings and Gibson, 1963) that in vivo luciferin coelenterazine reacts with oxygen, catalyzed by its luciferase (EC1.13.12.5), to form the hydroperoxide in a calcium-free compartment (the photocyte), where it is stored. Excitation allows Ca2+ to enter and bind to the protein (which possesses homology with calmodulin; Lorenz et al., 1991), changing its conformation so that the reaction continues, but without the need for free oxygen at this stage. It had been reported in the early literature (Harvey, 1952) that coe-

23

CNIDARIANS, CTENOPHORES, AND GFP

(a)

T T

T

M (b)

Figure 2.6. Fluorescence micrographs of photocytes visualized by GFP in living pennatulacean (sea pen) colonies. Width of field shown is about 0.8 mm in (a) and 1.6 mm in (b–d). (a) Photocytes in a cluster of five siphonozooids (water pumping polyps) of Renilla kollikeri. (b) Photocytes clustered in the lateral-axial region of the tentacles and oral disk of an autozooid (feeding polyp) of Renilla kollikeri (mouth [M] and base of three of eight tentacles [T] shown).

lenterates could inexplicably emit bioluminescence without oxygen. The explanation is now evident. Crystal structure determinations of aequorin (Head et al., 2000) and obelin (Liu et al., 2000) have confirmed them to be luciferase-bound peroxy coelenterazine intermediates; the two are very similar and represent an entirely new luciferase fold. As mentioned above, both possess homology with calmodulin and other calcium binding E-F hand proteins. The

24

PHOTONS FOR REPORTING MOLECULAR EVENTS

(c)

T T

T M

T T

T

T

(d)

Figure 2.6. (continued) (c) Photocytes clustered in only the two outer (of the eight) chambers within the calyx of the column (and not the tentacles) of an autozooid of Acanthoptilum gracile. (d) Photocytes clustered laterally along the length of each of the eight tentacles (T) of an autozooid of Ptilosarcus guerneyi (M = mouth).

monomer is considered to be the active species and is compact, having a 25Å radius. Each monomer is predominantly helical, being composed of 4 E-F hand motifs, of which three are able to bind calcium. The structure has been depicted as two cups joined “rim to rim” in which the bottom cup is composed of the first four helices and the top cup by the last four helices.

BACTERIA

Aequorin luminescence has been widely used for the detection and measurement of calcium, most especially in living cells, into which aequorin can be microinjected (Blinks et al., 1982). The first such experiment was reported by Ridgeway and Ashly (1968), in which they detected a calcium transient accompanying the contraction of single muscle fibers. Since then there have been many analogous applications (Blinks et al., 1982), making aequorin an important tool in analytical biochemistry, physiology, and developmental biology. Apoaequorin, which functions as the luciferase in this system, has been cloned and expressed in other cell types (Inouye et al., 1986, 1989; Tanahashi et al., 1990) where, in the presence of exogenously added coelenterazine, it serves to monitor intracellular calcium levels. For example, expressed as a transgene in Dictyostelium, it was used to monitor intracellular calcium changes in response to cyclic adenosinemonophosphate (cAMP) stimulation (Saran et al., 1994). In tobacco and Arabidopsis plants, the expressed transgene revealed circadian oscillations in free cytosolic calcium (Johnson et al., 1995); when targeted to the chloroplast, circadian chloroplast rhythms were likewise observed.

2.3

BACTERIA

Luminous bacteria (see color Fig. 2.7) occur ubiquitously in sea water samples. A primary habitat where most species abound is in association with another (higher) organism, dead (saprophytes) or alive (parasites or symbionts), where growth and propagation occur. Specific associations involve specialized light organs (e.g., in fish and squid; Ruby, 1996) in which a pure culture of luminous bacteria is maintained at a high density and at high light intensity (Nealson and Hastings, 1991). Parasitic and commensal relationships are also known. Terrestrial luminous bacteria are rare, the best described being those harbored by nematodes that are parasitic on insects such as caterpillars.

Figure 2.7. Streaks of luminescent bacteria photographed by their own light, showing two strains of Photobacterium fischeri, one of which emits yellow light by virtue of having YFP (yellow fluorescent protein). The other lacks YFP, emitting only blue light. See color insert.

25

26

PHOTONS FOR REPORTING MOLECULAR EVENTS

Luminous bacteria emit light continuously, peaking at about 490 nm if no accessory protein is present. When strongly expressed, a single bacterium may emit 104–105 photons s-1. The luciferase is a flavin mixed-function monooxygenase (EC 1.14.14.3), and its presence is diagnostic for a bacterial symbiotic involvement in the luminescence of a higher organism. The pathway constitutes a shunt of cellular electron transport at the level of flavin; reduced flavin mononucleotide (FMN) (Fig. 2.1) reacts with oxygen in the presence of bacterial luciferase to produce an intermediate peroxy flavin, which then reacts with a long-chain aldehyde (tetradecanal) to form the acid and the luciferase-bound hydroxy flavin in its excited state (Hastings et al., 1985; Baldwin and Zeigler, 1992). Although there are two substrates in this case, the flavin can claim the name luciferin on etymological grounds, since it forms (bears) the emitter. The bioluminescence quantum yield has been estimated to be about 30%, the same as the fluorescence quantum yield of FMN. Curiously, no other flavin monoxygenases have been found to emit light, even at very low quantum yields, and no genes with significant sequence similarities have been recorded in any of the databases. Bacterial luciferases are heterodimeric (alpha–beta) proteins (~80 kDa) in all species; they are relatively simple, having no metals, disulfide bonds, prosthetic groups, or nonamino acid residues. The crystal structure of V. harveyi luciferase has been determined at both 2.4Å and 1.5Å resolution (Fisher et al., 1995, 1996), but so far only in the absence of substrates. Each subunit adopts the shape of an a/b barrel, and the two barrels are roughly superimposable. The b strands in the core of the superimposed structures overlap very well, while the more peripheral helical and coil elements show more structural divergence. For both subunits the classic eight-stranded barrel topology is interrupted in several locations. For example, helix a4a is located between b4 and a4, and a long coil is located between helix a4 and b5. The most significant disruption in the classic (b/a)8 topology involves a long insertion between b7 and a7. In the b subunit, this insertion comprises about 50 residues and contains a7a, a7b, and b7a, while in the a subunit it contains an additional 29 residues. It is significantly disordered in both reported crystal structures and is located in a protease labile region that is thought to be involved in interactions with the reduced flavin substrate. The heterodimer interface contains both hydrogen bonds and hydrophobic interactions and is dominated by a four-helix bundle motif involving a2 and a3 from both subunits related by a pseudo two-fold rotation axis. An interesting feature of the reaction is its inherent slowness: At 20°C the time required for a single catalytic cycle is about 20 s. The luciferase peroxy flavin itself has a long lifetime; at low temperatures (0 to -20°C) it has been isolated, purified, and characterized (Hastings et al., 1973). It can be further stabilized by long-chain alcohols and amines, which bind at the aldehyde site. However, its crystal structure has not been determined. Two major operons contain genes for the luciferase and other proteins associated with the luminescent system, including enzymes that serve to maintain the supply of myristic aldehyde (Fig. 2.8a; Meighen, 1991). There are also genes that specifically control the development and expression of luminescence. This fascinating mechanism is called “autoinduction” (Nealson et al., 1970), in which the transcription of the luciferase and aldehyde synthesis genes of the lux operon is regulated by genes of the operon itself. A substance produced by the cells called autoinducer (a homoserine lactone; Eberhard et al., 1981; Fig. 2.8b) is a product of the lux I gene. The ecological implications are evident: In planktonic bacteria, a habitat where luminescence has no apparent value, autoinducer cannot accumulate, and no luciferase synthesis occurs (Nealson and Hastings, 1991). However, in the confines of a light organ, high autoinducer levels are reached and

BACTERIA

Figure 2.8. Organization of the lux genes (a) and the homoserine structure of autoinducer (b) in Vibrio fischeri. The operon on the right, transcribed from the 5¢ to the 3¢ end, carries genes for synthesis of autoinducer (lux I), for luciferase a and b peptides (lux A and B), and for aldehyde production (lux C, D, and E). Lux R, transcribed from the operon on the left, codes for a receptor molecule that binds autoinducer, controlling the transcription of the right operon. Other genes, lux F (N), G, and H (right), are associated with the operon but with still uncertain functions. The genes coding for accessory fluorescent proteins Lump and YFP are located to the left.

the luciferase genes are transcribed. Interestingly, it has recently been discovered that an autoinduction-type mechanism, now dubbed quorum sensing (Fuqua et al., 1994), similarly controls the expression of other specific genes in several different groups of bacteria. Bacterial lux genes have been used as reporters in numerous instances (Chatterjee and Meighen, 1995), such as for visualizing gene expression in Streptomyces (Schauer et al., 1988) and following the circadian regulation of transcription in cyanobacteria (Kondo et al., 1994), to name only two. In these cases it is necessary to supply exogenous aldehyde (as a vapor), but the reduced flavin substrate need not be added, since it is generally present in all cells. Bacterial luciferase is also useful in many analytical applications, where flavin or aldehyde, or any enzyme linked to nicotinamide adenine dinucleotide (NAD) or NAD phosphate (NADP), can be assayed (Hastings et al., 1997). Several species and strains of luminous bacteria also contain accessory proteins that, like GFP, serve as secondary emitters. These include both blue- and red-shifted emissions. As with GFP, light is emitted from the luciferase reaction alone (in the absence of a second emitter protein), with the emission peaking at about 490 nm. The blue-shifted emission, due to a lumazine protein (LUMP), peaks at about 475–480 nm; the dissociable chromophore is identified as 6,7-dimethyl-8-ribitylumazine (Small et al., 1980; Petushkov et al., 1995a). A yellow emission peaking at 540 nm in a strain of V. fischeri is due to an analogous yellow fluorescent protein (YFP; recently shown to have homologies with LUMP) in which the chromophore is flavin mononucleotide (FMN) or riboflavin (Hastings et al., 1985; Macheroux et al., 1987; Karatani and Hastings, 1993; Petushkov et al., 1995b). In the YFP system, evidence has been obtained that energy transfer alone cannot account for the yellow emission (Eckstein et al., 1990). In that case a direct population of

27

28

PHOTONS FOR REPORTING MOLECULAR EVENTS

the excited state of the accessory emitter may occur, without the intermediacy of the luciferase-bound excited state.

2.4

DINOFLAGELLATES

Dinoflagellates occur ubiquitously in the oceans as planktonic forms, and they contribute substantially to the bioluminescence commonly seen at night (especially in summer) when the water is disturbed. They occur primarily in surface waters, and many species are photosynthetic. In the phosphorescent bays (e.g., in Puerto Rico and Jamaica), high densities of a single species (Pyrodinium bahamense) usually occur. The so-called red tides are blooms of dinoflagellates, and some of these are bioluminescent. About 6% of all dinoflagellate genera contain luminous species, but since there are no luminous dinoflagellates among the fresh water species, the proportion of luminous forms in the ocean is higher. As a group, dinoflagellates are important as symbionts, notably for contributing photosynthesis and carbon fixation in animals, especially corals. But unlike bacteria, no luminous dinoflagellates are known from symbiotic niches. Bioluminescent flashing is postulated to help reduce predation either by directly diverting predators or by revealing the location of the predators to their predators (Buskey et al., 1983; Hastings and Morin, 1991; Mensinger and Case, 1992; Fleisher and Case, 1995). Luminescence in dinoflagellates is emitted from many small (~0.5 mm) cortical locations. The structures have been identified as novel organelles, termed the scintillons (flashing units). They occur as outpocketings of the cytoplasm into the cell vacuole, like balloons, with their necks remaining connected (Fig. 2.9). Scintillons contain only two major proteins, dinoflagellate luciferase (LCF) and luciferin binding protein (LBP) (Desjardins and Morse, 1993); the latter sequesters luciferin and prevents it from reacting

Figure 2.9. A cartoon depicting scintillons of dinoflagellate, the organelles responsible for flashing light emission. They are formed as outpocketings of the cytoplasm projecting into the acidic vacuole.

DINOFLAGELLATES

with luciferase. Ultrastructurally, these proteins can be identified by immunolabeling (Nicolas et al., 1987, 1991; Fritz et al., 1990) and visualized with image intensification by their bioluminescent flashing following stimulation (Johnson et al., 1985), as well as by the fluorescence of luciferin, the emission spectrum of which is the same as the bioluminescence. Dinoflagellate luciferin is a novel tetrapyrrole related to chlorophyll (Fig. 2.1). Activity (quantum yield, 0.2) can be obtained in extracts made at pH 8 simply by shifting the pH from 8 to 6; it occurs in both soluble and particulate (scintillon) fractions (Fogel and Hastings, 1971, 1972). The existence of activity in both fractions is explained by the rupture of some scintillons during extraction, while others seal off at the neck and form closed vesicles. With the scintillon fraction, the in vitro activity occurs as a flash (~100 ms), very similar to that of the living cell, and the kinetics are independent of the dilution of the suspension. For the soluble fraction, the kinetics depend on dilution, as in enzyme reactions. A distinctive feature of the reaction is that the binding of luciferin to LBP is pHdependent, being bound at pH 8 and free at pH 6. Thus, the flashing of dinoflagellates in vivo is postulated to result from a transient pH change in the scintillons, triggered by an action potential in the vacuolar membrane which, while sweeping over the scintillon, opens ion channels that allow protons from the acidic vacuole to enter (Fig. 2.9). The genes for the two dinoflagellate luminescence proteins have been cloned and sequenced (Lee et al., 1993; Bae and Hastings, 1994; Li et al., 1997); there are no introns in either gene. Both proteins are synthesized and destroyed each day, mediated translationally for LBP by proteins that bind to its mRNA 3¢ untranslated region (Johnson et al., 1984; Morse et al., 1989; Mittag et al., 1994). Both of the cloned genes produce active proteins; when expressed in vitro, LBP exhibits a pH-dependent binding of luciferin while LCF catalyzes the oxidation of luciferin to give light. LCF has an interesting and unusual feature (Li et al., 1997). The approximately 140kDa protein has three tandem repeat domains (~377, 377, and 375 aa long, with no spacer sequences between). Recombinant proteins expressed from the three individual domains of the messenger ribonucleic acid (mRNA) are all separately active as luciferases. This means that in the scintillon, three different sites in the molecule could be concurrently contributing to the activity of the luciferase: a three-ring circus with the same act in all three rings. Although the three dimensional structure of full-length dinoflagellate luciferase has not been determined, crystal structures for two of the three contiguous homologous domains have been solved (Lui et al., 2003; Schultz et al., 2005). The two are very similar, consistent with their high protein sequence identity (Li et al., 1997). The overall size of a domain is about 45 by 50 by 50Å, and each comprises three subdomains, an N-terminal region of 75 residues, a C-terminal 10 stranded b-barrel, and a highly conserved central region forming a barrel that is the most likely location for binding the luciferin and the active site. However, the tetrapyrrole luciferin would be unable to fit within the barrel in its geometry at pH 8. There is strong evidence that the barrel shape is altered at low pH because of changes in the interaction between the N-terminal subdomain and the helixturn-helix region that resides above the barrel. Genes for LBP and LCF have no homologies or similarities with other luciferases or other sequences in any of the data bases. This distinctiveness is consistent with the hypothesis that luciferases have arisen independently in evolution. However, the 5¢ ends of both genes are about 50% homologous over a 90-nt region; for the luciferase, this constitutes the entire remainder of its sequence outside the three repeat regions. Both proteins bind luciferin, but since this region is not needed for luciferase activity, it must have some other

29

30

PHOTONS FOR REPORTING MOLECULAR EVENTS

function. It might be a sequence for targeting the proteins to the vacuolar membrane in the formation of scintillons.

2.5

FIREFLIES AND OTHER INSECTS

Out of a total of approximately 75,000 insect genera, there are only about 100 classed as luminous. But where seen, their luminescence is impressive, most notably in the many species of beetles: the fireflies and their relatives. Fireflies themselves possess ventral light organs on posterior segments; the South American railroad worm, Phrixothrix, has paired green lights on the abdominal segments and red head lights; and the click and fire beetles, Pyrophorini, have both “running lights” (dorsal) and “landing lights” (ventral). From ceiling perches, the dipteran cave glow worms (true flies, not beetles; they occur in New Zealand and Australia) use their light to attract flying prey, which are then entrapped. In fireflies, communication in courtship is the major function of luminescence in fireflies; one sex emits a flash as a signal, to which the other responds, usually in a speciesspecific pattern (Lloyd, 1977, 1980; Case, 1984). The time delay between the two may be a signaling feature; for example, it is precisely 2 s in some North American species. But the flashing pattern (e.g., trains distinctive in duration and/or intensity) is also important in some cases, as is the kinetic character of the individual flash (duration; onset and decay kinetics). In some species, flickering occurs within the flashes, sometimes at very high frequencies (~40 Hz). Fireflies in Southeast Asia are particularly noteworthy for their synchronous flashing; congregations of many thousands form in single trees, where the males produce an all-night-long courtship display of synchronous flashing (Buck and Buck, 1976). The adult firefly light organ comprises a series of photocytes arranged in rosettes, positioned radially around a central well, through which run nerves and trachea, the latter carrying oxygen to the cells (Ghiradella, 1977). Within the photocytes, organelles containing luciferase have been identified with peroxisomes on the basis of immunochemical labeling (Hanna et al., 1976). This identification is supported by the presence of a Cterminal peroxisomal signal sequence in luciferase (Conti et al., 1996). Although flashing is initiated by a nerve impulse that travels to the light organ, the nerve terminals in the light organ are not on photocytes but on tracheolar cells, which regulate the supply of oxygen (Case and Strause, 1978), suggesting that these cells control the flash. In support of this theory, there is a strong positive relationship between the flashing ability and the extent of the tracheal supply system in different species. On the other hand, rapid kinetics, complex waveforms, multiple flashes, and high-frequency flickering all seem unlikely to be regulated by a gas in solution. However, although oxygen might diffuse slowly, it reacts very rapidly in this system; the half rise-time of luminescence with the anaerobic enzyme intermediate (luciferase–luciferyl adenylate) is less than 10 ms (Hastings et al., 1953). Also, possibilities alternate to oxygen seem unlikely. The flash is not directly triggered by an action potential, and none of the ions typically gated by membrane potential changes (Na+, K+, and Ca2+) appear to be candidates for controlling firefly luminescence chemistry. The firefly system was the first in which the biochemistry was extensively studied. It had been known since before 1900 that cell-free extracts could continue to emit light for several minutes or hours, and that after the complete decay of the light, emission could be restored by adding a second extract, prepared by using boiling water to extract the cells (cooled before adding). The enzyme luciferase was assumed to be in the first (cold water)

FIREFLIES AND OTHER INSECTS

extract (with all the luciferin substrate being used up during the emission), whereas the enzyme would be denatured by the hot-water extraction, leaving some substrate intact. This test was referred to as the luciferin–luciferase reaction, and it was already in the first part of this century that luciferins and luciferases from the different major groups would not cross react, indicative of their independent evolutionary origins (Harvey, 1952). McElroy (1947) discovered that the addition of adenosine triphosphate (ATP) to an “exhausted” cold-water extract resulted in bioluminescence. This showed that luciferin had not actually been used up in the cold-water extract. But ATP could not be the emitter, since it does not have the appropriate fluorescence. It was thus discovered that firefly luciferin, which is a unique benzothiazole (Fig. 2.1), was still present in large amounts in the “exhausted” cold-water extract, and that it was ATP that was used up, but available in the hot-water extract. With the elucidation of the luciferin structure, ATP was shown to be required to form the luciferyl adenylate intermediate, which with the adenylate as the leaving group then reacts with oxygen to form a cyclic luciferyl peroxy species (Fig. 2.1). This breaks down to yield CO2 and an excited state of the carbonyl product (McElroy and DeLuca, 1978; Wood, 1995). A remarkably high quantum yield of 0.88 was reported (Seliger and McElroy, 1960). In reactions in which luminescence has decreased to a low level (this may continue for days), it was found that emission is greatly increased by coenzyme A (CoA), but the reason for this was obscure. The recent discovery that long-chain acyl-CoA synthetase (EC 6.2.1.3) has homologies with firefly luciferase (EC 1.13.12.7) both explains this observation and indicates the evolutionary origin of the gene (Wood, 1995). Firefly luciferase has been cloned and expressed in other organisms, including Escherichia coli and tobacco, and its crystal structure has recently been determined (Conti et al., 1996; Franks et al., 1998). It comprises a large N-terminal domain and a smaller Cterminal domain linked to the former by a flexible, four-residue coil. The N-terminal domain contains three distinct regions, two b-sheets and one b-barrel. The b-sheets are flanked on each side by helices and are related by a dyad axis of pseudo-symmetry creating a region of ababa topology within this domain. These two b-sheets flank the third structural element within the N-terminal domain, a distorted antiparallel b-barrel made of 8 b-strands. The N-terminus serves as a “cap” to this large domain and contains a small a+b motif. There is a cleft exposed to water between the N- and C-terminal distinct domains. The residues that are most conserved among all beetle luciferases and the other ATPactivating enzymes are located on the surfaces facing this cleft and on the coil connecting the domains. As a result, early analysis suggested that the active site would be located in this region. However, the cleft is too wide to allow both surfaces to interact with luciferin simultaneously, suggesting that a conformation change occurs on binding ATP or adenylate. A second hypothesis, based on structures of firefly luciferase with bromoform bound, places the active site deeper inside the enzyme somewhat away from the cleft. Unfortunately, firely luciferase has not yet been solved with luciferin bound, leaving somewhat open the questions about active site location as well as concerning domain closure during the bioluminescence reaction. To visualize expression, luciferin must be added exogenously; tobacco “lights up” when the roots are dipped in luciferin (Ow et al., 1986; see color Fig. 2.10). Luciferase catalyzes both the luciferin activation with ATP and the subsequent steps leading to the excited product. There are some beetles in which the light from different organs is a different color, and there is additional color variation between individuals of the same species.

31

32

PHOTONS FOR REPORTING MOLECULAR EVENTS

Figure 2.10. Transgenic tobacco plant carrying the firefly luciferase gene photographed by its own light. The continuous luminescence occurs following the uptake of luciferin by the roots. [From Ow et al. (1986).] See color insert.

Figure 2.11. Bacterial colonies carrying four different beetle luciferase genes cloned from the ventral organ, distinguished by their different luminescence colors: green, yellow-green, yellow and orange (Wood et al., 1989). See color insert.

In Pyrophorus plagiophthalamus, the same ATP-dependent luciferase reaction with the same luciferin occurs in the different organs, but no accessory emitter proteins have been implicated in any of these cases. Instead, differences in the luciferases appear to responsible. Different (but closely homologous) genes from a single organism have been cloned in E. coli and shown to fall into four color classes (see color Fig. 2.11; Wood et al., 1989;

REFERENCES

Wood, 1995). The chemical basis for the color differences remains to be elucidated (McCapra, 1997). Firefly luciferase has been extensively used in analytical applications for the measurement of ATP (Brolin and Wettermark, 1992; Hastings et al., 1997). The cloned gene has also been used as a reporter gene in a number of studies, most recently to monitor circadian regulation of transcription of genes in higher plants (Millar et al., 1995). The use of the genes that result in different colors of luminescence has also been explored. More recently, the simultaneous use of two different luminous systems, for example, firefly and Renilla, for assays of two different substances, has been reported (Sherf et al., 1997).

2.6

CONCLUSION

Bioluminescence occurs in many different species in phylogenetically diverse groups. Among the different groups, the type and method of display of the light, its color, and its function may be very different. In two groups (and only two), bacteria and cnidaria, some of the luminous species possess accessory proteins carrying chromophores, which may serve as secondary emitters and shift the spectrum of the light. The diversity of luminous organisms is indicative of what has been firmly established over the past several decades: Bioluminescent systems in different major groups are not evolutionarily conserved, so that the genes coding for the proteins (e.g., luciferases and accessory proteins) are not homologous. The consequent biochemical diversity offers a marvelous menu for many different specific analytical and reporter applications (Hastings et al., 1997), featuring noninvasive reporting by light emission, as exemplified by GFP.

REFERENCES Bae, Y.-M., and Hastings, J. W. (1994). Cloning, sequencing and expression of dinoflagellate luciferase DNA from a marine alga, Gonyaulax polyedra. Biochem. Biophys. Acta 121:449–456. Baldwin, T., and Ziegler, M. (1992). The biochemistry and molecular biology of bacterial bioluminescence. In Chemistry and Biochemistry of Flavoenzymes, Vol. III, Müller, F. Ed., CRC Press, Boca Raton, FL, pp. 467–530. Blinks, J. R., Wier, W. G., Hess, P., and Prendergast, F. G. (1982). Measurement of C concentrations in living cells. Prog. Biophys. Mol. Biol. 40:1–114. Brehm, P., Lechleiter, J., Smith, S., and Dunlap, K. (1989). Intercellular signaling as visualized by endogenous calcium-dependent bioluminescence. Neuron 3:191–198. Brolin, S., and Wettermark, G. (1992). Bioluminescence Analysis, Vol. 71, VCH Publishers, New York, pp. 176–182. Buck, J. B., and Buck, E. (1976). Synchronous fireflies. Sci. Am. 234:74–85. Buskey, E. J., Mills, L., and Swift, E. (1983). The effects of dinoflagellate bioluminescence on the swimming behavior of a marine copepod. Limnol. Oceanogr. 28:575–579. Case, J. (1984). Firefly behavior and vision. In Insect Communication, Lewis, T., Ed. Harcourt, Brace, Jovanovich, London, pp. 195–222. Case, J. F., and Strause, L. G. (1978). Neurally controlled luminescent systems. In Bioluminescence in Action, Herring, P. J., Ed., Academic Press, New York, pp. 331–366. 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.

33

34

PHOTONS FOR REPORTING MOLECULAR EVENTS

Chatterjee, J., and Meighen, E. A. (1995). Biotechnological applications of bacterial bioluminescence (lux) genes. Photochem. Photobiol. 62:651–650. 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 green-fluorescent protein. Biochemistry 32:1212–1218. Conti, E., Franks, N. T., and Brick, P. (1996) Crystal structure of firefly luciferase throws light on a super family of adenylate-forming enzymes. Structure 4:287–298. Cormier, M. J. (1981). Renilla and Aequorea bioluminescence. In Bioluminescence and Chemiluminescence, DeLuca, M., and McElroy, W. D., Eds., Academic, New York, pp. 225–233. Cormier, M. J., Prasher, D. C., Longiaru, M., and McCann, R. O. (1989). The enzymology and molecular biology of the Ca++-activated photoprotein, aequorin. Photochem. Photobiol. 49:509–512. DeLuca, M., Ed. (1978). Bioluminescence and chemiluminescence. Methods Enzymol. 57:1–653. Desjardins, M., and Morse, D. (1993). The polypeptide components of scintillons, the bioluminescence organelles of the dinoflagellate. Gonyaulax polyedra. Biochem. Cell Biol. 71:176–182. Dunlap, K., Takeda, K., and Brehm, P. (1987). Activation of a calcium-dependent photoprotein by chemical signalling through gap junctions. Nature (London) 325:60–62. Eberhard, A., Burlingame, A. L., Eberhard, C., Kenyon, G. L., Nealson, K. H., and Oppenheimer, N. H. (1981). Structural identification of autoinducer of Photobacterium fischeri Luciferase. Biochemistry 20:2444–2449. Eckstein, J., Cho, K. W., Colepicolo, P., Ghisla, S., Hastings, J. W., and Wilson, T. (1990). A timedependent bacterial bioluminescence emission spectrum in an in vitro single turnover system: Energy transfer alone cannot account for the yellow emission of Vibrio fischeri Y-1. Proc. Natl. Acad. Sci. 87: 1466–1470. Fisher, A. J., Thompson, T. B., Thoden, J. B., Baldwin, T. O., and Rayment, I. (1996). The 1.5-Å resolution crystal structure of bacterial luciferase in low salt conditions. J. Biol. Chem. 271: 21956–21968. Fisher, J., Rauschel, F. M., Baldwin, T. O., and Rayment, I. (1995). The three-dimensional structure of bacterial luciferase from Vibrio harveyi at 2.4 Å resolution. Biochemistry 34:6581–6586. Fleisher, K. J., and Case, J. F. (1995). Cephalopod predation facilitated by dinoflagellate luminescence. Biol. Bull. 189:263–271. Fogel, M., and Hastings, J. W. (1971). A substrate binding protein in the Gonyaulax bioluminescence reaction. Arch. Biochem. Biophys. 142:310–321. Fogel, M., and Hastings, J. W. (1972). Bioluminescence: Mechanism and mode of control of scintillon activity. Proc. Nat. Acad. Sci. 69:690–693. Franks, N., Jenkins, A., Conti, E., Lieb, W., and Brick, P. (1998). Structural basis for the inhibition of firefly luciferase by a general anesthetic. Biophys. J. 75: 2205–2211. Fritz, L., Morse, D., and Hastings, J. W. (1990). The circadian bioluminescence rhythm of Gonyaulax is related to daily variations in the number of light emitting organelles. J. Cell Sci. 95:321–328. Fuqua, W. C., Winans, S. C., and Greenberg, E. P. (1994). Quorum sensing in bacteria: The LuxRLufamily of cell density-responsive transcriptional regulators. J. Bacteriol. 176:269–275. Ghiradella, H. (1977). Fine structure of the tracheoles of the lantern of a photurid firefly. J. Morphol. 153:187–204. Grober, M. S. (1988a). Brittle-star bioluminescence functions as an aposematic signal to deter crustacean predators. Anim. Behav. 36:493–501. Grober, M. S. (1988b). Responses of tropical reef fauna to brittle-star luminescence. J. Exp. Mar. Biol. Ecol. 115:157–168. Hanna, C. H., Hopkins, T. A., and Buck, J. (1976). Peroxisomes of the firefly lantern. J. Ultrastruct. Res. 57:150–162.

REFERENCES

Harvey, E. N. (1952). Bioluminescence. Academic, New York. Hastings, J. W. (1983). Biological diversity, chemical mechanisms and evolutionary origins of bioluminescent systems. J. Molec. Evol. 19:309–321. Hastings, J. W., and Gibson, Q. H. (1963). Intermediates in the bioluminescent oxidation of reduced flavin mononucleotide: J. Biol. Chem. 238:2537–2554. Hastings, J. W., and Morin, J. G. (1969a). Calcium-triggered light emission in Renilla. A unitary biochemical scheme for coelenterate bioluminescence. Biochem. Biophys. Res. Commun. 37:493–498. Hastings, J. W., and Morin, J. G. (1969b). Comparative biochemistry of calcium activated photoproteins from the ctenophore, Mnemiopsis and the coelenterates Aequorea, Obelia, Pelagia and Renilla. Biol. Bull. 137:402. Hastings, J. W., and Morin, J. G. (1971). Kinetics of calcium-triggered luminescence of aequorin and other photoproteins. In Contractility of Muscle Cells and Related Processes, Podolsky, R. J., Ed., Prentice-Hall, Engelwood Cliffs, NJ, pp. 99–104. Hastings, J. W., and Morin, J. G. (1991). Bioluminescence. In Neural and Integrative Animal Physiology, Prosser, C. L., Ed., Wiley-Interscience, New York, pp. 131–170. Hastings, J. W., McElroy, W. D., and Coulombre, J. (1953). The effect of oxygen upon the immobilization reaction in firefly luminescence. J. Cell Comp. Physiol. 42:137–150. Hastings, J. W., Potrikus, C. J., Gupta, S., Kurfürst, M., and Makemson, J. C. (1985). Biochemistry and physiology of bioluminescent bacteria. Adv. Microbial. Physiol. 26:235–291. Hastings, J. W., Kricka, L., and Stanley, P., Eds. (1997). Bioluminescence and Chemiluminescence: Molecular Reporting by Photons. Wiley, Chichester. Head, J. F., Inouye, S., Teranishi, K., and Shimomura, O. (2000). The crystal structure of the phosphoprotein aequorin at 2.3Å resolution. Nature 405: 372–376. Heim, R., Prasher, D. C., and Tsien, R. Y. (1994). Wavelength mutations and posttranslational autoxidation of green fluorescent protein. Proc. Natl. Acad. Sci. USA 9:12501–12504. Herring, P. J., Ed. (1978). Bioluminescence in Action. Academic, New York, 570 pp. Inouye, S., Sakaki, Y., Goto, T., and Tsuji, F. I. (1986). Expression of apoaequorin complementary DNA in Escherichia coli. Biochemistry 25:8425–8429. Inouye, S., Aoyama, S., Miyata, T., Tsuji, F., and Sakaki, Y. (1989). Overexpression and purification of the recombinant Ca2+-binding protein, Apoaequorin, J. Biochem. 105:473–477. Johnson, C. H., Roeber, J., and Hastings, J. W. (1984). Circadian changes in enzyme concentration account for rhythm of enzyme activity in Gonyaulax. Science 223:1428–1430. Johnson, C. H., Inoué, S., Flint, A., and Hastings, J. W. (1985). Compartmentation of algal bioluminescence: Autofluorescence of bioluminescent particles in the dinoflagellate Gonyaulax as studied with the image-intensified video microscopy and flow cytometry. J. Cell Biol. 100:1435–1446. Johnson, C. H., Knight, M. R., Kondo, T., Masson, P., Sedbrook, J., Haley, A., and Trewavas, A. (1995). Circadian oscillations of cytosolic and chloroplastic free calcium in plants. Science 269:1863–1865. Johnson, F. H., Shimomura, O., Saiga, Y., Gershman, G., Reynolds, G. T., and Waters, J. R. (1962). Quantum efficiency of Cypridina luminescence with a note on that of Aequorea. J. Cell. Comp. Physiol. 60:85–103. Karatani, H., and Hastings, J. W. (1993). Two active forms of the accessary yellow fluorescence protein of the luminous bacterium Vibrio fischeri strain Y1. J. Photochem. Photobiol. B. 18:227–232. Kondo, T., Tsinoremas, N., Golden, S., Johnson, C., Kutsuna, S., and Ishiura, M. (1994). Circadian clock mutants of cyanobacteria. Science 266:1233–1236.

35

36

PHOTONS FOR REPORTING MOLECULAR EVENTS

Lee, D.-H., Mittag, M., Sczekan, S., Morse, D., and Hastings, J. W. (1993). Molecular cloning and genomic organization of a gene for luciferin-binding protein from the dinoflagellate Gonyaulax polyedra. J. Biol. Chem. 268:8842–8850. Li, L., Hong, R., and Hastings, J. W. (1997). Three functional luciferase domains in a single polypeptide chain. Proc. Natl. Acad. Sci. USA 94:8954–8958. Liu, L., Im, H., Cegielski, M., LeMagueres, P., Schultz, L. W., Krause, K. L., and Hastings, J. W. (2003). Characterization and crystallization of active domains of a novel luciferase from a marine dinoflagellate. Acta Crystallogr. D 59: 761–764. Liu, Z.-J., Vysotski, E. S., Chen, C.-J., Rose, J. P., Lee, J., and Wang, B.-C. (2000). Structure of the Ca2+-regulated photoprotein obelin at 1.7Å resolution determined directly from its sulfur substructure. Protein Sci. 9: 1–9. Lloyd, J. E. (1977). Bioluminescence and communication. In How Animals Communicate, Sebeok, T. A., Ed., Indiana University Press, Bloomington, IN, pp. 164–183. Lloyd, J. E. (1980). Male Photuris fireflies mimic sexual signals of their females’ prey. Science 210:669–671. Lorenz, W. W., McCann, R. O., Longiaru, M., and Cormier, M. J. (1991). Isolation and expression of cDNA encoding Renilla reniformis luciferase. Proc. Natll. Acad. Sci. USA 88:4438–4442. Macheroux, P., Steinerstauch, P., Ghisla, S., Buntic, R., Colepiciolo, P., and Hastings, J. W. (1987). Purification of the yellow fluorescent protein from V. fischeri and identity of the flavin chromophore. Biochem. Biophys. Res. Commun. 146:101–106. McCapra, F. (1997). Mechanisms in chemiluminescence and bioluminescence—unfinished business. In Hastings, J. W., Kricka, L. J., and Stanley, P. E., Ed., Bioluminescence and Chemiluminiscence, Wiley, Chichester, pp. 7–15. McElroy, W. D. (1947). The energy source for bioluminescence in an isolated system. Proc. Natl. Acad. Sci. USA 33:342–346. McElroy, W. D., and DeLuca, M. (1978). Chemistry of firefly luminescence. In Herring, P. J., Ed., Bioluminescence in Action, Academic, New York, pp. 109–128. Meighen, E. A. (1991). Molecular biology bacterial bioluminescence. Microbiol. Rev. 55:123–142. Mensinger, A. F., and Case, J. F. (1992). Dinoflagellate luminescence increases susceptibility of zooplankton to teleost predation. Mar. Biol. 112:207–210. Millar, A. J., Carré, I. A., Strayer, C. A., Chua, N.-H., and Kay, S. A. (1995). Circadian clock mutants in Arabidopsis identified by luciferase imaging. Science 267:1161–1166. Mittag, M., Lee, D.-H., and Hastings, J. W. (1994). Circadian expression of the luciferin-binding protein correlates with the binding of a protein to its 3¢ untranslated region. Proc. Natl. Acad. Sci. USA 91:5257–5261. Morin, J. G. (1974). Coelenterate bioluminescence. In Coelenterate Biology: Reviews and New Perspectives, Muscatine, L., and Lenhoff, H., Eds., Academic, New York, pp. 397–438. Morin, J. G. (1976). Probable functions of bioluminescence in the Pennatulacea (Cnidaria, Anthozoa). In Coelenterate Ecology and Behavior, Mackie, G. O., Ed., Plenum, New York, pp. 629–638. Morin, J. G. (1983). Coastal bioluminescence: Patterns and functions. Bull. Mar. Sci. 33:787–817. Morin, J. G., and Cooke, I. M. (1971a). Behavioural physiology of the colonial hydroid Obelia. II. Stimulus-initiated electrical activity and bioluminescence. J. Exp. Biol. 54:707–721. Morin, J. G., and Cooke, I. M. (1971b). Behavioural physiology of the colonial hydroid Obelia. III. Characteristics of the bioluminescent system. J. Exp. Biol. 54:723–735. Morin, J. G., and Hastings, J. W. (1971a). Biochemistry of the bioluminescence of colonial hydroids and other coelenterates. J. Cell Physiol. 77:305–311. Morin, J. G., and Hastings, J. W. (1971b). Energy transfer in a bioluminescent system. J. Cell Physiol. 77:313–318.

REFERENCES

Morin, J. G., and Reynolds, G. T. (1969). Fluorescence and time distribution of photon emission of bioluminescent photocytes in Obelia geniculata. Biol. Bull. 137:410. Morin, J. G., and Reynolds, G. T. (1970). Luminescence and related fluorescence in coelenterates. Biol. Bull. 139:430–431. Morin, J. G., and Reynolds, G. T. (1974). The cellular origin of bioluminescence in the colonial hydroid Obelia. Biol. Bull. 147:397–410. Morin, J. G., Reynolds, G. T., and Hastings, J. W. (1968). Excitatory physiology and localization of bioluminescence in Obelia. Biol. Bull. 135:429–430. Morise, H., Shimomura, O., Johnson, F. H., and Winant, J. (1974). Intermolecular energy transfer in the bioluminescent system. Biochemistry 13:2656–2662. Morse, D., Milos, P. M., Roux, E., and Hastings, J. W. (1989). Circadian regulation of the synthesis of substrate binding protein in the Gonyaulax bioluminescent system involves translational control. Proc. Natl. Acad. Sci. USA 86:172–176. Nealson, K., and Hastings, J. W. (1991). The luminous bacteria. In The Prokaryotes, 2nd ed., Vol. I, Part 2, Chapter 25, Balows, A., Trüper, H. G., Dworkin, M., Harder, W., and Schleifer, K. H., Eds., Springer-Verlag, New York, pp. 625–639. Nealson, K., Platt, T., and Hastings, J. W. (1970). The cellular control of the synthesis and activity of the bacterial luminescent system. J. Bacteriol. 104:313–322. Nicol, J. A. C. (1958). Observations on the luminescence of Pennatula phosphorea, with a note on the luminescence of Virgularia mirabilis. J. Mar. Biol. Assoc. UK 37:551–563. Nicolas, M.-T., Nicolas, G., Johnson, C. H., Bassot, J-M., and Hastings, J. W. (1987). Characterization of the bioluminescent organelles in Gonyaulax polyedra (dinoflagellates) after fast-freeze fixation and antiluciferase immunogold staining. J. Cell Biol. 105:723–735. Nicolas, M.-T., Morse, D., Bassot, J.-M., and Hastings, J. W. (1991). Colocalization of luciferin binding protein and luciferase to the scintillons of Gonyaulax polyedra revealed by immunolabeling after fast-freeze fixation. Protoplasma 160:159–166. Ormö, 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. Ow, D. W., Wood, K. V., DeLuca, M., de Wet, J. R., Helinski, D. R., and Howell, S. H. (1986). Transient and stable expression of the firefly luciferase gene in plant cells and transgenic plants. Science 234:856–859. Petushkov, V. N., Gibson, B. G., and Lee, J. (1995a). Properties of recombinant fluorescent proteins from Photobacterium leiognathi and their interaction with luciferase intermediates. Biochemistry 34:3300–3309. Petushkov, V. N., Gibson, B. G., and Lee, J. (1995b). The yellow bioluminescence bacterium, Vibrio fischeri Y1, contains a bioluminescence active riboflavin protein in addition to the yellow fluorescence FMN protein. Biochem. Biophys. Res. Commun. 211:774–779. Ridgeway, E. B., and Ashley, C. C. (1968). Simultaneous recording of membrane potential, calcium transient and tension in single muscle fibres. Nature (London) 219:1168–1169. Ruby, E. G. (1996). Lessons from a cooperative, bacterial-animal association: the Vibrio fischeriEuprymna scolopes light organ symbiosis. Annu. Rev. Microbiol. 50:591–624. Saran, S., Nakao, H., Tasaki, M., Iida, H., Tsuji, F. I., Nanjundiah, V., and Takeuchi, I. (1994). Intracellular free calcium level and its response to cAMP stimulation in developing Dictyostelium cells transformed with jellyfish apoaequorin cDNA. FEBS Lett. 337:43–47. Schauer, A., Ranes, M., Santamaria, R., Guijarro, J., Lawlor, E., Mendez, C., Chater, K., and Losick, R. (1988). Gene expression in the filamentous bacterium Streptomyces coelicolor. Science 240:768–772. Schultz, L. W., Liu, L., Cegielski, M., and Hastings, J. W. (2005). Crystal structure of a pHregulated luciferase catalyzing the bioluminescent oxidation of an open tetrapyrrole. Proc. Natl. Acad. Sci. USA 102:1378–1383.

37

38

PHOTONS FOR REPORTING MOLECULAR EVENTS

Seliger, H. H., and McElroy, W. D. (1960). Spectral emission and quantum yield of firefly bioluminescence. Arch. Biochem. Biophys. 88:136. Sherf, B., Navarro, S., Hannah, R., and Wood, K. V. (1997). Co-reporter technology integrating firefly and Renilla luciferase assays. In Hastings, J. W., Kricka, L. J., and Stanley, P. E., Eds., Co-reporter Technology Integrating Firefly and Renilla Luciferase Assays, Wiley, Chichester, pp. 228–231. Shimomura, O., and Johnson, F. H. (1975). Regeneration of the photoprotein aequorin. Nature (London) 256:236–238. 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. 59:223–240. Shorey, J., and Morin, J. G. (1974). Quantification of light produced from hydrozoan photocytes. Biol. Bull. 147:499. Small, E. D., Koka, P., and Lee, J. (1980). Lumazine protein from the bioluminescent bacterium Photobacterium phosphoreum. J. Biol. Chem. 255:8804–8810. Tanahashi, H., Ito, T., Inouye, S., Tsuji, F. I., and Sakaki, Y. (1990). Photoprotein aequorin: use as a reporter enzyme in studying gene expression in mammalian cells. Gene 96:249–255. Thomson, C. M., Herring, P. J., and Campbell, A. K. (1997). The widespread occurrence and tissue distribution of the imidazolopyrazine luciferins. J. Biolum. Chemilum. 12:87–91. Titschack, H. (1964). Untersuchungen Über das Leuchten der Seefeder Veretillum cynomorium (Pallas). Vie Milieu 15:547–563. 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 green-fluorescent proteins. Photochem. Photobiol. 31:611–615. Wilson, T., and Hastings, J. W. (1998). Bioluminescence. Annu. Rev. Cell Dev. Biol. 14:197–230. Wood, K. V. (1995). The chemical mechanism and evolutionary development of beetle bioluminescence. Photochem. Photobiol. 62:662–673. Wood, K. V., Lam, Y. A., Seliger, H. H., and McElroy, W. D. (1989). Complementary DNAs encoding click beetle luciferases can elicit bioluminescence of different colors. Science 244:700–702. Yang, F., Moss, L. G., and Phillips Jr., G. N. (1996). The molecular structure of green fluorescent protein. Nat. Biotechn. 14:1246–1251.

3 BIOCHEMICAL AND PHYSICAL PROPERTIES OF GREEN FLUORESCENT PROTEIN William W. Ward Department of Biochemistry and Microbiology, Rutgers University, Cook College, New Brunswick, NJ

3.1

INTRODUCTION

The recent popularity of green fluorescent protein (GFP) as a research tool in cellular and developmental biology (Chalfie, 1995; Hassler, 1995; Kain et al., 1995; Prasher, 1995; Stearns, 1995) requires that we look very carefully at the chemical and physical properties of the GFP molecule and its chromophore. Unfortunately, the chemical and physical characterizations of native and recombinant forms of GFP and numerous mutants of the original Aequorea victoria derived clone (Chalfie et al., 1994) have not, and cannot, keep pace with the proliferation of GFP mutants and the accelerating pace in GFP applications. After 30 years of research on the prototype native GFP molecules from the jellyfish, Aequorea victoria (Morin and Hastings, 1971; Morise et al., 1974; Prendergast and Mann, 1978), and the sea pansy, Renilla reniformis (Wampler et al, 1971, 1973; Morin, 1974; Cormier et al., 1974; Ward, 1979; Ward and Cormier, 1979), these proteins are still incompletely characterized; much less is known about the chemical and physical properties of the available mutants of GFP. In this chapter, the known chemical and physical properties of GFP are summarized to provide a sound basis for the qualitative and quantitative interpretations of data generated in its applications. Nonetheless, because so much remains unknown about the biochemical properties of these molecules, users of GFP should cautiously interpret their GFP derived data.

Green Fluorescent Protein: Properties, Applications, and Protocols, Second Edition Edited by Martin Chalfie and Steven R. Kain Copyright © 2006 John Wiley & Sons, Inc.

39

40

BIOCHEMICAL AND PHYSICAL PROPERTIES OF GREEN FLUORESCENT PROTEIN

3.2

BIOLOGICAL FUNCTION OF GREEN FLUORESCENT PROTEIN

Biologically, GFP acts to shift the color of bioluminescence from blue to green in luminous coelenterates (jellyfish, hydroids, sea pansies, and sea pens) and to increase the quantum yield of light emission (Ward, 1979). All coelenterates utilize the same luciferin (coelenterate-type luciferin or coelenterazine) in their bioluminescence reactions (Cormier et al., 1973; Hori and Cormier, 1973; Hori et al., 1973; Wampler et al., 1973; Ward and Cormier, 1975; Inoue et al., 1977a,b; Shimomura and Johnson, 1979; Shimomura et al., 1980), producing a protein-bound oxyluciferin (Hori et al., 1973, 1975, 1977) that emits blue light in the absence of GFP. In the presence of GFP, however, the emitted light is green and identical in spectral properties to the fluorescence emission spectrum of GFP (Ward, 1979) when excited directly by exogenous radiation. Such spectral shifts are known to occur in spectroscopy by one of two general mechanisms: (1) radiative (trivial) energy transfer in which the donor molecule emits light that is subsequently absorbed and reemitted by the acceptor and (2) radiationless (often called Förster type) energy transfer in which excitation energy is transferred, without photon emission, from the donor molecule to the acceptor (Ward, 1979). Efficient trivial transfer from blue-emitting oxyluciferin to GFP requires a relatively high concentration of GFP and a sufficiently long pathlength. In a 1-cm pathlength fluorometric cuvette, for example, 90% of the incident blue light could be absorbed by wild-type GFP at a concentration of 5–10 mg ml-1 (absorbance = 1.0 at 480 nm). But, with a very short pathlength as would be seen in animal cells (10 mm diameter), the GFP concentration would need to be 1000¥ as great for trivial transfer to operate efficiently. Clearly, intracellular protein concentrations on the order of 10,000 mg ml-1 cannot be achieved. Even if all oxyluciferin emission were absorbed by GFP, the quantum yield in a trivial transfer system can be no greater than the product of the quantum yields of donor and acceptor. In the coelenterate systems, the bioluminescence quantum yield for oxyluciferin is about 0.10 (Hori et al., 1973; Hart et al., 1979) and the fluorescence quantum yield for GFP is about 0.80 (Ward and Cormier, 1979). Thus the maximum overall quantum yield, by a trivial mechanism, would be 0.10 ¥ 0.80 = 0.08, a slight decrease from oxyluciferin emission alone. But, a radiationless system traps excitation energy directly by resonance transfer, so long as donor and acceptor molecules are relatively close to each other (4 hours) limits its usefulness in cell biological experiments. Similarly, attempts have been made (Richmond

107

108

MOLECULAR BIOLOGY AND MUTATION OF GREEN FLUORESCENT PROTEIN

Figure 5.6. A backbone representation of the three-dimensional structure of GFP (1EMG) (Elsliger et al., 1998). The residues where circular permutations are permitted while retaining fluorescence are color highlighted. E142, hot pink; Y143, gray; Y145, dark blue; H148, fuchsia; D155, yellow; H169, red; E172, light blue; D173, orange; A227, Cyan; I229, light purple. These residues represent sites where the main chain can be interrupted. In most cases, resumption of GFP sequence can occur one to four residues following the initial interruption. See color insert.

et al., 2000) to rationally design sites, on the external surface of the barrel (Fig. 5.5A) of YFP (10C) that could bind to metals. Mutants on the 10C background (S147H/Q204H and S202D or F223E) were generated that successfully quenched YFP fluorescence when exposed to various divalent cations, but nothing was reported of the selectivity concerning the ability of these mutants to discriminate among the cations. It would be ideal if robust, selective sensors of such simplicity could be generated for cellular analytes. Similar to the case where Zn2+ sites were engineered directly into the chromophore, individual GFPs with only small modifications can be made into sensors for pH and halides (Fig. 5.5A).

5.12

pH

High pH, 11–12, causes a relative redistribution of the two absorbance/excitation maxima of wtGPF toward the longer wavelength (470 nm) peak while low pH (pKa ~ 5.5) causes a quenching of fluorescence (Bokman and Ward, 1981; Ward et al., 1982). A fluorophore with sensitivity to variations in pH is often viewed as a lemon, but in the case of GFPs, the lemon has been made into some very informative lemonade. While it is good to have a fluorescent probe that emits stably over a broad range of pHs, it was also apparent that the pH-dependent change in fluorescence behavior could be exploited to measure pH in

109

HALIDES

TABLE 5.6. pka Values of Gfps Used as ph Indicators FP

pKa

Mutations

Reference

GFP YFP GFP GFP GFP GFP CFP

4.8 5.7 5.9 6.0 6.1 6.15 6.4

(Kneen et al., 1998) (Griesbeck et al., 2001) (Kneen et al., 1998) (Kneen et al., 1998) (Kneen et al., 1998) (Llopis et al., 1998) (Llopis et al., 1998)

YFP GFP GFP GFP

7.1 7.3 8.0 8.0

T203I Q69M S65 F64L, S65T Y66H F64L, S65T, H231L K26R, F64L, S65T, Y66W, N146I, M153T, V163A, N164H, H231L S65G, S72A, T203Y, H231L S65T/C48S/H148C/T203C S65T, H148G, T203C H148G

(Llopis et al., 1998) (Hanson et al., 2002) (Hanson et al., 2002) (Maysuyama et al., 2000)

FP indicates the spectral mutant upon which the indicated mutations were incorporated. pKa is the pH value at which 50% of the molecules are fluorescent.

subcellular domains (Llopis et al., 1998). pKa values range for GFPs vary widely covering the range of most cellular pHs (Table 5.6). pHluorins comprise a set of pH-sensitive GFPs used to monitor exocytosis (Miesenbock et al., 1998; Sankaranarayanan et al., 2000). These sensors are targeted to the luminal side of secretory vesicles where the pH is below the pKa of pHluorin (which causes them to be nonfluorescent). Upon fusion, exocytosis, and exposure of the membrane-associated pHluorin to the extracellular milieu held at a desired physiological pH, the pHluorin becomes fluorescent (ecliptic pHluorin) or shifts its excitation maximum from 395 nm to 475 nm (ratiometric pHluorin). The broad range of pKa values for the fluorophores is generated by the diversity mutations in and around the fluorophore. Many of the physical reasons for the various pH-sensitive behaviors are summarized in a theoretical study (Scharnagl et al., 1999) that incorporated a broad range of existing physical data from many experimental sources. In YFP (class 4 chromophores) the mutation Q69M (named “Citrine”) retains virtually identical excitation and emission spectra but lowers the pKa of the chromophore to 5.7, renders it insensitive to chloride, increases the photostability over previous versions of YFP by about twofold, and improves expression at 37°C in cells (Griesbeck et al., 2001). In the crystal structure of Citrine, the Met at position 69 is well-ordered, tightly packed into the cavity, and unlikely to be able to undergo the same sort of conformational change that is seen with the apo- and iodidebound forms of EYFP.

5.13

HALIDES

Wachter and Remington (1999) first reported halide sensitivity of a YFP (S65G/V68L/ S72A/T302Y and H148Q or H148G). Shortly thereafter, they and Verkman’s group made more detailed biophysical characterization of the nature of the Cl- sensitivity of YFP (T203Y/S65G/V68L/S72A/H148Q) and made the first steps toward developing it as the first genetically encoded Cl- sensor (Jayaraman et al., 2000). Almost simultaneously, Kuner and Augustine (2000) discovered the Cl- sensitivity of YFP while using CFP (K26R/F64L/S65T/Y66W/N146I/M153T/V163A/N164H/H231L) and YFP (S65G/S72A/ K79R/T203Y/H231L) to study protein–protein interactions by FRET. They fused CFP and

110

MOLECULAR BIOLOGY AND MUTATION OF GREEN FLUORESCENT PROTEIN

YFP with a short intervening peptide linker (ENLYFQG) to make a protein that, in the resting state, had a high degree of FRET, and they called it “Clomeleon.” When Cl- bound to YFP, quenching its fluorophore and rendering it a nonfunctional FRET acceptor, CFP subsequently became dequenched or brighter. This ratiometric sensor for Cl- was able to measure small changes in Cl- at physiological concentrations. The crystal structure of YFP H148Q has been solved (Wachter et al., 2000), and the selecitive halide-binding site was found in a small, amphiphilic, buried cavity adjacent to R96. The halide ion was found to be hydrogen-bonded to the phenol group of tyrosine at position 203, illustrating why this mutation is critical to the formation of a halide-sensitive GFP (Wachter et al., 2000). Conversely, YFPs have been rendered virtually insensitive to the effects of halide binding by the mutation Q69M (Griesbeck et al., 2001) or F64L/M153T/V163A/S175G (“Venus”) (Nagai et al., 2002). In the first case, the bulkier side chain of Met fills the cavity into which chloride ion could reside with Q69, and it is conformationally stable. Even if a conformational change in the thioether side chain of Met were permitted in this space, it is unlikely to contribute to halide binding because it is incapable of hydrogen bonding in the same manner as the carboxamide nitrogen of a Gln side chain (Griesbeck et al., 2001). In the second case, F64L induced large conformational changes in the molecule, leading to the removal of halide sensitivity by preventing ion access to the binding site (Rekas et al., 2002). The “Venus” variant of GFP is also very insensitive to changes in pH.

5.14

INSERTION OF GFP INTO OTHER PROTEINS

Insertion of GFPs into other proteins is another important twist on this theme. An early example was the insertion of GFP [GFPDC; Chalfie et al. (1994)] into a nonconducting mutant of the Shaker K+ channel (Siegel and Isacoff, 1997) and subsequent, improved generations thereof (Guerrero et al., 2002). This first version of the fusion was able to monitor changes in membrane voltage with a maximal fractional fluorescence change of 5.1%. Similarly, wtGFP was inserted into an intracellular loop of a reversibly nonconducting form of the rat mu I skeletal muscle voltage-gated sodium channel (Ataka and Pieribone, 2002). The resulting protein called SPARC (sodium channel protein-based activity reporting construct) can report depolarizing pulses as short as 2 ms and does not inactivate during prolonged depolarizations, but the size of its optical response is very small.

5.15

TANDEM CONCATENATIONS OF TWO GFPS

The availability of GFP mutants of different colors, UV-excited blue emitters and blueexcited green emitters, enables fluorescence resonance energy transfer (FRET) from one to the other. FRET is strongly dependent on the angular orientation and distance of the fluorophores from one another, falling off steeply as the distance exceeds the Förster distance R0 at which FRET is 50% efficient (Tsien et al., 1993; Lakowicz, 1999). For the blue emitter P4-3, containing the point mutations Y66H and Y145F, donating energy either to S65T or S65C, R0 is calculated to be about 40 Å (Heim and Tsien, 1996), assuming that the mutual orientation of the fluorophores is random or freely tumbling. The larger the R0, the better; this is because GFP is a cylindrical structure of about 12-Å radius and 42-Å length (Ormo et al., 1996), and so much of R0 is used up simply within the two GFPs. A systematic study determined the Förster distances between all homo and hetero

SILENT AND LOSS-OF-FUNCTION MUTATIONS

pairings of BFP, CFP, GFP, YFP, and DsRed [Patterson et al., 2000; see also Wu (1994)] The maximum R0 measured for any pair was 56.4 Å between EGFP and EYFP. The Förster distance between CFP and YFP is 49.2 Å. This aspect, combined with greater distance between the peaks of excitation and a favorable overlap integral (J), makes CFP and YFP the pair used most commonly in FRET studies. Sensors that track the activity of proteases were one of the earliest applications for which FRET between concatenated GFPs was exploited (Xu et al., 1998 Heim, 1999; Harpur et al., 2001; Luo et al., 2001; Tawa et al., 2001) and reviewed (Jones et al., 2000). Sensors for caspase 3 have been used successfully as reporters in high-throughput drug discovery programs (Tawa et al., 2001). The first examples of these constructs included the concatenation of the genes encoding S65C or S65T (GFPs) and P4-3 (BFP) with an intervening 25-residue linker connecting the two GFP-derived domains (Heim and Tsien, 1996). Likewise, BFP5 and RSGFP4 have been fused with a 20-residue linker sensitive to factor Xa (Mitra et al., 1996). In either case, before protease cleavage, UV excitation gives rise to some blue emission but also substantial green emission due to FRET from the blue- to the green-emitting domain. After protease cleavage to separate the two domains, FRET is abolished, the blue emission is increased, and the green emission is nearly abolished. For the S65C:P4-3 construct, the ratio between blue and green emission intensities increased by a factor of 4.6 upon cleavage (Heim and Tsien, 1996), while the RSGFP4::BFP5 fusion showed about a 1.9-fold increase in emission ratio (Mitra et al., 1996). In the case of the S65C/P4-3 construct, the large change in ratio between blue and green emissions was shown to result from separation of the two fluorophores rather than from an effect on either one separately, because control experiments with the two separate proteins showed no spectral sensitivity to protease under matching conditions. Another concatenated, intramolecular FRET-based reporter that has been used broadly is the calcium sensor CaMeleon (Miyawaki et al., 1997). In this sensor, CFP and YFP flank calmodulin and a Ca2+-calmodulin-binding peptide, M13 from myosin light-chain kinase. Wheng CaMeleon encounters a change in Ca2+, M13 and calmodulin respond either by associating in increased [Ca2+] or dissociating in decreased [Ca2+]. The conformational change that occurs causes a change in the FRET efficiency (which is low at lower [Ca2+] and high at higher [Ca2+]) largely via a change in kappa squared or the intermolecular angle of fluorophore orientation (Atsushi Miyawaki, personal communication). This probe has been especially useful in reporting calcium changes in subcellular domains like the nucleus and endoplasmic reticulum (Arnaudeau et al., 2002; Demaurex and Frieden, 2003; Malli et al., 2003; Palme et al., 2004) and caveolae (Isshiki et al., 2002), as well as in Drosophila (Reiff et al., 2002; Liu et al., 2003) and C. elegans (Kerr et al., 2000). The most recent versions of CaMeleon incorporate YFPs (Citrine) that are insensitive to pH and Halides (Griesbeck et al., 2001). Most recently, conformationally sensitive concatenations of GFP have been used to track the activity of kinases (Ting et al., 2001; Zhang et al., 2001b; Violin et al., 2003), elegantly showing links between activity of these kinases and other signal transduction pathways.

5.16

SILENT AND LOSS-OF-FUNCTION MUTATIONS

During any random or semirandom mutagenesis screen, the great majority of colonies typically are either indistinguishable from the starting phenotype or of significantly reduced brightness. In principle, these mutants could be sequenced to provide a list of neutral or deleterious substitutions, but such a list would be laborious to collect and of negligible

111

112

MOLECULAR BIOLOGY AND MUTATION OF GREEN FLUORESCENT PROTEIN

interest to those wishing to improve GFP or obtain novel properties. Perhaps this list deserves compilation, because it might increase understanding of how GFP folds and builds its chromophore. A few mutations are known to be neutral, such as the ubiquitous Q80R, which may have arisen from a PCR error in the initially distributed cDNA clone (Chalfie et al., 1994). The substitutions within the natural isoforms are presumably all permissive for fluorescence, although other properties may well be altered. As mentioned earlier, mutations of S65 to bulky or highly polar residues, mutations of Y66 to any nonaromatic amino acid, and mutations of G67 to anything else are probably not tolerated.

5.17 NUCLEIC ACID CHANGES THAT DO NOT CHANGE THE PREDICTED AMINO ACID SEQUENCE—THAT IS, OPTIMIZATION OF CODON USAGE AND ELIMINATION OF CRYPTIC SPLICE SITES GFP expression levels can often be increased by redesigning the nucleic acid sequence in ways that should have no significant effect on the final protein sequence. For example, the codon usage in the jellyfish gene is not optimal for mammalian cells, so the gene has been resynthesized with mammalian-preferred codons (Crameri et al., 1996; Levy et al., 1996; Zolotukhin et al., 1996). Translation in eukaryotes can be optimized by inclusion of an optimal translation-initiation sequence (Kozak, 1989). This redesign sometimes involves inserting a new codon that begins with G immediately after the start (AUG) codon. This introduces an extra amino acid such as Ala or Val, which in some articles adds one to the numbering of all amino acids from 2 upwards (Crameri et al., 1996), whereas we prefer to call it 1a to preserve wild-type numbering. Fortunately, the N-terminus is tolerant of such additions. For ease of comparison of mutants, this chapter numbers residues according to their position in the original gfp gene. In plant cells, mRNA derived from the original gfp gene undergoes undesired splicing, which can be eliminated by codon changes (Haseloff and Amos, 1995; see also Chapter 12). GFP cDNA coding sequences have also been altered to reflect the codon bias, and thereby increase the level of expression, of a wide variety of organisms such as Chlamydomonas (Franklin et al., 2002), yeast (GeramiNejad et al., 2001), paramecium (Hauser et al., 2000), and sugar beets (Zhang et al., 2001a).

5.18

ODDS AND ENDS

Perhaps one of the most interesting and persistent questions concerning the existence of GFPs is that of their functional role in the animal: Why should they glow? What advantage is afforded to the creatures who harbor such a protein? The recent discoveries of GFPlike proteins from nonbioluminescent Anthozoan organisms indicates that the proteins primary function cannot be linked exclusively to bioluminescence. Similarly, discoveries of chromoproteins in these same animals indicate that the proteins function may not even necessarily be tied to fluorescence. Konstantin Lukyanov’s group (Gurskaya et al., 2003) has cloned a colorless, nonfluorescent GFP (acGFPL) from Aequorea coerulescens. They showed convincingly that the protein was not an artifact of cloning and that fluorescence could be imparted by a reintroducing the invariant G222 which existed naturally in acGFPL as E222. In the living organism, this protein cannot serve as an acceptor for the bioluminescence energy of aequorin, suggesting that this protein may have some completely unique role in the jellyfish. When one considers that the major absorbtion of wild-

REFERENCES

type Aequorea victoria GFP is at 396 nm with only a minor peak at 475 nm (the primary emission band of aequorin), it seems plausible that the primary role of normally fluorescing A. victoria GFP is not necessarily as an acceptor for aequorin. The absorbance maximum of wild-type A. victoria GFP can be shifted to the 475-nm peak by a singlepoint mutation at position 65, a seemingly easy evolutionary maneuver that would optimize the transfer of energy between the two proteins. Perhaps future screens will produce the optimal GFP from Aqueorea. It seems that the structure of GFP has persisted throughout the evolutionary tree and has possibly ended up as an invariant component called Nidogen of basement membranes in organisms up to humans (Hopf et al., 2001). Despite having only a 10% sequence identity with Aequorea GFP, all of the general structural components similar to GFP exist with remarkable similarity. The Nidogen residues that are equivalent to the chromophoreforming SYT residues in GFP are IGG. Several other residues within GFP that are known to be critical for forming fluorescence are also represented in Nidogen by residues that would eliminate the opportunity to form a functional fluorophore. These final two examples suggest that the striking beta-barrel structure of GFP has served several important functions in the many organisms known to express such a structure. We may simply count ourselves fortunate that at some point in the past the evolution chanced upon a version that glows.

REFERENCES Akemann, W., Raj, C. D., and Knopfel, T. (2001). Functional characterization of permuted enhanced green fluorescent proteins comprising varying linker peptides. Photochem. Photobiol. 74: 356–363. 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 99:12651–12656. Arnaudeau, S., Frieden, M., Nakamura, K., Castelbou, C., Michalak, M., and Demaurex, N. (2002). Calreticulin differentially modulates calcium uptake and release in the endoplasmic reticulum and mitochondria. J. Biol. Chem. 277:46696–46705. Ataka, K., and Pieribone, V. A. (2002). A genetically targetable fluorescent probe of channel gating with rapid kinetics. Biophys. J. 82:509–516. Baird, G. S., Zacharias, D. A., and Tsien, R. Y. (1999). Circular permutation and receptor insertion within green fluorescent proteins. Proc. Natl. Acad. Sci. USA 96:11241–11246. 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: 11984–11989. 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. 100:12111–12116. Battistutta, R., Negro, A., and Zanotti, G. (2000). Crystal structure and refolding properties of the mutant F99S/M153T/V163A of the green fluorescent protein. Proteins 41:429–437. Bell, A. F., He, X., Wachter, R. M., and Tonge, P. J. (2000). Probing the ground state structure of the green fluorescent protein chromophore using Raman spectroscopy. Biochemistry 39: 4423–4431. Bokman, S. H., and Ward, W. W. (1981). Renaturation of Aequorea green-fluorescent protein. Biochem. Biophys. Res. Commun. 101:1372–1380.

113

114

MOLECULAR BIOLOGY AND MUTATION OF GREEN FLUORESCENT PROTEIN

Brejc, K., Sixma, T. K., Kitts, P. A., Kain, S. R., Tsien, R. Y., Ormo, M., and Remington, S. J. (1997). Structural basis for dual excitation and photoisomerization of the Aequorea victoria green fluorescent protein. Proc. Natl. Acad. Sci. USA 94:2306–2311. Campbell, R. E., Tour, O., Palmer, A. E., Steinbach, P. A., Baird, G. S., Zacharias, D. A., and Tsien, R. Y. (2002). A monomeric red fluorescent protein. Proc. Natl. Acad. Sci. USA 99:7877–7882. Campbell, R. E., Tour, O., Palmer, A. E., Steinbach, P. A., Baird, G. S., Zacharias, D. A., and Tsien, R. Y. (2002). A monomeric red fluorescent protein. Proc. Natl. Acad. Sci. USA 99:7877–7882. 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. 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. 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. 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 green-fluorescent protein. Biochemistry 32:1212–1218. Cormack, B. P., Valdivia, R. H., and Falkow, S. (1996). FACS-optimized mutants of the green fluorescent protein (GFP). Gene 173:33–38. Cotlet, M., Hofkens, J., Habuchi, S., Dirix, G., Van Guyse, M., Michiels, J., Vanderleyden, J., and De Schryver, F. C. (2001). Identification of different emitting species in the red fluorescent protein DsRed by means of ensemble and single-molecule spectroscopy. Proc. Natl. Acad. Sci. USA 27:27. 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. 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. 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. Cubitt, A. B., Woollenweber, L. A., and Heim, R. (1999). Understanding structure-function relationships in the Aequorea victoria green fluorescent protein. Methods Cell Biol. 58:19–30. Davenport, D., and Nichol, J. A. C. (1955). Proc. R. Soc. London, Ser. B 144:399–411. De Angelis, D. A., Miesenbock, G., Zemelman, B. V., and Rothman, J. E. (1998). PRIM: proximity imaging of green fluorescent protein-tagged polypeptides. Proc. Natl. Acad. Sci. USA 95:12312–12316. 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 (NY) 13:151–154. Demaurex, N., and Frieden, M. (2003). Measurements of the free luminal ER Ca(2+) concentration with targeted “cameleon” fluorescent proteins. Cell Calcium 34:109–119. Dewey, T. G., and Hammes, G. G. (1980). Calculation on fluorescence resonance energy transfer on surfaces. Biophys. J. 32:1023–1035. Dewey, T. G., and Datta, M. M. (1989). Determination of the fractal dimension of membrane protein aggregates using fluorescence energy transfer. Biophys. J. 56:415–420. Doi, N., and Yanagawa, H. (1999). Design of generic biosensors based on green fluorescent proteins with allosteric sites by directed evolution. FEBS Lett. 453:305–307. Dopf, J., and Horiagon, T. M. (1996). Deletion mapping of the Aequorea victoria green fluorescent protein. Gene 173:39–44. Dove, S. G., Hoegh-Guldberg, O., and Ranganathan, S. (2001). Major colour patterns of reefbuilding corals are due to a family of GFP-like proteins. Coral Reefs 19:197–204.

REFERENCES

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. Fukuda, H., Arai, M., and Kuwajima, K. (2000). Folding of green fluorescent protein and the cycle3 mutant. Biochemistry 39:12025–12032. Fung, B. K., and Stryer, L. (1978). Surface density determination in membranes by fluorescence energy transfer. Biochemistry 17:5241–5248. Garcia-Parajo, M. F., Koopman, M., van Dijk, E. M., Subramaniam, V., and van Hulst, N. F. (2001). The nature of fluorescence emission in the red fluorescent protein DsRed, revealed by singlemolecule detection. Proc. Natl. Acad. Sci. USA 27:27. Gerami-Nejad, M., Berman, J., and Gale, C. A. (2001). Cassettes for PCR-mediated construction of green, yellow, and cyan fluorescent protein fusions in Candida albicans. Yeast 18:859–864. Gibbs, M. D., Nevalainen, K. M., and Bergquist, P. L. (2001). Degenerate oligonucleotide gene shuffling (DOGS): A method for enhancing the frequency of recombination with family shuffling. Gene 271:13–20. Giver, L., and Arnold, F. H. (1998). Combinatorial protein design by in vitro recombination. Curr. Opin. Chem. Biol. 2:335–338. Graf, R., and Schachman, H. K. (1996). Random circular permutation of genes and expressed polypeptide chains: Application of the method to the catalytic chains of aspartate transcarbamoylase. Proc. Natl. Acad. Sci. USA 93:11591–11596. 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. 276:29188–29194. 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 97:11990–11995. Guerrero, G., Siegel, M. S., Roska, B., Loots, E., and Isacoff, E. Y. (2002). Tuning Flash: Redesign of the dynamics, voltage range, and color of the genetically encoded optical sensor of membrane potential. Biophys. J. 83:3607–3618. Gurskaya, N. G., Fradkov, A. F., Pounkova, N. I., Staroverov, D. B., Bulina, M. E., Yanushevich, Y. G., Labas, Y. A., Lukyanov, S., and Lukyanov, K. A. (2003). A colourless green fluorescent protein homologue from the non-fluorescent hydromedusa Aequorea coerulescens and its fluorescent mutants. Biochem. J. 373:403–408. Hanson, G. T., McAnaney, T. B., Park, E. S., Rendell, M. E., Yarbrough, D. K., Chu, S., Xi, L., Boxer, S. G., Montrose, M. H., and Remington, S. J. (2002). Green fluorescent protein variants as ratiometric dual emission pH sensors. 1. Structural characterization and preliminary application. Biochemistry 41:15477–15488. Harpur, A. G., Wouters, F. S., and Bastiaens, P. I. (2001). Imaging FRET between spectrally similar GFP molecules in single cells. Nat. Biotechnol. 19:167–169. Haseloff, J., and Amos, B. (1995). GFP in plants. Trends Genet. 11:328–329. Hauser, K., Haynes, W. J., Kung, C., Plattner, H., and Kissmehl, R. (2000). Expression of the green fluorescent protein in Paramecium tetraurelia. Eur. J. Cell. Biol. 79:144–149. Heikal, A. A., Hess, S. T., Baird, G. S., Tsien, R. Y., and Webb, W. W. (2000). Molecular spectroscopy and dynamics of intrinsically fluorescent proteins: Coral red (dsRed) and yellow (Citrine). Proc. Natl. Acad. Sci. USA 97:11996–12001. Heim, R. (1999). Green fluorescent protein forms for energy transfer. Methods Enzymol. 302:408–423. 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. Heim, R., Prasher, D. C., and Tsien, R. Y. (1994). Wavelength mutations and posttranslational autoxidation of green fluorescent protein. Proc. Natl. Acad. Sci. USA 91:12501–12504.

115

116

MOLECULAR BIOLOGY AND MUTATION OF GREEN FLUORESCENT PROTEIN

Heim, R., Cubitt, A. B., and Tsien, R. Y. (1995). Improved green fluorescence. Nature 373:663–664. Hofmann, A., Iwai, H., Hess, S., Pluckthun, A., and Wlodawer, A. (2002). Structure of cyclized green fluorescent protein. Acta Crystallogr. D Biol. Crystallogr. 58:1400–1406. Hopf, M., Gohring, W., Ries, A., Timpl, R., and Hohenester, E. (2001). Crystal structure and mutational analysis of a perlecan-binding fragment of nidogen-1. Nat. Struct. Biol. 8:634–640. Hyun Bae, J., Rubini, M., Jung, G., Wiegand, G., Seifert, M. H., Azim, M. K., Kim, J. S., Zumbusch, A., Holak, T. A., Moroder, L., Huber, R., and Budisa, N. (2003). Expansion of the genetic code enables design of a novel “gold” class of green fluorescent proteins. J. Mol. Biol. 328:1071–1081. Isshiki, M., Ying, Y. S., Fujita, T., and Anderson, R. G. (2002). A molecular sensor detects signal transduction from caveolae in living cells. J. Biol. Chem. 277:43389–43398. Jayaraman, S., Haggie, P., Wachter, R. M., Remington, S. J., and Verkman, A. S. (2000). Mechanism and cellular applications of a green fluorescent protein-based halide sensor. J. Biol. Chem. 275:6047–6050. Jones, J., Heim, R., Hare, E., Stack, J., and Pollok, B. A. (2000). Development and application of a GFP-FRET intracellular caspase assay for drug screening. J. Biomol. Screen 5:307–318. Kaether, C., and Gerdes, H. H. (1995). Visualization of protein transport along the secretory pathway using green fluorescent protein. FEBS Lett. 369:267–271. Kahana, J. A., and Silver, P. A. (1996). Use of the A. Victoria green fluorescent protein to study protein dynamics in vivo. In Current Protocols in Molecular Biology, Aosobel, F. M., Brent, R., Kingston, R. E., Moore, D. E., Seidman, J. G., Smith, J. A., and Struhl, K., Eds., (New York, NY, John Wiley & Sons, Inc.), pp. 9.7.22–29.27.27. Kelmanson, I. V., and Matz, M. V. (2003). Molecular basis and evolutionary origins of color diversity in great star coral Montastraea cavernosa (Scleractinia: Faviida). Mol. Biol. Evol. 20:1125–1133. Kerr, R., Lev-Ram, V., Baird, G., Vincent, P., Tsien, R. Y., and Schafer, W. R. (2000). Optical imaging of calcium transients in neurons and pharyngeal muscle of C. elegans. Neuron 26:583–594. Kim, H. K., and Kaang, B. K. (1998). Truncated green fluorescent protein mutants and their expression in Aplysia neurons. Brain. Res. Bull. 47:35–41. Kimata, Y., Iwaki, M., Lim, C. R., and Kohno, K. (1997). A novel mutation which enhances the fluorescence of green fluorescent protein at high temperatures. Biochem. Biophys. Res. Commun. 232:69–73. Kozak, M. (1989). The scanning model for translation: An update. J. Cell Biol. 108:229–241. Kummer, A. D., Wiehler, J., Schuttrigkeit, T. A., Berger, B. W., Steipe, B., and Michel-Beyerle, M. E. (2002). Picosecond time-resolved fluorescence from blue-emitting chromophore variants Y66F and Y66H of the green fluorescent protein. Chembiochem 3:659–663. Kuner, T.. and Augustine G. J. (2000). A genetically encoded ratiometric indicator for chloride: capturing chloride transients in cultured hippocampal neurons. Neuron 27:447–459. Lakowicz, J. (1999). Principles of Fluorescence Spectroscopy, 2nd ed., Plenum, New York. Levy, J. P., Muldoon, R. R., Zolotukhin, S., and Link, C. J., Jr. (1996). Retroviral transfer and expression of a humanized, red-shifted green fluorescent protein gene into human tumor cells. Nat. Biotechnol. 14:610–614. Lim, C. R., Kimata, Y., Oka, M., Nomaguchi, K., and Kohno, K. (1995). Thermosensitivity of green fluorescent protein fluorescence utilized to reveal novel nuclear-like compartments in a mutant nucleoporin NSP1. J. Biochem. (Tokyo) 118:13–17. Liu, L., Yermolaieva, O., Johnson, W. A., Abboud, F. M., and Welsh, M. J. (2003). Identification and function of thermosensory neurons in Drosophila larvae. Nat. Neurosci. 6:267–273. Llopis, J., McCaffery, J. M., Miyawaki, A., Farquhar, M. G., and Tsien, R. Y. (1998). Measurement of cytosolic, mitochondrial, and Golgi pH in single living cells with green fluorescent proteins. Proc. Natl. Acad. Sci. USA 95:6803–6808.

REFERENCES

Luo, K. Q., Yu, V. C., Pu, Y., and Chang, D. C. (2001). Application of the fluorescence resonance energy transfer method for studying the dynamics of caspase-3 activation during UV-induced apoptosis in living HeLa cells. Biochem. Biophys. Res. Commun. 283:1054–1060. Malli, R., Frieden, M., Osibow, K., and Graier, W. F. (2003). Mitochondria efficiently buffer subplasmalemmal Ca2+ elevation during agonist stimulation. J. Biol. Chem. 278:10807– 10815. Martynov, V. I., Savitsky, A. P., Martynova, N. Y., Savitsky, P. A., Lukyanov, K. A., and Lukyanov, S. A. (2001). Alternative cyclization in GFP-like proteins family. The formation and structure of the chromophore of a purple chromoprotein from Anemonia sulcata. J. Biol. Chem. 276:21012–21016. Matz, M. V., Lukyanov, K. A., and Lukyanov, S. A. (2002). Family of the green fluorescent protein: Journey to the end of the rainbow. Bioessays 24:953–959. McAnaney, T. B., Park, E. S., Hanson, G. T., Remington, S. J., and Boxer, S. G. (2002). Green fluorescent protein variants as ratiometric dual emission pH sensors. 2. Excited-state dynamics. Biochemistry 41:15489–15494. McElroy, W. D. (1947). The energy source for biolominescence in an isolated system, Proc. Natl. Acad. Sci. USA 33:342–345. McRorie, D., and Voelker, P. (1993). Self-Associating Systems in the Analytical Ultracentrifuge, Beckman Instruments, Fullerton, CA. Miesenbock, G., De Angelis, D. A., and Rothman, J. E. (1998). Visualizing secretion and synaptic transmission with pH-sensitive green fluorescent proteins. Nature 394:192–195. Minshull, J., and Stemmer, W. P. (1999). Protein evolution by molecular breeding. Curr. Opin. Chem. Biol. 3:284–290. Mitra, R. D., Silva, C. M., and Youvan, D. C. (1996). Fluorescence resonance energy transfer between blue-emitting and red-shifted excitation derivatives of the green fluorescent protein. Gene 173:13–17. Miyawaki, A., Llopis, J., Heim, R., McCaffery, J. M., Adams, J. A., Ikura, M., and Tsien, R. Y. (1997). Fluorescent indicators for Ca2+ based on green fluorescent proteins and calmodulin. Nature 388:882–887. Miyawaki, A. (2003). Visualization of the spatial and temporal dynamics of intracellular signaling. Dev. Cell. 4:295–305. Mizuno, H., Sawano, A., Eli, P., Hama, H., and Miyawaki, A. (2001). Red fluorescent protein from Discosoma as a fusion tag and a partner for fluorescence resonance energy transfer. Biochemistry 40:2502–2510. Morise, H., Shimomura, O., Johnson, F. H., and Winant, J. (1974). Intermolecular energy transfer in the bioluminescent system of Aequorea. Biochemistry 13:2656–2662. Muhlrad, D., Hunter, R., and Parker, R. (1992). A rapid method for localized mutagenesis of yeast genes. Yeast 8:79–82. Nagai, T., Sawano, A., Park, E. S., and Miyawaki, A. (2001). Circularly permuted green fluorescent proteins engineered to sense Ca2+. Proc. Natl. Acad. Sci. USA 98:3197–3202. 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. Nakai, J., Ohkura, M., and Imoto, K. (2001). A high signal-to-noise Ca(2+) probe composed of a single green fluorescent protein. Nat. Biotechnol. 19:137–141. Niwa, H., Inouye, S., Hirano, T., Matsuno, T., Kojima, S., Kubota, M., Ohashi, M., and Tsuji, F. I. (1996). Chemical nature of the light emitter of the Aequorea green fluorescent protein. Proc. Natl. Acad. Sci. USA 93:13617–13622. 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.

117

118

MOLECULAR BIOLOGY AND MUTATION OF GREEN FLUORESCENT PROTEIN

Palm, G., and Wlodawer, A. (1999). Spectral variants of green fluorescent protein. Methods Enzymol. 302:378–394. 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. Palmer, A. E., Jin, C., Reed, J. C., and Tsien, R. Y. (2004). Bcl-2-mediated alterations in endoplasmic reticulum Ca2+ analyzed with an improved genetically encoded fluorescent sensor. Proc. Natl. Acad. Sci. USA 101:17404–17409. Patterson, G. H., and Lippincott-Schwartz, J. (2002). A photoactivatable GFP for selective photolabeling of proteins and cells. Science 297:1873–1877. 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. Patterson, G. H., Piston, D. W., and Barisas, B. G. (2000). Forster distances between green fluorescent protein pairs. Anal. Biochem. 284:438–440. Prasher, D. (1995). Using GFP to see the light. Trends Genet. 11:320–232. 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. Reiff, D. F., Thiel, P. R., and Schuster, C. M. (2002). Differential regulation of active zone density during long-term strengthening of Drosophila neuromuscular junctions. J. Neurosci. 22:9399–9409. Rekas, A., Alattia, J. R., Nagai, T., Miyawaki, A., and Ikura, M. (2002). Crystal structure of Venus, a yellow fluorescent protein with improved maturation and reduced environmental sensitivity. J. Biol. Chem. 4:4. Richmond, T. A., Takahashi, T. T., Shimkhada, R., and Bernsdorf, J. (2000). Engineered metal binding sites on green fluorescence protein. Biochem. Biophys. Res. Commun. 268:462–465. 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. Rizzuto, R., Brini, M., De Giorgi, F., Rossi, R., Heim, R., Tsien, R. Y., and Pozzan, T. (1996). Double labelling of subcellular structures with organelle-targeted GFP mutants in vivo. Curr. Biol. 6:183–188. Salih, A., Larkum, A., Cox, G., Kuhl, M., and Hoegh-Guldberg, O. (2000). Fluorescent pigments in corals are photoprotective. Nature 408:850–853. Sankaranarayanan, S., De Angelis, D., Rothman, J. E., and Ryan, T. A. (2000). The use of pHluorins for optical measurements of presynaptic activity. Biophys. J. 79:2199–2208. 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. Scharnagl, C., Raupp-Kossmann, R., and Fischer, S. F. (1999). Molecular basis for pH sensitivity and proton transfer in green fluorescent protein: Protonation and conformational substates from electrostatic calculations. Biophys. J. 77:1839–1857. 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. Shimomura, O. (1979). Structure of the chromophore of Aequorea green fluorescent protein. FEBS Lett. 104:220–222. Siegel, M. S., and Isacoff, E. Y. (1997). A genetically encoded optical probe of membrane voltage. Neuron 19:735–741. 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.

REFERENCES

Sikorski, R. S., and Boeke, J. D. (1991). In vitro mutagenesis and plasmid shuffling: from cloned gene to mutant yeast. Methods Enzymol. 194:302–318. Simon, S. M. (1996). Cellular probes on the move. Nat. Biotechnol. 14:1221. Snyder, B., and Freire, E. (1982). Fluorescence energy transfer in two dimensions. A numeric solution for random and nonrandom distributions. Biophys. J. 40:137–148. Tawa, P., Tam, J., Cassady, R., Nicholson, D. W., and Xanthoudakis, S. (2001). Quantitative analysis of fluorescent caspase substrate cleavage in intact cells and identification of novel inhibitors of apoptosis. Cell Death Differ. 8:30–37. Ting, A. Y., Kain, K. H., Klemke, R. L., and Tsien, R. Y. (2001). Genetically encoded fluorescent reporters of protein tyrosine kinase activities in living cells. Proc. Natl. Acad. Sci. USA 98:15003–15008. Topell, S., Hennecke, J., and Glockshuber, R. (1999). Circularly permuted variants of the green fluorescent protein. FEBS Lett. 457:283–289. Tsien, R. Y. (1998). The green fluorescent protein. Annu. Rev. Biochem. 67:509–544. Tsien, R. Y., Bacskai, B., and Adams, S. R. (1993). FRET for studying intracellular signalling. Trends Cell Biol. 3:242–245. 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. Violin, J. D., Zhang, J., Tsien, R. Y., and Newton, A. C. (2003). A genetically encoded fluorescent reporter reveals oscillatory phosphorylation by protein kinase C. J. Cell Biol. 161:899–909. Volkov, A. A., Shao, Z., and Arnold, F. H. (1999). Recombination and chimeragenesis by an in vitro heteroduplex formation and in vivo repair. Nucleic Acids Res. 27:e18 i–vi. Wachter, R. M., King, B. A., Heim, R., Kallio, K., Tsien, R. Y., Boxer, S. G., and Remington, S. J. (1997). Crystal structure and photodynamic behavior of the blue emission variant Y66H/Y145F of green fluorescent protein. Biochemistry 36:9759–9765. Wachter, R. M., Elsliger, M. A., Kallio, K., Hanson, G. T., and Remington, S. J. (1998). Structural basis of spectral shifts in the yellow-emission variants of green fluorescent protein. Structure 6:1267–1277. 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. 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. Waldo, G. S., Standish, B. M., Berendzen, J., and Terwilliger, T. C. (1999). Rapid protein-folding assay using green fluorescent protein. Nat. Biotechnol. 17:691–695. 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 101:16745–16749. Wang, L., Xie, J., Deniz, A. A., and Schultz, P. G. (2003). Unnatural amino acid mutagenesis of green fluorescent protein. J. Org. Chem. 68:174–176. Ward, W. W. (1998). In GFP Green Fluorescent Protein Properties, Applications and Protocols, Chalfie, M., and Kain, S, Ed. Wiley-Liss, New York, pp. 45–75. 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. Ward, W. W., and Cormier, M. J. (1979). An energy transfer protein in coelenterate bioluminescence. Characterization of the Renilla green-fluorescent protein. J. Biol. Chem. 254:781–788. Ward, W., Prentice, H., Roth, A., Cody, C., and Reeves, S. (1982). Spectral perturbations of the Aequorea green fluorescent protein. Photochem. Photobiol. 35:803–808.

119

120

MOLECULAR BIOLOGY AND MUTATION OF GREEN FLUORESCENT PROTEIN

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 97:14091–14096. 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 99:11646–11651. Wolber, P. K., and Hudson, B. S. (1979). An analytic solution to the Forster energy transfer problem in two dimensions. Biophys. J. 28:197–210. Wu, P., and Brand, L. (1994). Resonance energy transfer: Methods and applications. Anal. Biochem. 218:1–13. Xu, X., Gerard, A. L., Huang, B. C., Anderson, D. C., Payan, D. G., and Luo, Y. (1998). Detection of programmed cell death using fluorescence energy transfer. Nucleic Acids Res. 26:2034–2035. Yang, F., Moss, L. G., and Phillips, G. N., Jr. (1996a). The molecular structure of green fluorescent protein. Nat. Biotechnol. 14:1246–1251. Yang, T. T., Kain, S. R., Kitts, P., Kondepudi, A., Yang, M. M., and Youvan, D. C. (1996b). Dual color microscopic imagery of cells expressing the green fluorescent protein and a red-shifted variant. Gene. 173:19–23. Yang, T. T., Sinai, P., Green, G., Kitts, P. A., Chen, Y. T., Lybarger, L., Chervenak, R., Patterson, G. H., Piston, D. W., and Kain, S. R. (1998). Improved fluorescence and dual color detection with enhanced blue and green variants of the green fluorescent protein. J. Biol. Chem. 273:8212–8216. Yguerabide, J. (1994). Theory for establishing proximity relations in biological membranes by excitation energy transfer measurements. Biophys. J. 66:683–693. Yokoe, H., and Meyer, T. (1996). Spatial dynamics of GFP-tagged proteins investigated by local fluorescence enhancement. Nat. Biotechnol. 14:1252–1256. Youvan, D. C., Goldman, E., Delagrave, S., and Yang, M. M. (1995). Digital imaging spectroscopy for massively parallel screening of mutants. Methods Enzymol. 246:732–748. Yu, D., Baird, G. S., Tsien, R. Y., and Davis, R. L. (2003). Detection of calcium transients in Drosophila mushroom body neurons with camgaroo reporters. J. Neurosci. 23:64–72. Zacharias, D. A., Baird, G. S., and Tsien, R. Y. (2000). Recent advances in technology for measuring and manipulating cell signals. Curr. Opin. Neurobiol. 10:416–421. Zacharias, D. A., Violin, J. D., Newton, A. C., and Tsien, R. Y. (2001). Partitioning of monomeric lipid modified GFPs inot lipid rafts of living cells. Science 296:913–916. Zacharias, D. A. (2002). Sticky caveats in an otherwise glowing report: Oligomerizing fluorescent proteins and their use in cell biology. Sci. STKE 2002:PE23. Zapata-Hommer, O., and Griesbeck, O. (2003). Efficiently folding and circularly permuted variants of the Sapphire mutant of GFP. BMC Biotechnol. 3:5. Zhang, C. L., Chen, D. F., McCormac, A. C., Scott, N. W., Elliott, M. C., and Slater, A. (2001a). Use of the GFP reporter as a vital marker for Agrobacterium-mediated transformation of sugar beet (Beta vulgaris L.). Mol. Biotechnol. 17:109–117. Zhang, J., Ma, Y., Taylor, S. S., and Tsien, R. Y. (2001b). Genetically encoded reporters of protein kinase A activity reveal impact of substrate tethering. Proc. Natl. Acad. Sci. USA 98:14997–15002. 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. Zimet, D. B., Thevenin, B. J., Verkman, A. S., Shohet, S. B., and Abney, J. R. (1995). Calculation of resonance energy transfer in crowded biological membranes. Biophys. J. 68:1592–1603. Zimmer, M. (2002). Green fluorescent protein (GFP): Applications, structure, and related photophysical behavior. Chem. Rev. 102:759–781. Zolotukhin, S., Potter, M., Hauswirth, W. W., Guy, J., and Muzyczka, N. (1996). A “humanized” green fluorescent protein cDNA adapted for high-level expression in mammalian cells. J. Virol. 70:4646–4654.

6 DISCOVERY AND PROPERTIES OF GFP-LIKE PROTEINS FROM NONBIOLUMINESCENT ANTHOZOA Konstantin A. Lukyanov, Dmitry M. Chudakov, and Arkady F. Fradkov Shemyakin and Ovchinnikov Institute of Bioorganic Chemistry RAS, Moscow, Russia

Yulii A. Labas Institute of Biochemistry RAS, Moscow, Russia

Mikhail V. Matz Whitney Laboratory, University of Florida, St. Augustine, FL

Sergey Lukyanov Shemyakin and Ovchinnikov Institute of Bioorganic Chemistry RAS, Moscow, Russia

6.1

INTRODUCTION

Green fluorescent protein (GFP) was discovered in hydroid medusa Aequorea victoria (synonyms A. forskalea, A. aequorea) more than 40 years ago (Johnson et al., 1962; Chapter 1 of this volume). After that, GFPs were found in several bioluminescent Hydrozoa and Anthozoa species (Chalfie, 1995). In all these examples, GFPs played role of secondary emitter within bioluminescent systems. The association of GFPs with bioluminescence was possibly the main reason why researchers did not search for GFP-like proteins in nonbioluminescent corals for a long time. We were lucky to clone genes for GFP-like proteins from nonbioluminescent Anthozoa for the first time (Matz et al., 1999). Fortune really smiled on us: Several months after the work was started, we cloned the first GFP from the sea anemone Anemonia majana. In comparison to many other marine coelenterates, working on reef Anthozoa is particuGreen Fluorescent Protein: Properties, Applications, and Protocols, Second Edition Edited by Martin Chalfie and Steven R. Kain Copyright © 2006 John Wiley & Sons, Inc.

121

122

DISCOVERY AND PROPERTIES OF GFP-LIKE PROTEINS FROM NONBIOLUMINESCENT ANTHOZOA

larly convenient because one can readily buy the specimens for several dollars in aquarium shops throughout the world. We did not organize expeditions on a ship or bathyscaphe, but found brightly fluorescent and colored sea anemones and corallimorphs (mushroom anemones) in Moscow instead. Shortly after the publication of our paper, other groups independently reported finding of GFP-like proteins in Anthozoa species (Wiedenmann et al., 2000; Dove et al., 2001). At present, it is widely accepted that vivid fluorescent and nonfluorescent coloration of coral polyps is mainly determined by numerous GFP homologs (Matz et al., 1999, 2002; Fradkov et al., 2000; Lukyanov et al., 2000; Wiedenmann et al., 2000, 2002; Salih et al., 2000; Gurskaya et al., 2001a; Dove et al., 2001; Labas et al., 2002; Ando et al., 2002).

6.2

COLOR DIVERSITY WITHIN ANTHOZOA GFP HOMOLOGS

The most interesting feature of the coral GFP-like proteins is their color variety as determined by the proteins emission properties. Four main color groups have been recognized thus far (Labas et al., 2002): (a) green, yellow, and red fluorescent proteins (FPs) and (b) nonfluorescent chromoproteins (CPs) of different hues, from orange to blue (Fig. 6.1, left column). The green FPs are divided into three subgroups. The first subgroup contains cyan FPs, which are relatively blue-shifted and have broad spectra (Fig. 6.1A). These proteins possess excitation maxima at 440–460 nm and emission maxima at ~480–490 nm, along with spectral widths (width of the peak at half of maximal intensity) of emission curves of about 50 nm. The second subgroup includes the majority of the green FPs, proteins that have excitation–emission maxima at 490–510 and 500–520 nm, respectively, and a narrow fluorescence peak (spectral width about 25 nm; Fig. 6.1B). A characteristic feature of the third subgroup is dual-peaks excitation spectrum usually having a major peak at around 400 nm and a minor peak at 470–490 nm (Fig. 6.1C). There was found only one yellow FP, zoanYFP (zFP538) (Matz et al., 1999). It has a narrow emission spectrum with a peak at 538 nm and has an excitation spectrum with a major peak at 528 nm and a minor peak at 494 nm (Fig. 6.1D). Red FPs possess emission maxima greater than 570 nm. Often these proteins go through green-emitting stage during their posttranslational maturation. RFPs can be subdivided into two subgroups. The first subgroup is represented by drFP583 (Matz et al., 1999) (commercial name DsRed, available from Clontech Laboratories), the most well-studied RFP to date. These RFPs are characterized by a broad emission spectrum (spectral width about 50–60 nm) with a peak between 570 and 610 nm (Fig. 6.1E). The main characteristic feature of the second RFP subgroup is need of UV or violet light irradiation for red fluorophore formation, as it was first shown by Ando et al. (2002) and later confirmed for other similar proteins (Wiedenmann et al., 2004; M. Matz and K. Lukyanov, unpublished). In the dark these proteins mature only to the green FP form. Short-wavelength light irradiation causes their fast transformation into red FP form. The resulting red emission spectra are rather narrow (spectral width about 25 nm) and have a pronounced shoulder at about 630 nm (Fig. 6.1F). The last color class of GFP-like proteins unites chromoproteins (CPs) that effectively absorb but emit little or no light (Lukyanov et al., 2000; Wiedenmann et al., 2000; Gurskaya et al., 2001a; Dove et al., 2001). Known CPs possessed single absorption maxima at 560–590 nm (Fig. 6.1G). Curiously, in this region of spectra, a subtle shift of the absorption maximum may lead to a significant change in the perceived CP color, so

123

COLOR DIVERSITY WITHIN ANTHOZOA GFP HOMOLOGS

W ild types A

4 45

M utants dstrG FP

4 83

H

4 60

4 93

6 10 5 42

dstrG FP N 68S I112S G reen

+ Red

B

zoanG FP

I

zoanG FP A 63G N 65K N 68D

5 25 5 37 4 94

G reen

4 94 5 06

C

hcriG FP

4 99

4 05

zoanY FP D 68N

J

G reen + Yellow

4 81

D

zoanY FP

Yellow

5 24 5 37

K

zoanY FP K 65M

4 94 5 06

4 94

E

Yellow

G reen

5 58

5 83

D sR ed

L

D sR ed Y 121H

4 80 5 00

Red

G reen

F

5 08 5 20

5 80

m cavR FP

M

hcriC P

N

D sR ed S 148C I165N K 167M Chrom o S 203A

5 61

5 72

G

5 90

Chrom o

5 78

6 40

Far-red hcriC P C 148 S 400

500 600 Wavelength, nm

700

400

500 600 Wavelength, nm

700

Figure 6.1. Spectral properties of wild-type and mutant GFP-like proteins of the main color groups. Spectra for wild-type proteins are shown on the left, and spectra for corresponding mutants with altered color are shown on the right. Dashed lines represent excitation spectra for FPs or absorption spectra for CPs. Solid lines correspond to the emission spectra.

that to a human observer these proteins may appear as soft hues of purple, crimson, lilac, and even blue. In some CPs, extremely weak (quantum yield 48 hours to reach >90% of maximal fluorescence, a length of time that is prohibitive for use in a fast-growing organism such as yeast (Baird et al., 2000). The second drawback of DsRed is that it exists in solution and in vivo as a tetramer (Baird et al., 2000). While this oligomerization does not preclude the use of DsRed as an organelle or expression marker in yeast, it does complicate interpretation of the localization and function of DsRed fusion proteins. Directed optimization of DsRed, however, appears to have successfully addressed these shortcomings. Bevis and Glick (2002) reported on the isolation of a DsRed variants that mature 10–15 times faster than DsRed, albeit with lower fluorescence intensity. Encouragingly, these authors found that a mitochondrially targeted DsRed variant, in contrast to the parental DsRed, gave consistently strong fluorescence in S. cerevisiae cells from growing cultures. As mentioned earlier, the Yeast Resource Center has a template plasmid available for the targeted integration of one of these DsRed variants (termed DsRed.T1.N1) into the S. cerevisiae and S. pombe genomes (see Sections 9.2.1 and 9.2.2); however, researchers should be aware that this DsRed variant is still prone to tetramerization. To further improve DsRed as an in vivo fluorescent marker, Campbell et al. (2002) subjected one of these rapidly maturing DsRed variants to stepwise evolution, leading ultimately to the isolation of both a monomeric (mRFP) and tandem dimer (effectively

185

186

THE USES OF GREEN FLUORESCENT PROTEIN IN YEASTS

monomeric, termed tdimer2) DsRed. Several studies have confirmed that both mRFP and tdimer2 can be expressed successfully in S. cerevisiae [see, for example, Huh et al. (2003), Malinska et al. (2003), and Sheth and Parker (2003)]. Thus, the mRFP and tdimer2 DsRed variants appear to be the best currently available option for expression of red fluorescence in yeast. Resources necessary for the expression of these monomeric DsRed variants in yeast should be available, with the proper permissions, from the aforementioned authors.

9.3

UTILIZATION OF GFP AND GFP FUSION PROTEINS IN YEAST

GFP has revolutionized microscopic studies of dynamic, in vivo protein localization in the yeast cell. Recent years have witnessed a second revolution, in which GFP technology has been successfully applied to studies of organelle function and inheritance, in vivo DNA, RNA, and lipid localization, and the mapping of in vivo protein–protein interactions (see Fig. 9.1). Finally, GFP has become a permanent addition to the yeast geneticists’ tool kit,

Figure 9.1. In vivo macromolecular localization via GFP. (A) Proteins can be localized in living cells by expressing them as GFP fusions. (B) DNA can be visualized by expressing GFP fused to a DNA-binding protein (DBP) or to a DNA binding domain (DBD) that recognizes specific DNA sequences or regions. (C) RNA molecules engineered to contain hairpin tertiary structures can be localized in vivo by expressing GFP fusions to RNA binding domains (RBDs) that recognize such hairpins. (D) Phosphoinositides have been visualized in living cells by expressing GFP fusions to lipid-binding domains (LBDs) such as the plextrin homology (PH) and FYVE domains. Details regarding each of these approaches to localizing proteins, nucleic acids, and lipids are given in Section 9.3.2.

UTILIZATION OF GFP AND GFP FUSION PROTEINS IN YEAST

serving as a reporter for gene expression studies and in large-scale forward-genetic visual screens. In this section, we review several studies that highlight the diverse, creative, and powerful uses of GFP and GFP fusions in yeast.

9.3.1

Organelle Structure, Function, and Inheritance

GFP fusion proteins have served as valuable markers for intracellular compartments in studies of yeast organelle structure, function, and inheritance. Additionally, these reporters can be used to confirm colocalization of a protein or other macromolecule of interest to a specific subcellular compartment. GFP fusion proteins that have been utilized for studies of nearly every organelle and compartment in yeast have been compiled with corresponding references in two recent reviews (Kohlwein, 2000; Tatchell and Robinson, 2002). Typically, organelles are visualized by fusing GFP to either (1) a protein that is well characterized to localize exclusively to the organelle/compartment of interest or (2) the minimal amino acid sequence required for protein targeting to the organelle/compartment of interest. As an example of the first approach, Prinz et al. (2000) studied the dynamic structure of the cortical ER in S. cerevisiae by monitoring a GFP fusion to the wellcharacterized ER protein Sec63p (see Fig. 9.2). The second approach is exemplified by the recent description of a mitochondrially localized GFP construct (mtGFP) generated by fusing the mitochondrial presequence (targeting signal) at the N-terminus of GFP

Figure 9.2. Expression of Sec63-GFP highlights the ER of living yeast cells. The continuous perinuclear ER is best seen in images acquired while focusing on the center of the cells (top panels), while the cortical ER network is best seen with focal planes close to the periphery of the cell (bottom panels). Bars, 5 mm. [Reproduced from Prinz et al. (2000) with permission by The Rockefeller University Press.]

187

188

THE USES OF GREEN FLUORESCENT PROTEIN IN YEASTS

(Westermann and Neupert, 2000). This mtGFP reporter has been used, for example, to characterize factors required for normal mitochondrial morphology in the budding yeast cell (Messerschmitt et al., 2003). Similarly, the inheritance of peroxisomes has been monitored by in vivo time-lapse microscopy of budding yeast expressing a GFP fusion to the type I peroxisomal targeting sequence (GFP-PTS1; Hoepfner et al., 2001). Dual labeling of peroxisomes and the actin cytoskeleton, in part through the use of the GFP variants CFP and YFP, along with the use of yeast mutants, further allowed these authors to demonstrate that peroxisome movement into the daughter cell is dependent upon the action of the Myo2p myosin motor protein along actin filaments.

9.3.2

Localization Studies

9.3.2.1 Protein Localization. GFP fusions have been utilized to determine the in vivo steady-state localization of many proteins, thereby guiding hypotheses of protein function. In fact, a large-scale study cataloging the in vivo steady-state localizations of >4000 S. cerevisiae proteins (representing 75% of the yeast proteome) by GFP tagging and microscopy has recently been published (Huh et al., 2003). The true worth of GFP, however, has been revealed in studies of dynamic protein movement and trafficking within the yeast cell. For example, several groups have utilized GFP fusion to better understand the signals, mechanisms, and regulation of nucleocytoplasmic transport. Görner et al. (1998) found that GFP fusions to the stress-response transcription factors Msn2p and Msn4p are normally localized to the cytoplasm, but accumulate rapidly in the nucleus when stress is applied. By localizing Msn2p-GFP in a panel of yeast strains deleted or mutated for nuclear transport receptors (also termed importins/exportins or karyopherins), we found that the exportin Msn5p was responsible for Msn2p-GFP nuclear export [our unpublished results; see also Chi et al. (2001) and Gorner et al. (2002)]. Figure 9.3 depicts the constitutive nuclear localization of Msn2p-GFP in Dmsn5 cells, even in the absence of stress. Jacquet et al. (2003) tracked the stress-activated nuclear translocation of Msn2pGFP in real time by high-resolution time-lapse video microscopy. Strikingly, these authors found that an intermediate stress response (triggered by the microscope excitation light) caused the entire population of Msn2p-GFP to shuttle repeatedly into and out of the nucleus with a periodicity of a few minutes. This dynamic behavior of Msn2p and its implications for the autoregulation of Msn2p localization would not have been appreciated without the capability afforded by GFP to follow protein localization in living cells in real time. Further examples of the utility of GFP in studies of nucleocytoplasmic transport are plentiful. Convenient GFP reporter proteins have been generated to monitor the rates and requirements for nuclear localization signal (NLS)- and nuclear export signal (NES)dependent nuclear transport in yeast (Roberts and Goldfarb, 1998; Shulga et al., 1996; Stade et al., 1997; Taura et al., 1998). Fusions of GFP to components of the ribosome have led to the development of in vivo assays to identify the nuclear transport pathway(s) of this large riboprotein complex (Hurt et al., 1999; Stage-Zimmermann et al., 2000). Finally, GFP fusion proteins have been effectively used to study the localization and movements of nuclear transport factors and nuclear pore complexes themselves, shedding further light on the striking dynamics of the nucleocytoplasmic transport machinery (Belgareh and Doye, 1997; Bucci and Wente, 1997; Seedorf et al., 1999; Stade et al., 1997). Studies of the dynamic behavior, function, and regulation of the mitotic spindle in yeast have also benefited greatly by the use of GFP fusion proteins. Using a fusion of the

UTILIZATION OF GFP AND GFP FUSION PROTEINS IN YEAST

Figure 9.3. Stress- and Msn5-regulated nuclear transport of Msn2-GFP in S. cerevisiae. Budding yeast cells expressing a GFP fusion to the Msn2 transcription factor were visualized by fluorescence and Nomarski differential interference contrast (DIC) microscopy in the absence and presence of stress (0.4 M NaCl). Msn2-GFP can be seen to localize throughout the cell in the absence of stress and to relocalize tightly to the nucleus in the presence of stress (left panels). The constitutive localization of Msn2-GFP in Dmsn5 cells (right panels) indicates that the Msn5 nuclear transport receptor is required for nuclear export of this transcription factor.

budding yeast spindle pole body (SPB; the yeast equivalent of the centrosome) antigen Nuf2p to GFP and time-lapse fluorescence microscopy in living cells, we determined the rate and polarity of mitotic spindle growth in vivo [see Fig. 9.4 and Kahana et al. (1995)]. Similarly, Carminati and Stearns (1997) expressed GFP fusions to the microtubule component a-tubulin (Tub1p) to localize astral and mitotic spindle microtubules in vivo. From their time-lapse observations of GFP-labeled microtubules through the cell cycle, the authors were able to document the role of astral microtubules in mitotic spindle positioning. More recently, Maddox et al. (2000) studied the polarity and dynamics of the minus ends of budding yeast microtubules at the SPB through the in vivo monitoring of a GFPtagged a-tubulin subunit by fluorescent speckle microscopy (FSM) and fluorescence redistribution after photobleaching (FRAP). Based on their findings, the authors were able to conclude that the dynamic assembly and disassembly of astral and spindle microtubules most likely occurs at their plus, and not minus, ends. Finally, GFP-tagged tubulin expression has also allowed assessment of microtubule dynamics in the mitotic spindle of S. pombe (Mallavarapu et al., 1999) and observation of microtubules in budding and hyphal forms of C. albicans (Bachewich et al., 2003; Hazan et al., 2002). An exciting recent development to GFP-based studies in yeast is the use of the YFP and CFP spectral variants, as well as the related red fluorescent protein DsRed, to make real-time observations of the dynamic colocalization of proteins in living cells. The power in this approach is exemplified in an elegant study by Browning et al. (2003) in which the authors investigated the in vivo movement of cargo proteins along microtubules in fission yeast. In this study, a fusion of the Tea2p kinesin protein to YFP was observed by timelapse fluorescence microscopy in S. pombe cells that coexpressed a tubulin-CFP fusion. In real time, Tea2p-YFP could be seen loading onto microtubules in the middle of the cell near the nucleus and traveling at the tips of polymerizing microtubules toward the end of

189

190

THE USES OF GREEN FLUORESCENT PROTEIN IN YEASTS

Figure 9.4. Time-lapse fluorescence microscopy montage of budding yeast cells expressing the centrosomal antigen Nuf2-GFP. In frames in which the centrosomes (white dots) are parfocal, the distance between them is measured in CCD pixels (14.7 pixels = 1 mm). A time stamp is included at the bottom of each frame.

the cell. By introducing ATPase mutations into the Tea2p fusion protein, the movement of this kinesin to the cell ends was shown to require motor activity. By monitoring the movement of wild-type Tea2p-YFP in various mutant backgrounds, the authors were further able to identify proteins that regulate Tea2p cell-end transport both at early (microtubule loading) and late (cell-end anchoring) steps. 9.3.2.2 DNA Localization. GFP-based methods for monitoring in vivo DNA localization have revolutionized studies of chromosomal movement, localization, interactions, and architecture in the context of nuclear processes including mitosis, meiosis, DNA transcription, silencing, repair, and recombination. Generally, yeast DNA can be visualized in two ways with GFP-based systems. First, endogenous proteins that interact with DNA (such as histones, centromere components, or silencing factors) can be tagged with GFP and utilized as an indirect monitor of the dynamics of the DNA region(s) they bind.

UTILIZATION OF GFP AND GFP FUSION PROTEINS IN YEAST

As an example of this approach, Pidoux et al. (2000) utilized a GFP fusion to the Swi6p heterochromatin-binding protein as a centromeric and telomeric marker during real-time analysis of chromosome segregation in S. pombe; by monitoring GFP-Swi6p in chromosome segregation mutants, the authors were able to observe the wide range of behavior displayed by lagging chromosomes. GFP fusions to histone proteins (H2B and H4) have also been utilized to monitor nuclear migration and movement during cell division in wildtype and mutant budding yeast (Hoepfner et al., 2000; Thrower et al., 2003). As a second GFP-based approach to localizing DNA, specific chromosomal sites can be “tagged” with direct repeats of bacterial operator sequences (typically derived from the lac or tet operons); these tags can then be visualized using GFP fusions to the appropriate repressor (DNA-binding) protein [reviewed in Belmont (2001)]. An excellent example of this second approach can be found in a study of the dynamic localization of chromosomal regions during interphase in budding yeast. Heun et al. (2001) tagged early and late replication origins, as well as centromeric and telomeric loci, with tandem lac operator sequences; these four chromosomal regions were then monitored by expression of LacIGFP and time-lapse microscopy. Coexpression of a GFP-tagged nuclear pore protein allowed for the measurement of DNA movements relative to either the nuclear periphery or the calculated center of the nucleus. These experiments revealed that interphase chromatin is highly dynamic (in the case of replication origins, moving distances ≥0.5 mm within seconds) and that different chromosomal domains display varying degrees of constraint on their movement. This GFP-based chromosomal tagging technique has also been successfully applied to S. pombe (Nabeshima et al., 1998). Despite the obvious power of this approach, a valid concern is that the integration of bacterial operator sequences (often up to 10 kb) might significantly interfere with the normal, physiological behavior of a given chromosomal region. Gasser (2002) has suggested that such inserts do not substantially alter local chromatin structure, but such a finding would almost certainly depend on the specific genomic region. Therefore, inclusion of functional tests with the altered genomic region would significantly increase the credibility of any such studies. 9.3.2.3 RNA Localization. Studies of RNA localization and movement in yeast, especially for single transcripts, have been difficult due to the relative insensitivity of fluorescence in situ hybridization (FISH) techniques and the inability to perform injections of fluorescently labeled RNA. Furthermore, the extensive manipulations required for such techniques often call into question the physiological relevance of resulting observations. Within the past five years, a number of research groups have developed systems that allow for the GFP-based in vivo localization of RNA [reviewed in Brodsky and Silver (2002)]. Two components are required for GFP-based RNA imaging: an RNA-binding protein (RBP) fusion to GFP and an RNA (typically the RNA of interest) engineered to contain the binding sites (typically a hairpin structure) for the RBP. Two hairpin–RBP interactions have been used for GFP-based RNA imaging: one derived from the bacteriophage MS2 capsid protein and the other from human splicing protein U1A (each along with its cognate RNA hairpin binding sites). In yeast coexpressing these two components, the hairpincontaining RNA is bound by the RBP-GFP fusion; in this way, GFP fluorescence serves as an indirect indicator of RNA localization. Functional tests with the altered RNA of interest can be performed to attempt to allay concerns that the introduced hairpin structures (often in tandem repeats) interfere with physiological localization and/or function of the RNA. These GFP-based RNA localization techniques, combined with yeast genetics, have contributed significantly to the identification of factors required for mRNA nuclear export

191

192

THE USES OF GREEN FLUORESCENT PROTEIN IN YEASTS

(Brodsky and Silver, 2000) and bud-tip localization of the ASH1 mRNA (Beach and Bloom, 2001; Beach et al., 1999; Bertrand et al., 1998). Additionally, these techniques have allowed the relatively rapid screening and identification of additional nuclear encoded mRNAs that localize to the mitochondria or bud tip in yeast (Corral-Debrinski et al., 2000; Marc et al., 2002; Shepard et al., 2003). Finally, a recent and particularly elegant study employed a combination of GFP-based protein- and RNA-localization techniques to investigate the significance of cytoplasmic foci formed by mRNA decay factors in budding yeast (Sheth and Parker, 2003). By colocalizing mRNA degradation intermediates in vivo to these foci, Sheth and Parker identified these foci as actual sites of mRNA decay, rather than as storage or assembly sites for decay factors. As such, this finding adds tremendously to our growing appreciation for the importance of subcytoplasmic “compartments” in mRNA sequestration and regulation. 9.3.2.4 Lipid Localization. The development of GFP-based reporters for the in vivo detection of specific phosphorylated pools of the eukaryotic membrane component phosphatidylinositol (PtdIns) has greatly aided studies of the compartmentalized regulatory roles of these signaling lipids in both mammalian and yeast cells. Such GFP-based lipid reporters, sometimes termed FLAREs (for fluorescent lipid-associated reporters), take advantage of protein domains that have been identified in yeast and mammalian proteins to bind with high affinity and specificity to the inositide head group of particular phosphoinositides (see Balla et al., 2000; Balla and Varnai, 2002). By expressing a GFP fusion to any one of these phosphoinositide-binding modules, the lipid population corresponding to that module’s specificity can in principle be detectable by fluorescence microscopy of single living yeast cells. Stefan et al. (2002) visualized cellular pools of PtdIns(4,5)P2 at the budding yeast plasma membrane by expressing GFP fusions to a PtdIns(4,5)P2-binding plextrin homology (PH) domain derived from mammalian phospholipase C d1. Their finding that this lipid pool mislocalized in yeast mutated for a class of phosphoinositide 5-phospatases, concurrent with defects in cell morphology and membrane trafficking, underscores the importance of proper PtdIns(4,5)P2 compartmentalization to these biological processes. The same authors, using a GFP fusion to the PH domain derived from the mammalian FAPP1 protein, localized PtdIns(4)P pools to intracellular Golgi compartments, consistent with studies associating this phosphoinositide with transport of proteins from the Golgi (Stefan et al., 2002). In an earlier study from the same group, PtdIns(3)P was localized in vivo to endocytic compartments by a GFP-FYVE fusion protein, in agreement with its function in protein sorting in the late secretory pathway of yeast (Burd and Emr, 1998). Finally, GFP-FYVE has additionally been expressed in fission yeast to investigate the role of PtdIns(3)P in formation of the forespore membrane during sporulation (Onishi et al., 2003). It is worth noting that as protein domains with novel lipid-binding specificities continue to be identified, the repertoire of lipids (phosphoinositides and others) that can be detected and studied in living yeast cells by GFP will almost certainly expand.

9.3.3

Fluorescence Resonance Energy Transfer

The GFP derivatives CFP and YFP can be used in fluorescence resonance energy transfer (FRET) studies to detect in vivo interactions between two tagged proteins. During FRET, the excited donor fluorophore (CFP) directly transfers energy to the acceptor fluorophore (YFP); FRET can be detected, therefore, by measuring emission from YFP in the presence of wavelengths excitatory to CFP. Because FRET only occurs if the donor and accep-

UTILIZATION OF GFP AND GFP FUSION PROTEINS IN YEAST

tor moieties are very close in space (maximum separation of ~25–35 Å), the presence of a FRET indicates a high probability that the two tagged proteins directly interact. Yeastbased in vivo FRET studies especially benefit from the ability to express CFP- and YFPfusion proteins at controlled and consistent physiological levels from integrated genomic fusions. This technical advantage greatly increases the likelihood that the fusion protein will maintain functional localizations and interactions and that consistent FRET values will be obtained from cell to cell. Readers are directed to a recent method-based review by Hailey et al. (2002) of FRET applications in yeast. Several groups have applied the FRET assay to studies of yeast protein–protein interactions. In one study from our laboratory, potential pairwise interactions between YFPtagged nuclear pore proteins (termed nucleoporins) and CFP-tagged nuclear transport receptors were tested by microscopic FRET analysis (Damelin and Silver, 2000). Coimmune precipitation was able to confirm at least one novel receptor–nucleoporin interaction detected by FRET, bolstering the credibility of results obtained by this method. In all, this work revealed that distinct nuclear import and export receptors traverse the nuclear pore complex by overlapping, but not identical, pathways. A refined molecular model of protein-protein interactions within the nuclear pore complex was generated in a second study by testing for FRET interactions between nucleoporins themselves (Damelin and Silver, 2002). Finally, FRET has been utilized by Blumer and colleagues to demonstrate and map sequences required for the in vivo oligomerization of the G-protein coupled mating factor receptor Ste2p at the yeast plasma membrane (Overton and Blumer, 2000; Overton et al., 2003).

9.3.4

Gene Expression and Genetic Studies with GFP

GFP has become one of the most widely used reporters of gene expression, largely because the addition of exogenous substrates and/or cell disruption is not required for fluorescence, along with the ease of detection by fluorescence microscopy, spectrophotometry, and flow cytometry. A study by Li et al. (2000), for instance, monitored the in vivo kinetics of induction from the yeast GAL1 promoter using GFP as a reporter. The stability of GFP can be advantageous in studies of promoter activation, where low expression levels and sensitivity are issues. If, however, decreases and/or dynamic changes in gene expression must be monitored, the stability of the GFP protein can mask downregulation of transcription. To address this concern, a destabilized form of GFP has been generated for yeast by the addition of the ubiquitin/proteasome-targeting PEST motif of the constitutively unstable yeast Cln2p protein (Mateus and Avery, 2000). Notably, GFP-Cln2pPEST displayed a halflife of ~30 minutes as opposed to the ~7-hour half-life of unaltered GFP. Using this destabilized GFP reporter, the authors were able to monitor dynamic gene expression from the Cu2+-regulated CUP1 promoter as well as the cell-cycle regulated CLN2 promoter. Along similar lines, silencing- and promoter-responsive GFP expression constructs have been developed to specifically monitor processes such as mating type switching and the response to DNA damage in yeast (Laney and Hochstrasser, 2003; Walmsley et al., 1997). Finally, GFP has found important use as a reporter for several large-scale forward genetic screens in yeast. Because GFP can be rapidly observed by microscopy, it has become feasible to screen yeast temperature-sensitive or knockout libraries for mutants that improperly localize a given GFP fusion protein. For example, Ryan and Wente (2002) screened a library of temperature-sensitive mutants for mislocalized GFP-labeled nuclear pore complexes (NPCs) in an effort to identify yeast mutants affecting NPC assembly.

193

194

THE USES OF GREEN FLUORESCENT PROTEIN IN YEASTS

Interestingly, their identification of yeast mutated for regulators of the small GTPase Ran in this screen suggests for the first time a potential role for the Ran GTPase cycle in NPC assembly (Ryan et al., 2003). If the specific parameters permit, one can pre-enrich for mutants of interest by fluorescence activated cell sorting (FACS) of GFP-expressing yeast. Hammell et al. (2002) successfully identified yeast mutants affected for mRNA nuclear export by first enriching by FACS for mutants that failed to express a GFP fusion to the heat-shock response protein Ssa4p after shift to 42°C. More labor-intensive secondary screening (in situ hybridization to detect the SSA4 mRNA) then identified the subset of mutants that failed to express Ssa4p-GFP due to blocked mRNA export. The results of this genetic screen nicely reinforce the interdependence of mRNA 3¢ processing and nuclear export.

9.4

METHODS

The following two sections outline several methods for GFP fusion protein expression in S. cerevisiae and the detection of such proteins by fluorescence microscopy. For more detail on methods regarding S. pombe and C. albicans expression systems, as well as yeast DNA-, RNA-, and lipid-localization, FRET microscopy techniques and quantification, the reader is referred to reviews cited in the relevant sections above.

9.4.1

S. cerevisiae GFP Expression Methods

9.4.1.1 Extrachromosomal (Plasmid) GFP Constructs. We have designed a series of yeast plasmid vectors for the construction of GFP fusion proteins in budding yeast. The pCGF (C-terminal GFP fusion) series of high-copy vectors are designed for pGAL1-10 inducible expression of yeast gene fusions to the 3¢ end of GFP [orienting GFP at the N-terminus of the protein of interest; see Kahana and Silver (1996)]. Once an ORF is cloned into a pCGF vector and transformed into S. cerevisiae, high-level fusion gene expression can be induced by addition of galactose to the growth medium. Notably, the GAL1-10 inducible promoter can be extremely useful for expression of non-yeast cDNAs or yeast genes normally expressed at levels too low to be seen with GFP. While this system may be useful for some proteins, the atemporal and typically highlevel expression may cause several problems. First, such expression may, depending on the gene fused, have toxic side effects. Second, overexpression of a GFP fusion gene may cause it to mislocalize. For instance, when a Nuf2p-GFP fusion is expressed from a pCGF vector in the presence of 2% galactose, fluorescence is observed throughout the cell as opposed to its normal exclusive localization to the spindle pole body (SPB). Some of these problems can be ameliorated by repressing expression of the GFP fusion gene with glucose after sufficient induction in galactose. We have observed that expression of Nuf2p-GFP from a pCGF vector using a one-hour galactose “pulse” followed by a six-hour glucose “chase” leads to accurate localization of the fusion protein [at ~1000–5000 molecules Nuf2p-GFP/SPB; see Kahana et al. (1995)]. Presumably this effect is attributable to dilution of protein level after turnover and cell division during the glucose chase. Moreover, mixtures of galactose and glucose can give low- and intermediate-level expression, thus modulating protein levels even further. A second potential solution to problems with GFP fusion gene misexpression and protein mislocalization can be to direct fusion gene expression by the gene’s endogenous promoter. To do this in a plasmid-based system, we developed the centromeric (ARS/CEN)

METHODS

pNGF (N-terminal GFP fusion) vector that contains GFP followed by the NUF2 3¢ terminator sequence (Kahana and Silver, 1996). Once a gene and its upstream (5¢) promoter sequence (~500 bp is usually sufficient) are ligated in frame into pNGF, the resulting construct should express the fusion (with GFP oriented at the C-terminus of the gene of interest) at a level similar to that of the endogenous gene. Because the fusion is not overexpressed, toxic side effects are not likely, and functionality can be readily tested (see Section 9.2.4). 9.4.1.2 Integrating GFP Constructs. While centromeric plasmids such as pNGF are often able to approximate physiological expression levels of GFP fusion proteins in S. cerevisiae, experiments utilizing such constructs can still be hindered by the need to maintain plasmid selection, the variable expression of the fusion protein from cellto-cell, and/or complications/competition effects due to coexpression of the endogenous (untagged) chromosomal version of the gene (unless the experiment is performed in a knockout strain). To circumvent these remaining inconveniences, our laboratory has most often sought to express GFP fusion genes that are stably integrated into the yeast genome, most often at the 3¢ (translated C-terminus) of the chromosomal ORF for the gene of interest. As one approach to do this, we generated a plasmid based on pNGF that lacks ARS/CEN sequences—and as such can only be stably transmitted in budding yeast if integrated into the genome. By cloning a short region of the 3¢ end of the gene of interest (~200–400 bp) upstream of and in-frame with GFP in this “integrating” pNGF plasmid, and subsequently linearizing the resulting plasmid within the gene-specific sequence, integration can be targeted specifically at the 3¢ end of the ORF of interest by homologous recombination [e.g., Seedorf et al. (1999)]. As a second approach to generate integrated C-terminal fusions of GFP to genes of interest, we have also had significant success with the PCR-based method introduced by Knop et al. (1999). By this approach, a module containing GFP and a dominant selectable marker is amplified by PCR with primers that contain ~45 bp of flanking homology to the desired site of integration.

9.4.2

Assessing Expression

Several approaches can assess GFP fusion expression. The first, and most straightforward, is to look for fluorescence by microscopy. While this method will show that the GFP moiety is being accurately expressed, it does not prove that the entire fusion protein is being made. For instance, if an ORF is ligated out-of-frame into a pCGF vector, cells will fluoresce in the presence of galactose due to expression of unfused GFP. Conversely, if a protein is rapidly degraded within the cell, the GFP may not mature into its fluorescent form in time to be detected. Hence, the most accurate method of assessing expression is the immunoblot, which allows detection of immature as well as mature GFP. If antibodies against the protein being fused are available, they should recognize the fusion protein (which should run ~28 kDa larger than the unfused protein). Furthermore, such antibodies can be used to assess the relative levels of the fused to the unfused protein by the relative intensities of the signals on the blots. If antibodies to the untagged protein are not available, anti-GFP monoclonal and polyclonal antibodies are commercially available from a variety of commercial sources. To eliminate the possibility of interpreting a nonspecific anti-GFP reactive band as evidence for GFP fusion protein expression, a negative control experiment (e.g., yeast lysate without GFP) should be performed. Once fusion protein expression has been confirmed, it is additionally important to assess, if at all possible, its functionality. If a GFP fusion to a protein of interest retains

195

196

THE USES OF GREEN FLUORESCENT PROTEIN IN YEASTS

functionality, the researcher can be more confident that this reporter will accurately reflect the localizations and interactions of the endogenous (untagged) protein. A more detailed discussion of assessing GFP fusion protein functionality is provided in Section 9.2.4.

9.4.3 Microscopy Because yeast are among the smallest eukaryotes (~5–10 mm in diameter), a high level of magnification must be used in fluorescence microscopy. For visual observations, we typically use a 60¥ or 100¥ oil immersion objective lens and a 10¥ eyepiece lens. For maximal brightness and resolution, we always use objectives with a numerical aperture of 1.4 (the highest commercially available). For recording images with a digital camera, we typically set the magnification in such a manner as to project 1 mm of the specimen onto 8-15 pixels of the detector. Using a CCD camera with 6.8 ¥ 6.8-mm2 pixels, we need only a total magnification of 60¥ to achieve about 9 pixels/mm sample (or ~0.1 mm resolution). To achieve this, we use a 60¥ 1.4 N.A. oil immersion objective lens without an intermediate eyepiece or projection lens. Both yeast and many types of yeast media exhibit yellow autofluorescence when excited with ultraviolet or blue light. Thus, a fluorescence filter set that maximizes GFP detection while minimizing autofluorescence must be used. We have found that the use of a standard “barrier pass” FITC filter set (excitation 460–500 nm, Dichroic 505 nm; Barrier 510-560 nm; Chroma Technology No. #41001 or equivalent) with the S65T isolate of GFP (excitation 488 nm, emission 520 nm) gives the highest signal/noise ratio for detection. Furthermore, the use of low-fluorescence media is often advantageous. Rich media such as YPD (yeast extract, peptone, and dextrose) tends to have high-background fluorescence. Less rich media such as “synthetic complete” generally fluoresce much less brightly (Adams et al., 1997). Furthermore, media that lacks tryptophan tends to have the lowest levels of autofluorescence. Budding yeast that have mutations in the ADE1 or ADE2 genes tend to accumulate a metabolic intermediate that interferes with the observation of GFP. Under normal room lighting, colonies of ade1 or ade2 cells appear pink; when observed by epifluorescence in FITC filter sets, individual cells from these colonies appear bright yellow or green. While it has been reported that the addition of supplemental adenine to yeast media diminishes the pinkness of colonies, we have observed that this method does not completely ablate the autofluorescence observed by fluorescence microscopy. Thus, use of ade1 and ade2 strains should be avoided with GFP. GFP fusion proteins are often not affected by the presence of formaldehyde or other chemicals typically used in immunofluorescence protocols. Thus, a GFP fusion may be able to be colocalized with a protein being detected by immunofluorescence or with DNA stains such as 4¢,6-diamidino-2-phenylindole (DAPI).

9.5

CONCLUSIONS

To summarize, GFP is an extremely useful tool for studying dynamic protein, nucleic acid, and lipid localization in yeast. Furthermore, studies of organelle movement and function, protein–protein interaction, gene expression, and genetics have benefited greatly from this tool. Conversely, the use of yeast as a model system offers many advantages for expression of GFP. As a result, many of the newest uses for GFP have been developed in this

REFERENCES

model eukaryote. Undoubtedly the coming years will see even more diverse, creative, and powerful uses for this “genetic fluorophore” in yeast.

REFERENCES Adams, A., Gottschling, D. E., Kaiser, C. A., and Stearns, T. (1997). Methods in Yeast Genetics, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Bachewich, C., Thomas, D. Y., and Whiteway, M. (2003). Depletion of a polo-like kinase in Candida albicans activates cyclase-dependent hyphal-like growth. Mol. Biol. Cell 14:2163–2180. Bader, G. D., Heilbut, A., Andrews, B., Tyers, M., Hughes, T., and Boone, C. (2003). Functional genomics and proteomics: Charting a multidimensional map of the yeast cell. Trends Cell Biol. 13:344–356. Bahler, J., Wu, J. Q., Longtine, M. S., Shah, N. G., McKenzie, A., 3rd, Steever, A. B., Wach, A., Philippsen, P., and Pringle, J. R. (1998). Heterologous modules for efficient and versatile PCRbased gene targeting in Schizosaccharomyces pombe. Yeast 14:943–951. Bailer, S. M., Siniossoglou, S., Podtelejnikov, A., Hellwig, A., Mann, M., and Hurt, E. (1998). Nup116p and nup100p are interchangeable through a conserved motif which constitutes a docking site for the mRNA transport factor gle2p. EMBO J. 17:1107–1119. 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 97:11984–11989. Balla, T., Bondeva, T., and Varnai, P. (2000). How accurately can we image inositol lipids in living cells? Trends Pharmacol. Sci. 21:238–241. Balla, T., and Varnai, P. (2002). Visualizing cellular phosphoinositide pools with GFP-fused proteinmodules. Sci. STKE 2002:PL3. Basi, G., Schmid, E., and Maundrell, K. (1993). TATA box mutations in the Schizosaccharomyces pombe nmt1 promoter affect transcription efficiency but not the transcription start point or thiamine repressibility. Gene 123:131–136. Beach, D. L., and Bloom, K. (2001). ASH1 mRNA localization in three acts. Mol. Biol. Cell 12:2567–2577. Beach, D. L., Salmon, E. D., and Bloom, K. (1999). Localization and anchoring of mRNA in budding yeast. Curr. Biol. 9:569–578. Belgareh, N., and Doye, V. (1997). Dynamics of nuclear pore distribution in nucleoporin mutant yeast cells. J. Cell Biol. 136:747–759. Belmont, A. S. (2001). Visualizing chromosome dynamics with GFP. Trends Cell Biol. 11:250–257. Berman, J., and Sudbery, P. E. (2002). Candida Albicans: A molecular revolution built on lessons from budding yeast. Nat. Rev. Genet. 3:918–930. Bertrand, E., Chartrand, P., Schaefer, M., Shenoy, S. M., Singer, R. H., and Long, R. M. (1998). Localization of ASH1 mRNA particles in living yeast. Mol. Cell 2:437–445. Bevis, B. J., and Glick, B. S. (2002). Rapidly maturing variants of the Discosoma red fluorescent protein (DsRed). Nat. Biotechnol. 20:83–87. Brodsky, A. S., and Silver, P. A. (2000). Pre-mRNA processing factors are required for nuclear export. RNA 6:1737–1749. Brodsky, A. S., and Silver, P. A. (2002). Identifying proteins that affect mRNA localization in living cells. Methods 26:151–155. Browning, H., Hackney, D. D., and Nurse, P. (2003). Targeted movement of cell end factors in fission yeast. Nat. Cell Biol. 5:812–818. Bucci, M., and Wente, S. R. (1997). In vivo dynamics of nuclear pore complexes in yeast. J. Cell Biol. 136:1185–1199.

197

198

THE USES OF GREEN FLUORESCENT PROTEIN IN YEASTS

Burd, C. G., and Emr, S. D. (1998). Phosphatidylinositol(3)-phosphate signaling mediated by specific binding to RING FYVE domains. Mol. Cell 2:157–162. Campbell, R. E., Tour, O., Palmer, A. E., Steinbach, P. A., Baird, G. S., Zacharias, D. A., and Tsien, R. Y. (2002). A monomeric red fluorescent protein. Proc. Natl. Acad. Sci. USA 99:7877–7882. Carminati, J. L., and Stearns, T. (1997). Microtubules orient the mitotic spindle in yeast through dynein-dependent interactions with the cell cortex. J. Cell Biol. 138:629–641. Cherest, H., Kerjan, P., and Surdin-Kerjan, Y. (1987). The Saccharomyces cerevisiae MET3 gene: Nucleotide sequence and relationship of the 5¢ non-coding region to that of MET25. Mol. Gen. Genet. 210:307–313. Chi, Y., Huddleston, M. J., Zhang, X., Young, R. A., Annan, R. S., Carr, S. A., and Deshaies, R. J. (2001). Negative regulation of Gcn4 and Msn2 transcription factors by Srb10 cyclin-dependent kinase. Genes Dev. 15:1078–1092. Cormack, B. P., Bertram, G., Egerton, M., Gow, N. A., Falkow, S., and Brown, A. J. (1997). Yeastenhanced green fluorescent protein (yEGFP)a reporter of gene expression in Candida albicans. Microbiology 143 (Pt 2):303–311. Corral-Debrinski, M., Blugeon, C., and Jacq, C. (2000). In yeast, the 3¢ untranslated region or the presequence of ATM1 is required for the exclusive localization of its mRNA to the vicinity of mitochondria. Mol. Cell Biol. 20:7881–7892. Craven, R. A., Griffiths, D. J., Sheldrick, K. S., Randall, R. E., Hagan, I. M., and Carr, A. M. (1998). Vectors for the expression of tagged proteins in Schizosaccharomyces pombe. Gene 221:59– 68. Damelin, M., and Silver, P. A. (2000). Mapping interactions between nuclear transport factors in living cells reveals pathways through the nuclear pore complex. Mol. Cell 5:133–140. Damelin, M., and Silver, P. A. (2002). In situ analysis of spatial relationships between proteins of the nuclear pore complex. Biophys. J. 83:3626–3636. De Backer, M. D., Magee, P. T., and Pla, J. (2000). Recent developments in molecular genetics of Candida albicans. Annu. Rev. Microbiol. 54:463–498. Faryar, K., and Gatz, C. (1992). Construction of a tetracycline-inducible promoter in Schizosaccharomyces pombe. Curr. Genet. 21:345–349. Gasser, S. M. (2002). Visualizing chromatin dynamics in interphase nuclei. Science 296:1412–1416. Gerami-Nejad, M., Berman, J., and Gale, C. A. (2001). Cassettes for PCR-mediated construction of green, yellow, and cyan fluorescent protein fusions in Candida albicans. Yeast 18:859– 864. Gorner, W., Durchschlag, E., Martinez-Pastor, M. T., Estruch, F., Ammerer, G., Hamilton, B., Ruis, H., and Schuller, C. (1998). Nuclear localization of the C2H2 zinc finger protein Msn2p is regulated by stress and protein kinase A activity. Genes Dev. 12:586–597. Gorner, W., Durchschlag, E., Wolf, J., Brown, E. L., Ammerer, G., Ruis, H., and Schuller, C. (2002). Acute glucose starvation activates the nuclear localization signal of a stress-specific yeast transcription factor. EMBO J. 21:135–144. Hailey, D. W., Davis, T. N., and Muller, E. G. (2002). Fluorescence resonance energy transfer using color variants of green fluorescent protein. Methods. Enzymol. 351:34–49. Hammell, C. M., Gross, S., Zenklusen, D., Heath, C. V., Stutz, F., Moore, C., and Cole, C. N. (2002). Coupling of termination, 3¢ processing, and mRNA export. Mol. Cell Biol. 22:6441–6457. Hazan, I., Sepulveda-Becerra, M., and Liu, H. (2002). Hyphal elongation is regulated independently of cell cycle in Candida albicans. Mol. Biol. Cell 13:134–145. Heun, P., Laroche, T., Shimada, K., Furrer, P., and Gasser, S. M. (2001). Chromosome dynamics in the yeast interphase nucleus. Science 294:2181–2186. Hoepfner, D., Brachat, A., and Philippsen, P. (2000). Time-lapse video microscopy analysis reveals astral microtubule detachment in the yeast spindle pole mutant cnm67. Mol. Biol. Cell 11:1197–1211.

REFERENCES

Hoepfner, D., van den Berg, M., Philippsen, P., Tabak, H. F., and Hettema, E. H. (2001). A role for Vps1p, actin, and the Myo2p motor in peroxisome abundance and inheritance in Saccharomyces cerevisiae. J. Cell Biol. 155:979–990. Hoffman, C. S., and Winston, F. (1989). A transcriptionally regulated expression vector for the fission yeast Schizosaccharomyces pombe. Gene 84:473–479. Huh, W. K., Falvo, J. V., Gerke, L. C., Carroll, A. S., Howson, R. W., Weissman, J. S., and O’Shea, E. K. (2003). Global analysis of protein localization in budding yeast. Nature 425:686–691. Hurt, E., Hannus, S., Schmelzl, B., Lau, D., Tollervey, D., and Simos, G. (1999). A novel in vivo assay reveals inhibition of ribosomal nuclear export in ran-cycle and nucleoporin mutants. J. Cell Biol. 144:389–401. Jacquet, M., Renault, G., Lallet, S., De Mey, J., and Goldbeter, A. (2003). Oscillatory nucleocytoplasmic shuttling of the general stress response transcriptional activators Msn2 and Msn4 in Saccharomyces cerevisiae. J. Cell Biol. 161:497–505. Kahana, J. A., Schnapp, B. J., and Silver, P. A. (1995). Kinetics of spindle pole body separation in budding yeast. Proc. Natl. Acad. Sci. USA 92:9707–9711. Kahana, J. A., and Silver, P. A. (1996). Use of the A. Victoria green fluorescent protein to study protein dynamics in vivo. In Current Protocols in Molecular Biology, Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. E., Seidman, J. G., Smith, J. A., and Struhl, K., Eds., (New York, NY, John Wiley & Sons, Inc.), pp. 9.7.22–29.27.27. Karin, M., Najarian, R., Haslinger, A., Valenzuela, P., Welch, J., and Fogel, S. (1984). Primary structure and transcription of an amplified genetic locus: The CUP1 locus of yeast. Proc Natl Acad Sci USA 81:337–341. Knop, M., Siegers, K., Pereira, G., Zachariae, W., Winsor, B., Nasmyth, K., and Schiebel, E. (1999). Epitope tagging of yeast genes using a PCR-based strategy: More tags and improved practical routines. Yeast 15:963–972. Kohlwein, S. D. (2000). The beauty of the yeast: Live cell microscopy at the limits of optical resolution. Microsc. Res. Tech. 51:511–529. Laney, J. D., and Hochstrasser, M. (2003). Ubiquitin-dependent degradation of the yeast Mat(alpha)2 repressor enables a switch in developmental state. Genes Dev 17:2259–2270. Li, J., Wang, S., VanDusen, W. J., Schultz, L. D., George, H. A., Herber, W. K., Chae, H. J., Bentley, W. E., and Rao, G. (2000). Green fluorescent protein in Saccharomyces cerevisiae: real-time studies of the GAL1 promoter. Biotechnol. Bioeng. 70:187–196. Lippincott-Schwartz, J., and Patterson, G. H. (2003). Development and use of fluorescent protein markers in living cells. Science 300:87–91. Longtine, M. S., McKenzie, A., 3rd, Demarini, D. J., Shah, N. G., Wach, A., Brachat, A., Philippsen, P., and Pringle, J. R. (1998). Additional modules for versatile and economical PCR-based gene deletion and modification in Saccharomyces cerevisiae. Yeast 14:953–961. Maddox, P. S., Bloom, K. S., and Salmon, E. D. (2000). The polarity and dynamics of microtubule assembly in the budding yeast Saccharomyces cerevisiae. Nat. Cell Biol. 2:36–41. Malinska, K., Malinsky, J., Opekarova, M., and Tanner, W. (2003). Visualization of protein compartmentation within the plasma membrane of living yeast cells. Mol. Biol. Cell 14:4427–4436. Mallavarapu, A., Sawin, K., and Mitchison, T. (1999). A switch in microtubule dynamics at the onset of anaphase B in the mitotic spindle of Schizosaccharomyces pombe. Curr. Biol. 9:1423–1426. Marc, P., Margeot, A., Devaux, F., Blugeon, C., Corral-Debrinski, M., and Jacq, C. (2002). Genomewide analysis of mRNAs targeted to yeast mitochondria. EMBO Rep. 3:159–164. Mateus, C., and Avery, S. V. (2000). Destabilized green fluorescent protein for monitoring dynamic changes in yeast gene expression with flow cytometry. Yeast 16:1313–1323. Matz, M. V., Fradkov, A. F., Labas, Y. A., Savitsky, A. P., Zaraisky, A. G., Markelov, M. L., and Lukyanov, S. A. (1999). Fluorescent proteins from nonbioluminescent Anthozoa species. Nat. Biotechnol. 17:969–973.

199

200

THE USES OF GREEN FLUORESCENT PROTEIN IN YEASTS

Maundrell, K. (1990). nmt1 of fission yeast. A highly transcribed gene completely repressed by thiamine. J. Biol. Chem. 265:10857–10864. Maundrell, K. (1993). Thiamine-repressible expression vectors pREP and pRIP for fission yeast. Gene 123:127–130. Messerschmitt, M., Jakobs, S., Vogel, F., Fritz, S., Dimmer, K. S., Neupert, W., and Westermann, B. (2003). The inner membrane protein Mdm33 controls mitochondrial morphology in yeast. J. Cell Biol. 160:553–564. Morschhauser, J., Michel, S., and Hacker, J. (1998). Expression of a chromosomally integrated, single-copy GFP gene in Candida albicans, and its use as a reporter of gene regulation. Mol. Gen. Genet. 257:412–420. Nabeshima, K., Nakagawa, T., Straight, A. F., Murray, A., Chikashige, Y., Yamashita, Y. M., Hiraoka, Y., and Yanagida, M. (1998). Dynamics of centromeres during metaphase-anaphase transition in fission yeast: Dis1 is implicated in force balance in metaphase bipolar spindle. Mol. Biol. Cell 9:3211–3225. Onishi, M., Koga, T., Morita, R., Nakamura, Y., Nakamura, T., Shimoda, C., Takegawa, K., Hirata, A., and Fukui, Y. (2003). Role of phosphatidylinositol 3–phosphate in formation of forespore membrane in Schizosaccharomyces pombe. Yeast 20:193–206. Overton, M. C., and Blumer, K. J. (2000). G-protein-coupled receptors function as oligomers in vivo. Curr. Biol. 10:341–344. Overton, M. C., Chinault, S. L., and Blumer, K. J. (2003). Oligomerization, biogenesis, and signaling is promoted by a glycophorin A-like dimerization motif in transmembrane domain 1 of a yeast G protein-coupled receptor. J. Biol. Chem. 278:49369–49377. Park, N. H., and Choi, W. (2002). Vectors designed for efficient molecular manipulation in Candida albicans. Yeast 19:1057–1066. Pidoux, A. L., Uzawa, S., Perry, P. E., Cande, W. Z., and Allshire, R. C. (2000). Live analysis of lagging chromosomes during anaphase and their effect on spindle elongation rate in fission yeast. J. Cell Sci. 113 (Pt 23):4177–4191. Prein, B., Natter, K., and Kohlwein, S. D. (2000). A novel strategy for constructing N-terminal chromosomal fusions to green fluorescent protein in the yeast Saccharomyces cerevisiae. FEBS Lett. 485:29–34. Prinz, W. A., Grzyb, L., Veenhuis, M., Kahana, J. A., Silver, P. A., and Rapoport, T. A. (2000). Mutants affecting the structure of the cortical endoplasmic reticulum in Saccharomyces cerevisiae. J. Cell Biol. 150:461–474. Roberts, P. M., and Goldfarb, D. S. (1998). In vivo nuclear transport kinetics in Saccharomyces cerevisiae. Methods Cell Biol. 53:545–557. Rothstein, R. (1991). Targeting, disruption, replacement, and allele rescue: Integrative DNA transformation in yeast. Methods Enzymol. 194:281–301. Ruohonen, L., Aalto, M. K., and Keranen, S. (1995). Modifications to the ADH1 promoter of Saccharomyces cerevisiae for efficient production of heterologous proteins. J. Biotechnol. 39:193–203. Ryan, K. J., McCaffery, J. M., and Wente, S. R. (2003). The Ran GTPase cycle is required for yeast nuclear pore complex assembly. J. Cell Biol. 160:1041–1053. Ryan, K. J., and Wente, S. R. (2002). Isolation and characterization of new Saccharomyces cerevisiae mutants perturbed in nuclear pore complex assembly. BMC Genet 3:17. Seedorf, M., Damelin, M., Kahana, J., Taura, T., and Silver, P. A. (1999). Interactions between a nuclear transporter and a subset of nuclear pore complex proteins depend on Ran GTPase. Mol. Cell Biol. 19:1547–1557. Shepard, K. A., Gerber, A. P., Jambhekar, A., Takizawa, P. A., Brown, P. O., Herschlag, D., DeRisi, J. L., and Vale, R. D. (2003). Widespread cytoplasmic mRNA transport in yeast: Identification

REFERENCES

of 22 bud-localized transcripts using DNA microarray analysis. Proc. Natl. Acad. Sci. USA 100:11429–11434. Sheth, U., and Parker, R. (2003). Decapping and decay of messenger RNA occur in cytoplasmic processing bodies. Science 300:805–808. Shulga, N., Roberts, P., Gu, Z., Spitz, L., Tabb, M. M., Nomura, M., and Goldfarb, D. S. (1996). In vivo nuclear transport kinetics in Saccharomyces cerevisiae: A role for heat shock protein 70 during targeting and translocation. J. Cell Biol. 135:329–339. Stade, K., Ford, C. S., Guthrie, C., and Weis, K. (1997). Exportin 1 (Crm1p) is an essential nuclear export factor. Cell 90:1041–1050. Stage-Zimmermann, T., Schmidt, U., and Silver, P. A. (2000). Factors affecting nuclear export of the 60S ribosomal subunit in vivo. Mol. Biol. Cell 11:3777–3789. Stefan, C. J., Audhya, A., and Emr, S. D. (2002). The yeast synaptojanin-like proteins control the cellular distribution of phosphatidylinositol (4,5)-bisphosphate. Mol. Biol. Cell 13:542–557. Tatchell, K., and Robinson, L. C. (2002). Use of green fluorescent protein in living yeast cells. Methods Enzymol. 351:661–683. Taura, T., Krebber, H., and Silver, P. A. (1998). A member of the Ran-binding protein family, Yrb2p, is involved in nuclear protein export. Proc. Natl. Acad. Sci. USA 95:7427–7432. Thrower, D. A., Stemple, J., Yeh, E., and Bloom, K. (2003). Nuclear oscillations and nuclear filament formation accompany single-strand annealing repair of a dicentric chromosome in Saccharomyces cerevisiae. J. Cell Sci. 116:561–569. Tsien, R. Y. (1998). The green fluorescent protein. Annu. Rev. Biochem. 67:509–544. Walmsley, R. M., Billinton, N., and Heyer, W. D. (1997). Green fluorescent protein as a reporter for the DNA damage-induced gene RAD54 in Saccharomyces cerevisiae. Yeast 13:1535–1545. Westermann, B., and Neupert, W. (2000). Mitochondria-targeted green fluorescent proteins: Convenient tools for the study of organelle biogenesis in Saccharomyces cerevisiae. Yeast 16:1421–1427. Yocum, R. R., and Johnston, M. (1984). Molecular cloning of the GAL80 gene from Saccharomyces cerevisiae and characterization of a gal80 deletion. Gene 32:75–82.

201

10 USES OF GFP IN CAENORHABDITIS ELEGANS Oliver Hobert and Paula Loria Department of Biochemistry and Molecular Biophysics, Center for Neurobiology and Behavior, Columbia University, College of Physicians and Surgeons, New York, NY

10.1

INTRODUCTION

Green fluorescent protein has become one of the standard tools of research in C. elegans. Its use in C. elegans appears even more widespread than in other metazoan model organisms, including Drosophila, not just because GFP was first introduced into C. elegans (Chalfie et al., 1994), but because several characteristics of C. elegans specifically favor the use of GFP. A key advantage of C. elegans for the use of GFP is the animal’s transparency and thin diameter (500 bp that overlaps with the recombination substrate. This recombination substrate is a second PCR fragment amplified from worm genomic DNA. Both PCR products are co-injected. As a control that the resultant GFP expression is indeed a result of recombination, the second PCR fragment is also injected alone. The only limiting factor for the technique is the size of the genomic PCR product. In that regard, it is virtually impossible to beat the yeast homologous recombination technique, in which one can recombine GFP, at least in theory, into several hundred kilobase YAC clones. One pitfall to the in vivo recombination technique is that one has no control over whether the in vivo recombination has worked successfully, except in those cases where a selectable scheme is employed (i.e., rescue a mutant phenotype with a GFP-tagged locus that has been generated through in vivo recombination). Alternatively, one can visualize successful recombination if one knows that only the recombined DNA can produce a protein that is correctly localized to specific subcellular sites.

TECHNICAL ASPECTS

2. Recombination Cloning (Hawkins et al., 2003). While being more timeconsuming than the PCR fusion method, this technique—which relies on an elegant yeast recombination scheme (Fig. 10.5B)—has the invaluable advantage of being able to (a) handle virtually unlimited sizes of genomic DNA and (b) allow one to drop GFP anywhere within the genomic locus (the latter could in theory also be achieved in a triple-PCR fusion approach). 3. In Vivo Recombination Approach (Tsalik et al., 2003; Fig. 10.5C). This approach relies on the observation that overlapping pieces of DNA injected into C. elegans will undergo efficient homologous recombination, provided that the homologous region is >500 bp (Maryon et al., 1998; Mello et al., 1991). The approach uses a technical combination of the two above approaches and combines several of their strengths—that is, speed and fewer restrictions in terms of size (see legend to Fig. 10.5C). Creation of Transgenic Lines that Express GFP Fusion Proteins. DNA injected into the gonad of a worm will form extrachromosomal arrays that contain multiple copies of the injected DNA (Jin, 1999; Mello et al., 1991). The multicopy nature of the arrays has advantages in the form of potentially offering a higher level of sensitivity, but also disadvantages such as potential overexpression artifacts, gene silencing, or promoter-titration artifacts. Since extrachromosomal arrays created by DNA injection are not stably transmitted through mitosis (used as an advantage in mosaic analysis mentioned above), one often induces these arrays to become integrated into chromosomes through the use of highenergy irradiation. A problem of this integration approach is that its high mutagenicity rate necessitates thorough backcrossing following irradiation. Due to the repetitive structure of arrays and problems of gene silencing associated with it, mosaicism may even be observed with chromosomally integrated arrays (Hsieh et al., 1999). New developments in the field of array formation have alleviated some of these problems. The generation of “complex arrays” through the injection of the desired reporter gene, together with a complex mixture of heterologous DNA, allows expression of a transgene in the germline, which could not be observed with regular arrays (Kelly et al., 1997). In addition, reporters expressed from “complex arrays” appear less mosaic and more stably expressed. Moreover, DNA delivery to C. elegans via microparticle bombardment can also yield expression in the germline, likely due to spontaneous chromosomal integration of low-copy number arrays (Praitis et al., 2001). The low-copy number of the reporter gene delivered by particle bombardment has the added advantage of yielding expression levels of the reporter construct that are more likely to reflect endogenous gene expression levels, thus minimizing the potential problems arising from reporter gene overexpression mentioned in previous chapters. Visualization of GFP. The level of expression of GFP can in some cases be so low that GFP fluoresence cannot be observed. In some documented cases, the presence of GFP was nevertheless revealed through the use of antibody staining against GFP, thus illustrating the higher sensitivity of antibody-staining procedures (Levitan and Greenwald, 1998). Common Artifacts. Notable fluorescent artifacts are the autofluorescence of particles in the gut and the nucleoli of hypodermal cells. Another artifact commonly observed

219

220

USES OF GFP IN CAENORHABDITIS ELEGANS

is posterior gut fluorescence. Unlike the autofluorescence of the gut and hypodermal nucleoli, the source of this fluorescence clearly is GFP. The GFP signal may be a consequence of the canonical 3¢UTR derived from the unc-54 gene, present in the commonly used GFP vectors. It is curious to note that this posterior gut expression becomes more prevalent when less regulatory information is provided in the sequences fused 5¢ to the GFP coding sequences. Co-injected DNA (e.g., a gfp and rfp reporter) as well as extrachromosomal arrays can also show poorly understood interactions among themselves. For example, we observe that they can sometimes silence each other’s expression. In these cases, it is advisable to use independently created and chromosomally integrated arrays and to cross strains that carry these integrated arrays. Lastly, another inherent problem of GFP is the need for chromophore maturation, which makes its difficult to temporally correlate the onset of GFP fluorescence to the onset of gene expression as assessed by GFP reporters. Induction of GFP expression under a strong, heat-inducible promoter yields visible fluorescence in 95% cells are GFP-positive (Megason and McMahon, 2002). Expression typically fades in 3–4 days. GFP can be coexpressed from a ubiquitous enhancer/promoter such as CMV/b-actin along with a gene of interest utilizing an IRES (internal ribosome entry sequence) to mark which cells were transfected since the expression is mosaic (Megason and McMahon, 2002). Tissue-specific enhancers can also be used in electroporation (Itasaki et al., 1999). Electroporation has been most widely used in chick, but it is also applicable to other species including mice (Itasaki et al., 1999), ascidians (Corbo et al., 1997), zebrafish (Swartz et al., 2001; Tawk et al., 2002), and Xenopus (Eide et al., 2000). The quickness, ease, and flexibility of electroporation make it a very promising technique for transient transgenesis.

13.4.4

Mice

13.4.4.1 DNA Injection. Pronuclear injection has been a valuable technique for generating transgenic mice for over 20 years (Gordon and Ruddle, 1981). DNA injected into the pronuclei of mouse zygotes integrates to generate stable, nonmosaic transgenics at fairly high efficiency (10–50%). A number of lines of mice have been generated that ubiquitously express GFP (Takada et al., 1997; Okabe et al., 1997; Chiocchetti et al., 1997).

METHODS FOR CREATING TRANSGENICS IN DIFFERENT VERTEBRATE SPECIES

Figure 13.2. GFP expression in transgenic mouse and chick. (A, B) GFP expression in neural tube and neural crest following electroporation of chick with a GFP encoding plasmid (green) and anti-HNK1 immunostaining (red) to mark neural crest. (A) Lateral view of whole mount (Maria Elena de Bellard and Marianne Bronner-Fraser, unpublished). (B) Cross section through neural tube with DAPI staining (blue) to mark nuclei (Ed Coles and Marianne Bronner-Fraser, unpublished observations). (C) Yolk sac of an E9.5 transgenic mouse showing e–globin GFP expression in red blood cells (Dyer et al., 2001; Elizabeth Jones, unpublished observations). (D) Section through cerebellum of Calbindin BAC GFP transgenic mouse showing expression in Purkinje cells (Xiangdong William Yang and Nat Heintz, unpublished observations). See color insert.

Over the last several years, over 100 lines have also been generated that express GFP in a tissue-restricted manner (Fig. 13.2C). As in fish, using BACs to make GFP transgenics is also advantageous in mice to increase the likelihood of the transgene containing all of the elements required for proper expression of GFP (Yang et al., 1997) (Fig. 13.2D). A large-scale effort directed by Nathaniel Heintz, Mary-Beth Hatten, and Alexandra Joyner is currently underway to target GFP to a large number of BACs, generate transgenics for the BACs, and analyze their expression patterns (Gong et al., 2003). The project aims to analyze 1000 genes per year. This project has the potential to greatly benefit our knowledge of gene expression and provide a valuable collection of modified BACs. 13.4.4.2 Knock-in. Transgenic mice can also be generated using gene targeting in embryonic stem cells to “knock-in” GFP into a locus of interest. Knock-ins generally require more work to create than do transgenics created through pronuclear injection, but knock-ins allow for more precise control of the transgene. In knock-ins, GFP is inserted into the endogenous loci, ensuring that all of the regulatory elements required for proper expression of the gene of interest are present. Sequences inserted into a locus via target-

295

296

USES OF GFP IN TRANSGENIC VERTEBRATES

ing such as selection cassettes can interfere with the regulation of the locus, so care must be taken that interfering sequences are removed in the ES cells prior to generating the mice. Selection cassettes used in targeting can be removed using the cre-loxP system or through retargeting while using negative selection. GFP knock-ins in mice can be designed in several ways: GFP can be expressed in place of the endogenous gene such that the endogenous gene is knocked out (Godwin et al., 1998); GFP can be expressed as a protein fusion with the endogenous gene (Kundu et al., 2002a,b); or GFP can be expressed from an internal ribosome entry sequence (IRES) such that both GFP and the endogenous gene are expressed.

13.5 13.5.1

USES OF GFP TRANSGENICS IN VERTEBRATES GFP as a Coexpression Marker

GFP transgenics can be used for a variety purposes in vertebrates. Perhaps the most straightforward use of GFP is as a coexpression marker. In this scenario a transgenic is created that expresses both GFP and a gene of interest. The individual animals and cells expressing the gene of interest can then be identified simply and prescisely by visualizing GFP. Many forms of transgenesis such as electroporation result in mosaic expression. Coelectroporation a GFP expression plasmid along with an expression plasmid of interest has been used to identify which regions of the embryo were successfully transfected (Araki and Nakamura, 1999). For single-cell resolution of which cells were transfected by in ovo electroporation, a construct containing a gene of interest followed by an IRES-GFP cassette can be used (Megason and McMahon, 2002). GFP can also be used for identifying which animals are transgenic when generating transgenic vertebrates. In transgenic Xenopus embryos made using sperm nuclear transfer using both a GFP expression construct and another construct of interest, 94% of the transgenic embryos contained both transgenes cointegrated (Hartley et al., 2001). Double promoter plasmids containing a gene and promoter of interest along with GFP under the control of the crystalline promoter have been used in Xenopus to identify transgenics by looking for GFP in their eyes (FU et al., 2002). Cointegration of a detectable marker with the transgene of interest can also be used in mice to identify transgenics (Overbeek et al., 1991). GFP can also be used for genotyping and sex-typing: By generating mice embryos using a father containing a GFP insertion on his X-chromosome, the sex of embryos was determined noninvasively at embryonic day 2.75 by GFP visualization long before overt sexual differentiation occurs at E12.5 (Hadjantonakis et al., 1998).

13.5.2

GFP as a Marker for Cell Types

GFP can also be used as a marker for a cell type of interest by using an enhancer that drives expression of GFP in that cell type in transgenics. Currently, immunhistochemistry with cell-type-specific antibodies and in situ hybridization with cell-type specific probes are the standard methods used for marker analysis. These techniques are advantageous in that they allow a number of different markers to be assayed, but they can only be used on fixed specimens. Using GFP transgenics as markers allows cell types of interest to be studied in living specimens. GFP expressed under control of the Oct4 enhancer was used to mark primordial germ cells in the mouse (Anderson et al., 2000). Time-lapse imaging revealed that PGCs originate from the posterior primitive streak and begin migrating toward the future site of the allantois. Using GFP as a marker for cell types also allows

USES OF GFP TRANSGENICS IN VERTEBRATES

cellular morphology to be visualized at much higher resolution than by immunhistochemistry or in situ hybrization. GFP transgenics are particularly useful for marking neural cell types. Feng and colleagues used the Thy1 promoter to generate transgenics using four different spectral variants of GFP (Feng et al., 2000). They generated 25 different lines that each marked different populations of neurons presumably due to transgene integration effects. GFP diffused within expressing neurons to beautifully mark the entire cell from dentrititic spines to the nerve terminal of axons several centimeters long. GFP transgenics can also be used to mark cell types of interest to be purified by fluorescentactivated cell sorting (FACS). Transgenic mice expressing GFP under the control of the L7 promoter were used to purify live Purkinje cells (Tomomura et al., 2001). A knock-in of GFP into the Hoxa13 locus in mice was used to purify limb mesenchymal cells that were then used to show Hoxa13 null cells are defective in forming chondrogenic condensations in vitro (Stadler et al., 2001).

13.5.3

GFP as a Marker for Gene Expression Patterns

GFP transgenics can also be used to study gene expression patterns by placing GFP under the control of the regulatory elements of a gene of interest. This technique is similar to using GFP to mark cell types, but the focus is on the gene being marked rather than the cell type being marked. Knock-ins and BAC transgenics are the preferred method for marking a gene expression pattern with GFP because these methods are more likely to result in transgenics that faithfully recapitulate the expression pattern of the gene of interest compared to plasmid-based transgenics. GFP transgenics are advantageous over in situ hybrization for marking gene expression in several ways. Since GFP can reveal gene expression patterns in live animals, they could make it easier to study rapid and dynamic changes in gene expression during development, such as the oscillation of gene expression in the presegmental plate during somitogenesis, or in response to exogenous factors. Once a GFP transgenic line is established, it allows gene expression to be assayed much more easily than in situ hybridization. GFP transgenics also provide better spatial and temporal resolution of gene expression than does in situ hybridization because GFP transgenics can be imaged at cellular resolution continuously over development using time-lapse, confocal microscopy. However, GFP transgenics have some potential problems for marking gene expression patterns relative to in situ hybridization. One must ensure that GFP from a transgenic is expressed in the same pattern as the gene being marked, usually by comparison with in situ hybridization at individual time points. GFP takes 1–2 h to become fluorescent after its transcription is initiated. Because of the long half-life of normal GFP protein, GFP fluorescence can remain long after its transcription has ended, although destabilized variants of GFP may reduce this problem (Li et al., 1998). These effects can cause shifts in the timing of GFP fluorescence relative to its transcription. As described above, a large-scale effort is currently underway to analyze the gene expression patterns of thousands of genes in mice using BAC GFP transgenics. Similar efforts may also be performed in the coming years in zebrafish and Xenopus. Gene trapping using GFP is also being used to mark expression patterns from a number of genes in Xenopus (Bronchain et al., 1999) and zebrafish (Kawakami et al., 2004; Parinor et al., 2004; Balciunas et al., 2004).

13.5.4

GFP Transgenics for Enhancer Analysis

GFP transgenics can be used for mapping the regulatory elements that control the expression of a gene. Enhancer analysis is usually begun by isolating a large DNA fragment from

297

298

USES OF GFP IN TRANSGENIC VERTEBRATES

the locus being studied that is capable of directing expression of GFP in the proper pattern in transgenics. Successively smaller fragments are then used to map the location and tissue-specificity of different positive and negative regulatory elements in the enhancer. GFP transgenics were used to map the regulatory elements of the GATA1, GATA2, and rag1 loci in zebrafish (Long et al., 1997; Meng et al., 1997; Jessen et al., 1999). Data obtained from a number of transient transgenics can be pooled to simplify this analysis in zebrafish (Long et al., 1997). GFP transgenics were used in Xenopus to identify separate regulatory elements that direct endodermal and mesodermal expression of the transcription factor HNF3a (Ryffel and Lingott, 2000). Electroporation in chick can also be used for enhancer analysis (Itasaki et al., 1999). Electroporation in chick can be done using mouse enhancers and is much easier, quicker, and cheaper than creating transgenics in mice (Timmer et al., 2001). Electroporation in chick can be used for initial dissection of an enhancer followed by confirmation using mouse transgenics.

13.5.5

GFP Fusions for Examining Protein Function in Vivo

Although not yet fully realized, a potentially exciting use of GFP in transgenic vertebrates is the use of GFP protein fusions for studying the in vivo function of proteins. In this method the coding sequence of GFP is fused to the coding sequence of another protein to generate a chimeric protein containing both GFP and the protein of interest. If designed correctly, GFP fusions are often functional because of the compact, monomeric nature of GFP. Long et al. (2000) used transgenic zebrafish expressing a death receptor/GFP fusion protein to demonstrate the critical role of this protein in negative regulation of erythropoiesis. One use of GFP fusions is to examine the subcellular distribution of a protein in vivo. Many proteins change their subcellular distribution depending on their functional state. GFP fusions were used to show that the Wnt signaling component Axin redistributes from the cytoplasm to the membrane in response to Wnt signaling (Cliffe et al., 2003). GFP fusions were also used to show that the hedgehog signaling component Smoothened redistributes from internal compartments to the cell surface upon hedgehog signaling (Zhu et al., 2003). Functional GFP fusions can even be formed for secreted signaling molecules. A fusion of decapentaplegic to GFP was used to visualize the fromation of a morphogen gradient in the fly wing disc (Teleman and Cohen, 2000). GFP fusions are particularly useful in neurosecience. Protein fusions to ECFP and EYFP have been used in conjunction with fluorescent resonant energy transfer (FRET) to detect activation of G-proteincoupled potassium channels (Riven et al., 2003). pH-, Ca2+-, and voltage-sensitive variants of GFP generated through protein fusions can be used for monitoring neural activity noninvasively (Miesenbock et al., 1998; Miyawaki et al., 1999; Sakai et al., 2001). A novel GFP fusion protein can be used for mapping neural connectivity: A fusion of GFP to a nontoxic fragment of tetanus toxin is transferred across synapses in a retrograde direction in transgenic mice, allowing for mapping of neural circuits using transgenics (Maskos et al., 2002).

13.6

CONCLUSION

It has been less than a decade since the first use of GFP as a marker (Chalfie et al., 1994), yet this little protein has already revolutionized many areas of biology. Parallel advances in imaging and genetic manipulation over the past decade have further benefited the use of GFP. Because of the extra time and expense involved in using vertebrates compared to

REFERENCES

using invertebrates or in vitro approaches, many techniques for using GFP have only recently been applied to vertebrates. Already though, the use of GFP in vertebrates has contributed immensely to investigations into both their embryonic development and function as adults. The coming decade will undoubtedly see a dramatic broadening in the use of currently available GFP techniques in vertebrates in addition to the development of novel GFP techniques for studying biology.

REFERENCES Anderson, R., Copeland, T. K., Scholer, H., Heasman, J., and Wylie, C. (2000). The onset of germ cell migration in the mouse embryo. Mech. Dev. 91:61–68. Amsterdam, A., Lin, S., and Hopkins, N. (1995). The Aequorea victoria green fluorescent protein can be used as a reporter in live zebrafish embryos. Dev. Biol. 171:123–129. Araki, I., and Nakamura, H. (1999). Engrailed defines the position of dorsal di-mesencephalic boundary by repressing diencephalic fate. Development 126:5127–5135. Balciunas, D., Davidson, A. E., Sivasubbu, S., Hermanson, S. B., Welle, Z., and Ekker, S. C. (2004) Enhancer trapping in zebrafish using the Sleeping Beauty transposon. BMC Genomics 5:62–76. Boerkoel, C. F., Federspiel, M. J., Salter, D. W., Payne, W., Crittenden, L. B., Kung, H. J., and Hughes, S. H. (1993). A new defective retroviral vector system based on the Bryan strain of Rous sarcoma virus. Virology 195:669–679. Bordignon, V., Keyston, R., Lazaris, A., Bilodeau, A. S., Pontes, J. H., Arnold, D., Fecteau, G., Keefer, C., and Smith, L. C. (2003). Transgene expression of green fluorescent protein and germ line transmission in cloned calves derived from in vitro-transfected somatic cells. Biol. Reprod. 68:2013–2023. Bronchain, O. J., Hartley, K. O., and Amaya, E. (1999). A gene trap approach in Xenopus. Curr. Biol. 9:1195–1198. Campbell, K. H., McWhir, J., Ritchie, W. A., and Wilmut, I. (1996). Sheep cloned by nuclear transfer from a cultured cell line. Nature 380:64–66. Campbell, R. E., Tour, O., Palmer, A. E., Steinbach, P. A., Baird, G. S., Zacharias, D. A., and Tsien, R. Y. (2002). A monomeric red fluorescent protein. Proc. Natl. Acad. Sci. USA 99: 7877–7882. Chae, J., Zimmerman, L. B., and Grainger, R. M. (2002). Inducible control of tissue-specific transgene expression in Xenopus tropicalis transgenic lines. Mech. Dev. 117:235–241. 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. Chiocchetti, A., Tologano, E., Hirsch, E., Silengo, L., and Altruda, F. (1997). Green fluorescent protein as a reporter of gene expression in transgenic mice. Biochim. Biophys. Acta 1352:193–202. Cliffe, A., Hamada, F., and Bienz, M. (2003). A role of dishevelled in relocating axin to the plasma membrane during wingless signaling. Curr. Biol. 13:960–966. Colleaux, L., D’Auriol, L., Galibert, F., and Dujon, B. (1988). Recognition and cleavage site of the intron-encoded omega transposase. Proc. Natl. Acad. Sci. USA 85:6022–6026. Corbo, J. C., Levine, M., and Zeller, R. W. (1997). Characterization of a notochord-specific enhancer from the Brachyury promoter region of the ascidian, Ciona intestinalis. Development 124:589–602. Culp, P., Nusslein-Volhard, C., and Hopkins, N. (1991). High-frequency germ-line transmission of plasmid DNA sequences injected into fertilized zebrafish eggs. Proc. Natl. Acad. Sci. USA 88:7953–7957.

299

300

USES OF GFP IN TRANSGENIC VERTEBRATES

Cross, L. M., Cook, M. A., Lin, S., Chen, J. N., and Rubinstein, A. L. (2003). Rapid analysis of angiogenesis drugs in a live fluorescent zebrafish assay. Arterioscler. Thromb. Vasc. Biol. 23(5):911–912. Dai, C., Celestino, J. C., Okada, Y., Louis, D. N., Fuller, G. N., and Holland, E. C. (2001). PDGF autocrine stimulation dedifferentiates cultured astrocytes and induces oligodendrogliomas and oligoastrocytomas from neural progenitors and astrocytes in vivo. Genes Dev. 15:1913–1925. Downes, G. B., Waterbury, J. A., and Granato, M. (2002). Rapid in vivo labeling of identified zebrafish neurons. Genesis 34:196–202. Dyer, M. A., Farrington, S. M., Mohn, D., Munday, J. R., and Baron, M. H. (2001). Indian hedgehog activates hematopoiesis and vasculogenesis and can respecify prospective neurectodermal cell fate in the mouse embryo. Development 128:1717–1730. Eide, F. F., Eisenberg, S. R., and Sanders, T. A. (2000). Electroporation-mediated gene transfer in free-swimming embryonic Xenopus laevis. FEBS Lett. 486:29–32. Etkin, L. D., and Pearman, B. (1987). Distribution, expression and germ line transmission of exogenous DNA sequences following microinjection into Xenopus laevis eggs. Development 99:15–23. Feng, G., Mellor, R. H., Bernstein, M., Keller-Peck, C., Nguyen, Q. T., Wallace, M., Nerbonne, J. M., Lichtman, J. W., and Sanes, J. R. (2000). Imaging neuronal subsets in transgenic mice expressing multiple spectral variants of GFP. Neuron 28:41–51. Fu, L., Buchholz, D., and Shi, Y. B. (2002). Novel double promoter approach for identification of transgenic animals: A tool for in vivo analysis of gene function and development of gene-based therapies. Mol. Reprod. Dev. 62:470–476. Godwin, A. R., and Capecchi, M. R. (1998). Hoxc13 mutant mice lack external hair. Genes Dev. 12:11–20. Godwin, A. R., Stadler, H. S., Nakamura, K., and Capecchi, M. R. (1998). Detection of targeted GFP-Hox gene fusions during mouse embryogenesis. Proc. Natl. Acad. Sci. USA 95:13042–13047. Gong, S., Zheng, C., Doughty, M. L., Losos, K., Didkovsky, N., Schambra, U. B., Nowak, N. J., Joyner, A., Leblanc, G., Hatten, M. E., and Heintz, N. (2003). A gene expression atlas of the central nervous system based on bacterial artificial chromosomes. Nature 425: 917–925. Gordon, J. W., and Ruddle, F. H. (1981). Integration and stable germ line transmission of genes injected into mouse pronuclei. Science 214:1244–1246. Griesbeck, O., Baird, G. S., Campbell, R. E., Zacharias, D. A., and Tsien, R. Y. (2001). Reducing the environmental sensitivity of yellow fluorescent protein. Mechanisms and applications. J. Biol. Chem. 276:29188–29194. Hadjantonakis, A. K., Gertsenstein, M., Ikawa, M., Okabe, M., and Nagy, A. (1998). Non-invasive sexing of pre-implantation stage mammalian embryos. Nat. Genet. 19:220–222. Hartley, K. O., Hardcastle, Z., Friday, R. V., Amaya, E., and Papalopulu, N. (2001). Transgenic Xenopus embryos reveal that anterior neural development requires continued suppression of BMP signaling after gastrulation. Dev. Biol. 238:168–184. Higashijima, S., Okamoto, H., Ueno, N., Hotta, Y., and Eguchi, G. (1997). High-frequency generation of transgenic zebrafish which reliably express GFP in whole muscles or the whole body by using promoters of zebrafish origin. Dev. Biol. 192:289–299. Hirsch, N., Zimmerman, L. B., Gray, J., Chae, J., Curran, K. L., Fisher, M., Ogino, H., and Grainger, R. M. (2002). Xenopus tropicalis transgenic lines and their use in the study of embryonic induction. Dev. Dyn. 225:522–535. Ikawa, M., Kominami, K., Yoshimura, Y., Tanaka, K., Nishimune, Y., and Okabe, M. (1995). A rapid and non-invasive selection of transgenic embryos before implantation using green fluorescent protein (GFP). FEBS Lett. 375:125–128.

REFERENCES

Itasaki, N., Bel-Vialar, S., and Krumlauf, R. (1999). ‘Shocking’ developments in chick embryology: Electroporation and in ovo gene expression. Nat. Cell Biol. 1:E203–E207. Ivics, Z., Hackett, P. B., Plasterk, R. H., and Izsvak Z. (1997). Molecular reconstruction of Sleeping Beauty, a Tc1-like transposon from fish, and its transposition in human cells. Cell 91:501–510. Jessen, J. R., Meng, A., McFarlane, R. J., Paw, B. H., Zon, L. I., Smith, G. R., and Lin, S. (1998). Modification of bacterial artificial chromosomes through chi-stimulated homologous recombination and its application in zebrafish transgenesis. Proc. Natl. Acad. Sci. USA 95:5121–5126. Jessen, J. R., Willett, C. E., and Lin, S. (1999). Artificial chromosome transgenesis reveals longdistance negative regulation of rag1 in zebrafish, Nat. Genet. 23:15–16. 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. Kawakami, K., Takeda, H., Kawakami, N., Kobayashi, M., Matsuda, N., and Mishina, M. (2004). A transposon-mediated gene trap approach identifies developmentally regulated genes in zebrafish. Dev. Cell 7:133–144. Koga, A., Suzuki, M., Inagaki, H., Bessho, Y., and Hori, H. (1996). Transposable element in fish. Nature 383:30. Koster, R. W., and Fraser, S. E. (2001). Direct imaging of in vivo neuronal migration in the developing cerebellum. Curr. Biol. 11:1858–1863. Kroll, K. L., and Amaya, E. (1996). Transgenic Xenopus embryos from sperm nuclear transplantations reveal FGF signaling requirements during gastrulation. Development 122:3173–3183. Kundu, M., Javed, A., Jeon, J. P., Horner, A., Shum, L., Eckhaus, M., Muenke, M., Lian, J. B., Yang, Y., Nuckolls, G. H., Stein, G. S., and Liu, P. P. (2002a). Cbfbeta interacts with Runx2 and has a critical role in bone development. Nat. Genet. 32:639–644. Kundu, M., Chen, A., Anderson, S., Kirby, M., Xu, L., Castilla, L. H., Bodine, D., and Liu, P. P. (2002b). Role of Cbfb in hematopoiesis and perturbations resulting from expression of the leukemogenic fusion gene Cbfb-MYH11. Blood 100:2449–2456. Lai, L., Park, K. W., Cheong, H. T., Kuhholzer, B., Samuel, M., Bonk, A., Im, G. S., Rieke, A., Day, B. N., Murphy, C. N., Carter, D. B., and Prather, R. S. (2002). Transgenic pig expressing the enhanced green fluorescent protein produced by nuclear transfer using colchicine-treated fibroblasts as donor cells. Mol. Reprod. Dev. 62:300–306. Lee, K. Y., Huang, H., Ju, B., Yang, Z., and Lin, S. (2002). Cloned zebrafish by nuclear transfer from long-term-cultured cells. Nat. Biotechnol. 20:795–799. Li, X., Zhao, X., Fang, Y., Jiang, X., Duong, T., Fan, C., Huang, C. C., and Kain, S. R. (1998). Generation of destabilized green fluorescent protein as a transcription reporter. J. Biol. Chem. 273:34970–34975. Liu, N. A., Huang, H., Yang, Z., Herzog, W., Hammerschmidt, M., Lin, S., and Melmed, S. (2003). Pituitary corticotroph ontogeny and regulation in transgenic zebrafish. Mol. Endocrinol. 17:959–966. Logan, M., and Tabin, C. (1998). Targeted gene misexpression in chick limb buds using avian replication-competent retroviruses. Methods 14:407–420. Long, Q., Meng, A., Wang, H., Jessen, J. R., Farrell, M. J., and Lin, S. (1997). GATA-1 expression pattern can be recapitulated in living transgenic zebrafish using GFP reporter gene. Development 124:4105–4111. Long, Q., Huang, H. G., Shafizadeh, E., Liu, N., and Lin, S. (2000). Stimulation of erythropoiesis by inhibiting a new hematopoietic death receptor in transgenic zebrafish. Nature Cell Biology 2(8):549–552. Maskos, U., Kissa, K., St. Cloment, C., and Brulet, P. (2002). Retrograde trans-synaptic transfer of green fluorescent protein allows the genetic mapping of neuronal circuits in transgenic mice. Proc. Natl. Acad. Sci. USA 99:10120–10125.

301

302

USES OF GFP IN TRANSGENIC VERTEBRATES

Megason, S. G., and Fraser, S. E. (2003). Digitizing life at the level of the cell: High-performance laser-scanning microscopy and image analysis for in toto imaging of development. Mech. Dev. 120:1407–1420. Megason, D. G., and McMahon, A. P. (2002). A mitogen gradient of dorsal midline Wnts organizes growth in the developing CNS. Development 129:2087–2098. Meng, A., Tang, H., Ong, B. A., Farrell, M. J., and Lin, S. (1997). Promoter analysis in living zebrafish embryos identifies a cis-acting motif required for neuronal expression of GATA-2. Proc. Natl. Acad. Sci. USA 94:6267–6272. Miesenbock, G., De Angelis, D. A., and Rothman, J. E. (1998). Visualizing secretion and synaptic transmission with pH-sensitive green fluorescent proteins. Nature 394:192–195. Miyawaki, A., Griesbeck, O., Heim, R., and Tsien, R. Y. (1999). Dynamic and quantitative Ca2+ measurements using improved cameleons. Proc. Natl. Acad. Sci. USA 96:2135–2140. Moss, J. B., Price, A. L., Raz, E., Driever, W., and Rosenthal, N. (1996). Green fluorescent protein marks skeletal muscle in murine cell lines and zebrafish. Gene 173:89–98. Muramatsu, T., Mizutani, Y., Ohmori, Y., and Okumura, J.-I. (1997). Comparison of three non-viral transfection methods for foreign gene expression in early chicken embryos in ovo. Biochem. Biophys. Res. Commun. 230:376–380. Nagai, T., Ibata, K., Park, E. S., Kubata, 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. Offield, M. F., Hirsch, N., and Grainger, R. M. (2000). The development of Xenopus tropicalis transgenic lines and their use in studying lens developmental timing in living embryos. Development 127:1789–1797. 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. Okada, A., Lansford, R., Weimann, J. M., Fraser, S. E., and McConnell, S. K. (1999). Imaging cells in the developing nervous system with retrovirus expressing modified green fluorescent protein. Exp Neurol. 156:394–406. Overbeek, P. A., Aguilar-Cordova, E., Hanten, G., Schaffner, D. L., Patel, P., Lebovitz, R. M., and Lieberman, M. W. (1991). Coinjection strategy for visual identification of transgenic mice. Transgenic Res. 1:31–37. Ozato, K., Kondoh, H., Inohara, H., Iwamatsu, T., Wakamatsu, Y., and Okada, T. S. (1986). Production of transgenic fish: Introduction and expression of chicken delta-crystallin gene in medaka embryos. Cell Differ. 19:237–244. Parinov, S., Kondrichin, I., Korzh, V., and Emelyanov, A. (2004). Tol2 transposon-mediated enhancer trap to identify developmentally regulated zebrafish genes in vivo. Dev. Dyn. 231:449–459. Petropoulos, C. J., Payne, W., Salter, D. W., and Hughes, S. H. (1992). Appropriate in vivo expression of a muscle-specific promoter by using avian retroviral vectors for gene transfer. J. Virol. 66:3391–3397. Plautz, J. D., Kaneko, M., Hall, J. C., and Kay, S. A. (1997). Independent photoreceptive circadian clocks throughout Drosophila. Science 278:1632–1635. Pownall, M. E., Isaacs, H. V., and Slack, J. M. (1998). Two phases of Hox gene regulation during early Xenopus development. Curr Biol. 8:673–676. Reggio, B. C., James, A. N., Green, H. L., Gavin, W. G., Behboodi, E., Echelard, Y., and Godke, R. A. (2001). Cloned transgenic offspring resulting from somatic cell nuclear transfer in the goat: Oocytes derived from both follicle-stimulating hormone-stimulated and nonstimulated abattoirderived ovaries. Biol. Reprod. 65:1528–1533. Riven, I., Kalmanzon, E., Segev, L., and Reuveny, E. (2003). Conformational rearrangements associated with the gating of the G protein-coupled potassium channel revealed by FRET microscopy. Neuron 38:225–235.

REFERENCES

Ryffel, G. U., and Lingott, A. (2000). Distinct promoter elements mediate endodermal and mesodermal expression of the HNF1alpha promoter in transgenic Xenopus. Mech Dev. 90:65–75. Sakai, R., Repunte-Canonigo, V., Raj, C. D., and Knopfel, T. (2001). Design and characterization of a DNA-encoded, voltage-sensitive fluorescent protein. Eur. J. Neurosci. 13:2314–2318. 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. Stadler, H. S., Higgins, K. M., and Capecchi, M. R. (2001). Loss of Eph-receptor expression correlates with loss of cell adhesion and chondrogenic capacity in Hoxa13 mutant limbs. Development 128:4177–4188. Strauss, W. M., Dausman, J., Beard, C., Johnson, C., Lawrence, J. B., and Jaenisch, R. (1993). Germ line transmission of a yeast artificial chromosome spanning the murine alpha 1(I) collagen locus. Science 259:1904–1907. Stuart, G. W., McMurray, J. V., and Westerfield, M. (1988). Replication, integration and stable germline transmission of foreign sequences injected into early zebrafish embryos. Development 103:403–412. Stuart, G. W., Vielkind, J. R., McMurray, J. V., and Westerfield, M. (1990). Stable lines of transgenic zebrafish exhibit reproducible patterns of transgene expression. Development 109:577–584. Swartz, M., Eberhart, J., Mastick, G. S., and Krull, C. E. (2001). Sparking new frontiers: Using in vivo electroporation for genetic manipulations. Dev. Biol. 233:13–21. Takada, T., Iida, K., Awaji, T., Itoh, K., Takahashi, R., Shibui, A., Yoshida, K., Sugano, S., and Tsujimoto, G. (1997). Selective production of transgenic mice using green fluorescent protein as a marker. Nat. Biotechnol. 15:458–461. Tawk, M., Tuil, D., Torrente, Y., Vriz, S., and Paulin, D. (2002). High-efficiency gene transfer into adult fish: A new tool to study fin regeneration. Genesis 32:27–31. Teleman, A. A., and Cohen, S. M. (2000). Dpp gradient formation in the Drosophila wing imaginal disc. Cell 103:971–980. Thermes, V., Grabher, C., Ristoratore, F., Bourrat, F., Choulika, A., Wittbrodt, J., and Joly, J. S. (2002). I-SceI meganuclease mediates highly efficient transgenesis in fish. Mech. Dev. 118:91–98. Timmer, J., Johnson, J., and Niswander, L. (2001). The use of in ovo electroporation for the rapid analysis of neural-specific murine enhancers. Genesis. 29:123–132. Tomomura, M., Rice, D. S., Morgan, J. I., and Yuzaki, M. (2001) Purification of Purkinje cells by fluorescence-activated cell sorting from transgenic mice that express green fluorescent protein. Eur. J. Neurosci. 14:57–63. Wakayama, T., Perry, A. C., Zuccotti, M., Johnson, K. R., and Yanagimachi, R. (1998). Full-term development of mice from enucleated oocytes injected with cumulus cell nuclei. Nature. 394369–394374. Yang, X. W., Model, P., and Heintz, N. (1997). Homologous recombination based modification in Escherichia coli and germline transmission in transgenic mice of a bacterial artificial chromosome. Nat. Biotechnol. 15:859–865. Zhu, A. J., Zheng, L., Suyama, K., and Scott, M. P. (2003). Altered localization of Drosophila Smoothened protein activates Hedgehog signal transduction. Genes Dev. 17:1240–1252.

303

14 THE USES OF GREEN FLUORESCENT PROTEIN IN MAMMALIAN CELLS Theresa H. Ward Department of Infectious and Tropical Diseases, London School of Hygiene and Tropical Medicine, London, United Kingdom

Jennifer Lippincott-Schwartz Department of Cell Biology and Metabolism, NICHD, NIH, Bethesda, MD

14.1

INTRODUCTION

The use of green fluorescent protein (GFP) chimeras in the study of cell behavior and dynamics is now ubiquitous in all fields of mammalian cell biology. This extensive use is due to (a) the development of new GFP variants and optimized cell expression strategies that produce bright, stable fluorescent signals and (b) advances in fluorescent imaging methods and microscopy systems that make it simple to analyze protein geography, movement, and chemistry in living cells. Here, we discuss several GFP-based techniques including time-lapse imaging, photobleaching, photoactivation, and fluorescence resonance energy transfer (FRET) that have allowed protein dynamics, function, and expression to be analyzed in living mammalian cells. We further describe how these techniques have led to the identification of new pathways and mechanisms essential for mammalian cell homeostasis, which traditional biochemical approaches have been unable to address.

14.2

GFP AND GFP VARIANTS

Initial breakthrough discoveries demonstrated that the gene for GFP from the jellyfish Aequorea victoria contained all of the information necessary for proper synthesis of a Green Fluorescent Protein: Properties, Applications, and Protocols, Second Edition Edited by Martin Chalfie and Steven R. Kain Copyright © 2006 John Wiley & Sons, Inc.

305

306

THE USES OF GREEN FLUORESCENT PROTEIN IN MAMMALIAN CELLS

fluorescent protein in non-jellyfish species without the need for ancillary jellyfish proteins (Chalfie et al., 1994), that GFP could be utilized in many cell systems (Heim et al., 1994; Inouye and Tsuji, 1994), and that GFP could be attached to a protein of interest to yield a fluorescent chimera (Wang and Hazelrigg, 1994; Kaether and Gerdes, 1995; Ogawa et al., 1995; Olson et al., 1995; Cole et al., 1996; Rizzuto et al., 1996) while retaining the tagged protein’s function [e.g., NMDA (Marshall et al., 1995)]. However, widespread use of GFP as a fluorescent protein tag in mammalian cells only occurred once GFP variants with improved folding, spectral, and expression properties were generated (Table 14.1; see Fig. 14.1 for example fluorescence microscopy images demonstrating organelle localization). Constructed by random and site-directed mutagenesis, many of these variants include the amino acid substitution Ser65 to Thr65 (S65T), which converts the major and minor absorbance peaks of wild-type GFP (wtGFP) to a single absorbance peak at ~489 nm and results in accelerated fluorophore formation (Heim et al., 1995; Heim and Tsien, 1996). The variants also have the codon usage in wtGFP converted to forms more

TABLE 14.1. Spectral Characteristics of the Major Fluorescent Proteinsa Fluorescent Protein

Amino Acid Substitution

BFP

F64L, Y66H, Y145F, V163A F64L, S65T, Y66W, N146I, M153T, V163A F64L, S65T, Y66W, N146I, M153T,V163A, S72A, Y145A, H148D

CFP

Cerulean

wtGFP EGFP

YFP Citrine

Venus

DsRed2 HcRed PA-GFP Kaede KFP1 (kindling) DsRed timer a

F64L, S65T

S65G, V68L, S72A, T203Y S65G, V68L, Q69M, S72A, T203Y F46L, F64L, S65G, V68L, S72A

V163A, T203H

Excitation

Emission

References

384

448

Heim et al. (1994)

433

474

Heim et al. (1994), Ellenberg et al. (1998)

433

474

Rizzo et al. (2004)

397, (475) 489

504 508

514

527

516

529

Chalfie et al. (1994) Heim et al. (1995), Chiu et al. (1996), Cormack et al. (1996), Yang et al. (1996a) Ormö et al. (1996), Ellenberg et al. (1998) Griesbeck et al. (2001)

515

528

Nagai et al. (2002)

558 590 504

583 620 517

572 580 558

582 600 583

Bevis and Glick (2002) Gurskaya et al. (2001) Patterson and LippincottSchwartz (2002) Ando et al. (2002) Chudakov et al. (2003) Terskikh et al. (2000)

Wavelengths are given as the peak of the excitation or emission spectra in nanometers.

307

GFP AND GFP VARIANTS

A

ER

ssKDEL-GFP

ER exit sites

Sec13-GFP

Golgi

GalT-GFP

PM + Golgi

GPI-GFP

B

Figure 14.1. Examples of GFP chimeras and their subcellular localization. (A) Steady-state distribution of several proteins. (B) Confocal images of a cell expressing the secretory cargo protein VSVG-GFP imaged by time lapse as the protein leaves the Golgi apparatus. Eight images at 10-s intervals were overlaid. (Boxed areas) The route of a single post-Golgi carrier to the cell periphery. [Courtesy of Hirschberg et al. (1998).] See color insert.

efficiently used by mammalian cells, producing increased levels of intracellular protein expression. Finally, in all of the GFP mutants it is possible to make the additional mutations of Ala206 to Lys206, Leu221 to Lys221, or Phe223 to Arg223 to prevent GFP from dimerizing at high concentrations (Delagrave et al., 1995; Ehrig et al., 1995; Heim et al., 1995; Cormack et al., 1996; Crameri et al., 1996; Yang et al., 1996a; Zhang et al., 1996; Zolotukhin et al., 1996; Zacharias et al., 2002). The A. Victoria GFP variant known as enhanced GFP (EGFP), which contains the double mutant of Phe64 to Leu64 and Ser65 to Thr65 (F64L/S65T), has become the variant of choice for GFP expression in mammalian cells (Cormack et al., 1996; Yang et al., 1996a; Yang et al., 1996b; Zhang et al., 1996). It is more stable and fluoresces many-fold more intensely than wtGFP when excited at 488 nm, a wavelength of commonly used filter sets and the main emission wavelength of the argon ion laser used in fluorescence-activated cell sorter (FACS) machines as well as in the confocal scanning laser microscope. EGFP’s increased stability and brightness enables proteins labeled with this tag to be visualized in cells with low light intensities over many hours with minimal photobleaching (the photoinduced destruction of a fluorophore), permitting protein trafficking pathways and organelle dynamics to be analyzed in unprecedented detail. Development of other spectral variants of GFP has enabled multispectral imaging to be performed within living mammalian cells. GFP variants with blue emission spectra have

308

THE USES OF GREEN FLUORESCENT PROTEIN IN MAMMALIAN CELLS

been used together with red-shifted variants to label two different protein species, which is useful in protein colocalization experiments (Rizzuto et al., 1996; Yang et al., 1996b). These variants also offer the potential for assessing differential gene expression by flow cytometry (Ropp et al., 1996) and for measuring protein–protein interactions through fluorescence resonance energy transfer analysis (FRET) (Heim and Tsien, 1996; Mitra et al., 1996). However, blue fluorescent protein (BFP) is dim and tends to photobleach readily, so alternative multicolor pairs have been developed. One such pair is cyan fluorescent protein (CFP) and yellow fluorescent protein (YFP), which have superseded BFP and GFP, respectively, as better dual imaging partners. CFP has spectra that are intermediate between BFP and EGFP due to a Tyr66 to Trp66 substitution (Heim and Tsien, 1996; Ellenberg et al., 1998), and it is brighter and displays more photostability and less photodamage under imaging than BFP. YFP, which is much brighter than EGFP, was rationally designed on the basis of the GFP crystal structure to red-shift the absorbance and emission spectra with respect to EGFP (Ormö et al., 1996; Ellenberg et al., 1998). This makes it more efficiently excited by the 514-nm line of an argon ion laser. Together, CFP and YFP are readily imaged as dual signals using the two auxiliary lines of the argon laser, permitting simultaneous analysis of the temporal and spatial behavior of two different proteins. In addition to dual-color imaging, the pairing of CFP and YFP has been instrumental in the study of protein–protein interactions in fluorescent resonance energy transfer (FRET) experiments (Wu and Brand, 1994; Clegg, 1995; Lippincott-Schwartz et al., 2001; Rizzo et al., 2004). The development of brighter variants, Cerulean for CFP (Rizzo et al., 2004) and Citrine and Venus for YFP (Griesbeck et al., 2001; Nagai et al., 2002), provides a potentially superior alternative to the CFP/YFP pairing in multispectral experiments. Efforts to further red-shift GFP excitation and emission spectra to produce additional partners has led to the identification of a number of candidate proteins, predominantly from reef coral, with emission peaks ranging from 576 to 645 nm (Zhang et al., 2002). However, they are still under improvement due to their propensity to oligomerize. DsRed from Discosoma striata forms tetramers, while HcRed from Heteractis crispa dimerizes. Mutations to derive monomeric forms have proved helpful for some chimeric constructs (Campbell et al., 2002), or to concatemerize the red fluorescent protein gene such that in a chimeric protein it is able to oligomerize within the fusion protein (Gerlich et al., 2003), but many constructs (particularly those for membrane-bound proteins) that work well with a GFP label are not proving replicable with a red fluorescent label (Hayes et al., 2004). Analysis of the temporal expression pattern and turnover of proteins has become feasible with the development of GFP variants whose spectral properties change with time or are photoactivatable. Examples of these variants are the fluorescent timer protein (Terskikh et al., 2000) and the photoactivatable proteins including photoactivatable GFP (PA-GFP) (Patterson and Lippincott-Schwartz, 2002), Kaede (Ando et al., 2002), and KFP1 (Chudakov et al., 2003). The fluorescent timer protein was generated by random mutagenesis of the red fluorescent protein drFP583 (Terskikh et al., 2000) to a variant that is initially similar to GFP in terms of emitted light, but which over several hours (~16 h) converts to a red-emitting fluorophore. By observing the ratio of green to red fluorescence, it is possible to determine the age of a protein tagged with the timer protein. The use of PA-GFP, developed by improving on wtGFP’s photoconversion from a neutral to anionic species (Elowitz et al., 1997; Patterson and Lippincott-Schwartz, 2002), offers an even better approach to studying protein turnover. PA-GFP displays little initial fluorescence under excitation at the imaging wavelength (~488 nm) but increases its fluorescence up to 100-fold after activation by irradiation at a different wavelength (~400 nm). This allows

FLUORESCENCE MICROSCOPY-BASED TECHNIQUES USING GFP

Figure 14.2. Examples of three photobleaching techniques—FRAP (A), FLIP (B), and photoactivation (C)—that are commonly used with GFP and GFP chimeras to monitor discrete populations of molecules within cells.

direct highlighting of distinct pools of molecules within cells (Fig. 14.2). Because only photoactivated molecules are fluorescent, the lifetime and behavior of molecules can be studied independently of newly synthesized proteins. Many of the above-mentioned GFP variants are readily expressed as fusion products with other proteins in most mammalian cell types, including primary cells such as neurons, hepatocytes, muscle cells, and hematopoietic cells. This property has allowed them to be used as tools in numerous applications, including as minimally invasive markers to track and quantify individual or multiple protein species, as probes to monitor protein–protein interactions, as photomodulatable proteins to highlight and follow the fate of specific protein populations within a cell, and as biosensors to describe biological events and signals. Below, we describe the fluorescence imaging methods that have been used with these GFP variants in mammalian cells, the types of applications these methods have been used for, and the new insights they have gleaned in the analysis of mammalian cell biology.

14.3

FLUORESCENCE MICROSCOPY-BASED TECHNIQUES USING GFP

Addition of GFP to proteins is usually benign with no apparent disruption of function, despite its relatively large size. Since no exogenously added substrate or cofactors are necessary for detecting GFP fluorescence, cells are exposed to minimal invasive treatment. Furthermore, the GFP fluorophore is relatively photostable, with little photodamage occurring during imaging. Not surprisingly, a wide variety of imaging methods have been developed to take advantage of these properties of GFP (see Table 14.2), which are now being used in mammalian cells.

309

310

THE USES OF GREEN FLUORESCENT PROTEIN IN MAMMALIAN CELLS

TABLE 14.2. Uses of Different GFP Imaging Techniques in Mammalian Cells Imaging Method

Description

References

Time lapse

General visualization of processes and protein movement in cells Two or more chimeric proteins expressed together to compare localization and dynamics Quantitation of comparative amounts of two or more proteins in region of interest Measures diffusional mobility by selective photobleaching of a region of interest Visualization of exchange of proteins between compartments Visualization of the kinetics of one protein by photobleaching a selected pool while using another tagged with a different fluorescent protein as a marker, particularly useful for FRAP of a moving component FRAP of a region surrounding the area of interest to visualize movement of faint objects or exchange dynamics Repeated photobleaching to measure compartment connectivity Detection of protein-protein interactions by proximity of fluorophores Measures lifetime of fluorescence which can modulate with changes in oligomerization, pH, Ca2+, or degradation Measures molecular diffusion at low concentrations; sensitive to protein–protein interactions Visualization of processes close to, or at, the cell surface Low level expression of tagged cytoskeletal components to visualize turnover and movement of polymers Activation of selected pool of tagged protein enables pulse-chase of its subsequent dynamics Fluorescence changes color over time, thus monitoring populations of protein synthesized in response to stimuli (e.g., during development)

Presley et al. (1997), Hirschberg et al. (1998) Ellenberg et al. (1998)

Multispectral imaging Ratio imaging

FRAP

FLAP

iFRAP

FLIP FRET

FLIM

FCS

TIR-FM FSM

PA-GFP

Fluorescent Timer

White et al. (2001)

Edidin (1994), LippincottSchwartz et al. (2001)

Dunn et al. (2002)

Presley et al. (2002), Bubulya and Spector (2004)

Cole et al. (1996), White and Stelzer (1999) Stryer (1978), Pollok and Heim (1999), LippincottSchwartz et al. (2001) Lakowicz et al. (1992), Bastiaens and Squire (1999), Pepperkok et al. (1999) Krichevsky and Bonnet (2002), Weiss and Nilsson (2004) Axelrod (2001), Toomre and Manstein (2001) Waterman-Storer et al. (1998), Waterman-Storer and Danuser (2002) Elowitz et al. (1997), Patterson and LippincottSchwartz (2002) Terskikh et al. (2000)

FLUORESCENCE MICROSCOPY-BASED TECHNIQUES USING GFP

14.3.1

Time-Lapse, Multispectral, and Ratio Imaging

GFP can be used to label the steady-state distribution of a molecule within a cell (Fig. 14.1A). However, perhaps the most widely used imaging technique for GFP-based mammalian cell studies is now time-lapse imaging, in which a single focal plane of a live cell specimen is observed over time (Fig. 14.1B). This has allowed the localization and dynamics of GFP chimeras to be studied in real time, providing enormous insights into a protein’s distribution and transport pathways (including their response to cellular perturbations such as drug treatments and temperature shifts). Previous work examining these issues relied on static images or “snapshots” of large populations of cells, in which a specified cellular response is often difficult or impossible to piece together. In addition to time-lapse imaging, recent work has utilized 4D microscopy, which involves the collection of threedimensional datasets over time (Gerlich et al., 2001). This allows the behavior of a protein to be examined within the entire cell. To analyze the changes in a fluorescent protein’s spatial and temporal behavior in such experiments, researchers have used computer-based visualization programs that can quantify and discriminate fluorescent signals (Bergsma et al., 2001). Use of GFP mutants that fluoresce or are excited at different wavelengths offers the possibility of double labeling to compare the distribution and dynamics of two different populations of proteins simultaneously within cells (Ellenberg et al., 1998). In this method, cells are doubly transfected with proteins attached to different GFP variants that have different excitation or emission spectra and are imaged with alternative filter sets (Rizzuto et al., 1996). With the development of spectral imaging systems, it is now theoretically possible to resolve all six fluorescent protein colors (BFP, CFP, GFP, YFP, DsRed, HcRed) within the same cell. The limitation then becomes whether the cell can actually cope with the overexpression of all the constructs and whether their localization and behavior are affected as a result. Pairs of images can also be quantified using digital image processing techniques to see if the ratio of intensity of the two populations changes with time. Such ratio imaging approaches have already been standardized and used with rhodamine and fluorescein tags in the endosomal system (Mayor et al., 1993), and they promise to be an important application of GFP variants (White et al., 2001).

14.3.2

Photobleaching Techniques: FRAP, FLAP, iFRAP, and FLIP

The time-lapse and multispectral imaging techniques mentioned above can provide important information about the steady-state distribution of a protein over time, but they do not address the kinetic properties of a protein, such as whether the protein is free to diffuse through the cell or is attached to a matrix, or whether it is undergoing exchange between compartments or on/off a substrate. To obtain this type of information, a researcher must differentiate a selected pool of fluorescent proteins from other fluorescent molecules and then follow that pool as it equilibrates with other molecules over time. This differentiation can be accomplished using photobleaching techniques, in which an area of the cell is photobleached with a high-intensity laser pulse and the movement of unbleached molecules from neighboring areas into the bleached area is recorded by time-lapse microscopy (Fig. 14.2). Because photobleaching alters the fluorescence steady state in a cell, the dynamics of a GFP chimera (i.e., its diffusion rate, binding constant, or intracellular trafficking routes) can be unraveled in the absence of conditions that disrupt protein pathways or create protein gradients. Perhaps the most widely used photobleaching technique is f luorescence recovery after photobleaching (FRAP), in which fluorescent proteins in a small area are irreversibly

311

312

THE USES OF GREEN FLUORESCENT PROTEIN IN MAMMALIAN CELLS

bleached by an intense laser flash and recovery is measured using an attentuated laser beam (Edidin, 1994). This technique can provide an estimation of the effective diffusion coefficient (Deff) and mobile fraction (Mf) of a protein. Whereas Deff reflects the meansquared displacement that an idealized protein moves by random walk over time, Mf represents the fraction of fluorescent proteins that can diffuse into a bleached region during the time course of the experiment. By comparing the observed Deff and Mf to idealized values, one can determine whether the GFP chimera under study undergoes interactions with other molecules, is bound to a scaffold, or can freely diffuse within cells [Fig. 14.3; for review see Lippincott-Schwartz et al. (2001)]. With the advent of two-color GFP imaging (Ellenberg et al., 1998), another application of photobleaching that has become feasible is the technique called fluorescence localization after photobleaching (FLAP) (Dunn et al., 2002), which uses photobleaching to visualize protein trafficking or flux through a two-color bleaching protocol (White et al., 2001; Dunn et al., 2002). In this approach, proteins to be visualized are tagged with either the CFP or YFP (producing two chimeras of the same protein) and coexpressed in the same cells, or alternatively a single chimera is introduced carrying a CFP–YFP concatamer. The YFP and CFP molecules are then excited together, using a multitracking mode of a confocal microscope with a 514-nm laser line to excite YFP and a 458-nm laser line (or 413nm laser line) to excite CFP. YFP is selectively photobleached using the 514-nm laser line at maximum power, leaving CFP fluorescence unaltered. Fluorescence recovery is then tracked through image differencing by subtracting the image of the bleached fluorochrome from that of the unbleached fluorochrome. This technique is a great advance in photobleaching technology, since it exploits CFP as a visual reference to follow the dynamics of a targeted protein or organelle and, simultaneously, to follow and quantify YFP fluorescence recovery. Another application of photobleaching is called inverse FRAP (iFRAP), which can be used to reduce fluorescence from background noise to reveal faint populations of fluorescent proteins (Presley et al., 2002). In this approach, fluorescence surrounding a particular region of the cell is photobleached to allow visualization of fluorescent protein movement from the unbleached to bleached areas. As an example, photobleaching of fluorescence associated with the plasma membrane will allow visualization of organelle behavior inside the cell, which otherwise is masked by the plasma membrane fluorescence (Nichols et al., 2001). This approach has also been used to investigate intranuclear dynamics [e.g., exchange between the nucleolus and the nucleoplasm (Dundr et al., 2002)]. Fluorescence loss in photobleaching (FLIP) investigates fluorophore mobility and the continuity of various intracellular environments or compartments (Cole et al., 1996; Fig. 14.2). FLIP is similar to FRAP in that a region of interest is photobleached with a highpower laser; however, unlike FRAP, the region is repeatedly bleached over time to deplete the entire fluorescent pool. If photobleaching of one region depletes the entire fluorescence of the other, then the fluorescent molecules are capable of freely diffusing between the two regions. Thus, by using this technique, it is possible to address whether a protein can diffuse uniformly across a compartment or whether there are regions of restricted mobility.

14.3.3

FRET and FCS

The two fluorescence-based techniques, fluorescence resonance energy transfer (FRET) and fluorescence correlation spectroscopy (FCS), enable protein–protein interactions to be spatially and temporally resolved in living cells. Whereas FRET detects the close

FLUORESCENCE MICROSCOPY-BASED TECHNIQUES USING GFP

Figure 14.3. Distribution and mobilities of a nuclear envelope membrane protein, lamin B receptor (LBR) tagged with GFP in interphase membranes. At steady state, LBR-GFP is found localized within the ER network and in the inner nuclear envelope (NE). Qualitative FRAP experiments in ER and NE membranes in interphase cells expressing LBR-GFP show (left) photobleach recovery in ER membranes, and (right) photobleach recovery in NE membranes. Note the complete recovery of fluorescence in the ER and the lack of recovery in the NE. [Courtesy of Ellenberg et al. (1997).]

313

314

THE USES OF GREEN FLUORESCENT PROTEIN IN MAMMALIAN CELLS

proximity of interacting proteins, FCS detects either changes in the diffusion or the codiffusion of bound species. The use of GFP chimeras with these techniques is providing quantitative data on the physicochemical properties of molecules within cells. The type of data differs from that shown by biochemical approaches in that the microscopy approaches effectively measure the ability of molecules to cluster together (as detailed below). However, FCS can determine the absolute concentration of a species in vivo, which, in conjunction with the diffusion measurements, permits calculation of the Kd for protein–protein interactions. In contrast to biochemical methods, this may not be measurable for the Kd of a small enzyme and its substrate, but is measurable for a protein that is interacting with a large substrate such as a complex of proteins or a membrane, for example. Using FCS, the GTPase Arf1, which transiently binds and releases from Golgi membranes, has been shown to be freely soluble in mitotic cells, rather than bound to membranes (Altan-Bonnet et al., 2003). Furthermore, because FCS and FRET are measurable in real time in living cells, transient interactions over short periods of time can be captured, which may be missed by more traditional approaches (such as immunoprecipitation or chemical cross-linking) that are dependent on populations of cells exhibiting sufficient fractionable protein–protein interaction to be measurable [for example, FCS has demonstrated Gag–Gag interactions in the cytosol of Rous sarcoma virus-infected cells (Larson et al., 2003)]. FRET works by measuring the transfer of photon energy from one fluorophore to another molecule when both are located within a few nanometers of each other (Stryer, 1978; Uster and Pagano, 1986; Tsien et al., 1993). If the energy of the excited fluorophore coincides with the energy needed to excite the absorber, then energy is transferred. This transfer results in loss of fluorescence intensity of donor and fluorescence emission from the acceptor. The working scale of FRET is less than or equal to 100 Å, in contrast to conventional light microscopy, which is a few tenths of a micron (Stryer, 1978). The availability of several different mutants of GFP opens the possibility of using FRET to probe inter- and intramolecular distances in proteins, allowing the possibility of mapping protein–protein interactions within cells by fluorescence microscopy. An advantage of intramolecular FRET is that the stoichiometry between the donor and acceptor fluorophore is fixed, thereby enabling the ratio of acceptor to donor fluorescence to accurately measure FRET changes. An excellent example of intramolecular FRET is from the study by Heim and Tsien (1996). They attached the GFP mutants, Y66H/Y145F and S65C, to the same protein by a 25-residue cleavable spacer and used the first as donor and the second as acceptor in FRET experiments. Proteolytic cleavage of the spacer resulted in the two protein domains diffusing apart, causing loss of green emission by the acceptor S65C domain and enhancement of blue emission from the donor domain. Since this early seminal study, FRET using GFP chimeras has become a regular tool in cell biology. Intermolecular FRET can detect interactions between two proteins in real time. However, the involvement of mixed complexes between the endogenous protein and its labeled counterpart becomes an issue, and the ratio of donor to acceptor expression is no longer fixed. FRET is then better measured through acceptor photobleaching or by fluorescence lifetime imaging microscopy (FLIM), which measures the decay kinetics of the excited state. FLIM requires specialized equipment and complex mathematical analysis but can be used in living cells to measure changes in pH, Ca2+ concentration, protein–protein interactions (by FRET/FLIM, because lifetime greatly decreases when FRET is occurring), and proteolytic processing (Lakowicz et al., 1992; Bastiaens and Squire, 1999; Ng et al., 1999; Pepperkok et al., 1999; Calleja et al., 2003; Lin et al., 2003).

FLUORESCENCE MICROSCOPY-BASED TECHNIQUES USING GFP

In contrast to FRET, FCS measures the fluctuations in photons resulting from diffusion of fluorescently labeled molecules in and out of a small, defined volume (~1 femtoliter). Because the fluctuations reflect the average number of fluorescent molecules in the volume and the time of their diffusion, parameters such as the concentration of the fluorescent molecule and its diffusion constant can be derived using this technique (Krichevsky and Bonnet, 2002). It is also possible to look at protein–protein interactions using FCS, because binding to another protein alters the protein’s diffusional mobility. This technique was used to great effect in a study looking at the membrane association of the small GTPase Arf1 during mitosis (Altan-Bonnet et al., 2003). Inactive Arf1 is cytosolic (i.e., it has a diffusional coefficient D of a small cytoplasmic molecule), whereas if a fraction of Arf1-GFP were active and associated with mitotic membranes (e.g., vesicles), it would be detected by FCS as a species that diffused slower (with a D characteristic of vesicles). Because >98% Arf1 was found to diffuse as a small molecule, it could be inferred that Arf1 is persistently inactive during metaphase and does not associate with membranes. The degree of protein interaction or identity of the binding partner can be confirmed using fluorescence cross-correlation spectroscopy (FCCS), where two differently labeled proteins can be monitored together (Pyenta et al., 2001; Bacia et al., 2002; Weiss and Nilsson, 2004). FCS is currently less popular than FRET, but given its ability to measure both concentrations and diffusion constants, and the availability of confocal microscopes capable of sampling small volumes, FCS holds great promise for advancing our understanding of protein behavior in vivo (Elsner et al., 2003; Fradin et al., 2003; Weiss et al., 2003).

14.3.4 Total Internal Reflection Fluorescence and Fluorescent Speckle Microscopy Total internal reflection fluorescence microscopy (TIR-FM), or evanescent wave microscopy, allows processes to be imaged that only occur within very close proximity to the coverslip. This is accomplished by directing an excitatory laser beam through the coverslip at an angle steep enough so that it completely reflects off the water–coverslip interface. The result is the production of an evanescent field, in which a layer of ~100 nm is excited. Because only fluorescent molecules within this distance from the coverslip are excited, a high signal-to-background imaging of surface events is possible (Axelrod, 2001; Toomre and Manstein, 2001). GFP chimeras expressed in mammalian cells imaged using this technique have revealed important new insights into the mechanism(s) underlying fusion of (a) constitutive membrane transport carriers with the plasma membrane [Fig. 14.4; Schmoranzer et al., 2000; Toomre et al., 2000; Kreitzer et al., 2003; Schmoranzer and Simon, 2003) and (b) regulated secretory organelles (e.g., Weibel–Palade bodies (Manneville et al., 2003)]. In addition, interactions between microtubules and focal adhesion contacts at the cell surface have been analyzed with this approach (Krylyshkina et al., 2003). Also, dual-color TIR-FM has been used to image the interrelationship of clathrin-coated pits with actin and dynamin (Merrifield et al., 2002). Fluorescent speckle microscopy (FSM) is a technique that has been used for visualizing the movement, assembly, and turnover of macromolecular assemblies like the cytoskeleton in living cells. In this method, a fluorescently labeled protein is introduced into a cell at very low levels (0.1–0.5%) such that it can co-assemble as a small fraction of fluorescent subunits in a pool of unlabeled subunits. Because the labeled proteins are randomly incorporated into the polymer lattice, the individual molecules can be detected as a “fluorescent speckle” pattern. Movement of the speckles within cytoskeletal filaments

315

316

THE USES OF GREEN FLUORESCENT PROTEIN IN MAMMALIAN CELLS

Figure 14.4. Visualizing fusion of post-Golgi carriers with the plasma membrane using total internal reflection fluorescence microscopy (TIR-FM). Selected frames from a sequence showing the transport, docking, and fusion of a tubular carrier labelled with VSVG-GFP. Times are marked relative to the start of fusion. [Courtesy of Schmoranzer et al. (2000); reproduced from The Journal of Cell Biology 149:23–32 by copyright permission of The Rockefeller University Press.]

can be detected as they move from the site of assembly to regions of depolymerization (Waterman-Storer et al., 1998). The technique works with either microinjection of fluorescently labeled subunits or expression of subunits ligated to GFP. However, sometimes multiple GFP sequences need to be attached to the protein-encoding gene to get enough fluorescent signal from a single protein molecule.

14.4

APPLICATIONS OF GFP IN MAMMALIAN CELLS

Given the numerous fluorescence-based techniques available for monitoring GFP, it is not surprising that an enormous variety of applications for GFP in mammalian cells have been developed (see Table 14.3). These range from the determination of a protein’s geography, movement, and molecular interactions to the development of gene therapy vectors and cell sorting protocols, as discussed below.

14.4.1 Determining a Protein’s Localization, Dynamics, and Concentration Within Cells The most widespread application of GFP in mammalian cells has been for characterizing the location and dynamics of proteins expressed as fusion partners with GFP. GFP chimeras provide a major advance over previous methods for studying the intracellular localization and dynamics of proteins (Fig. 14.1). Previous techniques required fixation and permeabilization methods to gain access within the cell to the protein of interest. Numerous problems in specimen preparations often arise as a result of fixation including the danger of extracting or damaging antigen and the possibility that labeling efficiencies within different cell structures will differ. With GFP chimeras, these problems are avoided because the protein of interest is viewed in a living, unperturbed cell. The GFP reporter,

APPLICATIONS OF GFP IN MAMMALIAN CELLS

TABLE 14.3. Applications for GFP in Mammalian Cells Applications

Example References

Protein localization

Presley et al. (1997), Chao et al. (1999), Zaal et al. (1999), Griffis et al. (2002) Sutherland et al. (2001), Conrad et al. (2004), Sineshchekova et al. (2004) Cole et al. (1996), Ellenberg et al. (1997), Nehls et al. (2000), Phair and Misteli (2000), Daigle et al. (2001), Stenoien et al. (2001), Elsner et al. (2003), Weiss et al. (2003), Kenworthy et al. (2004), Shav-Tal et al. (2004) Vasudevan et al. (1998), Gaidarov et al. (1999), Wu et al. (2001b), Presley et al. (2002), Salmon et al. (2002), Altan-Bonnet et al. (2003), Elsner et al. (2003), Weiss and Nilsson (2003), Dundr et al. (2004), Engqvist-Goldstein et al. (2004) Hirschberg et al. (1998), Zaal et al. (1999), Dahm et al. (2001), Nichols et al. (2001) Majoul et al. (2001), Zacharias et al. (2002), Hayes et al. (2004), Snapp et al. (2004) Cole et al. (1996), Zaal et al. (1999), Nehls et al. (2000), Nichols et al. (2001) Presley et al. (1997), Scales et al. (1997), Hirschberg et al. (1998), Toomre et al. (1999), Nichols et al. (2001) Moriyoshi et al. (1996), Mosser et al. (1997), Mancia et al. (2004) Bartlett et al. (1995), Dorsky et al. (1996) Sönnichsen et al. (2000), Keller et al. (2001), Nichols et al. (2001), Presley et al. (2002), Stephens and Pepperkok (2002, 2004), Kreitzer et al. (2003), Mironov et al. (2003), Polishchuk et al. (2004) Elliott and O’Hare (1999), Pelkmans et al. (2002), Meulenbroek et al. (2004) Mahajan et al. (1999), Wiegand et al. (2003), Cook and Hinkle (2004) Ellenberg et al. (1997), Zaal et al. (1999), Bergeland et al. (2001), Jokitalo et al. (2001), Ward et al. (2001), Beaudouin et al. (2002), Salina et al. (2002), Gerlich et al. (2003), Walter et al. (2003) Presley et al. (1997), Chao et al. (1999), Tvaruskó et al. (1999), Eils et al. (2000), Stephens et al. (2000), Dahm et al. (2001), Gerlich et al. (2001), Wu et al. (2001a), Schmoranzer and Simon (2003) Anderson et al. (1996), Ropp et al. (1996), Mosser et al. (1997), Espinet et al. (2000) Niswender et al. (1995), Terasaki et al. (1996), Hirschberg et al. (1998)

GFP gene trap Protein diffusion rates

Binding and dissociation constants

Rate constants for intracellular transport steps Protein–protein interactions Compartment connectivity Visualization of transport intermediates

Expression marker Viral gene marker Intracellular sorting

Monitoring viral infection Protein turnover Organelle dynamics and assembly

Protein tracking

FACS Protein concentration within cells or compartments

317

318

THE USES OF GREEN FLUORESCENT PROTEIN IN MAMMALIAN CELLS

itself, usually does not interfere with the normal functioning, or targeting, of the tagged protein and can be added to either the C- or N-terminus of target proteins. Hence, use of GFP chimeras has become the primary means for identifying and studying a protein’s distribution within cells. In addition to offering a simple way to localize proteins within living cells, GFP fusion proteins have become an important tool for understanding protein dynamics, because distinct populations of GFP chimeras can be easily highlighted using photobleaching or photoactivation techniques. This has revolutionized our understanding of complex processes and structures within cells. In addition to describing a protein’s dynamics within cells, use of GFP chimeras allows the effective intracellular concentration of these molecules to be determined. This can be accomplished by comparing the intensity of a GFP chimera’s fluorescence with fluorescence of a known fluorescent protein standard (whose concentration in solution is known) using a sensitive camera system (e.g., a cooled CCD camera) or confocal microscope (Niswender et al., 1995; Terasaki et al., 1996). Knowledge of a protein’s concentration within cells is invaluable for the development and testing of mathematical models describing a protein’s biophysical parameters (including its binding and dissociation constants, or rate constants). As one example, Hirschberg et al. (1998) used a standardized recombinant GFP solution as a control to quantify the amount of VSVG-GFP protein in different subcellular compartments of mammalian tissue culture cells over time. This information was then used to model the kinetics by which a bolus of secretory cargo trafficked between compartments of the secretory pathway. Quantitative modeling can also be used to track protein residence times through organelle-specific photobleaching (Zaal et al., 1999; Dahm et al., 2001; Nichols et al., 2001; Presley et al., 2002; Weiss and Nilsson, 2003).

14.4.2

GFP as a Co-transfection or Expression Marker

GFP has been used as an expression marker in order to determine a protein’s expression level in the absence of direct tagging with GFP. This method is useful when a protein of interest cannot be directly tagged with GFP. Several approaches have been used to achieve this, including: transfection of two plasmids in tandem (White et al., 2001); transfection of two transcription initiation startpoints within the same vector, one transcribing GFP, the other the protein of interest (Mancia et al., 2004); or use of a polycistronic vector where the GFP is translated off the same RNA as, but not fused to, another protein through the use of an internal ribosome entry site [IRES (Mosser et al., 1997)].

14.4.3

Gene Targeting Using Viral Systems

The use of viral vector systems to express GFP has been used to monitor production and release of therapeutic molecules from cells and tissues. Bartlett et al. (1995) developed an adenovirus vector delivery system using GFP inserted downstream from the human muscle creatine kinase promoter and found efficient GFP expression in skeletal muscle injected with the vector. In contrast to traditional reporter methods including b-galactosidase, firefly luciferase, or chloramphenicol amino transferase (CAT) assays (which require cell lysis and introduction of the reporter enzyme substrate), GFP expression could be monitored consecutively over several days. In addition to the enormous potential of viral reporter genes carrying GFP in clinical studies for tracking the expression of gene products, such molecules are also extremely valuable for basic research. Moriyoshi et al. (1996), for example, used an adenovirus vector to transfer GFP into postmitotic neuronal cells in vivo

GFP REVELATIONS

to study cell migration and development of neuronal connections. Adeno-associated virus vectors expressing GFP were also used to target GFP to spinal neurons (Peel et al., 1997), allowing the fate of neurons to be followed and their response to various transducers analyzed.

14.4.4

Viral Infection and Pathogenesis

Viral vectors containing GFP have been used to monitor viral infection and pathogenesis with no need for processing of cells to detect infected cells. Using this approach, Dorsky et al. (1996) identified human immunodeficiency virus (HIV)-1-infected cells in tissue using GFP tagged HIV-1. GFP under the control of HIV-1 LTR promoter was readily detected in virally infected cells either by fluorescence microscopy or by fluorescenceactivated cell sorting. With the same goal in mind, Dhandayuthapani et al. (1995) used a mycobacterial shuttle-plasmid vector carrying GFP cDNA to assess mycobacterial interactions with macrophages. More recently, GFP has been used to comprehend the cytology of viral infection directly by visualizing cell uptake and viral factories in vivo (Elliott and O’Hare, 1999; Chen and Ahlquist, 2000; Ward and Moss, 2001; Potel et al., 2002; Taylor et al., 2003; La Boissière et al., 2004; Moradpour et al., 2004).

14.4.5

Flow Cytometry

Screening and selection of cells by flow cytometry has been greatly facilitated using GFP expression, since it provides an easy method for fluorescent labeling of viable cells. This method eliminates the task of characterizing cell lines through standard biochemical methods involving protein analysis. As an example, Mosser et al. (1997) generated a dicistronic mRNA encoding both a gene of interest and the gene for GFP. Clone selection involved the simple monitoring for GFP fluorescence using the fluorescence-activated cell sorter. Quantitative detection from two different genes within single mammalian cells has also been demonstrated using multiparameter flow cytometry with GFP and its red-shifted variant (Anderson et al., 1996) or EGFP and YFP (Espinet et al., 2000).

14.5

GFP REVELATIONS

Given the saturation of current cell biology publications with the use of GFP chimeras, this review is unable to cover the entire extent of the literature. The following sections, therefore, will focus on areas of mammalian cell biology that have been crucially changed by GFP-based experimental approaches.

14.5.1

Cytoskeleton

The cytoskeleton has been “illuminated” by the use of GFP. Not only can individual polymer subunits be independently labeled, but associated proteins including motor proteins and binding proteins can be tagged with GFP. Direct labeling of cytoskeletal subunits was initially used primarily as a noninvasive label of the cytoskeleton upon which to watch movements of vesicular intermediates or membranes (Robbins et al., 1999; Toomre et al., 1999) or of mitotic chromosomes (Haraguchi et al., 1999) in living cells (Ludin and Matus, 1998). With the introduction of fluorescence speckle microscopy (FSM) (see above), the dynamics of growth and shrinkage of GFP-tagged actin and microtubule

319

320

THE USES OF GREEN FLUORESCENT PROTEIN IN MAMMALIAN CELLS

filaments has now been analyzed (Waterman-Storer et al., 1998; Watanabe and Mitchison, 2002). Furthermore, use of dual-wavelength FSM has enabled the relative dynamics of GFP-labeled f-actin and microtubules to be monitored in migrating cells. These studies have shown that the movement and organization of f-actin helps to coordinate the dynamic organization of microtubules, suggesting that these two cytoskeletal components dynamically bind and interact with each other in vivo (Salmon et al., 2002). GFP-labeling of auxiliary proteins such as those that tether actin filaments to focal contacts and fibrillar adhesions (Zamir et al., 2000) have also revealed that these molecules are dynamic, moving at an average rate of 19 mm/h. Other important new insights into the cytoskeleton using GFP chimeras have been obtained using TIR-FM (Krylyshkina et al., 2003). Using either GFP-tubulin to label microtubules or GFP-CLIP-170 to label microtubule ends, these studies showed that microtubules consistently track to adhesion complexes labeled with DsRed-zyxin, suggesting that microtubules can provide tracks from the cell interior to specific zones of the plasma membrane.

14.5.2

Nucleus

Our understanding of nuclear architecture and dynamics has been dramatically improved due to the use of GFP in the study of nuclear behavior. Perhaps most significant has been the realization that interphase nuclear organization, once thought to be comprised of stably associated components assembled into rigid arrays, is extremely dynamic (both spatially and temporally) and is capable of self-organization (Misteli, 2001; Janicki and Spector, 2003; Bubulya and Spector, 2004). Using time-lapse imaging and photobleaching approaches to probe the behavior of different GFP-tagged nuclear components, researchers have found that many nuclear proteins are undergoing rapid movement within the nucleoplasm. The linker histone, H1, for example, was found to undergo rapid association and dissociation with chromatin (Misteli et al., 2000). Likewise, only a transient association of the transcription factor, glucocorticord receptor, with its promoter elements was observed (McNally et al., 2000). Moreover, RNA polymerase I and II components, DNA topoisomerase II, as well as DNA repair complexes were found to be highly mobile in the nucleoplasm (Houtsmuller et al., 1999; Becker et al., 2002; Christensen et al., 2002; Dundr et al., 2002; Essers et al., 2002; Kimura et al., 2002). An overriding theme from these studies is that many multiprotein complexes in the nucleus are very dynamic and do not form the stable holo-complexes previously predicted by biochemical methods. In the face of so many nuclear components being highly dynamic, the challenge has become one of determining what mechanisms allow the nucleus to establish and maintain its organization. This organization has been highlighted by the identification of nuclear subcompartments using a large-scale screen with a GFP gene trap technique, which found many discretely localized proteins (i.e., splice-rich speckles and nucleoli) (Sutherland et al., 2001). The current thinking is that some proteins act as structural components to create scaffolds onto which other components dynamically associate through transient interactions. This fits with photobleaching results such as those demonstrating that core histones labeled with GFP exhibit tight association with chromatin, whereas linker histones are only transiently associated (Lever et al., 2000; Misteli et al., 2000; Phair and Misteli, 2000; Kimura and Cook, 2001). An additional mechanism underlying nuclear organization could be through transient interactions between different components that are of variable strengths. This would allow different components to dynamically self-organize into steadystate assemblies within the nucleus, as has been suggested for Cajal bodies and speckles (Misteli, 2001; Lamond and Spector, 2003; Dundr et al., 2004).

GFP REVELATIONS

The regulation of chromosome position within the nucleus is another area in which GFP-based techniques is providing new insights. A study by Chubb et al. (2002) reported that chromosome dynamics in interphase correlated with nuclear positioning, with loci adjacent to nucleoli or to the nuclear periphery exhibiting more restricted movement than loci in nucleoplasmic locations. Two other studies revealed that during mitosis, metaphase chromosome positioning is inherited. One of these used the lac operator integrated into the genome to visualize a lac repressor-GFP fusion protein associated with chromosomes (Dietzel and Belmont, 2001), while the other used photobleaching of half the nucleus, either parallel to or perpendicular to the spindle axis, to address whether similar or different bleaching patterns reappeared after cell division (Gerlich et al., 2003). Results from both approaches were consistent with chromosome organization and behavior within mammalian cells being nonrandom. However, by visualizing a smaller population of GFPlabeled histone, a study by Walter et al. (2003) found that chromosome neighborhoods do not seem to be tightly maintained during mitosis. This highlights how different techniques applied to the same fusion proteins can reveal different conclusions. Another topic under active investigation is the dynamics of the nuclear envelope. GFP-based techniques have been used to study the organization and behavior of nuclear pore complexes. These studies have shown that nuclear pore complexes do not diffuse within the plane of the nuclear envelope and are composed of proteins that can be either stable [e.g., lamin B receptor, emerin, and lamins (Fig. 14.3; Ellenberg et al., 1997; Ostlund et al., 1999; Moir et al., 2000; Daigle et al., 2001)] or highly dynamic [nucleoporins (Daigle et al., 2001; Griffis et al., 2002)] proteins. They have further shown that nuclear lamina and nuclear pore complexes are interconnected, immobile two-dimensional networks that move synchronously during nuclear shape changes (Daigle et al., 2001). Live cell imaging and photobleaching studies of the nuclear envelope have also provided new insights into nuclear envelope breakdown and reassembly during mitosis (Ellenberg et al., 1997; Burke and Ellenberg, 2002). Such studies have shown that during mitosis, the nuclear envelope does not vesiculate, but is absorbed into the ER (Ellenberg et al., 1997). They have also demonstrated that initiation of nuclear envelope breakdown occurs by a microtubule-dependent tearing process combined with nuclear pore disassembly (Beaudouin et al., 2002; Salina et al., 2002), while nuclear envelope reassembly at the end of mitosis involves wrapping of ER membranes enriched in nuclear envelope components around chromatin (Ellenberg et al., 1997; Haraguchi et al., 2000; Moir et al., 2000; Holaska et al., 2002).

14.5.3

Membrane Trafficking and Organelle Dynamics

GFP-based studies are having an equally powerful impact on our understanding of the endomembrane of mammalian cells. Comprised of distinct organelles [including the endoplasmic reticulum (ER), Golgi apparatus, endosomes, lysosomes, and plasma membrane] and membrane-bound transport intermediates, the endomembrane system regulates the synthesis and sorting of all membrane-associated proteins within the cells. It is also responsible for protein secretion and the uptake of macromolecules from the extracellular environment. Among the major new findings obtained from GFP-based studies has been the recognition that most transport intermediates—long thought to be small, spherical vesicles of 60–90 nm—are, in fact, large, pleiomorphic structures capable of transforming into globular or tubular shapes depending on their interactions with cytoskeletal elements (Presley et al., 1997; Sciaky et al., 1997; Hirschberg et al., 1998; Shima et al., 1999; Toomre et al.,

321

322

THE USES OF GREEN FLUORESCENT PROTEIN IN MAMMALIAN CELLS

1999; Blum et al., 2000; Polishchuk et al., 2000, 2003; Schmoranzer et al., 2000; Mironov et al., 2003; Schmoranzer and Simon, 2003). These structures have been demonstrated using GFP-tagged cargo proteins as they move within membrane-bound carriers to different intracellular sites, and correlative electron microscopy has been used to verify that the large carriers seen by light microscopy are indeed tubular structures (Polishchuk et al., 2000, 2003; Mironov et al., 2003). Double-labeling of different cargo proteins has revealed that different proteins, particularly those destined for different cell surface domains, are capable of segregating into spatially distinct domains within a single carrier (Shima et al., 1999; Keller et al., 2001; Kreitzer et al., 2003; Polishchuk et al., 2004). Transport carriers all seem to be capable of interacting with microtubules and of moving along them in a directional manner (either plus- or minus-end directed). In studies examining the movement of GFP-tagged transport carriers destined for the plasma membrane, for example, the carriers underwent switching between microtubule tracks highlighted with rhodamine tubulin (Toomre et al., 1999) and moved in a stop-and-go manner (Hirschberg et al., 1998; Toomre et al., 1999) that was dependent on a kinesin-like motor (Kreitzer et al., 2000). Dual-color TIR-FM furthermore showed that such carriers remained associated with microtubules all the way to within 100 nm of the plasma membrane (Schmoranzer and Simon, 2003). While transport carriers can still dock and fuse with acceptor membranes in the absence of microtubules (Hirschberg et al., 1998), in cell types in which carriers must traverse significant distances to reach their destinations, such as neurons, microtubules are necessary (Rudolf et al., 2001; Hume et al., 2001; Wu et al., 2001a; Nakata and Hirokawa, 2003). Many transport carriers use cytosolic coat proteins (i.e., clathrin, COPI, and COPII) to pinch off from a membrane surface (Bonifacino and Lippincott-Schwartz, 2003). Recent photobleaching experiments using GFP-tagged versions of these coat proteins have revealed that their membrane binding, polymerization, and release from membranes occur rapidly and by a process that is not dependent on the carrier pinching off the membrane as a coated vesicle (Wu et al., 2001b; Presley et al., 2002). The role of small GTPases (Arf1 and Sar1) in the regulation of coat protein dynamics on membranes has also been addressed using GFP (Vasudevan et al., 1998; Ward et al., 2001; Presley et al., 2002; AltanBonnet et al., 2003; Elsner et al., 2003; Weiss and Nilsson, 2003). In cells expressing mutants of these proteins held in their GTP-bound state, the GTPases and their corresponding coat and effector molecules became irreversibly bound to membranes, exhibiting no recovery in photobleaching experiments (Presley et al., 2002). This contrasted with the normal behavior of these GTPases and their effectors, in which rapid cycling between membrane-bound and cytoplasmic pools was observed (Vasudevan et al., 1998; Stephens et al., 2000; Ward et al., 2001; Presley et al., 2002; Altan-Bonnet et al., 2003; García-Mata et al., 2003). Just as GFP-based studies have helped clarify key properties of transport carriers, they have also revealed important characteristics of membrane-bound organelles. In particular, they have shown that all secretory and endocytic organelles continuously exchange components with each other and they can undergo extensive changes in their structural organization. As one example, the ER was shown to transform from a network of branching tubules into stacked membrane arrays [termed organized smooth ER (OSER)] in response to elevated levels of resident components containing a dimerizing form of GFP on their cytoplasmic domains (Fig. 14.5) (Snapp et al., 2003). Because GFP is known to be capable of dimerizing in an antiparallel orientation (Yang et al., 1996a) through a low-affinity mechanism (Zacharias et al., 2002), one explanation of the OSER phenomenon is that it is caused by dimerization of GFP on apposing ER membranes (Snapp et al., 2003). This

GFP REVELATIONS

Figure 14.5. Morphological perturbation of the ER that occurs in response to the overexpression of a dimerizing form of GFP attached to the cytoplasmic tail of cytochrome b(5), an ER resident protein. Note that at low levels of chimera expression the ER looks normal (left panel), whereas at high levels (right panel) the ER is converted into swirls and tightly compacted stacks of membranes (called organized smooth ER or OSER). No change in the reticular pattern of the ER observed in the left panel was observed when a monomeric form of GFP was used to generate the chimera (not shown). [Courtesy of Snapp et al. (2003).]

could cause these membranes to then stack into geometric shapes. Consistent with this, when the authors expressed a nondimerizing form of GFP, no OSER structures were generated in the cells. Another example of organelle dynamics relates to the Golgi apparatus, which contains hundreds of diverse protein components with roles in the processing and sorting of secretory cargo. Photobleaching studies examining the residency time of different Golgi proteins on Golgi membranes revealed that no class of protein persisted stably (Storrie et al., 1998; Zaal et al., 1999; Miles et al., 2001; Nichols et al., 2001; Ward et al., 2001). Golgi processing enzymes stayed on the Golgi for ~60 min, cargo proteins for ~30 min, and cargo receptors and peripheral proteins for ~1 min before moving to other intracellular locations within the cell. These findings, together with other GFP-based observations on the transformation of Golgi membranes in response to specific stimuli (Sciaky et al., 1997; Presley et al., 1998; Feng et al., 2003) and the Golgi’s disassembly and reassembly during mitosis (Zaal et al., 1999; Jokitalo et al., 2001; Shorter and Warren, 2002), have led to the view that the Golgi apparatus is a dynamic, steady-state membrane system in a constant state of growth and consumption. New insights into the characteristics of lysosomes, which receive and digest endocytic cargo by hydrolytic enzymes, and of endosomes, which transfer material between the plasma membrane and other organelles, have also been obtained using GFP-based imaging approaches. Studies using PA-GFP photoactivation revealed that lysosomes undergo rapid interlysosome protein exchange (Patterson and Lippincott-Schwartz, 2002). This exchange was demonstrated in experiments in which a small population of lysosomes labeled with the lysosomal membrane protein (lgp120) tagged with PA-GFP was photoactivated and monitored over time. Because nearly all lysosomes were fluorescent within 20 min, transfer of lysosomal proteins between lysosomes is a rapid and extensive process. To clarify the properties of endosomes, Zerial and colleagues (Sönnichsen et al., 2000) used GFP-tagged Rab proteins and found that endosomes are comprised of a mosaic of

323

324

THE USES OF GREEN FLUORESCENT PROTEIN IN MAMMALIAN CELLS

domains enriched in Rab4, Rab5, and Rab11. These domains are dynamic yet do not significantly intermix, and they display differential pharmacological sensitivities. GFP-based studies have provided new revelations regarding the properties of the plasma membrane, including the mechanism(s) for recruitment of signaling complexes to receptors (Choy et al., 1999; Gillham et al., 1999; Varnai et al., 1999), the lifetime of activated receptors and their substrates on the plasma membrane (Wouters and Bastiaens, 1999; Smith et al., 2001), the complexity of endocytic uptake pathways (Benmerah et al., 1999; Gaidarov et al., 1999; Roberts et al., 1999; Nichols et al., 2001; Wu et al., 2001b; Bacia et al., 2002; Merrifield et al., 2002; Mundy et al., 2002; Nichols, 2002; Pelkmans et al., 2002; Thomsen et al., 2002; Engqvist-Goldstein et al., 2004), and the dynamics of lipid rafts (Zacharias et al., 2002; Glebov and Nichols, 2004; Kenworthy et al., 2004). Lipid rafts are thought to be small regions of membrane inhomogeneity enriched in cholesterol and glycosphingolipids. However, to what extent lipid raft domains concentrate membrane proteins under steady-state conditions is a controversial question (Kenworthy, 2002; Munro, 2003; Simons and Vaz, 2004). FRAP was used to systematically measure the diffusion coefficients of several types of GFP-tagged raft and nonraft components on the plasma membrane in response to raft perturbations (Kenworthy et al., 2004). Because different raft proteins (i.e., lipid-anchored, GPI-anchored, and transmembrane proteins) were found to freely diffuse over large distances (>4 mm) at completely different rates, the data ruled out models in which raft proteins undergo long-range diffusion as part of discrete, stable raft domains. Instead, they supported the view that raft proteins rapidly partition into and out of cholesterol-enriched membrane domains.

14.6

FUTURE DIRECTIONS

The hundreds of mammalian studies using GFP-based techniques that have been published to date (many of which are unmentioned in this review) attest to the revolutionary impact that GFP is now having on efforts to understand cellular processes and protein function in the complex environment within cells. Clearly, we are in a new era in which the continued developments in GFP techniques and applications, as well as in microscopy approaches, are having a tremendous impact in the study of protein dynamics and interactions within cells. Looking to the future, the engineering of new GFP-like fluorophores and reporter classes will be important given their potential for improving the detection limits and in vivo applicability of fluorescence-based reporters. Brighter and more red-shifted fluorescent proteins, for example, can provide probes for greater tissue penetration or as readouts for high-throughput approaches, as well as serve as additional tags for multispectral imaging and FRET-based methods. New developments in GFP-based indicators that are designed to respond to various biological events and signals will also be valuable. Currently, various biochemical parameters can be measured by the modulation of fluorescent spectra, such as pH using pHluorins, where fluorescence is reversibly quenched by low pH (Miesenböck et al., 1998), halide with the halide-sensitive YFP (Jayaraman et al., 2000), or indirect measurement of phosphorylation of tyrosine residues by recruitment of a YFP-tagged phosphotyrosinebinding SH2 domain construct (Kirchner et al., 2003). Indicators that rely on intramolecular FRET include those that measure protease activity (consisting of BFA and GFP with a protease-sensitive linker; Mahajan et al., 1999), direct phosphorylation using phocuses (Sato et al., 2002), or changes in calcium levels (i.e., cameleons, which consist of calmod-

REFERENCES

ulin and a CaM-binding protein sandwiched between YFP and CFP) (Miyawaki et al., 1997; Zaccolo et al., 2002, Zhang et al., 2002). By designing indicators for other cellular parameters, including those that are sensitive to metabolite concentrations or enzyme activity, or which show increased sensitivity for structural changes in a protein (Zhang et al., 2002), a set of powerful tools for probing the environment within a cell will become available. The incorporation of advanced microscopy techniques into everyday experiments will be needed to maximize the advantages of the new GFP-based reagents. One promising technique is single-molecule spectroscopy, which allows the visualization of specific molecular interactions such as EGF receptor dimerization (Sako et al., 2000) or conformational changes in voltage-gated ion channels (Harms et al., 2001). A different technique is correlative light-electron microscopy, which allows the distribution of molecules in a single fluorescent image to be analyzed at the electron microscopic level (Polishchuk et al., 2000). Fluorescence anisotropy microscopy is another emerging technique that elucidates the microenvironment of a protein by measuring the protein’s rotational diffusion (Rocheleau et al., 2003). Many other existing imaging techniques, including two-photon (Piston, 1999), TIR-FM, FCS, image correlation microscopy (ICM) (Petersen et al., 1993; Wiseman et al., 2004) and stimulated emission depletion (Dyba et al., 2003), should become more widely used given their ability, or potential, to visualize and quantify molecules and events at high spatial and temporal resolution. The enormous amount of data generated from these methods will necessitate the use of kinetic modeling and analysis tools in order to interpret the data. In summary, as newer versions of GFP and imaging techniques become available, the applications for GFP and GFP chimeras will continue to expand, from the analysis of small subcellular structures such as viruses, proteasomes, and mRNA, to movement in threedimensional matrices (Cukierman et al., 2001; Hegerfeldt et al., 2002; Petroll et al., 2003), imaging within living organs (Mempel et al., 2004), and whole-body in vivo imaging (Contag et al., 1998; Bacskai et al., 2003; Cook and Griffin, 2003; Zhang et al., 2004). In so doing, they will provide exciting new insights into the biology of mammalian cells.

REFERENCES Altan-Bonnet, N., Phair, R. D., Polishchuk, R. S., Weigert, R., and Lippincott-Schwartz, J. (2003). A role for Arf1 in mitotic Golgi disassembly, chromosome segregation, and cytokinesis. Proc. Natl. Acad. Sci. USA 100:13314–13319. Anderson, M. T., Tjioe, I. M., Lorincz, M. C., Parks, D. R., Herzenberg, L. A., Nolan, G. P., and Herzenberg, L. A. (1996). Simultaneous fluorescence-activated cell sorter analysis of two distinct transcriptional elements within a single cell using engineered green fluorescent proteins. Proc. Natl. Acad. Sci. USA 93:8508–8511. 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 99:12651–12656. Axelrod, D. (2001). Total internal reflection fluorescence microscopy in cell biology. Traffic 2:764–774. Bacia, K., Majoul, I. V., and Schwille, P. (2002). Probing the endocytic pathway in live cells using dual-color fluorescence cross-correlation analysis. Biophys. J. 83:1184–1193. Bacskai, B. J., Hickey, G. A., Skoch, J., Kajdasz, S. T., Wang, Y., Huang, G.-F., Mathis, C. A., Klunk, W. E., and Hyman, B. T. (2003). Four-dimensional multiphoton imaging of brain entry, amyloid

325

326

THE USES OF GREEN FLUORESCENT PROTEIN IN MAMMALIAN CELLS

binding, and clearance of an amyloid-b ligand in transgenic mice. Proc. Natl. Acad. Sci. USA 100:12462–12467. Bartlett, R. J., Singer, J. T., Shean, M. K., Weindler, F. W., and Secore, S. L. (1995). Expression of a fluorescent reporter gene in skeletal muscle cells from a virus-based vector: Implications for diabetes treatment. Transplant. Proc. 27:3344. Bastiaens, P. I. H., and Squire, A. (1999). Fluorescence lifetime imaging microscopy: spatial resolution of biochemical processes in the cell. Trends Cell Biol. 9:48–52. Beaudouin, J., Gerlich, D., Daigle, N., Eils, R., and Ellenberg, J. (2002). Nuclear envelope breakdown proceeds by microtubule-induced tearing of the lamina. Cell 108:83–96. Becker, M., Baumann, C., John, S., Walker, D. A., Vigneron, M., McNally, J. G., and Hager, G. L. (2002). Dynamic behavior of transcription factors on a natural promoter in living cells. EMBO Rep. 3:1188–1194. Benmerah, A., Bayrou, M., Cerf-Bensussan, N., and Dautry-Varsat, A. (1999). Inhibition of clathrincoated pit assembly by an Eps15 mutant. J. Cell Sci. 112:1303–1311. Bergeland, T., Widerberg, J., Bakke, O., and Nordeng, T. W. (2001). Mitotic partitioning of endosomes and lysosomes. Curr. Biol. 11:644–651. Bergsma, C. B. J., Streekstra, G. J., Smeulders, A. W. M., and Manders, E. M. M. (2001). Velocity estimation of spots in three-dimensional confocal image sequences of living cells. Cytometry 43:261–272. Bevis, B. J., and Glick, B. S. (2002). Rapidly maturing variants of the Discosoma red fluorescent protein (DsRed). Nat. Biotechnol. 20:83–87. Blum, R., Stephens, D. J., and Schulz, I. (2000). Lumenal targeted GFP, used as a marker of soluble cargo, visualizes rapid ERGIC to Golgi traffic by a tubulo-vesicular network. J. Cell Sci. 113:3151–3159. Bonifacino, J. S., and Lippincott-Schwartz, J. (2003). Coat proteins: shaping membrane transport. Nat. Rev. Mol. Cell Biol. 4:409–414. Bubulya, P. A., and Spector, D. L. (2004). “On the movements of nuclear components in living cells.” Exp. Cell Res. 296:4–11. Burke, B., and Ellenberg, J. (2002). Remodelling the walls of the nucleus. Nat. Rev. Mol. Cell Biol. 3:487–497. Calleja, V., Ameer-Beg, S. M., Vojnovic, B., Woscholski, R., Downward, J., and Larijani, B. (2003). Monitoring conformational changes of proteins in cells by fluorescence lifetime imaging microscopy. Biochem. J. 372:33–40. Campbell, R. E., Tour, O., Palmer, A. E., Steinbach, P. A., Baird, G. S., Zacharias, D. A., and Tsien, R. Y. (2002). A monomeric red fluorescent protein. Proc. Natl. Acad. Sci. USA 99: 7877–7882. 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. Chao, D. S., Hay, J. C., Winnick, S., Prekeris, R., Klumperman, J., and Scheller, R. H. (1999). SNARE membrane trafficking dynamics in vivo. J. Cell Biol. 144:869–881. Chen, J., and Ahlquist, P. (2000). Brome mosaic virus polymerase-like protein 2a is directed to the endoplasmic reticulum by helicase-like viral protein 1a. J. Virol. 74:4310–4318. Chiu, W.-L., 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. Choy, E., Chiu, V. K., Silletti, J., Feoktistov, M., Morimoto, T., Michaelson, D., Ivanov, I. E., and Philips, M. R. (1999). Endomembrane trafficking of Ras: The CAAX motif targets proteins to the ER and Golgi. Cell 98:69–80. Christensen, M. O., Larsen, M. K., Barthelmes, H. U., Hock, R., Andersen, C. L., Kjeldsen, E., Knudsen, B. R., Westergaard, O., Boege, F., and Mielke, C. (2002). Dynamics of human DNA topoisomerases IIa and IIb in living cells. J. Cell Biol. 157:31–44.

REFERENCES

Chubb, J. R., Boyle, S., Perry, P., and Bickmore, W. A. (2002). Chromatin motion is constrained by association with nuclear compartments in human cells. Curr. Biol. 12:439–445. Chudakov, D. M., Belousov, V. V., Zaraisky, A. G., Novoselov, V. V., Staroverov, D. B., Zorov, D. B., Lukyanov, S., and Lukyanov, K. A. (2003). Kindling fluorescent proteins for precise in vivo photolabeling. Nat. Biotechnol. 21:191–194. Clegg, R. M. (1995). Fluorescence resonance energy transfer. Curr. Opin. Biotechnol. 6:103–110. Cole, N. B., Smith, C. L., Sciaky, N., Terasaki, M., Edidin, M., and Lippincott-Schwartz, J. (1996). Diffusional mobility of Golgi proteins in membranes of living cells. Science 273: 797–801. Conrad, C., Erfle, H., Warnat, P., Daigle, N., Lörch, T., Ellenberg, J., Pepperkok, R., and Eils, R. (2004). Automatic identification of subcellular phenotypes on human cell arrays. Genome Res. 14:1130–1136. Contag, P. R., Olomu, I. N., Stevenson, D. K., and Contag, C. H. (1998). Bioluminescent indicators in living mammals. Nat. Med. 4:245–247. Cook, L. B., and Hinkle, P. M. (2004). Fate of internalized thyrotropin-releasing hormone receptors monitored with a timer fusion protein. Endocrinology 145:3095–3100. Cook, S. H., and Griffin, D. E. (2003). Luciferase imaging of a neurotropic viral infection in intact animals. J. Virol. 77:5333–5338. Cormack, B. P., Valdivia, R. H., and Falkow, S. (1996). FACS-optimized mutants of the green fluorescent protein (GFP). Gene 173:33–38. 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. Cukierman, E., Pankov, R., Stevens, D. R., and Yamada, K. M. (2001). Taking cell-matrix adhesions to the third dimension. Science 294:1708–1712. Dahm, T., White, J., Grill, S., Füllekrug, J., and Stelzer, E. H. K. (2001). Quantitative ER ´ Golgi transport kinetics and protein separation upon Golgi exit revealed by vesicular integral membrane protein 36 dynamics in live cells. Mol. Biol. Cell 12:1481–1498. Daigle, N., Beaudouin, J., Hartnell, L., Imreh, G., Hallberg, E., Lippincott-Schwartz, J., and Ellenberg, J. (2001). Nuclear pore complexes form immobile networks and have a very low turnover in live mammalian cells. J. Cell Biol. 154:71–84. 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. Biotechnol. 13:151–154. Dhandayuthapani, S., Via, L. E., Thomas, C. A., Horowitz, P. M., Deretic, D., and Deretic, V. (1995). Green fluorescent protein as a marker for gene expression and cell biology of mycobacterial interactions with macrophages. Mol. Microbiol. 17:901–912. Dietzel, S., and Belmont, A. S. (2001). Reproducible but dynamic positioning of DNA in chromosomes during mitosis. Nat. Cell Biol. 3:767–770. Dorsky, D. I., Wells, M., and Harrington, R. D. (1996). Detection of HIV-1 infection with a green fluorescent protein reporter system. J. Acquir. Immune Defic. Syndr. Hum. Retrovirol. 13:308–313. Dundr, M., McNally, J. G., Cohen, J., and Misteli, T. (2002). Quantitation of GFP-fusion proteins in single living cells. J. Struct. Biol. 140:92–99. Dundr, M., Hebert, M. D., Karpova, T. S., Stanek, D., Xu, H., Shpargel, K. B., Meier, U. T., Neugebauer, K. M., Matera, A. G., and Misteli, T. (2004). In vivo kinetics of Cajal body components. J. Cell Biol. 164:831–842. Dunn, G. A., Dobbie, I. M., Monypenny, J., Holt, M. R., and Zicha, D. (2002). Fluorescence localization after photobleaching (FLAP): a new method for studying protein dynamics in living cells. J. Microsc. 205:109–112. Dyba, M., Jakobs, S., and Hell, S. W. (2003). Immunofluorescence stimulated emission depletion microscopy. Nat. Biotechnol. 21:1303–1304.

327

328

THE USES OF GREEN FLUORESCENT PROTEIN IN MAMMALIAN CELLS

Edidin, M. (1994). Fluorescence photobleaching and recovery, FPR, in the analysis of membrane structure and dynamics. In Mobility and Proximity in Biological Membranes, Damjanovich, S., Edidin, M., Szollosi, J., and Tron, L., Eds., CRC, Boca Raton, FL, pp. 109–135. 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. Eils, R., Gerlich, D., Tvaruskó, W., Spector, D. L., and Misteli, T. (2000). Quantitative imaging of pre-mRNA splicing factors in living cells. Mol. Biol. Cell 11:413–418. Ellenberg, J., Siggia, E. D., Moreira, J. E., Smith, C. L., Presley, J. F., Worman, H. J., and Lippincott-Schwartz, J. (1997). Nuclear membrane dynamics and reassembly in living cells: targeting of an inner nuclear membrane protein in interphase and mitosis. J. Cell Biol. 138:1193– 1206. Ellenberg, J., Lippincott-Schwartz, J., and Presley, J. F. (1998). Two-color green fluorescent protein time-lapse imaging. Biotechniques 25:838–846. Elliott, G., and O’Hare, P. (1999). Live-cell analysis of a green fluorescent protein-tagged herpes simplex virus infection. J. Virol. 73:4110–4119. 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. Elsner, M., Hashimoto, H., Simpson, J. C., Cassel, D., Nilsson, T., and Weiss, M. (2003). Spatiotemporal dynamics of the COPI vesicle machinery. EMBO Rep. 4:1000–1004. Engqvist-Goldstein, A. E. Y., Zhang, C. X., Carreno, S., Barroso, C., Heuser, J. E., and Drubin, D. G. (2004). RNAi-mediated Hip1R silencing results in stable association between the endocytic machinery and the actin assembly machinery. Mol. Biol. Cell 15:1666–1679. Espinet, C., Gómez-Arbonés, X., Egea, J., and Comella, J. X. (2000). Combined use of the green and yellow fluorescent proteins and fluorescence-activated cell sorting to select populations of transiently transfected PC12 cells. J. Neurosci. Methods 100:63–69. Essers, J., Houtsmuller, A. B., van Veelen, L., Paulusma, C., Nigg, A. L., Pastink, A., Vermeulen, W., Hoeijmakers, J. H. J., and Kanaar, R. (2002). Nuclear dynamics of RAD52 group homologous recombination proteins in response to DNA damage. EMBO J. 21:2030–2037. Feng, Y., Yu, S., Lasell, T. K. R., Jadhav, A. P., Macia, E., Chardin, P., Melancon, P., Roth, M., Mitchison, T., and Kirchhausen, T. (2003). Exo1: A new chemical inhibitor of the exocytic pathway. Proc. Natl. Acad. Sci. USA 100:6469–6474. Fradin, C., Abu-Arish, A., Granek, R., and Elbaum, M. (2003). Fluorescence correlation spectroscopy close to a fluctuating membrane. Biophys. J. 84:2005–2020. Gaidarov, I., Santini, F., Warren, R. A., and Keen, J. H. (1999). Spatial control of coated-pit dynamics in living cells. Nat. Cell Biol. 1:1–7. García-Mata, R., Szul, T., Alvarez, C., and Sztul, E. (2003). ADP-ribosylation factor/COPI-dependent events at the endoplasmic reticulum-Golgi interface are regulated by the guanine nucleotide exchange factor GBF1. Mol. Biol. Cell 14:2250–2261. Gerlich, D., Beaudouin, J., Gebhard, M., Ellenberg, J., and Eils, R. (2001). Four-dimensional imaging and quantitative reconstruction to analyse complex spatiotemporal processes in live cells. Nat. Cell Biol. 3:852–855. Gerlich, D., Beaudouin, J., Kalbfuss, B., Daigle, N., Eils, R., and Ellenberg, J. (2003). Global chromosome positions are transmitted through mitosis in mammalian cells. Cell 112:751–764. Gillham, H., Golding, M. C. H. M., Pepperkok, R., and Gullick, W. J. (1999). Intracellular movement of green fluorescent protein-tagged phosphatidylinositol 3-kinase in response to growth factor receptor signaling. J. Cell Biol. 146:869–880. Glebov, O. O., and Nichols, B. J. (2004). Lipid raft proteins have a random distribution during localized activation of the T-cell receptor. Nat. Cell Biol. 6:238–243.

REFERENCES

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. 276:29188–29194. Griffis, E. R., Altan, N., Lippincott-Schwartz, J., and Powers, M. A. (2002). Nup98 is a mobile nucleoporin with transcription-dependent dynamics. Mol. Biol. Cell 13:1282–1297. Gurskaya, N. G., Fradkov, A. F., Terskikh, A., Matz, M. V., Labas, Y. A., Martynov, V. I., Yanushevich, Y. G., Lukyanov, K. A., and Lukyanov, S. A. (2001). GFP-like chromoproteins as a source of far-red fluorescent proteins. FEBS Lett. 507:16–20. Haraguchi, T., Ding, D. Q., Yamamoto, A., Kaneda, T., Koujin, T., and Hiraoka, Y. (1999). Multiple-color fluorescence imaging of chromosomes and microtubules in living cells. Cell Struct. Funct. 24:291–298. Haraguchi, T., Koujin, T., Hayakawa, T., Kaneda, T., Tsutsumi, C., Imamoto, N., Akazawa, C., Sukegawa, J., Yoneda, Y., and Hiraoka, Y. (2000). Live fluorescence imaging reveals early recruitment of emerin, LBR, RanBP2, and Nup153 to reforming functional nuclear envelopes. J. Cell Sci. 113:779–794. Harms, G. S., Cognet, L., Lommerse, P. H. M., Blab, G. A., Kahr, H., Gamsjäger, R., Spaink, H. P., Soldatov, N. M., Romanin, C., and Schmidt, T. (2001). Single-molecule imaging of L-type Ca2+ channels in live cells. Biophys. J. 81:2639–2646. Hayes, N., Howard-Cofield, E., and Gullick, W. (2004). Green fluorescent protein as a tool to study epidermal growth factor receptor function. Cancer Lett. 206:129–135. Hegerfeldt, Y., Tusch, M., Bröcker, E.-B., and Friedl, P. (2002). Collective cell movement in primary melanoma explants: Plasticity of cell–cell interaction, b1-integrin function, and migration strategies. Cancer Res. 62:2125–2130. 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. Heim, R., Prasher, D. C., and Tsien, R. Y. (1994). Wavelength mutations and posttranslational autoxidation of green fluorescent protein. Proc. Natl. Acad. Sci. USA 91:12501–12504. Heim, R., Cubitt, A. B., and Tsien, R. Y. (1995). Improved green fluorescence. Nature 373:663–664. Hirschberg, K., Miller, C. M., Ellenberg, J., Presley, J. F., Siggia, E. D., Phair, R. D., and Lippincott-Schwartz, J. (1998). Kinetic analysis of secretory protein traffic and characterization of Golgi to plasma membrane transport intermediates in living cells. J. Cell Biol. 143:1485– 1503. Holaska, J. M., Wilson, K. L., and Mansharamani, M. (2002). The nuclear envelope, lamins and nuclear assembly. Curr. Opin. Cell Biol. 14:357–364. Houtsmuller, A. B., Rademakers, S., Nigg, A. L., Hoogstraten, D., Hoeijmakers, J. H. J., and Vermeulen, W. (1999). Action of DNA repair endonuclease ERCC1/XPF in living cells. Science 284:958–961. Hume, A. N., Collinson, L. M., Rapak, A., Gomes, A. Q., Hopkins, C. R., and Seabra, M. C. (2001). Rab27a regulates the peripheral distribution of melanosomes in melanocytes. J. Cell Biol. 152:795–808. 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. Janicki, S. M., and Spector, D. L. (2003). Nuclear choreography: Interpretations from living cells. Curr. Opin. Cell Biol. 15:149–157. Jayaraman, S., Haggie, P., Wachter, R. M., Remington, S. J., and Verkman, A. S. (2000). Mechanism and cellular applications of a green fluorescent protein-based halide sensor. J. Biol. Chem. 275:6047–6050. Jokitalo, E., Cabrera-Poch, N., Warren, G., and Shima, D. T. (2001). Golgi clusters and vesicles mediate mitotic inheritance independently of the endoplasmic reticulum. J. Cell Biol. 154:317–330.

329

330

THE USES OF GREEN FLUORESCENT PROTEIN IN MAMMALIAN CELLS

Kaether, C., and Gerdes, H.-H. (1995). Visualization of protein transport along the secretory pathway using green fluorescent protein. FEBS Lett. 369:267–271. Keller, P., Toomre, D., Diaz, E., White, J., and Simons, K. (2001). Multicolour imaging of postGolgi sorting and trafficking in live cells. Nat. Cell Biol. 3:140–149. Kenworthy, A. (2002). Peering inside lipid rafts and caveolae. Trends Biochem. Sci. 27:435–437. Kenworthy, A. K., Nichols, B. J., Remmert, C. L., Hendrix, G. M., Kumar, M., Zimmerberg, J., and Lippincott-Schwartz, J. (2004). Dynamics of putative raft-associated proteins at the cell surface. J. Cell Biol. 165:735–746. Kimura, H., and Cook, P. R. (2001). Kinetics of core histones in living human cells: Little exchange of H3 and H4 and some rapid exchange of H2B. J. Cell Biol. 153:1341–1353. Kimura, H., Sugaya, K., and Cook, P. R. (2002). The transcription cycle of RNA polymerase II in living cells. J. Cell Biol. 159:777–782. Kirchner, J., Kam, Z., Tzur, G., Bershadsky, A. D., and Geiger, B. (2003). Live-cell monitoring of tyrosine phosphorylation in focal adhesions following microtubule disruption. J. Cell Sci. 116:975–986. Kreitzer, G., Marmorstein, A., Okamoto, P., Vallee, R., and Rodriguez-Boulan, E. (2000). Kinesin and dynamin are required for post-Golgi transport of a plasma-membrane protein. Nat. Cell Biol. 2:125–127. Kreitzer, G., Schmoranzer, J., Low, S. H., Li, X., Gan, Y., Weimbs, T., Simon, S. M., and RodriguezBoulan, E. (2003). Three-dimensional analysis of post-Golgi carrier exocytosis in epithelial cells. Nat. Cell Biol. 5:126–136. Krichevsky, O., and Bonnet, G. (2002). Fluorescence correlation spectroscopy: the technique and its applications. Rep. Prog. Phys. 65:251–297. Krylyshkina, O., Anderson, K. I., Kaverina, I., Upmann, I., Manstein, D. J., Small, J. V., and Toomre, D. K. (2003). Nanometer targeting of microtubules to focal adhesions. J. Cell Biol. 161:853–859. La Boissière, S., Izeta, A., Malcomber, S., and O’Hare, P. (2004). Compartmentalization of VP16 in cells infected with recombinant herpes simplex virus expressing VP16-green fluorescent protein fusion proteins. J. Virol. 78:8002–8014. Lakowicz, J. R., Szmacinski, H., Nowaczyk, K., Berndt, K. W., and Johnson, M. (1992). Fluorescence lifetime imaging. Anal. Biochem. 202:316–330. Lamond, A. I., and Spector, D. L. (2003). Nuclear speckles: a model for nuclear organelles. Nat. Rev. Mol. Cell Biol. 4:605–612. Larson, D. R., Ma, Y. M., Vogt, V. M., and Webb, W. W. (2003). Direct measurement of Gag–Gag interaction during retrovirus assembly with FRET and fluorescence correlation spectroscopy. J. Cell Biol. 162:1233–1244. Lever, M. A., Th’ng, J. P., Sun, X., and Hendzel, M. J. (2000). Rapid exchange of histone H1.1 on chromatin in living human cells. Nature 408:873–876. Lin, H.-J., Herman, P., and Lakowicz, J. R. (2003). Fluorescence lifetime-resolved pH imaging of living cells. Cytometry 52A:77–89. Lippincott-Schwartz, J., Snapp, E., and Kenworthy, A. (2001). Studying protein dynamics in living cells. Nat. Rev. Mol. Cell Biol. 2:444–456. Ludin, B., and Matus, A. (1998). GFP illuminates the cytoskeleton. Trends Cell Biol. 8:72–77. Mahajan, N. P., Harrison-Shostak, D. C., Michaux, J., and Herman, B. (1999). Novel mutant green fluorescent protein protease substrates reveal the activation of specific caspases during apoptosis. Chem. Biol. 6:401–409. Majoul, I., Straub, M., Hell, S. W., Duden, R., and Söling, H. D. (2001). KDEL-cargo regulates interactions between proteins involved in COPI vesicle traffic: Measurements in living cells using FRET. Dev. Cell 1:139–153. Mancia, F., Patel, S. D., Rajala, M. W., Scherer, P. E., Nemes, A., Schieren, I., Hendrickson, W. A., and Shapiro, L. (2004). Optimization of protein production in mammalian cells with a coexpressed fluorescent marker. Structure 12:1355–1360.

REFERENCES

Manneville, J.-B., Etienne-Manneville, S., Skehel, P., Carter, T., Ogden, D., and Ferenczi, M. (2003). Interaction of the actin cytoskeleton with microtubules regulates secretory organelle movement near the plasma membrane in human endothelial cells. J. Cell Sci. 116:3927–3938. Marshall, J., Molloy, R., Moss, G., Howe, J. R., and Hughes, T. E. (1995). The jellyfish green fluorescent protein: A new tool for studying ion channel expression. Neuron 14:211–215. Mayor, S., Presley, J. F., and Maxfield, F. R. (1993). Sorting of membrane components from endosomes and subsequent recycling to the cell surface occurs by a bulk flow process. J. Cell Biol. 121:1257–1269. McNally, J. G., Müller, W. G., Walker, D., Wolford, R., and Hager, G. L. (2000). The glucocorticoid receptor: Rapid exchange with regulatory sites in living cells. Science 287:1262–1265. Mempel, T. R., Henrickson, S. E., and von Andrian, U. H. (2004). T-cell priming by dendritic cells in lymph nodes occurs in three distinct phases. Nature 427:154–159. Merrifield, C. J., Feldman, M. E., Wan, L., and Almers, W. (2002). Imaging actin and dynamin recruitment during invagination of single clathrin-coated pits. Nat. Cell Biol. 4:691–698. Meulenbroek, R. A., Sargent, K. L., Lunde, J., Jasmin, B. J., and Parks, R. J. (2004). Use of adenovirus protein IX (pIX) to display large polypeptides on the virion—Generation of fluorescent virus through the incorporation of pIX-GFP. Mol. Ther. 9:617–624. Miesenböck, G., De Angelis, D. A., and Rothman, J. E. (1998). Visualizing secretion and synaptic transmission with pH-sensitive green fluorescent proteins. Nature 394:192–195. Miles, S., McManus, H., Forsten, K. E., and Storrie, B. (2001). Evidence that the entire Golgi apparatus cycles in interphase HeLa cells: sensitivity of Golgi matrix proteins to an ER exit block. J. Cell Biol. 155:543–555. Mironov, A. A., Mironov, A. A., Jr., Beznoussenko, G. V., Trucco, A., Lupetti, P., Smith, J. D., Geerts, W. J. C., Koster, A. J., Burger, K. N. J., Martone, M. E., Deerinck, T. J., Ellisman, M. H., and Luini, A. (2003). ER-to-Golgi carriers arise through direct en bloc protrusion and multistage maturation of specialized ER exit domains. Dev. Cell 5:583–594. Misteli, T. (2001). The concept of self-organization in cellular architecture. J. Cell Biol. 155:181–185. Misteli, T., Gunjan, A., Hock, R., Bustin, M., and Brown, D. T. (2000). Dynamic binding of histone H1 to chromatin in living cells. Nature 408:877–881. Mitra, R. D., Silva, C. M., and Youvan, D. C. (1996). Fluorescence resonance energy transfer between blue-emitting and red-shifted excitation derivatives of the green fluorescent protein. Gene 173:13–17. Miyawaki, A., Llopis, J., Heim, R., McCaffery, J. M., Adams, J. A., Ikura, M., and Tsien, R. Y. (1997). Fluorescent indicators for Ca2+ based on green fluorescent proteins and calmodulin. Nature 388:882–887. Moir, R. D., Yoon, M., Khuon, S., and Goldman, R. D. (2000). Nuclear lamins A and B1: Different pathways of assembly during nuclear envelope formation in living cells. J. Cell Biol. 151:1155–1168. Moradpour, D., Evans, M. J., Gosert, R., Yuan, Z., Blum, H. E., Goff, S. P., Lindenbach, B. D., and Rice, C. M. (2004). Insertion of green fluorescent protein into non-structural protein 5A allows direct visualization of functional hepatitis C virus replication complexes. J. Virol. 78:7400–7409. Moriyoshi, K., Richards, L. J., Akazawa, C., O’Leary, D. D., and Nakanishi, S. (1996). Labeling neural cells using adenoviral gene transfer of membrane-targeted GFP. Neuron 16:255–260. Mosser, D. D., Caron, A. W., Bourget, L., Jolicoeur, P., and Massie, B. (1997). Use of a dicistronic expression cassette encoding the green fluorescent protein for the screening and selection of cells expressing inducible gene products. Biotechniques 22:150–154.

331

332

THE USES OF GREEN FLUORESCENT PROTEIN IN MAMMALIAN CELLS

Mundy, D. I., Machleidt, T., Ying, Y. S., Anderson, R. G., and Bloom, G. S. (2002). Dual control of caveolar membrane traffic by microtubules and the actin cytoskeleton. J. Cell Sci. 115:4327– 4339. Munro, S. (2003). Lipid rafts: elusive or illusive? Cell 115:377–388. 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. Nakata, T., and Hirokawa, N. (2003). Microtubules provide directional cues for polarized axonal transport through interaction with kinesin motor head. J. Cell Biol. 162:1045–1055. Nehls, S., Snapp, E. L., Cole, N. B., Zaal, K. J., Kenworthy, A. K., Roberts, T. H., Ellenberg, J., Presley, J. F., Siggia, E., and Lippincott-Schwartz, J. (2000). Dynamics and retention of misfolded proteins in native ER membranes. Nat. Cell Biol. 2:288–295. Ng, T., Squire, A., Hansra, G., Bornancin, F., Prevostel, C., Hanby, A., Harris, W., Barnes, D., Schmidt, S., Mellor, H., Bastiaens, P. I., and Parker, P. J. (1999). Imaging protein kinase Ca activation in cells. Science 283:2085–2089. Nichols, B. J. (2002). A distinct class of endosome mediates clathrin-independent endocytosis to the Golgi complex. Nat. Cell Biol. 4:374–378. Nichols, B. J., Kenworthy, A. K., Polishchuk, R. S., Lodge, R., Roberts, T. H., Hirschberg, K., Phair, R. D., and Lippincott-Schwartz, J. (2001). Rapid cycling of lipid raft markers between the cell surface and Golgi complex. J. Cell Biol. 153:529–541. Niswender, K. D., Blackman, S. M., Rohde, L., Magnuson, M. A., and Piston, D. W. (1995). Quantitative imaging of green fluorescent protein in cultured cells: Comparison of microscopic techniques, use in fusion proteins and detection limits. J. Microsc. 180:109–116. Ogawa, H., Inouye, S., Tsuji, F. I., Yasuda, K., and Umesono, K. (1995). Localization, trafficking, and temperature-dependence of the Aequorea green fluorescent protein in cultured vertebrate cells. Proc. Natl. Acad. Sci. USA 92:11899–11903. Olson, K. R., McIntosh, J. R., and Olmsted, J. B. (1995). Analysis of MAP 4 function in living cells using green fluorescent protein (GFP) chimeras. J. Cell Biol. 130:639–650. Ormö, 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. Ostlund, C., Ellenberg, J., Hallberg, E., Lippincott-Schwartz, J., and Worman, H. J. (1999). Intracellular trafficking of emerin, the Emery–Dreifuss muscular dystrophy protein. J. Cell Sci. 112:1709–1719. Patterson, G. H., and Lippincott-Schwartz, J. (2002). A photoactivatable GFP for selective photolabeling of proteins and cells. Science 297:1873–1877. Peel, A. L., Zolotukhin, S., Schrimsher, G. W., Muzyczka, N., and Reier, P. J. (1997). Efficient transduction of green fluorescent protein in spinal cord neurons using adeno-associated virus vectors containing cell type-specific promoters. Gene Ther. 4:16–24. Pelkmans, L., Püntener, D., and Helenius, A. (2002). Local actin polymerization and dynamin recruitment in SV40-induced internalization of caveolae. Science 296:535–539. Pepperkok, R., Squire, A., Geley, S., and Bastiaens, P. I. (1999). Simultaneous detection of multiple green fluorescent proteins in live cells by fluorescence lifetime imaging microscopy. Curr. Biol. 9:269–272. Petersen, N. O., Hoddelius, P. L., Wiseman, P. W., Seger, O., and Magnusson, K. E. (1993). Quantitation of membrane receptor distributions by image correlation spectroscopy: concept and application. Biophys. J. 65:1135–1146. Petroll, W. M., Ma, L., and Jester, J. V. (2003). Direct correlation of collagen matrix deformation with focal adhesion dynamics in living corneal fibroblasts. J. Cell Sci. 116:1481–1491.

REFERENCES

Phair, R. D., and Misteli, T. (2000). High mobility of proteins in the mammalian cell nucleus. Nature 404:604–609. Piston, D. W. (1999). Imaging living cells and tissues by two-photon excitation microscopy. Trends Cell Biol. 9:66–69. Polishchuk, R. S., Polishchuk, E. V., Marra, P., Alberti, S., Buccione, R., Luini, A., and Mironov, A. A. (2000). Correlative light-electron microscopy reveals the tubular-saccular ultrastructure of carriers operating between Golgi apparatus and plasma membrane. J. Cell Biol. 148:45– 58. Polishchuk, E. V., Di Pentima, A., Luini, A., and Polishchuk, R. S. (2003). Mechanism of constitutive export from the Golgi: bulk flow via the formation, protrusion, and en bloc cleavage of large trans-Golgi network tubular domains. Mol. Biol. Cell 14:4470–4485. Polishchuk, R., Di Pentima, A., and Lippincott-Schwartz, J. (2004). Delivery of raft-associated, GPIanchored proteins to the apical surface of polarized MDCK cells by a transcytotic pathway. Nat. Cell Biol. 6:297–307. Pollok, B. A., and Heim, R. (1999). Using GFP in FRET-based applications. Trends Cell Biol. 9:57–60. Potel., C., Kaelin, K., Gautier, I., Lebon, P., Coppey, J., and Rozenberg, F. (2002). Incorporation of green fluorescent protein into the essential envelope glycoprotein B of herpes simplex virus type 1. J. Virol. Methods 105:13–23. Presley, J. F., Cole, N. B., Schroer, T. A., Hirschberg, K., Zaal, K. J. M., and Lippincott-Schwartz, J. (1997). ER-to-Golgi transport visualized in living cells. Nature 389:81–85. Presley, J. F., Smith, C., Hirschberg, K., Miller, C., Cole, N. B., Zaal, K. J. M., and LippincottSchwartz, J. (1998). Golgi membrane dynamics. Mol. Biol. Cell 9:1617–1626. Presley, J. F., Ward, T. H., Pfeifer, A. C., Siggia, E. D., Phair, R. D., and Lippincott-Schwartz, J. (2002). Dissection of COPI and Arf1 dynamics in vivo and role in Golgi membrane transport. Nature 417:187–193. Pyenta, P. S., Holowka, D., and Baird, B. (2001). Cross-correlation analysis of inner-leaflet-anchored green fluorescent protein co-redistributed with IgE receptors and outer leaflet lipid raft components. Biophys. J. 80:2120–2132. 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. Rizzuto, R., Brini, M., De Giorgi, F., Rossi, R., Heim, R., Tsien, R. Y., and Pozzan, T. (1996). Double labelling of subcellular structures with organelle-targeted GFP mutants in vivo. Curr. Biol. 6:183–188. Robbins, J. R., Barth, A. I., Marquis, H., de Hostos, E. L., Nelson, W. J., and Theriot, J. A. (1999). Listeria monocytogenes exploits normal host cell processes to spread from cell to cell. J. Cell Biol. 146:1333–1350. Roberts, R. L., Barbieri, M. A., Pryse, K. M., Chua, M., Morisaki, J. H., and Stahl, P. D. (1999). Endosome fusion in living cells overexpressing GFP-rab5. J. Cell Sci. 112:3667– 3675. Rocheleau, J. V., Edidin, M., and Piston, D. W. (2003). Intrasequence GFP in class I MHC molecules, a rigid probe for fluorescence anisotropy measurements of the membrane environment. Biophys. J. 84:4078–4086. Ropp, J. D., Donahue, C. J., Wolfgang-Kimball, D., Hooley, J. J., Chin, J. Y., Cuthbertson, R. A., and Bauer, K. D. (1996). Aequorea green fluorescent protein: simultaneous analysis of wild-type and blue-fluorescing mutant by flow cytometry. Cytometry 24:284–288. Rudolf, R., Salm, T., Rustom, A., and Gerdes, H.-H. (2001). Dynamics of immature secretory granules: Role of cytoskeletal elements during transport, cortical restriction, and F-actin-dependent tethering. Mol. Biol. Cell 12:1353–1365.

333

334

THE USES OF GREEN FLUORESCENT PROTEIN IN MAMMALIAN CELLS

Sako, Y., Minoghchi, S., and Yanagida, T. (2000). Single-molecule imaging of EGFR signalling on the surface of living cells. Nat. Cell Biol. 2:168–172. Salina, D., Bodoor, K., Eckley, D. M., Schroer, T. A., Rattner, J. B., and Burke, B. (2002). Cytoplasmic dynein as a facilitator of nuclear envelope breakdown. Cell 108:97–107. Salmon, W. C., Adams, M. C., and Waterman-Storer, C. M. (2002). Dual-wavelength fluorescent speckle microscopy reveals coupling of microtubule and actin movements in migrating cells. J. Cell Biol. 158:31–37. Sato, M., Ozawa, T., Inukai, K., Asano, T., and Umezawa, Y. (2002). Fluorescent indicators for imaging protein phosphorylation in single living cells. Nat. Biotechnol. 20:287–294. Scales, S. J., Pepperkok, R., and Kreis, T. E. (1997). Visualization of ER-to-Golgi transport in living cells reveals a sequential mode of action for COPII and COPI. Cell 90:1137– 1148. Schmoranzer, J., and Simon, S. M. (2003). Role of microtubules in fusion of post-Golgi vesicles to the plasma membrane. Mol. Biol. Cell 14:1558–1569. Schmoranzer, J., Goulian, M., Axelrod, D., and Simon, S. M. (2000). Imaging constitutive exocytosis with total internal reflection fluorescence microscopy. J. Cell Biol. 149:23–32. Sciaky, N., Presley, J., Smith, C., Zaal, K. J. M., Cole, N., Moreira, J. E., Terasaki, M., Siggia, E., and Lippincott-Schwartz, J. (1997). Golgi tubule traffic and the effects of brefeldin A visualized in living cells. J. Cell Biol. 139:1137–1155. Shav-Tal, Y., Darzacq, X., Shenoy, S. M., Fusco, D., Janicki, S. M., Spector, D. L., and Singer, R. H. (2004). Dynamics of single mRNPs in nuclei of living cells. Science 304:1797–1800. Shima, D. T., Scales, S. J., Kreis, T. E., and Pepperkok, R. (1999). Segregation of COPI-rich and anterograde-cargo-rich domains in endoplasmic-reticulum-to-Golgi transport complexes. Curr. Biol. 9:821–824. Shorter, J., and Warren, G. (2002). Golgi architecture and inheritance. Annu. Rev. Cell Dev. Biol. 18:379–420. Simons, K., and Vaz, W. L. (2004). Model systems, lipid rafts, and cell membranes. Annu. Rev. Biophys. Biomol. Struct. 33:269–295. Sineshchekova, O. O., Kawate, T., Vdovychenko, O. V., and Sato, T. N. (2004). Protein-trap version 2.1: screening for expressed proteins in mammalian cells based on their localizations. BMC Cell Biol. 5:8. Smith, J., Yu, R., and Hinkle, P. M. (2001). Activation of MAPK by TRH requires clathrindependent endocytosis and PKC but not receptor interaction with b-arrestin or receptor endocytosis. Mol. Endocrinol. 15:1539–1548. Snapp, E. L., Hegde, R. S., Francolini, M., Lombardo, F., Colombo, S., Pedrazzini, E., Borgese, N., and Lippincott-Schwartz, J. (2003). Formation of stacked ER cisternae by low affinity protein interactions. J. Cell Biol. 163:257–269. Snapp, E. L., Reinhart, G. A., Bogert, B. A., Lippincott-Schwartz, J., and Hegde, R. S. (2004). The organization of engaged and quiescent translocons in the endoplasmic reticulum of mammalian cells. J. Cell Biol. 164:997–1007. Sönnichsen, B., De Renzis, S., Nielsen, E., Rietdorf, J., and Zerial, M. (2000). Distinct membrane domains on endosomes in the recycling pathway visualized by multicolor imaging of Rab4, Rab5, and Rab11. J. Cell Biol. 149:901–914. Stenoien, D. L., Nye, A. C., Mancini, M. G., Patel, K., Dutertre, M., O’Malley, B. W., Smith, C. L., Belmont, A. S., and Mancini, M. A. (2001). Ligand-mediated assembly and real-time cellular dynamics of estrogen receptor a-coactivator complexes in living cells. Mol. Cell. Biol. 21:4404–4412. Stephens, D. J., and Pepperkok, R. (2002). Imaging of procollagen transport reveals COPIdependent cargo sorting during ER-to-Golgi transport in mammalian cells. J. Cell Sci. 115:1149–1160.

REFERENCES

Stephens, D. J., and Pepperkok, R. (2004). Differential effects of a GTP-restricted mutant of Sar1p on segregation of cargo during export from the endoplasmic reticulum. J. Cell Sci. 117: 3635–3644. Stephens, D. J., Lin-Marq, N., Pagano, A., Pepperkok, R., and Paccaud, J. P. (2000). COPI-coated ER-to-Golgi transport complexes segregate from COPII in close proximity to ER exit sites. J. Cell Sci. 113:2177–2185. Storrie, B., White, J., Röttger, S., Stelzer, E. H., Suganuma, T., and Nilsson, T. (1998). Recycling of Golgi-resident glycosyltransferases through the ER reveals a novel pathway and provides an explanation for nocodazole-induced Golgi scattering. J. Cell Biol. 143:1505–1521. Stryer, L. (1978). Fluorescence energy transfer as a spectroscopic ruler. Annu. Rev. Biochem. 47:819–846. Sutherland, H. G., Mumford, G. K., Newton, K., Ford, L. V., Farrall, R., Dellaire, G., Caceres, J. F., and Bickmore, W. A. (2001). Large-scale identification of mammalian proteins localized to nuclear sub-compartments. Hum. Mol. Genet. 10:1995–2011. Taylor, T. J., McNamee, E. E., Day, C., and Knipe, D. M. (2003). Herpes simplex virus replication compartments can form by coalescence of smaller compartments. Virol. 309:232– 247. Terasaki, M., Jaffe, L. A., Hunnicutt, G. R., and Hammer, J. A., III (1996). Structural change of the endoplasmic reticulum during fertilization: Evidence for loss of membrane continuity using green fluorescent protein. Dev. Biol. 179:320–328. Terskikh, A., Fradkov, A., Ermakova, G., Zaraisky, A., Tan, P., Kajava, A. V., Zhao, X., Lukyanov, S., Matz, M., Kim, S., Weissman, I., and Siebert, P. (2000). “Fluorescent timer”: Protein that changes color with time. Science 290:1585–1588. Thomsen, P., Roepstorff, K., Stahlhut, M., and van Deurs, B. (2002). Caveolae are highly immobile plasma membrane microdomains, which are not involved in constitutive endocytic trafficking. Mol. Biol. Cell 13:238–250. Toomre, D., and Manstein, D. J. (2001). Lighting up the cell surface with evanescent wave microscopy. Trends Cell Biol. 11:298–303. Toomre, D., Keller, P., White, J., Olivo, J. C., and Simons, K. (1999). Dual-color visualization of trans-Golgi network to plasma membrane traffic along microtubules in living cells. J. Cell Sci. 112:21–33. Toomre, D., Steyer, J. A., Keller, P., Almers, W., and Simons, K. (2000). Fusion of constitutive membrane traffic with the cell surface observed by evanescent wave microscopy. J. Cell Biol. 149:33–40. Tsien, R. Y., Bacskai, B. J., and Adams, S. R. (1993). FRET for studying intracellular signalling. Trends Cell Biol. 3:242–245. Tvaruskó, W., Bentele, M., Misteli, T., Rudolf, R., Kaether, C., Spector, D. L., Gerdes, H. H., and Eils, R. (1999). Time-resolved analysis and visualization of dynamic processes in living cells. Proc. Natl. Acad. Sci. USA 96:7950–7955. Uster, P. S., and Pagano, R. E. (1986). Resonance energy transfer microscopy: observations of membrane-bound fluorescent probes in model membranes and in living cells. J. Cell Biol. 103:1221–1234. Varnai, P., Rother, K. I., and Balla, T. (1999). Phosphatidylinositol 3-kinase-dependent membrane association of the Bruton’s tyrosine kinase pleckstrin homology domain visualized in single living cells. J. Biol. Chem. 274:10983–10989. Vasudevan, C., Han, W., Tan, Y., Nie, Y., Li, D., Shome, K., Watkins, S. C., Levitan, E. S., and Romero, G. (1998). The distribution and translocation of the G protein ADP-ribosylation factor 1 in live cells is determined by its GTPase activity. J. Cell Sci. 111:1277–1285. Walter, J., Schermelleh, L., Cremer, M., Tashiro, S., and Cremer, T. (2003). Chromosome order in HeLa cells changes during mitosis and early G1, but is stably maintained during subsequent interphase stages. J. Cell Biol. 160:685–697.

335

336

THE USES OF GREEN FLUORESCENT PROTEIN IN MAMMALIAN CELLS

Wang, S., and Hazelrigg, T. (1994). Implications for bcd mRNA localization from spatial distribution of exu protein in Drosophila oogenesis. Nature 369:400–403. Ward, B. M., and Moss, B. (2001). Vaccinia virus intracellular movement is associated with microtubules and independent of actin tails. J. Virol. 75:11651–11663. Ward, T. H., Polishchuk, R. S., Caplan, S., Hirschberg, K., and Lippincott-Schwartz, J. (2001). Maintenance of Golgi structure and function depends on the integrity of ER export. J. Cell Biol. 155:557–570. Watanabe, N., and Mitchison, T. J. (2002). Single-molecule speckle analysis of actin filament turnover in lamellipodia. Science 295:1083–1086. Waterman-Storer, C. M., and Danuser, G. (2002). New directions for fluorescent speckle microscopy. Curr. Biol. 12:R633–R640. Waterman-Storer, C. M., Desai, A., Bulinski, J. C., and Salmon, E. D. (1998). Fluorescent speckle microscopy, a method to visualize the dynamics of protein assemblies in living cells. Curr. Biol. 8:1227–1230. Weiss, M., and Nilsson, T. (2003). A kinetic proof-reading mechanism for protein sorting. Traffic 4:65–73. Weiss, M., and Nilsson, T. (2004). In a mirror dimly: tracing the movements of molecules in living cells. Trends Cell Biol. 14:267–273. Weiss, M., Hashimoto, H., and Nilsson, T. (2003). Anomalous protein diffusion in living cells as seen by fluorescence correlation spectroscopy. Biophys. J. 84:4043–4052. White, J., Keller, P., and Stelzer, E. H. (2001). Spatial partitioning of secretory cargo from Golgi resident proteins in live cells. BMC Cell Biol. 2:19. White, J., and Stelzer, E. (1999). Photobleaching GFP reveals protein dynamics inside live cells. Trends Cell Biol. 9:61–65. Wiegand, U. K., Duncan, R. R., Greaves, J., Chow, R. H., Shipston, M. J., and Apps, D. K. (2003). Red, yellow, green go! A novel tool for microscopic segregation of secretory vesicle pools according to their age. Biochem. Soc. Trans. 31:851–856. Wiseman, P. W., Brown, C. M., Webb, D. J., Hebert, B., Johnson, N. L., Squier, J. A., Ellisman, M. H., and Horwitz, A. F. (2004). Spatial mapping of integrin interactions and dynamics during cell migration by Image Correlation Microscopy. J. Cell Sci. 117:5521–5534. Wouters, F. S., and Bastiaens, P. I. (1999). Fluorescence lifetime imaging of receptor tyrosine kinase activity in cells. Curr. Biol. 9:1127–1130. Wu, P., and Brand, L. (1994). Resonance energy transfer: Methods and applications. Anal. Biochem. 218:1–13. Wu, X., Rao, K., Bowers, M. B., Copeland, N. G., Jenkins, N. A., and Hammer, J. A., III (2001a). Rab27a enables myosin Va-dependent melanosome capture by recruiting the myosin to the organelle. J. Cell Sci. 114:1091–1100. Wu, X., Zhao, X., Baylor, L., Kaushal, S., Eisenberg, E., and Greene, L. E. (2001b). Clathrin exchange during clathrin-mediated endocytosis. J. Cell Biol. 155:291–300. Yang, T. T., Cheng, L., and Kain, S. R. (1996a). Optimized codon usage and chromophore mutations provide enhanced sensitivity with the green fluorescent protein. Nucleic Acids Res. 24:4592–4593. Yang, T. T., Kain, S. R., Kitts, P., Kondepudi, A., Yang, M. M., and Youvan, D. C. (1996b). Dual color microscopic imagery of cells expressing the green fluorescent protein and a red-shifted variant. Gene 173:19–23. Zaal, K. J. M., Smith, C. L., Polishchuk, R. S., Altan, N., Cole, N. B., Ellenberg, J., Hirschberg, K., Presley, J. F., Roberts, T. H., Siggia, E., Phair, R. D., and Lippincott-Schwartz, J. (1999). Golgi membranes are absorbed into and reemerge from the ER during mitosis. Cell 99:589–601. Zaccolo, M., Magalhães, P., and Pozzan, T. (2002). Compartmentalisation of cAMP and Ca2+ signals. Curr. Opin. Cell Biol. 14:160–166.

REFERENCES

Zacharias, D. A., Violin, J. D., Newton, A. C., and Tsien, R. Y. (2002). Partitioning of lipidmodified monomeric GFPs into membrane microdomains of live cells. Science 296:913–916. Zamir, E., Katz, M., Posen, Y., Erez, N., Yamada, K. M., Katz, B. Z., Lin, S., Lin, D. C., Bershadsky, A., Kam, Z., and Geiger, B. (2000). Dynamics and segregation of cell-matrix adhesions in cultured fibroblasts. Nat. Cell Biol. 2:191–196. Zhang, G., Gurtu, V., and Kain, S. R. (1996). An enhanced green fluorescent protein allows sensitive detection of gene transfer in mammalian cells. Biochem. Biophys. Res. Commun. 227:707–711. 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. Zhang, G.-J., Safran, M., Wei, W., Sorensen, E., Lassota, P., Zhelev, N., Neuberg, D. S., Shapiro, G., and Kaelin, W. G., Jr. (2004). Bioluminescent imaging of Cdk2 inhibition in vivo. Nat. Med. 10:643–648. Zolotukhin, S., Potter, M., Hauswirth, W. W., Guy, J., and Muzyczka, N. (1996). A “humanized” green fluorescent protein cDNA adapted for high-level expression in mammalian cells. J. Virol. 70:4646–4654.

337

15 PRACTICAL CONSIDERATIONS FOR USE OF REEF CORAL FLUORESCENT PROTEINS IN MAMMALIAN CELLS: APPLICATIONS IN FLUORESCENCE MICROSCOPY AND FLOW CYTOMETRY Yu Fang, Olivier Déry, Michael Haugwitz, and Pierre Turpin BD Biosciences Clontech, Palo Alto, CA

Steven R. Kain Agilent Technologies, Palo Alto, CA

The discovery of fluorescent proteins from nonbioluminescent reef corals [reef coral fluorescent proteins (RCFPs)] has greatly expanded the panel of emission wavelengths available for in vivo analyses of cellular events and has opened the door to many new multicolor applications (Matz et al., 1999). Indeed, the excitation maxima of the RCFPs range from 458 nm to 588 nm, and their emission maxima range from 489 nm to 618 nm. Whereas the color variants of Aequorea victoria green fluorescent protein GFP were generated by mutagenesis of a single parental gene, the RCFPs AmCyan1, ZsGreen1, ZsYellow1, DsRed2 (orange-red), AsRed2 (true red), and HcRed1 (far-red) are encoded by discrete genes and were isolated from distinct species (Table 15.1). The RCFP family members share at most 30% amino acid sequence identity with A. victoria GFP. However, the three-dimensional structure obtained from DsRed1 crystals show that at least one member of the RCFP family has a structure very similar to that of A. victoria GFP. These RCFP proteins differ not only in their excitation and emission spectra, but also in the time course of fluorescence development, relative brightness, and long-term expression. In this chapter, we describe the unique properties of each RCFP, their expression in mammalian cells, and their use and detection in multiplex applications.

Green Fluorescent Protein: Properties, Applications, and Protocols, Second Edition Edited by Martin Chalfie and Steven R. Kain Copyright © 2006 John Wiley & Sons, Inc.

339

340

PRACTICAL CONSIDERATIONS FOR USE OF REEF CORAL FLUORESCENT PROTEINS IN MAMMALIAN CELLS

TABLE 15.1. Nomenclature of Reef Coral Fluorescent Proteins Commercial Name (Mutants)

Mutants (AA Sequence)

Species

Original Name (as published)

References

AmCyan1 ZsGreen1 ZsYellow1 DsRed2 DsRed-Express AsRed2 HcRed1

N34S, K68M N66M M129V R2A, K5E, K9T R2A, K5E, N6D F4L, K12R, F35L A2S, T36A, L122H

Anemonia Zoanthus Zoanthus Discosoma Discosoma Anemonia Heterectis

amFP486 zFP506 zFP540 drFP583 DsRed-T1 asFP595 HcRed2A

Matz et al. (1999) Matz et al. (1999) Matz et al. (1999) Matz et al. (1999) Bevis and Glick (2002) Lukyanov et al. (2000) Gurskaya et al. (2001)

Figure 15.1. Spectral properties of reef coral fluorescent proteins AmCyan1, ZsGreen1, ZsYellow1, DsRed2, AsRed2, and HcRed1. Except for the lower level of residual green emission, the spectrum for DsRed-Express (not shown) closely resembled that of DsRed2.

15.1

PROPERTIES OF RCFPs

With the exception of ZsYellow1 and ZsGreen1, which were both derived from the same Zoanthus gene, the genes encoding the RCFPs were isolated from different species: AmCyan1, DsRed2 (orange-red), AsRed2 (true red), and HcRed1 (far-red) were isolated in Anemonia majano, Discosoma, Anemonia sulcata, and Heteractis crispa, respectively. As can be seen in Fig. 15.1, RCFPs cover a broad range of excitation and emission wavelengths: Excitation maxima range between 458 nm and 588 nm, and emission maxima range from 489 nm to 618 nm. Also, the excitation spectrum of each RCFP is broad, whereas their emission spectra are quite narrow. Such properties are quite valuable for multiplexing applications. An important characteristic of RCFPs is their tendency to form homo-oligomers. Crystallographic analysis of the original DsRed1 revealed a tight tetrameric structure (Gross et al., 2000; Wall et al., 2000; Yarbrough at al., 2001). These structure analyses were supported by the findings of analytical ultracentrifugation experiments (Baird at al., 2000). The subunits of this tetramer are similar in structure to A. victoria GFP (Ormö et al., 1996). They form a beta-sheet barrel-like structure, containing an internal alpha helix that bears the chromophore-forming amino acids. It is likely that other RCFPs such

PROPERTIES OF RCFPs

as AmCyan1, ZsGreen1, ZsYellow1, and AsRed2 also form tetramers as suggested by pseudo-native gel electrophoresis (Yanushevich et al., 2002). Due to their tetrameric structure, it is difficult to predict the function and/or localization of proteins fused to RCFPs. It has also been shown that RCFPs tend to form higher-order structures (aggregates) in mammalian cells. The extent of aggregation varies depending on the cell type in which the RCFPs are expressed. Replacing positively charged amino acids at the extreme N-terminus by neutral or negatively charged amino acids via site-directed mutagenesis has increased the overall solubility of RCFPs (Yanushevich et al. 2002; Bevis and Glick, 2002). The tetrameric structure of RCFPs (dimeric in the case of HcRed1) might restrict their use as fusion tags with other cellular proteins and peptides. However, a variety of fusion proteins have been successfully expressed in mammalian cells using these RCFPs. For example, all RCFPs have been successfully targeted to intracellular organelles or compartments such as the nucleus (nuclear targeting signal fused to the C-terminus of RCFPs) and the mitochondria (mitochondrial targeting signal fused to the N-terminus of RCFPs). Fusion proteins of RCFPs (not tested with DsRed-Express) and protein kinase C (PKC) alpha translocate from the cytosol to the plasma membrane upon induction with PMA, as is the behavior of endogenous PKC alpha. Proper localization of RCFP fusions suggests that in many cases the biological function of the fusion partner is not adversely impacted by linkage to RCFPs. The biophysical properties of each fluorescent protein are summarized in Table 15.2.

15.1.1

AmCyan1

AmCyan1 is a mutant of the original wild-type amFP486 from Anemonia majano (Matz et al., 1999). Two amino acid substitutions, N34S and K68M, were introduced to enhance brightness of the expressed protein. AmCyan1 has an excitation maximum at 458 nm and an emission maximum at 489 nm, and it has greater fluorescence intensity than ECFP (Fig. 15.2A). When expressed in mammalian cells, AmCyan1 can be detected 8–12 h after transfection. In comparison to ECFP, AmCyan1 exhibits a higher relative fluorescence intensity (quantum yield 0.75, extinction coefficient 39,000). AmCyan1 is also more resilient than ECFP to destructive photobleaching, which is the fast loss of fluorescence upon an extended excitation period. Because AmCyan1 resists photobleaching, the fluorescence intensity remains stable during analysis with a spectrofluorometer or fluorescence microscope. AmCyan1 is rather insoluble and tends to aggregate in mammalian cells when it is overexpressed. AmCyan1 is well suited for genetic reporter assays, such as transcriptional reporting. However, AmCyan1 has also been successfully localized to intracellular organelles and compartments including the nucleus and the mitochondria using appropriate targeting sequences.

15.1.2

ZsGreen1

ZsGreen1 is a mutant of wild-type zFP506 from a species of Zoanthus (Matz et al., 1999). A single amino acid substitution (N66M) has been made to enhance the brightness of the expressed protein. ZsGreen1 has an excitation maximum at 493 nm and emission maximum at 505 nm, and it is much brighter than EGFP (Fig. 15.2B). The time required to detect the fluorescence in mammalian cells is about 8–12 h post transfection. In comparison with EGFP, ZsGreen1 is brighter (quantum yield 0.91, extinction coefficiency 43,000). In fact, ZsGreen1 is the brightest fluorescent protein of the family of RCFPs. However, its solubility is poor, and the protein tends to aggregate in mammalian cells.

341

579

557

563 576 588

DsRed-Express

DsRed2 AsRed2 HcRed1

a

489 505 539

Reef Coral Fluorescent Proteins AmCyan1 458 ZsGreen1 493 ZsYellow1 529

24 8–12 16

8–12

8–12 8–12 8–12

8–12 8–12 8–12

Time to detection (hr)*

+++ ++ +

+++

+++ ++++ ++

+ +++ ++

Brightness relative to EGFP

Tetramer Tetramer Dimer

Tetramer

Tetramer Tetramer Tetramer

Monomer Monomer Monomer

Structure

As measured by FACS analysis using transiently transfected mammalian cells cultures.

582 592 618

476 510 529

Emission Max (nm)

Aequorea victoria GFP Variants ECFP 439 EGFP 484 EYFP 512

Protein

Excitation Max (nm)

TABLE 15.2. Comparison of BD Living ColorTM Fluorescent Proteinsa

+++ +++ +

+++

+++ ++++ +++

+ +++ +++

Utility as a reporter

++ + +++

++

+ + +

++++ ++++ ++++

Utility in fusions

Far-red fluorescence ideal for multiplexing, no detectable aggregation

Preferred DsRed for FACS due to diminished green emission, faster maturation Low aggregation

Photostable alternative to ECFP Bright green True yellow emission; ideal for multicolor applications

Green / yellow

Not as photostable as EGFP, EYFP

Comments

342 PRACTICAL CONSIDERATIONS FOR USE OF REEF CORAL FLUORESCENT PROTEINS IN MAMMALIAN CELLS

PROPERTIES OF RCFPs

Figure 15.2. Spectral comparison using fluorometer. (A) 1 mM of AmCyan1 and ECFP excited at 280 nm. (B) 1 mM of ZsGreen1 and EGFP excited at 488 nm. (C) 1 mM of ZsYellow1 and EYFP excited at 280 nm.

Therefore, ZsGreen1 is recommended for reporter assays. Additionally, ZsGreen was successfully targeted for degradation by the proteasome by fusion with PEST sequences from mouse ornithine decarboxylase. This construct is the basis of an assay to monitor proteasome activity in live cells (pZsProSensor, the plasmid encoding this proteasome sensor, is commercially available from BD Biosciences Clontech).

15.1.3

ZsYellow1

ZsYellow1 is a mutant of wild-type zFP540 from a species of Zoanthus (Matz et al., 1999). A single amino acid substitution (M129V) has been made to enhance the brightness of the expressed protein. ZsYellow1 has its excitation maximum at 529 nm and an emission maximum at 539 nm. ZsYellow1 is a true yellow protein. With an emission maximum at 539 nm, its spectrum is red-shifted in comparison to EYFP, an EGFP variant with an emission maximum at 529 nm (see Table 15.2). The distinctive spectrum of ZsYellow1 allows the separation of three colors—AmCyan1, ZsGreen1, and ZsYellow1—by single laser excitation (488 nm) of cells expressing each protein by flow cytometry. The time required to detect ZsYellow1 fluorescence in mammalian cells is about 8–12 h. ZsYellow1 is quite bright (quantum yield 0.65, extinction coefficiency 20,000), but the fluorescent intensity is a little lower in comparison with EYFP (Fig. 15.2C). ZsYellow1 is not very soluble and tends to aggregate in mammalian cells. ZsYellow1 is very suitable for transcription reporter assays.

15.1.4

Red Fluorescent Proteins

The family of RCFPs includes three red fluorescent proteins, with emission spectra ranging from 582 nm to 618 nm. They are ideal for multiplexing and FRET applications (Kohl et al., 2002; Erickson et al., 2003). The first red fluorescent protein, DsRed1, was identified in a species of Discosoma and emits light in the orange-red range with a maximum at 582 nm (Matz et al., 1999). The second red fluorescent protein, AsRed2, which emits light in the true red range with an emission maximum at 592 nm, was generated by mutating a nonfluorescent chromoprotein of Anemonia sulcata (Lukyanov et al., 2000). More recently, the gene coding for a far-red-shifted fluorescent protein, HcRed1, with unique

343

344

PRACTICAL CONSIDERATIONS FOR USE OF REEF CORAL FLUORESCENT PROTEINS IN MAMMALIAN CELLS

far-red fluorescence emission at 618 nm was generated by a combination of site-directed and random mutagenesis of the gene for a nonfluorescent protein isolated from Heteractis crispa (Gurskaya et al., 2001).

15.1.5

DsRed2/DsRed-Express

DsRed2 contains six amino acid substitutions when compared with DsRed1. The mutations at V105A, I161T, and S197A result in a more rapid appearance of red fluorescence compared to DsRed1. The cause for the faster chromophore formation may be that the mutations create a more flexible and less “crowded” space around the chromophore, allowing faster folding. The N-terminal mutations R2A, K5E, and K9T reduce the tendency of the protein to form aggregates in mammalian cells. The time required to detect of DsRed2 fluorescence in mammalian cells is about 24 h post transfection. DsRed2 has an excitation maximum at 563 nm and an emission maximum at 582 nm. Although it likely forms the same tetrameric structure as DsRed1 (Yarbrough et al., 2001), DsRed2 is less prone to forming large insoluble aggregates, which often develop in mammalian cells expressing DsRed1. DsRed2 is well-suited for both fusion proteins and transcription reporter assays. DsRed2 has been expressed in mammalian cells as an N-terminal fusion protein with Bid, a proapoptotic “BH3-only” member of the Bcl-2 family. The Bid-DsRed2 fusion protein was proteolytically processed and translocated from the cytosol to the mitochondria as is characteristic of its endogenous Bid counterpart. In addition, DsRed2 has been used in a variety of different mammalian cell applications (Iida et al., 2003; Lu et al., 2003; Mathieu and El-Battari, 2003; Zhe et al., 2003). DsRed-Express was developed by a combination of random and site-directed mutagenesis of DsRed1 (Bevis and Glick, 2002). It contains nine amino acid substitutions. The N-terminal mutations R2A, K5E, and N6D enhance the solubility of DsRed-Express. Interestingly, two of these same mutations were found to increase the solubility of DsRed2 using separate mutagenesis procedures in different laboratories. The increase in solubility could be a result of decreased positive charge at the N-terminus of DsRed-Express and DsRed2 in comparison to DsRed1. Four additional mutations—T21S, H41T, N42Q, and V44A—were found to be crucial for a faster formation of a functional chromophore. The red fluorescence of DsRed-Express can be detected in mammalian cells about 8–12 h after transfection compared with 48 h for DsRed1. Another limitation of the DsRed1 and DsRed2 is a substantial residual green emission due to a chromophore maturation intermediate. This residual emission interferes with double and triple labeling applications using green fluorescent protein, especially in fluorescence microscopy and flow cytometry. Two mutations in DsRed-Express, C117S and T217A, yield a profound reduction of this residual green emission peak. However, there is no direct evidence that the mutations listed above changed the tendency of DsRed-Express to form a tetramer. The spectral properties of DsRed-Express are similar to those of DsRed2, with an excitation maximum at 557 nm and an emission maximum at 579 nm. DsRed-Express is bright (quantum yield 0.90, extinction coefficient 19,000). However, if compared to DsRed2, DsRed-Express has a lower extinction coefficient (43,800 versus 30,100) and a reduced relative brightness (0.68 for DsRed2 versus 0.36 for DsRed-Express; Bevis and Glick, 2002). DsRed-Express can be used as a fusion tag, if the presumed tetrameric structure is not problematic. DsRedExpress has been expressed in mammalian cells as an N-terminal fusion protein with Bid, a proapoptotic “BH3-only” member of the Bcl-2 family. The Bid-DsRed-Express fusion protein was proteolytically processed and translocated from the cytosol to the mitochondria as is characteristic of its endogenous Bid counterpart.

PROPERTIES OF RCFPs

15.1.6

The “Fluorescent Timer” (E5)

A set of two mutations (V105A; S197T) of the original DsRed1 protein gave rise to a fluorescent protein with a very unique property. This protein, called fluorescent timer, has a fluorescent spectrum that changes over time from an initial bright green to red (Terskikh et al., 2000). Purified recombinant “fluorescent timer” protein achieves maximum green fluorescence about 4 h after purification. Subsequent to this initial 4-h period, green fluorescence declines; simultaneously, red fluorescence starts to appear. This “color switch” of the fluorescent timer can be used to monitor activation as well as downregulation of a specific promoter of interest using the color switch as the timer for this event. The ratio of green to red fluorescence can be used to determine the time in the past at which a promoter was switched on and when it was switched off (Terskikh et al., 2000).

15.1.7

DsRed Monomer

Many efforts have been made to weaken the tetrameric structure of DsRed in order to obtain a monomeric red fluorescent protein. However, this goal has been challenging due to the very tight structure of DsRed. Recently Tsien et al. were successful in generating a monomeric DsRed by mutating DsRed T1 (DsRed-Express: Bevis and Glick, 2002). The monomeric mutant, mRFP1, contains 33 mutations in comparison to the original DsRed1. Three mutations are located at the A/B interface, and 10 mutations are located at the A/C interface of the original tetramer. These mutations seem to be essential to separate the DsRed subunits of the tetramer into monomers. The emission maximum of monomeric DsRed in comparison to the tetrameric DsRed2 is shifted from 582 nm to 607 nm. However, the extinction coefficient, quantum yield, and photostability is lower than in DsRed2 (Campbell et al., 2002).

15.1.8

AsRed2

AsRed2 is a mutant of AsRed1 from Anemonia sulcata (Lukyanov et al., 2000) and has been engineered for stronger fluorescence intensity. It contains eight amino acid substitutions: F4L, K12R, F35L, T68A, F84L, A143S, K163E, and M202L. The time to detect red fluorescence in mammalian cells is about 8–12 h. AsRed2 has an excitation maximum at 576 nm and an emission maximum at 592 nm. Thus, it is slightly red-shifted compared to DsRed2 and DsRed-Express. AsRed2 can be used as a fusion tag, although it likely forms the tetrameric structure proposed for all RCFPs. AsRed2 is well-suited for transcription reporter assays.

15.1.9

HcRed1

HcRed1 was generated by random and site-directed mutagenesis of a gene coding for a nonfluorescent chromoprotein in the reef coral Heteractis crispa (Gurskaya et al., 2001). Early rounds of random mutagenesis were used to produce variants with extreme far-red fluorescence and rapid maturation kinetics. After isolation of the brightest variant, investigators used site-directed mutagenesis to optimize the solubility of the protein. The final variant, HcRed1 (HcRed-2A; Gurskaya et al., 2001), was selected not just because of its bright far-red fluorescence, but also because this mutant forms a dimer rather than a tetramer due to the mutation L126H located at the interface of the originally tetrameric protein. In comparison to wild type HcRed, HcRed1 contains seven additional mutations:

345

346

PRACTICAL CONSIDERATIONS FOR USE OF REEF CORAL FLUORESCENT PROTEINS IN MAMMALIAN CELLS

A2S, T36A, L122H, C143S, R168H, L173H, and P201L. HcRed1 does not aggregate in mammalian cells, and it elutes in gel filtration column chromatography with the predicted size of a dimeric protein. The good solubility of HcRed1 makes it more suitable for fusion proteins and mammalian expression compared to other RCFPs with a predicted tetrameric structure. Numerous fusions proteins have been successfully fused to HcRed1 (Hori et al., 2003; Kogata et al., 2003; Mulholland et al., 2003; Rintoul et al., 2003; Valentijn et al., 2003; Wagner et al., 2003). The time required to detect red fluorescence of HcRed1 is about 16 h post transfection in mammalian cells. HcRed1 has an excitation maximum at 588 nm and an emission maximum at 618 nm. In comparison to DsRed-Express and AsRed2, HcRed1 is not as bright (quantum yield 0.03, extinction coefficient 20,000). However, if HcRed1 is expressed in mammalian cells, it can be easily detected by fluorescence microscopy as well as flow cytometry. It may be that the in vitro conditions used to determine the quantum yield and extinction coefficient of recombinant HcRed1 does not reflect the fluorescence intensity of HcRed1 that can be achieved when expressed in mammalian cells. The fluorescence of HcRed1 is easily detected 16 h after transfection. Due to the far-red emission spectral properties, it is very easy to separate HcRed1 from other fluorescent proteins in multiplexing applications.

15.2

EXPRESSION AND DETECTION OF RCFPs

RCFPs have been expressed in variety of hosts. We have expressed RCFPs in mammalian cell lines such as HEK 293, HeLa, 3T3, Jurkat, and HT1080 cells. From unpublished observations, RCFPs have been successfully expressed in a variety of organisms such as fungi, plants, yeast, E. coli, C. elegans, Drosophila, Xenopus, zebrafish, and mouse. Here we focus on the practical considerations for the use of RCFPs in mammalian cells. RCFPs can be expressed in all mammalian systems as long as a functional promoter is used to initiate transcription. A promoter can be constitutively active, meaning that it initiates a persisting transcription of the RCFP gene in the plasmid downstream of the promoter. The immediate early cytomegalovirus promoter (CMV IE) has been used frequently to drive the constitutive expression of exogenous genes in mammalian cells. However, it is also possible to express the RCFPs under the control of an inducible promoter. This option allows the researcher to monitor the activation or inactivation of a promoter via the appearance or disappearance of the respective fluorescent signal. The signal-to-noise ratio in this type of application can be increased dramatically by using fusion proteins consisting of RCFPs and protein degradation motifs such as that found in mouse ornithine decarboxylase. Indeed, such “destabilized” RCFPs are constitutively degraded by the proteasome complex, and their accumulation and degradation within the cells are better correlated with promoter functions than in the case of nondestabilized reporters. Several different promoterless vectors, lacking a functional promoter sequence for mammalian cells and encoding these destabilized RCFPs, are available from BD Biosciences Clontech (pZsGreen1-DR, pDsRed-Express-DR, and pHcRed1-DR) and can be used to monitor the activity of promoter/enhancer combinations of interest. When expressing RCFPs and other fluorescent proteins as a tag fused to a protein of interest, the behavior of the resulting fusion protein cannot always be predicted. Changes in the proper function and/or localization of the protein/peptide of interest upon fusion to RCFPs may occur in some cases. To minimize this risk, it is often helpful to consider alternative orientations of the protein/peptide of interest with respect to the RCFP. It is possible to fuse the protein/peptide of interest either to the N-terminus or to the C-terminus of

EXPRESSION AND DETECTION OF RCFPs

the respective RCFP. This option is of particular importance if the protein/peptide of interest has a functional domain at the extreme N- or C-terminus. The two fusion proteins that contain either a free N-terminus or a free C-terminus of the protein of interest can then be tested separately. When considering the expression level of an RCFP in mammalian cells, the reef coral species of origin must be considered. The optimal DNA codon usage between reef corals and mammalian cells is quite different. Therefore it may be necessary to optimize the codon usage of the RCFP for a specific expression system. Currently, all RCFPs are available in human codon optimized forms that have been tested in a variety of mammalian expression systems. However, the human codon usage might be suboptimal for specific expression systems, especially nonmammalian expression systems. In this case a possible series of silent base pair changes should be considered in order to increase the expression level of the RCFP protein. However, these steps may only be necessary in exceptional cases. The expression level in eukaryotic cells can also be increased by incorporating a Kozak consensus sequence (CGCCACCATGG) including the ATG start codon at the 5¢end of the RCFP gene. Commercially available mammalian expression vectors (from BD Biosciences Clontech) contain the human codon optimized versions of the respective RCFP genes. A high expression level using those mammalian expression vectors is ensured by the use of a 5¢ Kozak sequence. Multiple cloning sites, located at either the 5¢- or 3¢ end of the respective RCFP gene, allow the expression of either N- or C-terminal fusion proteins. It has often been a concern of many researchers that fluorescent proteins used as a research tool might have a toxic effect on cells, therefore altering their normal function, growth, or differentiation state. In order to evaluate cytotoxicity due to overexpression in mammalian cells, HEK 293 cells were transfected with cytoplasmic targeted RCFPs, and cells expressing high levels of the fluorescent protein were sorted by flow cyotmetry on a BD FACSVantage SE cell sorter. Fluorescent cells were collected and were further studied for their ability to grow normally when returned to culture, as well as to maintain their level of fluorescence. In all studies, fluorescence was evaluated by flow cytometry. EGFP was used as a control and a basis for comparison between the different tests. Cells expressing EGFP, AmCyan1, ZsGreen1, ZsYellow1, DsRed2, and AsRed2 were analyzed on a BD FACSCalibur flow cytometry system using a 488-nm laser. HcRed1 required a 568.2-nm laser line for proper excitation; cells expressing this far-red shifted variant were analyzed on a BD FACSVantage SE cell sorter. For a period of 14 –16 weeks, we monitored both the percentage of cells that remained fluorescent as well as the mean fluorescence of these positive cells. Our results show that for most of the fluorescent proteins, the percentage of positive cells was stable between 85% and 100%. Only DsRed2 and ZsYellow1 exhibited a decline to 80% and 60%, respectively (Fig. 15.3). These data suggest that for most of the RCFPs, almost all cells were able to maintain high levels of expression. The difference of intensities between the different cell populations is mostly due to the fact that all fluorescent proteins are not detected with the same efficiency on a specific flow cytometer (e.g., AmCyan1 is very bright but poorly detected on a BD FACSCalibur flow cytometry system).

15.2.1

Detection of RCFP by Fluorescence Microcopy

Protocol 1: Preparation of Cells Expressing Fluorescent Proteins for Fluorescence Microscopy. Prior to transfecting cells with the specific plasmid, 1 ¥ 105

347

348

PRACTICAL CONSIDERATIONS FOR USE OF REEF CORAL FLUORESCENT PROTEINS IN MAMMALIAN CELLS

Figure 15.3. Cells tolerate long-term expression of RCFPs. HEK 293 cells were stably transfected with N1 vectors encoding the indicated protein, and they were analyzed over time using BD FACSCaliburTM and BD FACSVantageTM. (A) The percentage of positive cells remaining in the population. (B) The non-normalized medium fluorescence intensity of the positive population. AmCyan1, EGFP, ZsGreen1, and ZsYellow1 were excited with a 488-nm line; DsRed2, DsRedExpress, AsRed2, and HcRed1 with a 568.2-nm laser line. AmCyan1, EGFP, ZsGreen1, and ZsYellow1 were detected with a 530/30 filter; DsRed2, DsRed-Express, and AsRed2 with a 595/25 filter; HcRed with a 630/20 filter.

cells are plated onto glass coverslips in a 6-well plate. Twenty-four hours after plating, cells are transfected using 0.75 mg of plasmid DNA using either lipid-based transfection agents like FuGene6 (Roche) or BD CLONfectin transfection reagent (BD Biosciences Clontech), or by using the calcium phosphate method (BD CalPhos Mammalian Transfection Kit, BD Biosciences Clontech), following the standard protocol. Twenty-four hours post transfection, the coverslips are rinsed twice with prewarmed (37°C) PBS with Ca2+/Mg2+. After these washes, the cells are fixed using 4% paraformaldehyde in PBS for 15 min at RT. After fixation, the coverslips are rinsed three times using PBS before being mounted onto glass slides (e.g., Molecular Probe mounting medium). The fixed cells are then stored at 4°C for 24 h before they are analyzed on a fluorescence microscope (e.g., Zeiss Axioskop) using different filter sets (e.g., Chroma RCFP filter sets).

15.2.2

Single-Color Analysis—Using 1 RCFP

In order to obtain good-quality imaging in microscopy using any fluorophore, it is important to use optimized filter sets. The sensitivity and specificity of the signal will be affected by this choice. Although it is possible to achieve good results using standard filter sets (such as FITC filter sets to detect ZsGreen1 and rhodamine or propidium iodide filter sets to detect DsRed), it is best to use filter sets that have been developed specifically for each fluorescent protein. Optimized filter sets for detecting all RCFPs including AmCyan1, ZsGreen1, ZsYellow1, DsRed2, AsRed2, and HcRed1 have been developed by Chroma Technology Corporation (Table 15.3). Detailed information on detecting RCFPs can be found on Chroma’s website (www.chroma.com). These filter sets have been developed to maximize the detection by adjusting the bandpass filters across the peak on the excitation and emission spectra of a given fluorescent protein. In addition, the width of the bandpass as well as the shift between the two bandpasses of the two filters has been carefully chosen

349

EXPRESSION AND DETECTION OF RCFPs

TABLE 15.3. Filter Sets for One-Color Analysis of Reef Coral Fluorescent Proteins Protein AmCyan1 ZsGreen1 ZsYellow1 DsRed2 AsRed2 HcRed1

Excitation Filter

Dichroic Mirror

Emission Filter

D440/40x HQ470/35 HQ550/40 D540/40x D540/40x HQ575/50x

470dcxr Q490lp 530dclp 570dclp 570dclp 610dclp

D500/40m HQ520/40 HQ550/40m D600/50m HQ620/60m HQ640/50m

Note: DsRed and AsRed filter sets are very close and interchangeable. All filter sets are available from Chroma Technology Corp. (www.chroma.com).

to minimize background and maximize fluorescent signal. Depending on the application, the RCFPs are not all equivalent; for fusion proteins, it is always best to try fluorescent proteins that are monomeric such as EGFP or dimers like HcRed1. Indeed, oligomerization of the fluorescent protein can result in the loss of the function of the protein studied, or it can lead to improper localization inside the cell. By contrast, for application where the subcellular localization of the fluorescent protein is not as important as a variation of fluorescence intensity, it is recommended to use the brightest and most photostable fluorescent proteins such as AmCyan1, ZsGreen1, or DsRed2. As an example, when under the control of a promoter specific of a signal transduction pathway, ZsGreen1 works well as an expression fluorescent reporter.

15.2.3

Multicolor Analysis—Using 2 or More RCFPs

Researchers are increasingly interested in multicolor analyses. With the introduction of the red fluorescent proteins, DsRed2, AsRed2, and HcRed, combined with the development of optimized filter sets by Chroma, it is now possible to separate as many as three fluorescent reporters (cyan, yellow, and red) by fluorescence microscopy. Here we offer several recommendations for two-and three-color analyses (Table 15.4). As a general rule, it is very difficult to use two fluorescent proteins simultaneously if their spectral characteristics are too similar. This will result in bleeding of one fluorescent protein signal in the filter set to the other, rendering signal separation problematic. For two-color analyses, AmCyan1 can be used in combination with ZsYellow1 or any of the red fluorescent proteins. AmCyan1 is less prone to photobleaching than ECFP and is potentially a better partner to yellow and red fluorescent proteins when long exposure times are required due to multicolor imaging. ZsYellow1 can be separated from AmCyan1 as well as the far-red-shifted HcRed1, so it is possible to use it in combination with either of these two for dual detection. ZsGreen1 can be used with all the red fluorescent proteins but not with AmCyan1 and ZsYellow1, whose spectral characteristics are too similar. The combination of ZsGreen1 and DsRed2 or DsRed-Express is especially convenient since it can be visualized with standard FITC and rhodamine filter sets of any conventional microscope. As shown in Fig. 15.4, ZsGreen1 targeted to the mitochondria (ZsGreen-Mito) and a DsRed2-bid fusion were coexpressed in HeLa cells, Bid is a proapoptotic “BH3only” member of the Bcl-2 family. In nonapoptotic cells, uncleaved Bid is localized in the cytosol. However, upon induction of apoptosis with an appropriate stimulus, Bid can be cleaved by the enzyme Caspase 8. Truncated tBid translocates to the mitochondria and may induce the release of Cytochrome C from the mitochondria into the cytosol. The

350

PRACTICAL CONSIDERATIONS FOR USE OF REEF CORAL FLUORESCENT PROTEINS IN MAMMALIAN CELLS

TABLE 15.4. Recommended Combination for Multicolor Microscopy First Color

Second Color

Third Color

AmCyan1

ZsYellow1 DsRed2 DsRed-Express HcRed1 DsRed2 DsRed-Express HcRed1 AmCyan1 HcRed1 AmCyan1 ZsGreen1 AmCyan1 ZsGreen1 ZsYellow1

HcRed1

ZsGreen1

ZsYellow1 DsRed2 DsRed-Express HcRed1

A

ZsYellow1

HcRed1 AmCyan1

ZsYellow1 AmCyan1

B

Control

Staurosporin

Figure 15.4. Dual-color analysis for monitoring Bid activation with DsRed2. HeLa cells were transiently cotransfected with plasmids encoding the fusion protein Bid-DsRed2 and a mitochondriatargeted ZsGreen1 (ZsGreen1-Mito). (A) Before induction of apoptosis, Bid-DsRed2 is localized in the cytosol and ZsGreen1-Mito labels the mitochondria. (B) After induction of apoptosis with 1 mM staurosporine for 3 h, the relocalization of Bid-DsRed2 to mitochondria as revealed by the colocalization with the mitochondria marker ZsGreen1-Mito. Images were taken with a 100¥ objective using Chroma filter sets hq460/40x, 490dclp, and hq515/30m for ZsGreen1 and using hq545/50x, 580dcxr, and hq630/60m for DsRed2. See color insert.

translocation of cleaved tBid from the cytosol to the mitochondria upon apoptotic stimulus is a very important hallmark for the role of mitochondria in the apoptotic destruction of cells. The expression of a chimeric fusion of Bid and a fluorescent protein, like DsRed2, allows one to monitor the translocation event by observing the redistribution of the fluorescence signals. The colocalization of green fluorescent mitochondria (ZsGreen-Mito) and red fluorescent tBid can therefore only be observed upon induction of apoptosis.

EXPRESSION AND DETECTION OF RCFPs

Figure 15.5. Detection of three fluorescent proteins by fluorescent microscopy. HeLa cells were separately transfected with plasmids pAmCyan1-N1, pZsYellow1-N1, and pHcRed1-N1, mixed, and observed by microscopy using Chroma Technology Corp. filter sets d440/40x, 470dcxr, and d500/40m for AmCyan1, using hq500/40, 530dclp, and hq550/40m for ZsYellow1, and using hq575/50x, 610dclp, and hq640/50m for HcRed1. See color insert.

Combining ZsGreen1 and DsRed2 for dual color labeling and using specific filter sets for their detection usually provides high-quality images, but in some cases the fluorescent signal of one or both fluorescent proteins is very strong. This might result in some bleedthrough of the green fluorescence in the red filter set or vice versa. To overcome this problem, we recommend visualizing ZsGreen1 with the AmCyan1 filter set and DsRed2 with the HcRed1 filter set. This configuration conveys adequate fluorescent signal from ZsGreen1 and DsRed2 to obtain good-quality images and efficiently reduces the bleeding of nonspecific fluorescence. For three-color analyses, there is only one recommended combination for microscopy: AmCyan1, ZsYellow1, and HcRed1. As shown in Fig. 15.5, HeLa cells were separately transfected with the plasmids pAmCyan1-N1, pZsYellow1-N1 and pHcRed1-N1. After mixing the transfected cells, the fluorescence from the three cell subpopulations were correctly visualized using Chroma Technology Corp. filter sets (Table 15.3). In multiplex experiments, the choice of filter sets is critical to optimize separation of the fluorescence signals. Aside from the fact that these filter sets were designed for a high signal-to-background ratio, they were also chosen to minimize the bleeding of fluorescent signal of one fluorescent protein into the other filter set. In the case of triple labeling for example, the images obtained for ZsYellow1 must not contain much signal from AmCyan1 or HcRed1. Otherwise it is impossible to distinguish any of the three fluorescent signals.

15.2.4

Detection of RCFP by Flow Cytometry

Protocol 2: Preparation of Cells Expressing Fluorescent Proteins for Flow Cytometry 1. BD FACSTM Analysis of Unfixed Cells. Routinely, 3 ¥ 105 cells are plated into a six-well plate. Twenty-four hours after plating, cells are transfected with the plasmid DNA of interest using 3 mg of DNA. Cells are transfected using either the calcium phosphate method or a lipid-based transfection reagent like FuGene6 (Roche) or BD CLONfectin transfection reagent (BD Biosciences Clontech). Twenty-four hours post transfection, cells are collected. Adherent cells are removed from the tissue culture plate by rinsing the plate with PBS w/o Ca2+/Mg2+

351

352

PRACTICAL CONSIDERATIONS FOR USE OF REEF CORAL FLUORESCENT PROTEINS IN MAMMALIAN CELLS

before adding trypsin, followed by a 1- to 5-min incubation at 37°C. The cells are than rinsed from the plate using DMEM medium. The collected cells are pelleted and washed once using ice-cold PBS. After washing, the cells are resuspended in 1 ml of ice-cold PBS or sheat fluid. This cell suspension is then analyzed using either a BD FACSCaliburTM flow cytometry system or a BD FACSVantageTM SE cell sorter. 2. BD FACSTM Analysis of Fixed Cells. Routinely, 3 ¥ 105 cells are plated into a 6well plate. twenty-four hours after plating, cells are transfected with the plasmid DNA of interest using 3 mg of DNA. Cells are transfected using either the calcium phosphate method or a lipid-based transfection reagent like FuGene6 (Roche) or BD CLONfectin transfection reagent (BD Biosciences Clontech). Twenty-four hours post transfection, cells are collected. Adherent cells are removed from the tissue culture plate by rinsing the plate with PBS w/o Ca2+/Mg2+ before adding trypsin, followed by a 1- to 5-min incubation at 37°C. The cells are then rinsed from the plate using DMEM medium. The collected cells are pelleted and washed once using ice-cold PBS. The cells are than resuspended in 1 ml of 4% paraformaldehyde in PBS. This cell suspension is then incubated at RT for 30 min either by mixing the cell suspension slowly every 5 min or by using a slow shaker. After fixation, cells are pelleted by centrifugation and washed 3 times with 1 ml of ice-cold PBS. The cell pellet of the last washing step is then resuspended in 1 ml of PBS or sheat fluid and analyzed using a BD FACSCalibur flow cytometry system or a BD FACSVantage SE cell sorter. Protocol 3: Generation of Stable Cell Populations and Clones Expressing RCFPs. In order to establish cell populations that express RCFPs, 3 ¥ 105 HEK 293 cells are plated in 6-well plates and transfected with 2 mg of the desired plasmid (e.g., pAmCyan1-N1). After 48 h, transfected cells are selected using media containing 0.5 mg/ml G418 and grown for another 2 weeks. At this stage, all the cells are able to grow in selective media, and a majority of the cells express the fluorescent protein. However, the expression level varies widely among the stable cells, resulting in a wide range of fluorescence intensity in the mixed population. If cells with brighter fluorescence signals are necessary for an experiment, the mixed population can be enriched using flow cytometry. For example, we used this procedure to isolate cells for studying potential long-term cytotoxicity effects of RCFP expression. The mixed population was run on a BD FACSVantage SE cell sorter, and the brightest 30% of cells were separated to achieve a more homogeneous sampling. Then, each week the new population was harvested and analyzed for the percentage of remaining fluorescent cells and the average fluorescence intensity. The BD FACSVantage SE cell sorter was also used to establish clones of HEK 293 cells expressing RCFPs. As described above, the brightest 30% of the cells were isolated and sorted into 96-well plates with 1 cell per well. After 2–3 weeks, individual clones were screened by fluorescence microscopy and the 12 brightest clones were expanded into 24well plates, 6-well plates, and 100-mm dishes. Each clone was analyzed for homogeneity and fluorescence intensity. Among all the RCFPS, HcRed1 is the only fluorescent protein that has offered challenges in establishing stable cell populations and clonal lines. We noticed that the expression of HcRed1 tends to reduce the growth rate of the cells. To overcome this problem,

353

EXPRESSION AND DETECTION OF RCFPs

we transfected cells with a lower quantity of plasmid and selected the stable population with a lower concentration of antibiotic. Interestingly, though the cells grew slowly, the level of HcRed1 expression and the percentage of positive cells remained stationary.

15.2.5

Single-Color Analysis by Flow Cytometry—Using 1 RCFP

When only one RCFP or other fluorescent protein is expressed in cells to be analyzed by flow cytometry, the two most important considerations are (a) the use of an appropriate laser line for excitation of the protein and (b) the channel selected to collect the fluorescent signal. We describe here our experience using each RCFP expressed in HEK 293 cells. AmCyan1. With an excitation maxima of 458 nm and emission maxima of 489 nm, AmCyan1 can be detected on flow cytometers equipped with laser lines such as 407 nm (BD FACSAria cell sorter) or 458 nm (Table 15.5). We have also successfully detected AmCyan1 using a BD FACSCalibur instrument. We used a clone that exhibited high levels of fluorescence that was detectable using the 488-nm laser line and the FL1 channel (530/30 bandpass filter).

TABLE 15.5. Detection of RCFPs on BD’s Flow Cytometers Excitation Max

Excitation Required

AmCyan1

458

ZsGree ZsYellow1

493 529

DsRed2

563

DsRedExpress

557

AsRed2

576

HcRed1

588

405 407 413 458 458 488 488 488 514 531 488 514 531 568 488 514 531 568 488 514 531 568 568 633 635

Protein

Suitable Laser VioFlame Point source Krypton Argon Spectrum Argon Argon Argon Argon Spectrum Argon Argon Spectrum Spectrum Argon Argon Spectrum Spectrum Argon Argon Spectrum Spectrum Spectrum HeNe Red diode

Emission Max 489

505 539

582

Suitable Emission Filters 485/22 or 530/30

530/30 519/20 or 520/40 530/30 or 550/30 550/30 or 585/42 585/42 555/20, 580/30, 585/42, or 610/20

579

610/20 or 630/30 555/20, 580/30, 585/42, or 610/20

592

610/20 or 630/30 580/30, 585/42, 610/20, or 630/30

618

610/20 or 630/30 610/20 or 630/30 660/20

354

PRACTICAL CONSIDERATIONS FOR USE OF REEF CORAL FLUORESCENT PROTEINS IN MAMMALIAN CELLS

ZsGreen1. Like EGFP, ZsGreen1 is ideal for flow cytometry because it is efficiently activated with a 488-nm laser found on most machines and is easily detected on the FL1 channel of a BD FACSCalibur instrument. ZsGreen1 is the brightest fluorescent protein of all the RCFPs, and we needed to establish and select clones expressing medium levels of ZsGreen1 (Fig. 15.6B); otherwise the entire cell population would be out of the FL1 channel scale. Its brightness makes it the RCFP of choice when the sensitivity of detection might be an issue. For example, we have successfully monitored subtle variations in ZsGreen1 fluorescence when using the protein as a gene expression reporter. ZsYellow1. ZsYellow1 can be activated using a 488-nm laser as well as the 531-nm line of a spectrum laser (Table 15.5). With an emission maximum of 539 nm, it can be detected using either the FL1 (530/30) or FL2 (585/42) channels of a BD FACSCalibur instrument. DsRed2, DsRed-Express and AsRed2. Typically, DsRed2, DsRed-Express, and AsRed2 are detected using a 488-nm laser and the FL2 (485/42) channel of a BD FACSCalibur instrument. All the clones we established were easily detectable, though these settings are not optimal. DsRed-Express matures faster than DsRed2 (Fig. 15.6C) and is therefore a better fit for reporter studies using flow cytometry. In cases where the level of expression of DsRed is very low and difficult to detect on a BD FACSCalibur

Figure 15.6. Two-color analysis of ZsGreen1 and DsRed-Express by flow cytometry. HEK 293 cells were transiently transfected with either pZsGreen1 or pDsRed-Express, mixed, and then analyzed by flow cytometry on a BD FACSCaliburTM. (A) Mock. (B) ZsGreen1. (C) DsRed-Express. (D) Mixed with ZsGreen1 and DsRed-Express. (E) Cotransfected with ZsGreen1 and DsRed-Express. (F) Mixed and cotransfected with ZsGreen1 and DsRed-Express.

EXPRESSION AND DETECTION OF RCFPs

instrument, we use the 568-nm line of an argon/krypton laser on the BD FACSVantage cell sorter in order to excite DsRed with higher efficiency. HcRed1. HcRed1 is not detectable on a BD FACSCalibur instrument because it is not excited efficiently with a 488-nm laser. We use the 568-nm laser of a BD FACSVantage SE cell sorter to excite HcRed1 and the FL5 channel equipped with a 640 long-pass filter for detection. Cells expressing HcRed1 are easily detected with these settings, even when the level of expression and the overall fluorescence of HcRed1 are low.

15.2.6

Two Color Analysis by Flow Cytometry—Using 2 RCFPs

Flow cytometry offers the possibility to detect several fluorescent proteins simultaneously and is an ideal complement to microscopy where quantitative information is required without the need to visualize the spatial distribution of fluoresence. ZsGreen1 and DsRedExpress form the most appropriate pair of fluorescent proteins because both can be excited with a 488-nm laser and are easily detected on a BD FACSCalibur instrument using the FL1 and FL2 channels (Fig. 15.6D). Although some instrument compensation might be required, it is generally easy to separate both signals so that cells cotransfected with ZsGreen1 and DsRed-Express can be visualized and potentially sorted using a BD FACSVantage SE cell sorter (Figs. 15.6E and 15.6F). DsRed-Express, which exhibits lower green emission than DsRed2, is the best partner for ZsGreen1 because it requires minimum compensation. Using machines with other laser lines and detection channels, it is possible to separate almost every fluorescent protein in performing two-color analysis. AmCyan1 can be combined with ZsYellow1 or any red fluorescent protein. Furthermore, using the 568-nm line of a spectrum laser on a BD FACSVantage SE cell sorter, combined with a 595/25 filter on FL4 and a 640 long-pass filter on FL5 channels, we have successfully separated DsRed-Express and HcRed1 signals (Fig. 15.7A) in such a manner that cells cotransfected with both were also visualized (Fig. 15.7B).

Figure 15.7. Two- and three-color analysis by flow cytometry. (A) HEK 293 cells were transiently transfected separately with either pDsRed2-N1 or pHcRed1-N1, mixed, and then analyzed by FACSVantage. (B) Cells were also cotransfected with both vectors, mixed with the independently transfected cells, and analyzed. Cells were excited using a BD FACSVantageTM SE cell sorter using the 568.2-nm line of a coherent krypton/argon laser. DsRed2 was detected with a Chroma Technology Corp. 595/25 bandpass filter; HcRed1, with a 640 long-pass filter. (C) HEK 293 cells were stably transfected with either pZsGreen-N1, pZsYellow-N1, or pAsRed2-N1, mixed, and then analyzed on a BD FACSCaliburTM using the 488-nm laser line.

355

PRACTICAL CONSIDERATIONS FOR USE OF REEF CORAL FLUORESCENT PROTEINS IN MAMMALIAN CELLS

15.2.7

Three- and Four-Color Analysis by Flow Cytometry

Compared to microscopy, one advantage of flow cytometry is that it measures the fluorescence intensity of cells. It is therefore possible to visualize the fluorescence of several fluorescent proteins using two-dimensional scatter plots, even if the spectral properties of the proteins are similar. For example, it is difficult to distinguish AmCyan1, ZsYellow1, and DsRed2 by microscopy because of the degree of spectral overlap. By contrast, we were able to separate signals from these proteins using a BD FACSCalibur instrument (Fig. 15.7C). AmCyan1 is mostly detected in the FL1 channel, and cells expressing it appear close to the FL1 axis in a FL1 versus FL2 scatter plot. Similarly, cells expressing DsRed2 appear close to the FL2 axis. Cells expressing ZsYellow1, which is equally detected by the FL1 and FL2 channels, are visualized in the diagonal of the same plot. It is therefore possible to establish three distinct regions and measure the individual fluorescence intensity of a mixed population of cells each expressing a different RCFP. Using a BD FACSVantage SE cell sorter, it has been possible to analyze cells expressing ECFP, EGFP, EYFP, and DsRed1 (Hawley et al., 2001). We achieved similar results with a BD FACSAria cell sorter using the three available laser lines (407, 488, and 633 nm) to activate AmCyan1, ZsYellow1, DsRed-Express, and HcRed1. Figures 15.8A and 15.8B show two scatter plots of the same mixed population of cells expressing the four fluorescent proteins. While DsRed-Express and ZsYellow1 can be separated as previously described on the BD FACSCalibur instrument, it was also possible to plot the HcRed1 channel versus the AmCyan1 channel to also distinguish these two cell populations. Overall, the growing number of flow cytometry instruments available on the market combined with the entire panel of reef coral fluorescent proteins provide an excellent array of tools for multiplexing applications. This technology has now reached a point where the automation of large-scale experiments is practical.

15.2.8

Other Methods of Detection

Since the fluorescent proteins have been used in research, microscopy and flow cytometry have been the two predominant detection methods. It is also possible to detect fluoB

A 5

10

AmCyan1

10

DsRed-Express

356

104

103

5

104

103

102

102

102

103

104

ZsYellow1

105

102

103

104

105

HcRed1

Figure 15.8. Four-color separation of RCFP-expressing HEK 293 cells using flow cytometry. A mixed population of cells stably expressing either DsRed2, ZsYellow1, HcRed1, or AmCyan1 was separated by flow cytometry with a BD FACSAriaTM cell sorter using three separate laser lines: 407 nm to excite AmCyan1; 488 nm to excite DsRed2 and ZsYellow1; and 633 nm to excite HcRed1.

REFERENCES

rescent proteins using spectrophotometry, and a number of investigators have developed applications where fluorescence intensity is measured in cells in a 96-well plate reader. Recently, a novel type of instrument has been applied to the detection of RCFPs expressed in cells. Acumen Bioscience has developed a laser-scanning device, the Acumen ExplorerTM, which is able to measure the fluorescence of cells attached to 96-well plates. With a resolution of less than one micron, the Explorer is able to visualize the fluorescence profile of a cell expressing a fluorescent protein. With the convenience of a plate reader, sensitivity close to that of a flow cytometer, and potential subcellular fluorescence localization capabilities, the Acumen Explorer is becoming a very popular tool for the development of cell-based assays in drug discovery. In summary, convenient coexcitation of green, yellow, and red fluorescent proteins with the common 488-nm laser line allows for easy detection of these proteins on nearly all flow cytometers. With the addition of more laser lines and appropriate detection filters, at least four cell populations, each harboring a different fluorescent protein, can be distinguished by the proteins’ unique emission spectra. These characteristics provide great flexibility for multiplexing in various applications. Fluorescent proteins are a valuable resource when combined with flow cytometry for both preparative and analytical applications. Applications range from the very simple enrichment of cell populations after transient transfection and cell clone isolation to the more complex isolation of cells from transgenic animals or analyses in gene therapy research.

ACKNOWLEDGMENTS We are grateful to Eric Machleder of BD Biosciences Clontech for help in preparing this chapter.

REFERENCES 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 97:11984–11989. Bevis, B. J., and Glick, B. S. (2002). Rapidly maturing variants of the Discosoma red fluorescent protein (DsRed). Nat. Biotechnol. 20:83–87. Campbell, R. E., Tour O., Palmer A. E., Steinbach P. A., Baird G. S., Zacharias D. A., and Tsien R. Y. (2002). A monomeric red fluorescent protein. Proc. Natl. Acad. Sci. USA 99:7877–7882. Erickson, M. G., Moon, D. L., and Yue, D. T. (2003). DsRed as a potential FRET partner with CFP and GFP. Biophys J. 85:599–611. 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. Natl. Acad. Sci. USA 97:11990–11995. Gurskaya, N. G., Fradkov, A. F., Terskikh, A., Matz, M. V., Labas, Y. A., Martynov, V. I., Yanushevich, Y. G., Lukyanov, K. A., and Lukyanov, S. A. (2001). GFP-like chromoproteins as a source of far-red fluorescent proteins. FEBS Lett. 507:16–20. Hawley, T. S., Telford, W. G., Ramezani, A., and Hawley, R. G. (2001). Four-color flow cytometric detection of retrovirally expressed red, yellow, green, and cyan fluorescent proteins. Biotechniques 30:1028–1034.

357

358

PRACTICAL CONSIDERATIONS FOR USE OF REEF CORAL FLUORESCENT PROTEINS IN MAMMALIAN CELLS

Hori, T., Haraguchi, T., Hiraoka, Y., Kimura, H., and Fukagawa, T. (2003). Dynamic behavior of Nuf2-Hec1 complex that localizes to the centrosome and centromere and is essential for mitotic progression in vertebrate cells. J. Cell Sci. 116:3347–3362. Iida, R., Yasuda, T., Tsubota, E., Takatsuka, H., Masuyama, M., Matsuki, T., and Kishi, K. (2003). M-LP, Mpv17-like protein, has a peroxisomal membrane targeting signal comprising a transmembrane domain and a positively charged loop and up-regulates expression of the manganese superoxide dismutase gene. J. Biol. Chem. 278:6301–6306. Kogata, N., Masuda M., Kamioka Y., Yamagishi A., Endo A., Okada M., and Mochizuki N. (2003). Identification of Fer tyrosine kinase localized on microtubules as a platelet endothelial cell adhesion molecule-1 phosphorylating kinase in vascular endothelial cells. Mol. Biol. Cell 14:3553–3564. Kohl, T., Heinze, K. G., Kuhlemann, R., Koltermann, A., and Schwille, P. (2002). A protease assay for two-photon crosscorrelation and FRET analysis based solely on fluorescent proteins. Proc. Natl. Acad. Sci. USA 19:12161–12166. Lu, J. Y., Chen, H. C., Chu, R. Y., Lin, T. C., Hsu, P. I., Huang, M. S., Tseng, C. J., and Hsiao, M. (2003). Establishment of red fluorescent protein-tagged HeLa tumor metastasis models: determination of DsRed2 insertion effects and comparison of metastatic patterns after subcutaneous, intraperitoneal, or intravenous injection. Clin. Exp. Metastasis 20:121–133. Lukyanov K. A., Fradkov, A. F., Gurskaya, N. G., Matz, M. V., Labas Y. A., Savitsky A. P., Markelov, M. L., Zaraisky A. G., Zhao X., Fang, Y., Tan, W., and Lukyanov, S. A. (2000). Natural animal coloration can be determined by a nonfluorescent green fluorescent protein homolog. J. Biol. Chem. 275:26879–25882. Mathieu, S., and El-Battari, A. (2003). Monitoring E-selectin-mediated adhesion using green and red fluorescent proteins. J. Immunol. Methods 272:81–92 Matz, M. V., Fradkov, A. F., Labas, Y. A., Savitsky, A. P., Zaraisky, A. G., Markelov, M. L., and Lukyanov, S. A. (1999). Fluorescent proteins from nonbioluminescent Anthozoa species. Nat. Biotechnol. 17:969–973. Mulholland, D. J., Read, J. T., Rennie, P. S, Cox, M. E., and Nelson, C. C. (2003). Functional localization and competition between the androgen receptor and T-cell factor for nuclear beta-catenin: A means for inhibition of the Tcf signaling axis. Oncogene 22:5602–5613. Ormö, 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. Rintoul, G. L., Filiano, A. J., Brocard, J. B., Kress, G. J., and Reynolds Y. J. (2003). Glutamate decreases mitochondrial size and movement in primary forebrain neurons. J. Neurosci. 23:7881–7888. Terskikh, A., Fradkov, A., Ermakova, G., Zaraisky, A., Tan, P., Kajava, A. V., Zhao, X., Lukyanov, S., Matz, M., Kim, S., Weissman, I., and Siebert, P. (2000). “Fluorescent timer”: Protein that changes color with time. Science 290:1585–1588. Valentijn, A. J., Metcalfe, A. D., Kott, J., Streuli, C. H., and Gilmore, A. P. (2003). Spatial and temporal changes in Bax subcellular localization during anoikis. J. Cell Biol. 162:599– 612. Wagner, L. E., Li, W., and Yule, D. I. (2003). Phosphorylation of type-1 inositol 1,4,5-trisphosphate receptors by cyclic nucleotide-dependent protein kinases: A mutational analysis of the functionally important sites in the S2+ and S2- splice variants. J. Biol. Chem., in press. 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. Yanushevich, Y. G., Staroverov, D. B., Savitsky, A. P., Fradkov, A. F., Gurskaya, N. G., Bulina, M. E., Lukyanov, K. A., and Lukyanov, S. A. (2002). A strategy for the generation of nonaggregating mutants of Anthozoa fluorescent proteins. FEBS Lett. 511:11–14.

REFERENCES

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-Å resolution. Proc. Natl. Acad. Sci. USA 98:462–467. Zhe, X., Yang, Y., Jakkaraju, S., and Schuger, L. (2003). Tissue inhibitor of metalloproteinase-3 downregulation in lymphangioleiomyomatosis: Potential consequence of abnormal serum response factor expression. Am. J. Respir. Cell. Mol. Biol. 28:504–511.

359

16 PHARMACEUTICAL APPLICATIONS OF GFP AND RCFP Nicola Bevan Stephen Rees Screening and Compound Profiling, GlaxoSmithKline, Stevenage, Herts, United Kingdom

16.1

INTRODUCTION

The color variants of the Aequorea victorea green fluorescent protein (GFP) and the reef coral fluorescent proteins (RCFP) are in widespread use within the pharmaceutical industry. In this chapter we describe a number of applications of the use of these proteins in the early phase of drug discovery. Many novel drug screening assays have been developed through the use of fluorescent proteins. These assays are being applied both (a) for the initial discovery of molecules with activity at a target protein in a process termed highthroughput screening (HTS) and (b) in the more detailed characterization of compound efficacy in a number of assays collectively termed high-content, or high-information, screening assays (referred to as HCS assays in this chapter). Fluorescent protein technology has also been applied within the target validation phase of drug discovery. Target validation describes a collection of techniques used to generate information on the likely involvement of a new target in disease. The objective of such studies is to generate a hypothesis that small-molecule intervention at that target may have efficacy within that disease state. Fluorescent protein technology has been applied for both in vitro and in vivo target validation. We will present some examples of the use of GFPs and RCFPs in model organisms such as C. elegans, Drosophila, and transgenic mice.

Green Fluorescent Protein: Properties, Applications, and Protocols, Second Edition Edited by Martin Chalfie and Steven R. Kain Copyright © 2006 John Wiley & Sons, Inc.

361

362

PHARMACEUTICAL APPLICATIONS OF GFP AND RCFP

16.2

HIGH-CONTENT AND HIGH-THROUGHPUT SCREENING

Perhaps the widest application of fluorescent protein technology within the pharmaceutical industry has been for the identification and characterisation of hit and lead molecules in processes termed HTS (Walters and Namchuk, 2003) and HCS (Kapur, 2002). HTS describes the process whereby chemical libraries, of many hundred thousand or indeed million molecules, are screened against drug targets to identify molecules with modulatory activity at the target protein that may form the basis of a lead optimisation drug discovery program. HTS assays are typically run in 384- or 1536-well microtiter plates in assay volumes of between 5 and 50 ml at compound throughputs of between 10,000 and 100,000 test molecules/day, using complex laboratory automation and detection systems. In contrast to HTS, which demonstrates whether or not a compound has activity at a target protein, HCS provides additional information on how the compound affects cell physiology. HCS assays enable multiple measurements of cellular phenotype as a consequence of compound activity. Assay end-points include the intracellular location and translocation of a fluorescent protein fusion protein, a change in cell shape, or neurite outgrowth, through the use of measurements taken on a cell-by-cell basis (Liptrot, 2001; Kapur, 2002). Several fluorescent protein translocation assays have been developed (Table 16.1). Fluorescent protein translocation assays require the generation of a fusion protein between the protein under study and a fluorescent protein. Plasmid vectors containing fluorescent proteins are available from suppliers such as Clontech (Palo Alto, CA) (www.clontech.com). For membrane proteins, GFP is typically fused to the carboxylterminus of the protein such that GFP is not required to pass through the membrane; for soluble proteins there are examples of the generation of GFP fusions at either the amino or carboxyl termini of the protein. In either case, care has to be taken to ensure that the generation of the fusion protein does not affect the function of the protein under study. It is a remarkable observation that the fusion of GFP or RCFP to other proteins has little detectable effect on the function of the protein partner. The detection of a change in the cellular localization of a fluorescent protein fusion protein in an HCS assay is determined using a plate-based fluorescence imaging system. Such instruments enable the generation of quantitative data from both individual cells and cell populations. Specific informatic tools have been developed in order to generate numerical data from fluorescence images captured by these analysis systems. These algorithms typically rely upon the identification of a cell by staining the nucleus with a nuclear specific fluorescent dye. The analysis algorithm captures the distribution of the fluorescent protein within the cell in relation to the nucleus to generate quantitative data on the intensity and location of the fluorescent protein. Such data are typically generated from pre- and posttreatment of cells with the test compound in order to determine the effect of that compound on the physiology under study. HCS assays may provide new insights into compound mode of action. It is hoped that the increasing application of HCS screening assays will enable a more thorough characterization of the activity of lead molecules prior to more expensive and complex animal studies, which ultimately will lead to the development of more efficacious and selective drugs (Kapur, 2002). In the last five years a number of fluorescence imaging systems have become available including the Cellomics ArrayscanTM, the GE Healthcare InCell Analyser 3000TM, or the Evotec OperaTM. These systems allow the detection of the subcellular localization of a GFP fusion protein within cells plated into 96-, 384-, or 1536-well microtiter plate formats. In the fastest of these systems, data can be collected from a 1536-well plate in

363

FLUORESCENT PROTEIN DRUG SCREENING ASSAYS

TABLE 16.1. Examples of Fluorescent Protein Translocation Assays Assay

Application

Translocation

Reference

Nuclear translocation

Agonist/antagonist identification at nuclear receptors

Cytoplasm to nucleus

GPCR/GFP internalization

Agonist/antagonist identification at GPCR

Membrane to cytoplasm

b-Arrestin recruitment

Agonist/antagonist identification at GPCR

Cytoplasm to membrane

GLUT4/GFP

Reporter of insulin activation Agonist/antagonist identification at GPCR

Cytoplasm to membrane Cytoplasm to membrane

Giuliano et al. (1997) Koster and Hauser (1999) Zhu et al. (1998) Htun et al. (1996) Conway et al. (1999) Conway et al. (2001) Tarasova et al. (1997) Drmota et al. (1998) Kallal et al. (1998) Barak et al. (1997b) Awaji et al. (1998) Slice et al. (1998) Barlic et al. (1999) Lamb et al. (2001) Schlador and Nathanson (1997) Zhang et al. (1999) Luttrell et al. (1999) Ferguson et al. (1998) Vrecl et al. (1998) Groarke et al. (1999) Evans et al. (2001) Barak et al. (1997a) Matharu et al. (2001) Mundell et al. (2001) Richardson et al. (2003) Patki et al. (2001)

PKC/GFP or PLC/GFP recruitment

Richardson et al. (2003) Feng et al. (1998) Almholt et al. (1999) Wang et al. (2001)

approximately 45 minutes. Further information about the characteristics of these detection instruments, the informatic tools developed to complement these detection systems, and the type of assays enabled by these systems can be obtained from the corresponding manufacturers (Table 16.2).

16.3

FLUORESCENT PROTEIN DRUG SCREENING ASSAYS

A number of drug screening assays based on the use of fluorescent protein technology have been described, and the use of such assays for HTS and HCS is increasing. Reporter gene assays, fluorescence resonance energy transfer (FRET) assays, bioluminescence resonance energy transfer (BRET) assays, and fluorescent protein degradation assays have been applied to HTS. In contrast, a range of fluorescent protein cellular redistribution assays have been developed for HCS.

364

PHARMACEUTICAL APPLICATIONS OF GFP AND RCFP

TABLE 16.2. Suppliers of Instrumentation and Assay Reagents for High Content Screening Company

Website

Summary

Accumen GE Healthcare BioImage Cellomics Evotec Imaging Research Molecular Devices Molecular Devices PerkinElmer

www.acumenbioscience.com www.amershambiosciences.com www.bioimage.com www.cellomics.com www.evotecoai.com www.imagingresearch.com www.moleculardevices.com www.moleculardevices.com www.perkinelmer.com

Instrumentation Reagents, instrumentation Reagents Instrumentation Instrumentation Instrumentation Instrumentation TransfluorTM Reagents, instrumentation

Fluorescent protein screening assays require the generation of a mammalian cell line transfected to express the target being screened together with the fluorescent reporter molecule. In many assays this involves the generation of fusion proteins between the protein being studied and the fluorescent protein. To generate a compound screening assay, mammalian cells in culture are either stably or transiently transfected with the fusion protein using either a plasmid or viral expression vector. For HTS assays or HCS assays in which many thousands of compounds are tested across many weeks or months, it is common practice to generate stable cell lines. HCS assays in which small numbers of compounds are tested are often generated using transient expression systems. A variety of methods have been applied to transfect immortalized mammalian cell lines in culture with a fluorescent protein expression plasmid. However, the efficiency of plasmid transfection in many immortalized cell lines, and most primary cell lines, is poor. Viral expression vectors, including those derived from adenovirus or retrovirus, are more complex to generate, and their use often requires stringent Biosafety conditions. However, such vectors are capable of transducing a wide variety of immortalised and primary cell lines. The use of viral vectors in HCS assays allows the study of compound activity in primary cell types which are often available in small numbers but which can generate a more physiologically relevant understanding of compound activity. HTS assays are typically performed on live cells. In contrast, HCS screening assays may be performed on fixed or live cells. The use of live cells allows the detection of translocation in real time and hence enables the study of the kinetics of protein translocation; however, this requires that the plate be imaged for the entire duration of the translocation event. The use of fixed cells allows many assay plates to be prepared and fixed at a predetermined point following compound stimulation. Images can then be collected for analysis. The use of fixed cells increases the daily capacity of the detection instrument as imaging times are generally shorter. In the following sections we review examples of such assays and their application to particular classes of drug targets to provide the reader with some insight into the application of fluorescent proteins for compound screening within the pharmaceutical industry.

16.3.1

Nuclear Translocation Assays

A number of HCS screening assays have been established that rely upon the detection of the translocation from the cytosol to the nucleus of a fusion protein between a transcription factor and a fluorescent protein. Following stable or transient expression in mam-

FLUORESCENT PROTEIN DRUG SCREENING ASSAYS

malian cells, translocation of the fusion protein is detected using a plate-based imaging system. A number of assays have been developed, including assays of translocation of the glucocorticoid receptor (Htun et al., 1996), NFkB (Schmid et al., 2000), ERK MAP kinase (Horgan and Stork, 2003), NFAT (Scott et al., 2001), and STAT (Koster and Hauser, 1999). Such assays have been configured to screen for direct modulators of transcription factor function or as reporters of compound activity at cell surface receptors known to regulate the signal transduction cascade under study. Within our laboratory we have developed a number of transcription factor and nuclear receptor translocation assays. In collaboration with GE Healthcare Biosciences (Cardiff, UK), we developed an assay for direct modulators of the glucocorticoid receptor (GR), through the generation of a fusion protein between human GR and GFP. This protein was expressed in HEK 293 cells, and a 96-well microtiter plate assay for GR modulators was established using the GE Healthcare InCell Anaylser 3000TM for signal detection. In unstimulated cells this fusion protein is found in the cytoplasm (Fig. 16.1). Following treatment with a steroid agonist such as dexamethasone, the fusion protein is seen to translocate to the nucleus within 30 minutes of drug treatment (Fig. 16.1). Using this assay, we have profiled a series of GR agonists and antagonists and demonstrated that these compounds are capable of regulating GR activity with the expected concentration dependence. This assay has the potential to be used in HTS to identify compounds capable of causing GR translocation. In combination with other nuclear receptor translocation assays, this assay has since been applied to profile the activity of GR agonists, for their activity against a panel of nuclear receptors and other transcription factors.

Figure 16.1. Confocal visualization of nuclear translocation of a glucocorticod receptor/GFP fusion protein. The glucocorticod receptor/GFP fusion protein was stably expressed in HEK 293 cells. Assays were performed on cells plated in a 96-well microtiter plate and images taken using the GE Healthcare InCell Analyser 3000TM. Basal fluorescence (A) and fluorescence distribution following 30 min incubation with dexamethasone (100 nM) (B) are shown. (Data provided by GE Healthcare, Cardiff, UK.)

365

366

PHARMACEUTICAL APPLICATIONS OF GFP AND RCFP

16.3.2

Cell Adhesion Assays

In a novel application of fluorescent protein technology, Mathieu and El-Battari (2003) have developed an E-selectin adhesion assay using both GFP and DsRed2 RCFP. Adhesion assays are in widespread use within the pharmaceutical industry for the identification of inhibitors of integrin and selectin mediated cell adhesion (Loster and Horstkorte, 2000). Such assays have relied upon the use of fluorescent dyes, such as BCECF, to label cells expressing the integrin or selectin, which are then examined for their ability to adhere to a matrix. The inhibition of cell adhesion is detected by a decrease in the adhesion of fluorescently labeled cells to the matrix. To avoid the use of fluorescent dyes, Mathieu and El-Bathtari describe the generation of a stable CHO cell line expressing both E-selectin and GFP or DsRed2 RCFP. The ability of such cells to adhere to a variety of E-selectin ligands was examined. This assay provides an alternative to the use of fluorescent dyes which avoids the requirement of loading cells with dye prior to assay and the dye leaching and photobleaching issues associated with fluorescent dye technology.

16.3.3

Reporter Gene Assays

The application of reporter gene assays for compound screening has been extensively reviewed elsewhere (Hill et al., 2001; Rees et al., 1999). A reporter gene is a sequence of DNA whose easily measured product is synthesized in response to the activation of a specific signal transduction cascade. The DNA sequence consists of a promoter element (where transcription factors bind to control transcription), a reporter gene, and a transcription stop signal. The choice of promoter element dictates the sensitivity and specificity of the reporter. The reporter gene should offer a unique property to the cell system being studied, or at least be easily distinguishable from other cell products, have a short half-life to minimize basal accumulation of the reporter product, and be easily measurable in simple, cheap assays. Commonly used reporter genes include the enzymes firefly luciferase, secreted placental alkaline phosphatase and b-lactamase (Rees et al., 1999). These assays have been applied throughout academia and the biotechnology and pharmaceutical industries for compound screening in 384 and 1536 well format for many target classes including G-protein coupled receptors (GPCRs), cytokine receptors and growth factor receptors (Terstappen et al., 2000; Subbaramaiah et al., 2001). GFP offers an attractive alternative to the enzymatic reporter genes as GFP assays do not require cell lysis and reagent addition prior to signal detection. In contrast, these assays rely upon the detection of the accumulation of GFP in living cells. This offers a simple assay protocol, does not require the purchase of often expensive detection reagents, and allows the detection of reporter gene activity in real time in live cells. The use of GFP as an inducible reporter gene has been limited due to its poor brightness and its extremely long half-life (>36 h), which results in substantial basal accumulation of GFP and hence a low signal window in the assay. In recent years a number of GFP mutants have been described that address these issues. To facilitate expression in mammalian cells, GFP vectors have been developed in which the open reading frame has been optimized for preferred human codon usage. In parallel with codon optimization, specific point mutations have been introduced within the GFP chromphore to generate proteins that show a 35-fold increase in fluorescence compared to the native protein. These proteins have been designated enhanced-GFP (E-GFP) and are available from Clontech (Palo Alto, CA). Furthermore, to address the issue of GFP accumulation from an un-induced reporter gene, Li et al. (1998) have developed a series of destabilised GFPs with a half-life of 2 h. This protein,

FLUORESCENT PROTEIN DRUG SCREENING ASSAYS

designated d2GFP, contains the PEST domain from mouse ornithine decarboxylase, which targets GFP for rapid degradation, fused to the C-terminus of GFP. Several groups have now reported the use of E-GFP or d2GFP in reporter gene assays. As an example of the application of E-GFP in a reporter gene assay, Nagy et al. (2002) described the development of a screen for Ah receptor agonists. The Ah receptor is a ligand-dependent transcription factor that mediates the biological and toxic effects of polycyclic and halogenated aromatic hydrocarbons such as dioxin. A HepG2 cell line containing an integrated Ah receptor responsive E-GFP reporter gene was used to screen chemical libraries to identify novel activators of this receptor.

16.3.4 Fluorescence Resonance Energy Transfer (FRET) and Bioluminescence Resonance Energy Transfer (BRET) Assays FRET and BRET describe the process of nonradiative energy transfer between a fluorescent donor protein (FRET) or a bioluminescent donor enzyme (BRET) and a fluorescent acceptor protein (Fig. 16.2; Boute et al., 2002). In a FRET assay the fluorescent donor protein is usually the enhanced blue or cyan fluorescent protein (EBFP/ECFP) and the fluorescent acceptor protein, one of the derivatives of the Green/Yellow fluorescent protein class (EGFP or EYFP). The bioluminescent donor enzyme in a BRET assay is the coelenterate luciferase cloned from Renilla reniformis. This luciferase catalyzes the oxidative degradation of coelenterazine to generate blue light with a lmax of 460 nM; hence it is able to act as an energy donor to one of the derivatives of the green/yellow fluorescent protein class (EGFP or EYFP) (Fig. 16.2). FRET or BRET occurs when the donor and acceptor

Figure 16.2. Principles of bioluminescence resonance energy transfer (BRET) (A) and fluorescence resonance energy transfer (FRET) (B) assays. For BRET fusion, proteins are generated between the proteins under study (A, B) and Renilla luciferase (Rluc) and a fluorescent protein (YFP). In the presence of coelenterazine, light is generated by Rluc, which, if the luciferase is in close proximity to YFP, will excite the fluorescent protein to generate light at the emission maxima of this protein. In FRET, fusion proteins are generated between the proteins under study (A, B) and donor [cyan fluorescent protein (CFP)] and acceptor [yellow fluorescent protein (YFP)] fluorescent proteins. Excitation of CFP results in emission at 475 nM which, when in close proximity to the YFP, will excite this protein to generate light at the emission maxima of YFP (525 nM).

367

368

PHARMACEUTICAL APPLICATIONS OF GFP AND RCFP

proteins are brought into close proximity. This can be achieved through the construction of a biosensor containing both donor and acceptor proteins within the same polypeptide, or following the fusion of the donor and acceptor proteins to two proteins known to interact. FRET and BRET have been used to establish a number of assays as described in the following sections. In each case, compound activity is determined as either an increase or a decrease in FRET/BRET, according to whether the compound promotes an association or dissociation of the FRET/BRET partners. 16.3.4.1 FRET Protease Assays. Fluorescent protein technology has been applied to develop FRET screening assays for protease inhibitors. Such assays rely upon the generation of a fusion protein between a cyan/blue fluorescent protein (ECFP) and a green/yellow fluorescent protein (EYFP) separated by a specific protease cleavage site. In this fusion protein, excitation of ECFP at 433 nM results in the generation of fluorescence emission from EYFP at an emission maxima of 527 nM due to FRET between the two closely associated fluorescent proteins. Cleavage of the fusion protein results in a loss of FRET because the CFP and YFP molecules are no longer in close proximity. This results in an increase in ECFP fluorescence emission at 475 nM and a decrease in EYFP emission at 527 nM as FRET is lost (Fig. 16.2). Thus FRET can be quantified according to a ratio of light emission at these two wavelengths. Instruments designed for the detection of FRET are capable of simultaneous excitation and signal detection at two emission wavelengths from the sample under study. Protease biosensors have been isolated and purified to generate in vitro assays for protease inhibitors as alternatives to the generation of peptide substrates labeled with small fluorescent probes. Protease biosensors have also been expressed in mammalian cell lines to generate whole-cell-based assays for protease inhibitors (Berdichevsky et al., 2003; Jones et al., 2000; Kohl et al., 2002). Caspase activity resulting in the proteolytic cleavage of target proteins is an integral step in the pathway leading to apoptotic cell death (Jones et al., 2000). Many pharmaceutical companies are attempting to identify caspase inhibitors for a variety of neurological disorders. To address this, FRET-based HTS and HCS assays for caspase inhibitors have been developed and applied in intact mammalian cells (Tawa et al., 2001; Jones et al., 2000; Mahajan et al., 1999). These assays rely upon the generation of FRET substrates containing ECFP and EGFP/EYFP linked by a specific caspase cleavage site. Such assays have been developed in 96- and 384-well assay plates. Caspase inhibitors are identified because they prevent degradation of the FRET substrate and hence prevent a loss of FRET. Caspase biosensors have been used in recombinant cell lines and also introduced into primary neurones to permit the study of caspase inhibitors in a more relevant disease cell type. 16.3.4.2 FRET Assays of Protein–Protein Interaction. There are many reports of the application of FRET to demonstrate the interaction of two proteins [for reviews see Sekar and Periasamy (2003) and Eidne et al. (2002)]. In our laboratory we have applied FRET technology to demonstrate the interaction of the two subunits of the GABA-B receptor. The GABA-B receptor is a heterodimer of two GPCRs, namely, the GABA-BR1 receptor and the GABA-BR2 receptor. The generation of a functional receptor at the cell surface requires the expression of both these receptors (White et al., 2002). To confirm that the receptor exists as a heterodimer at the cell surface we generated fusion proteins between the GABA-BR2 receptor and the cyan RCFP and the GABA-BR1 receptor and the yellow RCFP (Fig. 16.3). When expressed alone, the GABA-BR1 receptor does not reach the cell surface. When coexpressed with the GABA-BR2 receptor, both receptors

FLUORESCENT PROTEIN DRUG SCREENING ASSAYS

Figure 16.3. Confocal visualization of GABA-B receptor heterodimerization. HEK 293T cells were transfected with the fusion proteins, GABA-BR2/cyanRCFP and GABA-BR1/yellowRCFP. From left to right in the figure, the images show cellular expression of GABA-BR2/cyanRCFP (excitation at 433 nM, emission at 475 nM), GABA-BR1/yellowRCFP (excitation at 488 nM, emission at 525 nM; the overlay of the first two images demonstrate that both proteins are expressed at the same site), and the FRET signal (excitation at 433 nM; emission at 525 nM). The FRET event demonstrates that the proteins are in close proximity. See color insert.

can be visualized at the cell surface. That they exist in close proximity can be inferred from the observation of FRET between the two fusion proteins (Fig. 16.3). While many groups have applied FRET to identify and confirm protein–protein interactions, the application of FRET assays for compound screening has been limited. This is due to the relative weakness of the FRET signal in comparison to that obtained with BRET. However, such assays are being used within the pharmaceutical industry to identify and characterize novel protein–protein interactions during the target validation phase of drug discovery. 16.3.4.3 BRET Assays of Protein–Protein Interaction. G-protein-coupled receptors (GPCRs) are a family of approximately 750 membrane-spanning receptors that are known to have a fundamental role in physiology and pathophysiology. Members of this receptor family are responsible for the detection of visual and olfactory stimuli, and they play a pivotal role in cell signaling processes such as neurotransmission, chemotaxis, and inflammation. GPCRs are potential targets for therapeutic intervention in many diseases. Of the approximately 500 marketed drugs, around 30% are modulators of GPCR function having efficacy in diseases such as pain, psychiatric disorders, hypertension, asthma, cardiovascular disease, and peptic ulcers [Wise et al. (2002) and references therein]. The development of GPCR screening assays to enable the identification of novel GPCR drugs is a major challenge for the pharmaceutical and biotechnology industries. Agonist binding to a GPCR results in the activation of intracellular signal transduction cascades. To attenuate receptor signaling, serine and threonine residues within the Cterminal tail of many receptors become multiply phosphorylated by members of the GPCR kinases (GRK) family of protein kinases (Zhang et al., 1997). The phosphorylated Cterminal tail of the receptor is then able to bind members of the arrestin family. Arrestin binding promotes receptor internalization into the endosome compartment to attenuate receptor signaling. Within the acidic environment of the endosome, the ligand dissociates from the GPCR, the receptor C-terminal tail is dephosphorylated, and the receptor is recycled to the cell membrane (summarized in Fig. 16.6).

369

370

PHARMACEUTICAL APPLICATIONS OF GFP AND RCFP

BRET assays for compound screening at GPCRs have been described in which compound activity is determined according to the degree of interaction between a Renilla luciferase tagged GPCR and a YFP tagged b-arrestin (Bertrand et al., 2002). Assays have been established to identify both agonist and antagonist ligands at the receptor. Agonist ligands cause an increase in BRET following association between the receptor and the barrestin, wheras antagonists can be identified through the inhibition of agonist mediated BRET. BRET assays have also been established to study GPCR homo- and heterodimerization (Mercier et al., 2002; Jensen et al., 2002) and to monitor insulin receptor activation (Issad et al., 2002). A BRET assay for the interaction of the insulin receptor with protein tyrosine phosphatase 1B has been established by generating fusion proteins between the two interacting partners (Boute et al., 2003). The application of BRET to establish screening assays for regulators of such protein/protein interactions has been reviewed by Boute et al. (2002). Three combinations of donor and acceptor protein have been used in BRET assays (Table 16.3). In each case the donor protein is Renilla luciferase; however, different coelenterazine substrates and different acceptor proteins have been used in the different versions of BRET. In BRET1 the A. victoria E-YFP was used as the acceptor protein in combination with native coelenterazine as the Renilla luciferase substrate. BRETZS-Yellow again uses native coelenterazine as the Renilla luciferase substrate and the RCFP Zoanthus sp. YFP as the acceptor protein. Experiments in our laboratory have demonstrated that the BRET ratio change in equivalent experiments is greater with BRETZS-Yellow than with BRET1 due to the higher quantum yield of the Zoanthus sp. yellow RCFP and the red-shifted emission wavelength of this fluorescent protein. However, in both assays the emission spectra of native coelenterazine overlaps the emission spectra of the fluorescent protein, resulting in a low signal-to-noise ratio in the assay. To increase the Stokes shift between the emission maxima of native coelenterazine and the available fluorescent proteins, Perkin Elmer Biosciences (Monteal, Canada) have developed a novel Renilla luciferase substrate, DeepBlueC coelenterazine, and a novel YFP mutant with an excitation maxima at 400 nM (Bertrand et al., 2002). This version of BRET, designated BRET2, allows a wider separation of the emission maxima of the luciferase and fluorescent protein, thus leading to a large signal-to-noise ratio in the assay. However, this coelenterazine is substantially less bright, leading to a weak signal in the assay. Detection instruments, such as the Perkin Elmer Biosciences Fusion, are now available that enable the detection of BRET in 96- and 384-well plate formats. In collaboration with Perkin Elmer Biosciences (Montreal, Canada), we have developed a BRET assay to detect modulators of the interaction of the human glucocorticoid

TABLE 16.3. Fluorescence Characteristics of Donor and Acceptor Proteins Used in Bioluminescence Resonance Energy Transfer (BRET) Substrate

Substrate lmax Emission

BRET1

WT coelenterazine

475 nM

BRET2

DeepBlueC coelenterazine

400 nM

BRETZS-Yellow

WT coelenterazine

475 nM

Acceptor E-YFP (BD Biosciences) GFP-Topaz (Perkin Elmer) ZsYellow-NFP (BD Biosciences)

Acceptor lmax Emission 527 nM 508 nM 540 nM

FLUORESCENT PROTEIN DRUG SCREENING ASSAYS

receptor (GR) and the transcription factor NFkB. Because it is difficult to predict how the generation of a fusion protein will affect the biology of the interaction under study, Renilla luciferase and GFP-topaz were fused in-frame to both the amino and carboxy termini of GR and NFkB. A panel of transient transfection assays were performed to determine the vector combination that generated the greatest increase in BRET following agonist stimulation using the Perkin Elmer Fusion for BRET detection. These studies demonstrated that the GR agonist dexamethasone promoted an increase in BRET following transient cotransfection of the following fusion protein combinations: Renilla luciferase fused to the C-terminus of GR coexpressed with a fusion protein in which GFP was fused to either the N- or C-terminus of NFkB. It is important to note that in these studies other combinations of fusion proteins failed to generate a signal in the BRET assay. The reason for the lack of signal is undetermined but could be due to insufficient levels of expression of some fusion partners, or the generation of fusion partners in which the distance between the GFP and luciferase is too great for BRET. This example illustrates the requirement to generate a range of fusion proteins during assay development in order to identify the combination that generates the greatest signal in a BRET assay. Within the pharmaceutical industry, compounds are often stored in solvents such as dimethylsulfoxide (DMSO) to prevent compound degradation. Many cell-based screening assays are intolerant to DMSO concentrations of greater than 1% due to cytotoxicity. In this assay we have determined that the BRET ratio change in response to treatment with 1 mM dexamethasone was unchanged at DMSO concentrations of up to 2.5%, demonstrating the assay to be of appropriate robustness for drug screening. Similar robustness has been seen in other BRET assays and is a consequence of the ratio method used to calculate BRET. Increasing DMSO concentrations result in a decrease in emission from both Renilla luciferase and YFP; however, the ratio of emission is unchanged. In this assay a range of GR agonists promoted GR/NFkB interaction including dexamethasone (EC50 = 9 nM), fluticasone propionate (EC50 = 3.3 nM), Ru486 (EC50 = 2.6 nM), and Ru24858 (EC50 = 13.7 nM) (Fig. 16.4). As expected, progesterone antagonized the interaction between GR and NFkB (data not shown). This assay has been applied to screen for the effect of GR agonists on the interaction of this receptor with NFkB. This example demonstrates the application of BRET, and indeed FRET, to study the effect of compound activity directly on protein interactions.

16.3.5

Dual Colour Spectroscopy Assays

As an alternative to FRET, Kohl et al. (2002) have generated fusion proteins between EGFP and DsRed RCFP to develop a dual color correlation spectroscopy assay for protease inhibitors. This is a single-molecule-based detection technology that selectively probes the movement in solution of two fluorescent groups. The measurement principle is based on a spectrally resolved detection of single fluorescent molecules diffusing in and out of a diffraction-limited laser focus. Movement of the uncleaved FRET substrate through the laser allows the detection of both red and green fluorescence. Following protease cleavage GFP and DsRed RCFP become free to move in solution to cause only single molecules to be detected as they move through the laser focus. In contrast to FRET assays, the dual color correlation assay is not limited to certain ranges of distance between the two flurophores in the substrate because detection of cleavage does not rely upon FRET; instead, it relies upon the generation of distinct GFP and RCFP molecules. This assay is potentially more versatile and sensitive than FRET-based protease assays.

371

372

PHARMACEUTICAL APPLICATIONS OF GFP AND RCFP

Figure 16.4. BRET2 assay of ligand-mediated regulation of the interaction of the glucocorticoid receptor (GR) and NFkB. HEK 293 cells were transfected with the GR/Rluc and p65NFkB/GFP fusion proteins. Measurement of changes in BRET in response to various ligands was determined using the Packard BRETcount (Perkin Elmer Biosciences). The BRET signal was measured after stimulation with varying concentrations of the following agonists: RU24858 (䊏), RU486 (D), fluticasone propionate (䉱), and dexamethasone (䊉). (Data were generated in collaboration with Perkin Elmer Biosciences.)

16.3.6

Fluorescent Protein Degradation Assays

In addition to monitoring cellular localization, GFP fusion proteins can be used to track the lifetime of an expressed protein. An example of where this has been applied in drug discovery is the development of a screening assay for modulators of the interaction between IkBa and NFkB (Li et al., 1999). NFkB activity is upregulated by many proinflammatory cytokines including TNFa and IL-1 and is thought to play a fundamental role in many pro-inflammatory diseases. For this reason, many pharmaceutical companies are running drug discovery programs to identify antagonists of this signal transduction cascade. Prior to activation, NFkB exists as a complex with IkBa in the cytoplasm. Upon activation of the NFkB signal cascade, IkBa is phosphorylated and degraded to release NFkB, which is then able to exert its biological effects by modulating the expression of several genes. Degradation of IkBa is rapid and can be used as a reporter of NFkB activation. To establish a screening assay for modulators of the NFkB pathway, we have generated a fusion protein between IkBa and GFP. Following expression in mammalian cells, the degradation of this protein following the activation of the NFkB cascade can be measured on a confocal microscope (Fig. 16.5) or on an HCS machine such as the ArrayScanTM (Cellomics).

16.3.7 b-Arrestin Recruitment Assay Agonist-mediated translocation of arrestin has been applied to develop a highly novel and specific assay for GPCR ligands. This assay relies upon the detection of ligand-mediated recruitment of a fusion protein between b-arrestin1 or b-arrestin2 and GFP from the cytosol to the plasma membrane (Claing et al, 2002; Fig. 16.6). In this assay, stable mam-

FLUORESCENT PROTEIN DRUG SCREENING ASSAYS

Figure 16.5. Fluorescent protein assay of NFkB activity. A fusion protein between IkBa and GFP was transiently expressed in HeLa cells. Confocal images were taken various times after stimulation with TNFa and the fluorescence signal quantified. The images show the time course of the degradation induced by TNF alpha (A), and the graph shows the fluorescence intensity measured from these images over time (B).

malian cell lines are generated expressing both the GPCR and a fusion protein between b-arrestin and EGFP or b-arrestin and the dsRedRCFP. In the unstimulated cell the barrestin/fluorescent protein is distributed in the cytoplasm. Following agonist stimulation of the GPCR, the b-arrestin/fluorescent protein fusion is recruited to the cell membrane as part of the GPCR desensitization and internalization process. For many receptors, this is followed by internalization of the GPCR/arrestin/fluorescent protein complex into the endosome compartment. When visualized using a fluorescence or confocal plate reader, arrestin translocation is visualized as a change in the distribution of GFP fluorescence from the cytosol to the membrane (Fig. 16.7), or to generate a “spotted” appearance following internalization of the complex into the endosome compartment (Fig. 16.7). Because many, if not the majority, of GPCRs are internalized through the interaction of the receptor and b-arrestin, this assay appears to offer a generic HCS, and perhaps HTS, screening assay for GPCR ligands. The assay developed by Xsira Pharmaceuticals (Durham, North Carolina) and marketed as TransfluorTM has been applied to in excess of 20 receptors spanning the GPCR target class. This assay is now available from Molecular Devices, Sunnyvale, CA. This includes the b2-adrenoceptor (Barak et al., 1997b), the neurokinin NK-1 receptor (Richardson et al., 2003), the adenosine A2b receptor (Matharu et al., 2001), the metabotropic glutamate mGluR1 receptor (Mundell et al., 2001), the bradykinin B1 and B2 receptors (Lamb et al., 2001), and the CXCR1 chemokine receptor (Barlic et al., 1999). The application of this technology for GPCR screening is reviewed extensively by Conway and Demarest (2002).

373

374

PHARMACEUTICAL APPLICATIONS OF GFP AND RCFP

Figure 16.6. Schematic representation of b-arrestin-mediated GPCR internalization in response to agonist stimulation (see text for details). (Figure provided by Xsira Pharmaceuticals, North Carolina.)

We have also used the arrestin/GFP recruitment assay for GPCR HTS using the GE Healthcare InCell Analyser 3000TM to detect agonist-mediated redistribution of the arrestin/GFP fusion protein. In these studies a stable U2OS cell line was generated coexpressing the human b2-adrenoceptor together with the b-arrestin2/GFP fusion protein. A high-throughput screening assay was developed and applied to identify both agonists and antagonists of this receptor. This assay was demonstrated to be capable of identifying known b2-adrenoceptor ligands with the expected concentration dependence (Fig. 16.8). As mentioned previously, most pharmaceutical companies store their compound libraries in solvents such as DMSO. We have demonstrated that the performance of this assay is unaffected by DMSO concentrations of up to 2%. To demonstrate the application of this assay for HTS, a random set of 1280 compounds were screened in duplicate in 96-well plate format. Compounds were screened for both agonist and antagonist activity at this receptor. The screen was performed on two separate days to allow the reproducibility of the assay and the effect of compound interference to be assessed. Assay performance was determined according to Z¢, a statistical measure used to measure the quality of drug screening assays. A good HTS assay is defined as having a Z¢ factor of greater than 0.4 (Zhang et al., 1999). In the agonist and antagonist screens the Z¢ was consistently above 0.7 and 0.6, respectively. In the agonist screen, active molecules or “hits” were classified as all molecules with an activity of greater than 40% of the positive control agonist response. Using this measure, the agonist screen did not identify any active compounds.

FLUORESCENT PROTEIN DRUG SCREENING ASSAYS

Figure 16.7. Cytosol to membrane translocation of b-arrestin/fluorescent protein fusion proteins following ligand stimulation. (A) Cells expressing the human angiotensin AT1A receptor (top) and substance P receptor (bottom), together with a fusion protein between b-arrestin2 and GFP, were treated with the agonist ligands angiotensin and substance P, respectively. Images were taken at time 0 and after 30 min of agonist treatment. Data generated by Xsira Pharmaceuticals. (B) CHO cells stably expressing the human CCR2 chemokine receptor were transiently transfected with a fusion protein between b-arrestin2 and the Anemonia sulcata AsRed1 RCFP. Confocal images were taken at various times following stimulation with the CCR2 receptor ligand MCP-1 (10 mM). (Images were generated in collaboration with Professor G. Milligan, University of Glasgow.)

This was an expected observation because this compound set did not contain any b2adrenoceptor agonists. In the antagonist screen, active molecules were classified as all molecules that generated an inhibitory activity of greater than 40% of the control antagonist response. Using this measure, 1.2% of the compounds screened were determined to possess antagonist activity at the b2-adrenoceptor. As expected, the correlation of activity between compounds screened on different days illustrates that the majority of active compounds were identified on both days of assay. These assays are in use in many pharmaceutical companies for GPCR HTS in which chemical libraries of greater than half a million compounds have been screened to identify molecules of interest. Genome sequencing studies have identified approximately 160 so-called orphan GPCRs—that is, receptors with no known function and no known ligand (Wise et al., 2002). There is considerable activity both within the pharmaceutical industry and elsewhere aimed at the identification of the ligands and the physiological role of these new

375

376

PHARMACEUTICAL APPLICATIONS OF GFP AND RCFP

Figure 16.8. Application of the TransfluorTM b-arrestin recruitment assay for drug screening. U2OS cells stably expressing the human b2-adrenoceptor and a fusion protein between b-arrestin2 and GFP were provided for this study by Xsira Pharmaceuticals (North Carolina). Assays were performed in the 96-well microtiter plate format, and images were collected on using the GE Healthcare InCell Analyser 3000TM (GE Healthcare, Cardiff, UK). (A) Concentration response curve to the standard agonist isoprenaline. (B) Agonist screening assay. A set of 1280 random compounds were screened for agonist activity at 5 mM final concentration on two separate days. Plot shows the activity on day 1 plotted against the activity on day 2 (axis = % of maximal stimulation with isoprenaline). (C) Antagonist screening assay. A set of 1280 random compounds were screened for antagonist activity at 5 mM final concentration on two separate days. Cells were preincubated for 30 min with compound before stimulation with an EC80 concentration of isoprenaline (20 nM). Plot shows the activity on day 1 plotted against the activity on day 2 (axis = % of maximal inhibition of isoprenaline response). (Data were generated in collaboration with Xsira Pharmaceuticals, North Carolina.)

receptors (Wise et al., 2002). In the absence of an activating ligand, little is known about the signal transduction capabilities of an orphan receptor. Because b-arrestin recruitment following agonist binding at a GPCR is a phenomenon common to most GPCRs, the arrestin recruitment assay offers utility for the identification of ligands at orphan GPCRs. As a proof of concept experiment to investigate the use of this assay for these studies, HEK293 cells stably expressing the b-arrestin/GFP fusion protein were transiently transfected with the Neuromedin NMUR-1 receptor for which the ligand has recently been identified. Eighty random compounds were then screened for agonist activity in this assay. Five of the random compounds were spiked with a maximal concentration of NMU, the ligand for this receptor. The ability of the assay to detect the spiked samples was determined. Only those spiked wells showed any activity in this assay (Fig. 16.9).

16.3.8

GPCR/GFP Fusion Assays

A second HCS assay for the identification of GPCR ligands relies upon the generation of fusion proteins between the GPCR under study and the fluorescent protein to create a fluorescent receptor that is expressed on the cell surface. GFP has been fused to the intracellular C-terminal tail of more than 20 GPCRs with little or no effect on receptor pharmacology [reviewed in Kallal and Benovic (2000)]. Agonist binding to such a receptor results in receptor internalization into the endosome compartment as part of the signal transduction desensitization process. For example, McLean et al. (1999) generated fusion proteins between GFP and the native and a constituitively active mutant of the b2 adrenoceptor and used imaging techniques to study ligand activity at both forms of the receptor.

FLUORESCENT PROTEIN DRUG SCREENING ASSAYS

Figure 16.9. Application of the TransfluorTM b-arrestin recruitment assay for orphan GPCR screening. The neuromedin U receptor (NMR) was transiently transfected into HEK 293 cells stably expressing the b-arrestin/GFP fusion protein. In a 96-well microtiter plate screening assay, 80 random compounds were screened for agonist activity. The five compound wells that were spiked with Neuromedin U were active in this assay. Data were collected on using the GE Healthcare InCell Anaylser 3000TM (GE Healthcare, Cardiff, UK). (Data were generated in collaboration with Xsira Pharmaceuticals, North Carolina.)

In these studies the authors were able to determine both ligand efficacy and the ability of these ligands to regulate receptor degradation. This movement of fluorescent receptor is visualized as a membrane-to-endosome movement using HCS detection apparatus. This assay has been applied for HTS (Conway et al., 1999; Conway and Demarest, 2002).

16.3.9

Protein Complementation Assays

A number of protein complementation assays have been developed for both compound screening and target validation. Such assays have relied upon splitting enzymes such as b-galactosidase or b-lactamase into distinct domains with no enzymatic activity (Graham et al., 2001; Galarneau et al., 2002). Fusion proteins are generated between the domains of these enzymes and two proteins predicted to interact. The interaction of the protein partners facilitates the interaction of the enzyme domains to reconstitute enzyme activity. As such, the reconstitution of enzyme activity can be used as an indicator of protein–protein interaction. Such assays have been used to (a) screen random compounds to identify inhibitors of a known protein–protein interaction and (b) characterize the interaction between proteins presumed to interact. However, these assays do not permit the visualization of protein–protein interaction in living cells. To address this, Ozawa et al. (2001) have split GFP into two domains that independently do not exhibit fluorescence. When fused to interacting proteins, the two “halves” of GFP are brought into close proximity to reconstitute the GFP fluorophore. While the only successful demonstration of split GFP

377

378

PHARMACEUTICAL APPLICATIONS OF GFP AND RCFP

technology has been in E. coli, the application of this technology to mammalian cells could have great utility for drug discovery.

16.3.9

Fluorescence-Activated Cell Sorting

Fluorescence-activated cell sorting (FACS) is a long-established technology used to sort cells from within mixed cell populations and to determine the level of expression of proteins on the cell surface or, following permeabilization, within the cell (Herezenberg and de Rosa, 2000). Cells of different lineages express characteristic marker proteins. Through the use of fluorescently labeled antisera specific to these markers, it is possible to sort specific cells from within a mixed population. Similarly, it is possible to sort cells into different populations based on the level of expression of the marker protein. Cells expressing a greater level of the protein bind a greater level of the labeled antibody and hence can be sorted according to fluorescence intensity. GFP and RCFP have been widely applied for FACS as an alternative to the use of fluorescently labeled antibodies. In transient gene expression studies, cells are often cotransfected with the gene under study together with a GFP expression vector in order to sort transfected from untransfected cells by FACS. For example Guerra-Crespo et al. (2003) cotransfected primary rat hypothalamic neurones with a GFP reporter under the transcriptional control of the thyrotrophin-releasing hormone (TRH) promoter. FACS was applied to isolate TRH neurons for further study. Similar approaches have been developed to isolate specific cell lineages from transgenic animals. Tomomura et al. (2001) generated a transgenic mouse expressing GFP within Purkinje cells and applied FACS to isolate these cells. The use of GFP as a marker for FACS removes the requirement for the generation and labeling of specific antisera. The technique can be applied to sort cells on the basis of their expression of potentially any protein, and the technique requires no specific preparation of the cells prior to FACS. In addition to the use of FACS for monitoring the level of protein expression within a given cell type, FACS assays have been developed to detect the effect of drug molecules on the level of protein expression within given cells and the proportion of cells of a particular type within a tissue sample. Furthermore, GFP in combination with FACS has been used to isolate specific cell types from transgenic animals for the development of drug screening assays.

16.3.10

Fluorescent Protein Biosensors

GFP biosensors are genetically encoded probes, the fluorescent characteristics of which are modified by a change in the level of a cellular metabolite or second messenger protein. A GFP biosensor acts as a direct probe to enable the detection of a change in the level of expression of a second messenger metabolite in living mammalian cells in culture. The ability to genetically express reporter molecules in recombinant cell lines, primary cell lines, tissue slices, or indeed whole animals may be one of the most exciting applications of fluorescent protein technology (Conway and Demarest, 2002; Giuliano and Taylor, 1998). Such assays allow the direct visualization of signal transduction in conventional plate readers or confocal imaging systems without the need for assay reagent addition and cell lysis. Such assays are inexpensive to run and generate direct information regarding the level of the second messenger in the cell and, perhaps through the targeting of the biosensor, the level of the second messenger within a discrete cellular location. The use of GFP/RCFP as biosensors in drug screening assays for HTS offers a number of benefits compared to the use of chemical detection agents such as fluorescent dyes. Flu-

FLUORESCENT PROTEIN DRUG SCREENING ASSAYS

orescent dyes have been widely used to detect changes in cellular pH and changes in the intracellular concentration of sodium, calcium, potassium, and other ions, as indicators of cell number or cytotoxicity, and as detectors of the level of intracellular second messengers. Such dyes are expensive, the assays always involve a dye loading step that is often followed by a wash step to remove excess dye, the dyes are often subject to photobleaching, and they can be cytotoxic. In contrast, GFP/RCFP fluorescent indicator proteins are expressed within the cell; hence there is no need for dye loading and cell washing. The assays are inexpensive because no chemical detection agent is required, and there is little cytotoxicity. The absence of dye loading and wash steps in fluorescent protein assays is a particular advantage because this simplifies the assay protocol, leading to the generation of a more robust assay that is easier to transfer onto HTS laboratory automation. An early example of the use of GFP as a biosensor arose from the observation that the fluorescence characteristics of the YFP-H148Q mutant are pH-sensitive, leading to the application of this protein as a probe for changes in intracellular pH (Kneen et al., 1998). It was later observed that this protein exhibits halide sensitivity. At pH 7.5 the fluorescence emission of this protein decreases twofold (Jayaraman et al., 2000). In perhaps the first example of the use of a GFP biosensor for drug screening, a high-throughput screen has been run to identify activators of the cystic fibrosis transmembrane conductance regulator (CFTR). A stable cell line expressing both the CFTR and the YFP-H158Q protein was used to screen a 60,000 compound library in 96-well assay plates. CFTR activators caused an increase in intracellular chloride concentration to cause a decrease in YFPH148Q emission that was detected using a FluroStar fluorescence plate reader (BMG Lab Technologies). This screen resulted in the identification of a number of novel activators of CFTR and demonstrates the potential utility of this fluorescent protein for drug screening (Ma et al., 2002). Two of the most commonly used drug screening methods for GPCRs are the detection of drug-mediated changes in intracellular calcium or cAMP. Calcium assays are performed using the fluorescent indicator dyes Fluo-3 or Fluo-4 and the fluorescence imaging plate reader (FLIPR; Molecular Devices, Sunnyvale, CA). A variety of methods exist for the detection of changes in intracellular cAMP. Genetically encoded FRET indicators have been developed for the detection of changes in intracellular calcium (Miyawaki et al., 1997, 1999) and cAMP (Zaccolo et al., 2000). A calcium indicator, or chameleon, was developed through the construction of a fusion protein consisting of cyan fluorescent protein (CFP) and yellow fluorescent protein (YFP) separated by the calcium-binding protein calmodulin and the calmodulin-binding peptide M13. Binding of calcium to calmodulin facilitates the interaction between calmodulin and M13 to cause an intramolecular rearrangement to cause an increase in FRET between the flanking CFP and YFP. Such proteins have been used to monitor calcium levels in the cytosol, endoplasmic recticulum, and nucleus of living cells (Miyawaki et al., 1997). A similar fluorescent indicator for cAMP was developed through generating fusion proteins between the regulatory and catalytic subunits of protein kinase A and a blue fluorescent protein and GFP, respectively. Binding of cAMP to the regulatory subunit causes a dissociation of the PK-A complex to result in a decrease of FRET. In contrast to existing assays for changes in the intracellular concentration of cAMP, which involve cell lysis and the use of anti-cAMP antisera, fluorescent cAMP biosensors can be used non-invasively in living cells (Zuccolo et al., 2000). It will be of great interest to see whether the fluorescent protein biosensors will replace the use of these dye or biochemical assays for GPCR HTS.

379

380

PHARMACEUTICAL APPLICATIONS OF GFP AND RCFP

A GFP biosensor designed to detect changes in cellular membrane potential as a result of ion channel activity has been developed through the generation of a fusion protein between the Drosophila shaker potassium channel and GFP. In this example, the fluorescence characteristics of the fusion protein change according to the membrane potential of the cell (Guerrero and Isacoff, 2001). This probe offers an alternative to the use of fluorescent membrane potential dyes for compound screening at ion channels and other targets that regulate cell membrane potential.

16.4

HIGH CONTENT SCREENING DETECTION INSTRUMENTATION

Throughout this chapter we have described the development of HCS assays that rely upon the translocation of fluorescent protein fusion proteins, or the gain or loss of fluorescence from a fluorescent protein, in live and fixed cells. Prior to the development of instruments capable of the detection of changes in the level and intracellular localization of fluorescence in single cells plated in 96- and 384-well microtiter plates, assays such as receptor internalisation or nuclear translocation could be visualized on a fluorescence microscope but could not be quantified in a high-throughput manner. Moreover, the development of automated image analysis software available on such instruments has been critical to convert virtual features into numerical results that can be used to assess the efficacy of a drug. A number of instruments are now available that can be used for this purpose. Each of these instruments has different features such as differences in excitation light source or detection methods, cellular resolution, speed of image acquisition and data processing and cost of instrumentation. A brief comparison of currently available instruments can be found in Table 16.4. Most, if not all, are able to detect membrane to cytosol translocations (receptor internalisation), cytosol to membrane translocations (b-arrestin recruitment), and membrane to nucleus translocation (nuclear translocation assays) and will detect changes in fluorescence intensity. The majority of these readers can also be integrated onto robot tracks to allow fully automated high content screens to be run as high-throughput screens. It is difficult to provide an exhaustive list of the different applications of the various instruments because few comparative studies have been reported. Furthermore, the development of such instrumentation is proceeding at a rapid pace (Table 16.2).

16.5 APPLICATION OF FLUORESCENT PROTEINS IN TARGET VALIDATION STUDIES Genome sequencing and bioinformatic studies indicate that there are approximately 28,000 genes in the human genome. The role of the vast majority of these genes in both normal physiology and disease processes remains unknown. About 5000 of these genes fall into the so-called tractable target classes—that is, classes of protein such as GPCRs, ion channels, proteases, kinases, and so on, for which there are examples of clinically marketed drugs. There is intense effort within the pharmaceutical industry to identify the function of novel genes—in particular, those that fall within the tractable target classes—and their possible involvement in disease in order to identify the drug targets of the future. The techniques used to achieve this have been loosely grouped under the label “target validation” and have been reviewed elsewhere (Lindsay 2003; Wise et al., 2002). Target validation studies rely upon the detection of the spatial and temporal nature of target expression, the study of whether target expression is altered in disease, the identification of interacting

APPLICATION OF FLUORESCENT PROTEINS IN TARGET VALIDATION STUDIES

TABLE 16.4. Key Features of Commonly Used High Content Screening Detection Apparatus Company

Reader

Attributes

Cellomics

ArrayScanTM II

Mercury arc lamp, nonconfocal Four excitation and four emission filters CCD camera

ArrayScanTM Kinetics

Mercury arc lamp, nonconfocal Eight excitation and eight emission filters CCD camera Liquid handling, robotics

IN Cell Anaylser 1000TM

Xenon lamp, line scanning confocal Multiple excitation and emission filters CCD camera Liquid handling

IN Cell Anaylser 3000TM

Laser (3 lines), line scanning confocal Multiple emission filters 3¥ CCD cameras Liquid handling

Evotec

Opera

Laser (3 lines), Nipkow disc, confocal Multiple emission filters 2¥ CCD cameras Plate formats up to 2080 Separate image capture and analysis

Acumen

Explorer

Laser (488 nm), nonconfocal 4 PMT detection Independent of plate type

Molecular Devices

Discovery-1

Arc lamp Multiple excitation and emission filters CCD camera

GE Healthcare

partners of proteins of interest, the generation of transgenic animals, and many other studies. The ultimate goal of target validation is the generation of data that indicates that therapeutic intervention at the protein under study will have efficacy in human disease. Aequorea victorea GFP and the RCFPs are becoming valuable tools within the target validation phase of drug discovery. Fluorescent proteins have been applied to determine the tissue and subcellular localization of a novel protein. The expression in mammalian cell lines of fusion proteins between a fluorescent protein and the protein under study can allow the detection of the site of expression of that protein, and can be used to determine if this is altered by known stimuli in order to provide some indication of function. Similar studies in transgenic animals can identify all sites of protein expression, and studies in disease models can be used to determine if gene expression is altered during disease progression (Hadjantonakis and Nagy, 2001). Fluorescent proteins are being extensively used in pathway mapping. The objective of such studies is to take a protein of unknown function and identify interacting partners in order to identify whether that protein is involved in a known cellular process. In the past techniques such as the yeast two-hybrid system have been used for this purpose (Causier and Davies, 2002). Many groups are now developing both FRET- and BRETbased assays in order to perform such studies in mammalian cells and other species. FRET

381

382

PHARMACEUTICAL APPLICATIONS OF GFP AND RCFP

and BRET are also being applied to confirm the results of yeast two-hybrid studies in mammalian cells. One example of this approach is the demonstration of the interaction of the GABA-BR1 and GABA-BR2 receptors by FRET which confirmed earlier studies using the yeast two hybrid assay (White et al., 2002; Fig. 16.3). The investigation of the function of a target gene has been greatly facilitated through the development of overexpression and gene deletion techniques in model species such as the nematode worm Caenorhabditis elegans (David et al., 2003), the fruit fly Drosophila melanogaster (Misteli and Spector 1997; Brand 1999), and transgenic mice (Hadjantonakis and Nagy, 2001; Misteli and Spector, 1997; Zambrowicz and Sands, 2003). Such techniques are widely applied throughout academia and are in many pharmaceutical companies. Fluorescent protein technology has been applied to facilitate the detection of the site of expression of the gene of interest and to develop whole organism model systems to study the effects of novel drugs. The nematode worm C. elegans contains just 128 cells. The developmental fate and function of each of these cells has been well-characterized. As a result, this animal has been applied to study the effect of the deletion of genes on cell fate patterning during development and on the physiology of the mature worm. Fluorescent proteins have been applied as a reporter to enable the noninvasive generation of data from the live worm. As an example of the application of GFP in this species, David et al. (2003) have developed a strain of C. elegans expressing a hsp16-GFP-lacZ fusion protein for use in environmental monitoring. In this animal a chimeric GFP/b-galactosidase (lacZ) reporter gene was fused in frame into the C. elegans hsp16 gene such that expression of the reporter is under the transcriptional control of the hsp16 promoter. The reporter was modified to facilitate detection such that upon expression it accumulates in the cell nucleus. This worm has been used to examine the effects of environmental stressors such as heat and microwave radiation. In each case the authors were able to monitor the stress effect in the whole organism through the detection of the fluorescence intensity of the cell nuclei. GFP has been used extensively in transgenic mice as a reporter of gene expression. Such studies have been reviewed by Hadjantonakis and Nagy (2001) and Misteli and Spector (1997). Technologies have been developed that permit the targeted deletion of specific genes in mice. During the generation of such animals, the inclusion of a reporter gene in the gene deletion vector enables the creation of transgenic mice no longer expressing the gene under study and now expressing the reporter gene under the transcriptional control of the promoter of the deleted gene. Similarly, experiments have been performed to generate fusion proteins between the gene under study and a reporter gene. GFP has been widely applied as the reporter in transgenic animals because the sites of gene expression can be readily detected in cells and tissue slices by fluorescence microscopy. For example, a transgenic mouse expressing endothelial nitric oxide synthase (eNOS) fused to GFP was generated to identify the location and regulation of e-NOS expression (van Haperen et al, 2003). The expression of the fusion protein facilitated the isolation of eNOS expressing tissues from these animals through the detection of GFP fluorescence. These animals have been used to study the effects on eNOS expression of several vascular challenges. This application of GFP permits the study of the temporal and spatial expression of the targeted gene through imaging of GFP expression. In a second example a transgenic mouse was generated expressing GFP under the control of the melanocortin-4 (MC4) receptor promoter (Liu et al., 2003). This receptor is an important regulator of energy homeostasis, and antagonists of this receptor are proposed to have efficacy in obesity. These animals have been used to study the site of MC4 receptor expression and the role of this receptor on feeding behavior. In a final example, transgenic mice have been developed expressing

CONCLUDING REMARKS

GFP under the control of the mouse insulin I gene promoter (MIP). In such animals, GFP is expressed within the insulin-secreting beta cells of the pancreatic islet. These animals are being applied to study beta cell biology in normal and diabetic animals (Hara et al., 2003). Fluorescent proteins have been applied in transgenic mice to determine the effect of drug molecules on the level of expression of the deleted gene in the intact animal through monitoring changes in GFP expression. Lindsten et al. (2003) created transgenic mice expressing a GFP reporter carrying a constitutively active degradation signal to generate a model for the study of the ubiquitin/proteasome system. Impairment of this system has been proposed to play a role in many neurodegenerative disorders. Administration of proteosome inhibitors to these animals resulted in the accumulation of GFP in several tissues through the prevention of GFP degradation. This animal has been used as a model for the characterization of novel inhibitors of the ubiuquitin/proteosome system. Furthermore, primary neurones have been isolated from these animals and used in transfection studies to demonstrate that an aberrant ubiquitin found in Alzheimer’s disease patients causes the accumulation of the GFP reporter. These cells and animals could be applied to understand the role of ubiquitin in this disease and to screen for inhibitors of this complex. In a further novel application of GFP as a reporter in transgenic mice, Metzger et al. (2002) have expressed the pH- and halide-sensitive GFP described earlier (Jayaraman et al., 2000) in transgenic mice under the control of a potassium channel promoter. Neuronal tissue from these animals has been used to study drug effects by imaging changes in cellular fluorescence. The generation and characterization of transgenic animals expressing GFP reporters has facilitated the study of the effects of external stimuli, including novel small-molecule drugs, both within the intact animal and in tissue slices and primary cell lines generated from such animals. The advantages of GFP as a reporter for these purposes is that drug effects can be monitored noninvasively by simple imaging of the cell lines or tissue samples under study. In addition to the application of GFP for target validation, fluorescent protein technology may have application within the toxicology departments of pharmaceutical companies. It will be possible to develop novel noninvasive assays that permit the study of compound activity within both the whole animal and tissue samples by determining the effect of such compounds on the site and level of expression of GFP reporters. The challenge in this work will be the development and validation of assays suitable for this purpose.

16.6

CONCLUDING REMARKS

In the last five years, fluorescent protein technology has become integral to drug discovery. To date, the greatest impact of such proteins has been within the compound screening phase of drug discovery for which a plethora of fluorescent protein screening assays and fluorescent protein probes have been developed. In parallel, many instruments have become available that allow the detection of changes in the fluorescence characteristics of single and populations of cells within microtiter plate formats. It is now possible to determine the effect on the subcellular localization of a fluorescent protein fusion protein of 384 compounds within six minutes, a process which by confocal microscopy would have taken several days, if not weeks. In the coming years, with the development of more sensitive and faster detection apparatus, and improved fluorescent protein probes and biosensors, fluorescent protein screening assays will become integral to the high-throughput

383

384

PHARMACEUTICAL APPLICATIONS OF GFP AND RCFP

screening and pharmacology departments in most pharmaceutical companies. These developments should permit drug screening to be performed on biologically relevant cell types, perhaps including primary human cells, at very high throughput. The application of fluorescent protein technology within the development organisations of most drug companies is in its infancy. In the coming years, fluorescent protein technology will be applied to generate novel drug developability and toxicology assays. The application of fluorescent protein assays for primary screening, developability, and toxicology should lead to the identification of molecules with improved efficacy, toxicity profiles, and pharmacokinetics and thus should contribute to the reduction of both both cycle time and attrition within the drug discovery process.

ACKNOWLEDGMENTS The authors would like to acknowledge the contribution of many colleagues to the work described in this chapter. In particular, we would like to thank Mike Allen, Peter Chalk, Katy Gearing, Debbie Graham, Brian Hayes, Peter Lowe, Barbara Maschera, Rebecca Milton, Alan Wise, and Julie White, all of GlaxoSmithKline, Stevenage. These studies have been performed through a number of collaborations. The BRET studies were performed in collaboration with Benoit Houle and Luc Menard of Perkin Elmer Life Sciences, Montreal, and Greame Milligan and Douglas Ramsay of the University of Glasgow. The GR translocation studies were performed in collaboration with Suzanne Hancock and Fergus Mckenzie of GE Healthcare Biosciences, Cardiff and the arrestin recruitment studies with Robert Oakly, Rachel Cruickshank, Shay Rhem, and Carson Loomis of Xsira Pharmaceuticals, North Carolina.

REFERENCES Almholt, K., Arkhammar, P. O., Thastrup, O., and Tullin, S. (1999). Simultaneous visualization of the translocation of protein kinase Ca-green fluorescent protein hybrids and intracellular calcium concentrations. Biochem. J. 337:211–218. Awaji, T., Hirasawa, A., Kataoka, M., Shinoura, H., Nakayama. Y., Sugawara, T., Izumi, S.-I., and Tsujimoto, G. (1998). Real-time optical monitoring of ligand-mediated internalization of a1badrenoceptor with green fluorescent protein. Mol. Endocrinol. 12:1099–1111. Barak, L. S., Ferguson, S. S., Zhang, J., and Caron, M. G. (1997a). A b-arrestin/green fluorescent protein biosensor for detecting G protein-coupled receptor activation. J. Biol. Chem. 272: 27497–27500. Barak, L. S., Ferguson, S. S. G., Zhang, J., Martenson, C., Meyer, T., and Caron, M. G. (1997b). Internal trafficking and surface mobility of a functionally intact b2-adrenergic receptor-green fluorescent protein conjugate. Mol. Pharm. 51:177–184. Barlic, J., Khandaker, M. H., Mahon, E., Andrews, J., DeVries, M. E., Mitchell, G. B., Rahimpour, R., Tan, C. M., Ferguson, S. S., and Kelvin, D. J. (1999). b-Arrestins regulate interleukin-8induced CXCR1 internalization. J. Biol. Chem. 274:16287–16294. Berdichevsky, Y., Zemel, R., Bachmatov, L., Abramovich, A., Koren, R., Sathiyamoorthy, P., GolanGoldhirsh, A., Tur-Kaspa, R., and Benhar, I. (2003). Novel high throughput screening assay for HCV NS3 serine protease inhibitors. J. Virol. Methods 107:245–255. Bertrand, L., Parent, S., Caron, M., Legault, M., Joly, E., Angers, S., Bouvier, M., Brown, M., Houle, B., and Menard, L. (2002). The BRET2/arrestin assay in stable recombinant cells: A platform to

REFERENCES

screen for compounds that interact with G protein-coupled receptors. J. Rec. Signal Transduction Res. 22:533–541. Boute, N., Jockers, R., and Issad, T. (2002). The use of resonance energy transfer in highthroughput screening: BRET versus FRET. Trends Pharmacol. Sci. 23:351–354. Boute, N., Boubekeur, S., Lacasa, D., and Issad, T. (2003). Dynamics of the interaction between the insulin receptor and protein tyrosine-phosphatase 1B in living cells. EMBO Rep. 4:313–319. Brand, A. (1999). GFP as a cell and developmental marker in the Drosophila central nervous system. Methods Cell Biol. 58:153–163. Causier, B., and Davies, B. (2002). Analysing protein–protein interactions with the yeast two-hybrid system. Plant Mol. Biol. 50:855–870. Claing, A., Laporte, S. A., Caron, M. G., and Lefkowitz, R. J. (2002). Endocytosis of G proteincoupled receptors: Roles of G protein-coupled receptor kinases and b-arrestin proteins. Prog. Neurobiol. 66:61–79. Conway, B. R., and Demarest, K. T. (2002). The use of biosensors to study GPCR function: Application for high-content screening. Receptors and Channels 8:331–341. Conway, B. R., Minor, L. K., Xu, J. Z., Gunnet, J. W., DeBiasio, R., D’Andrea, M. R., Rubin, R., DeBiasio, R. L., Giuliano, K., Zhou, L., and Demarest, K. T. (1999). Quantification of G-protein coupled receptor internalization using G-protein coupled receptor-green fluorescent protein conjugates with the arrayscan high-content screening system. J. Biomol. Scr. 4:75–86. Conway, B. R., Minor, L. K., Xu, J. Z., D’andrea, M. R., Ghosh, R. N., and Demarest, K. T. (2001). Quantitative analysis of agonist-dependent parathyroid hormone receptor trafficking in whole cells using a functional green fluorescent protein conjugate. J. Cell. Physiol. 189:341– 355. David, H. E., Dawe, A. S., de Pmoerai, D. I., Jones, D., Candido, E. P., and Daniells, C. (2003). Construction and evaluation of transgenic hsp16-GFP-lacZ Caenorhabditis elegans strain for environmental monitoring. Environ. Toxicol. Chem. 22:111–118. Drmota, T., Novotny, J., Kim, G.-D., Eidne, K. A., Milligan, G., and Svoboda, P. (1998). Agonistinduced internalization of the G protein G11a and thyrotropin-releasing hormone receptors proceed on different time scales. J. Biol. Chem. 273:21699–21707. Eidne, K. A., Kroeger, K. M., and Hanyaloglu, A. C. (2002). Applications of novel resonance energy transfer techniques to study dynamic hormone receptor interactions in living cells. Trends Endocrinol. Metabol. 13:415–421. Evans, N. A., Groarke, D. A., Warrack, J., Greenwood, C. J., Dodgson, K., Milligan, G., and Wilson, S. (2001). Visualizing differences in ligand-induced b-arrestin-GFP interactions and trafficking between three recently characterized G protein-coupled receptors. J. Neurochem. 77:476–485. Feng, X., Zhang, J., Barak, L. S., Meyer, T., Caron, M. G., and Hannun, Y. A. (1998). Visualization of dynamic trafficking of a protein kinase C bII/green fluorescent protein conjugate reveals differences in G protein-coupled receptor activation and desensitization. J. Biol. Chem. 273:10755–10762. Ferguson, S. S., Zhang, J., Barak, L. S., and Caron, M. G. (1998). Molecular mechanisms of G protein-coupled receptor desensitization and resensitization. Life Sci. 62:1561–1565. Galarneau, A., Primeau, M., Trudeau, L. E., and Michnick, S. W. (2002). b-Lactamase protein fragment complementation assays as in vivo and in vitro sensors of protein protein interactions. Nat. Biotechnol. 20:619–622. Giuliano, K. A., and Taylor, D. L. (1998). Fluorescent-protein biosensors: New tools for drug discovery. Trends Biotechnol. 16:135–140. Giuliano, K. A., DiBiasio, R. L., Dunlay, R. T., Gough, A., Volosky, J. M., Zock, J., Pavlakis, G. N., and Taylor, D. L. (1997). High-content screening: A new approach to easing key bottlenecks in the drug discovery process. J. Biomol. Scr. 2:249–259.

385

386

PHARMACEUTICAL APPLICATIONS OF GFP AND RCFP

Graham, D. L., Bevan, N., Lowe, P. N., Palmer, M., and Rees, S. (2001). Application of bgalactosidase enzyme complementation technology as a high throughput screening format for antagonists of the epidermal growth factor receptor. J. Biomol. Scr. 6:401–411. Groarke, D. A., Wilson, S., Krasel, C., and Milligan, G. (1999). Visualization of agonist-induced association and trafficking of green fluorescent protein-tagged forms of both b-arrestin-1 and the thyrotropin-releasing hormone receptor-1. J. Biol. Chem. 274:23263–23269. Guerra-Crespo, M., Charli, J. L., Rosales-Garcia, V. H., Pedraza-Alva, G., and Perez-Martinez, L. (2003). Polyetheylenimine improved the transfection efficiency of primary cultures of postmitotic rat fetal hypothalamic neurons. J. Neurosci. Methods 127:179–192. Guerrero, G., and Isacoff, E. Y. (2001). Genetically encoded optical sensors of neuronal activity and cellular function. Curr. Opin. Neurobiol. 11:601–607. Hadjantonakis, A. K., and Nagy, A. (2001). The color of mice: In the light of GFP-variant reporters. Histochem. Cell Biol. 115:49–58. Hara, M., Wang, X., Kawamura, T., Bindokas, V. P., Dizon, R. F., Alcoser. S. Y., Magnuson, M. A., and Bell, G. I. (2003). Transgenic mice with green fluorescent protein-labelled pancreatic beta cells. Am. J. Physiol. 284:E177–E183. Herezenberg, L. A., and de Rosa, S. C. (2000). Moloclonal antibodies and FACS: Complementary tools for immunobiology and medicine. Immunol. Today 21:383–390. Hill, S. J., Baker, J. G., and Rees, S. (2001). Reporter gene systems for the study of G-protein coupled receptors. Curr. Opin. Pharmacol. 1:526–532. Horgan, A. M., and Stork, P. J. (2003). Examining the mechanism of Erk nuclear translocation using green fluorescent protein. Exp. Cell Res. 285:208–220. Htun, H., Barsony, J., Renyi, I., Gould, D. L., and Hager, G. L. (1996). Visualization of glucocorticoid receptor translocation and intranuclear organization in living cells with a green fluorescent protein chimera. Proc. Natl. Acad. Sci. USA 93:4845–4850. Issad, T., Boute, N., and Pernet, K. (2002). The activity of the insulin receptor assessed by bioluminescence resonance energy transfer. Ann. NY Acad. Sci. 973:120–123. Jayaraman, S., Haggie, P., Wachter, R. M., Remington, S. J., and Verkman, A. S. (2000). Mechanism and cellular applications of a green fluorescent protein-based halide sensor. J. Biol. Chem. 275:6047–6050. Jensen, A. A., Hansen, J. L., Sheikh, S. P., and Brauner-Osborne, H. (2002). Probing intermolecular protein–protein interactions in the calcium-sensing receptor homodimer using bioluminescence resonance energy transfer (BRET). Eur. J. Biochem. 269:5076–5087. Jones, J., Heim, R., Hare, E., Stack, J., and Pollok, B. A. (2000). Development and application of a GFP-FRET intracellular caspase assay for drug screening. J. Biomol. Scr. 5:307–318. Kallal, L., Gagnon, A. W., Penn, R. B., and Benovic, J. L. (1998). Visualization of agonist-induced sequestration and down-regulation of a green fluorescent protein-tagged b2-adrenergic receptor. J. Biol. Chem. 273:322–328. Kallal, L., and Benovic, J. L. (2000). Using green fluorescent proteins to study G-protein-coupled receptor localization and trafficking. Trends Pharmacol. Sci. 21:175–180. Kapur, R. (2002). Fluorescence imaging and engineered biosensors: functional and activity-based sensing using high content screening. Ann. NY Acad. Sci. 961:196–197. Kneen, M., Farinas, J., Li, Y., and Verkman, A. S. (1998). Green fluorescent protein as a noninvasive intracellular pH indicator. Biophys. J. 74:1591–1599. Kohl, T., Heinze, K. G., Kuhlemann, R., Koltermann, A., and Schwille, P. (2002). A protease assay for two-photon cross correlation and FRET analysis based solely on fluorescent proteins. Proc. Natl. Acad. Sci. USA 99:12161–12166. Koster, M., and Hauser, H. (1999). Dynamic redistribution of STAT1 protein in IFN signaling visualized by GFP fusion proteins. Eur. J. Biochem. 260:137–144.

REFERENCES

Lamb, M. E., De Weerd, W. F., and Leeb-Lundberg, L. M. (2001). Agonist-promoted trafficking of human bradykinin receptors: Arrestin- and dynamin-independent sequestration of the B2 receptor and bradykinin in HEK293 cells. Biochem. J. 355:741–750. Li, X., Zhao, X., Fang, Y., Jiang, X., Duong, T., Fan, C., Huang, C. C., and Kain, S. R. (1998). Generation of destabilized green fluorescent protein as a transcription reporter. J. Biol. Chem. 273:34970–34975. Li, X., Fang, Y., Zhao, X., Jiang, X., Duong, T., and Kain, S. R. (1999). Characterization of NFkB activation by detection of green fluorescent protein-tagged IkB degradation in living cells. J. Biol. Chem. 274:21244–21250. Lindstren, K., Menendez-Benito, V., Masucci, M. G., and Dantuma, N. P. (2003). A transgenic mouse model of the ubiquitin/proteasome system. Nat. Biotechnol. 21:897–902. Liptrot, C. (2001). High content screening—From cells to data to knowledge. Drug Disc. Today 6:832–834. Liu, H., Kishi, T., Roseberry, A. G., Cai, X., Lee, C. E., Montez, J. M., Friedman, J. M., and Elmquist. J. K. (2003). Transgenic mice expressing green fluorescent protein under the control of the melanocortin-4 receptor promoter. J. Neurosci. 23:7143–7154. Lindsay, M. (2003). Target discovery. Nat. Rev. Drug Disc. 2:831–838. Loster, K., and Horstkorte, R. (2000). Enzymatic quantification of cell–matrix and cell–cell adhesion. Micron 31:41–53. Luttrell, L. M., Ferguson, S. S. G., Daaka, Y., Miller, W. E., Maudsley, S., Della Rocca, G. J., Lin, F.-T., Kawakatsu, H., Owada, K., Luttrell, D. K., Caron, M. G., and Lefkowitz, R. J. (1999). barrestin-dependent formation of b2 adrenergic receptor-Src protein kinase complexes. Science 283:655–661. Ma, T., Vetrivel, L., Yang, H., Pedemonte, N., Zegarra-Moran, O., Galietta, L. J., and Verkman, A. S. (2002). High-affinity activators of cystic fibrosis transmembrane conductance regulator (CFTR) chloride conductance identified by high-throughput screening. J. Biol. Chem. 277: 37235–37241. Mahajan, N. P., Harrison-Shostak, D. C., Michaux, J., and Herman, B. (1999). Novel mutant green fluorescent protein protease substrates reveal the activation of specific caspases during apoptosis. Chem. Biol. 6:401–409. Matharu, A. L., Mundell, S. J., Benovic, J. L., and Kelly, E. (2001). Rapid agonist-induced desensitization and internalization of the A(2B) adenosine receptor is mediated by a serine residue close to the COOH terminus. J. Biol. Chem. 276:30199–30207. Mathieu, S., and El-Battari, A. (2003). Monitoring E-selectin-mediated adhesion using green and red fluorescent proteins. J. Immunol. Methods 272:81–92. McLean, A. J., Bevan, N., Rees, S., and Milligan, G. (1999). Visualizing differences in ligand regulation of wild-type and constitutively active mutant b2-adrenoceptor-green fluorescent protein fusion proteins. Mol. Pharm. 56:1182–1191. Mercier, J. F., Salahpour, A., Angers, S., Breit, A., and Bouvier, M. (2002). Quantitative assessment of b1- and b2-adrenergic receptor homo- and heterodimerization by bioluminescence resonance energy transfer. J. Biol. Chem. 277:44925–44931. Metzger, F., Repunte-Canaonigo, V., Matsushita, S., Akeman, W., Diez-Garcia, J., Ho, C. S., Iwasto, T., Grandes, P., Itohara, S., Joho, R. H., and Knopfel, T. (2002). Transgenic mice expressing a pH and Cl- sensing yellow fluorescent protein under the control of a potassium channel promoter. Eur. J. Neurosci. 15:40–50. Misteli, T., and Spector, D. L. (1997). Applications of green fluorescent protein in cell biology and biotechnology. Nat. Biotechnol. 15:961–964. Miyawaki, A., Llopis, J., Heim, R., McCaffery, J. M., Adams, J. A., Ikura, M., and Tsien, R. Y. (1997). Fluorescent indicators for Ca2+ based on green fluorescent proteins and calmodulin. Nature 388:882–887.

387

388

PHARMACEUTICAL APPLICATIONS OF GFP AND RCFP

Miyawaki, A., Griesbeck, O., Heim, R., and Tsien, R. Y. (1999). Dynamic and quantitative Ca2+ measurements using improved cameleons. Proc. Natl. Acad. Sci. USA 96:2135–2140. Mundell, S. J., Matharu, A. L., Pula, G., Roberts, P. J., and Kelly, E. (2001). Agonist-induced internalization of the metabotropic glutamate receptor 1a is arrestin- and dynamin-dependent. J. Neurochem. 78:546–551. Nagy, S. R., Liu, G., Lam, K. S., and Denison, M. S. (2002). Identification of novel Ah receptor agonists using a high-throughput green fluorescent protein-based recombinant cell bioassay. Biochemistry 41:861–868. Ozawa, T., Takeuchi, T. M., Kaihara, A., Sato, M., and Umezawa, Y. (2001). Protein splicing-based reconstitution of split green fluorescent protein for monitoring protein–protein interactions in bacteria: Improved sensitivity and reduced screening time. Anal. Chem. 73:5866–5874. Patki, V., Buxton, J., Chawla, A., Lifshitz, L., Fogarty, K., Carrington, W., Tuft, R., and Corvera, S. (2001). Insulin action on GLUT4 traffic visualized in single 3T3-l1 adipocytes by using ultrafast microscopy. Mol. Biol. Cell. 12:129–141. Rees, S., Brown, S., and Stables, J. (1999). Reporter gene systems for the study of G-protein coupled receptor signal transduction in mammalian cells. In Milligan, G., Eds. Signal Trandsduction: A Practical Approach, 2nd Ed., Oxford University Press, New York. pages 171–221. Richardson, M. D., Balius, A. M., Yamaguchi, K., Freilich, E. R., Barak, L. S., and Kwatra, M. M. (2003). Human substance P receptor lacking the C-terminal domain remains competent to desensitize and internalize. J. Neurochem. 84:854–863. Schlador, M. L., and Nathanson, N. M. (1997). Synergistic regulation of m2 muscarinic acetylcholine receptor desensitization and sequestration by G protein-coupled receptor kinase-2 and b-arrestin1. J. Biol. Chem. 272:18882–18890. Schmid, J. A., Birbach, A., Hofer-Warbinek, R., Pengg, M., Burner, U., Furtmuller, P. G., Binder, B. R., and de Martin, R. (2000). Dynamics of NFkB and IkBa studied with green fluorescent protein (GFP) fusion proteins. Investigation of GFP-p65 binding to DNA by fluorescence resonance energy transfer. J. Biol. Chem. 275:17035–17042. Scott, E. S., Malcomber, S., and O’Hare, P. (2001). Nuclear translocation and activation of the transcription factor NFAT is blocked by herpes simplex virus infection. J. Virol. 75:9955– 9965. Sekar, R. B., and Periasamy, A. (2003). Fluorescence resonance energy transfer (FRET) microscopy imaging of live cell protein localizations. J. Cell Biol. 160:629–633. Slice, L. W., Yee, H. F. Jr., and Walsh, J. H. (1998). Visualization of internalization and recycling of the gastrin releasing peptide receptor-green fluorescent protein chimera expressed in epithelial cells. Receptors and Channels 6:201–212. Subbaramaiah, K., Bulic, P., Lin, Y., Dannenberg, A. J., and Pasco, D. S. (2001). Development and use of a gene promoter-based screen to identify novel inhibitors of cyclooxygenase-2 transcription. J. Biomol. Scr. 6:101–110. Tarasova, N. I., Stauber, R. H., Choi, J. K., Hudsoni, E. A., Czerwinski, G., Miller, J. L., Pavlakis, G. N., Michejda, C. J., and Wank, S. A. (1997). Visualization of G protein-coupled receptor trafficking with the aid of the green fluorescent protein. J. Biol. Chem. 272:14817–14824. Tawa, P., Tam, J., Cassady, R., Nicholson, D. W., and Xanthoudakis, S. (2001). Quantitative analysis of fluorescent caspase substrate cleavage in intact cells and identification of novel inhibitors of apoptosis. Cell Death Differ. 8:30–37. Terstappen, G. C., Giacometti, A., Ballini, E., and Aldegheri, L. (2000). Development of a functional reporter gene HTS assay for the identification of mGluR7 modulators. J. Biomol. Scr. 5:255–262. Tomomura, M., Rice, D. S., Morgan, J. I., and Yuzaki, M. (2001). Purification of Purkinje cells by fluorescence activated cell sorting from transgenic mice that express green fluorescent protein. Eur. J. Neurosci. 14:57–63.

REFERENCES

van Haperen, R., Cheng, C., Mees, B. M., van Deel, E., de Waard, M., van Damme, L. C., van Gent, T., van Aken, T., Krams, R., Dunker, D. J., and de Crom, R. (2003). Functional expression of endothelial nitric oxide synthase fused to green fluorescent protein in transgenic mice. Am. J. Path. 163:1677–1686. Vrecl, M., Anderson, L., Hanyaloglu, A., McGregor, A. M., Groarke, A. D., Milligan, G., Taylor, P. L., and Eidne, K. A. (1998). Agonist-induced endocytosis and recycling of the gonadotropinreleasing hormone receptor: Effect of b-arrestin on internalization kinetics. Mol. Endocrinol. 12:1818–1829. Walters, W. P., and Namchuk, M. (2003). Designing screens: How to make your hits a hit. Nature Drug Disc. Today 2:259–266. Wang, X. J., Liao, H. J., Chattopadhyay, A., and Carpenter, G. (2001). EGF-dependent translocation of green fluorescent protein-tagged PLC-g1 to the plasma membrane and endosomes. Exp. Cell Res. 267:28–36. White, J. H., Wise, A., and Marshall, F. H. (2002). Heterodimerization of gamma-aminobutyric acid B receptor subunits as revealed by the yeast two-hybrid system. Methods 27:301–310. Wise, A., Gearing, K., and Rees, S. (2002). Target validation of G-protein coupled receptors. Drug Discov. Today 7:235–246. Zaccolo, M., De Giorgi, F., Cho, C. Y., Feng, L., Knapp, T., Negulescu, P. A., Taylor, S. S., Tsien, R. Y., and Pozzan, T. (2000). A genetically encoded, fluorescent indicator for cyclic AMP in living cells. Nat. Cell Biol. 2:25–29. Zambrowicz, B. P., and Sands, A. T. (2003). Knockouts model the 100 best-selling drugs—will they model the next 100? Nat. Rev. Drug Disc. 2:38–51. Zhang, J., Ferguson, S. S., Barak, L. S., Aber, M. J., Giros, B., Lefkowitz, R. J., and Caron, M. G. (1997). Molecular mechanisms of G protein-coupled receptor signaling: Role of G proteincoupled receptor kinases and arrestins in receptor desensitization and resensitization. Receptors and Channels 5:193–199. Zhang, J., Barak, L. S., Anborgh, P. H., Laporte, S. A., Caron, M. G., and Ferguson, S. S. G. (1999). Cellular trafficking of G protein-coupled receptor/b-arrestin endocytic complexes. J. Biol. Chem. 274:10999–11006. Zhu, X. G., Hanover, J. A., Hager, G. L., and Cheng, S. Y. (1998). Hormone-induced translocation of thyroid hormone receptors in living cells visualized using a receptor green fluorescent protein chimera. J. Biol. Chem. 273:27058–27063.

389

17 REASSEMBLED GFP: DETECTING PROTEIN–PROTEIN INTERACTIONS AND PROTEIN EXPRESSION PATTERNS Thomas J. Magliery* Department of Molecular Biophysics & Biochemistry, Yale University, New Haven, CT

Lynne Regan Department of Molecular Biophysics & Biochemistry and Department of Chemistry, Yale University, New Haven, CT Genomic research has resulted in the identification of tens of thousands of putative proteins from all three domains of life in recent years, many of which have no clear function. Key clues to the function of these proteins come from identifying their binding partners and expression patterns. Therefore, it is now of critical importance to develop robust, highthroughput methods to address these issues (Zhu et al., 2003). Immunoprecipitation and related methods like TAP-TAG (Puig et al., 2001) require purification of the analyte protein, demand relatively strong interactions between protein partners, and are not amenable to library approaches. Fusions to Aequorea victoria GFP and its variants have been used to examine expression patterns (Chalfie et al., 1994) and protein interactions through fluorescence resonance energy transfer (FRET) (Miyawaki et al., 1997), but these methods are limited by the photophysical properties of GFP variants and the promoters available to drive expression, particularly in whole organisms. Several combinatorial screens based on the reassembly of dissected proteins have been introduced in recent years to determine the identity of protein ligands, beginning with the yeast two-hybrid screen. In the last five years, experiments have demonstrated that GFP and its variants can be dissected and reassembled to yield fluorescent products. GFP reassembly can be used to demonstrate and identify protein–protein interactions and protein expression patterns in cells and whole organisms. * Present address: Department of Chemistry and Department of Biochemistry, The Ohio State University, Columbus, OH Green Fluorescent Protein: Properties, Applications, and Protocols, Second Edition Edited by Martin Chalfie and Steven R. Kain Copyright © 2006 John Wiley & Sons, Inc.

391

392

REASSEMBLED GFP: DETECTING PROTEIN–PROTEIN INTERACTIONS AND PROTEIN EXPRESSION PATTERNS

17.1

PROTEIN DISSECTION

Cleavage of a single peptide bond in a protein can result in protein unfolding and inactivation, because protein folding is typically a highly cooperative process. Despite the dramatic energetic consequences of covalent bond cleavage, some proteins remain folded even after being cleaved into two pieces, held together by noncovalent interactions. Moreover, a mixture of the polypeptides corresponding to the proteolytic fragments can sometimes result in reassembly of active protein (Richards, 1958; Anfinsen, 1973). This observation has been exploited to make useful biological tools. For example, independent expression of the alpha and omega fragments of b-galactosidase results in spontaneous assembly of the active enzyme, capable of turning over chromogenic galactosides such as X-gal, which is the basis of “blue/white” screening in plasmid subcloning (Ullmann et al., 1967). Fusion to ribonuclease S, arising from the noncovalent reassembly of the RNase S-peptide and S-protein resulting from cleavage of RNase A by subtilisin, can be used as an indicator of protein expression and folding using a fluorogenic substrate (Kelemen et al., 1999). The system is marketed as FRETWorks by Novagen (Madison, WI). In the cases of LacZ and RNase S, the “dissected” protein fragments spontaneously reassemble, resulting in active protein. Fields and Song (1989) realized that if a reporter protein could be split such that the dissected fragments did not spontaneously reassemble, the interaction of proteins fused to those fragments might drive the reassembly and function of the reporter. In the yeast two-hybrid (Y2H) screen, the DNA-binding and activation domains of the GAL4 transcription factor do not spontaneously reassemble, but if they are brought together by the interaction of “bait” and “prey” proteins fused to the two domains, the reassembled GAL4 can drive the transcription of a reporter like bgalactosidase. Y2H has been tremendously useful for identifying protein interaction partners (Uetz et al., 2000). However, the method has considerable limitations: It must be done in yeast, it requires nuclear importation and function (excluding whole protein classes, such as membrane-associated proteins), it does not demand a direct interaction (i.e., interactions through complexes are sometimes detected), and it can be confounded by proteins that activate transcription in the absence of a binding partner. Although Y2H allows detection of fairly weak interactions, it is hampered by abundant false positives. Several assays have been introduced to circumvent some of the problems with Y2H, including bacterial two-hybrid systems (Karimova et al., 1998; Joung et al., 2000; Hays et al., 2000) and functional interaction traps based on fusion to dissected fragments of ubiquitin (Johnsson and Varshavsky, 1994), b-galactosidase (Rossi et al., 1997), dihydrofolate reductase (Pelletier et al., 1998; Pelletier et al., 1999), or b-lactamase (Galarneau et al., 2002). The reassembly of these dissected proteins must be detected by the addition of a chromogenic, fluorogenic or chemiluminescent reagent, or through survival selection. Recently, our group introduced a screen for protein–protein interactions based on the reassembly of dissected fragments of GFP (Ghosh et al., 2000). GFP is especially attractive because no exogenous reagent must be added to detect the reassembled protein and because GFP is known to express and mature in virtually every cell type and subcellular structure in which it has been tested (as demonstrated in other chapters in this volume). Moreover, since most cells do not have significant fluorescence background at the GFP emission/excitation wavelengths, virtually all the signal can be attributed to the reassembled GFP, and subcellular localization can be directly visualized. The method has successfully been used by our group and others to detect protein–protein interactions in bacteria and eukaryotes, to identify unknown interaction partners, and to visualize cell type and subcellular protein localization in multicellular organisms.

DISSECTION AND REASSEMBLY OF GFP

Figure 17.1. Schematic of GFP dissection. (A) The original system used by Ghosh et al. (2000) split GFP at 157–158. The reassembled GFP, fused to antiparallel leucine zipper peptides (blue), is depicted with the N- and C-terminal fragments are colored green and red, respectively. (B) The dissection points discussed in the text are highlighted. Those in bold have been the most generally successful. Created with PyMOL (http://www.pymol.org) from PDB entries 1EMA and 1SER. See color insert.

17.2 17.2.1

DISSECTION AND REASSEMBLY OF GFP Variants and Topology

Our original implementation of the GFP fragment reassembly involved dissection of the sg100 GFP variant between residues 157 and 158 (Ghosh et al., 2000). Antiparallel leucine zipper peptides were fused to the C-terminus of GFP(1–157) and the N-terminus of GFP(158–238), an arrangement that was designed to allow interaction of the peptides in the reassembled complex (Fig. 17.1A). Coexpression of these peptide-fragment fusions results in GFP reassembly and cellular fluorescence. When the peptide-fragment fusions are expressed separately, or when one or both of the peptides are not fused, no reassembly occurs. Therefore, the leucine zipper peptide–peptide interaction is required for GFP reassembly.1 We and others typically call the GFP fragments “NGFP” and “CGFP,” but we will use the nomenclature GFP(1–157) and GFP(158–238) here to make it clear what spectral variant and dissection point we are referring to. See Fig. 17.1B for the dissection points discussed in this section. The reassembly reaction is not extremely sensitive to linker length. We have tested linkers of 4–15 amino acids between the GFP fragments and the zipper peptides that are compatible with reassembly (Magliery et al., 2005). Moreover, the linkers do not have to be the same length. Reassembly occurs when the linker between the GFP(158–238) and the peptide is 7 amino acids longer than the linker between the GFP(1–157) and peptide. The sg100 variant (G. J. Palm, personal communication) has the following mutations from the original gfp10 wild-type sequence: F64L, S65C, Q80R, Y151L, I167T, and 1

The plasmid vectors pET11a-link-NGFP and pMRBAD-link-CGFP, which permit facile cloning of analyte proteins and independent maintenance in E. coli, are available upon request. See the Regan Lab webpage (http://www.csb.yale.edu/people/regan/publications.html) for sequences and information, in addition to Wilson et al. (2004).

393

394

REASSEMBLED GFP: DETECTING PROTEIN–PROTEIN INTERACTIONS AND PROTEIN EXPRESSION PATTERNS

K238N. The actual construct used originally had a proline at position 151, but we later demonstrated that this had no effect on the fluorescence phenotype relative to leucine (Magliery et al., 2005). Using the same antiparallel leucine zipper fusions, Chalfie and coworkers found that CFP and YFP can also reassemble, and the CFP(158–238) can reassemble with GFP(1–157) and YFP(1–157) to yield fluorescent proteins with unique spectral properties (Zhang et al., 2004). Michnick and co-workers have used virtually the same dissection point (158–159) and topology in EGFP, an F64L, S65T GFP mutant codonoptimized for mammalian cells, adding 10-amino-acid (GGGGS)2 linkers between the fused proteins and the dissection point (Remy and Michnick, 2004a). Other points of dissection have also been explored. Tsien and co-workers showed that large insertions and circular permutation were possible in EYFP between residues 144– 145 (Baird et al., 1999). Based on this, Miyawaki and colleagues fused calmodulin to the N-terminus of EYFP(1–144) and fused the tightly interacting M13 peptide to the Nterminus of EYFP(145–238), and they observed reassembly in HeLa cells (Nagai et al., 2001). This is both a different point of dissection and a different fusion topology, especially notable because it is at the end of the b-barrel distal to the native termini. Kerppola and co-workers tested various dissection points in EYFP (S65G S72A T203Y) using yet another topology and parallel leucine zippers from Fos and Jun (Hu et al., 2002). Here, both zipper peptides were fused to the C-termini of the two YFP fragments with short 5to 7-amino-acid linkers between the fragments and the zipper peptides. Loops at both ends of the b-barrel, proximal and distal to the C-terminus, were selected for dissection (38–39, 101–102, 144–145, 154–155, 168–169, 172–173, and 192–193). The 154–155 dissection point, in the same loop used by our group, gave the best results. Hu and Kerppola (2003) later tested fragments derived from EYFP, EGFP, ECFP, and EBFP dissected at 154–155 and 172–173 using the same fusion topology as before to the C-termini of both fragments. Surprisingly, the GFP fragments split at 154–155 did not reassemble, but YFP(1–154) did reassemble with CFP(155–238) to give a reassembled protein with spectral characteristics distinct from YFP or CFP. At a second dissection point at the opposite end of the b-barrel (172–173), it was found that YFP(173–238) reassembled with GFP(1–172), CFP(1–172) and YFP(1–172), but that B/C/GFP(173–238) did not reassemble with any 1–172 fragment. Perhaps even more remarkably, the 1–172 fragments of GFP, CFP, and YFP reassembled with the 155–238 fragments of CFP and YFP despite the resulting duplication of a b-strand (Fig. 17.2). Moreover, BFP(1–172) and CFP(155–238) also reassembled, even though no reassembly occurred with B/C/ GFP(173–238). Since the extra “strand” is likely just an unstructured linker between the N-terminal fluorescent protein fragment and the zipper peptide, the mixed dissection-point data actually suggest that a longer linker between the peptides and fragments is beneficial with this topology. Umezawa and colleagues developed a different means of reassembling GFP. Specifically, EGFP fragments are covalently reassembled when fused interacting proteins drive the association and splicing of an intein (Ozawa et al., 2000). Originally, the N-terminal domain of the yeast VMA1 intein was sandwiched between EGFP(1–128) and one analyte protein, and the C-terminal intein domain was sandwiched between the second analyte protein and EGFP(129–238). The analyte proteins (originally calmodulin and M13 peptide) were separated from the intein fragments by 9- to 10-amino-acid linkers. To achieve efficient splicing, the EGFP sequence between I124 and I129 was altered from IEKKGI to IILKGC, resulting in weak cellular fluorescence with the CaM/M13 fusions. To improve the efficiency of the system, the Umezawa group later replaced the VMA1 intein with the smaller, more soluble, bacterial dnaE intein, and other EGFP dissection

DISSECTION AND REASSEMBLY OF GFP

Figure 17.2. Multicolor reassembly of fluorescent proteins. Reassembly of CFP(155–238) with (A) YFP(1–172), (B) GFP(1–172), (C) BFP(1–172), and (D) CFP(1–172) results in yellow, green, blue, and cyan cells. [Adapted from Hu and Kerppola (2003) with permission.] See color insert.

points were examined (Ozawa et al., 2001). Dissection at residues 144–145 or 224–225 gave poor fluorescence, but dissection at 157–158 was successful, with either K156Y Q157C mutations or a KFAEYC insertion after Q157. Recently, Cabantous et al. (2005) dissected the so-called “superfolder” variant of GFP in the last loop, at residue 214. Both the GFP(1–214) and GFP(214–230) fragments were optimized by directed evolution for enhanced fluorescence and solubility. These optimized GFP fragments spontaneously reassemble, resulting in chromophore maturation and cellular fluorescence upon their co-expression without the mediation of fused, interacting proteins. These GFP tags are therefore useful for detecting protein expression and solubility in cells or lysates, but they are not useful for detecting protein-protein interactions. Alteration of the unfused termini of GFP sometimes prevents the reassembly reaction. In our implementation in which the fusions are made at the point of dissection, we found that N-terminal hexahistidine tagging of the GFP(1–157) fragment is not detrimental, but that tagging the C-terminus of GFP(158–238) with a biotinylation sequence and His6-tag prevented reassembly. The Kerppola group fused interacting proteins at the C-termini of both fluorescent protein fragments and tagged both N-termini with His6. Because the N- and C-termini of GFP are close in space at the same “end” of the barrel as the 157–158 loop, the spatially near termini might affect the reassembly reaction. Further experimentation is needed to see if modification of the termini or even a random mutagenic approach, as Stemmer took to engineer GFPuv (Crameri et al., 1996), can improve the reassembly. Therefore, several GFP variants are amenable to dissection and reassembly, and several points of dissection are useful. The most generally useful point of dissection appears to be in a surface loop near residues 157–158, spatially near the N- and C-termini, regardless of fusion topology. The striking fact that different fusion topologies and a wide range of linker lengths lead to reassembly is useful, since it does not require extensive

395

396

REASSEMBLED GFP: DETECTING PROTEIN–PROTEIN INTERACTIONS AND PROTEIN EXPRESSION PATTERNS

optimization for each protein pair. It suggests that precise alignment of the GFP fragments is not necessary to nucleate reassembly and that the persistent interaction of the fused proteins is not necessary to maintain the reassembled complex.

17.2.2

Kinetics and Mechanism

For the reassembly of GFP to be useful as a screen for protein–protein interactions, three key mechanistic parameters need to be determined: the affinity required to drive reassembly, the time required to observe cellular fluorescence, and the reversibility of the reassembly reaction. We addressed the first of these, the minimum affinity required for reassembly, by screening a library of antiparallel leucine zipper mutants with different affinities (Magliery et al., 2005). The reassembly was found to be remarkably sensitive, resulting in fluorescent cells from peptide–peptide interactions with dissociation constants as weak as 500 mM to 1 mM. In addition to peptide–peptide interactions, protein–peptide interactions in the range of 500 mM result in weak cellular fluorescence, but clearly more than negative controls. Moreover, we found that peptides with different affinities for a given peptide-binding domain can be distinguished by the level of cellular fluorescence. In Umezawa’s intein-based implementation, association of the intein fragments is required for GFP reassembly, but no information is available about the affinity required for this interaction. In addition, the use of these polypeptides results in considerable background fluorescence even in the absence of fused interacting proteins [a feature that has been exploited to examine mitochondrial localization (vide infra)]. In contrast, no cellular fluorescence is detected for negative controls in our implementation of the screen, and little to no fluorescence results from the various unfused fragment pairs tested by Kerppola (Magliery et al., 2005; Hu and Kerppola, 2003). The acquisition of cellular fluorescence is quite slow. Intact GFP alone usually requires several hours to mature in the cell, with rate-limiting oxidation of the chromophore (Tsien, 1998). We see optimum cellular fluorescence in E. coli with fused interacting proteins after overnight growth at 30°C followed by 1–2 days at room temperature, or with about 3 days all at room temperature. Kerppola and co-workers observe cellular fluorescence in COS-1 cells 8 h after transfection, but they show data from cells 36–48 h after transfection (Hu et al., 2002). More recently, they report cell growth at 37°C for 24 h followed by 30°C for 0–24 h (Hu and Kerppola, 2003). Michnick also reports screening COS-1 cells by FACS 48 h after transfection (Remy and Michnick, 2004a). Chalfie and co-workers see fluorescence in embryos and newly hatched nematode larvae, which probably corresponds to less than 8 h of expression time at 20°C (Zhang et al., 2004). Kerppola and co-workers estimated the t1/2 for the reassembly reaction (i.e., folding into the GFP conformation after association of the fused proteins), which is a pseudo-first-order process, to be about 60 s (Hu et al., 2002). (Rate-limiting chromophore maturation followed reassembly, in the proposed kinetic scheme.) This analysis is complicated by the fact that the individual, fused GFP fragments are almost entirely insoluble. Thus, the rate was estimated from dilution out of 6 M guanidine, and it is difficult to know the initial concentration, since presumably most of the protein precipitated upon dilution. The improved, dnaE intein-mediated version of the system results in rapid acquisition of cellular fluorescence. Fluorescence can be detected in the lysate 4 h after IPTG induction, and bacteria are typically grown 12–16 h on agar before observation, presumably in part to achieve sufficient colony size (Ozawa et al., 2001). The GFP reassembly reaction is essentially irreversible in vitro (Magliery et al., 2005). We purified fluorescent, soluble, reassembled GFP complex His6 tagged at the N-terminus

DISSECTION AND REASSEMBLY OF GFP

of GFP(1–157) over NiNTA-agarose. The urea-induced denaturation of the complex is slow and irreversible, with an estimated t1/2 in aqueous buffer of almost 10 years (when one of the peptides is pre-cleaved from the purified complex with protease). However, Chalfie and co-workers suggest that the reassembled GFP may have a much shorter half-life in C. elegans than GFP itself (Zhang et al., 2004). The irreversibility of the reassembly reaction could explain why it is possible to detect such weak interactions, and presumably would allow detection of transient interactions as well. The intein-mediated formation of EGFP is also obviously irreversible. Some caution must be exercised in applying the reassembly of GFP to investigating protein–protein interactions. Association of the fused proteins (nucleation), the critical first step of reassembly, is affected by the solubility and expression level of the fused GFP fragments. Thus, GFP reassembly can distinguish the binding of cognate from noncognate peptide to a particular peptide-binding domain, but the absolute amount of fluorescence does not correspond to differences in protein-peptide affinity between different peptidebinding domains (Magliery et al., 2005). Presumably, this lack of correspondence is due, at least in part, to different expression levels and solubilities of the GFP fragment fusions. Interrogation of a library, where the expression and solubility properties of the fusions might vary considerably, must therefore especially be scrutinized for the possibility of false negatives (i.e., authentic interactions that do not lead to cellular fluorescence). The tolerance of the reassembly reaction to topology and linker length suggests that precise positioning of the GFP fragments is not required; the most important factor appears to be effective concentration. Moreover, the interaction of the fused analyte proteins is not necessary to maintain the complex, since fluorescence persists after cleavage of one of the fused proteins (Magliery et al., 2005). However, we have also shown by CD spectroscopy that the antiparallel leucine zippers on which we originally tested the screen do associate in the reassembled complex and increase its already considerable kinetic stability. This tolerance is critical for the function of the screen, since many false negatives would result if exact positioning were required for either reassembly or maintenance of the reassembled complex.

17.2.3

Scope: Proteins and Cells

A variety of different peptides and proteins have been used successfully with GFP reassembly. The artificial antiparallel leucine zipper used with our approach and the parallel leucine zipper from the bZIP proteins Fos and Jun used by Kerppola work very well (Ghosh et al., 2000; Hu et al., 2002; Magliery et al., 2005). Fusion of TPR peptide-binding domains (see below) to the N-terminus of GFP(158–238), as well as fusion of peptide ligands to the C-terminus of GFP(1–157), was also successful (Magliery et al., 2005). Due to the insolubility of the GFP fragments, we are more wary of applications in which proteins are fused to the C-termini of the fragments, since the proteins almost certainly have to re-fold to initiate reassembly. However, several successes have been reported, including (a) fulllength Fos and Jun, and domains of Rel family proteins NF-kB and IkBa, fused to the Ctermini of YFP fragments split at 154–155 (Hu et al., 2002) and (b) fusion of proteins from a cDNA library to the C-terminus of EGFP(1–158) against N-terminally fused PNK/Akt to EGFP(159–239) (Remy and Michnick, 2004a). The intein-mediated system has been used exclusively with calmodulin and its ligand M13 peptide, although in both fusion orientations (Ozawa et al., 2001). The reassembly reaction has been tested extensively in E. coli, but it also functions well in mammalian COS-1 cells, both cytoplasmically and in the nucleus. It has been used

397

REASSEMBLED GFP: DETECTING PROTEIN–PROTEIN INTERACTIONS AND PROTEIN EXPRESSION PATTERNS

successfully in mammalian NIH3T3, HeLa, and HEK293T cells. Chalfie and co-workers have used the system successfully in the nematode C. elegans, with localization to the specific sets of neurons, as well as with subcellular localization to nuclei and presynaptic vesicles. The intein-mediated reassembly has been tested in E. coli, and has been used slightly differently (see below) in mammalian BNL1ME mitochondria.

17.3

APPLICATIONS OF THE GFP REASSEMBLY

17.3.1

Understanding Protein–Protein Interactions

Identification and characterization of protein–protein interactions from libraries are the most obvious and perhaps ambitious applications of this system. Two key benefits of the screen relative to other two-hybrid methods are that the screen can be carried out, in principle, in any organism or subcellular compartment and that no exogenous reagent is necessary to detect the reporter. To date, the reassembly has been used to detect or discriminate protein–protein interactions in bacteria and mammalian cells. We have used GFP reassembly to explore the structural determinants of antiparallel leucine zipper formation. Coiled coils are generally stabilized by the burial of hydrophobic residues and the interaction of oppositely charged “edge” residues (O’Shea et al., 1993). What controls the orientation (parallel versus antiparallel) is less clear, but favorable edge interactions and buried hydrogen bonds can be used to favor the antiparallel orientation (Oakley and Hollenbeck, 2001). We constructed a library in which the eight edge positions on one peptide were randomized between Glu and Lys, which would formally vary from 0 to 8 charge–charge mismatches in the antiparallel orientation with a constant peptide (Magliery et al., 2005) (Fig. 17.3). Only peptides with three or fewer charge-charge mismatches passed the screen, and the positions of the mismatches were not equivalent. Specifically, mismatches near the ends of the zipper were less significant. Comparison of

E5 E12 E19 E26

(A) K27 K20 K13 K6

L24 L17 L10 L3

(B)

L2 N9 L16 L23

c' f'

c

Z-NGFP b'

Z-CGFP

f

L7 L14 N21 L28

b E4 E11 E18 E25

L27 L20 L13 L6

K24 K17 K10 K3

Number of clones (scaled to N=100)

398

40 Positive Negative Normal

30

20

10

0

1

2 3 4 5 6 Number of mismatches

7

8

Figure 17.3. A library of antiparallel leucine zipper interactions. (A) A helical wheel diagram of the library, in which the e and g positions in one peptide (boxed) were randomized between Lys and Glu. (B) Positive clones and three or fewer mutation, while negative have three or more. The distributions are shown in contrast to the distribution that would have resulted if the screen did not select for tight binders. [Adapted from Magliery et al. (2005).]

APPLICATIONS OF THE GFP REASSEMBLY

the screening data and in vitro biophysical data suggested that binding in the parallel orientation was not possible in our model system even with favorable charge pairings. Tetratricopeptide repeat (TPR) domains are composed of sequentially arrayed 34amino-acid motifs, typically in groups of three, that sometimes act as a peptide-binding interface (D’Andrea and Regan, 2003). Hsp-organizing protein (HOP) has three TPR domains (TPR1, TPR2A, and TPR2B), the first two of which are known to bind to the Ctermini of chaperones Hsc70 and Hsp90 (Brinker et al., 2002). (HOP-TPR2B has no known binding partner.) When challenged with C-terminal peptides from Hsc70 and Hsp90, as well as an unrelated leucine zipper peptide, the TPR domains could distinguish their related cognate from noncognate ligands and could discriminate both from the unrelated negative control (Magliery et al., 2005) (Fig. 17.4). This type of experiment may be a route to identifying the targets of TPR2B and other TPR domains without known binding partners, using either random or cDNA-based peptide libraries (see below). The use of multicolor fluorescent proteins makes it possible to examine multiple interactions in a cell at one time, or to compare the relative efficiency of dimerization with a given protein to multiple interaction partners. For example, the relative efficiencies of dimerization of the leucine zipper peptide from Jun with the peptides from Fos and ATF2 were compared (Hu and Kerppola, 2003). Coexpression of YFP(1–154)-bFos, CFP(1–173)-bATF2, and CFP(155–238)-bJun resulted in yellow fluorescence in the nucleoli, consistent with preferential bFos/bJun dimerization. More recently, the competition for in vivo dimerization of Myc/Mad/Max proteins was examined (Grinberg et al., 2004),

Figure 17.4. TPR domains and peptide ligands. (A) Schematic of interactions between the TPR domains of HOP and the chaperones Hsc70 and Hsp90. (B) Screening of the three TPR domains from HOP against peptides from the C-terminus of Hsc70 and Hsp90, and an unrelated peptide (Z). The cognate TPR-peptide interactions result in stronger fluorescence than the noncognate interactions. [From Magliery et al. (2005).]

399

400

REASSEMBLED GFP: DETECTING PROTEIN–PROTEIN INTERACTIONS AND PROTEIN EXPRESSION PATTERNS

and the combinatorial interaction of crystallin-regulating factors in lens development was studied (Rajaram and Kerppola, 2004).

17.3.2

Identifying Unknown Protein–Protein Interactions

Remy and Michnick searched for binding partners of the protein kinase PKB/Akt by challenging it with a human brain cDNA-derived library fused to an EGFP fragment (Remy and Michnick, 2004a). Screening was done by FACS in transiently transfected mammalian COS-1 cells. The positive clones, after two rounds of selection, contained a substantial number that did not produce readable sequence or were false positives for various reasons, but 22 out of 100 contained identifiable genes. One of these proteins, a human homolog of mouse Ft1, was shown to interact in a serum- and insulin-dependent manner, consistent with participation in the PI3K signal transduction pathway. The direct interaction of hFT1 and PKB was later established by immunoprecipitation, and a role in increasing the activity of PKB was established, probably involving apoptosis regulation (Remy and Michnick, 2004b). Despite the relatively high rate of false positives, along with the potential for false negatives from fusion-insolubility or refolding problems, this is an exciting demonstration of the general applicability of the screen. Problems may arise, both in the discovery of new ligands and in the characterization competing interactions, because unfused, endogenous proteins may compete with the GFP fragment-fused analyte protein. For example, quantification of Ca2+ concentration was not reliable with fusion of calmodulin and M13 peptide to dissected GFP fragments, because of competition with endogenous calmodulin (Nagai et al., 2001). Genetically tractable systems like E. coli, yeast, Drosophila, and C. elegans allow replacement of the gene of interest and testing for complementation by the fused protein, which may be a general solution to this problem. This technology should be readily extended to the discovery of inhibitors and activators of protein–protein interactions. Because the GFP reassembly reaction is essentially irreversible, one cannot expect to disrupt the interaction of proteins fused to GFP fragments by adding inhibitors. However, if inhibitors are added at the inception of cell growth, they might prevent the reassembly reaction altogether. This approach is likely to provide a very stringent screen for inhibitors. On the other hand, the irreversibility of the reassembly reaction makes it particularly suited to the discovery of interaction agonists, or so-called chemical inducers of dimerization (CIDs). Similarly, the ability to screen in eukaryotic cells raises the possibility of identifying interactions requiring posttranslational modifications, perhaps in an environment- or stimulus-dependent manner.

17.3.3

Detection of Subcellular Localization

Since GFP fluorophore maturation is an autocatalytic process dependent upon protein folding but not accessory proteins or environment (except that folding and oxidation must be possible), GFP reassembly can be used to detect the subcellular localization of interacting proteins. Specifically, this means that protein–protein interactions can be examined in their native context, away from potentially complicating interactions that arise from expression in other compartments (like the nucleus). For example, Kerppola demonstrated in mammalian COS-1 cells that the leucine zippers of Fos and Jun alone localize to the nucleoli, but that the full-length bZIP dimers localize to the nucleoplasmic region and are excluded from the nucleoli (Hu et al., 2002) (Fig. 17.5). Similarly, the ATF2/Jun dimer

APPLICATIONS OF THE GFP REASSEMBLY

Figure 17.5. Subcellular localization of protein–protein interactions. The leucine zipper domains from Fos and Jun localize to the nucleoli of COS-1 cells (A), while the full-length proteins localize to the nucleoplasm but are excluded from the nucleoli (B). ATF2 and Jun localize to the perinuclear region (C), but are translocated to the nucleus when p38 is overexpressed (D). [Adapted from Hu et al. (2002) with permission.]

localizes to the perinuclear region, but is translocated into the nucleus when p38 is overexpressed. Umezawa and co-workers have adapted their intein-based GFP reassembly for identification of mitochondrial proteins (Ozawa et al., 2003). Essentially, the two-component system is used as a logical “AND” gate (Zhang et al., 2004), such that cells are fluorescent if and only if two different proteins are both localized to the mitochondria. The Cterminal intein/EGFP fragment was fused to a mitochondrial targeting signal (MTS), and the N-terminal fragment was fused to proteins from a cDNA library. If the proteins in the cDNA library contain an MTS of their own, the two intein/EGFP fragments are colocalized to the mitochondria, resulting in splicing and EGFP folding and maturation. Seventy different clones were identified, including known mitochondrial proteins, known proteins with no previous information about subcellular localization, and proteins of unknown function. It should be noted that splicing occurs here because of background intein selfassociation. This limits the utility of this version of the screen in detecting protein–protein interactions, but it is perfectly usable for detecting expression patterns.

17.3.4

Visualizing Protein Expression Patterns

Chalfie and co-workers have exploited the combinatorial features of the reassembly of GFP to detect cellular and subcellular expression patterns in whole animals (C. elegans)

401

402

REASSEMBLED GFP: DETECTING PROTEIN–PROTEIN INTERACTIONS AND PROTEIN EXPRESSION PATTERNS

Figure 17.6. Cellular colocalization in whole organisms. (A) Punc-24gfp is expressed in many adult cells. (B) Punc-24nzgfp and Pmec-2czgfp are coexpressed only in six touch receptor neurons. (C) Pmec-3nzgfp and Pegl-44czgfp are coexpressed only in the two FLP neurons. [From Zhang et al. (2004) with permission.]

(Zhang et al., 2004). Specifically, by driving the expression of the two GFP-leucine zipper fusions from two separate promoters, it was possible to (1) establish gene coexpression in the same cells, (2) label cells that uniquely coexpress genes from both promoters, and (3) determine the expression pattern of one promoter when the expression of the other promoter is known. For example, there is no known promoter that is only active in the two FLP neurons of the nematode. However, mec-3 and egl-44, which are expressed in various cell types, are uniquely coexpressed in FLP neurons. Expression of the GFP fragments from these two promoters resulted in unique labeling of the FLP cells (Fig. 17.6). The combinatorial aspect of GFP reassembly frees the geneticist from the limitations of the expression patterns of available promoters. The authors also showed that fusion of one of the fragments to a nuclear localization signal or synaptobrevin, which localizes to presynaptic vesicles, could be used to label subcellular components only in cells in which both fragments’ promoters are active.

17.4

FUTURE DIRECTIONS

GFP reassembly can be used to identify protein binding partners in bacteria and mammalian cells, but the potential of the system has not been stretched by the early studies described here. In particular, the identification of interactions that requisitely occur in a particular cell type or subcellular structure should be possible. Similarly, the identification of interactions requiring cell-specific modifications is not difficult to imagine. Adaptation of the GFP reassembly to yeast, in which virtually every genetic knockout is available,

REFERENCES

may be particularly useful in this regard. Also, the irreversibility of the reassembly suggests that it may be useful in capturing transiently interacting proteins, although the rate of degradation may limit this approach in some kinds of cells. A major area of interest in developing small-molecule therapeutics is the identification of inhibitors and activators of protein–protein interactions. Although mechanistic complications arise in adapting the screen for this purpose, they are likely to be surmountable. Again, the finding that the reassembly can be carried out in mammalian cells or whole organisms is particularly attractive for discovering therapeutics. Finally, understanding the expression and coexpression patterns of proteins is a major advance for studying organismal development and cellular differentiation. Such localization may be useful, for example, in the precise characterization of cancer cell lines and will clearly be useful for researchers wishing to study particular a cell type during animal development. GFP has already proven useful for (a) understanding cellular and subcellular expression patterns with fusions of GFP to proteins of interest (Chalfie et al., 1994), (b) visualizing protein–protein interactions with fused FRET pairs of GFP variants (Miyawaki et al., 1997), and (c) establishing protein expression or solubility by direct fusion of GFP to the C-terminus of the analyte protein (Waldo et al., 1999). The use of reassembled GFP not only remedies some of the technical complications with these elegant methods, but also allows uses that are not possible with intact GFP.

ACKNOWLEDGMENTS T.J.M. is an N.I.H. Postdoctoral Fellow (GM065750). Work on the reassembled GFP-based protein interaction trap was supported in part by NIH grants GM62413 and GM57265 (L.R.).

REFERENCES Anfinsen, C. B. (1973). Principles that govern the folding of protein chains. Science 181:223–230. Baird, G. S., Zacharias, D. A., and Tsien, R. Y. (1999). Circular permutation and receptor insertion within green fluorescent proteins. Proc. Natl. Acad. Sci. USA 96:11241–11246. Brinker, A., Scheufler, C., Von Der Mulbe, F., Fleckenstein, B., Herrmann, C., Jung, G., Moarefi, I., and Hartl, F. U. (2002). Ligand discrimination by TPR domains. Relevance and selectivity of EEVD-recognition in Hsp70 ¥ Hop ¥ Hsp90 complexes. J. Biol. Chem. 277:19265–19275. Cabantous, S., Terwilliger, T. C., and Waldo, G. S. (2005). Protein tagging and detection with engineered self-assembling fragments of green fluorescent protein. Nat. Biotechnol. 23:102–107. 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. 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. D’Andrea, L., and Regan, L. (2003). TPR proteins: The versatile helix. Trends Biochem. Sci. 28:655–662. Fields, S., and Song, O. (1989). A novel genetic system to detect protein–protein interactions. Nature 340:245–246. Galarneau, A., Primeau, M., Trudeau, L. E., and Michnick, S. W. (2002). beta-Lactamase protein fragment complementation assays as in vivo and in vitro sensors of protein–protein interactions. Nat. Biotechnol. 20:619–622.

403

404

REASSEMBLED GFP: DETECTING PROTEIN–PROTEIN INTERACTIONS AND PROTEIN EXPRESSION PATTERNS

Ghosh, I., Hamilton, A. D., and Regan, L. (2000). Antiparallel leucine zipper-directed protein reassembly: Application to the green fluorescent protein. J. Am. Chem. Soc. 122:5658–5659. Grinberg, A. V., Hu, C. D., and Kerppola, T. K. (2004). Visualization of Myc/Max/Mad family dimers and the competition for dimerization in living cells. Mol. Cell. Biol. 24:4294–4308. Hays, L. B., Chen, Y. S., and Hu, J. C. (2000). Two-hybrid system for characterization of protein–protein interactions in E. coli. Biotechniques 29:288–296. Hu, C. D., and Kerppola, T. K. (2003). Simultaneous visualization of multiple protein interactions in living cells using multicolor fluorescence complementation analysis. Nat. Biotechnol. 21:539–545. Hu, C. D., Chinenov, Y., and Kerppola, T. K. (2002). Visualization of interactions among bZip and Rel family proteins in living cells using bimolecular fluorescence complementation. Mol. Cell 9:789–798. Johnsson, N., and Varshavsky, A. (1994). Split ubiquitin as a sensor of protein interactions in vivo. Proc. Natl. Acad. Sci. USA 91:10340–10344. Joung, J. K., Ramm, E. I., and Pabo, C. O. (2000). A bacterial two-hybrid selection system for studying protein–DNA and protein–protein interactions. Proc. Natl. Acad. Sci. USA 97:7382– 7387. Karimova, G., Pidoux, J., Ullmann, A., and Ladant, D. (1998). A bacterial two-hybrid system based on a reconstituted signal transduction pathway. Proc. Natl. Acad. Sci. USA 95:5752–5756. Kelemen, B. R., Klink, T. A., Behlke, M. A., Eubanks, S. R., Leland, P. A., and Raines, R. T. (1999). Hypersensitive substrate for ribonucleases. Nucleic Acids Res. 27:3696–3701. Magliery, T. J., Wilson, C. G. M., Pan, W., Mishler, D., Ghosh, I., Hamilton, A. D., and Regan, L. (2005). Detecting protein–protein interactions with a GFP-fragment reassembly trap: Scope and mechanism. J. Am. Chem. Soc. 127:146–157. Miyawaki, A., Llopis, J., Heim, R., McCaffery, J. M., Adams, J. A., Ikura, M., and Tsien, R. Y. (1997). Fluorescent indicators for Ca2+ based on green fluorescent proteins and calmodulin. Nature 388:882–887. Nagai, T., Sawano, A., Park, E. S., and Miyawaki, A. (2001). Circularly permuted green fluorescent proteins engineered to sense Ca2+. Proc. Natl. Acad. Sci. USA 98:3197–3202. Oakley, M. G., and Hollenbeck, J. J. (2001). The design of antiparallel coiled coils. Curr. Opin. Struct. Biol. 11:450–457. O’Shea, E. K., Lumb, K. J., and Kim, P. S. (1993). Peptide velcro—Design of a heterodimeric coiled coil. Curr. Biol. 3:658–667. Ozawa, T., Nogami, S., Sato, M., Ohya, Y., and Umezawa, Y. (2000). A fluorescent indicator for detecting protein–protein interactions in vivo based on protein splicing. Anal. Chem. 72:5151–5157. Ozawa, T., Takeuchi, T. M., Kaihara, A., Sato, M., and Umezawa, Y. (2001). Protein splicing-based reconstitution of split green fluorescent protein for monitoring protein–protein interactions in bacteria: Improved sensitivity and reduced screening time. Anal. Chem. 73:5866–5874. Ozawa, T., Sako, Y., Sato, M., Kitamura, T., and Umezawa, Y. (2003). A genetic approach to identifying mitochondrial proteins. Nat. Biotechnol. 21:287–293. Pelletier, J. N., Campbell-Valois, F. X., and Michnick, S. W. (1998). Oligomerization domaindirected reassembly of active dihydrofolate reductase from rationally designed fragments. Proc. Natl. Acad. Sci. USA 95:12141–12146. Pelletier, J. N., Arndt, K. M., Pluckthun, A., and Michnick, S. W. (1999). An in vivo library-versuslibrary selection of optimized protein-protein interactions. Nat. Biotechnol. 17:683–690. Puig, O., Caspary, F., Rigaut, G., Rutz, B., Bouveret, E., Bragado-Nilsson, E., Wilm, M., and Seraphin, B. (2001). The tandem affinity purification (TAP) method: A general procedure of protein complex purification. Methods 24:218–229.

REFERENCES

Rajaram, N., and Kerppola, T. K. (2004). Synergistic transcription activiation by Maf and Sox and their subnuclear localization are disrupted by a mutation in Maf that causes cataracts. Mol. Cell. Biol. 24:5694–5709. Remy, I., and Michnick, S. W. (2004a). A cDNA library functional screening strategy based on fluorescent protein complementation assays to identify novel components of signaling pathways. Methods 32:381–388. Remy, I., and Michnick, S. W. (2004b). Regulation of apoptosis by the Ft1 protein, a new modulator of protein kinase B/Akt. Mol. Cell. Biol. 24:1493–1504. Richards, F. M. (1958). On the enzymatic activity of subtilisin-modified ribonuclease. Proc. Natl. Acad. Sci. USA 44:162–166. Rossi, F., Charlton, C. A., and Blau, H. M. (1997). Monitoring protein–protein interactions in intact eukaryotic cells by beta-galactosidase complementation. Proc. Natl. Acad. Sci. USA 94:8405–8410. Tsien, R. Y. (1998). The green fluorescent protein. Annu. Rev. Biochem. 67:509–544. Uetz, P., Giot, L., Cagney, G., Mansfield, T. A., Judson, R. S., Knight, J. R., Lockshon, D., Narayan, V., Srinivasan, M., Pochart, P., Qureshi-Emili, A., Li, Y., Godwin, B., Conover, D., Kalbfleisch, T., Vijayadamodar, G., Yang, M., Johnston, M., Fields, S., and Rothberg, J. M. (2000) A comprehensive analysis of protein-protein interactions in Saccharomyces cerevisiae. Nature 403:623–627. Ullmann, A., Jacob, F., and Monod, J. (1967). Characterization by in vitro complementation of a peptide corresponding to an operator-proximal segment of the beta-galactosidase structural gene of Escherichia coli. J. Mol. Biol. 24:339–343. Waldo, G. S., Standish, B. M., Berendzen, J., and Terwilliger, T. C. (1999). Rapid protein-folding assay using green fluorescent protein. Nat. Biotechnol. 17:691–695. Wilson, C. G. M., Magliery, T. J., and Regan, L. (2004) Detecting protein–protein interactions with GFP-fragment reassembly. Nat. Methods 1:255–262. Zhang, S., Ma, C., and Chalfie, M. (2004). Combinatorial marking of cells and organelles with reconstituted fluorescent proteins. Cell 119:137–144. Zhu, H., Bilgin, M., and Snyder, M. (2003). Proteomics. Annu. Rev. Biochem. 72:783–812.

405

METHODS AND PROTOCOLS Steven R. Kain Agileut Technologies, Palo Alto, CA

Since the seminal work by Prasher et al. (1992) and Chalfie et al. (1994) to clone and express A. victoria GFP, this magnificent protein has moved from the spotlight of journal covers and article titles to join the ranks of a “reagent” in the methods section of most published material. As an editor of this text and a pioneer in the field of GFP application development, I feel that the more humble position now held by GFP is both sad and momentous. On the one hand, we miss the glory days of GFP experimentation, when reports of the use of GFP in any new species of cell or organism were newsworthy. Then again, the fact that GFP is now peripheral to most studies means that this biological marker has not only arrived, but is firmly established as a means to investigate fundamental questions in science. At the time of the first edition of this text, researchers were scrambling to understand how to use GFP. Methodological questions arose such as: Which GFP should I use? Will it work in my organism? How do I express the protein? How do I detect the signal? These questions have largely been answered through the more than 10,000 papers published concerning the application of GFP and its variants. Therefore, we have elected to include from the first volume only the broadest information, methods, and protocols that should be relevant for all applications that employ these biological markers.

PROTOCOL I: EXPRESSION OF GREEN FLUORESCENT PROTEIN I.C

Toxicity Due to GFP Expression

There have been published reports that overexpression of GFP may be toxic or interfere with regeneration required to generate transgenic plants (Haseloff and Amos, 1995; Chiu et al., 1996), as well as unpublished reports of toxicity associated with the overexpression of wild-type GFP in bacteria. Greatly overexpressed EGFP in bacteria—for example, from pUC-based vectors—can cause slower growth rates and osmosensitivity (Valdivia, Cormack, and Falkow, personal communication). Some bacterial species appear to tolerate high levels of GFP better than others. For example, high levels of GFP are tolerated by Yersinia species (Valdivia and Falkow, 1996) but not by Salmonella or Anabaena species (Valdivia and Falkow, 1996) (Buikema and Haselkorn, unpublished results). Toxicity due to high levels of GFP may also explain the difficulty that some workers have

Green Fluorescent Protein: Properties, Applications, and Protocols, Second Edition Edited by Martin Chalfie and Steven R. Kain Copyright © 2006 John Wiley & Sons, Inc.

407

408

METHODS AND PROTOCOLS

had in obtaining stably transfected mammalian cell lines. The problems that have been encountered with toxic effects associated with GFP overexpression may be a general problem associated with protein overexpression, rather than specifically due to GFP.

I.D

Folding and Temperature Sensitivity

The ability of GFP to absorb blue light and emit green light is believed to depend on the formation of a chromophore by cyclization and oxidation of S-Y-G (residues 65–67) (Heim et al., 1994; Cody et al., 1993). The time constant of chromophore formation has been reported to be about 2–4 h for wild type and 0.45 h for S65T mutant GFP (Heim et al., 1994, 1995). The rate-limiting formation of the chromophore may limit the ability to visualize GFP fluorescence in transfected cells or transgenic organisms until a defined time after its expression. For Drosophila, the time required for the appearance of GFP fluorescence has been reported to be 3–5 h following expression (Davis et al., 1995). Wild-type GFP has been reported to be sensitive to temperature when expressed in mammalian cells, producing a brighter fluorescent signal at 33°C compared to 37°C (Pines, 1995). Temperature sensitivity of the fluorescence of a wild-type GFP fusion protein has also been observed in studies utilizing a GFP–human glucocorticooid receptor (GFP–hGR) fusion construct. These studies showed efficient transactivation of the mouse mammary tumor virus promoter in the presence of dexamethasone at 30°C but not at 37°C (Ogawa et al., 1995), a result demonstrating that the activity of the GFP–fusion protein, as well as its fluorescence, is greatly reduced at higher temperature. These effects may be due to the folding or redox state of GFP in the cell. The studies also showed that cells fluorescing at 30°C continued to fluoresce for at least 48 h upon shifting to 37°C. The time course of GFP–hGR movement from the cytoplasm into the nucleus after induction could be determined by addition of hormone to cells grown at 30°C, followed by incubation for various time periods at 37°C. Studies in yeast have shown that wild-type GFP and a wild-type GFP fusion protein expressed in S. cerevisiae showed markedly reduced fluorescence when cells were grown at ⭓30°C, and that fluorescent cells grown at lower temperature retained their fluorescence after a shift to higher temperature (Lim et al., 1995). These observations allowed the workers to monitor relocalization of a GFP–nucleoplasmin fusion protein in a temperature-sensitive mutant of the nucleoporin gene by first culturing cells at 23°C to allow the fusion protein to accumulate, then shifting to 35°C. These studies illustrate the usefulness of the temperature sensitivity of wild-type GFP given appropriate experimental design. Such temperature sensitivity has not been reported for mutant forms of GFP expressed in mammalian cells. It should be noted, however, that the autocatalytic folding process may be less efficient under certain conditions, and that optimized expression protocols are likely to continue to be developed over the next few years.

I.E Purification of GFP The original purification of Aequorea GFP from photogenic organs of the jellyfish (Morise et al., 1974) is described in Chapter 1. The cloning of the gfp gene (Prasher et al., 1992) has permitted expression of GFP in bacteria for biochemical and biphysical studies. Bacterially expressed GFP has been characterized in clarified induced cell lysates without further purification (Heim et al., 1994) or following purification on a Ni-affinity column of a His6-tagged GFP containing a 34-residue peptide with six contiguous His residues

PROTOCOL I: EXPRESSION OF GREEN FLUORESCENT PROTEIN

fused to the N-terminus of GFP (Inouye and Tsuji, 1994). A fusion of GFP to GST (glutathione S-transferase) has also been made and purified by glutathione affinity chromatography (Niswender et al., 1995). Purification of bacterially-expressed GFP without an affinity tag has also been achieved and is detailed in Protocol 1.

Protocol 1: Purification of Recombinant GFP from Bacteria The following protocol gives high yields of purified GFP from a high-expression strain derived from TU#58 (Chalfie et al., 1994). The yield of purified protein has been as much as 150 mg/liter. This protocol has also been used to purify native GFP directly from the jellyfish, A. victoria, and several recombinant wild-type or mutant GFPs. One example is the mutant GFP expressed from the pBAD/gfp construct (Crameri et al., 1996). The mutant gfp in this vector is under the tight control of the arabinose promoter/repressor, araBAD, and can be induced continuously with a final concentration of 0.2% L(+) arabinose (w/v) in LB. We have transformed the E. coli strain DH5a with pBAD/gfp and have grown 10 liter of LB under continuous induction at 28°C for 24 h before harvesting. Ultimately, we were able to purify 50 mg of mutant GFP from the 10 liters. The most readily available high GFP expressing construct is TU#60 (Chalfie et al., 1994) sold by CLONTECH as pGFP. The gfp gene in pGFP is fused in-frame to the lacZ initiation codon from pUC19 (which adds an additional 24 amino acids to the N-terminus of GFP) that allows for high expression from the lac promoter. The pGFP vector also contains the bla gene for ampicillin selection and is a high copy number plasmid. Suitable E. coli strains which can be used to produce the protein include DH5a, JM109, and TB1 (New England BioLabs, Inc.). The GFP purified according to this protocol is suitable for biochemical and biophysical experiments, including crystallization trials. Materials Bacterial strain derived from TU#58 [pET3a/gfp, BL21 (DE3)] This strain expresses GFP as a nonfusion protein under the control of the T7 F 10-s10 promoter fragment. The transcription of the gfp gene is directly controlled by the T7 RNA polymerase. For transcription of gfp, TU#58 must be maintained in E. coli cells lysogenic for the l phage derivative, DE3 (Studier et al., 1990). The lDE3 lysogen carries the T7 RNA polymerase gene driven by the IPTG-inducible lacUV5 promoter. The pET3a plasmid also contains an ampR gene for selection. Cells are maintained continuously on ampicillin selection plates and selected on the basis of fluorescence prior to large-scale culture. LB (Miller, 1972) 10 g NaCl 10 g Bacto tryptone (Difco) 5 g Bacto yeast extract (Difco) Add DW to 1 liter, autoclave. Add ampicillin to 37 mg/ml for cell culture. LB + ampicillin plates 1 liter LB 14 g Bacto agar (Difco) 10 g Lactose Add ampicillin to 37 mg/ml prior to pouring plates. This concentration of ampicillin is suitable for the copy number of the plasmid. For pGFP, the optimal ampicillin concentration is 60 mg/ml.

409

410

METHODS AND PROTOCOLS

0.5 M IPTG (isopropyl-b-D-thiogalactopyranoside) in DW Extraction buffer 25 mM Tris-HCl, pH 8.0 1 mM b-mercaptoethanol (Eastman Kodak Co.) 0.1 M Phenylmethylsulfonylfluoride (PMSF) (Sigma) in 2-Propanol Low ionic strength buffer 10 mM Tris-HCl, pH 8.0 10 mM EDTA 50 mg/ml protamine sulfate (Sigma) in DW The protamine sulfate will not be in solution at room temperature. Warm the bottle under hot tap water immediately before dispensing. Ammonium sulfate (solid) Tris base (solid) Octyl agarose column buffer 10 mM Tris-HCl, pH 8.0 10 mM EDTA 1.0 M ammonium sulfate Sepharose column buffer 5 mM Tris-HCl, pH 8.0 0.02% NaN3 DEAE column buffer 5 mM Tris-HCl, pH 8.0 0.02% NaN3 Special Equipment Chromatography columns Octyl agarose HIC (Hydrophobic Interaction Chromatography) column (Pharmacia Biotech), 2.5 ¥ 12 cm Pre-equilibrate the column with octyl agarose column buffer. Sepharose CL-6B (Pharmacia Biotech), 3 ¥ 95 cm Pre-equilibrate the column with sepharose column buffer. DEAE Sepharose Fast Flow (Pharmacia Biotech), 2.5 ¥ 17 cm Pre-equilibrate the column with DEAE column buffer. 1. Select a brightly fluorescent colony from an LB + ampicillin plate and use it to inoculate 50 ml of LB + ampicillin. Grow overnight at 37°C. Visually select the most intensely green fluorescent colony from an ampicillin plate by placing the plate on a hand-held long-wave UV lamp (lmax = 365 nm). 2. Inoculate a flask containing 1 liter of LB + ampicillin with the 50 ml overnight culture. Grow at 37°C to OD660 = 0.8, then add IPTG to a final concentration of 0.5 mM. Induce cells at 37°C for 12 h.

PROTOCOL I: EXPRESSION OF GREEN FLUORESCENT PROTEIN

3.

4.

5.

6.

7.

8.

9.

No significantly higher production of GFP is obtained by longer incubation. The ideal temperature of expression is 37°C, but reasonable expression (up to 100 mg/l of LB) is achieved at 28°C. In our initial work with the TU#58-derived pET3a construct we obtained lower yields of 1–3 mg/l in LB. We increased the yields to 10 mg/l of LB broth by growing the E. coli cells at 28°C. With improvements in our colony selection technique, we now have cells that are considerably more productive and we are now able to grow these cells at 37°C with excellent GFP yields. Collect the GFP-producing E. coli cells by gentle centrifugation at 1000¥ g for 15 minutes at 4°C. Pellets can be stored at this step by freezing after harvesting. Extract the GFP from induced cells by repeated cycles of freezing and thawing. Slowly freeze (60 min at -20°C) and slowly thaw (60 min at room temperature) the packed pellets through three cycles (Johnson and Hecht, 1994). Follow this by two to four cold buffer washes (20¥ pellet volume) with the extraction buffer. Collect the wash supernatant by centrifugation (10,000¥ g, 15 min) and add PMSF to a final concentration of 1 mM. The PMSF will help prevent proteolytic cleavage of the protease-susceptible C-terminal “tail” of GFP. Usually it is necessary to freeze pellets between buffer washes to release all the GFP. While the freeze–thaw process is slower than other methods such as sonication or lysozyme treatment, the freeze–thaw extracts are remarkably clean (low viscosity, low DNA content, and high GFP content—up to 10% of total soluble protein). More than 90% of the GFP can be released from the cells by this method. Treat clarified extracts in low ionic strength buffer at 0–4°C with protamine sulfate to remove residual nucleic acids. Generally 1 mg protamine sulfate per 100 OD260 units is sufficient to precipitate most of the DNA, but not the GFP. Remove the precipitate by centrifugation (5000¥ g, 5 min). Add the protamine sulfate dropwise while stirring rapidly, so as not to precipitate GFP in localized regions of the extract. It is advisable to test each batch of GFP by small-scale titration in microfuge tubes to avoid “overshooting” the titration. It is not easy to recover GFP that is inadvertently precipitated by protamine sulfate. Precipitate the protamine-treated and clarified extract with ammonium sulfate (100% of saturation, 697 g/l extract) at 0°C. Add approximately 10 g of solid Tris base per liter of extract during the precipitation step to maintain a pH near 7.0. The pH of unbuffered saturated ammonium sulfate is close to 5.5, dangerously close to the low end of the GFP pH stability range. Generally, the precipitation of GFP is rapid. Collect the precipitate by centrifugation (10,000¥ g, 30 min) within an hour of precipitation. Expect near quantitative recovery if the GFP concentration in the crude extract is ⭓0.2 mg/ml. Dissolve the GFP-containing pellet in a minimal volume (just sufficient to dissolve the GFP) of octyl agarose column buffer containing 1 mM PMSF. Clarify the dissolved pellet by centrifugation (15,000¥ g, 20 min). Pelleted GFP, which appears yellow, will take on the familiar bright green color as it goes into solution and the entire suspension becomes clear. Load the clarified solution, at room temperature, onto an octyl agarose HIC column preequilibrated with column buffer. Elute the column stepwise, first with 250 ml of low ionic strength buffer +0.5 M ammonium sulfate, and then with 250 ml of low ionic strength buffer without ammonium sulfate. Complete elution of GFP requires 100–200 ml of the second buffer solution. The column is capable of binding more than 1 g of total protein and can be eluted free of GFP in less than 2 h. Hydrophobicity of octyl agarose columns varies greatly with the length and chemical

411

412

METHODS AND PROTOCOLS

nature of the spacer arm. We prefer a three-carbon spacer with ether linkage to the agarose beads. 10. Concentrate the GFP sample to a volume of 2–5 ml by ultrafiltration or ammonium sulfate precipitation. 11. Chromatograph at room temperature on a column of Sepharose CL-6B, preequilibrated in column buffer, at a flow rate of 1.4 ml/min. GFP elutes from this column at an apparent molecular weight of about 40 kDa, indicating significant dimerization. a. Alternate two-step purification Reversible dimerization can be used in an alternative purification scheme following gel filtration on Sepharose to achieve ~95% purity. Additional Materials Bio-Gel column buffer 0 mM Tris-HCl, pH 8.0 10 mM EDTA 1 M ammonium sulfate 0.02% NaN3 Additional Special Equipment Two columns of Bio-Gel P-100 medium resin (Bio-Rad), 10 ¥ 120 cm (~8 liters) and 3 ¥ 120 cm (~0.75 liter) 1. Run the larger Bio-Gel P-100 column at room temperature with a dilute GFP sample (0.2 mg/ml). Partial hydrophobic interaction in high salt causes GFP to elute at an apparent molecular weight of 21 kDa. 2. Then run the second smaller column at room temperature with a very concentrated GFP sample (20–100 mg/ml) in the Bio-Gel column buffer, with or without ammonium sulfate. GFP dimerizes at high-protein concentrations and elutes at an apparent molecular weight of 44 kDa. Nearly all contaminants that co-elute with GFP on the first column are removed on the second. 12. Final polishing to achieve >95% purity is on a DEAE Sepharose Fast Flow column at room temperature. Load the sample in the column buffer by gravity at a flow rate of 2–4 ml/min and elute with a 2.0-liter gradient of salt (0 to 0.5 M NaCl) in the same buffer. Gravity-driven flow rates of 5 ml/min can be achieved with excellent resolution, equaling that achieved with a 4-h-long shallow salt gradient on Pharmacia’s Mono Q FPLC column. In fact, minor isoforms of GFP that differ by one charged amino acid are quantitatively removed on DEAE Fast Flow. This column is capable of purifying up to 1 g of GFP. 13. Judge the purity of GFP by the ratio of the absorbance of the chromophore at its lmax to that of the aromatic region of the protein at 280 nm. Note that, in the chromophore absorption band, the wild-type recombinant GFP, the so-called “Stemmer” mutant [cycle 3 mut (F99S, M153T, V163A)] (Crameri et al., 1996), and native Aequorea GFP all fail to follow Beer’s law. For these three types of GFP, as protein concentration increases, the absorbance at 395 nm increases disproportionately while the absorbance at 475 nm decreases disproportionately. The suppression of the 475 nm shoulder upon dimerization (as occurs in cells overexpressing GFP) can be as great as five-fold, making these forms of GFP exceedingly poor absorbers of blue light (molar extinction coefficient ⭐3000 for the dimer). Thus, for accu-

413

PROTOCOL II: SPECIMEN PREPARATION

TABLE A.1.3. GFP Absorption Ratios Useful in Establishing Purity Chromophore lmax

e

l1/l2

Numerical Ratio

27,600 12,000 27,000 12,000 56,000

395/280

1.25

Recombinant wild-type GFP S65T

395 470 397 475 489

397/280

1.25

489/280

2.25

Stemmer mutant cycle 3 mut: F99S, M153T, V163A P4: Y66H

397 475 382

27,000 12,000 25,000

397/280

1.25

382/280

1.23

Protein Native Aequorea GFP

rate quantitation, it is necessary to measure absorbance at low protein concentration (0.05–0.20 mg/ml) in a concentration range that obeys Beer’s law. Table A.I.3 shows a partial list of GFP forms and spectral characteristics that may be used to judge purity. Contributed by Daniel G. González and William W. Ward

PROTOCOL II: SPECIMEN PREPARATION II.B

Fixed Specimens

Visualization of GFP in fixed cells or tissues has been successfully achieved for many types of cells after fixation with formaldehyde. In some cases, fixation of cells in sodium azide, methanol, ethanol or glutaraldehyde has also proven successful for subsequent visualization of GFP (some fixatives with sodium azide have been problematic). Although glutaraldehyde is a superior fixative to formaldehyde, it is not commonly used in fluorescence microscopy because it causes autofluorescence. At high concentrations, glutaraldehyde will destroy GFP fluorescence, but at low concentrations (e.g., 0.025%), this problem can be partially avoided. Denaturants such as 1% sodium dodecyl sulfate (SDS) or 8 M urea at room temperature can also be used in fixation procedures with the preservation of GFP fluorescence, but if GFP is fully denatured, or treated with 1% hydrogen perooxide or sulfhydryl reagents (Inouye and Tsuji, 1994), fluorescence is irreversibly destroyed. At high protein concentrations (above 5–10 mg/ml) or in high-salt solutions, GFP has been reported to dimerize, resulting in a fourfold reduction in absorption. GFP fluorescence has been reported to be sensitive to some nail polishes used to seal coverslips to slides (Chalfie et al., 1994; Wang and Hazelrigg, 1994). Molten agarose, rubber cement, or VALAP [1 : 1 : 1 Vaseline (petroleum jelly) : lanolin : paraffin chips, heated until clear] is recommended as a substitute for sealing coverslips. Alternatively, specimens can be viewed without applying a sealant by pressing down firmly on the coverslip to remove any excess mounting media between the slide and coverslip that might cause slippage of the coverslip.

414

METHODS AND PROTOCOLS

The use of slides marked with grids can aid in relocating cells of interest. Gridded slides available for use contain divisions of several millimeters in a printed Teflon coating (Cel-line Associates, Carlson Scientific), numbered divisions of 200–500 mm (“England finders”; Klarfield Rulings, Inc.), or squares of 55 mm (Eppendorf North America). The following methods have been successfully used to prepare fixed cells for GFP visualization.

Protocol 1: Bacteria (E. coli, EGFPmut1-3) Materials E. coli expressing EGFPmut1-3 5 mM Sodium azide in PBS (phosphate-buffered saline) Microscope slides Coverslips 1. Resuspend bacterial samples in 5 mM sodium azide in PBS to fix cells. 2. Mount cells on a glass slide with a glass coverslip. Fluorescence due to soluble EGFP is maintained as long as the cell surface integrity is not compromised. Fixing in sodium azide in PBS has no obvious effect on fluorescence. The cells can be fixed with a variety of crosslinking agents such as paraformaldehyde (2% w/v) and formaldehyde (1% v/v), and will retain some of the fluorescence from soluble EGFP. The GFP fluorescence in the fixed cells is also stable to photobleaching. Detergents and permeabilizing fixatives like methanol will destroy fluorescence. The effect of fixation on the ability to visualize GFP is most pronounced when imaging low levels of EGFP expression. CONTRIBUTED by RAPHAEL H. VALDIVIA, BRENDAN P. CORMACK, and STANLEY FALKOW

Protocol 2: Yeast (S. cerevisiae, TUB4-GFP) Materials Yeast expressing Tub4-GFP Formaldehyde (37%) Methanol Acetone DAPI (4¢,6-diamidino-2-phenylindole) Microscope slides Coverslips

415

PROTOCOL II: SPECIMEN PREPARATION

1. Fix cells by adding formaldehyde to a final concentration of 3.7%. 2. Incubate at RT or 30°C 1–2 h. 3. Treat with cold methanol for 5 min, then with cold acetone for 30 s. This procedure was performed to help flatten the cells, making the spindles easier to visualize in a single focal plane, rather than specifically for the GFP visualization. 4. To visualize DNA, incubate in 1 mg/ml DAPI for 1 min. 5. Mount cells on a glass slide with a glass coverslip. The ability to visualize GFP after fixation with formaldehyde may not pertain to all GFP fusion proteins in yeast. Although Tub4-GFP fluorescence could be observed following formaldehyde fixation, a-tubulin-GFP fusion proteins do not fluoresce visibly after fixation. CONTRIBUTED by TIM STEARNS (MARSHALL et al., 1996)

Protocol 3: Yeast (S. pombe, p93dis1-GFP) Materials Yeast expressing p93dis1-GFP Methanol or 2.5% glutaraldehyde Microscope slides Coverslips 1. Treat cells with methanol for 8 min at -80°C or with 2.5% glutaraldehyde at 33°C for 1 h. 2. Mount cells on a glass slide with a glass coverslip. TAKEN from NABESHIMA et al. (1995)

Protocol 4: Yeast (Schizosaccharomyces pombe, GFP) Protocol for preparation of fixed yeast cells for FACS analysis Materials S. pombe cells expressing GFP from an episomal expression vector EMM medium Sterile DDW 100% Ethanol 50 mM Sodium citrate 1. Grow cells in EMM to a density of 107 cells/ml. 2. Harvest 20 ml of cells by pelleting. 3. Suspend cells in sterile DDW to wash. Pellet again.

416

METHODS AND PROTOCOLS

4. Resuspend cells in 6 ml sterile DDW. 5. Add 14 ml 100% ethanol. The fixed cells can be stored indefinitely at 4°C. 6. Prior to analysis, harvest cells by centrifugation. 7. Wash with sterile DDW. 8. Resuspend in 5 volumes of 50 mM sodium citrate. 9. Sonicate briefly. 10. Analyze by FACS. Data was acquired for 20,000 cells for each sample and analyzed by plotting fluorescence against forward scatter. TAKEN from ATKINS and IZANT (1995)

Protocol 5: Drosophila (D. melanogaster, GFP-Exu in egg chambers) Materials exuSco2/exuSco2; P[Cas,NGE]3/+ females PBS (phosphate buffered saline) Fixative (Theurkauf and Hawley, 1992) 8% Formaldehyde 100 mM Potassium cacodylate, pH 7.2 100 mM Sucrose 40 mM Potassium acetate 10 mM Sodium acetate 10 mM EGTA Strips of Whatman filter paper 50% (v/v) Glycerol in PBS Microscope slides Coverslips 1. 2. 3. 4. 5. 6. 7. 8.

Collect 0- to 1-day-old females and place in well-yeasted vials with males. Keep 2 days prior to use, so females are 2–3 days old. Anesthetize females with CO2. Dissect ovaries in PBS. Place in fixative for 10 min. Wash 3 ¥ 10 min in PBS. Tease ovarioles apart in a drop of PBS on a glass slide. Remove excess PBS with strips of Whatman filter paper, then cover with a drop of 50% glycerol in PBS. 9. Cover with a glass coverslip. CONTRIBUTED by TULLE HAZELRIGG (WANG and HAZELRIGG, 1994)

417

PROTOCOL II: SPECIMEN PREPARATION

Protocol 6: Mammalian Cells (HeLa cells, GFP chimeras) Materials PBS (phosphate buffered saline) Fixative 2% Formaldehyde or 0.025% glutaraldehyde in PBS Fluoromount (Southern Biotechnology) Microscope slides Coverslips 1. Place cells in fixative for 10 min at room temperature. 2. Rinse twice in PBS. 3. Mount cells on a glass slide in Fluoromount with a glass coverslip. a. Staining of fixed cells with antibodies Additional Materials Primary antibody solution Primary antibody in PBS 10% Bovine serum 0.5% Saponin 10% Bovine serum in PBS Secondary antibody solution Rhodamine-labeled secondary antibody in PBS 10% Bovine serum 0.5% Saponin 1. Carry out fixation and PBS wash steps as described above. 2. Incubate cells for 1 h at room temperature in primary antibody solution. 3. Wash cells three times over 30 min in 10% bovine serum in PBS to remove unbound antibody. 4. Incubate cells for 1 h at room temperature in secondary antibody solution. 5. Wash cells three times over 30 min in 10% bovine serum in PBS to remove unbound secondary antibody. 6. Rinse cells quickly in PBS without serum. 7. Mount cells on a glass slide in Fluoromount with a glass coverslip. Observe using rhodamine and fluorescein filters to determine the distribution of antibody and GFP. CONTRIBUTED by JENNIFER LIPPINCOTT-SCHWARTZ

418

METHODS AND PROTOCOLS

III.B.2 Photobleaching, Photoactivation, Photodamage, and pH Dependence of GFP. Photobleaching of GFP has been reported to be slow (Chalfie et al., 1994; Niswender et al., 1995), certainly much slower than fluorescein under similar conditions. For instance, continuous observation in a confocal microscope for 20 min only reduced GFP intensity to one-half its original value (Niswender et al., 1995). Although GFP is resistant to photobleaching, it has proven very useful in fluorescence photobleaching recovery experiments (described below and in Chapter 12) because of its ability as a fusion protein to be targeted to specific organelles. Some problems with photobleaching of GFP have been reported. For instance, photobleaching with 395–440 nm light is accelerated by some agents used to anesthestize C. elegans, such as 10 mM NaN3, and another anesthetic agent, phenoxypropanol, has been reported to quench GFP fluorescence (Chalfie et al., 1994). An alternative anesthetic agent for use in visualizing GFP in live C. elegans is described in Protocol II. Some mutant forms of GFP may also be more susceptible to photobleaching. For imaging in cultured mammalian cells, 10 mM Trolox has been added to live cells when visualizing Y66H, Y145F (P4-3) GFP to reduce the rapid photobleaching of this mutant GFP (Rizzuto et al., 1996). Contrary to early reports, the photobleaching rate of wild-type GFP is close to the same whether it is excited in the UV or the blue, but the decrease in UV-excited fluorescence appears more rapid because of photoisomerization (Cubitt et al., 1995). Wild-type GFP has two absorption peaks, 395 and 475 nm, which are thought to be related via rotation about a bond (isomerization) within the chromophore structure. The isomerization can be induced by irradiation of wild-type GFP with either 395- or 490-nm light and the kinetics of this photoinduced reaction have recently been measured (Chattoraj et al., 1996). The photoisomerization affects the brightness of wild-type GFP during visualization, but does not occur in either the S65T or F64L, S65T (EGFPmut1) mutants of GFP. The recent solving of the crystal structures of wild-type GFP and the S65T mutant (described in Chapter 4) shows that the chromophore is tilted within the protein structure differently in the two proteins. This difference in tilting angle is likely the origin of the photoisomerization observed for wild-type GFP. The photoisomerization can also be used to photoactivate wild-type GFP. That is, the fluorescence properties of GFP change after irradiation. When wild-type GFP is irradiated with either UV (~395 nm) or blue light (488 nm), photoisomerization occurs and causes an increase in the 475-nm peak and a decrease in the 395-nm peak (Cubitt et al., 1995). In this manner, UV preexposure can be used to increase the blue excitation brightness of wild-type GFP (Chalfie et al., 1994). A detailed discussion of the chromophore structure and its photoinduced reactions is presented in Chapters 4 and 5. Since the photobleaching of GFP is low, cellular photodamage arising from GFP can also be expected to be low. Only a few extended studies have been reported to date, but photodamage associated with using GFP has not been identified as a problem. Anecdotal evidence suggests that photodamage associated with GFP fusion proteins is less than that found with labeling with other fluorophores. For example, fluorescence photobleaching recovery has been used to examine the diffusional mobilities of Golgi-targeted GFP chimeras (Cole et al., 1996), and in neither approach used was there evidence of cellular photodamage or photoinduced crosslinking of the GFP chimeras. Multiple bleaches of the same spot did not affect the measured diffusion coefficient, and staining with antibodies after photobleaching revealed intact membrane structures. Moreover, the mobile fraction did not decrease with successive bleaches, which would be expected if incomplete recovery was due to photoinduced immobilization of a fraction of the labeled molecules. Evi-

REFERENCES

dence such as this suggests that GFP will be excellent for time-lapse and four-dimensional imaging. The resistance of GFP to photobleaching may be due to the protection of the chromophore by a tightly packed barrel of b sheets [referred to as a b-can structure by Yang et al. (1996)], as revealed by the crystal structures (Ormö et al., 1966; Yang et al., 1996). Still, the intense light typically used for excitation of fluorescence can generate free radicals that in turn can damage cellular proteins. Excitation with UV light can also cause crosslinking and breakage of DNA, producing further detrimental effects. As with any fluorescence microscopy study of living samples, it is important to monitor sample viability during and after the experiment. One measure of viability is the ability of irradiated cells to divide further with the same doubling or mitotic cycle time as unirradiated cells. During time-lapse confocal imaging of mitotic spindles in live Drosophila embryos visualized using the Ncd motor protein fused to wild-type or S65T mutant GFP, photodamage caused the spindles in the irradiated region of the embryo to become delayed relative to the unirradiated region, resulting in asynchronous divisions (Endow, unpublished results). The cause of this damage was not determined (it could be caused by the GFP or by interactions between the incident light and endogenous absorbers), but lowering the excitation intensity to a level that reduced or eliminated the photodamage still gave a strong GFP signal. Delayed mitotic divisions or cell arrest has also been observed in Dictyostelium (Maniak et al., 1995) and yeast (Kahana and Silver, unpublished results) upon overirradiation of cells. In general, addition of free radical scavengers (Mikhailov and Gundersen, 1995) or antioxidants to the medium may aid in the imaging of GFP in live specimens with bright light for long periods of time. Oxyrase (Oxyrase Inc. P.O. Box 1345, Mansfield, OH 44901) (0.3 U/ml) and ascorbic acid (0.1–1.0 mg/ml) are two antioxidants that have been shown to reduce photodamage when added to the medium of living cells. Where possible, increasing gfp gene dose, or the use of brighter GFP variants is advantageous in imaging live specimens, permitting greatly reduced levels of exposure to fluorescent or laser light. Finally, the brightness of some GFP mutants appears to be sensitive to pH, although detailed studies have not yet appeared in the literature. As an example, wild-type GFP shows relatively even brightness from pH 5 to 10 (Ward, 1981), while the S65T mutant is 2-fold brighter at pH 7 than at pH 6 (Patterson and Piston, unpublished results). A similar fall-off in brightness at lower pH is exhibited by the F64L, S65T (EGFPmut1) double mutant.

REFERENCES Atkins, D., and Izant, J. G. (1995). Expression and analysis of the green fluorescent protein gene in the fission yeast Schizosaccharomyces pombe. Curr. Genet. 28:585–588. 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. 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. Chiu, W.-L., 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.

419

420

METHODS AND PROTOCOLS

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 green-fluorescent protein. Biochemistry 32:1212–1218. Cole, N. B., Smith, C. L., Sciaky, N., Terasaki, M., Edidin, M., and Lippincott-Schwartz, J. (1996). Diffusional mobility of Golgi proteins in membranes of living cells. Science 273:797–801. Crameri, A., Whitehorn, E. A., Tate, E., and Stemmer, W. P. C. (1996). Improved green fluorescent protein by molecular evolution using DNA shuffling. Nature Biotech. 14:315–319. 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. Davis, I., Girdham, C. H., and O’Farrell, P. H. (1995). A nuclear GFP that marks nuclei in living Drosophila embryos; maternal supply overcomes a delay in the appearance of zygotic fluorescence. Dev. Biol. 170:726–729. Haseloff, J., and Amos, B. (1995). GFP in plants. Trends Genet. 11:328–329. Heim, R., Prasher, D. C., and Tsien, R. Y. (1994). Wavelength mutations and posttranslational autoxidation of green fluorescent protein. Proc. Natl. Acad. Sci. USA 91:12501–12504. Heim, R., Cubitt, A. B., and Tsien, R. Y. (1995). Improved green fluorescence. Nature (London) 373:663–664. 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. 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. Johnson, B. H., and Hecht, M. H. (1994). Recombinant proteins can be isolated from E. coli cells by repeated cycles of freezing and thawing. Bio Technol. 12:1357–1360. Lim, C. R., Kimata, Y., Nomaguchi, K., and Kohno, K. (1995). Thermosensitivity of green fluorescent protein fluorescence utilized to reveal novel nuclear-like compartments in a mutant nucleoporin NSP1. J. Biochem. 118:13–17. Maniak, M., Rauchenberger, R., Albrecht, R., Murphy, J., and Gerisch, G. (1995). Coronin involved in phagocytosis: dynamics of particle-induced relocalization visualized by a green fluorescent protein tag. Cell 83:915–924. Marshall, L. G., Jeng, R. L., Mulholland, J., and Stearns, T. (1996). Analysis of Tub4p, a yeast gtubulin-like protein: implications for microtubule-organizing center function. J. Cell Biol. 134:443–454. Mikhailov, V. S., and Gunderson, G. G. (1995). Centripetal Transport of microtubules in motile cells. Cell Motil. Cytoskeleton 32:173–186. Morise, H., Shimomura, O., Johnson, F. H., and Winant, J. (1974). Intermolecular energy transfer in the bioluminescent system of Aequorea. Biochemistry 13:2656–2662. Nabeshima, K., Kurooka, H., Takeuchi, M., Kinoshita, K., Nakaseko, Y., and Yanagida, M. (1995). p93dis1, which is required for sister chromatid separation, is a novel microtubule and spindle pole body-associating protein phosphorylated at the Cdc2 target sites. Genes Dev. 9:1572–1585. Niswender, K. D., Blackman, S. M., Rohde, L., Magnuson, M. A., and Piston, D. W. (1995). Quantitative imaging of green fluorescent protein in cultured cells: comparison of microscopic techniques, use in fusion proteins and detection limits. J. Microscopy 180:109–116. Ogawa, H., Inouye, S., Tsuji, F. I., Yasuda, K., and Umesono, K. (1995). Localization, trafficking, and temperature-dependence of the Aequorea green fluorescent protein in cultured vertebrate cells. Proc. Natl. Acad. Sci. USA 92:11899–11903. Ormö, M., Cubitt, A. B., Kallio, K., Gross, L. A., Tsien, R. Y., and Remington, S. J. (1966). Crystal structure of the Aequorea victoria green fluorescent protein. Science 273:1392–1395. Pines, J. (1995). GFP in mammalian cells. Trends Genet. 11:326–327. 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.

REFERENCES

Rizzuto, R., Brini, M., De Giorgi, F., Rossi, R., Heim, R., Tsien, R. Y., and Pozzan, T. (1996). Double labelling of subcellular structures with organelle-targeted GFP mutants in vivo. Curr. Biol. 6:183–188. Theurkauf, W. E., and Hawley, R. S. (1992). Meiotic spindle assembly in Drosophila females: behavior of nonexchange chromosomes and the effects of mutations in the nod kinesin-like protein. J. Cell Biol. 116:1167–1180. Valdivia, R. H., and Falkow, S. (1996). Bacterial genetics by flow cytometry: Rapid isolation of Salmonella typhimurium acid-inducible promoters by differential fluorescence induction. Mol. Microbiol. 22:367–378. Wang, S., and Hazelrigg, T. (1994). Implications for bcd mRNA localization from spatial distribution of exu protein in Drosophila oogenesis. Nature (London) 369:400–403. Ward, W. W. (1981). Properties of coelenterate green fluorescent proteins. In Bioluminescence and Chemiluminescence, DeLuca, M. A. and McElroy, W. D., Eds., Academic, San Diego, pp. 225–234. Yang, T.-T., Kain, S. R., Kitts, P., Kondepudi, A., Yang, M. M., and Youvan, D. C. (1996). Dual color microscopic imagery of cells expressing the green fluorescent protein and a red-shifted variant. Gene 173:19–23.

421

INDEX

Absorption/extinction cross-section, green fluorescent protein, 57–59 Absorption of chromophores, suppression, molecular mechanism for, 59–61 Absorption ratios, GFP purification protocols, 413 Acid-inducible gene isolation, S. typhimurium GFP, 172–173 Actin cycling, Drosophila GFP, FRAP analysis, 243 Action potentials, in O. geniculata, 20–21 Activation control, green fluorescent protein fluorophore, 76 Acumen Explorer, reef coral fluorescent protein expression and detection, 357 Adenosine triphosphate (ATP), firefly luminescence, 31–33 ADH1 promoter, S. cerevisiae GFP gene expression, 181–182 adh1+ promoter, S. pombe GFP gene expression, 183 Aequorea aequorea: biochemistry, 19 green fluorescent protein, 1–4 biological function, 41 chromophore structure, 53–54 denaturation/renaturation, 49–50 mutant structures, 73–74 organic solvents, 51–52 physical characteristics, 42–52 protease resistance in, 50–51 light organs in, 6–8 protein concentration, and absorption suppression, 59–61 Aequorea coerulescens, green fluorescent protein mutant structures, 112–113 Aequorea forskalea, green fluorescent protein in, 10–11

Aequorea victoria: GFP-like proteins, 140–141 green fluorescent protein in, 10–11, 41–42 monomer/dimer equilibrium, 57–59 mutant structures, 83–84, 112–113 stability of, 48–49 stability properties, 49 Aequorin: biological function, 41 calcium channel buffer, 6–8 crystalline structure, 23–25 discovery of, 3–4 jellyfish species, 9–11 physical characteristics, 41–52 Aggregation, GPF mutation modulation, 102–105 AmCyan1 mutant: basic properties, 341 flow cytometry, single-color analysis, 353 multicolor analysis, 349–351 Amino acids: DsRed/DsRed-Express mutants, 344 fluorophore formation, 72–73 GFP dissection and reassembly, 393–396 in green fluorescent protein, 41–48 green fluorescent protein mutant structures, 73–75 heterogeneous residues, 84–85 nucleic acid changes, 112 phenolate anion with stacked p-electron system, 92–93 plant GFP, 270–272 predetermined stretch, randomization, 86 wavelength altered ratios, 97–100 nonbioluminescent anthozoa GFP-like proteins, sequence comparisons, 124–126 protein absorption suppression, 60–61 ZaGreen1 mutant, substitutions, 341, 343

Green Fluorescent Protein: Properties, Applications, and Protocols, Second Edition Edited by Martin Chalfie and Steven R. Kain Copyright © 2006 John Wiley & Sons, Inc.

423

424

INDEX

Anabaena sp., heterocyst formation, 167–168 Ancient diversity, GFP-like proteins, 154–155 “Ancient duplication” scenario, GFP-like proteins, 151 Anemonia majano, reef coral fluorescent proteins, 340–341 Anemonia sulcata, reef coral fluorescent proteins, 340–341 Animal hosts, GFP-host interactions, 169 Anthozoa, GFP-like proteins: FP modification, mutagenesis, 129–130 homolog color diversity, 122–124 mutagenesis color transitions, 126–128 oligomeric state, 126 photoactivatable probes, 130–134 Dronpa, 134 DsRed “greening,” 130–132 fluorescent protein kindling, 132–134 UV-induced green-to-red photoconversion, 132 phylogenetic relationships, 151–152 research background, 121–122 sequence comparisons, 124–126 Antibody staining, C. elegans GFP pattern analysis, 204–205 Antiparallel beta strands, green fluorescent protein folding, 78–79 Apoaequorin, isolation, 25 Arabidopsis, green fluorescent proteins in: cell imaging, 274–276 cell marking, 280–281 cryptic intron removal, 262–263 gene expression, 263–264 modification, 274 gfp mRNA cryptic splicing in, 260–262 localization, 264–265 maturation, 269–272 spectral modification, 272–273 subcellular dynamics visualization, 276–279 subcellular targeting, 265–269 b-Arrestin recruitment assay, fluorescent proteins, 372–376 ars1+ sequence, S. pombe GFP gene expression, 182–183 Artifacts, C. elegans GFP, 219–220 AsRed2 mutant: flow cytometry, single-color analysis, 354–355 reef coral fluorescent proteins, 345 asulCP protein, photoconversion, 132 Autofluorescence techniques, yeast GFP analysis, 196 Autoinduction, bacterial luciferases, 26–28

Autonomously replicating, centromeric (ARS/CEN) plasmids: Candida albicans GFP gene expression, 183–184 S. cerevisiae GFP gene expression, 181–182 S. cerevisiae GFP gene expression, extrachromosomal constructs, 194–195 S. pombe GFP gene expression, 182–183 B. subtilis, GFP isolation, 165 spore formation, 165–167 BAC injection, transgenic fish GFP, 292 Bacteria: biochemistry, 18–19 GFP purification protocols, 409–413 green fluorescent protein in: complex environments, 168–171 development and cell biology, 165–168 cell division, E. coli, 168 heterocyst formation, Anabaena, 167–168 spore formation, B. subtilis, 165–167 ecology and behavior, 170–171 fixed specimen preparation, 414 future research issues, 174–175 genetic applications, 171–174 acid-inducible gene isolation, S. typhimurium, 172–173 flow cytometry, 172 macrophage-inducible gene isolation, S. typhimurium, 173–174 host interactions, 169–170 research background, 163–165 luminescence in, 25–28 Balancers, Drosophila GFP, 247 BD FACS analysis protocol, reef coral fluorescent protein expression and detection: fixed cells, 352 unfixed cells, 351–352 BD FACSVantage SE cell sorter, reef coral fluorescent protein expression and detection, 352–353 multicolor analysis, 356 BD Living Color fluorescent proteins, comparison table, 342 b-can structure: green fluorescent protein, 69–71 nonbioluminescent anthozoa GFP-like proteins, sequence comparisons, 124–126 stability, 48–49

425

INDEX

Bioinformatic studies, fluorescent protein applications, 380–383 Biological functions, of green fluorescent protein, 40–41 Bioluminescence resonance energy transfer (BRET), drug assays, fluorescent proteins, 367–371 target validation studies, 381–383 Bioluminescent coelentrates: evolutionary function, 15–19 GFP-like proteins, 140–141 green fluorescent protein sources, 4–8 phylogenetic organization, 15–17 BioRad MRC-600 microscope, plant GFP imaging, 278–279 Bioremediation, bacterial GFP ecology and behavior, 170–171 Biosensors: fluorescent protein drug screening, 378–380 GFP mutant structures, 106–108 Blue fluorescent protein (BFP): biosensor mutations, 107–108 drug assays, 367–371 imidazole derivation from chromophores, 93–94 isolation of, 8–9 in mammalian cells, 308–309 plant GFP modification, 273 yeast GFP spectral variants, 185–186 Blue-shifted emissions, bacterial luciferases, 27–28 Brightness properties, green fluorescent protein, 54 Caenorhabditis elegans: green fluorescent protein in: artifacts, 219–220 cell biology processes, 213 cell culturing and sorting, 215 cell fate markers, 212 cellular anatomy visualization, 208–212 CFP/YFP variants, 216–217 co-labeling of expression patterns, 213–215 electrophysiological recording, 213 expression pattern analysis, 204–205 injection marker experiments, 215–216 laser ablation, 213 mosaic analysis, 216 neuronal function and plasticity measurement, 212–213

PCR fusion, 217 protein localization, 206–208 protein tagging, 215 recombination cloning, 219 regulatory sequence decoding, 205–206 reporter gene constructs, 217–219 research background, 203–204 transgenic cell lines, 219 visualization of, 219 in vivo recombination, 219 target validation studies, fluorescent proteins, 382–383 Calcium channels: aequorin discovery, 3–4 fluorescent protein biosensors, 379–380 green fluorescent photocyte dispersion, 22–25 Candida albicans, green fluorescent protein in, 180 gene expression system, 183–184 Carboxypeptidases, in green fluorescent proteins, 45–48 Cargo proteins, mammalian GFPs, organelle structure and membrane trafficking, 322–324 Cauliflower mosaic virus (CaMV), S. pombe GFP gene expression, 183 cDNA protein sequence: GFP-like proteins, gene coding, 142–143 green fluorescent protein: mutant structures, 83–85 durable photoisomerization, 99–100 prokaryotes, 164–165 Cell adhesion assays, fluorescent proteins, 366 Cell-based diagnostics, green fluorescent protein research, 61–62 Cell biology: bacterial GFP, 165–168 animal host interactions, 169 cell division, E. coli, 168 heterocyst formation, Anabaena, 167–168 spore formation, B. subtilis, 165–167 Drosophila GFP, 232–243 chromosome structure, 234–235 fusion protein FRAP analysis, 242–243 meiosis and mitosis, 235–238 morphogen gradients, 239–240 neuronal function and connectivity, 238–239 organelle function and transport, 234 original studies, 232–233 tagged RNAs, 241–242 tissue morphogenesis, 233–234

426

INDEX

Cell biology (continued) transcription factors, 241 mammalian cell GFP, 316–318 transgenic GFP markers, 296–297 Cell culturing and sorting: C. elegans GFP, 215 reef coral fluorescent protein expression and detection: flow cytometry analysis, 351–352 fluorescence microscopy, 347–348 Cell fate marker, C. elegans GFP, 212 Cellular anatomy, C. elegans GFP visualization, 208–212 Cerulean mutants: indole derivation from chromophores, 93 in mammalian cells, 308–309 Chaotropes, green fluorescent protein and, 52 Chemical mutagens, random mutagenesis, green fluorescent protein, 85 Chicks, transgenic GFP in, 294 Chromophore structure: green fluorescent protein, 8–9 absorption suppression, molecular mechanisms, 59–61 brightness properties, 54 molar extinction coefficient, 55–56 mutations, 83–85 prokaryotes, 164–165 quantum yield measurements, 54–55 spectral mutants, 89–94 imidazol derivation, Y66H, 93–94 indole derivation from Y66W, 93 neutral phenol, 92 phenolate anion, 89, 92 phenolate anionstacked pi-electron system, 92–93 phenyl derivation, Y66F, 94 wild-type neutral phenol/anionic phenolate mixture, 89 spectroscopic analysis, 53–56 plant GFP, maturation mechanisms, 270–272 Chromoproteins (CPs), color diversity, anthoza GFP homologs, 122–124 Chromosome dynamics: C. elegans GFP visualization, 213 Drosophila GFP, 234–235 mammalian cell GFP, 321 2m circle sequences, S. cerevisiae GFP gene expression, 182 Circular permutations, GFP mutations, 96–100 Cis-regulatory logic, C. elegans GFP decoding, regulatory sequences, 206

Citrine YFP variants: in mammalian cells, 308–309 in transgenic vertebrates, 288–289 Cln2p protein, yeast GFP gene expression, 193–194 Clonal analysis: Drosophila GFP, 247–248 reef coral fluorescent protein expression and detection, 352–353 Cloning mechanisms, in transgenic fish GFP, 293 Cnidarian species: biochemistry, 18–19 GFP-like proteins, 139–140 deep-level relationships, 149–151 luminescence, 19–25 phylogenetic classification, 18 Codon optimization, green fluorescent protein mutant structures, nucleic acid changes, 112 Coelentrates. See also Nonbioluminescent anthozoa green fluorescent protein in, 41–42 photoproteins in, 19–25 Coelentrazine, cnidarian luminescent systems, 22–25 Coexpression marker, transgenic GFP as, 296 Co-labeling techniques, C. elegans GFP gene expression, 213–214 Color diversity: anthoza GFP homologs, 122–124 GFP-like proteins, 144–146 phylogenetic relationships, 151–153 nonbioluminescent anthozoa GFP-like proteins, mutagenesis, 126, 128 Colorless nonfluorescent GFP (acGFPL), cloning of, 112–113 Complex environments, bacterial GFP in, 168–171 Confocal microscopy: green fluorescent protein brightness, 54 nuclear translocation assays, fluorescent proteins, 365 transgenic vertebrate GFP, 289–290 Conjugated double bonds, color diversity, anthoza GFP homologs, 124 Conjugative DNA transfer, bacterial GFP ecology and behavior, 170–171 Connectivity structures, Drosophila GFP cell biology, 238–239 Copepods, GFP-like proteins, 139–140 deep-level relationships, 149–151 functions, 147

427

INDEX

Co-transfection, mammalian cell GFP, 318 Courtship patterns, firefly luminescence, 30–33 Covalent mutant alteration, fluorophore structures, 94–96 Cryptic introns, in plant GFP, removal, 262–263 Cryptic splice sites: Arabidopsis gfp mRNA, 260–262 green fluorescent protein mutant structures, 112 Ctenophores, luminescence in, 19–25 C-termainal domains: in dinoflagellate luciferase, 29–30 in firefly luciferase, 31–33 GFP dissection and reassembly, 397–398 GFP truncation and fusion constructs, 76–77 in green fluorescent proteins, 45–52 CUG codons, Candida albicans GFP gene expression, 183–184 CUP1 promoter, yeast GFP gene expression, 193–194 Cyan fluorescent proteins (CFPs): C. elegans variants, 216–217 Candida albicans GFP gene expression, 183–184 cellular anatomy visualization, 210–212 color diversity, anthoza GFP homologs, 122–124 mutagenesis, 128 dissection and reassembly, 394–396 drug assays, 367–371 indole derivation from chromophores, 93 in mammalian cells, 308–309 neuronal function and plasticity studies, 212–213 phylogenetic relationships, 151–152 protease assays, 368 S. cerevisiae GFP gene expression, 181–182 S. pombe GFP gene expression, 183 yeast GFP localization, 189–190 fluorescence resonance energy transfer, 192–193 Cyclization, green fluorescent protein structures: aggregation modulation, GFP mutations, 103–105 fluorophore, activation control, 76 mutant structures, 84–85 truncation and fusion constructs, 76–77 Cypridina luciferin, green fluorescent protein chromophore, 8–9 Cytomegalovirus promoter, reef coral fluorescent protein expression and detection, 346

Cytoskeleton structure, mammalian cell GFP analysis, 319–320 Cytosolic coat proteins, mammalian GFPs, organelle structure and membrane trafficking, 322–324 Decapentaplegic (Dpp) gradient, Drosophila GFP cell biology, 239–240 Deep-level relations, GFP phylogeny, 149–151 Denaturation, in green fluorescent protein, 49–50 Dendronephtya, phylogenetic relationships, 152–153 Detergents, green fluorescent protein and, 52 Deterring luminescence, GFP-like proteins, 141 Diethylaminoethyl (DEAE) particles, aequorin, 6–8 Differential fluorescence induction (DFI), bacterial GFP: acid-inducible gene isolation, S. typhimurium, 173 genetic applications, 172 Dimeric structure: aggregation modulation, GFP mutations, 102–105 green fluorescent protein: dimerization control, 77–78 monomer/dimer equilibrium, 57–59 site-directed mutations, 86 in green fluorescent protein, 42–52 Dinoflagellates: biochemistry, 18–19 bioluminescence properties, 28–30 Directed evolution, green fluorescent protein mutant structure, 87 Discosoma sp., reef coral fluorescent proteins, 340–341 DNA-binding proteins, yeast GFP localization, 190–191 DNA injection, transgenic mice GFP, 294–295 DNA localization, yeast GFP, 190–191 DNA shuffling, green fluorescent protein mutation, 86–87 DNA transfer, GFP spore formation, B. subtilis, 167 Donor/acceptor proteins, bioluminescence resonance energy transfer assays, 370–371 Dronpa protein, photoactivation of, 134 Drosophila, green fluorescent protein applications in: cell biology and development, 232–243

428

INDEX

Drosophila, green fluorescent protein applications in (continued) chromosome structure, 234–235 fusion protein FRAP analysis, 242–243 meiosis and mitosis, 235–238 morphogen gradients, 239–240 neuronal function and connectivity, 238–239 organelle function and transport, 234 original studies, 232–233 tagged RNAs, 241–242 tissue morphogenesis, 233–234 transcription factors, 241 clonal analysis, 247–249 embryonic balancers, 247 fixed specimen preparation, 416 gene expression, 228–231 detection timing and sensitivity, 229–231 transformation vectors, 228–229 transient expression, 229 gene regulation studies, 244–247 research background, 227–228 screens, 250–251 Drug screening assays, fluorescent proteins, 363–380 b-arrestin recruitment assay, 372–376 BRET protein-protein interaction, 369–371 cell adhesion assays, 366 dual colour spectroscopy assays, 371–372 fluorescence-activated cell sorting, 378 fluorescent protein biosensors, 378–380 fluorescent protein degradation, 372 FRET/BRET assays, 367–371 FRET protease assays, 368 FRET protein-protein interaction, 368–369 GPCR/GFP fusion assays, 376–377 nuclear translocation assays, 364–365 protein complementation assays, 377–378 reporter gene assays, 366–367 DsRed-Express mutant: flow cytometry, single-color analysis, 354–355 reef coral fluorescent proteins, 344 DsRed mutants: C. elegans FP variants, 217 cell adhesion assays, 366 color diversity, anthoza GFP homologs, 124 flow cytometry, single-color analysis, 354–355 Fluorescent timer (E5) protein, 345 “greening” of, 130–132 monomer structure, 345 mutagenic modification, 129–130

nonbioluminescent anthozoa GFP-like proteins: E5 mutant, 128 oligomeric state, 126–127 reef coral fluorescent proteins, 344 dual color analysis, 351 dual color spectroscopy assays, 371–372 S. cerevisiae GFP gene expression, 181–182 S. pombe GFP gene expression, 183 in transgenic vertebrates, 289 yeast GFP spectral variants, 185–186 Dual-channel imaging, plant GFP imaging, 275–276 Dual protein labeling, C. elegans FP variants, 216–217 Durable photoisomerization, GFP mutations, 96–100 Ecological applications, bacterial GFP, 170–171 Electrophysiological recording, C. elegans GFP cell identification, 213 Electroporation, transgenic chick GFP expression, 294 Emission spectra: Drosophila GFP detection, 230 green fluorescent proteins, 41–42 Energy levels, green fluorescent protein mutant structures, 100–101 Enhanced green fluorescent protein (EGFP): AmCyan1 mutant comparison, 341 dissection and reassembly, 394–396 dual color spectroscopy assays, 371–372 in mammalian cells, 307–309 reef coral fluorescent protein expression and detection, 347 in transgenic vertebrates, 288–289 Enhancer analysis, transgenic GFP for, 297–298 Enhancer trap screens: Drosophila GFP, 250–251 plant GFP marking, 280–281 Entactin, G2F domains, 147–149 Error-prone polymerase chain reaction, green fluorescent protein mutants, 85–86 Escherichia coli: cell division, bacterial GFP and, 168 firefly luciferase in, 31–33 green fluorescent protein expression in, 72 Evolution, bioluminescense and, 15–19 Excitation spectra: Drosophila GFP detection, 230

429

INDEX

green fluorescent proteins, 41–42 mutant ratios, 96–100 plant GFP modification, 272–273 Extrachromosomal (plasmid) GFP constructs, S. cerevisiae GFP gene expression, 194–195 Exuperantia protein, Drosophila GFP, 232–243 fbp1+ promoter, S. pombe GFP gene expression, 183 Fireflies: biochemistry, 18–19 bioluminescence in, 30–33 Fish, transgenic green fluorescent protein in, 290–293 BAC injection, 292 meganuclease, 292 plasmid injection, 290–291 somatic nuclear transfer, 293 transposons, 292–293 Fission yeast, S. pombe GFP gene expression, 182–183 Fixatives, green fluorescent protein and, 52 Fixed specimen preparation protocols, GFP specimens, 413–418 Flashing mechanism, firefly luminescence, 30–33 Flavin mononucleotide (FMN): in bacteria, 26–28 in yellow fluorescent protein, 27–28 Flow cytometry techniques: bacterial GFP: animal host interactions, 169 genetic applications, 172 mammalian cell GFP analysis, 319 reef coral fluorescent proteins, 351–353 single-color analysis, 353–355 AmCyan1, 353 DsRed2, DsRed-Express and AsRed2, 354–355 HcRed1, 355 ZsGreen1, 354 ZsYellow1, 354 three- and four-color analysis, 356 two color analysis, 355 Fluorescence activated cell sorting (FACS): bacterial GFP: acid-inducible gene isolation, S. typhimurium GFP, 173 genetic applications, 172 fluorescent protein drug screening, 378 green fluorescent protein: brightness properties, 54

mutant structures, 88 transgenic GFP markers, 297 yeast GFP gene expression, 194 Fluorescence correlation spectroscopy (FCS), mammalian cell GFP, 312–315 Fluorescence cross-correlation spectroscopy (FCCS), mammalian cell GFP, 315 Fluorescence emission: covalent mutation of fluorophore, 95–96 green fluorescent protein, 6–8 spectroscopic analysis, 54–55 Fluorescence in situ hybridization (FISH), yeast GFP, RNA localization, 191–192 Fluorescence lifetime imaging microscopy (FLIM), mammalian cell GFP, 314–315 Fluorescence localization after photobleaching (FLAP), mammalian cell GFPs, 312 Fluorescence loss in photobleaching (FLIP), mammalian cell GFPs, 312 Fluorescence microscopy: green fluorescent protein brightness, 54 mammalian cell green fluorescent proteins, 309–316 FRET and FCS, 312–315 photobleaching techniques, 311–312 time-lapse multispectral, and ratio imaging, 311 total internal reflection fluorescence and fluorescent speckle microscopy, 315–316 plant GFP imaging, 274–279 reef coral fluorescent protein expression and detection, 347–348 yeast GFP, 179–180 Fluorescence recovery after photobleaching (FRAP): C. elegans GFP visualization, 213 Drosophila GFP, 242–243 mammalian GFPs, 311–312 yeast GFP localization, 189–190 Fluorescence resonance energy transfer (FRET): drug assays, fluorescent proteins, 367–371 dual color spectroscopy assays, 371–372 target validation studies, 381–383 DsRed “greening,” 130–132 GFP mutations: aggregation modulation, 104–105

430

INDEX

Fluorescence resonance energy transfer (FRET) (continued) biosensors, 106–108 tandem concatenations, 110–111 mammalian cell GFP, 308–309 protein-protein interactions, 312–315 yeast GFP, 180, 192–193 Fluorescence speckle microscopy (FSM): mammalian cell GFP, 315–316 cytoskeleton analysis, 319–320 yeast GFP localization, 189–190 Fluorescent dissecting microscopy, transgenic vertebrate GFP, 289 Fluorescent lipid-associated reporters (FLAREs), yeast GFP lipid localization, 192 Fluorescent protein degradataion assays, 372 Fluorescent proteins: color diversity, anthoza GFP homologs, 122–124 kindling, 132–133 mutagenesis modification, 129–130 Fluorescent timer protein: DsRed mutant, 128 in mammalian cells, 308–309 reef coral fluorescent proteins, 345 Fluorophore properties: covalent mutation, 94–96 green fluorescent protein structure, 71–73 activation control, 76 mammalian GFPs, 324–325 Folding mechanisms: GFP expression protocols, 408 green fluorescent protein structure, 78–79 mutant improvements, 101–102 plant GFP, 270–272 Forespore-specific fluorescence, GFP spore formation, B. subtilis, 166–167 Frozen sectioning, transgenic vertebrate GFP, 289–290 FtsA protein, E. coli cell division, bacterial GFP and, 168 FtsZ protein, E. coli cell division, bacterial GFP and, 168 Functionality studies, yeast green fluorescent protein localization, 184–185 Fusion constructs: C. elegans GFP, creation of, 217–219 Drosophila GFP, cell biology and development, 232–243 chromosome structure, 234–235 fusion protein FRAP analysis, 242–243 meiosis and mitosis, 235–238

morphogen gradients, 239–240 neuronal function and connectivity, 238–239 organelle function and transport, 234 original studies, 232–233 tagged RNAs, 241–242 tissue morphogenesis, 233–234 transcription factors, 241 Drosophila GFP gene expression, 229 fluorescent protein degradataion assays, 372 GFP reassembly, 397 green fluorescent protein, 76–77 genetic applications, 172 GFP spore formation, B. subtilis, 166–167 reef coral fluorescent protein expression and detection, 346–347 yeast GFP, 186–194 integration of, 195 organelle structure, 187–188 protein localization, 188–190 b-galactosidase (bGAL): Drosophila GFP detection, 229–230 GFP vs., 204 GAL4 gene expression, plant GFP, 280–281 GAL1 promoter, yeast GFP gene expression, 193–194 GAL1-10 promoter, S. cerevisiae GFP gene expression, extrachromosomal constructs, 194–195 Gal4 system, Drosophila GFP gene expression, 228 GATA promoters, transgenic fish GFP, plasmid injection, 290–291 Gene batteries, C. elegans GFP as cell fate marker, 212 Gene expression: C. elegans GFP: co-labeling techniques, 213–215 pattern analysis, 204–205 Drosophila GFP, 228–231 detection timing and sensitivity, 229–231 transformation vectors, 228–229 transient expression, 229 mammalian cell GFP markers, 318 plant GFP, 263–264 modified genes, 274 reassembled GFP visualization, 401–402 reef coral fluorescent proteins (RCFPs), 346–357 flow cytometry, 351–353 single-color analysis, 353–355 three- and four-color analysis, 356

431

INDEX

two color analysis, 355 fluorescence microscopy techniques, 347–348 laser scanning, 357 multicolor analysis, 349–351 single-color analysis, 348–349 spectrophotemetry, 357 in transgenic vertebrates: levels of, 287 markers for patterns, 297 yeast green fluorescent protein, 180–186 assessment of, 195–196 C. albicans, 183–184 functional and proper localization, 184–185 genetic applications, 193–194 S. cerevisiae, 181–182, 194–195 S. pombe, 182–183 spectral variants and red fluorescent proteins, 185–186 Gene regulation, Drosophila GFP reporters, 244–247 Gene targeting, mammalian cell GFP, 318–319 Genetic applications: bacterial green fluorescent protein, 171–174 acid-inducible gene isolation, S. typhimurium, 172–173 flow cytometry, 172 macrophage-inducible gene isolation, S. typhimurium, 173–174 yeast GFP gene expression, 193–194 Genome sequencing studies: b-arrestin recruitment assay, 375–376 fluorescent protein applications, 380–383 G2F domains, functions, 147–149 GFP400, neutral phenol composition, 92 GFPA mutant, plant GFP, 270–273 gfp 10 gene, green fluorescent protein brightness, 54 gfp genes, prokaryotic GFP, 164–165 GFP-like proteins: classification, 139–140 functions: bioluminescent organisms, 140–141 copepoda species, 147 G2F domains, 147–149 nonbioluminescent organisms, 141–147 anthozoa: FP modification, mutagenesis, 129–130 homolog color diversity, 122–124 mutagenesis color transitions, 126–128 oligomeric state, 126 photoactivatable probes, 130–134

Dronpa, 134 DsRed “greening,” 130–132 fluorescent protein kindling, 132–134 UV-induced green-to-red photoconversion, 132 research background, 121–122 sequence comparisons, 124–126 corals, gene encoding, 142–143 hydroid medusae, 141–142 oligomerization, 146–147 photoprotection hypothesis, 143–144 phylogeny, 149–155 deep-level relationships, 149–151 ancient diversity, 154–155 anthoza proteins, 151 color diversity, 151–154 GFP5 mutant, plant GFP modification, 272–273 b-Glucuronidase (guaA) gene, plant GFP, 259–260 Glucocorticoid receptor (GR), nuclear translocation assays, fluorescent proteins, 364–365 Glutamate receptors, C. elegans GFP protein tagging, 215 Golgi apparatus, mammalian GFPs, 323–324 GPCR kinases (GRK), protein-protein interaction, bioluminescence resonance energy transfer assays, 369–371 G-protein-coupled receptors (GPCRs): b-arrestin recruitment assay, 372–376 fluorescent protein biosensors, 379–380 GFP fusion assays, 376–377 protein-protein interaction, bioluminescence resonance energy transfer assays, 369–371 target validation studies, 380–383 Green fluorescent protein (GFP). See also GFP-like proteins absorption/extinction cross-section, 57–59 monomer/dimer equilibrium, 57–59 barrel structure, 20 b-can structure, 69–71 biological function, 40–41 future research issues, 61–62 chromophore absorption suppression, molecular mechanism, 59–61 chromophore structure, 8–9 brightness properties, 54 molar extinction coefficient, 55–56 quantum yield, 54–55 spectroscopic analysis, 53–56

432

INDEX

Green fluorescent protein (GFP) (continued) in cnidarians and ctenophores, 19–25 color diversity, anthoza GFP homologs, 122–124 dimerization control, 77–78 discovery of, 1–4 dissection and reassembly mechanisms: kinetics and mechanism, 396–397 proteins and cells, 397–398 variants and topology, 393–396 isolation and properties, 4–8 jellyfish species, 9–11 molecular biology and mutation: aggregation modulation, 102–105 biosensors, 106–108 covalent alteration, Y66FHW, 94–96 energy level modification, 100–101 excitation peak ratios, durable photoisomerizatai, 96–100 halides, 109–110 insertion of other proteins, 110 mutational strategies, 85–87 nucleic acid changes, 112 nucleotide sequencing, 83–85 pH levels, 108–109 screening methods, 87–88 silent and loss-of-function mutations, 111–112 spectral mutant classification by chromophore, 89–94 imidazol derivation, Y66H, 93–94 indole derivation from Y66W, 93 neutral phenol, 92 phenolate anion, 89, 92 phenolate anionstacked pi-electron system, 92–93 phenyl derivation, Y66F, 94 wild-type neutral phenol/anionic phenolate mixture, 89 tandem concentrations, 110–111 temperature-dependent folding, 101–102 truncations, 105–106 natural sources, 41 pharmaceutical applications: drug screening assays, 363–380 b-arrestin recruitment assay, 372–376 BRET protein-protein interaction, 369–371 cell adhesion assays, 366 dual colour spectroscopy assays, 371–372 fluorescence-activated cell sorting, 378 fluorescent protein biosensors, 378–380

fluorescent protein degradation, 372 FRET/BRET assays, 367–371 FRET protease assays, 368 FRET protein-protein interaction, 368–369 GPCR/GFP fusion assays, 376–377 nuclear translocation assays, 364–365 protein complementation assays, 377–378 reporter gene assays, 366–367 high-content/high-throughput screening, 362–363 high content screening detection instrumentation, 380 target validation studies, 380–383 physical characteristics of, 41–52 denaturation and renaturation, 49–50 detergents and chaotropes, 52 fixatives and preservatives, 52 organic solvents, 51–52 protease effects, 50–51 stability, 48–49 three-dimensional structure: activation control, 76 Aequorea mutant, 73–75 fluorophore properties, 71–73 folding mechanisms, 78–79 future research issues, 79 research background, 67–68 spectral and physical properties, 68–69 structure-based engineering, 76 truncation and fusion constructs, 76–77 “Greening” of DsRed, 130–132 Halides, green fluorescent protein mutant structures, 109–110 HcRed mutant: color transition to, 128 reef coral fluorescent proteins, 345–346 three-color analysis, 351 stable cell populations and clone expression, 352–353 Hermes element, Drosophila GFP gene expression, 228–229 Heteractis crispa: HcRed1 mutant, 345–346 reef coral fluorescent proteins, 340–341 ZaGreen1 mutant, 344 Heterocyst formation, Anabaena GFP, 167–168 Heterodimeric proteins, bacterial luciferases, 26–28 Heteroduplex recombination, green fluorescent protein mutant structure, 87

433

INDEX

hetR gene expression, heterocyst formation in Anabaena sp., 167–168 High-content screening (HCS): b-arrestin recruitment assay, 373–376 detection instrumentation, 380 GFP/RCFP pharmaceutical applications, 362–363 drug screening assays, 363–364 GPCR/GFP fusion assays, 376–377 High-throughput microscopy (HTM), green fluorescent protein mutation, 88 High-throughput screening (HTS): b-arrestin recruitment assay, 373–376 fluorescent protein biosensors, 378–380 GFP/RCFP pharmaceutical applications, 362–363 drug screening assays, 363–364 Homologous recombination: Candida albicans GFP gene expression, 184 S. pombe GFP gene expression, 182–183 Homo-oligomer formation, reef coral fluorescent proteins, 340–341 Human immunodeficiency virus (HIV-1), mammalian cell GFP and pathogenesis, 319 Hydrogen bonding, green fluorescent mutants, 75 Hydroid medusa, GFP-like proteins, 141–142 Hydrophobic interactions, protein absorption suppression, 59–61 Imidazole, chromophore derivation, 93–94 Immunity, Drosophila GFP reporters, 245–247 Indole, cyan fluorescent proteins, chromophore derivation, 93 Injection markers, C. elegans GFP experiments, 215–216 Insertion mechanisms, green fluorescent protein mutations, 110 In situ hybridization, transgenic GFP expression patterns, 297 Intein fragments, GFP reassembly, 396–397 subcellular localization, 401 Internal ribosome entry sequence (IRES), transgenic chick GFP expression, 294 Inverse fluorescence recovery after photobleaching (iFRAP), mammalian cell GFPs, 312 In vivo imaging: bacterial GFP applications, 174 Candida albicans GFP gene expression, 183–184

transgenic GFP for, 298 yeast GFP, 180 fusion constructs, 186–194 spectral variants and red fluorescent proteins, 185–186 In vivo recombination, C. elegans fusion gene constructs, 219 Isoform structure, in green fluorescent proteins, 45–52 Jellyfish species, aequorin and GFP sources, 9–11 “Kaede” protein, ultraviolet-induced green-tored photoconversion, 132 Kindling fluorescent proteins (KFP), photoactivation, 132–133 Kinetic parameters, GFP reassembly, 396–397 Knock-in gene expression, transgenic mice GFP, 295–296 Kruppel gene, Drosophila GFP balancers, 247 Laser ablation, C. elegans GFP cell identification, 213 Laser scanning confocal microscopy, plant GFP imaging, 275–279 Laser-scanning detection, reef coral fluorescent protein expression and detection, 357 Leptoseris fragilis, color diversity, 145–146 Leucine zipper peptides, GFP dissection and reassembly, 393–398 protein-protein interactions, 398–400 Lipids, yeast GFP localization, 192 Localization studies: plant GFP, 264–265 protein localization: absorption suppression and, 59–61 C. elegans GFP determination, 206–208 mammalian cell GFP, 316–318 yeast GFP, 188–190 yeast green fluorescent protein, 188–192 DNA, 190–191 lipids, 192 protein, 188–190 RNA, 191–192 Loss-of-function muations, green fluorescent proteins, 111–112 Luciferases: in bacteria, 26–28 crystalline structure, 23–25 in dinoflagellates, 28–30 in fireflies, 30–33

434

INDEX

Luciferin binding protein (LBP), in dinoflagellates, 28–30 Luciferins, biochemistry, 18–19 Lumazine protein (LUMP), blue-shifted emission, 27–28 Luminescent potentials, green fluorescent protein species, 21–25 Luminous systems, biochemistry, 18–19 lux I gene, bacterial luciferases, 26–28 Lysosomes, mammalian GFPs, 323–324 Macromolecular localization, yeast GFP, 186–187 Macrophage-inducible gene isolation, S. typhimurium GFP, 173–174 Mammalian cells: green fluorescent protein in: applications, 316–319 co-transfection or expression markers, 318 protein localization, dynamics, and concentration, 316–318 viral infection and pathogenesis, 319 viral system targeting, 318–319 characteristics, 305–309 cytoskeleton illumination, 319–320 fixed specimen preparation, 417 flow cytometry, 319 fluorescence microscopy-based techniques, 309–316 FRET and FCS, 312–315 photobleaching techniques, 311–312 time-lapse multispectral, and ratio imaging, 311 total internal reflection fluorescence and fluorescent speckle microscopy, 315–316 future research issues, 324–325 membrane trafficking and organelle dynamics, 321–324 nucleus revelation, 320–321 reef coral fluorescent proteins, fluorescence microscopy/flow cytometry applications, 339–357 Maturation mechanisms, plant GFP, 269–272 Meganuclease, in transgenic fish GFP, 292 Meiosis, Drosophila GFP cell biology, 235–238 Membrane trafficking, mammalian GFPs, 321–324 Messenger RNA (mRNA): Arabidopsis gfp, cryptic splicing, 260–262

dinoflagellate luciferases, 29–30 Drosophila GFP, tagged RNA, 241–242 yeast GFP RNA localization, 192 MET3 promoter, S. cerevisiae GFP gene expression, 182 mfgp4-ER gene, plant GFP: subcellular targeting, 266–269 visualization techniques, 276–279 mfgp4 gene, plant GFP: expression, 263–264 localization, 264–265 maturation mechanisms, 270–272 subcellular targeting, 265–269 Mice, transgenic GFP in, 294–296 target validation studies, fluorescent proteins, 382–383 Micro-RNA (miRNA) analysis, Drosophila GFP reporters, 245–247 Microscopic techniques, yeast GFP analysis, 196 Microtubule dynamics, yeast GFP localization, 189–190 Mini-white gene, Drosophila GFP gene expression, 228–229 Mitochondrial GFP: Drosophila GFP organelles, 234 organelle structure, 187–188 Mitosis, Drosophila GFP cell biology, 235–238 Mitotic spindle analysis, yeast GFP localization, 188–190 Moesin fusion, Drosophila GFP, tissue morphogenesis, 233–234 Molar extinction coefficient, green fluorescent protein, 55–56 Molecular function, GFP spore formation, B. subtilis, 166–167 Molecular weight analysis: fluorescent proteins, mutagenic modification, 129–130 green fluorescent proteins, 45, 47 Monomer/dimer equilibrium, green fluorescent protein absorption/extinction crosssection, 57–59 Montastraea cavernosa: GFP-like proteins, gene coding, 142–144 phylogenetic relationships, 151–152 Morphogen gradients, Drosophila GFP cell biology, 239–240 Mosaic analysis, C. elegans GFP experiments, 216 Mosaic analysis with a repressible cell marker (MARCM), Drosophila GFP, 248–249

435

INDEX

Mos1 gene, Drosophila GFP gene expression, 228–229 mRFP1 protein, mutagenic modification, 129–130 mtrA gene expression, S. typhimurium GFP gene expression, 174 Multicolor analysis: GFP dissection and reassembly, 394–396 protein-protein interactions, 399–400 reef coral fluorescent proteins, 349–351 flow cytometry, 356 Multiphoton microscopy, transgenic vertebrate GFP, 289–290 Multispectral imaging, mammalian cell green fluorescent proteins, 311 Mushroom bodies (MB), Drosophila GFP cell biology, 238–239 Mutagenesis: fluorescent protein modification, 129–130 kindling fluorescent proteins, 132–133 nonbioluminescent anthozoa GFP-like proteins, color transitions, 126, 128 Mutant structures, green fluorescent proteins, 73–75 aggregation modulation, 102–105 biosensors, 106–108 covalent alteration, Y66FHW, 94–96 energy level modification, 100–101 excitation peak ratios, durable photoisomerization, 96–100 halides, 109–110 insertion of other proteins, 110 mutational strategies, 85–87 nucleic acid changes, 112 nucleotide sequencing, 83–85 pH levels, 108–109 screening methods, 87–88 silent and loss-of-function mutations, 111–112 spectral mutant classification by chromophore, 89–94 imidazol derivation, Y66H, 93–94 indole derivation from Y66W, 93 neutral phenol, 92 phenolate anion, 89, 92 phenolate anionstacked pi-electron system, 92–93 phenyl derivation, Y66F, 94 wild-type neutral phenol/anionic phenolate mixture, 89 structure-based engineering, 76 tandem concentrations, 110–111 temperature-dependent folding, 101–102 truncations, 105–106

Mycobacterial promoters, bacterial green fluorescent protein, genetic applications, 171–174 Mycobacteria sp., GFP-host interactions, 169 Natural sources of GFP, 41–42 Neuronal function: C. elegans GFP measurement, 212–213 Drosophila GFP cell biology, 238–239 transgenic GFP analysis, 298 Neutral phenol: in chromophore, 92 wild-type green fluorescent proteins, 89 NFkB activity: b-arrestin recruitment assay, 372–376 fluorescent protein degradation assays, 372 Nidogen, G2F domains, 147–149 Nitrogen fixation, heterocyst formation in Anabaena sp., 167–168 nmt1+ promoter, S. pombe GFP gene expression, 183 Nomarski optical analysis, C. elegans GFP visualization, cellular anatomy, 208–212 Nonbioluminescent organisms, GFP-like proteins: anthozoa: FP modification, mutagenesis, 129–130 homolog color diversity, 122–124 mutagenesis color transitions, 126–128 oligomeric state, 126 photoactivatable probes, 130–134 Dronpa, 134 DsRed “greening,” 130–132 fluorescent protein kindling, 132–134 UV-induced green-to-red photoconversion, 132 research background, 121–122 sequence comparisons, 124–126 corals, gene encoding, 142–143 hydroid medusae, 141–142 oligomerization, 146–147 photoprotection hypothesis, 143–144 NOP1 promoter, S. cerevisiae GFP gene expression, 182 N-terminal domain: in dinoflagellate luciferase, 29–30 in firefly luciferase, 31–33 GFP truncation and fusion constructs, 76–77 mutant structures, 105–106 Nuclear architecture and dynamics, mammalian cell GFP, 320–321

436

INDEX

Nuclear-localized GFP (nlsGFP), Drosophila GFP detection, 230 Nuclear pore complexes (NPCs), yeast GFP gene expression, 193–194 Nuclear translocation assays, fluorescent proteins, 364–365 Nucleic acid sequences, green fluorescent protein mutations, 112 Nucleocytoplasmic transport, yeast GFP localization, 188–190 Obelia geniculata, green fluorescent protein in, 19–25 Oligomerization: GFP-like proteins, 146–147 GFP mutations, aggregation modulation, 104–105 nonbioluminescent anthozoa GFP-like proteins, 126 mutagenic modification, 129–130 “150% recovery,” organic solvents on green fluorescent proteins, 51–52 Open reading frame (ORF) technique: S. cerevisiae GFP gene expression, 181–182 yeast GFP gene expression, 195–196 Operons, bacterial luciferases, 26–28 Organelle structure: Drosophila GFP, 234 mammalian GFPs, 321–324 yeast GFP, 187–188 Organic solvents, green fluorescent protein, 51–52 absortpion/excitation spectral shifts, 60–61 Organism structure, in transgenic vertebrates, GFP expression, 287–288 Organized smooth endoplasmic reticulum (OSER), mammalian GFPs, 322–324 Orphan GPCRs, b-arrestin recruitment assay, 375–376 P. fluorescens: GFP-host interactions, 169–170 GFP isolation, 165 Pathway mapping, fluorescent proteins, 381–383 P-element transformation system, Drosophila GFP gene expression, 228–231 Pelican vectors, Drosophila GFP gene expression, 228–229 Peroxy flavin, lifetime, 26 pGALI-10 promoter, S. cerevisiae GFP gene expression, 182

Pharmaceutical applications, fluorescent proteins: drug screening assays, 363–380 b-arrestin recruitment assay, 372–376 BRET protein-protein interaction, 369–371 cell adhesion assays, 366 dual colour spectroscopy assays, 371–372 fluorescence-activated cell sorting, 378 fluorescent protein biosensors, 378–380 fluorescent protein degradation, 372 FRET/BRET assays, 367–371 FRET protease assays, 368 FRET protein-protein interaction, 368–369 GPCR/GFP fusion assays, 376–377 nuclear translocation assays, 364–365 protein complementation assays, 377–378 reporter gene assays, 366–367 high-content/high-throughput screening, 362–363 high content screening detection instrumentation, 380 target validation studies, 380–383 Phenolate anion: stacked p-electron system, 92–93 wild-type green fluorescent proteins: chromophore structures, 89, 92 neutral phenols, 89 Phenyl, chromophore derivation, 94 Phialidium sp.: GFP-like proteins, 141–142 green fluorescent protein structure, 45, 47–48 pH levels, green fluorescent protein structure, 45, 48 mutant structures, 108–109 specimen preparation, 418–419 Phosphatidylinositol (PtdIns), yeast GFP lipid localization, 192 Photoactivatable green fluorescent proteins: in mammalian cells, 308–309 nonbioluminescent anthozoa GFP-like proteins, 130–134 Dronpa, 134 DsRed “greening,” 130–132 fluorescent protein kindling, 132–134 UV-induced green-to-red photoconversion, 132 specimen preparation, 418–419 Photobleaching: Drosophila GFP, FRAP analysis, fusion proteins, 242–243

437

INDEX

GFP specimen preparation, 418–419 mammalian GFPs: kinetic analysis, 311–312 nuclear envelope and, 321 variant populations, 308–309 yeast GFP localization, 189–190 Photocyte dispersions: firefly luminescence, 30–33 green fluorescent protein species, 21–25 Photodamage, GFP specimen preparation, 418–419 Photoisomerization, green fluorescent protein mutant structures, 96–100 Photoprotection hypothesis, GFP-like proteins, 143–144 Photoprotein energy transfer, green fluorescent protein, 4–8 Photoreception, GFP-like proteins, 145–146 Phrixothrix, bioluminescence in, 30–33 Physiological processes, C. elegans GFP visualization, 213 PiggyBac element, Drosophila GFP gene expression, 228–229 Plants: GFP-host interactions, 169–170 green fluorescent proteins in: arabidopsis, cryptic splicing of gfp mRNA, 260–262 cell imaging, 274–276 cell marking, 280–281 cryptic intron removal, 262–263 gene expression, 263–264 modification, 274 localization, 264–265 maturation, 269–272 spectral modification, 272–273 subcellular dynamics visualization, 276–279 subcellular targeting, 265–269 Plasmid injection, transgenic fish GFP, 290–291 Plasmid-swap experiment, yeast green fluorescent protein localization, 188–192 Plasmid vectors, S. cerevisiae GFP gene expression, extrachromosomal constructs, 194–195 Plasticity studies, C. elegans GFP measurement, 212–213 Plextrin homology domain, yeast GFP lipid localization, 192 Pocillopora damicornis, GFP-like proteins, photoprotection hypothesis, 144

Polar follicle cells, Drosophila GFP, tissue morphogenesis, 233–234 Polymerase chain reaction (PCR): Arabidopsis gfp mRNA, cryptic splicing, 260–262 C. elegans fusion gene constructs, 217–218 Candida albicans GFP gene expression, 184 green fluorescent protein structure, 68 error-prone mutational strategies, 85–86 S. cerevisiae GFP gene expression, 181–182 Polymorphism, GFP-like proteins, gene coding, 142–143 Position effect variegation, Drosophila GFP reporters, 244–247 Potassium chloride, in Aequorea aequorea, 1–4 Preservatives, green fluorescent protein and, 52 Prokaryotes, green fluorescent protein in: bacterial development and cell biology, 165–168 cell division, E. coli, 168 heterocyst formation, Anabaena, 167–168 spore formation, B. subtilis, 165–167 bacterial ecology and behavior, 170–171 bacterial-host interactions, 169–170 complex environments, 168–171 future research issues, 174–175 genetic applications, 171–174 acid-inducible gene isolation, S. typhimurium, 172–173 flow cytometry, 172 macrophage-inducible gene isolation, S. typhimurium, 173–174 research background, 163–165 Protease assays: fluorescence resonance energy transfer assays, 368 in green fluorescent proteins, 45–48 physical effects of, 50 tandem concatenations, 111 Protein complementation assays, fluorescent protein screening, 377–378 Protein dissection, reassambled green fluorescent proteins, 392–393 Protein disulfide isomerase (PDI), Drosophila GFP, organelle fusion and transport, 234 Protein function, transgenic GFP analysis, 298 Protein localization: absorption suppression and, 59–61 C. elegans GFP determination, 206–208 mammalian cell GFP, 316–318 yeast GFP, 188–190

438

INDEX

Protein-protein interactions: bioluminescence resonance energy transfer assays, 369–371 dissected proteins, 392–393 fluorescence resonance energy transfer assays, 368–39 GFP reassembly, 398–400 unknown interactions, 400 mammalian cell GFP, FRET/FCS imaging, 312–315 Protein trap screens, Drosophila GFP, 250–251 Protocols for GFP expression: folding and temperature sensitivity, 408 purification, 408–413 specimen preparation, 413–419 fixed specimens, 413–418 photobleaching, photoactivation, photodamage, and pH dependence, 418–419 toxicity studies, 407–408 Proton transfer, green fluorescent protein mutant structures, 74–75 Purification protocols, GFP expression, 408–413 Pyrophorus plagiophthalamus, firefly luciferase, 32–33 Quantum mechanical modeling, green fluorescent protein structure, 68–69 Quantum yield of bioluminescence (Qbl), green fluorescent protein, 7–8 spectroscopic analysis, 54–55 Quorum sensing, bacterial luciferases, 27–28 Radiationless energy transfer, green fluorescent protein, 40–41 Random mutagenesis: DsRed/DsRed-Express mutants, 344 green fluorescent protein mutant structures, 85 amino acid predetermination, 86 nonbioluminescent anthozoa GFP-like protein color transitions, 126, 12800 Ratio imaging, mammalian cell green fluorescent proteins, 311 Reassambled green fluorescent proteins: applications, 398–402 protein expression patterns, 401–402 protein-protein interactions, 398–400 subcellular localization, 400–401 dissection and reassembly mechanisms: kinetics and mechanism, 396–397 proteins and cells, 397–398

variants and topology, 393–396 future research issues, 402–403 protein dissection, 392–393 Recombinant proteins: absorption suppression, 60–61 denaturation/renaturation, 49–50 dinoflagellate luciferases, 29–30 green fluorescent protein mutation, heteroduplex recombination, 87 monomer/dimer equilibrium, 58–59 S. cerevisiae GFP gene expression, 181–182 S. pombe GFP gene expression, 182–183 structural characteristics, 41–42, 44–52 Recombination cloning, C. elegans fusion gene constructs, 219 Red fluorescent proteins (RFPs): color diversity, anthoza GFP homologs, 122–124 mutagenesis, 128000 phylogenetic relationships, 152–153 yeast green fluorescent protein and, 185–186 Red-shifted emissions: bacterial luciferases, 27–28 green fluorescent protein mutant structures, wavelength altered ratios, 98–100 Red tides, dinoflagellates in, 28–30 Reef coral fluorescent proteins (RCFPs): basic properties of, 340–346 AmCyan 1 mutant, 341 AsRed2 mutant, 345 DsRed2/DsRed-express, 344 DsRed monomer, 345 fluorescent timer (E5), 345 HcRed1 mutant, 345–346 red fluorescent proteins, 343–344 ZsGreen 1 mutant, 341–343 ZsYellow1 mutant, 343 gene expression and detection, 346–357 flow cytometry, 351–353 single-color analysis, 353–355 three- and four-color analysis, 356 two color analysis, 355 fluorescence microscopy techniques, 347–348 laser scanning, 357 multicolor analysis, 349–351 single-color analysis, 348–349 spectrophotemetry, 357 GFP-like proteins, gene coding, 142–143 in mammalian cells, fluorescence microscopy/flow cytometry applications, 339–357 nomenclature, 340

439

INDEX

pharmaceutical applications: drug screening assays, 363–380 b-arrestin recruitment assay, 372–376 BRET protein-protein interaction, 369–371 cell adhesion assays, 366 dual colour spectroscopy assays, 371–372 fluorescence-activated cell sorting, 378 fluorescent protein biosensors, 378–380 fluorescent protein degradation, 372 FRET/BRET assays, 367–371 FRET protease assays, 368 FRET protein-protein interaction, 368–369 GPCR/GFP fusion assays, 376–377 nuclear translocation assays, 364–365 protein complementation assays, 377–378 reporter gene assays, 366–367 high-content/high-throughput screening, 362–363 high content screening detection instrumentation, 380 target validation studies, 380–383 phylogenetic relationships, 151–153 Regulatory sequences, C. elegans GFP decoding, 205–206 Renaturation, in green fluorescent protein, 49–50 Renilla: biochemistry, 19 GFP-like proteins, 140–1410 green fluorescent protein in, 7–8 biological function, 40–41 chromophore structure, 53–54 denaturation/renaturation, 49–50 fluorophore formation, 72–73 molar extinction coefficient, 56 organic solvents, 51–52 physical characteristics, 41–52 protease resistance in, 50–51 stability properties, 48–49 Reporter genes: C. elegans GFP decoding, regulatory sequences, 205–206 C. elegans GFP gene constructs, 217–219 C. elegans GFP pattern analysis, 204–205 C. elegans GFP protein localization, 206–208 cell fate markers, C. elegans GFP, 212

cellular anatomy visualization, C. elegans GFP, 211–212 Drosophila GFP, 244–247 fluorescent protein drug assays, 366–367 mammalian GFPs, 324–325 Retrotransposons, Drosophila GFP chromosomes, 234–235 Reverse transcriptase polymerase chain reaction (RT-PCR), Arabidopsis gfp mRNA, cryptic splicing, 260–262 Rhizobium sp., GFP-host interaction, 170 Ricordea proteins, phylogenetic relationships, 152–153 RNA-binding proteins, yeast GFP localization, 191–192 RNAi, Drosophila GFP reporters, 246–247 RNA localization, yeast GFP, 191–192 Rous sarcoma virus (RSV), transgenic chick GFP expression, 294 “RSGFP4” mutant, green fluorescent protein, 86 Saccharomyces cerevisiae: fixed specimen preparation, 414–415 green fluorescent protein in, 179–197 applications, 186–194 construct integration, 195 extrachromosomal (plasmid) constructs, 194–195 fluorescence resonance energy transfer, 192–193 gene expression systems, 180–186, 194–195 assessment of, 195–196 functional and proper localization, 184–185 S. cerevisiae, 181–182, 194–195 spectral variants and red fluorescent proteins, 185–186 genetic applications, 193–194 localization studies, 188–192 DNA, 190–191 lipids, 192 protein, 188–190 RNA, 191–192 microscopic analysis, 196 organelle structure, function, and inheritance, 187–188 Saccharomyces pombe: DNA localization, 191 fixed specimen preparation, 415 green fluorescent protein in, 179–180 gene expression systems, 182–183

440

INDEX

Salmonella typhimurium: acid-inducible gene isolation, 172–173 GFP-host interactions, 169 macrophage-inducible gene isolation, 173–174 Sapphire chromophores, neutral phenol, 92 Schizosaccharomyces pombe, fixed specimen preparation, 415–416 Scintillon fractionation, in dinoflagellates, 29–30 Screening methods: Drosophila GFP, 250–251 green fluorescent protein mutation, 87–88 Sequence comparison, nonbioluminescent anthozoa GFP-like proteins, 124–126 Ser-Tyr-Gly sequence, green fluorescent protein structure, 71–73 sg100 variant, GFP dissection and reassembly, 393–396 Sigma factors, GFP spore formation, B. subtilis, 165 Signature-tagged transposition, bacterial GFP, 175 Silent mutations, green fluorescent proteins, 111–112 Single-color analysis, reef coral fluorescent protein expression and detection, 348–349 Single-copy fusions, prokaryotic GFP, 165 Singlet oxygen, green fluorescent protein fluorophore formation, 72–73 Site-directed mutations: DsRed/DsRed-Express mutants, 344 green fluorescent protein structures, 86 nonbioluminescent anthozoa GFP-like protein color transitions, 126, 128 Sleeping Beauty transposon, in transgenic fish GFP, 292–293 Sodium dodecyl sulfate (SDS) gel electrophoresis, recombinant green fluorescent proteins, 45–46 Somatic nuclear transfer, in transgenic fish GFP, 293 Specimen preparation protocols, fixed GFP specimens, 413–418 Spectral emissions: green fluorescent protein structure, 68–69 photoproteins in coelentrates and, 20–25 plant GFP modification, 272–273 yeast green fluorescent protein, 185–186 Spectral mutants, green fluorescent protein, chromophore classification, 89–94

imidazol derivation, Y66H, 93–94 indole derivation from Y66W, 93 neutral phenol, 92 phenolate anion, 89, 92 phenolate anionstacked p-electron system, 92–93 phenyl derivation, Y66F, 94 wild-type neutral phenol/anionic phenolate mixture, 89 Spectrofluorimetric measurments, bacterial green fluorescent protein, genetic applications, 171–174 Spectroscopic analysis: green fluorescent protein chromophores, 53–56 brightness properties, 54 molar extinction coefficient, 55–56 quantum yield, 54–55 green fluorescent protein mutant structures, wavelength altered ratios, 98–100 Sperm nuclear transplantation, transgenic Xenopus GFP, 293–294 Spindle pole body (SPB): Drosophila GFP cell biology, 235–238 yeast GFP localization, 189 Spore formation, B. subtilis green fluorescent protein, 165–167 Stability properties, green fluorescent protein, 48–49 Stable cell populations, reef coral fluorescent protein expression and detection, 352–353 Stacked pi-electron system, phenolate anions, 92–93 Stress-response transcription factors, yeast GFP localization, 188–190 Structure-based engineering, green fluorescent protein, 76 Subcellular targeting: plant GFP, 265–269 visualization techniques, 276–279 reassambled GFPs, 400–401 “Superfolder” GFP variant, dissection and reassembly, 395–396 Synaptogyrin (SNG-1), C. elegans GFP protein localization, 207–208 Tagged proteins: C. elegans GFP, 215 Drosophila GFP, RNA, 241–242 Tandem concatenations, green fluoresent proteins, 110–111

441

INDEX

Tandem repeats, dinoflagellate luciferases, 29–30 Target validation studies, fluorescent protein applications, 380–383 Telomere formation, Drosophila GFP chromosomes, 234–235 Temperature loss (Tm): green fluorescent protein denaturation/renaturation, 49–50 protein folding, GFP mutants, 101–102 Temperature sensitivity: GFP expression protocols, 408 prokaryotic GFP, 164–165 Tetrameric structures, reef coral fluorescent proteins, 341 Tetratricopeptiderepeat (TPR) domains, GFP dissection and reassembly, 399–400 Three-dimensional structure, green fluorescent protein: activation control, 76 Aequorea mutant, 73–75 fluorophore properties, 71–73 folding mechanisms, 78–79 future research issues, 79 research background, 67–68 spectral and physical properties, 68–69 structure-based engineering, 76 truncation and fusion constructs, 76–77 Time-lapse imaging, mammalian cell green fluorescent proteins, 311 Timing issues, Drosophila GFP detection, 230 Tissue morphogenesis: Drosophila GFP, 233–234 in transgenic vertebrates, GFP expression, 287–288 Tobacco: firefly luciferase in, 31–33 GFP expression in, 263–264 Tol2 transposon, in transgenic fish GFP, 292–293 Topological analysis, GFP dissection and reassembly, 393–396 Total internal reflection fluorescence microscopy (TIR-FM), mammalian cell GFP, 315–316 Toxicity studies, GFP expression protocols, 407–408 Transcription activators, plant GFP marking, 280–281 Transcription factors: bioluminescence resonance energy transfer assays, 370–371 Drosophila GFP, 241

translocation assays, fluorescent proteins, 364–365 Transformation vectors, Drosophila GFP gene expression, 228–229 Transgenic lines, C. elegans fusion gene constructs, 219 Transgenic vertebrates, green fluorescent protein in: cell markers, 296–297 chicks, 294 coexpression markers, 296 definitions, 286 enhancer analysis, 297–298 expression levels, 287 fish species, 290–293 BAC injection, 292 meganuclease, 292 plasmid injection, 290–291 somatic nuclear transfer, 293 transposons, 292–293 gene expression markers, 297 mice, 294–296 DNA injection, 294–295 knock-in species, 295–296 techniques, 286–290 tissue and organism, 287–288 variants, 288–289 visualization methods, 289–290 in vivo protein, fusion constructs, 298 Xenopus species, 293–294 Transient gene expression, Drosophila GFP, 229 Transposons, in transgenic fish GFP, 292–293 Truncation constructs, green fluorescent protein, 76–77 mutant structures, 105–106 Tubulin binding, Drosophila GFP cell biology, 235–238 Turbo (T-sapphire) mutation, durable photoisomerization, 97–100 Two-color analysis: drug assays, fluorescent proteins, 371–372 reef coral fluorescent proteins, 349–351 flow cytometry, 355 T203Y GFP mutant, energy level modification, 100–101 Ultraviolet-induced green-to-red photoconversion, “Kaede” protein, 132 Upstream activator sequence (UAS), Drosophila GFP gene expression, 228

442

INDEX

vab-7 reporter gene, C. elegans GFP pattern analysis, 205 Variant green fluorescent proteins: mammalian cells, 305–309 in transgenic vertebrates, 288–289 Venus YFP variants: in mammalian cells, 308–309 in transgenic vertebrates, 288–289 Vibratome sectioning, transgenic vertebrate GFP, 289–290 Video imaging, green fluorescent protein mutation, 88 Viral infection: mammalian cell GFP and pathogenesis, 319 transgenic chick GFP, 294 Viral systems, mammalian cell GFP gene targeting, 318–319 Visualization techniques: C. elegans GFP, 219 plant GFP, 274–276 reassembled GFP expression patterns, 401–402 in transgenic vertebrate GFP, 289–290 Visual screening procedures, green fluorescent protein mutation, 87–88 Wavelength excitation: green fluorescent protein mutations, altered ratios, 96–100 organic solvents on green fluorescent proteins, 52 Wee-P trap, Drosophila GFP, 251 Wild-type neutral phenol/anionic phenolate, spectral properties, 89 Xenopus species, transgenic GFP in, 293–294 Y. pseudotuberculosis, GFP-host interactions, 169 Yeasts: green fluorescent protein in, 179–197 applications, 186–194 construct integration, 195 extrachromosomal (plasmid) S. cerevisiae constructs, 194–195 fixed specimen preparation, 414–416 fluorescence resonance energy transfer, 192–193 gene expression systems, 180–186 assessment of, 195–196 C. albicans, 183–184 functional and proper localization, 184–185

S. cerevisiae, 181–182, 194–195 S. pombe, 182–183 spectral variants and red fluorescent proteins, 185–186 genetic applications, 193–194 localization studies, 188–192 DNA, 190–191 lipids, 192 protein, 188–190 RNA, 191–192 microscopic analysis, 196 organelle structure, function, and inheritance, 187–188 transgenic fish GFP, 292 Yeast two-hybrid (Y2H) screening, protein dissection, 392–393 yEGFP variant, Candida albicans GFP gene expression, 183–184 Yellow fluorescent protein (YFP): aggregation modulation, 105 in bacteria, 25–28 C. elegans variants, 216–217 Candida albicans GFP gene expression, 183–184 color diversity, anthoza GFP homologs, 122–124 mutagenesis, 128 dissection and reassembly, 394–396 drug assays, 367–371 durable photoisomerization, 96–100 flavin mononucleotide (FMN), 27–28 halide mutations, 109–110 in mammalian cells, 308–309 neuronal function and plasticity studies, 212–213 phenolate anion with stacked pi-electron system, 92–93 protease assays, 368 S. cerevisiae GFP gene expression, 181–182 S. pombe GFP gene expression, 183 in transgenic vertebrates, 288–289 yeast GFP localization, 189–190 fluorescence resonance energy transfer, 192–193 Y66F chromophore, phenyl derivation, 94 Y66FHW, covalent mutations, 94–96 Y66H chromophore: covalent mutation of fluorophore, 95–96 imidazole derivation, 93–94 Y66W chromophore, indole derivation, 93

443

INDEX

Zoantharia orders: GFP-like proteins, 154–155 reef coral fluorescent proteins, 340–341 ZaGreen1 mutant, 341, 343 Zooxanthellae: color diversity, 144–145 GFP-like proteins, photoprotection hypothesis, 144

ZsGreen1 mutant: basic properties, 341, 343 flow cytometry, single-color analysis, 354 ZsYellow1 mutant: basic properties, 343 flow cytometry, single-color analysis, 354 multicolor analysis, 349–351