Regulation of Transcription in Plants Edited by KLAUS D. GRASSER Department of Life Sciences Aalborg University Aalborg,...
51 downloads
1347 Views
2MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Regulation of Transcription in Plants Edited by KLAUS D. GRASSER Department of Life Sciences Aalborg University Aalborg, Denmark
Regulation of Transcription in Plants
Regulation of Transcription in Plants Edited by KLAUS D. GRASSER Department of Life Sciences Aalborg University Aalborg, Denmark
C
2006 by Blackwell Publishing Ltd
Editorial Offices: Blackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK Tel: +44 (0)1865 776868 Blackwell Publishing Professional, 2121 State Avenue, Ames, Iowa 50014-8300, USA Tel: +1 515 292 0140 Blackwell Publishing Asia Pty Ltd, 550 Swanston Street, Carlton, Victoria 3053, Australia Tel: +61 (0)3 8359 1011 The right of the Author to be identified as the Author of this Work has been asserted in accordance with the Copyright, Designs and Patents Act 1988. All rights reserved. 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 or otherwise, except as permitted by the UK Copyright, Designs and Patents Act 1988, without the prior permission of the publisher. First published 2006 by Blackwell Publishing Ltd ISBN-13: 978-14051-4528-2 ISBN-10: 1-4051-4528-5 Library of Congress Cataloging-in-Publication Data Grasser, Klaus (Klaus D.) Regulation of transcription in plants / Klaus Grasser. – 1st ed. p. cm. Includes bibliographical references and index. ISBN-13: 978-1-4051-4528-2 (hardback : alk. paper) ISBN-10: 1-4051-4528-5 (hardback : alk. paper) 1. plant genetic regulation. I. title. QK981.4G73 2006 572.8 845—dc22 2006012521 A catalogue record for this title is available from the British Library Set in 10/12 pt Times by TechBooks, New Delhi, India Printed and bound in Singapore by Markono Print Media Pte. Ltd The publisher’s policy is to use permanent paper from mills that operate a sustainable forestry policy, and which has been manufactured from pulp processed using acid-free and elementary chlorine-free practices. Furthermore, the publisher ensures that the text paper and cover board used have met acceptable environmental accreditation standards. For further information on Blackwell Publishing, visit our website: www.blackwellpublishing.com
Contents Contributors Preface 1
2
General transcription factors and the core promoter: ancient roots WILLIAM B. GURLEY, KEVIN O’GRADY, EVA CZARNECKA-VERNER and SHAI J. LAWIT 1.1 Introduction 1.2 Origins of the eukaryotic promoter 1.3 Organization of the eukaryotic promoter 1.3.1 TATA-less promoters 1.4 Transcription factor IID 1.5 Role of TFIID in development 1.6 Mediator 1.6.1 Tail module 1.6.2 Middle complex 1.6.3 Head module 1.6.4 CDK8/Srb8-11 module 1.6.5 Mediator subunits unique to metazoans and plants 1.7 Transcription factor IIB 1.8 Summary References Transcription factors of Arabidopsis and rice: a genomic perspective JOSE´ LUIS RIECHMANN 2.1 Introduction 2.2 Arabidopsis and rice genomes: the angiosperm complement of transcription factors 2.2.1 General considerations for genome-wide analyses 2.2.2 Arabidopsis transcription factors 2.2.3 Rice transcription factors: a comparison to Arabidopsis 2.2.4 Gene duplications, functional redundancy, and the transcription factor phenome 2.3 Plant promoters References
xiii xv
1
1 1 4 8 8 9 11 12 15 16 16 17 18 19 21
28 28 29 29 30 35 38 40 43
vi 3
4
CONTENTS
Chromatin-associated architectural HMGA and HMGB proteins assist transcription factor function KLAUS D. GRASSER and DORTE LAUNHOLT 3.1 Introduction 3.2 HMGA proteins 3.2.1 Structure and expression 3.2.2 DNA and chromatin interactions 3.3 HMGB proteins 3.3.1 Structure and expression 3.3.2 DNA and chromatin interactions 3.4 Dynamic interaction of histone H1 and HMG proteins with chromatin 3.5 HMGA and HMGB proteins as architectural assistant factors Acknowledgements References Histone modifications and transcription in plants YII LENG CHUA and JOHN C. GRAY 4.1 Introduction 4.2 Histone acetylation and transcriptional activation 4.2.1 Plant histone acetyltransferases 4.2.1.1 GNAT/MYST family 4.2.1.2 TAFII 250 family 4.2.1.3 P300/CBP family 4.2.2 Bromodomain proteins 4.2.2.1 Bromodomain extra-terminal proteins 4.2.2.2 Plant bromodomain proteins 4.2.3 Plant histone deacetylases 4.2.3.1 RPD3/HDA1 family 4.2.3.2 HD2 family 4.2.3.3 SIR2 family 4.2.4 Histone acetylation/deacetylation and environmental adaptation 4.3 Histone methylation 4.3.1 Plant SET-domain proteins 4.3.1.1 E(Z)- and Trithorax-type HMTases 4.3.1.2 Su(var)3-9-type HMTases 4.3.1.3 ASH1-type and plant-specific HMTases 4.3.2 Histone demethylase 4.4 Interplay between histone acetylation and methylation in transcriptional regulation 4.5 Conclusions References
54 54 55 55 57 59 59 64 66 67 72 72 79 79 83 85 86 87 87 88 89 90 90 90 92 92 93 93 95 96 97 99 99 100 102 103
CONTENTS
5
6
Chromatin remodeling and histone variants in transcriptional regulation and in maintaining DNA methylation J.C. REYES, J. BRZESKI and A. JERZMANOWSKI 5.1 Introduction 5.2 ATP-dependent chromatin remodeling 5.2.1 SWI/SNF-like complexes in plants 5.2.2 Other ATPases of the SNF2 family that control plant development 5.3 Chromatin remodeling and DNA methylation 5.3.1 The effects of ddm1 mutation differ from those caused by met1 5.3.2 The model of DDM1 action 5.3.3 DDM1 and methylation of histone H3 5.3.4 Models of DDM1 targeting 5.3.5 DRD1 – another SNF2 family protein involved in control of DNA methylation 5.4 Histone variants in the regulation of chromatin functions 5.4.1 Occurrence and functional role of core histone variants 5.4.2 Role of H1 histone in chromatin and significance of nonallelic H1 variants References Matrix attachment regions and transcriptional gene silencing WILLIAM F. THOMPSON, STEVEN SPIKER and GEORGE C. ALLEN 6.1 Introduction 6.2 Costs and consequences of transgene expression variation 6.3 Position effects 6.4 Gene silencing 6.4.1 Post-transcriptional gene silencing 6.4.2 Transcriptional gene silencing 6.4.3 Repeats and rearrangement effects 6.5 Matrix attachment regions 6.6 MARs and transgene expression 6.6.1 MARs reduce gene silencing 6.6.2 MARs have a dual effect 6.6.3 Hypotheses to explain MAR effects 6.7 MAR effects in Arabidopsis 6.7.1 Testing the effect of developmental stage 6.7.2 How might MARs decrease net transgene expression? 6.7.3 MAR effects in PTGS mutants 6.8 Conclusions References
vii
112 112 113 115 119 120 121 121 123 124 124 125 125 128 130 136
136 137 137 138 138 139 140 140 142 142 148 149 151 152 152 153 153 153
viii 7
8
9
CONTENTS
Polymerase I transcription ´ ´ JULIO SAEZ-V ASQUEZ and MANUEL ECHEVERR´IA 7.1 Introduction 7.2 Organization of rRNA genes 7.3 Ribosomal RNA transcription and the nucleolus 7.4 rRNA gene promoters and transcription factors 7.5 RNA pol I holoenzymes: evidence and controversies 7.6 Coupling rDNA transcription and processing of pre-rRNA 7.7 Growth and hormonal control of RNA pol I transcription 7.8 Nucleolar dominance 7.9 Final remarks Acknowledgements References Transcription of plastid genes ¨ KARSTEN LIERE and THOMAS BORNER 8.1 Introduction 8.2 RNA polymerases 8.2.1 NEP: Nuclear-encoded RNA polymerase 8.2.2 PEP: Plastid-encoded RNA polymerase 8.3 Plastidial promoters 8.3.1 PEP promoters 8.3.2 NEP promoters 8.3.3 tRNAs with internal promoters 8.4 Regulation 8.4.1 Role of diverse and multiple promoters in developmental regulation of gene expression 8.4.2 Transcription factors for selective promoter recognition 8.4.2.1 Nuclear-encoded plastidial σ -factors 8.4.2.2 Role of σ -factor diversity in transcriptional regulation 8.4.2.3 NEP transcription factors 8.4.3 Effects of environmental factors and crosstalk between plastids and nucleus References Control of flowering time STEVEN VAN NOCKER and MARIA JULISSA EK-RAMOS 9.1 Introduction 9.2 Regulation of FLC expression through the ‘autonomous pathway’ 9.3 Chromatin-related pleiotropic regulators of FLC 9.4 Vernalization-associated repression of FLC 9.5 Transcriptional repression of flowering by FLC
162 162 162 163 166 169 171 172 173 176 176 177 184 184 185 187 188 189 189 190 192 192 193 196 196 198 203 204 209 225 225 225 229 232 235
CONTENTS
Transcriptional regulation in the photoperiodic induction of flowering 9.7 Activation of SOC1 by CO 9.8 Chromatin-related mechanisms of photoperiod pathway regulation 9.9 Transcriptional activation of AP1 by FT and FD 9.10 Transcriptional mechanisms in the promotion of flowering by GAs 9.11 PcG-mediated repression of floral homeotic genes 9.12 Summary and prospects Acknowledgements References
ix
9.6
10
11
12
Combinatorial control of floral organ identity by MADS-domain transcription factors ¨ GUNTER THEIßEN and RAINER MELZER 10.1 Introduction 10.2 ABC: early genetic models of floral organ identity 10.3 From the ABC to the ABCDE model 10.4 Elective affinities of homeotic proteins: the floral quartet model 10.5 Beyond the floral quartets sensu stricto: multimeric complexes of other MIKC-type proteins 10.6 Outlook: the devil is always in the details – what we still don’t know about floral quartets 10.7 Summary Acknowledgements References Networks of transcriptional regulation underlying plant defense responses toward phytopathogens IMRE E. SOMSSICH 11.1 Introduction 11.2 Defense transcriptome 11.3 Regulatory sequences mediating defense gene expression 11.4 Transcription factors involved in defense gene regulation 11.5 Transcriptional networks 11.6 Defense gene suppression by pathogens 11.7 Summary and outlook Acknowledgements References Temperature-regulated gene expression ¨ FRIEDRICH SCHOFFL and TRESSA JACOB PANIKULANGARA 12.1 Heat stress
236 237 238 240 242 243 244 244 244
253 253 253 256 257 260 261 263 263 263
266 266 267 268 270 275 277 277 279 279 285 285
x
CONTENTS
12.1.1 Introduction 12.1.2 Heat stress sensing and signaling 12.1.3 Heat stress transcription factors: structure and function 12.1.4 Class A HSF: immediate early regulators of the heat-shock response 12.1.5 Second-level regulation 12.1.6 Heat stress target genes 12.1.7 Common stress-response genes and pathways 12.2 Cold stress 12.2.1 Introduction 12.2.2 Low-temperature sensing and signaling 12.2.3 Transcription factors in ABA-independent pathways 12.2.4 Transcription factors in ABA-dependent pathways 12.2.5 Cold stress target genes 12.3 Perspectives for plant biotechnology References 13
14
Applications of inducible transcription in plant research and biotechnology BRIAN TOMSETT, ANGELA TREGOVA and MARK CADDICK 13.1 Introduction 13.2 Conditional expression of transgenes 13.2.1 The principles of regulated expression 13.2.2 Generic mechanisms 13.2.3 Physical stimuli for activation 13.2.4 Chemical activators related to plant metabolism 13.2.5 Chemical activators unrelated to plant metabolism 13.2.6 The conditional expression tool kit and its further development 13.3 Applications to plant functional genomics 13.3.1 Conditional over-expression 13.3.2 Conditional knockout/knockdown 13.3.3 Conditional complementation 13.4 Potential applications to plant biotechnology 13.5 Conclusions References Modulation of transcriptional networks in crop plants TONG ZHU 14.1 Introduction 14.2 Genes, regulatory factors, regulatory pathways, regulatory nodes, and regulatory networks 14.3 Functional characterization of transcription factor genes 14.4 Reverse engineering of transcription regulatory network
285 285 286 288 289 290 291 292 292 292 296 298 298 300 302
309 309 309 310 310 312 314 316 319 320 321 321 322 322 324 324 329 329 330 331 333
CONTENTS
14.5 Transcription network in crop plants 14.6 Transcription factors as gene switches for genetic engineering 14.7 Perspectives References Index The colour plate section appears after Page 142
xi 337 339 340 342 347
Contributors Dr George C Allen Department of Crop Science and Horticultural Science, 1203 Partners II (office), Campus Box 7550, North Carolina State University, Raleigh, NC 27695–7550, USA Dr J Brzeski Institute of Biochemistry and Biophysics, Polish Academy of Sciences, Pawinskiego 5a, 02-106 Warsaw, Poland Professor Thomas B¨orner Institute of Biology/Genetics, Humboldt University Berlin, Chausseestraße 117, 10115 Berlin, Germany Dr Mark Caddick School of Biological Sciences, The University of Liverpool, Biosciences Building, Liverpool L69 7ZB, UK Dr Yii Leng Chua Hutchinson MRC Research Centre, Department of Pathology, University of Cambridge, Hills Road, Cambridge CB2 2XZ, UK Professor Eva Czarnecka-Verner Department of Microbiology and Cell Science, University of Florida, PO Box 110700, Gainesville, FL 32611-0700, USA Professor Manuel Echeverr´ıa Laboratoire G´enome et D´eveloppement des Plantes, UMR CNRS-IRD 5096, Universit´e de Perpignan, 52, Avenue de Villeneuve, 66860 Perpignan Cedex, France Dr Mar´ıa Julissa Ek-Ramos Departamento de Bioqu´ımica, Facultad de Qu´ımica, Universidad Nacional Aut´onoma de Mexico, Ciudad Universitaria, 04510 M´exico, DF, M´exico Professor Klaus D Grasser Department of Life Sciences, Aalborg University, Sohngaardsholmsvej 49, DK-9000 Aalborg, Denmark Professor John C Gray Department of Plant Sciences, University of Cambridge, Downing Street, Cambridge CB2 3EA, UK Professor William B Gurley Department of Microbiology and Cell Science, University of Florida, PO Box 110700, Gainesville, FL 32611-0700, USA Professor A Jerzmanowski Laboratory of Plant Molecular Biology, Warsaw University, Pawinskiego 5a, 02-106 Warsaw, Poland Dr Dorte Launholt Department of Life Sciences, Aalborg University, Sohngaardsholmsvej 49, DK-9000 Aalborg, Denmark Dr Shai J Lawit Pioneer Hi-Bred International, Inc., a DuPont Company, 7300 NW 62nd Ave., PO Box 1004, Johnston, IA 50131–1004, USA Dr Karsten Liere Institute of Biology/Genetics, Humboldt University Berlin, Chausseestraße 117, 10115 Berlin, Germany Dr Rainer Melzer Lehrstuhl f¨ur Genetik, Friedrich-Schiller-Universit¨at Jena, Philosophenweg 12, D-07743 Jena, Germany Dr Kevin O’Grady Department of Microbiology and Cell Science, University of Florida, PO Box 110700, Gainesville, FL 32611-0700, USA
xiv
CONTRIBUTORS
Dr Tressa Jacob Panikulangara ZMBP, Allgemeine Genetik, Universit¨at T¨ubingen, Auf der Morgenstelle 28, D-72076 Tuebingen, Germany Dr Jos´e C Reyes Instituto de Bioqu´ımica Vegetal y Fotos´ıntesis, Consejo Superior de Investigaciones Cient´ıficas – Universidad de Sevilla, Av Am´erico Vespucio 49, 41092 Sevilla, Spain Dr Jos´e Luis Riechmann Division of Biology, California Institute of Technology, 1200 East California Boulevard, Pasadena, CA 91125, USA Dr Julio S´aez-V´asquez Laboratoire G´enome et D´eveloppement des Plantes, UMR CNRS-IRD 5096, Universit´e de Perpignan, 52, Avenue de Villeneuve, 66860 Perpignan Cedex, France Professor Friedrich Sch¨offl ZMBP, Allgemeine Genetik, Universit¨at T¨ubingen, Auf der Morgenstelle 28, D-72076 Tuebingen, Germany Dr Imre E Somssich Department of Plant Microbe Interactions, Max-PlanckInstitute for Plant Breeding, Carl-von-Linn´e Weg 10, D-50829 Koeln, Germany Professor Steven Spiker Department of Genetics, North Carolina State University, Box 7614, Raleigh, NC 27607, USA Professor Gunter ¨ Theißen Lehrstuhl f¨ur Genetik, Friedrich-Schiller-Universit¨at Jena, Philosophenweg 12, D-07743 Jena, Germany Professor William F Thompson Departments of Plant Biology, Genetics, and Crop Science, Campus Box 7550, North Carolina State University, Raleigh, NC 27607, USA Professor Brian Tomsett School of Biological Sciences, The University of Liverpool, Biosciences Building, Liverpool L69 7ZB, UK Dr Angela Tregova School of Biological Sciences, The University of Liverpool, Biosciences Building, Liverpool L69 7ZB, UK Professor Steven van Nocker Department of Horticulture, 390 Plant and Soil Sciences Building, Michigan State University, East Lansing, MI 48824, USA Dr Tong Zhu Syngenta Biotechnology, Inc., 3054 Cornwallis Road, Research Triangle Park, NC 27709, USA
Preface The regulation of transcription is a central step in the control of gene expression in plants and it therefore plays an essential role in development and in the way plants cope with environmental influences. Through combinatorial control, a wide variety of transcription factors precisely determines the transcription level of target genes as well as their spatial and temporal expression pattern. It has emerged in recent years that the nuclear environment of the genome critically affects the regulated transcription of genes. Thus, genes are packaged into a complex nucleoprotein structure termed chromatin, generally repressing their transcription. At the same time the dynamic chromatin structure offers an opportunity for the regulation of gene expression mediated, for instance, by histone modifiers and chromatin-remodelling activities. In this volume, authors from leading laboratories have prepared chapters addressing the different mechanisms that cooperate in the regulation of transcription in plants, and links between transcriptional regulation and biotechnological applications have also been considered. The book begins with chapters describing the structure and function of transcription factors and assistant architectural proteins as well as their interplay with promoter regions (Chapters 1–3). These chapters provide insights into the protein– DNA and the protein–protein interactions that are crucial for the regulation of transcript initiation. Chapters 4 and 5 focus on the role of histone modifications, histone variants and chromatin remodelling, priming the chromatin for transcription. Various effects of DNA methylation on transcription are addressed in Chapters 5 and 7. Chapter 6 is devoted to matrix attachment regions and their connection to transcriptional gene silencing. In Chapters 7 and 8, the special features of transcription by RNA polymerase I and the transcription of plastid genes are described. In the following chapters some well-studied examples of transcriptional regulation in plants are explored. Chapter 9 deals with the transcriptional regulation of flowering time, while Chapter 10 describes how floral organ identity is under the combinatorial control of MADS-domain transcription factors. Chapter 11 focuses on plant defense responses that are regulated by transcriptional networks. In Chapter 12, the temperature-regulated expression of genes upon heat- and cold-stress is described. The final two chapters are dedicated to the discussion of some aspects related to applied plant biotechnology. Chapter 13 summarises the principles and applications of inducible transcription systems for research and biotechnology. In Chapter 14, the potential of engineering transcriptional networks in crop plants is discussed. Through the availability of genome sequences and ‘high-throughput’ approaches, the research in the field of transcriptional regulation is currently moving in a new direction. This volume considers the ‘genomic perspectives’ as well as the
xvi
PREFACE
thorough ‘case-by-case’ studies that have revealed many of the principles known about the regulation of transcription in plants. The book is intended to provide a guide into this exciting topic for researchers and teachers in the field of plant biology. I would like to thank all the authors for their outstanding contributions, and the editorial staff of Blackwell Publishing for the pleasant collaboration on this project. Klaus D. Grasser
Annual Plant Reviews A series for researchers and postgraduates in the plant sciences. Each volume in this series focuses on a theme of topical importance and emphasis is placed on rapid publication. Editorial Board: Prof. Jeremy A. Roberts (Editor-in-Chief), Plant Science Division, School of Biosciences, University of Nottingham, Sutton Bonington Campus, Loughborough, Leicestershire, LE12 5RD, UK; Dr David Evans, School of Biological and Molecular Sciences, Oxford Brookes University, Headington, Oxford, OX3 0BP, UK; Prof. Hidemasa Imaseki, Obata-Minami 2419, Moriyama-ku, Nagoya 463, Japan; Dr Michael T. McManus, Institute of Molecular BioSciences, Massey University, Palmerston North, New Zealand; Dr Jocelyn K.C. Rose, Department of Plant Biology, Cornell University, Ithaca, New York 14853, USA. Titles in the series: 1. Arabidopsis Edited by M. Anderson and J.A. Roberts 2. Biochemistry of Plant Secondary Metabolism Edited by M. Wink 3. Functions of Plant Secondary Metabolites and their Exploitation in Biotechnology Edited by M. Wink 4. Molecular Plant Pathology Edited by M. Dickinson and J. Beynon 5. Vacuolar Compartments Edited by D.G. Robinson and J.C. Rogers 6. Plant Reproduction Edited by S.D. O’Neill and J.A. Roberts 7. Protein–Protein Interactions in Plant Biology Edited by M.T. McManus, W.A. Laing and A.C. Allan 8. The Plant Cell Wall Edited by J.K.C. Rose 9. The Golgi Apparatus and the Plant Secretory Pathway Edited by D.G. Robinson 10. The Plant Cytoskeleton in Cell Differentiation and Development Edited by P.J. Hussey 11. Plant–Pathogen Interactions Edited by N.J. Talbot 12. Polarity in Plants Edited by K. Lindsey 13. Plastids Edited by S.G. Moller 14. Plant Pigments and their Manipulation Edited by K.M. Davies 15. Membrane Transport in Plants Edited by M.R. Blatt 16. Intercellular Communication in Plants Edited by A.J. Fleming 17. Plant Architecture and its Manipulation Edited by C. Tuurnbull 18. Plasmodesmata Edited by K.J. Oparka 19. Plant Epigenetics Edited by P. Meyer
20. Flowering and Its Manipulation Edited by C. Ainsworth 21. Endogenous Plant Rhythms Edited by A. Hall and H. McWatters 22. Control of Primary Metabolism in Plants Edited by W.C. Plaxton and M.T. McManus 23. Biology of the Plant Cuticle Edited by M. Riederer 24. Plant Hormone Signaling Edited by P. Hadden and S.G. Thomas 25. Plant Cell Separation & Adhesion Edited by J.R. Roberts and Z. Gonzalez-Carranza 26. Senescence Processes in Plants Edited by S. Gan 27. Seed Development, Dormancy and Germination Edited by K.J. Bradford and H. Nonogaki 28. Plant Proteomics Edited by C. Finnie 29. Regulation of Transcription in Plants Edited by K. Grasser 30. Light and Plant Development Edited by G. Whitelam 31. Plant Mitochondria Edited by D. Logan
1 General transcription factors and the core promoter: ancient roots William B. Gurley, Kevin O’Grady, Eva Czarnecka-Verner and Shai J. Lawit
1.1
Introduction
Transcription is an ancient process that has its origins near the beginning of life on this planet, nearly 3.5 billion years ago. It is, therefore, no surprise that the basic components of the transcriptional apparatus and the underlying mechanisms of gene activation of modern organisms comprise a mixture of the very old interspersed with recent adaptation. Plants share a common heritage of transcriptional regulation with all Eukaryota. In this chapter, we examine the fundamentals of promoter structure and activated transcription, building on what is known from other eukaryotes and exploring the implications with respect to higher plants. Where information is available, we describe how plants have utilized and exploited the most basic components of the transcriptional machinery to meet their unique needs in the regulation of gene expression.
1.2
Origins of the eukaryotic promoter
Promoter regions of genes comprise the DNA sequences that are involved in regulating transcriptional activity. For eukaryotes, sequence elements (cis-elements) within the DNA of the promoter regulate transcriptional activity by determining the type, affinities, and spatial arrangement of regulatory proteins (transcription factors) that nucleate on the DNA near the start site of transcription. For genes read by RNA polymerase II (pol II), most promoter elements are clustered upstream from the start site of transcription. There are two major classes of transcription factors that regulate RNA pol II: those common to the expression of most genes (general transcription factors (GTFs)), and those that regulate specific subsets of genes (transactivators). The GTFs bind in a complex with the DNA of the promoter to form a preinitiation complex (PIC) immediately upstream from the start site (+1 nt). The promoter can be thought of as a scaffold that regulates gene activity by specifying the type and physical arrangement of proteins that assemble for transcription. In eubacteria, promoter DNA often plays an additional role in affecting the rate of DNA melting, which, in turn, affects the kinetics of transcriptional initiation by RNA polymerase. However, this direct involvement in transcriptional regulation appears to be minimal in eukaryotes.
2
REGULATION OF TRANSCRIPTION IN PLANTS
Traces of the archetypal eukaryotic promoter can be clearly seen in the configuration of archaeal promoters (Figure 1.1). This is not surprising since the basal transcriptional apparatus of the eukaryotic nucleus is closely related to that found in this group of prokaryotic organisms, and current evolutionary theory suggests that the archaea contributed to the formation of the eukaryotic cell (Margulis et al., 2000). For example, the subunit composition of archaeal RNA polymerase is complex and very similar to that present in eukaryotes: both have the same 12 subunits versus the 4 present in eubacteria (Werner and Weinzierl, 2002). The major differences are that in archaea, the largest two subunits of the eukaryotic polymerase are present as two proteins (A , A and B , B ), and the two eukaryotic subunits, RBP8 and RBP9, are absent in archaeal enzyme. These common components of the transcriptional apparatus are reflected in shared features in promoter design between archaea and eukaryotes. The archaeal promoter contains several DNA elements that are components of modern eukaryotic promoters: an A/ T-rich TATA-like element centered 25–30 bp upstream from the start of transcription where TATA-binding protein (TBP) binds, and an element immediately upstream of the TATA that interacts specifically with TFB (equivalent to eukaryotic transcription factor IIB (TFIIB)) known as the TFB response element (BRE) (Geiduschek and Ouhammouch, 2005). As in eukaryotes, TFIIB makes contact with TBP within the C-terminal stirrup (Nikolov et al., 1995; Nikolov and Burley, 1997), thus fixing the relative locations of the two promoter contact sites in the TBP–TFB protein complex. It is interesting that the TATA in archaea is most often in the inverted orientation (5 -TTATA-3 ) with respect to the BRE. TBP in both archaea and eukaryotes can bind to either orientation; hence, the polarity of transcription is determined by the relative location of the BRE with respect to the TATA (Cox et al., 1997; Bell et al., 1999), with the BRE always positioned upstream of the TATA in the promoter. As in the archaea, the eukaryotic promoter determines the direction of transcription by providing the transcriptional apparatus at least two points of contact with the template. In contrast to the usual perception of upstream cis-elements being flexible in position, the location of these upstream activation sequences (UASs) in archaea is spatially constrained (Ouhammouch et al., 2005). The general strategy of inducing transcription by binding regulatory proteins upstream of the core promoter can be traced back to the eubacteria, where positive transcriptional regulators such as cyclic AMP-binding protein (CAP) bind to spatially constrained sites upstream of the core promoter elements at the –10 and –35 positions (Busby and Ebright, 1999). In the case of eubacteria, the transcription factor recruits RNA polymerase directly by making contacts with the α-subunit. The location of the bacterial UAS is critical to ensure the proper positioning of RNA polymerase at the start site. In archaea, it appears that at least some promoters initially recruit TBP (or perhaps TFB), which subsequently recruits RNA polymerase (Ouhammouch et al., 2005). As in eubacteria, this direct recruitment of a protein that makes specific contact with the DNA of the core promoter imposes strict spatial constraints on the placement of the UAS. The general strategies for promoter design evident in the eubacteria and
GENERAL TRANSCRIPTION FACTORS AND THE CORE PROMOTER
3
Figure 1.1 Schematic comparison of the archaeal and eukaryotic promoter structure. (A) The archaeal promoter shares two core elements with eukaryotic promoters, the binding sites for the general transcription factors TBP and TFB. Another similarity is the positioning of upstream activation sequences 5 and proximal to the BRE. (B) Eukaryotic promoters typically contain at least one of the core archaeal elements, and sometimes additional elements for binding subunits of TFIID (TAFs). Abbreviations: BRE, TFB response element; Inr, initiator region; DPE, downstream promoter element.
4
REGULATION OF TRANSCRIPTION IN PLANTS
archaea are (1) the presence of an A/T-rich sequence upstream of the start site, (2) the placement of UASs upstream of the core promoter, and (3) the involvement of ciselements proximal to the core promoter to facilitate the precise positioning of RNA polymerase. Each of these features of promoter architecture has been conserved in eukaryotes, especially with respect to genes transcribed by RNA pol II.
1.3
Organization of the eukaryotic promoter
Typically, the eukaryotic promoter is more complex than the archaeal promoter in terms of the types of elements and their location. As seen in Figure 1.1B, there are certain features in common with the archaeal promoter regarding elements proximal to the start site of transcription (specifically the core promoter). For example, the TATA and sometimes the BRE are present, and their location is spatially constrained in a fashion reminiscent of that seen in archaea. Additional spatially constrained elements are sometimes seen in metazoan core promoters such as the initiator region (Inr) centered around +1, and the downstream promoter element (DPE), located in the 5 -untranslated region (UTR). In Drosophila four classes of core promoters are common: those containing TATA, TATA-Inr, Inr, and Inr-DPE (Smale and Kadonaga, 2003). The last two categories lack a TATA, but still initiate transcription with a precise start site and the correct polarity. However, TBP may still be recruited to these promoters via TBP-associated factors (TAFs), since both the Inr and DPE are recognized by specific groups of TAFs. TAFs 1 and 2 bind the Inr (Chalkley and Verrijzer, 1999), and TAFs 6 and 9 recognize the DPE (Burke and Kadonaga, 1997). The lack of Inr and DPE elements in archaea is consistent with the lack of TAFs in these organisms. Few, if any, eukaryotic promoters contain all of the possible sites of the core promoter. Since only two points of contact with the promoter are required to establish the polarity of transcription, only one to two elements are normally present in the core promoter. In those cases where there is only a single spatially constrained site, the determination of transcriptional polarity is more problematic and requires input from elements upstream of the core promoter. Although the second site of contact with the PIC is usually upstream of the core promoter, the spacing and orientation of this second element may be more flexible. Transactivators are predicted to orient the PIC by making contacts with GTFs that do not directly bind DNA and are, thus, more adaptable with regard to alignment and spacing. Most eukaryotic cis-elements exhibit considerable flexibility in orientation and spacing with respect to core promoter elements. There are, however, many examples of cis-elements that are located within 15–20 bp upstream of the TATA motif where direct contacts with TBP and TFIIB are possible. It is reasonable that these TATA-proximal elements may be involved in the precise alignment of the PIC by binding transactivators that make direct contact with either TBP or TFIIB. Although DNA looping may bring distal sites in close proximity with the core promoter (Blackwood and Kadonaga, 1998; Bulger and Groudine, 1999), the expectation is that transactivators bound to more distal cis-elements recruit (but do not precisely position) the PIC through contacts with proteins that do not directly
GENERAL TRANSCRIPTION FACTORS AND THE CORE PROMOTER
5
bind the core promoter. However, transactivators bound at TATA-proximal elements may participate in the direct recruitment of TBP and TFIIB, as is the case in the archaea. A close examination of one of the best analyzed promoters in eukaryotes, the Endo16 promoter from sea urchin (Yuh et al., 1998), will serve as an introduction of what we may expect in plants. The Endo16 gene encodes a polyfunctional secreted protein (Soltysik-Espanola et al., 1994) expressed in the primordial endoderm cells present in the midgut and late embryo and larval stages (Nocente-McGrath et al., 1989; Ransick et al., 1993). It is also specifically repressed in adjacent cell lineages. The full promoter includes the core promoter and 2300 bp of additional upstream sequences. Although 15 different proteins have been shown to bind specific ciselements within the upstream promoter, most have not been identified beyond their molecular weight and DNA-binding affinities. The general term ‘cis-element’ is used here to distinguish between those sites that may activate, synergize, or repress transcription as opposed to those that are known simply to activate transcription (UASs). A detailed mutational analysis of the upstream regions of the promoter (Yuh et al., 1998) revealed an underlying structure in terms of function and spatial organization of the cis-elements (Figure 1.2). Elements in the Endo16 promoter are organized into seven clusters, or modules. The term ‘module’ is preferred here to ‘enhancer’, since modules can either activate or repress. Although enhancers are also composed of clusters of cis-elements, they are usually associated with the activation of genes. The upstream modules in the Endo16 promoter are responsible for the developmental timing of gene activation and repression, and in marking specific cells for expression (spatial expression). It is noteworthy that all activities of modules B–G (both positive and negative) either require a functional TATAproximal element (module A), or rely on this proximal module for their synergistic enhancement. In addition to integrating inputs from other modules, it also appears that module A functions as a two-position switch that either accepts or blocks
Figure 1.2 Modular organization of the upstream cis-elements in the eukaryotic promoter. The core promoter includes binding sites for the basal transcriptional apparatus (GTFs). In plants, where little evidence exists for core elements other than TATA and the Inr, the core promoter typically extends from −45 to a little beyond +1. In other eukaryotes the core promoter includes additional cis-elements such as the BRE, initiator, and the DPE. The upstream regulatory region includes all sequences upstream of the core promoter involved in regulating expression of the gene and often extends 1–2 kb upstream (proximal and distal promoters). Note that the upstream cis-elements are organized into clusters referred to as ‘modules’. A module may contain like elements or a variety of different types of elements. The TATAproximal module (module A above) may integrate inputs from other modules positioned upstream.
6
REGULATION OF TRANSCRIPTION IN PLANTS
input from the adjacent module B. An important lesson that can be taken from the functional organization of the Endo16 promoter is the dominant influence of the TATA-proximal module. This is logical since module A is best positioned to have direct input to the basal transcriptional apparatus – a theme that can be traced back to the archaeal and bacterial roots of all eukaryotic promoters. It is also worth noting that only one of the seven modules is involved in spatial expression; the rest are dedicated to operation of the logic system of the promoter. This logic system is involved in regulating the timing of activation, synergistic enhancement, and in repression. This should serve as a caveat for promoter studies that focus only on identification of tissue-specific modules in plants and in other systems. How do plant promoters differ from the typical eukaryotic promoter? A compilation of plant core promoter elements is presented in Figure 1.3. The most obvious difference is the lack of evidence for core elements other than the TATA and Inr (sometimes called the ‘transcription start sequence motif’; TSS-motif) (Shahmuradov et al., 2003, 2005). This absence of additional core elements, however, does not imply that plant TFIIB and subunits of TFIID (TAFs) do not make contact with the core promoter, only that the contacts may not be base pair specific. Upstream cis-elements are not generally found closer to the TATA than about 15 bp, suggesting that TFIIB may occupy the intervening space by nonspecific general affinity for the DNA instead of specifically recognizing the BRE consensus. Computer-based analysis of a large number of eukaryotic core promoters and 5 flanking sequences has revealed an additional element, known as the ‘CCAAT box’, that has an optimal location immediately upstream of the core promoter (Benoist et al., 1980; Bucher, 1990; Mantovani, 1998). The CCAAT box is present in 25– 30% of eukaryotic promoters and is functional in both the forward and the reverse orientations. Its prevalence is slightly higher in TATA-less promoters, where it is often found in the reverse orientation. In some TATA-less promoters, it is closer to +1 and may even overlap the start site (Mantovani, 1998). The consensus in animals is 5 -CCA a/t t/a c/g a/g-3 , while in plants it simply reads ‘5 -CAAT3 ’ (Shahmuradov et al., 2003). In 71 unrelated plant promoters containing the CAAT box, this element was distributed between positions –46 and –106, with the average centered at −75 bp. The plant CAAT box was found to be present in 131 unrelated promoters: 71 in TATA-containing promoters and 60 in TATA-less promoters. Unexpectedly, recent findings indicate that from 40 to 50% of eukaryotic CCAAT boxes may also be found outside of the proximal promoter in introns or at far distant positions (both 5 and 3 ) (Testa et al., 2005). Nuclear factor Y (NF-Y) binds the CCAAT box and activates transcription in animals (Hooft van Huijsduijnen et al., 1987; Mantovani, 1999). NF-Y binds DNA as a trimer formed between the three subunits, NF-YA, NF-YB, and NF-YC (Sinha et al., 1995). Subunits NF-YB and NF-YC contain a conserved histone-fold domain involved in a tight interaction between these two subunits (Baxevanis et al., 1995). DNA binding requires all three subunits. Transcriptional activation in mammals is dependent on glutamine-rich (Q-rich) and serine–threonine-rich regions in the Cterminal domain (CTD) of NF-YB and a Q-rich domain in NFY-B (Coustry et al., 1995, 1996). Homologs to all three subunits have been cloned as multiple isoforms
GENERAL TRANSCRIPTION FACTORS AND THE CORE PROMOTER
7
General eukaryotic TATAAA GTATAAAAGGCGGGG C T TT TACGCCCC TATAAA from 171 plant promoters cTATAAATAcc TA General eukaryotic Inr TCAGTCTT G TTCTCC CA G TSS-motif (Inr) in 217 dicot promoters WnTCAw = +1 atcw TSS-motif (Inr) in 70 monocot promoters aNnCA TSS-motif (Inr) in 171 plant TATA promoters TCAnM c TSS-motif (Inr) in 130 plant TATA-less promoters TYAat a cca tg CCAAT box in general eukaryotes TCTAGCCAATCA ctcga tagg CCAAT box in 131 plant promoters nCAAT a cca Figure 1.3 Elements present in the core promoter of plants. This information is based on the results of Shahmuradov et al. (2003) obtained from computer analysis of a plant promoter database (PlantProm DB). The consensus sequences are represented in bold text, with the highest consensus positions capitalized. Alternate nucleotides are given below the preferred sequence. Abbreviations: W, A or T; N, any nucleotide; TATAAA, binding site for TBP usually located from −45 to −25 bp from the start of transcription; Inr, initiator region; recognition site for TAFs 1 and 2 which spans the transcription start site in yeast and metazoans; TSS, transcription start site – consensus of sequences that include and flank the initiation site of plant genes. It is roughly equivalent to the Inr, but has been derived from homology comparisons instead of analyses of promoter function; CCAAT box, promoter element that is often located upstream and proximal to the core promoter in eukaryotes.
8
REGULATION OF TRANSCRIPTION IN PLANTS
(7 NF-YAs, 10 NF-Ybs, and 9 NF-YCs) from Arabidopsis (Gusmaroli et al., 2001, 2002). Of these, only five plant isoforms of NF-YC contain Q-rich domains. Consistent with the occurrence of CCAAT boxes in the proximal promoter (where direct contact with GTFs is facilitated) is the finding that NF-YB and NF-YC subunits show in vitro affinity for TBP (Bellorini et al., 1997).
1.3.1
TATA-less promoters
In Drosophila almost half of the promoters contain a TATA motif spaced 25–30 bp upstream of the start site. Most of the remaining promoters contain either the Inr element, or Inr plus the DPE (Smale and Kadonaga, 2003). Whereas both the TATA and Inr can direct correct transcriptional initiation independently (Carcamo et al., 1991; Weis and Reinberg, 1997), the DPE must act in conjunction with the Inr (Burke and Kadonaga, 1996). In a study of 301 unrelated plant promoters, only 57% contained TATA, a percentage similar to that generally found in other eukaryotes (Shahmuradov et al., 2003). The average spacing between the plant TATA and the start of transcription was found to be 26 bp, very similar to other eukaryotes. Despite a lack of TATA consensus, TATA-less promoters can function with TBP. In vitro experiments indicate that in TATA-less promoters, activity was correlated with the %A/T nucleotides at the –30 position, suggesting that mammalian TBP shows some degree of recognition of even badly flawed TATA-like sequences (Zenzie-Gregory et al., 1993). However, activity in assays containing mutated TBP indicated that DNA contact by TBP may not be an absolute requirement for transcription (Zenzie-Gregory et al., 1993).
1.4
Transcription factor IID
The GTF TFIID is a large multiprotein complex composed of TBP and additional subunits (usually 12–15) called TAFs (for reviews see Burley and Roeder, 1996; Verrijzer and Tjian, 1996; Lee and Young, 1998; Roeder, 1998; Tora, 2002). TFIID seems to be present in all groups of eukaryotic organisms and is thought to play a key role in promoter recognition and transcriptional activation of gene expression. TAFs have been shown to bind activators in vitro (Verrijzer and Tjian, 1996) and to play a critical role in metazoan development. TAFs are the targets of activator proteins, serving as bridges between the activator and TBP in the recruitment of TFIID to the promoter (Verrijzer and Tjian, 1996). Activation of specific genes in animals can be reduced by mutations in certain TAFs including TAF1, TAF6, and TAF4 (Sauer et al., 1996); moreover, certain TAF mutations are lethal in both mammals and yeast, especially those involved in the cell cycle (Reese et al., 1994; Poon et al., 1995; Verrijzer and Tjian, 1996). Work from Buratowski’s group has shown that four histone-like TAFs (TAFs 6, 9,11, and 12) are generally required for most transcription (Michel et al., 1998; Komarnitsky et al., 1999) in yeast. However, some exceptions are noted: TAF9-depleted cells still retain the capacity for Ace1 induction by copper, and the activation of heat-shock (HS) genes occurs in the
GENERAL TRANSCRIPTION FACTORS AND THE CORE PROMOTER
9
absence of a TFIID complex (Moqtaderi et al., 1998). The essential nature of TAF11 is significant, since this TAF is found only in TFIID. Approximately 84% of yeast genes are dependent on one or more TAFs (Shen et al., 2003). Selective recruitment of TAFs by the upstream activator sequences of TAF-dependent promoters has been demonstrated in yeast, using chromatin immunoprecipitation (ChIP) assays (Li et al., 2002). This suggests that many TAFs, while not globally required for transcription, may play specialized roles in the regulation of subsets of genes. In plants, little information regarding the role of TAFs in transcription is available, except the in vitro studies of Washburn et al.(1997). In those experiments, activated transcription directed by the Herpes simplex VP16 acidic activator was dependent on the presence of TBP plus the TAFs – result closely in line with expectations from mammalian and yeast studies (Albright and Tjian, 2000; Sanders et al., 2002).
1.5
Role of TFIID in development
The degree to which TAFs participate in global gene activation is not resolved, but the evidence points to a strong bias toward involvement in developmental processes of multicellular organisms. Although TAFs are clearly not involved in all transcription, it does appear that certain TAFs and their alternate isoforms are critical components of cell-type-specific pathways of differentiation, while others fill more general requirements in the progression of the cell cycle. For example, the two largest TAFs (1 and 2) are required in mammals for the G1/S and G2/M transitions of the cell cycle, respectively (Wang et al., 1997; Martin et al., 1999). Out of the 12–13 remaining TAFs, almost half have been shown to be involved in cellular differentiation. TAF4 and TAF4b are required for embryogenesis in flies and mammals, as well as participate in gamete production (Metzger et al., 1999; Bell et al., 2001; Freiman et al., 2001; Guermah et al., 2003; Hiller et al., 2004). TAF4 is also required for activation of the master regulators Dorsal and Twister in Drosophila (Zhou et al., 1998; Pham et al., 1999) and is critical to neuronal cell differentiation of the embryonic brain and spinal cord in mice (Metsis et al., 2001). Although the subunits of TFIID play major roles in cellular differentiation and cell cycle in animals, it is unclear to what extent they play a similar role in plants. However, deep involvement in the processes of cellular differentiation is likely, since all of the alternative forms of TAFs that have specialized toward differentiation in other organisms are also present in Arabidopsis (Lago et al., 2004). It will be interesting to see how this potential for specialization manifests itself in plants where reproductive tissues and organs are derived from postembryonic meristems instead of being formed during embryogenesis, as is the case in metazoans. Experimentally induced alterations in levels of Arabidopsis TBP or specific TAFs show a wide range of phenotypes affecting the organization of vegetative meristems, leaf development, formation of floral organs and leaves, fertility, pollen tube growth, and light-mediated induction of specific genes. High levels of TBP-2 expression at the transcription level are seen in apical shoot tissues; whereas, overexpression
10
REGULATION OF TRANSCRIPTION IN PLANTS
of Arabidopsis TBP-2 in transgenic plants resulted in shoot proliferation, with supernumerary meristems forming in disorganized leaf primordia (Li et al., 2001). There are several lines of evidence consistent with the involvement of plant TAFs in developmental processes. TAF10 is a histone-fold protein that is a component of two major transcriptional complexes: TFIID and SAGA (Grant et al., 1998; Green, 2000). In animals, TAF10 shows a very restricted pattern of expression and is present only in specific cells during embryogenesis (Georgieva et al., 2000; Mohan et al., 2003). Although few similar studies have been reported in plants, there is limited evidence suggesting that plant TAF10 may also be specialized to regulate transcription with tissue specificity. In Flaveria trinervia (a composite of the Aster family), low levels of transcripts occur throughout the plant (Tamada et al., 2003). However, there are locations where TAF10 mRNA is abundant or at elevated levels: in the vascular tissues of hypocotyls, in the central stele of roots, and in bundle sheath cells of leaves (Furumoto et al., 2005). Overexpression of the F. trinervia TAF10 (ftTAF10) in Arabidopsis resulted in several abnormal phenotypes. One had limited growth of an indeterminate inflorescence, and the other developed dark green deformed leaves and chimeric floral organs similar to terminal flower mutants (Shannon and Meeks-Wagner, 1991). TAF6 contains a histone fold and is present in animals in two forms, TAF6 and TAF6-L (TAF6-like). TAF6 is an integral part of TFIID, and TAF6-L has been shown in humans to be a component of the histone acetyltransferase (HAT)-containing PCAF (p300/CBP-associated factor) complex. In animals TAF6 is implicated in a variety of developmental events in germline and somatic cells (Aoyagi and Wassarman, 2001). Arabidopsis has two TAF6 genes (6 and 6b) that are distantly related (35% identity) (Lago et al., 2004). In plants heterozygous for a T-DNA insertion in TAF6, the only observed affect was a drastic reduction in pollen tube growth rates (Lago et al., 2005). These results not only confirm the relatively recent finding that de novo RNA synthesis occurs in pollen tubes (Becker et al., 2003), but indicate that TAF6 plays an important role in pollen tube function. TAF1 is unique among the subunits of TFIID due to its polyfunctional enzymatic capabilities. In addition to promoter recognition at the Inr, it has the potential to acetylate, phosphorylate, and facilitate ubiquitylation of chromatin or PIC components. There are two paralogs of TAF1 in Arabidopsis, TAF1 and TAF1b. These are also known as HAP1 and HAP2, respectively, based on their classification as proteins possessing HAT activity (Pandey et al., 2002). TAF1 has been shown in other organisms to possess two additional activities: a protein kinase activity and a ubiquitin activating and conjugating activity (Pham and Sauer, 2000). These enzymatic activities potentially enable TAF1 to modify chromatin and other transcription factors. All three of these domains are conserved in the Arabidopsis genes (Lago et al., 2004). In humans, two bromodomains are present at the C-terminus that synergistically facilitate binding to acetylated histone H4 (Jacobson et al., 2000). In Arabidopsis, there is only a single bromodomain. It is not known whether an additional bromodomain is furnished by an auxiliary protein, as is the case in yeast where both bromodomains are located on a separate protein (Bdf1). Plant TAF1 is unique among eukaryotes in having a ubiquitin domain (UbD) embedded within the coding region. The function of the UbD is unknown, but it raises the possibility that
GENERAL TRANSCRIPTION FACTORS AND THE CORE PROMOTER
11
it serves for the attachment of proteasome subunits, perhaps as a way to facilitate turnover of transactivators at the promoter. Another feature of TAF1 is the presence of an N-terminal domain (TAND-1) that mimics the structure of the double-strand TATA sequence of the promoter (Liu et al., 1998; Mal et al., 2004). TAND-1 of yeast and Drosophila TAF1 interacts with the concave DNA-binding surface of TBP and inhibits TBP binding to the TATA of the promoter (Bagby et al., 2000). The binding of TAND-1 to TBP may be conserved at the structural level, but is not easily identified in terms of amino acid sequence in plant TAF1 (Lago et al., 2004). This domain seems to be present in Arabidopsis TAF1, but is not evident in TAF1b from examination of amino acid sequence. HAP2 (TAF1b) is expressed at extremely low levels in most plant tissues, but is easier to detect in shoots and flower buds (Bertrand et al., 2005). Microarray analysis of RNA from fourteen-day Arabidopsis shoots of the haf2 knockout mutant indicated that transcript levels from 854 genes from various functional categories were affected. However, it was not possible to distinguish between direct and secondary effects in this experiment (Bertrand et al., 2005). In addition, disruption of Arabidopsis HAF2 expression by T-DNA insertion interfered with the expression of several light-regulated genes, including CAB2 and RBCS-1A. ChIP analysis of these two promoters revealed that TAF1b may be involved in the acetylation of histone H3 in both genes, but only H4 acetylation at the RBSC-1A promoter. The fact that these perturbations in light-responsive signaling were specific to TAF1b mutations, and not TAF1, suggests that at least some degree of specialization has occurred between these plant paralogs of TAF1. The histone fold is present in 9 out of 14 TAFs of TFIID. These have the potential to form five heterodimer pairs, which are thought to make structural contributions to TFIID (Gangloff et al., 2001). Four TAFs may form a nucleosome-like structure composed of two heterodimers: TAF6 (H4)–TAF9 (H3) and TAF4 (H2A)–TAF12 (H2B). However, stoichiometric considerations and the TAF immunolocalization experiments of Leurent et al. (2004) suggest that the structure more resembles the archaeal tetramer versus the eukaryotic octomeric structure. In addition, six other TAFs may form three other heterodimers (TAF3–TAF10/10b, TAF8–TAF10/10b, and TAF11–TAF13). Immunolabeling experiments with yeast TAFs, combined with EM, have shown TFIID to possess a trilobed structure (Figure 1.4) (Leurent et al., 2004). Two copies of each histone-fold heterodimer are present that allow asymmetric assignment to each of the three lobes. Two copies of TAF5 seem to form a structural backbone for lobes A and B, with TAFs 1 and 7 and TBP associated with lobe A. Yeast two-hybrid studies using the Arabidopsis TAFs yielded an interaction matrix consistent with the yeast structure (Lawit, 2003), suggesting that TFIID has experienced overall structural conservation among eukaryotes.
1.6
Mediator
The mediator is a large complex (∼2 MDa) composed of 20–25 proteins that are required in vivo for basal transcription at the level of initiation and reinitiation.
12
REGULATION OF TRANSCRIPTION IN PLANTS
Figure 1.4 Map of TFIID subunits. The topology and assignment of subunit locations is based on the immunolabeling experiments of Leurent et al. (2004). The two copies of TAF5 serve important structural roles, as well as the two copies of histone-fold TAF heterodimers (4/12, 6/9, 8/10, and 11/13). TAF3 is absent here to reflect the lack of TAF3 identification in plants to date.
Its critical role in basal transcription, as well as its involvement in activation and repression qualify it as a GTF (Takagi and Kornberg, 2005). Mediator does not directly bind DNA of the promoter, but instead acts as a coactivator, serving as a bridge between transactivators and RNA pol II. One of its roles in gene activation is thought to involve interactions with the of the largest subunit of RNA pol II (Myers et al., 1998; Naar et al., 2002; Nair et al., 2005). Since the role of mediator in activated transcription and its structure in both yeast and metazoans have recently been extensively reviewed elsewhere (Bjorklund and Gustafsson, 2005; Chadick and Asturias, 2005; Conaway et al., 2005; Kim and Lis, 2005; Kornberg, 2005; Malik and Roeder, 2005), in this section we concentrate on what is known of mediator in plants. A number of biochemical, genetic, protein interaction, and electron microscopy (EM) studies have determined that mediator is composed of three modules designated as ‘tail’, ‘head’, and ‘middle’ (Chadick and Asturias, 2005). In addition, there is a fourth module known as ‘CDK8/Srb8-11’ that is present in some mediator preparations. We have used the amino acid sequences provided in Boube et al. (2002) to identify the corresponding homologs in Arabidopsis using BLAST and reciprocal BLAST searches. Here we present a compiled sequence analysis of the data representing various mediator modules and identify Arabidopsis mediator subunits, including putative Arabidopsis homologs not found in yeast (Bourbon et al., 2004). We have used the model for modular subunit distribution compiled in Chadick and Asturias (2005) to organize the data (Tables 1.1–1.5).
1.6.1
Tail module
The tail module of mediator has been characterized as an integrator of transcription factor signals in yeast, and evidence suggests that it can function as a separate coactivator complex (Zhang et al., 2004). In yeast it is composed of five subunits (MED2, 3, and 14–16), three of which have corresponding orthologs in metazoans (MED14–16). In plants, only MED14 and 15 have been tentatively identified, one of which is linked to a mutant phenotype (SWP).
GENERAL TRANSCRIPTION FACTORS AND THE CORE PROMOTER Table 1.1
13
Mediator tail subunits from humans, Drosophila, and Arabidopsis Tail
Saccharomyces cerevisiae MED2 (Med2) MED3 (Med3) MED14 (Rgr1) MED15 (Gal11) MED16 (Sin4)
Arabidopsis thaliana
Drosophila melanogaster
Homo sapiens
At3g04740 (SWP1) (B) At1g15780
TRAP170 Arc105 TRAP95
TRAP170-DRIP/CRSP150 ARC105 TRAP95-DRIP92
Note: The unified nomenclature proposed by Bourbon et al. (2004) is listed in the S. cerivisiae column along with the yeast homolog in parentheses. Genes identified in Boube et al. (2002) are followed with ‘(B)’. Arabidopsis genes are according to the Arabidopsis Genome Initiative (AGI) annotators and are the TAIR locus designation (http://www.Arabidopsis.org/).
The identification of MED15 (a yeast Gal11 ortholog) is based on a single block of amino acid conservation. Novatchkova and Eisenhaber (2004) identified a KIX domain in the N-terminus of Drosophila ARC105, yeast Gal11, plant Gal11-like (At1g15780.1), and plant CREB-binding protein/p300 (CBP/p300) and postulate it to be the defining feature of Gal11-like proteins. KIX is a protein–protein interaction domain with affinity for the KID domain of CREB, as well with a number of other transcriptional activators (Campbell and Lumb, 2002; Liu et al., 2003; Zor et al., 2004). MED14 is a subunit that has been implicated as a regulator of both positive and negative gene expressions in yeast (Covitz et al., 1994; Jiang et al., 1995; Lee et al., 1999). It is broadly required for Caenorhabditis elegans embryonic transcription, Table 1.2
Mediator middle subunits from humans, Drosophila, and Arabidopsis Middle
Saccharomyces cerevisiae MED1 (Med1) MED4 (Med4) MED5 (Nut1) MED7 (Med7) MED9 (Cse2/Med9) MED10 (Nut2/Med10) MED21 (Srb7) MED31 (Soh1)
Arabidopsis thaliana
Drosophila melanogaster
Homo sapiens
At5g02850 (B)
Trap220 Trap36
TRAP220-ARC/DRIP205 TRAP36-ARC/DRIP36
Med7
ARC/DRIP34-CRSP33
CG5134 Nut2
Med25 hNut2-hMed10
Trap19 Trap18
HSrb7 hSoh1
At5g03220 (B) At5g03500 At5g41910 (B) At1g26665.1/.2a At4g04780 (B) At5g19910 (B)
Note: The unified nomenclature proposed by Bourbon et al. (2004) is listed in the S. cerivisiae column along with the yeast homolog in parentheses. Genes identified in Boube et al. (2002) are followed with ‘(B)’. Alternate splice variants are designated by ‘a’. Arabidopsis genes are according to the AGI annotators and are the TAIR locus designation (http://www.Arabidopsis.org/).
Table 1.3 Mediator head subunits from humans, Drosophila, and Arabidopsis Head Saccharomyces cerevisiae
Arabidopsis thaliana
Drosophila melanogaster
Homo sapiens
MED6 MED8 MED11 MED17 (Srb4) MED18 (Srb5) MED19 (Rox3) MED20 (Srb2)
At3g21350 (B)
Med6 ARC32 Med21 Trap80 P28/CG14802 CG5546 Trfp
HMed6-ARC/DRIP33 ARC32 HSPC296 TRAP80-ARC/DRIP77 p28b LCMR1 hTrfp
Med24
Surf5
MED22 (Srb6)
At5g20170 (B) At2g22370 (B) At4g09070 (B) At2G28230 At1g07950 (B) At1g16430
Note: The unified nomenclature proposed by Bourbon et al. (2004) is listed in the S. cerivisiae column along with the yeast homolog in parentheses. Genes identified in Boube et al. (2002) are followed with ‘(B)’. Arabidopsis genes are according to the AGI annotators and are the TAIR locus designation (http://www.Arabidopsis.org/).
Table 1.4 Mediator CDK8 subunits from humans, Drosophila, and Arabidopsis CDK8 Saccharomyces cerevisiae
Arabidopsis thaliana
Drosophila melanogaster
Homo sapiens
CDK8 (Srb10) CycC (Srb11) MED12 (Srb8) MED13 (Srb9)
At5g63610 At5g48640 At4g00450 (B) At1g55325 (B)
Dk8 CycC Kto Skd/Pap/Bli
CDK8 CycC TRAP230 ARC/DRIP240 TRAP240 ARC/DRIP250
Note: The CDK8 module is dissociable and thought to be involved in repression of gene expression. The unified nomenclature proposed by Bourbon et al. (2004) is listed in the S. cerivisiae column along with the yeast homolog in parentheses. Genes identified in Boube et al. (2002) are followed with ‘(B)’. Arabidopsis genes are according to the AGI annotators and are the TAIR locus designation (http://www.Arabidopsis.org/).
Table 1.5 Unique mediator subunits from humans, Drosophila, and Arabidopsis Unique
Arabidopsis thaliana
Drosophila melanogaster
Homo sapiens
MED23 MED24 MED25 MED26
At1g23230.1
TRAP150β TRAP100 Arc92 Arc70
hSUR2/CRSP130 TRAP/CRSP/DRIP100 ARC92 CRSP70-ARC70
Trap37 Med23 Intersex Trap25
TRAP37-CRSP34 Fksg20 Hintersex TRAP25
MED27 MED28 MED29 MED30
At1g25540.1 (PFT1) At3g48060.1 At3g48050.1/.2a At3g09180.1
Note: These subunits are not present in yeast and, therefore, have been designated as ‘unique’ (Bourbon et al., 2004). Alternate splice variants are designated by ‘a’. Arabidopsis genes are according to the AGI annotators and are the TAIR locus designation (http://www.Arabidopsis.org/).
GENERAL TRANSCRIPTION FACTORS AND THE CORE PROMOTER
15
as well as phosphorylation of the CTD of RNA pol II (Shim et al., 2002). Arabidopsis MED14 has strong homology to MED14 (yeast Rgr1) in other organisms (Autran et al., 2002). Mutations in Arabidopsis MED14 (yeast Rgr1) have drastic developmental consequences. Analysis of a set of mutants with abnormal cell division patterns and cell number identified a recessive mutation named ‘STRUWWELPETER’ (SWP) (Autran et al., 2002). SWP plants have reduced cell numbers in all aerial organs and exhibit abnormal meristem function. SWP seems to define the duration of the cell proliferation phase of leaf primordia, but not affect cell division rates. Other abnormalities in the shoot indicated that it is a highly pleiotropic mutation; however, there were no apparent abnormalities in root growth or development despite SWP expression in the root (Autran et al., 2002). As may be expected, a recessive epigenetic mutation in two step II splicing factors that seems to specifically effect the expression of SWP has a similar phenotype to SWP (Clay and Nelson, 2005).
1.6.2
Middle complex
The middle and head complexes of mediator have been implicated by structural EM studies to make direct contact with RNA pol II (Chadick and Asturias, 2005). The middle module has been shown in yeast to have affinity for the CTD of the largest subunit of RNA pol II (Kang et al., 2001), participating in both activation and repression (Han et al., 2001; Kang et al., 2001). The middle complex is composed of two submodules designated MED9 (MED9, 1 and 4 subunits) and MED10 (MED10, 7 and 21) (Boube et al., 2002). These two submodules are thought to act in opposite ways: the MED9 subcomplex is involved in repression, while the MED10 subcomplex activates transcription (Han et al., 2001; Kang et al., 2001). Arabidopsis orthologs to components of the middle complex have been identified for five out of the eight subunits present in other eukaryotes. For MED7 and 10, there is more uncertainty regarding their identification as bona fide subunits of the plant mediator. The Arabidopsis MED7 (Boube et al., 2002) is nearly identical along its entire length to the N-terminus of the other listed Arabidopsis protein, At5g030500. This putative alternate form of MED7 has an RNA-binding domain (RBD) in its C-terminus with similarity to RBDs found in plant nucleolins (Bogre et al., 1996). The two listed Arabidopsis MED10 (Nut2) candidates show comparable amino acid sequence similarity to MED10 present in other eukaryotes. The involvement of MED10 in GCN4-mediated transcription in yeast raises the possibility that the MED10 submodule may play a wider role in transcriptional activation (Han et al., 1999). Less is known concerning the roles of the other members of the MED10 submodule in specific gene regulation. The MED21 (Srb7) Arabidopsis homolog is larger than MED10 and is similar to metazoan SRB7 at its C-terminus. In addition, it contains a GRF zinc-finger domain at its N-terminus. This presumed zinc-binding domain is found in a variety of DNA-binding proteins and may be involved in nucleic acid binding. MED31 (Soh1), long recognized as a component of human and Drosophila mediator, was recently identified as a genuine mediator subunit in yeast (Linder and Gustafsson, 2004). Like many of the mediator subunits, only a relatively
16
REGULATION OF TRANSCRIPTION IN PLANTS
small domain of the protein has been conserved across evolutionary distance: a 69-amino acid region that is rich in leucine and aromatic residues (Linder and Gustafsson, 2004). This conserved motif appears to be a protein interaction domain required for the association of MED31 with the rest of the mediator complex.
1.6.3
Head module
The head module (Srb4 module) plays a major role in binding RNA pol II. Early genetic studies strongly suggested the possibility of interactions between the CTD of RNA pol II and the Srb proteins, many of which reside in the head module (Thompson et al., 1993). In vitro studies later demonstrated direct binding of the Srb4, MED9, and MED10 modules to the CTD of pol II (Kang et al., 2001). EM imaging of highly purified yeast RNA pol II–mediator complexes suggests that multiple contacts between the head module and polymerase are likely (Asturias et al., 1999). It is unknown whether these contacts serve a strictly structural purpose, or if mediator may impart functionally significant conformational change to the polymerase. Although attachment of mediator to RNA pol II is a major function of the head module, it also serves as a target for recruitment by the acidic activation domain of the yeast Gal4 transactivator (Koh et al., 1998). Based on experiments in yeast, the MED17 (Srb4) subunit appears to exert global effects on transcription. For example, the nearly absolute requirement of MED17 was demonstrated by the observation that a temperature-sensitive mutation (srb4ts) was as detrimental to overall transcription as a mutation in RNA pol II itself (Holstege et al., 1998). Two head module subunits, MED20 and MED22, may have multiple isoforms in plants. The MED20 (Srb2) Arabidopsis homolog (encoded by At4g09070) does not have significant similarity to yeast MED20, but is clearly related to human and Drosophila homologs. In addition, we identified another MED20 gene in Arabidopsis, At2G28230, that does show significant similarity to both human and Drosophila genes and is highly similar to At4g09070. The third putative MED20 gene, At2g28020.1, shows a high degree of similarity that is restricted to the C-termini of the other two Arabidopsis paralogs. Interestingly, some head module genes are not found in Arabidopsis (such as MED8 and MED19), or have been assigned to the mediator by others (Boube et al., 2002), but are not confirmed by our analyses (i.e., MED11). The MED6 homolog in Arabidopsis shows strong similarity to the metazoan subunit. Mutants of MED6 in Drosophila exhibit severe developmental abnormalities, and the expression of many, but not all, developmentallyregulated genes are affected (Gim et al., 2001).
1.6.4
CDK8/Srb8-11 module
The fourth mediator module, CDK8/Srb8-11, appears to be reversibly bound and has been closely linked to transcriptional repression, both in yeast (Kuchin and Carlson, 1998; van de Peppel et al., 2005) and in metazoans (Mo et al., 2004; Yoda et al., 2005). This module, containing a kinase that phoshorylates the CTD of RNA pol II, may repress transcription by inhibiting the association between mediator and the
GENERAL TRANSCRIPTION FACTORS AND THE CORE PROMOTER
17
CTD. Mediator containing the Srb8-11 module from Shizosaccharomyces pombe can be isolated only in free form, not complexed with RNA pol II (Samuelsen et al., 2003). This is also the case in mammalian systems where the larger form of mediator, ARC-L, is unable to bind the CTD of RNA pol II, unlike CRSP, the smaller transcriptionally active form, which can (Naar et al., 2002). Genes encoding all subunits of the CDK8 module have been putatively identified in Arabidopsis and are listed below in Table 1.4.
1.6.5
Mediator subunits unique to metazoans and plants
A number of mediator subunits have been identified in metazoans that are not found in yeast (Bourbon et al., 2004) and, hence, could not be assigned to a specific module. We conducted reciprocal Psi blast searches (Altschul et al., 1997) of the A. thaliana database using human and Drosophila homologs to identify putative homologs of these subunits in Arabidopsis listed in the following table. It is interesting that several of these subunits have affinity for transactivators. For example, MED23 interacts with a number of transcription factors including C/EBPβ where it mediates C/EBPβ interaction with both transcriptionally active and inactive mediator complexes (Mo et al., 2004). Note that MED23 is the new designation of the Sur2/CRSP130 subunit originally misidentified as the homolog of Gal11 (Boube et al., 2002). Mediator in metazoans seems to be present in two distinct forms: one involved in gene activation and one associated with repression. The repressor form of mediator (ARC-L) is larger in mass than the activation-type mediator exemplified by CRSP, which was first characterized as a coactivator complex required for Sp1 activation (Ryu et al., 1999). Repressor-type mediators contain the CDK8 module and lack MED26. In contrast, activation-type mediators (i.e., CRSP) lack the CDK8 module, but contain the MED26 subunit. Therefore, MED26 is thought to be essential for the activation function. MED26 in metazoans and in Arabidopsis (At3g48060 and At3g48050.1/2) contains an N-terminal domain (Domain I) that forms a four-helix cluster that is also found in transcription elongation factor SII (TFIIS) and in elongin A (Booth et al., 2000). Both Arabidopsis homologs contain an additional domain, BAH/ELM1/BAM (bromoadjacent homology or motif), that may be a protein– protein interaction domain and is present in a number of proteins involved in gene transcription and repression (Goodwin and Nicolas, 2001). The Arabidopsis MED26 homologs appear to represent a gene duplication, one of which (At3g48050.1/2) has a splice variant. MED25 is a member of the ARC mediator complex and has been identified as the target subunit of the VP16 transactivation domain (Mittler et al., 2003; Yang et al., 2004). In Arabidopsis, MED25 (At1g25540) encodes the nuclear protein PFT1, which is involved in the regulation of flowering time downstream of phyB in a photoperiod-independent manner (Cerdan and Chory, 2003). PFT1 also modulates the expression of the floral integrator gene FT, consistent with its role as a mediator subunit. Metazoan MED25 interactions with the mediator complex are mediated through a von Willebrand type A domain (vWF-A), shown to be involved in protein– protein interactions (Hinshelwood et al., 1999; Mittler et al., 2003 ). Arabidopsis
18
REGULATION OF TRANSCRIPTION IN PLANTS
PFT1 also has this interaction domain, as well as a glutamine-rich region similar to the one present in Drosophila MED25. Overall, plant mediator appears to share a subunit composition similar to that characteristic of metazoans. This is based on the finding that four out of the eight subunits present in metazoans, but not in yeast, have putative orthologs in Arabidopsis. It will be interesting to see if the ‘missing’ metazoan subunits are actually present in plants (as highly diverged forms), lack additional subunits altogether, or whether plants possess truly unique mediator subunits.
1.7
Transcription factor IIB
Transcription factor IIB is a key member of the PIC from archaea through higher eukaryotes. The highly conserved C-terminal half of TFIIB (TFIIB core) is composed of two imperfectly repeated domains (R1 and R2) that show similarity to cyclin. When in the PIC with TBP, TFIIB makes contact upstream of the TATA in the major groove of the BRE and downstream of TATA in the minor groove (Nikolov et al., 1995; Lagrange et al., 1998; Tsai and Sigler, 2000). The N-terminus of TFIIB contains two highly conserved domains of approximately 30 residues each: the zinc ribbon and B-finger. These N-terminal domains protrude into the cleft of RNA pol II that forms the active site, and the TFIIB core makes contact with the C-terminal stirrup of TBP in the region between the two cyclin folds. The zinc ribbon binds to the dock region of pol II (Rbp1 subunit) and overlaps the RNA exit channel, and the adjacent B-finger forms a loop structure that fills the space between the RNA exit channel and the catalytic site of the enzyme. The B-finger is positioned in close proximity to the 8-bp RNA–DNA hybrid within the transcription bubble and is thought to play a role in start-site selection (Bushnell et al., 2004; Chen and Hahn, 2004). The B-finger and TFIIB core domains are also in close contact with TFIIF on the surface of pol II and within the active site (Freire-Picos et al., 2005). These intimate associations between TBP, TFIIB, and RNA polymerase seem to be conserved from archaea to higher eukaryotes, including plants. There are two families of TFIIB type proteins: TFIIB which is associated with RNA pol II transcription, and those associated with RNA pol III transcription known as B-related factors (BRFs). There are at least three of BRF genes in Arabidopsis: At2g45100, At3g09360, and At2g01280. Although TFIIB has specialized to function with RNA polymerases II and III, a single type of TBP functions with all three eukaryotic RNA polymerases. Plants are unusual in that they have multiple TFIIBs compared to the single proteins in metazoans and yeast. For example, in Arabidopsis there are five general classes of TFIIB based on phylogenetic analysis of the R2 core domain (Lawit, 2003). There appears to be a group of closely related TFIIBs found in all plants that we designate as Type I. This group is presumed to represent the main TFIIB based on the observation that it is, by far, the most highly conserved among plants and, therefore, is most likely involved in expression of a largest percentage of the genes. Mutational changes in the Type I TFIIBs are expected to be very constrained due to the wide-ranging affects mutations would impose on all aspects of
GENERAL TRANSCRIPTION FACTORS AND THE CORE PROMOTER
19
growth and development. At present, Type I TFIIB genes include two from Arabidopsis (At3g10330 and At2g41630), two from poplar (B3 and B1), and one each from grape (TC19782), soybean (U 31097), alfalfa (TC86832), citrus (CB292941), rice (AF464908), and wheat (TC68795). Plants also have several other more highly diverged TFIIB genes, which we designate as Type II. There are four Type II TFIIB genes in rice (PO430F03.39, AC136492, AAN59779, and PO430F03.36) and two in Arabidopsis (At3g57370 and At3g29380). From the limited amount of DNA sequence data available in plants, it seems that the number of Type I genes may normally be one to two, while the number of Type II genes may be more variable between plant species (four in rice; two in Arabidopsis) (Lawit, 2003). Plants also contain a very unusual class of TFIIB known as pBrp (TFIIB5; Type III) that is associated with the plastid (Lagrange et al., 2003). Phylogenetic analysis of the R2 domain suggests that this protein is quite ancient, having a branch point near those of yeast TFIIB and archaeal TFB (Lagrange et al., 2003; Lawit, 2003). Studies in Arabidopsis revealed that pBrp is predominantly located on the outer surface of the plastid envelop (Lagrange et al., 2003). In addition, pBrp contains PEST-related sequences in the linker domain of the N-terminus and at the C-terminus beyond the TFIIB core region. The PEST motif targets proteins for rapid turnover by the proteasomes. Inhibition of proteasome function by chemical treatments, or by utilizing mutants in COP9, resulted in the accumulation of pBrp in the nucleus. Lagrange et al. (2003) have suggested from these results that pBrp is controlled by nucleocytoplasmic partitioning which is dependent on proteasome degradation by an unknown process. The close association with the plastid cyosolic face supports the possibility that pBrp may be uniquely involved in the expression of nuclear-encoded plastid genes. The existence of three types of TFIIB, and the existence of multiple paralogs within these types, strongly implies that TFIIB has undergone a much higher degree of specialization in plants compared to metazoans and yeast. It will be interesting to test this hypothesis in the future and determine the parameters of promoter recognition that may have been exploited in the process. For example, is there a pBrp-specific BRE, and are there other GTFs or transactivators that specifically function in association with pBrp? Along these lines, perhaps pBrp functions with a novel form of RNA pol II. This is not as speculative as it may seem in view of the degree of divergence between pBrp proteins and other TFIIBs present in the pBrp zinc ribbon and the low level of similarity with the canonical B-finger at the amino acid level (Lawit, 2003). This possibility is strengthened by the presence of a novel RNA pol II Rbp2 subunit in Arabidopsis (Nature, 2000).
1.8
Summary
Higher plants have much in common with metazoans and yeast with respect to gene and promoter structure and with respect to the composition of the basal transcriptional apparatus. The core promoter in plants contains the TATA and Inr elements, but appears to lack other motifs such as the DPE and BRE. In addition, the TATAproximal CAAT box and its cognate transactivator (NF-Y) first characterized in
20
REGULATION OF TRANSCRIPTION IN PLANTS
mammals is also present in plants. It is expected that in plants, as in other eukaryotes, upstream elements are often distributed in module-like clusters and TATA-proximal modules exert a dominant influence on transcription. These common features of promoter structure are consistent with the assumption that the fundamental mechanisms of gene activation and repression have been conserved among the eukaryotes. A survey of completed plant genomes reveals that genes encoding subunits for the typical assemblage of GTFs (TFIIA, B, D, E, F, and H) are present. While most metazoans and the fungi usually possess a single gene encoding the GTFs and their respective subunits, plants are able to accommodate limited amounts of gene duplication and specialization. For example, the Arabidopsis genome contains two copies of TBP, and TAFs 1, 4, 6, 11, and 12. Of these, the additional copy of TAFs 6, 11, and 12 seems to be derived from the ancient genome duplication that doubled approximately 70% of the Arabidopsis genome (Nature, 2000; Lago et al., 2004). However, an even more ancient event is responsible for the duplication of TAFs 1 and 4, since these are present in two copies in monocots, metazoans, and yeast, and are not located in the regions of the genome that were duplicated in Arabidopsis (Lago et al., 2004). This ability of plants to tolerate, and possibly exploit, multiple alleles of GTFs is best illustrated with TFIIB, where widely divergent forms exist, and one (pBrp) is plastid associated (Lagrange et al., 2003). At present the biological significance of having multiple genes encoding a GTF is unknown; however, the expectation is that several have specialized for cell-type or developmentally specific roles. Many of the redundant GTFs in plants are not universal in distribution. For example, TAFs 6, 11, and 12, which are duplicated in Arabidopsis, are present only as single alleles in rice (Lago et al., 2004). This raises several possibilities: that ancient duplications of the plant GTFs may not be maintained through evolution, or that the duplication of plant GTFs is an ongoing process that may lead to the development of novel patterns of expression. Even though the underlying mechanisms of promoter structure and gene regulation can be traced back to the pre-eukaryotic period and are shared by all eukaryotes, much remains to be discovered regarding plant-specific mechanisms by which GTFs regulate gene expression. The technical difficulties inherent to in vitro transcription with plant systems have been a barrier for biochemical characterization of the complex machinery of transcription. However, the rapid accumulation of whole genome sequence information for a growing number of organisms has greatly facilitated the identification of genes encoding the major components of basal transcription in plants based on the conservation of functional domains among the eukaryotes. In some cases, orthologous genes corresponding to GTFs and their subunits can clearly be identified in plants; however, some subunits are more conserved in structure than in amino acid sequence. In these instances (some TAFs and mediator subunits), only small blocks (20–30 amino acids) of similarity in coding sequence can be recognized, and these domains are often also present in multiple proteins not associated with transcription. A future challenge is to immunopurify the multisubunit GTFs found in plants using epitope-tagged proteins introduced into transformed plants in order to confirm predictions in subunit composition and, possibly, discover novel plant-specific subunits. A strategy employing targeted mutagenesis of specific GTF subunits may ultimately provide information that will enable us to better understand
GENERAL TRANSCRIPTION FACTORS AND THE CORE PROMOTER
21
how plants as multicellular organisms have exploited the ancient mechanisms of transcription in their unique exploration of a sessile, land-adapted lifestyle.
References S.R. Albright and R. Tjian (2000) TAFs revisited: more data reveal new twists and confirm old ideas. Gene 242, 1–13. S. Altschul, T. Madden, A. Schaffer, et al. (1997) Gapped blast and psi-blast: a new generation of protein database search programs. Nucleic Acids Research 25, 3389–3402. N. Aoyagi and D.A. Wassarman (2001) Developmental and transcriptional consequences of mutations in Drosophila TAF(II)60. Molecular and Cellular Biology 21, 6808–6819. F.J. Asturias, Y.W. Jiang, L.C. Myers, C.M. Gustafsson and R.D. Kornberg (1999) Conserved structures of mediator and RNA polymerase II holoenzyme. Science 283, 985–987. D. Autran, C. Jonak, K. Belcram, et al. (2002) Cell numbers and leaf development in Arabidopsis: a functional analysis of the STRUWWELPETER gene. The EMBO Journal 21, 6036–6049. S. Bagby, T.K. Mal, D. Liu, E. Raddatz, Y. Nakatani and M. Ikura (2000) TFIIA-TAF regulatory interplay: NMR evidence for overlapping binding sites on TBP. FEBS Letters 468, 149–154. A.D. Baxevanis, G. Arents, E.N. Moudrianakis and D. Landsman (1995) A variety of DNA-binding and multimeric proteins contain the histone fold motif. Nucleic Acids Research 23, 2685–2691. J.D. Becker, L.C. Boavida, J. Carneiro, M. Haury and J.A. Feijo (2003) Transcriptional profiling of Arabidopsis tissues reveals the unique characteristics of the pollen transcriptome. Plant Physiology 133, 713–725. B. Bell, E. Scheer and L. Tora (2001) Identification of hTAF(II)80 delta links apoptotic signaling pathways to transcription factor TFIID function. Molecular Cell 8, 591–600.S.D. Bell, P.L. Kosa, P.B. Sigler and S.P. Jackson (1999) Orientation of the transcription preinitiation complex in archaea. Proceedings of the National Academy of Sciences of the United States of America 96, 13662–13667. M. Bellorini, D.K. Lee, J.C. Dantonel, et al. (1997) CCAAT binding NF-Y-TBP interactions: NF-YB and NF-YC require short domains adjacent to their histone fold motifs for association with TBP basic residues. Nucleic Acids Research 25, 2174–2181. C. Benoist, K. O’Hare, R. Breathnach and P. Chambon (1980) The ovalbumin gene-sequence of putative control regions. Nucleic Acids Research 8, 127–142. C. Bertrand, M. Benhamed, Y.-F. Li, et al. (2005) Arabidopsis HAF2 gene encoding TATA-binding protein (TBP)-associated factor TAF1, is required to integrate light signals to regulate gene expression and growth. The Journal of Biological Chemistry 280, 1465–1473. S. Bjorklund and C.M. Gustafsson (2005) The yeast Mediator complex and its regulation. Trends in Biochemical Sciences 30, 240–244. E.M. Blackwood and J. Kadonaga (1998) Going the distance: a current view of enhancer action. Science 281, 60–63. L. Bogre, C. Jonak, M. Mink, et al. (1996) Developmental and cell cycle regulation of alfalfa nucMs1, a plant homolog of the yeast Nsr1 and mammalian nucleolin. Plant Cell 8, 417–428. V. Booth, C.M. Koth, A.M. Edwards and C.H. Arrowsmith (2000) Structure of a conserved domain common to the transcription factors TFIIS, elongin A, and CRSP70. The Journal of Biological Chemistry 275, 31266–31268. M. Boube, L. Joulia, D.L. Cribbs and H.M. Bourbon (2002) Evidence for a mediator of RNA polymerase II transcriptional regulation conserved from yeast to man. Cell 110, 143–151. H.-M. Bourbon, A. Aguilera, A.Z. Ansari, et al. (2004) A unified nomenclature for protein subunits of mediator complexes linking transcriptional regulators to RNA polymerase II. Molecular Cell 14, 553–557. P. Bucher (1990) Weight matrix descriptions of four eukaryotic RNA polymerase II promoter elements derived from 502 unrelated promoter sequences. Journal of Molecular Biology 212, 563–578. M. Bulger and M. Groudine (1999) Looping versus linking: toward a model for long-distance gene activation. Genes and Development 13, 2465–2477.
22
REGULATION OF TRANSCRIPTION IN PLANTS
T.W. Burke and J.T. Kadonaga (1996) Drosophila TFIID binds to a conserved downstream basal promoter element that is present in many TATA-box-deficient promoters. Genes and Development 10, 711–724. T.W. Burke and J.T. Kadonaga (1997) The downstream core promoter element, DPE, is conserved from Drosophila to humans and is recognized by TAFII60 of Drosophila. Genes and Development 11, 3020–3031. S.K. Burley and R.G. Roeder (1996) Biochemistry and structural biology of transcription factor IID (TFIID). Annual Review of Biochemistry 65, 769–799. S. Busby and R.H. Ebright (1999) Transcription activation by catabolite activator protein (CAP). Journal of Molecular Biology 293, 199–213. D.A. Bushnell, K.D. Westover, R.E. Davis and R.D. Kornberg (2004) Structural basis of transcription: an RNA polymerase II-TFIIB cocrystal at 4.5 Angstroms. Science 303, 983–988. K.M. Campbell and K.J. Lumb (2002) Structurally distinct modes of recognition of the KIX domain of CBP by Jun and CREB. Biochemistry 41, 13956–13964. J. Carcamo, L. Buckbinder and D. Reinberg (1991) The initiator directs the assembly of a transcription factor IID-dependent transcription complex. Proceedings of the National Academy of Sciences of the United States of America 88, 8052–8056. P.D. Cerdan and J. Chory (2003) Regulation of flowering time by light quality. Nature 423, 881–885. J.Z. Chadick and F.J. Asturias (2005) Structure of eukaryotic mediator complexes. Trends in Biochemical Sciences 30, 264–271. G.E. Chalkley and C.P. Verrijzer (1999) DNA binding site selection by RNA polymerase II TAFs: a TAF(II)250–TAF(II)150 complex recognizes the initiator. The EMBO Journal 18, 4835–4845. H.T. Chen and S. Hahn (2004) Mapping the location of TFIIB within the RNA polymerase II transcription preinitiation complex: a model for the structure of the PIC. Cell 119, 169–180. N.K. Clay and T. Nelson (2005) The recessive epigenetic swellmap mutation affects the expression of two step II splicing factors required for the transcription of the cell proliferation gene STRUWWELPETER and for the timing of cell cycle arrest in the Arabidopsis leaf. Plant Cell 17, 1994–2008. R.C. Conaway, S. Sato, C. Tomomori-Sato, T. Yao and J.W. Conaway (2005) The mammalian mediator complex and its role in transcriptional regulation. Trends in Biochemical Sciences 30, 250–255. F. Coustry, S.N. Maity and B. de Crombrugghe (1995) Studies on transcription activation by the multimeric CCAAT-binding factor CBF. The Journal of Biological Chemistry 270, 468–475. F. Coustry, S.N. Maity, S. Sinha and B. de Crombrugghe (1996) The transcriptional activity of the CCAAT-binding factor CBF is mediated by two distinct activation domains, one in the CBF-B subunit and the other in the CBF-C subunit. The Journal of Biological Chemistry 271, 14485– 14491. P.A. Covitz, W. Song and A.P. Mitchell (1994) Requirement for RGR1 and SIN4 in RME1-dependent repression in Saccharomyces cerevisiae. Genetics 138, 577–586. J.M. Cox, M.M. Hayward, J.F. Sanchez, et al. (1997) Bidirectional binding of the TATA box binding protein to the TATA box. Proceedings of the National Academy of Sciences of the United States of America 94, 13475–13480. R.N. Freiman, S.R. Albright, S. Zheng, W.C. Sha, R.E. Hammer and R. Tjian (2001) Requirement of tissue-selective TBP-associated factor TAFII105 in ovarian development. Science 293, 2084– 2087. M.A. Freire-Picos, S. Krishnamurthy, Z.W. Sun and M. Hampsey (2005) Evidence that the Tfg1/Tfg2 dimer interface of TFIIF lies near the active center of the RNA polymerase II initiation complex. Nucleic Acids Research 33, 5045–5052. T. Furumoto, Y. Tamada, A. Izumida, H. Nakatani, S. Hata and K. Izui (2005) Abundant expression in vascular tissue of plant TAF10, an orthologous gene for TATA box-binding protein-associated factor 10, in Flaveria trinervia and abnormal morphology of Arabidopsis thaliana transformants on its overexpression. Plant and Cell Physiology 46, 108–117. Y.G. Gangloff, C. Romier, S. Thuault, S. Werten and I. Davidson (2001) The histone fold is a key structural motif of transcription factor TFIID. Trends in Biochemical Sciences 26, 250–257.
GENERAL TRANSCRIPTION FACTORS AND THE CORE PROMOTER
23
E.P. Geiduschek and M. Ouhammouch (2005) Archaeal transcription and its regulators. Molecular Microbiology 56, 1397–1407. S. Georgieva, D.B. Kirschner, T. Jagla, et al. (2000) Two novel Drosophila TAF(II)s have homology with human TAF(II)30 and are differentially regulated during development. Molecular and Cellular Biology 20, 1639–1648. B.S. Gim, J.M. Park, J.H. Yoon, C. Kang and Y.-J. Kim (2001) Drosophila Med6 is required for elevated expression of a large but distinct set of developmentally regulated genes. Molecular and Cellular Biology 21, 5242–5255. G.H. Goodwin and R.H. Nicolas (2001) The BAH domain, polybromo and the RSC chromatin remodelling complex. Gene 268, 1–7. P.A. Grant, D. Schieltz, M.G. Pray-Grant, et al. (1998) A subset of TAF(II)s are integral components of the SAGA complex required for nucleosome acetylation and transcriptional stimulation. Cell 94, 45–53. M.R. Green (2000) TBP-associated factors (TAFIIs): multiple, selective transcriptional mediators in common complexes. Trends in Biochemical Sciences 25, 59–63. M. Guermah, K. Ge, C.M. Chiang and R.G. Roeder (2003) The TBN protein, which is essential for early embryonic mouse development, is an inducible TAFII implicated in adipogenesis. Molecular Cell 12, 991–1001. G. Gusmaroli, C. Tonelli and R. Mantovani (2001) Regulation of the CCAAT-binding NF-Y subunits in Arabidopsis thaliana. Gene 264, 173–185. G. Gusmaroli, C. Tonelli and R. Mantovani (2002) Regulation of novel members of the Arabidopsis thaliana CCAAT-binding nuclear factor Y subunits regulation of the CCAAT-binding NF-Y subunits in Arabidopsis thaliana. Gene 283, 41–48. S.J. Han, J.S. Lee, J.S. Kang and Y.J. Kim (2001) Med9/Cse2 and Gal11 modules are required for transcriptional repression of distinct group of genes. The Journal of Biological Chemistry 276, 37020–37026. S.J. Han, Y.C. Lee, B.S. Gim, et al. (1999) Activator-specific requirement of yeast mediator proteins for RNA polymerase II transcriptional activation. Molecular and Cellular Biology 19, 979– 988. M. Hiller, X. Chen, M.J. Pringle, et al. (2004) Testis-specific TAF homologs collaborate to control a tissue-specific transcription program. Development 131, 5297–5308. J. Hinshelwood, D.I.R. Spencer, Y.J.K. Edwards and S.J. Perkins (1999) Identification of the c3b binding site in a recombinant vWF-A domain of complement factor B by surface-enhanced laser desorption-ionisation affinity mass spectrometry and homology modelling: implications for the activity of factor B. Journal of Molecular Biology 294, 587–599. F.C.P. Holstege, E.G. Jennings, J.J. Wyrick, et al. (1998) Dissecting the regulatory circuitry of a eukaryotic genome. Cell 95, 717–728. R.A. Hooft van Huijsduijnen, J. Bollekens, A. Dorn, C. Benoist and D. Mathis (1987) Properties of a CCAAT box-binding protein. Nucleic Acids Research 15, 7265–7282. R.H. Jacobson, A.G. Ladurner, D.S. King and R. Tjian (2000) Structure and function of a human TAFII250 double bromodomain module. Science 288, 1422–1425. Y.W. Jiang, P.R. Dohrmann and D.J. Stillman (1995) Genetic and physical interactions between yeast RGR1 and SIN4 in chromatin organization and transcriptional regulation. Genetics 140, 47–54. J.S. Kang, S.H. Kim, M.S. Hwang, S.J. Han, Y.C. Lee and Y.J. Kim (2001) The structural and functional organization of the yeast mediator complex. The Journal of Biological Chemistry 276, 42003– 42010. Y.-J. Kim and J.T. Lis (2005) Interactions between subunits of Drosophila mediator and activator proteins. Trends in Biochemical Sciences 30, 245–249. S.S. Koh, A.Z. Ansari, M. Ptashne and R.A. Young (1998) An activator target in the RNA polymerase II holoenzyme. Molecular Cell 1, 895–904. P.B. Komarnitsky, B. Michel and S. Buratowski (1999) TFIID-specific yeast TAF40 is essential for the majority of RNA polymerase II-mediated transcription in vivo. Genes and Development 13, 2484–2489.
24
REGULATION OF TRANSCRIPTION IN PLANTS
R.D. Kornberg (2005) Mediator and the mechanism of transcriptional activation. Trends in Biochemical Sciences 30, 235–239. S. Kuchin and M. Carlson (1998) Functional relationships of Srb10-Srb11 kinase, carboxy-terminal domain kinase CTDK-I, and transcriptional corepressor Ssn6-Tup1. Molecular and Cellular Biology 18, 1163–1171. C. Lago, E. Clerici, L. Dreni, et al. (2005) The Arabidopsis TFIID factor AtTAF6 controls pollen tube growth. Developmental Biology 285, 91–100. C. Lago, E. Clerici, L. Mizzi, L. Colombo and M.M. Kater (2004) TBP-associated factors in Arabidopsis. Gene 342, 231–241. T. Lagrange, M.A. Hakimi, D. Pontier, et al. (2003) Transcription factor IIB (TFIIB)-related protein (pBrp), a plant-specific member of the TFIIB-related protein family. Molecular and Cellular Biology 23, 3274–3286. T. Lagrange, A.N. Kapanidis, H. Tang, D. Reinberg and R.H. Ebright (1998) New core promoter element in RNA polymerase II-dependent transcription: sequence-specific DNA binding by transcription factor IIB. Genes and Development 12, 34–44. S. Lawit (2003) Protein–protein interaction map of Arabidopsis thaliana transcription factors A, B, D, E and F. In: Plant Molecular and Cellular Biology, p. 262. University of Florida, Gainesville, FL. D.-K. Lee, S. Kim and J.T. Lis (1999) Different upstream transcriptional activators have distinct coactivator requirements. Genes and Development 13, 2934–2939. T.I. Lee and R.A. Young (1998) Regulation of gene expression by TBP-associated proteins. Genes and Development 12, 1398–1408. C. Leurent, S.L. Sanders, M.A. Demeny, et al. (2004) Mapping key functional sites within yeast TFIID. The EMBO Journal 23, 719–727. X.Y. Li, S.R. Bhaumik, X. Zhu, et al. (2002) Selective recruitment of TAFs by yeast upstream activating sequences. Implications for eukaryotic promoter structure. Current Biology 12, 1240–1244. Y.F. Li, F. Dubois and D.X. Zhou (2001) Ectopic expression of TATA box-binding protein induces shoot proliferation in Arabidopsis. FEBS Letters 489, 187–191. T. Linder and C.M. Gustafsson (2004) The Soh1/MED31 protein is an ancient component of Schizosaccharomyces pombe and Saccharomyces cerevisiae mediator. The Journal of Biological Chemistry 279, 49455–49459. D. Liu, R. D., Ishima, K.I. R., Tong, et al. (1998) Solution structure of a TBP–TAF(II)230 complex: protein mimicry of the minor groove surface of the TATA box unwound by TBP. Cell 94, 573–583. Y.-P. Liu, C.-W. Chang and K.-Y. Chang (2003) Mutational analysis of the KIX domain of CBP reveals residues critical for SREBP binding. FEBS Letters 554, 403–409. T.K. Mal, Y. Masutomi, L. Zheng, et al. (2004) Structural and functional characterization on the interaction of yeast TFIID subunit TAF1 with TATA-binding protein. Journal of Molecular Biology 339, 681–693. S. Malik and R.G. Roeder (2005) Dynamic regulation of pol II transcription by the mammalian mediator complex. Trends in Biochemical Sciences 30, 256–263. R. Mantovani (1998) A survey of 178 NF-Y binding CCAAT boxes. Nucleic Acids Research 26, 1135–1143. R. Mantovani (1999) The molecular biology of the CCAAT-binding factor NF-Y. Gene 239, 15–27. L. Margulis, M.F. Dolan and R. Guerrero (2000) The chimeric eukaryote: origin of the nucleus from the karyomastigont in amitochondriate protists. Proceedings of the National Academy of Sciences of the United States of America 97, 6954–6959. J. Martin, R. Halenbeck and J. Kaufmann (1999) Human transcription factor hTAF(II)150 (CIF150) is involved in transcriptional regulation of cell cycle progression. Molecular and Cellular Biology 19, 5548–5556. M. Metsis, A. Brunkhorst and T. Neuman (2001) Cell-type-specific expression of the TFIID component TAF(II)135 in the nervous system. Experimental Cell Research 269, 214–221. D. Metzger, E. Scheer, A. Soldatov and L. Tora (1999) Mammalian TAF(II)30 is required for cell cycle progression and specific cellular differentiation programmes. The EMBO Journal 18, 4823–4834.
GENERAL TRANSCRIPTION FACTORS AND THE CORE PROMOTER
25
B. Michel, P. Komarnitsky and S. Buratowski (1998) Histone-like TAFs are essential for transcription in vivo. Molecular Cell 2, 663–673. G. Mittler, T. Stuhler, L. Santolin, et al. (2003) A novel docking site on Mediator is critical for activation by VP16 in mammalian cells. The EMBO Journal 22, 6494–6504. X. Mo, E. Kowenz-Leutz, H. Xu and A. Leutz (2004) Ras induces mediator complex exchange on C/EBP beta. Molecular Cell 13, 241–250. W.S. Mohan, Jr., E. Scheer, O. Wendling, D. Metzger and L. Tora (2003) TAF10 (TAF(II)30) is necessary for TFIID stability and early embryogenesis in mice. Molecular and Cellular Biology 23, 4307–4318. Z. Moqtaderi, M. Keaveney and K. Struhl (1998) The histone H3-like TAF is broadly required for transcription in yeast. Molecular Cell 2, 675–682. L.C. Myers, C.M. Gustafsson, D.A. Bushnell, et al. (1998) The med proteins of yeast and their function through the RNA polymerase II carboxy-terminal domain. Genes and Development 12, 45–54. A.M. Naar, D.J. Taatjes, W. Zhai, E. Nogales and R. Tjian (2002) Human CRSP interacts with RNA polymerase II CTD and adopts a specific CTD-bound conformation. Genes and Development 16, 1339–1344. D. Nair, Y. Kim and L.C. Myers (2005) Mediator and TFIIH govern carboxyl-terminal domaindependent transcription in yeast extracts. The Journal of Biological Chemistry 280, 33739–33748. The Arabidopsis Initiative (2000) Analysis of the genome sequence of the flowering plant Arabidopsis thaliana. Nature 408, 796–815. D.B. Nikolov and S.K. Burley (1997) RNA polymerase II transcription initiation: a structural view. Proceedings of the National Academy of Sciences of the United States of America 94, 15–22. D.B. Nikolov, H. Chen, E.D. Halay, et al. (1995) Crystal structure of a TFIIB–TBP–TATA-element ternary complex. Nature 377, 119–128. C. Nocente-McGrath, C.A. Brenner and S.G. Ernst (1989) Endo16, a lineage-specific protein of the sea urchin embryo, is first expressed just prior to gastrulation. Developmental Biology 136, 264–272. M. Novatchkova and F. Eisenhaber (2004) Linking transcriptional mediators via the GACKIX domain super family. Current Biology 14, R54–R55. M. Ouhammouch, G.E. Langham, W. Hausner, et al. (2005) Promoter architecture and response to a positive regulator of archaeal transcription. Molecular Microbiology 56, 625–637. R. Pandey, A. Muller, C.A. Napoli, et al. (2002) Analysis of histone acetyltransferase and histone deacetylase families of Arabidopsis thaliana suggests functional diversification of chromatin modification among multicellular eukaryotes. Nucleic Acids Research 30, 5036–5055. A.D. Pham, S. Muller and F. Sauer (1999) Mesoderm-determining transcription in Drosophila is alleviated by mutations in TAF(II)60 and TAF(II)110. Mechanisms of Development 84, 3–16. A.D. Pham and F. Sauer (2000) Ubiquitin-activating/conjugating activity of TAFII250, a mediator of activation of gene expression in Drosophila. Science 289, 2357–2360. D. Poon, Y. Bai, A.M. Campbell, et al. (1995) Identification and characterization of a TFIID-like multiprotein complex from Saccharomyces cerevisiae. Proceedings of the National Academy of Sciences of the United States of America 92, 8224–8228. A. Ransick, S. Ernst, R.J. Britten and E.H. Davidson (1993) Whole mount in situ hybridization shows Endo 16 to be a marker for the vegetal plate territory in sea urchin embryos. Mechanisms of Development 42, 117–124. J.C. Reese, L. Apone, S.S. Walker, L.A. Griffin and M.R. Green (1994) Yeast TAFIIS in a multisubunit complex required for activated transcription. Nature 371, 523–527. R.G. Roeder (1998) Role of general and gene-specific cofactors in the regulation of eukaryotic transcription. In: Mechanisms of Transcription, Cold Spring Harbor Symposia on Quantitative Biology, p. 679. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. S. Ryu, S. Zhou, A.G. Ladurner and R. Tjian (1999) The transcriptional cofactor complex CRSP is required for activity of the enhancer-binding protein Sp1. Nature 397, 446–450. C.O. Samuelsen, V. Baraznenok, O. Khorosjutina, et al. (2003) TRAP230/ARC240 and TRAP240/ARC250 mediator subunits are functionally conserved through evolution. Proceedings of the National Academy of Sciences of the United States of America 100, 6422–6427.
26
REGULATION OF TRANSCRIPTION IN PLANTS
S.L. Sanders, K.A. Garbett and P.A. Weil (2002) Molecular characterization of Saccharomyces cerevisiae TFIID. Molecular and Cellular Biology 22, 6000–6013. F. Sauer, D.A. Wassarman, G.M. Rubin and R. Tjian (1996) TAF(II)s mediate activation of transcription in the Drosophila embryo. Cell 87, 1271–1284. I.A. Shahmuradov, A.J. Gammerman, J.M. Hancock, P.M. Bramley and V.V. Solovyev (2003) PlantProm: a database of plant promoter sequences. Nucleic Acids Research 31, 114–117. I.A. Shahmuradov, V.V. Solovyev and A.J. Gammerman (2005) Plant promoter prediction with confidence estimation. Nucleic Acids Research 33, 1069–1076. S. Shannon and D.R. Meeks-Wagner (1991) A mutation in the Arabidopsis TFL1 gene affects inflorescence meristem development. Plant Cell 3, 877–892. W.C. Shen, S.R. Bhaumik, H.C. Causton, et al. (2003) Systematic analysis of essential yeast TAFs in genome-wide transcription and preinitiation complex assembly. The EMBO Journal 22, 3395– 3402. E.Y. Shim, A.K. Walker and T.K. Blackwell (2002) Broad requirement for the mediator subunit RGR1 for transcription in the Caenorhabditis elegans embryo. The Journal of Biological Chemistry 277, 30413–30416. S. Sinha, S.N. Maity, J. Lu and B. de Crombrugghe (1995) Recombinant rat CBF-C, the third subunit of CBF/NFY, allows formation of a protein-DNA complex with CBF-A and CBF-B and with yeast HAP2 and HAP3. Proceedings of the National Academy of Sciences of the United States of America 92, 1624–1628. S.T. Smale and J.T. Kadonaga (2003) The RNA polymerase II core promoter. Annual Review of Biochemistry 72, 449–479. M. Soltysik-Espanola, D.C. Klinzing, K. Pfarr, R.D. Burke and S.G. Ernst (1994) Endo16, a large multidomain protein found on the surface and ECM of endodermal cells during sea urchin gastrulation, binds calcium. Developmental Biology 165, 73–85. Y. Takagi and R.D. Kornberg (2006) Mediator as a general transcription factor. The Journal of Biological Chemistry 268, 80–89. Y. Tamada, K. Nakamori, K. Matsuda, T. Furumoto and K. Izui (2003) Characterization of TAF10, a general transcription factor, in plants. In: 7th International Congress of Plant Molecular Biology, Barcelona, pp. S05–S22. A. Testa, G. Donati, P. Yan, et al. (2005) Chromatin immunoprecipitation (ChIP) on chip experiments uncover a widespread distribution of NF-Y binding CCAAT sites outside of core promoters. The Journal of Biological Chemistry 280, 13606–13615. C.M. Thompson, A.J. Koleske, D.M. Chao and R.A. Young (1993) A multisubunit complex associated with the RNA polymerase II CTD and TATA-binding protein in yeast. Cell 73, 1361–1375. L. Tora (2002) A unified nomenclature for TATA box binding protein (TBP)-associated factors (TAFs) involved in RNA polymerase II transcription. Genes and Development 16, 673–675. F.T. Tsai and P.B. Sigler (2000) Structural basis of preinitiation complex assembly on human pol II promoters. The EMBO Journal 19, 25–36. J. van de Peppel, N. Kettelarij, H. van Bakel, et al. (2005) Mediator expression profiling epistasis reveals a signal transduction pathway with antagonistic submodules and highly specific downstream targets. Molecular Cell 19, 511–522. C.P. Verrijzer and R. Tjian (1996) TAFs mediate transcriptional activation and promoter selectivity. Trends in Biochemical Sciences 21, 338–342. E.H. Wang, S. Zou and R. Tjian (1997) TAFII250-dependent transcription of cyclin A is directed by ATF activator proteins. Genes and Development 11, 2658–2669. K.B. Washburn, E.A. Davis and S. Ackerman (1997) Coactivators and TAFs of transcription activation in wheat. Plant Molecular Biology 35, 1037–1043. L. Weis and D. Reinberg (1997) Accurate positioning of RNA polymerase II on a natural TATAless promoter is independent of TATA-binding-protein-associated factors and initiator-binding proteins. Molecular and Cellular Biology 17, 2973–2984. F. Werner and R.O. Weinzierl (2002) A recombinant RNA polymerase II-like enzyme capable of promoter-specific transcription. Molecular Cell 10, 635–646.
GENERAL TRANSCRIPTION FACTORS AND THE CORE PROMOTER
27
F. Yang, R. DeBeaumont, S. Zhou and A.M. Naar (2004) The activator-recruited cofactor/Mediator coactivator subunit ARC92 is a functionally important target of the VP16 transcriptional activator. PNAS 101, 2339–2344. A. Yoda, H. Kouike, H. Okano and H. Sawa (2005) Components of the transcriptional Mediator complex are required for asymmetric cell division in C. elegans. Development 132, 1885–1893. C.H. Yuh, H. Bolouri and E.H. Davidson (1998) Genomic cis-regulatory logic: experimental and computational analysis of a sea urchin gene. Science 279, 1896–1902. B. Zenzie-Gregory, A. Khachi, I.P. Garraway and S.T. Smale (1993) Mechanism of initiator-mediated transcription: evidence for a functional interaction between the TATA-binding protein and DNA in the absence of a specific recognition sequence. Molecular and Cellular Biology 13, 3841–3849. F. Zhang, L. Sumibcay, A.G. Hinnebusch and M.J. Swanson (2004) A triad of subunits from the Gal11/Tail domain of Srb mediator is an in vivo target of transcriptional activator Gcn4p. Molecular and Cellular Biology 24, 6871–6886. J. Zhou, J. Zwicker, P. Szymanski, M. Levine and R. Tjian (1998) TAFII mutations disrupt dorsal activation in the Drosophila embryo. Proceedings of the National Academy of Sciences of the United States of America 95, 13483–13488. T. Zor, R.N. De Guzman, H.J. Dyson and P.E. Wright (2004) Solution structure of the KIX domain of cbp bound to the transactivation domain of c-Myb. Journal of Molecular Biology 337, 521–534.
2 Transcription factors of Arabidopsis and rice: a genomic perspective Jos´e Luis Riechmann
2.1
Introduction
The regulation of many physiological and developmental processes in plants (and in multicellular organisms in general) is largely built upon the cellular capacity for differential gene expression. That capacity is hardwired and encoded in the genome (Davidson, 2001), in the form of cis-regulatory sequences that determine where and when genes are expressed, and exerted by transcription factors and transcriptional coregulators that act on those sequences. Transcription factors are usually defined (and are considered in this chapter) as proteins that show sequence-specific DNA binding and that are capable of activating or/and repressing transcription. They are modular proteins, with distinct and separable domains for DNA binding and activation/repression, and most of them can be grouped into families according to the amino acid sequence of their DNA-binding domain (Luscombe et al., 2000). In addition, they can bear motifs or domains for protein–protein interactions with other transcription factors, from the same or different families. Transcription factors are the most numerous of all the different classes of proteins involved in transcription. They are building blocks for the mechanisms for selectivity of gene activation, are the principal protein components of the combinatorial logic of transcription, and are often expressed in a tissue-specific, cell-type-specific, temporal-specific, or stimulus-dependent specific manner. In this chapter, transcription factors are considered in the context of the complete plant genome sequences that are available, i.e., in Arabidopsis and rice. A description of the complement of transcription factors in Arabidopsis and rice sheds light on issues such as gene redundancy, and the ensuing problems for the characterization of the transcription factor phenome, or the degree of functional conservation across species. Recent extensive reviews are available for aspects of transcription factor biology (such as their molecular mechanisms of action) that are not covered here (for example, Riechmann, 2002; Latchman, 2003; Wray et al., 2003). In addition, the genomic technologies developed in the last decade have enabled the analysis of transcription factor activity from the perspective of gene regulatory networks, which is beyond the scope of this chapter but has also been reviewed recently (Wellmer and Riechmann, 2005).
TRANSCRIPTION FACTORS OF ARABIDOPSIS AND RICE
2.2
29
Arabidopsis and rice genomes: the angiosperm complement of transcription factors
The availability of the genome sequences of Arabidopsis thaliana (Arabidopsis Genome Initiative, 2000) and rice (Oryza sativa) (Goff et al., 2002; Yu et al., 2002, 2005; International Rice Genome Sequencing Project, 2005) allows a global, or genomic, analysis of transcriptional regulation in plants. For such analyses, a comprehensive list of parts (in this case, of transcription factors and of the genes that they may regulate) can be generated through various bioinformatic analyses.
2.2.1
General considerations for genome-wide analyses
The identification of transcription factor coding genes in a genome sequence is carried out through sequence similarity searches and pattern-finding methods. Thus, the results obtained reflect, and are limited by, the previous knowledge on this class of proteins. All of the major families of transcription factors in plants have probably been already identified. However, in the case of Arabidopsis, approximately 40–45% of its 26 207 protein-coding genes (annotation release 5, January 2004) do not have meaningful Gene Ontology (GO) categories assigned to them (i.e., they remain classified as ‘unknown’) (Haas et al., 2005). Similarly, biological functions have not been assigned for ∼50% of the genes in rice (International Rice Genome Sequencing Project, 2005; Yu et al., 2005). It is likely that some of these uncharacterized proteins are transcriptional regulators, and, in fact, novel transcription factors are still being discovered, which are often found to belong to (small) gene families (for example, Desveaux et al., 2000; Bouche et al., 2002; Sangwan and O’Brian, 2002; Yang and Poovaiah, 2002; Zourelidou et al., 2002; C¸ akir et al., 2003; Carrasco et al., 2003; Curaba et al., 2003; Desveaux et al., 2004; Meister et al., 2004; Mitsuda et al., 2004; Mouchel et al., 2004; Carles et al., 2005; Carrasco et al., 2005; Yin et al., 2005). Conversely, new experimental evidence may question (or eventually disprove) whether proteins once considered transcription factors actually play such a role (for example, the TUBBY-like proteins or TULP; Carroll et al., 2004). Therefore, the exact number and identity of transcription factor genes present in Arabidopsis and rice (as well as, for the same reasons, in any other of the sequenced eukaryotic genomes) will be uncertain for some time. Additional issues in genome-wide surveys include whether all the proteins identified by sequence similarity searches do indeed belong to the functional groups into which they are being cataloged, and the degree of completeness of the genome sequence under study. For example, only a subset of the proteins that bear a C2H2 zinc-finger motif are actually transcription factors, whereas others are involved in other processes, such as splicing (Riechmann et al., 2000; Englbrecht et al., 2004). In addition, gene sequences might diverge more than expected, which might hamper the identification of homologous genes through sequence similarity searches (for
30
REGULATION OF TRANSCRIPTION IN PLANTS
a discussion of these issues, including examples in Arabidopsis, see Riechmann, 2002). For these and other reasons, different genome-wide searches for genes of the same family may result in slightly different gene sets (for an illustrative example about the bHLH proteins in Arabidopsis, see Bailey et al., 2003) (see also Table 2.1).
2.2.2
Arabidopsis transcription factors
The analysis of the Arabidopsis genome sequence indicates that it codes for at least 1589 transcription factors, which account for ∼6% of its estimated ∼26 200 protein-coding genes (Table 2.1) (Arabidopsis Genome Initiative, 2000; Riechmann et al., 2000). These 1589 transcription factor genes are classified into more than 40 different gene families and, in addition, there are a few single-copy or ‘orphan’ genes, such as LEAFY (LFY) and SPOROCYTELESS (SPL) (Table 2.1). Many of those gene families are large, with over, or close to, 100 members, and the largest of them, bHLH, AP2/ERF, and MYB-(R1)R2R3 factors, each represent ∼9% of the transcription factor complement, or 0.5% of the total number of genes in the genome (Table 2.1). A recently developed online database of Arabidopsis transcription factors (DATF) aims to present a comprehensive view of them by integrating multiple sources of information, in particular about gene families and gene sets (http://datf.cbi.pku.edu.cn/; Guo et al., 2005). Data in DATF are very similar to the results summarized in Table 2.1, with the differences being due to the use in Table 2.1 of slightly more stringent criteria for considering a gene or members of a gene family as bona fide transcription factors. Salient features of the Arabidopsis complement of transcription factors (that also apply to other plant genomes) were highlighted in a comparative study that included the genomes of Saccharomyces cerevisiae, Caenorhabditis elegans, and Drosophila melanogaster, and that illustrated the large degree of diversity in transcriptional regulators that is present among the different eukaryotic kingdoms (Riechmann et al., 2000; Riechmann, 2002). Arabidopsis transcription factors that belong to families that are common to all eukaryotes do not share significant similarity with those from the other kingdoms, except in the conserved DNA-binding domains that define the respective families (Riechmann et al., 2000). In addition, other sources of diversity include the presence of lineage-specific families, uneven amplification of common families among eukaryotic lineages (and between plant lineages, i.e., monocots and dicots, see below), and protein domain shuffling. In general, transcription factor families have collectively higher expansion rates in plants than in animals: the majority of the families that are shared between plants and animals are larger in plants (Riechmann et al., 2000; Shiu et al., 2005). For example, two of the largest Arabidopsis families, MYB and MADS, are very small in fungi and animals. However, in animals and fungi there tends to be one or two specific transcription factor families that have been enormously and disproportionately amplified, more so than any of the families in plants: for example, the nuclear hormone receptors in C. elegans (∼38% of its transcription factors), the C2H2 zincfinger proteins in Drosophila (∼46%) and humans, or the C6 and C2H2 families in yeast (∼25% each) (Riechmann et al., 2000; Riechmann, 2002). In plants, there
31
TRANSCRIPTION FACTORS OF ARABIDOPSIS AND RICE Table 2.1
Arabidopsis transcription factorsa,b
Gene family
Number of genesc
Genetically characterized factorsd
e
AP2/ERF AP2 subfamily f ERF-EREBP subfamily RAV subfamily BHLH bHLH
19 122
6 123
(-b)HLH MYB superfamily MYB-(R1)R2R3
39 135
MYB-related C2H2 (Zn)
52 113
NAC
109
HB
91
MADSg Type II (MIKC)h
46
Type Ii
bZIP
WRKY (Zn)
?-64
77
72
ANT, AP2, PLT1, PLT2, WRI1 ABI4, AtERF7, CBF1, CBF2, CBF3, CBF4, ERF1, DRN/ESR1, SHN2, SHN3, WIN1/SHN1 — ALC, AMS, BEE1, BEE2, BEE3, BIM1, BIM2, BIM3, EGL3, FIT1, FAMA, GL3, HFR1, ICE1, NAI1, PIF1/PIL5, PIF3, PIF4, MYC2/JIN1, SPT, TT8 IND AS1, ATR1, BOS1, DUO1, FLP, GL1, HOS10, LAF1, MYB2, MYB4, MYB12, MYB23, MYB26, MYB30, MYB32, MYB33, MYB60, MYB61, MYB65, MYB88, MYB98, MYB103, PAP1, PAP2, TT2, WER CCA1, CPC, EPR1, ETC1, LHY, TRY EMF2, FIS2, JAG, KNU, RBE, RHL41, SE, SUP, TT1, VRN2 ATAF2, CUC1, CUC2, CUC3, NAC1, NAP, NST1, NST2, RD26, VND6, VND7 ANL2, ATHB-2, ATHB-8, ATHB-13, ATHB-15, ATML1, BEL1, CNA, GL2, HOS9/PFS2, KNAT1/BP, OCP3, PDF2, PHB, PHV, PNF, PNY/BLR/RPL, PRS, REV, STIP, STM, WUS AG, AGL15, AGL24, ANR1, AP1, AP3, CAL, FLC, FUL, MAF1, MAF2, PI, SEP1, SEP2, SEP3, SEP4, SHP1, SHP2, SOC1, STK, SVP, TT16
—
ABI5, ABF2, ABF3, ABF4, EEL, FD, HY5, HYH, PAN, TGA2, TGA5, TGA6 TTG2, RRS1, WRKY22, WRKY29, WRKY70
Gene family references Riechmann and Meyerowitz, 1998; Sakuma et al., 2002; Gutterson and Reuber, 2004; Yamasaki et al., 2004b; Nole-Wilson et al., 2005 Bailey et al., 2003; Buck and Atchley, 2003; Heim et al., 2003; Toledo-Ortiz et al., 2003
Stracke et al., 2001; Jiang et al., 2004
Takatsuji, 1999; Englbrecht et al., 2004 Ooka et al., 2003; Ernst et al., 2004; Olsen et al., 2005 Chan et al., 1998; Schrick et al., 2004
Riechmann and Meyerowitz, 1997; Ng and Yanofsky, 2001; Becker and Theissen, 2003; De Bodt et al., 2003a,b; Kofuji et al., 2003; Martinez-Castilla and Alvarez-Buylla, 2003; Parenicova et al., 2003; Lehti-Shiu et al., 2005
Sib´eril et al., 2001; Jakoby et al., 2002; Vincentz et al., 2003; Deppmann et al., 2004 Eulgem et al., 2000; Ulker and Somssich, 2004 (Continued )
32
REGULATION OF TRANSCRIPTION IN PLANTS
Table 2.1 Arabidopsis transcription factorsa,b (Continued )
Gene family
Number of genesc
GARP G2-like
41
Genetically characterized factorsd
13
APL, GLK1, GLK2, KAN1, KAN2, LUX, PHR1 ARR1, ARR2, ARR10, ARR11, ARR12
C2C2 (Zn) Dof
36
CDF1, COG1, DAG1, DAG2, OBP3
CO-like
32
CO, COL1, COL9
GATA
30
HAN
ARR-type B/pseudoARRtype B
YABBY CCAAT-binding factor HAP2-type HAP3-type HAP5-type Dr1 GRAS
10 11 13 2 32
— LEC1, L1L — — GAI, LAS, PAT1, RGA, RGL1, RGL2, SCR, SHR
Trihelix TCP
29 24
PTL PTF1
ARF
23
HSF
21
ARF2/HSS, ARF3/ETT, ARF4, ARF5/MP, ARF6, ARF7/NPH4, ARF8, ARF10, ARF16, ARF19 HSF1, HSF3
STK-like
17
—
C3H-type 1 (Zn) SBP (Zn)
17 17
PEI1 SPL3, SPL8
GRF/ENBP (C3H-type2; Zn)
16
AtGRF1–3, AtGRF5
Nin-like ABI3/VP1 ZF-HB (Zn) SRS (Zn) E2F/DP CPP (Zn) Alfin-like GBP-like/BPC
14 14 14 10 8 8 7 7
— ABI3, FUS3, LEC2 — SHI, STY1, STY2 E2Fa, E2Fb, DPa, DEL1 TSO1 — —
EIL
6
6
CRC, FIL, INO
EIN3, EIL1
Gene family references Riechmann et al., 2000 Rossini et al., 2001 Hosoda et al., 2002; Hwang et al., 2002; Lohrmann and Harter, 2002; Mason et al., 2004 Yanagisawa, 2002; Lijavetzky et al., 2003 Lagercrantz and Axelsson, 2000; Griffiths et al., 2003 Reyes et al., 2004; Shikata et al., 2004 Bowman, 2000 Lotan et al., 1998; Gusmaroli et al., 2001, 2002
Pysh et al., 1999; Richards et al., 2000; Bolle, 2004; Tian et al., 2004 Zhou, 1999; Ayadi et al., 2004 Cubas et al., 1999; Cubas, 2002; Kosugi and Ohashi, 2002 Guilfoyle and Hagen, 2001; Remington et al., 2004; Okushima et al., 2005 Nover et al., 2001; Kotak et al., 2004 Zourelidou et al., 2002; Curaba et al., 2003 Li and Thomas, 1998 Cardon et al., 1999; Yamasaki et al., 2004a Christiansen et al., 1996; van der Knaap et al., 2000; Kim et al., 2003 Schauser et al., 2005 Stone et al., 2001 Windh¨ovel et al., 2001 Fridborg et al., 2001 de Jager et al., 2001 Cvitanich et al., 2000 Bastola et al., 1998 Sangwan and O’Brian, 2002; Meister et al., 2004 Chao et al., 1997
TRANSCRIPTION FACTORS OF ARABIDOPSIS AND RICE Table 2.1
33
(Continued )
Gene family
Number of genesc
Genetically characterized factorsd
Gene family references
NtLIM1-like (Zn) BES CAMTA
6 6 6
— BES1, BZR1 —
BRX-like DBP-like S1Fa-like
5 5 3
BRX — —
Whirlye HRT-like (Zn) VOZ (Zn) ULT LFY SPL/NZZ SAP
3 2 2 2 1 1 1
AtWhy1 — — ULT1 LFY SPL/NZZ SAP
Kawaoka et al., 2000 Yin et al., 2005 Bouche et al., 2002; Yang and Poovaiah, 2002 Mouchel et al., 2004 Carrasco et al., 2005 Zhou et al., 1995 Desveaux et al., 2002, 2005 Krause et al., 2005 Raventos et al., 1998 Mitsuda et al., 2004 Carles et al., 2005 Weigel et al., 1992 Schiefthaler et al., 1999 Byzova et al., 1999
a
Transcription factors are usually defined as proteins that show sequence-specific DNA binding and the capability to activate or repress transcription. However, the distinction between ‘transcription factors’ and other transcriptional regulators is not without problems (see, Riechmann et al., 2000; Riechmann, 2002). Most of the genes counted here fulfill those characteristics, but some do not (for example, some proteins of the bHLH family, indicated as the (-b)HLH subfamily, may not actually bind DNA); on the other hand, proteins that work together with some transcription factors, such as the IAA proteins that work with ARF proteins, have not been included in the table. b (Zn) indicates a zinc-coordinating motif. c The data shown in the table are an updated (October 2005) version of those in Riechmann (2002). Additional genome-wide analyses of several Arabidopsis transcription factor families have been published more recently. The results of such analyses have also been used to update the table, when appropriate, and if so, they are cited in the ‘Gene family references’ column. d Includes genes characterized by loss of function and/or, if informative results were obtained, by gain of function (as of October 2005). The correspondence between acronyms and gene names can be found at TAIR (http://www.arabidopsis.org). e The AP2/ERF family was initially referred to as AP2/EREBP, for AP2/ethylene-r esponsive element binding protein; the Whirly family was initially referred to as PBF-2-like, after the founding member, potato PBF-2. f The AP2 subfamily contains 19 members. Fourteen of them are proteins with two AP2 domains each (R1 and R2). The other five are proteins with only one clearly recognizable AP2 domain, but that domain is more related in sequence to either R1 or R2 than to the single AP2 domain of any of the ERF subfamily proteins. g The MADS-box family has been the subject of several genome-wide gene identification and phylogenetic analyses. Whereas these studies (and the analysis reported in Riechmann et al., 2000) agree on the size and composition of the MIKC subfamily of MADS-box genes (Type II MADS-box genes, according to Alvarez-Buylla et al., 2000), they differ in the number of recognized Type I genes, as well as on their classification in different phylogenetic subgroups. Only the broad subdivision of the MADS family into Type I and Type II genes is considered here (following the classification of Martinez-Castilla and Alvarez-Buylla, 2003). h Type II (MIKC) includes MADS-box genes of the classic MIKC-type (also referred to as MIKCc ) and also genes of the MIKC∗ -type (see Becker and Theissen, 2003; Kofuji et al., 2003). i The identification and classification of Type I genes has differed among the various studies, in part because of poor conservation of typical key MADS-domain residues. For example, some genes exhibit a highly divergent N-terminal region of the MADS box (and also lack other conserved regions); those genes were considered MADS like by De Bodt et al., but were included in the MADS family by Parenicova et al. and Kofuji et al. Furthermore, it is unclear (but possible) whether some of the Type I genes might be pseudogenes, nonfunctional genes, the result of (retro)transposition events, or perhaps even represent a new class of transposable elements (De Bodt et al., 2003b; Parenicova et al., 2003). For these reasons, since it is still unclear how many of Type I MADSbox genes are in fact functional transcription factor genes, the number of members of the subfamily is indicated as ‘?-64’.
34
REGULATION OF TRANSCRIPTION IN PLANTS
is no family of transcription factors that represents by itself one third or more of the total number of transcription factors in the organism. Shuffling of some of the DNA-binding domains that are present in all eukaryotes has generated novel transcriptional regulators with plant-specific combinations of modules, as, for example, in the homeodomain, MADS, and ARID protein families (Riechmann et al., 2000; Riechmann, 2002). Furthermore, in Arabidopsis, approximately 45% of its transcription factors belong to families that are specific to the plant kingdom (when compared to animals and fungi, i.e., among the crown-group eukaryotes), whereas ∼53% correspond to families that are present in plants, fungi, and animals (the remaining minority belonging to families that are present in two of the three kingdoms) (Riechmann et al., 2000; Riechmann, 2002) (see Table 2.1). However, some of the ‘plant-specific’ families of transcription factors have also been recently identified in noncrown-group eukaryotes. In addition, new sequence and structural data are showing how the DNA-binding domains of some of the plant-specific transcription factor families are related to proteins present in other organisms, in which those domains may not form part of transcription factors but rather of other types of (DNA-binding) proteins. Thus, some proteins might have evolved as transcription factors after the plant lineage split from animals and fungi, but the structural frameworks for the DNA-binding domains were already present in ancestral organisms (Yamasaki et al., 2005a). A large plant-specific family of transcription factors is the AP2/ERF group, characterized by the presence of the AP2 DNA-binding domain. The AP2 domain itself, however, is not plant specific, since it has been found in endonucleases of different organisms (cyanobacteria, ciliates, and viruses) and in putative transcription factors from apicomplexans (such as the malarial parasite Plasmodium) (Magnani et al., 2004; Wuitschick et al., 2004; Balaji et al., 2005). Similarly, the B3 DNA-binding domain, which is present in several plantspecific families of transcription factors (ARF, ABI3/VP1, and RAV) (Riechmann, 2002), is structurally related to the EcoRII DNA-binding domain (Yamasaki et al., 2004b). The WRKY DNA-binding domain, also initially considered plant specific, does not show strong structural similarity to other known domains (Yamasaki et al., 2005a), but it has been found in a primitive eukaryote (the unicellular protist Giardia lamblia) and in the slime mold Dictyostelium discoideum (as well as in the green alga Chlamydomonas reinhardtii) (Ulker and Somssich, 2004; Wu et al., 2005; Zhang and Wang, 2005); the RWP-RK domain, the putative DNA-binding region characteristic of the plant NIN-like proteins, is also found in D. discoideum and in C. reinhardtii (Schauser et al., 2005). In summary, some of the domains that define families of transcription factors initially considered specific to plants predate the plant–fungus/animal divide. On the other hand, the structures of the DNA-binding domains of the SBP, NAC, and EIL proteins (all apparently plant specific) represent novel folds (Ernst et al., 2004; Yamasaki et al., 2004a, 2005b). Because of their regulatory nature, it is often thought that transcription factors are generally expressed at low levels relative to other genes. However, various types of genomic data contradict such assumption, and indicate that, at least in plants, transcription factor genes are not substantially different from the rest of the genes in the genome in terms of overall expression and when considered as a whole. Analyses of Arabidopsis expressed sequence tags (ESTs) (Riechmann, 2002), of the RIKEN
TRANSCRIPTION FACTORS OF ARABIDOPSIS AND RICE
35
full-length (RAFL) cDNA clone collection (Seki et al., 2002), or of various macroand microarray gene expression datasets (Chen et al., 2002; Jiao et al., 2003; de Folter et al., 2004; Wellmer et al., 2004; Schmid et al., 2005), all indicate that the average expression of transcription factors is similar to the average expression for all genes. However, for transcription factors that are expressed in discrete, often small, spatial and temporal domains, detection of their expression may require real-time polymerase chain reaction (PCR) experiments (Czechowski et al., 2004). Quantitative real-time PCR is generally more sensitive than microarrays, and can thus alleviate the problem caused by isolating heterogenous tissue samples in which the abundance of the transcript(s) of interest is reduced (it is often difficult, if at all possible, to specifically isolate the cells/tissues in which a given transcription factor may be expressed – rather, those cells/tissues are usually isolated together with cells/tissues in which it is not; Schnable et al., 2004; Wellmer and Riechmann, 2005).
2.2.3
Rice transcription factors: a comparison to Arabidopsis
The rice genome has been sequenced using the whole-genome shotgun method (in draft and finished form, and for two different rice subspecies, indica and japonica) (Goff et al., 2002; Yu et al., 2002, 2005), and also with a map-based strategy (subsp. japonica) (International Rice Genome Sequencing Project, 2005). Initial global analyses of the sequence illustrate the general similarities and differences with Arabidopsis, in transcription factor gene content (Xiong et al., 2005). In addition, many families and subfamilies of transcription factors have been the subject of detailed analyses in rice; these include the WRKY (Wu et al., 2005; Xie et al., 2005; Zhang and Wang, 2005), NAC (Ooka et al., 2003), MADS (MIKC-type, Nam et al., 2004), MYB (Jia et al., 2004; Jiang et al., 2004), bHLH (Buck and Atchley, 2003), AP2/ERF (ERF subfamily, Gutterson and Reuber, 2004), Dof (Lijavetzky et al., 2003), GRAS (Tian et al., 2004), HB (START-HB subfamily, Schrick et al., 2004), GATA (Reyes et al., 2004), GRF (Choi et al., 2004), CO-like (Griffiths et al., 2003), and NIN-like families (Schauser et al., 2005). Overall, the transcription factor gene complements of Arabidopsis and rice are similar in terms of the total number of genes and the relative sizes of the different gene families, as expected, since the evolutionary divergence between the monocot and dicot lineages is a relatively recent event (∼150–200 million years ago). A recent study identified ∼1600 transcription factor coding genes in the rice genome sequence (Xiong et al., 2005). Applying the criteria used in Table 2.1 (which yielded 1589 Arabidopsis transcription factors), and also accounting for the differences in the quality of sequence annotation, it is possible to estimate the rice genome codes for ∼1800 transcription factors (representing 4.8–4.4% of the estimated 37 500– 40 000 nontransposable-element-related protein-coding sequences in the genome; Bennetzen et al., 2004; International Rice Genome Sequencing Project, 2005; Yu et al., 2005). The relative sizes of the families in the two organisms are generally maintained, with minor variations; for example, NAC and WRKY families are proportionally larger in rice, whereas the MADS family appears to be larger in Arabidopsis (Figure 2.1). However, both genomes have a complicated history of
Figure 2.1 Content and distribution of transcription factor genes in Arabidopsis and rice. The data represented in this figure are from Xiong et al. (2005). Arabidopsis and rice data were obtained in that study using the same analyses and criteria, which allows for a more consistent comparison between the two organisms than if results from two separate studies were used. Because of this, however, the Arabidopsis data in this figure do not match perfectly with those in Table 2.1, although the general conclusions that can be drawn from the Arabidopsis versus rice comparison remain valid. Differences between the study by Xiong et al. and Table 2.1 are minor differences in gene number (for example, 105 NAC family genes identified by Xiong et al., versus 109 in Table 2.1), recently identified families that were not included in the study by Xiong et al. (such as CAMTA, VOZ, BES, and BRX-like), and differences in family definition (for example, for the C2H2 proteins; only the EPF subfamily was considered by Xiong et al.). In addition, some families are not included in the figure (Trihelix, C3H-type1) due to larger discrepancies between the results of Xiong et al. and previous analyses. Single-copy genes are not included in the figure. The data for the Dof and GATA families are from Lijavetzky et al. (2003) and Reyes et al. (2004), respectively, and from Table 2.1.
TRANSCRIPTION FACTORS OF ARABIDOPSIS AND RICE
37
large- and gene-scale duplications, and it appears that many lineage-specific events have contributed to generating the respective current gene sets. Phylogenetic analyses of individual gene families with both Arabidopsis and rice sequences (or with Arabidopsis and maize; for example, Rabinowicz et al., 1999; Munster et al., 2002) show the presence of many common subgroups or clades, indicating that gene duplications and amplification of the gene families had already taken place in the most recent common ancestor (MRCA) of monocots and dicots. In fact, through such analyses it has been estimated that the common ancestor of rice and Arabidopsis had a set of at least ∼400 ancestral transcription factor genes (Xiong et al., 2005). Many cases of 1:1 orthologous relationships can be identified between Arabidopsis and rice genes (usually corresponding to well-conserved functions). However, there are also many instances of 1:n, n:1, and n:n relationships (originated by duplications that took place after the two lineages had separated and/or by lineage-specific geneloss events), as well as species-specific genes for which no putative orthologs can be identified in the other organism (Figure 2.2). Orthologous groups of the 1:n and n:1 types may correspond to functions that have diversified in one species but not in the other, whereas those of the n:n type correspond to functions that might have diversified in both, and genes that are classified as lineage specific in these analyses may correspond to species-specific or highly specialized functions. The MADS-box family provides examples of all these scenarios. Many cases of orthology can be identified between Arabidopsis MADS-box genes and those from rice or maize, and even from gymnosperms (Becker and Theissen, 2003; Nam et al., 2004). Putative orthologous MADS-box genes have regularly maintained conserved functions, even after substantial sequence divergence. On the other hand, there are a few MADS-box genes that are not members of clades shared by Arabidopsis and rice, for example, the Arabidopsis flowering time genes FLC and MAF1-5 (Ratcliffe et al., 2003), which form a monophyletic group, Arabidopsis AGL15, and rice OsMADS32 (Nam et al., 2004). Rice and Arabidopsis have 37 and 39 MADS-box genes of the MIKCc (or Type II) class, respectively (Figure 2.2; Nam et al., 2004), and at least 32 genes have been already identified in maize (Munster et al., 2002). The number of ancestral genes in the MRCA of monocots and dicots has been estimated as 12–20, and many MADS-box gene duplication and diversification events occurred earlier than that, after the separation of the moss and fern lineages from the lineage that originated the flowering plants (Munster et al., 2002; Becker and Theissen, 2003; Nam et al., 2003, 2004; Zahn et al., 2005). Diversity of MADS-box genes in Arabidopsis is thus rather ancient, and it appears that it is fairly representative for other flowering plants (Becker and Theissen, 2003). Other families of transcription factors had similar evolutionary histories, although the proportion of species- or lineage-specific genes, and the degree of amplification after a given lineage divergence step, varies by gene family (see Figure 2.2 and associated references). In addition, gene clades that are shared by monocots and dicots might evolve at different rates. For example, within the WRKY family, subgroup III is largely amplified in monocots and appears to be evolutionarily more active than in Arabidopsis (Wu et al., 2005; Zhang and Wang, 2005). Interestingly, it appears that WRKY genes are overrepresented in flowering plants in comparison with fern, moss, or pine; because WRKY genes are important regulators for biotic
38
REGULATION OF TRANSCRIPTION IN PLANTS
Figure 2.2 Summary of orthologous relationships between Arabidopsis and rice genes for selected transcription factor gene families. A hypothetical phylogenetic tree is depicted to illustrate the different types of relationships that are considered: 1 : 1, 1 : n and n : 1 (duplications in rice or Arabidopsis), n : n (duplications in both), and specific to one of the two plants. Because the complete map-based sequence of the rice genome was not available when the phylogenetic studies were performed, various sources of rice sequences were used to identify gene family members (whole-genome shotgun assemblies from the indica and/or japonica subspecies, sequences from the International Rice Genome Sequencing Project (IRGSP), and cDNA sequences and ESTs). Families and phylogenetic analyses that were used to create this figure are as follows: WRKY (Wu et al., 2005), MADS (MIKC-type proteins only) (Nam et al., 2004), GRAS (Tian et al., 2004), Dof (Lijavetzky et al., 2003), GATA (Reyes et al., 2004), and CO-like (Griffiths et al., 2003). The percentage of genes classified as ‘specific’ for each of the families and plants is indicated.
and abiotic stress responses, and flowering plants are dominant over nonflowering plants in their distribution, it has been hypothesized that WRKY genes might be important for such adaptability (Zhang and Wang, 2005).
2.2.4
Gene duplications, functional redundancy, and the transcription factor phenome
The Arabidopsis and rice genomes contain many tandem gene duplications, as well as many large duplicated chromosomal segments – the result of ancient whole
TRANSCRIPTION FACTORS OF ARABIDOPSIS AND RICE
39
genome duplication event(s) that occurred during their evolution (Simillion et al., 2002; Blanc et al., 2003; Bowers et al., 2003; Cannon et al., 2004; Yu et al., 2005). Many of these duplications have been followed by the loss of duplicated genes, by rearrangements, or by divergent evolution – but still up to 40–50% of the Arabidopsis genes comprise pairs of highly related sequences (Arabidopsis Genome Initiative, 2000; Blanc et al., 2000). The process of gene loss following duplication appears nonrandom, since genes from some functional categories (including, but not limited to, transcription factors) have been preferentially retained in Arabidopsis (Blanc and Wolfe, 2004). Closely related genes account for ∼45% of the total number of members in the major families of transcription factors (Riechmann et al., 2000). Importantly, pairs or groups of closely related transcription factor genes in Arabidopsis are most often located in different chromosomes (∼65% on average), or in the same chromosome but at very large distances (∼22%; the remaining 13% corresponding to tandem repeats) (Riechmann et al., 2000). This distribution makes it feasible to generate the respective double or triple mutants, which are often required for the analysis of highly related genes that may have overlapping or partially redundant functions. Approximately 230 Arabidopsis transcription factors have been genetically characterized (∼15% of the transcription factor gene complement; Table 2.1). Whereas in the early years of Arabidopsis research, characterization was achieved through the traditional genetic approach (whereby genes are first defined by a mutant phenotype and then isolated), reverse genetics methods are currently widely used (Zhang, 2003; Østergaard and Yanofsky, 2004) and have facilitated targeted analyses of closely related genes (see below). Still, for the majority of these transcriptional regulators, functional characterization is usually limited to the description of phenotypic differences between mutant and wild-type plants, and determination of their expression pattern, and there is limited knowledge on the genes that they regulate and on the mechanisms that they use to achieve that regulation. However, gene expression microarrays and other genomic approaches are starting to shed light on the dynamic relationships between the transcriptional regulators, the transcriptome, and the genome (Wellmer and Riechmann, 2005). Of particular relevance to transcription factors, DNA microarrays have been used to compare the transcriptome of transcription factor mutant versus wild-type plants (for example, Mele et al., 2003; Schmid et al., 2003; Wellmer et al., 2004) and to follow gene expression changes after the specific induction or activation of the transcription factor under study, as a way to identify genes potentially regulated by such transcription factor or that form part of the network that it controls or participates in. In these experiments, synthetic systems in which the transcription factor is fused to the rat glucocorticoid receptor (GR), and thus exogenously activated with dexamethasone, are generally used. Arabidopsis regulators that have already been studied using GR fusions include LFY, AGAMOUS (AG), and NPR1 (Ito et al., 2004; Wagner et al., 2004; William et al., 2004; Gomez-Mena et al., 2005; Wang et al., 2005b). In addition, the global gene expression profiling of physiological or developmental processes (Chen et al., 2002; Nemhauser et al., 2004; Hannah et al., 2005) can lead to the identification of transcription factors that may be involved in their regulation, but that had not been uncovered by genetic analyses (for example, Bergmann et al., 2004).
40
REGULATION OF TRANSCRIPTION IN PLANTS
Probably the two main difficulties for generating a comprehensive phenome of a plant’s transcriptional regulator set are the finite number of assays in which the mutants can usually be screened and the existence of functional redundancy among different genes (Riechmann and Ratcliffe, 2000; Riechmann, 2002). Many examples of functional redundancy or overlap among highly related Arabidopsis transcription factors have been uncovered in recent years, already encompassing ∼36% of the total number of transcription factors functionally characterized and including the majority of gene families (Table 2.2). In many of these cases, single mutants exhibit either subtle or no phenotypic alterations, whereas the double (or triple) mutants show much more drastic phenotypic changes. Furthermore, in addition to redundancy resulting from the incomplete functional divergence between paralogs, it can also arise from functional convergence of more distantly related gene family members. Although the extent of this type of redundancy in Arabidopsis is unknown, there is evidence that it exists. For example, AINTEGUMENTA acts redundantly with APETALA2 to repress AG expression in cells of the second whorl of developing Arabidopsis flowers (the two proteins belong to the AP2 subfamily, but to different clades within it) (Krizek et al., 2000).
2.3
Plant promoters
Transcription factors regulate gene expression in a combinatorial manner, through different cis-elements that coordinately regulate the expression of the corresponding gene. Cis-acting elements are usually organized in a modular fashion: the regulatory region of a gene can be partitioned into discrete sub-elements, each containing one or several binding sites for transcription factors and performing a certain regulatory function. These modules can interact synergistically, i.e., combinations of modules can direct gene expression in a manner not observed with the modules in isolation (for reviews see Riechmann, 2002; Wray et al., 2003). Regulatory sequences of plant genes usually span relatively short DNA intervals (as compared to animals, metazoans in particular), often less than 1 or 2 kb. This is illustrated by the compact organization of the Arabidopsis and rice genomes, in which gene density is high. In Arabidopsis, there is one gene per 4.5 kb of DNA: the gene length (from annotated transcriptional start to stop) is approximately 2.2 kb, and ∼2.3 kb corresponds to intergenic regions. Considering the whole genome, transposons account for ∼20% of the intergenic DNA, resulting in an average of 1.9 kb of DNA for the 5 and 3 regions of a particular gene (Arabidopsis Genome Initiative, 2000). In rice, gene density is lower (one nontransposable-element-related protein-coding gene per ∼10 kb), but the average gene length is similar to that in Arabidopsis (International Rice Genome Sequencing Project, 2005; Yu et al., 2005). Other plants have much larger genomes than those of Arabidopsis or rice (∼2500 Mbp in the case of maize, for example), but with a similar organization of genes. In maize, genic regions occupy less than 10–20% of the genome, usually in the form of compact gene-rich islands, with much of the remaining genomic DNA corresponding to repetitive sequences made up of retrotransposons and other elements (Palmer et al., 2003; Whitelaw et al., 2003; Messing et al., 2004).
41
TRANSCRIPTION FACTORS OF ARABIDOPSIS AND RICE Table 2.2
Functional redundancy among related Arabidopsis transcription factors Redundant or overlapping function
Gene family
Genes
AP2/ERF
PLT1, PLT2
AP2/ERF AP2/ERF bHLH bHLH bHLH
CBF1, CBF2, CBF3 WIN1/SHN1, SHN2, SHN3 BEE1, BEE2, BEE3 BIM1, BIM2, BIM3 GL3, EGL3, TT8
MYB-R2R3 MYB-R2R3
FLP, MYB88 GL1, MYB23
MYB-R2R3 MYB-related
MYB33, MYB65 ETC1, TRY, CPC
MYB-related
LHY, CCA1
Circadian rhythmicity
NAC
CUC1, CUC2, CUC3
NAC
NST1, NST2
HB
PNF, PNY
HB
PDF2, ATML1
HB
PHB, PHV, REV, CNA
HB MADS
KNAT1/BP, STM AP1, CAL, FUL
Shoot apical meristem formation; organ separation Secondary wall thickening and anther dehiscence Response to floral inductive signals Shoot epidermal cell differentiation Radial patterning (PHB, PHV, REV); meristem regulation (PHB, PHV, CNA) Stem cell function Floral meristem identity
MADS
AG, SHP1, SHP2, STK
MADS
SEP1, SEP2, SEP3, SEP4
BZIP WRKY GARP
TGA2, TGA5, TGA6 WRKY22, WRKY29 GLK1, GLK2
Root meristem: stem cell niche patterning and maintenance Freezing tolerance Regulation of wax biosynthesis Brassinosteroid signaling Brassinosteroid signaling Trichome and root hair development, anthocyanin production (GL3, EGL3); seed coat mucilage regulation (TT8, EGL3) Stomatal patterning Trichome initiation at leaf edges Anther development Trichome and root hair cell patterning
Fruit dehiscence (SHP1, SHP2); Carpel identity (AG, SHP1, SHP2); Ovule identity (AG, SHP1, SHP2, STK) Floral organ identity
Systemic acquired resistance Innate immunity Chloroplast development
References Aida et al. (2004)
Gilmour et al. (2004) Aharoni et al. (2004), Broun et al. (2004) Friedrichsen et al. (2002) Yin et al. (2005) Bernhardt et al. (2003), Zhang et al. (2003a)
Lai et al. (2005) Kirik et al. (2005) Millar and Gubler (2005) Schellmann et al. (2002), Esch et al. (2004), Kirik et al. (2004) Alabadi et al. (2002), Mizoguchi et al. (2002) Aida et al. (1997), Takada et al. (2001), Vroemen et al. (2003) Mitsuda et al. (2005) Smith et al. (2004) Abe et al. (2003) Emery et al. (2003), Prigge et al. (2005) Byrne et al. (2002) Bowman et al. (1993), Kempin et al. (1995), Ferr´andiz et al. (2000) Liljegren et al. (2000), Pinyopich et al. (2003)
Pelaz et al. (2000), Honma and Goto (2001), Pelaz et al. (2001), Ditta et al. (2004) Zhang et al. (2003b) Asai et al. (2002) Fitter et al. (2002) (Continued )
42
REGULATION OF TRANSCRIPTION IN PLANTS
Table 2.2 Functional redundancy among related Arabidopsis transcription factors (Continued )
Gene family
Genes
GARP GARP
KAN1, KAN2 ARR1, ARR2, ARR10, ARR11, ARR12 FIL, YAB3 GAI, RGA, RGL1, RGL2
YABBY GRAS
ARF
ARF5/MP, ARF7/NPH4, ARF19
ARF
ARF6, ARF8
ARF ARF SRS BES
ARF3/ETT, ARF4 ARF10, ARF16 STY1, STY2 BES1, BZR1
Redundant or overlapping function
References
Lateral organ polarity Cytokinin signal transduction
Eshed et al. (2001) Mason et al. (2005)
Lateral organ development Gibberellin signaling
Kumaran et al. (2002) Dill and Sun, (2001), King et al. (2001), Lee et al. (2002), Wen and Chang (2002), Tyler et al. (2004) Hardtke et al. (2004), Okushima et al. (2005), Weijers et al. (2005), Wilmoth et al. (2005)
Embryo axial patterning, vascular development (ARF5, ARF7); lateral root formation, leaf cell expansion, gravitropism (ARF7, ARF19) Jasmonic acid production and flower maturation Organ asymmetry Root cap formation Gynoecium development Brassinosteroid signaling
Nagpal et al. (2005) Pekker et al. (2005) Wang et al. (2005a) Kuusk et al. (2002) Yin et al. (2005)
As a result of this genomic organization, regulatory sequences in Arabidopsis, and in plants in general, are easier to identify and delimit experimentally than in animals (see Riechmann, 2002). Compact Arabidopsis and rice 5 promoter sequences often recapitulate faithfully the expression of the native gene when assayed in transgenic plants by reporter gene fusions, that is, in a chromatin context. However, this is not always the case, because regulatory elements can also be localized downstream of the transcription start site. For example, the expression of the Arabidopsis MADS-box gene AG has been shown to depend on its large, second intron, which contains binding sites for at least two AG regulators, LFY and WUSCHEL (WUS) (Sieburth and Meyerowitz, 1997; Bomblies et al., 1999; Busch et al., 1999; Deyholos and Sieburth, 2000; Lohmann et al., 2001). Other Arabidopsis genes that depend on intronic sequences for the regulation of their expression are FLC and SEEDSTICK (both MADS-box genes; Sheldon et al., 2002; Kooiker et al., 2005), as well as PETAL LOSS (a trihelix transcription factor; Brewer et al., 2004). In other cases, regulation is mediated by elements in the 5 or 3 untranslated regions (Larkin et al., 1993; Chen et al., 1998; Yu et al., 2001), in sequences farther downstream (Ito et al., 2004), or, more rarely, even in coding sequences (Ito et al., 2003). Interestingly, two recent reports in maize have described or suggested that sequences located far upstream (50–100 kb) of the transcription start site of a gene can influence its expression, perhaps after being separated from the transcribed sequences by the insertion of retroelements – these are the first exceptions to the general rule that gene regulatory regions in plants are relatively short (Stam et al., 2002; Clark et al., 2004).
TRANSCRIPTION FACTORS OF ARABIDOPSIS AND RICE
43
References M. Abe, H. Katsumata, Y. Komeda and T. Takahashi (2003) Regulation of shoot epidermal cell differentiation by a pair of homeodomain proteins in Arabidopsis. Development 130 (4), 635– 643. A. Aharoni, S. Dixit, R. Jetter, E. Thoenes, G. Van Arkel and A. Pereira (2004) The SHINE clade of AP2 domain transcription factors activates wax biosynthesis, alters cuticle properties, and confers drought tolerance when overexpressed in Arabidopsis. The Plant Cell 16 (9), 2463–2480. M. Aida, D. Beis, R. Heidstra, et al. (2004) The PLETHORA genes mediate patterning of the Arabidopsis root stem cell niche. Cell 119 (1), 109–120. M. Aida, T. Ishida, H. Fukaki, H. Fujisawa and M. Tasaka (1997) Genes involved in organ separation in Arabidopsis: an analysis of the cup-shaped cotyledon mutant. The Plant Cell 9 (6), 841– 857. D. Alabadi, M.J. Yanovsky, P. Mas, S.L. Harmer and S.A. Kay (2002) Critical role for CCA1 and LHY in maintaining circadian rhythmicity in Arabidopsis. Current Biology 12 (9), 757–761. E.R. Alvarez-Buylla, S. Pelaz, S.J. Liljegren, et al. (2000) An ancestral MADS-box gene duplication occurred before the divergence of plants and animals. Proceedings of the National Academy of Sciences of the United States of America 97 (10), 5328–5333. Arabidopsis Genome Initiative (2000) Analysis of the genome sequence of the flowering plant Arabidopsis thaliana. Nature 408 (6814), 796–815. T. Asai, G. Tena, J. Plotnikova, et al. (2002) MAP kinase signalling cascade in Arabidopsis innate immunity. Nature 415 (6875), 977–983. M. Ayadi, V. Delaporte, Y.F. Li and D.X. Zhou (2004) Analysis of GT-3a identifies a distinct subgroup of trihelix DNA-binding transcription factors in Arabidopsis. FEBS Letters 562 (1–3), 147– 154. P.C. Bailey, C. Martin, G. Toledo-Ortiz, et al. (2003) Update on the basic helix-loop-helix transcription factor gene family in Arabidopsis thaliana. The Plant Cell 15 (11), 2497–2501. S. Balaji, M.M. Babu, L.M. Iyer and L. Aravind (2005) Discovery of the principal specific transcription factors of Apicomplexa and their implication for the evolution of the AP2-integrase DNA binding domains. Nucleic Acids Research 33 (13), 3994–4006. D.R. Bastola, V.V. Pethe and I. Winicov (1998) Alfin1, a novel zinc-finger protein in alfalfa roots that binds to promoter elements in the salt-inducible MsPRP2 gene. Plant Molecular Biology 38 (6), 1123–1135. A. Becker and G. Theissen (2003) The major clades of MADS-box genes and their role in the development and evolution of flowering plants. Molecular Phylogenetics and Evolution 29 (3), 464–489. J.L. Bennetzen, C. Coleman, R. Liu, J. Ma and W. Ramakrishna (2004) Consistent over-estimation of gene number in complex plant genomes. Current Opinion in Plant Biology 7 (6), 732–736. D.C. Bergmann, W. Lukowitz and C.R. Somerville (2004) Stomatal development and pattern controlled by a MAPKK kinase. Science 304 (5676), 1494–1497. C. Bernhardt, M.M. Lee, A. Gonzalez, F. Zhang, A. Lloyd and J. Schiefelbein (2003) The bHLH genes GLABRA3 (GL3) and ENHANCER of GLABRA3 (EGL3) specify epidermal cell fate in the Arabidopsis root. Development 130 (26), 6431–6439. G. Blanc, A. Barakat, R. Guyot, R. Cooke and M. Delseny (2000) Extensive duplication and reshuffling in the Arabidopsis genome. The Plant Cell 12 (7), 1093–1101. G. Blanc, K. Hokamp and K.H. Wolfe (2003) A recent polyploidy superimposed on older large-scale duplications in the Arabidopsis genome. Genome Research 13 (2), 137–144. G. Blanc and K.H. Wolfe (2004) Functional divergence of duplicated genes formed by polyploidy during Arabidopsis evolution. The Plant Cell 16 (7), 1679–1691. C. Bolle (2004) The role of GRAS proteins in plant signal transduction and development. Planta 218 (5), 683–692. K. Bomblies, N. Dagenais and D. Weigel (1999) Redundant enhancers mediate transcriptional repression of AGAMOUS by APETALA2. Developmental Biology 216 (1), 260–264.
44
REGULATION OF TRANSCRIPTION IN PLANTS
N. Bouche, A. Scharlat, W. Snedden, D. Bouchez and H. Fromm (2002) A novel family of calmodulinbinding transcription activators in multicellular organisms. The Journal of Biological Chemistry 277 (24), 21851–21861. J.E. Bowers, B.A. Chapman, J. Rong and A.H. Paterson (2003) Unravelling angiosperm genome evolution by phylogenetic analysis of chromosomal duplication events. Nature 422 (6930), 433– 438. J.L. Bowman (2000) The YABBY gene family and abaxial cell fate. Current Opinion in Plant Biology 3 (1), 17–22. J.L. Bowman, J. Alvarez, D. Weigel, E.M. Meyerowitz and D. Smyth (1993) Control of flower development in Arabidopsis thaliana by APETALA1 and interacting genes. Development 119, 721–743. P.B. Brewer, P.A. Howles, K. Dorian, et al. (2004) PETAL LOSS, a trihelix transcription factor gene, regulates perianth architecture in the Arabidopsis flower. Development 131 (16), 4035–4045. P. Broun, P. Poindexter, E. Osborne, C.Z. Jiang and J.L. Riechmann (2004) WIN1, a transcriptional activator of epidermal wax accumulation in Arabidopsis. Proceedings of the National Academy of Sciences of the United States of America 101 (13), 4706–4711. M.J. Buck and W.R. Atchley (2003) Phylogenetic analysis of plant basic helix-loop-helix proteins. Journal of Molecular Evolution 56 (6), 742–750. M.A. Busch, K. Bomblies and D. Weigel (1999) Activation of a floral homeotic gene in Arabidopsis. Science 285 (5427), 585–587. M.E. Byrne, J. Simorowski and R.A. Martienssen (2002) ASYMMETRIC LEAVES1 reveals knox gene redundancy in Arabidopsis. Development 129 (8), 1957–1965. M.V. Byzova, J. Franken, M.G. Aarts, et al. (1999) Arabidopsis STERILE APETALA, a multifunctional gene regulating inflorescence, flower, and ovule development. Genes and Development 13 (8), 1002–1014. B. C¸ akir, A. Agasse, C. Gaillard, A. Saumonneau, S. Delrot and R. Atanassova (2003) A grape ASR protein involved in sugar and abscisic acid signaling. The Plant Cell 15 (9), 2165–2180. S.B. Cannon, A. Mitra, A. Baumgarten, N.D. Young and G. May (2004) The roles of segmental and tandem gene duplication in the evolution of large gene families in Arabidopsis thaliana. BMC Plant Biology 4 (1), 10. G. Cardon, S. Hohmann, J. Klein, K. Nettesheim, H. Saedler and P. Huijser (1999) Molecular characterisation of the Arabidopsis SBP-box genes. Gene 237 (1), 91–104. C.C. Carles, D. Choffnes-Inada, K. Reville, K. Lertpiriyapong and J.C. Fletcher (2005) ULTRAPETALA1 encodes a SAND domain putative transcriptional regulator that controls shoot and floral meristem activity in Arabidopsis. Development 132 (5), 897–911. J.L. Carrasco, G. Ancillo, M.J. Castello and P. Vera (2005) A novel DNA-binding motif, hallmark of a new family of plant transcription factors. Plant Physiology 137 (2), 602–606. J.L. Carrasco, G. Ancillo, E. Mayda and P. Vera (2003) A novel transcription factor involved in plant defense endowed with protein phosphatase activity. The EMBO Journal 22 (13), 3376– 3384. K. Carroll, C. Gomez and L. Shapiro (2004) Tubby proteins: the plot thickens. Nature Reviews Molecular Cell Biology 5 (1), 55–63. R.L. Chan, G.M. Gago, C.M. Palena and D.H. Gonzalez (1998) Homeoboxes in plant development. Biochimica et Biophysica Acta 1442 (1), 1–19. Q. Chao, M. Rothenberg, R. Solano, G. Roman, W. Terzaghi and J.R. Ecker (1997) Activation of the ethylene gas response pathway in Arabidopsis by the nuclear protein ETHYLENEINSENSITIVE3 and related proteins. Cell 89 (7), 1133–1144. R. Chen, D.L. Silver and F.J. de Bruijn (1998) Nodule parenchyma-specific expression of the Sesbania rostrata early nodulin gene SrEnod2 is mediated by its 3 untranslated region. The Plant Cell 10 (10), 1585–1602. W. Chen, N.J. Provart, J. Glazebrook, et al. (2002) Expression profile matrix of Arabidopsis transcription factor genes suggests their putative functions in response to environmental stresses. The Plant Cell 14 (3), 559–574.
TRANSCRIPTION FACTORS OF ARABIDOPSIS AND RICE
45
D. Choi, J.H. Kim and H. Kende (2004) Whole genome analysis of the OsGRF gene family encoding plant-specific putative transcription activators in rice (Oryza sativa L.). Plant and Cell Physiology 45 (7), 897–904. H. Christiansen, A.C. Hansen, I. Vijn, et al. (1996) A novel type of DNA-binding protein interacts with a conserved sequence in an early nodulin ENOD12 promoter. Plant Molecular Biology 32 (5), 809–821. R.M. Clark, E. Linton, J. Messing and J.F. Doebley (2004) Pattern of diversity in the genomic region near the maize domestication gene tb1. Proceedings of the National Academy of Sciences of the United States of America 101 (3), 700–707. P. Cubas (2002) Role of TCP genes in the evolution of morphological characters in angiosperms. In: Developmental Genetics and Plant Evolution (eds Q.C.B. Cronk, R.M. Bateman and J.A. Hawkins), pp. 247–266. Taylor and Francis, London. P. Cubas, N. Lauter, J. Doebley and E. Coen (1999) The TCP domain: a motif found in proteins regulating plant growth and development. The Plant Journal 18 (2), 215–222. J. Curaba, M. Herzog and G. Vachon (2003) GeBP, the first member of a new gene family in Arabidopsis, encodes a nuclear protein with DNA-binding activity and is regulated by KNAT1. The Plant Journal 33 (2), 305–317. C. Cvitanich, N. Pallisgaard, K.A. Nielsen, et al. (2000) CPP1, a DNA-binding protein involved in the expression of a soybean leghemoglobin c3 gene. Proceedings of the National Academy of Sciences of the United States of America 97 (14), 8163–8168. T. Czechowski, R.P. Bari, M. Stitt, W.R. Scheible and M.K. Udvardi (2004) Real-time RT-PCR profiling of over 1400 Arabidopsis transcription factors: unprecedented sensitivity reveals novel root- and shoot-specific genes. The Plant Journal 38 (2), 366–379. E.H. Davidson (2001) Genomic Regulatory Systems: Development and evolution. Academic Press, San Diego, CA. S. De Bodt, J. Raes, K. Florquin, et al. (2003a) Genomewide structural annotation and evolutionary analysis of the type I MADS-box genes in plants. Journal of Molecular Evolution 56 (5), 573–586. S. De Bodt, J. Raes, Y. Van de Peer and G. Theissen (2003b) And then there were many: MADS goes genomic. Trends in Plant Science 8 (10), 475–483. S. De Folter, J. Busscher, L. Colombo, A. Losa and G.C. Angenent (2004) Transcript profiling of transcription factor genes during silique development in Arabidopsis. Plant Molecular Biology 56 (3), 351–366. S.M. De Jager, M. Menges, U.-M. Bauer and J.A.H. Murray (2001) Arabidopsis E2F1 binds a sequence present in the promoter of S-phase-regulated gene AtCDC6 and is a member of a multigene family with differential activities. Plant Molecular Biology 47, 555–568. C.D. Deppmann, A. Acharya, V. Rishi, et al. (2004) Dimerization specificity of all 67 B-ZIP motifs in Arabidopsis thaliana: a comparison to Homo sapiens B-ZIP motifs. Nucleic Acids Research 32 (11), 3435–3445. D. Desveaux, J. Allard, N. Brisson and J. Sygusch (2002) A new family of plant transcription factors displays a novel ssDNA-binding surface. Nature Structural Biology 9 (7), 512–517. D. Desveaux, C. Despr´es, A. Joyeux, R. Subramaniam and N. Brisson (2000) PBF-2 is a novel singlestranded DNA binding factor implicated in PR-10a gene activation in potato. The Plant Cell 12, 1477–1489. D. Desveaux, A. Marechal and N. Brisson (2005) Whirly transcription factors: defense gene regulation and beyond. Trends in Plant Science 10 (2), 95–102. D. Desveaux, R. Subramaniam, C. Despres, et al. (2004) A ‘Whirly’ transcription factor is required for salicylic acid-dependent disease resistance in Arabidopsis. Developmental Cell 6 (2), 229–240. M.K. Deyholos and L.E. Sieburth (2000) Separable whorl-specific expression and negative regulation by enhancer elements within the AGAMOUS second intron. The Plant Cell 12 (10), 1799–1810. A. Dill and T.-P. Sun (2001) Synergistic derepression of gibberellin signaling by removing RGA and GAI function in Arabidopsis thaliana. Genetics 159, 777–785.
46
REGULATION OF TRANSCRIPTION IN PLANTS
G. Ditta, A. Pinyopich, P. Robles, S. Pelaz and M.F. Yanofsky (2004) The SEP4 gene of Arabidopsis thaliana functions in floral organ and meristem identity. Current Biology 14 (21), 1935– 1940. J.F. Emery, S.K. Floyd, J. Alvarez, et al. (2003) Radial patterning of Arabidopsis shoots by class III HD-ZIP and KANADI genes. Current Biology 13 (20), 1768–1774. C.C. Englbrecht, H. Schoof and S. Bohm (2004) Conservation, diversification and expansion of C2H2 zinc finger proteins in the Arabidopsis thaliana genome. BMC Genomics 5 (1), 39. H.A. Ernst, A. Nina Olsen, K. Skriver, S. Larsen and L. Lo Leggio (2004) Structure of the conserved domain of ANAC, a member of the NAC family of transcription factors. EMBO Reports 5, 297–303. J.J. Esch, M.A. Chen, M. Hillestad and M.D. Marks (2004) Comparison of TRY and the closely related At1g01380 gene in controlling Arabidopsis trichome patterning. The Plant Journal 40 (6), 860–869. Y. Eshed, S.F. Baum, J.V. Perea and J.L. Bowman (2001) Establishment of polarity in lateral organs of plants. Current Biology 11, 1251–1260. T. Eulgem, P.J. Rushton, S. Robatzek and I.E. Somssich (2000) The WRKY superfamily of plant transcription factors. Trends in Plant Science 5 (5), 199–206. C. Ferr´andiz, Q. Gu, R. Martienssen and M.F. Yanofsky (2000) Redundant regulation of meristem identity and plant architecture by FRUITFULL, APETALA1 and CAULIFLOWER. Development 127 (4), 725–734. D.W. Fitter, D.J. Martin, M.J. Copley, R.W. Scotland and J.A. Langdale (2002) GLK gene pairs regulate chloroplast development in diverse plant species. The Plant Journal 31 (6), 713–727. I. Fridborg, S. Kuusk, M. Robertson and E. Sundberg (2001) The Arabidopsis protein SHI represses gibberellin responses in Arabidopsis and barley. Plant Physiology 127, 937–948. D.M. Friedrichsen, J. Nemhauser, T. Muramitsu, et al. (2002) Three redundant brassinosteroid early response genes encode putative bHLH transcription factors required for normal growth. Genetics 162 (3), 1445–1456. S.J. Gilmour, S.G. Fowler and M.F. Thomashow (2004) Arabidopsis transcriptional activators CBF1, CBF2, and CBF3 have matching functional activities. Plant Molecular Biology 54 (5), 767–781. S.A. Goff, D. Ricke, T.H. Lan, et al. (2002) A draft sequence of the rice genome (Oryza sativa L. ssp. japonica). Science 296 (5565), 92–100. C. Gomez-Mena, S. de Folter, M.M. Costa, G.C. Angenent and R. Sablowski (2005) Transcriptional program controlled by the floral homeotic gene AGAMOUS during early organogenesis. Development 132 (3), 429–438. S. Griffiths, R.P. Dunford, G. Coupland and D.A. Laurie (2003) The evolution of CONSTANS-like gene families in barley, rice, and Arabidopsis. Plant Physiology 131 (4), 1855–1867. T.J. Guilfoyle and G. Hagen (2001) Auxin response factors. Journal of Plant Growth Regulation 20, 281–291. A. Guo, K. He, D. Liu, et al. (2005) DATF: a database of Arabidopsis transcription factors. Bioinformatics 21 (10), 2568–2569. G. Gusmaroli, C. Tonelli and R. Mantovani (2001) Regulation of the CCAAT-binding NF-Y subunits in Arabidopsis thaliana. Gene 264 (2), 173–185. G. Gusmaroli, C. Tonelli and R. Mantovani (2002) Regulation of novel members of the Arabidopsis thaliana CCAAT-binding nuclear factor Y subunits. Gene 283 (1–2), 41–48. N. Gutterson and T.L. Reuber (2004) Regulation of disease resistance pathways by AP2/ERF transcription factors. Current Opinion in Plant Biology 7 (4), 465–471. B.J. Haas, J.R. Wortman, C.M. Ronning, et al. (2005) Complete reannotation of the Arabidopsis genome: methods, tools, protocols and the final release. BMC Biology 3 (1), 7. M.A. Hannah, A.G. Heyer and D.K. Hincha (2005) A global survey of gene regulation during cold acclimation in Arabidopsis thaliana. PLoS Genetics 1 (2), e26. C.S. Hardtke, W. Ckurshumova, D.P. Vidaurre, et al. (2004) Overlapping and non-redundant functions of the Arabidopsis auxin response factors MONOPTEROS and NONPHOTOTROPIC HYPOCOTYL 4. Development 131 (5), 1089–1100.
TRANSCRIPTION FACTORS OF ARABIDOPSIS AND RICE
47
M.A. Heim, M. Jakoby, M. Werber, C. Martin, B. Weisshaar and P.C. Bailey (2003) The basic helixloop-helix transcription factor family in plants: a genome-wide study of protein structure and functional diversity. Molecular Biology and Evolution 20 (5), 735–747. T. Honma and K. Goto (2001) Complexes of MADS-box proteins are sufficient to convert leaves into floral organs. Nature 409 (6819), 525–529. K. Hosoda, A. Imamura, E. Katoh, et al. (2002) Molecular structure of the GARP family of plant Myb-related DNA binding motifs of the Arabidopsis response regulators. The Plant Cell 14 (9), 2015–2029. I. Hwang, H.C. Chen and J. Sheen (2002) Two-component signal transduction pathways in Arabidopsis. Plant Physiology 129 (2), 500–515. International Rice Genome Sequencing Project (2005) The map-based sequence of the rice genome. Nature 436 (7052), 793–800. T. Ito, H. Sakai and E.M. Meyerowitz (2003) Whorl-specific expression of the SUPERMAN gene of Arabidopsis is mediated by cis elements in the transcribed region. Current Biology 13 (17), 1524–1530. T. Ito, F. Wellmer, H. Yu, et al. (2004) The homeotic protein AGAMOUS controls microsporogenesis by regulation of SPOROCYTELESS. Nature 430 (6997), 356–360. M. Jakoby, B. Weisshaar, W. Droge-Laser, et al. (2002) bZIP transcription factors in Arabidopsis. Trends in Plant Science 7 (3), 106–111. L. Jia, M.T. Clegg and T. Jiang (2004) Evolutionary dynamics of the DNA-binding domains in putative R2R3-MYB genes identified from rice subspecies indica and japonica genomes. Plant Physiology 134 (2), 575–585. C. Jiang, X. Gu and T. Peterson (2004) Identification of conserved gene structures and carboxy-terminal motifs in the Myb gene family of Arabidopsis and Oryza sativa L. ssp. indica. Genome Biology 5 (7), R46. Y. Jiao, H. Yang, L. Ma, et al. (2003) A genome-wide analysis of blue-light regulation of Arabidopsis transcription factor gene expression during seedling development. Plant Physiology 133 (4), 1480–1493. A. Kawaoka, P. Kaothien, K. Yoshida, S. Endo, K. Yamada and H. Ebinuma (2000) Functional analysis of tobacco LIM protein Ntlim1 involved in lignin biosynthesis. The Plant Journal 22 (4), 289–301. S.A. Kempin, B. Savidge and M.F. Yanofsky (1995) Molecular basis of the cauliflower phenotype in Arabidopsis. Science 267 (5197), 522–525. J.H. Kim, D. Choi and H. Kende (2003) The AtGRF family of putative transcription factors is involved in leaf and cotyledon growth in Arabidopsis. The Plant Journal 36 (1), 94–104. K.E. King, T. Moritz and N.P. Harberd (2001) Gibberellins are not required for normal stem growth in Arabidopsis thaliana in the absence of GAI and RGA. Genetics 159, 767–776. V. Kirik, M.M. Lee, K. Wester, et al. (2005) Functional diversification of MYB23 and GL1 genes in trichome morphogenesis and initiation. Development 132 (7), 1477–1485. V. Kirik, M. Simon, M. Huelskamp and J. Schiefelbein (2004) The ENHANCER OF TRY AND CPC1 gene acts redundantly with TRIPTYCHON and CAPRICE in trichome and root hair cell patterning in Arabidopsis. Developmental Biology 268 (2), 506–513. R. Kofuji, N. Sumikawa, M. Yamasaki, et al. (2003) Evolution and divergence of the MADS-box gene family based on genome-wide expression analyses. Molecular Biology and Evolution 20 (12), 1963–1977. M. Kooiker, C.A. Airoldi, A. Losa, et al. (2005) BASIC PENTACYSTEINE1, a GA binding protein that induces conformational changes in the regulatory region of the homeotic Arabidopsis gene SEEDSTICK. The Plant Cell 17, 722–729. S. Kosugi and Y. Ohashi (2002) DNA binding and dimerization specificity and potential targets for the TCP protein family. The Plant Journal 30 (3), 337–348. S. Kotak, M. Port, A. Ganguli, F. Bicker and P. Von Koskull-Doring (2004) Characterization of Cterminal domains of Arabidopsis heat stress transcription factors (Hsfs) and identification of a new signature combination of plant class A Hsfs with AHA and NES motifs essential for activator function and intracellular localization. The Plant Journal 39 (1), 98–112.
48
REGULATION OF TRANSCRIPTION IN PLANTS
K. Krause, I. Kilbienski, M. Mulisch, A. Rodiger, A. Schafer and K. Krupinska (2005) DNA-binding proteins of the Whirly family in Arabidopsis thaliana are targeted to the organelles. FEBS Letters 579 (17), 3707–3712. B.A. Krizek, V. Prost and A. Macias (2000) AINTEGUMENTA promotes petal identity and acts as a negative regulator of AGAMOUS. The Plant Cell 12 (8), 1357–1366. M.K. Kumaran, J.L. Bowman and V. Sundaresan (2002) YABBY polarity genes mediate the repression of KNOX homeobox genes in Arabidopsis. The Plant Cell 14 (11), 2761–2770. S. Kuusk, J.J. Sohlberg, J.A. Long, I. Fridborg and E. Sundberg (2002) STY1 and STY2 promote the formation of apical tissues during Arabidopsis gynoecium development. Development 129 (20), 4707–4717. U. Lagercrantz and T. Axelsson (2000) Rapid evolution of the family of CONSTANS-LIKE genes in plants. Molecular Biology and Evolution 17 (10), 1499–1507. L.B. Lai, J.A. Nadeau, J. Lucas, et al. (2005) The Arabidopsis R2R3 MYB proteins FOUR LIPS and MYB88 restrict divisions late in the stomatal cell lineage. The Plant Cell 17 (10), 2754– 2767. J.C. Larkin, D.G. Oppenheimer, S. Pollock and M.D. Marks (1993) Arabidopsis GLABROUS1 gene requires downstream sequences for function. The Plant Cell 5, 1739–1748. D.S. Latchman (2003) Eukaryotic Transcription Factors. Academic Press, San Diego, CA. S. Lee, H. Cheng, K.E. King, et al. (2002) Gibberellin regulates Arabidopsis seed germination via RGL2, a GAI/RGA-like gene whose expression is up-regulated following imbibition. Genes and Development 16 (5), 646–658. M.D. Lehti-Shiu, B.J. Adamczyk and D.E. Fernandez (2005) Expression of MADS-box genes during the embryonic phase in Arabidopsis. Plant Molecular Biology 58 (1), 89–107. Z. Li and T.L. Thomas (1998) PEI1, an embryo-specific zinc finger protein gene required for heartstage embryo formation in Arabidopsis. The Plant Cell 10 (3), 383–398. D. Lijavetzky, P. Carbonero and J. Vicente-Carbajosa (2003) Genome-wide comparative phylogenetic analysis of the rice and Arabidopsis Dof gene families. BMC Evolutionary Biology 3 (1), 17. S.J. Liljegren, G.S. Ditta, Y. Eshed, B. Savidge, J.L. Bowman and M.F. Yanofsky (2000) SHATTERPROOF MADS-box genes control seed dispersal in Arabidopsis. Nature 404 (6779), 766–770. J.U. Lohmann, R.L. Hong, M. Hobe, et al. (2001) A molecular link between stem cell regulation and floral patterning in Arabidopsis. Cell 105, 793–803. J. Lohrmann and K. Harter (2002) Plant two-component signaling systems and the role of response regulators. Plant Physiology 128 (2), 363–369. T. Lotan, M. Ohto, K.M. Yee, et al. (1998) Arabidopsis LEAFY COTYLEDON1 is sufficient to induce embryo development in vegetative cells. Cell 93 (7), 1195–1205. N.M. Luscombe, S.E. Austin, H.M. Berman and J.M. Thornton (2000) An overview of the structures of protein-DNA complexes (reviews). Genome Biology 1 (1), 001.001–001.037. E. Magnani, K. Sjolander and S. Hake (2004) From endonucleases to transcription factors: evolution of the AP2 DNA binding domain in plants. The Plant Cell 16 (9), 2265–2277. L.P. Martinez-Castilla and E.R. Alvarez-Buylla (2003) Adaptive evolution in the Arabidopsis MADSbox gene family inferred from its complete resolved phylogeny. Proceedings of the National Academy of Sciences of the United States of America 100 (23), 13407–13412. M.G. Mason, J. Li, D.E. Mathews, J.J. Kieber and G.E. Schaller (2004) Type-B response regulators display overlapping expression patterns in Arabidopsis. Plant Physiology 135 (2), 927–937. M.G. Mason, D.E. Mathews, D.A. Argyros, et al. (2005) Multiple type-B response regulators mediate cytokinin signal transduction in Arabidopsis. The Plant Cell 17 (11), 3007–3018. R.J. Meister, L.A. Williams, M.M. Monfared, et al. (2004) Definition and interactions of a positive regulatory element of the Arabidopsis INNER NO OUTER promoter. The Plant Journal 37 (3), 426–438. G. Mele, N. Ori, Y. Sato and S. Hake (2003) The knotted1-like homeobox gene BREVIPEDICELLUS regulates cell differentiation by modulating metabolic pathways. Genes and Development 17, 2088–2093.
TRANSCRIPTION FACTORS OF ARABIDOPSIS AND RICE
49
J. Messing, A.K. Bharti, W.M. Karlowski, et al. (2004) Sequence composition and genome organization of maize. Proceedings of the National Academy of Sciences of the United States of America 101 (40), 14349–14354. A.A. Millar and F. Gubler (2005) The Arabidopsis GAMYB-like genes, MYB33 and MYB65, are microRNA-regulated genes that redundantly facilitate anther development. The Plant Cell 17, 705–721. N. Mitsuda, T. Hisabori, K. Takeyasu and M.H. Sato (2004) VOZ; isolation and characterization of novel vascular plant transcription factors with a one-zinc finger from Arabidopsis thaliana. Plant and Cell Physiology 45 (7), 845–854. N. Mitsuda, M. Seki, K. Shinozaki and M. Ohme-Takagi (2005) The NAC transcription factors NST1 and NST2 of Arabidopsis regulate secondary wall thickenings and are required for anther dehiscence. Plant Cell 17 (11), 2993–3006. T. Mizoguchi, K. Wheatley, Y. Hanzawa, et al. (2002) LHY and CCA1 are partially redundant genes required to maintain circadian rhythms in Arabidopsis. Developmental Cell 2 (5), 629–641. C.F. Mouchel, G.C. Briggs and C.S. Hardtke (2004) Natural genetic variation in Arabidopsis identifies BREVIS RADIX, a novel regulator of cell proliferation and elongation in the root. Genes and Development 18, 700–714. T. Munster, W. Deleu, L.U. Wingen, et al. (2002) Maize MADS-box genes galore. Maydica 47 (3–4), 287–301. P. Nagpal, C.M. Ellis, H. Weber, et al. (2005) Auxin response factors ARF6 and ARF8 promote jasmonic acid production and flower maturation. Development 132 (18), 4107–4118. J. Nam, C.W. DePamphilis, H. Ma and M. Nei (2003) Antiquity and evolution of the MADS-box gene family controlling flower development in plants. Molecular Biology and Evolution 20 (9), 1435–1447. J. Nam, J. Kim, S. Lee, G. An, H. Ma and M. Nei (2004) Type I MADS-box genes have experienced faster birth-and-death evolution than type II MADS-box genes in angiosperms. Proceedings of the National Academy of Sciences of the United States of America 101 (7), 1910– 1915. J.L. Nemhauser, T.C. Mockler and J. Chory (2004) Interdependency of brassinosteroid and auxin signaling in Arabidopsis. PLoS Biology 2 (9), E258. M. Ng and M.F. Yanofsky (2001) Function and evolution of the plant MADS-box gene family. Nature Reviews Genetics 2 (3), 186–195. S. Nole-Wilson, T.L. Tranby and B.A. Krizek (2005) AINTEGUMENTA-like (AIL) genes are expressed in young tissues and may specify meristematic or division-competent states. Plant Molecular Biology 57 (5), 613–628. L. Nover, K. Bharti, P. D¨oring, S.K. Mishra, A. Ganguli and K.D. Scharf (2001) Arabidopsis and the heat stress transcription factor world: how many heat stress transcription factors do we need? Cell Stress and Chaperones 6 (3), 177–189. Y. Okushima, P.J. Overvoorde, K. Arima, et al. (2005) Functional genomic analysis of the AUXIN RESPONSE FACTOR gene family members in Arabidopsis thaliana: unique and overlapping functions of ARF7 and ARF19. The Plant Cell 17 (2), 444–463. A.N. Olsen, H.A. Ernst, L.L. Leggio and K. Skriver (2005) NAC transcription factors: structurally distinct, functionally diverse. Trends in Plant Science 10 (2), 79–87. H. Ooka, K. Satoh, K. Doi, et al. (2003) Comprehensive analysis of NAC family genes in Oryza sativa and Arabidopsis thaliana. DNA Research 10 (6), 239–247. L. Østergaard and M.F. Yanofsky (2004) Establishing gene function by mutagenesis in Arabidopsis thaliana. The Plant Journal 39 (5), 682–696. L.E. Palmer, P.D. Rabinowicz, A.L. O’Shaughnessy, et al. (2003) Maize genome sequencing by methylation filtration. Science 302 (5653), 2115–2117. L. Parenicova, S. De Folter, M. Kieffer, et al. (2003) Molecular and phylogenetic analyses of the complete MADS-box transcription factor family in Arabidopsis: new openings to the MADS world. The Plant Cell 15 (7), 1538–1551.
50
REGULATION OF TRANSCRIPTION IN PLANTS
I. Pekker, J.P. Alvarez and Y. Eshed (2005) Auxin response factors mediate Arabidopsis organ asymmetry via modulation of KANADI activity. The Plant Cell 17 (11), 2899–2910. S. Pelaz, G.S. Ditta, E. Baumann, E. Wisman and M.F. Yanofsky (2000) B and C floral organ identity functions require SEPALLATA MADS-box genes. Nature 405 (6783), 200–203. S. Pelaz, R. Tapia-Lopez, E.R. Alvarez-Buylla and M.F. Yanofsky (2001) Conversion of leaves into petals in Arabidopsis. Current Biology 11 (3), 182–184. A. Pinyopich, G.S. Ditta, B. Savidge, et al. (2003) Assessing the redundancy of MADS-box genes during carpel and ovule development. Nature 424 (6944), 85–88. M.J. Prigge, D. Otsuga, J.M. Alonso, J.R. Ecker, G.N. Drews and S.E. Clark (2005) Class III homeodomain-leucine zipper gene family members have overlapping, antagonistic, and distinct roles in Arabidopsis development. The Plant Cell 17 (1), 61–76. L.D. Pysh, J.W. Wysocka-Diller, C. Camilleri, D. Bouchez and P.N. Benfey (1999) The GRAS gene family in Arabidopsis: sequence characterization and basic expression analysis of the SCARECROW-LIKE genes. The Plant Journal 18 (1), 111–119. P.D. Rabinowicz, E.L. Braun, A.D. Wolfe, B. Bowen and E. Grotewold (1999) Maize R2R3 Myb genes: Sequence analysis reveals amplification in the higher plants. Genetics 153 (1), 427–444. O.J. Ratcliffe, R.W. Kumimoto, B.J. Wong and J.L. Riechmann (2003) Analysis of the Arabidopsis MADS AFFECTING FLOWERING gene family: MAF2 prevents vernalization by short periods of cold. The Plant Cell 15 (5), 1159–1169. D. Raventos, K. Skriver, M. Schlein, et al. (1998) HRT, a novel zinc finger, transcriptional repressor from barley. The Journal of Biological Chemistry 273 (36), 23313–23320. D.L. Remington, T.J. Vision, T.J. Guilfoyle and J.W. Reed (2004) Contrasting modes of diversification in the Aux/IAA and ARF gene families. Plant Physiology 135 (3), 1738–1752. J.C. Reyes, M.I. Muro-Pastor and F.J. Florencio (2004) The GATA family of transcription factors in Arabidopsis and rice. Plant Physiology 134 (4), 1718–1732. D.E. Richards, J. Peng and N.P. Harberd (2000) Plant GRAS and metazoan STATs: one family? Bioessays 22 (6), 573–577. J.L. Riechmann (2002) Transcriptional regulation: a genomic overview. In: The Arabidopsis Book (eds C.R. Somerville and E.M. Meyerowitz). American Society of Plant Biologists, Rockville, MD. J.L. Riechmann, J. Heard, G. Martin, et al. (2000) Arabidopsis transcription factors: genome-wide comparative analysis among eukaryotes. Science 290 (5499), 2105–2110. J.L. Riechmann and E.M. Meyerowitz (1997) MADS domain proteins in plant development. Biological Chemistry 378 (10), 1079–1101. J.L. Riechmann and E.M. Meyerowitz (1998) The AP2/EREBP family of plant transcription factors. Biological Chemistry 379 (6), 633–646. J.L. Riechmann and O.J. Ratcliffe (2000) A genomic perspective on plant transcription factors. Current Opinion in Plant Biology 3 (5), 423–434. L. Rossini, L. Cribb, D.J. Martin and J.A. Langdale (2001) The maize golden2 gene defines a novel class of transcriptional regulators in plants. The Plant Cell 13 (5), 1231–1244. Y. Sakuma, Q. Liu, J.G. Dubouzet, H. Abe, K. Shinozaki and K. Yamaguchi-Shinozaki (2002) DNAbinding specificity of the ERF/AP2 domain of Arabidopsis DREBs, transcription factors involved in dehydration- and cold-inducible gene expression. Biochemical and Biophysical Research Communications 290 (3), 998–1009. I. Sangwan and M.R. O’Brian (2002) Identification of a soybean protein that interacts with GAGA element dinucleotide repeat DNA. Plant Physiology 129 (4), 1788–1794. L. Schauser, W. Wieloch and J. Stougaard (2005) Evolution of NIN-like proteins in Arabidopsis, rice, and Lotus japonicus. Journal of Molecular Evolution 60 (2), 229–237. S. Schellmann, A. Schnittger, V. Kirik, et al. (2002) TRIPTYCHON and CAPRICE mediate lateral inhibition during trichome and root hair patterning in Arabidopsis. The EMBO Journal 21 (19), 5036–5046. U. Schiefthaler, S. Balasubramanian, P. Sieber, D. Chevalier, E. Wisman and K. Schneitz (1999) Molecular analysis of NOZZLE, a gene involved in pattern formation and early sporogenesis
TRANSCRIPTION FACTORS OF ARABIDOPSIS AND RICE
51
during sex organ development in Arabidopsis thaliana. Proceedings of the National Academy of Sciences of the United States of America 96 (20), 11664–11669. M. Schmid, T.S. Davison, S.R. Henz, et al. (2005) A gene expression map of Arabidopsis thaliana development. Nature Genetics 37, 501–506. M. Schmid, N.H. Uhlenhaut, F. Godard, et al. (2003) Dissection of floral induction pathways using global expression analysis. Development 130 (24), 6001–6012. P.S. Schnable, F. Hochholdinger and M. Nakazono (2004) Global expression profiling applied to plant development. Current Opinion in Plant Biology 7 (1), 50–56. K. Schrick, D. Nguyen, W.M. Karlowski and K.F. Mayer (2004) START lipid/sterol-binding domains are amplified in plants and are predominantly associated with homeodomain transcription factors. Genome Biology 5 (6), R41. M. Seki, M. Narusaka, A. Kamiya, et al. (2002) Functional annotation of a full-length Arabidopsis cDNA collection. Science 296 (5565), 141–145. C.C. Sheldon, A.B. Conn, E.S. Dennis and W.J. Peacock (2002) Different regulatory regions are required for the vernalization-induced repression of FLOWERING LOCUS C and for the epigenetic maintenance of repression. The Plant Cell 14 (10), 2527–2537. M. Shikata, Y. Matsuda, K. Ando, et al. (2004) Characterization of Arabidopsis ZIM, a member of a novel plant-specific GATA factor gene family. Journal of Experimental Botany 55 (397), 631–639. S.H. Shiu, M.C. Shih and W.H. Li (2005) Transcription factor families have much higher expansion rates in plants than in animals. Plant Physiology 139 (1), 18–26. Y. Sib´eril, P. Doireau and P. Gantet (2001) Plant bZIP G-box binding factors. Modular structure and activation mechanisms. European Journal of Biochemistry 268 (22), 5655–5666. L.E. Sieburth and E.M. Meyerowitz (1997) Molecular dissection of the AGAMOUS control region shows that cis elements for spatial regulation are located intragenically. The Plant Cell 9 (3), 355–365. C. Simillion, K. Vandepoele, M.C. Van Montagu, M. Zabeau and Y. Van de Peer (2002) The hidden duplication past of Arabidopsis thaliana. Proceedings of the National Academy of Sciences of the United States of America 99 (21), 13627–13632. H.M. Smith, B.C. Campbell and S. Hake (2004) Competence to respond to floral inductive signals requires the homeobox genes PENNYWISE and POUND-FOOLISH. Current Biology 14 (9), 812–817. M. Stam, C. Belele, W. Ramakrishna, J.E. Dorweiler, J.L. Bennetzen and V.L. Chandler (2002) The regulatory regions required for B paramutation and expression are located far upstream of the maize b1 transcribed sequences. Genetics 162 (2), 917–930. S.L. Stone, L.W. Kwong, K.M. Yee, et al. (2001) LEAFY COTYLEDON2 encodes a B3 domain transcription factor that induces embryo development. Proceedings of the National Academy of Sciences of the United States of America 98 (20), 11806–11811. R. Stracke, M. Werber and B. Weisshaar (2001) The R2R3-MYB gene family in Arabidopsis thaliana. Current Opinion in Plant Biology 4, 447–456. S. Takada, K. Hibara, T. Ishida and M. Tasaka (2001) The CUP-SHAPED COTYLEDON1 gene of Arabidopsis regulates shoot apical meristem formation. Development 128 (7), 1127– 1135. H. Takatsuji (1999) Zinc-finger proteins: the classical zinc finger emerges in contemporary plant science. Plant Molecular Biology 39 (6), 1073–1078. C. Tian, P. Wan, S. Sun, J. Li and M. Chen (2004) Genome-wide analysis of the GRAS gene family in rice and Arabidopsis. Plant Molecular Biology 54 (4), 519–532. G. Toledo-Ortiz, E. Huq and P.H. Quail (2003) The Arabidopsis basic/helix-loop-helix transcription factor family. The Plant Cell 15 (8), 1749–1770. L. Tyler, S.G. Thomas, J. Hu, et al. (2004) DELLA proteins and gibberellin-regulated seed germination and floral development in Arabidopsis. Plant Physiology 135 (2), 1008–1019. B. Ulker and I.E. Somssich (2004) WRKY transcription factors: from DNA binding towards biological function. Current Opinion in Plant Biology 7 (5), 491–498.
52
REGULATION OF TRANSCRIPTION IN PLANTS
E. van der Knaap, J.H. Kim and H. Kende (2000) A novel gibberellin-induced gene from rice and its potential regulatory role in stem growth. Plant Physiology 122 (3), 695–705. M. Vincentz, C. Bandeira-Kobarg, L. Gauer, P. Schlogl and A. Leite (2003) Evolutionary pattern of angiosperm bZIP factors homologous to the maize Opaque2 regulatory protein. Journal of Molecular Evolution 56 (1), 105–116. C.W. Vroemen, A.P. Mordhorst, C. Albrecht, M.A. Kwaaitaal and S.C. De Vries (2003) The CUPSHAPED COTYLEDON3 gene is required for boundary and shoot meristem formation in Arabidopsis. The Plant Cell 15 (7), 1563–1577. D. Wagner, F. Wellmer, K. Dilks, et al. (2004) Floral induction in tissue culture: a system for the analysis of LEAFY-dependent gene regulation. The Plant Journal 39 (2), 273–282. D. Wang, N.D. Weaver, M. Kesarwani and X. Dong (2005a) Induction of protein secretory pathway is required for systemic acquired resistance. Science 308 (5724), 1036–1040. J.W. Wang, L.J. Wang, Y.B. Mao, W.J. Cai, H.W. Xue and X.Y. Chen (2005a) Control of root cap formation by microRNA-targeted auxin response factors in Arabidopsis. The Plant Cell 17 (8), 2204–2216. D. Weigel, J. Alvarez, D.R. Smyth, M.F. Yanofsky and E.M. Meyerowitz (1992) LEAFY controls floral meristem identity in Arabidopsis. Cell 69 (5), 843–859. D. Weijers, E. Benkova, K.E. Jager, et al. (2005) Developmental specificity of auxin response by pairs of ARF and Aux/IAA transcriptional regulators. The EMBO Journal 24 (10), 1874–1885. F. Wellmer and J.L. Riechmann (2005) Gene network analysis in plant development by genomic technologies. The International Journal of Developmental Biology 49 (5–6), 745–759. F. Wellmer, J.L. Riechmann, M. Alves-Ferreira and E.M. Meyerowitz (2004) Genome-wide analysis of spatial gene expression in Arabidopsis flowers. Plant Cell 16, 1314–1326. C.K. Wen and C. Chang (2002) Arabidopsis RGL1 encodes a negative regulator of gibberellin responses. The Plant Cell 14 (1), 87–100. C.A. Whitelaw, W.B. Barbazuk, G. Pertea, et al. (2003) Enrichment of gene-coding sequences in maize by genome filtration. Science 302 (5653), 2118–2120. D.A. William, Y. Su, M.R. Smith, M. Lu, D.A. Baldwin and D. Wagner (2004) Genomic identification of direct target genes of LEAFY. Proceedings of the National Academy of Sciences of the United States of America 101 (6), 1775–1780. J.C. Wilmoth, S. Wang, S.B. Tiwari, et al. (2005) NPH4/ARF7 and ARF19 promote leaf expansion and auxin-induced lateral root formation. The Plant Journal 43 (1), 118–130. A. Windh¨ovel, I. Hein, R. Dabrowa and J. Stockhaus (2001) Characterization of a novel class of plant homeodomain proteins that bind to the C4 phosphoenolpyruvate carboxylase gene of Flaveria trinervia. Plant Molecular Biology 45 (2), 201–214. G.A. Wray, M.W. Hahn, E. Abouheif, et al. (2003) The evolution of transcriptional regulation in eukaryotes. Molecular Biology and Evolution 20 (9), 1377–1419. K.L. Wu, Z.J. Guo, H.H. Wang and J. Li (2005) The WRKY family of transcription factors in rice and Arabidopsis and their origins. DNA Research 12 (1), 9–26. J.D. Wuitschick, P.R. Lindstrom, A.E. Meyer and K.M. Karrer (2004) Homing endonucleases encoded by germ line-limited genes in Tetrahymena thermophila have APETELA2 DNA binding domains. Eukaryotic Cell 3 (3), 685–694. Z. Xie, Z.L. Zhang, X. Zou, et al. (2005) Annotations and functional analyses of the rice WRKY gene superfamily reveal positive and negative regulators of abscisic acid signaling in aleurone cells. Plant Physiology 137 (1), 176–189. Y. Xiong, T. Liu, C. Tian, S. Sun, J. Li and M. Chen (2005) Transcription factors in rice: a genomewide comparative analysis between monocots and eudicots. Plant Molecular Biology 59 (1), 191– 203. K. Yamasaki, T. Kigawa, M. Inoue, et al. (2004a) A novel zinc-binding motif revealed by solution structures of DNA-binding domains of Arabidopsis SBP-family transcription factors. Journal of Molecular Biology 337 (1), 49–63. K. Yamasaki, T. Kigawa, M. Inoue, et al. (2004b) Solution structure of the B3 DNA binding domain of the Arabidopsis cold-responsive transcription factor RAV1. Plant Cell 16 (12), 3448–3459.
TRANSCRIPTION FACTORS OF ARABIDOPSIS AND RICE
53
K. Yamasaki, T. Kigawa, M. Inoue, et al. (2005a) Solution structure of an Arabidopsis WRKY DNA binding domain. Plant Cell 17, 944–956. K. Yamasaki, T. Kigawa, M. Inoue, et al. (2005b) Solution structure of the major DNA-binding domain of Arabidopsis thaliana ethylene-insensitive3-like3. Journal of Molecular Biology 348 (2), 253–264. S. Yanagisawa (2002) The Dof family of plant transcription factors. Trends in Plant Science 7 (12), 555–560. T. Yang and B.W. Poovaiah (2002) A calmodulin-binding/CGCG box DNA-binding protein family involved in multiple signaling pathways in plants. The Journal of Biological Chemistry 277 (47), 45049–45058. Y. Yin, D. Vafeados, Y. Tao, S. Yoshida, T. Asami and J. Chory (2005) A new class of transcription factors mediates brassinosteroid-regulated gene expression in Arabidopsis. Cell 120 (2), 249–259. D. Yu, C. Chen and Z. Chen (2001) Evidence for an important role of WRKY DNA binding proteins in the regulation of NPR1 gene expression. Plant Cell 13, 1527–1539. J. Yu, S. Hu, J. Wang, et al. (2002) A draft sequence of the rice genome (Oryza sativa L. ssp. indica). Science 296 (5565), 79–92. J. Yu, J. Wang, W. Lin, et al. (2005) The genomes of Oryza sativa: a history of duplications. PLoS Biology 3 (2), e38. L.M. Zahn, H. Kong, J.H. Leebens-Mack, et al. (2005) The evolution of the SEPALLATA subfamily of MADS-box genes: a preangiosperm origin with multiple duplications throughout angiosperm history. Genetics 169 (4), 2209–2223. F. Zhang, A. Gonzalez, M. Zhao, C.T. Payne and A. Lloyd (2003a) A network of redundant bHLH proteins functions in all TTG1-dependent pathways of Arabidopsis. Development 130 (20), 4859– 4869. J.Z. Zhang (2003) Overexpression analysis of plant transcription factors. Current Opinion in Plant Biology 6 (5), 430–440. Y. Zhang, M.J. Tessaro, M. Lassner and X. Li (2003b) Knockout analysis of Arabidopsis transcription factors TGA2, TGA5, and TGA6 reveals their redundant and essential roles in systemic acquired resistance. Plant Cell 15 (11), 2647–2653. Y. Zhang and L. Wang (2005) The WRKY transcription factor superfamily: its origin in eukaryotes and expansion in plants. BMC Evolutionary Biology 5 (1), 1. D.X. Zhou (1999) Regulatory mechanism of plant gene transcription by GT-elements and GT-factors. Trends in Plant Science 4 (6), 210–214. D.X. Zhou, C. Bisanz-Seyer and R. Mache (1995) Molecular cloning of a small DNA binding protein with specificity for a tissue-specific negative element within the rps1 promoter. Nucleic Acids Research 23 (7), 1165–1169. M. Zourelidou, M. De Torres-Zabala, C. Smith and M.W. Bevan (2002) Storekeeper defines a new class of plant-specific DNA-binding proteins and is a putative regulator of patatin expression. The Plant Journal 30 (4), 489–497.
3 Chromatin-associated architectural HMGA and HMGB proteins assist transcription factor function Klaus D. Grasser and Dorte Launholt
3.1
Introduction
Within the cell nucleus, the genomic DNA of eukaryotic organisms is packaged into a highly complex and dynamic nucleoprotein structure known as chromatin. The best studied level of chromatin organisation is the wrapping of the DNA around histone octamers, resulting in the formation of nucleosome particles, which consist of two copies each of the core histones (H2A/B, H3 and H4) and ∼146 bp of DNA. Assisted by linker histones, the nucleosomal arrays fold into higher order chromatin fibres comprising the chromosomes. The compaction of the DNA provided by chromatin represses the transcription of genes by restricting the access of DNA-binding regulatory factors to their DNA target sites, and by inhibiting the progression of RNA polymerases. Since both initiation and elongation of RNA polymerase II transcription are inhibited in the chromatin context in vitro, the dynamic assembly and disassembly of the genomic DNA into chromatin represents an important level of gene regulation. Numerous non-histone proteins regulate the transient changes in chromatin structure that prime the nuclear DNA for gene expression. Thus, post-translational modification of histones (acetylation, methylation, etc.) (see Chapter 4) and ATP-dependent chromatin-remodelling complexes (see Chapter 5) are involved in altering the chromatin structure stimulating activatordependent transcription. The high mobility group (HMG) proteins are (after the histones) the second most abundant family of chromosomal proteins. Their presence in all tissues of eukaryotes favours the possibility that the HMG proteins are required for proper cellular function. Due to their abundance, HMG proteins serve a global structural function in the nucleus, and they act as architectural factors, facilitating various DNA-dependent processes such as transcription and recombination. More than 30 years ago, HMG proteins were identified as contaminants in histone H1 preparations. Originally they were defined based on their extractability from chromatin by 0.35 M NaCl, their solubility in 2% trichloroacetic acid or 5% perchloric acid and their high content of charged amino acid residues (Goodwin et al., 1973). Because of their characteristic primary structures, mammalian HMG proteins have been subdivided into three distinct families (Bustin and Reeves, 1996; Bianchi and Agresti, 2005),
ARCHITECTURAL HMG PROTEINS
55
whose nomenclature has been revised (Bustin, 2001): r HMGA proteins (formerly HMGI/Y) containing AT-hook DNA-binding motifs r HMGB proteins (formerly HMG1/2) containing HMG-box domain(s) r HMGN proteins (formerly HMG14/17) containing a nucleosome-binding domain In plants, proteins belonging to the HMGA and HMGB families have been identified and characterised over the past years (Spiker, 1988; Grasser, 1995), whereas HMGN proteins so far have been found exclusively in vertebrates. The pioneering work on plant HMG proteins by Spiker and co-workers has identified and characterised the HMG proteins from wheat germ, revealing structural differences between the plant and animal HMG proteins (Spiker, 1984, 1988; Spiker and Everett, 1987). Isolation and characterisation of HMG proteins (and of the cDNAs encoding these proteins) from various mono- and dicotyledonous species essentially confirmed the initial findings obtained for the wheat germ proteins (Grasser, 1995).
3.2 3.2.1
HMGA proteins Structure and expression
HMGA proteins have been identified in higher eukaryotes including mammals, insects and plants. Plant HMGA proteins (∼20 kDa) display an overall structure that is different from that of their counterparts in other organisms (Klosterman and Hadwiger, 2002; Grasser, 2003). Thus, plant HMGA proteins contain four copies of the AT-hook DNA-binding motif, while mammalian and insect HMGA have three AT-hooks (Figure 3.1). The AT-hook motif is a short, positively charged sequence containing the invariant GRP core (Figure 3.2), which is usually flanked by proline, arginine and lysine residues (Bustin and Reeves, 1996). The short acidic region found at the C-terminus of mammalian and insect HMGA proteins does not occur in plant HMGA. However, the most obvious structural difference is the plantspecific presence of an N-terminal domain of ∼65 amino acid residues, which shares sequence similarity with the globular domain of H1 linker histones (Figure 3.2). The evolutionary relation of plant HMGA proteins and linker histones is also evident from the presence of a single intron in the plant HMGA genes, which is localised within the H1-like domain. The position of this intron is conserved between HMGA genes and the genes coding for linker histones (Krech et al., 1999). HMGA proteins from various plant species share 40–80% amino acid sequence identity (Gupta et al., 1997b; Yamamoto and Minamikawa, 1997b). In the absence of DNA, the nuclear magnetic resonance spectrum of the AT-hook region of human HMGA indicates random coil, but upon DNA binding, the part of the protein that contacts the DNA becomes ordered and adopts a well-defined conformation. Thereby, the RGR motif
56
REGULATION OF TRANSCRIPTION IN PLANTS
Figure 3.1 Schematic representation of HMGA proteins from various eukaryotes. The AT-hook regions representing the DNA-binding motifs of all HMGA proteins are indicated by black boxes, while the domain similar to the globular domain of linker histone H1 specific for plant HMGA is indicated by a light-grey box and the C-terminal acidic region specific for mammalian and insect HMGA is indicated by a dark-grey box.
Figure 3.2 Alignment of the amino acid sequences of plant HMGA proteins. The alignment contains the sequences of the HMGA proteins deduced from cDNAs of Glycine max (X58246, X58244), Canavalia gladiata (D83070, D83071), Pisum sativum (X89568), Arabidopsis thaliana (X99116), Oryza sativa (L24390), Triticum aestivum (AF502250) and Zea mays (AJ131371, AF291748). The region sharing sequence similarity with the globular domain of linker histone H1 and the GRP center of the AT-hook motifs are highlighted in black.
ARCHITECTURAL HMG PROTEINS
57
of the AT-hooks presents a narrow concave surface that perfectly fits into the minor groove of A/T-rich tracts (Huth et al., 1997). The HMGA proteins of mammals and insects are subject to various posttranslational modifications (Reeves and Beckerbauer, 2001). They are phosphorylated by several protein kinases including CDC2, MAP kinase, CK2 and PKC, and the phosphorylations modulate the nucleosome- and DNA interactions of the HMGA proteins (Bustin, 1999; Banks et al., 2000; Schwanbeck et al., 2001). Accordingly, the phosphorylations are likely to regulate functional properties of HMGA proteins in vivo. Moreover, it has been reported that the HMGA proteins can be modified by multiple acetylations and methylations (Edberg et al., 2004; Sgarra et al., 2004, 2006; Edberg et al., 2005; Miranda et al., 2005; Zou and Wang, 2005). Thus, (mammalian) HMGA proteins are highly modified by post-synthetic modifications, and some of the modifications are tissue specific or dependent on the developmental state (Edberg et al., 2004, 2005; Sgarra et al., 2004). Analogous to the histone code, the existence of a ‘modification code’ for HMGA was suggested that may be expressed differentially, for instance, in normal and malignant cells (Edberg et al., 2004). Comparatively little is currently known about post-synthetic modifications of plant HMGA proteins. In a study examining HMGA from maize endosperm tissue, it was found that HMGA is phosphorylated by the CDC2 kinase (Zhao and Grafi, 2000), suggesting that post-translational modifications contribute also to the regulation of plant HMGA proteins. The expression of mammalian HMGA genes is upregulated in undifferentiated, rapidly proliferating cells or during embryonic development, whereas they are expressed at low levels in fully differentiated or non-dividing cells (Reeves and Beckerbauer, 2001). Some plant species including Arabidopsis and pea have one type of HMGA protein (Gupta et al., 1997a,b), while other species including Canavalia gladiata and soybean encode two different HMGA proteins (Laux et al., 1991; Yamamoto and Minamikawa, 1997b) (cf. Figure 3.2). The plant HMGA genes appear to be expressed ubiquitously (Laux et al., 1991; Gupta et al., 1997a,b, 1998; Yamamoto and Minamikawa, 1997a,b; Zhang et al., 2003b), but the HMGA expression levels vary between different tissues, possibly correlated to the proliferative state of the cells (Gupta et al., 1997b; Zhang et al., 2003b). The level of maize HMGA is elevated during endoreduplication in developing endosperm tissue, suggesting that HMGA is involved in the formation of a more open chromatin configuration facilitating transcription and/or replication (Zhao and Grafi, 2000). In pea, HMGA was found to be expressed at decreased levels upon pathogen infection. Moreover, HMGA can reduce gene expression driven by a defence gene promoter, suggesting that HMGA is involved in the regulation of plant defence gene expression (Klosterman et al., 2003).
3.2.2
DNA and chromatin interactions
Mediated by the AT-hooks, HMGA proteins bind preferentially to the minor groove of A/T-rich stretches of double-stranded DNA (Reeves and Beckerbauer, 2001). Both electrostatic and non-electrostatic components contribute to the DNA binding of HMGA (Huth et al., 1997; Dragan et al., 2003). Depending on the DNA substrate,
58
REGULATION OF TRANSCRIPTION IN PLANTS
binding of mammalian HMGA proteins can bend, straighten, unwind and loop the DNA molecules in vitro (Reeves and Beckerbauer, 2001). The interaction of plant HMGA proteins of various sources with DNA has been examined in several studies using electrophoretic mobility shift assays and DNA footprinting (Pedersen et al., 1991; Nieto-Sotelo et al., 1994; Webster et al., 1997; Yamamoto and Minamikawa, 1997a; Gupta et al., 1997b; Zhang et al., 2003a). In these experiments, the HMGA proteins interact with varying affinities with different A/T-motifs in double-stranded DNA. Random oligonucleotide selection experiments with pea HMGA have revealed that the protein binds to A/T-stretches of five or more base pairs (Webster et al., 1997). Using surface plasmon resonance it has been discovered that HMGA has an approximately sevenfold higher affinity for A/T-rich double-stranded DNA than the structurally unrelated HMGB proteins (Webster et al., 2000). Experiments with truncated HMGA proteins have demonstrated that the individual H1-like domain does not interact with DNA, and that at least two AT-hook motifs are required for efficient DNA binding (Nieto-Sotelo et al., 1994; Zhang et al., 2003a). A detailed DNA-binding analysis has shown that the four AT-hooks of the wheat HMGA protein do not contribute equally to the interaction with a PetE promoter fragment, since the N-terminal AT-hooks play a major role in DNA binding. The regions flanking the AT-hook motifs (cf. Figure 3.2) can modulate the DNA binding that is primarily dictated by the AT-hooks, and therefore account for the distinct DNA binding of the individual AT-hook regions (Zhang et al., 2003a). These differences seen in DNAbinding experiments are in line with structural studies of various AT-hook regions demonstrating that the regions flanking the core of the AT-hooks contact the DNA to different extents (Huth et al., 1997). Mammalian HMGA proteins also interact preferentially with four-way junction DNA, contacting both the branch point, as well as sites on the arms of the DNA substrate. The mode of binding is distinct from that of HMGB proteins and histone H1, although all three protein families share the specific binding to four-way junction DNA, and compete for binding to four-way junctions (Hill et al., 1999). Using electrophoretic mobility shift assays and surface plasmon resonance the specific binding of wheat HMGA to four-way junction has been demonstrated (Zhang et al., 2003a). The interaction of wheat HMGA with four-way junction DNA was compared with that of HMGB proteins and histone H1 for the same substrate, revealing that the relative binding affinity of three types of proteins is in the order of HMGA > H1 > HMGB (Zhang et al., 2003b). The interaction of HMGA with native chromatin has been studied in intact maize nuclei. HMGA could be released from the highly nuclease-sensitive chromatin by limited digestion with micrococcal nuclease, indicating that HMGA primarily associates with the non-compacted, transcriptionally active chromatin. Moreover, HMGA could be selectively released from chromatin by spermidine and distamycin A (Lichota and Grasser, 2001). Distamycin A intercalates into A/T-rich DNA, and can compete for HMGA binding in naked DNA (Webster et al., 1997), suggesting that HMGA binds A/T-tracts both in naked DNA and in nucleosomal templates. Mammalian HMGA proteins interact with nucleosome core particles (containing ∼146 bp of DNA, lacking internucleosomal linker DNA) binding in close proximity to histones H2A/B and H3 (Reeves and Nissen, 1993). Plant HMGA can specifically
ARCHITECTURAL HMG PROTEINS
59
bind to isolated mononucleosome particles in vitro, both core particles (Arwood and Spiker, 1990), as well as nucleosomes containing ∼165 bp of DNA (including linker DNA) (Lichota and Grasser, 2001). HMGA interacts with isolated nucleosomes by a mode that differs from that of the HMGB proteins (Lichota and Grasser, 2001).
3.3 3.3.1
HMGB proteins Structure and expression
HMGB proteins have been identified from various higher plants and they have in common an overall structure that differs from that of HMGB proteins of other sources (Grasser, 1995). In plant HMGB proteins (∼13–27 kDa) the central HMGbox DNA-binding domain is flanked by a basic N-terminal domain and a highly acidic C-terminal domain (Figure 3.3). The HMG-box domain of ∼75 amino acid residues has a conserved L-shaped fold with an angle of ∼80◦ between the arms, consisting mainly of three α-helices (Travers, 2000; Thomas and Travers, 2001). The basic N-terminal and the acidic C-terminal domains are variable in size and amino acid sequence (Figure 3.4), accounting for the differences in molecular mass of the members of plant HMGB protein family (Grasser, 2003). In Arabidopsis, for instance, the N-terminal domain of the HMGB5 protein is composed of 29 amino acid residues, whereas the N-terminal domain of HMGB6 comprises 111 amino acid residues (Grasser et al., 2004). Using intramolecular crosslinking and spectrometric techniques, it has been demonstrated that the terminal domains of plant HMGB proteins can interact, which modulates various functional properties of the proteins (Thomsen et al., 2004). Higher plants express several structurally variable HMGB proteins. Five HMGB proteins have been identified from the monocot plant maize (Stemmer et al., 1999) and six related proteins have been characterised from the dicot plant Arabidopsis (Stemmer et al., 1997; Grasser et al., 2004). A recent study of two additional Arabidopsis HMGB-type proteins has revealed that despite striking sequence
Figure 3.3 Schematic representation of the structure of HMGB proteins from various eukaryotes. HMG-box domains are indicated by dark-grey boxes, while the C-terminal acidic domains found in the plant, insect and vertebrate proteins, but not in the yeast NHP6a/b proteins, are indicated by black boxes. HMG boxes of the ‘single HMG-box proteins’ have a higher degree of similarity to the second HMG box (B-domain) than to the first HMG box (A-domain) of vertebrate HMGB.
60
REGULATION OF TRANSCRIPTION IN PLANTS
Figure 3.4 Alignment of the amino acid sequences of HMGB proteins from Arabidopsis thaliana, Oryza sativa and Zea mays (cf. Table 3.1; for sequence details see Plant Chromatin Database, http://www.chromdb.org/). We have considered only sequences that share the typical overall structure of plant HMGB proteins, containing a central HMG-box domain that is flanked by a basic N-terminal and an acidic C-terminal domain. The potential primary and secondary intercalating residues within the HMG-box domain are indicated in bold, and the relative position of introns within the amino acid sequences are indicated by triangles.
61
ARCHITECTURAL HMG PROTEINS
similarity to the previously characterised HMGB proteins, one of the analysed proteins has properties that are very different from those of typical HMGB proteins. According to analyses of green fluorescent protein (GFP) fusions, this protein (encoded by the AGI locus At5g23405) is mainly localised in the cytoplasm (rather than the nucleus) and it does not bind to DNA (Grasser et al., 2006). Therefore, HMGB proteins that have been predicted solely based on their amino acid sequences should be evaluated experimentally. Currently, comprehensive information about the members of the plant HMGB protein family (from systematic experimental studies and/or genome projects) is available only for few species. We have summarised the characteristics of the HMGB-type proteins encoded by the maize, rice and Arabidopsis genomes in Table 3.1. The amino acid sequences of the same group of proteins were used to construct an amino acid sequence similarity tree that reveals the relationship between the protein sequences (Figure 3.5). There is a close relationship between the monocot HMGB sequences. Thus, there are rice proteins resembling all the HMGB proteins that have been characterised from maize, but instead of the pairs of maize HMGB2/3 and HMGB4/107 proteins (∼90% amino acid sequence identity for each Table 3.1
HMGB proteins of Arabidopsis, rice and maize
Proteina
Length Mass
Acc. no.
Locus
AtHMGB1 AtHMGB2 AtHMGB3 AtHMGB4 AtHMGB5 AtHMGB6 AtHMGB14∗ AtHMGB12∗ ZmHMGB1 ZmHMGB2 ZmHMGB3 ZmHMGB4 ZmHMGB5 ZmHMGB106∗ ZmHMGB107∗ OsHMGB1 OsHMGB710∗ OsHMGB707∗ OsHMGB706∗ OsHMGB705∗ OsHMGB711∗
178 aa 144 aa 141 aa 138 aa 125 aa 241 aa 151 aa 149 aa 157 aa 139 aa 138 aa 126 aa 123 aa 212 aa 127 aa 157 aa 146 aa 131 aa 145 aa 202 aa 133 aa
Y14071 Y14072 Y14073 Y14074 Y14075 AY086023 AY084626 AY052244 X58282 Y08297 Y08298 Y08807 AJ006708 AF527617 AF527616 AY262827 AP005742 XM 473515 AP004858 XM 479646 XM 473515
At3g51880 At1g20693 At1g20696 At2g17560 At4g35570 At5g23420 At2g34450 At5g23405
20265 15982 15681 15364 14203 26964 17481 16997 17146 15316 15007 14104 13637 23534 14215 17100 16349 13873 15765 22328 15566
Structurespecific DNANuclear DNA-bindingb bendingc localisationd +1 +1 +1 +1 +1 +2 +7 −7 +3 +3 +3 +3
Os06g51220 +4 Os09g37910 Os04g47690 Os02g44930 Os08g01100 Os01g47600
+u +u
+2 −7 −7 +3,5 +3,5 +3,5 +3,5
+u +u +2 +7 −7 +6 +6 +6 +6 +6
+4
Published name of protein or ∗ name used in Plant Chromatin Database (http://www.chromdb.org/). Experimental evidence for structure-specific DNA binding to four-way junction DNA, minicircles or supercoiled DNA. c Experimental evidence for DNA bending (circularisation, fluorescence resonance energy transfer (FRET)). d Experimental evidence for nuclear localisation (GFP fusions, cell fractionation). b,c,d References of the functional HMGB protein studies: 1 Stemmer et al., 1997; 2 Grasser et al., 2004; 3 Ritt et al., 1998b; 4 Wu et al., 2003a,b; 5 Wisniewski et al., 1999a; 6 Grasser et al., 1991; 7 Grasser et al., 2006; u our unpublished results.
a
b
62
REGULATION OF TRANSCRIPTION IN PLANTS
Figure 3.5 Amino acid sequence similarity tree of HMGB-type proteins from Arabidopsis thaliana (At), Oryza sativa (Os) and Zea mays (Zm). The amino acid sequences (Plant Chromatin Database, http://www.chromdb.org/) were aligned by multiple sequence alignment that was used to construct the tree, using the ClustalW software. The upper part contains all the sequences related to the previously characterised HMGB proteins (‘traditional HMGBs’), while the lower part contains the HMGB-type sequences (‘novel HMGBs’) related to Arabidopsis HMGB6 (Grasser et al., 2004).
pair) there is only a single rice protein each. The Arabidopsis HMGB proteins (in particular the HMGB4 and HMGB5 proteins) form a rather separate group. The group of ‘novel HMGB proteins’ specified by AtHMGB6 (Grasser et al., 2004) also forms a distinct set of sequences, grouping at the bottom of the tree (Figure 3.5). In general, the HMGB proteins of monocots and dicots (currently just based on Arabidopsis sequences) display to some extent different structural features (with the exception of the Arabidopsis HMGB2/3 proteins, which share similarities with maize and rice HMGB1). It requires further studies to clarify whether the structural differences seen between the maize/rice and Arabidopsis HMGB proteins are reflected by functional differences. Based on the available studies, currently there is no indication of significant functional differences between monocot and dicot HMGB proteins (Grasser, 2003).
ARCHITECTURAL HMG PROTEINS
63
HMGB proteins of mammals are subject to various post-translational modifications including acetylation, methylation, ADP ribosylation and glycosylation (van Holde, 1989). Mammalian HMGB1 can be acetylated at multiple sites (Bonaldi et al., 2003), and acetylation of a lysine residue in the N-terminal region of mammalian HMGB1 can enhance its binding to DNA structures (Ugrinova et al., 2001). HMGB proteins of the insect Chironomus tentans are phosphorylated by protein kinases PKC and CK2, altering the DNA-binding properties of the proteins (Wisniewski et al., 1994, 1999b). Post-translational modification of plant HMGB proteins initially was recognised, since the proteins purified from plant tissues (according to mass spectrometric measurements) displayed masses that were significantly larger than the calculated masses of the proteins (Webster et al., 1997; Stemmer et al., 1999). Further analysis of some maize HMGB proteins revealed that masses of the native proteins could be reduced essentially to the calculated masses by dephosphorylation with alkaline phosphatase (Stemmer et al., 2002b, 2003). Hence, plant HMGB proteins are phosphoproteins in vivo that occur in differentially phosphorylated variants (Stemmer et al., 2003). Mapping of the phosphorylation sites identified serine residues within the acidic C-terminal domains of the analysed proteins, and the same residues could be phosphorylated in vitro by protein kinase CK2α (Stemmer et al., 2002b). Therefore, protein kinase CK2 is involved in the phosphorylation of HMGB proteins of plants (and insects, see above). CK2 is a conserved eukaryotic Ser/Thr kinase that occurs in various isoforms in the plant cell nucleus and cytosol (Riera et al., 2001). CK2 is suspected to play a global role in transcription-related chromatin remodelling (Barz et al., 2003) and is involved in cell growth and proliferation, as well as in stress response and cell survival (Litchfield, 2003; Meggio and Pinna, 2003). Phosphorylation of plant HMGB proteins by CK2 increases protein stability against thermal denaturation (Stemmer et al., 2002b). In line with this finding, the phosphorylation enhances intramolecular interactions between the basic N-terminal and the acidic C-terminal domains of maize HMGB1 (Thomsen et al., 2004). Based on Northern and Western blot analyses, the HMGB genes appear to be expressed ubiquitously in plants (O’Neill and Zheng, 1998; Yamamoto and Minamikawa, 1998; Stemmer et al., 1999; Wu et al., 2003a). The various HMGB proteins that simultaneously occur in the same tissue appear to be of very different abundance. Accordingly, the five maize HMGB proteins are present in markedly different amounts (HMGB1 and HMGB2/3 are ∼20 times more abundant than HMGB4 and HMGB5), and the relative amounts of the proteins differ between the analysed tissues (Stemmer et al., 1999). The situation might be similar in Arabidopsis, as according to the very different number of expressed sequence tag hits in the database (http://www.arabidopsis.org/), the various HMGB genes are expressed at different levels. In the short-day plant Pharbitis nil, the expression of an HMGB gene is regulated by an endogenous circadian rhythm, whereas another HMGB gene is not under the control of photoperiod or an endogenous rhythm, suggesting that some HMGB proteins may be involved in the circadian-regulated gene expression (O’Neill and Zheng, 1998).
64
3.3.2
REGULATION OF TRANSCRIPTION IN PLANTS
DNA and chromatin interactions
In general, the DNA-binding properties of HMGB proteins are characterised by non-sequence-specific interaction with linear double-stranded DNA. However, the proteins specifically recognise various DNA structures, including four-way junctions, DNA minicircles and cis-platinated DNA. Another important feature of the HMGB proteins in their function as architectural factors is their ability upon DNA binding to bend linear DNA by over 90◦ (Travers, 2000; Thomas and Travers, 2001; Agresti and Bianchi, 2003; Grasser, 2003). Structural studies of protein–DNA complexes have revealed a number of details regarding the DNA interaction of nonsequence-specific HMG-box domains (Murphy et al., 1999; Ohndorf et al., 1999; Masse et al., 2002). The concave face of the L-shaped HMG-box domain primarily contacts the minor groove of the DNA, and a hydrophobic wedge is inserted deep into the minor groove. Two residues partially intercalate between base pairs introducing a kink, which contributes significantly to the overall bend (Travers, 2000; Thomas and Travers, 2001). The potential primary intercalating residue of the plant HMGB proteins is a well-conserved phenylalanine residue while in the majority of the proteins the secondary intercalating residue is valine (cf. Figure 3.4). The intercalating residues of HMG-box domains play a critical role in DNA binding and modulating DNA structure by bending and supercoiling (He et al., 2000; Stros and Musel´ıkov´a, 2000; Klass et al., 2003). In various studies it has been demonstrated that plant HMGB proteins bind to certain DNA structures such as DNA minicircles, four-way junctions and supercoiled DNA with higher affinity than that for the corresponding linear DNA (Grasser et al., 1994, 2004; Stemmer et al., 1997; Ritt et al., 1998b; Webster et al., 2001; Wu et al., 2003b; Zhang et al., 2003b). As studied by electrophoretic mobility shift assays, DNA footprinting and surface plasmon resonance, plant HMGB proteins bind to a variety of (A/T-rich) promoter regions (Pedersen et al., 1991; Grasser et al., 1994; Webster et al., 1997, 2000; Wu et al., 2003b). Random oligonucleotide binding site selection experiments have revealed that a pea HMGB protein in principle binds non-sequence-specifically to DNA, but it displays a preference for structurally flexible sites in linear DNA containing deformable dinucleotide steps (Webster et al., 1997). In line with that, plant HMGB proteins upon DNA binding can severely bend the DNA (Table 3.1), as demonstrated by fluorescent resonance energy transfer (Wisniewski et al., 1999a) and circularisation experiments (Grasser et al., 1994; Ritt et al., 1998b; Webster et al., 2001; Wu et al., 2003b). DNA fragments shorter than ∼150 bp (due to the inflexibility of the DNA) cannot be ligated intramolecularly. However, in the presence of a DNA-bending protein, intramolecular ring closure catalysed by DNA ligase can occur. In circularisation assays of this type, in the presence of DNA ligase the HMGB proteins can mediate covalent intramolecular ring closure of short DNA fragments. Maize HMGB1, for instance, could facilitate the formation of DNA minicircles as small as 70 bp, demonstrating its ability to bend the DNA remarkably (Grasser et al., 1994). HMGB proteins from various sources have been shown to bend DNA severely, and the bending activity is an important
ARCHITECTURAL HMG PROTEINS
65
feature of the HMGB proteins in their function as architectural factors (Ritt et al., 1998b; Webster et al., 2001; Wu et al., 2003b). The basic N-terminal domain of maize and rice HMGB1 significantly stimulates the binding to linear DNA, whereas the acidic C-terminal domain has the opposite effect. However, the domains flanking the central HMG-box DNA-binding domain have only little influence on the binding to DNA minicircles, since the interaction with minicircles is an inherent property of the HMG-box domain. Full-length HMGB1 displays DNA-binding properties similar to those of the individual HMGbox domain, suggesting that the terminal domains functionally neutralise each other (Ritt et al., 1998a; Wu et al., 2003b). This finding is in line with the interaction of the basic N-terminal and the acidic C-terminal domains in maize HMGB1, which is facilitated by CK2-mediated phosphorylation of residues within the acidic tail of the protein (Thomsen et al., 2004). Moreover, the phosphorylation negatively influences binding to linear DNA, while the phosphorylation does not affect the affinity for DNA minicircles (Stemmer et al., 2002b). The phosphorylation-induced interaction of the basic N-terminal and the acidic C-terminal domains limits the positive effect of the basic domain on interactions with linear DNA, while the basic region is dispensable for binding DNA minicircles (Ritt et al., 1998a; Wu et al., 2003b). The basic N-terminal domain can support binding to linear DNA by directly interacting with the DNA (Ritt et al., 1998a; Thomsen et al., 2004). Similar to the yeast HMGB protein, NHP6A (Yen et al., 1998; Masse et al., 2002), the HMG-box domain may bind to the minor groove, whereas the basic N-terminal domain wraps over the DNA backbone from the minor groove into the major groove of the DNA, thereby stabilising the protein–DNA interaction. The chromatin association of maize HMGB proteins was examined by releasing the proteins from isolated nuclei, using various reagents. The different HMGB proteins were similarly extracted from chromatin by NaCl, but there were clear differences in the extractability using ethidium bromide and spermine. Ethidium bromide released preferentially HMGB2/3 and HMGB4, whereas spermine released exclusively HMGB1 (Lichota and Grasser, 2001). Differential extraction of HMGB proteins was also observed from wheat and rice chromatin (Arwood et al., 1991; Van den Boeck et al., 1994). The chromatin association was also studied by limited digestion of the chromatin with micrococcal nuclease. HMGB2/3 appears to be associated primarily with the highly nuclease-sensitive chromatin, whereas HMGB1 and especially HMGB4 and HMGB5 were released only after more extensive digestion of the chromatin (Lichota and Grasser, 2001). In summary, these results indicate that the various plant HMGB proteins are differentially associated with chromatin. In addition to DNA binding, specific interactions with histones (and their variants) and other chromosomal proteins are likely to be important determinants of the differential chromatin association of plant HMGB proteins. In line with these results, the HMGB proteins of the dipteran insect Chironomus, termed cHMG1a and cHMG1b, are distinctly distributed in chromosomes (Ghidelli et al., 1997). The cHMG1a protein is uniformly distributed along the chromosome, whereas the cHMG1b protein is specifically localised in chromosomal puffs of interphase giant chromosomes,
66
REGULATION OF TRANSCRIPTION IN PLANTS
indicating a role in transcription. The in vitro interaction of HMGB proteins with nucleosomes has been examined by electrophoretic mobility shift assays, revealing that maize HMGB proteins can bind specifically to isolated mononucleosomes containing DNA of ∼165 bp (Lichota and Grasser, 2001). Since wheat HMGB proteins form rather non-specific complexes with nucleosome core particles (containing ∼146 bp of DNA) (Arwood and Spiker, 1990), the interaction with linker DNA may be a critical feature for the nucleosome binding of plant HMGB proteins. Analysis of the nucleosome interactions of full-length and truncated HMGB1 proteins demonstrated that only the full-length protein can bind to nucleosomes specifically, as both deletion of the basic N-terminal domain and deletion of the acidic C-terminal domain abolish specific nucleosome interactions (Lichota and Grasser, 2001). Most likely, proper nucleosome interaction of the plant HMGB proteins requires simultaneous interactions of the HMG-box domain and the acidic tail with DNA and core histone(s), respectively.
3.4
Dynamic interaction of histone H1 and HMG proteins with chromatin
Mammalian nuclei contain a relatively constant amount of linker histone H1 of almost one molecule of H1 per nucleosome (∼107 molecules), which indicates that the majority of nucleosomes are bound by H1 (van Holde, 1989; Bustin et al., 2005). On the other hand it has been estimated that nuclei contain ∼106 molecules of HMGB and ∼104 molecules of HMGA proteins, and their cellular amounts can differ, depending on the developmental state (van Holde, 1989; Bustin, 1999). H1 and HMGB proteins share several DNA-binding features (see also above) and there also are similarities in their chromatin interactions (Zlatanova and van Holde, 1998; Jerzmanowski et al., 2000). This has led to the proposal that the accessibility of nucleosomal DNA for regulatory proteins can be controlled by displacing H1 by other proteins, and HMG proteins are obvious candidates for this function (Zlatanova et al., 2000). In line with this possibility, H1 and HMGB proteins differentially affect the accessibility of the nucleosomal DNA (Ragab and Travers, 2003). The relatively high expression level of Xenopus and Drosophila HMGB proteins during certain stages of development (when the H1 levels are comparatively low) suggested that HMGB and linker histones share a structural role in organizing chromatin (Ner and Travers, 1994; Nightingale et al., 1996). The inverse regulation of the expression of HMGB proteins and linker histones, however, is a feature that is observed only in few cases and therefore does not explain in general the interaction of HMGB proteins and linker histones. Recent studies have revealed that the bulk of core histones is immobile on DNA in the cell nucleus, but the majority of other DNA-interacting proteins are mobile. Using photobleaching techniques such as fluorescence recovery after photobleaching (FRAP), it has been demonstrated that various nuclear proteins display very different mobility in mammalian cell nuclei (Misteli, 2001; Hager et al., 2002). In contrast to earlier assumptions, H1 is a mobile molecule that interacts with the
ARCHITECTURAL HMG PROTEINS
67
nucleosome only transiently, before it dissociates to bind to another site, and the mobility is modulated by post-translational modifications of H1 (Lever et al., 2000; Misteli et al., 2000; Horn et al., 2002). Previous experiments have revealed that HMGB1 does not stably associate with chromatin in the cell nucleus (Falciola et al., 1997). In line with that, recent FRAP experiments have shown that HMGB and HMGA proteins are highly mobile proteins with a residence time on chromatinbinding sites in the order of seconds, but still the main fraction of the proteins is bound to chromatin at steady state (Catez et al., 2004; Phair et al., 2004). The proteins find their binding sites by three-dimensional scanning of the nucleus, transiently binding to target sites and then rapidly moving on to the next site (Phair et al., 2004). Intriguingly, in this scenario HMGA and HMGB proteins compete with histone H1 for binding sites, weakening H1 binding to chromatin, whereas the HMGs of different families do not compete with each other (Catez et al., 2004). These findings are consistent with earlier studies demonstrating that HMGA could cause transcriptional activation by displacing inhibitory proteins such as linker histone H1 from A/T-rich scaffold attachment regions (SARs, also termed matrix attachment regions, MARs) (Zhao et al., 1993). Maize HMGA could also relieve in vitro the inhibitory effect exerted by H1 on transcription driven by an A/T-rich zein gene promoter, suggesting that transcription may be controlled to some extent by the interplay of HMGA and H1 (Zhao and Grafi, 2000). A/T-rich sequences can act in general as quantitative, non-tissue-specific enhancers of plant gene expression, and binding of HMGA to these sequences may play a critical role (Sandhu et al., 1998). By reducing the binding of H1, HMG proteins may induce a less compact chromatin structure facilitating the access of regulatory factors to their nucleosomal target sites. The chromatin structure and accessibility of the genomic DNA packaged into nucleosomes are constantly modulated by a dynamic and competitive interaction network of chromosomal proteins (Catez et al., 2004; Phair et al., 2004).
3.5
HMGA and HMGB proteins as architectural assistant factors
Various analyses studying altered expression levels of HMGA and HMGB genes have demonstrated that the chromosomal HMG proteins have important cellular roles. Knockout of the mouse HMGA1 gene revealed that HMGA1 is required for normal sperm development (Liu et al., 2003), and that disrupting the HMGA1 gene results in a decreased expression of the insulin receptor, largely impairing insulin signalling (Foti et al., 2005). Inactivation of the HMGA2 gene causes the mouse pygmy phenotype, and a deficiency in fat tissue, probably due to affected fat-cell proliferation (Zhou et al., 1995; Anand and Chada, 2000). The lack of HMGB1 causes pleiotropic defects in mice and they die soon after birth, but cell lines can grow normally without HMGB1 (Calogero et al., 1999). Mice lacking HMGB2 (which is ∼80% identical to HMGB1) are viable, but male mice have reduced fertility, as HMGB2 seems to play a role in germ cell differentiation (Ronfani et al., 2001). Reduction of the amount of the HMGB-type proteins HMGD/Z in Drosophila causes only minor morphological defects. HMGD/Z was found to interact genetically with the Brahma
68
REGULATION OF TRANSCRIPTION IN PLANTS
chromatin-remodelling complex (Ragab et al., 2006). Functional interaction with chromatin-remodelling complexes also has been observed for mammalian HMGB1, stimulating nucleosome sliding (Bonaldi et al., 2002). Ectopic expression of maize HMGB1 in tobacco seedlings resulted in reduced length of the primary root, whereas the shoot of the seedlings had wild-type appearance. This ‘short-root’ phenotype of the transgenic lines correlated with a decreased size of the cells in the cell division zone of the root. In these plants, the cell division rate rather than the cell elongation appears to be affected (Lichota et al., 2004). In yeast, knockout of one of the two NHP6A/B genes (encoding HMGB proteins) did not result in a phenotype distinct from the wild type, but the inactivation of both genes led to growth aberrations such as temperature-sensitive growth and various morphological defects (Costigan et al., 1994). Analyses of the gene expression in the strain lacking both NHP6A and NHP6B revealed that the induction of transcription and the expression levels of a variety of genes are altered relative to the control strain (Paull et al., 1996; Moreira and Holmberg, 2000). The expression of the majority of genes is not affected, but certain genes are upregulated, whereas other genes are downregulated in the doublemutant strain, presumably depending on whether the architectural factors assist the formation of stimulatory or repressive complexes. A well-established function of the HMGA and HMGB proteins is their architectural function in the proper three-dimensional assembly of nucleoprotein complexes regulating transcription (Reeves and Beckerbauer, 2001; Agresti and Bianchi, 2003; Grasser, 2003). The formation of these complexes is primarily under the control of DNA-binding factors, which recognise specifically cis-acting DNA target sequences. In eukaryotes, usually various regulators bound to different DNA sequence elements interact to form higher order nucleoprotein structures in which multiple protein–protein and protein–DNA contacts increase the specificity and stability of the regulatory complexes (combinatorial control). Dependent on the developmental stage and tissue type, or in response to environmental signals, specific sets of genes within the eukaryotic genomes are activated or repressed. This is achieved by assembling regulatory complexes containing various transcription factors bound to their cognate promoter/enhancer elements (Tjian and Maniatis, 1994; Singh, 1998; Merika and Thanos, 2001). Accordingly, activation or repression of a gene requires that all components of the final complex are simultaneously available. The correct assembly of such DNA-bound (multiprotein) complexes often is assisted by architectural factors such as HMGA and HMGB proteins that can modulate DNA structure and/or contribute to the protein interactions within the complex. A proteomic approach identified novel protein interaction partners of HMGA including proteins related to chromatin remodelling and mRNA processing, which indicates that HMGA may serve (in addition to its role in transcript initiation) other functions in gene expression (Sgarra et al., 2005). The human virus-inducible IFN-β gene can serve as a model illustrating the role of HMGA in the regulation of gene expression (Merika and Thanos, 2001). The assembly of the multiprotein complex at the enhancer region of the IFN-β gene consisting of several transcription factors is assisted by DNA interactions of HMGA, which unbend an unfavourable intrinsic DNA curvature. Moreover, the
ARCHITECTURAL HMG PROTEINS
69
final complex is stabilised by a complex network of mutual protein–protein interactions between activators and HMGA, resulting in a remarkable stable nucleoprotein structure (Yie et al., 1999). Dynamic control of IFN-β gene expression requires the regulated assembly and disassembly of the nucleoprotein structure, which is accomplished by differential acetylation of HMGA. Acetylation of HMGA at Lys65 by the acetyltransferase CBP destabilises the complex, whereas acetylation at Lys71 by PCAF/GCN5 potentiates transcription by stabilising the enhanceosome. Thus, the regulated acetylation of HMGA coordinates the transcriptional switch by causing either enhanceosome stabilisation or destabilisation (Munshi et al., 2001). Activation of the IFN-β gene is also accompanied by chromatin rearrangements in the promoter region allowing transcription initiation (Lomvardas and Thanos, 2002). There are several examples of an involvement of mammalian HMGA proteins in the formation of productive complexes initiating gene transcription (Reeves and Beckerbauer, 2001; Bianchi and Agresti, 2005). The role of HMGA in the regulation of transcription is to facilitate the DNA binding of transcription factors by inducing allosteric effects on DNA and/or by direct protein–protein interactions with the sequence-specific factors (Figure 3.6). Depending on the biological context and the developmental stage, HMGA proteins can regulate gene expression positively or negatively (Reeves and Beckerbauer, 2001; Bianchi and Agresti, 2005). In
Figure 3.6 Architectural function of HMGA and HMGB proteins in the formation of nucleoprotein structures (adapted from Grasser, 2003). Both HMGA and HMGB proved to be versatile and functionally flexible proteins presumably involved in a variety of DNA-dependent processes including transcriptional initiation. Due to their abundance and mobility, HMG proteins bind the DNA randomly, eventually resulting in the formation of productive complexes. (A) Sequence-specific transcription factors may bind their target sites creating a transient DNA bend, which represents a high-affinity binding site for HMG proteins. The DNA bending activity of the recruited HMG protein can stabilise the final complex (without requirement of protein–protein interactions between the HMG protein and the transcription factor). (B) Alternatively, HMG proteins are specifically recruited by protein–protein interactions with sequence-specific regulators. Free or DNA-bound regulators can recruit the architectural HMGA or HMGB proteins. There may be different orders and mechanisms for HMG protein recruitment, and HMGs can interact only transiently to establish the complex, or they can be stable components of the assembly.
70
REGULATION OF TRANSCRIPTION IN PLANTS
electrophoretic mobility shift assays, the rice HMGA protein (termed PF1) could enhance (approximately tenfold) the binding of the transcriptional activator GT-2 to the phyA gene promoter. Since no direct physical contacts could be detected between HMGA and GT-2, the stimulation of GT-2 DNA binding was attributed to a preconditioning of the DNA target site by HMGA (Martinez-Garcia and Quail, 1999). In collaboration with transcription factors, the HMGB proteins contribute to the regulation of transcription initiation. The HMGB proteins can influence the formation of productive regulatory complexes at promoter or enhancer regions in two ways (Figure 3.6): (i) by direct protein interaction with transcription factor(s) facilitating their binding to DNA target sites, and/or (ii) by folding DNA target site (without protein interaction with transcription factors) into a suitable conformation stimulating the formation of the regulatory complex (Agresti and Bianchi, 2003; Grasser, 2003). The latter mechanism has been studied in detail using bacterial sitespecific recombination reactions as model systems, since HMGB proteins (without directly interacting with the recombinases) stimulate the assembly of productive synaptic complexes in these recombination reactions (Grosschedl, 1995). The correct three-dimensional assembly of the recombination complexes is facilitated by the DNA-bending activity of HMGB proteins. DNA bending is the primary determinant of the stimulatory effect, since HMGB proteins could substitute for other, even structurally unrelated bending proteins such as bacterial HU or IHF in promoting the recombination reactions (Grosschedl, 1995). Maize HMGB proteins can facilitate the site-specific β-recombination reaction in vitro and in vivo. Depending on the distance of the recombinase-binding sites on the DNA, the various maize HMGB proteins display different efficiency in stimulating β-recombination, indicating that the members of the HMGB family have different abilities to support the assembly of nucleoprotein structures (Stemmer et al., 2002a). Analogous to site-specific recombination, HMGB can be recruited to promoter regions without physical interactions with the transcription factors. In the case of the Epstein–Barr virus BHLF-1 promoter, ZEBRA transcription factors bind and bend the target DNA weakly, and this bent DNA represents a high-affinity binding site for HMGB1. Thereby, HMGB1 is recruited to the promoter and stabilises the final complex (Ellwood et al., 2000). At the BHLF-1 enhancer, HMGB1 displays a different mode of action. HMGB1 binds the target DNA only transiently to establish binding of two molecules of the viral activator Rta by a chaperone mechanism, and the DNA-bending activity of HMGB1 is the primary determinant of the stimulatory effect (Mitsouras et al., 2002). HMGB-induced DNA bending plays an important role also in the functional interaction of HMGB1 and p53. HMGB1 stimulates the binding of p53 to linear DNA, but not to pre-bent DNA (McKinney and Prives, 2002). In many instances, the HMGB proteins interact directly with transcription factors, stimulating the binding of the transcription factors to their specific DNA sites. HMGB proteins have been found to associate with a variety of transcription factors including members of the Rel, HOX, octamer and SREBP families as well as TFIID/TFIIA and steroid hormone receptors (Shykind et al., 1995; Zwilling et al., 1995; Zappavigna et al., 1996; Boonyaratanakornkit et al., 1998; Decoville et al., 2000; Najima et al., 2005). The protein–protein interactions with transcription factors explain how the
ARCHITECTURAL HMG PROTEINS
71
non-sequence-specific HMGB proteins in these cases are recruited to their sites of action (Bustin, 1999; Thomas and Travers, 2001; Agresti and Bianchi, 2003; Grasser, 2003). Despite the interaction with certain transcription factors, the HMGB proteins are not always detectable in the final ternary complex with DNA. Therefore, depending on the promoter/enhancer geometry and on the transcription factors involved, HMGB proteins may leave the assembly after initiating the formation of the transcription factor–DNA complex. In line with the dynamic movement of HMG proteins and other factors (see above) in the cell nucleus, a study of the interaction of HMGB1 and the glucocorticoid receptor (GR) revealed intriguing novel perspectives. HMGB1 and GR were found to interact only in the chromatin context, but not in the nucleoplasm, and the two partners decrease each other’s mobility favouring the formation of a productive complex at GR-binding sites (Agresti et al., 2005). In transient co-transformation experiments, maize HMGB1 can stimulate reporter gene expression in protoplasts (Grasser et al., 1993), and there are also a few examples of functional interactions of HMGB proteins and plant transcription factors. Thus, a wheat HMGB protein (termed HMGb, which is related to maize and rice HMGB1) can stimulate the binding of the bZIP transcription factor EmBP-1 to its DNA target site, whereas other tested HMGB proteins are not effective (Schultz et al., 1996). Maize HMGB1 interacts with the zinc-finger transcription factors Dof1 and Dof2 (through their Dof DNA-binding domain) and facilitates Dof DNA binding (Yanagisawa, 1997). The HMG-box domain of HMGB1 mediates the interaction and the individual domain is sufficient for stimulating Dof2 DNA binding. Although all maize HMGB proteins can cooperate with Dof2, they do so with different efficacy (Krohn et al., 2002). HMGB5 is clearly most effective and can stimulate the binding of maize Dof2 to its target site >30 fold in naked DNA. Maize HMGB5 also assists Dof2 DNA binding, when the Dof2 target site is assembled into nucleosome particles (Cavalar et al., 2003). Moreover, phosphorylation of HMGB1 and HMGB2/3 by protein kinase CK2 (see above) abolishes the functional interaction with Dof2 (Krohn et al., 2002). The widespread expression, the abundance and relative evolutionary conservation of the HMGA and HMGB proteins in (higher) eukaryotes indicate that they serve important cellular functions. This view is supported by studies of organisms with increased or decreased HMGA and HMGB levels displaying developmental defects and altered gene expression. Analyses of the role of HMGA and HMGB proteins in the formation of higher order nucleoprotein structures indicate that the HMGA and HMGB proteins as architectural factors assist specific transcriptional regulators, and contribute to proper control of transcript initiation. Upon stimulation through signal transduction pathways, the HMGA and HMGB proteins cooperate with sequencespecific transcription factors to form regulatory complexes, in which the synergy of multiple protein–DNA and protein–protein interactions provides the precision for the proper control of gene transcription in complex eukaryotic genomes. In contrast to the models involving an ordered stepwise assembly of regulatory complexes, future studies have to take into consideration recent findings demonstrating that HMG proteins and many other chromatin proteins are highly dynamic molecules in the cell nucleus (Bianchi and Agresti, 2005; Bustin et al., 2005; Gorski and Misteli, 2005).
72
REGULATION OF TRANSCRIPTION IN PLANTS
Acknowledgements Research in our laboratory has been supported primarily by grants from the Danish Research Council, and Dorte Launholt is recipient of a DBRA stipend.
References A. Agresti and M.E. Bianchi (2003) HMGB proteins and gene expression. Current Opinion in Genetics and Development 13, 170–178. A. Agresti, P. Scaffidi, A. Riva, V.R. Caiolfa and M.E. Bianchi (2005) GR and HMGB1 interact only within chromatin and influence each other’s residence time. Molecular Cell 18, 109–121. A. Anand and K. Chada (2000) In vivo modulation of Hmgic reduces obesity. Nature Genetics 24, 377–380. L.J. Arwood, E. Hill and S. Spiker (1991) Ethidium bromide preferentially releases high mobility group chromosomal proteins HMGc and HMGd from wheat embryo nuclei. Plant Physiology 82, 419–422. L.J. Arwood and S. Spiker (1990) Binding of wheat and chicken high mobility group chromosomal proteins to DNA and to wheat and chicken mononucleosomes. The Journal of Biological Chemistry 265, 9771–9777. G.C. Banks, Y. Li and R. Reeves (2000) Differential in vivo modifications of the HMGI(Y) nonhistone chromatin proteins modulate nucleosome and DNA interactions. Biochemistry 39, 8333–8346. T. Barz, K. Ackermann, G. Dubois, R. Eils and W. Pyerin (2003) Genome-wide expression screens indicate a global role for protein kinase CK2 in chromatin remodeling. Journal of Cell Science 116, 1563–1577. M.E. Bianchi and A. Agresti (2005) HMG proteins: dynamic players in gene regulation and differentiation. Current Opinion in Genetics and Development 15, 496–506. T. Bonaldi, G. L¨angst, R. Strohner, P.B. Becker and M.E. Bianchi (2002) The DNA chaperone HMGB1 facilitates ACF/CHRAC-dependent nucleosome sliding. The EMBO Journal 21, 6865–6873. T. Bonaldi, F. Talamo, P. Scaffidi, et al. (2003) Monocytic cells hyperacetylate chromatin protein HMGB1 to redirect it towards secretion. The EMBO Journal 22, 5551–5560. V. Boonyaratanakornkit, V. Melvin, P. Prendergast, et al. (1998) High-mobility group chromatin proteins 1 and 2 functionally interact with steroid hormone receptors to enhance their DNA binding in vitro and transcriptional activity in mammalian cells. Molecular and Cellular Biology 18, 4471–4487. M. Bustin (1999) Regulation of DNA-dependent activities by the functional motifs of the highmobility-group chromosomal proteins. Molecular and Cellular Biology 19, 5237–5246. M. Bustin (2001) Revised nomenclature for high mobility group (HMG) chromosomal proteins. Trends in Biochemical Sciences 26, 152–153. M. Bustin, F. Catez and J.-H. Lim (2005) The dynamics of histone H1 function in chromatin. Molecular Cell 17, 617–620. M. Bustin and R. Reeves (1996) High-mobility-group chromosomal proteins: architectural components that facilitate chromatin function. Progress in Nucleic Acids Research 54, 35–100. S. Calogero, F. Grassi, A. Aguzzi, et al. (1999) The lack of chromosomal protein HMG1 does not disrupt cell growth but causes lethal hypoglycaemia in newborn mice. Nature Genetics 22, 276– 280. F. Catez, H. Yang, K.J. Tracey, R. Reeves, T. Misteli and M. Bustin (2004) Network of dynamic interactions between histone H1 and high-mobility-group proteins in chromatin. Molecular and Cellular Biology 24, 4321–4328. M. Cavalar, C. M¨oller, N.M. Krohn, K.D. Grasser and C. Peterh¨ansel (2003) The interaction of DOF transcription factors with nucleosomes depends on the positioning of the binding site and is facilitated by maize HMGB5. Biochemistry 42, 2149–2157.
ARCHITECTURAL HMG PROTEINS
73
C. Costigan, D. Kolodrubetz and M. Snyder (1994) NHP6A and NHP6B, which encode HMG1-like proteins, are candidates for downstream components of the yeast SLT2 mitogen-activated protein kinase pathway. Molecular and Cellular Biology 14, 2391–2403. M. Decoville, M.J. Giraud-Panis, C. Mosrin-Huaman, M. Leng and D. Locker (2000) HMG boxes of DSP1 protein interact with the Rel homology domain of transcription factors. Nucleic Acids Research 28, 454–462. A.I. Dragan, J.R. Liggins, C. Crane-Robinson and P.L. Privalov (2003) The energetics of specific binding of AT-hooks from HMGA1 to target DNA. Journal of Molecular Biology 327, 393– 411. D.D. Edberg, J.N. Adkins, D.L. Springer and R. Reeves (2005) Dynamic and differential in vivo modifications of the isoform HMGA1a and HMGA1b chromatin proteins. The Journal of Biological Chemistry 280, 8961–8973. D.D. Edberg, J.E. Bruce, W.F. Siems and R. Reeves (2004) In vivo posttranslational modifications of the high mobility group A1a proteins in breast cancer cells of differing metastatic potential. Biochemistry 43, 11500–11515. K.B. Ellwood, Y.-M. Yen, R.C. Johnson and M. Carey (2000) Mechanism for specificity by HMG-1 in enhanceosome assembly. Molecular and Cellular Biology 20, 4359–4370. L. Falciola, F. Spada, S. Calogero, et al. (1997) High mobility group 1 protein is not stably associated with the chromosomes of somatic cells. The Journal of Cell Biology 137, 19–26. D. Foti, E. Chiefari, M. Fedele, et al. (2005) Lack of the architectural factor HMGA1 causes insulin resistance and diabetes in humans and mice. Nature Medicine 11, 765–773. S. Ghidelli, P. Claus, G. Thies and J.R. Wisniewski (1997) High mobility group proteins cHMG1a, cHMG1b and cHMGI are distinctly distributed in chromosomes and differentially expressed during ecdysone dependent cell differentiation. Chromosoma 105, 369–379. G.H. Goodwin, C. Sanders and E.W. Johns (1973) A new group of chromatin-associated proteins with a high content of acidic and basic amino acids. European Journal of Biochemistry 38, 14–19. S. Gorski and T. Misteli (2005) Systems biology in the cell nucleus. Journal of Cell Science 118, 4083–4092. K.D. Grasser (1995) Plant chromosomal high mobility group (HMG) proteins. The Plant Journal 7, 185–192. K.D. Grasser (2003) Chromatin-associated HMGA and HMGB proteins: versatile co-regulators of DNA-dependent processes. Plant Molecular Biology 53, 281–295. K.D. Grasser, S. Grill, M. Duroux, et al. (2004) HMGB6 from Arabidopsis thaliana specifies a novel type of plant chromosomal HMGB protein. Biochemistry 43, 1309–1314. K.D. Grasser, W. Hetz, E.A. Griess and G. Feix (1993) Stimulatory effect of the maize HMGa protein on reporter gene expression in maize protoplasts. FEBS Letters 327, 141–144. K.D. Grasser, A.B. Krech and G. Feix (1994) The maize chromosomal HMGa protein recognises structural features of DNA and increases DNA flexibility. The Plant Journal 6, 351–358. K.D. Grasser, A. Wurz and G. Feix (1991) Isolation and characterisation of high-mobility-group proteins from maize. Planta 185, 350–355. M. Grasser, A. Lentz, J. Lichota, T. Merkle and K.D. Grasser (2006) The Arabidopsis genome encodes structurally and functionally diverse HMGB-type proteins. Journal of Molecular Biology 358, 654–664. R. Grosschedl (1995) Higher-order nucleoprotein complexes in transcription: analogies with sitespecific recombination. Current Opinion in Cell Biology 7, 362–370. R. Gupta, C.I. Webster and J.C. Gray (1997a) The single-copy gene encoding high-mobility-group protein HMG-I/Y from pea contains a single intron and is expressed in all organs. Plant Molecular Biology 35, 987–992. R. Gupta, C.I. Webster and J.C. Gray (1998) Characterisation and promoter analysis of the Arabidopsis gene encoding high-mobility-group protein HMG-I/Y. Plant Molecular Biology 36, 897–907. R. Gupta, C.I. Webster, A. Walker and J.C. Gray (1997b) Chromosomal location and expression of the single-copy gene encoding high-mobility-group protein HMG-I/Y in Arabidopsis thaliana. Plant Molecular Biology 34, 529–536.
74
REGULATION OF TRANSCRIPTION IN PLANTS
G.L. Hager, C. Elbi and M. Becker (2002) Protein dynamics in the nuclear compartment. Current Opinion in Genetics and Development 12, 137–141. Q. He, U.-M. Ohndorf and S.J. Lippard (2000) Intercalating residues determine the mode of HMG1 domains A and B binding to cisplatin-modified DNA. Biochemistry 39, 14426–14435. D.A. Hill, M.L. Pedulla and R. Reeves (1999) Directional binding of HMG-I(Y) on four-way junction DNA and the molecular basis for competitive binding with HMG-1 and histone H1. Nucleic Acids Research 27, 2135–2144. P.J. Horn, L.M. Carruthers, C. Logie, et al. (2002) Phosphorylation of linker histones regulates ATPdependent chromatin remodeling enzymes. Nature Structural Biology 9, 263–267. J.R. Huth, C.A. Bewley, M.S. Nissen, et al. (1997) The solution structure of an HMG-I(Y)-DNA complex defines a new architectural minor groove binding motif. Nature Structural Biology 4, 657–665. A. Jerzmanowski, M. Przewloka and K.D. Grasser (2000) Linker histones and HMG1 proteins of higher plants. Plant Biology 2, 586–597. J. Klass, F.V. Murphy, S. Fouts, et al. (2003) The role of intercalating residues in chromosomal high-mobility-group protein DNA binding, bending and specificity. Nucleic Acids Research 31, 2852–2864. S.J. Klosterman, J.J. Choi and L.A. Hadwiger (2003) Analysis of pea HMG-I/Y expression suggests a role in defence gene regulation. Molecular Plant Pathology 4, 249–258. S.J. Klosterman and L.A. Hadwiger (2002) Plant HMG proteins bearing the AT-hook motif. Plant Science 162, 855–866. A.B. Krech, D. Wulff, K.D. Grasser and G. Feix (1999) Plant chromosomal HMGI/Y proteins and histone H1 exhibit a protein domain of common origin. Gene 230, 1–5. N.M. Krohn, S. Yanagisawa and K.D. Grasser (2002) Specificity of the stimulatory interaction between chromosomal HMGB proteins and the transcription factor Dof2 and its negative regulation by protein kinase CK2-mediated phosphorylation. The Journal of Biological Chemistry 277, 32438– 32444. T. Laux, J. Seurinck and R.B. Goldberg (1991) A soybean embryo cDNA encodes a DNA-binding protein with histone and HMG protein-like domains. Nucleic Acids Research 19, 4768. M.A. Lever, J.P. Th’ng, X. Sun and M.J. Hendzel (2000) Rapid exchange of histone 1.1 on chromatin in living human cells. Nature 408, 873–876. J. Lichota and K.D. Grasser (2001) Differential chromatin association and nucleosome binding of the maize HMGA, HMGB, and SSRP1 proteins. Biochemistry 40, 7860–7867. J. Lichota, C. Ritt and K.D. Grasser (2004) Ectopic expression of the maize chromosomal HMGB1 protein causes defects in root development in tobacco seedlings. Biochemical and Biophysical Research Communications 318, 317–322. D.W. Litchfield (2003) Protein kinase CK2: structure, regulation and role in cellular decisions of life and death. The Biochemical Journal 369, 1–15. J. Liu, J.F. Schiltz, H.R. Ashar and K. Chada (2003) HMGa1 is required for normal sperm development. Molecular Reproduction and Development 66, 81–89. S. Lomvardas and D. Thanos (2002) Modifying gene expression programs by altering core promoter chromatin architecture. Cell 110, 261–271. J.F. Martinez-Garcia and P.H. Quail (1999) The HMG-I/Y protein PF1 stimulates binding of the transcriptional activator GT-2 to the PHYA gene promoter. The Plant Journal 18, 173–183. J.E. Masse, B. Wong, Y.-M. Yen, F.H.T. Allain, R.C. Johnson and J. Feignon (2002) The S. cerevisiae architectural HMGB protein NHP6A complexed with DNA: DNA and protein conformational changes upon binding. Journal of Molecular Biology 323, 263–284. K. McKinney and C. Prives (2002) Efficient specific DNA binding by p53 requires both its central and C-terminal domains as revealed by studies with high-mobility-group 1 protein. Molecular and Cellular Biology 22, 6797–6808. F. Meggio and L.A. Pinna (2003) One-thousand-and-one substrates of protein kinase CK2? The FASEB Journal 17, 349–368. M. Merika and D. Thanos (2001) Enhanceosomes. Current Opinion in Genetics and Development 11, 205–208.
ARCHITECTURAL HMG PROTEINS
75
T.B. Miranda, K.J. Webb, D.D. Edberg, R. Reeves and S. Clark (2005) Protein arginine methyltransferase 6 specifically methylates the nonhistone chromatin protein HMGA1a. Biochemical and Biophysical Research Communications 336, 831–835. T. Misteli (2001) Protein dynamics: implications for nuclear architecture and gene expression. Science 291, 843–847. T. Misteli, A. Gunjan, R. Hock, M. Bustin and D.T. Brown (2000) Dynamic binding of histone H1 to chromatin in living cells. Nature 408, 877–881. K. Mitsouras, B. Wong, C. Arayata, R.C. Johnson and M. Carey (2002) The DNA architectural protein HMGB1 displays two distinct modes of action that promote enhanceosome assembly. Molecular and Cellular Biology 22, 4390–4401. J.M.A. Moreira and S. Holmberg (2000) Chromatin-mediated transcriptional regulation by the yeast architectural factors NHP6A and NHP6B. The EMBO Journal 19, 6804–6813. N. Munshi, T. Agalioti, S. Lomvardas, M. Merika, G. Chen and D. Thanos (2001) Coordination of a transcriptional switch by HMGI(Y) acetylation. Science 293, 1133–1136. F.V. Murphy, R.M. Sweet and M.E.A. Churchill (1999) The structure of a chromosomal high mobility group protein–DNA complex reveals sequence-neutral mechanisms important for non-sequencespecific DNA recognition. The EMBO Journal 18, 6610–6618. Y. Najima, N. Yahagi, Y. Takeuchi, et al. (2005) High mobility group protein-B1 interacts with sterol regulatory element-binding proteins to enhance their DNA binding. The Journal of Biological Chemistry 280, 27523–27532. S.S. Ner and A.A. Travers (1994) HMG-D, the Drosophila melanogaster homologue of HMG1 protein, is associated with early embryonic chromatin in the absence of histone H1. The EMBO Journal 13, 1817–1822. J. Nieto-Sotelo, A. Ichida and P.H. Quail (1994) PF1: an A-T hook containing DNA binding protein from rice that interacts with a functionally defined d(AT)-rich element in the oat phytochrome A3 gene promoter. The Plant Cell 6, 287–301. K. Nightingale, S. Dimitrov, R. Reeves and A.P. Wolffe (1996) Evidence for a shared structural role for HMG1 and linker histones B4 and H1 in organising chromatin. The EMBO Journal 15, 548–561. U.-M. Ohndorf, M.A. Rould, Q. He, C.O. Pabo and S.J. Lippard (1999) Basis for recognition of cisplatin-modified DNA by high-mobility-group proteins. Nature 399, 708–712. S.D. O’Neill and C.C. Zheng (1998) Abundance of mRNAs encoding HMG1/HMG2 class highmobility-group DNA-binding proteins are differentially regulated in cotyledons of Pharbitis nil. Plant Molecular Biology 37, 235–241. T.T. Paull, M. Carey and R.C. Johnson (1996) Yeast HMG proteins NHP6A/B potentiate promoterspecific transcriptional activation in vivo and assembly of preinitiation complexes in vitro. Genes and Development 10, 2769–2781. T.J. Pedersen, L.J. Arwood, S. Spiker, M.J. Guiltinan and W.F. Thompson (1991) High mobility group proteins bind to AT-rich tracts flanking plant genes. Plant Molecular Biology 16, 95–104. R.D. Phair, P. Scaffidi, C. Elbi, et al. (2004) Global nature of dynamic protein–chromatin interactions in vivo: three-dimensional genome scanning and dynamic interaction networks of chromatin proteins. Molecular and Cellular Biology 24, 6393–6402. A. Ragab, E.C. Thompson and A.A. Travers (2006) HMGD and HMGZ interact genetically with the Brahma chromatin remodelling complex in Drosophila. Genetics 172, 1069–1078. A. Ragab and A. Travers (2003) HMG-D and histone H1 alter the local accessibility of nucleosomal DNA. Nucleic Acids Research 31, 7083–7089. R. Reeves and L. Beckerbauer (2001) HMGI/Y proteins: flexible regulators of transcription and chromatin structure. Biochimica et Biophysica Acta 1519, 13–29. R. Reeves and M.S. Nissen (1993) Interaction of high mobility group-I(Y) nonhistone proteins with nucleosome core particles. The Journal of Biological Chemistry 268, 21137–21146. M. Riera, G. Peracchia and M. Pag`es (2001) Distinctive features of plant protein kinase CK2. Molecular and Cellular Biochemistry 227, 119–127. C. Ritt, R. Grimm, S. Fern´andez, J.C. Alonso and K.D. Grasser (1998a) Basic and acidic regions flanking the HMG domain of maize HMGa modulate the interactions with DNA and the selfassociation of the protein. Biochemistry 37, 2673–2681.
76
REGULATION OF TRANSCRIPTION IN PLANTS
C. Ritt, R. Grimm, S. Fern´andez, J.C. Alonso and K.D. Grasser (1998b) Four differently chromatinassociated maize HMG domain proteins modulate DNA structure and act as architectural elements in nucleoprotein complexes. The Plant Journal 14, 623–631. L. Ronfani, M. Ferraguti, L. Croci, et al. (2001) Reduced fertility and spermatogenesis defects in mice lacking chromosomal protein Hmgb2. Development 128, 1265–1273. J.S. Sandhu, C.I. Webster and J.C. Gray (1998) A/T-rich sequences act as quantitative enhancers of gene expression in transgenic tobacco and potato plants. Plant Molecular Biology 37, 885–896. T.F. Schultz, S. Spiker and R.S. Quatrano (1996) Histone H1 enhances the DNA binding activity of the transcription factor EmBP-1. The Journal of Biological Chemistry 271, 25742–25745. R. Schwanbeck, M. Gymnopoulos, I. Petry, et al. (2001) Consecutive steps of phosphorylation affect confirmation and DNA binding of the Chironomus high mobility group A protein. The Journal of Biological Chemistry 276, 26012–26021. R. Sgarra, J. Lee, M.A. Tessari, et al. (2006) The AT-hook of the chromatin architectural transcription factor HMGA1a is arginine methylated by PRMT6. The Journal of Biological Chemistry 281, 3764–3772. R. Sgarra, A. Rustighi, M.A. Tessari, et al. (2004) Nuclear phosphoproteins HMGA and their relationship with chromatin structure and cancer. FEBS Letters 574, 1–8. R. Sgarra, M.A. Tessari, J. di Bernardo, et al. (2005) Discovering high mobility group A molecular partners in tumor cells. Proteomics 5, 1494–1506. B. Shykind, J. Kim and P.A. Sharp (1995) Activation of the TFIID–TFIIA complex with HMG-2. Genes and Development 9, 1354–1365. K.B. Singh (1998) Transcriptional regulation in plants: the importance of combinatorial control. Plant Physiology 118, 1111–1120. S. Spiker (1984) High mobility group chromosomal proteins of wheat. The Journal of Biological Chemistry 259, 12007–12013. S. Spiker (1988) Histone variants and high mobility group non-histone chromosomal proteins of higher plants: their potential for forming a chromatin structure that is either poised for transcription or transcriptionally inert. Plant Physiology 74, 200–213. S. Spiker and E.M. Everett (1987) Blotting index of dissimilarity: use to study immunological relatedness of plant and animal high mobility (HMG) chromosomal proteins. Plant Molecular Biology 9, 431–442. C. Stemmer, S. Fern´andez, G. Lopez, J.C. Alonso and K.D. Grasser (2002a) Plant chromosomal HMGB proteins efficiently promote the bacterial site-specific β-mediated recombination in vitro and in vivo. Biochemistry 41, 7763–7770. C. Stemmer, R. Grimm and K.D. Grasser (1999) Occurrence of five different chromosomal HMG1 proteins in various maize tissues. Plant Molecular Biology 41, 351–361. C. Stemmer, D.J. Leeming, L. Franßen, R. Grimm and K.D. Grasser (2003) Phosphorylation of maize and Arabidopsis HMGB proteins by protein kinase CK2α. Biochemistry 42, 3503–3508. C. Stemmer, C. Ritt, G.L. Igloi, R. Grimm and K.D. Grasser (1997) Variability in Arabidopsis thaliana chromosomal high-mobility-group-1-like proteins. European Journal of Biochemistry 250, 646– 652. C. Stemmer, A. Schwander, G. Bauw, P. Fojan and K.D. Grasser (2002b) Protein kinase CK2 differentially phosphorylates maize chromosomal high mobility group B (HMGB) proteins modulating their stability and DNA interactions. The Journal of Biological Chemistry 277, 1092– 1098. M. Stros and E. Musel´ıkov´a (2000) A role of basic residues and the putative intercalating phenylalanine of the HMG-1 box B in DNA supercoiling and binding to four-way junctions. The Journal of Biological Chemistry 275, 35699–35707. J.O. Thomas and A.A. Travers (2001) HMG1 and 2, and related architectural DNA-binding proteins. Trends in Biochemical Sciences 26, 167–174. M.S. Thomsen, L. Franßen, D. Launholt, P. Fojan and K.D. Grasser (2004) Interactions of the basic N-terminal and the acidic C-terminal domains of the maize chromosomal HMGB1 protein. Biochemistry 43, 8029–8037.
ARCHITECTURAL HMG PROTEINS
77
R. Tjian and T. Maniatis (1994) Transcriptional activation: a complex puzzle with few easy pieces. Cell 77, 5–8. A. Travers (2000) Recognition of distorted DNA structures by HMG domains. Current Opinion in Structural Biology 10, 102–109. I. Ugrinova, E.A. Pasheva, J. Armengaud and I.G. Pashev (2001) In vivo acetylation of HMG1 protein enhances its binding affinity to distorted DNA structures. Biochemistry 40, 14655–14660. D. Van den Boeck, D. Van der Straten, M. Van Montagu and A. Caplan (1994) A group of chromosomal proteins is specifically released by spermine and loses DNA-binding activity upon phosphorylation. Plant Physiology 106, 559–566. K.E. van Holde (1989) Chromatin. Springer, Heidelberg. C.I. Webster, M.A. Cooper, L.C. Packman, D.H. Williams and J.C. Gray (2000) Kinetic analysis of high-mobility-group proteins HMG-1 and HMG-I/Y binding to cholesterol-tagged DNA on a supported lipid monolayer. Nucleic Acids Research 28, 1618–1624. C.I. Webster, L.C. Packman and J.C. Gray (2001) HMG-1 enhances HMGI/Y binding to an A/Trich enhancer element from the pea plastocyanin gene. European Journal of Biochemistry 268, 3154–3162. C.I. Webster, L.C. Packman, K.-H. Pwee and J.C. Gray (1997) High mobility group proteins HMG-1 and HMGI/Y bind to a positive regulatory region of the pea plastocyanin gene. The Plant Journal 11, 703–715. J.R. Wisniewski, N.M. Krohn, E. Heyduk, K.D. Grasser and T. Heyduk (1999a) HMG1 proteins from evolutionary distant organisms distort B-DNA conformation in similar way. Biochimica et Biophysica Acta 1447, 25–34. J.R. Wisniewski, E. Schulze and B. Sapetto (1994) DNA binding and nuclear translocation of insect high-mobility-group-1 (HMG1) proteins are inhibited by phosphorylation. European Journal of Biochemistry 225, 687–693. J.R. Wisniewski, Z. Szewczuk, I. Petry, R. Schwanbeck and U. Renner (1999b) Constitutive phosphorylation of the acidic tails of the high mobility group proteins by casein kinase II alters their conformation, stability and DNA binding specificity. The Journal of Biological Chemistry 274, 20116–20122. Q. Wu, W. Zhang, K.-H. Pwee and P.P. Kumar (2003a) Cloning and characterization of rice HMGB1 gene. Gene 312, 103–109. Q. Wu, W. Zhang, K.-H. Pwee and P.P. Kumar (2003b) Rice HMGB1 protein recognizes DNA structures and bends DNA efficiently. Archives of Biochemistry and Biophysics 411, 105–111. S. Yamamoto and T. Minamikawa (1997a) The high mobility group protein HMG-Y binds to promoter regions of seed storage protein genes from Canavalia gladiata D.C. Plant Science 124, 165– 173. S. Yamamoto and T. Minamikawa (1997b) Two genes for the high mobility group protein HMG-Y are present in the genome of Canavalia gladiata D. C. Plant Molecular Biology 33, 537–544. S. Yamamoto and T. Minamikawa (1998) The isolation and characterisation of a cDNA encoding a high mobility group protein HMG-1 from Canavalia gladiata D.C. Biochimica et Biophysica Acta 1396, 47–50. S. Yanagisawa (1997) Dof DNA-binding domains of plant transcription factors contribute to multiple protein–protein interactions. European Journal of Biochemistry 250, 403–410. Y.-M. Yen, B. Wong and R.C. Johnson (1998) Determinants of DNA binding and bending by the Saccharomyces cerevisiae high mobility group protein NHP6A that are important for its biological activities. The Journal of Biological Chemistry 273, 4424–4435. J. Yie, M. Merika, N. Munshi, G. Chen and D. Thanos (1999) The role of HMGI(Y) in the assembly and function of the IFN-beta enhanceosome. The EMBO Journal 18, 3074–3089. V. Zappavigna, L. Falciola, M. Helmer-Citterich and M.E. Bianchi (1996) HMG1 interacts with HOX proteins and enhances their DNA binding and transcriptional activation. The EMBO Journal 15, 4981–4991. W. Zhang, Q. Wu, K.-H. Pwee, S.D.S. Jois and R.M. Kini (2003a) Characterization of the interaction of wheat HMGa with linear and four-way junction DNA. Biochemistry 42, 6596–6607.
78
REGULATION OF TRANSCRIPTION IN PLANTS
W. Zhang, Q. Wu, K.-H. Pwee and R.M. Kini (2003b) Interaction of wheat high-mobility-group proteins with four-way-junction DNA and characterization of the structure and expression of HMGA gene. Archives of Biochemistry and Biophysics 409, 357–366. J. Zhao and G. Grafi (2000) The high mobility group I/Y protein is hypophosphorylated in endoreduplicating maize endosperm cells and is involved in alleviating histone H1-mediated transcriptional repression. The Journal of Biological Chemistry 275, 27494–27499. K. Zhao, E. K¨as, E. Gonzales and U.K. Laemmli (1993) SAR-dependent mobilization of histone H1 by HMGI/Y in vitro. The EMBO Journal 12, 3237–3247. X. Zhou, K.F. Benson, H.R. Ashar and K. Chada (1995) Mutation responsible for the mouse pygmy phenotype in the developmentally regulated factor HMGI-C. Nature 376, 771–774. J. Zlatanova, P. Caiafa and K. van Holde (2000) Linker histone binding and displacement: versatile mechanism for transcriptional regulation. The FASEB Journal 14, 1697–1704. J. Zlatanova and K. van Holde (1998) Linker histones versus HMG1/2: a struggle for dominance? Bioessays 20, 584–588. Y. Zou and Y. Wang (2005) Tandem mass spectrometry for the examination of the posttranslational modifications of the high-mobility group A1 proteins: symmetric and asymmetric dimethylation of Arg25 in HMGA1a protein. Biochemistry 44, 6293–6301. S. Zwilling, H. K¨onig and T. Wirth (1995) High mobility group protein 2 functionally interacts with the POU domains of octamer transcription factors. The EMBO Journal 14, 1198–1208.
4 Histone modifications and transcription in plants Yii Leng Chua and John C. Gray
The N-terminal tails of histones H3 and H4 are modified post-translationally by acetylation and methylation. Acetylation of histones is reversible and dynamic, whereas histone methylation patterns persist through cell divisions, functioning as epigenetic memory. Plant histone acetyltransferases (HATs) and histone deacetylases (HDACs) regulate transcription of a wide range of genes, including genes that respond to environmental signals, and those that regulate development. Plants also have unique bromodomain (BRD) genes that control biological processes specific to plant physiology, such as seed germination and leaf morphogenesis. HATs, HDACs and BRD proteins modify growth patterns in response to a changing environment, and play important roles in adaptation, given the sessile nature of plants. HDACs play a central role in transcriptional control, because deacetylation of histones generally precedes histone methylation during gene silencing. Histone methylation by histone methyltransferases (HMTases) regulates spatial patterns of gene expression to specify organogenesis, genomic imprinting and flowering time, and confer cell identity. We review the functions of various plant proteins that modify acetylation and methylation patterns of histones, and discuss the interplay between histone acetylation and methylation in transcriptional control in plants.
4.1
Introduction
Interphase chromatin is organised into nucleosomes (Figure 4.1A), which are composed of a linker histone, a histone octamer and 146 bp of DNA wound round the octamer in a 1.65 turn (reviewed by McGhee and Felsenfeld, 1980). The histone octamer contains two molecules each of the core histones H2A, H2B, H3 and H4, and is linked to the next nucleosome by linker DNA of variable length (Figure 4.1A) (Kornberg, 1974). The core histones possess a histone-fold domain and a basic Nterminal tail (Wolffe and Hayes, 1999). The histone-fold domain folds into three α-helices and interacts with the nucleosomal DNA (Figure 4.1B). Extending out of the nucleosome are the basic N-terminal tails of histones H2A, H2B, H3 and H4, and the basic C-termini of histone H2A (Luger et al., 1997). The N-terminal tails are modified by acetylation, methylation, phosphorylation, adenosine diphosphate (ADP)-ribosylation, ubiquitination and addition of small ubiquitin modifier (SUMO). These post-translational covalent modifications regulate various cellular processes, such as transcription, DNA replication, chromatin assembly, doublestranded DNA break repair and cell division (Brownell and Allis, 1996; Strahl and Allis, 2000).
80
REGULATION OF TRANSCRIPTION IN PLANTS
Figure 4.1 Nucleosomes: (A) Schematic diagram of DNA organised into nucleosomes. DNA is wrapped around the histone octamer. The N-terminal tails of histones extend out of the nucleosomes. (Adapted from Luger, 2002.) (B) Schematic diagram of a nucleosome. Histones H3 (dark grey), H4 (black), H2A (light grey) and H2B (white) each fold into three α-helices. Positions of lysine residues in the N-termini of H3 and H4 are indicated by numbers. (Adapted from Luger, 2003.)
HISTONE MODIFICATIONS AND TRANSCRIPTION IN PLANTS
81
Although the nucleosome structure was elucidated using avian and mammalian chromatin, the structure was later discovered to be highly conserved amongst yeasts, insects, plants and mammals (Bavykin et al., 1985). This suggests that the nucleosome is an extremely ancient macromolecule, which was formed before plants and animals diverged during evolution. Mutations that alter the nucleosome structure are likely to be lethal. Not only is the tertiary structure conserved, but the amino acid sequences of histones H3 and H4 are similar between different organisms (deLange et al., 1968; Patthy and Smith, 1973). For example, pea and calf thymus H4 differ at only two amino acids (Patthy and Smith, 1973). In particular, the amino acid sequences of the N-terminal tails of histones H3 and H4 are extremely conserved, and post-translational covalent modifications occur at the same lysine residues in different organisms (Figures 4.2A and 4.2B). Although the amino acid sequence of histone H2A is less conserved, the locations of lysine residues in the N-terminus that are subjected to post-translational modifications, are similar in different species (Figure 4.2C). In contrast, the N-terminal sequences of plant and animal histones
Figure 4.2 N-terminal sequences of histones: alignment of N-terminal sequences of human, Drosophila, yeast and Arabidopsis (A) histone H3, (B) histone H4 and (C) histone H2A. Lysine residues subjected to acetylation are numbered and labelled Ac.
82
REGULATION OF TRANSCRIPTION IN PLANTS
H2B are different, suggesting that post-translational modifications of histones H2B and their biological functions are likely to be different between plants and animals. Nucleosomes are often positioned on specific sequences along a gene (Simpson et al., 1993). Nucleosome positioning is defined as the presence of a nucleosome on the same DNA sequence in all cells (Simpson et al., 1993). The nucleosomal DNA that wraps round the histone octamer in an 80 base-pair circle is distorted, containing sharp kinks in the double helix (Richmond et al., 1984; Luger et al., 1997). Thus, DNA sequences that encourage nucleosome positioning are often A/T-rich, with a propensity to curve (Richmond et al., 1984; Kornberg and Lorch, 1999). Promoter and enhancer regions are A/T-rich, and nucleosomes are often positioned on these regulatory sequences (Laybourn and Kadonaga, 1991; Workman and Kingston, 1998). The presence of nucleosomes on promoters and enhancers is generally inhibitory to transcription, because nucleosomal DNA is not available for binding to transcription activators and assembly of the RNA polymerase machinery. Inhibition is most effective when the recognition sequences are present at the pseudodyad (centre) than at the edge of the nucleosomes (Xu et al., 1998). In plants, nucleosomes are selectively positioned on the promoter of the β-phaseolin gene when the gene is inactive (Li et al., 1998, 1999). Nucleosomes also inhibit the interaction of the wheat transcription factor EmBP-1 with its recognition site in vitro (Niu et al., 1996). Interphase chromatin is also organised into euchromatin and heterochromatin. By definition, euchromatin is the chromatin that stains poorly with DNA-staining dyes, whereas heterochromatin stains deeply (Pardue and Hennig, 1990). Euchromatin consists of regions of the genome that are transcriptionally competent and possess an open chromatin structure. An open chromatin structure is currently thought to consist of DNA with widely spaced nucleosomes containing acetylated histones, and is organised into loops of 200–2000 kb, with the bases of the loops attached to a nuclear scaffold (Heslop-Harrison, 2003). In contrast, heterochromatin is highly condensed, is replicated late in S-phase and is inaccessible to transcription (Maison and Almouzni, 2004). The organisation of nucleosomal DNA into heterochromatin is poorly understood (Wolffe, 1995). Heterochromatic regions are also referred to as chromocentres. In Arabidopsis, there are nine chromocentres, consisting of centromeric regions of the five chromosomes, and four regions of ribosomal RNA (rRNA) genes (Heslop-Harrison, 2003). Gene-silencing consequent of heterochromatin formation is observed in X-chromosome inactivation in mammals (Jeppesen and Turner, 1993), position-effect variegation in Drosophila (Reuter and Spierer, 1992) and telomeric silencing in yeast and mammals (Perrod and Gasser, 2003). In plants, heterochromatin formation is involved in silencing of transposons (Lippman and Martienssen, 2004), and repression of rRNA genes in nucleolar dominance (Lawrence et al., 2004). The most well-studied post-translational modifications in plants are acetylation and methylation of the N-terminal tails of histones H3 and H4. These modifications regulate transcription by facilitating mobilisation of nucleosomes from promoters, recruiting transcription factors, or by regulating formation of euchromatin or heterochromatin. In contrast, very little is known about the functions of ADPribosylation, ubiquitination and SUMOylation of histones in plants. Phosphorylation
HISTONE MODIFICATIONS AND TRANSCRIPTION IN PLANTS
83
of histones regulates DNA replication and cell division in plants (Houben et al., 2005), and is beyond the scope of this chapter. This review focuses on histone acetylation and methylation, and their roles in regulation of transcription. We first give an overview of the functions of plant chromosomal proteins that regulate acetylation and methylation of histones, and the transcription factors that recognise these posttranslational modifications. We then discuss how histone acetylation and methylation may act in conjunction to regulate transcription in plants.
4.2
Histone acetylation and transcriptional activation
Acetylation of histones involves the transfer of acetyl (—COCH3 ) groups from acetyl-coenzyme A to the ε-amino groups of lysine residues K4, K9, K14, K18 or K23 of histone H3, or lysine residues K5, K8, K12 K16 or K20 of histone H4 by histone acetyltransferases (HATs) (Figure 4.2) (Lin et al., 1999; Rojas et al., 1999). Addition of the acetyl group results in loss of positive charge at the lysine side chain (Figure 4.3), abolishing interactions between the positively charged histone tails and the negatively charged DNA (McGhee and Felsenfeld, 1980; Cary et al., 1982). Acetylation of histone tails is generally correlated with activated transcription in yeasts, animals and plants. One of the earliest experiments indicating transcriptional activation by acetylation of histones in plants is the study of nucleolar dominance, a phenomenon where one parental set of rRNA genes is silenced in an interspecific hybrid (Chen and Pikaard, 1997). Brassica napus is a hybrid between Brassica
Figure 4.3 Acetylation of a lysine residue. Acetylation involves addition of —COCH3 to the ε-amino group of a lysine residue.
84
REGULATION OF TRANSCRIPTION IN PLANTS
oleracea and Brassica rapa. In B. napus, rRNA genes that are inherited from B. oleracea are silenced. Treatment of B. napus with trichostatin A or sodium butyrate, which are potent inhibitors of histone deacetylases (HDACs), increased histone acetylation levels and activated expression of B. oleracea rRNA genes (Chen and Pikaard, 1997). This suggests that rRNA genes are silenced by HDACs in nucleolar dominance (Chen and Pikaard, 1997). Subsequently, chromatin immunoprecipitation experiments showed that acetylated histones H3 and H4 were present on the promoter and enhancer regions of the actively transcribed plastocyanin gene (PetE) in green pea shoots, whereas these regions were associated with non-acetylated histones in etiolated shoots and roots, which had low PetE transcription (Chua et al., 2001). Interestingly, histone H3 acetylation levels at regions upstream of the enhancer, and at the coding region, remained unchanged during activation of PetE transcription (Chua et al., 2001). This indicates that acetylation of histones is targeted to specific nucleosomes present on important regulatory sequences during transcriptional activation (Chua et al., 2001). In transgenic tobacco plants containing the β-glucuronidase (GUS) reporter gene fused to the PetE promoter, treatment with trichostatin A and sodium butyrate increased histone acetylation levels at the promoter of the transgene and activated expression, indicating that acetylation of histones has a direct positive effect on transcription in plants (Chua et al., 2003). Unlike histone methylation, which activates or represses transcription depending on the lysine residue that is modified, transcriptional activation is generally correlated with acetylation of various lysine residues, such as K5, K8, K12 and K16 of histone H4, and K9 and K14 of histone H3 (Braunstein et al., 1996; Osano and Ono, 2003). Early biochemical studies suggested that acetylation of histones may increase transcription by altering the physical structure of chromatin to facilitate entry of transcription factors (Lee et al., 1993). Structural changes in chromatin may occur in the following ways: (1) acetylation of lysine residues neutralises the positive charges of histones, weakens histone–DNA contacts and facilitates the displacement of nucleosomes by transcription factors (McGhee and Felsenfeld, 1980; Cary et al., 1982; Puig et al., 1998); (2) acetylation also weakens nucleosome interconnections, inhibits formation of condensed chromatin and, hence, increases the accessibility of DNA recognition sequences to transcription factors (Luger et al., 1997; Mutskov et al., 1998; Tse et al., 1998); (3) histone acetylation changes the conformation of nucleosomal arrays by affecting the DNA path around nucleosomes, and releasing negative supercoils present in chromatin, thereby facilitating interactions of protein factors with chromatin (Norton et al., 1989; Bauer et al., 1994). In addition to altering chromatin structure, acetylation of histones may have a signalling function. The histone-code hypothesis proposes that post-translational modifications of histones provide heritable epigenetic information that specifies alternative states of chromatin to regulate cellular processes such as transcription and replication (Strahl and Allis, 2000). Acetylated histones may specify an activated transcriptional state by signalling to transcriptional activators or co-activators (Strahl and Allis, 2000). In support of this hypothesis is the observation that various transcriptional activators and co-activators contain the bromodomain (BRD) motif,
HISTONE MODIFICATIONS AND TRANSCRIPTION IN PLANTS
85
which interacts with acetylated histone tails, and are directed to actively transcribed genes (see Section 4.2.2). It is plausible that acetylation of histones has both a structural effect and a signalling function. Nonetheless, it is interesting that acetylation has evolved to be the post-translational modification that marks activated transcription, because it would be thermodynamically less favourable for deacetylation to specify an open chromatin structure and an activated transcriptional state, due to the physical properties of histone tails and DNA. Acetylation of histone tails is associated with activated transcription, with the exception of a small group of genes in yeasts and mammals (Montecino et al., 1999; Deckert and Struhl, 2001; Sheldon et al., 2001; Ricci et al., 2002). Continuous high histone acetylation levels at these genes has a negative effect on transcription, because deacetylation is required to convert these genes back to their original chromatin structures following transcription (Kurdistani and Grunstein, 2003). The original chromatin structures are necessary for binding of sequence-specific transcription factors for the next round of transcription initiation (Kurdistani and Grunstein, 2003). Thus, for some genes, activated transcription is regulated by a dynamic cycle of histone acetylation and deacetylation (Kurdistani and Grunstein, 2003). Acetylation of histones is also correlated with DNA replication and cell-cycle progression in plants (Jasencakova et al., 2000; Wako et al., 2002). DNA replication is associated with an increase in histone H4 acetylation levels over large domains of euchromatin and heterochromatin (Jasencakova et al., 2000, 2001). In contrast, changes in histone acetylation levels associated with transcription are targeted to specific nucleosomes on promoters, enhancers and/or coding regions of plant genes (Chua et al., 2001, 2004). In addition, whereas acetylation of various lysine residues in histones H3 and H4 is generally correlated with increased transcription, replication and mitosis are associated with changes in acetylation at lysine residues K5, K8, K12 or K16 of histone H4 (Jasencakova et al., 2000, 2001; Wako et al., 2002). The acetylation levels of histone H3 remain largely invariant during DNA replication and cell-cycle progression (Jasencakova et al., 2000, 2001).
4.2.1
Plant histone acetyltransferases
HATs are classified on the basis of sequence homology into four families (Sterner and Berger, 2000): the GNAT/MYST superfamily consisting of the General Control Non-repressed protein 5-related AcetylTransferases (GNATs), and the ‘MOZ, Ybf2, Sas2 and Tip60’-related HATs (MYSTs); the p300/CBP family consisting of the p300/cAMP-response element-binding (CREB) protein-related HATs; the TAFII 250 family; and the SRC3 family consisting of ‘SRC1 and ACTR’related HATs. Arabidopsis, rice and maize have genes encoding homologues of GNAT/MYST-, p300/CBP- and TAFII 250-type HATs, but not SRC3-type HATs, which are involved in nuclear hormone signalling in animal cells, suggesting that this pathway is likely to be non-existent in plants (Loidl, 2004). There are many genes encoding HATs in plants: 16 in Arabidopsis and 8 in rice (Plant Chromatin Database, http://www.chromdb.org/).
86
REGULATION OF TRANSCRIPTION IN PLANTS
Analysis of HAT mutants in Arabidopsis indicates that acetylation of histones plays a central role in transcriptional activation of many genes, such as cold-response, light-induced and pathogen-induced genes, and regulates various aspects of plant development, including floral organogenesis, leaf morphogenesis, root development and flowering time (Bertrand et al., 2003; Vlachonasios et al., 2003). Microarray analysis of chromatin-immunoprecipitated DNA by antibodies that recognise acetylated histone H4 tails indicates that histone acetylation is targeted to promoters of diverse sets of genes involved in photosynthesis, pathogen response, stress response and tissue-specific gene expression (Chua et al., 2004). Thus, acetylation of histones is a general mechanism that regulates transcription of a wide range of genes in plants.
4.2.1.1
GNAT/MYST family
The GNAT family is characterised by the acetyltrnsf 1 motif (Pfam accession: PF00583), an 80 amino acid residue consensus that binds to acetyl-coenzyme A. Plant GNAT-type HATs also possess BRDs that interact with acetylated histones, and also contain the radical S-adenosylmethionine (SAM) motif (Pfam accession: PF04055), which binds to an iron ion, and generates radical species by cleavage of SAM. There are three genes encoding GNAT-type HATs in Arabidopsis (HAG1, HAG2 and HAG3), and in rice (HAG702, HAG703 and HAG704) (Plant Chromatin Database). Yeast and Drosophila GNAT-type HAT GCN5 (General Control Non-repressed protein 5) proteins function as part of a multiprotein complex to regulate gene expression by acetylation of histones; they associate with Alteration/Deficiency in Activation (ADA) co-activators to form the Spt/Ada/GCN5 Acetyltransferase (SAGA) and ADA HAT complexes (Grant et al., 1997; Kusch et al., 2003). Plant GCN5 homologues are also part of multiprotein HAT complexes. For example, Arabidopsis HAG1 interacts with Arabidopsis homologues of ADA co-activators, ADA2a and ADA2b, to control expression of cold-regulated genes that contain CRT/DRE (C-repeat/dehydration-responsive) regulatory sequences (Stockinger et al., 2001). Recruitment of HAG1 to cold-regulated genes is likely to be mediated by CBF1, a transcription factor that binds to CRT/DRE (Stockinger et al., 2001). The maize ZmGCN5 also interacts with the maize homologue of ADA co-activator ZmADA2, and associates with the bZIP transcription factor OPAQUE2 to regulate gene expression (Bhat et al., 2003, 2004). Although purified recombinant HAG1 or ZmGCN5 acetylate free histones in vitro (Bhat et al., 2003), the associated ADA factors are likely to be required for acetylating histone tails in a chromatin template and for regulating substrate specificity (Balasubramanian et al., 2002). T-DNA insertional mutations in Arabidopsis HAG1 or ARADA2b reduced activation of cold-regulated genes by cold treatment (Vlachonasios et al., 2003). The observation that the expression of cold-regulated genes is not totally abolished in the mutants suggests that HAG1 controls activation but not initiation of transcription. In addition, HAG1 also regulates various aspects of plant development; HAG1 mutants have reduced root growth, smaller rosettes, curled leaves, smaller and fewer siliques, and homeotic transformation of petals into stamens, and sepals into needlelike structures (Bertrand et al., 2003; Vlachonasios et al., 2003). The floral homeotic
HISTONE MODIFICATIONS AND TRANSCRIPTION IN PLANTS
87
transformations are due to activation of WUSCHEL (WUS) and AGAMOUS (AG) in the outer whorls of the flower, suggesting that HAG1 regulates floral organogenesis through the WUS/AG pathway (Bertrand et al., 2003). The plant MYST family of HATs contains the MOZ SAS domain (Pfam accession: PF01853) and zinc fingers. In yeast and mammals, MYST-type HATs regulate a wide range of cellular processes, including gene expression, chromatin assembly and double-stranded DNA break repair (Osada et al., 2001; Bird et al., 2002; Gomez et al., 2005). Two genes encoding MYST-type HATs (HAM1, HAM2) have been predicted for Arabidopsis, and one (HAM701) for rice (Plant Chromatin Database). The functions of these genes have not yet been investigated.
4.2.1.2
TAF I I 250 family
The TAFII 250 family of HATs associates with the TATA-box binding protein (TBP), which is a component of the basal transcription complex (Ruppert et al., 1993). TAFII 250 possesses a HAT catalytic domain of 260 amino acid residues, and activates transcription through acetylation of histones (Mizzen et al., 1996). There are two Arabidopsis TAFII 250 genes (HAF1 and HAF2), and one rice gene (HAF701) (Plant Chromatin Database). Plant HAF proteins also possess BRDs. Arabidopsis HAF2 regulates expression of photosynthesis genes. T-DNA insertion mutations in HAF2 led to reduced chlorophyll levels, and downregulation of light-responsive genes, such as LHCB1 and RBCS1A (Bertrand et al., 2005). Moreover, histone H3 acetylation levels at the promoters of LHCB1 and RBCS1A are decreased in HAF2, as demonstrated by chromatin immunoprecipitation experiments, indicating that HAF2 regulates transcription by acetylation of histones (Bertrand et al., 2005). These results are consistent with the observation that acetylation of histones is a key mechanism that activates expression of the photosynthesis gene PetE in light (Chua et al., 2001). Interestingly, mutations in HAF1 do not have phenotypes similar to HAF2, suggesting that the downstream genes regulated by the two HAF transcription factors are different (Bertrand et al., 2005). Analysis of Arabidopsis det1 (de-etiolated1) also suggests that HATs play important roles in regulating expression of light-induced genes (Benvenuto et al., 2002; Schroeder et al., 2002). Seedlings grown in the dark are characterised by long hypocotyls, closed and yellow cotyledons, and have low expression of photosynthesis genes, whereas light-grown seedlings have short hypocotyls, and green and expanded cotyledons. When grown in the dark, det1 has activated expression of photosynthesis genes and morphologies similar to light-grown seedlings, suggesting DET1 is involved in repression of light-induced genes. DET1 is a nuclear protein that binds to non-acetylated tails of histone H2B in dark-grown plants (Benvenuto et al., 2002). Following illumination, DET1 associates with UV-DDB1 (UV-Damaged DNA-Binding Protein 1), which consequently recruits HATs, leading to acetylation of histones and activated expression of light-induced genes (Schroeder et al., 2002).
4.2.1.3
P300/CBP family
The p300/CBP family of HATs possesses the CBP-type HAT domain, which contains the plant homeodomain (PHD) motif and a putative acetyl-coenzyme A-binding
88
REGULATION OF TRANSCRIPTION IN PLANTS
motif (Yuan and Giordano, 2002). Arabidopsis has five genes encoding p300/CBPtype HATs: HAC1, HAC2, HAC4, HAC5 and HAC12 (Plant Chromatin Database). Arabidopsis HAC1 (also known as PCAT2) acetylated histones in vitro, and activated gene expression when it was targeted to promoters of reporter genes in human cell lines (Bordoli et al., 2001). Mutants of HAT genes in this family have not been characterised, and the downstream genes regulated by p300/CBP-type HATs have not been investigated. Although acetylation of histones is a general mechanism that upregulates expression of many plant genes, specificity in gene regulation is likely to be mediated by different types of HATs. Different families of HATs regulate different aspects of plant physiology; for example, the Arabidopsis GNAT-type HAT HAG1 regulates cold-response genes, whereas the TAFII 250 protein HAF2 controls expression of light-response genes (Vlachonasios et al., 2003; Bertrand et al., 2005). In addition, different members of the same family appear to have distinct functions, indicating that they are recruited to different groups of downstream genes (Bertrand et al., 2005). Moreover, plant HATs function in multiprotein complexes. Thus, each HAT can associate with different sequence-specific transcription factors to regulate groups of genes independently and in a coordinated manner. In yeast and mammals, HATs also regulate transcription by acetylating transcription factors such as p53 and high mobility group (HMG) proteins in addition to histones (Bode and Dong, 2004; Edberg et al., 2004). Future work may indicate whether plant HATs also control transcription by modifying transcription factors.
4.2.2
Bromodomain proteins
Acetylated histones H3 and H4 regulate transcription through interaction with the BRD (Pfam accession: PF00439), a 110 amino acid residue motif that binds acetyllysine side chains, and is present in various yeast, mammalian and plant transcription factors (Jeanmougin et al., 1997). The BRD of the human p300/CBP-associated factor (PCAF) was first demonstrated to interact with acetylated lysines in histone H3 peptides in vitro, by nuclear magnetic resonance spectroscopy (Dhalluin et al., 1999). Subsequently, the BRD was shown to bind selectively to acetylated histones in vivo by fluorescence resonance energy transfer assays (Kanno et al., 2004). The BRD folds into a four-helical bundle with two loops at each end, forming a hydrophobic pocket that selectively fits the non-charged side chain of an acetylated lysine residue, and not the positively charged side chain of an unmodified lysine residue (Owen et al., 2000). The BRD functions as a targeting module for various transcription factors. BRDs activate transcription by anchoring HATs to promoters of actively transcribed genes. The BRD motif is present in various plant HATs, including Arabidopsis HAG1, HAF1 and HAF2, maize HAF101 and HAG101 and rice HAF701 and HAG702. HATs are initially recruited to genes by sequence-specific transcription factors that recognise promoter sequences (Hassan et al., 2001, 2002). Following acetylation of histone tails, HATs are then tethered to acetylated histones through their BRDs, and remain associated with the promoter in the absence of sequence-specific
HISTONE MODIFICATIONS AND TRANSCRIPTION IN PLANTS
89
transcription factors (Hassan et al., 2001, 2002). Stable promoter occupancy by HATs maintains high histone acetylation status, and activated transcription. BRDs are also present in chromatin-remodelling proteins such as CHROMATIN REMODELLING COMPLEX SUBUNIT R (CHR) 2 in Arabidopsis, and CHR707 in rice. CHR is the ATPase component of the Sucrose Non-Fermenting 2 (SNF2) family of proteins, which disrupt histone–DNA contacts and mobilise nucleosomes. Chromatin-remodelling proteins may thus be recruited to actively transcribed genes through BRDs. Bromodomains also bind to acetyl-lysine side chains in other proteins such as p53, C-Myb and MyoD, and may regulate transcription through interactions with transcription factors (Yang, 2004).
4.2.2.1
Bromodomain extra-terminal proteins
In yeast and mammals, the Bromodomain Extra-Terminal (BET) family of proteins possesses BRDs and an ET domain, and regulates cell proliferation and growth (Chua and Roeder, 1995; Houzelstein et al., 2002). There are 12 Arabidopsis and 10 rice genes encoding BET-like proteins (Plant Chromatin Database). These genes are designated GENERAL TRANSCRIPTION FACTOR GROUP E (GTE). Unlike yeast and mammalian BET proteins, which have two BRDs, plant GTE proteins contain one BRD. Moreover, many plant GTE genes encode only the N-terminus of the ET domain, the NET domain, and not the serine-rich C-terminus, which is known as the SEED domain (Florence and Faller, 2001). The functions of the ET domain have not been elucidated, but the motif has been proposed to be involved in protein–protein interactions (Lygerou et al., 1994). Arabidopsis GTE1 and GTE6 regulate different aspect of plant development, possibly because they are targeted to different downstream genes. GTE1 is involved in regulation of seed germination by light and abscisic acid (ABA) (Duque and Chua, 2003). Inactivation of GTE1 by a T-DNA insertion resulted in repression of a broad class of genes, including genes encoding proteins involved in cell wall formation, plastid transcription and translation, and lipid metabolism, suggesting that GTE1 functions as a general transcriptional activator (Duque and Chua, 2003). GTE6 controls leaf patterning during juvenile-to-adult transition in Arabidopsis (Chua et al., 2005). In many plant species, the transition from the juvenile to the mature phase is characterised by changes in leaf structure (Kerstetter and Poethig, 1998). In Arabidopsis, juvenile leaves possess round laminae, whereas mature leaves have elliptical laminae (Lawson and Poethig, 1995; Theodoris et al., 2003). GTE6 regulates the development of elliptical leaf shapes in mature plants by activating transcription of the myb-domain gene ASYMMETRIC LEAVES1 (AS1). GTE6 associates with the promoter and the first exon of the gene, and activates transcription through acetylation of histones (Chua et al., 2005). AS1 in turn controls proximodistal patterning of leaves by repressing transcription of the KNOX genes KNAT1 and KNAT2 (Byrne et al., 2002; Theodoris et al., 2003). Although GTE1 and GTE6 have no HAT domains, they may activate transcription through acetylation of histones by interacting with HATs (Hassan et al., 2001). Alternatively, they may activate transcription by binding to acetylated histone tails and shielding them from the actions of HDACs (Ladurner et al., 2003). Interestingly,
90
REGULATION OF TRANSCRIPTION IN PLANTS
a tomato BET protein, VIRP1, binds to potato spindle tuber viroid RNA in the nuclei of infected leaves (Martinez de Alba et al., 2003). Because VIRP1 binds to RNA and is localised to the nucleus, it may be involved in RNA-mediated DNA methylation and gene silencing, or transportation of RNA to the nucleus (Martinez de Alba et al., 2003).
4.2.2.2
Plant bromodomain proteins
There are six Arabidopsis genes that encode proteins possessing a single BRD in conjunction with other functional motifs such as the ATPase AAA domain (BRD8), the Myb-like DNA-binding domain (BRD3, 4, 6 and 9) and WD-40 repeats (BRD11 and 12). Arabidopsis BRD1, 2, 5, 7, 10 and 13 encode proteins containing only one BRD. The functions of these plant BRD proteins have not been elucidated, and they may regulate transcription or cellular processes such as replication, cell cycle and protein degradation by associating with acetyl-lysine side chains in histones and other proteins.
4.2.3
Plant histone deacetylases
Opposing the activities of HATs are the histone deacetylases (HDACs), which remove acetyl groups from acetylated lysine residues (Taunton et al., 1996; Rundlett et al., 1998). HATs and HDACs act in conjunction to regulate global histone acetylation levels, and local histone acetylation at regulatory sequences of genes (Kurdistani and Grunstein, 2003). A total number of 18 HDACs has been predicted from the genome sequences of Arabidopsis and rice (Pandey et al., 2002). Plant HDACs are classified on the basis of sequence similarity into three families: the Reduced Potassium Deficiency 3/Histone Deacetylase 1 (RPD3/HDA1) family, the Histone Deacetylase 2 (HD2) family and the Silent Information Regulator 2 (SIR2) family.
4.2.3.1
RPD3/HDA1 family
The RPD3/HDA1 family of HDACs is characterised by the histone deacetylase domain, which spans over 300 amino acid residues (Pfam accession: PF00850). There are 12 genes encoding RPD3/HDA1-type HDACs in Arabidopsis, and 14 in rice (Plant Chromatin Database). Plant RPD3/HDA1-type HDACs function as part of multiprotein complexes, and are targeted to specific genes by co-repressors. For example, maize ZmRPD3I associates with ZmRBR1, the maize homologue of Retinoblastoma-Related (RBR), and with ZmRbAp1, a maize MSI/RbAp corepressor protein, to downregulate transcription (Rossi et al., 2003). Amongst the 12 genes encoding Arabidopsis RPD3/HDA1-type HDACs, HDA19 (also known as AtRPD3A or OOK) and HDA6 (also known as ATRPD3B) have been well investigated. HDA19 repressed transcription when it was directed to promoters of reporter genes in transient expression assays (Wu et al., 2000a). Chromatin immunoprecipitation experiments indicated that activated genes in HDA19 null mutants possessed increased histone acetylation levels, suggesting that HDA19 repressed gene expression by deacetylation of histones (Tian et al., 2005). Downregulation of HDA19 by antisense constructs and T-DNA insertions in Arabidopsis
HISTONE MODIFICATIONS AND TRANSCRIPTION IN PLANTS
91
increased global histone acetylation levels, and had pleiotropic effects on plant development; plants had serrated and elongated leaves, floral homeotic transformations, reduced fertility and were late flowering (Wu et al., 2000a; Tian and Chen, 2001; Tian et al., 2003). In HDA19 null mutants, almost 360 flower-specific genes such as SUPERMAN were activated in leaves, and 80 leaf-specific genes were expressed in flowers, indicating that HDA19 regulated tissue-specific gene expression (Tian et al., 2005). In rice, over-expression of OsHDAC1 results in increased growth rate, altered leaf structure and elongated roots, suggesting that RPD2/HDA-type HDACs also regulate growth and plant development in monocots (Jang et al., 2003). Arabidopsis HDA19 also plays a central role in pathogen and stress responses (Song et al., 2005; Zhou et al., 2005). HDA19 is activated following wounding, pathogen invasion and treatment with jasmonic acid or ethylene (Zhou et al., 2005). Plants over-expressing HDA19 were more resistant to pathogen invasion, and possessed increased expression of ETHYLENE RESPONSE FACTOR1 (ERF1) and PATHOGENESIS-RELATED (PR) genes encoding basic chitinase and b-1,3glucanase (Zhou et al., 2005). HDA19 also regulates gene expression in response to ABA, which accumulates in plants during drought, cold and salt stress (Song et al., 2005). HDA19 associates with the co-repressor AtSin3, and with an APETALA2/EREBP-type transcription factor AtERF7, to repress transcription of ABA-regulated genes that contain ethylene-response elements (Song et al., 2005). Apart from deacetylating histones, HDA19 may also regulate gene expression by interacting with transcription activators. In B. napus, cold treatment upregulates the expression of BnKCP1, which encodes a transcription activator containing a putative kinase-inducible domain (Gao et al., 2003). BnKCP1 associates with HDA19 to regulate expression of reporter genes (Gao et al., 2003). HDA19 also interacts with another transcription activator SCARECROW-like to control plant development (Gao et al., 2004). It is likely that HDA19 represses transcription through another mechanism, i.e. by sequestering transcriptional activators. Alternatively, activators may be expressed in a cell-specific manner to prevent or relieve repression of gene expression by HDA19 in certain tissues (Gao et al., 2004). Arabidopsis HDA6 is also a repressor of transcription. However, downregulation of HDA6 did not result in global changes in histone acetylation levels, but only increased histone H4 acetylation levels at certain regions of the genome, suggesting that HDA6 is not a general transcriptional repressor like HDA19 (Probst et al., 2004). Mutations in HDA6 resulted in reactivation of silenced transgenes and transposons, indicating that HDA6 is involved in maintaining gene silencing (Murfett et al., 2001; Aufsatz et al., 2002). HDA6 functions in conjunction with DNA methyltransferases, DDM1 and MET1, in RNAi-mediated gene silencing of transposons (Lippman et al., 2003). Thus, mutations in HDA6 also lead to decreased or modified DNA methylation patterns at rRNA repeats and transposons (Aufsatz et al., 2002; Lippman et al., 2003). The phenotypes of HDA6 mutants are different from HDA19; HDA6 mutants are late flowering and do not possess altered morphology or growth patterns. Thus, the downstream target genes of HDA19 and HDA6 are likely to be different.
92
4.2.3.2
REGULATION OF TRANSCRIPTION IN PLANTS
HD2 family
The HD2 family of HDACs is unique to plants, and is absent from mammals or fungi (Lusser et al., 1997; Pandey et al., 2002). Arabidopsis has four HD2-type HDAC genes, namely HDT1, HDT2, HDT3 and HDT4, and rice has two, HDT701 and HDT702. HD2-type HDACs are characterised by a nucleoplasmin motif (Pfam accession: PF03066). Nucleoplasmins are chromatin de-condensation proteins that regulate histone–DNA interactions (Earnshaw et al., 1980). HDT1, HDT2 and HDT701 also possess C2H2-type zinc fingers (Pfam accession: PF00096) for interactions with DNA. HD2 was first described as a nucleolar acid phosphoprotein with histone deacetylation activity in maize (Lusser et al., 1997). In Arabidopsis, HDT1, HDT2 and HDT3 localise to the nucleus, and repress gene expression when they are targeted to promoters of reporter genes (Wu et al., 2003; Zhou et al., 2004). Downregulation of HDT1 by an antisense construct, or expression of HDT1-GFP fusions, leads to aborted seeds in Arabidopsis (Wu et al., 2000b; Zhou et al., 2004). Expression of LATE EMBRYOGENESIS-ABUNDANT PROTEIN (LEA), SEED MATURATION PROTEIN and ABSCISIC ACID-RESPONSIVE ELEMENTS-BINDING FACTOR3 was misregulated in HDT1 mutants, indicating that HDT1 is a master regulatory gene that controls embryogenesis and/or endosperm development (Zhou et al., 2004). Arabidopsis HDT1 also regulates silencing of rRNA genes in nucleolar dominance (Lawrence et al., 2004). In Arabidopsis suecica, which is a hybrid between Arabidopsis thaliana and Arabidopsis arenosa, rRNA genes that are inherited from A. thaliana are silent, whereas rRNA genes derived from A. arenosa are expressed. Downregulation of HDT1 by RNAi leads to loss of DNA methylation, increased levels of trimethylated K4 in histone H3 and reduced levels of dimethylated K9 in histone H3 at A. thaliana rRNA genes, leading to gene expression and loss of nucleolar dominance (Lawrence et al., 2004). Trimethylated K4 in histone H3 is a hallmark of euchromatin, whereas dimethylated K9 in histone H3 is associated with transcriptionally repressed heterochromatin (Lawrence et al., 2004). HDT1 thus plays a central role in maintaining rRNA genes in a repressive chromatin structure by deacetylating amino acid residue K9 of histone H3, which consequently becomes available for methylation (Lawrence et al., 2004). Downregulation of HDT3 or HDT4 by RNAi does not affect nucleolar dominance in Arabidopsis, suggesting that nucleolar dominance is regulated specifically by HDT1, and not by the other members of the HD2 family (Lawrence et al., 2004).
4.2.3.3
SIR2 family
The SIR2 family consists of NADH-dependent HDACs, which are characterised by a SIR2-type HDAC domain (Pfam accession: PF02146). There are two genes encoding SIR2-type HDACs in Arabidopsis (HDA12 and HDA14), and two genes in rice (HDA701 and HDA702). Inhibition of SIR2-type HDACs in Arabidopsis by compounds containing the hydroxynaphthaldehyde moiety led to altered leaf structures and perturbation of body axis (Grozinger et al., 2001). In addition, the inhibitors prevented the formation of the vascular system in seedlings (Grozinger et al., 2001).
HISTONE MODIFICATIONS AND TRANSCRIPTION IN PLANTS
93
SIR2-type HDACs regulate growth and development in plants (Grozinger et al., 2001).
4.2.4
Histone acetylation/deacetylation and environmental adaptation
Acetylation of histones is an important mechanism that regulates the expression of genes that respond to external stimuli such as cold, light and pathogens. Due to the reversible nature of histone acetylation, genes can be rapidly activated in the presence of an external stimulus by acetylation of histones, and then subsequently repressed by histone deacetylation when the stimulus ceases. Transcription by RNA polymerase II is a complex process; the TBP binds weakly to promoter sequences, and requires sequence-specific transcription factors along with a host of general transcription factors to be assembled into a functional RNA polymerase II machinery (Kadonaga, 1998). Following acetylation of histones, HATs and BRD-containing activators are directed to promoters, surpassing the requirement for ordered recruitment of sequence-specific transcription factors (Agalioti et al., 2000; Hassan et al., 2002). Moreover, because histone acetylation is a common mechanism that activates transcription, HATs and BRD-containing activators can be shared amongst different genes. In this way, the cell does not have to increase the amounts of sequence-specific transcription factors for each individual activated gene. Histone acetylation is thus an efficient mechanism that maintains high transcription rates of many activated genes concurrently. Once the external stimulus ceases, deacetylation of histones occurs, and transcription is decreased, as HATs and co-activators are no longer recruited. Histone acetylation/deacetylation is a flexible regulatory mechanism, which allows plants to respond rapidly to a changing environment. Plants are sessile organisms and adapt to a changing environment by modifying their growth patterns. In Arabidopsis, maize and rice, HATs, HDACs and BRD proteins modify plant development such as rate of growth, leaf morphogenesis, root growth, flowering and seed development (see Sections 4.2.1–4.2.3). In addition, HATs and HDACs regulate the expression of many developmental genes in response to plant hormones, which are produced during abiotic and biotic stress. HATs and HDACs thus play central roles in adaptation to the environment, because they modify many aspects of plant development in response to external stimuli.
4.3
Histone methylation
Histones H3 and H4 are also modified by methylation, which involves the addition of methyl (—CH3 ) groups to the ε-amino groups of lysine residues by HMTases. Unlike acetylation, where only one acetyl group is added, the ε-amino group of lysine residues can be mono-, di- or trimethylated (Figure 4.4). Methylation of amino acid residues K4, K36 and K79 of histone H3 is generally associated with transcriptional activation, whereas methylation of amino acid residues K9 and K27 of histone H3 and K20 of histone H4 is correlated with heterochromatin formation and gene silencing (Bannister and Kouzarides, 2005; Lee et al., 2005). Because
94
REGULATION OF TRANSCRIPTION IN PLANTS
Figure 4.4 Methylation of a lysine residue. Chemical structure of lysine, monomethyl-lysine, dimethyl-lysine and trimethyl-lysine residues.
HISTONE MODIFICATIONS AND TRANSCRIPTION IN PLANTS
95
methylation does not result in charge neutralisation of lysines, and the addition of a small methyl group is not likely to alter the physical structure of chromatin, histone methylation regulates transcription by recruiting transcription factors (Strahl and Allis, 2000). Moreover, mono-, di- and trimethylation have different effects on gene regulation, possibly because modified lysine residues are recognised by different groups of proteins. For example, in Chlamydomonas, dimethylated K4 in histone H3 is correlated with transcription in euchromatin, whereas monomethylated K4 in histone H3 is associated with non-transcribed genes (van Dijk et al., 2005). In plants, methylation of H3K9 plays an important role in silencing the expression of the flowering repressor gene FLC (FLOWERING LOCUSC) during vernalisation, leading to early flowering (Bastow et al., 2004; Sung and Amasino, 2004). This topic is discussed in detail in Chapter 9.
4.3.1
Plant SET-domain proteins
HMTases are characterised by the SET domain (Pfam accession: PF00856), which is a motif of approximately 130 amino acid residues that catalyses methylation of lysine residues using the cofactors S-adenosylmethionine or S-adenosylhomocysteine (Min et al., 2002; Marmorstein, 2003). The SET domain was first described in three Drosophila chromatin-remodelling proteins: Su(var)3-9 (Suppressor of variegation 3-9), E(z) (enhancer of zeste) and Trithorax (Tschiersch et al., 1994). In solution, the SET domain of the yeast transcription factor Clr4 forms three clusters of β-sheets arranged in a triangle, forming a pocket for binding the cofactor and the N-terminal tail of histone H3 (Min et al., 2002; Marmorstein, 2003). SET-domain proteins exhibit different substrate specificity. For example, human SET7/9 methylates amino acid residue K4 of histone H3, whereas Neurospora DIM-5 methylates amino acid residue K9 of histone H3 (Wilson et al., 2002; Zhang et al., 2002). The substrate specificities of SET-domain proteins are likely to be regulated by sequences surrounding the domain (Min et al., 2002; Marmorstein, 2003). Not all SET-domain proteins function as HMTases; for example, the pea SET-domain protein Rubisco large subunit methyltransferase methylates amino acid residue K14 of the large subunit of Rubisco in chloroplasts (Trievel et al., 2002). Animal HMTases are divided into four families: E(Z), Trithorax, ASH1 (ABSENT OR SMALL HOMEOTIC DISCS1) and Su(var)3-9, based on sequence similarities. Depending on the substrate specificity, HMTases may function as transcription activators or repressors. There are 14 genes encoding SET-domain proteins in Drosophila, 17 in mouse and 4 in yeast (Springer et al., 2003). In Arabidopsis, there are a total of 41 genes that encode SET-domain proteins. Rice has 33 genes encoding SET-domain proteins, and 35 genes have been identified in maize so far (Plant Chromatin Database). Thus, plants have many more genes encoding SET-domain proteins than animals. Many plant SET-domain proteins possess PHD (plant homeodomain, Pfam accession: PF00628) (Aasland et al., 1995), PWWP (Pfam accession: PF00855) (Stec et al., 2000) and YDG (Pfam accession: PF02182) domains. The functions of these domains are unknown, although they may be involved in protein–protein interactions (Aasland et al., 1995; Stec et al., 2000). DNA-binding
96
REGULATION OF TRANSCRIPTION IN PLANTS
motifs such as zinc fingers and AT-hooks are also common in plant SET-domain proteins (Springer et al., 2003). In addition, plant SET-domain proteins possess the pre-SET motif (Pfam accession: PF05033), which is a cysteine-rich domain that is essential for the enzymatic functions of the SET domain (Rea et al., 2000; Marmorstein, 2003).
4.3.1.1
E(Z)- and Trithorax-type HMTases
In Drosophila, E(z)-type HMTases belong to the Polycomb-group (PcG) proteins, which are involved in repression of homeotic genes for specification of segmentation during embryogenesis (Francis and Kingston, 2001). In PcG mutants, spatial patterns of homeotic gene expression are established correctly, but are misregulated later in late embryogenesis, suggesting that PcG proteins maintain repression of inactivated genes through cell divisions (Francis and Kingston, 2001). In Drosophila and mammals, E(z)-type HMTases methylate amino acid residue K9 and/or K27 of histone H3, but not K4 of histone H3 (Czermin et al., 2002; Kuzmichev et al., 2002). Methylated amino acid residues K9 and K27 of histone H3 are hallmarks of heterochromatin, and are involved in transcriptional repression, because they are recognised by repressors possessing chromodomains (Nielsen et al., 2002). The chromodomain (chromatin organisation modifier, Pfam accession: PF00385) is a 60 amino acid residue module binding methylated lysine side chains, and is present in various transcriptional repressors such as the yeast, mammalian and Drosophila Heterochromatin Protein 1 (HP1), and the Drosophila Polycomb protein (Czermin et al., 2002; Maison and Almouzni, 2004). In Arabidopsis, there are three E(z)-type HMTases: SDG1 (also known as CLF for CURLY LEAF), SDG5 (also known as MEA for MEDEA or FIS1 for FERTILISATION INDEPENDENT SEED1) and SDG10. Three E(z)-type HMTases, SDG124, SDG125 and SDG126, are also predicted in rice (Springer et al., 2003). Although HMTase activities have not been demonstrated for CLF and MEA, these proteins mediate transcriptional silencing (Goodrich et al., 1997; Grossniklaus et al., 1998). CLF represses transcription of the floral homeotic gene AGAMOUS (AG) in vegetative tissues. AG is a MADS-box gene that is expressed in whorls 3 and 4 of the flower, and specifies identity of stamens and the carpel (Bowman et al., 1989). Loss-of-function mutation in CLF resulted in expression of AG in whorls 1 and 2 of flowers, and in leaves (Goodrich et al., 1997). CLF mutants possess curly leaves and exhibit homeotic transformation of sepals and petals into stamens and carpels. Like animal PcG proteins, CLF does not specify the spatial pattern of AG expression, but is required to repress its expression in cells where AG was initially silenced (Goodrich et al., 1997). MEA is an imprinted gene encoding a E(z)-type HMTase (Grossniklaus et al., 1998). Genomic imprinting is the phenomenon where a particular allele that is silenced depends on whether it is derived maternally or paternally (Moore, 2001). Because a plant may receive pollen from many sources, it is evolutionarily advantageous for the pollen-donating plant if the seed containing its embryo has welldeveloped endosperm. However, it is unfavourable for the mother plant to invest more energy than necessary in the development of a particular seed. The maternal
HISTONE MODIFICATIONS AND TRANSCRIPTION IN PLANTS
97
genome in the embryo or the endosperm silences paternal growth-promoting genes, whereas the paternal genome would inactivate maternal growth-suppressing genes (Mora-Garcia and Goodrich, 2000). MEA is maternally expressed, and represses the paternally expressed MADS-box gene PHERES1 (Kohler et al., 2005). Loss-offunction mutations in MEA result in aborted seeds, due to over-proliferation of the embryo (Grossniklaus et al., 1998; Kohler et al., 2005). Another class of plant SET-domain proteins, the Trithorax-Group (TrxG)-type HMTases, is antagonistic in function to PcG, i.e. TrxG positively regulates expression of homeotic genes for pattern specification in Drosophila (Francis and Kingston, 2001). TrxG genes are not involved in the initial activation of gene expression in progenitor cells, but are necessary for maintaining transcriptional activity in subsequent cell lineages following cell division (Francis and Kingston, 2001). TrxG-type HMTases such as the human MIXED LINEAGE LEUKAEMIA and the yeast SET1 methylate histone H3 at amino acid residue K4, which is a post-translational modification characteristic of euchromatin (Lachner and Jenuwein, 2002; Milne et al., 2002). There are five genes encoding TRITHORAX-type HMTases in Arabidopsis (SDG2, SDG14, SDG25, SDG27 and SDG30) and in rice (SDG106, SDG115, SDG116, SDG127 and SDG129) (Springer et al., 2003). Arabidopsis SDG27 (also known as ATX1 for ARABIDOPSIS TRITHORAX1) methylates histone H3 at K4 in vitro, and activates expression of several floral homeotic genes, such as AP1, AP2 and AG (Alvarez-Venegas et al., 2003). Like Drosophila TrxG proteins, ATX1 positively regulates gene expression, but does not initiate transcription in cells (AlvarezVenegas et al., 2003). In addition to homeotic transformation of floral organs, ATX1 mutants have small rosette leaves and are late flowering, suggesting that ATX1 also regulates genes that control vegetative development and flowering time (AlvarezVenegas et al., 2003). During embryogenesis in Drosophila, PcG and TrxG genes function in conjunction to maintain initially established pattern of gene expression (Hugh and Brock, 2005). PcG and TrxG genes are important in propagating cell fate, and therefore confer cell identity during formation of organs. Because methylation of amino acid residue K9 of histone H3 by PcG-type HMTases is associated with heterochromatin formation, and methylation of amino acid residue K4 of histone H3 by TrxG-type HMTases is correlated with open accessible chromatin structures, it is currently considered that PcG and TrxG confer cell memory by re-establishing the same chromatin structures at their target genes in daughter cells following cell division (Francis and Kingston, 2001; Hugh and Brock, 2005). In plants, the TrxG-type HMTase SDG27 activates AG, whereas the PcG-type HMTase CLF represses AG to define its expression boundaries to specify floral organ identity. Hence, although plants and animals have very different homeotic genes (Meyerowitz, 2002), they have evolved to utilise PcG- and TrxG-type HMTases to regulate spatial patterns of gene expression to specify formation of organs (Kohler and Grossniklaus, 2002).
4.3.1.2
Su(var)3-9-type HMTases
A total of 15 genes encoding SET-domain proteins have been predicted to belong to the Su(var)3-9 family of HMTases in Arabidopsis (Springer et al., 2003).
98
REGULATION OF TRANSCRIPTION IN PLANTS
These genes are designated SUVH1 to SUVH15. Unlike Drosophila and mammalian Su(var)3-9-type HMTases, Arabidopsis homologues do not possess chromodomain or ankyrin repeats. Ankyrin repeats are tandemly arranged modules of 33 amino acid residues involved in protein–protein interactions (Kalus et al., 1997). In animals, Su(var)3-9-type HMTases specifically methylate amino acid residue K9 of histone H3, which in turn recruits transcriptional repressors possessing chromodomains, such as HP1. HP1 is involved in organising chromatin into heterochromatin (Maison and Almouzni, 2004). Su(var)3-9-type HMTases play important roles in gene silencing and are the main regulators of heterochromatic domains in animal cells (Maison and Almouzni, 2004). In Arabidopsis, SUVH2 localises to pericentric heterochromatic regions in the interphase nucleus (Naumann et al., 2005). Loss-of-function mutations in SUVH2 led to reduction in levels of mono- and dimethylated K9 in histone H3, monoand dimethylated K27 in histone H3, and monomethylated K20 in histone H4, which are the hallmarks of heterochromatin in Arabidopsis (Naumann et al., 2005). Over-expression of SUVH2 resulted in increased amounts of heterochromatin and enhanced transcriptional gene silencing (Naumann et al., 2005). Thus, Arabidopsis SUVH2 has similar functions to its animal homologues in establishing heterochromatin regions in the nucleus. Plants over-expressing SUVH2 are smaller and possess curly leaves, suggesting that SUVH2 regulates genes involved in growth and development (Naumann et al., 2005). Over-expression of another Arabidopsis Su(var)39-type HMTase gene, SUVH1, has no effect on heterochromatin formation, and no obvious morphological phenotypes (Naumann et al., 2005). Thus, SUVH1 and SUVH2 have different functions, suggesting that different members of the Su(var)39-family of HMTases have distinct biological roles. The tobacco Su(var)3-9-type HMTase, NtSET1, methylates amino acid residues K9 and K27 of histone H3 in vitro (Yu et al., 2004). Transgenic tobacco plants that over-express NtSET1 have stunted vegetative growth, due to inhibition of cell division and expansion (Shen and Meyer, 2004). NtSET1 interacts with Like Heterochromatin Protein1 (LHP1), which is the Arabidopsis homologue of HP1, and both proteins co-localise to heterochromatin in the nuclei of tobacco cells. Thus, NtSET1 may mediate heterochromatin formation through HP1 in tobacco (Yu et al., 2004). In addition, NtSET1 regulates chromosome segregation during cell division (Shen and Meyer, 2004). The tobacco NtSET1 is likely to be a homologue of the Arabidopsis SUVH2. Arabidopsis SUVH4 (also known as KYP for KRYPTONITE) was isolated in a genetic screen for suppression of silencing of the SUPERMAN gene, and again in a screen for reactivation of the silenced PAI gene (Jackson et al., 2002; Malagnac et al., 2002). SUVH4 can mono- or dimethylate but does not trimethylate amino acid residue K9 of histone H3 in vitro (Jackson et al., 2004). Loss-of-function mutations led to decreased levels of dimethylated K9 in histone H3, and reduced CpNpG DNA methylation at retro-transposons, repetitive genomic sequences and transgenes, suggesting that DNA methylation is coupled to methylation of amino acid residue K9 of histone H3 (Jackson et al., 2002; Naumann et al., 2005). DNA methylation is likely to be linked to histone methylation through CHROMOMETHYLASE3 (CMT3),
HISTONE MODIFICATIONS AND TRANSCRIPTION IN PLANTS
99
which is a DNA methyltransferase that possesses a chromodomain, and is involved in maintenance of methylated cytosines at methylated genes (Bartee et al., 2001; Lindroth et al., 2001). SUVH4 and CMT3 may function in conjunction with ARGONAUTE1 (AGO1) in RNAi-mediated silencing of transposons (Lippman et al., 2003). In contrast to SUVH2, SUVH4 null mutants do not have significant decrease in levels of methylated K27 in histone H3 and methylated K20 in histone H4 at chromocentres, suggesting that SUVH4 is not likely to be involved in organisation of heterochromatin (Naumann et al., 2005). Thus, there appear to be at least two mechanisms of gene silencing by dimethylated K9 in histone H3 in plants, i.e. via interaction with chromodomain-containing repressors to mediate formation of heterochromatin, or through interaction with chromomethylases to maintain DNA methylation patterns.
4.3.1.3
ASH1-type and plant-specific HMTases
Arabidopsis has five genes (SDG4, SDG7, SDG8, SDG26 and SDG24) and rice has two genes (SDG102 and SDG110) that encode proteins with sequence similarity and with similar arrangements of protein domains to Drosophila and mammalian ASH1-type HMTases (Baumbusch et al., 2001; Springer et al., 2003). Arabidopsis SDG15 and SDG34 contain a PHD domain at the N-terminus, and a SET domain at the C-terminus, and are unique to plant and fungi (Springer et al., 2003). The functions of these predicted plant HMTases have not yet been studied.
4.3.2
Histone demethylase
For many years, methylation of histones H3 and H4 was regarded as a permanent form of post-translational modification, at least as compared to acetylation, because the enzymes that demethylate histones had not been characterised. However, recently, the human LSD1 (lysine-specific histone demethylase1) was shown to demethylate dimethylated K4 in histone H3 using flavin adenine dinucleotide (FAD) as a cofactor in a two-step reaction, where one methyl group was removed in each step (Shi et al., 2004). LSD1 catalyses amine oxidation, converting an N methyl group to an imine group, which is subsequently hydrolysed by water to form an amine group and a formaldehyde byproduct (Shi et al., 2004). Demethylation of amino acid residue K4 of histone H3, and not amino acid residue K9, by LSD1 is consistent with its function as a co-repressor in silencing neuronal-specific genes in non-neuronal cells (Ballas et al., 2001). However, recent work indicates that LSD1 can also function as a transcriptional activator by demethylating amino acid residue K9 of histone H3 (Metzger et al., 2005; Wysocka et al., 2005). The substrate specificity depends on the protein cofactors that associate with LSD1 (Metzger et al., 2005; Wysocka et al., 2005). LSD1 possesses four domains: a SWIRM domain (Pfam accession: PF04433), which is an 85 amino acid residue α-helical domain that is involved in protein interactions; a DAO domain, which is an FAD-dependent oxidoreductase motif (Pfam accession: PF01266); the FAD-binding domain (Pfam accession: PF01494);
100
REGULATION OF TRANSCRIPTION IN PLANTS
and the amine-oxidase domain (Pfam accession: PF01593). There are four genes in Arabidopsis (HMA1–HMA4) that have sequence similarities to LSD1 (Shi et al., 2004). Interestingly, HMA1 is FLD (FLOWERING LOCUS D), which is a repressor of the flowering repressor gene FLC. Although histone demethylase activity has not been demonstrated for FLD, loss-of-function mutations in FLD lead to increased histone H3 acetylation at FLC, resulting in activated FLC expression and a lateflowering phenotype (He et al., 2003).
4.4
Interplay between histone acetylation and methylation in transcriptional regulation
A gene can exist in a range of transcriptional states: an activated transcriptional state, a basal transcriptional state, a state competent for transcription and a silenced transcriptional state (Figure 4.5). Acetylation of histones is a mechanism that upregulates transcription from the basal to an activated state. This notion is consistent with the observation that mutations in Arabidopsis HAG1, HAF2 and GTE6 result in downregulation of gene expression, but do not eliminate transcription altogether. Moreover, as acetylation is reversible, transition from the activated state to the basal state is mediated by HDACs in a single-step reaction, and gene expression would be downregulated in the absence of stimuli. Thus, genes that are involved in cellular responses to environmental signals, such as cold-response, light-induced, pathogeninduced and stress-induced genes, are regulated by HATs and HDACs in plants. In contrast to histone acetylation, methylation at amino acid residue K4 of histone H3 is a mark for euchromatin, and is associated with competence for transcription (Schneider et al., 2004). Both transcribed and non-transcribed genes in euchromatin are associated with methylated K4 in histone H3 (Schneider et al., 2004; van Dijk et al., 2005). Over-expression of Arabidopsis SDG27, whose product methylates amino acid residue K4 of histone H3 in vitro, does not initiate transcription of target genes in cells that do not originally express these genes, suggesting that methylated K4 in histone H3 is associated with competence for transcription, and not transcription per se (Alvarez-Venegas et al., 2003). It can be envisaged that transition from the transcriptionally competent state to the basal transcriptional state involves recruitment of sequence-specific transcription factors, chromatin remodelling and assembly of the RNA polymerase machinery (Figure 4.5). In the activated transcriptional state, genes are dimethylated at amino acid residue K4 of histone H3, and acetylated at histones H3 and H4 (Osano and Ono, 2003). In the silent transcriptional state, genes are methylated, and associated with histones H3 with dimethylated K9, and are packaged into heterochromatin. Dimethylated K9 in histone H3 is recognised by HP1, which is involved in heterochromatin organisation. In addition, dimethylation of K9 in histone H3 is linked to DNA methylation; dimethylated K9 is recognised by the chromomethylase, CMT3, which methylates DNA (Bartee et al., 2001; Lindroth et al., 2001). Methylated DNA is bound by methyl-CpG-binding proteins, which associate with HDACs (Nan et al., 1998). HDACs in turn deacetylate K9 in histone H3, making the lysine residue free
HISTONE MODIFICATIONS AND TRANSCRIPTION IN PLANTS
Figure 4.5
101
Interplay between histone acetylation and methylation in transcriptional control.
for methylation (Lawrence et al., 2004). Moreover, HDACs may physically interact with HMTases, resulting in histone methylation (Czermin et al., 2001; Vaute et al., 2002). Thus, DNA methylation and dimethylation of K9 in histone H3 form a selfpropagating cycle that ensures a silenced gene stays silenced (Figure 4.5). Unlike the transition from the activated to the basal transcriptional state, which is mediated by deacetylation, transition to the silenced transcriptional state involves various epigenetic changes, including methylation of DNA, methylation of K9 in histone H3 and organisation into heterochromatin (Figure 4.5). In plants, DNA methylation and dimethylation of K9 in histone H3 are involved in long-term gene silencing
102
REGULATION OF TRANSCRIPTION IN PLANTS
that is maintained through cell division, resulting in silencing of transposon activity (Lippman and Martienssen, 2004), repression of floral homeotic genes in vegetative tissues (Goodrich et al., 1997) and silencing of rRNA gene in nucleolar dominance (Lawrence et al., 2004). Common lysine residues in histone H3 and H4 tails are subjected to acetylation or methylation for transcriptional regulation. Once acetylated, the lysine residue cannot be methylated, and vice versa. Following synthesis in the cytoplasm, histones H3 and H4 are acetylated at their N-terminal tails before they are transported into the nucleus and assembled into chromatin in animals and plants (Sobel et al., 1995; Jasencakova et al., 2000). Acetylation of histone H4 occurs at amino acid residues K5 and K12 for histone deposition into chromatin, whereas acetylation of histone H3 is less conserved (Sobel et al., 1995). HDACs play a central role in transcriptional regulation, because the acetylated N-terminal tails of histones H3 and H4 on the newly replicated chromatin are deacetylated before other post-translational modifications can occur. In addition, HDACs are involved in several aspects of transcriptional control; they regulate transition from the activated to the basal transcriptional state, and are also required for establishing silent transcriptional states (Figure 4.5). Histone deacetylation is a key event in silencing of active genes, which are associated with acetylated histones (Czermin et al., 2001). Hence, mutations in plant HDACs affect expression of pathogen-response and stress-response genes (Song et al., 2005; Zhou et al., 2005), and also abolish gene silencing (Lawrence et al., 2004). As a seedling grows into an adult plant, the chromatin structures of various genes are modified in response to cues from the developmental programmes and environmental factors, so that sets of genes are expressed at the right time and in the right place (Li et al., 2002). These epigenetic modifications have to be erased prior to or shortly after egg fertilisation, so that the newly formed zygote can start a similar life cycle again. In mammals, global chromatin remodelling such as genomewide DNA demethylation and acetylation of histones have been observed during spermatogenesis and embryogenesis (Hazzouri et al., 2000; Mhanni and McGowan, 2004). DNA methylation levels decreased by 80% during maturation of pollen grains, suggesting that global DNA demethylation also occurs during gametogenesis in plants (Oakeley et al., 1997). Future work should indicate whether the mechanisms that reset epigenetic patterns are similar in plants and animals.
4.5
Conclusions
Acetylation and methylation of histones are key mechanisms of transcriptional regulation in plants. Proteins that acetylate and methylate histones play important roles in plant development. Plants possess an open growth system, forming vegetative and floral organs continuously throughout their life cycle. Moreover, plant development is plastic, and growth patterns change with environmental factors. In contrast, animals possess a closed system of development; specification of organ formation occurs mainly during embryogenesis and follows a well-defined developmental programme, which is relatively unaffected by environmental factors. Thus, plants have
HISTONE MODIFICATIONS AND TRANSCRIPTION IN PLANTS
103
more genes encoding HATs, HDACs, BRD proteins and SET-domain proteins than yeast, Drosophila and mammals to regulate the continuous process of organogenesis, and to switch developmental fates in response to environmental signals. In addition, plants have evolved unique chromosomal proteins such as HD2-type HDACs and BRD proteins to regulate biological processes specific to plant physiology. Plant chromatin research is likely to stay at the forefront of chromatin research, because plants are good experimental systems for studying the biological functions of posttranslational modifications of histones in a multicellular organism. Currently, little is known about other post-translational modifications such as ADP-ribosylation, ubiquitination and SUMOylation, and modifications of residues such as proline, serine and arginine in histones. Furthermore, the functions of many plant transcription factors involved in post-translational modifications of histones have not yet been investigated, and future work will resolve the role of each individual transcription factor in plant development.
References R. Aasland, T.J. Gibson and A.F. Stewart (1995) The PHD finger: implications for chromatin-mediated transcriptional regulation. Trends in Biochemical Sciences 20, 56–59. T. Agalioti, S. Lomvardas, B. Parekh, J. Yie, T. Maniatis and D. Thanos (2000) Ordered recruitment of chromatin modifying and general transcription factors to the IFN-β promoter. Cell 103, 667–678. R. Alvarez-Venegas, S. Pien, M. Sadder, X. Witmer, U. Grossniklaus and Z. Avramova (2003) ATX1, an Arabidopsis homolog of Trithorax, activates flower homeotic genes. Current Biology 13, 627–637. W. Aufsatz, M.F. Mette, J. van der Winden, M. Matzke and A.J.M. Matzke (2002) HDA6, a putative histone deacetylase needed to enhance DNA methylation induced by double-stranded RNA. The EMBO Journal 21, 6832–6841. R. Balasubramanian, M.G. Pray-Grant, W. Selleck, P.A. Grant and S. Tan (2002) Role of the Ada2 and Ada3 transcriptional coactivators in histone acetylation. The Journal of Biological Chemistry 277, 7989–7995. N. Ballas, E. Battaglioli, F. Atouf, et al. (2001) Regulation of neuronal traits by a novel transcriptional complex. Neuron 31, 353–365. A.J. Bannister and T. Kouzarides (2005) Reversing histone methylation. Nature 436, 1103–1106. L. Bartee, F. Malagnac and J. Bender (2001) Arabidopsis CMT3 chromomethylase mutations block non-CG methylation and silencing of an endogenous gene. Genes and Development 15, 1753– 1758. R. Bastow, J.S. Mylne, C. Lister, Z. Lippman, R.A. Martienssen and C. Dean (2004) Vernalization requires epigenetic silencing of FLC by histone methylation. Nature 427, 164–167. W.R. Bauer, J.J. Hayes, J.H. White and A.P. Wolffe (1994) Nucleosome structural changes due to acetylation. Journal of Molecular Biology 236, 685–690. L.O. Baumbusch, T. Thorstensen, V. Krauss, et al. (2001) The Arabidopsis thaliana genome contains at least 29 active genes encoding SET domain proteins that can be assigned to four evolutionarily conserved classes. Nucleic Acids Research 29, 4319–4333. S.G. Bavykin, S.I. Usachenko, A.I. Lishanskaya, et al. (1985) Primary organization of nucleosomal core particles is invariable in repressed and active nuclei from animal, plant and yeast cells. Nucleic Acids Research 13, 3439–3459. G. Benvenuto, F. Formiggini, P. Laflamme, M. Malakhov and C. Bowler (2002) The photomorphogenesis regulator DET1 binds the amino-terminal tail of histone H2B in a nucleosome context. Current Biology 12, 1529–1534.
104
REGULATION OF TRANSCRIPTION IN PLANTS
C. Bertrand, M. Benhamed, Y.F. Li, et al. (2005) Arabidopsis HAF2 gene encoding TATA-binding protein (TBP)-associated factor TAF1, is required to integrate light signals to regulate gene expression and growth. The Journal of Biological Chemistry 280, 1465–1473. C. Bertrand, C. Bergounioux, S. Domenichini, M. Delarue and D.X. Zhou (2003) Arabidopsis HAT AtGCN5 regulates the floral meristem activity through the WUSCHEL/AGAMOUS pathway. The Journal of Biological Chemistry 278, 28246–28251. R.A. Bhat, J.W. Borst, M. Riehl and R.D. Thompson (2004) Interaction of maize Opaque-2 and the transcriptional co-activators GCN5 and ADA2, in the modulation of transcriptional activity. Plant Molecular Biology 55, 239–252. R.A. Bhat, M. Riehl, G. Santandrea, et al. (2003) Alteration of GCN5 levels in maize reveals dynamic responses to manipulating histone acetylation. The Plant Journal 33, 455–469. A.W. Bird, D.Y. Yu, M.G. Pray-Grant, et al. (2002) Acetylation of histone H4 by Esa1 is required for DNA double-strand break repair. Nature 419, 411–415. A.M. Bode and Z. Dong (2004) Post-translational modification of p53 in tumorigenesis. Nature Reviews Cancer 4, 793–805. L. Bordoli, M. Netsch, U. L¨uthi, W. Lutz and R. Eckner (2001) Plant orthologs of p300/CBP: conservation of a core domain in metazoan p300/CBP acetyltransferase-related proteins. Nucleic Acids Research 29, 589–597. J.L. Bowman, D.R. Smyth and E.M. Meyerowitz (1989) Genes directing flower development in Arabidopsis. The Plant Cell 1, 37–52. M. Braunstein, R. Sobel, C. Allis, B. Turner and J. Broach (1996) Efficient transcriptional silencing in Saccharomyces cerevisiae requires a heterochromatin histone acetylation pattern. Molecular and Cellular Biology 16, 4349–4356. J.E. Brownell and C.D. Allis (1996) Special HATs for special occasions: linking histone acetylation to chromatin assembly and gene activation. Current Opinion in Genetics and Development 6, 176–184. M.E. Byrne, J. Simorowski and R.A. Martienseen (2002) ASYMMETRIC LEAVES1 reveals KNOX gene redundancy in Arabidopsis. Development 29, 1957–1965. P.D. Cary, C. Crane-Robinson, E.M. Bradbury and G.H. Dixon (1982) Effect of acetylation on the binding of N-terminal peptides of histone H4 to DNA. European Journal of Biochemistry 127, 137–143. Z.J. Chen and C.S. Pikaard (1997) Epigenetic silencing of RNA polymerase I transcription: a role for DNA methylation and histone modification in nucleolar dominance. Genes and Development 16, 2124–2136. P. Chua and S.G. Roeder (1995) Bdf1, a yeast chromosomal protein required for sporulation. Molecular and Cellular Biology 15, 3685–3696. Y.L. Chua, A.P.C. Brown and J.C. Gray (2001) Targeted histone acetylation and altered nuclease accessibility over short regions of the pea plastocyanin gene. The Plant Cell 13, 599–612. Y.L. Chua, S. Channeliere, E. Mott and J.C. Gray (2005) The bromodomain protein GTE6 controls leaf development in Arabidopsis by histone acetylation at ASYMMETRIC LEAVES1. Genes and Development 19, 2245–2254. Y.L. Chua, E. Mott, A.P. Brown, D. MacLean and J.C. Gray (2004) Microarray analysis of chromatinimmunoprecipitated DNA identifies specific regions of tobacco genes associated with acetylated histones. The Plant Journal 37, 789–800. Y.L. Chua, L.A. Watson and J.C. Gray (2003) The transcriptional enhancer of the pea plastocyanin gene associates with the nuclear matrix and regulates gene expression through histone acetylation. The Plant Cell 15, 1468–1479. B. Czermin, R. Melfi, D. McCabe, V. Seitz, A. Imhof and V. Pirrotta (2002) Drosophila enhancer of Zeste/ESC complexes have a histone H3 methyltransferase activity that marks chromosomal Polycomb sites. Cell 111, 185–196. B. Czermin, G. Schotta, B.B. Hulsmann, et al. (2001) Physical and functional association of SU(VAR)3-9 and HDAC1 in Drosophila. EMBO Reports 2, 915–919. J. Deckert and K. Struhl (2001) Histone acetylation at promoters is differentially affected by specific activators and repressors. Molecular and Cellular Biology 21, 2726–2735.
HISTONE MODIFICATIONS AND TRANSCRIPTION IN PLANTS
105
R.J. deLange, D.M. Fambrough, E.L. Smith and J. Bonner (1968) Calf and pea histone IV. II: The complete amino acid sequence of calf thymus histone IV; presence of ε-N -acetyllysine. The Journal of Biological Chemistry 244, 319–334. C. Dhalluin, J.E. Carlson, L. Zeng, et al. (1999) Structure and ligand of a histone acetyltransferase bromodomain. Nature 399, 491–496. P. Duque and N.-H. Chua (2003) IMB1, a bromodomain protein induced during seed imbibition, regulates ABA- and phyA-mediated responses of germination in Arabidopsis. The Plant Journal 35, 787–799. W.C. Earnshaw, B.M. Honda, R.A. Laskey and J.O. Thomas (1980) Assembly of nucleosomes: the reaction involving X. laevis nucleoplasmin. Cell 21, 373–383. D.D. Edberg, J.E. Bruce, W.F. Siems and R. Reeves (2004) In vivo post-translational modifications of the high mobility group A1a proteins in breast cancer cells of differing metastatic potential. Biochemistry 43, 11500–11515. B. Florence and D.V. Faller (2001) You bet-cha: a novel family of transcriptional regulators. Frontiers in Bioscience 1, D1008–D1018. N.J. Francis and R.E. Kingston (2001) Mechanisms of transcriptional memory. Nature Reviews Molecular Cell Biology 2, 409–421. M.-J. Gao, I.A.P. Parkin, D.J. Lydiate and A. Hannoufa (2004) An auxin-responsive SCARECROWlike transcriptional activator interacts with histone deacetylase. Plant Molecular Biology 55, 417–431. M.-J. Gao, U.A. Schafer, I.A.P. Parkin, D.D. Hegedus, D.J. Lydiate and A. Hannoufa (2003) A novel protein form Brassica napus has a putative KID domain and responds to low temperature. The Plant Journal 33, 1073–1086. E.B. Gomez, J.M. Espinosa and S.L. Forsburg (2005) Schizosaccharomyces pombe mst2+ encodes a MYST family histone acetyltransferase that negatively regulates telomere silencing. Molecular and Cellular Biology 25, 8887–8903. J. Goodrich, P. Puangsomlee, M. Martin, D. Long, E.M. Meyerowitz and G. Coupland (1997) A polycomb-group gene regulates homeotic gene expression in Arabidopsis. Nature 386, 44–51. P.A. Grant, L. Duggan, J. Cote, et al. (1997) Yeast Gcn5 functions in two multisubunit complexes to acetylate nucleosomal histones: characterization of an Ada complex and the SAGA (Spt/Ada) complex. Genes and Development 11, 1640–1650. U. Grossniklaus, J.-P. Vielle-Calzada, M.A. Hoeppner and W.B. Gagliano (1998) Maternal control of embryogenesis by MEDEA, a Polycomb Group gene in Arabidopsis. Science 280, 446– 450. C.M. Grozinger, E.D. Chao, H.E. Blackwell, D. Moazed and S.L. Schreiber (2001) Identification of a class of small molecule inhibitors of the sirtuin family of NAD-dependent deacetylases by phenotypic screening. The Journal of Biological Chemistry 276, 38837–38843. A.H. Hassan, K.E. Neely and J.L. Workman (2001) Histone acetyltransferase complexes stabilize SWI/SNF binding to promoter nucleosomes. Cell 104, 817–827. A.H. Hassan, P. Prochasson, K.E. Neely, et al. (2002) Function and selectivity of bromodomains in anchoring chromatin-modifying complexes to promoter nucleosomes. Cell 111, 369–379. M. Hazzouri, C. Pivot-Pajot, A.K. Faure, et al. (2000) Regulated hyperacetylation of core histones during mouse spermatogenesis: involvement of histone deacetylases. European Journal of Cell Biology 79, 950–960. Y. He, S.D. Michaels and R.M. Amasino (2003) Regulation of flowering time by histone acetylation in Arabidopsis. Science 302, 1751–1754. J.S. Heslop-Harrison (2003) Planning for remodelling: nuclear architecture, chromatin and chromosomes. Trends in Plant Science 8, 195–197. A. Houben, D. Demidov, T. Rutten and K.H. Scheidtmann (2005) Novel phosphorylation of histone H3 at threonine 11 that temporally correlates with condensation of mitotic and meiotic chromosomes in plant cells. Cytogenetic and Genome Research 109, 148–155. D. Houzelstein, S.L. Bullock, D.E. Lynch, E.F. Grigorieva, V.A. Wilson and R.S.P. Beddington (2002) Growth and early post-implantation defects in mice deficient for the bromodomain-containing protein Brd4. Molecular and Cellular Biology 22, 3794–3802.
106
REGULATION OF TRANSCRIPTION IN PLANTS
W. Hugh and C.L.F. Brock (2005) Maintenance of gene expression patterns. Developmental Dynamics 232, 633–655. J.P. Jackson, L. Johnson, Z. Jasencakova, et al. (2004) Dimethylation of histone H3 lysine 9 is a critical mark for DNA methylation and gene silencing in Arabidopsis thaliana. Chromosoma 112, 308–315. J.P. Jackson, A.M. Lindroth, X. Cao and S.E. Jacobsen (2002) Control of CpNpG DNA methylation by the KRYPTONITE histone H3 methyltransferase. Nature 416, 556–560. I.-C. Jang, Y.-M. Pahk, S.I. Song, H.J. Kwon, B.H. Nahm and J.-K. Kim (2003) Structure and expression of the rice class-I type histone deacetylase genes OsHDAC1–3: OsHDAC1 overexpression in transgenic plants leads to increased growth rate and altered architecture. The Plant Journal 33, 531–541. Z. Jasencakova, A. Meister and I. Schubert (2001) Chromatin organization and its relation to replication and histone acetylation during the cell cycle in barley. Chromosoma 110, 83–92. Z. Jasencakova, A. Meister, J. Walter, B.M. Turner and I. Schubert (2000) Histone H4 acetylation of euchromatin and heterochromatin is cell cycle dependent and correlated with replication rather than with transcription. The Plant Cell 12, 2087–2100. F. Jeanmougin, J.-M. Wurtz, B. Le Douarin, P. Chambon and R. Losson (1997) The bromodomain revisited. Trends in Biochemical Sciences 22, 151–153. P. Jeppesen and B.M. Turner (1993) The inactive X chromosome in female mammals is distinguished by a lack of histone H4 acetylation, a cytogenetic marker for gene expression. Cell 74, 281–289. J.T. Kadonaga (1998) Eukaryotic transcription: an interlaced network of transcription factors and chromatin-modifying machines. Cell 92, 307–313. W. Kalus, R. Baumgartner, C. Renner, et al. (1997) NMR structural characterization of the CDK inhibitor p19INK4d. FEBS Letters 401, 127–132. T. Kanno, Y. Kanno, R.M. Siegel, M.K. Jang, M.J. Lenardo and K. Ozato (2004) Selective recognition of acetylated histones by bromodomain proteins visualized in living cells. Molecular Cell 13, 33–43. R.A. Kerstetter and R.S. Poethig (1998) The specification of leaf identity during shoot development. Annual Review of Cell and Developmental Biology 14, 373–398. C. Kohler and U. Grossniklaus (2002) Epigenetic inheritance of expression states in plant development: the role of Polycomb group proteins. Current Opinion in Cell Biology 14, 773–779. C. Kohler, D.R. Page, V. Gagliardini and U. Grossniklaus (2005) The Arabidopsis thaliana MEDEA Polycomb group protein controls expression of PHERES1 by parental imprinting. Nature Genetics 37, 28–30. R.D. Kornberg (1974) Chromatin structure: a repeating unit of histones and DNA. Science 184, 868– 871. R.D. Kornberg and Y. Lorch (1999) Twenty-five years of the nucleosome, fundamental particle of eukaryote chromosome. Cell 98, 285–294. S.K. Kurdistani and M. Grunstein (2003) Histone acetylation and deacetylation in yeast. Nature Reviews Molecular Cell Biology 4, 276–284. T. Kusch, S. Guelman, S.M. Abmayr and J.L. Workman (2003) Two Drosophila Ada2 homologues function in different multiprotein complexes. Molecular and Cellular Biology 23, 3305–3319. A. Kuzmichev, K. Nishioka, H. Erdjument-Bromage, P. Tempst and D. Reinberg (2002) Histone methyltransferase activity associated with a human multiprotein complex containing the enhancer of Zeste protein. Genes and Development 16, 2893–2905. M. Lachner and T. Jenuwein (2002) The many faces of histone lysine methylation. Current Opinion in Cell Biology 14, 286–298. A.G. Ladurner, C. Inouye, R. Jain and R. Tjian (2003) Bromodomains mediate an acetyl-histone encoded antisilencing function at heterochromatin boundaries. Molecular Cell 11, 365–376. R.J. Lawrence, K. Earley, O. Pontes, et al. (2004) A concerted DNA methylation/histone methylation switch regulates rRNA gene dosage control and nucleolar dominance. Molecular Cell 13, 599– 609. E.J.R. Lawson and R.S. Poethig (1995) Shoot development in plants: time for a change. Trends in Genetics 11, 263–265.
HISTONE MODIFICATIONS AND TRANSCRIPTION IN PLANTS
107
P.J. Laybourn and J.T. Kadonaga (1991) Role of nucleosomal cores and histone H1 in regulation of transcription by RNA polymerase II. Science 254, 238–245. D.Y. Lee, J.J. Hayes, D. Pruss and A.P. Wolffe (1993) A positive role for histone acetylation in transcription factor access to nucleosomal DNA. Cell 72, 73–84. D.Y. Lee, C. Teyssier, B.D. Strahl and M.R. Stallcup (2005) Role of protein methylation in regulation of transcription. Endocrine Reviews 26, 147–170. G. Li, K.J. Bishop, M.B. Chandrasekharan and T.C. Hall (1999) β-Phaseolin gene activation is a two-step process: PvALF-facilitated chromatin modification followed by abscisic acid-mediated gene activation. Proceedings of the National Academy of Sciences of the United States of America 96, 7104–7109. G. Li, S.P. Chandler, A.P. Wolffe and T.C. Hall (1998) Architectural specificity in chromatin structure at the TATA box in vivo: nucleosome displacement upon the β-phaseolin gene activation. Proceedings of the National Academy of Sciences of the United States of America 95, 4772–4777. G. Li, T.C. Hall and R. Holmes-Davis (2002) Plant chromatin: development and gene control. Bioessays 24, 234–243. Y. Lin, C.M. Fletcher, J. Zhou, C.D. Allis and G. Wagner (1999) Solution structure of the catalytic domain of GCN5 histone acetyltransferase bound to coenzyme A. Nature 400, 86–89. A.M. Lindroth, X. Cao, J.P. Jackson, et al. (2001) Requirement of CHROMOMETHYLASE3 for maintenance of CpXpG methylation. Science 292, 2077–2080. Z. Lippman and R. Martienssen (2004) The role of RNA interference in heterochromatic silencing. Nature 431, 364–370. Z. Lippman, B. May, C. Yordan, T. Singer and R. Martienssen (2003) Distinct mechanisms determine transposon inheritance and methylation via small interfering RNA and histone modification. PLoS Biology 1, 420–428. P. Loidl (2004) A plant dialect of the histone language. Trends in Plant Science 9, 84–90. K. Luger (2002) The tail does not always wag the dog. Nature Genetics 32, 221–222. K. Luger (2003) Structure and dynamic behavior of nucleosomes. Current Opinion in Genetics and Development 13, 127–135. K. Luger, A.W. M¨ader, R.K. Richmond, D.F. Sargent and T.J. Richmond (1997) Crystal structure of ˚ resolution. Nature 389, 251–260. the nucleosome core particle at 2.8 A A. Lusser, G. Brosch, A. Loidl, H. Haas and P. Loidl (1997) Identification of maize histone deacetylase HD2 as an acidic nucleolar phosphoprotein. Science 277, 88–91. Z. Lygerou, C. Conesa, P. Lesage, et al. (1994) The yeast BDF1 gene encodes a transcription factor involved in the expression of a broad class of genes including snRNA. Nucleic Acids Research 22, 5332–5340. C. Maison and G. Almouzni (2004) HP1 and the dynamics of heterochromatin maintenance. Nature Reviews Molecular Cell Biology 5, 296–305. F. Malagnac, L. Bartee and J. Bender (2002) An Arabidopsis SET domain protein required for maintenance but not establishment of DNA methylation. The EMBO Journal 21, 6842–6852. R. Marmorstein (2003) Structure of SET domain proteins: a new twist on histone methylation. Trends in Biochemical Sciences 28, 59–62. A.E. Martinez de Alba, R. Sagesser, M. Tabler and M. Tsagris (2003) A bromodomain-containing protein from tomato specifically binds potato spindle tuber viroid RNA in vitro and in vivo. Journal of Virology 77, 9685–9694. J.D. McGhee and G. Felsenfeld (1980) Nucleosome structure. Annual Review of Biochemistry 49, 1115–1156. E. Metzger, M. Wissmann, N. Yin, et al. (2005) LSD1 demethylates repressive histone marks to promote androgen-receptor-dependent transcription. Nature 437, 436–439. E.M. Meyerowitz (2002) Plants compared to animals: the broadest comparative study of development. Science 295, 1482–1485. A.A. Mhanni and R.A. McGowan (2004) Global changes in genomic methylation levels during early development of the zebrafish embryo. Development Genes and Evolution 214, 412–417. T.A. Milne, S.D. Briggs, H.W. Brock, et al. (2002) MLL targets SET domain methyltransferase activity to Hox gene promoters. Molecular Cell 10, 1107–1117.
108
REGULATION OF TRANSCRIPTION IN PLANTS
J. Min, X. Zhang, X. Cheng, S.I.S. Grewal and R.-M. Xu (2002) Structure of the SET domain histone lysine methyltransferase Clr4. Nature Structural Biology 9, 828–832. C.A. Mizzen, X.J. Yang, T. Kokubo, et al. (1996) The TAF(II)250 subunit of TFIID has histone acetyltransferase activity. Cell 87, 1261–1270. M. Montecino, B. Frenkel, A.J. van Wijnen, J.B. Lian, G.S. Stein and J.L. Stein (1999) Chromatin hyperacetylation abrogates vitamin D-mediated transcriptional upregulation of the tissue-specific osteocalcin gene in vivo. Biochemistry 38, 1338–1345. T. Moore (2001) Genetic conflict, genomic imprinting and establishment of the epigenotype in relation to growth. Reproduction 122, 185–193. S. Mora-Garcia and J. Goodrich (2000) Genomic imprinting: seeds of conflict. Current Biology 10, R71–R74. J. Murfett, X.-J. Wang, G. Hagen and T.J. Guilfoyle (2001) Identification of Arabidopsis histone deacetylase HDA6 mutants that affect transgene expression. The Plant Cell 13, 1047–1061. V. Mutskov, D. Gerber, D. Angelov, J. Ausio, J. Workman and S. Dimitrov (1998) Persistent interactions of core histone tails with nucleosomal DNA following acetylation and transcription factor binding. Molecular and Cellular Biology 18, 6293–6304. X. Nan, H.H. Ng, C.A. Johnson, et al. (1998) Transcriptional repression by the methyl-CpG-binding protein MeCP2 involves a histone deacetylase complex. Nature 393, 386–389. K. Naumann, A. Fischer, I. Hofmann, et al. (2005) Pivotal role of AtSUVH2 in heterochromatic histone methylation and gene silencing in Arabidopsis. The EMBO Journal 24, 1418–1429. P.R. Nielsen, D. Nietlispach, H.R. Mott, et al. (2002) Structure of the HP1 chromodomain bound to histone H3 methylated at lysine 9. Nature 416, 103–107. X. Niu, C.C. Adams, J.L. Workman and M.J. Guiltinan (1996) Binding of the wheat basic leucine zipper protein EmBP-1 to the nucleosomal binding sites is modulated by nucleosome positioning. The Plant Cell 8, 1569–1587. V.G. Norton, B.S. Imai, P. Yau and E.M. Bradbury (1989) Histone acetylation reduces nucleosome core particle linking number change. Cell 57, 449–457. E.J. Oakeley, A. Podesta and J.P. Jost (1997) Developmental changes in DNA methylation of the two tobacco pollen nuclei during maturation. Proceedings of the National Academy of Sciences of the United States of America 94, 11721–11725. S. Osada, A. Sutton, N. Muster, et al. (2001) The yeast SAS (something about silencing) protein complex contains a MYST-type putative acetyltransferase and functions with chromatin assembly factor ASF1. Genes and Development 15, 3155–3168. K. Osano and M. Ono (2003) State of histone modification in the rat Ig-β/growth hormone locus. European Journal of Biochemistry 270, 2532–2539. D.J. Owen, P. Ornaghi, J.-C. Yang, et al. (2000) The structural basis for the recognition of acetylated histone H4 by the bromodomain of histone acetyltransferase Gcn5p. The EMBO Journal 19, 6141–6149. R. Pandey, A. Muller, C.A. Napoli, et al. (2002) Analysis of histone acetyltransferase and histone deacetylase families of Arabidopsis thaliana suggests functional diversification of chromatin modification among multicellular eukaryotes. Nucleic Acids Research 30, 5036–5055. M.L. Pardue and W. Hennig (1990) Heterochromatin: junk or collectors item? Chromosoma 100, 3–7. L. Patthy and E.L. Smith (1973) Histone III. V: The amino acid sequence of pea embryo histone III. The Journal of Biological Chemistry 248, 6834–6840. S. Perrod and S.M. Gasser (2003) Long-range silencing and position effects at telomeres and centromeres: parallels and differences. Cellular and Molecular Life Sciences 60, 2303–2318. A.V. Probst, M. Fagard, F. Proux, et al. (2004) Arabidopsis histone deacetylase HDA6 is required for maintenance of transcriptional gene silencing and determines nuclear organization of rDNA repeats. The Plant Cell 16, 1021–1034. O.M. Puig, E. Belles, G. Lopez-Rodas, R. Sendra and V. Tordera (1998) Interaction between N-terminal domain of H4 and DNA is regulated by the acetylation degree. Biochimica et Biophysica Acta 1397, 79–90.
HISTONE MODIFICATIONS AND TRANSCRIPTION IN PLANTS
109
S. Rea, F. Eisenhaber, D. O’Carroll, et al. (2000) Regulation of chromatin structure by site-specific histone H3 methyltransferases. Nature 406, 593–599. G. Reuter and P. Spierer (1992) Position effect variegation and chromatin proteins. Bioessays 14, 605–612. A.R. Ricci, J. Genereaux and C.J. Brand (2002) Components of the SAGA histone acetyltransferase complex are required for repressed transcription of ARG1 in rich medium. Molecular and Cellular Biology 22, 4033–4042. T.J. Richmond, J.T. Finch, B. Rushton, D. Rhodes and A. Klug (1984) Structure of the nucleosome ˚ resolution. Nature 311, 532–537. core particle at 7A J.R. Rojas, R.C. Trievel, J. Zhou, et al. (1999) Structure of Tetrahymena GCN5 bound to coenzyme A and a histone H3 peptide. Nature 401, 93–98. V. Rossi, S. Locatelli, C. Lanzanova, et al. (2003) A maize histone deacetylase and retinoblastomarelated protein physically interact and cooperate in repressing gene transcription. Plant Molecular Biology 51, 401–413. S.E. Rundlett, A.A. Carmen, N. Suka, B.M. Turner and M. Grunstein (1998) Transcriptional repression of UME6 involves deacetylation of lysine 5 of histone H4 by RPD3. Nature 392, 381–385. S. Ruppert, E.H. Wang and R. Tjian (1993) Cloning and expression of human TAFII250: a TBPassociated factor implicated in cell-cycle regulation. Nature 362, 175–179. R. Schneider, A.J. Bannister, F.A. Myers, A.W. Thorne, C. Crane-Robinson and T. Kouzarides (2004) Histone H3 lysine 4 methylation patterns in higher eukaryotic genes. Nature Cell Biology 6, 73–77. D.F. Schroeder, M. Gahrtz, B.B. Maxwell, et al. (2002) De-etiolated 1 and damaged DNA binding protein 1 interact to regulate Arabidopsis photomorphogenesis. Current Biology 12, 1462–1472. L.A. Sheldon, M. Becker and C.L. Smith (2001) Steroid hormone receptor-mediated histone deacetylation and transcription at the mouse mammary tumor virus promoter. The Journal of Biological Chemistry 276, 32423–32426. W.-H. Shen and D. Meyer (2004) Ectopic expression of the NtSET1 histone methyltransferase inhibits cell expansion, and affects cell division and differentiation in tobacco plants. Plant and Cell Physiology 45, 1715–1719. Y. Shi, F. Lan, C. Matson, et al. (2004) Histone demethylation mediated by the nuclear amine oxidase homolog LSD1. Cell 119, 941–953. R.T. Simpson, S.Y. Roth, R.H. Morse, et al. (1993) Nucleosome positioning and transcription. Cold Spring Harbor Symposia on Quantitative Biology 58, 237–245. R.E. Sobel, R.G. Cook, C.A. Perry, A.T. Annunziato and C.D. Allis (1995) Conservation of depositionrelated acetylation sites in newly synthesized histones H3 and H4. Proceedings of the National Academy of Sciences of the United States of America 92, 1237–1241. C.-P. Song, M. Agarwal, M. Ohta, et al. (2005) Role of an Arabidopsis AP2/EREBP-type transcriptional repressor in abscisic acid and drought stress responses. The Plant Cell 17, 2384–2396. N.M. Springer, C.A. Napoli, D.A. Selinger, et al. (2003) Comparative analysis of SET domain proteins in maize and Arabidopsis reveals multiple duplications preceding the divergence of monocots and dicots. Plant Physiology 132, 907–925. I. Stec, S.B. Nagl, G.-J.B. van Ommen and J.T. den Dunnen (2000) The PWWP domain: a potential protein–protein interaction domain in nuclear proteins influencing differentiation? FEBS Letters 473, 1–5. D.E. Sterner and S.L. Berger (2000) Acetylation of histones and transcription-related factors. Microbiology and Molecular Biology Reviews 64, 435–459. E.J. Stockinger, Y. Mao, M.K. Regier, S.J. Triezenberg and M.F. Thomashow (2001) Transcriptional adaptor and histone acetyltransferase proteins in Arabidopsis and their interactions with CBF1, a transcriptional activator involved in cold-regulated gene expression. Nucleic Acids Research 29, 1524–1533. B.D. Strahl and C.D. Allis (2000) The language of covalent histone modifications. Nature 403, 41–45. S. Sung and R.M. Amasino (2004) Vernalization and epigenetics: how plants remember winter. Current Opinion in Plant Biology 7, 4–10.
110
REGULATION OF TRANSCRIPTION IN PLANTS
J. Taunton, C.A. Hassig and S.L. Schreiber (1996) A mammalian histone deacetylase related to the yeast transcriptional regulator Rpd3p. Science 272, 408–411. G. Theodoris, N. Inada and M. Freeling (2003) Conservation and molecular dissection of ROUGH SHEATH2 and ASYMMETRIC LEAVES1 function in leaf development. Proceedings of the National Academy of Sciences of the United States of America 100, 6837–6842. L. Tian and Z.J. Chen (2001) Blocking histone deacetylation in Arabidopsis induces pleiotropic effects on plant gene regulation and development. Proceedings of the National Academy of Sciences of the United States of America 98, 200–205. L. Tian, M.P. Fong, J.J. Wang, et al. (2005) Reversible histone acetylation and deacetylation mediate genome-wide, promoter-dependent and locus-specific changes in gene expression during plant development. Genetics 169, 337–345. L. Tian, J. Wang, M.P. Fong, et al. (2003) Genetic control of developmental changes induced by disruption of Arabidopsis histone deacetylase 1 (AtHD1) expression. Genetics 165, 399–409. R.C. Trievel, B.M. Beach, L.M.A. Dirk, R.L. Houtz and J.H. Hurley (2002) Structure and catalytic mechanism of a SET domain protein methyltransferase. Cell 111, 91–103. B. Tschiersch, A. Hofmann, V. Krauss, R. Dorn, G. Korge and G. Reuter (1994) The protein encoded by the Drosophila position-effect variegation suppressor gene Su(var)3–9 combines domains of antagonistic regulators of homeotic gene complexes. The EMBO Journal 13, 3822–3831. C. Tse, T. Sera, A.P. Wolffe and J.C. Hansen (1998) Disruption of higher-order folding by core histone acetylation dramatically enhances transcription of nucleosomal arrays by RNA polymerase III. Molecular and Cellular Biology 18, 4629–4638. K. van Dijk, K.E. Marley, B.-R. Jeong, et al. (2005) Monomethyl histone H3 lysine 4 as an epigenetic mark for silenced euchromatin in Chlamydomonas. The Plant Cell 17, 2439–2453. O. Vaute, E. Nicolas, L. Vandel and D. Trouche (2002) Functional and physical interaction between the histone methyl transferase Suv39H1 and histone deacetylases. Nucleic Acids Research 30, 475–481. K.E. Vlachonasios, M.F. Thomashow and S.J. Triezenberg (2003) Disruption mutations of ADA2b and GCN5 transcriptional adaptor genes dramatically affect Arabidopsis growth, development, and gene expression. The Plant Cell 15, 626–638. T. Wako, M. Fukuda, R. Furushima-Shimogawara, N.D. Belyaev and K. Fukui (2002) Cell cycledependent and lysine residue-specific dynamic changes of histone H4 acetylation in barley. Plant Molecular Biology 49, 645–653. J.R. Wilson, C. Jing, P.A. Walker, et al. (2002) Crystal structure and functional analysis of the histone methyltransferase SET7/9. Cell 111, 105–115. A. Wolffe (ed.) (1995) Chromatin Structure and Function. Academic Press, London. A.P. Wolffe and J.J. Hayes (1999) Chromatin disruption and modification. Nucleic Acids Research 27, 711–720. J.L. Workman and R.E. Kingston (1998) Alteration of nucleosome structure as a mechanism of transcriptional regulation. Annual Review of Biochemistry 67, 545–579. K. Wu, K. Malik, L. Tian, D. Brown and B. Miki (2000a) Functional analysis of a RPD3 histone deacetylase homologue in Arabidopsis thaliana. Plant Molecular Biology 44, 167–176. K. Wu, L. Tian, K. Malik, D. Brown and B. Miki (2000b) Functional analysis of HD2 histone deacetylase homologues in Arabidopsis thaliana. The Plant Journal 22, 19–27. K. Wu, L. Tian, C. Zhou, D. Brown and B. Miki (2003) Repression of gene expression by Arabidopsis HD2 histone deacetylases. The Plant Journal 34, 241–247. J. Wysocka, T.A. Milne and C.D. Allis (2005) Taking LSD1 to a new high. Cell 122, 654–658. M. Xu, R.T. Simpson and M.P. Kladde (1998) Gal4p-mediated chromatin remodeling depends on binding site position in nucleosomes but does not require DNA replication. Molecular and Cellular Biology 18, 1201–1212. X.-J. Yang (2004) Lysine acetylation and the bromodomain: a new partnership for signaling. Bioessays 26, 1076–1087. Y. Yu, A. Dong and W.-H. Shen (2004) Molecular characterization of the tobacco SET domain protein NtSET1 unravels its role in histone methylation, chromatin binding, and segregation. The Plant Journal 40, 699–711.
HISTONE MODIFICATIONS AND TRANSCRIPTION IN PLANTS
111
L.W. Yuan and A. Giordano (2002) Acetyltransferase machinery conserved in p300/CBP-family proteins. Oncogene 21, 2253–2260. X. Zhang, H. Tamaru, S.I. Khan, et al. (2002) Structure of the Neurospora SET domain protein DIM-5, a histone H3 lysine methyltransferase. Cell 111, 117–127. C. Zhou, H. Labbe, S. Sridha, et al. (2004) Expression and function of HD2-type histone deacetylases in Arabidopsis development. The Plant Journal 38, 715–724. C. Zhou, L. Zhang, J. Duan, B. Miki and K. Wu (2005) HISTONE DEACETYLASE19 is involved in jasmonic acid and ethylene signaling of pathogen response in Arabidopsis. The Plant Cell 17, 1196–1204.
5 Chromatin remodeling and histone variants in transcriptional regulation and in maintaining DNA methylation J.C. Reyes, J. Brzeski and A. Jerzmanowski
5.1
Introduction
The days when transcription activation was mostly about DNA-binding proteins recruiting RNA polymerases are not too far away (Goodrich et al., 1996; Ptashne and Gann, 1997). However, during the 1990s a combination of genetic and biochemical studies demonstrated that chromatin structure plays a crucial role in eukaryotic transcription (for a recent review, see Mellor, 2005). Far from being a mere passive structure designed for DNA package inside the nucleus, chromatin is a dynamic frame with essential functions in transcriptional activation, repression, and memory of expression patterns. It is now clear that changes in the chromatin structure of promoters and enhancer regions, what is generally known as ‘chromatin remodeling’, determine transcriptional control in a natural chromosomal context. The building block of the chromatin is the nucleosome, constituted by two copies of each of the four core histones H2A, H2B, H3, and H4 surrounded by 147 base pairs of DNA (Luger et al., 1997). In addition, a linker histone H1 directs the path of DNA between the adjacent nucleosomes that make up the chromatin fiber. During the last years, numerous studies in yeast and animals have evidenced that the nucleosome is a dynamic structure. Thus, dynamism affects different nucleosomal characteristics. First, canonical histones of the core can be exchanged by histone variant proteins with specific functions. This is known as ‘histone replacement’ and may occur during cell cycle interphase in specific genes or chromosomal regions. Second, histones are post-translationally modified (by acetylation, methylation, phosphorylation, etc.) mostly in the unstructured amino-terminal tail of histones H3 and H4. These modifications are reversible, and different patterns of modification can be recognized, in terms of structural complimentarity, by transcriptional coactivators, and therefore constitute a code (the histone code) (Jenuwein and Allis, 2001). Finally, positioning of a nucleosome is also dynamic. Thus, nuclear machines are able to unwrap the nucleosomal DNA, to slide the nucleosome along the DNA, or even to eject the histone octamer in an ATP-dependent manner. All these dynamic transitions in the structure of a nucleosome have been designated by the general term of chromatin remodeling. In the present chapter we will focus on how ATP-dependent chromatin remodeling and histone replacement control gene expression and silencing in plants.
CHROMATIN REMODELING AND HISTONE VARIANTS
5.2
113
ATP-dependent chromatin remodeling
ATP-dependent chromatin-remodeling complexes (CRCs) are multisubunit nuclear enzymes able to alter the interactions between the histone octamer and the DNA and therefore able to increase the fluidity of the chromatin (for recent reviews, see Flaus and Owen-Hughes, 2004; Cairns, 2005; Smith and Peterson, 2005). In vitro, nucleosomes are in a thermodynamic equilibrium between a fully wrapped state and a set of partially unwrapped states (Li et al., 2005). In fact, it has been shown that increased temperature can provoke changes in the nucleosomal positioning similar to those carried out by CRCs (Flaus and Owen-Hughes, 2003). Therefore, in this sense, CRCs act as classical enzymes accelerating a natural reaction. A number of in vitro assays have been designed to determine the activity of CRCs. In these assays, CRCs, in the presence of ATP, are able to increase the accessibility of nucleases to the nucleosomal DNA, to move nucleosomes in cis along a DNA template (sliding), to transfer a histone octamer from one DNA template to another in trans, or to create dinucleosome structures from mononucleosomes. However, the mechanism by which these processes occur remains unclear, and more importantly, the way that chromatin-remodeling machines work in vivo is also unknown. All the CRCs contain a protein of the SNF2 family of DNA-dependent ATPases that constitute the motor subunit of the complex. SNF2 family proteins carry out, per se, ATP-dependent nucleosome-remodeling activity in vitro in the absence of additional subunits, although less efficiently than the full complex (Phelan et al., 1999). SNF2 family, which consists of more than 500 currently known proteins, falls into the large superfamily of DEXD/H ATPases. The distinguishing feature of the SNF2 family is the SNF2 N domain, a variant of the typical DEXD/H domain, which contains a well-conserved C-terminal extension of approximately 100 amino acids. SNF2 N ATPase domain contains an ATP-binding pocket and mediates ATP hydrolysis. A variable elastic spacer connects SNF2 N to the C-terminally located HelicC domain. These two domains form the catalytic core of the enzyme. In DNA/RNA helicases which are more distant homologs of SNF2 family, the DEXD/H and HelicC domains form a molecular hinge that mediates helicase activity and processive movement of the enzyme along the nucleic acid molecule. The SNF2 family is usually arranged into several subfamilies based on their auxiliary (signature) domains, located in the less conserved part outside the catalytic core (Figure 5.1). In general, this classification reflects phylogenesis, indicating that the history of SNF2 family involved a series of independent fusion events. Some domain fusions occurred several times throughout the family evolution – for example, the chromodomain independently fused to ATPase domain to form Mi2/CHD and ATRX subfamilies. Numerous studies indicate that the signature domains may play a vital role in substrate recognition as well as in modulation of the enzymatic activity of SNF2-like proteins. One example is the deletion analysis of SANT/HAND/SLIDE signature module of Drosophila ISWI protein (Grune et al., 2003). The deletion of the SLIDE domain completely abolished the ATPase and remodeling activities of ISWI. Thus, assuming that the common ancestor of the
114
REGULATION OF TRANSCRIPTION IN PLANTS
Figure 5.1 Domain organization of the Arabidopsis SNF2 family proteins discussed in the text. Number of amino acids of the full-length proteins is indicated. Domain detection has been carried out using the SMART program (http://smart.embl-heidelberg.de/), the Pfam database (http://www.sanger.ac.uk/ Software/Pfam/), or the authors’ own sequence alignments.
family had similar enzymatic activity, an interesting evolutionary consequence of the fusion to auxiliary domains might be the divergent evolution of the ATPase domain. It could evolve to change the interaction with substrate and/or toward different reaction mechanism. Signature domains do not occur in all subfamilies. For example, a comparison of the DDM1 subfamily sequences revealed two highly conserved stretches of amino acids located N- and C-terminally with respect to the ATPase domain. However, a distant homology search using PSI-BLAST detected homologous sequences in other subfamilies. This result indicated that in contrast to other subfamilies the DDM1 subfamily has no unique signature domain. From an evolutionary perspective, this might imply that DDM1 mode of action may closely resemble the common ancestor. Interestingly, nucleosomes are not the only substrates of SNF2 family members. At least MOT1 subfamily acts to regulate TATA-binding protein (TBP) association with promoter DNA. The enzymatic activities of many subfamilies have not yet been assayed. Thus SNF2 family requires broader description. It can be defined as a family of chaperons that assists protein–DNA interactions. The analysis of the Arabidopsis genome has evidenced the existence of 42 putative genes encoding SNF2 family members (Plant Chromatin Database. http://chromdb.org). These proteins can be grouped into 13 clearly defined subfamilies. Twelve of these subfamilies contain partners from other phylogenetic groups
115
CHROMATIN REMODELING AND HISTONE VARIANTS
of organisms, while one of the subfamilies appears to be plant specific. So far, seven of these genes have been characterized (PICKLE, SPLAYED, ARABIDOPSIS THALIANA BRAHMA, DECREASE IN DNA METHYLATION1, DRD1, CHR11 and PHOTOPERIOD-INDEPENDENT EARLY FLOWERING1), although, multisubunit CRCs have not yet been purified from plants. In the next sections, we will review evidence that suggests that Arabidopsis SNF2-like proteins also form multisubunit complexes, as well as their role in plant transcription and development.
5.2.1
SWI/SNF-like complexes in plants
The SWI and SNF genes were first identified in screenings to isolate Saccharomyces cerevisiae mutants affected in mating type switching (SWItch genes) or in sucrose metabolization (Sucrose Non-Fermenting), respectively (recently reviewed in Smith and Peterson, 2005). Early studies demonstrated that all the SWI and SNF mutations displayed similar phenotypes and defects in the activation of a subset of genes regulated at the transcriptional level, suggesting that SWI and SNF proteins act in the same genetic pathways or form a functional complex. In fact, biochemical analysis demonstrated that SWI and SNF proteins function together as a multisubunit complex (Cairns et al., 1994; Peterson et al., 1994) able to facilitate the binding of transcription factors to nucleosomal DNA in an ATP-dependent way (Cote et al., 1994). The purified complex is approximately 2 MDa in size and contains 11 subunits (Table 5.1). Different SWI/SNF-like complexes have been identified in yeast and metazoans. Although we will refer to them
Table 5.1 Subunits of Saccharomyces cerevisiae SWI/SNF, Drosophila BAP, human BAF and putative Arabidopsis homologs S. cerevisiae
Drosophila (BAP)
Human (BAF)
Arabidopsis
SWI1 SWI2/SNF2
Osa Brahma
BAF250 BRG1, hBRM
SWI3
Moira
BAF155, BAF170
SNF5 SNF6 SWP73
SNR1 — BAP60
SNF11 SWP82 SWP29 ARP7, ARP9
— —
hSNF5 — BAF60a,BAF60b BAF60c —— ENL, EF9 BAF53
— AtBRM SYD CHR12 CHR23 AtSWI3A (CHB1) AtSWI3B (CHB2) AtSWI3C (CHB4) AtSWI3D (CHB3) BSH — CHC1 CHC2 — — — AtARP4 AtARP5
BAP55
116
REGULATION OF TRANSCRIPTION IN PLANTS
in general as SWI/SNF complexes, two different subclasses have been characterized (reviewed in Mohrmann and Verrijzer, 2005). One subclass comprises yeast SWI/SNF, Drosophila BAP (Brahma-associated proteins), and mammalian BAF (Brahma-associated factors) complexes, whereas the second subclass includes yeast RSC (remodels the structure of chromatin), Drosophila PBAP (Polybromo Brahmaassociated proteins), and mammalian PBAF (Polybromo Brahma-associated factors). Brahma is the name of the Drosophila SWI2/SNF2 orthologous protein. The difference between both subclasses of complexes is the presence of a higher number of bromodomains in the RSC, PBAP, and PBAF complexes than in the SWI/SNF, BAP, and BAF complexes. The bromodomain is a 110 amino acid domain involved in the interaction with acetylated lysines (Zeng and Zhou, 2002), which suggests that those complexes containing more bromodomains would be recruited to loci or genome regions with a higher level of histone acetylation. Although yeast SWI/SNF and Drosophila BAP complexes were first identified as transcription activators, today it is well established that SWI/SNF-like complexes are involved in both transcription activation and transcription repression. Thus, whole-genome expression analysis has shown that expression of about 5% of the yeast genes are affected by loss of SWI and SNF proteins, about half of them being upregulated (Sudarsanam et al., 2000). Animal SWI/SNF has been implicated in embryo development (Bultman et al., 2000; Klochendler-Yeivin et al., 2000) as well as in cell-specific gene expression of erythrocytes (Armstrong et al., 1998), myelocytes (Kowenz-Leutz and Leutz, 1999), myocytes (de la Serna et al., 2001; Simone et al., 2004), and adipocytes (Pedersen et al., 2001). Furthermore, mutations of several SWI/SNF subunits have been found in human cancers, suggesting that SWI/SNF factors are important tumor suppressors (Versteege et al., 1998; Wong et al., 2000). How does the SWI/SNF complex regulate transcription? Unfortunately, this question does not have a simple and single answer. The picture that is emerging from the still small number of well-known examples in yeasts and mammals is that the SWI/SNF complexes are recruited to promoters or/and regulatory regions by interacting with specific DNA-binding proteins (Krebs et al., 1999; Agalioti et al., 2000). Histone H3 and H4 acetylation state of neighboring nucleosomes can contribute to the recruiting process through bromodomains or may be important for the stabilization of the complex in the chromatin (Hassan et al., 2002). Once in the regulatory region, the SWI/SNF complex appears to remodel the structure of one or more nucleosomes in order to permit access to buried DNA sequences. This remodeled state allows the binding to the newly exposed sequence of an additional DNA, specific transcription factor, or a general transcription factor such as the TBP. As mentioned above, an SWI/SNF-like complex has not been so far purified from plants, although, the analysis of the Arabidopsis genome indicates the existence of genes encoding polypeptide homologs of the subunits of the complex. Table 5.1 summarizes the polypeptides of the yeast SWI/SNF complex and their homologs in Arabidopsis. The functional core of the SWI/SNF complex sufficient to remodel chromatin in vitro comprises four polypeptides: a dimer of SWI3 protein type, a SWI2/SNF2 type, and a SNF5 type (Phelan et al., 1999). This core seems to be universally conserved in all characterized SWI/SNF complexes and is also conserved in
CHROMATIN REMODELING AND HISTONE VARIANTS
117
plants. In addition, Arabidopsis also contains homologs of SWP73 and actin-related protein (ARP) subunits. The fact that Arabidopsis contains small gene families for several of the putative SWI/SNF subunits suggests the existence of a number of different SWI/SNF complexes and raises the issue of function specificity or redundancy. The ATPases of the SWI/SNF complexes are classified into the SWI2/SNF2 subfamily and are characterized by the following signature domains: Domain I and Domain II at the N-terminus (originally defined for the Drosophila Brahma protein (Tamkun et al., 1992), and a bromodomain at the C-terminus. From the 42 ATPases of the SNF2 family found in Arabidopsis only 4 belong to the SWI2/SNF2 subfamily based on phylogenetic analysis of SNF2 N and HelicC domains: SPLAYED (SYD), ARABIDOPSIS THALIANA BRAHMA (AtBRM), CHR12, and CHR23 (Figure 5.1). The four polypeptides contain a region related to the Domain II of Brahma, and two of them (SYD and AtBRM) also contain a region related to the Domain I. However, only AtBRM presents a bromodomain (see Figure 5.1). Two of these genes, SYD and AtBRM, have been characterized at the functional level. Mutants in SYD were identified as phenotypic enhancers of a weak LEAFY allele (Wagner and Meyerowitz, 2002). SYD plays a role in phase transition, and in carpel and ovule development. The syd inflorescence terminate in a central carpelloid structure after forming 2–20 flowers, suggesting that SYD is also required for the maintenance of the shoot apical meristem (SAM) during the reproductive phase. In agreement with this, it has been recently shown that SYD controls the SAM stem cell pool via direct transcriptional control of the WUSCHEL (WUS) gene (Kwon et al., 2005). Therefore, WUS is the first demonstrated direct target gene of a SNF2 protein in plants. Reduction of the levels of AtBRM by RNA interference also results in a pleiotropic phenotype (Farrona et al., 2004). Comparison of AtBRM-silenced plants with syd mutants indicates that while they share some phenotypes, others are very different or even opposite, suggesting that there is very little functional redundancy between both homologous proteins. AtBRM-silenced flowers exhibit short petals and stamens, fused stamens and immature anthers, as well as defects in organ identity such as second-whorl petals with patches of sepalloid tissue. AtBRM-silenced plants display an early flowering phenotype both under short day and long day, caused by the upregulation of CONSTANS, FT, and SOC1. However, at the moment it is unknown whether any of these genes is a direct target of AtBRM. The SWI3-like small gene family is composed of four polypeptides termed AtSWI3A (CHB1), AtSWI3B (CHB2), AtSWI3D (CHB3), and AtSWI3C (CHB4) (Sarnowski et al., 2002). These proteins contain four well-defined domains of unclear functions: a SWIRM (SWI3/Rsc8/Moira) domain, a zinc finger, a SANT (SWISNF, ADA, N-CoR, TFIIIB) domain, and a leucine zipper. A phylogenetic analysis indicates that AtSWI3A and AtSWI3B are closely related and clearly separated from AtSWI3C and AtSWI3D (Sarnowski et al., 2005). Interestingly, these phylogenetic differences have also functional consequences (see below). Analysis of interactions between AtSWI3 proteins by yeast two-hybrid demonstrates that they form dimers similarly to the yeast and Drosophila orthologous proteins. However, not all the combinations of AtSWI3 subunits seem to be permitted. Thus, only
118
REGULATION OF TRANSCRIPTION IN PLANTS
Figure 5.2 Schematic representation of the interactions within the putative Arabidopsis SWI/SNF complex. This model combines yeast two-hybrid system data from three different publications (Sarnowski et al., 2002, 2005; Farrona et al., 2004).
six dimer combinations are theoretically possible (AtSWI3A:AtSWI3A, B:B, A:B, A:C, B:C, and B:D; see Figure 5.2). Functional analysis indicates that AtSWI3 proteins have rather specific and nonredundant functions. Thus, AtSWI3A and AtSWI3B are required for embryo development while AtSWI3C and AtSWI3D are required for vegetative and reproductive development (Sarnowski et al., 2005). Thus, atswi3a and atswi3b homozygous embryos arrest in the late and early globular stadium, respectively. Furthermore, atswi3b mutations result in lethality of about half of both macro- and microspores. Mutations in AtSWI3D provoke severe dwarfism, and several floral defects including highly degenerated gynoecium and complete male and female sterility. Interestingly, inactivation of AtSWI3C results in a phenotype strikingly similar to that observed in AtBRM-silenced plants. Thus, atswi3c knockout plants present a delayed development, characterized by short stature and small downward curled leaves. Flowers exhibit small petals and stamens, fused stamens, and anthers replaced by sepalloid tissue, suggesting defects of class B homeotic genes. In agreement with these observations, atswi3c plants present a reduction in AP3 and PI mRNA levels. The finding that AtBRM-silenced plants display similar phenotypes to those in atswi3c plants suggests that they may form part of the same functional SWI/SNF-like complex. This
CHROMATIN REMODELING AND HISTONE VARIANTS
119
is consistent with the fact that AtBRM interacts with AtSWI3C in yeast two-hybrids (Farrona et al., 2004) (Figure 5.2). Arabidopsis contains a single SNF5 homolog termed BUSHY (BSH). BSH is able to partially complement yeast SNF5 mutations confirming the functional homology between both proteins (Brzeski et al., 1999). BSH interacts with AtSWI3B and AtSWI3A but not with AtSWI3D and AtSWI3C (Figure 5.2). Since SNF5 is one of the subunits of the core complex it is predicted that BSH should be present in all putative Arabidopsis SWI/SNF complexes. This suggests that mutations in BSH should have more dramatic effects than mutations in SWI3 or SWI2/SNF2 homologs, which are encoded by small gene families. Consistently, partial silencing of BSH led to a dramatic pleiotropic phenotype characterized by reduction in apical dominance and infertility.
5.2.2
Other ATPases of the SNF2 family that control plant development
Arabidopsis pickle (PKL) mutants were identified during a screen for mutants that exhibit abnormal root development (Ogas et al., 1997). The PKL roots express many embryonic traits after germination and are referred to as ‘pickle roots’. Another PKL mutant allele, gymnos, enhance the carpel abaxial–adaxial polarity defects of crabs claw (crc) mutations (Eshed et al., 1999). Thus, a gym crc double mutant displays ectopic formation of placentae and ovules outside the carpels. Single gym mutants exhibit a delay in differentiation of several carpel cell types, suggesting that GYM is not specifically involved in organ or cell polarity but in cell differentiation. PICKLE encodes an SNF2-like protein similar to the human Mi-2 proteins. Mi-2 belongs to the Mi-2/CHD subfamily of SNF2-like proteins, which are characterized by the presence of a chromodomain, and a SANT domain (Woodage et al., 1997) (Figure 5.1). Human Mi-2 associates with histone deacetylases (HDAC1 and HDAC2) in a multisubunit complex named NuRD (for nucleosome remodeling and histone deacetylation) involved in repression of transcription (Ahringer, 2000). The human NuRD complex also contains two histone chaperons of the MSI family (RbAp46 and RbAp48), a methyl-cytosine-binding protein and two zinc finger proteins called MTA1 and MTA2. Whether PICKLE is integrated in a multisubunit complex in Arabidopsis or not is currently unknown. PKL is involved in repression of embryo-specific genes after embryogenesis by controlling negatively the LEAFY COTYLEDON (LEC) genes (Ogas et al., 1999; Dean Rider et al., 2003). LEC genes promote embryonic identity in Arabidopsis. Loss-of-function mutations of the LEC genes, LEC1, LEC2, and FUSCA3 (FUS3), result in the replacement of cotyledons with vegetative leaves, as well as in reduced accumulation of storage proteins and desiccation protectants (Meinke et al., 1994). Therefore, overexpression of these regulatory genes in adult PKL plants has opposite effects. In addition, PKL plants exhibit the phenotypic characteristics of plant defective in gibberellic acid (GA) response: PKL plants are dark green and dwarfed, and time to flowering is increased (Henderson et al., 2004). FUS3 and LEC2 repress the expression of the enzyme AtGA3ox2 that catalyzes the conversion of inactive to bioactive GAs (Curaba et al., 2004; Gazzarrini et al., 2004). The levels of both LEC
120
REGULATION OF TRANSCRIPTION IN PLANTS
genes are increased in adult tissues of PKL plants, which leads to the repression of AtGA3ox2, decreasing the level of GA. Another member of the SNF2 family with important functions in Arabidopsis development is PHOTOPERIOD-INDEPENDENT EARLY FLOWERING1 (PIE1) (Noh and Amasino, 2003). The PIE1 mutant displays an early flowering phenotype both under long day and under short day. Interestingly, PIE1 is required for high levels of FLOWERING LOCUS C (FLC) expression in the shoot. FLC is a MADSdomain transcription factor involved in flowering repression. In addition to the SNF2 N and the HelicC domains, PIE1 has a HSA and a SANT domain (Figure 5.1). Our phylogenetic analysis indicates that PIE1 belongs to the SWR1 subfamily of SNF2 proteins. The yeast SWR1-associated complex (SWC) is capable of directing the replacement of H2A–H2B dimers by the histone variant containing dimers H2A.Z–H2B (Krogan et al., 2004; Mizuguchi et al., 2004) (see Section 5.4.1). In yeast, H2A.Z (Htz1p) helps to form boundaries that restrict the spreading of silent chromatin from the telomers. One of the subunits of the SWR1 complex is the actin-related protein ARP6. Arabidopsis ARP6 (also called suppressor of frigida 3) mutants show a phenotype strikingly similar to the pie1 mutants, including early flowering and reduced levels of expression of FLC (Choi et al., 2005; Deal et al., 2005). This is consistent with the existence of a SWC-like Arabidopsis complex.
5.3
Chromatin remodeling and DNA methylation
Classically, heterochromatin is defined as a highly compacted transcriptionally inert part of the genome. Cytosine methylation is one of the characteristic epigenetic markers of heterochromatin. In Arabidopsis, methylated DNA is mostly located in heterochromatic chromocenters, which contain centromeric and pericentromeric repeat sequences. In contrast to Metazoa, plants have three distinct classes of DNA methyltransferases that are specific to cytosines within different DNA sequence contexts: CpG, CpXpG, and asymmetric C (Finnegan and Kovac, 2000). The MET1 methyltransferase is mainly responsible for de novo methylation (Aufsatz et al., 2004) and its maintenance at CpG sites (Finnegan et al., 1996; Genger et al., 1999). CHROMOMETHYLASE 3 (CMT3) is required for maintenance of the methylation status at CpXpG sites (Lindroth et al., 2001), and it cooperates with DOMAINS REARRANGED METHYLASES 1 and 2 (DRM1 and DRM2) in maintaining CpXpG and asymmetric methylation at other loci (Cao and Jacobsen, 2002). DRM1 and DRM2 also act as de novo DNA methylases at cytosines in all sequence contexts (Cao and Jacobsen, 2002). The mechanisms by which particular DNA sequences are targeted and converted into heterochromatin are poorly understood. In the DNA methylation mutant met1, chromocenters are still present (although they are smaller than normal), which suggests that mechanisms other than DNA methylation operate to form heterochromatin (Soppe et al., 2002). RNA interference has been shown to participate in centromeric heterochromatin maintenance in Schizosaccharomyces pombe (Volpe et al., 2002). Similar mechanisms were proposed for the maintenance of H3K9 and DNA methylation in centromeric tandem
CHROMATIN REMODELING AND HISTONE VARIANTS
121
repeats of Arabidopsis (Martienssen, 2003). Paradoxically, such mechanisms would require continuous transcription of one DNA strand to produce siRNAs, which in turn would promote silencing of the other strand. According to this scenario, heterochromatin would have relatively high transcriptional activity although very little of the mature full-length transcripts would be produced. DECREASE IN DNA METHYLATION 1 (DDM1) was identified in early genetic screens for Arabidopsis mutants impaired in DNA methylation (Vongs et al., 1993) and transgene silencing (Mittelsten Scheid et al., 1998; Mittelsten Scheid and Paszkowski, 2000). It was subsequently shown to encode an SNF2-class protein, DDM1. Biochemical analysis of recombinant DDM1 demonstrated specific interaction with mononucleosomes and ATP-dependent nucleosome sliding activity (Brzeski and Jerzmanowski, 2003). As DDM1 has not been analyzed using other chromatin-remodeling assays, very little is known about its reaction mechanism.
5.3.1
The effects of ddm1 mutation differ from those caused by met1
Sequence homology and the available biochemical evidence suggest that DDM1 is closely related to chromatin-remodeling factors, and this prompted the formulation of the first model of DDM1 action in vivo. In this simple scenario, DDM1 accompanies DNA methyltransferase MET1 and its chromatin-remodeling activity facilitates access to hemimethylated nucleosomal DNA just after DNA replication. This model satisfactorily explains the strong reduction in DNA methylation observed in ddm1 plants. However, further examination of ddm1 and met1 lines has highlighted striking differences in their phenotypes. Depletion of MET1, either by expression of antisense RNA or by mutation, causes a strong pleiotropic phenotype (Finnegan et al., 1996; Kankel et al., 2003; Saze et al., 2003). In the met1-3 null allele homozygotic line, CpG methylation is almost completely absent in all analyzed loci (Saze et al., 2003). In contrast, the ddm1 lines are initially WT (Wild Type) for this phenotype, with the onset of hypomethylation observed only after prolonged inbreeding (Kakutani et al., 1996). In the first mutant generation, the DNA methylation level is decreased by 70% but the changes are restricted to the repetitive, heterochromatic fraction of the genome, which is the major acceptor of DNA methylation. Euchromatic sequences begin to be demethylated only in subsequent generations.
5.3.2
The model of DDM1 action
In light of the above data, the previously described model has to be modified slightly. MET1 would require DDM1 assistance mainly in the heterochromatic environment, while DDM1 appears to play only a minor role in the replication of euchromatic DNA methylation. This model is supported by an immunolocalization study on LSH, the mouse ortholog of DDM1. LSH is found mainly in late-replicating heterochromatic foci where it colocalizes with DNA methyltransferase DNMT1 during late S-phase (Yan et al., 2003). However, it is not clear to what extent mechanisms involving DDM1-like proteins have been conserved during evolution. In addition,
122
REGULATION OF TRANSCRIPTION IN PLANTS
this second model has been challenged by more recent studies. DAPI staining of ddm1 and met1 nuclei has revealed reduced chromocenters, indicating that heterochromatic foci have been ‘dissolved’ (Soppe et al., 2002). Both mutations led to an approximately 25% reduction in the average heterochromatin content. Consistent with these results, fluorescence in situ hybridization (FISH) analysis demonstrated that a number of low-copy pericentromeric sequences appear dislocated away from the chromocenters while tandem repeats, such as the 180-bp centromeric repeats and the transposon arrays, remained associated with chromocenters. However, in both mutated lines, tandem repeats underwent demethylation and in the ddm1 line this was almost fully restricted to the repetitive DNA. These observations led to the suggestion that dense DNA methylation is not an important factor in heterochromatinization. Alternatively, some other unique features of the tandem repeats might prevent their dissociation from chromocenters, even where DNA methylation has been erased. These features might include particular attributes of DNA sequences or unrecognized chromatin modifications such as a specific pattern of histone modifications or the presence of the CenH3 variant of histone H3. Strikingly, crossing the ddm1 and met1 lines caused a further decrease in the chromocenter size (Soppe et al., 2002). This additive effect suggests independent modes of action for DDM1 and MET1. It is noteworthy that the study discussed above did not use null alleles. The allele ddm1-2 has a single nucleotide mutation that creates an additional splicing acceptor site, and translation of the incorrect splicing product would produce a truncated protein lacking an essential part of the HelicC domain. Such a protein would lack enzymatic activity but it might still mediate some important molecular interactions that may be a part of DDM1 activity in vivo. The in vivo stability of this truncated DDM1 has not been assayed. In addition, the MET1 allele used was semifunctional and could mediate DNA methylation to some extent. FISH experiments using a som8 line, which is a null allele of DDM1, revealed the dissociation of CEN180 bp arrays away from chromocenters (Scheid et al., 2002), while CEN180 bp was fully associated with chromocenters in the nuclei of met1-3, a MET1 null line (Tariq et al., 2003). These data further suggest an additional function for DDM1 besides assisting MET1. Formation of chromocenters and heterochromatinization of DNA is developmentally regulated (Mathieu et al., 2003). In young 2-day-old seedlings only small prechromocenters can be observed. Chromocenters achieve their mature size between days 2 and 4 after germination. DDM1 is most likely to be involved in the latter steps of chromatin organization, since prechromocenters in a ddm1-2 line are of the same size as those in WT nuclei. However, the depletion of DDM1 function severely affects heterochromatin maturation. Prechromocenter size increases only very slightly in ddm1-2 nuclei. Maturation of heterochromatin not only involves the association of dispersed DNA with growing prechromocenters but also requires extensive reorganization of chromosomal structure. FISH painting experiments in 2-day-old nuclei demonstrated that 5S rDNA is entirely associated with prechromocenters. In 4-day-old nuclei a fraction of 5S rDNA disperses away from the chromocenters. The formation of 5S rDNA euchromatic loops correlates with increased transcription. In addition, the modification status of 5S rDNA changes during these
CHROMATIN REMODELING AND HISTONE VARIANTS
123
structural rearrangements. The overall DNA methylation level increases, although it is possible that postreplicative methylation is limited to the heterochromatic fraction of 5S rDNA because another heterochromatin marker, H3K9 methylation, is only weakly detectible in the euchromatic 5S rDNA loops. DDM1 seems to counteract 5S rDNA euchromatinization – the loops are much more extensive in mature ddm1-2 nuclei compared to WT nuclei. This observation supports the hypothesis assigning an active role to DDM1 in shaping chromosomal structures.
5.3.3
DDM1 and methylation of histone H3
An additional level of complexity, which has to be considered when constructing a model of DDM1 action, results from its influence on the patterns of other chromatin modifications. Mutation of DDM1 resets the methylation patterns of histone H3. In general, H3K9 methylation is decreased in many heterochromatic sequences and is often substituted by H3K4 methylation (Gendrel et al., 2002). In contrast to the overall decrease in cytosine methylation, H3 methylation levels are not significantly altered in the ddm1-2 line. Extensive mapping of chromatin modification using chromatin immunoprecipitation has in fact revealed heterochromatic loci that exhibit more H3K9 methylation in ddm1-2 compared to WT. Therefore, the effects of ddm1 mutation on the methylation of cytosine and histone H3 are clearly different: the former shows a global decrease while the pattern of the latter is altered. These modifications are closely connected, but a discussion of their interdependence is beyond the scope of this review. The global reduction in DNA methylation in genetic backgrounds producing hypomethylation, such as ddm1 and met1, is often accompanied by local hypermethylation in some loci including AGAMOUS (AG) and SUPERMAN (SUP) (Jacobsen et al., 2000). This phenomenon mirrors to some extent the resetting of patterns of H3 methylation. Genetic screens for suppressors of the hypermethylated SUP epiallele clark kent (clk) identified DNA chromomethylase CMT3 (Lindroth et al., 2001) and H3K9-specific histone methyltransferase (HMT) KRYPTONITE (KYP) (Jackson et al., 2002). In good agreement with the genetic data, clk showed increases in CpXpG cytosine methylation and in H3K9 methylation. RNA interference mechanisms are presumably involved in this process, since the Argonaut family protein AGO4 was also identified in the same genetic screen (Zilberman et al., 2003). It is tempting to speculate that global hypomethylation of the genome might lead to transcriptional derepression of many loci and to random production of aberrant RNAs. Stochastic transcription and degradation of these transcripts might produce siRNAs that would guide RNA-dependent chromatin modification apparatus. Such a mechanism would provide an additional layer of complexity to epigenetic repression. The results discussed above suggest that the mode of action of DDM1 is more complex than just increasing MET1 substrate accessibility. One attractive hypothesis is that DDM1 serves as a chromatin chaperon that actively shapes heterochromatin. The two models described above are not mutually exclusive because DDM1 activity apparently partially overlaps that of MET1. However, both are still highly speculative and further experimental evidence is required to establish their validity.
124
5.3.4
REGULATION OF TRANSCRIPTION IN PLANTS
Models of DDM1 targeting
Transient expression of DDM1 fused to GFP (green fluorescent protein) revealed two distinct patterns of DDM1 subnuclear distribution (Zemach et al., 2005). Type I nuclei showed an intense signal in the chromocenters with much weaker staining of euchromatin. In type II nuclei the DDM1–GFP signal was uniformly dispersed throughout the nucleus. What might be the targeting signal for DDM1? It could be provided by heterochromatic epigenetic markers such as DNA or H3K9 methylation. Recombinant DDM1 is unable to sense the methylation status of DNA (Brzeski and Jerzmanowski, 2003). However, DDM1 might interact with auxiliary proteins that can specifically bind to methylated DNA. Recently DDM1 was shown to interact in vitro and colocalize in vivo with methyl CpG binding (MBD) proteins, which recognize methylated CpG dinucleotides (Zemach et al., 2005). Specific interactions between DDM1 and modified histones have not yet been examined. If DDM1 assists MET1 in DNA methylation, specific recognition of hemimethylated DNA would be expected. MBD proteins recognize symmetrically methylated CpG but recent in silico simulations of the recognition process demonstrated that only one of the methyl groups significantly contributes to the binding energy (Rauch et al., 2005). A number of speculative scenarios can be envisaged to describe the role of MBDs in DDM1 targeting. MBDs could bind to newly methylated DNA while DDM1 participates in DNA methylation processes. DDM1–MBD interaction would then provide a spreading mechanism causing DDM1 and presumably MET1 to ‘slide’ along the chromatin fiber. At the same time, MBD proteins would decorate chromatin to lock up the heterochromatic state. Alternatively, DDM1 might act during the latter stages of heterochromatin formation. MBD-covered chromatin could attract DDM1, which would act to mature the heterochromatic environment to prevent demethylation. Such a mechanism could involve further compaction of heterochromatin promoted by the remodeling activity of DDM1. Contrary to the scenarios described above, LSH targeting is unchanged in DNA hypomethylating cell lines (Yan et al., 2003). Interestingly, LSH localization is affected in cells treated with a histone deacetylase inhibitor. In Arabidopsis, mutation of the sil1 histone acetyltransferase (HAT) leads to decreases in DNA and H3K9 methylation (Lippman et al., 2003). In contrast to ddm1- and met1-mediated DNA demethylation, the sil1 effect is unstable. Backcrossing to WT causes remethylation to reappear very quickly. Histone acetylation/deacetylation has a very high turnover rate and is therefore unlikely to serve as a stable epigenetic marker. At present it is not clear whether histone acetylation is involved in DDM1 targeting.
5.3.5
DRD1 – another SNF2 family protein involved in control of DNA methylation
Recently, another plant SNF2 family member, DRD1, has been implicated in DNA methylation control (Kanno et al., 2004, 2005). DRD1 has no obvious orthologs in fungi and animals and appears to belong to a plant-specific subfamily. DRD1 was identified in a genetic screen for suppressors of RNA-directed DNA methylation. The screen consisted of two transgenes integrated into the Arabidopsis genome: a
CHROMATIN REMODELING AND HISTONE VARIANTS
125
‘reporter transgene’ encoding GFP and a ‘silencer transgene’ that produced short double-stranded RNA complementary to the promoter region of the reporter transgene. In the wild-type population, the silencer transgene effectively guided RNAdirected DNA methylation to block expression of the target transgene. In drd1 mutants, DNA methylation in the siRNA complementary region was significantly reduced. However, the drd1 mutation did not affect global DNA methylation in heterochromatic regions. Thus, DRD1 is thought to act locally during de novo methylation of DNA complementary to siRNA. Intriguingly, DRD1 was required for the efficient erasure of a methylation mark, after outcrossing of the siRNA source. The precise role of DRD1 in the dynamic regulation of DNA methylation has yet to be revealed.
5.4
Histone variants in the regulation of chromatin functions
Both animals and plants, in addition to the major class of replication-dependent (RD) histones, have a class of characteristic histone forms, usually encoded by singlecopy genes, the expression of which is not restricted to S-phase. These variants, also known as replication-independent (RI) or replacement histones, may play a role in the modulation and inheritance of chromatin states. They should be distinguished from nonallelic subtypes of major histones that are common in many lineages. A distinctive feature of the variants is an early evolutionary origin, which explains why they are more closely related to each other in phylogenetically distant branches, than to the major core histone forms of their own species.
5.4.1
Occurrence and functional role of core histone variants
Of the four core histones, only H3 and H2A have characteristic, evolutionarily conserved variants (Table 5.2). The biological importance of these forms is best illustrated by H3 variant CenH3. This centromere-specific protein determines the maintenance and inheritance of centromeric chromatin independently of the type of DNA satellite sequences involved, and is critical for correct assembly of the proteins comprising kinetochores (Henikoff et al., 2004). The characteristic feature of CenH3 is the N-terminal part that bears no homology to the N-terminal tail of Table 5.2 Major replication-independent variants of H3 and H2A histones and their function in vertebrates and flowering plants Histone variant
Occurrence and function/animals
Occurrence and function/plants
CenH3
+
+
H3.3 H2AF/Z H2A.X MacroH2A H2ABdb
+ + + + +
Organization of centromeres kinetochore assembly/function Transcriptional activation Transcriptional activation and repression DNA repair/recombination X-chromosome inactivation Transcriptional activation
+ + + − −
Organization of centromeres kinetochore assembly/function Transcriptional activation Not determined Not determined
126
REGULATION OF TRANSCRIPTION IN PLANTS
canonical H3 forms and can be of variable length in different lineages. The histonefold domain (HFD) of CenH3 is only 50% identical to the HFD of H3, with the most distinctive change being the replacement of a conserved glutamine in the α1 helix and the extension by a few amino acids of the loop 1 region connecting helices α1 and α2. Interestingly, loop 1 is one of the conserved fragments through which H3 interacts with nucleosomal DNA. The enlargement of this particular element could cause altered DNA-binding properties, resulting in stronger contacts of CenH3 with nucleosomal DNA (Malik and Henikoff, 2003). The characteristic loop 1 region of CenH3 is both necessary and sufficient for its centromeric localization (Vermaak et al., 2002). The unique and variable N-terminal sequences of CenH3 contain binding domains for kinetochore proteins (Malik et al., 2002). Overall, the variability within the CenH3 family reflects the accelerated evolution of this variant, compared to major H3 forms. Plant CenH3 (HTR12) has been identified and analyzed in Arabidopsis (Talbert et al., 2002). Differences in the N-terminal tail between the CenH3 of Arabidopsis thaliana and the closely related Arabidopsis arenosa confirm the rapid adaptive evolution of this protein in plants. Arabidopsis HTR12 is present in both mitotic and meiotic centromeres. The fast changes in CenH3 variants may be driven by the selective pressure to counteract the spread of deleterious forms of centromeric DNA resulting from rapid evolution of satellite sequences occurring due to meiotic drive (Henikoff et al., 2001). H3.3 another evolutionarily old variant, differs from the conserved canonical H3 in only four amino acids. One of these residues is located in the N-terminal tail and three are in the HFD α1 helix. In H3.3 from different species the replacement amino acids vary, and so both residue substitution and lack of conservation in these positions in addition to RI expression distinguish H3.3 from H3. The correlation between high transcriptional activity and H3.3 deposition argues strongly for structural alterations affecting the accessibility of nucleosomal DNA for transcription. Interestingly, in Drosophila, the replacement of the three amino acids in the HFD region is necessary and sufficient to confer RI deposition of H3 into chromatin, suggesting that these amino acids are critical for recognition by nucleosome assembly factors (Ahmad and Henikoff, 2002). In Lilium longiflorum the expression of a gene encoding an H3.3-like protein is considerably upregulated in vegetative cells of the developing bicellular pollen, which are characterized by highly active chromatin (Sano and Tanaka, 2005). As shown for the first time in alfalfa (Waterborg, 1990) and later in Drosophila (McKittrick et al., 2004) and Arabidopsis (Johnson et al., 2004), H3 and H3.3 show different patterns of post-translational modification. Characteristically, modifications usually associated with active chromatin, like methylation of K4 and K79 and acetylation of K9, K14, K18, and K23, were found to be two- to fivefold enriched in H3.3 compared to canonical H3. On the other hand, the latter was selectively enriched in methylated K9, a modification associated with transcriptionally inactive chromatin. H2AF/Z is an evolutionarily old and universally occurring RI variant of H2A with an essential function in animals. It differs from the canonical H2A in a number of amino acids, with the most pronounced differences in the ‘docking domain’ near the C-terminus, a part of H2A involved in critical interaction with the H3–H4
CHROMATIN REMODELING AND HISTONE VARIANTS
127
dimer. Structural studies indicate looser compaction of H2A.F/Z-containing nucleosomes and suggest that H2A and H2A.F/X cannot coexist in the same nucleosome (Malik and Henikoff, 2003). While there is a strong indication that H2A.F/Z is required for maintaining a transcriptionally open chromatin state, it has also been linked with the repression of transcription (Swaminathan et al., 2005). In Arabidopsis, variants corresponding to H2A.F/Z are encoded by four genes. The expression of one of these genes was shown to be insensitive to inhibitors of DNA replication and strictly correlated with proliferative activity (Callard and Mazzolini, 1997). Another universally conserved RI variant of H2A is H2A.X, characterized by the C-terminal peptide SQ(E/D), where is a hydrophobic amino acid. This variant is critical for genome stability. The phosphorylation of its C-terminal serine in response to double-strand breaks occurring during replication or recombination mediates the recruitment of proteins directly involved in DNA repair (Malik and Henikoff, 2003). H2A.X is present in Arabidopsis, where it is encoded by two genes. Its importance for plant biology has not yet been assessed in detail. In addition to the above universal RI variants, there are some plant-specific forms of H2A that differ structurally from the canonical form. One is an H2A RI variant characterized in wheat, with a short C-terminal tail lacking the SPKK motifs (Huh et al., 1997). On the other hand, some major RI variants of H2A that occur in animals have not been detected in plants (Table 5.2). The replication- and repair-coupled assembly of nucleosomes using canonical core histones is mediated by chromatin assembly factor 1 (CAF-1). It is a complex composed of three conserved subunits: p150, p60, and p48 in human and Cac1, Cac2, and Cac3 in yeast. The CAF-1 complex is also active in nucleosome assembly in plants. Mutations in the Arabidopsis genes FAS1 and FAS2, homologs of Cac1 and Cac2, respectively, disrupt meristem activity resulting in similar phenotypes distinguished by aberrant shoot development and deficient root growth (van Nocker, 2003). MSI, the homolog of the third subunit Cac3, is also required for proper shoot and root development, but in addition it acts as a component of Polycomb complexes that control seed and embryo development (Hennig et al., 2003). Different complexes mediate the RI deposition of histone variants. The FACT complex, composed of SSRP1 and Spt16 subunits, displaces the H2A/H2B dimers during transcription, without changing nucleosome positioning (Belotserkovskaya et al., 2003). H2A.Z has been found in two complexes: one characterized by the presence of the H2A/H2B histone chaperon assembly protein Nap1 and the other containing the Swi/Snf-type ATPase Swr1. The Swr1-complex is critical for the substitution of H2A.Z/H2B by H2A/H2B in the nucleosome. It is possible that the action of Swr1 and FACT are coordinated during transcription to enable effective incorporation of H2A.Z/H2B heterodimers. As mentioned in Section 5.2.2, a close homolog of yeast Swr1 in Arabidopsis is PIE1, a gene required for activation of FLC and floral repression (Noh and Amasino, 2003). Interestingly, PIE1 has a SANT domain, which has been shown to be involved in interactions with HATs and HDACs. The Arabidopsis homolog of the FACT complex has been shown to associate with actively transcribed genes (Duroux et al., 2004). H3.3 is deposited by a HIRA
128
REGULATION OF TRANSCRIPTION IN PLANTS
complex mediating RI chromatin assembly, which contains a histone regulator A (HirA) protein. A gene encoding a HIRA homolog is present in Arabidopsis. Available experimental evidence indicates that deposition of both H2A.X and H3.3 occurs during transcription. In Drosophila cells there is sufficient H3.3 to package all transcribed genes. Given the fact that H3.3 mostly carries modifications associated with active chromatin, whereas the canonical H3 is enriched in modifications linked to transcriptional repression, it is conceivable that the modification status of histones on a particular gene is linked with the nucleosome assembly pathway (McKittrick et al., 2004). This implies that histone-modifying complexes (like those containing HDAC, HMT, and HAT activities) are associated with protein machines responsible for RC or RI nucleosome assembly. If this is indeed the case, and given the conservative nature of RC nucleosome assembly, the transcription-linked replacement of histone variants, rather than the replication (by as yet unknown mechanisms) of the pre-existing modification pattern, could be a major driving force in the epigenetic inheritance of chromatin states (Henikoff et al., 2004).
5.4.2
Role of H1 histone in chromatin and significance of nonallelic H1 variants
The assessment of global linker (H1) histone function in plants and animals has been hampered by compensation among numerous variants, a phenomenon reflecting strong pressure to maintain a physiological H1-to-nucleosome ratio in chromatin. Reversal of the normal ratio of major to minor H1 variants in tobacco, while not affecting the growth and development of the plant, led to disturbances in male gametogenesis and ultimately produced a male sterility phenotype (Przewloka et al., 2002). In mice, the individual somatic subtypes were found to be dispensable even in animals lacking the H10 variant. However, mice lacking three main somatic H1 variants and showing a 47% reduction in the H1–nucleosome ratio died, demonstrating that linker histones are essential for vertebrate development (Fan et al., 2003). Arabidopsis plants with a >90% reduction in H1 expression exhibited a spectrum of aberrant developmental phenotypes, some of which resembled those observed in DNA hypomethylation mutants (Wierzbicki and Jerzmanowski, 2005). While the downregulation of H1 genes did not cause substantial genome-wide DNA hypo- or hypermethylation, it was correlated with minor but statistically significant changes in the methylation patterns of repetitive and single-copy sequences, occurring in a stochastic manner. These findings suggest an important and previously unrecognized link between linker histones and specific patterns of DNA methylation. Notably, a similar effect of H1 depletion on DNA methylation has been recently reported for mice (Fan et al., 2005). H1-dependent higher order chromatin structures have been shown to interfere with Swi/Snf-mediated nucleosome remodeling (Horn et al., 2002). It is thus possible that an H1-stabilized ordered chromatin conformation is required for correct targeting of DNA and core histone modifications. The nonallelic H1 variants could provide this system with additional levels of subtle regulation. The variability of linker histones is the greatest among all histones. In general, one can distinguish between the macroheterogeneity resulting from lineage-specific
CHROMATIN REMODELING AND HISTONE VARIANTS
129
variation and microheterogeneity occurring within lineages. The alignment of conserved H1 globular domains (GH1s) shows a hotspot for insertion located in the WING subdomain. An analysis of the size of the insert yielded a tree that separates H1 histones into four major classes: (1) H1 without an insert, including all known plant H1s and Dictyostelium H1; (2) H1 with a three amino acid insert including yeast and other fungal H1s; (3) H1 with a five amino acid insert including the majority of animal H1s; and (4) H1 with an eight- to ten amino acid insert including some special ‘cleavage-stage’ animal H1 variants. An independent parsimonious tree based on full length GH1 sequences confirmed the above classification (Jerzmanowski, 2004). The WING of GH1 plays a key role in DNA recognition (Gajiwala et al., 2000). Deletion or insertion of several amino acids in this region can thus be of critical importance for the overall strength and specificity of H1 binding in chromatin. This is analogous to the situation revealed for variants of H3 and H2A, in which changes to a few amino acids in crucial interacting regions caused huge overall effects on chromatin structure (see above). The relationships between linker histones within kingdoms have been deduced from phylogenetic analyses (see Jerzmanowski, 2004, for review). Within animal H1s there are several well-separated isoforms. One is the major form of H1 in vertebrates. Another readily distinguished isoform is represented by the cell-cycleindependent forms of H1, characteristic of the whole subkingdom of deuterostomes: H1D of Echinodermata, B4-like and H5 and H10 of Vertebrata. The B4-like isoforms belong to a class characterized by an eight- to ten amino acid insert in the WING domain. Unlike the canonical animal H1 histones, B4 does not prevent the remodeling of chromatin by ATP-dependent complexes and thus might be at least partly responsible for the totipotency of early embryonic chromatin (Saeki et al., 2005). It would be interesting to know if this effect is due to changes in the WING subdomain. The general ‘plant-type’ H1 is also characterized by alterations of the WING, albeit by shortening rather than extension of the ‘insert’ region. However, the characteristic binding properties of the WING might be distorted by both extending and shortening of this critical region. It remains to be seen whether plant H1s affect chromatin accessibility in a similar way to that shown for B4. The occurrence of the novel, highly distinct isoforms of H1 throughout the history of animals reflects the adaptive evolution of H1. Plant H1s have apparently evolved in a similar way. Phylogenetic analysis has revealed a distinct branch of so-called ‘drought-inducible’ variants of H1, the expression of which is upregulated upon water stress. They represent an old isoform, which existed before the separation of mono- and dicotyledonous plants. Another striking feature of plant H1s is a branch representing ‘hybrid’ proteins. It groups sequences in which a typical GH1 domain is fused to domains characteristic of proteins from outside the H1 family, like MYB and HMG (Jerzmanowski et al., 2000). The general scheme based on comparisons of the most evolutionarily conserved GH1 domain does not reveal the rich microheterogeneity of linker histones, which stems from differences between the less well-conserved basic tails. Such variants, often referred to as somatic subtypes, occur in both plants and animals. The N- and
130
REGULATION OF TRANSCRIPTION IN PLANTS
C-terminal tails of the subtypes differ in length, amino acid composition, and the frequency and distribution of phosphorylation sites. The importance of the variability of plant linker histones is still not understood. A characteristic feature of the drought-inducible variants is a much shorter C-terminal tail (about 80 residues) compared to that of the major somatic H1s (125– 150 residues). The drought-inducible variants are not only smaller but also notably less basic than somatic H1s. This feature is due to the fact that the C-terminal domains of linker histones are highly enriched in basic residues. Thus, the Lys + Arg to Asx + Glx ratio for tobacco drought-inducible variant H1C is 1.2, whereas this ratio for the tobacco major somatic H1 (H1B) is 2.5 (Prymakowska-Bosak et al., 1999). The C-terminal domain of H1, which can adopt a segmental α-helical conformation, binds to the linker DNA, neutralizing its negative phosphate charges and facilitating chromatin condensation. In transgenic tobacco plants in which the proportion of smaller and less charged H1 variants (including drought-inducible variants) was considerably increased at the cost of the major somatic variants, the chromatin is less tightly packed than in plants where the dominating forms of H1 are major variants (Prymakowska-Bosak et al., 1999). Interestingly, among 64 transcripts expressed preferentially in Arabidopsis guard cells there are two variants of H1: a major somatic variant and a drought-inducible variant (Leonhardt et al., 2004). These two variants of H1 are the only chromatin-related proteins with a global regulatory function which are specifically upregulated in guard cells. It remains to be seen whether this correlates with the rapid modulation of transcriptional activity required for guard cell function.
References T. Agalioti, S. Lomvardas, B. Parekh, J. Yie, T. Maniatis and D. Thanos (2000) Ordered recruitment of chromatin modifying and general transcription factors to the IFN-beta promoter. Cell 103, 667–678. K. Ahmad and S. Henikoff (2002) The histone variant H3.3 marks active chromatin by replicationindependent nucleosome assembly. Molecular Cell 9, 1191–1200. J. Ahringer (2000) NuRD and SIN3 histone deacetylase complexes in development. Trends Genetics 16, 351–356. J.A. Armstrong, J.J. Bieker and B.M. Emerson (1998) A SWI/SNF-related chromatin remodeling complex, E-RC1, is required for tissue-specific transcriptional regulation by EKLF in vitro. Cell 95, 93–104. W. Aufsatz, M.F. Mette, A.J. Matzke and M. Matzke (2004) The role of MET1 in RNA-directed de novo and maintenance methylation of CG dinucleotides. Plant Molecular Biology 54, 793–804. R. Belotserkovskaya, S. Oh, V.A. Bondarenko, G. Orphanides, V.M. Studitsky and D. Reinberg (2003) FACT facilitates transcription-dependent nucleosome alteration. Science 301, 1090–1093. J. Brzeski and A. Jerzmanowski (2003) Deficient in DNA methylation 1 (DDM1) defines a novel family of chromatin-remodeling factors. The Journal of Biological Chemistry 278, 823–828. J. Brzeski, W. Podstolski, K. Olczak and A. Jerzmanowski (1999) Identification and analysis of the Arabidopsis thaliana BSH gene, a member of the SNF5 gene family. Nucleic Acids Research 27, 2393–2399. S. Bultman, T. Gebuhr, D. Yee, et al. (2000) A Brg1 null mutation in the mouse reveals functional differences among mammalian SWI/SNF complexes. Molecular Cell 6, 1287–1295.
CHROMATIN REMODELING AND HISTONE VARIANTS
131
B.R. Cairns (2005) Chromatin remodeling complexes: strength in diversity, precision through specialization. Current Opinion in Genetics and Development 15, 185–190. B.R. Cairns, Y.J. Kim, M.H. Sayre, B.C. Laurent and R.D. Kornberg (1994) A multisubunit complex containing the SWI1/ADR6, SWI2/SNF2, SWI3, SNF5, and SNF6 gene products isolated from yeast. Proceedings of the National Academy of Sciences of the United States of America 91, 1950–1954. D. Callard and L. Mazzolini (1997) Identification of proliferation-induced genes in Arabidopsis thaliana. Characterization of a new member of the highly evolutionarily conserved histone H2A.F/Z variant subfamily. Plant Physiology 115, 1385–1395. X. Cao and S.E. Jacobsen (2002) Role of the Arabidopsis DRM methyltransferases in de novo DNA methylation and gene silencing. Current Biology 12, 1138–1144. K. Choi, S. Kim, S.Y. Kim, et al. (2005) Suppressor of frigida3 encodes a nuclear actin-related PROTEIN6 required for floral repression in Arabidopsis. The Plant Cell 17, 2647–2660. J. Cote, J. Quinn, J.L. Workman and C.L. Peterson (1994) Stimulation of GAL4 derivative binding to nucleosomal DNA by the yeast SWI/SNF complex. Science 265, 53–60. J. Curaba, T. Moritz, R. Blervaque, et al. (2004) AtGA3ox2, a key gene responsible for bioactive gibberellin biosynthesis, is regulated during embryogenesis by LEAFY COTYLEDON2 and FUSCA3 in Arabidopsis. Plant Physiology 136, 3660–3669. I.L. de la Serna, K.A. Carlson and A.N. Imbalzano (2001) Mammalian SWI/SNF complexes promote MyoD-mediated muscle differentiation. Nature Genetics 27, 187–190. R.B. Deal, M.K. Kandasamy, E.C. McKinney and R.B. Meagher (2005) The nuclear actin-related protein ARP6 is a pleiotropic developmental regulator required for the maintenance of FLOWERING LOCUS C expression and repression of flowering in Arabidopsis. The Plant Cell 17, 2633–2646. S. Dean Rider, Jr., J.T. Henderson, R.E. Jerome, H.J. Edenberg, J. Romero-Severson and J. Ogas (2003) Coordinate repression of regulators of embryonic identity by PICKLE during germination in Arabidopsis. The Plant Journal 35, 33–43. M. Duroux, A. Houben, K. Ruzicka, J. Friml and K.D. Grasser (2004) The chromatin remodelling complex FACT associates with actively transcribed regions of the Arabidopsis genome. The Plant Journal 40, 660–671. Y. Eshed, S.F. Baum and J.L. Bowman (1999) Distinct mechanisms promote polarity establishment in carpels of Arabidopsis. Cell 99, 199–209. Y. Fan, T. Nikitina, E.M. Morin-Kensicki, et al. (2003) H1 linker histones are essential for mouse development and affect nucleosome spacing in vivo. Molecular and Cellular Biology 23, 4559–4572. Y. Fan, T. Nikitina, J. Zhao, et al. (2005) Histone H1 depletion in mammals alters global chromatin structure but causes specific changes in genome regulation. Cell 123, 1–14. S. Farrona, L. Hurtado, J.L. Bowman and J.C. Reyes (2004) The Arabidopsis thaliana SNF2 homolog AtBRM controls shoot development and flowering. Development 131, 4965–4975. E.J. Finnegan and K.A. Kovac (2000) Plant DNA methyltransferases. Plant Molecular Biology 43, 189–201. E.J. Finnegan, W.J. Peacock and E.S. Dennis (1996) Reduced DNA methylation in Arabidopsis thaliana results in abnormal plant development. Proceedings of the National Academy of Sciences of the United States of America 93, 8449–8454. A. Flaus and T. Owen-Hughes (2003) Dynamic properties of nucleosomes during thermal and ATP-driven mobilization. Molecular and Cellular Biology 23, 7767–7779. A. Flaus and T. Owen-Hughes (2004) Mechanisms for ATP-dependent chromatin remodelling: farewell to the tuna-can octamer? Current Opinion in Genetics and Development 14, 165–173. K.S. Gajiwala, H. Chen, F. Cornille, et al. (2000) Structure of the winged-helix protein hRFX1 reveals a new mode of DNA binding. Nature 403, 916–921. S. Gazzarrini, Y. Tsuchiya, S. Lumba, M. Okamoto and P. McCourt (2004) The transcription factor FUSCA3 controls developmental timing in Arabidopsis through the hormones gibberellin and abscisic acid. Developmental Cell 7, 373–385. A.V. Gendrel, Z. Lippman, C. Yordan, V. Colot and R.A. Martienssen (2002) Dependence of heterochromatic histone H3 methylation patterns on the Arabidopsis gene DDM1. Science 297, 1871–1873.
132
REGULATION OF TRANSCRIPTION IN PLANTS
R.K. Genger, K.A. Kovac, E.S. Dennis, W.J. Peacock and E.J. Finnegan (1999) Multiple DNA methyltransferase genes in Arabidopsis thaliana. Plant Molecular Biology 41, 269–278. J.A. Goodrich, G. Cutler and R. Tjian (1996) Contacts in context: promoter specificity and macromolecular interactions in transcription. Cell 84, 825–830. T. Grune, J. Brzeski, A. Eberharter, et al. (2003) Crystal structure and functional analysis of a nucleosome recognition module of the remodeling factor ISWI. Molecular Cell 12, 449– 460. A.H. Hassan, P. Prochasson, K.E. Neely, et al. (2002) Function and selectivity of bromodomains in anchoring chromatin-modifying complexes to promoter nucleosomes. Cell 111, 369– 379. J.T. Henderson, H.C. Li, S.D. Rider, et al. (2004) PICKLE acts throughout the plant to repress expression of embryonic traits and may play a role in gibberellin-dependent responses. Plant Physiology 134, 995–1005. S. Henikoff, K. Ahmad and H.S. Malik (2001) The centromere paradox: stable inheritance with rapidly evolving DNA. Science 293, 1098–1102. S. Henikoff, T. Furuyama and K. Ahmad (2004) Histone variants, nucleosome assembly and epigenetic inheritance. Trends in Genetics 20, 320–326. L. Hennig, P. Taranto, M. Walser, N. Schonrock and W. Gruissem (2003) Arabidopsis MSI1 is required for epigenetic maintenance of reproductive development. Development 130, 2555–2565. P.J. Horn, L.M. Carruthers, C. Logie, et al. (2002) Phosphorylation of linker histones regulates ATP-dependent chromatin remodeling enzymes. Nature Structural Biology 9, 263–267. G.H. Huh, T. Nakayama, T. Meshi and M. Iwabuchi (1997) Structural characteristics of two wheat histone H2A genes encoding distinct types of variants and functional differences in their promoter activity. Plant Molecular Biology 33, 791–802. J.P. Jackson, A.M. Lindroth, X. Cao and S.E. Jacobsen (2002) Control of CpNpG DNA methylation by the KRYPTONITE histone H3 methyltransferase. Nature 416, 556–560. S.E. Jacobsen, H. Sakai, E.J. Finnegan, X. Cao and E.M. Meyerowitz (2000) Ectopic hypermethylation of flower-specific genes in Arabidopsis. Current Biology 10, 179–186. T. Jenuwein and C.D. Allis (2001) Translating the histone code. Science 293, 1074–1080. A. Jerzmanowski (2004) The linker histones. In: Chromatin Structure and Dynamics, State-of-the-Art (eds J. Zlatanova and S.H. Leuba). Elsevier, Paris. A. Jerzmanowski, M.R. Przewloka and K.D. Grasser (2000) Linker histones and HMG1 proteins of higher plants. Plant Biology 2, 586–597. L. Johnson, S. Mollah, B.A. Garcia, et al. (2004) Mass spectrometry analysis of Arabidopsis histone H3 reveals distinct combinations of post-translational modifications. Nucleic Acids Research 32, 6511–6518. T. Kakutani, J.A. Jeddeloh, S.K. Flowers, K. Munakata and E.J. Richards (1996) Developmental abnormalities and epimutations associated with DNA hypomethylation mutations. Proceedings of the National Academy of Sciences of the United States of America 93, 12406–12411. M.W. Kankel, D.E. Ramsey, T.L. Stokes, et al. (2003) Arabidopsis MET1 cytosine methyltransferase mutants. Genetics 163, 1109–1122. T. Kanno, W. Aufsatz, E. Jaligot, M.F. Mette, M. Matzke and A.J. Matzke (2005) A SNF2-like protein facilitates dynamic control of DNA methylation. EMBO Reports 6, 649–655. T. Kanno, M.F. Mette, D.P. Kreil, W. Aufsatz, M. Matzke and A.J. Matzke (2004) Involvement of putative SNF2 chromatin remodeling protein DRD1 in RNA-directed DNA methylation. Current Biology 14, 801–805. A. Klochendler-Yeivin, L. Fiette, J. Barra, C. Muchardt, C. Babinet and M. Yaniv (2000) The murine SNF5/INI1 chromatin remodeling factor is essential for embryonic development and tumor suppression. EMBO Reports 1, 500–506. E. Kowenz-Leutz and A. Leutz (1999) A C/EBP beta isoform recruits the SWI/SNF complex to activate myeloid genes. Molecular Cell 4, 735–743. J.E. Krebs, M.H. Kuo, C.D. Allis and C.L. Peterson (1999) Cell cycle-regulated histone acetylation required for expression of the yeast HO gene. Genes and Development 13, 1412–1421.
CHROMATIN REMODELING AND HISTONE VARIANTS
133
N.J. Krogan, K. Baetz, M.C. Keogh, et al. (2004) Regulation of chromosome stability by the histone H2A variant Htz1, the Swr1 chromatin remodeling complex, and the histone acetyltransferase NuA4. Proceedings of the National Academy of Sciences of the United States of America 101, 13513–13518. C.S. Kwon, C. Chen and D. Wagner (2005) WUSCHEL is a primary target for transcriptional regulation by SPLAYED in dynamic control of stem cell fate in Arabidopsis. Genes and Development 19, 992–1003. N. Leonhardt, J.M. Kwak, N. Robert, D. Waner, G. Leonhardt and J.I. Schroeder (2004) Microarray expression analyses of Arabidopsis guard cells and isolation of a recessive abscisic acid hypersensitive protein phosphatase 2C mutant. The Plant Cell 16, 596–615. G. Li, M. Levitus, C. Bustamante and J. Widom (2005) Rapid spontaneous accessibility of nucleosomal DNA. Nature Structural and Molecular Biology 12, 46–53. A.M. Lindroth, X. Cao, J.P. Jackson, et al. (2001) Requirement of CHROMOMETHYLASE3 for maintenance of CpXpG methylation. Science 292, 2077–2080. Z. Lippman, B. May, C. Yordan, T. Singer and R. Martienssen (2003) Distinct mechanisms determine transposon inheritance and methylation via small interfering RNA and histone modification. PLoS Biology 1, E67. K. Luger, A.W. Mader, R.K. Richmond, D.F. Sargent and T.J. Richmond (1997) Crystal structure of the nucleosome core particle at 2.8 A resolution. Nature 389, 251–260. H.S. Malik and S. Henikoff (2003) Phylogenomics of the nucleosome. Nature Structural Biology 10, 882–891. H.S. Malik, D. Vermaak and S. Henikoff (2002) Recurrent evolution of DNA-binding motifs in the Drosophila centromeric histone. Proceedings of the National Academy of Sciences of the United States of America 99, 1449–1454. R.A. Martienssen (2003) Maintenance of heterochromatin by RNA interference of tandem repeats. Nature Genetics 35, 213–214. O. Mathieu, Z. Jasencakova, I. Vaillant, et al. (2003) Changes in 5S rDNA chromatin organization and transcription during heterochromatin establishment in Arabidopsis. The Plant Cell 15, 2929–2939. E. McKittrick, P.R. Gafken, K. Ahmad and S. Henikoff (2004) Histone H3.3 is enriched in covalent modifications associated with active chromatin. Proceedings of the National Academy of Sciences of the United States of America 101, 1525–1530. D.W. Meinke, L.H. Franzmann, T.C. Nickle and E.C. Yeung (1994) LEAFY COTYLEDON mutants of Arabidopsis. The Plant Cell 6, 1049–1064. J. Mellor (2005) The dynamics of chromatin remodeling at promoters. Molecular Cell 19, 147–157. O. Mittelsten Scheid, K. Afsar and J. Paszkowski (1998) Release of epigenetic gene silencing by trans-acting mutations in Arabidopsis. Proceedings of the National Academy of Sciences of the United States of America 95, 632–637. O. Mittelsten Scheid and J. Paszkowski (2000) Transcriptional gene silencing mutants. Plant Molecular Biology 43, 235–241. G. Mizuguchi, X. Shen, J. Landry, W.H. Wu, S. Sen and C. Wu (2004) ATP-driven exchange of histone H2AZ variant catalyzed by SWR1 chromatin remodeling complex. Science 303, 343–348. L. Mohrmann and C.P. Verrijzer (2005) Composition and functional specificity of SWI2/SNF2 class chromatin remodeling complexes. Biochimica et Biophysica Acta 1681, 59–73. Y.S. Noh and R.M. Amasino (2003) PIE1, an ISWI family gene, is required for FLC activation and floral repression in Arabidopsis. The Plant Cell 15, 1671–1682. J. Ogas, J.C. Cheng, Z.R. Sung and C. Somerville (1997) Cellular differentiation regulated by gibberellin in the Arabidopsis thaliana pickle mutant. Science 277, 91–94. J. Ogas, S. Kaufmann, J. Henderson and C. Somerville (1999) PICKLE is a CHD3 chromatinremodeling factor that regulates the transition from embryonic to vegetative development in Arabidopsis. Proceedings of the National Academy of Sciences of the United States of America 96, 13839–13844.
134
REGULATION OF TRANSCRIPTION IN PLANTS
T.A. Pedersen, E. Kowenz-Leutz, A. Leutz and C. Nerlov (2001) Cooperation between C/EBPalpha TBP/TFIIB and SWI/SNF recruiting domains is required for adipocyte differentiation. Genes and Development 15, 3208–3216. C.L. Peterson, A. Dingwall and M.P. Scott (1994) Five SWI/SNF gene products are components of a large multisubunit complex required for transcriptional enhancement. Proceedings of the National Academy of Sciences of the United States of America 91, 2905–2908. M.L. Phelan, S. Sif, G.J. Narlikar and R.E. Kingston (1999) Reconstitution of a core chromatin remodeling complex from SWI/SNF subunits. Molecular Cell 3, 247–253. M. Prymakowska-Bosak, M.R. Przewloka, J. Slusarczyk, et al. (1999) Linker histones play a role in male meiosis and the development of pollen grains in tobacco. The Plant Cell 11, 2317–2329. M.R. Przewloka, A.T. Wierzbicki, J. Slusarczyk, et al. (2002) The ‘drought-inducible’ histone H1s of tobacco play no role in male sterility linked to alterations in H1 variants. Planta 215, 371–379. M. Ptashne and A. Gann (1997) Transcriptional activation by recruitment. Nature 386, 569–577. C. Rauch, M. Trieb, F.R. Wibowo, B. Wellenzohn, E. Mayer and K.R. Liedl (2005) Towards an understanding of DNA recognition by the methyl-CpG binding domain 1. Journal of Biomolecular Structure and Dynamics 22, 695–706. H. Saeki, K. Ohsumi, H. Aihara, et al. (2005) Linker histone variants control chromatin dynamics during early embryogenesis. Proceedings of the National Academy of Sciences of the United States of America 102, 5697–5702. Y. Sano and I. Tanaka (2005) A histone H3.3-like gene specifically expressed in the vegetative cell of developing lily pollen. Plant and Cell Physiology 46, 1299–1308. T.J. Sarnowski, G. Rios, J. Jasik, et al. (2005) SWI3 subunits of putative SWI/SNF chromatinremodeling complexes play distinct roles during Arabidopsis development. The Plant Cell 17, 2454–2472. T.J. Sarnowski, S. Swiezewski, K. Pawlikowska, S. Kaczanowski and A. Jerzmanowski (2002) AtSWI3B, an Arabidopsis homolog of SWI3, a core subunit of yeast Swi/Snf chromatin remodeling complex, interacts with FCA, a regulator of flowering time. Nucleic Acids Research 30, 3412–3421. H. Saze, O.M. Scheid and J. Paszkowski (2003) Maintenance of CpG methylation is essential for epigenetic inheritance during plant gametogenesis. Nature Genetics 34, 65–69. O.M. Scheid, A.V. Probst, K. Afsar and J. Paszkowski (2002) Two regulatory levels of transcriptional gene silencing in Arabidopsis. Proceedings of the National Academy of Sciences of the United States of America 99, 13659–13662. C. Simone, S.V. Forcales, D.A. Hill, A.N. Imbalzano, L. Latella and P.L. Puri (2004) p38 pathway targets SWI-SNF chromatin-remodeling complex to muscle-specific loci. Nature Genetics 36, 738–743. C.L. Smith and C.L. Peterson (2005) ATP-dependent chromatin remodeling. Current Topics in Developmental Biology 65, 115–148. W.J. Soppe, Z. Jasencakova, A. Houben, et al. (2002) DNA methylation controls histone H3 lysine 9 methylation and heterochromatin assembly in Arabidopsis. The EMBO Journal 21, 6549–6559. P. Sudarsanam, V.R. Iyer, P.O. Brown and F. Winston (2000) Whole-genome expression analysis of SNF/SWI mutants of Saccharomyces cerevisiae. Proceedings of the National Academy of Sciences of the United States of America 97, 3364–3369. J. Swaminathan, E.M. Baxter and V.G. Corces (2005) The role of histone H2Av variant replacement and histone H4 acetylation in the establishment of Drosophila heterochromatin. Genes and Development 19, 65–76. P.B. Talbert, R. Masuelli, A.P. Tyagi, L. Comai and S. Henikoff (2002) Centromeric localization and adaptive evolution of an Arabidopsis histone H3 variant. The Plant Cell 14, 1053–1066. J.W. Tamkun, R. Deuring, M.P. Scott, et al. (1992) Brahma: a regulator of Drosophila homeotic genes structurally related to the yeast transcriptional activator SNF2/SWI2. Cell 68, 561–572. M. Tariq, H. Saze, A.V. Probst, J. Lichota, Y. Habu and J. Paszkowski (2003) Erasure of CpG methylation in Arabidopsis alters patterns of histone H3 methylation in heterochromatin. Proceedings of the National Academy of Sciences of the United States of America 100, 8823–8827.
CHROMATIN REMODELING AND HISTONE VARIANTS
135
S. van Nocker (2003) CAF-1 and MSI1-related proteins: linking nucleosome assembly with epigenetics. Trends in Plant Science 8, 471–473. D. Vermaak, H.S. Hayden and S. Henikoff (2002) Centromere targeting element within the histone fold domain of Cid. Molecular and Cellular Biology 22, 7553–7561. I. Versteege, N. Sevenet, J. Lange, et al. (1998) Truncating mutations of hSNF5/INI1 in aggressive paediatric cancer. Nature 394, 203–206. T.A. Volpe, C. Kidner, I.M. Hall, G. Teng, S.I. Grewal and R.A. Martienssen (2002) Regulation of heterochromatic silencing and histone H3 lysine-9 methylation by RNAi. Science 297, 1833–1837. A. Vongs, T. Kakutani, R.A. Martienssen and E.J. Richards (1993) Arabidopsis thaliana DNA methylation mutants. Science 260, 1926–1928. D. Wagner and E.M. Meyerowitz (2002) SPLAYED, a novel SWI/SNF ATPase homolog, controls reproductive development in Arabidopsis. Current Biology 12, 85–94. J.H. Waterborg (1990) Sequence analysis of acetylation and methylation in two histone H3 variants of alfalfa. The Journal of Biological Chemistry 265, 17157–17161. A.T. Wierzbicki and A. Jerzmanowski (2005) Suppression of histone H1 genes in Arabidopsis results in heritable developmental defects and stochastic changes in DNA methylation. Genetics 169, 997–1008. A.K. Wong, F. Shanahan, Y. Chen, et al. (2000) BRG1, a component of the SWI–SNF complex, is mutated in multiple human tumor cell lines. Cancer Research 60, 6171–6177. T. Woodage, M.A. Basrai, A.D. Baxevanis, P. Hieter and F.S. Collins (1997) Characterization of the CHD family of proteins. Proceedings of the National Academy of Sciences of the United States of America 94, 11472–11477. Q. Yan, E. Cho, S. Lockett and K. Muegge (2003) Association of Lsh, a regulator of DNA methylation, with pericentromeric heterochromatin is dependent on intact heterochromatin. Molecular and Cellular Biology 23, 8416–8428. A. Zemach, Y. Li, B. Wayburn, et al. (2005) DDM1 binds Arabidopsis methyl-CpG binding domain proteins and affects their subnuclear localization. The Plant Cell 17, 1549–1558. L. Zeng and M.M. Zhou (2002) Bromodomain: an acetyl-lysine binding domain. FEBS Letters 513, 124–128. D. Zilberman, X. Cao and S.E. Jacobsen (2003) ARGONAUTE4 control of locus-specific siRNA accumulation and DNA and histone methylation. Science 299, 716–719.
6 Matrix attachment regions and transcriptional gene silencing William F. Thompson, Steven Spiker and George C. Allen
6.1
Introduction
It was realized early in the history of transgenic work that the expression of genes transferred to plant genomes was not always stable or predictable. Such variation has often been attributed to position effects, such as those known to occur in Drosophila (Pal-Bhadra et al., 2004). Position effects are presumed to reflect pre-existing differences in chromatin structure or other genomic features at different sites of transgene integration. However, it is now clear that transgene expression is also affected, and may even be blocked altogether (silenced), by mechanisms that are largely independent of chromosomal integration site. For example, in 1989 the Matzke group reported inactivation and methylation of transgenes from a first transformation following a second transformation with another T-DNA. Soon thereafter, back to back papers from Napoli et al. (1990) and van der Krol et al. (1990) reported cosuppression of transgene and homologous host genes, and Lindbo et al. (1993) reported that transgene-induced virus resistance was associated with a cytoplasmic activity that targets specific RNA sequences for inactivation. Many subsequent reports have highlighted the fact that transgenes are quite frequently subject to position effects and silencing. One particularly dramatic example comes from work by Meyer et al. (1992), who set up an experiment designed to trap transposons inserting into a transgene in petunia. The maize A1 gene, transferred to a white-flowered mutant of petunia, conferred a salmon-red color phenotype in the flowers. Transposon insertions inactivating the transgene should, therefore, create white sectors in otherwise salmoncolored flowers. Expecting such events to be rare, Meyer et al. set up a large field experiment involving 30 000 transgenic individuals. Instead of the expected rare events, however, they found a variety of white-flowered phenotypes, including fully white and variegated flowers, at a very high frequency. A few of the white flowers were found to be associated with deletions of the transgene, but many others, including those showing variegation and developmental shifts in expression, were associated with variation in the methylation status of the 35S promoter. The likelihood of methylation seemed to increase with the age of the plant from which the seed was derived, as well as in response to several environmental factors such as temperature.
MARs AND GENE SILENCING
6.2
137
Costs and consequences of transgene expression variation
These and many other examples illustrate a complex of epigenetic processes that complicate both the use of transgenes as research tools and their application in plant improvement programs (Iyer et al., 2000; Gepts, 2002; Lorence and Verpoorte, 2004). Variability attributable to silencing and related processes necessitates the analysis of many more transformation ‘events’ than might otherwise be necessary to interpret phenotypic effects of transgenes, increases labor costs, and triggers regulatory concerns. In addition, because silencing can occur in later generations of lines initially showing good expression (Bourdon et al., 2002; Vain et al., 2002; Bregitzer and Tonks, 2003; Levin et al., 2005), selecting good expressors from a population of primary transformants, or even from the T1 generation, is not sufficient to guarantee stable performance over subsequent generations. Typically, a considerable number of candidate lines must be maintained through multiple generations before selecting those in which expression is stable and predictable.
6.3
Position effects
Position effects are not always clearly distinguished from other forms of gene silencing. However, position effects are manifest in cis, and, in the classic cases in Drosophila they involve spreading of heterochromatin from nearby loci in such a way that the transgene insertion site is occluded. Due to the stochastic nature of the spreading, position effects sometimes produce a somatically sectored or variegated phenotype. Also, position effects may sometimes reflect the activity of nearby enhancers or suppressors, as exemplified by the results of activation tagging experiments (Masaki et al., 2005; Pan et al., 2005; Xu et al., 2005; Yoo et al., 2005). Position effects are generally regarded as sequence independent, in the sense that they do not depend on insertion of multiple transgene copies; nor do they depend on homologies between transgene and endogenous sequences. Gene silencing, on the other hand, is often triggered by interactions between homologous sequences, which can be located either in cis (e.g., within the transgenic locus) or in trans (multiple transgene loci, or homologies with endogenous sequences elsewhere in the genome) (Wolffe, 1997; Henikoff, 1998; Allen et al., 2000; Hsieh and Fire, 2000; Stam and Mittelsten Scheid, 2005). The extent to which position effects influence transgene expression in plants is currently the subject of some controversy. Activation tagging experiments (Masaki et al., 2005; Pan et al., 2005; Xu et al., 2005; Yoo et al., 2005) clearly demonstrate that at least one kind of position effect can occur when a transgene inserts near a strong enhancer element. However, recent reports (De Buck et al., 2004; Schubert et al., 2004; Nagaya et al., 2005) show that insertions at many different sites in the Arabidopsis genome are expressed similarly when a strong promoter is used and the analysis is restricted to simple, non-rearranged single-copy inserts. In these and
138
REGULATION OF TRANSCRIPTION IN PLANTS
most other experiments, transformants are selected on the basis of drug resistance conferred by a marker linked to the gene of interest, which should prevent recovery of insertions into heterochromatin or other regions of the genome that do not support expression. Francis and Spiker (2005) have shown that such insertions do in fact occur, and can be recovered by screening directly for inserted DNA. Such insertions show that ‘cold spots’ – regions where transgenes are poorly expressed – exist even in the simple, mostly nonrepetitive Arabidopsis genome. It seems likely that such regions occur much more often in the more complex and highly repetitive genomes of typical crop plants.
6.4
Gene silencing
If position effects were the only source of variation in transgene expression, insertions at the same locus should display the same activity in independent transformants. This is not always the case, as elegantly demonstrated by the work of Day et al. (2000). These and other data highlight the existence of gene-silencing mechanisms other than position effects. In many cases, these mechanisms appear to depend on homology within or between transgenes, or between transgenes and endogenous genes. Most current models postulate that these types of gene silencing evolved to protect against invading nucleic acids, either viruses or transposons, and to silence transcription in large blocks of repetitive DNA (Iglesias et al., 1997; Matzke et al., 2000; Miura et al., 2001; Lippman et al., 2003; Volpe et al., 2003; Bender, 2004; Herr et al., 2005; Pontier et al., 2005; Slotkin et al., 2005).
6.4.1
Post-transcriptional gene silencing
Many examples of gene silencing in plants, including the cosuppression phenomena initially described in petunia (Napoli et al., 1990; Van de Krol et al., 1990), involve post-transcriptional gene silencing (PTGS). In PTGS, transcription apparently continues but mRNA fails to accumulate and is thought to be degraded in the cytoplasm. (Transcription measurements in plants usually involve run-on assays, which measure the association of RNA polymerase with a gene of interest, and its ability to elongate transcripts in vitro. Factors controlling elongation in vivo may be lost during nuclear isolation, and so such assays may not quantitatively reflect in vivo transcriptional activity.) After its discovery in plants, this type of silencing was demonstrated in nematodes and other animals, where it is commonly referred to as RNAi (for RNA interference) (Matzke and Birchler, 2005; Tomari and Zamore, 2005; Venken and Bellen, 2005; Wassenegger, 2005). It is now known that formation of short double-stranded RNA (dsRNA), often referred to as short interfering RNA (siRNA), is critical to the process. The siRNA is incorporated into the RNA-induced silencing complex (RISC), a protein complex capable of degrading mRNA and simultaneously producing additional fragments of dsRNA. These new siRNAs continue the cycle, resulting in an extremely effective silencing system that can amplify the effect of minute quantities of initiating dsRNA.
MARs AND GENE SILENCING
139
It is also known that once induced, PTGS can spread through a plant, moving through plasmadesmata and in the phloem. However, silencing is not always complete, and spatially heterogeneous, or patchy, phenotypes are commonly observed. In this case, in contrast to position effect variegation, the patterns of active and inactive areas do not necessarily correspond to developmental cell lineages. In both cases, however, spatial heterogeneity can seriously complicate assays in which tissue samples must be taken ‘blindly’, without visualizing the pattern of expression in vivo (Kerschen et al., 2004). Interestingly, although many examples of gene silencing involve multiple-copy insertions, there are many other examples in which PTGS seems to be triggered simply by high-level expression of a single-copy transgene. An excellent example is the work of Que et al. (1997). Readers are also referred to a recent review by Jorgensen (2003). Strong, high-frequency silencing from single-copy transgenes is most commonly observed with constructs designed for high-level protein overexpression. Such data have led to the concept of a threshold above which mRNA would become a template for RNA-dependent RNA polymerase (RDRP), and then be degraded by Dicer and the RISC complex. Constructs not designed for protein overexpression seem not to be silenced by the same mechanism. In these cases, silencing occurs most frequently when integrated transgenes are present as inverted repeats (Que et al., 1997; Stam et al., 1997a), suggesting that dsRNA production may occur without RDRP involvement. In many cases, it is unclear to what extent transgene repeats and rearrangements may have led to the production of dsRNA. However, it is clear that deliberate production of dsRNA from (for example) engineered hairpin constructs (Wesley et al., 2001; Stoutjesdijk et al., 2002; Guo et al., 2003; Lacomme et al., 2003) can lead to efficient RNA silencing.
6.4.2
Transcriptional gene silencing
Transcriptional gene silencing (TGS) can also be homology dependent (Breyne et al., 1992; Meyer et al., 1992; Meyer and Saedler, 1996; Park et al., 1996; Meyer, 2000), although in this case both run-on transcription and cytoplasmic mRNA accumulation are inhibited. Recent work has shown that many of the proteins and protein complexes involved are similar to those involved in PTGS (Sijen et al., 2001). Short dsRNAs appear to play a role in TGS quite similar to their role in PTGS, being incorporated into an RNA-induced transcriptional gene-silencing complex (RITS), which is analogous to the RISC complex involved in cytoplasmic mRNA degradation (Parker et al., 2004; Schramke et al., 2005). Current models postulate that the RITS complex attracts histone and DNA modification enzymes to the site of silencing, helping to convert open chromatin into condensed, transcriptionally inactive heterochromatin (Motamedi et al., 2004; Noma et al., 2004; Cam et al., 2005; Schramke et al., 2005). Plate 1 shows an example in which integration of multiple transgene copies resulted in a transcriptionally silenced locus (Probst et al., 2003). In this example, a locus containing 15 copies of a 35S-HPT, some of them rearranged during integration, was expressed in the initial transgenic plant but lost expression in progeny.
140
REGULATION OF TRANSCRIPTION IN PLANTS
The locus is hypermethylated and forms cytologically detectable heterochromatic knobs clearly distinguishable from centromeric heterochromatin. One interesting consequence of recent models of dsRNA-dependent transcriptional silencing is that chromatin must be transcribed in order to be silenced (Martienssen, 2003). It is likely that such transcription occurs at low levels in many, if not all, regions of the genome, even where transcripts do not accumulate.
6.4.3
Repeats and rearrangement effects
Given the role of dsRNA in both major forms of silencing, it is perhaps not surprising that transgenic loci containing multiple rearranged or scrambled copies are especially subject to being silenced, particularly if inverted repeats are present (Cluster et al., 1996; Stam et al., 1997a, 1998; Jacobsen, 1999; Meins, 2000; Muskens et al., 2000; De Buck et al., 2001; Beclin et al., 2002; Matzke et al., 2002; Muller et al., 2002). Inverted repeats in the RNA that reaches the cytoplasm would be targets for cytoplasmic PTGS, while aberrant transcripts that remain in the nucleus may attract the RITS complex and induce transcriptional silencing. A particularly interesting case arises when promoter sequences are transcribed in complex loci. In such cases, dsRNA is produced and transcriptional silencing occurs (Mette et al., 1999; Melquist and Bender, 2003), characteristically in association with methylation of cytosine residues in the promoter. Effects of this sort can occur in trans if there are corresponding promoter sequences elsewhere at the locus or in the genome.
6.5
Matrix attachment regions
Matrix attachment regions (MARs) are operationally defined as DNA sequences that bind to the nuclear matrix. The biological role of MARs has been envisioned as the organization of chromatin into a series of topologically isolated loop domains ranging in size from about 5 to 300 kb of DNA. Although not included in the operational definition, a number of characteristics have been associated with MARs. They are approximately 1 kb in size. They are AT rich (typically more than 70% AT) and they contain a number of sequence motifs and features (e.g., A-boxes, T-boxes, Drosophila topoisomerase II consensus sequences, ARS consensus sequences, TG/CA-motifs, CT-rich stretches, kinked DNA, and curved DNA) (Cockerill and Garrard, 1986; Gasser and Laemmli, 1986a,b; Amati and Gasser, 1988; Boulikas, 1993, 1995). Even though all these (and more) sequence motifs and features have been found in MARs, it is clear that the simple presence of one or more of the motifs is not sufficient to establish a DNA fragment as a MAR (Michalowski et al., 1999; Liebich et al., 2002). Thus, a number of bioinformatic approaches have been devised with the goal of identifying MARs based on sequence information (Singh et al., 1997; Glazko et al., 2001; http://futuresoft.org/MAR-Wiz/). These bioinformatic approaches rely on sequence motifs previously found in MARs. The problem with this tactic is
MARs AND GENE SILENCING
141
that even though a motif may be present in a MAR, the motif may not be a factor involved in the binding of the MAR to the nuclear matrix. Typically, programs such as MAR-Wiz are used to identify potential MARs in genomes or portions of genomes without subsequent validation of MARs by actual matrix-binding studies (e.g., Rudd et al., 2004). When such validation studies have been carried out (Namciu et al., 2004), MARs identified by matrix binding and MARs identified by bioinformatic approaches have not correlated well. There is a value in studies incorporating bioinformatic-based approaches for mapping MARs, but results of such studies should be viewed in light of their limitations. Another characteristic associated with MARs (the characteristic that has probably attracted the most attention) is their ability to increase and stabilize transgene expression. This property was first demonstrated in animal cell cultures (Stief et al., 1989) and has been observed in a large number of systems (reviewed by Allen et al., 2000 and Bode et al., 2000). The operational definition and general characteristics of MARs noted above are useful to a degree, but ambiguous for several reasons. First, binding studies are generally carried out in vitro, with isolated nuclear matrices. Thus, there is no compelling evidence that the DNA sequences that are operationally defined as MARs actually serve to anchor loop domains in vivo. Second, although there is a huge volume of literature describing the nuclear matrix both microscopically and biochemically (see Nickerson, 2001, for review), the existence of such a network of nucleoprotein fibers has been controversial (Pederson, 1998, 2000; Hancock, 2000, 2004). Some observers counsel neither full-fledged acceptance nor rejection of the reality of the nuclear matrix, but rather the adoption of an open mind (Martelli et al., 2002; Jackson, 2003, 2005). Third, the definition of a MAR as a matrix-binding DNA sequence has a limitation in that all DNA sequences will bind to the biochemically defined nuclear matrix in the absence of competitor DNA (Michalowski et al., 1999). Even DNA sequences that bind in the presence of competitor DNA have a continuum of affinities ranging from barely detectable to very strong (Brun et al., 1990; Michalowski et al., 1999). Furthermore, the concentration of competitor DNA and its characteristics will determine the relative affinity for the nuclear matrix of any MAR candidate. Thus, it is impossible to set meaningful cutoffs to distinguish MARs from non-MARs. This becomes an issue when bioinformatic approaches are used to identify MARs in silico. These approaches yield a ‘matrix association potential’, but ‘potential’ denotes potential to be a MAR and is not an estimate of binding strength. When predicted MARs are checked for validity by binding experiments, the program used can (1) correctly identify a MAR, (2) identify a DNA sequence as a MAR even though the DNA has negligible affinity for the nuclear matrix, or (3) fail to identify a sequence as a MAR even though it has appreciable affinity for the nuclear matrix (Glazko et al., 2001). How successful a program will be depends, in part, on how great the affinity of a DNA sequence for the nuclear matrix must be in order for it to be judged a MAR. One approach to identifying MARs based on sequence alone depends on calculated DNA duplex destabilization energy rather than the density and arrangement of sequence motifs (Benham et al., 1997). This approach shows considerable promise
142
REGULATION OF TRANSCRIPTION IN PLANTS
because quantitative estimates of stress-induced DNA duplex destabilization correlate well with experimentally measured binding strengths (Bode et al., 2006). The fourth ambiguity relating to MARs concerns their ability to increase and stabilize transgene expression. Although many studies have documented such an effect on transgene expression (reviewed in Allen et al., 2000 and Bode et al., 2000), augmentation of transgene expression has not always been noted and the overall picture is complex (Breyne et al., 1992; Bonifer et al., 1994; Huber et al., 1994; Phi-Van and Str¨atling, 1996; Vaucheret et al., 1998; Brouwer et al., 2002; De Bolle et al., 2003; Sidorenko et al., 2003). The complexity of MAR augmentation and stabilization of transgene expression is the subject of this review.
6.6
MARs and transgene expression
MARs are known to affect the expression of transgenes. Our early experiments, such as shown in Plate 2, showed effects as high as 60-fold when tobacco suspension culture cells were transformed by bombardment (Allen et al., 1996). Subsequent experiments in which plants were used in place of suspension cells and different methods of DNA delivery were tested have shown positive MAR effects in most cases, although the magnitude of the effect is usually much smaller, more on the order of a three- to tenfold increase in expression. Reports of MAR effects in various plant systems are summarized in Table 6.1. Multiple copies of the transgene were introduced in the early experiments, and it is likely that rearrangements often occurred during microprojectile-mediated transformation. On the basis of recent data highlighting the role of complex repeats in TGS and heterochromatin formation (Stam et al., 1997b; Henikoff, 1998; Hsieh and Fire, 2000; Volpe et al., 2002; Ascenzi et al., 2003; Fransz et al., 2003; Martienssen, 2003; Probst et al., 2003; Matzke and Birchler, 2005; Pecinka et al., 2005), it seems likely that this type of transformation favors transcriptional silencing. In addition, Mitsuhara et al. (2002) have shown that rapidly dividing cells show reduced levels of post-transcriptional silencing. If MARs protect against TGS but not PTGS, especially large MAR effects would be expected in situations in which TGS is strong and PTGS is weak. In such cases, strong TGS would reduce transcription of control transgenes lacking MARs, while weak PTGS would allow full expression of MAR constructs that might otherwise be inhibited by cytoplasmic mRNA degradation. More evidence for this hypothesis is presented in the following paragraphs. By comparison, transformation methods leading to lower copy numbers and less rearrangement would be expected to minimize TGS, while assays involving mostly nondividing cells in expanded leaves would be expected to be strongly influenced by PTGS.
6.6.1
MARs reduce gene silencing
In early reports, it was shown that only relatively small MAR effects are seen in transient expression assays, which are conventionally regarded as measuring the expression of transgenes prior to integration in the chromosome. These data were
Chicken (A element)
Tobacco (RB7) Tobacco (RB7) Tobacco (RB7) Arabidopsis (ARS-like)
Chicken (A element)
Tobacco (RB7) Yeast (ARS1) Tobacco (RB7)
Tobacco plants
Tobacco plants
Tobacco cells Poplar explants Tobacco leaf disks Tobacco plants
Maize cells
Rice plants
Tobacco plants
35S-GUS Phaseolin-GUS
Chicken (A element) Bean (Phaseolin)
Tobacco plants Tobacco cells Tobacco plants
35S-GUS 35S-GUS 35S-GUS
35S-cab1-GUS
35S-GUS 35S-GUS 35S-GUS 35S-GUS
Enh35S-GUS
Nos-GUS Nos-GUS Heat shock-GUS 35S-GUS Lhca3-GUS
Soybean (P1) Human (β-globin) Soybean (Gmhsp17.6-L) Yeast (ARS1) Chicken (A element)
Tobacco cells
Promoter-reporter
Source of MAR
Biolistic Biolistic Biolistic
Biolistic
Biolistic Agrobacterium Agrobacterium Agrobacterium
Agrobacterium
Agrobacterium Agrobacterium
Agrobacterium Biolistic Agrobacterium
Agrobacterium
DNA transfer
Effect of flanking MARs on transgene expression and variability in plants and plant cells
Plant system
Table 6.1
2.5-fold increase 3-fold increase 2-fold increase
60-fold increase 10-fold increase Small 5- to 10-fold increase 36-fold increase
2-fold increase
3-fold increase 3-fold increase
Small decrease No effect 9-fold increase 12-fold increase 4-fold increase
Effect on expression level
Small Small Small
Increase
No effect Small Small No effect
2-fold decrease
7-fold decrease 2-fold decrease
Small decrease Small decrease No effect Small 3-fold decrease
Effect on expression variability
(Continued)
¨ Ulker et al. (1999)
Odell and Krebbers (1998) Vain et al. (1999)
Liu and Tabe (1998)
Van der Geest et al. (1994) Mlyn´arov´a et al. (1995) Allen et al. (1996) Han et al. (1997)
Sch¨offl et al. (1993) Allen et al. (1993) Mlyn´arov´a et al. (1994)
Breyne et al. (1992)
Reference
Lhca3-GUS 35S-LUC 35S-LUC 35S-LUC Rsyn7-GUS Apt-Lc-cDNA Apt-Lc-cDNA 35S-LUC Lhca3-GUS
Chicken (A element) Yeast (ARS1) Maize (Adh1 5 ) Maize (Mha1 5 ) Maize (Adh1 5 ) Arabidopsis (HSC80 5 )
Yeast (ARS1) Chicken (A element)
Chicken (A element)
Arabidopsis plants
Tobacco plants
Maize cells
Enh35S-GUS
Chicken (A element)
Chrysanthemum plants
35S-GUS 35S-GUS 35S-NPTII Enh35S-LUC
Tobacco (RB7) Synthetic Tobacco (TJ1) Chicken (A element)
Actin-GUS
Tobacco (RB7)
Tobacco cells Tobacco plants Tobacco cells Tobacco plants
Enh35S-GUS:nptII Actin-GFP
Tobacco (RB7) Tobacco (RB7)
Pine callus Rice plants
Promoter-reporter
Source of MAR
Agrobacterium
Agrobacterium Biolistic Biolistic Biolistic Biolistic Agrobacterium
Agrobacterium
Biolistic Agrobacterium Biolistic Agrobacterium
Biolistic
Agrobacterium Biolistic
DNA transfer
None
Slight decrease None
None 5.8-fold 26.8-fold 1.5-fold 87-fold decrease None
3- to 4-fold 3.3- to 18-fold increase 376- to 650-fold increase 42-fold increase 2.3 to 3.8-fold Slight 0- to 2-fold decrease None
Effect on expression level
Effect of flanking MARs on transgene expression and variability in plants and plant cells (Continued)
Plant system
Table 6.1
ND 1- to 15-fold decrease 1- to 18-fold decrease
Slight increase None None None ND ND
ND Slight increase ND None to slight increase None
Slight
Decrease Slight
Effect on expression variability
Mlyn´arov´a et al. (2002)
Holmes-Davis and Comai (2002)
Brouwer et al. (2002)
Mendu et al. (2001) Nowak et al. (2001) Shimizu et al. (2001) Van Leeuwen et al. (2001) Annadana et al. (2002)
Levee et al. (1999) (Cheng et al., 2001)
Reference
Tobacco (RB7)
Tobacco (RB7)
Chicken (A element)
Maize (P1-rr)
Tobacco cells
Tobacco cells
Cacao plants
Tobacco plants
Maize plants
Sorghum Rice plants
Tobacco (RB7) Synthetic (SM1)
Maize (Adh1 MAR)
Tobacco (5 CHN5)
Tobacco plants Arabidopsis plants
Rice plants
Soybean (P1) Petunia (TBS) Tobacco (RB7) Yeast (ARS1) Tobacco (RB7) Chicken (A element)
Barley plants
(P1-rr,WP, Rsyn7)-GUS (P1-rr,WP, Rsyn7)-GUS Ubiquitin-GUS Gos2-GUS
Enh35S-GUS
35S-NPTII 35S-GUS 35S-GUS 35S-GUS 35S-GUS 35S-GUS Mas-GUS 35S-GUS (MAR-flanked) 35S-GUS (MAR-5 ) 35S-GUS 35S-GUS AtAhas-GUS PsFed1-GUS GmHspL-GUS Nos-GUS Ocs-GUS E12omega-GFP
Biolistic Biolistic
Biolistic
Agrobacterium
Agrobacterium
Biolistic
Agrobacterium
Biolistic Agrobacterium
Biolistic
Biolistic
2-fold increase 0- to 1.9-fold
None
None None
None
None
ND
ND ND ND ND ND ND ND ND Slight decrease
10-fold None 5.3-fold None None 3.1-fold 4.9-fold 15.3-fold 2.2-fold MAR blocked RNAsilencing None
4.6-fold None None None ND None None ND
3.4-fold None 2.5-fold 3.1-fold Slight None None 15-fold
(Continued)
Able et al. (2004) Van der Geest et al. (2004)
Sidorenko et al. (2003)
Maximova et al. (2003) Mlyn´arov´a et al. (2003)
Mankin et al. (2003)
(Fukuda and Nishikawa, 2003)
(Ascenzi et al. (2003) De Bolle et al. (2003)
Vain et al. (2002)
Petersen et al. (2002)
Synthetic (SM1) Maize (Adh1 MAR) Chicken (A element) Chicken (A element) Chicken (A element) Tobacco (RB7) Tobacco (TM-2) Tobacco (TM-2) Tobacco (TM-2) Tobacco (TM-2) Tobacco (TM-2) Tobacco (TM-2) Tobacco (TM-2) Tobacco (TM-2) Tobacco (TM-2) Chicken (A element)
Arabidopsis plants Maize Arabidopsis plants (WT) Arabidopsis plants (sgs2) Arabidopsis plants (sgs3) Tobacco cells Rice calli
Rice plants
Rice roots
Rice leaves
Source of MAR Act2-GUS Ubiquitin-GUS 35S-GUS 35S-GUS 35S-GUS 35S-GFP 35S-GUS Ubi-GUS PNZIP-GUS 35S-GUS Ubi-GUS PNZIP-GUS 35S-GUS Ubi-GUS PNZIP-GUS Rice SBE 1-glgB
Promoter-reporter
Agrobacterium
Agrobacterium Agrobacterium
Agrobacterium Biolistic Agrobacterium
DNA transfer 1.6- to 3.7-fold None 1.7-fold decrease 5-fold 12-fold 2- to 3.7-fold 3.2-fold 4-fold 2.3-fold 5.4-fold 9.9-fold 4-fold 6.2-fold 7.8-fold None Transgenic rice had twofold greater branching ratio
Effect on expression level
Effect of flanking MARs on transgene expression and variability in plants and plant cells (Continued)
Plant system
Table 6.1
None Increase None None None Not determined None None None None None None None None None None
Effect on expression variability
Kim et al. (2005)
Halweg et al. (2005) Xue et al. (2005)
Torney et al. (2004) Butaye et al. (2004)
Reference
Tobacco (RB7)
Tobacco (RB7)-full, direct repeats
Tobacco plants
Tobacco cells
Rice leaves
540-bp 5 end of RB7 MAR
BP-MAR 1300 bp of Chicken (A element) Tobacco (RB7)
Rice plants
35S-GUS
35S-GUS
TetOp35S-LUC
35S-Tomato Spotted Wilt Virus Coat protein
Act1-GFP
Biolistic
Biolistic
Agrobacterium
Agrobacterium Significantly greater resistance in R1 through R4 generations Significant increase, less silencing with delayed induction Full RB7 MAR resulted in a 77-fold increase in transgene expression Flanking 540-bp of 5 RB7 MAR in various orientations had no effect
None
All RB7 MAR constructs had twofold less variation
Not determined
Copy number dependence Not determined
Verma et al. (2005)
Abranches et al. (2005)
(Levin et al., 2005)
Oh et al. (2005)
148
REGULATION OF TRANSCRIPTION IN PLANTS
thus interpreted as indicating that at least the majority of the MAR effect was seen only after integration, and thus that MARs were not simply acting as classic enhancers of transcription. Subsequent reports showed that MARs reduced the loss of ¨ transgene expression from one generation to the next (Conner et al., 1998; Ulker et al., 1999; Vain et al., 1999). As the copy number or Southern pattern of the integrated transgenes did not change, this loss can be attributed only to gene silencing. Thus MARs are considered to reduce gene silencing, rather than merely stimulating transcription. MARs have also been reported to reduce the loss of virus resistance in tobacco plants containing a transgene engineered to induce resistance by producing transcripts homologous to viral RNA (Levin et al., 2005), a result most consistent with a protective effect of MARs at the transcriptional level. Consistent with this hypothesis, Ascenzi et al. (2003) showed that a viral suppressor of PTGS relieves silencing in tobacco plants containing MAR constructs, but not in controls without MARs. Thus, most of the available data are consistent with the hypothesis that MARs protect against transcriptional silencing, but do not affect PTGS (Butaye et al., 2004).
6.6.2
MARs have a dual effect
Both position effects and gene silencing are known to give rise to variegated or patchy expression (Meyer et al., 1992; Neuhuber et al., 1994; Van Blokland et al., 1994; Ten Lohuis et al., 1995; Chinn et al., 1996; Que et al., 1997). Most often, however, assays are conducted in a way that averages large populations of cells, entire leaves, etc., and it is usually impossible to distinguish changes in the percentage of cells affected from changes in the intensity of expression in individual cells. This question is important to the interpretation of MAR effects as well. If MARs reduce the statistical likelihood of gene silencing in each cell generation, we might expect an increase in the percentage of cells with detectable expression, while a uniform increase in the intensity of expression in all cells might reflect other mechanisms, such as an increase in chromatin availability or polymerase loading. We have addressed this problem by adapting flow cytometry to analyze green fluorescent protein (GFP) expression on a cell-by-cell basis in populations of tobacco protoplasts (Halweg et al., 2005). In the past, such measurements have been complicated by the fact that plant protoplasts lose GFP when membrane integrity is compromised, leading to artificially high estimates of nonexpressing cells. However, we were able to take advantage of the fact that compromised cells are selectively stained with propidium iodide to eliminate this population from the analysis. The resulting profiles provide a good indication of the statistical profile of GFP expression. Figure 6.1 shows an example of data from these experiments. Clearly, a major effect of MARs is to decrease the fraction of cells that appear to be silenced – i.e., the fraction of cells that show little or no GFP expression. This result is consistent with the idea that MARs decrease the likelihood of silencing, rather than increasing expression in unsilenced cells. In addition, however, the GFP profiles for the MAR lines in Figure 6.1 have a shoulder on the high end, indicating the presence of a subpopulation of cells in which expression is higher than in any of the cells in
MARs AND GENE SILENCING
149
Figure 6.1 The RB7 MAR increases the likelihood and magnitude of GFP expression in cultured cells. NT1 tobacco suspension cells were transformed by Agrobacterium containing a CaMV:smRSGFP reporter cassette. The cassette was flanked either by two copies of the RB7 MAR (MGFPM, dark grey) or by two copies of an 1194-bp fragment of lambda DNA, which served as a spacer control (LGFPL, light grey). The black line represents control cells transformed with the GFP cassette alone. Each line represents average fluorescence histograms of 30 individual cell lines, each of which was measured individually by flow cytometry. GFP expression, measured in relative fluorescence units (RFU), is shown on the X axis and the number of cells with a given fluorescence intensity is shown on the Y axis. Cells with fluorescence lower than 17 RFU (vertical line on X axis separating negative from positive) are considered negative for GFP expression, as untransformed controls display fluorescence at this level. (This figure is reproduced, with permission from the American Society of Plant Biologists, from Figure 6A in Halweg et al. (2005).
a control population. Thus, MARs appear to have a dual effect. They reduce the fraction of cells in which silencing occurs, while simultaneously promoting higher levels of expression in some cells.
6.6.3
Hypotheses to explain MAR effects
Early models to explain MAR effects on animal cells postulated that MARs functioned as boundary elements, anchoring the ends of chromosomal domains and preventing the spread of heterochromatin into transgenes flanked by MARs. However, this class of model works best in the context of position effects, such as the canonical position effect variegation phenomenon in Drosophila (Dorer and Henikoff, 1994; Csink and Henikoff, 1996; Dernburg et al., 1996; Walters et al., 1996; Girard et al., 1998). As noted above, much of the variation in transgene expression that occurs in plant cells seems to involve homology-dependent gene silencing, rather than position effects. This fact requires us to consider alternatives to the boundary element model. We have previously discussed several possible models (Allen et al., 2000); only two will be mentioned here. First, the ‘carpet tack’ hypothesis postulates that MARs
150
REGULATION OF TRANSCRIPTION IN PLANTS
Figure 6.2 The terminator hypothesis. The model envisions complex loci containing transgene and vector DNA in various orientations. (A) In the absence of MARs, transcription initiates correctly but in cases of poor termination continues into the surrounding DNA, which can produce promoter transcripts or inverted repeats and activate one or more RNA-silencing pathways. (B) When MARs flank a transgene, transcriptional read-through and subsequent silencing may be reduced if the MARs function to terminate transcription. An extension of the terminator hypothesis (not shown) envisions transcripts initiating in surrounding chromatin and extending into the transgene locus, with subsequent silencing. MARs might also prevent such transcripts from reaching the transgene.
keep a transgene locus physically close to the nuclear matrix, where there may be higher concentrations of RNA polymerase and transcription factors, and that they may also constrain the chromatin fiber to reduce pairing interactions within complex multicopy loci. Another attractive hypothesis postulates that MARs act as transcriptional terminators, as illustrated in Figure 6.2. This terminator function might prevent transcription from extending from one transgene to another in a complex locus (reducing the potential for inverted repeat formation in the transcripts) or from the transgene into adjoining repeated sequences in the genome. Repeat transcripts and dsRNA that may be produced in this manner might attract the RITS complex and cause heterochromatin formation and TGS, while MAR effects might result from minimizing the production of dsRNA or other aberrant RNAs. Support for this class of model comes from a report by Nap’s group (Mlyn´arov´a et al., 2003), who used site-specific recombinases to excise a MAR from a transgenic locus in tobacco after insertion and initial analysis. After excision of the MAR, the locus became more vulnerable to silencing. The authors were able to detect transcription extending into adjacent repeated DNA, as well as small dsRNAs similar to those associated with RNA-mediated silencing. Both of these hypotheses are consistent with results indicating that MARs work more effectively in cis than in trans. Vaucheret et al. (1998) showed that strong trans-silencing loci could also silence genes flanked by MARs. Ascenzi et al. (2003) detected a protective effect of MARs against other silencing loci, weaker than those
MARs AND GENE SILENCING
151
tested by Vauchert et al., but confirmed the major observation that MAR effects are most prominent in cis. An additional prediction of our hypothesis is that MARs would protect the transcriptional potential of genes with low expression levels, whereas actively transcribed genes would be susceptible to silencing by PTGS. To test this hypothesis, we used a tetracycline inducible promoter system (Abranches et al., 2005) in which a noninduced luciferase reporter transgene was stably transformed into tobacco NT1 cells (Nicotiana tabacum), which were grown for many division cycles. The cells were then induced and transgene expression was compared between the control and the MAR-flanked transgenes. Nearly 85% of the lines with the MAR-flanked transgenes could be induced, in contrast to only 40% induction observed in the two groups of control lines (Abranches et al., 2005).
6.7
MAR effects in Arabidopsis
Up to this point, our discussion has focused primarily on work with tobacco and other plants that have relatively large, repetitive genomes. Given the importance of Arabidopsis as a model for plant molecular genetics, we and others have been interested in examining MAR effects in this species as well. At this time, in wildtype Arabidopsis, there are three reports showing either no effect or a negative effect of MARs (Holmes-Davis and Comai, 2002; De Bolle et al., 2003; Butaye et al., 2004) and one report of a two- to fourfold increase in expression (Van der Geest et al., 2004) (Table 6.1). Might this small effect be because there is little transcriptional silencing in this extremely small, mostly nonrepetitive genome? Position effects have been reported to be minimal in Arabidopsis (Schubert et al., 2004; Nagaya et al., 2005), but examples of silencing, both TGS and PTGS, are widely reported (Mittelsten Scheid and Paszkowski, 2000; Melquist and Bender, 2003; Probst et al., 2003) and Arabidopsis mutants are, in fact, the main tools for molecular genetic analysis of silencing (Amedeo et al., 2000; Mittelsten Scheid and Paszkowski, 2000; Morel et al., 2000; Mourrain et al., 2000; Steimer et al., 2000; Miura et al., 2001; Beclin et al., 2002; Tariq et al., 2002; Turnage et al., 2002; Vaistij et al., 2002; Gazzani et al., 2004; Muangsan et al., 2004; Kidner and Martienssen, 2005; Yoshikawa et al., 2005). However, it remains possible that TGS is less important – at least relative to PTGS – in Arabidopsis than in other species. In addition, differences in analytical methods may also play a role. Gene expression assays in Arabidopsis, unlike those with other model systems, commonly involve extracting whole plants rather than isolated organs, and are often carried out at advanced developmental stages. Our previous work with tobacco had shown that suppressors of PTGS could relieve silencing of MAR constructs, but not control constructs lacking MARs, leading to the supposition that strong PTGS could ‘hide’ the otherwise significant MAR effects at the transcriptional level (Ascenzi et al., 2003). We therefore set out to test the hypothesis that PTGS might suppress gene expression in mature Arabidopsis plants
152
REGULATION OF TRANSCRIPTION IN PLANTS
so that MAR effects would not be detected, whereas they might be detected in young seedlings or certain growth conditions in which PTGS activity is lessened.
6.7.1
Testing the effect of developmental stage
We used the RB7 MAR from tobacco, as in our previous work, and carried out Agrobacterium-mediated transformation by the floral dip procedure (Clough and Bent, 1998). Agrobacterium transformation is expected to give mostly single- and low-copy insertions with little rearrangement. Our expectation is that it should therefore lead to less gene silencing in control constructs and corresponding smaller MAR effects than those that we have seen with microprojectile transformation. By using a luciferase reporter (Mankin et al., 1997) we were able to examine transgene expression in the same plants as they progressed through different developmental stages. Our results (Callaway et al., in preparation) showed a strong influence of developmental stage. MAR effects at the seeding stage were consistently positive, averaging about threefold. At the rosette stage, however, the MAR effect was more variable, and a statistically robust effect was not detected. Later in development, at the bolting stage, we saw a statistically significant negative MAR effect. In another study, this time with GUS as a reporter gene (Francis et al., in preparation), we found that the RB7 MAR increased reporter gene expression in leaf tissues of Arabidopsis plants measured 14 days after germination. The increases in transgene expression were highly statistically significant, but as expected they were only in the twofold range as compared to the tenfold or greater increases observed in tobacco cells in culture. The MAR-mediated increases in reporter gene expression were lost when measurements were made on leaves of 28-day-old plants, and by 42 days a negative MAR effect was sometimes noted. These effects were consistent in T2, T3, and T4 generations of transgenic Arabidopsis plants.
6.7.2
How might MARs decrease net transgene expression?
How can we account for developmental variation in the magnitude of MAR effects? One hypothesis is based on the observations that suppressors of post-transcriptional silencing (Ascenzi et al., 2003) or mutants that block PTGS function ((Butaye et al., 2004) can ‘rescue’ expression that would otherwise be masked by posttranscriptional silencing. When coupled with the observations of Mitsuhara et al. (2002) these results provide a likely explanation of the developmental variation we see. Mitsuhara et al. showed that PTGS of a luciferase transgene in tobacco is greatly reduced in proliferating tissues such as meristematic tissue and developing leaves. In young seedlings, therefore, we expect MAR effects on transcriptional activity to be reflected in reporter gene expression, with relatively little interference from PTGS. Later in development, however, the onset of severe PTGS in mature tissues may counteract transcriptional stimulation, so that MAR effects are no longer observed. Mitsuhara et al. (2002) also showed that PTGS was reduced in callus cultures, a result that may help to explain why some of the largest MAR effects have been seen in rapidly dividing suspension cultures (Allen et al., 1993, 1996, 2000).
MARs AND GENE SILENCING
153
Interactions between transcriptional and post-transcriptional effects may also explain the instances in which we see a negative influence of MARs. Suppose, for example, that strong PTGS is triggered when transcription (or mRNA production) exceeds a certain threshold. If MARs increase transcription (or reduce TGS), transgene mRNA production might exceed the threshold more often in the case of MAR constructs than in controls. In such cases, a positive MAR effect on transcription would lead to increased PTGS and a decrease in steady state mRNA abundance.
6.7.3
MAR effects in PTGS mutants
Recently, the relative importance of MAR effects and PTGS has been systematically addressed by Butaye et al. (2004), who analyzed expression of MAR constructs in Arabidopsis plants carrying the either sgs2 or sgs3 (suppressor of gene silencing) mutations, which block PTGS. Although little or no MAR effect could be seen in wild-type plants, mutant lines transformed with the same constructs showed large (ca. fivefold) MAR effects. The use of MARs with PTGS mutations had a multiplicative effect, so that transgene expression was increased 35-fold relative to wild-type controls. This combination, which is currently available for licensing as the ‘MARs Plus’ system (http://www.pbltechnology.com/), seems to hold great promise for protein expression or metabolic engineering studies.
6.8
Conclusions
In summary, our own data and our reading of the literature lead us to conclude that MARs increase both the frequency and the intensity of transgene expression in populations of transformants. We believe this effect is mostly or entirely at the level of TGS, and that it is heritable over multiple plant generations. The observed effect of MARs on mRNA abundance or reporter gene expression is strongly affected by variations in post-transcriptional silencing, which can be strongly influenced by developmental processes as well as by viral suppressor proteins and mutations in host genes. Under certain circumstances, MAR-mediated increases in transcription may trigger enough PTGS to create an overall negative effect on gene expression. However, by combining MAR constructs with mutations or suppressors of PTGS it is possible to achieve high levels of protein expression that are expected to be developmentally and genetically stable.
References J.A. Able, C. Rathhus, B.J. Carroll and I.D. Godwin (2004) Sorghum Tissue Culture and Transformation (eds N. Seetharama and I. Godwin), p. 168. Science Publishers, Enfield, NH. R. Abranches, R.W. Shultz, W.F. Thompson and G.C. Allen (2005) Matrix attachment regions and regulated transcription increase and stabilize transgene expression. Plant Biotechnology Journal 3, 535–543.
154
REGULATION OF TRANSCRIPTION IN PLANTS
G.C. Allen, G. Hall, S. Michalowski, et al. (1996) High-level transgene expression in plant cells: effects of a strong scaffold attachment region from tobacco. The Plant Cell 8, 899–913. G.C. Allen, G.E. Hall, L.C. Childs, A.K. Weissinger, S. Spiker and W.F. Thompson (1993) Scaffold attachment regions increase reporter gene expression in stably transformed plant cells. The Plant Cell 5, 603–613. G.C. Allen, S. Spiker and W.F. Thompson (2000) Use of matrix attachment regions (MARs) to minimize transgene silencing. Plant Molecular Biology 43, 361–376. B.B. Amati and S.M. Gasser (1988) Chromosomal ARS and CEN elements bind specifically to yeast nuclear scaffold. Cell 54, 967–978. P. Amedeo, Y. Habu, K. Afsar, O. Mittelsten Scheid and J. Paszkowski (2000) Disruption of the plant gene MOM releases transcriptional silencing of methylated genes. Nature 405, 203–206. S. Annadana, L. Mlyn´arov´a, M. Udayakumar, J. de Jong and J.P. Nap (2002) The potato Lhca3.St.1 promoter confers high and stable transgene expression in chrysanthemum, in contrast to CaMVbased promoters. Molecular Breeding 8, 335–344. ¨ R. Ascenzi, B. Ulker, J.J. Todd, et al. (2003) Analysis of trans-silencing interactions using transcriptional silencers of varying strength and targets with and without flanking nuclear matrix attachment regions. Transgenic Research 12, 305–318. C. Beclin, S. Boutet, P. Waterhouse and H. Vaucheret (2002) A branched pathway for transgene-induced RNA silencing in plants. Current Biology 12, 684–688. J. Bender (2004) Chromatin-based silencing mechanisms. Current Opinion in Plant Biology 7, 521– 526. C. Benham, T. Kohwi-Shigematsu and J. Bode (1997) Stress-induced duplex DNA destabilization in scaffold/matrix attachment regions. Journal of Molecular Biology 274, 181–196. J. Bode, C. Benham, A. Knopp and C. Mielke (2000) Transcriptional augmentation: modulation of gene expression by scaffold/matrix-attached regions (S/MAR elements). Critical Reviews in Eukaryotic Gene Expression 10, 73–90. J. Bode, S. Winkelmann, S. Gotze, S. Spiker, K. Tsutsui, C. Bi, A.K. Prashanth and C. Benham (2006) Correlations between scaffold/matrix attachment region (S/MAR) binding activity and DNA duplex destabilization energy. Journal of Molecular Biology 358, 597–613. C. Bonifer, N. Yannoutsos, G. Kruger, F. Grosveld and A.E. Sippel (1994) Dissection of the locus control function located on the chicken lysozyme gene domain in transgenic mice. Nucleic Acids Research 22, 4202–4210. T. Boulikas (1993) Nature of DNA sequences at the attachment regions of genes to the nuclear matrix. Journal of Cellular Biochemistry 52, 14–22. T. Boulikas (1995) Chromatin domains and prediction of MAR sequences. International Reviews in Cytology 162A, 279–388. V. Bourdon, Z. Ladbrooke, A. Wickham, D. Lonsdale and W. Harwood (2002) Homozygous transgenic wheat plants with increased luciferase activity do not maintain their high level of expression in the next generation. Plant Science 163, 297–305. P. Bregitzer and D. Tonks (2003) Inheritance and expression of transgenes in barley. Crop Science 43, 4–12. P. Breyne, M. Van Montague, A. Depicker and G. Gheysen (1992) Characterization of a plant scaffold attachment region in a DNA fragment that normalizes transgene expression in tobacco. The Plant Cell 4, 463–471. C. Brouwer, W. Bruce, S. Maddock, Z. Avramova and B. Bowen (2002) Suppression of transgene silencing by matrix attachment regions in maize: a dual role for the maize 5 ADH1 matrix attachment region. The Plant Cell 14, 2251–2264. C. Brun, Q. Dang and R. Miassod (1990) Studies of an 800-kilobase DNA stretch of the Drosophila X chromosome: comapping of a subclass of scaffold-attached regions with sequences able to replicate autonomously in Saccharomyces cerevisiae. Molecular and Cellular Biology 10, 5455–5463. K.M.J. Butaye, I.J.W.M. Goderis, P.F.J. Wouters, et al. (2004) Stable high-level transgene expression in Arabidopsis thaliana using gene silencing mutants and matrix attachment regions. The Plant Journal 39, 440–449.
MARs AND GENE SILENCING
155
H.P. Cam, T. Sugiyama, E.S. Chen, X. Chen, P.C. FitzGerald and S.I.S. Grewal (2005) Comprehensive analysis of heterochromatin- and RNAi-mediated epigenetic control of the fission yeast genome. Nature Genetics 37, 809–819. Z.Q. Cheng, J. Targolli and R. Wu (2001) Tobacco matrix attachment region sequence increased transgene expression levels in rice plants. Molecular Breeding 7, 317–327. A.M. Chinn, S.R. Payne and L. Comai (1996) Variegation and silencing of the heat shock cognate 80 gene are relieved by a bipartite downstream regulatory element. The Plant Journal 9, 325–339. S.J. Clough and A.F. Bent (1998) Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. The Plant Journal 16, 735–743. P.D. Cluster, M. O’Dell, M. Metzlaff and R.B. Flavell (1996) Details of T-DNA structural organization from a transgenic petunia population exhibiting co-suppression. Plant Molecular Biology 32, 1197–1203. P.N. Cockerill and W.T. Garrard (1986) Chromosomal loop anchorage of the kappa immunoglobulin gene occurs next to the enhancer in a region containing topoisomerase II sites. Cell 44, 273–282. A.J. Conner, L. Mlyn´arov´a, W.J. Stiekema and J.P. Nap (1998) Meiotic stability of transgene expression is unaffected by flanking matrix-associated regions. Molecular Breeding 4, 47–58. A.K. Csink and S. Henikoff (1996) Genetic modification of heterochromatic association and nuclear organization in Drosophila. Nature 381, 529–531. C.D. Day, E. Lee, J. Kobayashi, L.D. Holappa, H. Albert and D.W. Ow (2000) Transgene integration into the same chromosome location can produce alleles that express at a predictable level, or alleles that are differentially silenced. Genes and Development 14, 2869–2880. M.F.C. De Bolle, K.M.J. Butaye, W.J.W. Coucke, et al. (2003) Analysis of the influence of promoter elements and a matrix attachment region on the inter-individual variation of transgene expression in populations of Arabidopsis thaliana. Plant Science 165, 169–179. S. De Buck, M. Van Montagu and A. Depicker (2001) Transgene silencing of invertedly repeated transgenes is released upon deletion of one of the transgenes involved. Plant Molecular Biology 46, 433–445. S. De Buck, P. Windels, M. De Loose and A. Depicker (2004) Single-copy T-DNAs integrated at different positions in the Arabidopsis genome display uniform and comparable beta-glucuronidase accumulation levels. Cellular and Molecular Life Sciences 61, 2632–2645. A.F. Dernburg, K.W. Broman, J.C. Fung, et al. (1996) Perturbation of nuclear architecture by long-distance chromosome interactions. Cell 85, 745–759. D.R. Dorer and S. Henikoff (1994) Expansions of transgene repeats cause heterochromatin formation and gene silencing in Drosophila. Cell 77, 993–1002. K.E. Francis and S. Spiker (2005) Identification of Arabidopsis thaliana transformants without selection reveals a high occurrence of silenced T-DNA integrations. The Plant Journal 41, 464–477. P. Fransz, W. Soppe and I. Schubert (2003) Heterochromatin in interphase nuclei of Arabidopsis thaliana. Chromosome Research 11, 227–240. Y. Fukuda and S. Nishikawa (2003) Matrix attachment regions enhance transcription of a downstream transgene and the accessibility of its promoter region to micrococcal nuclease. Plant Molecular Biology 51, 665–675. S.M. Gasser and U.K. Laemmli (1986a) Cohabitation of scaffold binding regions with upstream/enhancer elements of three developmentally regulated genes of Drosophila melanogaster. Cell 46, 521–530. S.M. Gasser and U.K. Laemmli (1986b) The organisation of chromatin loops: characterization of a scaffold attachment site. The EMBO Journal 5, 511–518. S. Gazzani, T. Lawrenson, C. Woodward, D. Headon and R. Sablowski (2004) A link between mRNA turnover and RNA interference in Arabidopsis. Science 306, 1046–1048. P. Gepts (2002) A comparison between crop domestication, classical plant breeding, and genetic engineering. Crop Science 42, 1780–1790. F. Girard, B. Bello, U.K. Laemmli and W.J. Gehring (1998) In vivo analysis of scaffold-associated regions in Drosophila: a synthetic high-affinity SAR binding protein suppresses position effect variegation. The EMBO Journal 17, 2079–2085.
156
REGULATION OF TRANSCRIPTION IN PLANTS
G.V. Glazko, I.B. Rogozin and M.V. Glazkov (2001) Comparative study and prediction of DNA fragments associated with various elements of the nuclear matrix. Biochimica et Biophysica Acta 1517, 351–364. H.S. Guo, J.F. Fei, Q. Xie and N.H. Chua (2003) A chemical-regulated inducible RNAi system in plants. The Plant Journal 34, 383–392. G.E. Hall, Jr., G.C. Allen, D.S. Loer, W.F. Thompson and S. Spiker (1991) Nuclear scaffolds and scaffold-attachment regions in higher plants. Proceedings of the National Academy of Sciences of the United States of America 88, 9320–9324. C. Halweg, W.F. Thompson and S. Spiker (2005) The Rb7 matrix attachment region increases the likelihood and magnitude of transgene expression in tobacco cells: a flow cytometric study. The Plant Cell 17, 418–429. K.H. Han, C.P. Ma and S.H. Strauss (1997) Matrix attachment regions (MARs) enhance transformation frequency and transgene expression in poplar. Transgenic Research 6, 415–420. R. Hancock (2000) A new look at the nuclear matrix. Chromosoma 109, 219–225. R. Hancock (2004) Internal organisation of the nucleus: assembly of compartments by macromolecular crowding and the nuclear matrix model. Biology of the Cell 96, 595–601. S. Henikoff (1998) Conspiracy of silence among repeated transgenes. Bioessays 20, 532–535. A.J. Herr, M.B. Jensen, T. Dalmay and D.C. Baulcombe (2005) RNA polymerase IV directs silencing of endogenous DNA. Science 308, 118–120. R. Holmes-Davis and L. Comai (2002) The matrix attachment regions (MARs) associated with the heat shock cognate 80 gene (HSC80) of tomato represent specific regulatory elements. Molecular Genetics and Genomics 266, 891–898. J. Hsieh and A. Fire (2000) Recognition and silencing of repeated DNA. Annual Reviews in Genetics 34, 187–204. M.C. Huber, F.X. Bosch, A.E. Sippel and C. Bonifer (1994) Chromosomal position effects in chicken lysozyme gene transgenic mice are correlated with suppression of DNase I hypersensitive site formation. Nucleic Acids Research 22, 4195–4201. V.A. Iglesias, E.A. Moscone, I. Papp, et al. (1997) Molecular and cytogenetic analyses of stably and unstably expressed transgene loci in tobacco. The Plant Cell 9, 1251–1264. L.M. Iyer, S.P. Kumpatla, M.B. Chandrasekharan and T.C. Hall (2000) Transgene silencing in monocots. Plant Molecular Biology 43, 323–346. D. Jackson (2005) Understanding nuclear organization: when information becomes knowledge. Workshop on nuclear organization. EMBO Reports 6, 213–217. D.A. Jackson (2003) The principles of nuclear structure. Chromosome Research 11, 387–401. S.E. Jacobsen (1999) Gene silencing: maintaining methylation patterns. Current Biology 9, R617–R619. R.A. Jorgensen (2003) Sense cosuppression in plants: past, present, and future. In: RNAi: A guide to gene silencing (ed. G.J. Hannon), pp. 5–22. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. A. Kerschen, C.A. Napoli, R.A. Jorgensen and A.E. Muller (2004) Effectiveness of RNA interference in transgenic plants. FEBS Letters 566, 223–228. C.A. Kidner and R.A. Martienssen (2005) The role of ARGONAUTE1 (AGO1) in meristem formation and identity. Developmental Biology 280, 504–517. W.S. Kim, J. Kim, H.B. Krishnan and B.H. Nahm (2005) Expression of Escherichia coli branching enzyme in caryopses of transgenic rice results in amylopectin with an increased degree of branching. Planta 220, 689–695. C. Lacomme, K. Hrubikova and I. Hein (2003) Enhancement of virus-induced gene silencing through viral-based production of inverted-repeats. The Plant Journal 34, 543–553. V. Levee, E. Garin, K. Klimaszewska and A. Seguin (1999) Stable genetic transformation of white pine (Pinus strobus L.) after cocultivation of embryogenic tissues with Agrobacterium tumefaciens. Molecular Breeding 5, 429–440. J.S. Levin, W.F. Thompson, A.S. Csinos, M.G. Stephenson and A.K. Weissinger (2005) Matrix attachment regions increase the efficiency and stability of RNA-mediated resistance to tomato spotted wilt virus in transgenic tobacco. Transgenic Research 14, 193–206.
MARs AND GENE SILENCING
157
I. Liebich, J. Bode, I. Reuter and E. Wingender (2002) Evaluation of sequence motifs found in scaffold/matrix-attached regions (S/MARs). Nucleic Acids Research 30, 3433–3442. J.A. Lindbo, L. Silva-Rosales, W.M. Proebsting and W.G. Doughtery (1993) Induction of a highly specific antiviral state in transgenic plants: implications for regulation of gene expression and virus resistance. The Plant Cell 5, 1749–1759. Z. Lippman, B. May, C. Yordan, T. Singer and R. Martienssen (2003) Distinct mechanisms determine transposon inheritance and methylation via small interfering RNA and histone modification. PLoS Biology 1, E67. J.W. Liu and L.M. Tabe (1998) The influences of two plant nuclear matrix attachment regions (MARs) on gene expression in transgenic plants. Plant and Cell Physiology 39, 115–123. A. Lorence and R. Verpoorte (2004) Gene transfer and expression in plants. Methods in Molecular Biology 267, 329–350. L.M. Mankin, G.C. Allen and W.F. Thompson (1997) Introducton of a plant intron into the luciferase gene of Photinus pyralis. Plant Molecular Biology Reporter 15, 186–196. S.L. Mankin, G.C. Allen, T. Phelan, S. Spiker and W.F. Thompson (2003) Elevation of transgene expression level by flanking matrix attachment regions (MAR) is promoter dependent: a study of the interactions of six promoters with the RB7 3 MAR. Transgenic Research 12, 3–12. A.M. Martelli, E. Falcieri, M. Zweyer, et al. (2002) The controversial nuclear matrix: a balanced point of view. Histology and Histopathology 17, 1193–1205. R.A. Martienssen (2003) Maintenance of heterochromatin by RNA interference of tandem repeats. Nature Genetics 35, 213–214. T. Masaki, H. Tsukagoshi, N. Mitsui, et al. (2005) Activation tagging of a gene for a protein with novel class of CCT-domain activates expression of a subset of sugar-inducible genes in Arabidopsis thaliana. The Plant Journal 43, 142–152. M.A. Matzke, W. Aufsatz, T. Kanno, M.F. Mette and A.J. Matzke (2002) Homology-dependent gene silencing and host defense in plants. Advances in Genetics 46, 243–275. M.A. Matzke and J.A. Birchler (2005) RNAi-mediated pathways in the nucleus. Nature Reviews Genetics 6, 24–35. M.A. Matzke, M.F. Mette and A.J.M. Matzke (2000) Transgene silencing by the host genome defense: implications for the evolution of epigenetic control mechanisms in plants and vertebrates. Plant Molecular Biology 43, 401–415. M.A. Matzke, M. Primig, J. Trnovsky and A.J.M. Matzke (1989) Reversible methylation and inactivation of marker genes in sequentially transformed tobacco plants. The EMBO Journal 8, 643–649. S. Maximova, C. Miller, G. Antunez de Mayolo, S. Pishak, A. Young and M.J. Guiltinan (2003) Stable transformation of Theobroma cacao L. and influence of matrix attachment regions on GFP expression. Plant Cell Reports 21, 872–883. F. Meins, Jr. (2000) RNA degradation and models for post-transcriptional gene-silencing. Plant Molecular Biology 43, 261–273. S. Melquist and J. Bender (2003) Transcription from an upstream promoter controls methylation signaling from an inverted repeat of endogenous genes in Arabidopsis. Genes and Development 17, 2036–2047. N. Mendu, M. Massel and S. Spiker (2001) Increasing loop domain size does not diminish effects of matrix attachment regions on transgene expression in tobacco cells culture. FEBS Letters 496, 66–67. M.F. Mette, J. vanderWinden, M.A. Matzke and A.J.M. Matzke (1999) Production of aberrant promoter transcripts contributes to methylation and silencing of unlinked homologous promoters in trans. The EMBO Journal 18, 241–248. P. Meyer (2000) Transcriptional transgene silencing and chromatin components. Plant Molecular Biology 43, 221–234. P. Meyer, F. Linn, I. Heidmann, H. Meyer, I. Niedenhof and H. Saedler (1992) Endogenous and environmental factors influence 35S promoter methylation of a maize A1 gene construct in transgenic petunia and its colour phenotype. Molecular and General Genetics 231, 345–352. P. Meyer and H. Saedler (1996) Homology-dependent gene silencing in plants. Annual Review of Plant Physiology and Plant Molecular Biology 47, 23–48.
158
REGULATION OF TRANSCRIPTION IN PLANTS
S.M. Michalowski, G.C. Allen, G.E. Hall, W.F. Thompson and S. Spiker (1999) Characterization of randomly-obtained matrix attachment regions (MARs) from higher plants. Biochemistry 38, 12795–12804. I. Mitsuhara, N. Shirasawa-Seo, T. Iwai, S. Nakamura, R. Honkura and Y. Ohashi (2002) Release from post-transcriptional gene silencing by cell proliferation in transgenic tobacco plants: possible mechanism for noninheritance of the silencing. Genetics 160, 343–352. O. Mittelsten Scheid and J. Paszkowski (2000) Transcriptional gene silencing mutants. Plant Molecular Biology 43, 235–241. A. Miura, S. Yonebayashi, K. Watanabe, T. Toyama, H. Shimada and T. Kakutani (2001) Mobilization of transposons by a mutation abolishing full DNA methylation in Arabidopsis. Nature 411, 212–214. L. Mlyn´arov´a, A. Hricova, A. Loonen and J.P. Nap (2003) The presence of a chromatin boundary appears to shield a transgene in tobacco from RNA silencing. The Plant Cell 15, 2203–2217. L. Mlyn´arov´a, R.C. Jansen, A.J. Conner, W.J. Stiekema and J.P. Nap (1995) The MAR-mediated reduction in position effect can be uncoupled from copy number-dependent expression in transgenic plants. The Plant Cell 7, 599–609. L. Mlyn´arov´a, A. Loonen, J. Heldens, et al. (1994) Reduced position effect in mature transgenic plants conferred by the chicken lysozyme matrix-associated region. The Plant Cell 6, 417–426. L. Mlyn´arov´a, A. Loonen, E. Mietkiewska, R.C. Jansen and J.P. Nap (2002) Assembly of two transgenes in an artificial chromatin domain gives highly coordinated expression in tobacco. Genetics 160, 727–740. J.B. Morel, P. Mourrain, C. Beclin and H. Vaucheret (2000) DNA methylation and chromatin structure affect transcriptional and post-transcriptional transgene silencing in Arabidopsis. Current Biology 10, 1591–1594. M.R. Motamedi, A. Verdel, S.U. Colmenares, S.A. Gerber, S.P. Gygi and D. Moazed (2004) Two RNAi complexes, RITS and RDRC, physically interact and localize to noncoding centromeric RNAs. Cell 119, 789–802. P. Mourrain, C. Beclin, T. Elmayan, et al. (2000) Arabidopsis SGS2 and SGS3 genes are required for posttranscriptional gene silencing and natural virus resistance. Cell 101, 533–542. N. Muangsan, C. Beclin, H. Vaucheret and D. Robertson (2004) Geminivirus VIGS of endogenous genes requires SGS2/SDE1 and SGS3 and defines a new branch in the genetic pathway for silencing in plants. The Plant Journal 38, 1004–1014. A. Muller, M. Marins, Y. Kamisugi and P. Meyer (2002) Analysis of hypermethylation in the RPS element suggests a signal function for short inverted repeats in de novo methylation. Plant Molecular Biology 48, 383–399. M.W.M. Muskens, A.P.A. Vissers, J.N.M. Mol and J.M. Kooter (2000) Role of inverted DNA repeats in transcriptional and post-transcriptional gene silencing. Plant Molecular Biology 43, 243–260. S. Nagaya, K. Kato, Y. Ninomiya, et al. (2005) Expression of randomly integrated single complete copy transgenes does not vary in Arabidopsis thaliana. Plant and Cell Physiology 46, 438–444. S.J. Namciu, R.D. Friedman, M.D. Marsden, L.M. Sarausad, C.L. Jasoni and R.E. Fournier (2004) Sequence organization and matrix attachment regions of the human serine protease inhibitor gene cluster at 14q32.1. Mammalian Genome 15, 162–178. C. Napoli, C. Lemieux and R. Jorgensen (1990) Introduction of a chimeric chalcone synthase gene into petunia results in reversible co-suppression of homologous genes in trans. The Plant Cell 2, 279–289. F. Neuhuber, Y.D. Park, A.J.M. Matzke and M.A. Matzke (1994) Susceptibility of transgene loci to homology-dependent gene silencing. Molecular and General Genetics 244, 230–241. J. Nickerson (2001) Experimental observations of a nuclear matrix. Journal of Cell Science 114, 463–474. K. Noma, T. Sugiyama, H. Cam, et al. (2004) RITS acts in cis to promote RNA interference-mediated transcriptional and post-transcriptional silencing. Nature Genetics 36, 1174–1180. W. Nowak, M. Gawlowska, A. Jarmolowski and J. Augustyniak (2001) Effect of nuclear matrix attachment regions on transgene expression in tobacco plants. Acta Biochimica Polonica 48, 637–646.
MARs AND GENE SILENCING
159
J.T. Odell and E. Krebbers (1998) Enhanced transgene expression in a population of monocot cells employing scaffold attachment regions. World Patent Office, World Patent, Dupont. S.J. Oh, J.S. Jeong, E.H. Kim, et al. (2005) Matrix attachment region from the chicken lysozyme locus reduces variability in transgene expression and confers copy number-dependence in transgenic rice plants. Plant Cell Reports 24, 145–154. M. Pal-Bhadra, B.A. Leibovitch, S.G. Gandhi, et al. (2004) Heterochromatic silencing and HP1 localization in Drosophila are dependent on the RNAi machinery. Science 303, 669–672. X. Pan, Y. Li and L. Stein (2005) Site preferences of insertional mutagenesis agents in Arabidopsis. Plant Physiology 137, 168–175. Y.D. Park, I. Papp, E.A. Moscone, et al. (1996) Gene silencing mediated by promoter homology occurs at the level of transcription and results in meiotically heritable alterations in methylation and gene activity. The Plant Journal 9, 183–194. J.S. Parker, S.M. Roe and D. Barford (2004) Crystal structure of a PIWI protein suggests mechanisms for siRNA recognition and slicer activity. The EMBO Journal 23, 4727–4737. A. Pecinka, N. Kato, A. Meister, A.V. Probst, I. Schubert and E. Lam (2005) Tandem repetitive transgenes and fluorescent chromatin tags alter local interphase chromosome arrangement in Arabidopsis thaliana. Journal of Cell Science 118, 3751–3758. T. Pederson (1998) Thinking about a nuclear matrix. Journal of Molecular Biology 277, 147–159. T. Pederson (2000) Half a century of ‘the nuclear matrix’. Molecular Biology of the Cell 11, 799–805. K. Petersen, R. Leah, S. Knudsen and V. Cameron-Mills (2002) Matrix attachment regions (MARs) enhance transformation frequencies and reduce variance of transgene expression in barley. Plant Molecular Biology 49, 45–58. L. Phi-Van and W.H. Str¨atling (1996) Dissection of the ability of the chicken lysozyme gene 5 matrix attachment region to stimulate transgene expression and to dampen position effects. Biochemistry 35, 10735–10742. D. Pontier, G. Yahubyan, D. Vega, et al. (2005) Reinforcement of silencing at transposons and highly repeated sequences requires the concerted action of two distinct RNA polymerases IV in Arabidopsis. Genes and Development 19, 2030–2040. A.V. Probst, P.F. Fransz, J. Paszkowski and O. Mittelsten Scheid (2003) Two means of transcriptional reactivation within heterochromatin. The Plant Journal 33, 743–749. Q.D. Que, H.Y. Wang, J.J. English and R.A. Jorgensen (1997) The frequency and degree of cosuppression by sense chalcone synthase transgenes are dependent on transgene promoter strength and are reduced by premature nonsense codons in the transgene coding sequence. The Plant Cell 9, 1357–1368. S. Rudd, M. Frisch, K. Grote, B.C. Meyers, K. Mayer and T. Werner (2004) Genome-wide in silico mapping of scaffold/matrix attachment regions in Arabidopsis suggests correlation of intragenic scaffold/matrix attachment regions with gene expression. Plant Physiology 135, 715–722. F. Sch¨offl, G. Schr¨oder, M. Kliem and M. Rieping (1993) An SAR-sequence containing 395 bp-DNA fragment mediates enhanced, gene-dosage-correlated expression of a chimaeric heat shock gene in transgenic tobacco plants. Transgenic Research 2, 93–100. V. Schramke, D.M. Sheedy, A.M. Denli, et al. (2005) RNA-interference-directed chromatin modification coupled to RNA polymerase II transcription. Nature 435, 1275–1279. D. Schubert, B. Lechtenberg, A. Forsbach, M. Gils, S. Bahadur and R. Schmidt (2004) Silencing in Arabidopsis T-DNA transformants: the predominant role of a gene-specific RNA sensing mechanism versus position effects. The Plant Cell 16, 2561–2572. K. Shimizu, M. Takahashi, N. Goshima, S. Kawakami, K. Irifune and H. Morikawa (2001) Presence of an SAR-like sequence in junction regions between an introduced transgene and genomic DNA of cultured tobacco cells: its effect on transformation frequency. The Plant Journal 26, 375–384. L. Sidorenko, W. Bruce, S. Maddock, et al. (2003) Functional analysis of two matrix attachment region (MAR) elements in transgenic maize plants. Transgenic Research 12, 137–154. T. Sijen, I. Vijn, A. Rebocho, et al. (2001) Transcriptional and posttranscriptional gene silencing are mechanistically related. Current Biology 11, 436–440.
160
REGULATION OF TRANSCRIPTION IN PLANTS
G.B. Singh, J.A. Kramer and S.A. Krawetz (1997) Mathematical model to predict regions of chromatin attachment to the nuclear matrix. Nucleic Acids Research 25, 1419–1425. R.K. Slotkin, M. Freeling and D. Lisch (2005) Heritable transposon silencing initiated by a naturally occurring transposon inverted duplication. Nature Genetics 37, 641–644. M. Stam, R. de Bruin, S. Kenter, et al. (1997a) Post-transcriptional silencing of chalcone synthase in petunia by inverted transgene repeats. The Plant Journal 12, 63–82. M. Stam and O. Mittelsten Scheid (2005) Paramutation: an encounter leaving a lasting impression. Trends in Plant Science 10, 283–290. M. Stam, J.N.M. Mol and J.M. Kooter (1997b) The silence of genes in transgenic plants. Annals of Botany 79, 3–12. M. Stam, A. Viterbo, J.N.M. Mol and J.M. Kooter (1998) Position-dependent methylation and transcriptional silencing of transgenes in inverted T-DNA repeats: implications for posttranscriptional silencing of homologous host genes in plants. Molecular and Cellular Biology 18, 6165–6177. A. Steimer, P. Amedeo, K. Afsar, P. Fransz, O. Mittelsten Scheid and J. Paszkowski (2000) Endogenous targets of transcriptional gene silencing in Arabidopsis. The Plant Cell 12, 1165–1178. A. Stief, D.M. Winter, W.H. Str¨atling and A.E. Sippel (1989) A nuclear DNA attachment element mediates elevated and position-independent gene activity. Nature 341, 343–345. P.A. Stoutjesdijk, S.P. Singh, Q. Liu, C.J. Hurlstone, P.A. Waterhouse and A.G. Green (2002) hpRNA-mediated targeting of the Arabidopsis FAD2 gene gives highly efficient and stable silencing. Plant Physiology 129, 1723–1731. M. Tariq, Y. Habu and J. Paszkowski (2002) Depletion of MOM1 in non-dividing cells of Arabidopsis plants releases transcriptional gene silencing. EMBO Reports 3, 951–955. M. Ten Lohuis, A. Muller, I. Heidmann, I. Niedenhof and P. Meyer (1995) A repetitive DNA fragment carrying a hot spot for de novo DNA methylation enhances expression variegation in tobacco and petunia. The Plant Journal 8, 919–932. Y. Tomari and P.D. Zamore (2005) Perspective: machines for RNAi. Genes and Development 19, 517–529. F. Torney, A. Partier, V. Says-Lesage, I. Nadaud, P. Barret and M. Beckert (2004) Heritable transgene expression pattern imposed onto maize ubiquitin promoter by maize adh-1 matrix attachment regions: tissue and developmental specificity in maize transgenic plants. Plant Cell Reports 22, 931–938. M.A. Turnage, N. Muangsan, C.G. Peele and D. Robertson (2002) Geminivirus-based vectors for gene silencing in Arabidopsis. The Plant Journal 30, 107–114. ¨ B. Ulker, G.C. Allen, W.F. Thompson, S. Spiker and A.K. Weissinger (1999) A tobacco matrix attachment region reduces the loss of transgene expression in the progeny of transgenic tobacco plants. The Plant Journal 18, 253–263. P. Vain, V.A. James, B. Worland and J.W. Snape (2002) Transgene behaviour across two generations in a large random population of transgenic rice plants produced by particle bombardment. Theoretical and Applied Genetics 105, 878–889. P. Vain, B. Worland, A. Kohli, et al. (1999) Matrix attachment regions increase transgene expression levels and stability in transgenic rice plants and their progeny. The Plant Journal 18, 233–242. F.E. Vaistij, L. Jones and D.C. Baulcombe (2002) Spreading of RNA targeting and DNA methylation in RNA silencing requires transcription of the target gene and a putative RNA-dependent RNA polymerase. The Plant Cell 14, 857–867. R. Van Blokland, N. Van der Geest, J.N.M. Mol and J.M. Kooter (1994) Transgene-mediated suppression of chalcone synthase expression in petunia hybrida results from an increase in RNA turnover. The Plant Journal 6, 861–877. A.H.M. Van der Geest, G.E. Hall, S. Spiker and T.C. Hall (1994) The beta-phaseolin gene is flanked by matrix attachment regions. The Plant Journal 6, 413–423. A.H.M. Van der Geest, M.E. Welter, A.T. Woosley, et al. (2004) A short synthetic MAR positively affects transgene expression in rice and Arabidopsis. Plant Biotechnology Journal 2, 13–26. A.R. van der Krol, L.A. Mur, M. Beld, J.N.M. Mol and A.R. Stuitje (1990) Flavonoid genes in petunia: addition of a limited number of gene copies may lead to a suppression of gene expression. The Plant Cell 2, 291–299.
MARs AND GENE SILENCING
161
W. Van Leeuwen, L. Mlyn´arov´a, J.P. Nap, L.H. van der Plas and A.R. van der Krol (2001) The effect of MAR elements on variation in spatial and temporal regulation of transgene expression. Plant Molecular Biology 47, 543–554. H. Vaucheret, T. Elmayan, D. Thierry, et al. (1998) Flank matrix attachment regions (MARs) from chicken, bean, yeast or tobacco do not prevent homology-dependent trans-silencing in transgenic tobacco plants. Molecular and General Genetics 259, 388–392. K.J. Venken and H.J. Bellen (2005) Emerging technologies for gene manipulation in Drosophila melanogaster. Nature Reviews Genetics 6, 167–178. D. Verma, M. Verma, M. Dey, R.K. Jain and R. Wu (2005) Molecular dissection of the tobacco Rb7 matrix attachment region (MAR): effect of 5 half on gene expression in rice. Plant Science 169, 704–711. T. Volpe, V. Schramke, G.L. Hamilton, et al. (2003) RNA interference is required for normal centromere function in fission yeast. Chromosome Research 11, 137–146. T.A. Volpe, C. Kidner, I.M. Hall, G. Teng, S.I. Grewal, and R.A. Martienssen (2002) Regulation of heterochromatic silencing and histone H3 lysine-9 methylation by RNAi. Science 297, 1833–1837. M.C. Walters, W. Magis, S. Fiering, et al. (1996) Transcriptional enhancers act in cis to suppress position-effect variegation. Genes and Development 10, 185–195. M. Wassenegger (2005) The role of the RNAi machinery in heterochromatin formation. Cell 122, 13–16. S.V. Wesley, C.A. Helliwell, N.A. Smith, et al. (2001) Construct design for efficient, effective and high-throughput gene silencing in plants. The Plant Journal 27, 581–590. A.P. Wolffe (1997) Transcription control: repressed repeats express themselves. Current Biology 7, R796–R798. Y.-Y. Xu, X.-M. Wang, J. Li, et al. (2005) Activation of the WUS gene induces ectopic initiation of floral meristems on mature stem surface in Arabidopsis thaliana. Plant Molecular Biology 57, 773–784. H. Xue, Y.T. Yang, C.A. Wu, G.D. Yang, M.M. Zhang and C.C. Zheng (2005) TM2, a novel strong matrix attachment region isolated from tobacco, increases transgene expression in transgenic rice calli and plants. Theoretical and Applied Genetics 110, 620–627. S.Y. Yoo, K. Bomblies, S.K. Yoo, et al. (2005) The 35S promoter used in a selectable marker gene of a plant transformation vector affects the expression of the transgene. Planta 221, 523–530. M. Yoshikawa, A. Peragine, M.Y. Park and R.S. Poethig (2005) A pathway for the biogenesis of trans-acting siRNAs in Arabidopsis. Genes and Development 19, 2164–2175.
7 Polymerase I transcription Julio S´aez-V´asquez and Manuel Echeverr´ıa
7.1
Introduction
Regulation of RNA polymerase I (pol I) transcription is an important mechanism in controlling the abundance of ribosomes (Bailey-Serres, 1998; Fatica and Tollervey, 2002; Fromont-Racine et al., 2003), and thereby in controlling the protein synthetic capacity of the cell. In actively growing animal and yeast cells, ribosomal RNA synthesis driven by RNA pol I represents about 50% of total cellular RNA production (Warner, 1999; Raska et al., 2004). Stimulation of rRNA transcription can be due to either a higher rate of transcription of active genes or an increase in the number of active rRNA genes (Warner, 1999; Grummt, 2003; Moss, 2004; Russell and Zomerdijk, 2005). Plant cells can employ diverse mechanisms to control rRNA synthesis, from silencing individual rRNA genes (collectively known as rDNA) to silencing of complete rDNA loci, known as nucleolus organizer regions (NORs), which contain hundreds or even thousands of rRNA genes. The development of genomics in Arabidopsis thaliana and other plant species coupled with biochemical explorations has helped to include plants at the forefront of research aimed at understanding rRNA gene regulation in eukaryotes. In this chapter, we review the cis- and trans-acting elements that regulate transcription of rDNA by RNA pol I and the responses of rRNA genes to general and specific plant regulatory signals.
7.2
Organization of rRNA genes
In eukaryotes, ribosomal RNA genes are arranged in simple head-to-tail tandem arrays in copy numbers ranging from several hundred in mammals to several thousand in plants (Long and Dawid, 1980). The rRNA genes clustered at a single locus comprise NORs, so named because the nucleolus, the site of ribosome synthesis, is organized around active rRNA genes during interphase (Hernandez-Verdun, 2005; Lam et al., 2005; Shaw and Doonan, 2005). Each rRNA gene transcription unit consists of sequences encoding a precursor transcript that includes the 18S, 5.8S, and 25S structural rRNAs, and is separated from the adjacent gene in the array by an intergenic spacer (IGS) (Pikaard, 2002; Grummt, 2003). In plants, as in animals, coding sequences for the three structural rRNAs are very highly conserved between even distantly related species, but the IGS and sequences that are removed during processing, including the internal
POLYMERASE I TRANSCRIPTION
163
transcribed spacers (ITS) and the external transcribed spacers (ETS), are much less well conserved (Reeder, 1992; Cordesse et al., 1993; Da Rocha and Bertrand, 1995) (Figure 7.1). Analysis of complete IGS sequences reveals considerable length and sequence heterogeneity in different plant species, including radish (Delcasso-Tremousaygue et al., 1988), wheat (Vincentz and Flavell, 1989), A. thaliana (Gruendler et al., 1991), rice (Cordesse et al., 1993), Brassica (Tremousaygue et al., 1992; Da Rocha and Bertrand, 1995), Nicotiana (Borisjuk et al., 1997), and Solanum species (Volkov et al., 2003). However, all ribosomal IGS contain repeated sequences. IGS organization in plant rRNA genes resembles IGS organization in most higher eukaryotes, including at least one array of tandemly repeated sequences located upstream from the transcription initiation site (TIS). In Xenopus (Pikaard and Reeder, 1988) and mouse (Pikaard et al., 1990a), repeated sequences in this location have been shown to possess enhancer activity, increasing the expression of the adjacent promoter after injection into frog oocytes or embryos or after transfection into cultured cells. Though Arabidopsis spacer repeats cloned adjacent to a Xenopus rRNA gene promoter act as enhancers in frog oocytes (Doelling et al., 1993), there is no evidence that they have analogous enhancer activity in plant cells (Wanzenbock et al., 1997). Nonetheless, questions persist as to whether or not IGS heterogeneity affects rRNA transcription, possibly playing a role in the differential activation or silencing of rDNA genes inherited from the different parents in interspecific hybrids (nucleolar dominance) and/or in stimulating rRNA transcription levels among active rRNA genes. Other studies have shown that the IGS, in addition to possessing rDNA transcription regulatory elements, contains origins of replication as well as barriers to replication fork progression (Hernandez et al., 1993). As IGS sequences have diverged rapidly, even between closely related species, a major question is whether the trans-acting factors implicated in these different roles have coevolved with the interacting rDNA sequences.
7.3
Ribosomal RNA transcription and the nucleolus
Biogenesis of ribosomes takes place in the most prominent subnuclear structure: the nucleolus (Scheer and Hock, 1999; Lam et al., 2005). Here, the tandem rRNA genes encoding the 18S, 5.8S, and 25S structural rRNAs are transcribed by RNA pol I as a precursor or pre-rRNA, which is then processed into mature forms that assemble with ribosomal proteins to form ribosome particles (Shaw and Jordan, 1995; Gerbi et al., 2003; Raska et al., 2004). The driving force for the assembly of the nucleolus is transcription of the tandem rRNA gene repeats, as has been shown conclusively in yeast (Oakes et al., 1998; Trumtel et al., 2000) and Drosophila (Karpen et al., 1988) by the assembly of nucleoli at active rRNA genes integrated at ectopic locations. The synthesis of rRNA is completely inhibited during mitosis, when nucleoli disassemble and the rDNA is arranged in a partially condensed form at the NORs. When rRNA synthesis resumes at the end of mitosis, nucleoli reassemble (Hernandez-Verdun et al., 2002; Shaw and Doonan, 2005).
Figure 7.1 Schematic cartoon of rDNA gene repeats transcribed by RNA polymerase I (pol I). The top portion shows tandemly arrayed rDNA genes separated by intergenic spacers (IGSs). ‘T’ indicates potential terminator sites. In the middle part, an enlarged rDNA unit, delimited by successive terminator, is presented, with positions of the gene promoter (closed box) and DNA repeat elements found in all plant species. The primary rRNA (pre-rRNA), transcribed from the transcription initiation site (TIS) and containing18S, 5.8S, and 25S rRNA (gray boxes) sequences separated by the external transcribed spacers (ETS) and internal transcribed spacers (ITS), is shown in gray. In A. thaliana, two spacer promoters (open boxes) are located between the repeat elements containing Sal I restriction sites. In the bottom part, the A. thaliana rRNA gene promoter is enlarged to show the minimal sequences required to initiate in vivo- and in vitro-specific RNA pol I transcription (from −55 to +6). The initiator element TATATAGGGGG (+1 italicized) characterized in A. thaliana is highly similar to sequences surrounding putative transcription sites in different plants species. The B. oleracea RNA pol I holoenzyme binds and protects the rDNA promoter region from −30 to +20 (black rectangle) and initiates in vitro specific transcription from +1.
POLYMERASE I TRANSCRIPTION
165
The basic structure of the nucleolus can be divided into the fibrillar center (FC), dense fibrillar component (DFC), and granular component (GC) (Thiry and Lafontaine, 2005). Even though the relationships between the function and organization of plant and animal nucleoli seem to be well conserved, the nucleolar ultrastructure is very different in these organisms. Whereas in animal nucleoli the FC are surrounded by dense, relatively small regions of DFC (L´eger-Silvestre and Gas, 2004; Mosgoeller, 2004), in plants the DFC is extensive and can occupy most of the nucleolar volume (Shaw and Doonan, 2005). It has been accepted for some time that ribosome biosynthesis occurs in a vectorial pattern within the nucleolus. The transcription of pre-rRNA takes place in the FC, the early processing steps of pre-rRNA occur in the DFC, and the later processing and RNA modification steps, together with the formation of preribosomal particles, take place in the GC. However, the distribution of sites of active rDNA transcription inside the nucleolus remains somewhat controversial. In animal cells, different research groups have localized active ribosomal genes either in the FC and FC/DFC border zone or in the DFC together with the DFC/FC (Raska, 2003; Thiry and Lafontaine, 2005). In plant cells, the site of rRNA transcription is proposed to be in the border zone area between the FC and the DFC (de Carcer and Medina, 1999; Gonzalez-Melendi et al., 2001; Tao et al., 2001a,b). Many years ago, Miller and Beaty (1969) used a chromatin-spreading technique to visualize active rRNA genes by transmission electron microscopy, revealing ‘Christmas tree’ structures (Miller and Beaty, 1969), in which the trunks of the trees are rDNA and the branches are nascent pre-rRNA transcripts. Since then, similar Christmas tree structures have been visualized in many different species, including plants (Shaw et al., 2002). However, it has been very difficult to determine whether or not this structure occurs in the nucleolus in vivo. Shaw et al., using bromouridine labeling, presented evidence that rDNA transcription units are linear, compacted Christmas trees in intact Pisum sativum nucleoli (Gonzalez-Melendi et al., 2001). Importantly, this work clearly revealed the DFC as the subnucleolar domain containing the active transcription units, whereas the FC serves as a reserve of inactive or very weakly expressed genes (Shaw and Doonan, 2005). The work of Shaw and collaborators also revealed that only 200–300 rRNA genes are active in meristematic tissue of root tips of P. sativum. This is only ∼5% of the total number of rRNA genes in this species (Gonzalez-Melendi et al., 2001; Shaw et al., 2002). However, rRNA gene silencing seems not to be fully randomized in the NORs. Studies by Caperta et al. in rye, a diploid species with only one pair of homologous NOR, demonstrate that at metaphase, NORs always show condensed inactive rDNA chromatin in the centromere proximal region whereas the more active rDNA chromatin is located in the NORs region that decondenses progressively toward the satellite (Caperta et al., 2002). There is evidence that differential expression of rDNA genes at the different rye NORs is established after each cell division or reflects patterns that are imposed early in development and are then maintained (Caperta et al., 2002). Consistent with the latter view, numerous studies in mammalian cells have shown that during mitosis the pol I machinery is associated only with NORs that were transcribed in the previous interphase cycle
166
REGULATION OF TRANSCRIPTION IN PLANTS
and that will be transcriptionally active in the next cell cycle (Scheer and Rose, 1984; Roussel et al., 1993; Gebrane-Younes et al., 1997).
7.4
rRNA gene promoters and transcription factors
In animals and yeast, rRNA gene promoters share a common bipartite structure consisting of a core promoter (CP) sequence, which includes the TIS, and an upstream control element (UCE), which enhances rRNA transcription (Grummt, 2003; Moss, 2004; Russell and Zomerdijk, 2005). In mammals, at least four factors in addition to RNA pol I are required to initiate rRNA transcription. The multisubunit transcription factor TIF-IB/SL1, which is composed of TATA-binding protein (TBP) and three TBP-associated factors or TAFs; the HMG-box protein, upstream binding factor (UBF); the TIF-IA/RRN3 protein; and TIF-IC (reviewed in Cavanaugh et al., 2004). In yeast, rDNA transcription requires upstream activation factor (UAF) and core factor (CF), as well as Rrn3p, the yeast homolog of mammalian TIF-1A. The UAF factor is a multisubunit complex of at least six proteins and CF is composed of TBP and three stably associated proteins (Nomura et al., 2004). In plants, most of our knowledge of cis-acting DNA elements that regulate rRNA transcription comes from studies of the A. thaliana IGS, the size and organization of which is similar to that of Xenopus laevis (Pikaard, 2002). Both Xenopus and Arabidopsis rRNA gene IGS sequences include multiple spacer promoters in addition to the gene promoter, as well as reiterated sequence elements that act as enhancers of transcription, at least in Xenopus (Labhart and Reeder, 1985; Pikaard, 1994). Using transient expression in Arabidopsis protoplasts, Doelling et al. demonstrated that sequences extending from −33 upstream to +6 downstream from the transcription start site (+1) are sufficient to program minimal accurate transcription initiation in vivo, though sequences between −55 and +6 constitute a more reliable minimal promoter (Doelling et al., 1993; Doelling and Pikaard, 1995). More recently, analysis of the wheat rRNA promoter demonstrated that promoter sequences localized between −113 and +15 are sufficient to program specific RNA pol I transcription (Akhunov et al., 2001). Remarkably, sequences surrounding the TIS in Arabidopsis (TATATAGGGGG, +1 is italicized) are highly similar between different plant species (Delcasso-Tremousaygue et al., 1988; Cordesse et al., 1993; Fan et al., 1995). The importance of this sequence was demonstrated by experiments showing that mutations in this consensus region abolish or inhibit rRNA transcription and influence the position of the TIS (Doelling and Pikaard, 1995) (Figure 7.1). In addition to the gene promoter, the A. thaliana IGS has at least two spacer promoters and blocks of repeated elements bearing Sal I restriction sites (Sal I repeats) that are located between the gene promoter and each duplicated spacer promoter (Pikaard, 2002). The spacer promoter sequences share 90% similarity with the gene promoter (from −92 to +6), although they are only ∼10% as active as the gene promoter (Doelling et al., 1993). Deletions of the Sal repeat sequences show no qualitative or quantitative differences in transcription initiation activity,
POLYMERASE I TRANSCRIPTION
167
suggesting that upstream repeat sequences of the rDNA intergenic region do not act as transcriptional enhancers for RNA pol I in Arabidopsis (Doelling et al., 1993; Wanzenbock et al., 1997). Nevertheless, when Sal repeats are cloned adjacent to an X. laevis ribosomal RNA gene promoter, they enhance transcription activity at least five- to tenfold, as discussed previously (Doelling et al., 1993). Moreover, when they are cloned upstream from the β-glucuronidase GUS reporter gene, reporter expression increases fourfold in A. thaliana plants (Schlogelhofer et al., 2002). In summary, at present there is no firm evidence for a bipartite structure of rRNA gene promoters in plants, as has been demonstrated in animals and yeast. Instead, plant rRNA gene promoters may be more like the promoters of Acanthamoeba, with a relatively small CP that is sufficient for transcription initiation both in vivo and in vitro (Kownin et al., 1985; Radebaugh et al., 1997). The repeat elements present in the IGS regions of all studied species might have a structural role in establishing chromatin organization and/or nucleolus structure. Furthermore, we can speculate that upstream sequences in the plant IGS may play a regulatory role during development and/or under different environmental conditions. Therefore, a role for additional sequences may become clearer as additional assays are developed. Can rRNA gene promoters from a close or distant plant species be recognized and transcribed by the transcriptional machinery of another plant? In mammalian cells, it is known that pol I transcription is species specific due to the coevolution of the promoter and TIF-1B/SL1 transcription factor (Learned et al., 1985; Schnapp et al., 1991; Heix and Grummt, 1995). Species specificity in plants has been studied in vivo as well as in vitro and has revealed interesting results. In one study, the rRNA gene from tobacco was reported to be actively transcribed in tobacco nuclear extracts, but a broad bean rRNA gene was inactive in the tobacco extract (Fan et al., 1995), in keeping with the expectations derived from the studies in mammals. Interestingly, when a tomato rRNA gene promoter was transfected into A. thaliana protoplasts, it was shown to program transcription initiation, but from an unusual start site located 32 nucleotides downstream from the normal RNA pol I TIS (Doelling and Pikaard, 1996). In the same study, when the gene promoter from Brassica oleracea, a species closely related to A. thaliana, was transfected into A. thaliana protoplasts, transcription initiation was detected both from the normal RNA pol I TIS and from a novel site 29 nucleotides downstream from the expected start site (Doelling and Pikaard, 1996). Doelling and Pikaard reasoned that in a heterologous system, in which the pol I transcription machinery does not efficiently recognize the rRNA gene promoter, the conserved TATATA sequence located at the normal TIS might serve as a TATA box. If so, recognition of the TATA box by the TBP within the transcription factor complex TFIID might program pol II transcription, which typically initiates 25–30 nt downstream of a TATA box (Muller and Tora, 2004). Consistent with this hypothesis, mutations of the Brassica TATA sequence, which eliminated +29 transcription and expression of a linked reporter protein gene, were rescued by cotransfection of TBP genes that contained suppressor mutations. Importantly, the transcription initiating from +1, attributable to pol I transcription, was unaffected by the mutations or TBP suppressor gene (Doelling and Pikaard, 1996). These experiments showed that the TATATA sequence within the conserved initiator region
168
REGULATION OF TRANSCRIPTION IN PLANTS
of plant rRNA gene promoters is capable of serving as a TATA box recognized by TBP, but in the context of pol I transcription, the conserved TATA-containing sequence serves a function other than TBP binding. In plants, little is known concerning specific transcription factors that comprise the rRNA transcriptional machinery. Early biochemical and immunological studies revealed that B. oleracea RNA pol I core enzyme is an approximately 600-kDa multimeric enzyme containing 12–15 subunits (Guilfoyle et al., 1976; Guilfoyle and Jendrisak, 1978; Guilfoyle and Malcolm, 1980). In the following 20 years, a number of efforts to identify potential pol I transcription factors relied on electrophoretic mobility shift assays to detect proteins interacting with the promoter or promoterproximal sequences (Echeverria et al., 1992; Jackson and Flavell, 1992; Griess et al., 1993; Ashapkin et al., 1995; Kneidl et al., 1995), although the identity of these proteins remains unknown. Sequencing of genomes from A. thaliana (TAG Initiative, 2000) and other plant species (IRGS Project, 2005) has allowed the identification of potential transcription factors described in the animal and yeast systems. For instance, the A. thaliana genome codes for at least one protein with multiple HMG boxes, reminiscent of the vertebrate UBF family (Soullier et al., 1999), and two proteins with significant similarity to TIF-1A/RRN3 proteins (Saez-Vaquez and Pikaard, unpublished results). On the other hand, no Arabidopsis genes encoding proteins with detectable homologies to any of the TAFs of TIF-IB/SL1 or yeast CF, nor to the yeast UAF proteins, are readily apparent in the genome. It is possible that functional homologs could have diverged beyond recognition in the rapidly evolving pol I/rRNA gene transcription system. In animals, UBF is an HMG protein involved in pol I transcription activation, as mentioned previously. UBF can bend and wrap rDNA using its HMG boxes (Putnam et al., 1994; Stefanovsky et al., 1996) and is implicated in rRNA gene regulation, rDNA chromatin organization, and NOR morphology (Schnapp and Grummt, 1991; Voit et al., 1992, 1997; Mais et al., 2005). Recently, a protein distantly related to UBF, HMO1, was identified in yeast. This protein is a ∼25-kDa polypeptide that contains a single HMG domain, localizes to the nucleolus, and stimulates rRNA transcription (Gadal et al., 2002). Moreover, Hmo1p bends DNA and may play a role in chromatin structure (Kamau et al., 2004). Thus, factors that are functionally equivalent to UBF may be present in a wide range of eukaryotes. It is possible that a functional homolog of UBF might exist in plants, although this has not yet been demonstrated. The putative UBF-like protein in A. thaliana mentioned previously has a predicted molecular mass of 53 kDa and contains three HMG boxes, whereas the molecular mass of vertebrate UBF proteins range from ∼85 to 97 kDa (Pikaard et al., 1990b; O’Mahony and Rothblum, 1991; Kuhn et al., 1994) and have five or six HMG boxes (Pikaard et al., 1990b; Bachvarov and Moss, 1991; Hisatake et al., 1991; O’Mahony and Rothblum, 1991; Kuhn et al., 1994). Both rice and maize also have predicted proteins that contain three HMG boxes as in Arabidopsis. Moreover, antibodies against mammalian UBF have been shown to cross-react with a single 58-kDa polypeptide on immunoblots and interact with one or more proteins in the nucleolus of Allium cepa (Rodrigo et al., 1992; de Carcer
POLYMERASE I TRANSCRIPTION
169
and Medina, 1999; Tao et al., 2001a,b). Collectively, these data provide tantalizing hints that a UBF-like protein could be present in plants, but this remains highly speculative, and no role for a UBF-like protein in plant rRNA transcription has yet been demonstrated. Another important DNA element that has been shown to regulate RNA pol I promoter activity in other species, including mouse (Kuhn et al., 1988), yeast (van der Sande et al., 1989), and Xenopus (McStay and Reeder, 1986), is the pol I transcription terminator. In mouse and Xenopus, the terminator is an element located just downstream from the 28S rRNA coding sequences at the 5 end of the IGS but is repeated again just upstream from the rRNA gene promoter. In mouse, the terminator element is recognized by the transcriptional terminator factor 1 (TTF1) (Grummt, 2003). In plants, potential transcription termination sites have been identified in Raphanus sativus (Delcasso-Tremousaygue et al., 1988) and Vigna radiata (Schiebel et al., 1989), but the factors have not yet been identified and the role of these sites in rRNA gene regulation remains to be determined.
7.5
RNA pol I holoenzymes: evidence and controversies
The development of cell-free transcription systems has allowed the identification of basal and regulatory transcription factors involved in rRNA transcription in animals as well as in yeast. Whole cell extracts that appear to program specific rRNA transcription initiation have been described for Vicia faba (Yamashita et al., 1993) and Nicotiana tabacum (Fan et al., 1995). These systems, together with the purification of a cell-free transcription system for pol I prepared from broccoli (B. oleracea) (Saez-Vasquez and Pikaard, 1997), should facilitate the characterization of regulatory factors and mechanisms involved in transcription of rRNA genes in plants. In B. oleracea, a highly purified in vitro transcription system led to the first evidence in any eukaryote that the RNA pol I core enzyme and its essential transcription factors can assemble into a multicomponent holoenzyme complex that is capable of binding to the promoter and programming accurate transcription initiation. The key observation is that the B. oleracea RNA pol I holoenzyme can be purified through multiple different chromatography columns, and single fractions at each step remain capable of programming accurate, promoter-dependent transcription initiation (Saez-Vasquez and Pikaard, 1997). The fact that all activities required to direct accurate rDNA transcription copurify through five different chromatography steps suggests that all the necessary transcription factors are stably associated with RNA pol I to form a single multiprotein complex or holoenzyme. Based on gel filtration, the B. oleracea RNA pol I holoenzyme complex displays a molecular mass of approximately 2 MDa and contains approximately 30 distinct polypeptides. In addition to several proteins identified immunologically as subunits of RNA pol I, a casein kinase II (CKII)-like protein kinase (Saez-Vasquez et al., 2001) and a histone acetyltransferase (HAT) activity (Saez-Vasquez and Pikaard, unpublished results) copurify with pol I holoenzyme activity. Likewise, CKII and HAT activities were also found in an RNA pol I holoenzyme purified from X. laevis
170
REGULATION OF TRANSCRIPTION IN PLANTS
(Albert et al., 1999), suggesting that a general feature of RNA pol I holoenzymes is an ability to respond to growth signals and modify chromatin structure to activate transcription. DNase I footprinting revealed that the Brassica RNA pol I holoenzyme binds to and protects the CP region from ≈−30 to +20 (Saez-Vasquez and Pikaard, 2000), in good agreement with the boundaries of the minimal rDNA promoter defined in vivo and in vitro (Doelling et al., 1993; Doelling and Pikaard, 1995) (Figure 7.1). Furthermore, the minimal promoter sequences required to program transcription by the Brassica pol I holenzyme correspond to the same sequences that are required for transcription in vivo following transfection and transient expression of rRNA minigenes (Saez-Vasquez and Pikaard, 1997). Therefore, the purified holoenzyme activity appears to account for the known characteristics of Brassica and Arabidopsis rRNA transcription in vivo. Another observation that supports the existence of a pol I holenzyme is that in an agarose electrophoretic mobility shift assay, highly purified B. oleracea RNA pol I holoenzyme fractions contain a promoter-specific DNA-binding activity that associates and dissociates from the promoter in a single step, with no intermediate complex(es) observed (Saez-Vasquez and Pikaard, 2000). The discovery of RNA pol I holoenzymes, not only in plants but also in different organisms and transcriptional systems (Wang et al., 1997; Myer and Young, 1998), has raised interest in their role in rRNA transcription initiation as well as some debate concerning their significance. Time-lapse microscopy studies of fluorescently tagged proteins involved in pol I transcription in mammalian cells suggested that in contrast to the holoenzyme model, RNA pol I subunits assemble in vivo at the rRNA genes in a sequential manner via metastable intermediates, with increasing stability achieved as more subunits are added (Dundr et al., 2002). A model developed from these observations proposes that at the end of each transcription cycle, RNA pol I dissociates into individual subunits, which then leave the nucleolus. However, these results are inconsistent with a detailed study in yeast, which showed that Rrn3p–Pol I complexes form even in the absence of rRNA gene promoters and that RNA pol I subunits remain stably associated with one another through many transcription cycles (Schneider and Nomura, 2004). The observations in yeast are compatible with the existence of holoenzymes that presumably include the essential factor, Rrn3p/TIF-IA. Although one cannot exclude the possibility that after homogenization of tissues and breakage of cells, RNA pol I holoenzymes complexes might be artificially generated, this is unlikely because both plant and animal holoenzymes have been identified and purified under very stringent biochemical conditions (Saez-Vasquez et al., 2003) and by using different approaches (Seither et al., 1998; Hannan et al., 1999). Also, it is possible that the experimental conditions of Dundr et al. do not allow analysis of RNA pol I holoenzyme complexes because only a small fraction of RNA pol I is tightly associated with transcription factors to form a pool of holoenzyme complexes (Saez-Vasquez and Pikaard, 1997; Albert et al., 1999). Finally, it is always possible that both models are correct, with the holoenzyme potentially being the initiation-competent form of pol I, but representing a small proportion of the total pol I pool, much of which is engaged in transcriptional elongation, rather than initiation, on each rRNA gene. Dissociation of holoenzyme
POLYMERASE I TRANSCRIPTION
171
subcomplexes following initiation may be part of the transcription cycle, and would necessitate reassociation of the subcomplexes with pol I core enzyme complexes that are released from the template upon transcriptional termination.
7.6
Coupling rDNA transcription and processing of pre-rRNA
In the nucleolus, transcription of rDNA is linked to processing of the primary transcript (pre-rRNA) synthesized by RNA pol I. Soon after transcription of the fulllength pre-rRNA, containing 18S, 5.8S, and 25S–28S rRNA, external transcribed spacers (5 ETS and 3 ETS) and internal transcribed spacers (ITS-1 and ITS-2) are removed and involve multiple cleavage events (Figure 7.1). Likewise, modification of numerous rRNA residues takes place, mainly by pseudouridylation and 2 -Oribose methylation. All these events are directed by nucleolar proteins and small nucleolar RNAs (snoRNAs) that form different small nucleolar ribonucleoprotein (snoRNP) complexes that transiently interact with the pre-rRNAs (Venema and Tollervey, 1999; Fromont-Racine et al., 2003). One of the first events occurring cotranscriptionally on the nascent pre-rRNA is the formation of a 5 ‘terminal ball’ that can be visualized by electron microscopy of rRNA genes subjected to the so-called Miller chromatin-spreading technique (Miller and Beaty, 1969). These terminal balls correspond to a large U3 snoRNP that directs the early cleavages of the pre-rRNA (Mougey et al., 1993). Characterization of this complex in yeast identified fibrillarin and 27 other nucleolar proteins that associate with the highly conserved U3 snoRNA (Dragon et al., 2002). Further genetic dissection of the yeast U3 snoRNP identified a subcomplex of proteins (Utp complex) that forms in the absence of U3 and directly interacts with the rDNA 5 ETS region. Most remarkably, it was shown that this interaction is important for subsequent assembly of U3 snoRNP on the pre-rRNA (Gallagher et al., 2004). A similar situation probably occurs in plants. A U3 snoRNP of ∼600 kDa that has been purified and characterized from cauliflower inflorescence binds to prerRNA and accurately produces the first cleavage of the pre-rRNA in vitro (SaezVasquez et al., 2004b). In addition to U3 snoRNA, fibrillarin and a nucleolin-like protein were identified in this complex. These are two major nucleolar proteins implicated in pre-rRNA processing in animals and yeast (Tollervey et al., 1991; Ginisty et al., 1998). Interestingly the plant U3 snoRNP was purified based on its sequence-specific rDNA-binding activity. It was demonstrated that the plant U3 snoRNP complex specifically binds to a double-stranded rDNA sequence just upstream from the primary cleavage site in the pre-RNA 5 ETS (Caparros-Ruiz et al., 1997). Altogether, these data support a model whereby the U3 snoRNP processing complex might assemble first on rDNA and subsequently interact with nascent prerRNA that is produced when RNA pol I traverses the DNA element (Saez-Vasquez et al., 2004a,b). In this model the plant nucleolin-like protein could play an important regulatory role. In vertebrates, nucleolin has been shown to bind to pre-rRNA and is essential for 5 ETS cleavage. In addition, nucleolin also binds to rDNA and can interfere with
172
REGULATION OF TRANSCRIPTION IN PLANTS
RNA pol I-directed transcription: overexpression of nucleolin in Xenopus oocytes represses RNA pol I transcription (Roger et al., 2002). On the other hand, experiments using oocytes injected with nucleolin and actinomycin D revealed that overexpression of nucleolin affects processing of pre-rRNA only when it is present at the same time of transcription (Roger et al., 2003). Thus, in plants, animals, and yeast various lines of evidence indicate a link between RNA pol I transcription and assembly of processing complexes on nascent pre-rRNA, but the mechanisms for this linkage remain unknown. An important question is whether or not processing activities are directly interacting with RNA pol I. Such a link is well documented for RNA pol II and the processing of pre-mRNAs. It has been shown that capping, splicing, and polyadenylation factors are recruited to transcription sites by direct interaction with the carboxy-terminal domain (CTD) that protrudes from the largest RNA pol II (Neugebauer, 2002). The large subunit of RNA pol I lacks a CTD-like domain and a different mechanism is probably involved. In plants and yeast the U3 snoRNP and Utp complexes respectively interact with the rDNA independently of rRNA transcription and do not require RNA pol I. Transcription by RNA pol I is probably required to assemble processing complexes on the nascent pre-rRNA, but RNA pol I does not recruit these complexes to the rDNA. Thus, the next challenge in plants will be to elucidate the mechanism regulating the assembly of the 5 ETS processing complex on the rDNA and its subsequent binding to nascent pre-rRNA.
7.7
Growth and hormonal control of RNA pol I transcription
In all eukaryotes, rRNA synthesis is highly sensitive to environmental conditions, including nutrient availability or stress conditions. Synthesis of rRNA is also stimulated by diverse physiological effectors, including growth signals and phytohormones (Nomura et al., 1984; Moss, 2004). In plants, early work carried out by Guilfoyle and collaborators showed that auxin treatment increases the amount of extractable RNA pol I activity, assayed using heterologous template DNA (promoter-independent, ribonucleotide incorporation using a nicked or sheared DNA template) (Guilfoyle et al., 1975). Increased RNA pol I activity in response to auxin treatment could be correlated with increased amounts of RNA pol I, but no apparent modification of RNA polymerase subunits was detected (Guilfoyle and Malcolm, 1980). Later, Gaudino and Pikaard showed that exogenous cytokinin upregulates pol I transcription from the Arabidopsis rRNA gene promoter, whereas auxin had no discernible effect (Gaudino and Pikaard, 1997). Because auxin treatment can increase RNA pol I abundance and yet does not increase transcription from the rRNA gene promoter, as does cytokinin, it is possible that cytokinin regulates a rate-limiting step, such as post-translational modification of one or more key activities in order to upregulate promoter-dependent transcription (Gaudino and Pikaard, 1997). One prediction is that treatment with both auxin and cytokinin might show a synergistic effect in upregulating rRNA gene promoter activity, but this experiment has not yet been reported. Thirty years ago, Teissere et al. reported four fractions – alpha, beta, gamma, and delta – that stimulate the activity
POLYMERASE I TRANSCRIPTION
173
of RNA pol I in lentil roots (Teissere et al., 1975). At least one of these factors, the gamma factor, is an auxin-induced protein (Teissere et al., 1975). The possibility that the gamma factor could also be induced by cytokinin remains untested. The identification of CKII and HAT activities associated with the B. oleracea pol I holoenzyme suggests that the holoenzyme is equipped to respond to growth signals and modify chromatin as necessary, to activate transcription (Saez-Vasquez et al., 2001). Growth-dependent changes in rDNA transcription in animals have been correlated with the phosphorylation of UBF by CKII (Voit et al., 1995). Furthermore, the HAT p300/CBP-associated factor (PCAF) is implicated in the control of cell growth and differentiation and is known to acetylate TIF-1B/SL1 and activate RNA pol I transcription (Muth et al., 2001). Moreover, acetylation of UBF by CREB-binding protein (CBP) activates rRNA transcription, whereas the retinoblastoma (Rb) protein promotes UBF deacetylation and suppresses rRNA transcription (Pelletier et al., 2000). As mentioned previously, a TIF-1B/SL1-like activity has not been identified in plants; however, Rb protein has been described as participating in cell proliferation in A. thaliana (Ebel et al., 2004; Park et al., 2005). It is possible that Rb may play a role in regulating rRNA transcription during growth or cell proliferation in plants, as in animals, but there is no direct evidence at this time. It has been shown that plant growth and proliferation can be induced in the absence of cytokinins by overexpressing cyclin D3 (Riou-Khamlichi et al., 1999). Because RNA pol I and D cyclins are under the control of Target of Rapamacyn (TOR) in yeast and mammals (Schmelzle and Hall, 2000), a plant TOR pathway could also be involved in rRNA regulation in plants. The A. thaliana genome contains one TOR gene (AtTOR) (Menand et al., 2002) as well as other potential orthologs of the animal and yeast TOR pathway (Menand et al., 2004). Therefore, the TOR pathway in plants might be a nodal point that is able to integrate hormonal and developmental signals with rRNA transcription. Clearly, the study of signaling pathways involved in growth and hormonal control of rRNA transcription in plants merits further investigation.
7.8
Nucleolar dominance
Plants and animals typically have hundreds of rRNA genes, not all of which are active, such that the mechanisms responsible for the selective activation and/or silencing of subsets of the rRNA genes have been sought for many years. Considerable progress has been achieved during the last decade – thanks to the study of nucleolar dominance in plants, a phenomenon that offers a unique model for understanding large-scale transcriptional regulation of rRNA genes in higher eukaryotes. When two species are crossed to form an interspecific hybrid, a NOR-bearing chromosome from one progenitor species may form a nucleolus while its homolog inherited from the other parent does not. This phenomenon, known as nucleolar dominance, was first described in plants in 1934 by Navashin (Navashin, 1934) and was called ‘differential amphiplasty’. It has since been shown to occur in insects, amphibia, and mammals, as well as in plants (Reeder, 1985).
174
REGULATION OF TRANSCRIPTION IN PLANTS
Nucleolar dominance has been demonstrated in numerous plant species, including three allotetraploids of the genus Brassica (Chen and Pikaard, 1997a,b), several cereal hybrids (Houchins et al., 1997; Neves et al., 1997), and more recently in artificial Solanum allopolyploids (Komarova et al., 2004). Nucleolar dominance has also been reported in the natural species Arabidopsis suecica, an allotetraploid derived from A. thaliana and Arabidopsis arenosa (also know as Cardaminopsis arenosa), in which the A. thaliana rRNA genes are repressed and those of A. arenosa are expressed (Chen et al., 1998). Interestingly, nucleolar dominance does not occur in all strains of A. suecica (Pontes et al., 2003). Some species show almost complete silencing of A. thaliana genes, whereas in others significant A. thaliana rRNA gene expression is detected (Pontes et al., 2003). The basis of this natural variation in nucleolar dominance is not clear, but is likely the result of subtle genetic (or epigenetic) variation rather than dramatic genome-restructuring events. Although nucleolar dominance seems to be a manifestation of a dosage control system that limits the number of active rRNA genes (Lawrence et al., 2004), the system responds to growth or developmental conditions. For instance, in the allotetraploid B. napus, the B. oleracea-derived rDNA is normally suppressed and B. rapa-derived rDNA is active in vegetative tissue (Chen and Pikaard, 1997a). However, rDNA transcripts from B. oleracea are found in all organs derived from floral meristems (floral buds, sepals, petals, anthers, and siliques), suggesting that the B. oleracea genes silenced during vegetative growth are derepressed on flowering (Chen and Pikaard, 1997b). In agreement with these results, Komarova et al. recently observed developmental modulation of dominant and underdominant rDNA genes in Solanum allopolyploid species (Komarova et al., 2004). Several hypotheses have been proposed to explain nucleolar dominance in plants, based either on differences in rRNA gene–pol I transcription factor interactions or on chromatin modifications that completely repress one set of rRNA genes (Pikaard, 2000). Studies using transfection of cloned rRNA genes from B. rapa and B. oleracea demonstrated that both genes are equally active in B. oleracea, B. rapa, or B. napus (the allotetraploid hybrid of B. oleracea and B. rapa) cells, ruling out the hypothesis that species-specific transcription factors play a role in nucleolar dominance in B. napus (Frieman et al., 1999). In addition, in vitro transcription assays using a cell-free transcription system from B. oleracea also revealed that dominant and underdominant rRNA genes are equally transcribed (Frieman et al., 1999). The hypothesis that the rRNA genes with longer IGSs (which might contain more putative enhancer repeats) are dominant is inconsistent with experiments in vivo and in vitro showing that dominant or underdominant genes, with either minimal promoters or complete IGSs, are similarly transcribed (Chen et al., 1998; Frieman et al., 1999). Furthermore, dominant species can sometimes have a relatively long IGS and more repeats than do underdominant rRNA genes, or they can have an IGS that is shorter and contains fewer repeats (Neves et al., 1997; Pikaard, 1998; Komarova et al., 2004), allowing the conclusion that there is no clear correlation between IGS size or repeat number and nucleolar dominance. The earliest evidence suggesting that nucleolar dominance is controlled by chromatin modification came from observations of the level of methylation of NORs in
POLYMERASE I TRANSCRIPTION
175
wheat and maize (reviewed by Pikaard, 2002). In these two cases, rRNA genes at active NORs were shown to be slightly less methylated and more DNase accessible than rRNA genes at repressed NORs. More recent experiments showed that treatment of Brassica and Arabidopsis hybrids with an inhibitor of cytosine methyltransferase, 5-aza-2 -deoxycytoine (aza-dC), induced reactivation of underdominant rRNA genes and suppressed NORs (Chen and Pikaard, 1997a,b; Chen et al., 1998). However, DNA methylation is not the only player in activation or repression of rRNA genes. Treatment with trichostatin or sodium butyrate, chemicals that cause histone hyperacetylation due to inhibition of histone deacetylase (HDAC) activity, also induced rRNA gene derepression in Brassica and Arabidopsis hybrids (Chen and Pikaard, 1997a,b). Thus, these results suggested that underdominant rRNA genes are silenced as a consequence of covalent chromatin modifications. In Arabidopsis hybrids, the promoters of silenced rRNA genes are hypermethylated and are associated with histone H3 dimethylated on lysine 9 (H3dimethyl K9), whereas hypomethylated promoters (active genes) are associated with histone H3 trimethylated on lysine 4 (H3trimethyl K4) (Lawrence et al., 2004). Likewise, rRNA gene derepression is associated with loss of H3dimethyl K9 and gain of H3trimethyl K4 (Lawrence et al., 2004). Interestingly, blocking DNA methylation causes changes in histone modification, and blocking histone deacetylation induces cytosine demethylation. Since treatment with both aza-dC and trichostatin does not have a synergistic effect, this suggests that cytosine methylation and histone deacetylation act in the same pathway of rRNA gene repression. In fact, the available data suggest a model for a self-reinforcement pathway for rRNA silencing. In this model, H3K9 methylation and H3K9 acetylation are mutually exclusive, suggesting that H3K9 deacetylation precedes H3K9 methylation. A potential H3K9 deacetylase is HDT1 (AtHD2a), whose knockdown by RNA interference causes the derepression of underdominant rRNA genes (Lawrence et al., 2004). Interestingly, mutation in the RPD3-like HDAC AtHDA6 gene affects the H3K4 methylation status at rDNA loci, and although no effect of HDA6 on rRNA gene expression has yet been reported, it may play some role in rRNA gene silencing in nucleolar dominance (Probst et al., 2004). The identity of the H3K9 methylase remains unclear. What do we know about factors and mechanisms directing rDNA chromatin silencing or activation? Grummt et al. showed that the nucleolar chromatinremodeling complex, NoRC, is implicated in silencing of rRNA reporter genes in mammalian cells (Santoro et al., 2002). In this case, RNA silencing is initiated by recruitment of NoRC to the Pol I promoter by interacting with transcriptional terminator factor 1 (TTF-1) at the terminator site located just upstream from the rRNA gene promoter. Subsequently, NoRC interacts with HDAC and DNA methyltransferases that mediate histone H4 deacetylation, H3K9 methylation, and cytosine methylation (Santoro and Grummt, 2005). On the other hand, recent genetic data have demonstrated silencing of transposable elements and other repeated sequences via small interfering RNA (siRNA). This silencing mechanism involves catalytic subunits for a fourth RNA polymerase in plants (RNA pol IV). The data suggest that pol IV, together with factors RDR2 and DCL3, produces siRNAs that target de novo cytosine methylation of homologous sequences via a pathway that also involves
176
REGULATION OF TRANSCRIPTION IN PLANTS
AGO4 and DRM (Herr et al., 2005; Onodera et al., 2005; Pontier et al., 2005; Vaughn and Martienssen, 2005). This siRNA silencing mechanism is implicated in cytosine methylation of 5S rRNA genes, and so it is tempting to speculate that methylation of silenced RNA pol I genes is also directed by the same pathway, but no direct evidence has been presented for a role of pol IV in rRNA silencing or nucleolar dominance. However, Onodera et al. showed that in pol IV mutants, NORs partially dissociate from each other, suggesting a role of pol IV in establishing the heterochromatin environment for nucleolar dominance (Onodera et al., 2005). Future studies will be necessary to understand if the NoRC complex and/or RNA pol IV is involved in nucleolar dominance in plants.
7.9
Final remarks
The plant rRNA gene transcription machinery seems to have similar modes of regulation as in animals and/or the yeast system, and the pol I transcription machinery can apparently interact to form holoenzymes in plants, Xenopus, mouse, and rat. However, several of the key basal and regulatory transcription factors found in other systems seem to be missing, or have not been identified in plants, suggesting that some transcription factors for plant rRNA genes are likely to be plant specific. This is not unexpected, given the rapid evolution of the pol I–rRNA gene transcription system. Full characterization of purified in vitro transcription (and processing) systems combined with the use of genetic resources in Arabidopsis should help us to identify and characterize the different players involved in rRNA gene transcription regulation in plants. Although there is considerable information on mechanisms involved in nucleolar dominance in plants, major questions remain unanswered. For instance, what makes plants repress one parental set of rRNA genes rather than expressing both sets at reduced levels? How do chromatin-remodeling factors direct and/or maintain heterochomatin formation at underdominant rDNA genes? Recent work suggesting that there is crosstalk between transcription and pre-rRNA processing provides evidence for novel mechanisms involved in the regulation of rRNA transcription, and plants are currently at the forefront of this new area of investigation. Determining the identity and function of factors involved in the different stages of rRNA transcription regulation (transcription activation, coordination between different stages of rRNA synthesis, establishment and/or maintenance of silencing) should help us to better understand rRNA transcription regulation in plants. Building upon the considerable progress that has been achieved in the last decade, there are certainly many more advances awaiting us in the next decade.
Acknowledgment We thank Dr Craig Pikaard at Washington University (St Louis, MO, USA) for critical reading and advices that contributed to improve the manuscript. We also thank the head of the LGDP laboratory,
POLYMERASE I TRANSCRIPTION
177
M. Delseny, for supporting this work and Richard Cooke for comments and corrections of English. We apologize to colleagues whose work we did not mention in this chapter for reason of space.
References E.D. Akhunov, A.V. Chemeris, A.M. Kulikov and V.A. Vakhitov (2001) Functional analysis of diploid wheat rRNA promoter by transient expression. Biochimica et Biophysica Acta 1522, 226–229. A.C. Albert, M. Denton, M. Kermekchiev and C.S. Pikaard (1999) Histone acetyltransferase and protein kinase activities copurify with a putative Xenopus RNA polymerase I holoenzyme selfsufficient for promoter-dependent transcription. Molecular and Cellular Biology 19, 796–806. V.V. Ashapkin, T.T. Antoniv and B.F. Vanyushin (1995) Methylation-dependent binding of wheat nuclear proteins to the promoter region of ribosomal RNA genes. Gene 157, 273–277. D. Bachvarov and T. Moss (1991) The RNA polymerase I transcription factor xUBF contains 5 tandemly repeated HMG homology boxes. Nucleic Acids Research 19, 2331–2335. J. Bailey-Serres (1998) Cytoplasmic ribosomes of higher plants. In: A Look Beyond Transcription: Mechanisms Determing mRNA Stability and Translation in Plants (ed. D.R. Gallie), pp. 125–144. American Society of Plant Physiologists, Rockville, MD. N.V. Borisjuk, Y.M. Davidjuk, S.S. Kostishin, et al. (1997) Structural analysis of rDNA in the genus Nicotiana. Plant Molecular Biology 35, 655–660. D. Caparros-Ruiz, S. Lahmy, S. Piersanti and M. Echeverria (1997) Two ribosomal DNA-binding factors interact with a cluster of motifs on the 5 external transcribed spacer, upstream from the primary pre-rRNA processing site in a higher plant. European Journal of Biochemistry 247, 981–989. A.D. Caperta, N. Neves, L. Morais-Cecilio, R. Malho and W. Viegas (2002) Genome restructuring in rye affects the expression, organization and disposition of homologous rDNA loci. Journal of Cell Science 115, 2839–2846. A. Cavanaugh, I. Hirschler-Laszkiewicz and L. Rothblum (2004) Ribosomal DNA transcription in mammals. In: The Nucleolus (ed. M.O.J. Olson), pp. 88–127. Kluwer Academic/Plenum, New York. Z.J. Chen, L. Comai and C.S. Pikaard (1998) Gene dosage and stochastic effects determine the severity and direction of uniparental ribosomal RNA gene silencing (nucleolar dominance) in Arabidopsis allopolyploids. Proceedings of the National Academy of Sciences of the United States of America 95, 14891–14896. Z.J. Chen and C.S. Pikaard (1997a) Epigenetic silencing of RNA polymerase I transcription: a role for DNA methylation and histone modification in nucleolar dominance. Genes and Development 11, 2124–2136. Z.J. Chen and C.S. Pikaard (1997b) Transcriptional analysis of nucleolar dominance in polyploid plants: biased expression/silencing of progenitor rRNA genes is developmentally regulated in Brassica. Proceedings of the National Academy of Sciences of the United States of America 94, 3442–3447. F. Cordesse, R. Cooke, D. Tremousaygue, F. Grellet and M. Delseny (1993) Fine structure and evolution of the rDNA intergenic spacer in rice and other cereals. Journal of Molecular Evolution 36, 369– 379. P.S. Da Rocha and H. Bertrand (1995) Structure and comparative analysis of the rDNA intergenic spacer of Brassica rapa. Implications for the function and evolution of the Cruciferae spacer. European Journal of Biochemistry 229, 550–557. G. de Carcer and F.J. Medina (1999) Simultaneous localization of transcription and early processing markers allows dissection of functional domains in the plant cell nucleolus. Journal of Structural Biology 128, 139–151. D. Delcasso-Tremousaygue, F. Grellet, F. Panabieres, E.D. Ananiev and M. Delseny (1988) Structural and transcriptional characterization of the external spacer of a ribosomal RNA nuclear gene from a higher plant. European Journal of Biochemistry 172, 767–776.
178
REGULATION OF TRANSCRIPTION IN PLANTS
J.H. Doelling, R.J. Gaudino and C.S. Pikaard (1993) Functional analysis of Arabidopsis thaliana rRNA gene and spacer promoters in vivo and by transient expression. Proceedings of the National Academy of Sciences of the United States of America 90, 7528–7532. J.H. Doelling and C.S. Pikaard (1995) The minimal ribosomal RNA gene promoter of Arabidopsis thaliana includes a critical element at the transcription initiation site. The Plant Journal 8, 683– 692. J.H. Doelling and C.S. Pikaard (1996) Species-specificity of rRNA gene transcription in plants manifested as a switch in RNA polymerase specificity. Nucleic Acids Research 24, 4725–4732. F. Dragon, J.E. Gallagher, P.A. Compagnone-Post, et al. (2002) A large nucleolar U3 ribonucleoprotein required for 18S ribosomal RNA biogenesis. Nature 417, 967–970. M. Dundr, U. Hoffmann-Rohrer, Q. Hu, et al. (2002) A kinetic framework for a mammalian RNA polymerase in vivo. Science 298, 1623–1626. C. Ebel, L. Mariconti and W. Gruissem (2004) Plant retinoblastoma homologues control nuclear proliferation in the female gametophyte. Nature 429, 776–780. M. Echeverria, D. Delcasso-Tremousaygue and M. Delseny (1992) A nuclear protein fraction binding to dA/dT-rich sequences upstream from the radish rDNA promoter [corrected]. The Plant Journal 2, 211–219. H. Fan, K. Yakura, M. Miyanishi, M. Sugita and M. Sugiura (1995) In vitro transcription of plant RNA polymerase I-dependent rRNA genes is species-specific. The Plant Journal 8, 295–298. A. Fatica and D. Tollervey (2002) Making ribosomes. Current Opinion in Cell Biology 14, 313–318. M. Frieman, Z.J. Chen, J. Saez-Vasquez, L.A. Shen and C.S. Pikaard (1999) RNA polymerase I transcription in a Brassica interspecific hybrid and its progenitors: tests of transcription factor involvement in nucleolar dominance. Genetics 152, 451–460. M. Fromont-Racine, B. Senger, C. Saveanu and F. Fasiolo (2003) Ribosome assembly in eukaryotes. Gene 313, 17–42. O. Gadal, S. Labarre, C. Boschiero and P. Thuriaux (2002) Hmo1, an HMG-box protein, belongs to the yeast ribosomal DNA transcription system. The EMBO Journal 21, 5498–5507. J.E. Gallagher, D.A. Dunbar, S. Granneman, et al. (2004) RNA polymerase I transcription and prerRNA processing are linked by specific SSU processome components. Genes and Development 18, 2506–2517. R.J. Gaudino and C.S. Pikaard (1997) Cytokinin induction of RNA polymerase I transcription in Arabidopsis thaliana. The Journal of Biological Chemistry 272, 6799–6804. J. Gebrane-Younes, N. Fomproix and D. Hernandez-Verdun (1997) When rDNA transcription is arrested during mitosis, UBF is still associated with non-condensed rDNA. Journal of Cell Science 110, 2429–2440. S.A. Gerbi, A.V. Borovjagin and T.S. Lange (2003) The nucleolus: a site of ribonucleoprotein maturation. Current Opinion in Cell Biology 15, 318–325. H. Ginisty, F. Amalric and P. Bouvet (1998) Nucleolin functions in the first step of ribosomal RNA processing. The EMBO Journal 17, 1476–1486. P. Gonzalez-Melendi, B. Wells, A.F. Beven and P.J. Shaw (2001) Single ribosomal transcription units are linear, compacted Christmas trees in plant nucleoli. The Plant Journal 27, 223– 233. E.A. Griess, K.D. Grasser and G. Feix (1993) Repeat units from a maize rDNA external spacer region exhibit DNA curvature and interact with high-mobility-group proteins. Planta 191, 524– 531. P. Gruendler, I. Unfried, K. Pascher and D. Schweizer (1991) rDNA intergenic region from Arabidopsis thaliana. Structural analysis, intraspecific variation and functional implications. Journal of Molecular Biology 221, 1209–1222. I. Grummt (2003) Life on a planet of its own: regulation of RNA polymerase I transcription in the nucleolus. Genes and Development 17, 1691–1702. T.J. Guilfoyle and J.J. Jendrisak (1978) Plant DNA-dependent RNA polymerases: subunit structures and enzymatic properties of the class II enzymes from quiescent and proliferating tissues. Biochemistry 17, 1860–1866.
POLYMERASE I TRANSCRIPTION
179
T.J. Guilfoyle, C.Y. Lin, Y.M. Chen and J.L. Key (1976) Purification and characterization of RNA polymerase I from a higher plant. Biochimica et Biophysica Acta 418, 344–357. T.J. Guilfoyle, C.Y. Lin, Y.M. Chen, R.T. Nagao and J.L. Key (1975) Enhancement of soybean RNA polymerase I by auxin. Proceedings of the National Academy of Sciences of the United States of America 72, 69–72. T.J. Guilfoyle and S. Malcolm (1980) The amounts, subunit structures, and template-engaged activities of RNA polymerases in germinating soybean axes. Developmental Biology 78, 113–125. R.D. Hannan, A. Cavanaugh, W.M. Hempel, T. Moss and L. Rothblum (1999) Identification of a mammalian RNA polymerase I holoenzyme containing components of the DNA repair/replication system. Nucleic Acids Research 27, 3720–3727. J. Heix and I. Grummt (1995) Species specificity of transcription by RNA polymerase I. Current Opinion in Genetics and Development 5, 652–656. P. Hernandez, L. Martin-Parras, M.L. Martinez-Robles and J.B. Schvartzman (1993) Conserved features in the mode of replication of eukaryotic ribosomal RNA genes. The EMBO Journal 12, 1475–1485. D. Hernandez-Verdun (2005) Nucleolus in the spotlight. Cell Cycle 4, 106–108. D. Hernandez-Verdun, P. Roussel and J. Gebrane-Younes (2002) Emerging concepts of nucleolar assembly. Journal of Cell Science 115, 2265–2270. A.J. Herr, M.B. Jensen, T. Dalmay and D.C. Baulcombe (2005) RNA polymerase IV directs silencing of endogenous DNA. Science 308, 118–120. K. Hisatake, T. Nishimura, Y. Maeda, K. Hanada, C.Z. Song and M. Muramatsu (1991) Cloning and structural analysis of cDNA and the gene for mouse transcription factor UBF. Nucleic Acids Research 19, 4631–4637. K. Houchins, M. O’Dell, R.B. Flavell and J.P. Gustafson (1997) Cytosine methylation and nucleolar dominance in cereal hybrids. Molecular and General Genetics 255, 294–301. IRGS Project (2005) The map-based sequence of the rice genome. Nature 436, 793–800. S.D. Jackson and R.B. Flavell (1992) Protein-binding to reiterated motifs within the wheat rRNA gene promoter and upstream repeats. Plant Molecular Biology 20, 911–919. E. Kamau, K.T. Bauerle and A. Grove (2004) The Saccharomyces cerevisiae high mobility group box protein HMO1 contains two functional DNA binding domains. The Journal of Biological Chemistry 279, 55234–55240. G.H. Karpen, J.E. Schaefer and C.D. Laird (1988) A Drosophila rRNA gene located in euchromatin is active in transcription and nucleolus formation. Genes and Development 2, 1745– 1763. C. Kneidl, E. Dinkl and F. Grummt (1995) An intrinsically bent region upstream of the transcription start site of the rRNA genes of Arabidopsis thaliana interacts with an HMG-related protein. Plant Molecular Biology 27, 705–713. N.Y. Komarova, T. Grabe, D.J. Huigen, V. Hemleben and R.A. Volkov (2004) Organization, differential expression and methylation of rDNA in artificial Solanum allopolyploids. Plant Molecular Biology 56, 439–463. P. Kownin, C.T. Iida, S. Brown-Shimer and M.R. Paule (1985) The ribosomal RNA promoter of Acanthamoeba castellanii determined by transcription in a cell-free system. Nucleic Acids Research 13, 6237–6248. A. Kuhn, A. Normann, I. Bartsch and I. Grummt (1988) The mouse ribosomal gene terminator consists of three functionally separable sequence elements. The EMBO Journal 7, 1497–1502. A. Kuhn, R. Voit, V. Stefanovsky, R. Evers, M. Bianchi and I. Grummt (1994) Functional differences between the two splice variants of the nucleolar transcription factor UBF: the second HMG box determines specificity of DNA binding and transcriptional activity. The EMBO Journal 13, 416–424. P. Labhart and R.H. Reeder (1985) Xenopus ribosomal gene enhancers function when inserted inside the gene they enhance. Nucleic Acids Research 13, 8999–9009. Y.W. Lam, L. Trinkle-Mulcahy and A.I. Lamond (2005) The nucleolus. Journal of Cell Science 118, 1335–1337.
180
REGULATION OF TRANSCRIPTION IN PLANTS
R.J. Lawrence, K. Earley, O. Pontes, et al. (2004) A concerted DNA methylation/histone methylation switch regulates rRNA gene dosage control and nucleolar dominance. Molecular Cell 13, 599– 609. R.M. Learned, S. Cordes and R. Tjian (1985) Purification and characterization of a transcription factor that confers promoter specificity to human RNA polymerase I. Molecular and Cellular Biology 5, 1358–1369. I. L´eger-Silvestre and N. Gas (2004) The nucleolar ultrastructure in yeast. In: The Nucleolus (ed. M.O.J. Olson). Kluwer Academic/Plenum, New York. E.O. Long and I.B. Dawid (1980) Repeated genes in eukaryotes. Annual Review of Biochemistry 49, 727–764. C. Mais, J.E. Wright, J.L. Prieto, S.L. Raggett and B. McStay (2005) UBF-binding site arrays form pseudo-NORs and sequester the RNA polymerase I transcription machinery. Genes and Development 19, 50–64. B. McStay and R.H. Reeder (1986) A termination site for Xenopus RNA polymerase I also acts as an element of an adjacent promoter. Cell 47, 913–920. B. Menand, T. Desnos, L. Nussaume, et al. (2002) Expression and disruption of the Arabidopsis TOR (target of rapamycin) gene. Proceedings of the National Academy of Sciences of the United States of America 99, 6422–6427. B. Menand, C. Meyer and C. Robaglia (2004) Plant growth and the TOR pathway. Current Topics in Microbiology and Immunology 279, 97–113. O.L. Miller and B.R. Beaty (1969) Visualization of nucleolar genes. Science 164, 955–957. W. Mosgoeller (2004) Nucleolar ultrastructure in vertebrates. In: The Nucleolus (ed. M.O.J. Olson). Kluwer Academic/Plenum, New York. T. Moss (2004) At the crossroads of growth control; making ribosomal RNA. Current Opinion in Genetics and Development 14, 210–217. E.B. Mougey, M. O’Reilly, Y. Osheim, O.L. Miller, Jr., A. Beyer and B. Sollner-Webb (1993) The terminal balls characteristic of eukaryotic rRNA transcription units in chromatin spreads are rRNA processing complexes. Genes and Development 7, 1609–1619. F. Muller and L. Tora (2004) The multicoloured world of promoter recognition complexes. The EMBO Journal 23, 2–8. V. Muth, S. Nadaud, I. Grummt and R. Voit (2001) Acetylation of TAF(I)68, a subunit of TIF-IB/SL1, activates RNA polymerase I transcription. The EMBO Journal 20, 1353–1362. V.E. Myer and R.A. Young (1998) RNA polymerase II holoenzymes and subcomplexes. The Journal of Biological Chemistry 273, 27757–27760. M. Navashin (1934) Chromosomal alterations caused by hybridization and their bearing upon certain general genetic problems. Cytologia 5, 169–206. K.M. Neugebauer (2002) On the importance of being co-transcriptional. Journal of Cell Science 115, 3865–3871. N. Neves, M. Silva, J.S. Heslop-Harrison and W. Viegas (1997) Nucleolar dominance in triticales: control by unlinked genes. Chromosome Research 5, 125–131. M. Nomura, R. Gourse and G. Baughman (1984) Regulation of the synthesis of ribosomes and ribosomal components. Annual Review of Biochemistry 53, 75–117. M. Nomura, Y. Nogi and M. Oakes (2004) Transcription of rDNA in the yeast Saccharomyces cerevisiae. In: The Nucleolus (ed. M.O.J. Olson). Kluwer Academic/Plenun, New York. M. Oakes, J.P. Aris, J.S. Brockenbrough, H. Wai, L. Vu and M. Nomura (1998) Mutational analysis of the structure and localization of the nucleolus in the yeast Saccharomyces cerevisiae. Journal of Cell Biology 143, 23–34. D.J. O’Mahony and L.I. Rothblum (1991) Identification of two forms of the RNA polymerase I transcription factor UBF. Proceedings of the National Academy of Sciences of the United States of America 88, 3180–3184. Y. Onodera, J.R. Haag, T. Ream, P.C. Nunes, O. Pontes and C.S. Pikaard (2005) Plant nuclear RNA polymerase IV mediates siRNA and DNA methylation-dependent heterochromatin formation. Cell 120, 613–622.
POLYMERASE I TRANSCRIPTION
181
J.A. Park, J.W. Ahn, Y.K. Kim, et al. (2005) Retinoblastoma protein regulates cell proliferation, differentiation, and endoreduplication in plants. The Plant Journal 42, 153–163. G. Pelletier, V.Y. Stefanovsky, M. Faubladier, et al. (2000) Competitive recruitment of CBP and RbHDAC regulates UBF acetylation and ribosomal transcription. Molecular Cell 6, 1059–1066. C.S. Pikaard (1994) Ribosomal gene promoter domains can function as artificial enhancers of RNA polymerase I transcription, supporting a promoter origin for natural enhancers in Xenopus. Proceedings of the National Academy of Sciences of the United States of America 91, 464–468. C.S. Pikaard (1998) Nucleolar dominance. In: Transcription of Ribosomal RNA Genes by Eukaryotic RNA Polymerase I (ed. M. Paule). Landes Bioscience, Austin, TX. C.S. Pikaard (2000) The epigenetics of nucleolar dominance. Trends in Genetics 16, 495–500. C.S. Pikaard (2002) Transcription and tyranny in the nucleous: the organization, activation, dominance and repression of ribosomal RNA genes. In: The Arabidopsis Book. American Society of Plant Biologists, Rockville, MD. C.S. Pikaard, L.K. Pape, S.L. Henderson, et al. (1990a) Enhancers for RNA polymerase I in mouse ribosomal DNA. Molecular and Cellular Biology 10, 4816–4825. C.S. Pikaard and R.H. Reeder (1988) Sequence elements essential for function of the Xenopus laevis ribosomal DNA enhancers. Molecular and Cellular Biology 8, 4282–4288. C.S. Pikaard, S.D. Smith, R.H. Reeder and L. Rothblum (1990b) rUBF, an RNA polymerase I transcription factor from rats, produces DNase I footprints identical to those produced by xUBF, its homolog from frogs. Molecular and Cellular Biology 10, 3810–3812. O. Pontes, R.J. Lawrence, N. Neves, et al. (2003) Natural variation in nucleolar dominance reveals the relationship between nucleolus organizer chromatin topology and rRNA gene transcription in Arabidopsis. Proceedings of the National Academy of Sciences of the United States of America 100, 11418–11423. D. Pontier, G. Yahubyan, D. Vega, et al. (2005) Reinforcement of silencing at transposons and highly repeated sequences requires the concerted action of two distinct RNA polymerase IV in Arabidopsis. Genes and Development 19(17): 2030–2040. A.V. Probst, M. Fagard, F. Proux, et al. (2004) Arabidopsis histone deacetylase HDA6 is required for maintenance of transcriptional gene silencing and determines nuclear organization of rDNA repeats. The Plant Cell 16, 1021–1034. C.D. Putnam, G.P. Copenhaver, M.L. Denton and C.S. Pikaard (1994) The RNA polymerase I transactivator upstream binding factor requires its dimerization domain and high-mobility-group (HMG) box 1 to bend, wrap, and positively supercoil enhancer DNA. Molecular and Cellular Biology 14, 6476–6488. C.A. Radebaugh, X. Gong, B. Bartholomew and M.R. Paule (1997) Identification of previously unrecognized common elements in eukaryotic promoters. A ribosomal RNA gene initiator element for RNA polymerase I. The Journal of Biological Chemistry 272, 3141–3144. I. Raska (2003) Oldies but goldies: searching for Christmas trees within the nucleolar architecture. Trends in Cell Biology 13, 517–525. I. Raska, K. Koberna, J. Malinsky, H. Fidlerova and M. Masata (2004) The nucleolus and transcription of ribosomal genes. Biologie Cellulaire 96, 579–594. R.H. Reeder (1985) Mechanisms of nucleolar dominance in animals and plants. Journal of Cell Biology 101, 2013–2016. R.H. Reeder (1992) Regulation of transcription by RNA polymerase I. In: Transcriptional Regulation. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. C. Riou-Khamlichi, R. Huntley, A. Jacqmard and J.A. Murray (1999) Cytokinin activation of Arabidopsis cell division through a D-type cyclin. Science 283, 1541–1544. R.M. Rodrigo, M.C. Rendon, J. Torreblanca, G. Garcia-Herdugo and F.J. Moreno (1992) Characterization and immunolocalization of RNA polymerase I transcription factor UBF with anti-NOR serum in protozoa, higher plant and vertebrate cells. Journal of Cell Science 103, 1053–1063. B. Roger, A. Moisand, F. Amalric and P. Bouvet (2002) Repression of RNA polymerase I transcription by nucleolin is independent of the RNA sequence that is transcribed. The Journal of Biological Chemistry 277, 10209–10219.
182
REGULATION OF TRANSCRIPTION IN PLANTS
B. Roger, A. Moisand, F. Amalric and P. Bouvet (2003) Nucleolin provides a link between RNA polymerase I transcription and pre-ribosome assembly. Chromosoma 111, 399–407. P. Roussel, C. Andre, C. Masson, G. Geraud and D. Hernandez-Verdun (1993) Localization of the RNA polymerase I transcription factor hUBF during the cell cycle. Journal of Cell Science 104, 327–337. J. Russell and J.C. Zomerdijk (2005) RNA-polymerase-I-directed rDNA transcription, life and works. Trends in Biochemical Sciences 30, 87–96. J. Saez-Vasquez, A.C. Albert, K. Earley and C.S. Pikaard (2003) Purification and transcriptional analysis of RNA polymerase I holoenzymes from broccoli (Brassica oleracea) and frog (Xenopus laevis). Methods in Enzymology 370, 121–138. J. Saez-Vasquez, D. Caparros-Ruiz, F. Barneche and M. Echeverria (2004a) Characterization of a crucifer plant pre-rRNA processing complex. Biochemical Society Transactions 32, 578–580. J. Saez-Vasquez, D. Caparros-Ruiz, F. Barneche and M. Echeverria (2004b) A plant snoRNPcomplex containing snoRRNA, fibrillarin and nucleolin-like proteins is competent for both rDNA binding and pre-rRNA processing in vitro. Molecular and Cellular Biology 24(16): 7284–7297. J. Saez-Vasquez, M. Meissner and C.S. Pikaard (2001) RNA polymerase I holoenzyme–promoter complexes include an associated CK2-like protein kinase. Plant Molecular Biology 47, 449–459. J. Saez-Vasquez and C.S. Pikaard (1997) Extensive purification of a putative RNA polymerase I holoenzyme from plants that accurately initiates rRNA gene transcription in vitro. Proceedings of the National Academy of Sciences of the United States of America 94, 11869–11874. J. Saez-Vasquez and C.S. Pikaard (2000) RNA polymerase I holoenzyme–promoter interactions. The Journal of Biological Chemistry 275, 37173–37180. R. Santoro and I. Grummt (2005) Epigenetic mechanism of rRNA gene silencing: temporal order of NoRC-mediated histone modification, chromatin remodeling, and DNA methylation. Molecular and Cellular Biology 25, 2539–2546. R. Santoro, J. Li and I. Grummt (2002) The nucleolar remodeling complex NoRC mediates heterochromatin formation and silencing of ribosomal gene transcription. Nature Genetics 32, 393–396. U. Scheer and R. Hock (1999) Structure and function of the nucleolus. Current Opinion in Cell Biology 11, 385–390. U. Scheer and K.M. Rose (1984) Localization of RNA polymerase I in interphase cells and mitotic chromosomes by light and electron microscopic immunocytochemistry. Proceedings of the National Academy of Sciences of the United States of America 81, 1431–1435. K. Schiebel, G. von Waldburg, J. Gerstner and V. Hemleben (1989) Termination of transcription of ribosomal RNA genes of mung bean occurs within a 175 bp repetitive element of the spacer region. Molecular and General Genetics 218, 302–307. P. Schlogelhofer, V. Nizhynska, N. Feik, et al. (2002) The upstream Sal repeat-containing segment of Arabidopsis thaliana ribosomal DNA intergenic region (IGR) enhances the activity of adjacent protein-coding genes. Plant Molecular Biology 49, 655–667. T. Schmelzle and M.N. Hall (2000) TOR, a central controller of cell growth. Cell 103, 253–262. A. Schnapp and I. Grummt (1991) Transcription complex formation at the mouse rDNA promoter involves the stepwise association of four transcription factors and RNA polymerase I. The Journal of Biological Chemistry 266, 24588–24595. A. Schnapp, H. Rosenbauer and I. Grummt (1991) Trans-acting factors involved in species-specificity and control of mouse ribosomal gene transcription. Molecular and Cellular Biochemistry 104, 137–147. D.A. Schneider and M. Nomura (2004) RNA polymerase I remains intact without subunit exchange through multiple rounds of transcription in Saccharomyces cerevisiae. Proceedings of the National Academy of Sciences of the United States of America 101, 15112–15117. P. Seither, S. Iben and I. Grummt (1998) Mammalian RNA polymerase I exists as a holoenzyme with associated basal transcription factors. Journal of Molecular Biology 275, 43–53. P. Shaw and J. Doonan (2005) The nucleolus. Playing by different rules? Cell Cycle 4, 102–105. P.J. Shaw, R. Abranches, A. Paula Santos, et al. (2002) The architecture of interphase chromosomes and nucleolar transcription sites in plants. Journal of Structural Biology 140, 31–38.
POLYMERASE I TRANSCRIPTION
183
P.J. Shaw and E.G. Jordan (1995) The nucleolus. Annual Review of Cell and Developmental Biology 11, 93–121. S. Soullier, P. Jay, F. Poulat, J.M. Vanacker, P. Berta and V. Laudet (1999) Diversification pattern of the HMG and SOX family members during evolution. Journal of Molecular Evolution 48, 517–527. V.Y. Stefanovsky, D.P. Bazett-Jones, G. Pelletier and T. Moss (1996) The DNA supercoiling architecture induced by the transcription factor xUBF requires three of its five HMG-boxes. Nucleic Acids Research 24, 3208–3215. TAG Initiative (2000) Analysis of the genome sequence of the flowering plant Arabidopsis thaliana. Nature 408, 796–815. W. Tao, W. Xu, M.M. Valdivia, S. Hao and Z.H. Zhai (2001a) Distribution and transcription activity of nucleolar DNA in higher plant cells. Cell Biology International 25, 1167–1171. W. Tao, C.H. Yan, T. Cai, S. Hao and Z.H. Zhai (2001b) Structural components of the nuclear body in nuclei of Allium cepa cells. Cell Research 11, 68–73. M. Teissere, P. Penon, R.B. van Huystee, Y. Azou and J. Ricard (1975) Hormonal control of transcription in higher plants. Biochimica et Biophysica Acta 402, 391–402. M. Thiry and D.L. Lafontaine (2005) Birth of a nucleolus: the evolution of nucleolar compartments. Trends in Cell Biology 15, 194–199. D. Tollervey, H. Lehtonen, M. Carmo-Fonseca and E.C. Hurt (1991) The small nucleolar RNP protein NOP1 (fibrillarin) is required for pre-rRNA processing in yeast. The EMBO Journal 10, 573–583. D. Tremousaygue, M. Laudie, F. Grellet and M. Delseny (1992) The Brassica oleracea rDNA spacer revisited. Plant Molecular Biology 18, 1013–1018. S. Trumtel, I. Leger-Silvestre, P.E. Gleizes, F. Teulieres and N. Gas (2000) Assembly and functional organization of the nucleolus: ultrastructural analysis of Saccharomyces cerevisiae mutants. Molecular Biology of the Cell 11, 2175–2189. C.A. van der Sande, T. Kulkens, A.B. Kramer, et al. (1989) Termination of transcription by yeast RNA polymerase I. Nucleic Acids Research 17, 9127–9146. M.W. Vaughn and R.A. Martienssen (2005) Finding the right template: RNA Pol IV, a plant-specific RNA polymerase. Molecular Cell 17, 754–756. J. Venema and D. Tollervey (1999) Ribosome synthesis in Saccharomyces cerevisiae. Annual Review of Genetics 33, 261–311. M. Vincentz and R.B. Flavell (1989) Mapping of ribosomal RNA transcripts in wheat. The Plant Cell 1, 579–589. R. Voit, A. Kuhn, E.E. Sander and I. Grummt (1995) Activation of mammalian ribosomal gene transcription requires phosphorylation of the nucleolar transcription factor UBF. Nucleic Acids Research 23, 2593–2599. R. Voit, K. Schafer and I. Grummt (1997) Mechanism of repression of RNA polymerase I transcription by the retinoblastoma protein. Molecular and Cellular Biology 17, 4230–4237. R. Voit, A. Schnapp, A. Kuhn, et al. (1992) The nucleolar transcription factor mUBF is phosphorylated by casein kinase II in the C-terminal hyperacidic tail which is essential for transactivation. The EMBO Journal 11, 2211–2218. R.A. Volkov, N.Y. Komarova, I.I. Panchuk and V. Hemleben (2003) Molecular evolution of rDNA external transcribed spacer and phylogeny of sect. Petota (genus Solanum). Molecular Phylogenetics and Evolution 29, 187–202. Z. Wang, T. Luo and R.G. Roeder (1997) Identification of an autonomously initiating RNA polymerase III holoenzyme containing a novel factor that is selectively inactivated during protein synthesis inhibition. Genes and Development 11, 2371–2382. E.M. Wanzenbock, C. Schofer, D. Schweizer and A. Bachmair (1997) Ribosomal transcription units integrated via T-DNA transformation associate with the nucleolus and do not require upstream repeat sequences for activity in Arabidopsis thaliana. The Plant Journal 11, 1007–1016. J.R. Warner (1999) The economics of ribosome biosynthesis in yeast. Trends in Biochemical Sciences 24, 437–440. J. Yamashita, T. Nakajima, S. Tanifuji and A. Kato (1993) Accurate transcription initiation of Vicia faba rDNA in a whole cell extract from embryonic axes. The Plant Journal 3, 187–190.
8 Transcription of plastid genes Karsten Liere and Thomas B¨orner
8.1
Introduction
Virtually all plant cells contain plastids. Exceptions are generative and/or sperm cells of those plant species that exhibit uniparental maternal inheritance of the plastid genome (plastome; Hagemann, 2004). Plastids accomplish important metabolic functions. In the various tissues of higher plants exist different types of plastids with specialized roles, the most intensively studied of them being the photosynthetically active chloroplast in green tissues. Plastids divide in a similar manner as bacteria (Aldridge et al., 2005). Each plastid in a plant contains identical plastome copies. Restriction mapping of chloroplast DNA provided evidence for circular chromosomes, first shown for maize (Bedbrook and Bogorad, 1976) and spinach (Herrmann et al., 1976) and later found in most investigated plants including algae (for a recent review see Maier and Schmitz-Linneweber, 2004). In addition to monomeric circles, there exist not only dimers, trimers, and tetramers, but also numerous linear and even more complex molecules of different sizes (Bendich, 2004). Similar to bacterial chromosomes, plastid DNA is organized in nucleoids. Developing and mature chloroplasts contain several nucleoids, and each nucleoid contains several copies of the genome. The copy number of plastid chromosomes and transcript levels vary drastically between different tissues. Highest levels are found in chloroplasts of photosynthetically active cells (Aguettaz et al., 1987; Baumgartner et al., 1989; Isono et al., 1997a). Several DNA-binding proteins are known and proposed to organize and maintain nucleoid structure (‘plastid chromatin’), to bind DNA to membranes, and to be involved in replication, repair, and inheritance of the plastome (Kuroiwa, 1991; Sato et al., 2003; Heinhorst et al., 2004). Only few data exist about effects of conformation of plastid DNA on plastid gene expression (Stirdivant et al., 1985; Gauly and K¨ossel, 1989; Sekine et al., 2002). Effects of DNA methylation on plastidial gene activity have been described, but seem to be an exception rather than the rule (Ngernprasirtsiri et al., 1989; Kobayashi et al., 1990; Ngernprasirtsiri and Akazawa, 1990; Hess et al., 1993; Isono et al., 1997a). Plastomes are remnants of the chromosome of their cyanobacterial ancestor. There is ample evidence for plastids being descendants of a cyanobacterium taken up as endosymbiont by the ancestral cell of the green lineage of eukaryotes (Martin et al., 2002; Dyall et al., 2004; Gray, 2004). During further coevolution of plastid (endosymbiont) and host cell, a massive loss of genes from the plastome has occurred. Many of those genes have been transferred to the nucleus. Gene transfer from the organelle to the nucleus is a relatively frequent, ongoing
TRANSCRIPTION OF PLASTID GENES
185
process (Herrmann et al., 2003; Timmis et al., 2004; Leister, 2005). Plastid chromosomes of embryophytes have a size of about 150 kb (ranging from 117 kb in Pinus koraiesis to 164 kb in Oenothera elata subsp. hookeri) and a gene content between approximately 100 and 120 (Hupfer et al., 2000; Gray, 2004; Maier and Schmitz-Linneweber, 2004) (for a compilation of all sequenced chloroplast genomes, see http://megasun.bch.umontreal.ca/ogmp/projects/other/cp list.html). Plastid genes code for the 16S, 23S, 5S, and 4.5S RNAs, for tRNAs sufficient to translate all codons, for proteins needed in first line for gene expression and photosynthesis, and for only a few proteins involved in other metabolic processes. As in bacteria, genes are organized in operons, but promoters also serve transcription of single genes (Sugiura, 1995; Maier and Schmitz-Linneweber, 2004). Several plastid genes contain introns belonging to the group I and group II families. Transcript maturation includes splicing, C to U (rarely U to C) editing, and excision of monocistronic mRNAs from polycistronic precursors. Trimming and polyadenylation at 3 ends are steps leading to degradation of transcripts (for reviews see Sugita and Sugiura, 1996; Bock, 2000; Barkan, 2004; Marchfelder and Binder, 2004). Gene expression in plastids is highly regulated at the level of post-transcriptional processes that are not subject of this chapter (for recent reviews see Barnes and Mayfield, 2003; Barkan, 2004; Bollenbach et al., 2004; Herrin and Nickelsen, 2004; Marchfelder and Binder, 2004; Zerges, 2004). In the following discussion, we review our current understanding of the transcription in plastids of higher plants, with a focus on RNA polymerases, transcription factors, promoters, and the regulation of transcription by endogenous and environmental factors.
8.2
RNA polymerases
Cyanobacteria, the ancestors of chloroplasts, as other bacteria, use one RNA polymerase to transcribe all genes. The core enzyme consists of several subunits encoded by the rpoA, B, C1, and C2 genes (Bergsland and Haselkorn, 1991). For promoter recognition and initiation of transcription, the core polymerase is complemented by a σ -factor to constitute the holoenzyme. Cyanobacteria possess several σ -factors to transcribe different sets of genes (Imamura et al., 2003). Although chloroplast genomes are much smaller than bacterial ones, the transcription machinery in plastids is more complex as compared to that of bacteria. At least in angiosperms, transcription of plastid genes needs more than one type of RNA polymerase and different σ -factors. Originally, transcription of plastid genes was thought to depend exclusively on a nuclear-encoded RNA polymerase. RNA synthesis was detected in rye and barley plastids lacking all plastid-encoded proteins, indicating that one or more nuclear genes code for transcriptional activity in plastids (B¨unger and Feierabend, 1980; Siemenroth et al., 1981). The nuclear-encoded RNA polymerase was shown to synthesize plastidial rRNAs (Siemenroth et al., 1981) and, more recently, to transcribe many other genes in ribosome-free plastids (Falk et al., 1993; Han et al., 1993; Hess et al., 1993; H¨ubschmann and B¨orner, 1998; Zubko and Day, 1998).
186
REGULATION OF TRANSCRIPTION IN PLANTS
With the knowledge of first complete sequences of plastid chromosomes it became evident that plastid genomes contain rpoA, B, C1, and C2 genes for all subunits of a core RNA polymerase of the (cyano)bacterial type (Ohyama et al., 1986; Shinozaki et al., 1986). Later, these genes were shown to indeed code for subunits of an active bacterial-type RNA polymerase (Hu and Bogorad, 1990; Hu et al., 1991). Inactivation of rpo genes in transplastomic tobacco plants led to an albino phenotype, demonstrating the essential role this RNA polymerase is playing for chloroplast development and function. However, the plastid genes were still transcribed in rpo plants, providing further evidence of the importance of the nuclear-encoded plastidial RNA polymerase (Allison et al., 1996; Hajdukiewicz et al., 1997; Krause et al., 2000; Legen et al., 2002). Thus, plastids of most angiosperms need at least two types of transcriptases, nuclear-encoded plastid RNA polymerases (NEP) and plastid-encoded plastid RNA polymerases (PEP) (Hajdukiewicz et al., 1997). PEP is the bacterial-type RNA polymerase encoded by the plastid rpoA, B, C1, and C2 genes (Figure 8.1). NEP was originally defined as a transcriptase that is active in plastids even in the absence of PEP and plastid translation. Later, plastid-targeted phagetype RNA polymerases were recognized as potential candidates for NEP (Figure 8.1; Hedtke et al., 1997; reviewed by Gray and Lang, 1998; Hess and B¨orner, 1999). The transcriptional machinery might be even more complex, as Bligny et al. (2000) observed transcription of rRNA in spinach chloroplasts by an RNA polymerase activity that showed the expected features neither of a bacterial-type PEP nor of a phage-type NEP. Simpler, however, might be the transcriptional apparatus in plastids of certain parasitic, nonphotosynthetic angiosperms with reduced plastid genomes lacking rpo genes or containing only rpo pseudogenes. Consequently, they should lack PEP activity and rely solely on NEP to transcribe their plastid genes (Ems et
Figure 8.1 The nuclear-encoded plastid RNA polymerase (NEP) is related to phage-type singlesubunit enzymes. The catalytic subunit RpoT may need additional, yet unknown protein factors for promoter recognition. The plastid-encoded plastid RNA polymerase (PEP) is a multisubunit enzyme homologous to bacterial RNA polymerases. PEP needs one of several nuclear-encoded σ -factors to bind to the promoter. The composition shown here is typical for PEP in etioplasts. In chloroplasts, PEP is associated with more proteins. (TIS, transcription initiation site).
TRANSCRIPTION OF PLASTID GENES
187
al., 1995; Lusson et al., 1998; Krause et al., 2003; Berg et al., 2004). In contrast, algal chloroplasts may need only the PEP enzyme. In the case of Chlamydomonas reinhardtii no indications for a NEP were found (A. Weihe et al., unpublished data). NEP and PEP recognize different sets of promoters. NEP promoters resemble in their architecture the promoters in the genomes of mitochondria and T-bacteriophages, whereas PEP binds to sequence motives known from the −10 and −35 boxes of bacterial promoters.
8.2.1
NEP: Nuclear-encoded plastid RNA polymerase
There is accumulating evidence for NEP activity being represented by phage-type RNA polymerase(s). The genomes of several bacteriophages contain a gene for an RNA polymerase. The best characterized bacteriophage RNA polymerase is that of T7 phages (Steitz, 2004). In contrast to bacterial, archaeal, and eukaryotic RNA polymerases, the phage enzymes are single-subunit RNA polymerases; i.e., promoter recognition, initiation of transcription, and RNA synthesis (elongation) are performed by just one polypeptide (Steitz, 2004). The mitochondrial genome of baker’s yeast, Saccharomyces cerevisiae, was the first nonphage genome demonstrated to be transcribed by a phage-type RNA polymerase (Masters et al., 1987; Schinkel et al., 1988). Mitochondrial genes of almost all eukaryotes are transcribed by related phage-type RNA polymerases (Cermakian et al., 1996). The only known exception to date is Reclinomonas americana, a freshwater protozoon belonging to the jakobids. Reclinomonas uses a mitochondrial-gene-encoded multisubunit RNA polymerase of the bacterial type (Lang et al., 1997). Moreover, plants possess nuclear genes for phage-type RNA polymerases (Figure 8.1; referred to as RpoT genes and RpoT polymerases, for RNA polymerases of the T3/T7-type; Hedtke et al., 1997). Some of these genes code for RNA polymerases that are targeted to mitochondria, as shown for dicots (Hedtke et al., 1997, 1999, 2000, 2002; Weihe et al., 1997; Kobayashi et al., 2001a), monocots (Chang et al., 1999; Ikeda and Gray, 1999; Emanuel et al., 2004; Kusumi et al., 2004), and the moss Physcomitrella (Kabeya et al., 2002; Richter et al., 2002). These polymerases are regarded as transcriptases of plant mitochondria, although direct experimental evidence for this proposed function is lacking. We have recently observed that the Arabidopsis mitochondrial RpoT polymerase (RpoTm) is able to recognize mitochondrial promoters and initiate transcription in vitro, strongly suggesting that this enzyme is indeed active in mitochondrial transcription (K. K¨uhn et al., unpublished data). The first indication for a phage-type RNA polymerase in chloroplasts was reported by Lerbs-Mache (1993), who observed an RNA polymerase activity recognizing a T7 promoter in a 110-kDa protein fraction isolated from spinach chloroplasts. The first candidate RpoT gene for a plastidial RNA polymerase of the phage-type (RpoTp) was discovered in the nuclear genome of Arabidopsis. It is proposed to originate from duplication of the RpoTm gene (Hedtke et al., 1997). Homologous nuclear RpoTp genes have been described for Nicotiana tabacum, Nicotiana sylvestris, Hordeum vulgare, Zea mays, Triticum aestivum, and Oryza sativa (Chang et al.,
188
REGULATION OF TRANSCRIPTION IN PLANTS
1999; Ikeda and Gray, 1999; Kobayashi et al., 2001a; Hedtke et al., 2002; Emanuel et al., 2004; Kusumi et al., 2004) and seem to be present in all angiosperms. Plastid localization of RpoTp polymerases is supported by chloroplast targeting of green fluorescent protein (GFP) fused to the putative transit sequences. RpoTp polymerases have also been detected in chloroplasts of rice and maize by immunological methods (Chang et al., 1999; Kusumi et al., 2004). That RpoTp is indeed involved in transcription of plastid genes has recently been demonstrated using transgenic tobacco plants overexpressing this polymerase (Liere et al., 2004). Embryophytes are the only eukaryotes reported to possess more than one RpoT gene. Cereals have two RpoT genes coding for a mitochondrial (RpoTm) and a plastidial RNA polymerase (RpoTp; Chang et al., 1999; Ikeda and Gray, 1999; Emanuel et al., 2004; Kusumi et al., 2004). Arabidopsis and N. sylvestris have three (Hedtke et al., 2000, 2002; Kobayashi et al., 2001a,b, 2002) and the amphidiploid genome of N. tabacum harbors even six RpoT genes (two sets of three genes from each parental species; Hedtke et al., 2002). The RNA polymerase of the third gene in Arabidopsis and Nicotiana (and possibly other eudicots) is proposed to be targeted to both mitochondria and plastids (referred to as RpoTmp polymerase) as deduced from the localization of GFP fused to the putative N-terminal transit peptide (Hedtke et al., 2000, 2002; Kobayashi et al., 2001a). The localization of RpoTmp to Arabidopsis plastids has more recently been questioned by a study on transgenic plants expressing GFP fused to the putative transit peptide together with the 5 -UTR of RpoTmp. GFP fluorescence was detected only in mitochondria of these plants (Kabeya and Sato, 2005). Baba et al. (2004) analyzed mitochondrial and chloroplast transcripts in an Arabidopsis line with a defective RpoTmp gene bearing a T-DNA insertion. The mutation affected the transcription of several plastidial but not of mitochondrial genes. Evidently, more investigations are needed to clarify localization and role of RpoTmp polymerases. All three RpoT genes are active in roots, leaves, and flowers of Arabidopsis, though RpoTm and RpoTmp seem particularly active in meristems and young cells of all organs, whereas RpoTp is most expressed in green tissues (Emanuel et al., 2006).
8.2.2
PEP: Plastid-encoded plastid RNA polymerase
Plastid genomes contain the rpoA, B, C1, and C2 genes coding for the α (38 kDa), β (120 kDa), β (85 kDa), and β (185 kDa) subunits, respectively, of a bacterialtype RNA polymerase (Sugiura, 1995). Like in cyanobacteria, rpoB, C1, and C2 on one hand and rpoA together with genes for ribosomal proteins on the other hand form operons (Hudson et al., 1988; Purton and Gray, 1989; Kaneko et al., 1996). Two α subunits and one each of the β, β, and β subunits build up the core polymerase that performs the elongation step of RNA synthesis, but is able to bind to the promoter and to initiate transcription only when completed with one of several nuclear-encoded σ -factors (Figure 8.1). Similar to the core polymerase, σ -factors and promoters used by PEP are closely related to the corresponding components in bacteria (see below). PEP has been detected in the soluble fraction of chloroplast preparations and as an insoluble, DNA-bound activity, the so-called transcriptionally
TRANSCRIPTION OF PLASTID GENES
189
active chromosome (TAC; Little and Hallick, 1988; Suck et al., 1996; Pfalz et al., 2006). The soluble form of PEP isolated from etioplasts consists mainly of the core subunits, while additional proteins assemble with the core subunits to a larger complex in chloroplasts (Pfannschmidt and Link, 1994, 1997; Link, 1996; Baginsky et al., 1999; Pfannschmidt et al., 2000), probably reflecting the need of regulating PEP in response to changing light conditions and photosynthetic activity. Yet, it is unclear if NEP forms a complex with PEP or if the two RNA polymerases operate entirely independently of each other. Since PEP and NEP recognize different promoters (see below), the latter might be the case.
8.3 8.3.1
Plastidial promoters PEP promoters
The plastid-encoded plastid RNA polymerase (PEP) has evolved from a eubacterialtype RNA polymerase. Accordingly, plastidial promoters contain a variant of the −35 (TTGaca) and −10 (TAtaaT) consensus sequences of typical σ 70 -type Escherichia coli promoters (Reznikov et al., 1985; for reviews see Gruissem and Tonkyn, 1993; Link, 1994; Hess and B¨orner, 1999; Weihe and B¨orner, 1999; Liere and Maliga, 2001; Weihe, 2004). In fact, plastidial σ 70 -type promoters are accurately recognized by the E. coli RNA polymerase (e.g., Gatenby et al., 1981; Bradley and Gatenby, 1985; Boyer and Mullet, 1988; Eisermann et al., 1990). Besides, some plastidial σ 70 -type promoters contain additional regulatory cis-elements. Early data derived from in vitro transcription experiments established that the psbA gene, which encodes the D1 photosystem II reaction center polypeptide, is transcribed from a σ 70 -type promoter with the typical conserved −10 and −35 elements (Link, 1984; Gruissem and Zurawski, 1985; Boyer and Mullet, 1986; Boyer and Mullet, 1988). In vivo psbA transcription is developmentally timed and activated by light (Klein and Mullet, 1990; Schrubar et al., 1990; Baumgartner et al., 1993). A detailed in vitro characterization of the mustard psbA promoter identified a TATATA promoter element between the −10 and −35 hexamers resembling the TATA box of nuclear genes transcribed by RNA polymerase II (Eisermann et al., 1990; Link, 1994). The TATATA element and −10 region were sufficient to obtain basic transcription levels in vitro in plastidial extracts prepared from both dark- and light-grown plants. Nonetheless, presence of the −35 element was essential for enhanced transcription rates characteristic of chloroplasts of light-grown plants (Link, 1984; Eisermann et al., 1990). The barley psbA promoter also contains the TATA motif between the −35 and −10-elements, but, unlike in mustard, the −35 sequence is absolutely required for transcription in vitro (Kim et al., 1999b). Such TATA box is also present in the wheat psbA promoter, but does not seem to be important. Constitutive (light-independent) transcription by PEP isolated from the leaf base (base-type PEP; young plastids) required both the −10 and the −35 elements for promoter activity. However, transcription by PEP isolated from the leaf tip (tip-type PEP; mature plastids) was dependent only on the −10 region, which additionally
190
REGULATION OF TRANSCRIPTION IN PLANTS
Figure 8.2 Different types of PEP and NEP promoters. PEP promoter: schematic overview of the wheat psbA (TaepsbA), barley psbD BLRP (HvupsbD), and tobacco rrn16 PEP promoters (Ntarrn16). Conserved −10/−35 consensus elements as well as individual promoter elements as the TATA box, extended −10 sequence (TGn), AAG box, and RUA element are shown. The less conserved −35 element in the barley psbD BRLP is shown in gray. NEP promoter: schematic overview of Type I and Type II NEP promoters, with examples from tobacco given in brackets. The YRTa promoter core and GAA box are marked. (TIS, transcription initiation site, indicated by arrows).
contains a TGn motif upstream of the −10 element (Figure 8.2; extended −10 sequence; Bown et al., 1997; Satoh et al., 1999). It was proposed that the basaland tip-type PEPs differ by their associated transcription factors, and hence, that the extended −10 sequence may be involved in promoter recognition by the tip-type PEP in mature plastids (Satoh et al., 1999). Since the mustard, barley, and wheat psbA promoter sequences are highly conserved, differences in the utilization of ciselements possibly are the result of a divergent evolution of trans-factors in these species. In addition to the core promoter regions, regulatory sequence motifs were identified in the proximity of the psbD-psbC, rbcL, and psaA-psaB promoters (see Section 8.4.3 for details).
8.3.2
NEP promoters
Transcription initiation sites for a nuclear-encoded transcription activity (i.e., NEP) were unambiguously identified in plants in which the transcriptional activity by PEP had been reduced or eliminated. Such systems comprise the ribosome-deficient plastids of the monocot albostrians barley and iojap maize mutants (H¨ubschmann and B¨orner, 1998; Silhavy and Maliga, 1998a), tobacco rpo plants lacking PEP in dicots (Allison et al., 1996; Hajdukiewicz et al., 1997; Serino and Maliga, 1998), and photosynthetically inactive tobacco and rice suspension cultures, with elevated levels of NEP activity (Vera et al., 1996; Kapoor et al., 1997; Miyagi et al., 1998; Silhavy and Maliga, 1998b). While photosynthetic genes and operons have PEP promoters, most nonphotosynthetic genes involved in housekeeping functions such as transcription and translation have promoters for both RNA polymerases. NEP transcripts of these genes
TRANSCRIPTION OF PLASTID GENES
191
are, with a few exceptions, rarely detectable in chloroplasts and were therefore mostly analyzed in PEP-deficient plants (see above). Only a few genes are known to be transcribed exclusively from a NEP promoter, i.e., accD, encoding a subunit of the acetyl-CoA carboxylase in dicots; clpP, encoding the proteolytic subunit of the Clp ATP-dependent protease; ycf2, encoding a protein with a yet unknown function; rpl23, encoding a ribosomal protein in monocots; and, most interestingly, the rpoB operon encoding three of the four PEP core subunits in all higher plants (Hajdukiewicz et al., 1997; H¨ubschmann and B¨orner, 1998; Silhavy and Maliga, 1998a). Thus, PEP abundance and activity depends on the nuclear-encoded RNA polymerase. NEP promoters do not have conserved eubacterial σ 70 -type promoter elements, but resemble mitochondrial and phage promoters, and based on their sequence properties they can be grouped into three types (Figure 8.2; Weihe and B¨orner, 1999; Liere and Maliga, 2001). Type I promoters analyzed thus far are characterized by a conserved YRTa motif critical for rpoB promoter recognition embedded in a small DNA fragment (−15 to +5) upstream of the transcription initiation site (+1) (PatpB289 – Kapoor and Sugiura (1999), Xie and Allison (2002); PaccD-129 – Liere and Maliga (1999b); PrpoB-345 – Liere and Maliga (1999a)). Although, in vitro, no additional sequence elements outside the promoter core modified rpoB transcription (Liere and Maliga, 1999a), transient expression of chimeric Arabidopsis rpoB 5 -flanking region::GUS deletion-constructs in cultured tobacco cells suggested the existence of upstream regulatory regions for rpoB expression (Inada et al., 1997). Similar transient transcription assays to examine the 5 -flanking region of the tobacco accD gene revealed that possibly further sequence elements up- and downstream of the promoter determine its strength (Hirata et al., 2004). A subset of Type I NEP promoters possesses a second, conserved sequence motif (ATAN0−1 GAA) ∼18–20 bp upstream of the YRTa motif, designated box II or GAA box (Figure 8.2; Silhavy and Maliga, 1998a; Kapoor and Sugiura, 1999). Mutational analyses of the GAA box in in vitro and in vivo transcription experiments suggest, at least for the tobacco PatpB-289 promoter, a functional role of this element in promoter recognition (Kapoor and Sugiura, 1999; Xie and Allison, 2002). Hence, Type I promoters are classified into two subclasses: Ia, with only the YRTa motif; and Ib, carrying both YRTa- and GAA box (Weihe and B¨orner, 1999; Liere and Maliga, 2001). Type II NEP promoters lack the YRTa motif and differ completely in sequence and organization from Type I promoters. So far, this class is represented by a single example, a promoter of the ClpP protease subunit gene (Figure 8.2). The tobacco PclpP-53 was characterized using a transplastomic in vivo approach demonstrating that critical promoter sequences are located mainly downstream of the transcription initiation site (−5 to +25; Sriraman et al., 1998a). The clpP-53 promoter motif and transcription initiation site are conserved among monocots, dicots, conifers, and liverworts, pointing toward an early appearance of the NEP transcription machinery in evolution. But, although the tobacco PclpP-53 sequence motif also is present in rice and Chlamydomonas, it is not used as a promoter. Although, if the rice sequence is introduced into tobacco plastids, the tobacco NEP recognizes this conserved Type II promoter. One may speculate that the lack of transcription in rice from
192
REGULATION OF TRANSCRIPTION IN PLANTS
the PclpP-53 homolog is due to the lack of either a Type II specificity factor or a distinct NEP enzyme not present in monocots (e.g., RpoTmp, see Section 8.2.1; Sriraman et al., 1998a; Liere et al., 2004). However, experimental data supporting these scenarios are still missing.
8.3.3
tRNAs with internal promoters
Most plastidial tRNAs are transcribed by the PEP from upstream σ 70 -type promoters. However, transcription of some tRNA genes such as the spinach trnS, trnR, and trnT (Gruissem et al., 1986; Cheng et al., 1997b) as well as trnS, trnH, and trnR from mustard (Neuhaus and Link, 1990; Nickelsen and Link, 1990; Liere and Link, 1994), and the Chlamydomonas trnE gene (Jahn, 1992), is assumed to be mediated by internal promoters. The transcription start site of the spinach trnS gene was mapped 12 nucleotides upstream of the mature tRNA coding region (Wu et al., 1997). While the coding region (+1/+93) promoted basal levels (8%) of transcription in in vitro assays, inclusion of an AT-rich region between −31 and −11 upstream of the coding region restored wild-type promoter strength. No sequences resembling either NEP or PEP promoters were found in this region. However, the trnS coding region contains sequences resembling the A and B blocks of nuclear tRNA promoters transcribed by the eukaryotic RNA polymerase III (Galli et al., 1981; Geiduschek et al., 1995). Upon injection into Xenopus oocyte nuclei, the tRNAArg (ACG) gene from Pelargonium zonale was efficiently transcribed (Hellmund et al., 1984), suggesting that the plastidial tRNAs may be transcribed by an RNA polymerase III-type enzyme in plastids. Thus far, in silico analyses of the Arabidopsis genome did not reveal a plastid-targeted polymerase of this type (K. Liere and T. B¨orner, unpublished). Therefore, the biochemical properties and enzyme composition of this transcription activity remains to be determined. Alternatively, specialized NEP or PEP enzymes associated with distinct transcription factors may recognize internal promoter structures and transcribe such tRNAs.
8.4
Regulation
Transcriptional control is thought to be a key element to manage expression of nuclear-encoded plastid-localized gene products (Kuhlemeier, 1992). Although post-transcriptional events may contribute more significantly to regulation of plastidial gene expression (Deng and Gruissem, 1987; Stern et al., 1997; Barkan and Goldschmidt-Clermont, 2000; Monde et al., 2000), transcription of plastid genes was also shown to respond to factors such as light and plastid-type (Rapp et al., 1992; Mullet, 1993; Mayfield et al., 1995; Link, 1996). At an early stage of light-induced plastid development, transcription activity of the majority of plastid-encoded genes increases to support rapid construction of the photosynthesis apparatus. Furthermore, it has been shown that light-dependent plastid transcription occurs in mature leaves as well as leaves under greening (Greenberg et al., 1989; Schrubar et al., 1990; Baumgartner et al., 1993; DuBell and Mullet, 1995; Hoffer and Christopher, 1997;
TRANSCRIPTION OF PLASTID GENES
193
Shiina et al., 1998; Satoh et al., 1999; Baena-Gonzalez et al., 2001; Chun et al., 2001; Nakamura et al., 2003). Most prominent examples are photosynthesis-related genes as psbA, psbD-psbC, petG, and rbcL, but also housekeeping genes as atpB (Klein et al., 1988; Haley and Bogorad, 1990; Klein and Mullet, 1990; Sexton et al., 1990; Isono et al., 1997a). Transcriptional activation of photosynthesis-related genes by distinctive photoreceptors has been analyzed (Chun et al., 2001; Thum et al., 2001). Light-quality perception may depend on the developmental stage. In dark-adapted mature leaves, red light only partially increased overall plastid transcription activity whereas blue light further enhanced it. Therefore, global activation of plastidial transcription after dark adaptation is probably mediated by cryptochromes. A further role is suggested for phyA, since an Arabidopsis phyA mutant exposed to blue light/UV-A displayed, in comparison to the wild type, reduced psbA and rrn16 transcript activities (Chun et al., 2001). As illustrated while describing plastid promoters, transcriptional response to developmental and environmental changes is likely to involve interaction of the core RNA polymerase with specific regulatory molecules (e.g., σ -factors), which might be available only under certain conditions. Additionally, various pathways routing developmental and environmental cues may regulate these factors.
8.4.1
Role of diverse and multiple promoters in developmental regulation of gene expression
The region upstream of the rrn16 operon is greatly conserved between different plant species. In vivo and in vitro dissection of the σ 70 -type PEP promoter (P1 or Nt-Prrn114; Vera and Sugiura, 1995; Allison et al., 1996) in tobacco plastids identified an essential hexameric sequence upstream of the −35 element (Figure 8.2; GTGGGA, rRNA operon upstream activator – RUA) that is conserved in monocot and dicot species. Interaction between the PEP and an RUA-binding transcription factor is proposed to replace the σ -factor interaction with the −10 element (Suzuki et al., 2003). Also in barley, maize, and pea the rrn operon is transcribed from the P1 σ 70 -type PEP promoter (Figure 8.3; Strittmatter et al., 1985; Sun et al., 1989; H¨ubschmann and B¨orner, 1998). Additionally, the rrn operon in tobacco has a second, NEP promoter (Figure 8.3; P2 or Nta-Prrn-62), inactive in chloroplasts, but functional in BY2 tissue culture cells and in plants lacking PEP (Vera and Sugiura, 1995; Allison et al., 1996). Conversely, in maize plastids there is no active NEP promoter directly upstream of the rrn operon (Silhavy and Maliga, 1998a). In spinach chloroplasts, transcription of the rrn operon initiates in a region, which contains typical −10/−35-elements active as rrn operon promoters in other species. However, the σ 70 -type promoter sequences are not utilized in vivo. Instead, transcripts initiate from a site between the conserved −10/−35 hexamers (Figure 8.3; Pc promoter; Baeza et al., 1991; Iratni et al., 1994, 1997). Sequences relevant for transcription initiation from Pc have yet to be identified. Interestingly, the Pc site appears to be faithfully recognized by partially purified mustard PEP in vitro (Pfannschmidt and Link, 1997). A good candidate for the Pc-activating factor in
194
REGULATION OF TRANSCRIPTION IN PLANTS
Figure 8.3 Differential rRNA operon promoter usage in spinach, Arabidopsis, tobacco, and barley. The schematic representation of the transcription initiation sites between trnV and rrn16, indicated by arrows, shows the distinctive promoters used in different species. P1 (open circle) and P2 (filled circle) mark transcription initiation sites by PEP and NEP respectively. Pc marks transcript initiation in spinach and Arabidopsis from a yet uncharacterized promoter. A dashed vertical line indicates an RNAprocessing site. Also denoted are marker genes used in transplastomic plants introducing the tobacco rrn promoter sequence in Arabidopsis (aadA: spectinomycin resistance) and the spinach sequence in tobacco (uidA: β-glucuronidase).
spinach is CDF2, the binding of which is correlated with transcription from Pc (Iratni et al., 1994, 1997). CDF2 is supposed to exist in two distinct forms, CDF2-A and CDF2-B. While CDF2-A represses transcription initiation of PEP at the rrn16 P1 promoter (named P2 in spinach), CDF2-B possibly binds NEP-2, a yet-to-becharacterized nuclear transcription activity, and initiates transcription from the rrn16 Pc promoter (Bligny et al., 2000). In Arabidopsis, rrn operon transcripts were mapped to both the major tobacco PEP P1 and the spinach Pc initiation sites (Figure 8.3). A study of rrn promoters in heterologous plastids indicates that tobacco plastids lack the factor required for transcription from Pc, while spinach has an intact P1 promoter but lacks the cognate P1 activator (Figure 8.3; Sriraman et al., 1998b). However, in tobacco a RUA region that is conserved in monocot and dicot species has been identified in vitro (Figure 8.2; Suzuki et al., 2003). It was suggested that the −10 element plays only a limited role in rrn16 P1 recognition and that σ -factor interaction is replaced in part by direct PEP–RUA (protein–DNA) interaction or by protein–protein interaction between the PEP and a putative RUA-binding transcription factor. The rrn16 promoters are an example of the diversity of promoter usage even in closely related species. Similarly, the tobacco atpB gene is transcribed from at least five NEP (PatpB-255, -502/-488, -611) and PEP promoters (PatpB-289, -329; Figure 8.4; Hajdukiewicz et al., 1997), but only one PEP promoter is driving this gene in Arabidopsis (PatpB-520/-515; D. Kaden, K. K¨uhn, and K. Liere, unpublished),
TRANSCRIPTION OF PLASTID GENES
195
Figure 8.4 Genes with multiple promoters. Schematic synopsis that shows the multiple PEP promoters of the barley psbD/C operon (HvupsbD), PEP and NEP promoters of the tobacco atpB (NtaatpB), and clpP genes (NtaclpP). Boxes below the line represent genes on the opposite strand, open arrowheads denote PEP promoters, filled black arrowheads Type I NEP promoters, and filled gray arrowheads Type II NEP promoters. The promoters are named based on their position with respect to the translation initiation site (+1).
whereas in maize one NEP (PatpB-601) and one PEP promoter (PatpB-298) are responsible for atpB expression (Silhavy and Maliga, 1998a). One NEP promoter precedes the tobacco accD gene (PaccD-129; Hajdukiewicz et al., 1997); however, two NEP promoters were found to transcribe the Arabidopsis accD gene (PaccD172, -251; D. Kaden, K. K¨uhn, and K. Liere, unpublished). In case of clpP the tobacco gene has two Type I NEP (PclpP-173, -511), one PEP (PclpP-95), and one Type II NEP initiation sites (PclpP-53; Figure 8.4; Hajdukiewicz et al., 1997; Sriraman et al., 1998a), the Arabidopsis gene has one PEP (PclpP-117) and one Type II NEP initiation site (PclpP-57; M. Swiatecka and K. Liere, unpublished; Sriraman et al., 1998a), whereas the maize gene is transcribed by a sole Type I NEP promoter (PclpP-111; Silhavy and Maliga, 1998a). Nonetheless, transcription of plastidial genes and operons by multiple promoters seems to be a rather common feature.
196
REGULATION OF TRANSCRIPTION IN PLANTS
For example, the psbD-psbC operon is transcribed from up to three different PEP promoters (Figure 8.4; Yao et al., 1989; Berends Sexton et al., 1990; Christopher et al., 1992; Wada et al., 1994; To et al., 1996; Hoffer and Christopher, 1997) and the tobacco rpl32 gene from two promoters far upstream of the coding region (NEP-Prpl32-1101, PEP-Prpl32-1030; Vera et al., 1996). While some genes are cotranscribed within an operon, they additionally possess an individual promoter upstream of their coding region (e.g., trnG; Meng et al., 1991; Kapoor et al., 1994; Liere and Link, 1994), or, vice versa, although transcribed by a strong individual promoter, they are cotranscribed with upstream genes (e.g., psbA; Nickelsen and Link, 1991; Liere et al., 1995). Although the role of most multiple promoters upstream of plastidial genes and operons is not fully understood, some are well characterized. While the blue-lightresponsive promoter (BRLP) of psbD-psbC is thought to differentially maintain the ability to resynthesize and replace damaged D2 and CP43 photosystem components in mature chloroplasts, mixed NEP and PEP promoters generally are believed to differentially express their cognate gene during plant development (reviewed in Liere and Maliga, 2001). NEP promoters are generally recognized in youngest and nongreen tissues early in plant development, while PEP takes over in maturating photosynthetically active chloroplasts (Bisanz-Seyer et al., 1989; Baumgartner et al., 1993; Hajdukiewicz et al., 1997; Kapoor et al., 1997; Emanuel et al., 2004). Consequently, both promoter types typically are found upstream of housekeeping genes which need to be transcribed during full plastidial development (Maliga, 1998). Recently, this simple model has been challenged by results from transcriptional reanalyses of tobacco rpo mutants lacking PEP (Krause et al., 2000; Legen et al., 2002). Large spurious transcripts initiated by NEP cover the entire plastome in these mutants, suggesting that besides selective promoter utilization, post-transcriptional processes also determine the transcript pattern of plastids. Further data derived from plastids in the developmental gradient in maize leaves suggest that although, as plastids mature, NEP may become less abundant, transcriptional activity by NEP increases while RNA stability declines. Hence, in a proposed model for maize plastidial biogenesis, NEP-controlled transcript accumulation changes little during plastidial development, while PEP-controlled transcript accumulation increases (Cahoon et al., 2004).
8.4.2 8.4.2.1
Transcription factors for selective promoter recognition Nuclear-encoded plastidial σ -factors
In eubacteria, specific transcription initiation requires a transcription factor (σ ), which is responsible for promoter recognition and contributes to DNA melting around the initiation site. Most eubacterial genomes contain genes for several σ -factors recognizing distinct promoters. Bacterial σ -factors have conserved functional regions and are grouped into two families, σ 70 and σ 54 (W¨osten, 1998; Ishihama, 2000). The σ 70 -factors are furthermore classified into primary (group 1, essential for cell growth), nonessential primary (group 2), and alternative σ -factors (group 3), responsible for recognition of certain promoters in response to
TRANSCRIPTION OF PLASTID GENES
197
environmental signals (Lonetto et al., 1992; Gruber and Bryant, 1997). Cyanobacteria, the ancestors of plastids, also have multiple σ -factors with distinct promoter specificity (Kaneko et al., 1996). Early on, biochemically purified σ -like activities in plant plastids were reported in Chlamydomonas (Surzycki and Shellenbarger, 1976), spinach (Lerbs et al., 1983), and mustard (B¨ulow and Link, 1988; Tiller and Link, 1993b). Furthermore, immunological evidence for σ -like factors was obtained in chloroplast RNA polymerase preparations of maize, rice, C. reinhardtii, and Cyanidium caldarium (Troxler et al., 1994). Moreover, multiple nuclear-encoded genes encoding eubacterial σ 70 -type factors were identified in the red algae Cyanidium caldarium (CcaA-C; Liu and Troxler, 1996; Tanaka et al., 1996) and Cyandioschyzon merolae (CmeSig1-4; Matsuzaki et al., 2004), suggesting specialized promoter recognition as in bacteria. Correspondingly, σ -factor families were identified in genomes of land plants such as Arabidopsis (AthSig1-6; Isono et al., 1997b; Tanaka et al., 1997; Fujiwara et al., 2000; Hakimi et al., 2000), mustard (SalSig1-3; Kestermann et al., 1998; Homann and Link, 2003), tobacco (NtaSigA1, -A2; Oikawa et al., 2000), rice (OsaSig1-4; Tozawa et al., 1998; Kasai et al., 2004), maize (ZmaSig1-5; Lahiri et al., 1999; Tan and Troxler, 1999; Lahiri and Allison, 2000), Physcomitrella patens (PpaSig1, -2, -5; Hara et al., 2001a,b; Ichikawa et al., 2004), as well as wheat (TaeSigA; Morikawa et al., 1999) and Sorghum (SbiSig1; Kroll et al., 1999). The N-terminal sequences of these σ -factors are typical for plastid-targeting transit peptides, which indeed have been demonstrated to confer plastidial targeting of GFP fusion proteins in vivo (Isono et al., 1997b; Tanaka et al., 1997; Kanamaru et al., 1999; Fujiwara et al., 2000; Lahiri and Allison, 2000; Oikawa et al., 2000; Hara et al., 2001a) or with radiolabeled proteins in vitro (Kestermann et al., 1998). Surprisingly, organellar targeting of some plant σ -factors was observed to exist not only into plastids but also into mitochondria. The AthSig5 transcripts are alternatively spliced within intron 1 establishing two initiation methionines (M1 and M2). Shorter peptides starting with M2 exclusively targeted GFP into plastids. However, GFP fusion proteins starting with M1 were localized into mitochondria. RNA analyses revealed that the longer (plastid) AthSig5 transcripts are exclusively located in flowers, whereas the shorter (mitochondrial) transcripts were detectable in both flower and leaf tissue (Yao et al., 2003). Furthermore, AthSig1::GFP fusion proteins also are colocalized into both plastids and mitochondria in tobacco protoplast import assays (H. Tandara and K. Liere, unpublished data). Similarly, dual targeting was shown for the maize ZmaSig2B protein by immunological and GFP fusion protein import studies. Interestingly, ZmaSig2B was biochemically copurified with RpoTm, the mitochondrial phage-type RNA polymerase (Beardslee et al., 2002), suggesting a possible role of these mitochondrial localized σ -factors in regulation of plant mitochondrial transcription. Historically, plastidial σ -factors were designated alphabetically or by numbers. Thus, in Arabidopsis, SigA, SigB, and SigC (Tanaka et al., 1997) were also named SIG2, SIG1, and SIG3 (Isono et al., 1997b), respectively. In an effort to unify the nomenclature, σ -factors sequences were subjected to phylogenetic analyses (http://sfns.u-shizuoka-ken.ac.jp/pctech; Shiina et al., 2005). Higher plant σ -factors
198
REGULATION OF TRANSCRIPTION IN PLANTS
belong to a monophyletic group. Although they are related to bacterial primary (group 1) and nonessential primary (group 2) σ 70 -factors, none fit into alternative group 3 or are related to σ 54 -factors. Furthermore, the phylogenetic analyses revealed that plastidial σ -factors are split into at least five subgroups: Sig1, Sig2, Sig3, Sig5, and Sig6. Interestingly, only in Sig1 and Sig2 groups the monocot and dicot σ -factors are located on separate branches. Since most sequenced higher plant and moss genomes contain at least one gene for a Sig1-type σ -factor, and the Arabidopsis Sig1 homologs are highly expressed during chloroplast biogenesis, it is assumed that Sig1 represents the principal σ -factor in chloroplasts (Tanaka et al., 1997; Kestermann et al., 1998; Tozawa et al., 1998; Kanamaru et al., 1999; Morikawa et al., 1999). Sig2, Sig3, and Sig5 genes have been identified in various plant organisms, suggesting a correspondingly important role in plastidial transcription. Conversely, to date Sig4 has been identified only in Arabidopsis, and in comparison to the other σ -factors in light-grown plants its transcription is rather low (Tsunoyama et al., 2002). In addition, phylogenetic analysis suggests that the Sig2, Sig3, AthSig4, and Sig6 groups are related to Sig1 (Shiina et al., 2005), supported by the observation that intron sites of AthSig1, AthSig2, AthSig3, AthSig4, and AthSig6 are identical (Fujiwara et al., 2000). This suggests that these σ -factors may have originated from gene duplication event of one or more ancestral genes. The Sig5 group, however, only partially seems to be related to the bacterial alternative σ -factors (Tsunoyama et al., 2002; Shiina et al., 2005). Eubacterial σ 70 -factors contain four conserved domains (1–4) involved in binding the core RNA polymerase (Subdomain 2.1 and domain 3), hydrophobic core formation (Subdomain 2.2), DNA melting (Subdomain 2.3), recognition of the −10 promoter motif (Subdomain 2.4), and recognition of the −35 promoter motif (Subdomains 4.1 and 4.2; W¨osten, 1998; Paget and Helmann, 2003). Since these domains are also present in all known plastidial σ -factors, it is to be expected that they are responsible for transcription from σ 70 -type promoters in plastids. However, structural analysis seems not to provide answers if the role of the different plastidial σ -factors is to selectively modify promoter activity and if they possess distinct or overlapping promoter specificities. Based on the phylogenetic analyses, one might presume that both might be true, and that plastidial σ -factors group into general σ -factors involved in transcription of standard σ 70 -type promoters and specialized σ -factors responsible for recognition of exceptional promoters in response to developmental and/or environmental cues.
8.4.2.2
Role of σ -factor diversity in transcriptional regulation
In order to address this question (see Table 8.1 summarizing putative roles of σ -factors), several groups carried out in vitro reconstitution and transcription experiments using recombinant σ -factors, which furthermore demonstrated that indeed plant σ -factor genes encode functional plastidial σ -factors (Kestermann et al., 1998; Hakimi et al., 2000; Beardslee et al., 2002; Homann and Link, 2003; Privat et al., 2003). The mustard SalSig1, SalSig2, and SalSig3 σ -factors indeed conferred distinctive binding specificity of the E. coli core RNA polymerase for certain tested
ndhF psbD
General
Sig4 Sig5
Sig6
psbD BLRP
General
Sig3
Others AGF
General, preference for some tRNA promoters
General
PEP-regulating factors σ -factors Sig1
Sig2
Target
Binds to AAG box; transcription enhancer
Light-independent early primary/alternative σ -factor (?)
Light-independent early primary/alternative σ -factor (?) σ -Factor in plant stress response (?) σ -Factor in plant stress response (?) and regulating psbD BLRP via blue-/UVA-light
Primary/alternative σ -factor (?)
Alternative σ -factor, may possibly need activating factor(s)
Function
Transcription factors and regulators in higher plants
Protein
Table 8.1
Wheat
Barley
Maize
Arabidopsis
At2g36990
At5g13730 At5g24120
At3g53920
At1g08540
Mustard Arabidopsis
Mustard Arabidopsis Mustard Arabidopsis Arabidopsis
At1g64860
Gene
Arabidopsis
Plant
(continued )
Kim and Mullet, 1995 Nakahira et al., 1998
Hakimi et al., 2000 Privat et al., 2003 Kestermann et al., 1998 Homann and Link, 2003 Hakimi et al., 2000 Hanaoka et al., 2003 Kanamaru et al., 2001 Privat et al., 2003 Shirano et al., 2000 Tsunoyama et al., 2002 Homann and Link, 2003 Privat et al., 2003 Homann and Link, 2003 Favory et al., 2005 Nagashima et al., 2004a Tsunoyama et al., 2004 Tsunoyama et al., 2002 Ishizaki et al., 2005 Lahiri and Allison, 2000
Reference
rrn16(P1)
psaA psaA rbcL
CDF2-A
Region U-building protein Region D-building protein RLBP SibI T3K9.5 cpCK2 (PTK)
Interaction with a NEP-2 transcription activity? DNA-building; transcription regulation Inactivates NEP activity by binding to RpoTp
Interaction with CDF2?; transcription regulation
DNA building; transcription regulation DNA building; transcription regulation DNA building; transcription regulation AthSig1-building protein AthSig1-building protein
DNA building; transcription repression
Binds to AAG box; transcription enhancer part of AGF Binds to PGT box DNA building; transcription regulation
Function
Note: The question mark signifies a proposed function rather than a proven one.
tRNAGlu
CDF-2B
rrn16
psbD BLRP rbcL
PGTF CDF1
NEP-regulating factors RPL4
psbD BLRP
Target
Transcription factors and regulators in higher plants (Continued)
PTF1
Protein
Table 8.1
Arabidopsis
Spinach
Spinach
Mustard
Spinach Spinach Tobacco Arabidopsis Arabidopsis Arabidopsis
Barley Pea Maize Spinach
Arabidopsis
Plant
trnE (ArthCt097)
X93160 (gi2792019)
At3g56710 At2g41180 At5g67380
At3g02150
Gene
Hanaoka et al., 2005
Bligny et al., 2000
Trifa et al., 1998
Baeza et al., 1991 Bligny et al., 2000 Cheng et al., 1997a Cheng et al., 1997a Kim et al., 2002 Morikawa et al., 2002 Morikawa et al., 2002 Baginsky and Gruissem, 2002 Baginsky et al., 1997 Ogrzewalla et al., 2002
Kim and Mullet, 1995 Lam et al., 1988
Baba et al., 2001
Reference
TRANSCRIPTION OF PLASTID GENES
201
promoters. While the psbA promoter was recognized by all three σ -factors, only SalSig1 and SalSig2 recognized the rbcL promoter. However, trnK, trnQ, rps16, and rrn16(PEP-P1) promoters were rather recognized by SalSig1 and SalSig3, but less efficiently by SalSig2 (Homann and Link, 2003). In Arabidopsis, similar experiments suggested that rather AthSig2 and AthSig3 confer specific recognition of the rbcL and psbA promoters than AthSig1, which recognized these promoters less efficiently (Hakimi et al., 2000; Privat et al., 2003). Given that both reconstitution systems consisted of plant σ -factors and the E. coli core RNA polymerase, the observed discrepancies in promoter recognition may be due to the heterologous transcription system with hindered abilities to identify species and/or PEP-specific regulatory elements on cis- and trans-factor level. Further efforts to address specific functionality of plant σ -factors in regulation of plastidial gene expression utilized characterization of their expression profiles. Consistent with a prominent role of PEP in leaves, most plastidial σ -factors are expressed in a light-dependent manner in green tissue but are silent in nonphotosynthetic roots (Isono et al., 1997b; Tanaka et al., 1997; Fujiwara et al., 2000; Oikawa et al., 2000). Moreover, expression of plastidial σ -factors seems to be differentially regulated during early Arabidopsis development. AthSig2, AthSig3, AthSig4, and AthSig6 but not AthSig1 and AthSig5 transcripts accumulate in 4-day-old seedlings (Ishizaki et al., 2005), while in 8-day-old seedlings transcript levels increase for all σ -factors (Nagashima et al., 2004a). Additionally, expression of Sig2 transcripts prior to Sig1 in developing leaves was reported for both Arabidopsis and rice, suggesting an early function of Sig2 in seedling development (Kanamaru et al., 1999; Kasai et al., 2004). Interestingly, unlike AthSig1 and AthSig2, AthSig3 protein accumulates in seeds and during early germination, which suggests regulation by post-transcriptional processes (Privat et al., 2003). A similar expression pattern was observed for the mustard SalSig3 factor, which accumulates rather in the dark- than in light-grown seedlings (Homann and Link, 2003). Hence, Sig3 may play a distinctive role in regulation of gene expression in etio- and/or proplastids. Similarly, ZmaSig6 was also detected in root, leaf base, and etiolated leaf tissue in maize (Lahiri and Allison, 2000). Therefore, it might be possible that Sig3 and Sig6 represent light-independent, early σ -factors regulating plastid gene expression during seedling growth and development. Contrarily, AthSig5 is expressed later in plant development, and furthermore, its RNA expression is redox-controlled downregulated (Fey et al., 2005) and rapidly induced by blue but not red light, which coincides with the blue-light-activated expression of psbD (Tsunoyama et al., 2002, 2004). In addition, to date AthSig5 is the only plant σ -factor activated by various stress cues (Nagashima et al., 2004b). Expression of some plastid genes in higher plants seems to be regulated by circadian rhythms (Nakahira et al., 1998). It is expected that this timing of plastid gene expression is controlled by nuclear factors, amid σ -factors being good candidates. Indeed, TaeSig1, NtaSig1, AthSig1, AthSig2, and PpaSig5 transcripts were shown to exhibit circadian or diurnal expression patterns (Kanamaru et al., 1999; Morikawa et al., 1999; Oikawa et al., 2000; Ichikawa et al., 2004). A recent approach to investigate gene function in plants is employment of knockout mutants or antisense lines. If plastids contain a primary σ -factor similar to most eubacteria, one would assume that inactivation of this gene would result in a drastic,
202
REGULATION OF TRANSCRIPTION IN PLANTS
most likely albino phenotype, by causing defects in PEP-dependent transcription of photosynthesis-related genes. However, examination of various Arabidopsis mutants of AthSig1, AthSig2, AthSig3, AthSig4, AthSig5, and AthSig6 did not reveal such a phenotype (Shirano et al., 2000; Kanamaru et al., 2001; Hanaoka et al., 2003; Privat et al., 2003; Nagashima et al., 2004b; Tsunoyama et al., 2004; Favory et al., 2005; Ishizaki et al., 2005). Still, a major breakthrough in revealing transcription specificity by σ -factors came by characterization of these plants. Cotyledons of AthSig6 knockout mutants displayed a transient pale green phenotype during early plant development, combined with a delay in light-dependent chloroplast development (Ishizaki et al., 2005). During this developmental stage the transcript pattern was found to be similar to that of rpo mutants, since transcript levels of most PEPdependent genes for photosynthesis components, rRNAs, and some tRNAs were decreased. Since the maize homolog ZmSig6 protein is expressed exclusively in tissue containing immature plastids (Lahiri and Allison, 2000), it was proposed that (Ath)Sig6 might be a general σ -factor serving PEP in an early, initial developmental stage. Nonetheless, given that after 8 days the mutant phenotype is restored to wild type it is plausible that other σ -factor(s) are able to take over AthSig6 function later in seedling development and plant growth (Shiina et al., 2005). Similarly, AthSig2 mutants displayed a pale green phenotype accompanied by reduced accumulation of plastid-encoded photosynthesis genes (Shirano et al., 2000; Kanamaru et al., 2001; Privat et al., 2003). Yet, accumulation of most of the photosynthesis gene transcripts was not reduced (Nagashima et al., 2004a). Interestingly, although it was previously shown that Sig2 is able to confer promoter specificity for psbA and rbcL promoters in vitro in Arabidopsis (Privat et al., 2003) and mustard (Homann and Link, 2003), several PEP-transcribed tRNAs including trnDGUC, trnE-UUC, trnM-CAU, and trnV-UAC were prominently reduced in AthSig2 knockout mutants (Kanamaru et al., 2001; Hanaoka et al., 2003) and antisense plants (Privat et al., 2003). Furthermore, overexpression of AthSig2 enhanced the transcription of trnE-trnD (Tsunoyama et al., 2004). Reduction of the photosynthesis-related components is suggested to be caused by defects in chlorophyll biosynthesis and plastid translation due to the decrease of trnE, an initiator of δ-aminolevulinic acid (ALA), and consequently chlorophyll synthesis. Hence Sig2 may have a primary role in driving transcription of certain plastid tRNAs, although it cannot be excluded that Sig2 is able to recognize other PEP promoters as suggested for psbA, psbD, and rbcL (Kanamaru et al., 2001; Hanaoka et al., 2003; Tsunoyama et al., 2004). Recently, characterization of an AthSig4 knockout mutant revealed a specific reduction in transcription of the plastid ndhF gene, resulting in a strong downregulation of the plastid NADH-specific dehydrogenase (NDH) activity (Favory et al., 2005). Therefore, ndhF expression and thus NDH activity seems to be regulated on transcriptional level, controlled by specific σ -factor AthSig4. Interestingly, regulation of NDH activity is involved in plant stress response (Casano et al., 2001) and leaf senescence (Zapata et al., 2005). Whether AthSig4 expression is modulated by such environmental or developmental parameters remains to be investigated. Some plastid genes are transcribed from promoters with unique architecture, e.g., the psbD BLRP, activated by blue and UVA-light (see Section 8.4.3). As shown by
TRANSCRIPTION OF PLASTID GENES
203
analyses of transcription in light-treated plants (Tsunoyama et al., 2002; Nagashima et al., 2004b), AthSig5 knockout plants, and overexpression studies (Nagashima et al., 2004b; Tsunoyama et al., 2004), AthSig5, a phylogenetically distinct σ -factor, is regulated by blue light and activates psbD BLRP transcription. Interestingly, analysis of a further AthSig5 knockout mutant showed embryo lethality (Yao et al., 2003). However, it is not understood yet why the different AthSig5 mutants exhibit these diverse phenotypes. PEP activity depends on the developmental stage of the plastids: it is downregulated in etioplasts and is active in chloroplasts (Rapp et al., 1992; DuBell and Mullet, 1995). Furthermore, rates of PEP transcription are higher in the light than in the dark (Shiina et al., 1998). Changes in PEP transcription activity were in part explained by changes in the phosphorylation state of σ -factors. Phosphorylation of σ -factors and the PEP enzyme itself has been shown to be an important regulatory event in chloroplast transcription (Tiller and Link, 1993a; Baginsky et al., 1997; Christopher et al., 1997). In mustard, a CK2-type kinase has been identified to be part of the chloroplast PEP-A complex (Ogrzewalla et al., 2002). This plastid transcription kinase activity, termed cpCK2, is able to phosphorylate biochemically purified sigma-like factors (SLFs) as well as subunits of the PEP-A complex. cpCK2 was proposed to be part of the signaling pathway controlling PEP activity, based on the observation that cpCK2 itself is differentially regulated by phosphorylation and redox state (Baginsky et al., 1999). The modifications of phosphorylation and SH-group redox state were shown to work antagonistically. A nonphosphorylated cpCK2 appears to be more active, but is inhibited by treatment with reduced glutathione (GSH), and vice versa, a phosphorylated nonactive enzyme could be reactivated by adding GSH. In vivo observations that cpCK2 isolated from plants grown under moderate light conditions effectively phosphorylated the associated PEP-A, while this was not observed with cpCK2 from plants subjected to 3 h of high light, corroborated these findings (Baena-Gonzalez et al., 2001). Therefore, light-dependent reduction of GSH would inactivate cpCK2, and dephosphorylation of PEP under high-light conditions would enhance PEP-dependent transcription. It remains unknown whether cpCK2 also regulated via extraplastidic signal chains mediated by phyto- and/or cryptochromes. Since cpCK2 orthologs have been identified in various plant species (Loschelder et al., 2004), it might well be that this kinase has an evolutionary conserved role in plastid redox-sensitive signal transduction. In bacteria, σ -factor activity is controlled by anti-σ factors (Ishihama, 2000). Plastid σ -factor AthSig1 associated proteins with plastid localization were identified in Arabidopsis (SibI and T3K9.5; Morikawa et al., 2002). They are not related to any proteins of known function and are light dependent, developmental, and tissuespecifically expressed, and thus may be involved in regulation of AthSig1 activity.
8.4.2.3
NEP transcription factors
As to the NEP subunit composition, to date one can only speculate based on information on the related mitochondrial RNA polymerases. Mitochondrial transcription complexes from humans, mice, Xenopus laevis, and yeast consist of a minimum of two components: the catalytic core enzyme (mtRPO, ∼120–150 kDa), and
204
REGULATION OF TRANSCRIPTION IN PLANTS
a specificity factor, which confers promoter recognition (mtTFB, ∼40–45 kDa). Despite poor overall sequence similarity, it recently has been shown that mtTFB factors belong to a family of RNA methyltransferases (Falkenberg et al., 2002; McCulloch et al., 2002; Rantanen et al., 2003; Seidel-Rogol et al., 2003). An additional component, which binds the DNA further upstream, enhances mitochondrial transcription in vitro (mtTFA, 20–25 kDa). This DNA-binding protein belongs to the HMG (high mobility group) family and may also facilitate the interaction with other trans-acting factors (reviewed in Jaehning, 1993; Shadel and Clayton, 1993; Tracy and Stern, 1995; Hess and B¨orner, 1999). To date, no mtTFA or mtTFB homologs have been isolated from plant mitochondria or plastids, and the presence of such proteins in plant organelles is unclear. A factor that is discussed to be involved in NEP transcription (summarized in Table 8.1) is the plastidial ribosomal protein L4 (RPL4; encoded by the nuclear Rpl4 gene). A role for RPL4 in transcription was proposed, as it copurifies with the T7-like transcription complex in spinach (Trifa et al., 1998). The ribosomal protein L4 was shown to have extraribosomal functions in transcriptional regulation in prokaryotes (Zengel et al., 1980). The spinach and Arabidopsis Rpl4 genes have acquired a remarkable 3 extension during evolutionary transfer to the nuclear genome, which resembles highly acidic C-terminal ends of some transcription factors. A function for this protein in NEP or PEP transcription, however, has yet to be demonstrated. CDF2-B, a DNA-binding factor activity biochemically isolated from spinach chloroplast, has been reported to stimulate transcription of the rrn operon Pc promoter by a NEP activity yet to be characterized (Bligny et al., 2000). Besides, some nucleus-encoded σ -factors generally imported into plastids were found to additionally localize into mitochondria, suggesting an additional role in regulating transcription by phage-type RNA polymerases (H. Tandara and K. Liere, unpublished data; Beardslee et al., 2002; Yao et al., 2003). Yet, experimental data to link the activity of the bacterial-type plastidial σ -factors to the phage-type enzymes in mitochondria or plastids has still to be shown.
8.4.3
Effects of environmental factors and crosstalk between plastids and nucleus
Plant development is highly influenced by environmental factors. Plastid gene expression was shown to differentially respond to environmental cues (Chory et al., 1995; Link, 1996; Goldschmidt-Clermont, 1998; Barkan and GoldschmidtClermont, 2000). Therefore, in the last decades cis- and trans-elements regulating differential gene expression in plastids were the center of attention (see Table 8.1 for summary). In addition to the core promoter region, regulatory sequence motifs upstream the −35 region were found in the promoters of rbcL and psbD-psbC. The rbcL gene encodes the large subunit of ribulose-1,5-bisphosphate carboxylase and is transcribed from a single PEP promoter with well-conserved −35 and −10 elements and canonical spacing by 18 nucleotides (Shinozaki and Sugiura, 1982; Mullet et al., 1985; Reinbothe et al., 1993; Isono et al., 1997a). The importance of both the −35 and −10 box spacing and sequence for rbcL promoter strength was
TRANSCRIPTION OF PLASTID GENES
205
confirmed by in vitro studies (Gruissem and Zurawski, 1985; Hanley-Bowdoin et al., 1985). An upstream element, conserved between maize, pea, spinach, and tobacco, was proposed to function as a binding site for the chloroplast DNA-binding factor 1 (CDF1) in maize (Lam et al., 1988). Interestingly, a segment of CDF1, region II, is reminiscent of the AT-rich UP element, which stimulates transcription by a factor of 30 in E. coli (Ross et al., 1993). However, a study of transplastomic plants with chimeric PrbcL::uidA constructs demonstrated that the rbcL core promoter is sufficient to obtain wild-type rates of transcription (Shiina et al., 1998). Unless the constructs contained the rbcL 5 UTR, transcription rates of rbcL were significantly reduced in dark-adapted plants, resulting in lower steady-state uidA mRNA levels in the dark. Therefore, stabilization of the rbcL mRNA via its 5 UTR is compensating for reduced rates of transcription in the dark and leads to light-dependent transcript accumulation. Nonetheless, another DNA-binding protein RLBP (rbcL promoterbinding protein) binds specifically to the rbcL promoter core in tobacco (−3 to −32; Kim et al., 2002). Only detectable in light-grown seedlings, RLBP is suggested to play a role in light-dependent rbcL transcription. Contrary to most photosynthetic genes, the rate of transcription of psbD-psbC remains high in mature chloroplasts (Klein and Mullet, 1990; Baumgartner et al., 1993; DuBell and Mullet, 1995). Responsible is the activation of the BLRP (Sexton et al., 1990). The psbD BLRP promoter architecture of various species studied to date is similar (Figure 8.2; Christopher et al., 1992; Wada et al., 1994; Allison and Maliga, 1995; Kim and Mullet, 1995; To et al., 1996; Hoffer and Christopher, 1997; Kim et al., 1999b; Thum et al., 2001). The architecture of the psbD BLRP promoter, which has two conserved upstream elements (PGT box, AAG box) and poorly conserved and closely spaced −10/−35-elements (15 nucleotides instead of the usual 17–19), was studied in transplastomic tobacco in vivo (Allison and Maliga, 1995). Deletion of part of the PGT box reduced mRNA levels, while subsequent deletion of the AAG-box sequences further reduced transcript levels. Thus, sequences directly upstream of the psbD promoter core are accountable for a light-activated transcript accumulation in tobacco. In rice, wheat, and barley, transcription from the psbD promoter in vitro depends on the −10 but not on the −35 promoter element (To et al., 1996; Satoh et al., 1997; Nakahira et al., 1998; Kim et al., 1999b). The AAG box of the barley promoter was shown to be the binding site for a nuclear-encoded AAG-binding complex in vitro (AGF; Kim and Mullet, 1995). However, binding activity of AGF to the AAG box is not correlated with psbD BLRP transcription activity (Nakahira et al., 1998). One of the components of AGF in Arabidopsis was cloned and named plastid transcription factor 1 (PTF1; Baba et al., 2001). Studies in PTF1-deficient mutants revealed that indeed PTF1 is rather responsible in general transcriptional enhancement than in light-dependent activation of psbD transcription. Correspondingly, the PGT box is the binding site for PGTF, the PGT-binding factor, the activity of which is regulated by an ADP-dependent kinase (Kim et al., 1999a). A model based on these in vitro experiments in barley explains that constitutive binding of AGF to the upstream AAG element assists promoter recognition by PEP. Light-dependent transcriptional activation is mediated by PGTF binding to the PGT box. In the dark, PGTF phosphorylation results in loss of affinity for the
206
REGULATION OF TRANSCRIPTION IN PLANTS
PGT element, which is thought to hinder transcription. Although psbD promoter architecture is highly conserved, it is unlikely that PGT is required for light-dependent transcription in various other plants. As it was shown for rice (To et al., 1996), wheat (Satoh et al., 1997), and barley in vitro (Kim et al., 1999b), and in vivo in transplastomic tobacco (Thum et al., 2001), the PGF box is not required for light-dependent activation in these plants. Therefore, the role of PGT and PGTF remains largely unknown. Based on the data available, it was proposed that Sig5 might act as a mediator of blue-light signaling in activating psbD BLRP transcription in blue light (Tsunoyama et al., 2002, 2004; Nagashima et al., 2004b), whereas AGF enhances psbD BLRP transcription by constitutively binding to the AAG box (Shiina et al., 2005). The signal transduction pathway is assumed to involve reception of blue light by cryptochromes and phyA (Thum et al., 2001; Mochizuki et al., 2004), further mediation by a protein phosphatase PP7 (Moller et al., 2003), and subsequent induction of Sig5 expression (Mochizuki et al., 2004). Sig5 then is imported into plastids where it together with AGF (PTF1) initiates psbD transcription. Furthermore, psbD BLRP activity is also regulated in a developmental and tissue-specific manner, since the Arabidopsis DET1 gene product downregulates the activity of psbD BLRP in young seedlings (Christopher and Hoffer, 1998). Plastids are under environmental control as shown for their gene expression, e.g., psbD BLRP. However, most intense is the control by light of plastidial differentiation from proplastids to either etioplasts (dark) or chloroplasts (light). Analyses of photomorphogenic mutants established the existence of different pathways to communicate light perception to plastids in order to control their development (Leon et al., 1998; Rodermel, 2001; Gray et al., 2003; L´opez-Juez and Pyke, 2005). However, these analyses also showed that retrograde or ‘plastid signals’ are controlling nuclear gene expression, depending on the developmental status of the plastid (Figure 8.5; Rodermel, 2001; Gray, 2003; Gray et al., 2003; Brown et al., 2005). Although both plastidial transcription and translation are necessary for the production of a ‘plastid signal’, it is not an immediate translational product of a plastidial gene (Oelm¨uller et al., 1986; Lukens et al., 1987), but rather part(s) of signal transduction pathways. Light is used as not only an energy source for photosynthesis, but also an environmental signal to regulate plant biogenesis and environmental adaptation. Apart from blue/UVA-light, illumination has been early hypothesized to control plastid gene expression via the physiological status of the plastid, e.g., redox conditions (Link, 2003; Pfannschmidt and Liere, 2005). Redox control of plastidial gene expression is interpreted as a selection force throughout evolution to retaining their genomes (Allen, 1993). First confirmation for such a redox control was obtained by demonstrating that light supported incorporation of radioactive-labeled NADH into the RNA fraction of lettuce plastids (Pearson et al., 1993). Plastidial gene expression is controlled by photosynthetic activity on different levels such as RNA maturation (Deshpande et al., 1997; Liere and Link, 1997; Salvador and Klein, 1999) and translation (Danon and Mayfield, 1994; Bruick and Mayfield, 1999; Trebitsh et al., 2000; Zhang et al., 2000), by photosynthetic activity. Effects on plastidial gene transcription then were demonstrated by growing plants under light conditions generating an imbalance in excitation energy distribution between the photosystems
TRANSCRIPTION OF PLASTID GENES
207
Figure 8.5 Speculative model of the role of nuclear-encoded phage-type RNA polymerases in regulation of plastidial gene expression. The nuclear-located RpoTp gene encodes a phage-type RNA polymerase, which is part of the NEP transcription activity. NEP transcribes and therefore may regulate expression of the plastidial rpoB operon encoding subunits of the plastid-encoded RNA polymerase (PEP). PEP transcribes components of the photosynthetic complexes (PSI, PSII), which regulate nuclear transcription by generating diverse ‘plastid signals’ (ROS, reactive oxygen species). PEP also transcribes trnE encoding trnAGlu (Hess et al., 1992; Walter et al., 1995), which is required for the synthesis of δ-aminolevulinic acid (ALA). ALA is a precursor of the chlorophyll and heme biosynthesis, which are thought to provide plastid signals influencing nuclear transcription. Furthermore, tRNAGlu is assumed to developmentally inhibit NEP transcription activity by binding to RpoTp (Hanaoka et al., 2005). In turn, transcription of plastid genes and consequently the developmental stage of the plastid regulate expression and activity of nuclear-encoded, plastid phage-type RNA polymerase (RpoTp; Emanuel et al., 2004). Thus, the regulated network of nuclear and plastidial transcription machineries may be a key element for a concerted expression of genes located within plastids and the nucleus of the plant cell.
(PSII- and PSI-light, 680 and 700 nm, respectively; Pfannschmidt et al., 1999a,b; Fey et al., 2005). Preferential excitation of PSII results in a reduction of the electron transport chain, while a preferential excitation of PSI results in its oxidation. A change in photosystem stoichiometry correlated with respective changes in the transcriptional rates and transcript amounts of the plastidial genes for the reaction center proteins of PSII and PSI, psbA and psaAB. Further experiments demonstrated that the transcriptional regulation was independent from cytosolic factors such as photoreceptors, but that the redox state of the plastoquinone pool (PQ) is the major determinant for the changes in gene expression. When the PQ pool is mainly reduced, transcription of the psaAB operon is promoted while in the opposite case psbA
208
REGULATION OF TRANSCRIPTION IN PLANTS
transcription is increased. Opposite regulation of these genes has been recently found also in pea (Tullberg et al., 2000), Chlamydobotrys stellata (Kovacs et al., 2000), and Synechocystis PCC 6803 (Li and Sherman, 2000; El Bissati and Kirilovsky, 2001) suggesting that this mechanism represents an evolutionary old regulation. These data provide a first model on how plants adapt to light quality gradients occurring in natural environments under low light intensities. Signal transduction from the PQ pool toward the level of transcription is not clear yet. However, a long-term response possibly represents an extended branch of the short-term response (the state transition), which is also regulated by the redox state of the PQ pool (Allen and Forsberg, 2001; Pursiheimo et al., 2001). The PQ oxidation site at the cyt b6 f complex functions as a sensor for the PQ redox state during state transition (Vener et al., 1997; Zito et al., 1999). A small 9-kDa protein of PSII, TSP9, has been discussed to be a putative signal transducer towards transcription. The putative DNA-binding protein TSP9 is partially released from PSII upon PQ reduction in spinach (Carlberg et al., 2003; Zer and Ohad, 2003). Identification of an additional protein of 31 kDa capable of sequence-specific binding between positions +64 to +83 (region D) of the light dependentpsaAB PEP promoter region (Chen et al., 1993, 1997a) suggests the existence of yet unidentified transcription factors that forward redox signals. Furthermore, the Arabidopsis high chlorophyll fluorescence mutant hcf145 shows decreased mRNA stability and transcription of psaA (Lezhneva and Meurer, 2004). Thus, HCF145 might be involved in transcriptional regulation of the psaA operon. However, further analysis of this promoter has yet to be reported. Microarray analyses of the influence of light quality on the transcriptome to assess redox-regulated transcription of nuclear-encoded plastidial genes (Fey et al., 2005) revealed a regulatory impact on transcription of genes for PEP components: rpoB (plastid-encoded β-subunit), AthSig5 (nuclear-encoded σ -factor), and SibI (nuclear-encoded Sig1-binding protein; Morikawa et al., 2002). PEP therefore is not only responsible for the redox regulation at the psbA and psaAB promoters, but apparently is also regulated transcriptionally via redox control. Interestingly, rpoB is transcribed by a nuclear-encoded phage-type RNA polymerase (Figure 8.5, RpoTp; Liere et al., 2004), suggesting a redox regulation of this enzyme. Since this analyses did not indicate differences in the glutathione redox state, as suggested by the redox control of PEP mediated by glutathione (Baginsky et al., 1999), it seems likely that several distinct redox control pathways to control plastidial transcription exist (Link, 2003; Pfannschmidt and Liere, 2005), depending on environmental conditions such as responses to low or high light. Extensive further studies, however, are necessary to resolve this important question. The barley mutant albostrians, with alternating stripes of white and green tissue, contains no detectable ribosomes in plastids of white tissue cells (Siemenroth et al., 1981). Transcript levels of some photosynthesis-related plastidial and nuclear genes are reduced or missing suggesting the existence of ‘plastid signals’ controlling nuclear gene expression (Hess et al., 1993, 1994). However, similar to rpoB knockout tobacco plants (Hajdukiewicz et al., 1997), plastid transcription of certain genes is maintained by a nuclear-encoded transcription activity (NEP; Hess et al., 1993; H¨ubschmann and B¨orner, 1998). Recently, transcript levels of the nuclear-encoded
TRANSCRIPTION OF PLASTID GENES
209
RpoTp, which is likely to represent NEP activity, and its plastidial target genes were analyzed throughout the developmental gradient of albostrians leaves (Emanuel et al., 2004). The results revealed a significant influence of the developmental stage of plastids on expression and activity of RpoTp, indicating a plastid-to-nucleus signaling to coordinate expression of plastidial and nuclear-encoded RNA polymerases as a prerequisite of a concerted gene expression in both plastids and nucleus (Figure 8.5). Recently, a regulatory role, which links chlorophyll synthesis and the developmental switch from nucleus-encoded RNA polymerases to the plastid-encoded bacterial-type enzyme, has been proposed for the plastid-encoded tRNAGlu in Arabidopsis (Hanaoka et al., 2005). tRNAGlu is not only required for translation, but also for synthesis of ALA, a precursor of chlorophyll (Sch¨on et al., 1986). The tRNA was specifically bound by recombinant RpoTp in gel mobility shift experiments. Additionally, transcription from a putative plastidial accD NEP promoter sequence was inhibited by addition of tRNAGlu to in vitro transcription reactions with proplastid extracts from Arabidopsis. Hence, the authors suggested tRNAGlu to developmentally inhibit transcription by RpoTp (Figure 8.5). In bacteria, one of the most important processes to regulate gene expression is the so-called ‘stringent control’ enabling adaptation to nutrient-limiting conditions (Cashel et al., 1996). The effector molecule is guanosine 5 -diphosphate 3 -diphosphate (ppGpp), which binds to the core RNA polymerase modifying its promoter specificity (Toulokhonov et al., 2001). Stress-induced synthesis is mediated by ppGpp synthetases, RelA and SpoT, homologs of which were found in C. reinhardtii (Kasai et al., 2002), Arabidopsis (van der Biezen et al., 2000), and tobacco (Givens et al., 2004). Plastidial targeting has been demonstrated for some of these RSH termed proteins, suggesting an implication in ppGpp signaling in plastids. Given that RSH expression and plastidial ppGpp levels are clearly elevated by light and various abiotic and biotic stress conditions and that PEP activity is inhibited by ppGpp in vitro (Givens et al., 2004; Takahashi et al., 2004), it is conceivable that PEP might indeed be under control of a bacterial-like stringent response mediated by ppGpp. Interestingly, stress signals specifically induce transcription initiation from the psbD BRLP conferred by a special σ -factor, Sig5 (Nagashima et al., 2004b; Tsunoyama et al., 2004). However, target genes that are regulated by a plastidial stringent control have yet to be identified, which will elucidate the molecular mechanisms of transcriptional responses to plant hormones and environmental stress situations.
References P. Aguettaz, P. Seyer, H. Pesey and A.-M. Lescure (1987) Relations between the plastid gene dosage and levels of 16S rRNA and rbcL gene transcripts during amyloplast to chloroplast change in mixotrophic spinach cell suspensions. Plant Molecular Biology 8, 169–177. C. Aldridge, J. Maple and S.G. Møller (2005) The molecular biology of plastid division in higher plants. Journal of the Experimental Botany 56, 1061–1077. J.F. Allen (1993) Control of gene expression by redox potential and the requirement for chloroplast and mitochondrial genomes. Journal of Theoretical Biology 165, 609–631.
210
REGULATION OF TRANSCRIPTION IN PLANTS
J.F. Allen and G.J. Forsber (2001) Molecular recognition in thylakoid structure and function. Trends in Plant Science 6, 317–326. L.A. Allison and P. Maliga (1995) Light-responsive and transcription-enhancing elements regulate the plastid psbD core promoter. The EMBO Journal 14, 3721–3730. L.A. Allison, L.D. Simon and P. Maliga (1996) Deletion of rpoB reveals a second distinct transcription system in plastids of higher plants. The EMBO Journal 15, 2802–2809. K. Baba, T. Nakano, K. Yamagishi and S. Yoshida (2001) Involvement of a nuclear-encoded basic helixloop-helix protein in transcription of the light-responsive promoter of psbD. Plant Physiology 125, 595–603. K. Baba, T.J. Schmid, A. Espinosa-Ruiz, et al. (2004) Organellar gene transcription and early seedling development are affected in the RpoT;2 mutant of Arabidopsis. The Plant Journal 38, 38–48. E. Baena-Gonzalez, S. Baginsky, P. Mulo, H. Summer, E.-M. Aro and G. Link (2001) Chloroplast transcription at different light intensities. Glutathione-mediated phosphorylation of the major RNA polymerase involved in redox-regulated organellar gene expression. Plant Physiology 127, 1044–1052. L. Baeza, A. Bertrand, R. Mache and S. Lerbs-Mache (1991) Characterization of a protein binding sequence in the promoter region of the 16S rRNA gene of the spinach chloroplast genome. Nucleic Acids Research 19, 3577–3581. S. Baginsky and W. Gruissem (2002) Endonucleolytic activation directs dark-induced chloroplast mRNA degradation. Nucleic Acids Research 30, 4527–4533. S. Baginsky, K. Tiller and G. Link (1997) Transcription factor phosphorylation by a protein kinase associated with chloroplast RNA polymerase from mustard (Sinapis alba). Plant Molecular Biology 34, 181–189. S. Baginsky, K. Tiller, T. Pfannschmidt and G. Link (1999) PTK, the chloroplast RNA polymeraseassociated protein kinase from mustard (Sinapis alba), mediates redox control of plastid in vitro transcription. Plant Molecular Biology 39, 1013–1023. A. Barkan (2004) Intron splicing in plant organelles. In: Molecular Biology and Biotechnology of Plant Organelles (eds H. Daniell and C.D. Chase). Springer, Dordrecht. A. Barkan and M. Goldschmidt-Clermont (2000) Participation of nuclear genes in chloroplast gene expression. Biochimie 82, 559–572. D. Barnes and S.P. Mayfield (2003) Redox control of posttranscriptional processes in the chloroplast. Antioxidants & Redox Signaling 5, 89–94. B.J. Baumgartner, J.C. Rapp and J.E. Mullet (1989) Plastid transcription activity and DNA copy number increase early in barley chloroplast development. Plant Physiology 89, 1011–1018. B.J. Baumgartner, J.C. Rapp and J.E. Mullet (1993) Plastid genes encoding the transcription/translation apparatus are differentially transcribed early in barley (Hordeum vulgare) chloroplast development: evidence for selective stabilization of psbA mRNA. Plant Physiology 101, 781–791. T.A. Beardslee, S. Roy-Chowdhury, L. Jaiswal, S. Lerbs-Mache, D.B. Stern and L.A. Allison (2002) A nucleus-encoded maize protein with sigma factor activity accumulates in mitochondria and chloroplasts. The Plant Journal 31, 199–209. J.R. Bedbrook and L. Bogorad (1976) Endonuclease recognition sites mapped on Zea mays chloroplast DNA. Proceedings of National Academy of Sciences of the United States of America 73, 4309– 4313. A.J. Bendich (2004) Circular chloroplast chromosomes: the grand illusion. The Plant Cell 16, 1661– 1666. T. Berends Sexton, J.T. Jones and J.E. Mullet (1990) Sequence and transcriptional analysis of the barley ctDNA region upstream of psbD-psbC encoding trnK (UUU), rps16, trnQ (UUG), psbK, psbI, and trnS (GCU). Current Genetics 17, 445–454. S. Berg, K. Krause and K. Krupinska (2004) The rbcL genes of two Cuscuta species, C. gronovii and C. subinclusa, are transcribed by the nuclear-encoded plastid RNA polymerase (NEP). Planta 219, 541–546. K.J. Bergsland and R. Haselkorn (1991) Evolutionary relationships among eubacteria, cyanobacteria, and chloroplasts: evidence from the rpoC1 gene of Anabaena sp. strain PCC 7120. Journal of Bacteriology 173, 3446–3455.
TRANSCRIPTION OF PLASTID GENES
211
C. Bisanz-Seyer, Y.-F. Li, P. Seyer and R.G. Mache (1989) The components of the plastid ribosome are not accumulated synchronously during the early development of spinach plants. Plant Molecular Biology 12, 201–211. M. Bligny, F. Courtois, S. Thaminy, et al. (2000) Regulation of plastid rDNA transcription by interaction of CDF2 with two different RNA polymerases. The EMBO Journal 19, 1851–1860. R. Bock (2000) Sense from nonsense: how the genetic information of chloroplasts is altered by RNA editing. Biochimie 82, 549–557. T.J. Bollenbach, G. Schuster and D.B. Stern (2004) Cooperation of endo- and exoribonucleases in chloroplast mRNA turnover. Progress in Nucleic Acid Research and Molecular Biology 78, 305– 337. J. Bown, K. Barne, S. Minchin and S. Busby (1997) Extended – 10 promoters. Nucleic Acids and Molecular Biology 11, 41–52. S.K. Boyer and J.E. Mullet (1986) Characterization of P. sativum chloroplast psbA transcripts produced in vivo and in vitro and in E. coli. Plant Molecular Biology 6, 229–243. S.K. Boyer and J.E. Mullet (1988) Sequence and transcript map of barley chloroplast psbA gene. Nucleic Acids Research 16, 8184. D. Bradley and A.A. Gatenby (1985) Mutational analysis of the maize chloroplast ATPase-beta subunit gene promoter: the isolation of promoter mutants in E. coli and their characterization in a chloroplast in vitro transcription system. The EMBO Journal 4, 3641–3648. N.J. Brown, J.A. Sullivan and J.C. Gray (2005) Light and plastid signals regulate the expression of the pea plastocyanin gene through a common region at the 5 end of the coding region. The Plant Journal 43, 541–552. R.K. Bruick and S.P. Mayfield (1999) Light-activated translation of chloroplast mRNAs. Trends in Plant Science 4, 190–195. S. B¨ulow and G. Link (1988) Sigma-like activity from mustard (Sinapis alba L.) chloroplasts confering DNA-binding and transcription specificity to E. coli core RNA polymerase. Plant Molecular Biology 10, 349–357. W. B¨unger and J. Feierabend (1980) Capacity for RNA synthesis in 70S ribosome-deficient plastids of heat-bleached rye leaves. Planta 149, 163–169. A.B. Cahoon, F.M. Harris and D.B. Stern (2004) Analysis of developing maize plastids reveals two mRNA stability classes correlating with RNA polymerase type. EMBO Reports 5, 801–806. I. Carlberg, M. Hansson, T. Kieselbach, W.P. Schroder, B. Andersson and A.V. Vener (2003) A novel plant protein undergoing light-induced phosphorylation and release from the photosynthetic thylakoid membranes. Proceedings of National Academy of Sciences of the United States of America 100, 757–762. L.M. Casano, M. Martin and B. Sabater (2001) Hydrogen peroxide mediates the induction of chloroplastic NDH complex under photooxidative stress in barley. Plant Physiology 125, 1450–1458. M. Cashel, D.M. Gentry, V.J. Hernandez and D. Vinella (1996) The stringent response. In: Escherichia coli and Salmonella typhimurium: Cellular and Molecular Biology (ed. F.C. Neidhardt). ASM Press, Washington DC. N. Cermakian, T.M. Ikeda, R. Cedergren and M.W. Gray (1996) Sequences homologous to yeast mitochondrial and bacteriophage T3 and T7 RNA polymerases are widespread throughout the eukaryotic lineage. Nucleic Acids Research 24, 648–654. C.-C. Chang, J. Sheen, M. Bligny, Y. Niwa, S. Lerbs-Mache and D.B. Stern (1999) Functional analysis of two maize cDNAs encoding T7-like RNA polymerases. The Plant Cell 11, 911–926. M.C. Chen, M.C. Cheng and S.C. Chen (1993) Characterization of the promoter of rice plastid psaApsaB-rps14 operon and the DNA-specific binding proteins. Plant and Cell Physiology 34, 577– 584. M.C. Cheng, S.P. Wu, L.F.O. Chen and S.C.G. Chen (1997a) Identification and purification of a spinach chloroplast DNA-binding protein that interacts specifically with the plastid psaA-psaBrps14 promoter region. Planta 203, 373–380. Y.S. Cheng, C.H. Lin and L.J. Chen (1997b) Transcription and processing of the gene for spinach chloroplast threonine tRNA in a homologous in vitro system. Biochemical and Biophysical Research Communications 233, 380–385.
212
REGULATION OF TRANSCRIPTION IN PLANTS
J. Chory, R.K. Cook, R. Dixon, et al. (1995) Signal-transduction pathways controlling light-regulated development in Arabidopsis. Philosophical Transactions of the Royal Society of London Series B: Biological Sciences 350, 59–65. D.A. Christopher and P.H. Hoffer (1998) DET1 represses a chloroplast blue light-responsive promoter in a developmental and tissue-specific manner in Arabidopsis thaliana. The Plant Journal 14, 1–11. D.A. Christopher, M. Kim and J.E. Mullet (1992) A novel light-regulated promoter is conserved in cereal and dicot chloroplasts. The Plant Cell 4, 785–798. D.A. Christopher, X.L. Li, M. Kim and J.E. Mullet (1997) Involvement of protein kinase and extraplastidic serine/threonine protein phosphatases in signaling pathways regulating plastid transcription and the psbD blue light-responsive promoter in barley. Plant Physiology 113, 1273–1282. L. Chun, A. Kawakami and D.A. Christopher (2001) Phytochrome A mediates blue light and UV-Adependent chloroplast gene transcription in green leaves. Plant Physiology 125, 1957–1966. A. Danon and S.P. Mayfield (1994) Light-regulated translation of chloroplast messenger RNAs through redox potential. Science 266, 1717–1719. X.W. Deng and W. Gruissem (1987) Control of plastid gene expression during development: the limited role of transcriptional regulation. Cell 49, 379–387. N.N. Deshpande, Y. Bao and D.L. Herrin (1997) Evidence for light/redox-regulated splicing of psbA pre-RNAs in Chlamydomonas chloroplasts. RNA 3, 37–48. A.N. Dubell and J.E. Mullet (1995) Differential transcription of pea chloroplast genes during lightinduced leaf development. Plant Physiology 109, 105–112. S.D. Dyall, M.T. Brown and P.J. Johnson (2004) Ancient invasions: from endosymbionts to organelles. Science 304, 253–257. A. Eisermann, K. Tiller and G. Link (1990) In vitro transcription and DNA binding characteristics of chloroplast and etioplast extracts from mustard (Sinapis alba) indicate differential usage of the psbA promoter. The EMBO Journal 9, 3981–3987. K. El Bissati and D. Kirilovsky (2001) Regulation of psbA and psaE expression by light quality in Synechocystis species PCC 6803. A redox control mechanism. Plant Physiology 125, 1988–2000. C. Emanuel, U. von Groll, M. M¨uller, T. B¨orner and A. Weihe (2006) Development- and tissuespecific expression of the RpoT gene family of Arabidopsis encoding mitochondrial and plastid RNA polymerases. Planta 223, 998–1009. C. Emanuel, A. Weihe, A. Graner, W.R. Hess and T. B¨orner (2004) Chloroplast development affects expression of phage-type RNA polymerases in barley leaves. The Plant Journal 38, 460–472. S.C. Ems, C.W. Morden, C.K. Dixon, K.H. Wolfe, C.W. Depamphilis and J.D. Palmer (1995) Transcription, splicing and editing of plastid RNAs in the nonphotosynthetic plant Epifagus virginiana. Plant Molecular Biology 29, 721–733. J. Falk, A. Schmidt and K. Krupinska (1993) Characterization of plastid DNA transcription in ribosome deficient plastids of heat-bleached barley leaves. Journal of Plant Physiology 141, 176–181. M. Falkenberg, M. Gaspari, A. Rantanen, A. Trifunovic, N.-G. Larsson and C.M. Gustafsson (2002) Mitochondrial transcription factors B1 and B2 activate transcription of human mtDNA. Nature Genetics 31, 289–294. J.-J. Favory, M. Kobayshi, K. Tanaka, et al. (2005) Specific function of a plastid sigma factor for ndhF gene transcription. Nucleic Acids Research 33, 5991–5999. V. Fey, R. Wagner, K. Brautigam, et al. (2005) Retrograde plastid redox signals in the expression of nuclear genes for chloroplast proteins of Arabidopsis thaliana. The Journal of the Biological Chemistry 280, 5318–5328. M. Fujiwara, A. Nagashima, K. Kanamaru, K. Tanaka and H. Takahashi (2000) Three new nuclear genes, sigD, sigE and sigF, encoding putative plastid RNA polymerase σ factors in Arabidopsis thaliana. FEBS Letters 481, 47–52. G. Galli, H. Hofstetter and M.L. Birnstil (1981) Two conserved blocks within eukaryotic tRNA genes are major promoter elements. Nature 294, 626–631. A.A. Gatenby, J.A. Castleton and M.W. Saul (1981) Expression in E. coli of maize and wheat chloroplast genes for large subunit of ribulose bisphosphate carboxylase. Nature 291, 117–121.
TRANSCRIPTION OF PLASTID GENES
213
A. Gauly and H. K¨ossel (1989) Evidence for tissue-specific cytosine-methylation of plastid DNA from Zea mays. Current Genetics 15, 371–376. E.P. Geiduschek, C. Bardeleben, C.A. Joazeiro, G.A. Kassavetis and S. Whitehall (1995) Yeast RNA polymerase III: transcription complexes and RNA synthesis. Brazilian Journal of Medical and Biological Research 28, 147–159. R.M. Givens, M.H. Lin, D.J. Taylor, U. Mechold, J.O. Berry and V.J. Hernandez (2004) Inducible expression, enzymatic activity, and origin of higher plant homologues of bacterial RelA/SpoT stress proteins in Nicotiana tabacum. The Journal of the Biological Chemistry 279, 7495–7504. M. Goldschmidt-Clermont (1998) Coordination of nuclear and chloroplast gene expression in plant cells. International Review of Cytology 177, 115–180. J.C. Gray (2003) Chloroplast-to-nucleus signalling: a role for Mg-protoporphyrin. Trends in Genetics 19, 526–529. J.C. Gray, J.A. Sullivan, J.H. Wang, C.A. Jerome and D. Maclean (2003) Coordination of plastid and nuclear gene expression. Philosophical Transactions of the Royal Society of London Series B: Biological Sciences 358, 135–144; discussion 144–145. M.W. Gray (2004) The evolutionary origins of plant organelles. In: Molecular Biology and Biotechnology of Plant Organelles (eds H. Daniell and C.D. Chase). Springer, Dordrecht. M.W. Gray and B.F. Lang (1998) Transcription in chloroplasts and mitochondria: a tale of two polymerases. Trends in Microbiology 6, 1–3. B.M. Greenberg, V. Gaba, O. Canaani, S. Malkin, A.K. Mattoo and M. Edelman (1989) Separate photosensitizers mediate degradation of the 32-kDa photosystem II reaction center protein in the visible and UV spectral regions. Proceedings of the National Academy of Sciences of the United States of America 86, 6617–6620. T.M. Gruber and D.A. Bryant (1997) Molecular systematic studies of eubacteria, using sigma70-type sigma factors of group 1 and group 2. Journal of Bacteriology 179, 1734–1747. W. Gruissem, C. Elsner-Menzel, S. Latshaw, J.O. Narita, M.A. Schaffer and G. Zurawski (1986) A subpopulation of spinach chloroplast tRNA genes does not require upstream promoter elements for transcription. Nucleic Acids Research 14, 7541–7556. W. Gruissem and J.C. Tonkyn (1993) Control mechanisms of plastid gene expression. Critical Reviews in Plant Sciences 12, 19–55. W. Gruissem and G. Zurawski (1985) Analysis of promoter regions for the spinach chloroplast rbcL, atpB and psbA genes. The EMBO Journal 4, 3375–3383. R. Hagemann (2004) The sexual inheritance of plant organelles. In: Molecular Biology and Biotechnology of Plant Organelles (eds H. Daniell and C.D. Chase). Springer, Dordrecht. P.T.J. Hajdukiewicz, L.A. Allison and P. Maliga (1997) The two RNA polymerases encoded by the nuclear and the plastid compartments transcribe distinct groups of genes in tobacco plastids. The EMBO Journal 16, 4041–4048. M.A. Hakimi, I. Privat, J.G. Valay and S. Lerbs-Mache (2000) Evolutionary conservation of C-terminal domains of primary sigma70 -type transcription factors between plants and bacteria. The Journal of Biological Chemistry 275, 9215–9221. J. Haley and L. Bogorad (1990) Alternative promoters are used for genes within maize chloroplast polycistronic transcription units. The Plant Cell 2, 323–333. C.D. Han, W. Patrie, M. Polacco and E.H. Coe (1993) Abberations in plastid transcripts and deficiency of plastid DNA in striped and albino mutants in maize. Planta 191, 552–563. M. Hanaoka, K. Kanamaru, M. Fujiwara, H. Takahashi and K. Tanaka (2005) Glutamyl-tRNA mediates a switch in RNA polymerase use during chloroplast biogenesis. EMBO Reports 6, 545–550. M. Hanaoka, K. Kanamaru, H. Takahashi and K. Tanaka (2003) Molecular genetic analysis of chloroplast gene promoters dependent on SIG2, a nucleus-encoded sigma factor for the plastid-encoded RNA polymerase, in Arabidopsis thaliana. Nucleic Acids Research 31, 7090–7098. L. Hanley-Bowdoin, E.M.J. Orozco and N.H. Chua (1985) In vitro synthesis and processing of a maize chloroplast transcript encoded by the ribulose 1,5-bisphosphate carboxylase large subunit gene. Molecular and Cellular Biology 5, 2733–2745.
214
REGULATION OF TRANSCRIPTION IN PLANTS
K. Hara, M. Morita, R. Takahashi, M. Sugita, S. Kato and S. Aoki (2001a) Characterization of two genes, Sig1 and Sig2, encoding distinct plastid sigma factors(1) in the moss Physcomitrella patens: phylogenetic relationships to plastid sigma factors in higher plants. FEBS Letters 499, 87–91. K. Hara, M. Sugita and S. Aoki (2001b) Cloning and characterization of the cDNA for a plastid sigma factor from the moss Physcomitrella patens. Biochimica et Biophysica Acta 1517, 302–306. B. Hedtke, T. B¨orner and A. Weihe (1997) Mitochondrial and chloroplast phage-type RNA polymerases in Arabidopsis. Science 277, 809–811. B. Hedtke, T. B¨orner and A. Weihe (2000) One RNA polymerase serving two genomes. EMBO Reports 1, 435–440. B. Hedtke, J. Legen, A. Weihe, R.G. Herrmann and T. B¨orner (2002) Six active phage-type RNA polymerase genes in Nicotiana tabacum. The Plant Journal 30, 625–637. B. Hedtke, M. Meixner, S. Gillandt, E. Richter, T. B¨orner and A. Weihe (1999) Green fluorescent protein as a marker to investigate targeting of organellar RNA polymerases of higher plants in vivo. The Plant Journal 17, 557–561. S. Heinhorst, C.L. Chi-Ham, S.W. Adamson and G.C. Cannon (2004) The somatic inheritance of plant organelles. In: Molecular Biology and Biotechnology of Plant Organelles (eds H. Daniell and C.D. Chase). Springer, Dordrecht. D. Hellmund, M. Metzlaff and E. Serfling (1984) A transfer RNAArg gene of Pelargonium chloroplasts, but not a 5S RNA gene, is efficiently transcribed after injection into Xenopus oocyte nuclei. Nucleic Acids Research 12, 8253–8268. D.L. Herrin and J. Nickelsen (2004) Chloroplast RNA processing and stability. Photosynthesis Research 82, 301–314. R.G. Herrmann, H.J. Bohnert, A. Driesel and G. Hobom (1976) The location of rRNA genes on the restriction endonuclease map of the Spinacia oleracea chloroplast DNA. In: Genetics and Biogenesis of Chloroplasts and Mitochondria (eds T. Bucher, W. Neupert, W. Sebald and S. Werner). North-Holland, Amsterdam. R.G. Herrmann, R.M. Maier and C. Schmitz-Linneweber (2003) Eukaryotic genome evolution: rearrangement and coevolution of compartmentalized genetic information. Philosophical Transactions of the Royal Society of London Series B: Biological Sciences 358, 87–97; discussion 97. W.R. Hess and T. B¨orner (1999) Organellar RNA polymerases of higher plants. International Review of Cytology 190, 1–59. W.R. Hess, A. Muller, F. Nagy and T. Borner (1994) Ribosome-deficient plastids affect transcription of light-induced nuclear genes: genetic evidence for a plastid-derived signal. Molecular and General Genetics 242, 305–312. W.R. Hess, A. Prombona, B. Fieder, A.R. Subramanian and T. B¨orner (1993) Chloroplast rps15 and the rpoB/C1/C2 gene cluster are strongly transcribed in ribosome-deficient plastids: evidence for a functioning non-chloroplast-encoded RNA polymerase. The EMBO Journal 12, 563–571. W.R. Hess, R. Schendel, W. R¨udiger and T. B¨orner (1992) Protochlorophyllide oxidoreductase and chlorophyll synthetase are present in a barley albina mutant unable to synthesize δ-aminolevulinic acid by utilizing the transfer RNA for glutamic acid. Planta 188, 19–27. N. Hirata, D. Yonekura, S. Yanagisawa and K. Iba (2004) Possible involvement of the 5 -flanking region and the 5 UTR of plastid accD gene in NEP-dependent transcription. Plant and Cell Physiology 45, 176–186. P.H. Hoffer and D.A. Christopher (1997) Structure and blue-light-responsive transcription of a chloroplast psbD promoter from Arabidopsis thaliana. Plant Physiology 115, 213–222. A. Homann and G. Link (2003) DNA-binding and transcription characteristics of three cloned sigma factors from mustard (Sinapis alba L.) suggest overlapping and distinct roles in plastid gene expression. European Journal of Biochemistry 270, 1288–1300. J. Hu and L. Bogorad (1990) Maize chloroplast RNA polymerase: the 180-, 120-, and 38-kilodalton polypeptides are encoded in chloroplast genes. Proceedings of the National Academy of Sciences of the United States of America 87, 1531–1535.
TRANSCRIPTION OF PLASTID GENES
215
J. Hu, R.F. Troxler and L. Bogorad (1991) Maize chloroplast RNA polymerase: the 78-kilodalton polypeptide is encoded by the plastid rpoC1 gene. Nucleic Acids Research 19, 3431–3434. T. H¨ubschmann and T. B¨orner (1998) Characterisation of transcript initiation sites in ribosome-deficient barley plastids. Plant Molecular Biology 36, 493–496. G.S. Hudson, D.A. Holton, P.R. Whitfeld and W. Bottomley (1988) Spinach chloroplast rpoBC genes encode three subunits of the chloroplast RNA polymerase. Proceedings of the National Academy of Sciences 10, 525–558. H. Hupfer, M. Swiatek, S. Hornung, et al. (2000) Complete nucleotide sequence of the Oenothera elata plastid chromosome, representing plastome I of the five distinguishable euoenothera plastomes. Molecular and General Genetics 263, 581–585. K. Ichikawa, M. Sugita, T. Imaizumi, M. Wada and S. Aoki (2004) Differential expression on a daily basis of plastid sigma factor genes from the moss Physcomitrella patens. Regulatory interactions among PpSig5, the circadian clock, and blue light signaling mediated by cryptochromes. Plant Physiology 136, 4285–4298. T.M. Ikeda and M.W. Gray (1999) Identification and characterization of T7/T3 bacteriophage-like RNA polymerase sequences in wheat. Plant Molecular Biology 40, 567–578. S. Imamura, S. Yoshihara, S. Nakano, et al. (2003) Purification, characterization, and gene expression of all sigma factors of RNA polymerase in a cyanobacterium. Journal of Molecular Biology 325, 857–872. H. Inada, M. Seki, H. Morikawa, M. Nishimura and K. Iba (1997) Existence of three regulatory regions each containing a highly conserved motif in the promoter of plastid-encoded RNA polymerase gene (rpoB). The Plant Journal 11, 883–890. R. Iratni, L. Baeza, A. Andreeva, R. Mache and S. Lerbs-Mache (1994) Regulation of rDNA transcripion in chloroplasts: promoter exclusion by constitutive repression. Genes Development 8, 2928–2938. R. Iratni, L. Diederich, H. Harrak, M. Bligny and S. Lerbs-Mache (1997) Organ-specific transcription of the rrn operon in spinach plastids. The Journal of Biological Chemistry 272, 13676–13682. A. Ishihama (2000) Functional modulation of Escherichia coli RNA polymerase. Annual Review of Microbiology 54, 499–518. Y. Ishizaki, Y. Tsunoyama, K. Hatano, et al. (2005) A nuclear-encoded sigma factor, Arabidopsis SIG6, recognizes sigma-70 type chloroplast promoters and regulates early chloroplast development in cotyledons. The Plant Journal 42, 133–144. K. Isono, Y. Niwa, K. Satoh and H. Kobayashi (1997a) Evidence for transcriptional regulation of plastid photosynthesis genes in Arabidopsis thaliana roots. Plant Physiology 114, 623–630. K. Isono, M. Shimizu, K. Yoshimoto, et al. (1997b) Leaf-specifically expressed genes for polypeptides destined for chloroplasts with domains of sigma70 factors of bacterial RNA polymerases in Arabidopsis thaliana. Proceedings of the National Academy of Sciences of the United States of America 94, 14948–14953. J.A. Jaehning (1993) Mitochondrial transcription: is a pattern emerging? Molecular Microbiology 8, 1–4. D. Jahn (1992) Expression of the Chlamydomonas reinhardtii chloroplast tRNA(Glu) gene in a homologous in vitro transcription system is independent of upstream promoter elements. Archives of Biochemistry and Biophysics 298, 505–513. Y. Kabeya, K. Hashimoto and N. Sato (2002) Identification and characterization of two phage-type RNA polymerase cDNAs in the moss Physcomitrella patens: implication of recent evolution of nuclear-encoded RNA polymerase of plastids in plants. Plant and Cell Physiology 43, 245–255. Y. Kabeya and N. Sato (2005) Unique translation initiation at the second AUG codon determines mitochondrial localization of the phage-type RNA polymerases in the moss Physcomitrella patens. Plant Physiology 138, 369–382. K. Kanamaru, M. Fujiwara, M. Seki, et al. (1999) Plastidic RNA polymerase sigma factors in Arabidopsis. Plant and Cell Physiology 40, 832–842. K. Kanamaru, A. Nagashima, M. Fujiwara, et al. (2001) An Arabidopsis sigma factor (SIG2)-dependent expression of plastid-encoded tRNAs in chloroplasts. Plant and Cell Physiology 42, 1034–1043.
216
REGULATION OF TRANSCRIPTION IN PLANTS
T. Kaneko, S. Sato, H. Kotani, et al. (1996) Sequence analysis of the genome of the unicellular cyanobacterium Synechocystis sp. strain PCC6803. II: Sequence determination of the entire genome and assignment of potential protein-coding regions. DNA Research 3, 109–136. S. Kapoor and M. Sugiura (1999) Identification of two essential sequence elements in the nonconsensus Type II PatpB-290 plastid promoter by using plastid transcription extracts from cultured tobacco BY-2 cells. The Plant Cell 11, 1799–1810. S. Kapoor, J.Y. Suzuki and M. Sugiura (1997) Identification and functional significance of a new class of non-consensus-type plastid promoters. The Plant Journal 11, 327–337. S. Kapoor, T. Wakasugi, H. Deno and M. Sugiura (1994) An atpE-specific promoter within the coding region of the atpB gene in tobacco chloroplast DNA. Current Genetics 26, 263–268. K. Kasai, M. Kawagishi-Kobayashi, M. Teraishi, et al. (2004) Differential expression of three plastidial sigma factors, OsSIG1,OsSIG2A, and OsSIG2B, during leaf development in rice. Bioscience, Biotechnology, and Biochemistry 68, 973–977. K. Kasai, S. Usami, T. Yamada, Y. Endo, K. Ochi and Y. Tozawa (2002) A RelA-SpoT homolog (Cr-RSH) identified in Chlamydomonas reinhardtii generates stringent factor in vivo and localizes to chloroplasts in vitro. Nucleic Acids Research 30, 4985–4992. M. Kestermann, S. Neukirchen, K. Kloppstech and G. Link (1998) Sequence and expression characteristics of a nuclear-encoded chloroplast sigma factor from mustard (Sinapis alba). Nucleic Acids Research 26, 2747–2753. J.W. Kim, J.K. Park, B.H. Kim, J.-S. Lee and W.S. Sim (2002) Molecular analysis of the accumulation of the transcripts of the large subunit gene of ribulose-1,5-bisphosphate carboxylase/oxygenase by light. Molecules and Cells 14, 281–287. M. Kim, D.A. Christopher and J.E. Mullet (1999a) ADP-dependent phosphorylation regulates association of a DNA-binding complex with the barley chloroplast psbD blue-light-responsive promoter. Plant Physiology 119, 663–670. M. Kim and J.E. Mullet (1995) Identification of a sequence-specific DNA binding factor required for transcription of the barley chloroplast blue light-responsive psbD-psbC promoter. The Plant Cell 7, 1445–1457. M. Kim, K.E. Thum, D.T. Morishige and J.E. Mullet (1999b) Detailed architecture of the barley chloroplast psbD-psbC blue light-responsive promoter. The Journal of Biological Chemistry 274, 4684–4692. R.R. Klein, H.S. Mason and J.E. Mullet (1988) Light-regulated translation of chloroplast proteins. I: Transcripts of psaA-psaB, psbA, and rbcL are associated with polysomes in dark-grown and illuminated barley seedlings. Journal of Cell Biology 106, 289–301. R.R. Klein and J.E. Mullet (1990) Light-induced transcription of chloroplast genes. psbA transcription is differentially enhanced in illuminated barley. The Journal of Biological Chemistry 265, 1895– 1902. H. Kobayashi, J. Ngernprasirtsiri and T. Akazawa (1990) Transcriptional regulation and DNA methylation in plastids during transitional conversion of chloroplasts to chromoplasts. The EMBO Journal 9, 307–313. Y. Kobayashi, Y. Dokiya, Y. Kumazawa and M. Sugita (2002) Non-AUG translation initiation of mRNA encoding plastid-targeted phage-type RNA polymerase in Nicotiana sylvestris. Biochemical and Biophysical Research Communications 299, 57–61. Y. Kobayashi, Y. Dokiya and M. Sugita (2001a) Dual targeting of phage-type RNA polymerase to both mitochondria and plastids is due to alternative translation initiation in single transcripts. Biochemical and Biophysical Research Communications 289, 1106–1113. Y. Kobayashi, Y. Dokiya, M. Sugiura, Y. Niwa and M. Sugita (2001b) Genomic organization and organ-specific expression of a nuclear gene encoding phage-type RNA polymerase in Nicotiana sylvestris. Gene 279, 33–40. L. Kovacs, W. Wiessner, M. Kis, F. Nagy, D. Mende and S. Demeter (2000) Short- and long-term redox regulation of photosynthetic light energy distribution and photosystem stoichiometry by acetate metabolism in the green alga, Chlamydobotrys stellata. Photosynthesis Research 65, 231–247.
TRANSCRIPTION OF PLASTID GENES
217
K. Krause, S. Berg and K. Krupinska (2003) Plastid transcription in the holoparasitic plant genus Cuscuta: parallel loss of the rrn16 PEP-promoter and of the rpoA and rpoB genes coding for the plastid-encoded RNA polymerase. Planta 216, 815–823. K. Krause, R.M. Maier, W. Kofer, K. Krupinska and R.G. Herrmann (2000) Disruption of plastidencoded RNA polymerase genes in tobacco: expression of only a distinct set of genes is not based on selective transcription of the plastid chromosome. Molecular and General Genetics 263, 1022–1030. D. Kroll, M. Streubel and P. Westhoff (1999) A plastid sigma factor sequence from the C4 monocot Sorghum bicolor. Plant Biology 1, 180–186. C. Kuhlemeier (1992) Transcriptional and post-transcriptional regulation of gene expression in plants. Plant Molecular and Biology 19, 1–14. T. Kuroiwa (1991) The replication, differentiation, and inheritance of plastids with emphasis on the concept of organelle nuclei. International Review of Cytology – A Survey of Cell Biology 128, 1–62. K. Kusumi, A. Yara, N. Mitsui, Y. Tozawa and K. Iba (2004) Characterization of a rice nuclear-encoded plastid RNA polymerase gene OsRpoTp. Plant and Cell Physiology 45, 1194–1201. S.D. Lahiri and L.A. Allison (2000) Complementary expression of two plastid-localized sigma-like factors in maize. Plant Physiology 123, 883–894. S.D. Lahiri, J. Yao, C. Mccumbers and L.A. Allison (1999) Tissue-specific and light-dependent expression within a family of nuclear-encoded sigma-like factors from Zea mays. Molecular Cell Biology Research Communications 1, 14–20. E. Lam, L. Hanley-Bowdoin and N.H. Chua (1988) Characterization of a chloroplast sequence-specific DNA binding factor. The Journal of Biological Chemistry 263, 8288–8293. B.F. Lang, G. Burger, C.J. O’kelly, et al. (1997) An ancestral mitochondrial DNA resembling a eubacterial genome in miniature. Nature 387, 493–497. J. Legen, S. Kemp, K. Krause, B. Profanter, R.G. Herrmann and R.M. Maier (2002) Comparative analysis of plastid transcription profiles of entire plastid chromosomes from tobacco attributed to wild-type and PEP-deficient transcription machineries. The Plant Journal 31, 171–188. D. Leister (2005) Origin, evolution and genetic effects of nuclear insertions of organelle DNA. Trends in Genetics 21, 655–663. P. Leon, A. Arroyo and S. Mackenzie (1998) Nuclear control of plastid and mitochondrial development in higher plants. Annual Review of Plant Physiology and Plant Molecular Biology 49, 453–480. S. Lerbs, J.F. Briat and R. Mache (1983) Chloroplast RNA polymerase from spinach: purification and DNA-binding proteins. Plant Molecular Biology 2, 67–74. S. Lerbs-Mache (1993) The 110-kDa polypeptide of spinach plastid DNA-dependent RNA polymerase: single-subunit enzyme or catalytic core of multimeric enzyme complexes? Proceedings of the National Academy of Sciences of the United States of America 90, 5509–5513. L. Lezhneva and J. Meurer (2004) The nuclear factor HCF145 affects chloroplast psaA-psaB-rps14 transcript abundance in Arabidopsis thaliana. The Plant Journal 38, 740–753. H. Li and L.A. Sherman (2000) A redox-responsive regulator of photosynthesis gene expression in the cyanobacterium Synechocystis sp. strain PCC 6803. Journal of Bacteriology 182, 4268–4277. K. Liere, D. Kaden, P. Maliga and T. B¨orner (2004) Overexpression of phage-type RNA polymerase RpoTp in tobacco demonstrates its role in chloroplast transcription by recognizing a distinct promoter type. Nucleic Acids Research 32, 1159–1165. K. Liere, M. Kestermann, U. M¨uller and G. Link (1995) Identification and characterization of the Arabidopsis thaliana chloroplast DNA region containing the genes psbA, trnH and rps19 . Current Genetics 28, 128–130. K. Liere and G. Link (1994) Structure and expression characteristics of the chloroplast DNA region containing the split gene for tRNA(Gly) (UCC) from mustard (Sinapis alba L.). Current Genetics 26, 557–563. K. Liere and G. Link (1997) Chloroplast endoribonuclease p54 involved in RNA 3 -end processing is regulated by phosphorylation and redox state. Nucleic Acids Research 25, 2403–2408.
218
REGULATION OF TRANSCRIPTION IN PLANTS
K. Liere and P. Maliga (1999a) In vitro characterization of the tobacco rpoB promoter reveals a core sequence motif conserved between phage-type plastid and plant mitochondrial promoters. The EMBO Journal 18, 249–257. K. Liere and P. Maliga (1999b) Novel in vitro transcription assay indicates that the accD NEP promoter is contained in a 19 bp fragment. In: The Chloroplast: From Molecular Biology to Biotechnology (eds J.H. Argyroudi-Akoyunoglou and H. Senger). Kluwer Academic, Amsterdam. K. Liere and P. Maliga (2001) Plastid RNA polymerases. In: Regulation of Photosynthesis (eds B. Andersson and E.-M. Aro). Kluwer Academic, Dordrecht. G. Link (1984) DNA sequence requirements for the accurate transcription of a protein-coding plastid gene in a plastid in vitro transcription system from mustard (Sinapis alba L.). The EMBO Journal 3, 1697–1704. G. Link (1994) Plastid differentiation: organelle promoters and transcription factors. In: Plant Promoters and Transcription factors – Results and Problems in Cell Differentiation (ed. L. Nover). Springer Verlag, Berlin. G. Link (1996) Green life: control of chloroplast gene transcription. BioEssays 18, 465–471. G. Link (2003) Redox regulation of chloroplast transcription. Antioxidants & Redox Signaling 5, 79–87. M.C. Little and R.B. Hallick (1988) Chloroplast rpoA, rpoB, and rpoC genes specify at least three components of a chloroplast DNA-dependent RNA polymerase active in tRNA and mRNA transcription. The Journal of Biological Chemistry 263, 14302–14307. B. Liu and R.F. Troxler (1996) Molecular characterization of a positively photoregulated nuclear gene for a chloroplast RNA polymerase sigma factor in Cyanidium caldarium. Proceedings of the National Academy of Sciences of the United States of America 93, 3313–3318. M. Lonetto, M. Gribskov and C.A. Gross (1992) The sigma 70 family: sequence conservation and evolutionary relationships. Journal of Bacteriology 174, 3843–3849. E. L´opez-Juez and K.A. Pyke (2005) Plastids unleashed: their development and their integration in plant development. The International Journal of Development Biology 49, 557–577. H. Loschelder, A. Homann, K. Ogrzewalla and G. Link (2004) Proteomics-based sequence analysis of plant gene expression – the chloroplast transcription apparatus. Phytochemistry 65, 1785–1793. J.H. Lukens, D.E. Mathews and R.D. Durbin (1987) Effect of tagetitoxin on the levels of ribulose 1,5-bisphosphate carboxylase, ribosomes, and RNA in plastids of wheat leaves. Plant Physiology 84, 808–813. N.A. Lusson, P.M. Delavault and P.A. Thalouarn (1998) The rbcL gene from the non-photosynthetic parasite Lathraea clandestina is not transcribed by a plastid-encoded RNA polymerase. Current Genetics 34, 212–215. R.M. Maier and C. Schmitz-Linneweber (2004) Plastid genomes. In: Molecular Biology and Biotechnology of Plant Organelles (eds H. Daniell and C.D. Chase). Springer, Dordrecht. P. Maliga (1998) Two plastid polymerases of higher plants: an evolving story. Trends in Plant Science 3, 4–6. A. Marchfelder and S. Binder (2004) Plastid and plant mitochondrial RNA processing and RNA stability. In: Molecular Biology and Biotechnology of Plant Organelles (eds H. Daniell and C.D. Chase). Springer, Dordrecht. W. Martin, T. Rujan, E. Richly, et al. (2002) Evolutionary analysis of Arabidopsis, cyanobacterial, and chloroplast genomes reveals plastid phylogeny and thousands of cyanobacterial genes in the nucleus. Proceedings of National Academy of Sciences of the United States of America 99, 12246–12251. B.S. Masters, L.L. Stohl and D.A. Clayton (1987) Yeast mitochondrial RNA polymerase is homologous to those encoded by bacteriophages T3 and T7. Cell 51, 89–99. M. Matsuzaki, O. Misumi, I.T. Shin, et al. (2004) Genome sequence of the ultrasmall unicellular red alga Cyanidioschyzon merolae 10D. Nature 428, 653–657. S.P. Mayfield, C.B. Yohn, A. Cohen and A. Danon (1995) Regulation of chloroplast gene expression. Annual Review of Plant Physiology and Plant Molecular Biology 46, 147–166. V. Mcculloch, B.L. Seidel-Rogol and G.S. Shadel (2002) A human mitochondrial transcription factor is related to RNA adenine methyltransferases and binds S-adenosylmethionine. Molecular and Cellular Biology 22, 1116–1125.
TRANSCRIPTION OF PLASTID GENES
219
B.Y. Meng, T. Wakasugi and M. Sugiura (1991) Two promoters within the psbK-psbI-trnG gene cluster in tobacco chloroplast DNA. Current Genetics 20, 259–264. T. Miyagi, S. Kapoor, M. Sugita and M. Sugiura (1998) Transcript analysis of the tobacco plastid operon rps2/atpI/H /F/A reveals the existence of a non-consensus type II (NCII) promoter upstream of the atpI coding sequence. Molecular and General Genetics 257, 299–307. T. Mochizuki, Y. Onda, E. Fujiwara, M. Wada and Y. Toyoshima (2004) Two independent light signals cooperate in the activation of the plastid psbD blue light-responsive promoter in Arabidopsis. FEBS Letters 571, 26–30. S.G. Moller, Y.S. Kim, T. Kunkel and N.H. Chua (2003) PP7 is a positive regulator of blue light signaling in Arabidopsis. The Plant Cell 15, 1111–1119. R.A. Monde, G. Schuster and D.B. Stern (2000) Processing and degradation of chloroplast mRNA. Biochimie 82, 573–582. K. Morikawa, S. Ito, Y. Tsunoyama, Y. Nakahira, T. Shiina and Y. Toyoshima (1999) Circadianregulated expression of a nuclear encoded plastid sigma factor gene (sigA) in wheat seedlings. FEBS Letters 451, 275–278. K. Morikawa, T. Shiina, S. Murakami and Y. Toyoshima (2002) Novel nuclear-encoded proteins interacting with a plastid sigma factor, Sig1, in Arabidopsis thaliana. FEBS Letters 514, 300– 304. J.E. Mullet (1993) Dynamic regulation of chloroplast transcription. Plant Physiology 103, 309–313. J.E. Mullet, E.M. Orozco and N.-H. Chua, (1985) Multiple transcripts for higher plant rbcL and atpB genes and localization of the transcription initiation sites of the rbcL gene. Plant Molecular Biology 4, 39–54. A. Nagashima, M. Hanaoka, R. Motohashi, et al. (2004a) DNA microarray analysis of plastid gene expression in an Arabidopsis mutant deficient in a plastid transcription factor sigma, SIG2. Bioscience, Biotechnology, and Biochemistry 68, 694–704. A. Nagashima, M. Hanaoka, T. Shikanai, et al. (2004b) The multiple-stress responsive plastid sigma factor, SIG5, directs activation of the psbD blue light-responsive promoter (BLRP) in Arabidopsis thaliana. Plant and Cell Physiology 45, 357–368. Y. Nakahira, K. Baba, A. Yoneda, T. Shiina and Y. Toyoshima (1998) Circadian-regulated transcription of the psbD light-responsive promoter in wheat chloroplasts. Plant Physiology 118, 1079–1088. T. Nakamura, Y. Furuhashi, K. Hasegawa, et al. (2003) Array-based analysis on tobacco plastid transcripts: preparation of a genomic microarray containing all genes and all intergenic regions. Plant and Cell Physiology 44, 861–867. H. Neuhaus and G. Link (1990) The chloroplast psbK operon from mustard (Sinapis alba L.): multiple transcripts during seedling development and evidence for divergent overlapping transcription. Current Genetics 18, 377–383. J. Ngernprasirtsiri and T. Akazawa (1990) Modulation of DNA methylation and gene expression in cultured sycamore cells treated by hypomethylating base analog. European Journal of Biochemistry 194, 513–520. J. Ngernprasirtsiri, R. Chollet, H. Kobayashi, T. Sugiyama and T. Akazawa (1989) DNA methylation and the differential expression of C4 photosynthesis genes in mesophyll and bundle sheath cells of greening maize leaves. The Journal of Biological Chemistry 264, 8241–8248. J. Nickelsen and G. Link (1990) Nucleotide sequence of the mustard chloroplast genes trnH and rps19 . Nucleic Acids Research 18, 1051. J. Nickelsen and G. Link (1991) RNA–protein interactions at transcript 3 ends and evidence for trnK-psbA cotranscription in mustard chloroplasts. Molecular and General Genetics 228, 89–96. R. Oelm¨uller, I. Levitan, R. Bergfeld, V.K. Rajasekhar and H. Mohr (1986) Expression of nuclear genes as affected by treatments acting on the plastids. Planta 168, 482–492. K. Ogrzewalla, M. Piotrowski, S. Reinbothe and G. Link (2002) The plastid transcription kinase from mustard (Sinapis alba L.). A nuclear-encoded CK2-type chloroplast enzyme with redox-sensitive function. European Journal of Biochemistry 269, 3329–3337. K. Ohyama, H. Fukuzawa, T. Kohchi, et al. (1986) Chloroplast gene organization deduced from complete sequence of liverwort Marchantia polymorpha chloroplast DNA. Nature 322, 572– 574.
220
REGULATION OF TRANSCRIPTION IN PLANTS
K. Oikawa, M. Fujiwara, E. Nakazato, K. Tanaka and H. Takahashi (2000) Characterization of two plastid sigma factors, SigA1 and SigA2, that mainly function in matured chloroplasts in Nicotiana tabacum. Gene 261, 221–228. M.S. Paget and J.D. Helmann (2003) The sigma70 family of sigma factors. Genome Biology 4, 203.1– 203.6. C.K. Pearson, S.B. Wilson, R. Schaffer and A.W. Ross (1993) NAD turnover and utilisation of metabolites for RNA synthesis in a reaction sensing the redox state of the cytochrome b6f complex in isolated chloroplasts. European Journal of Biochemistry 218, 397–404. J. Pfalz, K. Liere, A. Kandlbinder, K.-J. Dietz and R. Oelm¨uller (2006) pTAC2, -6 and -12 are components of the transcriptionally active plastid chromosome that are required for plastid gene expression. The Plant Cell 18, 176–197. T. Pfannschmidt and K. Liere (2005) Redox regulation and modification of proteins controlling chloroplast gene expression. Antioxidants & Redox Signaling 7, 607–618. T. Pfannschmidt and G. Link (1994) Separation of two classes of plastid DNA-dependent RNA polymerases that are differentially expressed in mustard (Sinapis alba L.) seedlings. Plant Molecular Biology 25, 69–81. T. Pfannschmidt and G. Link (1997) The A and B forms of plastid DNA-dependent RNA polymerase from mustard (Sinapis alba L.) transcribe the same genes in a different developmental context. Molecular and General Genetics 257, 35–44. T. Pfannschmidt, A. Nilsson and J.F. Allen (1999a) Photosynthetic control of chloroplast gene expression. Nature 397, 625–628. T. Pfannschmidt, A. Nilsson, A. Tullberg, G. Link and J.F. Allen (1999b) Direct transcriptional control of the chloroplast genes psbA and psaAB adjusts photosynthesis to light energy distribution in plants. IUBMB Life 48, 271–276. T. Pfannschmidt, K. Ogrzewalla, S. Baginsky, A. Sickmann, H.E. Meyer and G. Link (2000) The multisubunit chloroplast RNA polymerase A from mustard (Sinapis alba L.): integration of a prokaryotic core into a larger complex with organelle-specific functions. European Journal of Biochemistry 267, 253–261. I. Privat, M.A. Hakimi, L. Buhot, J.J. Favory and S. Mache-Lerbs (2003) Characterization of Arabidopsis plastid sigma-like transcription factors SIG1, SIG2 and SIG3. Plant Molecular Biology 51, 385–399. S. Pursiheimo, P. Mulo, E. Rintamaki and E.M. Aro (2001) Coregulation of light-harvesting complex II phosphorylation and lhcb mRNA accumulation in winter rye. The Plant Journal 26, 317–327. S. Purton and J.C. Gray (1989) The plastid rpoA gene encoding a protein homologous to the bacterial RNA polymerase alpha subunit is expressed in pea chloroplasts. Molecular and General Genetics 217, 77–84. A. Rantanen, M. Gaspari, M. Falkenberg, C.M. Gustafsson and N.-G. Larsson (2003) Characterization of the mouse genes for mitochondrial transcription factors B1 and B2. Mammalian Genome 14, 1–6. J.C. Rapp, B.J. Baumgartner and J. Mullet (1992) Quantitative analysis of transcription and RNA levels of 15 barley chloroplast genes. Transcription rates and mRNA levels vary over 300fold; predicted mRNA stabilities vary 30-fold. The Journal of Biological Chemistry 267, 21404–21411. S. Reinbothe, C. Reinbothe, C. Heintzen, C. Seidenbecher and B. Parthier (1993) A methyl jasmonateinduced shift in the length of the 5 untranslated region impairs translation of the plastid rbcL transcript in barley. The EMBO Journal 12, 1505–1512. W. Reznikov, D.A. Siegle, D.W. Cowing and C.A. Gross (1985) The regulation of transcription initiation in bacteria. Annual Review of Genetics 19, 355–387. U. Richter, J. Kiessling, B. Hedtke, et al. (2002) Two RpoT genes of Physcomitrella patens encode phage-type RNA polymerases with dual targeting to mitochondria and plastids. Gene 290, 95–105. S. Rodermel (2001) Pathways of plastid-to-nucleus signaling. Trends in Plant Science 6, 471–478. W. Ross, K.K. Gosink, J. Salomon, et al. (1993) A third recognition element in bacterial promoters: DNA binding by the alpha subunit of RNA polymerase. Science 262, 1407–1413.
TRANSCRIPTION OF PLASTID GENES
221
M.L. Salvador and U. Klein (1999) The redox state regulates RNA degradation in the chloroplast of Chlamydomonas reinhardtii. Plant Physiology 121, 1367–1374. N. Sato, K. Terasawa, K. Miyajima and Y. Kabeya (2003) Organization, developmental dynamics, and evolution of plastid nucleoids. International Review of Cytology 232, 217–262. J. Satoh, K. Baba, Y. Nakahira, T. Shiina and Y. Toyoshima (1997) Characterization of dynamics of the psbD light-induced transcription in mature wheat chloroplasts. Plant Molecular Biology 33, 267–278. J. Satoh, K. Baba, Y. Nakahira, Y. Tsunoyama, T. Shiina and Y. Toyoshima (1999) Devolpmental stagespecific multi-subunit plastid RNA polymerases (PEP) in wheat. The Plant Journal 18, 407–416. A.H. Schinkel, M.J. Groot Koerkamp and H.F. Tabak (1988) Mitochondrial RNA polymerase of Saccharomyces cerevisiae: composition and mechanism of promoter recognition. The EMBO Journal 7, 3255–3262. A. Sch¨on, G. Krupp, S. Gough, S. Berry-Lowe, C.G. Kannangara and D. S¨oll (1986) The RNA required in the first step of chlorophyll biosynthesis is a chloroplast glutamate tRNA. Nature 322, 281–284. H. Schrubar, G. Wanner and P. Westhoff (1990) Transcriptional control of plastid gene expression in greening sorghum seedlings. Planta 183, 101–111. B.L. Seidel-Rogol, V. Mcculloch and G.S. Shadel (2003) Human mitochondrial transcription factor B1 methylates ribosomal RNA at a conserved stem-loop. Nature Genetics 33, 23–24. K. Sekine, T. Hase and N. Sato (2002) Reversible DNA compaction by sulfite reductase regulates transcriptional activity of chloroplast nucleoids. The Journal of Biological Chemistry 277, 24399–24404. G. Serino and P. Maliga (1998) RNA polymerase subunits encoded by the plastid rpo genes are not shared with the nucleus-encoded plastid enzyme. Plant Physiology 117, 1165–1170. T.B. Sexton, D.A. Christopher and J.E. Mullet (1990) Light-induced switch in barley psbD-psbC promoter utilization: a novel mechanism regulating chloroplast gene expression. The EMBO Journal 9, 4485–4494. G.S. Shadel and D.A. Clayton (1993) Mitochondrial transcription. The Journal of Biological Chemistry 268, 16083–16086. T. Shiina, L. Allison and P. Maliga (1998) rbcL transcript levels in tobacco plastids are independent of light: reduced dark transcription rate is compensated by increased mRNA stability. The Plant Cell 10, 1713–1722. T. Shiina, Y. Tsunoyama, Y. Nakahira and M.S. Khan (2005) Plastid RNA polymerases, promoters, and transcription regulators in higher plants. International Review of Cytology 244, 1–68. K. Shinozaki, M. Ohme, M. Tanaka, et al. (1986) The complete sequence of the tobacco chloroplast genome: its gene organization and expression. The EMBO Journal 5, 2043–2049. K. Shinozaki and M. Sugiura (1982) The nucleotide sequence of the tobacco chloroplast gene for the large subunit of ribulose-1,5-bisphosphate carboxylase/oxygenase. Gene 20, 91–102. Y. Shirano, H. Shimada, K. Kanamaru, et al. (2000) Chloroplast development in Arabidopsis thaliana requires the nuclear-encoded transcription factor Sigma B. FEBS Letters 485, 178–182. A. Siemenroth, R. Wollgiehn, D. Neumann and T. B¨orner (1981) Synthesis of ribosomal RNA in ribosome-deficient plastids of the mutant ‘albostrians’ of Hordeum vulgare L. Planta 153, 547–555. D. Silhavy and P. Maliga (1998a) Mapping of the promoters for the nucleus-encoded plastid RNA polymerase (NEP) in the iojap maize mutant. Current Genetics 33, 340–344. D. Silhavy and P. Maliga (1998b) Plastid promoter utilization in a rice embryonic cell culture. Current Genetics 34, 67–70. P. Sriraman, D. Silhavy and P. Maliga (1998a) The phage-type PclpP-53 plastid promoter comprises sequences downstream of the transcription initiation site. Nucleic Acids Research 26, 4874–4879. P. Sriraman, D. Silhavy and P. Maliga (1998b) Transcription from heterologous rRNA operon promoters in chloroplasts reveals requirement for specific activating factors. Plant Physiology 117, 1495–1499. T.A. Steitz (2004) The structural basis of the transition from initiation to elongation phases of transcription, as well as translocation and strand separation, by T7 RNA polymerase. Current Opinion in Structure Biology 14, 4–9.
222
REGULATION OF TRANSCRIPTION IN PLANTS
D.B. Stern, D.C. Higgs and J.J. Yang (1997) Transcription and translation in chloroplasts. Trends in Plant Science 2, 308–315. S.M. Stirdivant, L.D. Crossland and L. Bogorad (1985) DNA supercoiling affects in vitro transcription of two maize chloroplast genes differently. Proceedings of the National Academy of Sciences of the United States of America 82, 4886–4890. G. Strittmatter, A. Godzicka-Josefiak and H. K¨ossel (1985) Identification of an rRNA operon promoter from Zea mays chloroplast which excludes the proximal tRNAVal from the primary transcript. The EMBO Journal 4, 599–604. R. Suck, P. Zeltz, J. Falk, A. Acker, H. K¨ossel and K. Krupinska (1996) Transcriptionally active chromosomes (TACs) of barley chloroplasts contain the α-subunit of plastome encoded RNA polymerase. Current Genetics 30, 515–521. M. Sugita and M. Sugiura (1996) Regulation of gene expression in chloroplasts of higher plants. Plant Molecular Biology 32, 315–326. M. Sugiura (1995) The chloroplast genome. Essays in Biochemistry 30, 49–57. E. Sun, B.W. Wu and K.K. Tewari (1989) In vitro analysis of the pea chloroplast 16S rRNA gene promoter. Molecular and Cellular Biology 9, 5650–5659. S.J. Surzycki and D.L. Shellenbarger (1976) Purification and characterization of a putative sigma factor from Chlamydomonas reinhardtii. Proceedings of the National Academy of Sciences of the United States of America 73, 3961–3965. J.Y. Suzuki, P. Sriraman, Z. Svab and P. Maliga (2003) Unique architecture of the plastid ribosomal RNA operon promoter recognized by the multisubunit RNA polymerase in tobacco and other higher plants. The Plant Cell 15, 195–205. K. Takahashi, K. Kasai and K. Ochi (2004) Identification of the bacterial alarmone guanosine 5 -diphosphate 3 -diphosphate (ppGpp) in plants. Proceedings of the National Academy of Sciences of the United States of America 101, 4320–4324. S. Tan and R.F. Troxler (1999) Characterization of two chloroplast RNA polymerase sigma factors from Zea mays: photoregulation and differential expression. Proceedings of the National Academy of Sciences of the United States of America 96, 5316–5321. K. Tanaka, K. Oikawa, N. Ohta, H. Kuroiwa, T. Kuroiwa and H. Takahashi (1996) Nuclear encoding of a chloroplast RNA polymerase sigma subunit in a red alga. Science 272, 1932– 1935. K. Tanaka, Y. Tozawa, N. Mochizuki, et al. (1997) Characterization of three cDNA species encoding plastid RNA polymerase sigma factors in Arabidopsis thaliana: evidence for the sigma factor heterogeneity in higher plant plastids. FEBS Letters 413, 309–313. K.E. Thum, M. Kim, D.T. Morishige, C. Eibl, H.-U. Koop and J.E. Mullet (2001) Analysis of barley chloroplast psbD light-responsive promoter elements in transplastomic tobacco. Plant Molecular Biology 47, 353–366. K. Tiller and G. Link (1993a) Phosphorylation and dephosphorylation affect functional characteristics of chloroplast and etioplast transcription systems from mustard (Sinapis alba L.). The EMBO Journal 12, 1745–1753. K. Tiller and G. Link (1993b) Sigma-like transcription factors from mustard (Sinapis alba L.) etioplast are similar in size to, but functionally distinct from, their chloroplast counterparts. Plant Molecular Biology 21, 503–513. J.N. Timmis, M.A. Ayliffe, C.Y. Huang and W. Martin (2004) Endosymbiotic gene transfer: organelle genomes forge eukaryotic chromosomes. Nature Reviews Genetics 5, 123–135. K.Y. To, M.C. Cheng, D.F. Suen, D.P. Mon, L.F.O. Chen and S.C.G. Chen (1996) Characterization of the light-responsive promoter of rice chloroplast psbD-C operon and the sequence-specific DNA binding factor. Plant and Cell Physiology 37, 660–666. I.I. Toulokhonov, I. Shulgina and V.J. Hernandez (2001) Binding of the transcription effector ppGpp to Escherichia coli RNA polymerase is allosteric, modular, and occurs near the N terminus of the beta -subunit. The Journal of Biological Chemistry 276, 1220–1225. Y. Tozawa, K. Tanaka, H. Takahashi and K. Wakasa (1998) Nuclear encoding of a plastid sigma factor in rice and its tissue- and light-dependent expression. Nucleic Acids Research 26, 415– 419.
TRANSCRIPTION OF PLASTID GENES
223
R.L. Tracy and D.B. Stern (1995) Mitochondrial transcription initiation: promoter structures and RNA polymerases. Current Genetics 28, 205–216. T. Trebitsh, A. Levitan, A. Sofer and A. Danon (2000) Translation of chloroplast psbA mRNA is modulated in the light by counteracting oxidizing and reducing activities. Molecular and Cellular Biology 20, 1116–1123. Y. Trifa, I. Privat, J. Gagnon, L. Baeza and S. Lerbs-Mache (1998) The nuclear RPL4 gene encodes a chloroplast protein that co-purifies with the T7-like transcription complex as well as plastid ribosomes. The Journal of Biological Chemistry 273, 3980–3985. R.F. Troxler, F. Zhang, J. Hu and L. Bogorad (1994) Evidence that sigma factors are components of chloroplast RNA polymerase. Plant Physiology 104, 753–759. Y. Tsunoyama, Y. Ishizaki, K. Morikawa, et al. (2004) Blue light-induced transcription of plastidencoded psbD gene is mediated by a nuclear-encoded transcription initiation factor, AtSig5. Proceedings of the National Academy of Sciences of the United States of America 101, 3304–3309. Y. Tsunoyama, K. Morikawa, T. Shiina and Y. Toyoshima (2002) Blue light specific and differential expression of a plastid sigma factor, Sig5 in Arabidopsis thaliana. FEBS Letters 516, 225–228. A. Tullberg, K. Alexciev, T. Pfannschmidt and J.F. Allen (2000) Photosynthetic electron flow regulates transcription of the psaB gene in pea (Pisum sativum L.) chloroplasts through the redox state of the plastoquinone pool. Plant and Cell Physiology 41, 1045–1054. E.A. van der Biezen, J. Sun, M.J. Coleman, M.J. Bibb and J.D. Jones (2000) Arabidopsis RelA/SpoT homologs implicate (p)ppGpp in plant signaling. Proceedings of the National Academy of Sciences of the United States of America 97, 3747–3752. A.V. Vener, P.J. van Kan, P.R. Rich, I.I. Ohad and B. Andersson (1997) Plastoquinol at the quinol oxidation site of reduced cytochrome bf mediates signal transduction between light and protein phosphorylation: thylakoid protein kinase deactivation by a single-turnover flash. Proceedings of the National Academy of Sciences of the United States of America 94, 1585–1590. A. Vera, T. Hirose and M. Sugiura (1996) A ribosomal protein gene (rpl32) from tobacco chloroplast DNA is transcribed from alternative promoters – similarities in promoter region organization in plastid housekeeping genes. Molecular and General Genetics 251, 518–525. A. Vera and M. Sugiura (1995) Chloroplast rRNA transcription from structurally different tandem promoters: an additional novel-type promoter. Current Genetics 27, 280–284. T. Wada, Y. Tunoyama, T. Shiina and Y. Toyoshima (1994) In vitro analysis of light-induced transcription in the wheat psbD/C gene cluster using plastid extracts from dark-grown and short-term-illuminated seedlings. Plant Physiology 104, 1259–1267. G. Walter, A. M¨uller, P. Hoffmann and T. B¨orner (1995) Tetrapyrrole synthesis in leaves and roots of a barley albina mutant with extremely low amount of plastid glutamyl-tRNA. In: Photosynthesis: From Light to Biosphere (ed. P. Mathis). Kluwer Academic, Dordrecht/Boston/London. A. Weihe (2004) The transcription of plant organelle genomes. In: Molecular Biology and Biotechnology of Plant Organelles (eds H. Daniell and C.D. Chase). Kluwer Academic, Dordrecht. A. Weihe and T. B¨orner (1999) Transcription and the architecture of promoters in chloroplasts. Trends in Plant Science 4, 169–170. A. Weihe, B. Hedtke and T. B¨orner (1997) Cloning and characterization of a cDNA encoding a bacteriophage-type RNA polymerase from the higher plant Chenopodium album. Nucleic Acids Research 25, 2319–2325. M.M. W¨osten (1998) Eubacterial sigma-factors. FEMS Microbiology Reviews 22, 127–150. C.Y. Wu, C.H. Lin and L.J. Chen (1997) Identification of the transcription start site for the spinach chloroplast serine tRNA gene. FEBS Letters 418, 157–161. G. Xie and L.A. Allison (2002) Sequences upstream of the YRTA core region are essential for transcription of the tobacco atpB NEP promoter in chloroplasts in vivo. Current Genetics 41, 176–182. W.B. Yao, B.Y. Meng, M. Tanaka and M. Sugiura (1989) An additional promoter within the protein-coding region of the psbD-psbC gene cluster in tobacco chloroplast DNA. Nucleic Acids Research 17, 9583–9591. J. Yao, S. Roy-Chowdhury and L.A. Allison (2003) AtSig5 is an essential nucleus-encoded Arabidopsis σ -like factor. Plant Physiology 132, 739–747.
224
REGULATION OF TRANSCRIPTION IN PLANTS
J.M. Zapata, A. Guera, A. Esteban-Carrasco, M. Martin and B. Sabater (2005) Chloroplasts regulate leaf senescence: delayed senescence in transgenic ndhF-defective tobacco. Cell Death Differential 12, 1277–1284. J.M. Zengel, D. Mueckl and L. Lindahl (1980) Protein L4 of the E. coli ribosome regulates an eleven gene r protein operon. Cell 21, 523–535. H. Zer and I. Ohad (2003) Light, redox state, thylakoid-protein phosphorylation and signaling gene expression. Trends in Biochemical Sciences 28, 467–470. W. Zerges (2004) Regulation of translation in chloroplasts. In: Molecular Biology and Biotechnology of Plant Organelles (eds H. Daniell and C.D. Chase). Springer, Dordrecht. L. Zhang, V. Paakkarinen, K.J. van Wijk and E.M. Aro (2000) Biogenesis of the chloroplast-encoded D1 protein: regulation of translation elongation, insertion, and assembly into photosystem II. The Plant Cell 12, 1769–1782. F. Zito, G. Finazzi, R. Delosme, W. Nitschke, D. Picot and F.A. Wollman (1999) The Qo site of cytochrome b6 f complexes controls the activation of the LHCII kinase. The EMBO Journal 18, 2961–2969. M.K. Zubko and A. Day (1998) Stable albinism induced without mutagenesis: a model for ribosome free plastid inheritance. The Plant Journal 15, 265–271.
9 Control of flowering time Steven van Nocker and Maria Julissa Ek-Ramos
9.1
Introduction
The developmental switch from vegetative to reproductive growth – also known as flowering – is a critical transition in the life cycle of flowering plants. This transition is developmentally delayed until the plant has reached a stage of growth sufficient to support fruit and seed production. Superimposed on developmental regulation is the ability to delay or initiate flowering in response to environmental cues, such as photoperiod and temperature, in order to take full advantage of a seasonal climate optimal for reproduction. The switch to flowering involves the activation of a select set of key regulatory genes that initiate extensive changes in gene activity in the shoot apical meristem (Figure 9.1). Cumulative research has revealed that the molecular mechanisms of flowering are vastly complex and touch on most of the known eukaryotic mechanism of transcriptional regulation. This review will focus on transcriptional aspects of the individual regulatory pathways that have been defined to regulate flowering in the reference plant Arabidopsis thaliana, as well as how these pathways are integrated to activate genes that force inflorescence fate on the shoot apical meristem.
9.2
Regulation of FLC expression through the ‘autonomous pathway’
Although plant MADS-box genes are probably best recognized for their role in establishing identity of floral organs, a small but growing subset of genes are now known to play central roles in determining the timing of flowering. Among these is FLC, which acts as a strong suppressor of flowering. In rapid cycling accessions of Arabidopsis such as Columbia (Col) or Landsberg erecta (Ler), early flowering depends on the activity of a mechanism(s) called the autonomous pathway. This pathway involves a class of genes that act in concert to silence FLC. The identities of two of these, FVE and FLD, suggest that the encoded proteins participate in FLC silencing through direct modification of chromatin. FVE is one of five Arabidopsis genes encoding small WD-repeat proteins homologous to yeast MSI1, mammalian RbAp46 and RbAp48, and fly p55/NuRF-55 (Ausin et al., 2004). These proteins interact with histones and are components of the chromatin assembly factor CAF-1, which deposits histones onto newly replicated DNA (review in van Nocker, 2003). In higher eukaryotes, MSI1-like proteins are also found in various transcriptional corepressor complexes (e.g., SIN3, Mi2/NuRD, and ESC–E(z)) that also may contain histone deacetylases (HDAs) such as RPD3 or HDAC1 (Zhang
226
REGULATION OF TRANSCRIPTION IN PLANTS
Figure 9.1 Regulatory map of flowering in Arabidopsis. Pathways by which photoperiod, cold, and GAs influence flowering are shown. Genes acting in common mechanisms are grouped. FLC-like genes, SOC1, and FT/TSF act as regulatory hubs, culminating in activation of the meristem identity genes AP1 and CAL.
et al., 1997; 1998; Tie et al., 2001; Kuzmichev et al., 2002). It has been suggested that MSI1-like proteins mediate interactions between nascent chromatin and transcriptional repressor complexes, potentially to help establish repressive marks, including deacetylation of histones H3 and H4, on new chromatin. Consistent with this, loss of FLC silencing in an fve mutant is associated with hyperacetylation of histone H3 and H4 within FLC chromatin (Ausin et al., 2004). What might be the enzymatic activity responsible for carrying out deacetylation of FLC chromatin in wild-type plants? The Arabidopsis genome encodes for at least 16 canonical HDAs (Pandey et al., 2002). Although most have not been characterized, genetic disruption of at least the RPD3/HDAC1 class HDAs, HDA6 and HDA19 (also called RPD3a or HD1), confers delayed flowering (Wu et al., 2000; Tian and Chen, 2001; Tian et al., 2003; Probst et al., 2004). Thus, HDA6 and HDA19 might participate cooperatively in repressing FLC, although this has yet to be demonstrated. FLD has an intriguing story. This protein is one of four relatively uncharacterized Arabidopsis homologs of human lysine-specific demethylase 1 (LSD1), a nuclear, FAD-dependent amine oxidase that targets histone lysines for demethylation (Shi et al., 2004). FLD and its Arabidopsis siblings show a domain structure conserved with LSD1, and, in addition to its FAD-binding and amino oxidase domains, contain a SWIRM (SWI3, RSC8, MOIRA) domain considered to mediate protein– protein interactions.
CONTROL OF FLOWERING TIME
227
LSD1 is found in HDA-associated corepressor complexes in mammalian cells (Tong et al., 1998; You et al., 2001; Shi et al., 2003), and can act as a transcriptional repressor when tethered to an artificial promoter in vivo (Shi et al., 2004). The repressive function of LSD1 is linked to its ability to demethylate lysine 4 of histone H3 (H3K4). Lysine residues can be mono-, di-, or trimethylated, and in yeast and higher eukaryotes where studied, trimethylation of H3K4 is prevalent on nucleosomes in actively transcribed genes, especially at promoters and 5 regions (Santos-Rosa et al., 2002; Guenther et al., 2005). This modification becomes enriched in these regions through transcription-associated replacement of histone H3 with H3.3 (Mito et al., 2005) and the activity of specific histone methyltransferases (HMTs) such as Set1/COMPASS (budding yeast), Trx (fly), or MLL (human), and is thought to promote continual transcriptional activity (Ringrose and Paro, 2004; Kouskouti and Talianidis, 2005). The mechanism of amine oxidation of methylated lysines used by LSD1 (and probably related proteins as well) requires a protonated nitrogen, and so this mechanism is compatible with demethylation only of mono- or dimethylated lysines (Shi et al., 2004). Theoretically, then, LSD1 should be incapable of silencing an actively transcribed gene. For FLD, this would be consistent with an apparent role in maintaining, but not initiating, silencing. One of many possible functions for FLD on FLC chromatin is the demethylation of mono- or dimethylated H3K4 spontaneously arising through low-level ‘background’ transcription, constitutive HMT activity, or replication-associated incorporation of methylated H3 variants. An interesting observation is that like FVE, FLD appears to be required for hypoacetylation of histone H4 in FLC chromatin (He et al., 2003). A likely explanation is that loss of either FLD or FVE is sufficient to disrupt the activity of a HDA protein complex in which both proteins participate. Deletion of a limited region within the first intron of FLC proximal to the promoter results in loss of silencing and hyperacetylation similar to that seen in fve and fld mutants, suggesting that this region could be targeted by either or both proteins (He et al., 2003). Further characterization of FLD and other LSD1-like proteins in Arabidopsis is eagerly awaited and should provide unique insights into the autonomous mechanisms. An additional chromatin-related protein that probably participates in the autonomous mechanisms is REF6 (Noh et al., 2004), one of a small group of Arabidopsis proteins containing jumonji (jmjN and jmjC) domains. These domains were first identified in an interesting mouse protein, JUMONJI (JMJ), and related proteins (Balciunas and Ronne, 2000). A mutation in JMJ leads to severe developmental pleiotropy associated with abnormal stem cell proliferation (Takeuchi et al., 1995; Motoyama et al., 1997; Lee et al., 2000a), and JMJ has been shown to interact with E2F-regulated promoter sequences and repress transcription in collaboration with the retinoblastoma (Rb) protein, implicating JMJ in chromatin-based repression (Jung et al., 2005). jmjN/jmjC proteins also typically contain any of an assortment of domains associated with transcription and chromatin (Balciunas and Ronne, 2000). The jmjN domain in JUMONJI is required for DNA-binding and transcriptional repressor activity (Kim et al., 2003). Based on predicted structural similarity with a class of zinc-binding metalloenzyme domains known as cupins,
228
REGULATION OF TRANSCRIPTION IN PLANTS
Clissold and Ponting (2001) proposed that the jmjC domain could function as a metalloenzyme, and, further, that jmjC proteins also containing the jmjN domain and/or other chromatin-associated domains might participate in transcriptional repression through modification of chromatin or associated proteins, potentially through protein hydroxylation. It has recently been postulated that a jmjC protein from fission yeast, Epe1, which is required for confining heterochromatin at silenced domains (Ayoub et al., 2003), might function as a histone demethylase through oxidative demethylation of methylated lysine residues (Trewick et al., 2005). Similarly, it is conceivable that REF6 targets H3K4, including its trimethylated form, in FLC chromatin for demethylation. Features of four other autonomous pathway proteins suggest a role for premRNA processing in FLC expression. FCA and FPA encode proteins containing potential RNA-binding RRM domains, and the FLK protein contains K-homology putative RNA-binding domains (Macknight et al., 1997; Schomburg et al., 2001; Lim et al., 2004; Mockler et al., 2004). FY encodes a WD-repeat protein orthologous to budding yeast Pfs2, a component of the pre-mRNA cleavage and polyadenylation factor (CPF) protein complex required for 3 -end processing of pre-mRNAs (Simpson et al., 2003). FCA negatively regulates its own expression by promoting selection of a polyadenylation site near the 5 end of the FCA transcriptional unit, resulting in a truncated and dysfunctional transcript (Quesada et al., 2003). FY interacts with FCA in vitro, and this interaction is thought to promote the accumulation of the truncated FCA RNA in vivo and also to be required for repression of FLC (Simpson et al., 2003). Because regulation of FLC by FCA, FY, FLK, and FPA probably involves mainly post-transcriptional events, and because the potential mechanisms of programmed RNA processing involving these genes have been discussed at length (Amasino, 2003; Simpson et al., 2004; Quesada et al., 2005), it will not be covered further here. However, it should be mentioned that although an obvious target of post-transcriptional regulation is FLC, there is currently no evidence to support this, and thus regulation of FLC by these genes may be indirect. Another autonomous pathway gene, LD, encodes a nuclear protein containing a diverged homeodomain and a carboxyl-terminal region enriched in glutamine residues (Lee et al., 1994). Although these features suggest that LD could act as a transcriptional activator, potential targets have not yet been identified. Although originally considered to be devoted to nonenvironmental inputs and the regulation of FLC, it is now becoming clear that the ‘autonomous pathway’ moderates the effects of ambient temperatures on flowering. Like many plants, most Arabidopsis accessions flower appreciably later (i.e., producing more vegetative nodes) when grown at lowered ambient temperatures, such as 16◦ C. The earlier flowering associated with increasing growth temperatures was found to be mediated by FVE and FCA, but surprisingly, was not associated with decreased FLC RNA abundance (Blazquez et al., 2003). Thus, FVE and FCA (and not unlikely, other autonomous pathway components as well) have a role in thermoregulation of flowering time that is distinct from their repression of FLC. An additional indication that the autonomous mechanisms have more expansive functions was revealed by the observation that FY is an essential gene in Arabidopsis, with hypomorphic fy alleles
CONTROL OF FLOWERING TIME
229
affecting FLC expression and flowering time relatively specifically (Henderson et al., 2005). FLC is subject to positive regulation as well as negative, and several floweringtime genes are known that act as strong activators of FLC. Probably the best known of these is FRIGIDA (FRI). FRI has received much attention, because allelic variation at FRI is a major determinant of the flowering habit (i.e., annual vs. winter annual) among natural Arabidopsis accessions (Lee et al., 1993; Johanson et al., 2000). Most of the rapid cycling accessions of Arabidopsis used in laboratory work, such as Col or Ler, do not express FLC to appreciable levels because they carry loss-of-function mutations in FRI (Johanson et al., 2000). The predicted FRI protein is not conserved through evolution and does not exhibit strong homology with any other protein of known function, but exhibits coiled-coil domains, suggesting that it interacts with protein partner(s) (Johanson et al., 2000). FRI is a member of a small gene family in Arabidopsis, and at least two others in this family have been implicated in flowering repression through upregulation of FLC (Michaels et al., 2004). The late flowering conferred by FRI in wild-type plants reveals that activation of FLC by FRI is epistatic to the suppression of FLC by autonomous pathway genes; in the ‘standard’, rapid cycling Col-0 ecotype of Arabidopsis lacking FRI activity, loss of autonomous pathway gene function restores FLC expression and late flowering (Michaels and Amasino, 1999, 2001). This is consistent with a mechanism whereby FRI limits the activity of autonomous mechanisms, possibly through the negative regulation of one or more components.
9.3
Chromatin-related pleiotropic regulators of FLC
In addition to FRI, an emerging number of activators of FLC are becoming characterized, and most appear to be related in some way to chromatin dynamics. Unlike FRI, in most cases reported, these act pleiotropically to affect other aspects of growth and development, suggesting that the factors are not specific regulators of FLC. One example of such a factor is PIE1, the closest relative among a large family of Arabidopsis proteins to budding yeast Swr1 (Noh and Amasino, 2003). Swr1 is a Swi2/Snf2-related ATPase required for incorporation of the H2A variant, Htz1 (conserved as H2A.Z in plants and higher eukaryotes), into chromatin (Krogan et al., 2003; Kobor et al., 2004; Mizuguchi et al., 2004). In vitro studies have implicated H2A.Z in destabilizing interaction between chromatin fibers in highly condensed chromatin and promoting formation of 30-nm chromatin fibers (Fan et al., 2002). This activity might promote a conformation that allows access to gene regulatory regions by DNA-binding transcription factors and other regulatory proteins while simultaneously precluding efficient transcriptional initiation and elongation by RNA polymerase II (Pol II). Indeed, along with the observations that Htz1 occupies yeast promoter regions and becomes depleted from open reading frames upon transcriptional activation (Farris et al., 2005), these observations suggest that Htz1/H2A.Z (and by association, perhaps also Swr1) is involved in poising genes within the chromatin fabric for transcriptional activation (Fan et al., 2002). Although its
230
REGULATION OF TRANSCRIPTION IN PLANTS
biochemical activity has yet to be demonstrated, PIE1 shares several important features with Swr1, including the SNF2-homologous region, a HELICc domain found in an assortment of helicases and related proteins, a SANT domain implicated in DNA binding, and the SANT-associated HSA domain found in numerous DNA-binding proteins including helicases. Swr1 acts in concert with numerous other proteins in a chromatin-remodeling complex, SWR1c (Kobor et al., 2004; Mizuguchi et al., 2004), and given that many of these are also conserved (Kobor et al., 2004) it would not be surprising if the respective plant genes also are required for FLC transcription and functionally interact with PIE1. Indeed, it has recently been demonstrated that the Arabidopsis counterpart of the SWR1c actin-related protein Arp6 is required for full expression of FLC, and its mutation results in developmental pleiotropy similar to that seen in pie1 mutants (Choi et al., 2005; Deal et al., 2005). The biochemical function of Arp6 is similarly unknown, but the plant ARP6 protein is observed in foci at the nuclear periphery (Choi et al., 2005). This interesting and seemingly unusual property and the requirement for ARP6 in FLC activation are evocative of the observed physical association of GAL chromatin with nuclear pore complex proteins upon activation (Casolari et al., 2004; Misteli, 2005), potentially representing a more universal strategy of gene regulation (Casolari et al., 2005), and suggest that association with the nuclear envelope might be involved in FLC regulation. A potentially related mechanism that promotes FLC expression involves ESD4, one of three Arabidopsis proteins closely related to a SUMO protease from budding yeast called Ulp1. SUMO, a small ubiquitin-related protein, is post-translationally ligated to numerous cellular proteins including an assortment of nuclear factors related to transcription and chromatin structure (Gill, 2004). Similar to the nonproteolytic functions of ubiquitination, SUMOylation alters localization and/or activity of target proteins (Hilgarth et al., 2004), and SUMO-specific proteases such as Ulp1 participate in processingof immature SUMO and removal of SUMO from protein targets. ESD4 likely acts as a bona fide SUMO-specific protease, because esd4 mutants show a striking accumulation of SUMO–protein conjugates (Murtas et al., 2003). In relation to transcription, SUMOylation is generally regarded as a repressive mechanism. Perhaps the best characterized targets of SUMOylation are transcription factors such as Sp3, Elk-1, c-Myb, and c-Jun (Verger et al., 2003). In the majority of cases studied, modification of these proteins by SUMO is associated with decreased activation potential (Gill, 2004). SUMOylation has also been associated with reinforcing gene repression; the human Polycomb group (PcG; see below) repressor Pc2 is a SUMO ligase that targets the transcriptional corepressor CtBP (Kagey et al., 2003). In addition, SUMOylated histone H4 has been connected with transcription repression associated with recruitment of HDAs (Shiio and Eisenman, 2003). Interestingly, at least in the case of Sp3, repression is associated with sequestration of the transcription factor to the nuclear periphery (Ross et al., 2002). Ulp1 is localized to the nuclear periphery and has been shown to be affixed to nuclear pores (Takahashi et al., 2000; Panse et al., 2003). ESD4 also is localized to the nuclear periphery (Murtas et al., 2003), and it is tempting to speculate that here it cooperates
CONTROL OF FLOWERING TIME
231
with ARP6. Further research into the phenomenon of subnuclear spatial regulation of transcription therefore should provide new insights into potential mechanisms of FLC regulation. Zhang et al. (2003) carried out a genetic screen for loss of FLC expression in a winter-annual genotype that led to the identification of at least 20 distinct loci required for appreciable FLC expression, with 7 of these, designated the VERNALIZATION INDEPENDENCE (VIP) loci, defining a single epistatic group. Based on their identical developmental pleiotropy, and the high degree of overlap of transcriptional profiles between at least the vip5 and vip6 mutants, these loci likely define a class of genes with identical function (Zhang and van Nocker, 2002; Zhang et al., 2003; Oh et al., 2004). Loss of these genes leads to silencing not only of FLC but also of the additional five members of the FLC gene family, MADS AFFECTING FLOWERING (MAF) 1–5 (He et al., 2004; Oh et al., 2004), all of which have the capacity to act as floral repressors (see below) (Ratcliffe et al., 2001, 2003; Scortecci et al., 2001). Four of the VIP genes (VIP4, VIP5, VIP6 (also called ELF8), and ELF7 (formerly VIP2)) encode homologs of distinct subunits of a transcriptional regulator from budding yeast called Paf1C (He et al., 2004; Oh et al., 2004). In yeast, Paf1C subunits associate with the initiating and elongating forms of Pol II (Mueller and Jaehning, 2002), and are bound within 5 regions and open reading frames of various genes (Krogan et al., 2002; Simic et al., 2003). Paf1C acts in transcriptional activation through assisting in ubiquitination of the carboxyl-terminal domain of histone H2B by the ubiquitin conjugating/ligase proteins Rad6/Bre1 (Henry et al., 2003; Ng et al., 2003a; Wood et al., 2003). At least in yeast, H2B ubiquitination is a prerequisite for establishment of methylation of histone H3 at lysines 4 and 79 by the HMTs Set1/COMPASS and Dot1, respectively, within genes (Wood et al., 2003), and Paf1C is required for association of these methyltransferases with Pol II. Trimethylation of H3K4 is uniquely associated with transcriptional activity (Hampsey and Reinberg, 2003), and the association between elongating Pol II and Set1/COMPASS, mediated by Paf1C, has been proposed as a mechanism to perpetuate the active state of genes (Ng et al., 2003b). This may involve recruitment by trimethylated H3K4 of chromatin-remodeling coactivators such as Isw1 (Santos-Rosa et al., 2003) or, in flies, Brahma (Beisel et al., 2002). The plant counterpart of this mechanism may function in a similar manner to maintain the activation of FLC in winter-annual accessions. However, there is yet only scant evidence that the VIP complex has a role similar to its yeast counterpart. The VIP proteins do interact in vivo, and are found in the nucleus in a ∼0.5-mDa complex (Oh et al., 2004; our unpublished results). However, this complex also includes VIP3, a small WD-repeat protein that is conserved in higher eukaryotes but has no clear homology with any yeast protein (Zhang et al., 2003; Oh et al., 2004). Also, unlike yeast strains deleted for components of Paf1C, Arabidopsis vip mutants did not exhibit detectable ‘global’ defects in methylation at H3K4 or H3K79 when assayed on a bulk chromatin and total plant tissue basis (Oh et al., 2004). VIP3, VIP6/ELF8, and ELF7 are required for detectable levels of trimethylation of H3K4 within the FLC transcriptional unit (He et al., 2004; our unpublished results).
232
REGULATION OF TRANSCRIPTION IN PLANTS
However, it remains unclear if this reflects a direct influence of VIP activity. Loss of H3K4 methylation was also observed in FLC silenced in wild-type, vernalized plants (Bastow et al., 2004), suggesting that this might simply represent an indirect outcome of FLC silencing. What might carry out activating histone methylations at FLC? Mutations in the EFS gene, which encodes one of several Arabidopsis Set1/Trx-like HMTs, confer early flowering (Soppe et al., 1999), and this is associated with a loss of FLC expression and H3K4 trimethylation within FLC chromatin, implicating this gene as a candidate (Soppe et al., 1999; Kim et al., 2005).
9.4
Vernalization-associated repression of FLC
In many plants, including winter-annual-habit accessions of Arabidopsis, the vegetative phase can be significantly abbreviated by growth at very low temperatures, a phenomenon known as vernalization. In the last few years, several key studies have illuminated the involvement of chromatin dynamics, transcriptional silencing, and epigenetic mechanisms in the vernalization response in Arabidopsis. The molecular mechanism of vernalization involves the downregulation of FLC and at least a subset of the five paralogous MAF genes (Michaels and Amasino, 1999; Sheldon et al., 1999; Ratcliffe et al., 2001, 2003). Studies have focused on FLC, probably because of the obvious role of this gene in repressing flowering in winter-annual accessions. Genetic screens for loss of vernalization sensitivity have identified several loci, and the molecular identities of three of these – namely VRN1, VRN2, and VIN3 – have been reported (Gendall et al., 2001; Levy et al., 2002; Sung and Amasino, 2004). The VRN1 protein contains B3-type DNA-binding domains at its carboxyl and amino termini; this is a plant-specific class of DNA-binding domain found in several transcription factors such as VP1 and ABI3 (Levy et al., 2002). VRN2 is one of three proteins in Arabidopsis related to the fly PcG protein Suppressor of zeste 12 (Su(z)12) (Gendall et al., 2001). PcG is a class of proteins initially identified through a series of genetic screens for ectopic formation of sex combs. These proteins repress a subset of genes, including the homeotic hox genes, outside of appropriate expression domains, at least partly through maintenance of transcriptional-repressive chromatin modifications including methylation of lysines 9 and 27 of histone H3 (Czermin et al., 2002; Kuzmichev et al., 2002; Muller et al., 2002). Several of the PcG proteins are conserved among higher eukaryotes, including components of the HMT complex ESC–E(z) (also called PRC2). Among other subunits, this complex in flies contains Su(z)12, the SET-domain protein enhancer of zeste (E(z)), and the WD-repeat proteins extra sex combs (Esc) and p55/NuRF-55. Whereas E(z) provides the catalytic core of the enzyme, Su(z)12 and NuRF-55 are required for interaction with nucleosomes (Nekrasov et al., 2005). VRN2 and other Su(z)12-related proteins contain two structural features of note (Birve et al., 2001). The first is a predicted C2H2 zinc finger of the class known to mediate sequence-specific DNA binding. Potentially, this mediates binding to DNA
CONTROL OF FLOWERING TIME
233
in vivo, as mutations expected to disrupt the zinc finger of Su(z)12 abrogate its biological activity without affecting assembly of the associated ESC–E(z) complex, or its in vitro HMT activity. (Ketel et al., 2005). The second feature is a strongly conserved, ∼80-amino-acid region that has been termed the VEFS (for VRN2EMF2-FIS2-Su(z)12 box). Although this region of Su(z)12 is required for integrity of the ESC–E(z) complex, specific mutations in the VEFS box disrupt in vitro HMT activity without affecting complex formation (Ketel et al., 2005), suggesting a role beyond protein–protein interaction. VRN1 and VRN2 repress FLC in a manner characteristic of PcG mechanisms. In both vrn1 and vrn2 mutants, FLC RNA is downregulated during an extended cold treatment, similar to wild-type plants, but becomes strongly derepressed during subsequent growth at warmer temperatures (Gendall et al., 2001; Levy et al., 2002). Therefore, these proteins are required only for maintenance of FLC repression, and not for its initial silencing. At least VRN2 also has a promotive role in flowering in nonvernalized plants, potentially as a suppressor of FLC outside of its typical expression domains (Gendall et al., 2001). In addition to the homology between the VRN2 and Su(z)12 proteins, molecular evidence suggests that VRN2 (and VRN1) might participate in an ESC–E(z)-like HMT. Like ESC–E(z), the activity of VRN2 and VRN1 is associated with methylation of H3K9 and H3K27. This occurs within promoter and intronic regions of FLC formerly demonstrated to contain cis-elements required for maintenance of repression (Sheldon et al., 2002). VRN1 is presumed to work downstream of VRN2 to assist in methylation of H3K9, because H3K27 methylation within FLC is still observed in a vrn1 mutant (Bastow et al., 2004). What might be the catalytic core of this methyltransferase? Based on the homologous components seen in ESC–E(z) complexes in flies and humans, VRN2 probably interacts with one or more E(z)related proteins in plants. Good candidates are the E(z) homologs CURLY LEAF (CLF) and/or SWINGER (SWN), both of which physically interact with VRN2 in yeast (Chanvivattana et al., 2004). The VIN3 protein contains an internal motif resembling a PHD (plant homeodomain)-type zinc finger, common in proteins that participate in chromatin regulation (Aasland et al., 1995). In contrast to VRN1 and VRN2, VIN3 is required for the initial silencing of FLC during cold (Sung and Amasino, 2004). VIN3 RNA accumulates during extended cold exposure with kinetics similar to those seen for loss of FLC RNA expression, suggesting direct regulation. VIN3 activity is also associated with deacetylation of histone H3 within FLC promoter and intronic regions (Sung and Amasino, 2004). The mechanisms involved in upregulation of VIN3 RNA during cold, and VIN3 targeting of FLC for repression, are intriguing, as VIN3 appears to be the most ‘upstream’ component of the vernalization mechanism identified so far. Although the importance of the PHD domain in VIN3 has not been addressed, and these domains have traditionally been perceived as mediating protein–protein interactions, it is interesting that PHD fingers of several animal proteins have been shown to bind phospholipids, and at least in one case phospholipid binding was required for chromatin association and activity (Gozani et al., 2003). The involvement of
234
REGULATION OF TRANSCRIPTION IN PLANTS
phospholipid signaling in plant stress responses is becoming increasingly documented (van Leeuwen et al., 2004), and it is conceivable that VIN3 represents a target. Indeed, ectopic expression of VIN3 in transgenic plants under nonvernalizing conditions is insufficient to promote flowering, suggesting that vernalizationassociated accumulation of VIN3 is not alone a regulatory step, but that VIN3 function requires the participation of other cold-associated mechanisms (Sung and Amasino, 2004). VIN3, VRN1, and VRN2 likely function as general regulators of the vernalization response rather than FLC-specific factors; vernalized vin3, vrn1, or vrn2 mutants flower as late as nonvernalized plants, and loss of VIN3 decreases vernalization sensitivity even in an flc mutant (Gendall et al., 2001; Levy et al., 2002; Sung and Amasino, 2004). Obvious additional targets are the MAF genes, although this has yet to be demonstrated. Vernalization-associated downregulation of FLC and maintenance of its repression provides an attractive model to study the sequence of events associated with epigenetic gene silencing. An interesting and unexplored question is the means by which PcG silencing overcomes the reinforced transcriptional activity potentially set up by the Paf1C-like (VIP group)/Trx-like (EFS) mechanisms. A possible scenario can be constructed based on cumulative studies of antagonism between PcG and Trx mechanisms, in animals, which suggest that PcG mechanisms target a subset of genes for repression opportunistically in response to transcriptional inactivity (Klymenko and Muller, 2004). Upon cessation of Pol II activity, transcriptionalpromotive epigenetic marks on FLC chromatin, including trimethylation of H3K4 and acetylation of H3/H4, could be erased, either actively by enzymatic mechanisms (see above) and/or by replication-independent histone exchange, or through dilution by nascent histones over the course of several cell divisions. FLC chromatin would then be a suitable substrate for H3K9/K27 methylation mediated by VRN2 and VRN1. This scenario is consistent with studies in fission yeast and animals, revealing that acetylation of H3K9 precludes its methylation, and that H3K9 methylation also requires deacetylation of additional lysine residues in the H3 amino terminus (Rea et al., 2000; Nakayama et al., 2001). Given the observed interaction between HDAs and ESC–E(z) (Tie et al., 2001), it may follow that HDAs recruited to FLC, potentially by VIN3, assist in recruiting the PcG machinery. Although information in plants is limited, studies in animals and fission yeast suggest that methylation at H3K9 within euchromatic regions can promote the formation of localized heterochromatin and long-term gene silencing, suggesting a precedence for the stable repression of FLC in vernalized plants. This may involve LIKE HETEROCHROMATIN PROTEIN 1 (LHP1), also known as TERMINAL FLOWER 2 (TFL2), the Arabidopsis homolog of HP1 proteins (Gaudin et al., 2001; Takada and Goto, 2003). HP1 proteins can bind methylated H3K9 through a chromodomain, and are common protein constituents of heterochromatin (Bannister et al., 2001). Human Su(z)12 interacts with HP1 (Yamamoto et al., 2004), and it has been suggested that these interactions could be a mechanism for spreading of heterochromatin into adjacent euchromatic regions, a well-recognized phenomenon
CONTROL OF FLOWERING TIME
235
in flies termed position-effect variegation (Elgin, 1996; Yamamoto et al., 2004). It is tempting to speculate that such a mechanism could reproduce epigenetic silencing marks on nascent histone octamers following DNA replication, using the persisting octamers as a template. Although LHP1/TFL2 acts in flowering (see below), a potential role in vernalization-associated FLC silencing has not been explored. A fascinating but relatively unexplored question is the relationships between the autonomous and vernalization pathways. Although both target FLC, there is yet no indication that the pathways utilize common components. The autonomous pathway mutants that have been characterized so far are responsive to vernalization, suggesting that the respective proteins do not play a crucial role in initiating or maintaining FLC repression associated with vernalization. Silencing of FLC in the rapid cycling accession Ler, which is dependent on activity of the autonomous mechanisms, does not obviously involve H3K9/H3K27 methylation (Bastow et al., 2004), suggesting that PcG repression is not relevant to autonomous pathway functions. However, FVE-related proteins (i.e., MSI1 homologs) participate in ESC–E(z) in humans and flies (Kuzmichev et al., 2002; Muller et al., 2002), and so Arabidopsis MSI1-like proteins might be expected to cooperate with VRN2. Additionally, biochemical evidence (described above) suggests that both mechanisms utilize a HDA, probably of the RPD class. In this respect, exploration of potential roles of additional members of the MSI1 and RPD gene families in vernalization should be informative.
9.5
Transcriptional repression of flowering by FLC
A major downstream target of FLC is another MADS-box gene, SOC1 (formerly also known as AGL20) (Lee et al., 2000b; Samach et al., 2000). FLC represses SOC1, and probably does so directly. In vitro, FLC binds to a segment of DNA from the SOC1 promoter containing a CArG box (the known binding site of other MADS-box factors) (Hepworth et al., 2002). This is likely to be relevant in vivo, because deletion of this promoter region abrogated much of the repressive effect of FLC on SOC1 expression (Hepworth et al., 2002). SOC1 is also targeted by other flowering pathways (see below) and has been termed a ‘flowering integrator’ gene. Other targets of FLC are the FT and TSF genes. In genotypes that express high levels of FLC due to disruption in the autonomous pathway or presence of active FRI alleles, FT and TSF RNA levels were reduced; in addition, the delayed flowering of these genotypes is mostly suppressed by constitutive transgenic expression of FT and TSF (Samach et al., 2000; Hepworth et al., 2002; Michaels et al., 2005; Yamaguchi et al., 2005). FT and TSF play an additional, and probably more important, role in photoperiodic induction of flowering, and are discussed in more detail below. MADS-box proteins including FLC are expected to interact with their binding sites as homo- or heterodimers or tetramers, and also to interact with other transcription-related proteins. Surprisingly, in light of the fact that so many studies have targeted FLC, binding partners of FLC are yet unknown, and this clearly is an attractive area for further research.
236
9.6
REGULATION OF TRANSCRIPTION IN PLANTS
Transcriptional regulation in the photoperiodic induction of flowering
The promotion of flowering by long-day photoperiods in Arabidopsis is mediated through what is probably the best known transcriptional regulator of flowering, CONSTANS (CO) (Putterill et al., 1995). CO is one of numerous plant genes that are regulated by an endogenous (circadian) clock (Suarez-Lopez et al., 2001). Cyclic transcription of CO is regulated partly by the F-box protein FKF1, through its interaction with and destabilization of CDF1, a Dof-zinc-finger family transcription factor that binds to the CO proximal promoter to repress CO expression (Imaizumi et al., 2005). It has been proposed that gradual accumulation of FKF1 throughout the day progressively eliminates promoter-bound CDF1, derepressing CO in the afternoon (Imaizumi et al., 2005). When high levels of CO RNA and protein are reached during the light period, as in the evening hours in inductive, long-day photoperiods, CO is able to efficiently activate downstream flowering genes (below). The molecular mechanisms involved in circadian clock function upstream of CO expression, and the role of light in these mechanisms, have been the subject of numerous excellent reviews (Hayama and Coupland, 2003; Salome and McClung, 2004; Searle and Coupland, 2004) and will not be discussed in detail here. CO is a nuclear protein containing B-boxes, the conserved, ∼40-amino-acid zinc-finger motifs that are found in numerous transcription factors. B-boxes have been best characterized in animal proteins as protein interaction domains, and have not been demonstrated to participate in DNA binding (Borden, 1998). The carboxyl terminus of CO contains a CCT (CONSTANS, CO-like, TOC1) motif, a ∼43-aminoacid module that is enriched in basic amino acids. This domain is found also in a small number of Arabidopsis proteins including other CONSTANS-related proteins (see below), the PSEUDO-RESPONSE REGULATORS (APRRs) including the endogenous clock component TOC1, and in a few GATA-type presumed transcription factors. It has been proposed to participate in nuclear localization and/or protein interactions (Strayer et al., 2000). CO is one of at least 17 members of a group of related genes in Arabidopsis called CONSTANS-LIKE (COL). The encoded proteins are predicted to contain B-box zinc fingers and often also carboxyl-terminal CCT domains. Where studied, the COL genes do not have an obvious role in flowering (Ledger et al., 2001). A possible exception is COL9; constitutive overexpression of COL9 in transgenic plants delayed flowering, and late flowering could be abrogated by parallel overexpression of CO (Cheng and Wang, 2005). Also, COL9 overexpression did not affect the circadian phase or period of any clock-regulated gene tested, including CO, but did affect the amplitude of CO oscillations (Cheng and Wang, 2005). Therefore, COL9 may participate in flowering regulation through CO, but downstream or independently of the circadian clock. Samach et al. (2000) identified genes that are likely to be direct targets of CO, using a constitutively expressed fusion of CO to the glucocorticoid receptor. This fusion protein accumulates in the cytosol and can translocate into the nucleus in the presence of a steroid hormone such as dexamethasone (DEX). When DEX is
CONTROL OF FLOWERING TIME
237
applied together with the translational inhibitor cycloheximide (CYC), the fusion protein activates only its direct transcriptional targets. Early targets of CO that were identified through this approach included the FT gene (Samach et al., 2000). Wigge et al. (2005) similarly searched for early targets of CO, using transcriptional profiling to identify genes that were activated soon after a transition to an inductive photoperiod and dependently on CO expression. This approach also identified FT as an early target of CO. Under long-day conditions, the diurnal expression pattern of FT RNA roughly corresponds with that of CO RNA (Suarez-Lopez et al., 2001). FT showed little or no transcriptional response to lengthened photoperiods in a co mutant, suggesting a lack of even partially redundant pathways of activation (Wigge et al., 2005). The close functional relationship between CO and FT was also revealed by the substantially similar transcriptional profiles seen in co and ft mutants (Schmid et al., 2003). However, CO expression is capable of promoting flowering to a small extent even in an ft mutant, suggesting that CO targets additional flowering gene(s) (An et al., 2004). Indeed, Yamaguchi et al. (2005) found that the FT paralog TWIN SISTER OF FT (TSF) was induced quickly in response to CO activation. In addition, like FT, TSF exhibited increased expression in response to extended photoperiods, was required for appropriate timing of flowering under these conditions, and accelerated flowering when expressed ectopically (Michaels et al., 2005; Yamaguchi et al., 2005). Also similar to FT, rapid activation of TSF by CO was seen even in the presence of CYC. This suggests that TSF promotes flowering as a direct target of CO, partially redundant with FT. However, loss of TSF has a much lesser effect on flowering, probably at least partly due to its more limited expression pattern (Yamaguchi et al., 2005). Both genes are expressed in vascular (phloem) tissues, but in distinct parts of the plant, with FT expression seen in cotyledons and leaf blades and TSF expression seen mainly in the hypocotyl and petiole (Yamaguchi et al., 2005). Coregulation of FT and TSF genes probably is not surprising, as these proteins are closely related (Kobayashi et al., 1999). Another member of the FT/TSF gene family, TERMINAL FLOWER 1 (TFL1), is upregulated in the center of the shoot apex during the floral transition, and responds quickly to CO activation (Simon et al., 1996), and thus represents an additional potential target of CO.
9.7
Activation of SOC1 by CO
The SOC1 gene, described above as a downstream target of FLC, was identified as a mutation capable of suppressing much of the flowering-promotive effect of overexpression of CO (Onouchi et al., 2000), and also by Samach et al. (2000) as an early transcriptional target of CO. Activation of SOC1 by CO in the system used by Samach et al. (2000) was insensitive to CYC, and so is probably mediated directly. The transcriptional basis of this activation was addressed in a series of experiments utilizing gel-shift assays and expression of promoter truncations in transgenic plants (Hepworth et al., 2002). Deletion of a limited region of the SOC1 promoter abrogated most of the activation of SOC1 by CO, suggesting that CO
238
REGULATION OF TRANSCRIPTION IN PLANTS
targets this region. However, recombinant CO apparently does not bind to SOC1 promoter DNA, at least in vitro (Hepworth et al., 2002). This is consistent with the observation that the CO protein does not exhibit any known DNA-binding domains (see above), and suggests that CO should cooperate with additional, DNA binding factors to activate SOC1. In contrast to FT, which shows absolute dependency on CO, SOC1 could be induced by a switch to inductive photoperiods even in a co mutant (Wigge et al., 2005), an effect potentially mediated through gibberellins (see below). An interesting question is the mechanism of coregulation of SOC1 by CO and FLC. SOC1 is expressed most strongly in leaves, but is detected in leaf primordia and the shoot apex of photoperiod-induced plants (Samach et al., 2000; Wigge et al., 2005). Expression of FLC is preferential for apical regions including the shoot apex, but also is seen in leaves (Michaels and Amasino, 1999); so, it is unclear where the relevant integration of FLC and CO signaling takes place. The CO-targeted promoter region identified by Hepworth et al. (2002) is distinct from that containing the CArG box demonstrated to bind FLC (above), and so flowering inputs from FLC and CO are mediated independently at the SOC1 promoter. High levels of FLC blocks transcriptional activation of SOC1 by CO, and it was hypothesized that promoterbound FLC might assemble a repressive protein complex (Hepworth et al., 2002). In addition to its apparent direct activation by CO, SOC1 also appears to be activated through FT and TSF. Although SOC1 clearly can be induced by long days in the absence of FT or TSF, SOC1 RNA is decreased in an ft mutant (Lee et al., 2000b; Schmid et al., 2003; Yamaguchi et al., 2005) and increased in 35S::TSF and 35S::FT plants (Michaels et al., 2005; Moon et al., 2005; Yamaguchi et al., 2005). This positions SOC1 in the flowering regulatory pathway ‘downstream’ from FT and TSF. An additional and very interesting regulatory output of the photoperiodic pathway mediated through CO/FT is the upregulation of a gene encoding a precursor of a microRNA, mir172 (Schmid et al., 2003). mir172 targets a set of previously uncharacterized, APETALA2 (AP2)-like flowering repressors, including ¨ TOE1, TOE2, SCHLAFMUTZE, and SCHNARCHZAPFEN, for translational and post-transcriptional downregulation (Aukerman and Sakai, 2003; Schmid et al., 2003). In maize, mir172 is also implicated in regulating flowering time through downregulation of the AP2-like gene glossy 15 (gl15) (Lauter et al., 2005), suggesting a conserved and important role in flowering.
9.8
Chromatin-related mechanisms of photoperiod pathway regulation
Chromatin-based mechanisms of transcription are also important for photoperiodic induction of flowering. FT is repressed by mechanisms involving the chromatinrelated factors, EARLY BOLTING IN SHORT DAYS (EBS) and LHP1/TFL2 (described above) (Kotake et al., 2003; Pi˜neiro et al., 2003; Takada and Goto, 2003). FT RNA accumulates to higher levels in ebs mutants throughout the diurnal cycle, and this is especially evident in the dark phase when plants are grown in a short-day
CONTROL OF FLOWERING TIME
239
(8 h light/16 h dark) photoperiod (Pi˜neiro et al., 2003). Interestingly, the phase or amplitude of CO RNA cycling was apparently unaffected in ebs mutants. In addition, CO was not required for substantial accumulation of FT RNA in an ebs mutant. Although a mechanism to explain these observations has yet to be proposed, this implicates EBS in repression of FT expression independently of or downstream from CO, particularly during the dark phase (Pi˜neiro et al., 2003). An additional target of EBS is TSF (Yamaguchi et al., 2005). EBS RNA is prevalent in shoot and floral apices, and in young leaf and flower primordia, especially in outer cell layers (Pi˜neiro et al., 2003). The observation that EBS RNA is expressed constitutively with respect to the floral transition suggests that EBS does not have a regulatory role in flowering. In addition, the developmental pleiotropy seen in ebs mutants (Gomez-Mena et al., 2001), which is not observed in transgenic plants overexpressing FT or TSF, suggests that its role is not confined to regulation of these genes. Based on epistatic interactions and phenotypic similarity with mutants defective in gibberellic acid (GA) signaling, Gomez-Mena et al. (2001) suggested that EBS might play a role in the induction of flowering by GAs. However, to date, there is no clear evidence that GAs directly participate in regulation of FT (see below) and so this potential role remains uncertain. Unlike many other chromatin-related proteins that participate in flowering time, loss of EBS expression had no discernible effect on expression of FLC (Pi˜neiro et al., 2003). EBS encodes a nuclear protein that contains a so-called bromoadjacent homology (BAH) region. Although its function is unknown, the BAH region is seen in numerous proteins with chromatin-related activities, including yeast RSC1 and RSC2 subunits of the ATP-dependent chromatin-remodeling factor RSC, the conserved cytosine methyltransferases, and, in plants, the chromodomain DNA methyltransferases (chromomethylases). Most of these proteins contain several other conserved chromatin-related domains. However, EBS is a relatively small protein and does not contain any other recognizable features except for a PHD-type Zn finger. As the PHD finger is supposed to mediate protein–protein interactions, EBS may operate as part of a larger protein complex and contribute a BAH domain for this protein. In fact, overexpression of EBS in Arabidopsis leads to phenotypic defects similar to those seen in ebs loss-of-function mutants, and one explanation for this is a dominant negative effect of enhanced gene dosage on activity of a potential protein complex (Pi˜neiro et al., 2003). EBS is related to two additional proteins in Arabidopsis with similar domain structure; one of these, termed SHORT LIFE (SHL), is a nuclear protein required for normal flowering time (Mussig et al., 2000). However, unlike EBS, SHL probably normally acts to promote flowering, as reduced SHL activity due to antisense expression resulted in delayed flowering, and flowering was accelerated when SHL was ectopically expressed to high levels (Mussig et al., 2000). Although the position of SHL in flowering pathways has not been genetically defined, delayed flowering of SHL-antisense plants is associated with repression of SOC1 (Mussig et al., 2003), suggesting that SHL targets SOC1 or its upstream regulators. In mutants defective for the LHP1/TFL2 gene, FT is expressed prematurely to higher levels, and ectopically, suggesting that LHP1/TFL2 moderates the proper
240
REGULATION OF TRANSCRIPTION IN PLANTS
spatial and temporal activation of FT (Takada and Goto, 2003). Repression of FT by LHP1/TFL2 and its activation by CO are likely mediated through separate pathways, because FT expression is still responsive to daylength or ectopic CO expression in plants lacking lhp1/tfl2 activity, and ectopic expression of FT associated with loss of LHP1/TFL2 is seen even in a co mutant (Takada and Goto, 2003). Interestingly, LHP1/TFL2 does not seem to participate in repression of TSF (Yamaguchi et al., 2005). LHP1/TFL2 is also important for silencing a variety of additional genes important for plant development (Nakahigashi et al., 2005). Although HP1 outside of plants has been best characterized as a constituent of constitutive heterochromatin, it also has known functions within euchromatin, including assisting Rb in silencing of E2F-regulated cell cycle genes (Nielsen et al., 2001), suggesting a precedent for the direct involvement of LHP1/TFL2 in euchromatic gene silencing. Two additional transcriptional components that participate in photoperiodic flowering are the MADS-box genes, SHORT VEGETATIVE PHASE (SVP) and MAF1 (also called FLM). Mutation in either gene causes early flowering in noninductive photoperiods, and both can confer delayed flowering when constitutively expressed (Hartmann et al., 2000; Ratcliffe et al., 2001; Scortecci et al., 2001). These proteins may work closely together, because a svp/maf1 double mutant did not exhibit earlier flowering relative to either single mutant, and SVP is required for the late flowering conferred by constitutive expression of MAF1/FLM (Scortecci et al., 2003). The precise roles of these factors in flowering should become clear through analysis of their transcriptional relationships with other photoperiodic pathway genes and flowering integrator genes. It is worthwhile to point out that in spite of the close paralogous relationship between FLC and MAF1/FLM, these genes show clear differences in regulation and function. For example, FLC expression is regulated strongly by autonomous mechanisms genes or FRI (above), but is relatively insensitive to activity of genes that participate in photoperiodic flowering (Sheldon et al., 1999), whereas the autonomous mechanisms or FRI do not seem to be significant regulators of MAF1/FLM (Ratcliffe et al., 2001; Scortecci et al., 2001). Thus, these proteins have been recruited for very different flowering roles. The four remaining members of the MAF clade have not been extensively characterized, although it is known that all have the capacity to act as floral repressors (Ratcliffe et al., 2003).
9.9
Transcriptional activation of AP1 by FT and FD
CO regulatory sequences mediate expression that is predominately localized to vascular (phloem) tissues of cotyledons and leaves (Takada and Goto, 2003). Because change in fate of primordia is effected at the apex, CO must provide a mobile signal to initiate this change. It now appears that FT is an important factor in this phenomenon. Grafting and targeted expression studies revealed that CO promotes flowering by activating FT expression in phloem companion cells (An et al., 2004). Huang et al. (2005) used a heterologous, heat-inducible promoter to show that FT
CONTROL OF FLOWERING TIME
241
RNA produced transiently in the leaves was able to move into the shoot apex. Thus, FT may comprise this mobile flowering signal. Two recent studies showed that the transcription factor FD plays a central role in the promotion of flowering by FT in the shoot apex (Abe et al., 2005; Wigge et al., 2005). FD is a member of the bZIP family, which includes at least 75 members in Arabidopsis (Jakoby et al., 2002). Mutation in FD strongly suppressed early flowering conferred by constitutive, high-level expression of FT (Abe et al., 2005; Wigge et al., 2005). This suggested that FT works with or through FD to promote flowering. By contrast, the fd mutation only slightly affected precocious flowering in transgenic plants constitutively expressing SOC1 or LFY (Abe et al., 2005), suggesting a special relationship between FT and FD. FD appears to be constitutively expressed in the apex, and FT and FD proteins were shown to interact in the nucleus to promote transcription of the meristem identity gene, APETALA1 (AP1) (Abe et al., 2005; Wigge et al., 2005). This activation appears to be independent of other regional factors, because ectopic expression of FD in regions of seedlings that express FT (vasculature of cotyledons) is sufficient to drive expression of an AP1 reporter gene (Abe et al., 2005). A region of the AP1 promoter previously demonstrated to direct authentic expression (Hempel et al., 1997) contains several of the ACGT-core sequence motifs that bind plant bZIP factors (Foster et al., 1994; Wigge et al., 2005) and these may be directly targeted by FD. FD contains a potential phosphorylation site on its amino terminus that is required for binding to FT and for activity, although it remains to be determined if this residue is phosphorylated in vivo (Abe et al., 2005). As a bZIP factor, FD is expected to work as a dimer, and indeed FD is also able to interact in yeast cells both with itself and, intriguingly, with a paralogous bZIP protein (Abe et al., 2005). The mechanistic role that FT plays in activation of AP1 is a largely unresolved question. Although LFY also targets AP1, FT and FD probably act independently of LFY, because mutations in FT and LFY, or FD and LFY, have a synergistic effect on flowering when combined. Other targets of FT that may require FD are the MADS-box gene FRUITFULL (FUL) and the AP1 paralog CAULIFLOWER (CAL) (Ferrandiz et al., 2000; Teper-Bamnolker and Samach, 2005). An interesting observation is that in the inducible system used by Huang et al. (2005), transgenic expression of FT in leaves was associated not only with the accumulation of the transgenic transcript in the apex, but also with the induction of endogenous FT both in the leaves and in the shoot apex, suggesting that FT participates in a positive autoregulatory feedback loop. This appeared to be robust, as endogenous FT RNA was produced for at least 3 days after activation in plants maintained in noninductive conditions (Huang et al., 2005). This mechanism may explain the ability of some plants, including Arabidopsis, to initiate and maintain flowering in response to even a transient inductive signal. The mechanism of this positive autoregulation is an interesting and open question. FT may activate its own promoter directly, or this may occur through activation of downstream targets such as SOC1. SOC1 is the only one among the known targets of FT to be appreciably expressed in leaves where autoinduction can occur (Huang et al., 2005). In any
242
REGULATION OF TRANSCRIPTION IN PLANTS
case, given its lack of apparent DNA-binding motifs and dependence on FD in the apex, FT should collaborate with a leaf-expressed transcription factor to carry this out.
9.10
Transcriptional mechanisms in the promotion of flowering by GAs
As in many plants, GAs play a major role in flowering in Arabidopsis. Arabidopsis plants grown in noninductive short days eventually do undergo the flowering transition, and under these conditions flowering is absolutely dependent on GA signaling (Wilson et al., 1992). LFY is a central component in GA-promoted flowering. In noninductive photoperiods, LFY promoter activity and RNA expression show a gradual increase over time in the primordia emerging from the shoot apex, and this is dependent on GAs (Bl´azquez et al., 1998). GA promotion of LFY transcription is dependent on a GAresponse cis-element (GARE) in the LFY promoter (Bl´azquez et al., 1998; Bl´azquez and Weigel, 2000). This DNA binds the Myb-like protein AtMYB33 in vitro (Gocal et al., 2001), suggesting that GAs regulate LFY through AtMYB33. AtMYB33 is one of a small subset of Arabidopsis Myb proteins, designated the GAMYBs, which are known to participate in GA signaling (Gocal et al., 2001). Interestingly, AtMYB33 accumulates in the shoot apex of plants transferred from short to long days, an effect potentially driven by increased GA concentration (Gocal et al., 2001). This suggests that GAs should also contribute to promoting flowering under long days, and is consistent with observations that a GA biosynthetic mutant flowers slightly later than wild-type plants under long days (Wilson et al., 1992). Similar to the AP2-like flowering repressors (above), AtMYB33 is posttranscriptionally downregulated by a microRNA, in this case miR159 (Achard et al., 2004). miR159 accumulates in response to GA, and so this mechanism might moderate the GA-promotive effects of AtMYB33. Both miR159 and its target sequences within GAMYB-like RNAs are evolutionarily conserved, suggesting that microRNA-mediated downregulation of these GAMYBs is of fundamental importance (Achard et al., 2004). GA signaling is believed to involve the downregulation, through ubiquitin/26S proteasome-dependent proteolysis, of members of the DELLA family of nuclear repressors and subsequent derepression of GA-promoted genes (review in Sun and Gubler, 2004). There are five DELLA proteins encoded by the Arabidopsis genome, and these proteins have partially overlapping roles in various GA-dependent processes. Combined loss of function of two of these, GAI and RGA, rescued the late-flowering defect of a strong ga1 mutation (Dill and Sun, 2001), suggesting that these proteins mediate most of the flowering-promotive effect of GAs in Arabidopsis. The SOC1 gene also has been shown to participate in GA-mediated activation of LFY and flowering in short days (Borner et al., 2000; Moon et al., 2003a), and SOC1 RNA expression is upregulated in response to GAs (Moon et al., 2003a). Because
CONTROL OF FLOWERING TIME
243
GAs do not seem to strongly regulate expression of known upstream regulators of SOC1, including FT or FLC (Moon et al., 2003a), this activation may be fairly direct. However, GARE-like cis-elements in the SOC1 promoter that might mediate induction by GAs have not been reported. GA-mediated promotion of SOC1 expression is considered to be independent of miR159/MYB33, because late flowering associated with overexpression of miR159 (and consequently, decreased MYB33 RNA) was not associated with decreased SOC1 transcript levels (Achard et al., 2004). The identification of GA-signaling components upstream of SOC1 therefore is an interesting and unresolved matter. Although LFY is obviously a central component in the GA-flowering pathway, GAs also promote flowering by regulating genes that are downstream, or act independently, from LFY, as shown by the ability of a ga1 mutation to partially suppress the early flowering conferred by ectopic expression of LFY (Bl´azquez et al., 1998). This may be due at least in part to an LFY-independent activity of GA in promoting expression of floral homeotic genes such as PI, AP3, and AG (Yu et al., 2004).
9.11
PcG-mediated repression of floral homeotic genes
In addition to the maintenance of FLC silencing in vernalized plants (above), PcGlike mechanisms in Arabidopsis inhibit flowering by repressing floral homeotic genes outside of their appropriate expression domains. This activity is probably best illustrated by the EMF1 and EMF2 genes, which have similar function and encode a putative transcriptional regulator and Su(z)12 homolog, respectively (Aubert et al., 2001; Yoshida et al., 2001). In strong emf1 or emf2 mutants, the vegetative phase is largely bypassed and flowers form embryonically or during seedling development (Sung et al., 1992; Yang et al., 1995). Consistent with this phenotype, loss of at least EMF1 has been shown to result in derepression of genes acting in floral development, including AG, AP1, AP3, and PI, in vegetative tissues (Chen et al., 1997; Moon et al., 2003b). Early flowering associated with misexpression of floral homeotic genes was also observed in plants lacking activity of both CLF and SWN (Goodrich et al., 1997; Chanvivattana et al., 2004), and in mutants and transgenic plants cosuppressed for the ESC homolog FERTILIZATION-INDEPENDENT ENDOSPERM (FIE) (Kinoshita et al., 2001; Katz et al., 2004). Interestingly, at least in the emf1 mutant, ectopic expression of homeotic genes was not accompanied by substantial derepression of associated upstream regulators such as CO, FT, SOC1, or LFY (Moon et al., 2003b). This, along with the observation that loss of CO, SOC1, or LFY cannot rescue the emf1 phenotype (Moon et al., 2003b), suggests that EMF1 represses floral homeotic genes independently of flowering-time or flowering integrator genes. Thus, the early flowering observed in this class of mutants probably results from a bypass of normal flowering-timing mechanisms and the direct action of meristem identity and floral organ identity genes on the shoot apex. In wild-type plants, flowering may involve antagonism of PcG repression by flowering-time and flowering integrator genes, and the mechanism involved is an interesting question.
244
9.12
REGULATION OF TRANSCRIPTION IN PLANTS
Summary and prospects
The substantial amount of accumulated information related to transcriptional pathways involved in flowering in Arabidopsis has originated almost exclusively in phenotypic screens and genetic analyses. Thus, our knowledge of transcriptional regulation of flowering timing is ‘gene rich’ but ‘mechanism poor’. Although the identification of new flowering genes will continue to have a strong impact, the most significant advances will likely be based on emerging molecular and biochemical techniques. For example, the availability of genomic tiling microarrays covering essentially the entire Arabidopsis genome should allow rapid identification of sites bound by transcription- and chromatin-related factors and elucidation of the transcriptional networks in which they participate. The development of efficient, plantcompatible systems for purification of protein complexes based on affinity tags, an approach that has been very profitable in studies of transcriptional mechanisms in yeast and animals, should further accelerate the pace of discovery. Substantial effort is now being directed into the use of Arabidopsis as a blueprint for understanding flowering mechanisms in economically important crops such as rice, wheat, and maize. In general, such studies have revealed that many genes influencing flowering in Arabidopsis have been conserved through evolution, but that these genes sometimes are used in flowering in apparently novel ways (Hayama and Coupland, 2004; Marx, 2004). Thus, although continued studies in Arabidopsis obviously have merit, the ability to eventually control flowering in a sophisticated manner and in a range of species will necessitate the development of alternative flowering models. Novel findings are especially likely to await in the characterization of flowering mechanisms in perennial and woody plants.
Acknowledgments The authors acknowledge the contributions of many investigators that were not cited due to space limitations. This work was supported by funding from the U.S. Department of Agriculture (National Research Initiative Competitive Grant Program, No. 2003-35304-13299) and the U.S. National Science Foundation (MCB-0445867) to S.V.N.
References R. Aasland, T.J. Gibson and A.F. Stewart (1995) The PHD finger: implications for chromatin-mediated transcriptional regulation. Trends in Biochemical Sciences 20 (2), 56–59. M. Abe, Y. Kobayashi, S. Yamamoto, et al. (2005) A bZIP protein mediating signals from the floral pathway integrator FT at the shoot apex. Science 309 (5737), 1052–1056. P. Achard, A. Herr, D.C. Baulcombe and N.P. Harberd (2004) Modulation of floral development by a gibberellin-regulated microRNA. Development 131 (14), 3357–3365. R.M. Amasino (2003) Flowering time: a pathway that begins at the 3 end. Current Biology 13 (17), R670–R672. H. An, C. Roussot, P. Suarez-Lopez, et al. (2004) CONSTANS acts in the phloem to regulate a systemic signal that induces photoperiodic flowering of Arabidopsis. Development 131 (15), 3615–3626.
CONTROL OF FLOWERING TIME
245
D. Aubert, L. Chen, Y.H. Moon, et al. (2001) EMF1, a novel protein involved in the control of shoot architecture and flowering in Arabidopsis. The Plant Cell 13 (8), 1865–1875. M.J. Aukerman and H. Sakai (2003) Regulation of flowering time and floral organ identity by a microRNA and its APETALA2-like target genes. The Plant Cell 15 (11), 2730–2741. I. Ausin, C. Alonso-Blanco, J.A. Jarillo, L. Ruiz-Garcia and J.M. Martinez-Zapater (2004) Regulation of flowering time by FVE, a retinoblastoma-associated protein. Nature Genetics 36 (2), 162–166. N. Ayoub, K. Noma, S. Isaac, T. Kahan, S.I. Grewal and A. Cohen (2003) A novel jmjC domain protein modulates heterochromatization in fission yeast. Molecular and Cellular Biology 23 (12), 4356–4370. D. Balciunas and H. Ronne (2000) Evidence of domain swapping within the jumonji family of transcription factors. Trends in Biochemical Sciences 25 (6), 274–276. A.J. Bannister, P. Zegerman, J.F. Partridge, et al. (2001) Selective recognition of methylated lysine 9 on histone H3 by the HP1 chromo domain. Nature 410 (6824), 120–124. R. Bastow, J.S. Mylne, C. Lister, Z. Lippman, R.A. Martienssen and C. Dean (2004) Vernalization requires epigenetic silencing of FLC by histone methylation. Nature 427 (6970), 164–167. C. Beisel, A. Imhof, J. Greene, E. Kremmer and F. Sauer (2002) Histone methylation by the Drosophila epigenetic transcriptional regulator Ash1. Nature 419 (6909), 857–862. A. Birve, A.K. Sengupta, D. Beuchle, et al. (2001) Su(z)12, a novel Drosophila Polycomb group gene that is conserved in vertebrates and plants. Development 128 (17), 3371–3379. M.A. Blazquez, J.H. Ahn and D. Weigel (2003) A thermosensory pathway controlling flowering time in Arabidopsis thaliana. Nature Genetics 33 (2), 168–171. M.A. Bl´azquez, R. Green, O. Nilsson, M.R. Sussman and D. Weigel (1998) Gibberellins promote flowering of Arabidopsis by activating the LEAFY promoter. The Plant Cell 10 (5), 791–800. M.A. Bl´azquez and D. Weigel (2000) Integration of floral inductive signals in Arabidopsis. Nature 404 (6780), 889–892. K.L. Borden (1998) RING fingers and B-boxes: zinc-binding protein–protein interaction domains. Biochemistry and Cell Biology 76 (2–3), 351–358. R. Borner, G. Kampmann, J. Chandler, et al. (2000) A MADS domain gene involved in the transition to flowering in Arabidopsis. The Plant Journal 24 (5), 591–599. J.M. Casolari, C.R. Brown, D.A. Drubin, O.J. Rando and P.A. Silver (2005) Developmentally induced changes in transcriptional program alter spatial organization across chromosomes. Genes and Development 19 (10), 1188–1198. J.M. Casolari, C.R. Brown, S. Komili, J. West, H. Hieronymus and P.A. Silver (2004) Genomewide localization of the nuclear transport machinery couples transcriptional status and nuclear organization. Cell 117 (4), 427–439. Y. Chanvivattana, A. Bishopp, D. Schubert, et al. (2004) Interaction of Polycomb-group proteins controlling flowering in Arabidopsis. Development 131 (21), 5263–5276. L. Chen, J.C. Cheng, L. Castle and Z.R. Sung (1997) EMF genes regulate Arabidopsis inflorescence development. The Plant Cell 9 (11), 2011–2024. X.F. Cheng and Z.Y. Wang (2005) Overexpression of COL9, a CONSTANS-LIKE gene, delays flowering by reducing expression of CO and FT in Arabidopsis thaliana. The Plant Journal 43 (5), 758–768. K. Choi, S. Kim, S.Y. Kim, et al. (2005) SUPPRESSOR OF FRIGIDA3 encodes a nuclear actin-related protein6 required for floral repression in Arabidopsis. The Plant Cell 17 (10), 2647–2660. P.M. Clissold and C.P. Ponting (2001) JmjC: cupin metalloenzyme-like domains in jumonji, hairless and phospholipase A2beta. Trends in Biochemical Sciences 26 (1), 7–9. B. Czermin, R. Melfi, D. McCabe, V. Seitz, A. Imhof and V. Pirrotta (2002) Drosophila enhancer of Zeste/ESC complexes have a histone H3 methyltransferase activity that marks chromosomal Polycomb sites. Cell 111 (2), 185–196. R.B. Deal, M.K. Kandasamy, E.C. McKinney and R.B. Meagher (2005) The nuclear actin-related protein ARP6 is a pleiotropic developmental regulator required for the maintenance of FLOWERING LOCUS C expression and repression of flowering in Arabidopsis. The Plant Cell 17 (10), 2633–2646.
246
REGULATION OF TRANSCRIPTION IN PLANTS
A. Dill and T.-P. Sun (2001) Synergistic derepression of gibberellin signaling by removing RGA and GAI function in Arabidopsis thaliana. Genetics 159 (2), 777–785. S.C. Elgin (1996) Heterochromatin and gene regulation in Drosophila. Current Opinion in Genetics and Development 6 (2), 193–202. J.Y. Fan, F. Gordon, K. Luger, J.C. Hansen and D.J. Tremethick (2002) The essential histone variant H2A.Z regulates the equilibrium between different chromatin conformational states. Nature Structural Biology 9 (3), 172–176. S.D. Farris, E.D. Rubio, J.J. Moon, W.M. Gombert, B.H. Nelson and A. Krumm (2005) Transcriptioninduced chromatin remodeling at the c-myc gene involves the local exchange of histone H2A.Z. The Journal of Biological Chemistry 280 (26), 25298–25303. C. Ferrandiz, Q. Gu, R. Martienssen and M.F. Yanofsky (2000) Redundant regulation of meristem identity and plant architecture by FRUITFULL, APETALA1 and CAULIFLOWER. Development 127 (4), 725–734. R. Foster, T. Izawa and N.H. Chua (1994) Plant bZIP proteins gather at ACGT elements. FASEB Journal 8 (2), 192–200. V. Gaudin, M. Libault, S. Pouteau, et al. (2001) Mutations in LIKE HETEROCHROMATIN PROTEIN 1 affect flowering time and plant architecture in Arabidopsis. Development 128 (23), 4847– 4858. A.R. Gendall, Y.Y. Levy, A. Wilson and C. Dean (2001) The VERNALIZATION 2 gene mediates the epigenetic regulation of vernalization in Arabidopsis. Cell 107 (4), 525–535. Gill, G. (2004) SUMO and ubiquitin in the nucleus: different functions, similar mechanisms? Genes and Development 18 (17), 2046–2059. G.F.W. Gocal, C.C. Sheldon, F. Gubler, et al. (2001) GAMYB-like genes, flowering, and gibberellin signalling in Arabidopsis. Plant Physiology 127 (4), 1682–1693. C. Gomez-Mena, M. Pineiro, J.M. Franco-Zorrilla, J. Salinas, G. Coupland and J.M. Martinez-Zapater (2001) Early bolting in short days: an Arabidopsis mutation that causes early flowering and partially suppresses the floral phenotype of leafy. The Plant Cell 13 (5), 1011–1024. J. Goodrich, P. Puangsomlee, M. Martin, D. Long, E.M. Meyerowitz and G. Coupland (1997) Polycomb-group gene regulates homeotic gene expression in Arabidopsis. Nature 386 (6620), 44–51. O. Gozani, P. Karuman, D.R. Jones, et al. (2003) The PHD finger of the chromatin-associated protein ING2 functions as a nuclear phosphoinositide receptor. Cell 114 (1), 99–111. M.G. Guenther, R.G. Jenner, B. Chevalier, et al. (2005) Global and Hox-specific roles for the MLL1 methyltransferase. Proceedings of the National Academy of Sciences 102 (24), 8603–8608. M. Hampsey and D. Reinberg (2003) Tails of intrigue: phosphorylation of RNA polymerase II mediates histone methylation. Cell 113 (4), 429–432. U. Hartmann, S. Hohmann, K. Nettesheim, E. Wisman, H. Saedler and P. Huijser (2000) Molecular cloning of SVP: a negative regulator of the floral transition in Arabidopsis. The Plant Journal 21 (4), 351–360. R. Hayama and G. Coupland (2003) Shedding light on the circadian clock and the photoperiodic control of flowering. Current Opinion in Plant Biology 6 (1), 13–19. R. Hayama and G. Coupland (2004) The molecular basis of diversity in the photoperiodic flowering responses of Arabidopsis and rice. Plant Physiology 135 (2), 677–684. Y. He, M.R. Doyle and R.M. Amasino (2004) PAF1-complex-mediated histone methylation of FLOWERING LOCUS C chromatin is required for the vernalization-responsive, winter-annual habit in Arabidopsis. Genes and Development 18 (22), 2774–2784. Y. He, S.D. Michaels and R.M. Amasino (2003) Regulation of flowering time by histone acetylation in Arabidopsis. Science 302 (5651), 1751–1754. F.D. Hempel, D. Weigel, M.A. Mandel, et al. (1997) Floral determination and expression of floral regulatory genes in Arabidopsis. Development 124 (19), 3845–3853. I.R. Henderson, F. Liu, S. Drea, G.G. Simpson and C. Dean (2005) An allelic series reveals essential roles for FY in plant development in addition to flowering-time control. Development 132 (16), 3597–3607.
CONTROL OF FLOWERING TIME
247
K.W. Henry, A. Wyce, W.S. Lo, et al. (2003) Transcriptional activation via sequential histone H2B ubiquitylation and deubiquitylation, mediated by SAGA-associated Ubp8. Genes and Development 17 (21), 2648–2663. S.R. Hepworth, F. Valverde, D. Ravenscroft, A. Mouradov and G. Coupland (2002) Antagonistic regulation of flowering-time gene SOC1 by CONSTANS and FLC via separate promoter motifs. The EMBO Journal 21 (16), 4327–4337. R.S. Hilgarth, L.A. Murphy, H.S. Skaggs, D.C. Wilkerson, H. Xing and K.D. Sarge (2004) Regulation and function of SUMO modification. The Journal of Biological Chemistry 279 (52), 53899– 53902. T. Huang, H. Bohlenius, S. Eriksson, F. Parcy and O. Nilsson (2005) The mRNA of the Arabidopsis gene FT moves from leaf to shoot apex and induces flowering. Science 309 (5741), 1694–1696. T. Imaizumi, T.F. Schultz, F.G. Harmon, L.A. Ho and S.A. Kay (2005) FKF1 F-box protein mediates cyclic degradation of a repressor of CONSTANS in Arabidopsis. Science 309 (5732), 293–297. M. Jakoby, B. Weisshaar, W. Droge-Laser, et al. (2002) bZIP transcription factors in Arabidopsis. Trends in Plant Science 7 (3), 106–111. U. Johanson, J. West, C. Lister, S. Michaels, R. Amasino and C. Dean (2000) Molecular analysis of FRIGIDA, a major determinant of natural variation in Arabidopsis flowering time. Science 290 (5490), 344–347. J. Jung, T.G. Kim, G.E. Lyons, H.R. Kim and Y. Lee (2005) Jumonji regulates cardiomyocyte proliferation via interaction with retinoblastoma protein. The Journal of Biological Chemistry 280 (35), 30916–30923. M.H. Kagey, T.A. Melhuish and D. Wotton (2003) The polycomb protein Pc2 is a SUMO E3. Cell 113 (1), 127–137. A. Katz, M. Oliva, A. Mosquna, O. Hakim and N. Ohad (2004) FIE and CURLY LEAF polycomb proteins interact in the regulation of homeobox gene expression during sporophyte development. The Plant Journal 37 (5), 707–719. C.S. Ketel, E.F. Andersen, M.L. Vargas, J. Suh, S. Strome and J.A. Simon (2005) Subunit contributions to histone methyltransferase activities of fly and worm polycomb group complexes. Molecular and Cellular Biology 25 (16), 6857–6868. S.Y. Kim, Y. He, Y. Jacob, Y.S. Noh, S. Michaels and R. Amasino (2005) Establishment of the vernalization-responsive winter-annual habit in Arabidopsis requires a putative histone H3 methyl transferase. The Plant Cell 17, 3301–3310. T.G. Kim, J.C. Kraus, J. Chen and Y. Lee (2003) JUMONJI, a critical factor for cardiac development, functions as a transcriptional repressor. The Journal of Biological Chemistry 278 (43), 42247– 42255. T. Kinoshita, J.J. Harada, R.B. Goldberg and R.L. Fischer (2001) Polycomb repression of flowering during early plant development. Proceedings of the National Academy of Sciences 98 (24), 14156– 14161. T. Klymenko and J. Muller (2004) The histone methyltransferases Trithorax and Ash1 prevent transcriptional silencing by Polycomb group proteins. EMBO Reports 5 (4), 373–377. Y. Kobayashi, H. Kaya, K. Goto, M. Iwabuchi and T. Araki (1999) A pair of related genes with antagonistic roles in mediating flowering signals. Science 286 (5446), 1960–1962. M.S. Kobor, S. Venkatasubrahmanyam, M.D. Meneghini, et al. (2004) A protein complex containing the conserved Swi2/Snf2-related ATPase Swr1p deposits histone variant H2A.Z into euchromatin. PLoS Biology 2 (5), E131. T. Kotake, S. Takada, K. Nakahigashi, M. Ohto and K. Goto (2003) Arabidopsis TERMINAL FLOWER 2 gene encodes a heterochromatin protein 1 homolog and represses both FLOWERING LOCUS T to regulate flowering time and several floral homeotic genes. Plant and Cell Physiology 44 (6), 555–564. A. Kouskouti and I. Talianidis (2005) Histone modifications defining active genes persist after transcriptional and mitotic inactivation. The EMBO Journal 24 (2), 347–357. N.J. Krogan, M.C. Keogh, N. Datta, et al. (2003) A Snf2 family ATPase complex required for recruitment of the histone H2A variant Htz1. Molecular Cell 12 (6), 1565–1576.
248
REGULATION OF TRANSCRIPTION IN PLANTS
N.J. Krogan, M. Kim, S.H. Ahn, et al. (2002) RNA polymerase II elongation factors of Saccharomyces cerevisiae: a targeted proteomics approach. Molecular and Cellular Biology 22 (20), 6979– 6992. A. Kuzmichev, K. Nishioka, H. Erdjument-Bromage, P. Tempst and D. Reinberg (2002) Histone methyltransferase activity associated with a human multiprotein complex containing the Enhancer of Zeste protein. Genes and Development 16 (22), 2893–2905. N. Lauter, A. Kampani, S. Carlson, M. Goebel and S.P. Moose (2005) microRNA172 down-regulates glossy15 to promote vegetative phase change in maize. Proceedings of the National Academy of Sciences 102 (26), 9412–9417. S. Ledger, C. Strayer, F. Ashton, S.A. Kay and J. Putterill (2001) Analysis of the function of two circadian-regulated CONSTANS-LIKE genes. The Plant Journal 26 (1), 15–22. I. Lee, M.J. Aukerman, S.L. Gore, et al. (1994) Isolation of LUMINIDEPENDENS: a gene involved in the control of flowering time in Arabidopsis. The Plant Cell 6 (1), 75–83. I. Lee, A. Bleecker and R.M. Amasino (1993) Analysis of naturally occurring late flowering in Arabidopsis thaliana. Molecular and General Genetics 237 (1–2), 171–176. Y. Lee, A.J. Song, R. Baker, B. Micales, S.J. Conway and G.E. Lyons (2000a) Jumonji, a nuclear protein that is necessary for normal heart development. Circulation Research 86 (9), 932–938. H. Lee, S.S. Suh, E. Park, et al. (2000b) The AGAMOUS-LIKE 20 MADS domain protein integrates floral inductive pathways in Arabidopsis. Genes and Development 14 (18), 2366–2376. Y.Y. Levy, S. Mesnage, J.S. Mylne, A.R. Gendall and C. Dean (2002) Multiple roles of Arabidopsis VRN1 in vernalization and flowering time control. Science 297 (5579), 243–246. M.H. Lim, J. Kim, Y.S. Kim, et al. (2004) A new Arabidopsis gene, FLK, encodes an RNA binding protein with K homology motifs and regulates flowering time via FLOWERING LOCUS C. The Plant Cell 16 (3), 731–740. R. Macknight, I. Bancroft, T. Page, et al. (1997) FCA, a gene controlling flowering time in Arabidopsis, encodes a protein containing RNA-binding domains. Cell 89 (5), 737–745. J. Marx (2004) Plant biology. Remembrance of winter past. Science 303 (5664), 1607. S.D. Michaels and R.M. Amasino (1999) FLOWERING LOCUS C encodes a novel MADS domain protein that acts as a repressor of flowering. The Plant Cell 11 (5), 949–956. S.D. Michaels and R.M. Amasino (2001) Loss of FLOWERING LOCUS C activity eliminates the lateflowering phenotype of FRIGIDA and autonomous pathway mutations but not responsiveness to vernalization. The Plant Cell 13 (4), 935–941. S.D. Michaels, I.C. Bezerra and R.M. Amasino (2004) FRIGIDA-related genes are required for the winter-annual habit in Arabidopsis. Proceedings of the National Academy of Sciences 101 (9), 3281–3285. S.D. Michaels, E. Himelblau, S.Y. Kim, F.M. Schomburg and R.M. Amasino (2005) Integration of flowering signals in winter-annual Arabidopsis. Plant Physiology 137 (1), 149–156. T. Misteli (2005) Concepts in nuclear architecture. Bioessays 27 (5), 477–487. (review) Y. Mito, J.G. Henikoff and S. Henikoff (2005) Genome-scale profiling of histone H3.3 replacement patterns. Nature Genetics 37 (10), 1090–1097. G. Mizuguchi, X. Shen, J. Landry, W.H. Wu, S. Sen and C. Wu (2004) ATP-driven exchange of histone H2AZ variant catalyzed by SWR1 chromatin remodeling complex. Science 303 (5656), 343–348. T.C. Mockler, X. Yu, D. Shalitin, et al. (2004) Regulation of flowering time in Arabidopsis by K homology domain proteins. Proceedings of the National Academy of Sciences 101 (34), 12759– 12764. J. Moon, H. Lee, M. Kim and I. Lee (2005) Analysis of flowering pathway integrators in Arabidopsis. Plant and Cell Physiology 46 (2), 292–299. J. Moon, S.S. Suh, H. Lee, et al. (2003a) The SOC1 MADS-box gene integrates vernalization and gibberellin signals for flowering in Arabidopsis. The Plant Journal 35 (5), 613–623. (erratum in The Plant Journal 37 (3), 459). Y.H. Moon, L. Chen, R.L. Pan, et al. (2003b) EMF genes maintain vegetative development by repressing the flower program in Arabidopsis. The Plant Cell 15 (3), 681–693. (erratum in The Plant Cell 15 (5), 1257).
CONTROL OF FLOWERING TIME
249
J. Motoyama, K. Kitajima, M. Kojima, S. Kondo and T. Takeuchi (1997) Organogenesis of the liver, thymus and spleen is affected in jumonji mutant mice. Mechanisms of Development 66 (1–2), 27–37. C.L. Mueller and J.A. Jaehning (2002) Ctr9, Rtf1, and Leo1 are components of the Paf1/RNA polymerase II complex. Molecular and Cellular Biology 22 (7), 1971–1980. J. Muller, C.M. Hart, N.J. Francis, et al. (2002) Histone methyltransferase activity of a Drosophila Polycomb group repressor complex. Cell 111 (2), 197–208. G. Murtas, P.H. Reeves, Y.F. Fu, I. Bancroft, C. Dean and G. Coupland (2003) A nuclear protease required for flowering-time regulation in Arabidopsis reduces the abundance of SMALL UBIQUITIN-RELATED MODIFIER conjugates. The Plant Cell 15 (10), 2308–2319. C. Mussig and T. Altmann (2003) Changes in gene expression in response to altered SHL transcript levels. Plant Molecular Biology 53 (6), 805–820. C. Mussig, A. Kauschmann, S.D. Clouse and T. Altmann (2000) The Arabidopsis PHD-finger protein SHL is required for proper development and fertility. Molecular and General Genetics 264 (4), 363–370. K. Nakahigashi, Z. Jasencakova, I. Schubert and K. Goto (2005) The Arabidopsis HETEROCHROMATIN PROTEIN1 homolog (TERMINAL FLOWER2) silences genes within euchromatic region but not genes positioned in heterochromatin. Plant and Cell Physiology 46, 1747–1756. J. Nakayama, J.C. Rice, B.D. Strahl, C.D. Allis and S.I. Grewal (2001) Role of histone H3 lysine 9 methylation in epigenetic control of heterochromatin assembly. Science 292 (5514), 110–113. M. Nekrasov, B. Wild and J. Muller (2005) Nucleosome binding and histone methyltransferase activity of Drosophila PRC2. EMBO Reports 6 (4), 348–353. H.H. Ng, S. Dole and K. Struhl (2003a). The Rtf1 component of the Paf1 transcriptional elongation complex is required for ubiquitination of histone H2B. The Journal of Biological Chemistry 278 (36), 33625–33628. H.H. Ng, F. Robert, R.A. Young and K. Struhl (2003b). Targeted recruitment of Set1 histone methylase by elongating PolII provides a localized mark and memory of recent transcriptional activity. Molecular Cell 11 (3), 709–719. S.J. Nielsen, R. Schneider, U.M. Bauer, et al. (2001) Rb targets histone H3 methylation and HP1 to promoters. Nature 412 (6846), 561–565. B. Noh, S.H. Lee, H.J. Kim, et al. (2004) Divergent roles of a pair of homologous jumonji/zinc-fingerclass transcription factor proteins in the regulation of Arabidopsis flowering time. The Plant Cell 16 (10), 2601–2613. Y.S. Noh and R.M. Amasino (2003) PIE1, an ISWI family gene, is required for FLC activation and floral repression in Arabidopsis. The Plant Cell 15 (7), 1671–1682. S. Oh, H. Zhang, P. Ludwig and S. van Nocker (2004) A mechanism related to the yeast transcriptional regulator Paf1c is required for expression of the Arabidopsis FLC/MAF MADS box gene family. The Plant Cell 16 (11), 2940–2953. H. Onouchi, M.I. Igeno, C. Perilleux, K. Graves and G. Coupland (2000) Mutagenesis of plants overexpressing CONSTANS demonstrates novel interactions among Arabidopsis flowering-time genes. The Plant Cell 12 (6), 885–900. R. Pandey, A. Muller, C.A. Napoli, et al. (2002) Analysis of histone acetyltransferase and histone deacetylase families of Arabidopsis thaliana suggests functional diversification of chromatin modification among multicellular eukaryotes. Nucleic Acids Research 30 (23), 5036–5055. V.G. Panse, B. Kuster, T. Gerstberger and E. Hurt (2003) Unconventional tethering of Ulp1 to the transport channel of the nuclear pore complex by karyopherins. Nature Cell Biology 5 (1), 21–27. M. Pi˜neiro, C. Gomez-Mena, R. Schaffer, J.M. Martinez-Zapater and G. Coupland (2003) EARLY BOLTING IN SHORT DAYS is related to chromatin remodeling factors and regulates flowering in Arabidopsis by repressing FT. The Plant Cell 15 (7), 1552–1562. A.V. Probst, M. Fagard, F. Proux, et al. (2004) Arabidopsis histone deacetylase HDA6 is required for maintenance of transcriptional gene silencing and determines nuclear organization of rDNA repeats. The Plant Cell 16 (4), 1021–1034.
250
REGULATION OF TRANSCRIPTION IN PLANTS
J. Putterill, F. Robson, K. Lee, R. Simon and G. Coupland (1995) The CONSTANS gene of Arabidopsis promotes flowering and encodes a protein showing similarities to zinc finger transcription factors. Cell 80 (6), 847–857. V. Quesada, C. Dean and G.G. Simpson (2005) Regulated RNA processing in the control of Arabidopsis flowering. The International Journal of Developmental Biology 49 (5–6), 773–780. V. Quesada, R. Macknight, C. Dean and G.G. Simpson (2003) Autoregulation of FCA pre-mRNA processing controls Arabidopsis flowering time. The EMBO Journal 22 (12), 3142–3152. O.J. Ratcliffe, R.W. Kumimoto, B.J. Wong and J.L. Riechmann (2003) Analysis of the Arabidopsis MADS AFFECTING FLOWERING gene family: MAF2 prevents vernalization by short periods of cold. The Plant Cell 15 (5), 1159–1169. O.J. Ratcliffe, G.C. Nadzan, T.L. Reuber and J.L. Riechmann (2001) Regulation of flowering in Arabidopsis by an FLC homologue. Plant Physiology 126 (1), 122–132. S. Rea, F. Eisenhaber, D. O’Carroll, et al. (2000) Regulation of chromatin structure by site-specific histone H3 methyltransferases. Nature 406 (6796), 593–599. L. Ringrose and R. Paro (2004) Epigenetic regulation of cellular memory by the Polycomb and Trithorax group proteins. Annual Review of Genetics 38, 413–443. S. Ross, J.L. Best, L.I. Zon and G. Gill (2002) SUMO-1 modification represses Sp3 transcriptional activation and modulates its subnuclear localization. Molecular Cell 10 (4), 831–842. P.A. Salome and C.R. McClung (2004) The Arabidopsis thaliana clock. Journal of Biological Rhythms 19 (5), 425–435. A. Samach, H. Onouchi, S.E. Gold, et al. (2000) Distinct roles of CONSTANS target genes in reproductive development of Arabidopsis. Science 288 (5471), 1613–1616. H. Santos-Rosa, R. Schneider, A.J. Bannister, et al. (2002) Active genes are tri-methylated at K4 or histone H3. Nature 419 (6905), 407–411. H. Santos-Rosa, R. Schneider, B.E. Bernstein, et al. (2003) Methylation of histone H3 K4 mediates association of the Isw1p ATPase with chromatin. Molecular Cell 12 (5), 1325–1332. M. Schmid, N.H. Uhlenhaut, F. Godard, et al. (2003) Dissection of floral induction pathways using global expression analysis. Development 130 (24), 6001–6012. F.M. Schomburg, D.A. Patton, D.W. Meinke and R.M. Amasino (2001) FPA, a gene involved in floral induction in Arabidopsis, encodes a protein containing RNA-recognition motifs. The Plant Cell 13 (6), 1427–1436. K. Scortecci, S.D. Michaels and R.M. Amasino (2003) Genetic interactions between FLM and other flowering-time genes in Arabidopsis thaliana. Plant Molecular Biology 52 (5), 915–922. K.C. Scortecci, S.D. Michaels and R.M. Amasino (2001) Identification of a MADS-box gene, FLOWERING LOCUS M, that represses flowering. The Plant Journal 26 (2), 229–236. I. Searle and G. Coupland (2004) Induction of flowering by seasonal changes in photoperiod. The EMBO Journal 23 (6), 1217–1222. C.C. Sheldon, J.E. Burn, P.P. Perez, et al. (1999) The FLF MADS box gene: a repressor of flowering in Arabidopsis regulated by vernalization and methylation. The Plant Cell 11 (3), 445–458. C.C. Sheldon, A.B. Conn, E.S. Dennis and W.J. Peacock (2002) Different regulatory regions are required for the vernalization-induced repression of FLOWERING LOCUS C and for the epigenetic maintenance of repression. The Plant Cell 14 (10), 2527–2537. Y. Shi, F. Lan, C. Matson, et al. (2004) Histone demethylation mediated by the nuclear amine oxidase homolog LSD1. Cell 119 (7), 941–953. Y.J. Shi, J.-I. Sawada, G.C. Sui, et al. (2003) Coordinated histone modifications mediated by a CtBP co-repressor complex. Nature 422 (6933), 735–738. Y. Shiio and R.N. Eisenman (2003) Histone sumoylation is associated with transcriptional repression. Proceedings of the National Academy of Sciences 100 (23), 13225–13230. R. Simic, D.L. Lindstrom, H.G. Tran, et al. (2003) Chromatin remodeling protein Chd1 interacts with transcription elongation factors and localizes to transcribed genes. The EMBO Journal 22 (8), 1846–1856. R. Simon, M.I. Igeno and G. Coupland (1996) Activation of floral meristem identity genes in Arabidopsis. Nature 384 (6604), 59–62.
CONTROL OF FLOWERING TIME
251
G.G. Simpson, P.P. Dijkwel, V. Quesada, I. Henderson and C. Dean (2003) FY is an RNA 3 endprocessing factor that interacts with FCA to control the Arabidopsis floral transition. Cell 113 (6), 777–787. G.G. Simpson, V. Quesada, I.R. Henderson, P.P. Dijkwel, R. Macknight and C. Dean (2004) RNA processing and Arabidopsis flowering time control. Biochemical Society Transactions 32 (Pt 4), 565–566. W.J. Soppe, L. Bentsink and M. Koornneef (1999) The early-flowering mutant efs is involved in the autonomous promotion pathway of Arabidopsis thaliana. Development 126 (21), 4763–4770. C. Strayer, T. Oyama, T.F. Schultz, et al. (2000) Cloning of the Arabidopsis clock gene TOC1, an autoregulatory response regulator homolog. Science 289 (5480), 768–771. P. Suarez-Lopez, K. Wheatley, F. Robson, H. Onouchi, F. Valverde and G. Coupland (2001) CONSTANS mediates between the circadian clock and the control of flowering in Arabidopsis. Nature 410 (6832), 1116–1120. T.P. Sun and F. Gubler (2004) Molecular mechanism of gibberellin signaling in plants. Annual Review of Plant Biology 55, 197–222. S. Sung and R.M. Amasino (2004) Vernalization in Arabidopsis thaliana is mediated by the PHD finger protein VIN3. Nature 427 (6970), 159–164. Z.R. Sung, A. Belachew, S. Bai and R. Bertrand-Garcia (1992) EMF, an Arabidopsis gene required for vegetative shoot development. Science 258 (5088), 1645–1647. S. Takada and K. Goto (2003) Terminal flower2, an Arabidopsis homolog of heterochromatin protein1, counteracts the activation of flowering locus T by constans in the vascular tissues of leaves to regulate flowering time. The Plant Cell 15 (12), 2856–2865. Y. Takahashi, J. Mizoi, A. Toh-E and Y. Kikuchi (2000) Yeast Ulp1, an Smt3-specific protease, associates with nucleoporins. Journal of Biochemistry 128 (5), 723–725. T. Takeuchi, Y. Yamazaki, Y. Katoh-Fukui, et al. (1995) Gene trap capture of a novel mouse gene, jumonji, required for neural tube formation. Genes and Development 9 (10), 1211–1222. P. Teper-Bamnolker and A. Samach (2005) The flowering integrator FT regulates SEPALLATA3 and FRUITFULL accumulation in Arabidopsis leaves. The Plant Cell 17 (10), 2661–2675. L. Tian and Z.J. Chen (2001) Blocking histone deacetylation in Arabidopsis induces pleiotropic effects on plant gene regulation and development. Proceedings of the National Academy of Sciences 98 (1), 200–205. L. Tian, J. Wang, M.P. Fong, et al. (2003) Genetic control of developmental changes induced by disruption of Arabidopsis histone deacetylase 1 (AtHD1) expression. Genetics 165 (1), 399–409. F. Tie, T. Furuyama, J. Prasad-Sinha, E. Jane and P.J. Harte (2001) The Drosophila Polycomb group proteins ESC and E(Z) are present in a complex containing the histone-binding protein p55 and the histone deacetylase RPD3. Development 128 (2), 275–286. J.K. Tong, C.A. Ha, G.R. Schnitzler, R.E. Kingston and S.L. Schreiber (1998) Chromatin deacetylation by an ATP-dependent nucleosome remodelling complex. Nature 395 (6705), 917–921. S.C. Trewick, P.J. McLaughlin and R.C. Allshire (2005) Methylation: lost in hydroxylation? EMBO Reports 6 (4), 315–320. W. van Leeuwen, L. Okresz, L. Bogre and T. Munnik (2004) Learning the lipid language of plant signaling. Trends in Plant Science 9 (8), 378–384. S. van Nocker (2003) CAF-1 and MSI1-related proteins: linking nucleosome assembly with epigenetics. Trends in Plant Science 8 (10), 471–473. A. Verger, J. Perdomo and M. Crossley (2003) Modification with SUMO. A role in transcriptional regulation. EMBO Reports 4 (2), 137–142. P.A. Wigge, M.C. Kim, K.E. Jaeger, et al. (2005) Integration of spatial and temporal information during floral induction in Arabidopsis. Science 309 (5737), 1056–1059. R.N. Wilson, J.W. Heckman and C.R. Sommerville (1992) Gibberellin is required for flowering in Arabidopsis thaliana under short days. Plant Physiology 100 (1), 403–408. A. Wood, J. Schneider, J. Dover, M. Johnston and A. Shilatifard (2003) The Paf1 complex is essential for histone monoubiquitination by the Rad6–Bre1 complex, which signals for histone methylation by COMPASS and Dot1p. The Journal of Biological Chemistry 278 (37), 34739–34742.
252
REGULATION OF TRANSCRIPTION IN PLANTS
K. Wu, K. Malik, L. Tian, D. Brown and B. Miki (2000) Functional analysis of a RPD3 histone deacetylase homologue in Arabidopsis thaliana. Plant Molecular Biology 44 (2), 167–176. A. Yamaguchi, Y. Kobayashi, K. Goto, M. Abe and T. Araki (2005) TWIN SISTER OF FR (TSF), a new member of floral pathway integrators (Abstract 305). In: The 16th International Conference on Arabidopsis Research, Madison, WI. K. Yamamoto, M. Sonoda, J. Inokuchi, S. Shirasawa and T. Sasazuki (2004) Polycomb group suppressor of zeste 12 links heterochromatin protein 1alpha and enhancer of zeste 2. The Journal of Biological Chemistry 279 (1), 401–406. C.-H. Yang, L.J. Chen and Z.R. Sung (1995) Genetic regulation of shoot development in Arabidopsis: role of the EMF genes. Developmental Biology 169 (2), 421–435. N. Yoshida, Y. Yanai, L. Chen, et al. (2001) EMBRYONIC FLOWER2, a novel polycomb group protein homolog, mediates shoot development and flowering in Arabidopsis. The Plant Cell 13 (11), 2471–2481. A. You, J.K. Tong, C.M. Grozinger and S.L. Schreiber (2001) CoREST is an integral component of the CoREST-human histone deacetylase complex. Proceedings of the National Academy of Sciences 98 (4), 1454–1458. H. Yu, T. Ito, Y. Zhao, J. Peng, P. Kumar and E.M. Meyerowitz (2004) Floral homeotic genes are targets of gibberellin signaling in flower development. Proceedings of the National Academy of Sciences 101 (20), 7827–7832. Y. Zhang, R. Iratni, H. Erdjument-Bromage, P. Tempst and D. Reinberg (1997) Histone deacetylases and SAP18, a novel polypeptide, are components of a human Sin3 complex. Cell 89 (3), 357–364. Y. Zhang, G. LeRoy, H.P. Seelig, W.S. Lane and D. Reinberg (1998) The dermatomyositis-specific autoantigen Mi2 is a component of a complex containing histone deacetylase and nucleosome remodeling activities. Cell 95 (2), 279–289. H. Zhang, C. Ransom, P. Ludwig and S. van Nocker (2003) Genetic analysis of early flowering mutants in Arabidopsis defines a class of pleiotropic developmental regulator required for expression of the flowering-time switch flowering locus C. Genetics 164 (1), 347–358. H. Zhang and S. van Nocker (2002) The VERNALIZATION INDEPENDENCE 4 gene encodes a novel regulator of FLOWERING LOCUS C. The Plant Journal 31 (5), 663–673.
10 Combinatorial control of floral organ identity by MADS-domain transcription factors G¨unter Theißen and Rainer Melzer
10.1
Introduction
The involvement of multiprotein complexes that bind DNA in a sequence-specific manner and either repress or activate transcription is a recurring theme in studies of eukaryotic gene regulation (Wolberger, 1998). Concerning plants, MIKC-type MADS-domain proteins involved in flower development are a good case in point. The transcription factor complexes that specify floral organ identity have been comparatively well characterized during recent years. In this chapter, we focus on the major model system, thale cress Arabidopsis thaliana (henceforth termed Arabidopsis), with side views on other plants only when necessary for a general understanding.
10.2
ABC: early genetic models of floral organ identity
The distinct identities of the different floral organs confer flowers their unique appearance. A typical flower of a eudicot, which is the largest group of flowering plants (angiosperms), consists of four different organ classes arranged in four whorls at the tip of a floral shoot. The first, outermost whorl often consists of green, leaflike sepals. The second whorl is composed of usually colored and showy petals. The third whorl contains the stamens, i.e., the male reproductive organs, which produce the pollen. Finally, the fourth, innermost whorl contains the carpels, i.e., the female reproductive organs, which are often fused and inside which the ovules and seeds develop. Although the appearance of sepals, petals, stamens, and carpels may differ dramatically at maturity, each floral organ starts its development as a little bulge on a tiny clump of undifferentiated cells called ‘floral meristem’. Thus, each cell in the developing floral organ primordia must somehow ‘learn’ its position within the flower, and differentiate accordingly into a cell type that is appropriate for the specific organ. Almost two decades ago the first genetic model was suggested to explain how homeotic selector genes specify organ identity during flower development (Haughn and Somerville, 1988). This ‘early ABC model’ was based on the study of homeotic mutants in which the identity of floral organs is changed. In Arabidopsis such mutants come in three classes: A, B, and C (Figure 10.1). Ideal class A mutants have carpels in the first whorl instead of sepals, and stamens in the second whorl instead of petals.
254
REGULATION OF TRANSCRIPTION IN PLANTS
Figure 10.1 Floral homeotic mutants and the ABCDE model of floral organ identity of Arabidopsis. The photographs show flower phenotypes of loss-of-function mutants of floral homeotic genes. ap1-1 is a class A mutant affected in the APETALA1 (AP1) gene that has sepals transformed into carpelloid organs and petals transformed into stamenoid organs. ap3-3 and pi-1 are class B mutants affected in the APETALA3 (AP3) and PISTILLATA (PI) genes, respectively, which have petals transformed into sepaloid organs and stamens transformed into carpelloid organs. ag-1 is a class C mutant that has stamens transformed into petaloid organs; carpels are replaced by another mutant flower due to the loss of determinate growth of the floral shoot, and so the number of floral organs is vastly increased in these ‘filled flowers’. stk shp1 shp2 is a class D triple mutant in the genes SEEDSTICK (STK), SHATTERPROOF1 (SHP1), and SHATTERPROOF2 (SHP2), in which ovules are transformed into carpelloid or vegetative leaf-like organs. sep1 sep2 sep3 sep4 is a class E quadruple mutant affected in the genes SEPALLATA1 (SEP1), SEPALLATA2 (SEP2), SEPALLATA3 (SEP3), and SEPALLATA4 (SEP4), in which all floral organs develop like vegetative leaves and the number of floral organs is vastly increased due to a loss of determinate growth of the floral shoot. The graphic shows the ABCDE model of flower organ identity, which is based on the mutant phenotypes shown in the photographs. Corresponding to the five classes of mutants five classes of homeotic genes are postulated, with class A + E genes specifying sepals in whorl 1 of the flower; the combined activities of A + B + E specify petals in whorl 2; B + C + E specify stamens in whorl 3; C + E specify carpels in whorl 4; D + E specify ovules that develop within carpels. The activities A and C are mutually antagonistic: A prevents the activity of C in whorls 1 and 2, and C prevents the activity of A in whorls 3 and 4. (Photographs of ap1-1, ap3-3, pi-1, and ag-1 are reproduced from Riechmann and Meyerowitz (1997) with kind permission by Walter de Gruyter and the authors, stk shp1 shp2 (modified) is from Pinyopich et al. (2003) with kind permission from Macmillan Publishers Ltd, and sep1 sep2 sep3 sep4 is from Ditta et al. (2004) with kind permission from Elsevier.)
COMBINATORIAL CONTROL OF FLORAL ORGAN IDENTITY
255
Class B mutants have sepals rather than petals in the second whorl, and carpels rather than stamens in the third whorl. Class C mutants have petals instead of stamens in the third whorl, and sepals rather than carpels in the fourth whorl. In addition, the flowers of these mutants grow indeterminately, i.e., there is continued production of mutant floral organs inside the fourth whorl. The phenotype of class A, B, and C mutants suggested that development of the flower is sculpted by homeotic selector genes, aptly termed ‘floral organ identity genes’, whose expression gives the different floral organs their unique identity. Homeotic selector genes can be considered as acting as major developmental switches that activate the entire genetic program required for the development of a particular organ, and repress genes whose activity would be unnecessary or even detrimental for organ development. The early ABC model proposed that the state of expression (‘on’ or ‘off’) of three different genes a, b, and c specifies the identity of the four different floral organs. (Haughn and Somerville, 1988). The proposed states of expression of genes a, b, c in the different floral organs are ‘on, off, off’ (sepals), ‘on, on, off’ (petals), ‘off, on, on’ (stamens), and ‘off, off, on’ (carpels). If a, b, and c gene expression is off, leaves develop. (Haughn and Somerville, 1988). Not much later, a more elaborate and now widely known model, termed ‘classical ABC model’ here and elsewhere (Theißen, 2001), was proposed to explain how homeotic genes control floral organ identity (Figure 10.1) (Coen and Meyerowitz, 1991). Similar to the early ABC model, the classical ABC model maintains that organ identity in each whorl is determined by a unique combination of three organ identity gene activities, called A, B, and C. Expression of A alone specifies sepal formation. The combination AB specifies the development of petals, and the combination BC specifies the formation of stamens. Expression of C alone determines the development of carpels. In contrast to the early ABC model, however, there can be more than one gene behind the different gene activities (‘floral homeotic functions’) A, B, and C. Moreover, in order to explain the three classes of floral homeotic mutants, the classical ABC model proposes that the A- and C-function genes negatively regulate each other; that is, the C-function becomes expressed throughout the flower when the A-function is mutated, and vice versa (for reviews of the ABC model, see Weigel and Meyerowitz, 1994; Theißen, 2001). Arabidopsis genes providing the three homeotic activities A, B, and C are known. The A-function is contributed by two different genes, APETALA1 (AP1) and APETALA2 (AP2), the B-function also by two genes, APETALA3 (AP3) and PISTILLATA (PI), and the C-function by just one gene, AGAMOUS (AG). Molecular cloning of these genes revealed that they all encode putative transcription factors (Yanofsky et al., 1990; Jack et al., 1992; Mandel et al., 1992; Goto and Meyerowitz, 1994; Jofuku et al., 1994; for a review, see Weigel and Meyerowitz, 1994; Theißen, 2001). Thus, the products of the ABC genes probably all control the transcription of other genes (‘target genes’) whose products are directly or indirectly involved in the formation or function of floral organs. Except for AP2, all ABC genes belong to a special class of genes that is restricted to green plants, termed ‘MIKC-type MADS-box genes’ (M¨unster et al., 1997; Kaufmann et al., 2005b). These are named so because they encode MADS-domain transcription factors with a characteristic
256
REGULATION OF TRANSCRIPTION IN PLANTS
structural organization, including, from N- to C-terminus, a MADS (M-), intervening (I-), keratin-like (K-), and C-terminal (C-) domain (Theißen et al., 1996; Kaufmann et al., 2005b). The name-giving MADS domain is by far the most highly conserved region of all kinds of MADS-domain proteins. The MADS domain is the major determinant of DNA binding, but it also performs functions in dimerization of MADS-domain proteins and in accessory factor binding. MADS-domain proteins bind to DNA sites based on the consensus sequence 5 -CC(A/T)6 GG-3 , termed a ‘CArG box’ (for ‘CC-A rich-GG’). The I-domain, located directly downstream of the MADS domain, is quite variable in length and only relatively weakly conserved. It may constitute a key molecular determinant for the selective formation of DNA-binding dimers, at least in some MIKC-type proteins. The K-domain is characterized by a conserved, regular spacing of hydrophobic residues, which is proposed to allow for the formation of an amphipathic helix. It is assumed that such an amphipathic helix interacts with that of another K-domain containing protein to promote protein dimerization or multimeric complex formation. The most variable region is the C-domain at the C-terminal end of the MADS-domain proteins. In some MADSdomain proteins it is involved in transcriptional activation, or in the formation of multimeric complexes (structural and phylogenetic aspects of MIKC-type proteins have been reviewed in detail by Kaufmann et al., 2005b).
10.3
From the ABC to the ABCDE model
Despite its explanatory power, e.g., in correctly predicting the phenotype of most double and triple mutants of floral homeotic genes, the ABC model has two important shortcomings: mutant and transgenic studies indicated that the ABC genes are required, but not sufficient, for the specification of floral organ identity. Moreover, the ABC model could not explain as to how the floral homeotic genes interact at the molecular level to specify floral organ identity. The first shortcoming could be overcome by the identification of additional floral homeotic functions and led to an extension of the ABC into an ABCDE model. Elimination of the second shortcoming required a switch from considerations at gene level to the level of the encoded proteins, and led to the ‘floral quartet model’. The floral homeotic functions A, B, and C could be identified by conventional ‘forward’ genetics, because some single mutants revealed complete loss-of-function phenotypes (Figure 10.1). However, additional homeotic functions had escaped forward genetics approaches due to the functional redundancy of the underlying, closely related genes. Their identification hence required the generation of multiple mutants via ‘reverse genetics’ generating single mutants followed by crossing. Based on studies in petunia (Petunia hybrida), the ABC model was first extended to an ‘ABCD model’ by addition of a D-function specifying ovule identity (Angenent and Colombo, 1996). In Arabidopsis, the class D genes are represented by AGAMOUS (which was introduced as a class C gene already), and three additional genes closely related to AG, namely SEEDSTICK (STK; formerly known as AGL11),
COMBINATORIAL CONTROL OF FLORAL ORGAN IDENTITY
257
SHATTERPROOF1 (SHP1; formerly known as AGL1), and SHATTERPROOF2 (SHP2; formerly known as AGL5) (Favaro et al., 2003; Pinyopich et al., 2003). stk shp1 shp2 triple mutants, for example, are characterized by conversion of ovules into carpel-like or leaf-like structures (Figure 10.1) (Pinyopich et al., 2003). Knocking out another class of MIKC-type MADS-box genes, termed SEPALLATA-like genes (also known as AGL2-like genes), revealed yet another class of floral homeotic function (Pelaz et al., 2000; Ditta et al., 2004). As class E it has been added to the ABCD model, yielding the ‘ABCDE model’ (Figure 10.1) (Theißen, 2001; for reviews on the diversity of ABCDE functions in different plants, see Ferrario et al., 2004; Krizek and Fletcher, 2005). While single and double mutants of SEPALLATA1 (SEP1, formerly known as AGL2), SEPALLATA2 (SEP2; formerly known as AGL4), SEPALLATA3 (SEP3; formerly known as AGL9), or SEPALLATA4 (SEP4; formerly known as AGL3) have only very weak mutant phenotypes, if any, in sep1 sep2 sep3 triple mutants the organs in all whorls of the flower develop into sepals, and flower development becomes indeterminate (Pelaz et al., 2000); in sep1 sep2 sep3 sep4 quadruple mutants vegetative leaves rather than sepals develop in all whorls of indeterminate flowers (Figure 10.1) (Ditta et al., 2004). The sep1 sep2 sep3 triple mutant phenotype is strikingly similar to that of a class B and C double mutant, i.e., ap3 ag or pi ag, indicating that the floral homeotic B- and C-functions do not work in the sep triple mutant. Also, the sep1 sep2 sep3 sep4 quadruple mutant phenotype is very similar to a class A, B, C triple mutant, indicating that the floral homeotic A-, B-, and C-functions do not work in the sep quadruple mutant. However, the SEP genes are still expressed in B and C lossof-function mutants (see, e.g., Flanagan and Ma, 1994), and the initial expression patterns of B and C class genes are not altered in the sep1 sep2 sep3 triple mutant (Pelaz et al., 2000). This suggested that the SEP genes, rather than acting upstream or downstream of the floral homeotic genes, constitute a novel class of redundant floral organ identity genes. Obviously, different combinations or dosages of SEP, i.e., class E genes, in addition to the ABC genes, are required for the specification of all types of floral organs. The ABCDE model maintains that class A + E genes are required to specify sepals, A + B + E petals, B + C + E stamens, C + E carpels, and D + E ovules (Figure 10.1) (Theißen, 2001; Ditta et al., 2004). Like the ABC model, the ABCDE model mainly relies on genetic data. This raises the question that by which molecular mechanism do the different floral homeotic genes interact, such that, e.g., A + B + E gives petals and B + C + E stamens (Figure 10.1)?
10.4
Elective affinities of homeotic proteins: the floral quartet model
MADS-domain proteins bind to DNA in a sequence-specific way only as dimers. It was thus tempting to explain the interaction of floral homeotic genes by dimerization of their gene products. This soon failed, however, because the formation of DNAbinding dimers, as revealed by electrophoretic mobility shift assays and the yeast two-hybrid system, shows a high degree of partner specificity (Riechmann et al., 1996; Fan et al., 1997). For example, the class B proteins AP3 and PI can bind to
258
REGULATION OF TRANSCRIPTION IN PLANTS
DNA only as obligatory AP3–PI heterodimers, but DNA-binding dimers with other MIKC-type proteins have not been observed (Riechmann et al., 1996). So how, for example, do class B genes (AP3 and PI) interact with the C gene (AG) to control stamen identity, if B + C (AP3–AG or PI–AG) DNA-binding dimers cannot be formed? An obvious solution would be that B and C proteins (AP3–PI and AG–AG dimers) bind separately to different cis-regulatory elements in the promoters of the same target genes that are activated or repressed during stamen formation. Alternatively, class B and class C proteins could have distinct sets of target genes involved in stamen formation, so that the combinatorial interaction of the homeotic genes would be mechanistically realized only at the level of target genes, or even more downstream in the gene cascades involved. A new perspective was provided when Egea-Cortines et al. (1999) reported that DEFICIENS (DEF), GLOBOSA (GLO), and SQUAMOSA (SQUA) from snapdragon (Antirrhinum majus) form multimeric complexes in electrophoretic mobility shift assays and yeast three-hybrid analyses. Evidence was provided that the multimeric complex has a higher DNA-binding affinity than the individual dimers. The authors suggested a model according to which the protein complex is actually a protein tetramer, composed of a DEF–GLO heterodimer and a SQUA–SQUA homodimer, in which the DEF–GLO and SQUA–SQUA dimers recognize different CArG boxes (Egea-Cortines et al., 1999). Based on double mutant analysis originally thought just to be required for the establishment of the whorled phyllotaxis of floral organ arrangement in Antirrhinum (Egea-Cortines et al., 1999), the formation of higher order complexes was soon realized to be a more general principle of floral homeotic proteins. DEF, GLO, and SQUA are putative orthologs of AP3, PI, and AP1, respectively (Becker and Theißen, 2003), suggesting that multimeric complexes might also be formed by the respective Arabidopsis proteins. It was known before that class C proteins such as PLE from Antirrhinum and AG form DNA-binding dimers with AGL2-like (SEP) proteins (Davies et al., 1996; Huang et al., 1996), suggesting that the floral homeotic C-function in Arabidopsis may require interaction between AG and SEP proteins. So when Pelaz et al. (2000) reported that not only the ABC, but also the SEP genes are required for the formation of petals, stamens, and carpels, all these findings were combined in the floral quartet model (Theißen, 2001). It suggests that tetrameric complexes of floral homeotic proteins including SEP proteins rather than individual dimers control floral organ identity. According to the original ‘quartet model’, there is at least one unique quaternary complex for each type of floral organ (Theißen, 2001) (Figure 10.2). These quaternary protein complexes might exert their function as transcription factors by binding to CArG boxes in the promoters of target genes, which they either activate or repress as appropriate for the development of the identities of the different floral organs (Theißen, 2001). In these complexes, the AP1 or SEP proteins may provide the transcription activation domain, while the other proteins might be more important for organ specificity of gene regulation. It has been proposed that two protein dimers of each tetramer recognize two different DNA CArG boxes, which might be brought into close vicinity by bending the DNA between the CArG boxes (Egea-Cortines et al., 1999; Theißen, 2001; Theißen and
COMBINATORIAL CONTROL OF FLORAL ORGAN IDENTITY
259
Figure 10.2 An extended and updated version of the floral quartet model of Arabidopsis. The model suggests that different combinations of MIKC-type MADS-domain proteins determine the identity of the different floral organs, or of substructures (the endothelium is the inner tissue layer in the inner integument of ovules, and ovules are parts of carpels). The combinatorial protein complexes represent transcription factors that exert their function by specifically binding to the promoters of target genes, which they either activate or repress as appropriate for the development of the identities of the different floral organs or substructures. Binding is assumed to occur to pairs of DNA sequence elements termed CArG boxes, which are brought in close vicinity by DNA bending. The molecular mechanism by which the basal transcriptional apparatus is affected by floral quartets is unknown. (Abbreviations of protein names: ABS, Arabidopsis Bsister ; for all other names, see corresponding gene names in Figure 10.1.)
Saedler, 2001), so the quartet (or tetramer) is actually a dimer of dimers (which might give a clue as to how it originated during evolution). Shortly after development of the floral quartet model, Honma and Goto (2001) indeed demonstrated by yeast three-hybrid and four-hybrid analyses the formation of the complexes postulated for stamens and petals, namely AP3/PI/AG/SEP and AP3/PI/AP1 (or SEP), respectively. Even more intriguingly, it was shown that ectopic coexpression of class A/E + B genes AP1 + AP3 + PI or SEP3 + AP3 + PI in transgenic Arabidopsis plants leads, simply put, to a developmental conversion of vegetative leaves into petaloid organs (Honma and Goto, 2001; Pelaz et al., 2001); more accurately phrased, ectopic expression of the mentioned genes leads to a reprogramming of supposed-to-be leaf primordia in a way that they develop into petaloid organs. AP1 and SEP3 can substitute each other, which probably reflects their close evolutionary relationship (Becker and Theißen, 2003) and hence partial redundancy. Again simply put, the coexpression of the class B + C + E genes AP3 + PI + AG + SEP3 converts vegetative leaves into stamenoid organs (Honma and Goto,
260
REGULATION OF TRANSCRIPTION IN PLANTS
2001). These data not only add genetic support for the floral quartet model but also demonstrate that less than a handful of proteins are not only necessary, but also sufficient to superimpose petal and stamen identity upon a vegetative developmental program.
10.5
Beyond the floral quartets sensu stricto: multimeric complexes of other MIKC-type proteins
The floral quartet model describes the interaction of the floral homeotic proteins at the molecular level. However, the floral homeotic proteins are only a minor fraction of the large family of MIKC-type MADS-domain proteins, comprising 39 members in Arabidopsis alone (Becker and Theißen, 2003). Therefore, the question arises as to whether multimerization is restricted to the floral homeotic proteins of eudicots (an extreme hypothesis) or is a common feature of all MIKC-type proteins in all kinds of land plants and charophyte green algae (another extreme hypothesis) (Kaufmann et al., 2005b). There is indeed evidence that higher order complexes are also important for ovule development (Figure 10.2). The class D proteins from Arabidopsis formed multimeric complexes together with the SEP3 protein when tested in yeast threehybrid assays (Favaro et al., 2003). In addition, partial loss of SEP-gene activity leads to similar defects in ovule development as observed in stk shp1 shp2 triple mutants (Favaro et al., 2003). Moreover, another subfamily of MIKC-type MADS-domain proteins has been shown to be important for some aspects of ovule development as well. Members of this Bsister -subfamily, the putative sister clade of the class B genes, show femalespecific expression in all species examined so far (Becker et al., 2002). In accordance with this, loss-of-function mutants of the Arabidopsis Bsister gene (ABS) show defects in the development of the innermost layer of the integument of the ovule, the endothelium (Nesi et al., 2002; Kaufmann et al., 2005a). Fitting well to these genetic data, the class D proteins SHP1, SHP2, and STK all form a multimeric complex together with ABS mediated by SEP3, suggesting that endothelium development is controlled by a multimeric complex composed of the respective proteins (Figure 10.2) (Kaufmann et al., 2005a). This demonstrates that higher complex formation is not restricted to floral homeotic proteins, but applies also to MIKC-type MADS-domain proteins that work ‘downstream’ of the homeotic proteins in the gene hierarchy that controls flower development (Kaufmann et al., 2005a). Although these data help us already to get a deeper understanding of the molecular mechanisms underlying flower development, we are just beginning to elucidate the complex network of multimerization among MIKC-type proteins. It has been suggested recently that different MADS-domain protein complexes are formed in a temporal order of availability, adding another level of complexity to the regionspecific complex formation (Castillejo et al., 2005). If this holds true, it has important evolutionary implications, as it suggests a mechanism by which the present-day regulatory cascade could have originated. For example, AG and SEP are hypothesized
COMBINATORIAL CONTROL OF FLORAL ORGAN IDENTITY
261
to act together to specify carpel identity. Moreover, the expression of the respective genes is maintained through autoregulatory feedback loops (G´omez-Mena et al., 2005). One target gene of the complex is thought to be SHP2 (Savidge et al., 1995; Castillejo et al., 2005). On the other hand, SHP2 and SEP are supposed to act together to confer ovule identity (Favaro et al., 2003). As AG and the SHP genes originated by gene duplication, one could imagine that this ancient duplication followed by the loss of the autoregulatory capability of SHP2 created the present-day cascade. Even less explored is the interaction behavior of MADS-domain proteins expressed in reproductive parts of the plant. However, it is quite reasonable to assume that what is true for organ identity proteins is also true for other angiosperm MIKCtype MADS-domain proteins.
10.6
Outlook: the devil is always in the details – what we still don’t know about floral quartets
Like the ABCDE model, the floral quartet model specifies not only which genetic components are required, but also which are sufficient to specify floral organ identity. In contrast to the ABCDE model, however, the floral quartet model also demonstrates how the combinatorial interaction of these components works mechanistically. Therefore, it soon became the standard model for work on floral homeotic genes far beyond Arabidopsis (see, e.g., Jack, 2001; Winter et al., 2002; Ferrario et al., 2003; Lee et al., 2003; Ferrario et al., 2004; Shchennikova et al., 2004). Not few consider the floral quartet model now as (largely) accepted by the scientific community (see, e.g., Krizek and Fletcher, 2005; Robles and Pelaz, 2005). Our parental pride notwithstanding, however, we want to draw attention to the fact that the empirical basis of the floral quartet model is still quite weak; few successful direct experimental tests beyond studying protein–protein interactions in vitro and in yeast, employing electrophoretic mobility shift assays and yeast three-hybrid analyses, have been carried out. Whether multimeric complexes of floral homeotic proteins really regulate gene expression in the nuclei of plant cells, we just do not know. To become experimentally tractable, the problem should be subdivided into two major questions: 1. Do floral quartets or similar complexes really exist in planta? One of the most intriguing questions is as to which extent the interactions observed in yeast n-hybrid and gel retardation studies also exist in the living plant cell. Using fusions of MADS-domain proteins and yellow fluorescent protein (YFP) or cyan fluorescent protein (CFP), Immink et al. (2002) were able to demonstrate dimerization of MADS-domain transcription factors in plant protoplasts. The method being used is based on the fluorescence resonance energy transfer (FRET) occurring when YFP and CFP are in close vicinity to each other. Recently, this experimental strategy was extended to provide circumstantial evidence for higher order complex formation (Nougalli-Tonaco et al., 2006). In the novel assay, a third, unlabeled interaction partner was coexpressed with two others that dimerize only weakly. As the unlabeled
262
REGULATION OF TRANSCRIPTION IN PLANTS
protein interacts more strongly with both of the labeled proteins, a titration effect, i.e., very weak or no interaction, should be observed between the labeled proteins if only dimerization occurs. However, Nougalli-Tonaco et al. (2006) observed a strong FRET signal when these three proteins were coexpressed; the most plausible explanation for this observation is the formation of a higher order complex as predicted by the floral quartet model. These studies provide an excellent starting point for analyzing interactions of MADS-domain proteins in the plant. Expressing the floral homeotic proteins fused to CFP and YFP under the control of their natural promoter will reveal which complexes are formed in a floral context. It should also be mentioned that another noninvasive technique, bimolecular fluorescence complementation (BiFC), has been used to test protein interactions in plant cells (Bracha-Drori et al., 2004). This system is based on the observation that a split YFP, fused to two different proteins, regains its fluorescence activity if the proteins tested interact with each other. However, it remains to be determined whether BiFC is useful for studying the interaction of MADS-domain proteins. Another question that could be answered making use of GFP and its derivates is whether the ‘floral quartet’ is indeed a quartet (i.e., a tetramer). Using virus particles containing a known number of GFP molecules as an internal standard, the number of proteins in cellular structures has been determined (Dundr et al., 2002). If this method is also applicable to MADS-domain proteins, the stochiometric compostion of the complexes might be accessible. 2. What is the exact composition of multimeric complexes? Although the quartet model explains the available data in the most plausible way, it cannot explain all the phenomena observed. Questions that remain open are, for example, how target gene specificity is achieved or how only one multimer can specify a complex floral organ with its many different cell types. One possible answer, although not the only one, is that the floral quartets act together with other factors to regulate region-specific gene expression and target gene specificity. Several screens using the yeast two-hybrid system have been carried out, employing either one or several MIKC-type MADSdomain proteins as a bait (for examples see Davies et al., 1996; Honma and Goto, 2001). However, few convincing non-MADS interaction partners have been identified so far. This might reflect the fact that there are no such interaction partners, or, alternatively, that other methods need to be employed to isolate them. In the past years, tandem affinity purification (tap) (Rigaut et al., 1999) turned out to be a powerful technique to isolate protein complexes directly from their ‘natural cellular environment’. Using this method, tagged proteins can be stably expressed in the organism of interest. The tag design allows two different purification steps, yielding complexes of high purity. The big advantage of this method is its unbiased approach that aims to isolate the ‘real’ interaction partners of a protein of interest. Tandem affinity purification has been successfully used to isolate protein complexes from yeast and mammalian cells (reviewed in Gingras et al., 2005). Special tap tags for isolation of plant protein complexes have been published (Rohila et al., 2004; Rubio et al., 2005), and the near future will hopefully show whether this technique is also useful to tackle the in vivo complex formation of MADS-domain transcription factors. A number of other questions also remain: Which target genes do floral quartets control, and how is target gene specificity achieved? Little is known to answer the
COMBINATORIAL CONTROL OF FLORAL ORGAN IDENTITY
263
first question, and almost nothing to answer the second one (see, e.g., Sablowski and Meyerowitz, 1998; G´omez-Mena et al., 2005; Melzer et al., in press). How do the MIKC-type protein complexes bind to DNA, and how is transcription affected, e.g., how do they interact with RNA polymerase II and general transcription factors? How conserved are multimeric complexes of MIKC-type proteins in evolution, and to what extent did gene duplications within the MADS-box gene family contribute to the origin of these complexes? What are the implications of complex formation for the coevolution of MADS-box genes? Only time – filled with hard work – will tell.
10.7
Summary
To explain how the different floral organs adopt their unique identities during flower development, combinatorial interactions between five classes of floral homeotic genes, termed A, B, C, D, and E, have been proposed, with A + E specifying sepals, A + B + E petals, B + C + E stamens, C + E carpels, and D + E ovules. The capacity of MIKC-type MADS-domain proteins to form multimeric complexes provides a molecular basis for the combinatorial interactions of the class A, B, C, D, and E genes, which (almost) all encode MIKC-type proteins. According to the floral quartet model, the identity of the different types of floral organs is specified by tetrameric complexes of MIKC-type proteins (floral quartets). These protein complexes are hypothesized to bind to two cis-regulatory DNA sequence elements in the promoter regions of their target genes, which are brought into close vicinity by DNA bending. The floral quartets represent transcription factors that regulate all those target genes whose differential gene expression is required and sufficient for the specification of the identity of the different floral organs (sepals, petals, stamens, and carpels), or substructures (ovules, endothelium of the inner integument of ovules). An important goal of future research will be to define the exact structures of all the transcription factor complexes that work as major developmental switches during the plant life cycle.
Acknowledgments The authors thank Klaus D. Grasser for his kind invitation to contribute to this book. A fellowship to R.M. by the Studienstiftung des deutschen Volkes is greatly acknowledged.
References G.C. Angenent and L. Colombo (1996) Molecular control of ovule development. Trends in Plant Science 1, 228–232. A. Becker, K. Kaufmann, A. Freialdenhoven, et al. (2002) A novel MADS-box gene subfamily with a sister-group relationship to class B floral homeotic genes. Molecular Genetics and Genomics 266, 942–950. A. Becker and G. Theißen (2003) The major clades of MADS-box genes and their role in the development and evolution of flowering plants. Molecular Phylogenetics and Evolution 29, 464–489. K. Bracha-Drori, K. Shichrur, A. Katz, et al. (2004) Detection of protein–protein interactions in plants using bimolecular fluorescence complementation. The Plant Journal 40, 419–427.
264
REGULATION OF TRANSCRIPTION IN PLANTS
C. Castillejo, M. Romera-Branchat and S. Pelaz (2005) A new role of the Arabidopsis SEPALLATA3 gene revealed by its constitutive expression. The Plant Journal 43, 586–596. E.S. Coen and E.M. Meyerowitz (1991) The war of the whorls – genetic interactions controlling flower development. Nature 353, 31–37. B. Davies, M. EgeaCortines, E.D. Silva, H. Saedler and H. Sommer (1996) Multiple interactions amongst floral homeotic MADS box proteins. EMBO Journal 15, 4330–4343. G. Ditta, A. Pinyopich, P. Robles, S. Pelaz and M.F. Yanofsky (2004) The SEP4 gene of Arabidopsis thaliana functions in floral organ and meristem identity. Current Biology 14, 1935–1940. M. Dundr, J.G. McNally, J. Cohen and T. Misteli (2002) Quantitation of GFP-fusion proteins in single living cells. Journal of Structural Biology 140, 92–99. M. Egea-Cortines, H. Saedler and H. Sommer (1999) Ternary complex formation between the MADSbox proteins SQUAMOSA, DEFICIENS and GLOBOSA is involved in the control of floral architecture in Antirrhinum majus. EMBO Journal 18, 5370–5379. H.Y. Fan, Y. Hu, M. Tudor and H. Ma (1997) Specific interactions between the K domains of AG and AGLs, members of the MADS domain family of DNA binding proteins. The Plant Journal 12, 999–1010. R. Favaro, A. Pinyopich, R. Battaglia, et al. (2003) MADS-box protein complexes control carpel and ovule development in Arabidopsis. Plant Cell 15, 2603–2611. S. Ferrario, R.G. Immink and G.C. Angenent (2004) Conservation and diversity in flower land. Current Opinion in Plant Biology 7, 84–91. S. Ferrario, R.G.H. Immink, A. Shchennikova, J. Busscher-Lange and G.C. Angenent (2003) The MADS box gene FBP2 is required for SEPALLATA function in petunia. Plant Cell 15, 914–925. C.A. Flanagan and H. Ma (1994) Spatially and temporally regulated expression of the MADS-box gene AGL2 in wild-type and mutant Arabidopsis flowers. Plant Molecular Biology 26, 581–595. A.C. Gingras, R. Aebersold and B. Raught (2005) Advances in protein complex analysis using mass spectrometry. Journal of Physiology – London 563, 11–21. C. G´omez-Mena, S. de Folter, M.M.R. Costa, G.C. Angenent and R. Sablowski (2005) Transcriptional program controlled by the floral homeotic gene AGAMOUS during early organogenesis. Development 132, 429–438. K. Goto and E.M. Meyerowitz (1994) Function and regulation of the Arabidopsis floral homeotic gene PISTILLATA. Genes and Development 8, 1548–1560. G.W. Haughn and C.R. Somerville (1988) Genetic control of morphogenesis in Arabidopsis. Developmental Genetics 9, 73–89. T. Honma and K. Goto (2001) Complexes of MADS-box proteins are sufficient to convert leaves into floral organs. Nature 409, 525–529. H. Huang, M. Tudor, T. Su, Y. Zhang, Y. Hu and H. Ma (1996) DNA binding properties of two Arabidopsis MADS domain proteins: binding consensus and dimer formation. Plant Cell 8, 81–94. R.G.H. Immink, T.W.J. Gadella, S. Ferrario, M. Busscher and G.C. Angenent (2002) Analysis of MADS box protein–protein interactions in living plant cells. Proceedings of the National Academy of Sciences of the United States of America 99, 2416–2421. T. Jack (2001) Relearning our ABCs: new twists on an old model. Trends in Plant Science 6, 310–316. T. Jack, L.L. Brockman and E.M. Meyerowitz (1992) The homeotic gene APETALA3 of Arabidopsis thaliana encodes a MADS box and is expressed in petals and stamens. Cell 68, 683–697. K.D. Jofuku, B.G.W. Denboer, M. Van Montagu and J.K. Okamuro (1994) Control of Arabidopsis flower and seed development by the homeotic gene APETALA2. Plant Cell 6, 1211–1225. K. Kaufmann, N. Anfang, H. Saedler and G. Theissen (2005a) Mutant analysis, protein–protein interactions and subcellular localization of the Arabidopsis B-sister (ABS) protein. Molecular Genetics and Genomics 274, 103–118. K. Kaufmann, R. Melzer and G. Theissen (2005b) MIKC-type MADS-domain proteins: structural modularity, protein interactions and network evolution in land plants. Gene 347, 183–198. B.A. Krizek and J.C. Fletcher (2005) Molecular mechanisms of flower development: an armchair guide. Nature Reviews Genetics 6, 688–698. S. Lee, J.S. Jeon, K. An, et al. (2003) Alteration of floral organ identity in rice through ectopic expression of OsMADS16. Planta 217, 904–911.
COMBINATORIAL CONTROL OF FLORAL ORGAN IDENTITY
265
M.A. Mandel, C. Gustafson-Brown, B. Savidge and M.F. Yanofsky (1992) Molecular characterization of the Arabidopsis floral homeotic gene APETALA1. Nature 360, 273–277. R. Melzer, K. Kaufmann and G. Theißen (in press) Missing links: DNA-binding and target gene specificity of floral homeotic proteins. Advances in Botanical Research. T. M¨unster, J. Pahnke, A. DiRosa, et al. (1997) Floral homeotic genes were recruited from homologous MADS-box genes preexisting in the common ancestor of ferns and seed plants. Proceedings of the National Academy of Sciences of the United States of America 94, 2415–2420. N. Nesi, I. Debeaujon, C. Jond, et al. (2002) The TRANSPARENT TESTA16 locus encodes the ARABIDOPSIS BSISTER MADS domain protein and is required for proper development and pigmentation of the seed coat. Plant Cell 14, 2463–2479. I.A. Nougalli-Tonaco, J.W. Borst, S.C. de Vries, G.C. Angenent and R.G.H. Immink (2006) In vivo imaging of MADS-box transcription factor interactions. Journal of Experimental Botany 57, 33–42. S. Pelaz, G.S. Ditta, E. Baumann, E. Wisman and M.F. Yanofsky (2000) B and C floral organ identity functions require SEPALLATA MADS-box genes. Nature 405, 200–203. S. Pelaz, R. Tapia-Lopez, E.R. Alvarez-Buylla and M.F. Yanofsky (2001) Conversion of leaves into petals in Arabidopsis. Current Biology 11, 182–184. A. Pinyopich, G.S. Ditta, B. Savidge, et al. (2003) Assessing the redundancy of MADS-box genes during carpel and ovule development. Nature 424, 85–88. J.L. Riechmann and E.M. Meyerowitz (1997). MADS domain proteins in plant development. Biol. Chem. 378, 1079–1101. J.L. Riechmann, B.A. Krizek and E.M. Meyerowitz (1996) Dimerization specificity of Arabidopsis MADS domain homeotic proteins APETALA1, APETALA3, PISTILLATA, and AGAMOUS. Proceedings of the National Academy of Sciences of the United States of America 93, 4793–4798. G. Rigaut, A. Shevchenko, B. Rutz, M. Wilm, M. Mann and B. Seraphin (1999) A generic protein purification method for protein complex characterization and proteome exploration. Nature Biotechnology 17, 1030–1032. P. Robles and S. Pelaz (2005) Flower and fruit development in Arabidopsis thaliana. International Journal of Developmental Biology 49, 633–643. J.S. Rohila, M. Chen, R. Cerny and M.E. Fromm (2004) Improved tandem affinity purification tag and methods for isolation of protein heterocomplexes from plants. The Plant Journal 38, 172–181. V. Rubio, Y.P. Shen, Y. Saijo, et al. (2005) An alternative tandem affinity purification strategy applied to Arabidopsis protein complex isolation. The Plant Journal 41, 767–778. R.W.M. Sablowski and E.M. Meyerowitz (1998) A homolog of NO APICAL MERISTEM is an immediate target of the floral homeotic genes APETALA3/PISTILLATA. Cell 92, 93–103. B. Savidge, S.D. Rounsley and M.F. Yanofsky (1995) Temporal relationship between the transcription of 2 Arabidopsis MADS-box genes and the floral organ identity genes. Plant Cell 7, 721–733. A.V. Shchennikova, O.A. Shulga, R. Immink, K.G. Skryabin and G.C. Angenent (2004) Identification and characterization of four chrysanthemum MADS-box genes, belonging to the APETALA1/FRUITFULL and SEPALLATA3 subfamilies. Plant Physiology 134, 1632–1641. G. Theißen (2001) Development of floral organ identity: stories from the MADS house. Current Opinion in Plant Biology 4, 75–85. G. Theißen, J.T. Kim and H. Saedler (1996) Classification and phylogeny of the MADS-box multigene family suggest defined roles of MADS-box gene subfamilies in the morphological evolution of eukaryotes. Journal of Molecular Evolution 43, 484–516. G. Theißen and H. Saedler (2001) Plant biology. Floral quartets. Nature 409, 469–471. D. Weigel and E.M. Meyerowitz (1994) The ABCs of floral homeotic genes. Cell 78, 203–209. K.U. Winter, H. Saedler and G. Theißen (2002) On the origin of class B floral homeotic genes: functional substitution and dominant inhibition in Arabidopsis by expression of an orthologue from the gymnosperm Gnetum. The Plant Journal 31, 457–475. C. Wolberger (1998) Combinatorial transcription factors. Current Opinion in Genetics and Development 8, 552–559. M.F. Yanofsky, H. Ma, J.L. Bowman, G.N. Drews, K.A. Feldmann and E.M. Meyerowitz (1990) The protein encoded by the Arabidopsis homeotic gene AGAMOUS resembles transcription factors. Nature 346, 35–39.
11 Networks of transcriptional regulation underlying plant defense responses toward phytopathogens Imre E. Somssich
11.1
Introduction
Plants are constantly exposed to thousands of potentially harmful air-, water- and soil-borne pathogens within their environment. From an evolutionary standpoint, the survival of individual plant species can only be reasonably ensured provided they have the capacity to develop surveillance mechanisms and signaling networks allowing them to perceive and respond appropriately to such dangers. Conversely, pathogen survival is equally dependent on the ability of microbial pests to avoid, overcome, or partly subdue such plant defenses, thereby allowing them to gain access to their hosts and successfully complete their life cycles. This coevolutionary ‘tug-of-war’ is certainly a key driving force resulting, on the one hand, in continuous diversification and refinement of host defense systems, and on the other hand, in alterations of the countermeasures employed by pathogens to undermine detection or overcome newly devised defense barriers. Molecular and genetic studies, in particular with the model system Arabidopsis thaliana, have revealed that higher plants have evolved highly sophisticated surveillance mechanisms and signaling networks to initiate effective defense responses (Belkhadir et al., 2004; Jones and Takemoto, 2004). Considering the large diversity of potential pathogens, their distinct modes of infection, and their varying lifestyles, it is obvious that plants must have gained the ability to recognize and restrict invasions occurring in diverse organs, tissue, cell types, and in intra- and intercellular compartments. From our current understanding, plants appear to possess several layers of defenses to cope with these problems (Thordal-Christensen, 2003; Mysore and Ryu, 2004). The most common form of plant resistance toward phytopathogenic organisms is termed nonhost or species-level resistance (N¨urnberger and Brunner, 2002). Apart from preformed physical and chemical barriers, this kind of basic resistance involves the ability of the plant to recognize various surface or secreted molecules of diverse pathogens (also designated PAMPs for Pathogen-Associated Molecular Patterns), and also molecules of the plant itself that are released during microbial infection. This recognition process subsequently triggers an efficient and multicomponent defense response. Nonhost resistance has several remarkable similarities to the mammalian innate immunity response and hence is generally referred to as plant innate immunity (Jones and Takemoto, 2004; Zipfel and Felix,
TRANSCRIPTIONAL REGULATION UNDERLYING PLANT DEFENSE
267
2005). A second very effective form of resistance is termed R-gene mediated or gene-for-gene-type resistance. In this case, specific defense responses are triggered only if there is an allele-specific genetic interaction between a major host resistance gene (R-gene) and a pathogen avirulence (avr) gene (Belkhadir et al., 2004; Skamnioti and Ridout, 2005). Although we do not yet fully understand all of the features associated with these two types of resistances, it is becoming quite apparent that there is substantial overlap in the signaling networks employed by the two, and that in both cases transcriptional activation of defense genes plays a pivotal role (Nimchuk et al., 2003; Eulgem, 2005). Stimulation of defense responses is not only locally restricted to the site of infection but also occurs in distal areas of the plant, a reaction termed systemic acquired resistance (SAR) (Durrant and Dong, 2004). Activation of the SAR pathway confers immunity throughout the plant toward a broad range of microorganisms. Over the past few years, huge progress has been made in defining key upstream components and early signaling events vital for establishing plant resistance. However, although the plant cell nucleus represents a major target of different defense signal transduction pathways, our understanding of the terminal signaling stages is extremely rudimentary. We know very little about the biochemical signals and how they are relayed and linked to nuclear components and specific transcription factors (TFs). Equally fragmentary are our insights into the intricate transcriptional machinery responsible for executing proper temporal and spatial control of defense gene expression. In this review, I will present an update of our current state of knowledge on the key mechanisms and individual constituents involved in the transcriptional regulatory network governing the expression of the defense transcriptome.
11.2
Defense transcriptome
Intensive physiological, biochemical, molecular, and genetic studies over the past two decades using a number of plant pathosystems have revealed that plants respond to pathogen challenge by drastically reprogramming overall cellular metabolism. All nonessential cellular activities appear to be arrested or downregulated, probably to redirect substrates and energy to the de novo synthesis of a complex spectrum of defense-related metabolites (Somssich and Hahlbrock, 1998; Hahlbrock et al., 2003). This major redirection of cell metabolism is a consequence of extensive transcriptional reprogramming resulting in the altered expression of numerous genes coding for diverse products derived from both primary and secondary metabolism (Batz et al., 1998; Rushton and Somssich, 1999; Durrant et al., 2000). These earlier observations have been substantiated and extended by the development of large-scale gene expression profiling techniques (Scheideler et al., 2002) and the intensive use of the fully sequenced model plants Arabidopsis thaliana and rice for plant–pathogen interaction studies. Microrarray analyses have revealed massive transcriptional reprogramming, affecting up to 20% of the entire Arabidopsis genome, following challenge by diverse pathogens or by pathogen-mimicking stimuli (Nimchuk et al.,
268
REGULATION OF TRANSCRIPTION IN PLANTS
2003; Katagiri, 2004). The Arabidopsis defense transcriptome comprises genes with known functions related, among others, to signaling, transport, transcription, and primary and secondary metabolism. A similarly complex expression pattern was observed by comprehensive transcript profiling of pathogen-mediated host defense responses in rice and tomato (Chu et al., 2004; Gibly et al., 2004). The ability to monitor the expression of thousands of genes simultaneously under different defenserelated treatments and in a spatial and temporal manner has enabled us to detect novel genes associated with defense and to identify coregulated genes (regulons). Additionally, valuable information on the links between defense and other plant responses has been gained from such global analyses. For instance, previous genetic studies have indicated substantial crosstalk between defense and senescence and between different defense-signaling pathways (Obreg´on et al., 2001; Glazebrook, 2005). Indeed, microarray studies have revealed significant overlap in the genes activated upon pathogen challenge and during leaf senescence (Buchanan-Wollaston et al., 2003), as well as by treatments with salicylic acid (SA) and jasmonic acid (JA), molecules that trigger two distinct defense pathways originally thought to act antagonistically (Bostock, 2005). Extensive overlap in the genes associated with defense, wounding, and other stresses have also been observed (Durrant et al., 2000; Cheong et al., 2002). One vital piece of information gained from such studies is that the expression patterns associated with compatible (the plant is susceptible) and incompatible (the plant is resistant) plant–microbe interactions can be strikingly similar, but both the speed and the amplitude of expression are often greater during the R-genedependent resistance response (Nimchuk et al., 2003; Katagiri, 2004). Thus, the temporal aspect of the host response may be a key determinant in deciding whether the plant or the pathogen wins the battle.
11.3
Regulatory sequences mediating defense gene expression
Expression of any given gene is primarily determined by the interaction of different TFs and cofactors acting at specific DNA sites within its regulatory regions. Other processes that involve interactions with other proteins and signaling cascades in turn regulate the TFs. Several distinct cis-acting DNA elements have been pinpointed in the promoters of defense-associated plant genes that mediate pathogen-dependent expression (Table 11.1) (Eulgem, 2005). Expression of PR1 in Arabidopsis is strongly upregulated during pathogen infection or upon chemical treatments known to activate the SA-dependent defense-signaling pathway. Functional dissection of its promoter revealed the importance of two positive regulatory elements, the activator sequence-1-like (as-1) element and an undefined sequence, 5 -AGGACTTTTC-3 , and one negative regulatory element containing a W RKYbinding site (W-box) (Lebel et al., 1998) (Table 11.1). In parsley, several genes responding rapidly and transiently to pathogen stimuli were identified (Somssich and Hahlbrock, 1998). Detailed functional dissection of their promoters using a protoplast system for transient expression revealed the presence of several cis-acting
TRANSCRIPTIONAL REGULATION UNDERLYING PLANT DEFENSE Table 11.1
269
Regulatory DNA sequences mediating defense gene expression
DNA element
TF class
Reference
TGACGTCA (as1-like) AGGACTTTTC TTGACC/T (W-box) TTCAAACA (D-box) GCCACCA (S-box) AGCCGCC (GCC box) AGACCGCC (JERE box) GTCAAAAA/T (PB box) CTGAAGAAGAA (TL1)
TGA factor (bZIP) ? WRKY factor ? ? ERF factor (AP2 class) ORCA (AP2 class) Whirly factor ?
Jakoby et al. (2002) Lebel et al. (1998) Eulgem et al. (2000) Rushton et al. (2002) Kirsch et al. (2001) Gutterson and Reuber (2004) Memelink et al. (2001) Desveaux et al. (2005) Wang et al. (2005)
‘?’ indicates interacting factor is unknown.
DNA elements that were both necessary and sufficient (can act independently) to mediate elicitor-dependent responsiveness of the respective gene (Rushton et al., 1996; Kirsch et al., 2000, 2001). Positive elements uncovered in this system included the D-box, S-box, and W-box (Table 11.1). All of these individual elements alone were also capable of driving strong pathogen-dependent expression of a reporter gene in transgenic Arabidopsis plants (Rushton et al., 2002). The GCC element and variations such as the JERE box were found in numerous plant defense gene promoters that are activated via the JA- or ethylene-signaling pathways (Memelink et al., 2001; Gutterson and Reuber, 2004). Two additional DNA elements shown to be important in the activation of defense genes included the PB box, that was originally found in the potato PR-10a gene promoter but is also present in other plant defense gene promoters (Desveaux et al., 2004), and the TLI element present in the promoters of several Arabidopsis genes encoding endoplasmic-reticulum-resident proteins (Wang et al., 2005). Recently, differential expression of the rice R-gene Xa27 was shown to determine resistance toward the bacterium Xanthomonas oryzae (Gu et al., 2005). Intriguingly, the promoter sequences of the Xa27 alleles present in the resistant and susceptible rice cultivars are nearly identical except for the presence of an additional 10-bp sequence at –1400 bp and a 25-bp duplication at –97 bp in the susceptible allele. Infection by the bacterium carrying the corresponding avrXa27 avirulence gene selectively induced expression of the resistant allele, indicating that specificity resided within the Xa27 promoter (Gu et al., 2005). The availability of fully sequenced genomes and the use of microarrays to monitor the expression of thousands of genes under a variety of conditions have provided a rich resource for the in silico detection of regulatory DNA elements. Bioinformatics methods are being developed with the ultimate aim of producing comprehensive maps of the regulatory networks within respective organisms (Rombauts et al., 2003; Wasserman and Sandelin, 2004). Despite the development of numerous algorithms, detection of key cis-regulatory elements remains an enormous challenge. Nevertheless, computational analyses have already provided us with some important clues concerning regulatory elements involved in defense gene activation. Applying
270
REGULATION OF TRANSCRIPTION IN PLANTS
clustering methods, several groups identified the W-box element as a commonly occurring motif within the promoters of coregulated defense-response genes (Maleck et al., 2000; Petersen et al., 2000; Navarro et al., 2004; Andreasson et al., 2005). In the studies performed by Petersen et al. (2000) a second conserved motif, similar to the 5 -AGGACTTTTC-3 element found in the Arabidopsis PR1 promoter, was also found. The potential power of computational biology was exemplified by the recent work of Wang et al. (2005). Microarray experiments designed to search for downstream target genes coordinately regulated by the nuclear-localized protein NPR1, a key regulator of SAR (Durrant and Dong, 2004), revealed a set of genes coding for secretion-related products. Analysis of the promoters of 13 of these genes revealed the presence of a common novel element designated TL1 (Table 11.1). Transgenic Arabidopsis plants harboring expression constructs driven by the TL1 or mutated TL1 elements subsequently confirmed the importance of this motif in mediating SA-induced expression (Wang et al., 2005).
11.4
Transcription factors involved in defense gene regulation
The drastic transcriptional reprogramming associated with the plant defense response requires the interaction of specific TFs with DNA, with other TFs, and with diverse nuclear proteins including coactivators/-repressors or components of the general transcriptional machinery. Currently, we have just begun to identify members of several TF families that selectively modulate defense gene expression. Such studies are complicated by the fact that in the majority of cases, TFs belong to large gene families and that functional redundancy often masks the contribution of individual members. Nevertheless, it is becoming obvious that TGA proteins of the bZIP family (Jakoby et al., 2002), ethylene-response factors (ERFs) of the AP2 ¨ class of TFs (Gutterson and Reuber, 2004), and WRKY TFs (Ulker and Somssich, 2004) are key players (Table 11.2). Several lines of evidence support the involvement of TGA factors in SAR and in the direct regulation of PR1, a prominent SA-inducible Arabidopsis marker gene. Previous work showed that NPR1 interacts in vitro and in vivo with selected TGA factors leading to enhanced binding of these factors to the as-1 site within the PR1 promoter (Durrant and Dong, 2004; Fobert and Despres, 2005). SA- and NPR1dependent in vivo recruitment of TGA2 and TGA3 to the PR1 promoter was also demonstrated using chromatin immunoprecipitation (ChIP) (Johnson et al., 2003). Moreover, genetic studies using tga2, tga5, and tga6 mutant Arabidopsis lines revealed the existence of functional redundancy between these factors and also illustrated their essential roles in SAR (Zhang et al., 2003). Interestingly, overexpression of TGA5 conferred resistance in Arabidopsis to an oomycete pathogen, but this was independent of SA and NPR1 (Kim and Delaney, 2002), indicating that TGA factors can effect other plant resistance mechanisms. In rice, ectopic expression of a dominant-negative variant of rTGA2.1, or silencing of the corresponding endogenous gene, both resulted in plants showing elevated transcript levels of defense marker genes and increased tolerance toward a bacterial pathogen, consistent with a negative function of the wild-type gene in defense (Fitzgerald et al., 2005).
TRANSCRIPTIONAL REGULATION UNDERLYING PLANT DEFENSE
271
Table 11.2 Plant transcription factors implicated in regulating defense responses toward phytopathogens TF class
Name/synonyms
Origin
References (and citations within)
AP2/EREBP
ERF1 AtERF1 AtERF2 AtERF4 AtERF13 CaPF1 pCaERFLP1 NbCD1
Arabidopsis Arabidopsis Arabidopsis Arabidopsis Arabidopsis Hot pepper Hot pepper Nicotiana benthamiana Periwinkel Periwinkel Rice Tobacco Tobacco Tobacco Tomato Tomato Tomato Tomato Arabidopsis Arabidopsis Arabidopsis Arabidopsis Arabidopsis Arabidopsis Tobacco Tobacco Rice Arabidopsis Arabidopsis Periwinkel Tobacco Tomato Tomato Arabidopsis Arabidopsis Potato Arabidopsis Potato Arabidopsis Arabidopsis Arabidopsis Arabidopsis Arabidopsis Arabidopsis Arabidopsis Arabidopsis Barley
Berrocal-Lobo and Solano (2002) Gutterson and Reuber (2004) Gutterson and Reuber (2004) Yang et al. (2005) Gutterson and Reuber (2004) Li et al. (2004) Lee et al. (2004b) Nasir et al. (2005)
bHLH/MYC BZIP
MYB
NAC
Whirly WRKY
ORCA2 ORCA3 OsEREBP1 NtERF5 Tsi1 OPB1 Pti4 Pti5 Pti6 TSRF1 AtMYC2/JAI1/JIN1 TGA2 TGA3 TGA4 TGA5 TGA6 TGA2.1 TGA2.2 RTGA2 AtMYB30 BOS1/AtMYB108 CrBPF-1 myb1 JAMYC2 JAMYC10 ATAF2 TIP StNAC AtWhy1 PBF-2/StWhy1 AtWRKY6 AtWRKY18 AtWRKY22 AtWRKY25 AtWRKY29 AtWRKY33 AtWRKY70 RRS1/AtWRKY52 HvWRKY38/TDF N9-10
Memelink et al. (2001) Memelink et al. (2001) Zhou et al. (1997) Fischer and Dr¨oge-Laser (2004) Park et al. (2001) Guo et al. (2004) Zhou et al. (1997) Zhou et al. (1997) Zhou et al. (1997) Zhang et al. (2004) Lorenzo et al. (2004) Durrant and Dong (2004) Durrant and Dong (2004) Durrant and Dong (2004) Durrant and Dong (2004) Durrant and Dong (2004) Durrant and Dong (2004) Durrant and Dong (2004) Fitzgerald et al. (2005) Vailleau et al. (2002) Mengiste et al. (2003) Memelink et al. (2001) Eulgem (2005) Lorenzo and Solano (2005) Lorenzo and Solano (2005) Delessert et al. (2005) Olsen et al. (2005) Olsen et al. (2005) Desveaux et al. (2005) Desveaux et al. (2005) ¨ Ulker and Somssich (2004) ¨ Ulker and Somssich (2004) Asai et al. (2002) Andreasson et al. (2005) Asai et al. (2002) Andreasson et al. (2005) Li et al. (2004) Deslandes et al. (2002) Eckey et al. (2004)
(Continued)
272
REGULATION OF TRANSCRIPTION IN PLANTS
Table 11.2 Plant transcription factors implicated in regulating defense responses toward phytopathogens (Continued) TF class
Name/synonyms
Origin
References (and citations within)
PcWRKY1 PcWRKY2 PcWRKY3 PcWRKY4 PcWRKY5 NbWRKY1
Parsley Parsley Parsley Parsley Parsley Nicotiana benthamiana Nicotiana benthamiana Nicotiana benthamiana Tobacco Tobacco Tobacco Tobacco Tobacco Tobacco Tobacco Tobacco Hot pepper
Eulgem et al. (2000) Eulgem et al. (2000) Eulgem et al. (2000) Eulgem et al. (2000) Eulgem et al. (2000) Liu et al. (2004)
NbWRKY2 NbWRKY3
Z-C2H2
WRKY1 NtWIZZ tWRKY1 tWRKY3/NtDBP3 tWRKY4 NtWRKY3 NtEIG-D48/NtSubD48 TIZZ CaZFP1
Liu et al. (2004) Liu et al. (2004) Menke et al. (2005) Eulgem et al. (2000) Eulgem et al. (2000) Eulgem et al. (2000) Eulgem et al. (2000) ¨ Ulker and Somssich (2004) ¨ Ulker and Somssich (2004) ¨ Ulker and Somssich (2004) Kim et al. (2004)
Regulatory nuclear proteins involved in plant defense with no known DNA-binding capabilities HDA19 IQD1 MKS1 NPR1/NIM1 NIMIN1/2/3 NPR4 SN1 NH1 NRR GmCaM-4 GmCaM-5
Arabidopsis Arabidopsis Arabidopsis Arabidopsis Arabidopsis Arabidopsis Arabidopsis Rice Rice Soybean Soybean
Zhou et al. (2005) Levy et al. (2005) Andreasson et al. (2005) Durrant and Dong (2004) Weigel et al. (2005) Lorenzo and Solano (2005) Durrant and Dong (2004) Chern et al. (2005) Chern et al. (2005) Park et al. (2004) Park et al. (2004)
The best examined case for the involvement of ERF/AP2-type TFs in disease resistance is Pti4 from tomato. Pti4 was isolated along with two other ERF proteins by their ability to interact with Pto, a protein kinase encoded by a major R-gene conferring resistance to bacterial speck disease (Zhou et al., 1997). Pti4 was shown to be specifically phosphorylated by Pto and thereby to increase its binding affinity to the GCC boxes present in several defense gene promoters (Gu et al., 2000). Expression of several ERF genes was shown to be induced by ethylene or JA treatments and by pathogen infections (Gutterson and Reuber, 2004). Ectopic expression of Arabidopsis ERF1 or the tomato Pti4 in Arabidopsis, as well as the tomato ERF gene TSRF1 in both tomato and tobacco plants, resulted in improved resistance toward several tested pathogens (Berrocal-Lobo and Solano, 2002; Gu et al., 2002; Zhang et al., 2004). For tobacco NtERF5, overexpression had no effect on infections
TRANSCRIPTIONAL REGULATION UNDERLYING PLANT DEFENSE
273
with the bacterial pathogen Pseudomonas syringae (P. syringae) but significantly enhanced resistance toward Tobacco Mosaic Virus (Fischer and Dr¨oge-Laser, 2004). The ERF genes Tsi1 and OPBP1 from tobacco and CaPF1 and pCaERFLP1 from pepper are induced by biotic and abiotic stresses (Park et al., 2001; Guo et al., 2004; Lee et al., 2004b; Yi et al., 2004). All four TFs showed in vitro binding to the GCC box, and with the exception of OPBP1, also to the drought-, cold-, and osmoticresponse DNA element 5 -A/GCCGACAT-3 , designated DRE/CRT, known to be bound by a subset of the ERF/AP2 family (Shinozaki et al., 2003). Thus, a dual function of these TFs in response to various stresses could be expected. Indeed, Tsi1 overexpression in transgenic tobacco and pepper plants, and OPBP1 in tobacco, enhanced resistance to multiple tested phytopathogens and also resulted in salt-tolerant plants (Park et al., 2001; Guo et al., 2004). Similarly, high constitutive expression of pCaERFLP1 in tobacco rendered plants resistant to P. syringae infection and more tolerant to high salinity (Lee et al., 2004b). Transgenic Arabidopsis plants ectopically expressing CaPF1 showed enhanced resistance to P. syringae infection, but in this case the plants displayed tolerance against freezing temperatures (Yi et al., 2004). Consistent with a dual role, both CaPF1 and pCaERFLP1 transgenic plants had elevated transcript levels of pathogen- and stress-response genes containing either GCC or DRE/CRT boxes in their promoters. Most of the characterized ERF factors function as transcriptional activators. Recently however, ERF members capable of actively repressing transcription have also been reported (Yang et al., 2005). For one, NbCD1, a negative regulatory function associated with nonhost resistance and pathogen-triggered cell death, was identified (Nasir et al., 2005). Although experiments based exclusively on overproduction of TFs must be dealt with caution, the results outlined above are consistent with an important contribution of ERFs to plant defense. Less well studied is another group of AP2-type TFs, designated Octadecanoid-Responsive Catharanthus-APETALA2-domain proteins (ORCAs), which activate some JA-responsive genes involved in alkaloid metabolism via binding to Jasmonate- and Elicitor-Responsive Element (JERE) promoter sites (Memelink et al., 2001). Several members of the zinc-finger-type WRKY TFs have been implicated in defense-related plant responses. WRKY-binding sites (W-boxes) are often found enriched in promoters of coregulated defense genes. Moreover, in Arabidopsis, expression of over 70% of the entire WRKY gene family is strongly affected by ¨ various stresses, particularly by pathogen-related stimuli (Ulker and Somssich, 2004). In Arabidopsis, AtWRKY22 and AtWRKY29 were found to be involved in a PAMP-mediated defense-signaling pathway, and transient overexpression of AtWRKY29 conferred resistance to an otherwise virulent bacterial pathogen (Asai et al., 2002). Likewise, overexpression of AtWRKY70 in Arabidopsis plants resulted in enhanced resistance to two tested virulent bacteria (Li et al., 2004). In tobacco, virus-induced gene silencing of three WRKY genes (NtWRKY1–3) compromised N -gene-mediated resistance toward Tobacco Mosaic Virus (Liu et al., 2004). ChIP analyses in parsley revealed elicitor-dependent and transient in vivo binding of a WRKY factor (PcWRKY1) to the promoters of PcWRKY1 and PcPR1-1, two
274
REGULATION OF TRANSCRIPTION IN PLANTS
strongly pathogen-responsive genes (Turck et al., 2004). The Arabidopsis major R-gene, RRS1, which confers resistance to bacterial wilt, represents a novel type of WRKY gene. This gene, also designated AtWRKY52, codes for a protein containing structural features of NBS-LRR-type R-gene products and a C-terminal WRKY domain (Deslandes et al., 2002). RRS1/AtWRKY52 appears to be translocated to the plant cell nucleus together with the bacterial avr product PopP2, but it is currently unclear whether it regulates transcription of target genes (Deslandes et al., 2003). Another class of TFs found in diverse plants are the Whirly TFs (Why) (Desveaux et al., 2005). These preferentially bind to the single-stranded form of their respective PBF-2 binding DNA element (PB box). Elicitor-dependent in vivo binding activity of StWhy1 to the promoter of the pathogen-response gene PR-10a was demonstrated by ChIP experiments in potato. Furthermore, missense mutants of AtWhy1 in Arabidopsis showed enhanced susceptibility toward oomycete infections, suggesting a key role of the factor in plant defense (Desveaux et al., 2005). Members of several other classes of TF genes were found to be rapidly induced upon pathogen attack in global expression data (Chen et al., 2002) or from individual studies, indicating their potential in regulating the defense transcriptome (Table 11.2). Nonetheless, our knowledge on these factors is rather limited. However, a clear link has been established between two Arabidopsis MYB family members, AtMYB30 and BOS1, and plant defense. AtMYB30 appears to act as a positive regulator of the hypersensitive (HR) cell death program induced in response to pathogen challenge (Vailleau et al., 2002). Overexpression of AtMYB30 in Arabidopsis and tobacco led to enhanced HR cell death and increased resistance to different bacterial pathogens, whereas suppression of AtMYB30 expression had converse effects. In the case of BOS1, loss-of-function mutants displayed enhanced disease symptoms toward both biotrophic and necrotrophic pathogens (Mengiste et al., 2003). Expression of BOS1 is induced upon pathogen treatment via the JA-signaling pathway and BOS1 itself appears to regulate the expression of a subset of JA-responsive genes. One Arabidopsis member of the bHLH family, AtMYC2/JAI1/JIN1, was shown to have dual functions, acting as a positive regulator of abiotic stress signaling and as a negative regulator of the JA and ethylene defense pathways (Anderson et al., 2004; Lorenzo et al., 2004). Interestingly, AtMYC2/JAI1/JIN1 was shown to repress JA-mediated genes related to pathogen defense, but to activate genes involved in JA-mediated responses to wounding (Lorenzo et al., 2004). Mutants of AtMYC2/JAI1/JIN1 showed enhanced JA- and ethylene-dependent resistance to three different necrotrophic fungal pathogens. Ectopic expression of Arabidopsis ATAF2, a NAC-domain TF family member (Olsen et al., 2005), resulted in repression of several defense marker genes and in heightened susceptibility toward the soil-borne fungal pathogen Fusarium oxysporum (Delessert et al., 2005). Conversely, ATAF2 knockout plants exhibited elevated transcript levels of these genes, consistent with a repressor function of this TF in the defense response. Finally, overexpression of a pepper Cys2 /His2 -type zinc-finger TF, CAZFP1, in Arabidopsis, led to enhanced resistance of the plants to a bacterial pathogen (Kim et al., 2004).
TRANSCRIPTIONAL REGULATION UNDERLYING PLANT DEFENSE
11.5
275
Transcriptional networks
Understanding how the transcriptional regulatory network is linked to the various defense-signaling cascades governing cellular responses is of utmost importance. We need to know how the signal is transduced to the DNA-binding transcriptional regulators and to define the intermediate nuclear interactions that are essential for proper control of defense gene expression. Recently, downstream components of distinct MAP kinase signaling pathways involved in plant defense have been uncovered (Pedley and Martin, 2005). In rice, a MAP kinase designated BWMK1 was discovered, whose activity was rapidly induced by pathogen-mimicking signals (Cheong et al., 2003). BWMK1 localizes exclusively to the cell nucleus, and can phosphorylate the AP2-type TF OsEREBP1 in vitro, thereby increasing the binding affinity of this factor to GCC-box-containing DNA sequences. This is reminiscent of the already mentioned interaction of the Pto4 kinase with ERF factors in tomato (Zhou et al., 1997). In Arabidopsis, the MAP kinase 4 mutant, mpk4, exhibited constitutive SAR and resistance to virulent pathogens (Petersen et al., 2000). MPK4 interacts with MKS1, a nuclear-localized protein of unknown function, which can bind two WRKY TFs, namely AtWRKY25 and AtWRKY33 (Andreasson et al., 2005). MPK4 is capable of phosphorylating both MKS1 and the WRKY factors, and it is hypothesized that MKS1 acts as a coupling protein affecting the activities of the TFs in MPK4-mediated signaling. WRKY factors have also been implicated as downstream components in a defined PAMP-mediated MAP kinase defensesignaling cascade in Arabidopsis (Asai et al., 2002). In this case, a signal triggered by the interaction of a bacterial flagellin with the FLS2 receptor was transduced via MEKK1, MKK4/MKK5, and MPK3/MPK6, leading to gene activation of AtWRKY22 and AtWRKY29. However, it remains unclear what the direct nuclear partner(s) of MPK3/MPK6 signaling are. Similarly, in parsley, a PAMP-mediated MAP kinase cascade involving PcMKK5 and PcMPK3/PcMPK6 was found, and PcMPK3/PcMPK6 activities were observed in the nucleus upon stimulation, indicating that they exert their functions within this compartment (Lee et al., 2004a). In this system, two subsequent early nuclear steps were shown to be regulated by WRKY TFs (Eulgem et al., 1999). However, a direct link to the MAP kinases has not yet been established. Recently, the tobacco WRKY1 was shown to be directly phosphorylated by the MAP kinase SIPK and to mediate HR-like cell death (Menke et al., 2005). Currently, the best elucidated signaling cascade leading to defense gene activation is the SAR pathway (Durrant and Dong, 2004; Fobert and Despres, 2005). In Arabidopsis, pathogen infection leads to elevated levels of the endogenous signal molecule SA. This increase in SA alters the redox state of the cell allowing NPR1, which is sequestered in an inactive oligomeric form in the cytoplasm, to be reduced to its monomeric active state and thus to be translocated to the nucleus. Within the nucleus, NPR1 physically interacts with TGA factors such as TGA1 or TGA4 and thereby stimulates their binding to as-1 promoter sites within a set of immediateearly response type PR genes. This interaction requires not only the reduced state of NPR1, but also the SA-dependent reduction of two key cysteine residues within both TGA1 and TGA4 (Despr´es et al., 2003). NPR1 however does not appear to be part of
276
REGULATION OF TRANSCRIPTION IN PLANTS
the TGA complex once it is associated with DNA (Despr´es et al., 2003). Apart from this, NPR1 also directly regulates another set of immediate-early SAR-response genes containing TL1 elements within their promoters (Wang et al., 2005). Since NPR1 shows no obvious DNA-binding activity and TGA factors do not bind to TL1 elements, activation of this set of genes requires interaction of NPR1 with other so far unknown TF(s). Potential candidates for such factors are the NIMIN1/2/3 proteins, small nuclear-localized acidic proteins that have been shown to interact with NPR1 (Weigel et al., 2005). However, there is no evidence suggesting that these proteins can bind TL1 sites or even DNA. Interestingly, NPR1 itself appears to be transcriptionally controlled by WRKY factors, as W-box mutations within the NPR1 promoter strongly compromised its expression (Yu et al., 2001). The role of redox control in regulating other transcriptional processes implicated in plant defense may be more significant than previously anticipated. Microbial infections often trigger a rapid production of reactive oxygen intermediates (ROI) as signaling molecules in the host (Mittler et al., 2004). Increased ROI generation/signaling can modulate defense gene expression but the level of ROI must be kept in check due to their cellular toxicity. This is achieved by major ROI-scavenging enzymes and also by oxidoreductases such as thioredoxin and glutaredoxins, which are responsible for controlling and maintaining cellular redox homeostasis. Oxidoreductases are capable of modifying TFs, thereby directly affecting gene expression patterns (Tron et al., 2002). In this respect, it is interesting to note that activation of the oxidative stress and pathogen-responsive Arabidopsis thioredoxin h5 gene appears to be mediated by W-box promoter elements, and that plants ectopically overexpressing AtWRKY6 showed upregulated expression of this gene (Laloi et al., 2004). A number of other Arabidopsis nuclear-localized components have been partly characterized that seem to play important, albeit ill-defined, roles in the plant defense response. One gene designated SNI1 was uncovered in a mutant screen in search of suppressors of npr1. sni1 mutants restore SA responsiveness of PR genes and pathogen resistance in the npr1 background. SNI1 is localized to the nucleus but appears to lack DNA-binding activity and may act as a repressor of a general transcriptional function common between yeast and plants (Durrant and Dong, 2004). The AtWRKY70 TF has been reported to play a crucial regulatory role at the convergence point of the SA- and JA-signaling pathways, positively regulating SAresponse genes and negatively regulating JA-response genes (Li et al., 2004). A gene designated IQD1 was detected in a screen of T-DNA activation-tagged lines (Levy et al., 2005). Upregulation of IQD1 stimulated glucosinolate accumulation, a small and diverse class of defense-associated secondary metabolites, and also reduced insect herbivory. IQD1 is a nuclear-localized basic protein of 50.5 kDa containing several functional calmodulin-binding sites but appears to lack any DNA- or RNAbinding motifs. Finally, plants overexpressing HDA19 encoding a histone deacetylase had decreased histone acetylation levels, showed increased expression of the ERF1 TF gene, and were more resistant to the necrotroph Alternaria brassicicola (Zhou et al., 2005). In contrast, downregulation of HDA19 resulted in plants being more sensitive to this pathogen. In rice, a small nuclear-localized proline-rich protein, designated NRR, was identified by its ability to interact with NPR1 as well as with the rice NPR1 homolog
TRANSCRIPTIONAL REGULATION UNDERLYING PLANT DEFENSE
277
NH1 (Chern et al., 2005). NRR shares limited similarity to the Arabidopsis NIMIN2 and tobacco G8-1 proteins. Constitutive overexpression of NRR in rice suppressed defense responses and resistance to the bacterial pathogen Xanthomonas oryzae implying a negative function of NRR in plant defense. Silencing of NRR expression did not affect the response of plants toward this pathogen, suggesting that other rice proteins may play redundant roles to NRR.
11.6
Defense gene suppression by pathogens
How many of the vast number of host genes that are altered in their expression during a plant–pathogen interaction are actually involved in plant defense is difficult to say. What has become evident however is that some of these host genes, designated susceptibility genes, are actually targets of the pathogen, thereby enabling them to overcome certain resistance barriers (Abramovitch and Martin, 2004; Nomura et al., 2005). This is probably part of the ongoing coevolutionary tug-of-war for survival between plants and microbes. Indeed, global expression studies with Arabidopsis have revealed that bacterial pathogens deliver numerous effector molecules (products of the avr genes) into the host that act to suppress the expression of defense genes triggered by the PAMP surveillance system of the plant (Mudgett, 2005; N¨urnberger and Lipka, 2005). Similarly, microarray analyses revealed that infections by a virulent powdery mildew fungus either suppressed or failed to activate the JA/ethylene defense-signaling pathways (Zimmerli et al., 2004). Ectopic activation of the JA/ethylene pathway resulted in protection against this powdery mildew as well as against a virulent oomycete, indicating that avoidance/suppression of such defenses is required for successful pathogen colonization. One well-established case is the Arabidopsis NHO1 gene coding for a glycerol kinase required for resistance to nonhost P. syringae strains and also to the fungal pathogen Botrytis cinerea (Kang et al., 2003). Expression of this gene is strongly induced by nonhost bacterial strains but actively suppressed by virulent strains targeting the JA-dependent pathway (Li et al., 2005). Ectopic expression of NHO1 led to increased plant resistance to a virulent strain, further supporting its involvement in plant defense. Another example of the suppression of defense genes comes from findings that coronatine, a phytotoxin produced by several pathovars of P. syringae, completely mimics the plant signal molecule methyl jasmonate (MeJA). Plant resistance toward bacteria such as Pseudomonas is established by activation of the SA defense pathway. Coronatine, acting as MeJA, activates the JA defense pathway and thereby suppresses the SA pathway required to restrict growth of these bacteria (Nomura et al., 2005).
11.7
Summary and outlook
We have made substantial progress in understanding the importance of transcription in activating plant defense responses against phytopathogens. Equally evident however is that we are just at the beginning of our efforts to uncover and dissect the complex and highly interconnected defense transcriptional network. For example,
278
REGULATION OF TRANSCRIPTION IN PLANTS
only recently have we begun to realize that abiotic stress signaling controlled by the key hormone abscisic acid is also interwoven into this network (Mauch-Mani and Mauch, 2005). Abscisic acid signaling is generally considered to negatively regulate disease resistance, and thus the spectrum of defense genes and/or the transcriptional regulators involved may be quite different. Eukaryotic gene promoters are made up of a composite of cis-regulatory DNA elements capable of being bound by different classes of TFs, which are themselves, embedded in larger regulatory complexes. Often, both the amplitude and the duration of transcription will be determined by specific TF interactions as well as by discrete transient TF associations/dissociations at a given promoter. Recent advances in molecular and computational biology are opening the way for us to begin constructing and validating transcriptional regulatory networks (Barabasi and Oltvai, 2004; Blais and Dynlacht, 2005). Thanks to the availability of fully sequenced plant genomes, we can now deduce the complete set of TFs within these species. Highthroughput approaches to identify all potential TF–TF interactions have been established (Immink and Angenent, 2002) and should provide valuable clues concerning candidate interactors. Gene expression profiling supplemented with data derived from the use of the large collections of available insertional Arabidopsis TF mutants will help to pinpoint distinct sets of coregulated TFs and downstream defense genes. It is foreseeable that such studies will be complemented by the genome-wide application of the ChIP technique (ChIP-chip, STAGE) to determine the entire spectrum of in vivo TF target genes (Gao et al., 2004; Hanlon and Lieb, 2004; Kim et al., 2005). Using bioinformatics tools to merge the data of these several sources, we can start to create an extensive map of the transcriptional network defining the plant defense response. There are, however, several additional levels of regulation that need to be considered. Protein–protein and protein–DNA interactions can be highly dynamic, making definition of the entire transcription network a daunting endeavor. TFs and associated components can undergo various modifications including phosphorylation (Holmberg et al., 2002), acetylation (Sng et al., 2004), ubiquitination (Conaway et al., 2002), and sumoylation (Miura et al., 2005). Such alterations may profoundly alter TF functions. There is also an emerging recognition that chromatin structure can act both locally and globally to regulate gene expression (Felsenfeld and Groudine, 2003). For instance, modifications of chromatin proteins such as histone H3 can directly affect expression at specific promoter sites through their interaction with TFs. In particular, intracellular signaling through MAP kinase cascades can directly act to modify chromatin proteins (Clayton and Mahadevan, 2003; de Nadal et al., 2004). Considering the extensive involvement of MAP kinases in plant defense (Nakagami et al., 2005), we can expect that defense gene expression will in part be regulated by such chromatin modifications. Beyond this, an increasing body of evidence indicates that nuclear architecture is linked to the organization and sorting of regulatory information, thereby representing another level of complexity that needs to be taken into consideration (Stein et al., 2003). Finally, newly uncovered transcriptional regulatory mechanisms involving small RNAs (Willmann and Poethig, 2005) must also be taken into account.
TRANSCRIPTIONAL REGULATION UNDERLYING PLANT DEFENSE
279
In summary, although the key mechanisms and individual constituents of the defense transcriptome network are becoming apparent, our understanding is still inadequate to explain the elaborate molecular apparatus that allows regulatory discrimination and selective trafficking of proteins to translate into specific transcriptional outputs. In particular, there is a need to capture, purify, and characterize the highly dynamic, multimeric transcriptional complexes at distinct promoter sites following signaling through defined defense pathways. Furthermore, we need to know the biochemical basis of how the convergence of synergistic and antagonistic signals at a given promoter is resolved. Improvements in computational biology and in rapidly evolving ultrasensitive high-throughput methodologies, as illustrated by protein mass spectrometry, will certainly facilitate our basic understanding of the transcriptional regulatory defense network. Beyond this however, we will need to delineate the topology and dynamics of this gene network in order to fully grasp the biological complexity governing plant defense responses.
Acknowledgments I am grateful to Drs George Coupland and Richard O’Connell (MPIZ K¨oln) for critical reading of the manuscript. I apologize to all my colleagues whose original publications I failed to cite due to space limitations.
References R.B. Abramovitch and G.B. Martin (2004) Strategies used by bacterial pathogens to suppress plant defenses. Current Opinion in Plant Biology 7, 356–364. J.P. Anderson, E. Badruzsaufari, P.M. Schenk, et al. (2004) Antagonistic interaction between abscisic acid and jasmonate–ethylene signaling pathways modulates defense gene expression and disease resistance in Arabidopsis. The Plant Cell 16, 3460–3479. E. Andreasson, T. Jenkins, P. Brodersen, et al. (2005) The MAP kinase substrate MKS1 is a regulator of plant defense responses. The EMBO Journal 24, 2579–2589. T. Asai, G. Tena, J. Plotnikova, et al. (2002) MAP kinase signalling cascade in Arabidopsis innate immunity. Nature 415, 977–983. A.-L. Barabasi and Z.N. Oltvai (2004) Network biology: understanding the cell’s functional organization. Nature Reviews Genetics 5, 101–113. O. Batz, E. Logemann, S. Reinold and K. Hahlbrock (1998) Extensive reprogramming of primary and secondary metabolism by fungal elicitor or infection in parsley cells. Biological Chemistry 379, 1127–1135. Y. Belkhadir, R. Subramaniam and J.L. Dangl (2004) Plant disease resistance protein signaling: NBSLRR proteins and their partners. Current Opinion in Plant Biology 7, 391–399. M. Berrocal-Lobo and R. Solano (2002) Constitutive expression of ETHYLENE-RESPONSEFACTOR1 in Arabidopsis confers resistance to several necrotrophic fungi. The Plant Journal 29, 23–32. A. Blais and B.D. Dynlacht (2005) Constructing transcriptional regulatory networks. Genes and Development 19, 1499–1511. R.M. Bostock (2005) Signal crosstalk and induced resistance: straddling the line between cost and benefit. Annual Review of Phytopathology 43, 545–580.
280
REGULATION OF TRANSCRIPTION IN PLANTS
V. Buchanan-Wollaston, S. Earle, E. Harrison, et al. (2003) The molecular analysis of leaf senescence – a genomics approach. Plant Biotechnology Journal 1, 3–22. W. Chen, N.J. Provart, J. Glazebrook, et al. (2002) Expression profile matrix of Arabidopsis transcription factor genes suggests their putative functions in response to environmental stresses. The Plant Cell 14, 559–574. Y.H. Cheong, H.-S. Chang, R. Gupta, X. Wang, T. Zhu and S. Luan (2002) Transcriptional profiling reveals novel interactions between wounding, pathogen, abiotic stress, and hormonal responses in Arabidopsis. Plant Physiology 129, 661–677. Y.H. Cheong, B.C. Moon, J.K. Kim, et al. (2003) BWMK1, a rice mitogen-activated protein kinase, locates in the nucleus and mediates pathogenesis-related gene expression by activation of a transcription factor. Plant Physiology 132, 1961–1972. M.S. Chern, H.A. Fitzgerald, P.E. Canlas, D.A. Navarre and P.C. Ronald (2005) Ovrexpression of a rice NPR1 homolog leads to constitutive activation of defense response and hypersensitivity to light. Molecular Plant–Microbe Interactions 18, 511–520. Z. Chu, Y. Ouyang, J. Zhang, H. Yang and S. Wang (2004) Genome-wide analysis of defense-responsive genes in bacterial blight resistance of rice mediated by the recessive R gene xa13. Molecular Genetics and Genomics 271, 111–120. A.L. Clayton and L.C. Mahadevan (2003) MAP kinase-mediated phosphorylation of histone H3 and inducible gene regulation. FEBS Letters 546, 51–58. R.C. Conaway, C.S. Brower and J.W. Conaway (2002) Emerging roles of ubiquitin in transcription regulation. Science 296, 1254–1258. E. De Nadal, M. Zapater, P.M. Alepuz, L. Sumoy, G. Mas and F. Posas (2004) The MPK Hog1 recruits Rpd3 histone deacteylase to activate osmoresponsive genes. Nature 427, 370–374. C. Delessert, K. Kazan, I.W. Wilson, et al. (2005) The transcription factor ATAF2 represses the expression of pathogenesis-related genes in Arabidopsis. The Plant Journal 43, 745–757. L. Deslandes, J. Olivier, N. Peeters, et al. (2003) Physical interaction between RRS1-R, a protein conferring resistance to bacterial wilt, and PopP2, a type III effector targeted to the plant nucleus. Proceedings of the National Academy of Sciences of the United States of America 100, 8024–8029. L. Deslandes, J. Olivier, T. Theuli`eres, et al. (2002) Resistance to Ralstonia solanacearum in Arabidopsis thaliana is confered by the recessive RRS1-R gene, a member of a novel family of resistance genes. Proceedings of the National Academy of Sciences of the United States of America 99, 2404–2409. C. Despr´es, C. Chubak, A. Rochon, et al. (2003) The Arabidopsis NPR1 disease resistance protein is a novel cofactor that confers redox regulation of DNA binding activity to the basic domain/leucine zipper transcription factor TGA1. The Plant Cell 15, 2181–2191. D. Desveaux, A. Mar´echal and N. Brisson (2005) Whirly transcription factors: defense gene regulation and beyond. Trends in Plant Science 10, 95–102. D. Desveaux, R. Subramaniam, C. Despr´es, et al. (2004) A ‘Whirly’ transcription factor is required for salicylic acid-dependent disease resistance in Arabidopsis. Developmental Cell 6, 229–240. W.E. Durrant and X. Dong (2004) Systemic acquired resistance. Annual Review of Phytopathology 42, 185–209. W.E. Durrant, O. Rowland, P. Piedras, K.E. Hammond-Kosack and J.D.G. Jones (2000) cDNA-AFLP reveals a striking overlap in race-specific resistance and wound response gene expression profiles. The Plant Cell 12, 963–977. C. Eckey, M. Korell, K. Leib, et al. (2004) Identification of powdery mildew-induced barley genes by cDNA-AFLB: functional assessment of an early expressed MAP kinase. Plant Molecular Biology 55, 1–15. T. Eulgem (2005) Regulation of the Arabidopsis defense transcriptome. Trends in Plant Science 10, 71–78. T. Euglem, P.J. Rushton, S. Robatzek and I.E. (2000) The WRKY superfamily of plant transcription factors. Trends in Plant Science 5, 199–206. T. Eulgem, P.J. Rushton, E. Schmelzer, K. Hahlbrock and I.E. Somssich (1999) Early nuclear events in plant defense: rapid gene activation by WRKY transcription factors. The EMBO Journal 18, 4689–4699. G. Felsenfeld and M. Groudine (2003) Controlling the double helix. Nature 421, 448–453.
TRANSCRIPTIONAL REGULATION UNDERLYING PLANT DEFENSE
281
U. Fischer and W. Dr¨oge-Laser (2004) Overexpression of NtERF5, a new member of the tobacco ethylene response transcription factor family enhances resistance to Tobacco mosaic virus. Molecular Plant–Microbe Interactions 17, 1162–1171. H.A. Fitzgerald, P.E. Canlas, M.S. Chern and P.C. Ronald (2005) Alteration of TGA factor activity in rice results in enhanced tolerance to Xanthomonas oryzae pv. oryzae. The Plant Journal 43, 335–347. P.R. Fobert and C. Despres (2005) Redox control of systemic acquired resistance. Current Opinion in Plant Biology 8, 378–382. Y. Gao, J. Li, E. Strickland, et al. (2004) An Arabidopsis promoter microarray and its initial usage in the identification of HY5 binding targets in vitro. Plant Molecular Biology 54, 683–699. A. Gibly, A. Bonshtien, V. Balaji, P. Debbie, G.B. Martin and G. Sessa (2004) Identification and expression profiling of tomato genes differentially regulated during a resistance response to Xanthomonas campestris pv. vesicatoria. Molecular Plant–Microbe Interactions 17, 1212–1222. J. Glazebrook (2005) Contrasting mechanisms of defense against biotrophic and necrotrophic pathogens. Annual Review of Phytopathology 43, 205–227. K. Gu, B. Yang, D. Tian, et al. (2005) R gene expression induced by a type-III effector triggers disease resistance in rice. Nature 435, 1122–1125. Y.-Q. Gu, M.C. Wildermuth, S. Chakravarthy, et al. (2002) Tomato transcription factors Pti4, Pti5, and Pti6 activate defense responses when expressed in Arabidopsis. The Plant Cell 14, 817–831. Y.-Q. Gu, C. Yang, V.K. Thara, J. Zhou and G.B. Martin (2000) Pti4 is induced by ethylene and salicylic acid, and its product is phosphorylated by the Pto kinase. The Plant Cell 12, 771–785. Z.-J. Guo, X.-J. Chen, X.-L. Wu, J.-Q. Ling and P. Xu (2004) Overexpression of the AP2/EREBP transcription factor OPBP1 enhances disease resistance and salt tolerance in tobacco. Plant Molecular Biology 55, 607–618. N. Gutterson and L.T. Reuber (2004) Reglation of disease resistance pathways by AP2/ERF transcription factors. Current Opinion in Plant Biology 7, 465–471. K. Hahlbrock, P. Bednarek, I. Ciolkowski, et al. (2003) Non-self recognition, transcriptional reprogramming, and secondary metabolite accumulation during plant/pathogen interactions. Proceedings of the National Academy of Sciences of the United States of America 100, 14569–14576. S.E. Hanlon and J.D. Lieb (2004) Progress and challenges in profiling the dynamics of chromatin and transcription factor binding with DNA microarrays. Current Opinion in Genetics and Development 14, 697–705. C.I. Holmberg, S.E.F. Tran, J.E. Eriksson and L. Sistonen (2002) Multisite phosphorylation provides sophisticated regulation of transcription factors. Trends in Biochemical Sciences 27, 619–627. R.G.H. Immink and G.C. Angenent (2002) Transcription factors do it together: the hows and whys of studying protein–protein interactions. Trends in Plant Science 7, 531–534. M. Jakoby, B. Weisshaar, W. Dr¨oge-Laser, et al. (2002) bZIP transcription factors in Arabidopsis. Trends in Plant Science 7, 106–111. C. Johnson, E. Boden and J. Arias (2003) Salicylic acid and NPR1 induce the recruitment of transactivating TGA factors to a defense gene promoter in Arabidopsis. The Plant Cell 15, 1846–1858. D.A. Jones and D. Takemoto (2004) Plant innate immunity – direct and indirect recognition of general and specific pathogen-associated molecules. Current Opinion in Immunology 16, 48–62. L. Kang, J. Li, T. Zhao, et al. (2003) Interplay of the Arabidopsis nonhost resistance gene NHO1 with bacterial virulence. Proceedings of the National Academy of Sciences of the United States of America 100, 3519–3524. F. Katagiri (2004) A global view of defense gene expression regulation – a highly interconnected signaling network. Current Opinion in Plant Biology 7, 506–511. H.S. Kim and T.P. Delaney (2002) Over-expression of TGA5, which encodes a bZIP transcription factor that interacts with NIM1/NPR1, confers SAR-independent resistance to Arabidopsis thaliana to Peronospora parasitica. The Plant Journal 32, 151–163. J. Kim, A.A. Bhinge, X.C. Morgan and V.R. Iyer (2005) Mapping DNA–protein interactions in large genomes by sequence tag analysis of genomic enrichment. Nature Methods 2, 47–53. S.H. Kim, J.K. Hong, S.C. Lee, K.H. Sohn, H.W. Jung and B.K. Hwang (2004) CAZFP1, Cys2 /His2 type zinc-finger transcription factor gene functions as a pathogen-induced early-defense gene in Capsicum annuum. Plant Molecular Biology 55, 883–904.
282
REGULATION OF TRANSCRIPTION IN PLANTS
C. Kirsch, E. Logemann, B. Lippok, E. Schmelzer and K. Hahlbrock (2001) A highly specific pathogenresponsive promoter element from the immediate-early activated CMPG1 gene in Petroselinum crispum. The Plant Journal 26, 217–227. C. Kirsch, M. Takamiya-Wik, E. Schmelzer, K. Hahlbrock and I.E. Somssich (2000) A novel regulatory element involved in rapid activation of parsley ELI7 gene family members by fungal elicitor or pathogen infection. Molecular Plant Pathology 1, 243–251. C. Laloi, D. Mestres-Ortega, Y. Marco, Y. Meyer and J.-P. Reichheld (2004) The Arabidopsis cytosolic thioredoxin h5 gene induction by oxidative stress and its W-box-mediated response to pathogen elicitor. Plant Physiology 134, 1006–1016. E. Lebel, P. Heifetz, L. Thorne, S. Uknes, J. Ryals and E. Ward (1998) Functional analysis of the regulatory sequences controlling PR-1 gene expression in Arabidopsis. The Plant Journal 16, 223–233. J. Lee, J.J. Rudd, V.K. Macioszek and D. Scheel (2004a) Dynamic changes in the localization of MAP kinase cascade components controlling pathogenesis-related (PR) gene expression during innate immunity in parsley. Journal of Biological Chemistry 279, 22440–22448. J.-H. Lee, J.-P. Hong, S.-K. Oh, S. Lee, D. Choi and W. Kim (2004b) The ethylene-responsive factor like protein 1 (CaERFLP1) of hot pepper (Capsicum annuum L.) interacts in vitro with both GCC and DRE/CRT sequences with different binding affinities: possible biological roles of CaERFLP1 in response to pathogen infection and high salinity conditions in transgenic tobacco plants. Plant Molecular Biology 55, 61–81. M. Levy, Q. Wang, R. Kaspi, M.P. Parrella and S. Abel (2005) Arabidopsis IQD1, a novel calmodulinbinding nuclear protein, stimulates glucosinolate accumulation and plant defense. The Plant Journal 43, 79–96. J. Li, G. Brader and E.T. Palva (2004) The WRKY70 transcription factor: a node of convergence for jasmonate-mediated and salicylate-mediated signals in plant defense. The Plant Cell 16, 319–331. X. Li, H. Lin, W. Zhang, et al. (2005) Flagellin induces innate immunity in nonhost interactions that is suppressed by Pseudomonas syringae effectors. Proceedings of the National Academy of Sciences of the United States of America 102, 12990–12995. Y. Liu, M. Schiff and S.P. Dinesh-Kumar (2004) Involvement of MEK1 MAPKK, NTF6 MAPK, WRKY/MYB transcription factors, COI1 and CTR1 in N -mediated resistance to tobacco mosaic virus. The Plant Journal 38, 800–809. O. Lorenzo, J.M. Chico, J.J. Sanchez-Serrano and R. Solano (2004) JASMONATE-INSENSITIVE1 encodes a MYC transcription factor essential to discriminate between different jasmonateregulated defense responses in Arabidopsis. The Plant Cell 16, 1938–1950. O. Lorenzo and R. Solano (2005) Molecular players regulating the jasmonate signalling network. Current Opinion in Plant Biology 8, 532–540. K. Maleck, A. Levine, T. Eulgem, et al. (2000) The transcriptome of Arabidopsis thaliana during systemic acquired resistance. Nature Genetics 26, 403–410. B. Mauch-Mani and F. Mauch (2005) The role of abscisic acid in plant–pathogen interactions. Current Opinion in Plant Biology 8, 409–414. J. Memelink, R. Verpoorte and J.W. Kijne (2001) ORCAnization of jasmonate-responsive gene expression in alkaloid metabolism. Trends in Plant Science 6, 212–219. T. Mengiste, X. Chen, J. Salmeron and R. Dietrich (2003) TheBOTRYTIS SUSCEPTIBLE1 gene encodes an R2R3MYB transcription factor protein that is required for biotic and abiotic stress responses in Arabidopsis. The Plant Cell 15, 2551–2565. F.L.H. Menke, H.-G. Kang, X. Chen, J.M. Park, D. Kumar and D.F. Klessig (2005) Tobacco transcription factor WRKY1 is phosphorylated by the MAP kinase SIPK and mediates HR-like cell death in tobacco. Molecular Plant–Microbe Interactions 18, 1027–1034. R. Mittler, S. Vanderauwera, M. Gollery and F. Van Breusegem (2004) Reactive oxygen gene network of plants. Trends in Plant Science 9, 490–498. K. Miura, A. Rus, A. Sharkhuu, et al. (2005) The Arabidopsis SUMO E3 ligase SIZ1 controls phosphate deficiency responses. Proceedings of the National Academy of Sciences of the United States of America 102, 7760–7765. M.B. Mudgett (2005) New insights to the function of phytopathogenic bacterial type III effectors in plants. Annual Review of Plant Biology 56, 509–531.
TRANSCRIPTIONAL REGULATION UNDERLYING PLANT DEFENSE
283
K.S. Mysore and C.-M. Ryu (2004) Nonhost resistance: how much do we know. Trends in Plant Science 9, 97–104. H. Nakagami, A. Pitzschke and H. Hirt (2005) Emerging MAP kinase pathways in plant stress signalling. Trends in Plant Science 10, 339–346. K.H.B. Nasir, Y. Takahashi, A. Ito, et al. (2005) High-throughput in planta expression screening identifies a class II ethylene-responsive element binding factor-like protein that regulates plant cell death and non-host resistance. The Plant Journal 43, 491–505. L. Navarro, C. Zipfel, O. Rowland, et al. (2004) The transcriptional innate immune response to flg22. Interplay and overlap with avr gene-dependent defense responses and bacterial pathogenesis. Plant Physiology 135, 1113–1128. Z. Nimchuk, T. Eulgem, B.F. Holt Iii and J.L. Dangl (2003) Recognition and response in the plant immune system. Annual Review of Genetics 37, 579–609. K. Nomura, M. Melotto and S.-Y. He (2005) Suppression of host defense in compatible plant– Pseudomonas syringae interactions. Current Opinion in Plant Biology 8, 361–368. T. N¨urnberger and F. Brunner (2002) Innate immunity in plants and animals: emerging parallels between the recognition of general elicitors and pathogen-associated molecular patterns. Current Opinion in Plant Biology 5, 318–324. T. N¨urnberger and V. Lipka (2005) Non-host resistance in plants: new insights into an old phenomenon. Molecular Plant Pathology 6, 335–345. P. Obreg´on, R. Mart´ın, A. Sanz and C. Castresana (2001) Activation of defense-related genes during senescence: a correlation between gene expression and cellular damage. Plant Molecular Biology 46, 67–77. A.N. Olsen, H.A. Ernst, L.L. Leggio and K. Skriver (2005) NAC transcription factors: structurally distinct, functionally diverse. Trends in Plant Science 10, 79–87. C.Y. Park, W.D. Heo, J.H. Yoo, et al. (2004) Pathogenesis-related gene expression by specific calmodulin isoforms is dependent on NIM1, a key regulator of systemic acquired resistance. Molecules and Cells 18, 207–213. J.M. Park, C.-J. Park, S.-B. Lee, B.K. Ham, R. Shin and K.-H. Paek (2001) Overexpression of the tobacco Tsi1 gene encoding an EREBP/AP2-type transcription factor enhances resistance against pathogen attack and osmotic stress in tobacco. The Plant Cell 13, 1035–1046. K.F. Pedley and G.B. Martin (2005) Role of mitogen-activated protein kinases in plant immunity. Current Opinion in Plant Biology 8, 541–547. M. Petersen, P. Brodersen, H. Naested, et al. (2000) Arabidopsis MAP kinase 4 negatively regulates systemic acquired resistance. Cell 103, 1111–1120. S. Rombauts, K. Florquin, M. Lescot, K. Marchal, P. Rouz´e and Y. Van De Peer (2003) Computational approaches to identify promoters and cis-regulatory elements in plant genomes. Plant Physiology 132, 1162–1176. P.J. Rushton, A. Reinst¨adler, V. Lipka, B. Lippok and I.E. Somssich (2002) Synthetic plant promoters containing defined regulatory elements provide novel insights into pathogen- and wound-induced signaling. The Plant Cell 14, 749–762. P.J. Rushton and I.E. Somssich (1999) Transcriptional regulation of plant genes responsive to pathogens and elicitors. In: Plant–Microbe Interactions (eds G. Stacey and N.T. Keen). APS Press, St Paul, MN. P.J. Rushton, J.T. Torres, M. Parniske, P. Wernert, K. Hahlbrock and I.E. Somssich (1996) Interaction of elicitor-induced DNA binding proteins with elicitor response elements in the promoters of parsley PR1 genes. The EMBO Journal 15, 5690–5700. M. Scheideler, N.L. Schlaich, K. Fellenberg, et al. (2002) Monitoring the switch from housekeeping to pathogen defense metabolism in Arabidopsis thaliana using cDNA arrays. Journal of Biological Chemistry 277, 10555–10561. K. Shinozaki, K. Yamaguchi-Shinozaki and M. Seki (2003) Regulatory network of gene expression in the drought and cold stress responses. Current Opinion in Plant Biology 6, 410–417. P. Skamnioti and C.J. Ridout (2005) Microbial avirulence determinants: guided missiles or antigenic flak? Molecular Plant Pathology 6, 551–559. J.C.G. Sng, H. Tanura and Y. Yoneda (2004) A tale of early response genes. Biological and Pharmaceutical Bulletin 27, 606–612.
284
REGULATION OF TRANSCRIPTION IN PLANTS
I.E. Somssich and K. Hahlbrock (1998) Pathogen defence in plants – a paradigm of biological complexity. Trends in Plant Science 3, 86–90. G.S. Stein, S.K. Zaidi, C.D. Braastad, et al. (2003) Functional architecture of the nucleus: organizing the regulatory machinery for gene expression, replication and repair. Trends in Cell Biology 13, 584–592. H. Thordal-Christensen (2003) Fresh insights into processes of nonhost resistance. Current Opinion in Plant Biology 6, 351–357. A.E. Tron, C.W. Bertoncini, R.L. Chan and D.H. Gonzalez (2002) Redox regulation of plant homeodomain transcription factors. Journal of Biological Chemistry 277, 34800–34807. F. Turck, A. Zhou and I.E. Somssich (2004) Stimulus-dependent, promoter-specific binding of transcription factor WRKY1 to its native promoter and the defense-related gene PcPR1-1 in parsley. The Plant Cell 16, 2573–2585. ¨ B. Ulker and I.E. Somssich (2004) WRKY transcription factors: from DNA binding towards biological function. Current Opinion in Plant Biology 7, 491–498. F. Vailleau, X. Daniel, M. Tronchet, J.-L. Montillet, C. Triantaphylid`es and D. Roby (2002) A R2R3MYB gene, AtMYB30, acts as a positive regulator of the hypersensitive cell death program in plants in response to pathogen attack. Proceedings of the National Academy of Sciences of the United States of America 99, 10179–10184. D. Wang, N.D. Weaver, M. Kesarwani and X. Dong (2005) Induction of protein secretory pathway is required for systemic acquired resistance. Science 308, 1036–1040. W.W. Wasserman and A. Sandelin (2004) Applied bioinformatics for the identification of regulatory elements. Nature Reviews Genetics 5, 276–287. R.R. Weigel, U.M. Pfitzner and C. Gatz (2005) Interaction of NIMIN1 with NPR1 modulates PR gene expression in Arabidopsis. The Plant Cell 17, 1279–1291. M.R. Willmann and R.S. Poethig (2005) Time to grow up: the temporal role of smallRNAs in plants. Current Opinion in Plant Biology 8, 548–552. Z. Yang, L. Tian, M. Latoszek-Green, D. Brown and K. Wu (2005) Arabidopsis ERF4 is a transcriptional repressor capable of modulating ethylene and abscisic acid responses. Plant Molecular Biology 58, 585–596. S.Y. Yi, J.-H. Kim, Y.-H. Joung, et al. (2004) The pepper transcription factor CaPF1 confers pathogen and freezing tolerance in Arabidopsis. Plant Physiology 136, 2862–2874. D. Yu, C. Chen and Z. Chen (2001) Evidence for an important role of WRKY DNA binding proteins in the regulation of NPR1 gene expression. The Plant Cell 13, 1527–1539. H. Zhang, D. Zhang, J. Chen, et al. (2004) Tomato stress-responsive factor TSRF1 interacts with ethylene responsive element GCC box and regulates pathogen resistance to Ralstonia solanacearum. Plant Molecular Biology 55, 825–834. Y. Zhang, M.J. Tessaro, M. Lassner and X. Li (2003) Knockout analysis of Arabidopsis transcription factors TGA2, TGA5, and TGA6 reveals their redundant and essential roles in systemic acquired resistance. The Plant Cell 15, 2647–2653. C. Zhou, L. Zhang, J. Duan, B. Miki and K. Wu (2005) HISTONE DEACETYLASE19 is involved in jasmonic acid and ethylene signaling of pathogen response in Arabidopsis. The Plant Cell 17, 1196–1204. J. Zhou, X. Tang and G.B. Martin (1997) The Pto kinase conferring resistance to tomato bacterial speck disease interacts with proteins that bind a cis-element of pathogenesis-related genes. The EMBO Journal 16, 3207–3218. L. Zimmerli, M. Stein, V. Lipka, P. Schulze-Lefert and S. Somerville (2004) Host and non-host pathogens elicit different jasmonate/ethylene responses in Arabidopsis. The Plant Journal 40, 633–646. C. Zipfel and G. Felix (2005) Plants and animals: a different taste for microbes? Current Opinion in Plant Biology 8, 353–360.
12 Temperature-regulated gene expression Friedrich Sch¨offl and Tressa Jacob Panikulangara
12.1 12.1.1
Heat stress Introduction
Temperature extremes impose variable stress severity depending on the rate of temperature change, the duration of the stress, and the intensity. The optimal/ambient growth temperatures and the heat stress temperatures vary with different plant species, as they are evolutionarily adapted to different climatic conditions. Basal levels of thermotolerance are indicated by a survival of nonacclimated plants to heat stress temperatures to up to 10–15◦ C above the optimal growth temperatures: for example, the normothermic conditions of Arabidopsis, tomato, and soybean are 25, 28, and 22◦ C respectively; the maximal heat stress temperatures range from 37 to 41◦ C. Treatments at higher temperatures would be lethal for nonacclimated plants but can be survived if plants are conditioned by a previous sublethal heat treatment. This acquisition of enhanced levels of thermotolerance or heat acclimation correlates at the molecular level with a transient reprogramming of gene expression. The expression of heat-shock proteins (HSPs) is a signature of the heat-shock response. It is well established that the same conditions that induce HSPs also cause heat inactivation and denaturation of cellular enzymes/proteins and that the HSPs act as molecular chaperons, assisting the refolding of denatured proteins (for review, see Sch¨offl et al., 1999). In all species studied, heat stress results in the production of HSPs, which have been classified into a number of families based on their molecular mass (HSP100, HSP90, HSP70, HSP60, and small (s) HSP), most of which have chaperon function (for review, see Jaenicke and Creighton, 1993; Boston et al., 1996). Plants are unique in the number and complexity of sHSP that they produce upon heat stress; no isoforms are expressed in vegetative tissue under nonstress conditions (for review, see Jakob and Buchner, 1994; Sch¨offl et al., 1998).
12.1.2
Heat stress sensing and signaling
Despite the ubiquitous nature of the heat-shock response, little is known how plants sense heat stress or about the signaling pathways resulting in heat-shock gene expression. In all organisms the heat-shock response is primarily regulated at the transcriptional level by heat-shock transcription factors (HSFs), which are activated by stress for a specific binding to heat-shock promoter elements (HSEs). Compared to cold stress the mechanisms of stress sensing and signaling in response to heat stress are poorly understood, although, hormonal, calcium, and redox signals have been
286
REGULATION OF TRANSCRIPTION IN PLANTS
implicated (Foyer et al., 1997; Clarke et al., 2000; Larkindale and Knight, 2002). However, there is no evidence for an involvement of membrane-bound receptors. The HSFs are considered as the central components for sensing and regulation of the heat-shock response. Plant HSFs belong to a multigene family comprising constitutive and stress-inducible members. The existence of heat-inducible HSFs suggests multistep mechanisms in target gene expression. The ‘chaperon titration’ model (Figure 12.1), adopted from animal and mammalian systems (Sch¨offl et al., 1998), integrates the different phenomena and known molecular components, which are involved in the regulation of HSF activity. Crucial to this model is a stress-induced imbalance between available chaperons and the load of denatured proteins. HSP-like chaperons are present in limited amounts under nonstress conditions but HSPs become major components after stress. The model proposes that under nonstress conditions HSF is already present but in an inactive HSF–HSP-chaperon-complex. Heat stress or certain other environmental stresses cause denaturation of cellular proteins, hence causing a higher demand of chaperon-refolding capacity. A competition for chaperons by denatured proteins leads to withdrawal of HSPs from HSF-chaperon-complexes, which is prerequisite for the formation of active HSF oligomers (Sch¨offl et al., 1998). This transition from monomeric (repressed) to trimeric (activated) state of HSF is the first step in HSF activation. Activated HSF can bind to the HSE of target genes and stimulate their transcription. The subsequent accumulation of HSPs, the major products of the heat-shock response, meet the increased demand for chaperon capacity, and excess of HSPs would cause a feedback regulation by interacting with HSF, resulting in the formation of an inactive HSF-chaperon-complex. This model is in accordance with the transient nature of the heat-shock response. HSF modifications like phosphorylation or redox reactions (Reindl and Sch¨offl, 1997; Ahn and Thiele, 2003) may be involved in regulating HSF activity but have not been much studied in plants.
12.1.3
Heat stress transcription factors: structure and function
HSFs are the central regulators of the heat-shock response. Plant HSFs display a modular structure with a highly conserved N-terminal DNA-binding domain (DBD), which is characterized by a central helix-turn-helix motif, an adjacent bipartite oligomerization domain (HR-A/B) containing hydrophobic heptad repeats for HSF trimerization via the formation of a triple-stranded α-helical coiled coil, nuclear localization (NLS) and nuclear export (NES) sequences, and a C-terminal activator domain (CTAD) containing aromatic, hydrophobic, and acidic amino acids, the socalled AHA motifs (Nover et al., 1996, 2001). These structural characteristics are important for stress-dependent activation that converts the inactive HSF monomers to trimers, which bind after NLS to conserved heat-shock promoter elements (HSE consensus: -CAA- -TTC-) followed by transcriptional activation of the respective target genes (Wu, 1995). Contrasting other organisms, plants exhibit a multiplicity of heat stress transcription factors. Six HSF cDNA clones have been isolated from soybean (CzarneckaVerner et al., 1995), and in Arabidopsis thaliana, 21 different HSF genes have been
Figure 12.1 Levels of heat stress regulation. Octagonal boxes represent HSF transcription factors (classes A and B), open bars promoter-binding elements of genes, and ovals HSP (heat-shock proteins, chaperons).
288
REGULATION OF TRANSCRIPTION IN PLANTS
identified by cloning (Sch¨offl and Pr¨andl, 1999) and data mining of genomic sequences (Nover et al., 2001). More than 17 HSFs have been identified in tomato expressed sequence tag (ESTs) (Scharf et al., 1990; Nover et al., 2001) and up to 30 representatives were identified in the rice genome. Based on sequence relationship, plant HSFs are assigned to three classes A, B, and C (Nover et al., 1996, 2001). The structure of class A and C HSFs differs from that of class B by an insertion of 21 or 7 amino acids, respectively, separating hydrophobic regions HR-A and -B. Another structural difference between class A and class B HSFs is the absence of AHA motifs in their CTAD, which have appeared to be crucial for activator function of class A HSF (D¨oring et al., 2000). This lack of AHA motifs of class A HSF correlates with an inability for transcriptional activation of heat-shock promoter reporter genes in tobacco protoplasts (Czarnecka-Verner et al., 1997, 2000), and a class B HSF failed to rescue the yeast Hsf1 mutation (Boscheinen et al., 1997). By contrast, overexpression of class A HSFs in transgenic plants (Lee et al., 1995; Pr¨andl et al., 1998; Li et al., 2003) resulted in the expression of heat-shock promoterdriven stress and reporter genes. These structural and functional differences led to the hypothesis that class A HSFs represent the true transcriptional activators and class B HSFs are inert or repressor HSFs (Czarnecka-Verner et al., 1997, 2000). A peculiarity exclusive to plants is the finding that the expression of several members of the HSF family requires a heat stress for their own expression, suggesting a multistep mechanism of HSF involvement in the heat-shock response (Nover et al., 1996). The relatively high numbers of class A HSFs (there are 15 members in Arabidopsis) suggests that functional diversification and/or genetic redundancy have evolved in plants.
12.1.4
Class A HSF: immediate early regulators of the heat-shock response
Arabidopsis class A HSFs have the capacity to act as positive regulators of heat-shock gene expression, originally implicated by transgenic overexpression of AtHsfA1a constructs (previously designated Hsf1, in Wunderlich et al., 2003; Lohmann et al., 2004) or AtHsfA1b (previously designated Hsf3, in Pr¨andl et al., 1998; Panchuck et al., 2002; Lohmann et al., 2004; Panikulangara et al., 2004) constructs in Arabidopsis, which resulted in a constitutive expression of HSPs and enhanced levels of basal thermotolerance (Lee et al., 1995; Pr¨andl et al., 1998). Heat-stress-dependent binding of AtHsfA1a to heat-shock gene promoters has been confirmed by in vivo UV-laser crosslinking in Arabidopsis cells (Zhang et al., 2003). At present there is no evidence that in Arabidopsis a single HSF dominates the regulation of heat-shock response. Considering the high complexity of HSF genes in plants the isolation of HSF knockout mutants becomes crucial for determining their functional roles and biological importance in plants. T-DNA knockout mutations of class A HSF genes, AtHsfA1a or AtHsfA1b, showed only little effect on DNA binding, induction of heat-shock gene expression, and thermotolerance. Double knockout mutants (hsfA1a/1b) displayed an almost complete, selective loss of the early heat-shock-dependent DNA-binding capacity in plant cell extracts (Lohmann et al., 2004). The plants exhibited no detectable physiological or morphological
TEMPERATURE-REGULATED GENE EXPRESSION
289
phenotype and thermotolerance was only marginally impaired. However, the double knockout mutants exert clear negative effects in the immediate induction of HSF target gene expression. Thus these two class A HSFs (previously identified as positive transcriptional regulators of heat-shock gene expression, Lee et al., 1995; Pr¨andl et al., 1998) are required for only immediate early (within the first 2 h of heat stress) regulation of the heat-shock response. In this early phase, differences in heat-induced mRNA levels, evident for different heat-shock genes, further suggest that not all heat-shock genes have the same requirements of HSF and/or synergistic effects between them. The transient accumulation of mRNA indicates that during sustained stress the transcription of heat-shock genes is negatively regulated in wild type (WT), probably via a negative regulation of HSF activity. After extended heat stress, WT and mutant plants showed comparable expression levels (mRNA and proteins) of all target genes tested (Lohmann et al., 2004). In WT, the transient accumulation of mRNA indicates that during sustained stress the transcription of heat-shock genes is negatively regulated, probably via a negative regulation of HSF activity. At present it is not known whether in AthsfA1a/1b double mutants this negative regulation is still implemented. Putative negative regulators of HSF activity are HSP70 and HSP90, which have been found to be associated in a complex with HSF in its inactive form in vertebrate cells (Mosser et al., 1993; Baler et al., 1996; Satyal et al., 1998; Zou et al., 1998). There is experimental evidence for an interaction between HSP70 and AtHSFA1a of Arabidopsis, which implicates a feedback regulation of the heat-shock response in plants (Kim and Sch¨offl, 2002). An alternative model for the shutoff of the heat-shock response suggests a replacement or inactivation of activator HSF by ‘repressor’ HSF, which become activated during later stages. Members of class B HSFs, which lack certain structural features of the class A activator HSFs (Nover et al., 2001), are implicated as transcriptional repressors or attenuators of the heat-shock response (Czarnecka-Verner et al., 1997, 2000). These class B transcription factors are known to inhibit promoter activity by repression mechanism that involves the C-terminal regulatory region (CTR). Deletion analysis revealed that in the CTR of soybean, GmHSFB1, which is potentially involved in protein–protein interaction with the general transcription factors, there were two sites, namely, the repressor domain (RD, located in the N-terminal half of the CTR) and the TFIIB-binding domain (located C-terminally from the RD) that binds to the TFIIB (Czarnecka-Verner et al., 2004).
12.1.5
Second-level regulation
The heat-dependent expression of certain class A and class B HSFs indicates that the immediate early phase (induction and expression of primary target genes), which is attributed to the action of constitutively expressed class A HSFs, also triggers a second wave of regulatory events. The expression of class B HSFs (AtHsfB1 and AtHsfB2b), which is controlled by class A HSFs, suggests that their function is required for regulating delayed effects of the heat-shock response. Certain other class B HSFs may serve as coactivators of gene expression during prolonged heat stress or during recovery (Bharti et al., 2004). The fact that also certain class A
290
REGULATION OF TRANSCRIPTION IN PLANTS
HSFs are expressed at a higher level after heat stress may indicate that the functions of putative ‘activator’ HSFs are required for the second-level regulation. Further analysis of Arabidopsis HSF knockout mutants will shed more light on the functional roles of class B HSFs in plants. In conclusion, loss-of-function mutant analysis provided evidence that two transcription factors, AtHsfA1a and AtHsfA1b, are fast response regulators, which are important for the coordination of stress gene expression and the generation of stress tolerance under rapidly changing environmental condition in nature. HSF functions seem to be required not only for the fast induction of stress genes (e.g., HSP, HSF) but consequently also for the timing of a negative feedback regulation causing transient expression in WT. Plants may have, in contrast to vertebrate cells, which clearly show defects in the heat-shock response when either Hsf1 or Hsf3 genes have been disrupted (McMillan et al., 1998; Tanabe et al., 1998; Xiao et al., 1999), a stronger backup and a more complex control of the stress-response system.
12.1.6
Heat stress target genes
HSP genes are known as the conventional targets of HSFs. Activated HSF binds to the conserved HSE sequences in their promoter regions, which results in the transcription of genes. Besides HSPs a few unconventional heat-shock genes like ascorbate peroxidase 2 (Apx2) and galactinol synthase1 (GolS1) (Panchuck et al., 2002; Panikulangara et al., 2004) and certain HSF genes (Lohmann et al., 2004) have been identified as HSF targets. Primary target genes were identified as genes whose short-term heat stress induction (1 h, 37◦ C) was significantly impaired in AthsfA1a/1b knockout mutants of Arabidopsis (Lohmann et al., 2004). The transcriptome analysis of heat-stressed leaves of Arabidopsis WT and AthsfA1a/1b double knockout plants revealed a differential expression of a large number of genes (2567 and 3056 respectively) after short-term heat stress; however, only few of them (112) fulfilled the criteria for being direct targets of AtHsfA1a/1b (Busch et al., 2005). Among them were several of the known HSP genes and a number of novel genes with functions in different pathways, as, for example, in protein biosynthesis/degradation, membrane transport, oxidative stress response, and signaling (Busch et al., 2005). The exact biological functions of most of these genes are not known; however, some have been connected with environmental stress responses. For example, WRKY7 is involved in disease resistance (Maleck et al., 2000); scarecrow factors are involved in developmental processes but were also shown to be affected in expression by salt stress in Arabidopsis and osmotic stress in white spruce (Ueda et al., 2002; Stasolla et al., 2003); Dof zinc-finger proteins are associated with stress response and control of seed germination (Chen et al., 1996; Kang and Singh, 2000; Papi et al., 2000). Other HSF-upregulated targets are IAA2, a transcription factor, which is involved in auxin signaling (Reed, 2001), and AtERF4, an active repressor of transcription, which is induced by wounding, cold, high salinity, and drought stress (Fujimoto et al., 2000).
TEMPERATURE-REGULATED GENE EXPRESSION
291
The highest ranking unconventional heat-shock gene was GolS1; its mRNA is more than 100-fold induced by heat shock in WT, and it represents the highest score as an unconventional AtHsfA1a/1b-dependent heat-shock gene (Panikulangara et al., 2004; Busch et al., 2005). Within the family of sHSP, which comprises 18 members, the induction levels of 6 heat-shock-inducible sHSP genes were clearly dependent on AtHsfA1a/1b. Most striking was the AtHsfA1a/1b-dependent expression of other HSFs, in particular AtHsfB1, -B2a, and -A7a, which represent three out of six heat-inducible HSF genes. The identification of AtHsfA7a (class A) as a HSF-dependent gene indicates that class B factors, which lack a AHA motif (transcriptional activation domain), are not the only ones involved in controlling expression of secondary HSF target genes for delayed functions in the heat-shock response. Microarray expression analysis of long-term heat-shock treatment (6 h, 37◦ C) of Arabidopsis WT plants resulted in much lower numbers (approximately tenfold) of differentially expressed genes (Rizhsky et al., 2004) compared to short-term (1 h, 37◦ C) heat stress (Busch et al., 2005). This discrepancy between expressionprofiling data may be attributed to differences in the duration of heat stress. After short-term heat stress the profiling registers the fast, transient changes in the expression of many genes, whereas long-term heat stress may lead to an adjustment of the steady-state levels of mRNAs to control levels for a large number of transiently up- or downregulated genes. Thus, short-term heat stress may cover with a higher probability the primary HSF target genes, whereas long-term heat stress may have the potential to represent secondary target genes. Proteome and transcriptome analysis of knockout mutations in HSF-dependent HSF genes as, for example, AtHsfB1, -B2a, and -A7awill be required to clearly discriminate between primaryand secondary-level HSF target genes and functions.
12.1.7
Common stress-response genes and pathways
Several genes encoding enzymes involved in galactinol and RFO (raffinose family of oligosaccharides) synthesis were identified as heat stress and HSF dependent (Busch et al., 2005). The expression of two enzymes, myoinositol-1-phosphatesynthase and galactinol synthase, belonging to this biosynthetic pathway, are also induced by other stresses. Myoinositol-1-phosphate-synthase is also induced by cold, drought, and salt stress in Arabidopsis (Kreps et al., 2002) and was found to be related to salt stress response and osmoprotection in the halophyte Mesembryanthemum crystallinum (Ishitani et al., 1996). Members of the galactinol synthase (GolS) gene family were also upregulated in response to different environmental stresses (Taji et al., 2002). Enhanced levels of galactinol and raffinose, generated in leaves of transgenic plants by overexpression of GolS2, correlated with improved drought tolerance of Arabidopsis (Taji et al., 2002). The HSF-dependent expression of GolS1 correlates with an increase in the level of RFO, and knockout mutations of GolS1 are unable to accumulate stress-induced galactinol and raffinose levels in leaves (Panikulangara et al., 2004). It has been shown that GolS1 and GolS2 genes are induced by drought and high salinity stress but the expression does not
292
REGULATION OF TRANSCRIPTION IN PLANTS
seem to be regulated by the drought transcription factor DREB (Taji et al., 2002). It is conceivable that HSFs may integrate different signaling pathways that lead to the expression of GolS1. The expression/function of these genes overlaps between different environmental responses that require protection by osmoprotective solutes. A new viewpoint regarding metabolic mechanisms of plant adaptation to thermal stress comes from a recent metabolite-profiling analysis by Kaplan et al. (2004). Using gas chromatography–mass spectrometry, a number of known and unknown metabolites were identified to be influenced by temperature stress; apart from a few stress-specific metabolites identified, interestingly, a majority of the heat stress metabolites overlapped with the cold stress metabolite profiles, thus uncovering common responses between high- and low- temperature stresses.
12.2 12.2.1
Cold stress Introduction
Low-temperature stress is one of the major stresses plants are exposed to in the environment. Different plant species have different abilities to tolerate cold stress. Many plants grown in tropical regions, like tomato, maize, and rice, are unable to tolerate temperatures below 10◦ C. Many other herbaceous plants from moderate climatic regions, depending on the species, are able to survive freezing temperatures as low as –30◦ C. However, plants can become acclimated to withstand low or freezing temperatures by exposing them for a short period to low nonlethal temperatures. This phenomenon to survive a normally lethal temperature is known as cold acclimation or acquired chilling or freezing tolerance (Guy, 1990; Thomashow, 1998). For instance, wheat plants grown at normal temperature are killed at a temperature around –5◦ C but can survive freezing temperature as low as –20◦ C after being cold acclimated; nonacclimated rye is killed at about –5◦ C, whereas cold-acclimated plants can withstand temperatures dropping down to −30◦ C; Arabidopsis plants grown at normal temperature are killed at –3◦ C, but after acclimation at 4◦ C (day) and 2◦ C (night) cycle for 5 days, plants can survive –7◦ C (Gilmour et al., 1988; L˚ang et al., 1994). Upon exposure to low-temperature stress, plants immediately pursue genetic programs to protect them from damage. The cascade of events starts with perception of stress, transduction of signals leading ultimately to the induction/expression of genes/proteins involved in protection of cells. In addition, biochemical changes occur, which may be involved in the perception of stress and initiation of signaling events.
12.2.2
Low-temperature sensing and signaling
The mechanisms by which low-temperature signals are perceived and transduced into biochemical responses are not completely understood. The model sketched in Figure 12.2 integrates the current knowledge about sensing and signaling of the
Figure 12.2 Regulatory network of cold stress response in plants. Octagonal boxes represent the transcription factors, and open bars promoter-binding elements of genes. ABA, abscisic acid; bZIP, a family of transcription factors with basic region and leucine zipper motif; CBF, CRT-binding factor; CDPK, calcium-dependent protein kinase; COR, cold-regulated genes; CRT, cold-response elements; DRE, dehydration-responsive elements; DREB, DRE-binding factors; ESK1, Eskimo 1 protein; HOS1, high expression of osmotically responsive genes 1; ICE, inducer of CBF expression; ICEr1 and r2, inducer of CBF expression region 1 and 2; KIN, kalt (cold) induced gene; MEKK1, MAPK kinase kinase 1 or ERK kinase kinase 1; MKK2, MAPK kinase 2; MPK4/6, MAP kinase 4/6; MYB, transcription factors with tryptophan cluster motif; MYC, transcription factors with basic-helix-loop-helix (bHLH) and leucine zipper motif; RAB18, response to abscisic acid 18; RAP2.1/2.6, related to AP2 transcription factor; RAV1, related to ABI3/VP1 transcription factor; SFR6, sensitive to freezing 6; ZAT12, zinc-finger protein 12; ?, unknown factors/regulators.
294
REGULATION OF TRANSCRIPTION IN PLANTS
cold stress response. Current models suggest that low temperature may be perceived directly via membrane-bound receptors, which sense biochemical changes, or indirectly by biochemical changes that result from dehydration, which subsequently activate an osmosensor. In both cases sensing and signaling require biochemical changes and the regulation of receptor protein activity. To date, no membranebound cold stress receptors have been identified in plants. An early response, when plants are exposed to low temperature, is a decrease in membrane fluidity and coldinduced calcium influx, followed by secondary responses in which calcium and cold-regulated (COR) protein kinases and phosphatases are believed to be involved (Monroy et al., 1998). In alfalfa, Ovar et al. (2000) showed that subsets of coldresponsive genes can be induced by chemicals that alter the membrane fluidity, and that cytoskeletal reorganization occurs, which may be linked to membrane rigidification and calcium influx in response to cold. The calcium concentration in the cell increases due to an influx of Ca2+ through the plasma membrane and by Ca2+ release from the vacuole (Knight et al., 1996). The magnitude of Ca2+ influx correlated with the rate of temperature change in Arabidopsis (Plieth et al., 1999). The phytohormone abscisic acid (ABA) is involved in a second mechanism in cold sensing/signaling. In Arabidopsis the levels of ABA increase transiently during the first day of cold treatment and thereafter decline gradually (L˚ang et al., 1994). The role of ABA in cold signaling is confirmed by the fact that many cold-responsive genes are induced by external application of ABA (Thomashow, 1994). Another highly conserved stress-signaling mechanism among eukaryotes works via the activation of the mitogen-activated protein (MAP) kinase cascade. The cascade consists of three subsequently acting protein kinases, namely, MAP kinase kinase kinase (MAPKKK), MAP kinase kinase (MAPKK), and finally MAP kinase (MAPK). There have been indications of MAPK cascade being involved in mediating cold stress tolerance in plants, but upstream components of the cascade have not been identified. A cold-activated MAPK was demonstrated by Jonak et al. (1996) in alfalfa. In the Arabidopsis genome there are more than 60 putative MAPKKK, 10 MAPKK, and 20 MAPK. Out of the 20 MAPK, AtMPK3, AtMPK4, and AtMPK6 were induced by cold and a number of other stresses (Mizoguchi et al., 1996; Ichimura et al., 2000), and AtMKK2 was induced by cold and salt stress (Teige et al., 2004). Using yeast two-hybrid analysis and in vivo protein kinase assay it was shown that MKK2 directly targets MPK4 and MPK6 MAPK, respectively. MKK2-overexpressing transgenic Arabidopsis plants exhibited increased freezing and salt tolerance and showed enhanced expression of cold and salt stress marker genes including the transcription factors RAV1, CRT/DRE binding factor 2 (CBF2), CBF3, FAD8 (chloroplast-localized fatty-acid desaturase), and P5CS ( pyrroline5-carboxylate synthase). The roles of calcium as a messenger in response to cold and freezing acclimation are linked to the activity of calcium-dependent Ser/Thr protein kinases (CDPK), which are believed to participate in signal transduction. An increase in CDPK transcript levels in response to low temperature or other abiotic stresses has been reported for different plant species. The transcript levels of two CDPKs are differentially regulated by low temperature in alfalfa (Monroy and Dhindsa, 1995), and in rice CDPK
TEMPERATURE-REGULATED GENE EXPRESSION
295
RNA levels increase in response to cold and salt stress (Saijo et al., 1998). Overexpression of OsCDPK7 from rice confers both cold- and salt tolerance in rice (Saijo et al., 2000). In rice, a 12–18-h cold treatment activates a 56-kD membrane-bound CDPK, indicating that this enzyme is involved in late but not in the early events of the cold-adaptive response (Mart´ın and Busconi, 2001). Downstream components of this signaling path have still to be identified. In summary, the COR gene expression (the key to cold acclimation or freezing tolerance) is mediated through multiple pathways. The signaling pathway follows into two major routes: (1) calcium and CBF/DREB transcription factor dependent and (2) ABA mediated and bZIP, MYC, and MYB transcription factor dependent. Less well investigated are pathways involving other transcription factors like (3) RAV1 and ZAT12, and (4) ESK1. 1. In the ABA-independent signaling pathway, following perception of cold stress signal, a MYC-like basic helix-loop-helix transcription activator ICE (inducer of CBF expression), which is present in an inactive state at normal temperature, becomes activated for binding to the ICE box (CANNTG) in the promoter region of CBF3/DREB1a transcription factor gene (Gilmour et al., 1998; Chinnusamy et al., 2003). Following expression of CBF3/DREB1 the factor binds to the cis-element CRT/DRE (CCGAC) of the cold-responsive target genes, e.g., COR, KIN, and RAB18, which stimulates their transcription and expression. Further insight into the CBF-mediated induction of the cold-responsive genes was obtained from transcriptome analysis using CBF-overexpressing transgenic Arabidopsis plants (Fowler and Thomashow, 2002). In these plants the transcripts of RAP2.1 and RAP2.6 transcription factors are also upregulated, suggesting that they are also targets of CBF. These factors seem to act upstream of other cold-responsive genes as indicated by an involvement in COR gene expression. Only RAP2.1 contains the C-repeat/dehydration-responsive element (CRT/DRE) sequences in the promoter region, but the exact mechanism of expression and role in transcription are yet unknown. 2. The ABA-affected regulation of cold-responsive genes can be mediated through the activation of either CBF4/DREB1D or bZIP, MYC, and MYB transcription factors. The latter group of factors do not bind to the CRT/DRE sequences but recognize other sequences like the ABA response elements (ABRE) or the MYC/MYB-binding sites in the promoter region of the target genes. 3. An involvement of other pathways in cold stress response was indicated by results from transcriptome analysis (Fowler and Thomashow, 2002). Transcripts of cold-induced genes like the CBF2, RAV1, and ZAT12 appear at the same time (after 1 h of cold stress). Thus, they could be acting in parallel. ZAT12 and CBF2 seem to control partially overlapping sets of target genes. Overexpression of ZAT12 led to a downregulation of the CBF genes, indicating its role as a negative effector of CBF in the cold-response pathway (Vogel et al., 2005).
296
REGULATION OF TRANSCRIPTION IN PLANTS
4. Another CBF-independent pathway was identified in the esk1 (ESKIMO1) mutants in Arabidopsis. These mutants are constitutively freezing tolerant but the expression of genes with CRE/DRT sequences is not altered. This suggests that at least partial freezing tolerance can be achieved in a CBFindependent manner through ESK1 (Xin and Browse, 1998). Genetic screens have led to the identification of other regulators of cold signal transduction. The hos1 (high expression of osmotically responsive genes) mutant was isolated after EMS mutagenesis of transgenic Arabidopsis plants carrying a COR promoter of RD29A fused to luciferase reporter gene (Ishitani et al., 1998). The hos1 mutant shows enhanced cold induction of CBF and cold-responsive genes. The gene encodes a RING-finger protein, and it was proposed that HOS1 acts as a negative regulator by targeting the positive regulators of CBF for ubiquitination and degradation (Lee et al., 2001). Other regulators identified via mutant screening are LOS1 (low expression of osmotically responsive gene 1) and LOS2 (low expression of osmotically responsive gene 2), encoding translation elongation factor 2 and a bifunctional enolase, respectively (Ishitani et al., 1997). Guo et al. (2002) found that in the los1 mutant CBF transcripts are highly induced in response to cold but the expression of the target genes bearing CRT/DRE sequences was blocked. Similarly the los2 mutants are also affected in the downstream target genes bearing the CRT/DRE sequences, suggesting that the LOS2 protein may function in the expression of delayed coldresponsive genes by transcriptional repression of ZAT10 (Lee et al., 2002). The Arabidopsis mutant sfr6 (sensitive to freezing), identified in a screen by Warren et al. (1996) and characterized by Knight et al. (1999), shows a lower expression of the COR genes containing CRT/DRE promoter elements. However, SFR6 was not involved in the regulation of CBF expression. At present it is not possible to draw a complete picture of the regulatory network in cold stress signaling. Further genetic, proteome, transcriptome, metabolome, and protein interaction analyses are required for gaining deeper insights into this complex stress-response system.
12.2.3
Transcription factors in ABA-independent pathways
Transcription factors involved in ABA-independent cold regulation were identified by the binding of proteins to the CRT/DRE sequences present in the promoters of COR genes (Stockinger et al., 1997). This led to the identification of CBF1 (CRT/DRE binding factor 1) in Arabidopsis, which is present on chromosome 4 along with two other CBF genes organized in direct repetition (Gilmour et al., 1998). The transcript levels of all three CBF genes increase within 15 min in response to low temperature; thereafter, within 2 h, the transcripts of COR target genes accumulate (Gilmour et al., 1998). Transgenic overexpression of CBF1 in Arabidopsis led to the induction of multiple COR genes at nonacclimating temperatures and enhanced freezing tolerance of these plants (Jaglo-Ottosen et al., 1998), suggesting that CBF1 is the key regulator of COR gene expression leading to freezing tolerance.
TEMPERATURE-REGULATED GENE EXPRESSION
297
CBF1–3, synonymously designated DREB1b, DREB1c, and DREB1a, respectively (Liu et al., 1998), possess at the N-terminus a putative NLS signal, an AP2 DBD, and a C-terminal acidic region, which specifies an activation domain (Stockinger et al., 1997; Gilmour et al., 1998). It was shown that the N-terminal 115 amino acids of CBF are necessary to direct the protein to the target genes and the C-terminal 98 amino acids are necessary for trans-activation of transcription (Wang et al., 2005). Within the activation domain four amino acid motifs (clusters of hydrophobic residues delimited from one another by small stretches of Asp, Glu, Pro, and other residues) were identified, which play a role in transcriptional activation. CBF orthologs have been isolated and studied in other plant species. Lycopersicum esculentum harbors three CBF genes (LeCBF1–3) arranged in tandem array, but only LeCBF1 is low temperature induced (Zhang et al., 2004). Constitutive overexpression of LeCBF1 in transgenic Arabidopsis led to the accumulation of the COR genes and enhanced freezing tolerance. In a closely related species, Brassica napus, four orthologs of Arabidopsis CBF/DREB transcription factors BNCBF-5, -7, -16, and -17 have been identified (Jaglo et al., 2001; Gao et al., 2002). The structures of BNCBF-5, -7, and -16 are very similar to Arabidopsis CBF1 but BNCBF17 contains two extra regions of 16 and 21 amino acids in the acidic activation domain (Gao et al., 2002). The transcripts of all genes were cold induced, but the mRNA of BNCBF17 was short lived. Library screening of two distant species, rye (Secale cereale) and wheat (Triticum aestivum), led to the identification of cDNA clones, sharing 30 and 34% sequence similarity within the AP2 DBD of Arabidopsis CBF (Jaglo et al., 2001). A striking feature observed by the alignment of Arabidopsis, B. napus, S. cereale, T. aestivum, and L. esculentum sequences was the presence of short signature sequences PKK/RPAGRxKFxETRHP located immediately upstream of the AP2 DBD and the sequence motif DSAWR just downstream of the AP2 domain (Jaglo et al., 2001). These structural characteristics distinguish CBF from the other 140 or more AP2/EREBP proteins in Arabidopsis. Since the expression of CBF is induced by low temperature, Gilmour et al. (1998) proposed a cold-activated regulator acting upstream of CBF. Chinnusamy et al. (2003) carried out a genetic screen for regulators of CBF/DREB expression. Transgenic Arabidopsis plants bearing CBF3 promoter::luciferase (CBF3-LUC) constructs were chemically mutagenized and mutants with altered CBF3-LUC expression in response to cold were isolated. The mutant ice1 was impaired in CBF3LUC activity and cold acclimation. ICE1 encodes a MYC-like bHLH transcription factor binding to the MYC recognition elements in the promoter of CBF3 (Chinnusamy et al., 2003). Until recently the possibility that CBFs regulate their own expression was ruled out by two criteria: (i) the promoter regions of CBF1–3 lack the CRT/DRE-binding sequences and (ii) CBF1-overexpressing Arabidopsis has no influence on the expression of the other CBFs (Jaglo-Ottosen et al., 1998). However, Novillo et al. (2004) isolated an Arabidopsis T-DNA insertion in the CBF2/DREB1c gene. The cbf2 mutant showed enhanced transcript levels of the two other cold-induced CBFs at control temperatures and improved tolerance to freezing, dehydration, and salt stress. Hence, CBF2 may be a negative regulator of the other CBF/DREB (CBF1/DREB1b and CBF3/DREB1a) genes.
298
REGULATION OF TRANSCRIPTION IN PLANTS
In nature, multiple environmental factors are involved in governing a stress response. The work of Harmer et al. (2000) and Fowler et al. (2005) showed that both low temperature and circadian rhythm affect induction of CBF/DREB transcription factors. Other transcription factors like RAV1 and ZAT12, induced in parallel, were also subject to circadian regulation (Fowler et al., 2005). In addition to cold, CBF1–3 expression was also induced by mechanical agitation (Gilmour et al., 1998), ABA (Knight et al., 2004), and inhibition of protein synthesis (Zarka et al., 2003).
12.2.4
Transcription factors in ABA-dependent pathways
The ABA-dependent expression of cold-responsive genes is mediated by transcription factors of the bZIP, MYB, and MYC families, which recognize specific elements in the promoter of target genes. These transcription factors are known to respond to stresses like cold, dehydration, and exogenous application of ABA in plants like rice, maize, and Arabidopsis (Kusano et al., 1995; Lu et al., 1996; Nakagawa et al., 1996; Abe et al., 2003). Another ABA-independent induction of the COR genes is regulated by RAV1 and ZAT12 transcription factors. The transcription factor RAV1 (related to ABI3/VP1) has two distinct DBDs, the N-terminus is homologous to the AP2 DBD present in APETALA2 transcription factor and the C-terminus designated as B3 is homologous to that of ABI3/VP1 transcription factor. The two DBDs were shown to bind two unrelated sequences, CAACA and CACCTG, to achieve higher affinity and specificity to binding (Kagaya et al., 1999). Six proteins of this family have been identified in Arabidopsis (Riechmann et al., 2000). ZAT12 is a zinc-finger protein that seems to be involved in the regulation of CBF genes. The involvement of RAV1 and ZAT12 in cold stress was suggested by transcriptome analysis (Fowler and Thomashow, 2002). Data of Rizhsky et al. (2004) implicated a role of ZAT12 in oxidative stress response. Recent transcriptome analysis (Vogel et al., 2005) showed that CBF2 and ZAT12 share a number of the highly induced putative target genes. Constitutive overexpression of ZAT12 in Arabidopsis caused a small increase in freezing tolerance, indicating its role in cold acclimation. Transcription factors of the RAP2 (related to AP2) family were identified as putative regulators, by transcriptome analysis of CBF-overexpressing Arabidopsis plants (Fowler and Thomashow, 2002). Members of this family (12 genes in Arabidopsis) are differently expressed in different tissues and AP2 seems to contribute to their expression throughout development (Okamuro et al., 1997). The exact role/function of two of these transcription factors (RAP2.1 and RAP2.6) in cold stress is not known.
12.2.5
Cold stress target genes
Genes expressed in response to cold are synonymously designated as cold-regulated genes (COR), or low temperature induced (LTI), or cold inducible (KIN), or early dehydration responsible (ERD) in different publications. These genes contain one
TEMPERATURE-REGULATED GENE EXPRESSION
299
or more binding sequences for transcription factors in the promoter regions. In the ABA-independent pathway, the expression of target genes is mediated by one or more CBF/DREB-binding sites (CRT/DRE) with the core sequence A/GCCGAC (Yamaguchi-Shinozaki and Shinozaki, 1994; Stockinger et al., 1997; Sakuma et al., 2002). In the ABA-dependent pathway, bZIP transcription factors bind to the ABRE consensus sequences ACGTGGC (Busk et al., 1997; Hobo et al., 1999; Zhu, 2002) and to the less conserved coupling elements (CE) with the core ACGT. Other ABA dependently expressed genes are targets of the MYB and MYC transcription factors. The recognition sites in the promoters of target genes are TGGTTAG and CACATG respectively (Abe et al., 1997, 2003). The knowledge about how cold-response target genes contribute to the attainment of cold/freezing tolerance is scant. From the group of CBF/DREB-controlled COR genes like KIN1, COR6.6/KIN2, COR15a, COR47/RD17, COR78/RD29a, and ERD10,−6,−9, the mode of action has been studied only for COR15a. Evidence came from the work of Artus et al. (1996), who showed that the constitutive expression of COR15a (COR15am) in Arabidopsis enhanced freezing tolerance of chloroplast (in vivo) and protoplasts (in vitro) in nonacclimated plants. As a result of freeze-induced dehydration the plasma membrane comes in close apposition with the chloroplast envelope; COR15am protein decreases the transition of lamellar to hexagonal II phase lipids (caused usually as a result of cold-induced dehydration) in these regions and alters the intrinsic curvature of the inner membrane of the chloroplast envelope (Steponkus et al., 1998). Increase in cold/freezing tolerance is not caused only by the action of COR proteins but also due to an increase in metabolite/sugar levels. CBF3-overexpressing plants, which are freezing tolerant, show in addition to the elevated levels of COR gene expression also increased amounts of proline and total sugars (Gilmour et al., 2000). The eskimo1 mutant (Xin and Browse, 1998) of Arabidopsis, which is also freezing tolerant, accumulates enormous amounts of proline. Proline is well known as a stress-dependent substance involved in freezing and salt tolerance, as for example demonstrated by antisense plants for proline dehydrogenase (AtProDH) (Nanjo et al., 1999). It is not shown but thought that the role of sugars, in part, might be to stabilize the membranes. Maruyama et al. (2004) grouped genes on the basis of expression level with respect to the duration of stress applied. Genes belonging to class I show increased expression within 5 h and reach a maximum level at 10 or 24 h (these include a few known DREB1A target genes like rd29A, cor15a, cor15b, kin1, kin2, erd10, rd17, AtGolS3, RAP2.1, erd7, and genes of unknown function). Class II genes are induced within 2 h and reach their highest expression at 5 h (DREB1A transcription factor); class III includes genes that are induced gradually up to 24 h (AtGRP7, AtPlc1, and genes of unknown function). Recently, microarray expression profiling became a useful tool in the analysis of cold stress response (a) to obtain a broader picture on the expression patterns of genes in response to different abiotic or biotic stresses; (b) to identify downstream targets of stress-responsive transcription factors; and (c) to study the regulatory network that crosstalks during stress. So far the method has been exploited to obtain new
300
REGULATION OF TRANSCRIPTION IN PLANTS
cold-responsive genes and other potential targets of CBF/DREB1 transcription factors. A number of microarrays using cDNA prepared from Arabidopsis grown under stress conditions like cold, drought, and salt, as well as from normal temperature grown CBF/DREB-overexpressing transgenic Arabidopsis plants (Seki et al., 2001; Fowler and Thomashow, 2002; Seki et al., 2002; Maruyama et al., 2004), have contributed to the identification and categorization of COR genes that include (1) transcription factors, (2) late-embryogenesis-abundant (LEA) proteins, (3) PI metabolism, (4) KIN proteins, (5) RNA-binding proteins, (6) transport proteins, (7) hydrophilic proteins, (8) desaturases, (9) metabolic enzymes, (10) senescencerelated proteins, and many proteins with unknown function supporting earlier findings by Guy (1990), Mohapatra et al. (1989), and Thomashow (1999). Confirmation of the existence of regulatory pathways distinct from CBF/DREB-dependent expression of genes is implicated by the observation that 28% of the cold-responsive genes and 15 transcription factors are obviously not regulated by the CBF/DREB.
12.3
Perspectives for plant biotechnology
In the natural environment, extreme temperatures, drought, salinity, and oxidative stress are often interconnected and cause common detrimental effects on plant growth and yield. As a consequence, diverse environmental stresses often activate similar cell signaling pathways and cellular responses. It should be noted that abiotic stress is the major cause for reducing the efficient exploitation of the genetic potential for yield in crop plants worldwide (Boyer, 1982; Bray et al., 2000). Thus, the improvement of stress tolerance is an important goal in plant biotechnology-assisted breeding. The overlap in the responses between cold and drought and, to a lesser extent, between heat and drought stress responses implicates that the genetic manipulations may have a great potential to improve common stress tolerance. At present the most suitable targets for genetic engineering of stress tolerance are transcription factors because they are responsible for coordinating the expression of whole sets of target genes. Transgenic overexpression of transcription factors involved in the expression of heat- or cold-regulated genes has been successfully used for generating plants having improved basal tolerance to heat or drought and cold respectively. For example, overexpression of CBF/DREB factors resulted in a constitutive expression of COR genes and in improved cold/drought tolerance in Arabidopsis (Jaglo-Ottoson et al., 1998; Kasuga et al., 1999). However, a strong, S35-promoter-driven overexpression of DREB was associated with a dwarf phenotype of transgenic plants. This finding suggests that the native response system has two functions: (i) upregulation of genes, which are required for cellular protection from harmful effects, and (i) downregulation of genes, which are involved in growth and development. In CBF/DREBtransgenic Arabidopsis plants the growth effect problem could be partially solved by using inducible promoter constructs that limit the level of CBF/DREB overexpression or by treating plants with gibberellic acid.
TEMPERATURE-REGULATED GENE EXPRESSION
301
On the other hand transgenic overexpression of AtHSF1 and AtHSF3, respectively, in Arabidopsis plants resulted in constitutive expression of stress genes and enhanced basal thermotolerance (Lee et al., 1995; Pr¨andl et al., 1998). Interestingly, transgenic tomato plants overexpressing AtHSF3 (AtHSF1b) also showed a constitutive expression of different HSPs and an improved chilling tolerance (Li et al., 2003). However, HSF overexpression appears to become unstable with time in subsequent generations, possibly by homeostatic adjustments, gene-silencing effects, and/or counterselection under nonstress conditions (unpublished observations). The gain-of-function effects of transgenic overexpression of HSFs were very surprising, since several of the native transcription factors, for example HSF1 and HSF3, are constitutive proteins in WT where they require stress signals to become activated for target gene binding and/or initiation of transcription. There is no general answer to explain this phenomenon but it seems that overexpression of transcription factors shifts the equilibrium between the inactive and active forms of these transcription factors, possibly by a titration of their negative feedback regulators. There is evidence that HSP70 is involved in the regulation of HSF activity (Lee et al., 1986; Kim et al., 2002). Thus, the real target of genetic engineering may be the metastable HSF complexes whose activation depends on the balance between HSF and its negative regulators. Such complexes may act as cellular sensors for such environmental stresses, which cause a depletion of free chaperons by increasing the load of denatured proteins. The heat- and cold stress response systems and molecular mechanisms appear to be conserved in plants, and it seems possible to use the regulatory genes/proteins identified in one species for successful genetic engineering in another species. Transgenic expression of Arabidopsis CBF/DREB in tobacco or B. napus plants resulted in a constitutive expression of putative target genes and cold/drought stress gene expression and caused enhanced stress tolerance (Jaglo et al., 2001; Kasuga et al., 2004). Overexpression of the tomato CBF1 homolog had the same effect upon heterologous expression in Arabidopsis, but interestingly, stress tolerance was not improved when the tomato CBF was overexpressed in tomato plants (Zhang et al., 2004). This shows that the mechanisms of stress-response regulation are conserved in plants but not necessarily the sets of downstream target genes, which seem to differ between cold-acclimated species (Arabidopsis) and tomato, which cannot acclimate to cold. On the other hand the molecular mechanisms and functions in stress tolerance seem to be conserved between mono- and dicotyledonous plants. Rice DREB/CBF homologs function in a similar way in rice and transgenic Arabidopsis (Dobouzet et al., 2003; Oh et al., 2005). While many transgenic crops have been already commercialized, only a few stress-resistant transgenic crops have been moved from the laboratory to the field for evaluation under natural stress conditions (Dunwell, 2000). To date, most of the genetically modified stress-tolerant plants are nonagronomic plants, and it will take some time until the knowledge gained from model plants has been successfully converted into useful stress-tolerant crops.
302
REGULATION OF TRANSCRIPTION IN PLANTS
References H. Abe, T. Urao, T. Ito, M. Seki, K. Shinozaki and K. Yamaguchi-Shinozaki (2003) Arabidopsis AtMYC2 (bHLH) and AtMYB2 (MYB) function as transcriptional activators in abscisic acid signalling. The Plant Cell 15, 63–78. H. Abe, K. Yamaguchi-Shinozaki, T. Urao, T. Iwasaki, D. Hosokawa and K. Shinozaki (1997) Role of Arabidopsis MYC and MYB homologs in drought- and abscisic acid-regulated gene expression. The Plant Cell 9, 1859–1868. S.G. Ahn and D.J. Thiele (2003) Redox regulation of mammalian heat shock factor 1 is essential for Hsp gene activation and protection from stress. Genes and Development 17, 516–528. N.N. Artus, M. Uemura, P.L. Steponkus, S.J. Gilmour, C. Lin and M.F. Thomashow (1996) Constitutive expression of the cold regulated Arabidopsis thaliana COR15a genes affects both chloroplast and protoplast freezing tolerance. Proceedings of the National Academy of Sciences of the United States of America 93, 13404–13409. R. Baler, J. Zou and R. Voellmy (1996) Evidence for a role of Hsp70 in the regulation of the heat shock response in mammalian cells. Cell Stress and Chaperones 1, 33, 39. K. Bharti, von P. Koskull-D¨oring, S. Bharti, et al. (2004) Tomato heat stress transcription factor HsfB1 represents a novel type of general transcription coactivator with a histone-like motif interacting with plant CREB binding protein ortholog HAC1. The Plant Cell 16, 1521–1535. O. Boscheinen, R. Lyck, C. Queitsch, E. Treuter, V. Zimarino and K.-D. Scharf (1997) Heat stress transcription factors from tomato can functionally replace HSF1 in the yeast Saccharomyces cerevisiae. Molecular and General Genetics 255, 322–331. R.S. Boston, P.V. Viitanen and E. Vierling (1996) Molecular chaperons and protein folding in plants. Plant Molecular Biology 32, 191–222. J.S. Boyer (1982) Plant productivity and environment. Science 218, 443–448. E.A. Bray, J. Bailey-Serres and E. Weretilnyk (2000) Responses to abiotic stresses. In: Biochemistry and Molecular Biology of Plants (eds W. Gruissem, B. Buchannan and R. Jones), pp. 1158–1249. American Society of Plant Physiologists, Rockville, MD. W. Busch, M. Wunderlich and F. Sch¨offl (2005) Identification of novel heat shock factor dependent genes and biochemical pathways in Arabidopsis thaliana. The Plant Journal 41, 1–14. P.K. Busk, A.B. Jensen and M. Pag`es (1997) Regulatory elements in vivo in the promoter of abscisic acid responsive gene rab 17 from maize. The Plant Journal 11, 1285–1295. W. Chen, G. Chao and K. Singh (1996) The promoter of a H2 O2 -inducible, Arabidopsis glutathione S-transferase gene contains closely linked OBF- and OBP-binding sites. The Plant Journal 10, 955–966. V. Chinnusamy, M. Ohta, S. Kanrar, et al. (2003) ICE1: a regulator of cold-induced transcriptome and freezing tolerance in Arabidopsis. Genes and Development 17, 1043–1054. A. Clarke, R. Desikan, R.D. Hurst, J.T. Hancock and S.J. Neill (2000) NO way back: nitric oxide and programmed cell death in Arabidopsis thaliana suspension cultures. The Plant Journal 24, 667–677. E. Czarnecka-Verner, S. Pan, T. Salem and W.B. Gurley (2004) Plant class B HSFs inhibit transcription and exhibit affinity for TFIIB and TBP. Plant Molecular Biology 56, 57–75. E. Czarnecka-Verner, C.X. Yuan, P.C. Fox and W.B. Gurley (1995) Isolation and characterization of six heat shock transcription factor genes from soybean. Plant Molecular Biology 29, 37–51. E. Czarnecka-Verner, C.-X. Yuan, L. Nover, K.-D. Scharf, G. English and W.B. Gurley (1997) Plant heat shock transcription factors: positive and negative aspects of regulation. Acta Physiologiae Plantarum 19, 529–537. E.C.-X. Czarnecka-Verner, C.-X. Yuan, K.D. Scharf, G. English and W.B. Gurley (2000) Plants contain a novel multi-member class of heat shock factors without transcriptional activator potential. Plant Molecular Biology 43, 459–471. P. D¨oring, E. Treuter, C. Kistner, R. Lyck, A. Chen and L. Nover (2000) The role of AHA motifs in the activator function of tomato heat stress transcription factors HsfA1 and HsfA2. The Plant Cell 12, 265–278.
TEMPERATURE-REGULATED GENE EXPRESSION
303
J.G. Dubouzet, Y. Sakuma, Y. Ito, et al. (2003) OsDREB genes in rice, Oryza sativa L., encode transcription activators that function in drought-, high salt and cold-responsive gene expression. The Plant Journal 33, 751–763. J.M. Dunwell (2000) Transgenic approaches to crop improvement. Journal of Experimental Botany 51, 487–496. S. Fowler and M.F. Thomashow (2002) Arabidopsis transcriptome profiling indicates that multiple pathways are activated during cold acclimation in addition to the CBF cold response pathway. The Plant Cell 14, 1675–1690. S.G. Fowler, D. Cook and M.F. Thomashow (2005) Low temperature induction of Arabidopsis CBF1, 2, and 3 is gated by circadian clock. Plant Physiology 137, 961–968. C.H. Foyer, H. Lopez-Delgado, J.F. Dat and I.M. Scott (1997) Hydrogen peroxide- and glutathioneassociated mechanisms of acclimatory stress tolerance and signaling. Plant Physiology 100, 241–254. S.Y. Fujimoto, M. Ohta, A. Usui, H. Shinshi and M. Ohme-Takagi (2000) Arabidopsis ethyleneresponsive element binding factors act as transcriptional activators or repressors of GCC boxmediated gene expression. The Plant Cell 12, 393–404. M.J. Gao, G. Allard, L. Byass, A.M. Flanagan and J. Singh (2002) Regulation and characterization of four CBF transcription factors from Brassica napus. Plant Molecular Biology 49, 459–471. S.J. Gilmour, R.K. Hajela and M.F. Thomashow (1988) Cold acclimation in Arabidopsis thaliana. Plant Physiology 87, 745–750. S.J. Gilmour, A.M. Sebolt, M.P. Salazar, J.D. Everard and M.F. Thomashow (2000) Overexpression of the Arabidopsis CBF3 transcriptional activator mimics multiple biochemical changes associated with cold acclimation. Plant Physiology 124, 1854–1865. S.J. Gilmour, D.G. Zarka, E.J. Stockinger, M.P. Salazar, J.M. Houghton and M.F. Thomashow (1998) Low temperature regulation of the Arabidopsis CBF family of AP2 transcriptional activators as an early step in cold-induced COR gene expression. The Plant Journal 16, 433–442. Y. Guo, L. Xiong, M. Ishitani and J.K. Zhu (2002) An Arabidopsis mutation in translation elongation factor 2 causes superinduction of CBF/DREB1 transcription factors genes but blocks the induction of their downstream targets under low temperature. Proceedings of the National Academy of Sciences of the United States of America 99, 7786–7791. C.L. Guy (1990) Cold acclimation and freezing stress tolerance: role of protein metabolism. Annual Review of Plant Physiology and Plant Molecular Biology 41, 187–223. S.L. Harmer, J.B. Hogenesch, M. Straume, et al. (2000) Orchestrated transcription of key pathways in Arabidopsis by the circadian clock. Science 290, 2110–2113. T. Hobo, M. Asada, Y. Kowyama and T. Hattori (1999) ACGT-containing abscisic acid and largely impaired gene regulation in the abi1-1 mutant. Molecular and General Genetics 19, 679–689. K. Ichimura, T. Mizoguchi, R. Yoshida, T. Yuasa and K. Shinozaki (2000) Various abiotic stresses rapidly activate Arabidopsis MAP kinases ATMPK4 and ATMPK6. The Plant Journal 24, 655– 665. M. Ishitani, A.L. Majumder, A. Bornhouser, C.B. Michalowski, R.G. Jensen and H.J. Bohnert (1996) Coordinate transcriptional induction of myo-inositol metabolism during environmental stress. The Plant Journal 9, 537–548. M. Ishitani, L. Xiong, H. Lee, B. Stevenson and J.K. Zhu (1998) HOS1, a genetic locus involved in cold-responsive gene expression in Arabidopsis. The Plant Cell 10, 1151–1161. M. Ishitani, L. Xiong, B. Stevenson and J.K. Zhu (1997) Genetic analysis of osmotic and cold stress signal transduction in Arabidopsis: interactions and convergence of abscisic acid-dependent and abscisic acid-independent pathways. The Plant Cell 9, 1935–1949. R. Jaenicke and T.E. Creighton (1993) Junior chaperones. Current Biology 3, 234–235. K.R. Jaglo, S. Kleff, K.L. Amundsen, et al. (2001) Components of the Arabidopsis Crepeat/dehydration-responsive element binding factor cold-response pathway are conserved in Brassica napus and other plant species. Plant Physiology 127, 910–917. K.R. Jaglo-Ottosen, S.J. Gilmour, D.G. Zarka, O. Schabenberger and M.F. Thomashow (1998) Arabidopsis CBF1 over expression induces COR genes and enhances freezing tolerance. Science 280, 104–106.
304
REGULATION OF TRANSCRIPTION IN PLANTS
U. Jakob and J. Buchner (1994) Assisting spontaneity: the role of HSP90 and smHSPs as molecular chaperones. Trends in Biochemical Sciences 19, 205–211. C. Jonak, S. Kieferl, W. Ligterink, P.J. Barker, N.S. Huskisson and H. Hirt (1996) Stress signalling in plants: a mitogen-activated protein kinase pathway is activated by cold and drought. Proceedings of the National Academy of Sciences of the United States of America 93, 11274–11279. Y. Kagaya, K. Ohmiya and T. Hattori (1999) RAV1, a novel DNA-binding protein, binds to bipartite recognition sequence through two distinct DNA-binding domains uniquely found in higher plants. Nucleic Acids Research 27, 470–478. H.G.W. Kang and K. Singh (2000) Characterization of salicylic acid-responsive, Arabidopsis Dof domain proteins: overexpression of OBP3 leads to growth defects. The Plant Journal 21, 329– 339. F. Kaplan, J. Kopka, D.W. Haskell, et al. (2004) Exploring the temperature-stress metabolome of Arabidopsis. Plant Physiology 136, 4159–4168. M. Kasuga, Q. Liu, S. Miura, K. Yamaguchi-Shinozaki and K. Shinozaki (1999) Improving plant drought, salt, and freezing tolerance by gene transfer of a single stress-inducible transcription factor. Nature Biotechnology 17, 287–291. M. Kasuga, S. Miura, K. Shinozaki and K. Yamaguchi-Shinozaki (2004) A combination of the Arabidopsis DREB1A gene and stress-inducible rd29A promoter improved drought and lowtemperature stress tolerance in tobacco by gene transfer. Plant and Cell Physiology 45, 346–350. B.H. Kim and F. Sch¨offl (2002) Interaction between Arabidopsis heat shock transcription factor 1 and 70 kDa heat shock proteins. Journal of Experimental Botany 367, 371–375. H. Knight, A.J. Trewavas and M.R. Knight (1996) Cold calcium signalling in Arabidopsis involves two cellular pools and a change in calcium signature after acclimation. The Plant Cell 8, 489–503. H. Knight, E.L. Veale, G.J. Warren and M.R. Knight (1999) The sfr6 mutation in Arabidopsis suppresses low-temperature induction of genes dependent on CRT/DRE sequence motif. The Plant Cell 11, 875–886. H. Knight, D.G. Zarka, H. Okamoto, M.F. Thomashow and M.R. Knight (2004) Abscisic acid induces CBF gene transcription and subsequent induction of cold regulated genes via the CRT promoter element. Plant Physiology 135, 1710–1717. J.A. Kreps, Y. Wu, H.S. Chang, T. Zhu, X. Wang and J.F. Harper (2002) Transcriptome changes for Arabidopsis in response to salt, osmotic, and cold stress. Plant Physiology 130, 2129–2141. T. Kusano, T. Berberich, M. Harada, N. Suzuki and K. Sugawara (1995) A maize DNA-binding protein with bZIP motif is induced by low temperature. Molecular and General Genetics 248, 507–517. V. L˚ang, E. M¨antyl¨a, B. Welin, B. Sundberg and T.E. Palva (1994) Alterations in water status, endogenous abscisic acid content, and expression of rab 18 gene during the development of freezing tolerance in Arabidopsis thaliana. Plant Physiology 104, 1341–1349. J. Larkindale and M.R. Knight (2002) Protection against heat stress-induced oxidative damage in Arabidopsis involves calcium, abscisic acid, ethylene, and salicylic acid. Plant Physiology 128, 682–695. H. Lee, Y. Guo, M. Ohta, L. Xiong, B. Stevenson and J.K. Zhu (2002) LOS2, a genetic locus required for cold-responsive gene transcription encodes a bi-functional enolase. The EMBO Journal 21, 2692–2702. H. Lee, L. Xiong, Z. Gong, M. Ishitani, B. Stevenson and J.K. Zhu (2001) The Arabidopsis HOS1 gene negatively regulates cold signal transduction and encodes a RING finger protein that displays cold-regulated nucleo-cytoplasmic partitioning. Genes and Development 15, 912–924. J.H. Lee, A. H¨ubel and F. Sch¨offl (1995) Derepression of the activity of genetically engineered heat shock factor causes constitutive synthesis of heat shock proteins and increased thermotolerance in transgenic Arabidopsis. The Plant Journal 8, 603–612. J.H. Lee and F. Sch¨offl (1996) HSP70 antisense genes affect the expression of HSP70/HSC70, the regulation of HSF, and the acquisition of thermotolerance in transgenic Arabidopsis. Molecular and General Genetics 252, 11–19. H.-Y. Li, C.-S. Chang, L.-S. Lu, C.-A. Liu, M.-T. Chang, and Y.-Y. Charng (2003) Over-expression of Arabidopsis thaliana heat shock factor gene (AtHsf1b) enhances chilling tolerance in transgenic tomato. Botanic Bulletin of Academia Sinica 44, 129–140.
TEMPERATURE-REGULATED GENE EXPRESSION
305
Q. Liu, M. Kasuga, Y. Sakuma, et al. (1998) Two transcription factors, DREB1 and DREB2, with an EREBP/AP2 DNA binding domain separate two cellular signal transduction pathways in droughtand low temperature-responsive gene expression respectively, in Arabidopsis. The Plant Cell 10, 1391–1406. C. Lohmann, G. Eggers-Schumacher, M. Wunderlich and F. Sch¨offl (2004) Two different heat shock transcription factors regulate immediate early expression of stress genes in Arabidopsis. Molecular Genetics and Genomics 271, 11–21. G. Lu, A.L. Paul, D.R. McCarty and R.J. Ferl (1996) Transcription factor veracity: is GBF3 responsible for ABA regulated expression of Arabidopsis Adh? The Plant Cell 8, 847–857. K. Maleck, A. Levine, T. Eulgem, et al. (2000) The transcriptome of Arabidopsis thaliana during systemic acquired resistance. Nature Genetics 26, 403–410. M.L. Martn and L. Busconi (2001) A rice membrane-bound calcium-dependent protein kinase is activated in response to low temperature. Plant Physiology 125, 1442–1449. K. Maruyama, Y. Sakuma, M. Kasuga, et al. (2004) Identification of cold inducible downstream genes of the Arabidopsis DREB1A/CBF3 transcriptional factor using two microarray systems. The Plant Journal 38, 982–993. D.R. McMillan, X. Xiao, L. Shao, K. Graves and I.J. Benjamin (1998) Targeted disruption of heat shock transcription factor 1 abolishes thermotolerance and protection against heat-inducible apoptosis. The Journal of Biological Chemistry 273, 7523–7528. T. Mizoguchi, K Irie, T. Hirayama, et al. (1996) A gene encoding a mitogen-activated protein kinase kinase kinase is induced simultaneously with genes for a mitogen-activated protein kinase and an S6 ribosomal protein kinase by touch, cold, and water stress in Arabidopsis thaliana. Proceedings of the National Academy of Sciences of the United States of America 93, 765–769. S.S. Mohapatra, L. Wolfraim, R.J. Poole and R.S. Dhindsa (1989) Molecular cloning and relationship to freezing tolerance of cold acclimation specific genes of alfalfa. Plant Physiology 89, 375–380. A.F. Monroy and R.S. Dhindsa (1995) Low-temperature signal transduction: induction of cold acclimation-specific genes of alfalfa by calcium at 25◦ C. The Plant Cell 7, 321–331. A.F. Monroy, V. Sangwan and R.S. Dhindsa (1998) Low temperature signal transduction during cold acclimation: protein phosphatase 2A as an early target for cold-inactivation. The Plant Journal 13, 653–660. D.D. Mosser, J. Duchaine and B. Massie (1993) The DNA-binding activity of the human heat shock transcription factor is regulated in vivo by hsp70. Molecular and Cellular Biology 13, 5427–5438. H. Nakagawa, K. Ohmiya and T. Hattori (1996) A rice bZIP protein, designated OSBZ8, is rapidly induced by abscisic acid. The Plant Journal 9, 217–227. T. Nanjo, M. Kobayashi, Y. Yoshiba, Y. Kakubari, K. Yamaguchi-Shinozaki and K. Shinozaki (1999) Antisense suppression of proline degradation improves tolerance to freezing and salinity in Arabidopsis thaliana. FEBS Letters 461, 205–210. L. Nover, K. Bharti, P. D¨oring, S.K. Mishra, A. Ganguli and K.-D. Scharf (2001) Arabidopsis and the heat stress transcription factor world: how many heat stress transcription factors do we need? Cell Stress Chaperones 6,177–189. L. Nover, K.D. Scharf, D. Gagliardi, P. Vergne, E. Czarnecka-Verner and W.B. Gurley (1996) The Hsf world: classification and properties of plant heat stress transcription factors. Cell Stress and Chaperones 1, 215–223. F. Novillo, J.M. Alonso, J.R. Ecker and J. Salinas (2004) CBF2/DREB1C is a negative regulator of CBF1/DREB1B and CBF3/DREB1A expression and plays a central role in stress tolerance in Arabidopsis. Proceedings of the National Academy of Sciences of the United States of America 101, 3985–3990. S.J. Oh, S.I. Song, Y.S. Kim, et al. (2005) Arabidopsis CBF3/DREB1A and ABF3 in transgenic rice increased tolerance to abiotic stress without stunting growth. Plant Physiology 138, 341–351. J.K. Okamuro, B. Caster, R. Villarroel, M.V. Montagu and K.D. Jofuku (1997) The AP2 domain of APETALA2 defines a large new family of DNA binding proteins in Arabidopsis. Proceedings of the National Academy of Sciences of the United States of America 94, 7076–7081. B.L. Ovar, V. Sangwan, F. Omann and R.S. Dhindsa (2000) Early steps in cold sensing by plant cells: the role of actin cytoskeleton and membrane fluidity. The Plant Journal 23, 785–794.
306
REGULATION OF TRANSCRIPTION IN PLANTS
I. Panchuck, R. Volkov and F. Sch¨offl (2002) Heat stress and HSF-dependent expression of ascorbate peroxidase in Arabidopsis. Plant Physiology 129, 838–853. T.J. Panikulangara, G. Eggers-Schumacher, M. Wunderlich, H. Stransky and F. Sch¨offl (2004) Galactinol synthase 1, a novel heat-inducible and HSF-target gene responsible for heat-induced synthesis of Raffinose Family Oligosaccharides in Arabidopsis. Plant Physiology 136, 3148– 3158. M. Papi, S. Sabatini, D. Bouchez, C. Camilleri, P. Costantino and P. Vittorioso (2000) Identification and disruption of an Arabidopsis zinc finger gene controlling seed germination. Genes and Development 14, 28–33. C. Plieth, U.P. Hansen, H. Knight and M.R. Knight (1999) Temperature sensing by plants: the primary characteristics of signal perception and calcium response. The Plant Journal 18, 491– 497. R. Pr¨andl, K. Hinderhofer, G. Eggers-Schumacher and F. Sch¨offl (1998) HSF3, a new heat shock factor from Arabidopsis thaliana, derepresses the heat shock response and confers thermotolerance when overexpressed in transgenic plants. Molecular and General Genetics 258, 269–278. A. Reindl and F. Sch¨offl (1997) Interaction between the Arabidopsis thaliana heat shock transcription factor HSF1 and the TATA-binding protein TBP. FEBS Letters 436, 318–322. J.L. Riechmann, J. Heard, G. Martin, et al. (2000) Arabidopsis transcription factors: genome-wide comparative analysis among eukaryotes. Science 290, 2105–2110. L. Rizhsky, S. Davletova, H. Liang and R. Mittler (2004) The zinc finger protein ZAT12 is required for cytosolic ascorbate peroxidase 1 expression during oxidative stress in Arabidopsis. The Journal of Biological Chemistry 279, 11736–11743. L. Rizhsky, H. Liang, J. Shuman, V. Shulaev, S. Davletova and R. Mittler (2004) When defence pathways collide. The response of Arabidopsis to a combination of drought and heat stress. Plant Physiology 134, 1683–1696. Y. Saijo, S. Hata and K. Izui (1998) Characterization of a rice cold stress inducible calcium-dependent protein kinase 475(H1p18). Plant Cell Physiology Supplements 39, s127. Y. Saijo, S. Hata, J. Kyozuka, K. Shimamoto and K. Izui (2000) Over-expression of a single Ca+2 dependent protein kinase confers both cold and salt/drought tolerance on rice plants. The Plant Journal 23, 319–327. Y. Sakuma, Q. Liu, J.G. Dubouzet, H. Abe, K. Shinozaki and K. Yamaguchi-Shinozaki (2002) DNA binding specificity of the ERF/AP2 domain of Arabidopsis DREBs, transcription factors involved in dehydration and cold inducible gene expression. Biochemical and Biophysical Research Communications 290, 998–1009. S.H. Satyal, D. Chen, S.G. Fox, J.M. Kramer and R.I. Morimoto (1998) Negative regulation of the heat shock transcriptional response by HSBP1. Genes and Development 12, 1962–1974. K.-D. Scharf, S. Rose, W. Zott, F. Sch¨offl and L. Nover (1990) Three tomato genes code for heat stress transcription factors with a region of remarkable homology to the DNA-binding domain of the yeast HSF. The EMBO Journal 9, 4495–4501. K.-D. Scharf, M. Siddique and E. Vierling (2001) The expanding family of Arabidopsis thaliana small heat stress proteins and a new family of proteins containing a-crystallin domains (Acd proteins). Cell Stress and Chaperones 6, 225–237. F. Sch¨offl and R. Pr¨andl (1999) Derepression of the heat shock protein synthesis in transgenic plants. In: Plant Responses to Environmental Stress (eds M. Smallwood, C. Calvert and D. Bowles), pp. 65–73. BIOS Scientific, Oxford. F. Sch¨offl, R. Pr¨andl and A. Reindl (1998) Regulation of the heat shock response. Plant Physiology 117, 1135–1141. F. Sch¨offl, R. Pr¨andl and A. Reindl (1999). Molecular responses to heat stress. In: Molecular Responses to Cold, Drought, Heat and Salt Stress in Higher Plants (eds K. Shinozaki and K. YamaguchiShinozaki), pp. 81–98. R.G. Landes, Austin, TX. M. Seki, M. Narusaka, H. Abe, et al. (2001) Monitoring the expression pattern of 1300 Arabidopsis genes under drought and cold stresses by using a full length cDNA microarray. The Plant Cell 13, 61–72.
TEMPERATURE-REGULATED GENE EXPRESSION
307
M. Seki, M. Narusaka, J. Ishida, et al. (2002) Monitoring the expression profiles of 7000 Arabidopsis genes under drought, cold and high salinity stresses using a full-length cDNA microarray. The Plant Journal 31, 279–292. Q. Shen and T.J. Ho (1997) Promoter switches specific for abscisic acid (ABA)-induced gene expression in cereals. Physiologia Plantarum 101, 653–664. C. Stasolla, L. van Zyl, U. Egertsdotter, D. Craig, W. Liu and R. Sederoff (2003) The effects of polyethylene glycol on gene expression of developing white spruce somatic embryos. Plant Physiology 131, 49–60. P.L. Steponkus, M. Uemura, R.A. Joseph, S.J. Gilmour and M.F. Thomashow (1998) Mode of action of the COR15a gene on the freezing tolerance of Arabidopsis thaliana. Proceedings of the National Academy of Sciences of the United States of America 95, 14570–14575. E.J. Stockinger, S.J. Gilmour and M.F. Thomashow (1997) Arabidopsis thaliana CBF1 encodes an AP2 domain-containing transcriptional activator that binds to the C-repeat/DRE, a cis-acting DNA regulatory element that stimulates transcription in response to low temperature and water deficit. Proceedings of the National Academy of Sciences of the United States of America 94, 1035–1040. T. Taji, C. Ohsumi, S. Iuchi, et al. (2002) Important roles of drought- and cold-inducible genes for galactinol synthase in stress tolerance in Arabidopsis thaliana. The Plant Journal 29, 417– 426. M. Tanabe, Y. Kawazoe, S. Takeda, R.I. Morimoto, K. Nagata and A. Nakai (1998) Disruption of the HSF3 gene results in the severe reduction of heat shock gene expression and loss of thermotolerance. The EMBO Journal 17, 1750–1758. M. Teige, E. Scheikl, T. Eulgem, et al. (2004) The MKK2 pathway mediates cold and salt stress signalling in Arabidopsis. Molecular Cell 15, 141–152. M.F. Thomashow (1994) Arabidopsis thaliana as a model for studying mechanisms of plant cold tolerance. In: Arabidopsis (eds E. Meyerowitz and C. Somerville), pp. 807–834. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. M.F. Thomashow (1998) Role of cold responsive genes in plant freezing tolerance. Plant Physiology 118, 1–8. M.F. Thomashow (1999) Plant cold acclimation, freezing tolerance genes and regulatory mechanisms. Annual Review of Plant Physiology and Plant Molecular Biology 50, 571–599. A. Ueda, W. Shi, T. Nakamura and T. Takabe (2002) Analysis of salt-inducible genes in barley roots by differential display. Journal of Plant Research 115, 119–130. J.T. Vogel, D.G. Zarka, H.A. Van Buskirk, S.G. Fowler and M.F. Thomashow (2005) Roles of CBF2 and ZAT12 transcription factors in configuring the low temperature transcriptome of Arabidopsis. The Plant Journal 41, 195–211. Z. Wang, S.J. Triezenberg, M.F. Thomashow and E.J. Stockinger (2005) Multiple hydrophobic motifs in Arabidopsis CBF1 COOH-terminus provide functional redundancy in trans-activation. Plant Molecular Biology 58, 543–559. G. Warren, R. McKown, A. Marin and R. Teutonico (1996) Isolation of mutations affecting the development of freezing tolerance in Arabidopsis thaliana (L.) Heynh. Plant Physiology 111, 1011–1019. C. Wu (1995) Heat shock transcription factors: structure and regulation. Annual Review of Cell and Developmental Biology 11, 441–469. M. Wunderlich, W. Werr and F. Sch¨offl (2003) Generation of dominant negative effects on the heat shock response in Arabidopsis thaliana by transgenic expression of a chimeric HSF1 protein fusion construct. The Plant Journal 35, 442–451. X. Xiao, X. Zou, A.A. Davis, et al. (1999) HSF1 is required for extra-embryonic development, postnatal growth and protection during inflammatory responses in mice. The EMBO Journal 18, 5943– 5952. Z. Xin and J. Browse (1998) eskimo 1 mutants of Arabidopsis are constitutively freezing-tolerant. Proceedings of the National Academy of Sciences of the United States of America 95, 7799– 7804.
308
REGULATION OF TRANSCRIPTION IN PLANTS
K. Yamaguchi-Shinozaki and K. Shinozaki (1994) A novel cis-acting element in an Arabidopsis gene is involved in responsiveness to drought, low temperature, or high-salt stress. The Plant Cell 6, 251–264. D.G. Zarka, J.T. Vogel, D. Cook and M.F. Thomashow (2003) Cold induction of Arabidopsis CBF genes involves multiple ICE (Inducer of CBF Expression) promoter elements and a cold-regulatory circuit that is desensitized by low temperature. Plant Physiology 133, 910–918. L. Zhang, C. Lohmann, R. Pr¨andl and F. Sch¨offl (2003) Heat stress-dependent DNA binding of Arabidopsis Heat Shock Transcription Factor HSF1 to heat shock gene promoters in Arabidopsis suspension culture cells in vivo. Biological Chemistry 384, 959–963. X. Zhang, G.S. Fowler, H. Cheng, et al. (2004) Freezing sensitive tomato has a functional CBF regulon that differs from that of freezing tolerant Arabidopsis. The Plant Journal 39, 905–919. J. Zou, Y. Guo, T. Guettouche, D.F. Smith and R. Voellmy (1998) Repression of heat shock transcription factor HSF1 activation by HSP90 (HSP90 complex) that forms a stress-sensitive complex with HSF1. Cell 94, 471–480. J.K. Zhu (2002) Salt and drought stress signal transduction in plants. Annual Review of Plant Biology 53, 247–273.
13 Applications of inducible transcription in plant research and biotechnology Brian Tomsett, Angela Tregova and Mark Caddick
13.1
Introduction
Functional genomics is the key to our understanding of plants and their interactions in the environment. Analysis of plant genomes has revealed an amazing complexity, including a significant proportion of sequences that are classified as having an ‘unknown function’. Even if the general class of function can be guessed, the degree of apparent redundancy among multi-gene families is such that their exact role requires investigation – are they functional in different cell types, different stages of development or in response to different environmental cues/stresses, etc.? The absolute definition of function for any gene, encompassing all possible epistatic or pleiotropic effects in all tissues and at all stages of development, requires analysis through transgenic plants. For this, transcriptionally regulated expression systems are a necessary component of the plant functional genomics tool kit. Most research using transgenic plants utilises constitutive promoters to direct expression at maximal rate in all tissues of the plant. For genotypically dominant markers, the phenotype is visible immediately. For recessive characters, selfing the plant can produce homozygous, hemizygous and azygous progeny to explore all possible interactions. However, for some genes, there can be significant complications. Aberrant expression can disrupt normal function, for example by triggering embryo lethality or senescence, or by preventing meiosis, leading to sterility. Furthermore, deleterious effects at an early stage of development may mask a more subtle phenotype later in the life cycle. Inducible expression systems provide the ability to upregulate or downregulate one or more genes temporally, i.e. at a defined time in growth or development, minimising such problems. While this chapter ranges broadly across this topic of inducible expression systems in plants, it cannot be exhaustive, since there is much current activity and even greater potential for both its development and its application. It includes an overview of inducible transgene expression systems, and then examines some examples of their use in plant functional genomics and plant biotechnology.
13.2
Conditional expression of transgenes
There have been a series of excellent reviews dedicated to this topic, which have examined in detail the mechanisms of inducible expression systems (Gatz, 1996,
310
REGULATION OF TRANSCRIPTION IN PLANTS
1997; Gatz and Lenk, 1998; Zuo and Chua, 2000; Padidam, 2003; Wang et al., 2003; Tang et al., 2004). Here, we attempt to draw together principles and options in relation to their benefits, limitations and applications. As we will show, there is no ‘perfect’ system for conditional expression, and careful consideration is necessary to achieve satisfactory results. Furthermore, as we point out below, research and biotechnology are moving into a phase where we need multiple approaches, and so several of the systems described here will need to be used in conjunction.
13.2.1
The principles of regulated expression
Transcription is a process involving a complex of molecules acting in a concerted fashion to deliver an RNA transcript of defined length and sequence. For some genes related to core housekeeping functions, expression will be constitutive – all cells, at all times. For many genes however, their promoter and enhancer sequences combine to precisely define expression levels for a specific cell type under specific conditions. Thus, expression could be controlled by signals for differentiation or development (specific cell types, tissues or organs; e.g. meristems), spatially (cells in defined positions, and not similar cells in other locations; e.g. abscission zones), temporally (all cells at some times and not others; e.g. circadian clock genes) or in response to stimuli from the environment (abiotic or biotic stress; e.g. excess salt or pathogen attack). Potentially, any regulated promoter could be adapted for the conditional expression of a transgene. For some applications, an endogenous plant promoter will be the obvious choice, particularly where it is desirable to coordinate the expression of the transgene with that of a native gene and any corresponding cellular activity, while in others there may be a requirement to control expression entirely artificially. Either way, the principles will be similar. The regulation of transcription can be positive or negative. Negative regulatory circuits rely upon the production of a molecule that prevents transcription, for example by blocking the binding or destabilising the transcription initiation complex. A target gene is thus repressed until such time as an effector molecule interacts with the repressor to unblock transcription, referred to as de-repression. In positive-regulated systems, the promoter of the target gene remains quiescent until an activated regulatory protein stimulates transcription. In some cases, the inactive regulatory protein could be bound to the promoter/enhancer region of the target gene but not direct transcription until activated by the effector molecule. In others, the interaction of an effector ligand with the regulatory protein stimulates binding to the promoter and/or enhancer elements, and directly or indirectly promotes formation of the transcription initiation complex.
13.2.2
Generic mechanisms
The construction of all inducible expression systems relies upon a series of essential elements that will differ slightly, depending on the basis for the regulation (Figure 13.1). Whether positive or negative regulation is employed, the heart of the
INDUCIBLE TRANSGENE EXPRESSION
311
Figure 13.1 Generic mechanisms for conditional regulation of transcription. (R = the regulator gene and its protein product; TG = a transgene whose protein product is signified by ; RP = a reporter gene, which produces a product ; pR = is the promoter directing expression of the regulatory gene; pA = the promoter regulated by protein R and the inducer (I), which can be used for the expression of any transgene, a reporter gene, or both; t1, t2 and t3 = transcriptional terminator sequences for the three gene constructs, which should preferably be different.)
312
REGULATION OF TRANSCRIPTION IN PLANTS
system will be a regulatory protein (R) that binds to sequences associated with a promoter capable of activation (pA). In endogenous (single-component) systems, the regulatory protein is the product of a native gene within the host’s genome that is part of a regulatory mechanism responding to an internal or external stimulus (the inducer, I); a single transgene driven by pA will respond in the same manner as other genes in the circuit. In heterologous (two-component) systems, two transgenes will be required, one to produce a regulatory protein that responds to an inducer (I), the other being the gene of interest driven by pA. In the development of ‘gene switches’, reporter genes (RP) have been the transgenes of choice to determine the parameters of expression, including chloramphenicol acetyltransferase (CAT), glucuronidase (GUS), luciferase (LUC) and green fluorescent protein (GFP). In two-component systems, expression of the regulatory protein must be considered. In most studies to date, the promoter (pR) directing the expression of the regulator has been a constitutive promoter, delivering high levels in all tissues, often the Cauliflower Mosaic Virus (CaMV) 35S promoter. In part, this has been to ensure that sufficient amounts of regulatory protein would be available to ensure maximal induced expression. However, other options may be more favourable to optimise expression of a specific gene and to reduce background levels of expression. The structure of the regulatory protein and pA will depend on the system used. Proteins activating transcription generally require at least three major domains: a DNA-binding domain that recognises defined sequences within the target promoter; a transactivation domain for efficient transcription; and a ligand-binding domain that binds the molecule triggering a conformational change to initiate expression (see Tyagi, 2001). A specific regulatory protein may be utilised, but the modular nature of many transcription factors has meant that synthetic/chimeric regulatory proteins can be developed combining three domains from different genes. Similarly, in some systems, pA will be a native promoter from a characterised gene, containing appropriate response elements; in others, it will be an artificial construct containing plant promoter sequences, required to allow the initiation of transcription by the complex containing RNA polymerase, together with response elements that bind the specific regulatory protein that will in turn either repress or activate transcription. For inducible promoters, a minimal promoter, most often from the CaMV 35S promoter, can direct expression only when a regulatory protein containing an activation domain is bound to an adjacent response element. The final component of any transgene is the terminator sequence. Most constructs use an ∼200–300-bp sequence from a well-characterised gene (e.g. from 35S RNA of CaMV, or the nos gene of the T-DNA of Agrobacterium tumefaciens); this positions a polyadenylation signal downstream of the final translation stop codon to ensure that a mature and stable mRNA is produced.
13.2.3
Physical stimuli for activation
Recognition of physical stimuli relies upon the native responses of the host plant, and hence the regulatory protein will be an endogenous gene in the host genome. Those currently used are all stress related and so come with the limitation that
INDUCIBLE TRANSGENE EXPRESSION
313
the plant may be exhibiting ‘abnormal’ (stress) physiology at the time of transgene expression. Here we review three approaches: heat-, wound- and pathogen-inducible expression. Heat-shock inducible promoters are triggered by an 8–12◦ C rise in temperature above the ‘normal’ ambient temperature of growth. The plant responds by inducing the transcription of a defined set of genes, encoding the so-called heat-shock proteins (HSPs), while many other genes show decreased transcription. In an early study, Callis et al. (1988) introduced a construct in which the CAT reporter gene was transcribed from the maize hsp70 promoter in maize protoplasts. At 26◦ C, they detected very low CAT activity, but this increased a further 35-fold by 37◦ C, and 200-fold at 40◦ C. Simultaneous expression of a 35S-luc reporter showed an approximately threefold reduction in expression at 37◦ C relative to the 26◦ C level. As will be discussed below (Section 13.3), heat-inducible systems have been used successfully to express genes in whole plants. Plants have defence mechanisms that protect them against the invasion of competing organisms, part of which is a wounding response that alters the physiology of the cells and activates defence. They respond to wounding in several ways: by producing reactive oxygen species, phytoalexins, ethylene, pathogenesis-related (PR) proteins, methyljasmonate, systemin, abscissic acid, auxin and/or salicylic acid. Many of these also act as signalling molecules causing cascades of gene expression of subsets within the response. Furthermore, some pathogen-derived molecules also activate certain genes and trigger defence. The promoters of these genes are thus a potential route for the expression of transgenes in an inducible manner. An important consideration is that many such promoters are inducible by a number of stimuli, which can be advantageous or disadvantageous, depending on the purpose of transgene expression. For example, Sa et al. (2003) have used the GAFP-2 promoter from a Gastrodia elata antifungal protein gene to examine reporter gene activity in transgenic tobacco plants. There was a significant basal level (uninduced) GUS activity that was higher in roots than in stems and leaves, and strongly localised to vascular tissues. Expression was increased up to tenfold upon induction by Trichoderma viride, or on application of jasmonic acid or salicylic acid. In a similar approach, Perera and Jones (2004) have examined expression driven from a peroxidase gene promoter from the tropical forage legume Stylosanthes humilis. In transgenic tobacco, chewing or sucking insects induced GUS/GFP expression in cells around the feeding sites, but wounding or methyljasmonate could also activate reporter gene activity. Physical stress, for example from heat or wounding, can induce gene expression but these may not be as convenient as a response triggered by the application of a chemical in easily controlled doses, at specific sites and at defined times. Thus, even those laboratory studies employing pathogen/wound-inducible promoters, often use one of the chemical elicitors that underlie the response, namely methyljasmonate and salicylic acid. Yamada et al. (2002) suggested the pathogenesis-related protein 1a (PR1a) promoter could be used as a benzothiadiazole-inducible promoter. While they demonstrate its effectiveness, this raises questions about their widespread use, due to both the wide range of conditions that could induce expression and the range
314
REGULATION OF TRANSCRIPTION IN PLANTS
of native genes that will also respond to the specific stimulus. For most laboratory studies however, this may not matter.
13.2.4
Chemical activators related to plant metabolism
Chemical activators can be divided into those that are known to be part of the plant’s normal metabolism, and those that appear to be unrelated (which will be discussed in Section 13.2.5). It is also of importance here whether the regulator protein is native or heterologous. In some studies, a native regulatory protein serves to dissect the physiology/development of the plant under investigation. For example, an auxin-inducible reporter gene has been used to analyse auxin distribution in the moss Physcomitrella patens (Bierfreund et al., 2003). Here the endogenous inducer and the native regulatory protein were used to drive the expression from heterologous promoters, one synthetic and the other from soybean. In this instance, the native regulation was the focus of the study. However, for many transgenes, regulation is required under very defined conditions, and hence isolation of expression is (partially) achieved by using a regulatory protein unrelated to the plant’s own transcription regulators, derived from another organism. Ideally these would respond to an inducer that is not normally present in plants nor induces a significant response in the absence of the regulatory system. Two of the best characterised examples of this approach use fungal regulatory proteins that act positively on their responsive promoters. Mett et al. (1993) described a two-component copper-controlled gene expression system for plants, which was based upon the regulatory circuit of the yeast metallothionein gene. A transcription factor (ACE1) was expressed from the 35S promoter, which could activate transcription in the presence of copper ions from a chimeric promoter (pA) containing the metal responsive element (MRE) binding domain linked to a −90 minimal 35S promoter; GUS was used as the reporter. In transgenic tobacco, significant pA-directed GUS expression occurred with copper concentrations >5 μM when applied to the root solution, and from foliar sprays with as little as 0.5 μM CuSO4 . Induction was up to 50-fold over the basal level. Given that copper is a normal component of the plant nutrient solution (∼0.1 μM), one might expect a high basal level of activity. However, uninduced levels of GUS were low. Furthermore, levels of GUS were similar in plants (±Cu) containing pA:GUS but lacking the regulatory transgene, as in untransformed controls, indicating that the promoter did not respond to an endogenous transcription factor. The regulation from this system can be tight (non-leaky) and hence despite copper being normally available in the environment, and readily taken up by the plant, the system has great utility for transgene expression (Mackenzie et al., 1998). Granger and Cyr (2001) have further demonstrated its utility in Arabidopsis thaliana grown in agar; expression in roots is highest in the root cap and epidermis, which may indicate a higher exposure to copper in the medium. Unfortunately, expression was not extensively analysed in the above-ground parts of the plants, and while GFP could be detected, the transport of GFP itself rather than the inducer (Cu) cannot be excluded. More recently, even higher levels of expression have been reported through a modification to the MRE (Peng et al., 2003).
INDUCIBLE TRANSGENE EXPRESSION
315
One possible advantage to this system is that expression can be sustained for some considerable time, presumably because while copper ions may be sequestered away from the ACE1 protein, the turnover is not as rapid as in metabolite-based systems. Thus, a single application of 0.5 μM copper as a foliar spray led to GUS activity at high levels at day 4, and remained high until at least day 12, before declining by day 16 (Mett et al., 1993). Furthermore, a degree of reversibility was demonstrated after removal of surface copper by washing with water, which triggered a decline in activity to below 50% of the maximally induced activity 2 days later. Clearly, this system is not without its limitations. Symptoms of copper toxicity appeared at 20 and 15 days in response to 5-μM and 50-μM copper (respectively) being applied to the roots. This was not the case for foliar spraying at 0.5 μM. In addition, the utility of this system for tobacco cell culture has been questioned (Granger and Cyr, 2000). Another system is based upon the alc regulon of Aspergillus nidulans. The regulatory protein, AlcR, expressed constitutively using a plant promoter (generally 35S), binds to sites (derived from the alcohol dehydrogenase gene, alcA) in a chimeric promoter linked to a −31 minimal 35S promoter (Caddick et al., 1998; Salter et al., 1998). Transcription is activated by the presence of an effector ligand, either ethanol or acetaldehyde, and while some other compounds have been shown to trigger expression (e.g. 2-butanone, acetone and propan-1-ol), these are not generally as effective (Junker et al., 2003; Garoosi et al., 2005). Ethanol has been the most extensively studied inducer; expression is dose dependent and sufficiently sensitive that vapour can be used to activate expression (Roslan et al., 2001; Sweetman et al., 2002). The response is very rapid; for example, luciferase expression could be detected in the roots of A. thaliana seedlings within 1 h of ethanol addition (Roslan et al., 2001). Expression can be as high as an equivalent line carrying a 35S-driven transgene, and it peaks between 3 and 5 days postinduction, before declining (Salter et al., 1998; Garoosi et al., 2005). Expression can be sustained at a high level through the subsequent additions of inducer (Roslan et al., 2001; Schaarschmidt et al., 2004). Ethanol induction appears to give expression systemically throughout the plant, but some tissues have higher levels than others in mature plants (Garoosi et al., 2005). Application of acetaldehyde as an inducer has been demonstrated to give local induction (Schaarschmidt et al., 2004). The alc system is extremely effective for academic research and may have applicability to the field because the inducers are metabolites that can be degraded in the plant, and by microbes in the soil. However, these metabolites occur in plants under defined conditions. Hypoxia/anoxia leads to ethanolic fermentation, during which pyruvate is converted initially to acetaldehyde by pyruvate decarboxylase, which in turn is converted to ethanol by alcohol dehydrogenase (Drew, 1997). Artificial anoxia does induce alc, but a 3-day submergence of the root (but not aerial) systems of plants did not lead to induction (Salter et al., 1998; Roberts et al., 2005). However, seedlings grown in agar show higher basal levels of expression, and cells in both callus and suspension cultures have levels that are effectively constitutive (Roslan et al., 2001; Roberts et al., 2005). Since both copper and ethanol/acetaldehyde are part of the normal metabolism of the plant, it is inevitable that application of these compounds may change the
316
REGULATION OF TRANSCRIPTION IN PLANTS
physiology of the plant. The true extent of their effects has not been determined; however, an analysis of more than 60 metabolites in potato tubers revealed that ethanol treatment altered the levels of 13, but acetaldehyde induction affected only 3 (Junker et al., 2003). Clearly, this will not be critical for many studies, but should be considered prior to experimentation.
13.2.5
Chemical activators unrelated to plant metabolism
Given the potential problems associated with the use of an inducer or regulatory protein that is part of the normal plant repertoire, a number of heterologous and/or synthetic systems have been developed. Some of these rely upon regulatory elements from microbes, and others from animal systems. Initial interest stemmed from the negatively regulated tet (and lac) operon of enteric bacteria (see Gatz, 1997; Padidam, 2003). However, high levels of the repressor were required in the nucleus to maintain a low basal expression from the target transgene (David and Perrot-Rechenmann, 2001). But by combining the activation domain of the VP16 protein of human herpes simplex virus (HSV) with the DNA-binding domain of TetR, the chimeric regulatory protein bound to the tet operator sequences activated transcription in the absence of tetracycline, the addition of which prevented expression (Gossen and Bujard, 1992; Weinmann et al., 1994; Love et al., 2000, 2002). However, there have been reports of leakiness of the promoters (De Veylder et al., 2000), and methylation of the tet operator sequences (Bohner et al., 1999), both of which have been addressed during the development of a dual regulatory system discussed below. The key elements of the lac system have been used to define a novel hybrid transcriptional regulation system, termed pOp/LhG4 (Moore et al., 1998; Baroux et al., 2005). The pOp (=pA) consisted of two lac operators (2xRElac ) placed upstream of a minimal 35S promoter; a chimeric regulatory protein (LhG4) composed of a mutant DNA-binding domain from the Lac repressor (with increased binding efficiency) fused to the activation domain II from Gal4 of Saccharomyces cerevisiae. Unlike the other systems described in this section, this is not activated by a chemical; rather separate transgenic lines, one carrying the pOp-transgene and the other 35S-LhG4, are crossed to produce an F1 with excellent expression throughout the plant. While there is no temporal control here, as will be discussed below, it is perfectly stringent, and can be combined with spatial and/or chemically inducible systems. In an alternative approach, a number of groups have constructed chimeric transcription factors responsive to hormones from animal systems. Three groups have developed control systems based on the receptor for the insect hormone ecdysone (EcR) because of its potential for field application. Martinez et al. (1999) constructed a chimeric GVE regulatory protein using the DNA-binding domain and activation domain of the mammalian glucocorticoid receptor (GR) fused to the activation domain of V P16, and the hinge and ligand-binding domain (EcR) of Heliothis virescens. The target promoter (pA) had six copies of the glucocorticoid response element (GRE) attached to a minimal 35S promoter, and activation was achieved in transgenic tobacco through application of the commercially available
INDUCIBLE TRANSGENE EXPRESSION
317
non-steroidal ecdysone agonist tebufenozide. Unger et al. (2002) adopted a similar approach: their chimeric regulatory protein was made from VP16, Gal4 and the EcR from the European corn borer, Ostinia nubialis, and it was used to direct expression from 5xREgal (yeast Gal4-binding sites) linked to a minimal promoter after induction with the agonist methoxyfenozide. Padidam et al. (2003) constructed the GVE chimeric activator from Gal4, V P16 and the Choristoneura fumiferana EcR ligand-binding domain; another activator (LVE) was also constructed in which the Gal4 element was replaced by the LexA DNA-binding domain. Each was used with luciferase constructs with either 5xREgal or 6xRElex , and induced by application of methoxyfenozide in Arabidopsis and tobacco. In each of these systems, significant levels of gene expression could be achieved, at least equivalent to the best 35S promoter activity, and while there was variation in both the basal and the induced levels of response, low basal activity could be achieved. However, this could be a greater problem in some species, since endogenous phytoecdysteroids have been reported (Saez et al., 2000). Other systems have been based on oestrogen as an inducer. Bruce et al. (2000) used the human oestrogen receptor (ER) in a fusion protein with the maize C1 activation domain. Zuo et al. (2000) developed a chimeric regulatory protein, XVE, using the LexA DNA-binding domain, V P16 activation domain, with the ER ligandbinding domain to direct expression from a pA composed of 8xRElex in an oestrogendependent manner. High levels of expression were achieved (three- to fivefold 35S) without detectable basal levels. These systems appear to have high efficiency and low basal activity in transgenic plants; however, Zuo and Chua (2000) report deregulated expression in transiently transformed soybean cells, possible due to the presence of phytooestrogens. The most extensively used of the steroid-hormone-activated systems utilises the ligand-binding domain of the mammalian GR (Schena et al., 1991). Aoyama and Chua (1997) developed a chimeric regulatory protein, GVG, using the Gal4 DNAbinding domain, V P16 activation domain, together with the GR ligand-binding domain to direct expression from a pA composed of 6xREgal fused to a minimal 35S. Here, in the absence of an inducer ligand, GR associates with cytosolic proteins, but upon addition of a glucocorticoid inducer, the regulatory protein is translocated to the nucleus to bind to the target promoter (see Zuo and Chua, 2000). This has the advantage that compartmentation restricts non-ligand-directed interactions with pA, and thus the basal level of expression is minimised, while allowing rapid highlevel dose-dependent induction in the presence of inducer. Dexamethasone (dex), a synthetic glucocorticoid, is the most commonly used inducer, but other glucocorticoid derivatives have been shown to induce to differing extents (Aoyama and Chua, 1997; Tang and Newton, 2004). The system has been successfully adapted to rice by using the Gos2 gene promoter to drive the GVG construct (Ouwerkerk et al., 2001). However, the system is not without its limitations: dex toxicity has been an issue in rice at 10 μM (Ouwerkerk et al., 2001); severe growth defects, linked to the constructs themselves (rather than dex), have been observed in Arabidopsis; and furthermore the system has caused the induction of defence-related genes (Kang et al., 1999; Craft et al., 2005).
318
REGULATION OF TRANSCRIPTION IN PLANTS
The efficacy of the GR element for induction has led to its inclusion into other systems recently. pOp6/LhGR is a modification of the pOp/LhG4 system described above. The regulatory protein has been modified to include the GR ligand-binding domain at the N-terminus of the Lac DNA-binding domain and Gal4 activation domain; pOp6 (=pA) has been modified to include six ideal lac operators (6xRElac ) and the tobacco sequence in the 5 UTR (Craft et al., 2005; Samalova et al., 2005). pOp6/LhGR is thus a dex-inducible system, which shows very high inducible expression, and with a low basal level of activity. Roberts et al. (2005) have also used GR to solve a key problem with the alc system. Plant cells in tissue culture are often (partially) hypoxic and hence produce ethanol/acetaldehyde through ethanolic fermentation. Thus, alc tends to have high basal levels or be constitutive, depending on the culture method and its aeration. Thus, with the exception of the inflorescence infiltration methods used in Arabidopsis, the tissue culture stages associated with plant transformation will inevitably mean that any gene with a deleterious effect on cell growth may not yield viable/normal transgenic lines. By combining the GR ligand-binding domain at the C-terminus of AlcR, the transcription activator remains in the cytosol (even in the presence of ethanol) and thus cannot activate expression of the transgene. Upon dex addition, expression of the transgene is triggered. This not only obviates the transformation problem, but also enables the study of transgenes in cultured cells, increasing the applicability of alc to cell biology research. Finally, Bohner et al. (1999) have developed a dual control (positive and negative) system termed TGV, by combining elements of both the tet and GR systems. The regulatory protein is composed of the TetR DNA-binding domain, the GR ligand-binding domain and the V P16 activation domain; TGV can bind to a pA containing tet operator sequences. In the absence of dex and tetracycline, TGV will remain in the cytosol, and upon dex addition, it will translocate to the nucleus, bind to pA and activate transcription. The addition of the negative ligand tetracycline destabilises TGV binding, releasing it from the tet operator sites in pA. Different levels of expression were achieved with TGV by using different pA constructs; however, the higher levels of induced activity were associated with elevated background activity (Bohner and Gatz, 2001). Nevertheless, the ability to regulate the decline in activity following induction represents a novel contribution to the transgene expression tool kit. There is a strong argument for ‘artificial’ non-plant-derived transgene expression systems in terms of the minimal chance of inadvertent expression (low likelihood of initiation from the target promoter by plant transcription factors; little chance of effector ligands activating the introduced transcription factor) and minimal chance of interference of the introduced regulatory protein with host genes and less disruption to plant metabolism/development from the externally applied effector ligand. However, they are not without their drawbacks: dex and some GR constructs have proven to be toxic, and there appears to be induction of A. thaliana defence-related genes; some TetR-based systems have a significant basal level of expression; some plants contain phytooestrogens or phytoecdysteroids. As stated earlier, there is no ‘ideal’ conditional expression system, but the development of these, their modification and their combination are starting to make important contributions to functional genomic research.
INDUCIBLE TRANSGENE EXPRESSION
13.2.6
319
The conditional expression tool kit and its further development
The ideal conditional expression systems should have very low basal expression; high inducibility; high specificity with respect to the gene target and the inducer ligand; maximal expression levels that can be modulated in a dose-dependent manner; rapid response upon addition of the inducer; a rapid decline in expression following inducer withdrawal; inducers with zero toxicity that are not native to plants and have no physiological effect on the plant; environmentally safe inducers that can be used under field conditions; transgene components that are publicly acceptable for field use (e.g. see Zuo and Chua, 2000; Padidam, 2003; Tang et al., 2004). Clearly, no such system exists; many inducers are problematic in certain circumstances, and some chimeric regulatory proteins, containing e.g. mammalian or HSV sequences, may be unacceptable outside a controlled laboratory environment. The search for the perfect switch will continue, but there are some important considerations in relation to existing systems. Background expression is a key issue in conditional expression systems. One approach has been to prevent activation in the absence of the inducer by maintaining the regulatory protein and pA in separate lines and combining them by crossing (e.g. Moore et al., 1998), or through the use of site-specific recombinases to activate expression, e.g. the Cre/loxP recombination system from E. coli bacteriophage P1 (see Hare and Chua, 2002). However, both of these have the drawback that once crossing/excision has occurred, then the transgene is constitutively expressed, and reversal is not possible in those individuals. In many systems, background expression is not zero (under all circumstances). This can be because transcription from pA could be triggered by some endogenous process, or because the construction of a chimeric pA has a low constitutive activity, and/or because the level of expression of the regulatory protein is sufficiently high to initiate transcription in the absence of the inducer ligand. In a number of systems, pA has been designed to have strong transcription under inducing conditions, without regard to the effect on uninduced expression: the dual control TGV system demonstrated clearly that the stronger promoters had higher background activity (Bohner et al., 2001). In part, one of the problems is the utilisation of very simple constructs whereas native expression systems are often more complex, where the combinatorial effect of the different components leads to increased fidelity. For many applications, transcription from pA may not need to be so high. In the case of the alc expression system, the risk of induction from endogenous inducer has been partially addressed (Salter et al., 1998; Garoosi et al., 2005): under hypoxic conditions ethanol/acetaldehyde is produced, which induces the transgene. However, ethanol/acetaldehyde occurs in some tissues/organs when growing naturally, for example in the vascular cambium and xylem sap of trees (MacDonald and Kimmerer, 1991), in large tubers of 16-week-old potato plants (Junker et al., 2003) or in the fruits of some species (Pesis, 2005). Low-level inadvertent induction of alc can thus occur in some cell types arising from endogenous levels of these inducers. We also cannot exclude the possibility that any of these systems may produce the regulatory protein in excess to the extent that higher background levels may be observed. It is
320
REGULATION OF TRANSCRIPTION IN PLANTS
obvious from the literature that most chemically activated gene expression systems have been developed for maximal expression, rather than for optimal expression. In the basic design, almost all use a ‘high’ constitutive promoter (35S) to express the regulatory protein, but perhaps we now need to address whether other plant (rather than viral) constitutive promoters need to be tested to ensure that appropriate levels of regulatory protein are present. It is vital then that a range of constitutive promoters are available, or that synthetic promoters are designed to ensure appropriate levels of transcription (see Bhullar et al., 2003; Potenza et al., 2004). This may also avoid the silencing associated with the 35S promoter (Al-Kaff et al., 2000), and with the increasing requirement for the study of multiple genes simultaneously, it will be essential that all the expression signals (promoters, terminators, etc.) are not repeated between multiple transgene constructs. One approach to avoid high-level expression systemically throughout the plant has been the use of spatial/tissue-specific expression. For example, Mett et al. (1996) adapted the Cu-inducible Ace1 system to be expressed in an organ-specific manner in nodulated Lotus corniculatus plants, under the control of the nod45 promoter. Others have adapted alc to be expressed spatially, either in defined regions of meristems (Towers et al., 2003) or in a flower-specific manner (Maizel and Weigel, 2004). The plastome also offers advantages for the expression of certain proteins, and an ethanolinducible T7 RNA polymerase, targeted to plastids, was able to direct expression of transgenes under the control of T7 regulatory elements (L¨ossl, et al., 2005). Finally, many of the chemically inducible systems may suffer from side effects either by the application of the chemical inducer, or from inappropriate expression of the regulatory protein. While most reports specify whether any ‘visible’ symptoms occur as a result of induction/transgene activation, there clearly could be a myriad of invisible/subtle changes to the development or physiology of the plant as a result of the modification. To date the experimentation has been such that most researchers are measuring gross effects. As functional genomic applications increase, subtle but significant changes in metabolism, for example from one inducer, may be avoided by using a different inducer or a different expression system. There is little data on such effects but, as mentioned for alc above, ethanol application changed the levels of 13 of the 60 metabolites in potato tubers, while acetaldehyde altered only 3 of these (Junker et al., 2003). There is thus a need to undertake transcriptome analysis to determine the effect of inducers and/or over-expressed heterologous transcription factors to determine the nature of their effect on the host plant’s transcriptome, metabolome and physiology.
13.3
Applications to plant functional genomics
The high throughput genome analyses have revealed large numbers of genes requiring detailed experimentation, individually and in groups, to determine their function. Traditionally, such analysis has always relied upon the examination of the phenotypes of cells/organisms arising from loss-of-function or aberrant gain-of-function
INDUCIBLE TRANSGENE EXPRESSION
321
mutations. Conditional expression systems can provide such analyses, and their use is increasing. Here we give some examples of the various approaches adopted.
13.3.1
Conditional over-expression
Plants in which there is an over-expression of a transgene can develop novel phenotypes for a variety of reasons: for example it could activate downstream genes; there can be a direct, deleterious affect on the phenotype, perhaps leading to lethality at an early developmental stage; or, in the case of the expression of a heterologous gene, a gain-of-function phenotype can result through interference with a normal cellular mechanism. The over-expression of plant transcription factors has been reviewed previously (Zuo and Chua, 2000; Zhang, 2003), and can be used successfully to demonstrate the downstream target genes for these proteins. The major use of conditional expression to date has been to investigate genes whose function is deleterious to the plant when constitutively expressed. Tobacco plants expressing yeast invertase from a strong constitutive promoter generated only weakly expressing transgenic lines, because high expression was deleterious; whereas lines generated as alc-invertase were normal until ethanol addition (Caddick et al., 1998). More recently, by directing AlcR expression from a patatin promoter, Junker et al. (2004) demonstrated that in response to acetaldehyde induction, invertase expression could be localised within the tuber, and dissection of the complex changes in carbohydrate metabolism could be investigated. Using a similar approach, Deveaux et al. (2003) have used alc to investigate the effects of SHOOT MERISTEMLESS and CYCLIN D3;1 in defined subdomains of the shoot apical and floral meristems. Even if expression of the heterologous gene is not deleterious, normal plant function can be probed. Werner et al. (2003) used a heat-inducible bacterial ipt gene to elevate levels of cytokinin within Arabidopsis and identify novel cytokinin metabolites that are possibly involved in homeostatic mechanisms. Similarly, Koroleva et al. (2004) used an Antirrhinum majus CycD1;1 gene in Arabidopsis to demonstrate a divergence in D cyclin function between plants and animals.
13.3.2
Conditional knockout/knockdown
Insertional mutagenesis of the A. thaliana genome has been very successful, but many insertions do not yield a recognisable phenotype. This is in part due to functional redundancy, but it is compounded by our inability to recognise subtle phenotypic changes. Most phenotypic screens are relatively crude, analysing gross effects: hypothetically, a gene that is expressed only in a specific cell type in response to a specific cue may not reveal its phenotype unless a very detailed phenotypic/metabolomic analysis is undertaken throughout plant growth and development. For the analysis of some transcription factors for example, phenotypes have been observed only from multiple knockouts (see Zhang, 2003).
322
REGULATION OF TRANSCRIPTION IN PLANTS
Conditional knockout/knockdown has potential advantages, as subtle growth changes can be compared under isogenic conditions, in the same plant, at different times. Furthermore, such approaches have the potential to modulate the degree of knockdown by adjusting the concentration of the inducer, and timing of application. Several of the systems described above have been shown to be effective for RNAi on reporter genes: oestradiol-inducible Cre/loxP activated silencing of GFP and phytoene desaturase in Nicotiana benthamiana (Guo et al., 2003); heat-inducible RNAi of phytoene desaturase in Arabidopsis (Masclaux et al., 2004); alc triggered knockdown of GUS in tobacco (Lo et al., 2005); GVG controlled silencing of GFP in pine cells (Tang et al., 2005). Ethanol-inducible silencing of the chlorophyll biosynthetic genes magnesium-chelatase subunit 1 and glutamate 1-semialdehyde aminotransferase led to progressive loss of chlorophyll over 7–9 days, but the subsequent new leaves were normal (Chen et al., 2003). Furthermore, local silencing by application of ethanol to a single leaf left other parts of the plant unaffected. Ketelaar et al. (2004) have also demonstrated that alc-inducible RNAi of actin-interacting protein 1 leads to Arabidopsis plants with cells in the leaves, roots and shoots that fail to expand normally, and in the severest cases, the plants die. Guo et al. (2005) have also demonstrated microRNA activity in downregulating target RNA levels using the oestradiol system.
13.3.3
Conditional complementation
Since genes may have roles both early and late in the development of a plant, constitutive expression of a transgene, or mutation of the gene, may lead to the masking of some phenotypes (Zuo and Chua, 2000). One approach to mutant analysis is to complement the loss of function conditionally, thus revealing the temporal requirement for its gene product. In a study of ufo mutations in A. thaliana, Laufs et al. (2003) were able to use the alc system to express UFO at defined stages of flower development to dissect its function and determine aspects that had not previously been observed through mutation studies. With the wealth of developmental mutants in some species, there is clearly much potential for this approach.
13.4
Potential applications to plant biotechnology
Genetically modified crops have great potential to improve the growth range, quality and yield of both food and non-food products. Conditional expression is clearly essential for the functional genomics that will generate the transgenes for use in plant biotechnology. But will they ever be used in the field, and if so, to what end? Basically, conditional expression would be used to address a particular need for transgene expression only at one time and not another, and hence the technology would need to be stringent to avoid expression when it was not required. The requirement for conditional expression could be (i) to avoid toxicity/contamination to the plant, man or the environment; (ii) to optimise plant growth and productivity at
INDUCIBLE TRANSGENE EXPRESSION
323
different stages of development or in response to environmental conditions; or (iii) to manipulate metabolism for the enhancement of some products over others. One major criticism of GM crops has been the use of selectable markers as indicators of transformation, and hence there has been much interest in the prospect of marker excision. To this end, Cre/loxP excision systems have been developed that are either heat-shock inducible (Hoff et al., 2001; Liu et al., 2005) or oestradiolinduced via XVE (Zuo et al., 2001). Such technology is designed to be used in the development of the crop, prior to growth in the field. A second concern has been the ‘superweed’ argument – the introgression of herbicide or pesticide traits into native plants or weed communities around fields with a subsequent inability to control those weeds. While there have been few examples to date, chemically inducible herbicide resistance has been demonstrated by expressing a rat cytochrome P450 monoxygenase by benzothiadiazole activation of the tobacco PR1a promoter (Yamada et al., 2002). This approach would negate the superweed argument after transgene escape. Recently, engineered disease resistance has been reviewed in detail (Gurr and Rushton, 2005). The optimisation of plant growth or productivity can involve a number of strategies from plant protection against biotic or abiotic stress for growth in marginal environments, to modifying plant development to optimise yield. Salt tolerance, for example, could be achieved through increased production of osmoprotectants, but there may be advantage to produce these only when required. For protection against pathogens, there are clear possibilities through the use of wound/pathogen-inducible promoters. Breitler et al. (2004) used a wound-inducible promoter to direct cry gene expression in Bt transgenic rice, and showed that while it was as effective as constitutive expression, there were more external symptoms of attack due to the time lag to protection. Approaches such as these can lead to a more targeted approach to plant protection. Increased yield has been achieved through the development of shorter crop varieties in which lodging is reduced. The constitutive expression of the GAI protein in transgenic rice led to a series of detrimental effects, affecting flowering and seed production (Fu et al., 2001). However, conditional expression of GAI using the alc system in Arabidopsis enabled GAI expression to be temporally controlled thus avoiding the side effects (Ait-ali et al., 2003; Tomsett et al., 2004). The increased yield associated with F1 hybrid seed production has generated much interest in conditional male sterility, which also can suppress gene flow from biocrops to the wild. Of course, male sterility must be reversed to maintain the inbred parental lines to make F1 hybrids in the following season. Several crops exploit cytoplasmic male sterility (cms); for example, mutation of the maize MS45 gene results in the abortion of microspore development that can be restored through complementation with a functional MS45 gene. Unger et al. (2002) have demonstrated ecdysone-inducible restoration of male fertility in ms45 mutant maize. In a novel approach, cms has been engineered through the expression of β-ketothiolase in the chloroplast genome of tobacco (Ruiz and Daniell, 2005). In this study, restoration was achieved through growth in continuous light in a few flowers, but the
324
REGULATION OF TRANSCRIPTION IN PLANTS
combination of this with a chemically inducible RNAi to suppress β-ketothiolase expression would work even better. Plant metabolic engineering is complex and in its infancy. While there is much to do in the improvement of crop nutritional qualities, much of this is unlikely to need conditional promoters. However, the non-food uses, some of which will be toxic to the plant, or potentially hazardous in the environment, may profit from temporally controlled expression. Scheller and Conrad (2005) have recently reviewed the potential for production of silk and mammalian proteins, as well as for biodegradable plastic. While such synthesis can be targeted to organs or organelles to compartmentalise the product, it may be that temporal control may be required to ensure sufficient yield only after plant growth and development is complete.
13.5
Conclusions
We need to develop conditional gene expression systems further, not just to apply them to the analysis of single traits, but to manipulate multiple genes in an individual temporally, e.g. to understand development/metabolism both for research and for the development of novel crops. In the research stages at the very least, multiple and different regulation will be required – one or more genes to be over-expressed, while one or more others are downregulated. This will enable the flux of primary metabolism to be diverted to specific ends, for example. It will be essential therefore that each of the regulatory systems described above is developed further to allow their use simultaneously; that will mean promoters and other signals must avoid similar sequences to prevent silencing. Conditional expression systems will then make significant inroads on the wealth of data generated by ‘omic technologies.
References T. Ait-ali, C. Rands and N.P. Harberd (2003) Flexible control of plant architecture and yield via switchable expression of Arabidopsis gai. Plant Biotechnology Journal 1, 337–343. N.S. Al-Kaff, M.M. Kreike, S.N. Covey, R. Pitcher, A.M. Page and P.J. Dale (2000) Plants rendered herbicide-susceptible by cauliflower mosaic virus-elicited suppression of a 35S promoterregulated transgene. Nature Biotechnology 18, 995–999. T. Aoyama and N.-H. Chua (1997) A glucocorticoid-mediated transcriptional induction system in transgenic plants. The Plant Journal 11, 605–612. C. Baroux, R. Blanvillain, H. Betts, et al. (2005) Predictable activation of tissue-specific expression from a single gene locus using the pOp/LhG4 transactivation system in Arabidopsis. Plant Biotechnology Journal 3, 91–101. S. Bhullar, S. Chakravarthy, S. Advani, S. Datta, D. Pental and P.K. Burma (2003) Strategies for development of functionally equivalent promoters with minimum sequence homology for transgene expression in plants: cis-elements in a novel DNA context versus domain swapping. Plant Physiology 132, 988–998. N.M. Bierfreund, R. Reski and E.L. Decker (2003) Use of an inducible reporter gene system for the analysis of auxin distribution in the moss Physcomitrella patens. Plant Cell Reports 21, 1143–1152.
INDUCIBLE TRANSGENE EXPRESSION
325
S. Bohner and C. Gatz (2001) Characterisation of novel target promoters for the dexamethasoneinducible/tetracycline-repressible regulator TGV using luciferase and isopentenyl transferase as sensitive reporter genes. Molecular and General Genetics 264, 860–870. S. Bohner, I. Lenk, M. Rieping, M. Herold and C. Gatz (1999) Transcriptional activator TGV mediates dexamethasone-inducible and tetracycline-inactivatable gene expression. The Plant Journal 19, 87–95. J.C. Breitler, J.M. Vassal, M.D.M. Catala, et al. (2004) Bt rice harbouring cry genes controlled by a constitutive or wound-inducible promoter: protection and transgene expression under Mediterranean field conditions. Plant Biotechnology Journal 2, 417–430. W. Bruce, O. Folkerts, C. Garnaat, O. Crasta, B. Roth and B. Bowen (2000) Expression profiling of the maize flavonoid pathway genes controlled by estradiol-inducible transcription factors CRC and P. The Plant Cell 12, 65–79. M.X. Caddick, A.J. Greenland, I. Jepson, et al. (1998) An ethanol inducible gene switch for plants used to manipulate carbon metabolism. Nature Biotechnology 16, 177–180. J. Callis, M. Fromm and V. Walbot (1988) Heat inducible expression of a chimeric maize Hsp70cat gene in maize protoplasts. Plant Physiology 88, 965–968. S. Chen, D. Hofius, U. Sonnewald and F. B¨ornke (2003) Temporal and spatial control of gene silencing in transgenic plants by inducible expression of double-stranded RNA. The Plant Journal 36, 731– 740. J. Craft, M. Samalova, C. Baroux, et al. (2005) New pOp/LhG4 vectors for stringent glucocorticoiddependent transgene expression in Arabidopsis. The Plant Journal 41, 899–918. K.M. David and C. Perrot-Rechenmann (2001) Characterization of a tobacco bright yellow 2 cell line expressing the tetracycline repressor at a high level for strict regulation of transgene expression. Plant Physiology 125, 1548–1553. L. De Veylder, T. Beeckman, M. Van Montagu and D. Inz´e (2000) Increased leakiness of the tetracycline-inducible triple-Op promoter in dividing cells renders it unsuitable for high inducible levels of a dominant negative CDC2aAt gene. Journal of Experimental Botany 51, 1647–1653. Y. Deveaux, A. Peaucelle, G.R. Roberts, et al. (2003) The ethanol switch: a tool for tissue-specific gene induction during plant development. The Plant Journal 36, 918–930. M.C. Drew (1997) Oxygen deficiency and root metabolism: injury and acclimation under hypoxia and anoxia. Annual Review of Plant Physiology and Plant Molecular Biology 48, 223–250. X.D. Fu, D. Sudhakar, J. Peng, D.E. Richards, P. Christou and N.P. Harberd (2001) Expression of Arabidopsis GAI in transgenic rice represses multiple gibberellin responses. The Plant Cell 13, 1791–1802. G.A. Garoosi, M.G. Salter, M.X. Caddick and A.B. Tomsett (2005) Characterization of the ethanolinducible alc gene expression system in tomato. Journal of Experimental Botany 56, 1635–1642. C. Gatz (1996) Chemically inducible promoters in transgenic plants. Current Opinion in Biotechnology 7, 168–172. C. Gatz (1997) Chemical control of gene expression. Annual Review of Plant Physiology and Plant Molecular Biology 48, 89–108. C. Gatz and I. Lenk (1998) Promoters that respond to chemical inducers. Trends in Plant Science 3, 352–358. M. Gossen and H. Bujard (1992) Tight control of gene-expression in mammalian-cells by tetracyclineresponsive promoters. Proceedings of the National Academy of Sciences of the United States of America 89, 5547–5551. C.L. Granger and R.J. Cyr (2000) Microtubule reorganization in tobacco BY-2 cells stably expressing GFP-MBD. Planta 210, 502–509. C.L. Granger and R.J. Cyr (2001) Characterization of the yeast copper-inducible promoter system in Arabidopsis thaliana. Plant Cell Reports 20, 227–234. H.S. Guo, J.F. Fei, Q. Xie and N.-H. Chua (2003) A chemical-regulated inducible RNAi system in plants. The Plant Journal 34, 383–392. H.S. Guo, Q. Xie, J.-F. Fei and N.-H. Chua (2005) MicroRNA directs mRNA cleavage of the transcription factor NAC1 to downregulate auxin signals for Arabidopsis lateral root development. The Plant Cell 17, 1376–1386.
326
REGULATION OF TRANSCRIPTION IN PLANTS
S.J. Gurr and P.J. Rushton (2005) Engineering plants with increased disease resistance: how are we going to express it? Trends in Biotechnology 23, 283–290. P.D. Hare and N.-H. Chua (2002) Excision of selectable marker genes from transgenic plants. Nature Biotechnology 20, 575–580. T. Hoff, K.M. Schnorr and J. Mundy (2001) A recombinase-mediated transcriptional induction system in transgenic plants. Plant Molecular Biology 45, 41–49. B.H. Junker, C.C. Chu, U. Sonnewald, L. Wilmitzer and A.R. Fernie (2003) In plants the alc gene expression system responds more rapidly following induction with acetaldehyde than with ethanol. FEBS Letters 535, 136–140. B.H. Junker, R. Wuttke, A. Tiessen, et al. (2004) Temporally regulated expression of a yeast invertase in potato tubers allows dissection of the complex metabolic phenotype obtained following its constitutive expression. Plant Molecular Biology 56, 91–110. H.G. Kang, Y.W. Fang and A. Singh (1999) A glucocorticoid-inducible transcription system causes severe growth defects in Arabidopsis and induces defense-related genes. The Plant Journal 20, 127–133. T. Ketelaar, E.G. Allwood, R. Anthony, B. Voigt, D. Menzel and P.J. Hussey (2004) The actininteracting protein AIP1 is essential for actin organization and plant development. Current Biology 14, 145–149. O.A. Koroleva, M. Tomlinson, P. Parinyapong, et al. (2004) CycD1, a putative G1 cyclin from Antirrhinum majus, accelerates the cell cycle in cultured tobacco BY-2 cells by enhancing both G1/S entry and progression through S and G2 phases. The Plant Cell 16, 2364–2379. P. Laufs, E. Coen, J. Kronenberger, J. Traas and J. Doonan (2003) Separable roles of UFO during floral development revealed by conditional restoration of gene function. Development 130, 785– 796. H.K. Liu, C. Yang and Z.M. Wei (2005) Heat shock-regulated site-specific excision of extraneous DNA in transgenic plants. Plant Science 168, 997–1003. C. Lo, N. Wang and E. Lam (2005) Inducible double-stranded RNA expression activates reversible transcript turnover and stable translational suppression of a target gene in transgenic tobacco. FEBS Letters 579, 1498–1502. A. L¨ossl, K. Bohmert, H. Harloff, C. Eibl, S. M¨uhlbauer and H.-U. Koop (2005) Inducible transactivation of plastid transgenes: expression of the R. eutropha phb operon in transplastomic tobacco. Plant and Cell Physiology 46, 1462–1471. J. Love, G.C. Allen, C. Gatz and W.F. Thompson (2002) Differential Top10 promoter regulation by six tetracycline analogues in plant cells. Journal of Experimental Botany 53, 1871–1877. J. Love, A.C. Scott and W.F. Thompson (2000) Stringent control of transgene expression in Arabidopsis thaliana using the Top10 promoter system. The Plant Journal 21, 579–588. R.C. Macdonald and T.W. Kimmerer (1991) Ethanol in the stems of trees. Physiologia Plantarum 82, 582–588. M.J. Mackenzie, V. Mett, L.P. Lochhead and P.H.S. Reynolds (1998) Controlled cytokinin production in transgenic tobacco using a copper inducible promoter. Plant Physiology 116, 969– 977. A. Maizel and D. Weigel (2004) Temporally and spatially controlled induction of gene expression in Arabidopsis thaliana. The Plant Journal 38, 164–171. A. Martinez, C. Sparks, C.A. Hart, J. Thompson and I. Jepson (1999) Ecdysone agonist inducible transcription in transgenic tobacco plants. The Plant Journal 19, 97–106. F. Masclaux, M. Charpenteau, T. Takahashi, R. Pont-Lezica and J.-P. Galaud (2004) Gene silencing using a heat-inducible RNAi system in Arabidopsis. Biochemical and Biophysical Research Communications 321, 364–369. V.L. Mett, L.P. Lochhead and P.H.S. Reynolds (1993) Copper-controllable gene-expression system for whole plants. Proceedings of the National Academy of Sciences of the United States of America 90, 4567–4571. V.L. Mett, E. Podivinsky, A.M. Tennant, L.P. Lochhead, W.T. Jones and P.H.S. Reynolds (1996) A system for tissue-specific copper-controllable gene expression in transgenic plants: nodulespecific antisense of aspartate aminotransferase-P-2. Transgenic Research 5, 105–113.
INDUCIBLE TRANSGENE EXPRESSION
327
I. Moore, L. Galweiler, D. Grosskopf, J. Schell and K. Palme (1998) A transcription activation system for regulated gene expression in transgenic plants. Proceedings of the National Academy of Sciences of the United States of America 95, 376–381. P.B.F. Ouwerkerk, R.J. de Kam, J.H.C. Hodge and A.H. Meijer (2001) Glucocorticoid-inducible gene expression in rice. Planta 213, 370–378. M. Padidam (2003) Chemically regulated gene expression in plants. Current Opinion in Plant Biology 6, 169–177. M. Padidam, M. Gore, D.L. Lu and O. Smirnova (2003) Chemical-inducible, ecdysone receptor-based gene expression system for plants. Transgenic Research 12, 101–109. Y.L. Peng, J. Zhong, B. Liang, Y.L. Hu, Y. Gao and Z.P. Lin (2003) Improvement of copper-inducible gene expression system for plant. Acta Botanica Sinica 45, 1307–1311. M.R. Perera and M.G.K. Jones (2004) Expression of the peroxidase gene promoter (Shpx6b) from Stylosanthes humilis in transgenic plants during insect attack. Entomologia Experimentalis et Applicata 111, 165–171. E. Pesis (2005) The role of the anaerobic metabolites, acetaldehyde and ethanol, in fruit ripening, enhancement of fruit quality and fruit deterioration. Postharvest Biology and Technology 37, 1–19. C. Potenza, L. Aleman and C. Sengupta-Gopalan (2004) Targeting transgene expression in research, agricultural, and environmental applications: promoters used in plant transformation. In Vitro Cellular and Developmental Biology – Plant 40, 1–22. G.R. Roberts, G.A. Garoosi, O. Koroleva, et al. (2005) The alc-GR system. A modified alc gene switch designed for use in plant tissue culture. Plant Physiology 138, 1259–1267. H.A. Roslan, M.G. Salter, C.D. Wood, et al. (2001) Characterization of the ethanol-inducible alc gene-expression system in Arabidopsis thaliana. The Plant Journal 28, 225–235. O.N. Ruiz and H. Daniell (2005) Engineering cytoplasmic male sterility via the chloroplast genome by expression of beta-ketothiolase. Plant Physiology 138, 1232–1246. Q. Sa, Y. Wang, W. Li, L. Zhang and Y. Sun (2003) The promoter of an antifungal protein gene from Gastrodia elata confers tissue-specific and fungus-inducible expression patterns and responds to both salicylic acid and jasmonic acid. Plant Cell Reports 22, 79–84. E. Saez, M.C. Nelson, B. Eshelman, et al. (2000) Identification of ligands and coligands for the ecdysone- regulated gene switch. Proceedings of the National Academy of Sciences of the United States of America 97, 14512–14517. M.G. Salter, J.A. Paine, K.V. Riddell, et al. (1998) Characterisation of the ethanol-inducible alc gene expression system for transgenic plants. The Plant Journal 16, 127–132. M. Samalova, B. Brzobohaty and I. Moore (2005) pOp6/LhGR: a stringently regulated and highly responsive dexamethasone-inducible gene expression system for tobacco. The Plant Journal 41, 919–935. S. Schaarschmidt, N. Qu, D. Strack, U. Sonnewald and B. Hause (2004) Local induction of the alc gene switch in transgenic tobacco plants by acetaldehyde. Plant and Cell Physiology 45, 1566–1577. J. Scheller and U. Conrad (2005) Plant-based material, protein and biodegradable plastic. Current Opinion in Plant Biology 8, 188–196. M. Schena, A.M. Lloyd and R.W. Davis (1991) A steroid-inducible gene-expression system for plantcells. Proceedings of the National Academy of Sciences of the United States of America 88, 10421–10425. J.P. Sweetman, C.C. Chu, N. Qu, A.J. Greenland, U. Sonnewald and I. Jepson (2002) Ethanol vapor is an efficient inducer of the alc gene expression system in model and crop plant species. Plant Physiology 129, 943–948. W. Tang, K. Kinken and R.J. Newton (2005) Inducible antisense-mediated post-transcriptional gene silencing in transgenic pine cells using green fluorescent protein as a visual marker. Plant and Cell Physiology 46, 1255–1263. W. Tang, X.Y. Luo and V. Samuels (2004) Regulated gene expression with promoters responding to inducers. Plant Science 166, 827–834. W. Tang and R.J. Newton (2004) Glucocorticoid-inducible transgene expression in loblolly pine (Pinus taeda L.) cell suspension cultures. Plant Science 166, 1351–1358.
328
REGULATION OF TRANSCRIPTION IN PLANTS
B. Tomsett, A. Tregova, A. Garoosi and M. Caddick (2004) Ethanol-inducible gene expression: first step towards a new green revolution? Trends in Plant Science 9, 159–161. M.I. Towers, M. Ito, G. Roberts and J.H. Doonan (2003) Developmental control of the cell cycle. Cell Biology International 27, 283–285. A.K. Tyagi (2001) Plant genes and their expression. Current Science 80, 161–169. E. Unger, A.M. Cigan, M. Trimnell, et al. (2002) A chimeric ecdysone receptor facilitates methoxyfenozide-dependent restoration of male fertility in ms45 maize. Transgenic Research 11, 455–465. R.H. Wang, X.F. Zhou and X. Wang (2003) Chemically regulated expression systems and their applications in transgenic plants. Transgenic Research 12, 529–540. P. Weinmann, M. Gossen, W. Hillen, H. Bujard and C. Gatz (1994) A chimeric transactivator allows tetracycline-responsive gene-expression in whole plants. The Plant Journal 5, 559–569. T. Werner, J. Hanus, J. Holub, T. Schm¨ulling, H. van Onckelen and M. Strnad (2003) New cytokinin metabolites in IPT transgenic Arabidopsis thaliana plants. Physiologia Plantarum 118, 127–137. T. Yamada, Y. Ohashi, M. Ohshima, et al. (2002) Inducible cross-tolerance to herbicides in transgenic potato plants with the rat CYP1A1 gene. Theoretical and Applied Genetics 104, 308–314. J.Z. Zhang (2003) Overexpression analysis of plant transcription factors. Current Opinion in Plant Biology 6, 430–440. J.R. Zuo and N.-H. Chua (2000) Chemical-inducible systems for regulated expression of plant genes. Current Opinion in Biotechnology 11, 146–151. J.R. Zuo, Q.W. Niu and N.-H. Chua (2000) An estrogen receptor-based transactivator XVE mediates highly inducible gene expression in transgenic plants. The Plant Journal 24, 265–273. J.R. Zuo, Q.W. Niu, S.G. Moller and N.-H. Chua (2001) Chemical-regulated, site-specific DNA excision in transgenic plants. Nature Biotechnology 19, 157–161.
14 Modulation of transcriptional networks in crop plants Tong Zhu
14.1
Introduction
As a precisely controlled biological process, transcription converts the onedimensional information (nucleotide sequences) encoded by the genome into multidimensional quantitative information in the highly dynamic transcriptome, directing when, where, and how much a specific type of transcript should be transcribed, processed, and degraded. This process is the first key step to creating complex messages for plants and other organisms to control complex developmental programs, maintaining a proper metabolic flux of multiple pathways and interactions, and responding to various genetic and environmental stimuli and damages with speed and precision. Transcription regulates many biological processes and consequently affects traits, including many with economic values. Changing transcript levels of even a single gene could dramatically disrupt normal plant development, or alter a plant’s response to environmental stimuli. In fact, complete suppression or reduction of the abundance of a transcript type (i.e., knockout or knockdown, respectively) and the significant increase of the abundance of a transcript type (overexpression) have become routine methods to study gene functions (Zhang, 2003; Ostergaard and Yanofsky, 2004). The sensitivity to the transcription change exhibited by plants demonstrates not only the essential function of the transcription of life messages, but also its precise regulation in their life cycle and interaction with their environment. It also provides opportunities to change phenotype by genetic modification of existing transcription regulation. The genetic engineering of transcription regulation is a forward engineering process. It revises or builds new systems by introducing new features to the transcriptome systems or completely new pathways to the plants. This can be and has been one of the major paths toward improvement of crop performance and introduction of new functions (Dunwell, 2000; Gantet and Memelink, 2002; Gutterson and Zhang, 2004). In contrast to the modification of transcriptional regulation, the identification and prioritization of target genes to be manipulated in a genome are reverse engineering processes. Instead of building a system, they are concerned with raveling the operational principles underlying the existing system. This is crucial for the successful manipulation of the trait of interest, but it can also be a challenge. In general, many genes participate in the same transcriptional pathways. They play different
330
REGULATION OF TRANSCRIPTION IN PLANTS
roles, and the products of these genes may or may not be a rate-limiting factor in a metabolic or regulatory pathway. Furthermore, there are considerable crosstalks among different transcriptional regulatory pathways, demonstrating the existence of a complex transcription regulatory network (Harmer et al., 2000; Shinozaki et al., 2003; Chen and Zhu, 2004; Broun, 2005). The complexity of the transcriptional regulation implies that the simple up- or downregulated gene expression may be suitable to manipulate certain monogenic traits, but may not yield the desired phenotypes for complex traits. It also emphasizes the importance of understanding the gene interaction and regulation in selection and prioritization of target genes in trait improvement by modulation of gene transcription. Furthermore, unraveling a transcription regulatory network in crop species could enable the prioritization of the candidate genes according to their functions, and targeting the transcription of key genes, thus effectively and precisely introducing the expected effects and eliminating or minimizing the unexpected negative effects. In this chapter, basic methods for dissecting transcriptional regulatory networks will be introduced, and recent advances in this area will be discussed. The impact of understanding regulatory pathways in target gene selection for crop improvement will also be addressed.
14.2
Genes, regulatory factors, regulatory pathways, regulatory nodes, and regulatory networks
Although transcription regulations in plants are highly specific and believed to occur at the individual gene level, recent global surveys of plant transcriptome under normal growth and development and various environmental conditions revealed that no gene acts alone to perform biological functions. The complexity of the transcription regulations at the genome level can be demonstrated by the numbers of genes being induced by a single simple stimulus and their organized response. Numbers of transcriptome-profiling experiments illustrated a common theme of the transcription response. Any environmental stimuli, such as light, circadian, wound, pathogen, and abiotic stress, can induce transcription changes in hundreds to thousands genes (Harmer et al., 2000; Shinozaki et al., 2003; Chen and Zhu, 2004; Schnable et al., 2004; Eulgem, 2005). These genes can be generally categorized into two groups: genes encoding signaling molecules and genes encoding proteins functioning as effectors. The signaling genes include genes encoding protein kinases, transcription factors, and other signaling molecules. Their transcription is characterized by the fast response speed and short half-life. The effector genes encode functional proteins such as various enzymes. Their transcription in response to environmental or developmental stimuli is usually slow and steady (Harmer et al., 2000; Tepperman et al., 2001; Cheong et al., 2002; Fowler and Thomashow, 2002). It is generally accepted that transcription of genes in a signaling or a metabolic pathway is coordinately regulated to form a transcription pathway. However, it is less clear how different transcription pathways interact with each other. Characterization of these interactions has been proved challenging by the gene-by-gene
MODULATION OF TRANSCRIPTIONAL NETWORKS
331
approach. The systems-wide survey, such as transcriptome profiling, on the other hand, revealed such interactions or crosstalks. Genes previously associated with a particular established signal transduction pathway or a biochemical pathway are found to be induced or suppressed by stimuli activating other transcription pathways. For example, it was found that mechanical wounding significantly elevates transcript levels of genes involving wounding and other signals, including pathogen attack, abiotic stress factors, and plant hormones (Cheong et al., 2002). Therefore, different transcription regulatory pathways interact with each other, sometimes by sharing common regulatory components, to form the crosstalks among different pathways. The crosstalks between different regulatory pathways constitute transcription regulatory networks. The transcription networks enable the coordination of synthesis of molecules with differential functions in a biological process by regulating complex interactions among transcription pathways. They also enable processing biological functions by alternative paths. The transcription regulatory networks could be depicted as a graphic representation of the regulatory relationships reflecting the functional organization of the transcription in a cell. In such a diagram, the transcription factors are placed in the center as nodes, and the interactions between genes are depicted as connected lines called edges. The transcription factors and their peripherally located target genes form a regulatory module or motif. These genes either involve the same biological process, or belong to the same biochemical or regulatory pathways. When the genes in the regulatory module share similar expression pattern, they are considered as coexpressed, sometimes being hypothesized as coregulated (regulon). There are three major forms of the regulatory modules (Figure 14.1): single-input modules (a single transcription factor regulates a group of genes), feedforward loops (a single transcription factor regulates a target gene directly or indirectly through a second transcription factor), and bi-fans (both transcription factors regulate both target genes). These forms can occur independently or in combination in a regulatory network. Regulatory modules or motifs are the basic structural and functional units of the regulatory network. Each regulatory network could comprise multiple regulatory modules (Chen and Zhu, 2004).
14.3
Functional characterization of transcription factor genes
Characterization of the functions of the transcription factors is essential to interpret the biological meaning of the transcription regulatory network. Toward this aspect, genome-wide analyses of spatial and temporal expression patterns of transcription factor genes were used to imply putative functions to specific members of transcription factor gene families (Chen et al., 2002; Czechowski et al., 2004). The openreading frames of all transcription factor genes were cloned to systematically analyze their functions in Arabidopsis (Gong et al., 2004). Large-scale reverse genetics studies using a variety of transposon- or tDNA-tagged mutant population were also used to identify the association between the disrupted transcription factor genes and morphological characteristics (Ostergaard and Yanofsky, 2004). CBF2/DREB1C,
332
REGULATION OF TRANSCRIPTION IN PLANTS
Figure 14.1 A diagram showing the three major types of regulatory modules. Transcription factors (TF) are indicated by oval circles; genes encoding transcription factors or other products are indicated by the boxes. The interaction between the transcription factors and promoters of genes is illustrated by the solid arrows, and feedback regulation of the transcription factors by the factors themselves is shown as dashed arrows.
a gene with its product functioning as a negative regulator of CBF1/DREB1B and CBF3/DREB1A (Novillo et al., 2004), ABORTED MICROSPORES (AMS), a gene that encodes a MYC class transcription factor functioning in postmeiotic transcriptional regulation of microspore development (Sorensen et al., 2003), and DAG1 and DAG2, both of which encode Dof zinc-finger proteins controlling seed germination (Gualberti et al., 2002), are among many transcription factor genes being identified and characterized by this approach. However, in some cases, the systematic
MODULATION OF TRANSCRIPTIONAL NETWORKS
333
disruption could become a challenge, as many insertion lines lack obvious morphological phenotypic differences (Meissner et al., 1999; Bouche and Bouchez, 2001), partially due to the existence of high functional redundancy among the transcription factor gene family members (Riechmann and Ratcliffe, 2000). The high-throughput analysis of plants containing overexpressed transcription factor genes (Zhang, 2003) and the prediction of functional redundancy by phylogeny reconstructions (Riechmann et al., 2000) have been used to help to combat the obstacles.
14.4
Reverse engineering of transcription regulatory network
The reverse engineering of a transcription regulatory network is the process used to establish the understanding of the necessary components. These components make up the parts list (a list of genes or proteins that participate in the network); the characteristics of the parts (when and where the genes are expressed); the organization of the parts (pathways and processes that known genes participated in); and dynamics and functions of the organized parts. Genome-wide expression analyses are one of the most useful approaches to predict and characterize these regulatory modules. These analyses identify most if not all of the genes in a network and roughly group them together according to their temporal and spatial expression patterns, and the functional nature of the encoded protein products. A number of clustering algorithms are developed to classify the genes in a system – hierarchical clustering, self-organizing map, or principal component analysis, just to name a few (Clarke and Zhu, 2006). The input used for the clustering analysis varies, from gene expression values alone to composite functional indices that consider average expression of all genes known to fall into a certain functional category based on existing knowledge resources such as the Gene Ontology (Cheng et al., 2004; Fang et al., 2006). In many cases the composite functional indices result in clusters that better reflect the structure of the data. The expression patterns detected from the genome-wide expression analyses, regulatory sequences, and ontology-based functional predictions form clusters of putative regulatory linkage groups (regulons). It is expected that the coexpressed genes share common DNA-binding sites if they are coregulated; thus an overrepresented cis-regulatory motif could be an indication of the existence of regulatory linkage. A general strategy for predicting the regulatory linkage (edge) within a regulatory group is to quantify the association of a known DNA-binding motif with coexpressed genes. This can be done by comparing the regulatory sequences of coexpressed genes with the regulatory sequences of randomly selected genes from the same genome. The occurrence of a consensus sequence within a prescribed distance from 5 to the start of transcription is counted and compared between the two groups. If the consensus sequence is statistically significantly overrepresented in the coexpressed genes than those random genes, then consensus sequences could be predicted as potentially functional binding motifs that mediate the gene regulation (Harmer et al., 2000; Chen et al., 2002; Zhu et al., 2003). Several algorithms based on this string-based motif prediction have been
334
REGULATION OF TRANSCRIPTION IN PLANTS
developed to avoid the influence from the plant genome compositions and take the degenerate bases into consideration (Hudson and Quail., 2003; Kreps et al., 2003). The availability of expression data of the model plants and several crop species, such as barley, rice, and maize (Table 14.1), and the regulomics databases and tools (Lescot et al., 2002; Davuluri et al., 2003; Shahmuradov et al., 2003; Steffens et al., 2004; Guo et al., 2005; O’Connor et al., 2005) further empower this approach by identifying coexpressed target genes and their regulators in diverse genetic and environmental conditions. However, this simple model considers only the transcription regulation mediated by one regulatory motif bound by one transcription factor. It discounts the cooperative and competitive interactions between transcription factors and cofactors and their effects in the binding affinity and specificity. Furthermore, this model is difficult to apply to identify novel transcription factors with unknown binding consensus sequences (Chen and Zhu, 2004). Novel algorithms using more realistic models for transcription regulation by considering multiple binding sites of different strength and accounting for cooperative and competitive interactions have been further developed (Rombauts et al., 2003; Beer and Tavazoie, 2004). One such algorithm predicts the probability that a transcription factor binds within a promoter region by examining the free energy of binding for all possible bases at each position in a binding site, with or without considering the effect of cooperative and competitive interactions (Liu and Clarke, 2002; Granek and Clarke, 2005). An additional method for regulatory motif prediction is using phylogenetic footprinting and shadowing among different species (Wasserman and Sandelin, 2004). This is based on the fact that cis-regulatory elements are frequently conserved (Wray et al., 2003), and its success is recently further demonstrated in mammals (Boffelli et al., 2003) and plants (Hong et al., 2003). Two aspects of the predicted regulatory linkage groups require functional validation. First, the connectivity and relations between transcription factor genes and target genes should be examined experimentally. To identify the downstream target genes, the transcription of transcription factor genes could be disrupted, or induced by ectopic expression, and the transcription of the suspected target genes from the same cluster as the transcription factor genes could be examined. Gene disruption mutant collections provided useful resources for examining the linkage between regulatory factors and target genes in a number of plant species, including Arabidopsis, rice, and maize. Systematic overexpression of transcription factor genes in Arabidopsis overcomes difficulties due to the functional redundancy among transcription factor genes, and provides valuable additional resources (Zhang, 2003). Transcriptome profiling of such mutants and transgenic plants have led to the identification of target genes of APETALA3 and PISTILLATA, class B floral homeotic genes (Zik and Irish, 2003), of ripening-inhibitor (RIN) and nonripening (NOR) transcription factors (Moore et al., 2002), and target genes of transcription regulators for fiber differentiation (Ehlting et al., 2005). These results not only establish the linkage between the transcription factor gene and its downstream genes, but also characterize the type of the interaction (activation or repression).
Multiple species including plants, public depository Multiple species including plants, public depository Multiple species including plants, public depository Multiple plant species Arabidopsis, public depository Arabidopsis Arabidopsis Barley Soybean Maize Rice Tomato Multiple plant species, stress related
Gene Expression Omnibus
PLEXdb TAIR Gene Expression NASCArrays Botany Array Resource BarleyBase SGMD ZEAMAGE RED TED OSMID
SMD
ArrayExpress
Species
Major public microarray databases that archive plant data
Database
Table 14.1
Yazaki et al. (2002) Fei et al. (2003)
Shen et al. (2005) Rhee et al. (2003) Craigon et al. (2004) Toufighi et al. (2005) Shen et al. (2005) Alkharouf and Matthews (2004)
Ball et al. (2005)
Parkinson et al. (2005)
Barrett et al. (2005)
Reference
http://www.plexdb.org/ http://www.arabidopsis.org/info/expression/index.jsp http://affymetrix.arabidopsis.info/ http://bbc.botany.utoronto.ca www.barleybase.org http://psi081.ba.ars.usda.gov/SGMD/default.htm. http://www.maizearray.org/ http://red.dna.affrc.go.jp/RED/ http://ted.bti.cornell.edu/ http://www.osmid.org/
http://genome-www5.stanford.edu/
http://www.ebi.ac.uk/arrayexpress/
http://www.ncbi.nlm.nih.gov/geo/
URL
336
REGULATION OF TRANSCRIPTION IN PLANTS
Second, the physical association of transcription factors with corresponding promoter sequences should be further examined in order to support the establishment of the regulatory linkages. Yeast one-hybrid experiments are useful to detect protein– DNA interactions and to isolate new proteins that bind to a specific regulatory element (Luo et al., 1996). The technical advancements in combination of chromatin immunoprecipitation (ChIP) with microarray analysis enable the genome-wide survey of the transcription factor–DNA interaction (Ren et al., 2000). ChIP is a method that allows one to identify the specific locations in the genome where proteins of interest are bound within living cells. When combined with microarrays with tiled probes representing collected promoter sequences or whole genomes, it is possible to identify target sites of transcriptional regulatory proteins and specific features of chromatin, on a genome-wide basis in an unbiased manner. These methodologies have provided new insight into the mechanisms by which proteins find their correct target sites in living cells, and have played a very important role in elucidating the transcriptional circuitry that underlies cellular differentiation and the response to environmental cues. Using this approach, genes regulated by the transcription factors functioning in flower development and light signaling are identified (Zik and Irish, 2002; Gao et al., 2004). However, currently, the technical challenge and high cost associated with this approach limit the application for studying target genes for all transcription factors of a plant genome in a systematic way. This technical hurdle is especially true while using the whole-genome tiling arrays (Mockler et al., 2005). In order to elucidate the remote indirect interaction among genes (‘trans’ effect) in the regulatory network, the synergy between global gene expression studies and genetic linkage analysis has been used to identify expression quantitative trait loci (eQTL) (Bystrykh et al., 2005; Li and Burmeister, 2005). Since mRNA levels for many genes are heritable quantitative phenotypes, the genetic variants that influence gene expression can be characterized (Chen et al., 2005; Kliebenstein et al., 2006). A recent microarray analysis of Arabidopsis shoot development in 30 recombinant inbred lines derived from a cross of Col and Ler ecotypes produced quantitative measurements of the phenotypes. Using these data, five major QTLs with significant marker-by-gene linkages were identified. The most significant eQTLs, particularly those overlapped with shoot regeneration QTLs, showed genetic regulation in ‘cis’ because of variants in the gene’s own regulatory regions, while additional less significant eQTLs showed ‘ trans’ effect, i.e., by loci elsewhere in the genome (Decook et al., 2006). Although this methodology is still in development, there is no doubt that it will shed light on the genetic basis of the regulatory networks, and on the process that transforms genetic blueprints into cellular functions. Reconstruction of a genome-wide complex regulatory network from large-scale data sets depends on the development of computational algorithms. Several algorithms have been developed to utilize the results from analyzing single data type from literature, transcription profiling data, and yeast two-hybrid data or combined heterologous data types. The most commonly used is combinatorial analysis of promoter elements and transcriptome data. It establishes links between transcription factors that bind to the cis-elements and genes containing the combination of cis-elements based on the common expression patterns emerged from microarray
MODULATION OF TRANSCRIPTIONAL NETWORKS
337
data (Pilpel et al., 2001). Module network is another algorithm that first partitions the genes into different regulatory modules based on the expression pattern revealed by large-scale transcriptome analysis. It then further analyzes the relations among the genes and their regulators within each regulatory module based on a probabilistic relational model (Segal et al., 2003). This algorithm can be used to obtain in-depth information about the regulatory modules that are most relevant to the trait of interest. Similar to the module network algorithm, the genetic regulatory module algorithm also establishes the regulatory network, based on the identification of regulatory modules. In addition, it uses the global binding data from the ChIP microarray analyses to connect transcription factors with target genes with different expression patterns (Bar-Joseph et al., 2003). Finally, the Bayesian network model converts the quantitative data into a probabilistic framework, and thus is able to combine heterologous data types to predict an interaction (Troyanskaya et al., 2003). Although computational inference of molecular interaction could provide guidance for targeted experimental validation, and overcome some time-consuming and difficult steps faced by the experimental approaches, such as identification of all the in vivo interactions, and characterization of the kinetic parameters associated with in vivo biochemical, biomechanical, or electrophysiological interactions (Bolouri and Davidson, 2002), it is important to recognize that the interactions suggested by the computational prediction and large-scale in vitro or in vivo functional analysis may not be accurate and may not correlate. In a recent comparative analysis of the results obtained from yeast two-hybrid system, in silico prediction based on sequence motif and transcriptome profiling, it was found that no single approach produced reliable results as the results from these approaches frequently do not relate to each other and overlap to each other (von Mering et al., 2002). It demonstrated the importance of combining the evidence from two or three methods in improving the accuracy and coverage of the predicted interactions.
14.5
Transcription network in crop plants
To date, most of the studies on reverse engineering of plant transcriptional regulatory network were conducted in Arabidopsis. The translation of the genomic information and knowledge obtained from a model species into agriculturally relevant plants for improvement could be achieved by (1) discovering leads from the regulatory networks of model systems that correspond to orthologs in crops, if the regulatory network modules are evolutionarily conserved, and (2) discovering leads and reconstructing the regulatory network directly in crops, using an experimental approach, if the regulatory network module is not conserved (Clarke and Zhu, 2006). Comparative transcriptomics studies could be used to experimentally determine the overall similarity and differences of regulatory networks between different species. A recent study revealed that rice transcriptome consists of many unique transcripts compared to that of Arabidopsis. Microarray analysis confirmed that a large number of predicted genes without significant Arabidopsis homologs are in fact expressed. Among the best matched homologous genes of certain functional groups between the two species, the spatial and temporal expression pattern could
338
REGULATION OF TRANSCRIPTION IN PLANTS
differ significantly (Ma et al., 2005). These results illustrate the variations in degree of conservation among different regulatory modules between the two species and signal the potential pitfalls in translation of certain regulatory relations from model species to distant crops. When the regulatory modules are functionally conserved, the comparative analysis of the orthologs and their positions in the network between closely related species or genotypes within a species can determine the nature of the transcriptional flux: if the transcriptional flux is operated by similar genes with different timing in those species or genotypes, then the difference between them represents a mechanistic solution; if the transcriptional flux is operated by different genes with similar timing, then the difference represents a functional solution (Clarke and Zhu, 2006). In many crop species, the regulatory network can be revealed directly using the available functional genomic tools and data. Custom-designed and commercial microarrays have become increasingly available for crop species (Zhu, 2003) to construct large data sets for this purpose. For species having less functional genomics tools, microarrays of a closely related model species could be used to experimentally explore the regulatory networks in crops by heterologous detection (i.e., cross-species probe-target hybridization) (Zhu et al., 2001). Because the degree of sequence conservation is not always consistent among the genes between the model and applied species, this approach is considered a prescreening method. The relatively low-signal specificity during cross-species hybridizations makes distinguishing the expression of a specific gene from a sequence-conserved gene family a challenge task. Nevertheless, the results could provide valuable reference data based on not only sequence similarity but also expression pattern for regulatory network comparison. The existence of common regulatory modules for many traits is supported by the discovery of shared common key transcription factors among phylogenic distant species. Because these transcription factors represent the key nodes in the transcription regulatory network, their commonality could imply the similar regulatory modules among different species. Cold acclimation is a phenomenon referring to the increased freezing tolerance in response to low, nonfreezing temperatures in plants. It has been shown that cold acclimation is transcriptionally controlled by the CCAAT-binding factor (CBF)/dehydration-responsive element-binding (DREB) transcriptional activators and the CBF-targeted genes (CBF regulon) in Arabidopsis (Thomashow, 2001; Shinozaki et al., 2003). CBF-like proteins have been identified in many other species including Brassica, wheat, rye, tomato (Jaglo et al., 2001), barley (Choi et al., 2002), rice (Dubouzet et al., 2003), pepper (Hong and Kim, 2005), maize (Qin et al., 2004), and soybean (Li et al., 2005). Among them, tomatoes are freezing sensitive, and do not cold acclimate. Nevertheless, the CBF-like proteins showed conserved amino acid sequences that are unique among other AP2/EREBP protein family members (Jaglo et al., 2001). The transcripts of these genes and the homologs of the Arabidopsis CBF target genes can be induced by low-temperature exposure. Moreover, the constitutive overexpression of the Arabidopsis CBF genes has shown the induction of the expression of orthologs of Arabidopsis CBF-targeted genes and
MODULATION OF TRANSCRIPTIONAL NETWORKS
339
increases the freezing tolerance of both nonacclimated and cold-acclimated plants. Similarly, overexpression of genes encoding CBF-like proteins in Arabidopsis has resulted in improved cold and salt stress tolerance in many species (Zhang et al., 2004a). These results suggest that the components of the CBF cold-response pathway are highly conserved in flowering plants and not limited to those that cold acclimate. The identification of the common key regulatory nodes is essential for the detection of similar regulatory network modules, and is the basis for enabling a generic strategy to manipulate the traits of interest across a variety of crops. However, the target genes within the regulon or regulatory network module and their interaction with the transcription factors could vary among species; thus, precaution should be taken when applying such an approach. Slight difference in the regulatory elements could lead to dramatic changes in phenotype (Frary et al., 2000). A recent study illustrated that the key transcription factor genes could activate different set of the target genes in different species. The nonacclimated plants such as tomato contain functional CBF transcription activators, but the tomato CBF activates a different set of target genes compared to the cold-acclimated plants, such as Arabidopsis (Zhang et al., 2004b). In another case, the key regulator LeMADS-RIN (ripen-inhibitor), a MADS transcription factor that is found expressed in nonclimacteric fruit plants, is also present in tomato, a species of climacteric fruits whose ripening is initiated by ethylene (Moore et al., 2002). These examples demonstrate the difference between the conserved regulatory network key nodes and the conserved regulatory function, and postulate a fine map of the regulatory network modules.
14.6
Transcription factors as gene switches for genetic engineering
Because of the essential positions transcription factors occupied in a regulatory network, they have been targeted for developing gene switches for activation and repression of an entire transcription module by genetic engineering (Lam and Meisel, 1999; Memelink et al., 2001; Broun, 2004). There are several advantages to forwardengineer the transcription regulatory network via manipulating the transcription of transcription factor genes. (1) It is possible to regulate multiple pathways by manipulating the expression of a transcription factor at the high level of hierarchy of the pathway and central node. (2) Changing transcript levels of genes encoding transcription factors could have better fitness during natural selection. It was found that genes encoding transcription factors that are transiently transcribed are significantly enriched in Escherichia coli regulatory network motifs and hubs (Wang and Purisima, 2005). In Arabidopsis, it is also shown that the genes encoding transcription factors are targeted for selection (Chen et al., 2005). Because these genes have short transcript half-lives, they can alter their transcript levels quickly, to facilitate network adaptation to the rapid environmental changes. Therefore, the node could represent an excellent target for changing the network motif and further altering the phenotypes. (3) Transcription of transcription factor genes is flexible to manipulate through breeding. By this approach significantly fewer genes will need to be introgressed.
340
REGULATION OF TRANSCRIPTION IN PLANTS
Of the 1500 transcription factor genes identified in the Arabidopsis genome (Riechmann et al., 2000), the most extensively studied stress-related transcription factors are probably members of the CBF/DREB transcription factors family (Thomashow, 2001; Shinozaki et al., 2003). The transcriptions of CBF1 and CBF3 are induced by various abiotic stresses, including drought, cold, and salinity, but their products also activate the transcription of their direct target genes (COR for cold regulated, named by the original observation) (Jaglo-Ottosen et al., 1998) by binding to the CCGAC motif in the DRE/CRT cis-acting elements. Overexpression of one of these transcription factor genes under the control of various promoters could improve the stress tolerance in Arabidopsis and other plants (Zhang et al., 2004a). These results imply that the elevated transcript levels of the transcription factor genes are translated to the increased functional transcription factors. Genetic manipulation of the expression of the transcription factor genes requires precision to minimize the interaction between the transgene expression and the remainder expression of the existing transcription regulatory network in the host plants. This can be exemplified by the transgenic expression of the CBF1. Constitutive overexpression of CBF1 using 35S promoter can lead to undesirable agronomic performance and yield drag (Liu et al., 1998; Gilmour et al., 2000), partially due to the interference with endogenous regulatory pathways. However, when CBF1 is expressed under the control of the stress-inducible promoter rd29A, mimicking the expression regulation of native copies, the transgenic plants exhibit less yield drag under normal development and better survival when exposed to salt, freezing, and drought, than do the transgenic plants controlled by constitutive promoter (Kasuga et al., 1999; Lee et al., 2003). The modulation of the expression of transcription factors can also be achieved by engineering transcription factors such as zinc-finger DNA-binding proteins. The unique features of plant zinc-finger protein in DNA binding and sequence recognition enable more specific regulation. Using this strategy, a number of genes, including those encoding transcription factors, are regulated in a heritable fashion (Guan et al., 2002). Post-transcription regulation has been used to reduce the level of endogenous transcripts. RNAi is RNA interference by short RNAs. Using a chimeric hairpin RNA construct, all members of the multigene family encoding codeinone reductase are silenced through RNA interference, resulting in accumulation of nonnarcotic alkaloid reticuline in the transgenic opium poppy (Allen et al., 2004). Since it does not affect transcription of other genes in the same pathway, it demonstrated the potential utility of the RNAi in modifying the transcript level of the transcription factor genes in nonmodel crop plants.
14.7
Perspectives
Introduction of foreign genes into plants have been a major route to improve plant performance in the past decades (Dunwell, 2000). Since this approach can introduce independent transcription pathways, it has been used to engineer complete novel functions that do not exist in the plant host, such as novel agronomic traits
MODULATION OF TRANSCRIPTIONAL NETWORKS
341
(Ye et al., 2000; Paine et al., 2005), and producing biopharmaceuticals such as mammalian antibodies, vaccines, hormones, and antibodies (Ma et al., 2005), or other novel proteins (Scheller and Conrad, 2005). This approach could be complemented by targeted regulation of transcription of the endogenous genes, especially those with regulatory functions, such as genes encoding transcription factors to improve agronomic performance, especially complex traits such as yield and drought tolerance. Modification of transcription of the genes encoding functional proteins such as enzymes has been successfully used in metabolic engineering strategies to overcome single or few rate-limiting enzymatic steps in one pathway. The flux through an already existing pathway in the host plant can be increased or decreased and can result in a modified accumulation of particular metabolites of interest. For example, overexpression of phosphoethanolamine N -methyltransferase, a rate-limiting enzyme of the plant choline synthesis pathway, significantly increases the production of osmoprotectant glycine betaine by 30-fold, and leads to improved resistance to osmotic stresses and improved nutritional value (McNeil et al., 2001). However, in some cases, identification of those rate-limiting steps could be unenviable. Change of expression of transcription factor genes, on the other hand, could result in much broader effects and does not require the identification of the ratelimiting steps. For example, overexpression of maize LEAF COLOR (LC), a gene encoding MYC-type transcription factor, could result in increased accumulation of anthocyanin and other classes of flavonoids in various organs and tissues in tomato (Bovy et al., 2002). These examples demonstrated a general nonspecific strategy to coordinately upregulate or activate multiple existing host pathways. Recent efforts in unrevealing the transcriptional regulatory network have identified the key regulators and their target genes, and provided many candidates for improvement of a number of complex traits. However, to be able to predict the outcome of the modification of a transcriptional regulatory network continues to be a challenge: (1) The dynamic nature of the regulatory network and additional interconnected layers ranging from signal transduction pathways involving kinase cascades to transcription control and post-transcriptional regulation of RNA pools, e.g., by microRNAs, to regulatory units that regulate the expression of physical gene clusters at the chromosomal level (Chen and Zhu, 2004) remain to be understood. It is now known that the transgenes originated from other organisms have limited interaction with the host transcriptome (El Ouakfaoui and Miki, 2005). Such studies should be extended to characterize the interaction between the transgenic regulation and host regulatory network, when the transgene has a native copy. (2) The criteria of an optimal target for such genetic manipulations need to be defined and applied. Among the regulatory genes in the regulatory network that were related to the traits of interest, selecting one to maximize the positive impact and minimize the deleterious effects on plant performance is not an easy task. In addition to the genetic and epigenetic effects, such as the inserted transgene being tight linked to the undesired native variation, or being silenced by methylation, it is possible that many intermediate steps in these series of events directly influence other layers of gene regulation often linking simultaneously to more than one layer. The tight interconnection of signaling, transcription control, and epigenetic changes becomes evident in many
342
REGULATION OF TRANSCRIPTION IN PLANTS
examples related to regulatory networks. It is essential to include all layers of gene regulation, as functional gene–gene dependencies and feedback loops often involve different levels including protein and DNA modification as well as transcriptional changes. Only when the plant transcription network and its interaction with foreign genetic components are well characterized, the genetic improvements of crops through transcription enhancement can have the precision and effectiveness. This will be an essential task of the rapidly emerging field of systems biotechnology.
References N.W. Alkharouf and B.F. Matthews (2004) SGMD: the Soybean Genomics and Microarray Database. Nucleic Acids Research 32, D398–D400. R.S. Allen, A.G. Millgate, J.A. Chitty, et al. (2004) RNAi-mediated replacement of morphine with the nonnarcotic alkaloid reticuline in opium poppy. Nature Biotechnology 22, 1559–1566. C.A. Ball, I.A.B. Ihab Awad, J. Demeter et al. (2005) The Stanford Microarray Database accommodates additional microarray platforms and data formats. Nucleic Acids Research 33, D580–D582. Z. Bar-Joseph, G.K. Gerber, T.I. Lee, et al. (2003) Computational discovery of gene modules and regulatory networks. Nature Biotechnology 21, 1337–1342. T. Barrett, T.O. Suzek, D.B. Troup, et al. (2005) NCBI GEO: mining millions of expression profiles – database and tools. Nucleic Acids Research 33, D562–D566. M.A. Beer and S. Tavazoie (2004) Predicting gene expression from sequence. Cell 117, 185–198. D. Boffelli, J. McAuliffe, D. Ovcharenko, et al. (2003) Phylogenetic shadowing of primate sequences to find functional regions of the human genome. Science 299, 1391–1394. H. Bolouri and E. Davidson (2002) Modeling transcription regulatory networks. Bioessays 24, 1118– 1129. N. Bouche and D. Bouchez (2001) Arabidopsis gene knockout: phenotypes wanted. Current Opinion in Plant Biology 4, 111–117. A. Bovy, R. de Vos, M. Kemper, et al. (2002) High-flavonol tomatoes resulting from the heterologous expression of the maize transcription factor genes LC and C1. The Plant Cell 14, 2509–2526. P. Broun (2004) Transcription factors as tools for metabolic engineering in plants. Current Opinion in Plant Biology 7, 202–209. P. Broun (2005) Transcriptional control of flavonoid biosynthesis: a complex network of conserved regulators involved in multiple aspects of differentiation in Arabidopsis. Current Opinion in Plant Biology 8, 272–279. L. Bystrykh, E. Weersing, B. Dontje, et al. (2005) Uncovering regulatory pathways that affect hematopoietic stem cell function using ‘genetical genomics’. Nature Genetics 37, 225–232. W. Chen, N.J. Provart, J. Glazebrook, et al. (2002) Expression profile matrix of Arabidopsis transcription factor genes suggests their putative functions in response to environmental stresses. The Plant Cell 14, 559–574. W.J. Chen, S.H. Chang, M.E. Hudson, et al. (2005) Contribution of transcriptional regulation to natural variations in Arabidopsis. Genome Biology 6, R32. W.J. Chen and T. Zhu (2004) Networks of transcription factors with roles in environmental stress response. Trends in Plant Science 9, 591–596. J. Cheng, M. Cline, J. Martin, et al. (2004) A knowledge-based clustering algorithm driven by Gene Ontology. Journal of Biopharmaceutical Statistics 14, 687–700. Y.H. Cheong, H.S. Chang, R. Gupta, X. Wang, T. Zhu and S. Luan (2002) Transcriptional profiling reveals novel interactions between wounding, pathogen, abiotic stress, and hormonal responses in Arabidopsis. Plant Physiology 129, 661–677. D.W. Choi, E.M. Rodriguez and T.J. Close (2002) Barley Cbf3 gene identification, expression pattern, and map location. Plant Physiology 129, 1781–1787.
MODULATION OF TRANSCRIPTIONAL NETWORKS
343
J.D. Clarke and T. Zhu (2006) Microarray analysis of the transcriptome as a stepping stone towards understanding biological systems: practical considerations and perspectives. The Plant Journal 45, 630–650. D.J. Craigon, N. james, J. Okyere, J. Higgins, J. Jotham, S. May (2004) NASCArrays: a repository for microarray data generated by NASC’s transcriptomics service. Nucleic Acids Research 32, D575–D577. T. Czechowski, R.P. Bari, M. Stitt, W.R. Scheible and M.K. Udvardi (2004) Real-time RT-PCR profiling of over 1400 Arabidopsis transcription factors: unprecedented sensitivity reveals novel root- and shoot-specific genes. The Plant Journal 38, 366–379. R.V. Davuluri, H. Sun, S.K. Palaniswamy, et al. (2003) AGRIS: Arabidopsis gene regulatory information server, an information resource of Arabidopsis cis-regulatory elements and transcription factors. BMC Bioinformatics 4, 25. R. Decook, S. Lall, D. Nettleton and S.H. Howell (2006) Genetic regulation of gene expression during shoot development in Arabidopsis. Genetics 172, 1155–1164. J.G. Dubouzet, Y. Sakuma, Y. Ito, et al. (2003) OsDREB genes in rice, Oryza sativa L., encode transcription activators that function in drought-, high-salt- and cold-responsive gene expression. The Plant Journal 33 (4), 751–763. J.M. Dunwell (2000) Transgenic approaches to crop improvement. Journal of Experimental Botany 51, 487–496. J. Ehlting, N. Mattheus, D.S. Aeschliman, et al. (2005) Global transcript profiling of primary stems from Arabidopsis thaliana identifies candidate genes for missing links in lignin biosynthesis and transcriptional regulators of fiber differentiation. The Plant Journal 42, 618–640. S. El Ouakfaoui and B. Miki (2005) The stability of the Arabidopsis transcriptome in transgenic plants expressing the marker genes nptII and uidA. The Plant Journal 41, 791–800. T. Eulgem (2005) Regulation of the Arabidopsis defense transcriptome. Trends in Plant Science 10, 71–78. Z. Fang, J. Yang, Y. Li, Q. Luo and L. Liu (2006) Knowledge guided analysis of microarray data. Journal of Biomedical Information 39, 401–411. Z. Fei, X. Tang, R. Alba and J. Giovannoni (2003) Tomato Expression Database (TED) – an interactive management tool for tomato expression profiling data. Proceedings of the Computational Systems Bioinformatics 3, 424–425. S. Fowler and M.F. Thomashow (2002) Arabidopsis transcriptome profiling indicates that multiple regulatory pathways are activated during cold acclimation in addition to the CBF cold response pathway. The Plant Cell 14, 1675–1690. A. Frary, T.C. Nesbitt, S. Grandillo, et al. (2000) fw2.2: a quantitative trait locus key to the evolution of tomato fruit size. Science 289, 85–88. P. Gantet and J. Memelink (2002) Transcription factors: tools to engineer the production of pharmacologically active plant metabolites. Trends in Pharmacological Sciences 23, 563–569. Y. Gao, J. Li and E. Strickland (2004) An Arabidopsis promoter microarray and its initial usage in the identification of HY5 binding targets in vitro. Plant Molecular Biology 54, 683–699. S.J. Gilmour, A.M. Sebolt, M.P. Salazar, J.D. Everard and M.F. Thomashow (2000) Overexpression of the Arabidopsis CBF3 transcriptional activator mimics multiple biochemical changes associated with cold acclimation. Plant Physiology 124, 1854–1865. W. Gong, Y.P. Shen, L.G. Ma, et al. (2004) Genome-wide ORFeome cloning and analysis of Arabidopsis transcription factor genes. Plant Physiology 135, 773–782. J.A. Granek and N.D. Clarke (2005) Explicit equilibrium modeling of transcription-factor binding and gene regulation. Genome Biology 6, R87. G. Gualberti, M. Papi, L. Bellucci, et al. (2002) Mutations in the Dof zinc finger genes DAG2 and DAG1 influence with opposite effects the germination of Arabidopsis seeds. The Plant Cell 14, 1253–1263. X. Guan, J. Stege, M. Kim, et al. (2002) Heritable endogenous gene regulation in plants with designed polydactyl zinc finger transcription factors. Proceedings of the National Academy of Sciences of the United States of America 99, 13296–13301.
344
REGULATION OF TRANSCRIPTION IN PLANTS
A. Guo, K. He, D. Liu, et al. (2005) DATF: a database of Arabidopsis transcription factors. Bioinformatics 21, 2568–2569. N. Gutterson and J.Z. Zhang (2004) Genomics applications to biotech traits: a revolution in progress? Current Opinion in Plant Biology 7, 226–230. S.L. Harmer, J.B. Hogenesch, M. Straume, et al. (2000) Orchestrated transcription of key pathways in Arabidopsis by the circadian clock. Science 290, 2110–2113. J.P. Hong and W.T. Kim (2005) Isolation and functional characterization of the Ca-DREBLP1 gene encoding a dehydration-responsive element binding-factor-like protein 1 in hot pepper (Capsicum annuum L. cv. Pukang). Planta 220, 875–888. R.L. Hong, L. Hamaguchi, M.A. Busch and D. Weigel (2003) Regulatory elements of the floral homeotic gene AGAMOUS identified by phylogenetic footprinting and shadowing. The Plant Cell 15, 1296–1309. M.E. Hudson and P.H. Quail (2003) Identification of promoter motifs involved in the network of phytochrome a-regulated gene expression by combined analysis of genomic sequence and microarray data. Plant Physiology 133, 1605–1616. K.R. Jaglo, S. Kleff, K.L. Amundsen, et al. (2001) Components of the Arabidopsis C-repeat/ dehydration-responsive element binding factor cold-response pathway are conserved in Brassica napus and other plant species. Plant Physiology 127, 910–917. K.R. Jaglo-Ottosen, S.J. Gilmour, D.G. Zarka, O. Schabenberger and M.F. Thomashow (1998) Arabidopsis CBF1 overexpression induces COR genes and enhances freezing tolerance. Science 280, 104–106. M. Kasuga, Q. Liu, S. Miura, K. Yamaguchi-Shinozaki and K. Shinozaki (1999) Improving plant drought, salt, and freezing tolerance by gene transfer of a single stress-inducible transcription factor. Nature Biotechnology 17, 287–291. D.J. Kliebenstein, M.A. West, H. van Leeuwen, et al. (2006) Genomic survey of gene expression diversity in Arabidopsis thaliana. Genetics 172, 1179–1189. J. Kreps, P. Budworth, S. Goff and R. Wang (2003) Identification of putative plant cold responsive regulatory elements by gene expression profiling and a pattern enumeration algorithm. Plant Biotechnology Journal 1, 345–352. E. Lam and L. Meisel (1999) Gene switch and stress management: modulation of gene expression by transcription factors. In: Plant Responses to Environmental Stresses: From Phytohormones to Genome Organization (ed. H.R. Lerner), pp. 51–70. Marcel Dekker, New York. J.T. Lee, V. Prasad, P.T. Yang, et al. (2003) Expression of Arabidopsis CBF1 regulated by an ABA/stress promoter in transgenic tomato confers stress tolerance without affecting yield. Plant, Cell and Environment 26, 1181–1190. M. Lescot, P. Dehais, G. Thijs, et al. (2002) PlantCARE, a database of plant cis-acting regulatory elements and a portal to tools for in silico analysis of promoter sequences. Nucleic Acids Research 30, 325–327. J. Li and M. Burmeister (2005) Genetical genomics: combining genetics with gene expression analysis. Human Molecular Genetics 14, R163–R169. X.P. Li, A.G. Tian, G.Z. Luo, Z.Z. Gong, J.S. Zhang and S.Y. Chen (2005) Soybean DRE-binding transcription factors that are responsive to abiotic stresses. Theoretical and Applied Genetics 110, 1355–1362. Q. Liu, M. Kasuga, Y. Sakuma, et al. (1998) Two transcription factors, DREB1 and DREB2, with an EREBP/AP2 DNA binding domain separate two cellular signal transduction pathways in droughtand low-temperature-responsive gene expression, respectively, in Arabidopsis. The Plant Cell 10, 1391–1406. X. Liu and N.D. Clarke (2002) Rationalization of gene regulation by a eukaryotic transcription factor: calculation of regulatory region occupancy from predicted binding affinities. Journal of Molecular Biology 323, 1–8. Y. Luo, S. Vijaychander, J. Stile and L. Zhu (1996) Cloning and analysis of DNA-binding proteins by yeast one-hybrid and one-two-hybrid systems. Biotechniques 20, 564–568. J.K. Ma, E. Barros, R. Bock, et al. (2005) European union framework 6 pharma-planta consortium. Molecular farming for new drugs and vaccines. Current perspectives on the production of pharmaceuticals in transgenic plants. EMBO Reports 6, 593–599.
MODULATION OF TRANSCRIPTIONAL NETWORKS
345
L. Ma, C. Chen, X. Liu, et al. (2005) A microarray analysis of the rice transcriptome and its comparison to Arabidopsis. Genome Research 15, 1274–1283. S.D. McNeil, M.L. Nuccio, M.J. Ziemak and A.D. Hanson (2001) Enhanced synthesis of choline and glycine betaine in transgenic tobacco plants that overexpress phosphoethanolamine N -methyltransferase. Proceedings of the National Academy of Sciences of the United States of America 98, 10001–10005. R.C. Meissner, H. Jin, E. Cominelli, et al. (1999) Function search in a large transcription factor gene family in Arabidopsis: assessing the potential of reverse genetics to identify insertional mutations in R2R3 MYB genes. The Plant Cell 11, 1827–1840. J. Memelink, J.W. Kijne, R. van der Heijden and R. Verpoorte (2001) Genetic modification of plant secondary metabolite pathways using transcriptional regulators. Advances in Biochemical Engineering/Biotechnology 72, 103–125. T.C. Mockler, S. Chan, A. Sundaresan, H. Chen, S.E. Jacobsen and J.R. Ecker (2005) Applications of DNA tiling arrays for whole-genome analysis. Genomics 85, 1–15. S. Moore, J. Vrebalov, P. Payton and J. Giovannoni (2002) Use of genomics tools to isolate key ripening genes and analyse fruit maturation in tomato. Journal of Experimental Botany 53, 2023– 2030. F. Novillo, J.M. Alonso, J.R. Ecker and J. Salinas (2004) CBF2/DREB1C is a negative regulator of CBF1/DREB1B and CBF3/DREB1A expression and plays a central role in stress tolerance in Arabidopsis. Proceedings of the National Academy of Sciences of the United States of America 101, 3985–3990. T.R. O’Connor, C. Dyreson and J.J. Wyrick (2005) Athena: a resource for rapid visualization and systematic analysis of Arabidopsis promoter sequences. Bioinformatics 21, 4411–4413. L. Ostergaard and M.F. Yanofsky (2004) Establishing gene function by mutagenesis in Arabidopsis thaliana. The Plant Journal 39, 682–696. J.A. Paine, C.A. Shipton, S. Chaggar, et al. (2005) Improving the nutritional value of golden rice through increased pro-vitamin A content. Nature Biotechnology 23, 482–487. H. Parkinson, U. Sarkans, M. Shojatalab, et al. (2005) ArrayExpress – a public repository for microarray gene expression data at the EBI. Nucleic Acids Research 33, D553–D555. Y. Pilpel, P. Sudarsanam and G.M. Church (2001) Identifying regulatory networks by combinatorial analysis of promoter elements. Nature Genetics 29, 153–159. F. Qin, Y. Sakuma, J. Li, et al. (2004) Cloning and functional analysis of a novel DREB1/CBF transcription factor involved in cold-responsive gene expression in Zea mays L. Plant Cell Physiology 45, 1042–1052. B. Ren, F. Robert, J.J. Wyrick, et al. (2000) Genome-wide location and function of DNA binding proteins. Science 290, 2306–2309. S.Y. Rhee, BW. Beavis, T.Z. Berardini, et al. (2003) The Arabidopsis Information Resource (TAIR): a model organism database providing a centralized, curated gateway to Arabidopsis biology, research materials and community. Nucleic Acids Research 31, 224–228. J.L. Riechmann, J. Heard, G. Martin, et al. (2000) Arabidopsis transcription factors: genome-wide comparative analysis among eukaryotes. Science 290, 2105–2110. J.L. Riechmann and O.J. Ratcliffe (2000) A genomic perspective on plant transcription factors. Current Opinion in Plant Biology 3, 423–434. S. Rombauts, K. Florquin, M. Lescot, K. Marchal, P. Rouze and Y. van de Peer (2003) Computational approaches to identify promoters and cis-regulatory elements in plant genomes. Plant Physiology 132, 1162–1176. J. Scheller and U. Conrad (2005) Plant-based material, protein and biodegradable plastic. Current Opinion in Plant Biology 8, 188–196. P.S. Schnable, F. Hochholdinger and M. Nakazono (2004) Global expression profiling applied to plant development. Current Opinion in Plant Biology 7, 50–56. E. Segal, M. Shapira, A. Regev, et al. (2003) Module networks: identifying regulatory modules and their condition-specific regulators from gene expression data. Nature Genetics 34, 166– 176. I.A. Shahmuradov, A.J. Gammerman, J.M. Hancock, P.M. Bramley and V.V. Solovyev (2003) PlantProm: a database of plant promoter sequences. Nucleic Acids Research 31, 114–117.
346
REGULATION OF TRANSCRIPTION IN PLANTS
L. Shen, J. Gong, R.A. Caldo, et al. (2005) BarleyBase – an expression profiling database for plant genomics. Nucleic Acids Research 33, D614–D618. K. Shinozaki, K. Yamaguchi-Shinozaki and M. Seki (2003) Regulatory network of gene expression in the drought and cold stress responses. Current Opinion in Plant Biology 6, 410–417. A.M. Sorensen, S. Krober, U.S. Unte, P. Huijser, K. Dekker and H. Saedler (2003) The Arabidopsis ABORTED MICROSPORES (AMS) gene encodes a MYC class transcription factor. The Plant Journal 33, 413–423. N.O. Steffens, C. Galuschka, M. Schindler, L. Bulow and R. Hehl (2004) AthaMap: an online resource for in silico transcription factor binding sites in the Arabidopsis thaliana genome. Nucleic Acids Research 32, D368–D372. J.M. Tepperman, T. Zhu, H.S. Chang, X. Wang and P.H. Quail (2001) Multiple transcription-factor genes are early targets of phytochrome a signaling. Proceedings of the National Academy of Sciences of the United States of America 98, 9437–9442. M.F. Thomashow (2001) So what’s new in the field of plant cold acclimation? Lots! Plant Physiology 125, 89–93. K. Toufighi, S.M. Brady, R. Austin, E. Ly and N.J. Provart (2005) The Botany Array Resource: e-Northerns, Expression Angling, and promoter analyses. The Plant Journal 43,153–163. O.G. Troyanskaya, K. Dolinski, A.B. Owen, R.B. Altman and D. Botstein (2003) A Bayesian framework for combining heterogeneous data sources for gene function prediction (in Saccharomyces cerevisiae). Proceedings of the National Academy of Sciences of the United States of America 100, 8348–8353. C. von Mering, R. Krause, B. Snel, et al. (2002) Comparative assessment of large-scale data sets of protein–protein interactions. Nature 417, 399–403. E. Wang and E. Purisima (2005) Network motifs are enriched with transcription factors whose transcripts have short half-lives. Trends in Genetics 21, 492–495. W.W. Wasserman and A. Sandelin (2004) Applied bioinformatics for the identification of regulatory elements. Natural Revolution Genetics 5, 276–287. G.A. Wray (2003) Transcriptional regulation and the evolution of development. The International Journal of Developmental Biology 47, 675–684. J. Yazaki, N. Kishimoto, M. Ishikawa and S. Kikuchi (2002) The Rice Expression Database (RED): gateway to rice functional genomics. Trends in Plant Science, 7, 563–564. X. Ye, S. Al-Babili, A. Kloti, et al. (2000) Engineering the provitamin A (beta-carotene) biosynthetic pathway into (carotenoid-free) rice endosperm. Science 287, 303–305. J.Z. Zhang (2003) Overexpression analysis of plant transcription factors. Current Opinion in Plant Biology 6, 430–440. J.Z. Zhang, R.A. Creelman and J.K. Zhu (2004a) From laboratory to field. Using information from Arabidopsis to engineer salt, cold, and drought tolerance in crops. Plant Physiology 135, 615–621. X. Zhang, S.G. Fowler, H. Cheng, et al. (2004b) Freezing-sensitive tomato has a functional CBF cold response pathway, but a CBF regulon that differs from that of freezing-tolerant Arabidopsis. The Plant Journal 39, 905–919. T. Zhu (2003) Global analysis of gene expression using GeneChip microarrays. Current Opinion in Plant Biology 6, 418–425. T. Zhu, P. Budworth, W. Chen, et al. (2003) Transcriptional control of nutrient partitioning during rice grain filling. Plant Biotechnology Journal 1, 59–70. T. Zhu, H.-S. Chang, J. Schmeits, et al. (2001) Gene expression microarrays: improvements and applications towards agricultural gene discovery. Journal of Association Laboratory Automat 6, 95–98. M. Zik and V.F. Irish (2003) Global identification of target genes regulated by PETALA3 and PISTILLATA floral homeotic gene action. The Plant Cell 15, 207–222.
Index Note: genes and mutants are italicized. A AAG box, 205 ABC model, 253 ABCDE model, 256, 261 acetylation, 57, 63, 69, 278 activation domain, 28 activator sequence-1-like (as-1) element, 268 ADP-ribosylation, 82 AGAMOUS (AG), 39, 42, 123, 255 AlcR, 315, 318, 321 AP2/ERF, 30, 31, 34, 270, 338 APETALA (AP), 255, 334 ARID, 34 AT-hook motif, 55, 58, 96, 114 avirulence gene (avr gene), 267 B bHLH, 30, 31, 35, 274 bimolecular fluorescence complementation (BiFC), 262 BRAHMA (BRM), 117 BRE, 2 bromodomain (BRD), 10, 84, 86, 87, 88, 90, 114 bromodomain extra-terminal protein, 89 bZIP, 31, 71, 241, 270, 295, 298 C CAAT box, 6 CArG box, 256, 258 CBF/DREB, 295, 297, 300, 331, 338 CCAAT box, 6, 7 CDK8/Srb8-11 module, 16 CenH3, 122, 125 chromatin, 42, 54, 58, 65, 112, 278 chromatin assembly factor 1 (CAF-1), 127, 225 chromatin immunoprecipitation (ChIP), 9, 84, 86, 123, 278, 336
chromatin remodeling complex (CRC), 89, 113, 239 chromocenter, 82, 122 chromodomain, 99, 114, 119 CHROMOMETHYLASE3 (CMT3), 120 cis-element, 4, 28, 40, 68, 162, 268, 334, 336 CK2, 57, 63,65,71,169, 173, 203 cold stress, 292 cold-regulated (COR) gene, 295, 298, 300, 340 combinatorial control, 68 conditional gene expression, 309, 319, 321 CONSTANS (CO), 117, 236, 237, 238, 240, 243 copper-controlled gene expression, 314 core histone, 79 core promoter, 4 cosuppression, 136, 138 cotranscriptional processing, 171 C-terminal domain (CTD), 15, 16, 172 D database of Arabidopsis transcription factors (DATF), 30 DDM1, 114, 121, 123 defense response, 266, 270 DEFICIENS (DEF), 258 DELLA, 242 DNA bending, 64 DNA binding, 28, 57, 64, 116, 184, 205, 227, 256, 312 DNA binding domain, 28, 34, 58 DNA demethylation, 102 DNA methylation, 102, 120, 175 Dof, 32, 35, 71, 236, 332 DRD1, 124 DRM1/2, 120
348
INDEX
E ecdysone-controlled gene expression, 316 epigenetic modification, 101, 102, 123 enhancer, 4, 68, 82, 137 ethanol-induced gene expression, 315 ethylene-response factor (ERF), 270 euchromatin, 82, 92, 95 expressed sequence tag (EST), 34 expression quantitative trait locus (eQTL), 336 external transcribed spacer (ETS), 163 E(z)-type, 96 D downstream promoter element (DPE), 4 F FACT, 127 FCA, 228 feedforward loop, 330, 331 FLC (FLOWERING LOCUS C), 37, 42, 95, 120, 127, 225, 228, 229, 231 FLD (FLOWERING LOCUS D), 225, 226 floral homeotic mutants, 255 fluorescence resonance energy transfer (FRET), 261 FRI, 240 FT, 117, 235, 238, 240, 243fluorescent in situ hybridisation (FISH), 122 G general transcription factor (GTF), 1, 12, 19 gene silencing, 101, 137, 138, 148, 165 genomic imprinting, 96 GLOBOSA (GLO), 258 globular domain of H1 (GH1), 55, 129 glucocorticoid-controlled gene expression, 317 GNAT/MYST family, 86 GRAS, 32, 35 H H1 linker histone, 55, 58, 66, 112, 128, 129 HB, 31, 35 HD2 family, 92 head module, 16 heat stress, 285, 290 heat shock protein (HSP), 285, 289, 300, 313
heat shock transcription factor (HSF), 285, 288, 300 heterochromatin, 82, 93, 99, 122, 139, 240 Heterochromatin Protein 1 (HP1), 96, 98, 100, 234 high mobility group (HMG) protein, 54, 88 histone acetylation, 81, 100 histone acetyltransferase (HAT), 10, 83, 85, 88, 100, 116, 127, 169, 173 histone code hypothesis, 84, 112 histone deacetylase (HDAC), 83, 90, 100, 119, 127, 175, 225, 276 histone demethylase, 99, 226 histone H2A, 79, 81 H2AF/Z, 125, 126, 229 H2A.X, 125, 127 H2B, 79 H3, 79, 81 H3.3, 125, 126, 227 H4, 79, 81 histone methylation, 93, 100, 175, 231 histone methyltransferase (HMT), 93, 101, 128, 227, 232 histone replacement, 112 histone variant, 112 HMGA, 55, 67 HMGB, 54, 59, 67HMG-box domain, 58, 64, 168 homeodomain, 34 I initiator region (Inr), 4, 7 intergenic spacer (IGS), 162, 166, 174 internal transcribed spacer (ITS), 162 J jasmonic acid (JA), 268, 274, 277 jumonji domain, 227 K Kryptonite (KYP), 98, 123 L Lac repressor, 316 LEAFY (LFY), 30, 39, 42 leucine zipper, 117 Like Heterochromatin Protein 1 (LHP1), 98, 234, 240 loop domain, 140 LSD1 (lysine-specific histone demethylase 1), 99, 226 LSH, 121, 124
INDEX
M MADS, 30, 31, 34, 35, 37, 42, 235, 253, 256, 260, 262, 339 MAF1-5, 37, 231 MAP kinase, 275, 278, 294 matrix attachment region (MAR), 67, 140, 142, 150 mediator, 11 MET1, 120, 122, 129 methylation, 57, 63, 82, 93, 94 methyl CpG binding (MBD) protein, 124 microarray analysis, 11, 34, 39, 86, 267, 270, 336 micro RNA (miRNA), 242, 341 middle complex, 15 module, 4 multi-input module, 330, 331 MYB, 30, 31, 35, 295, 298 N NEP promoter, 190, 196 NIN, 32, 35 nuclear encoded plastid RNA polymerase (NEP), 186, 187, 207 nuclear export signal (NES), 286 nuclear factor Y (NFY), 6 nuclear localization signal (NLS), 286 nuclear scaffold, 82 nucleoid, 184 nucleolar dominance, 82, 92, 173 nucleolus, 163 nucleolus organizer region (NOR), 162, 174 nucleoprotein structure, 68 nucleosome, 11, 54, 66, 79, 80, 112 nucleosome positioning, 82 O oestrogen-controlled gene expression, 317 P p300/CBP family, 87 Paf1C, 231 pathogen-associated molecular pattern (PAMP), 266, 277 PEP promoter, 189, 196 PGT box, 205 phosphorylation, 57, 65, 82, 278 PIE1(PHOTOPERIOD-INDEPENDENT YEARLY FLOWERING 1), 120, 127, 228
349
PICKLE (PKL), 119 PISTILLATA (PI), 255, 334 plant chromatin database, 60, 62, 114 plant homeodomain (PHD), 87, 95, 114, 233, 239 plastid encoded plastid RNA polymerase (PEP), 186, 188, 207 plastidial promoter, 189 plastid transcription, 185, 192, 202 Polycomb-group (PcG), 96, 232, 243 position effect, 136, 137, 148 post-transcriptional gene silencing (PTGS), 138, 142, 151, 153 preinitiation complex (PIC), 1, 18 promoter, 1, 40, 68, 82, 86, 137, 172, 258, 268, 310 pytopathogen, 266 Q quantitative trait locus (QTL), 336 R reactive oxygen intermediate (ROI), 276 repression domain, 28 resistance gene (R gene), 267, 272 retinoblastoma (Rb) protein, 227 RNA-dependent RNA polymerase (RDRP), 139 RNA-induced silencing complex (RISC), 138 RNA-induced transcriptional gene silencing complex (RITS), 139, 150 RNA interference (RNAi), 123, 138, 322, 324, 340 RNA polymerase I, 162, 166, 169 RNA polymerase II, 1, 12, 18, 93, 229 RNA polymerase III, 18 RNA polymerase IV, 175 RPD3/HDA1 family, 90 S salicylic acid (SA), 267, 275, 277, 313 SANT domain, 114, 117, 127, 230 scaffold attachment region (SAR), 267, 270, 275 SEPALLATA (SEP), 257 SET domain, 95, 97 SIR2 family, 92 single-input module, 331, 332 siRNA, 121, 125, 138, 175
350
INDEX
SNF2, 113, 230 SNF5, 119 SOC1, 117, 235, 237, 242 SPLAYED (SYD), 117 SQUAMOSA (SQUA), 258 STRUWWELPETER (SWP), 15 SUMOylation, 82, 230, 278 Su(var)3-9 type, 97 SWI/SNF, 113, 115 systemic acquired resistance (SAR), 267 σ -factor, 185, 196, 198 T tail module, 12 TAFII 250 family, 87 tandem affinity purification (TAP), 262 TATA, 2, 7, 18, 167, 189 TATA-binding protein (TBP), 2, 18, 93, 114, 166, 167 TATA-less promoter, 8 TBP-associated factor (TAF), 4, 8, 9, 166 TetR, 316, 318 TFP response element (BRE), 2 TGA factor, 270, 275 TIF-IA/RRN3, 166 TIF-IB/SL1, 166, 173 TIF-IC, 166 transactivator, 1 transcriptome, 290, 329, 330 transcriptional network, 278, 329, 333, 337 transcription factor, 1, 28, 68, 267, 270, 278, 286, 311, 330 transcription factor IIB (TFIIB), 2, 18
transcription factor IID (TFIID), 6, 8, 12 transcription initiation site (TIS), 163, 166 transcription start sequence motif (TSS motif), 6, 7 transcriptional gene silencing (TGS), 139, 142, 150 transcriptional repressor, 227 transcriptional silencing, 96 transgene expression, 136, 142, 152 Tritorax-group (TrxG), 97 U ubiquitination, 82, 231, 278 upstream activating sequence (UAS), 2 upstream binding factor (UBF), 166, 168, 173 upstream control element (UCE), 166 V vernalization, 232 vernalization independence (VIP) genes, 231 W WRKY, 31, 34, 35, 37, 268, 270, 273, 275, 290 WUSCHEL (WUS), 42, 117 Y yeast two-hybrid analysis, 11 Z zinc finger, 29, 30, 92, 96, 117, 232, 274, 298, 340
Plate 1 Multicopy transgene loci can form visible heterochromatic knobs. Interphase (A and B) and early prophase (C and D) nuclei from transgenic Arabidopsis were analyzed by DAPI (4 -6-diamidino-2phenylindole) staining to visualize total DNA (panels A and C) and by fluorescence in situ hybridization to visualize centromeric repeats (pAL1, lavender, panel B) and hygromycin phosphotransferase DNA from the hygromycin phosphotransferase transgene (HPT, green, panels B and D). The transgenic HPT locus forms an independent heterochromatic knob (arrows in A and C) that localizes apart from the centromeric heterochromatin. (This plate is modified from Figure 1 in Probst et al. (2003) and reproduced with the permission from Blackwell Scientific Publishing.)
Plate 2 MARs increase transgene expression. NT1 tobacco suspension culture cells were cotransformed by microprojectile bombardment with a reporter and selectable marker plasmids as described by Allen et al. (1993, 1996). Independent hygromycin-resistant cell lines were picked and assayed for GUS (b-glucuronidase) by histochemical assay. The top panel shows 58 individual cell lines transformed with a MAR-flanked CaMV 35S-GUS reporter plasmid. The bottom panel shows 60 individual cell lines transformed with a control CaMV 35S-GUS reporter plasmid. The intensity of blue staining indicates the level of GUS expression in each cell line.