Inducible Gene Expression in Plants

Citation preview

00 Inducible Gene prelims

3/9/99 11:59 AM

Page i

INDUCIBLE GENE EXPRESSION IN PLANTS

00 Inducible Gene prelims

3/9/99 11:59 AM

Page ii

00 Inducible Gene prelims

3/9/99 11:59 AM

Page iii

Inducible Gene Expression in Plants Edited by

P.H.S. Reynolds Ministry of Research, Science and Technology PO Box 5336 Wellington New Zealand Formerly at: The Horticulture and Food Research Institute of New Zealand Palmerston North New Zealand

CABI Publishing

00 Inducible Gene prelims

3/9/99 11:59 AM

Page iv

CABI Publishing is a division of CAB International CABI Publishing CAB International Wallingford Oxon OX10 8DE UK Tel: +44 (0)1491 832111 Fax: +44 (0)1491 833508 Email: [email protected]

CABI Publishing 10 E 40th Street Suite 3203 New York, NY 10016 USA Tel: +1 212 481 7018 Fax: +1 212 686 7993 Email: [email protected]

© CAB International 1999. All rights reserved. No part of this publication may be reproduced in any form or by any means, electronically, mechanically, by photocopying, recording or otherwise, without the prior permission of the copyright owners. A catalogue record for this book is available from the British Library, London, UK. Library of Congress Cataloging-in-Publication Data Inducible gene expression in plants / edited by P.H.S. Reynolds. p. cm. Includes bibliographical references and index. ISBN 0–85199–259–5 (alk. paper) 1. Plant genetic regulation. 2. Plant gene expression. I. Reynolds, P. H. S. (Paul H. S.) QK981.4.I558 1998 572.89652––dc21

ISBN 0 85199 259 5 Typeset in 10/12pt Photina by Columns Design Ltd, Reading. Printed and bound in the UK at the University Press, Cambridge.

98–25843 CIP

00 Inducible Gene prelims

3/9/99 11:59 AM

Page v

Contents

Contributors 1 Inducible Control of Gene Expression: an Overview P.H.S. Reynolds 2 Use of the TN10-encoded Tetracycline Repressor to Control Gene Expression C. Gatz

vii 1

11

3 Ecdysteroid Agonist-inducible Control of Gene Expression in Plants A. Martinez and I. Jepson

23

4 Glucocorticoid-inducible Gene Expression in Plants T. Aoyama

43

5 Tissue-specific, Copper-controllable Gene Expression in Plants V.L. Mett and P.H.S. Reynolds

61

6 Nitrate Inducibility of Gene Expression Using the Nitrite Reductase Gene Promoter S.J. Rothstein and S. Sivasankar

83

7 Use of Heat-shock Promoters to Control Gene Expression in Plants R.T. Nagao and W.B. Gurley

97

8 Wound-inducible Genes in Plants L. Zhou and R. Thornburg

127

v

00 Inducible Gene prelims

vi

3/9/99 11:59 AM

Page vi

Contents

9 Developmental Targeting of Gene Expression by the Use of a Senescence-specific Promoter S. Gan and R.M. Amasino

169

10 Abscisic Acid- and Stress-induced Promoter Switches in the Control of Gene Expression Q. Shen and T.-H.D. Ho

187

11 Potential Use of Hormone-responsive Elements to Control Gene Expression in Plants T.J. Guilfoyle and G. Hagen

219

Index

237

00 Inducible Gene prelims

3/9/99 11:59 AM

Page vii

Contributors

Richard M. Amasino, Department of Biochemistry, 420 Henry Mall, University of Wisconsin, Madison, WI 53706-1569, USA. Takashi Aoyama, Institute for Chemical Research, Kyoto University, Uji, Kyoto 611, Japan. Susheng Gan, Tobacco and Health Research Institute and Department of Agronomy, Cooper and University Drives, University of Kentucky, Lexington, KY 40546-0236, USA. Christiane Gatz, Albrecht von Haller Institut für Pflanzenwissenschaften, Universität Göttingen, Untere Karspüle 2, 37073 Göttingen, Germany. Tom J. Guilfoyle, University of Missouri, Department of Biochemistry, 117 Schweitzer Hall, Columbia, MO 62511, USA. William B. Gurley, Department of Microbiology and Cell Science, University of Florida, Gainesville, FL 32611, USA. Gretchen Hagen, University of Missouri, Department of Biochemistry, 117 Schweitzer Hall, Columbia, MO 62511, USA. Tuan-Hua David Ho, Plant Biology Program, Department of Biology, Division of Biology and Biomedical Sciences, Washington University, St Louis, MO 63130, USA. Ian Jepson, Zeneca Agrochemicals, Jealott’s Hill Research Station, Bracknell, Berkshire RG42 6ET, UK. Alberto Martinez, Zeneca Agrochemicals, Jealott’s Hill Research Station, Bracknell, Berkshire RG42 6ET, UK. Vadim L. Mett, Plant Improvement Division, The Horticulture and Food Research Institute of New Zealand, Batchelar Research Centre, Highway 57, Private Bag 11030, Palmerston North, New Zealand. vii

00 Inducible Gene prelims

viii

3/9/99 11:59 AM

Page viii

Contributors

Ronald T. Nagao, Botany Department, University of Georgia, Athens, GA 30602, USA. Paul H.S. Reynolds, Plant Improvement Division, The Horticulture and Food Research Institute of New Zealand, Batchelar Research Centre, Highway 57, Private Bag 11030, Palmerston North, New Zealand. Steven J. Rothstein, Department of Molecular Biology and Genetics, University of Guelph, Guelph, Ontario N1G 2W1, Canada. Qingxi Shen, Monsanto Company, Mail Zone AA2G, 700 Chesterfield Village Parkway, Chesterfield, MO 63198, USA. Sobhana Sivasankar, Department of Molecular Biology and Genetics, University of Guelph, Guelph, Ontario N1G 2W1, Canada. Robert Thornburg, Department of Biochemistry and Biophysics, Iowa State University, Ames, IA 50011, USA. Lan Zhou, Department of Biochemistry and Biophysics, Iowa State University, Ames, IA 50011, USA.

01 Inducible Gene 01

3/9/99 11:59 AM

Page 1

Inducible Control of Gene Expression: an Overview

1

Paul H.S. Reynolds Plant Improvement Division, The Horticulture and Food Research Institute of New Zealand, Batchelar Research Centre, Highway 57, Private Bag 11030, Palmerston North, New Zealand

There is considerable interest in the use of inducible systems for the expression of genes introduced into plants, not only because they allow expression of genes which may, for example, be developmentally lethal, but also because they allow for controlled experiments to be performed in a true isogenic background. Such systems also find use in the manipulation of levels of expression in order to understand more fully individual gene function, or to provide a means for the overproduction or deletion, by reverse genetics, of a particular gene product. This is a rapidly developing area of plant molecular biological research. The need for inducible expression systems is high, not only for their obvious use as research tools, but also for their potential in the future in field-based systems for the inducible expression of desired characters. A wide range of promoter systems can be envisioned which could potentially allow inducible control of genes introduced into plants. These could be broadly described as falling into three general areas. Firstly, there are those which rely on plant-based developmental processes. Such promoters could, for example, include those regulated by plant hormones or which are otherwise developmentally regulated. The advantage of such systems is clearly that all components of the necessary signal transduction pathways are already present in the plant. They also provide a means for the coordinated expression of a gene product within a defined stage of plant growth and development. The second group of promoter systems includes control sequences which respond to particular environmental signals. These potential control systems include heat-shock- and senescence-specific promoters, as well as systems which are responsive to nutritional status. These sorts of promoter systems may well be attractive for the controlled expression of characters in the field, as opposed to the laboratory situation. This is because no application of specific © CAB International 1999. Inducible Gene Expression (ed. P.H.S. Reynolds)

1

01 Inducible Gene 01

2

3/9/99 11:59 AM

Page 2

P.H.S. Reynolds

inducers or defined conditions for growth are necessary, and the desired expression of a gene at a particular growth stage of the plant could be ‘selfregulating’. The third group of control systems comprises those promoters which are introduced from non-plant backgrounds. This includes animal hormone receptor/activators, antibiotic resistance control mechanisms from bacteria and promoters responsive to chemical inducers. Such systems require the introduction of the appropriate transcription factor systems into the plant background together with the inducible promoter. They have the potential advantage that the signal transduction systems are therefore unique to the gene which is being induced and allow timing of expression which is totally independent from the timetable of plant processes and from plant transcription factors.

CONTROL SYSTEMS FROM NON-PLANT BACKGROUNDS The advantage offered by the use of control systems from non-plant sources to be independent from plant processes also provides the disadvantage to their use outside the laboratory. That is, they frequently require modified growth conditions and/or the provision of specific inducers for their activation. For example, the copper-controllable system (Mett et al., 1993, 1996) is activated by copper levels commonly seen in the environment and so is not amenable for use in the field. The tetracycline (Gatz et al., 1992; Weinmann et al., 1994) and animal steroid hormone (Schena et al., 1991; Aoyama and Chua, 1997) systems require the provision of specific elicitors for their activation. None the less these systems offer enormous potential in laboratory-based studies to elucidate the roles of specific genes or to recover potentially lethal signal transduction mutants. The increasing sophistication of plant cell culture techniques and the emerging opportunities offered by the use of ‘plants as factories’ suggests that such promoters will have an important role to play in the commercial plant biotechnology of the future. A recently reported system uses an ethanol-inducible gene switch which may well be amenable to use in the field (Caddick et al., 1998). This system is based on the alc regulator from Aspergillus nidulans, a self-contained genetic system that controls cellular response to ethanol. The system developed for expression in plants utilizes the AlcR transcription factor expressed constitutively, together with the ‘gene of interest’ under the control of a promoter consisting of the CaMV 35S RNA promoter TATA sequences fused to the AlcR binding sites from the A. nidulans AlcA (alcohol dehydrogenase) promoter. Binding of the AlcR transcription factor to the chimeric promoter is responsive to the inducer, ethanol. In an experiment using the system with the chloramphenicol acetyl transferase (CAT) reporter as the ‘gene of interest’, CAT protein was barely detectable in the absence of ethanol. When ethanol was provided either by root drenching as a 1% solution or by foliar spray, there was strong induction of CAT

01 Inducible Gene 01

3/9/99 11:59 AM

Page 3

Inducible Control of Gene Expression: an Overview

3

activity to 50% of that obtained in plants transformed with the CAT reporter under control of the full CaMV 35S RNA promoter. Four promoter systems utilizing transcriptional control systems from outside the plant genome are reviewed here. These are the tetracycline repressor (Chapter 2) and the copper-controllable promoter (Chapter 5) systems. Chapters 3 and 4 discuss the use of mammalian nuclear receptor systems in controlled expression of genes introduced into plants.

Use of the Tn10-encoded tetracycline repressor to control gene expression In Gram-negative bacteria, the Tet repressor (TetR) negatively regulates expression of the tetracycline resistance gene. Induction of this resistance gene is mediated by tetracycline (tc) which binds to TetR and abolishes its DNAbinding activity, thus relieving the repression. This function of TetR is used in two ways to provide inducible expression in plants. 1. TetR is used to repress plant gene expression. This is achieved by the expression of the TetR gene in plants together with the ‘gene of interest’ under control of a chimeric promoter which contains target operator sequences which are bound by TetR about the TATA motif. Repression is relieved by providing tc to plants and the gene of interest is then expressed. 2. TetR is used to activate plant gene expression. This is achieved by expressing a fusion of TetR with the transcriptional activation domain of herpes simplex virus protein VP16 together with the ‘gene of interest’ under the control of a target promoter containing seven tet operators upstream of a minimal promoter. In the absence of tc the TetR-activation fusion binds to the operator sequence and is able to activate transcription. In the presence of tc the fusion protein no longer binds and transcriptional activation is not favoured.

Ecdysteroid agonist-inducible control of gene expression in plants A nuclear receptor functions both as a sensor for its ligand as well as a transcription factor regulating the expression of target genes by binding to specific DNA sequences. Nuclear receptors consist of at least four domains. The A/B and D domains function in transactivation and nuclear targeting, respectively. The DNA-binding (C) domain is the most conserved and consists of two zinc finger structures which bind specific DNA sequences. The carboxyl terminal (E) domain plays multiple roles in ligand binding, dimerization and transcriptional regulation. This domain structure of the nuclear receptors make them ideal candidates for engineering of novel receptors with unique behaviour. A low-background, highexpression system has been developed based on the Heliothis ecdysteroid ligandbinding domain and the glucocorticoid receptor transactivation and DNA-binding domains.

01 Inducible Gene 01

4

3/9/99 11:59 AM

Page 4

P.H.S. Reynolds

This two component system consists of an ‘effector’ which comprises a chimeric receptor containing the glucocorticoid receptor transactivation and DNA-binding domains fused to the Heliothis ecdysteroid receptor ligandbinding domain. The second ‘response’ component contains the ‘gene of interest’ under control of a chimeric promoter with six copies of the glucocorticoid response element fused to a minimal CaMV 35S RNA promoter. In the presence of a suitable ecdysone agonist the system is activated and transcription of the ‘gene of interest’ is initiated from the chimeric promoter.

Glucocorticoid-inducible gene expression in plants A novel glucocorticoid-inducible system which functions in transgenic plants has been developed. It uses only the hormone-binding domain of the glucocorticoid receptor protein as a regulatory domain in a chimeric transcription factor. The chimeric transcription factor ‘GVG’ consists of the yeast GAL4 DNAbinding domain, the transactivating domain of the herpes viral protein VP16 and the hormone-binding domain of the rat glucocorticoid receptor. This chimeric protein strongly activates transcription of the ‘gene of interest’ from a promoter which contains GAL4 upstream activating sequences only in the presence of glucocorticoid.

Tissue-specific copper-controllable gene expression in plants The copper-controllable system makes use of the yeast copper metallothionein regulatory system. In Saccharomyces cerevisiae this consists of a constitutively expressed metallo-responsive transcription factor, targeted to the nucleus, which activates yeast metallothionein transcription. This activation is mediated by copper ions which alter the conformation of the transcription factor allowing it to bind its cognate binding site in the metallothionein promoter, thus activating transcription. This mechanism has been translated into a plant background in a system in which there is constitutive expression of the ace1 gene and control of expression of the ‘gene of interest’ from a chimeric promoter consisting of a minimal promoter fused to the cognate binding site of the ACE1 transcription factor. In the presence of copper ions the ACE1 protein is competent to bind the chimeric promoter and so activate expression. Control over place of expression is effected by controlling the site of expression of the transcription factor.

PLANT PROMOTER SYSTEMS RESPONSIVE TO ENVIRONMENTAL SIGNALS Plants survive in the environment without the ability to avoid many of its rigours, unlike animals, who take shelter, hide or physically modify the

01 Inducible Gene 01

3/9/99 11:59 AM

Page 5

Inducible Control of Gene Expression: an Overview

5

environment to enhance survival. This means that plants have developed a wide range of mechanisms for defence against disease (Hammond-Kosack and Jones, 1997; Sticher et al., 1997), insect or other predator attack (Green and Ryan, 1972) and are able to respond to a wide range of chemical/nutritional threats provided by the environment. An increasing number of gene regulation systems which activate these processes have been elucidated and a number of these systems can readily be used to control the expression of introduced genes. Three systems amenable for use are described here in three review chapters which identify a wide range of current and potential future mechanisms for the control of gene expression by environmental signals. These are wound inducibility (Chapter 8), nutrient control of expression (specifically, nitrate inducibility, Chapter 6), together with a discussion of the heat-shock response and its applicability to inducible control of gene expression (Chapter 7).

Wound-inducible control of gene expression in plants A wide range of plant genes are induced in response to wounding. This chapter identifies classes of proteins produced and discusses the overall biochemical processes important in the wound response. Mechanisms of gene activation of seemingly unrelated proteins (the proteinase inhibitor genes of solanaceous plants and the vegetative storage proteins), in response to wounding are also examined.

Nitrate inducibility of gene expression using the nitrate reductase gene promoter The essential nutrient for plant growth, nitrate, is itself involved in the activation of genes required for its assimilation. The nitrite reductase gene promoter has been extensively studied and the promoter elements responsible for nitrateinducible expression have been identified. However, the mechanism of repression by both glutamine and asparagine has not yet been elucidated, nor have transcription factors binding identified sequence motifs important in nitrate inducibility been cloned. As more information becomes available and it is possible to construct chimeric promoters, the possibility exists for nutritional status control of expression to be obtained.

Use of heat-shock promoters to control gene expression in plants Heat induction of gene expression depends on the presence of heat-shock consensus elements in the promoter. There are three types of promoter: type A are totally dependent on heat-shock transcription factor binding to heat-shock

01 Inducible Gene 01

6

3/9/99 11:59 AM

Page 6

P.H.S. Reynolds

elements (HSEs) in the promoter; type B promoters exhibit HSE-dependent expression which is responsive to developmental signals under non-heat-shock conditions; and type C promoters have multiple mechanisms of induction, only one of which is dependent on HSEs. Heat-shock promoters have been used to provide an inducible expression system with which to answer a range of basic research questions from thermotolerance (Lee and Schoffl, 1996) to the role of the T-6B oncogene of Agrobacterium on plant growth and development (Tinland et al., 1992). Transient heat induction has been shown to be sufficient to express an introduced gene. Heat-shock promoters have been used in mutagenesis screens to isolate heatshock response regulatory mutants and to investigate the consequences of expression of genes which are normally down-regulated during the heat-shock response. There is a wide range of characterized heat-shock promoters available. To successfully express a gene using an inducible heat-shock promoter requires matching of the expression profile of the gene in question with the optimum parameters desired for transgenic expression.

PROMOTER SYSTEMS BASED ON PLANT DEVELOPMENTAL PROCESSES As more research is carried out in the area of development a greater understanding is being obtained of the genes which regulate these processes. The explosion of information in plant vegetative (Taylor, 1997) and floral (Ma, 1998) development will in the future provide elegant mechanisms for the targeted expression of characters in the reproductive growth phase of plant development. Continued research into the signalling mechanisms involved in plant–microbe interactions (Hahn, 1996) will create a fertile hunting ground for potential new inducible control systems. Recent research (Guilfoyle, 1997) has described specific sequence elements in hormone-responsive promoters which will, in the near future, allow the controlled expression of characters by endogenous regulatory pathways. Other plant developmental processes such as organogenesis (for example, the development of the leguminous root nodule (Long, 1996)) and senescence (Gan and Amasino, 1997) offer characterized promoter systems that are amenable for the controlled expression of introduced genes. Three inducible promoter systems which allow the expression of genes introduced into plants by developmental processes are covered. These are senescence-specific promoters (Chapter 9), together with promoters responsive to the plant hormones abscisic acid (ABA) (Chapter 10) and auxin (Chapter 11).

Developmental targeting of gene expression by the use of a senescencespecific promoter The process of senescence is driven by changes in gene expression, involving the activation and inactivation of specific sets of genes. Using differential screening

01 Inducible Gene 01

3/9/99 11:59 AM

Page 7

Inducible Control of Gene Expression: an Overview

7

techniques a number of senescence-associated genes (SAG) have been identified which are activated only during senescence. Two specific genes, SAG12 and SAG13 have transcripts which were only detected in senescing tissues. That is, these two genes were expressed in a highly senescence-specific manner. The promoter regions of both these genes have been fused to the GUS reporter and were shown, in Arabidopsis and tobacco, to direct expression in a senescence-specific manner. Although the signal transduction pathway which activates expression of these genes is not yet known, the identification of their promoter regions now makes it possible to specifically target the expression of genes introduced into plants in senescing tissues.

Abscisic acid-inducible promoters in the control of gene expression in plants Abscisic acid (ABA) appears to be a ‘stress hormone’. In addition to drought, other stresses such as cold and salinity also cause an increase in ABA content. ABA induces the expression of a variety of genes, including those encoding seed storage proteins and late embryogenesis abundant (lea) and RAB (response to ABA) proteins. It has also been implicated in the suppression of gene expression. Studies of the promoters of these genes has allowed the characterization of a core ACGT box which, together with a coupling element, gives ABA signal response specificity. Molecular switches have been constructed which demonstrate different levels of ABA induction and transcription strength. It is particularly significant that these switches function not only in the model barley aleurone tissue but in vegetative tissue as well.

Potential use of hormone-response elements to control gene expression in plants Hormone-response elements are minimal DNA sequence motifs that confer hormone responsiveness to a promoter. Recently, there has been considerable progress in the understanding of auxin response elements (AuxRE), such that there are now identified sequences which have been shown to function in synthetic composite promoters. Two major types of AuxRE, the ocs or as-1 elements (Ellis et al., 1987) and TGTCTC elements (Hagen et al., 1991), have been characterized and minimal sequences identified and tested for functionality in vivo. Ocs/as-1 elements, fused to minimal promoter-GUS reporter genes can be induced by most of the biologically active auxins and, most importantly, by inactive auxin analogues such as 2,3-D. The potential also exists to manipulate promoters with composite AuxRE to enable auxin regulation in unique tissuespecific, organ-specific or developmentally specific fashion.

01 Inducible Gene 01

3/9/99 11:59 AM

8

Page 8

P.H.S. Reynolds

CONCLUDING COMMENTS The utility of inducible promoter systems in laboratory-based research is selfevident and there is clear potential for their usefulness in the genetically modified plants of the future. For example, there is considerable concern about the place of genetically engineered plants in modern agriculture/horticulture and forestry. This concern is wide-ranging, from the effects of prolonged constitutive expression of pest resistance genes to the effects of expressed genes on the metabolism and fitness of the engineered plants. Studies which address the multiple effects caused at multiple trophic levels by the introduction of a new gene into a plant are only now beginning. The precise timing and control over place of expression are important aspects of the increasing sophistication in genetic engineering which in the future will be combined with the ability to control the chromosomal site of insertion. The boundaries to the development of different methods of control over expression of introduced genes are limited only by the scope of human ingenuity and its ability to trap and utilize the masterful and intricate systems that control the growth, development and survival of all organisms on the planet. For example, the universality of the heat-shock response has the potential to allow almost continuous expression of an introduced gene. The potential usefulness of woundinducible promoter systems for the control of expression of genes introduced to effect pest resistance is obvious. The precise timing of expression offered by the use of specific signal compounds which activate introduced non-plant control mechanisms can allow targeted expression of genes outside of intrinsic programmed control processes. In contrast, hormonal promoters allow control of expression of introduced genes as part of programmed developmental cycles. No one inducible control system can provide all of the answers. Clearly what is required is the careful analysis of all the available systems and selection of the one which is most amenable to the particular gene being expressed.

REFERENCES Aoyama, T. and Chua, N.-M. (1997) A glucocorticoid-mediated transcriptional induction system for transgenic plants. The Plant Journal 11, 605–612. Caddick, M.X., Greenland, A.J., Jepson, I., Krause K.-P., Qu, N., Riddell, K.V., Salter M.G., Schuch, W., Sonnewald, U. and Tomsett, A.B. (1998) An ethanol inducible gene switch for plants used to manipulate carbon metabolism. Nature Biotechnology 16, 177–180. Ellis, J.G., Llewellyn, D.J., Walker, J.C., Dennis, E.S. and Peacock, W.J. (1987) The ocs element: a 16 base pair palindrome essential for activity of the octopine synthase enhancer. The EMBO Journal 6, 3203–3208. Gan, S. and Amasino, R.M. (1997) Making sense of senescence. Molecular genetic regulation and manipulation of leaf senescence. Plant Physiology 113, 313–319. Gatz, C., Frohberg, C. and Wendenburg, R. (1992) Stringent repression and homogeneous de-repression by tetracycline of a modified CaMV 35S promoter in intact transgenic tobacco plants. The Plant Journal 2, 397–404.

01 Inducible Gene 01

3/9/99 11:59 AM

Page 9

Inducible Control of Gene Expression: an Overview

9

Green, T. and Ryan, C. (1972) Wound-induced proteinase inhibitor in plant leaves: a possible defense mechanism against insects. Science 175, 776–777. Guilfoyle, T.J. (1997) The structure of plant gene promoters. In: Setlow, J.K. (ed.) Genetic Engineering, Principles and Methods, Vol. 19. Plenum Press, New York, pp. 15–47. Hagen, G., Martin, G., Li, Y. and Guilfoyle, T.J. (1991) Auxin induced expression of the soybean GH3 promoter in transgenic tobacco plants. Plant Molecular Biology 17, 567–579. Hahn, M.G. (1996) Microbial elicitors and their receptors in plants. Annual Review of Phytopathology 34, 387–412. Hammond-Kosack, K.E. and Jones, J.D.G. (1997) Plant disease resistance genes. Annual Review of Plant Physiology and Plant Molecular Biology 48, 575–607. Lee, J.H. and Schoffl, F. (1996) An Hsp70 antisense gene affects the expression of HSP70/HSC70, the regulation of HSF, and the acquisition of thermotolerance in transgenic Arabidopsis thaliana. Molecular and General Genetics 252, 11–19. Long, S.R. (1996) Rhizobium symbiosis: Nod factors in perspective. The Plant Cell 8, 1885–1898. Ma, H. (1998) To be, or not to be, a flower – control of floral meristem identity. Trends in Genetics 14, 26–32. Mett, V.L., Lochhead, L.P. and Reynolds, P.H.S. (1993) Copper controllable gene expression system for whole plants. Proceedings of the National Academy of Sciences USA 90, 4567–4571. Mett, V.L., Podivinsky E., Tennant, A.M., Lochhead, L.P., Jones, W.T. and Reynolds, P.H.S. (1996) A system for tissue-specific copper controllable gene expression in transgenic plants: nodule-specific antisense of aspartate aminotransferase-P2. Transgenic Research 5, 105–113. Schena, M., Lloyd, A.M. and Davis, R.W. (1991) A steroid-inducible gene expression system for plant cells. Proceedings of the National Academy of Sciences USA 88, 10421–10425. Sticher, L., Mauch-Mani, B. and Metraux, J.P. (1997) Systemic acquired resistance. Annual Review of Phytopathology 35, 235–270. Taylor, C.B. (1997) Plant vegetative development: from seed and embryo to shoot and root. The Plant Cell 9, 981–988. Tinland, B., Fournier, P., Heckel, T. and Otten, L. (1992) Expression of a chimeric heatshock-inducible Agrobacterium 6b oncogene in Nicotiana rustica. Plant Molecular Biology 18, 921–930. Weinmann, P., Gossen, M., Hillen, W., Bujard, H. and Gatz, C. (1994) A chimeric transactivator allows tetracycline-responsive gene expression in whole plants. The Plant Journal 5, 559–569.

01 Inducible Gene 01

3/9/99 11:59 AM

Page 10

02 Inducible Gene 02

3/10/99 8:57 AM

Page 11

Use of the Tn10-encoded Tetracycline Repressor to Control Gene Expression

2

Christiane Gatz Albrecht von Haller Institut für Pflanzenwissenschaften, Universität Göttingen, Untere Karspüle 2, 37073 Göttingen, Germany

The Tn10-encoded Tet repressor (TetR) negatively regulates expression of the Tn10-encoded tetracycline resistance gene in Gram-negative bacteria (for review see Hillen and Berens, 1994). Induction is mediated by tetracycline (tc), which binds to TetR, thus abolishing its DNA-binding activity. Taking advantage of the high specificity of the TetR–tet operator interaction, the high affinity of tc to TetR and the favourable transport properties of tc, tc-regulatable gene expression systems have been developed for a variety of eukaryotes. This chapter reviews the features of TetR important for its use to regulate eukaryotic gene expression and describes two different approaches to use TetR for this purpose. Most recent applications of these systems in plants are briefly described.

THE Tn10-ENCODED Tet REPRESSOR (TetR) TetR regulates expression of its own gene (tetR) as well as expression of the tc resistance gene tetA (Fig. 2.1). Both genes are oriented with divergent polarity; between them is a central regulatory region with overlapping promoters and two tet operators. TetR, a dimer of two 24 kDa subunits, binds via a helix–turn–helix motif to two tet operators, resulting in repression of both genes. Induction is based on binding of tc to TetR, resulting in a TetR–tc complex being unable to bind to DNA. This efficient tc-dependent genetic switch might have evolved because of selective pressure against constitutive expression of the resistance gene; which is an integral membrane protein pumping tc out of the cell. The molecular mechanism of the TetR–tet operator interaction has been studied thoroughly (for review see Hillen and Berens, 1994). The sequence of the © CAB International 1999. Inducible Gene Expression (ed. P.H.S. Reynolds)

11

02 Inducible Gene 02

12

3/10/99 8:58 AM

Page 12

C. Gatz

Fig. 2.1. Genetic organization and mechanism of regulation of the Tn10-encoded tc-resistance determinant. Upper panel: autoregulatory expression of TetR (grey circles) leads to TetR levels that repress transcription of tetR and tetA by binding to the two operators O1 and O2. Lower panel: tc binds to TetR, enforcing its dissociation from the DNA, which leads to transcription of both genes. Tcresistance protein TetA, which is an integral membrane protein that exports tc out of the cell, is represented by cylinders.

two Tn10-encoded tet operators are shown in Fig. 2.2. Each operator is a 19 bp palindrome consisting of two 9 bp half sites flanking a central bp. Five of the 9 bps of each half site are directly contacted by amino acids of the N-terminal helix–turn–helix motif of TetR, thus contributing strongly to the specificity of the interaction. The binding constants at an assumed physiological salt concentration of 160 mM sodium chloride are 3 3 102 M21 for non-specific, and 2 3 1011 M21 for specific binding. The ratio of specific over non-specific binding constants (7 3 108) guarantees that non-specific DNA does not effectively compete with operator DNA for repressor binding. Considering the genome size of higher plants (6 3 1010 bp in the allo-diploid species tobacco) this high specificity of binding is an essential feature of TetR for its use in eukaryotes. TetR mutants with altered recognition specificities are also available, thus providing potentially valuable tools for further refinements of tc-dependent expression systems in higher plants. TetR-regulated promoter systems respond to tc, because binding of tc to TetR leads to a conformational change rendering the protein into a non-DNA-

02 Inducible Gene 02

3/10/99 8:58 AM

Page 13

Use of the Tn10-encoded Tetracycline Repressor

13

Fig. 2.2. Sequence of the two tet operators. Asterisks indicate the central bp of the palindrome, arrows illustrate the palindromic nature of the sequence and boxes indicate bp that are directly contacted by TetR.

binding conformation. The high association constant of the inducer tc to TetR (Kass = 3 3 109 M21) makes induction sensitive to even nanomolar concentrations of the drug. The crystal structure of the TetR–tc complex has offered insight into the conformational changes associated with the switch between inducing and repressing structures of TetR. Moreover it might provide clues to develop new inducing tc derivatives lacking antibiotic activities. Using a mutagenesis screen to isolate TetR mutants which repress prokaryotic gene expression in the presence of tc, a TetR mutant was isolated that requires tc for efficient binding to tet operator DNA. This mutant has been successfully used in mammalian systems to regulate gene expression in a reverse manner as compared to TetR (Gossen et al., 1995; see below). Apart from the Tn10-encoded tc-resistance determinant, similar operons have been found to be encoded by other transposons (Tn1721) or plasmids (RA1, pSC101, pJA 8122, pSL1456). They all show the same arrangement of tetA and tetR with tet operators being located overlapping to the promoters in the central regulatory region (Fig. 2.1). The amino acid sequences of the encoded proteins are 43–78% identical. Based on sequence analysis the different tc-resistance determinants are grouped into classes A to G. The operator sequences of the different classes show sequence similarities, but tet operators of classes A, C and G are only poorly recognized by class B and D repressors. This set of naturally occurring TetR derivatives with different operator binding specificities, combined with mutants differing in their response to various tc derivatives opens potential avenues for controlling several transgenes by individual repressor molecules.

USING TetR TO REPRESS PLANT GENE EXPRESSION In the prokaryotic system, TetR represses transcription by sterically interfering with binding of RNA polymerase to the promoter due to the overlapping arrangement of tet operators with promoter sequences. This principle of

02 Inducible Gene 02

14

3/10/99 8:58 AM

Page 14

C. Gatz

regulating gene expression is a common mechanism in bacteria but is found infrequently in higher eukaryotes, where protein–protein interactions are the primary mechanism to mediate stimulating or inhibitory effects on the transcription machinery. Nevertheless, initial experiments using transiently transformed protoplasts revealed that this principle of steric hindrance could also be applied to control a plant promoter (Gatz and Quail, 1988). Two operators were positioned flanking the TATA box of the cauliflower mosaic virus (CaMV) 35S promoter. Expression of TetR was achieved by putting tetR under the control of the wild-type (wt) CaMV 35S promoter. TetR was able to repress the modified CaMV 35S promoter, presumably by interfering with the assembly of a functional transcription initiation complex in this region of the promoter. Repression was relieved by addition of tc. Transfer of these regulatory modules to transgenic plants did not immediately result in an efficient expression system indicating that the principle of repression is less efficient when the target promoter is integrated into the chromosome. Two important adjustments had to be made. The chimeric CaMV 35S:tetR construct used in transient assays provided considerable lower TetR expression levels when integrated into the genome. By shortening the untranslated leader by 50 bp, steady-state levels of one million TetR molecules per cell were achieved in the transgenic situation (Gatz et al., 1991). Second, the location of the operators had to be adjusted. Systematic analysis of 22 CaMV 35S promoter derivatives containing a single tet operator in different positions (Fig. 2.3) demonstrated that repression by TetR depended very much on the exact location of the operator. For instance, when the distance between the 3′ end of the operator and the 5′ end of the TATA box was only 1 bp, repression was very efficient; repression was less efficient at a distance of 3 bp and at a distance of 5 bp no repression was observed (Frohberg et al., 1991). Also, when the distance between the 5′ end of the tet operator and the transcriptional start site exceeded 9 bp the operator was not able to contribute to repression in the presence of TetR (Heins et al., 1992). Whereas occupation of a single operator within the CaMV 35S promoter mediated repression in transiently transformed protoplasts, one operator was not efficient when integrated into the plant genome. However, integration of three operators into the CaMV 35S promoter, with each operator being able to contribute to repression as determined by transient analysis, led to the development of a tightly repressible CaMV 35S promoter derivative (Gatz et al., 1992). Thus, high TetR levels as well as multiple operator sites are required for efficient repression. In contrast with the prokaryotic system, where it only has to interfere with binding of RNA polymerase, TetR has to compete against at least 40 proteins in eukaryotic systems (Fig. 2.4), which cooperate to form a functional transcription initiation complex (Roeder, 1991). Sequence alterations in the vicinity of the TATA

Fig. 2.3. (Opposite) Schematic drawing of the promoter derivatives constructed to define functional operator locations. Using transient assays, 22 CaMV 35S:chloramphenicol acetyl transferase (cat) constructs were tested in TetR-encoding tobacco protoplasts.

02 Inducible Gene 02

3/10/99 8:58 AM

Page 15

Use of the Tn10-encoded Tetracycline Repressor

15

Fig. 2.3. (Continued) Expression was monitored after incubation of protoplasts with or without tc. Only the region from 282 to +30 of the CaMV 35S promoter derivatives is shown with each bp being represented by one square. Except for activating sequence-1 (as-1) and the TATA box, sequences can be replaced without altering promoter activity. The 19 bp tet operator is indicated as a black, grey or white box. Black boxes indicate locations, where TetR interferes with transcription; the grey box indicates a location, where TetR has a weaker negative effect on transcription; and white boxes indicate locations, where TetR has no effect on transcription. Only, the promoter with three operators mediates stringent repression in the transgenic situation. TSS, transcriptional start site.

02 Inducible Gene 02

16

3/10/99 8:59 AM

Page 16

C. Gatz

Fig. 2.4. Schematic representation of the use of TetR to repress transcription. TetR is synthesized under the control of a strong constitutive promoter (upper panel) and controls a target promoter in a tc-dependent manner (lower two panels). The DNA is represented as a string of white squares, the operators are indicated in black and the enhancer module is indicated as a white box. In the absence of the inducer tc (filled triangles), binding of TetR (grey circles) to the operators interferes with assembly of the transcription initiation complex at the TATA box (left panel). Binding of tc to TetR triggers a conformational change in the protein, so that it can no longer bind to DNA, enforcing rapid dissociation from the DNA. Thus, the multifactorial initiation complex, which contains TFIID, TFIIA, TFIIB, TFIIF, TFIIE, other associated factors and RNA polymerase II, can assemble and transcription is initiated (right panel). Tissue specificity of the system can be achieved by choosing appropriate enhancer modules of the target promoter.

box did not reduce expression from the CaMV 35S promoter (Gatz et al., 1992). In tobacco expression of this promoter can be modulated 500-fold by tc. The induction factor is independent of position effects. High-expressing plants have background GUS levels of 2000 pM 4-MU produced min21 (mg protein)21 and can be induced to 180,000 pM 4-MU produced min21 (mg protein)21; lowexpressing plants show, in the absence of tc, GUS levels barely distinguishable from GUS levels detectable in untransformed plants but can only be induced to 1000 to 2000 U. Induction of gene expression is achieved by tc treatment; only 0.1 mg l21 tc is required when single leaves are infiltrated (Gatz et al., 1991). Under these conditions, induction is extremely fast (10 min) reflecting the short signal transduction chain. At the whole plant level, various modes of tc treatment can

02 Inducible Gene 02

3/10/99 8:59 AM

Page 17

Use of the Tn10-encoded Tetracycline Repressor

17

be applied. Systemic induction can be achieved by cultivating plants in hydroponic culture (1 mg l21; Gatz et al., 1992). This method is somehow tedious, as the solution has to be renewed every other day. Depending on the size of the plants, full induction is achieved after 1 or 2 weeks. Alternatively, plants can be grown in sand or rockwool in a setup that allows drainage of the solution (Corlett et al., 1996). Unfortunately, tc treatment reduces root growth, but plant height, chlorophyll content and assimilation rates are only marginally affected. Daily painting of leaves with tc (10 mg l21) is an alternative method of induction. Tc stays in the painted leaf, so that local induction is possible. In tissue culture containers transpiration is not sufficient for homogenous distribution of the inducer but expression in roots and leaves touching the medium is highly induced (Gatz et al., 1992). If fresh tc is not added constantly, TetR turns transcription off again indicating that tc is not very stable in planta. The tc-inducible system has been used to express a dominant negative mutant of the TGA family of transcription factors in transgenic tobacco plants, leading to the conditional reduction of transcription factor complex ASF-1 (Rieping et al., 1994). Conditional reduction of ASF-1 provided the possibility to directly correlate the subsequent reduction of expression from a reporter construct with the reduction of ASF-1, thus concluding that low expression of the reporter construct was simply due to position effects. Plants expressing the Agrobacterium rhizogenes-encoded rolB gene grew normally in the absence of tc, and a very severe phenotype (chlorosis, stop of growth, no flower development) could be induced by adding tc to the hydroponic nutrient solution (Röder et al., 1994). Upon removal of tc, healthy leaves developed again. Tc-inducible expression of the Agrobacterium rhizogenes-encoded rolC gene allowed the analysis of primary effects of RolC on cytokinin levels, thus ruling out the possibility that homeostatic mechanisms might mask primary events (Faiss et al., 1996). Local tc-inducible expression of the Agrobacterium tumefaciensencoded ipt gene helped to prove that the consequences of enhanced cytokinin synthesis remained restricted to the site of hormone production (Faiss et al., 1997). The system further served to obtain transgenic plants overexpressing oat arginine decarboxylase (Masgrau et al., 1996) and S-adenosylmethionine decarboxylase (Kumar et al., 1995) and to evaluate their phenotypes under vegetative and reproductive growth. Originally, the system was developed for tobacco plants. Only one paper describes its use in potato (Kumar et al., 1995). Establishment of the system in tomato and Arabidopsis has failed. In tomato, high levels of TetR caused reduced shoot dry weight, leaf chlorophyll content and leaf size, and an altered photosynthetic capacity when grown in the summer (Corlett et al., 1996). This phenotype was almost completely reversed by the application of tc. In addition, the phenotype was not visible when plants were grown in the winter. Thus TetR seems to interfere with high growth rates under strong light conditions. In Arabidopsis, it seems that repressor concentrations sufficient for transcriptional control cannot be tolerated, a phenomenon that has been also reported for mammalian cells (Gossen et al., 1993).

02 Inducible Gene 02

18

3/10/99 8:59 AM

Page 18

C. Gatz

USING TetR TO ACTIVATE PLANT GENE EXPRESSION As originally described by Gossen and Bujard (1992), TetR can be turned into a tc-controlled transcriptional activator (tTA) when fused to the potent transcriptional activation domain of herpes simplex virus protein 16. Despite this C-terminal extension, TetR retains its DNA-binding activity and tcinducibility. tTA can regulate gene expression from a target promoter containing seven tet operators upstream of a minimal promoter over a range of five orders of magnitude in the mammalian HeLa cell line, which was stably transformed with the construct. The same principle was shown to work in transgenic tobacco plants, thus establishing a promoter system that can be shut off in the presence of tc (Weinmann et al., 1994). The advantage of this system is that background levels are lower than with the tc-inducible system described above. This is due to the fact that inactivation of tTA by tc leads to a target promoter that is not activated (Fig. 2.5). Basal expression from the TATA box in the absence of any activators is very low when the DNA is packed in chromatin. In contrast, repression depends on competition of TetR with a number of proteins assembling around the TATA box and even 99% occupancy of the binding sites only guarantees 100-fold repression. This can be explained by the free access of tTA to the operator sites, thus abolishing the requirement for high levels of tTA. In addition, 50% occupancy of binding sites can be sufficient for transcriptional activation, but is definitely not sufficient for stringent repression. The system has been shown to work in Arabidopsis (M. Roever, U. Treichelt, C. Gatz, J. Schiemann and R. Hehl, Braunschweig/Göttingen, 1995, personal communication). Thus, Arabidopsis seems to tolerate the amount of TetR derivatives needed for transcriptional activation. The tTA-based system has been successfully applied for measuring mRNA decay rates in tobacco BY-2 cells (Gil and Green, 1995). Because of the fast uptake of tc by suspension cultured cells, the target promoter can be shut off very efficiently which allows the observation of first-order decay of transcripts within 15 min after tc treatment. The tTA-dependent promoter provides an important alternative to using general inhibitors of polymerase II like actinomycin D. Actually, it proved to be essential in the analysis of the effect of the 3′-untranslated region of one of the small auxin up-regulated RNAs (SAUR) transcripts on mRNA stability. The destabilizing effect of the sequence was not visible when actinomycin D was used for half-life studies, which indicates that some mRNA decay pathways require ongoing transcription to function. Despite its favourable properties for measuring RNA or protein decay rates, the tTA-dependent expression system has not yet reached its optimum performance. First, expression levels in the absence of tc only reach 30% of the levels reached by the inducible system and drop as transgenic plants age (Weinmann et al., 1994). This problem has been solved recently by reconstructing the target promoter (S. Böhner, I. Lenk and C. Gatz, Göttingen, 1997, personal communication). In addition, cultivating plants permanently on tc to keep the promoter silent can be disadvantageous. A promising alternative

02 Inducible Gene 02

3/10/99 8:59 AM

Page 19

Use of the Tn10-encoded Tetracycline Repressor

19

Fig. 2.5. Schematic representation of the use of TetR to activate transcription. The fusion protein consisting of TetR and an activation domain (tTA) is synthesized under the control of a strong constitutive promoter (upper panel) and controls a target promoter in a tc-dependent manner (lower two panels). The DNA is represented as a string of white squares, the multimerized operators are indicated as a black box in brackets. In the absence of the effector tc (filled triangles), binding of tTA (grey pear-shaped symbol) to the operators activates transcription (left panel), by favouring the functional assembly of the initiation complex consisting of TFIID, TFIIA, TFIIB, TFIIF, TFIIE, other associated factors and polymerase II. Binding of tc to tTA triggers a conformational change in the protein, so that it can no longer bind to DNA and transcription is not activated (right panel). Whether some basal transcription factors keep sitting on the DNA is pure speculation. Tissue specificity of the system can be achieved by choosing appropriate promoters to drive expression of tTA.

was to use the above mentioned TetR mutant that binds to DNA only in the presence of tc (Gossen et al., 1995). Thus, by fusing this mutant to the VP16 domain, a chimeric transcriptional activator (rtTA) was made available. The activity of a target promoter can be induced by tc when rtTA is used, a principle that has been shown to work in mammalian cells. When either Arabidopsis or tobacco was transformed with this construct, no tc-inducible activation of the target promoter was observed. Although mRNA levels similar to tTA mRNA levels were found, no protein was detectable in Western blot analysis using TetR antibodies, indicating that rtTA cannot accumulate in plant cells. We have recently fused tTA to the glucocorticoid receptor hormonebinding domain, resulting in the transcriptional activator TGV, which renders

02 Inducible Gene 02

3/10/99 8:59 AM

20

Page 20

C. Gatz

transcriptional activation dexamethasone (dx)-inducible. A dx-inducible promoter was already established by combining the DNA-binding domain of yeast transcription factor GAL4 with the transcriptional activation domain of VP16 and the glucocorticoid receptor hormone-binding domain (Aoyama and Chua, 1997). This trimeric protein (GVG) activates an artificial promoter consisting of six GAL4 binding sites upstream of the 246 to +1 region of the 35S promoter. As the binding constant of GAL4 (Parthun and Jachnig, 1990) to its target sequence is 100-fold lower than the binding constant of TetR to the tet operators, it could well be that TGV might mediate higher expression levels as compared with GVG. However, this will have to be tested. In addition, background levels of the respective target promoters will have to be compared. An additional feature of TGV is that its activity can be abolished by addition of tc. It remains to be shown whether shutting down transcription by the addition of tc is kinetically more favourable than depletion of dx. The target promoter for tTA or TGV offers a number of useful options. A gene of interest under the control of this promoter can be introduced in either tTA or TGV expressing plants. If the transgene only interferes with regeneration, tTA expressing host plants are recommended. Regeneration could be done in the presence of tc and further analysis could be done without the need of any inducer. If regeneration is done in the absence of tc, expression is constitutive, but can be silenced later, e.g. if the transgene interferes with reproduction. Introduction of the construct into TGV expressing plants keeps the transgene silent in the absence of any chemical. Transcription can be induced and turned off by the subsequent addition of the effectors dx and tc. Moreover, induction can be done at the whole plant level with transcription turned off in selected leaves. In summary, several years of experience with the use of Tn10-encoded regulatory elements for regulating gene expression in eukaryotes has led to a variety of adjustments after the first publication of tc-controlled gene expression in 1988. As the principle of turning TetR into an activator by fusing it to other protein domains has proven to be a more successful strategy than using TetR as a bona fide repressor, the latest development of TGV controlling a target promoter in a dx- and tc-dependent manner seems to be the most promising way to flexibly regulate the expression of transgenes. REFERENCES Aoyama, T. and Chua, N.-H. (1997) A glucocorticoid-mediated transcriptional induction system for transgenic plants. The Plant Journal 11, 605–612. Corlett, J.E., Myatt, S.C. and Thompson, A.J. (1996) Toxicity symptoms caused by high expression of Tet repressor in tomato (Lycopersicon esculentum Mill. L.) are alleviated by tetracycline. Plant Cell Environment 19, 447–454. Faiss, M., Strnad, M., Redig, P., Dolezal, K., Hanus, J., Van Onckelen, H. and Schmülling, T. (1996) Chemically induced expression of the rolC-encoded b-glucosidase in transgenic tobacco plants and analysis of cytokinin metabolism: rolC does not hydrolyze endogenous cytokinin glucosides in plants. The Plant Journal 10, 33–46.

02 Inducible Gene 02

3/10/99 8:59 AM

Page 21

Use of the Tn10-encoded Tetracycline Repressor

21

Faiss, M., Zalubilova, J,, Strnad, M. and Schmülling, T. (1997) Conditional transgenic expression of the ipt gene indicates a function for cytokinins in paracrine signaling in whole tobacco plants. The Plant Journal 12, 401–415. Frohberg, C., Heins, L. and Gatz, C. (1991) Characterization of the interaction of plant transcription factors using a bacterial repressor protein. Proceedings of the National Academy of Sciences USA 88, 10470–10474. Gatz, C. and Quail, P.H. (1988) Tn10-encoded Tet repressor can regulate an operatorcontaining plant promoter. Proceedings of the National Academy of Sciences USA 85, 1394–1397. Gatz, C., Kaiser, A. and Wendenburg, R. (1991) Regulation of a modified CaMV 35S promoter by the Tn10-encoded Tet repressor in transgenic tobacco. Molecular and General Genetics 227, 229–237. Gatz, C., Frohberg, C. and Wendenburg, R. (1992) Stringent repression and homogeneous de-repression by tetracycline of a modified CaMV 35S promoter in intact transgenic tobacco plants. The Plant Journal 2, 397–404. Gil, P. and Green, P.J. (1995) Multiple regions of the Arabidopsis SAUR-AC1 gene control transcript abundance: the 3′-untranslated region functions as an mRNA instability determinant. The EMBO Journal 15, 1678–1686. Gossen, M. and Bujard, H. (1992) Tight control of gene expression in mammalian cells by tetracycline-responsive promoters. Proceedings of the National Academy of Sciences USA 89, 5547–5551. Gossen, M., Bonin, A.L. and Bujard, H. (1993) Control of gene activity in higher eukaryotic cells by prokaryotic regulatory elements. Trends in Biochemical Science 18, 471–475. Gossen, M., Freundlieb, S., Bender, G., Müller, G., Hillen, W. and Bujard, H. (1995) Transcriptional activation by tetracycline in mammalian cells. Science 268, 1766–1769. Heins, L., Frohberg, C. and Gatz, C. (1992) The Tn10 encoded Tet repressor blocks early but not late steps of assembly of the RNA Polymerase II initiation complex in vivo. Molecular and General Genetics 232, 328–331. Hillen, W. and Berens, C. (1994) Mechanisms underlying expression of Tn10 encoded tetracycline resistance. Annual Review of Microbiology 48, 345–369. Kumar, A., Taylor, M.A., Arif, S.A.M. and Davies, H.V. (1995) Potato plants expression antisense and sense S-adenosylmethionine decarboxylase (SAMDC) transgenes show altered levels of polyamines and ethylene: antisense plants display abnormal phenotypes. The Plant Journal 9, 147–158. Masgrau, C., Altabella, T., Farrás, R., Flores, D., Thompson, A.J., Besford, R.T. and Tiburcio, A.F. (1996) Inducible overexpression of oat arginine decarboxylase in transgenic tobacco plants. The Plant Journal 11, 465–473. Parthun, M.R. and Jachnig, J.A. (1990) Purification and characterization of the yeast transcriptional activator GAL4. Journal of Biological Chemistry 265, 209–213. Rieping, M., Fritz, M., Prat, S. and Gatz, C. (1994) A dominant negative mutant of PG13 suppresses transcription from a cauliflower mosaic virus 35S truncated promoter in transgenic tobacco plants. Plant Cell 6, 1087–1098. Roeder, R.G. (1991) The complexities of eukaryotic transcription initiation: regulation of preinitiation complex assembly. Trends in Biochemical Sciences 16, 402–408. Röder, F.T., Schmülling, T. and Gatz, C. (1994) Efficiency of the tetracycline-dependent gene expression system: complete suppression and efficient induction of the rolB phenotype in transgenic plants. Molecular and General Genetics 243, 32–38.

02 Inducible Gene 02

22

3/10/99 8:59 AM

Page 22

C. Gatz

Weinmann, P., Gossen, M., Hillen, W., Bujard, H. and Gatz, C. (1994) A chimeric transactivator allows tetracycline-responsive gene expression in whole plants. The Plant Journal 5, 559–569.

03 Inducible Gene 03

3/9/99 12:20 PM

Page 23

Ecdysteroid Agonist-inducible Control of Gene Expression in Plants

3

Alberto Martinez and Ian Jepson Zeneca Agrochemicals, Jealott’s Hill Research Station, Bracknell, Berkshire RG42 6ET, UK

TRANSCRIPTIONAL CONTROL OF TRANSGENE EXPRESSION IN PLANTS A number of approaches have been reported for the chemical control of transgene expression in plants (Gatz, 1996). Several systems are available including those which rely on plant-inducible promoters as well as relief of repression systems which use bacterial operator repressors and heterologous promoter/transcription factor combinations. Plant promoters which display chemically inducible expression can be exploited to develop heterologous systems for gene regulation. This approach has been adopted in the case of the PR-1 promoter, the activity of which is induced by salicylic acid (Williams et al., 1992). Although this approach has been used successfully with both reporter and insecticidal genes, it may be of limited use due to unspecific induction by pathogens and other chemical triggers. Chemical-dependent regulation of the maize GST-27 promoter has been described using herbicide safeners (Jepson et al., 1994a, b). Inducible regulation was detected in transgenic plants, however, constitutive expression was observed in root tissues. Inducible regulation of transgenes in plants has been achieved by relief of repression or activation of transcription. In the case of the tetracycline (Gatz et al., 1992) and lacI (Wilde et al., 1992) systems, operator sequences were inserted within a target promoter region. The repressor protein binds to these operator sequences in the absence of ligand (i.e. tetracycline or isopropyl-β-Dthiogalacto pyranoside (IPTG)) preventing transcription of the target gene. Addition of the ligand prevents repressor binding to the operator sequence, thus transcription is initiated (Gatz et al., 1992; Wilde et al., 1992). The tetracycline © CAB International 1999. Inducible Gene Expression (ed. P.H.S. Reynolds)

23

03 Inducible Gene 03

3/9/99 12:20 PM

24

Page 24

A. Martinez and I. Jepson

system has been used in protoplasts and transgenic plants. Recently, the system was shown to tightly regulate mRNA levels of arginine decarboxylase in tobacco plants (Masgrau et al., 1997). The lacI system has been exemplified in tobacco protoplasts (Wilde et al., 1992). Due to the nature of their inducing chemicals (IPTG, tetracycline) it is likely these systems will be restricted to research applications only. Chemical-dependent induction of transcription can also be achieved by using ligand-dependent transcription factors and responsive promoter sequences. One such system is that based on the introduction of ACE1, a copperdependent transcriptional activator from yeast, into plants. The addition of inducer (i.e. Cu2+) leads to activation of reporter gene expression (Mett et al., 1993). A second gene control system based on components of the alcohol dehydrogenase regulon of Aspergillus nidulans has been used to provide chemical-inducible gene expression in plants. The alcR regulatory protein in the presence of certain alcohols and ketones will bind to the alcA promoter and achieve gene expression. This system has been used successfully in tobacco, oilseed rape (Sweetman et al., 1997; Tomsett et al., 1997) and tomato (Garoosi et al., 1997; Tomsett et al., 1997). Another example of a switch system based on heterologous transcription factors will be described later in the nuclear receptor section. Although the utility of these systems for research purposes has been documented, a number are not tightly regulated, exhibit low levels of inducible expression or utilize chemistry which is phytotoxic or incompatible with agricultural use.

NUCLEAR RECEPTORS Nuclear receptors are a large well-defined family of transcription factors with over 150 members. Although the presence of lipophilic hormones in mammalian and insect systems has been established for many years, the first nuclear receptors (glucocorticoid and oestrogen receptors) were not isolated until the mid-1980s (Mangelsdorf et al., 1995). Members of the steroid/retinoic acid/thyroid receptor superfamily have been isolated from both invertebrates and vertebrates and include receptors with different developmental functions. Many of these receptors lack a recognized ligand and are known as orphan receptors. The nuclear receptor superfamily encompasses four classes of receptors (Mangelsdorf et al., 1995). Class I are the steroid receptors which generally form homodimers. These receptors are bound by heat-shock proteins (Hsp70, Hsp90) and p59 to form a complex in the cytoplasm (e.g. glucocorticoid receptor (GR)). When bound by ligand, the receptor is released from the complex allowing translocation into the nucleus and binding to a cognate response element as a dimer (see Evans, 1988; Beato, 1991; Green and Chambon, 1988, for reviews). These receptors are only found in vertebrates and thus represent a new evolutionary branch of the superfamily (Mangelsdorf et

03 Inducible Gene 03

3/9/99 12:20 PM

Page 25

Ecdysteroid Agonist-inducible Control

25

al., 1995). Class II receptors, or retinoic-X receptor (RXR) heterodimers, are those belonging to the retinoic/thyroid receptor family. These receptors may interact with response elements as heterodimers. Class II receptors are normally found bound to DNA in the absence of ligand (Mangelsdorf et al., 1995). An example of class II receptors is in the ecdysteroid receptor (EcR) of insects which interacts with ultraspiracle (USP; insect homologue of RXR) to form a heterodimer responsive to ecdysone (Yao et al., 1992, 1993). Class III receptors or dimeric orphan receptors, bind as homodimers to direct repeats in the target promoter. Examples of this family are RXR or COUP (chicken ovalbumin upstream promoter transcription factor-1). Finally, Class IV, or monomeric orphan receptors, bind to extended core site, an example of which is SF1 (Mangelsdorf et al., 1995). Nuclear receptors have six different protein domains (Fig. 3.1a) (see Evans, 1988; Beato, 1991; Green and Chambon, 1988; Mangelsdorf et al., 1995 for reviews). Domains A and B are involved in ligand-independent transactivation. The DNA-binding domain, or domain C, is the best conserved region within the superfamily. This domain is between 66 and 68 amino acids long and has eight invariant cysteine residues implicated in the formation of zinc fingers which are responsible for interacting with DNA response elements. This domain also contributes to dimer formation. Domain D, or the hinge region, is variable and contains the sequences required for direct targeting to the nucleus. The ligandbinding domain is also well-conserved and is not only involved in interactions with ligands but also in dimerization and ligand-dependent activation. The F domain is highly variable in size, and has yet to be ascribed a function. The domain structure of the nuclear receptors make them ideal candidates for engineering of novel receptors with unique behaviour.

Inducible nuclear receptor systems in mammalian systems The use of a number of nuclear receptors for chemical-inducible transcription control has been illustrated in animal cells. Chimeric receptors based on a fusion of the Drosophila ecdysone ligand-binding domain with the glucocorticoid receptor (GR) DNA-binding domain were shown to activate reporter gene activity in HEK 293 and CV-1 cells in the presence of muristeroneA (an ecdysone agonist) (Christopherson et al., 1992). Activation levels were modulated by altering the transactivation domain of the chimeric receptor (Christopherson et al., 1992). The whole Drosophila EcR has been transformed into Chinese hamster ovary (CHO) cells. Addition of ponasteroneA (an ecdysone agonist) resulted in reporter gene activation (Yang et al., 1995), however, levels of activation were not high due to the reliance of the system on the weak transactivation domain of the EcR. No et al. (1996) recently constructed chimeric receptors based on the Drosophila ecdysone ligandbinding domain. The chimeric receptors contain strong transactivator sequences and a modified GR DNA-binding region. The expression levels in

03 Inducible Gene 03

26

3/9/99 12:20 PM

Page 26

A. Martinez and I. Jepson

Fig. 3.1. (a) Nuclear receptor structure. The receptors have six different domains: (A and B) transactivation domain; (C) DNA-binding domain; (D) hinge domain; (E) ligand-binding domain; and (F) C-terminus. (b) Ecdysteroids. MuristeroneA and 20hydroxyecdysone. (c) Non-steroidal compound belonging to the dibenzylhydrazine chemistry. RH5992 (Tebufenozide).

transformed mammalian cells were elevated by 20,000-fold following treatment with muristeroneA. This system was shown to activate gene expression in transgenic mice after the application of muristeroneA. Finally, a progesterone receptor (PR)-based transcription control system has been described (Wang et al., 1997) based on a PR mutant which activates gene expression in the presence of RU486 (an antiprogestin) (Vegeta et al., 1992). Here, the altered specificity ligand-binding domain was fused to the GAL4 DNA-binding domain and a strong transactivator sequence and was shown to activate gene

03 Inducible Gene 03

3/9/99 12:20 PM

Page 27

Ecdysteroid Agonist-inducible Control

27

expression in liver cells of transgenic mice by up to 33,000-fold following application of RU486 (Wang et al., 1997). Both systems used in transgenic mice have been optimized to reduce background levels and have high inducible levels in host cells. The work in mammalian systems has exemplified the modular nature of nuclear receptors and the advantages of manipulating these receptors. In plants, nuclear receptors have been used to control gene expression, and both transcriptional and post-translational approaches have been studied.

Transcriptional control Initial work in tobacco protoplasts demonstrated that expression of the rat glucocorticoid receptor in plants activated reporter gene activity in the presence of dexamethasone (Schena et al., 1991). However, the levels of induced activity were very low when compared to 35S CaMV:CAT controls. Transgenic plants containing the GR failed to induce in the presence of dexamethasone (Lloyd et al., 1994). An alternative system has been recently described in which the GAL4 DNA-binding domain was fused to the herpes simplex VP16 transactivation domain and the GR ligand-binding domain. Arabidopsis and tobacco transgenic plants containing VP16-GAL4-GR were induced with micromolar amounts of dexamethasone (Aoyama and Chua, 1997). Further development of chimeric receptors has used lacI mutants fused to transactivation domain of GAL4 and the GR ligand-binding domain (Moore et al., 1997). The mutant lacI sequences confer high binding affinity to operator sequences (Lehming et al., 1987, 1990). Expression levels in transformed Arabidopsis protoplasts were elevated in the presence of dexamethasone while in the absence of ligand the system remained silent (Moore et al., 1997). These systems show the utility of the nuclear receptor transcriptional control in plants. However, the nature of the GR-inducing compounds (agonists to mammalian glucocorticoid hormone) will restrict their use to research applications.

Post-transcriptional control Steroid receptor ligand-binding domains are amenable to use in the posttranscriptional control of gene expression. In this approach the GR ligandbinding domain is fused to transcription factors controlling plant development. The GR ligand-binding domain fusion causes compartmentalization of the transcription factor in the cytoplasm due to the binding of heat-shock proteins to GR sequences. The transcription factor fusion is released and translated into the nucleus upon inducer interaction with GR ligand-binding domain, where it binds to and activates target genes. This approach has been shown to work in fusions with transcription factors controlling trichoma development (Lloyd et al., 1994), leaf morphology (Aoyama et al., 1995) and flowering time (Simon et

03 Inducible Gene 03

3/9/99 12:20 PM

28

Page 28

A. Martinez and I. Jepson

al., 1996). The GR ligand-binding domain fusions show the flexibility of nuclear receptor components to control gene activity in plants. Although this approach is useful for research studies, it is limited to the regulation of transcription factors and it is difficult to assess the required amounts of transcription factor to deliver the effect.

ECDYSONE RECEPTORS The pleiotropic effect of the moulting hormone, 20-hydroxyecdysone (herein ecdysone) in insects, has been the focus of study for many decades. Ashburner et al. (1974 and references therein) showed that the addition of ecdysone to Drosophila third instar larvae salivary glands resulted in the induction of two sets of genes. The ‘early genes’ are induced upon addition of ecdysone and are necessary for the induction of the ‘late genes’. The induction of early gene expression is mediated by the ecdysone receptor which when bound by ligand activates transcription (Yao et al., 1993). The ecdysteroid receptor (EcR) was first isolated from Drosophila melanogaster and has been shown to be a member of the steroid/retinoic/ thyroid receptor superfamily (Koelle et al., 1991). A number of homologues have been isolated from other insects which show strong similarity to the Drosophila EcR. However, only the Drosophila (Koelle et al., 1991; Thomas et al., 1993; Yao et al., 1992, 1993) and Bombyx mori (Swevers et al., 1996) EcRs have been shown to be functional. We have isolated the Heliothis virescens EcR homologue by a combination of degenerate polymerase chain reaction (PCR), library screening and 5′RACE (Martinez et al., 1999). Figure 3.2 shows the alignment of the hinge and ligandbinding domains of the Drosophila, Bombyx and Heliothis EcR proteins. The Drosophila (Thomas et al., 1993; Yao et al., 1992, 1993), Bombyx (Swevers et al., 1996), Chironomus tentans (Elke et al., 1997), Aedes aegipti (Kapitskaya et al., 1996), Choristoneura fumiferana (Kothapalli et al., 1995) and Heliothis virescens (Martinez et al., 1999) EcR proteins have been shown to form a heterodimer with ultraspiracle (USP) suggesting that all these ecdysteroid receptors are part of the active ecdysone receptor. The EcR–USP complex binds ecdysone and subsequently activates reporter gene expression in mammalian cells (Yao et al., 1992, 1993). Yao et al. (1993) showed that EcR complexed with RXR, the mammalian counterpart of USP, was able to bind muristeroneA (an agonist of ecdysone, see below) but unable to bind ecdysone. The binding of muristeroneA to EcR–RXR is at lower affinity when compared to muristeroneA binding to the EcR–USP complex. Similar data have been obtained with the Heliothis EcR receptor in Chinese hamster ovary (CHO) cells in the presence of muristeroneA (Martinez et al., 1999). A chimeric receptor, containing the GR transactivation and DNA-binding domain fused to the hinge and ligand-binding domains of Drosophila EcR (Christopherson et al., 1992) or Heliothis EcR, were shown to activate reporter gene expression in mammalian cells in the presence of muristeroneA (Fig. 3.3) but did not activate expression in the presence of ecdysone.

03 Inducible Gene 03

3/9/99 12:20 PM

Page 29

Ecdysteroid Agonist-inducible Control

29

Fig. 3.2. Alignment of ecdysone ligand-binding domains from EcR proteins shown to be active. Bombyx mori (BmLBD, Swevers et al., 1995), Drosophila melanogaster (DmLBD, Koelle et al., 1991) and Heliothis virescens (HvLBD, Martinez et al., 1999). The sequence in bold is that of the ligand-binding domain (Domain E). Multiple sequence alignment was carried out using CLUSTAL in PCGENE version 1.0. * indicates residues are identical. . indicates conserved substitution.

ECDYSONE AGONISTS Ecdysteroid compounds Plants are a rich source of agonists of 20-hydroxyecdysone. One such compound is muristeroneA which was isolated from Ipomeoea calonyction. Blackford and coworkers (Blackford et al., 1996; Blackford and Dinan, 1997) have assayed a number of plants in order to ascertain the distribution of ecdysteroidal active compounds in the plant kingdom (Table 3.1). Two assay

03 Inducible Gene 03

30

3/9/99 12:20 PM

Page 30

A. Martinez and I. Jepson

Fig. 3.3. Ecdysone agonist transcriptional activation of reporter gene in mammalian cells. HEK 293 cells transfected with reporter and chimeric receptor show activation of lacZ reporter gene following treatment with 10 mM muristeroneA and 20 mM RH5992 mimic. The chimeric receptor contains the transactivation and DNA-binding domain of GR fused to the ligand-binding domain of the Heliothis ecdysteroid receptor.

systems were adopted. The first relies on immunodetection of ecdysteroid compounds, while the second is based on a cell division bioassay using Drosophila Kc cells. A number of plants have been found to contain ecdysteroidal compounds as judged by both assay methods. While certain species contain relatively high levels of ecdysone agonists, crop plants appear not to contain significant levels of these compounds. The Drosophila Kc cell bioassay has been used to determine the activity of purified ecdysteroids and was found to be a sensitive system for determination of ecdysone agonists (Harmatha and Dinan, 1997) and a good indicator of ecdysone agonist activity (Blackford et al., 1996). While these steroidal agonists are important to gain further understanding on ligand/receptor interactions, they are not suitable candidates for insecticides or transcription system triggers.

Non-steroidal compounds Steroidal compounds are complex, expensive to produce, hydrophilic in nature and detoxification processes in insects are well adapted to deal with them (Hsu, 1991). Despite efforts to discover non-steroidal chemistry mimicking the activity of ecdysone, little advance has been made until recently. A number of compounds from the dibenzylhydrazine chemistry have been shown to have insecticidal activity (Wing, 1988). These compounds bind with high affinity to

03 Inducible Gene 03

3/9/99 12:20 PM

Page 31

Ecdysteroid Agonist-inducible Control

31

Table 3.1. Activity of plant extracts and purified ecdysteroid. Plant

Beta vulgaris ssp. maritima Brassica oleracea cv. botrytis Brassica oleracea cv. capitata Lycopersicon esculentum Solanum tuberosum Dianthus caryophyllus Gossipium hirsutum Helianthus annus Brassica napus Oryza sativa Zea mays Sorghum bicolor Glycine max Nicotiana tabacum 20-Hydroxyecdysone PonasteroneA MakisteroneA a,

Common name

Radioimmunoassaya

Drosophila Kc cells

Sea beet

–

–

Blackford and Dinan, 1997

Cauliflower

–

–

Blackford and Dinan, 1997

Cabbage Tomato Potato Carnation Cotton Sunflower Rape, oilseed rape Rice Maize Sorghum Soybean Tobacco

– – – – 0.071 0.093 –

– – – – – – –

Blackford and Dinan, 1997 Blackford and Dinan, 1997 Blackford and Dinan, 1997 Blackford and Dinan, 1997 Blackford et al., 1996 Blackford et al., 1996 Blackford et al., 1996

0.094 – – – NL nt nt nt

– – – – – 7.5 3 1029 Mb 3.1 3 10210 Mb 1.3 3 1028 Mb

Reference

Blackford et al., 1996 Blackford et al., 1996 Blackford et al., 1996 Blackford et al., 1996 Blackford et al., 1996 Harmatha and Dinan, 1997 Harmatha and Dinan, 1997 Harmatha and Dinan, 1997

µg ecdysone equivalents g21 dry weight; b, ED50; NL, non-linear response; nt, not tested; 2, negative.

ecdysone receptors from insects (Wing, 1988; Wing et al., 1988; Dhadialla and Tzertzinis, 1997). The compounds when applied to growing larvae cause premature head encapsulation, preventing feeding, which leads to death (Wing, 1988). RH5992 (Fig. 3.1b) is a highly substituted member of the family with high activity in lepidopteran species (Carlson et al., 1992). This compound has a narrow spectrum of activity. Two other compounds are in development, RH0345 (Heller et al., 1992) and RH2485 (Carlson et al., 1996; Le et al., 1996), both of which have a different spectrum of activities. Other non-steroidal compounds have been reported in the literature, 3,5di-tert-butyl-4-hydroxy-N-isobutyl-benzamide (DTBHIB) (Mikitani, 1996) and 8-O-acetylharpagide (Elbrecht et al., 1996), but these show poor affinity when compared to ecdysone in cell extracts containing the Drosophila ecdysone receptor. The use of dibenzylhydrazines to control gene expression offers advantages over ecdysteroidal compounds as they are non-phytotoxic yet stable enough for use in the field.

ECDYSONE RECEPTOR SWITCH (ERS) The use of nuclear receptors to control gene transcription has been demonstrated in plants and discussed previously (Schena et al., 1991; Aoyama

03 Inducible Gene 03

32

3/9/99 12:20 PM

Page 32

A. Martinez and I. Jepson

and Chua, 1997; Moore et al., 1997). The systems described use dexamethasone, a steroidal compound which is unsuitable for field use. The ecdysone receptor presents an approach to design a novel inducible system for plants. The system is based on two components. The first component, the effector cassette, is a chimeric receptor containing the GR-transactivation and DNA-binding domain fused to the Heliothis EcR ligand-binding domain (Fig. 3.4). The second component, or reporter cassette, has six copies of the glucocorticoid response element (GRE) fused to the 260 minimal 35S CaMV promoter and βglucuronidase (GUS) gene (Fig. 3.4). A chimeric ecdysone receptor-based system has a number of attractive features as a gene switch. Synthetic, non-steroidal and non-phytotoxic chemistry is available and the system is modular in nature and may be modified. For example, the basal level can be manipulated by altering the minimal promoter context. Furthermore, the use of the GR components favours homodimer formation and thus negates the requirement of USP, the natural partner of EcR.

Transient expression of ecdysone chimeric receptor in maize protoplasts The effector and reporter constructs (Fig. 3.4) were tested in both maize and tobacco protoplasts in the presence of RH5992. Figure 3.5 shows that, in the absence of inducer, low levels of GUS expression were observed. Following treatment of maize and tobacco protoplasts with 100 mM or 10 mM RH5992, respectively, a significant increase in gene expression was observed. The absolute levels of expression observed in maize and tobacco protoplasts were 10% that of the 35S CaMV:GUS controls. These levels are similar to those reported in tobacco protoplasts transformed with GR (Schena et al., 1991). These data demonstrate the system functions in both monocotyledonous and dicotyledonous cells.

Transformation of tobacco with ERS The activation of reporter gene activity in protoplasts by the effector construct containing the Heliothis EcR ligand-binding domain showed the potential of the ERS system. However, Lloyd et al. (1994) reported that GR was also capable of inducing reporter gene expression in protoplasts but not in transgenic plants. This may be a result of poor transactivating potency of the GR sequences. In order to address this, we constructed a new effector cassette containing the strong transactivator domain from herpes simplex VP16 protein (Fig. 3.6) which has been shown to function in plants (Ma et al., 1988; Wilde et al., 1994; Aoyama et al., 1995; Weinmann et al., 1994; Aoyama and Chua, 1997). The effector and reporter gene cassettes were transferred to a Bin19 plant transformation vector to give pERS3.

03 Inducible Gene 03

3/9/99 12:21 PM

Page 33

Ecdysteroid Agonist-inducible Control

33

Fig. 3.4. The effector construct consists of transactivation and DNA-binding domains of the GR fused to the hinge and ligand-binding domains of Heliothis EcR. The reporter construct contains six repeats of glucocorticoid response element (GRE) fused to the 260 minimal 35S CaMV promoter and the reporter gene βglucoronidase.

Transgenic plant analysis Agrobacterium-mediated transformation with pERS3 yielded a transgenic plant population containing 60 independent transformants. PCR, using primer pairs 1–2 and 3–4 (Fig. 3.6), revealed 38 plants containing both effector and reporter cassettes. Seed was collected from and screened for reporter gene activity in the presence of RH5992 (0.1 mM). The assay system involved growing seed from the primary transformants in the presence or absence of inducer. The germinated seedlings were collected 2 days post-germination and GUS activity compared between induced and uninduced treatments

03 Inducible Gene 03

34

3/9/99 12:21 PM

Page 34

A. Martinez and I. Jepson

Fig. 3.5. RH5992 activates reporter gene expression in both maize and tobacco. (a) Maize protoplasts, transformed with both effector and reporter constructs, show activation of GUS reporter gene activity. RH5992 was applied at 100 mM. (b) Tobacco protoplasts were transformed with the dicot version of the effector and reporter constructs. The tobacco protoplast growth media was supplemented with 10 mM of RH5992. GUS activity is expressed as nmol 4-methylumbelliferyl h21 mg21 protein.

(Jefferson et al., 1987). The screen indicated that nine primary transgenic plants did not induce whilst two demonstrated constitutive GUS activity. Twenty-three plants were found to induce GUS activity in the presence of RH5992 ranging from 20 to 150% that of 35S CaMV:GUS seedlings. The induction levels observed throughout the population varied between two- and 430-fold. An example of one transgenic line is shown in Fig. 3.7. High inducibility of the GUS reporter gene was observed following treatment with muristeroneA and RH5992. Phenoxycarb, a juvenile hormone (JH) agonist, does not induce GUS activity even when applied at high levels (0.13 mM). The treatment of ERS3 plants with ecdysone has a small but significant effect on GUS reporter gene expression. The lack of a stronger ecdysone effect in ERS plants may be explained by the requirement for USP by EcR for efficient activation (Thomas et al., 1993; Yao et al., 1992, 1993). Ecdysone treatment of animal cells transfected with similar chimeric constructs failed to activate reporter gene activity (Christopherson et al., 1992), it is perhaps due to the metabolism of ecdysone in tobacco that renders it a better activator than expected. The GUS activity observed in transgenic seedlings treated with muristeroneA and RH5992 is comparable to that of 35S CaMV:GUS transgenic seedlings. The activity observed in plants is higher than that observed in transients and is likely to be due to the presence of VP16 in the modified effector construct. Similar results were observed in transients experiments with the VP16 chimeric receptor (data not shown). It is

03 Inducible Gene 03

3/9/99 12:21 PM

Page 35

Ecdysteroid Agonist-inducible Control

35

Fig. 3.6. ERS plant transformation vector. The effector and reporter cassettes were combined in a pBin19-based vector to give pEGS3. LB and RB denotes the left and right borders of the construct. Arrows denote the PCR primer positions.

important to note that the inducible GUS activity is from a segregating population of ERS3 seed while the 35S CaMV:GUS plants are homozygous. This implies that in homozygous ERS3 seedlings the RH5992 may exhibit improved inducibility with lower variability.

Ligand induction of ERS is dose dependent The effect of muristeroneA and RH5992 on reporter gene activity of a selected ERS3 plant is shown in Fig. 3.8. The experiment shows that RH5992 has a higher affinity for the chimeric receptor than muristeroneA, where maximal induction was observed with muristeroneA at 75 mM and RH5992 at 25 mM. The IC 50 (amount of compound required for 50% induction) was observed to be 7.5 mM for muristeroneA and 2 mM for RH5992. Comparisons between muristeroneA affinity and the non-steroidal compounds in insect systems have not been reported. However, it has been established that a homodimer of EcR binds muristeroneA with lower affinity than the native receptor (EcR/USP).

03 Inducible Gene 03

36

3/9/99 12:21 PM

Page 36

A. Martinez and I. Jepson

Fig. 3.7. Seed from EGS3-37 plant were grown for 2 days post-germination in the presence or absence of the following compounds: DMSO (1.8% v/v); muristeroneA (0.4 mM); RH5992 (0.05 mM); 20-hydroxyecdysone (0.1 mM); and phenoxycarb (0.13 mM). GUS activity is expressed as nmol 4-methyl umbelliferyl h21 mg21 protein.

Induction of ERS is specific The ERS can be activated by certain ecdysone agonists but not others. This has been demonstrated by the lack of activation in the presence of ecdysone, makisteroneA and ponasteroneA in maize protoplasts transformed with effector and reporter plasmid (data not shown). The data indicate that the chimeric ecdysone receptor has different ligand specificity to that of the native insect ecdysone receptor. The narrow specificity of the chimeric homodimeric receptor may be an advantage when the ERS is introduced into plant species containing endogenous ecdysteroids. Environmental factors may interfere with the activity of an inducible system by triggering activation when it is not required. To address this, an ERS3 line was subjected to a number of stresses. Two-day-old seedlings were subjected to 24 h 4°C incubation, heat-shock at 40°C for 2 h followed by 12 h recovery and 12 h heat-shock at 40°C with no recovery. The three replicate samples of ten seedlings were collected and assayed for GUS activity (Jefferson et al., 1987) following exposure to stress conditions. None of the treatments induced GUS activity above control seedling levels (data not shown). Six-week-old greenhouse ERS3 plants were wounded (second leaf was cut from midrib to edge six times each 1 cm apart), grown under waterlog or drought conditions. Samples from droughttreated, wounded and control leaves were taken 24 h post-treatment. Plant roots were submerged for 48 h and then samples were collected from leaves of the treated and untreated plant. All samples were assayed for GUS activity (Jefferson et al., 1987). The abiotic stresses tested to date give no induction of the ERS.

03 Inducible Gene 03

3/9/99 12:21 PM

Page 37

Ecdysteroid Agonist-inducible Control

37

Fig. 3.8. A dose response for a steroidal and non-steroidal inducer of the ecdysone receptor switch. Seeds from EGS3-44 plant were grown for 2 days in the presence of different amounts of muristeroneA or RH5992. The concentration of the compounds ranged from 0.1 mM to 100 nM. GUS activity is expressed as nmol 4-methyl umbelliferyl h21 mg21 protein.

CONCLUSIONS The experiments shown here demonstrate the use of the ecdysone-based transcription control system. We have shown that the ERS system activates reporter gene activity in the presence of inducer in protoplasts isolated from both monocotyledonous and dicotyledonous species. ERS-transformed tobacco plants treated with micromolar levels of a commercial ecdysteroid agonist RH5992 show comparable levels of expression to that seen with a strong constitutive promoter (35S CaMV:GUS). The ERS system is specific to ecdysone agonists and environmental factors (drought, water logging, high and low temperature, and wounding) do not trigger activation of the system. The ERS provides the basis for a plant-inducible system which has broad utility for both basic research and commercial applications.

ACKNOWLEDGEMENTS The authors acknowledge the support of Zeneca Agrochemicals in conducting this work. We also thank P. Broad, D. Scanlon, S. Green (Zeneca Pharmaceuticals), D. Pearson, B. Gross, P. Drayton, C. Sparks, J. Thompson and A. Greenland (Zeneca Agrochemicals) for discussions and practical help.

03 Inducible Gene 03

3/9/99 12:21 PM

38

Page 38

A. Martinez and I. Jepson

REFERENCES Aoyama, T. and Chua, N.-M. (1997) A glucocorticoid-mediated transcriptional induction system for transgenic plants. The Plant Journal 11, 605–612. Aoyama, T., Dong, C.-H., Wu, Y., Carabelli, M., Sessa, G., Ruberti, I., Morelli, G. and Chua, N.-H. (1995) Ectopic expression of the Arabidopsis transcriptional activator Athb-1 alters leaf cell fate in tobacco. Plant Cell 7, 1773–1785. Ashburner, M., Chihara, C., Meltzer, P. and Richards, G. (1974) Temporal control of puffing activity in polytene chromosomes. Cold Spring Harbor Symposium. Quantitative Biology 38, 655–662. Beato, M. (1991) Transcriptional control by nuclear receptors. Federation of American Societies for Experimental Biology Journal 5, 2044–2051. Blackford, M. and Dinan, L. (1997) The tomato moth Locanobia oleracea (Lepidoptera: Noctuidae) detoxifies ingested 20-hydroxyecdysone, but is susceptible to the agonist RH5849 and RH5992. Insect Biochemistry and Molecular Biology 27, 167–177. Blackford, M., Clarke, B. and Dinan, L. (1996) Tolerance of the Egyptian cotton leafworm Spodoptera littoralis (Lepidoptera: Noctuidae) to ingested phytoecdysteroids. Journal of Insect Physiology 42, 931–936. Carlson, G.R., Aller, H.E., Ramsay, R., Slawecki, R., Thirugnanam, M. and Wing, K. (1992) The insecticidal properties of non-steroidal ecdysone agonists. In: X Ecdysone workshop, Liverpool, April 6–9, 1992, p. 31. Carlson, G.R., Dhadialla, T.S., Ramsay, J.R., Thirugnanam, M., James, W.N., Aller, H.E., Hunter, R., Le, D.P. and Lidert, Z. (1996) New ecdysteroid agonist. In: XII Ecdysone workshop, Barcelona, July 22–26, 1996, p. 39. Christopherson, K.S., Melanie, M.R., Bajaj, V. and Godowski, P.J. (1992) Ecdysteroiddependent regulation of genes in mammalian cells by a Drosophila ecdysone receptor and chimeric transactivators. Proceedings of the National Academy of Sciences USA 89, 6314–6318. Dhadialla, T.S. and Tzertzinis, G. (1997) Characterisation and partial cloning of ecdysteroid receptor from a cotton boll weevil embryonic cell line. Archive Insect Biochemistry Physiology 35, 45–57. Elbrecht, A., Chen, Y., Jurgen, T., Hensens, O.D., Zink, B.L., Beck, H.T., Balick, M.J. and Borris, R. (1996) 8-O-Acetylharpagide is a nonsteroidal ecdysteroid agonist. Insect Biochemistry and Molecular Biology 26, 519–523. Elke, C., Vogtli, M., Rauch, P., Spindle-Barth, M. and Lezzi, M. (1997) Expression of EcR and USP in Escherichia coli: purification and functional studies. Archive Insect Biochemistry Physiology 35, 59–69. Evans, R.M. (1988) The steroid and thyroid hormone receptor superfamily: transcriptional regulators of development and physiology. Science 240, 889–895. Garoosi, G.A., Salter, M.G., Collin, H.A., Caddick, M.X. and Tomsett, A.B. (1997) The use of an inducible gene expression cassette for transgenic tomato. Journal of Experimental Botany 48S, 51. Gatz, C. (1996) Chemically inducible promoters in transgenic plants. Current Opinion in Biotechnology 7, 168–172. Gatz, C., Frohberg, C. and Wendenburg, R. (1992) Stringent repression and homogeneous de-repression by tetracycline of a modified CaMV 35S promoter in intact transgenic tobacco plants. The Plant Journal 2, 397–404. Green, S. and Chambon, P. (1988) Nuclear receptors enhance our understanding of transcriptional regulation. Trends in Genetics 4, 309–314.

03 Inducible Gene 03

3/9/99 12:21 PM

Page 39

Ecdysteroid Agonist-inducible Control

39

Harmatha, J. and Dinan, L. (1997) Biological activity of natural and synthetic ecdysteroids in the BII Bioassay. Archive Insect Biochemistry Physiology 35, 219–225. Heller, J.J., Mattioda, H., Klein, E. and Sagenmuller, A. (1992) Proceedings of the 1992 Brighton Crop Protection Conference – Pest and Diseases. British Crop Protection Council, Farnham Royal, pp. 59–65. Hsu, A.C.-T. (1991) 1,2-Diacyl-1-alkylhydrazines, A new class of insect growth regulator. In: Synthesis and Chemistry of Agrochemicals. American Chemical Society, pp. 478–490. Jefferson, R.A., Kavanagh, R.H. and Bevan, M.W. (1987) GUS fusions: β-glucuronidase as a sensitive and versatile gene fusion marker in higher plants. The EMBO Journal 8, 3901–3907. Jepson, I., Lay, V.J., Holt, D.C., Bright, S.W.J. and Greenland, A.J. (1994a) Cloning and characterisation of maize herbicide safener-induced cDNAs encoding subunits of glutathione S-transferase isoforms I, II and IV. Plant Molecular Biology 26, 1855–1866. Jepson, I., Bell, P., Bright, S., Holt, D., Wright, S. and Greenland, A. (1994b) Control of gene expression in transgenic plants using a chemically inducible promoter. Journal of Cellular Biochemistry 18A, 110. Kapitskaya, M., Wang, S., Cress, D.E., Dhadialla, T.S. and Raikhel, A.S. (1996) The mosquito ultraspiracle homologue, a partner of ecdysteroid receptor heterodimer: cloning and characterisation of isoforms expressed during vitellogenesis. Molecular and Cellular Endocrinology 121, 119–132. Koelle, M.R., Talbot, W.S., Segraves, W.A., Bender, M.T., Cherbas, P. and Hogness, D.S. (1991) The Drosophila EcR gene encodes an ecdysone receptor, a new member of the steroid receptor superfamily. Cell 67, 59–77. Kothapalli, R., Palli, S.R., Ladd, T.R., Sohi, S.S., Cress, D., Dhadialla, T.S., Tzertzinis, G. and Rethakaran, A. (1995) Cloning and developmental expression of the ecdysone receptor gene from the spruce budworm, Choristoneura fumiferana. Developmental Genetics 17, 319–330. Le, D.P., Thirugnanam, M., Lidert, Z., Carlson, G.R. and Byan, J.B. (1996) RH2485: a new selective insecticide for caterpillar control. Brighton Crop Protection Conference, Pest and Diseases 1996. British Crop Protection Council, Farnham Royal, pp. 481–486. Lehming, N., Sartoris, J., Niemoeller, M., Genenger, G., Wilcken-Bergmann, V. and Mueller-Hill, B. (1987) The interaction of the recognition helix of the lac operator. The EMBO Journal 6, 3145–3153. Lehming, N., Sartoris, J., Kisters-Woike, B., Wilcken-Bergmann, V. and Mueller-Hill, B. (1990) Mutant lac repressors with new specificities hint at rules for protein–DNA recognition. The EMBO Journal 9, 615–621. Lloyd, A.M., Schena, M., Walbot, V. and Davis, R.W. (1994) Epidermal cell fate determination in Arabidopsis: patterns defined by a steroid inducible regulator. Science 266, 436–439. Ma, J., Przibilla, E., Hu, J., Bogorad, L. and Ptashne, M. (1988) GAL4-VP16 is an unusually potent transcriptional activator. Nature (London) 334, 631–633. Mangelsdorf, D.J., Thummel, C., Beato, M., Herrlich, P., Schutz, G., Umesono, K., Blumberg, B., Kastner, P., Mark, M., Chambon, P. and Evans, R. (1995) The nuclear receptor superfamily: the second decade. Cell 83, 835–839. Martinez, A,. Scanlon, D., Gross, B., Perara, A., Palli, S.R., Greenland, A., Windass, J., Pongs, O., Broad. P. and Jepson, I. (1999) Transcriptional activation of the cloned Heliothis virescens (Lepidoptera) ecdysteroid receptor (HvEcR) by muristeroneA. Insect Biochemistry and Molecular Biology (submitted).

03 Inducible Gene 03

40

3/9/99 12:21 PM

Page 40

A. Martinez and I. Jepson

Masgrau, C., Altabella, T., Farras, R., Flores, D., Thompson, A.J., Besford, R.T. and Tiburcio, A.F. (1997) Inducible overexpression of oat arginine decarboxylase in transgenic tobacco plants. The Plant Journal 11, 465–473. Mett, V.L. Lockhead, L.P. and Reynolds, P.H.S. (1993) Copper-controllable gene expression system for whole plants. Proceedings of the National Academy of Sciences USA 90, 4567–4571. Mikitani, K. (1996) A new nonsteroidal class of ligand for the ecdysteroid receptor 3,5di-tert-butyl-4-hydroxy-N-isobutyl-benzamide shows apparent insect molting hormone activities at molecular and cellular levels. Biochemical and Biophysical Research Communications 227, 427–432. Moore, I., Baroux, C., Gaelweiter, L., Grosskopt, D., Mader, P., Schell, J. and Palme, K. (1997) A transactivation system for regulating expression of transgenes in whole plants. Journal of Experimental Botany 48S, 51. No, D., Yao, T.-P. and Evans, R.M. (1996) Ecdysone inducible gene expression in mammalian cells and transgenic mice. Proceedings of the National Academy of Sciences USA 93, 2246–3351. Schena, M., Lloyd, A.M. and Davis, R.W. (1991) A steroid-inducible gene expression system for plant cells. Proceedings of the National Academy of Sciences USA 88, 10421–10425. Simon, R., Igeno, I.-M. and Coupland, G. (1996) Activation of floral meristem identity genes in Arabidopsis. Nature 384, 59–62. Sweetman, J.P., Paine, J.A.M., Greenland, A.J., Jones, H. and Jepson, I. (1997) Characterisation of an ethanol inducible gene switch in tobacco and oil seed rape. Journal of Experimental Botany 48S, 52. Swevers, L., Drevet, J.R., Lunke, M.D. and Iatrou, K. (1995) The silkmoth homolog of the Drosophila ecdysone receptor (B1 isoform): cloning and analysis of expression during follicular cell differentiation. Insect Biochemistry and Molecular Biology 25, 857–866. Swevers, L., Cherbas, L., Cherbas, P. and Iatrou, K. (1996) Bombyx EcR (BmEcR) and Bombyx USP (BMCF1) combine to form a functional ecdysone receptor. Insect Biochemistry and Molecular Biology 26, 217–221. Thomas, H.E., Stunnenberg, H.G. and Steward, A.F. (1993) Heterodimerisation of the Drosophila ecdysone receptor with retinoid X receptor and ultraspiracle. Nature 362, 471–475. Tomsett, A.B., Salter, M.G., Garoosi, G.A., Caddick, M.X., Paine, J.A.M., Sweetman, J., Greenland, A.J. and Jepson, I. (1997) A chemically-inducible gene cassette for transgenic plants. Journal of Experimental Botany 48S, 46. Vegeta, E., Allan, G.T., Schrader, W.T., Tsai, M.-J., McDonnell, D.P. and O’Malley, B.W. (1992) The mechanism of RU486 antagonism is dependent on the conformation of the carboxy-terminal tail of the human progesterone receptor. Cell 69, 703–713. Wang, Y., DeMayo, F.J., Tsai, S.Y. and O’Malley, B.W. (1997) Ligand-induced and liver expression in transgenic mice. Nature Biotechnology 15, 239–243. Weinmann, P., Gossen, M., Hillen, W., Bujard, H. and Gatz, C. (1994) A chimeric transactivator allows tetracycline-responsive gene expression in whole plants. The Plant Journal 5, 559–569. Wilde, R.J., Shufflebottom, D., Cooke, S., Jasinska, I., Merryweather, A., Beri, R., Brammar, W.J., Bevan, M. and Schuch, W. (1992) Control of gene expression in tobacco cells using a bacterial operator-repressor system. The EMBO Journal 11, 1251–1259.

03 Inducible Gene 03

3/9/99 12:21 PM

Page 41

Ecdysteroid Agonist-inducible Control

41

Wilde, R.J., Cooke, S.E., Brammar, W.J. and Schuch, W. (1994) Control of gene expression in plant cells using a 434:VP16 chimeric protein. Plant Molecular Biology 24, 381–388. Williams, S., Friedrich, L., Dincher, S., Carozzi, N., Kessmann, H., Ward, E. and Ryals, J. (1992) Chemical regulation of Bacillus thuringiensis d-endotoxin expression in transgenic plants. Biotechnology 10, 540–543. Wing, K.D. (1988) RH5948, a non steroidal ecdysone agonist: effects on a Drosophila cell line. Science 241, 467–469. Wing, K.D., Slawecki, R.A. and Carlson, G.R. (1988) RH5849, a nonsteroidal ecdysone agonist: effects on larval lepidoptera. Science 241, 470–472. Yang, G., Hannan, G.N., Lockett, T.J. and Hill, R.J. (1995) Functional transfer of an elementary ecdysone gene regulatory system to mammalian cells: transient transfections and stable cell lines. European Journal of Entomology 92, 379–389. Yao, T.P., Segraves, W.A., Oro, A.E., McKeown, M. and Evans, R.M. (1992) Drosophila ultraspiracle modulates ecdysone receptor function via heterodimer formation. Cell 71, 63–72. Yao, T.P., Forman, B.M., Jiang, Z., Cherbas, L., Chen, J.D., McKeown, M., Cherbas, P. and Evans, R.M. (1993) Functional ecodysone receptor is the product of EcR and ultraspiracle genes. Nature 366, 476–479.

03 Inducible Gene 03

3/9/99 12:21 PM

Page 42

04 Inducible Gene 04

3/16/99 11:09 AM

Page 43

Glucocorticoid-inducible Gene Expression in Plants

4

Takashi Aoyama Institute for Chemical Research, Kyoto University, Uji, Kyoto 611, Japan

INTRODUCTION Transgenic techniques have become a general approach in both basic and applied plant sciences. Many genes have been introduced into plants and expressed under the control of various promoters. Quite often ectopic overexpression of transgenes by a constitutively active promoter is sufficient to provide evidence of their actions. On the other hand, the judicious choice of the most appropriate promoter for transgene expression is critical in realizing the huge potential of transgenic plants. Many native promoters characterized in the literature can be used to express a transgene. Promoters active in specific cell types and responding to specific environmental stimuli or plant hormones are useful in limiting the location or the timing of transgene expression. Transgenic experiments with well-characterized promoters have actually provided us with valuable evidence for functions of plant genes. Although the use of native promoters is a promising way of expressing transgenes, other types of expression systems are still required for a variety of purposes. For example, in the case of a transgene product whose ectopic expression is toxic to plants, it is difficult to use native promoters acting either constitutively or at different developmental stages, because plants expressing the transgene cannot be maintained. A more common problem is the need to induce transgene expression without any pleiotropic effects on plants, i.e. the need to observe the effect of transgene expression. Transcriptional induction using native plant promoters responding to either environmental stimuli or plant hormones is necessarily accompanied by the normal physiological responses of the plant, which may complicate analysis of the effects of the transgenes. For these reasons, induction systems which differ from those © CAB International 1999. Inducible Gene Expression (ed. P.H.S. Reynolds)

43

04 Inducible Gene 04

44

3/16/99 11:09 AM

Page 44

T. Aoyama

normally found in plants, and which have no background level or pleiotropic effects, are highly desirable (for a review see Gatz, 1996). Three types of transcriptional induction systems have been developed as candidates for such an expression system. They are: (i) systems using the tetracycline repressor of the bacterial transposon TetR (Gatz et al., 1992; Weinmann et al., 1994); (ii) the copper ion-responding yeast transcription factor ACE1 (Mett et al., 1993); and (iii) the mammalian glucocorticoid receptor (GR) (Aoyama and Chua, 1997). This chapter describes the system using the GR regulatory mechanism. For many years, an induction system in plants based on GR has been thought of as ideal, since glucocorticoid is highly permeable in cells and has no detectable effects on plants. In fact, a system based on GR and glucocorticoid response elements (GREs) was used as an effective transient expression system in cultured plant cells (Schena et al., 1991), although the system did not work in transgenic plants (Lloyd et al., 1994). Recently, a novel glucocorticoidinducible transcription system that functions in transgenic plants has been developed, which uses only the hormone-binding domain (HBD) of the GR protein as a regulatory domain in a chimeric transcription factor (Aoyama and Chua, 1997). In the following sections, the regulation and utility of the GR HBD are outlined first. Then, the construction, use and characteristics of the glucocorticoid-inducible gene expression system are described. Finally, the potential of this steroid-inducible system as an ideal induction system in plants is discussed.

REGULATORY MECHANISM OF THE GR GR is a member of the nuclear receptor super-family. This family includes receptors for various hydrophobic ligands including thyroid hormones, vitamin D, retinoic acid and steroids (for reviews see: Evans, 1988; Green and Chambon, 1988; Beato, 1989; Laudet et al., 1992). Each nuclear receptor functions as the sensor for its ligand as well as a transcription factor regulating the expression of target genes by binding to specific cis-acting sequences. Nuclear receptors consist of at least four domains (Fig. 4.1a). The A/B and D domains function as a transactivating domain and a hinge domain, respectively. The DNA-binding (C) domain is the most conserved and contains two zinc finger structures which bind specific DNA sequences (Green et al., 1988; Umesono and Evans, 1989). The carboxyl terminal (E) domain plays a role in ligand-binding, dimerization, and transcriptional regulation (Rusconi and Yamamoto, 1987; Forman and Samuels, 1990). The E domains of hormone receptors are also called hormonebinding domains (HBDs). Many systems for the artificial regulation of gene expression have been developed which employ nuclear receptors and their specific cis-acting elements (Picard et al., 1990; No et al., 1996). Rather than using the nuclear receptors themselves, a more attractive approach for regulation of protein functions is to make use of the HBDs of

04 Inducible Gene 04

3/16/99 11:09 AM

Page 45

Glucocorticoid-inducible Gene Expression

45

(a) C9

N9 A/B

C

D

E (HBD)

(b) HSP90 complex Inactive

Functional domain

HBD

Hormone

Active Fig. 4.1. (a) General structure of nuclear receptor proteins (Evans, 1988; Green and Chambon, 1988; Forman and Samuels, 1990). The A/B, C and D domains function as a transactivating domain, a DNA-binding domain and a hinge domain, respectively. The E domain (HBD) plays a role in ligand binding, dimerization and regulation of transcription. (b) Model for the regulatory mechanism of HBDs (Picard et al., 1988; Beato, 1989; Picard, 1993, 1994). In the absence of hormone, the HBDs repress the function of sterically neighbouring domains by forming a complex with multiple proteins including the heat-shock protein HSP90 (inactive state). Hormone-binding releases the complex resulting in de-repression (active state).

steroid hormone receptors. HBDs can function as regulatory domains in cis with fusion proteins, as well as with their own receptors (Picard et al., 1988). A model of the regulatory mechanism of HBDs is illustrated in Fig. 4.1b. It is believed that, in the absence of ligands, HBDs repress the function of sterically neighbouring domains by forming a complex with multiple proteins including the heat-shock protein HSP90. Ligand binding releases the complex resulting in de-repression (Picard et al., 1988; Beato, 1989; Picard, 1993, 1994). Although the HSP90 complex is necessary for the regulation of HBDs, the interaction between the complex and the HBDs does not seem to be species-specific, since mammalian GR functions in other eukaryotes, including yeast and plants (Schena and Yamamoto, 1988; Schena et al., 1991). It is believed that this role of the HSP90 complex is evolutionarily conserved among eukaryotes (Stancato et al., 1996). Table 4.1 shows examples of experiments in which HBDs have been used for the regulation of heterologous proteins. The proteins listed include

Systems

References

E1A c-Myc Rev c-Fos c-Myb C/EBP v-Rel GATA-2 GAL4-VP16 MyoD c-Abl RafI GCN4 R ATHB1-VP16 Cre CO Fas GAL4-VP16

Transcription factor Transcription factor Post-transcriptional activator Transcription factor Transcription factor Transcription factor Transcription factor Transcription factor Transcription factor Transcription factor Tyrosine kinase Serine/threonine kinase Transcription factor Transcription factor Transcription factor Site-specific recombinase Putative transcription factor Cell surface receptor Transcription factor

GR ERa GR GR, ER ER GR, ER ER ER ER ER ER ER ER, MRb GR GR ER GR ER GR

Tissue culture cells Tissue culture cells Tissue culture cells Tissue culture cells Tissue culture cells Tissue culture cells Tissue culture cells Tissue culture cells Yeast Tissue culture cells Tissue culture cells Tissue culture cells Tissue culture cells Transgenic Arabidopsis Transgenic tobacco Tissue culture cells Transgenic Arabidopsis Tissue culture cells Transgenic plants

Picard et al., 1988 Eilers et al., 1989 Hope et al., 1990 Superti-Furga et al., 1991 Burk and Klempnauer, 1991 Umek et al., 1991 Boehmelt et al., 1992 Briegel et al., 1993 Louvion et al., 1993 Hollenberg et al., 1993 Jackson et al., 1993 Samuels et al., 1993 Fankhauser et al., 1994 Lloyd et al., 1994 Aoyama et al., 1995 Metzger et al., 1995 Simon et al., 1996 Kawaguchi et al., 1997 Aoyama and Chua, 1997

a

ER, oestrogen receptor; b MR, mineralocorticoid receptor.

Page 46

HBDs

3/16/99 11:09 AM

Functions

T. Aoyama

Proteins

04 Inducible Gene 04

46

Table 4.1. Heterologous proteins regulated by HBDs.

04 Inducible Gene 04

3/16/99 11:09 AM

Page 47

Glucocorticoid-inducible Gene Expression

47

not only transcription factors but also protein kinases (Jackson et al., 1993; Samuels et al., 1993), a site-specific DNA recombinase (Metzger et al., 1995) and a cell surface receptor (Kawaguchi et al., 1997). Initially, most experiments were performed with mammalian and avian tissue cultures. Although this method should work in transgenic mammals, it is difficult to analyse the results of such induction experiments in animals because of the effects of endogenous steroid hormones and their receptors. On the other hand, this induction system could become a powerful tool in transgenic plants, which have no natural receptors for the vertebrate steroids. Experiments in which transcription factors and a putative transcription factor were regulated with the rat GR HBD in transgenic plants have been described (Lloyd et al., 1994; Aoyama et al., 1995; Simon et al., 1996). The maize regulatory factor R belongs to the family of Myc-type transcription factors (Ludwig et al., 1989). Expression of this gene product complements the Arabidopsis regulatory mutation transparent testa glabra (ttg) (Lloyd et al., 1992). Lloyd et al. (1994) constructed a gene encoding a fusion protein between R and the rat GR (R-GR) HBD, and introduced it into the ttg mutant. In generated transgenic plants, trichome formation on the developing leaf epidermis was artificially induced by glucocorticoid treatment. In an experiment with ATHB-1, an Arabidopsis homeodomain protein (Ruberti et al., 1991), a chimeric transcription factor consisting of the ATHB-1 DNA-binding domain, the transactivating domain of the herpes viral protein VP16, and the rat GR HBD (HDZip1-VP16-GR) was expressed in transgenic tobacco plants (Aoyama et al., 1995). The transgenic plants showed aberrant palisade parenchyma development and de-etiolated phenotypes when grown in the dark only when they were treated with glucocorticoid. In another example, the GR HBD was fused to the putative transcription factor encoded by the Arabidopsis flowering-time gene CONSTANS (CO) (Putterill et al., 1995; Simon et al., 1996). Transgenic co mutant plants expressing the fusion protein (CO-GR) flowered earlier when treated with glucocorticoid. The artificial control of a protein function by HBDs is a very powerful technique for studying the regulatory cascade involving that protein. Fusion proteins formed from transcription factors and HBDs are especially useful for the analysis of a transcription network. Since the induction of transcriptional activation does not require de novo protein synthesis, we can identify those transcripts directly activated by that transcription factor from others which are activated indirectly, by using conditions in which protein synthesis is inhibited. Direct target genes of the Myc transcription factor have been identified using a fusion protein to an oestrogen receptor HBD in animal tissue cultures (Eilers et al., 1991; Grandori and Eisenman, 1997). It is anticipated that appropriate studies will be performed that reveal the network of transcriptional regulation involved in plant morphogenesis by using the induction systems for R-GR, HDZip1-VP16-GR and CO-GR. In general, it is difficult to construct a fusion protein with novel characteristics. We cannot always design a successful fusion protein, even if the 3-D

04 Inducible Gene 04

48

3/16/99 11:09 AM

Page 48

T. Aoyama

structure of each domain is known. One promising method is to mimic the design of a fusion protein that has already been found to work successfully. Therefore, fusion with the HBDs of steroid hormone receptors is one of the most promising ways of artificially regulating a protein’s function. The various applications of HBD fusion shown in Table 4.1 suggest that the HBD mechanism of regulating protein function in eukaryotes has the potential for widespread use.

CONSTRUCTION OF THE GVG SYSTEM Because the HBD of mammalian GR functions in transgenic plants, a transcriptional induction system was constructed using the HBD as a regulatory domain in a chimeric transcription factor. To avoid cross-communication with endogenous regulatory systems in plants, the other components of the system were also obtained from non-plant sources. The chimeric transcription factor consisted of the DNA-binding domain of the yeast GAL4 (Keegan et al., 1986), the transactivating domain of the herpes viral protein VP16 (Triezenberg et al., 1988), and the HBD of rat GR (Picard et al., 1988) (Fig. 4.2a). GAL4 belongs to the class of zinc finger proteins and binds to specific DNA sequences designated as GAL4 upstream activating sequences (UASGs) (Giniger et al., 1985). The minimum DNA-binding domain of GAL4 (amino acids 1–74) (Laughon and Gesteland, 1984) was used. The VP16 domain, an acidic-type transactivating domain, is expected to act as a strong transactivator in all cell types, because it interacts directly with general transcription factors, which are thought to be evolutionarily conserved among eukaryotes (Sadowski et al., 1988; Lin et al., 1991; Goodrich et al., 1993). Amino acids 413–490 of VP16 (Dalrymple et al., 1985) were fused to the C terminus of the GAL4 domain. The resulting fusion protein, GAL4–VP16, strongly activated transcription from a promoter containing six copies of UASG in a transient expression experiment assayed by particle bombardment of tobacco leaves (T. Aoyama et al., unpublished). The HBD of the rat GR (amino acids 519–795) (Miesfeld et al., 1986) was added to this strong transcription factor. The resulting transcription factor was designated as GVG because it consisted of one domain each from GAL4, VP16 and GR. The GVG protein strongly activates transcription from a promoter containing UASGs only in the presence of glucocorticoid. As shown in Fig. 4.2a, the coding sequence of GVG was placed between the 35S promoter of the cauliflower mosaic virus (Odell et al., 1985) and the polyA addition sequence of the pea ribulose bisphosphate carboxylase small subunit rbcS-E9 gene (Coruzzi et al., 1984). In the trans construct, the 35S promoter can be replaced by other promoters using restriction sites at its ends. The expression of GVG by a tissue-specific promoter allows us to induce transgene expression only in a specific tissue. A DNA fragment containing six copies of UASG was chosen as the cisacting element and placed upstream of the TATA box sequence of the 35S promoter. As shown in Fig. 4.2b, there are two restriction sites between the promoter and the polyA addition sequence of the pea rbcS-3A that can be used

04 Inducible Gene 04

3/16/99 11:09 AM

Page 49

Glucocorticoid-inducible Gene Expression

49

(a) Sse8387l 35S promoter

PmeI GAL4

XhoI

(b) GxGAL4 UAS

TATA

VP16

GR

E9

SpeI

3A

Fig. 4.2. Structures of the trans and cis constructs in the GVG system. (a) Structure of the trans construct. The DNA fragments encoding the chimeric transcription factor GVG was placed between the cauliflower mosaic virus 35S promoter (Odell et al., 1985) and the poly(A) addition sequence of the pea ribulose bisphosphate carboxylase small subunit rbcS-E9 (Coruzzi et al., 1984). The 35S promoter can be replaced with other promoters using the restriction sites indicated as Sse8387I and PmeI. (b) Structure of the cis construct. The inducible promoter contains six copies of the UASG and the TATA box region (246 to +1) of the 35S promoter. A DNA fragment to be transcribed inducibly can be placed between the promoter and the polyA addition sequence of the pea rbcS-3A (Fluhr et al., 1986) using the restriction sites indicated as XhoI and SpeI.

for cloning (Fluhr et al., 1986). We can make transgenic plants that express a specific, hormone-inducible, gene by cloning the coding region in the cis construct and introducing it into plants along with the trans construct.

INDUCTION EXPERIMENTS WITH THE GVG SYSTEM The inducibility of the GVG system has been studied in transgenic tobacco (Aoyama and Chua, 1997). The 35S-driven GVG gene was introduced into transgenic tobacco together with a cis construct containing the luciferase (luc) gene (de Wet et al., 1987) as a reporter. Induction of luciferase activity was observed when the transgenic tobacco plants were treated with dexamethasone (DEX), a strong synthetic glucocorticoid. The maximum expression level was over 100 times that of non-induction levels. The induction levels correlated with DEX concentrations ranging from 0.1 to 10 mM when the plants were grown on an agar medium containing DEX. In Northern hybridization analysis with RNAs from hydroponic plants, luc mRNA was detected 1 h after DEX treatment and levels increased to a maximum over 4 h (Aoyama and Chua, 1997). In this section, important aspects of induction experiments are described as well as the results of experiments with transgenic Arabidopsis.

04 Inducible Gene 04

50

3/16/99 11:10 AM

Page 50

T. Aoyama

The same cis and trans constructs used in transgenic tobacco were introduced into Arabidopsis thaliana (ecotype Columbia) and induction experiments were carried out with homozygous T3 plants. In experiments using whole plants, water flow through the vascular system and molecular diffusion are both factors involved in the delivery of glucocorticoid to tissues. Since glucocorticoid is a small hydrophobic chemical and diffuses directly into vertebrate cells without any special transport mechanisms, it is thought that glucocorticoid can also diffuse through plant cell walls and membranes. There are two general methods for treating plants with glucocorticoid. In the first, glucocorticoid is absorbed from the plant surface. Spraying is an easy way to deliver a glucocorticoid solution to the plant surface. This method is especially effective when the exposed epidermal tissue is the target of induction. Figure 4.3a shows the result of a spraying experiment with transgenic Arabidopsis carrying the luc reporter gene. In this experiment, induced luciferase activity was detectable within 30 min of spraying. The other group of methods involve the uptake of glucocorticoid by the vascular system, e.g. from roots or the cut ends of shoots. Induction can be stimulated by simply pouring DEX solution into a pot (Fig. 4.3b). This method allows us to perform induction experiments with healthy plants grown under natural conditions, although we cannot be certain how much DEX is taken up by individual plants. Hydroponic plants, cuttings and leaves all take up glucocorticoids through their roots or cut ends. As long as plants are grown under open air conditions, glucocorticoid is delivered through the vascular system to peripheral tissues quickly. In the experiment shown in Fig. 4.3b, the induction of luciferase activity was detectable in leaves 30 min after adding DEX. Under open air conditions, however, hormone concentrations are thought to vary throughout a plant. The hormone accumulates in leaves in higher concentrations, as a result of transpirational water flow. It is very difficult to deliver glucocorticoid uniformly throughout a plant. One possible way of doing so is by growing enclosed plants on an agar medium containing DEX under airtight conditions. Under these conditions, there is little

Fig. 4.3. (Opposite) Luciferase activity induced in Arabidopsis. Induction experiments were performed with transgenic Arabidopsis carrying the 35S-driven GVG gene and the luc reporter gene. (a) Transgenic plants grown in a pot for 4 weeks were sprayed with the luciferin solution containing 0.5 µM potassium luciferin and 0.01% (w/v) Tween 20. After 30 min, luciferase luminescence from the plants was imaged using a high-sensitivity camera system (Hamamatsu Photonic Systems) (left). Then the plants were sprayed with a solution containing 30 µM DEX and 0.01% (w/v) Tween 20. Twenty-four hours later, the plants were sprayed again with the luciferin solution and imaged (right). (b) Transgenic plants grown in a pot for 4 weeks were sprayed with the luciferin solution and luciferase luminescence from the plants was imaged as described above (left). Then a solution containing 30 µM DEX was poured into the pot. Twenty-four hours later, the plants were sprayed again with the luciferin solution and imaged (right).

04 Inducible Gene 04

3/16/99 11:10 AM

Page 51

Glucocorticoid-inducible Gene Expression

Fig. 4.3. (Continued)

51

04 Inducible Gene 04

52

3/16/99 11:10 AM

Page 52

T. Aoyama

transpirational water flow, so hormone concentrations remain relatively similar throughout the plant. The induction level of transgene expression can be controlled by varying the DEX concentration in the agar medium. Figure 4.4 shows how induction depends on the concentration of DEX. A good correlation between DEX concentrations and induction levels was obtained over concentrations from 3 nM to 300 nM. A significant level of induced luciferase activity was detected at a concentration of 3 nM or higher and the maximum induction level was over 1000 times higher than the non-induction level in Arabidopsis. In a similar experiment with tobacco (Aoyama and Chua, 1997), a 30-fold higher concentration was required for detectable induction and the maximum induction level was about 100 times that of the non-induction level. It is thought that both the high sensitivity and inducibility of Arabidopsis are due to the low non-induction level. Non-induction levels were generally lower in Arabidopsis than in tobacco, although levels vary among transgenic lines in both species.

Fig. 4.4. Luciferase activity induced by different concentrations of DEX. Transgenic Arabidopsis plants carrying the 35S-driven GVG gene and the luc reporter gene were germinated and grown on an agar medium for 14 days, then transferred to a fresh agar medium containing different concentrations of DEX for an additional 2 days. Relative luciferase activities were plotted against DEX concentrations. The value obtained at 0 nM DEX (the non-induction level) was arbitrarily set as 1. Extraction of luciferase and assays for relative luciferase activities were carried out as described by Millar et al. (1992).

04 Inducible Gene 04

3/16/99 11:10 AM

Page 53

Glucocorticoid-inducible Gene Expression

53

CHARACTERISTICS OF THE GVG SYSTEM One of the characteristics of an ideal induction system is that the inducer must not evoke pleiotropic effects that might complicate the analysis of the resulting phenomena. DEX, at least at the concentrations used in induction experiments, does not have any observable physiological effects in wild-type (wt) tobacco or Arabidopsis. Even in experiments in which DEX was assumed to accumulate at high concentrations in leaves, no adverse effects have been observed in the leaves. A class of steroids, the brassinosteroids, has strong physiological effects in plants (for reviews see Mandava, 1988; Li et al., 1996; Hooley, 1996). It is thought that glucocorticoids do not interact with the signal transduction pathway of brassinosteroids, since the molecular structure required for the biological activity of brassinosteroids (Yokota and Mori, 1992) is not found in glucocorticoids. Another requirement of an ideal induction system is that the noninduction level of transgene expression is minimal or absent. Most of the transgenic Arabidopsis lines carrying the 35S-promoter-driven GVG gene and the luc reporter gene have very low levels of luciferase activity under noninduction conditions. In some Arabidopsis lines, no activity was detected at noninduction levels. Nevertheless, there may be a basal level of constitutive induction because the inducible promoter of the GVG system contains an ideal TATA box sequence. It might be possible to find a transgenic plant whose noninduction level is zero by screening many transgenic lines, as both the noninduction and induction levels vary from line to line. As a case in point, transgenic Arabidopsis plants carrying the 35S-driven GVG gene and an inducible diphtheria toxin gene have been produced (T. Aoyama et al., unpublished). Expression of diphtheria toxin kills a cell even at a very low level (Palmitter et al., 1987; Thorsness et al., 1991). The plants that survived under non-induction conditions were killed immediately by DEX treatment, so the non-induction level of the toxin gene expression is believed to be almost zero in these plants. The characteristics of glucocorticoid as an inducer provide an advantage to the GVG system. Since glucocorticoid permeates cells easily, rapid induction of gene expression can be initiated in a variety of ways. By measuring luciferase activity, induction of gene expression was detectable within 30 min of DEX treatment under open air conditions. Glucocorticoid is one of the most-studied biological compounds and has many derivatives. The intensity and sustainability of induction vary with the use of different glucocorticoids (Aoyama and Chua, 1997). In an experiment with transgenic tobacco, plants treated with DEX maintained an induced level of luciferase activity for a longer period than those treated with triamcinolone acetonide, while both groups of plants showed the same initial level of induction. It is hypothesized that triamcinolone acetonide is less stable in plants than DEX. Over 100 different glucocorticoid derivatives are available from commercial sources. Some of these may be very stable in plants, while others are degraded rapidly. The respective types of

04 Inducible Gene 04

54

3/16/99 11:10 AM

Page 54

T. Aoyama

glucocorticoids would be useful for stable and transient induction. Moreover, glucocorticoid antagonists might be used for suppressing induction. It is possible to control the amount of transgene expression using the GVG system. As shown by the experiment in Fig. 4.4, the induction level can be regulated using different concentrations of DEX. This feature allows us to analyse dose-dependent effects of induced gene products under a constant genetic background. Such regulation, however, can only be performed effectively under enclosed conditions, where the concentration of DEX is relatively even throughout a plant. In open-air conditions, it is difficult to control the concentration of glucocorticoid throughout a plant, as discussed above. In such cases, the induction level might be regulated by using different glucocorticoids. The level of induced luciferase activity using an excess concentration of hydrocortisone, a natural glucocorticoid, was over 30-times less than that in response to DEX in tobacco (Aoyama and Chua, 1997).

PROSPECT OF THE STEROID-INDUCIBLE SYSTEM IN PLANTS The GVG system is the first steroid-inducible transcription system developed in transgenic plants. Although the system has been designed to work effectively in a variety of experiments, aspects of the system can be improved for specific experiments. First, modifications in the cis construct might reduce the noninduction level. It is hypothesized that replacement of the ideal TATA box sequence of the inducible promoter would decrease both the non-induction and induction levels. Such a modification would be effective in cases that require a low non-induction level rather than a high induction level. Conversely, increasing the number of UASG repeats might elevate induction levels. The trans construct, i.e. the GVG gene, can also be modified in many ways. Since each of the three domains that make up the GVG protein can function independently, they should be interchangeable with other domains of similar functions. A chimeric transcription factor consisting of the DNA-binding domain from ATHB-1 and the same VP16-GR cassette of the GVG protein worked as a regulatable transcription factor in transgenic tobacco (Aoyama et al., 1995). Many bacterial repressors, whose structures and functions are well characterized (e.g. lacI (Labow et al., 1990), LexA (Godowski et al., 1988) and TetR (Gossen and Bujard, 1992; Weimann et al., 1994)), might be used as alternative DNA-binding domains in the GVG system. With TetR it might be possible to develop a dual-control system, in which transgene expression is induced and repressed by glucocorticoid and tetracycline, respectively. Such a system would be very useful when quick induction and repression of transgene expression are both required. The HBD of the GVG protein can also be replaced by that of other hormone receptors. As shown in Table 4.1, the HBD of an oestrogen receptor (e.g. Eilers et al., 1989; Superti-Furga et al., 1991) and a mineralocorticoid receptor (Fankhauser et al., 1994) have both been used as regulatory domains in chimeric transcription factors. Using a different DNA-

04 Inducible Gene 04

3/16/99 11:10 AM

Page 55

Glucocorticoid-inducible Gene Expression

55

binding domain and the HBD of another steroid hormone receptor, it is possible to develop a second steroid induction system that could be used in combination with the GVG system. The promoter of the GVG gene can also be modified, as previously described. Induction of transgene expression in a specific tissue, so-called spacio-temporal gene expression, is possible using a tissue-specific promoter for the GVG gene. In unicellular systems, like yeast and tissue culture cells, we can perform induction uniformly and analyse the events that are induced in a single type of cell, but in multicellular organisms, such as higher plants, it is very difficult to perform induction uniformly in all cell types. Even assuming that uniform induction is possible, it would be difficult to assess the results due to the variety of responses by different cell types. Inducible gene expression in multicellular organisms is thus fundamentally different from that in unicellular systems. To overcome this weakness, induction should be limited to specific types of cells. Spatio-temporal gene expression by the GVG system will allow us to perform simpler induction experiments in complex organisms. The GVG system is designed to be very flexible and hence the steroidinducible system has the potential to become the ideal induction system in plants. As described above, an ideal induction system for plants should have no non-induction levels or pleiotropic effects. Even if such an ideal system exists theoretically, it is very difficult to prove that any system really satisfies these criteria. It is important that users of induction systems understand the characteristics of the systems thoroughly and carefully design each experiment to obtain the optimal results.

ACKNOWLEDGEMENTS Research by this laboratory has been supported in part by a Grant-in-Aid for scientific research on priority areas from the Ministry of Education, Science and Culture, Japan (09251210). I would like to thank Drs Nam-Hai Chua and Kazuhiko Umesono for their suggestions on the chapter.

REFERENCES Aoyama, T. and Chua, N.-H. (1997) A glucocorticoid-mediated transcriptional induction system for transgenic plants. The Plant Journal 11, 605–612. Aoyama, T., Dong, C.-H., Wu, Y., Carabelli, M., Sessa, G., Ruberti, I., Morelli, G. and Chua, N.-H. (1995) Ectopic expression of the Arabidopsis transcription activator Athb-1 alters leaf cell fate in tobacco. Plant Cell 7, 1773–1785. Beato, M. (1989) Gene regulation by steroid hormones. Cell 56, 335–344. Boehmelt, G., Walker, A., Kabrun, N., Mellitzer, G., Beug, H., Zenke, M. and Enrietto, P.J. (1992) Hormone-regulated v-rel estrogen receptor fusion protein: reversible induction of cell transformation and cellular gene expression. The EMBO Journal 11, 4641–4652.

04 Inducible Gene 04

56

3/16/99 11:10 AM

Page 56

T. Aoyama

Briegel, K., Lim, K.C., Plank, C., Beug, H., Engel, J.D. and Zenke, M. (1993) Ectopic expression of a conditional GATA-2/estrogen receptor chimera arrests erythroid differentiation in a hormone-dependent manner. Genes & Development 7, 1097–1109. Burk, O. and Klempnauer, K.H. (1991) Estrogen-dependent alterations in differentiation state of myeloid cells caused by a v-myb/estrogen receptor fusion protein. The EMBO Journal 10, 3713–3719. Coruzzi, G., Broglie, R., Edwards, C. and Chua, N.-H. (1984) Tissue-specific and lightregulated expression of a pea nuclear gene encoding the small subunit of ribulose1,5-bisphosphate carboxylase. The EMBO Journal 3, 1671–1679. Dalrymple, M.A., McGeoch, D.J., Davison, A.J. and Preston, C.M. (1985) DNA sequence of the herpes simplex virus type 1 gene whose product is responsible for transcriptional activation of immediate early promoters. Nucleic Acids Research 13, 7865–7879. de Wet, J.R., Wood, K.V., DeLuca, M., Helinski, D.R. and Subramani, S. (1987) Firefly luciferase gene: structure and expression in mammalian cells. Molecular and Cellular Biology 7, 725–737. Eilers, M., Picard, D., Yamamoto, K.R. and Bishop, J.M. (1989) Chimaeras of myc oncoprotein and steroid receptors cause hormone-dependent transformation of cells. Nature 340, 66–68. Eilers, M., Sabine, S. and Bishop, J.M. (1991) The MYC protein activates transcription of the a-prothymosin gene. The EMBO Journal 10, 133–141. Evans, R.M. (1988) The steroid and thyroid hormone receptor super-family. Science 240, 889–895. Fankhauser, C.P., Briand, P.A. and Picard, D. (1994) The hormone binding domain of the mineralcorticoid receptor can regulate heterologous activities in cis. Biochemical and Biophysical Research Communications 200, 195–201. Fluhr, R., Moses, P., Morelli, G., Coruzzi, G. and Chua, N.-H. (1986) Expression dynamics of the pea rbcS multigene family and organ distribution of the transcripts. The EMBO Journal 5, 2063–2071. Forman, B.M. and Samuels, H.H. (1990) Interaction among a subfamily of nuclear hormone receptors: the regulatory zipper model. Molecular Endocrinology 4, 1293–1301. Gatz, C. (1996) Chemically inducible promoters in transgenic plants. Current Opinion in Biotechnology 7, 168–172. Gatz, C., Frohberg, C. and Wendenburg, R. (1992) Stringent repression and homogeneous de-repression by tetracycline of a modified CaMV 35S promoter in intact transgenic tobacco plants. The Plant Journal 2, 397–404. Giniger, E., Varnum, S. and Ptashne, M. (1985) Specific DNA binding of GAL4, a positive regulatory protein of yeast. Cell 40, 767–774. Godowski, P.J., Picard, D. and Yamamoto, K.R. (1988) Signal transduction and transcriptional regulation by glucocorticoid receptor–LexA fusion proteins. Science 241, 812–816. Goodrich, J.A., Hoey, T., Thut, C.J., Admon, A. and Tjian, R. (1993) Drosophila TAFII40 interacts with both a VP16 activation domain and the basal transcription factor TFIIB. Cell 75, 519–530. Gossen, M. and Bujard, H. (1992) Tight control of gene expression in mammalian cells by tetracycline responsive promoters. Proceedings of the National Academy of Sciences USA 89, 5547–5551.

04 Inducible Gene 04

3/16/99 11:10 AM

Page 57

Glucocorticoid-inducible Gene Expression

57

Grandori, C. and Eisenman, R.N. (1997) Myc target genes. Trends in Biochemical Science 22, 177–181. Green, S. and Chambon, P. (1988) Nuclear receptors enhance our understanding of transcription regulation. Trends in Genetics 4, 309–314. Green, S., Kumar, V., Theulaz, I., Wahli, W. and Chambon, P. (1988) The N-terminal DNA-binding ‘zinc finger’ of the oestrogen and glucocorticoid receptors determines target gene specificity. The EMBO Journal 7, 3037–3044. Hollenberg, S.M., Cheng, P.F. and Weintraub, H. (1993) Use of a conditional MyoD transcription factor in studies of MyoD trans-activation and muscle determination. Proceedings of the National Academy of Sciences USA 90, 8028–8032. Hooley, R. (1996) Plant steroid hormones emerge from the dark. Trends in Genetics 12, 281–283. Hope, T.J., Huang, X.J., McDonald, D. and Parslow, T.G. (1990) Steroid receptor fusion of the human immunodeficiency virus type 1 Rev transactivator: mapping cryptic functions of the arginine-rich motif. Proceedings of the National Academy of Sciences USA 87, 7787–7791. Jackson, P., Baltimore, D. and Picard, D. (1993) Hormone-conditional transformation by fusion proteins of c-Abl and its transforming variants. The EMBO Journal 12, 2809–2819. Kawaguchi, Y., Takebayashi, H., Kakizuka, A., Arii, S., Kato, M. and Imamura, M. (1997) Expression of Fas–estrogen receptor fusion protein induces cell death in pancreatic cancer cell lines. Cancer Letters 116, 53–59. Keegan, L., Gill, G. and Ptashne, M. (1986) Separation of DNA binding from the transcription-activating function of eukaryotic regulatory protein. Science 231, 699–704. Labow, M.A., Bain, S.B., Shenk, T. and Levine, A.J. (1990) Conversion of the lac repressor into an allosterically regulated transcriptional activator for mammalian cells. Molecular and Cellular Biology 10, 3343–3356. Laudet, V., Hanni, C., Coll, J., Catzeflis, F. and Stehelin, D. (1992) Evolution of the nuclear receptor gene superfamily. The EMBO Journal 11, 1003–1013. Laughon, A. and Gesteland, R. (1984) Primary structure of the Saccharomyces cerevisiae GAL4 gene. Molecular and Cellular Biology 4, 260–267. Li, J., Nagpal, P., Vitart, V., McMorris T.C. and Chory, J. (1996) A role for brassinosteroids in light-dependent development of Arabidopsis. Science 272, 398–401. Lin, Y.-S., Maldonado, E., Reinberg, D. and Green, M.R. (1991) Binding of general transcription factor TFIIB to an acidic activating region. Nature 353, 569–571. Lloyd, A.M., Walbot, V. and Davis, R.W. (1992) Arabidopsis and Nicotiana anthocyanin production activated maize regulators R and C1. Science 258, 1773–1775. Lloyd, A.M., Schena, M., Walbot, V. and Davis, R.W. (1994) Epidermal cell fate determination in Arabidopsis: patterns defined by a steroid-inducible regulator. Science 266, 436–439. Louvion, J.-F., Havaux-Copf, B. and Picard, D. (1993) Fusion of GAL4-VP16 to a steroidbinding domain provides a tool for gratuitous induction of galactose-response genes in yeast. Gene 131, 129–134. Ludwig, S.R., Habera, L.F., Dellaporta, S.L. and Wessler, S.R. (1989) Lc, a member of the maize R gene family responsible for tissue-specific anthocyanin production, encodes a protein similar to transcriptional activators and contains the mychomology region. Proceedings of the National Academy of Sciences USA 86, 7092–7096.

04 Inducible Gene 04

58

3/16/99 11:10 AM

Page 58

T. Aoyama

Mandava, N.B. (1988) Plant growth-promoting brassinosteroids. Annual Review of Plant Physiology and Plant Molecular Biology 39, 23–52. Mett, V.L., Lockhead, L.P. and Reynolds, P.H.S. (1993) Copper controllable gene expression system for whole plants. Proceedings of the National Academy of Sciences USA 90, 4567–4571. Metzger, D., Clifford, J., Chiba, H. and Chambon, P. (1995) Conditional site-specific recombination in mammalian cells using a ligand-dependent chimeric Cre recombinase. Proceedings of the National Academy of Sciences USA 92, 6991–6995. Miesfeld, R., Rusconi, S., Godowski, P.J., Maler, B.A., Okret, S., Wikstroem, A.-C., Gustafsson, J.-A. and Yamamoto, K.R. (1986) Genetic complementation to a glucocorticoid receptor deficiency by expression of cloned receptor cDNA. Cell 46, 389–399. Millar, A.J., Short, S.R., Chua, N.-H. and Kay, S.A. (1992) A novel circadian phenotype based on firefly luciferase expression in transgenic plants. Plant Cell 4, 1075–1087. No, D., Yao, T.-P. and Evans, R.M. (1996) Ecdysone-inducible gene expression in mammalian cells and transgenic mice. Proceedings of the National Academy of Sciences USA 93, 3346–3351. Odell, J.T., Nagy, F. and Chua, N.-H. (1985) Identification of DNA sequences required for activity of the cauliflower mosaic virus 35S promoter. Nature 313, 810–812. Palmitter, R.D., Behringer, R.R., Quaife, C.J., Maxwell, F., Maxwell, I.H. and Brinster, R.L. (1987) Cell lineage ablation in transgenic mice by cell-specific expression of a toxin gene. Cell 50, 435–443. Picard, D. (1993) Steroid-binding domains for regulating the functions of heterologous proteins in cis. Trends in Cell Biology 3, 278–280. Picard, D. (1994) Regulation of protein function through expression of chimeric proteins. Current Opinion in Biotechnology 5, 511–515. Picard, D., Salser, S.J. and Yamamoto, K.R. (1988) A movable and regulable inactivation function within the steroid binding domain of the glucocorticoid receptor. Cell 54, 1073–1080. Picard, D., Schena, M. and Yamamoto, K.R. (1990) An inducible expression vector for both fision and budding yeast. Gene 86, 257–261. Putterill, J., Robson, F., Lee, K., Simon, R. and Coupland, G. (1995) The CONSTANS gene of Arabidopsis promotes flowering and encodes a protein showing similarities to zinc finger transcription factors. Cell 80, 847–857. Ruberti, I., Sessa, G., Lucchetti, S. and Morelli, G. (1991) A novel class of plant proteins containing a homeodomain with a closely linked leucine zipper motif. The EMBO Journal 10, 1787–1791. Rusconi, S. and Yamamoto, K.R. (1987) Functional dissection of the hormone and DNA binding activities of the glucocorticoid receptor. The EMBO Journal 6, 1309–1315. Sadowski, I., Ma, J., Triezenberg, S. and Ptashne, M. (1988) GAL4-VP16 is an unusually potent transcription activator. Nature 335, 563–564. Samuels, M.L., Weber, J.M., Bishop, J.M. and McMahon, M. (1993) Conditional transformation of cells and rapid activation of the mitogen-activated protein kinase cascade by an estradiol-dependent human raf-1. Molecular and Cellular Biology 13, 6241–6252. Schena, M. and Yamamoto, K.R. (1988) Mammalian glucocorticoid receptor derivatives enhance transcription in yeast. Science 241, 965–967. Schena, M., Lloyd, A.M. and Davis, R.W. (1991) A steroid-inducible gene expression system for plant cells. Proceedings of the National Academy of Sciences USA 88, 10421–10425.

04 Inducible Gene 04

3/16/99 11:10 AM

Page 59

Glucocorticoid-inducible Gene Expression

59

Simon, R., Igeno, M.I. and Coupland, G. (1996) Activation of floral meristem identity genes in Arabidopsis. Nature 384, 59–62. Stancato, L.F., Hutchison, K.A., Krishna, P. and Pratt, W.B. (1996) Animal and plant cell lysates share a conserved chaperone system that assembles the glucocorticoid receptor into a functional heterocomplex with hsp90. Biochemistry 35, 544–561. Superti-Furga, G., Gergers, G., Picard, D. and Busslinger, M. (1991) Hormone-dependent transcriptional regulation and cellular transformation by Fos-steroid receptor fusion proteins. Proceedings of the National Academy of Sciences USA 88, 5114–5118. Thorsness, M.K., Kansasamy, M.K., Nasrallah, M.E. and Nasrallah, J.B. (1991) A brassica S-locus gene promoter targets toxic gene expression and cell death to the pistil and pollen of transgenic Nicotiana. Developmental Biology 143, 173–184. Triezenberg, S.J., Kingsbury, R.C. and McKnight, S.L. (1988) Functional dissection of VP16, the transactivator of herpes simplex virus immediate early gene expression. Genes and Development 2, 718–729. Umek, R.M., Friedman, A.D. and McKnight, S.L. (1991) CCAAT-enhancer binding protein: a component of a differentiation switch. Science 251, 288–292. Umesono, K. and Evans, R.M. (1989) Determinants of target gene specificity for steroid/thyroid hormone receptors. Cell 57, 1139–1146. Weinmann, P., Gossen, M., Hillen, W., Bujard, H. and Gatz, C. (1994) A chimeric transactivator allows tetracycline-responsive gene expression in whole plants. The Plant Journal 5, 559–569. Yokota, T. and Mori, K. (1992) Molecular structure and biological activity of brassinolide and related brassinosteroids. In: Duax, W.L. and Bohl, M. (eds) Molecular Structure and Biological Activity of Steroids (Uni-science Report Series). CRC Press, Washington, DC, pp. 317–340.

04 Inducible Gene 04

3/16/99 11:10 AM

Page 60

05 Inducible Gene 05

3/16/99 11:12 AM

Page 61

Tissue-specific, Coppercontrollable Gene Expression in Plants

5

Vadim L. Mett and Paul H.S. Reynolds Plant Improvement Division, The Horticulture and Food Research Institute of New Zealand, Batchelar Research Centre, Highway 57, Private Bag 11030, Palmerston North, New Zealand

The copper-controllable gene expression system has been shown to give tight control over time and place of expression of a gene of interest in response to the application of copper to transformed plants either in the nutrient solution or as a foliar spray. Whilst the levels of expression from this system are not high when compared to the cauliflower mosaic virus (CaMV) 35S RNA promoter, they have been shown to be sufficient to, for example, drive effective antisense of a metabolic gene, to express plant hormone biosynthetic genes resulting in phenotype changes and to express potentially lethal avirulence genes. Perhaps the most significant aspect of this system is the tight control it effects allowing the recovery of transgenic plants expressing genes which are conditionally lethal or of plants carrying genes which, if expressed in tissue culture, would compromise plant developmental processes. The system has been successfully used in tobacco, Lotus and Arabidopsis backgrounds. Its functionality in Arabidopsis is particularly useful in that other systems, such as the tetracycline repressor, do not function in this background. On the other hand, experiments in poplar (S.H. Strauss, Oregon, USA, 1998, personal communication) have suggested that, in this plant, the copper system operates constitutively. Here, the basis of the copper-controllable system is described together with examples of its successful use in plants. A vector system for convenient use is presented, together with practical information on the conducting of experiments in plants.

© CAB International 1999. Inducible Gene Expression (ed. P.H.S. Reynolds)

61

05 Inducible Gene 05

62

3/16/99 11:12 AM

Page 62

V.L. Mett and P.H.S. Reynolds

BASIS AND FUNCTIONING OF THE COPPER-CONTROLLABLE EXPRESSION SYSTEM The copper-controllable gene expression mechanism is based on the yeast copper–metallothionein (MT) regulatory system with copper ion as the ‘inducer molecule’. The yeast copper–MT system, consists of a constitutively expressed activating copper-metallothionein expression-1 (ace1) gene which encodes a metallo-responsive transcription factor, targeted to the nucleus, which activates yeast MT gene transcription (Wright et al., 1988). This activation is mediated by copper ions which alter the conformation of the regulatory protein (Szczypka and Thiele, 1989) allowing it to bind to specific upstream sequences in the yeast MT promoter (Butt et al., 1984; Furst et al., 1988). The yeast copper–MT regulatory system thus represents a simple model in which the effective binding conformation of the regulatory protein to the MT promoter (and hence, activation of the MT gene) is controlled by the copper ion concentration (Thiele and Hamer, 1986). To translate this mechanism into a plant background (Fig. 5.1) would require constitutive expression of the ace1 gene using, for example, the CaMV 35S RNA promoter, together with a chimeric promoter to drive expression of the ‘gene of interest’ consisting of some basic plant-compatible TATA sequence together with the binding site for the ACE1 protein. MTs are characteristically low-molecular-weight, cysteine-rich polypeptides which fall into three classes based on the arrangement of their cysteine residues. Animals, yeasts and other fungi synthesize class I or II MTs which are encoded in the nuclear genome. Higher plants, on the other hand, produce phytochelatins (class III MTs), which are enzyme-synthesized peptides with the structure poly(γglutamyl-cysteinyl)glycine (Thiele, 1992). Recent evidence suggests that plants contain metal-binding proteins in addition to phytochelatins. Studies of the copper-tolerant flowering plant Mimulus guttatus have revealed that root extracts contain several copper-binding elements, only one of which is a phytochelatin (Grill et al., 1987). Interestingly, transcription of the MT-like genes investigated in this work did not appear to be induced by copper ions. In fact, half of the transcripts analysed were repressed by high copper ion concentrations. Genes encoding MT-like proteins have also been isolated from a range of other plants. Some plant MT-like genes are induced by metal treatment (DeFramond, 1991; Robinson et al., 1993; Zhou and Goldsbrough, 1994; Hsein et al., 1995) whilst others respond to different environmental and developmental signals such as senescence, abscission, wounding and tobacco mosaic virus (TMV) infection (Buchanan-Wollaston, 1994; Coupe et al., 1995; Choi et al., 1996). Even when ‘induced’ by copper, the level of induction driven from the promoters of these genes is low. For example, treatment of Arabidopsis seedlings with 50 µM CuSO4 for 30 h produced only a 5.5-fold increase in the level of MT2 mRNA, as determined by densitometry (Zhou and Goldsbrough, 1994). The exact role of MT-like plant proteins is yet to be determined though at present it does not seem to relate directly to metal tolerance but rather, may be important in the directed delivery of copper to protein complexes in which it is important.

05 Inducible Gene 05

3/16/99 11:12 AM

Page 63

Tissue-specific, Copper-controllable Gene Expression

Inactive

ACE1

Application of copper

ACE1

63

Active

Reporter protein

ACE1 p35S

ace 1

MRE

Reporter

Fig. 5.1. The copper-inducible expression system. The system consists of a constitutively expressed cysteine-rich nucleoprotein encoded by the ace1 gene, under the control of the CaMV 35S RNA promoter (p35S). This protein activates transcription of the ‘gene of interest’ (reporter) by binding to its cognate binding site (the metal regulatory element, MRE) in a chimeric promoter which also contains sequences necessary for transcription in plants. The competence of the ACE1 protein to bind the MRE is mediated by copper which alters its conformation to allow effective binding and activation of transcription from the chimeric promoter.

The key question, therefore, in investigating the feasibility of a coppercontrol system in plants (Fig. 5.1) is to determine: 1. Whether or not activation of the introduced yeast regulatory copper–MT system by manipulation of copper ion concentration is possible in plants against a background of multiple metal binding systems, and; 2. If such manipulation of copper ion levels has widespread physiological effects in the plant. To test the functioning of the system in plants a construct, containing the β-glucuronidase (GUS) reporter gene under control of the chimeric promoter (containing the ACE1-binding site and domain A from CaMV 35S RNA promoter) together with the ace1 gene fused to the CaMV 35S RNA promoter, was prepared and transgenic tobacco plants produced (Mett et al., 1993). Transgenic plants were also produced using a control construct which contained the GUS reporter under control of the chimeric promoter but from which the ACE1 coding region had been omitted. Clonal replicates were used in the experiments to allow direct comparisons of data to be made, since variation of expression due to differing sites of insertion into the tobacco genome could be avoided. Non-transformed plants gave an apparent fluorometric GUS activity of 20 pmol 4-methylumbelliferyl-β-glucuronide cleaved min21 mg21 protein (20 units mg21) (Fig. 5.2). This background activity, which was clearly not due to GUS expression, was identical in the presence or absence of copper

05 Inducible Gene 05

3/16/99 11:12 AM

Page 64

V.L. Mett and P.H.S. Reynolds

64

GUS activity (pmol min–1 mg–1 protein)

1200

900

600

300

0 Copper treatment Plant

(–) Wildtype

(+)

(–)

(+)

Controlconstruct

(–)

(+)

Fullconstruct

Fig. 5.2. Copper responsiveness of gene expression. Wild-type plants are nontransformed Nicotiana tabacum. Control-construct plants are N. tabacum transformed with a construct which contained the GUS reporter gene under control of a chimeric promoter consisting of a copy of the MRE fused to domain A of the CaMV 35S RNA promoter, but which does not contain the ace1 gene. Fullconstruct plants are N. tabacum transformed with a construct which contains the ace1 gene under control of the CaMV 35S RNA promoter, together with the GUS reporter under control of the chimeric promoter. Following an acclimatization period of 7 days after transfer of plants from agarose to solution culture, CuSO4 was added to the nutrient solution to a final concentration of 50 µM. After 5 days these plants (+) and plants grown in the absence of copper (2) were harvested and the leaves assayed fluorometrically for GUS activity.

ions. The control-construct plants showed the GUS assay background of 20 units mg21 in the absence of copper but, after 5 days in the presence of 50 µM copper, these plants showed a GUS activity of 40 units mg21 (Fig. 5.2). In contrast, the full-construct plants, grown in the presence of 50 µM CuSO4 for 5 days, showed an increase in leaf GUS specific activity to 1200 units mg21. Northern analysis confirmed the correct functioning of the whole yeast metallo-regulatory system transferred into the plants. The ace1 gene transcript was constitutively present, whereas the GUS transcript was detectable only in the presence of inducing copper ion concentrations. The fact that no copper induction was observed in plants transformed with the control-construct which did not contain the ace1 regulatory gene suggested that the yeast transcription

05 Inducible Gene 05

3/16/99 11:12 AM

Page 65

Tissue-specific, Copper-controllable Gene Expression

65

factor was essential for the functioning of the system in plants and that its nuclear targeting was unaffected in the plant background. The activity of the described chimeric promoter was shown to be directly dependent on the copper ion concentration. The copper ion concentration required for activation (in our hands 5 µM or higher) was shown to be significantly above that usually found in plant nutrient solutions (for example, standard MS medium contains copper at a concentration of 0.1 mM). Maintenance of plants for extended periods in the presence of the ‘inducing’ copper ion concentration in the nutrient solution resulted in development of copper toxicity symptoms. This problem could be circumvented by the application of copper ions in a foliar spray. The concentration of copper required in these sprays was considerably lower at 0.5 µM. It was further shown that if, following activation of GUS expression by addition of copper to the nutrient solution (or its application as a foliar spray), copper was then removed from the system then expression of the GUS gene was repressed; that is, the system could be used in experiments demanding precise timing of expression. Data showing the time-course of activation and repression of GUS expression in response to the addition and removal of 50 µM CuSO4 from the nutrient solution is shown in Fig. 5.3. GUS activity before the addition of copper was 48 units mg21. A twofold increase in specific activity was observed after 24 h, increasing to 1030 units mg21 after 4 days. Removal of CuSO4 from the nutrient solution resulted in a dramatic decrease in GUS activity to 80 units mg21 after 4 days. The same induction/repression was also observed in plants which had copper applied as a foliar spray. If leaves were sprayed to drip point daily with a 0.5 µM CuSO4 solution, maximal induction occurred after 5 days. If plants were thereafter sprayed to drip point daily with water, GUS activity decreased to background levels after a further 5 days. When plants were sprayed only once to drip point with the 0.5 µM copper solution, GUS activity was induced within 5 days and remained high for a further 7 days, before decreasing to background levels. In experiments using Arabidopsis, foliar application of 5 µM CuSO4 gave maximal induction of a GUS reporter gene after 4 days, but the period of maximal induction was short lived (F. Johnson-Potter, Australia, 1997, personal communication).

MODIFICATION OF THE SYSTEM TO OVERCOME BACKGROUND EXPRESSION IN ROOTS Whilst the system described above showed a very low background activity in the uninduced state in the leaves of transgenic plants, activity of the reporter enzyme GUS in roots was significant, even in half-strength MS growth medium with a concentration of CuSO4 (0.05 µM), which is below the induction threshold observed in leaves. Apparently this resulted from the presence of the ASF1 transcription factor binding site, which lies within the 35S promoter 90

05 Inducible Gene 05

3/16/99 11:12 AM

Page 66

V.L. Mett and P.H.S. Reynolds

66

1200

GUS activity (pmol min–1 mg–1 protein)

1000

800

600 Add copper

400

Remove copper

200

0 (Cu)–

1

2 Days

4

2

4 Days

Fig. 5.3. Time-course of activation/repression of gene expression. Following acclimatization, duplicate clonal Nicotiana tabacum plants transformed with the full construct (see Fig. 5.2) were harvested and the leaves assayed for GUS activity. The nutrient solution of the remaining ten plants was adjusted to contain 50 µM CuSO4, and duplicate plants were harvested after 1, 2 and 4 days, and leaves were assayed for GUS activity. The nutrient solution of the remaining plants was then changed to one lacking copper; duplicate plants were harvested, and the leaves were assayed for GUS activity after a further 2 and 4 days.

base pair (bp) domain A (Lam et al., 1989). Indeed, it has been shown that the 35S promoter 90 bp domain A is sufficient for low level constitutive expression in roots (Benfey and Chua, 1990). To eliminate this background expression of the system in the roots the ASF1 binding site was removed from the chimeric promoter leaving the 35S promoter 246 bp TATA fragment only. In addition, the effect of increasing the number of repeats of the MRE (metal regulatory element) fused in tandem to the TATA fragment (Mett et al., 1996) was investigated. Three variants of the chimeric promoter containing one, two or four copies of the MRE were constructed to evaluate the effect of the number of MRE on the level of expression. Due to the very close position of the expression cassette to the ace1 coding region under control of the potent CaMV 35S RNA promoter, the influence of the orientation (D (direct) or R (reverse)) of the chimeric promoter with respect to the direction of ace1 transcription was also investigated.

05 Inducible Gene 05

3/16/99 11:12 AM

Page 67

Tissue-specific, Copper-controllable Gene Expression

67

Tobacco plants from transformations in which the various chimeric promoter constructs were fused to the GUS reporter coding sequence, were used in an experiment in which GUS activity was measured in extracts from the roots and leaves of solution cultured plants before and after addition of copper to the medium. All plants showed very low GUS activity (typically between 20 and 100 units mg21) in the leaves and roots, in the absence of copper. Following induction with 50 µM CuSO4, all plants showed elevated GUS levels in the roots. However, there was no significant difference between plants for measured GUS activity regardless of the numbers of MRE present and, by inference, for the levels of activity of chimeric promoters containing one, two or four copies of MRE. This result could be explained by incorrect spacing between individual elements in comparison with the spacing of the two MRE in the native yeastmetallothionein promoter (Thiele and Hamer, 1986) or, despite high transcription, be due to low concentrations of ace1 protein actually being present in the nucleus. The lack of effect of orientation suggested low impact of the adjoining full 35S promoter which drives ace1 transcription. Over the total experiment, ‘4R’-transformed plants gave the highest GUS activity in roots after induction and the lowest background activity in the uninduced state, making this variant

Fig. 5.4. Copper-controllable gene expression system: vectors for convenient use. pPMB 768 and pPMB 7066 are pUC-based plasmids allowing for cloning of the gene of interest behind the desired chimeric promoter. pPMB 768 contains four tandem copies of the metal regulatory element (MRE) and the 246 bp fragment of the CaMV 35S RNA promoter (TATA), whereas pPMB 7066 contains only one copy of the MRE fused to domain A of the CaMV 35S promoter. pPMB 765 is a binary vector containing, in addition to the selectable marker gene and NotI site for cloning the gene of interest, the ace1 gene under control of the full CaMV 35S RNA promoter. The pPMB 7088 vector contains a promoterless ace1 gene with a HindIII site for cloning the desired tissue/organ-specific promoter.

05 Inducible Gene 05

68

3/16/99 11:12 AM

Page 68

V.L. Mett and P.H.S. Reynolds

the best candidate for copper-controllable gene expression in the roots of transgenic plants (see Fig. 5.4; pPMB 768/pPMB 765). The ‘4R’ tobacco plants were further investigated using four independent transformants (4R1, 4R2, 4R3, 4R4). Copper-induction experiments were performed on three re-regenerated clones from each independent transformant. One set of three clones was used as a minus copper control and a second set of clones was used to measure GUS activity after 5 days treatment with 50 µM CuSO4 in the nutrient solution. The third set of three clones was assayed for GUS activity 5 days after copper was removed from the nutrient solution. As shown in Table 5.1, very low GUS activity was observed in the roots before induction by the addition of 50 µM CuSO4 to the nutrient solution ((2) CuSO4). Five days after the addition of copper to the nutrient solution an up-to-160fold increase in GUS activity was observed ((+) CuSO4). When copper was then removed from the nutrient solution, a significant drop in GUS activity was observed after 5 days ((±) CuSO 4). These results clearly demonstrate that elimination of the ASF1-binding site has allowed tight control of expression in the roots. However, in all of these experiments using transgenic tobacco plants, very low GUS activity was measurable in the leaves before and after the induction (less than 50 units mg21). The introduction of an ASF2-binding site between MRE and TATA in the chimeric promoters did not restore expression in the leaves (data not shown, New Zealand, 1997), in spite of the fact that the ASF2binding site confers leaf expression when fused to domain A of the 35S promoter (Lam et al., 1989). It is therefore highly significant that recent experiments in Arabidopsis (Fumiaki Katagiri, USA, 1998, personal communication), using the pPMB 768/765 system (see Fig. 5.4) have allowed tight, copper-inducible control in leaves. Also, experiments performed in Arabidopsis seedlings with the pPMB 768/765 system using a GUS reporter gene, have revealed tight copperdependent expression both in leaves and in roots (S. Kurup and M. Holdsworth, UK, 1997, personal communication).

USE OF THE COPPER SYSTEM TO EFFECT CONTROL OVER PLACE, AS WELL AS TIME, OF EXPRESSION The ‘ideal’ gene expression system must provide both temporal and spatial control of a ‘gene of interest’ in transgenic plants. Tissue-specificity was introduced into the copper-controllable gene expression system by the use of a tissue-specific promoter to effect spatial control of the expression of the ACE1 transcription factor (Mett et al., 1996). As we were interested in the development of a tissue-specific coppercontrollable expression system for use in the nitrogen-fixing nodules of leguminous plants, the promoter of the nod45 gene from lupin (Rice et al., 1993) was used to drive the expression of the ace1 gene. The product of the nod45 gene, a late nodulin

05 Inducible Gene 05

3/16/99 11:12 AM

Page 69

Tissue-specific, Copper-controllable Gene Expression

69

Table 5.1. Copper inducibility of GUS expression in the roots of transgenic tobacco plants. GUS activitya (nmol min21 mg21 protein) Plant

(2) CuSO4

(+) CuSO4

(±) CuSO4

4R1 4R2 4R3 4R4

15.0 21.0 20.0 7.5

1665 820 1391 1216

140 120 85 20

a

Values given are the average of three clonal replicates for each treatment.

with unidentified function, is localized only in nitrogen-fixing nodules and activity of the promoter has been shown to be associated strictly with developed nodules (MacKnight et al., 1995). To demonstrate the feasibility of this approach, the GUS-expression cassette containing a chimeric promoter with four copies of the MRE was cloned into a plasmid containing the ace1 gene under control of the nod45 promoter in the reverse orientation with respect to the direction of ace1 transcription (pACENOD). A second construct containing the same GUS-expression cassette and the ace1 gene driven by the constitutive 35S promoter (pACE-in-ART), was used as a control. Both constructs were tested in ‘transgenic roots–wild-type tops’ Lotus corniculatus plants. After the development of nitrogen-fixing nodules (4 weeks) the plants were transferred into liquid culture and copper-induction experiments were performed. As shown in Fig. 5.5b histochemical GUS staining was localized only in nodules when expression of the ace1 gene was controlled by the nod45 promoter, whilst when transgenic roots were produced using the pACE-in-ART vector, GUS staining was also present in the tips of secondary roots (Fig. 5.5a). The apparent absence of GUS activity in the rest of the root tissue could be explained by the higher level of GUS expression in metabolically active cells of root tips. These results clearly demonstrated that the use of a specific promoter to limit the expression of the transcription factor ACE1 to a particular tissue/organ to be compatible with the copper-controllable system and to provide a mechanism to introduce tissue/organ-specificity into this system. Theoretically, it seems possible that the combination of an appropriate tissue/organ-specific promoter to drive the expression of the metallo-regulatory transcription factor ACE1 together with the elements of the chimeric promoter could be applicable to the temporal and spatial control of gene expression in any plant organ or tissue, provided that a TATA domain in the chimeric promoter has the capability to support expression in all tissues of the particular plant being used.

05 Inducible Gene 05

70

3/16/99 11:12 AM

Page 70

V.L. Mett and P.H.S. Reynolds

Fig. 5.5. Spatial control of GUS reporter gene expression in transgenic roots of Lotus corniculatus using the copper system. (a) Histochemical localization of GUS reporter gene activity in both roots and nodules of transgenic L. corniculatus roots transformed with the pPMB 768/pPMB 765 or ‘pACE-in-ART’ construct. (b) Histochemical localization of GUS reporter gene activity in nodules only of transgenic L. corniculatus roots transformed with the pPMB 768/pPMB 7088 or ‘pNOD-ACE’ construct in which expression of ACE1 is under the control of the nodule-specific nod45 promoter.

THE ‘EASE OF USE’ VECTORS FOR COPPER-CONTROLLABLE GENE EXPRESSION The copper-controllable gene expression system has been formulated for convenient use in four basic vectors (Fig. 5.4). Two of these are pUC-based plasmids allowing for cloning of the gene of interest and two are binary vectors for transfer to plants, one of which allows for control of the ace1 gene by an organ-specific promoter.

pPMB 768 This is a pUC119-based plasmid containing four copies of the MRE fused to the 246 bp fragment from the CaMV 35S RNA promoter, separated from a nos terminator by a cloning cassette. Following cloning of a gene of interest, the

05 Inducible Gene 05

3/16/99 11:12 AM

Page 71

Tissue-specific, Copper-controllable Gene Expression

71

sequence can then be excised NotI and cloned into the appropriate binary vector (pPMB 765 (p-ACE-in-ART) or pPMB 7088). In our experience, genes cloned under control of the chimeric promoter in this plasmid give inducible expression in roots with no expression in the absence of inducing copper concentrations. However, we see no leaf expression in tobacco, although there is clearly induction in leaves (F. Katagiri, Maryland, USA, 1998, personal communication) from this construct in Arabidopsis.

pPMB 7066 This is a pUC119-based plasmid containing one copy of the MRE fused to the 290 bp domain A from the CaMV 35S RNA promoter, separated from a nos terminator by a cloning cassette. Following cloning of a gene of interest, the sequence can then be excised NotI and cloned into the appropriate binary vector (pPMB 765 (pACE-in-ART) or pPMB 7088). Genes cloned under control of the chimeric promoter in this plasmid will give background expression in roots (at least in tobacco) in the absence of copper due to the 290 bp 35S promoter. Full control of expression has been obtained in leaves with no background expression in the absence of inducing copper concentrations. We have noticed a rather low percentage of tobacco transformants which demonstrate copper-inducible phenotype. At present, we cannot explain this phenomenon which results in the need to analyse a large number of primary transformants in order to find the appropriate phenotype.

pPMB 765 (pACE-in-ART) This is a binary vector for transfer to plants. It is based on the pART system of Gleave (1992). A NotI site allows for cloning of the gene of interest under control of the chosen chimeric promoter (from pPMB 768 or pPMB 7066). The ace1 gene is constitutively expressed throughout the plant and so gives expression of the induced gene in all plant organs under copper inducing conditions.

pPMB 7088 This is a binary vector for transfer to plants. It is based in the pART system of Gleave (1992). A NotI site allows for cloning of the gene of interest under control of the chosen chimeric promoter (from pPMB 768 or pPMB 7066). A HindIII site is provided 5′ to the ace1 gene to allow cloning of an organ-specific promoter. In this way the ace1 gene will be expressed only in the plant organ defined by the introduced promoter region and will give expression of the introduced gene (under inducing copper concentrations) only in that organ.

05 Inducible Gene 05

72

3/16/99 11:12 AM

Page 72

V.L. Mett and P.H.S. Reynolds

All the vectors are provided in an E. coli DH5a background. pPMB 768 and pPMB 7066 grow on LB amp, 100 µg ml21. pPMB 765 and pPMB 7088 grow on LB spec, 100 µg ml21.

USE OF THE SYSTEM FOR TISSUE-SPECIFIC ANTISENSE EXPERIMENTS Functional use of the system was demonstrated by its ability to drive effective antisense constructs in an organ-specific manner. The nodule-specific system was used to express antisense constructs of aspartate aminotransferase-P2 (AAT-P2) in transgenic L. corniculatus plants (Mett et al., 1996). Aspartate aminotransferase plays a key role in plant carbon and nitrogen metabolism. It exists as at least two isoenzymic forms in nodules and one of these, AAT-P2, is thought to function as part of the pathway which assimilates ammonia produced by the nitrogen fixation process (Reynolds and Farnden, 1979). Controlled, organ-specific expression of the antisense construct of this isoform offered the possibility of an in vivo demonstration of its direct role in the assimilation of ammonia into the amino acid asparagine. Three antisense constructs (7048, 7049 and 7050), derived from the lupin AAT-P2 cDNA (Reynolds et al., 1992), were expressed in transgenic L. corniculatus plants using the pACE-NOD vector (see Fig. 5.4 above and results in Table 5.2). Of the three constructs, 7050 gave the most effective antisense effect, with AAT-P2 enzyme activity below the level of detection in two of the three plants tested. In plant 7050-1 there was a 77% decrease in nodule asparagine concentration. In the 7049 plants, an antisense effect on AAT-P2 activity was seen in all three plants. However, in only one of these plants, 7049-1, was there a dramatic decrease in nodule asparagine concentration. This could be due to sufficient AAT-P2 activity remaining in the other plants to allow unimpeded asparagine synthesis. Full analysis of the 7048 plants was compromised by the lack of nodules in plant 7048-3(+) and the lack of an AAT-P2 determination in 7048-1(+), due to low nodule-weight. However, a significant antisense effect on AAT-P2 enzyme activity was seen in plant 7048-2 with only 18% of the AAT-P2 activity remaining after copper induction. A dramatic effect of AAT-P2 antisense expression on the nodule asparagine concentration was also observed, with 83% and 91% reductions observed in plants 7048-1 and 7048-2, respectively. Consistently, across the whole experiment, plants with very low or undetectable AAT-P2 activity showed large decreases in nodule asparagine concentration. In plants where the antisense effect was not high, or where significant residual levels of AAT-P2 activity (for example, plant 7049-3) remained, asparagine levels were either unaffected or only slightly reduced. In untransformed L. corniculatus plants there was no effect of copper on the activity of AAT-P2, and the asparagine levels in the nodules of these plants were comparable to those levels seen in the nodules of transformed plants which had not been exposed to antisense expression-inducing levels of copper.

05 Inducible Gene 05

3/16/99 11:12 AM

Page 73

Tissue-specific, Copper-controllable Gene Expression

73

Table 5.2. Effect of expression of AAT-P2 antisense constructs on nodule AAT-P2 enzyme activity and on nodule asparagine levels. AAT-P2 activity (mmol min21 g21 nodules) Plant 7048–1 7048–2 7048–3

Asparagine (mmol g21 nodules)

(+) Cu

(2) Cu

(+) Cu

(2) Cu

nd 0.051 No nodules

0.236 0.284 0.663

4.9 1.8 No nodules

28.6 19.1 40.4

0.593 0.609 1.042

8.7 28.8 36.3

37.6 44.9 34.3

0.342 No nodules 0.294

4.0 7.6 8.8

17.3 No nodules 19.4

7049–1 7049–2 7049–3

0.080 0.368 0.175

7050–1 7050–2 7050–3