The Molecular Biology and Biotechnology of Flowering [Second ed.] 9781845930424, 1-84593-042-8

Citation preview

THE MOLECULAR BIOLOGY OF FLOWERING Second Edition

AND

BIOTECHNOLOGY

This page intentionally left blank

THE MOLECULAR BIOLOGY AND BIOTECHNOLOGY OF FLOWERING Second Edition

Edited by

Brian R. Jordan Lincoln University Canterbury, New Zealand

CABI is a trading name of CAB International CABI Head Office Nosworthy Way Wallingford Oxfordshire OX10 8DE UK Tel: þ44 (0)1491 832111 Fax: þ44 (0)1491 833508 E-mail: [email protected] Website: www.cabi.org

CABI North American Office 875 Massachusetts Avenue 7th Floor Cambridge, MA 02139 USA Tel: þ1 617 395 4056 Fax: þ1 617 354 6875 E-mail: [email protected]

ßCAB International 2006. 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. A catalogue record for this book is available from the Library of Congress, Washington, DC. The molecular biology and biotechnology of flowering / edited by Brian R. Jordan. -- 2nd ed. p. cm. Rev. ed. of: The molecular biology of flowering. c1993. Includes bibliographical references. ISBN-13: 978-1-84593-042-4 (alk. paper) ISBN-10: 1-84593-042-8 (alk. paper) 1. Plants, Flowering of. 2. Plant molecular biology. 3. Plant biotechnology. I. Jordan, Brian R. II. Molecular biology of flowering. III. Title. QK830.M64 2006 575.6--dc22 2005017931

ISBN-10: 1-84593-042-8 ISBN-13: 978-1-84593-042-4 Typeset by SPI Publisher Services, Pondicherry, India Printed and bound in the UK by Biddles Ltd, King’s Lynn.

Contents

Contributors

vii

Preface

ix

Part I: External and Internal Regulation of Flowering

1

1.

Photoperiodism and Flowering B. Thomas, I. Carre´ and S. Jackson

3

2.

Vernalization A.R. Gendall and G.G. Simpson

26

3.

Signal Transduction Regulating Floral Development R.G. Anthony

50

Part II: Floral Development

79

4.

Floral Meristem Identity Genes B. Davies

81

5.

Molecular Biology of Floral Organogenesis B. Krizek

100

6.

Molecular Developmental Genetics and the Evolution of Flowers G. Theissen and K. Kaufmann

124

7.

Flower Senescence S. Verlinden

150

8.

Developmental Control and Biotechnology of Floral Pigmentation K. Davies and K. Schwinn

178

v

vi

Contents 9.

Biotechnology of Floral Development C. Winefield and B.R. Jordan

237

Part III: Fertilization and Gametophyte Development

267

10.

Control of Fertilization by Self-incompatibility Mechanisms T. Gaude, I. Fobis-Loisy and C. Mie`ge

269

11.

Stamen Development: Primordium to Pollen R.J. Scott, M. Spielman and H.G. Dickinson

298

12.

Genes Regulating Ovule Development J. Broadhvest and B.A. Hauser

332

13.

The Molecular Biology of Apomixis R.A. Bicknell and A.S. Catanach

354

Index

391

Contributors

Anthony, R.G., School of Biological Sciences, Royal Holloway, University of London, Egham, Surrey TW20 0EX, UK; e-mail: [email protected] Bicknell, R.A., New Zealand Institute for Crop and Food Research Ltd, Private Bag 4704, Christchurch, New Zealand; e-mail: [email protected] Broadhvest, J., Bayer BioScience NV, Technologie Park 38, B-9052 Ghent, Belgium; e-mail: [email protected] Carre´, I., Department of Biological Sciences, University of Warwick, Coventry, CV4 7AL, UK; e-mail: [email protected] Catanach, A., New Zealand Institute for Crop and Food Research Ltd, Private Bag 4704, Christchurch, New Zealand; e-mail: [email protected] Davies, B., Centre for Plant Sciences, University of Leeds, Leeds LS2 9JT, UK; e-mail: [email protected] Davies, K., New Zealand Institute for Crop and Food Research Ltd, Private Bag 11600, Palmerston North, New Zealand; e-mail: [email protected] Dickinson, H.G., Department of Plant Sciences, South Parks Road, Oxford, OX1 3RB, UK Fobis-Loisy, I., Reproduction et De´veloppement des Plantes, Ecole Normale Supe´rieure de Lyon, UMR 5667 CNRS-INRA-ENSL-UCB, Lyon I, IFR 128 BioSciences Lyon-Gerland, 46 alle´e d’Italie, 69364, Lyon cedex 07, France; e-mail: [email protected] Gaude, T., Reproduction et De´veloppement des Plantes, Ecole Normale Supe´rieure de Lyon, UMR 5667 CNRS-INRA-ENSL-UCB, Lyon I, IFR 128 BioSciences Lyon-Gerland, 46 alle´e d’Italie, 69364, Lyon cedex 07, France; e-mail: [email protected] Gendall, A.R., Department of Botany, La Trobe University, Bundoora, Victoria, Australia; e-mail: [email protected] Hauser, B.A., Department of Botany, University of Florida, Gainesville, FL 32611-8526, USA; e-mail: [email protected]

vii

viii

Contributors

Jackson, S., Warwick HRI, University of Warwick, Wellesbourne, Warwick, CV35 9EF, UK. Jordan, B.R., Cell Biology Group, Agriculture and Life Sciences Division, Lincoln University, PO Box 84, Canterbury, New Zealand; e-mail: [email protected] Kaufmann, K., Friedrich-Schiller-Universita¨t Jena, Lehrstuhl fu¨r Genetik, Philosophenweg 12, D-07743, Jena, Germany. Krizek, B., Department of Biological Sciences, University of South Carolina, Columbia, SC 29208, USA; e-mail: [email protected] Mie`ge, C., Reproduction et De´veloppement des Plantes, Ecole Normale Supe´rieure de Lyon, UMR 5667 CNRS-INRA-ENSL-UCB, Lyon I, IFR 128 BioSciences Lyon-Gerland, 46 alle´e d’Italie, 69364, Lyon cedex 07, France. Schwinn, K., New Zealand Institute for Crop and Food Research Ltd, Private Bag 11600, Palmerston North, New Zealand. Scott, R.J., Department of Biology and Biochemistry, University of Bath, Bath BA2 7AY, UK; e-mail: [email protected] Simpson, G.G., Dundee University, c/o Gene Expression Programme, SCRI, Invergowrie, Dundee, DD2 5DA, UK; e-mail: [email protected] Spielman, M., Department of Biology and Biochemistry, University of Bath, Bath, BA2 7AY, UK. Theissen, G., Friedrich-Schiller-Universita¨t Jena, Lehrstuhl fu¨r Genetik, Philosophenweg 12, D-07743 Jena, Germany; e-mail: guenter.theissen @uni-jena.de Thomas, B., Warwick HRI, University of Warwick, Wellesbourne, Warwick, CV35 9EF, UK; e-mail: [email protected] Verlinden, S., Division of Plant and Soil Sciences, West Virginia University, PO Box 6108, Morgantown, WV 26506-6108, USA; e-mail: [email protected] Winefield, C., Agriculture and Life Sciences Division, Lincoln University, PO Box 84, Canterbury, New Zealand; e-mail: [email protected]

Preface

In 1993 The Molecular Biology of Flowering was published. At that time the physiological events associated with flowering had been well characterized and yet the underlying molecular mechanisms remained unknown. This book captured the spirit of molecular research that was beginning to provide a greater understanding of the flowering process. In the late 1980s and early 1990s, homeotic genes such as floricaula ( flo) had been isolated and established as playing important roles in the transition of the vegetative to the floral apex. Furthermore, genes that are involved in organogenesis had been identified and provided insight for the development of the ABC model of floral morphogenesis. The advent of molecular approaches opened new avenues of research and seemed to provide the ‘Dawn of a New Age’. This second edition of the book (with slightly modified title) documents the progress that has been made since those early days. In developing the overall theme for the book, I have tried to update the research for subject areas covered by the original book and, whenever possible, involve the same authors. Using this approach I hope to allow the reader to appreciate the developments that have taken place in the last 12 years. Two examples will emphasize this point. In Chapter 1, Brian Thomas et al. cite recent molecular studies that provide convincing evidence that F T mRNA induced in the leaf is then transferred to the apex where it initiates flowering. Has the molecular mechanism of florigen or anti-florigen action now been discovered after 80 years of investigation? In Chapter 11, Rod Scott et al. state ‘Twelve years ago little was known about genetic control of stamen development beyond specification of stamen primordia by floral homeotic genes. New footholds have been established in several areas, notably patterning of the microsporangium, regulation of meiosis and anther dehiscence’. This increase in knowledge, both in breadth and depth, is clearly reflected in these reviews. In addition, it is very apparent that since 1993, molecular genetics has been applied to a much wider range of related areas. Thus, topics such as the evolution of flowers, floral senescence and ix

x

Preface

apomixis have been included in the new edition. These reviews add different perspectives and provide further stimulus to investigate and enhance our understanding. To some extent I feel research on flowering has not had the prominence it deserves; molecular biology and the commercial opportunities of biotechnology have started to redress this situation. There is still much to learn, but I believe that this volume will contribute substantially to increase the understanding of this vitally important process. Most importantly, I would like to thank the authors for their time and considerable effort in producing these excellent reviews. In a period when scientists are under considerable work pressure, with a variety of demands on their time, I am very appreciative of their willingness to contribute their knowledge and expertise to this volume. It is also particularly rewarding for me to read the contributions from colleagues working on flowering from all parts of the world. Thanks, once again, to Tim Hardwick of CABI for his support. Finally, I would like to thank my invaluable assistant, Bronwyn Hamilton, without whose patient endeavours the smooth preparation of this book would not have taken place. Brian R. Jordan Professor of Plant Biotechnology Lincoln University New Zealand September 2005

I

External and Internal Regulation of Flowering

This page intentionally left blank

1

Photoperiodism and Flowering B. THOMAS,1 I. CARRE´ 2 AND S. JACKSON1 1

Warwick HRI, University of Warwick, Wellesbourne, Warwick, CV35 9EF, UK; 2Department of Biological Sciences, University of Warwick, Coventry, CV4 7AL, UK

Introduction Photoperiodism can be defined as the response to changes in daylength that enables plants (or any other living organism) to adapt to seasonal changes in their environment. Except at the equator, the passage of the year is marked by a continuous but highly reproducible variation in the length of the day. In order to locate the time of year accurately, a timekeeping mechanism operates with precision as part of the plant’s photoperiodic sensing mechanism in a way that is insensitive to less predictable variations in the environment such as temperature. Photoperiod alone is not an unambiguous signal as any particular daylength occurs twice in an annual cycle. Progressive changes in daylength, which are at their greatest around the equinoxes in spring and autumn, do, however, provide a certain environmental signal for the passage of the seasons (Thomas and Vince-Prue, 1997). The seasonal range and rate of change of daylength is lower in the tropics than at higher latitudes and photoperiodic mechanisms need to be sufficiently precise and flexible to operate across the entire range of daylengths. The ability to detect seasonal change and respond to it confers a selective advantage to plants because it provides a means of anticipating, and consequently preventing, the adverse effects of a particular seasonal environment. Photoperiodic plants are common, even in tropical latitudes where the seasonal daylength changes are small, and daylength is used to synchronize reproductive or other activities with seasonal events such as dry or rainy periods. Coincident flowering in members of a population increases the chances of outbreeding and hence genetic recombination. For this reason, synchronization of floral initiation through photoperiodic sensitivity can confer advantages independently of whether reproduction is matched with a particular favourable environment. A further potential benefit of photoperiodic responses is that they can enable organisms to occupy an ecological niche in space and time. For example, ßCAB International 2006. The Molecular Biology and Biotechnology of Flowering, 2nd edn (ed. B.R. Jordan)

3

4

B. Thomas et al.

a response to short days can enable a woodland plant to flower and seed before the dense leaf canopy is formed in the spring. Photoperiodic control of flowering is also important in agriculture and horticulture. Breeding to extend latitudinal range or altered timing of flowering involves understanding and exploiting variation in the photoperiodic responses of a particular species. Daylength manipulation in order to schedule flower production is also a common practice. This chapter deals with mechanisms underlying the photoperiodic control of flowering and is divided into three sections: (i) the first describes the physiological background to molecular and genetic studies of the photoperiodic control of flowering; (ii) the second details recent progress in understanding the genetic regulatory networks controlling flowering in response to daylength; and (iii) the third covers the molecular and genetic basis of the underlying timekeeping mechanisms for plant photoperiodism.

Physiology of Photoperiodism Discovery and variation The response of plants to the daily duration of light was proposed independently by Julien Tournois and Hans Klebs at the beginning of the 20th century. It was the American physiologists Garner and Allard (1920), however, who first saw clearly that flowering and many other responses in plants could be accelerated either by long days (LDs) or by short days (SDs), depending on the plant. They were led to their discoveries by studies on ‘Maryland Mammoth’ variety of tobacco that failed to flower and grew large in summer, while plants grew under glass during winter and early spring, and soybean, in which for a particular variety, flowering tended to occur at the same time in the field, irrespective of planting date. After experimentally eliminating temperature and light intensity as causal factors they concluded that the tobacco and soybean plants would only flower if the duration of the daylight period was sufficiently short. They introduced the terms photoperiod and photoperiodism and classified plants into the photoperiodic groups in use today. Short-day plants (SDPs) are those that flower or in which flowering is accelerated by days which are shorter than a critical daylength. Long-day plants (LDPs) are plants that flower or in which flowering is accelerated when the daylight period exceeds a critical daylength. Plants that flower at the same time irrespective of the photoperiodic conditions are called day-neutral plants (DNPs). Plants that respond to daylength can be further subdivided into obligate (or qualitative) types, where a particular daylength is essential for flowering, or facultative (or quantitative) types, where a particular daylength accelerates but is not essential for flowering. Many important crop species are potentially photoperiodic, e.g. many cereals such as wheat and barley are LDPs, while SDPs include rice and soybean (see Table 1.1). The model plant for molecular genetic studies, Arabidopsis thaliana, is a typical facultative LDP under this classification.

Photoperiodism and Flowering

5

Table 1.1. Examples of LDP and SDP, including some important crop species. (Modified from Thomas and Vince-Prue, 1997.) SDP (qualitative or absolute)

SDP (quantitative)

Coffea arabica Fragaria  ananassa Glycine max Biloxi Hibiscus cannabinus; esculentus Humulus japonicus; lupulus Kalanchoe¨ blossfeldiana Nicotiana tabacum Maryland Mammoth Oryza sativa Perilla (red) Pharbitis nil Xanthium strumarium

Cannabis sativa Cucumis sativus (some cvs) Glycine max Gossypium Helianthus annuus; tuberosus Nicotiana tabacum Oryza sativa Rhododendron spp. (florist’s azalea) Ricinus communis Rosa gallica; rugosa Sesamum indicum Sorghum bicolor ; halepense Zea mays

LDP (qualitative or absolute) Avena sativa (spring strains) Brassica juncea carinata; pekinensis Delphinium elatum, garden hybrids Fuchsia hybrida

LDP (quantitative) Allium ampeloprasum Antirrhinum majus Arabidopsis thaliana Dianthus carthusianorum Napoleon III; caryophyllus (glasshouse carnations) Eustoma grandiflorum Hemerocallis fulva Hordeum vulgare (spring strains) Linum usitatissimum Lolium temulentum Ba 3081 Medicago sativa Petunia hybrida Pisum sativum Secale cereale (spring cvs) Solanum tuberosum Triticum aestivum (spring cvs) Vicia faba

Hyoscyamus niger (annual) Jasminium grandiflorum Lemna gibba; minor Lolium temulentum Ceres Nicotiana sylvestris Papaver somniferum Silene armeria; coeli-rosa Trifolium pratense (English Montmorency)

The length of the day and night are mutually linked within the 24-h daily cycle. Photoperiodic responses could therefore be theoretically determined by either the length of the day or the length of the night. Classic experiments with SDP Xanthium revealed that flowering only occurred if the night length was greater than 8.5 h, irrespective of the relative durations of light and darkness in the experimental cycle (Hamner and Bonner, 1938). SDs did not cause flowering if they were coupled with short nights but when the night was sufficiently long, flowering occurred even when the accompanying light periods were long. However, although a sufficiently long dark period appeared to be the decisive factor for flowering to occur, the level of flowering was also affected by the length of the light period. This indicated that the interaction between light and darkness formed part of the daylength-sensing mechanism. If a long night is

6

B. Thomas et al.

interrupted by a short (e.g. 30 min) period of light (or night break, NB) near the middle, SDPs respond as if they have been exposed to an LD. For LDPs, such NBs are only effective if given in combination with daylengths that are just longer than those needed to permit flowering or if they are of several hours duration. Also, in LDPs, unlike in SDPs, the amount and spectral composition of the light given during the day period, especially in the latter part, has a large effect on flowering. If the response to daylength depends primarily on the length of the dark period, the plants are called dark-dominant and conversely, if the light period is the main influence, they are called light-dominant (Thomas and Vince-Prue, 1997). In general, most SDPs are dark-dominant and most LDPs are light-dominant.

Transmissible signals A common feature of photoperiodism appears to be that daylength perception is a separate process from the response to photoperiod. When either the leaves or the shoot tips of photoperiodically sensitive plants are exposed to different daylengths, flowering depends on the daylength given to the leaves and not to the apex (Knott, 1934). In several instances, leaves from plants, which have been given a daylength treatment that initiates flowering have been grafted on to plants that have not been exposed to permissive daylengths, with the result being flowering in the receptor plants. Daylength therefore is perceived in leaves and results in a localized change in the properties of that leaf. Flowering then occurs as a result of a signal transmitted from the leaves to the apex. The change in the leaf is termed induction, while the response at the apex leading to the initiation of flowering is sometimes called evocation. From grafting experiments, it is known that in some species, induced leaves are independently capable of generating a flowering stimulus over many days or even weeks while in others, favourable cycles must be continued until the apex has become recognizably floral, indicating the need for a continual supply of a floral stimulus (presumably from the leaf) if flower development is to be sustained. An intact and functioning nutrient transport system is a requirement for communication between the sites of perception in the leaves and the sites of response. Daylength response is lost when the pathway is disrupted by removing the source leaf, inhibiting transport by localized heat or cold treatments applied at intermediate points in the transport path or stem girdling. Also, when leaves taken from an induced plant are grafted on to a receptor plant, promotion of flowering or other responses occurs only when a graft union has developed. It was proposed by Chailakhyan (1936) more than half a century ago that the signal passing between leaves and response sites is a specific flowering hormone: florigen. This idea was based on a series of experiments showing that grafting of leaves from one donor species to a separate receptor species could cause flowering (see Table 1.2). This strongly suggested the participation of common signals in different species. Other grafting experiments suggested that other substances inhibiting flowering may be involved; the appropriate day-

Photoperiodism and Flowering

7

Table 1.2. Examples of successful transfer of flowering stimulus between species or genera following grafting. (After Thomas and Vince-Prue, 1997.) Donor Intraspecific grafts Glycine max Agate Chenopodium rubrum 608 47 N Pisum sativum various genetic lines Interspecific grafts Gossypium hirsutum Nicotiana tabacum Delcrest Intergeneric grafts Blitum virgatum Chenopodium polyspermum Cucumis sativus Centauria cyanus

Response type

Receptor

Response type

DNP SDP* DNP or LDP*

G. max Biloxi C. rubrum 348 90 N P. sativum line G

SDP SDP LDP

DNP DNP

G. davidsonii N. sylvestris

SDP LDP

LDP SDP DNP LDP

Chenopodium rubrum Blitum capitatum Sicyos angulatus Xanthium strumarium

SDP LDP SDP SDP

length would then lead to removal of an antiflorigen rather than (or in addition to) synthesis of a floral hormone.

Genes Controlling Floral Initiation Pathways to flowering in Arabidopsis Flowering is generally regarded as a default process that will occur at some point in the plant’s life. The time that a plant flowers, however, is affected by many environmental and endogenous factors and consequently there are numerous genetic pathways that are involved in the control of flowering time. These pathways interact in different ways depending upon endogenous signals and the environmental conditions thus enabling the plant to flower in the most favourable conditions. Our understanding of the genetic and signalling pathways controlling flowering has increased dramatically over the last decade, based largely on the analysis of the flowering responses of winter- and springannual genotypes of A. thaliana, a facultative LDP. The predominance of the different pathways changes with the developmental state of the plant. Early on in the life cycle of the plant flowering is actively repressed to enable the plant to grow sufficiently large to be able to support the development of flowers, fruits and seeds. As the plant develops this repression is gradually lifted by what have been termed floral-enabling pathways such as the vernalization and autonomous pathways (Boss et al., 2004). In certain environments there is also activation of floral-promotion pathways such as the photoperiodic, gibberellin (GA), ambient temperature and lightquality pathways. At some stage the point is reached when promotion is

8

B. Thomas et al.

Floral-promotion pathways FRI Light quality

GA

Photoperiod FLC

Ambient temperature

Floral-enabling pathways Vernalization pathway Autonomous pathway

Floral-integrator genes LFY, FT, SOC1

Floral meristem identity genes Flower development

Fig. 1.1. Integration of the photoperiodic flowering pathway with other environmental and developmental pathways to flowering in Arabidopsis.

greater than repression and the flowering transition occurs. All the different flowering pathways converge on a small group of genes called floral-pathway integrators. These genes are FLOWERING LOCUS T (FT ), LEAFY (LFY ) and SUPPRESSOR OF OVEREXPRESSION OF CONSTANS 1 (SOC1), and they are responsible for the activation of the floral meristem identity genes that direct floral organ formation (see Fig. 1.1). Activation of the floral integrators will thus directly result in flowering. Expression of the floral-integrator genes is actively repressed by the floral inhibitor FLOWERING LOCUS C (FLC). Consequently flowering is prevented until this repression is lifted by the floral-enabling pathways, thus allowing activation of the floral-pathway integrators by the floral-promotion pathways. The levels of FLC are maintained at a high level by the FRIGIDA (FRI) gene. The activity of the FRI gene and consequently the resulting levels of FLC are major determinants in flowering time in Arabidopsis. Mutations in the FRI gene are responsible for most of the variation in flowering time observed in different ecotypes of Arabidopsis (Johanson et al., 2000). Loss-of-function mutations in the FRI gene are found in early flowering ecotypes of Arabidopsis, such as Landsberg erecta and Colombia. These mutations result in low FLC levels and only mild repression of the floral-pathway integrators. This low-level repression can be directly overridden by activation of a floral-promotion pathway, e.g. the photoperiodic pathway, without the need for a floral-enabling pathway to first lift the repression by FLC. In addition to FRI other genes are involved in the upregulation of FLC. These include EARLY IN SHORT DAYS 4 (ESD4), PHOTOPERIOD INDEPENDENT FLOWERING 1 (PIE1), EARLY FLOWERING IN SHORT DAYS (EFS) and VERNALIZATION INDEPENDENCE (VIP) genes. The mechanism of action of these genes is currently poorly understood.

Photoperiodism and Flowering

9

Floral-enabling pathways Two floral-enabling pathways result in the downregulation of the FLC repressor, the vernalization pathway and the autonomous pathway (Fig. 1.1). Vernalization, resulting from exposure to low temperatures (between 08C and 108C), is a quantitative response and causes progressive downregulation of the FLC gene until the response is saturated after several weeks (Chapter 2, this volume for further details). Once the vernalized state has been established it is stable throughout the life of the plant even if returned to warmer temperatures. It is thus mitotically stable through numerous cell divisions indicating that the repression of FLC expression is epigenetic. This epigenetic silencing of FLC is mediated by the VERNALIZATION 1 (VRN1) and VERNALIZATION 2 (VRN2) genes, which are involved in histone methylation and the formation of mitotically stable transcriptionally silent heterochromatin. Mutations in these genes prevent the stable repression of the FLC gene, and in the vrn1 and vrn2 mutants FLC expression increases back to normal as soon as the plant is returned to warm temperatures. The initial decrease in FLC expression is mediated in part by the VERNALIZATION INSENSITIVE 3 (VIN3) gene (Sung and Amasino, 2004). The VIN3 gene is induced by long periods of cold treatment and is thought to be an early step in the vernalization process. The autonomous pathway is the other floral-enabling pathway that acts to reduce the levels of FLC expression. There are several genes in this pathway, FCA, FLOWERING LOCUS D (FLD), FPA, FVE, FY, LUMINIDEPENDENS (LD) and FLOWERING LOCUS K (FLK), which act in different ways to repress FLC. FVE is thought to act together with FLD in a histone deacetylation complex and repress FLC expression by deacetylating FLC chromatin. FCA, FPA, FY and FLK all appear to have roles in RNA processing although there is no direct evidence that they directly affect FLC RNA. Active FCA mRNA levels are low at germination but increase significantly in meristems 4–5 days after germination, at a time when the floral-enabling pathways are likely to become active (Macknight et al., 2002).

Floral-promotion pathways Of the floral-promotion pathways, the photoperiodic pathway is probably the best understood. Key elements include a gene named CONSTANS, or CO, which encodes a nuclear protein with two zinc fingers at the amino terminus and a conserved carboxyl-terminal domain, known as the CCT domain for the three plant proteins in which it was identified (CO, COL, TIMING OF CAB1 (TOC1)). CO is regulated both transcriptionally and post-transcriptionally. Transcription of the CO mRNA is controlled by the circadian clock and it is also upregulated by the nuclear protein GIGANTEA (GI). CO protein stability is affected by the action of different photoreceptors. Under SD conditions, CO protein levels remain low at all times, but under LD conditions, the combined action of the blue light photoreceptor CRYPTOCHROME 2 and of the far-red light photoreceptor PHYTOCHROME A (PHYA) promotes accumulation of

10

B. Thomas et al.

the CO protein at the end of the light period. This in turn activates transcription of the floral-pathway integrators FT and SOC1 (Suarez-Lopez et al., 2001; Valverde et al., 2004). The mechanism by which accumulation of the CO protein is modulated by daylength will be described in further detail below. Recently evidence has emerged for the existence of a light-quality pathway that acts independently of CO and the photoperiodic pathway. This pathway acts through PHYTOCHROME B (PHYB) and a nuclear protein called PHYTOCHROME AND FLOWERING TIME 1 (PFT1) to upregulate FT levels in response to low red light to far-red light (R/FR) ratios (Cerdan and Chory, 2003). The effects of light quality on the activation of FT (either acting through CO or PFT1) can explain the promotion of flowering by vegetative shade which is part of the shade-avoidance response observed when plants are grown in environments with low R/FR ratios. Temperature has also been shown to affect PHYB-mediated control of flowering. At 228C PHYB is the predominant photoreceptor mediating the repression of flowering by red light, but at 168C this repression is mediated through PHYTOCHROME E (PHYE) (Halliday et al., 2003). PHYA-mediated promotion of flowering through the photoperiodic pathway is also temperature-dependent and PHYA promotes flowering at 238C but not at 168C (Blazquez et al., 2003). The other floral-promotion pathway is the GA pathway. In Arabidopsis GA promotes flowering, and mutations in genes involved in GA biosynthesis or response, such as gibberellic acid 1 (ga1) or gibberellic acid insensitive (gai), respectively, cause late flowering in SDs. The effect of these mutations is not observed in LDs because the photoperiodic pathway is actively promoting flowering in these conditions, however in SDs the GA pathway is the major floral-promotion pathway (Reeves and Coupland, 2001). It is thought that GA acts by inducing the expression of a MYB-like transcription factor AtMYB33 that binds to a motif in the LFY promoter and induces LFY expression. This GA-responsive element is distinct from a photoperiod-responsive element, which is also present in the LFY promoter. Thus, LFY acts to integrate signals from the GA and from the photoperiodic-response pathways. GA may regulate the expression of the other floral-integrator genes SOC1 and FT as well. Interestingly, the expression of GA 20-oxidase, an enzyme involved in GA biosynthesis, is reported to be higher in LDs than SDs; and genes encoding both GA 20-oxidase and 3b-hydroxylase (which catalyses a later step in the pathway) are induced by red light and downregulated by far-red light. Whether the influence of photoperiod and light quality on GA biosynthesis results in altered control of flowering by the GA pathway is not yet known. Floral-promotion pathways do not act in isolation but interact with floralenabling pathways. The photoperiodic pathway promotes flowering through activation of the floral-integrator genes FT and SOC1, but this is not possible when high levels of FLC are present. Thus, activation of the SOC1 gene by overexpression of CO can be blocked by overexpression of FLC. The FLC protein binds to an element in the SOC1 promoter that presumably prevents induction of the gene by CO (Hepworth et al., 2002). FLC also directly downregulates the photoperiodic pathway as high levels of FLC repress

Photoperiodism and Flowering

11

CRY2 expression thus preventing the CRY2-mediated promotion of CO activity (El-Assal et al., 2003). The photoperiodic pathway can thus only induce flowering once the FLC-mediated repression has been alleviated by the floral-enabling pathways, or in early flowering ecotypes where mutations to the FRI gene result in only low levels of FLC expression. This interaction between FLC and the photoperiodic pathways provides a mechanism for plants such as winter annual or biennials to delay flowering until winter has passed (and vernalization has reduced the levels of FLC) and the warm weather of late spring or summer has arrived (when LDs will activate the photoperiodicpromotion pathway).

Mechanism of Photoperiodic Time Perception The circadian clock as the timing mechanism The timing mechanism used in photoperiodism seems in most cases to be based on endogenous circadian rhythms in light sensitivity, as first postulated by Bu¨nning (1936). He suggested that the circadian clock consisted of two half cycles, photophil and scotophil. When light was received in the scotophilic phase, the daily cycle was perceived as an LD, but the absence of light during the scotophilic phase produced an SD response. This idea was further refined by Pittendrigh and Minis (1964) to form the external coincidence model. In this model, a signal was produced when an environmental signal (light) coincided with the sensitive phase of an endogenous circadian rhythm of photoresponsiveness. Support for these circadian clock-based models was initially provided by physiological studies (reviewed in Thomas and Vince-Prue, 1997). Thus, responsiveness to NBs was rhythmic in SD species and oscillated with a period of approximately 24 h. Results obtained with LDPs also supported this model, but suggested that the characteristics of the interaction of light and circadian rhythms were not the same in SDPs and LDPs. SDPs often exhibit a qualitative requirement for inductive photoperiods and can frequently be induced to flower in response to a single inductive light–dark cycle. In such plants the circadian rhythm of responsiveness to NBs is entrained by the dusk signal, so that the photoinducible phase always occurs at about the same time in darkness (Lumsden et al., 1982). Light received during the photoinducible phase (dawn or an NB) prevents the SD response. Thus it is easy to think that the circadian rhythm in light sensitivity provides the timing base that allows the length of the dark period to be distinguished by the plant. For LDPs or light-dominant plants the situation is not quite as straightforward. Because sensitivity to NBs is much reduced, it is very difficult to do NB experiments. Rhythms in response to an NB have been described in the LDPs but the pattern of response varies with the duration of the experimental dark period (Perilleux et al., 1994) suggesting an interaction with the subsequent light period. A circadian rhythm in light sensitivity in constant light can be seen when far-red light is added to a background of white fluorescent or red light

12

B. Thomas et al.

(Deitzer et al., 1982) indicating that the rhythm in light sensitivity continues to run in constant light. The link between circadian rhythms and the circadian clock has been firmly established by recent studies using Arabidopsis, a facultative LDP. First, abnormal function of the circadian clock in Arabidopsis mutants was correlated with defective floral responses to photoperiod. For example, the short-period mutant timing of cab 1-1 (toc1-1) flowered earlier than wild-type under SD conditions, whereas the long-period mutant zeitlupe (ztl) exhibited delayed flowering under LD conditions (Somers et al., 1998b, 2004). The complete arrhythmia of the lhy-1 and elf3 mutants, observed in constant light, was correlated with daylength-insensitive flowering (Hicks et al., 1996; Schaffer et al., 1998). Yet the causal link between the circadian clock defect and the flowering-time phenotype remained unclear as seemingly equivalent circadian phenotypes were in some cases associated with opposite effects on flowering times. In order to address this question, Roden et al. (2002) investigated the effects of experimental conditions that artificially altered the phase of circadian rhythms relative to the light and dark portion of the environmental cycle. One of the fundamental properties of the circadian clock is that it will advance or delay its phase (that is, shift its rhythm forward or back relative to dawn and dusk) when entrained to light–dark cycles that are either longer or shorter than 24 h, respectively. When Arabidopsis plants were exposed to a range of these atypical light–dark cycles, floral responses were not determined by the number of light or dark hours within a cycle but reflected how much transcription of the clock-controlled gene CO coincided with light. These results were consistent with a model in which expression of CO under the control of the circadian clock mediates rhythmic changes in photoresponsiveness and perception of external coincidence with light.

Clock mechanism Circadian rhythms (from circa, approximately, and dies, day) have been described in a wide range of organisms ranging from cyanobacteria to mammals and at every level of organization. These rhythms all share the same fundamental properties: (i) their ability to become entrained, or synchronized, to diurnal changes in environmental conditions; (ii) persistence upon transfer to constant conditions; and (iii) a constant period over the physiological range of temperatures. In plants, the circadian clock controls expression of approximately 6% of the transcriptome (Harmer et al., 2000). This includes genes encoding components of all major metabolic pathways as well as genes involved in hormone biosynthesis, photoreceptors and floral regulators such as CO, FT and GI. The circadian oscillator of higher plants comprises transcriptional– translational feedback loops similar to those described earlier for fungal and animal clocks (Young and Kay, 2001). In Arabidopsis, the central oscillator comprises three key components named LATE ELONGATED HYPOCOTYL (LHY), CIRCADIAN CLOCK-ASSOCIATED 1 (CCA1) and TIMING OF CAB1 (TOC1). The LHY and CCA1 genes encode single MYB transcription

Photoperiodism and Flowering

13

factors with largely overlapping functions. Expression of the LHY and CCA1 mRNA levels oscillates and levels of both transcripts peak shortly after dawn. The LHY and CCA1 proteins also are synthesized rhythmically with a lag of approximately 2 h after their cognate mRNAs. Both transcription factors bind to a promoter element known as the evening element and act to inhibit transcription of evening-specific genes, including TOC1. In the evening, the level of both repressors declines and transcription of the TOC1 gene resumes. Accumulation of the TOC1 protein at night promotes transcription from the LHY and CCA1 promoters, thus initiating a new cycle (Alabadi et al., 2001). It is not clear at this point how TOC1 performs this function as the TOC1 protein does not comprise a DNA-binding domain and does not exhibit features typical of any transcription-factor family. The TOC1 protein comprises an N-terminal domain similar to response-regulator proteins of plant and bacterial two-component signalling systems, but lacks an aspartate residue that is required for phosphotransfer (Makino et al., 2000; Strayer et al., 2000). Another domain known as the CCT-domain is thought to play a role in protein–protein interactions since in yeast it mediated binding of CO and TOC1 to the transcriptional regulator ABI3 (Kurup et al., 2000). TOC1 may regulate transcription through its interaction with ABI3. It has also been shown to interact with a number of basic helixloop-helix (bHLH) transcription factors including the phytochrome-interacting protein PIF3 (Makino et al., 2002; Yamashino et al., 2003). This is interesting because PIF3 binds a G-box motif (CCACTG?) within the promoters of the LHY and CCA1 genes (Martinez-Garcia et al., 2000). TOC1 may therefore regulate expression of the LHY and CCA1 mRNAs by modulating the activity of a light-responsive transcription factor, thus placing the oscillator very close to light-response mechanisms. Disruption of the LHY/CCA1/TOC1 feedback loop severely affected the ability of the oscillator to free-run in constant conditions. Thus, lhy cca1 double mutants became gradually arrhythmic upon transfer to constant light or darkness, whereas plants in which expression of TOC1 was inhibited using RNA interference (RNAi) technology also became arrhythmic in constant darkness and in constant red light but not under blue or white light (Alabadi et al., 2002; Mizoguchi et al., 2002; Mas et al., 2003a). These results suggest that the LHY/CCA1/TOC1 feedback loop functions as part of the oscillatory mechanism of the clock, but that its importance for self-sustained rhythmicity may vary with light conditions. Evidence is accumulating that the plant circadian clock comprises additional interlocking feedback loops, similar to those described in animal and fungal clocks (Young and Kay, 2001). A number of other elements of the circadian systems have been identified that contribute to the positive regulation of LHY and CCA1 expression. The ELF3, ELF4 and GI transcripts are expressed at night with phases similar to TOC1 (Fowler et al., 1999; Hicks et al., 2001; Doyle et al., 2002). The ELF3 gene is particularly interesting because its function is essential for rhythmicity in constant light, but not in constant darkness. ELF3 has been proposed to negatively regulate light signals to the clock and to act to dampen effects of light at times when its effects on the clock might be deleterious (Covington et al., 2001). A set of four rhythmically expressed

14

B. Thomas et al.

pseudo response-regulator proteins related to TOC1 has also been proposed to compose a regulatory feedback loop within the oscillator mechanism (Matsushika et al., 2000). Expression of the PRR9 transcript peaks shortly after dawn, followed at approximately 3-h intervals by PRR7, PRR5 and PRR3. The TOC1 transcript, also described as PRR1, describes the latest wave of gene expression. An increasing body of evidence suggests that all of the PRR genes function as part of the clock mechanism. Mutations in each of these genes caused relatively subtle effects on circadian period (Eriksson et al., 2003; Michael et al., 2003). However, plants lacking function of both PRR7 and PRR9 exhibited very longperiod phenotypes in constant light and severely dampened rhythmicity in constant darkness (Farre´ et al., 2005). Expression of PRR7 and PRR9 was increased in LHY- and CCA1-overexpressing plants and decreased in double loss-of-function mutants, suggesting a positive effect of CCA1 and LHY on PRR7 and PRR9 expression levels. Despite its dramatic long-period phenotype the prr7 prr9 double mutant exhibited nearly normal levels of LHY, CCA1 and TOC1 expression. Further work is therefore required to determine the organization of the multiple feedback loops within the plant circadian clock. Synchronization of circadian clocks to light–dark cycles (often described as entrainment) is mediated by the action of multiple photoreceptors. In Arabidopsis, this includes at least four of the five phytochromes (PHYA, B, D and E) and both cryptochromes (CRY1, 2) (Somers et al., 1998a,b; Devlin and Kay, 2000). In most organisms, effects of light on the clock are mediated by lightinduced changes in the level of one of the oscillator components. Plants are unusual in this respect because light affects the oscillator at several levels. First, light promotes transcription of the LHY and CCA1 genes in the morning (Martinez-Garcia et al., 2000; Kim et al., 2003). Expression of the LHY protein is boosted further by light-stimulated translation of the LHY mRNA (Kim et al., 2003). Light also modulates the accumulation of the TOC1 protein through the effect of the F-box, Kelch-repeat LOV-domain protein ZEITLUPE (ZTL) (Somers et al., 2000; Mas et al., 2003b). The LOV domain of the ZTL protein is related to that found in the blue light photoreceptors CRYPTOCHROMES and PHOTOTROPINS and is thought to bind a flavin chromophore. F-box proteins act to target specific molecules for ubiquitination and degradation by the 26S proteasome. ZTL promotes degradation of the TOC1 protein in darkness but not in the light, suggesting that light may act to inhibit ZTL activity. It is likely that light modulates proteolytic degradation of additional clock molecules through the effects of DEETIOLATED 1 (DET1) and CONSTITUTIVELY MORPHOGENETIC 1 (COP1), since plants lacking either of these activities exhibit short-period phenotypes and since both of these proteins participate in the proteasome-mediated degradation of positive effectors of morphogenesis in the dark (Millar et al., 1995; Schwechheimer and Deng, 2000).

Photoperception The multiple effects of light on the timing mechanism are important determinants of photoperiodic responses as they serve to set the phase of the

Photoperiodism and Flowering

15

photoperiodic-response rhythm. Yet they must be distinguished from separate effects of light mediating the floral response downstream of the clock. Two types of photoreceptors have a role in the latter: (i) phytochromes, which have maximum sensitivity in the red and far-red parts of the spectrum; and (ii) cryptochromes, the blue light photoreceptors. There are five members of the phytochrome gene family in Arabidopsis and equivalent gene families in other species (Kendrick and Weller, 2004). Phytochromes are chromoproteins that contain identical tetrapyrrole chromophores (Lagarias and Rapoport, 1980). Mutants or transgenic plants in which chromophore biosynthesis is impaired will be incapacitated with regard to all of the functional phytochromes. Such plants show altered daylength responses, usually flowering earlier than wild types (Montgomery et al., 1999, 2001; Sawers et al., 2002). Different phytochromes have different roles in controlling plant development and this is the case for the photoperiodic-perception process. Thus, PHYA is required for normal daylength perception in Arabidopsis and in the LDP, pea (Johnson et al., 1994; Weller et al., 2001). PHYB is also essential for daylength perception in barley since the BMDR-1 mutant of barley, which contains a defective PHYB is insensitive to photoperiod (Hanumappa et al., 1999). In contrast, Arabidopsis mutants deficient in PHYB flower earlier than wild type in both SD and LD but retain sensitivity to daylength (Reed et al., 1994). PHYB, along with PHYD and PHYE promotes early flowering in Arabidopsis in response to low-red to far-red ratios (Halliday and Whitelam, 2003). This response plays a role in shade avoidance and is distinct from the role of PHYB in photoperiodism, which is inhibitory (Mockler et al., 2003). The cryptochromes are flavoproteins that mediate plant responses to blue light (Kendrick and Weller, 2004). Two members of the cryptochrome gene family (CRY1 and CRY2) are present in Arabidopsis. CRY2 is thought to be the major blue photoreceptor for flowering in Arabidopsis, although cry1 cry2 double mutants flowered earlier in blue light than the single mutants, indicating that both cryptochromes play a role to promote flowering (Mockler et al., 1999). Further evidence for a role for CRY2 comes from a study by El-Assal et al. (2001) in which a quantitative trait loci (QTL) for flowering time in Arabidopsis was accounted for by an allele of CRY2. The early flowering phenotype resulted from a single amino acid substitution that reduces the light-induced turnover of the CRY2 protein under short photoperiods. The participation of cryptochromes in the control of flowering in Arabidopsis is consistent with physiological studies, which have shown that blue light has a promotive effect on flowering for LDP of the Cruciferae, however, this is not necessarily true for other families (Thomas and Vince-Prue, 1997; Runkle and Heins, 2001). There is no physiological evidence for a specific role for blue light, and by inference cryptochromes, in SDP, but this remains to be confirmed in genetic and comparative genomic studies.

Perception of external coincidence One of the genes whose expression is under control of the circadian clock is CO, which is a key regulator of the photoperiodic pathway (Suarez-Lopez et al.,

16

B. Thomas et al.

2001). The rhythmic expression of CO cycles such that low levels of transcripts are observed during the day. Expression begins to increase approximately 8 h after dawn, followed by a broad peak between 12 and 16 h after dawn (Fig. 1.2). In SD conditions this increase coincides with the beginning of the night, however, in LD the CO transcript accumulates to relatively high levels in the light. CO directly activates transcription of the floral integrators FT and SOC1, but only when expression of its mRNA coincides with the light period under LD conditions (Fig. 1.2). This is because the CO protein is subject to post-transcriptional regulation and is ubiquitinated and degraded by the 26S proteasome in the dark (Valverde et al., 2004). Consequently, the protein does not accumulate when expression of the CO transcript takes place in darkness and FT transcription is not induced. Accumulation of the CO protein is promoted under monochromatic far-red and blue light by the action of PHYA- and CRY2-mediated pathways, respectively, but is prevented in red light via a PHYB-dependent pathway. Thus under natural conditions of white light where all of these photoreceptors are stimulated at the same time, the balance between the activity of these different pathways must determine the floral response. This is in good agreement with mutant analyses, which suggested that far-red light acting through PHYA and blue light acting through CRY1 and CRY2 may act to antagonize the repression of flowering mediated by PHYB (Mockler et al., 2003). A surprising observation is that plants that expressed the CO mRNA at a constant level from a heterologous promoter (35S::CO) only accumulate high levels of the CO protein towards the end of the day (Valverde et al., 2004). This pattern was proposed to result from a gradual shift in the balance of different photoreceptor pathways through the course of the day so that PHYB promoted the degradation of CO in the morning but this effect was antagonized about 12 h onwards through the effects of CRY2 and PHYA, thus allowing the accumulation of CO protein in the light. This gradual shift in the

Day 0

Night

8 h 16 h 24 h

CO Long days Flowering FT expression Day 0 CO

8h

Night 24 h

Short days FT expression

No flowering

Fig. 1.2. Proposed rhythmic expression of CO under long- and short-daylight/dark cycles and the resulting expression of FT.

Photoperiodism and Flowering

17

balance of photoreceptor pathways may result from the circadian patterns of expression of the photoreceptors, since expression of the PHYB mRNA peaks in the morning whereas that of PHYA and CRY2 mRNAs peak later in the day (To´th et al., 2001). The levels of the cognate proteins do not oscillate significantly, but it is possible that post-translational modifications alter the properties of these photoreceptors over time and that the older proteins are inactive with regard to photoperiodic induction. In addition to its effects on CO protein turnover, light promotes expression of the CO mRNA at the end of an LD through the action of a rhythmically expressed protein known as FKF1 (Imaizumi et al., 2003). FKF1 is part of a family of three flavin-binding, Kelch-repeat, F-box proteins (FKF1, ZTL and LKP2) that regulate circadian rhythms by targeting specific proteins for degradation by the proteasome. Interestingly, the FKF1 protein has an LOV domain, which is the light-sensing module of phototropin blue light photoreceptors, and this was shown to bind a flavin-mononucleotide chromophore and to exhibit blue light–induced changes in absorbance. FKF1 thus has properties of a blue light photoreceptor and may be able to perceive blue light. In wild-type plants grown under LD cycles, FKF1 protein levels reach maximum levels before dusk and the peak of FKF1 expression coincides with a shoulder of the broad peak of CO mRNA at the end of the light interval. This shoulder was not detected in a T-DNA insertion allele, fkf1-2, indicating that FKF1 activity was required to promote this particular aspect of CO transcription. As a result, the onset of CO transcription was delayed into the dark interval and this correlated with low expression of FT and late flowering. The effect of FKF1 on CO transcription required exposure to light, since no difference in CO expression patterns was observed between wild-type and fkf1 mutant plants upon transfer to shorter photoperiods where FKF1 expression does not coincide with light. Rhythmic transcription of CO under the control of the circadian clock provides the basis for rhythmic responsiveness to light in photoperiodism, since the effects of CRY2 and PHYA on CO protein accumulation can take place only when the CO mRNA is expressed and actively translated. Photoperiodic time perception in Arabidopsis involves more than one rhythm of light sensitivity, however, since the peak of FKF1 protein must also coincide with light in order to promote CO transcription. Levels of accumulation of the CO protein have been shown to closely correlate with the transcriptional induction of the floral integrator FT. A combination of approaches including grafting and tissue-specific expression have shown that CO acts in the phloem to regulate a systemic flowering signal through cell-autonomous activation of the flowering integrator FT (An et al., 2004). The FT gene encodes a 23 kD protein with amino acid sequence similarity to mammalian RAF kinase inhibitor proteins (Kardailsky et al., 1999; Kobayashi et al., 1999). Recent work has provided convincing support for the concept of a systemic flowering signal (florigen or antiflorigen) as described earlier in this chapter. FT mRNA induced locally in the leaf, moves to the apex where the FT protein interacts with the shoot apex-expressed transcription factor FD to initiate flowering (Huang et. al., 2005). This strongly suggests that FT mRNA itself constitutes an important part of the floral stimulus (i.e. florigen).

18

B. Thomas et al.

Daylength perception in short-day plants With the recent completion of its genome project, rice has emerged as a powerful genetic system and an excellent SD plant counterpart to Arabidopsis. Genes that control photoperiodic flowering have been identified through QTL analysis of flowering time, also described as heading date. Fourteen loci controlling heading date (labelled Hd1 to Hd14) were identified, five of which were

ELF3

LHY mRNA CCA1 mRNA

PRR9

PHYA,B,D,E

LHY protein CCA1 protein

CRY 1,2

PRR7 PRR5

ZTL

TOC1 protein PRR3

Proteasomal degradation

TOC1 mRNA Circadian oscillator

FKF1 mRNA

PHYB

FKF1 protein

CO mRNA CRY1,2 PHYA CO protein

Proteasomal degradation FT mRNA

Floral transition

Photoperiodism and Flowering

19

shown to control photoperiodic responses, including Hd1, Hd2, Hd3, Hd5 and Hd6 (Lin et al., 2000; Yamamoto et al., 2000). Hd6 identified the a-subunit of casein kinase 2 (CK2), a protein that interacts with and phosphorylates CCA1 in Arabidopsis and is thought to regulate the function of the circadian clock (Sugano et al., 1998, 1999; Takahashi et al., 2001). Hd1 was allelic to Se1, a flowering-time locus identified through mutant analyses. Mapbased cloning of Hd1 identified a homologue of CO in Arabidopsis. Another QTL, Hd3a, identified a gene related to Arabidopsis FT that functions as a positive effector of flowering (Kojima et al., 2002). The rice counterpart of CO (Hd1 or Se1) is expressed rhythmically with a phase similar to that of CO in Arabidopsis. As in Arabidopis, expression of Hd3a and the related rice genes RFT1, FTL are regulated by CO. A key difference, however, is that Hd1 inhibits expression of Hd3a under LDs and promotes it under SD conditions (Izawa et al., 2002; Kojima et al., 2002). Unlike Arabidopsis CO, which only plays a role to promote FT expression and flowering under LD conditions, Hd1 may also have differing functions under SD and LD conditions. The se1 mutant flowered earlier than wild-type under LDs but later than wild-type under SDs, suggesting that the wild-type gene product may be required to delay flowering in one condition while promoting it in the other (Yano et al., 2000). In addition to these components, which are closely related to genes identified in Arabidopsis, Early Heading Date1 (Ehd1) promotes early flowering under SD conditions (Doi et al., 2004). The Ehd1 gene encodes a protein containing a B-type response-regulator domain, which may be involved in relaying a phosphorylation signal, as well as a GARP DNA-binding domain. No orthologue of Edh1 was detected in the Arabidopsis genome. Edh1

Fig. 1.3. Mechanism of photoperiodic timing in Arabidopsis. Perception of daylength is mediated by an interaction of a light and a circadian rhythm. The timing mechanism of the clock is composed of multiple transcriptional–translational feedback loops, whose oscillations are entrained to diurnal light–dark cycles through the action of phytochrome and cryptochrome photoreceptors. Light resets the clock through at least two mechanisms, including transcriptional induction of LHY and CCA1 expression and light-induced degradation of the TOC1 protein. Effects of light on the clock are dampened at night through the action of the rhythmically expressed protein ELF3. Downstream of the clock, expression of the floral regulator CO is rhythmic under diurnal light– dark cycles. Light promotes expression of the CO mRNA at the end of a long day through the action of the rhythmically expressed photoreceptor FKF1. In addition, the circadian oscillator mediates light-independent expression of the CO mRNA at night. The CO protein does not accumulate in the dark because it is subject to proteasomal degradation. Blue light acting through cryptochromes and far-red light acting through PHYA prevent this degradation and allow accumulation of CO protein, which can then activate transcription of FT and promote the conversion of vegetative meristems to floral meristems. This action of CRY1 and CRY2 photoreceptors is antagonized by red light acting through the PHYB photoreceptor. Thus, photoperiodic induction of flowering in Arabidopsis takes place when the circadian rhythm of CO transcription coincides with a blue or far-red light signal.

20

B. Thomas et al.

induces transcription of FT-like genes independently of Hd1 and probably represents a second mechanism by which Hd3a expression is increased in an SD-specific manner. The photoreceptor mediating photoperiodic responses in rice is the red/ far-red photoreceptor phytochrome. As for cry2 mutants of Arabidopsis, loss of phytochrome (SE5) function in rice abolished responses to LDs. The se5 mutant plants flowered as early under inhibitory LDs and as wild-type under inductive SD conditions (Izawa et al., 2002). Thus, under LD conditions, coincidence between light and Hd1 expression may lead to inhibition of Hd3a transcription and suppression of flowering. Rhythmic expression of Hd1 is not altered in the se5 mutant, suggesting that phytochrome may act downstream to mediate perception of external coincidence (Izawa et al., 2002). The mechanism by which Edh1 participates in daylength responses independently of Hd1 remains to be elucidated.

Conclusions Genetic and molecular studies, largely with the model plant A. thaliana, over the last decade have gone a long way to confirm and explain the essential elements of photoperiodism as it applies to flowering in plants. The physiological conclusions that photoperiodic mechanisms involve multiple interactions between photoreceptors and an underlying circadian rhythm in light sensitivity through an external-coincidence model have been borne out. However, the level of complexity in these interactions is much greater than conceived in classical physiology (Fig. 1.3). Homologues of the central genetic elements of the model in the LDP Arabidopsis have been shown to affect photoperiodic regulation in SDPs such as rice and Pharbitis, implying a common photoperiodic mechanism for all plants. The role of a florigenic signal has also been confirmed and although its exact nature is still unknown, there are prospects of it being discovered in the foreseeable future. We can also soon expect to understand more fully the basis for the different requirements for light quantity and quality in different species. This understanding will be of great benefits to breeders and agronomists in designing and growing plants with flowering properties tailored to the food and ornamental industries and the wider needs of society.

References Alabadi, D., Oyama, T., Yanovsky, M.J., Harmon, F.G., Mas, P. and Kay, S.A. (2001) Reciprocal regulation between TOC1 and LHY/CCA1 within the Arabidopsis circadian clock. Science 293, 880–883. Alabadi, D., Yanovsky, M.J., Mas, P., Harmer, S.L. and Kay, S.A. (2002)

Critical role for CCA1 and LHY in maintaining circadian rhythmicity in Arabidopsis. Current Biology 12, 757–761. An, H.L., Roussot, C., Suarez-Lopez, P., Corbesler, L., Vincent, C., Pineiro, M., Hepworth, S., Mouradov, A., Justin, S., Turnbull, C. and Coupland, G.

Photoperiodism and Flowering (2004) CONSTANS acts in the phloem to regulate a systemic signal that induces photoperiodic flowering of Arabidopsis. Development 131, 3615–3626. Blazquez, M.A., Ahn, J.H. and Weigel, D. (2003) A thermosensory pathway controlling flowering time in Arabidopsis thaliana. Nature Genetics 33, 168–171. Boss, P.K., Bastow, R.M., Mylne, J.S. and Dean, C. (2004) Multiple pathways in the decision to flower: enabling, promoting, and resetting. Plant Cell 16, S18–S31. Bu¨nning, E. (1936) Die endogene Tagesrhythmik als Grundlage der photoperiodischen Reaktion. Berichte der Deutschen Botanischen Gesellschaft 54, 590–607. Cerdan, P.D. and Chory, J. (2003) Regulation of flowering time by light quality. Nature 423(6942), 881–885. Chailakhyan, M.Kh. (1936) On the hormonal theory of plant development. Doklady Academy of Science USSR 12, 443–447. Covington, M.F., Panda, S., Liu, X.L., Strayer, C.S., Wagner, D.R. and Kay, S.A. (2001) ELF3 modulates resetting of the circadian clock in Arabidopsis. Plant Cell 13, 1305–1315. Deitzer, G.F., Hayes, R. and Jabben, M. (1982) Phase shift in the circadian rhythm of floral promotion by far-red light in Hordeum vulgare L. Plant Physiology 69, 597–601. Devlin, P.F., and Kay, S.A. (2000) Cryptochromes are required for phytochrome signalling to the circadian clock but not for rhythmicity. Plant Cell 12, 2499–2509. Doi, K., Izawa, T., Fuse, T., Yamanouchi, U., Kubo, T., Shimatani, Z. and Yano, M. (2004) EhD1, a B-type response regulator in rice, confers short-day promotion of flowering and controls Ft-like gene expression independently of Hd1. Genes and Development 18, 926–936.

21 Doyle, M.R., Davis, S.J., Bastow, R.M., McWatters, H.G., Kozma-Bognar, L., Nagy, F., Millar, A.J. and Amasino, R.M. (2002) The ELF4 gene controls circadian rhythms and flowering time in Arabidopsis thaliana. Nature 419, 74–77. El-Assal, S.E.D., Alonso-Blanco, C., Peeters, A.J.M., Raz, V. and Koornneef, M. (2001) A QTL for flowering time in Arabidopsis reveals a novel allele of CRY2. Nature Genetics 29, 435–440. El-Assal, S.E.D., Alonso-Blanco, C., Peeters, A.J.M., Wagemaker, C., Weller, J.L. and Koornneef, M. (2003) The role of cryptochrome 2 in flowering in Arabidopsis. Plant Physiology 133, 1504–1516. Eriksson, M.E., Hanano, S., Southern, M.M., Hall, A. and Millar, A.J. (2003) Response regulator homologues have complementary, light-dependent functions in the Arabidopsis circadian clock. Planta 218, 159–162. Farre´, E.M., Harmer, S.L., Harmon, F.G., Yanovsky, M.J. and Kay, S.A. (2005) Overlapping and distinct roles of PRR7 and PRR9 in the Arabidopsis clock. Current Biology 15, 47–54. Fowler, S., Lee, K., Onouchi, H., Samach, A., Richardson, K., Morris, B., Coupland, G. and Putterill, J. (1999) GIGANTEA: a circadian clockcontrolled gene that regulates photoperiodic flowering in Arabidopsis and encodes a protein with several possible membrane-spanning domains. EMBO Journal 18, 4679–4688. Garner, W.W. and Allard, H.A. (1920) Effect of the relative length of the day and night and other factors of the environment on growth and reproduction in plants. Journal of Agricultural Research 18, 533–603. Halliday, K.J. and Whitelam, G.C. (2003) Changes in photoperiod or temperature alter the functional relationships between phytochromes and reveal

22

B. Thomas et al. roles for phyD and phyE. Plant Physiology 131, 1913–1920. Halliday, K.J., Salter, M.G., Thingnaes, E. and Whitelam, G.C. (2003) Phytochrome control of floweritemperature sensitive and correlates with expression of the floral integrator FT. Plant Journal 33, 875–885. Hamner, K.C. and Bonner, J. (1938) Photoperiodism in relation to hormones as factors in floral initiation. Botanica Gazette 100, 388–431. Hanumappa, M., Pratt, L.H., Cordonnier-Pratt, M.M. and Deitzer, G.F. (1999) A photoperiod-insensitive barley line contains a light-labile phytochrome B. Plant Physiology 119, 1033–1039. Harmer, S.L., Hogenesch, J.B., Straume, M., Chang, H.-S., Han, B., Zhu, T., Wang, X., Kreps, J. A. and Kay, S.A. (2000) Orchestrated transcription of key pathways in Arabidopsis by the circadian clock. Science 290, 2110– 2113. Hepworth, S.R., Valverde, F., Ravenscroft, D., Mouradov, A. and Coupland, G. (2002) Antagonistic regulation of flowering-time gene SOC1 by CONSTANS and FLC via separate promoter motifs. EMBO Journal 21, 4327–4337. Hicks, K.A., Millar, A.J., Carre´, I.A., Somers, D.E., Straume, M., MeeksWagner, D.R. and Kay, S.A. (1996) Conditional circadian dysfunction of the Arabidopsis earlyflowering 3 mutant. Science 274, 790–792. Hicks, K.A., Albertson, T.M. and Wagner, D.R. (2001) EARLY FLOWERING 3 encodes a novel protein that regulates circadian clock function and flowering in Arabidopsis. Plant Cell 13, 1281– 1292. ¨ hlenius, H., Eriksson, S., Huang, T., Bo Parcy, F. and Nilsson, O. (2005) The mRNA of the Arabidopsis gene FT moves from leaf to shoot apex and induces flowering. Science 309, 1694–1696.

Imaizumi, T., Tran, H.G., Swartz, T.E., Briggs, W.R. and Kay, S.A. (2003) FKF1 is essential for photoperiodicspecific light signalling in Arabidopsis. Nature 426(6964), 302–306. Izawa, T., Oikawa, T., Sugiyama, N., Tanisaka, T., Yano, M. and Shimamoto, K. (2002) Phytochrome mediates the external light signal to repress FT orthologs in photoperiodic flowering of rice. Genes and Development 16, 2006–2020. Johanson, U., West, J., Lister, C., Michaels, S., Amasino, R. and Dean, C. (2000) Molecular analysis of FRIGIDA, a major determinant of natural variation in Arabidopsis flowering time. Science 290, 344–347. Johnson, E., Bradley, M., Harberd, N.P. and Whitelam, G.C. (1994) Photoresponses of light-grown Phya mutants of Arabidopsis – Phytochrome-a is required for the perception of daylength extensions. Plant Physiology 105, 141–149. Kardailsky, I., Shukla, V.K., Ahn, J.H., Dagenais, N., Christensen, S.K., Nguyen, J.T., Chory, J., Harrison, M.J. and Weigel, D. (1999) Activation tagging of the floral inducer FT. Science 286, 1962–1965. Kendrick, R.E. and Weller, J.L. (2004) Phytochromes and other photoreceptors. In: Thomas, B., Murphy, D.J. and Murray, B.G. (eds) Encyclopedia of Applied Plant Sciences. Elsevier, Oxford, UK, pp. 1063–1069. Kim, J.Y., Song, H.R., Taylor, B.L. and Carre, I.A. (2003) Light-regulated translation mediates gated induction of the Arabidopsis clock protein LHY. EMBO Journal 22, 935–944. Knott, J.E. (1934) Effect of a localized photoperiod on spinach. Proceedings of the American Society for Horticultural Science 31, 152–154. Kobayashi, Y., Kaya, H., Goto, K., Iwabuchi, M. and Araki, T. (1999) A pair of related genes with antagonistic roles in mediating flowering signals. Science 286, 1960–1962.

Photoperiodism and Flowering Kojima, S., Takahashi, Y., Kobayashi, Y., Monna, L., Sasaki, T., Araki, T. and Yano, M. (2002) Hd3a, a rice orthologue of the Arabidopsis FT gene, promotes flowering downstream of Hd1 under short day conditions. Plant Cell Physiology 43, 1096–1105. Kurup, S., Jones, H.D. and Holdsworth, M.J. (2000) Interactions of the developmental regulator ABI3 with proteins identified from developing Arabidopsis seed. Plant Journal 21, 143–155. Lagarias, J.C. and Rapoport, H. (1980) Chromopeptides from phytochrome. The structure and linkage of the Pr form of the phytochrome chromophore. Journal of American Chemical Society 102, 4821–4828. Lin, H.X., Yamamoto, T., Sasaki, T. and Yano, M. (2000) Characterisation and detection of epistatic interactions of 3QTLs, Hd1, Hd2, and Hd3, controlling heading date in rice using nearly isogenic lines. Theoretical and Applied Genetics 101(7), 1021–1028. Lumsden, P.J., Thomas, B. and VincePrue, D. (1982) Photoperiodic control of flowering in dark grown seedlings of Pharbitis nil Choisy. The effect of skeleton and continuous light photoperiods. Plant Physiology 70, 277– 282. Macknight, R., Duroux, M., Laurie, R., Dijkwel, P., Simpson, G. and Dean, C. (2002) Functional significance of the alternative transcript processing of the Arabidopsis floral promoter FCA. Plant Cell 14, 877–888. Makino, S., Matsushika, A., Kojima, M., Yamashino, T. and Mizuno, T. (2002) The APRR1/TOC1 quintet implicated in circadian rhythms of Arabidopsis thaliana: I. Characterization with APRR1-overexpressing plants. Plant Cell Physiology 43, 58–69. Martinez-Garcia, J.F., Huq, E. and Quail, P.H. (2000) Direct targeting of light signals to a promoter element-bound transcription factor. Science 288, 859–863.

23 Mas, P., Alabadi, D., Yanovsky, M.J., Oyama, T. and Kay, S.A. (2003a) Dual role of TOC1 in the control of circadian and photomorphogenic responses in Arabidopsis. Plant Cell 15, 223–236. Mas, P., Kim, W.Y., Somers, D.E. and Kay, S.A. (2003b) Targeted degradation of TOC1 by ZTL modulates circadian function in Arabidopsis thaliana. Nature 426, 567–570. Matsushika, A., Makino, S., Kojima, M. and Mizuno, T. (2000) Circadian waves of expression of the APRR1/ TOC1 family of pseudo-response regulators in Arabidopsis thaliana: insight into the plant circadian clock. Plant and Cell Physiology 41, 1002– 1012. Michael, T.P., Salome, P.A., Yu, H.J., Spencer, T.R., Sharp, E.L., McPeek, M.A., Alonso, J.M., Ecker, J.R. and McClung, C.R. (2003) Enhanced fitness conferred by naturally occurring variation in the circadian clock. Science 302, 1049–1053. Millar, A.J., Straume, M., Chory, J., Chua, N.H. and Kay, S.A. (1995) The regulation of circadian period by phototransduction pathways in Arabidopsis. Science 267, 1163–1166. Mizoguchi, T., Wheatley, K., Hanzawa, Y., Wright, L., Mizoguchi, M., Song, H.R., Carre, I.A. and Coupland, G. (2002) LHY and CCA1 are partially redundant genes required to maintain circadian rhythms in Arabidopsis. Developmental Cell 2, 629–641. Mockler, T.C., Guo, H.W., Yang, H.Y., Duong, H. and Lin, C.T. (1999) Antagonistic actions of Arabidopsis cryptochromes and phytochrome B in the regulation of floral induction. Development 126, 2073–2208. Mockler, T., Yang, H.Y., Yu, X.H., Parikh, D., Cheng, Y.C., Dolan, S. and Lin, C.T. (2003) Regulation of photoperiodic flowering by Arabidopsis photoreceptors. Proceedings of the National Academy of Sciences of USA 100, 2140–2145.

24

B. Thomas et al. Montgomery, B.L., Yeh, K.C., Crepeau, M.W. and Lagarias, J.C. (1999) Modification of distinct aspects of photomorphogenesis via targeted expression of mammalian biliverdin reductase in transgenic Arabidopsis plants. Plant Physiology 121, 629–639. Montgomery, B.L., Franklin, K.A., Terry, M.J., Thomas, B., Jackson, S.D., Crepeau, M.W. and Lagarias, J.C. (2001) Biliverdin reductase-induced phytochrome chromophore deficiency in transgenic tobacco. Plant Physiology 125, 266–277. Perilleux, C., Bernier, G. and Kinet, J.M. (1994) Circadian-rhythms and the induction of flowering in the long-day grass Lolium-Temulentum L. Plant Cell and Environment 17, 755–761. Pittendrigh, C.S. and Minis, D.H. (1964) The entrainment of circadian clocks by light and their role as photoperiodic clocks. The American Naturalist 98, 261–294. Reed, J.W., Nagatani, A., Elich, T.D., Fagan, M. and Chory, J. (1994) Phytochrome-a and Phytochrome-b have overlapping but distinct functions in Arabidopsis development. Plant Physiology 104, 1139–1149. Reeves, P.H. and Coupland, G. (2001) Analysis of flowering time control in Arabidopsis by comparison of double and triple mutants. Plant Physiology 126(3), 1085–1091. Roden, L.C., Song, H.-R., Jackson, S., Morris, K. and Carre´, I.A. (2002) Floral responses to photoperiod are correlated with the timing of rhythmic gene expression relative to dawn and dusk, in Arabidopsis. Proceedings of the National Academy of Sciences USA 99, 13313–13318. Runkle, E.S. and Heins, R.D. (2001) Specific functions of red, far red, and blue light in flowering and stem extension of long-day plants. Journal of the American Society for Horticultural Science 126, 275–282. Sawers, R.J.H., Linley, P.J., Farmer, P.R., Hanley, N.P., Costich, D.E., Terry, M.J.

and Brutnell, T.P. (2002) Elongated mesocotyl1, a phytochrome-deficient mutant of maize. Plant Physiology 130, 155–163. Schaffer, R., Ramsay, N., Samach, A., Corden, S., Putterill, J., Carre´, I.A. and Coupland, G. (1998) The late elongated hypocotyl mutation of Arabidopsis disrupts circadian rhythms and the photoperiodic control of flowering. Cell 93, 1219–1229. Schwechheimer, C. and Deng, X.-W. (2000) The COP/DET/FUS proteins – regulators of eukaryotic growth and development. Seminars in Cell and Developmental Biology 11, 495– 503. Somers, D.E., Devlin, P.F. and Kay, S.A. (1998a) Phytochromes and cryptochromes in the entrainment of the Arabidopsis circadian clock. Science 282, 1488–1490. Somers, D.E., Webb, A.A., Pearson, M. and Kay, S.A. (1998b) The shortperiod mutant, toc1-1, alters circadian clock regulation of multiple outputs throughout development in Arabidopsis thaliana. Development 125, 485– 494. Somers, D.E., Schultz, T.F., Milnamow, M. and Kay, S.A. (2000) ZEITLUPE encodes a novel clock-associated PAS protein from Arabidopsis. Cell 101, 319–329. Somers, D.E., Kim, W.Y. and Geng, R.S. (2004) The F-box protein ZEITLUPE confers dosage-dependent control on the circadian clock, photomorphogenesis, and flowering time. Plant Cell 16, 769–782. Strayer, C., Oyama, T., Schultz, T.F., Raman, R., Somers, D.E., Mas, P., Panda, S., Kreps, J.A. and Kay, S.A. (2000) Cloning of the Arabidopsis clock gene TOC1, an autoregulatory response regulator homolog. Science 289, 768–771. Suarez-Lopez, P., Wheatley, K., Robson, F., Onouchi, H., Valverde, F. and Coupland, G. (2001) CONSTANS mediates between the circadian clock

Photoperiodism and Flowering and control of flowering in Arabidopsis. Nature 410, 1116–1120. Sugano, S., Andronis, C., Green, R.M., Wang, Z.Y. and Tobin, E.M. (1998) Protein kinase CK2 interacts with and phosphorylates the Arabidopsis circadian clock-associated 1 protein. Proceedings of the National Academy of Sciences USA 95, 11020–11025. Sugano, S., Andronis, C., Ong, M.S., Green, R.M. and Tobin, E.M. (1999) The protein kinase CK2 is involved in regulation of circadian rhythms in Arabidopsis. Proceedings of the National Academy of Sciences USA 96, 12362–12366. Sung, S. and Amasino, R.M. (2004) Vernalization in Arabidopsis thaliana is mediated by the PHD finger protein VIN3. Nature 427(6970), 159–164. Takahashi, Y., Yomura, A., Sasaki, T. and Yano, M. (2001) Hd6, a rice quantitative trait locus involved in photoperiod sensitivity, encodes the alpha subunit of protein kinase CK2. Proceedings of the National Academy of Sciences USA 98, 7922–7927. Thomas, B. and Vince-Prue, D. (1997) Photoperiodism in Plants. Academic Press, San Diego, California. To´th, R., Kevei, E., Hall, A., Millar, A.J., Nagy, F. and Kozma-Bognar, L. (2001) Circadian clock-regulated expression of phytochrome and cryptochrome genes in Arabidopsis. Plant Physiology 127, 1607–1616. Valverde, F., Mouradov, A., Soppe, W., Ravenscroft, D., Samach, A. and

25 Coupland, G. (2004) Photoreceptor regulation of CONSTANS protein in photoperiodic flowering. Science 303, 1003–1006. Weller, J.L., Beauchamp, N., Kerckhoffs, L.H.J., Platten, J.D. and Reid, J.B. (2001) Interaction of phytochromes A and B in the control of de-etiolation and flowering in pea. Plant Journal 26, 283–294. Yamamoto, T., Lin, H.X., Sasaki, T. and Yano, M. (2000) Identification of heading date quantitative trait locus Hd6 and characterization of its epistatic interactions with Hd2 in rice using advanced backcross progeny. Genetics 154, 885–891. Yamashino, T., Matsushika, A., Fujimori, T., Sato, S., Kato, T., Tabata, S. and Mizuno, T. (2003) A link between circadian-controlled bHLH factors and the APRR1/TOC1 quintet in Arabidopsis thaliana. Plant and Cell Physiology 44, 619–629. Yano, M., Katayose, Y., Motoyuki, A., Yamanouchi, U., Monna, L., Fuse, T., Baba, T., Yamamoto, K., Umehara, Y., Nagamura, Y. and Sasaki, T. (2000) Hd1, a major photoperiod sensitivity quantitative trait locus in rice, is closely related to the Arabidopsis flowering time gene CONSTANS. Plant Cell 12, 2473–2483. Young, M.W. and Kay, S.A. (2001) Time zones: a comparative genetics of circadian clocks. Nature Reviews. Genetics 2, 702–715.

2

Vernalization A.R. GENDALL1 AND G.G. SIMPSON2 1

Department of Botany, La Trobe University, Bundoora, Victoria, Australia; Dundee University, Gene Expression Programme, Scottish Crop Research Institute, Invergowrie, Dundee, UK 2

Introduction Some definitions For clarity, we begin this chapter with some definitions. The term vernalization is derived from the Latin word vernus meaning ‘of the spring’. Chouard (1960) defined vernalization as ‘the acquisition or acceleration of the ability to flower by a chilling treatment’. The promotion of flowering by vernalization is the result of subjecting an imbibed seed or young plant to a long period of cold (typically weeks). Floral initiation does not occur in the cold treatment but only after returning the plant to a higher temperature and in many cases a specific photoperiod. Therefore, cold temperatures do not cause plants to initiate floral primordia, but create the capacity for subsequent flowering. Vernalization can be facultative or obligate. Winter annuals have a facultative vernalization requirement, as cold accelerates, but this is not required for flowering. Biennials, however, cannot flower without cold treatment and therefore have an obligate requirement for vernalization. Summer annuals flower rapidly without a vernalization treatment. Some Arabidopsis summer-annual accessions may complete more than one life cycle in the same growing season in the wild and as such a more appropriate term for them may be ‘rapid cyclers’. It is the connection to the acceleration of floral initiation that distinguishes vernalization from other cold-related phenomena such as dormancy chilling and stratification. In woody perennials, floral primordia are formed during a growing season, but further development is arrested as plants enter dormancy in winter. The chilling of winter breaks dormancy, resulting in bud burst and the subsequent appearance of fully developed flowers. Stratification is the promotion of seed germination by cold, usually of only a few days duration. 26

ßCAB International 2006. The Molecular Biology and Biotechnology of Flowering, 2nd edn (ed. B.R. Jordan)

Vernalization

27

Vernalization and flowering time control Plants meet the challenge of environmental change with developmental change as an adaptation to their sessile lifestyle. For example, many plants control the time at which they flower by monitoring and responding to environmental stimuli like light quality, daylength and temperature (Yanovsky and Kay, 2003; Boss et al., 2004). In this way, plants ensure that they reproduce in favourable conditions. This developmental plasticity is made possible by the supply of undifferentiated stem cells from shoot meristems through the course of a plant’s life (Baurle and Laux, 2003). The floral transition is executed by the product of genes that confer floral meristem identity (Ferra´ndiz et al., 2000). Therefore, at the molecular level, the control of flowering time ultimately depends on the activation of floral meristem-identity genes. Precision and robustness in this regulation is delivered by genetically separable pathways that provide for the quantitative integration of multiple environmental responses with an endogenous programme of development (Yanovsky and Kay, 2003; Boss et al., 2004). Temperate environments are characterized by seasonal changes in rainfall, temperature and daylength. These cues provide relatively reliable indicators of the prevailing season and plants respond to them in controlling flowering time. For example, many plants promote flowering in response to long-day (LD) photoperiods as this can distinguish late spring from midwinter (Mouradov et al., 2002; Yanovsky and Kay, 2003). However, the predictability as well as the reliability of these seasonal changes can provide information not only on the prevailing season but also on the seasonal progression. This is important, for although daylength and ambient temperature at the spring and autumn equinoxes may be similar, the conditions that follow are strikingly different. The importance of monitoring seasonal progression is evident in winter annuals that germinate prior to winter, but actively repress flowering until winter has passed. If winter annuals are grown in otherwise favourable conditions, but not exposed to a long cold treatment, the transition to flowering is extremely delayed. However, if exposed to a long cold treatment, such plants show accelerated flowering. By invoking a default delay in flowering from germination until winter has passed, winter annuals in effect ‘predict’ that winter is yet to come (Simpson and Dean, 2002). Likewise, by responding to favourable environmental cues only after exposure to a long cold treatment, they effectively ‘remember’ that winter has passed. The close connection of flower development to seasonal progression makes flowering-time control an important aspect of how plants adapt to their environment and logically, therefore, is an important part of crop plantbreeding programmes. Consistent with this, ecotypes and agritypes of the same species show different flowering responses and one of the most striking distinctions relates to winter/summer annualism. The widespread uptake of Arabidopsis thaliana as the laboratory-based model of modern plant biology is partly due to its rapid life cycle. However, in the wild, most Arabidopsis ecotypes are winter annuals that exhibit extremely delayed flowering unless exposed to a vernalization treatment (Johanson et al., 2000). The adaptive

28

A.R. Gendall and G.G. Simpson

value of winter annualism may derive both from preventing flowering in winter and from providing a mechanism to monitor seasonal progression. Several potential explanations may account for the evolution of summer and winter annualism and the selective forces may depend on different local environmental conditions (see below). It is important to mention that winter- and summerannual classifications cover extremes of flowering behaviour. In reality, considerable variation exists in flowering time and vernalization requirement, and this is also likely to relate to adaptation to particular environments. These distinctions are clearly pertinent to agriculture and winter/summer annualism is an important trait in breeding programmes as it influences the geographical range in which crop plants can be grown. Vernalization-requiring crops are typically sown in the autumn and are known as winter types. In contrast, non-vernalization-requiring agritypes are typically sown in the spring and known as spring types. The availability of both winter and spring crops can influence the dynamic management of multiple crops on one farm in 1 year.

Dissecting the Molecular Processes Involved in Vernalization The process of vernalization has two essential features: (i) requirement; and (ii) response. Vernalization requirement must involve a mechanism by which flowering is repressed in a manner that can be overcome by a long period of cold. Vernalization response depends on the ability to sense cold temperature and to signal this to factors that execute the vernalization response. An important feature of vernalization is its quantitative nature: plants must measure the duration of cold in order to distinguish an occasional cold-spell from the true passage of winter, requiring then, the measurement and recording of temperature over a period of weeks. Plants show a progressive acceleration of flowering time that is correlated with the duration of the cold treatment they received. Finally, plants remember that winter has passed – vernalization does not directly promote flowering but facilitates or enables the response to genuinely promotive cues like daylength. Therefore, there is a significant intervening period between exposure to low temperatures and subsequent floral development that means that the memory of winter is maintained through mitotic divisions (but reset after meiosis). This has long indicated an epigenetic basis to this aspect of the vernalization response.

Addressing the mechanisms of vernalization using Arabidopsis as a model Our understanding of the molecular processes that underpin much of plant biology has been transformed in the last 10 years through molecular genetic approaches and the use of Arabidopsis as a model. This is especially true of recent progress in dissecting the mechanisms of flowering-time control. Arabidopsis is native to temperate latitudes and its flowering is promoted in response to LDs and vernalization (Yanovsky and Kay, 2003; Boss et al., 2004; Henderson and Dean, 2004). Flowering in Arabidopsis is controlled by genetically

Vernalization

29

separable pathways that include light quality, photoperiod, gibberellin signal transduction, autonomous and (in winter annuals), vernalization requirement and response pathways (Yanovsky and Kay, 2003; Boss et al., 2004; Henderson and Dean, 2004). Activity through these pathways is integrated in a quantitative manner by the upregulation of an overlapping set of target genes known as floral pathway integrators that include FLOWERING LOCUS T (FT ), AGAMOUS-LIKE (AGL20) and LEAFY (LFY) (Yanovsky and Kay, 2003; Henderson and Dean, 2004). These factors in turn activate floral meristem-identity genes like APETALA1 (AP1) that execute the floral transition. The study of vernalization in Arabidopsis has facilitated the discovery and characterization of components involved in this process. In addition, the wider exploitation of Arabidopsis as a model has meant that the understanding of vernalization can be embedded and integrated with other aspects of flowering-time control.

Vernalization Requirement A genetic approach has been used to determine the molecular basis of vernalization requirement in several plant species. By crossing vernalization-requiring accessions with non-vernalization-requiring accessions of the same species, F2 populations segregating for this trait have been generated that have allowed a map-based cloning approach to identify the genes involved. This strategy has been successful in identifying major genes controlling vernalization requirement in Arabidopsis, brassicas and temperate cereals.

Arabidopsis vernalization requirement Early characterization of the vernalization requirement of Arabidopsis was described by Napp-Zinn during the 1960s (Napp-Zinn, 1987). These physiological and genetic experiments established that the process of vernalization in Arabidopsis was similar to that observed in other species, and that only a few genes with large effects determined vernalization requirement. Recent discoveries have revealed many of the underlying mechanisms regulating this phenotypic response, and are largely consistent with these early observations. Since 1999, there has been rapid progress in the identification and characterization of many genes that regulate vernalization. For ease of discussion, these genes are described in a table (Table 2.1) and their effects presented in a model (Fig. 2.1). The model proposed is the result of a large number of laboratories’ efforts to identify and characterize flowering-time genes in Arabidopsis. Our model only includes genes that have been shown to have an effect on vernalization or the expression of FLOWERING LOCUS C (FLC), a major regulatory point of vernalization, and does not include other flowering time–control pathways. Here we will divide the genes affecting vernalization into two broad categories: (i) those that confer a vernalization requirement by controlling the ‘constitutive

30

Table 2.1. Genes determining vernalization requirement and mediating response in Arabidopsis. Normal effect on FLC Gene Requirementa FRI family Autonomous pathway

Partially FLC independent

Unclear/unknown a

Interactions/complex

FLC

MADS domain

Homo/heterodimers?

FRI FRL1, FRL2 FCA FY FPA LD

n.d. n.d. Direct interaction

FLK FVE FLD ELF5 ESD4 PIE1 VIP3 VIP4

Coiled-coil Coiled-coil RNA binding Polyadenylation RNA binding Homeodomain (RNA binding?) RNA binding WD-repeat Polyamine oxidase Nuclear localized SUMOylation ISWI WD-repeat Novel hydrophilic protein

VIN3 VRN1 VRN2

PHD finger B3 DNA–binding Polycomb group

MAF2 FLH

MADS domain Not clonedc

d

 Histone deacetylasee e direct Histone deacetylase n.d. n.d. n.d. n.d. n.d. Chromatin remodelling Polycomb Homo/heterodimers?d

mRNA expression

Chromatin

n.a.

n.a.

Increase Increasef Decrease Decrease Decrease Decrease

No change in H4Ac n.d. No change in H3Ac or H4Ac n.d. No change in H4Ac No change in H4Ac

Decrease Decrease Decrease Slight increase? Increase Increase Increase Increase

n.d. # H3Ac, # H4Ac # H4Ac n.d. n.d. n.d. n.d. n.d.

(Decrease during cold) (Maintain after cold) (Maintain after cold)

#H3Ac, "H3K27Me, "H3K9Me, #H3Ac, "H3K9Me Decreases DNase I accessibility; #H3Ac, "H3K27Me, "H3K9Me n.d n.d.

No effect n.d.

Genes determining the level of FLC expression prior to vernalization. Genes mediating the cold-induced reduction or post-cold level of FLC expression. c FLH is a quantitative trait locus (QTL) with some alleles that enhance the vernalization response. d MADS-domain proteins often act as homodimers or with other MADS proteins as heterodimers. e Inferred on the basis of homology to other species. f In conjunction with FRI. Abbreviations: n.d. – not determined, H3 – Histone H3, H4 – Histone H4, Ac – acetylation, Me – methylation, K9 – Lysine 9, K27 – Lysine 27. b

A.R. Gendall and G.G. Simpson

Responseb Cold-induced repression Maintenance

Protein function/type

Vernalization

31

Vernalization Post-cold maintenance

Cold-induced repression Cold

VRN1 VRN2

VIN3

AGL20 FLC expression

FT LFY Floral pathway integration

FRL1

FRI Autonomous pathway FLD, FVE, FPA, FY, FCA, LD, FLK

ESD4, ELF5 PIE1 VIP3 VIP4

‘Constitutive’ expression

Fig. 2.1. The regulation of FLC expression controls Arabidopsis vernalization requirement and response. Genes that promote elevated levels of FLC mRNA expression are indicated by an arrow, while genes that repress FLC mRNA accumulation are indicated with a crossed bar.

level’ of FLC expression; and (ii) those that mediate the vernalization response by repressing FLC expression in response to cold (see Fig. 2.1). The demarcation or distinction between these categories is not absolute, but seems to fit the current data well. The characterization of loci that confer a vernalization requirement in Arabidopsis led to the discovery of FRIGIDA (FRI) and FLC (Michaels and Amasino, 1999; Sheldon et al., 1999; Johanson et al., 2000). Arabidopsis accessions with active alleles of both of these genes are generally late flowering and are very responsive to a vernalization treatment. Molecular analysis has revealed that the presence of active FRI alleles leads to an increased level of FLC mRNA expression. This accounts for the function of FRI, as early flowering null alleles of flc are epistatic to FRI (Michaels and Amasino, 2001). FRI encodes a plant-specific coiled-coil domain protein, and is a member of a small group of related genes (FRIGIDA LIKE, FRL) (Johanson et al., 2000; Michaels et al., 2004). The activity of one of these genes (FRL1) is required for FRI to upregulate FLC expression, but FRL1 itself is not sufficient to delay flowering as the overexpression of FRL1 does not delay flowering in an early flowering accession (Michaels et al., 2004). FLC encodes a member of the

32

A.R. Gendall and G.G. Simpson

MADS-domain family of transcription factors, a large family with over 100 members in Arabidopsis (Parenicova et al., 2003). The importance of FLC in determining the vernalization requirement of Arabidopsis is a consequence of its strong repressive effect on flowering time, and its ability to be downregulated by exposure to low temperatures (Michaels and Amasino, 1999; Sheldon et al., 1999). The comparison of winter- and summer-annual (or rapid cycling) Arabidopsis accessions has revealed that the rapid-cycling habit of Arabidopsis evolved from a winter-annual habit most commonly through loss of FRI function or, more rarely, as a result of weak alleles of flc. Several different FRI and FLC alleles have been described (Johanson et al., 2000; LeCorre et al., 2002; Gazzani et al., 2003; Michaels et al., 2003, 2004). Inactive fri alleles are usually found in early flowering accessions and active FRI alleles are generally associated with late-flowering, vernalization-responsive accessions. However, in some cases the lateness that can be conferred by active FRI alleles is masked by weak flc alleles (Gazzani et al., 2003; Michaels et al., 2004). Naturally occurring allelic variation at FLC was identified first in the Landsberg erecta (Ler) accession in a series of elegant genetic experiments (Lee et al., 1993). Recently, the nature of this allelic variation was attributed to the insertion of a transposon within the first intron of FLC, a region previously shown to be important for FLC regulation (Sheldon et al., 2002; Gazzani et al., 2003; Michaels et al., 2003). This insertion reduces the ability of FLC to be upregulated by FRI, but exactly how this occurs is unclear. Presumably, this is the result of altered transcription, pre-mRNA maturation or turnover of FLC mRNA. The weak flc allele of the Da(1)-12 accession also appears to be due to a large retrotransposon insertion within intron 1 of FLC (Michaels et al., 2003). Analysis of crosses and recombinant inbred line (RIL) populations using the Shahdara accession has also revealed that it too carries a weak flc allele – the exact nature of this flc allele has not yet been resolved but it is not the result of any large insertions or deletions as it is in Ler and Da(1)-12 (Johanson et al., 2000; Loudet et al., 2002; Michaels et al., 2003). Other accessions with as yet unexplained weak flc alleles include the C24 and Niederzenz accessions (Sanda and Amasino, 1996; Schlappi, 2001). Further study of allelic variation at these loci will undoubtedly inform us on both the evolution of flowering-time variation and the function of these important genes. It is notable that, in contrast to FRI, no naturally occurring null FLC alleles have been identified. This may indicate that in the wild Arabidopsis requires a low level of FLC activity (as experimentally induced flc null mutations are not lethal). It is not clear why this may be, but one additional function for FLC is in regulating circadian-clock function (Swarup et al., 1999). FLC is part of a smaller subgroup of six very closely related genes named MADS AFFECTING FLOWERING (MAFs), some of which also affect flowering time (Ratcliffe et al., 2001; Scortecci et al., 2001). FLC appears to be the most important of these with respect to vernalization requirement and response. It is important to mention that although the regulation of FLC is the most important feature of Arabidopsis vernalization requirement, it is not the only one; flc null mutants exhibit a vernalization response (Michaels and Amasino, 2001). Interestingly, some of the MAF genes are also regulated by

Vernalization

33

vernalization in a manner similar to FLC, while others are regulated in the reciprocal pattern – i.e. upregulated by vernalization (Ratcliffe et al., 2003). The second group of factors determining FLC mRNA expression levels is the autonomous pathway which was originally described as a genetic pathway that regulates flowering time independently of photoperiod (Koornneef et al., 1991; Simpson, 2004). The autonomous pathway consists of at least seven members including FCA, FY, FPA, FVE (these genes do not have complete names), LUMINIDEPENDENS (LD), FLOWERING LOCUS D (FLD) and FLOWERING LOCUS K (FLK). While most of the corresponding mutants have been known for some time, FLK (Lim et al., 2004) was only recently identified, suggesting that additional components are yet to be discovered. None of the autonomous pathway components appear to regulate one another, revealing that this is a genetically defined ‘pathway’ and not a pathway in the sense of a sequential series of biochemical changes (Simpson, 2004). All autonomous pathway components prevent the accumulation of FLC mRNA (Michaels and Amasino, 2001). The corresponding loci of all seven known members have now been identified, but it is still unclear how they regulate FLC. FCA, FY, FPA and FLK encode RNA-binding or RNA-processing factors, suggesting that they may act to regulate FLC mRNA synthesis or stability – however, a direct effect on FLC mRNA has not been demonstrated (Macknight et al., 1997; Schomburg et al., 2001; Simpson et al., 2003; Lim et al., 2004). LD encodes a homeodomain protein (Lee et al., 1994), but it remains unclear what DNA target (if any) LD binds. The observation that many other autonomous pathway genes bind or regulate RNA might indicate that LD interacts with RNA as has been demonstrated for some homeodomain proteins (Dubnau and Struhl, 1996; RiveraPomar et al., 1996). Both FVE and FLD encode putative components of a histone deacetylase complex and have recently been shown to be required for the normal deacetylation of histone 3 and/or histone 4 at the FLC gene (He et al., 2003; Ausin et al., 2004). The association of histone deacetylation with reduced levels of gene expression is well established in many systems (Carrozza et al., 2003). Neither FCA, FY, FPA nor LD affects the acetylation status of histones around the FLC locus revealing that multiple mechanisms are used in the autonomous pathway to control FLC expression (He et al., 2003). Autonomous pathway mutants flower late because they have elevated levels of FLC mRNA (Michaels and Amasino, 2001). Like active FRIcontaining winter-annual accessions, this lateness can be overcome by vernalization as this reduces FLC mRNA levels (Koornneef et al., 1991; Michaels and Amasino, 1999; Sheldon et al., 1999). Mutations in autonomous pathway genes effectively confer a ‘synthetic’ vernalization requirement and such similar mutations could theoretically account for winter annualism in some Arabidopsis accessions. However, there is no evidence that this occurs. This may be because these genes play other essential roles (Simpson, 2004) and so the selective pressure on these functions may reduce allelic variation. As most recent attention has focused on extreme accessions that flower particularly late or early, weak allelic variation of autonomous pathway genes may account for quantitative trait loci (QTL) that modify vernalization requirement or flowering time through more modest effects on FLC expression levels.

34

A.R. Gendall and G.G. Simpson

Some additional genes, although not similar to each other in sequence or function, affect FLC expression. These genes – EARLY FLOWERING 5 (ELF5), EARLY IN SHORT DAYS 4 (ESD4), PHOTOPERIOD-INDEPENDENT EARLY FLOWERING 1 (PIE1), VERNALIZATION INDEPENDENCE 3 (VIP3) – are required for the promotion of FLC expression. As a result, the early flowering phenotype of these mutants is clearer in FRI-containing or autonomous pathway-mutant backgrounds, when FLC mRNA levels would otherwise be elevated. All of these genes except ELF5 affect floralorgan development also, indicating that they regulate the expression of genes other than FLC (Reeves et al., 2002; Zhang and van Nocker, 2002; Noh and Amasino, 2003; Zhang et al., 2003; Noh et al., 2004). Such genes may be intermediate targets by which FRI or autonomous pathway components regulate FLC expression, but currently there is no evidence for this. Genes that affect AGAMOUS pre-mRNA maturation also affect FLC levels, as hua1 hua2 double mutants (hua is Chinese for flower) have reduced FLC expression (Cheng et al., 2003).

A role for FLC in controlling flowering time is conserved in brassicas Flowering-time variation in an F2 population derived from a cross between an annual and a biennial cultivar of oilseed Brassica rapa identified two major QTLs: vernalization-responsive flowering in B. rapa 1 (VFR1) and vernalization-responsive flowering in B. rapa 2 (VFR2). The VFR2 phenotype co-segregates with a marker derived from Arabidopsis FLC (Kole et al., 2001). The expression of B. rapa FLC-like sequences is elevated in biennials compared to annuals and reduced by vernalization (Kole et al., 2001). These observations indicate that vernalization requirement in B. rapa is controlled by allelic variation at genes related to Arabidopsis FLC. The molecular basis of the allelic variation that might account for this is currently unknown. Multiple FLClike genes have been identified in Brassica napus that delay flowering when expressed from the cauliflower mosaic virus 35S promoter in transgenic Arabidopsis plants (Tadege et al., 2001). Multiple FLC-like genes have been identified in B. rapa and display an additive effect on flowering time (Schranz et al., 2002). There is no evidence yet for a role of a gene related to Arabidopsis FRI in controlling vernalization requirement in Brassica.

Vernalization requirement evolved independently in temperate cereals Temperate cereals like wheat, barley, rye and oats exhibit a vernalization requirement, whereas rice and maize, which are of a more subtropical origin, do not. The analysis of vernalization requirement in cereals reveals that it involves different genes to Arabidopsis, indicating that this requirement has evolved independently. The major loci controlling vernalization requirement in the diploid wheat Triticum monococcum are VRN1 and VRN2 (in cereals the VERNALIZATION, VRN, designation applies to genes conferring a vernaliza-

Vernalization

35

tion requirement and differs from Arabidopsis genes of the same name that are involved in vernalization response, see below). These two genes have strong epistatic interactions and are likely to be part of the same regulatory pathway (Tranquilli and Dubcovsky, 2000). VRN1 is located on wheat chromosome 5. Genes controlling vernalization requirement in barley (Spring growth habit2 [Sgh2]) and rye (Spring growth habit1 [Sp1]) map to syntenic regions, as judged by their co-segregation with common markers, indicating that related genes control vernalization requirement in these different cereals. Wheat VRN2 also maps to chromosome 5 and a gene influencing vernalization requirement in barley maps to a syntenic region on barley chromosome 4H (Sgh1). VRN1 is the major determinant of vernalization requirement in hexaploid wheat. No VRN2 equivalent has been detected in hexaploid wheat, presumably because homozygosity for recessive alleles on all three genomes would be needed for the phenotype to be seen. A third gene controlling vernalization requirement in barley, Sgh3, has been mapped to the long arm of chromosome 1H (Shahla and Tsuchiya, 1990). Significant progress in characterizing the molecular basis of temperate cereal vernalization requirement has been made by Jorge Dubcovsky and coworkers using diploid wheat, T. monococcum, as a genetically tractable background. VRN2 was identified by positional cloning (Yan et al., 2004) and encodes a protein with a putative zinc finger and a so-called CCT (CO, COlike, TOC1) domain. A related gene lies adjacent to VRN2 and the pair are also known as ZCCT1 and ZCCT2, with VRN2 corresponding to ZCCT1 (Yan et al., 2004). VRN2 mRNA levels are reduced by vernalization treatment and this is consistent with the idea that VRN2 is a repressor of flowering eliminated by vernalization (Yan et al., 2004). The function of the CCT domain is unknown, but it is essential for the function of CONSTANS (CO), an Arabidopsis protein that controls flowering in response to daylength (Robson et al., 2001). Among the 49 spring T. monococcum accessions tested, 22 of them carry a point mutation at position 35 of the CCT domain that replaced an arginine (R) with a tryptophan (W). This R is highly conserved in CCT domains and is essential for CO function as evidenced by the late-flowering phenotype of the Arabidopsis co-7 allele that carries a point mutation at this position (Robson et al., 2001). A further 17 spring accessions harboured a deletion of the entire VRN2 gene. These data indicate that in wheat, spring habit can evolve from winter habit through loss of VRN2 function. Consistent with this, transgenic RNA interference (RNAi)-mediated knock-down of VRN2 expression in a hexaploid winter-wheat background resulted in earlier flowering in the absence of a vernalization treatment (Yan et al., 2004). A clue to the independent evolution of vernalization requirement came from the absence of Arabidopsis FLC-related sequences in cereals. However, the discovery of VRN2 and analysis of the relatedness of known CCT domains underlines the independent evolution of this regulatory mechanism. The CCT domains of T. monococcum ZCCT1 and ZCCT2, and homologues from winter barley fall into a group that do not include any Arabidopsis or rice proteins (Yan et al., 2004). This group is most closely related to a group comprised only of grass species CCT domains. This indicates that the ancestor

36

A.R. Gendall and G.G. Simpson

of VRN2 originated in the grasses and diverged substantially in the temperate cereal species. This may have occurred through duplication of a related CCT protein as ZCCT orthologues are absent from the colinear region of rice. Dubcovsky’s group also identified a candidate gene for T. monococcum VRN1 (Yan et al., 2003). Using a positional-cloning approach, a gene encoding a MADS-box transcription factor that most closely resembles the Arabidopsis floral meristem-identity gene, AP1, was found to co-segregate with vernalization requirement (allelic variation at AP1 or its paralogues, FRUITFULL and CAULIFLOWER, has not been associated with Arabidopsis vernalization requirement). This gene was expressed in apices and the level of expression increased after a vernalization treatment. Three spring accessions of T. monococcum showed deletions of sequence within the promoter of this gene that a winter accession did not (Yan et al., 2003). Therefore on the basis of map position, expression pattern in response to vernalization and promoter sequence differences between spring and winter wheat accessions, this MADSbox protein was putatively identified as VRN1 (Yan et al., 2003). Caveats to the definitive identification of this gene as VRN1 remain: (i) it has not yet proved possible to generate complete physical coverage of the chromosomal region flanked by markers that co-segregate with the VRN1 phenotype; (ii) no loss-offunction VRN1 mutation has been described that might reveal a phenotype consistent with a role in flowering-time control; and (iii) no functional or causal connection between the promoter deletions and an effect on VRN1 regulation by VRN2 has yet been established. This does not mean that this is not VRN1, but simply that there are gaps in our knowledge. Brian Fowler, Fathey Sarhan and co-workers identified the same MADS-domain transcription factor as being associated with vernalization in hexaploid wheat at around the same time as the Dubcovsky group were working with T. monococcum (Danyluk et al., 2003). Using near-isogenic lines of wheat and working also with barley, they showed that the expression of mRNA encoding this gene (which they named TaVRT-1) was dependent on a vernalization treatment in winter genotypes, but could be detected in the absence of vernalization in spring genotypes. Furthermore, they demonstrated that expression of TaVRT-1 was sensitive to photoperiod, with expression being higher in LDs than in short day (SD) photoperiods (Danyluk et al., 2003). Using wheat deletion lines they mapped this gene to the same region as Vrn-A1 and Vrn-D1. The analysis of expression of the same gene in multiple barley cultivars revealed a less clear pattern. However, Trevaskis et al. (2003) detected expression of the same gene (which they referred to as BM5) in a winter barley that had not been vernalized and failed to detect expression levels similar to vernalized material in several spring barleys. The characterization of VRN2 has made it possible to conceive a model to explain aspects of cereal vernalization at the molecular level. VRN2 is a repressor of flowering. Vernalization reduces VRN2 mRNA expression, removing this repression (Yan et al., 2004). VRN1 expression then increases and VRN1 promotes the floral transition. This interpretation is consistent with genetic data revealing epistasis between VRN1 and VRN2 (Tranquilli and Dubcovsky, 2000). In addition, transgenic RNAi knock-down of VRN2 mRNA results in elevated levels of VRN1 mRNA expression, providing

Vernalization

37

molecular evidence that VRN1 lies downstream of VRN2 and is ultimately repressed by VRN2 (Yan et al., 2004). VRN2 may repress VRN1 directly by binding to the VRN1 promoter through its zinc-finger domain. Spring habit has evolved from winter habit either through loss of VRN2 function (Yan et al., 2004) (through either mutation of the VRN2 CCT domain or deletion of VRN2 entirely) or through loss of VRN1 sensitivity to VRN2-mediated repression. It has been proposed that this may be the result of deletion of sequences in the VRN1 promoter required for VRN2-mediated regulation (Yan et al., 2003). A model depicting these interactions is shown in Fig. 2.2. As vernalization requirement in barley maps to positions syntenic with VRN1 and VRN2 known as Sgh2 and Sgh1, respectively or VRN-H1 and VRN-H2, this indicates that vernalization requirement involves related genes in wheat and barley and that differences between winter and spring habit involved mutation of similar genes. A gene homologous to VRN2 (ZCCT1) is present in a colinear region of winter barley and can be identified by DNA hybridization using ZCCT1 as a probe (Yan et al., 2004). All but one spring barley examined by Yan et al. (2004) has a deletion of ZCCT-like sequences. Research with barley reveals another gene interacting with VRN-H1 and VRN-H2, known as Sgh3. Winter alleles at all three loci are required for a strong vernalization response to be shown (Takahashi and Yasuda, 1971). Perhaps VRN-H2 also represses Sgh3. Allelic variation at three loci would provide for considerable flexibility or variation in the quantitative nature of the vernalization requirement. Sgh3 may be restricted to barley, or it may simply be that allelic variation at the corresponding locus in T. monococcum has not yet been identified.

Vernalization Daylength VRN2

FRI Vernalization Daylength

FLC

FLC MADS domain VRN1

VRN2 Zinc finger, CCT domain

Flowering

Flowering Arabidopsis

FRI Novel, coiled-coil domains

VRN1 MADS domain Wheat

Fig. 2.2. Vernalization requirement evolved independently in Arabidopsis and temperate cereals. Different genes control vernalization requirement in wheat and Arabidopsis. Promotive functions are indicated by arrows and repressive functions are indicated by crossed bars.

Vernalization Response Progress towards the characterization of the molecular basis of vernalization response has so far depended on induced mutations and the analysis of natural variation in this trait. Arabidopsis mutants have been identified that no longer

38

A.R. Gendall and G.G. Simpson

display accelerated flowering in response to a long cold treatment (Chandler et al., 1996). In addition, QTLs in vernalization response have been identified in RIL populations of Arabidopsis (Alonso-Blanco et al., 1998). Currently, nothing is known of the mechanisms of vernalization response in species other than Arabidopsis.

Arabidopsis vernalization response Consistent with the expectation that vernalization response has an epigenetic component, the few Arabidopsis genes so far identified as being involved in this process are required for the epigenetic regulation of FLC. One of the genes that mediates the vernalization response, VERNALIZATION-INSENSITIVE 3 (VIN3), appears to be important for the perception of long periods of low temperature (Sung and Amasino, 2004). VIN3 encodes a putatively nuclear-localized protein containing both a plant homeodomain (PHD) and fibronectin type II domain (Sung and Amasino, 2004). The expression of VIN3 mRNA increases slowly during the cold, and thus correlates with the quantitative nature of vernalization. vin3 mutants are unable to repress FLC during exposure to low temperatures, and as a consequence flower late (Sung and Amasino, 2004). Exactly how VIN3 represses FLC under these conditions is not known, but it is likely that VIN3 acts directly on FLC, as FLC DNA can be immunoprecipitated with an anti-VIN3 antibody in a chromatin immunoprecipitation (ChIP) assay (Sung and Amasino, 2004). The cold-induced reduction in FLC mRNA levels is accompanied by changes in histone modification that are indicators of chromatin structure. However, the sequence of these events is not entirely resolved. Two recent reports examined FLC chromatin structure by ChIP using antibodies against modified histones. Following vernalization, there is apparently a decrease in the level of acetylation of histone 3 (on lysine 9, H3K9 and lysine 14, H3K14) (Sung and Amasino, 2004), but this result is not entirely consistent as other work has shown no significant alterations in histone acetylation near FLC (Bastow et al., 2004). This apparent discrepancy may be the result of the different ages of the plants used in the experiments – as it is clear that changes in histone modifications take place both during and subsequent to the exposure to low temperatures (Sung and Amasino, 2004). More clear-cut are the increases in H3K9 methylation and histone 3 lysine 27 (H3K27) dimethylation in regions near the start of FLC transcription (Bastow et al., 2004; Sung and Amasino, 2004). Regions within intron 1 of FLC are also modified by vernalization, as exhibited by a decrease in methylation of H3K4; but interestingly H3K4 methylation was not altered in other regions of FLC, indicating localized changes in chromatin modifications (Fig. 2.3) (Bastow et al., 2004). Exactly how long exposure to low temperature induces changes in histone modifications remains unknown. VIN3 certainly contributes, as these vernalization-induced changes are not observed in a vin3 mutant (Sung and Amasino, 2004), indicating that VIN3 is required (perhaps indirectly) for the acetylation and methylation changes of histones. What is clear is that the

Vernalization

39

FLC transcription

FLC expression +++

Low temperature

VIN3

-

Return to warm temperature

K27 K9

VRN2-containing H3-methylating complex

-

Acetylation Lysine dimethylation

recruitment? K27 K9

VRN1-containing complex?

VRN1 repression?

-

K27 K9

Fig. 2.3. Chromatin changes during and after vernalization. The diagram indicates a region of chromatin near the start of FLC transcription, with histones shown as spheres bound to DNA indicated as a black line. The decrease in histone acetylation observed in some experiments is indicated by unfilled flags. The precise sequence of histone modifications and complex recruitment, and the composition of the modifying complexes are yet to be fully determined.

products of two genes VRN1 and VRN2 are required to maintain FLC in a repressed state (Gendall et al., 2001; Levy et al., 2002). The Arabidopsis VRN genes are involved in mediating the vernalization response, and as such are distinct functionally and genetically from the wheat VRN genes. VRN1 and VRN2 act to stably repress FLC upon the transition to warm temperatures – in vrn1 and vrn2 mutants FLC expression begins to rise when plants are returned

40

A.R. Gendall and G.G. Simpson

to warm conditions rather than remaining low as it would in a wild-type plant. VRN1 encodes a B3 DNA–binding protein and VRN2 encodes a Polycombgroup protein (Gendall et al., 2001; Levy et al., 2002). The VRN2-related protein from Drosophila, Su(z)12, functions as part of a chromatin-remodelling complex that induces H3K9 methylation (Czermin et al., 2002; Kuzmichev et al., 2002; Muller et al., 2002). It is perhaps not surprising then that the histone modifications in vrn1 and vrn2 mutants are altered when compared to the wild-type – with VRN2 and VRN1 being required for increased methylation of H3K9, and VRN2 also required for methylation of H3K27 (Fig. 2.3). The model from this biochemical and genetic analysis suggests that the exposure to low temperature probably induces changes in histone acetylation, mediated by VIN3. This is accompanied by (or may induce) reduced FLC expression. These acetylated histone tails at FLC may then serve as markers for the recruitment or activation of a histone methylating-complex, probably containing VRN2. This may then also recruit VRN1 specifically to FLC, as VRN1 has been shown to bind DNA non-sequence specifically in vitro, but appears to regulate relatively few genes apart from FLC in planta (Levy et al., 2002). The binding of VRN1 and a VRN2-containing Polycomb complex would be required for the normal repression of FLC after vernalization (Fig. 2.3). Two other loci have been implicated in vernalization response in Arabidopsis. The MADS AFFECTING FLOWERING 2 (MAF2) gene was recently shown to be a repressor of the response to vernalization, as maf2 mutants flower earlier after vernalization than wild-type plants, particularly when exposed to short durations (up to 21 days) of low temperature (Ratcliffe et al., 2003). This acceleration of flowering was not correlated with lower than normal FLC levels, suggesting that MAF2 acts independently of FLC to repress flowering induced by vernalization. In addition, plants overexpressing MAF2 were unresponsive to vernalization, despite being able to repress FLC normally. It appears then that MAF2 represses flowering, particularly during relatively short exposure to low temperature by an FLC-independent pathway or perhaps by acting downstream of FLC. The second locus implicated in vernalization response in Arabidopsis is the FLH QTL (Alonso-Blanco et al., 1998). Alleles from the Cape Verde Island (CVI) accession enhanced the vernalization response, relative to alleles from the Ler accession (Alonso-Blanco et al., 1998). It is not known whether this effect is FLC independent like the maf2 acceleration, but interestingly, the FLH QTL maps to the same region of the Arabidopsis genome as MAF2 (ARG, unpublished). Why the CVI accession that originates from close to the equator might have alleles to enhance the vernalization response is not clear, but perhaps it evolved as a survival mechanism to trigger flowering prior to the harsh conditions of summer.

Vernalization response in species other than Arabidopsis Nothing is currently known of the molecular mechanisms involved in the vernalization response of species other than Arabidopsis. Sequences highly

Vernalization

41

related to Arabidopsis VRN1 do not appear to be conserved in cereals. It is less clear how conserved Arabidopsis VRN2 may be. This is because there are three related genes in Arabidopsis, EMBRYONIC FLOWER 2, FERTILIZATION INDEPENDENT SEED 2 and VRN2, that control the epigenetic regulation of several genes and function in different processes (Gendall et al., 2001). It is, however, not clear that this is true of all plants – so the existence of genes related to VRN2 is not necessarily indicative of a function in vernalization response. In contrast, genes related to VIN3 appear to be widely conserved. The isolation of mutations in sequences related to VIN3 may provide a means to determine if a similar epigenetic mechanism controls vernalization response in other species. In addition, the identification of T. monococcum VRN2 as a target of vernalization will make it possible to determine whether histone modifications, consistent with the epigenetic regulation that is an essential feature of the vernalization response of Arabidopsis, also control the vernalization response in cereals.

Connecting cold to vernalization While we have recently acquired insight into the molecular basis of vernalization requirement and response, a fundamental gap in our knowledge remains the elucidation of how cold is perceived and signalled to factors that execute vernalization. Although work with Arabidopsis has recently shed light on the process of cold acclimation, which prepares the plant to withstand cold (Thomashow, 2001), there is little to indicate that this cold response connects with vernalization. First, cold acclimation is a short-term response that is not remembered after return to warmer conditions. In contrast, vernalization takes place over a period of weeks and the memory of cold exposure is retained through mitosis. Second, the overexpression of CBF1, a transcription factor required for the cold-acclimation response, does not affect FLC expression, and none of the mutations so far identified that affect vernalization response affect cold acclimation (Liu et al., 2002). The discovery, however, that the expression of VIN3 mRNA (but not Arabidopsis VRN1 or VRN2 mRNA) is upregulated in response to cold may provide the key to unlocking this process. By focusing on the factors required for the cold-dependent upregulation of VIN3, it should be possible to define the signal transduction cascade that links cold perception to FLC chromatin modification in the Arabidopsis vernalization response.

Evolution of Vernalization Some aspects of flowering-time control appear to be widely conserved. For example, although Arabidopsis promotes flowering in response to LDs, and rice does so in response to SDs, many of the genes involved in photoperiod response are conserved (Hayama et al., 2003). Furthermore, the order in which these genes act is also conserved, with only a small reversal of function

42

A.R. Gendall and G.G. Simpson

of one component in the pathway in rice conferring floral repression rather than floral promotion in LDs (Hayama et al., 2003). As a result, Arabidopsis is a useful model for the study of photoperiodism. In contrast, vernalization requirement has evolved independently in Arabidopsis and cereals. This difference in the conservation of different aspects of flowering-time control might be explained by the number of potential alternative targets that could confer a vernalization requirement. As flowering-time control features a network of multiple pathways feeding into overlapping floral pathway integrators, which in turn control floral meristem-identity genes, there are many alternative mechanisms by which floral repression could evolve. If these were sensitive to cold, they could establish independent mechanisms of conferring a vernalization requirement. In contrast, the photoperiodic control of flowering is dependent on a circadian clock time measurement mechanism (Mouradov et al., 2002; Yanovsky and Kay, 2003). The complexity of the clock and its role in controlling many other processes may make it recalcitrant to reinvention (Simpson, 2003). A second consideration may be the distinct rationales that provide a selective advantage for winter-annual versus summer-annual habit. This may have led to the independent evolution of winter- or summer-annual habit on multiple occasions, but in different environments for different reasons. At higher latitudes, where the growing season is relatively short, winter annualism can provide a head start in development. Plants that germinate prior to winter and overwinter in the vegetative state may be able to complete their life cycle in the relatively short growing season of the following spring and summer. At lower temperate latitudes, like North Africa, this head start in development can provide a drought-avoidance mechanism that allows life cycle completion in spring, prior to the hazardous dry and hot conditions of summer. In contrast, higher-latitude environments with winters too extreme for overwintering survival may favour a rapid-cycling habit. This rapidcycling strategy can also provide a mechanism by which opportunists can complete their life cycle on disturbed ground. This is not an exhaustive list, but conjecture on different potential drivers that may favour winter- or summerannual habit. A survey of 749 landraces of wheat from different parts of the world revealed that winter wheat is typically grown in regions with relatively mild mean midwinter temperatures ( 78C to þ48C). Lower mean winter temperatures result in frost damage, while higher mean winter temperatures do not satisfy vernalization requirement. In these environments, spring varieties dominate (Iwaki et al., 2001). Similar analyses of Arabidopsis accessions failed to identify latitudinal clines between winter and rapid-cycling accessions (Johanson et al., 2000). However, a latitudinal cline has been identified within FRI-containing accessions; Arabidopsis accessions from southern latitudes flowered faster than accessions isolated from northern latitudes when grown together in a garden experiment in Vermont, USA (Stinchcombe et al., 2004). A strong correlation between flowering time in the garden experiment and January precipitation levels at the site of accession collection was identified. It is therefore possible that southern accessions are more responsive to

Vernalization

43

vernalization as this may align flowering more closely with rainfall. In such cases, flowering in mild winters may not be as disadvantageous as flowering in hot, dry summers (Stinchcombe et al., 2004). The adaptation to the environment that winter and summer annualism provides may also be linked to other responses. For example, overwintering in the vegetative state may require cold tolerance and if early flowering is a drought-avoidance mechanism, late flowering may be associated with increased water-use efficiency. Consistent with this, there is some evidence that lowtemperature tolerance in wheat is associated with the same chromosomal region that contains VRN1 (Danyluk et al., 2003) and it has been suggested that water-use efficiency may be related to Arabidopsis flowering time in FRI- and FLC-containing backgrounds (McKay et al., 2003). It is possible that these phenomena involve pleiotropy of function of genes involved in vernalization.

Short-day Vernalization While vernalization is the acquisition or acceleration of the ability to flower following a chilling treatment, a similar effect has been observed by exposing some plants, like wheat and barley, to SDs. Flowering of wheat and barley is promoted in response to LDs; spring varieties and vernalized winter varieties generally accelerate flowering in response to LDs. However, the response of unvernalized winter varieties to daylength is less consistent. Some winter varieties of wheat (such as cv Templar and Huntsman) (Evans, 1987) and barley (such as the landrace Arabi Abiad) (Roberts et al., 1988) display earlier flowering if growth in SDs precedes transfer to LD conditions. In some cases, SD treatment can completely suppress a requirement for low temperaturemediated vernalization (Evans, 1987). This phenomenon has been referred to as SD vernalization (Purvis and Gregory, 1937). All winter wheat varieties tested accelerate flowering in response to a low-temperature treatment, but only a subset respond to short-day vernalization (Evans, 1987). Although it has been identified in several temperate cereal species, we are not aware of any report of a similar phenomenon occurring in Arabidopsis. The process of SD vernalization has not been characterized beyond observation. Therefore, although it indicates a dependence on both daylength and temperature to monitor seasonal progression, the nature of this is not understood. For example, one interpretation is that SD vernalization is indicative of positive responses to independent cues signalling the passage of winter. Alternatively, this effect may be restricted to varieties that carry loss-of-function alleles of a gene that normally inhibits flowering in SDs. For example, two lines derived from a barley winter (cv Igri)  spring (cv Triumph) cross that showed a SD vernalization response had loss-of-function alleles of a gene that inhibits flowering in SD, PpdH1 (Laurie et al., 1995). The phenomenon is mechanistically distinct from low temperature-mediated vernalization as wheat floral primordia are produced in SD vernalization, but not low-temperature vernalization (Evans, 1987).

44

A.R. Gendall and G.G. Simpson

Perspective The cloning of Arabidopsis FLC and wheat VRN2 has identified floral repressors that confer a vernalization requirement in these plants. In addition, the characterization of these genes has revealed that they are the targets of quantitative vernalization response and that vernalization requirement evolved independently in Arabidopsis and cereals. Therefore, the identification of genes conferring a vernalization requirement in other species will probably require a forward genetic approach rather than depending on a model-derived candidate gene approach. The memory, however, of vernalization involves epigenetic control, executed in part by factors that modify chromatin in many eukaryotes. It is entirely possible therefore that this mechanism confers memory of winter in other plants as well. Nothing is yet known of the mechanisms by which cold is perceived and signalled to mediate the vernalization response. It will be interesting to dissect these mechanisms in other species and also to compare the molecular basis of bud break through dormancy chilling in woody perennials to determine the distinctive mechanisms and unifying themes by which plants control reproductive development through sensing and remembering winter.

Acknowledgements Work in ARG’s laboratory on natural variation and flowering time is supported by the Australian Research Council (DP0210592 and DP0449651). SCRI is supported by a grant-in-aid from the Scottish Executive Environment and Rural Affairs Department. Thanks to Ben Ong for comments on the manuscript and to Trevor Phillips for help with artwork.

References Alonso-Blanco, C., El Assal, S.E., Coupland, G. and Koornneef, M. (1998) Analysis of natural allelic variation at flowering time loci in the Landsberg erecta and Cape Verde Islands ecotypes of Arabidopsis thaliana. Genetics 149, 749–764. Ausı´n, I., Alonso-Blanco, C., Jarillo, J.A., Ruiz-Garcı´a, L. and Martı´nez-Zapater, J.M. (2004) Regulation of flowering time by FVE, a retinoblastoma-associated protein. Nature Genetics 36, 162–166. Bastow, R., Mylne, J.S., Lister, C., Lippman, Z., Martienssen, R.A. and Dean, C. (2004) Vernalization re-

quires epigenetic silencing of FLC by histone methylation. Nature 427, 164–167. Baurle, I. and Laux, T. (2003) Apical meristems: the plant’s fountain of youth. Bioessays 25, 961–970. Boss, P.K., Bastow, R.M., Mylne, J.S. and Dean, C. (2004) Multiple pathways in the decision to flower: enabling, promoting, and resetting. Plant Cell 16 (Suppl.), S18–S31. Carrozza, M.J., Utley, R.T., Workman, J.L. and Cote, J. (2003) The diverse functions of histone acetyltransferase complexes. Trends in Genetics 19, 321–329.

Vernalization Chandler, J., Wilson, A. and Dean, C. (1996) Arabidopsis mutants showing an altered response to vernalization. Plant Journal 10, 637–644. Cheng, Y., Kato, N., Wang, W., Li, J. and Chen, X. (2003) Two RNA binding proteins, HEN4 and HUA1, act in the processing of AGAMOUS pre-mRNA in Arabidopsis thaliana. Developmental Cell 4, 53–66. Chouard, P. (1960) Vernalization and its relations to dormancy. Annual Review of Plant Physiology 11, 191–238. Czermin, B., Melfi, R., McCabe, D., Seitz, V., Imhof, A. and Pirrotta, V. (2002) Drosophila enhancer of Zeste/ESC complexes have a histone H3 methyltransferase activity that marks chromosomal Polycomb sites. Cell 111, 185– 196. Danyluk, J., Kane, N.A., Breton, G., Limin, A.E., Fowler, D.B. and Sarhan, F. (2003) TaVRT-1, a putative transcription factor associated with vegetative to reproductive transition in cereals. Plant Physiology 132, 1849–1860. Dubnau, J. and Struhl, G. (1996) RNA recognition and translational regulation by a homeodomain protein. Nature 379, 694–699. Evans, L.T. (1987) Short day induction of inflorescence initiation in some winter wheat varieties. Australian Journal of Plant Physiology 14, 277–286. Ferra´ndiz, C., Gu, Q., Martienssen, R. and Yanofsky, M.F. (2000) Redundant regulation of meristem identity and plant architecture by FRUITFULL, APETALA1 and CAULIFLOWER. Development 127, 725–734. Gazzani, S., Gendall, A.R., Lister, C. and Dean, C. (2003) Analysis of the molecular basis of flowering time variation in Arabidopsis accessions. Plant Physiology 132, 1107–1114. Gendall, A.R., Levy, Y.Y., Wilson, A. and Dean, C. (2001) The VERNALIZATION 2 gene mediates the epigenetic regulation of vernalization in Arabidopsis. Cell 107, 525–535.

45 Hayama, R., Yokoi, S., Tamaki, S., Yano, M. and Shimamoto, K. (2003) Adaptation of photoperiodic control pathways produces short-day flowering in rice. Nature 422, 719–722. He, Y., Michaels, S.D. and Amasino, R.M. (2003) Regulation of flowering time by histone acetylation in Arabidopsis. Science 302, 1751–1754. Henderson, I.R. and Dean, C. (2004) Control of Arabidopsis flowering: the chill before the bloom. Development 131, 3829–3838. Iwaki, K., Haruna, S., Niwa, T. and Kato, K. (2001) Adaptation and ecological differentiation in wheat with special reference to geographical variation of growth habit and Vrn genotype. Plant Breeding 120, 107–114. Johanson, U., West, J., Lister, C., Michaels, S., Amasino, R. and Dean, C. (2000) Molecular analysis of FRIGIDA, a major determinant of natural variation in Arabidopsis flowering time. Science 290, 344–347. Kole, C., Quijada, P., Michaels, S.D., Amasino, R.M. and Osborn, T.C. (2001) Evidence for homology of flowering-time genes VFR2 from Brassica rapa and FLC from Arabidopsis thaliana. Theoretical and Applied Genetics 102, 425–430. Koornneef, M., Hanhart, C.J. and van der Veen, J.H. (1991) A genetic and physiological analysis of late flowering mutants in Arabidopsis thaliana. Molecular and General Genetics 229, 57–66. Kuzmichev, A., Nishioka, K., ErdjumentBromage, H., Tempst, P. and Reinberg, D. (2002) Histone methyltransferase activity associated with a human multiprotein complex containing the enhancer of Zeste protein. Genes and Development 16, 2893– 2905. Laurie, D.A., Pratchett, N., Bezant, J.H. and Snape, J.W. (1995) RFLP mapping of five major genes and eight quantitative trait loci controlling flowering time in a winter  spring barley

46

A.R. Gendall and G.G. Simpson (Hordeum vulgare L.) cross. Genome 38, 575–585. LeCorre, V., Roux, F. and Reboud, X. (2002) DNA polymorphism at the FRIGIDA gene in Arabidopsis thaliana: extensive nonsynonymous variation is consistent with local selection for flowering time. Molecular Biology and Evolution 19, 1261–1271. Lee, I., Bleecker, A. and Amasino, R. (1993) Analysis of naturally occurring late flowering in Arabidopsis thaliana. Molecular and General Genetics 237, 171–176. Lee, I., Aukerman, M.J., Gore, S.L., Lohman, K.N., Michaels, S.D., Weaver, L.M., John, M.C., Feldmann, K.A. and Amasino, R.M. (1994) Isolation of LUMINIDEPENDENS: a gene involved in the control of flowering time in Arabidopsis. Plant Cell 6, 75–83. Levy, Y.Y., Mesnage, S., Mylne, J.S., Gendall, A.R. and Dean, C. (2002) Multiple roles of Arabidopsis VRN1 in vernalization and flowering time control. Science 297, 243–246. Lim, M.H., Kim, J., Kim, Y.S., Chung, K.S., Seo, Y.H., Lee, I., Kim, J., Hong, C.B., Kim, H.J. and Park, C.M. (2004) A new Arabidopsis gene, FLK, encodes an RNA-binding protein with K homology motifs and regulates flowering time via FLOWERING LOCUS C. Plant Cell 16, 731– 740. Liu, J., Gilmour, S.J., Thomashow, M.F. and van Nocker, S. (2002) Cold signalling associated with vernalization in Arabidopsis thaliana does not involve CBF1 or abscisic acid. Physiology Plantarum 114, 125–134. Loudet, O., Chaillou, S., Camilleri, C., Bouchez, D. and Daniel-Vedele, F. (2002) Bay-0  Shahdara recombinant inbred line population: a powerful tool for the genetic dissection of complex traits in Arabidopsis. Theoretical and Applied Genetics 104, 1173–1184. Macknight, R., Bancroft, I., Page, T., Lister, C., Schmidt, R., Love, K., Westphal, L., Murphy, G., Sherson, S.,

Cobbett, C. and Dean, C. (1997) FCA, a gene controlling flowering time in Arabidopsis, encodes a protein containing RNA-binding domains. Cell 89, 737–745. McKay, J.K., Richards, J.H. and MitchellOlds, T. (2003) Genetics of drought adaptation in Arabidopsis thaliana: I. Pleiotropy contributes to genetic correlations among ecological traits. Molecular Ecology 12, 1137–1151. Michaels, S.D. and Amasino, R.M. (1999) FLOWERING LOCUS C encodes a novel MADS domain protein that acts as a repressor of flowering. Plant Cell 11, 949–956. Michaels, S.D. and Amasino, R.M. (2001) Loss of FLOWERING LOCUS C activity eliminates the late-flowering phenotype of FRIGIDA and autonomous pathway mutations but not responsiveness to vernalization. Plant Cell 13, 935–941. Michaels, S.D., He, Y., Scortecci, K.C. and Amasino, R.M. (2003) Attenuation of FLOWERING LOCUS C activity as a mechanism for the evolution of summer-annual flowering behavior in Arabidopsis. Proceedings of the National Academy of Sciences USA 100, 10102–10107. Michaels, S.D., Bezerra, I.C. and Amasino, R.M. (2004) FRIGIDA-related genes are required for the winterannual habit in Arabidopsis. Proceedings of the National Academy of Sciences USA 101, 3281–3285. Mouradov, A., Cremer, F. and Coupland, G. (2002) Control of flowering time: interacting pathways as a basis for diversity. Plant Cell 14 (Suppl.), S111– S130. Muller, J., Hart, C.M., Francis, N.J., Vargas, M.L., Sengupta, A., Wild, B., Miller, E.L., O’Connor, M.B., Kingston, R.E. and Simon, J.A. (2002) Histone methyltransferase activity of a Drosophila Polycomb group repressor complex. Cell 111, 197–208. Napp-Zinn, K. (1987) Vernalization: environmental and genetic regulation.

Vernalization In: Atherton, J.G. (ed.) Manipulation of Flowering. Butterworths, London, pp. 123–132. Noh, Y.S. and Amasino, R.M. (2003) PIE1, an ISWI family gene, is required for FLC activation and floral repression in Arabidopsis. Plant Cell 15, 1671– 1682. Noh, Y.S., Bizzell, C.M., Noh, B., Schomburg, F.M. and Amasino, R.M. (2004) EARLY FLOWERING 5 acts as a floral repressor in Arabidopsis. Plant Journal 38, 664–672. Parˇenicova´, L., de Folter, S., Kieffer, M., Horner, D.S., Favalli, C., Busscher, J., Cook, H.E., Ingram, R.M., Kater, M.M., Davies, B., Angenent, G.C. and Colombo, L. (2003) Molecular and phylogenetic analyses of the complete MADS-box transcription factor family in Arabidopsis: new openings to the MADS world. Plant Cell 15, 1538–1551. Purvis, O.N. and Gregory, F.G. (1937) Studies in vernalization of cereals. I. A comparative study of vernalization of winter rye by low temperature and by short days. Annals of Botany 1, 569– 592. Ratcliffe, O.J., Nadzan, G.C., Reuber, T.L. and Riechmann, J.L. (2001) Regulation of flowering in Arabidopsis by an FLC homologue. Plant Physiology 126, 122–132. Ratcliffe, O.J., Kumimoto, R.W., Wong, B.J. and Riechmann, J.L. (2003) Analysis of the Arabidopsis MADS AFFECTING FLOWERING gene family: MAF2 prevents vernalization by short periods of cold. Plant Cell 15, 1159–1169. Reeves, P.H., Murtas, G., Dash, S. and Coupland, G. (2002) early in short days 4, a mutation in Arabidopsis that causes early flowering and reduces the mRNA abundance of the floral repressor FLC. Development 129, 5349–5361. Rivera-Pomar, R., Niessing, D., SchmidtOtt, U., Gehring, W.J. and Jackle, H. (1996) RNA binding and translational

47 suppression by bicoid. Nature 379, 746–749. Roberts, E.H., Summerfield, R.J., Cooper, J.P. and Ellis, R.H. (1988) Environmental control of flowering in barley (Hordeum vulgare L.). I. Photoperiod limits to long-day responses, photoperiod-insensitive phases and effects of low-temperature and short-day vernalization. Annals of Botany 62, 127–144. Robson, F., Costa, M.M., Hepworth, S.R., Vizir, I., Pineiro, M., Reeves, P.H., Putterill, J. and Coupland, G. (2001) Functional importance of conserved domains in the flowering-time gene CONSTANS demonstrated by analysis of mutant alleles and transgenic plants. Plant Journal 28, 619–631. Sanda, S.L. and Amasino, R.M. (1996) Ecotype-specific expression of a flowering mutant phenotype in Arabidopsis thaliana. Plant Physiology 111, 641–644. Schlappi, M. (2001) RNA levels and activity of FLOWERING LOCUS C are modified in mixed genetic backgrounds of Arabidopsis thaliana. International Journal of Plant Science 162, 527–537. Schomburg, F.M., Patton, D.A., Meinke, D.W. and Amasino, R.M. (2001) FPA, a gene involved in floral induction in Arabidopsis, encodes a protein containing RNA-recognition motifs. Plant Cell 13, 1427–1436. Schranz, M.E., Quijada, P., Sung, S.B., Lukens, L., Amasino, R. and Osborn, T.C. (2002) Characterization and effects of the replicated flowering time gene FLC in Brassica rapa. Genetics 162, 1457–1468. Scortecci, K.C., Michaels, S.D. and Amasino, R.M. (2001) Identification of a MADS-box gene, FLOWERING LOCUS M, that represses flowering. Plant Journal 26, 229–236. Shahla, A. and Tsuchiya, T. (1990) Genetic analysis in six telotrisomic lines in barley. Journal of Heredity 81, 127–130.

48

A.R. Gendall and G.G. Simpson Sheldon, C.C., Burn, J.E., Perez, P.P., Metzger, J., Edwards, J.A., Peacock, W.J. and Dennis, E.S. (1999) The FLF MADS box gene: a repressor of flowering in Arabidopsis regulated by vernalization and methylation. Plant Cell 11, 445–458. Sheldon, C.C., Conn, A.B., Dennis, E.S. and Peacock, W.J. (2002) Different regulatory regions are required for the vernalization-induced repression of FLOWERING LOCUS C and for the epigenetic maintenance of repression. Plant Cell 14, 2527–2537. Simpson, G.G. (2003) Evolution of flowering in response to day length: flipping the CONSTANS switch. Bioessays 25, 829–832. Simpson, G.G. (2004) The autonomous pathway: epigenetic and posttranscriptional gene regulation in the control of Arabidopsis flowering time. Current Opinion in Plant Biology 7, 570–574. Simpson, G.G. and Dean, C. (2002) Arabidopsis, the Rosetta stone of flowering time? Science 296, 285–289. Simpson, G.G., Dijkwel, P.P., Quesada, V., Henderson, I. and Dean, C. (2003) FY is an RNA 3’ end-processing factor that interacts with FCA to control the Arabidopsis floral transition. Cell 113, 777–787. Stinchcombe, J.R., Weinig, C., Ungerer, M., Olsen, K.M., Mays, C., Halldorsdottir, S.S., Purugganan, M.D. and Schmitt, J. (2004) A latitudinal cline in flowering time in Arabidopsis thaliana modulated by the flowering time gene FRIGIDA. Proceedings of the National Academy of Sciences USA 101, 4712–4717. Sung, S. and Amasino, R.M. (2004) Vernalization in Arabidopsis thaliana is mediated by the PHD finger protein VIN3. Nature 427, 159–164. Swarup, K., Alonso-Blanco, C., Lynn, J.R., Michaels, S.D., Amasino, R.M., Koornneef, M. and Millar, A.J. (1999) Natural allelic variation identifies new

genes in the Arabidopsis circadian system. Plant Journal 20, 67–77. Tadege, M., Sheldon, C.C., Helliwell, C.A., Stoutjesdijk, P., Dennis, E.S. and Peacock, W.J. (2001) Control of flowering time by FLC orthologues in Brassica napus. Plant Journal 28, 545–553. Takahashi, R. and Yasuda, S. (1971) Genetics of earliness and growth habit in barley. In: Nilan, R.A. (ed.) Proceedings of the 2nd International Barley Genetics Symposium. Washington State University Press, Washington, pp. 388–408. Thomashow, M.F. (2001) So what’s new in the field of plant cold acclimation? Lots! Plant Physiology 125, 89–93. Tranquilli, G. and Dubcovsky, J. (2000) Epistatic interaction between vernalization genes Vrn-Am1 and Vrn-Am2 in diploid wheat. Journal of Heredity 91, 304–306. Trevaskis, B., Bagnall, D.J., Ellis, M.H., Peacock, W.J. and Dennis, E.S. (2003) MADS box genes control vernalization-induced flowering in cereals. Proceedings of the National Academy of Sciences USA 100, 13099–13104. Yan, L., Loukoianov, A., Tranquilli, G., Helguera, M., Fahima, T. and Dubcovsky, J. (2003) Positional cloning of the wheat vernalization gene VRN1. Proceedings of the National Academy of Sciences USA 100, 6263–6268. Yan, L., Loukoianov, A., Blechl, A., Tranquilli, G., Ramakrishna, W., SanMiguel, P., Bennetzen, J.L., Echenique, V. and Dubcovsky, J. (2004) The wheat VRN2 gene is a flowering repressor down-regulated by vernalization. Science 303, 1640–1644. Yanovsky, M.J. and Kay, S.A. (2003) Living by the calendar: how plants know when to flower. Nature Reviews Molecular Cell Biology 4, 265–275. Zhang, H. and van Nocker, S. (2002) The VERNALIZATION INDEPENDENCE 4 gene encodes a novel regulator of

Vernalization FLOWERING LOCUS C. Plant Journal 31, 663–673. Zhang, H., Ransom, C., Ludwig, P. and van Nocker, S. (2003) Genetic analysis of early flowering mutants in Arabi-

49 dopsis defines a class of pleiotropic developmental regulator required for expression of the flowering-time switch flowering locus C. Genetics 164, 347–358.

3

Signal Transduction Regulating Floral Development R.G. ANTHONY School of Biological Sciences, Royal Holloway, University of London, Egham, Surrey, TW20 0EX UK, e-mail: [email protected]

Introduction The emergence of complete genome sequences from strategic eukaryote models and recent advances in the molecular analysis of plant signalling pathways allow comparative analysis of the signal transduction pathways in plants and animals. It is now clear that the logic of signalling is combinatorial and modular rather than built around linear chains of events (Pawson, 2004). At present, we have insight only into a limited number of floral signalling modules. Furthermore, these modules, when treated individually, appear rather aspecific to stimuli and are clearly utilized in many signalling events. The components need to be considered in the network as a whole because complex systems have emergent properties that are not entirely obvious when the elements are considered in isolation. Research over the last decade has resulted in major advances in elucidating the mechanisms that regulate gene expression and in identifying the components of floral signal transduction. This chapter outlines our current knowledge of the genetically defined signal transduction pathways and the mechanisms by which signals are relayed. Before we can consider the mechanisms surrounding intracellular signalling and the cell–cell communication processes that lead to flowering, it is important to consider flowering in a whole environmental context. The precise moment in a growing season when a plant initiates flowering is a critical developmental decision, if the plant is to ensure reproductive success. In flowering plants, the timing of the transition from vegetative growth to flowering is controlled by the prevailing environmental conditions and/or intrinsic developmental signals (Colasanti and Sundaresan, 2000; Mouradov et al., 2002; Simpson and Dean, 2002). For some plants the timing of flowering and the subsequent seed set appear more critical than others. For annuals, such as Arabidopsis thaliana that flower once, set seed and die, the timing is obviously extremely important. It is now well known that flowering in Arabidopsis is induced by long-day (LD) 50

ßCAB International 2006. The Molecular Biology and Biotechnology of Flowering, 2nd edn (ed. B.R. Jordan)

Signal Transduction Regulating Floral Development

51

photoperiods (Hayama and Coupland, 2003), exposure to cold, vernalization (Henderson and Dean, 2004), and by internal factors: the autonomous pathway (Boss et al., 2004) and plant hormones such as gibberellin (GA) (Boss et al., 2004). In contrast, temperate maize is day neutral and flowers after a specific number of leaves are produced. Thus in many species the signalling pathways that control flowering have evolved to provide considerable flexibility. Built into this flexibility is the requirement that plants not only have to sense and respond to appropriate signals but also have the ability to remain unaltered by environmental fluctuations that do not represent a signal. The control of flowering has been intensively studied by genetic analysis in several plant species, particularly Arabidopsis (Koornneef et al., 1998; Amasino, 2003). These studies have revealed many of the genes involved in regulating flowering, but research into the biochemical processes in which the products of these genes participate is currently lagging behind. Furthermore, although many of the elements that trigger the floral signal transduction cascade have been identified, the specific downstream pathways and their interactions are still largely unknown. It is now evident that the pathway to flowering is highly redundant and this explains why no single mutation that prevents flowering under all conditions has been identified. However, the triple mutant co-2 fca1 ga1–3 compromised in the photoperiod-dependent, autonomous and GA-dependent pathways does not flower under either short-day (SD) or LD conditions (Reeves and Coupland, 2001). The different floral-induction pathways are ‘bottlenecked’ by a small set of genes that integrates the signals coming from all four major pathways into a unitary output. The strongest output signal arises when all four pathways are activated. This output in turn regulates the downstream floral homeotic genes which in turn produce the floral organs (Boss et al., 2004). The signalling events that ultimately result in flowering involve an extremely complex system of interacting factors that regulate the genetic elements within different signalling modules. This chapter outlines the principal signalling events regulating floral development focusing on the internal biochemical and genetic pathways and highlighting the biochemical steps that occur following reception of the signal and which ultimately lead to altered gene expression and enzyme activity.

The Genetic Signalling Pathways in Arabidopsis External stimuli/receptors Numerous environmental factors influence plant development. Temperature, light, touch, water and gravity are among the stimuli that serve as signals for the activation of endogenous developmental programmes. Of these, light and temperature have an especially important role in floral induction, although other stimuli such as water stress can also induce early flowering providing a strategic mechanism to ensure seed set before plant death. The decision to flower is based on the coordination of the different environmental signals.

52

R.G. Anthony

How do photoreceptors affect floral initiation? The photoperiodic stimulus in both LD plants and SD plants is perceived by the leaves. The pathway involves the interaction of the photoreceptors (phytochrome and cryptochrome) with a circadian clock initiating a pathway that eventually results in the expression of the gene CONSTANS (CO), which encodes a zinc-finger transcription factor that promotes flowering. CO acts through other genes to increase the expression of the floral meristem-identity gene LEAFY (LFY) (Fig. 3.1) (Izawa et al., 2002; Mouradov et al., 2002; Hayama and Coupland, 2003; Yanovsky and Kay, 2003). Thus, in response to the photoperiod, the leaf transmits a signal that regulates the transition to flowering at the shoot apex. Detailed experiments on the involvement of CO are now beginning to shed light on the nature of this floral stimulus (florigen), which has eluded researchers for decades (see below). The role and mode of action of the photoreceptors is now very well understood and is dealt with in more detail (see Chapter 1, this volume). In brief, the phytochromes (PHY) are red and far-red light photoreceptors and can exist in two photoreversible forms that are able to detect light using a covalently bound tetrapyrrole chromophore. The carboxy-terminal domain of PHY functions in dimerization and contains a region that resembles prokaryotic twocomponent histidine kinases (see below). The amino-terminal region is thought to define the photosensory activity of the PHY molecule. The cryptochromes (CRY1 and CRY2 in Arabidopsis) perceive blue/UV-A light and are found in various taxa and are thought to have evolved from photolyases. The photochemical mechanism of signal capture and transfer by CRY is likely to involve a redox reaction (Gyula et al., 2003). After activation by light, receptors initiate downstream signal propagation that results in transient or sustained physiological responses. Current models postulate that PHY induces a signalling cascade mediated by the second messengers Ca2þ and cyclic 3’,5’-GMP (cGMP) in the cytoplasm and also functions as a light-regulated kinase (Neuhaus et al., 1997; Wang and Deng, 2003). Its Pfr conformer can rapidly translocate into the nucleus, where it interacts with transcription factors and thus can directly regulate light-induced gene transcription (Nagy and Schafer, 2002). This complex signalling network is attractive and is supported by considerable data (Gyula et al., 2003). According to current interpretation of the data, the signalling cascade controlled by CRY1 and CRY2 is organized differently from that controlled by PHY. It is postulated that the blue-light perception by CRY photoreceptors triggers the rapid deactivation/degradation of COP (constitutive photomorphogenesis1: the COP1 gene represses photomorphogenesis in seedlings and flowering in darkness) by an unknown mechanism, allowing the accumulation of PHY5 in the nucleus, which in turn enhances the transcription of target genes ( (Nakagawa and Komeda, 2004). Although this model highlights the role of proteolysis, it is likely that other molecular mechanisms also play a significant role in CRY1- and CRY2-mediated signalling. It is clear that the different photoreceptors can induce different signalling cascades that partly overlap. The terminal step of signalling, the regulation of target-gene

Signal Transduction Regulating Floral Development

53

Cold vernalization pathway

External cues

FRI, FRL1

Photoperiod pathway FRL2, VIP3 Circadian clock VIP4, ART1 Ambient PIE1, ESD4 temperature Florigen pathway /CO

VRN1 VRN2 VIN3

FLC CO

Internal cues FCA, FY FLD, FVE FPA, LD FLK

Leaf number autonomous pathway Gibberellin pathway

Floral integration

FT

SOC1

LFY

Floral meristem identity

AP1 CAL FUL LFY

Floral homeotic genes

AP2, AG, AP3, PI

AGL24

ADG1 SEX1 Sucrose PGM pathway TOE1, TOE2 SMZ, SNZ, FLM SVP, TFL1, TFL2 EMF1, EMF2

Floral organs

Fig. 3.1. An overview of the major floral signalling pathways in Arabidopsis. The photoperiod, autonomous/vernalization, gibberellin, sucrose, light quality and ambient temperature pathways activate floral-pathway integrators. The CONSTANS (CO) gene functions in the photoperiod pathway, long-day photoperiods promote flowering by circadian clock-dependent and -independent mechanisms, which control the activity of CO. FLOWERING LOCUS C (FLC) is the primary upstream target of a large number of repressor and activator genes in the vernalization and autonomous pathways. See text for details. Vernalization pathway: VERNALIZATION1 (VRN1), VERNALIZATION2 (VRN2), VERNALIZATION INSENSITVE3 (VIN3). Genes promoting FLC expression and resulting in delayed flowering: FRIGIDA (FRI), FRIGIDA-LIKE1 (FRL1), FRIGIDA-LIKE2 (FRL2) VERNALIZATION INDEPENDENCE3 (VIP3), VERNALIZATION INDEPENDENCE4 (VIP4), AERIAL ROSETTE1 (ART1), PHOTOPERIOD INSENSITIVE EARLY FLOWERING (PIE1), EARLY UNDER SHORT DAYS4 (ESD4). The autonomous pathway: FCA, FY, FLOWERING LOCUS D (FLD) FVE, FPA, LUMINIDEPENDENS (LD), FLOWERING LOCUS K (FLK). Floral integrators: FLOWERING LOCUS T (FT), SUPPRESSOR OF OVEREXPRESSION OF CO1 (SOC1) AND LEAFY (LFY). Floral repressing genes: TARGET OF EAT1 (TOE1), TARGET OF EAT2 (TOE2), SCHLAFMUTZE (SMZ), SCHNARCHZAPFEN (SNZ), FLOWERING LOCUS M (FLM), SHORT VEGETATIVE PHASE (SVP), TFL1, TFL2, EMBRYONIC FLOWER1 (EMF1), EMBRYONIC FLOWER2 (EMF2). Meristem-identity genes: APETALA1 (AP1), CAULFLOWER (CAL), FRUITFUL (FUL), LFY. Floral homeotic genes: APETALA2 (AP2), AGAMOUS (AG), APETALA3 (AP3), PISTILLATA (PI). Floral promoter independent of FLC: AGAMOUS LIKE 24 (AGL24). Adapted from Henderson and Dean (2004).

54

R.G. Anthony

expression occurs predominately at the level of transcription but signal relay is significantly affected by regulated degradation and the compartmentalization of the signalling intermediates. The vernalization pathway Extended periods of cold temperature can accelerate flowering later in development in a process known as vernalization. The genetic analysis of natural variation of flowering among different geographical Arabidopsis ecotypes revealed two major genes, FRIGIDA (FRI) and FLOWERING LOCUS C (FLC), that confer the winter-annual flowering habit (Henderson and Dean, 2004). FRI encodes a novel protein with two coiled-coil domains, whereas FLC encodes a MADs-box transcription factor (Michaels and Amasino, 1999; Sheldon et al., 1999). In Arabidopsis the pathways that regulate vernalization converge on FLC (Fig. 3.1). The identification of the FLC gene has helped address a long-standing question in flowering-time research: how do plants that have been vernalized remember this signal and flower perhaps months later? The perception of cold occurs in the cells of the shoot apex and, after an extended exposure to cold, a vernalized state is induced in these cells. This state can be passed on through mitotic cell divisions even in the absence of cold, but is lost after meiosis. The recent identification and molecular analysis of two genes, VERNALIZATION1 (VRN1) and VERNALIZATION2 (VRN2), and their effects on FLC expression, suggests that epigenetic changes in chromatin structure at the FLC locus are the basis of this cellular memory of vernalization (Sung and Amasino, 2004a). Although the vrn1 and vrn2 mutants are able to perceive cold and respond by downregulating FLC mRNA levels, they are defective in their ability to remember the cold as once exposed to warm temperatures FLC mRNA returns to prevernalization levels (Gendall et al., 2001). Good clues to the mechanism of action of VRN2 come from its structure. VRN2 encodes a nuclear-localized zinc-finger protein with similarity to the Drosophila Polycombgroup (PcG) protein SU(Z)12 (Gendall et al., 2001). PcG proteins are components of complexes that repress gene expression by maintaining the chromatin in a state incompatible with transcription. This state is maintained after mitotic but not meiotic cell divisions. It is therefore proposed that VRN2 functions in a similar manner to other PcG proteins, to maintain the vernalization-induced repression of FLC. It is possible that VRN1 functions in a chromatin-modifying complex as it encodes a protein with B3 DNA-binding domains and, in vitro, binds DNA in a strong, but non-sequence-specific manner (Levy et al., 2002). A major unanswered question is how do plants perceive cold? Cold perception is required for the vernalization process and the induction of freezing tolerance (cold acclimation). Genetic analysis suggests that cold may be perceived by a common mechanism; however, cold acclimation and vernalization probably use different downstream signal transduction pathways. Intriguingly, cortical microtubules that have an intimate association with the plasma membrane, the major platform for signal perception and transduction, may play a

Signal Transduction Regulating Floral Development

55

role in sensing temperature, and have recently been shown to induce cold acclimation (Abdrakhamanova et al., 2003). Activation of the flowering signalling network by environmental stress The decision to flower is based on the fine-tuned coordination of different environmental signals, in an interactive network that ensures the onset of flowering at the proper time of the year. In the unfavourable conditions imposed by crowding, for example, these coordinated cues can be overruled allowing rapid flowering in response to low red to far-red ratios (Martinez et al., 2004). Other stress factors that can promote flowering include drought, high irradiation, extreme temperature and pathogen infection. These stresses lead to increased levels of several metabolites including ethylene, abscisic acid (ABA), oxylipins and phenylpropanoid derivatives such as salicyclic acid (SA) (Pastori and Foyer, 2002). Recent work demonstrates that UV-C light stress activates the transition to flowering in Arabidopsis through SA. Moreover, SA also regulates the flowering time in non-stressed plants, as SA-deficient plants are late flowering. The regulation of flowering time by SA involves the photoperiod and autonomous pathways, but it does not require the function of the flowering-time gene CO, FCA or FLC (Martinez et al., 2004). It is possible that SA may be a link between stress-activated responses and developmental programmes in Arabidopsis. External signals influence the balance between positive and negative regulators In the absence of a relevant environmental cue to flower, the initiating response is actively repressed at multiple levels including chromatin structure, mRNA stability and protein degradation. The environmental signal to flower will normally initiate several pathways that share signalling elements. Positive and/or negative regulatory loops are also activated. Such complex networks can provide robustness and buffer the system against environmental noise (Hasty et al., 2002). There is increasing evidence that certain floral-inductive signals can exert their influence by antagonizing negative regulators thus allowing the positive regulator to exert its response (Casal et al., 2004). For example, vernalization downregulates the expression of FLC and the maintenance of this repression requires epigenetic regulation mediated by VRN2 (Gendall et al., 2001). The levels of FLC are crucial for the vernalization response and the interaction between the two autonomous genes FCA and FY appears to regulate FLC mRNA 3’-end formation providing a fine-tuning mechanism for the levels of FLC (Simpson et al., 2003). Mutations in EARLY BOLTING IN SHORT DAYS (EBS) and TERMINAL FLOWER 2 (TFL2) accelerate flowering by specifically derepressing FT (Pineiro et al., 2003; Takada and Goto, 2003). EBS and TFL2 contain domains that are normally involved in chromatin remodelling. Thus the response to environmental signals, such as daylength, requires FT expression to be repressed by EBS and TFL2 to allow for upregulation of FT by these signals. Mutant analysis shows that negative regulators not only prevent the response in the absence of the signal but also ensure that in the presence of the signal an

56

R.G. Anthony

exaggerated response does not occur (Casal et al., 2004). Some of the negative regulation is enhanced by the signal itself as part of a feedback loop. Mathematical modelling and experiments with engineered gene circuits indicate that a network with negative feedback is more stable than an unregulated network, and that complex networks involving multiple feedback loops provide robustness, control, exploitation and tolerance of intracellular noise (Hasty et al., 2002).

Endogenous factors affect transition The autonomous pathway The autonomous pathway of Arabidopsis appears to regulate flowering independently of environmental cues and as such is especially important for Arabidopsis ecotypes that are rapid cyclers, i.e. they will complete a life cycle of vegetative and reproductive growth more than once a growing season. The pathway was identified via a group of mutants that are late flowering under all photoperiods and are highly responsive to vernalization (Mouradov et al., 2002). All of the genes associated with the pathway are expressed in the meristem. The autonomous pathway acts in parallel to vernalization to repress FLC expression (Koornneef et al., 1991). In the absence of FRI, this pathway is the major regulator of FLC levels and therefore confers a vernalization requirement (Koornneef et al., 1991). Seven genes that are components of this pathway have now been identified (FCA, FY, FLOWERING LOCUS D (FLD), FVE, FPA, LUMINIDEPENDENS (LD), FLOWERING LOCUS K (FLK)) that, when mutated, produce late-flowering phenotypes, which have higher levels of FLC mRNA (Fig. 3.1) (Boss et al., 2004). Although all members of this pathway act to limit FLC expression, genetic analysis has revealed that they have distinct functions (Mouradov et al., 2002). The notion that the ‘autonomous pathway’ is a typical linear pathway by which a signal leads to a response is now questioned because both genetic and molecular analyses suggest that the autonomous genes function in different subgroups to promote flowering. Several of the genes have been cloned and structural analysis reveals that these genes signal by a process of RNA processing (see below). A thermosensory pathway Most studies on the effects of environmental cues have focused on the importance of photoperiod and vernalization; it is also well known that ambient growth temperature has similarly profound and extremely complex effects on flowering (Blazquez et al., 2003). A most notable example of this is the increase in spring temperatures in temperate zones which has caused an acceleration of the onset of flowering in many plants, despite an unchanged photoperiod (Fitter and Fitter, 2002). Ambient temperature affects the flowering of the Arabidopsis strain Landsberg. At 238C Landsberg flowers after having produced ten leaves but at 168C flowering is delayed until 15 leaves have formed. It appears that ambient temperature is sensed through a genetic pathway requiring FCA and FVE and this is integrated with the environmental signals such as daylength by FT, a strong

Signal Transduction Regulating Floral Development

57

promoter of flowering (Blazquez et al., 2003). Although there are several genetically distinct temperature-response pathways in Arabidopsis, the primary mechanism for detecting temperature changes is not known for any of these pathways. It has been suggested that it might be similar to thermosensing in other organisms, e.g. microbes, where one of the mechanisms involves changes in mRNA conformation (Johansson et al., 2002). This is interesting as FCA encodes an RNA-binding protein.

Biochemical Signalling to the Shoot Apical Meristem Phytohormones influence many diverse developmental processes ranging from seed germination to root, shoot and flower formation. Mutational analysis using Arabidopsis has been instrumental in determining the individual components of specific hormone signal transduction pathways. While no hormone transduction pathway is completely understood, the genes identified to date suggest that simple molecular rules can be established to explain how plant hormone signals are transduced (McCourt, 1999; Hay et al., 2004).

Gibberellins Floral initiation and floral-organ development are both regulated by the phytohormone GA in some but not all species. GA is diffusible and accelerates flowering in wild type particularly in SDs and thus would be a potential candidate for a flowering signal. However, because it is synthesized throughout the plant, it is unlikely to serve as the long-sought after universal and primary longdistance flowering signal (Colasanti and Sundaresan, 2000). Moreover, the fact that GA mutants can still flower in response to photoperiod implies that the daylength pathway must use signals additional to GA. Nevertheless the role for GA as a major signal in the floral process is indisputable. GA control of flowering has been studied in many species, most importantly in Arabidopsis. This is crucial for understanding and connecting the hormone signalling pathways to the well-established floral genetic development routes. Additionally, research in the grasses highlights the importance of the GA signal. Gibberellins promote flowering of Arabidopsis by activating the LFY promoter Arabidopsis mutants defective in GA biosynthesis or signalling have demonstrated that endogenous GAs are involved in the promotion of flowering, although the requirement for GAs in SDs is more critical (Jacobsen and Olszewski, 1993). It has previously been shown that LFY promoter activity is reduced in mutants defective in GA biosynthesis and that the failure of ga1–3 mutants to flower in SDs can be overcome by constitutive expression of LFY. Conversely, constitutive GA signalling in SPINDLY (SPY) mutants causes an increase in LFY promoter activity. GA thus affects flowering through a pathway that controls LFY transcription (Fig. 3.2A) (Blazquez et al., 1998). The

58

R.G. Anthony

Fig. 3.2. (A) gibberellin (GA) signalling pathway that regulates flowering in Arabidopsis. Activation of a hypothetical transmembrane receptor by GA inhibits repressors of GA signalling. These repressors are encoded by the DELLA proteins. The SPINDLY (SPY) gene also represses GA signalling and genetically acts upstream of RGA and GAI. It may act to promote the activity of GAI/RHA/RGL by N-acetylglucosamine (GlcNAc) modification, in which case GA signalling may inhibit GAI/RGA/RGL by repressing SPY function. LFY is upregulated at the transcriptional level by GA. The flowering-time gene SOC1 is also upregulated by GA. GAMYB is proposed to mediate between GAs and the regulation of flowering time. (B) Molecular and GA signalling events following a single long-day in Lolium temulentum. Timing of events of daylength response including GA increase related to shoot apex molecular change. See text for details.

first protein that has been directly implicated in the transcriptional control of downstream target genes is the barley transcription factor GAMYB, whose RNA accumulation is under the control of GAs and which positively regulates

Signal Transduction Regulating Floral Development

59

an a-amylase promoter in the aleurone (Gubler et al., 2002). GAMYB binds specifically to a GA-response element (GARE) in the 5’ regulatory region of GA-activated genes. Binding of GAMYB to the GARE activates the transcription of genes encoding the hydrolytic enzymes that release seed nutrients from the endosperm during seed germination (Gubler et al., 1995). There are three genes in Arabidopsis that resemble GAMYB: AtMYB33, AtMYB65 and AtMYB101 (Gocal et al., 2001b). AtMYB33 has been shown to bind in vitro to a GARE in the LFY promoter suggesting that GAMYB plays a key role in LFY activation (Blazquez and Weigel, 2000; Gocal et al., 2001b). Upstream of GAMYB, GA regulates various plant developmental programmes by suppressing a group of DELLA protein nuclear repressors (GAI, RGA, RGL1, RGL2 and RGL3) (Yu et al., 2004b). These proteins contain a conserved N-terminal DELLA domain, which is possibly involved in the inactivation of these proteins by GA signals (Yu et al., 2004b). GAI and RGA are negative regulators of GA responses in the control of stem elongation, flowering time and root growth. Removing both gene functions causes a synergistic suppression of the corresponding defects in ga1–3 mutants (Fu and Harberd, 2003). In a similar way, RGL2 is a major repressor of seed germination because rgl2 null mutations can significantly promote germination (Lee et al., 2002). GA is able to overcome repression by DELLA by targeting the DELLA protein for destruction in the 26S proteosome (Fu and Harberd, 2003; McGinnis et al., 2003). Because DELLA proteins function in a wide range of plant developmental programmes, their involvement in flower development may be mediated by additional flower-specific regulators. For example, the floral integrator SOC1 plays an important role in GA promotion (Moon et al., 2003). Additional regulators have also been identified such as microRNAs. These are single-stranded molecules of 20–22 nucleotides that can cause complementary-dependent cleavage (or translational inhibition) of target RNA molecules. The microRNA, miR159, directs cleavage of AtMYB33-encoding transcripts, and elevated expression of miR159 results in delayed flowering in SD photoperiods that were associated with a reduction in the levels of LFY transcripts. In addition, miR159 modulates GA-dependent development regulation via its effects on GAMYB activity. The precise role of second messengers in GA signal transduction is largely unknown, although the GA-dependent transduction pathway in barley aleurone cells involves cytosolic Ca2þ and cGMP. Mitogen-activated protein kinase (MAPK) pathways have also been implicated in GA signal transduction and these protein kinase cascades probably contribute to the floral response. Gibberellin and the grasses There have been many studies implicating the importance of GAs in floral induction in both grasses and cereals. Flowering of the grass Lolium temulentum is strictly regulated, occurring rapidly on exposure to a single LD; there is now considerable data linking GA to the photoperiodic induction of leaves, to floral evocation and to floral differentiation (Fig. 3.2B). The GA content increases in the leaf early in the LD and then, hours later, at the shoot apex. In rosette dicot species, LDs increase shoot contents of highly bioactive GAs that

60

R.G. Anthony

results in rapid elongation of stems (bolting) (King and Evans, 2003). In grasses and cereals there is little early stem elongation in LDs, a potentially adaptive response for survival of close grazing by herbivores. It has now been demonstrated that GAs with low effectiveness for stem elongation, GA5 and GA6, that reach the shoot apex are not degraded by 2-oxidase enzymes. By contrast, GA1 and GA4 cause stem elongation and appear to be inactive for floral evocation, and do not reach the vegetative shoot apex because they are susceptible to degradation (King and Evans, 2003). How does the increase in GA relate to the potential molecular events? Exposure of the leaf blade to one LD increases GA5 content fourfold and, 8–16 h later, GA5 and GA6 double in the shoot apex to a concentration known to be inductive for apices in vitro. Following the critical photoperiod, leaf GA20 and GA5 increase within 16 h, 1 day after the start of the LD and apex GA5 and GA6 also begin to rise with a corresponding increase in metabolism and gene expression. At time point 36 h, GA5 and GA6 reach maximal value at the apex, which correlates with a dramatic increase in LtCDC2 indicating rapid cell division (King and Evans, 2003). Then 48 h after the start of the original LD, expression of LtMADS1 and LtMADS2 (APETALA1 (AP1)-like genes) increases (Gocal et al., 2001a). Other genes downstream including LtGAMYB and LtLFY increase at the time of inflorescence formation, which correlates with a rapid rise in GA1 and GA4 at the shoot apex. Thus GAs may regulate the increased expression of LtGAMYB at this time. It is interesting that LtLFY is detected quite late (about 12 days after LD induction) within the spikelet meristems, glumes and lemma primordial. These patterns of expression contrast with Arabidopsis, where LFY and AP1 are consecutively activated early during flower formation (King and Evans, 2003).

The influence of cytokinin signalling on flowering Cytokinins are essential plant hormones that control cell division, shoot meristem initiation, leaf and root differentiation, chloroplast biogenesis, stress tolerance and senescence. Recent rapid advances have discovered hybrid histidine protein kinases (HKs) as cytokinin receptors, histidine phosphotransfer proteins (HPTs) and nuclear response regulators (RRs) as transcription activators and repressors in the cytokinin signal transduction pathway (Sheen, 2002). There are four major steps of the cytokinin phosphorelay: HK sensing and signalling, HPT nuclear translocation, RR transcription activation and a negative feedback loop through cytokinin-inducible RR gene products. Despite our knowledge of the mechanisms and importance of this central signalling pathway, the participation of cytokinins in the control of the floral transition of shoot apical meristems (SAMs) is still a controversial issue. The clearest example is found in Sinapis alba. In S. alba induced to flower by exposure to a single LD, the phloem sap feeding the apex is enriched in cytokinins of the isopentenylade-nine-type (iP-type) between 9 h and 25 h after start of the LD (Leujeune, 1994). One of the earliest events observed at the shoot apex following photoperiodic induction is a transitory increase in the number of

Signal Transduction Regulating Floral Development

61

cells undergoing mitosis. Interestingly, cytokinin causes stimulation of the cell division rate and transcription of the SaMADS A gene, responses that are identical to those normally occurring after LD induction (Bonhomme et al., 2000; Jacqmard et al., 2002). Research on the participation of cytokinins in the control of flowering time in Arabidopsis is relatively scarce. Exogenous applications have been reported to accelerate flowering in various ecotypes (He and Loh, 2002). In the Columbia ecotype, cytokinin stimulation is observed only when light irradiance is low (Dennis et al., 1996). Alternatively, high endogenous levels of cytokinins are associated with early flowering in Columbia plants treated with triacontanol, cerium and lanthanum (He and Loh, 2002). Corbesier et al. (2003) recently analysed cytokinins in leaf extracts, leaf phloem exudate and in the SAM at different times during the floral transition. It was found that, in leaf tissues and leaf exudate, isopentenyladenine forms of cytokinins increased from 16 h after the start of the LD. At 30 h, the SAM of induced plants contained more isopentenyladenine and zeatin than vegetative controls. These cytokinin increases correlate well with the early events of floral transition. Thus these results suggest that endogenous cytokinins might play a role in the control of floral transition in Arabidopsis and act as a component of the floral stimulus of leaf origin (Corbesier et al., 2003).

Other hormones implicated in flowering In addition to cytokinin and GAs, other growth hormones can either inhibit or promote flowering. Mutants that reduce ABA biosynthesis are earlier flowering under non-inductive conditions, suggesting that ABA inhibits flowering (Martinez-Zapater et al., 1994). In support of this, abi1 and abi2 (ABA signalling mutants) have been shown to reduce the flowering time of fca-1 mutants (Chandler et al., 2000; Boss et al., 2004). Ethylene signalling mutants (Ogawara et al., 2003), the brassinosteroid biosynthesis mutant det2 (Chory et al., 1991) and plants altered in SA biosynthesis (Martinez et al., 2004) are late flowering, implicating these plant growth regulators in floral signalling pathways.

The Transmission of Long-distance Flowering Signals Attempts to isolate the transmissible floral regulators Perception of daylength in the leaf suggested that a systemic signal, often called the floral stimulus or florigen, is synthesized in the leaf and transmitted to the SAM where it triggers flower development (An et al., 2004). Early physiological studies showed that mobile signals travelled in the vasculature of the plant, specifically, through the phloem, which also transports nutrients and other molecules from source tissues to sink tissues (Zeevaart, 1976). The many attempts to isolate and characterize the floral stimulus have been largely

62

R.G. Anthony

unsuccessful. The most common approach has been to make extracts from induced leaf tissue and test for their ability to elicit flowering in non-induced plants (Colasanti and Sundaresan, 2000). More sophisticated approaches have recently been undertaken involving mass spectrometry studies of phloem sap (see below). Most of these extractions have focused on small molecules. Recent studies using fluorescent tracers have shown that in Arabidopsis there is actually a decrease in the movement of small molecules from the leaf-to-apex via the symplast at the time of floral induction (Gisel et al., 2002). The lack of tracer movement from the leaf to the shoot apex may indicate either a reduction in overall symplastic transport to the shoot apex or a change in the selectivity of the plasmodesmata during floral induction. There is increasing evidence that macromolecular traffic between cells via plasmodesmata plays essential roles in normal meristem development and function. It is possible that the floral stimulus is a macromolecule, such as RNA or protein that is translocated via the phloem from the leaf to the apical meristem, where it functions as a regulator of gene expression (Zambryski, 2004).

Are CO and florigen connected? CO plays a central role in the photoperiodic perception pathway (Hayama and Coupland, 2003; Yanovsky and Kay, 2003; Valverde et al., 2004). It is expressed in leaves, yet can activate the expression of two genes, LFY and AP1, expressed in reproductive meristems that directly control the initiation of flower development. Therefore CO could regulate the synthesis of a chemical that might fit the definition of florigen. Alternatively, CO could control the levels of several substances, which together regulate the flowering transition (Hayama and Coupland, 2003). Similar arguments can be made for a role of the maize gene INDETERMINATE in regulating the synthesis of transportable substances because of its function in regulating floral initiation and its expression in leaves. CO activates the transcription of FT, which encodes a RAF-kinaseinhibitor-like protein. Recent data suggest that CO acts in the phloem companion cells to trigger floral development at the apex, and controls a systemic signal that crosses graft junctions (An et al., 2004). The mechanism by which CO acts in the phloem involves cell-autonomous activation of its target gene FT and, based on analysis of a green fluorescent protein (GFP), CO fusion protein does not require movement of the CO protein. Thus CO, partly through the activation of FT, regulates the synthesis or transport of a systemic flowering signal, thereby positioning this signal within the established hierarchy of regulatory proteins that controls flowering (An et al., 2004).

Redefining ‘florigen’ clues from microarrays and mass spectrometry Rapid advances in proteomic techniques will make it possible to identify proteins in specific signalling pathways from different species that have similar

Signal Transduction Regulating Floral Development

63

function in the initiation of flowering and allow the analysis of global protein function. For example, it will be possible to determine which genes have been co-opted by plants for the unique purpose of mediating flowering in response to a particular environmental stimulus (Colasanti and Sundaresan, 2000). How floral-inductive signals are transmitted from genes such as CO and FT to downstream effectors such as LFY and AP1 is not well understood; a recent microarray experiment has shed some light on potential downstream targets (Schmid et al., 2003). Schmid et al. (2003) used the parallel analysis of many known floral-regulatory genes, along with the analysis of a large group of newly identified genes that respond to a change in photoperiod. A new subset of genes dependent on CO and FT but not LFY has been identified. Within this group there are several paralogous gene pairs with very similar CO and FT responses, which suggest that many of these genes were not identified in forward genetic screens because of redundancy (Schmid et al., 2003). Within this there exist two groups of potential microRNA (miRNA) targets, a clade of AP2 domain-encoding genes and a large group of SPOROCYTELESS (SPL) genes as being regulated by CO and FT. This observation raises the possibility that miRNAs perform a critical function in mediating the effects of floral induction, which is supported by a recent report on the consequences of miR172 overexpression (Chen, 2004). A recent approach using microbore HPLC-MALDI-TOF-MS overcomes the problems of bioassaying phloem sap. Using this approach HoffmannBenning et al. (2002) compared the contents of the phloem sap from flowering and non-flowering Perilla and lupin plants. The microbore HPLC separations allowed detection of proteins/peptides that were very small and present at very low levels. This approach allowed the identification of more than 100 components in the phloem sap of both Perilla and lupin; of these, four small proteins potentially play a role in the induction of flowering, one of which showed no similarity to any known protein sequences. Furthermore, a second small protein in exudates of induced plants was similar to Ser/Thr receptor-like protein kinases. Two additional protein sequences appeared to be related to two different purine permeases. These may be important for the transport of nucleic acid bases or related signalling molecules, such as cytokinins, into and within the sieve elements. The kinase and purine permeases may act in concert to induce flowering directly or facilitate the transport of the signal. These data clearly demonstrate the power of this approach and similar studies in other species with differing flowering requirements would be of considerable value.

Integration of Pathways Controlling Signalling Current genetic models propose that multiple promotion signalling pathways converge. The different floral-induction pathways are integrated by a small set of genes, including FT, SOC1 and LFY (Blazquez and Weigel, 2000; Lee et al., 2000; Samach et al., 2000). The SOC1 (or AGL20) gene that encodes a MADS-box protein is both activated by CO and repressed by FLC, suggesting

64

R.G. Anthony

that the pathways represented by these genes converge on SOC1 (Michaels and Amasino, 2001). SOC1 is also regulated by GA and therefore is a common target of all three flowering pathways (Borner et al., 2000). In contrast to SOC1, the GA pathway does not regulate expression of other flowering integrators, FLC and FT (Moon et al., 2003). The role of SOC1 is conserved in other plants (Kim et al., 2003). FT is an immediate target of CO, and the expression of FT is negatively regulated by FLC, suggesting that FT integrates the LD and autonomous pathways (Samach et al., 2000). Recent microarray data have indicated that FRUITFUL (FUL) may also act as a floral integrator (Schmid et al., 2003). LFY was one of the first genes to be identified as a floral integrator. LFY is the earliest of the known floral identity genes to be expressed and directly activates one of the later genes AP1 (Wagner et al., 1999). LFY mRNA levels remain low and the floral transition is abolished in co ga1 double mutants in which both the photoperiod pathway and the GA pathway are blocked (Blazquez and Weigel, 2000). Thus LFY expression during the vegetative phase is regulated by both photoperiod and GA signalling pathways. Deletion analysis demonstrated that the photoperiod pathway and the GA pathway act on different cis elements in the LFY promoter (Blazquez and Weigel, 2000). Thus signalling appears to converge on several parallel transcription factors rather than just a single component.

Mechanistic Aspects of Floral Signal Transduction We are rapidly gaining insight into the genetic make-up that comprises the major signalling routes employed by plants for flowering, the continued use of forward and reverse genetics and global expression strategies will add new elements to the skeletal networks already established. Detailed analysis of specific gene to gene interactions is now starting to reveal some of the mechanisms and strategies that are employed to convey the signal message within the cell and from cell to cell.

Control of gene expression: chromatin structure Genetic analysis of flowering has identified a large source of transcription factors including MADS-box transcription factors (e.g. AGL24, SOC1, FLC), MYB transcription factors (e.g. CCA1) and zinc-finger transcription factors (e.g. CO). Recent research is showing how the signal transduction mechanism can impinge directly on to the control of gene expression. Methylation of DNA, especially 5-methylcytosine, is a specific structural modification that regulates gene expression. Methylation generally results in inactivation and demethylation in activation. There is clear evidence that chromatin remodelling can affect flowering by changing the methylation patterns. The histones are the most abundant protein in chromatin and primarily responsible for the

Signal Transduction Regulating Floral Development

65

folding and packaging of DNA. The N-terminal tails of histones H3 and H4 undergo extensive post-translational modifications including acetylation and methylation, which are two intensively studied alterations with well-defined roles in the control of gene-expression states (Henderson and Dean, 2004). Specific residues of histone H3 tails are modified by acetylation and methylation and changes in these modifications serve as part of a histone code specifying active or repressed gene activity states (Fischle, 2003). Vernalization increases histone H3 deacetylation in the 5’ region of FLC and increased methylation of histone H3 lysine residues 9 and 27 (histone methylation marks) after cold exposure, all modifications associated with gene repression (Sung and Amasino, 2004b). The histone methylation marks can act as signals to recruit further mediators of gene expression (Orlando, 2003). Further evidence for the control of transcription-factor expression by chromatin remodelling has come to light from the autonomous pathway gene FLD. FLD encodes a protein with homology to a human protein that functions in the histone deacetylase 1,2 (HDAC1/2) corepressor complex (He et al., 2003). The FLD protein possesses a SWIRM domain, which is associated with chromatin-remodelling enzymes. In fld mutants, the 5’-end of FLC displays hyperacteylation of histone H3, indicating that FLD is required to deacetylate FLC chromatin and thereby repress its expression (He et al., 2003).

RNA-binding and processing Several genes implicated in the autonomous pathway interact to repress the activity of FLC (Fig. 3.1). There is strong evidence to suggest that the primary mechanism of repression is at the RNA level. The expression of FLC is complex and exhibits an autoregulatory mechanism involving the choice of polyadenylation site. FCA and FY interact by virtue of a WW protein-interaction domain at the C-terminus. In addition, FCA possesses two RNA recognition-motif domains (Sudol and Hunter, 2000). In contrast to FCA, FY is highly conserved throughout eukaryotes. The yeast homologue Pfs2p acts as a scaffold protein in the CPF (cleavage and polyadenylation factor) complex, which is required for 3’ cleavage and polyadenylation of pre-mRNA transcripts (Ohnacker et al., 2000). It is therefore possible that FY may perform a similar function, in Arabidopsis, in RNA processing while also functioning in regulated polyadenylation through interaction with FCA. FPA, another autonomous pathways gene, also possesses a RNA recognition-motif domain (Schomburg et al., 2001) and FLK encodes a nuclear KH-type RNA-binding protein (Lim et al., 2004). The mechanism of RNA processing is best understood for FCA. There are four FCA transcripts, and intron 3 is a major site of alternative processing. Premature cleavage and polyadenylation within this intron generates the truncated, non-functional FCA transcript. FCA is able to negatively autoregulate its own expression by promoting intron 3 polyadenylation (Quesada et al., 2003). There is recent evidence to suggest that non-coding RNA may have a role in chromatin regulation. It is possible to speculate that PHOTOPERIOD INSENSITIVE EARLY FLOWERING (PIE1) and VERNALIZATION INDEPENDENCE (VIP),

66

R.G. Anthony

positive upregulators of FLC, may act to promote active chromatin, whereas FVE and FLD act to deacetylate histones, thus promoting a silent chromatin state (Henderson and Dean, 2004). Thus multiple RNA-binding proteins are required for repression of FLC expression by the autonomous pathway, whether this reflects a cascade of post-transcriptional regulators or a complex of RNA-binding factors remains uncertain.

A role for receptor-like kinases The mechanisms used by plants to transmit extracellular signals into the cytoplasm require receptors located in the plasma membrane. Receptor-like kinases (RLKs) are an important group of protein kinases with direct functions in transmission of signals across the plasma membrane. They undergo autophosphorylation on the intracellular kinase domain in a reaction thought to result from homologous dimerization of the receptors in the plasma membrane when the ligand binds (Takasaki et al., 2000; Becraft, 2002). It is now apparent that plants contain a much larger complement of receptor-kinase genes than other organisms. Current counts reveal that there are over 400 genes predicted to encode RLKs in the genome and that LRR (leucine-rich repeat)type RLKs are the largest class, containing over 200 genes (Shiu and Bleecker, 2001). The possibility exists that plants use hundreds of receptors as entry points into signalling pathways and this has broad implications for developmental and environmental responses. Fewer than 2% of the total RLKs identified have known functions and much less is known about their signalling components or ligands. RLKs participate in a diverse range of processes, including self versus non-self recognition, disease resistance, hormone perception and regulation of floral development (Shiu and Bleecker, 2001). Maintenance of shoot and floral meristem cell proliferation and fate Plants have adopted unique developmental mechanisms that allow them to make the commitment to flowering in response to external signals. The key structure that provides this flexibility is the SAM, which gives rise to the aerial structures of the plant such as leaves, stems and ultimately, flowers (Baurle and Laux, 2003; Veit, 2004). It forms organs continuously by carefully balancing two activities. The first is the maintenance of a constant sized pool of undifferentiated stem cells; the second is the commitment of appropriately positioned progeny cells towards differentiation, so that they are competent to form specialized lateral organs such as leaves. Although some components of signalling within the shoot meristem have been uncovered by genetic analysis, intracellular mechanisms connecting growth and development signals to the activation of specific genes are largely unknown. The differentiation of stem cells is regulated by the CLAVATA genes (CLV1, CLV2, CLV3), which code for components of a signal transduction pathway (Fig. 3.3). Mutation in any of the CLAVATA loci leads to a progressive overgrowth of the SAM and floral meristems. Genetic analysis revealed

Signal Transduction Regulating Floral Development

67

CLV3 ligand

CLV2 CLV1

KAPP ROP

MAPKs? POL ? WUS

AG LFY

Meristem proliferation

Floral meristem AG Gynoecium differentiation

Fig. 3.3. Signal transduction pathways of the receptor-like kinase (RLK) Clavata (CLV). The model predicts that a CLV1-CLV2 heterodimer binds to a CLV3 multimer. This ligand– receptor interaction drives the transphosphorylation of the CLV1-kinase domain. Phosphorylated residues on the kinase domain act as direct or indirect binding sites for downstream effector molecules such as kinase-associated protein phosphatase (KAPP) and a Rho GTPase-related protein (ROP). How signalling is transmitted from the plasma membrane is unclear, although it could involve a mitogen-activated protein kinase (MAPK) cascade. This signalling inactivates WUS expression and is negatively regulated by POL.

that these genes function in the same pathway (Clark et al., 1995; Kayes and Clark, 1998). CLV1 encodes a LRR transmembrane receptor with an intracellular serine/threonine-kinase domain (Clark et al., 1997). CLV2 codes for a similar protein lacking the kinase domain. The model predicts that a CLV1CLV2 heterodimer binds to a small peptide CLV3 (Fletcher et al., 1999; Trotochaud et al., 1999). The CLV3 ligand functions in a non-cell autonomous manner; it is transported through the secretory pathway to the apoplast and this export to the extracellular space is essential for its function in activating the CLV1/CLV2 receptor complex (Rojo et al., 2002). Thus CLV3 relays a signal

68

R.G. Anthony

from the summit of the meristem to the underlying tissues and activates the CLV1 receptor to keep cells mitotically active. This ligand–receptor interaction drives the transphosphorylation of the CLV1-kinase domain. Phosphorylated residues on the kinase domain act as direct or indirect binding sites for downstream effector molecules such as kinase-associated protein phosphatase (KAPP) and a Rho GTPase-related protein (ROP), indicating that CLV1 may relay signal transduction through this small GTPase (Trotochaud et al., 1999). How signalling is transmitted from the plasma membrane is unclear, although it could involve a MAPK cascade. This signalling inactivates the WUSCHEL (WUS) gene and is negatively regulated by POL. WUS is required to initiate and maintain stem cells in the central zone of the meristem. WUS encodes a novel homeodomain-like transcription factor expressed in few cells beneath the CLV1 expression zone (Mayer et al., 1998). WUS and CLV3 establish a feedback regulatory loop, such that WUS signal promotes CLV3 expression and specifies stem cell maintenance in the apex of the meristem through the CLV pathway (via noncell autonomy of CLV3), which in turn restricts the strength and expression of WUS to the underlying tissues (Gross-Hardt et al., 2002). This ensures a tight control of stem-cell number through carefully balanced WUS and CLV3 activities of promotion and restriction of stem-cell identity, respectively. This regulatory loop operates through the communication between the apical stem cells and the underlying organizing centre. Regulation of floral meristem cell fate Developmental signals that cause the vegetative to floral transition originate outside the SAM and so the SAM remains uncommitted to flowering prior to its perception of external signals. Floral induction then causes a cascade of processes within the SAM that result in its restructuring, accompanied by changes in the rate and pattern of cell division, and the formation of floral structures instead of leaves (Carles and Fletcher, 2003). At the structural and organizational level, both the SAM and the floral meristem are similar, because both contain a stem-cell reservoir at the apex that contributes cells to organogenesis on the flanks. Floral meristems develop from the flanks of the SAM similar to leaf primordial, but with different fates. The central pool of stem cells is maintained by the CLV pathway. Signalling by CLV3 through the CLV1 receptor complex limits the size of the WUS-expression domain in the interior of the meristem. In turn WUS activity, perhaps with the help of an unknown diffusible signal (X), preserves the population of CLV3-expressing stem cells in the superficial cell layers (Fig. 3.3). TFL simultaneously represses the expression of the floral meristem-identity gene LFY in the SAM. LFY is then upregulated in early floral meristem and along with WUS; it activates AGAMOUS (AG) in the centre of the floral meristem where the stamens and carpels will form. LFY and WUS may require another factor (Y) to activate AG in the correct expression domain. Later in floral development, when the carpel primordials are due to form, AG and another unidentified factor (Z) repress the expression of WUS, terminating stem cell maintenance and allowing gynoecium differentiation.

Signal Transduction Regulating Floral Development

69

RLKs can transduce signals directly to transcription factors AGL24 is a gene with similar properties to SOC1; data suggest that it is a flowering-time target of vernalization. Like SOC1, it is upregulated by vernalization but unlike SOC1, this upregulation is independent of FLC activity. Mutants of agl24 are late flowering, but this is not strongly suppressed by vernalization (Yu et al., 2004b). AGL24 appears to act downstream of SOC1 but upstream of LFY, but overexpression of AGL24 increases SOC1 expression indicating that the relationship between these genes may not be straightforward (Michaels et al., 2003a,b). Indeed, both AP1 and LFY repress AGL24 activity (Yu et al., 2004a). Two-hybrid analysis indicates that AGL24 is a substrate of meristematic receptor-like kinase (MRLK). MRLK encodes an RLK with LRRs in its extracellular domain. The precise function of MRLK is still unclear. The AGL24 protein specifically interacted with, and was phosphorylated by, the MRLK kinase domain in in vitro assays. The simultaneous expression of AGL24 and MRLK in shoot apices during floral transition suggests that the interaction occurs in plants. Furthermore, use of GFP markers indicates that MRLK signalling promotes translocation of AGL24 from the cytoplasm to the nucleus. It is possible that AGL24 may form complexes with other MADSbox transcription factors, which play a role in floral transition and that the formation and/or subcellular localization of this complex may in part be regulated by MRLK (Fujita et al., 2003).

The two-component Ehd1 signalling cascade in the short-day promotion pathway The sensory and transduction systems evolved by bacteria enable them to survive and adapt to various different environmental conditions. Twocomponent systems have also emerged as important sensing/response mechanisms in higher plants (Grefen and Harter, 2004). The first component of a two-component system is usually a receptor protein. The receptor contains a periplasmic domain that binds ligands, includes a variable number of transmembrane domains and has a C-terminal extension. When activated by binding a ligand, a kinase activity located in the C-terminus autophosphorylates the receptor, transferring orthophosphate from ATP to a histidine residue. The active receptor is a homodimer in which each monomer phosphorylates the other. The second component, a response-regulator protein is activated when the receptor kinase transfers phosphate from its histidine residue to a conserved aspartate residue on the regulator. Removal of the phosphate group inactivates the response regulator. Two-component systems play a major role in cytokinin perception and signalling and contribute to ethylene signal transduction, osmosensing and function as components of the Arabidopsis circadian clock (Urao et al., 2001). Furthermore, developmental processes like megametogenesis in A. thaliana and flowering promotion in rice (Oryza sativa) involve elements of two-component systems (Grefen and Harter, 2004). Because of the molecular mode of signalling, plant two-component systems appear to serve as intensive crosstalk and signal-integration machinery. In a recent study, Doi et al. (2004) showed that a B-type response

70

R.G. Anthony

regulator, the gene for which has been named Early heading date 1 (Ehd1), promotes flowering under SD conditions in rice (Oryza sativa). Interestingly, although Ehd1 has a classical receiver domain and a functional GARP (Golden2, Arabidopsis RESPONSE REGULATOR (ARR)) DNA-binding motif, it very likely has no orthologue in Arabidopsis. Ehd1 transcript accumulates under SD conditions but, in contrast to ARRs, Ehd1 appears not to be an intrinsic circadian clock component. Ehd1 very likely promotes flowering by direct regulation of certain MADS-box genes and of Flowering Locus T group-like inducer genes, which are implicated in the regulation of photoperiodic flowering. These findings indicate that a novel two-component signalling cascade, which is different from the systems of Arabidopsis, is integrated into the conserved pathway of photoperiodic control of flowering in rice (Doi et al., 2004).

Cell–cell communication: the trafficking of transcription factors It is clear that floral development and indeed development of multicellular organisms in general depends on the communication of reliable information among neighbouring cells. In plants cell fate is generally determined by position rather than cell lineage, and both cellular differentiation and organ development depend on intercellular signalling. It is widely accepted that the control of gene expression during development by intercellular signalling is normally specified by the release of small signal molecules by the signalling cell, followed by perception by the target cell, transduction to the nucleus, followed by modulation of transcriptional control of target genes. An alternative hypothesis proposes that transcription factors themselves can act as signal molecules moving between cells to activate gene expression at a distance from the site where they were originally produced. This hypothesis was proposed based on the observation that plant viruses could mediate intercellular movement of viral proteins by physical interaction with plasmodesmata. The communication of information between the cell layers of the meristem and its influence on meristem function can be studied by exploiting periclinal chimeras in which a developmental gene has been inactivated in one or more meristems. (Vincent et al., 2003). Two transcription factors KNOTTED1 (KN1) in maize and DEFICIENS (DEF) in Antirrhinum have been shown to move to cells in which their RNAs are not found (Perbal et al., 1996). The trafficking of transcription factors is now thought to be a widespread phenomenon in plants. To address the question of functional activity of a transcription factor after intercellular movement, Sessions et al. (2000) investigated the cellular autonomy of action of LFY and AP1. Both LFY and AP1 encode transcription factors that regulate overlapping sets of target genes. Sessions et al. used two approaches to study cell autonomy: (i) heat-inducible, site-specific recombination to produce AP1þ or LFYþ clonal sectors, marked by corresponding loss of a visible marker gene (b-glucuronidase (GUS)); and (ii) expression of AP1 or LFY under the control of an L1 cell layer-specific promoter in ap1 or lfy mutant plants. LFY and AP1 are able to directly activate the expression of AP3 and AG. Analysis of the RNA expression patterns of these target genes by

Signal Transduction Regulating Floral Development

71

in situ hybridization indicates that AP1 activates its targets only in cells expressing AP1, whereas LFY activates the same target genes throughout all cell layers of the developing flower, even when LFY expression is confined to a single cell layer. Although LFY RNA is found only in cells that are genetically capable of expressing LFY, LFY protein is detectable in all cell layers of the developing flower. Most importantly LFY was shown to act in lfy mutant cells (possessing LFY protein, but not LFY RNA) on a reporter transgene, whose expression requires that LFY bind directly on the target gene’s promoter. This result proves that LFY protein is active as a DNA-binding transcription factor after moving between cells. Also LFY acts not only between adjacent cells, but also at long range because it can reorganize an entire meristem even if expressed in only half of the meristem. Wu et al. (2003) recently characterized the mode of LFY movement in Arabidopsis SAMs and floral primordial. Using functional LFY–GFP fusion proteins, they showed that LFY moves more readily from the L1 into deeper cell layers than laterally into adjacent, clonally related cells (Fig. 3.4). By

Fig. 3.4. Confocal images of green fluorescence protein (GFP) fluorescence in Arabidopsis inflorescence apices (A, C, D) and leaf epidermis (B). AP1:GFP does not move from the L1 meristematic layer (A). C-terminal and N-terminal GFP fusion of LFY move several cell layers into the underlying tissues from the L1 in the apex. The LFY–GFP fusions are localized to both the nucleus and the cytoplasm in leaf epidermal cells and localize with plasmodesmata pit fields (arrows). Movement occurs by diffusion rather than active translocation. Photo copyright Xuelin Wu and Detlef Weigel, Salk Institute.

72

R.G. Anthony

contrast, a functional AP1–GFP fusion is unable to move from its source cells. Comparison of the dynamics of LFY–GFP fusion proteins with other GFP fusion suggests that this movement is driven by diffusion. Deletion experiments failed to identify a specific movement signal in LFY, which is compatible with the conclusion that LFY movement is non-targeted. The hypothesis of non-targeted movement is also supported by the finding of a correlation between cytoplasmic localization and the ability of these proteins to move to adjacent cells (Wu et al., 2003). Furthermore, Wu et al. propose that movement is a default mechanism for many proteins in the Arabidopsis shoot apex unless they are either efficiently targeted to specific subcellular locations or retained through formation of protein complexes, although more case studies are needed to determine the universal application of this hypothesis.

Conclusions and Future Perspectives Genetic analysis in Arabidopsis has enabled the isolation of genes that control flowering time, and the identification of interacting pathways that promote flowering in response to different environmental conditions. However, our present understanding of these pathways represents only a minimal framework and future work will concentrate on understanding the biochemical function of pathway components and the manner in which the signalling pathways convey information that ultimately regulates flowering time. Bacterial receptor and transduction systems provide models for plant receptors including proteins that sense phytochrome for example. Investigations into the roles of GTPases in plant signal transduction are still in their infancy, but already a strong relationship is implicated between GTPase activity and phospholipid signalling. Phosholipases A, C and D influence many aspects of plant development and signalling and are likely to play a role in floral signal transduction. Cyclic nucleotides act as second messengers in plant cells and are implicated in phytochrome and GA signalling; they most likely interact with Ca2þ signalling network in plants, although how the second messenger activity interacts with the multitude of floral transcription factors is unknown. Post-translational modifications, such as phosphorylation, have been implicated in the regulation of floral-specific proteins and are likely to involve MAPK signalling cascades, although this has yet to be demonstrated. It is clear that the complexity of the data regarding the network of flowering pathways will increase dramatically with the continued development of sophisticated imaging techniques for the visualization of signalling pathways and the increasing use of proteomic research to characterize signalling modules.

Acknowledgements I would like to thank Dr Xuelin Wu and Professor Detlef Weigel for providing Fig. 3.4, and colleagues at Royal Holloway for their helpful comments.

Signal Transduction Regulating Floral Development

73

References Abdrakhamanova, A., Wang, Q.Y., Khokhlova, L. and Nick, P. (2003) Is microtubule disassembly a trigger for cold acclimation? Plant & Cell Physiology 44, 676–686. Amasino, R.M. (2003) Flowering time: a pathway that begins at the 3’ end. Current Biology 13, R670–672. An, H., Roussot, C., Suarez-Lopez, P., Corbesier, L., Vincent, C., Pineiro, M., Hepworth, S., Mouradov, A., Justin, S., Turnbull, C. and Coupland, G. (2004) CONSTANS acts in the phloem to regulate a systemic signal that induces photoperiodic flowering of Arabidopsis. Development 131, 3615– 3626. Baurle, I. and Laux, T. (2003) Apical meristems: the plant’s fountain of youth. Bioessays 25, 961–970. Becraft, P.W. (2002) Receptor kinase signalling in plant development. Annual Review of Cell Developmental Biology 18, 163–192. Blazquez, M.A. and Weigel, D. (2000) Integration of floral inductive signals in Arabidopsis. Nature 404, 889–892. Blazquez, M.A., Green, R., Nilsson, O., Sussman, M.R. and Weigel, D. (1998) Gibberellins promote flowering of Arabidopsis by activating the LEAFY promoter. Plant Cell 10, 791–800. Blazquez, M.A., Ahn, J.H. and Weigel, D. (2003) A thermosensory pathway controlling flowering time in Arabidopsis thaliana. Nature Genetics 33, 168– 171. Bonhomme, F., Kurz, B., Melzer, S., Bernier, G. and Jacqmard, A. (2000) Cytokinin and gibberellin activate SaMADS A, a gene apparently involved in regulation of the floral transition in Sinapis alba. Plant Journal 24, 103–111. Borner, R., Kampmann, G., Chandler, J., Gleissner, R., Wisman, E., Apel, K. and Melzer, S. (2000) A MADS domain gene involved in the transition to

flowering in Arabidopsis. Plant Journal 24, 591–599. Boss, P.K., Bastow, R.M., Mylne, J.S. and Dean, C. (2004) Multiple pathways in the decision to flower: enabling, promoting, and resetting. Plant Cell 16 (Suppl.), S18–31. Carles, C.C. and Fletcher, J.C. (2003) Shoot apical meristem maintenance: the art of a dynamic balance. Trends in Plant Science 8, 394–401. Casal, J.J., Fankhauser, C., Coupland, G. and Blazquez, M.A. (2004) Signalling for developmental plasticity. Trends in Plant Science 9, 309–314. Chandler, J., Martinez-Zapater, J.M. and Dean, C. (2000) Mutations causing defects in the biosynthesis and response to gibberellins, abscisic acid and phytochrome B do not inhibit vernalization in Arabidopsis fca-1. Planta 210, 677–682. Chen, X. (2004) A microRNA as a translational repressor of APETALA2 in Arabidopsis flower development. Science 303, 2022–2025. Chory, J., Nagpal, P. and Peto, C.A. (1991) Phenotypic and genetic analysis of det2, a new mutant that affects light regulated seedling development in Arabidopsis. Plant Cell 3, 445–459. Clark, S.E., Running, M.P. and Meyerowitz, E.M. (1995) CLAVATA3 is a specific regulator of shoot and floral meristem development affecting the same processes as CLAVATA1. Development 121, 2057–2067. Clark, S.E., Williams, R.W. and Meyerowitz, E.M. (1997) The CLAVATA1 gene encodes a putative receptor kinase that controls shoot and floral meristem size in Arabidopsis. Cell 89, 575–585. Colasanti, J. and Sundaresan, V. (2000) ‘Florigen’ enters the molecular age: long-distance signals that cause plants to flower. Trends in Biochemical Sciences 25, 236–240.

74

R.G. Anthony Corbesier, L., Prinsen, E., Jacqmard, A., Lejeune, P., Van Onckelen, H., Perilleux, C. and Bernier, G. (2003) Cytokinin levels in leaves, leaf exudate and shoot apical meristem of Arabidopsis thaliana during floral transition. Journal of Experimental Botany 54, 2511–2517. Dennis, E., Finnegan, E.J., Bilodeau, P., Chaudhury, A., Genger, R., Helliwell, C., Sheldon, C.C., Bagnall, D.J. and Peacock, W.J. (1996) Vernalization and the initiation of flowering. Seminars in Cell and Developmental Biology 7, 441–448. Doi, K., Izawa, T., Fuse, T., Yamanouchi, U., Kubo, T., Shimatani, Z., Yano, M. and Yoshimura, A. (2004) Ehd1, a B-type response regulator in rice, confers short-day promotion of flowering and controls FT-like gene expression independently of Hd1. Genes & Development 18, 926–936. Fischle, W., Wang, Y. and Allis, C.D. (2003) Binary switches and modification cassettes in histone biology and beyond. Nature 425, 475–479. Fitter, A.H. and Fitter, R.S. (2002) Rapid changes in flowering time in British plants. Science 296, 1689– 1691. Fletcher, J.C., Brand, U., Running, M.P., Simon, R. and Meyerowitz, E.M. (1999) Signalling of cell fate decisions by CLAVATA3 in Arabidopsis shoot meristems [see comments]. Science 283, 1911–1914. Fu, X. and Harberd, N.P. (2003) Auxin promotes Arabidopsis root growth by modulating gibberellin response. Nature 421, 740–743. Fujita, H., Takemura, M., Tani, E., Nemoto, K., Yokota, A. and Kohchi, T. (2003) An Arabidopsis MADS-box protein, AGL24, is specifically bound to and phosphorylated by meristematic receptor-like kinase (MRLK). Plant Cell Physiology 44, 735–742. Gendall, A.R., Levy, Y.Y., Wilson, A. and Dean, C. (2001) The VERNALIZATION 2 gene mediates the epigenetic

regulation of vernalization in Arabidopsis. Cell 107, 525–535. Gisel, A., Hempel, F.D., Barella, S. and Zambryski, P. (2002) Leaf-to-shoot apex movement of symplastic tracer is restricted coincident with flowering in Arabidopsis. Proceedings of the National Academy of Sciences USA 99, 1713–1717. Gocal, G.F., King, R.W., Blundell, C.A., Schwartz, O.M., Andersen, C.H. and Weigel, D. (2001a). Evolution of floral meristem identity genes: analysis of Lolium temulentum genes related to APETALA1 and LEAFY of Arabidopsis. Plant Physiology 125, 1788– 1801. Gocal, G.F., Sheldon, C.C., Gubler, F., Moritz, T., Bagnall, D.J., MacMillan, C.P., Li, S.F., Parish, R.W., Dennis, E.S., Weigel, D. and King, R.W. (2001b). GAMYB-like genes, flowering, and gibberellin signalling in Arabidopsis. Plant Physiology 127, 1682– 1693. Grefen, C. and Harter, K. (2004) Plant two-component systems: principles, functions, complexity and cross talk. Planta 219, 733–742. Gross-Hardt, R., Lenhard, M. and Laux, T. (2002) WUSCHEL signalling functions in interregional communication during Arabidopsis ovule development. Genes & Development 16, 1129–1138. Gubler, F., Kalla, R., Roberts, J.K. and Jacobsen, J.V. (1995) Gibberellinregulated expression of a myb gene in barley aleurone cells: evidence for Myb transactivation of a high-pI alphaamylase gene promoter. Plant Cell 7, 1879–1891. Gubler, F., Chandler, P.M., White, R.G., Llewellyn, D.J. and Jacobsen, J.V. (2002) Gibberellin signalling in barley aleurone cells. Control of SLN1 and GAMYB expression. Plant Physiology 129, 191–200. Gyula, P., Schafer, E. and Nagy, F. (2003) Light perception and signalling in higher plants. Current Opinion in Plant Biology 6, 446–452.

Signal Transduction Regulating Floral Development

75

Hasty, J., McMillen, D. and Collins, J.J. (2002) Engineered gene circuits. Nature 420, 224–230. Hay, A., Craft, J. and Tsiantis, M. (2004) Plant hormones and homeoboxes: bridging the gap? Bioessays 26, 395–404. Hayama, R. and Coupland, G. (2003) Shedding light on the circadian clock and the photoperiodic control of flowering. Current Opinion in Plant Biology 6, 13–19. He, Y.W. and Loh, C.S. (2002) Induction of early bolting in Arabidopsis thaliana by triacontanol, cerium and lanthanum is correlated with increased endogenous concentration of isopentenyl adenosine (iPAdos). Journal of Experimental Botany 53, 505–512. He, Y., Michaels, S.D. and Amasino, R.M. (2003) Regulation of flowering time by histone acetylation in Arabidopsis. Science 302, 1751–1754. Henderson, I.R. and Dean, C. (2004) Control of Arabidopsis flowering: the chill before the bloom. Development 131, 3829–3838. Hoffmann-Benning, S., Gage, D.A., McIntosh, L., Kende, H. and Zeevaart, J.A. (2002) Comparison of peptides in the phloem sap of flowering and nonflowering Perilla and lupine plants using microbore HPLC followed by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Planta 216, 140–147. Izawa, T., Oikawa, T., Sugiyama, N., Tanisaka, T., Yano, M. and Shimamoto, K. (2002) Phytochrome mediates the external light signal to repress FT orthologs in photoperiodic flowering of rice. Genes & Development 16, 2006–2020. Jacobsen, S.E. and Olszewski, N.E. (1993) Mutations at the SPINDLY locus of Arabidopsis alter gibberellin signal transduction. Plant Cell 5, 887–896. Jacqmard, A., Detry, N., Dewitte, W., Van Onckelen, H. and Bernier, G. (2002) In situ localisation of cytokinins

in the shoot apical meristem of Sinapis alba at floral transition. Planta 214, 970–973. Johansson, J., Mandin, P., Renzoni, A., Chiaruttini, C., Springer, M. and Cossart, P. (2002) An RNA thermosensor controls expression of virulence genes in Listeria monocytogenes. Cell 110, 551–561. Kayes, J.M. and Clark, S.E. (1998) CLAVATA2, a regulator of meristem and organ development in Arabidopsis. Development 125, 3843–3851. Kim, K.W., Shin, J.H., Moon, J., Kim, M., Lee, J., Park, M.C. and Lee, I. (2003) The function of the flowering time gene AGL20 is conserved in crucifers. Molecules and Cells 16, 136–141. King, R.W. and Evans, L.T. (2003) Gibberellins and flowering of grasses and cereals: prizing open the lid of the ‘Florigen’ black box. Annual Review of Plant Physiology and Plant Molecular Biology 54, 307–328. Koornneef, M., Hanhart, C.J. and van der Veen, J.H. (1991) A genetic and physiological analysis of late flowering mutants in Arabidopsis thaliana. Molecular and General Genetics 229, 57–66. Koornneef, M., Alonso-Blanco, C., Peeters, A.J. and Soppe, W. (1998) Genetic control of flowering time in Arabidopsis. Annual Review of Plant Physiology and Plant Molecular Biology 49, 345–370. Lee, H., Suh, S.S., Park, E., Cho, E., Ahn, J.H., Kim, S.G., Lee, J.S., Kwon, Y.M. and Lee, I. (2000) The AGAMOUS-LIKE 20 MADS domain protein integrates floral inductive pathways in Arabidopsis. Genes & Development 14, 2366–2376. Lee, S., Cheng, H., King, K.E., Wang, W., He, Y., Hussain, A., Lo, J., Harberd, N.P. and Peng, J. (2002) Gibberellin regulates Arabidopsis seed germination via RGL2, a GAI/RGAlike gene whose expression is up-regulated following imbibition. Genes & Development 16, 646–658.

76

R.G. Anthony Leujeune, P., Bernier, G. Requier M-C., Kinet, J-M. (1994) Cytokinins in phloem and xylem saps of Sinapis alba during floral induction. Physiologia Plantarum 90, 522–528. Levy, Y.Y., Mesnage, S., Mylne, J.S., Gendall, A.R. and Dean, C. (2002) Multiple roles of Arabidopsis VRN1 in vernalization and flowering time control. Science 297, 243–246. Lim, M.H., Kim, J., Kim, Y.S., Chung, K.S., Seo, Y.H., Lee, I., Hong, C.B., Kim, H.J. and Park, C.M. (2004) A new Arabidopsis gene, FLK, encodes an RNA binding protein with K homology motifs and regulates flowering time via FLOWERING LOCUS C. Plant Cell 16, 731–740. Martinez, C., Pons, E., Prats, G. and Leon, J. (2004) Salicylic acid regulates flowering time and links defence responses and reproductive development. Plant Journal 37, 209–217. Martinez-Zapater, J.M., Coupland, G., Dean, C. and Koornneef, M. (1994) The Transition to Flowering in Arabidopsis. Cold Spring Harbor Laboratory Press , Cold Spring Harbor, New York. Mayer, K.F., Schoof, H., Haecker, A., Lenhard, M., Jurgens, G. and Laux, T. (1998) Role of WUSCHEL in regulating stem cell fate in the Arabidopsis shoot meristem. Cell 95, 805–815. McCourt, P. (1999) Genetic analysis of hormone signalling. Annual Review of Plant Physiology and Plant Molecular Biology 50, 219–243. McGinnis, K.M., Thomas, S.G., Soule, J.D., Strader, L.C., Zale, J.M., Sun, T.P. and Steber, C.M. (2003) The Arabidopsis SLEEPY1 gene encodes a putative F-box subunit of an SCF E3 ubiquitin ligase. Plant Cell 15, 1120– 1130. Michaels, S.D. and Amasino, R.M. (1999) FLOWERING LOCUS C encodes a novel MADS domain protein that acts as a repressor of flowering. Plant Cell 11, 949–956. Michaels, S.D. and Amasino, R.M. (2001) Loss of FLOWERING LOCUS C activ-

ity eliminates the late-flowering phenotype of FRIGIDA and autonomous pathway mutations but not responsiveness to vernalization. Plant Cell 13, 935–941. Michaels, S.D., He, Y., Scortecci, K.C. and Amasino, R.M. (2003a) Attenuation of FLOWERING LOCUS C activity as a mechanism for the evolution of summer-annual flowering behavior in Arabidopsis. Proceedings of the National Academy of Sciences USA 100, 10102–10107. Michaels, S.D., Ditta, G., GustafsonBrown, C., Pelaz, S., Yanofsky, M. and Amasino, R.M. (2003b) AGL24 acts as a promoter of flowering in Arabidopsis and is positively regulated by vernalization. Plant Journal 33, 867– 874. Moon, J., Suh, S.S., Lee, H., Choi, K.R., Hong, C.B., Paek, N.C., Kim, S.G. and Lee, I. (2003) The SOC1 MADSbox gene integrates vernalization and gibberellin signals for flowering in Arabidopsis. Plant Journal 35, 613– 623. Mouradov, A., Cremer, F. and Coupland, G. (2002) Control of flowering time: interacting pathways as a basis for diversity. Plant Cell 14 (Suppl.), S111– S130. Nagy, F. and Schafer, E. (2002) Phytochromes control photomorphogenesis by differentially regulated, interacting signalling pathways in higher plants. Annual Review of Plant Biology 53, 329–355. Nakagawa, M. and Komeda, Y. (2004) Flowering of Arabidopsis cop1 mutants in darkness. Plant Cell Physiology 45, 398–406. Neuhaus, G., Bowler, C., Hiratsuka, K., Yamagata, H. and Chua, N.H. (1997) Phytochrome-regulated repression of gene expression requires calcium and cGMP. EMBO Journal 16, 2554– 2564. Ogawara, T., Higashi, K., Kamada, H. and Ezura, H. (2003) Ethylene advances the transition from vegetative

Signal Transduction Regulating Floral Development

77

growth to flowering in Arabidopsis thaliana. Journal of Plant Physiology 160, 1335–1340. Ohnacker, M., Barabino, S.M., Preker, P.J. and Keller, W. (2000) The WD-repeat protein pfs2p bridges two essential factors within the yeast pre-mRNA 3’-end-processing complex. EMBO Journal 19, 37–47. Orlando, V. (2003) Polycomb, epigenomes, and control of cell identity. Cell 112, 599–606. Pastori, G.M. and Foyer, C.H. (2002) Common components, networks, and pathways of cross-tolerance to stress. The central role of ‘redox’ and abscisic acid-mediated controls. Plant Physiology 129, 460–468. Pawson, T. (2004) Specificity in signal transduction: from phosphotyrosineSH2 domain interactions to complex cellular systems. Cell 116, 191–203. Perbal, M.C., Haughn, G., Saedler, H. and Schwarz-Sommer, Z. (1996) Non-cell-autonomous function of the Antirrhinum floral homeotic proteins DEFICIENS and GLOBOSA is exerted by their polar cell-to-cell trafficking. Development 122, 3433–3441. Pineiro, M., Gomez-Mena, C., Schaffer, R., Martinez-Zapater, J.M. and Coupland, G. (2003) EARLY BOLTING IN SHORT DAYS is related to chromatin remodeling factors and regulates flowering in Arabidopsis by repressing FT. Plant Cell 15, 1552–1562. Quesada, V., Macknight, R., Dean, C. and Simpson, G.G. (2003) Autoregulation of FCA pre-mRNA processing controls Arabidopsis flowering time. EMBO Journal 22, 3142–3152. Reeves, P.H. and Coupland, G. (2001) Analysis of flowering time control in Arabidopsis by comparison of double and triple mutants. Plant Physiology 126, 1085–1091. Rojo, E., Sharma, V.K., Kovaleva, V., Raikhel, N.V. and Fletcher, J.C. (2002) CLV3 is localized to the extracellular space, where it activates the Arabidop-

sis CLAVATA stem cell signalling pathway. Plant Cell 14, 969–977. Samach, A., Onouchi, H., Gold, S.E., Ditta, G.S., Schwarz-Sommer, Z., Yanofsky, M.F. and Coupland, G. (2000) Distinct roles of CONSTANS target genes in reproductive development of Arabidopsis. Science 288, 1613–1616. Schmid, M., Uhlenhaut, N.H., Godard, F., Demar, M., Bressan, R., Weigel, D. and Lohmann, J.U. (2003) Dissection of floral induction pathways using global expression analysis. Development 130, 6001–6012. Schomburg, F.M., Patton, D.A., Meinke, D.W. and Amasino, R.M. (2001) FPA, a gene involved in floral induction in Arabidopsis, encodes a protein containing RNA-recognition motifs. Plant Cell 13, 1427–1436. Sessions, A., Yanofsky, M.F. and Weigel, D. (2000) Cell–cell signalling and movement by the floral transcription factors LEAFY and APETALA1. Science 289, 779–782. Sheen, J. (2002) Phosphorelay and transcription control in cytokinin signal transduction. Science 296, 1650– 1652. Sheldon, C.C., Burn, J.E., Perez, P.P., Metzger, J., Edwards, J.A., Peacock, W.J. and Dennis, E.S. (1999) The FLF MADS box gene: a repressor of flowering in Arabidopsis regulated by vernalization and methylation. Plant Cell 11, 445–458. Shiu, S. and Bleecker, A.B. (2001) Plant receptor-like kinase gene family: diversity, function, and signalling. Science’s STKE. Available at: www.stke.org/ cgi/content/full/OC_sigtrans;2001/ 113/re22, 1–13 Simpson, G.G. and Dean, C. (2002) Arabidopsis, the Rosetta stone of flowering time? Science 296, 285–289. Simpson, G.G., Dijkwel, P.P., Quesada, V., Henderson, I. and Dean, C. (2003) FY is an RNA 3’ end-processing factor that interacts with FCA to con-

78

R.G. Anthony trol the Arabidopsis floral transition. Cell 113, 777–787. Sudol, M. and Hunter, T. (2000) New wrinkles for an old domain. Cell 103, 1001–1004. Sung, S. and Amasino, R.M. (2004a) Vernalization and epigenetics: how plants remember winter. Current Opinion in Plant Biology 7, 4–10. Sung, S. and Amasino, R.M. (2004b). Vernalization in Arabidopsis thaliana is mediated by the PHD finger protein VIN3. Nature 427, 159–164. Takada, S. and Goto, K. (2003) Terminal flower2, an Arabidopsis homolog of heterochromatin protein1, counteracts the activation of flowering locus T by constans in the vascular tissues of leaves to regulate flowering time. Plant Cell 15, 2856–2865. Takasaki, T., Hatakeyama, K., Suzuki, G., Watanabe, M., Isogai, A. and Hinata, K. (2000) The S receptor kinase determines self-incompatibility in Brassica stigma. Nature 403, 913–916. Trotochaud, A.E., Hao, T., Wu, G., Yang, Z.B. and Clark, S.E. (1999) The CLAVATA1 receptor-like kinase requires CLAVATA3 for its assembly into a signalling complex that includes KAPP and a Rho-related protein. Plant Cell 11, 393–405. Urao, T., Yamaguchi-Shinozaki, K. and Shinozaki, K. (2001) Plant histidine kinases: an emerging picture of twocomponent signal transduction in hormone and environmental responses. Science’s STKE 2001, RE18. Valverde, F., Mouradov, A., Soppe, W., Ravenscroft, D., Samach, A. and Coupland, G. (2004) Photoreceptor regulation of CONSTANS protein in photoperiodic flowering. Science 303, 1003–1006.

Veit, B. (2004) Determination of cell fate in apical meristems. Current Opinion in Plant Biology 7, 57–64. Vincent, C.A., Carpenter, R. and Coen, E.S. (2003) Interactions between gene activity and cell layers during floral development. Plant Journal 33, 765–774. Wagner, D., Sablowski, R.W. and Meyerowitz, E.M. (1999) Transcriptional activation of APETALA1 by LEAFY. Science 285, 582–584. Wang, H. and Deng, X.W. (2003) Dissecting the phytochrome A-dependent signalling network in higher plants. Trends in Plant Science 8, 172–178. Wu, X., Dinneny, J.R., Crawford, K.M., Rhee, Y., Citovsky, V., Zambryski, P.C. and Weigel, D. (2003) Modes of intercellular transcription factor movement in the Arabidopsis apex. Development 130, 3735–3745. Yanovsky, M.J. and Kay, S.A. (2003) Living by the calendar: how plants know when to flower. Nature Reviews Molecular Cell Biology 4, 265–275. Yu, H., Ito, T., Wellmer, F. and Meyerowitz, E.M. (2004a) Repression of AGAMOUS-LIKE 24 is a crucial step in promoting flower development. Nature Genetics 36, 157–161. Yu, H., Ito, T., Zhao, Y., Peng, J., Kumar, P. and Meyerowitz, E.M. (2004b) Floral homeotic genes are targets of gibberellin signalling in flower development. Proceedings of the National Academy of Sciences USA 101, 7827–7832. Zambryski, P. (2004) Cell-to-cell transport of proteins and fluorescent tracers via plasmodesmata during plant development. Journal of Cell Biology 164, 165–168. Zeevaart, J. (1976) Physiology of flower formation. Annual Review of Plant Physiology 27, 321–348.

II

Floral Development

This page intentionally left blank

4

Floral Meristem Identity Genes B. DAVIES Centre for Plant Sciences, University of Leeds, Leeds LS2 9JT, UK, e-mail: [email protected]

Introduction The flexibility of plant development, which enables plants to modify their development in response to environmental conditions, derives from groups of pluripotent cells maintained in the meristems. Meristems are organized structures that must both maintain themselves and provide a source of cells that can differentiate into a variety of tissues and organs, in response to internal and external cues (Carles and Fletcher, 2003). The exact pattern and type of lateral structure produced by the meristems at each stage of development dictates the overall growth habit of the plant. Meristems are defined by the types of differentiated structures produced on their flanks. The shoot apical meristem (SAM) is formed during embryogenesis and, following seed germination, is maintained at the apex of the shoot giving rise to the leaves and branches that form during the vegetative phase of plant development. A whole host of signals, including such factors as the age of the plant, daylength, temperature and light quality and intensity, are combined by a complex signal-integration network to initiate the flowering process (Simpson and Dean, 2002). One of the first steps in this process is the conversion of the SAM into an inflorescence meristem (IM). In some plant species the apical meristem becomes a floral meristem (FM) without the production of further lateral IMs. In many other species the IM produces FMs on its flanks whilst retaining its IM identity to produce an indeterminate inflorescence. The transition from SAM to IM is often accompanied by a variety of characteristic morphological changes such as an altered spatial distribution of lateral organs (phylotaxis). The FM, unlike the SAM and IM, is a determinate structure and will produce the characteristic floral organ primordia, which will form sepals, petals and stamens on its flanks before differentiating to produce the carpels of the flower. For consistency, the Arabidopsis gene names and models have been predominantly used in this chapter, but the reader is encouraged to consider ßCAB International 2006. The Molecular Biology and Biotechnology of Flowering, 2nd edn (ed. B.R. Jordan)

81

82

B. Davies

Fig. 4.1. The transition to flowering in Arabidopsis. A shoot apical meristem (A) forms leaf primordia on its flanks. The inflorescence meristem (B) forms floral meristems (C) on its flanks, which eventually develop into flowers.

the differences that exist between species, in terms of both gene complement and mechanisms. Arabidopsis is an example of a plant that maintains an IM and produces many lateral flowers from FMs produced on the flanks of the IM (Fig. 4.1), a type of growth habit known as monopodial. Therefore, the stages leading up to the formation of an Arabidopsis flower can be crudely summarized as the sequential conversion of a SAM (an indeterminate structure producing leaves) to an IM (an indeterminate structure producing cauline leaves and FMs) followed by the establishment and maintenance of FMs (determinate structures producing floral organs). The purpose of this chapter is to examine what is known about the processes involved in this conversion. Obviously, there is a potential for considerable overlap between this chapter and others in this volume, especially those concerning external and internal regulation of flowering and the molecular biology of floral organogenesis. This is particularly the case because some of the genes that play an early role in FM identity may also have a later role in floral organ development. In order to minimize these potential overlaps, detailed views into floral induction and the establishment of floral organ identity have been restricted and any additional functions of the key genes involved have largely been ignored.

Genes Promoting Floral Meristem Identity: LEAFY is the Driving Force In order to study the genes that act to specify FM identity, it is first necessary to identify mutants in which the normal steps in the establishment of meristem identity are affected. Such mutants have long been known in several

Floral Meristem Identity Genes

83

plant species, but it was the floricaula ( flo) mutant of Antirrhinum majus that was the first to reveal its molecular secrets (Coen et al., 1990). In normal Antirrhinum development the indeterminate IM gives rise to a spiral of lateral, determinate FMs, each flanked by a leaf-like bract. It is this developmental pattern that gives rise to the characteristic flower spike observed in this species. Inflorescence development in the flo mutant is clearly blocked at the point of initiation of lateral FMs, because further IMs form where FMs are expected. These secondary IMs form mutant inflorescence shoots that in turn produce tertiary inflorescences in place of flowers. The result of this developmental defect is a proliferation of inflorescence shoots bearing no flowers – FMs are never established. The FLORICAULA (FLO) gene was cloned by transposon tagging and found to encode a novel protein of unknown function. A mutant, leafy (lfy), had already been described in Arabidopsis that showed some similarity to the flo mutant of Antirrhinum (Haughn and Somerville, 1988). Subsequent isolation of the gene affected in this mutant showed that it encoded a protein that was closely related to FLO and demonstrated that the mechanism of establishment of FM identity is similar in diverse plant species (Weigel et al., 1992). LFY/FLO has now been shown to be a transcription factor with a dual role in flower development (Parcy et al., 1998). On the one hand, LFY promotes FM identity but, on the other, it also has a later role in the activation of the homeotic floral organ identity genes. Further confirmation of the important role of LFY in promoting flowering was obtained by constitutive-expression experiments. Constitutive expression of LFY in Arabidopsis, or in a variety of other plant species, resulted in precocious flowering often accompanied by conversion of IMs to FMs (Weigel and Nilsson, 1995; Nilsson and Weigel, 1997; He et al., 2000; Pena et al., 2001). The feature common to both mutants, lfy in Arabidopsis and flo in Antirrhinum, is the inappropriate adoption of IM identity by lateral meristems, which should be floral. However, the two mutants differ markedly in terms of severity. The flo mutant shows a complete conversion of FMs into IMs whereas the conversion in lfy is only partial. Initially, IMs form in lfy mutants at positions where FMs would form in wild-type plants. However, meristems arising later develop as FMs, although these give rise to abnormal flowers with inflorescence-like characteristics (Weigel et al., 1992) (Fig. 4.2). One explanation for this difference is that a greater degree of redundancy exists in the Arabidopsis genome for the specification of FM identity. Redundancy is often caused by the existence of related genes with overlapping function. We shall see later that although the specification of FM identity in Arabidopsis is complicated by redundancy, in this case the redundancy is not caused by multiple LFY-like genes. LFY is a single copy gene in Arabidopsis. Unlike some of the other meristem identity genes, LFY-like genes, with an apparently analogous function in promoting reproductive shoots, have been identified outside of the angiosperm lineage (Mellerowicz et al., 1998; Mouradov et al., 1998). This suggests that LFY had an ancient role in reproduction and confirms its position as a central regulator of flowering.

84

B. Davies

Wild-type

ap1

cal

ful

ap1 cal

ap1 ful

ful cal

ap1 calful

IM Flower ap1-like flower lfy

lfy ap1

lfy cal

lfy ful

lfy ap1 cal lfy ap1 cal ful

lfy-like flower Proliferating IM

Fig. 4.2. Schematic of inflorescence phenotypes. The inflorescence phenotypes seen in single mutants and mutant combinations of genes affecting floral meristem (FM) identity are shown. In the interests of clarity many important details have been omitted. The two lateral positions shown give an indication of the type of structure (wild-type flower, mutant flower, proliferating meristem, inflorescence) that might be anticipated at early and late stages of inflorescence development. The designation of a flower as wild-type ignores defects apparently unrelated to FM identity, such as altered carpel valves in ful mutants and mutant combinations. Flowers shown as ap1-like have bract-like or leaf-like sepals and often lack second whorl organs, instead forming further mutant FMs in the axils of the first whorl organs. Flowers shown as lfy-like are often formed in the axils of bracts and have more leaf-like floral organs with a tendency towards spiral phylotaxis. Mutant combinations lfy ap1, lfy ap1 cal, lfy ap1 cal ful and ap1 cal ful are depicted without flowers, although they all eventually flower under some conditions (see text). Other effects, such as the effect of ful on flowering time, are not shown.

As a transcription factor, LFY would be expected to influence the expression of downstream target genes. A few of these target genes were identified by genetic and transgenic experiments (Parcy et al., 1998; Busch et al., 1999; Wagner et al., 1999; Lamb et al., 2002; Schmid et al., 2003). Some of these genes, such as APETALA3 (AP3), AGAMOUS (AG) and the SEPALLATA genes (SEP1, SEP2 and SEP3) belong to a subset of LFY targets that relate to LFY’s later role in organ identity. Until recently, only a single LFY target gene, APETALA1 (AP1), was known to be involved in the establishment of FM identity. The use of sophisticated molecular techniques for the controlled induction of gene expression, combined with transcriptomics, has both verified existing targets and identified new ones (Wagner et al., 1999, 2004; William et al., 2004). At present few of these targets can be integrated into our understanding of the determination of meristem identity, but the relative over abundance of transcription factors amongst the list of candidates supports the view that LFY is a master regulatory gene. LFY’s fundamental role in the promotion of flowering is also apparent from its ability to contribute to the production of flowers in inappropriate places. Expression of LFY together with WUSCHEL (WUS), a homeodomain-encoding gene that specifies shoot identity, in Arabidopsis roots is sufficient to produce floral organs and tissues at root tips (Gallois et al., 2004).

Floral Meristem Identity Genes

85

Genes Promoting Floral Meristem Identity: APETALA-1 and CAULIFLOWER the Supporting Cast As suggested above, the fact that later-arising meristems on lfy inflorescences adopt a more floral character suggests that other genes, which are activated later than LFY, can partially compensate for the lack of LFY activity and confer FM identity. AP1 was a prime candidate for such a gene, since the flowers formed in ap1 mutants also show inflorescence characteristics (Irish and Sussex, 1990) (Fig. 4.2). LFY is expressed in the periphery of the SAM at about the time of the switch to IM. A couple of days later both LFY and AP1 expression is boosted in the cells on the flanks of the IM that are destined to become FMs (Ratcliffe et al., 1999). Mutations in AP1 enhance the lfy phenotype and make it look more like that of flo, but even these plants do not show the total absence of FMs seen in flo mutants (Huala and Sussex, 1992; Weigel et al., 1992; Bowman et al., 1993). The flowers which form in lfy ap1 double mutants contain mainly leaf-like or carpeloid organs, with a tendency towards spiral phylotaxis and often have secondary mutant flowers in the axils of the floral organs. In addition to AP1 and LFY, there are other factors that contribute to the establishment of FM identity in Arabidopsis. AP1 has been shown to be a direct target of LFY (Wagner et al., 1999). Since lfy mutants are more severe than ap1 mutants it is clear, however, that AP1 cannot be the only target of LFY. The AP1 gene encodes a MADS-box transcription factor (SchwarzSommer et al., 1990; Mandel et al., 1992), one of several members of this large plant family of transcription factors that are important for flowering (Parenicova et al., 2003). The phenotype of ap1 single mutants is itself enhanced in double mutants with cauliflower (cal), a mutant affecting another AP1-like MADS-box gene (Kempin et al., 1995). Inflorescences formed in ap1 cal double mutants produce a heading phenotype, reminiscent of cauliflower heads, composed of a proliferation of multiple higher order IMs (Bowman et al., 1993). Despite the fact that cal enhances ap1, the lfy ap1 cal triple mutant looks like the lfy ap1 double mutant and proliferation of lateral IMs is not observed (Fig. 4.2). Therefore cal does not appear to enhance lfy mutants (Bowman et al., 1993) (Fig. 4.2). Once again, like LFY, further evidence that AP1 and CAL are involved in the establishment of FM identity comes from overexpression experiments. Transgenic plants that constitutively express either AP1 or CAL flower early and show transformations of IMs to FMs, although the effect is stronger in 35S::AP1 than 35S::CAL (Mandel and Yanofsky, 1995b, Liljegren et al., 1999). Interestingly, the effects of overexpression of AP1 are not suppressed by lfy mutants, whereas the effects of overexpression of LFY are partially suppressed by mutations in AP1 (Mandel and Yanofsky, 1995b, Weigel and Nilsson, 1995). The early expression patterns of AP1 and CAL are similar, both becoming first detectable in the earliest visible floral primordia (Mandel et al., 1992; Kempin et al., 1995). Single cal mutants show no detectable differences

86

B. Davies

from wild-type plants (Fig. 4.2), suggesting that AP1 acts redundantly with CAL and can compensate for the loss of CAL activity. CAL can partially compensate for the loss of AP1 in the FM, with the absence of both AP1 and CAL causing a proliferation of IM tissue in place of flowers. Since this is not observed in lfy ap1 cal triple mutants, LFY is required for the meristemproliferation phenotype. CAL cannot replace later AP1 activity in the flower, possibly because of differences in expression patterns in developing floral organs (Yanofsky, 1995). Expression of both AP1 and CAL is reduced in lfy mutants and it has now been shown that both AP1 and CAL are likely to be direct targets of LFY ( Wagner et al., 2004; William et al., 2004). Furthermore, the enhancement in expression of LFY, AP1 and CAL, that is normally observed as flower production begins, does not happen in ap1 cal mutants (Ferra´ndiz et al., 2000). Thus, LFY directly activates AP1 and CAL, which in turn boost expression of LFY. Even so, the fact that flowering is not completely abolished in lfy ap1 or lfy ap1 cal mutants suggests even more redundant control. AP1-like genes, unlike LFY, appear to be confined to the angiosperms (Litt and Irish, 2003). Some functional analysis has been carried out in other angiosperm models and, whilst the role of AP1-like genes in FM identity is conserved, some differences remain. In Antirrhinum the closest MADS-box gene to AP1 is SQUAMOSA (SQUA). Mutants in SQUA are much more severe than ap1 mutants and very rarely flower. As in the case of LFY/FLO, this suggests a greater degree of redundancy in the control of FM identity in Arabidopsis. However, a more significant difference exists in the later role of AP1/SQUA in flower development (Schwarz-Sommer et al., 2003). Floral organs are specified by the combinatorial activity of a series of genes that have become familiar as the A, B and C class organ identity genes (Coen and Meyerowitz, 1991; Gutierrez-Cortines and Davies, 2000). The A-function genes, usually identified as APETALA2 (AP2) and AP1, are supposed to specify the identity of the sepals and, together with the B-function genes, the petals. It has been proposed, however, that the main role of AP1 in this process is the repression of AG and activation of UNUSUAL FLORAL ORGANS (UFO), a gene involved in floral patterning (Durfee et al., 2003). Whatever the mechanism, Arabidopsis is still the only species where mutations in AP1-like genes have been shown to play a direct role in organ identity. The rare flowers produced on squa mutant inflorescences are often normal and show no evidence of homeotic conversion of the perianth organs (Huijser et al., 1992). The control of FM identity shows a higher degree of redundancy in Arabidopsis and ap1 mutants display only a partial conversion of floral to IM. It is therefore possible that apparent homeotic conversions of the perianth, observed in ap1 mutants, are in fact simply a result of a failure to properly establish FM identity (Motte et al., 1998; Davies et al., 1999; Litt and Irish, 2003). AP1 and CAL can be seen to form a partially redundant autoregulatory gene network with LFY (Fig. 4.3). Much of the downstream activity of LFY is accounted for by the action of AP1 and CAL. AP1 and CAL also, in part, account for the boost in LFY expression that is observed at the transition to

Floral Meristem Identity Genes

A

87

B

C

TFL1 FUL AGL24

TFL1

FM

FUL AGL24

LFY AP1 CAL

LFY TFL1 FUL

AGL24 SAM

TFL1, FUL and AGL24 become independently activated in the SAM. AGL24 promotes IM identity. AGL24 activates LFY.

IM LFY is excluded from IM by TFL1. T FL1 is excluded from FM precursor by LFY. LFY activates AP1 and CAL.

IM AP1/CAL boosts LFY, which boosts AP1/CAL. AP1 and CAL exclude FUL and AGL24 from FM. AP1, CAL and LFY promote FM identity.

Fig. 4.3. A model for the establishment and maintenance of the floral meristem (FM). Initially the genes TFL1, FUL and AGL24 become independently activated in the shoot apical meristem (SAM) as a result of the action of the floral pathway–integrator genes (Simpson and Dean, 2002), shown by open arrows (A). The activity of these genes is described in the text. LFY is activated by AGL24, but is excluded from the inflorescence meristem (IM) by the action of TFL1, which is already established there (B). LFY expression therefore becomes restricted to the founder cells that will become the FM. LFY, in turn, ensures that TFL1 is not expressed in the establishing FM. LFY activates AP1 and CAL in the developing FM (C). AP1 and CAL boost LFY expression and exclude expression of FUL and AGL24 from the FM. AP1, CAL, LFY and other unidentified genes promote further floral development and the activation of the organ identity genes. AGL24 promotes IM identity. Thick lines indicate the activation and repression signals that are already known to be direct.

flowering. LFY activity is not abolished in ap1 cal double mutants and this indicates that other genes can enhance LFY expression in their absence.

Genes Promoting Floral Meristem Identity: FRUITFULL an Unexpected Influence A third AP1-like gene, originally called AGL8, exists in the Arabidopsis genome (Mandel and Yanofsky, 1995a). Initial functional analysis of this gene revealed a role in fruit development and the gene was renamed FRUITFULL (FUL) (Gu et al., 1998). Mutations in the FUL gene affect the development of the valve, replum and style. It was noted, however, that in addition to its later expression, FUL is expressed early in the SAM and is upregulated on induction

88

B. Davies

of flowering. Consistent with this expression pattern, ful mutants flower slightly later than wild type and have abnormal cauline leaves (Gu et al., 1998; Ferra´ndiz et al., 2000). Although FUL is expressed before both AP1 and CAL and plays a role in the induction of flowering, it took extensive and careful genetic analysis to reveal its true position in the network of genes that promote FM identity (Ferra´ndiz et al., 2000). There is no change in the expression of LFY in ful mutants, indicating that the delayed flowering in these mutants is not a consequence of reduced LFY. On the face of it this would indicate that FUL is not involved in the activation of LFY. However, LFY expression, which is already reduced in ap1 cal mutants, is even further reduced in ap1 cal ful triple mutants. Therefore FUL is capable of activating LFY expression in the absence of AP1 and CAL. This probably occurs because AP1 and CAL repress FUL in the floral primordia of wild-type plants and FUL becomes ectopically expressed there in ap1 or ap1 cal mutants. It is therefore not clear whether the ability of FUL to activate LFY plays a significant role in wild-type plants where AP1 and CAL would exclude FUL from the FM. The ap1 cal ful triple mutants produce a proliferation of shoots bearing cauline leaves and axillary cauliflower-like meristems and never flower under normal growth conditions (Fig. 4.2), although flowers can still form under extreme conditions. The ful mutation does not enhance ap1 and ful cal double mutants look like ful single mutants (Fig. 4.2). Double mutants between lfy and ful flower later than either single mutant, showing that both genes act independently on flowering time. To test whether the reduction in LFY expression was responsible for the lack of flowering observed in ap1 cal ful triple mutants, the triple mutant was crossed to plants constitutively expressing LFY (Ferra´ndiz et al., 2000). These plants produced ap1 cal-like flowers showing that LFY overexpression can compensate for the loss of AP1 CAL and FUL. However, although enforced LFY expression causes the non-flowering ap1 cal ful triple mutant to flower, flowering time varies in the different mutant combinations. Hence 35S::LFY flowers earlier than 35S::LFY ap1, which flowers earlier than 35S::LFY ap1 cal that flowers earlier than 35S::LFY ap1 cal ful. This indicates that the AP1, CAL and FUL MADS-box genes are required for the early flowering effect seen in 35S::LFY. LFY expression, though reduced, still plays a role in ap1 cal ful triple mutants (Fig. 4.2). This was demonstrated by making the lfy ap1 cal ful quadruple mutant (Ferra´ndiz et al., 2000). This mutant looks similar to lfy ap1 double mutants, showing that neither FUL nor CAL contributes to FM identity in the absence of LFY and AP1. In contrast to ap1 cal ful triple mutants, the lfy ap1 cal ful quadruple mutant does not have leafy cauliflowers, confirming that LFY is necessary for the proliferation of meristems observed in the triple mutant. Unexpectedly, the lfy ap1 cal ful quadruple mutant, although seriously impaired in its ability to flower, is more likely to flower eventually than the ap1 cal ful triple mutant. This is probably because the triple mutant, even though it expresses LFY at a low level, becomes blocked at the stage of proliferation of multiple IMs. The quadruple mutant does not express LFY and so does not produce a proliferation of IMs and eventually

Floral Meristem Identity Genes

89

produces some inflorescence-like flowers. The ability of lfy ap1 cal ful quadruple mutants to flower suggests that we still have not identified all of the FMpromoting factors. In fact, there are two further MADS-box genes in the AP1/ CAL/FUL clade (AGL79 and AGL12) that might be good candidates for this role (Parenicova et al., 2003).

Other Genes Affecting Floral Meristem Identity Many other genes are known to affect FM identity. It is currently difficult to fit these into the model shown in Fig. 4.3 and this might reflect their indirect relationship or a lack of knowledge. Examples of such genes are AP2 and UFO (Jofuku et al., 1994; Ingram et al., 1995; Wilkinson and Haughn, 1995). Mutations in ap2 and ufo have been reported to enhance the FM to IM transformations observed in ap1 or lfy mutants (Bowman et al., 1993; Pidkowich et al., 1999). UFO is involved in the regulation of B-function organ identity genes and AP2 is part of the mechanism that regulates the spatial expression pattern of the C-function organ identity gene (Jofuku et al., 1994; Ingram et al., 1995; Wilkinson and Haughn, 1995). This makes it problematic to establish whether potential roles for UFO and AP2 in meristem identity are distinct from their later activities in the control of floral organ identity. Earlier in development, genes that promote or repress flowering also obviously influence expression of the meristem identity genes. LFY has been identified as one of the flowering-pathway integrators, a gene at which multiple flowering input signals converge (Simpson and Dean, 2002). Recently, a pair of paralogous homeodomain-encoding genes, PENNYWISE (PNY) and POUND-FOOLISH (PNF ) have been shown to act redundantly between the flowering-time genes and the FM identity genes like LFY (Smith et al., 2004). However, the exact connections between flowering-time genes and the activation of the FM identity genes is still not fully understood. Two of the best-known mutants that affect flowering in a very dramatic way are embryonic flower 1 (emf1) and embryonic flower 2 (emf2). Mutations in EMF1 and EMF2, which encode a predicted transcriptional regulator and a Polycomb-group (PcG) protein (Aubert et al., 2001; Yoshida et al., 2001), result in flowering immediately after germination. PcG-group proteins repress gene expression in Drosophila by a chromatin-remodelling mechanism. Since mutations in EMF1 and EMF2 cause precocious flowering, it is likely that EMF1 and EMF2 act to repress flowering in wild-type plants. Recent transcriptomic analysis of emf1 and emf2 mutants suggests that the targets of EMFmediated repression are likely to be the organ identity genes, acting later than the FM identity genes (Moon et al., 2003). The idea that floral organs can form directly as a result of expression of the floral organ identity genes, without going through the transition of IM and FM stages, is supported by ectopic expression experiments (Honma and Goto, 2001; Pelaz et al., 2001b). EMF1 and EMF2, therefore, appear to be genes involved in the epigenetic repression of the floral organ identity genes throughout plant development.

90

B. Davies

Differentiating the IM from the FM: TERMINAL FLOWER 1 an Enigmatic Regulator of Phase Change So far, we have considered only the establishment of FMs on the flanks of IMs. Independent, early expression of LFY in incipient FMs results in the establishment of FM identity by inducing expression of the genes introduced above. What prevents the IM from also adopting FM identity? Again the first clue came from analysis of mutants of both Antirrhinum and Arabidopsis. Both of these plants normally have an indeterminate inflorescence in which the IM maintains itself whilst producing FMs on its flanks. Mutants exist in both species, in which the IM becomes converted into an FM and inflorescence development terminates in a flower. The centroradialis (cen) mutant of Antirrhinum was the first to be cloned (Bradley et al., 1996) and subsequently, the Arabidopsis terminal flower 1 (tfl1) mutant was found to be caused by a lesion in a related gene (Bradley et al., 1997). Unlike all the foregoing genes, CEN/TFL1 does not encode a transcription factor but shows similarity to mammalian RAF-like kinase inhibitor (RKIP) proteins. In mammals RKIP is involved in the regulation of a kinase cascade that transmits a signal from growth factor receptors to effect changes in growth and differentiation (Corbit et al., 2003). RKIP homologues are found in plants, animals, fungi and bacteria, but little is known about their mode of action (Serre et al., 2001). The connection between the LFY and MADS-box transcription factors described above and such a signalling system is not understood but, however, it works; the interaction between these FM and IM genes has a profound effect on inflorescence architecture. Weak TFL1 expression is first detectable in the centre of the SAM during the vegetative phase when LFY expression is also weakly detectable in the periphery (Ratcliffe et al., 1999) (Fig. 4.3). As the SAM becomes an IM, TFL1 expression is enhanced in the centre of the IM and, a few days later, LFY and AP1 expression increases in the incipient FMs. Secondary shoot meristems also show strong central expression of TFL1 and weak peripheral LFY expression. Thus, from very early stages of inflorescence development the IM is distinguished from the FM by the expression of TFL1 in the IM and LFY/ AP1/CAL in the FMs. This separation of expression of TFL1 and LFY/AP1/ CAL is maintained throughout the life of the plant. In plants constitutively expressing LFY, TFL1 expression is not detected, suggesting that LFY acts to restrict expression of TFL1 in wild-type plants (Ratcliffe et al., 1999). 35S::LFY plants flower early and show conversion of IMs to FMs. These effects are partly reversed in 35S::LFY 35S::TFL1 double transgenic plants, showing that TFL1 can actively prevent the FM-promoting activity of LFY in addition to its role in simply repressing LFY in the IM. If LFY expression is responsible for excluding TFL1 from the developing FMs, TFL1 should become ectopically expressed there in lfy mutants. In fact, this is only the case at lower nodes that would develop as secondary shoots in lfy plants. Ectopic expression of TFL1 is also not observed in ap1 mutants, but in ap1 cal double mutants ectopic TFL1 expression is found at very early stages in the development of the lateral meristems. TFL1 is thus excluded from developing FMs in wild-type plants by

Floral Meristem Identity Genes

91

the combined action of LFY and AP1/CAL (Fig. 4.3). Terminal flowers produced in the tfl1 mutant are the result of the loss of repression of LFY and AP1 in the IM, which results in the ectopic expression of these FM identity genes in the IM. It is not currently known how LFY and AP1/CAL act to restrict expression of TFL1. Experiments in Antirrhinum, using periclinal chimeras expressing FLO in different cell layers, have shown that FLO can act non-cell-autonomously (Carpenter and Coen, 1995; Hantke et al., 1995). Assuming that the same is true for LFY, it is possible to speculate that TFL1 expression is restricted by an unknown signalling mechanism involving cell–cell movement of either the LFY protein itself or some other signalling molecule. Cell–cell movement has already been shown for other plant transcription factors (Lucas et al., 1995; Perbal et al., 1996; Jackson and Hake, 1997; Wu et al., 2003). Perhaps, the answer to this mystery lies amongst the potential LFY target genes that have already been identified (Wagner et al., 2004; William et al., 2004). TFL1 can be seen to have two roles. It acts as part of a system of mutual repression to exclude LFY and AP1/CAL from the IM and is itself restricted to the IM by the action of LFY and AP1/CAL in the FMs (Fig. 4.3). TFL1 is also capable of countering the flower-promoting activity of LFY independently of its control of LFY expression. In fact, TFL1 acts to slow down phase changes from early stages of plant development (Ratcliffe et al., 1998; Michaels et al., 2003). Analysis of the duration of the vegetative, inflorescence and floral phases of development in wild-type, tfl1 mutant and 35S::TFL1 overexpressing plants showed that all phases were shorter than wild-type in the mutant and longer than wild-type in the 35S::TFL1 plants. Manipulating TFL1 expression can have huge consequences for the overall pattern of plant development. Loss of TFL1 in Arabidopsis results in early flowering plants that produce few leaves, lateral shoots and flowers, whilst overexpression of TFL1 delays flowering and produces large bushy plants with numerous flowers. The formation of cauliflower-like heads in ap1 cal and ap1 cal ful mutants and their absence from lfy ap1 cal ful quadruple mutants can also be related to TFL1 (Ratcliffe et al., 1999; Ferra´ndiz et al., 2000). In both ap1 cal and ap1 cal ful mutants, TFL1 becomes ectopically expressed and overlaps with the reduced expression of LFY in what would have been the FMs. In fact, the further loss of either LFY (in lfy ap1 cal ful mutants) or TFL1 (in tfl1 ap1 cal ful mutants) prevents cauliflower formation, supporting the view that coexpression of LFY and TFL1 causes this phenotype. Evidence that the ratio of LFY to TFL1 is the deciding factor for the developmental pathway that follows is provided by ap1 cal ful mutants that are heterozygous for tfl1 (Ferra´ndiz et al., 2000). In contrast to the ap1 cal ful mutants, these plants flower and set seeds under normal conditions. This supports the view that flowering can be promoted even when LFY expression is reduced, by reducing the levels of TFL1. TFL1-like genes have also been studied in plants with different shoot patterns to that of Arabidopsis. Unlike the monopodial habit of Arabidopsis and Antirrhinum, tomato shows a sympodial growth habit in which reproductive

92

B. Davies

and vegetative stages alternate. The initial vegetative shoot terminates in an inflorescence and a new vegetative shoot arises in the axil of a leaf below the inflorescence. Subsequently, the new vegetative shoot terminates in another inflorescence and a further new vegetative shoot arises from an axil of a lower leaf on that shoot. The developmental pattern reiterates, in effect producing a series of linked alternating vegetative and reproductive shoots. The tomato selfpruning (sp) mutation, caused by a defect in a CEN/TFL1-like gene, results in a shortening of the intervening vegetative phases resulting in termination of development by the production of two consecutive inflorescences (Pnueli et al., 1998). Two pea mutants have also been shown to result from the loss of TFL1like gene expression (Foucher et al., 2003). In this case one gene, DETERMINATE (DET) was shown to be involved in the transition from IM to FM, whilst the other, LATE FLOWERING (LF) slowed down the transition from vegetative to IM. TFL1-like genes exist as small gene families in different plant species (Carmel-Goren et al., 2003) and their demonstrated effects on growth habit in mutant and transgenic experiments suggests that variation in TFL1-like gene activity could be a major influence in natural variation between species.

Promoting IM Identity: AGL24 the First IM Identity Gene? AP1, CAL and FUL are all MADS-box transcription factors that act to promote FM identity. AGL24 is another member of this gene family, but its role in flowering is slightly more complicated. Mutants in AGL24, or AGL24 RNA interference (RNAi) lines, flower late and overexpression of AGL24 results in early flowering, suggesting that AGL24 promotes flowering (Yu et al., 2002; Michaels et al., 2003). Using an LFY reporter line, in which the b-glucuronidase (GUS)-expression marker gene is driven by the LFY promoter, it was shown that LFY upregulation is delayed in lines with reduced AGL24 expression. It is therefore possible that the late-flowering phenotype observed in agl24 plants results from a failure to boost LFY expression at the floral transition. This is supported by the finding that overexpression of LFY can rescue the late-flowering phenotype of lines deficient in AGL24 (Yu et al., 2002). Despite this delay, LFY manages to attain wild-type levels of expression in AGL24 RNAi lines later in flower development, indicating that other factors are able to compensate for a loss of AGL24 (Yu et al., 2002). However, constitutive expression of AGL24 has another effect on flower development. Flowers formed on 35S::AGL24 plants showed a partial reversion to inflorescence fate, similar to the effect seen in ap1 mutants (Yu et al., 2004). Floral reversion can be viewed as a reversal of the normal developmental pathway such that the FM identity is lost after the production of the outermost organs and vegetative organs are initiated. Floral reversion is observed particularly clearly in some species, such as Impatiens, but is only observed in Arabidopsis in defined mutant backgrounds (Pouteau et al., 1997, 1998; Parcy et al., 2002). Floral reversion indicates that the acquisition of FM identity is reversible and the FM state must be maintained by the activity of genes. The partial conversion of FMs to IMs caused by overexpression of

Floral Meristem Identity Genes

93

AGL24 indicated that AGL24 might be promoting IM identity. In wild-type plants during vegetative development, AGL24 is expressed throughout the SAM and leaf primordia (Yu et al., 2004). After the floral transition, AGL24 expression can still be observed in the IM and cauline leaves, but becomes excluded from the central layers of the emerging FMs (Yu et al., 2004). In both lfy and ap1 mutants this exclusion fails to occur, suggesting that LFY and AP1 repress AGL24 expression in part of the FMs (Yu et al., 2004). Using an inducible expression system, Yu et al. (2004) were able to demonstrate that although both AP1 and LFY are capable of repressing ALG24, only in the case of AP1 is this repression likely to be direct. It is therefore possible that the indirect repression of AGL24 by LFY takes place by the LFY-induced activation of AP1, which is a direct repressor of AGL24. The AGL24 protein has been shown to associate with a meristematic receptor-like kinase, MRLK, which is predominantly expressed in SAM and root apical meristems (Fujita et al., 2003). Since AGL24 is not expressed in root apical meristems, this interaction is only likely to have a biological function in the SAM. MRLK can phosphorylate the AGL24 protein and AGL24 is only localized to the nucleus in meristematic tissue, raising the possibility that phosphorylation might be used to regulate its cellular localization. Other plant MADS-box transcription factors have also been suggested to be subject to regulation by control of their subcellular localization (McGonigle et al., 1996; Perry et al., 1996; Zachgo et al., 1997; Immink et al., 2002). The results above are consistent with a model of FM establishment and maintenance in which AGL24 is involved early in flowering in the SAM to IM switch, is subsequently instrumental in activating LFY expression (which in turn activates AP1) and later becomes repressed in the developing FMs by the activity of AP1 (Fig. 4.3). As yet it is unclear exactly how AGL24 promotes IM identity, or what significance its cellular localization and modification might play in this process.

Conclusions It is apparent that enormous progress has been made in understanding the way in which the FM is established and maintained. At least in Arabidopsis, we have a very robust framework on which to base future studies (Fig. 4.3). This model is already sufficient to begin to use genes in this pathway to manipulate flowering and inflorescence architecture. Despite this, more research is required. Some of the genetic interactions are direct, but many are not and we currently have no idea how these are mediated. The exact mechanism by which TFL1 regulates phase change and acts in the mutual exclusion of LFY and AP1 is still a mystery. Even the interactions between the many MADS-box genes that act sequentially throughout the flowering pathway are still far from fully understood. MADS-box genes such as FLOWERING LOCUS C (FLC), SUPPRESSOR OF OVEREXPRESSION OF CONSTANS (SOC1), FLOWERING LOCUS M (FLM) and SHORT VEGETATIVE PHASE (SVP) are involved in the initiation of flowering (Michaels and Amasino, 1999; Sheldon et al.,

94

B. Davies

1999; Borner et al., 2000; Hartmann et al., 2000; Lee et al., 2000; Samach et al., 2000; Ratcliffe et al., 2001; Scortecci et al., 2001). Similarly, most of the later acting ‘ABC’ floral organ identity genes, PISTILLATA (PI), AP3, AG, SEP1, SEP2 and SEP3 are also members of the same gene family (Yanofsky et al., 1990; Jack et al., 1992; Goto and Meyerowitz, 1994; Pelaz et al., 2000). MADS-box factors are known to form dimers, heterodimers and higher order complexes with each other (Davies et al., 1996; Fan et al., 1997; EgeaCortines et al., 1999; Honma and Goto, 2001; Immink et al., 2002). Many of these flowering time and organ-identity factors will be coexpressed with the meristem identity MADS-box genes at some stages during the transition from SAM to developing flower. Indeed, specific interactions have already been shown to take place between both meristem-identity factors, AP1 and CAL, and factors involved in the early initiation of flowering (SOC1, SVP and AGL24) and the later determination of floral organ identity (SEP3) (Pelaz et al., 2001a). It is therefore likely that specificity in spatial and temporal expression, protein–protein interaction, protein–DNA interaction and interaction with cofactors will play an important role in this developmental pathway. Given the very large number of potential interactions, sorting this out represents a huge challenge for the future. It also remains to be seen to what extent the Arabidopsis model holds true for other species and thorough evolutionary developmental analyses of other model species are eagerly awaited.

Acknowledgements Trying to review FM identity without trampling too much on the related areas of floral induction and flower development inevitably means that these important topics will receive little attention. I apologize for not discussing or citing important work in these areas and for having to omit some primary references in order to save space. Grateful thanks go to Zsuzsanna Schwarz-Sommer, Martin Kieffer, Barry Causier and Cristina Ferra´ndiz for their critical reading and helpful comments on this chapter.

References Aubert, D., Chen, L., Moon, Y.H., Martin, D., Castle, L.A., Yang, C.H. and Sung, Z.R. (2001) EMF1, a novel protein involved in the control of shoot architecture and flowering in Arabidopsis. Plant Cell 13, 1865– 1875. Borner, R., Kampmann, G., Chandler, J., Gleissner, R., Wisman, E., Apel, K. and Melzer, S. (2000) A MADS domain gene involved in the transition to

flowering in Arabidopsis. Plant Journal 24, 591–599. Bowman, J.L., Alvarez, J., Weigel, D., Meyerowitz, E. and Smyth, D.R. (1993) Control of flower development in Arabidopsis thaliana by APETALA1 and interacting genes. Development 119, 721–743. Bradley, D., Carpenter, R., Copsey, L., Vincent, C., Rothstein, S. and Coen, E. (1996) Control of inflorescence

Floral Meristem Identity Genes architecture in Antirrhinum. Nature 379, 791–797. Bradley, D., Ratcliffe, O., Vincent, C., Carpenter, R. and Coen, E. (1997) Inflorescence commitment and architecture in Arabidopsis. Science 275, 80–83. Busch, M.A., Bomblies, K. and Weigel, D. (1999) Activation of a floral homeotic gene in Arabidopsis. Science 285, 585–587. Carles, C.C. and Fletcher, J.C. (2003) Shoot apical meristem maintenance: the art of a dynamic balance. Trends in Plant Science 8, 394–401. Carmel-Goren, L., Liu, Y.S., Lifschitz, E. and Zamir, D. (2003) The self-pruning gene family in tomato. Plant Molecular Biology 52, 1215–1222. Carpenter, R. and Coen, E.S. (1995) Transposon induced chimeras show that floricaula, a meristem identity gene, acts non-autonomously between cell layers. Development 121, 19–26. Coen, E.S. and Meyerowitz, E.M. (1991) The war of the whorls: genetic interactions controlling flower development. Nature 353, 31–37. Coen, E.S., Romero, J.M., Doyle, S., Elliott, R., Murphy, G. and Carpenter, R. (1990) Floricaula: a homeotic gene required for flower development in Antirrhinum majus. Cell 63, 1311– 1322. Corbit, K.C., Trakul, N., Eves, E.M., Diaz, B., Marshall, M. and Rosner, M.R. (2003) Activation of Raf-1 signaling by protein kinase C through a mechanism involving Raf kinase inhibitory protein. Journal of Biological Chemistry 278, 13061–13068. Davies, B., Egea-Cortines, M., de Andrade Silva, E., Saedler, H. and Sommer, H. (1996) Multiple interactions amongst floral homeotic proteins. EMBO Journal 15, 4330–4343. Davies, B., Sommer, H. and SchwarzSommer, Zs. (1999) Floral development: genetic views and molecular news. In: Russo, V.A.E. (ed.) Development: Genetics, Epigenetics and En-

95 vironmental Regulation. Springer, Berlin, pp. 167–183. Durfee, T., Roe, J.L., Sessions, R.A., Inouye, C., Serikawa, K., Feldmann, K.A., Weigel, D. and Zambryski, P.C. (2003) The F-box-containing protein UFO and AGAMOUS participate in antagonistic pathways governing early petal development in Arabidopsis. Proceedings of the National Academy of Sciences USA 100, 8571– 8576. Egea-Cortines, M., Saedler, H. and Sommer, H. (1999) Ternary complex formation between the MADS-box proteins SQUAMOSA, DEFICIENS and GLOBOSA is involved in the control of floral architecture in Antirrhinum majus. EMBO Journal 18, 5370–5379. Fan, H.Y., Hu, Y., Tudor, M. and Ma, H. (1997) Specific interactions between the K domains of AG and AGLs, members of the MADS domain family of DNA binding proteins. Plant Journal 12, 999–1010. Ferra´ndiz, C., Gu, Q., Martienssen, R. and Yanofsky, M.F. (2000) Redundant regulation of meristem identity and plant architecture by FRUITFULL, APETALA1 and CAULIFLOWER. Development 127, 725–734. Foucher, F., Morin, J., Courtiade, J., Cadioux, S., Ellis, N., Banfield, M.J. and Rameau, C. (2003) DETERMINATE and LATE FLOWERING are two TERMINAL FLOWER1/CENTRORADIALIS homologs that control two distinct phases of flowering initiation and development in pea. Plant Cell 15, 2742–2754. Fujita, H., Takemura, M., Tani, E., Nemoto, K., Yokota, A. and Kohchi, T. (2003) An Arabidopsis MADS-box protein, AGL24, is specifically bound to and phosphorylated by meristematic receptor-like kinase (MRLK). Plant Cell Physiology 44, 735–742. Gallois, J.L., Nora, F.R., Mizukami, Y. and Sablowski, R. (2004) WUSCHEL induces shoot stem cell activity and

96

B. Davies developmental plasticity in the root meristem. Genes & Development 18, 375–380. Goto, K. and Meyerowitz, E.M. (1994) Function and regulation of the Arabidopsis floral homeotic gene PISTILLATA. Genes & Development 8, 1548–1560. Gu, Q., Ferra´ndiz, C., Yanofsky, M.F. and Martienssen, R. (1998) The FRUITFULL MADS-box gene mediates cell differentiation during Arabidopsis fruit development. Development 125, 1509–1517. Gutierrez-Cortines, M.E. and Davies, B. (2000) Beyond the ABCs: ternary complex formation in the control of floral organ identity. Trends in Plant Science 5, 471–476. Hantke, S.S., Carpenter, R. and Coen, E.S. (1995) Expression of floricaula in single cell layers of periclinal chimeras activates downstream homeotic genes in all layers of floral meristems. Development 121, 27–35. Hartmann, U., Hohmann, S., Nettesheim, K., Wisman, E., Saedler, H. and Huijser, P. (2000) Molecular cloning of SVP: a negative regulator of the floral transition in Arabidopsis. Plant Journal 21, 351–360. Haughn, G.W. and Somerville, C.R. (1988) Genetic control of morphogenesis in Arabidopsis. Developmental Genetics 9, 73–89. He, Z., Zhu, Q., Dabi, T., Li, D., Weigel, D. and Lamb, C. (2000) Transformation of rice with the Arabidopsis floral regulator LEAFY causes early heading. Transgenic Research 9, 223–227. Honma, T. and Goto, K. (2001) Complexes of MADS-box proteins are sufficient to convert leaves into floral organs. Nature 409, 525–529. Huala, E. and Sussex, I.M. (1992) LEAFY interacts with floral homeotic genes to regulate Arabidopsis floral development. Plant Cell 4, 901–913. ¨ nnig, W.-E., MeiHuijser, P., Klein, J., Lo jer, H., Saedler, H. and Sommer, H. (1992) Bractomania, an inflorescence

anomaly, is caused by the loss of function of the MADS-box gene squamosa in Antirrhinum majus. EMBO Journal 11, 1239–1249. Immink, R.G., Gadella, T.W. Jr, Ferrario, S., Busscher, M. and Angenent, G.C. (2002) Analysis of MADS box protein– protein interactions in living plant cells. Proceedings of the National Academy of Sciences USA 99, 2416– 2421. Ingram, G.C., Goodrich, J., Wilkinson, M.D., Simon, R., Haughn, G.W. and Coen, E.S. (1995) Parallels between unusual floral organs and fimbriata, genes controlling flower development in Arabidopsis and Antirrhinum. Plant Cell 7, 1501–1510. Irish, V.F. and Sussex, I.M. (1990) Function of the apetala-1 gene during Arabidopsis floral development. Plant Cell 2, 741–753. Jack, T., Brockman, L.L. and Meyerowitz, E.M. (1992) The homeotic gene APETALA3 of Arabidopsis thaliana encodes a MADS box and is expressed in petals and stamens. Cell 68, 683– 697. Jackson, D. and Hake, S. (1997) Morphogenesis on the move: cell-to-cell trafficking of plant regulatory proteins. Current Opinion in Genetics & Development 7, 495–500. Jofuku, K.D., den Boer, B.G.W., Van Montagu, M. and Okamuro, J.K. (1994) Control of Arabidopsis flower and seed development by the homeotic gene APETALA2. Plant Cell 6, 1211–1225. Kempin, S.A., Savidge, B. and Yanofsky, M.F. (1995) Molecular basis of the cauliflower phenotype in Arabidopsis. Science 267, 522–525. Lamb, R.S., Hill, T.A., Tan, Q.K. and Irish, V.F. (2002) Regulation of APETALA3 floral homeotic gene expression by meristem identity genes. Development 129, 2079–2086. Lee, H., Suh, S.S., Park, E., Cho, E., Ahn, J.H., Kim, S.G., Lee, J.S., Kwon, Y.M. and Lee, I. (2000) The

Floral Meristem Identity Genes AGAMOUS-LIKE 20 MADS domain protein integrates floral inductive pathways in Arabidopsis. Genes & Development 14, 2366–2376. Liljegren, S.J., Gustafson-Brown, C., Pinyopich, A., Ditta, G.S. and Yanofsky, M.F. (1999) Interactions among APETALA1, LEAFY, and TERMINAL FLOWER1 specify meristem fate. Plant Cell 11, 1007–1018. Litt, A. and Irish, V.F. (2003) Duplication and diversification in the APETALA1/ FRUITFULL floral homeotic gene lineage: implications for the evolution of floral development. Genetics 165, 821–833. Lucas, W.J., Bouche´-Pillon, S., Jackson, D.P., Nguyen, L., Baker, L., Ding, B. and Hake, S. (1995) Selective trafficking of KNOTTED1 homeodomain protein and its mRNA through plasmodesmata. Science 270, 1980–1983. Mandel, M.A. and Yanofsky, M.F. (1995a) The Arabidopsis AGL8 MADS box gene is expressed in inflorescence meristems and is negatively regulated by APETALA1. Plant Cell 7, 1763–1771. Mandel, M.A. and Yanofsky, M.F. (1995b) A gene triggering flower formation in Arabidopsis. Nature 377, 522–524. Mandel, M.A., Gustafson-Brown, C., Savidge, B. and Yanofsky, M.F. (1992) Molecular characterization of the Arabidopsis floral homeotic gene APETALA1. Nature 360, 273–277. McGonigle, B., Bouhidel, K. and Irish, V.F. (1996) Nuclear localization of the Arabidopsis APETALA3 and PISTILLATA homeotic gene products depends on their simultaneous expression. Genes & Development 10, 1812–1821. Mellerowicz, E.J., Horgan, K., Walden, A., Coker, A. and Walter, C. (1998) PRFLL – a Pinus radiata homologue of FLORICAULA and LEAFY is expressed in buds containing vegetative shoot and undifferentiated male cone primordia. Planta 206, 619–629.

97 Michaels, S.D. and Amasino, R.M. (1999) FLOWERING LOCUS C encodes a novel MADS domain protein that acts as a repressor of flowering. Plant Cell 11, 949–956. Michaels, S.D., Ditta, G., GustafsonBrown, C., Pelaz, S., Yanofsky, M. and Amasino, R.M. (2003) AGL24 acts as a promoter of flowering in Arabidopsis and is positively regulated by vernalization. Plant Journal 33, 867–874. Moon, Y.H., Chen, L., Pan, R.L., Chang, H.S., Zhu, T., Maffeo, D.M. and Sung, Z.R. (2003) EMF genes maintain vegetative development by repressing the flower program in Arabidopsis. Plant Cell 15, 681–693. Motte, P., Saedler, H. and SchwarzSommer, Zs. (1998) STYLOSA and FISTULATA: regulatory components of the homeotic control of Antirrhinum floral organogenesis. Development 125, 71–84. Mouradov, A., Glassick, T., Hamdorf, B., Murphy, L., Fowler, B., Marla, S. and Teasdale, R.D. (1998) NEEDLY, a Pinus radiata ortholog of FLORICAULA/LEAFY genes, expressed in both reproductive and vegetative meristems. Proceedings of the National Academy of Sciences USA 95, 6537–6542. Nilsson, O. and Weigel, D. (1997) Modulating the timing of flowering. Current Opinion in Biotechnology 8, 195– 199. Parcy, F., Bomblies, K. and Weigel, D. (2002) Interaction of LEAFY, AGAMOUS and TERMINAL FLOWER1 in maintaining floral meristem identity in Arabidopsis. Development 129, 2519–2527. Parcy, F., Nilsson, O., Busch, M.A., Lee, I. and Weigel, D. (1998) A genetic framework for floral patterning. Nature 395, 561–566. Parenicova, L., de Folter, S., Kieffer, M., Horner, D.S., Favalli, C., Busscher, J., Cook, H.E., Ingram, R.M., Kater, M.M., Davies, B., Angenent, G.C. and Colombo, L. (2003) Molecular

98

B. Davies and phylogenetic analyses of the complete MADS-box transcription factor family in Arabidopsis: new openings to the MADS world. Plant Cell 15, 1538–1551. Pelaz, S., Ditta, G.S., Baumann, E., Wisman, E. and Yanofsky, M.F. (2000) B and C floral organ identity functions require SEPALLATA MADS-box genes. Nature 405, 200–203. Pelaz, S., Gustafson-Brown, C., Kohalmi, S.E., Crosby, W.L. and Yanofsky, M.F. (2001a) APETALA1 and SEPALLATA3 interact to promote flower development. Plant Journal 26, 385–394. Pelaz, S., Tapia-Lopez, R., AlvarezBuylla, E.R. and Yanofsky, M.F. (2001b) Conversion of leaves into petals in Arabidopsis. Current Biology 11, 182–184. Pena, L., Martin-Trillo, M., Juarez, J., Pina, J.A., Navarro, L. and MartinezZapater, J.M. (2001) Constitutive expression of Arabidopsis LEAFY or APETALA1 genes in citrus reduces their generation time. Nature Biotechnology 19, 263–267. Perbal, M.-C., Haughn, G., Saedler, H. and Schwarz-Sommer, Zs. (1996) Noncell-autonomous function of the Antirrhinum floral homeotic proteins DEFICIENS and GLOBOSA is exerted by their polar cell-to-cell trafficking. Development 122, 3433–3441. Perry, S.E., Nichols, K.W. and Fernadez, D.E. (1996) The MADS domain protein AGL15 localizes to the nucleus during early stages of seed development. Plant Cell 8, 1977–1989. Pidkowich, M.S., Klenz, J.E. and Haughn, G.W. (1999) The making of a flower: control of floral meristem identity in Arabidopsis. Trends in Plant Science 4, 64–70. Pnueli, L., Carmel-Goren, L., Hareven, D., Gutfinger, T., Alvarez, J., Ganal, M., Zamir, D. and Lifschitz, E. (1998) The SELF-PRUNING gene of tomato regulates vegetative to reproductive switching of sympodial meristems and

is the ortholog of CEN and TFL1. Development 125, 1979–1989. Pouteau, S., Nicholls, D., Tooke, F., Coen, E. and Battey, N. (1997) The induction and maintenance of flowering in Impatiens. Development 124, 3343–3351. Pouteau, S., Tooke, F. and Battey, N. (1998) Quantitative control of inflorescence formation in Impatiens balsamina. Plant Physiol 118, 1191– 1201. Ratcliffe, O.J., Amaya, I., Vincent, C.A., Rothstein, S., Carpenter, R., Coen, E.S. and Bradley, D.J. (1998) A common mechanism controls the life cycle and architecture of plants. Development 125, 1609–1615. Ratcliffe, O.J., Bradley, D.J. and Coen, E.S. (1999) Separation of shoot and floral identity in Arabidopsis. Development 126, 1109–1120. Ratcliffe, O.J., Nadzan, G.C., Reuber, T.L. and Riechmann, J.L. (2001) Regulation of flowering in Arabidopsis by an FLC homologue. Plant Physiology 126, 122–132. Samach, A., Onouchi, H., Gold, S.E., Ditta, G.S., Schwarz-Sommer, Zs., Yanofsky, M.F. and Coupland, G. (2000) Distinct roles of CONSTANS target genes in reproductive development of Arabidopsis. Science 288, 1613–1616. Schmid, M., Uhlenhaut, N.H., Godard, F., Demar, M., Bressan, R., Weigel, D. and Lohmann, J.U. (2003) Dissection of floral induction pathways using global expression analysis. Development 130, 6001–6012. Schwarz-Sommer, Zs., Huijser, P., Nacken, W., Saedler, H. and Sommer, H. (1990) Genetic control of flower development by homeotic genes in Antirrhinum majus. Science 250, 931–936. Schwarz-Sommer, Zs., Davies, B. and Hudson, A. (2003) An everlasting pioneer: the story of Antirrhinum research. Nature Reveiws Genetics 4, 657–666.

Floral Meristem Identity Genes Scortecci, K.C., Michaels, S.D. and Amasino, R.M. (2001) Identification of a MADS-box gene, FLOWERING LOCUS M, that represses flowering. Plant Journal 26, 229–236. Serre, L., Pereira de Jesus, K., Zelwer, C., Bureaud, N., Schoentgen, F. and Benedetti, H. (2001) Crystal structures of YBHB and YBCL from Escherichia coli, two bacterial homologues to a Raf kinase inhibitor protein. Journal of Molecular Biology 310, 617–634. Sheldon, C.C., Burn, J.E., Perez, P.P., Metzger, J., Edwards, J.A., Peacock, W.J. and Dennis, E.S. (1999) The FLF MADS box gene: a repressor of flowering in Arabidopsis regulated by vernalization and methylation. Plant Cell 11, 445–458. Simpson, G.G. and Dean, C. (2002) Arabidopsis, the Rosetta stone of flowering time? Science 296, 285–289. Smith, H.M., Campbell, B.C. and Hake, S. (2004) Competence to respond to floral inductive signals requires the homeobox genes PENNYWISE and POUND-FOOLISH. Current Biology 14, 812–817. Wagner, D., Sablowski, R.W. and Meyerowitz, E.M. (1999) Transcriptional activation of APETALA1 by LEAFY. Science 285, 582–584. Wagner, D., Wellmer, F., Dilks, K., William, D., Smith, M.R., Kumar, P.P., Riechmann, J.L., Greenland, A.J. and Meyerowitz, E.M. (2004) Floral induction in tissue culture: a system for the analysis of LEAFY-dependent gene regulation. Plant Journal 39, 273–282. Weigel, D., Alvarez, J., Smyth, D.R., Yanofsky, M.F. and Meyerowitz, E.M. (1992) LEAFY controls floral meristem identity in Arabidopsis. Cell 69, 843–859. Weigel, D. and Nilsson, O. (1995) A developmental switch sufficient for flower initiation in diverse plants. Nature 377, 495–500. Wilkinson, M.D. and Haughn, G.W. (1995) UNUSUAL FLORAL OR-

99 GANS controls meristem identity and organ primordia fate in Arabidopsis. Plant Cell 7, 1485–1499. William, D.A., Su, Y., Smith, M.R., Lu, M., Baldwin, D.A. and Wagner, D. (2004) Genomic identification of direct target genes of LEAFY. Proceedings of the National Academy of Sciences USA 101, 1775–1780. Wu, X., Dinneny, J.R., Crawford, K.M., Rhee, Y., Citovsky, V., Zambryski, P.C. and Weigel, D. (2003) Modes of intercellular transcription factor movement in the Arabidopsis apex. Development 130, 3735–3745. Yanofsky, M.F. (1995) Floral meristems to floral organs: genes controlling early events in Arabidopsis flower development. Annual Review of Plant Physiology and Plant Molecular Biology 46, 167–188. Yanofsky, M.F., Ma, H., Bowman, J.L., Drews, G.N., Feldmann, K.A. and Meyerowitz, E.M. (1990) The protein encoded by the Arabidopsis homeotic gene agamous resembles transcription factors. Nature 346, 35–39. Yoshida, N., Yanai, Y., Chen, L., Kato, Y., Hiratsuka, J., Miwa, T., Sung, Z.R. and Takahashi, S. (2001) EMBRYONIC FLOWER2, a novel polycomb group protein homolog, mediates shoot development and flowering in Arabidopsis. Plant Cell 13, 2471–2481. Yu, H., Xu, Y., Tan, E.L. and Kumar, P.P. (2002) AGAMOUS-LIKE 24, a dosage-dependent mediator of the flowering signals. Proceedings of the National Academy of Sciences USA 99, 16336–16341. Yu, H., Ito, T., Wellmer, F. and Meyerowitz, E.M. (2004) Repression of AGAMOUS-LIKE 24 is a crucial step in promoting flower development. Nature Genetics 36, 157–161. Zachgo, S., Saedler, H. and SchwarzSommer, Z. (1997) Pollen-specific expression of DEFH125, a MADS-box transcription factor in Antirrhinum with unusual features. Plant Journal 11, 1043–1050.

5

Molecular Biology of Floral Organogenesis B. KRIZEK Department of Biological Sciences, University of South Carolina, Columbia, SC 29208, USA, e-mail: [email protected]

Introduction Our understanding of the molecular nature of floral organ development has increased tremendously over the last 15 years. In the early 1990s, genetic studies on Arabidopsis thaliana and Antirrhinum majus led to the isolation and characterization of floral organ-identity genes (also called floral homeotic genes) and the establishment of the seminal ABC model for flower development. This model proposed that different organ-identity genes act alone and in various combinations to specify each of the four types of floral organs. Tests using ectopic expression support the basic tenets of the model. More recent work has addressed additional issues regarding floral organogenesis. What is the molecular basis for the combinatorial nature of the ABC model? How are the spatially restricted expression patterns of the floral organ-identity genes established and maintained? What factors act downstream of the floral organidentity genes to actually build each floral organ? This chapter focuses primarily on studies carried out in Arabidopsis because these studies have provided the most detailed picture of floral organogenesis. The availability of important tools and resources for this model plant (including a completely sequenced genome, large numbers of insertional mutant lines and DNA microarrays) has contributed greatly to the fantastic progress made in flower development over the last several years.

Floral Organ-identity Genes and the ABC Model During reproductive development, the shoot apical meristem (SAM) initiates floral meristems (FMs) along its flanks. From each FM, organ primordia arise in a characteristic pattern with different organ types forming in each of the four concentric rings (called whorls) (Fig. 5.1). In Arabidopsis flowers, four sepal 100

ßCAB International 2006. The Molecular Biology and Biotechnology of Flowering, 2nd edn (ed. B.R. Jordan)

Molecular Biology of Floral Organogenesis

101

Fig. 5.1. Schematic of a wild-type Arabidopsis flower. The black circle indicates the position of the inflorescence meristem.

primordia are initiated from cells corresponding to the outermost whorl (whorl one). Four petal primordia initiate in whorl two at locations between each of the four sepal primordia. Six stamen primordia are initiated in whorl three with four longer medial stamens and two shorter lateral stamens. Two congenitally fused carpel primordia are formed from cells in the innermost fourth whorl. The FM is determinate with all meristematic cells being used up in the process of organ initiation. The identities of these different organs are specified by the actions of floral organ-identity genes in different regions of a developing flower. The ABC model for flower development proposed that three classes of these organidentity genes function in overlapping domains to specify sepals in whorl one, petals in whorl two, stamens in whorl three and carpels in whorl four (Carpenter and Coen, 1990; Schwarz-Sommer et al., 1990; Bowman et al., 1991; Meyerowitz et al., 1991) (Fig. 5.2). The A class genes APETALA1 (AP1) and APETALA2 (AP2) act to specify sepal and petal development, the B class genes APETALA3 (AP3) and PISTILLATA (PI) act to specify petal and stamen development and the C class gene AGAMOUS (AG) acts to specify stamen and carpel development (Table 5.1). Thus, sepal identity is conferred by class A activity, petal identity by the combined action of A þ B, stamen identity by the combined action of B þ C and carpel identity by C activity. The region-specific activities of the class A and C genes result from antagonism between these genes (Fig. 5.2). Mutations in the class A gene AP2 result in C activity spreading into whorls one and two and consequently homeotic transformations in organ identity with carpels replacing sepals in whorl one and stamens replacing petals in whorl two. Mutations in the class C gene AG result in class A activity in all four whorls. The ag mutants produced indeterminate flowers repeating the pattern of organs (Se Pe Pe)n indicating that AG is required for floral determinacy as well as reproductive organ development. Mutations in the class B genes result in flowers with sepals in the outer two whorls and carpels in the inner two whorls. ABC triple mutants produce indeterminate flowers consisting only of leaf-like organs indicating that the developmental ground state of a floral organ is a leaf and that the activities of the ABC genes are required to convert leaves into floral organs. In all but one case, the organ-identity genes are expressed at the RNA level in spatially restricted regions of a FM consistent with the activities of these

102

B. Krizek

Wild type AP3 B PI

Se 1

AP2 A AP1

AG

Pe 2

St 3

Ca 4

St 3

Ca 4

C

Class A mutant

Ca 1

AP3 PI

B

AG

C

St 2

Class B mutant

AP2 A AP1 Se 1

AG C

Se 2

Ca 3

Ca 4

Class C mutant AP3 PI

B AP2 A AP1

Se 1

Pe 2

Pe 3

Sa 4

Fig. 5.2. The ABC model of flower development. The class A, B and C genes function in different regions of a floral meristem to specify the identities of each type of floral organ. Mutations in these genes result in homeotic transformations in organ identity as indicated for each class. Note: Se, sepals; Pe, petals; St, stamens; Ca, carpels.

genes as proposed in the ABC model (Fig. 5.3). These genes are expressed prior to the emergence of organ primordia and their expression is maintained during organ development. The class C gene AG is expressed in cells that will give rise to the third and fourth whorls of a flower and subsequently in developing stamen and carpel primordia (Yanofsky et al., 1990). The two class B genes are expressed in largely overlapping domains that correspond to second and third whorl cells and later in petal and stamen primordia (Jack et al., 1992, 1994; Goto and Meyerowitz, 1994). In addition to second and third whorl expression, AP3 mRNA is detected at later stages at the base of sepals (Weigel and Meyerowitz, 1993). PI mRNA is present in some fourth whorl cells during stages 3 and 4 but is not maintained in these cells at later developmental stages (Goto and Meyerowitz, 1994). The class A gene AP1 is expressed throughout very young FMs (stages 1 and 2), consistent with its early

Molecular Biology of Floral Organogenesis

103

Table 5.1. Genes involved in floral organogenesis.

Floral meristem–identity genes LEAFY (LFY) APETALA1 (AP1) ABC genes APETALA1 (AP1) APETALA2 (AP2) APETALA3 (AP3) PISTILLATA (PI) AGAMOUS (AG) Caa identity CRABS CLAW (CRC) SPATULA (SPT) ABC cofactors SEPALLATA1 (SEP1) SEPALLATA2 (SEP2) SEPALLATA3 (SEP3) ABC regulators UNUSUAL FLORAL ORGANS (UFO) WUSCHEL (WUS) LEUNIG (LUG) SEUSS (SEU) STERILE APETALA (SAP) CURLY LEAF (CLF) INCURVATA2 (ICU2) BELLRINGER (BLR) AINTEGUMENTA (ANT) HUA1 HUA2 HUA ENHANCER2 (HEN2) HUA ENHANCER4 (HEN4) HUA ENHANCER1 (HEN1) Downstream targets genes NAC-like, activated by AP3/PI (NAP) SHATTERPROOF2 (SHP2) SPOROCYTELESS (SPL)

Function

Gene product

Specify floral meristem identity Specify floral meristem identity

DNA binding MADS domain

Class Class Class Class Class

MADS domain AP2/ERF MADS domain MADS domain MADS domain

A gene A gene B gene B gene C gene

Ca development Ca development

YABBY domain bHLH

Specify Pe, St, Ca identitya Specify Pe, St, Ca identitya Specify Pe, St, Ca identitya

MADS domain MADS domain MADS domain

Activate class B expression

F box

Activate class C expression Repress AG expression Repress AG expression Repress AG expression Repress AG expression Repress AG expression Repress AG expression Repress AG expression Processing of AG RNA Processing of AG RNA Processing of AG RNA Processing of AG RNA miRNA metabolism

Homeodomain Tup1-like corepressor Ldb-like Plant specific Polycomb group Homeodomain AP2/ERF CCCH zinc finger RPR domain RNA helicase? KH domain Novel

Regulate petal growth

NAC

Fruit dehiscence zone development Sporogenesis, ovule patterning

MADS domain Nuclear

a

Note: Se, sepals; Pe, petals; St, stamens; Ca, carpels.

role in specification of FM identity (Mandel et al., 1992b). By stage 3, AP1 mRNA becomes restricted to the just-arising sepal primordia and cells corresponding to the second whorl (Mandel et al., 1992b). The only floral organidentity gene whose mRNA pattern is not spatially localized within a flower is AP2 (Jofuku et al., 1994). The restriction of AP2 activity to whorls one and two appears to occur at the translational level and involves regulation by a microRNA (Aukerman and Sakai, 2003; Chen, 2004).

104

B. Krizek

Fig. 5.3. Expression patterns of the class ABC MADS-box genes in stage 3. Sepal primordia have initiated in stage 3 flowers. All organ primordia have initiated by stage 6 of flower development. Note: Se, sepals; Pe, petals; St, stamens; Ca, carpels.

Ectopic Expression Studies Provide Support for the ABC Model The basic hypotheses of the ABC model have been supported by several ectopic expression studies. Misexpression of the class C gene AG, under the control of the cauliflower mosaic virus 35S promoter, demonstrated that this gene is sufficient to turn off class A function when it is expressed in the outer two floral whorls (Mandel et al., 1992a; Mizukami and Ma, 1992). 35S::AG plants closely resemble ap2 mutants. Combined misexpression of both of the class B genes has shown that AP3 and PI are sufficient to specify petal and stamen identity (Krizek and Meyerowitz, 1996). 35S::PI 35S::AP3 plants produce flowers with petals in whorls one and two and stamens in whorls three and four.

SPT and CRC Promote Carpel Identity Independently of AG Carpelloid features are present on floral organs of ap2-2 ag-1 double mutants and ap2-2 pi-1 ag-1 triple mutants, suggesting that other genes besides AG can specify certain aspects of carpel development. These features are lost when the triple mutant is combined with mutations in SPATULA (SPT) and CRABS CLAW (CRC) (Table 5.1) (Alvarez and Smyth, 1999). Individual mutations in either SPT or CRC show defects in carpel fusion in the upper part of the organ and small effects on carpel morphology. The crc spt double mutants show a much more severe phenotype with further reductions in carpel fusion and decreased amounts of many carpel tissue types including stigma, style, septum, transmitting tract and ovules. CRC is a member of the YABBY gene family that specifies abaxial identity while SPT encodes a bHLH transcription factor (Bowman and Smyth, 1999; Heisler et al., 2001).

Molecular Biology of Floral Organogenesis

105

SEPALLATA Genes Work with ABC Genes to Specify Organ Identity A significant revision of the ABC model was the addition of the class E or SEPALLATA (SEP) genes, which were first identified in tomato and petunia. Decreased expression of the TM5 gene in tomato and the FBP2 and FBP5 genes in petunia resulted in partial homeotic transformations in the second, third and fourth whorls of the flower (Angenent et al., 1994; Pnueli et al., 1994; Ferrario et al., 2003). The specific role of these genes in flower development, however, was not well defined until corresponding genes in Arabidopsis were mutated. Arabidopsis has three SEP genes (SEP1, SEP2 and SEP3) that are functionally redundant (Table 5.1). SEP1 and SEP2 are expressed in all four whorls of the flower (Flanagan and Ma, 1994; Savidge et al., 1995) while SEP3 mRNA is detected in whorls two, three and four (Mandel and Yanofsky, 1998). Mutations in any one or two of these genes do not significantly alter floral organ identity but sep1 sep2 sep3 triple mutants produce indeterminate flowers consisting only of sepals (Pelaz et al., 2000). These flowers resemble those produced in ap3 ag and pi ag double mutants, indicating that the B and C functions are not active. The initial expression patterns of AP3, PI and AG are normal in sep1 sep2 sep3 triple mutants showing that the SEP genes are not required for activation of the B and C class genes (Pelaz et al., 2000). Conversely, the SEP genes do not function downstream of the class B and C genes as they are still expressed in class B and C mutants. Thus the three SEP genes function in combination with the ABC genes to specify petal, stamen and carpel identity. With the addition of the SEP genes to the classical ABC model, sepal identity is conferred by A function alone, petal identity by A þ B þ E, stamen identity by B þ Cþ E and carpel identity by C þ E (Fig. 5.4). Although floral organs are thought to be derived from leaves, misexpression of different combinations of the ABC genes in leaves did not significantly convert these vegetative organs into reproductive organs. Other factors required to specify a floral fate must be missing from leaves. The flower-specific expression of the SEP genes suggested that these genes might be such factors and this was shown to be the case. Misexpression of AP1, AP3, PI and SEP is sufficient to convert rosette leaves into petal-like organs (Honma and Goto, 2001; Pelaz et al., 2001b) while misexpression of AP3, PI, AG and SEP3 is sufficient to convert cauline leaves into stamen-like organs (Honma and Goto, 2001).

MADS-domain Protein Complexes All of the class A, B, C and E genes except for AP2 code for members of the MADS-domain family of dimeric transcription factors. MADS is an acronym for the first four identified members of the family: MCM1 in yeast, AG in Arabidopsis, DEFICIENS (DEF) in Antirrhinum and SRF in humans. These proteins share a 56 amino acid DNA-binding domain called the MADS domain and bind to CArG box [CC(A/T)6GG] sequences (reviewed in Riechmann and

106

B. Krizek

Fig. 5.4. The ABCE model of flower development. SEPALLATA genes (class E genes) act with the class A, B and C genes to specify organ identity. Shown below each whorl are the complexes of MADS-domain proteins that have been proposed to specify organ identity in the quartet model of flower development. Note: Se, sepals; Pe, petals; St, stamens; Ca, carpels.

Meyerowitz, 1997). In plants, there are two classes of MADS-domain proteins based on sequence similarity outside of the MADS domain (Alvarez-Buylla et al., 2000). AP1, AP3, PI, AG, SEP1, SEP2 and SEP3 are all members of the MIKC-type (or type II), which share a characteristic sequence of domains (I, K and C) in addition to the MADS domain. The less conserved I (intervening) region contributes to dimerization specificity. The K (keratin-like) domain is predicted to form a coiled-coil structure and is also involved in dimerization of these proteins. The carboxy-terminal region (C) acts as a transcriptionalactivation domain in some family members. Each whorl of the flower contains a unique combination of MADS-domain organ-identity proteins that act as transcription factors regulating the expression of different sets of target genes. The precise molecular nature of these MADS-domain protein complexes, however, has still not been defined. Early studies showed that although all of the class A, B and C MADS-domain proteins could interact with each other; only certain combinations (AP1 homodimers, AP3/PI heterodimers and AG homodimers in Arabidopsis) could bind DNA in vitro (Riechmann et al., 1996). Heterodimers formed between A and B class proteins and those between B and C class proteins did not bind to DNA, suggesting that unique heterodimers between A, B and C class proteins are not present within each whorl. However, more recent data suggest that unique higher order protein complexes involving A, B, C and E class proteins may form in each of the four whorls. Interactions between the class A proteins AP1 and SEP3 have been detected using the yeast two-hybrid system (Pelaz et al., 2001a) and by immunoprecipitation (Honma and Goto, 2001). Likewise, the class C protein AG can interact with SEP1, SEP2 and SEP3 in the yeast two-hybrid assay (Fan et al., 1997) and similar interactions have been detected with the corresponding

Molecular Biology of Floral Organogenesis

107

proteins in Antirrhinum (Davies et al., 1996). Protein–protein interactions are also detected when AP3 and PI are coexpressed with either SEP3 or AP1 in yeast and when AP3 and PI are coexpressed with both AG and SEP3 in yeast (Honma and Goto, 2001). These results suggest that floral organ-identity proteins may regulate gene expression as higher order multimeric complexes. The quartet model proposes that distinct complexes of four MADS-domain proteins function in each whorl of the flower (Theissen, 2001) (Fig. 5.4). An AP3-PI-AP1-SEP complex is proposed to function in second whorl cells and confer petal identity, an AP3-PI-AG-SEP complex is proposed to function in the third whorl and specify stamen identity and an AG-AG-SEP-SEP complex is proposed to act in the fourth whorl cells to specify carpel identity. These proposed tetramers may correspond to a pair of dimers with each dimer binding to adjacent CArG box sequences in the promoters of target genes. Different protein domains appear to mediate the formation of these complexes. While the I and K regions are crucial for dimerization of MADS-domain proteins (Riechmann et al., 1996), the formation of higher order MADSdomain complexes requires the C domain (Egea-Cortines et al., 1999). The actual stoichiometry and structural nature of these proposed complexes awaits experimental determination. Interactions between different MADS-domain protein dimers could have several consequences for the activities of these proteins. Some MADS-domain proteins such as AP3 and PI do not have transcriptional-activation domains. The ability of an AP3/PI heterodimer to regulate target gene expression may therefore be dependent upon the formation of a complex with MADS-domain proteins possessing transcriptional-activation activity. Both AP1 and SEP3 have C-terminal domains capable of transcriptional activation (Cho et al., 1999; Honma and Goto, 2001; Pelaz et al., 2001a). Thus the activation of petal genes may require interaction between an AP3/PI heterodimer and an AP1/SEP heterodimer. Alternatively, interactions between MADS-domain proteins may alter the DNA-binding properties of the component dimers. Such interactions could increase the overall affinity of dimers for binding to DNA or may lead to cooperative binding of dimers to DNA. In support of this idea, interactions between the Antirrhinum class A protein (SQUAMOSA (SQUA)) and class B proteins (DEF and GLOBOSA (GLO)) were shown to lead to increased in vitro DNA-binding affinity compared to either the DEF/ GLO heterodimer or SQUA homodimer alone (Egea-Cortines et al., 1999).

Regulation of Floral Organ-identity Gene Expression A key advance in the field of flower development has been the uncovering of a direct role for the FM-identity gene LEAFY (LFY ) in activation of the floral organ-identity genes (Table 5.1). LFY encodes a novel plant-specific DNA-binding protein that can bind in vitro to the promoters of several floral organ-identity genes (Parcy et al., 1998; Busch et al., 1999; Lamb et al., 2002). While LFY possesses DNA-binding activity, it does not appear to have intrinsic transcriptional-activation activity. Thus LFY may need to interact with

108

B. Krizek

additional proteins to function in transcriptional regulation. LFY can directly regulate the transcription of at least one representative of each of the class A, B and C genes but uses distinct mechanisms to activate expression of each of these genes in a region-specific manner. LFY alone is sufficient to activate expression of the class A gene AP1 throughout the flower. In contrast, the spatially localized induction of the class B and C genes appears to result from interactions between globally expressed LFY and region-specific factors. Experiments using a constitutively activated form of LFY (LFY containing the transcriptional-activation domain of VP16) (Parcy et al., 1998) as well as promoter dissection studies have made important contributions to our understanding of floral organ-identity gene activation. Expression of each of the floral organ-identity genes continues throughout most of flower development (Yanofsky et al., 1990; Bowman et al., 1991; Drews et al., 1991; Jack et al., 1992; Mandel et al., 1992b; Goto and Meyerowitz, 1994). Since LFY is not expressed after stage 5, other factors presumably act to maintain their expression (Weigel et al., 1992). The plant hormone gibberellin (GA) appears to be an important regulator of homeotic gene expression in later stages of flower development (Yu et al., 2004). The GAdeficient mutant ga1 produces flowers that have organs with the correct identity but which are growth arrested and immature (Goto and Pharis, 1999). Signalling through the GA pathway involves inactivation of a family of DELLA proteins that act as negative regulators of GA responses (reviewed in Sun, 2000). Mutations in these DELLA genes can repress the floral defects of ga1–3, suggesting that GA responses are essential for completion of floral organ development. Expression of AP3, PI and AG is increased in ga1 flowers after GA treatment and decreased in dexamethasone-treated plants expressing a steroid-inducible DELLA protein. Thus GA promotes expression of the class B and C genes and the continued development of floral organs by negatively regulating DELLA proteins.

Regulation of AP1 expression AP1 mRNA is initially detected throughout a young FM and later becomes restricted to sepal primordia and adjacent second whorl cells. Expression of AP1 is delayed in lfy mutants (Liljegren et al., 1999), increased in flowers expressing a constitutively active LFY (LFY::LFY-VP16) (Parcy et al., 1998) and activated ectopically in the young leaves of plants misexpressing LFY (Parcy et al., 1998). These results suggest that LFY is both necessary and sufficient to activate AP1 expression (Fig. 5.5). This activation appears to be direct as LFY can bind a sequence within the AP1 promoter (Parcy et al., 1998) and activation of AP1 by LFY does not require protein synthesis (Wagner et al., 1999). By stage 3 of flower development, AP1 mRNA is only detected in the first and second whorls of the flower. The restriction of AP1 expression requires AG activity as AP1 mRNA is detected in all four whorls of ag mutants (Gustafson-Brown et al., 1994). It is not known whether the repression of AP1 by AG is direct.

Molecular Biology of Floral Organogenesis

109

LFY WUS AP1

UFO

AG AP3

Fig. 5.5. Activation of the ABC class genes by LFY. LFY directly activates AP1. LFY acts in combination with UFO to activate expression of the class B gene AP3. LFY and WUS together activate expression of the class C gene AG.

Initiation of B class gene expression by LFY, UFO and AP1 Both LFY and AP1 act in the initiation of AP3 expression in the second and third whorl cells of a stage 3 flower. Expression of both of the class B genes is somewhat reduced in lfy mutants and more severely reduced in lfy ap1 double mutants (Weigel and Meyerowitz, 1993). Class B gene expression is also reduced in the unusual floral organ (ufo) mutant suggesting that UFO is another positive regulator of B class gene expression (Levin and Meyerowitz, 1995) (Table 5.1). Several pieces of evidence indicate that LFY and UFO work together in activation of the class B genes (Fig. 5.5). UFO is expressed in stage 2 FMs in a pattern suggesting that it could be the region-specific factor that functions with globally expressed LFY (Samach et al., 1999). LFY is not sufficient for AP3 and PI activation as 35S::LFY and LFY::LFYVP16 plants do not activate AP3 and PI in vegetative tissues or throughout flowers (Parcy et al., 1998). On the other hand, ectopic expression of UFO activates AP3 at a slightly earlier time in development and in an expanded domain as compared to wild-type (Parcy et al., 1998). 35S::UFO plants produce extra petals and stamens at the expense of sepals and carpels, similar to the phenotype resulting from misexpression of both of the class B genes. The induction of AP3 expression in 35S::UFO plants is dependent upon LFY activity. Misexpression of both LFY and UFO activates AP3 expression in seedlings suggesting that these two factors are sufficient for AP3 activation (Parcy et al., 1998). The steroid-inducible LFY::GR fusion has been used to show that LFY acts both directly and indirectly to induce AP3 expression. AP3 expression is increased after treatment of 35S::LFY-GR 35S::UFO seedlings with dexamethasone but to a lesser extent after treatment with both cycloheximide and dexamethasone (Lamb et al., 2002). Although an LFY-binding site has been identified within an AP3 promoter element required for early expression, mutation of this site did not affect AP3 expression in vivo (Lamb et al., 2002). It is not clear how LFY and UFO work in combination to activate AP3 expression. UFO is an F-box protein that is a component of SCF (Skp1-cullin-F box) protein complex that acts as an E3 ubiquitin-ligating enzyme (Samach

110

B. Krizek

et al., 1999; Wang et al., 2003). E3 enzymes serve as bridges between E2 ubiquitin-conjugating enzymes and specific protein substrates, catalysing the transfer of ubiquitin from the E2 to a target protein. Such ubiquitinated proteins are typically targeted for proteasome-mediated degradation. SCFUFO can interact with the COP9 signalosome (CSN) and CSN is required for UFO-mediated activation of AP3 expression (Wang et al., 2003). It has been hypothesized that UFO’s role in AP3 activation involves the targeted degradation of a negative regulator of AP3 expression. AP1 also participates in activation of class B gene expression in petals. AP1 is not sufficient for class B gene activation as 35S::AP1 plants do not show transformations in organ identity (Mandel and Yanofsky, 1995). However, an activated AP1 (AP1-VP16) can turn on AP3 and PI expression in the first whorl (Ng and Yanofsky, 2001). Furthermore, AP3 expression in 35S::LFY 35S::UFO seedlings is modulated by mutations in AP1 (Parcy et al., 1998). AP1 can bind in vitro to three CArG box sequences in the AP3 promoter (Hill et al., 1998; Tilly et al., 1998). Maintenance of class B gene expression The maintenance of AP3 and PI expression in flowers requires the activity of both AP3 and PI indicating that these two proteins function in a positive autoregulatory loop. The existence of such a loop was first postulated when it was observed that AP3 expression was not maintained in the third whorl of pi mutants (Jack et al., 1992). PI expression is not maintained in ap3 or pi mutants and the early fourth whorl expression of PI in wild-type flowers is also not maintained in older flowers (Goto and Meyerowitz, 1994). When AP3 is misexpressed in fourth whorl cells, PI expression is maintained in this region (Jack et al., 1994). Because the AP3 promoter but not the PI promoter contains CArG boxes, it has been speculated that the autoregulation of AP3 is direct while that of PI is indirect. This is supported by the binding of AP3/PI heterodimers to CArG boxes within the AP3 promoter (Hill et al., 1998; Tilly et al., 1998) and the requirement of de novo protein synthesis for PI autoregulation (Honma and Goto, 2000). In addition to transcriptional control, AP3 is regulated post-transcriptionally (Jack et al., 1994). This level of regulation also requires PI and could be explained by an instability of the AP3 protein when not found in a complex with PI. Initiation of AG expression by LFY and WUS AG mRNA is first detected in the central region of a stage 3 FM (Yanofsky et al., 1990). In lfy mutants, AG expression is reduced and delayed suggesting that LFY is a positive regulator of AG expression (Weigel and Meyerowitz, 1993). Plants expressing an activated form of LFY exhibit homeotic transformations in organ identity similar to those resulting from AG misexpression (Parcy et al., 1998). AG expression is activated in all four whorls of these LFY::LFY-VP16 flowers as well as in vegetative tissues of 35S::LFY-VP16

Molecular Biology of Floral Organogenesis

111

plants. Because AG is normally expressed in just a subset of LFY-expressing cells, it was proposed that LFY requires a cofactor to activate AG expression in just the third and fourth whorls of a flower. The meristem regulatory protein WUSCHEL (WUS) has been identified as an LFY cofactor (Table 5.1). The wus mutants have defective SAMs that cannot serve as permanent sources of stem cells for organ initiation (Laux et al., 1996). These mutants undergo a stop and go pattern of development in which shoot meristems produce a few leaves and then stop with subsequent adventitious meristems producing a few leaves and then stopping. During reproductive development, wus mutants produce flowers that largely lack third and fourth whorl organs. WUS encodes a homeodomain protein that is expressed in the central region of shoot meristems and young FMs (Mayer et al., 1998). That WUS might be an LFY cofactor was suggested by the 35S::AG-like phenotype of plants misexpressing WUS under the control of the LFY or AP3 promoter (Lenhard et al., 2001; Lohmann et al., 2001). These phenotypes are dependent upon LFY activity (Lenhard et al., 2001), lending support to a model in which LFY and WUS work in combination to activate AG specifically in the inner floral whorls (Fig. 5.5). Both LFY and WUS bind AG regulatory sequences in vitro (Busch et al., 1999; Lohmann et al., 2001). Two LFY and two WUS binding sites are found in close proximity within the AG second intron. All elements required for proper spatial and temporal regulation of AG expression are contained within this intron (Sieburth and Meyerowitz, 1997; Busch et al., 1999; Deyholos and Sieburth, 2000). Although the LFY- and WUS-binding sites are physically close, LFY and WUS appear to bind independently to their respective sites. No physical interaction or cooperation of binding was observed in gel shift experiments (Lohmann et al., 2001). Mutations that disrupt binding by either LFY or WUS to these sites in vitro result in significantly reduced in vivo expression of reporter genes that are under the control of these elements (Busch et al., 1999; Lohmann et al., 2001). Coexpression of LFY and WUS in yeast stimulated the expression of a reporter gene under the control of AG regulatory sequence (Lohmann et al., 2001). This transcriptional activity required binding of both factors to the DNA. After being activated by WUS, AG in turn downregulates the expression of WUS in the centre of the FM (Lenhard et al., 2001; Lohmann et al., 2001). WUS expression is at its highest levels during stages 2 and 3 of flower development and is not expressed after stage 6 of flower development when all FM cells have been used for organ initiation. In indeterminate ag flowers, WUS expression is maintained in the central region of the FM. It is not known whether AG regulates WUS expression directly or indirectly.

Transcriptional repression of AG The lack of AG expression in vegetative tissues and the outer two floral whorls is a consequence of both the absence of AG activators and the activity of AG repressors. Repressors of AG expression include APETALA2 (AP2), LEUNIG

112

B. Krizek

(LUG), SEUSS (SEU), STERILE APETALA (SAP), CURLY LEAF (CLF), INCURVATA2 (ICU2) and BELLRINGER (BLR) (Table 5.1; Fig. 5.6) (Drews et al., 1991; Jofuku et al., 1994; Liu and Meyerowitz, 1995; Goodrich et al., 1997; Byzova et al., 1999; Serrano-Cartagena et al., 2000; Franks et al., 2002; Bao et al., 2004). Mutations in any of these genes result in ectopic AG expression in the first and second whorl cells leading to partial or complete homeotic transformations of first whorl sepals into carpels and second whorl petals into stamens or the loss of second whorl organs. These factors also function in AG repression in other tissues and at other times in development. AP2, LUG and SEU act to prevent precocious expression of AG in stage 2 FMs. SAP acts to repress AG expression in inflorescence stems while BLR acts to repress AG in the inflorescence meristem (IM) of older shoots. CLF and ICU2 act to prevent AG expression in leaves. Another factor likely to be involved in AG regulation is AINTEGUMENTA (ANT). Homeotic changes in organ identity are not usually observed in ant mutants. However, AG is expressed precociously in stage 2 ant-9 mutants (Liu et al., 2000) and enhanced AG misexpression is observed in second whorl organs of ap2-1 ant-6 mutants compared to weak ap2-1 single mutants (Krizek et al., 2000). Several of the known AG repressors encode proteins that are likely to play roles in transcriptional repression including DNA-binding proteins, transcriptional corepressors and chromatin-modifying factors. AP2 and ANT are members of the AP2/ERF family of plant-specific transcription factors (Jofuku et al., 1994; Elliott et al., 1996; Klucher et al., 1996). ANT has been shown to have both DNA-binding (Nole-Wilson and Krizek, 2000) and transcriptionalactivation functions (Vergani et al., 1997; Krizek, 2003). Other potential transcription factors that may bind directly to AG regulatory sequence are BLR, a homeodomain protein of the BELL1 class (Byrne et al., 2003; Roeder et al., 2003; Smith and Hake, 2003) and SAP, a novel protein with similarity to transcriptional regulators (Byzova et al., 1999). LUG and SEU may function together as a transcriptional corepressor complex. LUG has sequence similarity with Tup1 in yeast and has recently been demonstrated to possess

Fig. 5.6. Mechanisms regulating AP2 and AG activity. AG is negatively regulated by a number of factors including AP2, LUG, SEU, SAP, ANT, BLR, CLF and ICU2. These genes act to prevent AG from being transcribed in the outer two whorls of a flower. HUA1, HUA2, HEN2 and HEN4 promote AG activity by participating in the processing of AG RNA. AP2 activity is restricted to the outer two floral whorls by the presence of miRNA172 in the inner whorls.

Molecular Biology of Floral Organogenesis

113

transcriptional-repression activity in both yeast and plants (Conner and Liu, 2000; Sridhar et al., 2004). This activity was inhibited by incubation with trichostatin A suggesting that AG repression involves histone deacetylation (Sridhar et al., 2004). The SEU protein contains two glutamine-rich regions and a putative dimerization domain similar to that found in the Ldb family of transcriptional coregulators (Franks et al., 2002). SEU does not have transcriptional-repression activity on its own but can interact with LUG and may function as an adapter protein, recruiting LUG to AG regulatory sequences by interacting with DNA-bound transcription factors (Sridhar et al., 2004). Currently, no physical interactions between LUG (or SEU) and any of the above-mentioned transcription factors have been described, although it seems likely that multi-protein transcriptional-repression complexes are formed at AG regulatory sequences. The specific subunit composition of these complexes may vary with tissue type and developmental time. An additional AG regulator, CLF, is likely to function in modification of chromatin structure as it encodes a protein with homology to Polycomb-group proteins like Enhancer of Zeste (Goodrich et al., 1997).

Post-transcriptional regulation of AG In addition to transcriptional regulation, AG is regulated post-transcriptionally. Two different genetic screens have identified several factors that act in the processing of AG pre-mRNA. The first was an enhancer screen utilizing the partial loss-of-function ag-4 allele. The ag-4 flowers have a normal third whorl but the fourth whorl carpel is replaced by a new flower repeating the pattern of organs: sepals, petals and stamens. Two partially redundant genes, HUA1 and HUA2 (hua is Chinese for flower), were identified that when mutated in combination with ag-4 cause a strong ag-like phenotype (Chen and Meyerowitz, 1999). Mutations in either HUA1 or HUA2 alone do not have any phenotype while hua1 hua2 double mutants have slight defects in stamen and carpel development. The hua1 hua2 double mutant was the basis for a second enhancer screen that identified several hua enhancers (hen) mutants, which in combination with hua1 hua2 cause a more severe ag-like phenotype. Subsequent cloning and characterization of the HUA and HEN genes identified in these screens is starting to shed light on their roles in AG regulation. Four of these genes (HUA1, HUA2, HEN2 and HEN4) appear to play roles in maturation of AG pre-mRNA (Table 5.1; Fig. 5.6). HUA2 is a novel protein with low overall similarity to transcriptional coactivators (Chen and Meyerowitz, 1999). In addition, HUA2 contains an RPR (regulation of premRNA processing) domain that is sometimes found in proteins involved in RNA metabolism. HEN2 is a putative nuclear RNA helicase homologous to yeast Dob1p, which can act in RNA degradation (Western et al., 2002). HUA1 encodes a CCCH zinc finger RNA-binding protein while HEN4 encodes a K homology (KH) domain-containing putative RNA-binding protein ( Li et al., 2001; Cheng et al., 2003). The two RNA-binding proteins HEN4 and HUA1 associate together in nuclear speckles. Several RNA bands of larger

114

B. Krizek

size than the fully processed AG mRNA were present at increased levels in single, double and triple mutant combinations involving hua1, hua2 and hen4 as well as in hua1 hua2 hen2 triples. These larger transcripts correspond to RNAs that include the first two exons of AG as well as part of the second intron but lack exons three to seven and are polyadenylated. It has been proposed that HUA1, HUA2 and HEN4 proteins act by inhibiting the use of cryptic polyadenylation sites within the AG second intron or by promoting proper splicing of AG pre-mRNA. Because HUA1 can bind to AG RNA in vitro, it is likely that HUA1 and HEN4 bind directly to AG pre-mRNA to carry out their function.

Translational repression of AP2 The only floral organ-identity gene that does not encode a MADS-domain protein is the class A gene AP2. Although AP2 is expressed at the mRNA level in all four whorls of developing flowers, its activity is restricted to the outer two whorls of the flower. The mechanism behind this spatial restriction of AP2 activity is now becoming clearer. AP2 is regulated post-transcriptionally by the activity of a microRNA (miRNA). miRNAs are non-coding RNAs of 21–22 nucleotides that are processed from longer hairpin transcripts and are thought to play important roles in development (reviewed in Mallory and Vaucheret, 2004). The miRNAs may regulate target gene expression by binding to regions of complementarity in the target transcripts and causing RNA cleavage and/or translational inhibition. A clue that miRNAs might be involved in regulation of floral homeotic genes came from the isolation and characterization of an enhancer of the hua1 hua2 double mutant called HEN1. The single mutants hen1 have a pleiotropic phenotype with defects in organ size, leaf shape, number of axillary inflorescences and fertility (Chen et al., 2002). A role for HEN1 in miRNA metabolism was suggested by the similarity of the hen1 phenotype with carpel factory (caf ) mutants (Table 5.1). CAF encodes the Arabidopsis homologue of Dicer, a protein with RNase III activity that is involved in processing of miRNA precursors. The mutants caf and hen1 have reduced levels of a number of miRNAs (Park et al., 2002). One of the miRNAs (miRNA172) affected in hen1 has significant sequence homology to AP2 and three related genes, matching the AP2 sequence in 19 of the 21 positions. Although AP2 RNA is present in similar amounts in wild-type and hen1 mutants, AP2 protein is present at higher levels in hen1 mutants than wild type. The reduced amount of miRNA172 and increased levels of AP2 protein in hen1 suggested that AP2 might be negatively regulated by miRNA172. Experiments in which miRNA172 was overexpressed under the control of the 35S promoter confirm this hypothesis. 35S::miRNA172 plants produce flowers with phenotypes similar to ap2 mutants (Aukerman and Sakai, 2003; Chen, 2004). While normal levels of AP2 mRNA were present in these overexpression lines, AP2 protein was not detectable. A possible model for AP2 regulation by miRNA172 can be envisioned in which expression of miRNA172 in the inner two floral whorls causes translational inhibition

Molecular Biology of Floral Organogenesis

115

of AP2 mRNA in these whorls (Fig. 5.6). In this way, AP2 protein (and thus activity) could be restricted to the outer two floral whorls. This model is supported by the higher levels of miRNA172 in the inner two whorls of the flower (Chen, 2004); however, the spatial pattern of expression of AP2 protein in inflorescences has not yet been determined.

Downstream Targets of Floral Organ-identity Genes Although we have a good understanding of how floral organ identity is specified by the ABC class genes and how the ABC class genes are regulated, we know far less about the events occurring downstream of these factors. Few direct targets of the ABC class transcription factors have been identified. Several types of transcriptional profiling techniques such as differential screening, subtractive hybridization, differential display and expression microarrays have been performed using transgenic and mutant flowers with altered organ identity (Nacken et al., 1991; Rubinelli et al., 1998; Hu et al., 2003; Zik and Irish, 2003; Hennig et al., 2004; Wellmer et al., 2004). Such studies have identified a number of genes expressed in flowers that are likely to play important developmental roles. Some of these genes show organ-specific expression patterns and may be direct targets of the floral organ-identity proteins. In most cases, however, such evidence is still lacking.

Genes regulated by the B function One target of the class B genes was identified via a combined use of differential display and a steroid-regulated AP3 protein (Sablowski and Meyerowitz, 1998). To identify genes directly regulated by AP3 and PI that function in petal organogenesis, a genetic background of ap3-3 ag-3 35S::PI 35S::AP3-GR was used. In the absence of steroid these plants produce sepals in all four whorls because they do not have active class B and C functions. After steroid treatment, these plants produce petals in all four whorls. Inflorescences of these plants were treated with steroid and cycloheximide or cycloheximide alone and RNA populations from both treatments were compared using differential display. NAP (NAC-like, activated by AP3/PI), a gene with homology to petunia NO APICAL MERISTEM (NAM) and Arabidopsis CUP SHAPED COTYLEDON (CUC) genes was identified as being positively regulated by AP3 and PI (Table 5.1). NAP is expressed in both petals and stamens and has been hypothesized to function in these organs during the developmental transition from growth due to cell division versus growth due to cell expansion. A global analysis of genes induced by AP3 and PI was carried out by Zik and Irish (2003). By comparing gene expression data among mutants that don’t make petals and stamens (ap3-3 and pi-1), those that only make petals (steroid treated ap3-3 ag-3 35S::PI 35S::AP3-GR), those that don’t make stamens (D6::DTA) and wild-type, 47 genes likely to be regulated by AP3/PI in either petal or stamen development were identified. Some of these genes are

116

B. Krizek

expressed in both petals and stamens while others are petal or stamen specific. Since the array used in this study represented approximately 25% of the Arabidopsis genome, the authors estimate that approximately 200 genes may be regulated by AP3/PI either directly or indirectly. Very few transcription factors were identified in their experiment suggesting that AP3 and PI might directly regulate most of the genes needed to make petals and stamens. However, another global expression experiment using mutants in each of the class ABC genes identified a much larger number of genes (including a large number of transcription factors) whose expression is affected by activity of the floral organ-identity proteins (Wellmer et al., 2004).

Targets of AG regulation A putative direct target of AG in the fourth whorl is SHATTERPROOF2 (SHP2). SHP1 and SHP2 are closely related and redundant MADS-domain proteins that are expressed in thin stripes at the boundaries between carpel valves and replums (Flanagan and Ma, 1994; Savidge et al., 1995). These valve margin cells develop into the dehiscence zone where cell separation events lead to detachment of the valve from the replum and subsequent pod shatter. shp1 shp2 double mutants produce indehiscent fruit that do not shatter to release seeds (Liljegren et al., 2000). SHP2 has been proposed to be a direct target of AG. There are several CArG box sites in the SHP2 promoter to which AG binds in vitro and ectopic expression of AG can activate a SHP2:: b-glucuronidase (GUS) reporter gene in cauline leaves (Savidge et al., 1995). Microarray experiments using 35S::AG-GR ag-1 plants have shown that SPOROCYTELESS (SPL)/NOZZLE (NZZ) is a direct target of the class C protein AG (Ito et al., 2004). 35S::AG-GR ag-1 plants given a single treatment with dexamethansone produce locules containing pollen grains on the margins of the third whorl petaloid organs. SPL/NZZ is known to be required for microsporogenesis as well as playing roles in ovule patterning, megasporogenesis and anther wall development (Table 5.1) (Schiefthaler et al., 1999; Yang et al., 1999; Balasubramanian and Schneitz, 2000, 2002). The activation of SPL/NZZ in the steroid-activated 35S::AG-GR ag-1 plants is independent of cycloheximide suggesting that AG directly regulates the expression of this gene. AG binds to a site within the 3’ region of SPL/NZZ that nearly matches a CArG box consensus. Mutations in this site that destroy AG binding in vitro also result in reduced levels and a smaller domain of SPL/NZZ expression in vivo. Based on a comparison of the expression patterns of AG and SPL, it was proposed that AG is responsible for the initiation but not the maintenance of SPL expression. SPL encodes a protein that may function as a transcription factor and ectopic expression of SPL in an ag-1 background is sufficient to induce microsporogenesis. This suggests that additional transcriptional regulators turned on by AG mediate various aspects of AG function in stamen and carpel organogenesis.

Molecular Biology of Floral Organogenesis

117

Summary The last 15 years have been an exciting time for plant biologists studying floral organogenesis. The ABC model for flower development was proposed approximately 13 years ago. Since that time a new class of genes (SEP) have been added to the model and a new level of understanding of class A, B and C activity at the protein level has been achieved. Investigations into the regulation of the floral organ-identity genes have revealed a large number of factors and complex interactions that regulate these genes at both the transcriptional and post-transcriptional levels. Despite progress in many areas, several aspects of floral organogenesis remain relatively uncharacterized. We know little about very early patterning events that establish the sites of floral organ initiation. What genes establish this pre-pattern? How are spatial and temporal patterns of cell division controlled within these meristematic cells such that discrete organ primordia are initiated at defined positions? New technologies that can monitor cell division and growth in real time have been described recently (Grandjean et al., 2004; Reddy et al., 2004). These methods will prove extremely useful in defining the alterations in cell behaviour that occur in mutants with altered patterns of floral organ initiation. Another continuing challenge is to identify components of the regulatory cascades initiated by the floral organ-identity proteins that lead to the elaboration and final form of each floral organ. The few direct targets of the ABC class proteins that have been identified up to now function at relatively late stages in floral organ development. This suggests that the floral organ-identity proteins function throughout floral organogenesis, first to specify organ identity and then later to control specific aspects of organogenesis such as regional growth or differentiation of particular cell types. Several new techniques are likely to contribute significantly to progress in this area as well. The continued use of DNA microarrays especially when used in combination with new technologies such as laser capture microdissection (LCM), which allow gene profiling within specific cell types rather than whole organs, should be particularly useful (Schnable et al., 2004). Direct targets of transcription factors can also be identified using chromatin immunoprecipitation. Finally, more in depth characterizations of genes already identified by microarray and differential screening methods will help to provide a clearer picture of how each floral organ type acquires its unique morphology and pattern of cell and tissue types.

Acknowledgement I acknowledge support for my work from the United States Department of Energy.

118

B. Krizek

References Alvarez, J. and Smyth, D.R. (1999) CRABS CLAW and SPATULA, two Arabidopsis genes that control carpel development in parallel with AGAMOUS. Development 126, 2377– 2386. Alvarez-Buylla, E.R., Pelaz, S., Liljegren, S.J., Gold, S.E., Burgeff, C., Ditta, G.S., Ribas de Pouplana, L., Martinez-Castilla, L. and Yanofsky, M.F. (2000) An ancestral MADS-box gene duplication occurred before the divergence of plants and animals. Proceedings of the National Academy of Sciences USA 97, 5328–5333. Angenent, G.C., Franken, J., Busscher, M., Weiss, D. and van Tunen, A.J. (1994) Co-suppression of the petunia homeotic gene fbp2 affects the identity of the generative meristem. Plant Journal 5, 33–44. Aukerman, M.J. and Sakai, H. (2003) Regulation of flowering time and floral organ identity by a microRNA and its APETALA2-like target genes. Plant Cell 15, 2730–2741. Balasubramanian, S. and Schneitz, K. (2000) NOZZLE regulates proximaldistal pattern formation, cell proliferation and early sporogenesis during ovule development in Arabidopsis thaliana. Development 127, 4227– 4238. Balasubramanian, S. and Schneitz, K. (2002) NOZZLE links proximal-distal and adaxial-abaxial pattern formation during ovule development in Arabidopsis thaliana. Development 129, 4291–4300. Bao, X., Franks, R.G., Levin, J.Z. and Liu, Z. (2004) Repression of AGAMOUS by BELLRINGER in floral and inflorescence meristems. Plant Cell 16, 1478–1489. Bowman, J.L. and Smyth, D.R. (1999) CRABS CLAW, a gene that regulates carpel and nectary development in Arabidopsis, encodes a novel protein

with zinc finger and helix-loop-helix domains. Development 126, 2387– 2396. Bowman, J.L., Smyth, D.R. and Meyerowitz, E.M. (1991) Genetic interactions among floral homeotic genes of Arabidopsis. Development 112, 1–20. Busch, M.A., Bomblies, K. and Weigel, D. (1999) Activation of a floral homeotic gene in Arabidopsis. Science 285, 585–587. Byrne, M.E., Groover, A.T., Fontana, J.R. and Martienssen, R.A. (2003) Phyllotactic pattern and stem cell fate are determined by the Arabidopsis homeobox gene BELLRINGER. Development 3941–3950. Byzova, M.V., Franken, J., Aarts, M.G.M., de Almeida-Engler, J., Engler, G., Mariana, C., Van Lookeren Campagne, M.M. and Angenent, G.C. (1999) Arabidopsis STERILE APETALA, a multifunctional gene regulating inflorescence, flower, and ovule development. Genes & Development 13, 1002–1014. Carpenter, R. and Coen, E.S. (1990) Floral homeotic mutations produced by transposon-mutagenesis in Antirrhinum majus. Genes & Development 4, 1483–1493. Chen, X. (2004) A microRNA as a translational repressor of APETALA2 in Arabidopsis flower development. Science 303, 2022–2025. Chen, X. and Meyerowitz, E.M. (1999) HUA1 and HUA2 are two members of the floral homeotic AGAMOUS pathway. Molecular Cell 3, 349–360. Chen, X., Liu, J., Cheng, Y. and Jia, D. (2002) HEN1 functions pleiotropically in Arabidopsis development and acts in C function in the flower. Development 129, 1085–1094. Cheng, Y., Kato, N., Wang, W., Li, J. and Chen, X. (2003) Two RNA binding proteins, HEN4 and HUA1, act in the processing of AGAMOUS pre-mRNA

Molecular Biology of Floral Organogenesis in Arabidopsis thaliana. Developmental Cell 4, 53–66. Cho, S., Jang, S., Chae, S., Chung, K.M., Moon, Y.-H., An, G. and Jang, S.K. (1999) Analysis of the C-terminal region of Arabidopsis thaliana APETALA1 as a transcriptional activation domain. Plant Molecular Biology 40, 419–429. Conner, J. and Liu, Z. (2000) LEUNIG, a putative transcriptional corepressor that regulates AGAMOUS expression during flower development. Proceedings of the National Academy of Sciences USA 97, 12902–12907. Davies, B., Egea-Cortines, M., de Andrade Silva, E., Saedler, H. and Sommer, H. (1996) Multiple interactions amongst floral homeotic MADS box proteins. EMBO Journal 15, 4330–4343. Deyholos, M.K. and Sieburth, L.E. (2000) Separable whorl-specific expression and negative regulation by enhancer elements within the AGAMOUS second intron. Plant Cell 12, 1799– 1810. Drews, G.N., Bowman, J.L. and Meyerowitz, E.M. (1991) Negative regulation of the Arabidopsis homeotic gene AGAMOUS by the APETALA2 product. Cell 65, 991–1002. Egea-Cortines, M., Saedler, H. and Sommer, H. (1999) Ternary complex formation between the MADS-box proteins SQUAMOSA, DEFICIENS and GLOBOSA is involved in the control of floral architecture in Antirrhinum majus. EMBO Journal 18, 5370–5379. Elliott, R.C., Betzner, A.S., Huttner, E., Oakes, M.P., Tucker, W.Q.J., Gerentes, D., Perez, P. and Smyth, D.R. (1996) AINTEGUMENTA, an APETALA2-like gene of Arabidopsis with pleiotropic roles in ovule development and floral organ growth. Plant Cell 8, 155–168. Fan, H.-Y., Hu, Y., Tudor, M. and Ma, H. (1997) Specific interactions between the K domains of AG and AGLs, mem-

119 bers of the MADS domain family of DNA binding proteins. Plant Journal 12, 999–1010. Ferrario, S., Immink, R.G.H., Shchennikova, A., Busscher-Lange, J. and Angenent, G.C. (2003) The MADS box gene FBP2 is required for SEPALLATA function in petunia. Plant Cell 15, 914–925. Flanagan, C.A. and Ma, H. (1994) Spatially and temporally regulated expression of the MADS box gene AGL2 in wild-type and mutant Arabidopsis flowers. Plant Molecular Biology 26, 581–595. Franks, R.G., Wang, C., Levin, J.Z. and Liu, Z. (2002) SEUSS, a member of a novel family of plant regulatory proteins, represses floral homeotic gene expression with LEUNIG. Development 129, 253–263. Goodrich, J., Puangsomlee, P., Martin, M., Long, D., Meyerowitz, E.M. and Coupland, G. (1997) A Polycombgroup gene regulates homeotic gene expression in Arabidopsis. Nature 386, 44–51. Goto, K. and Meyerowitz, E.M. (1994) Function and regulation of the Arabidopsis floral homeotic gene PISTILLATA. Genes & Development 8, 1548–1560. Goto, N. and Pharis, R.P. (1999) Role of gibberellins in the development of floral organs of the gibberellin-deficient mutant, ga1-1, of Arabidopsis thaliana. Canadian Journal of Botany 77, 944–954. Grandjean, O., Vernoux, T., Laufs, P., Belcram, K., Mizukami, Y. and Traas, J. (2004) In vivo analysis of cell division, cell growth, and differentiation at the shoot apical meristem in Arabidopsis. Plant Cell 16, 74–87. Gustafson-Brown, C., Savidge, B. and Yanofsky, M.F. (1994) Regulation of the Arabidopsis floral homeotic gene APETALA1. Cell 76, 131–143. Heisler, M.G.B., Atkinson, A., Bylstra, Y.H., Walsh, R. and Smyth, D.R. (2001) SPATULA, a gene that

120

B. Krizek controls development of carpel margin tissues in Arabidopsis, encodes a bHLH protein. Development 128, 1089–1098. Hennig, L., Gruissem, W., Grossniklaus, U. and Kohler, C. (2004) Transcriptional programs of early reproductive stages in Arabidopsis. Plant Physiology 135, 1765–1775. Hill, T.A., Day, C.D., Zondlo, S.C., Thackeray, A.G. and Irish, V.F. (1998) Discrete spatial and temporal cis-acting elements regulate transcription of the Arabidopsis floral homeotic gene APETALA3. Development 125, 1711–1721. Honma, T. and Goto, K. (2000) The Arabidopsis floral homeotic gene PISTILLATA is regulated by discrete ciselements responsive to induction and maintenance signals. Development 127, 2021–2030. Honma, T. and Goto, K. (2001) Complexes of MADS-box proteins are sufficient to convert leaves into floral organs. Nature 409, 525–529. Hu, W., Wang, Y., Bowers, C. and Ma, H. (2003) Isolation, sequence analysis, and expression studies of florally expressed cDNAs in Arabidopsis. Plant Molecular Biology 53, 545–563. Ito, T., Wellmer, F., Yu, H., Das, P., Ito, N., Alves-Ferreira, M., Riechmann, J.L. and Meyerowitz, E.M. (2004) The homeotic protein AGAMOUS controls microsporogenesis by regulation of SPOROCYTELESS. Nature 430, 356–360. Jack, T., Brockman, L.L. and Meyerowitz, E.M. (1992) The homeotic gene APETALA3 of Arabidopsis thaliana encodes a MADS box and is expressed in petals and stamens. Cell 68, 683–687. Jack, T., Fox, G.L. and Meyerowitz, E.M. (1994) Arabidopsis homeotic gene APETALA3 ectopic expression: transcriptional and posttranscriptional regulation determine floral organ identity. Cell 76, 703–716.

Jofuku, K.D., den Boer, B.G.W., Van Montagu, M. and Okamuro, J.K. (1994) Control of Arabidopsis flower and seed development by the homeotic gene APETALA2. Plant Cell 6, 1211–1225. Klucher, K.M., Chow, H., Reiser, L. and Fischer, R.L. (1996) The AINTEGUMENTA gene of Arabidopsis required for ovule and female gametophyte development is related to the floral homeotic gene APETALA2. Plant Cell 8, 137–153. Krizek, B.A. (2003) AINTEGUMENTA utilizes a mode of DNA recognition distinct from that used by proteins containing a single AP2 domain. Nucleic Acids Research 31, 1859– 1868. Krizek, B.A. and Meyerowitz, E.M. (1996) The Arabidopsis genes APETALA3 and PISTILLATA are sufficient to specify the B class organ identity function. Development 122, 11–22. Krizek, B.A., Prost, V. and Macias, A. (2000) AINTEGUMENTA promotes petal identity and acts as a negative regulator of AGAMOUS. Plant Cell 12, 1357–1366. Lamb, R.S., Hill, T.A., Tan, Q.K.-G. and Irish, V.F. (2002) Regulation of APETALA3 floral homeotic gene expression by meristem identity genes. Development 129, 2079–2086. Laux, T., Mayer, K.F.X., Berger, J. and Jurgens, G. (1996) The WUSCHEL gene is required for shoot and floral meristem integrity in Arabidopsis. Development 122, 87–96. Lenhard, M., Bohnert, A., Jurgens, G. and Laux, T. (2001) Termination of stem cell maintenance in Arabidopsis floral meristems by interactions between WUSCHEL and AGAMOUS. Cell 105, 805–814. Levin, J.Z. and Meyerowitz, E.M. (1995) UFO: an Arabidopsis gene involved in both floral meristem and floral organ development. Plant Cell 7, 529–548.

Molecular Biology of Floral Organogenesis Li, J., Jia, D. and Chen, X. (2001) HUA1, a regulator of stamen and carpel identities in Arabidopsis, codes for a nuclear RNA binding protein. Plant Cell 13, 2269–2281. Liljegren, S.J., Gustafson-Brown, C., Pinyopich, A., Ditta, G.S. and Yanofsky, M.F. (1999) Interactions among APETALA1, LEAFY, and TERMINAL FLOWER1 specify meristem fate. Plant Cell 11, 1007–1018. Liljegren, S.J., Ditta, G.S., Eshed, Y., Savidge, B., Bowman, J. and Yanofsky, M.F. (2000) SHATTERPROOF MADS-box genes control seed dispersal in Arabidopsis. Nature 404, 766–770. Liu, Z. and Meyerowitz, E.M. (1995) LEUNIG regulates AGAMOUS expression in Arabidopsis flowers. Development 121, 975–991. Liu, Z., Franks, R.G. and Klink, V.P. (2000) Regulation of gynoecium marginal tissue formation by LEUNIG and AINTEGUMENTA. Plant Cell 12, 1879–1891. Lohmann, J.U., Hong, R.L., Hobe, M., Busch, M.A., Parcy, F., Simon, R. and Weigel, D. (2001) A molecular link between stem cell regulation and floral patterning in Arabidopsis. Cell 105, 793–803. Mallory, A.C. and Vaucheret, H. (2004) MicroRNAs: something important between the genes. Current Opinion in Plant Biology 7, 120–125. Mandel, M.A. and Yanofsky, M.Y. (1995) A gene triggering flower formation in Arabidopsis. Nature 377, 522– 524. Mandel, M.A. and Yanofsky, M.F. (1998) The Arabidopsis AGL9 MADS box gene is expressed in young flower primordia. Sexual Plant Reproduction 11, 22–28. Mandel, M.A., Bowman, J.L., Kempin, S.A., Ma, H., Meyerowitz, E.M. and Yanofsky, M.F. (1992a) Manipulation of flower structure in transgenic tobacco. Cell 71, 133–143.

121 Mandel, M.A., Gustafson-Brown, C., Savidge, B. and Yanofsky, M.F. (1992b) Molecular characterization of the Arabidopsis floral homeotic gene APETALA1. Nature 360, 273–277. Mayer, K.F.X., Schoof, H., Haecker, A., Lenhard, M., Jurgens, G. and Laux, T. (1998) Role of WUSCHEL in regulating stem cell fate in the Arabidopsis shoot meristem. Cell 95, 805–815. Meyerowitz, E.M., Bowman, J.L., Brockman, L.L., Drews, G.N., Jack, T.J., Sieburth, L. and Weigel, D. (1991) A genetic and molecular model for flower development in Arabidopsis thaliana. Development Supplement 1, 157– 167. Mizukami, Y. and Ma, H. (1992) Ectopic expression of the floral homeotic gene AGAMOUS in transgenic Arabidopsis plants alters floral organ identity. Cell, 119–131. Nacken, W.K.F., Huijser, P., Beltran, J.-P., Saedler, H. and Sommer, H. (1991) Molecular characterization of two stamen-specific genes, tap1 and fil1, that are expressed in the wildtype, but not in the deficiens mutant of Antirrhinum majus. Molecular and General Genetics 229, 129–136. Ng, M. and Yanofsky, M.F. (2001) Activation of the Arabidopsis B class homeotic genes by APETALA1. Plant Cell 13, 739–753. Nole-Wilson, S. and Krizek, B.A. (2000) DNA binding properties of the Arabidopsis floral development protein AINTEGUMENTA. Nucleic Acids Research 28, 4076–4082. Parcy, F., Nilsson, O., Busch, M.A., Lee, I. and Weigel, D. (1998) A genetic framework for floral patterning. Nature 395, 561–566. Park, W., Li, J., Song, R., Messing, J. and Chen, X. (2002) CARPEL FACTORY, a dicer homolog, and HEN1, a novel protein, act in microRNA metabolism in Arabidopsis thaliana. Current Biology 12, 1484–1495.

122

B. Krizek Pelaz, S., Ditta, G.S., Baumann, E., Wisman, E. and Yanofsky, M.F. (2000) B and C floral organ identity functions require SEPALLATA MADS-box genes. Nature 405, 200–203. Pelaz, S., Gustafson-Brown, C., Kohalmi, S.E., Crosby, W.L. and Yanofsky, M.F. (2001a) APETALA1 and SEPALLALTA3 interact to promote flower development. Plant Journal 26, 385–394. Pelaz, S., Tapia-Lopez, R., AlvarezBuylla, E.R. and Yanofsky, M.F. (2001b) Conversion of leaves into petals in Arabidopsis. Current Biology 11, 182–184. Pnueli, L., Hareven, D., Broday, L., Hurwitz, C. and Lifschitz, E. (1994) The TM5 MADS box gene mediates organ differentiation in the three inner whorls of tomato flowers. Plant Cell 6, 175– 186. Reddy, G.V., Heisler, M.G., Ehrhardt, D.W. and Meyerowitz, E.M. (2004) Real-time lineage analysis reveals oriented cell divisions associated with morphogenesis at the shoot apex of Arabidopsis thaliana. Development 131, 4225–4237. Riechmann, J.L. and Meyerowitz, E.M. (1997) MADS domain proteins in plant development. Biological Chemistry 378, 1079–1101. Riechmann, J.L., Wang, M. and Meyerowitz, E.M. (1996) DNA-binding properties of Arabidopsis MADS domain homeotic proteins APETALA1, APETALA3, PISTILLATA, and AGAMOUS. Nucleic Acids Research 24, 3134–3141. Roeder, A.H., Ferrandiz, C. and Yanofsky, M.F. (2003) The role of the REPLUMLESS homeodomain protein in patterning the Arabidopsis fruit. Current Biology 13, 1630–1635. Rubinelli, P., Hu, Y. and Ma, H. (1998) Identification, sequence analysis and expression studies of novel anther-specific genes of Arabidopsis thaliana. Plant Molecular Biology 37, 607– 619.

Sablowski, R.W.M. and Meyerowitz, E.M. (1998) A homolog of NO APICAL MERISTEM is an immediate target of the floral homeotic genes APETALA3/PISTILLATA. Cell 92, 93– 103. Samach, A., Klenz, J.E., Kohalmi, S.E., Risseeuw, E., Haughn, G.W. and Crosby, W.L. (1999) The UNUSUAL FLORAL ORGANS gene of Arabidopsis thaliana is an F-box protein required for normal patterning and growth in the floral meristem. Plant Journal 20, 433–445. Savidge, B., Rounsley, S.D. and Yanofsky, M.F. (1995) Temporal relationship between the transcription of two Arabidopsis MADS box genes and the floral organ identity genes. Plant Cell 7, 721–733. Schiefthaler, U., Balasubramanian, S., Sieber, P., Chevalier, D., Wisman, E. and Schneitz, K. (1999) Molecular analysis of NOZZLE, a gene involved in pattern formation and early sporogenesis during sex organ development in Arabidopsis thaliana. Proceedings of the National Academy of Sciences USA 96, 11664–11669. Schnable, P.S., Hochholdinger, F. and Nakazono, M. (2004) Global expression profiling applied to plant development. Current Opinion in Plant Biology 7, 50–56. Schwarz-Sommer, Z., Huijser, P., Nacken, W., Saedler, H. and Sommer, H. (1990) Genetic control of flower development: homeotic genes of Antirrhinum majus. Science 250, 931–936. Serrano-Cartagena, J., Candela, H., Robles, P., Ponce, M.R., Perez-Perez, J.M., Piqueras, P. and Micol, J.L. (2000) Genetic analysis of incurvata mutants reveals three independent genetic operations at work in Arabidopsis leaf morphogenesis. Genetics 156, 1363–1377. Sieburth, L.E. and Meyerowitz, E.M. (1997) Molecular dissection of the AGAMOUS control region shows that cis

Molecular Biology of Floral Organogenesis elements for spatial regulation are located intragenically. Plant Cell 9, 355–365. Smith, H.M.S. and Hake, S. (2003) The interaction of two homeobox genes, BREVIPEDICELLUS and PENNYWISE, regulates internode patterning in the Arabidopsis inflorescence. Plant Cell 15, 1717–1727. Sridhar, V.V., Surendrarao, A., Gonzalez, D., Conlan, R.S. and Liu, Z. (2004) Transcriptional repression of target genes by LEUNIG and SEUSS, two interacting regulatory proteins for Arabidopsis flower development. Proceedings of the National Academy of Sciences USA 101, 11494– 11499. Sun, T. (2000) Gibberellin signal transduction. Current Opinion in Plant Biology 3, 374–380. Theissen, G. (2001) Development of floral organ identity: stories from the MADS house. Current Opinion in Plant Biology 4, 75–85. Tilly, J.J., Allen, D.W. and Jack, T. (1998) The CArG boxes in the promoter of the Arabidopsis floral organ identity gene APETALA3 mediate diverse regulatory effects. Development 125, 1647–1657. Vergani, P., Morandinin, P. and Soave, C. (1997) Complementation of a yeast Dpkc1 mutant by the Arabidopsis protein ANT. FEBS Letters 400, 243– 246. Wagner, D., Sablowski, R.W.M. and Meyerowitz, E.M. (1999) Transcriptional activation of APETALA1 by LEAFY. Science 285, 582–584. Wang, X., Feng, S., Nakayama, N., Crosby, W.L., Irish, V., Deng, X.W. and Wei, N. (2003) The COP9 signalosome interacts with SCFUFO and participates in Arabidopsis flower

123 development. Plant Cell 15, 1071– 1082. Weigel, D. and Meyerowitz, E.M. (1993) Activation of floral homeotic genes in Arabidopsis. Science 261, 1723– 1726. Weigel, D., Alvarez, J., Smyth, D.R., Yanofsky, M.F. and Meyerowitz, E.M. (1992) LEAFY controls floral meristem identity in Arabidopsis. Cell 69, 843– 859. Wellmer, F., Riechmann, J.L., Alves-Ferreira, M. and Meyerowitz, E.M. (2004) Genome-wide analysis of spatial gene expression in Arabidopsis flowers. Plant Cell 16, 1314–1326. Western, T.L., Cheng, Y., Liu, J. and Chen, X. (2002) HUA ENHANCER2, a putative DExH-box RNA helicase, maintains homeotic B and C gene expression in Arabidopsis. Development 129, 1569–1581. Yang, W.-C., Ye, D., Xu, J. and Sundaresan, V. (1999) The SPOROCYTELESS gene of Arabidopsis is required for initiation of sporogenesis and encodes a novel nuclear protein. Genes & Development 13, 2108–2117. Yanofsky, M.F., Ma, H., Bowman, J.L., Drews, G.N., Feldman, K.A. and Meyerowitz, E.M. (1990) The protein encoded by the Arabidopsis homeotic gene AGAMOUS resembles transcription factors. Nature 346, 35–39. Yu, H., Ito, T., Zhao, Y., Peng, J., Kumar, P. and Meyerowitz, E.M. (2004) Floral homeotic genes are targets of gibberellin signaling in flower development. Proceedings of the National Academy of Sciences USA 101, 7827–7832. Zik, M. and Irish, V.F. (2003) Global identification of target genes regulated by APETALA3 and PISTILLATA floral homeotic gene action. Plant Cell 15, 207–222.

6

Molecular Developmental Genetics and the Evolution of Flowers G. THEISSEN

AND

K. KAUFMANN

Friedrich-Schiller-Universita¨t Jena, Lehrstuhl fu¨r Genetik, Philosophenweg 12, D-07743 Jena, Germany

Introduction Origin and diversification of angiosperms: an ‘abominable perplexing phenomenon’ The structural diversity of multicellular organisms on our planet originated to a large extent in two major ‘bursts’. During the ‘Cambrian Explosion’ about 540 million years ago (MYA) the body plans of (almost) all animal taxa (extant and extinct) originated within a few million years (Valentine et al., 1999); in many respects, such as number of species, the insects became by far the most successful group of animals. More than 300 million years later the origin and diversification of the flowering plants (angiosperms) provided the second example of an apparently ‘sudden’ origin and rapid early morphological radiation. The origin and early diversification of angiosperms was considered an ‘abominable mystery’ and ‘perplexing phenomenon’ by Charles Darwin about 150 years ago, and remained very speculative until today (Crepet, 1998, 2000; Theissen et al., 2002; Frohlich, 2003; Stuessy, 2004). The contributions of the origin and early diversification of angiosperms to biodiversity in terrestrial ecosystems were profound. Moreover, angiosperms provide us, directly or indirectly, with most human food (such as vegetables, fruits and cereal grains) and a lot of other important products (e.g. used to prepare clothes, furniture and drugs). Due to the obvious ecological and economic importance of angiosperms, considerable efforts have been made by botanists and evolutionary biologists to clarify their evolution. However, understanding the evolutionary origin and diversification of flowers has remained a considerable scientific challenge (Crepet, 1998, 2000; Frohlich, 1999, 2003; Frohlich and Parker, 2000; Ma and dePamphilis, 2000; Theissen et al., 2002; Winter et al., 2002a; Stuessy, 2004; Theissen and Becker, 2004). Major reasons for this are a quite uninformative (and probably extremely incomplete) fossil record and the great morphological gap 124

ßCAB International 2006. The Molecular Biology and Biotechnology of Flowering, 2nd edn (ed. B.R. Jordan)

Molecular Developmental Genetics and the Evolution of Flowers

125

Fig. 6.1. Diversity of reproductive cones and flowers in spermatophytes. Upper row from left to right: flower of Arabidopsis thaliana (core eudicot); zygomorphic flower of Antirrhinum majus (core eudicot); Tulipa turkestanica, a monocot, with petaloid tepals in both outer whorls. Lower row from left to right: Nymphaea odorata, a basal angiosperm, with undifferentiated perianth; male cone of Cycas circinalis, a gymnosperm; female cone of Gnetum gnemon, another gymnosperm.

between angiosperm flowers and the reproductive structures of angiosperms’ closest relatives, the gymnosperms, which leads to problems with homology assignments between reproductive organs from flowering plants and their putative ancestors (Frohlich, 2003). Gymnosperms comprise extant conifers (the most species-rich group), gnetophytes (Fig. 6.1), cycads (Fig. 6.1), Ginkgos, and diverse extinct groups such as Bennettitales, Cordaitales, corystosperms, glossopterids and some others (Doyle, 1998). There is no perfect agreement among students of flower evolution as to which are the key characters that distinguish angiosperms from gymnosperms; however, the presence of carpels enclosing the ovules is generally considered essential (Crane et al., 1995; Endress, 2001; Stuessy, 2004). Another feature that distinguishes typical angiosperm flowers from most of the reproductive cones of gymnosperms is the fact that male and female reproductive organs are usually united in one structure (or secondarily separated, as in the unisexual flowers of monoecious and dioecious angiosperms), while they might be primarily separated in different structures in gymnosperms (Fig. 6.1). Yet another typical feature of angiosperm flowers distinguishing them from gymnosperm cones is the presence of a perianth surrounding the reproductive organs, often including attractive organs of petaloid appearance (petals or tepals) (Fig. 6.1).

126

G. Theissen and K. Kaufmann

The origin of organ types and the series of morphological changes by which the floral organs of angiosperms originated are largely unknown. Equally unknown is how they have been assembled into flowers. Despite the sudden appearance of the combination of all angiosperm features (and hence angiosperms) in the fossil record, the origin of flowering plants was not necessarily a rapid process. On the contrary, Stuessy (2004) suggested that angiosperms may have evolved slowly from seed ferns in the Jurassic, beginning first with the carpel followed later by double fertilization and lastly by the appearance of flowers, a process that may well have taken more than 100 million years to complete. Only when the final combination of the essential angiosperm features was achieved, an ‘explosive’ evolutionary diversification may have set in. This raises the question as to why the angiosperms have been so ‘successful’ during evolution. One important aspect may have been that the arrangement of male and female organs in hermaphroditic flowers facilitated outcrossing by animal pollinators (mainly insects such as flies, butterflies, moths and beetles). Optimization of outcrossing and attendant modifications in compatibility, breeding systems and seed dispersal (often also employing animal vectors) might have enabled angiosperms to colonize even scattered and rare but diverse and manifold ecological niches due to the avoidance of inbreeding suppression at remote habitats (and thus helped to establish many more niches). Such niches may not have been accessible by the gymnosperms due to their simple pollination and seed dispersal systems. The ability to exist at scattered and rare places might have boosted speciation and thus might have significantly contributed to the enormous diversity within angiosperms. Coevolution with their pollinators might thus have catalysed waves of radiation of both angiosperms and insects, which eventually allowed them to dominate the vast majority of terrestrial ecosystems (Behrensmeyer et al., 1992). In addition to animal pollinators, selection from animal predators might very well have played a great role during angiosperm radiation (Stuessy, 2004). Testing such hypotheses about flower origin and diversification may require many more well-preserved fossils documenting the floras and faunas of past times, and providing ‘missing links’ between extant forms. But even if we knew the precise evolutionary series of morphological transitions and the ‘ultimate’ causes of the evolution of angiosperm flowers, the molecular genetic mechanisms of the different evolutionary innovations that generated floral structures and led to their diversification in the first place would remain unknown. These ‘proximate’ causes of the evolution of flowers that provided the raw material for selection are the major focus of this chapter.

Phylogeny of spermatophytes Understanding the evolution of flowers requires clarification of the phylogeny of the clade of angiosperms þ gymnosperms (spermatophytes, seed plants). During recent years the use of molecular markers has considerably changed our view on the evolution of spermatophytes. Nevertheless, the deep evolutionary relationships among the major groups of seed plants and the identification of

Molecular Developmental Genetics and the Evolution of Flowers

127

the most basal angiosperms have remained quite controversial. Unfortunately, these are the most relevant topics for a better understanding of the origin of angiosperm flowers. Extant gymnosperms are a structurally very diverse group and hence have usually been considered to be paraphyletic by traditional botanists, with gnetophytes often regarded as the sister group of the angiosperms (Doyle, 1998). However, quite a number of independent molecular marker studies rather surprisingly suggested that gnetophytes are more closely related to conifers than to angiosperms (Winter et al., 1999), and that all extant gymnosperms constitute a monophyletic group (see e.g. Chaw et al., 1997, 2000; Bowe et al., 2000; Frohlich and Parker, 2000). According to these studies extant gymnosperms and angiosperms are sister groups. They separated about 300 million years ago, as also suggested by molecular evidence (see e.g. Goremykin et al., 1997). Only recently, an about 270 million-year-old fossil cone has been reported that shares morphological features with both gnetophytes and conifers, suggesting a close relationship between both groups of gymnosperms (Wang, 2004). Moreover, investigations on the morphogenesis of the male reproductive structures of two gnetalean species (Welwitschia mirabilis and Ephedra distachya) proposed that conifers, Gnetales and Cordaitales originated from a common ancestor (Mundry and Stutzel, 2004). Thus palaeobotanical, neobotanical and molecular evidence has eventually started to converge on the same view of seed plant evolution. According to a number of investigations employing molecular data to reconstruct angiosperm phylogeny, the most basal extant angiosperm is Amborella trichopoda followed by Nymphaeales (water lilies) as the next branch, and then by a clade uniting Illiciaceae, Schisandraceae, Trimeniaceae and Austrobaileyaceae (Fig. 6.2) (for a review, see Kuzoff and Gasser, 2000; Bremer et al., 2003; Soltis and Soltis, 2003). For this grade of possibly most basal flowering plants the term ‘ANITA’ has been coined (Qiu et al., 1999). However, the basal position of the ANITA taxa is not undisputed. Some studies favour a clade of Amborella þ Nymphaeales as basal (Barkman et al., 2000; Graham and Olmstead, 2000), and studies based on whole chloroplast genome sequences even consider Amborella and Nymphaeales as eudicots and the monocots as closer to the base of the angiosperm tree (Goremykin et al., 2003, 2004). If the ANITA taxa are indeed the most basal angiosperms, the most recent common ancestor of extant angiosperms probably already had hermaphroditic flowers with an undifferentiated perianth, in which organs were arranged in more than two cycles or a spiral. Differentiated sepals and petals may have evolved later during the evolution of angiosperms (Kuzoff and Gasser, 2000). If so, and assuming monophyly of extant gymnosperms, the morphological gap between the most basal angiosperms and their closest living relatives appears even wider than previously often assumed (Theissen et al., 2002). In any case, employing molecular markers to reconstruct seed plant phylogeny could not solve the mystery of the origin of angiosperm flowers so far. The same is true for palaeobotanical approaches. The oldest known unequivocal angiosperm fossils represent two species of the genus Archaefructus

128

G. Theissen and K. Kaufmann

Asteridae: Antirrhinum, Petunia Cornales

Eudicots

Rosidae: Arabidopsis Saxifragales Caryophyllales Gunnerales Proteales Ranunculales Ceratophyllaleles

Monocots: Oryza, Tulipa

Magnoliidae Chloranthaceae Austrobaileyales (ITA)

Nymphaeales

ANITA grade

Amborellaceae

Gymnosperms: Gnetum

Fig. 6.2. Outline of the angiosperm phylogeny according to the APGII system (Bremer et al., 2003). Larger groups of angiosperms are marked in bold. Genus names of model plant species and species that are mentioned in this publication are placed in the groups they belong to. Gymnosperms form the sister group to angiosperms.

from the uppermost Jurassic about 125 million years ago, at most (Sun et al., 2002). Its only moderate age and a combination of both primitive and nonprimitive characters, however, do not qualify Archaefructus as a plausible very early angiosperm. It has even been argued that Archaefructus is a crowngroup angiosperm specialized for aquatic habit rather than a more primitive relative (Friis et al., 2003). In the framework of the ANITA hypothesis, the clade comprising all flowering plants except the ANITA grade is called the ‘euangiosperms’, comprising the Magnoliidae, monocots, Chloranthaceae, Ceratophyllales and

Molecular Developmental Genetics and the Evolution of Flowers

129

eudicots (Fig. 6.2). The eudicots comprise a grade of successive branches, with Ranunculales as sister to all other eudicots, and a large clade of ‘core eudicots’ containing the majority of all angiosperm species including the model plants thale cress (Arabidopsis thaliana; henceforth termed Arabidopsis), snapdragon (Antirrhinum majus; Antirrhinum) and petunia (Petunia hybrida; Petunia) (Kuzoff and Gasser, 2000; Soltis and Soltis, 2003).

Why evo-devo, and how? The difficulties of traditional botany and palaeobotany to explain flower origin and diversification led to the application of alternative approaches such as evolutionary developmental genetics (‘evo-devo’, for short). Its rationale in the context of flower evolution has been outlined in quite some detail elsewhere (Theissen and Saedler, 1995; Theissen et al., 2000, 2002). Briefly, multicellular organisms usually develop from single cells (zygotes) in each generation anew, implying that morphological changes in evolution always occur by changes in developmental processes. Since development is largely under genetic control, the basic corollary of evo-devo follows, i.e. that novel morphological forms in evolution usually result from changes in function of genes that control developmental processes. Thus typical evo-devo projects investigate the phylogeny of developmental-control genes and its role in the evolution of morphological features. Concerning flower development and evolution, some classes of MADS-box genes are of special interest (Theissen et al., 2000), because they specify the identity of floral organs in some core eudicotyledonous model plants such as Arabidopsis, Antirrhinum and Petunia, and probably also in most (if not all) other angiosperms. The typical flower of a core eudicot is composed of four different classes of organs arranged in four whorls: there are usually small, green sepals in the first and outermost whorl; often large and showy petals in the second whorl; male reproductive organs (stamens) in the third whorl; and female reproductive organs (carpels) in the fourth and innermost whorl. However, outside of the core eudicots floral architecture is much more diverse (Fig. 6.1). Based on the analysis of three different classes of homeotic mutants in which the identity of the floral organs has changed in a systematic way, simple hypotheses were suggested to explain how the identity of organs is specified during flower development by unique combinations of the activities of three different classes of ‘floral organ-identity genes’ (Fig. 6.3B–F). These genes are subdivided into class A, B and C genes and are expressed in the organ primordia of whorls one and two (class A), two and three (class B) and three and four (class C), respectively (Fig. 6.3H) (for a review, see Theissen, 2001). According to the ‘ABC model’ (Coen and Meyerowitz, 1991), expression of class A genes alone leads to sepal formation. The combination of A with B specifies the development of petals. The combination B and C leads to the formation of stamens, and expression of C alone determines the development of carpels. The ABC model also proposes that the class A and class C genes regulate each other in an antagonistic way, such that the class A genes become

130

G. Theissen and K. Kaufmann

Fig. 6.3. Class A (C), B (D, E) and C mutants (F) in Arabidopsis compared to wild-type flower (B), the ABCDE model (H upper part) and the ‘floral quartet’ model of flower development in Arabidopsis (H lower part; G); (A–F: from Riechmann and Meyerowitz, 1997 with kind permission from the publisher (Walter de Gruyter, Berlin); G: from Theissen and Saedler, 2001 with kind permission from the publisher (Nature Publishing Group); H: modified from Theissen, 2001).

expressed throughout the flower when the class C gene is defective, and vice versa (for a review of the ABC model, see Theissen, 2001). The ABC genes are considered as encoding developmental switches that activate the entire genetic programme for one particular organ, and repress non-appropriate genetic programmes. In Arabidopsis there are two different class A genes, APETALA1 (AP1) and APETALA2 (AP2), two class B genes, APETALA3 (AP3) and PISTILLATA (PI), but just one class C gene, AGAMOUS (AG). All these genes encode putative transcription factors (for a review, see Theissen and Saedler, 1999; Theissen, 2001), and except AP2, they are all members of the MADS box-gene family (for reviews about MADS box genes in plants, see Riechmann and Meyerowitz, 1997; Theissen et al., 2000, 2002; Ng and Yanofsky, 2001; Becker and Theissen, 2003; De Bodt et al., 2003). The ABC model was a milestone in the progress of our understanding of flower development, but it has two major shortcomings: the ABC genes are required, but not sufficient for the specification of floral organ identity. Moreover, the ABC model did not provide a molecular mechanism for the interaction of floral homeotic genes during the specification of floral organ identity. The ABC model was later extended by class D genes involved in ovule development (Angenent and Colombo, 1996), and class E genes (Theissen, 2001) required for petal, stamen and carpel development (Fig. 6.3H) (Pelaz et al., 2000). Furthermore, the gene-based ‘ABCDE’ model obtained this way was transformed into a protein-based ‘floral quartet model’ (Fig. 6.3G) (Theissen, 2001; Theissen and Saedler, 2001). It explains the interaction of

Molecular Developmental Genetics and the Evolution of Flowers

131

floral homeotic genes at the molecular level by the capacity of floral homeotic proteins to form tetrameric complexes of transcription factors (floral quartets). Some of these protein complexes are not only required but also sufficient to superimpose floral organ identity upon the developmental programme of vegetative leaves (Honma and Goto, 2001). Thus understanding the evolutionary origin of these protein complexes might be quite revealing with respect to understanding flower origin and diversification (Theissen and Saedler, 2001). In the following sections, the unfolding drama of floral evolution is presented in two major acts: (i) flower origin; and (ii) floral diversification.

Flower Origin Since the identity of floral organs is specified by conserved floral organ-identity genes, clarifying the phylogeny of these genes and their protein products may provide us with valuable insights into the evolution of flowers (for reviews, see Theissen and Saedler, 1995; Theissen et al., 2000, 2002). Analysis of numerous MADS box genes from mosses and ferns suggested that orthologues of floral homeotic genes are absent in non-seed plants (for a review, see Theissen et al., 2000). However, orthologues of class B and class C floral homeotic genes were identified in different conifers and in the gnetophyte Gnetum gnemon, orthologues of class C floral homeotic genes were also reported for Cycas and Ginkgo (Tandre et al., 1995, 1998; Rutledge et al., 1998; Mouradov et al., 1999; Sundstrom et al., 1999; Winter et al., 1999; Theissen et al., 2000; Fukui et al., 2001; Jager et al., 2003; Theissen and Becker, 2004; Zhang et al., 2004). This suggests that B- and C-type genes were established by gene duplications and sequence divergence in the lineage that led to extant seed plants, after the lineage that led to extant ferns had already branched off, i.e. 300–400 million years ago, even though some molecular clock estimates suggested a considerably earlier origin of these genes (Nam et al., 2003). The expression patterns of gymnosperm B and C genes are characteristically similar to those of class B and class C genes in angiosperms; while C genes are generally expressed in both male and female reproductive organs, B genes are predominantly expressed in male reproductive organs (and angiosperm petals) (for a review, see Theissen et al., 2000; Theissen and Becker, 2004). When expressed in the flowering plant Arabidopsis under the control of constitutive promoters, some B and C genes from Gnetum and Picea (representing gnetophytes and conifers, respectively) can at least partially substitute for their angiosperm orthologues in different kinds of experiments (Sundstrom and Engstrom, 2002; Winter et al., 2002a). Even complementation of class B and class C gene null mutants of Arabidopsis could be achieved with a B gene from Gnetum and a C gene from Cycas, respectively; complementation of the class B gene mutant by the Gnetum B gene was only partial, however (Winter et al., 2002a; Zhang et al., 2004). Together with the expression data in gymnosperms, these findings suggest that the ABC system specifying floral organ identity evolved from a precursor system (BC system) that was already

132

G. Theissen and K. Kaufmann

established in the most recent common ancestor of extant seed plants about 300 million years ago (Winter et al., 1999). The BC system may have specified female reproductive organs by the expression of C genes and male reproductive organs by the expression of both C and B genes (Theissen et al., 2002; Theissen and Becker, 2004). Alternatively, the floral homeotic function of B and C class genes may have been recruited from an ancestral non-homeotic function in reproductive organ development in gymnosperms. If differential expression of B genes specifying male organ identity represents the primary sex-determination mechanism of all seed plants (Winter et al., 1999), switching from male to female organ identity, or vice versa, could in principle result from changes in the expression of just one gene. This B gene might encode a transcription factor that controls, directly or indirectly, all the target genes required to bring about male rather than female organ identity during development. Assuming such a simple switch mechanism allowed the development of novel hypotheses on the origin of flowers (Theissen et al., 2002). These hypotheses start with truly unisexual axes as existing in most extant gymnosperms. The ‘out-of-male’ scenario assumes that hermaphroditic flowers originated from a male gymnosperm cone, and that reduction of B gene expression in the upper region of the male cone led to the development of female instead of male reproductive units in this upper region (Theissen et al., 2002). The ‘outof-female’ scenario maintains that flowers originated from a female cone. Under this hypothesis ectopic expression of B genes in the basal region of the female cone led to the development of male rather than female reproductive units in this basal region (Theissen et al., 2002). In both scenarios, a perianthless flower-like structure with male reproductive units in the basal region (outer whorls) and female reproductive units in the apical region (inner whorls) would have been established. Mutant analysis indicates that the structural transition predicted in these hypotheses is developmentally possible. In the conifer, Picea abies ‘acrocona’ hermaphroditic cones with female reproductive units at the top and male organs underneath have occasionally been observed (Theissen and Becker, 2004). Among the molecular changes, which might have caused the modifications in B gene expression are changes in genes that control B gene expression (encoding ‘trans-acting factors’) or changes in the cis-regulatory elements of the B genes themselves (for a more detailed discussion, see Theissen and Becker, 2004). Accordingly, conserved differences in upstream regulators of B gene expression between angiosperms and gymnosperms, specifically those that prevent B gene expression in the central part of the floral meristem or promote B gene expression specifically in the second and third whorl could provide evidence for the ‘out-of-male’ scenario. Under the hypotheses discussed here the perianth originated later than the arrangement of both male and female reproductive organs in flowers. A major reason for this assumption is that the identity of all the organs of perianthless flowers could be specified by organ-identity genes (B and C genes) that were probably established already in the most recent common ancestor of extant seed plants. In line with this, the flowers of Archaefructus, the oldest unequivocal

Molecular Developmental Genetics and the Evolution of Flowers

133

angiosperm fossils known, have no recognizable perianth and they display flowers as predicted by the out-of-male and out-of-female hypotheses, with female organs (carpels) at the top of floral axes and male organs (stamens) attached underneath (Sun et al., 2002). Even though Archaefructus may not really represent an ancestral angiosperm (Friis et al., 2003), perianthless flowers may well represent a character state that was more common in the early days of angiosperm evolution than it is today (Taylor and Hickey, 1996). Class A floral homeotic genes from the MADS box-gene family (so-called SQUAMOSA-like genes), specifying the identity of sepals and petals in core eudicots, originated later during seed plant evolution, i.e. after the lineage that led to extant angiosperms had split off from the gymnosperm lineage, probably by recruitment from floral meristem-identity genes (Theissen et al., 2000). Even later class B genes, in addition to their ancestral function in specifying male reproductive organs, may have been recruited to specify petaloid perianth organs (Theissen et al., 2002). The out-of-male and out-of-female hypotheses are not the only scenarios on flower origin that can be considered from molecular evidence. Two alternative hypotheses have been inspired by phylogeny reconstructions of the floral meristem-identity gene FLORICAULA (FLO)/LEAFY (LFY) and its orthologues (Frohlich and Parker, 2000; Albert et al., 2002). Extant gymnosperms have two paralogous copies of the LFY gene, termed LEAF and NEEDLE genes (Frohlich, 2003). Phylogenetic reconstructions suggest that the most recent common ancestor of extant gymnosperms and angiosperms had both a LEAF and a NEEDLE gene, but that the NEEDLE paralogue was lost in the angiosperm lineage shortly after its separation from the gymnosperm lineage (Frohlich and Parker, 2000). This, and the expression patterns of the LEAF (PRFLL) and NEEDLE (NLY) genes from the conifer Pinus radiata (mainly in male or female cones, respectively), has been taken as evidence that flower organization derives more from the male structure of ancestral gymnosperms than from the female structure – hence this hypothesis has been termed the ‘Mostly Male Theory’ (Frohlich and Parker, 2000; Frohlich, 2003). Though this aspect is similar to the out-of-male hypothesis, their supposed molecular mechanisms are quite different. The ‘Mostly Male Theory’ maintains that female organs (ovules) develop ectopically in male cones, caused by, e.g. the ectopic expression of ovule-identity genes (such as D genes, or C/D gene precursors) in male cones. In contrast, under the out-of-male hypothesis, spatial changes in B gene expression caused the initial morphological change (as outlined above). Yet another hypothesis of the origin of flowers, though based on the same data, comes to conclusions that are quite different from those of the Mostly Male Theory. Albert et al. (2002) consider the fact that both LEAF and NEEDLE paralogues from gymnosperms can complement lfy null mutants of Arabidopsis. The authors conclude thus that the male–female segregation in gymnosperms is not determined by differences in the coding regions of the LFY-like genes, but by differential regulation of LEAF (male) and NEEDLE (female) promoters. They propose that ‘sexual condensation’ (hermaphroditism) in angiosperms might have originated when previously distinct spatial regulation of copies of LFY-like genes in separate male and female apices

134

G. Theissen and K. Kaufmann

was amalgamated into singular LEAF control in all reproductive meristems following loss of the NEEDLE gene. Albert et al. (2002) hypothesize that positive selection on the partly redundant LEAF paralogue trapped reproductively unisexual angiosperm ancestors into a condensed, bisexual state. In clear contrast to the Mostly Male Theory and the out-of-male hypothesis (or the outof-female hypothesis, respectively), the hypothesis by Albert et al. (2002) does not predict that early stages of flower development are controlled mainly by genes orthologous to those expressed in male (or female) structures of gymnosperms. A more detailed comparison of current hypotheses on flower origin that take molecular genetic evidence into account is given by Theissen and Becker (2004). Unfortunately, all these hypotheses suffer from our very limited knowledge about the developmental genetics of gymnosperms. However, in contrast to previous hypotheses they make predictions that can be tested by applying molecular biology tools to extant plants (Soltis et al., 2002; Frohlich, 2003).

Floral diversification The flowers of the 250,000 or so extant angiosperm species on Earth vary in many different ways such as the identity, number, arrangement and shape of floral organs, and the symmetry of the flower as a whole (see, e.g. Fig. 6.1). These differences are all of great potential interest for future evo-devo studies aiming at understanding the developmental genetic basis of floral biodiversity. Mutants in model plants exist for all of these characters. However, detailed studies have been carried out so far only on organ identity and shape and floral symmetry, which are hence also the focus of this chapter.

Organ identity Changes in the expression patterns of floral homeotic genes Changes in the expression domains of floral homeotic genes in mutant or transgenic plants can bring about homeotic transformations of floral organs (Fig. 6.3C–F). For example, the expression of class C genes in the whorls of the perianth leads to a transformation of sepals into carpelloid organs and of petals into staminoid organs (Bradley et al., 1993). The ectopic expression of class B genes in the first and fourth floral whorls of Arabidopsis leads to a transformation of sepals into petaloid organs and of carpels into staminoid organs (Krizek and Meyerowitz, 1996). But are such changes suitable models for evolutionary processes? Under a paradigm of gradualistic evolution homeotic transformations are generally considered as of little evolutionary importance. It is assumed that they undermine the fitness of the affected organisms in such a serious way that there is always strong selection against them. However, in an evo-devo framework drastic (saltational), yet coordinated morphological changes appear more likely (Theissen et al., 2000, 2002; Bateman and DiMichele, 2002; Kramer

Molecular Developmental Genetics and the Evolution of Flowers

135

et al., 2003), and they possibly appear more plausible in the case of plants whose growth form is more ‘open’ and modular than that of animals. In line with this, De Craene (2003) provided morphological evidence for the evolutionary significance of homeosis in the flowers of diverse angiosperms such as Rosaceae, Papaveraceae and Lacandonia. For example, there is strong phylogenetic and morphological evidence that the petals of the Rosaceae (comprising well-known cultivated plants such as roses, strawberries and apples) were derived from stamens (De Craene, 2003). To determine whether changes in the spatial or temporal expression of floral homeotic genes have contributed to the structural diversification of the flower during angiosperm evolution, comparative expression studies in flowers with different architectures are required. In the following we discuss a few molecular case studies that corroborate the view that homeotic transitions played a role during the evolution of flowers. The perigon of Liliaceae Tulip (e.g. Tulipa gesneriana), lily (e.g. Lilium regale) and most of their relatives in the lily family (Liliaceae) have flowers displaying organ identities quite similar to the ones of higher eudicots. However, first whorl organs are typically petaloid like second whorl organs rather than sepaloid, i.e. the perianth is a perigon composed of two whorls of tepals (Fig. 6.1). This suggests a ‘modified ABC model’ showing expression of class B genes not only in the organs of whorls two and three, but also in the organs of the first whorl. When class B genes were investigated in lily and tulip, they were found to be expressed in the organs of the first three whorls of the flower, as predicted by the modified ABC model (Theissen et al., 2000; Kanno et al., 2003). Defining the exact roles of class B genes in tulip and lily flower development still requires mutant analysis. However, these findings already support the view that shifts in the boundaries of class B floral homeotic gene expression and hence floral homeotic changes contributed to the difference between typical eudicot and monocot flowers. However, in the flowers of garden asparagus (Asparagus officinalis), another member of the lily family, class B genes are only expressed in the third (stamen) and second (inner tepal) whorl but not in the first (outer tepal) whorl, although asparagus has also two whorls of almost identical petaloid tepals (Park et al., 2003, 2004a). How asparagus specifies the identity of first whorl tepals, and whether presence or absence of B gene expression in the first whorl of flowers of the lily family represents the ancestral state, remains to be seen. Petaloid organs in Ranunculaceae Many flowers of the basal eudicot family Ranunculaceae have distinctly different petaloid organs in the first two whorls (even though the first whorl organs are usually called ‘sepals’). Kramer et al. (2003) identified many duplication events of class B genes at different phylogenetic levels, with AP3-like genes displaying early duplications near the base of eudicots and PI-like genes more recent duplications. Expression studies suggest not only that petaloidy of first

136

G. Theissen and K. Kaufmann

whorl organs is due to a shift of B gene expression towards the first floral whorl (as in lily and tulip) but also that differential expression of a particular lineage of AP3-like genes has contributed to the distinction of the petaloid organs in the first and second floral whorl (Kramer et al., 2003). Again, future mutant analysis will be required to corroborate or falsify these hypotheses.

Changes in floral homeotic proteins Translational frameshifts Besides changes in expression patterns, differentiation in the coding regions of floral homeotic genes also may have contributed to the evolution of flowers (Vandenbussche et al., 2003). Most B proteins including all from gymnosperms, basal angiosperms, monocots and the TM6-like proteins from eudicots contain a conserved C-terminal sequence named ‘paleoAP3 motif’ (Kramer et al., 1998). However, some B proteins from core eudicots – including all class B floral homeotic proteins known to date – have an ‘euAP3 motif’ that differs strongly in its sequence from the paleoAP3 motif. Vandenbussche et al. (2003) provided evidence strongly suggesting that the euAP3 motif originated from the paleoAP3 motif by a translational frameshift mutation in a DEF-like class B gene at the base of the core eudicots. Both paleoAP3 (TM6 lineage) and euAP3 genes have been isolated from several core eudicots, suggesting that euAP3 genes originated by duplication of a paleoAP3 ancestral gene followed by a frameshift mutation in one of the copies. The two motifs are functionally not equivalent. When a paleoAP3 motif replaced the euAP3 motif, the respective chimerical AP3 construct could partly substitute stamens in the third whorl of Arabidopsis ap3-3 (class B gene) mutants but second whorl organs remained mutant, i.e. sepaloid (Lamb and Irish, 2003). This finding is quite reminiscent of the observations made with GGM2, a B gene from the gymnosperm G. gnemon, which has also a paleoAP3 motif (Winter et al., 2002a). It thus seems that a euAP3 motif is required for the formation of petals in core eudicots. Interestingly, Vandenbussche et al. (2003) and Litt and Irish (2003) identified an additional C-terminal frameshift mutation in another subfamily of MADS-box genes, termed ‘SQUA-like genes’, including also class A genes. Similar to the situation in class B proteins, SQUA-like proteins with ‘paleoAP3 motifs’ were found in all types of flowering plants, but proteins with ‘euAP1 motifs’ only in core eudicots. Thus genes encoding class B proteins with euAP3 motifs, genes encoding class A proteins with euAP1 motifs and the type of petal present in core eudicots probably all originated almost simultaneously (on evolutionary time scales) near the base of core eudicots (Litt and Irish, 2003; Vandenbussche et al., 2003). Remarkably, class A and B proteins constitute multimeric transcription factor complexes involved in petal formation in core eudicots, making it conceivable that the two translational frameshift mutations, the formation of multimeric transcription factor complexes, the origin of petals in core eudicots and the recruitment of these complexes for the specification of petal identity were somehow linked events.

Molecular Developmental Genetics and the Evolution of Flowers

137

Protein–protein interactions MADS-domain proteins bind as dimers to DNA sequences known as CArGboxes (consensus sequence: 5’-CC(A/T)6GG-3’), or to very similar sequences. The well-studied class B proteins from some eudicot model plants are stable and functional in the cell only as heterodimers of a DEF/AP3-like and a GLO/PIlike protein. These obligate heterodimers regulate their own expression at the transcriptional level by binding to CArG-boxes in the promoters of their own genes, and are thus co-expressed in petals and stamens during flower development. Heterodimerization is also absolutely required for movement of B proteins into the nucleus and DNA binding in vitro (reviewed by Theissen and Becker, 2004). In contrast, a diversity of interaction patterns, including obligate and facultative heterodimerization as well as homodimerization, has been found for class B proteins from monocots such as tulip (T. gesneriana) and lily (L. regale) (Winter et al., 2002b; Kanno et al., 2003). All orthologous proteins from gymnosperms tested so far, comprising proteins from the gnetophyte G. gnemon and the conifer P. abies, revealed the ability to homodimerize (Sundstrom and Engstrom, 2002; Winter et al., 2002b). These data suggest that the interaction of B proteins evolved from homodimerization in gymnosperms via facultative and obligate heterodimerization in monocots to exclusively obligate heterodimerization in core eudicots (Winter et al., 2002b). But why did B protein interaction evolve in this intriguing way? One hypothesis maintains that by restricting functional class B gene expression to sharp domains (whorls) within the flower; obligate heterodimerization may have contributed to the evolution of standardized and canalized structures of core eudicot flowers (Winter et al., 2002b). Another intriguing topic is the origin of multimeric complexes of MADSdomain proteins. The ‘floral quartet model’ explains the interaction of floral homeotic genes at the molecular level by the capacity of floral homeotic proteins to form tetrameric complexes of transcription factors (Theissen, 2001; Theissen and Saedler, 2001). Up to now, such complexes have only been shown in vitro and in yeast-3-hybrid and yeast-4-hybrid systems, so demonstrating their relevance in planta will be an important goal for the near future. But assuming that these higher order complexes are of functional relevance, the question arises as to when during evolution did dimers of MADS-domain protein get the capacity to form multimeric complexes? So far such complexes are only known from core eudicots (Egea-Cortines et al., 1999; Honma and Goto, 2001; Ferrario et al., 2003), so one hypothesis may hold that they are restricted to the angiosperms (or even restricted to some groups of angiosperms); an extreme alternative is the assumption that higher order complex formation is an intrinsic property of all MIKC-type MADS-domain proteins (Kaufmann et al., 2005). These proteins have been named after their conserved domain structure (Munster et al., 1997). MIKC-type genes have been found so far in all major groups of green plants (but nowhere else), and they include all of the plant MADS-box genes for which detailed information about their function (as revealed by a mutant phenotype) is available. It appears likely that clarifying the evolution of these protein complexes

138

G. Theissen and K. Kaufmann

will help to better understand flower origin and diversification (Theissen and Saedler, 2001; Kaufmann et al., 2005).

Carpel structure The carpel represents the most outstanding key innovation in angiosperm evolution (Taylor and Kirchner, 1996; Endress, 2001; Stuessy, 2004). The primary function of the carpel is to protect the ovules, to mediate and regulate pollination and to facilitate seed dispersal. The origin of the carpel is still unclear. According to one scenario the carpel originated from a ‘megasporophyll’, a leaf-like structure bearing ovules on the adaxial surface. Another theory suggests a composite nature of the carpel, consisting of an ovule-bearing shoot system that is surrounded by a bract-like organ (Friis and Endress, 1990; Taylor and Kirchner, 1996; Endress, 2001). Carpel morphology is highly variable among angiosperms. Number and positioning of ovules may influence carpel morphology, as do coevolution with pollinators and seed dispersal strategies. A pioneer study of Yamaguchi et al. (2003, 2004) suggests that the morphological difference of Arabidopsis and rice carpels is reflected in differences at the level of gene regulation. Here the functions of CRABS CLAW (CRC) and related genes in Arabidopsis and rice are reviewed as a case study for variation in molecular programmes directing carpel development in angiosperms. CRC is a member of a small gene family in Arabidopsis, termed the YABBY genes, encoding transcription factors (Bowman and Smyth, 1999). YABBY (YAB) genes have been proposed to promote abaxial cell fates in leaves and reproductive organs (Siegfried et al., 1999; Bowman, 2000; Eshed et al., 2001). They act together with and are regulated by KANADI (KAN) genes, which constitute another gene family important for promoting abaxial cell identity (Eshed et al., 2001). In contrast to other members of this gene family, CRC is specifically expressed in the carpels and nectaries of Arabidopsis flowers (Bowman and Smyth, 1999). In the carpel, CRC expression becomes restricted to two domains: an epidermal and an internal one. Epidermal expression is mostly, but not exclusively, in the outer layer (abaxial), similar to other members of this family like FILAMENTOUS FLOWER (FIL), YAB2 or YAB3 (Bowman and Smyth, 1999; Siegfried et al., 1999; Watanabe and Okada, 2003). Internal expression is adjacent to where the placental tissue develops; no expression is detectable in placentas, septum, stigma or ovules (Bowman and Smyth, 1999). Single mutants of CRC do not show a polarity phenotype in the carpel, but suggest a role of CRC in suppression of early radial growth and promotion of longitudinal growth. The mutant carpels are furthermore unfused at the apex. When combined with kan-2, crc mutations result in adaxial tissues developing in abaxial positions in the carpel, leading to ovule formation at the abaxial side of the carpel (Eshed et al., 1999). There may be redundancy of CRC with other members of the YAB gene family (Eshed et al., 1999). The fil mutants also show regular deformation of carpels (Chen et al., 1999; Sawa et al., 1999a,b).

Molecular Developmental Genetics and the Evolution of Flowers

139

They have multiple defects in flower development and may provide an interesting link between the specification of abaxial–adaxial identity and the establishment of the posterior-distal axis in the flower (Chen et al., 1999). Double and triple mutant analyses have identified genetic interactions between floral homeotic genes and CRC. CRC is negatively regulated by class B genes (Alvarez and Smyth, 1999; Bowman and Smyth, 1999). These data suggest that CRC has an important function in shaping the carpel morphology of Arabidopsis. Carpel morphology in the monocot rice (Oryza sativa) is quite different from that in Arabidopsis. The Arabidopsis carpel consists of specialized structures such as septa, abaxial repla, transmitting tissues and placenta (Bowman and Smyth, 1999; Bowman et al., 1999; Alvarez and Smyth, 2002), which are adaptations for a larger number of ovules and for seed dispersal. In contrast, the rice carpel is relatively simply organized. It consists of a single compound pistil of two carpels and a superior ovary with one locule containing a single ovule. Accordingly, the CRC orthologue in rice, DROOPING LEAF (DL) is expressed in all the carpel except from the central part from which the ovule arises (Yamaguchi et al., 2004). Unlike Arabidopsis, severe loss-of-function mutations of DL in rice result in the homeotic transformation of carpels into stamens and loss of determinacy (Nagasawa et al., 2003; Yamaguchi et al., 2003, 2004). Similar to CRC, however, DL acts antagonistically to class B floral homeotic genes. So at which level has functional diversification occurred between CRC and DL? It appears that DL expression is clearly more evenly distributed throughout the carpel. Some basic regulatory circuits like the antagonistic interaction with class B genes are apparently conserved. Yamaguchi et al. (2004) speculate that DL may have been recruited to acquire critical functions in the specification of carpel identity during grass evolution. Another aspect of functional diversification between CRC and DL is the role of DL outside the flower. Unlike CRC, DL is expressed in leaves, particularly in central regions of leaf primordia (Yamaguchi et al., 2004). Accordingly, dl mutants fail to form the midrib in the leaf. Similar ‘drooping leaf’ phenotypes have been observed in other grass species, which may suggest that orthologues with similar functions are present at least in the grass family. Recent data from a basal angiosperm support the view that the additional roles of DL in rice are derived rather than ancestral and may have originated in the monocot lineage. Fourquin et al. (2005) studied the expression pattern of a CRC orthologue in the putative basal angiosperm Amborella trichopoda. They found expression in the carpel wall very similar to the situation in Arabidopsis. Thus the important role of CRC orthologues in the determination of carpel morphology may be largely conserved among angiosperms, while additional functions have evolved later associated with floral diversification. The expression of another member of the YAB gene family, related to the Arabidopsis ovule-specific regulator of abaxial identity, INNER NO OUTER (INO) (Baker et al., 1997; Villanneva et al., 1999), has been studied in a basal angiosperm, Nymphaea (Yamada et al., 2003). Interestingly, the expression of this gene also appears to be conserved with respect to Arabidopsis.

140

G. Theissen and K. Kaufmann

YAB genes are also present in gymnosperms, and there is evidence that the diversification in this gene family predates angiosperm origin. Studying the evolution of YAB genes in more detail may significantly contribute to our understanding of carpel origin and diversification.

Floral symmetry The evolution and diversification of floral symmetry was dynamic and variable throughout angiosperm history. Taxa belonging to the ANITA clade of basal angiosperms have usually acyclic or partially acyclic flowers, caused by spiral arrangement of floral organs (Fig. 6.1). The number of floral organs is variable. Spiral arrangement of floral organs in mature flowers and variable organ number may represent ancestral conditions in angiosperms. Aside from the ANITA group, a polysymmetric (actinomorphic, radial) or monosymmetric (zygomorphic) cyclical arrangement of floral organs is most frequent (Fig. 6.1). Floral symmetry is not fixed in development; early stages of flower initiation may differ from mature flowers (Endress, 1999). This emphasizes the need to integrate morphological and molecular approaches to flower development to draw valid conclusions about evolutionary scenarios (Endress, 1999; Buzgo et al., 2004). Flowers of the model plant A. thaliana (Fig. 6.1) are radially symmetric (actinomorphic) with respect to perianth organs, but bilaterally symmetric (disymmetric) with respect to stamen and carpel whorls. Full actinomorphy of all floral whorls is infrequent among higher eudicots and mainly confined to groups of monocots and some basal eudicots (Rudall and Bateman, 2004). This morphological diversity is reflected in a dynamic evolutionary pattern in changes of symmetry across angiosperms (Neal et al., 1998; Reeves and Olmstead, 2003; Rudall and Bateman, 2004). Its flexibility is a major determinant in the coevolution of plants and their pollinators (Neal et al., 1998). Key regulators of floral symmetry have been identified by the analysis of peloric mutants, which have radially symmetric flowers in species with otherwise bilaterally symmetric flowers. In Antirrhinum, bilateral symmetry is mainly controlled by CYCLOIDEA (CYC) and its paralogue DICHOTOMA (DICH) (Luo et al., 1996, 1999). These genes determine dorsal organ identity by affecting growth rate and primordium initiation. Mutations in CYC produce an intermediate (semipeloric) phenotype; cyc dich double mutants are fully ventralized and radially symmetrical. CYC and DICH belong to the TCP family of plant-specific transcription factors, which is named after its first characterized members: TEOSINTE BRANCHED1 (TB1) from maize, CYC from Antirrhinum and PCF1 from O. sativa. CYC/DICH and TB1 belong to a subfamily of TCP genes, which is additionally characterized by an arginine-rich R domain additional to the DNA-binding TCP domain (Cubas, 2002). The putative orthologue of CYC in Arabidopsis, TCP1, is expressed transiently at the adaxial base of floral and axillary meristems (Cubas et al., 2001). This suggests that asymmetric

Molecular Developmental Genetics and the Evolution of Flowers

141

expression of CYC-like genes may predate the divergence of rosids and asterids, the major clades of core eudicots (Fig. 6.2). TB1 has been identified as a key gene involved in maize domestication (Doebley et al., 1995, 1997) that prevents outgrowth of axillary buds at lower nodes and promotes ear development at upper nodes (Hubbard et al., 2002). Interestingly, gene expression is conserved among CYC, TCP1 and TB1 in second and third whorl floral organs, whose identity is controlled by class B floral homeotic genes. The phylogenetic relationships of TCP genes at the level of angiosperms are not well resolved, due to lineage and site-specific differences in evolutionary rates as well as large gaps in gene sampling (Citerne et al., 2003). This may reflect a general pattern for genes that have undergone multiple adaptive changes in response to modifications of internal and external functional constraints. Studying patterns of molecular evolution of this gene family in relation to morphological differentiation at a lower evolutionary time scale has become very attractive (Citerne et al., 2000; Gubitz et al., 2003; Hileman and Baum, 2003; Reeves and Olmstead, 2003; Rudall and Bateman, 2003, 2004; Ree et al., 2004; Smith et al., 2004). Developmental changes in the evolution of zygomorphy have been mostly studied in relatives of Antirrhinum (Scrophulariaceae, order Lamiales) so far. Lamiales show an interesting distribution of different levels of zygomorphy (Endress, 1998). Linnaeus already described a naturally occurring peloric mutant in Linaria vulgaris (Scrophulariaceae) more than 250 years ago (reviewed by Theissen, 2000). The plants reproduce vegetatively as the mutant flowers are inaccessible to pollinators. Cubas et al. (1999) could show that this mutation was caused by hypermethylation at the LCYC locus, the CYC orthologue of Linaria. Partial demethylation causes recovery of the LCYC transcription and the flowers revert to wild type (for a more detailed review and discussion, especially of the evolutionary implications, see Theissen, 2000). In general, the evolutionary contribution of peloric mutants to floral diversification and speciation is still not well understood. Rudall and Bateman (2003) surveyed complete or partial peloric mutants in natural populations and found examples throughout the angiosperms with an enhanced number of examples in mints and orchids. In some cases these mutants may have kept their ability to reproduce to form populations and could be regarded as ‘hopeful monsters’ for the establishment of new evolutionary lineages (Rudall and Bateman, 2003). The successful establishment of peloric mutants in the wild may in many cases require co-adaptation of pollinators to the new floral forms. Alternatively, additional mutations are required to allow access of generalist pollinators to the flowers (Cronk and Moller, 1997; Cubas, 2002). An example may be the genus Ramonda (Gesneriaceae) with its nearly actinomorphic flowers, which has a much reduced corolla tube compared to its close relatives with zygomorphic flowers (Cubas, 2002). CYC-like genes may have a conserved function in the establishment of zygomorphy in the order Lamiales. Citerne et al. (2000) studied CYC-like genes in the family Gesneriaceae. They found a frameshift mutation in a CYC-like gene of a naturally occurring homeotic mutant of Sinningia

142

G. Theissen and K. Kaufmann

speciosa. Genetic analyses will have to reveal whether there is a correlation between the mutant locus and the peloric phenotype. The phylogeny of CYClike genes in Gesneriaceae is characterized by the occurrence of family-specific gene duplications. Hileman and Baum (2003) studied the expression of CYC and DICH orthologues in Mohavea confertifolia (Scrophulariaceae). They found slight shifts in expression domains in both genes, which they correlate to differences in flower morphology between Mohavea and Antirrhinum. In particular, they correlate lateral stamen abortion to an expansion of CYC and DICH expression, and changes in adaxial petal symmetry to a reduction of DICH expression. This analysis may represent a good case study to show that simple changes in the expression of regulatory genes may relate to floral diversification and coevolution with pollinators. A still open question is whether floral asymmetry is also outside the Lamiales controlled by CYC-like genes. For example, it would be particularly interesting to determine whether CYC-like genes are involved in the establishment of zygomorphy in the capitulum (flower head) of the Asteraceae (Cubas, 2002).

Outlook There is evidence that changes in both regulatory and coding regions of developmental control genes, modifying the expression patterns of these genes and the function of the proteins they encode, have contributed to the origin as well as the diversification of flowers during the evolution of angiosperms. Many of these genes are members of moderate to large families encoding transcription factors, important examples being the MADS-box, YAB and TCP genes. Further investigations on these ‘evo-devo genes’ thus promise a significant leap forwards towards a better understanding of flower origin and diversification.

Acknowledgements We are grateful to Hannelore Simon and Rainer Melzer for providing pictures of tulip and Gnetum, respectively.

References Albert, V.A., Oppenheimer, D.G. and Lindqvist, C. (2002) Pleiotropy, redundancy and the evolution of flowers. Trends in Plant Science 7, 297–301. Alvarez, J. and Smyth, D.R. (1999) CRABS CLAW and SPATULA, two

Arabidopsis genes that control carpel development in parallel with AGAMOUS. Development 126, 2377– 2386. Alvarez, J. and Smyth, D.R. (2002) Crabs claw and Spatula genes regulate growth and pattern formation during

Molecular Developmental Genetics and the Evolution of Flowers gynoecium development in Arabidopsis thaliana. International Journal of Plant Sciences 163, 17–41. Angenent, G.C. and Colombo, L. (1996) Molecular control of ovule development. Trends in Plant Science 1, 228–232. Baker, S.C., RobinsonBeers, K., Villanueva, J.M., Gaiser, J.C. and Gasser, C.S. (1997) Interactions among genes regulating ovule development in Arabidopsis thaliana. Genetics 145, 1109–1124. Barkman, T.J., Chenery, G., McNeal, J.R., Lyons-Weiler, J., Ellisens, W.J., Moore, G., Wolfe, A.D. and dePamphilis, C.W. (2000) Independent and combined analyses of sequences from all three genomic compartments converge on the root of flowering plant phylogeny. Proceedings of the National Academy of Sciences USA 97, 13166–13171. Bateman, R.M. and DiMichele, W.A. (2002) Generating and filtering major phenotypic novelties: neoGoldschmidtian saltation revisited. In: Cronk, Q.C., Bateman, R.M. and Hawkins, J.M. (eds) Developmental Genetics and Plant Evolution. Taylor & Francis, London, pp. 109–159. Becker, A. and Theissen, G. (2003) The major clades of MADS-box genes and their role in the development and evolution of flowering plants. Molecular Phylogenetics and Evolution 29, 464–489. Behrensmeyer, A.K., Damuth, J.D., DiMichele, W.A., Potts, R., Sues, H.D. and Wing, S.L. (1992) Terrestrial Ecosystems Through Time. Chicago University Press, Chicago, Illinois. Bowe, L.M., Coat, G. and dePamphilis, C.W. (2000) Phylogeny of seed plants based on all three genomic compartments: extant gymnosperms are monophyletic and Gnetales’ closest relatives are conifers. Proceedings of the National Academy of Sciences USA 97, 4092–4097.

143

Bowman, J.L. (2000) The YABBY gene family and abaxial cell fate. Current Opinion in Plant Biology 3, 17–22. Bowman, J.L. and Smyth, D.R. (1999) CRABS CLAW, a gene that regulates carpel and nectary development in Arabidopsis, encodes a novel protein with zinc finger and helix-loop-helix domains. Development 126, 2387– 2396. Bowman, J.L., Baum, S.F., Eshed, Y., Putterill, J. and Alvarez, J. (1999) Molecular genetics of gynoecium development in Arabidopsis. Current Topics in Developmental Biology 45, 155–205. Bradley, D., Carpenter, R., Sommer, H., Hartley, N. and Coen, E. (1993) Complementary floral homeotic phenotypes result from opposite orientations of a transposon at the Plena-Locus of Antirrhinum. Cell 72, 85–95. Bremer, B., Bremer, K., Chase, M.W., Reveal, J.L., Soltis, D.E., Soltis, P.S., Stevens, P.F., Anderberg, A.A., Fay, M.F., Goldblatt, P., Judd, W.S., Kallersjo, M., Karehed, J., Kron, K.A., Lundberg, J., Nickrent, D.L., Olmstead, R.G., Oxelman, B., Pires, J.C., Rodman, J.E., Rudall, P.J., Savolainen, V., Sytsma, K.J., van der Bank, M., Wurdack, K., Xiang, J.Q.Y., Zmarzty, S. and Grp, A.P. (2003) An update of the angiosperm phylogeny group classification for the orders and families of flowering plants: APG II. Botanical Journal of the Linnean Society 141, 399–436. Buzgo, M., Soltis, D.E., Soltis, P.S. and Ma, H. (2004) Towards a comprehensive integration of morphological and genetic studies of floral development. Trends in Plant Science 9, 164–173. Chaw, S.M., Zharkikh, A., Sung, H.M., Lau, T.C. and Li, W.H. (1997) Molecular phylogeny of extant gymnosperms and seed plant evolution: analysis of nuclear 18S rRNA sequences. Molecular Biology and Evolution 14, 56–68. Chaw, S.M., Parkinson, C.L., Cheng, Y.C., Vincent, T.M. and Palmer, J.D.

144

G. Theissen and K. Kaufmann (2000) Seed plant phylogeny inferred from all three plant genomes: monophyly of extant gymnosperms and origin of Gnetales from conifers. Proceedings of the National Academy of Sciences USA 97, 4086–4091. Chen, Q.Y., Atkinson, A., Otsuga, D., Christensen, T., Reynolds, L. and Drews, G.N. (1999) The Arabidopsis FILAMENTOUS FLOWER gene is required for flower formation. Development 126, 2715–2726. Citerne, H.L., Moller, M. and Cronk, Q.C.B. (2000) Diversity of cycloidealike genes in Gesneriaceae in relation to floral symmetry. Annals of Botany 86, 167–176. Citerne, H.L., Luo, D., Pennington, R.T., Coen, E. and Cronk, Q.C. (2003) A phylogenomic investigation of CYCLOIDEA-like TCP genes in the Leguminosae. Plant Physiology 131, 1042–1053. Coen, E.S. and Meyerowitz, E.M. (1991) The war of the whorls: genetic interactions controlling flower development. Nature 353, 31–37. Crane, P.R., Friis, E.M. and Pedersen, K.R. (1995) The origin and early diversification of angiosperms. Nature 374, 27–33. Crepet, W.L. (1998) The abominable mystery. Science 282, 1653–1654. Crepet, W.L. (2000) Progress in understanding angiosperm history, success, and relationships: Darwin’s abominably ‘perplexing phenomenon’. Proceedings of the National Academy of Sciences USA 97, 12939–12941. Cronk, Q.C. and Moller, M. (1997) Genetics of floral asymmetry revealed. Trends in Ecology and Evolution 12, 85–86. Cubas, P. (2002) Role of TCP genes in the evolution of morphological characters in angiosperms. In: Hawkins, J.M. (ed.) Developmental Genetics and Plant Evolution. Taylor & Francis, London and New York, pp. 247–266. Cubas, P., Vincent, C. and Coen, E. (1999) An epigenetic mutation respon-

sible for natural variation in floral symmetry. Nature 401, 157–161. Cubas, P., Coen, E. and Zapater, J.M.M. (2001) Ancient asymmetries in the evolution of flowers. Current Biology 11, 1050–1052. De Bodt, S., Raes, J., Van de Peer, Y. and Theissen, G. (2003) And then there were many: MADS goes genomic. Trends in Plant Science 8, 475–483. De Craene, L.P.R. (2003) The evolutionary significance of homeosis in flowers: morphological perspective. International Journal of Plant Sciences 164, S225–S235. Doebley, J., Stec, A. and Gustus, C. (1995) Teosinte Branched1 and the origin of maize – evidence for epistasis and the evolution of dominance. Genetics 141, 333–346. Doebley, J., Stec, A. and Hubbard, L. (1997) The evolution of apical dominance in maize. Nature 386, 485–488. Doyle, J.A. (1998) Phylogeny of vascular plants. Annual Review of Ecology and Systematics 29, 567–599. Egea-Cortines, M., Saedler, H. and Sommer, H. (1999) Ternary complex formation between the MADS-box proteins SQUAMOSA, DEFICIENS and GLOBOSA is involved in the control of floral architecture in Antirrhinum majus. EMBO Journal 18, 5370–5379. Endress, P.K. (1998) Antirrhinum and Asteridae – evolutionary changes of floral symmetry. Symposium of the Society of Experimental Biology 51, 133–140. Endress, P.K. (1999) Symmetry in flowers: diversity and evolution. International Journal of Plant Sciences 160, S3–S23. Endress, P.K. (2001) Origins of flower morphology. Journal of Experimental Zoology 291, 105–115. Eshed, Y., Baum, S.F. and Bowman, J.L. (1999) Distinct mechanisms promote polarity establishment in carpels of Arabidopsis. Cell 99, 199–209.

Molecular Developmental Genetics and the Evolution of Flowers Eshed, Y., Baum, S.F., Perea, J.V. and Bowman, J.L. (2001) Establishment of polarity in lateral organs of plants. Current Biology 11, 1251–1260. Ferrario, S., Immink, R.G., Shchennikova, A., Busscher-Lange, J. and Angenent, G.C. (2003) The MADS box gene FBP2 is required for SEPALLATA function in petunia. Plant Cell 15, 914–925. Fourquin, C., Vinauger-Douard, M., Fogliani, B., Dumas C. and Scutt, C.P. (2005) Evidence that CRABS CLAW and TOUSLED have conserved their roles in carpel development since the ancestor of the extant angiosperms. Proceedings of the National Academy of Sciences USA 102, 4649–4654. Friis, E.M. and Endress, P.K. (1990) Origin and evolution of angiosperm flowers. Advances in Botanical Research Incorporating Advances in Plant Pathology 17, 99–162. Friis, E.M., Doyle, J.A., Endress, P.K. and Leng, Q. (2003) Archaefructus – angiosperm precursor or specialized early angiosperm? Trends in Plant Science 8, 369–373. Frohlich, M.W. (1999) MADS about Gnetales. Proceedings of the National Academy of Sciences USA 96, 8811–8813. Frohlich, M.W. (2003) An evolutionary scenario for the origin of flowers. Nature Reviews Genetics 4, 559–566. Frohlich, M.W. and Parker, D.S. (2000) The mostly male theory of flower evolutionary origins: from genes to fossils. Systematic Botany 25, 155–170. Fukui, M., Futamura, N., Mukai, Y., Wang, Y.Q., Nagao, A. and Shinohara, K. (2001) Ancestral MADS box genes in sugi, Cryptomeria japonica D. Don (Taxodiaceae), homologous to the B function genes in angiosperms. Plant and Cell Physiology 42, 566–575. Goremykin, V.V., Hansmann, S. and Martin, W.F. (1997) Evolutionary an-

145

alysis of 58 proteins encoded in six completely sequenced chloroplast genomes: revised molecular estimates of two seed plant divergence times. Plant Systematics and Evolution 206, 337–351. Goremykin, V.V., Hirsch-Ernst, K.I., Wolfl, S. and Hellwig, F.H. (2003) Analysis of the Amborella trichopoda chloroplast genome sequence suggests that Amborella is not a basal angiosperm. Molecular Biology and Evolution 20, 1499–1505. Goremykin, V.V., Hirsch-Ernst, K.I., Wolfl, S. and Hellwig, F.H. (2004) The chloroplast genome of Nymphaea alba: whole-genome analyses and the problem of identifying the most basal angiosperm. Molecular Biology and Evolution 21, 1445–1454. Graham, S.W. and Olmstead, R.G. (2000) Utility of 17 chloroplast genes for inferring the phylogeny of the basal angiosperms. American Journal of Botany 87, 1712–1730. Gubitz, T., Caldwell, A. and Hudson, A. (2003) Rapid molecular evolution of CYCLOIDEA-like genes in Antirrhinum and its relatives. Molecular Biology and Evolution 20, 1537–1544. Hileman, L.C. and Baum, D.A. (2003) Why do paralogs persist? Molecular evolution of CYCLOIDEA and related floral symmetry genes in Antirrhineae (Veronicaceae). Molecular Biology and Evolution 20, 591–600. Honma, T. and Goto, K. (2001) Complexes of MADS-box proteins are sufficient to convert leaves into floral organs. Nature 409, 525–529. Hubbard, L., McSteen, P., Doebley, J. and Hake, S. (2002) Expression patterns and mutant phenotype of teosinte branched1 correlate with growth suppression in maize and teosinte. Genetics 162, 1927–1935. Jager, M., Hassanin, A., Manuel, M., Le Guyader, H. and Deutsch, J. (2003) MADS-box genes in Ginkgo biloba and the evolution of the AGAMOUS

146

G. Theissen and K. Kaufmann family. Molecular Biology and Evolution 20, 842–854. Kanno, A., Saeki, H., Kameya, T., Saedler, H. and Theissen, G. (2003) Heterotopic expression of class B floral homeotic genes supports a modified ABC model for tulip (Tulipa gesneriana). Plant Molecular Biology 52, 831–841. Kaufmann, K., Melzer, R. and Theißen, G. (2005) MIKC-type MADS-domain proteins: structural modularity, protein interactions and network evolution in land plants. Gene 347, 183–198. Kramer, E.M., Di Stilio, V.S. and Schluter, P.M. (2003) Complex patterns of gene duplication in the APETALA3 and PISTILLATA lineages of the Ranunculaceae. International Journal of Plant Sciences 164, 1–11. Kramer, E.M., Dorit, R.L. and Irish, V.F. (1998) Molecular evolution of genes controlling petal and stamen development: duplication and divergence within the APETALA3 and PISTILLATA MADS-box gene lineages. Genetics 149, 765–783. Krizek, B.A. and Meyerowitz, E.M. (1996) The Arabidopsis homeotic genes APETALA3 and PISTILLATA are sufficient to provide the B class organ identity function. Development 122, 11–22. Kuzoff, R.K. and Gasser, C.S. (2000) Recent progress in reconstructing angiosperm phylogeny. Trends in Plant Science 5, 330–336. Lamb, R.S. and Irish, V.F. (2003) Functional divergence within the APETALA3/PISTILLATA floral homeotic gene lineages. Proceedings of the National Academy of Sciences USA 100, 6558–6563. Litt, A. and Irish, V.F. (2003) Duplication and diversification in the APETALA1/ FRUITFULL floral homeotic gene lineage: implications for the evolution of floral development. Genetics 165, 821–833. Luo, D., Carpenter, R., Vincent, C., Copsey, L. and Coen, E. (1996) Origin of floral asymmetry in Antirrhinum. Nature 383, 794–799.

Luo, D., Carpenter, R., Copsey, L., Vincent, C., Clark, J. and Coen, E. (1999) Control of organ asymmetry in flowers of Antirrhinum. Cell 99, 367–376. Ma, H. and dePamphilis, C. (2000) The ABCs of floral evolution. Cell 101, 5–8. Mouradov, A., Hamdorf, B., Teasdale, R.D., Kim, J.T., Winter, K.U. and Theissen, G. (1999) A DEF/GLO-like MADS-box gene from a gymnosperm: Pinus radiata contains an ortholog of angiosperm B class floral homeotic genes. Developmental Genetics 25, 245–252. Mundry, M. and Stutzel, T. (2004) Morphogenesis of the reproductive shoots of Welwitschia mirabilis and Ephedra distachya (Gnetales), and its evolutionary implications. Organisms Diversity & Evolution 4, 91–108. Munster, T., Pahnke, J., Di Rosa, A., Kim, J.T., Martin, W., Saedler, H. and Theissen, G. (1997) Floral homeotic genes were recruited from homologous MADS-box genes preexisting in the common ancestor of ferns and seed plants. Proceedings of the National Academy of Sciences USA 94, 2415–2420. Nagasawa, N., Miyoshi, M., Sano, Y., Satoh, H., Hirano, H., Sakai, H. and Nagato, Y. (2003) SUPERWOMAN1 and DROOPING LEAF genes control floral organ identity in rice. Development 130, 705–718. Nam, J., dePamphilis, C.W., Ma, H. and Nei, M. (2003) Antiquity and evolution of the MADS-box gene family controlling flower development in plants. Molecular Biology and Evolution 20, 1435–1447. Neal, P.R., Dafni, A. and Giurfa, M. (1998) Floral symmetry and its role in plant-pollinator systems: terminology, distribution, and hypotheses. Annual Review of Ecology and Systematics 29, 345–373. Ng, M. and Yanofsky, M.F. (2001) Function and evolution of the plant MADS-box gene family. Nature Reviews Genetics 2, 186–195.

Molecular Developmental Genetics and the Evolution of Flowers Park, J.H., Ishikawa, Y., Ochiai, T., Kanno, A. and Kameya, T. (2004a) Two GLOBOSA-like genes are expressed in second and third whorls of homochlamydeous flowers in Asparagus officinalis L. Plant and Cell Physiology 45, 325–332. Park, J.H., Ishikawa, Y., Yoshida, R., Kanno, A. and Kameya, T. (2003) Expression of AODEF, a B-functional MADS-box gene, in stamens and inner tepals of the dioecious species Asparagus officinalis L. Plant Molecular Biology 51, 867–875. Pelaz, S., Ditta, G.S., Baumann, E., Wisman, E. and Yanofsky, M.F. (2000) B and C floral organ identity functions require SEPALLATA MADSbox genes. Nature 405, 200–203. Qiu, Y.L., Lee, J.H., BernasconiQuadroni, F., Soltis, D.E., Soltis, P.S., Zanis, M., Zimmer, E.A., Chen, Z.D., Savolainen, V. and Chase, M.W. (1999) The earliest angiosperms: evidence from mitochondrial, plastid and nuclear genomes. Nature 402, 404– 407. Ree, R.H., Citerne, H.L., Lavin, M. and Cronk, Q.C.B. (2004) Heterogeneous selection on LEGCYC paralogs in relation to flower morphology and the phylogeny of Lupinus (Leguminosae). Molecular Biology and Evolution 21, 321–331. Reeves, P.A. and Olmstead, R.G. (2003) Evolution of the TCP gene family in Asteridae: cladistic and network approaches to understanding regulatory gene family diversification and its impact on morphological evolution. Molecular Biology and Evolution 20, 1997–2009. Riechmann, J.L. and Meyerowitz, E.M. (1997) MADS domain proteins in plant development. Biological Chemistry 378, 1079–1101. Rudall, P.J. and Bateman, R.M. (2003) Evolutionary change in flowers and inflorescences: evidence from naturally occurring terata. Trends in Plant Science 8, 76–82.

147

Rudall, P.J. and Bateman, R.M. (2004) Evolution of zygomorphy in monocot flowers: iterative patterns and developmental constraints. New Phytologist 162, 25–44. Rutledge, R., Regan, S., Nicolas, O., Fobert, P., Cote, C., Bosnich, W., Kauffeldt, C., Sunohara, G., Seguin, A. and Stewart, D. (1998) Characterization of an AGAMOUS homologue from the conifer black spruce (Picea mariana) that produces floral homeotic conversions when expressed in Arabidopsis. Plant Journal 15, 625–634. Sawa, S., Ito, T., Shimura, Y. and Okada, K. (1999a) FILAMENTOUS FLOWER controls the formation and development of Arabidopsis inflorescences and floral meristems. Plant Cell 11, 69–86. Sawa, S., Watanabe, K., Goto, K., Kanaya, E., Morita, E.H. and Okada, K. (1999b) FILAMENTOUS FLOWER, a meristem and organ identity gene of Arabidopsis: encodes a protein with a zinc finger and HMGrelated domains. Genes & Development 13, 1079–1088. Siegfried, K.R., Eshed, Y., Baum, S.F., Otsuga, D., Drews, G.N. and Bowman, J.L. (1999) Members of the YABBY gene family specify abaxial cell fate in Arabidopsis. Development 126, 4117–4128. Smith, J.F., Hileman, L.C., Powell, M.P. and Baum, D.A. (2004) Evolution of GCYC, a Gesneriaceae homolog of CYCLOIDEA, within Gesnerioideae (Gesneriaceae). Molecular Phylogenetics and Evolution 31, 765–779. Soltis, D.E. and Soltis, P.S. (2003) The role of phylogenetics in comparative genetics. Plant Physiology 132, 1790–1800. Soltis, D.E., Soltis, P.S., Albert, V.A., Oppenheimer, D.G., dePamphilis, C.W., Ma, H., Frohlich, M.W., Theissen, G. and Grp, F.G.P.R. (2002) Missing links: the genetic architecture of flower and floral diversification. Trends in Plant Science 7, 22–31.

148

G. Theissen and K. Kaufmann Stuessy, T.F. (2004) A transitional-combinational theory for the origin of angiosperms. Taxon 53, 3–16. Sun, G., Ji, Q., Dilcher, D.L., Zheng, S.L., Nixon, K.C. and Wang, X.F. (2002) Archaefructaceae, a new basal angiosperm family. Science 296, 899–904. Sundstrom, J. and Engstrom, P. (2002) Conifer reproductive development involves B-type MADS-box genes with distinct and different activities in male organ primordia. Plant Journal 31, 161–169. Sundstrom, J., Carlsbecker, A., Svensson, M.E., Svenson, M., Johanson, U., Theissen, G. and Engstrom, P. (1999) MADS-box genes active in developing pollen cones of Norway spruce (Picea abies) are homologous to the B-class floral homeotic genes in angiosperms. Developmental Genetics 25, 253– 266. Tandre, K., Albert, V.A., Sundas, A. and Engstrom, P. (1995) Conifer homologs to genes that control floral development in angiosperms. Plant Molecular Biology 27, 69–78. Tandre, K., Svenson, M., Svensson, M.E. and Engstrom, P. (1998) Conservation of gene structure and activity in the regulation of reproductive organ development of conifers and angiosperms. Plant Journal 15, 615–623. Taylor, D.W. and Hickey, L.J. (1996) Flowering plant origin, early evolution and phylogeny. In: Taylor, D.W. and Hickey, L.J. (eds) Flowering Plant Origin, Evolution and Phylogeny. Chapman & Hall, New York, pp. 1–7. Taylor, D.W. and Kirchner, K. (1996) The origin and evolution of the angiosperm carpel. In: Taylor, D.W. and Hickey, L.J. (eds) Flowering Plant Origin, Evolution and Phylogeny. Chapman & Hall, New York, pp. 116–140. Theissen, G. (2000) Evolutionary developmental genetics of floral symmetry: the revealing power of Linnaeus’ monstrous flower. Bioessays 22, 209–213.

Theissen, G. (2001) Development of floral organ identity: stories from the MADS house. Current Opinion in Plant Biology 4, 75–85. Theissen, G. and Becker, A. (2004) Gymnosperm orthologues of class B floral homeotic genes and their impact on understanding flower origin. Critical Reviews in Plant Sciences 23, 129– 148. Theissen, G. and Saedler, H. (1995) MADS-box genes in plant ontogeny and phylogeny: Haeckel’s ‘biogenetic law’ revisited. Current Opinion in Genetics and Development 5, 628– 639. Theissen, G. and Saedler, H. (1999) The golden decade of molecular floral development (1990–1999): a cheerful obituary. Developmental Genetics 25, 181–193. Theissen, G. and Saedler, H. (2001) Plant biology: floral quartets. Nature 409, 469–471. Theissen, G., Becker, A., Di Rosa, A., Kanno, A., Kim, J.T., Munster, T., Winter, K.U. and Saedler, H. (2000) A short history of MADS-box genes in plants. Plant Molecular Biology 42, 115–149. Theissen, G., Becker, A., Winter, K.U., Munster, T., Kirchner, C. and Saedler, H. (2002) How the land plants learned their floral ABCs: the role of MADS box genes in the evolutionary origin of flowers. In: Cronk, Q.C., Bateman, R.M. and Hawkins, J.M. (eds) Developmental Genetics and Plant Evolution. Taylor & Francis, London. Valentine, J.W., Jablonski, D. and Erwin, D.H. (1999) Fossils, molecules and embryos: new perspectives on the Cambrian explosion. Development 126, 851–859. Vandenbussche, M., Theissen, G., Van de Peer, Y. and Gerats, T. (2003) Structural diversification and neo-functionalization during floral MADS-box gene evolution by C-terminal frameshift mutations. Nucleic Acids Research 31, 4401–4409.

Molecular Developmental Genetics and the Evolution of Flowers Villanneva, J.M., Broadhvest, J., Hauser, B.A., Meister, R.J., Schneitz, K. and Gasser, C.S. (1999) INNER NOOUTER regulates abaxial–adaxial patterning in Arabidopsis ovules. Genes & Development 13, 3160–3169. Wang, Z.-Q. (2004) A new Permian Gnetalean cone as fossil evidence for supporting current molecular phylogeny. Annals of Botany 94, 281–288. Watanabe, K. and Okada, K. (2003) Two discrete cis elements control the abaxial side-specific expression of the FILAMENTOUS FLOWER gene in Arabidopsis. Plant Cell 15, 2592– 2602. Winter, K.U., Becker, A., Munster, T., Kim, J.T., Saedler, H. and Theissen, G. (1999) MADS-box genes reveal that gnetophytes are more closely related to conifers than to flowering plants. Proceedings of the National Academy of Sciences USA 96, 7342–7347. Winter, K.U., Saedler, H. and Theissen, G. (2002a) On the origin of class B floral homeotic genes: functional substitution and dominant inhibition in Arabidopsis by expression of an orthologue from the gymnosperm Gnetum. Plant Journal 31, 457–475. Winter, K.U., Weiser, C., Kaufmann, K., Bohne, A., Kirchner, C., Kanno, A.,

149

Saedler, H. and Theissen, G. (2002b) Evolution of class B floral homeotic proteins: obligate heterodimerization originated from homodimerization. Molecular Biology and Evolution 19, 587–596. Yamada, T., Ito, M. and Kato, M. (2003) Expression pattern of INNER NOOUTER homologue in Nymphaea (water lily family, Nymphaeaceae). Development Genes and Evolution 213, 510–513. Yamaguchi, T., An, G., Hirochika, H. and Hirano, H.Y. (2003) Genetic interactions of the DROOPING LEAF gene and the ABC genes in floral organ specification in rice. Plant and Cell Physiology 44, S183-S183. Yamaguchi, T., Nagasawa, N., Kawasaki, S., Matsuoka, M., Nagato, Y. and Hirano, H.Y. (2004) The YABBY gene DROOPING LEAF regulates carpel specification and midrib development in Oryza sativa. Plant Cell 16, 500–509. Zhang, P.Y., Tan, H.T.W., Pwee, K.H. and Kumar, P.P. (2004) Conservation of class C function of floral organ development during 300 million years of evolution from gymnosperms to angiosperms. Plant Journal 37, 566–577.

7

Flower Senescence S. VERLINDEN Division of Plant and Soil Sciences, West Virginia University, PO Box 6108, Morgantown, WV 26506-6108, USA, e-mail: [email protected]

Introduction Flowers, no matter their size, shape, colour or structure, have one important function for plants: sexual reproduction. Sexual reproduction consists of several distinct developmental processes including pollen production, ovule formation, pollination, fusion of gametes and the development and dispersal of viable seeds. Many flowers have the additional function of attracting pollinators. The initial attraction between flowers and their pollinators can be the result of petal or corolla colour or shape, or chemical composition of the flower, whereas the end of a flower’s attraction to pollinators is the result of petal or corolla inrolling or closure, colour changes, changes in chemical composition of flower scent or nectar, and ultimately withering or abscission of the petals. Therefore, pollination and subsequent changes in the flower corolla can be seen as a way of communication between the flower and pollinators. Disposing of the petals or corollas has the additional benefit of decreasing the chance of herbivory of the developing ovary and seeds. These combined processes following pollination are often referred to as flower senescence. The opportunity for pollination depends on flower longevity, the time a flower remains open and functional. In order to maintain floral structures for considerable amounts of time a lot of resources have to be expended by the plant. Therefore, flower longevity can be understood in terms of a cost–benefit analysis. In other words, plants have evolved a mechanism for flowers to remain open and functional in order to facilitate pollen dissemination and ovule fertilization as long as the resources used by the plant to maintain a flower are less than the resources expended to construct a new flower. Flower senescence in this context will happen when it becomes more advantageous for the plant to construct a new flower – including renewed odds of getting pollinated – than to maintain an existing one (Ashman and Schoen, 1994).

150

ßCAB International 2006. The Molecular Biology and Biotechnology of Flowering, 2nd edn (ed. B.R. Jordan)

Flower Senescence

151

Flower senescence can therefore be defined as the events that lead to the death of flower parts signalling the end of an open and functional disseminator and/or receptor of pollen. It should be noted here that the most visible and largest flower part to senesce is the corolla or petals and that this organ will receive most attention in this chapter. However, other flower parts such as styles, anthers, and in the case of unpollinated flowers, ovaries, sepals and pedicels, will also in time senesce. Flower senescence or petal senescence in the context of this chapter will refer to a genetically based programme that leads to programmed cell death (PCD). PCD is an integral part of plant growth and development and can lead to the death of individual cells, tissue, organs and the whole plant. Petal senescence has been studied extensively as a type of PCD because of the distinctive morphological, physiological, biochemical and molecular changes associated with the demise of the petals (Rubinstein, 2000; Xu and Hanson, 2000; Wagstaff et al., 2003).

Categories of Flower Senescence Several categories of flower senescence have been used to describe how a flower senesces. Three categories can be distinguished in the literature: (i) age related versus pollination induced; (ii) ethylene sensitive versus ethylene insensitive; and (iii) abscised versus persistent petals. These are somewhat artificial divisions but are based on distinct morphological, biochemical or molecular changes associated with each type of senescence. Age related versus pollination induced Pollination is the most obvious external factor to affect flower longevity and as such this type of flower senescence is one of the most widely studied. Pollination leads to dramatic accelerated flower senescence in a number of commercially important plants. In some plants, however, pollination does not lead to flower senescence or may even prolong the life of the flower. The effect of pollination is most pronounced in long-lived flowers and seems absent from ephemeral flowers such as daylily and morning glory (Stead, 1992). Ethylene sensitive versus ethylene insensitive Flower senescence can be categorized as ethylene sensitive or ethylene insensitive based on the exogenous application of ethylene. Ethylene-sensitive species have flowers that react to exogenous ethylene by wilting, withering or abscising petals or corollas. This is thought to be an indication that endogenous ethylene plays a role in their natural age-related or pollination-induced senescence process (Woltering and van Doorn, 1988; van Doorn, 2001). These categories of senescence are for the most part consistent within plant families. Most monocotyledonous species are ethylene insensitive, whereas dicotyledonous species tend to be ethylene sensitive. However, exceptions to these generalizations can be found in both groups (van Doorn, 2001).

152

S. Verlinden

An additional category was established more recently and further divides ethylene-insensitive flower senescence into lipoxygenase-dependent and lipoxygenase-independent flower senescence. Alstroemeria was found to senesce without the accumulation of significant lipoxygenase activity and can therefore be distinguished from daylily, and other ethylene-insensitive senescence which is associated with a significant increase in lipoxygenase activity throughout development and senescence (Leverentz et al., 2002). Abscised versus persistent petals The rapid shedding of petals or corolla, referred to as petal or corolla abscission, occurs both in ethylene-sensitive and -insensitive flowers and should be distinguished from flower abscission where the whole flower is shed by the plant (van Doorn, 2001, 2002b). Most petal abscission is mediated by ethylene. Here, we will discuss mostly persistent petals or petals that abscise after most of the petal tissues have senesced since abscission is clearly different from senescence in both temporal and spatial aspects and therefore not within the scope of this chapter (Rubinstein, 2000).

From Anthesis to Senescence: Setting the Stage The development of the flower from anthesis to senescence often receives little attention. However, the internal changes that occur during this phase of development set the stage for flower senescence. It should be noted here that ephemeral flowers such as morning glory and daylily enter into senescence almost immediately following anthesis and therefore do not have the maintenance phase of flower development often observed in other species (Suttle and Kende, 1980; Bieleski and Reid, 1992; K. Smith and S. Verlinden, unpublished). Physiological changes Physiological changes that have been observed between anthesis and senescence include changes in the ultrastructure of flower tissues, fresh weight, dry weight and nutrient content. Ultrastructural observations of carnation petals indicate that vacuolar and cytoplasmic vesiculation starts in pre-climacteric petals (Paliyath and Thompson, 1990; Smith et al., 1992). However, most studies on ultrastructural changes have concluded that these changes are limited compared to changes observed in senescing tissues (Rubinstein, 2000). After the rapid development of the petals and corolla up to anthesis, associated with an enormous influx of water, carbohydrates and other solutes, the petals often continue an albeit smaller increase in fresh weight and dry weight of varying duration (Borochov and Woodson, 1989; Bieleski, 1993; Rubinstein, 2000; Verlinden, 2003). The growth phase is followed by a maintenance phase for longer-lived species or the immediate entry into a senescence phase for ephemeral flower species (Bieleski, 1993, 1995; K. Smith and

Flower Senescence

153

S. Verlinden, unpublished). The maintenance phase is associated with a gradual decline in fresh weight, dry weight, carbon content and phosphorus content (Verlinden, 2003; K. Smith and S. Verlinden, unpublished).

Biochemical changes Most of the biochemical changes that have been described from anthesis to senescence are related to membranes and their constituents. Plasmalemma protein content decreases in both rose and petunia during petal and corolla development (Borochov et al., 1994, 1997; Itzhaki et al., 1990). Throughout flower development, increases in lipid order and phase-transition temperature of lipids in rose petals have been observed with nuclear magnetic resonance (NMR) and X-ray diffraction measurement of isolated membranes (Faragher et al., 1987; Borochov et al., 1995; Itzhaki et al., 1995). These changes are associated with a decrease in phospholipids, fluidity and the ratio of unsaturated to saturated lipids (Suttle and Kende, 1980; Itzhaki et al., 1990; Leverentz et al., 2002). Sterol levels do not change throughout development and therefore lead to ever increasing sterol to phospholipids ratios (Itzhaki et al., 1990). Increased sterol to phospholipids ratios have been described before and were suggested to decrease fluidity throughout petal development (Thompson et al., 1982). Decreases in fluidity in turn are thought to lead to membrane malfunctions and ion leakage and ultimately senescence and death of the tissue. Other mechanisms responsible for the rigidification or decrease in fluidity of the membranes during petal development such as increased saturation of lipids (Fobel et al., 1987; Sylvestre and Paulin, 1987), possibly due to lipid peroxidation (Paulin et al., 1986; Paulin and Droillard, 1989), have been suggested. Indeed increasing levels of lipoxygenase activity and peroxidized lipids are observed from anthesis until early senescence in daylily (Rubinstein, 2000). Lipoxygenases have been correlated to decreases in membrane fluidity and inhibitors of lipoxygenase activity prevent these membrane-fluidity changes indicating that they may play an important role in events leading up to senescence (Fobel et al., 1987). However, at least one study showed that lipoxygenases may not be directly involved in senescence or the events leading up to it. Application of linoleic and linolenic acids, substrates for lipoxygenases, enhanced senescence and only unsaturated lipids showed this effect in orchid flowers. Interestingly, no increases in lipoxygenase activities were observed during pollination-induced senescence, and inhibitors of lipoxygenase activity did not delay senescence (Porat et al., 1995b). It has been suggested that other lipid oxidases may play a role in peroxidation of lipids in orchids and this may still lead to the same outcome (Rubinstein, 2000). In contrast, specific lipoxygenase activity increases prior to wilting in tulip and carnation and subsequently declines during wilting symptoms. Here, again the massive and catastrophic loss of water during senescence was not associated with lipoxygenase activity (Jones and McConchie, 1995). Taken together, these observations suggest that lipoxygenases may play a role, at least in some species, in

154

S. Verlinden

events leading up to senescence. Interestingly, acyl chain desaturation by desaturases has been suggested throughout development as a possible way to counteract the negative effects of an increasingly rigid plasma membrane (Brown et al., 1990). This in turn would present more possible substrate for lipoxygenases leading to enhanced catabolism. Clearly, lipid metabolism during petal development is very active and complex. A lipoxygenase-independent senescence mechanism has been described in Alstroemeria peruviana, somewhat similar to the observations made in orchids where increases in lipoxygenase activity were absent following pollination (Porat et al., 1995b; Leverentz et al., 2002). However, in this species lipoxygenase activity declined significantly throughout petal development and did not increase leading up to senescence. This is in contrast to some other ethylene-sensitive and -insensitive species where lipoxygenase activity peaks at early to midsenescent stages of development (Peary and Prince, 1990; Rouet-Mayer et al., 1992). In Alstroemeria, an ethylene-insensitive species, lipoxygenase activity could not be correlated with the loss of membrane semipermeability indicating that it falls into a new class of flower senescence (Leverentz et al., 2002). Studies on the metabolism of membrane lipids show that phospholipid catabolic activity does not change from early flower development through senescence, despite changes in lipoxygenase activity, and that all classes of phospholipids decline in parallel (Borochov et al., 1990b; Brown et al., 1991a,b). In addition, studies have shown that cells maintain the ability to biosynthesize lipids well into senescence (Borochov et al., 1990b). However, phospholipids decline threefold relative to sterols suggesting that some species of lipids are more susceptible to catabolism than others and that the pool of those substrates gets replenished. A preference in degradation was observed for phospholipids containing two di-unsaturated or at least one polyunsaturated acyl chain (Brown et al., 1987). In addition, acyl chain composition has been shown to determine if phospholipase D will catabolize phosphatidylcholine (Brown et al., 1990). Indeed, varying susceptibility to degradation based on head group and acyl chain composition has been demonstrated (Brown et al., 1991a,b). Especially lipids with unsaturated acyl chains are prone to degradation (Brown et al., 1987). An important role for calcium in membrane catabolic activity has been demonstrated (Leshem, 1987). Phospholipase D activity, for example, is enhanced by the application of a calmodulin antagonist leading to possible increased levels of cytosolic calcium in carnation (de Vrije and Munnik, 1997). Direct and indirect evidence also supports a role for calcium and calmodulin in the activation of phospholipases A, B and C. The ability of microsomal vesicles from carnation petals to maintain ATP-dependent uptake of calcium decreases throughout development. This suggests that the cell’s ability to maintain low calcium levels in the cytoplasm may be impaired and could possibly lead to phospholipase activation (Paliyath and Thompson, 1988). However, it should be noted that phospholipase D activity, and maybe others, decreases throughout development (Suttle and Kende, 1980), a process that could counteract the overall observed in vivo catabolic activity. Calcium has

Flower Senescence

155

also been shown to decrease membrane fluidity possibly by increasing the order of polar heads of membrane lipids; however, this ordering effect is limited to the membrane surface and therefore different from the reduced molecular motion associated with development and senescence (Drory et al., 1992). It should be clear from a number of studies that calcium has both positive and negative effects on the progression towards petal senescence, depending on location and concentration and that both of those are tightly regulated (Leshem, 1987). Action of phospholipase C results in the formation of diacylglycerol (DAG) and inositol 1,4,5-triphosphate (IP3), both implicated in secondary signalling (Leshem, 1987). Other phospholipases, such as phospholipase A, have also been implicated in the release of secondary messengers (Borochov et al., 1994). Further breakdown after lipoxygenase and phospholipase action results in a number of compounds including jasmonates that can elicit senescence (Halaba and Rudnicki, 1989; Porat et al., 1993, 1995b). DAG, the result of phospholipase activity has been shown to accumulate well before the ethylene climacteric in petunia. Application of the elicitor PMA, which acts similarly in kinase activation to DAG, showed that DAG has a role to play in enhancing senescence (Borochov et al., 1997). In addition, rapid turnover rates of polyphosphoinositides versus structural lipids in carnation flowers suggest that the catabolic products of phospholipases behave like typical signal precursors in flowers (Munnik et al., 1994).

Molecular changes Very little attention has been given to molecular changes from anthesis through senescence with the exception of one recent study (van Doorn et al., 2003). Most genes isolated from petals or corollas of ethylene-sensitive and -insensitive species have been cloned from early to mid-senescent plant materials (Lawton et al., 1989; Meyer et al., 1991; Wang et al., 1993; Fukuchi-Mizutani et al., 1995, 2000; Jones et al., 1995; Panavas et al., 1999). Therefore, most of the transcripts of these genes are found to accumulate during senescence and not in the events leading up to senescence. A few of these genes have shown developmental regulation independent of senescence in both ethylene-sensitive and -insensitive species, even though their highest transcript levels are observed during senescence. The putative beta-glucosidase and cysteine proteinase from carnation, SR5 and DCCP1, the putative desaturase from rose, RP4, and the putative aspartic proteinase and allene oxide synthase from daylily, DSA4 and DSA5, among others are all examples of this class of genes (Lawton et al., 1990; Jones et al., 1995; Panavas et al., 1999; Fukuchi-Mizutani et al., 2000). More recently, microarray analysis of tepal-specific genes of Iris showed that the expression patterns of isolated genes falls into three clusters. At least two of these clusters of genes showed increased expression during growth and development well before senescence set in, again suggesting that major changes occur during development well before senescence (van Doorn et al., 2003).

156

S. Verlinden

Summary Membranes are by far the most widely studied cell component during flower development from anthesis through senescence. Membranes are also one of the few cell structures that show significant changes leading up to senescence. These observations also suggest that they may play an important role in triggering flower senescence in both ethylene-sensitive and -insensitive petals and corollas. Flower senescence may therefore be due to the release of signalling molecules and/or changes in the lipid environment of a number of membrane-residing enzymes.

Timing of Senescence: the Opening Act Genotypes The fact that petal senescence is under genetic control is well established. Genotypes with varying flower longevity have been identified in Dianthus caryophyllus (Wu et al., 1989; Brandt and Woodson, 1992), Dianthus barbatus (Friedman et al., 2001), Antirrhinum majus (Schroeder and Stimart, 2001), Petunia  hybrida (Porat et al., 1993, Krahl and Randle, 1999), Lilium longiflorum (van der Meulen-Muisers et al., 1998), Gerbera  hybrida (Wernett et al., 1996) and Hatoria  graeseri (Easter cactus) (Karle and Boyle, 1999) among others. These studies showed significant additive gene effects for a number of species (Wernett, et al., 1996; Krahl and Randle, 1999; Schroeder and Stimart, 2001), suggesting that flower longevity in these species is under the control of at least two genes and in at least one case epistatic effects are absent (Schroeder and Stimart, 2001). More detailed studies of genotypes of ethylene-sensitive species showed an important role for the ability to produce ethylene (Wu et al., 1989; Brandt and Woodson, 1992) and/or respond to exogenous ethylene as factors affecting flower longevity (Brandt and Woodson, 1992; Porat et al., 1993; Friedman et al., 2001). However, other genes must play a role in flower longevity since genotypic variation in flower longevity was also observed in flowers that are ethylene insensitive and either have the ability to produce an ethylene climacteric (Friedman et al., 2001) or are naturally non-climacteric (Wernett et al., 1996; van der Meulen-Muisers et al., 1998; Krahl and Randle, 1999). The overall conclusion of these studies was that flower longevity can be significantly improved through classical breeding efforts (van der Meulen-Muisers et al., 1998) often because improvements in flower longevity have not been focused on before (Krahl and Randle, 1999).

Carbohydrates Significant changes in carbohydrate levels can be observed throughout the development of petals and corollas. In most petals and corollas, starch (Ho and Nichols, 1977; Ferreira et al., 1986; Tirosh and Mayak, 1988), sucrose

Flower Senescence

157

(Nichols 1973; van der Meulen-Muisers et al., 2001) and a number of other sugars (van der Meulen-Muisers et al., 2001) decline after anthesis in both cut and uncut flowers (Tirosh and Mayak, 1988; van Doorn, 2004). However, relatively high levels of sugars are maintained through senescence in several species (van Doorn, 2004). Exogenously applied sugars, especially sucrose and glucose, have long been shown to delay flower senescence (Nichols, 1973; Borochov and Woodson, 1989; van Doorn, 2001). It was widely thought that sugar treatments prolonged vase life by increasing levels of respiratory substrate to levels that allow for prolonged flower maintenance (Borochov and Woodson, 1989; van Doorn, 2001). However, the fact that in multiple species significant amounts of sugars remain in floral tissues at the end of senescence seems to contradict this hypothesis (Nichols, 1973; Bieleski, 1993; Eason et al., 1997; Ichimura, 1998; Ichimura and Suto, 1999; van der Meulen-Muisers et al., 2001; van Doorn, 2004). Several studies have shown that flowers treated with sucrose show decreased ethylene sensitivity (Mayak and Dilley, 1976; van Doorn, 2004; Verlinden and Garcia, 2004). 1-aminocyclopropane-1-carboxylic acid (ACC) synthase and ACC oxidase transcript accumulation is lower when pre-senescent sucrose-treated flowers are exposed to ethylene when compared to untreated control flowers. A direct effect on ethylene biosynthesis through a decrease in ethylene responsiveness of the transcription of ethylene biosynthetic genes has therefore been postulated (Verlinden and Garcia, 2004). Recent work in Arabidopsis thaliana showed a close interaction between glucose, sucrose and ethylene signalling pathways that supports these observations (Smeekens, 2000; Gibson et al., 2001; Leon and Sheen, 2003). We have some evidence that certain ethylene-signalling components are regulated by sucrose application to flowers. Carnation flowers treated continuously with sucrose showed a delay in transcript accumulation during petal development of at least one EIN3-like (EIL) gene (Iordachescu and Verlinden, 2005). EILs have been shown to be an integral part of ethylene signalling in a number of species (see below). Although this may explain the role for sucrose in ethylene-sensitive flower senescence, it is not clear how sugars prolong flower longevity, although to a lesser extent, in ethyleneinsensitive species. The suggested role of abscisic acid (ABA) in this type of senescence (Panavas et al, 1998b) and the link between glucose and ABA signalling established in Arabidopisis (Leon and Sheen, 2003) may provide answers to this question. However, as long as it is not clear where and how sugars accumulate, are mobilized and used within the flowers and cells, it will be difficult to pinpoint the specific role of sugars in triggering senescence (van Doorn, 2004). However, it is clear that changes in ethylene sensitivity mediated through changes in the ethylene signalling pathway play an important role (van Doorn, 2004; Verlinden and Garcia, 2004).

Circadian rhythms Several lines of evidence suggest that circadian rhythms are involved in the timing of petal or corolla senescence of a number of species. The precise time

158

S. Verlinden

of flower opening followed by a predetermined time span to petal senescence has been established for a number of plant species with ephemeral flowers, suggesting that plants use their internal clock to dictate timing of senescence (Bieleski and Reid, 1992). However, except for a few studies showing the effect of light on longevity and ethylene production, or a circadian rhythm in ethylene production, very little is known about the mechanism that may involve the plants’ internal clock in flower senescence. Interorgan communication A significant role in timing of senescence of petals and corollas has been attributed to the style (Gilissen, 1976; Lovell et al., 1987; Woltering et al., 1997) and the gynoecium (Sacalis and Lee, 1985; Woodson and Brandt, 1991; Shibuya et al., 2000; Nukui et al., 2004). Some studies report prolonged vase life (Shibuya et al., 2000; Nukui et al., 2004) while others observed faster senescence when the gynoecium is removed (Mor et al., 1980; Sacalis and Lee, 1985; Woodson and Brandt, 1991). Expression studies of the ethylene biosynthetic genes in flowers showing prolonged vase life indicate that repressed ethylene production in the gynoecium is a major determining longevity factor (Nukui et al., 2004). Applications of plant growth hormones that stimulate ethylene production in the gynoecium lead to shorter flower longevity. Removal of the gynoecium before the application of these substances prevented the earlier senescence symptoms indicating that the gynoecium may have a regulatory role to play in triggering flower senescence (Woodson and Brandt, 1991; Shibuya et al., 2000; Nukui et al., 2004). In addition, the ovary is the first flower organ to produce detectable levels of ethylene after the flower is exposed to ethylene (Jones, 2002). A role for styles in triggering age-related flower senescence has been suggested beyond pollination-induced senescence (Gilissen, 1976; Lovell et al., 1987; Woltering et al., 1997). The evidence here is based mostly on the fact that wounding or removal of the style can cause accelerated wilting independent of pollination. A healthy, undamaged stigma and style therefore seem essential in maximizing flower longevity (Lovell et al., 1987). Further investigations have shown that wound or stress ethylene is the most likely translocated signal in this type of accelerated flower senescence (Woltering et al., 1997). Interorgan communication in ethylene-insensitive flowers has not been studied as far as we can ascertain. Abiotic stresses A number of external stimuli can either enhance or delay petal or corolla senescence. In many cases the intensity and timing of these external stimuli dictate enhanced or delayed petal or corolla senescence. High temperatures (Hansen et al., 1992), water stress (Coker et al., 1985; Spikman, 1986), low light (Heo et al., 2004), UV light exposure (Borochov and Faragher, 1983), metal halide light exposure during production (Garello et al., 1995), high

Flower Senescence

159

relative humidity (Torre et al., 1999), wounding of the stigma (Woltering et al., 1997) and severing from the plant (Borochov et al., 1997) have all been implicated in decreasing flower longevity. On the other hand, brief exposure to high temperatures after harvest (Panavas et al., 1998a; Verlinden and Woodson, 1998), exposure to red light (Heo et al., 2004) and low relative humidity during production (Torre et al., 1999) delayed flower senescence. In most cases the exact mechanisms that lead to enhanced or delayed senescence are not known or highly speculative. High calcium levels and heat shock have apparent stabilizing effects and have been used to explain enhanced flower longevity of plants grown in low relative humidity environments and flowers exposed to high temperatures for brief periods of time, respectively. Red light from light-emitting diodes delays senescence whereas the extra red light from metal halide versus high pressure sodium lights enhanced petal senescence. No clear explanation was given in either study for the observed effect. In all cases where decreased flower longevity was observed, a role for ethylene or ABA was evoked.

Pollination Successful pollination is perhaps the most dramatic external factor affecting the timing of flower senescence and has received by far the most attention of all external factors affecting flower senescence. Pollination can reduce flower longevity in longer-lived flowers such as orchids by 90% or more. The effect is less dramatic or absent for short-lived flowers such as daylily or Tradescantia (Stead, 1992; Larsen et al., 1993). The two factors that determine successful pollination are pollen load and pollen compatibility (Gilissen, 1976, 1977; Stead and Moore, 1979; Hoekstra and van Roekel, 1986; Singh, et al., 1992; Stead, 1992; Kao and McCubbin, 1996). A minimum amount of pollen has to land on the stigma to induce senescence in the pollinated flower (Gilissen, 1977; Stead and Moore, 1979). Self-incompatibility results in unsuccessful pollination and therefore no senescence of the floral tissues. Therefore, recognition of self and non-self pollen, and the rejection of self is essential in pollination-induced senescence. Although the early events, generally a small ethylene peak produced by the style, are the same in either type of pollination, further ethylene production is non-existent in self-incompatible pollinations. This observation indicates that complex recognition and rejection signals beyond ethylene are at play (O’Neil and Nadeau, 1997).

Hormones All classical hormones, cytokinin, ABA, gibberellins, auxins and ethylene, have at one point or another been implicated in timing of flower senescence and their individual roles and possible interactions have not, by far, been settled. In ethylene-sensitive flowers, ABA has been shown to increase during flower development. However, major increases seem to be tied to ethylene

160

S. Verlinden

production and therefore most likely do not play a major role in triggering senescence (Hanley and Bramlage, 1989; Muller et al., 1999). Exogenous application of ABA and induction by applying water stress decreases flower longevity, but no cause and effect relationship has been established (Vardi and Mayak, 1989; Muller et al., 1999). ABA accumulation in pollinated transgenic cytokinin-overproducing petunia was much lower than wild-type plants, again suggesting that ABA is a primary response and not an initiator of senescence (Chang et al., 2003). In ethylene-sensitive flowers, on the other hand, ABA seems to play an important role in the signalling that leads to PCD (Panavas et al., 1998b). Cytokinin levels decrease throughout petal and corolla development suggesting a role for cytokinins in maintaining the fitness of the flower (van Staden, 1989). In addition, both the exogenous application and transgenic plants overproducing cytokinins have shown that cytokinins can delay petal and corolla senescence (Cook et al., 1985; Upfold and van Staden, 1990; Lukaszewska et al., 1994; Taverner et al., 1999; Chang et al., 2003). In ethylene-sensitive flowers cytokinins, both exogenous and endogenous, decrease ethylene sensitivity thereby delaying senescence (Chang et al., 2003; Cook et al., 1985). Cytokinins and concentration changes during development have not been studied in ethylene-insensitive flowers. Exogenous auxin applications have, in most cases, been shown to induce ethylene production thereby resulting in faster petal or corolla senescence (Nichols and Manning, 1986; Harel, et al., 1989; van Staden, 1995). No significant changes in the levels of auxins throughout flower development have been observed (Harel et al., 1989). However, some evidence suggests significant movement, interaction with cytokinins and changes in metabolism of auxins throughout flower development (Nichols and Manning, 1986; van Staden, 1995). None the less, no clear role in changing timing of senescence has been attributed to these changes. Gibberellin application to ethylene-sensitive flowers delays flower senescence and is only effective in delaying senescence when applied to young flowers (Saks et al., 1992; Saks and van Staden, 1993a,b). Limited evidence for changes in gibberellin content during flower development exists (Saks et al., 1992). However, it has been suggested that there is negative correlation between gibberellin content and petal senescence in carnation. No evidence was presented, however, to indicate a causal role for declining gibberellin content in triggering petal senescence.

Regulation of Senescence Age-related senescence Ethylene sensitive Ethylene-sensitive flower senescence has received most attention by far in studies of flower longevity, most likely because a number of highly valuable flower crops fall into this category (Reid and Wu, 1992). Ethylene

Flower Senescence

161

measurements are relatively easy to make, and the clear link between senescence and ethylene production in ethylene-sensitive plants has contributed to the increased attention paid to them (Borochov and Woodson, 1989; van Altvorst and Bovy, 1995). However, it should be clear that many flowers senesce without involvement of ethylene. The fact that the climacteric rise in ethylene plays a regulatory role in the events leading to the demise of the flowers of ethylene-sensitive species has been established for many years (Borochov and Woodson, 1989). Studies of the expression of genes in the ethylene biosynthetic pathway and senescencerelated (SR) genes have shown that they are clearly regulated by this ethylene and have confirmed the importance of ethylene in ethylene-sensitive flowers (Lawton et al., 1990; Woodson et al., 1992). In addition to the mandatory perception of ethylene to complete the senescence programme, a large increase in ethylene responsiveness has been observed in petals during flower development (Lawton et al., 1990, Verlinden et al., 2002). It is thought that this increase in responsiveness ultimately results in the ethylene climacteric. As ethylene responsiveness in petals increases during development, it presumably reaches a threshold at which the flower tissues respond to the low basal levels of ethylene production of the flower, triggering the ethylene climacteric (Lawton et al., 1990; Woodson et al., 1992; Verlinden et al., 2002). In this section we will discuss some of the recent work dealing with ethylene biosynthesis and signalling during development and age-related senescence of ethylene-sensitive flowers. Ethylene biosynthesis The early work on ethylene biosynthesis has been nicely reviewed by Borochov and Woodson (1989) and others (Reid and Wu, 1992; van Altvorst and Bovy, 1995). Briefly, a continued and increased induction of the ethylene biosynthetic genes, ACC synthase and ACC oxidase, during flower senescence results in autocatalytic or ‘self-induced’ ethylene. The autocatalytic ethylene is responsible for the climacteric rise in ethylene seen during flower senescence. Continued perception of ethylene is necessary to sustain the climacteric ethylene production and the expression of several SR genes (Borochov and Woodson, 1989; Lawton et al., 1990). Both ACC synthase and ACC oxidase are members of multigene families in all higher plants studied to date. Most studies have followed ACC synthase and ACC oxidase transcript levels in flower organs upon pollination (O’Neil et al., 1993; Tang and Woodson, 1996; Jones and Woodson, 1997; Bui and O’Neil, 1998; Jones and Woodson, 1999a,b; Lindstrom et al., 1999; LlopTous et al., 2000; Wang et al., 2001; Sanchez and Mariani, 2002; Weterings, et al., 2002) (see below for further discussion). Only a handful of studies have observed the patterns of mRNA accumulation of ACC synthase and ACC oxidase genes in flower tissues before and during age-related senescence (Woodson et al., 1992; ten Have and Woltering, 1997; Jones and Woodson, 1999b; Muller et al., 2000a; Jones, 2002). These experiments have led to the conclusion that several members of the ACC synthase and ACC oxidase gene families are differentially expressed in highly coordinated spatial and temporal

162

S. Verlinden

patterns. The observations in these studies also support an important role for the ovary in interorgan signalling in flower longevity. The ovary is the first organ that shows accumulation of ACC synthase transcript during age-related senescence (ten Have and Woltering, 1997) and is also the organ that shows increased ethylene responsiveness well before the petals and styles (ten Have and Woltering, 1997; Jones, 2002). Ethylene signalling As stated above, mandatory ethylene perception is crucial for the senescence programme to take effect in ethylene-sensitive flowers. Both the use of ethyleneaction inhibitors and more recently transgenic plants lacking the ability to perceive ethylene show delayed flower senescence supporting this observation (Borochov and Woodson, 1989; Wilkinson et al., 1997; Jones et al., 2001). In the last decade, great strides have been made in elucidating ethylene signalling in a number of model systems. A number of ethylene signalling components, among them ETR, ERS, CTR, EIN2, EIN3, EIL and EREBP genes have as a result become available to study flower senescence. Studies of the expression of some of these genes have confirmed early observations on ethylene signalling during flower senescence but also added new insights. Both ETR and ERS genes are ethylene-receptor genes cloned from Arabidopsis, and a number of other species, and their function has been extensively reviewed (Guo and Ecker, 2004). These genes are negative regulators of ethylene responses. In other words, ethylene exposure abolishes their kinase activity resulting in ethylene responses (Hua and Meyerowitz, 1998). Changes in transcript accumulation of ethylene receptors have been observed for carnation, rose, pea and Pelargonium (Orzaez et al., 1999; Dervinis et al., 2000; Muller et al., 2000a,b; Shibuya et al., 2002). The observations in these studies ranged from little to no regulation of mRNA levels for some ethylene receptors (Orzaez et al., 1999; Dervinis et al., 2000) to a significant decrease during senescence (Muller et al., 2000a; Shibuya et al., 2002) or increase in mRNA levels upon ethylene or ABA exposure (Muller et al., 2000a,b) in flower tissues. These observations point to complex regulation of ethylene-receptor expression that integrates a number of internal and external cues. More data are clearly needed not only at the level of expression of the genes but also their products and ability to bind ethylene in vivo. Interesting to note here is that no significant changes in ethylene binding were observed in carnation petals throughout development in earlier studies (Brown et al., 1986) and that at least one protein accumulates to similar levels throughout development, when petals are exposed to ethylene apparently independent of changes in ethylene responsiveness of the petals (Woodson, 1987). Even less is known about CTR1-like genes encoding the second step in ethylene signalling and also negative regulators of ethylene responses, and EIN2, a downstream positive regulator of ethylene signalling. To date EIN2like genes have not been studied in relationship to ethylene-sensitive flower senescence. In rose, two CTR1-like genes have been cloned and both show increased mRNA accumulation in response to ethylene exposure in flower

Flower Senescence

163

tissues. The expression of at least one of these CTR1-like genes seems to be upregulated during senescence (Muller et al., 2002). One of the last steps in ethylene signalling, encoded by EIN3 and EIL genes, are transcriptional activators of ethylene responses. In rose at least one EIL gene is constitutively expressed (Muller et al., 2003), while at least one EIL in carnation shows a decrease in mRNA levels during senescence (Waki et al., 2001). We have cloned three additional EIL genes from carnation and have shown significant developmental regulation of one of these EIL genes. DC-EIL3 is regulated developmentally and its mRNA levels are correlated with the increased responsiveness observed during petal development, suggesting that this gene could play a role in triggering the flower senescence programme (Iordachescu and Verlinden, 2005). Changes in mRNA levels of all three genes were observed upon pollination. These changes may play a role in changes in ethylene production and responsiveness that have been observed upon pollination in several species (Halevy and Whitehead, 1989; Whitehead and Halevy, 1989; Porat et al., 1994, 1995a; Halevy, 1995; Ketsa and Rugkong, 2000; Iordachescu and Verlinden, 2005). Carnation ERE–binding protein (CEBP) was cloned from carnation petals and shown to interact with an ethylene-response enhancer element (ERE) in the promoter of an SR gene, GST1, a putative glutathione-S-transferase (Maxson and Woodson, 1996). Interestingly, the ERE is in the same region that putatively binds the EILs. Since the expression of CEBP declines during development (Maxson and Woodson, 1996) and the mRNA levels of at least one EIL, DC-EIL3, increase (Iordaschescu and Verlinden, 2005), it is tempting to speculate that there may be a novel transcriptional mechanism at work in ethylene signalling during flower development and senescence. Ethylene and other gene expression Pioneering work by Woodson and co-workers resulted in the cloning of a number of SR genes of which mRNA accumulation is clearly regulated by ethylene (Lawton et al., 1989, 1990). These genes can be categorized in several ways. Some of the SR gene mRNAs accumulate only in response to ethylene whereas others accumulate based on developmental cues, but their expression is enhanced by endogenous or exogenous ethylene exposure of the tissue (Lawton et al., 1990). The first category includes SR8 and SR12, a putative glutathione-S-transferase and beta-galactosidase, respectively. The latter category includes genes such as SR5 and DCCP, a putative betaglucosidase and cysteine proteinase, respectively (Lawton et al., 1990; Jones et al., 1995). A different system, and possibly easier and more universally applicable, is to categorize the cloned genes by function or timing of expression (Rubinstein, 2000; van Doorn et al., 2003). Based on these qualifications, genes that are involved in membrane metabolism, protein metabolism, nucleic acid metabolism, cell wall metabolism, signal transduction, volatile production, defence and ethylene biosynthesis or signalling can be classified separately. A number of additional SR genes since the original work by Woodson and co-workers have been cloned from ethylene-sensitive species. However, it is remarkable how very few have been characterized when ethylene biosynthetic

164

S. Verlinden

and signalling genes are excluded from this list. Two cysteine proteinases, a desaturase, a lipase, a lipoxygenase and a defender against death are some of the additional genes documented (Fukuchi-Mizutani et al., 1995, 2000; Jones et al., 1995; Orzaez and Gannell, 1997; Hong et al., 2000; Sugawara et al., 2002; Chang et al., 2003). These genes can be easily classified in the previously described categories of SR or senescence-associated senescence. Biotechnology: controlling ethylene-regulated flower senescence Several successful attempts have been made to create transgenic plants with increased flower longevity. These attempts fall into two categories. The first category includes transgenic plants with altered ethylene biosynthesis (Savin et al., 1995; Aida et al., 1998). ACC oxidase has been the main target in these studies, although ACC synthase and ACC deaminase have been used successfully in creating transgenic plants showing altered fruit ripening (Klee and Clark, 2002). The second category consists of plants with altered ethylene perception (Wilkinson et al., 1997; Bovy et al., 1999; Shaw et al., 2002). Heterologous expression of mutated ethylene-receptor genes is the only approach that has been reported to date. In addition, many plants engineered for altered patterns of fruit ripening through decreased ethylene biosynthesis or ethylene responsiveness show extended flower longevity (Klee and Clark, 2002). This observation also points to one of the major drawbacks of the approaches used to date. Since the effect of ethylene is pleiotropic in nature, many unintended and often negative attributes are observed in transgenic plants. Poor rooting, delayed and poor fruit set, decreased pollen viability, root dry weight, seed weight and seed germination have all been observed in transgenic ethylene-insensitive plants (Clark et al., 1999; Gubrium et al., 2000; Klee and Clark, 2002; Clevenger et al., 2004). In addition, increased susceptibility to soil-borne pathogens has been noted in ethylene-insensitive plants (Knoester et al., 1998). It has been suggested that these negative effects can be overcome by directed expression of the transgenes to flower tissues (Klee and Clark, 2002). However, correct timing of the expression will also be essential to create the desired effect. In addition, downstream ethylene signalling components, such as EIN3, should be considered as targets of transgenic approaches since they only affect a subset of ethylene responses. Other approaches to delay flower senescence in plant species showing ethylene-regulated flower senescence may be on the horizon. In Arabidopsis flower longevity was extended by constitutive and tissue-specific expression of the MADS-domain factor AGL15. These studies showed that the effect of this gene was independent of ethylene effects in flowers (Fernandez et al., 2000; Fang and Fernandez, 2002). The observations suggest that timing of flower senescence may be regulated by factors other than ethylene biosynthesis and perception very similar to the observations of studies on the genetics of flower senescence (Friedman et al., 2001). Ethylene insensitive Significant work established daylily as an excellent model flower to study ethylene-insensitive petal or corolla senescence (Bieleski and Reid, 1992; Lay-Lee et al., 1992; Bieleski, 1993, 1995; Panavas and Rubenstein, 1998;

Flower Senescence

165

Panavas et al., 1998a,b). Daylily was shown to switch from a sink to a source in less than 12 h. Most soluble carbohydrates and amino acids and additional materials from proteins, nucleic acids and cell walls were transported out of the senescing tissue. Most of these materials wound up in newly developing buds. The rapidly senescing parenchyma in the petals was shown to be distinct from the vascular tissue, especially the phloem that largely escapes the senescence process (Bieleski and Reid, 1992; Bieleski, 1993). In addition, increases in hydrogen peroxide, lipid peroxidation, hydrolases, proteinases and RNase activity (Panavas and Rubinstein, 1998; Panavas et al., 1998a,b; Rubinstein, 2000), and decreases of the enzymatic components of the ubiquitin-proteosome pathway, catalase and ascorbate peroxidase (Panavas and Rubinstein, 1998; Stephenson and Rubinstein, 1998) have been observed. In recent years, tremendous strides have also been made in the molecular arena in understanding ethylene-insensitive flower senescence. A number of senescence-associated genes have been cloned and characterized from daylily (Valpuesta, et al., 1995; Guerrero et al., 1998; Panavas et al., 1999), Iris (van Doorn et al., 2003), daffodil (Hunter et al., 2002) and Alstroemeria (Wagstaff et al., 2001; Leverentz et al., 2002). These also fall in the above-described categories for SR or senescene-associated genes. However, Hunter et al. (2002) describe a transport function category, which includes genes such as nitrate and auxin transporters. This additional category should be included in a gene classification system, especially in light of the rapid export of materials in a number of ethylene-sensitive and -insensitive flowers during senescence. The coordinated events leading up to, and during, ethylene-insensitive senescence indicate tight control and suggest that an intricate interorgan, intraorgan and intracellular signalling mechanism is at work. However, no plant growth hormone has been shown to play the same crucial role as ethylene in ethylene-sensitive flower senescence. Abscisic acid is the most likely candidate for a similar but possibly lesser role in ethylene-insensitive flower senescence. In addition, a delicate interplay with other plant growth hormones should not be excluded as part of a complex regulatory mechanism. ABA has the ability to prematurely induce flower senescence in ethyleneinsensitive daylily petals. Increases in lipid peroxidation, membrane permeability, proteinase and RNase activities have been observed upon ABA treatment with a high degree of similarity to natural senescence. A decrease in RNA content is also observed earlier in ABA-treated petals than naturally senescing petals. Changes in the mRNA population of ABA-treated flowers also mimic changes observed during natural petal senescence, although the changes appear approximately 24 h earlier. In addition, transcript levels of five distinct genes increase earlier in response to ABA treatment (Panavas et al., 1998b, 1999). Transcriptional regulation by ABA of senescence – enhanced or – associated genes in other ethylene-insensitive flowers has to date not been reported. Gibberellins are the only other plant growth hormone with a putative role in ethylene-insensitive flower senescence. The transcript levels of two thiol proteases increased prematurely, compared to natural senescence upon exposure to gibberellic acid, indicating a possible role for gibberellins in the events leading up to petal senescence (Guerrero et al., 1998).

166

S. Verlinden

Pollination-induced senescence Early work on pollination-induced senescence has been reviewed in detail by Stead (1992) and additional insight in the occurrence and significance of pollination-induced senescence and abscission has been provided by van Doorn (1997). A number of others have contributed significantly to the field of post-pollination development and detailed models of post-pollination interorgan signalling exist (O’Neil and Nadeau, 1997; Bui and O’Neil, 1998; Jones and Woodson, 1999b). Here, we want to focus on events triggered by pollination that lead to senescence and that are different from events during agerelated senescence. Pollination signals Pollination-induced senescence is the result of highly coordinated events that start with pollen germinating on the stigma and result in changes in the ovary and petals before the pollen tubes reach the ovary. This observation clearly indicates a role for rapidly moving signals. Both ACC and auxin, found in relatively large quantities in pollen, have been suggested as a primary signal that elicits ethylene production in the stigma. An initial burst of ethylene promotes the early phase of pollen tube growth and is present in both compatible and incompatible pollination events (Holden et al., 2003). However, a second burst of ethylene or a third as seen in carnation (Jones and Woodson, 1997) that leads to petal senescence occurs only upon compatible pollination events. The first burst of ethylene, possibly initiated by auxin, is not causally involved in triggering the second or third larger ethylene production event (Singh et al., 1992; Holden et al., 2003). It is clear from a number of species that ethylene, and possibly but to a lesser extent ACC from this second or third ethylene peak, is the signal that is propagated through the flower and leads to senescence of the perianth (Porat et al., 1993, 1995a; Jones and Woodson, 1997; O’Neil and Nadeau, 1997; Jones and Woodson, 1999b). However, the nature of the pollen factor that induces the second and/ or third burst of signalling ethylene has not been unequivocally established (Porat et al., 1998) but may well be related to the compatibility reaction that distinguishes between self and non-self. Short-chain fatty acids and electrical pulses have also been implicated as signals in the changes following pollination and may be candidates for secondary signals induced by pollen (Woltering et al., 1997). Changes upon pollination Although petal senescence is the ultimate result of a compatible pollination event, and in that respect very similar to age-related senescence, there are changes associated with pollination uniquely related to this developmental process. The rapid increase in ethylene sensitivity that has been observed following pollination in several species is one of those unique characteristics (Halevy and Whitehead, 1989; Whitehead and Halevy, 1989; Porat et al., 1994, 1995a,b; Halevy, 1995a,b; Ketsa and Rugkong, 2000). The increased ethylene responsiveness associated with pollination and independent of ethylene

Flower Senescence

167

production observed upon pollination is thought to result in enhanced flower senescence. The increase in EIL transcript levels observed upon pollination in styles indicates that the increase in sensitivity to ethylene may be mediated by changes in ethylene signalling components (Iordachescu and Verlinden, 2005). A number of compounds in the pollen have been looked at as the putative ‘sensitivity factor’. However, to date, no conclusive evidence has been presented for any compound to unequivocally claim the title of ‘sensitivity factor’.

Concluding Remarks Clearly, great strides have been made in our understanding of flower senescence in the last two decades. As the picture of flower senescence grows even more complex and divergent forms of flower senescence are discovered, certain overriding similarities are noteworthy. First, the rapid conversion from sink to source is apparently a very common theme in both ethylene-sensitive and -insensitive flowers that do not senesce through rapid abscission. Second, gene expression in all types of flower senescence can be classified in similar functional categories. Third, changes in membrane composition and function are found in all types of flower development and senescence studied to date. Despite this remarkable progress a number of processes are not well understood and need additional attention. The use of flower senescence mutants and transgenic plants with altered SR gene expression will certainly help solidify our understanding and open up new areas of research in flower senescence. The use of easily transformed and genetically well-mapped model plants would definitely aid in this effort.

References Aida, R., Yoshida, T., Ichimura, K., Goto, R. and Shibata, M. (1998) Extension of flower longevity in transgenic torenia plants incorporating ACC oxidase transgene. Plant Science 138, 91– 101. Ashman, T. and Schoen, D.J. (1994) How long should flowers live? Nature 371, 788–790. Bieleski, R.L. (1993) Fructan hydrolysis drives petal expansion in the ephemeral daylily flower. Plant Physiology 103, 213–219. Bieleski, R.L. (1995) Onset of phloem export from senescent petals of daylily. Plant Physiology 109, 557–565. Bieleski, R.L. and Reid, M.S. (1992) Physiological changes accompanying

senescence in the ephemeral daylily flower. Plant Physiology 98, 1042– 1049. Borochov, A. and Faragher, J. (1983) Comparison between ultraviolet irradiation and ethylene effects on senescence parameters in carnation flowers. Plant Physiology 71, 536–540. Borochov, A. and Woodson, W.R. (1989) Physiology and biochemistry of flower petal senescence. Horticultural Reviews 11, 15–43. Borochov, A., Drori, A., Tirosh, T., Borochov-Neori, H. and Mayak, S. (1990a) Quantitative and qualitative changes in membrane proteins during petal senescence. Journal of Plant Physiology 136, 203–207.

168

S. Verlinden Borochov, A., Itzhaki, H. and Mayak, S. (1990b) Age-related changes in petal membranes from attached and detached rose flowers. Plant Physiology 94, 1233–1236. Borochov, A., Cho, M.H. and Boss, W.F. (1994) Plasma membrane lipid metabolism of petunia petals during senescence. Physiologia Plantarum 90, 279–284. Borochov, A., Itzhaki, H., Mayak, S. Davis, J.H. and Pauls, K.P. (1995) Deuterium magnetic resonance studies of senscence-related changes in the physical properties of rose petal membrane lipids. Plant Physiology 108, 1029–1033. Borochov, A., Spiegelstein, H. and Philosoph-Hadas, S. (1997) Ethylene and flower petal senescence: interrelationship with the membrane lipid catabolism. Physiologia Plantarum 100, 606–612. Bovy, A.G, Angenent, G.C., Dons, H. and van Altvorst, A. (1999) Heterologous expression of the Arabidopsis etr1–1 allele inhibits the senescence of carnation flowers. Molecular Breeding 5, 301–308. Brandt, A.S. and Woodson, W.R. (1992) Variation in flower senescence and ethylene biosynthesis among carnations. HortScience 27, 1100–1102. Brown, J.H., Legge, R.L., Sisler, E.C., Baker, J.E. and Thompson, J.E. (1986) Ethylene binding to senescing carnation petals. Journal of Experimental Botany 37(177), 526–534. Brown, J.H., Lynch, D.V. and Thompson, J.E. (1987) Molecular species specificity of phospholipid breakdown in microsomal membranes of senescing carnation flowers. Plant Physiology 85, 679–683. Brown, J.H., Paliyath, G. and Thompson, J.E. (1990) Influence of acyl chain composition on the degradation of phosphatidylcholine by phospholipase D in carnation microsomal membranes. Journal of Experimental Botany 41(229), 979–986.

Brown, J.H., Chambers, J.A. and Thompson, J.E. (1991a) Acyl chain and head group regulation of phospholipids catabolism in senescing carnation flowers. Plant Physiology 95, 909– 916. Brown, J.H., Chambers, J.A. and Thompson, J.E. (1991b) Distinguishable patterns of phospholipids susceptibility to catabolism in senescing carnation petals. Phytochemistry 30(8), 2537–2543. Bui, A.Q. and O’Neill, S.D. (1998) Three 1-aminocyclopropane-1-carboxylate synthase genes regulated by primary and secondary pollination signals in orchid flowers. Plant Physiology 116, 419–428. Chang, H., Jones, M.L., Banowetz, G.M. and Clark, D.G. (2003) Overproduction of cytokinins in Petunia flowers with PSAG12-IPT delays corolla senescence and decreases sensitivity to ethylene. Plant Physiology 132, 2174– 2183. Clark, D.G., Gubrium, E.K., Barrett, J.E., Nell, T.A. and Klee, H.J. (1999) Root formation in ethylene-insensitive plants. Plant Physiology 121, 53–59. Clevenger, D.J., Barrett, J.E., Klee, H.J. and Clark, D.G. (2004) Factors affecting seed production in transgenic ethylene-insensitive petunias. Journal of the American Society for Horticultural Science 129, 401–406. Coker, T., Mayak, S. and Thompson, J.E. (1985) Effect of water stress on ethylene production and on membrane microviscosity in carnation flowers. Scientia Horticulturae 27, 317–324. Cook, D., Rasche, M. and Eisinger, W. (1985) Regulation of ethylene biosynthesis and action in cut carnation flower senescence by cytokinins. Journal of the American Society for Horticultural Science 110, 24–27. Dervinis, C., Clark, D.G., Barrett, J.E. and Nell, T.A. (2000) Effect of pollination and exogenous ethylene on accumulation of ETR1 homologue transcripts during flower petal abscis-

Flower Senescence sion in geranium (Pelargonium  hortorum L.H. Bailey). Plant Molecular Biology 42, 847–856. de Vrije, T. and Munnik, T. (1997) Activation of phospholipase D by calmodulin antagonists and mastoparan in carnation petals. Journal of Experimental Botany 48(314), 1631–163. Drory, A., Thompson, J.E., Mayak, S. and Borochov, A. (1992) Carnation petal membranes: a study of surface properties using dansyl pyrrolidine, a new fluorescent probe. Journal of Plant Physiology 140, 544–549. Eason, J.R., Heyes, J.A., Somerfield, S.D. and de Vre, L.A. (1997) Physiological changes associated with Sandersonia aurantiaca flower senescence in response to sugar. Postharvest Biology and Technology 12, 43–50. Fang, S. and Fernandez, D.E. (2002) Effect of regulated overexpression of the MADS domain factor AGL15 on flower senescence and fruit maturation. Plant Physiology 130, 78–89. Faragher, J.D., Wachtel, E. and Mayak, S. (1987) Changes in the physical state of membrane lipids during senescence of rose petals. Plant Physiology 83, 1037–1042. Fernandez, D.E., Heck, G.R., Perry, S.E., Patterson, S.E., Bleecker, A.B. and Fang, S. (2000) The embryo MADS domain factor AGL15 acts postembryonically: inhibition of perianth senescence and abscission via constitutive expression. Plant Cell 12, 183–197. Ferreira, D.I., van der Merwe, J.J. and de Swardt, G.H. (1986) Starch metabolism in flowers of senescing gladioli inflorescences. Acta Horticultura 177, 203–210. Fobel, M., Lynch, D.V. and Thompson, J.E. (1987) Membrane deterioration in senescing carnation flowers. Plant Physiology 85, 204–211. Friedman, H., Barak, A., Umiel, N., Hagiladi, A. and Resnick, N. (2001) Ethylene-insensitive related phenotypes exist naturally in a genetically

169 variable population of Dianthus barbatus. Theoretical and Applied Genetics 103, 282–287. Fukuchi-Mizutani, M., Savin, K., Cornish, E., Tanaka, Y., Ashikari, T., Kusumi, T. and Murata, N. (1995) Senescenceinduced expression of a homologue of D9 desaturase in rose petals. Plant Molecular Biology 29, 627–635. Fukuchi-Mizutani, M., Ishiguro, K., Nakayama, T., Utsunomiya, Y., Tanaka, Y., Kusumi, T. and Ueda, T. (2000) Molecular and functional characterization of a rose lipoxygenase cDNA related to flower senescence. Plant Science 160, 129–137. Garello, G., Menard, C., Dansereau, B. and Le Page-Degivry, M. (1995) The influence of light quality on rose flower senescence: involvement of abscisic acid. Plant Growth Regulation 16, 135–139. Gibson, S.I., Laby, R.J. and Kim, D. (2001) The sugar sensitive1 (sis1) mutant of Arabidopsis is allelic to ctr1. Biochemical and Biophysical Research Communications. 280, 196–203. Gilissen, L. (1976) The role of the style as a sense-organ in relation to wilting of the flower. Planta 131, 201–202. Gilissen, L. (1977) Style-controlled wilting of the flower. Planta 133, 275–280. Gubrium, E.K., Clevenger, D.J., Clark, D.G., Barrett, J.E. and Nell, T.A. (2000) Reproduction and horticultural performance of transgenic ethyleneinsensitive petunias. Journal of the American Society for Horticultural Science 125, 277–281. Guo, H. and Ecker, J.R. (2004) The ethylene signaling pathway: new insights. Current Opinion in Plant Biology 7, 40–49. Guerrero, C., de la Calle, M., Reid, M.S. and Valpuesta, V. (1998) Analysis of the expression of two thiolprotease genes from daylily (Hemerocallis spp.) during flower senescence. Plant Molecular Biology 36, 565–571.

170

S. Verlinden Halaba, J. and Rudnicki, RM. (1989) Jasmonate stimulates ethylene emanation by carnation flowers. Acta Horticulturae 251, 65–66. Halevy, A.H. (1995) The role of sensitivity to ethylene in pollination-induced corolla senescence syndrome. Acta Horticulturae 405, 210–215. Halevy, A.H. and Whitehead, C.S. (1989) Pollination-induced corolla abscission and senescence and the role of shortchain fatty acids in the process. In: Osborne, D.J. and Jackson, M.B. (eds) Cell Separation in Plants. NATO ASI Series, Vol. H35, pp. 221–232. Hanley, K.M. and Bramlage, W.J. (1989) Endogenous levels of abscisic acid in aging carnation flower parts. Journal of Plant Growth Regulation 8, 225–236. Hansen, J.D., Hara, A.H. and Tenbrink, V.L. (1992) Vapor heat: a potential treatment to disinfest tropical cut flowers and foliage. HortScience 27, 139–143. Harel, R., Mayak, S. and Riov, J. (1989) Involvement of auxin in the control of senescence in Petunia flowers. Acta Horticulturae 261, 113–117. Heo, J.W., Chakrabarty, D. and Paek, K.Y. (2004) Longevity and quality of cut ‘Master’ carnation and ‘Red Sandra’ rose flowers as affected by red light. Plant Growth Regulation 42, 169–174. Ho, L.C. and Nichols, R. (1977) Translocation of 14C-sucrose in relation to changes in carbohydrate content in rose corollas cut at different stages of development. Annals of Botany 41, 227–242. Hoekstra, F.A. and van Roekel, T. (1986) Effects of interspecific pollination on stylar ethylene production and flower longevity in Petunia hybrida. Acta Horticulturae 181, 105–112. Holden, M.J., Marty, J.A. and SinghCundy, A. (2003) Pollination-induced ethylene promotes the early phase of pollen tube growth in Petunia inflata.

Journal of Plant Physiology 160, 261–269. Hong, Y., Wang, T., Hudak, K.A., Schade, F., Froese, C.D. and Thompson, J.E. (2000) An ethylene-induced cDNA encoding a lipase expressed at the onset of senescence. Proceedings of the National Academy of Sciences USA 97, 8717–8722. Hua, J., Meyerowitz and E.M. (1998) Ethylene responses are negatively regulated by a receptor gene family in Arabidopsis. Cell 94, 261–271. Hunter, D.A., Steele, B.C. and Reid, M.S. (2002) Identification of genes associated with perianth senescence in daffodil (Narcissus pseudonarcissus L. ‘Dutch Master’). Plant Science 163, 13–21. Ichimura, K. (1998) Improvement of postharvest life in several cut flowers by the addition of sucrose. Japanese Agricultural Research Quarterly 32, 275– 280. Ichimura, K. and Suto, K. (1999) Effects of the time of sucrose treatment on vase life, soluble carbohydrate concentrations, and ethylene production in cut sweet pea flowers. Plant Growth Regulation 28, 117–122. Iordachescu, M. and Verlinden, S. (2005) Transcriptional regulation of three EIN3-like genes of carnation (Dianthus caryophyllus L.) during flower development and upon wounding, pollination, and ethylene exposure. Journal of Experimental Botany 56 (418), 2011–2018. Itzhaki, H., Borochov, A. and Mayak, S. (1990) Age-related changes in petal membranes from attached and detached rose flowers. Plant Physiology 94, 1233–1236. Itzhaki, H., Davis, J.H., Borochov, A., Mayak, S. and Pauls, P.K. (1995) Deuterium magnetic resonance studies of senescence-related changes in the physical properties of rose petal membrane lipids. Plant Physiology 108, 1029–1033. Jones, M. (2002) Ethylene responsiveness in carnation styles is associated with

Flower Senescence stigma receptivity. Sexual Plant Reproduction 15, 107–112. Jones, M.L. and Woodson, W.R. (1997) Pollination-induced ethylene in carnation. Plant Physiology 115, 205–212. Jones, M.L. and Woodson, W.R. (1999a) Differential expression of three members of the 1-aminocyclopropane1-carboxylate synthase gene family in carnation. Plant Physiology 119, 755–764. Jones, M.L. and Woodson, W.R. (1999b) Interorgan signaling following pollination in carnations. Journal of the American Society for Horticultural Science 124, 598–604. Jones, M.L., Larsen P.B. and Woodson, W.R. (1995) Ethylene-regulated expression of a carnation cysteine proteinase during flower petal senescence. Plant Molecular Biology 28, 505–512. Jones, M.L., Kim, E. and Newman, S.E. (2001) Role of ethylene and 1-MCP in flower development and petal abscission in zonal geraniums. HortScience 36, 1305–1309. Jones, R. and McConchie, R. (1995) Characteristics of petal senescence in a non-climacteric cut flower. Acta Horticulturae 405, 216-–223. Kao, T. and McCubbin, A.G. (1996) How flowering plants discriminate between self and non-self pollen to prevent inbreeding. Proceedings of the National Academy of Sciences USA 93, 12059–12065. Karle, R. and Boyle, T.H. (1999) Relationships between floral morphology, breeding behavior and flower longevity in Easter cactus. Journal of the American Society for Horticultural Science 124, 296–300. Ketsa, S. and Rugkong, A. (2000) Ethylene production, senescence and ethylene sensitivity of Dendrobium ‘Pompadour’ flowers following pollination. Journal of Horticultural Science and Biotechnology 75, 149–153. Klee, H.J. and Clark, D.G. (2002) Manipulation of ethylene synthesis and

171 perception in plants: the ins and outs. HortScience 37, 450–452. Knoester, M., Bol, J.F., Linthorst, H.J.M. Hennig, J., van Loon, L.C. and van den Heuvel, J. (1998) Ethylene-insenstive tobacco lacks nonhost resistance against soil-borne fungi. Proceedings of the National Academy of Sciences USA 95, 1933–1937. Krahl, K.H. and Randle, W.M. (1999) Genetics of floral longevity in Petunia. HortScience 34, 339–340. Larsen, P.B. Woltering, E.J. and Woodson, W.R. (1993) Ethylene and interorgan signaling in flowers following pollination. In: Raskin, I. and Schultz, J. (eds) Plant Signals in Interactions with Other Organisms. American Society of Plant Physiologists, Rockville, Maryland, pp. 171–181. Lawton, K.A., Huang, B., Goldsbrough, P.B. and Woodson, W.R. (1989) Molecular cloning and characterization of senescence-related genes from carnation flower petals. Plant Physiology 90, 690–696. Lawton, K.A., Raghothama, K.G., Goldsbrough, P.B. and Woodson, W.R. (1990) Regulation of senescenceelated gene expression in carnation flower petals by ethylene. Plant Physiology 93, 1370–1375. Lay-Lee, M., Stead, A.D. and Reid, M.S. (1992) Flower senescence in daylily (Hemerocallis). Physiologia Plantarum 84, 308–314. Leon, P. and Sheen, J. (2003) Sugar and hormone connections. Trends in Plant Science 8, 110–116. Leshem, Y.Y. (1987) Membrane phospholipid catabolism and Ca2þ activity in control of senescence. Physiologia Plantarum 69, 551–559. Leverentz, M.K., Wagstaff, C., Rogers, H.J., Stead, A.D., Chanasut, U., Silkowski, H., Thomas, B., Weichert, H., Feussner, I. and Griffiths, G. (2002) Characterization of a novel lipoxygenase-independent senescence mechanism in Alstroemeria peruviana

172

S. Verlinden floral tissue. Plant Physiology 130, 273–283. Lindstrom, J.T. Woodson, W.R., Jones, M.L. and Lei, C.H. (1999) Accumulation of 1-aminocyclopropane-1-carboxylic acid (ACC) in Petunia pollen is associated with expression of a pollenspecific ACC synthase late in development. Journal of the American Society for Horticultural Science 124, 145–151. Llop-Tous, I., Barry, C.S. and Grierson, D. (2000) Regulation of ethylene biosynthesis in response to pollination in tomato flowers. Plant Physiology 123, 971–978. Lovell, P.J., Lovell, P.H. and Nichols, R. (1987) The control of flower senescence in petunia (Petunia hybrida). Annals of Botany 60, 49–59. Lukaszewska, A.J., Bianco, J., Barthe, P. and LePage-Degivry, M.T. (1994) Endogenous cytokinins in rose petals and the effect of exogenously applied cytokinins on flower senescence. Plant Growth Regulation 14, 119– 126. Maxson, J.M. and Woodson, W.R. (1996) Cloning of a DNA-binding protein that interacts with the ethylene-responsive enhancer element of the carnation GST1 gene. Plant Molecular Biology 31, 751–759. Mayak, S. and Dilley, D.R. (1976) Effect of sucrose on response of cut carnation (Dianthus caryophyllus) to kinetin, ethylene, and abscisic acid. Journal of the American Society for Horticultural Science 101, 583–585. Meyer, R.C., Woodson, W.R. and Goldsbrough, P.B. (1991) An ethyleneresponsive flower senescence-related gene from carnation encodes a protein homolgous to glutathione-stransferases. Plant Molecular Biology 17, 277–281. Mor, Y. Porat, D. and Halevy, A.H. (1980) Characterization of the light reaction in promoting the mobilizing ability of rose shoot tips. Plant Physiology 66, 996–1000.

Muller, R., Stummann, B., Andersen, A.S. and Serek, M. (1999) Involvement of ABA in postharvest life of miniature potted roses. Plant Growth Regulation 29, 143–150. Muller, R., Lind-Iversen, S., Stummann, B. and Serek, M. (2000a) Expression of genes for ethylene biosynthetic enzymes and an ethylene receptor in senescing flowers of miniature potted rose. Journal of Horticultural Science Biotechnology. 75, 12–18. Muller, R., Stummann, B. and Serek, M. (2000b) Characterization of an ethylene receptor family with differential regulation in rose (Rosa hybrida L.) flowers. Plant Cell Reports 19, 1232–1239. Muller, R., Owen, C.A., Xue, Z., Welander, M. and Stummann, B. (2002) Characterization of two CTR-like protein kinases in Rosa hybrida and their expression during flower senescence and in response to ethylene. Journal of Experimental Botany 53(371), 1223–1225. Muller, R., Owen, C.A., Xue, Z., Welander, M. and Stummann, B. (2003) The transcription factor EIN3 is constitutively expressed in miniature roses with differences in postharvest life. Journal of Horticultural Science Biotechnology 78, 10–14. Munnik, T., Musgrave, A. and de Vrije., T. (1994) Rapid turnover of polyphosphoinositides in carnation flower petals. Planta 193, 89–98. Nichols, R. (1973) Senescence of the cut carnation flower: respiration and sugar status. Journal of Horticultural Science 48, 111–121. Nichols., R. and Manning, K. (1986) Growth substances and post harvest flower senescence. Acta Horticulturae 181, 161–167. Nukui, H., Kudo, S., Yamashita, A. and Satoh, S. (2004) Repressed ethylene production in the gynoecium of longlasting flowers of the carnation ‘White Candle’: role of the gynoecium in carnation flower senescence. Journal of

Flower Senescence Experimental Botany 55(397), 641– 650. O’Neil, S.D. and Nadeau, J.A. (1997) Postpollination flower development. Horticultural Reviews 19, 1–44. O’Neil, S.D., Nadeau, J.A., Zhang, X.S., Bui, A.Q. and Halevy, A.H. (1993) Interorgan regulation of ethylene biosynthetic genes by pollination. Plant Cell 5, 419–432. Orzaez, D. and Granell, A. (1997) The plant homologue of the defender against apoptotic death gene is downregulated during senescence of flower petals. FEBS Letters 404, 275–278. Orzaez, D., Blay, R. and Granell, A. (1999) Programme of senescence in petals and carpels of Pisum sativum L. flowers and its control by ethylene. Planta 208, 220–226. Paliyath, G. and Thompson, J.E. (1988) Senescence-related changes in ATPdependent uptake of calcium in microsomal vesicles from carnation petals. Plant Physiology 88, 295–302. Paliyath, G. and Thompson, J.E. (1990) Evidence of early changes in membrane structure during postharvest development of cut carnation (Dianthus caryophyllus L.) flowers. New Phytologist 114, 555–562. Panavas, T. and Rubinstein, B. (1998) Oxidative events during programmed cell death of daylily (Hemerocallis hybrid) petals. Plant Science 133, 125–138. Panavas, T., Reid, P.D. and Rubinstein, B. (1998a) Programmed cell death of daylily petals: activities of wall-based enzymes and effects of heat shock. Plant Physiology Biochemical 36, 379–388. Panavas, T., Walker, E.L. and Rubinstein, B. (1998b) Possible involvement of abscisic acid in senescence of daylily petals. Journal of Experimental Botany 49(329), 1987–1997. Panavas, T., Pikula, A., Reid, P.D., Rubinstein, B. and Walker, E.L. (1999) Identification of senescence-associated genes from daylily petals. Plant Molecular Biology 40, 237–248.

173 Paulin, A. and Droillard, M.J. (1989) Membrane lipids peroxidation during the senescence of cut carnation (Dianthus caryophyllus L.) Acta Horticulturae 261, 51–57. Paulin, A., Droillard, M.J. and Bureau, J.M. (1986) Effect of a free radical scavenger, 3,4,5-trichlorophenol, on ethylene production and on changes in lipids and membrane integrity during senescence of petals of cut carnations (Dianthus caryophyllus). Physiologia Plantarum 67, 465– 471. Peary, J.S. and Prince, T.A. (1990) Floral lipoxygenase activity during senescence and inhibition by phenidone. Journal of the American Society for Horticultural Science 115, 455–457. Porat, R., Reuveny, Y., Borochov, A. and Halevy, A.H. (1993) Petunia flower longevity: the role of sensitivity to ethylene. Physiologia Plantarum 89, 291–294. Porat, R., Borochov, A., Halevy, A.H. and O’Neill, S.D. (1994) Pollinationinduced senescence of Phalaenopsis petals. Plant Growth Regulation 15, 129–136. Porat, R., Halevy, A.H., Serek, M. and Borochov, A. (1995a) An increase in ethylene sensitivity following pollination is the intitial event triggering an increase in ethylene production and enhanced senescence of Phalaenopsis orchid flowers. Physiologia Plantarum 93, 778–784. Porat, R., Reiss, N., Atzorn, R., Halevy, A.H. and Borochov, A. (1995b) Examination of the possible involvement of lipoxygenase and jasmonates in pollination-induced senescence of Phalaenopsis and Dendrobium orchid flowers. Physiologia Plantarum 94, 205–210. Porat, R., Nadeau, J.A., Kirby, J.A., Sutter, E.G. and O’Neill, S.D. (1998) Characterization of the primary pollen signal in the postpollination syndrome of Phalaenopsis flowers. Plant Growth Regulation 24, 109–117.

174

S. Verlinden Reid, M.S. and Wu, M.J. (1992) Ethylene and flower senescence. Plant Growth Regulation 11, 37–43. Rouet-Mayer, M.A., Bureau, J.M. and Lauriere, C. (1992) Identification and characterization of lipoxygenase isoforms in senescing carnation petals. Plant Physiology 98, 971–978. Rubinstein, B. (2000) Regulation of cell death in flower petals. Plant Molecular Biology 44, 303–318. Sacalis, J.N. and Lee, J.S. (1985) Effects of gynoecium excision on vase life of cut carnation flowers. Acta Horticulturae 167, 141–145. Saks, Y. and van Staden, J. (1993a) Effect of gibberellic acid on ACC content, EFE activity and ethylene release by floral parts of the senescing carnation flower. Plant Growth Regulation 12, 99–104. Saks, Y. and van Staden, J. (1993b) Evidence for the involvement of gibberellins in developmental phenomena associated with carnation flower senescence. Plant Growth Regulation 12, 105–110. Saks, Y., van Staden, J. and Smith, M.T. (1992) Effect of gibberellic acid on carnation flower senescence: evidence that the delay in flower senescence by gibberellic acid depends on the stage of flower development. Plant Growth Regulation 11, 45–51. Sanchez, A.M. and Mariani, C. (2002) Expression of the ACC synthase and ACC oxidase coding genes after selfpollination and incongruous pollination of tobacco pistils. Plant Molecular Biology 48, 351–359. Savin, K.W., Baudinette, S.C., Graham, M.W., Michael, M.Z., Nugent, G.D., Lu, C., Chandler, S.F. and Cornish, E.C. (1995) Antisense ACC oxidase mRNA delays carnation petal senescence. HortScience 30, 970–972. Schroeder, K.R. and Stimart, D.P. (2001) Genetic analysis of cut-flower longevity in Antirrhinum majus. Journal of the American Society for Horticultural Science 126, 200–204.

Shaw, J., Chen, H., Tsai, M., Kuo, C. and Huang, L. (2002) Extended flower longevity of Petunia hybrida plants transformed with boers, a mutated ERS gene of Brassica oleracea. Molecular Breeding 9, 211–216. Shibuya, K., Yoshioka, T., Hashiba, T. and Satoh, S. (2000) Role of the gynoecium in natural senescence of carnation (Dianthus caryophyllus L.). Journal of Experimental Botany 51(353), 2067–2073. Shibuya, K., Nagata, M., Tanikawa, N., Yoshioka, T., Hashiba, T. and Satoh., S. (2002) Comparison of mRNA levels of three ethylene receptors in senescing flowers of carnation (Dianthus caryophyllus L.). Journal of Experimental Botany 53(368), 399–406. Singh, A., Evensen, K.B. and Kao, T. (1992) Ethylene synthesis and floral senescence following compatible and incompatible pollinations in Petunia inflata. Plant Physiology 99, 38–45. Smeekens, S. (2000) Sugar-induced signal transduction in plants. Annual Review of Plant Physiology and Molecular Biology 51, 49–81. Smith, M.T., Saks, Y. and van Staden, J. (1992) Ultrastructural changes in the petals of flowers of Dianthus caryphyllus L. Annals of Botany 69, 277–285. Spikman, G. (1986) The effect of water stress on ethylene production and ethylene sensitivity of Freesia inflorescences. Acta Horticulturae 181, 135–140. Stead, A.D. (1992) Pollination-induced flower senescence: a review. Plant Growth Regulation 11, 13–20. Stead, A.D. and Moore, K.G. (1979) Studies on flower longevity in Digitalis. Planta 146, 409–414. Stephenson, P.T. and Rubinstein, B. (1998) Characterization of proteolytic activity during senescence of daylily. Physiologia Plantarum 104, 463– 473. Sugawara, H., Shibuya, K., Yoshioka, T., Hashiba, T. and Satoh, S. (2002) Is a

Flower Senescence cysteine proteinase inhibitor involved in the regulation of petal wilting in senescing carnation (Dianthus caryophyllus L.) flowers? Journal of Experimental Botany 53(368), 407–413. Suttle, J.C. and Kende, H. (1980) Ethylene action and loss of membrane integrity during petal senescence in Tradescantia. Plant Physiology 65, 1067–1072. Sylvestre, I. and Paulin, A. (1987) Accelerated ethylene production as related to changes in lipids and electrolyte leakage during senescence of petals of cut carnations (Dianthus caryophyllus). Physiologia Plantarum 70, 530–536. Tang, X. and Woodson, W.R. (1996) Temporal and spatial expression of 1aminocyclopropane-1-carboxylate oxidase mRNA following pollination of immature and mature Petunia flowers. Plant Physiology 112, 503–511. Taverner, E., Letham, D.S., Wang, J., Cornish., E. and Willcocks, D.A. (1999) Influence of ethylene on cytokinin metabolism in relation to Petunia corolla senescence. Phytochemistry 51, 341–347. ten Have, A. and Woltering, E.J. (1997) Ethylene biosynthetic genes are differently expressed during carnation (Dianthus caryophyllus L.) flower senescence. Plant Molecular Biology 34, 89–97. Thompson, J.E., Mayak, S., Shinitzky, M. and Halevy, A.H. (1982) Acceleration of membrane senescence in cut carnation flowers by treatment with ethylene. Plant Physiology 69, 859–863. Tirosh, T. and Mayak, S. (1988) Changes in starch content during the development of carnation petals. Journal of Plant Physiology 133, 361–363. Torre, S., Borochov, A. and Halevy, A.H. (1999) Calcium regulation of senescence in rose petals. Physiologia Plantarum 107, 214–219. Upfold, S.J. and van Staden, J. (1990) Cytokinins in cut carnations. VII. The

175 effect of zeatin and dihydrozeatin derivatives on flower longevity. Plant Growth Regulation 9, 77–81. Valpuesta, V, Lange, N.E., Guerrero, C. and Reid, M.S. (1995) Up-regulation of a cysteine protease accompanies the ethylene insensitive senescence of daylily (Hemerocallis) flowers. Plant Molecular Biology 28, 575–582. van Altvorst, A.C. and Bovy, A.G. (1995) The role of ethylene in the senescence of carnation flowers, a review. Plant Growth Regulation 16, 43–53. van Doorn, W.G. (1997) Effects of pollination on floral attraction and longevity. Journal of Experimental Botany 48(314), 1615–1622. van Doorn, W.G. (2001) Categories of petal senescence and abscission: a re-evaluation. Annals of Botany 87, 447–456. van Doorn, W.G. (2002a) Does ethylene treatment mimic the effects of pollination on floral lifespan and attractiveness. Annals of Botany 89, 375–383. van Doorn, W.G. (2002b) Effect of ethylene on flower abscission: a survey. Annals of Botany 89, 689–693. van Doorn, W.G. (2004) Is petal senescence due to sugar starvation? Plant Physiology 134, 35–42. van Doorn, W.G., Balk, P.A., van Houwelingen, A.M., Hoeberichts, F.A., Hall, R.D., Vorst, O., van der Schoot, C. and van Wordragen, M.F. (2003) Gene expression during anthesis and senescence in Iris flowers. Plant Molecular Biology 53, 845–863. van der Meulen-Muisers, J., van Tuyl, J.M. and van Oeveren, J.C. (1998) Genotypic variation in postharvest longevity of Asiatic hybrid lilies. Journal of the American Society for Horticultural Science 123, 283–287 van der Meulen-Muisers, J., van Tuyl, J.M., van der Plas, L.H.W. and van Oeveren, J.C. (2001) Postharvest flower development in Asiatic hybrid lilies as related to tepal carbohydrate status. Postharvest Biology and Technology 21, 201–211.

176

S. Verlinden van Staden, J. (1989) Cytokinins and auxins in carnation senescence as related to chemical treatments. Acta Horticulturae 261, 69–80. van Staden, J. (1995) Hormonal control of carnation flower senescence. Acta Horticulturae 405, 232–239. Vardi, Y. and Mayak, S. (1989) Involvement of abscisic acid during water stress and recovery in Petunia flowers. Acta Horticulturae 261, 107–112. Verlinden, S. (2003) Changes in mineral nutrient concentrations in Petunia corollas during development and senescence. Hort Science 38, 71–74. Verlinden, S. and Garcia, J.J. (2004) Sucrose loading decreases ethylene responsiveness in carnation (Dianthus caryophyllus cv, White Sim.) petals. Postharvest Biology and Technology 31, 305–312. Verlinden, S. and Woodson, W.R. (1998) The physiological and molecular responses of carnation flowers to high temperatures. Postharvest Biology and Technology 14, 185–192. Verlinden, S., Boatright, J. and Woodson, W.R. (2002) Changes in ethylene responsiveness of senescence-related genes during carnation flower development. Physiologia Plantarum 116, 503–511. Wagstaff, C., Leverentz, M.K., Griffiths, G., Thomas, B., Chanasut, U., Stead, A.D. and Rogers, H.J. (2001) Cysteine protease gene expression and proteolytic activity during senescence of Alstroemeria petals. Journal of Experimental Botany 53(367), 233– 240. Wagstaff, C., Leverentz, M.K., Griffiths, G., Thomas, B., Chanasut, U., Stead, A.D. and Rogers, H.J. (2003) Programmed cell death (PCD) processes begin extremely early in Alstroemeria petal senescence. New Phytologist 160, 49–59. Waki, K. Shibuya, K., Yoshioka, T., Hashiba, T. and Satoh, S. (2001) Cloning of a cDNA encoding EIN3-like protein (DC-EIL1) and decrease in its

mRNA level during senescence in carnation flower tisues. Journal of Experimental Botany 52(355), 377– 379. Wang, H., Woodson, W.R. and Brandt, A.S. (1993) A flower senescencerelated mRNA from carnation encodes a novel protein related to enzymes involved in phosphonate biosynthesis. Plant Molecular Biology 22, 719– 724. Wang, N.N., Yang, S.F. and Charng, Y. (2001) Differential expression of 1-aminocyclopropane-1-carboxylate synthase genes during orchid flower senescence induced by the protein phosphatase inhibitor okadaic acid. Plant Physiology 126, 253–260. Wernett, H.C., White, T.L., Powell, G.L., Wilcox, C.J., Martin, F.G., Wifret, G.J., Sheehan, T.J. and Lyrene, P.M. (1996) Postharvest longevity of cutflower Gerbera. II: heritability of vase life. Journal of the American Society for Horticultural Science 121, 222– 224. Weterings, K., Pezzotti, M., Cornelissen, M. and Mariani, C. (2002) Dynamic 1-aminocyclopropane-1-carboxylatesynthase and -oxidase transcript accumulation patterns during pollen tube growth in tobacco styles. Plant Physiology 130, 1190–1200. Whitehead, C.S. and Halevy, A.H. (1989) Ethylene sensitivity: the role of shortchain saturated fatty acids in pollination-induced senescence of Petunia hybrida flowers. Plant Growth Regulation 8, 41–54. Wilkinson, J.Q., Lanahan, M.B., Clark, D.G., Bleecker, A.B., Chang, C., Meyerowitz, E.M. and Klee, H.J. (1997) A dominant mutant receptor from Arabidopsis confers insensitivity in heterologous plants. Nature Biotechnology 15, 444–448. Woltering, E.J., de Vrije, T., Harren, F. and Hoekstra, F.A. (1997) Pollination and stigma wounding: same response, different signal? Journal of Experimental Botany 48(310), 1027–1033.

Flower Senescence Woltering, E.J. and Van Doorn, W.G. (1988) Role of ethylene in senescence of petals: morphological and taxonomical relationships. Journal of Experimental Botany 39(208), 1605–1616. Woodson, W.R. (1987) Changes in protein and mRNA populations during carnation petal senescence. Plant Physiology 71, 445–502. Woodson, W.R. and Brandt, A.S. (1991) Role of the gynoecium in cytokinininduced carnation petal senescence. Journal of the American Society for Horticultural Science 116, 676–679.

177 Woodson, W.R., Park, K.Y., Drory, A., Larsen, P.B. and Wang, H. (1992) Expression of ethylene biosynthetic pathway transcripts in senescing carnation flowers. Plant Physiology 99, 526– 532. Wu, M., Van Doorn, W., Mayak, S. and Reid, M.S. (1989) Senescence of ‘Sandra’ carnation. Acta Horticulturae 261, 221–225. Xu, Y. and Hanson, M.R. (2000) Programmed cell death during pollination-induced petal senescence in Petunia. Plant Physiology 122, 1323–1333.

8

Developmental Control and Biotechnology of Floral Pigmentation K. DAVIES

AND

K. SCHWINN

New Zealand Institute for Crop and Food Research Ltd, Private Bag 11600, Palmerston North, New Zealand

Introduction For many angiosperms, pigment formation is a key part of flower development. Flower colour, along with fragrance, floral shape and nectar reward, is important to the interaction between plants and pollinators; and preferences towards specific colours are exhibited by pollinators, whether they are birds, bees, butterflies or other insects (reviewed in Bohm, 1998). Commonly contributing to floral phenotypes are colour combinations and patterning such as distinctive spots on the flower ‘lip’ or pigment lines in the flower tube (e.g. orchids, Plate 1 – see frontispiece to this book). These may provide more specific signals within the flower, e.g. acting as nectar guides to insects. In an extraordinary example of pollinator signalling, the flowers of the orchid genus Ophrys use colour, scent and shape to mimic female bees, causing the male bee to attempt copulation, thus achieving pollination (Paxton and Tengo, 2001). The major pigments responsible for flower colour are carotenoids, flavonoids and betalains. Although other pigment types such as chlorophylls (e.g. Cymbidium orchid, Plate 1), phenylphenalenones and quinochalcones can generate flower colours, their occurrence is rare (Davies, 2004). Carotenoids are lipid soluble, plastid-located terpenoids present in photosynthetic plants, algae and bacteria. For pigmentation of flowers (and fruits), carotenoids accumulate to high levels in specialized plastids called chromoplasts. More generally, these pigments participate in the harvesting and dissipation of light energy in chloroplasts, protecting the photosynthetic machinery from photo-oxidation. Carotenoids are the basis of pigmentation in most yellow flowers, but they also generate orange, red, brown and bronze flower colours. The flavonoids are phenylpropanoid compounds of great variation in structure and function (Bohm, 1998). Those involved in flower colour are water soluble and generally located in the vacuole. The predominant flavonoid pigments are the anthocyanins, which are the basis for nearly all of the pink, red, orange, scarlet, purple, 178

ßCAB International 2006. The Molecular Biology and Biotechnology of Flowering, 2nd edn (ed. B.R. Jordan)

Developmental Control and Biotechnology of Floral Pigmentation

179

blue and blue-black flower colours. The aurones and chalcones are yellow flavonoid pigments, which provide pigmentation in a few species. The betalains are nitrogenous compounds, also vacuolar located, which produce yellow, orange, red and purple flower colours. Their occurrence is restricted to the Caryophyllales (and some species of the fungal genera Amanita and Hygrocybe). Most of the families in the Caryophyllales have evolved to use betalains as the key floral pigments rather than the anthocyanins, even though the flavonoid pathway is active in these plants. The molecular basis for the mutual exclusion of betalains and anthocyanins in the Caryophyllales remains to be determined. Anthocyanins and carotenoids commonly occur together in petals, and are generally localized to different cell layers of the petal. Anthocyanins are typically found in epidermal cells and carotenoids subepidermally. When they cooccur, the pigmentation phenotype depends on the spatial and temporal distribution of the different pigment types, but may include orange, scarlet and brown/black colours as well as colour patterns and colours that change during flower development. Plate 1 shows examples of the floral colours generated by carotenoids, flavonoids and betalains. Although pigmentation in flowers is primarily to be found in the petals, in some species sepals or fused floral organs (e.g. tepals) may provide the main pigmentation. In others, non-floral organs are coloured. Within the Araceae family, a brilliantly coloured heart-shaped bract (the spathe) is used to attract pollinators. The bracts also provide the main coloration in Bougainvillea (Plate 1). In many species the pollen is also coloured by carotenoids and/or flavonoids. Although colourless flavonoids in pollen have been shown to be required for fertility in some species (e.g. Taylor and Jorgensen, 1992; Jorgensen et al., 2002), the function of the coloured compounds is not clear. Mutants lacking either carotenoids or anthocyanins can retain pollen fertility, although maternal fertility may be affected (Jorgensen et al., 2002; Wakelin et al., 2003). In some cases, pollen colour may be part of the signal to pollinators (Lunau, 2000) and it is also possible that the pigments have protective activities against various stresses. Pigment production is often timed to coincide with flower fertility. Thus, pigments may appear just prior to the flower becoming fertile; or in some cases, flower colour may fade or the flower may turn from white to coloured or even change in colour following fertilization. These transformations may be the result of changes in petal cell pH as the flower ages, degradation of the pigments or de novo pigment biosynthesis. An example of the latter can be found in Lantana. The flowers are yellow initially when fertile but change to purple over an ageing period of 3 days (Plate 1). The yellow flowers offer nectar and pollen rewards that older flowers do not, and are preferred by the butterfly pollinators (Weiss, 1995). Retention of the purple older flowers maintains a larger inflorescence, which is more attractive to pollinators at a distance while still directing pollinators to fertile flowers. At least 200 plant genera contain species that show colour change during flower development. Thus, variation in flower colour associated with a change in nectar and pollen availability may be a common occurrence (Weiss, 1995; Bohm, 1998). The production of pigments in complex patterns that are coincident with fertility and localized to particular cell types within the flower requires the

180

K. Davies and K. Schwinn

coordinated induction of the genes for the pigment biosynthetic enzymes. Furthermore, in addition to the developmental signals, the pigment biosynthetic genes may respond to environmental factors such as the light quality and quantity. This chapter reviews the knowledge on the regulation of pigment production during flower development, with an emphasis on the transcriptional regulation of the biosynthetic genes. It also provides a brief overview of the major flower pigment biosynthetic pathways. A comprehensive review of the biochemistry and molecular genetics of pigment biosynthesis would require several chapters of this book. Thus, only outlines of the biosynthetic pathways are given with reference to recent reviews. Flower colour has been the target of many different genetic modification (GM) approaches, either for advancing understanding of pigment biosynthesis or for the development of novel ornamental varieties. In the second part of this chapter we address the progress to date on the GM of floral pigments.

The Biosynthesis of Plant Pigments Flavonoid and betalain pigments, like many plant secondary metabolites, are derived from aromatic amino acids. Carotenoids, and a range of other compounds such as gibberellins and tocopherols, are produced from the 5-carbon (C5) compound isopentenyl diphosphate (IPP) within the general isoprenoid pathway. The carotenoid and flavonoid biosynthetic pathways are now well defined at the genetic and enzymatic level. Of course, some knowledge gaps do exist, but these are becoming fewer and gene sequences are now available for most of the biosynthetic steps for the coloured compounds. The molecular data for other floral pigments are very limited, although there have been significant recent advances in understanding betalain and apocarotenoid biosynthesis. The biosynthetic steps for which gene sequences are available are listed in Table 8.1, including details of the protein families to which the enzymes belong and reference to the first publications detailing identification of the corresponding DNA sequences. More extensive recent reviews on the molecular biology of flavonoid and carotenoid biosynthesis can be found in Marles et al. (2003), Springob et al. (2003), Cuttriss and Pogson (2004), Fraser and Bramley (2004), Schwinn and Davies (2004), Tanner (2004) and Dixon et al. (2005). For flavonoids and carotenoids, evidence is emerging on various subcellular factors involved in their biosynthesis. Carotenoids are synthesized and stored in plastids by nuclear-encoded enzymes; and flavonoids are synthesized in the cytosol, likely in association with the endoplasmic reticulum. The biosynthetic enzymes for both pathways are thought to form multi-enzyme complexes that may enable the selective channelling of intermediates along alternative biosynthetic paths (see the section below on Regulation of Pigment Biosynthesis in Flowers and Winkel, 2004). The requirement for appropriate enzyme complexes to be formed may have important implications for metabolic engineering experiments. For the anthocyanins, vacuolar compartmentalization and pH of the environment are also key elements in forming the coloured compounds,

Enzyme

Abbreviation

EC number

Protein family

Referencea

Flavonoid precursors Acetyl-CoA carboxylase (cytosolic)

ACC

6.4.1.2

Phenylalanine ammonia-lyase Cinnamate 4-hydroxylase

PAL C4H

4.3.1.5 1.14.13.11

Biotin-containing carboxylases Ammonia-lyases CytP450 (CYP73A)b

4-Coumarate:CoA ligase Pathway to anthocyanins Chalcone synthase

4CL

6.2.1.12

Adenylate-forming enzymes

Roesler et al. (1994); Shorrosh et al. (1994) Kuhn et al. (1984) Fahrendorf and Dixon (1993); Mizutani et al. (1993); Teutsch et al. (1993) Kuhn et al. (1984)

CHS

2.3.1.74

Polyketide synthase

Chalcone isomerase Flavanone 3-hydroxylase

CHI F3H (FHT)

5.5.1.6 1.14.11.9

Flavanone 4-reductase Dihydroflavonol 4-reductase

FNR DFR

1.1.1.234 1.1.1.219

No named family 2-Oxoglutarate-dependent dioxygenase (2OGD) NADPH reductase (RED) RED

Anthocyanidin synthase (leucoanthocyanidin dioxygenase) UDP-Glc:anthocyanidin 3-O-glucosyltransferase/ UDP-Glc:flavonol 3-O-glucosyltransferase UDP-Glc:anthocyanin 5-O-glucosyltransferased UDP-Glc:anthocyanin 3’-O-glucosyltransferase UDP-Glc:anthocyanin 3’,5’-O-glucosyltransferase UDP-Rha:anthocyanidin 3-O-glucoside rhamnosyltransferase Hydroxycinnamoyl-CoA:anthocyanin 5-Oglucoside-6’’’-O-hydroxycinnamoyltransferasee Hydroxycinnamoyl-CoA:anthocyanidin 3-Oglucoside-6’’-O-hydroxycinnamoyltransferase Malonyl-CoA:anthocyanin 5-O-glucoside-4’’’O-malonyltransferase Malonyl-CoA:anthocyanin 5-O-glucoside-6’’’O-malonyltransferase

ANS (LDOX)

1.14.11.19

2OGD

F3GT A5GT A3’GT A3’,5’GT A3RT

2.4.1.115/ 2.4.1.91 2.4.1.2.4.1.2.4.1.2.4.1.-

UDPG-Oglycosyltransferase (UGT) UGT UGT UGT UGT

A5AT (Gt5AT)

2.3.1.153

A3AT (Pf3AT)

2.3.1.-

Versatile acyltransferase (VAT) VAT

A5MT (Ss5MaT2)

2.3.1.-

VAT

Yonekura-Sakakibara et al. (2000) Suzuki et al. (2004b)

A5MT (Ss5MaT1)

2.3.1.-

VAT

Suzuki et al. (2001)

Kreuzaler et al. (1983); Reimold et al. (1983) Mehdy and Lamb (1987) Martin et al. (1991) O’Reilly et al. (1985)c Martin et al. (1985); O’Reilly et al. (1985) Menssen et al. (1990); Martin et al. (1991) Fedoroff et al. (1984) Yamazaki et al. (1999, 2002) Fukuchi-Mizutani et al. (2003) Noda et al. (2004) Brugliera et al. (1994); Kroon et al. (1994) Fujiwara et al. (1998)

Developmental Control and Biotechnology of Floral Pigmentation

Table 8.1. Flavonoid, carotenoid and betalain biosynthetic enzymes involved in flower pigmentation for which DNA sequences have been published.

181 continued

Protein family

Referencea

2.3.1.-

VAT

2.3.1.-

VAT

Suzuki et al. (2002, 2003, 2004a) Suzuki et al. (2004a)

F3’H F3’,5’H A3’OMT

1.14.13.21 1.14.13.2.1.1.-

CytP450 (CYP75B) CytP450 (CYP75A) SAM O-Methyltransferase (OMT)

SAM:anthocyanin 3’,5’-O-methyltransferase

A3’5’OMT

2.1.1.-

OMT

Flavones and flavonols Flavonol synthase Flavone synthase I Flavone synthase II

FLS FNSI FNSII

1.14.11.1.14.11.1.14.13.-

2OGD 2OGD CytP450 (CYP93B)

Holton et al. (1993b) Martens et al. (2001) Akashi et al. (1999); Martens and Forkmann (1999)

C2’GT

2.4.1.-

UGT

Patent application WO03/018682

PKR (CHR, CHKR)

1.1.1.-

Aldo/ketoreductase

Welle et al. (1991)

AUS

1.21.3.6

Polyphenol oxidase

Nakayama et al. (2000)

ANR

1.1.1.-

RED

Devic et al. (1999); Xie et al. (2003)

DXPS

4.1.3.37

Transketolase

DXR

1.1.1.267

Reductoisomerase

IPPI GGPPS

5.3.3.2 2.5.1.29

IPP isomerase type 1 Prenyltransferase

Bouvier et al. (1998); Lange et al. (1998) Lange and Croteau (1999); Schwender et al. (1999) Blanc and Pichersky (1995) Kuntz et al. (1992)

Enzyme

Abbreviation

Malonyl-CoA:anthocyanidin 3-O-glucoside-6’’O-malonyltransferase Malonyl-CoA:anthocyanidin 3-O-glucoside-3’’,6’’O-dimalonyltransferase Flavonoid 3’-hydroxylase Flavonoid 3’,5’-hydroxylase SAM:anthocyanin 3’-O-methyltransferase

A3MT (Sc3MaT, Dm3Mat1, Dv3MaT) A3diMT (Dm3MaT2)

Carotenoid precursors 1-Deoxy-D-xylulose 5-phosphate synthase 1-Deoxy-D-xylulose 5-phosphate reductoisomerase IPP isomerase GGPP synthase

Brugliera et al. (1999) Holton et al. (1993a) Quattrochio et al. (1993); Patent application WO03/062428 Quattrochio et al. (1993); Patent application WO04/062428

K. Davies and K. Schwinn

Chalcones UDP-Glc:chalcone 2’-O-glucosyltransferase 6’-Deoxychalcones Polyketide reductase Aurones Aureusidin synthase Proanthocyanidins Anthocyanidin reductase

EC number

182

Table 8.1. continued . Flavonoid, carotenoid and betalain biosynthetic enzymes involved in flower pigmentation for which DNA sequences have been published.

PSY PDS

2.5.1.32 1.3.99.-

z-Carotene desaturase Carotenoid isomerase Plastid terminal oxidase

ZDS CRTISO PTOX

1.14.99.30 – –

Lycopene e-cyclase Lycopene b-cyclase e-Ring hydroxylase b-Ring hydroxylase b-Carotene 3,4-desaturase/4-hydroxylase

eLCY bLCY eOH bOH AdKeto1/ AdKeto2 CCS ZEP VDE NXS

Capsanthin-capsorubin synthase Zeaxanthin epoxidase Violaxanthin de-epoxidase Neoxanthin synthase Carotenoid cleavage products Lycopene cleavage dioxygenase Bixin aldehyde dehydrogenase Norbixin carboxyl methyltransferase Zeaxanthin 7,8,7’,8’-cleavage dioxygenase Betalain pathway DOPA-4,5-dioxygenase UDP-Glc:betanidin 5-O-glucosyltransferase UDP-Glc:betanidin 5-O-glucosyltransferase

Bird et al. (1991); Bartley et al. (1992); Ray et al. (1992) Bartley et al. (1991)

5.5.1.5.5.1.1.14.-.1.14.13.1.14.13.-

Prenyltransferase Flavin adenosine dinucleotide (FAD) binding protein (BP) FAD BP FAD BP Non-haem di-iron oxygenase (membrane-bound di-iron carboxylase) FAD BP FAD BP CytP450 (CYP97C) Non-haem di-iron oxygenase Non-haem di-iron oxygenase

Cunningham et al. (1996) Hugueney et al. (1995) Tian et al. (2004) Sun et al. (1996) Cunningham and Gantt (2005)

– – – –

FAD BP Lipocalin Lipocalin FAD BP

Bouvier et al. (1994) Marin et al. (1996); Bouvier et al. (1996) Bugos and Yamamoto (1996) Al-Babili et al. (2000); Bouvier et al. (2000)

LCD

–

Bouvier et al. (2003b)

BAHD BMT CsZCD

– 2.1.1.–

Carotenoid cleavage dioxygenase (CCD) Aldehyde dehydrogenase MT CCD

Bouvier et al. (2003b) Bouvier et al. (2003b) Bouvier et al. (2003a)

DOD B5GT B6GT

– – –

Extradiol dioxygenase UGT UGT

Christinet et al. (2004) Vogt et al. (1999) Vogt (2002)

Albrecht et al. (1995) Isaacson et al. (2002); Park et al. (2002) Carol et al. (1999); Josse et al. (2000)

Developmental Control and Biotechnology of Floral Pigmentation

Main carotenoid pathway Phytoene synthase Phytoene desaturase

a

Reference to the first publications on the isolation and characterization of the corresponding cDNA/gene. CytP450s are classified according to molecular phylogeny into numbered families ($40% amino acid positional identity) and subfamilies designated by a letter ($55% amino acid identity). c Reference is given to the first example of isolation of a DFR cDNA that was shown to have FNR activity. d Recombinant A5GT proteins show varying degrees of anthocyanin substrate specificity. e The nomenclature for the AATs follows Nakayama et al. (2003). The positional numbering of the sugar hydroxyl that is modified is given followed by prime symbols to indicate which sugar is affected. The double and triple primes indicate the 3-O-glycosyl and 5-O-glycosyl, respectively. Recombinant AATs show varying degrees of substrate specificity. b

183

184

K. Davies and K. Schwinn

and aspects of vacuole acidification and the localization process are now understood (Fukada-Tanaka et al., 2000; Winefield, 2002). In flowers, carotenoid accumulation is associated with esterification of the pigments, specific proteins (carotenoid-associated proteins) and the development of chromoplasts, all of which may assist in sequestration of the high levels of pigments that occur compared to the levels in leaves (Vishnevetsky et al., 1999a; Fraser and Bramley, 2004). Flavonoids also interact with specific proteins in some species (Winefield, 2002). Despite the notable progress on pigment biosynthesis, little is known of the turnover and degradation of most pigments. Anthocyanin levels can change rapidly in flowers. For example, in the yesterday, today and tomorrow plant (Brunfelsia calcina), the flowers turn from white to purple and back to white within 3 days. For betalains, a number of degradation activities have been biochemically characterized (Zry¨d and Christinet, 2004). However, in general, research on the biochemistry and chemistry of pigment breakdown (reviewed in Simpson et al., 1976) has not been followed by extensive molecular findings. An exception is the recent cloning of cDNAs for activities that convert carotenoids into apocarotenoids and other cleavage products (Cuttriss and Pogson, 2004; Table 8.1).

The biosynthesis of flavonoids Flavonoid biosynthesis is part of the larger phenylpropanoid pathway, which produces a range of secondary metabolites from the aromatic amino acid phenylalanine. There are many branches to the flavonoid-specific pathway, producing coloured and colourless compounds with diverse biological functions. As mentioned previously, anthocyanins are the most significant flavonoid pigments, with aurones, chalcones and some flavonols playing a limited role. Also important to pigmentation are the colourless (or weakly coloured) flavones and flavonols, for their function as co-pigments. They stabilize and maintain anthocyanins in their coloured forms, in a process of complex molecular interactions known as co-pigmentation (Brouillard and Dangles, 1993). Flavones and flavonols, strong absorbers of UV light, also are the basis for some floral insect nectar guides. At their nucleus, flavonoids are 15-carbon (C15) compounds composed of two aromatic rings (called the A- and B-rings) joined by a 3-carbon unit (which usually forms a third ring called the C-ring; Fig. 8.1). The various types of flavonoids are determined by the degree of oxidation of the C-ring; and individual compounds of the same type are determined by the degree of hydroxylation and the type and extent of modifications such as glycosylation, acylation and methylation. Anthocyanin B-ring hydroxylation patterns are of prime importance in flower colour (see section below on B-ring hydroxylation). In the following sections, we briefly describe the parts of the pathway that are relevant to flower colour. Our focus is on the biosynthetic steps leading to different pigment/co-pigment types and anthocyanin modifications, as knowledge on the biochemistry and molecular biology of these aspects is most

Developmental Control and Biotechnology of Floral Pigmentation

185

R1 3 2 HO

4⬘

5⬘

3⬘

OH

4

2⬘

B 6⬘ OH

5

HO

7

6

A

8 A

B O C

6

3⬘ 2⬘ OH

5 O

4

2 6⬘

4⬘ OH 5⬘ R2

3 R3

OH

Fig. 8.1. Structures of naringenin chalcone (left) and the main anthocyanidins (right). The lettering of the carbon rings is shown, as well as the numbering of the key carbons. Note that the numbering used for chalcones is the same as for hydroxycinnamic acids (Fig. 8.2), while the numbering for anthocyanidins is as for the other flavonoids shown in Fig. 8.3 (except aurones). Substitutions at R1 and R2 determine the various common anthocyanidins (R3 ¼ OH) as follows: pelargonidin (R1 and R2 ¼ H), cyanidin (R1 ¼ OH and R2 ¼ H), delphinidin (R1 and R2 ¼ OH), peonidin (R1 ¼ OCH3 and R2 ¼ H), petunidin (R1 ¼ OCH3 and R2 ¼ OH) and malvidin (R1 and R2 ¼ OCH3). The 3-deoxyanthocyanidins (R3 ¼ H), which are rare compounds, are apigeninidin (R1 and R2 ¼ H), luteolinidin (R1 ¼ OH and R2 ¼ H) and tricetinidin (R1 and R2 ¼ OH).

advanced. Bohm (1998) and Forkmann and Heller (1999) provide extensive reviews on the biochemistry (and chemistry) of flavonoids. Formation of flavonoid pathway precursors Precursors for the flavonoid pathway are malonyl-CoA and a hydroxycinnamic acid (HCA)-CoA ester, usually 4-coumaroyl-CoA (Fig. 8.2). 4-Coumaroyl-CoA is derived from phenylalanine through the sequential activities of phenylalanine ammonia-lyase (PAL), cinnamate 4-hydroxylase (C4H) and 4-coumarate:CoA ligase (4CL). Citrate is the source of malonyl-CoA, the conversion likely involving ATP citrate lyase (ACL) and acetyl-CoA carboxylase (ACC). Malonyl-CoA and CoA esters of HCAs can also feed into the flavonoid pathway at later stages as acid group donors in the acylation of the end products. Formation of chalcones and aurones The first step committed to flavonoid synthesis is the formation of chalcone pigment (pale yellow), which establishes the C15 flavonoid structure (Fig. 8.2). Chalcone synthase (CHS) is the enzyme involved. It catalyses the condensation of one molecule of an HCA-CoA ester with three molecules of malonyl-CoA. Naringenin chalcone is the first flavonoid formed in most plants, through the use of the HCA-CoA substrate 4-coumaroyl-CoA. To function as floral pigments, naringenin chalcone, or chalcones derived from it, must be stabilized to prevent spontaneous conversion to colourless flavanone isomers. This is accomplished by enzyme-catalysed modifications involving glycosylation or methylation. In some species, the co-action of polyketide reductase (PKR) with CHS results in 6’-deoxychalcones, which are relatively stable due to an intramolecular hydrogen bond between the 2’-hydroxyl and the carbonyl group.

186

K. Davies and K. Schwinn

OH

HO

HO

OH STS Stilbenes

CoA

4CL C4H O O O Cinnamate 4-Coumarate 4-Coumaroyl-CoA Malonyl-CoA (3x) ACC ACL Citrate Acetyl-CoA Malonyl-CoA (3x)

Phenylalanine

PAL

CHS 6⬘-Deoxychalcones +PKR CHS Naringenin chalcone

Anthocyanins Aurones Flavones Flavonols Proanthocyanins

Fig. 8.2. Biosynthesis of the flavonoid pathway precursors 4-coumaroyl-CoA and malonylCoA. 4-Coumaroyl-CoA is synthesized within the phenylpropanoid pathway and malonyl-CoA within primary metabolism. Enzyme abbreviations are defined in the text or Table 8.1. Also shown are the points of action for PKR and STS, transgenes for which have been used to modify flower colour.

Aurones, bright yellow in colour, are formed directly from chalcones (Fig. 8.3). The reaction is reported to be catalysed by a bifunctional polyphenol oxidase-like enzyme (Nakayama et al., 2000; Davies et al., 2001). In antirrhinum (Antirrhinum majus), the enzyme termed aureusidin synthase (AUS) performs oxidation and B-ring hydroxylation reactions on naringenin chalcone substrate (Nakayama et al., 2001). However, confirmation of the role of AUS awaits analysis of gain of function or knockout plants generated through transgenic or mutagenic approaches. A preliminary report suggests the involvement of chalcone 4’-O-glucosyltransfease for generating the AUS substrate (Ono et al., 2005). Formation of flavones and dihydroflavonols As mentioned above, chalcones can spontaneously isomerize to colourless compounds of the flavanone type. The isomerization results in the formation of the heterocyclic C-ring (and the loss of the chromophore) (Fig. 8.3). The spontaneous reaction may occur in planta, but the conversion generally precedes enzymatically, through the activity of chalcone isomerase (CHI). Flavanones serve as substrates for the formation of flavones and dihydroflavonols (DHFs) (Fig. 8.3), which are also colourless. The conversion to flavone requires action of a flavone synthase (FNS). There are two types of FNS. In members of the Apiaceae, the enzyme (termed FNSI) is a 2-oxoglutaratedependent (2OG)-dioxygenase (Martens et al., 2001). In all other species studied to date, it is a cytochrome P-450-dependent monooxygenase

Developmental Control and Biotechnology of Floral Pigmentation

187

OH OH HO

OH HO

O

OH

AUS OH

O

OH

Aureusidin Aurone

CHI

OH HO

O

O

Naringenin chalcone

FNSI

HO

OH

O

FNSII OH

OH

O

F3H

OH HO

O

O

Naringenin

Apigenin Flavone

FLS

HO

OH OH

O

OH

Kaempferol Flavonol

FNR

O

OH OH

OH

HO

O

O

Dihydrokaempferol DFR HO

OH

OH

OH

Apiforol

O

ANS OH

LAR OH

F5GT

OH

OH

Leucopelargonidin

PAs

ANS HO

HO

O

OH

O O-Glc

ANR OH

Apigeninidin 5-O-glucoside 3-Deoxyanthocyanin

OH

Pelargonidin F3GT

HO

OH

O

O-Glc OH

Pelargonidin 3-O-glucoside Anthocyanin

Fig. 8.3. Simplified biosynthetic scheme for the major flavonoid pigment types. Only the routes to compounds with 4’-hydroxylation are shown, and the anthocyanidin/anthocyanin structures are presented in the cation form. Points of action are given for the proanthocyanidin (PA) biosynthetic enzymes leucoanthocyanidin reductase (LAR) and anthocyanidin reductase (ANR), transgenes for which may be used to modify flower colour. Other enzyme abbreviations are defined in the text or Table 8.1.

188

K. Davies and K. Schwinn

(CytP450), termed FNSII (Akashi et al., 1999; Martens and Forkmann, 1999). Flavanones are converted to DHFs by flavanone 3-hydroxylase (F3H). Formation of flavonols and leucoanthocyanidins DHFs are substrates for the formation of flavonols and leucoanthocyanidins (Fig. 8.3). Flavonols are formed through the action of flavonol synthase (FLS). A reduction catalysed by dihydroflavonol 4-reductase (DFR) yields leucoanthocyanidins. These colourless and unstable compounds are the direct precursors for anthocyanidin production. In terms of flower colour, the formation of leucoanthocyanidin is one of the key reactions in anthocyanin synthesis, as DFR substrate specificity is one of the determinants of the base pattern of B-ring hydroxylation, and hence colour, of the anthocyanin formed (Fig. 8.4, see also sections below on B-ring hydroxylation and altering B-ring hydroxylation). The relevant DHF substrates are dihydrokaempferol (DHK), dihydroquercetin (DHQ) and dihydromyricetin (DHM), having one, two or three B-ring hydroxyl groups, respectively. In the formation of the rare 3-deoxyanthocyanins, which give orange and red floral colours (Winefield, et al., 2005), a variant of DFR (initially referred to as flavanone 4-reductase or FNR) is capable of reducing flavanone substrates to flavan-4-ols (3-deoxyleucoanthocyanidins). Formation of anthocyanins The first anthocyanin formed in most plants is an anthocyanidin 3-O-glycoside. It is formed from leucoanthocyanidin through the activity of anthocyanidin synthase (ANS, also referred to as leucoanthocyanidin dioxygenase) and an anthocyanidin 3-O-glycosyltransferase (3GT) (Fig. 8.3). The ANS product is an anthocyanidin in a colourless pseudobase form, which serves as the substrate for a 3GT. 3GTs mediate the transfer of a glycosyl residue from an activated nucleotide sugar to the C-3 position of the pseudobase. A well-characterized

F3⬘H

Naringenin

Eriodictyol

F3⬘,5⬘H

Pentahydroxyflavanone

F3H F3⬘,5⬘H F3H F3H F3⬘,5⬘H F3⬘H Dihydroquercetin Dihydrokaempferol Dihydromyricetin DFR F3⬘,5⬘H Leucopelargonidin

Leucocyanidin

OH HO

O

OH

Pelargonidin

OH

OH HO

OH

ANS

OH

OH O

Leucodelphinidin

ANS

ANS

HO

DFR

DFR

O

OH

Cyanidin

OH OH

OH OH

Delphinidin

Fig. 8.4. Key enzymatic conversions involved in the formation of the three major anthocyanidins (shown in the cation rather than pseudobase form). Enzyme abbreviations are defined in the text or Table 8.1.

Developmental Control and Biotechnology of Floral Pigmentation

189

3GT, which adds a glucose residue, is uridine diphosphate (UDP)-glucose:flavonoid 3-O-glucosyltransferase (F3GT). The first F3GTs characterized were found to also use flavonol substrate, hence the use of flavonoid in the name. However, recent studies indicate there are 3GTs specific to anthocyanidin substrate (e.g. Ford et al., 1998). After the initial 3-O-glycosylation, anthocyanins are commonly modified by the addition of other sugar residues, to the O-linked sugar moiety at the C-3 position and/or to other positions (e.g. frequently at C-5). They also can be modified by aromatic or aliphatic acylation at one or more of the sugar moieties, and/or methylation of hydroxyl groups. Commonly, enzymes with strict substrate specificities catalyse these modifications. Nakayama et al. (2003) present a detailed review of the biochemistry and molecular biology of anthocyanin acyltransferases, while Ibrahim and Muzac (2000) and Vogt (2000) review the methyltransferase and glycosyltransferase gene superfamilies of plants, respectively, including those related to flavonoid biosynthesis. For simple 3-deoxyanthocyanins, the initial O-glycosylation occurs at the C-5 position. In what can be thought of as the final step in anthocyanin biosynthesis, anthocyanins are deposited in the vacuole, where the acidic environment allows these pH-responsive compounds to assume coloured forms (discussed further in section below on Vacuolar import). B-ring hydroxylation The degree of B-ring hydroxylation is a key determinant of flower colour, as it determines the colour manifested by an individual anthocyanin. An increase in hydroxylation shifts the colour away from the red end of the spectrum towards the blue. Figure 8.4 shows the three anthocyanidin structures from which most anthocyanins are derived. Pelargonidin-derived anthocyanins are associated with orange, pink and red flower colours, those derived from cyanidin with red, magenta and red-purple colours and those from delphinidin with purple and blue colours. The key hydroxylations occur at the C-3’ position or at both the C-3’ and C-5’ positions (Fig. 8.1) through the activity of the CytP450s enzymes, flavonoid 3’-hydroxylase (F3’H) and flavonoid 3’,5’-hydroxylase (F3’,5’H), respectively. Genes/cDNAs for both enzymes (sometimes referred to as the red and blue genes) were first cloned and characterized from petunia (Petunia hybrida) and have been subsequently isolated from several other species (Tanner, 2004). Anthocyanins themselves are not the substrates for these enzymes but rather flavanone and DHF intermediates in the anthocyanin biosynthetic pathway (Fig. 8.4). The hydroxylases from some species also have been found to use flavone and flavonol substrate. In petunia, the electron donor cyt b5 is required for full F3’,5’H (but not F3’H) activity in the flowers, replacing the more commonly used NADPH-cytochrome P-450 reductase electron donor (de Vetten et al., 1999). Vacuolar import Flavonoids involved in flower pigmentation are generally stored in the vacuole. For anthocyanins in particular, importation into the vacuole is crucial. This is

190

K. Davies and K. Schwinn

vividly demonstrated by the phenotypes caused by the importation mutants, bronze2 and anthocyanin9 (an9) of maize (Zea mays) and petunia, respectively. In the maize mutant, the kernels exhibit a brown or bronze colour attributed to the oxidation and condensation of anthocyanin in the cytosol (Marrs et al., 1995), while in the petunia mutant, the flower is white (Alfenito et al., 1998). The loci encode different types of glutathione S-transferases (GSTs), however, they are functionally homologous (Alfenito et al., 1998). These GSTs may function through their glutathionating activity or alternatively, may be carrier proteins, binding and escorting anthocyanins to the vacuole (Mueller et al., 2000). ABC-type transporters have been implicated in sequestration of anthocyanins and other flavonoids that have been glutathionated (reviewed in Winefield, 2002); and recently, an ABC protein of the multidrug resistance-associated protein type was confirmed by mutant analysis to be required for anthocyanin transport in maize (Goodman et al., 2004). However, alternative mechanisms, e.g. involving multidrug and toxic compound extrusion (MATE) transporters or vacuolar Hþ-ATPases (Baxter et al., 2005), have been implicated for other flavonoid types. It is not yet clear whether mechanisms are specific to the species or to the flavonoid type.

The biosynthesis of carotenoids Carotenoids are polyene molecules, with most having a C40 backbone and several conjugated double bonds. There are two types of carotenoids, the hydrocarbon carotenes and their oxygenated derivatives, the xanthophylls. While both types of carotenoids may contribute to flower colour, individual species tend to produce predominantly one type or the other, and some species produce distinctive rare xanthophyll structures (Goodwin, 1976). The following sections briefly describe the formation of carotenoid precursors, the branches of the carotenoid pathway common to all plants and the biosynthesis of carotenoid cleavage products, with emphasis on aspects pertaining to chromoplasts and floral pigmentation. Comprehensive recent reviews on most aspects of carotenoids can be found in Cuttriss and Pogson (2004) and Fraser and Bramley (2004). Formation of IPP The plastidic 2-C-methyl-D-erythritol 4-phosphate (MEP) pathway (Fig. 8.5) is the main source of IPP used in carotenoid synthesis. However, the cytosolic mevalonic acid pathway also produces IPP, and in some instances this is linked with carotenoid production (Fraser and Bramley, 2004). In the MEP pathway, 1-deoxy-D-xylulose 5-phosphate (DXP), which is derived from the condensation of pyruvate and glyceraldehyde-3-phosphate, is converted to MEP through the action of 1-deoxy-D-xylulose 5-phosphate reductoisomerase (DXR). A series of enzymatic conversions then form IPP and its allylic isomer dimethylallyl diphosphate (DMAPP) from MEP. DMAPP is also formed through the isomerization of IPP via the action of IPP isomerase (IPPI), a key biosynthetic route with regard to carotenoid biosynthesis.

Developmental Control and Biotechnology of Floral Pigmentation

OH

191

OH

DXR O OP

Pyruvate + GlyceraldehydeTPP 3-phosphate

DXPS

OP

OH OH

OH

MEP

DXP

OPP OH

HMBPP IDS CH2OPP

IDS IPPI

CH2OPP

IPP

DMAPP

GGPPS CH2OPP

GGPPS

GPP

CH2OPP

GGPP

--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Carotenoid pathway

2xGGPP PSY

PDS

Phytoene cis-ζ-Carotene ZDS + CRTISO 17

18 7

3 16

1

5

11

All-trans-lycopene

15

9

13

13⬘ 11⬘

15⬘

9⬘

5⬘

1⬘ 16⬘

3⬘

7⬘

17⬘

18⬘

bLCY

γ-Carotene eLCY

α-Carotene

17 1 3 5

bLCY

18'

16 7

9

18

11

13

15

15⬘

13⬘ 11⬘

9⬘

7⬘ 16⬘

5⬘ 3⬘ 1⬘ 17⬘

β-Carotene

Fig. 8.5. Biosynthesis of isopentenyl diphosphate (IPP) and carotenoids of the carotene type. Enzyme and compound abbreviations are defined in the text or Table 8.1, except for HMBPP (1-hydroxy-2-methyl-2-(E)-butenyl-4-phosphate), IDS (isopentenyl diphosphate/dimethylallyl diphosphate) and TPP (thiamine pyrophosphate). The numbering of the key carbons of some representative compounds is shown.

192

K. Davies and K. Schwinn

Formation of phytoene The first product formed in the carotenoid-specific pathway is the C40 carotene, phytoene (Fig. 8.5), a colourless molecule having only 3 conjugated double bonds. The chain elongation that is required to form phytoene begins with the condensation of IPP and DMAPP to form the C10 compound geranyl pyrophospate (GPP). Further sequential additions of IPP yield the C20 molecule geranylgeranyl pyrophosphate (GGPP). The conversion of IPP/ DMAPP to GGPP is catalysed by GGPP synthase (GGPPS). The subsequent formation of phytoene (in the 15-cis isomer form) is through a two-step process involving the condensation of two molecules of GGPP by phytoene synthase (PSY). Formation of carotene pigments A series of desaturation, isomerization and cyclization reactions form a variety of other carotenes from phytoene (Fig. 8.5). The desaturation reactions increase the number of conjugated double bonds in the carotene backbone, creating the chromophores that are the basis for the coloured carotenoids. The conversion of phytoene to the first pigment produced in the pathway, z-carotene (pale yellow), is catalysed by phytoene desaturase (PDS) that carries out two desaturations. The z-carotene desaturase (ZDS) carries out two additional desaturations to convert z-carotene to the orange pigment prolycopene (tetra-cis-lycopene). Before further conversions can take place, prolycopene must be isomerized to the all-trans form. In chloroplasts, all-trans-lycopene may be formed through photoisomerization. However, at least in chromoplasts, it appears that the production of alltrans-lycopene is dependent on the activity of carotenoid isomerase (CRTISO) (Isaacson et al., 2002, 2004; Park et al., 2002). Lycopene, which gives red pigmentation to organs such as tomato (Lycopersicon esculentum) fruit, is less relevant to flower colour than the lycopenederived carotenes, which give yellow-orange to orange colours. These carotenes are formed by two enzymes, which introduce different cyclic end groups. These conversions form the first branch point in the pathway (Fig. 8.5). In the b,b branch of the pathway, lycopene b-cyclase (bLCY) converts lycopene to the orange pigment b-carotene through the introduction of b-ionone rings at each end of the lycopene molecule. In the b,e branch, a-carotene (yellow) is formed from lycopene through the action of lycopene e-cyclase (eLCY), which forms a e-ring at one end of lycopene, and bLCY, which cyclizes the other end. eLCY generally cannot cyclize both ends of lycopene, although rare examples are known (Cunningham and Gantt, 2001). Formation of xanthophyll pigments The first xanthophylls formed from the b,e and b,b branches of the pathway are the yellow-orange monohydroxylated compounds: zeinoxanthin derived from a-carotene; and b-cryptoxanthin, derived from b-carotene (Fig. 8.6).The conversions involve introduction of a hydroxyl group to a single b-ring of the molecules. The hydroxylation is catalysed by b-hydroxylase (bOH), a nonhaem di-iron oxidase. Additional hydroxylation catalysed by e-hydroxylase (eOH), a CytP450 enzyme, or bOH leads to the formation of lutein (yellow) from zeinoxanthin and zeaxanthin (orange) from b-cryptoxanthin, respectively.

Developmental Control and Biotechnology of Floral Pigmentation

193

A

α-Carotene OH

bOH

Zeinoxanthin e OH

HO

OH

Lutein

B

HO

bOH

β-Carotene

bOH

β-Cryptoxanthin OH

HO

ZEP

Zeaxanthin OH

O HO

ZEP

Antheraxanthin OH O

O HO NXS

Violaxanthin OH O

C OH

Neoxanthin

HO

Fig. 8.6. Biosynthesis of xanthophylls, starting from a-carotene (A) or b-carotene (B). The pathway shows only the xanthophylls that are common in higher plants. Enzyme abbreviations are defined in the text or Table 8.1. It should be noted that in some species a-cryptoxanthin accumulates, due to eOH activity on a-carotene.

194

K. Davies and K. Schwinn

Epoxides of zeaxanthin, the monoepoxidated antheraxanthin (yellow), and diepoxidated violaxanthin (yellow), frequently contribute to flower colour and are formed through the action of zeaxanthin epoxidase (ZEP). The last xanthophyll formed by the b,b branch of the pathway that is common to many plants is neoxanthin (yellow), produced by the activity of neoxanthin synthase, an enzyme that awaits unequivocal characterization (Cuttriss and Pogson, 2004). Lutein is the most common xanthophyll floral pigment derived from the b,e branch of the pathway. Some species produce rare xanthophylls in their flowers (and fruit), which may be limited in occurrence to the specific species and its close relatives. However, relatively little is known about the enzymes involved. The bestdescribed enzyme is capsanthin-capsorubin synthase, which produces the characteristic pigments of red pepper (Capsicum spp.) fruit from antheraxanthin and violaxanthin. Furthermore, esterification is a common characteristic of floral carotenoids, and may aid in the accumulation of the high levels of pigment found in chromoplasts (see also section below on Regulation of Pigment Synthesis in Flowers). Pigments derived from carotenoid catabolism The vividly coloured apocarotenoids are formed by cleavage of the normal C40 carotenoid structure. They occur in various plant tissues including roots, stems and flowers. Saffron, an expensive spice made from the dried red styles of saffron flowers, derives its distinctive colour from C20 apocarotenoid crocetin glycosides. These are probably formed by cleavage of zeaxanthin by zeaxanthin 7,8(7’,8’)-cleavage dioxygenase, a plastid-localized enzyme that removes the cyclic rings from both ends (Bouvier et al., 2003a). The crocetin dialdehyde product is probably subsequently acted upon by an aldehyde oxidoreductase and a UDPG-glucosyltransferase to produce water-soluble crocetin glycosides that are transferred from the plastid to the vacuole via direct interactions between the two organelles (Bouvier et al., 2003a). Another example is found in the tropical bush Brunfelsia orellana, which accumulates bixin apocarotenoids in the floral parts and seed coat. The extract of the seed coat is the basis of the yellow-orange food colourant annatto. Cleavage of lycopene by a dioxygenase yields bixin aldehyde, which is subsequently acted upon by bixin aldehyde dehydrogenase to produce norbixin. Conversion of norbixin to the major pigment bixin (and a bixin dimethyl ester) is achieved by the action of norbixin carboxyl methyltransferase (Bouvier et al., 2003b). The biosynthesis of betalains Betalains are divided into two major groups: (i) the red-violet betacyanins; and (ii) the yellow betaxanthins. The base chromophore of both types is betalamic acid. Condensation of betalamic acid with cyclo-DOPA glucosides yields the common simplest betacyanins (betanidin and isobetanidin glucosides). These compounds can then be modified by additional glycosylation and acylation to yield other betacyanins. Betaxanthins result from the conjugation of betalamic acid with amino acids/amines.

Developmental Control and Biotechnology of Floral Pigmentation

195

The betalain biosynthetic pathway is relatively simple with only a few reactions that are enzyme catalysed (Fig. 8.7). In the formation of betalamic acid, the aromatic amino acid tyrosine is hydroxylated to form L-DOPA. Cleavage of the cyclic ring of L-DOPA then forms an unstable seco-DOPA intermediate, which spontaneously converts to betalamic acid. The enzyme catalysing the hydroxylation reaction has not been unequivocally determined, but studies suggest that it is a bifunctional tyrosinase or a tyrosine hydroxlyase (Zry¨d and Christinet, 2004). The cleavage reaction is catalysed by DOPA-4, 5-estradiol dioxygenase (DOD). The isolation and characterization of an encoding cDNA from Portulaca grandiflora has been reported, and it defined a novel gene family for non-haem dioxygenases that may be involved in aromatic compound metabolism (Christinet et al., 2004). Interestingly, the plant dioxygenase is phylogenetically unrelated to a DOPA-dioxygenase from Amanita muscaria, which was previously shown to complement betalain production in flowers of a P. grandiflora dod mutant (Mueller et al., 1997). The formation of betaxanthin pigments from the condensation of betalamic acid with amines/amino acids occurs spontaneously, as does the condensation of betalamic acid with cyclo-DOPA glucosides in betacyanin synthesis (Schliemann et al., 1999). Cyclo-DOPA is formed from L-DOPA through an oxidation reaction that has been attributed to the activity of the bifunctional tyrosinase suggested to be involved in the formation of L-DOPA from tyrosine (Zry¨d and Christinet, 2004). The conversion proceeds via an unstable DOPA-quinone intermediate, which spontaneously cyclizes to form cyclo-DOPA. Betacyanins are generally glycosylated and frequently acylated. O-glycosyltransferases that can use betanidin as well as flavonoid substrate are present in cell cultures of Dorotheanthus bellidiformis, and encoding cDNAs have been isolated (Strack et al., 2003). However, recent biochemical studies involving several species indicate that glycosylation (at the C-5 or C-6) predominantly occurs at the cycloDOPA step, prior to the formation of betanidin (Sasaki et al., 2004). In contrast to the flavonoid pathway, in which acylation reactions are thioester dependent, acylation of betacyanins involves b-acetal esters as acyl donors, and may involve serine carboxypeptidase-like proteins (Strack et al., 2003).

Regulation of Pigment Biosynthesis in Flowers The induction of the pigment biosynthetic genes during flowering requires both developmental and environmental signals. Although factors such as temperature and water stress influence flower coloration, light is the principle environmental signal. Light, through direct exposure of the flowers or the leaves, has been shown to be required for full floral coloration in many species (discussed in Weiss, 2000; Meng and Wang, 2004). However, there is little data on the molecular mechanisms that may be involved in mediating the light signals in flowers. This is in contrast to the wealth of data on the light regulation of pigment production in vegetative tissues of model species such as arabidopsis (Arabidopsis thaliana) and maize, especially during seedling photomorphogenesis (see section below on Regulation of the Production of Flavonoid

196

K. Davies and K. Schwinn

Shikimate pathway

Chorismate Arogenate pathway COOH H

COOH H

COOH H

NH2

NH2

NH2

TYR

TYR

(TYOH?)

O

OH

S HO

O

OH

OH

Tyrosine

L-DOPA

COOH

Dopaquinone N H

HO

DOD

cyclo-DOPA

COOH H

CD5GT/ CD6GT

O

NH2

R1 O

COOH

S H O

OH O

N H

R2 O

HOOC

N H

OH

4,5-seco-DOPA

H

COOH

H

cyclo-DOPA 5-O -glucoside (R1 = Glc; R2 = H) cyclo-DOPA 6-O -glucoside (R1= H; R2 = Glc)

Betalamic acid

Amino acids/ amines S

S Betaxanthins R1O

H N+

5

Betacyanins

4

3 2 N+

−

COO

R2O

6

HOOC

13 14 H 15 N H

COOH

e.g. Indicaxanthin

HOOC

H

N H

COO− N+

R2 O

7 12

H

R1 O

COO−

H

11 18 17 COOH

H HOOC

Betanidin 5-O -glucoside (R1 = Glc; R2 = H) Betanidin 6-O -glucoside (R1 = H; R2 = Glc)

N H

COOH

Isobetanidin glucosides

Modification reactions Betanidin/isobetanidin glucosides with acyl groups or additional sugar moieties

Fig. 8.7. Biosynthesis of betalains. Most betacyanins have glucose as the O-linked sugar at the C-5 or C-6 positions. The route to the less common descarboxy-betacyanins is not shown. Abbreviations are defined in the text except for CD5GT (cyclo-DOPA 5-O-glucosyltransferase), CD6GT (cyclo-DOPA 6-O-glucosyltransferase), S (spontaneous conversion), TYR (tyrosinase) and TYOH (tyrosine hydroxylase).

Developmental Control and Biotechnology of Floral Pigmentation

197

Precursors and the reviews of Weisshaar and Jenkins, 1998; Weiss, 2000; Davies and Schwinn, 2003). The developmental signals for pigmentation are not well understood. The MADS-box genes (A, B and E class) that control petal organogenesis have a role in later events of floral development, including anthocyanin biosynthesis, but the specifics of their action remain to be elucidated. In antirrhinum, the B-class gene Deficiens is expressed well after petal identity is established (Zachgo et al., 1995), and reduced expression of CHS and F3H has been found in petal regions lacking activity of the encoded protein (Bey et al., 2004). In petunia, inhibition of floral binding protein2 expression, a gene equivalent to the sepallata E class genes of arabidopsis, results in plants with small, aberrant green corollas that fail to produce anthocyanins or express CHS (Angenent et al., 1994; Ferrario et al., 2003). The endogenous developmental signals that synchronize pigmentation with the appropriate period of flower fertility have been examined in a few species. In particular, removal of the anthers will prevent normal anthocyanin pigment formation in flowers of petunia (reviewed in Weiss, 2000) and Viola cornuta (Farzad et al., 2002, 2003). In V. cornuta, the open flowers change from white to purple only when they are both exposed to light and pollinated, with pollination triggering specific changes in anthocyanin biosynthetic gene expression. In petunia, it is thought that anther-produced gibberellic acid (GA) is a developmental signal translocated to the corolla to modulate petal growth and induce anthocyanin pigmentation. GA1 and GA4 have been identified in petunia anthers and corollas, and exogenous GA1, GA3 or GA4 will compensate for anther excision and promote gene transcription for several anthocyanin biosynthetic enzymes. Other plant hormones may also influence anthocyanin biosynthesis, with exogenous methyl jasmonate application promoting anthocyanin biosynthetic gene expression and pigmentation, and abscisic acid (ABA) inhibiting the action of GA3 (Weiss, 2000). However, it is not known whether these results with exogenous application reflect processes occurring normally in flower development. GA3 also upregulated carotenoid production and gene expression for two carotenoid-associated proteins (CHRC and CHRD) in cucumber (Cucumis sativus) petals, although transcript levels for PSY were not affected (Vishnevetsky et al., 1997). It is not clear what other developmental signals may be active, in conjunction with GA, or in species in which GA may not be important. Ethylene may trigger colour change in some, but not all, species that show colour development in open flowers (Farzad et al., 2002). Adequate carbohydrate level is required for normal development and full anthocyanin pigmentation of flowers, and exogenously supplied sugar upregulates CHS expression in arabidopsis, petunia and soybean (Glycine max) (Weiss, 2000). However, although sugars are required directly as part of anthocyanin biosynthesis, it is likely that sugar status is a general influence on flower development, rather than a specific signal that controls the timing of pigment formation. Although there is physiological data on the factors influencing betalain production in plants and plant cell cultures, there are few studies on flowers and no molecular data on regulation of gene expression. For non-floral tissues,

198

K. Davies and K. Schwinn

both light and hormones, particularly auxin and cytokinin, have been found to influence betalain production (Strack et al., 2003; Zry¨d and Christinet, 2004). Light can affect the quantity and type of pigment produced, but is not an absolute requirement for their production. Cytokinins induce betalain production through new gene transcription in, e.g. seedlings of Amaranthus caudatus, while auxin has been shown to induce betalain production in P. grandiflora. For carotenoids, their biosynthesis must be coordinated with plastid biogenesis and development to provide both the biosynthetic apparatus and a carotenoid sink, and there is evidence that the two processes can be tightly linked. An increase in plastid replication and a transition from chloroplasts to chromoplasts usually accompanies coloration in carotenoid-pigmented flowers. In marigold (Tagetes erecta), genes related to plastid division are upregulated during flower development (Moehs et al., 2001). Similarly, in fruit of the tomato hp1 mutant, the increased pigmentation relative to wild-type fruit is due to a greater concentration of chromoplasts (Cookson et al., 2003). In cucumber flowers, the amount and/or type of the carotenoids accumulating influences CHRC protein accumulation post-transcriptionally (Vishnevetsky et al., 1999b). Thus, plastid–nuclear signalling is likely to occur when environmental stimuli, oxidative stress or metabolite feedback indicate required changes in plastid structure, number or function. Similarly, there is evidence that flavonoid biosynthesis and vacuolar development may be tightly linked (Abrahams et al., 2003). How are the various environmental and developmental signals mediated to trigger the production of active pigment biosynthetic enzymes? Amongst the various steps leading to production of pigment in flowers, it is thought that transcription rates for the biosynthetic genes are the key regulatory targets. For anthocyanin synthesis, it has been shown that major increases in transcript abundance for the biosynthetic genes precede pigment production, and that this is due to increased transcription of the genes. There are few studies for flowers pigmented by carotenoids or betalains. However, recent studies have shown that increases in transcript abundance occur for several carotenoid biosynthetic genes concomitant with flower coloration in daffodil (Narcissus pseudonarcissus), Gentiana lutea, marigold and Sandersonia aurantiaca (Schledz et al., 1996; Moehs et al., 2001; Zhu et al., 2002, 2003; Nielsen et al., 2003). Similarly, transcript abundance for the betalain biosynthetic enzyme DOD correlates with pigmentation in flowers of P. grandiflora and cell lines of beet (Christinet et al., 2004). Whether these changes in transcript abundance are due to changes in transcription rate or transcript stability has not been determined. However, in tomato fruit, which like flowers accumulate high levels of carotenoids in chromoplasts, major increases in transcription of some of the biosynthetic genes are observed prior to carotenoid pigmentation (reviewed in Bramley, 2002; Fraser and Bramley, 2004). Furthermore, the promoter of the tomato pds gene confers significant upregulation of b-glucuronidase (GUS) reporter gene activity in carotenoid-containing flowers and fruit of transgenic tomato plants, as well as a responsiveness to changes in end product concentrations in leaves (Corona et al., 1996).

Developmental Control and Biotechnology of Floral Pigmentation

199

In addition to transcriptional control, there is strong evidence that posttranscriptional and post-translational mechanisms contribute to the regulation of both carotenoid and flavonoid production. Along with the previously mentioned example of CHRC, there is evidence for such mechanisms in carotenoid synthesis during seedling photomorphogenesis, and changes in PSY and PDS enzyme activity in daffodil flowers may be associated with membrane association of carotenoid biosynthetic enzyme complexes and metabolic channelling (Schledz et al., 1996; Cunningham and Gantt, 1998; Bramley, 2002; Fraser and Bramley, 2004). Membrane associated multi-enzyme complexes are also thought to occur for flavonoid biosynthesis, with attachment via the CytP450 enzymes such as C4H (Achnine et al., 2004). Such complexes may be important in metabolic channelling of substrates within the phenylpropanoid pathway (Winkel, 2004). The existence in some species of multi-gene families for some carotenoid and flavonoid biosynthetic genes may allow for isoenzymes with differing activities to be incorporated into distinctive biosynthetic enzyme complexes. Large scale studies of arabidopsis gene expression have revealed rhythmic variation in transcript abundance for several phenylpropanoid and carotenoid genes in vegetative tissues, including the flavonoid regulatory gene Pap1, and the rhythms are potentially generated by changes in RNA stability (Harmer et al., 2000; Staiger, 2002). Post-transcriptional regulation may also influence the accumulation of BZ2 protein in maize (Pairoba and Walbot, 2003). Regarding flowers, there are few examples, but diurnal variation in transcript abundance has been found for DFR in the coloured spathe of anthurium (Anthurium andraeanum), which is the main coloured organ associated with the flowers of this species (Collette et al., 2004). The relative importance of post-transcriptional/translational mechanisms in regulating the production of the high concentrations of carotenoids and anthocyanins in flowers is not known. Although the emphasis on transcriptional regulation may in part be due to lack of appropriate studies, transcription remains the principal point of control identified to date for directing production of these pigments in flowers. The rate of transcript production from the biosynthetic genes is controlled through the action of transcription factors (TFs). TFs are proteins that bind in a sequence-specific manner to DNA motifs (cis-elements) within target genes, usually in gene promoters, and, through direct or indirect interaction with the basal transcription machinery, cause a change in the rate of transcription initiation. TFs are categorized into classes or families on the basis of common amino acid sequences for DNA-binding and protein–protein interaction domains (e.g. MADS box and MYB). A plant species may have many members within a TF family that have diverse regulatory roles (e.g. the arabidopsis MYB family, with approximately 125 members). TFs may either activate or repress transcription, and generally have defined domains for these functions that are functionally separable from the DNA-binding domains. Commonly, TFs form regulatory complexes with other TFs and cofactors, and such combinatorial control allows for greater regulatory diversity. Also, a range of post-transcriptional translational mechanisms, such as reversible phosphorylation, may

200

K. Davies and K. Schwinn

modulate TF activity. TFs often regulate the activity of multiple target genes, a key advantage for their use in plant biotechnology. Despite the importance of carotenoids in biology, there has been little progress in elucidating the transcriptional control contributing to the regulation of their synthesis. Regulatory data from studies on fruit ripening (tomato and pepper) and seedling development (arabidopsis) include identification of candidate psy gene promoter cis-elements, identification of candidate regulatory mutants and characterization of some of the general fruit ripening control genes (reviewed recently in Fraser and Bramley, 2004). However, no specific TFs relating to carotenoid biosynthesis have been identified. Furthermore, aside from the previously mentioned indications of transcriptional control of the biosynthetic genes, there is a lack of other data specifically for regulation of carotenoid production in flowers, although some pale-flowered varieties of marigold have reduced levels of transcript from several carotenoid genes, suggesting that they may be regulatory mutants (Moehs et al., 2001). For betalains, there is no published data on any TF. In contrast, much is understood about transcriptional control of flavonoid production, particularly that of anthocyanins in flowers and vegetative tissues, and proanthocyanidins (PAs or condensed tannins) in seeds, through studies in the model species antirrhinum, arabidopsis, maize and petunia. In the remainder of this section we discuss the TFs identified to date that regulate flavonoid biosynthesis in flowers.

Regulation of flavonoid production in flowers For flavonoids, much data have been published on the regulation of the pathway. We cover here only those data relevant to flowers, and reviews of the regulation of flavonoids in other parts of the plant may be found in Martin et al. (2001), Vom Endt et al. (2002), Yamazaki (2002), Davies and Schwinn (2003), and Springob et al. (2003). In addition, a detailed review of the knowledge on the regulatory mutations affecting pigmentation in antirrhinum and petunia is presented in the previous edition of this book (Martin and Gerats, 1993); and information on the structure and function of some of the types of TF mentioned in this section can be found in the reviews of Eulgem et al. (2000), Petroni et al. (2002) and Heim et al. (2003). Regulation of the production of flavonoid precursors The production of anthocyanins in flowers requires not only the activity of the specific flavonoid genes but also those of the general phenylpropanoid pathway for production of HCAs. Furthermore, the route to malonyl-CoA formation may also need to be upregulated. Detailed studies on the regulation of the genes required for HCA production, however, have been conducted principally in relation to their induction in response to biotic and abiotic (e.g. UV light) stress, and there is little information for flower development. Many of the studies have included CHS, as it is the first specific step of the flavonoid pathway. Only a brief mention of some of the types of TF identified is given here, as their relevance to flower pigmentation is uncertain.

Developmental Control and Biotechnology of Floral Pigmentation

201

Light upregulates phenylpropanoid biosynthesis in a range of plant tissues, and for arabidopsis the photoreceptors phytochrome B and cryptochrome have been shown to be involved in the induction process (Jenkins et al., 2001; Hemm et al., 2004). Key to mediating gene expression in response to light is the light-responsive unit (LRU), comprised of an MYB recognition element (MRE) and an ACGT-containing element (ACE) (reviewed in Martin et al., 2001; Davies and Schwinn, 2003). TFs of the basic region/leucine zipper (bZIP) type bind to ACE boxes in a range of promoters, and some of these are regulators of HCA and flavonoid production (reviewed in Menkens et al., 1995; Jakoby et al., 2002). For example, HY5 is a bZIP protein that is a direct activator of genes of phenylpropanoid biosynthesis and other lightresponsive pathways in arabidopsis (Hemm et al., 2004; Ulm et al., 2004). In parsley, bZIPs termed common plant regulatory factors (CPRFs) have been implicated in regulation of phenylpropanoid genes via the LRU ACE (reviewed in Jakoby et al., 2002). Several MYB factors have been identified that can activate PAL and CHS genes through the LRU MRE, or through related promoter elements in response to other signals. These include AmMYB305 and AmMYB340 from antirrhinum (Sablowski et al., 1994; Moyano et al., 1996), PcMYB1 from parsley (Feldbru¨gge et al., 1997), PhMYB3 from petunia (Solano et al., 1995) and NtMYBAS1, NtMYBAS2 and NtMYB2 (Sugimoto et al., 2000; Yang et al., 2001) from tobacco (Nicotiana). In a unique example to date, TFs having similarities with the mammalian Ku autoantigen protein, which controls DNA recombination and transcription, are involved in phenylpropanoid and CHS gene regulation in response to external abiotic and biotic signals in French bean (Phaseolus vulgaris) (Yu et al., 1993; Lindsay et al., 2002). Not all of the TFs identified are transcriptional activators. AtMYB4 from arabidopsis and AmMYB308 and AmMYB330 from antirrhinum are repressors of phenylpropanoid biosynthetic gene activity (Tamagnone et al., 1998; Jin et al., 2000). AtMYB4 plays a role in the UV-B light regulation of sinapate ester formation. These MYB proteins share amino acid sequence motifs in the C-terminal domain that are putatively involved in the repressor activity, perhaps through protein–protein interactions with components of the basal transcription machinery. Another component of the regulation of gene expression is the targeting of the activating or repressing TFs for degradation. TFs involved in photomorphogenic development, including some regulating phenylpropanoid genes, are targeted by the constitutive photomorphogenesis (COP) system. The COP system is an assemblage of proteins that controls the ubiquitin-mediated lightdependent degradation of downstream signalling components of many photoreceptors, e.g. HY5 (reviewed in Hardtke and Deng, 2000; Schwehheimer and Deng, 2001). TFs regulating anthocyanin production Transcriptional control of anthocyanin biosynthetic gene activity is one of the best-characterized examples of this form of regulation in plants. It was first described for vegetative and seed tissues of maize, subsequently for petals of

202

K. Davies and K. Schwinn

antirrhinum and petunia, and more recently for vegetative tissues of arabidopsis. All of these studies were aided significantly by the easy identification of mutants, due to the visual phenotypes. Transcriptional activation of anthocyanin biosynthetic genes is through TFs of the R2R3-MYB and basic Helix-Loop-Helix (bHLH) (or MYC) type, a mechanism that is conserved between widely divergent plants. Indeed, the combinatorial activities of MYB/bHLH partnerships regulate many other plant processes, an example being the formation of trichomes (Szymanski et al., 2000). In antirrhinum, arabidopsis, maize and petunia, gene families encoding both types of TFs have been found to regulate anthocyanin synthesis (Davies and Schwinn, 2003; Springob et al., 2003). Variations in gene expression patterns of family members, and likely in the activity of the encoded proteins, allow for fine control of the timing and amount of anthocyanin produced, and the complex patterns of spatial distribution that can occur. The MYB and bHLH TFs function cooperatively, directly interacting within the transcriptional complex (Goff et al., 1992). The complex is bound to DNA in part through the MYB member, which recognizes MYB-specific sequences (Sainz et al., 1997). It has not been resolved whether the bHLH member also directly binds DNA. It does function at least in part through a distinct cis element known as ARE, either directly or through interaction with an as yet unidentified ARE-binding protein (Hernandez et al., 2004). WD-repeat (or WD40) proteins also play a role in the control of anthocyanin production. This role was first determined from studies on the whiteflowered an11 mutant of petunia (de Vetten et al., 1997); and they have since been found in other species (Walker et al., 1999; Sompornpailin et al., 2002; Carey et al., 2004). WD-repeat proteins lack DNA-binding domains, but can bind with other proteins and promote interactions between the bound proteins (van Nocker and Ludwig, 2003). Recently, it has been indicated in studies of the TFs regulating PA production in arabidopsis that the WD-repeat protein is a part of the transcriptional complex (Baudry et al., 2004). MYB, bHLH and WD-repeat factors with a proven role in regulating anthocyanin production in petals (and published before mid-2003) are listed in Davies and Schwinn (2003). Notable recent additions are the MYB factor GMYB10 of Gerbera hybrida (gerbera) (Elomaa et al., 2003) and the bHLH factor IVORY SEED-VARIEGATED of morning glory (Ipomoea tricolor, Park et al., 2004). Of particular interest are the ROSEA1, ROSEA2, VENOSA, DELILA and MUTABILIS TFs of antirrhinum, which determine the complex floral patterning of pigmentation observed in this species (Martin et al., 2001). Despite the conserved role of MYB, bHLH and WD-repeat proteins in controlling anthocyanin pigmentation, and the highly conserved structures of the proteins, the anthocyanin biosynthetic genes that are the reported targets of the TFs vary between species. In morning glory petals, and maize and perilla (Perilla frutescens) vegetative tissues, the characterized TFs essentially activate all the genes from CHS to those involved in transport of the anthocyanin to the vacuole (Grotewold et al., 1998; Saito and Yamazaki, 2002; Park et al., 2004). However, based on studies of flower colour regulatory mutants, the

Developmental Control and Biotechnology of Floral Pigmentation

203

TF genes in Antirrhinum, gerbera and petunia that are similar to those in maize regulate only a subset of the genes, i.e. the ‘late biosynthetic genes’ (LBGs), starting from either F3H or DFR through to, presumably, the end of the pathway including genes encoding transport proteins (reviewed in Martin and Gerats, 1993; Davies and Schwinn, 2003). The corresponding ‘early biosynthetic genes’ (EBGs) in these species are under separate regulation, and as yet no TFs have been reported with a role in coordinating EBG expression for pigmentation, although MYB proteins are known that can activate these genes (see section above on the Regulation of the production of flavonoid precursors). Division of the pathway into modules under independent control enables metabolic flexibility, e.g. for separate regulation of flavone or flavonol biosynthesis for the formation of floral nectar guides. Variations in the partitioning of pathway regulation are becoming apparent as more species are characterized. DFR is regulated separately from the other biosynthetic genes in the spathe and spadix of Anthurium (Collette et al., 2004), and in V. cornuta ANS may be a key regulatory target (Farzad et al., 2003). Regarding the B-ring hydroxylases, studies show that in petunia they are regulated separately from each other (de Vetten et al., 1999). The genes encoding F3’,5’H (and the cyt b5 protein) are regulated with the LBGs by the anthocyanin TFs, while the gene for F3’H is not. However, both F3’H and F3’,5’H are upregulated in leaves of transgenic petunia expressing a maize anthocyanin regulator, the bHLH factor LC (Bradley et al., 1998), but, interestingly, not in tomato plants overexpressing LC and C1 (Bovy et al., 2002). Despite the success in elucidating aspects of the transcriptional regulation of anthocyanin biosynthesis, much is still not known, including how activities of the TFs are modulated or the role of repressor TFs in regulating pigmentation. Although there are MYB and bHLH proteins that are known to have a repressive effect on anthocyanin biosynthetic gene transcription (Paz-Ares et al., 1990; Burr et al., 1996; Aharoni et al., 2001; Chen et al., 2004), none have been reported that are active in flower coloration. It has recently been proposed that one of the roles of the bHLH factors in the activation of the target genes is to relieve the MYB factors from the effect of an inhibitory factor (Hernandez et al., 2004). Aside from the apparent involvement of MADS-box genes in petal pigmentation (see section above on Regulation of Pigment Biosynthesis in Flowers), there have been few developments in deciphering the regulatory cascade leading to flavonoid pigmentation in flowers. In petunia the TFs regulating anthocyanin synthesis in the petals are the bHLH factors: AN1 and JAF13 and AN2, which is an MYB protein (Quattrocchio et al., 1998, 1999; Spelt et al., 2000, 2002). Transgenic experiments with AN1-glucocorticoid receptor constructs have demonstrated that AN1 directly activates anthocyanin biosynthetic genes, as well as an myb gene of unknown function (Pmyb27). However, An1 gene expression is dependent on the activity of An2, while Jaf13 expression is not. Furthermore, Jaf13 expression does not compensate for loss of An1 activity in the flowers. It has not been determined

204

K. Davies and K. Schwinn

whether JAF13 and AN2 are themselves direct regulators of the biosynthetic genes. This hierarchical control has not been found for the TFs in maize (Carey et al., 2004), nor has it been reported for the factors in the other species studied to date. Some progress has been made in deciphering the regulatory cascades leading to seed pigmentation. However, the relevance of the data to floral pigmentation is uncertain. In maize, the only characterized direct activator of an anthocyanin TF gene is Viviparous1 (Vp1) (and its orthologues from other species), which encodes a unique type of TF that is required for seed maturation (McCarty et al., 1991; Hattori et al., 1992). Genes for a range of proteins active upstream in signalling cascades for regulation of PA production in arabidopsis have been identified. These include proteins of the homeobox (HDGLABRA2 type), WRKY, WIP zinc finger and MADS-box type, and reviews of their roles in flavonoid biosynthesis are available in Marles et al. (2003), Tanner (2004) and Dixon et al. (2005). In addition to their role in controlling anthocyanin biosynthesis, some of the regulatory factors have been found to control other aspects in epidermal cell differentiation. Based on mutant studies, the WD-repeat proteins of arabidopsis (TRANSPARENT TESTA GLABRA1, TTG1) and petunia (AN11) are involved in the control of several developmental aspects, likely through interaction with different bHLH factors involved in the various regulatory pathways. TTG1 regulates trichome, seed mucilage and PA production (Walker et al., 1999; Baudry et al., 2004), and AN11 influences seed coat development (Spelt et al., 2002). Furthermore, the bHLH and MYB factors themselves can have additional regulatory roles outside of anthocyanin production. Some of the petunia TFs regulate vacuolar pH in petal cells (AN1, AN2) and influence seed coat development (AN1) (Spelt et al., 2002), and one of the antirrhinum MYBs appears to regulate pH (ROSEA1; K. Schwinn, unpublished results). Regulation of the production of other floral flavonoid pigments/co-pigments The regulation of the production of the non-anthocyanin flavonoids in flowers is not as well understood as for anthocyanins. Some of the TFs identified as regulators of phenylpropanoid genes may have a role in regulating the flavonoid EBGs for production of flavones and flavonols in flowers, although there is limited data on this. AmMYB305, and the equivalent protein from tobacco, can bind to and activate a ‘P-box’ MRE that is linked to petal-enhanced expression of phenylpropanoid genes (Sablowski et al., 1994). Furthermore, AmMYB305 activates through the promoters of the CHI and F3H genes of antirrhinum (Moyano et al., 1996). Structurally similar is AmMYB340, which may serve a related function to AmMYB305, as it also binds to the P-box and regulates CHI (Moyano et al., 1996). In contrast to the anthocyanin-related TFs, these MYB factors were able to bind and activate without a bHLH partner. In petunia, the pollen-specific flavonol 3-O-galactosyltransferase, which may be involved in production of the fertility-important kaempferol and quercetin 3-O-(2’’-O-glucopyranosyl)-galactopyranosides, is regulated independently of a MYB TF identified that controls pollen anthocyanin pigmentation (AN4) (Quattrocchio et al., 1993; Miller et al., 2002).

Developmental Control and Biotechnology of Floral Pigmentation

205

Genetic Modification of Flower Pigmentation Novelty is a key driver of the commercial success of ornamental crops. There is much scope to introduce novel flower colours, as some of the leading ornamental crops show only a narrow colour spectrum, whilst in others specific colours, like blue or yellow, are lacking. GM of pigment biosynthesis offers a route to colours outside the existing range, and in the long term is likely to have a major influence on ornamental breeding. Indeed, the first ornamental products from GM of flavonoid biosynthesis are already in the marketplace. Carnations (Dianthus) with novel mauve and violet flower colours are being grown in South America and Australia for sale in Australia, Japan and the USA (Tanaka et al., 2005). Further products are ready for general release, including roses (Rosa hybrida) that accumulate delphinidin-based anthocyanins in the flowers and have mauve colours (Suntory Ltd, press release 8826, 6 June 2004). The potential exists to use GM approaches to introduce novelty into other commercially important species, as genetic transformation systems are available for most of the other major cut flower crops as well as some leading bedding plants, bulbs, pot plants and ground-covers (Deroles et al., 2002). Successful GM approaches to modify flower colour have primarily targeted the flavonoid pathway. Here we present an overview of the major GM approaches that have been taken to modify flower colour for the different pigment pathways, with a few representative examples discussed in detail. A comprehensive listing of individual experiments is given in Tables 8.2 to 8.5. Additional discussion of GM of flavonoid-based flower colour can be found in Schwinn and Davies (2004). Non-plant genes that code for coloured proteins or biosynthetic activities for novel pigments have also been used to alter plant/flower colour. Production of green fluorescent protein from the jellyfish Aequorea victoria has been introduced into lisianthus (Eustoma grandiflorum) flowers (Mercuri et al., 2001). Also, algal or bacterial genes have been used to modify carotenoid biosynthesis (Table 8.5). However, as the focus of this volume is on the molecular biology of flowering, we do not present the non-plant gene approaches in detail. Flavonoids There are numerous examples of modification of flavonoid biosynthesis in flowers of transgenic plants. There are also several examples for grains, fruit and vegetable crops, recently reviewed in Schijlen et al. (2004). The major methods are based on manipulation of pathway flux. The approaches to increasing, preventing or redirecting flux into or within the pathway have used up- or downregulation of the pathway using regulatory factors; introducing new biosynthetic activities; increasing specific endogenous biosynthetic activities; and abolishing branches of the pathway. The latter may cause substrate to accumulate or be directed into alternative biosynthetic branches. These approaches may target the coloured flavonoids directly, or change the

206 Table 8.2. Examples of genetic modification of flower colour using inhibition of flavonoid biosynthetic gene activity by sense or antisense RNA. Transgene

a

Effect on flower coloura

Reference

Carnation Chrysanthemum Gentian Gerbera Lisianthus Petunia Petunia Rose Torenia Torenia

Change from pink to white Change from pink to white Change from blue to pale blue or white Change from red to pink or cream Change from purple to white or patterns Change from red to white or patterns Change from purple to white or patterns Change from red to pale red Change from blue to pale blue or patterns Change from blue to pale blue, white or patterns

Gutterson (1995) Courtney-Gutterson et al. (1994) Nishihara et al. (2003) Elomaa et al. (1993) Deroles et al. (1998) van der Krol et al. (1988) Napoli et al. (1990); van der Krol et al. (1990) Firoozabady et al. (1994) Aida et al. (2000a,b) Suzuki et al. (2000); Fukusaki et al. (2004)

Carnation

Change from orange to white

Zuker et al. (2002)

Petunia Torenia Torenia

Change from purple to white or patterns Change from blue to pale blue or patterns Change from blue to white or patterns

Jorgensen et al. (2002) Aida et al. (2000a,b) Suzuki et al. (2000)

Petunia Torenia

Change from dark blue to pale blue or pink Change from blue to pink

Shimada et al. (2001); Jorgensen et al. (2002) Suzuki et al. (2000)

Lisianthus Petunia Petunia

Reddening effect Change from purple to red Change from white to pale pink

Nielsen et al. (2002) Holton et al. (1993b) Davies et al. (2003a)

Torenia

Paler flower colour

Ueyama et al. (2002)

Petunia

Change from purple to pink or patterns

Brugliera et al. (1994)

Only a general indication of the phenotype is given. Many publications include use of CHS in P. hybrida as a phenotypic marker for studying the antisense or sense RNA silencing process. Only the first reports are referenced here. b

K. Davies and K. Schwinn

CHSb Sense Sense and antisense Antisense Antisense Antisense Antisense Sense Sense Sense and antisense Sense or RNA interference (RNAi) F3H Antisense DFR Sense Sense and antisense Sense F3 ’,5 ’H Sense Sense FLS Antisense Antisense Antisense FNSII Antisense 3RT Antisense

Species modified

Developmental Control and Biotechnology of Floral Pigmentation

207

abundance of the colourless flavonoid co-pigments with which anthocyanins interact to develop their final colour. There is potential for using approaches that do not target flux. Changes in vacuolar pH, the subcellular site of flavonoid accumulation, or the non-flavonoid cellular components with which the flavonoid interacts can have dramatic effects on the colour the pigment manifests. Only limited gene technology is available as yet for targeting these aspects (e.g. for GSTs – section above on Vacuolar import – or for pH, see Schwinn and Davies, 2004). Preventing anthocyanin production The simplest approach for modifying flower colour is to prevent flavonoid pigment formation by inhibiting production of a key biosynthetic enzyme, such as CHS. This has been done numerous times in several species (reviewed recently in Tanaka et al., 2005 and listed in Table 8.2). Indeed, the first examples in plants of gene inhibition by antisense RNA (van der Krol et al., 1988) and sense RNA (Napoli et al., 1990; van der Krol et al., 1990) were for CHS in petunia. Aside from the expected pale or white flower colours, this approach has had unexpected phenotypes. Loss of flavonoid production has resulted in male (e.g. Taylor and Jorgensen, 1992; Fischer et al., 1997) or female (Jorgensen et al., 2002) sterility in some (but not all) species. Furthermore, ordered and erratic corolla pigmentation patterns have been obtained in some species (e.g. lisianthus, see Plate 1.K). The factors governing the type of anthocyanin patterning produced, and the stability of the phenotype, are still not clear, but may relate to morphological and environmental signals, as well as differences in transgene structure (Jorgensen, 1995). Interestingly, no pigmentation patterns have been observed to date in GM experiments with chrysanthemum, gerbera and rose, species that traditionally lack patterned varieties. The introduction of transgenes for single-chain antibody fragments developed against enzymes such as DFR has also been used to try to achieve controlled reduction of flavonoid enzyme activity, but without clear phenotypic effects (e.g. De Jaeger et al., 1999). Redirecting substrate within the pathway Altering the activities of enzymes that produce colourless flavonoids or compete with the anthocyanin-forming enzymes for substrate can be used to modify anthocyanin levels in flowers. The first approach has been demonstrated in antisense FLS transgenic plants. Reduction of flavonol biosynthesis in lisianthus, petunia or tobacco resulted in an increase in anthocyanin content in the flowers (Holton et al., 1993b; Nielsen et al., 2002; Davies et al., 2003a). Using the second approach, genes encoding stilbene synthase, PKR or anthocyanidin reductase (ANR) controlled by a CaMV35S promoter were introduced into petunia or tobacco. A reduction of anthocyanin levels and paler flower colours was achieved due to diversion of substrate (Fischer et al., 1997; Davies et al., 1998; Joung et al., 2003; Xie et al., 2003). As would be anticipated, the products of the respective enzymes, stilbenes, 6’-deoxychalcones and flavan-3-ols (or their expected derivatives) accumulated in the flowers.

208

K. Davies and K. Schwinn

An alternative to reducing the activity of competing enzymes for increasing anthocyanin production overall is to increase the abundance of anthocyanin biosynthetic enzymes that may be flux control points. For example, anthocyanin production was enhanced in flowers of petunia and tobacco when CaMV35S:DFR was introduced (Polashock et al., 2002; Davies et al., 2003a), and co-introduction of both CaMV35S:DFR and CaMV35S:ANS into forsythia resulted in increased levels of anthocyanins, including novel production in the flowers (Rosati et al., 1997, 2003). The observed increases in anthocyanin production may have been due to either improved competition with other flavonoid enzymes for common substrate or improved ability to utilize an excess of substrate. Introducing novel flavonoid compounds Chalcones may offer GM routes to developing yellow flower colours in target ornamentals. Chalcones are pale yellow, usually ephemeral, intermediates in the biosynthesis of all flavonoids. While chalcones accumulate in flowers of some chi mutant lines in sufficient quantities to provide yellow pigmentation (e.g. carnation), it has not yet proved possible to generate their accumulation in transgenic plants by sense or antisense inhibition of CHI gene activity. An alternative approach to triggering chalcone accumulation, which has proved successful, has been to introduce CaMV35S:PKR (Davies et al., 1998). PKR catalyses the production of 6’-deoxychalcones, which are physiologically more stable than the common 6’-hydroxychalcones and cannot be used as substrates by CHI in many species. A white-flowered line of petunia expressing a CaMV35S:PKR transgene accumulated up to 50% of the petal flavonoids as 6’-deoxychalcones, resulting in pale yellow colours. Another approach for eliciting chalcone accumulation may be to introduce a gene encoding an enzyme catalysing a modification to chalcones (e.g. glycosylation) that would prevent isomerization (both spontaneous or CHI-catalysed). There are preliminary reports of the isolation of a cDNA for one such enzyme, chalcone 2’-O-glucosyltransferase, but no details of transgenic experiments have been published (Tanaka et al., 2005; Table 8.1). A key consideration for this approach in plants expressing CHI activity would be whether the introduced enzyme could compete with CHI for substrate or even access substrate. Aurones are bright yellow compounds, and they also appear to be an excellent prospect for GM approaches to yellow flower colours. Their precursors, chalcones, are ubiquitous, and the molecular cloning of the key aurone biosynthetic enzyme, AUS, has been reported (Nakayama et al., 2000). Ono et al. (2005) have published a preliminary account of the successful introduction of aurone production into torenia through introduction of AUS and chalcone 4’-O-glucosyltransfease transgenes.

CHALCONES AND AURONES.

Some of the leading ornamental species lack anthocyanins based on one or more of the major anthocyanidin types, most commonly delphinidin, but also on occasion pelargonidin. An obvious approach for biotechnology is to introduce F3’,5’H activity or inhibit F3’H and/or

ALTERING B-RING HYDROXYLATION.

Developmental Control and Biotechnology of Floral Pigmentation

209

F3’,5’H activity to trigger production of precursors for delphinidin- or pelargonidin-derived anthocyanins, respectively. Delphinidin-derived anthocyanins are usually needed to produce blue flower colours. As blue flower colours are highly valued in the floriculture industry, not surprisingly, one of the major GM targets has been the introduction of delphinidin biosynthesis into species such as carnation, chrysanthemum and rose. Notable successes include the previously mentioned carnation and rose cultivars with violet or mauve colours. Approaches have also been developed to maximize levels of delphinidinderived anthocyanins relative to those of other types (Okinaka et al., 2003; Tanaka et al., 2005). Despite the success using strategies based on the F3’,5’H, it has not been possible to engineer blue flower colours in the target species to date. This is because the presence of delphinidin-derived anthocyanins does not necessarily result in a blue flower colour. Indeed, some ornamental species that lack blue flower colours, such as cyclamen (Cyclamen persicum), impatiens, lisianthus, pelargonium and tulip, naturally produce delphinidin-derived anthocyanins in the flowers. Furthermore, there are a few examples in which blue flower colours are derived from pigments other than delphinidin. For example, the ‘Heavenly Blue Anthocyanin’ in blue-flowered morning glory is peonidin (Fig. 8.1) with six molecules of glucose and three molecules of caffeic acid attached (Goto and Kondo, 1991). Chemical studies on several species that have delphinidin-based blue flower colours have suggested that, in addition to the presence of those anthocyanin types, the correct vacuolar pH and intra- or intermolecular (co-pigmentation) interactions are required to trigger blue anthocyanin forms. Although the transgenic carnations meet some of these requirements, accumulating complex delphinidin-derived anthocyanins, and having flavone co-pigments and a relatively high petal vacuolar pH of 5.5, they are still not blue in colour (Fukui et al., 2003). This illustrates the difficulty of engineering blue pigmentation. Inhibition of the activities of the F3’H and/or F3’,5’H could enable production of pelargonidin-derived anthocyanins in ornamental crops that lack them, such as chrysanthemum (Schwinn et al., 1994). Of course, with these approaches it must first be ascertained whether the downstream enzymes can use the differently hydroxylated substrates. In a few cases, the lack of pelargonidin-derived anthocyanins may also be related to the endogenous DFR having weak activity with 4’-hydroxylated substrates. In the first published case of GM of flower colour, Meyer et al. (1987) introduced a maize CaMV35S:DFR transgene into petunia, enabling use of DHK and the subsequent production of pelargonidin-derived anthocyanins at significant levels. This resulted in the formation of orange flower colours, a first for this species. DFR in cranberry (Vaccinium macrocarpon), Cymbidium orchid (Cymbidium hybrida) and tobacco also shows absent or low selectivity for DHK, and so might be a target in these species for this GM approach (Johnson et al., 1999, 2001; Polashock et al., 2002). SECONDARY MODIFICATIONS. Secondary modifications influence the colour resulting from anthocyanins by altering the light absorbencies of the pigments and

210

K. Davies and K. Schwinn

shaping their tertiary structures in the vacuole (Goto and Kondo, 1991; Brouillard and Dangles, 1993). In particular, methylation has a slight reddening effect on the anthocyanin, and glycosylation of the A- and B-rings and associated acylation, is often linked with blue flower colours. Many cDNAs are now available for enzymes carrying out a range of secondary modifications of anthocyanins, and some of these have been used successfully to alter anthocyanin profiles in flowers of transgenic plants (see Tables 8.2 and 8.3 and the reviews of Schwinn and Davies, 2004; Tanaka et al., 2005). Glycosylation, acylation and (indirectly) methylation patterns were modified. However, only the experiments of Brugliera et al. (1994) produced a change in flower colour. Utilization of transcription factors As the activity of the anthocyanin biosynthetic genes appears to be determined primarily by the expression patterns of the regulatory genes, the identification of defined TF genes provides tools for modulating the amount as well as the temporal and spatial occurrence of anthocyanins in plants. The validity of this approach has been well established, not only for the use of transgenes for endogenous TFs in a target species but also for the utility of TF transgenes to engineer anthocyanin production in heterologous species. TF gene technology offers the great advantage of being able to upregulate multiple genes with a single transgene. This is illustrated by an experiment in petunia, in which a transgene for a maize bHLH factor (LC) upregulated at least eight biosynthetic genes in the leaves, resulting in plants with deep purple vegetation (Bradley et al., 1998). Table 8.4 provides an extensive listing of published studies. The remainder of this section highlights a few points of note arising from the use of anthocyanin-related TFs in biotechnology experiments. The overexpression of anthocyanin-related TFs has successfully enhanced production of not only anthocyanin pigments, but also other flavonoids such as PAs, flavonols and isoflavonoids, in a range of species. The reasons for this are not clear. It is possible that they increase flux into or alter flux within the flavonoid pathway, allowing the accumulation of other end products. Alternatively, they may activate genes involved in other branches of the phenylpropanoid pathway as part of their normal function, but such roles have not previously been characterized. However, the TF transgenes are often used in heterologous species, driven by a strong constitutive promoter (e.g. CaMV35S) that results in transcript and (presumably) protein accumulation grossly exceeding the normal physiological levels. Thus, the introduced TF may interfere with the endogenous regulatory environment. This may promote activity on gene promoter sites not associated with the functioning of the native TF protein, or result in ‘squelching’ effects on endogenous TF activity through the titration of interacting TFs (Gill and Ptashne, 1988). ‘Dose dependent’ repression of phenylpropanoid genes other than those regulated in wild-type plants has been demonstrated in CaMV35S:AtMYB4 transgenic arabidopsis (Jin et al., 2000). Furthermore, substantial upregulation of major pathways like those for phenylpropanoids or carotenoids may have serious secondary effects, e.g. changes in carbon supply to other pathways, alterations in the production of ABA or GA, or changes in the transport of auxin.

Transgene

Species modified

Effect on flower coloura

Reference

PKR

Petunia Tobacco Tobacco Petunia Petunia

Change Change Change Change Change

Tobacco Forsythia

Change from pink to dark pink Flower (and vegetative) anthocyanin pigmentation induced Change from pink/white to blue/purple Change from lilac to pink Reddening effect Change from pink to blue/purple Change from pink to blue Change from pale pink/red to magenta/deep red or patterns Change in pink shade No change in phenotypeb No change in phenotypec No change in phenotyped Change from pink to white; PA-like compounds produced

Davies et al. (1998); Tanaka et al. (2005) Joung et al. (2003) Fischer et al. (1997) Davies et al. (2003a) Meyer et al. (1987); Helariutta et al. (1993); Tanaka et al. (1995); Johnson et al. (2001) Polashock et al. (2002); Xie et al. (2004) Rosati et al. (2003)

STS DFR

DFR and ANS DFR and F3 ’,5’H F3 ’H F3 ’,5’H

F3GT A3 ’GT and A5GT AM3T ANR

Carnation Petunia Torenia Carnation Lobelia Petunia Tobacco Lisianthus Petunia Petunia Tobacco

from from from from from

white to pale yellow pink to pale pink pink to pale pink white to pink pale pink to orange or red

Tanaka et al. (2005) Brugliera et al. (1999) Ueyama et al. (2002) Tanaka et al. (2005) Tanaka et al. (2005) Holton et al. (1993a); Park and Kim (2000); Shimada et al. (2001); Mori et al. (2004) Shimada et al. (1999); Okinaka et al. (2003) Schwinn et al. (1997) Fukuchi-Mizutani et al. (2003) Suzuki et al. (2002) Xie et al. (2003)

Developmental Control and Biotechnology of Floral Pigmentation

Table 8.3. Examples of genetic modification of flower colour by the introduction of novel flavonoid biosynthetic activities or by increasing endogenous activities, using sense transgenes. Experiments involving simple complementation of flower colour mutants are not included.

a

Only a general indication of the phenotype is given. A change in flavonoid glycosylation and acylation occurred. c One new anthocyanin type, delphinidin 3,5,3’-tri-O-glucoside, was found. d Anthocyanins with novel malonylation were formed. b

211

TF type: transgene (source species)

Species modified

Effect on flavonoid productiona

Reference

MYB: Rosea1 (antirrhinum)

Lisianthus Petunia Arabidopsis

Anthocyanins increased Anthocyanins increased Anthocyanins and other phenylpropanoids increased Anthocyanins increased Anthocyanins increased No visible change in phenotype No visible change in phenotype No visible change in phenotype No increase in flavonoid content Anthocyanins increased Anthocyanin production in flowers reduced 3-Deoxyflavonoids, C-glycosylflavones, other phenylpropanoids and fluorescent compounds increased Anthocyanins increased Anthocyanin and flavonol production inhibited Anthocyanins increased Anthocyanins increased Anthocyanin and PA production inhibited

Schwinn et al. (2001) Schwinn et al. (2001) Borevitz et al. (2000)

MYB: AtMYB75 (PAP1) or AtMYB90 (PAP2) (arabidopsis) MYB: Gmyb10 (gerbera) MYB: C1 (maize)

MYB: c1-I (maize) MYB: P1 or P2 (maize)

MYB: Myb.Ph2 (petunia) MYB: FaMYB1 (strawberry) MYB: ant1 (tomato)

bHLH: B-Peru (maize) bHLH: Lc (maize)

White clover Tobacco Tobacco Tomato Arabidopsis Tobacco Tomato Lucerne White clover Arabidopsis Lucerne Lisianthus

Anthocyanins increased Anthocyanins increased No visible change in phenotype Anthocyanins increased Anthocyanins increased Anthocyanins and PAs increased and flavones decreased No visible change in phenotype

Borevitz et al. (2000) Elomaa et al. (2003) Ray et al. (2003) Lloyd et al. (1992) Lloyd et al. (1992) Bovy et al. (2002) de Majnik et al. (2000) Chen et al. (2004) Grotewold et al. (1998); Zhang et al. (2003) de Majnik et al. (2000) Aharoni et al. (2001) Mathews et al. (2003) Mathews et al. (2003) Hiratsu et al. (2003); Matsui et al. (2004) Mooney et al. (1995) Mooney et al. (1995) Ray et al. (2003) de Majnik et al. (2000) Lloyd et al. (1992) Ray et al. (2003) Bradley et al. (1999)

K. Davies and K. Schwinn

Modified MYB: PAP1 þ EAR-motif (arabidopsis) bHLH: Delila (antirrhinum)

Tobacco Tobacco Lucerne Arabidopsis Tobacco Tomato White clover Tobacco Maize cell lines

212

Table 8.4. Examples of genetic modification of flavonoid production in stably transformed plants or cell lines by the introduction of sense transgenes encoding transcription factors that regulate flavonoid biosynthetic genes.

BHLH-containing Tag1 transposon: R (maize) bHLH: Myc-rp and Myc-gp (perilla) bHLHþMyb cointroduced: C1þDelila bHLHþMyb cointroduced: C1þR

bHLHþMyb cointroduced: C1þR and suppressed F3H bHLHþMyb cointroduced: LcþC1

Bird’s foot trefoil Tobacco Tomato Tobacco Arabidopsis Soybean Maize cell lines Soybean Arabidopsis Tomato

No visible change in phenotype Anthocyanins increased Anthocyanins increased Anthocyanins increased No increase in flavonoid content PAs increased in roots (and decreased in leaves of some lines) PAs and anthocyanins increased Anthocyanins increased (variegated patterns)

Bradley et al. (1999) Bradley et al. (1998) Lloyd et al. (1992) Goldsbrough et al. (1996) Bovy et al. (2002) Damiani et al. (1999)

Anthocyanins increased Anthocyanins increased No visible change in phenotype Ratio of genistein to daidzein reduced Anthocyanins, phlobaphenes and C-glycosylflavones increased Isoflavonoids increased

Gong et al. (1999) Gong et al. (1999) Mooney et al. (1995) Yu et al. (2003) Grotewold et al. (1998); Bruce et al. (2000) Yu et al. (2003)

Anthocyanins increased Flavonols increased in fruit; anthocyanins increased in leaves

Lloyd et al. (1992) Bovy et al. (2002)

Robbins et al. (2003) Liu et al. (2001)

Developmental Control and Biotechnology of Floral Pigmentation

bHLH: Sn (maize)

Pelargonium Petunia Tobacco Tomato Tomato Bird’s foot trefoil

a Only a general indication of the effect on flavonoid production is given (in some studies other phenylpropanoid types occurring were also affected), as results are variable and may include: small increases in flavonoid levels in tissues already producing flavonoids; production of anthocyanin earlier in flower development than normal; ectopic flavonoid production; increased flavonoid production only under stress conditions. Furthermore, a number of studies measured the effects on anthocyanin production only.

213

214

K. Davies and K. Schwinn

Although flavonoid biosynthesis has been enhanced in a range of species by introduction of transgenes encoding a single TF, there have been several cases reported where no change in phenotype has occurred or where the phenotype only becomes apparent under certain environmental conditions (Table 8.4). Indeed, the same TF transgenes may have markedly different effects in transgenic plants of different species. This variation in effect may be related to the availability of adequate precursors, the characteristics of the specific TF introduced, the presence or absence of interacting endogenous TFs or the characteristics of the endogenous TFs. Thus, for generation of the desired phenotypes in target species, appropriate TFs may need to be identified for each species and/or combinations of TF transgenes may be required. TF transgenes can also be used to repress flavonoid biosynthesis. This may be by using sequences for proteins that are thought to have a repressive impact on anthocyanin biosynthesis as part of their normal in planta function (Aharoni et al., 2001) or by modifying anthocyanin-related activator TFs through the addition of repression domains from other proteins, such as the ERF-associated amphiphilic repression (EAR) motif (Hiratsu et al., 2003; Matsui et al., 2004).

Carotenoids For species that have carotenoid pigmentation in flowers, overexpression or knockout transgenes for carotenoid biosynthetic enzymes might be used to generate different carotenoid profiles and, therefore, alter flower colour. Of greater interest would be the ability to introduce novel carotenoids based colours into species that lack significant carotenoid production in the flowers. However, a major hurdle may be the lack of floral plastids of sufficient number and type. The availability of carotenoid precursors in these tissues may also need to be addressed, although there has been some success in the metabolic engineering of carotenoid precursor supply (e.g. Botella-Pavia et al., 2004). Except for a preliminary report on increasing carotenoid levels in flowers of petunia (Davies et al., 2003b), there have been no reports to date of the GM of flower colour through the alteration of carotenoid biosynthesis – although the colour of nectary tissue has been modified in tobacco, in an attempt to produce commercially important ketocarotenoids in plants (Ralley et al., 2004). In addition to in planta ketocarotenoid production, the carotenoid biotechnology field is primarily focused on increasing levels of human health-related carotenoids in food or gaining an understanding of their in planta function as antioxidants (reviewed in Naik et al., 2003; Fraser and Bramley, 2004). One well-known success has been the development of ‘Golden Rice’, which has increased b-carotene levels in the endosperm due to introduction of daffodil PSY and Erwinia uredorva PDS transgenes (Ye et al., 2000). Table 8.5 lists published GM experiments for carotenoid biosynthesis in plants, and some of these approaches would be expected to generate novel phenotypes if applied to ornamentals that have carotenoid-based floral pigmentation. A complicating factor for GM experiments targeting carotenoid biosynthesis is the possibility of pleiotropic effects, and the potential associated need

Transgene

Species modified

Effect on carotenoid productiona

Reference

PSY

Arabidopsis

Increased carotenoid (including lutein and violaxanthin) and abscisic acid levels in seeds, reduced germination rate Carotenoid production increased, flower colour changed from white to pale yellow Increased levels of phytoene and other carotenoids, pleotropic effects Increased carotenoid production in fruit and other tissues, pleotropic effects 50-fold increase in carotenoid levels (principally carotenes) in the seed and an orange colour imparted No change in carotenoids, conferred herbicide resistance Twofold to fourfold increase in fruit carotenoid levels No change in phenotype Reduced levels of carotenoids with e,b-rings and increased levels of those with b, bb-rings Increased accumulation of b-carotene and its derivatives, overall decrease in carotenoid content Up to sevenfold increase in b-carotene levels in fruit, colour change from red to orange Increased lutein levels b-carotene and other carotenoid production in seed endosperm and a golden colour imparted Low levels of ketocarotenoids (including 4-ketolutein) produced in seeds

Lindgren et al. (2003)

Petunia Tobacco Tomato PSY (bacterial)

Oilseed rape

PDS PDS (bacterial)

Tobacco Tomato Tobacco Tobacco Tomato

bLYC

Tomato

eLYC PSY and PDS or PSY, PDS and bLCY b-carotene ketolase (algal)

Arabidopsis Rice (Oryza sativa) Arabidopsis

Davies et al. (2003b) Busch et al. (2002) Fray et al. (1995) Shewmaker et al. (1999) Wagner et al. (2002) Fraser et al. (2002) Busch et al. (2002) Misawa et al. (1994) Ro¨mer et al. (2000); Fraser et al. (2001) Rosati et al. (2000)

Developmental Control and Biotechnology of Floral Pigmentation

Table 8.5. Genetic modification of carotenoid biosynthesis in plants using carotenoid structural genes. Sense transgenes were used unless indicated otherwise. Experiments involving complementation of mutants are not included.

Pogson and Rissler (2000) Ye et al. (2000) Sta˚lberg et al. (2003) continued 215

Transgene

Species modified

Effect on carotenoid productiona

Reference

Tobacco

Astaxanthin produced in the flowers (the nectary), colour change from yellow to red Up to 13-fold increase in ketocarotenoid levels in seeds compared to b-carotene ketolase gene alone Ketocarotenoids produced in the leaves and nectary and changes in total carotenoid levels Ketocarotenoids produced in the leaves Change in carotenoid profile with increased xanthophyll levels Increased zeaxanthin synthesis under high light Increased zeaxanthin and b-cryptoxanthin levels Zeaxanthin produced Carotenoid levels reduced, pleiotropic effects, lethality in some lines Carotenoid levels reduced, gibberellin levels increased Phytoene levels increased, pleiotropic effects, lethality in some lines Loss of b-carotene, small increase in lycopene content of fruit Increase in b-carotene levels relative to xanthophylls Reduced zeaxanthin and antheraxanthin levels

Mann et al. (2000)

b-carotene ketolase and PSY

Arabidopsis

Carotene 4,4’-oxygenase and 3,3’-hydroxylase (bacterial)

Tobacco Tomato Arabidopsis

bOH (bacterial) bLYC and bOH ZEP antisense or co-suppression PSY antisense

Tobacco Tomato Potato Tobacco

PDS antisense

Tomato Tobacco

bLCY antisense

Tomato

bOH antisense VDE antisense

Arabidopsis Tobacco

a

Only a general indication of the phenotype is given.

Sta˚lberg et al. (2003) Ralley et al. (2004) Ralley et al. (2004) Davison et al. (2002) Go¨tz et al. (2002) Dharmapuri et al. (2002) Ro¨mer et al. (2002) Busch et al. (2002) Bird et al. (1991) Busch et al. (2002) Ronen et al. (2000); Rosati et al. (2000) Rissler and Pogson (2001) Chang et al. (2000); Verhoeven et al. (2001)

K. Davies and K. Schwinn

bOH (plant)

216

Table 8.5. continued . Genetic modification of carotenoid biosynthesis in plants using carotenoid structural genes. Sense transgenes were used unless indicated otherwise. Experiments involving complementation of mutants are not included.

Developmental Control and Biotechnology of Floral Pigmentation

217

for flower-specific transgene expression. Besides the obvious detrimental effects on photosynthesis that major disruption to carotenoid production could generate, carotenoids share precursors with, or are substrates for, the biosynthesis of important growth regulators. There may also be complications through metabolic channelling effects or concomitant regulatory changes. Expression of CaMV35S:PSY in tobacco increased carotenoid production, but also caused dwarfism, altered leaf morphology and reduced anthocyanin levels (Busch et al., 2002). Expression of CaMV35S:PSY in tomato also had multiple ramifications, including dwarfism, probably due to the diversion of GGPP substrate from gibberellin biosynthesis (Fray et al., 1995). In oilseed rape (Brassica napus), massive increases in a- and b-carotene levels were seen with seed-specific expression of a bacterial PSY gene, but tocopherol and chlorophyll levels were lowered and fatty acid composition altered (Shewmaker et al., 1999). Furthermore, a tomato line lacking activity of the fruit-related PSY-1 gene failed to make carotenoids even though the PSY-2 gene was active in the fruit, perhaps indicating the occurrence of metabolic channelling (Fraser et al., 1999). Betalains The betalain biosynthetic pathway is an attractive target for ornamental biotechnology approaches, particularly the introduction of this pathway into species that normally lack it. The pathway from general metabolites to the first coloured compounds involves only a few biosynthetic steps, and the introduction of betalain biosynthesis is unlikely to cause pleiotropic effects. If DOPA substrate is available, and DOD activity is introduced, then the steps to the yellow betaxanthins can occur spontaneously, even in species that normally do not make the compounds. Indeed, we have shown this is the case for antirrhinum flowers (N. Pathirana and K. Schwinn, unpublished results). However, DOPA will probably be present in limiting quantities in most species, preventing significant betalain formation. As yet no cDNA clones have been characterized for the putative tyrosinase involved in DOPA production. For synthesis of the red/purple betacyanins, other transgenes will also be required. Sequences for non-specific glucosyltransferases that can use flavonoid and betacyanin substrates are available, but sequences for other biosynthetic activities are lacking. The only published GM study for betalains is the transient expression of a fungal (A. muscaria) DOD cDNA in flowers of P. grandiflora genotypes lacking activity of the plant DOD (Mueller et al., 1997). Transformed petal cells developed pigmentation and were shown to have accumulated the plant betalains dopaxanthin and betanin, as well as miraxanthin V, which is normally specific to fungi.

Concluding Comments Flower–pollinator interactions have driven the evolution of complex biosynthetic pathways and elegant regulatory mechanisms to form pigments with specific colours and pigmentation patterns ranging from simple to highly

218

K. Davies and K. Schwinn

intricate. In some species leaves have been recruited to provide additional (or alternative) specialized coloured organs for pollinator attraction. Floral pigmentation is under strict developmental control and is commonly linked to fertility status. The developmental signalling involved can be complex, as more than one pigment pathway, and other pathways regulating other components that affect colour like vacuolar pH, may be coordinately controlled. The molecular basis of the biosynthesis of the common flower colour pigments, carotenoids and flavonoids, is now well understood. A significant advancement regarding the flavonoid pathway has been the identification and characterization of transcription factors that regulate flavonoid production. However, less understood is the control of flower colour patterning, or the linkage of pigmentation to floral development or environmental signals, although this is likely to change over the next few years as current studies progress. Based on the extensive molecular knowledge on flavonoid biosynthesis, flower and plant colour has been successfully modified in several species, with TF transgenes proving particularly effective. With the ongoing progress in the elucidation of the genes involved in carotenoid and betalain biosynthesis, along with a greater understanding of these pathways, it is expected that success in the GM of these pathways to produce desirable colours in flowers will soon be achieved.

Acknowledgements We thank our Crop & Food Research colleagues Murray Boase, Tony Corbett and Simon Deroles, and Dr Martha R. Weiss (Georgetown University, Washington, DC) for photographs included in Plate 1.

References Abrahams, S., Lee, E., Walker, A.R., Tanner, G.J., Larkin, P.J. and Ashton, A.R. (2003) The Arabidopsis TDS4 gene encodes leucoanthocyanidin dioxygenase (LDOX) and is essential for proanthocyanidin synthesis and vacuole development. Plant Cell 35, 624–636. Achnine, L., Blancaflor, E.B., Rasmussen, S. and Dixon, R.A. (2004) Colocalization of L-phenylalanine ammonia-lyase and cinnamate 4-hydroxylase for metabolic channeling in phenylpropanoid biosynthesis. Plant Cell 16, 3098– 3109.

Aharoni, A., De Vos, C.H., Wein, M., Sun, Z., Greco, R., Kroon, A., Mol, J.N. and O’Connell, A.P. (2001) The strawberry FaMYB1 transcription factor suppresses anthocyanin and flavonol accumulation in transgenic tobacco. Plant Journal 28, 319–332. Aida, R., Kishimoto, S., Tanaka, Y. and Shibata, M. (2000a) Modification of flower colour in torenia (Torenia fournieri Lind.) by genetic transformation. Plant Science 153, 33–42. Aida, R., Yoshida, K., Kondo, T., Kishimoto, S. and Shibata, M. (2000b) Copigmentation gives blue flowers on

Developmental Control and Biotechnology of Floral Pigmentation transgenic torenia plants with the antisense dihydroflavonol-4-reductase gene. Plant Science 160, 49–56. Akashi, T., Fukuchi-Mizutani, M., Aoki, T., Ueyama, Y., Yonekura-Sakakibara, K., Tanaka, Y., Kusumi, T. and Ayabe, S.-I. (1999) Molecular cloning and biochemical characterization of a novel cytochrome P450, flavone synthase II, that catalyzes direct conversion of flavanones to flavones. Plant and Cell Physiology 40, 1182–1186. Al-Babili, S., Hugueney, P., Schledz, M., Welsch, R., Frohnmeyer, H., Laule, O. and Beyer, P. (2000) Identification of a novel gene coding for neoxanthin synthase from Solanum tuberosum. FEBS Letters 485, 168–172. Albrecht, M., Klein, A., Hugueney, P., Sandmann, G. and Kuntz, M. (1995) Molecular cloning and functional expression in E. coli of a novel plant enzyme mediating zeta-carotene desaturation. FEBS Letters 372, 199– 202. Alfenito, M.R., Souer, E., Goodman, C.D., Buell, R., Mol, J., Koes, R. and Walbot, V. (1998) Functional complementation of anthocyanin sequestration in the vacuole by widely divergent glutathione S-transferases. Plant Cell 10, 1135–1149. Angenent, G.C., Franken, J., Busscher, M., Weiss, D. and van Tunen, A.J. (1994) Co-suppression of the petunia homeotic gene fbp2 affects the identity of the generative meristem. Plant Journal 5, 33–44. Bartley, G.E., Viitanen, P.V., Pecker, I., Chamovitz, D., Hirschberg, J. and Scolnik, P.A. (1991) Molecular cloning and expression in photosynthetic bacteria of a soybean cDNA coding for phytoene desaturase, an enzyme of the carotenoid biosynthesis pathway. Proceedings of the National Academy of Sciences USA 88, 6532– 6536. Bartley, G.E., Viitanen, P.V., Bacot, K.O. and Scolnik, P.A. (1992) A tomato gene expressed during fruit ripening

219

encodes an enzyme of the carotenoid biosynthesis pathway. Journal of Biological Chemistry 267, 5036–5039. Baudry, A., Heim, M.A., Dubreucq, B., Caboche, M., Weisshaar, B. and Lepiniec, L. (2004) TT2, TT8, and TTG1 synergistically specify the expression of BANYULS and proanthocyanidin biosynthesis in Arabidopsis thaliana. Plant Journal 39, 366–380. Baxter, I.R., Young, J.C., Armstrong, G., Foster, N., Bogenschutz, N., Cordova, T., Peer, W.A., Hazen, S.P., Murphy, A.S. and Harper, J.F. (2005) A plasma membrane Hþ-ATPase is required for the formation of proanthocyanidins in the seed coat endothelium of Arabidopsis thaliana. Proceedings of the National Academy of Sciences USA 102, 2649–2654. Bey, M., Stu¨ber, K., Fellenberg, K., Schwarz-Sommer, Z., Sommer, H., Saedler, H. and Zachgo, S. (2004) Characterization of antirrhinum petal development and identification of target genes of the class B MADS box gene DEFICIENS. Plant Cell 16, 3197–3215. Bird, C.R., Ray, J.A., Fletcher, J.D., Boniwell, J.M., Bird, S., Hughes, S., Morris, P.C. and Grierson, D. (1991) Using antisense RNA to study gene function: inhibition of carotenoid biosynthesis in transgenic tomatoes. Biotechnology 9, 635–639. Blanc, V.M. and Pichersky, E. (1995) Nucleotide sequence of a Clarkia breweri cDNA clone of Ipi1, a gene encoding isopentenyl pyrophosphate isomerase. Plant Physiology 108, 855–856. Bohm, B.A. (1998) Introduction to Flavonoids. Harwood Academic Publishers, Amsterdam. Borevitz, J.O., Xia, Y., Blount, J., Dixon, R.A. and Lamb, C. (2000) Activation tagging identifies a conserved MYB regulator of phenylpropanoid biosynthesis. Plant Cell 12, 2383–2394. ´ ., Phillips, Botella-Pavia, P., Besumbes, O M.A., Carretero-Paulet, L., Boronat, A. and Rodriiguez-Concepcio´n, M. (2004)

220

K. Davies and K. Schwinn Regulation of carotenoid biosynthesis in plants: evidence for a key role of hydroxymethylbutenyl diphosphate reductase in controlling the supply of plastidial isoprenoid precursors. Plant Journal 40, 188–199. Bouvier, F., Hugueney, P., d’Harlingue, A., Kuntz, M. and Camara, B. (1994) Xanthophyll biosynthesis in chromoplasts: isolation and molecular cloning of an enzyme catalyzing the conversion of 5,6-epoxycarotenoid into ketocarotenoid. Plant Journal 6, 45–54. Bouvier, F., d’Harlingue, A., Hugueney, P., Marin, E., Marion-Poll, A. and Camara, B. (1996) Xanthophyll biosynthesis: cloning, expression, functional reconstitution, and regulation of b-cyclohexenyl carotenoid epoxidase from pepper (Capsicum annuum). Journal of Biological Chemistry 271, 28861–28867. Bouvier, F., d’Harlingue, A., Suire, C., Backhaus, R.A. and Camara, B. (1998) Dedicated roles of plastid transketolases during early onset of isoprenoid biogenesis in pepper fruits. Plant Physiology 117, 1423–1431. Bouvier, F., D’Harlingue, A., Backhaus, R.A., Kumagai, M.H. and Camara, B. (2000) Identification of neoxanthin synthase as a carotenoid cyclase paralog. European Journal of Biochemistry 267, 6346–6352. Bouvier, F., Suire, C., Mutterer, J. and Camara, B. (2003a) Oxidative remodeling of chromoplast carotenoids: identification of the carotenoid dioxygenase CsCCD and CsZCD genes involved in crocus secondary metabolite biogenesis. Plant Cell 15, 47–62. Bouvier, F., Dogbo, O. and Camara, B. (2003b) Biosynthesis of the food and cosmetic plant pigment bixin (Annatto). Science 300, 2089–2092. Bovy, A., de Vos, R., Kemper, M., Schijlen, E., Pertejo, M.A., Muir, S., Collins, G., Robinson, S., Verhoeyen, M., Hughes, S., Santos-Buelga, C. and van Tunen, A. (2002) High-flavonol

tomatoes resulting from the heterologous expression of the maize transcription factor genes LC and C1. Plant Cell 14, 2509–2526. Bradley, J.M., Davies, K.M., Deroles, S.C., Bloor, S.J. and Lewis, D.H. (1998) The maize Lc regulatory gene up-regulates the flavonoid biosynthetic pathway of Petunia. Plant Journal 13, 381–392. Bradley, J.M., Deroles, S.C., Boase, M.R., Bloor, S., Swinny, E. and Davies, K.M. (1999) Variation in the ability of the maize Lc regulatory gene to upregulate flavonoid biosynthesis in heterologous systems. Plant Science 140, 31–39. Bramley, P.M. (2002) Regulation of carotenoid formation during tomato fruit ripening and development. Journal of Experimental Botany 53, 2107– 2113. Brouillard, R. and Dangles, O. (1993) Flavonoids and flower colour. In: Harborne, J.B. (ed.) The Flavonoids: Advances in Research Since 1986. Chapman & Hall, London, pp. 565– 587. Bruce, W., Folkerts, O., Garnaat, C., Crasta, O., Roth, B. and Bowen, B. (2000) Expression profiling of the maize flavonoid pathway genes controlled by estradiol-inducible transcription factors CRC and P. Plant Cell 12, 65–80. Brugliera, F., Holton, T.A., Stevenson, T.W., Farcy, E., Lu, C.Y. and Cornish, E.C (1994) Isolation and characterization of a cDNA clone corresponding to the Rt locus of Petunia hybrida. Plant Journal 5, 81–92. Brugliera, F., Barri-Rewell, G., Holton, T.A. and Mason, J.G. (1999) Isolation and characterization of a flavonoid 3’hydroxylase cDNA clone corresponding to the Ht1 locus of Petunia hybrida. Plant Journal 19, 441–451. Bugos, R.C. and Yamamoto, H.Y. (1996) Molecular cloning of violaxanthin deepoxidase from romaine lettuce and expression in Esherichia coli. Proceed-

Developmental Control and Biotechnology of Floral Pigmentation ings of the National Academy of Sciences USA 93, 6320–6325. Burr, F.A., Burr, B., Scheffler, B.E., Blewitt, M., Wienand, U. and Matz, E.C. (1996) The maize repressor-like gene intensifier1 shares homology with the R1/B1 multigene family of transcription factors and exhibits missplicing. Plant Cell 8, 1249–1259. Busch, M., Seuter, A. and Hain, R. (2002) Functional analysis of the early steps of carotenoid biosynthesis in tobacco. Plant Physiology 128, 439–453. Carey, C.C., Strahle, J.T., Selinger, D.A. and Chandler, V.L. (2004) Mutations in the pale aleurone color1 regulatory gene of the Zea mays anthocyanin pathway have distinct phenotypes relative to the functionally similar TRANSPARENT TESTA GLABRA1 gene in Arabidopsis thaliana. Plant Cell 16, 450–464. Carol, P., Stevenson, D., Bisanz, C., Breitenbach, J., Sandmann, G., Mache, R., Coupland, G. and Kuntz, M. (1999) Mutations in the Arabidopsis gene IMMUTANS cause a variegated phenotype by inactivating a chloroplast terminal oxidase associated with phytoene desaturation. Plant Cell 11, 57–68. Chang, S.-H., Bugos, R.C., Sun, W.H. and Yamamoto, H.Y. (2000) Antisense suppression of violaxanthin deepoxidase in tobacco does not affect plant performance in controlled growth conditions. Photosynthesis Research 64, 95–103. Chen, B., Wang, X., Hu, Y., Wang, Y. and Lin, Z. (2004) Ectopic expression of a c1-I allele from maize inhibits pigment formation in the flower of transgenic tobacco. Molecular Biotechnology 26, 187–192. Christinet, L., Burdet, F., Zaiko, M., Hinz, U. and Zry¨d, J.P. (2004) Characterization and functional identification of a novel plant 4,5-extradiol dioxygenase involved in betalain pigment biosynthesis in Portulaca grandiflora. Plant Physiology 134, 265–274.

221

Collette, V.E., Jameson, P.E., Schwinn, K.E., Umaharan, P. and Davies, K.M. (2004) Temporal and spatial expression of flavonoid biosynthetic genes in flowers of Anthurium andraeanum. Physiologia Plantarum 122, 297– 304. Cookson, P.J., Kiano, J.W., Shipton, C.A., Fraser, P.D., Romer, S., Schuch, W., Bramley, P.M. and Pyke, K.A. (2003) Increases in cell elongation, plastid compartment size and phytoene synthase activity underlie the phenotype of the high pigment-1 mutant of tomato. Planta 217, 896–903. Corona, V., Aracri, B., Kosturkova, G., Bartley, G.E., Pitto, L., Giorgetti, L., Scolnik, P.A. and Giuliano, G. (1996) Regulation of a carotenoid biosynthesis gene promoter during plant development. Plant Journal 9, 505–512. Courtney-Gutterson, N., Napoli, C., Lemieuz, C., Morgan, A., Firoozabady, E. and Robinson, K.E.P. (1994) Modification of flower color in florist’s chrysanthemum: production of a whiteflowering variety through moleculargenetics. Biotechnology 12, 268– 271. Cunningham, F.X. and Gantt, E. (1998) Genes and enzymes of carotenoid biosynthesis in plants. Annual Review of Plant Physiology and Plant Molecular Biology 49, 557–583. Cunningham, F.X. and Gantt, E. (2001) One ring or two? Determination of ring number in carotenoids by lycopene ecyclases. Proceedings of the National Academy of Sciences USA 98, 2905– 2910. Cunningham, F.X. and Gantt, E. (2005) A study in scarlet: enzymes of ketocarotenoid biosynthesis in flowers of Adonis aestivalis. Plant Journal 41, 478– 492. Cunningham, F.X., Pogson, B., Sun, Z.R., McDonald, K.A., DellaPenna, D. and Gantt, E. (1996) Functional analysis of the b and e lycopene cyclase enzymes of Arabidopsis reveals a mechanism for control of cyclic

222

K. Davies and K. Schwinn carotenoid formation. Plant Cell 8, 613–626. Cuttriss, A. and Pogson, B. (2004) Carotenoids. In: Davies, K. (ed.) Plant Pigments and their Manipulation. Blackwell, Oxford, UK, pp. 57–91. Damiani, F., Paolocci, F., Cluster, P.D., Arcioni, S., Tanner G.J., Joseph R.J., Li, Y.G., de Majnik J. and Larkin P.J. (1999) The maize transcription factor Sn alters proanthocyanidin synthesis in transgenic Lotus corniculatus plants. Australian Journal of Plant Physiology 26, 159–169. Davies, K.M. (2004) Important rare plant pigments. In: Davies, K.M. (ed.) Plant Pigments and their Manipulation. Annual Plant Reviews, Blackwell, Oxford, UK, pp. 214–247. Davies, K.M. and Schwinn, K.E. (2003) Transcriptional regulation of secondary metabolism. Functional Plant Biology 30, 913–925. Davies, K.M., Bloor, S.J., Spiller, G.B. and Deroles, S.C. (1998) Production of yellow colour in flowers: redirection of flavonoid biosynthesis in Petunia. Plant Journal 13, 259–266. Davies, K.M., Spiller, G.B., Bradley, J.M., Winefield, C.S., Schwinn, K.E., Martin, C.R. and Bloor, S.J., (2001) Genetic engineering of yellow flower colours. Acta Horticulturae 560, 39– 44. Davies, K.M., Schwinn, K.E., Deroles, S.C., Manson, D.G., Lewis, D.H., Bloor, S.J. and Bradley, J.M. (2003a) Enhancing anthocyanin production by altering competition for substrate between flavonol synthase and dihydroflavonol 4-reductase. Euphytica 131, 259–268. Davies, K.M., Marshall, G.B., Lewis, D.H., Winefield, C.S., Deroles, S.C., Boase, M.R., Zhang, H., Nielsen, K.M., Schwinn, K.E., Bloor, S.J., Swinny, E. and Martin, C.R. (2003b). Generation of new ornamental varieties through genetic modification of pigment biosynthesis. Acta Horticulturae 624, 435–447.

Davison, P.A., Hunter, C.N. and Horton, P. (2002) Overexpression of beta-carotene hydroxylase enhances stress tolerance in Arabidopsis. Nature 418, 203–206. De Jaeger, G., Buys, E., Eeckhout, D., De Wilde, C., Jacobs, A., Kapila, J., Angenon, G., Van Montagu, M., Gerats, T. and Depicker, A. (1999) High level accumulation of single-chain variable fragments in the cytosol of transgenic Petunia hybrida. European Journal of Biochemistry 259, 426–434. de Majnik, J., Weinman, J.J., Djordjevic, M.A., Rolfe, B.G., Tanner, G.J., Joseph, R.G. and Larkin, P.J. (2000) Anthocyanin regulatory gene expression in transgenic white clover can result in an altered pattern of pigmentation. Australian Journal of Plant Physiology 27, 659–667. Deroles, S.C., Bradley, J.M., Schwinn, K.E., Markham, K.R., Bloor, S.J., Manson, D.G. and Davies, K.M. (1998) An antisense chalcone synthase gene leads to novel flower patterns in lisianthus (Eustoma grandiflorum). Molecular Breeding 4, 59–66. Deroles, S.C., Boase, M.R., Lee, C.E. and Peters, T.A. (2002) Gene transfer to plants. In: Vainstein, A. (ed.) Breeding for Ornamentals: Classical and Molecular Approaches. Kluwer, Dordrecht, The Netherlands, pp. 155–196. de Vetten, N., Quattrocchio, F., Mol, J. and Koes, R. (1997) The an11 locus controlling flower pigmentation in petunia encodes a novel WD-repeat protein conserved in yeast, plants, and animals. Genes & Development 11, 1422–1434. de Vetten, N., ter Horst, J., van Schaik, H.-P., de Boer, A., Mol, J. and Koes, R. (1999) A cytochrome b5 is required for full activity of flavonoid 3’,5’-hydroxylase, a cytochrome P450 involved in the formation of blue flower colours. Proceedings of the National Academy of Sciences USA 96, 778–783. Devic, M., Guilleminot, J., Debeaujon, I., Bechtold, N., Bensaude, E., Koorn-

Developmental Control and Biotechnology of Floral Pigmentation neef, M., Pelletier, G. and Delseny, M. (1999) The BANYULS gene encodes a DFR-like protein and is a marker of early seed coat development. Plant Journal 19, 387–398. Dharmapuri, S., Rosati, C., Pallara, P., Aquilani, R., Bouvier, F., Camara, B. and Giuliano, G. (2002) Metabolic engineering of xanthophyll content in tomato fruits. FEBS Letters 519, 30–34. Dixon, R.A., Xie, D.-Y. and Sharma, S.B. (2005) Proanthocyanidins: a final frontier in flavonoid research? New Phytologist 165, 9–28. Elomaa, P., Honkanen, J., Puska, R., Seppanen, P., Helariutta, Y., Mehto, M., Kotilainen, M., Nevalainen, L. and Teeri, T.H. (1993) Agrobacteriummediated transfer of antisense chalcone synthase cDNA to Gerbera hybrida inhibits flower pigmentation. Biotechnology 11, 508–511. Elomaa, P., Uimari, A., Mehto, M., Albert, V.A., Laitinen, R.A.E. and Teeri, T.H. (2003) Activation of anthocyanin biosynthesis in Gerbera hybrida (Asteraceae) suggests conserved protein-protein and protein-promoter interactions between the anciently diverged monocots and eudicots. Plant Physiology 133, 1831–1842. Eulgem, T., Rushton, P.J., Robatzek, S. and Somssich, I.E. (2000) The WRKY superfamily of plant transcription factors. Trends in Plant Sciences 5, 199–206. Fahrendorf, T. and Dixon, R.A. (1993) Stress responses in lucerne (Medicago sativa L.) XVIII: molecular cloning and expression of the elicitor-inducible cinnamic acid 4-hydroxylase cytochrome P450. Archives of Biochemistry and Biophysics 305, 509–515. Farzad, M., Griesbach, R. and Weiss, M.R. (2002) Floral color change in Viola cornuta L. (Violaceae): a model system to study regulation of anthocyanin production. Plant Science 162 225–231.

223

Farzad, M., Griesbach, R., Hammond, J., Weiss, M.R. and Elmendorf, H.G. (2003) Differential expression of three key anthocyanin biosynthetic genes in a color-changing flower, Viola cornuta cv. Yesterday, Today and Tomorrow. Plant Science 165, 1333–1342. Fedoroff, N., Furtek, D. and Nelson, O.E. (1984) Cloning of the bronze locus in maize by a simple and generalizable procedure using the transposable element Activator (Ac). Proceedings of the National Academy of Sciences USA 81, 3825–3829. Feldbru¨gge, M., Sprenger, M., Hahlbrock, K. and Weisshaar, B. (1997) PcMYB1, a novel plant protein containing a DNA-binding domain with one MYB repeat, interacts in vivo with a light regulatory promoter unit. Plant Journal 11, 1079–1093. Ferrario, S., Immink, R.G., Shchennikova, A., Busscher-Lange, J. and Angenent, G.C. (2003) The MADS box gene FBP2 is required for SEPALLATA function in petunia. Plant Cell 15, 914–925. Firoozabady, E., Moy, Y., Courtney-Gutterson, N. and Robinson, K. (1994) Regeneration of transgenic rose (Rosa hybrida) plants from embryogenic tissue. Biotechnology 12, 609–613. Fischer, R., Budde, I. and Hain, R. (1997) Stilbene synthase gene expression causes changes in flower colour and male sterility in tobacco. Plant Journal 11, 489–498. Ford, C.M., Boss, P.K. and Høj, P.B. (1998) Cloning and characterization of Vitis vinifera UDP-glucose:flavonoid 3-O-glucosyltransferase, a homologue of the enzyme encoded by the maize Bronze-1 locus that may primarily serve to glucosylate anthocyanidins in vivo. Journal of Biological Chemistry 273, 9224–9233. Forkmann, G. and Heller, W. (1999) Biosynthesis of flavonoids. In: Sankawa, U. (ed.) Comprehensive Natural Products Chemistry, Vol. 1: Polyketides

224

K. Davies and K. Schwinn and Other Secondary Metabolites Including Fatty Acids and their Derivatives. Elsevier, Oxford, UK, pp. 713–748. Fraser, P.D. and Bramley, P.M. (2004) The biosynthesis and nutritional uses of carotenoids. Progress in Lipid Research 43, 228–265. Fraser, P.D., Kiano, J.W., Truesdale, M.R., Schuch, W. and Bramley, P.M. (1999) Phytoene synthase-2 enzyme activity in tomato does not contribute to carotenoid synthesis in ripening fruit. Plant Molecular Biology 40, 687–698. ¨ mer, S., Kiano, J.W., Fraser, P.D., Ro Shipton, C.A., Mills, P.B., Drake, R., Schuch, W. and Bramley, P.M. (2001) Elevation of carotenoids in tomato by genetic manipulation. Journal of the Science of Food and Agriculture 81, 822–827. ¨ mer, S., Shipton, C.A., Fraser, P.D., Ro Mills, P.B., Kiano, J.W., Misawa, N., Drake, R.G., Schuch, W. and Bramley, P.M. (2002) Evaluation of transgenic tomato plants expressing an additional phytoene synthase in a fruit-specific manner. Proceedings of the National Academy of Sciences USA 99, 1092–1097. Fray, R.G., Wallace, A., Fraser, P.D., Valero, D., Hedden, P., Bramley, P.M. and Grierson, D. (1995) Constitutive expression of a fruit phytoene synthase gene in transgenic tomatoes causes dwarfism by redirecting metabolites from the gibberellin pathway. Plant Journal 8, 693–701. Fujiwara, H., Tanaka, Y., Yonekura-Sakakibara, K., Fukuchi-Mizutani, M., Nakao, M., Fukui, Y., Yamaguchi, M., Ashikari, T. and Kusumi, T. (1998) cDNA cloning, gene expression and subcellular localization of anthocyanin 5-aromatic acyltransferase from Gentiana triflora. Plant Journal 16, 421– 431. Fukada-Tanaka, S., Inagaki, Y., Yamaguchi, T., Saito, N. and Iida, S. (2000) Colour-enhancing protein in blue petals. Nature 407, 581.

Fukuchi-Mizutani, M., Okuhara, H., Fukui, Y., Nakao, M., Katsumoto, Y., Yonekura-Sakakibara, K., Kusumi, T., Hase, T. and Tanaka, Y. (2003) Biochemical and molecular characterization of a novel UDP-glucose:anthocyanin 3’-O-glucosyltransferase, a key enzyme for blue anthocyanin biosynthesis, from gentian. Plant Physiology 132, 1652–1663. Fukui, Y., Tanaka, Y., Kusumi, T., Iwashita, T. and Nomoto, K. (2003) A rationale for the shift in colour towards blue in transgenic carnation flowers expressing the flavonoid 3’,5’-hydroxylase gene. Phytochemistry 63, 15–23. Fukusaki, E., Kawasaki, K., Kajiyama, S., An, C.I., Suzuki, K., Tanaka, Y. and Kobayashi, A. (2004) Flower color modulations of Torenia hybrida by downregulation of chalcone synthase genes with RNA interference. Journal of Biotechnology 111, 229–240. Gill, G. and Ptashne, M. (1988) Negative effect of the transcriptional activator GAL4. Nature 334, 721–724. Goff, S.A., Cone, K.C. and Chandler, V.L. (1992) Functional analysis of the transcriptional activator encoded by the maize B gene: evidence for a direct functional interaction between two classes of regulatory proteins. Genes & Development 6, 864–875. Goldsbrough, A., Tong, Y. and Yoder, J.I. (1996) Lc as a non-destructive visual reporter and transposition excision marker gene for tomato. Plant Journal 9, 927–933. Gong, Z.Z., Yamagishi, E., Yamazaki, M. and Saito, K. (1999) A constitutively expressed Myc-like gene involved in anthocyanin biosynthesis from Perilla frutescens: molecular characterization, heterologous expression in transgenic plants and transactivation in yeast cells. Plant Molecular Biology 41, 33–44. Goodman, C.D., Casati, P. and Walbot, V. (2004) A multidrug resistance-associated protein involved in anthocyanin transport in Zea mays. Plant Cell 16, 1812–1826.

Developmental Control and Biotechnology of Floral Pigmentation Goodwin, T.W. (1976) Distribution of carotenoids. In: Goodwin, T.W. (ed.) Chemistry and Biochemistry of Plant Pigments, Vol. 1. Academic Press, London, pp. 225–261. Goto, T. and Kondo, T. (1991) Structure and molecular stacking of anthocyanins – flower color variation. Angewandte Chemie – International Edition in English 30, 17–33. ¨ tz, T., Sandmann, G. and Romer, S. Go (2002) Expression of a bacterial carotene hydroxylase gene (crtZ) enhances UV tolerance in tobacco. Plant Molecular Biology 50, 129–142. Grotewold, E., Chamberlin, M., Snook, M., Siame, B., Butler, L., Swenson, J., Maddock, S., St Clair, G. and Bowen, B. (1998) Engineering secondary metabolism in maize cells by ectopic expression of transcription factors. Plant Cell 10, 721–740. Gutterson, N. (1995) Anthocyanin biosynthetic genes and their application to flower colour modification through sense suppression. Horticultural Science 30, 964–966. Hardtke, C.S. and Deng, X.-W. (2000) The cell biology of the COP/DET/ FUS proteins. Regulating proteolysis in photomorphogenesis and beyond? Plant Physiology 124, 1548–1557. Harmer, S.L., Hogenesch, J.B., Straume, M., Chang, H.-S., Han, B., Zhu, T., Wang, X., Kreps, J.A. and Kay, S.A. (2000) Orchestrated transcription of key pathways in Arabidopsis by the circadian clock. Science 290, 2110– 2113. Hattori, T., Vasil, V., Rosenkrans, L., Hannah, L.C., McCarty, D.R. and Vasil, I.K. (1992) The Viviparous–1 gene and abscisic acid activate the C1 regulatory gene for anthocyanin biosynthesis during seed maturation in maize. Genes & Development 6, 609–618. Hemm, M.R., Rider, S.D., Ogas, J., Murry, D.J. and Chapple, C. (2004) Light induces phenylpropanoid metab-

225

olism in Arabidopsis roots. Plant Journal 38, 765–778. Heim, M.A., Jakoby, M., Werber, M., Martin, C., Weisshaar, B. and Bailey, P.C. (2003) The basic helix-loop-helix transcription factor family in plants: a genome-wide study of protein structure and functional diversity. Molecular Biology and Evolution 20, 735–747. Helariutta, Y., Elomaa, P., Kotilainen, M., Seppanen, P. and Teeri, T.H. (1993) Cloning of a cDNA coding for dihydroflavonol-4-reductase (DFR) and characterization of dfr expression in corrollas of Gerbera hybrida var. Regina (Compositae). Plant Molecular Biology 22, 183–193. Hernandez, J.M., Heine, G.F., Irani, N.G., Feller, A., Kim, M.-G., Matulnik, T., Chandler, V.L. and Grotewold, E. (2004) Different mechanisms participate in the R-dependent activity of the R2R3 MYB transcription factor C1. Journal of Biological Chemistry 279, 48205–48213. Hiratsu, K., Matsui, K., Koyama, T. and Ohme-Takagi, M. (2003) Dominant repression of target genes by chimeric repressors that include the EAR motif, a repression domain, in Arabidopsis. Plant Journal 34, 733–739. Holton, T.A., Brugliera, F., Lester, D.R., Tanaka, Y., Hyland, C.D., Menting, J.G.T., Lu, C.-Y., Farcy, E., Stevenson, T.W. and Cornish, E.C. (1993a) Cloning and expression of cytochrome P450 genes controlling flower colour. Nature 366, 276–279. Holton, T.A., Cornish, E.C. and Tanaka, Y. (1993b) Cloning and expression of flavonol synthase from Petunia hybrida. Plant Journal 4, 1003– 1010. Hugueney, P., Badillo, A., Chen, H-.C., Klein, A., Hirschberg, J., Camara, B. and Kuntz, M. (1995) Metabolism of cyclic carotenoids: a model for the alteration of this biosynthetic pathway in Capsicum annuum chromoplasts. Plant Journal 8, 417–424.

226

K. Davies and K. Schwinn Ibrahim, R.K. and Muzac, I. (2000) The methyltransferase gene superfamily: a tree with multiple branches. In: Romeo, J.T., Ibrahim, R., Varin, L. and De Luca, V. (eds) Evolution of Metabolic Pathways. Elsevier, Oxford, UK, pp. 349–384. Isaacson, T., Ronen, G., Zamir, D. and Hirschberg, J. (2002) Cloning of tangerine from tomato reveals a carotenoid isomerase essential for the production of beta-carotene and xanthophylls in plants. Plant Cell 14, 333–342. Isaacson, T., Ohad, I., Beyer, P. and Hirschberg, J. (2004) Analysis in vitro of the enzyme CRTISO establishes a poly-cis-carotenoid biosynthesis pathway in plants. Plant Physiology 136, 4246–4255. ¨ ge-Laser, Jakoby, M., Weisshaar, B., Dro W., Vicente-Carbajosa, J., Tiedemann, J., Kroj, T. and Parcy, F. (2002) bZIP transcription factors in Arabidopsis. Trends in Plant Science 7, 106–111. Jenkins, G.I., Long, J.C., Wade, H.K., Shenton, M.R. and Bibikova, T.N. (2001) UV and blue light signalling: pathways regulating chalcone synthase gene expression in Arabidopsis. New Phytologist 151, 121–131. Jin, H., Cominelli, E., Bailey, P., Parr, A., Mehrtens, F., Jones, J., Tonelli, C., Weisshaar, B. and Martin, C. (2000) Transcriptional repression by AtMYB4 controls production of UVprotecting sunscreens in Arabidopsis. EMBO Journal 19, 6150–6161. Johnson, E.T., Yi, H., Shin, B., Oh, B.-J., Cheong, H. and Choi, G. (1999) Cymbidium hybrida dihydroflavonol 4reductase does not efficiently reduce dihydrokaempferol to produce orange pelargonidin-type anthocyanins. Plant Journal 19, 81–85. Johnson, E.T., Ryu, S., Yi, H., Shin, B., Cheong, H. and Choi, G. (2001) Alteration of a single amino acid changes the substrate specificity of dihydroflavonol 4-reductase. Plant Journal 25, 325–333.

Jorgensen, R. (1995) Cosuppression, flower color patterns, and metastable gene expression states. Science 268, 686–691. Jorgensen, R.A., Que, Q.D. and Napoli, C.A. (2002) Maternally-controlled ovule abortion results from cosuppression of dihydroflavonol-4-reductase or flavonoid-3’,5’-hydroxylase genes in Petunia hybrida. Functional Plant Biology 29, 1501–1506. Josse, E.M., Simkin, A.J., Gaffe´, J., Laboure´, A.M., Kuntz, M. and Carol, P. (2000) A plastid terminal oxidase associated with carotenoid desaturation during chromoplast differentiation. Plant Physiology 123, 1427–1436. Joung, J.Y., Kasthuri, G.M., Park, J.Y., Kang, W.J. Kim, H.S., Yoon, B.S., Joung, H. and Jeon, J.H. (2003) An overexpression of chalcone reductase of Pueraria montana var. Lobata alters biosynthesis of anthocyanin and 5’-deoxyflavonoids in transgenic tobacco. Biochemistry and Biophysics Research Communications 303, 326–331. Kreuzaler, F., Ragg, H., Fautz, E., Kuhn, D.N. and Hahlbrock, K. (1983) UVinduction of chalcone synthase mRNA in cell suspension cultures of Petroselinum hortense. Proceedings of the National Academy of Sciences USA 80, 2591–2593. Kroon, J., Souer, E., de Graaff, A., Xue, Y., Mol, J. and Koes, R. (1994) Cloning and structural analysis of the anthocyanin pigmentation locus Rt of Petunia hybrida: characterization of insertion sequences in two mutant alleles. Plant Journal 5, 69–80. Kuhn, D.N., Chappell, J., Boudet, A. and Hahlbrock, K. (1984) Induction of phenylalanine ammonia-lyase and 4-coumarate: CoA ligase mRNAs in cultured plant cells by UV light or fungal elicitor. Proceedings of the National Academy of Sciences USA 81, 1102–1106. Kuntz, M., Romer, S., Suire, C., Hugueney, P., Weil, J.H., Schantz, R. and

Developmental Control and Biotechnology of Floral Pigmentation Camara, B. (1992) Identification of a cDNA for the plastid-located geranylgeranyl pyrophosphate synthase from Capsicum annuum: correlative increase in enzyme activity and transcript level during fruit ripening. Plant Journal 2, 25–34. Lange, B.M., Wildung, M.R., McCaskill, D. and Croteau, R. (1998) A family of transketolases that directs isoprenoid biosynthesis via a mevalonate-independent pathway. Proceedings of the National Academy of Sciences USA 95, 2100–2104. Lange, B.M. and Croteau, R. (1999) Isoprenoid biosynthesis via a mevalonateindependent pathway in plants: cloning and heterologous expression of 1deoxy-D-xylulose-5-phosphate reductoisomerase from peppermint. Archives of Biochemistry and Biophysics 365, 170–174. Lindgren, L., Sta˚hlberg, K.G. and ¨ glund, A.-S. (2003) Seed-specific Ho overexpression of an endogenous Arabidopsis phytoene synthase gene results in delayed germination and increased levels of carotenoids, chlorophyll, and abscisic acid. Plant Physiology 132, 779–785. Lindsay, W.P., McAlister, F.M., Zhu, Q., ¨ ge-Laser, W., Hedrick, He, X.Z., Dro S., Doerner, P., Lamb, C. and Dixon, R.A. (2002) KAP-2, a protein that binds to the H-box in a bean chalcone synthase promoter, is a novel plant transcription factor with sequence identity to the large subunit of human Ku autoantigen. Plant Molecular Biology 49, 503–514. Liu, D., Galli, M. and Crawford, N.M. (2001) Engineering variegated floral patterns in tobacco plants using the Arabidopsis transposable element Tag1. Plant and Cell Physiology 42, 419–423. Lloyd, A.M., Walbot, V. and Davis, R.W (1992) Arabidopsis and Nicotiana anthocyanin production activated by maize regulators R and C1. Science 258, 1773–1775.

227

Lunau, K. (2000) The ecology and evolution of visual pollen signals. Plant Systematics and Evolution 222, 89– 111. Mann, V., Harker, M., Pecker, I. and Hirschberg, J. (2000) Metabolic engineering of astaxanthin production in tobacco flowers. Nature Biotechnology 18, 888–892. Marin, E., Nussaume, L., Quesada, A., Gonneau, M., Sotta, B., Hugueney, P., Frey, A. and Marion-Poll, A. (1996) Molecular identification of zeaxanthin epoxidase of Nicotiana plumbaginifolia, a gene involved in abscisic acid biosynthesis and corresponding to the ABA locus of Arabidopsis thaliana. EMBO Journal 15, 2331–2342. Marles, M.A., Ray, H. and Gruber, M.Y. (2003) New perspectives on proanthocyanidin biochemistry and molecular regulation. Phytochemistry 64, 367– 383. Marrs, K.A., Alfenito, M.R., Lloyd, A.M. and Walbot, V. (1995) A glutathione Stransferase involved in vacuolar transfer encoded by the maize gene Bronze2. Nature 375, 397–400. Martens, S. and Forkmann, G. (1999) Cloning and expression of flavone synthase II from Gerbera hybrids. Plant Journal 20, 611–618. Martens, S., Forkmann, G., Matern, U. and Lukacin, R. (2001) Cloning of parsley flavone synthase I. Phytochemistry 58, 43–46. Martin, C. and Gerats, T. (1993) The control of flower coloration. In: Jordan, B.R. (ed.) The Molecular Biology of Flowering. CAB International, Wallingford, UK, pp. 219–256. Martin, C., Carpenter, R., Sommer, H., Saedler, H. and Coen, E.S. (1985) Molecular analysis of instability in flower pigmentation of Antirrhinum majus, following isolation of the pallida locus by transposon tagging. EMBO Journal 4, 1625–1630. Martin, C., Prescott, A., Mackay, S., Bartlett, J. and Vrijlandt, E. (1991) Control of anthocyanin biosynthesis in flowers

228

K. Davies and K. Schwinn of Antirrhinum majus. Plant Journal 1, 37–49. Martin, C., Jin, H. and Schwinn, K. (2001) Mechanisms and applications of transcriptional control of phenylpropanoid metabolism. In: Romeo, J.T., Saunders, J.A. and Matthews, B.F. (eds) Regulation of Phytochemicals by Molecular Techniques. Elsevier, Oxford, UK, pp. 155–170. Mathews, H., Clendennen, S.K., Caldwell, C.G., Liu, X.L., Connors, K., Matheis, N., Schuster, D.K., Menasco, D.J., Wagoner, W., Lightner, J. and Wagner, D.R. (2003) Activation tagging in tomato identifies a transcriptional regulator of anthocyanin biosynthesis, modification, and transport. Plant Cell 15, 1689–1703. Matsui, K., Tanaka, H. and Ohme-Takagi, M. (2004) Suppression of the biosynthesis of proanthocyanidin in Arabidopsis by a chimeric PAP1 repressor. Plant Biotechnology Journal 2, 487– 493. McCarty, D.R., Hattori, T., Carson, C.B., Vasil, V., Lazar, M. and Vasil, I.K. (1991) The Viviparous-1 developmental gene of maize encodes a novel transcriptional activator. Cell 66, 895– 905. Mehdy, M.C. and Lamb, C.J. (1987) Chalcone isomerase cDNA cloning and mRNA induction by elicitor, wounding and infection. EMBO Journal 6, 1527–1533. Meng, X. and Wang, X. (2004) Regulation of flower development and anthocyanin accumulation in Gerbera hybrida. Journal of Horticultural Science and Biotechnology 79, 131–137 Menkens, A.E., Schindler, U. and Cashmore, A.R. (1995) The G-box: a ubiquitous regulatory DNA element in plants bound by the GBF family of bZIP proteins. Trends in Biochemical Science 20, 506–510. Menssen, A., Hohmann, S., Martin, W., Schnable, P.S., Peterson, P.A., Saedler, H. and Gierl, A. (1990) The En/ Spm transposable element of Zea

mays contains splice sites at the termini generating a novel intron from a dSpm element in the A2 gene. EMBO Journal 9, 3051–3057. Meyer, P., Heidmann, I., Forkmann, G. and Saedler, H. (1987) A new petunia flower colour generated by transformation of a mutant with a maize gene. Nature 330, 677–678. Mercuri, A., Sacchetti, A., De Benedetti, L., Schiva, T. and Alberti, S. (2001) Green fluorescent flowers. Plant Science 161, 961–968. Miller, K.D., Strommer, J. and Taylor, L.P. (2002) Conservation in divergent solanaceous species of the unique gene structure and enzyme activity of a gametophytically-expressed flavonol 3-O-galactosyltransferase. Plant Molecular Biology 48, 233–242. Misawa, N., Masamoto, K., Hori, T., Ohtani, T., Boger, P. and Sandmann, G. (1994) Expression of an Erwinia phytoene desaturase gene not only confers multiple resistance to herbicides interfering with carotenoid biosynthesis but also alters xanthophyll metabolism in transgenic plants. Plant Journal 6, 481–489. Mizutani, M., Ward, E., DiMaio, J., Ohta, D., Ryals, J. and Sato, R. (1993) Molecular cloning and sequencing of a cDNA encoding mung bean cytochrome P450 (P450C4H) possessing cinnamate 4-hydroxylase activity. Biochemistry and Biophysics Research Communications 190, 875–880. Moehs, C.P., Tian, L., Osteryoung, K.W. and DellaPenna, D. (2001) Analysis of carotenoid biosynthetic gene expression during marigold petal development. Plant Molecular Biology 45, 281–293. Mooney, M., Desnos, T., Harrison, K., Jones, J., Carpenter, R. and Coen, E. (1995) Altered regulation of tomato and tobacco pigmentation genes caused by the delila gene of Antirrhinum. Plant Journal 7, 333–339. Mori, S., Kobayashi, H., Hoshi, Y., Kondo, M. and Nakano, M. (2004)

Developmental Control and Biotechnology of Floral Pigmentation Heterologous expression of the flavonoid 3’,5’-hydroxylase gene of Vinca major alters flower color in transgenic Petunia hybrida. Plant Cell Reports 22, 415–421. Moyano, E., Martinez-Garcia, J.F. and Martin, C. (1996) Apparent redundancy in myb gene function provides gearing for the control of flavonoid biosynthesis in Antirrhinum flowers. Plant Cell 8, 1519–1532. Mueller, L.A., Hinz, U., Uze, M., Sautter, C. and Zry¨d, J.P. (1997) Biochemical complementation of the betalain biosynthetic pathway in Portulaca grandiflora by a fungal 3,4-dihydroxyphenylalanine dioxygenase. Planta 203, 260–263. Mueller, L.A., Goodman, C.D., Silady, R.A. and Walbot, V. (2000) AN9, a petunia glutathione S-transferase required for anthocyanin sequestration, is a flavonoid-binding protein. Plant Physiology 123, 1561–1570. Naik, P.S., Chanemougasoundharam, A., Paul Khurana, S.M. and Kalloo, G. (2003) Genetic manipulation of carotenoid pathway in higher plants. Current Science 85, 1423–1430. Nakayama, T., Yonekura-Sakakibara, K., Sato, T., Kikuchi, S., Fukui, Y., Fukuchi-Mizutani, M., Ueda, T., Nakao, M., Tanaka, Y., Kusumi, T. and Nishino, T. (2000) Aureusidin synthase: a polyphenol oxidase homolog responsible for flower coloration. Science 290, 1163–1166. Nakayama, T., Sato, T., Fukui, Y., Yonekura-Sakakibara, K., Hayashi, H., Tanaka, Y., Kusumi, T. and Nishino, T. (2001) Specificity analysis and mechanism of aurone synthesis catalyzed by aureusidin synthase, a polyphenol oxidase homolog responsible for flower coloration. FEBS Letters 499, 107–111. Nakayama, T., Suzuki, H. and Nishino, T. (2003) Anthocyanin acyltransferases: specificities, mechanism, phylogenetics, and applications. Journal of Molecular Catalysis 23, 117–132.

229

Napoli, C., Lemieux, C. and Jorgensen, R. (1990) Introduction of a chimeric chalcone synthase gene into Petunia results in reversible co-suppression of homologous genes in trans. Plant Cell 2, 279–289. Nielsen, K.M., Deroles, S.C., Markham, K.R., Bradley, J.M., Podivinsky, E. and Manson, D. (2002) Antisense flavonol synthase alters copigmentation and flower color in lisianthus. Molecular Breeding 9, 217–229. Nielsen, K.M., Lewis, D.H. and Morgan, E.R. (2003) Characterization of carotenoid pigments and their biosynthesis in two yellow flowered lines of Sandersonia aurantiaca (Hook). Euphytica 130, 25–34. Nishihara, M., Nakatsuka, T., Mishiba, K., Kikuchi, A. and Yamamura, S. (2003) Flower color modification by suppression of chalcone synthase gene in gentian. Plant Cell Physiology 44(Suppl.1), 59. Noda, N., Kato, N., Kogawa, K., Kazuma, K. and Suzuki, M. (2004) Cloning and characterization of the gene encoding anthocyanin 3’,5’-O-glucosyltransferase involved in ternatin biosynthesis from blue petals of butterfly pea (Clitoria ternatea). Plant Cell Physiology 45(Suppl.1), 32. Okinaka, Y., Shimada, Y., Nakano-Shimada, R., Ohbayashi, M., Kiyokawa, S. and Kikuchi, Y. (2003) Selective accumulation of delphinidin derivatives in tobacco using a putative flavonoid 3’,5’-hydroxylase cDNA from Campanula medium. BioScience, Biotechnology and Biochemistry 67, 161–165. Ono, E., Mizutani, M., Nakamura, N., Fukui, Y., Sakaibara, K., Nakayama, T., Yamaguchi, M., Kusumi, T. and Tanaka, Y. (2005) Molecular breeding of yellow flowers by expressing chalcone 4’-glucosyltransfease and aureusidin synthase genes. Plant Cell Physiology 46(Suppl.), 53. O’Reilly, C., Sheperd, N., Pereira, A., Schwarz-Somer, Z., Bertram, I.,

230

K. Davies and K. Schwinn Robertson, D.S., Peterson, P.A. and Saedler, H. (1985) Molecular cloning of the A1 locus of Zea mays using the transposable elements En and Mu1. EMBO Journal 4, 877–882. Pairoba, C.F. and Walbot, V. (2003) Post-transcriptional regulation of expression of the Bronze2 gene of Zea mays L. Plant Molecular Biology 53, 75–86. Park, H., Kreunen, S.S., Cuttriss, A.J., DellaPenna, D. and Pogson, B.J. (2002) Identification of the carotenoid isomerase provides insight into carotenoid biosynthesis, prolamellar body formation, and photomorphogenesis. Plant Cell 14, 321–332. Park, K.-I., Choi, J.-D., Hoshino, A., Morita, Y. and Iida, S. (2004) An intragenic tandem duplication in a transcriptional regulatory gene for anthocyanin biosynthesis confers pale-colored flowers and seeds with fine spots in Ipomoea tricolor. Plant Journal 38, 840–849. Park, S.Y. and Kim, Y. (2000) The expression of egg plant flavonoid 3’,5’hydroxylase gene in tobacco plants (Nicotiana tabacum cv. Xanthi). Journal of Plant Biotechnology 2, 25–28. Paxton, R.J. and Tengo, J. (2001) Doubly duped males: the sweet and sour of the orchid’s bouquet. Trends in Ecology and Evolution 16, 167–169. Paz-Ares, J., Peterson, P.A. and Saedler, H. (1990) Molecular analysis of the C1-I allele from Zea mays: a dominant mutant of the regulatory C1 locus. EMBO Journal 9, 315–321. Petroni, K., Tonelli, C. and Paz-Ares, J. (2002) The MYB transcription factor family: from maize to Arabidopsis. Maydica 47, 213–232. Pogson, B.J. and Rissler, H.M. (2000) Genetic manipulation of carotenoid biosynthesis and photoprotection. Philosophical Transactions of the Royal Society of London Series B 355, 1395–1403. Polashock, J.J., Griesbach, R.J., Sullivan, R.F. and Vorsa, N. (2002) Cloning of a cDNA encoding the cranberry dihydro-

flavonol-4-reductase (DFR) and expression in transgenic tobacco. Plant Science 163, 241–251. Quattrocchio, F., Wing, J.F., Leppen, H.T.C., Mol, J.N.M. and Koes, R.E. (1993) Regulatory genes controlling anthocyanin pigmentation are functionally conserved among plant species and have distinct sets of target genes. Plant Cell 5, 1497–1512. Quattrocchio, F., Wing, J.F., van der Woude, K., Mol, J.N.M. and Koes, R. (1998) Analysis of bHLH and MYBdomain proteins: species-specific regulatory differences are caused by divergent evolution of target anthocyanin genes. Plant Journal 13, 475–488. Quattrocchio, F., Wing, J., van der Woude, K., Souer, E, de Vetten, N., Mol, J. and Koes, R. (1999) Molecular analysis of the anthocyanin2 gene of petunia and its role in the evolution of flower color. Plant Cell 11, 1433– 1444. Ralley, L., Enfissi, E.M.A., Misawa, N., Schuch, W., Bramley, P.M. and Fraser, P.D. (2004) Metabolic engineering of ketocarotenoid formation in higher plants. Plant Journal 39, 477–486. Ray, H., Yu, M., Auser, P., Blahut-Beatty, L., McKersie, B., Bowley, S., Westcott, N., Coulman, B., Lloyd, A. and Gruber, M.Y. (2003) Expression of anthocyanins and proanthocyanidins after transformation of lucerne with maize Lc. Plant Physiology 132, 1448– 1463. Ray, J., Moureau, P., Bird, C., Bird, A., Grierson, D., Maunders, M., Truesdale, M., Bramley, P. and Schuch, W. (1992) Cloning and characterisation of a gene involved in phytoene synthesis from tomato. Plant Molecular Biology 19, 401–404. Reimold, U., Kroger, M., Kreuzaler, F. and Hahlbrock, K. (1983) Coding and 3’ non-coding nucleotide sequence of chalcone synthase mRNA and assignment of amino acid sequence of the enzyme. EMBO Journal 2, 1801– 1805.

Developmental Control and Biotechnology of Floral Pigmentation Rissler, H.M. and Pogson, B.J. (2001) Antisense inhibition of the beta-carotene hydroxylase enzyme in Arabidopsis and the implications for carotenoid accumulation, photoprotection and antenna assembly. Photosynthesis Research 67, 127–137. Robbins, M.P., Paolocci, F., Hughes, J.W., Turchetti, V., Allison, G., Arcioni, S., Morris, P. and Damiani, F. (2003) Sn, a maize bHLH gene, modulates anthocyanin and condensed tannin pathways in Lotus corniculatus. Journal of Experimental Botany 54, 239–248. Roesler, K.R., Shorrosh, B.S. and Ohlrogge, J.B. (1994) Structure and expression of an Arabidopsis acetylcoenzyme A carboxylase gene. Plant Physiology 105, 611–617. ¨ mer, S., Fraser, P.D., Kiano, J.W., Ro Shipton, C.A., Misawa, N., Schuch, W. and Bramley, P.M. (2000) Elevation of the provitamin A content of transgenic tomato plants. Nature Biotechnology 18, 666–669. ¨ mer, S., Lubeck, J., Kauder, F., SteiRo ger, S., Adomat, C. and Sandmann, G. (2002) Genetic engineering of a zeaxanthin-rich potato by antisense inactivation and co-suppression of carotenoid epoxidation. Metabolic Engineering 4, 263–272. Ronen, G., Carmel-Goren, L., Zamir, D. and Hirschberg, J. (2000) An alternative pathway to b-carotene formation in plant chromoplasts discovered by map-based cloning of Beta and oldgold color mutations in tomato. Proceedings of the National Academy of Sciences USA 97, 11102–11107. Rosati, C., Cadic, A., Duron, M., Renou, J.P. and Simoneau, P. (1997) Molecular cloning and expression analysis of dihydroflavonol 4-reductase gene in flower-organs of Forsythia  intermedia. Plant Molecular Biology 35, 303–311. Rosati, C., Aquilani, R., Dharmapuri, S., Pallara, P., Marusic, C., Tavazza, R., Bouvier, F., Camara, B. and Giuliano, G.

231

(2000) Metabolic engineering of betacarotene and lycopene content in tomato fruit. Plant Journal 24, 413–419. Rosati, C., Simoneau, P., Treutter, D., Poupard, P., Cadot, Y., Cadic, A. and Duron, M. (2003) Engineering of flower colour in forsythia by expression of two independently transformed dihydroflavonol 4-reductase and anthocyanidin synthase genes of flavonoid pathway. Molecular Breeding 12, 197–208. Sablowski, R.W.M., Moyano, E., Culianez-Macia, F.A., Schuch, W., Martin, C. and Bevan, M. (1994) A flower-specific Myb protein activates transcription of phenylpropanoid biosynthetic genes. EMBO Journal 13, 128–137. Sainz, M.B., Grotewold, E. and Chandler, V.L. (1997) Evidence for direct activation of an anthocyanin promoter by the maize C1 protein and comparison of DNA binding by related Myb domain proteins. Plant Cell 9, 611– 625. Saito, K. and Yamazaki, M. (2002) Biochemistry and molecular biology of the late-stage of biosynthesis of anthocyanin: lessons from Perilla frutescens as a model plant. New Phytologist 155, 9–23. Sasaki, N., Adachi, T., Koda, T. and Ozeki, Y. (2004) Detection of UDPglucose:cyclo-DOPA 5-O-glucosyltransferase activity in four o’clocks (Mirabilis jalapa L.). FEBS Letters 568, 159–162. Schledz, M., Al-Babili, S., Von Lintig, J., Haubruck, H., Rabbani, S., Kleinig, H. and Beyer, P. (1996) Phytoene synthase from Narcissus pseudonarcissus: functional expression, galactolipid requirement, topological distribution in chromoplasts and induction during flowering. Plant Journal 10, 781–792. Schijlen, E.G.W.M., de Vos, C.H.R., van Tunen, A.J. and Bovy, A.G. (2004) Modification of flavonoid biosynthesis in crop plants. Phytochemistry 65, 2631–2648.

232

K. Davies and K. Schwinn Schliemann, W., Kobayashi, N. and Strack, D. (1999) The decisive step in betaxanthin biosynthesis is a spontaneous reaction. Plant Physiology 119, 1217–1232. Schwechheimer, C. and Deng, X.W. (2001) COP9 signalosome revisited: a novel mediator of protein degradation. Trends in Cell Biology 11, 420–426. Schwender, J., Muller, C., Zeidler, J. and Lichtenthaler, H.K. (1999) Cloning and heterologous expression of a cDNA encoding 1-deoxy-D-xylulose5-phosphate reductoisomerase of Arabidopsis thaliana. FEBS Letters 455, 140–144. Schwinn, K.E. and Davies, K.M. (2004) Flavonoids. In: Davies, K. (ed.) Plant Pigments and their Manipulation. Blackwell, Oxford, UK, pp. 92–149. Schwinn, K.E., Markham, K.R. and Given, N.K. (1994) Floral flavonoids and the potential for pelargonidin biosynthesis in commercial chrysanthemum cultivars. Phytochemistry 35, 145–150. Schwinn, K.E., Davies, K.M., Deroles, S.C., Markham, K., Miller, R.M., Bradley, M., Manson, D.G. and Given, N.K. (1997) Expression of an Antirrhinum majus UDP-glucose:flavonoid-3-O-glucosyltransferase transgene alters flavonoid glycosylation and acylation in lisianthus (Eustoma grandiflorum Grise.). Plant Science 125, 53–61. Schwinn, K., Alm, V., Mackay, S., Davies, K. and Martin, C. (2001) Regulation of anthocyanin biosynthesis in Antirrhinum. Acta Horticulturae 560, 201–206. Shewmaker, C.K., Sheehy, J.A., Daley, M., Colburn, S. and Ke, D.Y. (1999) Seed-specific overexpression of phytoene synthase: increase in carotenoids and other metabolic effects. Plant Journal 20, 401–412. Shimada, Y., Nakano-Shimada, R., Ohbayashi, M., Okinaka, Y., Kiyokawa, S. and Kikuchi, Y. (1999) Expression of chimeric P450 genes encoding flavonoid-3’,5’-hydroxylase

in transgenic tobacco and petunia plants. FEBS Letters 461, 241–245. Shimada, Y., Ohbayashi, M., Nakano-Shimada, R., Okinaka, Y., Kiyokawa, S. and Kikuchi, Y. (2001) Genetic engineering of the anthocyanin biosynthetic pathway with flavonoid-3’,5’-hydroxylase: specific switching of the pathway in petunia. Plant Cell Reports 20, 456–462. Shorrosh, B.S., Dixon, R.A. and Ohlrogge, J.B. (1994) Molecular cloning, characterization, and elicitation of acetyl-CoA carboxylase from lucerne. Proceedings of the National Academy of Sciences USA 91, 4323– 4327. Simpson, K.L., Lee, T.-C., Rodriguez, D.B. and Chichester, C.O. (1976) Metabolism in senescent and stored tissues. In: Goodwin, T.W. (ed.) Chemistry and Biochemistry of Plant Pigments, Vol. 1. Academic Press, London, pp. 780–842. Solano, R., Nieto, C., Avila, J., Canas, L., Diaz, I. and Paz-Ares, J. (1995) Dual DNA binding specificity of a petal epidermis-specific MYB transcription factor (MYB.Ph3) from Petunia hybrida. EMBO Journal 14, 1773–1784. Sompornpailin, K., Makita, Y., Yamazaki, M. and Saito, K. (2002) A WD-repeatcontaining putative regulatory protein in anthocyanin biosynthesis in Perilla frutescens. Plant Molecular Biology 50, 485–495. Spelt, C., Quattrocchio, F., Mol, J.N.M. and Koes, R. (2000) Anthocyanin1 of petunia encodes a basic helix-loophelix protein that directly activates transcription of structural anthocyanin genes. Plant Cell 12, 1619–1631. Spelt, C., Quattrocchio, F., Mol, J. and Koes, R. (2002) ANTHOCYANIN1 of petunia controls pigment synthesis, vacuolar pH, and seed coat development by genetically distinct mechanisms. Plant Cell 14, 2121–2135. Springob, K., Nakajima, J., Yamazaki, M. and Saito, K. (2003) Recent advances in the biosynthesis and accumulation of

Developmental Control and Biotechnology of Floral Pigmentation anthocyanins. Natural Products Reports 20, 288–233. Staiger, D. (2002) Circadian rhythms in Arabidopsis: time for nuclear proteins. Planta 214, 334–344. Sta˚lberg, K., Lindgren, O., Ek, B. and ¨ glund, A.S. (2003) Synthesis of Ho ketocarotenoids in the seed of Arabidopsis thaliana. Plant Journal 36, 771–779. Strack, D., Vogt, T. and Schliemann, W. (2003) Recent advances in betalain research. Phytochemistry 62, 247– 269. Sugimoto, K., Takeda, S. and Hirochika, H. (2000) MYB-related transcription factor NtMYB2 induced by wounding and elicitors is a regulator of the tobacco retrotransposon Tto1 and defense-related genes. Plant Cell 12, 2511–2528. Sun, Z.R., Gantt, E. and Cunningham, F.X. (1996) Cloning and functional analysis of the beta-carotene hydroxylase of Arabidopsis thaliana. Journal of Biological Chemistry 271, 24349– 24352. Suzuki, H., Nakayama, T., YonekuraSakakibara, K., Fukui, Y., Nakamura, N., Nakao, N., Tanaka, Y., Yamaguchi, M., Kusumi, T. and Nishino, T. (2001) Malonyl-CoA:anthocyanin 5-O-glucoside-6’’’-O-malonyltransferase from scarlet sage (Salvia splendens) flowers. Journal of Biological Chemistry 276, 49013–49019. Suzuki, H., Nakayama, T., YonekuraSakakibara, K., Fukui, Y., Nakamura, N., Yamaguchi, M., Tanaka, Y., Kusumi, T. and Nishino, T. (2002) cDNA cloning, heterologous expressions, and functional characterization of malonyl-coenzyme A:anthocyanidin 3-O-glucoside-6’’-O-malonyltransferase from dahlia flowers. Plant Physiology 130, 2142–2151. Suzuki, H., Sawada, K., Yonekura-Sakakibara, K., Nakayama, T., Yamaguchi, M.-A. and Nishino, T. (2003) Identification of a cDNA encoding malonylcoenzyme A:anthocyanidin 3-O-gluco-

233

side-6’’-O-malonyltransferase from cineraria (Senecio cruentus) flowers. Plant Biotechnology 20, 229–234. Suzuki, H., Nakayama, T., Yamaguchi, M.-A. and Nishino, T. (2004a) cDNA cloning and characterization of two Dendranthema  morifolium anthocyanin malonyltransferases with different functional activities. Plant Science 166, 89–96. Suzuki, H., Sawada, S., Watanabe, K., Nagae, S., Yamaguchi, M., Nakayama, T. and Nishino, T. (2004b) Identification and characterization of a novel anthocyanin malonyltransferase from scarlet sage (Salvia splendens) flowers: an enzyme that is phylogenetically separated from other anthocyanin acyltransferases. Plant Journal 38, 994– 1003. Suzuki, K., Xue, H., Tanaka, Y., Fukui, Y., Fukuchi-Mizutani, M., Katsumoto, Y., Tsuda, S. and Kusumi, T. (2000) Flower color modifications of Torenia hybrida by cosuppression of anthocyanin biosynthesis genes. Molecular Breeding 6, 239–246. Szymanski, D.B., Lloyd, A.M. and Marks, M.D. (2000) Progress in the molecular genetic analysis of trichome initiation and morphogenesis in Arabidopsis. Trends in Plant Science 5, 214–219. Tamagnone, L., Merida, A., Parr, A., Mackay, S., Culianez-Macia, F.A., Roberts, K. and Martin, C. (1998) The AmMYB308 and AmMYB330 transcription factors from Antirrhinum regulate phenylpropanoid and lipid biosynthesis in transgenic tobacco. Plant Cell 10, 135–154. Tanaka, Y., Fukui, Y., Fukuchi-Mizutani, M., Holton, T.A., Higgins, E. and Kusumi, T. (1995) Molecular cloning and characterization of Rosa hybrida dihydroflavonol 4-reductase gene. Plant and Cell Physiology 36, 1023–1031. Tanaka, Y., Katsumoto, Y., Brugliera, F. and Mason, J. (2005) Genetic engineering in floriculture. Plant, Cell, Tissue and Organ Culture 80, 1–24.

234

K. Davies and K. Schwinn Tanner, G.J. (2004) Condensed tannins. In: Davies, K.M. (ed.) Plant Pigments and their Manipulation. Annual Plant Reviews, Sheffield Academic Press, Sheffield, UK, pp. 150–184. Taylor, L.P. and Jorgensen, R. (1992) Conditional male fertility in chalcone synthase-deficient petunia. Journal of Heredity 83, 11–17. Teutsch, H.G., Hasenfratz, M.P., Lesot, A., Stoltz, C., Garnier, J.M., Jeltsch, J.M., Durst, F. and Werck-Reichhart, D. (1993) Isolation and sequence of a cDNA encoding the Jerusalem artichoke cinnamate 4-hydroxylase, a major plant cytochrome P450 involved in the general phenylpropanoid pathway. Proceedings of the National Academy of Sciences USA 90, 4102–4106. Tian, L., Musetti, V., Kim, J., MagallanesLundback, M. and DellaPenna, D. (2004) The Arabidopsis LUT1 locus encodes a member of the cytochrome P450 family that is required for carotenoid epsilon-ring hydroxylation activity. Proceedings of the National Academy of Sciences USA 101, 402–407. Ueyama, Y., Suzuki, K., Fukuchi-Mizutani, M., Fukui, Y., Miyazaki, K., Ohkawa, H., Kusumi, T. and Tanaka, Y. (2002) Molecular and biochemical characterization of torenia flavonoid 3’-hydroxylase and flavone synthase II and modification of flower color by modulating the expression of these genes. Plant Science 163, 253–263. Ulm, R., Baumann, A., Oravecz, A., Mate, Z., Adam, E., Oakeley, E.J., Schafer, E. and Nagy, F. (2004) Genome-wide analysis of gene expression reveals function of the bZIP transcription factor HY5 in the UV-B response of Arabidopsis. Proceedings of the National Academy of Sciences USA 101, 1397–1402. van der Krol, A.R., Lenting, P.E., Veenstra, J.G., van der Meer, I.M., Koes, R.E., Gerats, A.G.M., Mol, J.N.M. and Stuitje, A.R. (1988) An antisense

chalcone synthase gene in transgenic plants inhibits flower pigmentation. Nature 333, 866–869. van der Krol, A.R., Mur, L.A., Beld, M., Mol, J.N.M. and Stuitje, A.R. (1990) Flavonoid genes in petunia: addition of a limited number of additional copies may lead to a suppression of gene activity. Plant Cell 2, 291– 299. van Nocker, S. and Ludwig, P. (2003) The WD-repeat protein superfamily in Arabidopsis: conservation and divergence in structure and function. BMC Genomics 4, 50 (http://www.biomedcentral.com/ 1471–2164/4/50). Verhoeven, A.S., Bugos, R.C. and Yamamoto, H.Y. (2001) Transgenic tobacco with suppressed zeaxanthin formation is susceptible to stress-induced photoinhibition. Photosynthesis Research 67, 27–39. Vishnevetsky, M., Ovadis, M., Itzhaki, H. and Vainstein, A. (1997) CHRC, encoding a chromoplast-specific carotenoid-associated protein, is an early gibberellic acid-responsive gene. Journal of Biological Chemistry, 272, 24747–24750. Vishnevetsky, M., Ovadis, M. and Vainstein, A. (1999a) Carotenoid sequestration in plants: the role of carotenoid-associated proteins. Trends in Plant Science 4, 232–235. Vishnevetsky, M., Ovadis, M., Zuker, A. and Vainstein, A. (1999b) Molecular mechanisms underlying carotenogenesis in the chromoplast: multilevel regulation of carotenoid-associated genes. Plant Journal 20, 423–431. Vogt, T. (2000) Glycosyltransferases involved in plant secondary metabolism. In: Romeo, J.T., Ibrahim, R., Varin, L. and De Luca, V. (eds) Evolution of Metabolic Pathways. Elsevier, Oxford, UK, pp. 317–347. Vogt, T. (2002) Substrate specificity and sequence analysis define a polyphyletic origin of betanidin 5- and 6-O-glucosyltransferase from Dorotheanthus bellidiformis. Planta 214, 492–495.

Developmental Control and Biotechnology of Floral Pigmentation Vogt, T., Grimm, R. and Strack, D. (1999) Cloning and expression of a cDNA encoding betanidin 5-O-glucosyltransferase, a betanidin- and flavonoid-specific enzyme with high homology to inducible glucosyltransferases from the Solanaceae. Plant Journal 19, 509–519. Vom Endt, D., Kijne, J.W. and Memelink, J. (2002) Transcription factors controlling plant secondary metabolism: what regulates the regulators? Phytochemistry 61, 107–114. ¨ vel, U. and Ro ¨ mer, S. Wagner, T., Windho (2002) Transformation of tobacco with a mutated cyanobacterial phytoene desaturase gene confers resistance to bleaching herbicides. Zeitschrift fu¨r Naturforschung 57c, 671–679. Wakelin, A.M., Lister, C.E. and Conner, A.J. (2003) Inheritance and biochemistry of pollen pigmentation in California poppy (Eschscholzia californica CHAM.). International Journal of Plant Science 164, 867–875. Walker, A.R., Davison, P.A., BolognesiWinfield, A.C., James, C.M., Srinivasan, N., Blundell, T.L., Esch, J.J., Marks, M.D. and Gray, J.C. (1999) The TRANSPARENT TESTA GLABRA1 locus, which regulates trichome differentiation and anthocyanin biosynthesis in Arabidopsis, encodes a WD40 repeat protein. Plant Cell 11, 1337–1350. Weiss, D. (2000) Regulation of flower pigmentation and growth: multiple signaling pathways control anthocyanin synthesis in expanding petals. Physiologica Plantarum 110, 152–157. Weiss, M.R. (1995) Floral colour change: a widespread functional convergence. American Journal of Botany 82, 167–185. Weisshaar, B. and Jenkins, G.I. (1998) Phenylpropanoid biosynthesis and its regulation. Current Opinion in Plant Biology 1, 251–257. ¨ der, G., Schiltz, E., GriseWelle, R., Schro ¨ der, J. (1991) Inbach, H. and Schro duced plant responses to pathogen

235

attack: analysis and heterologous expression of the key enzyme in the biosynthesis of phytoalexins in soybean (Glycine max L. Merr. cv. Harosoy 63). European Journal of Biochemistry 196, 423–430. Winefield, C.S., Lewis, D.H., Swinny, E.E., Zhang, H., Arathoon, H.S., Fischer, T.C., Halbwirth, H., Stich, K., Gosch, C., Forkmann, G. and Davies, K.M. (2005) Investigation of 3-deoxyanthocyanin biosynthesis in Sinningia cardinalis. Physiologia Plantarum 124, 419–430. Winefield, C. (2002) The final steps in anthocyanin formation: a story of modification and sequestration. Advances in Botanical Research 37, 55–74. Winkel, B.S.J. (2004) Metabolic channeling in plants. Annual Review of Plant Biology 55, 85–107. Xie, D.-Y., Sharma, S.B., Pavia, N.L., Ferreira, D. and Dixon, R.A. (2003) Role of anthocyanidin reductase, encoded by BANYULS in plant flavonoid biosynthesis. Science 299, 396–399. Xie, D.-Y., Jackson, L.A., Cooper, J.D., Ferreira, D. and Dixon, R.A. (2004) Molecular and biochemical analysis of two cDNA clones encoding dihydroflavonol-4-reductase from Medicago truncatula. Plant Physiology 134, 979–994. Yamazaki, M., Gong, Z., Fukuchi-Mizutani, M., Fukui, Y., Tanaka, Y., Kusumi, T. and Saito, K. (1999) Molecular cloning and biochemical characterization of a novel anthocyanin 5O-glucosyltransferase by mRNA differential display for plant forms regarding anthocyanin. Journal of Biological Chemistry 274, 7405–7511. Yamazaki, M., Yamagishi, E., Gong, Z., Fukuchi-Mizutani, M., Fukui, Y., Tanaka, Y., Kusumi, T., Yamaguchi, T. and Saito, K. (2002) Two flavonoid glucosyltransferases from Petunia hybrida: molecular cloning, biochemical properties and developmentally

236

K. Davies and K. Schwinn regulated expression. Plant Molecular Biology 48, 401–411. Yang, S., Sweetman, J.P., Amirsadeghi, S., Barghchi, M., Huttly, A.K., Chung, W.I. and Twell, D. (2001) Novel anther-specific myb genes from tobacco as putative regulators of phenylalanine ammonia-lyase expression. Plant Physiology 126, 1738–1753. Ye, X.D., Al-Babili, S., Kloti, A., Zhang, J., Lucca, P., Beyer, P. and Potrykus, I. (2000) Engineering the provitamin A (beta-carotene) biosynthetic pathway into (carotenoid-free) rice endosperm. Science 287, 303–305. Yonekura-Sakakibara, K., Tanaka, Y., Fukuchi-Mizutani, M., Fujiwara, H., Fukui, Y., Ashikari, T., Murakami, Y., Yamaguchi, M. and Kusumi, T. (2000) Molecular and biochemical characterization of a novel hydroxycinnamoylCoA: anthocyanin 3-O-glucoside-6’’O-acyltransferase from Perilla frutescens. Plant and Cell Physiology 41, 495–502. Yu, L.M., Lamb, C.J. and Dixon, R.A. (1993) Purification and biochemical characterization of proteins which bind to the H-box cis-element implicated in transcriptional activation of plant defense genes. Plant Journal 3, 805–816. Yu, O., Shi, J., Hession, A.O., Maxwell, C.A., McGonigle, B. and Odell, J.T. (2003) Metabolic engineering to increase isoflavone biosynthesis in soybean seed. Phytochemistry 63, 753–763. Zhang, P.F., Wang, Y.B., Zhang, J.B., Maddock, S., Snook, M. and Peterson,

T. (2003) A maize QTL for silk maysin levels contains duplicated Myb-homologous genes which jointly regulate flavone biosynthesis. Plant Molecular Biology 52, 1–15. Zhu, C.F., Yamamura, S., Koiwa, H., Nishihara, M. and Sandmann, G. (2002) cDNA cloning and expression of carotenogenic genes during flower development in Gentiana lutea. Plant Molecular Biology 48, 277– 285. Zhu, C.F., Yamamura, S., Nishihara, M., Koiwa, H. and Sandmann, G. (2003) cDNAs for the synthesis of cyclic carotenoids in petals of Gentiana lutea and their regulation during flower development. Biochimica et Biophysica Acta – Gene Structure and Expression 1625, 305–308. Zachgo, S., de Andrade Silva, E., Motte, ¨ bner, W., Saedler, H. and P., Tro Schwarz-Sommer, Z. (1995) Functional analysis of the Antirrhinum floral homeotic DEFICIENS gene in vivo and in vitro by using a temperature-sensitive mutant. Development 121, 2861–2875. Zry¨d, J-.P. and Christinet, L. (2004) Betalains. In: Davies, K. (ed.) Plant Pigments and their Manipulation. Blackwell, Oxford, UK, pp. 185–213. Zuker, A., Tzfira, T., Ben-Meir, H., Ovadis, M., Schklarman, E., Itzhaki, H., Forkmann, G., Martens, S., NetaSharir, I., Weiss, D. and Vainstein, A. (2002) Modification of flower colour and fragrance by antisense suppression of the flavanone 3-hydroxylase gene. Molecular Breeding 9, 33–41.

9

Biotechnology of Floral Development C. WINEFIELD AND B.R. JORDAN Cell Biology Group, Agriculture and Life Sciences Division, Lincoln University, PO Box 84, Lincoln, Canterbury, New Zealand

Introduction The understanding of floral development at a molecular level has proceeded at an unprecedented rate (see Jordan, 1993, for reviews of progress early last decade). Molecular genetic studies on two model plants, Antirrhinum and Arabidopsis in particular, have provided the foundation for this progress through the initial identification of genes involved in flowering (see below). Recent efforts at elucidating the regulatory networks controlling the transition to flowering in the model plant Arabidopsis have identified over 80 independent loci, which have important roles in the control of flowering in this species (Henderson and Dean, 2004). An even larger number of genes has been discovered controlling events downstream of floral initiation. This extensive compliment of genes also forms very complicated regulatory networks that integrate environmental and developmental mechanisms to coordinate the reproductive process. There have been many excellent review articles published recently, including those found in this volume, that provide an in-depth study of those genes involved in floral timing and floral organogenesis (Zik and Irish, 2003; Boss et al., 2004; Buzgo et al., 2004; Henderson and Dean, 2004; Komeda, 2004). The reader is encouraged to seek these out in order to appreciate the complexity inherent in the control of this vitally important developmental process. Many agriculturally important crops have had flowering and plant architectural phenotypes selected as part of ongoing domestication programmes. Selection of appropriate flowering time, floral, fruit and plant architecture has been a focus for many breeding programmes. Coincidently, many significant plant traits have arisen through fortuitous mutations that have been identified and selected for, in these programmes. Two good examples of this are the rapid domestication of maize, of which the modern form of this plant has been recently shown to be due to mutations of the TEOSINTE locus (Doebley et al., 1997; Lukens and Doebley, 2001), and the generation of a high yield ßCAB International 2006. The Molecular Biology and Biotechnology of Flowering, 2nd edn (ed. B.R. Jordan)

237

238

C. Winefield and B.R. Jordan

semi-dwarf rice variety in the 1960s that has now been attributed to a mutation in the SD-1 gene (Sasaki et al., 2002; Spielmeyer et al., 2002). Plant reproductive development clearly occupies a central role within agriculture. The majority of plant-based foods are floral structures and propagation of many food crops is via sexual reproduction. In addition, commercially important plant fibres such as cotton are also derived from reproductive structures (specialized terminally differentiated trichomes on the seed coat). Furthermore, the reproductive process itself can have profound impacts on the quality and quantities of non-floral organs as is the case in forestry (Brunner and Nilsson, 2004). The long juvenile period prior to reproductive development, commonly observed in many tree species, not only represents a considerable problem for breeding programmes but can also have an impact on wood volume and quality (Brunner and Nilsson, 2004). Overall, from the brief outline above, it is no surprise that a large and increasing effort has been deployed to gain a better understanding of the underlying principles and control of the flowering and reproductive processes. While many flowering mutants have been known for hundreds of years, the first genes to be cloned were FLORICAULA (FLO) from Antirrhinum (Coen et al., 1990) and LEAFY (LFY) from Arabidopsis (Weigel et al., 1992). Mutations in these genes cause a homeotic conversion of flowers to leaf-like structures (Weigel et al., 1992). These discoveries were rapidly followed by the identification of the genes affected by a raft of floral homeotic mutations that generally belong to a group of transcriptional regulators known as MADS-box genes. This, in turn, eventually led to the formation of the ABC model of floral development (Weigel and Meyerowitz, 1994), which has recently been modified into the quartet model of flowering (Theissen and Saedler, 2001; Ferrario et al., 2004). The homeotic transformations caused by mutations of this sort and the results of ectopic overexpression experiments automatically suggest the potential for biotechnology applications capable of rapidly altering reproductive development in agriculturally important crops without the difficulties associated with traditional breeding approaches. Although many genes involved in flowering have multiple roles both within flowering and in other developmental processes, there are a number of attractive ‘control points’ that may be useful from a biotechnological point of view. This chapter outlines current methodology and examples of targets for manipulating flowering (further targets for biotechnological manipulation are also identified in other chapters). To provide a different perspective, it also discusses flowering in trees, grapes and kiwifruit, agriculturally important crops that differ quite markedly from the rapid-cycling habit of currently employed model systems.

Methods for Gene Discovery and Manipulation Induced and natural mutation Modern agri-biotechnology techniques are based on commonly used experimental techniques for gene discovery and functional analysis. These normally

Biotechnology of Floral Development

239

take the form of methods aimed at removing (or reducing) a given gene activity or promoting a given gene activity to a very high level in an ectopic manner. Natural variations, due to acquired mutation, or introduced mutations such as T-DNAs or transposons are preferred methods for generating nulls in a particular gene for the analysis of function. Though this type of approach is useful in model plant systems, the applicability in a biotechnological sense is limited. Crops with endogenously active transposons, such as maize, can be screened for mutations in a given gene via a reverse genetics-type approach (Meely and Briggs, 1995). However, this approach is very labour intensive and limited to those crops with highly active characterized transposons. In other crops the generation of useful tagged mutant populations, either through the introduction of transposons or through random insertion of T-DNAs, is dependent on highly efficient transformation systems. Therefore for commercial biotechnology, these approaches realistically represent a blind alley. Recently, a reverse genetic system has been developed in Arabidopsis that makes use of chemically mutagenized populations, bioinformatics and a PCR screening method for identification of individual plant lines containing the lesions of interest. This system, termed Targeting Induced Local Lesions In Genomes (TILLING), has proved to be very useful in a wide range of plant systems and has the potential to be used in any plant system (Henikoff et al., 2004). The system is based on ethylmethanesulphonate (EMS)-mutagenized populations. This mutagen predominantly produces G:C to A:T transitions. A mutagenized population (generally M2) is grown up and individual DNA samples taken. These samples are then pooled for later PCR analysis and mutant identification based on a procedure using the CEL I mismatch cleavage enzyme (Oleykowski et al., 1998). Using a set of user-friendly web-based tools, CODDLE for finding optimal regions to be TILLed and PARSESNP for mapping and interpreting the mutations delivered, primers are designed to a particular gene of interest (Till et al., 2003). These primers are then used to PCR screen the pooled DNA samples from the mutagenized population and individual plants containing lesions most likely to be deleterious to gene function identified. This method offers a number of advantages over insertional or ‘tagged’ mutagenized populations in that a range of alleles can be obtained, not all of which will be lethal and transformation systems are not required for the production of the mutant lines. This system is likely to prove applicable to a wide range of crop species, many of which have already been subjected to EMS mutagenesis as part of long-standing breeding programmes.

Gene silencing A more favoured manipulation method is the use of antisense and co-suppression-mediated silencing of a target gene. Over the last 3–4 years the molecular basis underlying the ability to silence genes using these methods has begun to be resolved, with some startling implications for gene regulation (Carthew, 2001; Waterhouse et al., 2001; Wang and Waterhouse, 2002; Kidner and Martienssen, 2005). The consequence of these discoveries is that

240

C. Winefield and B.R. Jordan

antisense and co-suppression silencing are now considered to act through similar mechanisms termed post translational gene silencing (PTGS – in plants) and RNA interference (RNAi) in initial work on Caenorhabditis elegans. These mechanisms are proposed to have evolved to control the proliferation of extra chromosomal elements such as retrotransposons and RNA viruses (BurchSmith et al., 2004). However, very recent work has indicated that the role of this system extends to the transcriptional regulation of certain genes in plants with a strong bias towards transcriptional regulators, some of which appear to be involved in floral development (Reinhart et al., 2002; Carrington and Ambros, 2003). The mechanism is triggered by the presence of doublestranded RNA species that are then cut into small 19–25 nucleotide RNAs called small interfering RNAs (siRNAs). These siRNAs then prime the system and target homologous sequences for degradation and in some cases interfere with translation (Waterhouse et al., 2001; Wang and Waterhouse, 2002). Typically, antisense constructs will place entire cDNA fragments in an antisense orientation with respect to a strong constitutive promoter like the cauliflower mosaic virus 35S promoter (CaMV 35S). This approach, though as effective as the newer silencing constructs, does suffer from lack of specificity. Genes that share significant levels of sequence-identity risk being silenced along with the target transcript. The latest techniques place target sequences of between 200 and 300 bp that are specific for a particular sequence (i.e. do not share more than a 21-bp stretch of sequence identity) in a head-to-head orientation with an intron between the sequences. This construct is then expressed in plants under control of a strong constitutive promoter that gives rise to a transcript with a hammerhead type structure. The intron is then cleaved leaving a doublestranded RNA that then triggers the silencing machinery. This 200–300 bp double-stranded RNA species is reduced to 21–25 bp fragments (that are diagnostic of the correct functioning of the process in transgenic plants), which then act on the target RNA. This technique alongside the related use of virus-induced gene silencing (VIGS) (see below) has been used to successfully reduce gene activity for a range of genes involved in flowering, resulting in phenotype changes that have been useful in identifying potential roles of the targeted genes (Kapoor et al., 2002; Keck et al., 2003; Sung et al., 2003; Byzova et al., 2004; Ehsan et al., 2004; Fornara et al., 2004; Liu et al., 2004). VIGS makes use of the same technology but rather than using Agrobacterium-mediated transformation to deliver the silencing construct, the constructs are maintained and propagated throughout the plant using viral vectors (Burch-Smith et al., 2004; Constantin et al., 2004; Liu et al., 2004). At the present time the majority of the work using this method has been carried out using a tobacco rattle virus-based vector system. Although this virus has proved useful for potato, tobacco and Arabidopsis, the wider utility of this approach for other crop plants remains to be seen. It is likely that new virus vectors will need to be constructed to further improve the utility of this method. One great advantage, however, is that a transformation method is not necessarily required as infection, resulting in systemic spread of the silencing vector throughout the plant, can be carried out utilizing leaf infiltration methods (Robertson, 2004).

Biotechnology of Floral Development

241

Spreading of silencing – methylation-induced silencing One aspect of PTGS/ VIGS that was relatively unexpected was ‘spreading’. This process targets non-transcribed regions of DNA acting to methylate DNA in a sequence-specific manner (Vaistij et al., 2002; Garcia-Perez et al., 2004). This has the potential to allow promoters or other important regulatory regions of genes to be targeted. This method offers the potential to target combinations of transcription factor-binding sites and thereby silence a swathe of genes without necessarily having to target individual transcription factors. It also allows for a greater range of gene space within which gene-specific targeting can be achieved. The most surprising aspect to the development of our understanding of this area is that plants themselves make use of this system to control developmentally important processes (Jones, 2002; Carrington and Ambros, 2003). Micro RNAs (miRNAs) are single-stranded molecules that are formed by the Dicermediated cleavage of endogenously produced stem-loop structures (Hunter and Poethig, 2003). Over 100 miRNAs have been discovered in Arabidopsis that are conserved in rice, maize and tobacco (Jones, 2002; Reinhart et al., 2002; Carrington and Ambros, 2003; Kidner and Martienssen, 2005). Additional work has identified up to another 280 miRNAs in Arabidopsis (Tang and Zamore, 2004). The targets of these miRNAs have not all been discovered but, of those that have, a disproportionate number appear to target transcriptional regulators involved with controlling aspects of plant development (Kidner and Martienssen, 2005). This in turn leads to the intriguing possibility that we might be able to use this information to control aspects of flowering (Aukerman and Sakai, 2003; Achard et al., 2004; Peragine et al., 2004).

Cell ablation For a number of applications, especially in the forestry context, complete prevention of flowering is a desirable trait (Martin-Trillo and Martinez-Zapater, 2002; Brunner and Nilsson, 2004). The complete prevention of flowering is likely to be a difficult task if alterations to single genes in the flowering regulatory pathway are targeted. This is due to the fact that most plants are obligate in their flowering response and that vegetative growth appears to be a function of repression of flowering rather than the promotion of vegetative growth (Sung et al., 2003). Alteration of the expression of genes involved with floral promotion is more likely to result in an alteration of time to flowering rather than preventing the flowering response completely. However, the use of technologies that are aimed at the destruction of floral cell types has been used to successfully repress flowering, or particular aspects of flowering, in model species (Lemmetyinen et al., 2001; Burgess et al., 2002). The genes, encoding the toxin of choice, are fused to a cell-specific promoter. When expressed the agents act to kill the specified tissues. This approach has been shown to be useful in the production of male and female sterile poplar (Campbell et al., 2003; Brunner and Nilsson, 2004), tobacco and Arabidopsis (Lemmetyinen et al.,

242

C. Winefield and B.R. Jordan

2001; Burgess et al., 2002; Guerineau et al., 2003). Three main toxins have found favour amongst both animal and plant researchers: (i) the BARNASE gene, encoding an RNase gene from Bacillus amyloliquefaciens (Lemmetyinen et al., 2001); (ii) a gene encoding the diphtheria toxin component A, which inactivates ribosomes (Guerineau et al., 2003); and (iii) the ricin A toxin that acts as a specific ribosome-inactivating protein (Sandvig et al., 2004). While targeting of the cytotoxic compound directly to a particular cell type has proved to be effective in certain cases, there remains the problem of obtaining complete control over the expression of the toxin in question. Many controlling elements in promoters are multifunctional so a large effort must be expended to ensure that a promoter is identified that allows expression of the toxin only in the desired tissues. Otherwise potential harmful side effects may result that might diminish the effectiveness of this type of approach. More recent advances have proposed the use of a two-hybrid system in which two parts of the toxin are expressed under the control of two promoters with overlapping expression. This then allows the expression of harmless precursors of the toxin and the formation of the complete ‘toxic unit’ where the promoter activities overlap. This is reported to ensure a greater degree of tissue specificity than can be afforded by a single promoter alone (Burgess et al., 2002). In addition to the expression of cytotoxic compounds, alternative approaches that impact on general metabolism have also been shown to be effective in disrupting specific cell types and thereby affecting developmental processes. One approach makes use of the fact that biotin is an essential cofactor for many enzymatic cellular processes (Ginzberg et al., 2004). Through overexpression of streptavidin (the gene isolated from Streptomyces avidinii), naturally occurring cellular biotin is irreversibly bound to the expressed streptavidin leading to severe pleiotropic developmental effects. Application of biotin was shown to reverse this effect (Ginzberg et al., 2004). Through the tissue-specific expression of the streptavidin gene, it should be possible to cause a similar developmental disruption as has been reported for cytotoxin expression.

Misexpression/overexpression In particular cases it may be useful to either ectopically express a particular gene or unlink the expression of a gene from its normal regulatory control pathway. This is most commonly achieved through the coupling of the gene of interest to a strongly expressed constitutive promoter such as CaMV35S. This promoter has been the workhorse for gene studies for many years. While it may be invaluable in describing the function of a particular gene, there are drawbacks for its utilization when expressing transcriptional factors with potentially multiple functions throughout the plant at all stages of its life cycle. In terms of floral biotechnology, a good example of misexpression of a component of the flowering regulatory network is ectopic overexpression of the LFY gene from Arabidopsis (Weigel et al., 1992). Expression of this gene in hybrid aspen resulted in plants with accelerated flowering (Coupland,

Biotechnology of Floral Development

243

1995; Brunner and Nilsson, 2004). While accelerated flowering was achieved, the floral structures that were produced were not necessarily fertile or perfectly formed, indicating that ectopically expressed LFY alone was only partly responsible for early flowering (Brunner and Nilsson, 2004). Another confounding factor has been that LFY-induced early flowering appears to be species dependent in that overexpression of LFY in crops other than hybrid aspen has not resulted in an early flowering phenotype (Brunner and Nilsson, 2004). It would appear that a constitutive high level of expression of transgenes may not necessarily be a constructive way to generate the desired phenotype. In a number of cases no-phenotype or undesirable traits may occur through the use of promoters such as CaMV 35S. A good example of this was attempts to alter lignin composition in transgenic lines of Arabidopsis through the introduction of the Arabidopsis ferulate-5-hydroxylase (F5H) under control of the CaMV 35S promoter. The aim of the experiments was to increase the amount of syringyl-lignin monomers and show that this gene was involved in the production of S-subunit lignin in Arabidopsis and tobacco (Meyer et al., 1998; Franke et al., 2000). Surprisingly, it was found that the CaMV 35S– driven constructs failed to produce increased levels of S-lignin in the expected tissues while constructs driven with the cinnamate-4-hydroxylase (C4H) promoter strongly affected lignin production in both species in the expected manner (Meyer et al., 1998; Franke et al., 2000). The C4H gene is involved in lignin and general phenylpropanoid precursor production and shares an expression pattern with genes, like F5H, that are involved with lignin production (Bell-Lelong et al., 1997; Mizutani et al., 1997). Overexpression of flowering genes, especially of the MADS-box class of regulators, also requires special consideration. There are large numbers of MADS-box genes in plants and it has been shown that these genes often act as either homodimers or heterodimers that differ in their DNA-binding affinities and activation capabilities (Honma and Goto, 2001; Ng and Yanofsky, 2001). This, in reality, makes the effects of heterologous expression of MADS-box genes from different species highly unpredictable. In addition, competition effects driven by the high level of expression of a particular gene may result in deleterious phenotypes (Martin-Trillo and Martinez-Zapater, 2002). It is likely that the use of heterologous promoters which have been designed specifically to give high levels of expression in a tissue- and temporally specific manner will be a better approach to deliver the control of flowering phenotype required. To further improve the specificity of expression, twocomponent systems may give opportunities to tune the desired effects. Inducible expression systems may also offer a route around some of the problems associated with the extensive use of the CaMV 35S promoter.

Gene Targets for Biotechnology in Flowering The vast majority of research carried out to characterize flowering has taken place in Arabidopsis. Research in other model crops (Antirrhinum, petunia, lucerne, rice, maize and poplar) supports the idea that there is a conservation

244

C. Winefield and B.R. Jordan

of gene sequence, if not gene function, across a very wide range of genera (Pineiro and Coupland, 1998; Hempel et al., 2000; Barnes, 2002; MartinTrillo and Martinez-Zapater, 2002; Zik and Irish, 2003; Brunner and Nilsson, 2004; Henderson and Dean, 2004; Pennazio, 2004). So although the majority of work done to elucidate the mechanisms controlling flowering has taken place in a small herbaceous annual rapid-cycling plant (Arabidopsis), much of the data generated may be directly applicable across a wide range of species, or at least forms a good starting point from which to determine target genes for manipulation in any given crop. In this section, therefore, we outline some of the target genes that have formed the focus of attempts at manipulation of flowering, mainly in Arabidopsis. Table 9.1 provides a summary of the gene targets discussed in this section and should be considered with details provided in other chapters of this book. The main challenge in determining which to target of the 80 plus genes involved in flowering is the fact that a number of key genes have a variety of roles in controlling multiple processes in the plant. A good example of this duality is the AGAMOUS (AG) gene from Arabidopsis that has an important role in specifying C whorl organ identity but has also been shown recently to play a significant role in the regional activation of sporocyte formation (Dinneny and Yanofsky, 2004). Careful analysis of the phenotypes observed in flowering also reveals that in certain plants manipulation of single genes alone may not be sufficient to obtain a commercially viable product. Ectopic expression of LFY in Arabidopsis leads to partial conversion of vegetative meristems to determinate floral inflorescences; however, conversion of the vegetative meristems to those that form complete and fertile flowers is often incomplete (Weigel et al., 1992). Heterologous expression of LFY in other species such as hybrid aspen, as we have already mentioned, proved to be an excellent example of the power of a biotechnological approach to manipulate flowering. This experiment demonstrated that CaMV 35S-driven expression of Arabidopsis LFY in hybrid aspen could accelerate flowering from several years to just over 1 year (Coupland, 1995). However, as described earlier, the overexpression of LFY in Arabidopsis and poplar results not only in the acceleration of flowering but also the malformation of floral organs suggesting that LFY expression alone in non-competent meristems is insufficient to trigger complete floral development and that additional unidentified factors are required (Martin-Trillo and Martinez-Zapater, 2002; Campbell et al., 2003; Brunner and Nilsson, 2004). One important feature of flowering that these experiments bring to the fore is the idea of the modular nature of plant development. These modules or phases of plant development can be divided into several distinct phases which show distinctive morphological characteristics such as changes in leaf morphology (Lawson and Poethig, 1995; Vega et al., 2002). The phase changes from juvenile–adult and adult–floral (Fig. 9.1A) represent two points at which manipulation of developmental regulators may offer sensible opportunities for the manipulation of flowering.

Proposed role

Species

Gene/mutant

Comment

Reference

Control of juvenile–adult phase change

Arabidopsis

ZIPPY (ZIP) SUPPRESSOR OF GENE SILENCING3 (SGS3) SUPPRESSOR OF GENE SILENCING2 (SGS2)/SILENCING DEFECTIVE1 (SDE1)/RNA-DEPENDENT POLYMERASEE6 (RDR6 ) PAUSED (PSD ) HASTY (HST ) gibberellin-deficient dwarf mutants an1, d1, d3 and d5 viviparous8 (vp8 ) teopod1, 2, 3 (tp1–3 ) corngrass (cg) PENNYWISE (PNY )/ POUNDFOOLISH (PNF ) CONSTANS (CO) FLOWERING LOCUS T (FT ) FLOWERING LOCUS C (FLC ) SUPPRESSOR OF OVEREXPRESSION OF CO1 (SOC1) HEADING DATE1 (HD1)

Argonaute-like gene Involvement in plant transgene silencing (PTGS) Involvement in plant transgene silencing (PTGS)

Hunter et al. (2003a) Peragine et al. (2004)

Exportin-t orthologue Putative miRNA export receptor Not cloned to date

Hunter et al. (2003b) Bollman et al. (2003) Vega et al. (2002)

Not cloned to date Not cloned to date Not cloned to date BEL-1 like homeobox genes pny:pnf double mutants fail to flower Accelerates flowering Floral pathway integrator Key repressor of flowering Floral pathway integrator

Evans and Poethig (1997)

Maize

Control of vegetative– reproductive phase change Signal perception including photoperiodism and vernalization

Arabidopsis Arabidopsis

Wheat

Homologous to CO

Peragine et al. (2004)

Biotechnology of Floral Development

Table 9.1. Targets for floral and reproductive manipulation.

Vega et al. (2002) Smith et al. (2004) Putterill et al. (1995) Kardailsky et al. (1999) Boss et al. (2004) Robson et al. (2001) Yano et al. (2000)

continued

245

246

Table 9.1. continued. Targets for floral and reproductive manipulation. Proposed role

Species

Gene/mutant

Comment

Reference

Inflorescence and floral organ architecture

Arabidopsis

TERMINAL FLOWER1 (TFL)

Homologous to CEN – converts SAM from indeterminate to determinant growth A-class floral identity gene B-class floral identity gene C-class floral identity gene Triple lfy:ap1:cal mutants abolish flowering Homologous to TFL1 Closely related to CYC Homologous to TB1 Homologous to CYC Not cloned to date radially symmetric to monosymmetric floral conversion Homologous to CYC Homologous to LFY Homologous to AP1 Homologous to Arabidopsis AG and SHP Homologous to LFY Homologous to AP1 Homologous to TFL1

Ohshima et al. (1997)

LEAFY (LFY) APETALA1 (AP1) AGAMOUS (AG) CAULIFLOWER (CAL) Antirrhinum

Tomato Sunflower Maize

Grape

CENTRORADIALIS (CEN) DICHOTOMA (DICH) CYCLOIDEA (CYC) SELF PRUNING (SP) chrysanthemoides (chry) teosinte branched1 (tb1) Zea FLO/LFY (ZFL1 and ZFL2) zap1 Vitis vinifera MADS1 (VvMADS1) Vitis vinifera LFY (VvLFY) Vitis vinifera AP1 (VvAP1) Vitis vinifera TFL1 (VvTFL1)

Weigel et al. (1992) Mandel et al. (1992) Yanofsky et al. (1990) Kempin et al. (1995) Bradley et al. (1996) Luo et al. (1996) Luo et al. (1996) Pnueli et al. (1998) Fambrini et al. (2003) Doebley et al. (1997) Bomblies et al. (2003) Mena et al. (1995) Boss et al. (2001) Carmona et al. (2002) Joly et al. (2004)

C. Winefield and B.R. Jordan

Biotechnology of Floral Development

247

Floral stimuli e.g. photoperiod, light quality, temperature, gibberellin

(A)

Juvenile

Reproductive

Adult HASTY, SGS3, SGS2/SDE1/RDR6 Phase change

Autonomous pathway

(B) Photoperiod Photoreceptors

FVE FPA

Light quality

Vernalization FRI

Phytochrome A FLC

VIP 3/4 ESD4

Circadian clock genes

EFS Gibberellin

CO

Floral pathway integrators FT AGL20/SOC1

AP1 CAL FUL LFY

Floral organogenesis

Fig. 9.1. Interacting pathways controlling flowering. (A) Genes shown to be involved in the control of phase changes in Arabidopsis. Environmental stimulation of flowering can only take place after the juvenile to vegetative phase change. In Arabidopsis this may represent a period of 1–2 weeks while in tree species this phase may last as long as 25–30 years. (B) Environmental and hormonal control of flowering in Arabidopsis, adapted from Boss et al. (2004).

Juvenile–adult vegetative phase change Plant development is characterized by successive ‘phase’ changes from germination through to reproduction. The period of development from germination to ‘adulthood’ in plants, commonly known as the juvenile phase, differs from plant to plant from a period of a few weeks for small herbaceous annual plants such as Arabidopsis, through to periods that may last longer than 20 years, as is evident for many tree crops (Poethig, 2002a, 2003; Brunner and Nilsson, 2004). Juvenility represents a phase of shoot maturation during which the plant is incapable of flowering and the vegetative meristem can be considered ‘blind’ to promotive signals (Martin-Trillo and Martinez-Zapater, 2002; Poethig, 2002b; Boss et al., 2004). Mutations in genes that are involved with phase change from the juvenile–adult phases generally present as early

248

C. Winefield and B.R. Jordan

flowering phenotypes indicating that flowering is actively repressed in vegetative meristems (Sung et al., 2003; Boss et al., 2004). Therefore, manipulation of genes shown to be involved in the control of the length of the juvenile phase would be extremely useful for long-lived perennial crops such as trees. In addition to the possibilities for reduction in the juvenile phase (and therefore acceleration of the onset of flowering), there may be options for greatly extending the juvenile phase. There are a number of features of juvenile vegetative development/growth that make the indefinite extension of the juvenile phase of development particularly attractive. Typically, the juvenile phase is characterized by significantly different vegetative growth patterns compared to the adult and reproductive phases of development. In Arabidopsis, as described below, juvenile vegetative development results in the production of distinct leaf phenotypes (Lawson and Poethig, 1995). In trees there are distinct changes to wood quality between juvenile and mature phases which would be attractive to industry (Martin-Trillo and Martinez-Zapater, 2002; Brunner and Nilsson, 2004). In Arabidopsis a number of genes that appear to be involved with the control of juvenile–adult phase change have been identified. Mutations in these genes generally result in a shortened time to flowering independent of environmental cues. Interestingly, a number of these genes appear to be involved in the machinery responsible for post-translational gene silencing (PTGS). Mutations in the genes SUPPRESSOR OF GENE SILENCING3 (SGS3) and SUPPRESSOR OF GENE SILENCING2 (SGS2)/SILENCING DEFECTIVE1 (SDE1)/RNA-DEPENDANT POLYMERASE6 (RDR6) produce similar phenotypes to the Argonaute gene ZIPPY (ZIP) (Hunter et al., 2003a; Peragine et al., 2004). These phenotypes are typically an acceleration of the production of leaves with an adult character. In Arabidopsis this phase change manifests itself as a switch from the formation of small, round leaves with trichomes restricted to the adaxial leaf surface to adult leaves that are typified by an elongated leaf shape with a serrate edge and trichomes on both the abaxial and adaxial leaf surfaces (Hunter et al., 2003a). While ZIP appears to act independently of environmental cues, both SGS3 and SGS2/RDR6 appear to be able to shorten time to flowering as well as the transition between juvenile and adult growth phases. Epistasis analysis suggests that ZIP, SGS3, SGS2/RDR6 and the putative miRNA export receptor HASTY (HST) operate within the same pathway (Peragine et al., 2004). These data strongly suggest that PTGS plays a major role in the temporal control of shoot development in Arabidopsis. Mutations that affect the juvenile–adult phase transition have been identified in a number of other plant species such as maize, pea and ivy (Wiltshire et al., 1994; Lawson and Poethig, 1995; Poethig, 1997; Vega et al., 2002). In maize these mutations are either defined as recessive, loss-of-function type alleles and include the mutants gibberellin-deficient dwarf (an1, d1, d3 and d5) and viviparous8 (vp8), or as genes that have been defined as primarily dominant gain-of-function mutations and promote juvenile development and include teopod1 (tp1), teopod2 (tp2), teopod3 (tp3) and corngrass (cg) (Poethig, 1988; Lawson and Poethig, 1995). In pea, nine mutations have

Biotechnology of Floral Development

249

been identified that can act to perturb the juvenile–adult phase change (Wiltshire et al., 1994). However, for maize, pea and ivy the identity of the genetic loci has not been characterized, so comparisons to the identified mechanisms in Arabidopsis are not possible at this time. Given the involvement of PTGS machinery in the control of this process in Arabidopsis and that PTGS has been reported for a very wide range of plant species, it is highly likely that similar mechanisms for the control of vegetative phase change exist. The exact identity of the components involved may differ forcing a case-by-case study of the target organism to better understand the exact nature of the controlling mechanisms.

Vegetative to reproductive phase change The recently characterized BEL-1 like homeobox class gene regulators, PENNYWISE (PNY) and POUNDFOOLISH (PNF), when combined as double mutants result in Arabidopsis plants whose meristems remain capable of receiving floral stimuli but are unable to complete the floral transition (Smith et al., 2004). This situation is unusual in that the double-mutant plants do not form flowers at all. Generally, the other mutations discussed previously will result in some flowers being formed late in development (Komeda, 2004). In pnf:pny plants shoot meristems only initiated leaves after the time at which wild-type Arabidopsis plants had flowered. These leaves, though smaller than vegetative rosette leaves, none the less retained rosette leaf characteristics. Taken together, these data suggest that the double mutants are acting to disrupt the normal transition from vegetative to reproductive development. It is unclear at this stage what effect ectopic overexpression of these two homeobox genes might have on Arabidopsis or other plants. From genetic analysis of the single mutants in various backgrounds it is apparent that the roles of both PNY and PNF appear to be in the regulation of the allocation of meristematic cells into the developing primordium. Given the strongly dose-dependent effects of either gene on inflorescence development, ectopic overexpression of these genes using promoters such as CaMV 35S may not yield usable plants, alterations to flowering aside. The use of other promoter sequences, in combination with these genes that offer similar spatial expression patterns but with differing levels of expression and altered temporal regulation, may offer a better way to utilize the dose-dependent nature of these genes to allow controlled manipulation of flowering in Arabidopsis and other crops. From the extensive body of data describing the control of flowering in Arabidopsis and other model species it is clear that control of meristem identity is a key factor for any attempts at regulating flowering (see Jordan and Anthony, 1997, for a review of molecular aspects of floral evocation). We have already dealt with genes that appear to be important in the conversion of vegetative meristems from a juvenile non-responding meristem to a florally competent meristem. At this point those genes that direct the conversion of the vegetative meristem to an inflorescence or floral meristem can then act and offer opportunities to modify time to flowering. Mutants in these genes can act

250

C. Winefield and B.R. Jordan

to delay or accelerate flowering under promotive or non-promotive environmental conditions. The flowering response in Arabidopsis can be considered in four main parts (all of which tend to be interconnected to give an integrated response to the environmental and developmental status): (i) the autonomous pathway; (ii) the vernalization pathway; (iii) the gibberellin (GA) pathway; and (iv) the light-dependent pathway (see Fig. 9.1B and Chapters 1, 2 and 3, this volume). Inputs from each of these four pathways then act coordinately to finally promote the conversion of a competent vegetative meristem to an inflorescence meristem (Boss et al., 2004). Central to the integration of these pathways is the floral repressor FLOWERING LOCUS C (FLC). This regulator has been shown to repress floral pathway integrator genes and floral meristem identity genes such as LFY and APETALA1 (AP1). The signal cascades upstream of FLC act to lower the amount of FLC mRNA present in the meristem thereby allowing the transition to floral development. Null mutations in FLC lead to very early flowering phenotypes (Andersen et al., 2004). This gene is an obvious candidate for manipulation in crop species, either through upregulation to inhibit flowering or by downregulation to accelerate flowering. However, to date, although homologous genes have been identified in other species, it is evident that these homologues may be functioning in a different manner to that observed in Arabidopsis (Pineiro and Coupland, 1998). A common theme that appears to be developing is that although the machinery of flowering regulation is conserved between species and genera, the exact mechanism that is adopted by a particular species may differ quite widely requiring a specific analysis of the processes in each species before a rational approach to regulation of flowering can be applied. While FLC represents a convergence point in the regulation of flowering in response to developmental and temperature stimuli, light perception in terms of daylength represents an alternative point of control. The gene CONSTANS (CO) has been shown in Arabidopsis to be central to the regulation of flowering in response to daylength, with mutants resulting in a late-flowering phenotype (Putterill et al., 1995; Robson et al., 2001). Ectopic overexpression of this gene results in early flowering phenotypes in Arabidopsis (Boss et al., 2004). CO acts to regulate the levels of expression of FLOWERING LOCUS T (FT ) and SUPPRESSOR OF OVEREXPRESSION OF CO1 (SOC1), which together form part of the floral integration pathway (Robson et al., 2001). In Arabidopsis, CO belongs to a gene family consisting of 17 putative members that is divided into three broad groupings based on the conservation of gene structure surrounding the B-box motif and zinc-finger domains characteristic of this group of transcription factors (Robert et al., 1998; Liu et al., 2001). COlike genes have been identified in several dicot species and in two cases, Brassica napus and Pharbitus nil, the isolated gene sequences have been expressed in Arabidopsis co mutant backgrounds to show functional conservation (Yano et al., 2000). In rice the Heading date1 (Hd1) gene has been shown to be homologous to CO and possesses a role in photoperiodic responses in this plant (Yano et al., 2000; Griffiths et al., 2003). Similar genes have been isolated from hexaploid wheat and have been shown to be able to compliment the hd1 mutation in rice (Fambrini et al., 2003). These data

Biotechnology of Floral Development

251

suggest that manipulation of CO should be a viable method for alteration of photoperiodic control of flowering in many plant species. It is likely that the manipulation of the autonomous, vernalization and photoperiod pathway gene members will result not only in the alteration of flowering time but also as one moves backward up the regulatory pathways there will be other associated pleiotropic effects to plant development. Picking key genes that form distinct bottlenecks in the regulatory pathways will be essential to successful manipulation of flowering without significant alteration to other aspects of plant growth and development. Manipulation of meristem identity directly through the alteration of expression of key members of the floral pathway integrators such as FT and AGAMOUS Like-20 (AGL20)/SOC1, as well as key meristem identity genes such as LFY and AP1 in Arabidopsis, has been shown to have some utility (Mandel et al., 1992; Weigel et al., 1992; Kardailsky et al., 1999; Moon et al., 2003; Murai et al., 2003; Komeda, 2004). However, practically all of the experiments published to date attempt to alter the expression of these genes and have resulted in downstream pleiotropic effects to floral and reproductive organ development. While this may not necessarily be unsurprising, it clearly indicates that indirect manipulation of meristematic fate may be the most productive avenue for the manipulation of flowering in response to the environmental and developmental cues.

Inflorescence and floral organ architecture Other targets for agri-biotechnology approaches are control of inflorescence and floral organ architecture. These two aspects of floral development are the drivers behind the vast diversity in floral form observed in the angiosperms. The genetic regulation of both of these processes is based around the control of meristem identity and cell fate. Inflorescence architecture is essentially controlled through the determinacy of the inflorescence meristems derived from the shoot apical meristem (SAM), while floral structure involves the fate of cell lineages formed on the flanks of the floral meristem (Reinhardt and Kuhlemeier, 2002). Typically, plant growth can be characterized as two modes of development: (i) monopodial where the SAM produces leaves and then flowers on its flanks representing an indeterminate form of meristem development; and (ii) sympodial where the meristem exhibits a determinate form of growth terminating in floral structures and vegetative growth is continued from axillary meristems (Reinhardt and Kuhlemeier, 2002). The genetic control of meristem determinancy can have a major impact on plant and floral architecture. Mutants such as CENTRORADIALIS (CEN ) from Antirrhinum and TERMINAL FLOWER (TFL) from Arabidopsis convert the normally indeterminate SAM of these plants to a determinate meristem resulting in the premature termination of the SAM in a flower (Bradley et al., 1996, 1997). Tomato represents the only determinate plant in which the role of CEN/ TFL orthologues has been extensively studied. The SELF PRUNING (SP) gene represents the tomato homologue of CEN/ TFL (Pnueli et al., 1998; Carmel-Goren et al., 2003).

252

C. Winefield and B.R. Jordan

The sp mutant results in a gradual decrease in the vegetative nodes produced within each sympodial unit in tomato from three, to two, to one, to eventually the point where the vegetative phase is completely bypassed (Pnueli et al., 1998). In the context of the sympodial unit, SP therefore parallels the function of CEN/ TFL. Interestingly, unlike cen /tfl, sp does not appear to affect floral development. It is thought that this difference does not necessarily reflect a difference in SP function compared to CEN/ TFL but rather differences in the shoot models between Antirrhinum and Arabidopsis compared to tomato. In other words, vegetative and reproductive monopodial shoots with an indeterminate apical meristem compared with a sympodial vegetative and reproductive shoots and a determinate apical meristem, respectively (Pnueli et al., 1998). Furthermore, overexpression of CEN in the determinate plant tobacco results in a dramatic extension of the indeterminate (vegetative) phase of growth (Amaya et al., 1999). In most transgenic lines of tobacco expressing CEN, flowering will eventually occur. However, in transgenic lines with particularly high gene expression levels of CEN there was an indication that sufficient levels of CEN expression might lead to the indefinite suspension of flowering (Amaya et al., 1999). The expression patterns of the nearest tobacco homologues of CEN, CET2 and CET4, differ quite markedly to SP in tomato with the CET2 and CET4 being found to be expressed only in axillary meristems compared to SP which is expressed in all meristems (Pnueli et al., 1998; Amaya et al., 1999). This point reinforces the fact that an in-depth study of the system one wishes to manipulate is essential prior to embarking on attempts to manipulate plant architecture. Even with this in mind given the conserved functional role that CEN/ TFL-like genes have in regulation of meristem determinacy, orthologues of these genes may offer ways to manipulate flowering in a wide range of crop species depending on the growth habit (monopodial or sympodial) of the target species. While CEN/ TFL genes act to prevent the formation of flowers, the genes LFY and AP1 act antagonistically on CEN/ TFL-like genes to promote the formation of flowers (Amaya et al., 1999). In Arabidopsis a further gene CAULIFLOWER (CAL) acts in concert with LFY and AP1 to give a redundant regulation of meristem identity in this plant (Ferra´ndiz et al., 2000). As we have already mentioned, mutations in these genes leads from partial to full conversion of floral structures to leaves or leaf-like structures. The phenotypes of the individual genes are enhanced in the double (lfy:ap1) and triple (lfy:ap1:cal) mutants to a point where flowering can be completely abolished (Yanofsky, 1995; Ferra´ndiz et al., 2000). However, combinations of double mutants in Arabidopsis have been shown to give rise to commonly recognized mutant floral phenotypes such as seen in Brassica oleracea var. botrytis, the cauliflower (Anthony et al., 1995, 1996; Kempin et al., 1995). Interestingly, many edible Brassica species (cauliflower, broccoli, etc.) are arrested stages of floral development. Commercial loss takes place when the control of this state is lost leading to ‘bolting’ of the inflorescence. Molecular regulation of floral transition would seem a very appropriate target for manipulation. As for CEN/ TFL, it is evident that the basic mechanisms for floral patterning involving MADS-box transcriptional regulators are highly conserved; however, the ways in which

Biotechnology of Floral Development

253

individual plant species make use of these regulators can differ quite markedly thereby giving rise to the plethora of floral forms evident in nature. Indeed, quite simple mutations in this regulatory network can lead to striking changes in floral subunit formation. For instance, a recent report of a mutation in sunflower (Helianthus annuus) termed chrysanthemoides (chry) described the transformation of the normally radially symmetric corolla of tubular disc flowers into a monosymmetric ligulate-like corolla. This results in a pom-pom like floral structure. The actual molecular basis for this transformation is still being investigated, but is thought to be due to a single semi-dominant gene (Fambrini et al., 2003). The grasses include some of the most important food crops in the world and the productivity of these crops is determined by their inflorescence structures (Bommert et al., 2005). Grass inflorescences possess a number of unique floral structures in comparison to dicot species, such as the spiklet. Maize has been the most extensively studied of the grasses due in part to the large number of identified mutants that affect the development of the floral structures in this plant. Additionally, maize represents one of the most stunning transformations of a wild plant species into a high-productivity modern crop. The development of the modern maize plant has been a story of selective breeding for the conversion of the comparatively sparse reproductive structures of the proposed ancestral form of maize (teosinte) to the instantly recognizable ear structure on modern maize (Zea mays). This change in flowering habit has been attributed to four quality trait loci (QTL) that have been selected for during the domestication process (Burnham, 1961; Doebley et al., 1997; Doebley, 2004). Of the four identified QTLs, one was shown by complementation testing to correspond to the teosinte branched1 (tb1) locus. This locus has been shown to be the major contributor to the transformation of teosinte architecture to that of maize. Indeed, the effects of TB1 alone are sufficient that when inserted into teosinte it is capable of a complete transformation of this plant to a maize architecture (Doebley et al., 1997; Doebley, 2004). The TB1 gene shares significant homology with the CYCLOIDEA (CYC) gene from Antirrhinum and appears to affect similar aspects of development. Investigations into the genetic basis for Antirrhinum mutants that led to a switch in floral symmetry from bilateral to partially or fully ventralized flowers led to the identification of CYC and a closely related gene DICHOTOMA (DICH) (Luo et al., 1996). Both of these genes are part of the highly conserved TCP family of transcriptional activators that have been shown to have roles in the control of growth and development such as floral symmetry, axillary shoot outgrowth and leaf development (Doebley et al., 1997; Cubas et al., 1999; Cubas, 2004; Doebley, 2004). Genetic and molecular control of grass inflorescence and floral development has been most extensively studied in maize. This plant has an extensive collection of mutants that affect the development of floral structures, some of which produce some very startling phenotypes. Like the control of maize architecture, many of these genes share significant levels of sequence identity as well as functionality with homologous dicot genes (Bommert et al., 2005). The maize homologues of FLO/LFY and SQUAMOSA (SQUA)/AP1, ZEA FLO/LFY (ZFL1), ZFL2 and ZAP1 represent good examples of the sequence

254

C. Winefield and B.R. Jordan

conservation between the dicot and monocot flowering genes (Schmidt et al., 1993; Mena et al., 1995; Schmidt and Ambrose, 1998; Bommert et al., 2005). Much of what we have dealt with to date are candidate genes that might be amenable to agri-biotechnology approaches aimed at manipulation of flowering. In the following sections we look in more detail at approaches that have been and are being adopted to manipulate flowering in woody species, specifically grapevine, kiwifruit vines and tree species. Each of these case studies represents situations that draw heavily on what is currently known from the annual rapid-cycling model systems such as Arabidopsis. Although, as we have commented, this information is invaluable and there appear to be high levels of sequence and functional conservation between species, the different and extended life cycles of woody perennial plants offer a number of interesting challenges.

Floral development and biotechnology in woody perennial species In the following sections a brief overview of molecular studies in woody perennials is described. Readers, however, are recommended to refer to more comprehensive reviews as cited below. Grape and kiwifruit For a comprehensive review of floral development in grapes see Shavrukov et al. (2004), and for a recent molecular account see Boss et al. (2003) and Joly et al. (2004). Like many woody species, grape requires an extended period of juvenile growth before the vine is competent to enter reproductive development. Unlike Arabidopsis where flowering commences after the perception of a suitable flowering stimulus, grapevines produce both vegetative and reproductive meristematic structures on the same shoot (Boss and Thomas, 2002). During shoot development the SAM produces a regular pattern of both leaf primordia and small meristematic protuberances termed uncommitted primordium (Boss et al., 2003). If the uncommitted primordium forms within latent buds, they can develop into inflorescence structures. However, if they form on rapidly developing shoots they are more likely to form tendrils (Boss et al., 2003). First-order buds generally burst the year they are produced. Higher-order buds (latent buds) are formed in the axil of modified leaves formed on the lateral shoots derived from the first-order buds. In these latent buds, the uncommitted primordia form after 3–8 leaf primordia and subsequently opposite every two of three leaf primordia. Dependent on the cultivar and environmental conditions, the first 1–3 uncommitted primordia formed on shoots in latent buds will undergo repeated branching and will develop into an immature inflorescence before the buds enter winter dormancy (Boss et al., 2003). In the following season these buds burst and the immature inflorescences continue to develop and form flowers in a pyramidal branched inflorescence pattern (Boss and Thomas, 2000; Boss et al., 2003; Calonje et al., 2004). Cultivated grapevine

Biotechnology of Floral Development

255

is unusual in that it produces hermaphrodite flowers. Most members of the Vitis genus are dioecious and the hermaphroditic flowers of cultivated species may have been selected for at a very early stage in the domestication of grapevine. Grapevine tendrils are not modified leaves as are pea tendrils, but rather are considered analogous to inflorescence structures. Several lines of evidence including the coexpression of grapevine homologues of LFY and AP1 in these tissues strongly suggest that grapevine tendrils are derived from uncommitted primordia and are closely related to floral structures (Boss et al., 2001, 2002; Carmona et al., 2002; Joly et al., 2004). These differences in growth and developmental habit present researchers with some very interesting questions about flowering and, as has been seen for tree species (see below), it is likely that some of the mechanisms that control flowering in model crops such as Arabidopsis will differ in grapevine. What is interesting is that there appears to be a high level of conservation of gene sequence identity between Arabidopsis and grape. Recent reports have identified grape homologues of LFY, AP1, AG, TF1 and SEPALLATA3 (SEP3) and expression of these orthologues throughout grape inflorescence development (Boss et al., 2002). Additionally, a range of MADS box-related genes has been isolated from grape, by homology screening a young inflorescence library, that share identity with SEP and AG-like genes 6, 11 and 13 from Arabidopsis (Calonje et al., 2004). The cloning and identification of two genes sharing sequence identity with Arabidopsis FRUITFULL and AP1 and their involvement in tendril development has also been reported (Boss et al., 2001). In each case the genes have been sequenced and compared to reported sequences. Analysis of the expression patterns of each of the genes supports the view that they have a role in grape flowering. However, given the difficulties in producing stable transgenics in grape and the long generational time that would be required to analyse floral phenotypes arising from transgenic approaches, definitive descriptions of the function of these genes in the flowering process are still to be demonstrated. One approach that has been utilized to circumvent these issues has been the heterologous expression of isolated genes in model systems such as tobacco (Boss et al., 2001). The Vitis vinifera MADS1 (VvMADS1) gene shares significant sequence identity with AG and SHATTERPROOF from Arabidopsis. Expression of VvMADS1 in tobacco, under control of CaMV 35S, results in alterations to the outer two floral whorls. In some transgenic lines these alterations result in the formation of carpelloid sepals and stamenoid petals. In the most extreme examples the corolla, which is usually fused, was split or converted into stamen-like structures. These alterations are very reminiscent of plants ectopically expressing AG orthologues throughout the four floral whorls (Boss and Thomas, 2002). Recently, Chardonnay grapes that overexpress VvMADS1 have been generated (Boss et al., 2003). Although a similar floral phenotype to the tobacco 35S::VvMADS1 was produced, some differences were found. For instance, in some plants an extended calyx/sepal structure encased the whole cap morphology produced by the petals (Fig. 9.2). It is clear that VvMADS1 is likely to be a ‘C’ class gene; however, it is presently not possible

256

C. Winefield and B.R. Jordan

Fig. 9.2. (A) Expression analysis of VvMADS1 in different floral organs of Cabernet Sauvignon. (B) A wild-type Chardonnay flower just prior to anthesis. (C) A flower from a transgenic Chardonnay plant overexpressing VvMADS1. Bar ¼ 200 mm. Reproduced with permission from Functional Plant Biology 30(6), 593–606 (Boss et al., 2003). Copyright CSIRO 2003. Published by CSIRO Publishing, Melbourne, Australia.

to determine if it is an orthologue of AG or SHP1/2 from Arabidopsis (Boss et al., 2003). The relatedness between vine architecture and floral development combined with the apparent conservation of floral regulator sequence identity and expression pattern suggests fertile ground for agri-biotechnology approaches to assist in vine and grape improvement. The recent report of a mutation in the grape homologue of the Arabidopsis GAI gene indicates the important involvement of phytohormones, in particular GA, in the control of floral induction in this species (Coupland, 1995; Rottmann et al., 2000; Honma and Goto, 2001; Bomblies et al., 2003; Murai et al., 2003). This is consistent with previous physiological studies (Srinivasan and Mullins, 1980). Another woody perennial, like grape, is the kiwifruit (Actinidia deliciosa). Recently, studies on kiwifruit have suggested that LFY and AP1 homologues (ALF and AAP1, respectively) play a role in kiwifruit floral development (Walton et al., 2001). In kiwifruit the flowering process takes place over two growing seasons, with ‘evocation’ considered to take place at an early stage within the first year. Both ALF and AAP1 show expression in both years and the results clearly suggest a role in the first year that is indicative of a commitment to flowering (evocation). This expression pattern over 2 years is similar to that found in grape (Joly et al., 2004). Thus by the use of these floral gene biomarkers understanding of a very complex developmental process in grape and kiwifruit is being elucidated. While many interesting targets for agri-biotechnology approaches aimed at the alteration of flowering responses are available, problems associated with the long generational time to flowering and comparatively poor rates of transformation currently hinder the advance of these approaches. Investigating the roles of the identified regulators such as VvLFY and VvAG is also vitally important for understanding the actual roles of these genes in systems other than rapid-cycling annuals. Forestry biotechnology There are a number of very important reasons why controlling flowering in tree species is a high priority: (i) to increase the reproductive cycle; (ii) to reduce vegetative losses from reproduction; and (iii) to control pollen.

Biotechnology of Floral Development

257

Conventional breeding approaches have had an enormous impact on agricultural and horticultural crops. These techniques have been refined over many centuries and have improved crop productivity and quality enormously. In contrast, the domestication of forestry production has only taken place over the last century. Two major problems relating to tree biology have prevented substantive progress in breeding elite lines. The most obvious of these problems is the long time before maturation (Greenwood, 1995). Trees pass through stages of juvenility and it may be many years before they are reproductively mature and hence assessment of breeding characteristics within the breeding programmes is very slow. By way of example the most extensive breeding programmes currently underway for tree species have only just reached their fourth generation in over 100 years of effort! If the breeding cycle was reduced from 14 to 2 years, selection efficiency would increase several hundred percent and consequently impact on economic and environmental aspects of forestry (Pena et al, 2001). The extensive delay in flowering, of over 20 years in some cases, has proved to be a great hindrance, not only in terms of studying flowering processes at a molecular level but also in terms of domestication of tree species in general. Another difficulty is that many commercially important tree crops outcross preferentially and are subject to inbred suppression. This creates problems to establish desirable characteristics through conventional breeding approaches such as selfing and backcrossing (Williams and Savolainen, 1996). Thus, undoubtedly, the most pressing issue for forest biotechnology is the extensive juvenile phase of development during which flowering will not occur. To this end most of the research to date has been directed at understanding the genetic regulators that are involved in phase change in trees. Obvious candidates such as LFY and AP1 have been targeted due to their demonstrable ability to promote flowering in model systems and in certain tree species (Coupland, 1995; Rottmann et al., 2000). While many of the MADS-box and LFY-like orthologues cloned from various tree species have exhibited high degrees of functional and sequence conservation, the ability to consistently shorten the time to flowering in tree species has proved to be elusive (Rottmann et al., 2000). In the case of LFY homologues, ectopic overexpression of either the Arabidopsis LFY or tree-derived sequences has only promoted flowering reliably in hybrid aspen or in lines of transgenic poplar where LFY is expressed at very high levels (Martin-Trillo and Martinez-Zapater, 2002). This indicates that there are most likely other factors governing the switch from vegetative to reproductive development and LFY homologues are therefore insufficient alone to promote the complete switch from a vegetative to a floral meristem. Other genes from Arabidopsis that have been shown to have some involvement with promotion of flowering such as CO and FCA have also been expressed in hybrid aspen without effect (Pena et al., 2001). The high degree of species specificity for control of flowering is highlighted by the expression of LFY and AP1 in root stocks of citrus. In this species an acceleration of flowering from 6–7 years to 1 year was obtained through overexpression of these genes (Pena et al., 2001). Interestingly, the expression of either of these genes did not convert all meristems to floral meristems and the transgenic trees

258

C. Winefield and B.R. Jordan

were still able to respond to environmental cues that induce flowering each year (Telfer and Poethig, 1998; Hunter et al., 2003b). From the work described above and the ever growing number of reports of genes isolated from gymnosperms that show conservation of function when expressed in Arabidopsis, it is apparent that genes that show involvement with phase change in Arabidopsis and other model systems such as maize may represent other candidates for investigation of phase change in tree species. These include HASTY (Telfer and Poethig, 1998), PAUSED (Hunter et al., 2003b), SERRATE (Clarke et al., 1999) and SQUINT (Berardini et al., 2001). Mutations in these genes act to reduce the juvenile phase in Arabidopsis while not having major effects on other developmental processes. In order to effectively modulate flowering through molecular approaches, an in-depth understanding of the regulation of flowering in the target tree species is required. This therefore increases the importance of experiments aimed at acceleration of flowering time in tree species; until this is achieved it will be difficult if not impossible to characterize the genetic pathways controlling flowering in the diverse range of tree species that are currently grown on a commercial basis. Another major problem of tree reproduction is that pollen production leads to enormous loss of vegetative growth potential. Assimilation of nitrogen, etc. into the pollen is a net loss to vegetative growth and so inhibition of flowering and hence pollen production would be a substantial economic advantage. The other advantage to preventing pollen production would be to reduce public concern over forest biotechnology. If long-distance gene flow could be prevented through inhibiting pollen production, many important genetic traits could be introduced rapidly through modern molecular genetics without concern over their dispersal into native forest populations (Skinner et al., 2000; Strauss et al., 1995). Like grapevine and kiwifruit, tree species represent a different situation to the ‘classic’ flowering models of Arabidopsis, Antirrhinum, etc. Progress is, however, being made and the new developments in molecular technology will undoubtedly bring greater understanding and commercial success. As a final note, a very recent report in Nature describes a new system for genome-wide non-Mendelian inheritance of extra-genomic information in Arabidopsis (Lolle et al., 2005). This new phenomena revolves around the inheritance of allelic-specific DNA sequence information that is not contained within the chromosomal genome of either parent. This report describes the reversion of the homozygous HOTHEAD (hth) mutant to wild-type using allelic-specific information that is absent from either parent but has been present in previous generations. The thorough characterization of this phenomenon rules out trivial explanations such as increased mutation rates and gene conversion with closely related gene sequences. In order to determine whether the sequence changes observed for the hth mutant occur elsewhere in the genome, a number of hybrid hth plants were produced which were scored for changes to molecular markers that have been shown to be polymorphic between the Colombia and Landsberg erecta parents. In the progeny of the hth / hth F2 plants a high rate of sequence change in both directions (Col to Ler and Ler to

Biotechnology of Floral Development

259

Col) was observed. The authors postulate that the template for these reversions is a cache of previously undescribed RNAs that are propagated through many generations. In terms of biotechnological manipulation of flowering, this phenomenon should be viewed both in terms of caution (i.e. the plant may well be able to revert to wild-type from transgenic/mutant status at relatively high frequency) and in terms of potential opportunity. It might be possible that this system may be used to introduce sequence-specific lesions into genomes in a targeted fashion at higher frequencies than is currently possible in higher plant species. The unravelling of the underlying mechanism for this phenomenon may add yet another powerful tool to our repertoire for the targeted manipulation of flowering.

References Achard, P., Herr, A., Baulcombe, D.C. and Harberd, N.P. (2004) Modulation of floral development by a gibberellinregulated microRNA. Development 131, 3357–3365. Amaya, I., Ratcliffe, O.J. and Bradley, D.J. (1999) Expression of CENTRORADIALIS (CEN) and CEN-like genes in tobacco reveals a conserved mechanism controlling phase change in diverse species. Plant Cell 11, 1405–1418. Andersen, C.H., Jensen, C.S. and Petersen, K. (2004) Similar genetic switch systems might integrate the floral inductive pathways in dicots and monocots. Trends in Plant Science 9, 105–107. Anthony, R.G., James, P.E. and Jordan, B.R. (1995) The cDNA sequence of a cauliflower apetala-1/squamosa homologue. Plant Physiology 108, 441– 442. Anthony, R.G., James, P.E. and Jordan, B.R. (1996) Cauliflower (Brassica oleracea L. var. botrytis) curd development: expression of meristem identity genes. Journal of Experimental Botany 47, 181–188. Aukerman, M.J. and Sakai, H. (2003) Regulation of flowering time and floral organ identity by a microRNA and its APETALA2-like target genes. Plant Cell 15, 2730–2741.

Barnes, S. (2002) Comparing Arabidopsis to other flowering plants. Current Opinion in Plant Biology 5, 128– 133. Bell-Lelong, D.A., Cusumano, J.C., Meyer, K. and Chapple, C. (1997) Cinnamate-4-hydroxylase expression in Arabidopsis. Regulation in response to development and the environment. Plant Physiology 113, 729–738. Berardini, T.Z., Bollman, K., Sun, H. and Poethig, R.S. (2001) Regulation of vegetative phase change in Arabidopsis thaliana by cyclophilin 40. Science 291, 2405–2407. Bollman, K.M., Aukerman, M.J., Park, M.Y., Hunter, C., Berardini, T.Z. and Poethig, R.S. (2003) HASTY, the Arabidopsis ortholog of exportin 5/ MSN5, regulates phase change and morphogenesis. Development 130, 1493–1504. Bomblies, K., Wang, R.L., Ambrose, B.A., Schmidt, R.J., Meeley, R.B. and Doebley, J. (2003) Duplicate FLORICAULA / LEAFY homologs zfl1 and zfl2 control inflorescence architecture and flower patterning in maize. Development 130, 2385–2395. Bommert, P., Satoh-Nagasawa, N., Jackson, D. and Hirano, H-Y. (2005) Genetics and evolution of inflorescence and flower development in grasses. Plant Cell Physiology 46, 69–78.

260

C. Winefield and B.R. Jordan Boss, P.K. and Thomas, M.R. (2000) Tendrils, inflorescences and fruitfulness: a molecular perspective. Australian Journal of Grape and Wine Research 6, 168–174. Boss, P.K. and Thomas, M.R. (2002) Association of dwarfism and floral induction with a grape ‘green revolution’ mutation. Nature 416, 847–850. Boss, P.K., Vivier, M., Matsumoto, S., Dry, I.B. and Thomas, M.R. (2001) A cDNA from grapevine (Vitis vinifera L.), which shows homology to AGAMOUS and SHATTERPROOF, is not only expressed in flowers but also throughout berry development. Plant Molecular Biology 45, 541–553. Boss, P.K., Sensi, E., Hua, C., Davies, C. and Thomas, M.R. (2002) Cloning and characterisation of grapevine (Vitis vinifera L.) MADS-box genes expressed during inflorescence and berry development. Plant Science 162, 887–895. Boss, P.K., Buckeridge, E.J., Poole, A. and Thomas, M.R. (2003) New insights into grapevine flowering. Functional Plant Biology 30, 593–606. Boss, P.K., Bastow, R.M., Mylne, J.S. and Dean, C. (2004) Multiple pathways in the decision to flower: enabling, promoting, and resetting. Plant Cell 16, S18–S31. Bradley, D., Carpenter, R., Copsey, L., Vincent, C., Rothstein, S. and Coen, E. (1996) Control of inflorescence architecture in Antirrhinum. Nature 379, 791–797. Bradley, D., Ratcliffe, O., Vincent, C., Carpenter, R. and Coen, E. (1997) Inflorescence commitment and architecture in Arabidopsis. Science 275, 80–83. Brunner, A.M. and Nilsson, O. (2004) Revisiting tree maturation and floral initiation in the poplar functional genomics era. New Phytologist 164, 43–51. Burch-Smith, T.M., Anderson, J.C., Martin, G.B. and Dinesh-Kumar, S.P. (2004) Applications and advantages of

virus-induced gene silencing for gene function studies in plants. Plant Journal 39, 734–746. Burgess, D.G., Ralston, E.J., Hanson, W.G., Heckert, M., Ho, M., Jenq, T., Palys, J.M., Tang, K.L. and Gutterson, N. (2002) A novel, two-component system for cell lethality and its use in engineering nuclear male-sterility in plants. Plant Journal 31, 113–125. Burnham, C.R. (1961) Linkage relations of teosinte branched. Maize Genetics Cooperation News Letter 35, 87. Buzgo, M., Soltis, D.E., Soltis, P.S. and Ma, H. (2004) Towards a comprehensive integration of morphological and genetic studies of floral development. Trends in Plant Science 9, 164–173. Byzova, M., Verduyn, C., De Brouwer, D. and De Block, M. (2004) Transforming petals into sepaloid organs in Arabidopsis and oilseed rape: implementation of the hairpin RNA-mediated gene silencing technology in an organ-specific manner. Planta 218, 379–387. Calonje, M., Cubas, P., Martinez-Zapater, J.M. and Carmona, M.J. (2004) Floral meristem identity genes are expressed during tendril development in grapevine. Plant Physiology 135, 1491– 1501. Campbell, M.M., Brunner, A.M., Jones, H.M. and Strauss, S.H. (2003) Forestry’s fertile crescent: the application of biotechnology to forest trees. Plant Biotechnology Journal 1, 141–154. Carmel-Goren, L., Liu, Y.S., Lifschitz, E. and Zamir, D. (2003) The SELFPRUNING gene family in tomato. Plant Molecular Biology 52, 1215– 1222. Carmona, M.J., Cubas, P. and MartinezZapater, J.M. (2002) VFL, the grapevine FLORICAULA / LEAFY ortholog, is expressed in meristematic regions independently of their fate. Plant Physiology 130, 68–77. Carrington, J.C. and Ambros, V. (2003) Role of microRNAs in plant and animal development. Science 301, 336–338.

Biotechnology of Floral Development Carthew, R.W. (2001) Gene silencing by double-stranded RNA. Current Opinion in Cell Biology 13, 244–248. Clarke, J.H., Tack, D., Findlay, K., Van Montagu, M. and Van Lijsebettens, M. (1999) The SERRATE locus controls the formation of the early juvenile leaves and phase length in Arabidopsis. Plant Journal 20, 493–501. Coen, E.S., Romero, J.M., Doyle, S., Elliott, R., Murphy, G. and Carpenter, R. (1990) Floricaula: a homeotic gene required for flower development in Antirrhinum majus. Cell 63, 1311–1322. Constantin, G.D., Krath, B.N., MacFarlane, S.A., Nicolaisen, M., Johansen, I.E. and Lund, O.S. (2004) Virus-induced gene silencing as a tool for functional genomics in a legume species. Plant Journal 40, 622–631. Coupland, G. (1995) LEAFY blooms in aspen. Nature 377, 482–483. Cubas, P. (2004) Floral zygomorphy, the recurring evolution of a successful trait. Bioessays 26, 1175–1184. Cubas, P., Lauter, N., Doebley, J. and Coen, E. (1999) The TCP domain: a motif found in proteins regulating plant growth and development. Plant Journal 18, 215–222. Dinneny, J.R. and Yanofsky, M.F. (2004) Floral development: an ABC gene chips in downstream. Current Biology 14, R840–R841. Doebley, J. (2004) The genetics of maize evolution. Annual Review of Genetics 38, 37–59. Doebley, J., Stec, A. and Hubbard, L. (1997) The evolution of apical dominance in maize. Nature 386, 485–488. Ehsan, H., Reichheld, J.P., Durfee, T. and Roe, J.L. (2004) TOUSLED kinase activity oscillates during the cell cycle and interacts with chromatin regulators. Plant Physiology 134, 1488– 1499. Evans, M.M.S. and Poethig, R.S. (1997) The viviparous8 mutation delays vegetative phase change and accelerates the rate of seedling growth in maize. Plant Journal 12, 769–779.

261 Fambrini, M., Bertini, D. and Pugliesi, C. (2003) The genetic basis of a mutation that alters the floral symmetry in sunflower. Annals of Applied Biology 143, 341–347. Ferra´ndiz, C., Gu, Q., Martienssen, R. and Yanofsky, M.F. (2000) Redundant regulation of meristem identity and plant architecture by FRUITFULL, APETALA1 and CAULIFLOWER. Development 127, 725–734. Ferrario, S., Immink, R.G. and Angenent, G.C. (2004) Conservation and diversity in flower land. Current Opinion in Plant Biology 7, 84–91. Fornara, F., Parenicova, L., Falasca, G., Pelucchi, N., Masiero, S., Ciannamea, S., Lopez-Dee, Z., Altamura, M.M., Colombo, L. and Kater, M.M. (2004) Functional characterization of OsMADS18, a member of the AP1/ SQUA subfamily of MADS box genes. Plant Physiology 135, 2207–2219. Franke, R., McMichael, C.M., Meyer, K., Shirley, A.M., Cusumano, J.C. and Chapple, C. (2000) Modified lignin in tobacco and poplar plants over-expressing the Arabidopsis gene encoding ferulate 5-hydroxylase. Plant Journal 22(3), 223–234. Garcia-Perez, R.D., Houdt, H.V. and Depicker, A. (2004) Spreading of post-transcriptional gene silencing along the target gene promotes systemic silencing. Plant Journal 38, 594–602. Ginzberg, I., Perl, A., Genser, M., Wininger, S., Nemas, C. and Kapulnik, Y. (2004) Expression of streptavidin in tomato resulted in abnormal plant development that could be restored by biotin application. Journal of Plant Physiology 161, 611–620. Greenwood, M.S. (1995) Juvenility and maturation in conifers: current concepts. Tree Physiology 15, 433–438. Griffiths, S., Dunford, R.P., Coupland, G. and Laurie, D.A. (2003) The evolution of CONSTANS-like gene families in barley, rice, and Arabidopsis. Plant Physiology 131, 1855–1867.

262

C. Winefield and B.R. Jordan Guerineau, F., Sorensen, A.M., Fenby, N. and Scott, R.J. (2003) Temperature sensitive diphtheria toxin confers conditional male-sterility in Arabidopsis thaliana. Plant Biotechnology Journal 1, 33–42. Hempel, F.D., Welch, D.R. and Feldman, L.J. (2000) Floral induction and determination: where is flowering controlled? Trends in Plant Science 5, 17–21. Henderson, I.R. and Dean, C. (2004) Control of Arabidopsis flowering: the chill before the bloom. Development 131, 3829–3838. Henikoff, S., Till, B.J. and Comai, L. (2004) TILLING. Traditional mutagenesis meets functional genomics. Plant Physiology 135, 630–636. Honma, T. and Goto, K. (2001) Complexes of MADS-box proteins are sufficient to convert leaves into floral organs. Nature 409, 525–529. Hunter, C. and Poethig, R.S. (2003) miSSING LINKS: miRNAs and plant development. Current Opinion in Genetics & Development 13, 372– 378. Hunter, C., Sun, H. and Poethig, R.S. (2003a) The Arabidopsis heterochronic gene ZIPPY is an ARGONAUTE family member. Current Biology 13, 1734–1739. Hunter, C.A., Aukerman, M.J., Sun, H., Fokina, M. and Poethig, R.S. (2003b) PAUSED encodes the Arabidopsis exportin-t ortholog. Plant Physiology 132, 2135–2143. Joly, D., Perrin, M., Gertz, C., Kronenberger, J., Demangeat, G. and Masson, J.E. (2004) Expression analysis of flowering genes from seedlingstage to vineyard life of grapevine cv. Riesling. Plant Science 166, 1427–1436. Jones, L. (2002) Revealing micro-RNAs in plants. Trends in Plant Science 7, 473–475. Jordan, B.R. (ed.) (1993) The Molecular Biology of Flowering. CAB International, Wallingford, UK.

Jordan, B.R. and Anthony, R.G. (1997) Molecular events associated with floral evocation. In: Geneve, R.L., Preece, J.E. and Merkle, S.A. (eds) The Biotechnology of Ornamental Plants. CAB International, Wallingford, UK, pp. 215–235. Kapoor, M., Tsuda, S., Tanaka, Y., Mayama, T., Okuyama, Y., Tsuchimoto, S. and Takatsuji, H. (2002) Role of petunia pMADS3 in determination of floral organ and meristem identity, as revealed by its loss of function. Plant Journal 32, 115–127. Kardailsky, I., Shukla, V.K., Ahn, J.H., Dagenais, N., Christensen, S.K., Nguyen, J.T., Chory, J., Harrison, M.J. and Weigel, D. (1999) Activation tagging of the floral inducer FT. Science 286, 1962–1965. Keck, E., McSteen, P., Carpenter, R. and Coen, E. (2003) Separation of genetic functions controlling organ identity in flowers. EMBO Journal 22, 1058– 1066. Kempin, S.A., Savidge, B. and Yanofsky, M. (1995) Molecular basis of the cauliflower phenotype in Arabidopsis. Science 267, 522–525. Kidner, C.A. and Martienssen, R.A. (2005) The developmental role of microRNA in plants. Current Opinion in Plant Biology 8, 38–44. Komeda, Y. (2004) Genetic regulation of time to flower in Arabidopsis thaliana. Annual Review of Plant Biology 55, 521–535. Lawson, E.J.R. and Poethig, R.S. (1995) Shoot development in plants: time for a change. Trends in Genetics 11, 263–265. Lemmetyinen, J., Pennanen, T., Lannenpaa, M. and Sopanen, T. (2001) Prevention of flower formation in dicotyledons. Molecular Breeding 7, 341–350. Liu, J., Yu, J., McIntosh, L., Kende, H. and Zeevaart, J.A.D. (2001) Isolation of a CONSTANS ortholog from Pharbitis nil and its role in flowering. Plant Physiology 125, 1821–1830.

Biotechnology of Floral Development Liu, Y., Nakayama, N., Schiff, M., Litt, A., Irish, V.F. and Dinesh-Kumar, S.P. (2004) Virus induced gene silencing of a DEFICIENS ortholog in Nicotiana benthamiana. Plant Molecular Biology 54, 701–711. Lolle, S.J., Victor, J.L., Young, J.M. and Pruitt, R.E. (2005) Genome-wide nonMendelian inheritance of extra-genomic information in Arabidopsis. Nature 434, 505–509. Lukens, L. and Doebley, J. (2001) Molecular evolution of the teosinte branched gene among maize and related grasses. Molecular Biology and Evolution 18, 627–638. Luo, D., Carpenter, R., Vincent, C., Copsey, L. and Coen, E. (1996) Origin of floral asymmetry in Antirrhinum. Nature 383, 794–799. Mandel, M.A., Gustafson-Brown, C., Savidge, B. and Yanofsky, M.F. (1992) Molecular characterization of the Arabidopsis floral homeotic gene APETALA1. Nature 360, 273–277. Martin-Trillo, M. and Martinez-Zapater, J.M. (2002) Growing up fast: manipulating the generation time of trees. Current Opinion in Biotechnology 13, 151–155. Meely, B. and Briggs, S. (1995) Reverse genetics for maize. Maize Genetics Cooperation News Letter 69, 67–82. Mena, M., Mandel, A., Lerner, D., Yanofsky, M.F. and Schmidt, R.J. (1995) A characterization of the MADS-box gene family in maize. Plant Journal 8, 845–854. Meyer, K., Shirley, A.M., Cusumano, J.C., Bell-Lelong, D.A. and Chapple, C. (1998) Lignin monomer composition is determined by the expression of a cytochrome P450-dependent monooxygenase in Arabidopsis. Proceedings of the National Academy of Sciences USA 95, 6619–6623. Mizutani, M., Ohta, D. and Sato, R. (1997) Isolation of a cDNA and a genomic clone encoding cinnamate 4-hydroxylase from Arabidopsis and

263 its expression manner in planta. Plant Physiology 113, 755–763. Moon, J., Suh, S., Lee, H., Choi, K., Hong, C., Paek, N., Kim, S., Lee, I., Moon, J., Suh, S.S., Lee, H., Choi, K.R., Hong, C.B., Paek, N.C., Kim, S.G. and Lee, I. (2003) The SOC1 MADS-box gene integrates vernalization and gibberellin signals for flowering in Arabidopsis. Plant Journal 35, 613–623. Murai, K., Miyamae, M., Kato, H., Takumi, S. and Ogihara, Y. (2003) WAP1, a wheat APETALA1 homolog, plays a central role in the phase transition from vegetative to reproductive growth. Plant and Cell Physiology 44, 1255–1265. Ng, M. and Yanofsky, M.F. (2001) Function and evolution of the plant MADSbox gene family. Nature Reviews Genetics 2, 186–195. Ohshima, S., Murata, M., Sakamoto, W., Ogura, Y. and Motoyoshi, F. (1997) Cloning and molecular analysis of the Arabidopsis gene TERMINAL FLOWER 1. Molecular and General Genetics 254, 186–194. Oleykowski, C., Bronson Mullins, C., Godwin, A. and Yeung, A. (1998) Mutation detection using a novel plant endonuclease. Nucleic Acids Research 26, 4597–4602. Pena, L., Martin-Trillo, M., Juarez, J., Pina, J.A., Navarro, L. and MartinezZapater, J.M. (2001) Constitutive expression of Arabidopsis LEAFY or APETALA1 genes in citrus reduces their generation time. Nature Biotechnology 19, 263–267. Pennazio, S. (2004) ‘Florigen’: an intriguing concept of plant biology. Rivista Di Biologia-Biology Forum 97, 33–51. Peragine, A., Yoshikawa, M., Wu, G., Albrecht, H.L. and Poethig, R.S. (2004) SGS3 and SGS2 / SDE1/ RDR6 are required for juvenile development and the production of transacting siRNAs in Arabidopsis. Genes & Development 18, 2368–2379.

264

C. Winefield and B.R. Jordan Pineiro, M. and Coupland, G. (1998) The control of flowering time and floral identity in Arabidopsis. Plant Physiology 117, 1–8. Pnueli, L., Carmel-Goren, L., Hareven, D., Gutfinger, T., Alvarez, J., Ganal, M., Zamir, D. and Lifschitz, E. (1998) The SELF-PRUNING gene of tomato regulates vegetative to reproductive switching of sympodial meristems and is the ortholog of CEN and TFL1. Development 125, 1979–1989. Poethig, R.S. (1988) Heterochronic mutations affecting shoot development in maize. Genetics 119, 959–973. Poethig, S. (1997) Heterochrony in plant development and evolution. Developmental Biology 186, Suppl. S46. Poethig, S. (2002a) Genetic regulation of vegetative phase change in plants. Developmental Biology 247, Suppl. 521. Poethig, S. (2002b) Temporal and genetic regulation of vegetative phase change. Plant and Cell Physiology 43, Suppl. S8. Poethig, R.S. (2003) Phase change and the regulation of developmental timing in plants. Science 301, 334–336. 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. Reinhardt, D. and Kuhlemeier, C. (2002) Plant architecture. EMBO Reports 3, 846–851. Reinhart, B.J., Weinstein, E.G., Rhoades, M.W., Bartel, B. and Bartel, D.P. (2002) MicroRNAs in plants. Genes & Development 16, 1616–1626. Robert, L.S., Robson, F., Sharpe, A., Lydiate, D. and Coupland, G. (1998) Conserved structure and function of the Arabidopsis flowering time gene CONSTANS in Brassica napus. Plant Molecular Biology 37, 763– 772. Robertson, D. (2004) VIGS vectors for gene silencing: many targets, many

tools. Annual Review of Plant Biology 55, 495–519. Robson, F., Costa, M.M.R., Hepworth, S.R., Vizir, I., Pineiro, M., Reeves, P.H., Putterill, J. and Coupland, G. (2001) Functional importance of conserved domains in the flowering-time gene CONSTANS demonstrated by analysis of mutant alleles and transgenic plants. Plant Journal 28, 619– 631. Rottmann, W.H., Meilan, R., Sheppard, L.A., Brunner, A.M., Skinner, J.S., Ma, C., Cheng, S., Jouanin, L., Pilate, G., Strauss, S.H., Ma, C.P. and Cheng, S.P. (2000) Diverse effects of overexpression of LEAFY and PTLF, a poplar (Populus) homolog of LEAFY / FLORICAULA, in transgenic poplar and Arabidopsis. Plant Journal 22, 235–245. Sandvig, K., Spilsberg, B., Lauvrak, S.U., Torgersen, M.L., Iversen, T.G. and van Deurs, B. (2004) Pathways followed by protein toxins into cells. Zentralblatt fur Bakteriologie. International Journal of Medical Microbiology 293, 483. Sasaki, A., Ashikari, M., Ueguchi-Tanaka, M., Itoh, H., Nishimura, A., Swapan, D., Ishiyama, K., Saito, T., Kobayashi, M., Khush, G.S., Kitano, H. and Matsuoka, M. (2002) Green revolution: a mutant gibberellin-synthesis gene in rice – new insight into the rice variant that helped to avert famine over thirty years ago. Nature 416, 701– 702. Schmidt, R.J. and Ambrose, B.A. (1998) The blooming of grass flower development. Current Opinion in Plant Biology 1, 60–67. Schmidt, R.J., Veit, B., Mandel, M.A., Mena, M., Hake, S. and Yanofsky, M.F. (1993) Identification and molecular characterization of ZAG1, the maize homologue of the Arabidopsis floral homeotic gene AGAMOUS. Plant Cell, 5729–5737. Shavrukov, Y.N., Dry, I.B. and Thomas, M.R. (2004) Inflorescence and bunch architecture development in

Biotechnology of Floral Development Vitis vinifera L. Australian Journal of Grape and Wine Research 10, 116–124. Skinner, J.S., Meilan, R., Brunner, A.M., Strauss, S.H. (2000) Options for genetic engineering of floral sterility in forest trees. In: Jain, S.M. and Minocha, S.C. (eds) Molecular Biology of Woody Plants. Vol.1, Kluwer, Dordrecht, The Netherlands, pp. 135– 153. Smith, H.M.S., Campbell, B.C. and Hake, S. (2004) Competence to respond to floral inductive signals requires the homeobox genes PENNYWISE and POUND-FOOLISH. Current Biology 14, 812–817. Spielmeyer, W., Ellis, M.H. and Chandler, P.M. (2002) Semidwarf (sd-1), ‘‘green revolution’’ rice, contains a defective gibberellin 20-oxidase gene. Proceedings of the National Academy of Sciences USA 99, 9043–9048. Srinivasan, C. and Mullins, M.G. (1980) Effects of temperature and growth regulators on formation of anlagen, tendrils and inflorescences in Vitis vinifera L. Annals of Botany 45, 439–446. Strauss, S.H., Rottmann, W.H., Brunner, A.M. and Sheppard, C.A. (1995) Genetic engineering of reproductive sterility in forest trees. Molecular Breeding 1, 5–26. Sung, Z.R., Chen, L., Moon, Y.H. and Lertpiriyapong, K. (2003) Mechanisms of floral repression in Arabidopsis. Current Opinion in Plant Biology 6, 29–35. Tang, G. and Zamore, P.D. (2004) Biochemical dissection of RNA silencing in plants. Methods in Molecular Biology 257, 223–244. Telfer, A. and Poethig, R.S. (1998) HASTY: a gene that regulates the timing of shoot maturation in Arabidopsis thaliana. Development 125, 1889–1898. Theissen, G. and Saedler, H. (2001) Plant biology: floral quartets. Nature 409, 469–471.

265 Till, B.J., Reynolds, S.H., Greene, E.A., Codomo, C.A., Enns, L.C., Johnson, J.E., Burtner, C., Odden, A.R., Young, K., Taylor, N.E., Henikoff, J.G., Comai, L. and Henikoff, S. (2003) Large-scale discovery of induced point mutations with highthroughput TILLING. Genome Research 13, 524–530. Vaistij, F.E., Jones, L. and Baulcombe, D.C. (2002) Spreading of RNA targeting and DNA methylation in RNA silencing requires transcription of the target gene and a putative RNA-dependent RNA polymerase. Plant Cell 14, 857–867. Vega, S.H., Sauer, M., Orkwiszewski, J.A.J. and Poethig, R.S. (2002) The early phase change gene in maize. Plant Cell 14, 133–147. Walton, E.F., Podivinsky, E. and Wu, R-M. (2001) Biomodel patterns of floral gene expression over the two seasons that kiwifruit flowers develop. Physiologia Plantarum 111, 396– 404. Wang, M.B. and Waterhouse, P.M. (2002) Application of gene silencing in plants. Current Opinion in Plant Biology 5, 146–150. Waterhouse, P.M., Wang, M.B. and Finnegan, E.J. (2001) Role of short RNAs in gene silencing. Trends in Plant Science 6, 297–301. Weigel, D. and Meyerowitz, E.M. (1994) The ABC’s of floral homeotic genes. Cell 78, 203–209. Weigel, D., Alvarez, J., Smyth, D.R., Yanofsky, M.F. and Meyerowitz, E.M. (1992) LEAFY controls floral meristem identity in Arabidopsis. Cell 69, 843–859. Williams, C.G. and Savolainen, O. (1996) Inbreeding depression in conifers: implications for using selfing as a breeding strategy. Forestry Science 42, 102–117. Wiltshire, R.J.E., Murfet, I.C. and Reid, J.B. (1994) The genetic control of heterochrony: evidence from developmental mutants of Pisum sativum L.

266

C. Winefield and B.R. Jordan Journal Evolutionary Biology 7, 447–465. Yano, M., Katayose, Y., Ashikari, M., Yamanouchi, U., Monna, L., Fuse, T., Baba, T., Yamamoto, K., Umehara, Y., Nagamura, Y. and Sasaki, T. (2000) Hd1, a major photoperiod sensitivity quantitative trait locus in rice, is closely related to the Arabidopsis flowering time gene CONSTANS. Plant Cell 12, 2473–2484. Yanofsky, M.F. (1995) Floral meristems to floral organs: genes controlling early events in Arabidopsis flower de-

velopment. Annual Review of Plant Physiology and Plant Molecular Biology 46, 167–188. Yanofsky, M.F., Ma, H., Bowman, J.L., Drews, G.N., Feldmann, K.A. and Meyerowitz, E.M. (1990) The protein encoded by the Arabidopsis homeotic gene agamous resembles transcription factors. Nature 346, 35–39. Zik, M. and Irish, V.F. (2003) Flower development: initiation, differentiation, and diversification. Annual Review of Cell and Developmental Biology 19, 119–140.

III

Fertilization and Gametophyte Development

This page intentionally left blank

10

Control of Fertilization by Self-incompatibility Mechanisms

T. GAUDE, I. FOBIS-LOISY AND C. MIE`GE Reproduction et De´veloppement des Plantes, Ecole Normale Supe´rieure de Lyon, UMR 5667 CNRS-INRA-ENSL-UCB Lyon I, IFR 128 BioSciences Lyon-Gerland, 46 alle´e d’Italie 69364 Lyon cedex 07, France

Introduction In approximately 96% of flowering plant species, male and female organs are carried on the same individual, and in most cases (approximately 75% of species) within the hermaphroditic structure of flowers. Due to this close proximity of reproductive organs, there is a high probability for pollen to land on the stigma of the same flower or of a flower carried by the same individual. This feature will favour self-pollination and, as a consequence, will reduce the genetic variability in the species due to inbreeding. Besides, most angiosperms are anchored in the soil by their roots and therefore cannot actively search out their sexual partners. One might think that these traits should have promoted or imposed self-pollination, and thus self-fertilization, and might therefore have been deleterious for the evolution of angiosperms. However, since the end of the tertiary era, angiosperms have diversified and conquered almost all environments. They are the most successful groups of terrestrial flora both in terms of the number of species (approximately 300,000 described so far) and by their diversity of forms and ecological niches. One of the reasons for this success is thought to be the acquisition by the angiosperms of mechanisms that strongly limit or even prevent self-fertilization. The most sophisticated and widespread of these mechanisms is self-incompatibility (SI), which leads to the rejection of self-pollen by the pistil. SI mechanisms were described and studied as far back as the end of the 18th century by Charles Darwin, who observed that some plant species were completely sterile to their own pollen but fertile with that of any other individual of the same species (Darwin, 1876). Since this pioneer work, SI has been found to occur in more than half of the flowering plant species. Angiosperms have probably acquired SI early in their evolution (Weller et al., 1995), and genetic and molecular data clearly indicate that ßCAB International 2006. The Molecular Biology and Biotechnology of Flowering, 2nd edn (ed. B.R. Jordan)

269

270

T. Gaude et al.

SI arose independently several times during the course of evolution (de Nettancourt, 2001; Steinbachs and Holsinger, 2002). SI constitutes a quite original system of recognition since its operation is opposite to the immune system of vertebrates, to reject self while accepting non-self (Gaude and Dumas, 1987). SI systems allow the pistil to reject self-pollen grains, which are pollen issued from the same flower or from individuals genetically related to the female plant. In most cases, SI is genetically controlled by a single multiallelic locus, the S-locus (S for SI). SI systems can be classified into heteromorphic and homomorphic types (see de Nettancourt, 2001). In the heteromorphic type, flowers of the same species can have two or three different morphological forms, and successful pollination occurs only between flowers of different morphological forms. In the homomorphic type, there is no distinguishable difference between flowers of the same species. Homomorphic systems can further be classified into gametophytic self-incompatibility (GSI) or sporophytic selfincompatibility (SSI), depending on the genetic control of the pollen behaviour during SI interactions. In GSI, the phenotype of the pollen is determined by its own haploid genome, whereas in SSI the pollen phenotype depends on the diploid (sporophytic) genome of the pollen parent plant. In all cases, pollen rejection occurs when the same S-allele specificity is expressed by both the pollen and the pistil tissue (stigma, style or ovary). Since the beginning of the 1980s, the rapid expansion of new techniques in molecular biology and protein chemistry, and their use to study SI systems, has allowed significant advances in our knowledge of which molecules are involved in male–female recognition in flowering plants. Four of the families that display GSI, Solanaceae, Rosaceae, Scrophulariaceae and Papaveraceae, and one of the families that exhibits SSI, Brassicaceae (previously Crucifereae), have been extensively studied at the molecular level. These families often contain plant species of substantial interest for horticulture or agriculture, and thus, have been the object of intense classical genetic work in the past. For all these five families, SI is controlled by a single polymorphic S-locus. Although it is clear now that other loci are also required for a complete manifestation of the SI response, it is the S-locus that governs the self/non-self recognition step between the male and female partners. At least two distinct genetically linked genes are present at the S-locus, one encoding the male determinant while the second encodes the female determinant. Because of the presence of two genes (S-genes) at the S-locus, the term ‘haplotypes’ is generally used to designate variants of the S-locus (Boyes and Nasrallah, 1993). The mechanistic and molecular aspects of SI in these model families have been reviewed recently (see Kachroo et al., 2002; Franklin-Tong and Franklin, 2003; Hiscock and McInnis, 2003; Kao and Tsukamoto, 2004). The evolutionary aspects of SI have also been the concern of recent work and reviews (see Charlesworth et al., 2003; Vieira et al., 2003; Fobis-Loisy et al., 2004; Nasrallah et al., 2004; Shimizu et al., 2004). In this chapter, we concentrate on the most recent data concerning the molecular bases of SI studied in three model systems, two illustrating GSI (Solanaceae and Papaveraceae) and one illustrating SSI (Brassicaceae).

Control of Fertilization by Self-incompatibility Mechanisms

271

Gametophytic Self-incompatibility GSI is the most widespread system in flowering plants and has been described in more than 60 families including the Solanaceae, Rosaceae, Scrophulariaceae, Papaveraceae, Liliaceae and Poaceae. GSI has been particularly studied in two Solanaceae species (Nicotiana alata and Petunia inflata) and one Papaveraceae species (the poppy, Papaver rhoeas). Self-pollen rejection generally occurs by the arrest of pollen tube growth in the transmitting tissue of the style. However, in poppy and grasses, pollen tube growth is stopped earlier at the stigmatic surface. In the case of grasses, SI is genetically controlled by two loci, namely the S- and Z-genes; and this superior level of complexity, as compared to the single S-locus systems, has seriously slowed down the elucidation of the molecular bases of self-pollen rejection in this family (Baumann et al., 2000). The female component of SI has been characterized in five families. Interestingly, the same molecular mechanism based on a stylar ribonuclease (S-RNase) seems to be involved in the Solanaceae, Rosaceae, Scrophulariaceae and Campanulaceae. By contrast, a secreted glycoprotein (S-glycoprotein) with no known catalytic activity is involved in the Papaveraceae.

The S-RNase-based Self-incompatibility of the Solanaceae The female determinant is the S-RNase Since the first description of SI in the solanaceous Nicotiana sanderae by East and Mangelsdorf (1925), and the first report on the identification of pistilspecific S-proteins isolated in N. alata (Bredemeijer and Blass, 1981), it was only in the mid-1980s that the first S-gene of GSI systems was cloned from N. alata (Anderson et al., 1986). At that time, nucleotide or amino acid sequence databases were in their infancy; and it was only 3 years later that the biochemical nature of S-proteins was actually revealed, when the sequence of the extracellular RNase T2 of the fungus Aspergillus oryzae was found to share sequence similarity with S-proteins (McClure et al., 1989). During the meantime, several S-proteins and clones were isolated from other S-haplotypes, species and families exhibiting GSI. The S-gene products indeed have RNase activity and are glycoproteins abundantly expressed in the pistil (several micrograms per pistil). They are secreted in the extracellular space, principally in the stigma and transmitting tissue of the style. One remarkable feature of S-RNases is their high level of allelic sequence diversity. For example, in the Solanaceae, amino acid identity between S-RNases ranges from 38% to 98%. This level of polymorphism is unusual for most genes, but is expected for products of genes that are involved in self-recognition mechanisms (Clark and Kao, 1991). A new S-allele arising in a population is favoured, because pollen bearing this new S-allele will have high probability to encounter a stigma bearing a different S-allele. Sequence comparisons of S-RNases revealed that there are a large number of variable sites throughout the protein; however, two hypervariable

272

T. Gaude et al.

(HV) regions, namely HVa and HVb, were identified first in the solanaceous S-RNases (Ioerger et al., 1990; Tsai et al., 1992) and later in the Antirrhinum S-RNases (Xue et al., 1996). In the Rosaceae, Ishimizu et al. (1998) identified four regions of the S-RNase sequences that appear to be under positive selection, of which two overlap with the HVa and HVb domains. More recently, the crystal structure of two S-RNases has been determined by X-ray diffraction in N. alata and Pyrus pyrifolia, respectively (Ida et al., 2001; Matsuura et al., 2001). These studies reveal that both HVa and HVb regions are exposed on the surface of the protein and support a role for the HV regions in determining allelic specificity of the female determinant. However, a recent phylogenetic analysis of S-RNases based on new sequence information suggests that S-specificity may involve amino acids scattered through the S-RNases and shows evidence of recombination and/or diversifying selection in two distinct regions of S-RNases, corresponding to HVa and C-terminus, respectively (Roalson and McCubbin, 2003). S-RNases were good candidates as female determinants of GSI because they segregated with S-haplotypes; were polymorphic molecules; were expressed exclusively in the female tissues; and were developmentally correlated with the onset of SI. Direct confirmation of their involvement was first achieved by transgenic experiments in both Petunia and Nicotiana (Lee et al., 1994; Murfett et al., 1994). In Petunia, a loss-of-function approach (based on an antisense strategy) and a gain-of-function approach (using a sense strategy) were used by the group of Teh-hui Kao in the USA, to prove that S-RNases are sufficient for the recognition and rejection of self-pollen by the pistil (Lee et al., 1994). RNase activity is required for self-pollen rejection, because transgenic plants expressing a mutated S3-protein lacking RNase activity are not able to reject S3-pollen (Huang et al., 1994). Kao’s team also investigated whether the carbohydrate moiety and HV regions of S-RNases were important for recognition specificity. Expression of a mutated non-glycosylated S3-protein in transgenic plants did not alter the SI phenotype, indicating that the glycan chains do not confer S-specificity and are not necessary for S-RNase function in recognition and rejection of self-pollen (Karunanandaa et al., 1994). To address the role of HV regions in S-specificity, plants were transformed with constructs encoding chimeric S-RNases composed of the major part of the S3-RNase, but containing HV regions from the S1-RNase. The resulting transgenic plants lost the ability to reject S3-pollen but did not acquire the ability to reject S1-pollen, despite the fact that chimeric S-RNases exhibited normal levels of RNase activity (Kao and McCubbin, 1996). This result suggests that HV regions of S-RNases are necessary but not sufficient for determining S-specificity. An apparently contradictory result was obtained by Matton et al. (1997), who used two very closely related S-RNases (S11 and S13) of Solanum chacoense. These two S-RNases only differ by ten amino acid residues, of which four are located in HV regions. Plants transformed with chimeric gene constructs in which the S11-RNase had those four amino acids changed to those of the S13-RNase exhibited an S13-specificity. In this particular case, it appears that HV regions are sufficient for controlling S-haplotype specificity. However, it cannot be ruled out that amino acids outside HV regions and conserved

Control of Fertilization by Self-incompatibility Mechanisms

273

between S11-RNase and S13-RNase might also be involved in the recognition specificity of S-RNases (Verica et al., 1998). Although it is clear from these data that HV regions are important elements in determining S-specificity of S-RNases, further structure/function analyses are required to define more precisely domains of S-RNases that play a role in specific pollen recognition. A remarkable feature of all these transgenic experiments is that only the SI phenotype of the pistil was affected, the pollen phenotype remaining unchanged. This is consistent with the idea that two different genes located at the S-locus encode pollen and pistil S-molecules. How the S-RNases recognize and specifically inhibit self-pollen tubes remains unknown. Initially, based on our knowledge of cell–cell recognition systems in animals, it was thought that the SI reaction might be triggered by a ligand–receptor-type interaction. In this hypothesis, it was presumed that the pollen S-determinant is a receptor present at the surface of the pollen tube (in the wall or plasma membrane). Stylar S-RNases, which share the same S-haplotype specificity as that of the pollen S-receptor, are translocated into the pollen tube. This would lead to hydrolysis of ribosomal and messenger RNAs, blocking protein synthesis and therefore pollen tube growth (Kao and McCubbin, 1996). However, Luu et al. (2000) have shown that this selective entry of S-RNases into self-pollen tubes does not occur. By an elegant immunocytochemical study, these authors demonstrated in vivo uptake of the S11-RNase by S12-pollen tubes, indicating that S-RNases enter the pollen tubes independently of their S-allele specificity. This observation is compatible with another proposed hypothesis, which however gained fewer supporters. This hypothesis presumes that the pollen S-determinant is a cytosolic RNase inhibitor, which inactivates all S-RNases except the one sharing the same S-haplotype specificity as that of the pollen. In this ‘inhibitor’ model, only the S-RNases recognized as ‘self’ inside the pollen tube will not be inhibited, and thus will degrade RNA and stop pollen tube growth. This model supposes a dual function for the pollen S-product: (i) an S-specific recognition function; and (ii) an S-RNase inhibitor function. Another possibility is to have these two functions in two distinct molecules; the pollen S-determinant is responsible for S-specific recognition of the cognate S-RNase while a general RNase inhibitor is present in the pollen cytosol (Luu et al., 2000). In this latter model, specific interaction between the pollen S-product and its cognate S-RNase in self-incompatible tubes prevents the inhibitor from blocking the action of stylar S-RNases. The discovery of the pollen S-determinant has recently provided new support for this ‘inhibitor’ model. The male determinant is the S-locus F-box/S-haplotype-specific F-box protein During the last few years, a considerable amount of effort has been devoted to the isolation of the pollen S-gene. Recent advances in techniques of molecular biology, especially the development of bacterial artificial chromosome (BAC) libraries and facilities for large-scale genomic DNA sequencing, have played a major role in the identification of the pollen S-gene. As female and male

274

T. Gaude et al.

S-genes are genetically linked, both genes are predicted to be in close vicinity. The regions flanking S-RNases were sequenced and a good pollen S-gene candidate was searched for, i.e. an S-haplotype-specific polymorphic gene, specifically expressed in pollen (or male tissues) and located at the S-locus. The strongest candidate was isolated in Antirrhinum hispanicum (Scrophulariaceae) by Lai et al. (2002). This gene is located approximately 9 kb downstream from the S-RNase gene, encoding an F-box-containing protein and was designated AhSLF (A. hispanicum S-locus F-box gene). Most F-boxcontaining proteins are components of ubiquitin-ligase (E3) complexes that, together with ubiquitin-activating enzymes (E1) and ubiquitin-conjugating enzymes (E2), catalyse the polyubiquitination of specific substrates leading to their degradation by the 26S proteasome (Bai et al., 1996). Interestingly, homologues of AhSLF have been identified in several Prunus (Rosaceae) species (referred to as SLF in Prunus mume and SFB in Prunus dulcis) and in P. inflata (Entani et al., 2003; Ushijima et al., 2003; Sijacic et al., 2004). For clarity, we use only the term SLF to designate SLF/SFB pollen S-determinants. In P. inflata, PiSLF was located approximately 161 kb downstream of the S-RNase gene (Sijacic et al., 2004). Comparison of 13 SLF sequences of Prunus species revealed the presence of two HV regions, named HVa and HVb, located at the C-terminus (Ikeda et al., 2004; Kao and Tsukamoto, 2004). These HV regions might be responsible for the S-haplotype specificity of SLFs. The F-box motif is at the N-terminus and is relatively conserved among the various SLFs analysed. Demonstration that SLF is the pollen S-determinant was recently achieved in P. inflata (Sijacic et al., 2004) and A. hypersicum (Qiao et al., 2004a). In both studies, a similar approach was developed based on the ‘competitive interaction’ observed in the Solanaceae and other GSI species. Competitive interaction causes breakdown of SI by affecting exclusively the function of the pollen S-determinant. This phenomenon, which has been well described at the classical genetic level, occurs when a pollen grain carries two different S-haplotype specificities, arising for instance from duplications of the S-locus or from tetraploidy (see de Nettancourt, 2001). The defect only concerns heterollalelic pollen grains, e.g. diploid S1S2-pollen are compatible on S1S1-, S2S2-, S1S2-diploid or S1S1S2S2-tetraploid pistils. Pistil SI function is unchanged in tetraploids. Sijacic et al. (2004) introduced the PiSLF2 allele in self-incompatible S1S1 plants and obtained self-compatible plants in the progeny. Breakdown of SI was only detected on the pollen side and occurred only on S1-pollen grains expressing PiSLF2. The results were exactly as predicted by the competitive interaction. Interestingly, they showed by RT-PCR that both endogenous PiSLF1 and the PiSLF2 transgene were transcribed, indicating that competitive interaction is not dependent on silencing of pollen S-genes. How then does the discovery of the pollen S-determinant being an F-box protein fit with the ‘inhibitor’ model? If SLF is a component of the protein degradation machinery, the inhibition of non-self S-RNases could occur by a direct interaction between SLF and S-RNases, followed by the ubiquitination and degradation of S-RNases. Interestingly, Qiao et al. (2004b) showed in

Control of Fertilization by Self-incompatibility Mechanisms

275

Antirrhinum that AhSLF2 physically interacts with both self and non-self S-RNases, but seems to mediate degradation of only non-self S-RNases. We may suppose that PiSLF might act similarly. Nevertheless, these data do not allow us to distinguish as to whether the SLF molecule carries the two functions of S-haplotype recognition and S-RNase inhibition or only S-haplotype recognition, inhibition of S-RNases being controlled by a general RNase inhibitor that may not necessarily be encoded by an S-locus gene. Recent research carried out on two self-compatible selections of Prunus avium (sweet cherry) provides new insights on how SLF might work in SI (Sonneveld et al., 2005). Both selections have a functional S-RNase and self-compatibility is attributed to a loss-of-function of the pollen S-determinant. A molecular analysis demonstrated that there is no evidence of a duplication in either mutant selection, but one mutant has a sequence rearrangement downstream of the S-RNase gene, which includes the complete deletion of the SLF gene, while the other mutant has a 4-bp deletion in the HVa region of SLF, which leads to a shift in the translation reading frame and premature termination of the protein. The loss of pollen SI function in these two selections, associated with the loss or significant alteration of the SLF gene, provides compelling evidence that SLF is the gene encoding the pollen S-determinant in Prunus. In the inhibitor model, if SLF has a dual function, the loss-of-function of SLF should be coupled with the loss of inhibition of S-RNases. As a consequence, stylar S-RNases that enter the pollen tubes would remain active and block pollen tube growth, leading to an SI phenotype. Interestingly, a self-compatibility phenotype is observed in both sweet cherry mutants impaired in their pollen S function, which indicates that SLF does not carry the S-RNase inhibition function. These results are somehow contradictory to those obtained in Antirrhinum, which suggested that AhSLF2 interacts with non-self S-RNases to target non-self S-RNase proteasomal degradation during compatible pollination (Qiao et al., 2004b). By contrast, the work in Prunus suggests that SLF is likely to control only the specific S-haplotype recognition of S-RNases, preventing their inactivation by a general RNase inhibitor present in pollen tubes. These recent data are in favour of the two-component inhibitor model proposed by Luu et al. (2000). In this context, the role of the F-box domain of SLF is difficult to understand if SLF is not directly involved in S-RNase inactivation. We propose a possible mode of action of SLF, based on its specific interaction with its cognate S-RNase, within a protein complex made up of the heterotrimer S-RNase/general RNase inhibitor/SLF (Fig. 10.1). The role of SLF in this complex would be to promote polyubiquitination of the general RNase inhibitor, and hence, its degradation, leading to the maintenance of active S-RNases in the pollen tube and arrest of growth. In the absence of molecular recognition between SLF and S-RNase, only S-RNase/general RNase inhibitor heterodimers would form authorizing the growth of compatible pollen tubes. Although protein degradation via the 26S proteasome activity has been shown to be important in compatible but not in incompatible interactions in Antirrhinum, the direct involvement of SLF in an E3 complex remains to be demonstrated (Qiao et al., 2004b). It is worth mentioning that classic and molecular genetic studies have revealed the existence of genes at other loci than the S-locus that are required for the full

276

T. Gaude et al.

S2-RNase

General RNase inhibitor

S1-pollen tube

No RNase activity

S1-RNase

Pollen SLF1

RNase activity

Ub

AAA

Degradation

26S proteasome

mRNA rRNA

Tube tube growth growth arrest arrest

S1S2 style

Fig. 10.1. Model for the S-RNase-dependent mechanism of gametophytic self-incompatibility (GSI). Stylar S-RNases (S1- and S2-RNases) present in the extracellular space of transmitting tissue cells are taken up into the growing S1-pollen tube, independently of their S-allele specificity. Identity of proteins involved in this translocation is still unknown. Once in the pollen tube cytosol, S-RNases sharing no S-allele specificity with the pollen S-locus F-box (SLF) protein will interact with a general pollen RNase inhibitor that inactivates the S-RNases. By contrast, if the S-RNase and SLF share the same S-allele specificity (here S1), the general pollen RNase inhibitor will not be able to inactivate the self S1-RNase. We propose that this occurs via the formation of a heterotrimeric protein complex consisting of the S1-RNase/general RNase inhibitor/SLF1 proteins. The role of SLF1 in this complex would be to promote specific polyubiquitination of the general RNase inhibitor leading to its degradation by the 26S proteasome. In the absence of inhibition, the S1-RNase remains active and can degrade the pollen tube RNA, resulting in the arrest of selfpollen tube growth. PM ¼ plasma membrane.

manifestation of the SI response (Kao and Tsukamoto, 2004). In particular, Sims and Ordanic (2001) isolated in a yeast two-hybrid screen a pollenexpressed protein of Petunia hybrida, namely PhSBP1, which contains a RING-finger domain and binds to S-RNases in a non-S-allele-specific manner. RING-finger domains are often present in proteins, like F-box proteins, which participate in E3 complexes. These authors suggest that PhSBP1 is a good candidate for the general inhibitor of S-RNases, the existence of which is presumed to be necessary in the two-component inhibitor model.

Control of Fertilization by Self-incompatibility Mechanisms

277

Further molecular and biochemical investigations are required to define the role of SLF in the specific recognition of S-RNases and to elucidate the mechanism of S-RNase inactivation. Further work is also needed to ascertain whether SI in the Solanaceae, Scrophulariaceae and Rosaceae operates according to similar molecular mechanisms.

The S-glycoprotein-based self-incompatibility of the Papaveraceae In the GSI system of the Papaveraceae, the model species field poppy P. rhoeas has been particularly well studied by research groups in Birmingham, UK, first by Mike Lawrence and co-workers and then by Chris Franklin, Veronica Franklin-Tong and co-workers. By contrast to the S-RNase GSI systems described above, recognition and rejection of self-pollen grains occurs very rapidly within minutes, after deposition of pollen on to the stigmatic papillae of poppy flowers. The stigmatic S-determinant has been identified and several alleles of the female S-gene cloned (Foote et al., 1994). The stigmatic S-proteins are small (approximately 15 kDa) glycoproteins secreted in the extracellular space, which do not have RNase activity or any other known catalytic activity. However, using an in vitro pollen germination assay, stigmatic S-glycoproteins have been shown to inhibit pollen tube growth in an S-haplotype-specific manner. This led to the hypothesis that the stigmatic S-glycoproteins interact with the pollen S-gene product, which is believed to be a plasma membrane receptor. In the search for such a receptor, a pollen plasma membrane protein called S-protein-binding protein (SBP) was shown to interact specifically with the stigmatic S-glycoproteins, but with no S-haplotype specificity (Hearn et al., 1996). For this reason, it is likely that SBP is not the pollen S-determinant but it may be an accessory receptor required for a full manifestation of the SI response. This idea gained support from the analysis of mutant forms of the stigmatic S-glycoprotein, which were produced in Escherichia coli and assayed for pollen inhibition and SBP binding in vitro. All mutant forms that exhibit reduced ability to inhibit incompatible pollen tube growth are also reduced in SBP binding activity (Jordan et al., 1999). A better understanding of the initial event that takes place at the plasma membrane during selfpollen recognition necessitates cloning of SBP and the pollen S-receptor. Although the pollen S-determinant is still unknown in the poppy, the use of in vitro germination assay and recombinant forms of stigmatic S-glycoproteins has allowed us to get considerable information on the signal transduction pathway that occurs in pollen tubes after the initial recognition response (see the review by Franklin-Tong and Franklin, 2003). Briefly, SI induction triggers a transient increase in cytosolic [Ca2þ]i, involving influx of extracellular Ca2þ at the shank of the pollen tube (Franklin-Tong et al., 2002). This Ca2þ influx is rapidly followed by inhibition of pollen tube growth within 1–2 min. Ca2þ is likely to act as a second messenger and to trigger a signalling cascade, the final result of which is irreversible rejection of self-pollen tubes. The fact that the Ca2þ influx occurs at the shank, and not at the tip, may indicate that the pollen S-receptors are located in this region and might be directly involved in the

278

T. Gaude et al.

control of the Ca2þ influx. The Ca2þ influx is coupled with the loss of the apical gradient of [Ca2þ]i, and reorganization of the pollen actin cytoskeleton together with extensive depolymerization of F-actin. Snowman et al. (2002) showed that within 5 min after SI induction F-actin concentrations were less than 50% of the controls, and remained low even at 1 h after SI induction. The authors calculated that the level of depolymerization was several orders of magnitude greater than that required to inhibit tip growth of pollen tubes. Actin-binding proteins that might be involved in this particular alteration of the actin cytoskeleton have been investigated. Profilin, which is an abundant pollen protein, might participate in this process. Another actin-binding protein, a gelsolin-like protein isolated from Papaver pollen, named PrABP80, has recently been shown to actively depolymerize actin in vitro (Huang et al., 2004). This protein has an actin filament severing, nucleating, capping and depolymerizing activity, which is regulated by Ca2þ. During SI, PrABP80 is likely to enhance profilin-mediated depolymerization of actin filaments in pollen tubes. Another rapid change that occurs within 90 s of SI induction is the increase in phosphorylation of a cytosolic pollen protein, p26 (Rudd et al., 1996). Cloning of the p26-encoding gene predicted that the protein is a soluble inorganic pyrophosphatase. Biochemical assays confirmed the catalytic activity of the recombinant p26 protein and its phosphorylation was associated with a reduction in its pyrophosphatase activity. Soluble inorganic pyrophosphatases are key enzymes in generating ATP and are involved in the synthesis of biopolymers. It was suggested that the reduction in p26 pyrophosphatase activity might contribute to the inhibition of SI tip growth by an arrest of or reduction in the biosynthesis of pollen wall components (Rudd and Franklin-Tong, 2003). The activity of another pollen protein is affected following SI response. This is a pollen mitogen-activated protein kinase (MAPK), designated p56, which curiously is activated several minutes after the arrest of SI pollen tube growth (Rudd et al., 2003). Due to this peculiar timing, p56 is unlikely to play a role in the early events that led to the rapid inhibition of SI pollen tubes. Instead, it is proposed that p56 might play a role in initiating the signalling cascade making the arrest of pollen tube irreversible. This idea of a two-step inhibition mechanism controlling SI in poppy has recently received strong support from the demonstration that SI response in poppy involves programmed cell death (PCD) (Thomas and Franklin-Tong, 2004). PCD is specifically triggered in self-pollen tubes following SI induction and involves caspase 3-like activity, cytochrome c leakage in the cytosol and PARP (a classic substrate of caspase3)-cleavage activity, all three of which are diagnostic features of PCD. Moreover, the data show that SI response is a biphasic process, with a first step of rapid inhibition of pollen tube growth followed by a second step initiated by the activation of a caspase-like enzyme. This second step is likely to make selfpollen inhibition irreversible, as it ends with cell death (Fig. 10.2). This remarkable work points out that although genetically based on the same gametophytic control of self-pollen rejection, the molecular mechanisms selected to inhibit self-pollen grains in the Solanaceae and other S-RNase GSI systems differ fundamentally from those recruited in the Papaveraceae.

Control of Fertilization by Self-incompatibility Mechanisms

279

S1-pollen tube

Pollen S1-receptor

Ca2+

Ph

1 min stigmatic S2-glycoprotein

P

2 min V

rylation

Stigmatic S1-glycoprotein

Caspase-like activity

o Phosph

osp hor yl a tion Depo lyme rizat ion

Ca 2+

90 min

5−30 min PCD

P p26

F-actin

p56

0–10 min Reversible Reversible tube growth tube growth arrest arrest

Tube Tube death death

S1S2 stigmatic cell

Fig. 10.2. Model for the S-glycoprotein-dependent mechanism of gametophytic selfincompatibility (GSI) system in Papaver rhoeas. S-glycoproteins secreted by the stigmatic cells are thought to interact, in an S-dependent manner, with the still unidentified pollen S-receptor. The specific stigmatic S-glycoprotein/pollen S-receptor interaction (here between S1-allelic products) initiates an almost instantaneous Ca2þ influx and increase of cytosolic free Ca2þ within the self-pollen. This triggers an intracellular Ca2þ-dependent signalling cascade, which includes the phosphorylation of p26 (a soluble inorganic pyrophosphatase) and p56 (a MAPK) proteins, and depolymerization of F-actin (involving the PrABP80 enzyme). The activity of p56 MAPK increases 5 min after SI induction, reaches a peak at 10 min and remains high for a least 30 min. It is proposed that the activation of p56 MAPK initiates a programmed cell death (PCD), which makes the arrest of self-pollen tube growth irreversible.

Sporophytic Self-incompatibility Genetic complexity of sporophytic self-incompatibility In the Brassicaceae, the number of S-haplotypes is usually large, being estimated as over 90 in Brassica oleracea (Ruffio-Chable and Gaude, 2001). Detailed molecular analyses of SI performed during the last two decades in Brassica species, and more recently in the related species Arabidopsis lyrata and Raphanus sativus, have demonstrated that the S-locus consists of several distinct polyallelic genes. This multi-gene complex at the S-locus is usually

280

T. Gaude et al.

inherited as one unit, which was the basis for the introduction by Boyes and Nasrallah (1993) of the term S-haplotype. The genomic organization of the S-locus region has been studied by several groups (e.g. Yu et al., 1996; Cui et al., 1999; Suzuki et al., 1999; Kusaba et al., 2001; Shiba et al., 2003). Sequencing of a 76-kbp region of the S9-haplotype in Brassica rapa revealed that this region contains at least 14 expressed genes (1 gene/5.4 kbp) and three putative open reading frames (ORFs) (Suzuki et al., 1999). The presence of transposable elements and retrotransposons has also been reported (Pastuglia et al., 1997; Shiba et al., 2003). The region around the S-determinant genes is highly polymorphic and filled with S-haplotype-specific intergenic sequences. It is likely that this sequence polymorphism contributes to the suppression of recombination at the S-locus. Of the different genes characterized at the S-locus, three genes have been studied in detail as they are thought to be involved in the recognition of self-pollen grains by the stigma. Two of these genes, namely SLG (S-locus glycoprotein) and SRK (S-locus receptor kinase), are specifically expressed in the stigma (Nasrallah et al., 1985; Stein et al., 1991) whereas the third gene, SCR (S-locus cysteine rich), is only expressed in anthers and pollen grains (Schopfer et al., 1999). The complexity of SSI systems is also revealed by the existence of dominance interactions that occur between S-haplotypes in a heterozygous state. Thus S-haplotypes can be ordered in a non-linear dominance hierarchy (Thompson and Taylor, 1966). Two major groups have been distinguished in the dominance hierarchy: (i) class I haplotypes tend to be dominant and confer a strong SI phenotype; and (ii) class II haplotypes tend to confer a weaker phenotype (i.e. some seeds can be produced following self-pollination) and to be recessive to the class I haplotypes. While S-haplotypes are usually co-dominant in stigmas, dominance interactions can be observed in anther tissues. This particular feature of the genetics of SSI plants allows the production of homozygous S-haplotypes after crossing of plants that share a common S-haplotype. For example, an S2S4 heterozygous plant, with S2 co-dominant with S4 in the stigma, crossed with pollen from the pollen donor S2S3 heterozygous plant, with S3 dominant to S2 in the anther, will set seed as the pollen S2-determinant is not expressed in the pollen-producer plant. Hence, S2S2 homozygous plants will be present in the progeny. The molecular aspects of these genetic interactions between S-haplotypes are just starting to be understood (see below).

The female determinant is the S-locus receptor kinase protein Cytological studies of pollen rejection in Brassicaceae species showed a very early inhibition of the incompatible pollen. In all cases analysed, either the selfpollen fails to hydrate completely or, if it germinates, pollen tube elongation is rapidly arrested at the stigma surface (see de Nettancourt, 2001). Analysis of proteins extracted from the stigmas of plants carrying different S-haplotype revealed the presence of abundant polymorphic glycoproteins characteristic of each S-haplotype, the so-called SLGs. Cloning of the SLG gene (Nasrallah

Control of Fertilization by Self-incompatibility Mechanisms

281

et al., 1985) and later of the related S-locus-linked gene, SRK (Stein et al., 1991), were key steps in the molecular analysis of SI. As the accumulation of SLG and SRK transcripts and proteins coincides with the acquisition of SI by the maturing stigma, both genes have been considered as putative female determinants of self-pollen recognition (Stein et al., 1991, 1996). The SLG gene encodes a glycoprotein secreted into the cell wall of stigmatic papillae. Although the majority of the SLG protein accumulates in the stigma, SLG can also be detected, although at very low levels, in the transmitting tissue of the style and ovary (Kleman-Mariac et al., 1995). The only other organ in which SLG transcripts have been detected is the anther, but SLG protein does not accumulate to a detectable level (Delorme et al., 1995). The SRK gene encodes a membrane-spanning protein of the plasmalemma. SRK consists of an extracellular domain that shares homology with SLG (designated the S-domain), a single transmembrane alpha-helix, and a cytoplasmic domain exhibiting a serine/threonine kinase activity (Goring and Rothstein, 1992; Stein and Nasrallah, 1993; Giranton et al., 2000). SRK protein is only detected in stigmas, although low levels of SRK transcripts can be detected in anthers. The structure of SRK resembles the receptor kinases from animals and it is presumed that the extracellular domain acts as a ligand-binding region (Cock et al., 2002). Analysis of a number of alleles of SLG and SRK genes has shown that both genes are highly polymorphic and that their respective alleles can be grouped into two classes which correspond to the class I and II groups defined on the basis of SI phenotype (Stein et al., 1991; Kusaba et al., 1997; Cabrillac et al., 1999). Recent analysis has also shown a great complexity at the level of SLG and SRK gene expression, as certain alleles of both genes encode more than one protein product. In the case of the SLG gene, which is usually intronless, a few S-haplotypes have SLG alleles that contain one intron. Tantikanjana et al. (1993) showed that the SLG allele of the class II S2-haplotype (SLG2) possesses two exons and that alternative transcripts of this allele can encode both a secreted form of SLG and a membrane-anchored form designated mSLG. More recently, Cabrillac et al. (1999) analysed the SLG alleles and protein products of the three (S2, S5 and S15) class II S-haplotypes of B. oleracea. Two different SLG genes were isolated in the S15-haplotype, designated SLGA and SLGB. Both of these genes possess two exons interrupted by a single intron but only SLGA possesses a second exon that encodes a membrane-spanning domain. SLGA is predicted to encode a secreted SLGA protein and a membrane-anchored mSLGA protein, whereas SLGB only encodes a secreted SLGB protein. Analysis of stigma proteins from the S15-haplotype confirmed this prediction. Interestingly, the two other class II S2- and S5-haplotypes carry only one or the other of the SLG genes. In fact, recent studies on the S2-haplotype have revealed the existence of sequence heterogeneity at the S-locus genes for this haplotype (Kusaba et al., 2000; Mie`ge et al., 2001). Mie`ge et al. (2001) analysed ten S2-haplotypes of B. oleracea and found an intrahaplotype polymorphism at the sequence level of SLG and SRK genes. Two groups of S2-haplotype can be defined, depending on the presence of either only SLGA or only SLGB. Surprisingly, SRK2 alleles from the two

282

T. Gaude et al.

groups can be distinguished at the sequence level, suggesting that recombination is suppressed or rarely occurs between haplotypes of the two groups. Complexity of the SRK gene expression is also well documented. For example, in the class I S3-haplotype, SRK3 encodes at least seven different transcripts including transcripts from both strands of the gene (Delorme et al., 1995; Cock et al., 1997). Several of the sense transcripts retain all or part of the first intron and, because there is a termination codon just after the 5’ end of the intron, are predicted to encode a soluble, truncated form of SRK that corresponds to the predicted extracellular domain (Giranton et al., 1995). This truncated form of SRK, which has been called eSRK for extracellular SRK, resembles SLG and has been detected in stigma proteins of S3 plants. No eSRK protein was detected in stigmas of plants homozygous for the class II haplotype S15 (Cabrillac et al., 1999). Analysis of protein products coded by the various transcripts of SRK alleles will be necessary to determine how frequent the presence of eSRK is in the stigma of different S-haplotypes and whether eSRK is associated exclusively with class I S-haplotypes. The function of mSLG, eSRK and other putative variants of SLG or SRK proteins remains unknown. Actually, the functional role played by SLG proteins in the self-pollen recognition has been seriously questioned over the last few years. The first observation that called into question the importance of SLG was that plants, which express very low or undetectable levels of SLG can be self-incompatible, while plants that express high levels of SLG can be selfcompatible (Gaude et al., 1993, 1995; Okazaki et al., 1999). Moreover, molecular analysis of B. oleracea plants homozygous for two different S-haplotypes, S18 and S60, has shown that these plants are unable to produce functional SLG due to mutations in the SLG coding region, although their SI phenotype is perfectly maintained (Nishio and Kusaba, 2000). Furthermore, certainly the most convincing evidence comes from sequencing of the S-locus region from different S-haplotypes of A. lyrata, which revealed the absence of the SLG gene in this SI species (Kusaba et al., 2001). Taken together, these data suggest that SLG is not essential for SI in the Brassicaceae and that SRK, rather than SLG, plays a key role in the recognition of self-pollen. The importance of SRK in the accomplishment of the SI response gained strong support from the analysis of self-compatible variants of oilseed rape and cabbage (Goring et al., 1993; Nasrallah et al., 1994). In both studies, these plants express normal levels of SLG proteins, but they carry mutations that lead either to the absence of SRK transcripts or to the production of truncated transcripts. In all cases, these self-compatible plants are unable to produce a functional SRK protein. These data indicate that a functional SRK is required for the SI response. Compelling evidence that SRK is actually the female determinant of SI has only been obtained recently by transgenic experiments. Efforts to express novel SRK alleles (gain-of-function approach) or dominant negative forms of SRK (loss-of-function approach) in transgenic plants have run into problems for many years due to insufficient expression and cosuppression. To avoid co-suppression due to the high similarities between the transgene and endogenous SRK alleles, Takasaki et al. (2000) developed an elegant strategy based on the use of a transgenic class I SRK28 allele to

Control of Fertilization by Self-incompatibility Mechanisms

283

transform a class II S60-haplotype receiver plant. They demonstrated that selfincompatible B. rapa plants of the S60-haplotype transformed with the SRK28 gene acquire the ability to reject S28-pollen in addition to rejecting their self (S60)-pollen. In this study, the potential role of SLG in the SI was also addressed by analysing the phenotype of transgenic plants expressing both SRK28 and SLG28. While expression of SLG28 alone is not sufficient to confer a new S-haplotype specificity in transgenic plants, the degree of S28-pollen rejection by plants carrying the SRK28 transgene is enhanced by the presence of the SLG28 transgene. These experiments suggest that SRK alone determines the S-haplotype specificity of the stigma, and that SLG, when present, may act to promote a full manifestation of the SI response.

The male determinant is the S-locus protein 11/ S-locus cysteine-rich protein Among the different genes located at the S-locus, those whose expression is specific to male tissues have been considered as good candidates for encoding the male determinant of SI. However, most of them were eliminated as the potential pollen determinant either because they were present in both selfincompatible and self-compatible lines or because they showed no sequence variation between different S-haplotypes or, lastly, because direct evidence of their role in SI was lacking (Pastuglia et al., 1997; Mie`ge et al., 1999; Suzuki et al., 1999). By combining a biochemical approach with a pollination bioassay system, Stephenson et al. (1997) showed that pollen proteins coating the cavities of the pollen wall could induce the SI response in the stigma in a S-haplotype-specific manner. Fractionation of these pollen coat proteins (designated PCPs) revealed that molecules responsible for this activity, presumably the pollen S-determinants, are basic proteins with a molecular mass of less than 10 kDa. Interestingly, a Japanese group (Suzuki et al., 1999) identified and characterized in detail genes present in the SLG/SRK region of the S9haplotype of B. rapa and found that one of these genes, SP11 (S-locus protein 11), encodes a small cysteine-rich protein resembling the PCP isolated by Stephenson et al. (1997). By RNA gel blot and in situ hybridization analyses, SP11 transcripts were detected in early stages of anther development as well as in the developing pollen at late stages (Takayama et al., 2000). The unequivocal implication of SP11 in the SI response was demonstrated by loss-of-function and gain-of-function experiments performed by the group of June and Mike Nasrallah in the USA (Schopfer et al., 1999). By a systematic sequencing analysis of the S-locus region of different B. oleracea S-haplotypes, they isolated a gene, designated SCR, which is specifically expressed in anthers and corresponds to the previously characterized SP11 gene. The SP11/SCR gene encodes a novel class of highly polymorphic peptides within a family of proteins named the PCP family (Vanoosthuyse et al., 2001). For clarity, we use only the term SCR to designate the SP11/SCR gene or gene product. Schopfer et al. (1999) studied a self-compatible mutant of B. oleracea, generated by gamma irradiation, and showed that this mutant was unable to produce detectable levels of SCR transcripts. This lack of SCR transcripts resulted in the loss

284

T. Gaude et al.

of S-specificity in the pollen but not in the stigma. By contrast, transformation of a B. oleracea S2S2-homozygous line with the SCR6 cDNA led to the production of S2-pollen that expressed the SCR6 transgene and that was rejected on S6S6 stigmas. Similar results were obtained by Shiba et al. (2001) using the gain-of-function approach in B. rapa. These data clearly demonstrate that the SCR gene is necessary and sufficient for determining SI specificity in pollen of Brassica species. The analysis of the deduced amino acid sequences of several cloned SCR sequences predicts that the mature SCR peptides are hydrophilic, small (5.7–8.0 kDa), basic (isoelectric points over 8.0), secreted peptides containing generally eight conserved cysteine residues and a conserved glycine (Schopfer et al., 1999; Takayama et al., 2000, 2001). In spite of the very high level of polymorphism observed between SCR alleles, these conserved residues are presumably required to maintain a similar three-dimensional structure in the mature protein. This assumption has gained strong support from the recent determination of the solution structure of SCR8 of B. rapa (Mishima et al., 2003). Homology modelling of allelic SCR sequences and structure-based alignment identified a protruding loop, loop 1, as a HV region in allelic SCRs. The authors suggest that the HV region is likely to serve as a specific binding site for the SRK protein.

SCR/SRK interaction and self-pollen rejection The identification of SRK and SCR as the stigma and pollen S-determinants of SI, respectively, suggests that self-pollen rejection in the Brassicaceae depends on a receptor/ligand-like interaction. In this model, the SCR peptide is the pollen-borne ligand that specifically activates the stigmatic SRK receptor in the case of a self-incompatible pollination. Indeed, studies of animal receptor kinases have revealed that the general mode of action of these receptors is through the activation of their cytosolic-kinase domain following ligand binding (Heldin, 1995; Cock et al., 2002). Experiments carried out to determine the phosphorylation status of SRK in planta, at the basal state and following selfor cross-pollination, revealed that native SRK is not phosphorylated before pollination (Cabrillac et al., 2001). This inactive state of SRK was presumed to be maintained by the stigmatic thioredoxins THL1 and THL2, which act as negative regulators of the SRK kinase activity in vitro (Cabrillac et al., 2001). Recently, this assumption has gained support from the analysis of antisense THL1 or THL2 B. napus cv Westar transgenic plants, which shows that suppression of THL1/2 transcripts is associated with a low level constitutive rejection of normally compatible B. napus pollen grains (Haffani et al., 2004). The pollen rejection observed in fluorescence microscopy was a typical Brassica SI rejection response with reduced pollen adhesion, germination and pollen tube growth arrest at the surface of stigma papillae. Although the authors could not detect any change in SRK activation in the partially SI antisense THL1/2 transgenic plants, these results indicate that the THL1 and THL2 are required for full pollen acceptance in B. napus cv Westar.

Control of Fertilization by Self-incompatibility Mechanisms

285

The existence of SRK receptor complexes consisting of one or two SRK molecules associated with other stigma proteins in non-pollinated stigmas has been shown by cross-linking experiments and by velocity sedimentation on sucrose gradients (Giranton et al., 2000). These findings suggest that signal transduction during the SI response might be mediated by modification of a pre-existing inactive SRK oligomeric complex. We propose that the interaction of SCR with the extracellular domain of SRK, within the SRK oligomeric complex, leads to a conformational change of the SRK kinase domain that allows the release of THL1/2 inhibition and activation, through autophosphorylation, of the SRK receptor kinase. Using an in vitro phosphorylation assay, Cabrillac et al. (2001) showed for the first time that PCPs containing SCR can release the negative regulation of THL1 and activate recombinant integral SRK proteins in an S-specific manner. The direct allele-specific interaction between SCR and the extracellular domain of SRK as well as activation of the kinase domain of SRK following SCR binding have been confirmed in vitro (Kachroo et al., 2001; Takayama et al., 2001). How SCR peptides accumulate in the pollen coat during pollen development, and then reach the plasma membrane of the stigma papilla following pollination has been examined by immunoelectron microscopy (Shiba et al., 2001; Iwano et al., 2003). SCR is secreted from the tapetal cells into the anther locule and then translocated to the pollen surface at the early developmental stage of the anther. In pollen grains at late developmental stages, prior to anther dehiscence, SCR is mainly localized in the pollen coat. During the pollination process, SCR is translocated from the pollen surface to the surface of the papillar cell, and then penetrates the cuticle layer of the stigma papilla to migrate across the pectin cellulose layer. Interestingly, SCR alone cannot initiate an incompatible response when applied directly on to the stigma surface, apparently because it is unable to penetrate the stigmatic cuticle on its own. This indicates that an additional factor carried by the pollen grain is required to ensure the proper diffusion of SCR from the pollen surface to the plasma membrane of the stigma papilla, where it will interact physically with its complementary S-allelic SRK. Recently, regions that determine recognition specificity in SCR have been investigated by producing SCR variants generated by the swapping of specific domains between different SCR alleles and by in vitro mutagenesis (Chookajorn et al., 2004; Sato et al., 2004). These variants have been analysed for their capacity to bind SRK by using an in vitro binding assay and a pollination test with transgenic plants expressing the chimeric SCR peptides. This work revealed that some SCR variants exhibit a new or dual recognition specificity. From their results, Chookajorn et al. (2004) have proposed a model to explain the generation of new S-haplotypes in Brassica by the occurrence of selfincompatible intermediates that derived from the ancestor alleles by mutations affecting both SCR and SRK, but preserving allelic recognition. The work of Sato et al. (2004) based on the analysis of pairs of very homologous SCR/SRK sequences of B. rapa, B. oleracea and R. sativus supports this model. Once SRK phosphorylated, following the S-allele-specific interaction between SCR and SRK, the kinase domain of SRK recruits cytosolic targets to initiate a cascade of phosphorylation/dephosphorylation events responsible

286

T. Gaude et al.

for the self-pollen rejection by the stigma papilla. Among the cytosolic proteins of stigmatic cells, the Armadillo repeat-containing protein 1 (ARC1) was shown to bind specifically to the phosphorylated SRK kinase domain and to be phosphorylated in turn by SRK (Gu et al., 1998). In addition, ARC1 is known as a positive regulator of SI, because suppression of ARC1 transcripts in the pistil of self-incompatible B. napus plants is correlated with a partial breakdown of SI (Stone et al., 1999). More recently, Stone et al. (2003) demonstrated that ARC1 is an E3 ubiquitin ligose, which promotes the ubiquitination of stigma proteins during rejection of self-incompatible pollen grains. Moreover, inhibition of the proteasomal proteolytic activity on pollinated B. napus pistils disrupts the SI response. The authors propose that ARC1 promotes the ubiquitination and proteasomal degradation of compatibility factors (CFs) in the pistil, which are required for pollen germination and/or pollen tube growth, and that their degradation in turn leads to pollen rejection. Identification of these CFs remains the challenging step in elucidating the precise function of ARC1 in SI. Another protein, the kinase-associated protein phosphatase (KAPP), has also been shown to interact with the phosphorylated SRK kinase domain (Braun et al., 1997; Vanoosthuyse et al., 2003). However, this type 2C protein phosphatase interacts with various plant receptor kinases (PRKs) and is likely to have a more general role in downregulating PRK activity following ligand-binding receptor activation (Li et al., 1999; Shah et al., 2002; Tichtinsky et al., 2003). Interestingly, two novel interactors of the kinase domain of SRK have recently been isolated in B. oleracea, a calmodulin (CaM) and a sorting nexin (SNX1) protein (Vanoosthuyse et al., 2003). In animal cells, both CaM and sorting nexins are known regulators of receptor kinase-dependent signalling pathways, acting in the downregulation of the receptor kinase activity by promoting protein degradation via endocytosis (Tichtinsky et al., 2003). Like KAPP, CaM and SNX1 are able to bind to various PRKs. However, unlike KAPP, their binding is independent of the phosphorylation status of SRK. By analogy with the known functions of these proteins in animal systems, it is tempting to speculate that CaM and SNX1 function in downregulating PRK activity. Hence, they might control the duration and magnitude of signalling pathways initiated by PRKs during plant growth, development and reproduction. A new key component of the SI response in Brassica has recently been discovered by a Japanese group (Murase et al., 2004). They studied the recessive mutation modifier (m) that confers a strict self-compatibility phenotype in the B. rapa variety Yellow Sarson characterized about 20 years ago by Hinata et al. (1983). The M-locus is independent of and epistatic to the S-locus, and mm plants have a breakdown of the SI function in the stigma but not in the pollen. By positional cloning, Murase et al. (2004) identified the M-locus protein kinase (MLPK) as the M gene, and demonstrated that the m mutation is a single amino acid substitution (Gly to Arg) at position 194. MLPK is a membrane-anchored cytoplasmic serine/threonine protein kinase that belongs to the receptor-like cytoplasmic kinase VII (RLCK VII) subfamily, which has no apparent signal sequence or transmembrane domain (see Shiu and Bleecker, 2001). They showed that the single G194R mutation results in a loss of kinase

Control of Fertilization by Self-incompatibility Mechanisms

287

activity in vitro and a complete absence of detectable MLPK protein in stigma extracts from mm plants. Using MLPK fused to the green fluorescent protein (GFP) transiently expressed in tobacco BY-2 cells, they demonstrated that the wild-type form of MLPK–GFP is located at the plasma membrane, probably involving an N-myristoylation site. In addition, transient expression of wild-type MLPK in stigma papillae of mm plants restores the self-pollen rejection of the mm papillar cells. Together, these data clearly show that MLPK is a positive mediator of SI signalling in Brassica. Although the connection with the SRK signalling pathway remains unknown, it is likely that SRK and MLPK operate in concert to initiate the signalling pathway that leads to self-pollen rejection in the Brassicaceae (Fig. 10.3).

Molecular basis of dominance relationships In the self-incompatible Brassicaceae species, including Brassica species and A. lyrata, pollen SI specificity is determined by the diploid genotype of the pollen-producing parent rather than the genotype of individual haploid pollen grains. Genetic interactions between S-haplotypes are known to occur in the expression of SI phenotype in the stigma as well as in pollen. In Brassica species, co-dominance relationships are commonly observed in stigmas (Thompson and Taylor, 1966). In A. lyrata, classical genetic studies revealed a complex pattern of dominance relationships between seven haplotypes in the stigma, including at least three dominance levels with some co-dominant relationships between alleles at the same level (Schierup et al., 2001). Dominance relationships between alleles in the pollen can be different from dominance relationships in the stigma. For example, in Brassica species, S-haplotypes recessive in the pollen exhibited co-dominance in the stigma (Thompson and Taylor, 1966). Likewise, in A. lyrata (Kusaba et al., 2001, 2002) two S-haplotypes have been reported, which exhibited co-dominance in the stigma and dominance/recessiveness relationships in the pollen. The existence of independent dominance hierarchies in the stigma and pollen indicates that allelic interactions operate by different mechanisms in the two organs. This hypothesis has recently received some molecular support (Hatakeyama et al., 2001; Kusaba et al., 2002; Shiba et al., 2002). Dominance relationships in pollen are regulated at the mRNA level for SCR alleles, the expression of the pollen recessive allele being severely downregulated in heterozygotes containing a dominant allele. Interestingly, drastic inhibition of expression of the recessive SCR alleles has been reported both in A. lyrata (Kusaba et al., 2002) and in Brassica (Shiba et al., 2002). This indicates a common mechanism operating in both species to control allelic interactions in pollen. On the female side, a different mechanism may operate to control allelic interactions as the expression level of SRK gene cannot by itself account for the dominance relationships in Brassica (Hatakeyama et al., 2001) as well as in A. lyrata (Kusaba et al., 2002). The molecular basis of pollen recognition as well as more subtle controls of S-haplotype interactions influencing SI expression in the male and female partners seem to have been maintained in quite evolutionary distant species.

288

T. Gaude et al.

SCR2 Incompatible pollen

SCR1

?

P

THL

SNX1 ARC1 P CaM

P

SRK1

P

PM SRK1

MLPK

P

P ?

KAPP

P

SRK1

THL

KAPP

MLPK

SRK1

SRK1

KAPP

SRK1

Cellwall MLPK

THL

P

P

Pollen pollen rejection rejection

ARC1 CF Ub CF

Degradation

Stigmatic cell

26S Proteasome

Fig. 10.3. Model for the SRK-dependent mechanism of sporophytic self-incompatibility (SSI) in the Brassicaceae. Following self-pollination, the S-locus cysteine-rich (SCR) peptides carried by the pollen grain migrate from the pollen surface to the plasma membrane of the stigma papilla. SCR peptides that share the same S-allele specificity (here S1) as that of the S-locus receptor kinase (SRK) interact with the extracellular domain of SRK. SRK is part of a receptor complex located at the stigma plasma membrane. At the basal state, SRK is inactivated by thioredoxins (THL). The SRK1/SCR1 interaction releases the negative regulation of THL and activates the kinase domain of SRK. The cytosolic Armadillo repeat-containing (ARC1) protein 1, which is an E3-ubiquitin ligase, is then phosphorylated by SRK. It is proposed that the phosphorylated ARC1 recruits stigmatic compatibility factors (CFs) to promote their polyubiquitination and degradation via the 26S proteasome. The degradation of these still unknown factors would cause the arrest of pollen tube growth and failure of self-fertilization. The recently discovered M-locus protein kinase (MLPK) is supposed to be another substrate of the active SRK complex. MLPK is likely to participate in the signalling cascade leading to CF degradation. The kinase-associated protein phosphatase (KAPP) is probably involved in the negative regulation of the signalling pathway by dephosphorylating SRK. The role of calmodulin (CaM) and sorting nexin 1 (SNX1) remains to be determined; they might also be negative regulators of the SRK-dependent signalling pathway by analogy to their known functions in downregulating animal receptor kinases.

Control of Fertilization by Self-incompatibility Mechanisms

289

Arabidopsis thaliana: a new model for self-incompatibility studies? Because the S-locus is a master recognition locus that encodes the female (SRK) and male (SCR) SI determinants, inactivation of at least one of these SI-determining genes is predicted to lead to a breakdown of SI and to a shift to self-compatibility. In this context, it was particularly interesting to compare the S-locus structure in self-incompatible (e.g. Brassica species, A. lyrata) and selfcompatible species (e.g. A. thaliana). Comparative mapping of the S-locus in Brassica and in A. lyrata has unexpectedly revealed that the S-locus occupies different chromosomal locations in Brassica ssp. and in Arabidopsis ssp. (Conner et al., 1998; Kusaba et al., 2001). In Brassica, the S-locus is located in a region that is syntenous to an ETR1-linked chromosomal region on A. thaliana chromosome 1, whereas in A. lyrata the S-locus maps to the ARK3-containing region of chromosome 4. Location of the S-locus in Arabidopsis ssp. has allowed the identification of the orthologues of the SRK and SCR genes in A. thaliana. Both of them are non-functional genes in this selfcompatible species (Kusaba et al., 2001). Interestingly, transformation of A. thaliana with the SRK / SCR gene pair was sufficient to confer SI phenotype (Nasrallah et al., 2002, 2004). This indicates that the signalling cascade downstream of SRK activation and responsible for self-pollen rejection was maintained in A. thaliana, although it diverged from its SI A. lyrata relative about 5 million years ago. These data unequivocally demonstrate that the SC trait in A. thaliana has appeared as a consequence of the inactivation of the SRK and SCR genes. Moreover, the SRK gene has probably become inactivated recently because its promoter is still transcriptionally active (Kusaba et al., 2001). In contrast, previous attempts at generating a self-compatible A. thaliana by transforming with Brassica genes had failed (Bi et al., 2000). This failure might have been due to the inability of the Brassica SRK protein to interact properly with Arabidopsis downstream effectors. The availability of SI transgenic A. thaliana lines constitutes a real turning point in the study of SI. Indeed, it provides new and exciting opportunities for deciphering the biochemical machinery involved in self-pollen rejection by exploiting the numerous tools of this model plant.

Conclusion Tremendous progress has been made over the past few years in the molecular characterization of signal transduction pathways controlling SI in several plant families. Indeed, we now know the identity of the male and female determinants involved in self-pollen recognition in the Solanaceae, Rosaceae, Scrophulariaceae and Brassicaceae. It is likely that the still elusive pollen determinant of poppy will be identified soon. It is remarkable to note how different sophisticated molecular strategies have emerged during evolution to limit inbreeding and promote genetic mixing. With regard to the only very limited number of SI model systems analysed so far, we may assume that SI mechanisms, other than those based on the S-RNase/SLF, SRK / SCR or

290

T. Gaude et al.

S-glycoprotein-like poppy genes, exist among the single S-locus SI systems. It would be most interesting to initiate molecular studies in other families close to or distant from those already investigated. This would allow us to determine how many different SI systems evolved and whether they are phylogenetically linked. These data should give us clues to understand how these molecular mechanisms appeared during the evolution of angiosperms. From the recent data reviewed in this chapter, it is clear that self-pollen recognition constitutes a model of choice to study cell–cell communication in plants. New findings in the field of SI will have important implications for our general understanding of how plant cells communicate.

References Anderson, M.A., Cornish, E.C., Mau, S.L., Williams, E.G., Hoggart, R., Atkinson, A., Bonig, I., Grego, B., Simpson, R., Roche, P.J., Haley, J.D., Penschow, J.D., Niall, H.D., Tregear, G.W., Coghlan, J.P., Crawford, R.J. and Clarke, A.E. (1986) Cloning of cDNA for a stylar glycoprotein associated with expression of self-incompatibility in Nicotiana alata. Nature 321, 38–44. Bai, C., Sen, P., Hofmann, K., Ma, L., Goebl, M., Harper, J.W. and Elledge, S.J. (1996) SKP1 connects cell cycle regulators to the ubiquitin proteolysis machinery through a novel motif, the F-box. Cell 86, 263–274. Baumann, U., Juttner, J., Bian, X. and Langridge, P. (2000) Self-incompatibility in the grasses. Annals of Botany 85, 203–209. Bi, Y.M., Brugie`re, N., Cui, Y., Goring, D.R. and Rothstein, S.J. (2000) Transformation of Arabidopsis with a Brassica SLG / SRK region and ARC1 gene is not sufficient to transfer the selfincompatibility phenotype. Molecular and General Genetics 263, 648–654. Boyes, D.C. and Nasrallah, J.B. (1993) Physical linkage of the SLG and SRK genes at the self-incompatibility locus of Brassica oleracea. Molecular and General Genetics 236, 369–373. Braun, D.M., Stone, J.M. and Walker, J.C. (1997) Interaction of the maize and Arabidopsis kinase interaction

domains with a subset of receptor-like protein kinases: implications for transmembrane signaling in plants. Plant Journal 12, 83–95. Bredemeijer, G.M. and Blass, J. (1981) S-specific proteins in styles of self-incompatible Nicotiana alata. Theoretical and Applied Genetics 59, 185–190. Cabrillac, D., Delorme, V., Garin, J., Ruffio-Chable, V., Giranton, J.L., Dumas, C., Gaude, T. and Cock, J.M. (1999) The S15 self-incompatibility haplotype in Brassica oleracea includes three S gene family members expressed in stigmas. Plant Cell 11, 971–986. Cabrillac, D., Cock, J.M., Dumas, C. and Gaude, T. (2001) The S-locus receptor kinase is inhibited by thioredoxins and activated by pollen coat proteins. Nature 410, 220–223. Charlesworth, D., Bartolome, C., Schierup, M.H. and Mable, B.K. (2003) Haplotype structure of the stigmatic self-incompatibility gene in natural populations of Arabidopsis lyrata. Molecular Biology and Evolution 20, 1741–1753. Chookajorn, T., Kachroo, A., Ripoll, D.R., Clark, A.G. and Nasrallah, J.B. (2004) Specificity determinants and diversification of the Brassica selfincompatibility pollen ligand. Proceedings of the National Academy of Sciences USA 101, 911–917. Clark, A.G. and Kao, T.H. (1991) Excess nonsynonymous substitution of shared

Control of Fertilization by Self-incompatibility Mechanisms polymorphic sites among self-incompatibility alleles of Solanaceae. Proceedings of the National Academy of Sciences USA 88, 9823–9827. Cock, J.M., Swarup, R. and Dumas, C. (1997) Natural antisense transcripts of the S locus receptor kinase gene and related sequences in Brassica oleracea. Molecular and General Genetics 255, 514–524. Cock, J.M., Vanoosthuyse, V. and Gaude, T. (2002) Receptor kinase signalling in plants and animals: distinct molecular systems with mechanistic similarities. Current Opinion in Cell Biology 14, 230–236. Conner, J.A., Conner, P., Nasrallah, M.E. and Nasrallah, J.B. (1998) Comparative mapping of the Brassica S locus region and its homeolog in Arabidopsis: implications for the evolution of mating systems in the Brassicaceae. Plant Cell 10, 801–812. Cui, Y., Brugiere, N., Jackman, L., Bi, Y.M. and Rothstein, S.J. (1999) Structural and transcriptional comparative analysis of the S locus regions in two self-incompatible Brassica napus lines. Plant Cell 11, 2217– 2231. Darwin, C. (1876) The Effect of Cross and Self-fertilization in the Vegetable Kingdom. John Murray, London. Delorme, V., Giranton, J.L., Hatzfeld, Y., Friry, A., Heizmann, P., Ariza, M.J., Dumas, C., Gaude, T. and Cock, J.M. (1995) Characterization of the S locus genes, SLG and SRK, of the Brassica S3 haplotype: identification of a membrane-localized protein encoded by the S locus receptor kinase gene. Plant Journal 7, 429– 440. de Nettancourt, D. (2001) Incompatibility and Incongruity in Wild and Cultivated Plants, 2nd edn. SpringerVerlag, Berlin/Heidelberg. East, E.M. and Mangelsdorf, A.J. (1925) A new interpretation of the hereditary behavior of self-sterile plants. Pro-

291

ceedings of the National Academy of Sciences USA 11, 166–171. Entani, T., Iwano, M., Shiba, H., Che, F.S., Isogai, A. and Takayama, S. (2003) Comparative analysis of the self-incompatibility (S-)locus region of Prunus mume: identification of a pollen-expressed F-box gene with allelic diversity. Genes to Cells 8, 203–213. Fobis-Loisy, I., Mie`ge, C. and Gaude, T. (2004) Molecular evolution of the S locus controlling mating in the Brassicaceae. Plant Biology 6, 109–118. Foote, H.C., Ride, J.P., Franklin-Tong, V.E., Walker, E.A., Lawrence, M.J. and Franklin, F.C. (1994) Cloning and expression of a distinctive class of selfincompatibility (S) gene from Papaver rhoeas L. Proceedings of the National Academy of Sciences USA 91, 2265–2269. Franklin-Tong, V.E. and Franklin, F.C. (2003) Gametophytic self-incompatibility inhibits pollen tube growth using different mechanisms. Trends in Plant Science 8, 598–605. Franklin-Tong, V.E., Holdaway-Clarke, T.L., Straatman, K.R., Kunkel, J.G. and Hepler, P.K. (2002) Involvement of extracellular calcium influx in the self-incompatibility response of Papaver rhoeas. Plant Journal 29, 333– 345. Gaude, T. and Dumas, C. (1987) Molecular and cellular events of self-incompatibility. In: Giles, K. and Prakash, J. (eds) Pollen: Cytology and Development. International Review of Cytology 107, Academic Press, Orlando, Florida, pp. 333–366. Gaude, T., Friry, A., Heizmann, P., Mariac, C., Rougier, M., Fobis, I. and Dumas, C. (1993) Expression of a self-incompatibility gene in a selfcompatible line of Brassica oleracea. Plant Cell 5, 75–86. Gaude, T., Rougier, M., Heizmann, P., Ockendon, D.J. and Dumas, C. (1995) Expression level of the SLG gene is not correlated with the

292

T. Gaude et al. self-incompatibility phenotype in the class II S haplotypes of Brassica oleracea. Plant Molecular Biology 27, 1003–1014. Giranton, J.L., Ariza, M.J., Dumas, C., Cock, J.M. and Gaude, T. (1995) The S locus receptor kinase gene encodes a soluble glycoprotein corresponding to the SKR extracellular domain in Brassica oleracea. Plant Journal 8, 827– 834. Giranton, J.L., Dumas, C., Cock, J.M. and Gaude, T. (2000) The integral membrane S-locus receptor kinase of Brassica has serine/threonine kinase activity in a membranous environment and spontaneously forms oligomers in planta. Proceedings of the National Academy of Sciences USA 97, 3759– 3764. Goring, D.R. and Rothstein, S.J. (1992) The S-locus receptor kinase gene in a self-incompatible Brassica napus line encodes a functional serine/threonine kinase. Plant Cell 4, 1273–1281. Goring, D.R., Glavin, T.L., Schafer, U. and Rothstein, S.J. (1993) An S receptor kinase gene in self-compatible Brassica napus has a 1-bp deletion. Plant Cell 5, 531–539. Gu, T., Mazzurco, M., Sulaman, W., Matias, D.D. and Goring, D.R. (1998) Binding of an arm repeat protein to the kinase domain of the S-locus receptor kinase. Proceedings of the National Academy of Sciences USA 95, 382– 387. Haffani, Y.Z., Gaude, T., Cock, J.M. and Goring, D.R. (2004) Antisense suppression of thioredoxin h mRNA in Brassica napus cv. Westar pistils causes a low level constitutive pollen rejection response. Plant Molecular Biology 55, 619–630. Hatakeyama, K., Takasaki, T., Suzuki, G., Nishio, T., Watanabe, M., Isogai, A. and Hinata, K. (2001) The S receptor kinase gene determines dominance relationships in stigma expression of selfincompatibility in Brassica. Plant Journal 26, 69–76.

Hearn, M.J., Franklin, F.C.H. and Ride, J.P. (1996) Identification of a membrane glycoprotein in pollen of Papaver rhoeas which binds stigmatic selfincompatibility (S-) proteins. Plant Journal 9, 467–475. Heldin, C.H. (1995) Dimerization of cell surface receptors in signal transduction. Cell 80, 213–223. Hinata, K., Okazaki, K. and Nishio, T. (1983) Gene analysis of self-compatibility in Brassica campestris var yellow sarson (a case of recessive epistatic modifier). In Proceedings of the 6th International Rapeseed Conference, Paris, pp. 354–359. Hiscock, S.J. and McInnis, S.M. (2003) Pollen recognition and rejection during the sporophytic self-incompatibility response: Brassica and beyond. Trends in Plant Science 8, 606–613. Huang, S., Lee, H.S., Karunanandaa, B. and Kao, T.H. (1994) Ribonuclease activity of Petunia inflata S proteins is essential for rejection of self-pollen. Plant Cell 6, 1021–1028. Huang, S., Blanchoin, L., Chaudhry, F., Franklin-Tong, V.E. and Staiger, C.J. (2004) A gelsolin-like protein from Papaver rhoeas pollen (PrABP80) stimulates calcium-regulated severing and depolymerization of actin filaments. Journal of Biological Chemistry 279, 23364–23375. Ida, K., Norioka, S., Yamamoto, M., Kumasaka, T., Yamashita, E., Newbigin, E., Clarke, A.E., Sakiyama, F. and Sato, M. (2001) The 1.55 A resolution structure of Nicotiana alata S(F11)RNase associated with gametophytic self-incompatibility. Journal of Molecular Biology 314, 103–112. Ikeda, K., Igic, B., Ushijima, K., Yamane, H., Hauck, N.R., Nakano, R., Sassa, H., Iezzoni, A.F., Kohn, J.R. and Tao, R. (2004) Primary structural features of the S haplotype-specific F-box protein, SFB, in Prunus. Sexual Plant Reproduction 16, 235–243. Ioerger, T.R., Clark, A.G. and Kao, T.H. (1990) Polymorphism at the self-

Control of Fertilization by Self-incompatibility Mechanisms incompatibility locus in Solanaceae predates speciation. Proceedings of the National Academy of Sciences USA 87, 9732–9735. Ishimizu, T., Shinkawa, T., Sakiyama, F. and Norioka, S. (1998) Primary structural features of rosaceous S-RNases associated with gametophytic self-incompatibility. Plant Molecular Biology 37, 931–941. Iwano, M., Shiba, H., Funato, M., Shimosato, H., Takayama, S. and Isogai, A. (2003) Immunohistochemical studies on translocation of pollen S-haplotype determinant in self-incompatibility of Brassica rapa. Plant and Cell Physiology 44, 428–436. Jordan, N.D., Kakeda, K., Conner, A., Ride, J.P., Franklin-Tong, V.E. and Franklin, F.C. (1999) S-protein mutants indicate a functional role for SBP in the self-incompatibility reaction of Papaver rhoeas. Plant Journal 20, 119–125. Kachroo, A., Schopfer, C.R., Nasrallah, M.E. and Nasrallah, J.B. (2001) Allele-specific receptor-ligand interactions in Brassica self-incompatibility. Science 293, 1824–1826. Kachroo, A., Nasrallah, M.E. and Nasrallah, J.B. (2002) Self-incompatibility in the Brassicaceae: receptor-ligand signaling and cell-to-cell communication. Plant Cell 14, S227–238. Kao, T.H. and McCubbin, A.G. (1996) How flowering plants discriminate between self and non-self pollen to prevent inbreeding. Proceedings of the National Academy of Sciences USA 93, 12059–12065. Kao, T.H. and Tsukamoto, T. (2004) The molecular and genetic bases of S-RNase-based self-incompatibility. Plant Cell 16, S72–83. Karunanandaa, B., Huang, S. and Kao, T. (1994) Carbohydrate moiety of the Petunia inflata S3 protein is not required for self-incompatibility interactions between pollen and pistil. Plant Cell 6, 1933–1940. Kleman-Mariac, C., Rougier, M., Cock, J.M., Gaude, T. and Dumas, C.

293

(1995) S-locus glycoproteins are expressed along the path of pollen tubes in Brassica pistils. Planta 196, 614– 621. Kusaba, M., Nishio, T., Satta, Y., Hinata, K. and Ockendon, D. (1997) Striking sequence similarity in inter- and intraspecific comparisons of class I SLG alleles from Brassica oleracea and Brassica campestris: implications for the evolution and recognition mechanism. Proceedings of the National Academy of Sciences USA 94, 7673–7678. Kusaba, M., Matsushita, M., Okazaki, K., Satta, Y. and Nishio, T. (2000) Sequence and structural diversity of the S locus genes from different lines with the same self-recognition specificities in Brassica oleracea. Genetics 154, 413–420. Kusaba, M., Dwyer, K., Hendershot, J., Vrebalov, J., Nasrallah, J.B. and Nasrallah, M.E. (2001) Self-incompatibility in the genus Arabidopsis: characterization of the S locus in the outcrossing Arabidopsis lyrata and its autogamous relative Arabidopsis thaliana. Plant Cell 13, 627–643. Kusaba, M., Tung, C.W., Nasrallah, M.E. and Nasrallah, J.B. (2002) Monoallelic expression and dominance interactions in anthers of self-incompatible Arabidopsis lyrata. Plant Physiology 128, 17–20. Lai, Z., Ma, W., Han, B., Liang, L., Zhang, Y., Hong, G. and Xue, Y. (2002) An F-box gene linked to the self-incompatibility (S) locus of Antirrhinum is expressed specifically in pollen and tapetum. Plant Molecular Biology 50, 29–42. Lee, H.S., Huang, S. and Kao, T. (1994) S proteins control rejection of incompatible pollen in Petunia inflata. Nature 367, 560–563. Li, J., Smith, G.P. and Walker, J.C. (1999) Kinase interaction domain of kinase-associated protein phosphatase, a phosphoprotein-binding domain. Proceedings of the National

294

T. Gaude et al. Academy of Sciences USA 96, 7821–7826. Luu, D.T., Qin, X., Morse, D. and Cappadocia, M. (2000) S-RNase uptake by compatible pollen tubes in gametophytic self-incompatibility. Nature 407, 649–651. Matsuura, T., Sakai, H., Unno, M., Ida, K., Sato, M., Sakiyama, F. and Norioka, S. (2001) Crystal structure at 1.5A resolution of Pyrus pyrifolia pistil ribonuclease responsible for gametophytic self-incompatibility. Journal of Biological Chemistry 276, 45261– 45269. Matton, D.P., Maes, O., Laublin, G., Xike, Q., Bertrand, C., Morse, D. and Cappadocia, M. (1997) Hypervariable domains of self-incompatibility RNAses mediate allele-specific pollen recognition. Plant Cell 9, 1757– 1766. McClure, B.A., Haring, V., Ebert, P.R., Anderson, M.A., Simpson, R.J., Sakiyama, F. and Clarke, A.E. (1989) Style self-incompatibility gene products of Nicotiana alata are ribonucleases. Nature 342, 955–957. Mie`ge, C., Dumas, C. and Cock, J.M. (1999) Identification of a gene linked to the Brassica S (self-incompatibility) locus by differential display. Compte Rendu de l’Acade´mie des Sciences III 322, 1051–1060. Mie`ge, C., Ruffio-Chable, V., Schierup, M.H., Cabrillac, D., Dumas, C., Gaude, T. and Cock, J.M. (2001) Intrahaplotype polymorphism at the Brassica S locus. Genetics 159, 811–822. Mishima, M., Takayama, S., Sasaki, K., Jee, J.G., Kojima, C., Isogai, A. and Shirakawa, M. (2003) Structure of the male determinant factor for Brassica self-incompatibility. Journal of Biological Chemistry 278, 36389– 36395. Murase, K., Shiba, H., Iwano, M., Che, F.S., Watanabe, M., Isogai, A. and Takayama, S. (2004) A membraneanchored protein kinase involved in

Brassica self-incompatibility signaling. Science 303, 1516–1519. Murfett, J., Atherton, T.L., Mou, B., Gasser, C.S. and McClure, B.A. (1994) S-RNase expressed in transgenic Nicotiana causes S-allele-specific pollen rejection. Nature 367, 563–566. Nasrallah, J.B., Kao, T.-H., Goldberg, M.L. and Nasrallah, M.E. (1985) A cDNA clone encoding an S-locus specific glycoprotein from Brassica oleracea. Nature 318, 617–618. Nasrallah, J.B., Rundle, S.J. and Nasrallah, M.E. (1994) Genetic evidence for the requirement of Brassica S-locus receptor kinase gene in self-incompatibility response. Plant Journal 5, 373– 384. Nasrallah, M.E., Liu, P. and Nasrallah, J.B. (2002) Generation of self-incompatible Arabidopsis thaliana by transfer of two S locus genes from Arabidopsis lyrata. Science 297, 247–249. Nasrallah, M.E., Liu, P., ShermanBroyles, S., Boggs, N.A. and Nasrallah, J.B. (2004) Natural variation in expression of self-incompatibility in Arabidopsis thaliana: implications for the evolution of selfing. Proceedings of the National Academy of Sciences USA 101, 16070–16074. Nishio, T. and Kusaba, M. (2000) Sequence diversity of SLG and SRK in Brassica oleracea L. Annals of Botany 85, 141–146. Okazaki, K., Kusaba, M., Ockendon, D. and Nishio, T. (1999) Characterization of S tester lines in Brassica oleracea: polymorphism of restriction fragment length of SLG homologues and isoelectric points of S-locus glycoproteins. Theoretical and Applied Genetics 98, 1329–1334. Pastuglia, M., Ruffio-Chable, V., Delorme, V., Gaude, T., Dumas, C. and Cock, J.M. (1997) A functional S locus anther gene is not required for the self-incompatibility response in Brassica oleracea. Plant Cell 9, 2065–2076.

Control of Fertilization by Self-incompatibility Mechanisms Qiao, H., Wang, F., Zhao, L., Zhou, J., Lai, Z., Zhang, Y., Robbins, T.P. and Xue, Y. (2004a) The F-box protein AhSLF-S2 controls the pollen function of S-RNase-based self-incompatibility. Plant Cell 16, 2307–2322. Qiao, H., Wang, H., Zhao, L., Zhou, J., Huang, J., Zhang, Y. and Xue, Y. (2004b) The F-box protein AhSLF-S2 physically interacts with S-RNases that may be inhibited by the ubiquitin/26S proteasome pathway of protein degradation during compatible pollination in Antirrhinum. Plant Cell 16, 582– 595. Roalson, E.H. and McCubbin, A.G. (2003) S-RNases and sexual incompatibility: structure, functions, and evolutionary perspectives. Molecular Phylogenetics and Evolution 29, 490–506. Rudd, J.J. and Franklin-Tong, V.E. (2003) Signals and targets of the self-incompatibility response in pollen of Papaver rhoeas. Journal of Experimental Botany 54, 141–148. Rudd, J.J., Franklin, F., Lord, J.M. and Franklin-Tong, V.E. (1996) Increased phosphorylation of a 26-kD pollen protein is induced by the self-incompatibility response in Papaver rhoeas. Plant Cell 8, 713–724. Rudd, J.J., Osman, K., Franklin, F.C. and Franklin-Tong, V.E. (2003) Activation of a putative MAP kinase in pollen is stimulated by the self-incompatibility (SI) response. FEBS Letters 547, 223–227. Ruffio-Chable, V. and Gaude, T. (2001) Shaplotype polymorphism in Brassica oleracea. Acta Horticulturae 546, 257–261. Sato, Y., Okamoto, S. and Nishio, T. (2004) Diversification and alteration of recognition specificity of the pollen ligand SP11/SCR in self-incompatibility of Brassica and Raphanus. Plant Cell 16, 3230–3241. Schierup, M.H., Mable, B.K., Awadalla, P. and Charlesworth, D. (2001) Identification and characterization of a poly-

295

morphic receptor kinase gene linked to the self-incompatibility locus of Arabidopsis lyrata. Genetics 158, 387– 399. Schopfer, C.R., Nasrallah, M.E. and Nasrallah, J.B. (1999) The male determinant of self-incompatibility in Brassica. Science 286, 1697–1700. Shah, K., Russinova, E., Gadella, T.W. Jr, Willemse, J. and De Vries, S.C. (2002) The Arabidopsis kinase-associated protein phosphatase controls internalization of the somatic embryogenesis receptor kinase 1. Genes & Development 16, 1707–1720. Shiba, H., Takayama, S., Iwano, M., Shimosato, H., Funato, M., Nakagawa, T., Che, F.S., Suzuki, G., Watanabe, M., Hinata, K. and Isogai, A. (2001) A pollen coat protein, SP11/SCR, determines the pollen S-specificity in the self-incompatibility of Brassica species. Plant Physiology 125, 2095–2103. Shiba, H., Iwano, M., Entani, T., Ishimoto, K., Shimosato, H., Che, F.S., Satta, Y., Ito, A., Takada, Y., Watanabe, M., Isogai, A. and Takayama, S. (2002) The dominance of alleles controlling self-incompatibility in Brassica pollen is regulated at the RNA level. Plant Cell 14, 491–504. Shiba, H., Kenmochi, M., Sugihara, M., Iwano, M., Kawasaki, S., Suzuki, G., Watanabe, M., Isogai, A. and Takayama, S. (2003) Genomic organization of the S-locus region of Brassica. Bioscience, Biotechnology, and Biochemistry 67, 622–626. Shimizu, K.K., Cork, J.M., Caicedo, A.L., Mays, C.A., Moore, R.C., Olsen, K.M., Ruzsa, S., Coop, G., Bustamante, C.D., Awadalla, P. and Purugganan, M.D. (2004) Darwinian selection on a selfing locus. Science 306, 2081–2084. Shiu, S.H. and Bleecker, A.B. (2001) Receptor-like kinases from Arabidopsis form a monophyletic gene family related to animal receptor kinases. Proceedings of the National Academy of Sciences USA 98, 10763–10768.

296

T. Gaude et al. Sijacic, P., Wang, X., Skirpan, A.L., Wang, Y., Dowd, P.E., McCubbin, A.G., Huang, S. and Kao, T.H. (2004) Identification of the pollen determinant of S-RNase-mediated self-incompatibility. Nature 429, 302–305. Sims, T.L. and Ordanic, M. (2001) Identification of a S-ribonuclease-binding protein in Petunia hybrida. Plant Molecular Biology 47, 771–783. Snowman, B.N., Kovar, D.R., Shevchenko, G., Franklin-Tong, V.E. and Staiger, C.J. (2002) Signal-mediated depolymerization of actin in pollen during the self-incompatibility response. Plant Cell 14, 2613–2626. Sonneveld, T., Tobutt, K.R., Vaughan, S.P. and Robbins, T.P. (2005) Loss of pollen-S function in two self-compatible selections of Prunus avium is associated with deletion/mutation of an S haplotype-specific F-box gene. Plant Cell 17, 37–51. Stein, J.C. and Nasrallah, J.B. (1993) A plant receptor-like gene, the S-locus receptor kinase of Brassica oleracea L., encodes a functional serine/threonine kinase. Plant Physiology 101, 1103–1106. Stein, J.C., Howlett, B., Boyes, D.C., Nasrallah, M.E. and Nasrallah, J.B. (1991) Molecular cloning of a putative receptor protein kinase gene encoded at the self-incompatibility locus of Brassica oleracea. Proceedings of the National Academy of Sciences USA 88, 8816–8820. Stein, J.C., Dixit, R., Nasrallah, M.E. and Nasrallah, J.B. (1996) SRK, the stigma-specific S locus receptor kinase of Brassica, is targeted to the plasma membrane in transgenic tobacco. Plant Cell 8, 429–445. Steinbachs, J.E. and Holsinger, K.E. (2002) S-RNase-mediated gametophytic self-incompatibility is ancestral in eudicots. Molecular Biology and Evolution 19, 825–829. Stephenson, A.G., Doughty, J., Dixon, S., Elleman, C., Hiscock, S. and Dickinson, H.G. (1997) The male de-

terminant in self-incompatibility in Brassica oleracea is located in the pollen coating. Plant Journal 12, 1351– 1359. Stone, S.L., Arnoldo, M. and Goring, D.R. (1999) A breakdown of Brassica self-incompatibility in ARC1 antisense transgenic plants. Science 286, 1729– 1731. Stone, S.L., Anderson, E.M., Mullen, R.T. and Goring, D.R. (2003) ARC1 is an E3 ubiquitin ligase and promotes the ubiquitination of proteins during the rejection of self-incompatible Brassica pollen. Plant Cell 15, 885–898. Suzuki, G., Kai, N., Hirose, T., Fukui, K., Nishio, T., Takayama, S., Isogai, A., Watanabe, M. and Hinata, K. (1999) Genomic organization of the S locus: identification and characterization of genes in SLG/SRK region of S(9) haplotype of Brassica campestris (syn. rapa). Genetics 153, 391–400. Takasaki, T., Hatakeyama, K., Suzuki, G., Watanabe, M., Isogai, A. and Hinata, K. (2000) The S receptor kinase determines self-incompatibility in Brassica stigma. Nature 403, 913–916. Takayama, S., Shiba, H., Iwano, M., Shimosato, H., Che, F.S., Kai, N., Watanabe, M., Suzuki, G., Hinata, K. and Isogai, A. (2000) The pollen determinant of self-incompatibility in Brassica campestris. Proceedings of the National Academy of Sciences USA 97, 1920–1925. Takayama, S., Shimosato, H., Shiba, H., Funato, M., Che, F.S., Watanabe, M., Iwano, M. and Isogai, A. (2001) Direct ligand-receptor complex interaction controls Brassica self-incompatibility. Nature 413, 534–538. Tantikanjana, T., Nasrallah, M.E., Stein, J.C., Chen, C.H. and Nasrallah, J.B. (1993) An alternative transcript of the S locus glycoprotein gene in a class II pollen-recessive self-incompatibility haplotype of Brassica oleracea encodes a membrane-anchored protein. Plant Cell 5, 657–666.

Control of Fertilization by Self-incompatibility Mechanisms Thomas, S.G. and Franklin-Tong, V.E. (2004) Self-incompatibility triggers programmed cell death in Papaver pollen. Nature 429, 305–309. Thompson, K.F. and Taylor, J.P. (1966) Non-linear relationships between S alleles. Heredity 21, 345–362. Tichtinsky, G., Vanoosthuyse, V., Cock, J.M. and Gaude, T. (2003) Making inroads into plant receptor kinase signalling pathways. Trends in Plant Science 8, 231–237. Tsai, D.-S., Lee, H.-S., Post, L.C., Kreiling, K.M. and Kao, T.H. (1992) Sequence of an S-protein of Lycopersicon peruvianum and comparison with other solanaceous S-proteins. Sexual Plant Reproduction 5, 256– 263. Ushijima, K., Sassa, H., Dandekar, A.M., Gradziel, T.M., Tao, R. and Hirano, H. (2003) Structural and transcriptional analysis of the self-incompatibility locus of almond: identification of a pollen-expressed F-box gene with haplotype-specific polymorphism. Plant Cell 15, 771–781. Vanoosthuyse, V., Mie`ge, C., Dumas, C. and Cock, J.M. (2001) Two large Arabidopsis thaliana gene families are homologous to the Brassica gene superfamily that encodes pollen coat proteins and the male component of the self-incompatibility response. Plant Molecular Biology 46, 17–34.

297

Vanoosthuyse, V., Tichtinsky, G., Dumas, C., Gaude, T. and Cock, J.M. (2003) Interaction of calmodulin, a sorting nexin and kinase-associated protein phosphatase with the Brassica oleracea S locus receptor kinase. Plant Physiology 133, 919–929. Verica, J.A., McCubbin, A.G. and Kao, T. (1998) Are the hypervariable regions of S RNases sufficient for allele-specific recognition of pollen? Plant Cell 10, 314–317. Vieira, C.P., Charlesworth, D. and Vieira, J. (2003) Evidence for rare recombination at the gametophytic selfincompatibility locus. Heredity 91, 262–267. Weller, S.G., Donoghue, M.J. and Charlesworth, D. (1995) The evolution of self-incompatibility in flowering plants: a phylogenetic approach. In: Hoch, P.C. and Stephenson, A.G. (eds) Experimental and Molecular Approaches to Plant Biosystematics. Missouri Botanical Garden, St Louis, Missouri, pp. 355–382. Xue, Y., Carpenter, R., Dickinson, H.G. and Coen, E.S. (1996) Origin of allelic diversity in Antirrhinum S locus RNases. Plant Cell 8, 805–814. Yu, K., Schafer, U., Glavin, T.L., Goring, D.R. and Rothstein, S.J. (1996) Molecular characterization of the S locus in two self-incompatible Brassica napus lines. Plant Cell 8, 2369–2380.

11

Stamen Development: Primordium to Pollen

R.J. SCOTT1, M. SPIELMAN2 AND H.G. DICKINSON1 1

Department of Biology and Biochemistry, University of Bath, Bath BA2 7AY, UK, e-mail: [email protected] 2 Department of Plant Sciences, South Parks Road, Oxford, OX1 3RB, UK

Introduction The stamen is the male reproductive organ of flowering plants and consists of an anther, the site of pollen development, and in most species a stalk-like filament that transmits water and nutrients to the anther and positions it to aid pollen dispersal. Within the anther, male sporogenous cells differentiate and undergo meiosis to produce microspores, which give rise to pollen grains, while other cell types contribute to pollen maturation or release. Pollen development involves numerous extraordinary events, some of which are either unique or else restricted to reproductive development, including cell division and differentiation independent of a meristem; the transition from sporophytic to gametophytic generation; and modifications of cell division to produce structures that are unusual in plant development, including coenocytic tissues (the tapetum and the microsporocyte mass), and subsequently free cells (microspores) that give rise to self-contained units of dispersal, the pollen grains. A comprehensive review of stamen morphology across the angiosperms is provided by Bhandari (1984), and for Arabidopsis thaliana by Scott et al. (2004). Other reviews of stamen development, function and gene expression include Scott et al. (1991a), Goldberg et al. (1993) and Irish (1999). This chapter focuses mainly on recent advances in understanding stamen development and function in A. thaliana. This model plant species has unsurpassed experimental resources such as freely available genetic tools, a fully sequenced genome, microarrays, well-characterized mutants associated with cloned genes and ever more comprehensive knockout collections for reverse genetics. The advances of the last 12 years since the first edition of The Molecular Biology of Flowering include an increased understanding of stamen specification, stamen-specific gene expression, internal patterning of the anther, 298

ßCAB International 2006. The Molecular Biology and Biotechnology of Flowering, 2nd edn (ed. B.R. Jordan)

Stamen Development

299

regulation of male meiosis, the role of the callose wall in the production of free microspores and in the patterning of the pollen wall and anther dehiscence.

The Specification of Stamens Within the Floral Meristem Flowers of eudicots are organized into four concentric whorls of organs (sepals, petals, stamens, carpels), which arise sequentially from the floral meristem (Fig. 11.1). The third whorl in A. thaliana flowers contains six stamens, four long medials (long) and two shorter laterals. Stamen primordia appear during floral stage 5 (defined by Smyth et al., 1990), with the medial stamens arising first. At stage 7 the long stamen primordia differentiate a stalked basal region, which will give rise to the filament, and a wider upper region that will become the anther. Stage 8 is defined by the appearance of anther locules (in which pollen develops) as convex protrusions on the inner (adaxial) surface of long stamens. During this stage the sporogenous cells, which will give rise to pollen, are visible within locules of sectioned anthers (Sanders et al., 1999). The sporogenous cells differentiate into microsporocytes (also known as pollen mother cells or male meiocytes), which undergo meiosis to form tetrads of haploid microspores (Fig. 11.1). These are released into the anther locule to begin male gametophyte A

Long stamen

Short stamen

B

Sepal

T Petal

Abaxial surface

ML

Short stamen

C

V

StR

S

Td

Microsporangium

Adaxial surfaceAnther stage 7 Floral stage 9

Gynoecium

E En

D

E En

C C

V

St TPG TPG Anther stage 14 Floral stage 14

En E Anther stage 12 Floral stage 12

Fig. 11.1. Anther development. (A) Schematic representation of a transverse section through Arabidopsis thaliana floral bud showing number, position and orientation of the floral organs (after Hill and Lord, 1989). (B–D). Schematic representation of transverse sections through A. thaliana anthers at different stages (after Sanders et al. 1999). Floral stages as in Smyth et al. (1990); anther stages as in Sanders et al. (1999). C, connective; E, epidermis; En, endothecium; ML, middle layer; S, septum; St, stomium; StR, stomium region; T, tapetum; Td, tetrads; TPG, tricellular pollen grains; V, vascular bundle.

300

R.J. Scott et al.

development. By floral stage 12 the anthers are nearly at their final length and contain tricellular pollen, and the filaments begin to elongate rapidly. At stage 13 the flower opens and anther dehiscence occurs to release the pollen. The filaments continue to extend, so that the anthers brush past the receptive stigma at stage 14. Jenik and Irish (2000) studied the patterns of cell divisions involved in floral organ development. They found that the floral meristem of A. thaliana, like the shoot apical meristem (SAM), is composed of three ‘histogenic layers’ of cells with separate lineages: L1 (epidermis), L2 (subepidermis) and L3 (core). Stamen primordia are initiated by periclinal divisions in the floral meristem, usually within the L2. By stage 7 the primordia are composed of an epidermis derived from the L1, one layer of L2-derived subepidermis and an L3 core. The growth of the anther after stage 8 is due mainly to division of L2 cells in different planes; these give rise to most of the cell types of the anther, including the sporogenous cells (discussed in detail below). The L3 contributes to the vasculature and sometimes to the connectives. In other species, it has also been observed that the male germ cells derive almost exclusively from the L2 (Goldberg et al., 1993). There are many reviews of floral organ specification (e.g. Jack, 2001, 2004; Theißen, 2001; Lohmann and Weigel, 2002) and therefore regulation of stamen identity will be treated very briefly here. The identity of floral organ primordia is controlled by three classes (termed A, B and C) of homeotic selector genes with overlapping areas of activity; stamens develop in the third whorl, where both B and C genes are expressed. In stamens (as well as petals and carpels) additional transcription factors, encoded by SEPALLATA (SEP) genes in A. thaliana, are required to confer full activity on the ABC genes. The ABC model was based on genetic studies in A. thaliana and Antirrhinum majus (Bowman et al., 1991; Coen and Meyerowitz, 1991). The B class genes are APETALA3 (AP3) and PISTILLATA (PI ) in A. thaliana and DEFICIENS (DEF) and GLOBOSA (GLO) in A. majus; the C genes are AGAMOUS (AG) and PLENA (PLE), respectively. Mutations in any one of the B or C genes result in homeotic conversion of the third-whorl organs to a different type. In both A. thaliana and A. majus, loss of B function causes transformation of stamens to carpelloid organs, loss of C function converts stamens to petals and loss of both transforms stamens to sepals (Bowman et al., 1989; Carpenter and Coen, 1990). Single gene mutations of the SEP genes have only very subtle phenotypes, but in sep1 sep2 sep3 triple mutants all floral organs resemble sepals, suggesting that B and C functions have been abolished (Pelaz et al., 2000). The signals that determine the number of floral organ primordia in each whorl are unknown (Irish, 1999). However, the number of stamens can be perturbed by various mechanisms; extra stamens develop in clavata1 mutants, which form enlarged meristems (Clark et al., 1993), and in superman mutants, which have an expanded BC domain (Bowman et al., 1992). In contrast, arrest of stamen primordia is a feature of normal development in some species. In the radially asymmetric flowers of Antirrhinum, five stamens initiate but the dorsal stamen arrests early in development in response to activity of the cycloidea gene (Luo et al., 1995). In some species that produce unisexual flowers, such as maize and white campion, stamens in female flowers initiate but later arrest or abort (Calderon-Urrea and Dellaporta, 1999; Tanurdzik and Banks, 2004).

Stamen Development

301

Although the ABC model was developed for eudicots, it is also applicable to monocots and basal dicots with some modifications (reviewed by Theißen, 2001; Fornara et al., 2003; Zik and Irish, 2003). Throughout the angiosperms, stamens are specified by a combination of B class proteins that are orthologous to AP3 and PI, and C class proteins that are either orthologous or paralogous to AG. Strikingly, orthologues of B class genes isolated from gymnosperms (which do not form stamens) are expressed exclusively in male reproductive structures, suggesting that the role of B genes in specifying male reproductive development predates the divergence of gymnosperm and angiosperm lineages (Theißen, 2001). The B class, C class and SEP proteins all belong to the MADS family of transcription factors, which bind a target DNA sequence, CC(A / T)6CG (CArG box), as homo- or heterodimers. In vitro, the A. majus B class proteins DEF and GLO bind CArG box sequences only as a heterodimer (Theißen, 2001). Similarly, A. thaliana AP3 and PI bind CArG box sequences only as a heterodimer, and the C class AG protein binds as a homodimer (Riechmann et al., 1996). Yeast two-hybrid assays show that PI/AP3 interacts directly with SEP3 but not AG, and it has been proposed that SEP3 mediates interactions between PI/AP3 and AG dimers so that PI/AP3, AG and SEP bind DNA in quaternary complexes; this would explain the combinatorial action of B and C class genes (along with SEP genes) in stamens (Honma and Goto, 2001).

Identifying Genes Involved in Stamen Development and Function Twelve years ago one of the major unanswered questions about stamen development was the nature of the genes downstream of the B and C class transcription factors (Goldberg et al., 1993). After stamen specification, the B and C class as well as SEP genes continue to be expressed during stamen development (Bowman et al., 1991; Pelaz et al., 2000; Jack, 2001), so could be directly responsible for activating many of the genes involved in stamen morphogenesis and function. Among the targets of B class genes in A. thaliana are AP3 and PI themselves, as both genes are required for the continued expression of each in the developing flower (Lohmann and Weigel, 2002). Two main approaches have been used to identify further targets of B and C class genes. These involve either transcriptional profiling methods including subtractive hybridization, differential display of RNAs and differential screening of cDNA libraries or arrays to find genes with stamen-specific or -preferred expression or screening for mutations that affect stamens. Both approaches have yielded interesting results, but there is not yet enough information to reconstruct the developmental pathways that start with homeotic gene expression.

Gene expression studies Stamen expression studies have been conducted in many crop and model species including tomato (Ursin et al., 1989; Chmelnitsky et al., 2003),

302

R.J. Scott et al.

tobacco (Koltunow et al., 1990), A. majus (Nacken et al., 1991), oilseed rape (Scott et al., 1991b; Shen and Hsu, 1992), maize (Wright et al., 1993), rice (Tsuchiya et al., 1994), lily (Crossley et al., 1995), white campion (Barbacar et al., 1997), Lotus japonicus (Endo et al., 2002) and A. thaliana (Rubinelli et al., 1998; Sablowski and Meyerowitz, 1998; Amagai et al., 2003; Zik and Irish, 2003). Some of these studies have manipulated B and/or C class genes themselves to help identify their targets; these experiments will be discussed in more detail below. Proteomics analysis has also been used to compare and identify proteins extracted from rice anthers at different developmental stages (Kerim et al., 2003). The expression studies have identified genes involved in protein, starch and sucrose metabolism, osmoregulation, cell wall biosynthesis and expansion, sugar transport, lipid transfer, flavonoid synthesis and cytoskeleton structure. These are consistent with features of stamen and/or pollen development such as rapid growth, water movements associated with desiccation and dehiscence, stress response and accumulation of storage compounds. Many genes with unknown function were also recovered. Few of the genes have obvious developmental or signalling roles. Exceptions to this include several candidate transcription factors in white campion and A. thaliana (Barbacar et al., 1997; Zik and Irish, 2003), and a subunit of the COP9 signalosome (a repressor of photomorphogenesis in plants) found to be upregulated in rice anthers at the panicle heading stage, suggesting a role in mediating light signals to coordinate flowering with anther dehiscence (Kerim et al., 2003). In situ hybridization shows that genes identified in the expression studies are specific to a variety of stamen regions and cell types including microsporocytes, microspores, tapetum, endothecium, connectives, epidermis and filaments. A common theme is the predominance of tapetum-specific transcripts in early anthers, reflecting the high metabolic activity of this tissue (Scott et al., 1991a; and below). In some of the experiments referred to above, transcriptional profiling techniques were used in conjunction with mutants or transgenics for the B and C class genes to aid identification of their targets. For example, differential hybridization of an inflorescence-enriched cDNA library was used to isolate mRNAs expressed in wild-type flowers but not def mutants of Antirrhinum, which lack B function. This experiment yielded 12 differentially expressed genes including tap1, expressed in the tapetum and encoding a putative secreted protein and filamentous flower1 ( fil1), expressed mainly in stamen filaments and petal bases and encoding a candidate cell wall protein (Nacken et al., 1991). Rubinelli et al. (1998) compared wild-type and ag mutant flowers of A. thaliana, which lack C function, using subtractive hybridization, and after a further screen making use of RNA from ap3 mutants (which have pistils but not stamens) identified 13 genes that are differentially expressed in stamens. These included genes predicted to encode hydrolytic enzymes and a lipid transferase. Sablowski and Meyerowitz (1998) analysed gene expression in flowers with no C function and inducible B function, which were also treated with a protein synthesis inhibitor, to discover direct targets of AP3/PI. Three mRNAs were consistently upregulated by induction of B function in ap3-3; ag-3

Stamen Development

303

mutant flowers in the absence of protein synthesis, one of which, NAP, was studied in detail. In situ hybridizations to wild-type flowers showed expression in stamens and petals, both sites of B function, though also in some organs outside the domain of PI/AP3 activity. Based on ectopic and antisense expression studies the authors proposed that NAP functions in the transition between cell division and expansion, for instance in elongating stamen filaments. Zik and Irish (2003) used microarray analysis to identify genes affected by misexpression of AP3 and PI. Gene expression was compared in flowers from wild-type plants and from mutants and transgenics with altered B function. This study identified 47 genes likely to be directly or indirectly regulated by AP3/PI in petal and/or stamen development. Since the microarray used represented roughly 25% of the A. thaliana genome, some 200 genes may be affected by changes in AP3/PI activity. AP3/PI-responsive genes found to be expressed in stamens, or stamens and petals, included many implicated in the rapid cell expansion that is a feature of petal and stamen growth. Only two candidate transcription factors were identified, suggesting that AP3 and PI act relatively directly in regulating gene expression. Twenty-eight of the 47 genes had one or more candidate CArG boxes in a 1-kb upstream region, indicating that they could be directly activated by AP3/PI, although these sequences do not provide specific targets for particular MADS proteins.

Analysis of developmental mutants Mutant screens conducted over many decades have revealed a large number of sporophytic genes required for male fertility (described for example by Kaul, 1988; Chaudhury, 1993; Dawson et al., 1993; Chaudhury et al., 1994; Taylor et al., 1998; Sanders et al., 1999; Bhatt et al., 2001; Sorensen et al., 2002; Caryl et al., 2003). Pollen development is also affected by male gametophytic mutations (McCormick, 2004), and indirectly by mutations in the mitochondrial genome that result in degeneration of the tapetum (Hanson and Bentolila, 2004). The reported sporophytic mutations disturb a variety of processes in the stamen such as chromosome pairing or segregation in meiosis, viability of the tapetum, pollen wall formation, filament elongation and anther dehiscence. Genes required for pollen development that were first identified through mutant analysis include DIF1/SYN1, encoding a cohesin required for chromosome segregation (Bhatt et al., 1999; Bai et al., 1999); MALE STERILITY 2 (MS2), encoding a predicted fatty acyl reductase with a potential role in pollen wall formation (Aarts et al., 1997); and MS1 and ABORTED MICROSPORES (AMS), both candidate transcription factors expressed in the tapetum and possibly in microspores (Wilson et al., 2001; Sorensen et al., 2003). General screens for male sterility yield surprisingly few mutants affecting differentiation of anther cell types (Sanders et al., 1999). Therefore, we conducted an extinction screen for loss of expression of a b-glucuronidase (GUS) reporter gene fused to early tapetal-specific promoters, with the aim of specifically detecting mutations that disrupt microsporangium development

304

R.J. Scott et al.

(Sorensen et al., 2002). This screen identified three loci involved in patterning of the anther locule. Of these, EXTRA SPOROGENOUS CELLS (EXS)/ EXCESS MICROSPOROCYTES1 (EMS1) has been most thoroughly studied (Canales et al., 2002; Zhao et al., 2002); as this appears to be a key gene in patterning the stamen, it will be discussed in more detail below.

Genetic Control of Anther Development Most plant organs derive from meristems: undifferentiated populations of cells that divide to replenish themselves and also provide founder cells for organ primordia. Anther development is unusual in that the microsporangia arise from single archesporial cells rather than meristems. Key stages in this process are the establishment of adaxial–abaxial polarity, specification of cell types and formation of the radially symmetrical microsporangia. Asymmetry first becomes evident in the anther primordium with the formation of the microsporangia, for the two abaxial locules (facing the petals) are larger than the adaxial pair and are further separated by the connective tissue (Fig. 11.1). In A. thaliana leaves, asymmetry is established by cadastral interactions among genes promoting adaxial or abaxial identity, with the former suppressing the latter (Bowman et al., 2002). Mutations in either set of genes result in either ‘adaxialized’ or ‘abaxialized’ filamentous leaves, and the study of flowers in these mutant lines suggests that anther polarity is similarly regulated (Sawa et al., 1999; Siegfried et al., 1999). For example, double mutants for fil1 and yabby1 (yab1) or kanadi1 (kan1) and kan2 produce radially symmetrical filamentous structures, the internal structure of which has yet to be determined. As with leaves, FIL is expressed abaxially in the connective tissue of the anther, supporting the view that stamens evolved from leaves bearing microsporangia on their upper surfaces. The floral meristem of A. thaliana, like the SAM, is composed of three ‘histogenic layers’ of cells with separate lineages: L1 (epidermis), L2 (subepidermis) and L3 (core). Stamen primordia are initiated by periclinal divisions in the floral meristem, usually within the L2 (Jenik and Irish, 2000). The L2 gives rise to most of the cell types of the anther, including the sporogenous cells (discussed in detail below). The L3 contributes to the vasculature and sometimes to the connectives. With the growth of the anther primordium, cells of the L2 undergo a complex series of divisions leading to the formation of the four radially symmetrical microsporangia (Fig. 11.2), and conducting tissue that will eventually become linked to the filament. The founder cells of the four microsporangia are single L2 archesporial cells, each of which divides periclinally to form a primary parietal cell (PPC) subjacent to the L1, and a primary sporogenous cell (PSC) facing inwards (Canales et al., 2002). There has been significant progress in the last few years towards identifying genes involved in the first steps of microsporangium differentiation. In A. thaliana, the candidate transcription factor NOZZLE / SPOROCYTELESS (NZZ/SPL) (Schiefthaler et al., 1999; Yang et al., 1999) is required for archesporial specification, as nzz/spl mutant anthers fail to form archesporial cells.

Stamen Development

305

Microsporangium Adaxial surface

E E PP Ep

E SP

SP PP

A SF Archesporial cell (A) specified beneath epidermal layer (Ep)

PS

Division to form primary parietal (PP) and primary sporogenous (PS) cells. Sporogenous 'field' (SF) generated

PP PS

SP E

SP PS SP

PS SF

M T

SF

SP PS SP SP SP E

SF induces PP to divide into secondary parietal (SP) and endothecium (E), and adjacent cells into 'PP' cells

SF induces SP to divide into tapetal (T) and middle (M) cells, and 'PP' into secondary parietal (SP) and endothecium (E) cells

Fig. 11.2. Microsporangium development. Model for differentiation of the microsporangial cell layers in Arabidopsis thaliana.

Ito et al. (2005) have also demonstrated that sporogenesis is induced by AG through activation of the NZZ/SPL gene and that NZZ/SPL can induce microsporogenesis in the absence of AG function. One interpretation is that AG controls a specific process during organogenesis by activating another regulator that performs a subset of its functions. In maize, msca1 mutants produce archesporial cells but these do not divide into a PSC and PPC, and no microsporangia are formed (Chaubal et al., 2003). Intriguingly, the msca1 mutation also appears to convert the entire anther to another structure, as stomata form ectopically on the epidermis, although the filament maintains its identity. The number of cells that can acquire archesporial fate in the anther appears to be regulated by a similar mechanism to that used in the SAM to control numbers of pluripotent cells. In the SAM, cell number is determined by the leucine-rich repeat (LRR) receptor kinase CLAVATA 1 (CLV1), its signalling partner CLV2 and ligand CLV3 (Clark, 2001). In the anther, the number of cells acquiring archesporial fate is restricted to one per premicrosporangial domain by EXS/EMS1, a putative LRR receptor kinase (Canales et al., 2002; Zhao et al., 2002). In exs / ems1 mutants, which are affected only in male development, multiple L2 cells in the anther enter archesporial development, with the result that an excess of PSCs is formed following the first periclinal division. The ligand for EXS/EMS1 is unknown, but mutants in the TAPETUM DETERMINANT 1 (TPD1) gene have phenotypes similar to exs/ems1

306

R.J. Scott et al.

lines (Yang et al., 2003a; D.Ye, personal communication), suggesting that TPD1, a novel putative secreted protein, is involved at some point in this signalling pathway, whether or not as a ligand for EXS / EMS1. However, TPD1 expression differs from that of EXS/EMS1 (D. Ye, personal communication) and, while still consonant with a role in archesporial cell fate determination, this points to TPD1 playing a wider part in microsporangial development. An EXS/EMS1 orthologue, MSP1, has been reported in rice (Nonomura et al., 2003) and, in common with mac1 mutants in maize (which also share phenotypic features with exs / ems1 mutants (Sheridan et al., 1999)), msp1 lines show defects in both male and female development. The developmental fates of the two products of the archesporial cell division in A. thaliana are very different. The PSC undergoes a small number of divisions to generate the meiocytes, while the PPC divides periclinally to form an endothecial cell subjacent to the L1 and a secondary parietal cell (SPC). The SPC again divides periclinally to generate a middle layer cell next to the endothecium, and a tapetal cell adjacent to the sporogenous cells. Mutants have been described with defects in anther wall layer development – e.g. ms23 and ms32 in maize, where the prospective tapetal layer undergoes an extra periclinal division but neither layer differentiates as tapetum (Chaubal et al., 2000) – but no clear picture is emerging as to how cell fates are specified. Interestingly, this process is affected in exs / ems1 anthers for the tapetal and, frequently, middle layer cells are absent (Canales et al., 2002; Zhao et al., 2002). Whether EXS / EMS1 is itself required for tapetal specification, or whether the mass of extra sporogenous cells formed disrupts tapetal and middle layer differentiation, is unclear. Periclinal divisions of the single archesporial cell thus give rise to a linear array of different cell types (Fig. 11.2). However, this process alone – even accompanied by anticlinal divisions – cannot generate the radially symmetrical microsporangium, for cells of the microsporangium adjacent to the connective have been shown to have a different origin from those on the outer face in a number of species (Nanda and Gupta, 1978; Goldberg et al., 1993). The model for microsporangial development most consistent with the data confers a key organizational role on the sporogenous cells (Fig. 11.2). In this scheme, the PSC, following its formation from the archesporial cell, sets up a radial field of signals about itself. This field, which continues to be generated by the subsequent division products of the PSC, induces periclinal division and development in adjacent cells. Thus, the PPC – sister to the PSC – is induced to divide to form an endothecial cell, and adjacent to the source of signals, the meristematic SPC. The SPC then executes the final division of the programme, forming tapetal and middle layer cells. However, these signals are also received and interpreted by other cells adjacent to the sporogenous cells, whatever their origin, which are then recruited into the developmental pathway described above. This model may seem inconsistent with some observations; for example, the cell lineage described here for A. thaliana is not universally accepted (Yang et al., 1999), and certainly does not hold for all other species (Davis, 1966). However, whatever the division sequence followed by these cells, the sporogenous cells or their immediate antecedents are specified at an

Stamen Development

307

early stage in all plants, and the wall layers subsequently develop radially around them. There are many examples in plants where tissue organization is established in the absence of a coherent pattern of cell division (Torres-Ruiz and Ju¨rgens, 1994). A second problem is a time disparity between the development of the inner- and outer-facing domains of the microsporangium (Nanda and Gupta, 1978). However, data from exs / ems1 mutant anthers point to cells of the subepidermal L2 being ‘primed’ for development, and thus, together with the PPCs they may be able to respond to signals generated from the PSC earlier than other L2 cells and those of the connective tissue. The nature of the hypothetical radial signal is unknown, but mutants would have phenotypes resembling those of exs / ems1 lines. The fact that transcripts of TPD1, a putative partner for the EXS/EMS1 receptor kinase, are concentrated in the spore mass (D. Ye, personal communication) provides a first indication of a radial disposition of signals centred on the developing microsporocytes.

Microsporangial Development Development of the different cell types proceeds at varying speeds; while the micosporocytes and tapetal cells rapidly enlarge and display differing patterns of gene expression, the endothecial layer remains undifferentiated through meiosis and much of microsporogenesis. The middle layer, of which little is known, is apparently crushed by the expansion of the microsporocytes and tapetum. Following the initial phase of cell expansion, cytoplasmic contact is lost between the microsporocytes and tapetum as plasmodesmata are ruptured, seemingly by the rapid synthesis of the microsporocyte callose wall (see below). Both the microsporocytes and tapetum eventually develop into coenocytes. As the microsporocytes enter meiosis, plasmodesmata between them enlarge to form ‘cytomictic channels’ up to 0.5 mm in diameter, through which cytoplasmic exchange must occur. These channels are held to promote synchrony within the micosporocyte mass (Heslop-Harrison, 1966). Later in tapetal development, particularly in plants such as A. thaliana with secretory tapeta (see below), walls between the tapetal protoplasts become gelatinous, and interconnecting plasmodesmata enlarge to form irregular channels. Postmeiotic development thus involves crosstalk between two coenocytes, sealed within the microsporangium by a lipid/sporopollenin peritapetal membrane (Dickinson, 1970). Surprisingly, this interaction does not become essential until after meiosis, for exs / ems1 microsporocytes progress to the tetrad stage in the absence of any visible tapetum, and it is believed to be the lack of tapetally derived b-1,3-glucanase, which releases the microspores from the tetrad callose walls (see below); that causes developmental arrest (Canales et al., 2002).

Meiosis in the sporogenous cells The principal events of meiosis – chromosome pairing, recombination and segregation – are common to all eukaryotes and the genes involved have

308

R.J. Scott et al.

been the subject of a number of reviews (Vershon and Pierce, 2000; Rabitsch et al., 2001), as well as the focus of an interkingdom database (www.germonline.unibas.ch). However, unlike in animals, meiosis in plants stands not at the threshold of gametogenesis, but of the alternation of generations. The microspores formed as a product of male meiosis develop into pollen, an independent microgametophyte. Perhaps for this reason, developmental events surrounding plant meiosis, and some details of the process itself, differ from those in other kingdoms. In plant microsporocytes dramatic cytoplasmic reorganization accompanies nuclear events, including dedifferentiation and division of plastids and mitochondria, and dramatic falls in r- and mRNA (Dickinson, 1987). Little DNA synthesis occurs during this period, except in the organelles, and prophase cDNA libraries reveal a dramatic decrease in the numbers of nuclear genes expressed (Crossley et al., 1995). This de facto purging of sporophytic information from the microsporocyte cytoplasm has been interpreted either as facilitating gametophytic development post-meiosis, or freeing the germline of detrimental RNA species including viruses and silencing elements (Dickinson, 1987). Organellar dedifferentiation and replication may also be caused by this decrease in information from the nucleus, but microsporogenesis is also unusual in that it is highly sensitive to mitochondrial mutation. This is held to result from a need for extreme efficiency in energy production by the tapetum. Mutations in mitochondrial DNA; and nuclear restorer genes, which ‘correct’ the defect; have been exploited in the development of cytoplasmic male sterility systems used in plant breeding (Kaul, 1988; Hanson and Bentolila, 2004). As in all eukaryotes, homologous chromosomes pair during plant meiosis. While there have been many reports of presynaptic alignment of homologues, little unequivocal data exists, save for work with polyploid cereals where homologous pairing clearly takes place in floral tissue prior to meiosis (Martinez-Perez et al., 1999). The balance between homologues and homologous pairing in cereals is regulated by the polyhomeotic (ph) locus (Vega and Feldman, 1998), and this premeiotic alignment of homologues has been cited as evidence of an ancestral system of presynaptic alignment of homologues (Martinez-Perez et al., 1999). Evidence is now emerging of homologous pairing during interphase in somatic cells, as also occurs in animals and microorganisms (Francz et al., 2002), but any mechanistic relationship between these two types of pairing has yet to be established. Following pairing, DNA processing in the male meiocytes is identical to that of other organisms, including the formation of double strand breaks, resection, strand invasion, ligation and resolution of the complexes formed via Holliday junctions. As in most eukaryotes, these events are accompanied by the formation of synaptonemal complexes and recombination nodules at zygotene/pachytene. Generally, the genes involved in meiotic processing in plants are orthologous to animal sequences, but differences in gene number and expression profile do occur (Bhatt et al., 2001; Caryl et al., 2003). Evidence from a range of meiotic mutants suggests that meiotic checkpoints may also differ between plants and other eukaryotes (Bhatt et al., 2001), with at least one checkpoint absent from male development in A. thaliana (Bhatt et al.,

Stamen Development

309

1999). However, mutations in the homeodomain protein MALE MEIOCYTE DEATH, which cause meiocytes to arrest at diplotene/diakinesis (Yang et al., 2003b), suggest that some level of checkpoint control is present in anthers.

The cell wall in male meiosis Plant cells use different methods for mitotic and meiotic cytokinesis (reviewed in Brown and Lemmon, 2001). In mitotic cytokinesis, daughter cells are separated by a cell plate formed by fusion of golgi-derived vesicles at the phragmoplast. This structure comprises an array of cytoskeletal elements, vesicles and endoplasmic reticulum that expands centrifugally from the interzonal region of the mitotic spindle towards the parent cell walls (Staehelin and Hepler, 1996). The vesicles contribute both membranes and wall components to the nascent wall. Callose is the main component of the new wall, but it is later degraded and replaced by cellulose and its derivatives. Male meiotic cytokinesis differs from mitotic cytokinesis in several ways, the most striking being that on entering meiosis a microsporocyte deposits callose between the plasma membrane and cellulose wall (Fig. 11.3). Callose is also the major constituent of the intersporal walls formed after meiosis, which are later degraded to release the microspores (Echlin and Godwin, 1968; Steiglitz, 1977). The mechanism of cytokinesis is also different, with a variety of species-specific patterns observed. Most monocots undergo successive cytokinesis during male meiosis, in which a wall is formed between the dyad cells after meiosis I and the microspores of the nascent tetrad after meiosis II. As in mitotically dividing cells, these walls form centrifugally. In contrast, most dicots, including A. thaliana, undergo simultaneous microsporocyte cytokinesis, meaning that no walls are formed until the end of meiosis II. In simultaneous cytokinesis, intersporal walls first appear as ingrowths from the callose wall surrounding the microsporocyte, and expand centripetally until the microspores are separated (Brown and Lemmon, 1988, 2001; Owen and Makaroff, 1995). Control of division plane also occurs differently in sporophytic mitosis and in male meiosis. In meiotic divisions, the future site of cell separation is not marked by a preprophase band. Instead, after each meiotic division in species with successive cytokinesis, or after meiosis II in those with simultaneous cytokinesis, each microspore nucleus becomes surrounded by a radial array of microtubules that partition the surrounding cytoplasm into ‘spore domains’ (Dickinson and Sheldon, 1984; Brown and Lemmon, 1988, 2001). Cytokinesis proceeds along the planes defined by the intersection of the arrays, as vesicles contributing membranes and wall components coalesce at the spore domain interfaces. In tetraspore (tes)/stud (std) mutants of A. thaliana, there is partial or complete failure of male meiotic cytokinesis, so that all four microspore nuclei begin development in an undivided cytoplasm (Hu¨lskamp et al., 1997; Spielman et al., 1997). The TES gene encodes a putative kinesin that appears to be required for establishment of the radial microtubule arrays surrounding the microspore nuclei at the end of meiosis, as these are lacking in tes mutants ( Yang et al., 2003c). The tes / std mutations are recessive and

310

R.J. Scott et al.

Fig. 11.3. Synthesis and dissolution of the callose wall during microgametogenesis. (A) Callose is first synthesized beneath the cellulose wall of the pre-meiotic microsporocyte and persists throughout meiosis. Cross walls of callose are formed during post-meiotic cytokinesis, yielding a typical tetrad of microspores. A cocktail of wall-degrading enzymes, including an endo-b-1,3glucanase, is then secreted into the locular space by the tapetum and degrades the callose wall to release the microspores. The microspores then complete development, yielding bi- or trinucleate pollen grains. (B) Fluorescence microscopy of aniline blue–stained tetrads shows the characteristic pattern of callose distribution in wild-type tobacco.

TES is expressed throughout the anther before meiosis, indicating that male meiotic cytokinesis is under sporophytic control.

Callose wall synthesis, dissolution and function During prophase I of meiosis, the microsporocytes of most angiosperm species secrete a wall of callose, a b-(1,3)-glucan, between the plasmalemma and the original cellulosic wall (Rowley and Southworth, 1967). A second round of callose wall synthesis occurs as part of post-meiotic cytokinesis, so that individual dyads and microspores in the case of most monocots (which have successive cytokinesis) and the tetrad of microspores in the case of most dicot species are also surrounded by a layer of callose. In wounded plant cells callose is synthesized by a plasma membrane localized glycosyl transferase UDP-glucose

Stamen Development

311

(1,3)-b-glucan synthase (callose synthase) (Delmer, 1987). Recently, Østergaard et al. (2002) described an A. thaliana gene, Atgsl5, that encodes a plasma membrane protein homologous to yeast b-(1,3)-glucan synthase and which partially complements a yeast b-(1,3)-glucan synthase mutant. The gene is highly expressed in flowers and may represent the enzyme responsible for microsporocyte callose synthesis. The external microsporocyte and the internal cross-walls of the tetrad are dissolved to release individual microsporocytes by a cocktail of enzymes (callase) released from the tapetum (Fig. 11.3). Numerous theories have been advanced that suggest the function of the callose wall to be as follows: (i) a barrier or ‘molecular filter’ between the sporogenous cells and the rest of the sporophyte that is necessary for meiosis (Heslop-Harrison, 1964; Heslop-Harrison and Mackenzie, 1967); (ii) a temporary wall that both isolates the products of meiosis to prevent cell cohesion and fusion and, upon its dissolution, results in the release of free cells (Waterkeyn, 1962); and (iii) a template or mold for the formation of the species-specific exine sculpturing patterns seen on mature pollen grains (Waterkeyn and Bienfait, 1970). This last potential function will be considered in more detail later. Natural exceptions, natural and induced mutants and plants engineered to degrade the callose wall prematurely have provided some insight into the role of the callose wall. Several lines of evidence suggest the callose wall is not required for male meiosis; both the primitive species Pandanus odoratissimus, which naturally lacks a callose wall (Periasamy and Amalathas, 1991), and transgenic tobacco (Nicotiana tabacum), tomato (Lycopersicon esculentum) and oilseed rape (Brassica napus) plants in which the callose wall is absent throughout meiosis due to the ectopic expression of a b-1,3-glucanase, successfully initiate and complete meiosis to produce a tetrad of microspores (Worrall et al., 1992; R.J. Scott, unpublished results; Fig. 11.4A and B). Engineered absence of the callose wall in lettuce also failed to disrupt meiosis (Curtis et al., 1996). The callose wall does appear important for the production of individual microspores. Species that naturally produce permanent tetrads (four microspores fused together), e.g. members of Juncaceae, Ericaceae, Oenotheraceae develop little or no callose within the post-meiotic cross-walls within the tetrad (Blackmore and Crane, 1988; Fig. 11.4). The microspores possess individual exine walls but these are fused along the line of the cross-walls, which presumably prevents their separation. Blackmore and Crane (1988) proposed a model in which the fusion of exines is related to the extent of callose wall and the timing of its deposition. Absence of callose in the tetrad cross-walls in some species that normally produce them also results in permanent tetrads, e.g. tomato (Fig. 11.4). However, both P. odoratissimus (Periasamy and Amalathas, 1991) and B. napus expressing ectopic b-1,3-glucanase (R.J. Scott et al., unpublished results) produce individual microspores in the absence of cross-wall callose suggesting that in some species callose is not essential for ensuring microspores remain separate during exine wall formation within the tetrad (Fig. 11.4).

312

R.J. Scott et al.

Meiosis + Cytokinesis (Wild-type plants)

A

Uninucleate microspores

Meiosis + Cytokinesis (A9::glucanase plants)

Monocot Brassica napus

With callose wall

Dicot Lycopersicon esculentum

Ericaceae

General angiosperm (including N. tabacum, L. esculentum, B. napus, Zea mays)

Nicotiana tabacum

Pandanus odoratissimus

Key

Juncaceae Callose wall

Without callose wall

Pollen wall Pergularia daemia

L. esculentum

N. tabacum

Wild-type

B. napus

C

A9::glucanase

A9::glucanase

Wild-type

N. tabacum

B

Fig. 11.4. The role of the callose wall in the separation of microspores following cytokinesis. (A) Most species of angiosperms produce tetrads with both outer and cross walls composed of callose. Dissolution of the callose wall results in the release of four separate uninucleate microspores, which develop into individual pollen grains. The microsporocytes and tetrads of members of the Ericaceae also contain callose, but following tetrad wall dissolution the microspores, and subsequently pollen grains, remain permanently fused as tetrads. The microsporocytes and tetrads of a few species do not synthesize a callosic wall. Most produce fused tetrads, such as members of the Juncaceae, and Pergularia daemia, but Pandanus odoratissimus develops individual uninucleate micropores in the absence of the callose wall. Artificial removal of the callose wall by ectopic tapetum-specific expression of an endo-b-1,3glucanase (A9::glucanase) in three species that normally produce individual microspores resulted in a range of outcomes with respect to the separation of microspores. Brassica napus retained the ability to produce individual microspores, whilst Lycopersicon esculentum developed regular fused tetrads similar to members of the Ericaceae. Nicotiana tabacum produced a highly irregular mixture of microspores showing evidence of both a lack of cytokinesis (multinucleate spores) and fusion of individual microspores along abutting exine walls. (B) Fluorescence microscopy of aniline blue-stained tetrads showing the characteristic pattern of callose distribution in wild-type tobacco (top) but no callose in engineered N. tabacum carrying the A9::glucanase gene (bottom). (C) Transmission electron micrographs images of wild-type microspores and microspores from plants expressing an ectopic tapetum-specific endo-b-1,3-glucanase that removes the callose wall during early microsporogenesis.

Stamen Development

313

The quartet (qrt) mutants in A. thaliana produce permanent tetrads despite apparently normal patterns of callose wall synthesis and dissolution indicating that an alternative mechanism, possibly the deposition of pectins within the cross-walls, participate in microspore separation (Rhee and Somerville, 1998). At the end of meiosis, the external and intersporal walls of the tetrad are dissolved to release individual microspores by a mixture of enzymes (callase) containing endoglucanases and exoglucanases secreted by the tapetum. In the anthers of petunia and lily, expression and secretion of callase activity is under strict developmental control (Frankel et al., 1969; Steiglitz and Stern, 1973). The callase complex of lily consists of a 32 kDa endo-b-glucanase and a 62 kDa exotype b-1,3-glucanase (Steiglitz and Stern, 1973). The endotype enzyme appears to play the most important role in the degradation of the callosic walls, while the exotype is involved in the further hydrolysis of released oligosaccharides. Alterations in the timing of b-1,3-glucanase expression or failure to express b-1,3-glucanase lead to abnormal dissolution of the tetrad callose walls, which has been shown to be a primary cause of male sterility in cytoplasmic male-sterile lines of petunia (Izhar and Frankel, 1971), sorghum (Warmke and Overman, 1972) and soybean (Jin et al., 1999). Callase must also contain cellulases to degrade the original cellulosic wall of the microsporocyte since premature removal of the callose component of the wall does not release microspores into the locule until the normal time of callase production (Worrall et al., 1992). Several candidates for genes encoding the endo b-1,3-glucanase component of the callase enzyme cocktail have been identified, although none have been confirmed directly, e.g. by knockout or knock-down analysis. The A. thaliana A6 gene encodes a polypeptide with a domain similar in sequence to b-1,3-glucanases and a 114 amino-acid C-terminal domain, which is not present in other known b-1,3-glucanases. Reporter gene studies established that A6 gene expression is tapetum specific and temporally correlated with the expression of callase activity (Hird et al., 1993). Transcripts of the tobacco Tag1 gene that encodes a polypeptide related to established tobacco b-1,3glucanases and containing the conserved pentapeptide motif of the active site of these enzymes are also expressed exclusively in the tapetum and show a callase-like pattern of expression (Bucciaglia and Smith, 1994). Yamaguchi et al. (1997) reported the cloning of cDNA (Osag1) from rice with significant similarity to anther b-1,3-glucanases and speculated that ‘Osag1 may be involved in tetrad dissolution’. However, Yamaguchi et al. (2002) subsequently showed that Osag1 expression occurs in anthers after callose wall dissolution and that the transcript is present in roots and leaves, a pattern of expression inconsistent with its identity as the b-1,3-glucanase component of callase. Whilst one or more of the genes described above may participate in callose wall dissolution, more effort is required to confirm this and then to understand how this key stage in pollen production is regulated. The availability of expression array data and gene knockout lines in A. thaliana are likely to assist in identifying bona fide components of callase.

314

R.J. Scott et al.

Tapetal development and pollen coat formation Although presumably acting as a nurse tissue to the developing meiocytes, the tapetum is likely to play only a nutritive role. Developmental signalling, such as that occurring between nurse cells and the egg in Drosophila (St Johnston and Nu¨sslein-Vollhard, 1992), is unlikely to take place since cytoplasmic connections between these groups of cells are lost at an early stage. Microsporocytes and tapetal cells do, however, share many developmental pathways (Dickinson and Bell, 1976), particularly during pollen wall formation. Exceptions are intense phases of protein synthesis early in development, and meiosis itself (although homologous pairing can occur in the tapetum (Martinez-Perez et al., 1999)). Significant nuclear changes also occur in the tapetum, with endomitosis and endoreduplication commonly occurring in many species; in Petunia tapetal nuclei attain DNA levels of 8C (Liu et al., 1987), while 16C has been reported for maize (Moss and Heslop-Harrison, 1967). This intensity of nucleic acid synthesis is probably unique to the tapetum and, combined with high levels of protein synthesis early in development, must create an extraordinarily high demand for energy (Liu and Dickinson, 1989), which presumably results in the particular sensitivity of the tapetum to mitochondrial mutations. Tapeta may be secretory, as in the Liliales and the Gramineae, and remain at the periphery of the microsporangium throughout development, or amoeboid/invasive where tapetal cells move into the loculus and intermingle with the developing microspores, as in the Asteraceae (Pacini, 1990). In other plants, such as the Brassicaceae, the tapetum fragments at an early stage and components of the ruptured protoplast move into the loculus (see below). In addition to its central role in pollen wall formation, the tapetum contributes a lipid-rich exine coating in many species – the nature of this coating depending on the tapetal type. Thus, the secretory tapetum of Lilium deposits a blend of carotenins, flavinols and lipids (termed pollenkitt) on to the exine surface (Reznickova and Dickinson, 1982), while the invasive tapetal protoplasts of the Asteraceae penetrate within the cavea of the complex-chambered pollen wall. Most is known of the complex coatings applied to the surface of brassica pollen, termed tryphines (Dickinson and Lewis, 1973; Murgia et al., 1991). These are formed when tapetal fragments become applied to the pollen exine, and contain a range of lipids, glycolipids and proteins essential for successful development of the pollen on the stigma surface (Ruiter et al., 1997a,b). A. thaliana pop1 mutants defective in coating components can fail to hydrate after pollination, while in the Brassica spp. a class of small, cysteine-rich pollen coat proteins (PCPs) are involved in interactions with the stigmas, and particularly with the female determinants of the self-incompatibility (SI) system (Doughty et al., 1998). Indeed the male SI determinants in Brassica species (SRC and SP11) belong to the PCP family of proteins. How pollen coatings adhere so specifically to the exine is not understood, but the fact that in the Brassicaceae the entire tapetum is transferred wholesale to the pollen surface belies a high affinity for the sporopollenin surface. Detailed study of this interface has revealed an electron lucent exinic outer

Stamen Development

315

layer, but nothing is known of its molecular structure. As the loculus is dehydrated in preparation for anther dehiscence, the tryphine changes in aspect, with many of the cytoplasmic components becoming no longer visible. On alighting on the stigma, the tryphine forms the interface between the pollen and pistil, and elements penetrate fissures in the stigmatic pellicle to establish molecular and hydraulic continuity (Roberts et al., 1984; Dickinson, 1995).

Pollen Wall Structure, Synthesis and Patterning Structure of the pollen wall Following male cytokinesis the individual microspores of the tetrad initiate development of the pollen wall. As described earlier this occurs in close association with the temporary callose wall and is essentially complete before its dissolution to release the microspores into the locule. This ephemeral wall represents the first of several layers deposited at the microspore surface. This is followed by the primexine, a precursor of the outer of the two layers of exine ‘the ectexine’, followed by the second exine layer ‘the endexine’, and finally ‘the intine’ (Blackmore and Barnes, 1990) (Fig. 11.5A). This sequence is invariant, but individual phases of deposition may be reduced or deleted to produce species-specific wall architectures; however, a well-developed intine is always present. The ectexine is the most complex of the pollen wall layers and provides most of the species-specific variation in pollen wall patterning through more or less elaboration of its column-like columellae or roof-like tectum (Fig. 11.5B). The primexine is largely composed of polysaccharide, and apparently acts as a template, or glycocalyx, that contains patterning information to guide the accumulation of sporopollenin, the main structural component of the pollen wall. In the A. thaliana mutant dex1, primexine deposition is delayed and significantly reduced and pollen wall patterning is disrupted (Paxson-Sowders et al., 2001). The DEX1 gene encodes a predicted membrane-associated calcium-binding protein that may act as a nucleation site for sporopollenin. During the initial phase of wall growth sporopollenin is apparently polymerized from precursors synthesized and secreted by the microspore; however, the bulk of exine sporopollenin is derived from precursors secreted by the tapetum and incorporated into the wall following the dissolution of the tetrad (reviewed in Scott et al., 1991a). The combination of acute vulnerability and central importance of the male spore in the life cycle of sexually reproducing plants has driven the assembly of various adaptive features that endow its protective capsule with remarkable properties. Foremost is the unparalleled combination of physical strength, chemical inertness and resistance to biological attack of the exine, which is due principally to its main structural component, sporopollenin. These adaptive features of sporopollenin have greatly hampered progress in understanding both its chemical composition and details of its biosynthesis. Early literature frequently cites carotenoids as the main constituent of sporopollenin (see Shaw, 1971; reviewed in Scott, 1994). However, the demonstration that the

316

R.J. Scott et al.

Lacuna

A Columellae

Sexine Nexine

Nexine I Nexine II Intine

Exine

Tectum

B

C

Pollen wall development Callose wall 3

2 1

4

5

9

10

6 7 8

11

12

Fig. 11.5. Structure of the pollen wall and a model for its development. (A) Schematic representation of the main architectural features of a generalized mature pollen wall. The terminology is according to Erdtman (1969). (B) Scanning electron micrograph of Lilium pollen wall showing reticulate pattern formed by the fused heads of the columellae. The nexine I is visible through the lacunae. Scale bar: 5 mm. (C) Model of pollen wall development based on Lilium (after Scott, 1994). Events begin before meiosis and proceed left to right until the mature wall is formed: (1) transcription of ‘pattern’ genes in the pre-meiotic microsporocyte nucleus; (2) insertion of labile pattern information into the plasma membrane via endoplasmic reticulum-derived vesicles, followed by tessellation to produce a negative stencil of plates; (3) first phase of primexine synthesis over entire microspore surface except areas destined to become apertures; (4) conversion of primexine to a sporopollenin-receptive state through activity of factor(s) secreted from sites between the plates; (5) second phase of primexine synthesis, more rapid than the first, resulting in limited primexine conversion and specification of columellae; (6) nascent columellae apparent as lamellated strands that lack substantial sporopollenin; (7) consolidation of columellae by appearance of partially polymerized (proto)sporopollenin on the receptive surfaces; (8) final phase of primexine synthesis during which the pattern stencil dissipates or is circumvented to produce the solid nexine I; (9) callose wall dissolved; (10) wall elements further consolidated by tapetally derived sporopollenin; (11) nexine II synthesized without participation of the primexine; and (12) intine synthesis initiated.

potent inhibitor of carotenoid biosynthesis, norflurazon, failed to prevent sporopollenin biosysnthesis in Cucubita pepo (Prahl et al., 1985) began a re-evaluation of sporopollenin composition. Subsequently, a large body of experimental evidence (reviewed in Scott, 1994), including labelling and degradation studies, and several types of spectroscopy, has established that

Stamen Development

317

sporopollenin has two main constituents, the major part being long-chain fatty acids and the other more minor component, aromatic, perhaps p-coumaric acid or related compounds. More recent analyses such as Domı´nguez et al. (1999) and Ahlers et al. (1999) have confirmed this view. The phenolic monomers are coupled by ester bonds characteristic of polyphenolics such as lignin and suberin, and this places sporopollenin together with cutin, suberin and lignin in a family of functionally (protective) and compositionally related biopolymers (Scott, 1994). Sporopollenin, cutin and suberin also share a common ‘polylamellate’ fine structure – multiple alternating electron-opaque and electron-translucent lamellae (Kolattukudy and Espelie, 1985; Scott, 1994). In a model of suberin fine structure proposed by Kolattukudy and Koller (1983), the lucent layer is interpreted as being composed of the hydrophobic regions of long-chain fatty acids, whilst the hydrophilic ends of these molecules comprise the opaque layer in a complex with the phenolic component. How is sporopollenin synthesized within the growing exine walls? One model assumes that the process resembles suberin biosynthesis; the primexine acts as a loose scaffold on to which sporopollenin monomers (fatty acids and phenolics) are covalently attached by the localized action of superoxide radicals generated at the plasma membrane of the microspore (Fig. 11.5C). This creates multilamellate probaculae, the precursors of the columellae and tectum. Subsequently, the probaculae act as conduits for the cross-linking agent, which facilitates extensive polymerization, and eventually the production of the highly resistant pollen wall (Scott, 1994). Patterning of the pollen walls Pollen wall development involves the specification of three main architectural features: (i) consistent numbers of precisely positioned germinal apertures (in A. thaliana the three apertures are 1208 apart); (ii) patterned arrays of columellae, frequently arranged as hexagons (Fig. 11.5A and B); (iii) layering into tectum, columellae, nexine and intine (Fig. 11.5A). The pollen wall over apertures consists of intine only; consistent with the proposed role for primexine, this matrix is not deposited in areas destined to become apertures, thereby preventing exine development. For lily, HeslopHarrison (1963) proposed that primexine synthesis is prevented by apposition of a colpal shield (a plate of endoplasmic reticulum) to the plasma membrane at the position of the future aperture. Subsequently, Dover (1972) and Sheldon and Dickinson (1986) implicated the meiotic spindle in aperture positioning, perhaps by the spindle forcing cytoplasmic components, including endoplasmic reticulum, against the plasma membrane to form a colpal shield or similar structure. Despite the huge diversity in pollen wall patterning, a basic design feature shared by many species is the reticulate arrangement (frequently as hexagons) of the columellae, to form a series of interlocking lacunae (Fig. 11.5B). These are particularly obvious in species such as lilies and brassicas where the exine lacks the roof-like tectum; however, in species such as tobacco with a well-developed tectum, and hence a smooth appearance, the tectum is

318

R.J. Scott et al.

nevertheless supported by the same patterned arrays of columellae (Scott, 1994). The positioning of these elements is complete at the tetrad stage. Several lines of evidence suggest that the genes responsible for speciesspecific patterning of the pollen wall are transcribed within the diploid nucleus of the pre-meiotic microsporocyte, and that the pattern information is inherited by the microspores (Heslop-Harrison, 1963, 1968; Rogers and Harris, 1969). There is no evidence for participation of any subcellular organelle in specifying the columellae position. However, centrifugation experiments with developing lily microsporocytes suggest the agent(s), potentially protein-coated vesicles, responsible for imposing pattern on the primexine appear in the cytoplasm at the beginning of meiosis and are progressively inserted into the plasma membrane during the course of meiosis (Sheldon and Dickinson, 1983; Dickinson and Sheldon, 1986). These vesicles may insert protein, or another substance, into the plasma membrane that somehow influences the position of the columellae. How might protein inserted in the plasma membrane specify pattern? Heslop-Harrison (1972) made the following observation: ‘Sometimes when we contemplate biological pattern it is difficult to imagine how great a part of it could arise from physical causes, and correspondingly easy to slip into accepting that genetical control extends down to all detail. Yet this is obviously not so’. Sheldon and Dickinson (1983) acknowledged the role of purely physical phenomena in the specification of exine pattern, suggesting that the vesicles fuse randomly with the plasma membrane and the deposited protein undergoes selfassembly into pattern-specifying units. This would occur in a manner analogous to the behaviour of bubbles within a foam, or in the cracks formed in the drying of a uniform surface such as mud, or more dramatically in the basalt layers that formed geological wonders such as the Devil’s Postpile (USA) and Giants Causeway (Northern Ireland), which naturally generate hexagonal patterns (reviewed in Scott, 1994; Fig. 11.5C). In the bubble model the vesicles insert material into the membrane that initially generate circular plates that eventually collide to form hexagonal plates; the columellae form along lines specified by the interfaces between the plates. In the cracking model, the vesicular material forms a uniform layer, which upon shrinkage forms the hexagonal arrays; the columellae are specified along the fracture lines. A model for how this patterning information is elaborated to generate the pollen wall is shown in Fig. 11.5C. This figure also shows how the final phases of pollen wall development are orchestrated. The callose wall plays an important role in pollen wall formation since its loss strongly disrupts pollen wall structure. In tobacco plants expressing b1,3-glucanase ectopically, the usually near complete tectum is deleted leaving exposed columellae, suggesting that the callose wall may provide a solid surface against which the tectum forms (Worrall et al., 1992; Fig. 11.6). However, removal of the callose wall by the same technique in B. napus, a species that lacks a tectal plate, did not disrupt the regular positioning of columellae (Fig. 11.6), indicating that the callose wall does not provide a stencil to guide the positioning of wall elements as proposed by Waterkeyn and Bienfait (1970) for Ipomea.

Stamen Development

319

Fig. 11.6. Role of the callose wall in exine patterning. Electron micrographs of pollen grains from Nicotiana tabacum and Brassica napus wild-type and A9::glucanase-expressing plants showing the effect of callose wall removal on exine wall patterning. The top and middle rows show scanning electron micrographs at low and high power, respectively. In N. tabacum loss of the callose wall disrupts the formation of the outer tectal layer as evidenced by the exposed columellae. In B. napus, which naturally lacks a tectum, the absence of the callose wall results in columellae of irregular length, suggesting that the callose wall may provide a physical barrier to primexine growth that ensures the elaboration of columellae of regular length. The regular pattern of lacunae on the surface of B. napus A9::glucanase pollen grains suggests that the positioning of columellae does not involve information provided by the callose wall.

Anther Dehiscence Mature pollen is released from the anther by dehiscence, a programme of cell destruction culminating in rupture of the stomium, a furrow separating each pair of anther locules (Fig. 11.1). Sanders et al. (1999) codified the major events that occur during the dehiscence programme in A. thaliana. These begin with degeneration of the middle layer and tapetum, expansion of the endothecial layer and deposition of fibrous bands (wall thickenings) in endothecial and connective cells. Degeneration of the septum generates a bilocular anther, which is followed by stomium cell breakage. In tobacco, the dehiscence process is very similar, differing only in the greater degree of connective breakdown. An elegant series of cell ablation experiments in tobacco showed that a functional stomium region is essential for dehiscence (Beals and Goldberg, 1997). Although events associated with dehiscence are normally coordinated with pollen development, male-sterile tobacco anthers that lack pollen and tapetum undergo a normal dehiscence process (Goldberg et al., 1993), indicating that dehiscence does not require signals derived from these cell types.

320

R.J. Scott et al.

Comparative anatomical studies, microscopic observations, micromanipulation experiments and measurements of water movements in stamens of a range of species suggested a model in which the endothecium contributes to two steps in the dehiscence process (Keijzer, 1987; Bonner and Dickinson, 1989). Just prior to dehiscence, concomitant with lysis of the cells of the stomium, the endothecial and epidermal cells become turgid (Fig. 11.7). This generates an inwardly directed force in the anther wall that causes rupture of the weakened stomium. Subsequent desiccation of the endothecium causes differential shrinkage of thickened and unthickened regions of the cell wall, creating an outward bending force that leads to retraction of the anther wall and full opening of the stomium to permit pollen release (Keijzer, 1987). The ms35 mutant of A. thaliana (Dawson et al., 1999) provides supporting evidence for the role of the endothecium in this second phase of dehiscence. Mature anthers of ms35 mutants contain viable pollen grains, breakdown of the septum and stomium occur normally, and the pattern of water movement is also apparently normal. However, the endothecial cells fail to develop lignified secondary wall thickenings and do not undergo anther wall shrinkage normally associated with retraction of the anther wall. The MS35 gene (renamed MYB26) encodes a putative R2R3-type MYB transcription factor (SteinerLange et al., 2003). Since other MYBs are known to regulate the phenylpropanoid pathway, e.g. AtMYBb4 (Jin et al., 2000), the role of MYB26 may be to activate this pathway in endothecial cells in order to provide lignin residues for wall thickening. Perhaps the most significant recent advance in understanding the control of dehiscence has been the discovery of the involvement of jasmonic acid (JA),

Pistil Anther Petal Sepal

Stamen filament Water flow Organ elongation Jasmonate transmission

Fig. 11.7. Anther dehiscence. Model for the synchronous regulation of pollen maturation, anther dehiscence and flower opening by jasmonic acid (JA) in Arabidopsis thaliana. The shaded areas represent regions that actively take up water and elongate in response to JA (adapted from Ishiguro et al., 2001).

Stamen Development

321

a lipid-derived signalling compound widely distributed in the plant kingdom. All known mutants in JA biosynthetic or signalling pathways exhibit a similar phenotype with: (i) reduced elongation of the filaments; (ii) delayed anther dehiscence; and (iii) reduced pollen viability resulting in male sterility. Several mutants in JA biosynthetic enzymes have been isolated that cause male sterility and can be rescued by application of JA (McConn and Browse, 1996; Sanders et al., 2000; Stintzi and Browse, 2000; Park et al., 2002; von Malek et al., 2002). The role of JA in male fertility is supported by the JA signal transduction mutant coronatine-insensitive 1 (coi1), which is male sterile and insensitive to JA treatment (Feys et al., 1994; Xie et al., 1998). The DEFECTIVE IN ANTHER DEHISCENCE 1 (DAD1) gene encodes a phospholipase A1 catalysing the initial step of JA biosynthesis (Ishiguro et al., 2001). Desiccation of dad1 anthers appears delayed so that at the time of flower opening, when in wild-type anthers dehydration and shrinkage of the endothecium and connective cells break the stomium, the endothecium and connective cells of dad1 anthers are fully expanded and the locules are filled with liquid. Expression studies suggest that DAD1 activity is restricted to the anther filament immediately before flower opening; therefore the filament may represent the primary source of JA within the flower. Ishiguro et al. (2001) proposed a model in which JA synthesized in the filaments regulates water transport in the stamens and petals to bring about the coordinated opening of flowers, filament elongation and anther dehiscence (Fig. 11.7). In this model, JA is required for the expression of genes involved in water transport in anthers. Interestingly, AtSUC1, a plasma membrane Hþ-sucrose symporter, which is theoretically capable of transporting sucrose to increase water uptake, accumulates in some of the connective cells surrounding the vascular tissue during the final stages of anther development (Stadler et al., 1999). Filament and petal elongation are also inhibited in dad1 mutant flowers, suggesting that JA regulates water transport into these organs. This accords well with the observation that dehiscence of onion anthers correlates with extension rate of the filament (Keijzer, 1987), and the suggestion that anther dehiscence is preceded by dehydration of the locules and that water is exported through the filaments to the petals (Bonner and Dickinson, 1990). An alternative model suggests that JA regulates programmed cell death in the anther as part of the dehiscence process (Zhao and Ma, 2000). Rieu et al. (2003) reported evidence for involvement of ethylene signalling in dehiscence. Tobacco plants insensitive to ethylene developed structurally normal anthers but dehiscence was late and no longer synchronized with flower opening, correlated with delays in the degeneration of stomium cells and dehydration of the anther. Treatment of nearly mature anthers with ethylene accelerated dehiscence in wild-type plants. Ethylene and JA may perform the same role in tobacco and A. thaliana, respectively, or act redundantly in both species. The fact that A. thaliana mutants such as dde1, which cannot synthesize JA within the stamens, eventually undergo dehiscence might represent evidence in support of redundant pathways.

322

R.J. Scott et al.

Conclusion Twelve years ago little was known about genetic control of stamen development beyond specification of stamen primordia by floral homeotic genes. New footholds have been established in several areas, notably patterning of the microsporangium, regulation of meiosis and anther dehiscence. Other processes, such as genesis of the pollen wall, have proved less tractable, perhaps because of a lack of informative mutants. Important questions remain to be addressed, particularly those concerning the establishment of cell fates within the microsporangium, the molecular regulation of the very different nuclear events in the meiocytes and tapetal cells, the link between sporophytically controlled callose degradation and male gametophyte development, the ‘molecular mechanics’ of pollen wall formation and bridging the gap between JA levels and water movement during dehiscence. Understanding stamen development is key to the successful exploitation of plant breeding systems and, in a world of diminishing resources and an increasing need for efficient food production, solving these questions should be accorded a high priority.

Acknowledgements We are grateful to all those in the Scott and Dickinson laboratories who have contributed over the years to our work on anther and pollen development, particularly Anuj Bhatt, Claudia Canales, Franc¸ois Guerineau, Diane Hird, Rachel Hodge, Wyatt Paul, Liz Porter, Ursula Potter, Mike Roberts, Judith Sheldon, Anna Sorensen and Dawn Worrall.

References Aarts, M.G.M., Hodge, R., Kalantidis, K., Florack, D., Wilson, Z.A., Mulligan, B.J., Stiekema, W.J., Scott, R. and Pereira, A. (1997) The Arabidopsis MALE STERILITY 2 protein shares similarity with reductases in elongation/condensation complexes. Plant Journal 12, 615–623. Ahlers, F., Thom, I., Lambert, J., Kuckuk, R. and Wiermann, R. (1999) 1H NMR analysis of sporopollenin from Typha angustifolia. Phytochemistry 50, 1095–1098. Amagai, M., Ariizumi, T., Endo, M., Hatakeyama, K., Kuwata, C., Shibata, D., Toriyama, K. and Watanabe, M. (2003) Identification of anther-specific genes in a cruciferous model plant, Arabidopsis thaliana, by using a com-

bination of Arabidopsis macroarray and mRNA derived from Brassica oleracea. Sexual Plant Reproduction 15, 213–220. Bai, X., Peirson, B.N., Dong, R., Xue, C. and Makaroff, C.A. (1999) Isolation and characterization of SYN1, a RAD21-like gene essential for meiosis in Arabidopsis. Plant Cell 11, 417– 430. Barbacar, N., Hinnisdaels, S., Farbos, I., Mone´ger, F., Lardon, A., Deliche`re, C., Mouras, A. and Negrutiu, I. (1997) Isolation of early genes expressed in reproductive organs of the dioecious white campion (Silene latifolia) by subtraction cloning using an asexual mutant. Plant Journal 12, 805–817.

Stamen Development Beals, T.P. and Goldberg, R.B. (1997) A novel cell ablation strategy blocks tobacco anther dehiscence. Plant Cell 9, 1527–1545. Bhandari, N.N. (1984) The microsporangium. In: Johri, B.J. (ed.) Embryology of Angiosperms. Springer, Berlin, pp. 53–121. Bhatt, A.M., Lister, C., Page, T., Fransz, P., Findlay, K., Jones, G.H., Dickinson, H.G. and Dean, C. (1999) The DIF1 gene of Arabidopsis is required for meiotic chromosome segregation and belongs to the REC8 / RAD21 cohesin gene family. Plant Journal 19, 463–472. Bhatt, A.M., Canales, C. and Dickinson, H.G. (2001) Plant meiosis: the means to 1N. Trends in Plant Science 16, 114–121. Blackmore, S. and Barnes, S.H. (1990) Pollen wall development in angiosperms. In: Blackmore, S. and Knox, R.B. (eds) Microspores: Evolution and Ontogeny. Academic Press, London, pp. 173–192. Blackmore, S. and Crane, P.R. (1988) The systematic implications of pollen and spore ontogeny. In: Humphries, C.J. (ed.) Ontogeny and Systematics. Columbia University Press, New York, pp. 83–115. Bonner, L.J. and Dickinson, H.G. (1989) Anther dehiscence in Lycopersicon esculentum Mill. New Phytology 113, 97–115. Bonner, L.J. and Dickinson, H.G. (1990) Anther dehiscence in Lycopersicon esculentum. 2. Water relations. New Phytology 115, 367–375. Bowman, J.L., Smyth, D.R. and Meyerowitz, E.M. (1989) Genes directing flower development in Arabidopsis. Plant Cell 1, 37–52. Bowman, J.L., Drews, G.N. and Meyerowitz, E.M. (1991) Expression of the Arabidopsis floral homeotic gene AGAMOUS is restricted to specific cell types late in flower development. Plant Cell 3, 749–758.

323 Bowman, J.L., Sakai, H., Jack, T., Weigel, D., Mayer, U. and Meyerowitz, E.M. (1992) SUPERMAN, a regulator of floral homeotic genes in Arabidopsis. Development 114, 599–615. Bowman, J.L., Eshed, Y. and Baum, S.F. (2002) Establishment of polarity in angiosperm lateral organs. Trends in Genetics 18, 134–141. Brown, R.C. and Lemmon, B.E. (1988) Microtubules associated with simultaneous cytokinesis of coenocytic microsporocytes. American Journal of Botany 75, 1848–1856. Brown, R.C. and Lemmon, B.E. (2001) The cytoskeleton and spatial control of cytokinesis in the plant life cycle. Protoplasma 215, 35–49. Bucciaglia, P.A. and Smith, A.G. (1994) Cloning and characterisation of TAG1, a tobacco anther b-1,3-glucanase expressed during tetrad dissolution. Plant Molecular Biology 24, 903–914. Calderon-Urrea A. and Dellaporta, S.L. (1999) Cell death and cell protection genes determine the fate of pistils in maize. Development 126, 435– 441. Canales, C., Bhatt, A.M., Scott, R.J. and Dickinson, H.G. (2002) EXTRA SPOROGENOUS CELLS (EXS), a putative LRR receptor kinase, regulates male germline cell number and tapetal identy, and promotes seed development in Arabidopsis. Current Biology 12, 1718–1727. Carpenter R. and Coen, E.S. (1990) Floral homeotic mutations produced by transposon mutagenesis in Antirrhinum majus. Genes & Development 4, 1483–1493. Caryl, A.P., Jones, G.H. and Franklin, F.C.H. (2003) Dissecting plant meiosis using Arabidopsis thaliana mutants. Journal of Experimental Botany 54, 25–38. Chaubal, R., Zanella, C., Trimnell, M.R., Fox, T.W., Albertsen, M.C. and Bedinger, P. (2000) Two male-sterile

324

R.J. Scott et al. mutants of Zea mays (Poaceae) with an extra cell division in the anther wall. American Journal of Botany 87, 1193–1201. Chaubal, R., Anderson, J.R., Trimnell, M.R., Fox, T.W., Albertsen, M.C. and Bedinger, P. (2003) The transformation of anthers in the msca1 mutant of maize. Planta 216, 778–788. Chaudhury, A. (1993) Nuclear genes controlling male fertility. Plant Cell 5, 1277–1283. Chaudhury, A.M., Lavithis, M., Taylor, P.E., Craig, S., Singh, M.B., Signer, E.R., Knox, R.B. and Dennis, E.S. (1994) Genetic control of male fertility in Arabidopsis thaliana: structural analysis of premeiotic developmental mutants. Sexual Plant Reproduction 7, 17–28. Chmelnitsky, I., Sobolev, I., Barg, R., Shabtai, S. and Salts, Y. (2003) Isolation and preliminary characterization of tomato petal- and stamen-specific cDNAs from a subtracted and equilibrated cDNA library. Euphytica 129, 229–236. Clark, J. (2001) Meristems, start your signalling. Current Opinion in Plant Biology 4, 28–32. Clark, S.E., Running, M.P. and Meyerowitz, E.M. (1993) CLAVATA1, a regulator of meristem and flower development in Arabidopsis. Development 119, 397–418. Coen, E.S. and Meyerowitz, E.M. (1991) The war of the whorls: genetic interactions controlling flower development. Nature 353, 31–37. Crossley, S.J., Greenland, A.J. and Dickinson, H.G. (1995) The characterisation of tapetum-specific cDNAs isolated from a Lilium henryi L. meiocyte subtractive cDNA library. Planta 196, 523–529. Curtis, I.S., Caiping, H., Scott, R.J., Power, J.B. and Davey, M.R. (1996) Genomic male sterility in lettuce, a baseline for the production of F1 hybrids. Plant Science 113, 113–119.

Davis, G.L. (1966) Systematic Embryology of the Angiosperms. John Wiley & Sons, New York. Dawson, J., Wilson, Z.A., Aarts, M.G.M., Braithwaite, A.F., Briarty, L.G. and Mulligan, B.J. (1993) Microspore and pollen development in six male-sterile mutants of Arabidopsis thaliana. Journal of Botany 71, 629–638. ¨ zen, E., Vizir, I., Van Dawson, J., So Waeyenberge, S., Wilson, Z.A. and Mulligan, B.J. (1999) Characterization and genetic mapping of a mutation (ms35) which prevents anther dehiscence in Arabidopsis thaliana by affecting secondary wall thickening in the endothecium. New Phytology 144, 213–222. Delmer, D.P. (1987) Cellulose biosynthesis. Annual Review of Plant Physiology 38, 259–290. Dickinson, H.G. (1970) The fine structure of a peritapetal membrane investing the microsporangium of Pinus banksiana. New Phytology 69, 1065– 1068. Dickinson, H.G. (1987) The physiology and biochemistry of meiosis in the anther. International Review of Cytology 107, 79–109. Dickinson, H.G. (1995) Dry stigmas, water and self-incompatibility in Brassica. Sexual Plant Reproduction 8, 1–10. Dickinson, H.G. and Bell, P.R. (1976) The changes in the tapetum of Pinus banksiana accompanying formation and maturation of the pollen. Annals of Botany 40, 1101–1109. Dickinson, H.G. and Lewis, D. (1973) The formation of the tryphine coating the pollen grains of Raphanus and its properties relating to the selfincompatibility system. Proceedings of the Royal Society of London B 184, 149–165. Dickinson, H.G. and Sheldon, J.S. (1984) A radial pattern of microtubules extending between the nuclear envelope and the plasma membrane during

Stamen Development early male haplophase in flowering plants. Planta 161, 86–90. Dickinson, H.G. and Sheldon, J.S. (1986) The generation of patterning at the plasma membrane of the young microspore of Lilium. In: Blackmore, S. and Ferguson, I.K. (eds) Pollen and Spores: Form and Function. Academic Press, New York, pp. 1–17. Domı´nguez, E., Mercado, J.A., Quesada, M.A. and Heredia, A. (1999) Pollen sporopollenin: degradation and structural elucidation. Sexual Plant Reproduction 12, 171–178. Doughty, J., Dixon, S., Hiscock, S.J., Willis, A.C., Parkin, I.A.P. and Dickinson, H.G. (1998) PCP-A1, a defensinlike Brassica pollen coat protein that binds the S locus glycoprotein, is the product of gametophytic gene expression. Plant Cell 10, 1333–1347. Dover, G.A. (1972) The organisation and polarity of pollen mother cells of Triticum aestivum. Journal of Cell Science 11, 699–711. Echlin, P. and Godwin, H. (1968) The ultrastructure and ontogeny of pollen in Helleborus foetidus. L. II. Pollen grain development through the special callose wall stage. Journal of Cell Science 3, 175–186. Endo, M., Matsubara, H., Kokubun, T., Masuko, H., Takahata, Y., Tsuchiya, T., Fukuda, H., Demura, T. and Watanabe, M. (2002) The advantages of cDNA microarray as an effective tool for identification of reproductive organ-specific genes in a model legume, Lotus japonicus. FEBS Letters 514, 229–237. Erdtman, G. (1969) Handbook of Palynology. Hafner, Copenhagen. Feys, B.F, Benedetti, C.E, Penford, C.N. and Turner, J.G. (1994) Arabidopsis mutants selected for resistance to the phytotoxin coronatine are male sterile, insensitive to methyl jasmonate, and resistant to a bacterial pathogen. Plant Cell 6, 751–759. Fornara, F., Marziani, G., Mizzi, L., Kater, M. and Colombo, L. (2003) MADS-

325 box genes controlling flower development in rice. Plant Biology 5, 16–22. Francz, P., de Jong, H., Lysak, M., Castiglione, M.R. and Schubert, I. (2002) Interphase chromosomes of Arabidopsis are organized in well-defined chromocenters from which euchromatin loops emanate. Proceedings of the National Academy of Sciences USA 99, 14584–14589. Frankel, R., Izhar, S. and Nitsan, J. (1969) Timing of callase activity and cytoplasmic male sterility in Petunia. Biochemical Genetics 3, 451–455. Goldberg, R.B., Beals, T.P. and Sanders, P.M. (1993) Anther development: basic principles and practical applications. Plant Cell 5, 1217–1229. Hanson, M.R. and Bentolila, S. (2004) Interactions of mitochondrial and nuclear genes that affect male gametophyte development. Plant Cell 16 (Suppl.), 154–169. Heslop-Harrison, J. (1963) An ultrastructural study of pollen wall ontogeny in Silene pendula. Grana Palynol 4, 1–24. Heslop-Harrison, J. (1964) Cell walls, cell membranes, and protoplasmic connections during meiosis and pollen development. In: Linskens, H.F. (ed.) Pollen Physiology and Fertilization. North Holland, Amsterdam, pp. 39–47. Heslop-Harrison, J. (1966) Cytoplasmic connections between angiosperm meiocytes. Annals of Botany 30, 221–230. Heslop-Harrison, J. (1968) Pollen wall development. Science 16, 230–207. Heslop-Harrison, J. (1972) Pattern in plant cell walls: morphogenesis in miniature. Proceedings of the Royal Institute of Great Britain 45, 335–352. Heslop-Harrison, J. and Mackenzie, A. (1967) Autoradiography of soluble [2-I4C] thymidine derivatives during meiosis and microsporogenesis in Lilium anthers. Journal of Cell Science 2, 387–400.

326

R.J. Scott et al. Hill, J.P. and Lord, E.M. (1989) Floral development in Arabidopsis thaliana: a comparison of the wild type and the homeotic pistillata mutant. Canadian Journal of Botany 67, 2922– 2936. Hird, D.L., Worrall, D., Hodge, R., Smartt, S., Paul, W. and Scott, R. (1993) The anther-specific protein encoded by the Brassica napus and Arabidopsis thaliana A6 gene displays similarity to b-1,3-glucanase expressed during tetrad dissolution. Plant Journal 4, 1023–1033. Honma, T. and Goto, K. (2001) Complexes of MADS-box proteins are sufficient to convert leaves into floral organs. Nature 409, 525–529. Hu¨lskamp, M., Parekh, N.S., Grini, P., Schneitz, K., Zimmermann, I., Lolle, S.J. and Pruitt, R.E. (1997) The STUD gene is required for malespecific cytokinesis after telophase II of meiosis in Arabidopsis thaliana. Developmental Biology 187, 114–124. Irish, V. (1999) Petal and stamen development. Current Topics in Developmental Biology 41, 133–161. Ishiguro, S., Kawai-Oda, A., Ueda, J., Nishida, I. and Okada, K. (2001) The DEFECTIVE IN ANTHER DEHISCENCE1 gene encodes a novel phospholipase A1 catalyzing the initial step of jasmonic acid biosynthesis, which synchronizes pollen maturation, anther dehiscence, and flower opening in Arabidopsis. Plant Cell 13, 2191– 2209. Ito, T., Wellmer, F., Yu, H., Das, P., Ito, N., Alves-Ferreira, M., Riechmann, J.L. and Meyerowitz, E.M. (2005) The homeotic protein AGAMOUS controls microsporogenesis by regulation of SPOROCYTELESS. Nature 430, 356–360. Izhar, S. and Frankel, R. (1971) Mechanism of male sterility in Petunia: the relationship between pH, callase activity in the anthers, and the breakdown of the microsporogenesis. Theoretical and Applied Genetics 41, 104–108.

Jack, T. (2001) Relearning our ABCs: new twists on an old model. Trends in Plant Science 6, 310–316. Jack, T. (2004) Molecular and genetic mechanisms of floral control. Plant Cell 16 (Suppl.), 1–17. Jenik, P.D. and Irish, V.F. (2000) Regulation of cell proliferation patterns by homeotic genes during Arabidopsis floral development. Development 127, 1267–1276. Jin, H.L., Cominelli, E., Bailey, P., Parr, A., Mehrtens, F., Jones, J., Tonelli, C., Weisshaar, B. and Martin, C. (2000) Transcriptional repression by AtMYB4 controls production of UV-protecting sunscreens in Arabidopsis. EMBO Journal 19, 6150–6161. Jin, W., Horner, H.T., Palmer, R.G. and Shoemaker, R.C. (1999) Analysis and mapping of gene families encoding beta-1,3-glucanases of soybean. Genetics 153, 445–452. Kaul, M.L.H. (1988) Male Sterility in Higher Plants. Monographs on Theoretical and Applied Genetics 10. Springer, Berlin. Keijzer, C.J. (1987) The processes of anther dehiscence and pollen dispersal. Part I: the opening mechanism of longitudinally dehiscing anthers. New Phytologist 105, 487–498. Kerim, T., Imin, N., Weinman, J.J. and Rolfe, B.G. (2003) Proteome analysis of male gametophyte development in rice anthers. Proteomics 3, 738–751. Kolattukudy, P.E. and Espelie, K.E. (1985) Biosynthesis of cutin, suberin, and associated waxes. In: Higuchi, T. (ed.) Biosynthesis and Biodegradation of Wood Components. Academic Press, New York, pp. 161–207. Koltunow, A.M., Truettner, J., Cox, K.H., Wallroth, M. and Goldberg, R.B. (1990) Different temporal and spatial gene expression patterns occur during anther development. Plant Cell 2, 1201–1224. Liu, X.C. and Dickinson, H.G. (1989) Cellular energy levels and their effect on male cell abortion in cytoplasmically

Stamen Development male sterile lines of Petunia hybrida. Sexual Plant Reproduction 2, 167– 172. Liu, X.C., Jones, K.G. and Dickinson, H.G. (1987) DNA synthesis and cytoplasmic differentiation in tapetal cells of normal and CMS lines of Petunia hybrida. Theoretical and Applied Genetics 74, 846–851. Lohmann, J.U. and Weigel, D. (2002) Building beauty: the genetic control of floral patterning. Developmental Cell 2, 135–142. Luo, D., Carpenter, R., Vincent, C., Copsey, L. and Coen, E. (1995) Origin of floral asymmetry in Antirrhinum. Nature 383, 794–799. Martinez-Perez, E., Shaw, P., Reader, S., Aragon-Alcaide, L., Miller, T. and Moore, G. (1999) Homologous chromosome pairing in wheat. Journal of Cell Science 112, 1761–1769. McConn, M. and Browse, J. (1996) The critical requirement for linolenic acid is pollen development, not photosynthesis, in an Arabidopsis mutant. Plant Cell 8, 403–416. McCormick, S. (2004) Control of male gametophyte development. Plant Cell 16 (Suppl.), 142–153. Moss, G. and Heslop-Harrison, J. (1967) A cyto-chemical study of DNA, RNA and protein in the developing maize anther. II. Observations. Annals of Botany 31, 555–572. Murgia, M., Charzynska, M., Rougier, M. and Cresti, M. (1991) Secretory tapetum of Brassica oleracea L – polarity and ultrastructural features. Sexual Plant Reproduction 4, 28–35. Nacken, W.K.F., Huijser, P., Beltran, J.-P., Saedler, H. and Sommer, H. (1991) Molecular characterization of two stamen-specific genes, tap1 and fil1, that are expressed in the wild type, but not in the deficiens mutant of Antirrhinum majus. Molecular & General Genetics 229, 129–136. Nanda, K. and Gupta, S.C. (1978) Studies in the Bignoniaceae II: ontogeny of the dimorphic anther tapetum in Tecoma.

327 American Journal of Botany 65, 400–405. Nonomura, K.I., Miyoshi, K., Eiguchi, M., Suzuki, T., Miyao, A., Hirochika, H. and Kurata, N. (2003) The MSP1 gene is necessary to restrict the number of cells entering into male and female sporogenesis and to initiate anther wall formation in rice. Plant Cell 15, 1728–1739. Østergaard, L., Petersen, M., Mattsson, O. and Mundy, J. (2002) An Arabidopsis callose synthase. Plant Molecular Biology 49, 559–566. Owen, H.A. and Makaroff, C.A. (1995) Ultrastructure of microsporogenesis and microgametogenesis in Arabidopsis thaliana (L.) Heynh. Ecotype Wassilewskija (Brassicaceae). Protoplasma 185, 7–21. Pacini, E. (1990) Tapetum and microspore function. In: Blackmore, S. and Knox, R.B. (eds) Microspores: Evolution and Ontogeny. Academic Press, London. Park, J.-H., Halitschke, R., Kim, H.B., Baldwin, I.T., Feldmann, K.A. and Feyereisen, R. (2002) A knock-out mutation in allene oxide synthase results in male sterility and defective wound signal transduction in Arabidopsis due to a block in jasmonic acid biosynthesis. Plant Journal 31, 1–12. Paxson-Sowders, D.M., Dodrill, C.H., Owen, H.A. and Makaroff, C.A. (2001) DEX1, a novel plant protein, is required for exine pattern formation during pollen development in Arabidopsis. Plant Physiology 127, 1739– 1749. Pelaz, S., Ditta, G.S., Baumann, E., Wisman, E. and Yanofsky, M.F. (2000) B and C floral organ identity functions require SEPALLATA MADS-box genes. Nature 405, 200–203. Periasamy, K. and Amalathas, J. (1991) Absence of callose and tetrad in the microsporogenesis of Pandanus odoratissimus with well-formed pollen exine. Annals of Botany 67, 29–33.

328

R.J. Scott et al. Prahl, A.K., Springstubbe, H., Grumbach, K. and Wiermann, R. (1985) Studies on sporopollenin biosynthesis: the effect of inhibitors of carotenoid biosynthesis on sporopollenin accumulation. Zeitschrift fu¨r Naturforschung 40c, 621–626. Rabitsch, K.P., Toth, A., Galova, M., Schleiffer, A., Schaffner, G., Aigner, E., Rupp, C., Penkner, A.M., MorenoBorchart, A.C., Primig, M., Esposito, R.E., Klein, F., Knop, M. and Nasmyth, K. (2001) A screen for genes required for meiosis and spore formation based on whole-genome expression. Current Biology 11, 1001–1009. Reznickova, S.A. and Dickinson, H.G. (1982) Ultrastructural aspects of storage lipid metabolism in the tapetum of Lilium hybridum var. enchantment. Planta 155, 400–408. Rhee, S.Y. and Somerville, C.R. (1998) Tetrad pollen formation in quartet mutants of Arabidopsis thaliana is associated with persistence of pectic polysaccharides of the pollen mother cell wall. Plant Journal 15, 79–88. Riechmann, J.L., Wang, M.Q. and Meyerowitz, E.M. (1996) DNA-binding properties of Arabidopsis MADS domain homeotic proteins APETALA1, APETALA3, PISTILLATA and AGAMOUS. Nucleic Acids Research 24, 3134–3141. Rieu, I., Wolters-Arts, M., Derksen, J., Mariani, C. and Weterings, K. (2003) Ethylene regulates the timing of anther dehiscence in tobacco. Planta 217, 131–137. Roberts, I.N., Harrod, G. and Dickinson, H.G. (1984) Pollen stigma interactions in Brassica oleracea. I. Ultrastructure and physiology of the stigmatic papillar cells. Journal of Cell Science 66, 241–253. Rogers, C.M. and Harris, B.D. (1969) Pollen exine deposition, a clue to its control. American Journal of Botany 56, 101–106. Rowley, J.R. and Southworth, D. (1967) Deposition of sporopollenin on lamel-

lae of unit membrane dimensions. Nature 213, 703–704. Rubinelli, P., Hu, Y. and Ma, H. (1998) Identification, sequence analysis and expression studies of novel anther-specific genes of Arabidopsis thaliana. Plant Molecular Biology 37, 607–619. Ruiter, R.K., Van Eldik, G.J., Van Herpen, R.M., Schrauwen, J.A. and Wullems, G.J. (1997a) Characterization of oleosins in the pollen coat of Brassica oleracea. Plant Cell 9, 1621–1631. Ruiter, R.K., Mettenmeyer, T., van Laarhoven, D., Van Eldik, G.J., Doughty, J., van Herpen, M.M.A., Schrawwen, J.A.M., Dickinson, H.G. and Wullems, G.J. (1997b) Proteins of the pollen coat of Brassica oleracea. Journal of Plant Physiology 150, 85–91. Sablowski, R.W.M. and Meyerowitz, E.M. (1998) A homolog of NO APICAL MERISTEM is an immediate target of the floral homeotic genes APETALA3/ PISTILLATA. Cell 92, 93–103. Sanders, P.M., Bui, A.Q., Weterings, K., McIntire, K.N., Hsu, Y.-C., Lee, P.Y., Truong, M.T., Beals, T.P. and Goldberg, R.B. (1999) Anther developmental defects in Arabidopsis thaliana male-sterile mutants. Sexual Plant Reproduction 11, 297–322. Sanders, P.M., Lee, P.Y., Biesgen, C., Boone, J.D., Beals, T.P., Weiler, E.W. and Goldberg, R.B. (2000) The Arabidopsis DELAYED DEHISCENCE1 gene encodes an enzyme in the jasmonic acid synthesis pathway. Plant Cell 12, 1041–1061. Sawa, S., Ito, T., Shimura, Y. and Okada, K. (1999) FILAMENTOUS FLOWER controls the formation and development of Arabidopsis inflorescences and floral meristems. Plant Cell 11, 69–86. Schiefthaler, U., Balasubramanian, S., Sieber, P., Chevalier, D., Wisman, E. and Schneitz, K. (1999) Molecular analysis of NOZZLE, a gene involved in pattern formation and early sporogenesis during sex organ development in

Stamen Development Arabidopsis thaliana. Proceedings of the National Academy of Sciences USA 96, 11664–11669. Scott, R.J. (1994) Pollen exine – the sporopollenin enigma and the physics of pattern. In: Scott, R.J. and Stead A.D. (eds) Molecular and Cellular Aspects of Plant Reproduction. Cambridge University Press, Cambridge, UK, pp. 49–81. Scott, R., Hodge, R., Paul, W. and Draper. J. (1991a) The molecular biology of anther differentiation. Plant Science 80, 167–191. Scott, R., Dagless, E., Hodge, R., Paul, W., Soufleri, I. and Draper, J. (1991b) Patterns of gene expression in developing anthers of Brassica napus. Plant Molecular Biology 17, 195– 207. Scott, R.J., Spielman, M. and Dickinson, H.G. (2004) Stamen structure and function. Plant Cell 16(Suppl.), 46– 60. Shaw, G. (1971) The chemistry of sporopollenin. In: Brooks, J., Grant, P.R., Muir, M., van Gijzel, P. and Shaw, G. (eds) Sporopollenin. Academic Press, London. Sheldon, J.M. and Dickinson, H.G. (1983) Determination of patterning in the pollen wall of Lilium henryi. Journal of Cell Science 63, 191–208. Sheldon, J.M. and Dickinson, H.G. (1986) Pollen wall formation in Lilium: the effect of chaotropic agents, and the organisation of the microtubular cytoskeleton during pattern development. Planta 168, 11–23. Shen, J.B. and Hsu, F.C. (1992) Brassica anther-specific genes: characterization and in situ localization of expression. Molecular & General Genetics 234, 379–389. Sheridan, W.F., Golubeva, E.A., Abrhamova, L.I. and Golubovskaya, I.N. (1999) The mac1 mutation alters the developmental fate of the hypodermal cells and their cellular progeny in the maize anther. Genetics 153, 933–941.

329 Siegfried, K.R., Eshed, Y., Baum, S.F., Otsuga, D., Drews, G.N. and Bowman, J.L. (1999) Members of the YABBY gene family specify abaxial cell fate in Arabidopsis. Development 126, 4117–4128. Smyth, D.R.,Bowman,J.L.andMeyerowitz, E.M. (1990) Early flower development in Arabidopsis. Plant Cell 2, 755–767. Sorensen, A., Guerineau, F., CanalesHolzeis, C., Dickinson, H.G. and Scott, R.J. (2002) A novel extinction screen in Arabidopsis thaliana identifies mutant plants defective in early microsporangial development. Plant Journal 29, 581–594. ¨ ber, S., Unte, U.S., Sorensen, A.-M., Kro Huijser, P., Dekker, K. and Saedler, H. (2003) The Arabidopsis ABORTED MICROSPORES (AMS) gene encodes a MYC class transcription factor. Plant Journal 33, 413–423. Spielman, M., Preuss, D., Li, F.-L., Browne, W.E., Scott, R.J. and Dickinson, H.G. (1997) TETRASPORE is required for male meiotic cytokinesis in Arabidopsis thaliana. Development 124, 2645–2657. St Johnston, D. and Nu¨sslein-Volhard, C. (1992) The origin of pattern and polarity in the Drosophila embryo. Cell 68, 201–219. Stadler, R., Truernit, E., Gahrtz, M. and Sauer, N. (1999) The AtSUC1 sucrose carrier may represent the osmotic driving force for anther dehiscence and pollen tube growth in Arabidopsis. Plant Journal 19, 269–278. Staehelin, L.A. and Hepler, P.K. (1996) Cytokinesis in higher plants. Cell 84, 821–824. Steiner-Lange, S., Unte, U.S., Eckstein, L., Yang, C., Wilson, Z.A., Schmelzer, E., Dekker, K. and Saedler, H. (2003) Disruption of Arabidopsis thaliana MYB26 results in male sterility due to non-dehiscent anthers. Plant Journal 34, 519–528. Stieglitz, H. (1977) Role of b-1,3-glucanase release. Developmental Biology 57, 87–97.

330

R.J. Scott et al. Stieglitz, H. and Stern, H. (1973) Regulation of b-1,3-glucanase activity in developing anthers of Lilium. Developmental Biology 34, 169–173. Stintzi, A. and Browse, J. (2000) The Arabidopsis male-sterile mutant, opr3, lacks the 12-oxophytodienoic acid reductase required for jasmonate synthesis. Proceedings of the National Academy of Sciences USA 97, 10625–10630. Tanurdzik, M. and Banks, J. (2004) Sexdetermining mechanisms in land plants. Plant Cell 16, S61–S71. Taylor, P.E., Glover, J.A., Lavithis, M., Craig, S., Singh, M.B., Knox, R.B., Dennis, E.S. and Chaudhury, A.M. (1998) Genetic control of male fertility in Arabidopsis thaliana: structural analyses of postmeiotic developmental mutants. Planta 205, 492–505. Theißen, G. (2001) Development of floral organ identity: stories from the MADS house. Current Opinion in Plant Biology 4, 75–85. Torres-Ruiz, R and Ju¨rgens, G. (1994) Mutations in the FASS gene uncouple pattern formation and morphogenesis in Arabidopsis development. Development 120, 2967– 2978. Tsuchiya, T., Toriyama, K., Ejiri, S. and Hinata, K. (1994) Molecular characterization of rice genes specially expressed in the anther tapetum. Plant Molecular Biology 26, 1737–1746. Ursin, V.M., Yamaguchi, J. and McCormick, S. (1989) Gametophytic and sporophytic expression of antherspecific genes in developing tomato anthers. Plant Cell 1, 727–736. Vega, J.M. and Feldman, M. (1998) Effect of the pairing gene Ph1 and premeiotic colchicine treatment on intra- and interchromosome pairing of isochromosomes in common wheat. Genetics 150, 1199–1208. Vershon, A.K and Pierce, M. (2000) Transcriptional regulation of meiosis in yeast. Current Opinions in Cell Biology 12, 334–339.

von Malek, B., van der Graaff, E., Schneitz, K. and Keller, B. (2002) The Arabidopsis male-sterile mutant dde2–2 is defective in the ALLENE OXIDE SYNTHASE gene encoding one of the key enzymes of the jasmonic acid biosynthesis pathway. Planta 216, 187–192. Warmke, H.E. and Overman, M.A. (1972) Cytoplasmic male sterility in sorghum. 1: callose behavior in fertile and sterile anthers. Journal of Heredity 63, 103–108. Waterkeyn, L. (1962) Les parois microsporocytaires de nature callosique chez Helleborus et Tradescantia. Cellule 62, 225–255. Waterkeyn, L. and Bienfait, A. (1970) On a possible function of the callosic special wall in Ipomoea purpurea (L) Roth. Grana 10, 13–20. Wilson, Z.A., Morroll, S.M., Dawson, J., Swarup, R. and Tighe, P.J. (2001) The Arabidopsis MALE STERILITY1 (MS1) gene is a transcriptional regulator of male gametogenesis, with homology to the PHD-finger family of transcription factors. Plant Journal 28, 27–39. Worrall, D., Hird, D.L., Hodge, R., Paul, W., Draper, J. and Scott, R. (1992) Premature dissolution of the microsporocyte callose wall causes male sterility in transgenic tobacco. Plant Cell 4, 759–771. Wright, S.Y., Suner, M.-M., Bell, P.J., Vaudin, M. and Greenland, A.J. (1993) Isolation and characterization of male flower cDNAs from maize. Plant Journal 3, 41–49. Xie, D.-X., Feys, B.F., James, S., NietoRostro, M. and Turner, J.G. (1998) COI1: an Arabidopsis gene required for jasmonate-regulated defense and fertility. Science 280, 1091–1094. Yamaguchi, T., Tanaka Y., Hayashi, T. and Koike, S. (1997) Molecular cloning of rice anther b-1,3-glucanase gene expressed at the dissolution of tetrads. Plant Physiology 114 (Suppl. S.), 1236–1236.

Stamen Development Yamaguchi, T., Nakayama, K., Hayashi, T., Tanaka, Y. and Koike, S. (2002) Molecular cloning and characterisation of a novel b -1,3-glucanase gene from rice. Bioscience, Biotechnology, and Biochemistry 66, 1403–1406. Yang, C.-Y., Spielman, M., Coles, J.P., Li, Y., Ghelani, S., Bourdon, V., Brown, R.C., Lemmon, B.E., Scott, R.J., and Dickinson, H.G. (2003c). TETRASPORE encodes a kinesin required for male meiotic cytokinesis in Arabidopsis. Plant Journal 34, 229–240. Yang, S.-L., Mao, H.-Z., Xie, L.-F., Pauh, C.S., Jiang, L., Sundaresan, V. and Ye, D. (2003a) The Tapetum Determinant1 gene controls the identity of the tapetal cells in the Arabidopsis anther. Abstract, ISPMB 2003, Barcelona, Spain. Yang, W.C., Ye, D., Xu, J. and Sundaresan, V. (1999) The SPOROCYTELESS gene of Arabidopsis is required for initiation of sporogenesis and en-

331 codes a novel nuclear protein. Genes and Development 13, 2108–2117. Yang, X., Makaroff, C.A. and Ma, H. (2003b) The Arabidopsis MALE MEIOCYTE DEATH1 gene encodes a PHD-finger protein that is required for male meiosis. Plant Cell 15, 1281–1295. Zhao, D. and Ma, H. (2000) Male fertility: a case of enzyme identity. Current Biology 10, R904–R907. Zhao, D.-Z., Wang, G.-F., Speal, B. and Ma, H. (2002) The EXCESS MICROSPOROCYTES1 gene encodes a putative leucine-rich repeat receptor kinase that controls somatic and reproductive cell fates in the Arabidopsis anther. Genes & Development 16, 2021–2031. Zik, M. and Irish, V.F. (2003) Global identification of target genes regulated by APETALA3 and PISTILLATA floral homeotic gene action. Plant Cell 15, 207–222.

12

Genes Regulating Ovule Development

J. BROADHVEST1 AND B.A. HAUSER2 1

Bayer Bioscience NV, Technologie Park 38, B-9052 Ghent, Belgium; Department of Botany, University of Florida, Gainesville, FL 32611-8526, USA 2

Introduction Ovules are the precursors of seeds. More specifically, they are sporophytic structures and the site of megagametogenesis or female gamete formation that culminates in the formation of the haploid embryo sac. The prototypical angiosperm ovule consists of three parts: (i) the nucellus, where megasporogenesis occurs; (ii) one or two integuments, which cover and nourish the megagametophyte; and (iii) the funiculus, a supporting stalk that connects the ovule to the placenta (Fig. 12.1). The presence and morphogenesis of these structures are species dependent and result in the range of different ovule shapes observed in seed plants. Once fertilized, sporophytes play important roles during seed development: (i) the integumentary tapetum nourishes the developing embryo and endosperm; (ii) the integuments differentiate into the seed coat, which can regulate dormancy and seed germination; and (iii) the funiculus may contribute to the retention of the seed until dissemination of the progeny is required. Seeds are needed for the propagation of most plant species, but they have also become, through crop cultivation, a food source for humans and livestock. Because of this important role in human and animal nutrition, plant ovule and embryo sac development has been intensively investigated by basic and applied researchers. This review covers the known molecular mechanisms involved in ovule organogenesis, but it does not discuss megagametogenesis or embryogenesis. While most of the molecular work has been done using the model plant Arabidopsis, research from other plant systems is also incorporated into this discussion. The following generalizations can be made about ovule ontology. Ovule primordia emerge from the placental tissue (Fig. 12.1A). The distal end of the primordia differentiates into the nucellus, inside which meiosis and embryo sac development takes place. The central domain, termed the chalaza, is the site of integument initiation and growth. The proximal region of the primordia will develop into the funiculus (Fig. 12.1B). Finally, the developmental pattern of each of those structures, as well as the integument numbers (1 or 2), affects the 332

ßCAB International 2006. The Molecular Biology and Biotechnology of Flowering, 2nd edn (ed. B.R. Jordan)

Genes Regulating Ovule Development

333

Fig. 12.1. Development of a bitegmic ovule. (A) Ovule primordia emerge from the placenta. These primordia differentiate into three distinct zones: the nucellus (n), chalaza (c) and funiculus (f). (B) The inner integument (ii) and outer integument (oi) initiate from the chalaza. A cell within the nucellus undergoes meiosis, forming a large megaspore (m). (C) This megaspore differentiates into the embryo sac, which contains two synergids (s), an egg cell (e) and the central cell (cc). While this is happening, the oi undergoes asymmetric growth, causing the distal tip of the ovule to curve until it is adjacent to the funiculus.

overall morphology of the ovule in a species-specific manner. Before addressing ovule development per se, we briefly describe hypotheses that account for the origin of the ovule.

Ovule Evolution Fossil record and evolution of the integuments The nucellus derives from a megasporangium. Based on fossil evidence, palaeobotanists have proposed that the inner integument evolved from the fusion of lobed or filamentous structures in the vicinity of this central megasporangium (Stewart, 1983; Herr, 1995; Kenrick and Crane, 1997). Most of the hypotheses differ on the origin of the filaments from which the ancestral integument arose in progymnosperm (Andrews, 1963). The axillary theory, which, has fallen out of favour after evaluating fossils, proposed that integuments derive from foliar appendages (Worsdell, 1904). The telome theory (Zimmerman, 1952) postulates that peripheral lateral sterile branches and sterilized sporangia fused together to enclose a central fertile megasporangium. The synangium theory (Benson, 1904) postulates that the primordial integument arose exclusively from fusion of modified sporangia around a central fertile sporangium. More recently, Kenrick and Crane (1997) proposed that clusters of sporangia found in progymnosperms were the source of the formation of an integument in a single evolutionary step. Thus, there is a general consensus that fusion of these filaments was the process leading to the origin of this structure. There also seems to be a consensus that this primordial, or solitary, integument in gymnosperms corresponds to the inner integument in bitegmic ovules.

334

J. Broadhvest and B.A. Hauser

While a single integument is found in gymnosperms, both basal and most derived angiosperms bear bitegmic ovules, indicating that the precursor of angiosperms had two integuments (Doyle, 1978; Doyle and Donoghue, 1986). Although some extant angiosperms have only one integument, this unitegmy results from the fusion of the two integuments or the loss of one of the two. In short, the outer integument arose during the evolution of angiosperms. The identity of the original structures from which the outer integument arose in angiosperms remains nebulous. In some extinct species, cupules surrounded unitegmic ovules and had lobed, and occasionally recurved, filaments that may have fused to form the oi (Andrews, 1963). Interestingly, cupules have been found on a female reproductive structure where telomes had only partially fused, indicating that the inner integument may not have completely evolved in the angiosperm precursor.

Molecular Evolution Research on the morphogenesis of ovules and the molecular genetic model of ovule development offer some insights into the evolution of integuments. For example, some Arabidopsis ovule mutant phenotypes are suggestive of potential ovule evolutionary steps found in the fossil record. The bitegmic Arabidopsis ovule can be rendered unitegmic by mutations in the INNER NO OUTER (INO) or ABERRANT TESTA SHAPE (ATS) genes. In ino mutant ovules, an outer integument primordium emerges, but does not develop. Consequently, the inner integument and nucellus stand upright (Baker et al., 1997; Schneitz et al., 1997). The ino mutant phenotype clearly illustrates that development of the integuments is under separate genetic control, which is consistent with their evolution. While the aberrant testa shape (ats) mutant is also unitegmic, this integument has anatomical properties of both the inner and outer integuments (Leon-Kloosterziel et al., 1994). While no ovule mutant bearing only an outer integument has been identified up to now, the mimicry of earlier steps in ovule evolution has been suggested for these two ovule mutants. Based on phenotype of bell1 (bel1) mutant ovules, Herr (1995) argues that some of the filamentous structures that fused to form the integuments were fertile. In bel1 mutants, the embryo sac arrests at a late stage of megagametogenesis and a single highly-lobed integument-like structure (ils) forms (Robinson-Beers et al., 1992; Modrusan et al., 1994). It has been suggested the bel1 ils, which arises as a single structure, corresponds to the oi (Modrusan et al., 1994; Herr, 1995; Balasubramanian and Schneitz, 2002). Herr (1995) observed ‘sporangium-like’ structures in some of the ils protrusions of bel1 mutants, which led him to hypothesize that some of the ancestral filamentous structure were, in fact, fertile (Herr, 1995). In strubellig (sub) mutant ovules, unfused protrusions developing from the outer integuments are often found (Schneitz et al., 1997). This phenotype is amplified in the pretty few seeds2 ( pfs2) sub double mutant where, instead of sheathing integuments, unfused filaments grow (Park et al., 2004). While

Genes Regulating Ovule Development

335

interpretation of any evolutionary aspect of these mutations may be subject to controversy, orthologues of some of these genes were found in other plant species and the studies of their regulation may help in understanding ovule evolution. Also, the above ‘evolutionary’ mutant phenotypes were attributable to the putative loss of activity of the genes. It will be of interest to see if evolution can also be mimicked by way of gain-of-function experiments, which may be expected from the evolutionary process. Based on genetic and molecular analyses, it was noticed that many genes regulating ovule development might also regulate other developmental pathways, including flower organogenesis (or vice versa). Since ovules arose before the four angiosperm floral whorls, it has been suggested that many of the genes regulating ovule development spontaneously duplicated, that their functions diverged during floral evolution and that ‘ovule’ organogenesis pathways were later diverted to floral development (Gasser et al., 1998; Broadhvest et al., 2000). Evidence for this proposed evolutionary pathway is supplied not only by the phenotypic mutants but also by the high redundancy and presence of many floral/ovule paralogues in plant genomes. Four-fifths of the genes regulating ovule development (Table 12.1) fall into this category. For example, many MADS-box genes that encode putative transcription factors involved in regulation of floral organogenesis also regulate ovule development (Purugganan et al., 1995; Purugganan, 1997). A number of them share partially redundant genetic activities. Recently, the MADS-box genes were split into two phylogenetic clades: (i) a reproductive-organ patterning, or C-class, function; and (ii) a group of genes regulating ovule development (Kramer et al., 2004). It was shown that in many plant species MADS-box proteins can form homo- or heteromultimeric complexes both in vivo and in vitro, and some of these complexes are required to confer activity on these proteins (Egea-Cortines et al., 1999; Favaro et al., 2002, 2003; Immink et al., 2002). These studies have shown that some MADS-box genes are functionally redundant when expressed in a tissue-dependent manner. These data support the evolutionary sharing of ovule–floral molecular pathways. While neither of the hypotheses described in the above section satisfactorily accounts for all available data, they provide models that will continue to be tested by researchers as they gather new fossil and molecular data.

Ovule Development The genes that are known to regulate ovule development are listed in Table 12.1, as are their functions. As noted above, many of these genes also regulate other plant developmental pathways, especially floral morphogenesis. Unless they shed light on ovule formation, the functions of these genes will not be reported in this review. Other reviews have covered the subject of Arabidopsis ovule development (Angenent and Colombo, 1996; Gasser et al., 1998; Grossniklaus and Schneitz, 1998; Schneitz et al., 1998b; Schneitz, 1999), and will serve as the starting point for the present discussion.

336

Table 12.1. Regulators of Arabidopsis ovule development. Ovule differentiation and patterning

Growth and maturation

Growth/patterning Class

Initiation

CUC1, CUC2

NAC domains

Initiation

BEL1

TALE homeobox

Initiation

STK

MADS

Initiation

SHP1, SHP2 SEP1, 2 and 3

MADS MADS

WUS AG

Homeobox MADS

Initiation Carpel initiation Initiation Initiation identity

HLL

Ribosomal protein Novel

NZZ (SPL)

Proximal

Chalaza

Patterning identity

Patterning growth

Apical

Identity

Nucellus

Cell division

Inner

Identity

Outer

Identity

Patterning growth

Primary references Aida et al. (1997); Ishida et al. (2000); Takada et al. (2001) Modrusan et al. (1994); Ray et al. (1994); Reiser et al. (1995); Western and Haughn (1999); Bellaoui et al. (2001) Pinyopich et al. (2003) Pinyopich et al. (2003) Pelaz et al. (2000)

Patterning

Survival

Survival

Survival

Patterning

Patterning

Initiation

Initiation

Initiation

Initiation

Groß-Hardt et al. (2002) Western and Haughn (1999) Schneitz et al. (1998a); Skinner et al. (2001) Yang et al. (1999); Schiefthaler et al. (1999); Balasubramanian and Schneitz (2000)

J. Broadhvest and B.A. Hauser

Gene

Integument

N.P.

ANT

AP2 domain

DCL1

Identity (non-fusion) Initiation growth

Identity (non-fusion)

Leon-Kloosterziel et al. (1994)

Initiation growth

dsRNA RNAse

Cell expansion

Cell expansion

ACR

Receptor kinase

Organization

Organization

SIN2 AP2 TSO1

N.P. AP2 domain TSO1 domain

Cell division Growth Direction

Cell division Growth Direction of growth

SUP

Zinc finger

Asymmetric growth

INO

YABBY

Asymmetric growth

LUG

Putative repressor Repressor

Elliott et al. (1996); Klucher et al. (1996); Krizek (1999); Mizukami et al. (2000) Ray et al. (1996b); Golden et al. (2002) Tanaka et al. (2002); Gifford et al. (2003) Broadhvest et al. (2000) Jofuku et al. (1994) Hauser et al. (1998, 2000); Song et al. (2000) Gaiser et al. (1995); Balasubramanian and Schneitz (2002); Meister et al. (2002) Villanueva et al. (1999); Meister et al. (2002) Liu and Meyerowitz (1995); Conner and Liu (2000) Franks et al. (2002)

SEU

Patterning growth*

Patterning growth

Growth

Growth Growth

Genes Regulating Ovule Development

ATS

N.P. Not published. *Only WUS and SUP appear to be non-cell autonomous.

337

338

J. Broadhvest and B.A. Hauser

Placentation and ovule initiation In the angiosperms, the diversity of fruit shape, size and anatomy derives from variations in the location and arrangement of the placenta. In Arabidopsis, the placenta is located on the adaxial portion of the replum, where the carpel wall fuses with the septum. Ovule primordia initiate from the placenta and generally derive from the L1, L2 and L3 layers (Bouman, 1984). These primordia emerge as finger-like outgrowths and exhibit radial symmetry (Fig. 12.1A). Extensive mutagenesis suggests that ovule initiation in Arabidopsis is under the control of redundant gene activities. All known mutations identified to affect ovule initiation also affect pistil formation and/or placentation. It is expected that genes involved in placenta formation may affect ovule initiation, development and numbers. For example, Arabidopsis mutations that lead to the ectopic formation of placental tissues bear, as one might expect, ectopic ovules (Bowman et al., 1999). Only genes expressed in or affecting both placenta and ovule development are discussed in this section as it is more suggestive of a direct role in ovule initiation. In spatula (spt) mutants, the carpels do not completely fuse and form fewer ovules than in wild-type. These defects derive from reduced growth of the carpel margins (Heisler et al., 2001). Interestingly, SPT is expressed in ovule primordia and in expanding integuments, but redundant activities with other transcription factors may mask its role in ovule development (Alvarez and Smyth, 1999; Heisler et al., 2001). While the respective single mutations bear defective ovules (see below), the synergistic seuss leunig or ant leunig double mutants have unfused carpels with altered internal structures, including the placenta, and a reduced number or absence of ovules. (Liu et al., 2000; Franks et al., 2002). In cup-shaped cotyledon1 (cuc1) cuc2 double mutants, most commonly the septum fails to fuse, resulting in pistils with reduced numbers of ovules that range from none to ten (Ishida et al., 2000). Given the pistil phenotype, the reduction in ovule number may result from a reduction in the size of the placenta in which both CUC1 and 2 genes are expressed. The CUC3 gene is expressed between ovule primordia and hypothesized to establish a developmental boundary there (Vroemen et al., 2003). In petunia, FLORAL BINDING PROTEIN7 (FBP7) and FBP11 genes are exclusively expressed during placenta, ovule and seed development (Angenent et al., 1995; Colombo et al., 1995, 1997). In transgenic plants co-suppressing these genes, spaghetti-shaped structures that resemble style and stigma tissues were developed instead of ovules (Angenent et al., 1995). The presence of wild-type ovules was also observed in these plants, which might be due to variability in co-suppression or stochastic effects on the activity of other MADS-box genes. The recent description of FBP11 loss-of-function mutants that displayed no discernable ovule or floral phenotype (Vandenbussche et al., 2003) indicates another gene, perhaps FBP7, function during ovule initiation. In petunia, high levels of FBP11 expression cause ovule-like structures to develop on the petals (Colombo et al., 1995), indicating that overexpression of FBP11 alone can lead to ovule initiation there. FBP11 is the sole protein identified to have this activity. These data are consistent with

Genes Regulating Ovule Development

339

FBP11 and related MADS-box genes acting redundantly to stimulate ovule initiation or promote ovule identity. In rice, the ovuleless (ovl) mutant was recently isolated from in vitro cultures (Yamaki and Nagato, 2002). In this mutant, 85% of the mature flowers fail to form ovules. Unfortunately, the OVL gene has not yet been identified.

Ovule formation on sepals Ovules are one of the last sporophytic tissues formed in plants. Many genes active in ovule formation can also be necessary for earlier stages of plant development, thus making it difficult to evaluate their role in ovule morphogenesis. For example, the AGAMOUS (AG) gene is expressed in ovules, stamens and carpels. Plants with the ag mutation lack carpels, so it is difficult to assess the effect of AG on ovule formation (Bowman et al., 1991a). To examine the possible effects of such genes, researchers have used the homeotic transformation of sepals into carpels to examine ovule development. This transformation occurs in ap2 loss-of-function plants and in transgenic plants overexpressing AG and AG-like genes. The sepals of these plants exhibit carpel-like features, including ovules and ovule-like structures (Bowman et al., 1991b; Mizukami and Ma, 1992; Kempin et al., 1993; Western and Haughn, 1999; Honma and Goto, 2001; Favaro et al., 2003). While the presence of ovules on sepals permits the study of these loci, these analyses are far from optimal. In ap2–6 mutant plants, the morphology of the ovules on carpelloid sepals ranges from normal to highly aberrant (Kunst et al., 1989). Depending on the ap2 mutant allele, ovules in the fourth whorl may or may not exhibit a phenotype that differs from wild-type. For those ap2 ovules that exhibit a phenotype, integument morphology is altered near the micropyle. These results indicate that some additional morphogenic activities that are necessary for wild-type ovule formation are missing or at insufficient levels in the outer floral whorl of ap2 mutants. Despite this limitation, analyses of mutants in the ap2 background have provided evidence of the genetic functions of AG and BEL1 during ovule formation. Both of these genes are expressed in the chalaza of developing ovule primordia, then AG transcripts become restricted to the endothelium later in ovule formation (Bowman et al., 1991a; Reiser et al., 1995). Indirect evidence suggested to Ray et al. (1994) that AG is active during ovule development. Evaluation of the bel1–3 and ag-1 mutants in the ap2–6 mutant background led to an increase in filamentous outgrowths on sepal margins and a concomitant decrease in the number of mature ovules (Western and Haughn, 1999). Based on analysis of the double and triple mutant combinations, Western and Haughn propose that the BEL1 and AG proteins act in a partially redundant manner to maintain ovule identity as this organ develops. In addition, formation of the abnormal integuments, which are characteristic of the bel1 mutant, requires AG activity (Western and Haughn, 1999). Both the SHATTERPROOF1 (SHP1) and SHP2 genes, which are AG paralogues (Rounsley et al., 1995; Purugganan, 1997), regulate ovule formation.

340

J. Broadhvest and B.A. Hauser

Ectopic expression of either SHP gene causes homeotic conversion of the sepals into carpelloid tissues (Pinyopich et al., 2003), similar to AG overexpression (Mizukami and Ma, 1992). This result indicates that while the shp1,2 double mutant does not have an ovule phenotype, the SHP genes do affect ovule development. In the ap2–6 ag shp1 shp2 quadruple mutant, the ovule primordia on the sepals had fewer ovule characteristics than in the ap2–6 ag plants (Pinyopich et al., 2003). Considering that these genes are highly expressed in developing ovules and based on this quadruple mutant phenotype, it appears that SHP1, SHP2 and AG all specify ovule identity or regulate its development in a redundant fashion. While these types of experiments give insights into the role of some genes in ovule development, it is clear that additional factors are required for ovule morphogenesis than are present in ap2–6 sepals. With the availability of more ovule-specific promoters, the function of redundant genes could be tested using tissue-specific RNA interference (RNAi) approaches. These experiments would complement the observations made in ap2 carpelloid sepal ovules. Growth and patterning of ovule primordia After emergence from the placental tissue, the Arabidopsis ovule primordium forms a featureless cylindrical structure that becomes patterned along the proximal–distal axis resulting in three separate developmental regions (Fig. 12.1A). Each region will later give rise to different ovule structures. The distal end of the primordia differentiates into the nucellus, inside which meiosis and embryo sac development takes place. The central domain, the chalaza, is the site of integument initiation and growth. The proximal region of the primordia will further develop into the funiculus. Based on molecular expression data, patterning is thought to occur prior to the actual tripartite zonation of the primordia. For example, the BEL1 gene is expressed in the chalaza before this structure is anatomically established (Reiser et al., 1995). HLL is required for the growth of ovule primordia. In hll mutants, ovule primordia are smaller than wild-type and their development stops after the first visible signs of the tripartite zonation (Schneitz et al., 1998a). In the distal zone, hll primordia display anatomical signs of cell death (Schneitz et al., 1998a; Skinner et al., 2001). The HLL gene encodes a ribosomal protein, similar to the L14 protein found in eubacteria, which localizes to the mitochondria (Skinner et al., 2001). The HLL gene is expressed at high levels in pistils and ovules while a functionally redundant paralogue of HLL, called HLP, is expressed elsewhere at high levels. In the hll mutant, the levels of HLL and HLP proteins are insufficient to sustain growth of the ovule. Consequently, the HLL protein links the biochemical energy from the mitochondria to ovule primordia growth (Skinner et al., 2001). The ant and short integuments2 (sin2) loci, which as single mutants do not exhibit obvious defects in primordium growth, have shown synergistic effects when present in hll mutants. In the ant hll double mutants, ovule primordia growth stops earlier than in hll mutants and the chalaza-specific expression of BEL1 transcripts is detected initially at the base of emerging ovule primordia. This led Schneitz et al. (1998a) to hypothesize that patterning of

Genes Regulating Ovule Development

341

ovule primordia occurs as the primordia emerge from the placenta. Zonation of the nucellus occurs first, then the chalaza, and finally the funiculus. While zonation occurs appropriately in this double mutant, most of the proximal zone, the funiculus, fails to form. Since the ANT gene regulates growth (Krizek, 1999; Mizukami and Fischer, 2000), loss of this activity could slow the development of ovules, allowing cell death to terminate development even earlier in hll ant double mutants than is true in hll mutants The sin2 hll ovule primordia are even more severely affected than ant hll ovules. Growth of these ovule primordia ceases before they completely emerge from the placenta (Broadhvest et al., 2000). Based on the patterning model proposed by Schneitz et al. (1998a), these primordia contain only the apical portion of nucellus. Analysis of the sin2 mutant phenotype indicates that this locus regulates cell division in ovules (Broadhvest et al., 2000). If the growth of sin2 ovules occurs even more slowly than in ant ovules, it seems reasonable to speculate that hll-induced cell death leads to very early termination of ovule development in hll sin2 plants. The NOZZLE (NZZ ) (also called SPOROCYTELESS), which is a putative transcription factor, plays an important role in patterning of the ovule primordium and in nucellus development (Schiefthaler et al., 1999; Yang et al., 1999). In nzz mutants, the microspores and megaspores do not differentiate. This effect on patterning was suggested because ANT and BEL1, which are normally expressed in the chalaza, are detected in cells at the apical end of the nzz ovule primordium, indicating that the nucellar domain does not get established properly in these mutants (Balasubramanian and Schneitz, 2000, 2002). Overlapping roles for proximal/distal pattern or zonation were also suggested for BEL1 and NZZ since nzz bel1 ovules lack the chalaza region of ovule primordia (Balasubramanian and Schneitz, 2000). NZZ recently was shown to bind INO protein (Sieber et al., 2004), which regulates development of the outer integument. A number of double mutant phenotypes similarly indicate that NZZ not only regulates megaspore mother cell (MMC) differentiation, but also integument morphogenesis (Balasubramanian and Schneitz, 2000, 2002; Park et al., 2004). The current genetic model of ovule development links the patterning of primordia with concomitant growth, suggesting that mutants lacking only the funiculus might be found. Primordia growth and patterning are currently regarded as two interrelated processes displaying some redundant gene activities. The availability of more cell- or tissue-specific markers that are confined to a developmental field during ovule development could be used to test the current patterning models. For example, WUSCHEL (WUS), which is expressed in the nucellus anlagen and could specify nucellus identity is not expressed in nzz and sin2 hll mutant ovules. Nucellus growth and differentiation The MMC, located in the centre of the nucellus, undergoes meiosis forming four megaspores, three of which will degenerate (Fig. 12.1B). Concomitant to further nucellus growth, the surviving megaspore will differentiate into the gametophyte (Fig. 12.1C). In nzz mutants, the nucellus and the embryo sac fail to

342

J. Broadhvest and B.A. Hauser

develop leading to a very short nucellus where growth terminates not long after zonation of the primordia (Schiefthaler et al., 1999; Yang et al., 1999). Based on the phenotype of the nzz ovules, it is not clear whether the meiotic defect causes the characteristic nucellus phenotype or vice versa (Schiefthaler et al., 1999; Yang et al., 1999; Balasubramanian and Schneitz, 2000, 2002). Genetic analyses with the NZZ gene are complicated by interactions with numerous other loci and multiple development pathways, some of which are affected in a stochastic manner (Balasubramanian and Schneitz, 2000, 2002). This has led to a variety of phenotypes, even among ovules of the same flower (see below), complicating the determination of the role of this gene. Discrepancies in expression data make it hard to pinpoint functions of NZZ with regard to the multiple phenotypes. Using a reporter b-glucuronidase (GUS) construct, Yang et al. (1999) showed that NZZ localizes to the centre of the nucellus just prior to differentiation of the MMC. These data conflict with reports that NZZ transcripts are detected in ovule primordia, nucellus, chalaza and integuments in a complex pattern (Schiefthaler et al., 1999; Balasubramanian and Schneitz, 2000, 2002). The localization of the NZZ protein has not been determined yet, which is hoped to help in understanding the role and mode of action of the NZZ locus. None the less, genetic data indicate that the NZZ gene interacts with loci specifying the proximal–distal and the abaxial–adaxial axes (Balasubramanian and Schneitz, 2000, 2002). Based on the analyses of mutants, PFS2 appears to be required for differentiation of the MMC (Park et al., 2004). PFS2 and NZZ are the only genes known to directly or indirectly affect nucellus development. While other Arabidopsis ovule mutants show a reduced nucellus, each of these also has a defect in integument growth or morphogenesis. Based on these observations, it was suggested that the presence of at least one sheathing integument was necessary for development of the embryo sac and/or nucellus and that a mutation impeding integument growth would have an indirect effect on nucellar growth (Baker et al., 1997). Integument initiation and growth While in some species the integuments can originate from the L1 and L2 layers, in Arabidopsis integuments arise solely from the L1 (Jenik and Irish, 2001). Integuments initiate their growth sequentially in Arabidopsis. First, the inner integument begins from two adjacent cell rows of the upper portion of the chalaza in a synchronized manner to form a ring that encircles the primordium. Later, the outer integument originates from two rows of cells from the lower part of the chalaza. This occurs only on the abaxial, or bottom, side of the primordium. Growth of both integuments occurs simultaneously, although the outer integument undergoes asymmetric growth on the abaxial side while the inner integument grows symmetrically, eventually sheathing the nucellus. Asymmetric growth of the outer integument causes the ovule, including the embryo sac, to curve so that the small opening termed ‘the micropyle’, where the integuments terminate, is adjacent to the funiculus (Fig. 12.1C). The row of integument cells adjacent to the nucellus differentiates into an endothelium, which, in many species, has transfer cell characteristics. This tissue is thought

Genes Regulating Ovule Development

343

to be important for transferring sugars and nutrients into the embryo and endosperm (Kapil and Tiwari, 1978). Genes regulating integument initiation While the initiation of the integuments is not synchronous, many genes regulate the development of both integuments. For example, loss-of-function analyses indicate that ANT activity is necessary for development of both integuments (Elliott et al., 1996; Klucher et al., 1996; Baker et al., 1997). Overexpression experiments indicate that ANT promotes cell growth (Krizek, 1999; Mizukami and Fischer, 2000). This gene is expressed in plant meristems and is especially abundant in the chalaza during integument organogenesis (Elliott et al., 1996; Klucher et al., 1996). In an elegant set of experiments, WUS, a homeodomain protein controlling meristem cell fate, was recently found to be required for integument initiation (Groß-Hardt et al., 2002). Since wus mutant seedlings fail to make flowers, it was impossible to ascertain its role in ovule development. To correct for this defect, WUS was expressed under the control of the CLAVATA1 (CLV1) promoter. In wus mutants containing the CLV1::WUS transgene, the ovules failed to initiate integuments (Groß-Hardt et al., 2002), indicating that WUS is required for integument formation. In this genetic line, ectopic expression of WUS in the chalaza region induced the formation of integument primordia from the funiculus (Groß-Hardt et al., 2002). In wild-type ovules, WUS protein remains in the nucellus, indicating that this protein generates a downstream signal that induces differentiation of chalaza cells. While WUS activity promotes cell division within the floral meristem, it seems that the same transcription factor activates a different set of genes during ovule development. The nature of this signalling remains unclear, but appears to involve ANT and other ovule-patterning genes (Groß-Hardt et al., 2002). Possibly because of its effect on ovule primordia patterning (see above), NZZ regulates the temporal and spatial initiation of both integuments (Balasubramanian and Schneitz, 2000). In the nzz mutants, the two integuments appear to initiate at the same time and more towards the apical portion of the primordia (Balasubramanian and Schneitz, 2000). Unitegmic mutants As described earlier, loss of function of some genes cause unitegmy in Arabidopsis. In inner no outer (ino) mutants, the outer integument primordium forms but fails to grow. In strong ino alleles, the inner integument is phenotypically normal and the ino ovule is straight or orthotropous, instead of curved. It is clear that INO has a role in outer integument initiation and is also required for asymmetric growth of the outer integument (see below). In the ats mutant, a single integument initiates from the chalaza (LeonKloosterziel et al., 1994). While this integument has properties of both the inner and outer integuments, developmental defects result in misshapen seeds (Leon-Kloosterziel et al., 1994). The researchers were unable to determine if this integument results from a fusion of two integuments from inception or if one integument differentiates but takes on properties of both integuments.

344

J. Broadhvest and B.A. Hauser

None the less, the mutant line is of particular interest to scientists studying unitegmic angiosperms, which resulted from the loss of an integument during angiosperm evolution and diversification. Integument growth Once the integuments are initiated, controlled growth must occur to allow the ovule to function and exhibit species-dependent characteristics. Regulators of most growth components (cell elongation, division, etc.) have been identified for integument development (Table 12.1). In ovules, SIN2 is one of the major regulators of cell division. Loss of SIN activity causes an arrest of integument growth after their initiation, but before they sheathe the nucellus (Broadhvest et al., 2000). Flowers of sin2 mutants also have defects in sepal tip development, which also derive from the L1 cell layer. Cell elongation in the integuments requires DICER-LIKE1 (DCL1) (formerly SHORT INTEGUMENTS1) activity (Golden et al., 2002). DCL1 is an important component of the RNAi process and its loss of function causes reduced cell expansion of the integuments, in addition to other pleiotropic effects (Lang et al., 1994; Ray et al., 1996a). Recently, it has been shown that cuc1 cuc2 double mutants, which have pleiotropic effects on plant development that include integument morphogenesis, were regulated via microRNAs (Ishida et al., 2000; Mallory et al., 2004). Thus, RNAi seems to be involved in ovule organogenesis. A number of loss-of-function loci cause altered cellular morphogenesis in ovules. Weak mutant alleles of the floral gene TSO1 cause the integument cells to be disoriented and misshapen, indicating that this gene is involved in regulating the direction of cell expansion and planes of cell division of flowers and integuments (Liu et al., 1997; Hauser et al., 1998). In blasssig (bag) mutants, the misshapen integument cells cause a general disorganization of ovule tissues (Schneitz et al., 1997). In addition, bag ovules commonly fuse together. Fusion of ovules results because the wax on the cell wall of the outer integument layer is not synthesized. To a lesser extent, in the acr4 mutant, integument cell morphogenesis was disorganized and ovule fusion sometimes observed (Gifford et al., 2003). ACR4 encodes a L1-specific leucine-rich receptor kinase (Gifford et al., 2003). In the ovule, ACR4 mRNA is most abundant in the outer layers of the oi, ii and endothelium. In addition to the integument phenotype, the sepal tips, which are also L1 tissues, are misshapen or absent. These observations led to the premise that ACR4 is part of the cell–cell communication in the L1 layer. The sin2 mutant was also shown to affect development of sepal tips and integuments (Broadhvest et al., 2000). It would be interesting to see if these genes are part of a similar developmental pathway. A leucine-rich receptor kinase, SUB, was shown to regulate outer integument morphogenesis as well as whole plant development (Chevalier and Schneitz, 2000). In sub mutant ovules, initiation and growth of the outer integument cells are poorly regulated, leading to the formation of outgrowths. In tousled (tsl) mutant ovules, growth of the integuments was variable, but compared to wild-type the oi often was shorter and the ii longer (Roe et al., 1997b). TSL encodes nuclear serine/threonine protein kinase

Genes Regulating Ovule Development

345

(Roe et al., 1993, 1997a). The loss of two putative transcription repressors, LEUNIG (LUG) or SEUSS (SEU), resulted in tsl-like integument phenotypes (Roe et al., 1997b; Franks et al., 2002). TSL and LUG are elements of the same genetic pathway during ovule development. SEU, LUG and APETALA2 (AP2) are all negative regulators of AG and these genes may interact in a dosage-dependent manner during integument development. In seu-1/seu-1, lug-1/þ, or ap2–1/þ plants, ovules have shortened integuments (Franks et al., 2002). UNICORN (UCN) is a locus that seems to have a very specific role. Loss of UCN activity leads to the development of an outgrowth from the chalaza, which often abnormally emerges from the abaxial side of mature ovules (Schneitz et al., 1997). STYLISH1 (STY1), a RING-motif protein involved in Arabidopsis apical pistil formation is expressed in the ovule primordia, but later becomes restricted to the outer layer of the funiculus, the oi and the tip of the nucellus (Kuusk et al., 2002). While sty1 does not have an ovule phenotype on its own, in sty1 spt double mutants, the outer integuments do not grow normally (Kuusk et al., 2002). Asymmetric growth of the outer integument In Arabidopsis ovules, INO and SUP proteins are required for asymmetric growth of the oi. In ino mutant ovules, the integuments do not curve. The asymmetric growth of the oi results in curvature so extensive that the micropyle is adjacent to the funiculus and the ovule shaped by this growth is anatropous. INO, a member of the YABBY family of proteins, establishes an abaxial identity during lateral organ formation (Siegfried et al., 1999; Villanueva et al., 1999). INO is expressed specifically in the oi initial cells and later becomes spatially restricted to the outer layer of the oi (Villanueva et al., 1999; Balasubramanian and Schneitz, 2000). INO positively regulates the expression of itself, ANT and BEL1 (Meister et al., 2002). In sup mutants the outer integument exhibits reduced curvature as a result of growth on both the abaxial and adaxial sides of the oi. As revealed by in situ hybridization, SUP mRNA accumulates in the funiculus in an area adjacent to the chalaza (Sakai et al., 1995). SUP affects oi growth, indicating that this locus acts non-cell autonomously. Meister et al. (2002) proposed that SUP attenuates INO activity on the abaxial side of the integument, although the mechanism for this activity is not yet clear (Meister et al., 2002). Recently, Nakagawa et al. (2004) identified the SUP in petunia (PhSUP). In the phsup mutants, the ovules also fail to curve and are more orthotropous. Interestingly, petunia ovules are unitegmic. Because SUP affects the outer integument only, this indicates that the petunia integument either resulted from fusion of both integuments or derived solely from the oi. Many classes of proteins were found to bind to the INO promoter, but it is not yet clear what role these proteins have (Meister et al., 2004). In splayed (syd) mutants, vegetative and floral organs grow more slowly than normal. The ovules in these mutants fail to form an embryo sac and exhibit more reduced curvature of the oi than wild type (Wagner and Meyerowitz, 2002). SYD encodes a chromatin-remodelling factor that is expressed almost

346

J. Broadhvest and B.A. Hauser

ubiquitously in plants (Wagner and Meyerowitz, 2002). Loss of activity of this chromatin-remodelling ATPase appears to reduce the growth of the outer integument and disrupt megametogenesis.

Funiculus growth There are no genes that specifically regulate funiculus development, although a few mutants modulate funiculus morphology and cellular organization. In bel1 mutants, the funiculus exhibits an altered arrangement of cell files, resulting in a funiculus that is wider and shorter than normal (Balasubramanian and Schneitz, 2000). With respect to funiculus development, SIN2 activity appears to be redundant with BEL1 because bel1 sin2 funiculi also are shorter and thicker than wild-type or either single mutant (Broadhvest et al., 2000). NZZ is known to restrict funicular growth, so it is logical that genetic synergism on funicular length is observed in nzz, as well as in bel1 and ant mutants (Balasubramanian and Schneitz, 2000). Recently, SEEDSTICK (STK) was also described as regulating funicular width and as having an important role in seed dispersal (Pinyopich et al., 2003). SPT has also been suggested to have an effect on abscission of the funiculus, a conjecture that was based on the presence of the SPT transcript in this structure and the possible role of SPT in seed dehiscence (Heisler et al., 2001). Homeotic conversion or loss of ovule identity Loss of ovule identity is used to describe the only type of homeotic transformation observed up to now in ovules: the transformation of the chalaza, integuments and nucellus into carpels or carpel-like structures. In the ABC model, a major player in carpel identity is AG. Loss of AG activity leads to homeotic transformation of the central region of floral meristems and the reiteration of the ag mutant flower, which develops from the central region of an existing flower (Bowman et al., 1991a,b). During wild-type ovule development, AG is highly expressed in primordia and integuments As described above, bel1 mutants cause the transformation of the integuments into an integument-like structure (ILS). However, sometimes the ILS will develop into a complete miniature pistil, formed of two fused carpels, which will contain bel1 ovules that can reiterate the pistil formation process. The bel1 ILS acquires the identity of the fourth floral whorl, which would be able to respond to the AG protein present in the ILS. In the bel1 mutant where integument identity is lost during development, AG expression is dramatically upregulated (Ray et al., 1994; Reiser et al., 1995). The STK, SHP1 and SHP2 are the genes most closely related to AG (Purugganan et al., 1995). These four genes are each expressed in ovules, and mutational analysis indicates that they share partially redundant roles with one another (Pinyopich et al., 2003). In the stk shp1 shp2 triple mutant, integument identity is lost and the ovule reverts to a carpel identity. This result is consistent with these three genes’ roles of maintaining ovule identity during development or directly regulating ovule development (Pinyopich et al., 2003).

Genes Regulating Ovule Development

347

Hormonal regulation of ovule development We know very little about how hormones affect ovule development. While it is obvious that hormones are involved in plant cellular and tissue growth, there is no obvious link between the different mutant phenotypes observed up to now and their effect on plant hormones. In tobacco, ethylene was shown to be involved in integument growth. Downregulation of the 1-aminocyclopropane1-carboxylate oxidase (ACO) enzyme, which regulates the terminal step in ethylene synthesis, caused female sterility. In these ovules, the integuments were shorter than normal and the embryo sac failed to form. This phenotype could be reversed by applying 2-chloroethylphosphonic acid (De Martinis and Mariani, 1999). Ethylene and auxin appear to be involved in post-pollination ovule initiation in Phalaenopsis orchids. However, because of impacts of these hormonal pathways on the overall fertilization process, it is not clear if these hormones have direct or indirect effects on ovule initiation (Yu and Goh, 2001). Future Much of what we know about ovule development is based on reverse and forward genetic screens in Arabidopsis and a few other dicot plant species. Most loss-of-function screens are based on female sterility and thus may have inadvertently identified only a subset of the morphogenic pathways. Except for putative orthologues of dicot genes, we have very little molecular data on the function of the molecular mechanism involved in monocots. Since more and more plant genomic sequences data will be available in the future, it will be interesting to look at the function of putative orthologues or even of ovulespecific expressed genes using targeted reverse genetic approaches. Much remains to be learned in regard to loss- and gain-of-function of such genes in the ovule. While functional redundancies will always be a hurdle to overcome, the knowledge gained will serve as a base to better understand the evolution and regulation of ovule development in many species, particularly those whose economic value rests in their seeds.

References Aida, M., Ishida, T., Fukaki, H., Fujisawa, H. and Tasaka, M. (1997) Genes involved in organ separation in Arabidopsis: an analysis of the cup-shaped cotyledon mutant. Plant Cell 9, 841– 857. Alvarez, J. and Smyth, D.R. (1999) CRABS CLAW and SPATULA, two Arabidopsis genes that control carpel

development in parallel with AGAMOUS. Development 126, 2377– 2386. Andrews, H.N.J. (1963) Early seed plants. Science 142, 925–931. Angenent, G.C. and Colombo, L. (1996) Molecular control of ovule development. Trends in Plant Science 1, 228–232.

348

J. Broadhvest and B.A. Hauser Angenent, G.C., Franken, J., Busscher, M., Van Dijken, A., Van Went, J.L., Dons, H.J.M. and Van Tunen, A.J. (1995) A novel class of MADS-box genes is involved in ovule development in petunia. Plant Cell 7, 1569–1582. Baker, S.C., Robinson-Beers, K., Villanueva, J.M., Gaiser, J.C. and Gasser, C.S. (1997) Interactions among genes regulating ovule development in Arabidopsis thaliana. Genetics 145, 1109– 1124. Balasubramanian, S. and Schneitz, K. (2000) NOZZLE regulates proximal– distal pattern formation, cell proliferation and early sporogenesis during ovule development in Arabidopsis thaliana. Development 127, 4227– 4238. Balasubramanian, S. and Schneitz, K. (2002) NOZZLE links proximal–distal and adaxial–abaxial pattern formation during ovule development in Arabidopsis thaliana. Development 129, 4291–4300. Bellaoui, M., Pidkowich, M.S., Samach, A., Kushalappa, K., Kohalmi, S.E., Modrusan, Z., Crosby, W.L. and Haughn, G.W. (2001) The Arabidopsis BELL1 and KNOX TALE homeodomain proteins interact through a domain conserved between plants and animals. Plant Cell 13, 2455–2470. Benson, M. (1904) Telangium scotti, a new species of Telangium (Calymmatotheca) showing structure. Annals of Botany 18, 161–167. Bouman, F. (1984) The ovule. In: Johri, B.M. (ed.) Embryology of the Angiosperms. Springer-Verlag, New York, pp. 123–157. Bowman, J.L., Drews, G.N. and Meyerowitz, E.M. (1991a) Expression of the Arabidopsis floral homeotic gene AGAMOUS is restricted to specific cell types late in flower development. Plant Cell 3, 749–758. Bowman, J.L., Smyth, D.R. and Meyerowitz, E.M. (1991b) Genetic interactions among floral homeotic genes of Arabidopsis. Development 112, 1–20.

Bowman, J.L., Baum, S.F., Eshed, Y., Putterill, J. and Alvarez, J. (1999) Molecular genetics of gynoecium development in Arabidopsis. Current Topics in Developmental Biology 45, 155– 205. Broadhvest, J., Baker, S.C. and Gasser, C.S. (2000) SHORT INTEGUMENTS 2 promotes growth during Arabidopsis reproductive development. Genetics 155, 899–907. Chevalier, D. and Schneitz, K. (2000) The Arabidopsis STRUBBELIG gene encodes a distinct putative leucine–rich repeat kinase involved in the control of cell morphogernesis. In: 11th International Conference on Arabidopsis Research. Madison, Wisconsin, Abstract 339. Colombo, L., Franken, J., Koetje, E., Van Went, J., Dons, H.J.M., Angenent, G.C. and Van Tunen, A.J. (1995) The petunia MADS-box gene FBP11 determines ovule identity. Plant Cell 7, 1859–1868. Colombo, L., Franken, J., Van Der Krol, A.R., Wittich, P.E., Dons, H.J.M. and Angenent, G.C. (1997) Downregulation of ovule-specific MADS-box genes from petunia results in maternally controlled defects in seed development. Plant Cell 9, 703–715. Conner, J. and Liu, Z.C. (2000) LEUNIG, a putative transcriptional corepressor that regulates AGAMOUS expression during flower development. Proceedings of the National Academy of Sciences USA 97, 12902–12907. De Martinis, D. and Mariani, C. (1999) Silencing gene expression of the ethylene-forming enzyme results in a reversible inhibition of ovule development in transgenic tobacco plants. Plant Cell 11, 1061–1072. Doyle, J.A. (1978) Origin of the angiosperms. Annual Review of Ecological Systems 9, 365–392. Doyle, J.A. and Donoghue, M.J. (1986) Seed plant phylogeny and the origin of angiosperms: an experimental cladistic approach. Botanical Review 52, 321– 431.

Genes Regulating Ovule Development Egea-Cortines, M., Saedler, H. and Sommer, H. (1999) Ternary complex formation between the MADS-box proteins SQUAMOSA, DEFICIENS and GLOBOSA is involved in the control of floral architecture in Antirrhinum majus. EMBO Journal 18, 5370– 5379. Elliott, R.C., Betzner, A.S., Huttner, E., Oakes, M.P., Tucker, W.Q.J., Gerentes, D., Perez, P. and Smyth, D.R. (1996) AINTEGUMENTA, an APETALA2-like gene of Arabidopsis with pleiotropic roles in ovule development and floral organ growth. Plant Cell 8, 155–168. Favaro, R., Immink, R.G., Ferioli, V., Bernasconi, B., Byzova, M., Angenent, G.C., Kater, M. and Colombo, L. (2002) Ovule-specific MADS-box proteins have conserved protein–protein interactions in monocot and dicot plants. Molecular Genetics and Genomics 268, 152–159. Favaro, R., Pinyopich, A., Battaglia, R., Kooiker, M., Borghi, L., Ditta, G., Yanofsky, M.F., Kater, M.M. and Colombo, L. (2003) MADS-box protein complexes control carpel and ovule development in Arabidopsis. Plant Cell 15, 2603–2611. Franks, R.G., Wang, C., Levin, J.Z. and Liu, Z. (2002) SEUSS, a member of a novel family of plant regulatory proteins, represses floral homeotic gene expression with LEUNIG. Development 129, 253–263. Gaiser, J.C., Robinson-Beers, K. and Gasser, C.S. (1995) The Arabidopsis SUPERMAN gene mediates asymmetric growth of the outer integument of ovules. Plant Cell 7, 333–345. Gasser, C.S., Broadhvest, J. and Hauser, B.A. (1998) Genetic analysis of ovule development. Annual Review of Plant Physiology & Plant Molecular Biology 49, 1–24. Gifford, M.L., Dean, S. and Ingram, G.C. (2003) The Arabidopsis ACR4 gene plays a role in cell layer organisation during ovule integument and sepal

349 margin development. Development 130, 4249–4258. Golden, T.A., Schauer, S.E., Lang, J.D., Pien, S., Mushegian, A.R., Grossniklaus, U., Meinke, D.W. and Ray, A. (2002) SHORT INTEGUMENTS1/ SUSPENSOR1/CARPEL FACTORY, a dicer homolog, is a maternal effect gene required for embryo development in Arabidopsis. Plant Physiology 130, 808–822. Groß-Hardt, R., Lenhard, M. and Laux, T. (2002) WUSCHEL signaling functions in interregional communication during Arabidopsis ovule development. Genes & Development 16, 1129–1138. Grossniklaus, U. and Schneitz, K. (1998) The molecular and genetic basis of ovule and megagametophyte development. Seminars in Cellular and Developmental Biology 9, 227–238. Hauser, B.A., Villanueva, J.M. and Gasser, C.S. (1998) Arabidopsis TSO1 regulates directional processes in cells during floral organogenesis. Genetics 150, 411–423. Hauser, B.A., He, J.Q., Park, S.O. and Gasser, C.S. (2000) TSO1 is a novel protein that modulates cytokinesis and cell expansion in Arabidopsis. Development 127, 2219–2226. Heisler, M.G., Atkinson, A., Bylstra, Y.H., Walsh, R. and Smyth, D.R. (2001) SPATULA, a gene that controls development of carpel margin tissues in Arabidopsis, encodes a bHLH protein. Development 128, 1089–1098. Herr, J.M. (1995) The origin of the ovule. American Journal of Botany 82, 547–564. Honma, T. and Goto, K. (2001) Complexes of MADS-box proteins are sufficient to convert leaves into floral organs. Nature 409, 525–529. Immink, R.G.H., Gadella, T.W.J. Jr, Ferrario, S., Busscher, M. and Angenent, G.C. (2002) Analysis of MADS-box protein-protein interactions in living plant cells. Proceedings of the National Academy of Sciences USA 99, 2416–2421.

350

J. Broadhvest and B.A. Hauser Ishida, T., Aida, M., Takada, S. and Tasaka, M. (2000) Involvement of CUP-SHAPED COTYLEDON genes in gynoecium and ovule development in Arabidopsis thaliana. Plant and Cell Physiology 41, 60–67. Jenik, P.D. and Irish, V.F. (2001) The Arabidopsis floral homeotic gene APETALA3 differentially regulates intercellular signaling required for petal and stamen development. Development 128, 13–23. Jofuku, K.D., den Boer, B.G.W., Van Montagu, M. and Okamuro, J.K. (1994) Control of Arabidopsis flower and seed development by the homeotic gene APETALA2. Plant Cell 6, 1211–1225. Kapil, R.N. and Tiwari, S.C. (1978) The integumentary tapetum. Botanical Review 44, 457–490. Kempin, S.A., Mandel, M.A. and Yanofsky, M.F. (1993) Conversion of perianth into reproductive organs by ectopic expression of the tobacco floral homeotic gene NAG1. Plant Physiology 103, 1041–1046. Kenrick, P. and Crane, P.R. (1997) The Origin and Early Diversification of Land Plants: a Cladistic Study. Smithsonian Institution Press, Washington, DC. Klucher, K.M., Chow, H., Reiser, L. and Fischer, R.L. (1996) The AINTEGUMENTA gene of Arabidopsis required for ovule and female gametophyte development is related to the floral homeotic gene APETALA2. Plant Cell 8, 137–153. Kramer, E.M., Jaramillo, M.A. and Di Stilio, V.S. (2004) Patterns of gene duplication and functional evolution during the diversification of the AGAMOUS subfamily of MADS-box genes in angiosperms. Genetics 166, 1011–1023. Krizek, B.A. (1999) Ectopic expression of AINTEGUMENTA in Arabidopsis plants results in increased growth of floral organs. Developmental Genetics 25, 224–236.

Kunst, L., Klenz, J.E., Martinez-Zapater, J. and Haughn, G.W. (1989) AP2 gene determines the identity of perianth organs in flowers of Arabidopsis thaliana. Plant Cell 1, 1195–1208. Kuusk, S., Sohlberg, J.J., Long, J.A., Fridborg, I. and Sundberg, E. (2002) STY1 and STY2 promote the formation of apical tissues during Arabidopsis gynoecium development. Development 129, 4707–4717. Lang, J.D., Ray, S. and Ray, A. (1994) sin1, a mutation affecting female fertility in Arabidopsis, interacts with mod1, its recessive modifier. Genetics 137, 1101–1110. Leon-Kloosterziel, K.M., Keijzer, C.J. and Koornneef, M. (1994) A seed shape mutant of Arabidopsis that is affected in integument development. Plant Cell 6, 385–392. Liu, Z.C. and Meyerowitz, E.M. (1995) LEUNIG regulates AGAMOUS expression in Arabidopsis flowers. Development 121, 975–991. Liu, Z., Running, M.P. and Meyerowitz, E.M. (1997) TSO1 functions in cell division during Arabidopsis flower development. Development 124, 665–672. Liu, Z., Franks, R.G. and Klink, V.P. (2000) Regulation of gynoecium marginal tissue formation by LEUNIG and AINTEGUMENTA. Plant Cell 12, 1879–1892. Mallory, A.C., Dugas, D.V., Bartel, D.P. and Bartel, B. (2004) MicroRNA regulation of NAC-domain targets is required for proper formation and separation of adjacent embryonic, vegetative, and floral organs. Current Biology 14, 1035–1046. Meister, R.J., Kotow, L.M. and Gasser, C.S. (2002) SUPERMAN attenuates positive INNER NO OUTER autoregulation to maintain polar development of Arabidopsis ovule outer integuments. Development 129, 4281–4289. Meister, R.J., Williams, L.A., Monfared, M.M., Gallagher, T.L., Kraft, E.A., Nelson, C.G. and Gasser, C.S. (2004) Definition and interactions of a positive

Genes Regulating Ovule Development regulatory element of the Arabidopsis INNER NO OUTER promoter. Plant Journal 37, 426–438. Mizukami, Y. and Fischer, R.L. (2000) Plant organ size control: AINTEGUMENTA regulates growth and cell numbers during organogenesis. Proceedings of the National Academy of Sciences USA 97, 942–947. Mizukami, Y. and Ma, H. (1992) Ectopic expression of the floral homeotic gene AGAMOUS in transgenic Arabidopsis plants alters floral organ identity. Cell 71, 119–131. Modrusan, Z., Reiser, L., Feldmann, K.A., Fischer, R.L. and Haughn, G.W. (1994) Homeotic transformation of ovules into carpel-like structures in Arabidopsis. Plant Cell 6, 333–349. Nakagawa, H., Ferrario, S., Angenent, G.C., Kobayashi, A. and Takatsuji, H. (2004) The petunia ortholog of Arabidopsis SUPERMAN plays a distinct role in floral organ morphogenesis. Plant Cell 16, 920–932. Park, S.O., Hwang, S. and Hauser, B.A. (2004) The phenotype of Arabidopsis ovule mutants mimics the morphology of primitive seed plants. Proceedings of the Royal Society London. Series B Biological Sciences 271, 311–316. Pelaz, S., Ditta, G.S., Baumann, E., Wisman, E. and Yanofsky, M.F. (2000) B and C floral organ identity functions require SEPALLATA MADS-box genes. Nature 405, 200–203. Pinyopich, A., Ditta, G.S., Savidge, B., Liljegren, S.J., Baumann, E., Wisman, E. and Yanofsky, M.F. (2003) Assessing the redundancy of MADS-box genes during carpel and ovule development. Nature 424, 85–88. Purugganan, M.D. (1997) The MADS-box floral homeotic gene lineages predate the origin of seed plants: phylogenetic and molecular clock estimates. Journal of Molecular Evolution 45, 392–396. Purugganan, M.D., Rounsley, S.D., Schmidt, R.J. and Yanofsky, M.F. (1995) Molecular evolution of flower development: diversification of the

351 plant MADS-box regulatory gene family. Genetics 140, 345–356. Ray, A., Robinson-Beers, K., Ray, S., Baker, S.C., Lang, J.D., Preuss, D., Milligan, S.B. and Gasser, C.S. (1994) Arabidopsis floral homeotic gene BELL (BEL1) controls ovule development through negative regulation of AGAMOUS gene (AG). Proceedings of the National Academy of Sciences USA 91, 5761–5765. Ray, A., Lang, J.D., Golden, T. and Ray, S. (1996a) SHORT INTEGUMENT (SIN1), a gene required for ovule development in Arabidopsis, also controls flowering time. Development 122, 2631–2638. Ray, S., Golden, T. and Ray, A. (1996b) Maternal effects of the short integument mutation on embryo development in Arabidopsis. Developmental Biology 180, 365–369. Reiser, L., Modrusan, Z., Margossian, L., Samach, A., Ohad, N., Haughn, G.W. and Fischer, R.L. (1995) The BELL1 gene encodes a homeodomain protein involved in pattern formation in the Arabidopsis ovule primordium. Cell 83, 735–742. Robinson-Beers, K., Pruitt, R.E. and Gasser, C.S. (1992) Ovule development in wild-type Arabidopsis and two femalesterile mutants. Plant Cell 4, 1237– 1249. Roe, J.L., Rivin, C.J., Sessions, R.A., Feldmann, K.A. and Zambryski, P.C. (1993) The TOUSLED gene in Arabidopsis thaliana encodes a protein kinase homolog that is required for leaf and flower development. Cell 75, 939–950. Roe, J.L., Durfee, T., Zupan, J.R., Repetti, P.P., McLean, B.G. and Zambryski, P.C. (1997a) TOUSLED is a nuclear serine-threonine protein kinase that requires a coiled-coil region for oligomerization and catalytic activity. Journal of Biological Chemistry 272, 5838–5845. Roe, J.L., Nemhauser, J.L. and Zambryski, P.C. (1997b) TOUSLED

352

J. Broadhvest and B.A. Hauser participates in apical tissue formation during gynoecium development in Arabidopsis. Plant Cell 9, 335–353. Rounsley, S.D., Ditta, G.S. and Yanofsky, M.F. (1995) Diverse roles for MADSbox genes in Arabidopsis development. Plant Cell 7, 1259–1269. Sakai, H., Medrano, L.J. and Meyerowitz, E.M. (1995) Role of SUPERMAN in maintaining Arabidopsis floral whorl boundaries. Nature 378, 199–203. Schiefthaler, U., Balasubramanian, S., Sieber, P., Chevalier, D., Wisman, E. and Schneitz, K. (1999) Molecular analysis of NOZZLE, a gene involved in pattern formation and early sporogenesis during sex organ development in Arabidopsis thaliana. Proceedings of the National Academy of Sciences USA 96, 11664–11669. Schneitz, K. (1999) The molecular and genetic control of ovule development. Current Opinion in Plant Biology 2, 13–17. Schneitz, K., Hulskamp, M., Kopczak, S.D. and Pruitt, R.E. (1997) Dissection of sexual organ ontogenesis: a genetic analysis of ovule development in Arabidopsis thaliana. Development 124, 1367–1376. Schneitz, K., Baker, S.C., Gasser, C.S. and Redweik, A. (1998a) Pattern formation and growth during floral organogenesis: HUELLENLOS and AINTEGUMENTA are required for the formation of the proximal region of the ovule primordium in Arabidopsis thaliana. Development 125, 2555–2563. Schneitz, K., Balasubramanian, S. and Schiefthaler, U. (1998b) Organogenesis in plants: the molecular and genetic control of ovule development. Trends in Plant Science 3, 468–472. Sieber, P., Petrascheck, M., Barberis, A. and Schneitz, K. (2004) Organ polarity in Arabidopsis: NOZZLE physically interacts with members of the YABBY family. Plant Physiology 135, 2172–2185.

Siegfried, K.R., Eshed, Y., Baum, S.F., Otsuga, D., Drews, D.N. and Bowman, J.L. (1999) Members of the YABBY gene family specify abaxial cell fate in Arabidopsis. Development 126, 4117–4128. Skinner, D.J., Baker, S.C., Meister, R.J., Broadhvest, J., Schneitz, K. and Gasser, C.S. (2001) The Arabidopsis HUELLENLOS gene, which is essential for normal ovule development, encodes a mitochondrial ribosomal protein. Plant Cell 13, 2719–2730. Song, J.-L., Leung, T., Ehler, L.K., Wang, C. and Liu, Z. (2000) Regulation of meristem organization and cell division by TSO1, an Arabidopsis gene with cysteine-rich repeats. Development 127, 2207–2217. Stewart, W.N. (1983) Paleobotany and the Evolution of Plants. Cambridge University Press, New York. Takada, S., Hibara, K., Ishida, T. and Tasaka, M. (2001) The CUPSHAPED COTYLEDON1 gene of Arabidopsis regulates shoot apical meristem formation. Development 128, 1127–1135. Tanaka, H., Watanabe, M., Watanabe, D., Tanaka, T., Machida, C. and Machida, Y. (2002) ACR4, a putative receptor kinase gene of Arabidopsis thaliana, that is expressed in the outer cell layers of embryos and plants, is involved in proper embryogenesis. Plant Cell Physiology 43, 419–428. Vandenbussche, M., Zethof, J., Souer, E., Koes, R., Tornielli, G.B., Pezzotti, M., Ferrario, S., Angenent, G.C. and Gerats, T. (2003) Toward the analysis of the petunia MADS-box gene family by reverse and forward transposon insertion mutagenesis approaches: B, C, and D floral organ identity functions require SEPALLATA-like MADS-box genes in petunia. Plant Cell 15, 2680–2693. Villanueva, J.M., Broadhvest, J., Hauser, B.A., Meister, R.J., Schneitz, K. and Gasser, C.S. (1999) INNER NO OUTER regulates abaxial–adaxial patterning in Arabidopsis ovules.

Genes Regulating Ovule Development Genes & Development 13, 3160– 3169. Vroemen, C.W., Mordhorst, A.P., Albrecht, C., Kwaaitaal, M.A. and de Vries, S.C. (2003) The CUP-SHAPED COTYLEDON3 gene is required for boundary and shoot meristem formation in Arabidopsis. Plant Cell 15, 1563–1577. Wagner, D. and Meyerowitz, E.M. (2002) SPLAYED, a novel SWI/SNF ATPase homolog, controls reproductive development in Arabidopsis. Current Biology 12, 85–94. Western, T.L. and Haughn, G.W. (1999) BELL1 and AGAMOUS genes promote ovule identity in Arabidopsis thaliana. Plant Journal 18, 329–336.

353 Worsdell, W.C. (1904) The structure and morphology of the ovule: an historical sketch. Annals of Botany 18, 57–86. Yamaki, S. and Nagato, Y. (2002) OVULELESS gene regulates the initial step of ovule development in rice. Rice Genetics Newsletter 19, 33–35. Yang, C.-D., Ye, D., Xu, J. and Sundaresan, V. (1999) The SPOROCYTELESS gene of Arabidopsis is required for initiation of sporogenesis and encodes a novel nuclear protein. Genes & Development 13, 2108–2117. Yu, H. and Goh, C.J. (2001) Molecular genetics of reproductive biology in orchids. Plant Physiology 127, 1390–1393. Zimmerman, W. (1952) Main results of the ‘Telome theory’. The Palaeobotanist 1, 456–470.

13

The Molecular Biology of Apomixis

R.A. BICKNELL AND A.S. CATANACH New Zealand Institute for Crop & Food Research Ltd, Private Bag 4704, Christchurch, New Zealand

Introduction Apomixis is the subject of a number of recent reviews. Some authors have focused on the potential agronomic and economic benefits of apomixis (Hanna, 1995; Savidan, 2000a,b). Others have discussed the developmental and genetic basis of apomixis (Koltunow et al., 1995a; Grimanelli et al., 2001a; Koltunow and Grossniklaus, 2003; Bicknell and Koltunow, 2004) or reviewed current theory regarding the evolutionary and ecological implications of the trait (Van Dijk and Van Damme, 2000; Van Dijk, 2003). This chapter provides a general overview of the field of apomixis research, with an emphasis on our current understanding of the genetic mechanism(s) that underlie its expression in flowering plants, and a sketch of the strategies being taken to introduce this characteristic into crop species.

What is apomixis and where is it found? Apomixis is the asexual formation of seeds, avoiding the processes of meiotic reduction and recombination at fertilization, and leading to the formation of genetically uniform progeny. The term ‘apomixis’ was first coined by Winkler (1908) to describe ‘substitution of sexual reproduction by an asexual multiplication process without nucleus or cell fusion’. This is a broad definition that can be interpreted to include all forms of asexual reproduction. Many of the earlier reported references to apomixis use the term in this manner, and this has unfortunately led to some confusion. It is now generally accepted that the definition of apomixis be restricted to the formation of seeds by an asexual process, a definition analogous to the term ‘agamospermous’ (Richards, 1997). The first description of apomixis in plants is attributed to Smith (1841), who noted that a single plant of Alchornea ilicifolia (syn Caelebogyne ilicifolia) 354

ßCAB International 2006. The Molecular Biology and Biotechnology of Flowering, 2nd edn (ed. B.R. Jordan)

The Molecular Biology of Apomixis

355

from Australia set abundant levels of germinable seed at Kew gardens, UK. This species is dioecious and the individual involved had a morphology typical of the female form, which produces no pollen. Smith surmised, therefore, that this plant was able to form seed by a process analogous to parthenogenesis, which had been known in animals such as aphids since the reported observations of Bonnet (1745). Since Smith’s original observation, more than 400 flowering plant taxa have been reported to feature apomictic forms, including species of both monocotyledonous and dicotyledonous genera (Carman, 1997). This is almost certainly a significant underestimate of the prevalence of the trait. To conclusively prove the presence of apomixis it is necessary to conduct a combination of genetic and cytological analyses (Nogler, 1984a), yet very few plants have been adequately examined at this level. In addition, most, if not all apomicts are also capable of some level of sexual reproduction. This ‘facultativeness’ can mask the presence of the trait, particularly when it is expressed only at a low level. It seems, therefore, that as our understanding of apomixis increases and our ability to discern and detect it improves many new apomicts will be discovered and some plants that were originally described as apomicts may need to be reclassified. The distribution of the trait indicates a polyphyletic pattern of origin. Apomixis occurs widely throughout the angiosperms. It is particularly prevalent in Poaceae (a monocotyledonous family) and the Asteraceae (a dicotyledonous family), both of which are regarded as evolutionarily advanced, yet it appears to be either absent or very uncommon in more basal groups such as the Nymphaceae and the Winteraceae. In addition, the diversity of mechanisms of apomixis (see below) indicates that it has arisen many times. In most cases apomicts within a single genus utilize a similar mechanism to produce asexual seed. However, members of closely related genera can often differ widely in the mechanism they employ. As mentioned above, asexual reproduction in animals has been known since the time of Bonnet (1745), who noted its occurrence amongst aphids. It is now known to occur in a range of animal groups, being particularly prevalent amongst molluscs, lumbricid and planarian worms, arthropods, rotifers and insects. It is also known to occur in some vertebrates, most notably species of reptiles and amphibians (Suomalainen et al., 1987). Amongst the lower green plants forms of asexual reproduction are relatively common. There is some debate about the use of the term apomixis in these cases as the structures present are different from those of higher plants, although they are often regarded as analogous. Amongst ferns in particular, there are developmental mechanisms that show clear similarities with some of the different modes of apomixis known to occur in higher plants (Asker and Jerling, 1992). Curiously, apomixis is presently unknown amongst the gymnosperms. However, androgenesis or ‘paternal apomixis’ has recently been reported in the rare Mediterranean cypress tree Cupressus dupreziana, where embryos develop parthenogenetically from diploid pollen nourished by maternal reproductive tissues (El Maataoui and Pichot, 2001; Pichot et al., 2001). The widespread distribution of apomixis amongst flowering plants, its apparent polyphyletic origins and the occurrence of analogous phenomena

356

R.A. Bicknell and A.S. Catanach

in other organisms has led to the proposition that it has arisen de novo many times from the modification of conserved functions, common to a wide variety of multicellular organisms. Specifically, the functions of meiosis, egg cell specification and zygote specification appear to be involved in this transition. Recently, molecular evidence has emerged to support this hypothesis. This is discussed below. Before proceeding further, a brief comment regarding the taxonomy of apomicts is necessary. As indicated above, apomicts are typically classified by genus and species in a manner similar to other flowering plants. The Linnaean binomial classification system, however, is poorly suited to these plants, where asexual increase frequently leads to the formation and maintenance of numerous morphologically distinct, yet interfertile varieties growing true-to-type from seed. The taxonomy of such ‘agamic complexes’ is a difficult and contentious task (Dickinson, 1998; Horandl, 1998). Examples of genera in which the apomictic mode of reproduction is strongly combined with morphological polymorphism include: Alchemilla, Hieracium, Poa, Potentilla, Ranunculus, Rubus and Taraxacum (Czapik, 1994).

Apomixis in agriculture Apomixis leads to the formation of large genetically uniform populations and perpetuates hybrid vigour through successive seed generations. It could therefore be used in agriculture to perpetuate elite hybrid cultivars indefinitely through seed. In some cases apomixis can proceed without pollination, indicating that it may be possible to use it to avoid some current complications associated with sexual reproduction such as pollinator availability and cross compatibility. Finally, as viruses are very seldom transmitted by seed, apomixis may overcome viral transfer in plants that are typically propagated vegetatively such as potatoes (Hanna, 1995; Jefferson and Bicknell, 1995; Koltunow et al., 1995a; Savidan, 2000a,b). The value of these opportunities will vary between crops and between production systems. For farmers in the developed world the greatest benefit is expected to be the economic production of new, advanced, high-yielding cultivars for use in mechanized agricultural systems. Conversely, for farmers in the developing world the greatest benefits are expected to relate to the breeding of robust, high-yielding cultivars for specific environments, improvements in the security of food supply and greater autonomy over cultivar ownership (Bicknell and Bicknell, 1999; Toenniessen, 2001). Apomixis, however, is very uncommon among crop species. The main exceptions appear to be tropical and subtropical fruit trees, such as mango, mangosteen and Citrus, and tropical forage grasses such as Panicum, Brachiaria, Dichanthium and Pennisetum. There are also few apomictic species of significant relatedness available for use in introgression programmes, which may explain at least some of the difficulties experienced when attempts have been made to introduce apomixis into crops though hybridization. For example, major programmes aimed at introducing apomixis into maize (Zea mays) (Sokolov et al., 1998; Savidan, 2000a, 2001) from the wild relative Tripsacum

The Molecular Biology of Apomixis

357

dactyloides have been underway now for decades, yet unfortunately have proven unsuccessful in terms of generating apomictic plants with agronomically acceptable levels of seed set. Difficulties have also been encountered in efforts to produce apomictic lines of hybrid millet (Morgan et al., 1998; Savidan, 2001).

Biogeographical and ecological considerations Although apomixis appears to be widespread amongst angiosperms, it is not evenly distributed. Several authors have noted a marked bias in the distribution of apomixis (Asker and Jerling, 1992; Mogie, 1992; Carman, 1997; Richards, 1997). Of the plants known to utilize gametophytic apomixis, 75% belong to three families, the Asteraceae, Rosaceae and Poaceae, which collectively comprise only 10% of flowering plant species (Richards, 1997). Conversely, while apomixis is known among the Orchidaceae, the largest flowering plant family, it appears to be uncommon among these plants. Some authors have, therefore, postulated that the current patterns of distribution of apomicts may reflect the predisposition of certain plant groups to the unique developmental and genetic changes that characterize apomixis (Grimanelli et al., 2001a; Sharbel and Mitchell-Olds, 2001). Possible predisposing characteristics may include cytological features like the presence of a nutritive nucellus in Citrus (Koltunow et al., 1995b) and a nutritive integument in Hieracium (Koltunow et al., 1998) and Taraxacum (Cooper and Brink, 1949), or they may be genetic such as an apparent linkage grouping in Tripsacum necessary for the interspecific transfer of the trait (Savidan, 2001). Apomicts are often reported to predominate at higher latitudes (Stebbins, 1950, 1971; Richards, 1997). This appears to be largely due to a greater tolerance amongst apomicts to these environments as many of the apomicts known from Northern Europe also occur further south. Furthermore, while apomicts were found to make up a higher percentage of the Northern Scandinavian flora than in more southern sites, the absolute number of apomicts was greater in the warmer southern regions. This difference reflects the greater diversity of the southern flora, which includes many more sexual species (Asker and Jerling, 1992). There are several reports of apomicts existing over a much greater range than their sexual counterparts, such as for Rubrus in Europe (Gustafsson, 1943), Crepis in North America (Babcock and Stebbins, 1938) and Dichanthium in India and Africa (De Wet, 1968), often with the sexual form occupying a central region of the geographical range of the apomict. Similar observations have also been made for some parthenogenetic animal species (Suomalainen et al., 1987). Some examples, however, have also been reported of the opposite pattern, with an apomict displaying a restricted geographical distribution while the related sexual species is more widespread. In the genus Antennaria, the sexual species Antennaria dioica is widespread while the apomictic forms are restricted to some Scandinavian mountains (Porsild, 1965). There is evidence that apomicts often either originated in glacial refugia and/or were able to radiate from those regions more rapidly than related

358

R.A. Bicknell and A.S. Catanach

sexuals once deglaciation began (Asker and Jerling, 1992). It has been proposed that apomicts may act as better colonizers than their sexual counterparts as they are more able to successfully colonize remote sites with single individuals (Stebbins, 1950, 1971). In the case of autonomous apomicts (see below) in particular, founding individuals are unconstrained by the need for crosspollination. In addition, they are less likely to suffer from genetic bottlenecks following the founder event. Carman (1997, 2001) proposed that the noted association between the radiation of apomicts and glacial refuges results from the formation of agamic complexes within refuges during glacial periods. Apomicts in nature often appear to have a hybrid origin (Ernst, 1918; Gustafsson, 1947a,b; Nygren, 1948), a feature that may provide them with fitness advantages in some habitats. Carman (1997, 2001), however, proposed that hybridity itself is the cause of apomixis because it leads to the formation of unique developmental gene regulatory cascades. As implied above, apomicts are often among the early colonizers of disturbed habitats, a characteristic that may explain why many are regarded as aggressive agricultural and amenity weeds around the world. Such weedy species include members of the genera Taraxacum (the common dandelion), Hieracium (hawkweed), Rubus (bramble), Cortaderia (pampas grass), Chondrilla (skeleton weed), Opuntia (prickly pear) and Erigeron (fleabane). Apomicts are almost invariably perennials. They are also typically polyploid (discussed below), which is strongly correlated with perenniality across all flowering plants (Mu¨ntzing, 1936; Levin, 1983). Both woody and herbaceous apomicts are known. However, herbaceous types appear to be more common amongst the plants exhibiting gametophytic apomixis (see below), whilst sporophytic apomixis is common amongst trees of tropical origin. It is also common for apomicts to be capable of other forms of asexual increase, such as through stolon proliferation (Hieracium, Poa, Hypericum), bulb division (Allium, Zephyranthes) or rhizomes (Rubus). There is a clear association between the expression of apomixis and the operation of mechanisms that limit self-fertilization (autogamy). In many cases apomicts are closely related to sexual species that display either physiological or morphological mechanisms limiting autogamy. In Limonium, sexual forms exist in heteromorphic populations with a mixture of plants displaying papillateor cob-type floral stigma (Baker, 1966). Papillate plants produce pollen grains that can germinate only on a cob-type stigma, while the pollen of cob plants will germinate only on a papillate stigma. This mechanism drives outcrossing (allogamy), and also promotes the maintenance of similar frequencies of the two types in a sexual population. Conversely, most apomictic species of Limonium are known to occur only in a single form. This characteristic has been used as a diagnostic measure of apomixis in this group (Palacios et al., 2000). Similarly, many apomicts are closely related to sexual species that have male and female floral forms (dioecy) such as Antennaria (O’Connell and Eckert, 1999), Cortaderia (Philipson, 1978) and Coprosma (Heenan et al., 2002). In dioecious species the sex ratio is again driven to approximately 50% in completely sexual populations. The related apomicts, however, demonstrate sex ratios significantly biased towards the formation of female plants (Heenan et al., 2002). In

The Molecular Biology of Apomixis

359

cases such as those described above, the extent of bias towards a single mating type is governed not only by the prevalence of apomixis in the population but also by the level of facultative expression within individuals and by the need in many apomicts for the fertilization of polar nuclei to form an endosperm tissue (see below). Finally, even amongst hermaphroditic species, physiological mechanisms of self-incompatibility are often associated with apomixis. For example, such a mechanism is reported for sexual forms of Hieracium (Gadella, 1984, 1991; Krahulcova et al., 1999), and has recently been demonstrated to be active amongst the apomictic forms of this plant as well (Bicknell et al., 2003).

The Mechanisms of Apomixis Apomixis is defined at a functional level as the formation of genetically identical seeds without meiosis or fertilization. Although this outcome is the same across the known apomictic species, many different mechanisms for achieving it have been identified. This diversity is likely to have arisen partly from differences in the floral biology and the life histories of the species involved and partly it may reflect a polyphyletic origin for the trait. To provide a framework for describing apomixis at the cellular level, known mechanisms have been divided into groupings based on the tissue that gives rise to the asexually derived embryo and the origin of the supporting endosperm tissue (See Fig. 13.1, also Nogler, 1984a; Koltunow, 1993; Crane, 2001). Two main mechanisms are recognized: (i) ‘sporophytic apomixis’, in which the asexually derived embryo arises directly from a non-gametophytic cell of the ovule; and (ii) ‘gametophytic apomixis’, where the asexually derived embryo arises from a gametophytic cell. In gametophytic apomicts an embryo sac (megagametophyte) forms, which is unreduced. All of the cells within it, therefore, have the somatic chromosome complement. Gametophytic apomixis is then typically further subdivided on the basis of the cell, giving rise to this unreduced embryo sac. In ‘diplosporous’ types an unreduced embryo sac forms from the differentiation of a megaspore mother cell (MMC), following either the complete avoidance of meiosis (Antennaria type) or a dysfunctional form of meiosis that leads to a restitution product (Taraxacum and Ixeris types). In ‘aposporous’ types, the MMC or its derivatives degrade and a cell nearby initiates the development of an embryo sac. The further subdivision of the mechanisms of gametophytic apomixis is based on the steps of meiosis that are observed, the cells that contribute to the embryo sac and the final form of that organ (Crane, 2001). For most flowering plants the formation of an endosperm tissue is necessary to ensure the complete development of the embryo. Apomicts are no exception to this requirement. The formation of a functional endosperm, however, involves different constraints to the formation of a clonal embryo. In angiosperms, ordinarily the embryo is diploid while the endosperm tissue is triploid. This ‘2:3’ ratio is known to be critical for the survival of the developing seed in many species. In maize a 2:3 ratio is an absolute requirement (Lin, 1984) for the survival of the developing embryo, while in Arabidopsis it

360

R.A. Bicknell and A.S. Catanach

Sexual reproduction n

n 2n

Sporophytic apomixis n

2n n

2n

Diplospory 2n

2n

2n

Apospory

2n

2n n

2n n

Fig. 13.1. Sexual reproduction proceeds through meiosis, megaspore selection and gametogenesis to form a reduced embryo sac. Double fertilization leads to the formation of an n þ n zygote and a 2n þ n endosperm. In sporophytic apomixis the sexual process remains intact. Typically, near the end of gametogenesis one or more embryos (or ‘embryoids’) form from somatic cells of the ovule. Their further development is reliant on the formation of a 2n þ n endosperm arising from the meiotically derived embryo sac. Sporophytic apomicts are, therefore, always pseudogamous. In diplosporous apomicts, meiosis is either avoided entirely or is restitutional. A single unreduced cell remains, which then undergoes gametogenesis to form an unreduced embryo sac. The egg within that structure spontaneously divides (parthenogenesis), initiating the pathway of embryogenesis. The endosperm may either form following fertilization of the polar nuclei (pseudogamy) or it may arise spontaneously (autonomous apomixis). Apospory differs from diplospory in the origin of the initial cell, giving rise to the unreduced embryo sac. In aposporic apomicts a separate cell of the nucellus, typically adjacent to the meiotic apparatus, undergoes gametogenesis. This can lead to the formation of both a reduced and an unreduced embryo sac within the same ovule as pictured. Most frequently, however, one structure predominates and the other degenerates. Aposporic apomicts may also form an endosperm by either a pseudogamous or an autonomous process. The nuclear state of the principal structures is marked.

strongly favours development but is not an absolute necessity for the survival of the embryo (Scott et al., 1998). The reason for this requirement appears to relate to the role of differential genomic imprinting on the maternal and paternal contributions to the embryo and endosperm tissues (discussed below). In apomicts the embryo-to-endosperm ploidy ratio can differ depending on the mechanism of apomixis employed. In the case of sporophytic apomicts, asexually derived embryo(s) form within the somatic tissues of the

The Molecular Biology of Apomixis

361

ovule, typically alongside a sexually derived structure. An endosperm then forms as part of the sexual process and it is this structure that nourishes the somatic embryos. As the endosperm is derived from the fertilization of two reduced polar nuclei, this results in the typical 2:3 ploidy ratio and therefore normal development. In gametophytic apomicts several different mechanisms of endosperm formation are known. In all cases the asexually derived embryo sac is unreduced so that the egg and polar nuclei it contains all have the somatic chromosome complement. In some species differences in the development of the embryo sac allow the 2:3 embryo/endosperm ploidy ratio to be achieved. Panicum is a grass genus, the apomicts of which form a unique embryo sac with only four cells, one of which acts as an egg while the central cell has only one polar nucleus. The fertilization of that unreduced polar nucleus (and the avoidance of egg cell fertilization) leads to the typical 2:3 embryo/endosperm ratio. In Dichanthium annulatum the two sperm nuclei fuse with different unfused polar nuclei, again establishing a 3n endosperm about the 2n parthenogenetic embryo (Reddy and D’Cruz, 1969). Finally, in many apomictic plants the typical 2:3 embryo/endosperm ploidy balance requirement is relaxed (see Matzk et al., 2000). In the apomict Poa pratensis two unreduced polar nuclei fuse with a sperm nucleus resulting in an embryo/endosperm ploidy ratio of 5:2 (see Matzk et al., 2001), while in Taraxacum two unreduced polar nuclei fuse but are not fertilized. The endosperm that develops in Taraxacum is therefore 4n and the embryo/endosperm ratio is 2:4. As fertilization is neither required for the formation of an embryo nor an endosperm, Taraxacum is considered to be an example of an ‘autonomous apomict’. Plants such as Panicum maximum and Citrus, which do require fertilization to form an endosperm yet produce an embryo asexually, are referred to as ‘pseudogamous apomicts’.

Apomixis is a facultative process As indicated above, apomixis almost always exists in parallel with sexual reproduction. The first discussion of the ‘facultative’ nature of apomixis appears to have been made by Rosenberg (1906, 1907), who noted the coexistence of meiotic and ameiotic structures in the ovules of Hieracium. Together with Ostenfeld (1904, 1906, 1910) they correctly surmised that this resulted in the formation of both sexually and asexually derived progeny. Intriguingly, the material studied by Rosenberg and Ostenfeld was similar to that already studied extensively by Mendel (1869). Although Mendel did not attribute the segregation patterns he observed in these plants to asexual reproduction, he did note that, in contrast to his observations in pea, the Hieracium F1 hybrids showed extensive segregation while the F2 ‘hybrids’ did not segregate and uniform progeny were consistently obtained. In correspondence with Na¨geli, a Hieracium specialist (July, 1870), he noted the ‘almost opposed behaviour’ in the two systems ‘both (of which represented) the emanation of a higher universal law’, an indication that he was aware that this material had unique attributes (reviewed by Correns, 1905; Nogler, 1994).

362

R.A. Bicknell and A.S. Catanach

In sporophytic apomicts it is essential that the sexual pathway of seed development remains intact as this is the sole source of endosperm tissue. In most cases it appears that only two types of progeny form in these plants: (i) those arising from the fertilization of a reduced egg (sexually derived progeny); and (ii) those arising from a somatic embryo (asexually derived progeny). In Citrus the relative levels of these two types vary from species that are regarded as essentially obligate sexuals to plants such as grapefruit that reproduce primarily by apomixis. There are, however, some indications that unreduced embryo sacs occasionally form in this material and that this leads to the production of progeny with different ploidy levels through the mediation of unreduced gametes (Naumova et al., 1992). In gametophytic apomicts several different types of progeny are commonly reported. In these plants egg cells may form either from a reduced megaspore following meiosis or they may form within an unreduced embryo sac following either meiotic restitution (diplospory) or somatic cell gametogenesis (apospory). Gametophytic apomicts are also capable of developing an embryo directly from an egg cell (parthenogenesis). In most plants, however, it is also possible for fertilization to occur if a compatible sperm cell is available at the right time to effect syngamy. There are, therefore, two possible ploidy states for the egg cells of these plants and two states for the role of fertilization. In combination, four possible progeny types may result: (i) the fertilization of a reduced egg (sexual reproduction or ‘n þ n hybridization’); (ii) the fertilization of an unreduced egg (2n þ n hybridization); (iii) the parthenogenetic development of a reduced egg (‘polyhaploidy’ or ‘n þ 0 progeny’); or (iv) the parthenogenetic development of an unreduced egg (apomixis or ‘2n þ 0 progeny’).

The detection and measurement of apomixis In most cases apomixis has been detected through the formation of ‘maternal’ progeny, which appears to be morphologically identical to the mother parent. This is a notoriously inaccurate approach for assessing genetic variation and, therefore, for inferring the action of a particular breeding system. Several mechanisms can give rise to apparently identical progeny, most commonly though self-fertilization. Conversely, the form of some plants, such as the apomicts of Taraxacum officinale (De Kovel and De Jong, 1999), are sufficiently plastic in different environments to tempt the conclusion of genetic variation, despite the clonal nature of these populations. The extent of facultativeness in the expression of apomixis also has an important bearing on the interpretation of inheritance studies and on extrapolations regarding the ecological and evolutionary impact of the trait. The most cited classical example of this is the statement by Darlington (1939) expressing that ‘with the loss of sexual recombination the apomict, like the permanent hybrid, is cut off from ultimate survival. Apomixis is an escape from sterility, but it is an escape into a blind alley of evolution’. Similar comments were later expressed by Stebbins (1950). In these cases apomixis was being treated as a qualitative trait that was mutually exclusive of sexuality. This is now known to

The Molecular Biology of Apomixis

363

seldom occur. Ironically, the near universal observation that apomixis is expressed facultatively has led to a very different prediction. Several authors have recently suggested that even a low level of asexual reproduction can lead to the loss of sexual reproduction in a population, as apomixis leads to a greater investment in the replication of genotypes with predetermined fitness (Burt, 2000; Peck and Waxman, 2000). Given these concerns there is a clear need to base any quantitative study of apomixis on a reliable and sensitive method of detecting the trait. Several methods have been developed, often tailored to a specific genus or mechanism of apomixis. As mentioned above, morphological characters have been widely used and remain useful in some systems. In dimorphic systems such as Antennaria (O’Connell and Eckert, 1999), Limonium (Baker, 1966), Coprosma (Heenan et al., 2002) and Cortaderia (Philipson, 1978), apomixis is readily detected by the predominance of a single form in a population when the normal ratio of forms is expected to be closer to 1:1. In other plants cellular characters are used as a diagnostic for apomixis. In Panicum and Pennisetum, apomixis proceeds through the mediation of a four-celled embryo sac. Sexuals in these plants form an eight-celled embryo sac so the prevalence of apomixis is measured through the quantification of these different cyto-types (Leblanc and Mazzucato, 2001). Similarly, in Hieracium and Taraxacum an asexually derived embryo will form without fertilization, often prior to the opening of the flower. In these plants apomixis can be scored through the detection of that structure after the avoidance of fertilization. In diplosporous grasses, such as Elymus rectisetus and T. dactyloides, a callose cap is seen above the MMC and the selected megaspore following meiosis, but not in association with asexually derived structures. This observation forms the basis for a test used to score the incidence of apomixis in many of these systems (Carman et al., 1991; Leblanc and Mazzucato, 2001). It is important to realize that in all of these cases apomixis is being measured through a correlative character rather than directly. In the case of Panicum the frequency of four-celled embryo sacs is a strong measure of apospory, but does not guarantee that the unreduced egg will proceed to form an embryo by parthenogenesis. Similarly, the formation of embryos in Hieracium is a good indicator of parthenogenesis but does not provide a measure of apospory. Flow cytometry has been used to screen populations for signs of apomixis, the most notable example of which is the work of Savidan and his colleagues who screened tens of thousands of seedlings during a crossing programme aimed at introgressing apomixis into maize (Savidan, 2000b, 2001). Recently Matzk et al. (2000) reported a sensitive, yet flexible method for identifying different classes of apomixis using flow cytometry to detect differences in the embryo/endosperm ploidy ratio of seeds. A recent refinement to the technique permits its use on single-seed samples (Matzk et al., 2001). Bicknell et al. (2003) also reported a method based on the inheritance of selectable marker loci that permits the rapid estimation of progeny-type frequencies, and the recovery of rare classes from very large seedling populations at low cost. Genomic mapping strategies, such as restriction fragment length polymorphism (RFLP), AFLP and SSRs, are also being applied to the study of apomixis

364

R.A. Bicknell and A.S. Catanach

(for a review see Grimanelli et al., 2001b). These techniques provide an insight into the genetic variance of a population, a measure of inheritance from the parent(s) and a tool for determining the chromosomal location of genetic determinants controlling the trait. The unit cost of genetic testing is steadily reducing, and the techniques are proving to be highly reliable in skilled hands. It seems likely that some form of genetic testing will be a routine part of most studies into apomixis in the near future.

The Genetic Control of Apomixis Research into the genetic control of apomixis is progressing along two complementary lines of enquiry. In the first instance several groups are studying factors that influence apomixis in a range of species that naturally express the trait. A second line of enquiry is evaluating the control of sexual reproduction in standard sexual models and using this information to ‘synthesize’ apomixis in these plants. The two approaches are proving to be mutually supportive and both are likely to contribute to our understanding for some time to come. For simplicity, the two will be treated separately in this publication.

The genetic control of apomixis in native apomicts Significant progress has recently been reported in the study of the genetic control of apomixis in a number of species. For gametophytic apomicts most reports indicate that apomixis is relatively simply acquired through the inheritance of either one or a small number of factors, most commonly as dominant determinants. Amongst the aposporic grasses, a single dominant factor appears to be responsible for the inheritance of apomixis in Panicum (Savidan, 1981), Pennisetum (Sherwood et al., 1994) and Brachiaria (Valle et al., 1994). A single dominant factor is also reported to be associated with the inheritance of apospory in the dicotyledonous genus Ranunculus (Nogler, 1984b). In the diplosporous daisy genera Taraxacum (Van Dijk et al., 1999) and Erigeron (Noyes, 2000; Noyes and Rieseberg, 2000), the component processes of diplospory and parthenogenesis are separately inherited. We have recently discovered a similar pattern of inheritance in the aposporous daisy genus Hieracium (Bicknell, unpublished observation). Simple dominant inheritance has also been reported for the diplosporous grasses Eragrostis curvula (Voigt and Burson, 1983) and T. dactyloides (Leblanc et al., 1995). There is evidence of segregation ratio distortion in some of these systems, often because the dominant factor(s) associated with apomixis also appears to confer gamete lethality, restricting its transfer to some gamete genotypes (Nogler, 1984b; Grimanelli et al., 1998a; Roche et al., 2001a; Jessup et al., 2002). Much less information is available for sporophytic apomixis. Garcia et al. (1999) reported the detection of six quantitative trait loci (QTLs) contributing to the total variation associated with the expression of apomixis in a cross between Poncirus trifoliata and Citrus volkameriana.

The Molecular Biology of Apomixis

365

Even when the pattern of inheritance for apomixis appears to be simple, care must be exercised when interpreting these findings. The choice of parental material is particularly critical. Apomixis reduces the frequency of meiosis in the ovule and also the availability of the egg cell for fertilization. Segregation during female meiosis is, therefore, seldom measured in inheritance studies of apomicts. Fortunately, many apomicts produce adequate quantities of reduced pollen and this has been demonstrated in many systems to transmit factors associated with the trait. Most commonly, therefore, tests for the inheritance of apomixis have used a sexual plant as the maternal parent and an apomict as the pollen donor. The need to use a sexual plant of sufficient relatedness to an apomict means great care must be taken to ensure that that plant does not have any latent tendencies towards apomixis. Very often they do (see discussion by Savidan, 2000b). Apomixis is a complex trait involving several developmental processes. Progeny should, therefore, be assessed for all of the components of the mechanism of apomixis under study. Correctly, this would involve an embryological examination, together with tests of ploidy and a genomic profiling test (such as with AFLP) for each progeny plant (see Nogler, 1984a). Unfortunately, very often progeny in inheritance studies have been assessed either for apomixis as a whole or have been scored against only one component of apomixis (for a review see Bicknell, 2001; Leblanc and Mazzucato, 2001).

Why are gametophytic apomicts almost always polyploid? Gametophytic apomicts are almost invariably polyploid and most commonly tetraploid (Asker and Jerling, 1992). The reason for this association remains unclear. It is, however, potentially a critical issue if apomixis is ever to be introduced into diploid crop species. Several theories have been forwarded to explain this association. These subdivide on the basis of cause and effect into arguments proposing that polyploidy stimulates the development, penetrance and maintenance of apomixis, and arguments proposing that apomixis stimulates the formation and proliferation of polyploids. Both positions may be true. Although diploid apomicts are very rare in natural populations, they have been reported following the experimental manipulation of several polyploid apomicts (Bicknell, 1997; Kojima and Nagato, 1997; Naumova et al., 1999; Koltunow, 2000). These findings indicate that polyploidy is not an absolute requirement for the expression of apomixis. These results also proposed that on almost every reported occasion the identified diploid apomict was weak and most were also pollen sterile. It would appear, therefore, that while polyploidy is probably not an absolute necessity for the expression of apomixis in most systems, it may act to enhance the expression or promote the transmission of the trait. Some indications of how this might operate come from yeast (Galitski et al., 1999) and from Arabidopsis (Lee and Chen, 2001), where alterations in ploidy status are known to affect methylation and the expression of different alleles. There have also been reports of apomixis arising following the chromosome doubling of a sexual diploid plant. Nygren (1948) reported this finding in

366

R.A. Bicknell and A.S. Catanach

Calamagrostis purpurea and Quarin et al. (2001) noted it in Paspalum notatum. In both cases, however, the diploid sexual plant was a close relative or a diploid form of a known apomict. It may, therefore, have been predisposed to the expression of the trait. It also needs to be noted that a sexual plant has been recovered following the chromosome doubling of an apomict (Asker, 1967). Finally, polyploidy has been induced in a large number of plants without apomixis being noted in the products. Many apomicts appear to have had a hybrid origin, and this in combination with polyploidy is believed to have resulted in allopolyploidy in many gametophytic apomicts (Ellerstrom and Zagorcheva, 1977; Carman, 1997, 2001; Roche et al., 2001b). There is also evidence, however, for tetrasomic inheritance in many systems, indicating the presence of autopolyploidy, or possibly segmental allopolyploidy, in some apomicts (Pessino et al., 1999). Carman (1997, 2001) postulated that a combination of hybridity and polyploidy can lead to the disjunction of key regulatory events during critical stages of megasporogenesis, megagametogenesis and fertilization. This in turn may lead not only to apomixis but also to other unusual developmental events such as polyspory and polyembryony. Roche et al. (2001b) further suggested that supernumerary chromatin may be the principal component in this process. A hybrid origin, segmental allopolyploidy and the activity of reproductive drivers are all reported characteristics of supernumerary chromatin biology (McVean, 1995). There is growing evidence for the presence of supernumerary chromatin in several apomictic species and it is clearly involved in the inheritance of apomixis in the grasses Pennisetum squamulatum and Cenchrus ciliaris (Roche et al., 2001a,b and references therein). A possible hybrid origin may also explain the apparent conundrum that apomixis seems to result from the summation of several components, each of which on its own is expected to confer a selective disadvantage to the plant (Mogie, 1992). There is also evidence that apomixis promotes polyploidy. While apomixis can function in a diploid plant, the activity of apomixis may promote the formation of polyploids by reducing the fitness of diploid apomicts or of haploid gametes transferring the trait. Apomicts reproduce asexually, avoiding the activity of the meiotic sieve (Richards, 1997). As such, they can be expected to suffer from progressive genetic degeneration through the accumulation of random mutations, the proliferation of transposable elements and the activities of viruses (Muller, 1964; Matzk et al., 2003). Polyploidy increases the number of copies of each gene and effectively increases the tolerance of the genome to the impacts of this degenerative process. As noted above, most gametophytic apomicts form polyhaploid (n þ 0) progeny. This mechanism has been demonstrated to lead to the formation of diploid apomicts from tetraploid apomictic parents in several species (Nogler, 1982; Bicknell, 1997; Kojima and Nagato, 1997). There is, therefore, a mechanism in most gametophytic apomicts to form diploids, yet they do not typically survive under competition. The reason for this, as mentioned above, is that these plants are invariably much weaker than their polyploid, maternally derived cohorts, an apparent consequence of a high genetic load inherited from their polyploid parent. Conversely, most gametophytic apomicts are also capable of forming a small number of

The Molecular Biology of Apomixis

367

unreduced hybrids (2n þ n) through the fertilization of an unreduced egg. In this case the parent genome remains intact but is complemented with the addition of new genomic material. We commonly observe significant hybrid vigour in these plants, often making them more vigorous than their maternal cohorts (Chapman and Bicknell, 2000). It, therefore, appears that apomixis leads to the operation of mechanisms that generate progeny of greater or lesser ploidy to the parent. However, selection favours the latter. Nogler (1984b, 1986) noted that apomixis in the buttercup species Ranunculus auricomus could be inherited as a simple dominant trait; however, the locus for apomixis could only be transferred using a polyploid gamete. Haploid gametes were formed and proved functional, but the progeny were always sexual. Nogler concluded that the determinant for apomixis in this system was gamete lethal in the homozygous form. Similar results have been reported for Tripsacum (Grimanelli et al., 1998a) and Pennisetum (Roche et al., 2001a; Jessup et al., 2002). It is therefore argued that the restricted transfer of apomixis to diploid or polyploid gametes will lead to the exclusive formation of polyploid, apomictic progeny (Nogler, 1986). Furthermore, it is not necessary for the actual alleles for apomixis to be gamete lethal to generate this effect. Apomixis results in the asexual formation of progeny, which is expected to lead to an accumulation in genetic load over successive asexual cycles. Alleles conferring such a trait are, therefore, expected to become associated with regions of high genetic load by linkage drag. This effect is compensated for in the apomictic individual through the co-inheritance of functional alleles, but during the formation and function of haploid gametes linked recessive lethals associated with this region will be exposed and will reduce gamete fitness (Bicknell et al., 2000; Matzk et al., 2003). One wellstudied example of this is a region associated with the inheritance of apomixis in Pennisetum. The flanks of this ‘apospory-specific genomic region’ (ASGR) are characterized by repeats of an Opie-2-like retrotransposon. Interestingly, this transposon sequence is common throughout the genome of the related species C. ciliaris, yet restricted to the ASGR of Pennisetum (Akiyama et al., 2004).

The identification of ‘apomixis genes’ With the finding that gametophytic apomixis was often simply inherited, several groups are attempting to identify the sequences involved in a number of plant species. As mentioned above, Ozias-Akins and colleagues (Ozias-Akins et al., 1993, 1998; Roche et al., 2001a, 2002) noted in Pennisetum that the inheritance of apospory was associated with the transfer of a genomic region located near the telomere of a P. squamulatum chromosome (Goel et al., 2003). This ASGR was found only in the apomict, no corresponding allele was found in related sexual species. Furthermore, the ASGR was also found in the related apomict C. ciliaris, again as a hemizygous region restricted to apomictic biotypes (Roche et al., 1999, 2001b). Together with the evidence of transposon representation mentioned above, the authors note that the data

368

R.A. Bicknell and A.S. Catanach

suggest that speciation of apomictic Pennisetum was preceded by a wide crossmediated introgression event of the ASGR-associated chromatin from another species. A map-based cloning strategy is underway to identify the genes located in the ASGR. Twelve molecular markers were found to always segregate with the ASGR, of which eight were converted to SCARS and used to isolate corresponding bacterial artificial chromosome (BAC) clones (Roche et al., 2002). Of these BAC clones, none have more than one marker and no two of the BAC clones overlap, a result that indicates that ASGR is a region in excess of 50 Mbp. It would appear, therefore, that the inheritance of apomixis in this system is associated with the transfer of a large, non-recombinant segment of a chromosome. In the diplosporous grass T. dactyloides, a similar effort is underway. Again markers were reported in linkage with the region associated with the inheritance of diplospory in this plant (Grimanelli et al., 1998a; Blakey et al., 2001). Suppressed recombination is clearly also a feature of this region, frustrating efforts to clone the critical genes involved in apomixis. Estimates of the size of this region are less accurate than with Pennisetum but it appears to be of a similar magnitude (D. Grimanelli, personal communication). Paspalum, another grass genus, is also being studied. Apospory in this plant is inherited as a simple dominant factor (Martinez et al., 2001). Grasses are known for their high synteny. Pupilli et al. (2001) noted the co-inheritance of five rice markers with apospory in Paspalum and are attempting to clone the locus using these markers as probes. Again, non-recombination and hemizygosity were detected in association with the locus (Labombarda et al., 2002). Interestingly, although all of the grasses mentioned above show considerable synteny with the rice genome, in each case the region associated with the inheritance of apomixis aligns with a different region of that genome (Grimanelli et al., 2001a). Albertini et al. (2001a,b) also reported the isolation of markers linked to apospory in Kentucky bluegrass (Poa pratensis) and Pessino et al. (1997, 1998) noted similar findings for Brachiaria. Amongst the eudicotyledonous systems under study, progress is being made towards the mapping of ‘apomixis’ genes in the daisies Taraxacum (Van Dijk et al., 2003; Vijverberg et al., 2004) and Erigeron (Noyes and Rieseberg, 2000). There is some evidence for non-recombinant sector(s) associated with apomixis genes in Erigeron, and gamete selection in Taraxacum, indicating that similar difficulties with cloning may also arise in these plants. The nature of these large regions of suppressed recombination is poorly understood. They may include several genes, collectively required for the functioning of apomixis, as seen in the incompatibility locus of Brassica (Fobis-Loisy et al., 2004) and/or they may contain large regions of repetitive DNA as seen in the ASGR of Pennisetum (Akiyama et al., 2004). Retrotransposon accumulation in apomicts, with no counterbalancing of retrotransposon elimination via recombination during meiosis, is believed to be the cause of a higher DNA content of evolutionarily older apomictic species of Hypericum compared to their younger counterparts (Matzk et al., 2003). While conventional linkage mapping is very difficult when the genetic determinants for apomixis are associated with a large non-recombinant region, it is hoped that the common existence of the ASGR in several Pennisetum species

The Molecular Biology of Apomixis

369

will allow genes associated with apomixis to be identified by comparing different ASGRs for conserved sequences (Roche et al., 2002). Another related approach to identifying genes associated with apomixis is based on mutagenesis, coupled to a method for identifying the mutated genes. This approach has the advantage that it is not based on recombination and therefore should not be influenced by the issues mentioned above. T. dactyloides is a close relative of maize. The two can be crossed and apomictic hybrids selected from the progeny (Savidan, 1992, 2000a, 2001). Grimanelli et al. (2001b) are using this feature to introduce the mutator transposon of maize into this apomictic hybrid background. It is hoped that the transposon will then disrupt genes associated with apomixis through insertional inactivation and provide a molecular ‘tag’ to assist the isolation of the sequences involved. A similar ‘transposon tagging’ system was tried in the aposporous daisy Hieracium, using an engineered heterologous transposon tagging system based on the Ac/Ds elements of maize (Bicknell et al., 2001; Weld et al., 2002). In this case the transposons were found to be active in the apomict but their level of activity was too low to be of any practical value. Deletion mutagenesis is also being tried in Hieracium (Bicknell et al., 2001). From preliminary work it is clear that apomixis can be disrupted by genetic deletion in these plants. Efforts are now underway to use deletion mutants to identify region(s) associated with apomixis in these plants (R.A. Bicknell, unpublished results).

The ‘Synthesis’ of Apomixis in Sexual Systems As described above, apomixis can be considered as the sum of a number of component parts, specifically the avoidance of meiosis (apomeiosis), the avoidance of fertilization (parthenogenesis), the spontaneous formation of the endosperm (in some cases) and the relaxation of the 2:3 embryo/endosperm ploidy balance (in some cases). The discovery that these processes are under relatively simple genetic control in many apomictic species, and the identification of mutants in sexual species with functions similar to these components, led to the conclusion that apomixis represents an altered form of sexuality rather than a unique developmental programme (reviewed by Koltunow and Grossniklaus, 2003). Recent work by Tucker et al. (2003) demonstrating that similar gene expression profiles are involved in sexual and apomictic development in Hieracium further supports this conclusion. This is not a new idea. Nogler (1984a) expressed his belief that apomixis represented an ‘opening of the bonds’ linking megasporogenesis and embryo sac formation. Similarly, the views of Ernst (1918) and later Carman (1997, 2001), that apomixis results from the hybridization of sexual species due to the creation of unique combinations of factors otherwise involved in sexual reproduction, reflect this position. If apomixis is an altered form of sexuality, as appears to be the case, then it seems reasonable to propose that it could be ‘synthesized’ by compiling mutations and/or redirecting gene expression in a target sexual species. This approach is discussed below with respect to the component processes.

370

R.A. Bicknell and A.S. Catanach

The avoidance of meiosis In the case of sporophytic apomixis, both the avoidance of meiosis and spontaneous embryogenesis are achieved through the direct formation of an embryo from the somatic tissue of the parent plant (see discussion below on parthenogenesis). In gametophytic apomicts, a somatic cell of the ovule, normally a cell of the nucellus, enters gametogenesis without completing the events of meiosis, resulting in the formation of an unreduced embryo sac. Unreduced eggs are known to occur in many sexual flowering plant species (reviewed by Harlan and De Wet, 1975; Veilleux, 1985). However, their prevalence is much more pronounced in gametophytic apomicts. In maize the prevalence of unreduced egg formation amongst inbred lines varied from 0% to 3.48% (Alexander and Beckett, 1963). Similarly, amongst diploid species of Medicago the rate of unreduced egg usage was estimated at 0–9% (Veronesi et al., 1988). In contrast, the rates of unreduced egg cell usage in the apomicts Hieracium aurantiacum and Hieracium piloselloides were estimated at 97.4% and 97.1%, respectively (Bicknell et al., 2003). Amongst sexual species the formation of unreduced egg cells typically occurs following a dysfunction of meiosis, either at the first meiotic division (first division restitution, FDR) or second meiotic division (second division restitution, SDR). FDR is similar to the events seen in some diplosporous apomicts such as Taraxacum and Ixeris (Crane, 2001). If crossover is suppressed during the first meiotic prophase the resulting cell is genetically equivalent to the maternal soma. Interestingly, there is evidence that crossover does occur in some apomictic species of Taraxacum, leading to autosegregation amongst the parthenogen¨ rensen and Gudjonsson, 1946; Malecka, 1973). etically derived progeny (So Second division restitution results in genetic differences between the egg and maternal tissues due to chromatid segregation during the first meiotic division. It is not known amongst native apomicts and is not generally regarded as being useful in the possible synthesis of apomixis. There is clear evidence that the frequency of unreduced egg cell formation in sexual species is genetically controlled with both major and minor gene effects implicated in different species (reviewed by Veilleux, 1985). The rate of reduced gamete formation can also be influenced through selection, indicating that it is a heritable trait (Calderini and Mariani, 1997; Lamote et al., 2002). One of the best studied examples is the elongate gene of maize, the mutation of which leads to a single meiotic division that is neither completely typical of FDR nor SDR (Rhoades and Dempsey, 1966; Nel, 1975). Other maize mutants are known to influence the entrance of the MMC into meiosis and the completion of the first meiotic division. In ameiotic 1 (amc1) mutants meiosis is replaced by developmental arrest or by one-to-many mitotic divisions (Palmer, 1971; Golubovskaya et al., 1992, 1993, 1997). In the absence of first division (afd) mutants the first division of meiosis is replaced by a mitotic division (Golubovskaya, 1979), an outcome similar to the mechanism of diplospory seen in Taraxacum and Ixeris (Grossniklaus, 2001). In the Arabidopsis dyad mutant, a dyad of unreduced megaspores forms rather than a tetrad of reduced cells (Siddiqi et al., 2000). Unfortunately this

The Molecular Biology of Apomixis

371

mutant is also female sterile. The mutation of AtDMC1 also influences meiosis in Arabidopsis by disrupting the formation and/or stabilization of bivalents, and reducing the possibility of crossover (Klimyuk and Jones, 1997; Doutriaux et al., 1998; Couteau et al., 1999).

Autonomous embryo formation (parthenogenesis) As outlined above, parthenogenesis is an essential component of all natural apomictic systems, both sporophytic and gametophytic, and it will be equally essential in any synthetic approach. Egg cell activation is well studied in animal systems (reviewed by Nuccitelli, 1991; Whitacker and Swann, 1993; Jaffe, 1996), providing the basis of many recent discoveries in plants. Advances in the isolation and use of functional flowering plant gametes have also been an important factor in improving our understanding of this fundamental process in plants (Faure et al., 1992; Dumas and Faure, 1995; Kranz and Dresselhaus, 1996; Rougier et al., 1996). In animals the adhesion of the sperm triggers a transient rise in free calcium ions in the egg and this initiates a cascade of downstream events (Whitacker and Swann, 1993). In plants a similar mechanism of calcium accumulation in the egg cytosol has recently been observed (Digonnet et al., 1997; Antoine et al., 2000), but the components of the signal transduction cascade have yet to be determined. Parthenogenesis has been widely recorded in flowering plants, occurring in most species at a low level. In the sexual plant Datura the rate of parthenogenesis was estimated at 0.018% (Kimber and Riley, 1963) while in maize rates are reported to vary from 0.11% to 3.2%, depending on genotype (Sarkar and Coe, 1966; Chase, 1969). By comparison, the rates of parthenogenesis in the native apomicts H. aurantiacum and H. piloselloides were recorded at 97.6% and 98.0%, respectively (Bicknell et al., 2003). Several mutations are known to induce a parthenogenetic phenotype or to induce developmental phenomena with similarities to parthenogenesis in other tissues of the plant. One of the most intriguing examples is the ‘Salmon’ system of wheat, which can produce haploids by parthenogenesis at rates of up to 90% (Matzk et al., 1995). This material carries a 1BL-1RS wheat/rye translocation and utilizes the cytoplasm of Aegilops caudata or Aegilops kotschyi. Neither the translocation nor the cytoplasm is sufficient to induce the effect alone, indicating that both nuclear and cytoplasmic factors are involved (Matzk et al., 1995, 1997; Matzk, 1996). The nuclear factors involved include Ptg, a promoter of parthenogenesis, and Spg, a suppressor of parthenogenesis. Parthenogenesis is expressed when Ptg is present with the Aegilops cytoplasm and Spg is absent (Matzk, 1996). In barley the haploid initiator (hap) mutation leads to the formation of up to 30% haploid embryos (Hagberg and Hagberg, 1980). Intriguingly, there appears to be a mechanism in this mutant, which suppresses syngamy between the egg and sperm yet permits the polar nuclei to be fertilized (Morgensen, 1988). In Arabidopsis, several mutants have been described that influence the formation and development of embryos and/or that result in the ectopic formation of embryo-like structures. The LEAFY COTYLEDON (LEC) genes

372

R.A. Bicknell and A.S. Catanach

LEC1, LEC2 and FUSCA3 are regulators of embryogenesis, required for the specification of suspensor cell fate, cotyledon identity and for other critical functions later in embryo development (Meinke et al., 1994; Lotan et al., 1998; Stone et al., 2001; West et al., 2001). The ectopic expression of LEC1 and LEC2 in vegetative cells induces the expression of embryo-specific genes and the development of embryo-like structures (Lotan et al., 1998; Stone et al., 2001). PICKLE (PKL) is a chromatin-remodelling factor that acts as a master regulator of the LEC genes LEC1, LEC2 and FUSCA3 (Ogas et al., 1999; Rider et al., 2003). Roots in pkl mutants form embryo-like structures indicating that this gene plays a role in repressing embryogenic development throughout the life cycle of the plant (Henderson et al., 2004). Polycombgroup proteins form multimeric complexes that modify histone 3 and participate in the stable repression of specific genes through chromatin remodelling. The Fertilization Independent Seed (FIS) genes of Arabidopsis are members of this group that are known to play a role in the formation of the endosperm and in establishing early parent-of-origin effects (see below). Recently, another member of the group, MEDICIS, was reported. A loss-of-function mutation in this gene led to similar defects in endosperm development as other reported members of the FIS gene group but also led to the parthenogenetic development of a two-celled embryo from an unfertilized egg cell (Guitton et al., 2004). Further embryo growth was arrested in the mutant. BABY BOOM (BBM ) is an AP2-domain transcription factor identified in a screen for differentially expressed messages during somatic embryogenesis in microspore cultures of Brassica napus (Boutilier et al., 2002). BBM transcript was observed to occur preferentially in developing embryos and seeds of B. napus, and the ectopic expression of BBM induced the development of embryo-like structures on the cotyledons and leaves of Arabidopsis (Boutilier et al., 2002). SOMATIC EMBRYOGENESIS RECEPTOR KINASE (SERK) was similarly found as an upregulated transcript in carrot (Daucus carota) cultures induced to form somatic embryos (Schmidt et al., 1997). The D. carota SERK transcript (DcSERK) was found to be associated with cells destined to undergo embryogenesis, and also to be present at the time of the initiation of zygotic embryogenesis (Schmidt et al., 1997). The Arabidopsis orthologue AtSERK1 was found to be expressed during gametogenesis and early embryogenesis in Arabidopsis, and the ectopic expression of AtSERK1 was found to improve the rate of somatic embryogenesis in Arabidopsis cell cultures (Hecht et al., 2001). AtSERK, however, did not induce a discernible phenotype in Arabidopsis when ectopically expressed (Hecht et al., 2001).

Endosperm formation As mentioned above, the formation of an endosperm is critical to seed development in almost every angiosperm species. Only members of the Orchidaceae and Podostemonaceae are known to either fail to form an endosperm or form an endosperm that degenerates early in the ontogeny of the seed (Foster and Gifford, 1974). Furthermore, all known apomicts have been recorded to

The Molecular Biology of Apomixis

373

form an endosperm, with the exception of Nigritella, an orchid (Teppner and Klein, 1993), indicating that it will be an essential tissue for the successful installation of apomixis into target crop species. As discussed above, the endosperm of an apomict may either be derived autonomously, following the spontaneous fusion of polar nuclei, or may result from the fertilization of the central cell, as in sexual plants (pseudogamy). Pseudogamous systems are far more common than autonomous systems, possibly because they enable the establishment of a 2:3 embryo/endosperm ploidy balance in the developing seed. This appears to be particularly critical in the grasses, many of which demonstrate a profound intolerance to alterations to this 2:3 ploidy requirement (Nishiyama and Yabuno, 1978; Johnston et al., 1980; Lin, 1984; Haig and Westoby, 1991; Birchler, 1993). Grasses are critical to agriculture as they include all the major cereal crops for human and animal consumption as well as forage species for managed production in rangeland agriculture. For these reasons the installation of apomixis into grasses is a widely stated goal in apomixis research (Khush, 1994; Savidan, 2000a,b; Grossniklaus et al., 2001). It appears likely, therefore, that the most amenable form of ‘synthetic’ apomixis, at least in the intermediate term, will be a form that incorporates a pseudogamous mechanism of endosperm formation (Dresselhaus et al., 2001; Grossniklaus, 2001). It follows that an understanding of the factors controlling endosperm formation is needed before apomixis can be engineered into cereal crops. Given the relative complexity of the endosperm tissue and the critical role it plays in seed development, it is not surprising that many mutants are known with alterations in endosperm form and/or function. Many of these also influence embryo development (Sorensen et al., 2002), a possible indication of the postulated common evolutionary origin of these tissues (Friedman, 1994; Carmichael and Friedman, 1995; Williams and Friedman, 2002). It appears likely, although unproven, that many of the events of endosperm development are similar during sexual and apomictic development, as has recently been demonstrated for gametophyte development in the facultative apomict Hieracium (Tucker et al., 2003). For this reason, the focus here is on the events surrounding endosperm initiation and the role of imprinting during embryo and endosperm development in flowering plants. In Arabidopsis, as in many other flowering plants, fruit development begins following fertilization in association with the initiation of seed formation. Using this feature, mutant screens for ‘apomixis genes’ have been performed by several groups based on the recovery of plants that are able to initiate fruit development in the absence of fertilization (Chaudhury and Peacock, 1993; Ohad et al., 1996; Chaudhury et al., 1997; Grossniklaus, 2001). The recovered genes, collectively known as FIS genes, all appear to be Polycomb-group proteins as described above, and all play a role in the repression of endosperm tissue proliferation. In fis mutants the endosperm overproliferates and is often poorly cellularized (Kinoshita et al., 1999; Sorensen et al., 2001). In fis1/mea and fis2 mutants a cellularized endosperm forms while in fie/fis3 development arrests at the formation of a syncytial endosperm. As described above, a further

374

R.A. Bicknell and A.S. Catanach

member of this group MEDICIS has recently been recorded (Guitton et al., 2004), the mutation of which not only leads to a similar endosperm phenotype but also to the formation of a rudimentary parthenogenetic embryo. Further to their role in endosperm cell division, FIS genes appear to play a role in determining the anterior–posterior axis of the developing endosperm (Sorensen et al., 2001). FIS genes demonstrate parent-of-origin expression in the developing seed. In homozygous form, FIS mutations are typically lethal, mutant stocks are generally maintained as heterozygotes. When conveyed through pollen, mutant fis alleles transmit with normal frequency. However, they cannot be recovered when a plant heterozygous for the mutation is used as the pistillate parent. This differential ability to transmit mutant alleles through male and female gametes is an example of a ‘maternal effect’. med/fis1 was the first maternal effect mutant described in plants and it was this characteristic, rather than its impact on endosperm formation, that led to its initial discovery (Grossniklaus et al., 1998). MEDEA and the other FIS genes are now known to be imprinted with paternal copies inactive at the time of fertilization, an effect mediated by differential DNA methylation (Vielle-Calzada et al., 1999; Kinoshita et al., 1999; Luo et al., 2000). Imprinting may be acting at a large number of loci during fertilization and early seed development. Vielle-Calzada et al. (2000) noted that 20 paternally derived alleles distributed throughout the Arabidopsis genome were not expressed during early seed development, leading the authors to suggest that imprinting may be a general phenomenon influencing the entire paternal gamete genome. Weijers et al. (2001), however, noted the expression of a paternally derived chimeric gene, AtRPS5A::GUS, as early as the two-cell embryo stage, indicating that the effect of imprinting is not a global effect on the paternal genome but differs between genes.

Comparative Gene Expression between Apomicts and Sexuals As apomixis appears to be essentially an altered form of sexuality, genes involved in aspects of sexual reproduction are likely to be under altered regulation in apomicts. Tucker et al. (2003) assessed b-glucuronidase (GUS) expression in sexual and apomictic forms of Hieracium, following the introduction of several fusion constructs combining the reporter gene with Arabidopsis promoters of regulatory factors in embryo and endosperm development. The Arabidopsis promoters assessed included three FIS class genes, SERK and SPOROCYTELESS (SPL) / NOZZLE (NZZ), which is required for both male and female sporogenesis (Schiefthaler et al., 1999; Yang et al., 1999). Patterns of expression during ovule and seed development in the different backgrounds were compared. The expression profiles of the FIS gene fusions during embryogenesis and endosperm development demonstrated a high degree of conservation, highlighting that apomixis shares developmental pathways with sexual reproduction. However, during earlier events associated with megasporogenesis, reporter gene expression showed more variation. Expression of FIS2:GUS in Arabidopsis was not found in the tetrad of spores following meiosis. In contrast, expression of FIS2:GUS in sexual

The Molecular Biology of Apomixis

375

Hieracium ovules occurred in the three micropylar spores destined for degeneration after meiosis. Apomictic Hieracium also followed this pattern; FIS2 was expressed in all four meiotic spores, all of which were destined for degeneration to be replaced by an apospore. The authors suggested that the expression of FIS2 in the meiotic tetrad indicates that FIS2 in Hieracium may play a different role than in Arabidopsis, possibly relating to the capacity of Hieracium for apomixis. Alternatively, SPOROCYTELESS (SPL)/NOZZLE showed conserved expression in MMCs of both Arabidopsis and Hieracium, but no analogous expression in aposporous initial cells was found. While these data provide a negative gene marker for apospory, positive expression associated with apospory or diplospory of any reproductive genetic marker, to our knowledge, remains undetected. Wide-scale gene expression analysis between ovules of apomicts and their sexual counterparts is a potential means of isolating novel genes associated with apomixis. The success of these approaches is dependent in part on the availability of plant material that is genetically similar in all respects other than mode of reproduction. This increases the likelihood that differences in gene expression are not due to alternative polymorphisms. Genetic similarity between plant materials for analysis has been achieved with the use of bulks from an F1 population segregating for mode of reproduction (Leblanc et al., 1997) or the use of sexual accessions of the same species of the apomict (VielleCalzada et al., 1996; Rodrigues et al., 2003). The salmon system in wheat offers homozygous isogenic lines, the standard line with normal sexual reproduction and two alloplasmic lines that show autonomous embryo development (Matzk, 1996). Alternatively, mutagenesis of an apomict to a sexual may be an effective method to generate near-isogenic lines that differ in mode of reproduction. Further considerations for wide-scale gene expression analysis include the timing of mRNA capture for analysis and the mode of detection of differentially expressed sequences. Differential display was used to detect expression differences between sexual and apomictic ovules of the aposporic grasses Pennisetum ciliare (Vielle-Calzada et al., 1996), Brachiaria brizantha (Leblanc et al., 1997; Rodrigues et al., 2003) and P. notatum (Pessino et al., 2001). Typically, low numbers of verifiable expression differences were detected, which is a further indication of similar developmental pathways being shared by the apomictic and sexual modes of reproduction. Differential gene expression of apomictic and sexual development in Panicum maximum was assessed by probing a cDNA library representing genes expressed in flower buds of an apomictic accession (Chen et al., 1999). The library was constructed from mRNA isolated at the point of appearance of aposporous initial cells, and then probed with cDNA derived from apomictic and sexual flower buds prior and during formation of aposporous initial cells. This strategy yielded a gene that was only expressed in aposporous flower buds at the time of aposporous initial cell development. An alternative approach is currently underway to isolate candidate genes from a diplosporic monosomic line of sugarbeet Beta vulgaris carrying a chromosome from Beta corolliflora that confers diplospory. A binary BAC library has been constructed and clones of the diplospory-conferring chromosome were isolated. These clones have

376

R.A. Bicknell and A.S. Catanach

been screened further, with a pool of cDNA enriched via subtractive hybridization, for BAC clones that contain ovular genes specific to the alien chromosome which may be key factors of diplospory in B. corolliflora (Fang et al., 2004). On the basis of homology searches against protein and nucleotide databases, some different products from Brachiaria could be assigned functions (Rodrigues et al., 2003). Those that were expressed more highly during megasporogenesis in apomictic ovules included a mitogen-activated protein (MAP) kinase and a protein translation factor, which may conceivably be acting as part of an apospory-specific signal cascade. Another one that is more highly expressed in apomictic ovules during megagametogenesis showed homology to a myosin heavy chain of Arabidopsis. Myosins are involved in cellular structure and transport and the authors suggest that differentially expressed myosin may be a reflection of differences in polarity between the eight-nucleate sexual Polygonum embryo sac and the four-nucleate unreduced aposporic Panicum-type embryo sac, in which antipodal cells are absent. Other comparative gene expression analyses have identified expression of structural genes associated with aposporous megagametogenesis. Pessino et al. (2001) isolated a sequence from floral spikelets of apomictic P. notatum, which contained homology to the KSP domain of several cytoskeletal elements, and Matzk et al. (1997) isolated an a-tubulin from ovules with differentiated embryo sacs from alloplasmic parthenogenetic wheat. While results of comparative gene expression analyses are starting to be meaningful, it is unclear whether differences detected so far are critical factors for the processes of apomixis, or are a reflection of the identities of different structures of ovules of apomicts. Differentially expressed sequences specifically linked to genomic loci of determinants of apomixis are yet to be reported. It is possible that recent technical advances such as laser-assisted microdissection, which enables the capture and subsequent analysis of single cells, may assist in further piecing together signal transduction events during both apomictic and sexual reproduction.

Potential Mechanisms of Apomixis: the Complexity of Epigenetic Regulation Conventionally translated proteins such as transcription factors and protein kinases clearly play an important role in gene regulation of developmental pathways. However, recent research into both plant and animal systems reveals that gene regulation is often also mediated by alternative molecular means. As mentioned above, parthenogenesis is a phenomenon found in animals as well as plants. Mammals have not been found to utilize parthenogenesis, however, as in sexual plants, the induction of parthenogenesis in mammals is receiving substantial experimental attention. Several aspects of fertilization and early embryogenesis appear to be conserved between animals and plants indicating that current advances being made in understanding animal systems will be very valuable for researchers studying plant models. Successful mammalian embryogenesis in part appears to be dependent on

The Molecular Biology of Apomixis

377

paternal gene silencing during early embryogenesis. Recently, a viable parthenogenetic mouse was developed from a reconstructed oocyte with two haploid maternal genomes. The success of this development was attributed to increased expression of Igf2 and monoallelic expression of the non-coding RNA, H19, which affected the expression of a wide range of genes. Igf2 and H19 are genes that are normally imprinted in the mammalian zygote (Kono et al., 2004). Non-coding RNA in eukaryotes may play a critical role in some cases of imprinting (Sleutels et al., 2002) and in other forms of epigenetic regulation such as dosage compensation of mammalian X-chromosome inactivation (Brockdorff et al., 1992). Other forms of epigenetic regulation include alteration of chromatin structure and chromosome pairing, whereby homologous pairing of alleles can enable one allele to influence the state of another (GrantDownton and Dickinson, 2004). Recent data indicate that transposable elements may also influence gene expression. A comparison of Igf2 regions in a range of mammalian genomes indicates that lack of short interspersed transposable elements (SINES) is associated with Igf2 imprinting (Weidman et al., 2004). Despite the association of retrotransposons with the characterized ASGR of Pennisetum (Akiyama et al., 2004), it is speculative to suggest that transposable elements have a role in apomixis. However, recent discoveries of the potential utility of transposable elements in gene regulation are surprising. Different families of transposable elements were represented differentially between cDNA libraries of oocytes and early embryos in the mouse. Furthermore, chimeric transcripts of transposable elements spliced to the 5’ ends of host genes were found, suggesting transposon-mediated regulation (Peaston et al., 2004). In the context of apomixis, the different components of the trait are likely to be sexual processes with altered gene regulation, and this may be conducted via epigenetic control. The growing understanding of epigenetic regulation of processes such as imprinting and parthenogenesis in other systems indicates that the mechanistic possibilities of how altered gene regulation of apomixis might be achieved are many and diverse, and may vary between different native apomictic systems.

The ‘Synthesis’ of Apomixis The successful installation of apomixis into sexual crops will require the integration of several component processes. Using natural processes of apomixis as a guide, it would appear that the simplest and most flexible system to emulate would be diplospory with pseudogamy, as seen in many grass genera. This system ensures the formation of a single embryo within the seed as the early meiotic apparatus is utilized as the basis for the formation of an unreduced embryo sac. This also ensures that the developing asexual structures are correctly positioned within the ovule and that their ontogeny is synchronously timed with critical related events such as the formation of other floral tissues and the dehiscence of pollen. As mentioned above, the constraints imposed by the imprinting of the paternal genome in most sexual species will either

378

R.A. Bicknell and A.S. Catanach

necessitate the use of systems that create a triploid endosperm about a diploid clonal embryo, such as that in the grass Panicum (Savidan, 1980), or mechanisms will be needed that relax the criteria for a precise 2:3 embryo/endosperm ploidy balance, as is seen in many apomicts, such as the grasses Tripsacum and Paspalum (Grimanelli et al., 1997; Quarin, 1999; Matzk et al., 2000). Furthermore, the engineering of apomixis in sexual systems in a commercially useful form will require the use of regulatory elements with sufficient specificity to ensure that targeted changes are succinctly directed to specific cell types and only at specific times. Ideally, apomixis should be introduced into crops in an inducible format, permitting it to be used during seed increase but silenced during hybridization. Promoters that provide some of these functions (Dresselhaus et al., 2001) are available but more are clearly required. Finally, as the discussion above indicates, most of the processes underlying apomixis remain incompletely understood but reassuring progress is being made in both the study of sexual and apomictic systems. There is quiet optimism among those researching this field that these hurdles will be overcome in the intermediate term and the great potential of apomixis will then be realized.

References Akiyama, Y., Conner, J.A., Goel, S., Morishige, D.T., Mullet, J.E., Hanna, W.W. and Ozias-Akins, P. (2004) High-resolution physical mapping in Pennisetum squamulatum reveals extensive chromosomal heteromorphism of the genomic region associated with apomixis. Plant Physiology 134, 1733–1741. Albertini, E., Barcaccia, G., Porceddu, A., Sorbolini, S. and Falcinelli, M. (2001a) Mode of reproduction is detected by Parth1 and Sex1 SCAR markers in a wide range of facultative apomictic Kentucky bluegrass varieties. Molecular Breeding 7, 293–300. Albertini, E., Porceddu, A., Ferranti, F., Reale, L., Barcaccia, G., Romano, B. and Falcinelli, M. (2001b) Apospory and parthenogenesis may be uncoupled in Poa pratensis: a cytological investigation. Sexual Plant Reproduction 14, 213–217. Alexander, D.E. and Beckett, J.B. (1963) Spontaneous triploidy and tetraploidy

in maize. Journal of Heredity 54, 103–106. Antoine, A.F., Faure, J.-E., Cordeiro, S., Dumas, C., Rougier, M. and Feijo´, J.A. (2000) A calcium influx is triggered and propagates in the zygote as a wavefront during in vitro fertilization of flowering plants. Proceedings of the National Academy of Sciences USA 97, 10643–10648. Asker, S. (1967) Induced sexuality after chromosome doubling in an apomictic Potentilla argentea-biotype. Hereditas 57, 339–342. Asker, S.E. and Jerling, L. (1992) Apomixis in Plants. CRC Press, Boca Raton, Florida. Babcock, E.B. and Stebbins, G.L. (1938) The American Species of Crepis: their Interrelationships and Distribution as Affected by Polyploidy and Apomixis. Carnegie Institution of Washington, Washington, DC. Baker, H.G. (1966) The evolution, functioning and breakdown of hetero-

The Molecular Biology of Apomixis morphic incompatibility systems. 1. The Plumbaginaceae. Evolution 20, 349–368. Bicknell, R.A. (1997) Isolation of a diploid, apomictic plant of Hieracium aurantiacum. Sexual Plant Reproduction 10, 168–172. Bicknell, R.A. (2001) Model systems to study the genetics and developmental biology of apomixis. In: Savidan, Y., Carman, J.G. and Dresselhaus, T. (eds) Flowering of Apomixis: from Mechanisms to Genetic Engineering. CIMMYT, IRD, European Commission DG VI, Mexico, pp. 111–120. Bicknell, R.A. and Bicknell, K.B. (1999) Who will benefit from apomixis? Biotechnology and Development Monitor 37, 17–21. Bicknell, R.A. and Koltunow, A.M. (2004) Understanding apomixis: recent advances and remaining conundrums. Plant Cell 16(Suppl.), S228–S245. Bicknell, R.A., Borst, N.K. and Koltunow, A.M. (2000) Monogenic inheritance of apomixis in two Hieracium species with distinct developmental mechanisms. Heredity 84, 228–237. Bicknell, R., Podivinsky, E., Catanach, A., Erasmuson, S. and Lambie, S. (2001) Strategies for isolating mutants in Hieracium with dysfunctional apomixis. Sexual Plant Reproduction 14, 227–232. Bicknell, R.A., Lambie, S.C. And Butler, R.C. (2003) Quantification of progeny classes in two facultatively apomictic accessions of Hieracium. Hereditas, 11–20. Birchler, J.A. (1993) Dosage analysis of maize endosperm development. Annual Review of Genetics 27, 181–204. Blakey, C.A., Goldman, S.L. and Dewald, C.L. (2001) Apomixis in Tripsacum: comparative mapping of a multigene phenomenon. Genome 44, 222–230. Bonnet, C. (1745) Traite d’insectologie ou observations sur les pucerons. Durand, Paris.

379 Boutilier, K., Offringa, R., Sharma, V.K., Kieft, H., Ouellet, T., Zhang, L., Hattori, J., Liu, C.M., Van Lammeren, A.A., Miki, B.L., Custers, J.B. and Van Lookeren Campagne, M.M. (2002) Ectopic expression of BABY BOOM triggers a conversion from vegetative to embryonic growth. Plant Cell 14, 1737–1749. Brockdorff, N., Ashworth, A., Kay, G.F., Mccabe, V.M., Norris, D.P., Cooper, P.J., Swift, S. and Rastan, S. (1992) The product of the mouse Xist gene is a 15 kb inactive X-specific transcript containing no conserved ORF and located in the nucleus. Cell 71, 515–526. Burt, A. (2000) Perspective: sex, recombination, and the efficacy of selection – was Weismann right? Evolution 54, 337–351. Calderini, O. and Mariani, A. (1997) Increasing 2n gamete production in diploid lucerne by cycles of phenotypic recurrent selection. Euphytica 93, 113–118. Carman, J.G. (1997) Asynchronous expression of duplicate genes in angiosperms may cause apomixis, bispory, tetraspory, and polyembryony. Biological Journal of Linnaean Society 61, 51–94. Carman, J.G. (2001) The gene effect: genome collisions and apomixis. In: Savidan, Y., Carman, J.G. and Dresselhaus, T. (eds) Flowering of Apomixis: from Mechanisms to Genetic Engineering. CIMMYT, IRD, European Commission DG VI, Mexico, pp. 95–110. Carman, J.G., Crane, C.F. and RiereLizarazu, O. (1991) Comparataive histology of cell walls during meiotic and apomeiotic megasporogenesis in two hexaploid Australasian Elymus species. Crop Science 31, 1527–1532. Carmichael, J.S. and Friedman, W.E. (1995) Double fertilisation in Gnetum gnemon. The relationships between the cell cycle and sexual reproduction. Plant Cell 7, 1975–1988.

380

R.A. Bicknell and A.S. Catanach Chapman, H. and Bicknell, R. (2000) Recovery of a sexual and an apomictic hybrid from crosses between the facultative apomicts Hieracium caespitosum and H. praealtum. New Zealand Journal of Ecology 24, 81– 85. Chase, S. (1969) Monoploids and monoploid-derivatives of maize (Zea mays L). Botanical Review 35, 117–167. Chaudhury, A.M. and Peacock, J.W. (1993) Approaches towards isolating apomictic mutants in Arabidopsis thaliana: prospects and progress. In: Khush, G.S. (ed.) Apomixis: Exploiting Hybrid Vigor in Rice. International Rice Research Institute, Los Ban˜os, Philippines, pp. 67–71. Chaudhury, A.M., Ming, L., Miller, C., Craig, S. and Dennis, E.S. (1997) Fertrilization-independant seed development in Arabidopsis thaliana. Proceedings of the National Academy of Sciences USA 94, 4223–4228. Chen, L., Miyazaki, C., Kojima, A., Saito, A. and Adachi, T. (1999) Isolation and characterisation of a gene expressed during early embryo sac development in an apomictic guinea grass (Panicum maximum). Journal of Plant Physiololy 154, 55–62. Cooper, D.C. and Brink, R.A. (1949) The endosperm-embryo relationship in the autonomous apomict, Taraxacum officinale. Botanica Gazette 111, 139–152. Correns, C. (1905) Gregor Mendels briefe an Carl Nageli 1866–1873. Abhandlungen der MathematischPhysikalischen Classe der Ko¨niglich Sa¨chsischen Gesellschaft der Wissenschaften 29, 189–256. Couteau, F., Belzile, F., Horlow, C., Granjean, D., Vezon, D. and Doutriaux, M.P. (1999) Random chromosome segregation without meiotic arest in both male and female meiocytes of a dmc1 mutant of Arabidopsis. Plant Cell 11, 1623–1634. Crane, C.F. (2001) Classification of apomictic mechanisms. In: Savidan, Y.,

Carman, J.G. and Dresselhaus, T. (eds) Flowering of Apomixis: from Mechanisms to Genetic Engineering. CIMMYT, IRD, European Commission DG VI, Mexico, pp. 24–34. Czapik, R. (1994) How to detect apomixis in Angiospermae. Polish Botanical Studies 8, 13–21. Darlington, C.D. (1939) The Evolution of Genetic Systems. Cambridge University Press, London. De Kovel, C.G.F. and De Jong, G. (1999) Responses of sexual and apomictic genotypes of Taraxacum officinale to variation in light. Plant Biology 1, 541–546. De Wet, J.M.J. (1968) Diploid-tetraploidhaploid cycles and the origin of variability in Dichanthium agamospecies. Evolution 22, 394–397. Dickinson, T.A. (1998) Taxonomy of agamic complexes in plants: a role for metapopulation thinking. Folia Geobotanica 33, 327–332. Digonnet, C., Alson, D., Leduc, N., Dumas, C. and Rougier, M. (1997) First evidence of a calcium transient in flowering plants at fertilization. Development 124, 2867–2874. Doutriaux, M.P., Couteau, F., Nergounioux, C. and White, C. (1998) Isolation and characterization of the RAD51 and DMC1 homologues from Arabidopsis thaliana. Molecular and General Genetics 257, 283–291. Dresselhaus, T., Carman, J.G. and Savidan, Y. (2001) Genetic engineering of apomixis in sexual crops: a critical assessment of the apomixis technology. In: Savidan, Y., Carman, J.G. and Dresselhaus, T. (eds) Flowering of Apomixis: from Mechanisms to Genetic Engineering. CIMMYT, IRD, European Commission DG VI, Mexico, pp. 229–243. Dumas, C. and Faure, J-E. (1995) Use of in vitro fertilization and zygote culture in crop improvement. Current Opinion in Biotechnology 6, 183–188. El Maataoui, M. and Pichot, C. (2001) Microsporogenesis in the endangered

The Molecular Biology of Apomixis species Cupressus dupreziana A. Camus: evidence for meiotic defects yielding unreduced and abortive pollen. Planta 213, 543–549. Ellerstrom, S. and Zagorcheva, L. (1977) Sterility and apomictic embryo-sac formation in Raphanobrassica. Hereditas 87, 107–120. Ernst, A. (1918) Bastardierung als Ursache der Apogamie im Pflanzenreich. Fischer, Jena, Germany. Fang, X., Gu, S., Xu, Z., Chen, F., Guo, D., Zhang, H.B. and Wu, N. (2004) Construction of a binary BAC library for an apomictic monosomic addition line of Beta corolliflora in sugar beet and identification of the clones derived from the alien chromosome. Theoretical and Applied Genetics 108, 1420–1425. Faure, J.-E., Digonnet, C. and Dumas, C. (1992) An in vitro system for adhesion and fusion of maize gametes. Science 263, 1598–1600. Fobis-Loisy, I., Miege, C. and Gaude, T. (2004) Molecular evolution of the S locus controlling mating in the brassicaceae. Plant Biology 6, 109–118. Foster, A.S. and Gifford, E.M. (1974) Comparative Morphology of Vascular Plants. W.H. Freeman, San Francisco, California. Friedman, W.E. (1994) The evolution of embryogeny in seed plants and the developmental origin and early history of endosperm. American Journal of Botany 8, 1468–1486. Gadella, T.W.J. (1984) Cytology and the mode of reproduction of some taxa of Hieracium subgnenus Pilosella. Proceedings of the Koninklijke Nederlandse Akademie van Wetenschappen C. 87, 387–399. Gadella, T.W.J. (1991) Variation, hybridization and reproductive biology of Hieracium pilosella L. Proceedings of the Koninklijke Nederlandse Akademie van Wetenschappen C. 94, 455–488. Galitski, T., Saldanha, A.J., Styles, C.A., Lander, E.S. and Fink, G.R. (1999)

381 Ploidy regulation of gene expression. Science 285, 251–254. Garcia, R., Asins, M.J., Forner, J. and Carbonell, E.A. (1999) Genetic analysis of apomixis in Citrus and Poncirus by molecular markers. Theoretical and Applied Genetics 99, 511– 518. Goel, S., Chen, Z.B., Conner, J.A., Akiyama, Y., Hanna, W.W. and Ozias-Akins, P. (2003) Delineation by fluorescence in situ hybridization of a single hemizygous chromosomal region associated with aposporous embryo sac formation in Pennisetum squamulatum and Cenchrus ciliaris. Genetics 163, 1069–1082. Golubovskaya, I.N. (1979) Genetic control of meiosis. International Review of Cytology 58, 247–290. Golubovskaya, I.N., Avalkin, N.A. and Sheridan, W.F. (1992) Effect of several meiotic mutants on female meiosis in maize. Developmental Genetics 13. Golubovskaya, I.N., Grebennikova, Z.K., Avalkin, N.A. and Sheridan, W.F. (1993) The role of the ameiotic 1 gene in the initiation of meiosis and in subsequent meiotic events in maize. Genetics 135, 1151–1166. Golubovskaya, I.N., Avalkin, N.A. and Sheridan, W.F. (1997) New insights into the role of the maize ameiotic 1 locus. Genetics 147, 1339–1350. Grant-Downton, R.T. and Dickinson, H.G. (2004) Plants, pairing and phenotypes – two’s company? Trends in Genetics 20, 188–195. Grimanelli, D., Tohme, J. and GonzalezDe-Leon, D. (2001b) Applications of molecular genetics in apomixis research. In: Savidan, Y., Carman, J.G. and Dresselhaus, T. (eds) Flowering of Apomixis: from Mechanisms to Genetic Engineering. CIMMYT, IRD, European Commission DG VI, Mexico, pp. 83–94. Grimanelli, D., Hernandez, M., Perotti, E. and Savidan, Y. (1997) Dosage effects in the endosperm of the diplosporous apomictic Tripsacum (Poaceae). Sex-

382

R.A. Bicknell and A.S. Catanach ual Plant Reproduction 10, 279– 282. Grimanelli, D., Leblanc, O., Perotti, E. and Grossniklaus, U. (2001a) Developmental genetics of gametophytic apomixis. Trends in Genetics 17, 597–604. Grimanelli, D., Leblanc, O., Espinosa, E., Perotti, E., Gonzalez De Leon, D. and Savidan, Y. (1998a) Non-Mendelian transmission of apomixis in maizeTripsacum hybrids caused by a transmission ratio distortion. Heredity 80, 40–47. Grossniklaus, U. (2001) From sexuality to apomixis: molecular and genetic approaches. In: Savidan, Y., Carman, J.G. and Dresselhaus, T. (eds) Flowering of Apomixis: from Mechanisms to Genetic Engineering. CIMMYT, IRD, European Commission DG VI, Mexico, pp. 168–211. Grossniklaus, U., Vielle-Calzada, J.-P., Hoeppner, M.A. and Gagliano, W.B. (1998) Maternal control of embryogenesis by MEDEA, a Polycomb group gene in Arabidopsis. Science 280, 446–450. Grossniklaus, U., Spillane, C., Page, D.R. and Kohler, C. (2001) Genomic imprinting and seed development: endosperm formation with and without sex. Current Opinion in Plant Biology 4, 21–27. Guitton, A.E., Page, D.R., Chambrier, P., Lionnet, C., Faure, J.E., Grossniklaus, U. and Berger, F. (2004) Identification of new members of Fertilisation Independent Seed Polycomb Group pathway involved in the control of seed development in Arabidopsis thaliana. Development 131, 2971–2981. ˚ . (1943) The Genesis of the Gustafsson, A European Blackberry Flora. Lund University, Svalof, Sweden. ˚ . (1947a) Apomixis in Gustafsson, A Higher Plants. II, 69–179. Lund University, Svalof, Sweden. ˚ . (1947b) Apomixis in Gustafsson, A Higher Plants. III, 183–370. Lund University, Svalof, Sweden.

Hagberg, A. and Hagberg, G. (1980) High frequency of spontaneous haploids in the progeny of an induced mutation in barley. Hereditas 93, 341– 343. Haig, D. and Westoby, M. (1991) Genomic imprinting in endosperm: its effect on seed development in crosses between species, and between different ploides of the same species, and its implications for the evolution of apomixis. Philosophical Transactions of the Royal Society of London, Series B 333, 1–13. Hanna, W.W. (1995) Use of apomixis in cultivar development. Advances in Agronomy. 54, 333–350. Harlan, J.R. and De Wet, J.M.J. (1975) ¨ . Winge and a prayer: the origins On O of polyploidy. Botanical Review 41, 361–390. Hecht, V., Vielle-Calzada, J.P., Hartog, M.V., Schmidt, E.D.L., Boutilier, K., Grossniklaus, U. and De Vries, S.C. (2001) The Arabidopsis SOMATIC EMBRYOGENESIS RECEPTOR KINASE 1 gene is expressed in developing ovules and embryos and enhances embryogenic competence in culture. Plant Physiology 127, 803–816. Heenan, P.B., Dawson, M.I. and Bicknell, R.A. (2002) Evidence for apomictic seed formation in Coprosma waima (Rubiaceae). New Zealand Journal of Botany 40, 347–355. Henderson, J.T., Li, H.C., Rider, S.D., Mordhorst, A.P., Romero-Severson, J., Cheng, J.C., Robey, J., Sung, Z.R., De Vries, S.C. and Ogas, J. (2004) PICKLE acts throughout the plant to repress expression of embryonic traits and may play a role in gibberellin-dependent responses. Plant Physiology 134, 995–1005. Horandl, E. (1998) Species concepts in agamic complexes: applications in the Ranunculus auricomus complex and general perspectives. Folia Geobotanica 33, 335–348. Jaffe, L.F. (1996) Egg membranes during fertilization. In: Schultz, S.G.E.A. (ed.)

The Molecular Biology of Apomixis Molecular Biology of Membrane Transport Disorders. Plenum Press, New York, pp. 367–378. Jefferson, R.A. and Bicknell, R.A. (1995) The potential impacts of apomixis: a molecular genetics approach. In: Sobral, B.W.S. (ed.) The Impact of Plant Molecular Genetics. Birklha¨user, Boston, Massachusetts, pp. 87–101. Jessup, R.W., Burson, B.L., Burow, G.B., Wang, Y.W., Chang, C., Li, Z., Paterson, A.H. and Hussey, M.A. (2002) Disomic inheritance, suppressed recombination, and allelic interactions govern apospory in buffelgrass as revealed by genome mapping. Crop Science 42, 1688–1694. Johnston, S.A., Den Nijs, T.P.M., Peloquin, S.J. and Hanneman Jr, R.E. (1980) The significance of genic balance to endosperm development in interspecific crosses. Theoretical and Applied Genetics 57. Khush, G.S. (1994) Apomixis: Exploiting Hybrid Vigor in Rice. IRRI, Los Banos. Kimber, G. and Riley, R. (1963) Haploid angiosperms. Botanical Reviews 29, 480–531. Kinoshita, T., Yadegari, R., Harada, J.J., Goldberg, R.B. and Fischer, R.L. (1999) Imprinting of the MEDEA polycomb gene in Arabidopsis endosperm. Plant Cell 11, 1945–1952. Klimyuk, V.I. and Jones, J.D.G. (1997) AtDMC1, the Arabidopsis homologue of the yeast DMC1 gene: characterization, transposon-induced allelic variation and meiosis associated expression. Plant Journal 11, 1–14. Kojima, A. and Nagato, Y. (1997) Discovery of highly apomictic and highly amphimictic dihaploids in Allium tuberosum. Sexual Plant Reproduction 10, 8–12. Koltunow, A.M. (1993) Apomixis: embryo sacs and embryos formed without meiosis of fertilisation in ovules. Plant Cell 5, 1425–1437.

383 Koltunow, A.M. (2000) The genetic and molecular analysis of apomixis in the model plant Hieracium. Acta Biologica Cracoviensia Series Botanica 42, 61–72. Koltunow, A. and Grossniklaus, U. (2003) Apomixis: a developmental perspective. Annual Review of Plant Biology 54, 547–574. Koltunow, A.M., Bicknell, R.A. and Chaudhury, A.M. (1995a) Apomixis: molecular strategies for the generation of genetically identical seeds without fertilization. Plant Physiology 108, 1345–1352. Koltunow, A.M., Soltys, K., Nito, N. and Mcclure, S. (1995b) Anther, ovule, seed and nucellar embryo development in Citrus sinensis cv. Valencia. Canadian Journal of Botany 73, 1567– 1582. Koltunow, A.M., Johnson, S.D. and Bicknell, R.A. (1998) Sexual and apomictic development in Hieracium. Sexual Plant Reproduction 11, 213–230. Kono, T., Obata, Y., Wu, Q., Niwa, K., Ono, Y., Yamamoto, Y., Park, E.S., Seo, J.S. and Ogawa, H. (2004) Birth of parthenogenetic mice that can develop to adulthood. Nature 428, 860– 864. Krahulcova, A., Chrtek, J. and Krahulec, F. (1999) Autogamy in Hieracium subgen. Pilosella. Folia Geobotanica & Phytotaxonomica 34, 373–376. Kranz, E. and Dresselhaus, T. (1996) In vitro fertilization with isolated higher plant gametes. Trends in Plant Science 1, 82–89. Labombarda, P., Busti, A., Caceres, M.E., Pupilli, F. and Arcioni, S. (2002) An AFLP marker tightly linked to apomixis reveals hemizygosity in a portion of the apomixis-controlling locus in Paspalum simplex. Genome 45, 513–519. Lamote, V., Baert, J., Roldan-Ruiz, I., De Loose, M. and Van Bockstaele, E. (2002) Tracing of 2n egg occurrence in perennial ryegrass (Lolium perenne L.)

384

R.A. Bicknell and A.S. Catanach using interploidy crosses. EUPHYTICA 123, 159–164. Leblanc, O. and Mazzucato, A. (2001) Screening procedures to identify and quantify apomixis. In: Savidan, Y., Carman, J.G. and Dresselhaus, T. (eds) Flowering of Apomixis: from Mechanisms to Genetic Engineering. CIMMYT, IRD, European Commission DG VI, Mexico, pp. 121–136. Leblanc, O., Grimanelli, D., Gonzalez De Leon, D. and Savidan, Y. (1995) Detection of the apomictic mode of reproduction in maize-Tripsacum hybrids using maize RFLP markers. Theoretical and Applied Genetics 90, 1198–1203. Leblanc, O., Armstead, I., Pessino, S., Ortiz, J.P.A., Evans, C., Do Valle, C. and Hayward, M.D. (1997) Non-radioactive mRNA fingerprinting to visualise gene expression in mature ovaries of Brachiaria hybrids derived from B. brizantha, an apomictic tropical forage. Plant Science 126, 49–58. Lee, H.S. and Chen, Z.L. (2001) Proteincoding genes are epigenetically regulated in Arabidopsis polyploids. Proceedings of the National Academy of Sciences USA 98, 6753–6758. Levin, D.A. (1983) Polyploidy and novelty in flowering plants. American Naturalist 122, 1–25. Lin, B.-Y. (1984) Ploidy barrier to endosperm development in maize. Genetics 107, 103–115. Lotan, T., Ohto, M., Matsudaira Yee, K., West, M.A.L., Lo, R., Kwong, R.W., Yamagishi, K., Fischer, R.L., Goldberg, R.B. and Harada, J.J. (1998) Arabidopsis LEAFY COTYLEDON1 is sufficient to induce embryo development in vegetative cells. Cell 93, 1195–1205. Luo, M., Bilodeau, P., Dennis, E.S., Peacock, W.J. and Chaudhury, A. (2000) Expression and parent-of-origin effects for FIS2, MEA, and FIE in the endosperm and embryo of developing Arabidopsis seeds. Proceedings of the

National Academy of Sciences USA 97, 10637–10642. Malecka, J. (1973) Problems of the mode of reproduction in microspecies of Taraxacum section Palustria Dahlstedt. Acta Biologica Cracov 14, 37–87. Martinez, E.J., Urbani, M.H., Quarin, C.L. and Ortiz, J.P.A. (2001) Inheritance of apospory in bahiagrass, Paspalum notatum. Hereditas 135, 19–25. Matzk, F. (1996) The ‘Salmon system’ of wheat – a suitable model for apomixis research. Hereditas 125, 299–301. Matzk, F., Meyer, H.-M., Ba¨umlein, H., Balzer, H.-J. and Schubert, I. (1995) A novel approach to the analysis of the initiation of embryo development in Gramineae. Sexual Plant Reproduction 8, 266–272. Matzk, F., Meyer, H.M., Horstmann, C., Balzer, H.J., Bau¨mlein, H. and Schubert, I. (1997) A specific alpha-tubulin is associated with the initiation of parthenogenesis in ‘Salmon’ wheat lines. Hereditas 126, 219–224. Matzk, F., Meister, A. and Schubert, I. (2000) An efficient screen for reproductive pathways using mature seeds of monocots and dicots. Plant Journal 21, 97–108. Matzk, F., Meister, A., Brutovska, R. and Schubert, I. (2001) Reconstruction of reproductive diversity in Hypericum perforatum L. opens novel strategies to manage apomixis. Plant Journal 26, 275–282. Matzk, F., Hammer, K. and Schubert, I. (2003) Coevolution of apomixis and genome size within the genus Hypericum. Sexual Plant Reproduction 16, 51–58. McVean, G.T. (1995) Fractious chromosomes: hybrid disruption and the origin of selfish genetic elements. BioEssays 17, 579–582. Meinke, D.W., Franzmann, L.H., Nickle, T.C. and Yeung, E.C. (1994) Leafy cotyledon mutants of Arabidopsis. Plant Cell 6, 1049–1064.

The Molecular Biology of Apomixis ¨ ber einige aus ku¨nsMendel, G. (1869) U tlicher Befruchtung gewonnene Hieracium-Bastarde. Verhandlungen des naturforchenden Vereins in Bru¨nn 8, 26–31. Mogie, M. (1992) The Evolution of Asexual Reproduction in Plants. Chapman and Hall, London. Morgan, R., Ozias-Akins, P. and Hanna, W.W. (1998) Seed set in an apomictic BC3 pearl millet. International Journal of Plant Sciences 159, 89–97. Morgensen, H.L. (1988) Exclusion of male mitochondria and plastids during syngamy in barley as a basis for maternal inheritance. Proceedings of the National Academy of Sciences USA 85, 2594–2597. Muller, H.J. (1964) The relation of recombination to mutational advance. Mutatation Research 1, 2–9. Mu¨ntzing, A. (1936) The evolutionary significance of autoployploidy. Hereditas 21, 263–378. Naumova, T.N., Den Nijs, J.C.M. and Willemse, M.T.M. (1992) Quantitative analysis of aposporous parthenogenesis in Poa pratensis genotypes. Acta Botanica Neerlandica 43, 1–14. Naumova, T.N., Osadtchiy, J.V., Sharma, V.K., Dijkhuis, P. and Ramulu, K.S. (1999) Apomixis in plants: structural and functional aspects of diplospory in Poa nemoralis and P. palustris. Protoplasma 208, 186–195. Nel, P.M. (1975) Crossing over and diploid egg formation in the elongate mutant of maize. Genetics 79, 435–450. Nishiyama, I. and Yabuno, T. (1978) Causal relationships between the polar nuclei in double fertilization and interspecific cross-incompatibility in Avena. Cytologia 43, 453–466. Nogler, G.A. (1982) How to obtain diploid apomictic Ranunculus auricomus plants not found in the wild state. Botanica Helvetica 92, 13–22. Nogler, G.A. (1984a) Gametophytic Apomixis. In: Johri, B.M. (ed.) Embryology

385 of Angiosperms. Springer-Verlag, Berlin, pp. 475–518. Nogler, G.A. (1984b) Genetics of apospory in apomictic Ranunculus auricomus. V. Conclusion. Botanica Helvetica 94, 411–422. Nogler, G.A. (1986) Genetics of apomictic reproduction. Giornale Botanico Italiano 120, 49–52. Nogler, G.A. (1994) Genetics of gametophytic apomixis – a historical sketch. Polish Botanical Studies 8, 5–11. Noyes, R.D. (2000) Diplospory and parthenogenesis in sexual x agamospermous (apomictic) Erigeron (Asteraceae) hybrids. International Journal of Plant Sciences 161, 1–12. Noyes, R.D. and Rieseberg, L.H. (2000) Two independent loci control agamospermy (apomixis) in the triploid flowering plant Erigeron annuus. Genetics 155, 379–390. Nuccitelli, R. (1991) How do sperm activate eggs? Current topics in Developmental Biology 25, 1–16. Nygren, A. (1948) Further studies in spontaneous and synthetic Calamagrostis purpurea. Hereditas 34, 113–134. O’Connell, L.M. and Eckert, C.G. (1999) Differentiation in sexuality among populations of Antennaria parlinii (Asteraceae). International Journal of Plant Sciences. 160, 567–575. Ogas, J., Kaufmann, S., Henderson, J. and Somerville, C. (1999) PICKLE is a CHD3 chromatin-remodeling factor that regulates the transition from embryonic to vegetative development in Arabidopsis. Proceedings of the National Academy of Sciences USA 96, 13839–13844. Ohad, N., Margossian, L., Hsu, Y.-C., Williams, C., Repetti, P. and Fischer, R.L. (1996) A mutation that allows endosperm development without fertilization. Proceedings of the National Academy of Sciences USA 93, 5319–5324. Ostenfeld, C.H. (1904) Weitere Beitrage zur Kenntnis der Fruchtentwicklung bei der

386

R.A. Bicknell and A.S. Catanach Gattung Hieracium. Berichte der Deutschen Botanischen Gesellschaft 22, 537–541. Ostenfeld, C.H. (1906) Experimental and cytological studies in the Hieracia. I. Castration and hybridization experiments with some species of Hieracia. Botanisk Tidsskrift 27, 225–248. Ostenfeld, C.H. (1910) Further studies on the apogamy and hybridization of the Hieracia. Zeitschrift fuer Indukt Abstamm und Vererbungslehre 3, 241–285. Ozias-Akins, P., Lubbers, E.L., Hanna, W.W. and Mcnay, J.W. (1993) Transmission of the apomictic mode of reproduction in Pennisetum: co-inheritance of the trait and molecular markers. Theoretical and Applied Genetics 85, 632–638. Ozias-Akins, P., Roche, D. and Hanna, W.W. (1998) Tight clustering and hemizygosity of apomixis-linked molecular markers in Pennisetum squamulatum genetic control of apospory by a divergent locus that may have no allelic form in sexual genotypes. Proceedings of the National Academy of Sciences USA 95, 5127–5132. Palacios, C., Rossello, J.A. and GonzalezCandelas, F. (2000) Study of the evolutionary relationships among Limonium species (Plumbaginaceae) using nuclear and cytoplasmic molecular markers. Molecular Phylogenetics & Evolution 14, 232–249. Palmer, R.G. (1971) Cytological studies of ameiotic and normal maize with reference to premeiotic pairing. Chromosoma 35, 233–246. Peaston, A.E., Evsikov, A.V., Graber, J.H., De Vries, W.N., Holbrook, A.E., Solter, D. and Knowles, B.B. (2004) Retrotransposons regulate host genes in mouse oocytes and pre-implantation embryos. Development Cell 7, 597– 606. Peck, J.R. and Waxman, D. (2000) What’s wrong with a little sex? Journal of Evolutionary Biology 13, 63–69.

Pessino, S.C., Ortiz, J.P., Leblanc, O., Do Valle, C.B., Evans, C. and Hayward, M.D. (1997) Identification of a maize linkage group related to apomixis in Brachiaria. Theoretical and Applied Genetics 94, 439–444. Pessino, S.C., Evans, C., Ortiz, J.P.A., Armstead, I., Do Valle, C.B. and Hayward, M.D. (1998) A genetic map of the apospory-region in Brachiaria hybrids: identification of two markers closely associated with the trait. Hereditas 128, 153–158. Pessino, S.C., Ortiz, J.P.A., Hayward, M.D. and Quarin, C.L. (1999) The molecular genetics of gametophytic apomixis. Hereditas 130, 1–11. Pessino, S.C., Espinoza, F., Martinez, E.J., Ortiz, J.P.A., Valle, E.M. and Quarin, C.L. (2001) Isolation of cDNA clones differentially expressed in flowers of apomictic and sexual Paspalum notatum. Hereditas 134, 35–42. Philipson, M.N. (1978) Apomixis in Cortaderia jubata (Gramineae). New Zealand Journal of Botany 16, 45–59. Pichot, C., El Maataoui, M., Raddi, S. and Raddi, P. (2001) Surrogate mother for endangered Cupressus. Nature 412, 39. Porsild, A.E. (1965) The genus Antennaria in eastern Arctic and subarctic America. Botanisk Tidsskrift 61, 22–55. Pupilli, F., Labombarda, P., Caceres, M.E., Quarin, C.L. and Arcioni, S. (2001) The chromosome segment related to apomixis in Paspalum simplex is homoeologous to the telomeric region of the long arm of rice chromosome 12. Molecular Breeding 8, 53–61. Quarin, C.L. (1999) Effect of pollen source and pollen ploidy on endosperm formation and seed set in pseudogamous apomictic Paspalum notatum. Sexual Plant Reproduction 11, 331–335. Quarin, C.L., Espinoza, F., Martinez, E.J., Pessino, S.C. and Bovo, O.A. (2001) A rise of ploidy level induces the expre-

The Molecular Biology of Apomixis ssion of apomixis in Paspalum notatum. Sexual Plant Reproduction 13, 243–249. Reddy, P.S. and D’Cruz, R. (1969) Mechanism of apomixis in Dichanthium annulatum (Forssk) Staph. Botanica Gazette 139, 71–79. Rhoades, M.M. and Dempsey, E. (1966) Induction of chromosome doubling at meiosis by the elongate gene in maize. Genetics 54, 505–522. Richards, A.J. (1997) Plant Breeding Systems, 2nd edn. Chapman & Hall, London. Rider, S.D.J., Henderson, J.T., Jerome, R.E., Edenberg, H.J., Romero-Severson, J. and Ogas, J. (2003) Coordinate repression of regulators of embryonic identity by PICKLE during germination in Arabidopsis. Plant Journal 35, 33–43. Roche, D., Cong, P.S., Chen, Z.B., Hanna, W.W., Gustine, D.L., Sherwood, R.T. and Ozias-Akins, P. (1999) An apospory-specific genomic region is conserved between Buffelgrass (Cenchrus ciliaris L.) and Pennisetum squamulatum Fresen. Plant Journal 19, 203–208. Roche, D., Chen, Z.B., Hanna, W.W. and Ozias-Akins, P. (2001a) Non-Mendelian transmission of an aposporyspecific genomic region in a reciprocal cross between sexual pearl millet (Pennisetum glaucum) and an apomictic F1 (P. glaucum  P. squamulatum). Sexual Plant Reproduction 13, 217–223. Roche, D., Hanna, W.W. and OziasAkins, P. (2001b) Is supernumerary chromatin involved in gametophytic apomixis of polyploid plants? Sexual Plant Reproduction 13, 343–349. Roche, D., Conner, J.A., Budiman, M.A., Frisch, D., Wing, R., Hanna, W.W. and Ozias-Akins, P. (2002) Construction of BAC libraries from two apomictic grasses to study the microcolinearity of their apospory-specific genomic regions. Theoretical and Applied Genetics 104, 804–812.

387 Rodrigues, J.C.M., Cabral, G.B., Dusi, D.M.A., De Mello, L.V., Rigden, D.J. and Carneiro, V.T.C. (2003) Identification of differentially expressed cDNA sequences in ovaries of sexual and apomictic plants of Brachiaria brizantha. Plant Molecular Biology 53, 745–757. ¨ ber die EmbryoRosenberg, O. (1906) U bildung in der Gattung Hieracium. Berichte der Deutschen Botanischen Gesellschaft 24, 157–161. Rosenberg, O. (1907) Experimental and cytological studies in the Hieracia. II: cytological studies on the apogamy in Hieracium. Botanisk Tidsskrift 143–170. Rougier, M., Antoine, A.F., Aldon, D. and Dumas, C. (1996) New insights in early steps of in vitro fertilization in plants. Sexual Plant Reproduction 9, 324– 329. Sarkar, K.R. and Coe, E.H. (1966) A genetic analysis of the origin of maternal haploids in maize. Genetics 54, 453– 464. Savidan, Y. (1980) Chromosomal and embryological analyses in sexual  apomictic hybrids of Panicum maximum Jacq. Theoretical and Applied Genetics 57, 153–156. Savidan, Y. (1981) Genetics and utilization of apomixis for the improvement of Guineagrass (Panicum maximum Jacq.). International Grasslands Congress, Lexington, Kentucky, p. 182. Savidan, Y.H. (1992) Progress in research on apomixis and its transfer to major grain crops. In: Dattee, Y., Dumas, C. and Gallais, A. (eds) Reproductive Biology and Plant Breeding. Springer-Verlag, Berlin, pp. 269– 280. Savidan, Y. (2000a) Apomixis: genetics and breeding. In: Janick, J. (ed.) Plant Breeding Reviews. John Wiley & Sons, New York, pp. 13–86. Savidan, Y. (2000b) Apomixis, the way of cloning seeds. Biofutures 2000, 38–43.

388

R.A. Bicknell and A.S. Catanach Savidan, Y. (2001) Transfer of apomixis through wide crosses. In: Savidan, Y., Carman, J.G. and Dresselhaus, T. (eds) Flowering of Apomixis: from Mechanisms to Genetic Engineering. CIMMYT, IRD, European Commission DG VI, Mexico, pp. 153–167. Schiefthaler, U., Balasubramanian, S., Sieber, P., Chevalier, D., Wisman, E. and Schneitz, K. (1999) Molecular analysis of NOZZLE, a gene involved in pattern formation and early sporogenesis during sex organ development in Arabidopsis thaliana. Proceedings of the National Academy of Sciences USA 96, 11664–11669. Schmidt, E.D., Guzzo, F., Toonen, M.A. and De Vries, S.C. (1997) A leucinerich repeat containing receptor-like kinase marks somatic plant cells competent to form embryos. Development 124, 2049–2062. Scott, R.J., Spielman, M., Bailey, J. and Dickinson, H.G. (1998) Parent-of-origin effects on seed development in Arabidopsis thaliana. Development 125, 3329–3341. Sharbel, T.F. and Mitchell-Olds, T. (2001) Recurrent polyploid origins and chloroplast phylogeography in the Arabis holboellii complex (Brassicaceae). Heredity 87, 59–68. Sherwood, R.T., Berg, C.C. and Young, B.A. (1994) Inheritance of apospory in buffelgrass. Crop Science 34, 1490– 1494. Siddiqi, I., Ganesh, G., Grossniklaus, U. and Subbiah, V. (2000) The dyad gene is required for progression through female meiosis in Arabidopsis. Development 127, 197–207. Sleutels, F., Zwart, R. and Barlow, D.P. (2002) The non-coding Air RNA is required for silencing autosomal imprinted genes. Nature 415, 810–813. Smith, J. (1841) Notice of a plant which produces seeds without any apparent action of pollen. Transactions of the Linnaean Society, 18. Sokolov, V.A., Kindiger, B. and Khatypova, I.V. (1998) Apomictically

reproducing 39-chromosome maizeTripsacum hybrids. Genetika 34, 499–506. Sørensen, T. and Gudjonsson, G. (1946) Spontaneous chromosomal aberrants in apomictic Taraxaca. Kongelige Danske Videnskabernes Selskab Biologiske Skrifter 4, 3–48. Sorensen, M.B., M., C.A., Robert, H., Bancharel, E. and Berger, F. (2001) Polycomb group genes control pattern formation in plant seed. Current Biology 11, 277–281. Sorensen, M.B., Mayer, U., Lukowitz, W., Robert, H., Chambrier, P., Jurgens, G., Somerville, C., Lepiniec, L. and Berger, F. (2002) Cellularisation in the endosperm of Arabidopsis thaliana is coupled to mitosis and shares multiple components with cytokinesis. Development 129, 5567–5576. Stebbins, G.L. (1950) Variation and Evolution in Plants. Columbia University Press, New York. Stebbins, G.L. (1971) Chromosomal Evolution in Higher Plants. Edward Arnold, London. Stone, S.L., Kwong, L.W., Yee, K.M., Pelletier, J., Lepiniec, L., Fischer, R.L., Goldberg, R.B. and Harada, J.J. (2001) LEAFY COTYLEDON2 encodes a B3 domain transcription factor that induces embryo development. Proceedings of the National Academy of Sciences USA 98, 11806– 11811. Suomalainen, E., Saura, A. and J., L. (1987) Cytology and Evolution in Parthenogenesis. CRC Press, Boca Raton, Florida. Teppner, H. and Klein, E. (1993) Nigritella gabasiana spec. nova, N. nigra subsp. iberica subsp. nova (Orchidaceae – Orchideae) und deren Embryologie. Phyton 33, 179–209. Toenniessen, G.H. (2001) Feeding the world in the 21st century: plant breeding, biotechnology, and the potential role of apomixis. In: Savidan, Y., Carman, J.G. and Dresselhaus, T. (eds) Flowering of Apomixis: from Mechan-

The Molecular Biology of Apomixis isms to Genetic Engineering. CIMMYT, IRD, European Commission DG VI, Mexico, pp. 1–7. Tucker, M.R., Araujo, A.-C.G., Paech, N.A., Hecht, V., Schmidt, E.D.L., Rossell, J.-B., De Vries, S.C. and A.M.G., K. (2003) Sexual and apomictic reproduction in Hieracium subgenus Pilosella are closely interrelated developmental pathways. Plant Cell 15, 1524–1537. Valle, C.B., Glienke, C. and Leguizamon, G.O.C. (1994) Inheritance of apomixis in Brachiaria, a tropical forage grass. Apomixis Newsletter 7, 42–43. Van Dijk, P.J. (2003) Ecological and evolutionary opportunities of apomixis: insights from Taraxacum and Chondrilla. Philosophical Transactions of the Royal Society, Series B 358, 1113–1121. Van Dijk, P. and Van Damme, J. (2000) Apomixis technology and the paradox of sex. Trends in Plant Science 5, 81–84. Van Dijk, P.J., Tas, I.C.Q., Falque, M. and Bakx-Schotman, T. (1999) Crosses between sexual and apomictic dandelions (Taraxacum). II. The breakdown of apomixis. Heredity 83, 715– 721. Van Dijk, P.J., Van Baarlen, P. and De Jong, J.H. (2003) The occurrence of phenotypically complementary apomixis-recombinants in crosses between sexual and apomictic dandelions (Taraxacum officinale). Sexual Plant Reproduction 16, 71–76. Veilleux, R. (1985) Diploid and polyploid gametes in crop plants: mechanisms of formation and utilization in plant breeding. Plant Breeding Reviews 3, 253–288. Veronesi, F., Mariani, A. and Tavoletti, S. (1988) Screening for 2n gamete producers in diploid species of genus Medicago. Genetica Agraria 42, 187–200. Vielle-Calzada, J.P., Nuccio, M.L., Budiman, M.A., Thomas, T.L., Burson, B.L., Hussey, M.A. and Wing, R.A. (1996) Comparative gene expression

389 in sexual and apomictic ovaries of Pennisetum ciliare (L.) Link. Plant Molecular Biology 32, 1085–1092. Vielle-Calzada, J.-P., Thomas, J., Spillane, C., Coluccio, A., Hoeppner, M.A. and Grossniklaus, U. (1999) Maintenance of genomic imprinting at the Arabidopsis medea locus requires zygotic DDM1 activity. Genes & Development 13, 2971–2982. Vielle-Calzada, J.-P., Basker, R. and Grossniklaus, U. (2000) Delayed activation of the paternal genome during seed development. Nature 404, 91–94. Vijverberg, K., Van Der Hulst, R., Lindhout, P. and Van Dijk, P.J. (2004) A genetic linkage map of the diplosporous chromosomal region in Taraxacum (common dandelion; Asteraceae). Theoretical and Applied Genetics 108, 725–732. Voigt, P.W. and Burson, B.L. (1983) Breeding of apomictic Eragrostis curvula. 14th International Grasslands Congress, pp. 160–163. Weidman, J.R., Murphy, S.K., Nolan, C.M., Dietrich, F.S. and Jirtle, R.L. (2004) Phylogenetic footprint analysis of IGF2 in extant mammals. Genome Res 14, 1726–1732. Weijers, D., Geldner, N., Offringa, R. and Jurgens, G. (2001) Seed development: early paternal gene activity in Arabidopsis. Nature 414, 709–710. Weld, R.J., Bicknell, R.A., Heinemann, J.A. and Eady, C.C. (2002) Ds transposition mediated by transient transposase expression in Heiracium aurantiacum. Plant Cell Tissue and Organ Culture 69, 45–54. West, M., Yee, K.M., Danao, J., Zimmerman, J.L., Fischer, R.L., Goldberg, R.B. and Harada, J.J. (2001) LEAFY COTYLEDON1 is an essential regulator of late embryogenesis and cotyledon identity in Arabidopsis. Plant Cell 6, 1731–1745. Whitacker, M. and Swann, K. (1993) Lighting the fuse at fertilization. Development 117, 1–12.

390

R.A. Bicknell and A.S. Catanach Williams, J.H. and Friedman, W.E. (2002) Identification of diploid endosperm in an early angiosperm lineage. Nature 415, 522–526. ¨ ber parthenogenesis Winkler, H. (1908) U und apogamie im pflanzenreiche. Prog. Rei. Bot. 2, 293–454.

Yang, W.C., Ye, D., Xu, J. and Sundaresan, V. (1999) The SPOROCYTELESS gene of Arabidopsis is required for initiation of sporogenesis and encodes a novel nuclear protein. Genes & Development 13, 2108– 2117.

Index

ABC model of floral development 94, 100–103, 104, 106, 129–131, 238, 300, 301, 346 ectopic expression studies 104 history 117 modification 135 and reproductive organs 105 ABCDE model 130–131 ABCE model 106 abiotic stress 158–159 and pigmentation 200 abscisic acid (ABA) 55, 61, 159–160 and anthocyanin biosynthesis 197 and flower senescence 165 ACC biosynthetic genes and floral senescence 166 synthase and oxidase 161–162, 164 Actinidia deliciosa floral development 256 AGAMOUS (AG) gene 339 age and senescence 151 agriculture and apomixis 356–357, 377–378 allelic-specific DNA sequence information inheritance 258–259 Allium 358 ampeloprasum 5 Alstroemeria floral senescence 152, 154, 155, 165 Amborella trichopoda 127 angiosperms and apomixis 355–356 origin and diversification 124–126 phylogeny 126–129, 128 self-incompatibility mechanism 269–270 ANITA taxa 127, 128–129

acyclic flowers 140 and floral symmetry 140 annual flowering plants 26 summer 42 winter 27–28, 42 Antennaria and apomixis 359, 363 anthers dehiscence 319–322 development 299, 299–300 development genetic control 304–307 and tapetum-specific transcripts 302 anthesis 152–153 anthocyanidins structure 185 anthocyanin biosynthesis 197, 200 increasing 207–208 post-transcriptional mechanisms 199 prevention 207 transcription factors regulation 201–204 transcription factors utilization 210, 214 anthocyanins 178–179, 180, 184 acylation 210 cyanidin derived 188, 189 delphinidin-derived 188, 189, 205, 209 formation 188–189 glycosylation 210 methylation 210 pelargonidin-derived 188, 189, 209 vacuole storage 190 Anthurium pigmentation 199, 203 Antirrhinum hispanicum (Scrophulariaceae) A. hispanicum S-locus F-box gene (AhSLF) 274, 275 gametophytic self-incompatibility 270, 271

391

392 Antirrhinum majus 5, 125, 201 aureusidin synthase (AUS) 186 DEF protein DNA-binding 107 floral symmetry genes 140 genes AGAMOUS (AG) 300 centroradialis (cen) mutant 90, 251, 253 CYC-like 140, 142 CYCLOIDEA (CYC) 253 FLORICAULA (FLO) 238 floricaula mutant 83 manipulated for flowers and reproduction 246 PLENA (PLE) 300 SQUAMOSA (SQUA) 86 Globosa (GLO) protein DNA binding 107 Squamosa (SQA) protein DNA-binding 107 stamen expression studies 302 stamen identity genes 300 stylar ribonuclease (S-RNase) and selfincompatibility 272 transcription factors 70 antisense transgene 206, 207, 284, 303 modifying caratinoid biosynthesis 216 and plant transgene silencing 239–240 apomixis and agriculture 373, 377–378 autonomous 360 biogeographical distribution 357–359 comparative gene expression with sexual plants 374–376 definition 354–355 detection 362–364 endosperm formation 372–374 epigenetic regulation 376–377 as a facultative process 361–362 gametophytic 358, 359, 362 gametophytic and polyploid 365–366 gene identification 367–369 genetic control of native apomicts 364–365 introduction into sexual crops 377–378 maternal 362, 374 and morphological polymorphism 356 paternal 355 and polyploidy promotion 366–367 pseudogamous 361 sporophytic 359, 360–361, 370 synthesis 377–378 ‘synthesis’ in sexual systems 369–374 value to agriculture 356–357 apospory 359, 362, 363, 364, 368 Arabidopsis dyad 370–371 Arabidopsis lyrata self-incompatibility 289

Index self-incompatibility dominance relationship 287 sporophytic incompatibility 279, 282, 287 Arabidopsis thalia 4, 5, 125 ABC genes 130 allelic-specific DNA sequence information inheritance 258–259 anther dehiscence 319–321 anther development 304–307 anthocyanin biosynthesis regulation 204 apomixis genes 373 asymmetry and anther development 304 autonomous flowering pathway 56 autonomous flowering pathway mutants 33 bitegmic ovule 334 callose wall synthesis 310–311, 312, 313 chemically produced mutations 239 critical flowering time 50–51 cytokinin perception 69 cytokinin signalling 61 endosperm formation 373–374 environmental stress flowering pathway 55 ethylene-receptor genes 162 ethylene signal transduction 69 evolution of vernalization response 42 external flowering signals regulation 55–56 external stimuli for flowering 51 floral and inflorescent meristems 82–84 floral-enabling pathways 9, 10–11 floral meristem and stamen specification 299–301 floral meristem identity genes 85–89 floral-promotion pathways 9–11 floral quartet mutants and tetrads 313 floral symmetry 140 flowering pathways 7–8, 15–18 flowering time independent of photoperiod pathway 33 genes (see separate list for genes by name) determining vernalization requirement 30 genes in floral signalling pathways 53 mutations and organ identity 101–102, 102, 105, 109–112 as genetic model of flowering 243–244 gibberellins promoting flowering 57–60 glucose and ABA signalling 157 inflorescent and floral meristems 90–92 inflorescent meristem identity 92–93 integuments initiation and growth genetic regulation 343, 344–345 juvenile-adult phase change 247, 247–249, 257–258 light-dominant plant 12 manipulated for flowers and reproduction 245–246 mechanisms of vernalization 28–29

Index megaspore mother cells (MMCs) 341–343, 375 microsporangium development 305 microsporocyte cytokinesis 309–310 mutant tapetal development 314 mutants and primexine deposition 315 ovule development 335, 336–337 ovule mutant genes 334–335 ovule primordia growth and patterning 340–341 photoperiodic perception pathway 11–20, 62 photoreceptors and floral initiation 51, 52–54 phytochromes 52, 53, 54 pigment regulation 199 placental development 338 ploidy status and methylation 365 post-transitional gene silencing (PTGS) 248–249 proteins FKF1 17 INO 345 MADS-box 63–64 MADS-box transcription factors 64, 69 MALE MEIOCYTE DEATH (MMD) 309 phytochrome-interacting (PIF3) 13 SUP 345 quantitative trait loci 15, 33, 38 repression of phenylpropanoid 210 self-compatibility studies 289 shoot apical meristems 71–72 short-day promotion pathway 69–70 signalling pathways overview 53 stamen development gene research 301–304 stamen expression studies 302 thermosensory flowering pathway 56–57 transcriptional regulator AB13 13 unitegmic mutants 343–344 vegetative to reproductive phase changes 249–251 vernalization 54–55, 55 vernalization requirement 29–34 vernalization response 37–40, 44 as winter annuals 27–28, 32 zinc-finger transcription factors 64 Archaefructus 127–128, 132–133 Armadillo repeat-containing protein 1 (ARC1) 286, 288 Asparagus officinalis 135 asymmetry 304 Asteraceae and apomixis 355, 357 and tapetal development 314 aurones 179, 184 formation 185–186 introduced by genetic modification 208 Austrobaileyaceae 127, 128

393 autogamy (self-fertilization) 358 autonomous pathway to flowering 250, 251 auxins 198 and floral senescence 159–160, 166 Avena sativa 5 B-box genes 250 B class gene expression 131, 132, 134, 135, 136 B-ring hydroxylation 189 alteration by biotechnology 208–209 regulation 203 Bacillus amyloliquefaciens 242 barley see Hordeum basal dicots and ABC model 301 basic Helix-Loop-Helix (vHLH) transcription factors 202 Beta vulgaris and apomixis 375–376 and carotenoids biosynthesis 198 betalain pigments biosynthesis 194–195, 196, 197–198 genetically modified 217 biosynthetic pathways 180, 181–183 description 179 biennial flowering plants 26 bioinformatics 239 biotin 242 bitegmic ovule 333, 334 Blitum capitatum and virgatum 7 blue pigmentation strategies 209, 210 Brachiaria 368 Brachiaria brizantha 375, 376 bract pigmentation 179 Brassica juncea carinata; pekinensis 5 Brassica napus callose wall 311, 312 male meiosis 313 somatic embryogenesis 372 Brassica spp. apomixis and suppressed recombination 368 mutant floral phenotypes 252 pollen coat proteins and stigma development 314 pollen wall patterning 317–318 vernalization-responsive flowering 34 Brassicaceae S-haplotypes 279–280, 282 self-pollen rejection mechanism 280, 284–290 sporophytic self-incompatibility 270, 271 sporophytic self-incompatibility SRKdependent mechanism 287, 288, 289 C class genes 255–256 expression 131, 134

394 Calamagrostis purpurea 366 calcium and delayed senescence 159 in the egg cytosol 371 and membrane catabolic activity 154–155 callose wall 309–311, 310–313 development 315, 318, 319 calmodulin (CaM) protein 286, 288 Cannabis sativa 5 capsanthin-capsorubin synthase 194 carbohydrates and floral senescence 156–157 and pigmentation 197 carotenoid pigments and catabolism 194 description 178 in Lilium tapetum 314 transcriptional control 200 carotenoid pigments biosynthesis 190, 191, 192, 192, 193–194, 198 pathways 180, 181–183 post transcriptional mechanisms 199 and transgene modification 214, 215–216, 217 carpels 101–102, 103, 104, 105 genetic identity promotion 104 structure 138–140 cauliflower mosaic virus 35S (CaMV 35S) 34, 240, 242–243, 244, 249 cell ablation 241–242 cell–cell communication 70–72 Cenchrus ciliaris 366 Centauria cyanus 7 cereals gibberellins and floral process 59–60 vernalization requirement 34–37 vernalization response 41, 44 chalaza 332, 340, 341 homeotic conversion to carpels 346 chalcones 179, 184 formation 185 introduced by genetic modification 208 naringenin structures 185, 187 Chenopodium polyspermum and rubrum 7 Chondrilla and apomixis 358 chromatin 366 changes after vernalization 39 remodelling 55, 64–65, 377 chromoplasts 178 chrysanthemum 209 circadian clock 9, 11–17, 42, 52, 69 oscillator 12–14 and senescence 157–158 Citrus and apomixis 361, 364 flowering control 257

Index Coffea arabica 5 cold acclimation 41, 54–55 collumellae 315, 316, 317–318 colonization and apomixis 358 conifers 125, 126, 131, 137 constitutive photomorphogenesis (COP) system 201 Coprosoma 363 Cortaderia and apomixis 358, 363 cryptochromes 9, 11–14, 14, 15, 52 cucumber see Cucumis sativus Cucumis sativus 5, 7 and carotenoids biosynthesis 198 CYC-like genes 140–142 Cycas circinalis male cone 125 Cyclamen persicum 209 Cymbidium orchid 209 cytokinesis 309 cytokinin 198 and flower senescence timing 159–160 signalling 60–61 cytosol 180, 190 cytosolic Ca2þ 277–278, 279 dark-dominant plants 6, 11 Datura and parthenogenesis 371 Daucus carota somatic embryogenesis 372 day-neutral plants 4, 51 daylength 3 and flowering 56–57 DELLA proteins 58, 59 Delphinium elatum 5 Dianthus carthusianorum 5 Dianthus caryophyllus 179 ethylene signalling 162–163 and senescence 152, 154, 155, 157, 163 Dianthus spp. genetic modification for colour 205 pigmentation 209 dimers and heterodimers 94, 107, 137, 243, 301 diphtheria toxin 242 diplospory 359, 362–363, 368 DNA microarrays 117 dominance relationships, molecular basis 287, 289 dormancy 26 drought and flowering 55 drought-avoidance mechanism 42, 43 ecological niche 3 ectopic expression studies 104 embryo sac, reduced 360 embryo sac, unreduced 359, 362, 370 embryogenesis 376–377 environmental stimulation of flowering 247 Ericaceae 312 Erigeron and apomixis 358, 368 ethylene 55, 347

Index and pigmentation 197 ethylene biosynthesis 156, 158 biocontrol for flower senescence 164 and flower senescence 161–162 in the gynoecium 158 and sucrose application 157 ethylene insensitive plants daylily studies 152, 164–165 and flower senescence 151, 164–165 and soil-borne pathogens 164 ethylene sensitive senescence 151, 160–161 and ABA production 159–160, 162 ethylene signalling and dehiscence 321 and flower senescence 162–163, 165 euangiosperms 128 euAP3 motif in homeotic proteins 136 eudicots 125, 129 petal formation 136 qualitative dimerization 137 eukaryotes non-coding RNA 377 Eustoma grandiflorum 5 evolutionary developmental genetics 129–131 exines 311, 314–315 and ectexine 315 patterning 318, 319 F-actin depolymerization 278 facultative plants see quantitative plants female determinant in self-incompatibility 271–273 S-locus receptor kinase gene 280–283, 284–287 female sterility 347 flavenoid pigments biosynthesis post transcriptional mechanisms 199 flavinol pigments in Lilium tapetum 314 flavone pigments and dihydroflavonols formation 186, 188 regulation 203, 204 flavonoid pigments biosynthesis 184–190 biosynthesis in transgenic plants 205–214 biosynthesis modified by transgenes 212–213, 214 biosynthetic pathways 180, 181–183 biosynthetic scheme 187 description 178–179 gene activity inhibition by sense or antisense RNA 206 novel compound introduction 208 pathway precursors 185 production regulation 200–204 redirecting substrate within the pathway 207–208

395 flavonols biosynthesis reduction 207 biosynthesis regulation 203 and leucoanthocyanidins formation 188 and overexpression of transcription factors 210 floral development biotechnology 237–259 floral diversification 134–136 floral-enabling pathways 9, 10–11 floral homeotic genes expression patterns 134–135 floral homeotic proteins 131, 136–138 floral initiation gene control 7–11 floral meristem cell proliferation and fate 66–69 identity genes 27, 29, 81–94, 133 floral organ-identity genes 100–104, 129, 130, 131 AG expression initiation 110–111 AG post-transcriptional regulation 113–114 AG regulation targets 116 AG transcriptional repression 111–113 AP1 expression regulation 108, 109–110 AP2 translational repression 114–115 class B gene expression initiation 109–110 class B gene expression maintenance 110 class B gene regulation 115–116 expression 107–109, 110–111 floral organogenesis 100–117 genes 103 floral-pathway integrators 8, 29 floral pigmentation genetic modification 205–218 floral-promotion pathways 9–11 floral quartet model 130–131, 137, 238 floral senescence and abiotic stress 158–159 abscised versus persistent petals 152 age related 160–165 age related versus pollination induced 151 biochemical changes 153–155 carbohydrate levels 156–157 and circadian rhythms 156–157 definition 150–151 ethylene insensitive 151–152 ethylene sensitive 151 genotypes 156 hormone influence on timing 159–160 molecular changes 155 physiological changes after anthesis 152–153 and pigmentation 179 pollination-induced 153, 158, 159, 166–167 floral signal pathways integration 63–64 transduction mechanistic aspects 64–72 floral symmetry 140–142 floral transition 26, 27

396 florigen hormone 6, 17, 61–63 flow cytometry 363 flower origin 131–134 flowering time control 27–28, 241 independent of photoperiod pathway 33 Fragaria x ananassa 5 Fuchsia hybrida 5 funiculus 332, 340, 341, 342 growth 346 gametophytic self-incompatibility (GSI) 270, 271–278, 279 S-glycoprotein-dependent mechanism 279, 289–290 S-RNase-dependent mechanism 273–277, 276 genes discovery and manipulation methods 238–243 expression control 64–65 manipulated for flowers and reproduction 245–246 misexpression/overexpression 242–243 post-translational silencing 248 silencing 239–241 silencing mutations for juvenile-adult phase change 248 targets for biotechnology in flowering 243–259 Gentiana lutea and carotenoids biosynthesis 198 genetic bottlenecks and apomixis 358 genetic modification of flower colour 206 using sense transgenes 211 genetic pathways and flowering 7–11 gerbera anthyocyanin production regulation 203, 207 gibberellic acid (GA) hormone 7, 180 and flower senescence 159–160, 165 GA 20 oxidase enzyme and LD plants 10 pathway to flowering 10, 250, 251 and pigmentation 197 as regulator of homeotic gene expression 108 signal transduction 29 signalling floral process 59–60 Glycine max 4, 5 Agate 7 Biloxi 5, 7 male sterility 313 gnetophytes 127 Gnetum gnemon 131, 137 female cone 125 Gossypium 5 davidsonii 7 hirsutum 7

Index Gramineae tapetal secretion 314 grasses and apomixis 361, 363, 364, 367–368, 375 diplospory with pseudogamy 377–378 gibberellins and floral process 59–60 inflorescence 253 and pseudogamy 373 green fluorescent protein (GFP) 62, 71, 205, 287 guanosine triphosphatase activity 72 gymnosperms 125, 140 flowering control 256–258 phylogeny 126–129 gynoecium removal and vase life 158 haplotypes 270 heat shock and delayed senescence 159 Helianthus spp. genes manipulated for flowers and reproduction 246, 253 mutant flowers 253 tuberosus 5 Hemerocallis fulva 5 hermaphroditic cones 132 hermaphroditic flowers 126, 127, 132, 133 self-incompatibility mechanism 269–270 hermaphroditic plants and apomixis 359 heterodimers 137, 243, 275, 301 Hibiscus cannabinus; esculentus 5 Hieracium and apomixis 357, 358, 361, 363, 369, 374–375 endosperm formation 373 and unreduced egg cell usage 370 histidine phosphotransfer proteins 60–61 histone deacetylase complex 33 modifications 38, 40 homeodomain proteins 33 homodimers 137, 243, 301 Hordeum 4, 5 GAMYB gene 58, 59 short day vernalization 43 vernalization requirement genes 35–36, 37 Humulus japonicus; lupulus 5 Hyoscyamus niger 5 Hypericum 358, 368 Illiciaceae 127 Impatiens 209 floral reversal 92 inflorescence and floral organ architecture 251–254 inflorescent meristems 81–82, 84 and floral meristem identity 90–92 identity promotion 92–93

Index integuments 332 evolution and fossil record 333–334 growth 344–345 homeotic conversion to carpels 346 initiation and growth gene regulation 343 Ipomoea anthyocyanin production regulation 202 pigmentation 202, 209 Iris and senescence 155, 165 irradiation and flowering 55 isoflavonoids and overexpression of transcription factors 210 isopentynyl diphosphate (IPP) formation 190, 191 Ixeris and apomixis 359 Jasminium grandiflorum 5 jasmonic acid (JA) 320–321 juvenile-adult vegetative phase change 247–249, 257–258 Kalanchoe¨ blossfeldiana 5 kiwifruit see Actinidia deliciosa laser capture microdissection 117 leaf grafts and flowering 6, 7 Lemna gibba minor 5 leucine-rich receptor kinases and integument growth 344 light and flower development 201 and pigmentation 195, 197, 198 quality 29 and senescence 158 signals, flowering and anther dehiscence 302 light-dependent pathway to flowering 250, 251 light-dominant plants 6, 11–12 light-quality flowering pathways 7, 10 Liliaceae 135, 137 gametophytic self-incompatibility 271 Lilium callase complex 313 pollen wall patterning 317 stamen expression studies 302 tapetal secretion 314 Limonium 358, 363 Linaria vulgaris (Scrophulariaceae) 141 Linum usitatissimum 5 lipids in Lilium tapetum 314 lipoxygenase and senescence 153, 154 lisianthus 209 Lolium temulentum 59–60 Ba 3081 5 Ceres 5 long day plants 4–6, 11–12 crop species list 5 and gibberellin activity 59–60

397 photoperiods 27 long-distance flowering signal transmission 61–63 Lotus japonicus expression studies 302 lycopene 192, 194 Lycopersicon esculentum callose wall 311, 312 and carotenoids biosynthesis 198 genes manipulated for flowers and reproduction 246 SELF PRUNING (SP) 251–252 SEPALLATA (SEP) 105 male meiosis 313 stamen expression studies 301 sympodial growth habit 91–92 M-locus protein kinase (MPLK) gene 286–287 MADS-box gene family 129, 133, 238, 338–339 expression patterns of class ABC 104 in grapes 255 overexpression 243 and ovule development 335 and petal pigmentation 197, 203 transcription factor 85, 90, 92, 93–94, 301 transcriptional regulators 252–253 MADS-domain proteins definition 105–106 dimer binding 137 factor AGL15 164 MIKC-type 137 as transcription factors 106–107 male determinant in self-incompatibility 273–277 S-locus protein 11/S-locus cysteine-rich protein 283–287 male gametophytes development 298–322 mutations 303–304 male organs 132 male sterility 303–304 Medicago spp. 5 and unreduced egg formation 370 megagametogenesis 376 megaspore mother cells (MMCs) 341–343, 359, 363, 375 megasporogenesis 369 meiosis and apomixis 359, 360, 362 avoidance 370–371 male, and cell wall 309–310 regulation 322 in sporogenous cells 307–309 meiosis-mitotic cell division 5454 membrane study during senescence 154–156 meristem identity control 249, 251 meristematic receptor-like kinase (MRLK) 93

398 methylation-induced gene silencing 241 methylation of DNA 64–65 micro RNAs (miRNAs) 241 micropyle 342 microsporangium development 305, 307–311, 313–315 patterning 322 Mohavea and CYC-like genes 142 monocots and ABC model 301 monopodial development 251 morning glory see Ipomoea Mostly Male Theory 133–134 mutagenesis 369 mutation induction as research tool 238–239 MYB 10, 12–13, 201–204 transcription factor 320 myosins 376 N-acetylglucosamine (GlcNAc) 58 Narcissus and senescence 155, 165 Narcissus pseudonarcissus and carotenoids biosynthesis 198 nexine 316 Nicotiana alata gametophytic selfincompatibility 271 Nicotiana sylvestris 5, 7 Nicotiana tabacum callose wall 5, 7, 201, 208, 311, 312 control of inflorescence 252 Maryland Mammoth sp. 4, 5 pollen wall patterning 317–318 stamen expression studies 302 transgenic male meiosis 313 night breaks 6, 11–12 Norflurazon 316 nucellus 332, 357 growth and differentiation 341–342 homeotic conversion to carpels 346 nutrient transport system 6 Nymphaea odorata 125 Nymphaeales 127, 128 obligate plants see qualitative plants oilseed rape see Brassica napus stamen expression studies 302 Opuntia and apomixis 358 Orchidaceae and apomixis 357 endosperm 372–373 organellar differentation 308 orthophosphate 69 Oryza sativa 4, 5 carpel morphology 139 daylength perception 18–20 floral repression in LDs 42 floral symmetry genes 140

Index genes DROOPING LEAF (DL) 139 Early heading date 1 (ehd1) mutant 19–20 Early heading date 1(ehd1) mutant 69–70 Heading date (Hd1 - Hd4) mutants 18–19, 20 Heading date1 (Hd1) mutant 250 ‘Golden rice’ produced by transgenes 214 semi-dwarf rice and SD-1 gene mutation 237–238 as short-day plant 18–20 short-day promotion pathway 69–70 stamen expression studies 302 vernalization requirement 34 ‘out-of-female’ hypothesis 133 ‘out-of-male’ hypothesis 132, 133, 134 ovule development 332–347 development hormonal regulation 347 developmental genes 335, 336–337 formation on sepals 339–340 identity loss 346 molecular evolution 334–335 primordia growth and patterning 340–341 ovule-floral molecular pathways 335 P. grandiflora and carotenoids biosynthesis 198 paleoAP3 motif in homeotic proteins 136 Panicum and apomixis 361, 363, 364 Papaveraceae 5 gametophytic self-incompatibility 270, 271, 277–278, 279 parthenogenesis 360, 362, 364, 369, 370, 371–372, 376 Paspalum notatum 366, 368 pathogen infection and flowering 55 PCR screening as research tool 239 peas see Pisum sativum pelargonium see Puccinia spp. peleoric mutants 141 Pennisetum apomictic ovules 375 and apomixis 363, 364, 366, 367–369 ASGR and retrotransposon association 377 perennials and apomixis 358 perianth 125, 127, 133 Perilla anthyocyanin production regulation 202 phloem sap analysis 63 red 5 petals 101, 102, 104, 105 abscised versus persistent 152 organogenesis 115–116

Index pigmentation 179 senescence 156 Petunia spp. 5, 201, 208 anther removal and pigmentation 197 anthyocyanin production regulation 203 callase complex 313 gametophytic self-incompatibility 271, 272, 274, 276 genes for placenta, ovule and seed development 338–339 male sterility 313 orange flowers produced by transgenes 209 and senescence 153 SEPALLATA (SEP) genes 105 SUP protein and integument 345 tapetal nuclei and DNA levels of 8C 314 Pharbitis nil 5 Phaseolus vulgaris 201 phenylalanine 184 phenylpropanoid biosynthesis 201 phenylpropanoid genes 199, 204 regulation 204 repressed by transcription factors 210 phloem sap analysis 62, 63 phospholipid catabolic activity 153, 154, 155 photoperception 14–15 photoperiodism 11–20, 42, 250 definition 3 flowering pathways 7 physiology 4–7 response 3–4, 250 transmissible signals 6–7 and vernalization 29 phytochromes 9–10, 14, 15, 16, 52 phytoene formation 192 pigment biosynthesis 180–195 enzymes with published DNA sequences 181–183 regulation 195–204 pigmentation control and biotechnology 178–218 Pinus radiata genes 133 Pisum sativum 5, 7, 162 mutant genes and inflorescence 92 tendrils 254, 255 placenta 332, 333 and ovule initiation 338–339 plant development modules 244 plant fertility and pigment production 179–180 plant transgene silencing (PTGS) 240 plasmalemma 153, 310 Poaceae and apomixis 355, 357, 358, 361, 368 Podostemonaceae endosperm 372–373 pollen coat formation 314–315 coat proteins (PCPs) 283

399 development and male gametophic mutations 303–304 mitogen-activated protein kinase (MAPK) 278 mother cells (microsporocytes) 299 phenotypes and stylar ribonuclease (S-RNase) 273, 274–277 pigmentation 179, 204 production and tree reproduction 258 profilin protein 278 wall patterning and development 317–318 wall structure 315–317, 316 pollen tubes 277 self inhibition 273, 276 pollination 150, 154 and flower senescence 153, 158, 159 and pigmentation 197 signals and changes for senescence 166–167 pollinator attraction 218 signalling in orchids 178 polyembryony 366 polyspory 366 Poncirus trifoliata 364 primary sporogenous cells (PSC) 304–307 primexine 315, 317 programmed cell death 151, 321, 340 and self-incompatibility 278 protein–protein interactions 137–138 proteomics 302 Prunus S-locus F-box (SLF) gene 274, 275 pseudogamous apomicts 361 pseudogamy 373 Puccinia spp. 162, 209 qualitative plants 4, 26, 362–363 crop species list 5 heterodimerization 137 quantitative plants 4, 26, 33 crop species list 5 heterodimerization 137 quantitative trait loci (QTLs) 253, 364 RAF-like kinase inhibitor (RKIP) 90 rainfall and flowering 43 Ranunculaceae and apomixis 364 auricomus 367 petaloid organs 135–136 rapid-cycling habit 42 receptor-like kinases (RLKs) 66–69 recombination, suppressed 368–369 reproductive organs 105 Rhododendron 5 rice see Oryza sativa Ricin A toxin 242 Ricunus communis 5

400 RING-finger domains 276 RING-motif proteins 345 RNA-binding and signalling pathways 65–66 root apical meristems 93 Rosa gallica; rugosa 5 Rosa hybrida genetic modification for colour 205 Rosaceae and apomixis 357 gametophytic self-incompatibility 270, 271 and senescence 153 and senescence ethyline signalling 162 R2R3-MYB transcription factors 202 Rubus and apomixis 358 S-genes 271–277, 278, 279–287 S-Glycoprotein-based self-incompatibility 277–278, 279, 289–290 S-haplotypes 279–283 S-locus cystein rich gene (SCR) 280, 283, 286, 288 S-locus F-box/S-haplotype-specific F-box protein 273–277 S-locus glycoproteins gene (SLGs) 280–283 S-locus receptor kinase gene (SRK) 280–283, 284–287 salicyclic acid (SA) 55 Schisandraceae 127 Secale cereale 5 seed pigmentation 204 self-pollen rejection 284–287 self-pollen tubes inhibition 273, 276 sense transgene 206, 207, 208 modifying caratinoid biosynthesis 215–216 regulating flavonoid biosynthetic genes 212–213 RNA 211 sepals 101, 104, 105 converted to carpelloid tissues 330 organogenesis and class C gene expression 134 Sesamum indicum 5 sexual reproduction 150 shoot apical meristem (SAM) 251 and biochemical signalling 57–61 cell proliferation and fate 66–69 conversion to inflorescence meristem 81–82 and cytokinin signalling 60–61 short-chain fatty acids and floral senescence 166 short day plants 4–6, 11–12 crop species list 5 promotion pathway 69–70 vernalization 43 Sicyos angulatus 7 signal transduction regulating floral development 50–72 Silene armeria; coeli-rosa 5 Sinapis alba 60–61

Index Solanaceae gametophytic self-incompatibility 270, 271 stylar ribonuclease (S-RNase) and selfincompatibility 271–273 Solanum tuberosum 5 Sorghum bicolor; halepense 5 Sorghum male sterility 313 sorting nexin (SNX1) protein 286, 288 soybean see Glycine max spermatophytes flower and cone diversity 125 phylogeny 126–129 sporocyte formation 244 sporophytic mutations 303–304 sporophytic self-incompatibility 270 genetic complexity 279–280 SRK-dependent mechanism 280–283, 284–287, 288 sporopollenin 315, 317 SR genes and ethylene regulation 161, 163–164 stamens development and function 298 organogenesis 101–102, 104, 105, 115–116 stigma self-incompatibility mechanism 280, 284–286 streptavidin gene 242 stress and flowering stimulus 55 and pigmentation 200 stylar ribonuclease (S-RNase) female determinant 271–273, 289–290 sugar treatment for longevity 157 sympodial development 251 synangium theory 333 Tagetes erecta and carotenoids biosynthesis 198 tapetum 299, 302, 307, 308, 312, 313, 332 development 314–315 Taraxacum and apomixis 357, 358, 359, 361, 363, 364, 368, 370 targeting induced local lesions (TILLING) 239 tectum 316 formation 318 telome theory 333 temperature and flowering 7, 10, 27, 51, 55, 56–57 and pigmentation 195 and senescence 158–159 and vernalization 43 tendrils, grape and pea 254, 255 teosinte (maize) 253 tetrads 310, 311, 312, 313, 313 tetrasomic inheritance 366 tobacco see Nicotiana tabacum tobacco rattle virus 240 tocopherols 180

Index tomato see Lycopersicon esculentum toxins and prevention of flowering 241–242 transcription factors (TFs) 70–72 and anthocyanin modulation 210 and pigment biosynthesis 199–200, 201–204 profiling techniques 302 trees acceleration of flowering 242–243, 244 biotechnology to control flowering 256–258 juvenile period 238 juvenile to maturity vegetative phase change 247–249, 257–258 prevention of flowering 241–242 Trifolium pratense (English Montmorency) 5 Trimeniaceae 127 Triticum 4, 5 genes manipulated for flowers and reproduction 245 photoperiodic response manipulation 250–251 ‘Salmon’ system and parthenogenesis 371, 375 short day vernalization 43 winter growth in mild temperatures 42 Triticum monococcum MADS-box transcription factor 36 vernalization requirement genes 34–35, 36–37 Tulipa spp. 125, 137, 209 unitegmic mutants 343–344 unreduced egg cell formation 367, 370 UV light 158, 184 vacuoles 178, 179, 180, 184, 210 import of flavonoids 189–190 vegetative to reproductive phase change 249–251 vernalization 7, 9 definition 26 evolution 41–43 and histone H3 deacetylation 65 molecular process dissection 28–29 pathway to flowering 250, 251 requirement 29–37 response 37–41 Vicia faba 5 Viola cornuta 203 light and pigmentation 197 viral transfer and apomixis 356 virus-induced gene silencing (VIGS) 240

401 Vitis vinifera floral development and biotechnology 254–256 genes manipulated for flowers and reproduction 246 water loss during senescence 153 water stress and flowering 51 and pigmentation 195 water-use efficiency and flowering 43 WD-repeat proteins 202 wheat see Triticum white campion stamen expression studies 302 whorl development microbiology 101–107 wilting 153 woody perennial species biotechnology 254–256 Xanthium strumarium 5, 7 xanthophyll pigments formation 192, 193, 194 yeast proteins 105, 107, 113, 137, 276, 311, 365 Zea mays 5 anthyocyanin production regulation 202, 204 genes floral symmetry and domestication 140–141 INDETERMINATE 62 KNOTTED 1 (KN1) 70 manipulated for flowers and reproduction 245–246, 253 manipulation of inflorescence and floral development 253–254 msca1 mutations and anther development 305 mutation and apomixis 370 Viviparous1 (Vp1) 204 introduction of apomixis 356–357, 363 juvenile-adult phase change genes 248–249 signalling pathway for flowering 51 stamen expression studies 302 tapetal nuclei and DNA levels of 16C 314 TEOSINTE locus mutations 237 vernalization requirement 34 Zephyranthes 358 zinc-finger domain 250

Index of Flower Genes

A6, 313 ABERRANT TESTA SHAPE (ATS), 334 ABORTED MICROSPORES (ABM), 303 ACR4, 344 AERIAL ROSETTE1 (ART1), 53 AGAMOUS (AG), 84, 94 C whorl identity and sporocyte formation, 244 expression initiation, 110–111 floral meristem identity, 103 and floral meristem identity, 68 homodimers and stamen identity, 106–107 post-transcriptional regulation, 113–114 stamens and carpal development, 101, 300, 339, 346 transcriptional repression, 111–113 vernalization pathway gene, 53, 53 AGAMOUS LIKE-20 (AGL20), 29, 251 AGAMOUS LIKE-24 (AGL24), 53, 69, 87, 92–93 AINTEGUMENTA (ANT), 103, 112, 340, 343, 345 APETALA1 (AP1) cellular autonomy study, 70–72 floral meristem identity, 29, 53, 62, 85–87, 88–89, 250, 251, 252 grape homologue, 255 MIKC type, 106–107 restricting TLL1 expression, 91 sepal and petal identity, 101, 103 APETALA2 (AP2), 53, 86, 89, 101, 103, 111–112 and integument growth, 345 translational repression, 114–115 APETALA3 (AP3), 300, 301 Class B gene expression maintenance, 110

402

Class B gene petal and stamen development, 101, 103 floral homeotic gene, 53 and floral organ identity, 84, 94 MIKC-type MADS-domain protein, 106–107 and PI activity, 303 Atgs15, 311 BELL1 (BEL1), 340, 341, 346 BELLRINGER (BLR), 103, 111–112 CARPAL FACTORY (CAF ), 113 CAULIFLOWER (CAL), 53, 88–89, 91, 252 cauliflower (cal) mutant, 85–87 CIRCADIAN CLOCK-ASSOCIATED1(CCA1), 12–14 CLAVATA 1-3 (CLV1, CLV2, CLV3), 66–68, 305 CLAVATA1 (CLV1), 343 clavata mutants, 300 CONSTANS (CO), 9, 10, 13, 15–17, 18, 19, 52, 53, 56, 62, 63, 64, 250, 251 CONSTITUTIVELY MORPHOGENETIC1 (COP1), 14, 52 CRABS CLAW (CRC), 103, 104, 138–139 CRYPTOCHROME1 (CRY1) photoreceptor, 14, 15, 52, 54 CRYPTOCHROME2 (CRY2) photoreceptor, 9, 11, 14, 15, 16, 52, 54 CUP SHAPED COTYLEDON (CUC), 115 CURLY LEAF (CLF), 103, 111–113 DEETIOLATED1 (DET1), 14 DEFECTIVE IN ANTHER DEHISCENCE1 (DAD1), 321

Index of Flower Genes

403

DELLA, 58, 59, 108 DICER-LIKE1 (DCL1), 340, 344 dif1.syn1 mutants, 303

INCURVATA1 (ICU1), 111–112 INCURVATA2 (ICU2), 103 INNER NO OUTER (INO), 139, 334

EARLY BOLTING IN SHORT DAYS (EBS) mutation, 55 EARLY FLOWERING5 (EFL5), 34 EARLY FLOWERING IN SHORT DAYS (EFS), 8 EARLY IN SHORT DAYS4 (ESD4), 8, 34, 53 ELF3, 13 ELF4, 13 embryonic flower 1 (emf1) mutant, 89 embryonic flower 2 (emf1) mutant, 89 EXCESS MICROSPOROCYTES1 (EMS1), 304, 305–307 EXTRA SPOROGENOUS CELLS (EXS), 304, 305–307

KANADI (KAN) family, 138 kanadi1 (kan1) mutant, 304 kanadi2 (kan2) mutant, 304

FCA, 33, 53, 56 FDA, 9 Fertilization independent seed (FIS) group, 372, 373, 374–375 fil1 mutant, 304 FILAMENTOUS FLOWER (FIL), 138 FLORICAULA (FLO), 133 FLOWERING LOCUS C (FLC), 9, 54 and the autonomous pathway, 56 floral inhibition, 8, 9, 93–94, 250 and vernalization, 29, 30, 31–33, 37, 38, 53 FLOWERING LOCUS D (FLD), 9, 10, 33, 53, 56 FLOWERING LOCUS K (FLK), 9, 33, 53, 56 FLOWERING LOCUS M (FLM), 93–94 FLOWERING LOCUS T (FLT), 8, 10, 16, 17, 29, 53, 63, 64 FPA, 56 FRIGIDA (FRI), 8, 31, 37, 53, 54 FRIGIDA-LIKE2 (FR2), 53 FRIGIDA-LIKE (FRL), 31 FRIGIDA-LIKE1 (FRL1), 53 FRUITFULL (FUL), 53, 87–89, 255 FUSCA3, 372 FVE, 9, 33, 53, 56 FY, 9, 33, 53, 56 GI, 13 gibberellic acid1 (ga1), 10 gibberellic acid insensitive (gai), 10 HLL, 340 HUA1, 103, 113 HUA2, 103, 113 HUA ENHANCER1 (HEN1), 103, 113–114 HUA ENHANCER2 (HEN2), 103, 113–114 HUA ENHANCER4 (HEN4), 103 HUA ENHANCERS (HEN), 113–114

LATE ELONGATED HYPOCOTYL (LHY), 12–14 LEAFY COTYLEDON 1 AND 2 (LEC1, LEC2), 371–372 LEAFY (LFY) acting on CEN/TFL genes, 252 B class gene expression, 109 cellular autonomy study, 70–72 floral meristem identity, 52, 53, 62, 63, 64, 103, 107–108 floral pathway integrator, 8, 29 GA pathway, 57, 58 and hybrid aspen accelerated flowering, 243, 244 mutant cloning, 238 repressing AGL24 activity, 68, 69, 110–111 leafy (lfy) mutant, 109–110 leafy (lfy) mutant, 82–84, 85–87, 88, 91 LEUNIG (LUG), 103, 111–113 LUMINIDEPENDENS (LD), 33, 53, 56 MADS AFFECTING FLOWERING2 (MAF2), 40 MADS AFFECTING FLOWERING (MAFs), 32–33 MALE STERILITY2 (MS2), 303 MEDEA, 374 MEDICIS, 372, 373–374 MYB, 64 MYB26, 320 NOZZLE (NZZ), 116, 341, 342, 374, 375 NOZZLE/SPOROCYTELESS (NZZSPL), 304–305 PENNYWISE (PNY), 89, 249 PHOTOPERIOD INDEPENDENT EARLY FLOWERING1 (PIE1), 8, 34, 65–66 PHOTOPERIOD INSENSITIVE EARLY FLOWERING1 (PIE1), 53 PHYTOCHROME A (PHYA) photoreceptor, 9–10, 14, 15, 16 PHYTOCROME AND FLOWERING TIME1 (PFT1), 10 PHYTOCROME B (PHYB) photoreceptor, 10, 14, 15 photoreceptor and flavonoid precursor, 201 PHYTOCROME D (PHYD) photoreceptor, 14 PHYTOCROME E (PHYE) photoreceptor, 10, 14, 15 PI, 301–303

404 PICKLE (PKL), 372 PISTILLATA (PI ), 53, 94, 101, 103, 106–107, 110, 300, 301, 303 PLENA (PLE), 300 POUNDFOOLISH (PNF ), 89, 249 SEEDSTICK (STK), 346 SEPALLATA 1-3 (SEP1, SEP2, SEP3), 84, 94, 103, 105, 106–107, 300 SERK, 374 SEUSS (SEU), 103, 111–113, 345 SHATTERPROOF1 (SHP1), 255, 339–340 SHATTERPROOF2 (SHP2), 103, 116, 255 SHORT INTEGUMENTS2 (SIN2) see DICER-LIKE1 (DCL1) SHORT VEGETATIVE PHASE (SVP), 93–94 SPATULA (SPT), 103, 104 SPINDLY (SPY), 57, 58 SPOROCYTELESS (SPL), 63, 103, 116, 374, 375 STERILE APETALA (SAP), 103, 111–112 STYLISH1 (STY1), 345 superman mutant, 300 SUPPRESSOR OF GENE SILENCING2 (SG2), 248 SUPPRESSOR OF GENE SILENCING3 (SG3), 248 SUPPRESSOR OF OVEREXPRESSION OF CONSTANS1 (SOC1) and the autonomous floral signalling pathway, 53, 58 floral-pathway integrator, 8, 10, 16, 93–94, 250, 251 transcription factor, 63, 64, 69

Index of Flower Genes TAPETAL DETERMINANT1 (TAP1), 305–306 TERMINAL FLOWER (TFL), 68 TERMINAL FLOWER1 (TFL1), 87 terminal flower (tfl) mutant, 251 terminal flower1 (tfl1) mutant, 90, 91 terminal flower2 (tfl2) mutation, 55 TES, 309–310 tetraspore (tes)/stud (std) mutants, 309 TIMING OF CAB1 (TOC1), 9, 12–14 TSO1, 344 UNICORN (UCN), 345 UNUSUAL FLORAL ORGANS (UFO), 86, 89, 103, 109–110 VERNALIZATION INDEPENDENCE (VIP), 8, 65–66 VERNALIZATION INDEPENDENCE3 (VIP3), 34, 53 VERNALIZATION1 (VRN1), 9, 53, 54 VERNALIZATION2 (VRN2), 9, 53, 54 VERNALIZATION-INSENSITIVE1 (VIN1), 38, 39, 40 VERNALIZATION-INSENSITIVE2 (VIN2), 39–40 VERNALIZATION-INSENSITIVE3 (VIN3), 9, 38, 40, 53 WUSCHEL (WUS), 68, 84, 103, 341, 343 YABBY family, 104, 138–140 yabby1 (yab1) mutant, 302