SEARCH

SEARCH BY CITATION

Keywords:

  • evolutionary developmental genetics;
  • carpel development;
  • YABBY transcription factor;
  • CRABS CLAW;
  • Eschscholzia californica;
  • California poppy

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

The Arabidopsis transcription factor CRABS CLAW (CRC) is a major determinant of carpel growth and fusion, and, in concert with other redundantly acting genes, of floral meristem termination. Its rice ortholog, however, has additional functions in specifying carpel organ identity. We were interested in understanding the history of gene function modulation of CRC-like genes during angiosperm evolution. Here, we report the identification and functional characterization of EcCRC, the Californica poppy (Eschscholzia californica) CRC ortholog. The downregulation of EcCRC by virus-induced gene silencing (VIGS) produces additional organ whorls that develop exclusively into gynoecia, resulting in a reiteration of the fourth whorl. Additionally, defects in carpel polarity and ovule initiation are apparent, and the observed phenotype is restricted to the gynoecium. Our results further show that the history of CRC-like genes during angiosperm evolution is characterized by gains of function, independent of duplication processes in this gene subfamily. Moreover, our data indicate that the ancestral angiosperm CRC-like gene was involved in floral meristem termination and the promotion of abaxial cell fate in the gynoecium, and that in the lineage leading to Arabidopsis, additional genes have been recruited to adopt some of these functions, resulting in a high degree of redundancy.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

The most important specific character common to all flowering plants is the carpel, which is located in the centre of the flower, and protectively surrounds the ovules (Crane et al., 1995). Most angiosperms develop carpels that are differentiated into the following structures: the ovary, where the seeds develop; the style; and the stigma, which is a specialized region were pollen germination takes place. The carpel may also provide a system for preventing self-fertilization, as a mechanical barrier and through a molecular self-incompatibility system (Dilcher, 2000).The carpel is also generally the last organ to be formed by the floral meristem, which is consumed in the process of carpel development. When fertilization of the ovules has commenced, the carpel differentiates into the fruit that protects the seeds and ensures their dispersal by a vast variety of mechanisms.

The female reproductive structures of the sister group of the angiosperms, the gymnosperms, are comparatively simple, as the seeds develop on a scale, and pollen germination takes place close to, or at, the ovule surface.

One possible reason for the general success of angiosperms, which dominate the terrestrial ecosystems of our planet, is the evolution of the morphological innovation of the carpel. To learn more about the evolution of the carpel will thus help to better understand the emergence and effective radiation of angiosperms. As the fossil record has not yielded any carpel precursors from non-angiosperms, an alternative approach needs to be considered. Functional comparisons of gene networks directing carpel development in widely diverged angiosperm species could eventually unravel a basic set of gene functions necessary to orchestrate carpel development in all angiosperms. Carpel development control genes are being identified in the core eudicot Arabidopsis thaliana and in the monocot rice. However, the morphological differences between rice and Arabidopsis are vast, e.g. the ovules in Arabidopsis develop from secondary meristems within the carpel, whereas the rice ovule develops directly from the floral meristem (Itoh et al., 2005), and additional reference species are required.

The YABBY gene CRABS CLAW (CRC) encodes a putative transcription factor regulating several important aspects of carpel development in the rosid A. thaliana. The YABBY proteins are a small family of plant-specific transcription factors, and are generally expressed abaxially in developing lateral organs. Phylogenetic analysis of YABBY genes suggest that the CRC subfamily represents a single orthologous lineage, without ancient duplications (Lee et al., 2005b). Several mutant alleles have been identified in Arabidopsis and rice, yielding a wealth of functional data from these two highly divergent plant species (Bowman and Smyth, 1999; Yamaguchi et al., 2004; Lee et al., 2005b). CRC is involved in the control of radial and longitudinal growth of the Arabidopsis gynoecium, and also regulates carpel fusion, in part. crc mutants have gynoecia that are shorter and wider than the wild type, and show defects in carpel fusion. CRC is also essential for nectar gland formation in rosids and asterids, and in crc mutants of Arabidopsis, nectary formation is abolished completely (Alvarez and Smyth, 1999, 2002; Bowman and Smyth, 1999; Lee et al., 2005a,b). DROOPING LEAF (DL) is the CRC ortholog from rice, and is necessary for midrib formation in the rice leaf, floral meristem determinacy and carpel organ identity (Yamaguchi et al., 2004). Functional studies have also been carried out in petunia and tobacco via a small set of virus-induced gene silencing (VIGS)-treated plants (Lee et al., 2005b). In addition to these functional studies, information on expression patterns of CRC-like genes is available for additional species, e.g. carpel expression for Aquilegia formosa, Petunia hybrida and Amborella trichopoda, and additional nectary expression for Cleome sparsifolia, Lepidium africanum and Capparis flexuosa (Fourquin et al., 2005; Lee et al., 2005a).

The Arabidopsis CRC gene is also involved in the termination of the floral meristem in the latest stage during gynoecium initiation. A small number of mutant crc gynoecia show more than two carpels, indicating a mild effect of CRC on floral meristem termination. Recently, it was shown that CRC acts in concert with three genes, REBELOTE (RBL), SQUINT (SQN) and ULTRAPETALA1 (ULT1), to control the early and late phase of floral meristem termination. Interestingly, CRC is not expressed in the centre of the flower, where the activity of the floral meristem will cease, which hints at the possibility that CRC itself might not act in a cell-autonomous way (Alvarez and Smyth, 1999; Bowman and Smyth, 1999; Prunet et al., 2008). Several studies so far have shown that the various CRC-like genes from a diverse set of angiosperm species appear to be involved in many important aspects of plant development, including functions in carpel, nectar gland and leaf-blade development, carpel organ identity, and floral meristem determination.

The plant analyzed in this study, Eschscholzia californica Cham. (California poppy) is a representative of a lineage that derived prior to the eudicots, and belongs to the family of Papaveraceae within the Ranunculales. The Ranunculales clade is the earliest diverging eudicot lineage according to recent phylogenies employing molecular markers (Angiosperm Phylogeny Group, 2003), and within the Papaveraceae, Eschscholzia is a rather early diverging genus (Hoot et al., 1997).

In this study, the function of the Eschscholzia ortholog of CRC, EcCRC, was examined in order to deduce the evolutionary ancestral role of the CRC-like genes, and to understand the complex history of neofunctionalization in this gene subfamily. We determined EcCRC expression patterns and used VIGS to transiently knock-down EcCRC function. Based on these observations relative to what is known from other species, we propose that the ancestral functions of CRC-like genes included: (i) the establishment and maintenance of floral meristem determinacy, (ii) specifying abaxial cell fate within the carpel, and (iii) promoting differentiation of carpel marginal tissue. Mapping functional traits of CRC-like genes along phylogenetic trees, we can also infer that the CRC-like genes underwent a series of neofunctionalization events, leading to several divergent gene functions in the monocot and dicot lineages.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Cloning of the Eschscholzia CRC ortholog

3′ and subsequently 5′ RACE PCR cloning was used to amplify sequences homologous with the Arabidopsis and rice CRC and DL genes. Thorough Bayesian phylogenetic analysis was performed based on the nucleotide sequences of high overall quality present in the NCBI database. The potential Eschscholzia CRC ortholog, EcCRC, shows a domain structure typical for YABBY transcription factors. The phylogeny reconstruction presented in Figure 1a shows that EcCRC is the CRC ortholog, and non-stringent Southern blot hybridization (data not shown) demonstrates that it is a single-copy gene. To date, CRC-like YABBY transcription factors have not been identified outside the angiosperms. EcCRC is the only Papaveraceae CRC-like sequence so far, and it clusters robustly within the sequences of two other early diverging eudicot species, A. formosa and Grevillea robusta, being more closely related to the A. formosa sequence. Generally, the topology of our limited sample of CRC sequences is consistent with recent dicot species phylogenies (Soltis et al., 2000), and with the YABBY gene phylogeny of Lee et al. (2005b).

image

Figure 1.  Bayesian phylogenetic tree of angiosperm orthologous CRC-like sequences. (a) Phylogeny reconstruction of all available CRC-like sequences: the Arabidopsis YABBY genes FIL and INNER NO OUTER (INO) were used as the outgroups. The values above the branches denote posterior probabilities, and indicate clade support. (b) Graphic representation of published expression patterns of CRC-like genes (covered in the Results and Discussion sections). Black boxes indicate that CRC-like expression has been experimentally detected. White boxes indicate no expression, whereas the gray boxes indicate that expression patterns have not been recorded. White stars show species lacking nectaries. Abbreviations: L, leaves; N, nectaries; C, carpels; S, seedling apices.

Download figure to PowerPoint

EcCRC is expressed in floral and flower-derived tissues

RT-PCR experiments with cDNA amplified from diverse tissues were carried out to analyze the presence or absence of detectable EcCRC expression in Eschscholzia (Figure 2a). Within flowers at anthesis, EcCRC expression is restricted to the gynoecium, and no transcripts could be detected in sepals, petals, stamens or developing fruits. EcCRC expression was also absent in leaves and green seeds. However, mature seeds expressed EcCRC, which could hint to a function of EcCRC in late embryogenesis or seed maturation. EcCRC is consistently expressed from the earliest stages of flower development, i.e. stages 1–5, from the initiation of the floral meristem formation, when buds are 0–1 mm in diameter, until the buds are 3 mm in diameter, when male meiosis occurs. The expression level of EcCRC decreases in stage 9 (when female meiosis occurs), and is lower in gynoecia at anthesis than in developing buds (Figure 2, staging according to Becker et al., 2005).

image

Figure 2.  Expression of EcCRC in wild-type flowers shown by semi-quantitative RT-PCR and in situ hybridization. (a) RT-PCR-based expression analysis of EcCRC, with Actin analyzed as an endogenous control. Tissues from which the RNA samples were collected are listed above. (b–i) In situ hybridization pattern of EcCRC. (b) Longitudinal section of a bud in stage 5, when all floral organs are initiated. (c, d) Longitudinal sections of a bud in early (c) and late (d) stage 6. (d) shows the region directly adjacent to the placenta. (e) Transverse section of the gynoecium of a stage-6 bud. (f) Longitudinal section of a bud at stage 7. An arrow shows the presumptive placenta region, to the left, and the section shows part of the ovary wall. (g) Transverse section of a stage-7 gynoecium. (h) Transverse section of the gynoecium of a stage-8 bud. (i) Enlargement of the replum region of a stage-8 bud. (j) Schematic overview of a stage-9 Eschscholzia californica gynoecium. The gynoecium is composed of two carpels, one of which has been colour coded: green, ovules; red, placenta; blue, ovary wall with abaxial ridges. Abbreviations: g, gynoecium; ov, ovule; p, petal; pl, placenta; pp, petal primordium; r, replum; se, sepal; sp, stamen primordium; st, stamen. Scale bars: 100 μm.

Download figure to PowerPoint

For a more detailed analysis of the expression pattern of EcCRC, in situ hybridization was performed. The Eschscholzia wild-type gynoecium consists of two fused carpels, which later differentiate into valves connected with a replum that will subsequently allow for fruit opening and seed dispersal (Becker et al., 2005). The strong expression of EcCRC is first detected in stage 5, when the gynoecium initiates, and is observed in all subsequent flower development stages, although it remains restricted to the gynoecium (Figure 2b–i). In stage 6, when the gynoecium starts to elongate, EcCRC expression is found in two distinct domains: (i) in an abaxial domain covering about two-thirds of the gynoecium wall, but not in the most apical and basal regions of the elongating gynoecium wall; (ii) in the centre of the gynoecium base, where the floral meristem cell division has ceased (Figure 2c). In late stage 6, as the gynoecium elongates further, EcCRC expression is no longer confined to the abaxial side, or to the base of the gynoecium, but is present more widely in the region adjacent to the placenta. However, the apicalmost part of the gynoecium still does not show a hybridization signal (Figure 2d). In transverse orientation, a more complex expression pattern is revealed: (i) an even distribution of EcCRC expression over the medial domain of each carpel, (ii) strong expression in two broad strips enclosing the entire placenta region, but no expression can be detected in the central domain of the placenta (Figure 2e). In stage 7 prior to ovule formation, the expression can be found in three distinct domains: (i) the presumptive replum region, where a narrow strip of EcCRC expression can be detected in the abaxial domain; (ii) the region that will later form the ovary wall, which shows weak and uniform expression; and (iii) a few cells in the centre of the gynoecium that continue to express EcCRC (Figure 2f). The horizontal view into an older gynoecium shows that the expression domain of EcCRC is reduced to the central and abaxial domain of the ovary wall. No expression was detected in the developing ovules, and in the few cell layers of the adaxial ovary wall surface. Additionally, the presumptive replum region shows no EcCRC expression (Figure 2g). In a stage-8 bud, EcCRC expression is found exclusively on the abaxial side of the gynoecium. Strong domains of expression occur in the medial and lateral ridges of the ovary wall. However, the EcCRC expression domain continues to exclude the replum regions, placentae and ovules, creating a sharp border between the presumptive replum and the adaxial part of the ovary wall (Figure 2h,i).

EcCRC loss-of-function phenotypes result in reduced longitudinal and radial growth of the fruit, and loss of floral meristem termination

To understand the role of EcCRC in gynoecium development, we used VIGS to obtain a transient knock-down of EcCRC gene expression. We infected 220 poppy plants with a mix of Agrobacteria carrying pTRV1 and pTRV2-EcCRC1, and 38 control plants with Agrobacteria harboring pTRV1 and pTRV2-E. Of the 220 plants inoculated with pTRV2-EcCRC1, 208 survived the inoculation treatment and 177 flowered. Of the 177 plants that produced flowers, 85 (48%) showed various degrees of defects in fruit development. We phenotypically characterized the first three flowers/fruits of each treated plant, where applicable, totaling 495 analyzed flowers/fruits. Previous studies (Wege et al., 2007) indicated that the phenotypic effect decreases progressively in later formed flowers, and our results indicate a similar trend. When observing only the fruits formed first, we found that 47% show an EcCRC-VIGS phenotype, 34% showed wild-type fruits and 19% aborted. Of the fruits formed third, only 16% exhibited an EcCRC-VIGS phenotype, 65% did not show a phenotype and 21% aborted at an early developmental stage. Figure S1 shows the detailed distribution of fruit phenotypes of the first three fruits. Inoculation of poppy plants using the alternative construct pTRV2-EcCRC2 resulted in the same phenotypes with very similar ratios (see Figure S1). All plants treated with pTRV1 and pTRV2-E showed a wild-type phenotype.

We observed varying degrees of abnormal phenotypes in the EcCRC-silenced plants: in all cases restricted to the gynoecium and fruit development (Figure 3a–d). Mild phenotypes (Figure 3b) show an approximately 50% reduction in fruit length as compared with untreated plants, whereas strong phenotypes (Figure 3d) grew only to ∼20% of the wild-type fruit length.

image

Figure 3.  Phenotype of the EcCRC-VIGS plants. (a) Wild-type fruit of Eschscholzia californica (10-cm long, containing 100–120 seeds). (b) A mild EcCRC-VIGS phenotype, showing an apparently normally developed fruit (7.5-cm long) enclosing a second, fully differentiated inner fruit. (c) A severe phenotype of EcCRC-VIGS fruit that is reduced both in length (7.5 cm) and in width, with a highly reduced seed number. Nine seeds are bulging out of the fruit in (c), and only one seed is present in the 1.7-cm long fruit in (d). (e) Transverse section of wild-type fruit showing medial and lateral ridges protruding from the ovary. Black arrows indicate the replum region from which the seeds have been removed. (f) Transverse section of an EcCRC-VIGS fruit with two additional tissue layers within the outer gynoecium (arrows). (g) Longitudinal section of a wild-type gynoecium at stage 7, with developing ovules within the gynoecium. (h) Longitudinal section of an EcCRC-VIGS gynoecium at stage 7, showing an active meristem (indicated by an arrow). (i) Transverse section of a wild-type fruit in the medial ridge region, showing cellulose-fortified cells (green staining), and a single vascular bundle, indicated by an arrow. (j) Transverse section of an EcCRC-VIGS fruit with lignified parenchyma cells (pink staining) and several vascular bundles (arrows). (k) Transverse section of a wild-type gynoecium showing developing ovules extending from the placenta into the cavity of the gynoecium. (l) Transverse section of EcCRC-VIGS gynoecium lacking the proper development of placental tissue and ovules. (m) Transverse sections of a wild-type fruit illustrating the lateral ridges embedding the replum region. The arrows mark a file of heavily lignified cells, presumably involved in valve dehiscence. (n) Transverse section of a mild EcCRC-VIGS phenotype fruit that demonstrates a disrupted differentiation of the replum region. (o) RT-PCR showing the expression of EcCRC in young buds, the negative control plants pTRV2-1 to pTRV2-4 have been treated with pTRV1 and pTRV2-E; plants 1–20 were treated with EcCRC1-VIGS constructs. Plants showing a severe silencing phenotype are marked with ‘S’, those with a mild silencing phenotype are marked with ‘M’, ‘U’ designates the unknown phenotype, and ‘W’ refers to the wild-type-like phenotype. Actin was used to normalize the experiment. Abbreviations: cfc, cellulose-fortified cells; g, gynoecium; ig, inner gynoecium; lpc, lignified parenchymatic cells; lr, lateral ridges; mr, middle ridges; og, outer gynoecium; ov, ovules; p, placental tissue; r, replum; st, stamen; vb, vascular bundle. Scale bars: 100 μm.

Download figure to PowerPoint

All the EcCRC-silenced plants showed a duplication of the fourth floral whorl, resulting in a gynoecium surrounding a second internal gynoecium (Figure 3a,b,e,f). Carpels are initiated at stage 5 of normal Eschscholzia flower development, before the meristematic activity in the center of the flower ceases. In the case of EcCRC-silenced flowers, the carpels initiated correctly at stage 5, but the meristem failed to arrest and continued to produce consecutive carpel whorls (Figure 3f). In several instances a third, and in rare cases even a fourth, gynoecium was observed (data not shown). Later in fruit development, the longitudinal growth and increase in circumference of the inner fruit ruptured the wall of the outer fruit. The additional gynoecia produced viable seeds, albeit less than untreated plants. The normal apical–basal patterning of the fruits was not affected, and carpel fusion was complete. In the more severe phenotypes (Figure 3c, d) the fruit was tightly associated with the seeds, indicating that lateral growth of the fruit was also severely impaired. The number of seeds produced was reduced to a single seed in the most severe cases observed. These most severely affected fruits were also extremely short, growing to a maximum length of 2 cm. A large number of flowers (19%) aborted fruit development (about 10% of the pTRV2-E-treated control plants aborted fruits) even after hand pollination, suggesting that at least some of the most severely affected gynoecia did not develop further into fruits. Transverse sections of the severely affected fruits show that they also contained additional concentric tissue layers adaxial of the inner ovary wall, reminiscent of additional fruits, albeit without any further differentiation (Figure 3e,f). Longitudinal sections show that the center of the developing gynoecium, the floral meristem, continues to produce gynoecia. These inner gynoecia emerge at stage 7 of flower development, and we did not observe additional inner gynoecia in EcCRC-silenced plants at earlier developmental stages (Figure 3g,h).

We also tested if the strength of the observed phenotypes correlated with the degree of reduction in the EcCRC expression levels, by RT-PCR with EcCRC-specific primers on the first floral bud (with a diameter of 1–3 mm), that appeared on a sample of plants, and scored the next fruits to develop. In 175 of the 177 EcCRC-VIGS-treated plants examined in the main sample, both the first and second flowers produced the same phenotype (98.9%, e.g. the first and second flowers show a phenotype, or both flowers show no phenotype), thereby allowing us to analyze gene expression in the first flower, and to assess its phenotype based on the phenotype of the following flowers. All four control plants treated with TRV1 and TRV2-E show a strong expression of EcCRC in the buds (Figure 3o). Of the 20 plants treated with TRV1 and TRV2-EcCRC, all show either a strongly reduced or no expression of EcCRC when compared with TRV2-E-treated plants. All 12 plants that displayed a silencing phenotype observed from the second formed fruit showed no EcCRC expression in their first buds, indicating an inter-relationship between the EcCRC phenotype and the reduction of EcCRC expression. Also, two TRV2-EcCRC-treated plants that still showed expression of EcCRC (Figure 3o, nos 2 and 5) did not show a phenotype in their subsequent development. However, in two more plants (Figure 3o, nos 6 and 20) that did not have a phenotype, EcCRC expression was also absent, suggesting that an EcCRC mRNA concentration below the RT-PCR detection limit is sufficient for proper fruit development. Four more floral buds (Figure 3o, nos 4, 7, 17 and 18) showed no expression, but were the only buds produced, thereby impeding phenotypic assessment.

EcCRC expression is required for the elaboration of the abaxial ovary wall

In the late developmental stages of the flower (stage 9), 10 ridges develop at the abaxial surface of the gynoecium, five for each valve, distributed into three medial and two lateral ridges; the lateral ones are situated next to the replum. The ridges consist of parenchyma cells lined with thick cellulose deposits (collenchymas cells) arranged in an approximately circular manner surrounded by large, irregularly shaped parenchyma cells. The arrangement of collenchyma cells merges adaxially with the vascular bundles, and is abaxially covered with a layer of subepidermal cells (Figure 3i). Strong EcCRC-VIGS phenotypes lack the characteristic ridges completely, and show an irregularly spaced array of large patches of lignified cells; however, the subepidermal and epidermal cell layers are not affected (Figure 3f,j). Instead of collenchyma cells in untreated plants, lignified cell walls are found in EcCRC-silenced fruits. The localization of the vascular bundles is now oriented towards patches of lignified cells, and several vascular bundles are associated with one patch of lignified cells (Figure 3i,j).

The strong EcCRC-VIGS phenotype in Figure 3f shows two additional fruits that emerged as concentric whorls within the outer fruit. These additional whorls appear as two layers of parenchyma cells, without any obvious vascular bundles or cells with specially fortified cell walls. Adaxial/abaxial tissue differentiation, ridge and replum formation is also absent. This suggests that the ectopic inner fourth whorl organs in the strong EcCRC-VIGS phenotypes emerge without adaxial/abaxial and central/lateral polarity.

Taken together, our results indicate that EcCRC is necessary for abaxial ridge formation, proper spacing of vascular bundles and the deposition of cellulose in the specialized parenchyma cells of Eschscholzia fruits.

EcCRC function is required for ovule initiation

In the developing Eschscholzia gynoecia, ovules emerge from two placental tissue strands adaxial to the replum region. The placenta consists of two tissue protrusions, and is covered with a loose array of large club-shaped cells (Figure 3k). The gynoecia of EcCRC-VIGS plants showed a strong reduction in seed set, and in some fruits only one seed was produced (Figure 3c, d). We were interested if the reduced seed set was the result of impaired pollination, or the result of a failure of the gynoecium to produce ovules. Transverse histological sections were made of gynoecia of untreated plants in late stage 8, when the ovules have already initiated (Figure 3k), and relatively mild phenotypes of EcCRC-VIGS (no double gynoecium) plants of the same stage (Figure 3l). These sections show that the characteristic club-shaped cells are present, but that the placental protrusions are reduced. However, ovules are absent in EcCRC-VIGS plants. We were unable to find remnants of aborted ovules in these gynoecia, indicating that ovules, once initiated, will develop fully. Our results indicate that EcCRC function is important for ovule initiation.

Replum formation is impaired in EcCRC-VIGS fruits

In ripening fruits at stage 12, two replum regions differentiate between the lateral ridges of the carpels, allowing explosive dehiscence of the two valves to catapult the seeds away. The replum consists of cells markedly smaller in size than neighboring valve cells, some of which are strongly lignified. The replum region is narrower than the valve region of the ovary wall, and the epidermis is shaped like a W facing in the abaxial direction. Heavily lignified cells are located at the base of this W, emanating from a narrow band of lignified cells towards the adaxial side of the replum, which possibly marks the breaking point between the replum and the valve (Figure 3m). Mild EcCRC-VIGS phenotypes exhibit a narrow region in the ovary, reminiscent of a replum structure (Figure 3n), and are completely lacking in strong phenotypes (Figure 3f). However, transverse sections of mild phenotypes show a lack of lignified cells and a loss of the characteristic W-shaped indentation of the abaxial ovary wall in the replum region (Figure 3n). Even the fruits of the mild phenotypes have lost their explosive valve dehiscence completely, and need to be opened manually. Our results suggest that EcCRC function is necessary throughout gynoecium development, and that EcCRC is involved in a wide variety of developmental processes comprising floral meristem termination, longitudinal and radial growth of the gynoecium, ovule initiation, elaboration of the adaxial ovary wall, and replum formation.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

The molecular mechanisms underlying carpel development have been studied in a number of highly derived species, like the eudicots Arabidopsis and petunia, and in the grass species rice. However, extensive information on organ patterning and tissue differentiation is only available for Arabidopsis (e.g. Sessions et al., 1997; Heisler et al., 2001; Alvarez and Smyth, 2002; Pekker et al., 2005; Sohlberg et al., 2006). The current study aims to dissect the function of one of the key genes involved in carpel development in an evolutionary context. The organization of the Eschscholzia gynoecium is to a large extent similar to that of Arabidopsis (Becker et al., 2005). However, the results presented in this work show that the molecular mechanisms governing gynoecium morphogenesis and ovule initiation between superficially rather similar structures, can follow quite different pathways. Moreover, our analysis reveals a complex history of several gains of gene function during the evolution of CRC-like genes. Figure 4 schematically summarizes our hypothesis about the history of CRC-like gene function acquisition.

image

Figure 4.  Schematic drawing mapping the history of the gene function acquisitions of CRC-like genes. A simplified phylogeny of the major clades of angiosperms, indicating our theory of the latest time point of the proposed gains/losses of CRC-like gene functions during the evolution of flowering plants. The order of the respective symbols on an individual branch does not reflect the order of appearance of the gene function acquisitions/losses. The open circle represents a function in floral meristem termination. An open box indicates the promotion of abaxial cell fate during carpel development. A white star represents a function in the specification of carpel organ identity, and a black star symbolizes a function in leaf midrib formation. A dark-gray circle indicates a function in lateral carpel margin formation; a light-gray circle represents the promotion of longitudinal and/or lateral–medial growth of the gynoecium. Involvement in placenta development and ovule initiation is shown by a black circle. The recruitment of CRC-like genes for directing nectary gland development is represented by a light-gray box. The putative loss of function is indicated by the corresponding symbols that are crossed out.

Download figure to PowerPoint

Specification of gynoecium abaxial cell identity by CRC-like genes

The Eschscholzia ovary wall shows clear differentiation along the adaxial/abaxial axis, with prominent cellulose-fortified ridges bulging out of the gynoecium surface (Figure 3e). A loss of EcCRC function clearly reduces these abaxial ovary wall elaborations, leading to a complete loss of the ridge structure, and a loss of the regularity in the arrangement of vascular bundles associated with these ridges. Interestingly, the reduction of abaxial cell types did not result in an adaxialization of the gynoecium wall, as a proper epidermis is formed, and the large and highly vacuolized cells usually found on the adaxial side of the ovary wall are not found in the abaxial parts of EcCRC-silenced plants (Figure 3f).

All core eudicots for which expression data exist, and the early diverging angiosperm A. trichopoda, show CRC-like gene expression in the abaxial domain of the gynoecium or carpel (Bowman and Smyth, 1999; Fourquin et al., 2005; Lee et al., 2005b). However, the loss of CRC activity in Arabidopsis alone does not lead to obvious adaxial/abaxial polarity defects in the ovary wall, and the cell layers develop normally, but show earlier vascular differentiation (Eshed et al., 1999; Alvarez and Smyth, 2002). Alvarez and Smyth (2002) argue that earlier vascular differentiation and the observed larger cell sizes might reflect a partial loss of carpel identity, and an acquisition of more sepal- or leaf-like characteristics. This could also be the case in Eschscholzia, as EcCRC seems to restrict the number and size of vascular bundles, and organizes its association with abaxial ridge structures.

In contrast to the expression pattern detected in dicots, DL transcripts are not confined to the abaxial side of the carpel at any developmental stage (Yamaguchi et al., 2004). The lack of abaxial/adaxial differentiation in DL expression might represent a gain of adaxial function along the monocot lineage.

The abaxial expression domain of CRC-like genes in dicots and Amborella is in accordance with the function of other YABBY genes that are all involved in the abaxial cell identity of lateral organs (Eshed et al., 2004). As functional data for AmbCRC are not available, we cannot exclude the possibility that no function is assigned to the polar expression pattern of AmbCRC. However, as members of the eudicots also share this expression domain, and as a corresponding function was demonstrated for Arabidopsis, and now for Eschscholzia, it is very likely that one function of the ancestral CRC-like gene is the specification of abaxial cell identity in the gynoecium.

CRC-like genes are involved in floral meristem termination

The EcCRC-VIGS plants show the formation of multiple gynoecia in the centre of the flower nested within each other, reminiscent of Russian matrioshka dolls. This striking phenotype indicates that the activity of the floral meristem is prolonged. The strong crc-1 mutant of Arabidopsis also shows effects in floral meristem termination, albeit only in combination with the ag+/− mutant, but another inner whorl of the gynoecia, however, has not been observed in any of the crc mutant alleles (Alvarez and Smyth, 1999, 2002).

In Arabidopsis, the floral homeotic class-C gene AG is the key regulator responsible for the limitation of stem-cell proliferation in the flower, and ag mutants show an indeterminate appearance among other defects in floral morphology. This is because in later stages of flower development AG, in addition to an unknown factor, represses the transcription of WUSCHEL (WUS), a gene specifying stem-cell identity, which leads to the depletion of the stem-cell population in the floral meristem (Mizukami and Ma, 1997; Lenhard et al., 2001; Lohmann et al., 2001). Recently, it has been shown that three genes, REBELOTE, SQUINT and ULTRAPETALA act as modifiers of CRC action, and combinations of mutants of these genes show extreme defects in floral meristem termination. These effects are partially the result of a reduction in AG expression in the population of cells that is responsible for floral meristem termination (Prunet et al., 2008).

In the monocot rice, the regulation of floral meristem termination involves the class-C gene OsMADS58, as well as the CRC ortholog DROOPING LEAF (DL). Plants that are RNA-silenced for osmads58 show a dramatic loss in floral meristem determinacy, resulting in indeterminate flowers consisting of lodicules, stamens and carpel-like structures. Strong dl mutants, however, also show serious defects in floral meristem determinacy, and produce additional ectopic stamens instead of a central carpel, indicating a role for DL in carpel organ identity (Yamaguchi et al., 2006). Presently, the most parsimonious evolutionary path would indicate that the floral meristem termination function of CRC-like genes evolved once before the split of the monocot and eudicot lineage, and was maintained in both lineages. However, in the lineage leading to Arabidopsis, additional genes have been recruited to act redundantly to CRC, which might indicate a tendency towards a more pronounced homeostasis in the important developmental process of floral meristem termination.

Several functions of CRC-like genes are specific to certain angiosperm lineages

Based on functional studies and expression analysis in a phylogenetic context, it becomes apparent that several functions of CRC-like genes have been recruited in specific lineages only.

(i) Many representatives of early diverging angiosperm (including Aquilegia), monocot and dicot species grow nectaries to ensure maximum pollination success. However, only the core eudicots have recruited CRC-like genes for nectary development. Several rosids and asterids have been tested for expression of CRC-like genes in nectaries, and it has been demonstrated that flower associated nectaries and extrafloral nectaries express CRC-like genes. Nectaries are absent in Arabidopsis, petunia and tobacco if CRC-like genes are downregulated. However, nectary development in species outside the core eudicots is not related to CRC-like gene expression (Lee et al., 2005b). The California poppy does not develop nectaries (Becker et al., 2005).

(ii) CRC-like genes in monocots were recruited for additional functions other than the ones observed in dicots. The dl mutant alleles from rice reveal that the CRC orthologs in at least part of the lineage leading to grasses have gained specific functions not found in dicots or early diverging angiosperms. DL has an important function in the differentiation of the leaf midrib, as dl mutants show a strongly reduced mechanical stability of the leaf, resulting in the ‘drooping leaf’ phenotype.

(iii) The other major function of DL in specifying carpel organ identity is not observed to a similar extent in Eschscholzia, Arabidopsis or tobacco. However, two mutants of other grass species (Pennisetum americanum and Panicum aestivum) have also been reported to exhibit the same phenotype combination as dl (Yamaguchi et al., 2004).

Non-cell autonomous actions of CRC-like genes in carpel margin differentiation

Another feature of the EcCRC-VIGS phenotype related to a loss of adaxial/abaxial polarity, is a reduced seed set, most likely caused by the loss of placental tissue differentiation, entailing disrupted ovule initiation. EcCRC-VIGS plants lack the characteristic outgrowth of the placentae and only produce ovules in low numbers. An additional characteristic of the EcCRC-VIGS phenotype concerning the carpel margins is a severely reduced differentiation of the replum, resulting in fruits that are unable to dehisce. Interestingly, EcCRC expression is absent from the placenta and the replum region (Figure 2c, f, g, h). Moreover, not only adaxial but also abaxial tissue differentiation of the carpel margins is affected in EcCRC-silenced plants (Figure 3f, k, l). The strong Arabidopsis crc-1 mutant shows only a mildly affected replum region, and is apparently capable of normal seed dispersal. Also, placenta development and ovule initiation do not seem to be altered, and the reduced seed set is more likely to result from the reduced longitudinal growth of the pollen tubes in crc-1 gynoecia (Bowman and Smyth, 1999; Alvarez and Smyth, 2002). Thus, the function of EcCRC in carpel marginal tissue development is more pronounced than that of CRC in Arabidopsis, possibly because of the recruitment of redundantly acting genes. Our data demonstrate that EcCRC is necessary for the differentiation of all tissue types originating from the carpel margins, such as placenta, replum and ovules.

Eschscholzia and Arabidopsis are phylogenetically quite distant from each other, and their bicarpellate syncarpous gynoecium architecture is remarkably similar, even though they evolved independently of each other from an apocarpous ancestor (Endress and Igersheim, 1999; Armbruster et al., 2002; Magallon, 2007). However, the functions of CRC and EcCRC are similar to some extent: the development of tissues derived from the carpel margins is promoted in both species. The most parsimonious explanation for the similarities between the Eschscholzia and Arabidopsis CRC-like gene function would be that CRC-like gene function also promotes carpel margin differentiation in the apocarpous ancestral gynoecium of the eudicot lineage.

How EcCRC directs the differentiation of the carpel margins without being expressed there is still to be explained. One possibility is that EcCRC promotes carpel lateral domain identity, and inhibits placenta and ovule formation on the abaxial side of the valve margins. This has been shown for the Arabidopsis CRC in combination with GYMNOS (GYM) or KANADI (KAN). The crc, gym and kan single mutants do not show ectopic ovules on the abaxial side of the ovary wall. However, if crc is combined with either gym or kan, ectopic ovules are observed that develop on the abaxial side of the carpel margins, as a result of the duplication of adaxial tissue types on the abaxial side of the carpel (Eshed et al., 1999). Whether EcCRC also acts in combination with orthologs of GYM or KAN is not known, but could be examined through simultaneous knock-down of EcCRC with orthologs of either gene. Another way by which EcCRC may influence replum and placenta differentiation would be a direct or indirect activation of the genes responsible for placenta and replum identity, possibly by providing the valves with the competence to support medial and lateral tissue formation. This hypothesis could account for the significant reduction in placenta and replum development, as well as ovule number, in EcCRC-VIGS plants.

YABBY gene duplication and functional diversification at the base of the angiosperm lineage could indicate an important role for YABBY genes in angiosperm evolution. In particular, the establishment of CRC-like genes, which are key developmental regulators for the carpel, an autapomorphy of the angiosperms, might have contributed to the evolution of the carpel itself: apparently CRC-like genes are involved in promoting the formation of carpel marginal tissue, including the placenta. If one thinks of the carpel as a modified leaf, as Goethe proposed more than 200 years ago (Goethe, 1790), it is the marginal tissue differentiation supporting the placenta, and subsequently the ovules, that accounts for the major difference between the leaf and the carpel.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Cloning of EcCRCand phylogenetic analysis

The EcCRC gene was isolated using a combination of 3′ and 5′ RACE PCR (Frohmann et al., 1988). Total RNA was isolated from California poppy buds using the RNeasy Plant Kit (Qiagen, http://www.qiagen.com). A 4-μg portion of RNA was reverse transcribed with the Omniscript Kit (Qiagen) using the poly-T anchor primer AB05. A PCR with primers ABCRC06 and the 3′ RACE adapter primer AB07 yielded the 3′ region of the EcCRC coding sequence. We then amplified the missing portion of EcCRC with 5′ RACE using 2 μg of RNA isolated from buds as a template. The first-strand synthesis was performed with the primer EcCRC5R1 using the Omniscript kit (Qiagen), and a poly-A tail was added to the cDNAs using the NEB terminal transferase (New England BioLabs, http://www.neb.com), following the manufacturer’s protocol. Two rounds of nested PCR were performed, using the primers AB05/EcCRC5R2 for the first PCR and the primers AB07/EcCRC5R3 for the second PCR.

The nucleotide sequence has been deposited in the EBI database (acc. no. AM946412). Nucleotide sequences of CRC homologs from various other species were kindly provided by John L. Bowman (Lee et al., 2005b). Deduced amino acids were aligned with m-coffee (Wallace et al., 2006; Moretti et al., 2007), and were manually adjusted using BioEdit (Hall, 1999). Bayesian analysis was performed with MrBAYES 3.1 (Huelsenbeck and Ronquist, 2001; Ronquist and Huelsenbeck, 2003), according to the general-time-reversal model, with a gamma distribution of site substitution rates and a proportion of invariable sites (GTR + G + I), examined by MrMODELTEST 2.2 (Nylander, 2004).

Expression analysis of EcCRC by RT-PCR and in situ hybridization

Total RNA was isolated using the RNeasy Plant Mini Kit (Qiagen), and 1 μg of total RNA was reverse transcribed into cDNA with the SuperScript III Kit (Invitrogen, http://www.invitrogen.com). As an endogenous control for the RT-PCR, the E. californica expressed sequence tag (EST) sequence (NCBI accession: CD476630) closest to the Arabidopsis gene Actin2 was chosen. A total of 35 PCR amplification cycles were used for each RT-PCR, and the expected size of the amplified products was 191 bp for Actin2 (primer combination: actin2RTQfw/actin2RTQrev) and 192 bp for EcCRC (primer combination: eccrcRTQfw/eccrcRTQrev). The primer sequences of this study are listed in Table S1. For the EcCRC-VIGS plants, the total RNA of the very first bud (0–3 mm in diameter) was isolated using the RNeasy Micro Kit (Qiagen). As a negative control, we used the first buds of plants inoculated with pTRV1 and the empty pTRV2 (pTRV2-E).

Non-radioactive in situ hybridization essentially followed the protocol of Groot et al. (2005). The EcCRC coding sequence was cloned into the pDrive vector (Qiagen), and digoxigenin-labeled RNA probes were transcribed using T7 RNA polymerase (Roche, http://www.roche.com). A concentration of 5 μg μl−1 of Proteinase K was used for treating the tissue before hybridization.

Vector construction and plant inoculation

For the VIGS of EcCRC, we amplified a 557-bp fragment containing the major portion of the EcCRC open reading frame with a 3′EcoRI and a 5′BamHI restriction site. The resulting fragment was then cloned into the pTRV2 vector, creating pTRV2-EcCRC1. Additionally, an alternative version of pTRV2-EcCRC1 was produced to exclude the possibility that the observed phenotypes are dependent on the location of the fragment used to silence the EcCRC gene. This second fragment of 445 bp, encompassing the 3′ part of the EcCRC coding sequence and the 3′ untranslated region (UTR), but excluding the 5′ region of the EcCRC coding sequence, was cloned in the same way to produce the vector pTRV2-EcCRC2. pTRV2-E is the empty vector and was used as negative control.

The pTRV2-EcCRC1 and pTRV2-EcCRC2 vectors were transformed separately into Agrobacterium tumefaciens strain GV3101. The infiltration of A. tumefaciens was essentially performed as described previously (Wege et al., 2007), except that 100–150 μl of the combined A. tumefaciens strains suspension, containing pTRV1 and pTRV2-EcCRC1 or pTRV2-EcCRC2 plasmids, was injected into the shoot by inserting the 0.45 × 25-mm needle of a 2-ml syringe vertically into the apicalmost region of 3-week-old plants, taking care not to destroy the shoot apical meristem (SAM). The plants were grown under conditions described previously (Wege et al., 2007), and flowers were cross-pollinated by hand to ensure the maximum possible seed set.

Histology and light microscopy

Fresh buds (> 3 mm in diameter) and fruits (> 3 cm in diameter) of untreated and EcCRC-VIGS plants were fixed in FAE (3% formaldehyde, 5% acetic acid, 60% ethanol) and embedded in Paraplast Plus (Tyco Healthcare, http://www.tyco.com). Microtome sections of 7 μm thickness were stained with Safranin-O (Carl Roth, http://www.carlroth.com) for 24 h and counterstained with alcoholic Fast-Green (Chroma, http://www.chroma.com) solutions for 3 min.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

We are indebted to David R. Smyth, Günter Theißen and Sinead Drea for their helpful comments and discussions on the manuscript. We also thank Werner Vogel and Angelika Trambacz for their help with growing the poppy plants. This work was made possible with funding from the German Research Foundation (DFG) grant: BE 2547/6-1 (to AB).

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Alvarez, J. and Smyth, D.R. (1999) CRABS CLAW and SPATULA, two Arabidopsis genes that control carpel development in parallel with AGAMOUS. Development, 126, 23772386.
  • Alvarez, J. and Smyth, D.R. (2002) CRABS CLAW and SPATULA genes regulate growth and pattern formation during Gynoecium development in Arabidopsis thaliana. Int. J. Plant Sci. 163, 1741.
  • Angiosperm Phylogeny Group (2003) An update of the Angiosperm Phylogeny Group classification for the orders and families of flowering plants: APG II. Bot. J. Linn. Soc. 141, 399436.
  • Armbruster, W.S., Debevec, E.M. and Willson, M.F. (2002) Evolution of syncarpy in angiosperms: theoretical and phylogenetic analyses of the effects of carpel fusion on offspring quantity and quality. J. Evol. Biol. 15, 657672.
  • Becker, A., Gleissberg, S. and Smyth, D.R. (2005) Floral and vegetative morphogenesis in California poppy (Eschscholzia californica CHAM.). Int. J. Plant Sci. 166, 537555.
  • 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, 23872396.
  • Crane, P.R., Friis, E.M. and Pedersen, K.R. (1995) The origin and early diversification of angiosperms. Nature, 374, 2733.
  • Dilcher, D. (2000) Toward a new synthesis: major evolutionary trends in the angiosperm fossil record. Proc. Natl Acad. Sci. 97, 70307036.
  • Endress, P.K. and Igersheim, A. (1999) Gynoecium diversity and systematics of the basal eudicots. Bot. J. Linn. Soc. 130, 305393.
  • Eshed, Y., Baum, S.F. and Bowman, J.L. (1999) Distinct mechanisms promote polarity establishment in Carpels of Arabidopsis. Cell, 99, 199209.
  • Eshed, Y., Izhaki, A., Baum, S.F., Floyd, S.K. and Bowman, J.L. (2004) Asymmetric leaf development and blade expansion in Arabidopsis are mediated by KANADI and YABBY activities. Development, 131, 29973006.
  • 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. Proc. Natl Acad. Sci. 102, 46494654.
  • Frohmann, M., Dush, M. and Martin, G. (1988) Rapid production of full-length cDNAs from rare transcripts: amplification using a single gene-specific oligonucleotide primer. Proc. Natl Acad. Sci. 85, 89989002.
  • Goethe, J.W. (1790) Versuch, die Metamorphose der Pflanzen zu erklären. Gotha: Ettinger.
  • Groot, E.P., Sinha, N. and Gleissberg, S. (2005) Expression patterns of STM-like KNOX and Histone H4 genes in shoot development of the dissected-leaved basal eudicot plants Chelidonium majus and Eschscholzia californica (Papaveraceae). Plant Mol. Biol. 58, 317331.
  • Hall, T.A. (1999) BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucl. Acids. Symp. Ser. 41, 9598.
  • 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, 10891098.
  • Hoot, S.B., Kadereit, J.W., Blattner, F.R., Jork, K.B., Schwarzbach, A.E. and Crane, P.R. (1997) Data Congruence and Phylogeny of the Papaveraceae s.l. Based on four data sets: atpB and rbcL sequences, trnK restriction sites, and morphological characters. Syst. Bot. 22, 575590.
  • Huelsenbeck, J.P. and Ronquist, F. (2001) MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics, 17, 754755.
  • Itoh, J.-I., Nonomura, K.-I., Ikeda, K., Yamaki, S., Inukai, Y., Yamagishi, H., Kitano, H. and Nagato, Y. (2005) Rice plant development: from Zygote to Spikelet. Plant Cell Physiol. 46, 2347.
  • Lee, J.-Y., Baum, S.F., Alvarez, J., Patel, A., Chitwood, D.H. and Bowman, J.L. (2005a) Activation of CRABS CLAW in the Nectaries and Carpels of Arabidopsis. Plant Cell, 17, 2536.
  • Lee, J.-Y., Baum, S.F., Oh, S.-H., Jiang, C.-Z., Chen, J.-C. and Bowman, J.L. (2005b) Recruitment of CRABS CLAW to promote nectary development within the eudicot clade. Development, 132, 50215032.
  • 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, 805814.
  • 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, 793803.
  • Magallon, S. (2007) From fossils to molecules: phylogeny and the core eudicot floral groundplan in Hamamelidoideae (Hamamelidaceae, Saxifragales). Syst. Bot. 32, 317347.
  • Mizukami, Y. and Ma, H. (1997) Determination of Arabidopsis floral meristem identity by AGAMOUS. Plant Cell, 9, 393408.
  • Moretti, S., Armougom, F., Wallace, I.M., Higgins, D.G., Jongeneel, C.V. and Notredame, C. (2007) The M-Coffee web server: a meta-method for computing multiple sequence alignments by combining alternative alignment methods. Nucl. Acids Res. 35, W645648.
  • Nylander, J.A.A. (2004) MrModeltest v2. Program distributed by the author. Evolutionary Biology Centre, Uppsala University.
  • Pekker, I., Alvarez, J.P. and Eshed, Y. (2005) Auxin response factors mediate Arabidopsis organ asymmetry via modulation of KANADI activity. Plant Cell, 17, 28992910.
  • Prunet, N., Morel, P., Thierry, A.-M., Eshed, Y., Bowman, J.L., Negrutiu, I. and Trehin, C. (2008) REBELOTE, SQUINT, and ULTRAPETALA1 function redundantly in the temporal regulation of floral meristem termination in Arabidopsis thaliana. Plant Cell, 20, 901919.
  • Ronquist, F. and Huelsenbeck, J.P. (2003) MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics, 19, 15721574.
  • Sessions, A., Nemhauser, J.L., McColl, A., Roe, J.L., Feldmann, K.A. and Zambryski, P.C. (1997) ETTIN patterns the Arabidopsis floral meristem and reproductive organs. Development, 124, 44814491.
  • Sohlberg, J.J., Myrenas, M., Kuusk, S., Lagercrantz, U., Kowalczyk, M., Sandberg, G. and Sundberg, E. (2006) STY1 regulates auxin homeostasis and affects apical-basal patterning of the Arabidopsis gynoecium. Plant J. 47, 112123.
  • Soltis, D.E., Soltis, P.S., Chase, M.W. et al. (2000) Angiosperm phylogeny inferred from 18S rDNA, rbcL, and atpB sequences. Bot. J. Linn. Soc. 133, 381461.
  • Wallace, I.M., O’Sullivan, O., Higgins, D.G. and Notredame, C. (2006) M-Coffee: combining multiple sequence alignment methods with T-Coffee. Nucl. Acids Res. 34, 16921699.
  • Wege, S., Scholz, A., Gleissberg, S. and Becker, A. (2007) Highly efficient virus-induced gene silencing (VIGS) in California Poppy (Eschscholzia californica): an evaluation of VIGS as a strategy to obtain functional data from non-model plants. Ann Bot, 100, 641649.
  • 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, 500509.
  • Yamaguchi, T., Lee, D.Y., Miyao, A., Hirochika, H., An, G. and Hirano, H.-Y. (2006) Functional diversification of the two C-Class MADS Box Genes OSMADS3 and OSMADS58 in Oryza sativa. Plant Cell, 18, 1528.

Accession number for EcCRC: AM946412.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Figure 1. Percentages of EcCRC-VIGS phenotypes. EcCRC phenotypes obtained by silencing with the pTRV2-EcCRC1 (top, n = 495 flowers) and pTRV2-EcCRC2 (bottom, n = 114 flowers) vectors. Phenotype definitions on the X-axis relate to Figure 3: mild phenotypes correspond to Figure 3b and c, strong phenotypes are shown in Figure 3d. The first flowers of each plant scored are marked in blue, the second in red and the third in yellow.

Table 1. List of oligonucleotide sequences.

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

FilenameFormatSizeDescription
TPJ_3807_sm_fig_S1.pdf21KSupporting info item
TPJ_3807_sm_table_S1.doc33KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.