• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of plntphysLink to Publisher's site
Plant Physiol. Mar 2008; 146(3): 1165–1181.
PMCID: PMC2259068

SEUSS and AINTEGUMENTA Mediate Patterning and Ovule Initiation during Gynoecium Medial Domain Development1,[W][OA]


The Arabidopsis (Arabidopsis thaliana) gynoecium, the female floral reproductive structure, requires the action of genes that specify positional identities during its development to generate an organ competent for seed development and dispersal. Early in gynoecial development, patterning events divide the primordium into distinct domains that will give rise to specific tissues and organs. The medial domain of the gynoecium gives rise to the ovules, and several other structures critical for reproductive competence. Here we report a synergistic genetic interaction between seuss and aintegumenta mutants resulting in a complete loss of ovule initiation and a reduction of the structures derived from the medial domain. We show that patterning events are disrupted early in the development of the seuss aintegumenta gynoecia and we identify PHABULOSA (PHB), REVOLUTA, and CRABS CLAW (CRC) as potential downstream targets of SEUSS (SEU) and AINTEGUMENTA (ANT) regulation. Our genetic data suggest that SEU additionally functions in pathways that are partially redundant and parallel to PHB, CRC, and ANT. Thus, SEU and ANT are part of a complex and robust molecular system that coordinates patterning cues and cellular proliferation along the three positional axes of the developing gynoecium.

During organ development, positional identity information must be coordinated with cellular proliferation to achieve proper organ shape and function. In the Arabidopsis (Arabidopsis thaliana) gynoecium, the female reproductive floral structure, early patterning events divide the gynoecial primordium into distinct zones that distinguish adaxial (inner) versus abaxial (outer), medial versus lateral, and apical versus basal domains (Sessions and Zambryski, 1995; Sessions et al., 1997; Bowman et al., 1999; Ferrandiz et al., 1999; Sessions, 1999; Alvarez and Smyth, 2002). Subsequently, patterns of coordinated cell division and differentiation generate the mature structures that comprise the gynoecium. The Arabidopsis gynoecium is composed of two carpel organs that arise congenitally fused along their margins. The fused margins of the carpels comprise the medial domain of the gynoecium (Fig. 1). A meristematic ridge of tissue, termed the medial ridge (mr in the image), develops along the adaxial (inner) portion of the medial domain. Although clonal analysis data are not available, the patterns of gene expression and cell division, as well as genetic data, strongly suggest that the medial ridge generates the placenta, ovules, septum, and associated transmitting tract, as well as portions of the style and stigma; all of which are critical structures for reproductive competence (Bowman et al., 1999).

Figure 1.
seu ant double mutants display enhanced vegetative, floral, and gynoecial phenotypes. A to C, False-colored photomicrographs. D to K, Photomicrographs of floral/rosette morphology; some sepals and petals have been removed from front half of flowers to ...

Many mutations that affect the development of the medial ridge-derived structures have been identified (Bowman et al., 1999; Ferrandiz et al., 1999; Sessions, 1999; Alvarez and Smyth, 2002; Balanza et al., 2006). However, many of these genes share functional redundancy, and more severe alterations of medial ridge development have been reported in a variety of double mutants. For example, LEUNIG (LUG) and AINTEGUMENTA (ANT) share an important and partially redundant function during the development of the gynoecial medial domain (Liu et al., 2000). The lug and ant single mutants display relatively mild disruptions of the medial domain (Liu and Meyerowitz, 1995; Elliott et al., 1996; Klucher et al., 1996). However, lug ant double-mutant gynoecia lack nearly all of the medial domain and its derived tissues: septum, stigma, ovules, and style (Liu et al., 2000).

ANT encodes a sequence-specific DNA-binding protein expressed early during organ development that functions in organ initiation and potentiates cellular divisions during organ development (Elliott et al., 1996; Klucher et al., 1996; Mizukami and Fischer, 2000; Nole-Wilson and Krizek, 2000). LUG encodes a transcriptional coregulator with sequence similarity to two protein families: the Tup1/Groucho and Ssdp/Chip protein families (Conner and Liu, 2000; van Meyel et al., 2003). Typically transcriptional coregulators do not interact directly with DNA, but rather regulate transcription by physically interacting with sequence-specific DNA-binding proteins (Courey and Jia, 2001; Matthews and Visvader, 2003). The SEUSS (SEU) gene shares a number of functional similarities to LUG during floral organ identity specification (Franks et al., 2002). SEU also encodes a transcriptional coregulator, albeit one with sequence similarity to the LIM-domain-binding protein family. SEU forms a transcriptional regulatory complex through a direct physical interaction with LUG that requires a functionally conserved LisH/LUFS domain found in LUG and the Drosophila protein Chip (van Meyel et al., 2003; Sridhar et al., 2004). The SEU/LUG protein complex is recruited to AGAMOUS (AG) regulatory sequences by the DNA-binding proteins SEPALLATA3 (SEP3) and APETALA1 (AP1) to repress AG in perianth organs (Sridhar et al., 2006).

Although LUG, SEU, and ANT all function in the repression of AG during floral development (Liu and Meyerowitz, 1995; Krizek et al., 2000; Franks et al., 2002) a number of experiments indicate that ectopic AG expression is not responsible for the loss of ovule primordia in the lug ant double mutant (Liu et al., 2000). These results suggest that LUG, SEU, and ANT direct the development of the gynoecium in part through the regulation of an AG-independent pathway(s). Data from petal development suggest that SEU and LUG are required for maintenance of the adaxial and abaxial polarity genes PHABULOSA (PHB) and YABBY1/FILAMENTOUS FLOWER (YAB1/FIL), respectively (Franks et al., 2006). In addition to its role in cellular proliferation, ANT also participates in organ polarity decisions (Nole-Wilson and Krizek, 2006). Together with YAB1/FIL, ANT is required for PHB expression in lateral organ primordia. These data taken together suggest that SEU, LUG, and ANT may regulate organ polarity and/or cellular proliferation during the development of the medial domain of the gynoecium.

Here we describe synergistic genetic interactions between seu and ant mutants. We detail the developmental abnormalities in seu ant double-mutant plants with a focus on patterning events during early gynoecial development. We demonstrate that SEU and ANT provide two partially redundant, but parallel activities required for the development of the medial ridge and subsequent ovule initiation. Our results demonstrate that adaxial/inner identity is compromised early during the development of the gynoecium in seu ant mutants. We identify PHB, REVOLUTA (REV), and CRABS CLAW (CRC) as potential downstream targets of SEU and ANT regulation. In addition to the proposed linear regulatory relationship between SEU and the downstream targets PHB, REV, and CRC, our genetic analysis argues for an additional action of SEU in partially redundant pathway(s) parallel to PHB, CRC, and ANT.


seu ant Mutants Display Synergistic Floral and Vegetative Phenotypes

The seu single mutant has reduced stem length and shorter, rounder leaves (Franks et al., 2002). The seu plants also display multiple floral defects including partial homeotic transformations in perianth organs, narrower floral organs, reduction in stamen height and pollen production, partial splitting of the gynoecial apex, and ovule defects. Additionally, the seu plants display phenotypes associated with a decreased response to auxin, including decreased apical dominance, fewer lateral root primordia, and a reduced sensitivity to exogenous auxin (Pfluger and Zambryski, 2004). The ant single mutants display narrow floral organs, occasionally split carpel tips, ovule defects, slightly narrower leaves with reduced blade on the petiole section, and stamen locule defects (Elliott et al., 1996; Klucher et al., 1996).

To better understand the relationship of SEU and ANT, we generated seuss aintegumenta double-mutant plants. Organ counts in mature flowers revealed a synergistic role for SEU and ANT in the regulation of organ number in whorls 1 and 2 and in the control of organ size in all four whorls. The number of whorl-1 organs was significantly reduced in the seu ant double mutant, relative to wild type (Table I). Nearly all whorl-1 organs developed as narrow sepals (Fig. 2C), and no enhancement of homeotic transformation was detected. This is in contrast to whorl-1 organs in previously described lug ant flowers that typically display homeotic carpelloid transformations (Liu et al., 2000). The number of organs in whorl 2 was also significantly reduced in the seu ant mutant relative to wild type (Table I). In seu-3 ant-1 mutants, whorl-2 organs developed as small radialized filamentous structures that failed to display cellular characteristics of mature petals (Fig. 1I; data not shown). Weaker allelic combinations employing the ant-3 weak hypomorphic allele (Klucher et al., 1996) also displayed synergistic enhancement with respect to organ loss and organ shape and size (Fig. 1H). Scanning electron microscopy (SEM) analysis indicated that reduced numbers of floral organs resulted from early failure to initiate these organs (Fig. 2). In contrast to whorls 1 and 2, the effects of seu and ant on organ number in whorl 3 were additive (i.e. the number of stamens in the seu-3 ant-1 double was similar to that in the ant-1 single). Thus, most of the loss of stamens can be accounted for by the loss of ANT activity. The anthers of ant mutant stamens comprise just two locules compared to four locules in the wild-type anther (data not shown). The anther defect is further enhanced in the seu-3 ant-1 mutant resulting in complete male sterility (Figs. 1, I and J, and 2, I and K). Carpel number was unchanged in all mutant combinations examined (Table I). However, the average length of mature stage-12 gynoecia (stages according to Smyth et al., 1990) from seu-3 ant-1 mutants was statistically shorter than ecotype Columbia of Arabidopsis (Col-0) and either single mutant. The enhancement of carpel morphology observed in the seu ant double mutant is further described in the following section.

Table I.
Morphometric and phenotypic analysis of seu, ant, and seu ant mutant plants
Figure 2.
Early floral and ovule defects in seu ant double mutants. A to O, SEM micrographs. Numbers refer to stages of floral development. P to U, Nomarski contrast images of optically cleared tissue. Scale bars in A to C and H to O, 100 microns; bars in D to ...

The rosette leaves of the seu ant mutants are shorter and narrower than those of either single mutant or wild type (Fig. 1K). These shape changes are sometimes associated with a reduction in abaxial or adaxial fate assignment in the leaf (Waites and Hudson, 1995). To determine if the rosette leaves displayed adaxial or abaxial fate alterations, we examined epidermal cell morphology in the leaf blade by SEM. In wild-type rosette leaves, the leaf blade cells on both the adaxial and abaxial surfaces resemble jigsaw puzzle pieces. The cells of the adaxial surface are larger, more uniform in size, and less lobed than those of the abaxial surface (McConnell and Barton, 1998; Fig. 2, L and M). Additionally, the adaxial blade surface is flatter than the abaxial blade surface; the later is more undulating. In seu-3 ant-1 rosette leaves the adaxial leaf epidermal cells were slightly more lobed, more variable in size, and the leaf surface was more undulating when compared to wild type. These phenotypes suggest a partial loss of adaxial identity in seu-3 ant-1 leaves (Fig. 2O). A similar but weaker effect was observed on the adaxial surface of the seu-3 leaf (Fig. 2N). The edges of seu-3 adaxial leaf cells appeared more lobed and the adaxial surface was more undulating compared to the flat adaxial surface of wild-type leaves. No alterations of epidermal cell morphology were observed on the abaxial leaf surfaces (data not shown). In addition to leaf defects, seu ant double mutants exhibited more severe internode elongation defects than either single mutant resulting in a semidwarf phenotype (data not shown).

seu ant Mutants Display Early Gynoecial Defects and Lack Ovule Primordia

To determine the effect of seu and ant on ovule initiation, we counted ovules from chloral hydrate-cleared stage-10 gynoecia. The average number of ovules observed in the seu-3 single mutants was not statistically different from wild type (Table I). However, the ant-1 single mutants averaged half the wild-type number of ovules. Notably, the seu ant double mutants lack ovules completely. Analysis of the double mutants by thin section analysis, SEM, and chloral hydrate clearing confirmed the complete loss of ovule primordia initiation (Figs. 1, L–O, and and2K).2K). The seu ant double-mutant gynoecia, although split apart toward the apex, were fused normally in basal portions of the gynoecium. In these fused areas abaxial replum and some septal tissue formed (Fig. 1M). Alcian blue staining of cross sections was used to detect transmitting tract cells (Sessions and Zambryski, 1995). In Col-0, ant-1, and seu-3 gynoecia, transmitting tract cells could be detected in all sections examined after stage 12 (Fig. 1L). In the seu-3 ant-1 double mutants, transmitting tract cells could be detected in about 50% of the appropriately staged gynoecia (Fig. 1M). In many cases the extent of transmitting tract was reduced relative to the single mutants or wild type. In the other 50% of instances no transmitting tract was detected. Development of the stigmatic and stylar tissues was reduced in the seu-3 ant-1 double mutant, but was detected (Fig. 1I). In comparison to the gynoecial phenotypes reported for the lug ant double mutant, the seu ant phenotypes are less severe. The seu ant gynoecia nearly always displayed stigma, style, and septum (albiet reduced), whereas lug ant gynoecia did not display stigmatic or stylar tissues, and only exhibited partial septum development when intermediate-strength lug alleles were examined (Liu et al., 2000). Furthermore, the degree carpel fusion is greater in the seu ant gynoecia, with component carpels nearly always fused together in the bottom third of the gynoecium, whereas the lug ant gynoecia are only infrequently fused to this degree. This is unlikely to be due to allele-specific or ecotype-specific effects because seu-1 ant-9 double mutants in the Ler ecotype display a phenotype that is similar to the seu-3 ant-1 Col-0 ecotype plants reported here (R.G. Franks, unpublished data).

Weaker allelic combinations employing the ant-3 hypomorphic allele (Klucher et al., 1996) also had synergistic ovule initiation defects. The average number of ovules in ant-3 plants was not significantly different from the reference ecotype Col-glabrous (Col-gl; Table I). However, the seu-3 ant-3 plants initiated significantly fewer ovules per carpel than either single mutant. The seu-3 ant-3 double mutants also displayed enhanced splitting of the carpels (Fig. 1H). We examined mature, chloral hydrate-cleared seu-3 ant-3 ovules and found no evidence of female gametophyte development in 76% of these ovules (Table I; Fig. 2, R–U). The other 24% of the seu-3 ant-3 ovules displayed incompletely developed female gametophytes that appeared stalled at various stages of development. The ant-3 and seu-3 single-mutant ovules also displayed disrupted female gametophyte development, albeit less frequently than was observed in the double mutant (Table I). The extent of growth of the outer integument in seu-3 and both the outer and inner integuments in ant-3 was reduced relative to wild type (Fig. 2, R–U). This disruption of integument growth was more frequent and more pronounced in the seu-3 ant-3 ovules than in the seu-3 or ant-3 single mutant. However, integument disruptions in the seu-3 ant-3 ovules were less pronounced than in the strong loss-of-function ant-1 single mutant that displays almost no integument development (Klucher et al., 1996).

We examined early stages of floral and gynoecial development in the seu, ant, and seu ant flowers with SEM to determine when morphological defects were first observed. The earliest morphological defect we observed in seu ant mutant floral meristems occurred at floral stage 3. In wild-type stage-3 flowers, four sepal primordia arise on the flanks of the floral meristem in whorl 1 (Fig. 2A). The seu-3 and ant-1 single mutants typically also displayed this morphology (data not shown). However in the seu ant double-mutant flowers, often only two or three sepal primordia were observed suggesting an early defect in organ initiation (Fig. 2, B and C).

Gynoecial development in wild-type flowers initiates at stage 6 as the gynoecial tube arises from the remaining portion of the floral meristem. In all genotypes we examined, including the seu ant double mutant, the gynoecial tube arose normally (Fig. 2, D–G). Shortly after the gynoecial tube forms, the medial ridges are observed as two distinct bulges on the adaxial (inner) surface of the tube in a medial position (Fig. 2D). These medial ridges will give rise to the placental tissue and ovule primordia, as well as septum and transmitting tract tissue. Development of gynoecia from seu or ant single-mutant flowers appears morphologically indistinguishable from wild type at this stage (Fig. 2, E and F). In some seu ant gynoecia the early development of the medial ridge appears wild type. In other cases it appears to be reduced in its extent at this early stage of development, but still morphologically distinguishable (Fig. 2G). During stages 7 and 8, the gynoecial tube begins to elongate. In wild-type gynoecia, as well as in the seu and ant single mutants, this elongation is coordinated with respect to the medial and lateral domains ensuring a fused tube grows up evenly (Fig. 2H; data not shown). In the seu ant double mutants, splitting of the gynoecial apex results from the separation of the medial and lateral domains (Fig. 2I). Growth within the medial domains is retarded relative to the lateral domains and extended horn-like structures develop from the lateral domains in later stages (Fig. 2K).

In mature wild-type gynoecia the vascular bundles extend throughout the basal to apical extent of the ovary within the medial and the lateral domains (Kuusk et al., 2002). The lateral vascular bundle terminates at the distal portion of the ovary at the junction of the ovary and the style. The medial vascular bundle extends into the stylar tissue and terminates in a fan-shaped vascular array. Development of the medial and lateral vascular bundles was similar to wild type in the seu-3 ant-3 double mutant (containing the hypomorphic ant-3 allele; Fig. 2P, arrowhead). However, in the seu-3 ant-1 double mutants, vasculature development in the two medial domains was retarded relative to that in wild-type and vascular bundles consistently terminated within the basal third of the gynoecium with no apparent development of the terminal vascular array (Fig. 2Q, arrowhead). The development of the lateral vascular bundles in the seu-3 ant-1 double mutants was indistinguishable from wild type and terminated in the apex of the lateral horns (Fig. 2Q, arrow; data not shown).

seu ant Gynoecia Fail to Maintain Expression of Adaxial Fate Regulators PHB and REV

PHB, PHAVOLUTA (PHV), and REV are expressed within and are required for the specification of adaxial domains of Arabidopsis lateral organs (Zhong and Ye, 1999; Eshed et al., 2001; McConnell et al., 2001; Otsuga et al., 2001; Prigge et al., 2005). Additionally, they are expressed within adaxial portions of the gynoecium and are likely important for the specification of adaxial fate during gynoecium development (Otsuga et al., 2001; Prigge et al., 2005). As SEU and ANT regulate the expression of the polarity genes YAB1/FIL and PHB in petals and in leaves (Franks et al., 2006; Nole-Wilson and Krizek, 2006), we wondered if SEU and ANT regulate patterning events during gynoecial development. To test this we examined the expression of the adaxial patterning genes PHB and REV by in situ hybridization.

Within the wild-type floral meristem PHB is expressed within the adaxial core of the floral primordia from the preprimordium stage (anlagen) through stage 5 (Fig. 3, A and B; Prigge et al., 2005; Nole-Wilson and Krizek, 2006). At stage 5, the stamen primordia begin to separate from the future gynoecium (gynoecial anlagen), and PHB expression is detected in the core of the gynoecial anlagen as well as within the core region of the stamens (arrows); expression within the adaxial portions of the sepal is also detected (Fig. 3B). During stage 6, PHB expression continues to be detected within the core of the gynoecium and becomes more restricted within the stamen primordia to adaxial portions (Fig. 3, C and D). During stage 7, the expression domain in the gynoecial core resolves to two more lateral domains that likely mark the adaxial portions of the valve domains (Fig. 3E, arrowheads). Expression within the stamens is detected most strongly in an arc that may mark the boundary between adaxial and abaxial domains (Fig. 3E, arrow). The expression of PHB is strongest along this arc within marginal portions of the stamen. At later stages expression is detected in the placenta and early ovule primordia (Fig. 3F) and continues to be expressed in portions of the ovules (Sieber et al., 2004).

Figure 3.
Adaxial fate determinants PHB and REV in seu ant gynoecia. In situ hybridization with PHB antisense (A–M) or REV antisense (N–T) probes. A, D, F, G, J, K, L, and M, Floral longitudinal sections. B, C, E, H, I, and N to T, Floral cross ...

PHB expression within seu mutant flowers was indistinguishable from wild type at all stages examined (data not shown). Expression of PHB within ant mutant stage-3 flowers was detected in a pattern similar to that of stage-3 wild-type flowers (Fig. 3G). However, from stage 5 onward, the expression of PHB was weaker overall in the ant mutant flowers (Fig. 3, H–J). The reduced expression was more severe in the gynoecium than in the stamen primordia in which near wild-type levels of expression were typically detected (Fig. 3I, arrow). This molecular phenotype was not completely penetrant. However, we observed a loss or reduction of PHB signal during stages 5 or 6 in more than half of the ant mutant gynoecia examined. To control for variations in staining from experiment to experiment, we judged expression levels of PHB relative to expression levels in stage-3 flowers from the same in situ hybridization slide. In later stage gynoecia, stage 10 and older, PHB expression was strongly reduced or not detected in the ant-1 gynoecium (Fig. 3J) although occasional exceptions with intermediate levels of staining were noted.

Within the seu ant double mutants, expression of PHB in stages 1 to 4 was indistinguishable from wild type (Fig. 3K). However, again we detected a reduction of expression within the adaxial core of the gynoecium at stages 5 and 6 that was more obvious in the seu ant double mutant than in the ant single mutant (compare Fig. 3L with Fig. 3D). This phenotype was also not fully penetrant at this stage, but was observed in more than half of the gynoecia examined. Expression of PHB was not detected in seu ant gynoecia older than stage 9 (Fig. 3M) except in vasculature tissues (data not shown). These results suggest that SEU and ANT play a role in the maintenance of expression of PHB expression during the transition of the floral meristem into the gynoecial primordia and at later stages of gynoecial development. These alterations in expression pattern, particularly those at stages 5 and 6, preceded the earliest morphological alterations that are observed in seu ant gynoecia, suggesting that the loss of PHB expression is not simply due to a loss of the medial ridge precursors.

REV is also expressed within the adaxial core of early floral primordia, as well as the core of the gynoecial primordia (Otsuga et al., 2001; Prigge et al., 2005). REV continues to be expressed in adaxial portions of the gynoecium during stages 6 through 11 (Fig. 3, N–P). In seu and ant single mutants, expression of REV was indistinguishable from wild type (data not shown). In the seu ant double mutant, REV expression is strongly detected in the adaxial core of gynoecia (stages 6 and 7) at levels similar to wild type (Fig. 3, Q and R). By stages 8 and 9, expression within the adaxial gynoecium appeared reduced relative to wild type, however, weak expression was detected in adaxial portions of the gynoecium suggesting that adaxial fate is partially maintained in these double mutants (Fig. 3, S and T).

Quantitative real-time PCR (qRT-PCR) analysis confirmed the in situ hybridization studies. We examined expression levels from two different RNA sources: inflorescence apex (i.e. inflorescence meristem through floral stage 6) and dissected gynoecia from flowers (stages 8–10; Fig. 3, U and V). In the inflorescence apex, PHB levels in the seu-3 and seu-3 ant-1 genotypes were significantly reduced (P < 0.001) to 68% and 29% of wild-type levels, respectively. REV levels in the inflorescence apex were not significantly different from wild type in the seu or ant single mutants. In the seu-3 ant-1 double mutant, REV levels did display a significant reduction compared to wild type. Consistent with our in situ hybridization data, the magnitude of this reduction was less than that observed for PHB. In the dissected gynoecia (stages 8–10), PHB levels in the seu-3 and ant-1 genotypes displayed a weakly significant (P < 0.05) reduction. The levels of PHB in the seu ant double-mutant gynoecia were further reduced to 12% of wild type (P < 0.001). Levels of REV in the dissected gynoecia also displayed highly significant reductions in the ant-1 and seu-3 ant-1 genotypes.

seu ant Gynoecia Display Altered Expression of Abaxial Fate Determinant CRC

The reduced levels of PHB and REV expression in the seu ant mutant suggest that adaxial identities may be compromised. Because of an antagonistic regulatory interaction between adaxial and abaxial fates in lateral organs, a reduction in adaxial fate is often accompanied by an ectopic expansion of abaxial fate (Bowman et al., 2002). To determine if abaxial fate was expanded in seu ant mutant carpels, we examined the expression of the gynoecial abaxial fate determinants YAB1/FIL and CRC by in situ hybridization. Both CRC and YAB1/FIL are members of the YABBY family of putative transcriptional regulators (Bowman and Smyth, 1999; Chen et al., 1999; Sawa et al., 1999; Siegfried et al., 1999). Genetic analysis indicates that CRC plays a role in the maintenance of abaxial fate in the medial domain of the carpel, a function that is partially redundant with KANADI and GYMNOS activities (Eshed et al., 1999).

In wild-type gynoecia CRC expression initiates during late stage 5 or early stage 6 within the abaxial valve domains (Fig. 4A; Bowman and Smyth, 1999). During late stage 7, expression is detected in a ring that includes the abaxial-most one or two cell layers of both the valve and margin domains (Fig. 4B, arrow). Expression in the abaxial epidermis continues in the margin domain through stage 10 and in the lateral domain into stage 11 (data not shown). During stage 7, CRC is also expressed within an internal expression domain that initiates as two stripes that appear to mark the boundary between the valve and margin domains (Fig. 4B, arrowheads; Bowman and Smyth, 1999). The internal expression domain may be important for ovule initiation or development (Bowman and Smyth, 1999; Alvarez and Smyth, 2002). During stage 8, the internal expression domain consists of the adaxial valve epidermal cell layer (Fig. 4C, asterisk) and four subepidermal foci (Fig. 4C, arrowheads) located close to the sites of ovule initiation. These foci span the boundary between the medial and lateral domains and consist of between 15 and 20 CRC-positive cells in an 8-micron tissue section. A slightly different pattern of CRC is detected in apical portions of the gynoecium (within 10–15 microns of the apex). In the gynoecial apex (stages 7 and 8), CRC is expressed throughout most of the of the valve domain, and in the abaxial medial domain, but is not expressed in the adaxial portions of the medial domain (i.e. the medial ridge; Fig. 4D).

Figure 4.
Abaxial fate determinants YAB1/FIL and CRC in seu ant gynoecia. In situ hybridization gynoecial cross sections. Probes in A to K, CRC; L, ANT; M to S, YAB1/FIL. A to L are at same scale. Scale bar in A, M to N, and Q to S, 0.1 mm; bar in O to P, 0.01 ...

The seu and ant single mutants display wild-type patterns of CRC expression (data not shown). However, alterations of CRC expression were observed in the seu ant double mutants. Expression in stage-6 seu ant gynoecia initiates normally (data not shown). However, in seu ant gynoecia (stages 7 and 8), CRC expression within internal domains was very reduced or absent whereas expression in the abaxial epidermis was still strongly detected. This was most clearly exhibited in the basal third of the gynoecium in which the internal expression domain was often not detected (Fig. 4E). At midgynoecial sections (Fig. 4F, H and J) expression of CRC in the internal expression domain was detected, but expression was weaker and disorganized. The expression of CRC in the adaxial valve epidermis was more consistently detected than expression within the subepidermal foci. Typically from zero to five CRC-positive cells were detected in these foci in the seu ant mutant. Occasionally the extent of the medial ridge was reduced (Fig. 4F, bottom half of the gynoecium) or missing (Fig. 4H, bottom half of the gynoecium). In the apical portion of the seu ant gynoecium, we detected ectopic expression of CRC within the medial ridge. Although CRC is not normally expressed in the apical medial ridge in wild type (Fig. 4D), in the seu ant double mutants CRC was strongly detected in the medial ridge cells at the gynoecial apex (Fig. 4, G and K). These results indicate that CRC expression is differentially affected in the seu ant mutants along the apical/basal axis of the gynoecium. We note that ANT (Fig. 4L) and SEU (data not shown) are expressed in the medial domains at the apex of wild-type gynoecia and thus could be playing a direct role in the repression of CRC in these cells.

YAB1/FIL expression within the developing gynoecium is detected in abaxial subepidermal portions of the lateral domains that will later form the carpel valves (Fig 4M; Sawa et al., 1999; Siegfried et al., 1999). We also detected narrow rays or fingers of weaker expression that extend from the lateral domains into the portions of the medial domain that subtend the ovule primordia (Fig. 4, M, Q, R, and S). Serial cross sections indicated that these fingers were often, but not always, detected subtending sites of ovule primordium initiation. It appears that the cells of the YAB1/FIL-expressing fingers are a subset of the CRC-positive subepidermal foci cells, or closely abut them. YAB1/FIL expression in the seu and ant mutants appears to be unchanged relative to wild type. In the seu ant double-mutant YAB1/FIL is expressed normally in the abaxial valve domain, however, expression within the medial domain fingers was less consistently observed than in wild type (Fig. 4N). We also examined YAB1/FIL expression within the apex of the gynoecium. YAB1/FIL is not detected in the medial ridge at the gynoecial apex in wild-type gynoecia, nor in the seu, ant or seu ant mutants (Fig. 4, O and P). Thus, in contrast to CRC, YAB1/FIL is not ectopically expressed in the apical medial ridge in the seu ant double mutant.

SEU Protein Localizes to the Nucleus and Is Expressed Widely throughout the Plant

Northern analysis indicated that SEU RNA is expressed widely throughout all tissues examined (Franks et al., 2002). Examination of expression data available through the GENEVESTIGATOR site (Zimmermann et al., 2004) also suggest a widespread expression pattern for SEU RNA. SEU appears to play a role in determining expression of downstream target genes within domains of flower (repression of AG in perianth organs, but not reproductive structures; Franks et al., 2002; Sridhar et al., 2004, 2006) and domains of the gynoecium (this study). Thus, we wanted to examine the expression of the SEU protein during floral development to determine if localized expression of the SEU protein contributed to its specificity of action with respect to gene regulation. We generated C-terminal and N-terminal GFP translational fusion genomic rescue constructs: pSEU::GFP:SEU and pSEU::SEU:GFP. Three independent transformants were generated for each of the rescue constructs in a seu-1 mutant background and all rescued the aboveground defects of the seu-1 mutant (root phenotypes were not examined; data not shown). For all transformants, GFP florescence was detected in all tissues examined: root, young leaf, vegetative, and reproductive SAM, floral meristems, and all floral organs (Fig. 5; data not shown). GFP florescence was detected in the nucleoplasm, but was largely absent from the nucleolus and cytoplasm, or was detected at significantly reduced levels in these subcellular regions. Expression in the nucleoplasm is consistent with the predicted nuclear localization signal and the proposed role of SEU as a transcriptional coregulator (Franks et al., 2002). Within the developing floral meristem (Fig. 5B) and gynoecium (Fig. 5C), GFP fluorescence was detected throughout all stages and subdomains. The widespread expression of SEU protein indicates that localized expression of SEU is not responsible for its domain-specific action.

Figure 5.
Expression of genomic SEUSS_GFP rescue construct. All images are from seu-1 mutant plants phenotypically rescued by the pSEU::GFP:SEU fusion construct. A, pSEU::GFP:SEU expression in root division zone is detected in nearly all cells in ...

Genetic Interactions between seu, crc, and ant

Our in situ hybridization results indicate a complex relationship between SEU, ANT, and CRC. To better understand the regulatory relationships of these three genes, we analyzed seu crc double mutants and found that seu and crc displayed synergistic interactions with respect to carpel fusion and development of stigmatic tissues. In comparison to wild-type gynoecia, the crc mutant has a shorter and wider gynoecium that contains fewer ovules (Alvarez and Smyth, 1999, 2002; Fig. 6B). The crc gynoecium is also split within the apical-most third due to a failure of the two component carpels to maintain fusion. The seu single-mutant gynoecium often displays a small amount of splitting at the apex (Franks et al., 2002; Fig. 6C). This split is not between the two component carpels, but rather a split within the style or stigmatic tissue between the two fused marginal domains. The seu crc double-mutant gynoecium displays dramatically enhanced splitting between the two component carpels that extends for approximately 90% of the apical/basal extent of the gynoecium (Fig. 6D). The valves display horn-like extensions at the apex and there is an enhanced loss of stigmatic tissue. This synergistic enhancement between seu and crc is not found in ant crc double mutants that display a largely additive phenotype (Eshed et al., 1999; R.G. Franks, unpublished data).

Figure 6.
The seuss crabs claw double mutants display enhanced gynoecial defects. Photomicrographs of mature gynoecia of indicated genotypes. Scale bars are 1 mm for all panels. A, Landsberg erecta (Ler) reference ecotype. B, Gynoecium of the crabs claw (crc-1 ...

Can the seu ant Double-Mutant Phenotype Be Rescued by Supplemental PHB Activity?

We sought to determine if the reduction of PHB expression in the seu ant double mutant was sufficient to explain the observed ovule loss. To test this, we attempted to rescue the seu ant carpel phenotype by replacing PHB expression with a 35S:PHB construct. A 35S:PHB expression construct (kindly provided by M. Prigge and S. Clark, University of Michigan, Ann Arbor, MI) that is functional to rescue the phb phv cna athb8 mutant (Prigge et al., 2005) was crossed into the seu ant double mutant. Plants expressing the 35S:PHB construct are wild type in appearance, are self-fertile, and display normal ovule initiation and development (McConnell et al., 2001; Prigge et al., 2005) likely due to the targeted degradation of PHB transcripts by microRNAs 165/166 in abaxial domains (Emery et al., 2003). In all observable phenotypes the seu ant 35S:PHB/+ plants were indistinguishable from the seu ant double mutants (data not shown). These results indicate that the 35S:PHB construct is not sufficient for rescue of the carpel defects of the seu ant mutant.

We additionally attempted to rescue the seu ant mutant with a phb-1d semidominant gain-of-function allele of PHB (McConnell and Barton, 1998; McConnell et al., 2001). To generate self-fertile phb-1d heterozygotes, plants were grown at 16°C (McConnell and Barton, 1998). The phb-1d allele generates a PHB transcript that is resistant to microRNA degradation and leads to the partial adaxialization of leaves, sepals, petals, and stamens (McConnell and Barton, 1998; McConnell et al., 2001; Emery et al., 2003). The phb-1d/+ mutant gynoecium also displays signs of adaxialization including the growth of ectopic ovules (adaxial structures) from the abaxial base of the gynoecium (McConnell and Barton, 1998; McConnell et al., 2001). However, adaxially (within the phb-1d/+ gynoecium) the initiation and development of ovules progresses normally, thus supporting female self-fertility. The presence of ovule primordia within the phb-1d/+ gynoecium allowed us to test genetic epistasis of the phb-1d allele with respect to the ovule-less seu ant mutant phenotype. We reasoned that the phb-1d allele might replace lost adaxial activity in the seu ant mutant and thus rescue ovule development. We also reasoned that if SEU and ANT function upstream of PHB transcription that the loss of SEU and ANT might suppress or partially mitigate the phb-1d phenotype in the leaves and petals.

In the rosette leaves of the phb-1d/+ heterozygotes grown at 16°C, the area of the leaf blade is reduced and the petiole is longer relative to wild type (Fig. 7B). The rosette leaf phenotype of the phb-1d/phb-1d homozygote is more severe. These rosette leaves display only small amounts of laminar expansion of the leaf blade, and trumpet-shaped leaves were observed (Fig. 7D). In contrast to our expectations, we observed enhanced radialization and narrowing of rosette leaves in the phb-1d/+ seu ant mutants (Fig. 7C) approaching that exhibited by the phb-1d homozygote. We also observed an enhancement of the phb-1d homozygous phenotype in phb-1d seu double (Fig. 7E) and phb-1d seu ant mutants (Fig. 7F). Thus, the loss of SEU, or SEU and ANT together, caused a further radialization and loss of laminar expansion of rosette leaves in the phb-1d homozygote background. In these plants almost all leaves were completely radialized and rod shaped. These plants also did not produce an observable inflorescence whereas the phb-1d homozygotes did.

Figure 7.
SEU and ANT function in parallel to PHB. A to F, Rosette phenotypes. G to L, Floral phenotypes of indicated genotypes. All plants grown at 16°C. Scale bars, A and B, 3 mm; C, 1.5 mm; D to K, 1 mm; and L, 0.5 mm. A, Wild-type Landsberg erecta (L ...

Floral phenotypes also indicated that the seu and ant mutations enhanced (in petals) or were epistatic to (in gynoecia) the phb-1d mutation, rather than functioning as suppressors. Enhancement in the degree of radialization was observed in the petals of phb-1d/+ seu and phb-1d/+ ant mutants relative to the phb-1d/+ petals (Fig 7, J and K). With respect to the gynoecial phenotype, the seu ant double mutant was epistatic to the phb-1d/+ mutant. The phb-1d/+ seu ant mutant gynoecia were very similar phenotypically to those of the seu ant double mutants in that medial domains formed but failed to generate ovule primordia (Fig. 7L).


ANT regulates organ initiation and organ size in part by maintaining the proliferative potential of organ primordia (Krizek, 1999; Mizukami and Fischer, 2000). More recent analysis of ant fil double mutants indicates that ANT, in concert with YAB1/FIL, is also involved in the regulation of floral organ identity and organ polarity along the abaxial/adaxial axis (Nole-Wilson and Krizek, 2006). SEU together with LUG has also been shown to regulate cell number during organogenesis, as well as organ identity and organ polarity (Franks et al., 2002, 2006). Here we report synergistic interactions between the seu and ant mutants in both vegetative and reproductive tissues. Our results indicate that SEU and ANT share a number of partially redundant functions during Arabidopsis development. The seu ant gynoecia display a complete loss of ovule initiation and a reduction of the gynoecial medial domain. These morphological defects may in part be due to a reduction in adaxial identity within the gynoecial core as well as a reduction of cell proliferation or growth within the medial domain. We identify PHB, REV, and CRC as potential downstream targets of SEU and ANT regulation in the gynoecium. Additionally, we propose that SEU is required to support the activities of genes that function in parallel to ANT, PHB, and CRC during the development of the medial domain of the gynoecium.

Expression of PHB and REV Are Reduced in the seu ant Gynoecium

SEU in concert with LUG has been shown to function in the maintenance of PHB expression in petals (Franks et al., 2006). ANT in concert with YAB1/FIL is required for the expression of PHB in leaves and developing flowers (Nole-Wilson and Krizek, 2006). Our data indicate that the accumulation of PHB RNA in the seu ant mutants is significantly reduced relative to wild type. This effect is most strongly observed in the adaxial core of developing gynoecium during stage 6 and in the adaxial valves during stages 7 through 10. The analysis of REV RNA accumulation indicates that REV RNA is also significantly reduced in the seu ant mutants. REV expression in early stages 6 and 7 appeared at wild-type levels by in situ hybridization whereas at later stages REV expression was slightly, but significantly reduced (Fig. 3). The detection of REV within the adaxial core of the gynoecium through stage 7 indicates that the reduction of PHB levels in the seu ant double are not simply due to a loss of cells in this zone of the gynoecium. Thus we propose that one action of SEU and ANT is to support the expression or accumulation of PHB and REV in the developing gynoecial core and thus potentiate adaxial identity within the gynoecium. Alvarez and Smyth (2002) propose a model in which the expression of adaxial identity genes within the adaxial valve domain is required for development of the placenta within the juxtaposed medial domain. Our data are consistent with this model. However, we report previously undocumented expression of PHB and REV within the adaxial portion of both the medial and lateral domains of the stage-6 gynoecium (Fig. 3). Thus, another interpretation is that placental specification and subsequent ovule initiation requires the expression of adaxial identity genes within the medial domain itself. The molecular identities of SEU and ANT as transcriptional regulators suggest that the regulation of PHB and REV RNA accumulation may be direct and at the transcriptional level, however, we cannot exclude other possibilities. A sequence that matches the ANT binding site consensus in 12 of 14 conserved positions (allowing for a single base-pair gap) is located 1,335 bp upstream of the PHB translation start site (data not shown). However, Nole-Wilson and Krizek (2006) failed to detect significant binding of a bacterial expressed ANT protein to this template in a gel shift assay.

We used two independent methods to replace PHB activity in the seu ant mutant to determine if adding back PHB could rescue the ovule-less phenotype. We found that neither a 35S:PHB overexpression construct, nor the phb-1d allele provided rescue of the seu ant ovule loss (Fig. 7). These results suggest that the reduced expression of PHB is not likely sufficient, by itself, to explain loss of ovule primordia. SEU and ANT likely regulate additional genes during gynoecial development (e.g. REV and CRC) that may contribute to ovule initiation and medial domain development. An alternative, but not mutually exclusive, hypothesis is that SEU and ANT support cell division or growth within the medial domain independently from their role in the regulation of adaxial identity. The observation that seu and ant enhanced the leaf and petal defects of the phb-1d plants indicates a role of SEU and ANT in parallel to PHB in the regulation of laminar growth.

Expression of CRC Is Regulated in Different Directions by SEU and ANT in Apical versus Ovarian Portions of the Gynoecium

SEU and ANT together cooperate to regulate the expression of CRC within the developing gynoecium. However, the effect of SEU and ANT on CRC expression appears to be different in the apex of the gynoecium than it is in the ovary (Fig. 4). Within ovarian portions of the gynoecium, SEU and ANT support the expression of CRC in internal domains. However, within the apex of the gynoecium, SEU and ANT repress the expression of CRC in the medial ridge. Thus, the action of SEU and ANT on CRC expression likely depends on additional factors or may be indirect. In support of the former, SEU shares homology to the LIM-domain-binding protein family of transcriptional coregulators that have been shown to participate directly in both stimulatory and repressive transcriptional events in other systems by binding to a diversity of DNA-binding proteins (van Meyel et al., 2003). One speculation is that SEU and ANT mediate the action of an unidentified regulator of CRC expression that is differentially localized or active along the apical/basal axis of the gynoecium. Because a variety of experiments indicate that auxin signaling provides positional information along this axis (Sessions et al., 1997; Nemhauser et al., 2000; Balanza et al., 2006), it is possible that this proposed factor would be auxin dependent in some fashion.

The proximity of the CRC internal domain expression sites to the sites of ovule initiation, as well as a reduction in ovule initiation in crc mutants, suggest that these internal domains may facilitate ovule initiation (Bowman and Smyth, 1999). The loss of the CRC internal domain expression in the seu ant gynoecia is consistent with these domains being important for ovule development. However, the presence of ovule formation, albeit reduced, in crc null alleles indicates that redundant activities for CRC likely exist. We show that small rays or fingers of YAB1/FIL expression extend into the medial domain and may overlap with the subepidermal foci expressing CRC. Given the sequence similarity between YAB1/FIL and CRC, it is possible that YAB1/FIL provides a partially redundant activity for ovule initiation.

The ectopic expression of CRC within the seu ant mutant indicates that SEU and ANT are required for the repression of CRC in the apical portion of the gynoecium. LUG has previously been shown to function as a negative regulator of CRC in perianth organs and a positive regulator of CRC in the internal gynoecium domains (Bowman and Smyth, 1999). In contrast to the ectopic CRC perianth expression reported for the lug mutant, we did not observe ectopic expression of CRC in the perianth organs of the seu, ant, or seu ant mutants (data not shown). We do not yet know the phenotypic significance of the ectopic CRC expression observed in the apical portions of the seu ant gynoecia. It is possible that this ectopic expression contributes to the apical defects of the seu ant mutant (enhanced loss of stigmatic and style tissues) or that misregulation of CRC at the apex contributes to the loss of ovules in the ovarian portions of the gynoecium. Further genetic analysis will be required to test these possibilities.

Differential Contributions of SEU and LUG to Gynoecial and Floral Development

The seu ant mutants fail to initiate ovule primordia and have reduced growth of other medial-ridge-derived tissues. Still the seu ant defect is less severe than the complete loss of the medial domain reported for the lug ant double (Liu et al., 2000). Thus SEU and ANT may regulate a subset of the genes regulated by LUG and ANT or may be less stringently required for the regulation of these genes. We note that in contrast to the lug ant mutant, the seu ant double mutants seldom display homeotic transformations of perianth organs and we did not observe ectopic expression of AG by in situ hybridization (data not shown). Thus, although SEU and LUG function as a molecular complex, SEU and LUG must make independent contributions to the activity of this complex and may be required to differing extents for the action of this complex during diverse regulatory interactions. The nonequivalence of SEU and LUG is further supported by the strong synergistic enhancement of floral phenotypes reported in the seu lug double mutants (Franks et al., 2002). One molecular explanation for a differential requirement for SEU and LUG is that they may have multiple protein partners and may participate in a number of different complexes that are required for diverse developmental events. A family of SEUSS-LIKE (SLK) genes has been described in Arabidopsis (Franks et al., 2002), and Antirrhinum members of this family have been shown to physically interact with STYLOSA, the Antirrhinum LEU ortholog (Navarro et al., 2004). Thus, the SLK gene family may support SEU-independent activities of LUG in the seu mutant plants.

SEU May Be Required for Functions That Are Redundant with ANT during Ovule Initiation and Medial Domain Development

ANT is expressed throughout the gynoecium in early stages and at highest levels in the adaxial core of the gynoecium before ovule initiation (Elliott et al., 1996). During ovule initiation and early ovule development ANT is strongly expressed in the medial ridge and the developing ovule primordia. Thus, ANT likely provides a key proliferative potential to the ovule primordia early in their development. In ant null allele mutants, however, ovule primordia initiate and continue to develop until the time of integument initiation indicating that ANT activity is not absolutely required for ovule initiation, per se (Elliott et al., 1996; Klucher et al., 1996). The complete loss of ovule initiation in the seu ant double suggests that seu may be required to potentiate the activity of ANT-independent pathways that provide a redundant proliferative or organ initiation activity during early ovule development. There are at least two pathways that might provide these redundant activities: YABBY genes and AINTEGUMENTA-LIKE (AIL) genes.

YABBY Gene Family

The observation that fil ant mutants display a near complete loss of the gynoecial medial domain demonstrates that YAB1/FIL provides an important function that is partially redundant with ANT (Nole-Wilson and Krizek, 2006). The similar disruptions of medial domain development observed in the seu ant and fil ant double mutants suggest that SEU can potentiate the action of ANT in a manner similar to YAB1/FIL. The fil ant flowers also display a loss of floral meristem identity and floral organ identity that is characterized by reduced expression of APETALA3 and ectopic AG expression. These phenotypes are not observed in the seu ant double mutants (data not shown), highlighting a degree of functional differentiation between SEU and YAB1/FIL. Evidence that SEU and LUG may physically cooperate with YAB1/FIL comes from the analysis of orthologous genes in Antirrhinum majus (Navarro et al., 2004). STYLOSA (STY), the Antirrhinum ortholog of LUG, has been shown to interact physically with members of the Antirrhinum YABBY and SLK gene families suggesting that the orthologous Arabidopsis proteins (SEU, YAB1/FIL, CRC, and LUG) may be part of a multimeric complex important for the development of the medial domain of the gynoecium.

Our report of a synergistic genetic interaction in the seu crc double mutant is the first report to our knowledge that demonstrates functional synergy between seu and a member of the YABBY gene family. Although our in situ hybridization results indicate that SEU and ANT work together to promote the expression of CRC within the internal expression domains, the genetic analysis of the seu crc double mutant suggests that SEU and CRC also function in partially redundant, parallel pathways with respect to carpel fusion and stigmatic tissue development.

AIL Gene Family

ANT is a member of a family of transcriptional regulators, the AIL gene family (Nole-Wilson et al., 2005). The expression of several members of the AIL family within the gynoecium overlaps with that of ANT suggesting that these genes may provide some functional redundancy. Interestingly, ail6 ant double mutants display a severe disruption of medial domain development that in some ways resembles that of the fil ant, lug ant, and seu ant double mutants (B. Krizek, unpublished data). One possibility is that AIL6 provides functional redundancy in the ant mutant, where SEU is functional. However, in the seu ant mutant, AIL6 activity may be compromised if its ability to function requires SEU.

A Model for Development of the Gynoecium Medial Domain

We propose a model for the action of SEU, ANT, and LUG during the development of the gynoecial medial domain (Fig. 8). We speculate that the SEU/LUG coregulator complex physically interacts with YAB1 and/or ANT. This multimeric complex can: (1) potentiate the expression of the adaxial fate determinants (PHB and REV) in the adaxial core of stages 6 to 8 developing carpel; (2) support cellular proliferation within the medial ridge and the initiating ovule primordia through a PHB-independent pathway; and (3) engender a position-dependent regulation of CRC expression, supporting CRC expression in the internal expression domains of the gynoecium and repressing CRC expression at the gynoecial apex. These three functions support medial domain development and ovule initiation in wild-type gynoecia. Additional members of the SLK (Franks et al., 2002; Navarro et al., 2004), AIL (Nole-Wilson et al., 2005), YABBY (Siegfried et al., 1999), and LUG/LEUNIG-HOMOLOGUE (Conner and Liu, 2000) gene families are likely to provide molecular redundancy by also participating in this multicomponent molecular complex. Mutant combinations that disrupt at least two of these protein subunits can result in a dramatic loss of the gynoecium medial domain (e.g. lug ant, fil ant, and ail6 ant; Liu et al., 2000; Nole-Wilson and Krizek, 2006; B. Krizek, unpublished data), a loss of ovules from the medial domain (e.g. seu ant; this study), or a near complete deletion of the entire gynoecium (e.g. seu lug; Franks et al., 2002).

Figure 8.
A model for the action of SEU, LUG, ANT, and YAB1/FIL in ovule initiation and gynoecium medial domain development SEU physically interacts with LUG to form a transcriptional coregulator complex. In wild-type plants the SEU/LUG coregulator complex works ...


In Situ Hybridization

In situ hybridizations were carried out as previously reported (Franks et al., 2002). A detailed in situ hybridization protocol is available at http://www4.ncsu.edu/~rgfranks/protocols.html. In situ hybridization probes were in vitro transcribed from the following plasmids: ANT, p5delta4; CRC, pCRII_CRCc1; PHB, pPHB-AS (gift of J. Bowman, Monash University, Victoria, Australia); REV, pCRII_REVc4; YAB1/FIL, pY1-Y (gift of J. Bowman).


SEM was carried out as previously reported (Franks et al., 2006) except that the images were collected on a JEOL 5900LV. Chloral hydrate clearing (Berleth and Jurgens, 1993) and Alcian blue staining (Sessions and Zambryski, 1995) were performed as described. In situ, Nomarski, and Alcian-blue-stained samples were imaged on an Axioscop2 microscope (Zeiss) and captured with a MicroPublisher 5.0 RTV digital camera and QCapture software (QImaging). Stereoscope images were collected with the same camera on a MX12.5 stereoscope (Leica). Confocal images were collected on a IMBE inverted confocal microscope (Leica) at standard GFP settings.

Tissue Dissection, RNA Extraction, and qRT-PCR

Whole inflorescences were collected in 100% ice cold ethanol, fixed overnight at 4°C, and then dissected under a MX12.5 stereomicroscope (Leica) while submersed in 100% ethanol. Tissue was stored in 100% ethanol at 4°C or −20°C for up to 1 week while appropriate development stages were dissected. Ethanol was removed and tissue frozen in liquid nitrogen and then ground. RNA extraction was performed using TRI reagent (Molecular Research Center) according to manufacturer's instructions. The cDNA synthesis was performed using the SuperScript first-strand synthesis system for qRT-PCR (Invitrogen) with a oligo(dT)-primer, according to the manufacturer's instructions. The qRT-PCR was performed on an ABI 7900 using the QuantiTect SYBR Green PCR kit (QIAGEN). In a single PCR run, all samples were done in triplicate, averaged, and normalized using the levels of ADENOSINE PHOSPHORIBOSYL TRANSFERASE gene (APT; At1g27450); APT expression levels of the parental ecotype (Col) were set to a relative level of 1.0. Expression levels of PHB and REV were displayed as a fraction of the wild-type (Col) value. Dissociation curve analysis was performed at the end of each run to ensure the specificity of each reaction. Figure 3, U and V, shows the average relative expression of six PCR runs (three from each of two biological replicates). A Student's t test for significance was performed on these six normalized averages. Primers used for qRT-PCR analysis are as follows: for APT, RTAPT-3 and RTAPT-4; for PHB, RTPHB-1 and RTPHB-2; for REV, RTREV-1 and RTREV-2. Sequences of these primers are reported in the supplemental data.

Genetic Analysis and Plant Growth

All of the alleles used in this study have been previously described: ant-1 and ant-3 (Klucher et al., 1996) were backcrossed into Col-0 three and seven times, respectively, before our genetics analysis; ant-9 (Elliott et al., 1996); crc-1 (Alvarez and Smyth, 1999); phb-1d (McConnell and Barton, 1998); seu-1 (Franks et al., 2002); seu-3 (Pfluger and Zambryski, 2004). All genotypes were confirmed by PCR-based markers. Details for these markers are provided in the supplemental data. Plants were grown in growth chambers at 22°C to 24° C with a 16-h/8-h photoperiod under fluorescent light at 80 to 150 μmol m−2 s−1.

Generation of seu ant 35S:PHB Plants

Seeds containing the 35S:PHB construct in a phb-12 phv-11 cna-2 athb8-12 quadruple mutant background were obtained from Dr. Steven Clark (Prigge et al., 2005). 35S:PHB phb-12 phv-11 cna-2 athb8-12 plants were crossed to seu-3. F1 progeny were then crossed to ant-1. F2 progeny segregating both the seu-3 and ant-1 alleles were verified by PCR genotyping. This and subsequent generations were used for phenotype analysis. At each step, plants containing the 35S::PHB transgene were identified by selection on plates containing Basta prior to growth in soil. Loss of the original quadruple mutant background was also verified by PCR genotyping as previously described (Prigge et al., 2005).

Generation of phb-1d/+ seu-3 ant-1 Plants

phb-1d/+ (heterozygous) seeds were obtained from the Arabidopsis Biological Resource Center (stock CS3761) and grown at 16°C. phb-1d/+ plants were crossed to ant-1. F1 progeny with a phb-1d/+ ant-1/+ genotype were then crossed to seu-3. phb-1d/+ seu-3 ant-1 plants from the F2 and subsequent generations were used for phenotype analysis. All genotypes were confirmed by PCR-based markers. Details for these markers are provided in the supplemental data.

pSEU::SEU:GFP and pSEU::GFP:SEU Constructs

For details of the construction of pSEU::SEU:GFP and pSEU::GFP:SEU constructs, see the supplemental data. Three independent transformant lines for both pSEU::SEU:GFP and pSEU::GFP:SEU were obtained by the simplified Argobacterium floral dip method (Clough and Bent, 1998). All six of these lines rescued all of the aboveground seu-1 mutant phenotypes (data not shown). Expression from these six transformant lines was consistent between the individual lines. Transformant plants were examined in the T2 and T3 generations.

Supplemental Data

The following materials are available in the online version of this article.

  • Supplemental Figure S1. Details of construction of pSEU::SEU:GFP and pSEU::GFP:SEU.
  • Supplemental Figure S2. Genotyping for mutant alleles used in this study.
  • Supplemental Table S1. Oligonucleotides used in this study.

Supplementary Material

[Supplemental Data]


We thank Z. Liu and R. Fischer for support of this project from its inception; J. Pfluger, G. Drews, and C. Grossinger for technical advice; M. Prigge, S. Clark, and J. Bowman for sharing reagents and seeds; B. Krizek for sharing unpublished results; Z. Liu, J. Reed, J. Mahaffey, and anonymous reviewers for helpful comments on the manuscript; the Arabidopsis Biological Resource Center for distribution of seeds; and the North Carolina State University Center for Electron Microscopy and Cellular and Molecular Imaging Facility.


1This work was supported by the National Research Initiative of the U.S. Department of Agriculture Cooperative State Research, Education, and Extension Service (grant no. 2006–03378 to S.N.W.), the National Science Foundation (grant no. IOB–0416759 to R.G.F.), and the U.S. Department of Agriculture Agricultural Research Service (NC06759).

The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantphysiol.org) is: Robert G. Franks (ude.uscn@sknarfgr).

[W]The online version of this article contains Web-only data.

[OA]Open Access articles can be viewed online without a subscription.



  • Alvarez J, Smyth DR (1999) CRABS CLAW and SPATULA, two Arabidopsis genes that control carpel development in parallel with AGAMOUS. Development 126 2377–2386 [PubMed]
  • Alvarez J, Smyth DR (2002) CRABS CLAW and SPATULA genes regulate growth and pattern formation during gynoecium development in Arabidopsis thaliana. Int J Plant Sci 163 17–41
  • Balanza V, Navarrete M, Trigueros M, Ferrandiz C (2006) Patterning the female side of Arabidopsis: the importance of hormones. J Exp Bot 57 3457–3469 [PubMed]
  • Berleth T, Jurgens G (1993) The role of monopteros in organizing the basal body region of the Arabidopsis embryo. Development 118 575–587
  • Bowman JL, Baum SF, Eshed Y, Putterill J, Alvarez J (1999) Molecular genetics of gynoecium development in Arabidopsis. Curr Top Dev Biol 45 155–205 [PubMed]
  • Bowman JL, Eshed Y, Baum SF (2002) Establishment of polarity in angiosperm lateral organs. Trends Genet 18 134–141 [PubMed]
  • Bowman JL, Smyth DR (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 [PubMed]
  • Chen Q, Atkinson A, Otsuga D, Christensen T, Reynolds L, Drews GN (1999) The Arabidopsis FILAMENTOUS FLOWER gene is required for flower formation. Development 126 2715–2726 [PubMed]
  • Clough SJ, Bent AF (1998) Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. The Plant Journal 16 735–743 [PubMed]
  • Conner J, Liu Z (2000) LEUNIG, a putative transcriptional corepressor that regulates AGAMOUS expression during flower development. Proc Natl Acad Sci USA 97 12902–12907 [PMC free article] [PubMed]
  • Courey AJ, Jia S (2001) Transcriptional repression: the long and the short of it. Genes Dev 15 2786–2796 [PubMed]
  • Elliott RC, Betzner AS, Huttner E, Oakes MP, Tucker WQ, Gerentes D, Perez P, Smyth DR (1996) AINTEGUMENTA, an APETALA2-like gene of Arabidopsis with pleiotropic roles in ovule development and floral organ growth. Plant Cell 8 155–168 [PMC free article] [PubMed]
  • Emery JF, Floyd SK, Alvarez J, Eshed Y, Hawker NP, Izhaki A, Baum SF, Bowman JL (2003) Radial patterning of Arabidopsis shoots by class III HD-ZIP and KANADI genes. Curr Biol 13 1768–1774 [PubMed]
  • Eshed Y, Baum SF, Bowman JL (1999) Distinct mechanisms promote polarity establishment in carpels of Arabidopsis. Cell 99 199–209 [PubMed]
  • Eshed Y, Baum SF, Perea JV, Bowman JL (2001) Establishment of polarity in lateral organs of plants. Curr Biol 11 1251–1260 [PubMed]
  • Ferrandiz C, Pelaz S, Yanofsky MF (1999) Control of carpel and fruit development in Arabidopsis. Annu Rev Biochem 68 321–354 [PubMed]
  • Franks RG, Liu Z, Fischer RL (2006) SEUSS and LEUNIG regulate cell proliferation, vascular development and organ polarity in Arabidopsis petals. Planta 224 801–811 [PubMed]
  • Franks RG, Wang C, Levin JZ, 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 [PubMed]
  • Klucher KM, Chow H, Reiser L, Fischer RL (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 [PMC free article] [PubMed]
  • Krizek BA (1999) Ectopic expression of AINTEGUMENTA in Arabidopsis plants results in increased growth of floral organs. Dev Genet 25 224–236 [PubMed]
  • Krizek BA, Prost V, Macias A (2000) AINTEGUMENTA promotes petal identity and acts as a negative regulator of AGAMOUS. Plant Cell 12 1357–1366 [PMC free article] [PubMed]
  • Kuusk S, Sohlberg JJ, Long JA, Fridborg I, Sundberg E (2002) STY1 and STY2 promote the formation of apical tissues during Arabidopsis gynoecium development. Development 129 4707–4717 [PubMed]
  • Liu Z, Franks RG, Klink VP (2000) Regulation of gynoecium marginal tissue formation by LEUNIG and AINTEGUMENTA. Plant Cell 12 1879–1892 [PMC free article] [PubMed]
  • Liu Z, Meyerowitz EM (1995) LEUNIG regulates AGAMOUS expression in Arabidopsis flowers. Development 121 975–991 [PubMed]
  • Matthews JM, Visvader JE (2003) LIM-domain-binding protein 1: a multifunctional cofactor that interacts with diverse proteins. EMBO Rep 4 1132–1137 [PMC free article] [PubMed]
  • McConnell JR, Barton MK (1998) Leaf polarity and meristem formation in Arabidopsis. Development 125 2935–2942 [PubMed]
  • McConnell JR, Emery J, Eshed Y, Bao N, Bowman J, Barton MK (2001) Role of PHABULOSA and PHAVOLUTA in determining radial patterning in shoots. Nature 411 709–713 [PubMed]
  • Mizukami Y, Fischer RL (2000) Plant organ size control: AINTEGUMENTA regulates growth and cell numbers during organogenesis. Proc Natl Acad Sci USA 97 942–947 [PMC free article] [PubMed]
  • Navarro C, Efremova N, Golz JF, Rubiera R, Kuckenberg M, Castillo R, Tietz O, Saedler H, Schwarz-Sommer Z (2004) Molecular and genetic interactions between STYLOSA and GRAMINIFOLIA in the control of Antirrhinum vegetative and reproductive development. Development 131 3649–3659 [PubMed]
  • Nemhauser JL, Feldman LJ, Zambryski PC (2000) Auxin and ETTIN in Arabidopsis gynoecium morphogenesis. Development 127 3877–3888 [PubMed]
  • Nole-Wilson S, Krizek BA (2000) DNA binding properties of the Arabidopsis floral development protein AINTEGUMENTA. Nucleic Acids Res 28 4076–4082 [PMC free article] [PubMed]
  • Nole-Wilson S, Krizek BA (2006) AINTEGUMENTA contributes to organ polarity and regulates growth of lateral organs in combination with YABBY genes. Plant Physiol 141 977–987 [PMC free article] [PubMed]
  • Nole-Wilson S, Tranby TL, Krizek BA (2005) AINTEGUMENTA-like (AIL) genes are expressed in young tissues and may specify meristematic or division-competent states. Plant Mol Biol 57 613–628 [PubMed]
  • Otsuga D, DeGuzman B, Prigge MJ, Drews GN, Clark SE (2001) REVOLUTA regulates meristem initiation at lateral positions. Plant J 25 223–236 [PubMed]
  • Pfluger J, Zambryski P (2004) The role of SEUSS in auxin response and floral organ patterning. Development 131 4697–4707 [PubMed]
  • Prigge MJ, Otsuga D, Alonso JM, Ecker JR, Drews GN, Clark SE (2005) Class III homeodomain-leucine zipper gene family members have overlapping, antagonistic, and distinct roles in Arabidopsis development. Plant Cell 17 61–76 [PMC free article] [PubMed]
  • Sawa S, Watanabe K, Goto K, Kanaya E, Morita EH, Okada K (1999) FILAMENTOUS FLOWER, a meristem and organ identity gene of Arabidopsis, encodes a protein with a zinc finger and HMG-related domains. Genes Dev 13 1079–1088 [PMC free article] [PubMed]
  • Sessions A (1999) Piecing together the Arabidopsis gynoecium. Trends Plant Sci 4 296–297 [PubMed]
  • Sessions A, Nemhauser JL, McColl A, Roe JL, Feldmann KA, Zambryski PC (1997) ETTIN patterns the Arabidopsis floral meristem and reproductive organs. Development 124 4481–4491 [PubMed]
  • Sessions RA, Zambryski PC (1995) Arabidopsis gynoecium structure in the wild and in ettin mutants. Development 121 1519–1532 [PubMed]
  • Sieber P, Gheyselinck J, Gross-Hardt R, Laux T, Grossniklaus U, Schneitz K (2004) Pattern formation during early ovule development in Arabidopsis thaliana. Dev Biol 273 321–334 [PubMed]
  • Siegfried KR, Eshed Y, Baum SF, Otsuga D, Drews GN, Bowman JL (1999) Members of the YABBY gene family specify abaxial cell fate in Arabidopsis. Development 126 4117–4128 [PubMed]
  • Smyth DR, Bowman JL, Meyerowitz EM (1990) Early flower development in Arabidopsis. Plant Cell 2 755–767 [PMC free article] [PubMed]
  • Sridhar VV, Surendrarao A, Gonzalez D, Conlan RS, Liu Z (2004) Transcriptional repression of target genes by LEUNIG and SEUSS, two interacting regulatory proteins for Arabidopsis flower development. Proc Natl Acad Sci USA 101 11494–11499 [PMC free article] [PubMed]
  • Sridhar VV, Surendrarao A, Liu Z (2006) APETALA1 and SEPALLATA3 interact with SEUSS to mediate transcription repression during flower development. Development 133 3159–3166 [PubMed]
  • van Meyel DJ, Thomas JB, Agulnick AD (2003) Ssdp proteins bind to LIM-interacting co-factors and regulate the activity of LIM-homeodomain protein complexes in vivo. Development 130 1915–1925 [PubMed]
  • Waites R, Hudson A (1995) phantastica: a gene required for dorsoventrality of leaves in Antirrhinum majus. Development 121 2143–2154
  • Zhong R, Ye ZH (1999) IFL1, a gene regulating interfascicular fiber differentiation in Arabidopsis, encodes a homeodomain-leucine zipper protein. Plant Cell 11 2139–2152 [PMC free article] [PubMed]
  • Zimmermann P, Hirsch-Hoffmann M, Hennig L, Gruissem W (2004) GENEVESTIGATOR. Arabidopsis microarray database and analysis toolbox. Plant Physiol 136 2621–2632 [PMC free article] [PubMed]

Articles from Plant Physiology are provided here courtesy of American Society of Plant Biologists
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...