Abstract

Members of the plant-specific gibberellic acid-stimulated Arabidopsis ( GASA ) gene family play roles in hormone response, defense and development. We have identified six new Arabidopsis GASA genes, bringing the total number of family members to 14. Here we show that these genes all encode small polypeptides that share the common structural features of an N-terminal putative signal sequence, a highly divergent intermediate region and a conserved 60 amino acid C-terminal domain containing 12 conserved cysteine residues. Analysis of promoter::GUS (β- glucuronidase ) transgenic plants representing six different GASA loci reveals that the promoters are activated in a variety of stage- and tissue-specific patterns during development, indicating that the GASA genes are involved in diverse processes. Characterization of GASA4 shows that the promoter is active in the shoot apex region, developing flowers and developing embryos. Phenotypic analyses of GASA4 loss-of-function and gain-of-function lines indicate that GASA4 regulates floral meristem identity and also positively affects both seed size and total seed yield.

Introduction

Hormone-regulated gene families play key roles in diverse biological processes such as flower induction, seed development and germination in a wide range of monocotyledonous and dicotyledonous plant species. Members of the gibberellic acid-stimulated Arabidopsis ( GASA ) gene family form a plant-specific group of genes present in a wide range of species including tomato, rice, petunia, gerbera and Arabidopsis . Several GASA -like genes are hormone responsive, and are proposed to participate in pathogen responses as well as various aspects of plant development. The tomato gene GAST1 was the first member of the family to be characterized, and its expression was induced upon application of exogenous gibberellin in a gibberellin-deficient background (Shi et al. 1992 ). A related tomato gene, RSI-1 , shares high sequence identity with GAST1 and is activated during lateral root formation (Taylor and Scheuring 1994 ). The GASA homologs GIP1 , GIP2 , GIP4 and GIP5 from petunia are activated by gibberellin in vitro (Ben-Nissan and Weiss 1996 , Ben-Nissan et al. 2004 ), and GIP2 RNAi lines displayed late flowering and reduced stem elongation, suggesting a developmental role in promoting internode elongation and the transition to flowering (Ben-Nissan et al. 2004 ). The GASA homolog GEG from Gerbera hybrida is expressed in corollas and carpels, and its expression coincides with the cessation of longitudinal cell expansion (Kotilainen et al. 1999 ). Plants ectopically expressing GEG developed shorter corollas with decreased cell length compared with the wild type, indicating a role for GEG as an inhibitor of cell elongation.

Eight GASA genes have previously been identified in Arabidopsis thaliana . GASA1GASA4 were first identified based on their similarity to tomato GAST1 (Herzog et al. 1995 ). Expression data indicated that GASA1 accumulates in flower buds and immature siliques, GASA2 and GASA3 in siliques and dry seeds, and GASA4 in growing roots and flower buds. Further studies showed GASA4 to be expressed in all meristematic regions—in shoots, in flower buds, and in primary and lateral roots (Aubert et al. 1998 ). Similarly to tomato GAST1 , GASA1 and GASA4 are activated by the bioactive gibberellin GA 3 added exogenously in a gibberellin-deficient background (Aubert et al. 1998 ). In an attempt to unravel cross-talk between brassinosteroids and gibberellins, Bouquin et al. ( 2001 ) also found that gibberellin activates and brassinosteroids inhibit GASA1 expression. Four additional GASA genes were uncovered based on sequence analysis: GASA5 and GASA6 (Aubert et al. 1998 ), as well as GASA7 and GASA8 (Berrocal-Lobo et al. 2002 ). However, no function has been assigned to any of the GASA genes in Arabidopsis .

The observation that several Arabidopsis GASA genes are expressed in flower buds suggests a possible role for members of this gene family in flower development. In A. thaliana , flower development is controlled in four steps (Jack 2004 ), the first of which is a switch from vegetative to reproductive growth by the action of flowering time genes in response to various environmental and endogenous signals. The next step integrates the activities of flowering time pathways to activate a small group of genes that specify floral meristem identity. LEAFY ( LFY ) and APETALA1 ( AP1 ) are major floral meristem identity genes that specify lateral meristems to develop into flowers rather than axillary inflorescence shoots (Weigel et al. 1992 , Mandel and Yanosky 1995 , Weigel and Nilsson 1995 ). LFY is a key integrator of signals from pathways that promote flowering, including the gibberellin pathway (Blazquez and Weigel 2000 ). The meristem identity genes activate the class A [ AP1 and APETALA2 ( AP2 )], class B [ APETALA3 ( AP3 ) and PISTILLATA ( PI )] and class C [ AGAMOUS ( AG )] floral organ identity genes that pattern discrete regions of the flower. Complexes of the floral organ identity transcription factors subsequently activate downstream ‘organ-building’ genes, which specify various cell types and tissues that will constitute the four organs of the mature flower. AP2 is unique among these genes in that it also plays a key role in seed development controlling seed mass (Jofuku et al. 2005 , Ohto et al. 2005 ). The expression of GASA4 and GASA6 is affected in plants with altered LFY or AP3/PI activity, respectively (Zik and Irish 2003 , Wagner et al. 2004 ), suggesting a role for these GASA genes in flower meristem and/or flower organ development.

In this work, we identify six new members of the GASA gene family in A. thaliana , GASA9GASA14 . Analyses of the GASA1 , GASA3 , GASA4 , GASA8 , GASA10 and GASA14 promoters show that all are activated in root tissue and the vasculature, although in distinct spatial and temporal patterns, as well as in other tissues including meristems, flowers and seeds. Through phenotypic analysis of mutant and transgenic plants, we show that GASA4 plays a role in regulating floral meristem and floral organ identity, and promotes seed size and weight. This is the first assignment of a function to the Arabidopsis GASA genes, and suggests that members of the GASA gene family have regulatory roles in addition to the structural roles indicated by previous studies.

Results

The GASA gene family consists of 14 genes

In an effort to analyze the GASA gene family thoroughly, in A. thaliana , we identified six new members of the family through whole-genome data mining. We named them GASA9GASA14 according to the nomenclature established in the literature, and propose that the GASA gene family in Arabidopsis consists of 14 members ( Table 1 ). The genes were identified using the criteria of overall nucleotide sequence similarity, and a common predicted protein structure consisting of an N-terminal signal sequence and a C-terminal conserved cysteine-rich domain (Herzog et al. 1995 ). We discovered that GASA11 (At2g18420/ GASA11 ) was incorrectly annotated. While gene predictions based on the Arabidopsis genomic sequence annotated the putative gene product as containing 25 additional amino acid residues downstream of the conserved cysteine-rich domain, we found by cDNA cloning and sequencing that GASA11 encodes a stop codon after the last residues of the conserved domain, as is the case for the other 13 Arabidopsis GASA genes. However, the gene we have named GASA13 lies in a region currently annotated as not encoding genes. Although the putative GASA13 gene sequence corresponds perfectly to the structure of the genes in the GASA family, there is no confirmation of expression data in the databases and nor were we successful in cloning a corresponding cDNA product. Thus it remains unclear whether this genomic sequence encodes a functional gene product.

Table 1

Overview of the GASA family, presenting GASA number, Arabidopsis gene number, current annotation in databases, expression information based on isolated clones and predicted signal sequence cleavage sites

GASA gene number  Arabidopsis gene number  Annotated as Expression Signal sequence cleavage after amino acid 
GASA1 At1g75750 GASA1 cDNA 23 
GASA2 At4g09610 GASA2 cDNA 25 
GASA3 At4g09600 GASA3 cDNA 26 
GASA4 At5g15230 GASA4 cDNA 25 
GASA5 At3g02885 GASA5 cDNA 27 
GASA6 At1g74670 Gibberellin-responsive protein, putative cDNA 23 
GASA7 At2g14900 Gibberellin-regulated family protein cDNA 23 
GASA8 At2g39540 Gibberellin-regulated family protein cDNA 25 
GASA9 At1g22690 Gibberellin-responsive protein, putative EST 24 
GASA10 At5g59845 Gibberellin-regulated family protein cDNA 25 
GASA11 At2g18420 Gibberellin-responsive protein, putative Our cDNA 23 
GASA12 At2g30810 Gibberellin-regulated family protein EST 22 
GASA13 – Non-annotated – 23 
GASA14 At5g14920 Gibberellin-regulated family protein cDNA 21 
GASA gene number  Arabidopsis gene number  Annotated as Expression Signal sequence cleavage after amino acid 
GASA1 At1g75750 GASA1 cDNA 23 
GASA2 At4g09610 GASA2 cDNA 25 
GASA3 At4g09600 GASA3 cDNA 26 
GASA4 At5g15230 GASA4 cDNA 25 
GASA5 At3g02885 GASA5 cDNA 27 
GASA6 At1g74670 Gibberellin-responsive protein, putative cDNA 23 
GASA7 At2g14900 Gibberellin-regulated family protein cDNA 23 
GASA8 At2g39540 Gibberellin-regulated family protein cDNA 25 
GASA9 At1g22690 Gibberellin-responsive protein, putative EST 24 
GASA10 At5g59845 Gibberellin-regulated family protein cDNA 25 
GASA11 At2g18420 Gibberellin-responsive protein, putative Our cDNA 23 
GASA12 At2g30810 Gibberellin-regulated family protein EST 22 
GASA13 – Non-annotated – 23 
GASA14 At5g14920 Gibberellin-regulated family protein cDNA 21 

GASA9GASA14 have been defined as part of the GASA family in this work.

All 14 Arabidopsis GASA family members have structural characteristics of secreted polypeptide molecules. Each putative protein contains an N-terminal 21–27 amino acid hydrophobic region that is predicted by SignalP v.2.0 to act as a cleavable signal sequence ( Table 1 ), designating them to be secreted into the extracellular space. This information suggests that the functional roles of the GASA gene family are related to processes in the extracellular environment. Downstream of the signal sequence, the putative GASA gene products contain a region displaying high divergence between family members. This region is both divergent in amino acid composition, from hydrophilic to hydrophobic, and ranges in sequence length from 190 amino acid residues in GASA14 to only three residues in GASA8. The primary structural feature of the GASA gene family constitutes a 59–60 amino acid conserved domain at the C-terminal end of the predicted proteins. A multiple sequence alignment of this C-terminal cysteine-rich region illustrates the conservation of this domain ( Fig. 1 a). Approximately 30% of the domain, 19 amino acid residues, is perfectly conserved between the predicted 14 proteins of the GASA family. Among these 19 residues are 12 conserved cysteines, with the possibility of six disulfide bridges being formed between them. In addition, several proline and glycine residues are fully or partially conserved.

Fig. 1

The Arabidopsis GASA gene family. (a) Multiple sequence alignment of the conserved C-terminal domain of the 14 putative GASA proteins. Identical residues are shown in black and conserved residues are shaded, with dark gray indicating 75–99% conservation and light gray 55–74% conservation. The specific amino acids included in the alignment for each GASA protein are indicated in parentheses. (b) An unrooted phylogenetic tree constructed using neighbor-joining. The branches were reconfirmed by 1,000 bootstrap resamplings, and the bootstrap value of each branch is indicated. (c) Multiple sequence alignment of the proline-rich regions from GASA14, soybean proline-rich protein 2, SbPRP2 (AAA34011); Arabidopsis arabinogalactan protein 9, AGP9 (NP973463); and Arabidopsis extensin 4, AtEx4 (NP849895). The amino acids included are indicated in parentheses.

Fig. 1

The Arabidopsis GASA gene family. (a) Multiple sequence alignment of the conserved C-terminal domain of the 14 putative GASA proteins. Identical residues are shown in black and conserved residues are shaded, with dark gray indicating 75–99% conservation and light gray 55–74% conservation. The specific amino acids included in the alignment for each GASA protein are indicated in parentheses. (b) An unrooted phylogenetic tree constructed using neighbor-joining. The branches were reconfirmed by 1,000 bootstrap resamplings, and the bootstrap value of each branch is indicated. (c) Multiple sequence alignment of the proline-rich regions from GASA14, soybean proline-rich protein 2, SbPRP2 (AAA34011); Arabidopsis arabinogalactan protein 9, AGP9 (NP973463); and Arabidopsis extensin 4, AtEx4 (NP849895). The amino acids included are indicated in parentheses.

The evolutionary relationship within the GASA family was examined by constructing a phylogenetic tree based on multiple sequence alignments of the C-terminal conserved cysteine-rich region ( Fig. 1 b). The GASA family members appear to form four primary phylogenetic branches supported by high bootstrap values. One clade contains five members formed by two pairs of closely related proteins, plus the more divergent GASA9. These two pairs, GASA2/GASA3 and GASA1/GASA11, display >60% amino acid identity between them. The second group is formed by the closely related GASA8 and GASA10, together with the somewhat evolutionarily more distant GASA7. The third cluster contains GASA5, GASA12, GASA13, GASA4 and GASA6. The highly divergent GASA14 does not group with any other GASA protein and forms a phylogenetic branch of its own.

GASA14, the protein that is most divergent from the other members of the GASA family, contains an intermediate domain displaying significant similarities with known protein domains, having segmental sequence identity with a range of proline-rich putative proteins in Arabidopsis and other plant species. The sequences similar to GASA14 are mostly annotated as cell wall-associated proteins such as extensins and arabinogalactans ( Fig. 1 c). This compositional similarity suggests that GASA14 may be located in the cell wall after secretion, by means of association of its long proline-rich region with the cell wall.

In silico searches within the GASA upstream sequences for cis -regulatory elements known to be of importance throughout the plant kingdom, using PLACE (Higo et al. 1999 ), identified several gibberellin-responsive elements such as GARE, the pyrimidine box and the TATCCAY elements (e.g. Gubler and Jacobsen 1992 , Gubler et al. 1999 , Yanagisawa and Schmidt 1999 ) in all 14 GASA promoters. Moreover, all the GASA upstream sequences contain multiple elements known to direct seed- and root-specific expression (e.g. Dickinson et al. 1988 , Opsahl-Sorteberg et al. 2004 , Nakabayashi et al. 2005 ). Eight GASA upstream sequences also contain one or both of the LFY and WUSCHEL (WUS) sequence elements found in the AG intron (Lenhard et al. 2001 , Lohmann et al. 2001 ). The GASA5 upstream region contains one putative LFY and two WUS sequence-binding elements, and the GASA4 upstream region contains two putative LFY-binding elements.

The GASA promoters are activated in a diverse range of tissues and developmental stages

Among the 14 identified GASA family members, promoter–reporter gene analyses of only GASA1 and GASA4 have been reported to date ( Table 2 ). To investigate the range of GASA promoter activation more thoroughly, we examined the spatial localization of the β- glucuronidase ( GUS ) gene product driven by the GASA1 , GASA3 , GASA4 , GASA8 , GASA10 and GASA14 promoters. The results shown here represent consistent patterns validated by examining multiple independent transgenic lines. Our analysis of p GASA4::GUS plants ( Fig. 2 ) confirms the pattern established by Aubert et al. ( 1998 ) and provides additional data. Consistent with the previous study, which used a different construct and region of the promoter, we detected promoter activity (GUS) predominantly in rapidly dividing cells and meristematic tissues. p GASA4 -directed GUS was detected in vegetative shoot apical meristems and initiating leaves ( Fig. 2 a, b), as well as in the vasculature of cotyledons ( Fig. 2 c), hypocotyls and rosette leaves (data not shown). In the root, GUS was observed in the distal region of the root apical meristem and the columella root cap cells ( Fig. 2 d), emerging lateral roots ( Fig. 2 e) and in the phloem of the vasculature ( Fig. 2 f, g). The flowers of p GASA4::GUS plants displayed GUS in the stylar region of the female reproductive organs, as well as strong staining in the stamen filaments and the petal vasculature ( Fig. 2 h, i). In seeds, p GASA4 :: GUS was detected in developing embryos ( Fig. 2 j, k). The GUS signal appeared to be uniformly distributed in the embryo and low or absent in the endosperm and maternal tissues of the seed.

Table 2

Patterns of GUS localization from Arabidopsis GASA promoters as determined by Raventos et al. ( 2000 ) (Rv), Aubert et al. ( 1998 ) (A) and this work (R), and Northern expression analysis of GASA genes as determined by Hertzog et al. (1995).

GASA gene  Roots Vasculature Hypocotyl/cotyledon Leaves Seeds Shoot apex Abscission zone Flowers 
GASA1 +(Rv,R) +(Rv,R) +(Rv,R) +(Rv,R) −(R) +(R) +(R) +(H,Rv,R) 
GASA2 −(H) −(H) −(H) −(H) +(H) −(H) −(H) −(H) 
GASA3 +(R) +(R) +(R) +(R) +(H) −(R) −(R) −(R) 
GASA4 +(H,A,R) +(R) +(A,R) +(R) +(R) +(A,R) −(R) +(H,A,R) 
GASA8 +(R) −(R) −(R) −(R) +(R) −(R) −(R) −(R) 
GASA10 +(R) +(R) +(R) +(R) +(R) −(R) −(R) −(R) 
GASA14 +(R) +(R) +(R) −(R) −(R) −(R) +(R) +(R) 
GASA gene  Roots Vasculature Hypocotyl/cotyledon Leaves Seeds Shoot apex Abscission zone Flowers 
GASA1 +(Rv,R) +(Rv,R) +(Rv,R) +(Rv,R) −(R) +(R) +(R) +(H,Rv,R) 
GASA2 −(H) −(H) −(H) −(H) +(H) −(H) −(H) −(H) 
GASA3 +(R) +(R) +(R) +(R) +(H) −(R) −(R) −(R) 
GASA4 +(H,A,R) +(R) +(A,R) +(R) +(R) +(A,R) −(R) +(H,A,R) 
GASA8 +(R) −(R) −(R) −(R) +(R) −(R) −(R) −(R) 
GASA10 +(R) +(R) +(R) +(R) +(R) −(R) −(R) −(R) 
GASA14 +(R) +(R) +(R) −(R) −(R) −(R) +(R) +(R) 

The absence or presence of expression/promoter activity is indicated by − for not detected and + for detected, respectively.

Fig. 2

Histochemical localization of GUS activity in p GASA4::GUS plants. (a) Dark-field image showing GUS in the vegetative shoot apical meristem and in initiating and emerging leaves. (b) Higher magnification image of GUS uniformly distributed throughout the SAM. (c) Cross-section of a cotyledon at 18 d after germination showing GUS specifically in the phloem cells of the vasculature. (d) Root tip showing GUS in the root cap cells and the distal parts of the root meristem. (e) GUS in emerging lateral roots. (f) Whole roots exhibit GUS in the vascular tissue. (g) Cross-section of the root revealing GUS in the phloem. (h) Flower showing very strong GUS in the stamen filaments. (i) Flower showing GUS in the vascular tissue of the petals and in the style. (j) Dark field image of a developing seed, displaying GUS in the embryo. (k) Higher magnification showing GUS uniformly distributed throughout the globular stage embryo. Scale bars = 20 μm.

Fig. 2

Histochemical localization of GUS activity in p GASA4::GUS plants. (a) Dark-field image showing GUS in the vegetative shoot apical meristem and in initiating and emerging leaves. (b) Higher magnification image of GUS uniformly distributed throughout the SAM. (c) Cross-section of a cotyledon at 18 d after germination showing GUS specifically in the phloem cells of the vasculature. (d) Root tip showing GUS in the root cap cells and the distal parts of the root meristem. (e) GUS in emerging lateral roots. (f) Whole roots exhibit GUS in the vascular tissue. (g) Cross-section of the root revealing GUS in the phloem. (h) Flower showing very strong GUS in the stamen filaments. (i) Flower showing GUS in the vascular tissue of the petals and in the style. (j) Dark field image of a developing seed, displaying GUS in the embryo. (k) Higher magnification showing GUS uniformly distributed throughout the globular stage embryo. Scale bars = 20 μm.

p GASA1 -driven GUS was observed in the vasculature of rosette leaves ( Fig. 3 a), cotyledons and hypocotyls (data not shown) as well as in the vegetative shoot apex ( Fig. 3 b). The flowers displayed strong GUS staining in the style, the stamen filaments and the vascular tissue of the sepals ( Fig. 3 c). A strong p GASA1 :: GUS signal was also observed in the flower abscission zone ( Fig. 3 d). In the underground tissue, staining was observed in the root apical meristem initials and the distal region of the root meristem ( Fig. 3 e), as well as in the root vasculature and emerging lateral roots ( Fig. 3 f). Cross-sections of the roots pinpointed GUS to the phloem cells of the vasculature ( Fig. 3 g).

Fig. 3

Histochemical localization of GUS activity in p GASA1::GUS plants. (a) GUS activity in the vascular tissue of leaves. (b) Whole seedling showing GUS in the region around the shoot apex. (c) Flowers displaying GUS in the vascular tissue of the sepals, in the stamen filaments and in the style. (d) Developing siliques exhibit strong GUS activity in the abscission zone. (e) GUS localization in the root meristem. (f) GUS localized to emerging lateral roots. (g) Cross-sections of the root localize GUS to the phloem cells of the vasculature. Scale bar = 20 μm.

Fig. 3

Histochemical localization of GUS activity in p GASA1::GUS plants. (a) GUS activity in the vascular tissue of leaves. (b) Whole seedling showing GUS in the region around the shoot apex. (c) Flowers displaying GUS in the vascular tissue of the sepals, in the stamen filaments and in the style. (d) Developing siliques exhibit strong GUS activity in the abscission zone. (e) GUS localization in the root meristem. (f) GUS localized to emerging lateral roots. (g) Cross-sections of the root localize GUS to the phloem cells of the vasculature. Scale bar = 20 μm.

The novel GUS localizations obtained with four additional GASA promoters are distinct, with only a few features in common ( Fig. 4 ). GUS localization driven by the GASA8 promoter was restricted to the underground tissue in the vegetative stage. Strong GUS signals were detected in the zone of elongating cells above the root tip, gradually becoming weaker in the more mature parts of the root tissue ( Fig. 4 a, b). In addition, p GASA8 -driven GUS was detected in large parts of the developing seeds ( Fig. 4 c). p GASA14::GUS seedlings displayed high GUS levels in the elongating hypocotyl and at the base of the cotyledons and leaves ( Fig. 4 d). Flowers displayed p GASA14 -driven GUS in the abscission zone, the style and the filaments of the stamens, resulting in a floral pattern very similar to that of GASA1 and GASA4 ( Fig. 4 e). However, p GASA14::GUS was not detected in sepals or petals. The roots of p GASA14::GUS plants displayed GUS in the young dividing root vascular cells above the root tip, and faded in the more mature vascular cells of the root ( Fig. 4 f). Emerging lateral roots showed strong p GASA14::GUS staining ( Fig. 4 g). p GASA3 -driven GUS was detected in the vascular tissues of leaves from young rosette plants as well as in the root ( Fig. 4 h, i). At later developmental stages, p GASA3 -driven GUS became considerably weaker and was not detected in inflorescences. p GASA10 -directed GUS was detected in the vasculature of both rosette leaves and roots ( Fig. 4 j, k), and in the tips of the cotyledons and roots ( Fig. 4 k, l). p GASA10 :: GUS was also detected in a wide area in the basal parts of developing seeds ( Fig. 4 m).

Fig. 4

GUS patterns from the GASA8 , GASA14 , GASA3 and GASA10 promoters. (a) At the seedling stage, p GASA8::GUS activity is restricted to roots. (b) p GASA8::GUS activity is strong in the cell elongation zone of the root, but it declines in the more mature root cells closer to the hypocotyl and is absent from the root tip. (c) Developing seeds show p GASA8::GUS activity in a domain covering large parts of the seed. (d) p GASA14::GUS activity is strong in the hypocotyl and the base of the cotyledons in 9-day-old seedlings. (e) Flowers displaying strong p GASA14::GUS activity in the abscission zone, the stamen filaments and the style. (f) p GASA14::GUS activity is observed specifically in the root vasculature, with higher accumulation towards the root tip. (g) p GASA14::GUS activity is localized to initiating lateral roots. (h) p GASA3::GUS activity is exclusively in the vascular tissue throughout the developing seedling. (i) p GASA3::GUS activity in the root vasculature. (j) p GASA10::GUS activity is localized in patches in the rosette leaf vascular tissue. (k) The root displays p GASA10::GUS activity in the vascular tissue and at the root tip. (l) p GASA10::GUS activity is detected at the tip of the cotyledons. (m) Developing seeds have GASA10 -driven GUS in a wide domain in the basal parts of the seed.

Fig. 4

GUS patterns from the GASA8 , GASA14 , GASA3 and GASA10 promoters. (a) At the seedling stage, p GASA8::GUS activity is restricted to roots. (b) p GASA8::GUS activity is strong in the cell elongation zone of the root, but it declines in the more mature root cells closer to the hypocotyl and is absent from the root tip. (c) Developing seeds show p GASA8::GUS activity in a domain covering large parts of the seed. (d) p GASA14::GUS activity is strong in the hypocotyl and the base of the cotyledons in 9-day-old seedlings. (e) Flowers displaying strong p GASA14::GUS activity in the abscission zone, the stamen filaments and the style. (f) p GASA14::GUS activity is observed specifically in the root vasculature, with higher accumulation towards the root tip. (g) p GASA14::GUS activity is localized to initiating lateral roots. (h) p GASA3::GUS activity is exclusively in the vascular tissue throughout the developing seedling. (i) p GASA3::GUS activity in the root vasculature. (j) p GASA10::GUS activity is localized in patches in the rosette leaf vascular tissue. (k) The root displays p GASA10::GUS activity in the vascular tissue and at the root tip. (l) p GASA10::GUS activity is detected at the tip of the cotyledons. (m) Developing seeds have GASA10 -driven GUS in a wide domain in the basal parts of the seed.

GASA4 affects shoot branching, floral determinacy and floral organ identity

To determine the functional roles of GASA4 , we analyzed the phenotypes of plants with altered GASA4 activity. We identified a GASA4 T-DNA insertion line, gasa4-1 , in the Col-0 ecotype. gasa4-1 plants produced no detectable GASA4 mRNA ( Fig. 5 a) and thus we conclude that gasa4-1 represents an RNA null allele. To assess the phenotypic effects of GASA4 gain of function, we generated 35S::GASA4 overexpression lines in the Wassilewskija (Ws) ecotype, which produced highly elevated levels of GASA4 mRNA compared with wild-type plants ( Fig. 5 b).

Fig. 5

RT–PCR analyses of plants with reduced or increased GASA4 expression. (a) GASA4 transcripts are detected in wild-type Col but not in gasa4-1 T-DNA insertion mutant plants. (b) Relative GASA4 steady-state transcript levels in two independent 35S::GASA4 -overexpressing plants compared with wild-type Ws plants. EF1α was amplified as a control. The amplification reactions were performed with RNA isolated from inflorescence tissues containing developing flower buds for 26 cycles.

Fig. 5

RT–PCR analyses of plants with reduced or increased GASA4 expression. (a) GASA4 transcripts are detected in wild-type Col but not in gasa4-1 T-DNA insertion mutant plants. (b) Relative GASA4 steady-state transcript levels in two independent 35S::GASA4 -overexpressing plants compared with wild-type Ws plants. EF1α was amplified as a control. The amplification reactions were performed with RNA isolated from inflorescence tissues containing developing flower buds for 26 cycles.

The main phenotypic aberration observed in the gasa4-1 plants was a significant increase in the number of axillary inflorescence shoots produced prior to flower formation ( Fig. 6 a). Both the total number of higher order axillary inflorescence shoots originating from the primary shoot and the number of secondary inflorescences emerging directly from the primary shoot were approximately 50% higher in gasa4-1 plants than in wild-type plants ( Fig. 6 a). However, the number of leaves produced during vegetative development was similar in the gasa4-1 plants compared with wild-type plants, suggesting that the observed phenotypic effect is specific to reproductive development.

Fig. 6

Phenotypes of the gasa4-1 null mutant and 35S::GASA4 -overexpressing lines. (a) Mean number of axillary inflorescences in wild-type (Col) and gasa4-1 plants, with standard deviations shown as error bars. The number of total higher order shoots produced by gasa4-1 plants ( n = 13) is significantly higher than that of wild-type Col plants ( n = 13). The Student t -test gives the two-tailed P -value of 0.0001. (b) The weight of individual seeds was calculated from 3–5 different seed batches of 100–700 seeds each and is presented as average individual seed weight. (c) Seed yield was calculated as average total seed yield harvested per plant ( n = 5). Significant differences compared with wild-type control plants are indicated with asterisks (* P < 0.05; ** P < 0.01). Error bars represent the SD.

Fig. 6

Phenotypes of the gasa4-1 null mutant and 35S::GASA4 -overexpressing lines. (a) Mean number of axillary inflorescences in wild-type (Col) and gasa4-1 plants, with standard deviations shown as error bars. The number of total higher order shoots produced by gasa4-1 plants ( n = 13) is significantly higher than that of wild-type Col plants ( n = 13). The Student t -test gives the two-tailed P -value of 0.0001. (b) The weight of individual seeds was calculated from 3–5 different seed batches of 100–700 seeds each and is presented as average individual seed weight. (c) Seed yield was calculated as average total seed yield harvested per plant ( n = 5). Significant differences compared with wild-type control plants are indicated with asterisks (* P < 0.05; ** P < 0.01). Error bars represent the SD.

In contrast to gasa4 loss of function, ectopic expression of GASA4 under the control of the cauliflower mosaic virus (CaMV) 35S promoter did not affect secondary inflorescence branching. Unlike gasa4-1 plants, 35S::GASA4 lines occasionally exhibited meristem identity changes in which the plants underwent the transition to flowering and produced floral meristems, followed by a reversion to normal indeterminate inflorescence development. 35S::GASA4 flowers exhibited mosaic floral organs, most frequently ectopic carpel or stamen structures. Abnormalities such as the development of structures with part carpel and part inflorescence identity were also observed, but at a very low frequency. The latter phenotypes are likely to be due to partial co-suppression of endogenous GASA4 activity in the 35S::GASA4 lines. The average occurrence of flowers with developmental abnormalities in two independent 35S::GASA4 lines was 13 and 20 flowers per 10 plants, while a third line exhibited seven abnormal flowers per 10 plants.

GASA4 activity increases seed size and total seed yield

Analysis of plants with altered GASA4 expression levels showed that GASA4 affects seed size, seed weight and total seed yield ( Fig. 6 b, c). Seeds from gasa4-1 null mutant plants had significantly reduced seed weight, while, conversely, seeds from 35S::GASA4 -overexpressing lines had significantly increased seed weight ( Fig. 6 b). In addition to an increase in individual seed size, total seed yield was also increased in plants overexpressing GASA4 ( Fig. 6 c). These data indicate that the increased seed weight is a direct effect of the ectopic GASA4 activity and not a result caused by allocating more resources per seed due to lower seed set. Even gasa4-1 plants, with smaller seeds, still exhibited higher total seed yields than wild-type plants, accounted for by increased branching leading to an increased total number of seeds (data not shown). These results indicate a function for GASA4 in controlling seed development and seed yield through independent effects, where GASA4 directly affects seed size in addition to affecting inflorescence branching and seed set, both of which contribute to increasing the total seed mass production.

Discussion

The Arabidopsis GASA genes encode putative secreted polypeptides with a conserved cysteine-rich domain

In this work we have identified six novel members of the GASA gene family in Arabidopsis by whole genome data mining, bringing the total number of family members to 14. The conservation lies primarily in 12 cysteine residues distributed throughout the C-terminus of all 14 predicted GASA proteins ( Fig. 1 a). This domain probably provides a conserved mode of action at the protein level. The distribution of the cysteines throughout the domain suggests that a specific globular protein structure is formed by the generation of up to six disulfide bridges between the cysteine residues ( http://www.sanger.ac.uk/cgi-bin/Pfam/getacc?PF02704 ). The variable positions between the conserved cysteines, in addition to the variation in the intermediate domain, may give rise to differences in functions for the GASA proteins.

The phylogenetic tree generated based on alignment of the C-terminal conserved domain of the 14 Arabidopsis GASA gene products depicts four phylogenetic clusters. Several GASA genes form closely related pairs: GASA2 and GASA3 , GASA1 and GASA11 , GASA8 and GASA10 , and also GASA4 and GASA6 . These pairs might have related functions, as many genes in Arabidopsis have close homologs with which they are functionally redundant (e.g. Ohgishi et al. 2004 , Shpak et al. 2004 ). The use of out-group sequences would probably have provided a more informative phylogenetic tree. However, other sequences related to the GASA family, such as GAST , GIP and GEG sequences, are all found in other related angiosperms, and thus are not evolutionarily distant enough to root the tree.

The GASA promoters drive specific and distinct GUS localization patterns

The data obtained from our GASA promoter::GUS analyses reveal that the different GASA promoters are active in distinct tissues and developmental stages, suggesting that they fulfill distinct functions in the plant. Even though all GASA promoters investigated are active in roots and in the vasculature, the patterns are highly specific and different with respect to cell type and temporal regulation. The novel regulation patterns reported here for GASA3 , GASA8 , GASA10 and GASA14 further demonstrate the variability within the family. In addition, we have expanded the data available on GASA1 and GASA4 (Aubert et al. 1998 , Raventos et al. 2000 ). Our results reveal that there is considerable overlap between the GUS patterns driven by the GASA1 and GASA4 promoters. Even though GASA1 and GASA4 belong to different phylogenetic clusters within the GASA family, the overlapping GUS localizations suggest that they may provide partially overlapping functions. In light of the proposed role of the GASA homolog GEG in G. hybrida cell elongation (Kotilainen et al. 1999 ), our GUS analyses identify GASA8 and GASA14 as candidates for being involved in similar processes to GEG because the two promoters are highly active in zones typically associated with cell expansion, such as growing cotyledons and cells above the root initials.

Extensive research has established an important function for gibberellin to induce stem elongation, leaf expansion, flowering, seed development and germination (e.g. Isabel-LaMoneda et al. 2003 ). Yet, little is known about how these gibberellin effects are mediated and the genes involved. A search through the promoter regions of the 14 GASA genes revealed that they all contain one or more sequence elements putatively associated with gibberellin responses during seed development. This information suggests that many, if not all, of the GASA genes may be regulated by gibberellin. Different sensitivity and interactions between different signaling pathways and hormone effects require extensive studies to unravel the different mechanisms during flower and seed development (Swain and Singh 2005 ). This is illustrated by GASA4 expression, which is up-regulated by gibberellin in meristematic regions yet down-regulated by gibberellin in cotyledons and leaves (Aubert et al. 1998 ). GASA4 is likewise expected to be differentially regulated during flower and seed development. One possible explanation is that GASA4 may be associated with cell division, in which case it would be expected to be abundant in meristematic tissues with higher cell division rates and reduced in tissues undergoing differentiation (Aubert et al. 1998 ). Alternatively, the differential effects of gibberellin on GASA4 transcription may involve as yet unidentified tissue-restricted contributing factors, or may reflect a minor role for gibberellin in regulating GASA4 gene expression.

GASA4 is a regulator of floral meristem identity and seed development

The localization of p GASA4::GUS in meristems, initiating organs and seeds correlates with the abnormal shoot and flower architecture phenotypes as well as the seed size and yield phenotypes observed in the gasa4-1 null mutants and 35S::GASA4 plants. The GASA4 promoter is active throughout early stage embryos, the vegetative shoot apical meristem and young leaf primordia. In the reproductive phase, the GASA4 promoter is active during the early stages of flower development, as well as in specific mature organs such as the style and the stamen filaments.

Analysis of GASA4 loss-of-function plants indicates that GASA4 activity is important to promote the transition between axillary shoot meristem formation and flower formation. gasa4-1 null mutants produce an increased number of axillary shoots before generating flowers, indicating that GASA4 activity is required to promote floral meristem identity. In contrast, constitutive and ectopic 35S::GASA4 expression does not cause conversion of axillary meristems to a floral meristem fate, revealing that GASA4 is not sufficient to induce a premature floral transition, or the correct fine tuning of the regulation might be a prerequisite to obtain the floral transition.

The LFY , AP1 and UNUSUAL FLORAL ORGANS ( UFO ) genes all promote Arabidopsis floral meristem identity, and the increased branching phenotypes of plants with reduced GASA4 expression are similar to those observed in weak lfy mutants (Weigel et al. 1992 ) and strong ufo mutants (Levin and Meyerowitz 1995 ). In silico searches within the GASA upstream promoters for cis -regulatory elements using PLACE (Higo et al. 1999 ) revealed the presence of a conserved sequence element known to be bound by the LFY transcription factor (Lohmann et al. 2001 ) at two locations in the GASA4 promoter. The similarity between the lfy and gasa4 phenotypes, and the presence of two putative LFY-binding elements in the GASA4 promoter, thus suggests that GASA4 may be regulated by LFY. Indeed, GASA4 was recently identified as a putative LFY target in a microarray experiment to uncover genes exhibiting significant changes in expression levels upon LFY activation (Wagner et al. 2004 ). Thus, based on the combined genetic and molecular evidence, we propose that GASA4 is a target of LFY transcriptional regulation and is a part of the physiological output of LFY function.

GASA4 -driven GUS is also present in embryos, and GASA4 plays an important role in seed development. In Arabidopsis seeds, the embryo accounts for most of the seed mass, and seed mass is negatively correlated with the number of seeds produced (Ohto et al. 2005 ). We find that GASA4 has a direct effect on both seed size and total seed mass. gasa4-1 plants have reduced seed size and 35S::GASA4 plants have increased seed size, demonstrating that seed size increases quantitatively with increasing GASA4 expression levels. We do not yet know whether this is caused by an increase in cell size or an increase in cell number due to excess cell division, or possibly both. Both increased and decreased GASA4 expression significantly increase total seed mass, even though decreased GASA4 expression results in the production of smaller seeds. This effect can be explained by the increased shoot branching that occurs in gasa4-1 plants, which increases the total number of flowers that are formed and therefore enhances overall seed set. The AP2 gene, in addition to affecting floral meristem and floral organ identity (e.g. Jofuku et al. 1994 , Bowman et al. 1989 ) and regulating the stem cell niche in the shoot apical meristem (Würschum et al. 2006 ), has also been found to regulate seed mass (Jofuku et al. 2005 , Ohto et al. 2005 ). It has been suggested that AP2 acts on seed mass in part by suppressing gibberellin signaling and affecting cell size and cell number during seed growth (Okamuro et al. 1997 , Jofuku et al. 2005 ). Loss-of-function ap2 plants display increased seed size (Jofuku et al. 2005 , Ohto et al. 2005 ), while the dominant-negative AP2 allele l28 causes reduced seed size (Würschum et al. 2006 ). These phenotypes are opposite to those caused by mutations in GASA4 , suggesting that GASA4 and AP2 may play opposing roles in regulating seed size during plant development.

GASA proteins as putative regulatory molecules

The GASA -like gene GEG has been proposed to encode a cell wall-localized protein that plays a structural role during cell elongation in G. hybrida , based on its putative secretion, the relative abundance of its transcripts and its induction by gibberellin, which is typically involved in cell wall properties such as division and extension (Kotilainen et al. 1999 , Ben-Nissan et al. 2004 ). Our results suggest that at least one member of the GASA family plays a regulatory rather than a structural role. We have shown that GASA4 regulates seed development, plant architecture and the flowering transition, and based on their tissue- and stage-specific promoter activation patterns it appears that additional GASA gene products may also regulate aspects of development. The regulatory aspect introduced by the observed GASA4 phenotypes suggests that it and the other GASA proteins might function by participating in signaling pathways, e.g. as extracellular signaling molecules. Thus, a future goal is to determine whether the GASA proteins act as small secreted ligands in developmental signal transduction pathways.

Materials and Methods

Plant material and growth conditions

Arabidopsis thaliana seeds were surface sterilized for 5 min in a 2% sodium hypochlorite solution and plated on Murashige and Skoog ( 1962 ) agar plates with 20 g liter −1 sucrose and 80 mg liter −1 kanamycin for selection of transgenic plants. Seeds of ecotype Ws were surface sterilized and then directly transferred to growth chambers holding 20°C and long-day conditions (16 h light, 8 h dark) or, alternatively, constant light conditions. Seeds of ecotype Colombia (Col) were cold treated at 4°C for 4 d before being transferred to the growth chambers. At 10–14 d after germination, the seedlings were transferred to soil, and placed in a greenhouse complex with 20°C and long-day conditions, or to growth chambers holding 20°C and constant light. The T-DNA mutant allele gasa4-1 (N042431), containing a T-DNA insertion in the first exon of GASA4 , was obtained from the T-DNA Express Collection at the Salk Institute ( http://signal.salk.edu/cgi-bin/tdnaexpress ).

Phylogenetic analysis

Putative GASA protein sequences were obtained from public databases ( http://www.ncbi.nlm.nih.gov/ , http://www.arabidopsis.org , http://mips.gsf.de/proj/thal/db/index.html ) and from our cloning results. Multiple sequence alignments were constructed using ClustalW (Thompson et al. 1994 ) and displayed with GeneDoc ( www.psc.edu/biomed/genedoc ). On the basis of the multiple sequence alignment, a neighbor-joining phylogenetic tree was obtained using the PAUP v.4.0 software ( http://www.paup.csit.fsu.edu ), and statistical confidence was calculated by bootstrap analysis with 1,000 resamplings.

PCR amplification, cloning and transformation

The coding regions corresponding to the GASA5 , GASA9 , GASA10 and GASA11 genes were amplified by PCR from poly(A)-primed reverse-transcribed total RNA isolated from Ws flowers, while Ws genomic DNA was used as a template to amplify GASA8 , GASA12 and GASA13 . The seven remaining GASA genes GASA1 , GASA2 , GASA3 , GASA4 , GASA6 , GASA7 and GASA14 were amplified from expressed sequence tag (EST) clones in the vector Lambda ZAP II Reference pBluescript, provided by The Arabidopsis Biological Resource Center ( http://www.biosci.ohio-state.edu/~plantbio/Facilities/abrc/abrchome.htm ). Gene-specific primers were used in all amplifications from Arabidopsis genomic DNA and cDNA, while amplifications of GASA genes from EST clones were performed using Sp6, gatttaggtgacactatag; T3, attaaccctcactaaaggga; and T7, taatacgactcactataggg primers. The gene-specific primers used to amplify GASA coding regions from cDNA or genomic DNA were: GASA9 f, B1-tcaaaatcactactcaag; GASA9 r, B2-rcttttaaggacatttgag; GASA11 f, B1-cctcttttgttttctctg; GASA11 r, B2-ttttaaggacactttctgc; GASA12 f, B1-gaaatatatcataagaaaatc; GASA12 r, B2-ttttcatggacattttggtc; GASA8 f, B1-caaactctttcgagaaaacaa; GASA8 r, B2-tcaaggacacttggatcc; GASA5 f, B1-taagaatcattcattggcaatc; GASA5 r, B2-tttaaggacattttggacg; GASA10 f, B1-ccaaacagattattattaaaac; GASA10 r, B2-ctataaggataatcttaagaac; GASA13 f, B1-acctgagcttcattacaatc; and GASA13 r, B2-aattaagaggagaaactaag. B1 and B2 refer to the flanking 25 bp recombination sites necessary for GATEWAY™ recombination (Invitrogen).

GASA promoter regions extending between 1.0–2.0 kb upstream of the expected translation start site were amplified for the GASA1 , GASA3 , GASA4 , GASA8 , GASA10 and GASA14 genes using genomic Arabidopsis DNA as a template. The precise amplified regions were as follows: for GASA1 , a fragment from −1,298 to −15 upstream of the expected translation start site, for GASA3 from −1,218 to −5, for GASA4 from −1,717 to −20, for GASA14 from −1,075 to −3, for GASA8 from −1,574 to −6, and for GASA10 from −1,669 to −59. The primers used to amplify these fragments were GASA1 fw, cataactcataagaagaagtgc; GASA1 r, gtttatgatgagagaacttagc; GASA3 fw, tttgttccaagtttggcaacc; GASA3 r, aaaaagcatgcagtttcgtgc; GASA4 fw, gataatgaaatgatccaactgg; GASA4 r, atccaaagaaccaaacactcc; GASA14 fw, gagattgtggagtactgtgg; GASA14 r, tgtgggagatgagaaagtgg; GASA8 fw, B1-gtgctgatgaggaccacc; GASA8 r, B2-ttctcgaaagagtttgctacg; GASA10 fw, B1-gaaccggagaccgtttacc; and GASA10 r, B2-agaagatgaagagaaaactcg. All PCRs were performed with the thermal cycler GeneAmp PCR system 9600 provided by Applied Biosystems.

Amplified PCR fragments were cloned into appropriate constructs using GATEWAY™ cloning technology (Invitrogen) as described by the manufacturer. The binary GATEWAY™ destination vectors (Karimi et al. 2002 ) used were PKGWFS7 for obtaining a GASA:: GUS-GFP fusion construct and PK7 WG2D giving the CaMV35S:: GASA construct for overexpression. The destination vectors were provided by Flanders Interuniversity, Institute for Biotechnology ( http://www.psb.rug.ac.be/gateway ). Standard Agrobacterium tumefasciens -mediated transformation of binary constructs was performed with ecotype Ws (Clough and Bent 1998 ).

RNA isolation, reverse transcription and PCR

Total RNA was isolated from inflorescence tissues containing developing flower buds using the RNeasy RNA isolation kit (Qiagen) according to the manufacturer's instructions. Single-stranded cDNA was synthesized from 2 μg of total RNA with the Superscript™ III first-strand synthesis system (Invitrogen). A 80 ng aliquot of cDNA was used in a semi-quantitative PCR, using the primers GASA4 f, gaggagtgtttggttctttgg; GASA4 r, taaaaagggaacgaagggag; EF1αf, caggctgattgtgctgttcttatcat; and EF1αr, cttgtagacatcctgaagtggaaga, in a 20 μl reaction using Taq polymerase (Promega). In order to rule out possible contaminating genomic DNA templates, each primer set was designed to span one or more introns. The PCR (94°C, 1 min; followed by cycles of 94°C, 30 s; 55°C, 30 s; 72°C, 1 min; followed by 72°C, 7 min) was performed for 26 cycles.

Histology and microscopy

For promoter::GUS analysis, plant material was stained in GUS staining solution (100 mM NaPO 4 pH 7.2, 1 mg ml −1 X-Gluc in dimethylformamide (DMF), 5 mM potassium hexacyanoferrate (II), 0.5 nM potassium hexacyanoferrate (III), 0.1% Triton X-100) and incubated overnight at 37°C. The material was washed repeatedly in 70% ethanol followed by 100% ethanol, and seeds were further cleared in clearing solution (8 g of chloral hydrate, 1 ml of glycerol, 2 ml of H 2 O). The tissue was analyzed under a LEICA DM LB microscope, and photographed with a Nikon Coolpix 995 digital camera.

Tissue was prepared for sectioning by fixation in 1.25% glutaraldehyde and 2% paraformaldehyde, dehydrated through a graded ethanol series, and infiltrated and embedded in LR White Hard Grade Acrylic Resin (Electron Microscopy Sciences) according to the manufacturer's instructions. Sections of 1 μm were prepared using an ultramicrotome and a diamond knife, the sections mounted on slides, and GUS staining visualized in dark-field mode.

Seed mass and seed yield analyses

Average seed weight was determined by weighing mature dry seeds in batches of 100–700 seeds. The weight of 3–5 batches was further measured for each seed lot. Total seed yield from individual plants was collected using the Aracon system (Aracon containers, BetaTech, Gent, Belgium).

Acknowledgments

We thank Lene T. Olsen, Hege Divon (Norwegian University of Life Sciences) and Enrico Magnani (Plant Gene Expression Center) for technical assistance, as well as Ragnhild Nestestog (Norwegian University of Life Sciences/NARC) and the National Programme for Research on Functional Genomics in Norway (FUGE) in the Research Council of Norway for support. We thank Unifarm (Wageningen University and Research Centre) for greenhouse facilities, and the Arabidopsis Biological Resource Center and the SALK T-DNA collection for providing seed stocks. We thank the Flanders Interuniversity for supplying the GATEWAY vectors. This work was supported by The Norwegian Research Council, and by an USDA CRIS grant to J.C.F.

References

Aubert
D
Chevillard
M
Dorne
AM
Arlaud
G
Herzog
M
Expression patterns of GASA genes in Arabidopsis thaliana : the GASA4 gene is up-regulated by gibberellins in meristematic regions
Plant. Mol. Biol.
 , 
1998
, vol. 
36
 (pg. 
871
-
883
)
Ben-Nissan
G
Lee
JY
Borohov
A
Weiss
D
GIP, a Petunia hybrida GA-induced cysteine-rich protein: a possible role in shoot elongation and transition to flowering
Plant J.
 , 
2004
, vol. 
37
 (pg. 
229
-
238
)
Ben-Nissan
G
Weiss
D
The petunia homologue of tomato gast1: transcript accumulation coincides with gibberellin-induced corolla cell elongation
Plant Mol. Biol.
 , 
1996
, vol. 
32
 (pg. 
1067
-
1074
)
Berrocal-Lobo
M
Segura
A
Moreno
M
Lopez
G
Garcia-Olmedo
F
Molina
A
Snakin-2, an antimicrobial peptide from potato whose gene is locally induced by wounding and responds to pathogen infection
Plant Physiol.
 , 
2002
, vol. 
128
 (pg. 
951
-
961
)
Blazquez
MA
Weigel
D
Integration of floral inductive signals in Arabidopsis
Nature
 , 
2000
, vol. 
404
 (pg. 
889
-
892
)
Bouquin
T
Meier
C
Foster
R
Nielsen
ME
Mundy
J
Control of specific gene expression by gibberellin and brassinosteroid
Plant Physiol.
 , 
2001
, vol. 
127
 (pg. 
450
-
458
)
Bowman
JL
Smyth
DR
Meyerowitz
EM
Genes directing flower development in Arabidopsis
Plant Cell
 , 
1989
, vol. 
1
 (pg. 
37
-
52
)
Clough
SJ
Bent
AF
Floral dip. a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana
Plant J.
 , 
1998
, vol. 
16
 (pg. 
735
-
743
)
Dickinson
CD
Evans
RP
Nielsen
N
RY repeats are conserved in the 5′-flanking regions of legume seed-protein genes
Nucleic Acids Res.
 , 
1988
, vol. 
16
 pg. 
371
 
Gubler
F
Jacobsen
JV
Gibberellin-responsive elements in the promoter of a barley high-pI alpha-amylase gene
Plant Cell
 , 
1992
, vol. 
4
 (pg. 
1435
-
1441
)
Gubler
F
Raventos
D
Keys
M
Watts
R
Mundy
J
Jacobsen
JV
Target genes and regulatory domains of the GAMYB transcriptional activator in cereal aleurone
Plant J.
 , 
1999
, vol. 
17
 (pg. 
1
-
9
)
Herzog
M
Dorne
AM
Grellet
F
GASA, a gibberellin-regulated gene family from Arabidopsis thaliana related to the tomato GAST1 gene
Plant Mol. Biol.
 , 
1995
, vol. 
27
 (pg. 
743
-
752
)
Higo
K
Ugawa
Y
Iwamoto
M
Korenaga
T
Plant cis-acting regulatory DNA elements (PLACE) database
Nucl Acids Res.
 , 
1999
, vol. 
27
 (pg. 
297
-
300
)
Isabel-LaMoneda
I
Diaz
I
Martinez
M
Mena
M
Carbonero
P
SAD: a new DOF protein from barley that activates transcription of a cathepsin B-like thiol protease gene in the aleurone of germinating seeds
Plant J.
 , 
2003
, vol. 
33
 (pg. 
329
-
340
)
Jack
T
Molecular and genetic mechanisms of floral control
Plant Cell
 , 
2004
, vol. 
16
 (pg. 
S1
-
S17
)
Jofuku
KD
den Boer
BG
Van Montagu
M
Okamuro
JK
Control of Arabidopsis flower and seed development by the homeotic gene APETALA2
Plant Cell
 , 
1994
, vol. 
6
 (pg. 
1211
-
1225
)
Jofuku
KD
Omidyar
PK
Gee
Z
Okamuro
JK
Control of seed mass and seed yield by the floral homeotic gene APETALA2
Proc. Natl Acad. Sci. USA
 , 
2005
, vol. 
102
 (pg. 
3117
-
3122
)
Karimi
M
Inze
D
Depicker
A
GATEWAY vectors for Agrobacterium-mediated plant transformation
Trends Plant Sci
 , 
2002
, vol. 
7
 (pg. 
193
-
195
)
Kotilainen
M
Helariutta
Y
Mehto
M
Pollanen
E
Albert
VA
Elomaa
P
Teeri
TH
GEG participates in the regulation of cell and organ shape during corolla and carpel development in gerbera hybrida
Plant Cell
 , 
1999
, vol. 
11
 (pg. 
1093
-
1104
)
Lenhard
M
Bohnert
A
Jurgens
G
Laux
T
Termination of stem cell maintenance in Arabidopsis floral meristems by interactions between WUSCHEL and AGAMOUS
Cell
 , 
2001
, vol. 
105
 (pg. 
805
-
814
)
Levin
JZ
Meyerowitz
EM
UFO: an Arabidopsis gene involved in both floral meristem and floral organ development
Plant Cell
 , 
1995
, vol. 
7
 (pg. 
529
-
548
)
Lohmann
JU
Hong
RL
Hobe
M
Busch
MA
Parcy
F
Simon
R
Weigel
D
A molecular link between stem cell regulation and floral patterning in Arabidopsis
Cell
 , 
2001
, vol. 
105
 (pg. 
793
-
803
)
Mandel
MA
Yanofsky
MF
A gene triggering flower formation in Arabidopsis
Nature
 , 
1995
, vol. 
377
 (pg. 
522
-
524
)
Murashige
T
Skoog
F
A revised medium for rapid growth and bioassays with tobacco tissue cultures
Physiol. Plant.
 , 
1962
, vol. 
15
 (pg. 
473
-
479
)
Nakabayashi
K
Okamoto
M
Koshiba
T
Kamiya
Y
Nambara
E
Genome-wide profiling of stored mRNA in Arabidopsis thaliana seed germination: epigenetic and genetic regulation of transcription in seed
Plant J.
 , 
2005
, vol. 
41
 (pg. 
697
-
709
)
Ohgishi
M
Saji
K
Okada
K
Sakai
T
Functional analysis of each blue light receptor, cry1, cry2, phot1, and phot2, by using combinatorial multiple mutants in Arabidopsis
Proc. Natl Acad. Sci. USA
 , 
2004
, vol. 
101
 (pg. 
2223
-
2228
)
Ohto
M
Fischer
RL
Goldberg
RB
Nakamura
K
Harada
JJ
Control of seed mass by APETALA2
Proc. Natl Acad. Sci. USA
 , 
2005
, vol. 
102
 (pg. 
3123
-
3128
)
Okamuro
JK
Szeto
W
Lotys-Prass
C
Jofuku
KD
Photo and hormonal control of meristem identity in the Arabidopsis flower mutants apetala2 and apetala1
Plant Cell
 , 
1997
, vol. 
9
 (pg. 
37
-
47
)
Opsahl-Sorteberg
HG
Divon
HH
Nielsen
PS
Kalla
R
Hammond-Kosack
M
Shimamoto
K
Kohli
A
Identification of a 49-bp fragment of the HvLTP2 promoter directing aleurone cell specific expression
Gene
 , 
2004
, vol. 
341
 (pg. 
49
-
58
)
Raventos
D
Meier
C
Jensen
O
Ma
B
Mundy
J
Fusion genetic analysis of gibberellin signaling mutants
Plant J.
 , 
2000
, vol. 
22
 (pg. 
427
-
438
)
Shi
L
Gast
R
Gopalraj
M
Olszewski
N
Characterization of a shoot-specific, GA3- and ABA-regulated gene from tomato
Plant J.
 , 
1992
, vol. 
2
 (pg. 
153
-
159
)
Shpak
ED
Berthiaume
CT
Hill
EJ
Torii
KU
Synergistic interaction of three ERECTA-family receptor-like kinases controls Arabidopsis organ growth and flower development by promoting cell proliferation
Development
 , 
2004
, vol. 
131
 (pg. 
1491
-
1501
)
Swain
SM
Singh
DP
Tall tales from sly dwarves: novel functions of gibberellins in plant development
Trends Plant Sci.
 , 
2005
, vol. 
10
 (pg. 
123
-
129
)
Taylor
BH
Scheuring
CF
A molecular marker for lateral root initiation: the RSI-1 gene of tomato ( Lycopersicon esculentum Mill) is activated in early lateral root primordia
Mol. Gen. Genom.
 , 
1994
, vol. 
243
 (pg. 
148
-
157
)
Thompson
JD
Higgins
DG
Gibson
TJ
ClustalW, improving the sensitivity of progressive multiple sequence alignment through sequence weighting
Nucleic Acids Res.
 , 
1994
, vol. 
22
 (pg. 
4673
-
4680
)
Wagner
D
Wellmer
F
Dilks
K
William
D
Smith
MR
Kumar
PP
Riechmann
JL
Greenland
AJ
Meyerowitz
EM
Floral induction in tissue culture: a system for the analysis of LEAFY-dependent gene regulation
Plant J.
 , 
2004
, vol. 
39
 (pg. 
273
-
282
)
Weigel
D
Alvarez
J
Smyth
DR
Yanofsky
MF
Meyerowitz
EM
LEAFY controls floral meristem identity in Arabidopsis
Cell
 , 
1992
, vol. 
69
 (pg. 
843
-
859
)
Weigel
D
Nilsson
O
A developmental switch sufficient for flower initiation in diverse plants
Nature
 , 
1995
, vol. 
377
 (pg. 
495
-
500
)
Wurschum
T
Gross-Hardt
R
Laux
T
APETALA2 regulates the stem cell niche in the Arabidopsis shoot meristem
Plant Cell
 , 
2006
, vol. 
18
 (pg. 
295
-
307
)
Yanagisawa
S
Schmidt
RJ
Diversity and similarity among recognition sequences of Dof transcription factors
Plant J.
 , 
1999
, vol. 
17
 (pg. 
209
-
14
)
Zik
M
Irish
VF
Global identification of target genes regulated by APETALA3 and PISTILLATA floral homeotic gene action
Plant Cell
 , 
2003
, vol. 
15
 (pg. 
207
-
222
)

Abbreviations:

    Abbreviations:
  • EST

    expressed sequence tag

  • GASA

    gibberellic acid-stimulated Arabidopsis ( GASA )

  • GUS

    β-glucuronidase

Author notes

5 These authors contributed equally to this work.