In the model plant Arabidopsis thaliana, a core eudicot, the floral homeotic C-class gene AGAMOUS (AG) has a dual role specifying reproductive organ identity and floral meristem determinacy. We conduct a functional analysis of the putative AG ortholog ThtAG1 from the ranunculid Thalictrum thalictroides, a representative of the sister lineage to all other eudicots. Down-regulation of ThtAG1 by virus-induced gene silencing resulted in homeotic conversion of stamens and carpels into sepaloid organs and loss of flower determinacy. Moreover, flowers exhibiting strong silencing of ThtAG1 phenocopied the double-flower ornamental cultivar T. thalictroides 'Double White.' Molecular analysis of 'Double White' ThtAG1 alleles revealed the insertion of a retrotransposon causing either nonsense-mediated decay of transcripts or alternative splicing that results in mutant proteins with K-domain deletions. Biochemical analysis demonstrated that the mutation abolishes protein-protein interactions with the putative E-class protein ThtSEP3. C-and E-class protein heterodimerization is predicted by the floral quartet model, but evidence for the functional importance of this interaction is scarce outside the core eudicots. Our findings therefore corroborate the importance and conservation of the interactions between C-and E-class proteins. This study provides a functional description of a full C-class mutant in a noncore ("basal") eudicot, an ornamental double flower, affecting both organ identity and meristem determinacy. Using complementary forward and reverse genetic approaches, this study demonstrates deep conservation of the dual C-class gene function and of the interactions between C-and E-class proteins predicted by the floral quartet model. floral organ identity genes | MADS-box genes | solo long terminal repeats | RNA silencing C urrent understanding of floral patterning has emerged primarily from studies in the core eudicot model plants Arabidopsis thaliana and Antirrhinum majus. In these species, the genetic ABCE model predicts how combinatorial expression of four classes of transcription factors specifies organ identity in the floral meristem (1-4). According to the latest Arabidopsis model, which incorporates the role of the E-class proteins, once flowering has initiated, A-and E-class proteins specify sepals; A-, B-, and E-class proteins specify petals; B-, C-, and E-class proteins specify stamens; and C-and E-class proteins specify carpels and terminate floral meristem development (2, 5, 6). The underlying biochemical mechanism for specifying organ identity has been described by the floral quartet model, which predicts that correct transcription of organ-specific genetic programs requires the formation of hetero-multimeric complexes between these four interacting classes of transcription factors (5,(7)(8)(9). Mutations affecting the class-A, -B, -C, and -E functions are homeotic, resulting in the replacement of one organ type by another. Loss of expression of the Arabidopsis C-class gene AGAMOUS (AG) results in conversi...

Fruit set and development are dependent on auxin, gibberellin, and cytokinin, which cause parthenocarpic development in many species when applied ectopically. Commercial sprays containing these hormones are used to improve apple fruit set, size, and shape, but have been implicated negatively in other aspects of fruit quality. We applied gibberellic acid (GA 3 ), synthetic auxin (NAA), and the auxin-transport inhibitor NPA to ‘Honeycrisp’ apple flowers. Fruit retention and size were quantified throughout development, and seed number and fruit quality parameters were measured at maturity. GA 3 alone caused the development of seedless parthenocarpic apples. At maturity, GA 3 -treated apples were narrower due to reduced ovary width, indicating that GA 3 induced normal growth of the hypanthium, but not the ovary. GA 3 -treated fruits were also less acidic than hand-pollinated controls, but had similar firmness, starch, and sugar content. To further understand the regulation of parthenocarpy, we performed tissue-specific transcriptome analysis on GA 3 -treated, NAA-treated, and control fruits, at 18 days after treatment and again at maturity. Overall, transcriptome analysis showed GA 3 -treated and hand-pollinated fruits were highly similar in RNA expression profiles. Early expression differences in putative cell division, cytokinin degradation, and cell wall modification genes in GA 3 -treated ovaries correlated with the observed shape differences, while early expression differences in the acidity gene Ma1 may be responsible for the changes in pH. Taken together, our results indicate that GA 3 triggers the development of parthenocarpic apple fruit with morphological deviations that correlate with a number of candidate gene expression differences.