Duplication and Functional Divergence in the Floral Organ Identity Genes

Abstract

Among all extant land plant lineages, the flowering plants, or angiosperms, are by far the most abundant and diverse. A leading force in their speciation has been attributed to morphological changes in flowers, a key reproductive innovation. The floral organ identity genes are thus a logical focus to gain insight into the angiosperm radiation, since alterations to both the structure and regulation of developmental genes are known to underlie morphological diversity. Genetic studies on flower development in the model plants Arabidopsis and Antirrhinum have yielded the ‘ABC model’, which predicts that the four types of floral organs (sepals, petals, stamens and carpels) are specified by the products of four gene classes, A, B, C, and the subsequently added E class, that act in combination. All but one of these transcription factors belong to the MADS-box family, their function analogous to the metazoan body patterning of HOX genes. Plant genomes are prone to duplication and paralogs are particularly widespread in the floral organ identity genes. My aim was to elucidate the function of representatives of an early clade of duplicated floral organ identity genes, with a focus on determining the degree of redundancy vs. divergence amongst them and of conservation in relation to the later-diverging models. To that end, I use the species T. thalictroides, from the clade sister to all other eudicots, with two C class gene orthologs and three B class gene orthologs. In the C class genes, one paralog showed deep conservation of the dual C class function, affecting both reproductive organ identity and meristem determinacy, while the second revealed specialization to ovule identity. This sub-functionalization is unlike the fate of the B class genes following duplication, which retained partial redundancy, showed deep conservation in stamen identity, and gained a novel role in ectopic petaloidy. Taken together, these studies illustrate the different fates that floral organ identity gene paralogs can assume following duplication that can ultimately be relevant to the generation of biodiversity. Moreover, they underscore the need to survey the functional relevance of all branches of a gene’s evolutionary history.