Based on the similarities in sequence and function of a number of floral homeotic genes, it has been proposed that the molecular mechanisms controlling floral development are similar across dicots, and perhaps across all angiosperm species. Nonetheless, flowers display remarkable morphological diversity, and it is thought that particular floral organ types may have evolved more than once. In particular, petals are thought to have arisen numerous times in the angiosperms. These independent origins of petals imply that the molecular mechanisms specifying petal identity may be different in different species. This proposal seeks to explicitly test the hypothesis that the expression and function of the APETALA3 (AP3) and PISTILLATA (PI) genes which are known to control petal identity in higher eudicot species may have different roles in other angiosperm clades. Three sets of experiments are proposed to begin to understand how the functions of AP3 and PI orthologous genes may have been modulated over evolutionary time. First, expression analyses of AP3 and PI orthologs will be carried out in selected angiosperm species to see if changes in their expression correlate with the independent origins of petals. Second, recent work has identified another gene, TM6, which appears to be an AP3 paralog and may also function in petal development. Loss-of-function phenotypes in TM6 orthologs in Petunia and Nicotiana will be generated and characterized to determine the role of TMd-like genes. Third, recent phylogenetic analyses have identified conserved A.P3 and PI carboxy-terminal motifs; deletion and swapping experiments will be carried out in transgenic Arabidopsis to determine the role(s) of these conserved sequences.
