Our understanding of plant developmental, biochemical and physiological pathways has been greatly enhanced over the last decade by work on the small crucifer Arabidopsis thaliana. In particular, the molecular genetic analysis of floral development has led to a detailed ‘ABC’ model of organ identity during flower determination. However, it is not clear how well these models in Arabidopsis (a higher eudicot) can be extended to other distantly related plant species. To address this issue, Kramer and Irish 1 Kramer E.M. Irish V.F. Evolution of genetic mechanisms controlling petal development. Nature. 1999; 353: 31-37 Google Scholar examined the expression of two B class genes, APETALA3 (AP3) and PISTILLATA1 (PI) in flowers of several species of the lower eudicot subclass Ranunculidae. These species, include two from the family Papaveraceae (Iceland poppy and bloodroot), one from the family Fumariaceae (fringed bleeding heart) and two from the family Ranunculaceae (buttercup). Gene- and genome-duplication events both seem to have contributed to the diversification of the AP3 and PI lineages within the Ranunculidae. Thus, expression patterns were examined for all gene family members in these lineages. In developing stamen primordia, the temporal and spatial expression patterns of RNA and protein were similar between the Ranunculidae and higher eudicot (Arabidopsis and Antirrhinum) AP3 and PI orthologs. However, temporal and spatial patterns of AP3 and PI expression in the Ranunculid species in petal primordia were very different from those found in Arabidopsis and Antirrhinum: AP3 and PI were both expressed in young petal primoridia, but expression diminished during petal development, eventually becoming restricted to the extreme edges of the petal. In Arabidopsis and Antirrhinum, the continual expression of AP3 and PI throughout petal development is essential for the development of normal petal form. As the authors suggest, these differences in gene-expression patterns indicate that the ABC model is not rigidly conserved across Angiosperms and probably reflects the independent derivation of petals from either stamens or sterile bracts. Alternatively, PI might play a similar role in higher and lower eudicots during the initiation of petal primordia, but could have been replaced by novel B group genes that would function to direct petal differentiation during the later stages. Clearly, the evolution of floral pattern is not quite as simple as A,B,C.
Read full abstract