The Role of Gene Activity in Governing the Development and Evolution of Seeds

The angiosperm seed is an inherently complex hybrid structure, made up of three major regions differing in genotype and ploidy. It consists of the two products of double fertilization: the diploid embryo and the triploid endosperm, and the maternally derived seed coat. This complex structure represents a highly successful propagule and partially accounts for the evolutionary success of the angiosperms. The goal of my dissertation is to characterize the patterns of gene activity involved in seed formation on both evolutionary and developmental time scales in order to determine how the seed is formed.;Thanks to newly available genome sequences of basal land plants I have been able to identify, clone and characterize a lycophyte homolog of LEAFY COTYLEDON 1, a key regulator of angiosperm seed development. The presence of this key developmental gene in the lycophytes, a group of non-seed plants that diverged ∼380 million years ago, provides insight into the process by which existing developmental processes were repurposed during the evolution of the angiosperm seed habit.;My dissertation also focuses on the development of the angiosperm seed by identifying the patterns of gene activity among genes that are active in seven subregions of the Arabidopsis seed over five distinct stages of development. These data represent the first characterization of gene activity for several seed subregions and highlight the degree to which subregions are differentiated to carry out specific developmental processes. Based on these patterns of gene activity, I explore the mechanism of specification for one of the most distinctive seed subregions, the chalazal endosperm.;The chalazal endosperm occupies a similar position within the seed as the antipodal cells of the female gametophyte occupy within the ovule. Previous accounts indicate that gene activity specifically marking the antipodals will often mark the chalazal endosperm as well. On this basis, I examined the possibility that mechanisms known to convey positional identity in the ovule are conserved and within seed development and found significant differences between the two processes.