Seed Origin And Evolution Of Developmental B3 Gene

Seed formation is one of the key innovations during land plant evolution, which endowed plants with higher adaptability toward unfavorable environment and strengthened the ability of dispersal and competition. The process of seed development could be divided into two phases, i.e. the early embryo morphogenesis phase and the late maturation phase. During the late maturation phase, the embryo accumulates various storage compounds and acquires desiccation tolerance, and eventually enters dormancy.B3 genes encode plant-specific transcription factors, which play a variety of roles in plant growth and development. Proteins encoded by B3 genes vary in sequence similarity and domain structure but all contain a highly conserved B3 domain, forming the B3 superfamily. This superfamily consists of several distinct families, including the LAV, ARF, RAV and REM families, with a few subfamilies to be identified in each family. The AFL subfamily belongs to the LAV family, containing three members:ABSCISIC ACID INSENSITIVE 3(ABI3), FUSCA 3 (FUS3) and LEAFY COTYLEDON 2 (LEC2). The three members of the AFL subfamily are considered master genes in controlling seed development, especially for seed maturation. It remains unclear, however, whether all these genes are necessary and sufficient for induction of the maturation phase of seed development for all seed plants. The questions such as whether these genes appeared at the same time as the origin of the seed plants and how they originated remain unknown.In this study, genome-wide screens for the AFL genes were performed based on currently available plant genome sequences. By conducting a comprehensive analysis of the distribution patterns, gene structures and phylogenetic relationships of ABI3, FUS3 and LEC2, we intend to address questions, such as when these genes originated, how they duplicated and diverged during evolution, and whether all these genes are indispensable components of the regulatory network involved in the control of seed development for all seed plants. We also cloned and characterized the AFL homologous genes from several gymnosperm plants, for which the complete genome sequences were not available.The results showed that the AFL subfamily genes are widely distributed in land plant genomes, and that the AFL genes underwent a series of independent duplication and diversification during land plant evolution. The duplicate ABI3 genes of non-seed plants seem to have structurally and functionally diverged prior to the diversification of seed plants, leading to the reduction of the numbers of ABI3 in the seed plant genomes and giving rise to the appearance of the FUS3 gene in gymnosperms. In accompany with the divergence of gymnosperm and angiosperm, the FUS3 gene experienced a new round of duplication with subsequent diversification through neo-and/or sub-functionalization. As a result, one copy retained the original function of FUS3 in both gymnosperm and angiosperm genomes and the other copy diverged into the ancestors of LEC2 and IDEF-like. Present data suggested that the LEC2 gene appeared initially and limitedly distributed in the dicot genomes, while the functionally diverged IDEF-like gene limited to the monocot genomes. Therefore, the ABI3, FUS3 and LEC2 genes have appeared in the land plant genomes in a stepwise pattern. There is a clear trend toward the reduction of the conserved domain contained in ABI3, FUS3 and LEC2. Compared to the wide range of expression pattern of ABI3, FUS3 and LEC2 exhibit restricted spatial and temporal expression. Functional diversification of the AFL genes seems to correlate with the variation in domain structure.ABI3, FUS3 and LEC2 are currently considered the master regulatory genes for seed maturation. Based on the currently available data, the LEC2 genes are only distributed in the dicot genomes. It is thus doubtful whether all the ABI3, FUS3 and LEC2 genes are indispensable components of the regulatory network controlling seed development for all seed plants. To answer this question, we need more complete genome sequences, especially the genomic information of gymnosperms. Meanwhile, the evolutionary relationships between LEC2 and IDEF remain to be further assessed. These two genes probably descended from different ancestors that appeared at the early stage of angiosperm diversification, with the LEC2 and IDEF homologs sequences to be lost in the monocot and dicot genomes, respectively. But we also cannot exclude the possibility that IDEF was derived from LEC2 through duplication and subsequent diversification during the monocots-dicots split. To investigate when these genes originated and how they diverged during evolution will provide crucial insights into the molecular mechanisms controlling seed formation and evolution.