Changes in Gene Expression Underlie the Diversification of the LORELEI Gene Family in the Brassicaceae

Abstract

Sexual reproduction in flowering plants relies on cell-cell interactions between the pollen grain and the pistil, the female reproductive organ. After pollination, the pollen grain develops a pollen tube to transport the two immotile sperm cells through the pistil tissues to fertilize the female gametes contained within a female gametophyte in an ovule. The first direct contact between the male gametophyte and the female gametophyte occurs during pollen tube reception, which involves pollen tube growth arrest and release of sperm cells. Since pollen tube reception is a key pre-zygotic barrier in interspecific crosses, studying the genes that mediate this process can assist in establishing or overcoming species barriers. LORELEI (LRE) is a putative glycosylphosphatidylinositol (GPI)-anchored membrane protein important for pollen tube reception in Arabidopsis thaliana. LRE and its most closely related paralog, LORELEI-LIKE GPI-ANCHORED PROTEIN 1 (LLG1), arose from the alpha whole genome duplication event in the Brassicaceae. By binding to FERONIA, a member of the receptor-like kinase family first discovered in Catharanthus roseus, LRE and LLG1 perform similar molecular functions; still, lre and llg1 mutants do not show overlapping mutant phenotypes. In the first chapter of my dissertation, I hypothesized that LRE and LLG1 were retained because they split the expression patterns of the ancestral single copy gene (LRE/LLG1). To test this hypothesis, I studied the expression of LRE, LLG1, and the single copy LRE/LLG1 ortholog in Cleome violacea. I determined that regulatory subfunctionalization of the ancestral expression domains was one mechanism to explain why LRE and LLG1 were maintained after gene duplication. To investigate if LRE and LLG1 also split the molecular functions of the ancestral LRE/LLG1, we performed reciprocal complementation experiments and found that LRE and LLG1 can perform each other’s functions. Although the single copy LRE/LLG1 ortholog in Cleome violacea could fully complement llg1 phenotypes, it only partially complemented lre mutant phenotypes. This finding suggested that the LRE and LLG1 clades may have experienced differences in selective pressures. In the second chapter of my dissertation, I showed that the branch leading to the LRE clade experienced positive selection while the branch leading to the LLG1 clade experienced purifying selective pressures. By exploring the amino acid residues in LRE that are under positive selection, I identified a residue in the LRE clade that showed a lineage-specific conservation pattern. I used this residue to uncover a previously uncharacterized role for LRE in sensing of approaching pollen tubes by the synergid cells. This residue can be useful in testing if LRE acquired new functions (neofunctionalization) post gene duplication. Another prediction from the findings in the first chapter is that changes in cis-regulatory elements in the promoters of LRE and LLG1 genes underlie the differences in the expression of these two genes. In my final chapter, we identified a motif that was present in the LRE promoter but not in the promoters of its paralogs. We called this motif the Short Evening Element-Like (SEEL) motif. I showed that SEEL motif is important for synergid cell expression of LRE. I also investigated the importance of the MYB-related REVEILLE transcription factor family in A. thaliana in regulating LRE expression in synergid cells, as these transcription factors are known to bind to the SEEL motif. I found that mutants of synergid cell-expressed REVEILLEs showed reduced LRE expression. My dissertation provides insight into how regulatory subfunctionalization can cause diversification of regulatory elements of gene duplicates, which may lead to evolution of new functions post gene duplication.