UCLA

The step-wise assembly of the spliceosome onto pre-mRNA occurs co-transcriptionally, while the nascent transcript is synthesized from RNA Polymerase II. The mechanisms underlying the molecular coupling and coordination of these co-transcriptional reactions have not been thoroughly elucidated in the context of the dynamic chromatin environment. The Johnson Lab previously discovered that the major yeast histone acetyltransferase (HAT) Gcn5 demonstrates genetic interactions with MSL1 and LEA1—which encode two core U2 small nuclear ribonucleoprotein particle (snRNP) components of the spliceosome. Additionally, the lab observed that Gcn5 HAT activity is required for proper co-transcriptional recruitment of the spliceosome in yeast where U2 snRNP association with the pre-mRNA and subsequent spliceosomal rearrangements are sensitive to Gcn5-dependent acetylation. While Gcn5-dependent histone acetylation is important for the fate of spliceosomal rearrangements of these two proteins, the genome-wide implications and overall mechanism underlying this result is not yet clear. Here, I employed RNA-seq in S. cerevisiae to identify mechanistic insights into how Gcn5-dependent histone acetylation can affect pre-mRNA splicing. I prepared libraries from the following yeast strains: wildtype, gcn5Δ and H3Δ9-16 (deletion of residues 9-16 from histone H3 N-terminal tail) such that genes for which splicing is similarly affected in both gcn5Δ and H3Δ9-16 will represent a defect specific to the absence of Gcn5-histone acetyltransferase (HAT) activity. Surprisingly, the results support a role for the Gcn5-HAT in decreasing gene expression of all intron-containing ribosomal protein genes (IC-RPGs). Consequently, gcn5Δ and H3Δ9-16 improved the splicing efficiency of a subset of IC-non-RPGs in a manner regulated by the competition of IC-RPGs for limited spliceosomes. Lastly, gcn5Δ and H3Δ9-16 resulted in the decreased splicing efficiency for another subset of IC-non-RPGs. With publicly available ChIP-seq and MNase-seq data, I show that despite the increased availability of limited spliceosomes via down-regulation of IC-RPGs, gcn5Δ and H3Δ9-16 dependent splicing outcomes of IC-non-RPGs are distinguished by differences in RNAPII, H3K9ac and MNase enrichment profiles. In this thesis, I uncover multiple effects of the gcn5Δ and H3Δ9-16 on pre-mRNA splicing genome-wide--encompassing mechanisms regarding the economics of limited spliceosome availability and distinct chromatin landscapes directly regulating splicing outcomes.