Splicing regulatory elements (SREs) are short, degenerate sequences on pre-mRNA molecules that enhance or inhibit the splicing process via the binding of splicing factors, proteins that regulate the functioning of the spliceosome. Existing methods for identifying SREs in a genome are either experimental or computational. This work tackles the limitations in the current approaches for identifying SREs. It addresses two major computational problems, identifying variable length SREs utilizing a graph-based model with de Bruijn graphs and discovering co-occurring sets of SREs (combinatorial SREs) utilizing graph mining techniques. In addition, I studied and analyzed the effect of alternative splicing on tissue specificity in human. First, I have used a formalism based on de Bruijn graphs that combines genomic structure, word count enrichment analysis, and experimental evidence to identify SREs found in exons. In my approach, SREs are not restricted to a fixed length (i.e., k-mers, for a fixed k). Consequently, the predicted SREs are of different lengths. I identified 2001 putative exonic enhancers and 3080 putative exonic silencers for human genes, with lengths varying from 6 to 15 nucleotides. Many of the predicted SREs overlap with experimentally verified binding sites. My model provides a novel method to predict variable length putative regulatory elements computationally for further experimental investigation. Second, I developed CoSREM (Combinatorial SRE Miner), a graph mining algorithm for discovering combinatorial SREs. The goal is to identify sets of exonic splicing regulatory elements whether they are enhancers or silencers. Experimental evidence is incorporated through my graph-based model to increase the accuracy of the results. The identified SREs do not have a predefined length, and the algorithm is not limited to identifying only SRE pairs as are current approaches. I identified 37 SRE sets that include both enhancer and silencer elements in human genes. These results intersect with previous results, including some that are experimental. I also show that the SRE set GGGAGG and GAGGAC identified by CoSREM may play a role in exon skipping events in several tumor samples. Further, I report a genome-wide analysis to study alternative splicing on multiple human tissues, including brain, heart, liver, and muscle. I developed a pipeline to identify tissue-specific exons and hence tissue-specific SREs. Utilizing the publicly available RNA-Seq data set from the Human BodyMap project, I identified 28,100 tissue-specific exons across the four tissues. I identified 1929 exonic splicing enhancers with 99% overlap with previously published experimental and computational databases. A complicated enhancer regulatory network was revealed, where multiple enhancers were found across multiple tissues while some were found only in specific tissues. Putative combinatorial exonic enhancers and silencers were discovered as well, which may be responsible for exon inclusion or exclusion across tissues. Some of the enhancers are found to be co-occurring with multiple silencers and vice versa, which demonstrates a complicated relationship between tissue-specific enhancers and silencers. / Ph. D.

京都大学 / 新制・課程博士 / 博士(医学) / 甲第24996号 / 医博第5030号 / 新制||医||1069(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 齊藤 博英, 教授 滝田 順子, 教授 小川 誠司 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

Pre-mRNA splicing is an important level in posttranscriptional gene regulation that is essential for accurate protein synthesis and generating protein diversity. The abundance of cryptic splice sites and long intronic DNA sequences makes their splicing a complex one. The identification of correct exons and introns needs additional information in the form of splicing regulatory elements (SREs) along with canonical splice signals. The interplay among these SREs and the trans factors (which bind to SREs) gives the identity to introns and exons which in turn leads to precise pre-mRNA splicing. Previous studies from our laboratory showed, that when expressed in mammalian cells from an expression vector, PDGF-B ORF was re-spliced at 4/5 exon junction with the downstream SV40 splice acceptor site in the vector. However, deletion of the 66-nt PDGF-B 3’ UTR region resulted in about 25% reduction in re-splicing. Sequence analysis of this region revealed presence of binding sites for splicing factors ASF/SF2 and SRp55, and an AU-rich element (ARE), mutation each of which affected re-splicing partially. In mammals, AREs are commonly found in the 3’UTR of mRNAs encoding proteins involved in diverse functions and are involved in selective mRNA degradation. Several ARE binding proteins are crucial for ARE’s function. Since mutation of the single ARE in the 3’UTR region altered the re-splicing efficiency, the role of AU-rich elements and ARE-binding proteins (AU-BPs) in modulation of splicing was investigated using siRNAs against AU-BPs, BRF1, hnRNPD, HuR, GAPDH and TTP. Down regulation of expression of these factors indeed affected the level of re-spliced product. We have studied the interactions between the full-length splicing factors (U1-70K and U2AF35) and the AU-BPs (BRF1, hnRNPD and HuR) as well as among the AU-BPs using three different assay methods: Yeast-two hybrid, co-immunoprecipitation and pull down assays. Our study has revealed that the BRF1 interacts with U1-70K and U2AF35 as well as the other AU-BPs hnRNPD and HuR but with different affinities. We have also analyzed the ability of AU-BPs to interact with SR proteins SRp20 and 9G8. We did find strong interaction of BRF1 with SRp20 and 9G8. Generation of a large number of nested deletion mutants of all the proteins allowed us to identify the interaction regions on the surface of BRF1, U1-70K, hnRNPD, U2AF35 and HuR. The results of Y2H analyses were further confirmed by pull down assay using purified interacting regions. It was found that a single region from aa 181-254 in BRF1 interacts with multiple partners i.e., splicing factors and the AU-BP hnRNPD. However, the RNA-binding zinc-finger domain from residue 120-181 independently interacts with HuR. Further, the multiple protein interacting region (MPIR) (aa 181-254) in BRF1 exhibits different affinities towards its interacting partners with that for U1-70K and hnRNPD being stronger than that for U2AF35 and HuR. This observation suggests that BRF1 activity can be modulated by interaction with different partners at different sites. U1-70K interacted only with BRF1 among the proteins tested in this study and this interaction appears to be RNA independent .This could have implications in splice site selection and RNA stability since BRF1 has been shown to promote RNA degradation. While the Arg/Glu-rich C-terminal region in U1-70K is sufficient for its interaction with BRF1, U2AF35 requires both the zinc-finger 2 and the arg/Gly/Ser-rich C-terminal regions for its association with BRF1. hnRNPD also interacts with multiple partners that include BRF1, HuR and U2AF35 using the N-terminal region that harbors a Ala-rich domain. The interaction of hnRNPD with HuR is RNA dependent while with BRF1 and U2AF35, it is RNA independentt. Further, its interaction with all the partners is equally strong. This suggests that hnRNPD could exert differential influence depending on the context of its interaction and abundance of the interacting partner. HuR, primarily known as an mRNA stabilizing factor, interacts with both BRF1 and hnRNPD with equal affinity involving the hinge region, the interaction with the former being RNA-independent and the later being RNA-dependent. This differential RNA-dependent and independent interactions with the two AU-BPs using a single interacting domain suggests a balancing act of HuR on the activities of BRF1 and hnRNPD. These interactions can further be differentially modulated by posttranslational modifications on one or all of the interacting partners depending on the physiological status of the cell. We have also analyzed the multiple protein complexes formed in absence of cellular RNA. Though we are unable to see direct protein-protein interaction between HuR and U1-70K in Yeast two hybrid analysis, we could detect the presence of U1-70K in HuR immunoprecipitate. It appears that U1-70K associates with HuR via BRF. We also detected the presence of HuR in U1-70K complexes which could be due to its association with BRF1. We are unable to find hnRNPD and U2AF35 in these complexes indicating that they may have been excluded. In anti-U2AF35 immunoprecipitates, we detected the presence of U1-70K as well as hnRNPD but no HuR. This may be due to RNase treatment as hnRNPD and HuR interactions are RNA dependent. Our findings that AU-rich elements in conjunction with AU-BPs function as intronic splicing modulators or enhancers, reveal hitherto unidentified new players in the poorly understood complex mechanisms that mediate alternative splicing. The possibility of dynamic nature of the interactions among splicing factors and AU-BPs mediated by post-translational modifications provide a basis for rapid cellular responses to changing environmental cues through generation of differentially spliced mRNAs and corresponding protein products that differ in their stability and hence their relative abundance. Our results also unfold enormous possibilities for future investigations on interactions among the many splicing factors and AU-BPs, and in understanding these complex interactions in modulation of pre-mRNA splicing, mRNA translation and degradation. The finding of coupling of AU-BPs to splicing machinery could further lead to better understanding of the mechanism of AU-BP-mediated targeting of mRNAs to processing bodies and ultimate degradation of the mRNAs.