Pre-mRNA splicing is an essential step in eukaryotic gene expression. Splicing is catalyzed by the spliceosome, a large and highly dynamic macromolecular machine in the cell. Errors in splicing have been linked to many diseases, including cancer. The spliceosome is responsible for precisely joining the protein coding sequences that will be translated to protein. It is composed of five small nuclear RNA (U1, U2, U4, U5, and U6) along with over 100 unique proteins. Understanding the RNA/RNA and RNA/protein interactions that position the pre-mRNA substrate in the spliceosome for the two catalytic steps of splicing is key to determining the molecular mechanisms of this dynamic complex.

In this thesis, we compared two states of the spliceosome complexes, catalytic and post-catalytic. We investigated the role of pre-mRNA sequences in the spliceosome complex stalled prior to the second step of splicing. We next developed a new strategy that traps the spliceosome after the second step of splicing, but before the mRNA is released. We show that structural rearrangements of the RNA substrate exist as the spliceosome transitions from the first to second step of splicing. Using mass spectrometry assays, we identify proteins that stably associate with affinity-purified catalytic and post-catalytic complexes. We show enrichment of SF3B components and loss of an ATPase, DHX35, in the post-catalytic complex. In conjunction with biochemical studies, we use electron microscopy to determine structures of both complexes to further gain insights into the spliceosome's role in splicing. This study serves as a foundation for understanding the mechanisms underlying the later stages of spliceosomal progression.

In eukaryotes, a crucial step in gene expression is pre-mRNA splicing by the spliceosome. The spliceosome is a macromolecular machine that removes intervening intron sequences from over 95% of human transcripts to create a functional template for protein synthesis. Precise splicing is essential for correct gene expression, and mutations in the spliceosome are associated with diseases including cancer. Mass spectrometry has identified over 100 spliceosome proteins, but deciphering the function for the vast majority of them has been challenging due to the highly dynamic and complex nature of the spliceosome. My dissertation focuses on small-molecule inhibitors as tools to expand the currently limited mechanistic understanding of the spliceosome:

First, I developed a high-throughput assay to rapidly and efficiently screen large compound libraries for small molecules that inhibit splicing. I found three new splicing inhibitors and determined their effect on spliceosome assembly and splicing chemistry. Second, I used synthetic inhibitor analogs for structure-activity studies to identify chemical groups that are responsible for compound activity. I found that the same structural features are required for in vitro and in vivo splicing inhibition, and that the cellular response was mainly due to inhibition of the spliceosome. Third, I used biochemical assays to show that three structurally distinct inhibitors bind to the same site of the spliceosome core protein SF3B1, and that SF3B1 has a functional role throughout the multi-step splicing process. SF3B1 is of particular interest because it is often mutated in cancer, which makes it a promising target for the development of novel chemotherapeutics.

My work provides a fast, reliable assay to identify small-molecule splicing inhibitors, a series of in vitro and in vivo assays to characterize their effect on complex assembly, splicing chemistry, and cellular phenotype, and an excellent example of how inhibitors can be utilized to decipher the function of spliceosome proteins. In addition to expanding the mechanistic model of one of the most complex macromolecular machines in the cell, identifying the function of individual spliceosome parts in healthy situations is the necessary first step to determine how changes of these functions in aberrant situations can lead to cancer.

Human gene expression is a complicated process, and when things go wrong, disease may arise. Many diseases such as cancer arise from abnormal gene expression, so understanding the components of our cells that help facilitate normal gene expression is essential. My research focuses on the component known as the spliceosome, a very dynamic and complex molecular machine of ribonucleic acids and proteins. Splicing is a choreographed routine as there are many moving components. SF3B inhibitors target components of the spliceosome and interfere with spliceosome assembly and stability. I hypothesize that the inhibition and instability may be linked to differences in RNA arrangements in the inhibitor stalled complex and A-complex. My dissertation focuses on three goals: First, I investigated the connection between structural rearrangements in U2 snRNA and the mechanism of SF3B inhibitors that target the U2 snRNP. Second, I characterized the intrinsic capabilities of U2 snRNA to adopt different secondary structures. Last, I employed a bimolecular exon ligation reaction to test if SF3B inhibitors interfere with exon ligation. My studies have demonstrated that SF3B inhibitors do not affect U2 snRNA structure in the spliceosome and the snRNP, and the U2 snRNA has an intrinsic ability to adopt alternate and possibly novel structures than previously predicted. The basic research I conduct will ultimately inform therapeutic research to generate treatments for diseases.

Pre-mRNA splicing is critical for controlling normal gene expression within the cell. Several splicing associated RNA helicases and their binding partners are misregulated in a variety of cancers including leukemia, prostate, glioma, and non-small-cell lung cancer. Some of the RNA helicases from the DHX-family are involved in multiple cellular pathways, and to define their specific activity they rely on binding partners that contain G-patch domains. For instance, DHX15 has been characterized for its role in spliceosome disassembly after splicing and ribosome biogenesis directed by G-patch proteins TFIP11 and NKRF, respectively. My dissertation focuses on a new model for DHX15 in which it ensures fidelity of early spliceosome assembly. I hypothesize that DHX15 acts in a quality-control (QC) mechanism to disassemble aberrant spliceosomes. Due to the dynamic nature of the spliceosome, in that intronic landmarks are divergent as well as the multitude of rearrangements required to remove an intron, mistakes are bound to happen and the lack of a QC step during early assembly has long been a mystery. To address the role that DHX15 may be playing during early spliceosome assembly, I utilized a combination of RNA substrates and energy sources (ATP, GTP and AMP-PNP) to parse apart the different energy-dependent steps of A-complex assembly. Then through immunodepletion asked how removal of DHX15 affects the system. In attempting to determine where DHX15 binds within our complexes, I utilized a system of truncated RNA substrates and endogenous RNase H digestion of the U2 snRNA. Through this work, I found that depletion of DHX15 results in an accumulation of A- and B-complexes yet results in a decrease in splicing efficiency. Further, certain regions of the U2 snRNA are prone to NTP-dependent rearrangements, whereas others are protected or unaffected. In addition, the work in this thesis solely relied on HeLa cell nuclear extract and protein levels including DHX15 vary across cell types. Therefore, I worked to develop protocols and understand why other cell lines have yet to yield robust activity in nuclear extract in addition to why we observe inconsistent activity between HeLa nuclear extract preps. Finally, to understand how DHX15 may be impacting early complex formation, I devised a purification scheme to allow for early complexes to be purified for further study. Taken together, the data produced during my graduate career and described in this dissertation has yielded another path of study within the spliceosome and may provide a scaffold for future therapeutics in targeting diseases such as cancers and myelodysplastic syndromes.

The spliceosome is a dynamic ribonucleo-protein complex that is in charge of removing introns from pre-mRNA. Five small nuclear ribonucleic proteins (snRNPs) (U1, U2, U4, U5, U6) build the spliceosome, and each contain their own special RNA. In particular, U2 snRNP is unique, in that it guides the spliceosome to the correct branch point sequence. SF3B1 is the largest and keystone protein of U2 snRNP. Mutations in SF3B1 have been associated with misregulation in splicing and disease such as cancer. Spliceostatin A, Pladienolide B and herboxidiene are small drug molecules that target SF3B1. They are used as tools to help study the importance of SF3B1 and initiation of splicing assembly. However, the mechanism of action of these molecules is not known. More studies need to research how these drugs bind and inhibit. I used in vitro order of addition splicing assays, and studied SF3B1 in different conditions: temperature, time, and competition assay. My findings clearly demonstrate that temperature affects competition between active and inactive compounds. Time course experiments clearly imply that SF3B inhibitors exchange more slowly than inactive analogs. Lastly, we show positions C1 and C6 of herboxidiene are important for binding and inhibiting SF3B1 in the binding pocket. We also model drug-protein interaction. Our work will help elucidate how SF3B1 selects the correct branch point sequence and give medicinal chemists new insights to make small molecule drugs more potent.