|dc.description.abstracteng||In eukaryotes, most protein-coding genes are interrupted by non-coding sequences known as introns. The splicing of intronic sequences from a pre-mRNA is catalyzed by the spliceosome, an elaborate molecular machine formed by the interaction of five snRNPs and numerous splicing factors with the pre-mRNA. Initial assembly of the spliceosome can occur across an intron sequence (intron definition) or alternatively across an exon (exon definition). The latter is likely the prominent pathway for pre-mRNAs with long intronic sequences, which is characteristic for most human pre-mRNAs. As splicing catalysis can only occur across introns, a rearrangement from the exon-defined to intron-defined state is required. This rearrangement can be a critical point during the decision whether to include or to skip an exon in the mature mRNA during the process of alternative splicing. Recently exon-defined spliceosomes were shown to contain not only U1 and U2 snRNPs, but also the U4/U6.U5 tri-snRNP (37S exon complex), and evidence was provided that it is possible for a cross-exon complex to be converted directly into an intron-defined B complex.
During my studies, I addressed the question what are the requirements for stable integration of the U4/U6.U5 tri-snRNP and formation of a pre-catalytic B complex during exon- and intron-defined spliceosome assembly, since this is a crucial point in which the alternative assembly pathways converge.
To address this aim, I used a reductionist in vitro system to assemble the 37S exon complex on a single exon-containing RNA substrate in HeLa nuclear extract. The addition in trans of an RNA oligonucleotide containing a 5’splice site sequence (5’ss oligonucleotide) mimics an adjacent upstream 5’ss and induces the formation of a 45S B-like complex, which shares similarities with the intron-defined B complex. The transition from a 37S exon to a 45S B-like complex is accompanied by the stable integration of the U4/U6.U5 tri-snRNP within the complex and a significant shift in the sedimentation behavior, suggesting structural changes in the complex during its stabilization. Indeed, electron microscopy (EM) revealed major structural differences between 37S exon and 45S B-like complexes, likely due to a different orientation of the U2 snRNP and the U4/U6.U5 tri-snRNP with respect to each other, while the structure of the 45S B-like complex was highly similar to that of an intron-defined B complex. Thus, formation of a B-like complex and consequently the stable integration of the U4/U6.U5 tri-snRNP is indeed accompanied by significant structural remodeling of the spliceosome.
To identify factors that might contribute to the structural remodeling and the stabilization of U4/U6.U5 tri-snRNP binding, we identified highly-abundant proteins associated with the 37S exon and 45S B-like complexes by 2D gel-electrophoresis. The comparison of their protein compositions showed the recruitment of a distinct set of B complex-specific proteins, namely RED, MFAP1, FBP21, hSmu-1, hPrp38 and hSnu23, upon formation of the 45S B-like complex. These proteins are also recruited during formation of the intron-defined B complex, underlining the compositional and structural similarities of the 45S B-like and B complex. However, the B complex-specific proteins do not appear to contribute to stable U4/U6.U5 tri-snRNP integration, since solely the interaction of the 5'ss oligonucleotide with affinity-purified 37S exon complexes in the absence of splicing extract triggers the structural rearrangement that supports stable U4/U6.U5 tri-snRNP integration.
The defined changes in structure, stability and protein composition during transition from a 37S exon to 45S B-like complex are induced solely by the addition of a 5’ss oligonucleotide to the splicing reaction. Thus, I investigated the requirements for functional interaction of a 5’ss sequence with components of the U4/U6.U5 tri-snRNP to induce the formation of a stable 45S B-like complex. Studies using 5’ss oligonucleotides with mutated sequences indicated that base pairing with the ACAGAG box motif of U6 snRNA is not required to trigger the observed structural rearrangement. But instead, highly conserved guanine residues at the exon-intron junction are essential for the formation of a stable 45S B-like complex and I could show that these residues within the 5’ss oligonucleotide are contacted only by the U5-specific protein hPrp8, pointing to a crucial role of hPrp8 in stabilizing the binding of the U4/U6.U5 tri-snRNP. Further, my studies provide evidence that the stable integration of the U4/U6.U5 tri-snRNP not only depends on the recognition of the exocyclic part of these guanines, but also on interactions with the ribose backbone of the RNA.
Based on the results in the exon-defined assembly pathway and the similarities between the 45S B-like and B complex, I assumed that stable incorporation of the U4/U6.U5 tri-snRNP during intron-defined B complex formation also requires its interaction with the 5'ss of the pre-mRNA. Consequently, I set out to stall intron-defined spliceosome assembly just prior to stable B complex formation in order to test if the stable integration of the U4/U6.U5 tri-snRNP during B complex formation also relies on the interaction with a 5’ss. The DEAD-box helicase hPrp28 is involved in the displacement of U1 snRNP from the 5'ss and was recently shown to be essential for the formation of a stable intron-defined B complex. Thus, I chose hPrp28 as a target to inhibit spliceosome assembly prior to formation of a B complex. The use of a dominant-negative mutant of hPrp28, lacking its ATPase activity, inhibited B complex formation, but not formation of the A complex, and allowed me to affinity-purify a novel intermediate in the intron-defined spliceosome assembly pathway, namely the 37S cross-intron complex. Characterization of this complex showed that it contains all five snRNPs, but in contrast to the pre-catalytic B complex, the U4/U6.U5 tri-snRNP is not yet stably-integrated. Thus, in the absence of hPrp28 function, the U4/U6.U5 snRNP can associate with the spliceosome, while U1 snRNP is still present in the complex. Psoralen-mediated RNA-RNA crosslinking showed that in the 37S cross-intron complex U1 snRNA is base paired with the pre-mRNA, while the U6 snRNA establishes base pairing interactions with the U2 snRNA. In the B complex, no base pairing interactions between U1 and the pre-mRNA were detected. Instead we detected interactions between the U4/U6.U5 tri-snRNP and the pre-mRNA, while the U4 and U6 snRNA were still base paired. These data suggest that due to the retention of U1 snRNP at the 5’ss of the pre-mRNA, this sequence is not available for interaction with the U4/U6.U5 tri-snRNP and thus the transition to a pre-catalytic B complex is inhibited. Investigations of the protein composition of the 37S cross-intron complex by mass spectrometry and 2D gel-electrophoresis revealed distinct differences compared to the B complex, in particular the lack of the B complex-specific proteins. Surprisingly, the 37S cross-intron complex is compositionally very similar to the 37S exon complex, indicating that the exon- and intron-defined assembly pathways both involve a similar intermediate where U4/U6.U5 tri-snRNP has docked, but is not yet stably-integrated. However, structural investigation of the 37S cross-intron complex identified differences in its appearance in comparison to the 37S exon complex, suggesting that despite their similar protein inventory, the organization of these complexes is different.
The addition in trans of an accessible 5’ss in the form of a short RNA oligonucleotide to the splicing reaction resulted in the formation of a stable B complex, even in presence of the dominant-negative hPrp28 mutant. These results show that the 37S cross-intron complex is competent for stable B complex formation, but due to the lack of an accessible 5’ss that can interact with the U4/U6.U5 tri-snRNP, the stable integration of the latter cannot occur. The stable integration of the U4/U6.U5 tri-snRNP is also supported by addition of the 5’ss sequence to an affinity-purified 37S cross-intron complex, showing that all factors required for the stable integration of the U4/U6.U5 tri-snRNP are already present in the this complex. EM analyses of this stabilized cross-intron complex revealed structural features similar to those of the 37S exon complex after addition of the 5’ss oligonucleotide, showing that the conformational remodeling that occurs during stable U4/U6.U5 tri-snRNP binding, leads to a nearly identical architecture of both complexes. These results indicate that the exon- and intron-defined assembly pathway of the spliceosome converge at the stage where the 37S complexes become committed to a 5’ss sequence via interaction with the U4/U6.U5 tri-snRNP. In summary, our studies provide new insights into the mechanisms underlying the stable association of the U4/U6.U5 tri-snRNP during B complex formation, which is a prerequisite for the catalytic activation of the spliceosome.||de