| dc.description.abstracteng | The spliceosome is a highly dynamic, multi-MDa eukaryotic RNA-protein (RNP) machinery that catalyzes precursor messenger RNA (pre-mRNA) splicing. Pre-mRNA splicing entails the excision of non-coding introns and the joining of the neighboring coding exons via two consecutive transesterification reactions. For each catalytic cycle, a spliceosome is assembled on a substrate pre-mRNA by the stepwise recruitment of five small nuclear (sn) ribonucleoproteins (RNPs) and numerous non-snRNP factors. In contrast to ribosomal subunits, for example, none of the snRNPs or non-snRNP complexes contain a pre-formed catalytic center for the splicing catalysis. Instead, the active center of the spliceosome is formed anew during each spliceosome assembly cycle.
Spliceosome assembly occurs stepwise via several discrete intermediates that have been experimentally defined. During the transition from one assembly stage to the next, the spliceosome undergoes profound compositional and conformational remodeling. These remodeling events are driven and controlled by eight highly conserved Superfamily (SF) 2 RNP remodeling enzymes. In particular, an initial assembly containing all snRNPs (the so-called B complex) is still catalytically inactive and requires major rearrangements of its RNA-RNA, RNA-protein and protein-protein interaction networks in order to produce a catalytically competent spliceosome. Spliceosome catalytic activation requires a 650 kDa sub-complex that is part of the U5 snRNP and is composed of a large scaffolding protein, Prp8, a G-protein, Snu114, and a Ski-2 RNA helicase, Brr2. The molecular mechanisms underlying spliceosome catalytic activation are poorly understood. To elucidate the architecture of the complex formed by Prp8, Snu114 and Brr2 we aimed at recombinant reconstitution of this micro-machinery. We managed to successfully co-express human (h) Prp8, hSnu114 and hBrr2 in insect cells. However, we were not able to co-purify all the components. Only hBrr2 could be efficiently purified, indicating that under these working conditions, hBrr2 did not stably interact with hPrp8 and hSnu114. Using an ortholog screening approach, we tried to co-express the yeast (y) Prp8-ySnu114 sub-complex. Although yPrp8 was poorly expressed, we were able to co-purify small amounts of yPrp8 with ySnu114. Although these two proteins form a stable complex that can be purified, size exclusion chromatography revealed that the complex was possibly aggregated and unsuitable for further structural analysis since it migrated in the void volume of the column.
While our initial strategy of co-expressing Prp8, Snu114 and Brr2 as a complex failed, we have succeeded in isolating human Brr2. Brr2 is an essential RNA helicase needed for U4/U6 di-snRNP disruption during spliceosome catalytic activation. Brr2 is also the only spliceosomal helicase that is permanently associated with the spliceosome and thus requires faithful regulation. Concomitantly, Brr2 represents a unique subclass of SF2 nucleic acid helicases, containing tandem helicase cassettes. Presently, the mechanistic and regulatory consequences of this unconventional architecture are unknown. Henceforth, we then aimed at producing highly purified and homogeneous human and yeast Brr2 for further structural and functional investigations.
Full length human and yeast Brr2 could be expressed and purified to near homogeneity. Both enzymes were active in ATP-dependent U4/U6 duplex unwinding but failed to crystallize. In order to remove putatively flexible regions that may hinder crystallization, we treated hBrr2 and yBrr2 with proteases, several of which gave rise to a protease-resistant ca. 200 kDa fragment encompassing the two helicase cassettes. One of the six truncated hBrr2 proteins, whose borders were designed based on the proteolysis experiments, crystallized readily and the crystals diffracted to 2.65 Å resolution. Elucidation of the crystal structure and biochemical analyses showed that in hBrr2 two ring-like helicase cassettes intimately interact and functionally cooperate. Only the N-terminal cassette harbored ATPase and helicase activities in isolation. Structural comparisons and mutational analyses suggested that the N-terminal cassette of hBrr2 threads single-stranded RNA through a central tunnel and across a helix-loop-helix domain during duplex unwinding. While the C-terminal cassette did not seem to engage RNA in this fashion, it bound ATP and it strongly stimulated the N-terminal helicase. Stimulation depended on two inter-cassette communication lines, disruption of which affected ATPase and helicase activities in diverse ways. Additionally, mutations at the C-terminal ATP pocket affected the crosstalk between the two cassettes, suggesting that ATP binding may induce a specific C-terminal cassette conformation that solidifies important inter-cassette contacts. Using pre-steady state kinetics, we also probed the nucleotide binding preferences and worked out possible nucleotide binding mechanisms of either cassette, confirming that the C-terminal cassette strongly binds ATP in solution. Taken together, our results revealed the structural and functional interplay between two helicase cassettes in a tandem SF2 enzyme and suggested how Brr2 interactors may exploit the C-terminal cassette as a “remote control” to regulate the N-terminal helicase of the enzyme. | de |