Yeast Prp2 liberates the 5' splice site and the branch site adenosine for catalysis of pre-mRNA splicing

Abstract

The RNA helicase Prp2 facilitates the remodeling of the spliceosomal B^act complex to the catalytically activated B* complex just before step one of splicing. As a high-resolution cryo-EM structure of the B* complex is currently lacking, the precise spliceosome remodeling events mediated by Prp2 remain poorly understood. To investigate the latter, we used chemical structure probing to compare the RNA structure of purified yeast B^act and B* complexes. Our studies reveal deviations from conventional RNA helices in the functionally important U6 snRNA internal stem-loop and U2/U6 helix Ib in the activated B^act complex, and to a lesser extent in B*. Interestingly, the N7 of U6-G60 of the catalytic triad becomes accessible to DMS modification in the B* complex, suggesting that the Hoogsteen interaction with U6-A52 is destabilized in B*. Our data show that Prp2 action does not unwind double-stranded RNA, but enhances the flexibility of the first step reactants, the pre-mRNA's 5' splice site and branch site adenosine. Prp2 therefore appears to act primarily as an RNPase to achieve catalytic activation by liberating the first step reactants in preparation for catalysis of the first step of splicing.

Keywords: Prp2 RNA helicase; RNA structure; activated spliceosome; catalytic activation.

FIGURE 1.

Structure of the core RNA–RNA network in yeast activated spliceosomes. (A) Secondary structure of the group II intron domain V (DV; adapted from Pyle 2010 with permission from Taylor and Francis © 2010). Essential RNA–RNA interactions with other intron elements are denoted with small Greek letters and nonstructural magnesium ions are indicated. The J2/3 linker is indicated and positions of intron domains III and IV are indicated schematically. Tertiary interactions are shown as gray lines, and stacking interactions as gray open rectangles. (B) Secondary structure of the yeast U6 ISL and U6/U2 helices Ia and Ib (adapted from Fica et al. 2014 with permission from Nature Publishing Group © 2014). Phosphates of nucleotides (colored in yellow with red outlines) that coordinate the two essential magnesium ions. Tertiary interactions of the AGC triad are shown as gray lines. Stacking of U6–G52 and U6–U80 is indicated (Fica et al. 2014). (C) Watson/Crick accessibilities of the nucleotides in the core of the spliceosomal RNA network in purified B^act (orange) and B* complexes (green). The size of the dots corresponds to the degree of accessibility to chemical modification. The asterisk denotes nucleotide U2–C41 that was hyperreactive toward DMS in the B^act complex. 5′ exon nucleotides are shown in lowercase letters. The last nucleotides of the 5′ exon are base-pairing to U5 loop 1 nucleotides. (D–F) Representative gels of the DMS (D) and CMCT (E) chemical modification of U6 snRNA, as well as DMS modification of U2 snRNA (F) in B^act and B* complexes (as indicated above the lanes), analyzed by primer extension. Key nucleotides are highlighted. See Supplemental Figures S1–S3 for the full data set of all RNAs and chemical probes. B* complexes were reconstituted by complementing purified B^act complexes with recombinant Prp2 and its cofactor Spp2 and ATP.

FIGURE 2.

SHAPE analysis of the RNA–RNA network in yeast B^act and B* spliceosomes. (A,B) Shape analysis of U6 snRNA (A) and U2 snRNA (B) in B^act and B* complexes, as indicated. The data represent the mean of three independent experiments. Normalized reactivities are color-coded (inset) into three different categories of flexibility as previously described (Mortimer and Weeks 2007): black (0–0.3; not flexible, helically constrained), orange (0.3–0.7; intermediate flexibility, no internal Watson/Crick base pairs), and red (>0.7; fully unconstrained, single stranded). Key nucleotides are labeled above the top graph in each panel. Nucleotide numbers and the positions of structural elements are shown below the bottom graph in each panel. (C,D) Representative gels of the 1M7 reactivities of the pre-mRNA in B^act and B* around the 5′ splice site (C) and the branch site sequence (D) analyzed by primer extension. Primer extension will stop 1 nt before the 2′-O-acylated nucleotide relative to the dideoxy sequence. The bracket in C denotes the stretch of nucleotides with reduced 2′-OH flexibility in the B* complex. Quantifications of the autoradiographs in C and D were used for the difference plots (B* − B^act) shown below the respective autoradiographs. The sequences shown below the difference plots correspond to the actual nucleotide involved in attack or protection from 2′-O-acylation by 1M7.

FIGURE 3.

Analysis of tertiary RNA interactions in the catalytic core of the yeast B^act and B* spliceosomes. (A) N7-purine accessibilities of the nucleotides of the spliceosomal RNAs in the B^act and B* complexes as indicated in Figure 1C. (B,C) Representative gels of the chemical modification of the pre-mRNA around the 5′ ss (B) and of the U6 snRNA (C) in purified B^act and B* complexes (as indicated above the lanes) analyzed by primer extension. Key nucleotides are highlighted.

FIGURE 4.

RNA–protein interactions at/near the pre-mRNA branch site in the yeast B^act complex and the pivotal role of A30 of U2 snRNA. (A) Schematic of the region upstream and downstream from the branch site (A501), as found in the cryo-EM structure of the yeast B^act complex (pdb: 5GM6; EM density: EMD-9524), containing an ensemble of natural pre-mRNA (Yan et al. 2016). (Black dots) 5′ phosphates. (Blue dots) The homologous nucleotides of the actin pre-mRNA found protected or weakly accessible on their W/C positions (this manuscript, Supplemental Fig. S4). (Boxed nucleotides) Suboptimal base-pairing due to the large distance between the pairing partners. Amino acid side chain interactions are shown with arrows, with the protein only indicated when it is not Hsh155. The branch site is nucleotide A501 of the pre-mRNA; the U2/BS helix and the extended U2/BS helix are indicated. (B) Path of the extended U2/BS helix. The electrostatic surface of the cavity harboring the extended U2/BS helix and formed by Prp11 (U2 snRNA contacts) and Hsh155 (intron contacts). Intron sequences are not conserved and hence the backbone contacts are the major determinants for the paths of the RNAs. Charges are shown as heat maps, with blue for positive and red for negative potential (±5 kT). (C) The path of the single-stranded polypyrimidine tract across the superhelical α-solenoid structure of the Hsh155 HEAT repeats. Charges are as in B. (D) A close-up view of the counter-clockwise rotational movement of the region of U2 snRNA downstream from U2–A30 (red) encompassing the U2/BS helix and extended U2/BS helix. The coordinates from the cryo-EM structures from the B^act (5GM6, Yan et al. 2016) to C (5GMK, Wan et al. 2016a) to C* (5WSG, Yan et al. 2017) were aligned on U5 snRNA and only U2 snRNA (yellow) is shown. The U2/U6 helix Ia is behind the plane of the drawing. The pivotal U2–A30 is in red and is part of the flexible linker 1, AAGU (nts 30–33 of yeast U2 snRNA; Fig. 2B).