CEF1/CDC5 alleles modulate transitions between catalytic conformations of the spliceosome

Abstract

Conformational change within the spliceosome is required between the first and second catalytic steps of pre-mRNA splicing. A prior genetic screen for suppressors of an intron mutant that stalls between the two steps yielded both prp8 and non-prp8 alleles that suppressed second-step splicing defects. We have now identified the strongest non-prp8 suppressors as alleles of the NTC (Prp19 complex) component, CEF1. These cef1 alleles generally suppress second-step defects caused by a variety of intron mutations, mutations in U6 snRNA, or deletion of the second-step protein factor Prp17, and they can activate alternative 3' splice sites. Genetic and functional interactions between cef1 and prp8 alleles suggest that they modulate the same event(s) in the first-to-second-step transition, most likely by stabilization of the second-step spliceosome; in contrast, alleles of U6 snRNA that also alter this transition modulate a distinct event, most likely by stabilization of the first-step spliceosome. These results implicate a myb-like domain of Cef1/CDC5 in interactions that modulate conformational states of the spliceosome and suggest that alteration of these events affects splice site use, resulting in alternative splicing-like patterns in yeast.

FIGURE 1.

Synthetic lethality of suppressors of second-step splicing defects with prp22 alleles allows for their identification. (A) Schematic of splicing chemistry for BS-G mutant ACT1-CUP1 reporter in vivo. In wild-type cells, BS-G lariat intermediates stall prior to the second step of splicing and accumulate; in the isolated suppressor strains, BS-G lariat intermediates proceed through the second step, forming excised lariat and mRNA. (B) Copper growth phenotype of the BS-G mutant in the previously isolated BS suppressors. Suppressor strains 1 and 2 (containing prp8-161 and -162 alleles) (Query and Konarska 2004) and strains 6, 9, 10, 11, and M2 (containing alleles of cef1, reported here) support growth on 0.2 mM or higher copper, whereas wild-type cells do not. (C) Schematic representation of splicing pathway and assignment of mutant alleles of prp8, prp16, prp22 and U6 snRNA that modulate spliceosomal transitions. (SS) splice site, (BS) branch site. The non-prp8 suppressor alleles, which are thought to improve the second step by a relative stabilization of the second-step conformation, were predicted to exacerbate defects in the DEAH-ATPase Prp22p. (D) The non-prp8 suppressor “M2” is synthetically lethal with prp22 mutants at 30°C. (E) Schematic of screen to identify the isolated suppressors, using rescue of the synthetic growth defect of the suppressor mutations in combination with helicase-defective prp22 alleles. Genomic library plasmids that rescued this growth defect carried CEF1 or wild-type PRP22.

FIGURE 2.

cef1 alleles are general suppressors of second-step splicing defects. (A) Schematic of ACT1-CUP1 pre-mRNA, indicating intron mutations at 5′SS, BS, and 3′SS used in B and other mutations suppressed by cef1-V36R and -S48R. (B) cef1 alleles suppress multiple intron mutations. Upper, primer extension analysis of RNA from cells containing wild-type CEF1, cef1-V36R, or cef1-S48R and ACT1-CUP1 reporters as indicated. Primer complementary to the 3′ exon was used to monitor levels of pre-mRNA, mRNA, and lariat intermediate (indicated by icons on the left; icons on the right indicate use of alternative 3′ splice sites that are activated by cef1-V36R or -48R alleles). Lower, copper growth phenotypes of strains carrying the reporters used above; representative copper concentrations are shown, along with the highest concentration allowing for growth. (C) Sequence of the 3′SS region of the ACT1-CUP1 pre-mRNA, indicating the wild-type 3′SS and the additional weak 3′ splice sites that are used in the presence of cef1 alleles. (D) Schematic of RNA:RNA interactions in the spliceosome core, indicating nucleotides in U6 snRNA whose mutation inhibits the second step of splicing, used in panels E and F, and in Figure 6. Pre-mRNA is shown in black, U2 snRNA in red, U5 snRNA in gray, and U6 snRNA in green; numbering corresponds to S. cerevisiae snRNAs. (E) cef1 alleles suppress the lethal growth defects of U6-A51C, -G52C, and -A59C. (F) U6-A51C inhibits the second step of splicing in vivo. Primer extension analysis (as in Fig. 2B) of RNA from cells containing U6-A51C allele and the BS-C reporter. First-step efficiency was calculated as products of the first step/total RNA. Second-step efficiency was calculated as products of the second step/total products from the first step. (G) The cef1-V36R allele suppresses the growth defect of prp17 deletion at 30°C and rescues viability at 34°C; a series of 1:5 dilutions is shown.

FIGURE 3.

Comparison of the strength of second-step improvement by cef1 alleles to that of alleles of prp8, U6 snRNA, and prp22. (A,B,C) Graphs (upper) summarizing the copper growth (lower) of strains carrying the indicated pairs of reporters and second-step alleles.

FIGURE 4.

Saturation mutagenesis of Cef1 positions 36 and 48 confirms the importance of arginine at these sites. (A) Schematic of domain structure of Cef1/CDC5. Bottom, expansion of the first myb-like domain, alignment of sequences from S. cerevisiae, S. pombe, and Homo sapiens, and indication of positions of mutants identified here (V36R and S48R) and in the cdc5 S. pombe mutant (W33R) (Ohi et al. 1994). (*) Positions of amino acid identity in all three species, (:) positions of nonidentity but similar amino acid properties. (B) Schematic of randomization of Cef1 positions 36 and 48 by in vivo gap repair, and selection for improvement of BS-G splicing (limiting for the second step of splicing) by growth in the presence of copper. Selection at position 36 yielded only arginine, whereas selection at position 48 yielded arginine and lysine. (C) Primer extension analysis of BS-G reporters from strains carrying various cef1-V36x mutations, -S48x mutations. Primer extension and copper growth assays are as described in Figure 2B.

FIGURE 5.

Saturation mutagenesis of Cef1 myb-like domains confirms the importance of positions 36 and 48. (A) Upper, schematic of domain structure of Cef1/CDC5 and the two myb-like domains that were mutagenized. Lower, expansion of the first myb-like domain, alignment of sequences from S. cerevisiae, S. pombe, and H. sapiens, and indication of positions of mutants identified (H31N, W33R, V36R, A37P, and S48R). (*) Positions of amino acid identity in all three species, (:) positions of nonidentity, but similar amino acid properties. (B) Schematic of randomization of the cef1 myb-like domains 1 and 2 by in vivo gap repair, and selection for improvement of BS-G splicing (limiting for the second step of splicing) by growth in the presence of copper. (C) Primer extension analysis of BS-G reporter from strains carrying various cef1 mutations identified in panel B. Primer extension and copper growth assays are as described in Figure 2B. First-step efficiency was calculated as products of the first step/total RNA. Second-step efficiency was calculated as products of the second step/total products from the first step.

FIGURE 6.

The combination of first-step alleles of prp8 and second-step alleles of cef1 restores a nearly wild-type pattern of splicing. (A) Genetic interactions between cef1 and prp8 alleles. Second-step cef1 alleles suppress the temperature-sensitive growth defect of first-step prp8 alleles at 38°C (upper), whereas cef1 alleles are cold-sensitive in combination with second-step prp8 alleles (lower, 16°C). The temperature-sensitive phenotype of first-step prp8 alleles in the presence of wild-type CEF1 is suppressed by both of the second-step cef1 alleles (upper). Strains carrying plasmid-borne cef1 and prp8 alleles were spotted on plates incubated at 16, 25, 30, 37, and 38°C. (B) Primer extension analysis and copper growth phenotypes of BS-C reporter (limiting for both steps of splicing) from strains carrying combinations of prp8 first-step alleles (prp8-R1753K, 8-101, and 8-syf77) or second-step alleles (prp8-161 and 8-162) with cef1 second-step alleles (cef1-V36R and -S48R), as indicated. First-step efficiency was calculated as products of the first step/total RNA. Second-step efficiency was calculated as products of the second step/total products from the first step. (C) Schematic of cancellation of altered splicing upon combination of prp8 first-step alleles with cef1 second-step alleles.

FIGURE 7.

cef1 and U6-U57C alleles affect splicing at distinct steps. (A) Genetic interactions between cef1 and U6 snRNA alleles. Second-step cef1 alleles are synthetically lethal with the second-step U6-U57A allele at 37°C, whereas the same cef1 alleles suppress the temperature-sensitive growth defect of the first-step U6-U57C allele at 37°C (right). (B) Effects of combination of cef1 second-step allele with U6-U57C or U6-U57A on splicing of BS-C reporters. For BS-C introns (limiting for both steps), combinations of U6-U57C and cef1 second-step alleles improve overall splicing (lanes 8–9), in contrast to the combination of prp8 first-step alleles and cef1 second-step alleles (see Fig. 6). First-step efficiency was calculated as products of the first step/total RNA. Second-step efficiency was calculated as products of the second step/total products from the first step. (C) Proposed contributions of Cef1, Prp8, U6 snRNA, and Prp22 to distinct steps in spliceosomal transitions. cef1 alleles modulate an event in the first-to-second-step transition indistinguishable from that modulated by prp8 alleles; this event is distinct from that modulated by alleles of prp16 and U6 snRNA. We propose that these two distinct events represent the theoretically required opening of the catalytic center and repositioning of first-step products and second-step substrates. The same cef1, prp8, and U6 alleles, as well as alleles of prp22, alter the exit of mRNA from the second-step spliceosome, which may involve analogous events.