Extensive Genetic Interactions Between PRP8 and PRP17/CDC40, Two Yeast Genes Involved in Pre-mRNA Splicing and Cell Cycle Progression

Corresponding author: Martin Kupiec, Department of Molecular Microbiology and Biotechnology, Tel Aviv University, Ramat Aviv 69978, Israel., martin{at}ccsg.tau.ac.il (E-mail)

Communicating editor: F. WINSTON

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Biochemical and genetic experiments have shown that the PRP17 gene of the yeast Saccharomyces cerevisiae encodes a protein that plays a role during the second catalytic step of the splicing reaction. It was found recently that PRP17 is identical to the cell division cycle CDC40 gene. cdc40 mutants arrest at the restrictive temperature after the completion of DNA replication. Although the PRP17/CDC40 gene product is essential only at elevated temperatures, splicing intermediates accumulate in prp17 mutants even at the permissive temperature. In this report we describe extensive genetic interactions between PRP17/CDC40 and the PRP8 gene. PRP8 encodes a highly conserved U5 snRNP protein required for spliceosome assembly and for both catalytic steps of the splicing reaction. We show that mutations in the PRP8 gene are able to suppress the temperature-sensitive growth phenotype and the splicing defect conferred by the absence of the Prp17 protein. In addition, these mutations are capable of suppressing certain alterations in the conserved PyAG trinucleotide at the 3' splice junction, as detected by an ACT1-CUP1 splicing reporter system. Moreover, other PRP8 alleles exhibit synthetic lethality with the absence of Prp17p and show a reduced ability to splice an intron bearing an altered 3' splice junction. On the basis of these findings, we propose a model for the mode of interaction between the Prp8 and Prp17 proteins during the second catalytic step of the splicing reaction.

PRE-mRNA splicing takes place by two consecutive trans-esterification reactions. In the first step the border between the 5' exon and the intron is cleaved, yielding the 5' exon and lariat intron-exon intermediates. In the second step, the 3' splice site is cleaved and the two exons are joined, creating a mature RNA and the lariat intron (MADHANI and GUTHRIE 1994A ). Accurate splicing requires conserved sequences within the introns of yeast pre-mRNAs at the 5' (GUAUGAGU) and at the branchpoint (UACUAACA). Recognition of the 3' splice site requires a specific trinucleotide (PyAG) usually preceded by a pyrimidine-rich tract. Spliceosomal small nuclear ribonucleoprotein particles (snRNPs) U1, U2, U4/6, and U5 recognize these elements and assemble onto the pre-mRNA substrate in a stepwise fashion. Initially the U1 snRNP interacts with the pre-mRNA at the 5' splice site, followed by the U2 snRNP contacting the branchpoint. The U4/U6 and U5 snRNPs then join, as a tri-snRNP, thus forming the mature spliceosome (MADHANI and GUTHRIE 1994A ; KRAMER 1996 ; NILSEN 1998 ). Much knowledge has been gathered about the RNA interactions responsible for recognizing and aligning the 5' splice site and branch site before the first catalytic step; in contrast, much less is known about the events that lead to 3' splice site selection and the second catalytic step of splicing (reviewed in UMEN and GUTHRIE 1995C ).

Pre-mRNA splicing requires the activity of a large number of trans-acting factors (KRAMER 1996 ; WANG and MANLEY 1997 ; WILL and LUHRMANN 1997 ). Many of the proteins involved in the process of splicing have been identified by genetic screens for conditional mutations in yeast (VIJAYRAGHAVAN et al. 1989 ; WOOLFORD and PEEBLES 1992 ; BEGGS 1995 ). These prp (pre-mRNA processing) mutants are partially or completely defective in the removal of intervening sequences from pre-mRNAs.

Four genes that are required specifically for the second catalytic step in the yeast Saccharomyces cerevisiae have been identified: PRP16, SLU7, PRP17, and PRP18 (VIJAYRAGHAVAN et al. 1989 ; SCHWER and GUTHRIE 1991 ; FRANK and GUTHRIE 1992 ). These genes share a unique set of genetic interactions with each other, suggesting a physical or functional association of the encoded proteins (FRANK et al. 1992 ; JONES et al. 1995 ; XU et al. 1998 ). Previous work has shown that the four proteins act in concert in the second step of the splicing reaction. This step could be further separated into an ATP-dependent stage, which requires the activity of Prp16p and Prp17p, and a subsequent ATP-independent stage at which the Slu7 and Prp18 proteins participate (JONES et al. 1995 ). PRP16 encodes an RNA-dependent ATPase of the DEAH-box family, which has been shown recently to unwind RNA duplexes in vitro (WANG et al. 1998 ). SLU7 encodes a protein with a cysteine-rich zinc knuckle element, a motif that has been implicated in RNA binding (FRANK and GUTHRIE 1992 ). It has been shown that the requirement for Slu7p during the second step of splicing increases with the distance between the branchpoint and the 3' splice site (BRYS and SCHWER 1996 ). PRP18 is a nonessential gene that encodes a small (29-kD) protein. The absence of Prp18p confers a temperature-sensitive phenotype; in addition, in prp18 mutants the second step of the splicing reaction is inhibited, although not abolished (HOROWITZ and ABELSON 1993 ). A physical interaction between Slu7p and Prp18p has been demonstrated using the two-hybrid assay (ZHANG and SCHWER 1997 ).

Mutations in the PRP17 gene cause accumulation of splicing intermediates (VIJAYRAGHAVAN et al. 1989 ; BEN-YEHUDA et al. 1998 ). Functional interactions between Prp17p and the U2 and U5 snRNAs were suggested by the synergistic lethality of alleles of PRP17 in combination with specific U2 or U5 snRNA mutations (FRANK et al. 1992 ; XU et al. 1998 ). It has been found recently that PRP17 is allelic to the CDC40 gene, which was characterized previously as a gene involved in cell cycle progression (KASSIR and SIMCHEN 1978 ). The cdc40-1 mutation affects both the mitotic and meiotic cell cycles (KASSIR and SIMCHEN 1978 ). For the sake of clarity, in this article we refer to the PRP17/CDC40 gene as PRP17. A full deletion allele of the gene shows a temperature-sensitive phenotype, and a cell cycle arrest at the G2 phase of the cell cycle, after the completion of DNA replication (VAISMAN et al. 1995 ; SESHADRI et al. 1996 ). In addition, the Prp17 protein is needed for the maintenance of the mitotic spindle after the cell cycle arrest at the restrictive temperature (VAISMAN et al. 1995 ). The PRP17 gene codes for a protein with several copies of the WD repeat (VAISMAN et al. 1995 ; BEN-YEHUDA et al. 1998 ; ZHOU and REED 1998 ). This repeated motif is found in a large family of proteins that play important roles in signal transduction, cell cycle progression, splicing, transcription, and development (for review see NEER et al. 1994 ).

In addition to the four mentioned genes involved exclusively in the second step of the splicing reaction, a major role in executing this step was demonstrated for the PRP8 gene. Prp8p interacts with both splice sites, contacting the 3' splice site after the first step is concluded (TEIGELKAMP et al. 1995A , TEIGELKAMP et al. 1995B ; UMEN and GUTHRIE 1995A ). Mutational analysis of PRP8 has revealed that it plays a role in governing the specificity and the fidelity of 3' splice site selection (UMEN and GUTHRIE 1995B , UMEN and GUTHRIE 1996 ).

Here we report extensive genetic interactions between the PRP17 and the PRP8 genes. We show that some mutations in a confined region of the PRP8 gene suppress the temperature-sensitive phenotype conferred by the PRP17 allele, while others show synthetic lethality with the absence of the Prp17 protein. We propose that the Prp17 and Prp8 proteins cooperate in the recognition of the 3' splice site during the second catalytic step of the splicing reaction.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Yeast strains and plasmids:
The yeast strains used in this study are listed in Table 1. Plasmids pJU225 and pJU255 carry the PRP8 and prp8-122 alleles, respectively, on a high-copy-number TRP1-marked vector (pRS424). pJU169 bears PRP8 on a centromeric URA3-marked plasmid (UMEN and GUTHRIE 1996 ). pJU97, pJU98, pJU143, pJU146, pSB30, pSB33, pSB38, pSB47, and pCC71 are 2µ, LEU2-marked CUP1-ACT1 reporter plasmids (BURGESS and GUTHRIE 1993 ; LESSER and GUTHRIE 1993A , LESSER and GUTHRIE 1993B ; UMEN and GUTHRIE 1995A , UMEN and GUTHRIE 1995B , UMEN and GUTHRIE 1996 ; C. COLLINS and C. GUTHRIE, unpublished data). These plasmids were generously provided by C. Collins. Deletion of the PRP17 gene was carried out by a one-step replacement method using plasmid pSBY15 (PRP17::LEU2) (VAISMAN et al. 1995 ). Plasmids pSBY18, pSBY19, and p1426 carry the PRP17 gene on centromeric vectors and are marked with TRP1, TRP1, and ADE3, or URA3 and ADE3, respectively. pSBY55 is YEp24 (URA3, 2µ), carrying the PRP8 gene.

Media, growth conditions, and general procedures:
Standard molecular biology procedures such as restriction enzyme analysis and Southern blot analysis were carried out as described in SAMBROOK et al. 1989 . Yeast media and molecular biology procedures (transformations, DNA preparations, etc.) were as described in SHERMAN et al. 1986 . Plasmid DNA extraction from yeast cells was as described by ROBZYK and KASSIR 1992 . Yeast cells were grown at 25°, 30°, or 34° in YEPD (1% yeast extract, 2% Bacto peptone, 2% dextrose) or SD media (0.67% yeast nitrogen base, 2% dextrose, and the appropriate nutrients added). Bacto agar (1.8%) was added for solid media. Selective media lacking one nutrient are designated SD-nutrient (e.g., SD-Ura). Ura^- colonies were selected on SD complete medium with uracil (50 mg/liter) and 5-fluoroorotic acid (5-FOA, 0.8 g/liter) (BOEKE et al. 1987 ). Sporulation was carried out in SPO medium (1% K acetate, 0.1% yeast extract, 0.05% dextrose).

Isolation of prp8-scf alleles:
Yeast strains D103 and D110 (PRP17) were plated on YEPD plates (1 x 10⁷ cells/plate) and subjected to UV irradiation to yield a survival rate of 10%. The cells were incubated for 3 hr at the permissive temperature (25°) and were transferred to the restrictive temperature (34°). After 3 days of incubation, 20 independent temperature-resistant colonies were isolated. The mutants (SCF strains) were then crossed to the parental strains. The diploids thus obtained were sporulated and the 2:2 segregation of the temperature-sensitive phenotype was verified. Derivatives of all the mutants were crossed in all possible combinations and were subjected to complementation and allelism tests.

Cloning of the SCF gene:
SCF103A was transformed with a YEp24(URA3)-based genomic library. The transformants were screened for the inability to grow at 34°. Out of 30,000 transformants that were screened, only one colony exhibited a temperature-sensitive phenotype that correlated with the presence of the plasmid. The plasmid was isolated from this transformant, subjected to partial DNA sequencing, and compared to the yeast complete sequence in the Saccharomyces Genome Database (http://genome-www.stanford.edu/Saccharomyces). The complementing genomic fragment contained a single complete open reading frame (ORF) (YHR165c) corresponding to the PRP8 gene. pSBY57, carrying the HpaI-XhoI fragment of the PRP8 gene on an integrative vector (YIplac204); (GIETZ and SUGINO 1988 ), was integrated in the genome of a wild-type strain and crossed to the scf mutants. Tetrad analysis confirmed allelism between the integrated plasmid at the PRP8 locus (scored as Trp⁺ spores) and the original scf mutation.

Mutagenesis, screening, and genetic characterization of prp8-syf alleles:
Strain YH2 (PRP17 ura3 trp1 ade2 ade3) carrying the centromeric plasmid p1426 (PRP17, ADE3, URA3) was subjected to UV irradiation to yield a survival rate of 10%. The surviving cells were allowed to form colonies on YEPD plates and were screened for red, nonsectoring colonies. After restreak, the colonies were tested for sensitivity to 5-FOA. All the nonsectoring, 5-FOA^s colonies were transformed with plasmids pSBY18 (PRP17, TRP1) or the vector YCplac22 (TRP1, GIETZ and SUGINO 1988 ). The presence of PRP17 on pSBY18 allows the loss of p1426, conferring red/white sectoring and resistance to 5-FOA. Only strains that showed sectoring and 5-FOA resistance with pSBY18 but not with YCplac22 were taken for further studies.

Each putative syf mutant was crossed to YH3 (MATa, isogenic to YH2), and the 2:2 segregation of the sectoring/nonsectoring phenotype was verified. All the diploids exhibited red/white sectoring, demonstrating that the mutations were recessive. Derivatives of all the mutants were crossed in all possible combinations, and complementation groups were established. One member of each complementation group was chosen for cloning of the complementing gene.

Cloning of the SYF genes:
Nonsectoring syf mutants carrying plasmids p1426 (PRP17, URA3, ADE3) or pSBY19 (PRP17, TRP1, ADE3) were transformed with a YEp24(URA3)-based genomic library. The transformants were screened for the ability to form red/white sectors on rich medium and tested for resistance to 5-FOA. The plasmids were isolated from sectoring or white transformants, subjected to partial DNA sequencing, and compared to the yeast complete sequence in the Saccharomyces Genome Database (http://genome-www.stanford.edu/Saccharomyces). Three different, overlapping plasmids complemented the synthetic lethality of strain SYF14. The three plasmids overlapped in a region containing a single ORF, YHR165c, corresponding to the PRP8 gene. Allelism between this gene and the alleles present in prp8-syf mutants was confirmed using the integrative plasmid pSBY57, as described above.

Mapping of the prp8-scf and prp8-syf alleles:
The prp8-scf or prp8-syf alleles were cloned by a gene conversion strategy. Yeast strains bearing the relevant mutations in the PRP8 locus were transformed with plasmid pSBY55 (PRP8, 2µ) linearized with the SacI, Asp718, or SnaBI restriction enzymes. Transformants of the SCF strains (PRP17 prp8-scf) were screened for the ability to grow at the restrictive temperature in the presence of the plasmid, indicating a gene conversion event in which the genomic prp8 mutation was transferred to the plasmid during its repair. In a similar way, SYF transformants were screened for the inability to form sectors in the presence of the plasmid (red colonies). The plasmids were isolated from each mutant strain and were subjected to restriction analysis to rule out the possibility of plasmid deletions. Each plasmid was then retransformed to the mutant parental strain to confirm the mutant phenotype.

The mutations in each allele were mapped by cotransformation. Each plasmid was linearized and cotransformed to its parental mutant strain together with one of a series of overlapping fragments obtained from the wild-type PRP8 gene (ratio 1:10 plasmid/fragment). The following fragments were used (numbers reflect nucleotides in the PRP8 ORF): full length (1–7242); fragment I (1–5723); fragment II (2510–5723); fragment III (3711–7242). A recombination event between the mutant prp8 allele on the plasmid and the wild-type information on the fragment can restore PRP8 activity, provided the fragment information overlaps the location of the mutation on the prp8 allele. The transformants obtained in each cotransformation reaction were tested for their ability to complement the mutations as indicated by the inability to grow at the restrictive temperature (SCF strains) or by the ability to segregate sectored or white colonies (SYF strains).

Copper-resistance assay:
Yeast strains transformed with wild-type, UAG 3' splice site, pJU97 (3' UGG) or pJU98 (3' UUG), containing the ACT1 intron fused to the CUP1 gene (UMEN and GUTHRIE 1996 ), or pJU143 (+TPyDOWN), or pJU146 (+A WT) bearing the CUP1 gene with an intron containing duplicated 3' splice sites, or pSB30 (5' G1A), or pSB33 (5' G5A), or pCC71 (5' U2A), or pSB38 (branchpoint A6C), or pSB47 (branchpoint C3A; BURGESS and GUTHRIE 1993 ; LESSER and GUTHRIE 1993A , LESSER and GUTHRIE 1993B ; UMEN and GUTHRIE 1995A , UMEN and GUTHRIE 1995B , UMEN and GUTHRIE 1996 ) were tested for copper sensitivity as follows: 1 x 10⁷ and 1 x 10⁶ cells of the tested strains were spotted on SD-Leu plates containing different copper concentrations. Plates were made to a chosen copper concentration (0.025–2 mM) by adding to SD-Leu media a dilution of filtered 1 M CuSO₄ after autoclaving.

RNA analysis:
RNA was extracted by a hot phenol method (SCHMITT et al. 1990 ) from cells grown to an OD₆₀₀ of 0.4–0.6 in SD-Leu-Trp medium at 30°. Primer extension analysis was performed essentially as described by BOORSTEIN and CRAIG 1989 (but omitting actinomycin D), using the ACT1-CUP1 primer as in LESSER and GUTHRIE 1993A and, as a control, a U1 snRNA primer (CACGCCTTCCGCGCCGT).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Mutations in the PRP8 gene are able to suppress the temperature-sensitive phenotype conferred by the PRP17 allele:
The PRP17 gene encodes a protein that is essential only at elevated temperatures and affects two major processes: pre-mRNA splicing and cell cycle progression (VAISMAN et al. 1995 ; BEN-YEHUDA et al. 1998 ; BOGER-NADJAR et al. 1998 ). To identify genes that may interact with PRP17, we performed a screen for suppressor mutants that are able to grow at the restrictive temperature in the absence of the PRP17 gene product.

Yeast strains (D103, D110; both PRP17) were mutagenized using UV irradiation and incubated at the restrictive temperature (34°). Twenty independent temperature-resistant colonies were isolated. When the mutants were crossed to the parental strains (PRP17), all the obtained diploids failed to grow at the restrictive temperature, indicating that the mutations were recessive. Tetrad analysis of each diploid revealed a Mendelian segregation of the temperature-sensitive phenotype, implying that each mutant bears a mutation in a single gene. When mutants were then crossed with each other in all possible combinations, the resulting diploids were able to grow at the restrictive temperature, indicating that they fall into a single complementation group. Tetrad analysis of individual diploids failed to produce temperature-sensitive recombinant spores, confirming that the 20 mutants represent alleles of the same gene. The mutations were designated scf (suppressors of cdc forty).

Since all the scf mutations were recessive, the complementing gene from a yeast genomic library was cloned by transforming a PRP17 scf strain (SCF-103A) and screening for transformants unable to grow at the restrictive temperature. Out of 30,000 transformants that were tested, only one colony exhibited a temperature-sensitive phenotype that correlated with the presence of the plasmid. The complementing genomic fragment contained a single complete ORF (YHR165c) corresponding to the PRP8 splicing factor gene. Allelism tests confirmed that the scf mutations were allelic to the PRP8 gene (see MATERIALS AND METHODS). PRP8 encodes a large, evolutionarily conserved protein that is an essential component of the spliceosome during both steps of the splicing reaction (LOSSKY et al. 1987 ; WHITTAKER et al. 1990 ; HODGES et al. 1995 ; TEIGELKAMP et al. 1995A , TEIGELKAMP et al. 1995B ). It is thought that Prp8p anchors the exons in the catalytic center of the spliceosome and stabilizes the fragile interactions between the U5 snRNA and the nonconserved exon sequences (BEGGS 1995 ; TEIGELKAMP et al. 1995A , TEIGELKAMP et al. 1995B ; DIX et al. 1998 ). In addition, similarly to PRP17, the PRP8 gene has been found to play a role in cell cycle progression. Certain mutant alleles of PRP8 (dbf3, dna39) were isolated as cell-cycle-specific mutants with phenotypes very similar to those of PRP17 mutants (DUMAS et al. 1992 ; SHEA et al. 1994 ). Thus, the finding that mutations in PRP8 can rescue the cell cycle defect conferred by deletion of the PRP17 gene suggests that Prp8p and Prp17p participate in a common cellular process or processes involving pre-mRNA splicing and cell cycle progression.

The PRP17 mutants are sensitive to DNA-damaging agents, such as methylmethane sulfonate (MMS), and are unable to undergo sporulation and meiosis (KASSIR and SIMCHEN 1978 ; KASSIR et al. 1985 ; KUPIEC and SIMCHEN 1986 ). The prp8-scf mutations suppressed these phenotypes, too (data not shown). Thus, the prp8-scf mutations, isolated by their ability to allow growth at the restrictive temperature, suppress all the tested phenotypes conferred by the deletion of the PRP17 gene. In a Prp17⁺ background the prp8-scf alleles showed no detectable growth differences nor any sensitivity to DNA-damaging agents, in comparison to the isogenic wild-type strain, at several temperatures. Meiosis and sporulation in prp8-scf/prp8-scf strains were also normal (data not shown).

syf alleles of PRP8 exhibit synthetic lethality with the PRP17 allele:
In an effort to identify genes that interact with PRP17, we have performed recently a synthetic lethality screen seeking mutants unable to survive even at the permissive temperature in the absence of the Prp17/Cdc40 protein. The mutants thus obtained (syf: synthetic lethal with cdc forty) were separated into complementation groups by genetic analysis (see MATERIALS AND METHODS). The largest complementation group was found to contain eight alleles of the PRP8 gene. Thus, some mutations in PRP8 could suppress the temperature-sensitive phenotype conferred by the deletion of the PRP17 gene (prp8-scf alleles), while others cause lethality in the absence of Prp17p (prp8-syf alleles).

prp8 alleles act in a dosage-dependent fashion:
The prp8-scf mutations were recessive in diploids, and the PRP8 gene on a high-copy-number plasmid prevented a haploid PRP17 prp8-scf strain from growing at the restrictive temperature. However, the same haploid strain bearing a single-copy plasmid carrying the wild-type PRP8 gene was able to grow, albeit slowly, at 34°. This indicates that under these conditions the prp8-scf allele was semidominant over the wild-type PRP8 allele. Hence, the suppression of the cell cycle phenotype is a dosage-dependent trait, with prp8-scf alleles being recessive in diploids, but semidominant or dominant in haploid cells, depending on the expression level. Consistent with these results, the mutant prp8-scf alleles were dominant over the wild-type allele when expressed from high-copy-number plasmids (Fig 1A).

View larger version (71K): In this window In a new window Download PPT slide   Figure 1. (A) Overexpression of prp8-scf alleles is dominant. Strain D110 (PRP17 PRP8) was transformed with a 2µ-URA3-marked plasmid carrying the mutant prp8-scf103A or prp8-scf103C alleles, or with a plasmid bearing the PRP8 gene (pSBY55), or with the vector alone (YEp24). The transformants were diluted and spotted on SD-Ura plates and assayed for viability at either the permissive (25°) or the restrictive (34°) temperature. (B) Overexpression of prp8-syf alleles is dominant. Strain D110 (PRP17 PRP8) was transformed with a 2µ-URA3-marked plasmid carrying the mutant prp8-syf77 allele, or with a similar plasmid bearing the PRP8 gene (big colony). Growth after 4 days is shown.

In a similar way, although the synthetic lethal phenotype of prp8-syf mutants was recessive in diploids, high-copy-number plasmids carrying different prp8-syf alleles resulted in semilethality in PRP17 PRP8 yeast cells: upon transformation very small colonies were obtained, which displayed many death sectors (Fig 1B). This phenotype was not observed in Prp17⁺ cells; thus, the prp8-syf alleles behave as semidominant when present in high copy number.

These results suggest that the phenotypes displayed by the different prp8 alleles are affected by their dosage. The wild-type and mutant proteins may compete in the creation of spliceosomal complexes: the proportion of complexes carrying wild-type Prp8p vs. those carrying Prp8 mutant proteins determines the cell phenotype. A similar effect has been shown for the PRP16 gene: overexpression of nonviable alleles of PRP16 impaired the growth of wild-type PRP16 cells (HANS-RUDOLF and SCHWER 1998 ). The mutant proteins, therefore, may be able to assemble into the spliceosomal complexes, but impair their function.

Cloning and identification of the prp8-scf and prp8-syf alleles:
Since the PRP8 gene encodes one of the largest proteins (2413 residues, a molecular weight of 280 kD) in S. cerevisiae, it was interesting to determine whether the genetic interactions with the PRP17 gene are restricted to specific regions of PRP8. Four different prp8-scf alleles and three different prp8-syf alleles were cloned by a gene conversion strategy, and their mutations mapped using a cotransformation assay (see MATERIALS AND METHODS). The mutations in all seven prp8 alleles mapped to the same 2-kb region of the PRP8 gene (HpaI-MscI fragment, encoding amino acids 1237–1908). Thus, mutations that suppress the phenotypes conferred by the deletion of PRP17, or mutations that result in a synthetic lethal phenotype with the absence of Prp17p, map to the same region of PRP8. In a previous study, mutations in this region were found to affect 3' splice site fidelity (region C, UMEN and GUTHRIE 1996 ).

Sequence analysis of two of the prp8-scf alleles revealed that scf-103A carries three amino acid changes, W1575C, E1576K, and G1636S, whereas the scf-103C allele has a single amino acid change, K1563I. Similarly, the prp8-syf77 allele encodes a protein containing phenylalanine instead of leucine at position 1557, only a few amino acids upstream of the prp8-scf mutations. Thus, the mutations were clustered to a small region of the Prp8 protein, a region highly conserved in evolution from yeast to humans (HODGES et al. 1995 ; UMEN and GUTHRIE 1996 ; LUO et al. 1999 ).

prp8-scf alleles affect the fidelity of 3' splice site utilization:
The mutations W1575C and E1576K of the prp8-scf103A allele affect the same residues as those in the prp8-122 (W1575R) and prp8-123 (E1576V) mutant alleles, described by UMEN and GUTHRIE 1996 . These mutant alleles were isolated in a screen for PRP8 alleles that affect the fidelity of 3' splice site utilization. The prp8-122 and the prp8-123 alleles were able to suppress the splicing defect caused by a single base alteration in the conserved PyAG motif located at the 3' splice site of an ACT1-CUP1 reporter gene. CUP1 is a nonessential gene that allows cells to grow in the presence of copper in a dosage-dependent manner; the intron-containing ACT1-CUP1 construct thus enables quantitative analysis of pre-mRNA splicing (LESSER and GUTHRIE 1993A ; UMEN and GUTHRIE 1996 ). Whereas the wild-type PRP8 strain was unable to efficiently splice pre-mRNA molecules with a modified UUG sequence at the 3' splice junction, prp8-122 and prp8-123 strains were able to do so, providing a copper-resistant phenotype (UMEN and GUTHRIE 1996 ).

Using the ACT1-CUP1 reporter system, we tested whether the prp8-scf alleles identified in our screen affect 3' splice site fidelity. Yeast strain YJU75 (prp8) bearing the PRP8 gene on a TRP1-marked plasmid (UMEN and GUTHRIE 1996 ) was transformed with URA3-marked plasmids carrying each of the four prp8-scf alleles. Following plasmid shuffling, the strains harboring the mutant prp8-scf alleles or the wild-type PRP8 gene were transformed with either the ACT1-CUP1 wild-type 3' UAG reporter or the ACT1-CUP1 3' UUG reporter. The resistance to various copper concentrations was measured. In strains bearing the prp8-scf alleles and the wild-type 3' UAG reporter, no effect on copper resistance was seen, in comparison to the control strain bearing the wild-type PRP8 gene (data not shown).

The results obtained with the different prp8 alleles in the presence of the ACT1-CUP1 3' UUG reporter are presented in Table 2 and Fig 2. As for the prp8-122 allele (UMEN and GUTHRIE 1996 ), all four prp8-scf alleles enabled growth in the presence of this reporter at higher copper concentrations than those allowed by the wild-type PRP8 allele. Thus, as for prp8-122, the prp8-scf alleles suppressed the splicing defect caused by a mutation in the 3' splice site. The degree of suppression varied among the alleles and showed a good correlation with the ability to suppress the temperature sensitivity of PRP17 mutants, as assayed by generation time and plating efficiency at the restrictive temperature. Accordingly, the strongest prp8-scf allele, scf-103A, exhibited resistance to the highest copper concentrations; in fact, yeast strains carrying this allele were able to grow in the presence of 2 mM copper, a concentration at which even prp8-122 strains cannot grow (Table 2).

View larger version (71K): In this window In a new window Download PPT slide   Figure 2. The effect of different prp8-scf alleles on 3' splice site fidelity. A prp8 strain (YJU75) carrying different PRP8 alleles on URA3-marked plasmids was tested for the ability to grow on different copper concentrations in the presence of the LEU2-marked plasmid bearing a ACT1-CUP1 3' UUG reporter.

To confirm that the increased resistance to copper in the prp8-scf mutants was really due to higher splicing efficiency, the RNA was analyzed by primer extension. As shown in Fig 3, the UUG 3' splice site reporter RNA was indeed spliced significantly better (at least fourfold) in the strain carrying the prp8-scf103A allele, in comparison to the PRP8 control. The effect of the prp8-scf103C allele was less obvious, indicating that the copper-resistance test is more sensitive than the primer extension assay. Moreover, splicing of the wild-type ACT1-CUP1 RNA was also elevated in the prp8-scf103A strain. Presumably, this effect was not seen with the growth assay because the level of copper resistance was already maximal with the wild-type reporter and wild-type PRP8.

View larger version (40K): In this window In a new window Download PPT slide   Figure 3. Primer extension analysis of the splicing of ACT1-CUP1 reporter transcripts in PRP8 or prp8-scf strains. RNA was extracted from YJU75 (prp8) cells carrying wild-type (WT) or mutant (UUG) 3' splice site ACT1-CUP1 reporter plasmids and plasmids bearing wild-type PRP8, prp8-scf103A, or prp8-scf103C alleles. Primer extension products are shown for the ACT1-CUP1 reporter mRNAs and (as a loading control) for U1 snRNA. The wild-type and 3' splice site reporter mRNAs differ slightly in length due to minor differences in the ACT1-CUP1 junction sequence, but are comparable between PRP8 alleles.

Since the prp8-scf alleles showed an effect on the fidelity of 3' splice site utilization (the activity for which prp8-122 allele was isolated), the reciprocal experiment was performed, testing whether the prp8-122 fidelity mutant allele was able to suppress the temperature sensitivity of a PRP17 strain. The results, shown in Fig 4, demonstrate that prp8-122 was able to partially suppress the temperature-sensitive phenotype. Hence, a correlation was established between the ability to suppress the temperature sensitivity of the PRP17 allele and the ability to suppress the 3' splice site mutation.

View larger version (48K): In this window In a new window Download PPT slide   Figure 4. The 3' splice site fidelity mutant prp8-122 is able to suppress the temperaturesensitivity conferred by the PRP17 mutation. A PRP17PRP8 strain (D110) carrying a TRP1-marked plasmid bearing the prp8-122 allele was tested for the ability to grow at either the permissive (25°) or the restrictive (34°) temperature. Plasmids bearing PRP17, PRP8, or the vector alone (pRS424) were used as control.

The prp8-scf alleles specifically affect efficient splicing of reporters harboring 3' splice site alterations:
The suppression by prp8-scf alleles could be due to a higher efficiency of splicing in general, rather than to a specific effect on 3' splice site fidelity. To rule out this possibility, we have tested the ability of the prp8-scf alleles to splice reporter constructs carrying introns with different alterations. ACT1-CUP1 reporter plasmids bearing mutations in the conserved 5' splice site sequence (G1A, U2A, and G5A; LESSER and GUTHRIE 1993B ; C. COLLINS and C. GUTHRIE, unpublished observations), or in the conserved UAC₂₅₆UAA₂₅₉C branchpoint sequence (C256A and A259C; BURGESS and GUTHRIE 1993 ) were introduced into strains carrying either PRP8 or prp8-scf alleles. The ability of the prp8-scf strains to grow with increasing copper concentrations was tested and was found to be identical to that of the wild-type PRP8 control strain for all the mutant constructs.

In addition, the ability of the prp8-scf mutants to affect the recognition of the pyrimidine tract, an additional feature of the 3' splice site, was tested. Although the pyrimidine residues in yeast 3' splice sites are generally less conserved than those in mammalian introns, they play an important role in efficient 3' splice site utilization. Two ACT1-CUP1 reporter plasmids were tested: +TPyDOWN detects splicing of an intron containing duplicated 3' splice sites, one of which is uridine rich (proximal) and the other adenosine rich (distal), whereas +A WT carries the 3' splice sites in the reverse order: the proximal site is adenosine rich, and the distal site is uridine rich. Loss of pyrimidine recognition generates a higher level of in-frame message and Cup1 protein (UMEN and GUTHRIE 1995B ). None of the prp8-scf alleles showed any alteration of pyrimidine tract recognition (data not shown).

Therefore, we conclude that the prp8-scf alleles preferentially affect recognition of the conserved PyAG sequence at the 3' splice site and do not affect recognition of other sequence elements important for splicing. These results also suggest that the prp8-scf alleles may suppress the temperature sensitivity of PRP17 strains by increasing the efficiency of 3' splice site recognition of certain intron(s).

prp8-syf strains show defects in 3' splice site utilization:
As the PRP17 mutations can be suppressed by prp8 alleles that allow splicing of altered 3' splice sites, the prp8-syf alleles, which are lethal in the absence of Prp17p, may exhibit the opposite effect. Therefore, we tested the resistance of strains bearing the prp8-syf alleles to increasing copper concentrations in the presence of the reporter genes previously described. The results, shown in Table 3, demonstrate that the prp8-syf alleles confer hypersensitivity to low copper concentrations in the presence of the 3' UUG alteration. Whereas wild-type PRP8 supports growth in up to 0.125 mM copper, the prp8-syf strains show growth defects in the presence of 0.05 mM copper. The prp8-syf alleles did not affect the splicing of the wild-type ACT1-CUP1 reporter or the splicing of reporter genes carrying alterations in the 5' splice site or in the branchpoint. In addition, they did not show any alteration of pyrimidine tract recognition (data not shown). Therefore, we propose that the prp8-syf alleles affect 3' splice site usage antagonistically to the prp8-scf alleles, reducing splicing in the presence of the 3' UUG alteration.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The results described here demonstrate extensive genetic interactions between the PRP17 and the PRP8 splicing factor genes. The involvement of both genes in the second step of the splicing reaction has been shown previously (VIJAYRAGHAVAN et al. 1989 ; UMEN and GUTHRIE 1995A , UMEN and GUTHRIE 1995B , UMEN and GUTHRIE 1996 ; BEN-YEHUDA et al. 1998 ). Here we report that in the absence of Prp17p, mutations in PRP8 can either suppress the temperature-sensitive phenotype (prp8-scf mutations) or confer synthetic lethality (prp8-syf mutations). In addition, we demonstrate a strong correlation between the ability of prp8 alleles to suppress the phenotype of the PRP17 allele and their ability to recognize mutant 3' splice sites. Conversely, we show that the previously described prp8-122 3' fidelity mutant allele (UMEN and GUTHRIE 1996 ) is able to suppress the temperature sensitivity of a PRP17 strain. The correlation between PRP17 suppression and 3' splice site recognition is strengthened by the observation that the prp8-syf mutations, which are synthetic lethal with the PRP17 allele, reduce the ability of the cells to recognize mutant 3' splice sites. Neither the prp8-scf nor the prp8-syf alleles show any effect on splicing of introns bearing alterations in the 5' splice site, in the branchpoint conserved sequences, or in the uridine tract, suggesting that the prp8 mutations specifically affect 3' splice site recognition.

On the basis of these observations we propose a model for the mechanism of interaction between the Prp8 and the Prp17 proteins. According to this model, the two proteins cooperate during the second step of splicing. Whereas Prp8p plays a role in recognizing the conserved PyAG motif, Prp17p serves as an accessory protein that enhances the efficiency of this recognition. Prp17p may act by helping to position Prp8p at the correct site on the pre-mRNA. Alternatively, the Prp17 protein might act by producing a conformational change in Prp8p, which enhances the recognition step. In the absence of Prp17p, Prp8p is still able to identify the proper splice site, albeit less efficiently, and it is therefore proficient enough to enable growth at the permissive temperature (25°). At the restrictive temperature, the interaction of Prp8p with the PyAG motif becomes unstable, and the assistance of the Prp17 protein becomes essential. The requirement for Prp17p may be critical for the splicing of specific introns. Failure to splice these introns could account for the cell cycle arrest phenotype seen in the absence of Prp17p or in the presence of certain Prp8 mutant proteins (DUMAS et al. 1992 ; SHEA et al. 1994 ; see below).

This hypothesis is supported by the following observations:

1. Cross-linking studies have demonstrated that Prp8p interacts directly with the pre-mRNA during both steps of the splicing reaction (TEIGELKAMP et al. 1995A , TEIGELKAMP et al. 1995B ; UMEN and GUTHRIE 1995A , UMEN and GUTHRIE 1995B ). During the second step of splicing, an interaction occurs between Prp8p and the 3' splice site after ATP hydrolysis by the RNA-dependent ATPase Prp16p. This interaction is enhanced by the presence of Prp17p (UMEN and GUTHRIE 1995A ).

2. PRP17 strains show enhanced defects in 3' splice site recognition. We have found that the splicing defect of PRP17 strains at the permissive temperature is stronger for a reporter gene carrying the 3' UUG splice site mutation; this effect is even more pronounced at the semipermissive temperature (data not shown). These results suggest that the PRP17 strain is defective in 3' splice site recognition in a way similar to that seen in strains bearing the synthetic lethal prp8-syf alleles.

3. The prp8-101 allele was isolated in a search for mutants with altered uridine tract recognition (UMEN and GUTHRIE 1995B ). This strain was later shown to exhibit a decreased level of cross-linking of Prp8p to the 3' splice site in vitro and to be synthetically lethal with mutations in the PRP17 gene (UMEN and GUTHRIE 1995A ). We have found that, as for the other synthetic lethal prp8-syf alleles, this mutation exhibits a strong defect in the ability to splice the UUG 3' splice site mutation (UMEN and GUTHRIE 1995B ; data not shown). Hence, all the tested prp8 alleles, which are synthetic lethal with PRP17, show greater discrimination in 3' splice recognition. Conversely, all the PRP17 suppressor mutations found in our work were alleles of PRP8 that are less discriminating in recognition of the 3' PyAG motif.

In addition to the Prp8 and Prp17 proteins, genetic data point to the involvement of several snRNAs in 3' splice site recognition. Mutations in two spliceosomal snRNAs, U2 and U6, were able to suppress PyAG alterations, similar to the prp8-scf alleles (LESSER and GUTHRIE 1993B ; MADHANI and GUTHRIE 1994B ). Interestingly, mutations in U2 snRNA that perturb the U2-U6 snRNA helix II interactions were found to be synthetic lethal with mutations in either PRP17 or PRP8 (XU et al. 1998 ), and one of the syf mutants obtained in our synthetic lethality screen also maps to the U2 snRNA (S. BEN-YEHUDA, unpublished data). It has been shown that the RNA-RNA interactions between U2 and U6 snRNAs form an active site involved for the second step of the splicing reaction (MADHANI and GUTHRIE 1994B ). Therefore, the Prp17 and Prp8 proteins may assist in bringing the active site of the spliceosome to the appropriate conformation during the second step of the splicing reaction.

Other biochemical and genetic observations provide links between yeast Prp8p and various factors in or near the catalytic centers of the spliceosome: (1) Prp8p has been proposed to trigger the unwinding of the U4/U6 duplex prior to the interaction of U6 with U2 to form part of the reaction center in the spliceosome (KUHAN et al. 1999 ); (2) before the second catalytic step, Prp8p binds the ends of both the exons that are to be joined, possibly being responsible for anchoring these in the catalytic center (TEIGELKAMP et al. 1995A , TEIGELKAMP et al. 1995B ; UMEN and GUTHRIE 1995A , UMEN and GUTHRIE 1995B ); and (3) Prp8p binds to the invariant loop of U5 snRNA that is critical for precisely aligning the end of exon 1 with the 3' splice site for the second step of splicing (DIX et al. 1998 ). Thus, Prp8p may serve as an anchor in the spliceosome, not only of the ends of the exons, but also to position snRNAs appropriately to form the reaction centers for each step, while Prp17 may be involved only in the second-step interactions that are mediated by Prp8p.

Why are the prp8-scf and the prp8-syf alleles more and less permissive, respectively, for changes in the PyAG motif? One possible mechanism is that the prp8-scf mutants encode proteins with a more flexible conformation. This increased flexibility may allow a higher efficiency of 3' splice site recognition at the expense of accuracy and thus could bypass the need for Prp17p that is required for 3' splice site recognition at the restrictive temperature. Conversely, prp8-syf mutations have the opposite effect: they cause a conformational change in Prp8p that reduces its flexibility and thereby increases the stringency of 3' splice recognition, such that the activity of Prp17p becomes essential even at the permissive conditions. A similar model was proposed to explain the ability of mutations in the U2 and U6 snRNAs to partially suppress the splicing defect caused by mutations in the 3' PyAG motif. It has been suggested that specific mutations in these snRNAs change the fit of the 3' splice site in the spliceosomal active site such that noncanonical 3' splice sites can be accommodated (MADHANI and GUTHRIE 1994B ). Examples of mutations that act through such a mechanism have been reported previously in other systems. For example, mutants encoding a bacterial serine protease with a broader range of substrate recognition were isolated. Crystallographic analysis showed that the decreased specificity is due to a greater flexibility of the active site (BONE et al. 1991 ). Similarly, a specific mutation in DNA polymerase ß of Escherichia coli decreases accuracy during DNA synthesis, resulting in an increase in both base substitutions and frameshift errors. X-ray crystallographic studies suggest that the mutation causes a conformational change that increases the flexibility of the polymerase (PELLETIER et al. 1996 ; OPRESKO et al. 1998 ). Not all of the 3' splice site alterations are equally affected by the prp8-scf mutations. For example, no increased copper resistance was observed in the prp8-scf strains with the reporter gene carrying a UGG mutation at the 3' splice site (pJU97). The only exception was the strong prp8-scf103A allele, which was resistant to higher copper concentrations than the wild-type allele in the presence of the 3' UGG reporter. This latter 3' splice site was more difficult to recognize than the 3' UUG reporter, by both mutants and wild-type PRP8 alleles (data not shown).

An alternative hypothesis to explain the effect of the prp8-scf mutations is based on the mechanism proposed for the role of the Prp16 protein. Mutations in the PRP16 gene that produce a protein with reduced ATPase activity are able to splice pre-mRNA molecules carrying altered branch site sequences. The mutations slow down the reaction, and thus may allow more time for aberrant lariats to proceed to productive splicing (BURGESS and GUTHRIE 1993 ). Similarly, it is possible that the reduced accuracy seen with the prp8-scf mutations results from a conformational change in Prp8p that reduces the kinetics of the second step of the splicing reaction. The slow reaction provides more time for both aberrant and optimal 3' splice sites to be processed in the splicing pathway. On the other hand, according to this proposal, one has to speculate that Prp17p exerts its function by slowing down the second step of the splicing reaction and that prp8-syf alleles encode proteins that increase the pace of this step, possibilities that are less plausible, but nonetheless should be tested.

Although it is tempting to speculate that the Prp17 protein could interact directly with the PyAG motif during 3' splice site recognition, Prp17p does not contain any known RNA-binding motif that may justify this assumption. The Prp17 protein contains several copies of the WD repeat (VAISMAN et al. 1995 ; BEN-YEHUDA et al. 1998 ; ZHOU and REED 1998 ), which probably serves in protein-protein recognition (SIKORSKI et al. 1990 ; WILLIAMS et al. 1991 ). However, we were unable to detect physical interactions between Prp8p and Prp17p using the two-hybrid methodology (unpublished data). Thus, the interaction between Prp17p and Prp8p, which allows efficient and accurate recognition of the PyAG motif, may take place indirectly through other proteins. The clustering of scf and syf alleles to a small region of Prp8p defines the region that mediates the interactions between this protein and Prp17p. A possible candidate for a protein that may serve as a link between Prp8p, Prp17p, and the 3' splice site is Slu7p. Slu7p has been shown to interact with both Prp8p and Prp17p during the second step of the splicing reaction (FRANK and GUTHRIE 1992 ; FRANK et al. 1992 ; UMEN and GUTHRIE 1995A ). It has been suggested that Slu7p plays a role in the recognition of the 3' splice site, since its requirement increases with the distance between the branchpoint and the 3' splice site (BRYS and SCHWER 1996 ).

The PRP17 gene was identified originally through the temperature-sensitive mutation PRP17-1, which affects both the mitotic and meiotic cell cycles (KASSIR and SIMCHEN 1978 ). Interestingly, some mutant alleles of PRP8 (dbf3, dna39) were also isolated as cell-cycle-specific mutants (DUMAS et al. 1992 ; SHEA et al. 1994 ). The phenotype of dbf3-1 strongly resembles that of PRP17 strains: dbf3-1 cells held at the restrictive temperature show a cell cycle arrest at the G2/M transition, a delayed entry into the S-phase, and sensitivity to hydroxyurea (HU), a well-characterized inhibitor of DNA synthesis. These phenotypes were absent from other tested mutant alleles of PRP8 (SHEA et al. 1994 ). The extensive genetic interactions between the PRP17 and PRP8 genes presented here suggest that the phenotypes of the dbf3-1 allele originate from the loss of interaction between these proteins; in the dbf3-1 allele of PRP8, Prp17p cannot exert its function, resulting in similar phenotypes to those seen in the absence of the Prp17 protein.

How could mutations in the second step of the splicing reaction account for the associated cell cycle arrest? The yeast S. cerevisiae has highly conserved 5' splice site and branch site sequences, but it shows limited conservation of 3' splice site sequences (GUTHRIE 1991 ). This variability could be a target of regulation, allowing the controlled splicing of specific RNA molecules. One possible mechanism for splicing regulation is that a small number of cell-cycle-specific, intron-containing genes may require special splicing factors for their correct expression. Therefore, it is possible that Prp17p is essential for the efficient splicing of genes involved in cell cycle progression, which contain some unique features at the 3' splice site of the introns. A search in the databases reveals many intron-containing yeast genes that may affect cell cycle progression. The identity of the genes potentially involved in such a mechanism, however, remains to be determined, since there are many possible variations in 3' sequences/environments that could be important for the recognition process. In an alternative mechanism, mutations in some splicing genes may disrupt the normal process of pre-mRNA splicing, eliciting a checkpoint response that arrests the cell cycle, similar to the one observed when the integrity of other cell components, such as the spindle or the DNA, is compromised (ELLEDGE 1996 ).

The connection between pre-mRNA splicing and cell cycle regulation was strengthened recently by the finding of a physical association in mammalian cells between cyclin E-Cdk2 and components of the U2 snRNA-associated proteins: SAP114, SAP145, and SAP155. The splicing proteins are phosphorylated, and inhibitors of Cdk activity, such as p21, inhibit their phosphorylation (SEGHEZZI et al. 1998 ). Therefore, it is possible that certain introns may be removed at a specific stage of the cell cycle, at which splicing factors become activated via phosphorylation and dephosphorylation activities. The conservation of the Prp17 and Prp8 proteins throughout the evolutionary scale (ANDERSON et al. 1989 ; PINTO and STEITZ 1989 ; HODGES et al. 1995 ; BEN-YEHUDA et al. 1998 ; ZHOU and REED 1998 ) suggests that a similar cell cycle control through splicing may exist in other organisms and it will be interesting to determine whether phosphorylation plays a role in this process.

	ACKNOWLEDGMENTS

We thank C. Guthrie and C. Collins for their generous gifts of reagents. This work was supported by grants to M.K. by the Israel Cancer Association, the Israel Cancer Research Fund, and the Recanati Foundation, and to J.D.B. by the Wellcome Trust (no. 047685). S.B.-Y. was a recipient of a travel scholarship from the British Council. J.D.B. is supported by a Royal Society Cephalosporin Fund Senior Research Fellowship.

Manuscript received June 9, 1999; Accepted for publication September 10, 1999.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ANDERSON, G. J., M. BACH, R. LUHRMANN, and J. D. BEGGS, 1989  Conservation between yeast and man of a protein associated with U5 small nuclear ribonucleoprotein. Nature 342:819-821[Medline].

BEGGS, J. D., 1995 Yeast splicing factors and genetic strategies for their analysis, pp. 79–95 in Pre-mRNA Processing, edited by A. I. LAMOND. R. G. Landes Company, Austin, TX.

BEN-YEHUDA, S., I. DIX, C. S. RUSSELL, S. LEVY, and J. D. BEGGS et al., 1998  Identification and functional analysis of hPRP17, the human homologue of the PRP17 yeast gene involved in splicing and cell cycle control. RNA 4:1304-1312[Abstract].

BOEKE, J. D., J. TRUEHEART, G. NATSOULIS, and G. R. FINK, 1987  5-Fluoroorotic acid as a selective agent in yeast molecular genetics. Methods Enzymol. 154:165-175.

BOGER-NADJAR, E., N. VAISMAN, S. BEN-YEHUDA, Y. KASSIR, and M. KUPIEC, 1998  Efficient initiation of S-phase in yeast requires Cdc40p, a protein involved in cell cycle progression and pre-mRNA splicing reactions. Mol. Gen. Genet. 260:232-241[Medline].

BONE, R., A. FUJISHIGE, C. A. KETTNER, and D. A. AGARD, 1991  Structural basis of broad specificity in alpha lytic protease mutants. Biochemistry 30:10388-10398[Medline].

BOORSTEIN, W. R. and E. A. CRAIG, 1989  Primer extension analysis of RNA. Methods Enzymol. 180:347-369[Medline].

BRYS, A. and B. SCHWER, 1996  Requirement for the SLU7 in yeast pre-mRNA splicing is detected by the distance between the branchpoint and the 3' splice site. RNA 2:707-717[Abstract].

BURGESS, S. M. and C. GUTHRIE, 1993  A mechanism to enhance mRNA fidelity: the RNA-dependent ATPase Prp16 governs usage of a discard pathway for aberrant lariat intermediates. Cell 73:1377-1391[Medline].

DIX, I., C. S. RUSSELL, R. T. O'KEEFE, A. J. NEWMAN, and J. D. BEGGS, 1998  Protein-RNA interactions in the U5 snRNP of S. cerevisiae.. RNA 4:1239-1250[Abstract].

DUMAS, L. B., J. P. LUSKY, E. J. MCFARLAND, and J. SHAMPAY, 1992  New temperature-sensitive mutants of Saccharomyces cerevisiae affecting DNA replication. Mol. Gen. Genet. 187:42-46.

ELLEDGE, S. J., 1996  Cell cycle checkpoints: preventing an identity crisis. Science 274:1664-1672[Abstract/Free Full Text].

FRANK, D. and C. GUTHRIE, 1992  An essential splicing factor SLU7, mediates 3' splice site choice in yeast. Genes Dev. 6:2112-2124[Abstract/Free Full Text].

FRANK, D., B. PATTERSON, and C. GUTHRIE, 1992  Synthetic lethal mutations suggest interactions between U5 small nuclear RNA and four proteins required for the second step of splicing. Mol. Cell. Biol. 12:5197-5205[Abstract/Free Full Text].

GIETZ, R. D. and A. SUGINO, 1988  New yeast-Escherichia coli shuttle vectors constructed with in vitro mutagenized yeast genes lacking six-base pair restriction sites. Gene 74:527-534[Medline].

GUTHRIE, C., 1991  mRNA splicing in yeast: clues to why the spliceosome is a ribonucleoprotein. Science 253:157-163[Abstract/Free Full Text].

HANS-RUDOLF, H. and B. SCHWER, 1998  Mutational analysis of the yeast DEAH-box splicing factor Prp16. Genetics 149:807-815[Abstract/Free Full Text].

HODGES, P. E., S. P. JACKSON, J. D. BROWN, and J. D. BEGGS, 1995  Extraordinary sequence conservation of the PRP8 splicing factor. Yeast 11:337-342[Medline].

HOROWITZ, D. S. and J. ABELSON, 1993  Stages in the second reaction of pre-mRNA splicing: the final step is ATP independent. Genes Dev. 7:320-329[Abstract/Free Full Text].

JONES, M. H., D. N. FRANK, and C. GUTHRIE, 1995  Characterization and functional ordering of Slu7p and Prp17p during the second step of pre-mRNA splicing in yeast. Proc. Natl. Acad. Sci. USA 92:9687-9691[Abstract/Free Full Text].

KASSIR, Y. and G. SIMCHEN, 1978  Meiotic recombination and DNA synthesis in a new cell cycle mutant of Saccharomyces cerevisiae.. Genetics 90:49-58[Abstract/Free Full Text].

KASSIR, Y., M. KUPIEC, H. SHALOM, and G. SIMCHEN, 1985  Cloning and mapping of PRP17, a S. cerevisiae gene with a role in DNA repair. Curr. Genet. 9:253-257[Medline].

KRAMER, A., 1996  The structure and function of proteins involved in mammalian pre-mRNA splicing. Annu. Rev. Biochem. 65:367-409[Medline].

KUHAN, A. N., L. ZAIRONG, and D. A. BROW, 1999  Splicing factor Prp8 governs U4/U6 RNA unwinding during activation of the spliceosome. Mol. Cell 3:65-75[Medline].

KUPIEC, M. and G. SIMCHEN, 1986  DNA-repair characterization of PRP17-1, a cell cycle mutant of S. cerevisiae.. Mutat. Res. 162:33-40[Medline].

LESSER, C. F. and C. GUTHRIE, 1993a  Mutational analysis of pre-mRNA splicing in Saccharomyces cerevisiae using a sensitive new reporter gene, CUP1.. Genetics 133:851-863[Abstract].

LESSER, C. F. and C. GUTHRIE, 1993b  Mutations in U6 snRNA that alter splice site specificity: implications for the active site. Science 262:1982-1988[Abstract/Free Full Text]