INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

RNase MRP was first identified in mouse cells as an endoribonuclease involved in mitochondrial DNA replication (4-6). It cleaves RNA transcripts from mitochondrial origins of replication in a manner consistent with the formation of the primer for the initiation of DNA synthesis (23-25, 60), hence, the name MRP, for mitochondrial RNA processing. Like the mammalian RNase MRP, the yeast enzyme cleaves an RNA substrate complementary to a yeast mitochondrial origin of replication at an exact site of RNA-to-DNA linkage (2, 52, 57). However, the majority of RNase MRP is localized in the nucleus (20, 26, 39), where it directly processes the rRNA precursor at the A3 site (7, 9, 28, 44). The loss of RNase MRP activity leads to the loss of the production of one of two different species of 5.8S rRNA (41, 44). Some question has arisen as to whether the loss of processing of the rRNAs is the essential function of RNase MRP (28, 32). Recent work has identified a late cell cycle arrest in RNase MRP mutants, and it is some function in the control of the cell cycle that may be responsible for the essentiality of RNase MRP (2a).

RNase MRP is a ribonucleoprotein complex with a single RNA component and several protein subunits (3, 8, 10, 29, 43, 45). The 340-nucleotide (nt) RNA component of RNase MRP is encoded by an essential nuclear gene, NME1 (43). RNase MRP shares many functional and structural similarities with another ribonucleoprotein complex, RNase P, which is involved in pre-tRNA processing (14, 37, 38, 46). Both ribonucleoproteins have sequence-specific endoribonuclease activity and have functions in both the mitochondria and the nucleus. The RNA components of both complexes can be folded into similar secondary structures (14, 46). In the yeast Saccharomyces cerevisiae, RNase P and RNase MRP share several protein components: Pop1p (100.5 kDa), Pop3p (22.6 kDa), Pop4p (32.8 kDa), Pop5p (19.6 kDa), Pop6p (18.2 kDa), Pop7p (15.8 kDa), Pop8p (15.5 kDa), and Rpp1p (32.2 kDa) (3, 8, 10, 29, 53). Pop1p, Pop3p, and Pop4p were identified by genetic screening. Rpp1p was identified as a protein subunit of RNase P by virtue of its homology with a human scleroderma autoimmune antigen, Rpp30, which copurifies with human RNase P. Yeast RNase P has been purified biochemically to apparent homogeneity. Four more proteins that purify with RNase P activity, Pop5p (19.6 kDa), Pop6p (18.2 kDa), Pop7p/Rpp2p (15.8 kDa), and Pop8p (15.5 kDa), have been demonstrated to be shared by RNase P and RNase MRP (3).

Snm1p, the only protein component unique to the RNase MRP complex (not shared with RNase P), was identified as a high-copy-number suppressor of a temperature-sensitive nme1 mutant (45). SNM1 at a high copy number can fully complement the rRNA-processing defect of this nme1 mutant at the nonpermissive temperature. Snm1p binds the RNase MRP RNA but not the RNase P RNA. The SNM1 gene encodes an essential protein of 198 amino acids with a predicted molecular mass of 22.5 kDa. The primary sequence reveals three possible structural motifs. The amino-terminal portion of the protein contains a series of leucines spaced 6 to 8 amino acids apart that could potentially form a coiled-coil domain. The central portion of the protein contains two cysteine clusters that may function in binding zinc ions. Similar cysteine clusters can be found in a binuclear cluster arrangement seen in DNA-binding proteins (31, 54). The carboxy-terminal portion of the protein is rich in serines and lysines. Of the last 63 amino acids, 15 are serines and 16 are lysines. There is no obvious RNA binding motif in Snm1p, although it was found to specifically bind the MRP RNA (45). In order to learn more about the function of this protein that is exclusive to the RNase MRP enzyme complex, we undertook mutagenesis screening of the SNM1 gene. Here we report the isolation and characterization of a number of temperature-conditional SNM1 mutants. These mutants have led us to find a previously undiscovered role for the yeast RNase MRP in chromosomal segregation.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Strains and media. Yeast media and genetic manipulations have been described elsewhere (48). The Escherichia coli strain used for cloning, DH5, has the genotype psi80dlacZM15 endA1 recA1 hsdR17 (r_K m_K⁺) supE44 thi-1 gyrA96 relA1 (lacZYA-argF)U169 F. Basic molecular biology techniques were performed as described elsewhere (42).

PCR mutagenesis of SNM1. PCR mutagenesis was performed as described by Muhlrad et al. (34). Plasmid DNA carrying a full-length SNM1 gene was used as a template. The PCR mixtures (total reaction volume, 100 µl) contained 1 µg of DNA template (pMES193 [pBS containing SNM1]) (45); Vent DNA polymerase buffer (New England Biolabs, Beverly, Mass.); either 1.0 mM each dGTP, dCTP, and dTTP and 0.2 mM dATP or 1.0 mM each dATP, dCTP, and dTTP and 0.2 mM dGTP; 3 U of Deep VentR (exo) DNA polymerase (New England Biolabs); 0.5 mM MnCl₂; and 1 µg each of primers O-SNM-2 (5'-CTG CAG GCA AGG CGT ACA TGG CGT G-3') and O-SNM-5 (5'-GAT GAG TAC TAC TAG TGG CTA-3'). Thirty cycles were completed at an annealing temperature of 40°C. PCR products of the correct size (~845 bp) were purified from agarose gels by filter centrifugation.

ade2-1 leu2-3,112 his3-200 trp1-1 ura3-52 snm1-1::HIS3

Growth testing of yeast mutants. Mutant yeast strains were analyzed for temperature-conditional growth on SCD plates at 17, 24, 30, and 37°C. One hundred microliters of sterile water was distributed into the wells of a 96-well Falcon tissue culture plate. Two fresh colonies of each mutant yeast strain were mixed into the first-row wells. Tenfold serial dilutions of these cell mixtures were placed into the second-, third-, and fourth-row wells. A 48-pin Frogger (Dan-kar Corp.) was used to transfer diluted cell mixtures to SCD plates. The cells were grown at the temperatures specified for 2 days or at 17°C for 5 days.

Analysis of yeast RNA. RNA was extracted as previously described (47). Approximately 10 µg of whole-cell RNA was separated on 6% acrylamide-7 M urea gels. The gels were stained with ethidium bromide to visualize the 5.8S rRNA, or Northern blot analysis was performed (44). The probes used were SNM1 (a 705-bp BamHI-NsiI fragment of the SNM1 gene), NME1 (a 1,042-bp EcoRI fragment), and SCR1 (a 736-bp EcoRI-SpeI fragment) (20, 21, 39). Probes were radiolabeled for hybridization with [-³²P]dCTP by use of the Prime-It Kit (Stratagene, Inc.). Radioactive blots were analyzed on a Molecular Dynamics PhosphorImager.

Preparation of yeast cell extracts. Yeast cells were grown in 50 ml of liquid SCD at 30°C until they reached exponential growth (10⁷ cells/ml). The cultures were then shifted to the nonpermissive temperature and grown for 6 h. The cells were collected, washed with 1 ml of buffer A (20 mM Tris-HCl [pH 8], 150 mM KCl, 5 mM EDTA, 0.1% Triton X-100, 10% glycerol, 1 mM dithiothreitol [DTT], 1 mM phenylmethylsulfonyl fluoride [PMSF]), pelleted in a microcentrifuge, and resuspended in 400 µl of buffer A. Two hundred milligrams of glass beads was added to the cell mixture, which was then vortexed for 15 min at 4°C and centrifuged at 16,000 × g for 2 min at 4°C. The supernatant was transferred to a clean tube and centrifuged at 16,000 × g for an additional 15 min. Three hundred microliters of the supernatant was collected, and the protein concentration was determined (30).

Immunoprecipitation by anti-Snm1p antibody. Immunoprecipitation was performed as described before (45). Fifty microliters of protein A-Sepharose CL4B beads (200 mg/ml) was incubated with 3 µl of rabbit serum containing anti-Snm1p antibody for 1 h at 4°C, and the mixture was washed three times with buffer A. The bead volume was brought up to 1.25 ml with buffer A, and the mixture was incubated with 375 µg of yeast extract for 1 h at 4°C. The samples were washed four times with 1 ml of buffer A, the volume of each sample was brought up to 100 µl with buffer A, 100 µl of 5% sodium dodecyl sulfate (SDS)-10% -mercaptoethanol was added, and the samples were heated to 95°C for 5 min. After the beads were removed by centrifugation, the RNA was extracted with phenol-chloroform equilibrated to pH 5.3, precipitated with ethanol, and examined by Northern analysis.

Affinity purification of anti-Snm1p antibody. Snm1p was expressed in E. coli BL21(DE3) cells by use of a T7 promoter system as described before (45). After induction with isopropyl--D-thiogalactopyranoside, (IPTG), cells were pelleted and washed with lysis buffer (50 mM Tris-HCl [pH 8.0], 100 mM KCl, 5 mM EDTA, 1 mM DTT, 1 mM PMSF). Cells were then sonicated and washed with lysis buffer containing 0.5% Triton X-100. Inclusion bodies were pelleted, solubilized in 6 M guanidine chloride, and diluted in 50 mM Tris-HCl-0.5 mM DTT-10 µM ZnCl₂; about 60% of Snm1p was soluble. Snm1p was concentrated by use of an Amicon concentrator with a molecular mass cutoff of 10,000 Da. Approximately 10 mg of Snm1p was coupled to 6 ml of Affi-Gel 10 (Bio-Rad Corp., Hercules, Calif.) according to the instructions of the manufacturer, and 25 ml of serum was loaded on an Snm1p column and washed with 25 column volumes of 10 mM Tris (pH 7.5) and then with 25 column volumes of 10 mM Tris (pH 7.5)-0.5 M NaCl. Antibody was eluted with 10 column volumes of 100 mM glycine (pH 2.5). The column was equilibrated with 10 mM Tris (pH 8.8), and antibody was eluted with 10 column volumes of 100 mM triethylamine (pH 11.5). The two eluates were combined and dialyzed overnight against phosphate-buffered saline (PBS). The antibody was precipitated with 50% ammonium sulfate, resuspended in 10 ml of PBS, and dialyzed against PBS.

Western blot analysis. SDS-polyacrylamide gel electrophoresis (PAGE) was performed as previously described (47). Forty micrograms of yeast protein extract was denatured in 2% SDS-5% -mercaptoethanol for 5 min at 95°C and resolved on a 15% polyacrylamide gel. The separated proteins were electrophoretically transferred to a BA-S NC-supported nitrocellulose membrane (Schleicher & Schuell, Inc., Keene, N.H.) by use of a Bio-Rad semidry transfer cell. The membrane was stained for 10 min with Ponceau S solution (0.1% Ponceau S, 5% acetic acid) to ensure proper transfer. The membrane was blocked with 5% milk-TBST (20 mM Tris-Cl [pH 7.6], 150 mM NaCl, 0.1% Tween 20) for 1 h at 24°C, incubated with affinity-purified anti-Snm1p antibody (1:500 dilution in 5% milk-TBST) for 1 h, washed with TBST three times for 10 min each time, incubated with peroxidase-anti-rabbit antibody (1:10,000 dilution in TBST; Boehringer Mannheim Biochemicals, Indianapolis, Ind.) for 30 min, and washed with TBST three times for 10 min each time. The peroxidase-anti-rabbit antibody was detected with a Boehringer chemiluminescence Western blotting kit and exposed to film for 10 min and then again overnight.

Plasmid segregation assay. A 3.6-kb BamHI fragment containing the ADE2 gene was cloned into the pRS316 and YCPlac33 vectors to make the reporter plasmids pTC185 and pTC186, respectively (15, 51). Both reporter plasmids were transformed into THR200-derived yeast strains carrying the appropriate SNM1 allele by selecting for the URA3 marker on the plasmids (11). snm1 mutant strains carrying an ADE2 reporter plasmid were grown at 30°C in SCD overnight and diluted in sterile water, and 500 cells of each yeast strain were plated onto 150-mm YPD (1% yeast extract, 2% peptone, 2% dextrose) plates. The plates were scanned by use of a Linotype Hell JADE² scanner after 3 days of growth at 30°C, when the color was fully developed.

Tubulin immunofluorescence. Cells were prepared for immunofluorescence as previously described (36). Rat monoclonal antitubulin antibody YOL1/34 and indocarbocyanine (Cy3)-conjugated rabbit anti-rat secondary antibody were obtained from Accurate Chemical Corp., Westbury, N.Y. Cells were viewed by use of a Ziess Axioskop microscope equipped with epifluorescence and Narmoski optics and a Zeiss Plan-Apochromat 100× objective. Images were captured in real time by use of a Diagnostics Instruments Spot Camera 2 directly linked to a Macintosh PowerMac G3 computer.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Isolation of conditional mutations in the SNM1 gene. Random PCR mutagenesis of the SNM1 gene was performed to further define the functional regions of the Snm1 protein. Plasmid pTHR100 (pRS315 containing SNM1) was digested with restriction enzymes BamHI and NsiI to remove a 696-bp fragment containing the SNM1 coding region. At the same time, plasmid pMES193 (pBS containing SNM1) was used as a template to produce a linear fragment containing the SNM1 gene by use of PCR. The nucleotide concentrations in the PCR were made unequal to encourage mutations. The mutagenized snm1 PCR product was cotransformed with gap-containing plasmid pTHR100 into haploid yeast strain THR200, with selection for the LEU2 gene. Yeast cells will repair the gap-containing plasmid with the PCR product by recombining homologous regions found at the ends of the PCR product and the ends of the gap-containing plasmid. Yeast strain THR200 contains a chromosomal deletion of the SNM1 gene and a wild-type copy of SNM1 on a URA3-based CEN plasmid (YCp50) (40). Transformants were replicated to two successive 5-FOA plates and then replicated to SCD plates at various temperatures between 17 and 37°C.

View larger version (121K): [in this window] [in a new window]   FIG. 1.   Growth phenotypes of yeast strains carrying a mutant snm1 gene. Two colonies from each mutant yeast strain were mixed with 100 µl of sterile double-distilled H2O in the first-row wells of a 96-well Falcon tissue culture plate; 10-fold serial dilutions of these cell mixtures were placed in the second-, third-, and fourth-row wells. A 48-pin Frogger was used to transfer diluted cell mixtures to SCD plates. The cells were grown for 2 days at 24, 30, and 37°C and for 4 days at 17°C. SNM1 alleles are listed in Table 1. All SNM1 alleles are carried on LEU2 CEN vector pRS315 (51) in strain THR200. WT, wild type.

Defects in 5.8S rRNA processing in snm1 mutants. The severity of the mutations and the resulting growth phenotypes suggested that the defects in Snm1p affected the function of the RNase MRP enzyme complex. Although Snm1p has been shown to be an essential protein that is closely associated with the MRP RNA, it has not been directly shown to be required for the function of the MRP enzyme (45). Mutations in nme1, the gene for the MRP RNA, or in Pop1p, another protein component of yeast RNase MRP, lead to defective 5.8S rRNA processing (7, 29, 44). In addition, transcriptional depletion of mRNAs for other RNase MRP protein components leads to a similar phenotype (3, 8). Two species of 5.8S rRNA that differ in length by only 7 nucleotides exist in yeast (41). Normally, they are present at a ratio of 10:1, small to large. Loss of yeast RNase MRP components results in a loss of processing of the smaller 5.8S rRNA. To see if mutations in the snm1 gene produce the same result, mutants were screened for anomalies in 5.8S rRNA processing. The mutants were grown at 30°C until they reached the exponential phase and then were shifted to the nonpermissive temperature to determine the effect of growth at the nonpermissive temperature on 5.8S rRNA processing. Whole-cell RNA extracts from the mutants were analyzed on 6% acrylamide-7 M urea gels. Nearly all the mutants displayed some defect in 5.8S rRNA processing. The severity of the temperature-dependent growth seen in the mutants closely coincided with the severity of the defect seen in 5.8S rRNA processing (Fig. 2A). In addition, nearly all of the mutants displayed some loss of 5.8S rRNA processing at the permissive temperature (data not shown). These results demonstrate that Snm1p is required for the function of the RNase MRP enzyme in rRNA processing.

View larger version (69K): [in this window] [in a new window]   FIG. 2.   rRNA processing, RNase MRP RNA association, and Northern analysis of MRP RNA in snm1 mutant strains. Yeast strains were grown to 107 cells/ml at 30°C in YPD and then shifted to the nonpermissive temperature for 6 h. Cells were harvested, total RNA was isolated from one-half of the culture, and whole-cell protein extracts were made from the other half (see Materials and Methods). SNM1 alleles are listed in Table 1. All SNM1 alleles are carried on LEU2 CEN vector pRS315 (51) in strain THR200. (A) Ten micrograms of total RNA from the indicated strains was separated on 6% acrylamide-7 M urea gels and examined by staining with ethidium bromide. The locations of the relevant 5.8S(S), 5.8S(L), and 5.8Sb RNAs are shown. The ratios of the two 5.8S bands are provided in Table 1. WT, wild type. (B) Northern analysis of the RNase MRP RNA that coimmunoprecipitated with the anti-Snm1p antibody. Four hundred micrograms of protein extract from each mutant was incubated with polyclonal anti-Snm1p antibody. Antibody-protein RNA complexes were collected as described in Materials and Methods. RNA that coimmunoprecipitated with the antibody was separated on 6% acrylamide-7 M urea gels, and the MRP RNA was detected by hybridization with a 32P-labeled probe corresponding to a 1,042-bp EcoRI fragment containing the NME1 gene (43). (C) Northern analysis of the RNase MRP RNA in yeast snm1 mutant strains. Ten micrograms of whole-cell RNA isolated from yeast snm1 mutant strains was separated on 6% acrylamide-7 M urea gels and blotted to nylon membranes. The MRP RNA was visualized by hybridization with 32P-labeled probes corresponding to a 1,042-bp EcoRI fragment containing the NME1 gene (43) and a 736-bp EcoRI-SpeI fragment containing the SCR1 gene as a loading control (12).

The stability of the MRP RNA is not affected in snm1 mutants. The RNA component of yeast RNase MRP is essential for the proper functioning of the enzyme and cellular viability. It is known that Snm1p is associated with the MRP RNA in vivo (45). It was postulated that mutations in Snm1p could compromise the stability of the RNA component, explaining the mutant phenotypes seen. To test this possibility, Northern analyses were performed on whole-cell RNAs isolated from the mutant yeast strains after they were shifted to nonpermissive temperatures. Whole-cell RNA extracts from the mutants were analyzed for the presence of the MRP RNA and SCR1 RNA, an abundant cytoplasmic RNA, used as a loading control (12). As indicated in Fig. 2C, MRP RNA stability was unaffected in all of the snm1 mutants. This result is in contrast to the rapid loss of MRP RNA stability caused by a mutation in the pop1 gene (29).

The association of Snm1p with the MRP RNA changes in the mutants. Antibody against Snm1p has been previously shown to coimmunoprecipitate the MRP RNA. Coimmunoprecipitation experiments with mutant Snm1 proteins were performed to test if this interaction was maintained (45). Anti-Snm1p antibody was bound to protein A-Sepharose beads. Whole-cell mutant yeast extracts were incubated with the anti-Snm1p antibody attached to protein A-Sepharose beads and washed several times. The RNA was removed from the beads and analyzed by Northern analysis for the presence of both the MRP RNA and SCR1 RNA as a negative control. As shown in the Northern blot analyses of the coimmunoprecipitation experiments (Fig. 2B), wild-type controls were able to precipitate the MRP RNA by use of the anti-Snm1p antibody. However, many of the snm1 mutants failed to precipitate appreciable amounts of the MRP RNA, indicating that these mutants have a reduced MRP RNA association in vivo. Similar results were seen whether extracts were made from cells at either the permissive or the nonpermissive temperature (data not shown). Immunoprecipitations were repeated at least three times for each mutant, with similar results each time. In addition, degradation of nonprecipitated MRP RNA was never seen. Precipitated MRP RNA was undetectable in this assay for the p3, p9, p21, p22, p39, p31, p50, S56, p14, S33, 172, and p10 mutants. Mutants p2, S53, S8, S64, and 170 were able to precipitate some MRP RNA, although to a lesser extent than the wild type. The p18 mutant consistently precipitated more MRP RNA than the wild type. Unlike the processing of the 5.8S rRNA, there was no discernable correlation between Snm1p binding to the MRP RNA and the growth phenotype of the mutant. Many mutant strains, such as S56, grew quite well but still failed to bind any detectable RNase MRP RNA, while the p18 mutant appeared to bind more MRP RNA than the wild type yet grew slowly at most temperatures and had a severe rRNA-processing defect.

Essentiality testing of the SNM1 gene. In the snm1 mutants isolated, several frameshift and nonsense mutations were found. Some of these mutations were so severe (p14 had a frameshift mutation at Lys8, p3 had a nonsense mutation at Gln84, and S8 had a nonsense mutation at Glu53) that we expected no functional Snm1p was being made. However, all of these mutants, regardless of the severity of the mutations, could grow at least moderately well, if not as well as the wild type, at 30°C, despite little or no detectable MRP RNA binding. This result suggests that Snm1p may not be essential for the viability of yeast cells at 30°C. This notion would be in conflict with a previous deletion analysis of the SNM1 gene (45) and suggested the possibility that some other gene is tightly linked to the SNM1 gene and that a loss of the function of the linked gene is responsible for the lack of viability observed for snm1 deletion strains. We tested this possibility by using three different approaches. First, we placed the SNM1 coding sequence (from ATG to TAG) in between an ACT1 promoter and terminator and transformed this construct into yeast strain THR200, which has a chromosomal deletion of the SNM1 gene and is maintained by a wild-type SNM1 gene on a URA3-based plasmid. After removal of the wild-type SNM1 plasmid by growth on 5-FOA plates, we found that the heterologous promoter construct could fully complement the chromosomal deletion of the SNM1 gene at all temperatures (data not shown). This result indicated that the SNM1 coding region was required for viability. Second, we combined three different snm1 nonsense and frameshift mutations (p14, S8, and p18) to make all of the possible double and triple mutations. All of these mutations were unable to complement the chromosomal deletion, demonstrating that Snm1p was required and eliminating the possibility that the SNM1 RNA transcript was required for cell viability. Indeed, this result suggested that the snm1 mutants were translationally suppressed. Third, we had previously constructed a carboxy-terminal hemagglutonin (3×HA)-tagged version of the SNM1 gene. This tagged version complemented an snm1 deletion, but with reduced protein levels. We combined this 3×HA-tagged version of SNM1 with snm1 nonsense and frameshift mutations (p3 and p14). Neither of these tagged mutations could complement a chromosomal snm1 deletion, indicating that translational readthrough to the carboxy-terminal end of Snm1p was required for complementation of these mutations.

Reduction in Snm1p levels in snm1 mutants. An affinity-purified anti-Snm1p antibody was used in the Western analysis to examine Snm1p levels in snm1 mutants (45). As shown in Fig. 3, this antibody readily detects Snm1p in yeast cell extracts. In addition, this antibody cross-reacts with another protein of approximately 27 kDa, providing a convenient loading control. A yeast strain with the SNM1 gene on a high-copy-number plasmid (WT/2µ) produces a signal with an intensity stronger than that of a wild-type single-copy SNM1 strain (wild type or WT/CEN), confirming that this band is Snm1p. In all the mutants shown (S8, p3, p14, p18, and p21), Snm1p levels are very low. However, after a long exposure, a full-length Snm1p band is detected for most of these nonsense and frameshift mutations. This result suggests that very little Snm1p is required for cell viability and that these nonsense and frameshift mutations are translationally suppressed, but at a very low rate. Surprisingly, very little p18 protein is detected despite increased immunoprecipitation of the MRP RNA from this mutant. A low-abundance band with the molecular weight expected for the truncated p18 protein can be reproducibly seen in the Western analysis. This low-abundance protein may have a higher affinity than the wild-type protein for the MRP RNA and/or polyclonal anti-Snm1 antibodies when in its native configuration. The results obtained with mutant p18 show that the last third of Snm1p is dispensable for MRP RNA binding but not for RNase activity.

View larger version (57K): [in this window] [in a new window]   FIG. 3.   Western analysis of snm1 mutants. Whole-cell protein extracts were made from yeast strains grown to 107 cells/ml at 30°C in SCD (see Materials and Methods). Forty micrograms of protein extracts was denatured in 2% SDS-5% -mercaptoethanol for 5 min at 95°C. Proteins were separated by SDS-PAGE through a 15% polyacrylamide gel. Proteins were subsequently transferred to a nitrocellulose membrane and probed with an affinity-purified polyclonal anti-Snm1p antibody at a 1:500 dilution. The blot was probed with a secondary peroxidase-anti-rabbit antibody and visualized with the Boehringer chemiluminescence kit. The Snm1p band is indicated. The band above Snm1p is a nonspecific protein that serves as a convenient loading control. SNM1 alleles are listed in Table 1. SNM1 alleles WT/CEN, S8, S56, p3, p14, p18, p21, and 172 are carried on LEU2 CEN vector pRS315 (51) in strain THR200. The WT/2µ strain is identical to the other strains, except that the wild-type SNM1 gene is maintained on yeast 2µm plasmid YEp351 (18). WT is DBY2006, which contains a wild-type chromosomal SNM1 gene (43).

Other genes are not translationally suppressed in snm1 mutants. Since Snm1p and the RNase MRP enzyme are involved in ribosome biogenesis, we suspected that defects in RNase MRP could lead to the production of ribosomes that have compromised translational fidelity (55). To test this possibility, we looked for cross suppression of the yeast nonsense mutations lys1-1 (UAA), trp1-1 (UAG), tyr7-1 (UAG), ilv1-1 (UAG), ade3-26 (UAG), and met8-1 (UAG) (27) both by snm1 mutations and by mutations in the MRP RNA (NME1) (44). We were unable to detect translational suppression of any of these mutations by snm1 or nme1 mutations (data not shown). This result indicated that the translational suppression was specific for the SNM1 gene and that there is no general compromise of translational fidelity in snm1 mutants.

Increase in snm1 mRNA levels in snm1 mutants. Since we unexpectedly found translational suppression of the snm1 mutants, we decided to perform Northern analysis on the suppressed snm1 mutants. As shown in Fig. 4, the 800-nucleotide SNM1 transcript was barely detectable in a wild-type yeast strain (WT/CEN). However, in most of the mutants examined, snm1 mRNA levels were increased between 7- and 25-fold, to levels even higher than the level of expression from a wild-type SNM1 2µm plasmid (WT/2µ). This result suggests that high snm1 mRNA levels in nonsense and frameshift mutants may lead to sufficient translational suppression to sustain cell viability.

View larger version (89K): [in this window] [in a new window]   FIG. 4.   Northern analysis of SNM1 mRNA. Total RNA was isolated from yeast snm1 mutant strains grown to 107 cells/ml at 30°C in SCD (see Materials and Methods). Twenty micrograms of total RNA per lane was loaded on a 6% acrylamide-7 M urea gel and blotted to a nylon membrane. The mRNA corresponding to transcripts from the SNM1 allele was detected by hybridization with a 32P-labeled probe corresponding to the 696-bp BamHI-NsiI fragment containing the SNM1 coding region. The lower panel shows the location of the SNM1 transcripts. The upper panel shows the signal corresponding to the SNM1 mRNA, measured with a PhosphorImager and normalized to the mRNA level in WT/CEN, which was assigned a level of 1. SNM1 alleles are listed in Table 1. The WT/CEN and mutant SNM1 genes (S8, p3, p18, p21, and p31) are carried on LEU2 CEN vector YCplac111 (15) in strain THR200. The WT/2µ strain is identical to the other strains, except that the wild-type SNM1 gene is maintained on yeast 2µm plasmid YEp351 (18). I, the given allele was integrated into the yeast chromosome at the LEU2 gene in THR200 (see Materials and Methods).

Plasmid specificity of snm1 mutants. The mutant snm1 genes were maintained on a CEN LEU2 (pRS315) plasmid in all of the mutants (51). To examine if the mRNA level increase was plasmid specific, we transferred several of the snm1 mutations into plasmid YCplac111 and tested them for replacement of a wild-type SNM1 gene (15). All of the mutations still complemented a deletion strain at 30°C, except for the p14 mutation (frameshift at amino acid 8) (data not shown). We also placed several of the mutant snm1 genes into plasmid YIplac128 and integrated them into the chromosomal LEU2 gene. After integration, only the nonsense mutations were still able to complement the snm1 deletion at 30°C, and most of these nonsense mutants grew very slowly (only p18 grew at a rate comparable to that when maintained on a plasmid). None of the frameshift mutations were able to complement when integrated. These results were consistent with the placement of SNM1 on the plasmid playing a direct role in snm1 mRNA amplification and with past reports that frameshift mutations are more difficult to suppress than nonsense mutations (33).

Increase in plasmid copy number in snm1 mutants. To test if the large increase in mRNA expression was caused by plasmid amplification, copy numbers of the snm1-carrying plasmids were examined by Southern analysis. The signal was normalized to the single chromosomal copy of SNM1 in strain DBY2006 and normalized between strains by use of the single chromosomal copy of the SCR1 gene. This strategy allowed us to estimate plasmid copy numbers in the various strains. All the mutants showed increased snm1 plasmid copy numbers compared to the wild type, ranging from 5 to 15 copies per cell (Fig. 5). The wild-type CEN plasmid was estimated at about 1.5 copies per cell, and the wild-type 2µm plasmid was estimated at about 6 copies. However, most of the plasmids containing mutant snm1 genes were highly amplified even when CEN was present. We were able to find no other reports in the literature of this type of CEN-containing plasmid amplification and suppression. These results suggest that mutations in the snm1 gene may cause plasmid amplification, either by directly causing overreplication or missegregation because of a defective RNase MRP enzyme or by selecting for yeast that missegregates or overreplicates or by both methods. The integrated snm1 mutations were all maintained at one copy per cell.

View larger version (75K): [in this window] [in a new window]   FIG. 5.   Southern analysis of SNM1 gene copy number in various mutants. (A) Genomic DNA was isolated from yeast snm1 mutant strains grown to 2 × 107 cells/ml at 30°C in SCD (see Materials and Methods). Ten micrograms of genomic DNA per lane was digested with EcoRI and SpeI overnight, separated on a 1% agarose gel, transferred to a nylon membrane, and hybridized with 32P-labeled SNM1 probe and SCR1 probe for standardization. The locations of the SNM1 and SCR1 genes are indicated. (B) Band intensity was measured with a PhosphorImager and standardized against the intensities of the SCR1 and DBY2006 SNM1 bands as single-copy controls to calculate copy number. Copy numbers were plotted against mRNA levels from Fig. 4. SNM1 alleles are listed in Table 1. The WT/CEN and mutant SNM1 genes (S8, p3, p18, p21, and p31) are carried on LEU2 CEN vector YCplac111 (15). The WT/2µ strain is identical to the other strains, except that the wild-type SNM1 gene is maintained on yeast 2µm plasmid YEp351 (18). I, the given allele was integrated into the yeast chromosome at the LEU2 gene in THR200 (see Materials and Methods).

Plasmid missegregation in snm1 mutants. If the plasmid amplification is directly due to selection to amplify the mutant snm1 gene on the plasmid, maintenance of a separate plasmid should be normal. To test plasmid segregation in snm1 mutant strains, we constructed reporter plasmid pTC185. It contains a wild-type ADE2 gene in the CEN URA3 vector pRS316 (51) that is independent of the mutant snm1 gene. All the snm1 mutant strains have an ade2-1 mutation on the chromosome and form red colonies on YPD plates. After having been transformed with plasmid pTC185 containing a wild-type ADE2 gene, they form white colonies. Upon loss of the plasmid, red sectors are formed in white colonies. If there is a segregation problem, more red sectors will be seen.

View larger version (121K): [in this window] [in a new window]   FIG. 6.   Plasmid segregation in snm1 mutants. The yeast snm1 mutant strains were transformed with a plasmid carrying a wild-type ADE2 gene as a reporter for plasmid maintenance. These mutants were grown at 30°C in SCD with uracil to remove selection for the plasmid for 18 h and were diluted in sterile water, and 500 cells of each yeast strain were plated on 150-mm petri dishes. Pictures were taken after 3 days of growth at 30°C, when the color was fully developed. Strain p14 is shown at twice the magnification, since it grows slower than the other strains at 30°C. SNM1 alleles are listed in Table 1. All SNM1 alleles are carried on LEU2 CEN vector pRS315 (51) in strain THR200. WT, wild type.

View larger version (154K): [in this window] [in a new window]   FIG. 7.   Plasmid missegregation is independent of high-copy-number plasmid maintenance or a need to amplify the SNM1 gene. Yeast strains with snm1 mutations integrated into the LEU2 locus on chromosome III were transformed with a plasmid carrying a wild-type ADE2 gene as a reporter for plasmid maintenance. These mutants were grown at 30°C in SCD with uracil to remove selection for the plasmid for 18 h and were diluted in sterile water, and 500 cells of each yeast strain were plated on 150-mm petri dishes. Pictures were taken after 3 days of growth at 30°C, when the color was fully developed. SNM1 alleles are listed in Table 1.

View larger version (88K): [in this window] [in a new window]   FIG. 8.   Immunolocalization of mitotic spindles. Yeast strains were grown to 2 × 106 cells/ml at 30°C in YPD and either maintained at 30°C for an additional 3 h or shifted to 37°C for 3 h. Cells were fixed, and the mitotic spindle and DNA were localized as described in Materials and Methods. 4',6-Diamidino-2-phenylindole (DAPI) staining of DNA is shown in blue, and Cy3 staining of tubulin is shown in red. The wild-type gene and the p18 mutation are carried on LEU2 CEN vector pRS315 (51) in strain THR200.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Generation of snm1 mutations. Random PCR mutagenesis of the SNM1 gene yielded 18 mutants with various temperature-conditional growth phenotypes. DNA sequence analysis of the mutants revealed mutations located throughout the gene and ranging from missense mutations to frameshift and nonsense mutations. Because the SNM1 gene is essential, such serious frameshift and nonsense mutations were not expected. The phenotypes of most of the frameshift and nonsense mutants were very similar. They all displayed cold-sensitive growth (except for p18 and p14, which are also temperature sensitive), a decrease in 5.8S rRNA production, a decrease in Snm1p levels, and a decrease in Snm1p associated with MRP RNA (except for p18). There have been no other reports of cold-sensitive translational suppression, so the cold sensitivity may be the result of a reduction in binding of the MRP RNA by Snm1p at colder temperatures.

Function of Snm1p in RNase MRP. We show for the first time that mutations in the snm1 gene can adversely effect RNase MRP activity, causing a loss of 5.8S rRNA processing. Since it has been previously established that Snm1p interacts with the MRP RNA and that it is a component of the RNase MRP enzyme complex, one role for Snm1p may be to stabilize the MRP RNA component. This possibility was explored by analyzing whole-cell RNA isolated from the snm1 mutants for the presence of the MRP RNA. Northern analyses of the RNA isolated from the snm1 mutants showed that the stability of the MRP RNA was unaffected in all of the mutants. The stability of the MRP RNA is crucial for maintaining proper RNase MRP activity and yeast cell viability. Depletion of other known protein components of the yeast RNase MRP enzyme complex, Pop1p, Pop4p, Pop5p, Pop6p, Pop7/Rpp2p, Pop8p, and Rpp1p, results in a loss of the stability of the RNA component (3, 8, 53). The only other known protein component that has no role in MRP RNA stability is the Pop3 protein, which is believed to be loosely associated with RNase MRP and tightly associated with RNase P (10). Since Snm1p is the only unique protein component of RNase MRP, we speculate that the function of Snm1p may involve substrate recognition and/or binding. The eight other protein components shared by RNase MRP and RNase P probably have functions in stabilizing the component RNAs, assembling, transporting, and localizing the ribonucleoprotein complexes, and potentially catalyzing the RNA cleavage reactions. Either Rpr2p, which is a unique component of RNase P, or Pop3p may play the counterpart role of Snm1p in RNase P.

The essential function of RNase MRP. RNase MRP was first found as an enzyme involved in mitochondria DNA replication (4) and subsequently has been found to have a direct role in rRNA processing (44). However, yeast cells without mitochondrial DNA are viable, and others have questioned whether the role of RNase MRP in rRNA processing is essential (1, 17). The RNA component and all the protein components of yeast RNase MRP identified so far are essential, indicating that the RNase MRP complex has a function which is essential for normal cell growth.