C-terminal truncation of the Sup35 protein increases the frequency of de novo generation of a prion-based [PSI+] determinant in Saccharomyces cerevisiae

The yeast non-Mendelian [PSI ⁺ ] determinant is presumed to be the manifestation of the aggregated prion-like state of the Sup35 protein. Plasmid-mediated amplification of the SUP35 gene greatly increases the frequency of Sup35p transition to this prion-like state. Here we show that the 3′-deletions of plasmid SUP35, leading to the C-terminal truncation of Sup35p, further increase the frequency of [PSI ⁺ ] induction despite a marked decrease in Sup35p expression levels. The data suggest that the presence of Sup35p N-terminal proteolytic fragments can cause [PSI ⁺ ] appearance in wild-type yeast cells. Received: 2 February / 24 June 1998     

[NSI +]: a novel non-Mendelian nonsense suppressor determinant in Saccharomyces cerevisiae

Non-Mendelian determinants that control heritable traits in yeast are subdivided into two major groups—one that includes DNA- or RNA-based elements and another that comprises protein-based factors that are analogous to mammalian prion. All yeast non-Mendelian determinants show dominant inheritance, and some of them demonstrate cytoplasmic infectivity. Only prions, however, harbor-specific features, such as high frequency of induction following overproduction of prion-encoding protein, loss of the protein’s normal function, and reversible curability. Here, we describe a novel nonchromosomal determinant that, in addition to [PSI ⁺] and [ISP ⁺], is involved in epigenetic control of nonsense suppression. This determinant, which we have designated [NSI ⁺], causes nonsense suppression in the strains bearing the N-terminal-deleted or -modified SUP35 gene, but has no manifestation in the strains with the intact copy of SUP35. [NSI ⁺] shows dominant non-Mendelian inheritance, reversible curability and may be transmitted by cytoduction, albeit with low frequency. Similar to yeast prions, this determinant can be cured by deletion or mutational inactivation of Hsp104. We have shown that [NSI ⁺] does not correspond to the already identified yeast prions. Based on the data obtained, we hypothesize that [NSI ⁺] is a novel prion factor involved in epigenetic control of nonsense suppression.     

Modifiers of ochre suppressors in Saccharomyces cerevisiae that exhibit ochre suppressor-dependent amber suppression

Summary Mutants of Saccharomyces cerevisiae were selected that would interact with ochre (UAA) suppressors so as to allow ochre -suppressor dependant amber (UAG) suppression, but which do not exhibit opal (UGA) suppression. Strains mutant at four distinct loci were isolated, and two of these are recessive mutations while the other two behave as dominants or semidominants. MOS3 has some suppressor activity in the absence of a resident SUP4-o gene and shares other characteristics with previously described omnipotent suppressors. MOS4, mos1 and mos2, on the other hand, exhibit no suppressor activity in the absence of a resident SUP4-o gene but do exhibit suppression of UAG alleles when there is a resident SUP4-o gene. These latter modifier strains do not interact with a SUP4-o gene to suppress UGA alleles. By genetic and physiological criteria the MOS4, mosl, and most mutations appear to be different than previously described allosuppressors or modifiers of suppression.     

NSI+] determinant has a pleiotropic phenotypic manifestation that is modulated by SUP35, SUP45, and VTS1 genes

We recently discovered the novel non-chromosomal determinant in Saccharomyces cerevisiae [NSI ⁺] (nonsense suppression inducer), which causes omnipotent nonsense suppression in strains where the Sup35 N-terminal domain is deleted. [NSI ⁺] possesses yeast prion features and does not correspond to previously identified yeast prion determinants. Here, we show that [NSI ⁺ ] enhances nonsense codon read-through and inhibits vegetative growth in S. cerevisiae. Using a large-scale overexpression screen to identify genes that impact the phenotypic effects of [NSI ⁺], we found that the SUP35 and SUP45 genes encoding the translation termination factors eRF3 and eRF1, respectively, modulate nonsense suppression in [NSI ⁺] strains. The VTS1 gene encodes an NQ-enriched RNA-binding protein that enhances nonsense suppression in [NSI ⁺] and [nsi ⁻] strains. We demonstrate that VTS1 overexpression, like [NSI ⁺] induction, causes translational read-through and growth defects in S. cerevisiae.     

Sensitivity of sup35 and sup45 suppressor mutants in Saccharomyces cerevisiae to the anti-microtubule drug benomyl

 SUP35 and SUP45 genes determine the accuracy of translation at the stage of termination. We present indirect evidence indicating that these genes may also control some cellular process mediated by microtubules. A majority of sup35 and sup45 suppressor mutations confer supersensitivity to benomyl, the drug which de-polymerizes microtubules. In addition, data correlating phenotypic manifestations of sup45 suppressor mutations, involving sensitivity to benomyl, respiratory deficiency and a suppressor effect, are also presented. Received: 16 October 1995/7 December 1995     

Ssb1 chaperone is a [PSI +] prion-curing factor

Yeast SUP7 or SUP11 nonsense suppressors have no phenotypic expression in strains deficient in the isopentenylation of A37 in tRNA. Here we show that such strains spontaneously produce cells with a nonsense suppressor phenotype which is related to the cytoplasmically inherited determinant and manifests all the key features of the [PSI ⁺] prion. A screen of a multicopy yeast genomic library for genes that inactivate the [PSI ⁺]-related suppressor phenotype resulted in the isolation of the SSB1 gene. Moreover, we demonstrate that multicopy plasmid encoding the Ssb1 chaperone cures cells of the [PSI ⁺] prion. Received: 9 August 2000 / Accepted: 20 October 2000     

Recessive mutations in SUP35 and SUP45 genes coding for translation release factors affect chromosome stability in Saccharomyces cerevisiae

Chromosome stability in suppressor mutants for SUP35 and SUP45 genes coding for translation release factors was studied. We obtained spontaneous and UV-induced sup35 or sup45 mutants in a haploid strain disomic for chromosome III and tested the stability of an extra copy of this chromosome. The majority of the mutants showed increased chromosome instability. This phenotype was correlated with an increased sensitivity to the microtubule-poisoning drug benomyl which affects chromosome segregation at anaphase. Our data suggest that termination-translation factors eRF3 and eRF1 control chromosome transmission at mitotic anaphase in Saccharomyces cerevisiae. Received: 26 July 1999 / 27 January 2000     

The PNM2 mutation in the prion protein domain of SUP35 has distinct effects on different variants of the [PSI+] prion in yeast

We have previously described different variants of the yeast prion [PSI ⁺ ] that can be obtained and maintained in the same genetic background. These [PSI ⁺ ] variants, which differ in the efficiency of nonsense suppression, mitotic stability and the efficiency of curing by GuHCl, may correspond to different [PSI ⁺ ] prion conformations of Sup35p or to different types of prion aggregates. Here we investigate the effects of overexpressing a mutant allele of SUP35 and find different effects on weak and strong [PSI ⁺ ] variants: the suppressor phenotype of weak [PSI ⁺ ] factors is increased, whereas the suppressor effect of strong [PSI ⁺ ] factors is reduced. The SUP35 mutation used was originally described as a “Psi no more” mutation (PNM2) because it caused loss of [PSI ⁺ ]. However, none of the [PSI ⁺ ] variants in the strains used in our study were cured by PNM2. Indeed, when overexpressed, PNM2 induced the de novo appearance of both weak and strong [PSI ⁺ ] variants with approximately the same efficiency as the overexpressed wild-type SUP35 allele. Our data suggest that the change in the region of oligopeptide repeats in the Sup35p N-terminus due to the PNM2 mutation modifies, but does not impair, the function of the prion domain of Sup35p. Received: 12 October / 15 December 1998     

Inactivation of the RAD1 excision-repair gene does not affect correction of mismatches on heteroduplex plasmid DNA in yeast

Summary The efficiency and direction of mismatch correction in the Saccharomyces cerevisiae SUP4-o gene were not altered by an excision-repair defect (rad1). Although excision-repair functions remove methylated adenine from yeast, adenine methylation at a GATC sequence in SUP4-o did not direct the correction of mismatches via excision repair.     

Interactions between chromosomal omnipotent suppressors and extrachromosomal effectors in Saccharomyces cerevisiae

Summary Chromosomal omnipotent suppressor mutations recovered in ⁺ strains of Saccharomyces cerevisiae were brought into ^– cytoplasm. SUP46, SUP138 and SUP139 acted as dominant omnipotent suppressors in the ^– cytoplasm though their suppressor activity was substantially reduced. SUP46 and SUP138 conferred recessive thermosensitivity and antibiotic sensitivity in ^– cytoplasm as in ⁺ cytoplasm. On the other hand, sup111 through sup115, which acted as recessive omnipotent suppressors in the ⁺ cytoplasm, manifested no, or very low, suppressor activity in the ^– cytoplasm. They, however, still enhanced the efficiency of the SUP29 tRNA suppressor in ^– cytoplasm. A multicopy plasmid carrying the wild-type SUP35 gene enhanced the efficiency of sup111 in ^– cytoplasm.     

Two new loci that give rise to dominant omnipotent suppressors in Saccharomyces cerevisiae

Summary Ten dominant omnipotent suppressors of Saccharomyces cerevisiae, which were previously shown to be different from SUP46, have been examined. Nine are mapped in a region between lys5 and cyh2 on the left arm of chromosome VII. These suppressors, like SUP46, manifest sensitivity to increased temperature and the antibiotics paromomycin and hygromycin B. In addition, they have an identical action spectrum. These results strongly suggest that they are allelic to each other and they are designated SUP138. The tenth is mapped to a position between his1 and arg6 on the right arm of chromosome V. This suppressor, named SUP139, does not manifest temperature sensitivity nor antibiotic sensitivity. SUP139 and SUP138, which are clearly distinguished by means of action spectrum, act on much fewer nonsense mutations than SUP46. It is now clear that dominant omnipotent suppressors arising at a single locus are homogeneous and that their efficiency is locus-dependent. The order of efficiency is SUP46>SUP138>SUP139.     

Failure to detect an antimutator phenotype following disruption of the Saccharomyces cerevisiae DDR48 gene

The antimutator phenotype, reportedly conferred by disruption of the Saccharomyces cerevisiae DDR48 gene, was suggested to affect only a specific spontaneous mutational pathway. We attempted to identify the types of mutation that are DDR48-dependent by determining the specificity of the ddr48 antimutator. However, disruption of DDR48 did not decrease the rates of spontaneous forward mutation in a plasmid-borne copy of the yeast SUP4-o gene, the reversion or suppression of the lys2–1 allele, or forward mutation at the CAN1 locus. Interestingly, the latter gene had been reported previously to be subject to the antimutator effect. DNA sequence analysis of spontaneous SUP4-o mutations arising in DDR48 and ddr48 backgrounds provided no evidence for a reduction in the rates of individual mutational classes. Thus, we were unable to verify that disruption of DDR48 causes an antimutator phenotype.     

Antisuppression of class I suppressors in an isopentenylated-transfer RNA deficient mutant of Saccharomyces cerevisiae

Summary The effect of a previously isolated antisuppressor mutation from bakers' yeast, that reduced the efficiency of the tyrosine-inserting ochre suppressor, SUP7-o, on other tyrosine-inserting ochre suppressors has been determined. As expected, the antisuppressor mutation, mod5-1, restricted the capacity of all eight tyrosine-inserting ochre suppressors to suppress nonsense mutations. Based on the suppression of five ochre alleles in the presence of mod5, the eight class I suppressors can be grouped into three subclasses. The most efficient subclass had only one member, SUP4-o. Members of the second group included SUP2-o, SUP3-o, SUP7-o, and SUP8-o. The third and least efficient subclass included SUP5-o, SUP6-o, and SUP1 1-o. These differences in efficiencies are a function of the relative expression of the eight genes encoding tRNA^TYR.     

Polypeptide chain termination in Saccharomyces cerevisiae

Summary The study of translational termination in yeast has been approached largely through the identification of a range of mutations which either increase or decrease the efficiency of stop-codon recognition. Subsequent cloning of the genes encoding these factors has identified a number of proteins important for maintaining the fidelity of termination, including at least three ribosomal proteins (S5, S13, S28). Other non-ribosomal proteins have been identified by mutations which produce gross termination-accuracy defects, namely the SUP35 and SUP45 gene products which have closely-related higher eukaryote homologues (GST1-h and SUP45-h respectively) and which can complement the corresponding defective yeast proteins, implying that the yeast ribosome may be a good model for the termination apparatus existing in higher translation systems.While the yeast mitochondrial release factor has been cloned (Pel et al. 1992), the corresponding cytosolic RF has not yet been identified. It seems likely, however, that the identification of the gene encoding eRF could be achieved using a multicopy antisuppressor screen such as that employed to clone the E. coli prfA gene (Weiss et al. 1984). Identification of the yeast eRF and an investigation of its interaction with other components of the yeast translational machinery will no doubt further the definition of the translational termination process.While a large number of mutations have been isolated in which the efficiency of termination-codon recognition is impaired, it seems probable that a proportion of mutations within this class will comprise those where the accuracy of A site codon-anticodon interaction is compromised: such defects would also have an effect on termination-codon suppression, allowing mis- or non-cognate tRNAs to bind stop-condons, causing nonsense suppression. The remainder of mutatoons affecting termination fidelity should represent mutations in genes coding for components of the termination apparatus, including the eRF: these mutations reduce the efficiency of termination, allowing nonsense suppression by low-efficiency natural suppressor tRNAs. Elucidation of the mechanism of termination in yeast will require discrimination between these two classes of mutations, thus allowing definition of termination-specific gene products.     

A mutant allele of the SUP45 (SAL4) gene of Saccharomyces cerevisiae shows temperature-dependent allosuppressor and omnipotent suppressor phenotypes

Using a plasmid-based termination-read-through assay, the sal4-2 conditional-lethal (temperature-sensitive) allele of the SUP45 (SAL4) gene was shown to enhance the efficiency of the weak ochre suppressor tRNA SUQ5 some 10-fold at 30°C. Additionally, this allele increased the suppressor efficiency of SRM2-2, a weak tRNA^Gln ochre suppressor, indicating that the allosuppressor phenotype is not SUQ5-specific. A sup ⁺ sal4-2 strain also showed a temperature-dependent omnipotent suppressor phenotype, enhancing readthrough of all three termination codons. Combining the sal4-2 allele with an efficient tRNA nonsense suppressor (SUP4) increased the temperature-sensitivity of that strain, indicating that enhanced nonsense suppressor levels contribute to the conditional-lethality conferred by the sal4-2 allele. However, UGA suppression levels in a sup ⁺ sal4-2 strain following a shift to the non-permissive temperature reached a maximum significantly below that exhibited by a non-temperature sensitive SUP4 suppressor strain. Enhanced nonsense suppression may not therefore be the primary cause of the conditional-lethality of this allele. These data indicate a role for Sup45p in translation termination, and possibly in an additional, as yet unidentified, cellular process.     

Mouse GSPT2, but not GSPT1, can substitute for yeast eRF3 in vivo

BACKGROUND: The termination of protein synthesis in eukaryotes involves at least two polypeptide release factors (eRFs), eRF1 and eRF3. In mammals two genes encoding eRF3 structural homologues were identified and named GSPT1 and GSPT2. RESULTS: In the present study, we demonstrate that mouse mGSPT2 but not mGSPT1 could functionally substitute the essential yeast gene SUP35. However, we show that the complementation property of mGSPT1 protein is modified when NH2-tagged by GST. Since mGSPT1 and mGSPT2 differ mainly in their N-terminal regions, we developed a series of N-terminal deleted constructs and tested them for complementation in yeast. We found that at least amino acids spanning 84-120 of mGSPT1 prevent the complementation of sup35 mutation. The fact that chimeras between mGSPT1, mGSPT2 and yeast Sup35 complement the disruption of the SUP35 gene indicates that the N-terminal region of mGSPT1 is not sufficient by itself to prevent complementation. Complementation of the mutant with a double disruption of SUP35 and SUP45 genes is obtained when mGSPT2 and human eRF1 are co-expressed but not by co-expression of mGSPT1 and human eRF1. CONCLUSIONS: Our results strongly suggest that the two proteins (mGSPT1 and mGSPT2) are different. We hypothesize that the full length mGSPT1 does not have the properties expected for eRF3.     

Centromeric DNA of Kluyveromyces lactis

Summary A direct selection method was used to isolate centromeres from a genomic library of the yeast Kluyveromyces lactis. The method is based on the lethality at high copy number of the ochre-suppressing tRNA gene SUP11. Five different chromosomal fragments were found that confer mitotic stability to plasmids containing a replication origin of K. lactis (KARS). In addition, KARS plasmids containing these fragments have a copy number or approximately one, and each of the five fragments hybridizes to a different chromosome of K. lactis. From these results we conclude that five of the six centromeres of K. lactis have been isolated. These centromeres do not function in S. cerevisiae.     

Cloning and mapping of the CYS4 gene of Saccharomyces cerevisiae

Summary A DNA fragment containing the CYS4 gene of Saccharomyces cerevisiae was isolated from a genomic library. The cloned fragment hybridized to the transverse-alternating-field-electrophoresis band corresponding to chromosomes VII and XV. According to the 2 m DNA chromosome-loss procedure, the cys2 and cys4 mutations, which are linked together and co-operatively confer cysteine dependence, were assigned to chromosome VII. By further mapping involving tetrad analysis, the cys2-cys4 pair was localized between SUP77 (SUP166) and ade3 on the right arm of chromosome VII.