INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Saccharomyces cerevisiae exhibits multiple responses to nutrient limitation depending on cell type. Haploid cells exit the mitotic cell cycle at G₁, entering a quiescent phase reminiscent of the higher eukaryotic G₀ (48). Diploid cells undergo a dimorphic shift, developing pseudohyphae (13), or sporulate (22). Sporulation depends on the transcriptional master regulator Ime1p, and in some strains, its overproduction results in gratuitous sporulation (18). Although not understood in detail, the IME1 promoter is complex, containing multiple elements that respond to a variety of signals (33). One of these elements, IRE, is nutritionally responsive and homologous to the stress response element (STRE), which supports transcription that is influenced by cyclic AMP (cAMP)-dependent protein kinase A (PKA) activity (20, 35).

Yeast PKA is a heterodimer consisting of a regulatory subunit, Bcy1p, complexed to one of three catalytic subunits encoded by TPK1, TPK2, and TPK3 (7, 43). The binding of cAMP to Bcy1p promotes dissociation of the Bcy1p-Tpkp dimer and concomitantly activates kinase activity. Elevated PKA levels correlate with (i) inability to accumulate trehalose or glycogen (7) and (ii) repression of STRE-dependent gene expression (4). Overproduction of Rgs2p (a negative regulator of PKA) increases glycogen accumulation at stationary phase (1.75 versus 2.1% [wet weight]), whereas glycogen accumulation decreases in a rgs2 mutant (2.1 versus 1.3% [wet weight]) (45).

Rim15p kinase is another PKA regulator as cells enter stationary phase, (32). Although rim15 strains do not have an observable growth defect, the growth phenotype of a conditional adenyl cyclase mutation (cyr1) is suppressed in these mutants (32). Conversely, overproduction of Rim15p suppresses the phenotypes of mutations that cause elevated PKA activity, e.g., bcy1. In log-phase rim15/rim15 cells, glycogen levels are similar to those in wild-type cells (0.93 versus 1.63 mg of glycogen/g of protein) (32). At stationary phase, however, a greater fold difference is observed (11.36 versus 38.53 mg of glycogen/g of protein, i.e., 3.4- versus 1.7-fold) (32).

We previously identified a nitrogen starvation-induced, dual-specific protein phosphatase, Yvh1p, required for efficient sporulation (14, 28). YVH1 disruption reduces, but does not abolish, IME1 expression; an even greater effect is observed on IME2 (28). In addition, YVH1 disruption strains exhibit diminished ability to produce the fluorescence associated with dityrosine accumulation occurring in mature ascospore walls (1).

Here, we show that yvh1 disruption mutants cannot effectively accumulate glycogen. YVH1 alleles complementing this defect complement the growth and spore maturation defects as well. Complementing alleles, surprisingly, include catalytically inactive yvh1 variants and one encoding only the C-terminal 159 amino acids (aa), suggesting that Yvh1p's phosphatase activity, per se, is not required for complementation. Overproduction of phosphodiesterases (responsible for cAMP degradation) suppressed the glycogen accumulation and sporulation defects of yvh1 disruption mutants. Finally, disruption of YVH1 decreases stress-mediated induction of a STRE-driven reporter gene.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Strains and plasmids. Strains and plasmids used in this work are listed in Tables 1 and 2, respectively.

Plasmid construction. pAB7 contains the parent allele of YVH1 (1). To create a catalytically inactive yvh1 variant, the XbaI and XhoI fragment of pAB7 (in pSelect) was subjected to site-directed mutagenesis producing a C117S change, which abolishes phosphatase activity (9, 49). The entire fragment was sequenced to ensure that no additional mutations were present and recloned back into pAB7 to create plasmid pAB18.

C-terminal deletions of YVH1. Progressive C-terminal truncations were made with PCR oligonucleotides removing 50 C-terminal aa at a time; both 5' and 3' ends contained BamHI sites. The PCR products were cloned into an intermediate vector, litmus 28, and the resulting plasmids were verified by DNA sequencing; clones containing the desired mutations were digested with BamHI and cloned into Yep24. Full-length YVH1 pAB100 containing the promoter and 3' untranslated regions was constructed by cloning the 2.2-kb XbaI fragment from pAB7 into Yep24 digested with NheI.

Strain construction. To construct strains GYC134 (pho85) and GYC135 (pho85 yvh1::HIS3), the exact allele described by Timblin et al. (42) was constructed and used to transform strains GYC121 and GYC122, or GYC123 and GYC124, to Trp prototrophy. All constructs were confirmed by Southern blot analysis. Strain GYC134 was constructed by mating strains GYC130 and GYC131. Both GYC134 and GYC135 were demonstrated to sporulate when transformed with a plasmid containing PHO85 (data not shown).

pho85 yvh1

Construction of the STRE-responsive reporter. Two oligonucleotides, 31 and 32 (5'-GGCCGCTTCTTTTCCCCTGTTTCCATTTTTGTCTTTTCTCACCCCTTATGGGCCG-3' and 5'-TCGACGGTCCCCATAAGGGGTGAGAAAAGACAAAAATGGAAACAGGGGAAAAGAAGC-3') (19), were used. They differed from those published earlier by the extensions used to clone the element. The oligonucleotides were annealed, kinased by T4 polynucleotide kinase, purified, and ligated into pNG15 digested with SalI and EagI and treated with shrimp alkaline phosphatase (Boehringer).

Glycogen and enzyme assays. Glycogen and enzyme assays were performed as previously described (38) except that in some cases glycogen levels were normalized to soluble protein levels, which were assayed using the Bio-Rad assay. Glucose was quantitated using hexokinase or affinity glucose reagent diagnostic kits (Sigma); both kits yielded the same results.

Sporulation protocols. Two growth protocols were used to determine whether overexpression of PDE1 and PDE2 suppresses the dityrosine production (fluorescence) defect of yvh1 strains. In the standard protocol (47), transformants were grown on minimal glucose medium, transferred to YEPD for 15 h of presporulation, and then transferred onto sporulation plates for 144 h (1). Under these conditions, GAL1,10-GST-PDE1 and GAL1,10-GST-PDE2 expression is expected to increase only as cells overcome glucose repression on the sporulation plates, and then to decrease as cells sporulate. In the second protocol, transformants are grown on minimal raffinose medium, transferred to YEPGal for 15 h, and then transferred onto sporulation plates for 144 h. Here, GAL1,10-GST-PDE1 and GAL1,10-GST-PDE2 expression is expected to be moderate at the outset, increase to quite high levels on YEPGal presporulation medium, and decrease as cells sporulate. Note that although conditions in the second protocol are optimal for increased PDE1,2 expression, sporulation is less than optimal; wild-type cells sporulate less efficiently.

GST purification, silver staining, and Western analyses. GST fusion proteins were produced as described previously (23) with the following modifications. Cells were grown to an A₆₀₀ of 1.0 in raffinose Ura medium, induced for 3 h with 4% galactose, and harvested. Silver staining was done with Silver Stain (Bio-Rad). Phosphatase (pNPP) assays were performed as previously reported (25) with an incubation time of 1 h. Western analyses were performed as previously described (37) using horseradish peroxidase (HRP)-conjugated anti-GST antibody (Pharmacia) following 3 h of induction with galactose.

Epifluorescence microscopy. Microscopy was performed as previously described (37) by using a Zeiss Axiophot microscope equipped with a fluorescein filter and cells induced for 3 h with 4% galactose.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Disruption of YVH1 leads to defects in glycogen metabolism. Similar phenotypes of mutations in multiple genes affecting glycogen accumulation (32, 42) and a yvh1 disruption prompted determination of whether yvh1 strains accumulate glycogen. Glycogen did not increase in the yvh1 mutant strain as it did when the wild type reached stationary phase (Fig. 1A and B). The observation that a yvh1 pho85 double mutant behaved similarly to a pho85 single mutant argued that pho85 mutations were epistatic to those in yvh1 (Fig. 1B through D). These results suggested that YVH1 is required for stationary-phase cells to accumulate glycogen. In fact, yvh1 disruption exhibits a phenotype that more closely resembles that of snf1 and bcy1 mutants (15), i.e., the effects are observed only as cells approach stationary phase.

View larger version (39K): [in this window] [in a new window]   FIG. 1.   Glycogen accumulation in stationary-phase cells. Wild-type (GYC86) (A), yvh1 (HPY120) (B), pho85 (GYC134) (C), and yvh1 pho85 (GYC135) (D) cultures were grown in YEPD medium, and aliquots were withdrawn at the A600 indicated on the ordinate and quantified for glycogen as described in Materials and Methods. Each value represents the average of two determinations per aliquot. (E) Acid phosphatase assay. Wild-type (GYC86), yvh1 (HPY120), pho85 (GYC134), and yvh1 pho85 (GYC134) cells were streaked on high Pi X-P synthetic complete plates and incubated at 30°C. The appearance of blue color indicates acid phosphatase expression.

Only the C-terminal portion of Yvh1p is needed to perform its role in glycogen accumulation, growth, and spore maturation. To identify the portions of Yvh1p required for glycogen accumulation, we constructed progressive C-terminal yvh1 deletion plasmids and used them to transform strain HPY120. Since the glycogen accumulation defect of yvh1 mutants is most pronounced late in growth, these and subsequent assays were preformed 72 h after inoculation, and results were standardized to soluble protein levels. Nearly three times more glycogen accumulated with full-length YVH1 (containing or devoid of its 3' flanking sequences) (pAB100 and pAB101) than with vector Yep24 (Fig. 2A). Deletion of DNA encoding the C-terminal 50 aa of Yvh1p (pAB102) resulted in loss of glycogen accumulation above that of the negative control (YEp24); this also occurred with the remaining five deletion plasmids (Fig. 2A). Similar results were observed when growth was assayed with glucose (Fig. 2C and D) or raffinose (data not shown) as a sole carbon source. Finally, the same deletion plasmids also failed to complement the yvh1 spore maturation defect (fluorescence) phenotype in diploid cells (Fig. 2B). These data argued for the necessity of the C-terminal portion of Yvh1p to perform these three functions or, alternatively, to maintain the protein's stability. These alternatives were indistinguishable because the levels of Yvh1p are below the level of detection even in the wild type.

View larger version (51K): [in this window] [in a new window]   FIG. 2.   Effects of progressive C-terminal yvh1 deletions on their ability to complement the yvh1 null mutation in strain HPY120. (A) Normalized glycogen accumulation 72 h after inoculation of strain HPY120, transformed with plasmids indicated on the abscissa, into synthetic complete (SC) Ura medium. (B) Spore maturation (dityrosine fluorescence). (C and D) Growth on SC Ura plates. All transformants were streaked onto the respective plate and incubated at 30°C.

View larger version (41K): [in this window] [in a new window]   FIG. 3.   Phosphatase activity of wild-type and catalytically inactive alleles of YVH1. (A) pNPP activities of purified wild-type (pAB36) and catalytically inactive (pAB37) Yvh1p-GST fusion proteins. (B) Silver-stained polyacrylamide gel electrophoresis (PAGE) gels of the proteins used in panel A. (C) Western analyses of various YVH1 alleles fused to GST.

View larger version (61K): [in this window] [in a new window]   FIG. 4.   Abilities of various GST tagged alleles of YVH1 to suppress the slow-growth defect of yvh1 strain HPY120 on glucose (A) and raffinose (B) plates. (C) Growth phenotypes of various strains used in this work, including the integrated catalytically inactive allele of YVH1, on YEPD plates. All strains and transformants were grown at 30°C.

View larger version (51K): [in this window] [in a new window]   FIG. 5.   Abilities of various GST tagged alleles of YVH1 to suppress the glycogen accumulation and spore maturation defects of yvh1 strain HPY120. (A and B) Normalized glycogen accumulation of transformants 72 h postinoculation into synthetic complete (SC) Ura medium with glucose (A) or raffinose (B) as the carbon source. (C) Spore maturation (dityrosine fluorescence) assays using the standard spore maturation protocol described in Materials and Methods.

View larger version (36K): [in this window] [in a new window]   FIG. 6.   Protein phosphatase activity is not required for glycogen accumulation or spore maturation. (A) Normalized glycogen content for wild-type (GYC86), yvh1 (HPY120), mck1 (YSB40), pho85 (GYC134), and HAyvh1C117S (GYC136) strains 72 h postinoculation. (B) Spore maturation phenotypes of wild-type (GYC86), yvh1 (HPY120), HAYVH1 (YSB32), HAyvh1C117S (GYC136), and mck1 (YSB40) strains under standard conditions.

yvh1 mutants, lacking the phosphatase domain, complement YVH1 null mutations. The above results suggested that Yvh1p phosphatase activity itself was not required for growth, glycogen accumulation, or spore maturation (fluorescence). Therefore, we identified the portions of Yvh1p that were required by determining whether fragments encoding C-terminal portions of Yvh1p could complement a yvh1 mutation. The C-terminal 159 (pAB107) or 74 (pAB108) aa were fused to GAL1,10-GST in pEGKG. pAB107, lacking sequences encoding the Yvh1p phosphatase domain, complemented the growth (Fig. 4A and B) and fluorescence (Fig. 5C) phenotypes, whereas plasmid pAB108 did not. pAB107 complements the glycogen accumulation phenotype, but only when the plasmid-borne YVH1 fragment is expressed at moderate levels, i.e., when the GAL1,10 reporter is derepressed by growing transformants with raffinose as a sole carbon source (Fig. 5A and B).

Lowering cAMP levels partially suppresses the glycogen accumulation and spore maturation defects of yvh1. One form of cellular regulation previously implicated in growth, glycogen accumulation, and ability to sporulate is cAMP and its stimulation of PKA. Cells with constitutively active PKA activity (bcy1 mutants) fail to accumulate glycogen, are sporulation deficient, and possess various growth defects (7). Conversely, cells with low PKA activity (ras2 mutants) exhibit growth defects, hyperaccumulate glycogen, and sporulate precociously in the presence of nutrients (10, 24). In other words, the phenotypes associated with elevated levels of cAMP-PKA are qualitatively similar to those caused by disruption of YVH1.

View larger version (50K): [in this window] [in a new window]   FIG. 7.   Conditional expression of genes responsible for cAMP turnover, PDE1 (pAB112) and PDE2 (pAB113), can partially suppress the glycogen accumulation and spore maturation defects of yvh1 in an expression-dependent manner. (A and B) Normalized glycogen accumulation 72 h postinoculation in glucose (A) or raffinose (B) medium. (C and D) Spore maturation phenotypes of the same transformants under the standard dityrosine protocol (C) and the unconventional protocol (D) as described in Materials and Methods.

View larger version (48K): [in this window] [in a new window]   FIG. 8.   Conditional expression of PDE1 and PDE2 does not suppress the slow-growth phenotype of yvh1 mutants. Strain HPY120 was transformed with GST fused to one of the following genes: none (pEGKG; vector), YVH1 (pAB36), yvh1C117S (pAB37), GSY2 (pAB111), PDE1 (pAB112), and PDE2 (pAB113). Transformants were streaked onto minimal glucose (A) or raffinose (B) medium and grown at 30°C; these conditions yield low and moderate levels of fusion gene expression. Full induction of the GAL1,10 promoter, driving the above gene expression, is highly detrimental to growth of both the yvh1 (HPY120) and wild-type (GYC86) strains transformed with pAB112 and pAB113.

Disruption of YVH1 affects STRE-mediated gene expression. To further test whether yvh1 mutations elicit the same effects as elevated cAMP-PKA levels, we determined whether cAMP-PKA-dependent heterologous reporter gene expression was affected by disruption of YVH1. We chose the DDR2 promoter fragment, previously reported by Kobayashi and McEntee to support approximately 15 and 120 U of -galactosidase production at 23 and 37°C, respectively, an induction ratio of 8.1 (19). This DNA fragment, which contains two STREs, was cloned into the heterologous expression vector pNG15; the resulting plasmid, pAB60, was used to transform wild-type (GCY86) and yvh1 (HPY120) strains. Transformants were grown without stress (22°C) and subjected to a transient heat shock (40°C) for 1 h; -galactosidase activities were then measured. Heat shock has little if any effect on basal lacZ expression supported by the parent vector pNG15 in either the wild-type or the mutant strain (Fig. 9). Under non-stress conditions (22°C), -galactosidase production is only marginally higher (less than twofold) in the yvh1 mutant than in the wild type (Fig. 9). Heat shock increased lacZ expression in both wild-type (25- to 26-fold) and yvh1 (6- to 7-fold) strains, but the yvh1 mutation lowered cAMP-PKA-dependent stress-responsive transcription 4- to 5-fold.

View larger version (20K): [in this window] [in a new window]   FIG. 9.   Deletion of YVH1 results in decreased induction of transcription mediated by a multi-STRE. Wild-type (GYC86) and yvh1 (HPY120) strains were transformed with the lacZ reporter plasmid pNG15 or the same plasmid containing the DDR2 promoter element (pAB60). Transformants were grown at 22°C (non-stress condition) and then shifted to 40°C for 1 h (stress condition) before expression was assayed. Values represent the average of four independent transformants assayed in duplicate.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Data presented here expand the targets of the dual-specific phosphatase Yvh1p to include those associated with glycogen accumulation in stationary phase. In this characteristic, Yvh1p joins the Glc7p, Pph21, Pph22, Pgp1p/YNRO32w, and Sit4p phosphatases, which also regulate glycogen accumulation (8, 29-31). Experiments with YVH1 truncation alleles demonstrate that the alleles which suppress slow growth also suppress the spore maturation and glycogen accumulation defects. The congruence of suppression suggests that all three phenotypes are probably manifestations of a common function. Surprisingly, the function is independent of the catalytic activity demonstrated for YVH1, namely, a protein tyrosine phosphatase (14). During preparation of this article, J. E. Dixon's laboratory reported isolation of the human variant of YVH1 and observed that suppression of the vegetative growth defect was not impaired by the same mutations which abolish catalytic activity with either the yeast or the human YVH1 (25). Our results are in agreement with theirs and extend their findings by demonstrating that the C-terminal 159 aa of Yvh1p are sufficient for suppression, an important result, since this allele completely lacks the phosphatase domain. It is this result that prevents us from reaching the same conclusion as Muda et al. regarding the biochemical role of the Yvh1p C-terminal region (25). Although noncatalytic domain positioning of the catalytic domain in proximity to its substrate is one way in which signal insulation is achieved (6), pAB107 lacking the catalytic domain remains competent for suppression. Therefore, we favor a model in which Yvh1p phosphatase activity is not required for the functions that are suppressed by the C-terminal peptide. Although catalytic activity is not required for the functions we assayed, it is entirely possible that Yvh1p phosphatase activity is required for a function that has so far escaped detection. This surprising conclusion is not unique. Schwartzberg et al. demonstrated that kinase-deficient variants of src did not induce osteocarcinomas in rats, whereas deletion of SRC did (36). Although these two cases are certainly exceptions to the rule, they prompt caution in the use of homology as the means of establishing physiological function.

If phosphatase activity is not required for the Yvh1p function, what is the target of the Yvh1p C-terminal peptide? While we have not established how the Yvh1p C-terminal peptide suppresses the yvh1 null mutant phenotypes, the suppressed functions possess striking similarities. The three suppressed phenotypes, growth, sporulation and glycogen accumulation, are affected by cAMP and alteration of cAMP-dependent PKA (7). A ras2 strain precociously sporulates (24), whereas cells with elevated PKA activity, e.g. a bcy1 mutant and constitutively active alleles of RAS2, are completely sporulation defective (21). Additionally, DIT1 and DIT2 expression (responsible for dityrosine production, exploited in our fluorescence assays) is driven by a promoter containing 4 STREs through which PKA functions (PATCHMATCH at http://genome-ww2.stanford.edu). The negative effects of yvh1 mutations on heterologous STRE-dependent DDR2 expression are consistent with the argument that Yvh1p may well interact directly or indirectly with one or more component molecules of the cAMP-PKA signal cascade and thereby produce the same effect as that resulting from decreased cAMP levels. Supporting this interpretation is the observation (Fig. 7A and B) that artificially decreasing cAMP levels by overproduction of phosphodiesterases 1 and 2 partially suppresses yvh1 defects. We favor the suggestion that Yvh1p's physiological function is not, in the current context, to dephosphorylate critical phosphoproteins but to modulate a component(s) of the cAMP-PKA signal transduction cascade via a physical interaction (Fig. 10). Although this suggestion fits well with our data, our information does not permit exclusion of other possible alternatives. Most specifically, our experiments do not distinguish whether Yvh1p affects cAMP-PKA levels or some function downstream of cAMP in the cAMP-dependent PKA cascade.

View larger version (16K): [in this window] [in a new window]   FIG. 10.   Model accounting for physiological functions in which YVH1p participates.

This interpretation fits well with existing physiological models and may serve to resolve several outstanding issues. First is the nature of the kinases regulating glycogen synthase activity. Pho85p, Snf1p, and PKA all regulate GSY2p, but it appears that this regulation is heavily influenced by the cell's growth status, since (i) pho85 mutants reaching saturation possess glycogen levels similar to those in the wild type (Fig. 1) and (ii) GSY2p activity decreases 20 to 50% when snf1 cells enter stationary phase (15). These decreases may seem modest, but as glycogen is a key regulatory process, small differences have been shown to be physiologically important (32, 42, 45). bcy1 mutations, on the other hand, generate much larger defects in stationary phase; this is expected of mutations which completely abolish the cAMP dependence of PKA rather than merely increasing its activity by elevating cAMP levels. The effects of yvh1 disruptions on Gsy2p activity are comparable to those in snf1 mutants in stationary phase (15). Second is the nature of the nitrogen starvation signal. Starvation specifically for nitrogen dramatically induces YVH1 expression, making it a reasonable participant in transduction of the signal. Loss of Yvh1p decreases, but does not abolish, expression of early meiotic regulators, and Yvh1p is situated upstream of Mck1p, a protein known to modulate expression of the master sporulation regulator IME1 in response to nutritional signals (1, 28). Like Mck1p, Yvh1p also has an additional role in spore maturation (1, 26).

In light of the cAMP-related phenotypes generated by YVH1 disruption, we can suggest an explanation of the mechanism by which Yvh1p regulates early sporulation genes independently of its ability to dephosphorylate proteins. IME1 transcription is repressed by cAMP (21, 33). Sagee et al. localized this effect to an IRE which is homologous to the STREs in DDR2 (19, 33). Disruption of YVH1 modestly decreases IME1 expression and more drastically decreases IME2 transcription, leading to inefficient sporulation (28). This is the same pattern of control we observe with the stress-responsive STRE-driven reporter system (Fig. 9). Direct overexpression of IME1, or increasing its expression by overexpressing MCK1, suppresses both the sporulation and spore maturation defects of yvh1 mutants (1). The spore maturation phenotype of yvh1 mutants is interesting in light of the fact that DIT gene expression, associated with dityrosine production, may be STRE regulated. Further, the two phenotypes, early and late in sporulation, are linked. Cells which sporulate inefficiently are unlikely to synthesize much dityrosine because expression of the DIT1 and DIT2 genes is temporally regulated by the sporulation transcriptional cascade (Ime1pIme2p) (39) as well as, conceivably, by PKA.

The identification of the PKA phenotypes of rim15 mutations is also intriguing in light of the fact that Rim15p was independently identified as being required for maximal expression of IME2 (46), another characteristic shared by Yvh1p. The similarity of rim15 and yvh1 phenotypes may derive from the possibility that the two proteins are components of the same cascade as mentioned above. The rim15 and yvh1 phenotypes also fit the Reinders et al. model for regulating PKA (32). There are, however, very real quantitative differences between the rim15 and yvh1 phenotypes. rim15 mutants sporulate much less efficiently than yvh1 strains (<3% [32] versus 50% [28] with respect to congenic wild-type strains), and rim15 mutants do not possess a vegetative growth defect (32, 46).

The observation that MCK1 and IME1 overexpression suppresses the spore maturation but not the slow-growth defect of yvh1 mutants argues that perhaps more than one Yvh1p target exists. It will be informative to isolate targets of the slow-growth phenotype in light of Muda et al.'s report of a human Yvh1p orthologue which suppresses the slow-growth phenotype of S. cerevisiae yvh1 strains (25). Despite differences in the number of variants and mechanisms of adenyl cyclase activation in yeast and humans, the observation that human YVH1 maps to a chromosomal location that is duplicated in certain liposarcomas prompts the question as to whether cAMP levels, PKA activity, or downstream components of the cascade are also affected in these cell lines (11, 12). If a physiological function of Yvh1p is to modulate cAMP or a downstream component in the cAMP-dependent signal cascade, one might expect Yvh1p orthologues from other eukaryotes to also suppress the S. cerevisiae yvh1 null mutant phenotypes. A Yvh1p homologue has been identified in the Schizosaccharomyces pombe genome, and preliminary experiments suggest that the S. pombe protein is also capable of suppressing the growth and spore maturation defects in HPY120 (Beeser and Cooper, unpublished).