Cdc1 and the Vacuole Coordinately Regulate Mn²⁺ Homeostasis in the Yeast Saccharomyces cerevisiae

Corresponding author: Stephen Garrett, Department of Microbiology and Molecular Genetics, UMDNJ-New Jersey Medical Center, 185 South Orange Ave., University Heights, Newark, NJ 07103-2714, garretst{at}umdnj.edu (E-mail).

Communicating editor: M. JOHNSTON

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The yeast CDC1 gene encodes an essential protein that has been implicated in the regulation of cytosolic [Mn²⁺]. To identify factors that impinge upon Cdc1 or the Cdc1-dependent process, we isolated second-site suppressors of the conditional cdc1-1(Ts) growth defect. Recessive suppressors define 15 COS (CdcOne Suppressor) genes. Seven of the fifteen COS genes are required for biogenesis of the vacuole, an organelle known to sequester intracellular Mn²⁺. An eighth gene, COS16, encodes a vacuolar membrane protein that seems to be involved in Mn²⁺ homeostasis. These results suggest mutations that block vacuolar Mn²⁺ sequestration compensate for defects in Cdc1 function. Interestingly, Cdc1 is dispensable in a cos16 deletion strain, and a cdc1 cos16 double mutant exhibits robust growth on medium supplemented with Mn²⁺. Thus, the single, essential function of Cdc1 is to regulate intracellular, probably cytosolic, Mn²⁺.

CELLS of the yeast Saccharomyces cerevisiae divide by budding. Bud development is a complex process that appears to require the metal ion Mn²⁺. Early studies with ion chelators and Ca²⁺ ionophores (IIDA et al. 1990 ) suggested that Ca²⁺ was essential for cell growth and proliferation, but a subsequent report suggested Mn²⁺, not Ca²⁺, was the limiting ion (YOUATT and MCKINNON 1993 ). Yeast cells depleted of both Mn²⁺ and Ca²⁺ exhibit a defect in bud growth, and Mn²⁺ is 500–1000-fold more effective than Ca²⁺ in reversing the growth defect, suggesting that Mn²⁺ is the physiologically important ion (LOUKIN and KUNG 1995 ).

Yeast Mn²⁺ is present in several subcellular compartments, any one of which might represent, or influence, the Mn²⁺ pool essential for bud growth. The cytosol contains Mn²⁺ as well as such Mn²⁺-dependent enzymes as pyruvate carboxylase, glutamine synthetase, and arginase (WEDLER 1994 ). Mn²⁺ is also present in the Golgi, where it activates glycosyltransferases that are involved in the processing of secreted proteins (WEDLER 1994 ). Mitochondrial Mn²⁺ is required by enzymes of the citric acid cycle (WEDLER 1994 ) as well as proteases involved in mitochondrial protein import (SUPEK et al. 1996 ). Finally, Mn²⁺ is found in the yeast vacuole (OKOROKOV et al. 1977 ), an acidic, membrane-bound organelle that has been implicated in Mn²⁺ detoxification. Of these known Mn²⁺-dependent enzymes and processes, only glycosylation has been shown to affect bud growth. Mutants of och1 lack a Mn²⁺-dependent Golgi glycosyltransferase and exhibit a conditional defect in bud growth, but Och1 is not essential under standard growth conditions (NAGASU et al. 1992 ). Thus, it is not clear if Och1, or even Golgi Mn²⁺, is related to the Mn²⁺ requirement identified by the depletion studies.

Several proteins have been implicated in intracellular Mn²⁺ homeostasis. Mn²⁺ uptake across the plasma membrane appears to be mediated by Smf1, an integral plasma membrane protein (SUPEK et al. 1996 ). A second protein, Smf2, shares significant homology with Smf1 and might also be involved in Mn²⁺ uptake (EIDE and GUERINOT 1997 ). Of the intracellular organelles, only the Golgi has been characterized in any detail with respect to Mn²⁺ flux. Two structurally unrelated proteins, Pmr1 (a Ca²⁺-ATPase homologue) and Ccc1, seem to play a role in the transport of Mn²⁺ from the cytosol into the Golgi (LAPINSKAS et al. 1995 ; LAPINSKAS et al. 1996 ). A third protein, Atx2, has been implicated in the release of Golgi Mn²⁺ into the cytosol (LIN and CULOTTA 1996 ). Although two of these proteins (Pmr1 and Smf1) are essential for growth under Mn²⁺-limiting conditions, the phenotypes of Mn²⁺-depleted pmr1 or smf1 strains have not been reported. Thus, it is not clear if the activity of either protein affects the Mn²⁺ pool that is required for bud growth.

One factor that does affect the growth-related Mn²⁺ pool is the essential gene CDC1. Conditional cdc1(Ts) mutants were originally identified on the basis of a defect in bud growth (HARTWELL et al. 1970 ), and several lines of evidence implicate Cdc1 in the regulation of cytosolic Mn²⁺. First, Mn²⁺ supplement partially suppresses the cdc1(Ts) growth defect (LOUKIN and KUNG 1995 ; PAIDHUNGAT and GARRETT 1998 ), suggesting that Mn²⁺ is limiting in cdc1 mutants. Second, defects in CDC1 cause cells to become sensitive to depletion of cytosolic Mn²⁺ (SUPEK et al. 1996 ; PAIDHUNGAT and GARRETT 1998 ). Finally, CDC1 overexpression makes certain chelator-sensitive mutants more tolerant to Mn²⁺ depletion (PAIDHUNGAT and GARRETT 1998 ). Thus, we propose that the bud-growth defect of cdc1(Ts) mutants results from depletion of cytosolic Mn²⁺, which implies, in turn, that cytosolic Mn²⁺ is critical for bud growth (PAIDHUNGAT and GARRETT 1998 ). However, neither the mechanism by which Cdc1 regulates cytosolic Mn²⁺ nor the Mn²⁺-dependent processes are known.

To identify cellular components that affect the essential Cdc1 function, we isolated suppressors of the cdc1-1(Ts) temperature-sensitive growth defect. Several recessive suppressors define genes previously implicated in biogenesis of the vacuole, an organelle known to sequester divalent cations such as Mn²⁺. In addition, one suppressor gene, COS16, encodes a vacuolar membrane protein that appears to be involved in vacuolar Mn²⁺ sequestration. These results suggest mutations that either directly or indirectly block vacuolar Mn²⁺ sequestration compensate for defects in Cdc1 function. Our results also show that Cdc1 is dispensable in cells lacking Cos16, suggesting that Cdc1 function is not required in cells that exhibit a defect in vacuolar Mn²⁺ sequestration. Thus, it seems likely that the essential function of Cdc1 is to maintain cytosolic [Mn²⁺] above a minimum threshold that is required for growth.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Media:
Standard yeast media were prepared as previously described (PAIDHUNGAT and GARRETT 1998 ). LiCl and MgCl₂ were added to rich yeast medium (YEPD) after autoclaving. Medium containing CaCl₂ was made by autoclaving YEPD medium, buffering to pH 5.5 with 50 mM succinate/NaOH and then adding CaCl₂ to the indicated concentration.

Strains and plasmids:
Revertants were derived from strains FY11 [MATa ade1 trp1 leu2 his3 ura3 cdc1-1(Ts)] and FY12 [MAT ade8 trp1 leu2 his3 ura3 cdc1-1(Ts)]. Strain FY 70 (FY11 CDC1) was derived from FY11 by reversion and contains the wild-type CDC1 allele. Strain FY 71 (MAT ade8 trp1 leu2 his3 ura3 CDC1) has been described (PAIDMUNGAT and GARRETT 1998) and was used to determine suppressor linkage to CDC1. Diploid strain FY1 (MATa/MAT cdc1::HIS3/CDC1 ade2/ADE2 trp1/TRP1 ura3/ura3 leu2/leu2 his3/his3 lys2/lys2) and a cos16 derivative (FY1 cos16::LEU2/COS16) were derived from strain Y1029 (GARRETT and BROACH 1989 ) by transformation. Bacterial strains MC1066 and DH5 were used for plasmid manipulations and have been described (CASADABAN et al. 1983 ; WOODCOCK et al. 1989 ).

Plasmid YEp13-CDC1 (pFB28) has been described (PAIDHUNGAT and GARRETT 1998 ). Plasmid Y Ip5-CDC1 (pFB56) was constructed by inserting the 3.5-kb HindIII CDC1 fragment into the HindIII site of Y Ip5 (NEB catalogue; New England Biolabs, Beverly, MA). The resulting plasmid was integrated at the CDC1 locus by linearizing with XbaI before yeast transformation.

Genetic manipulations:
The cdc1(Ts) revertants were isolated by patching independent colonies of FY11 and FY12 onto YEPD agar plates, which were then incubated at either 30° or 36° for 2 days. Temperature-resistant revertants arose as papillae from the patches; a single papilla was picked from each patch to ensure that revertants were independent. At 23°, two revertants gave rise to small, temperature-resistant colonies as well as large, temperature-sensitive colonies. Crosses showed that the small colonies were disomic for chromosome IV, whereas the large colonies contained a single copy of chromosome IV.

Cloning COS4, COS5, and COS15:
COS4, COS5, and COS15 were cloned from a YCp50-based yeast genomic library (ROSE et al. 1987 ) by their ability to complement the temperature-sensitive growth defects of cos4-8 cdc1-1 [YEp13-CDC1], cos5-17 cdc1-1 [YEp13-CDC1], and cos15-118 cdc1-1 [YEp13-CDC1] strains, respectively. Four COS4 plasmids, pFB77–pFB80, contained distinct, but overlapping, genomic regions that also reversed the temperature-resistant growth of a cos4-8 cdc1-1(Ts) strain. Plasmid pFB87 was generated by subcloning the 4.3-kb HindIII-Bgl II genomic fragment from pFB78 into the HindIII and BamHI sites of the low-copy URA3 vector pRS316 (SIKORSKI and HIETER 1989 ). This fragment complemented all of the cos4 mutant phenotypes and was physically linked to the PEP3 region (RILES et al. 1993 ). The PEP3 gene was disrupted by inserting the HIS3 marker into a unique BamHI site in pFB78 to generate plasmid pFB107. The pep3::HIS3 disruption was excised from pFB107 using EcoRI.

Seven (pFB94–pFB100) of eleven COS5 plasmids recovered contained distinct, but overlapping, genomic regions. Plasmid pFB109 was constructed by inserting the 4.6-kb Bgl II-BamHI fragment from pFB99 into the BamHI site of pRS316. This plasmid complemented all of the cos5 mutant phenotypes. Plasmid pFB162 was used to mark the COS5 locus and was constructed by inserting the 1.9-kb EcoRI fragment from pFB109 into a derivative (pFB119) of the integrating vector pRS306, which lacked the SpeI and XbaI sites in the multiple cloning sequence (MCS). Plasmid pFB162 was linearized at the unique XbaI site in the insert and integrated by homology-directed recombination into the yeast genome.

Two identical COS15 plasmids, pFB135 and pFB136, contained the PEP5 gene, as determined by comparing its restriction map to that of a YCp50-PEP5 plasmid obtained from E. JONES (Carnegie Mellon University, Pittsburgh).

Cloning and disruption of COS8:
The COS8 gene was originally cloned as a high-copy suppressor of the cdc1-1(Ts) growth defect. Plasmid pFB57 was isolated from a high-copy (pRS202) yeast genomic library (C. CONNELLY and P. HIETER, unpublished results) by its ability to suppress the temperature-sensitive growth defect of strain FY11 (cdc1-1) at 30°. A 1.8-kb HindIII genomic fragment from pFB57 was subcloned into the HindIII site of pRS202 to generate plasmid pFB67, which suppressed the cdc1-1(Ts) temperature-sensitive growth defect. Interestingly, plasmid pFB67 complemented the salt sensitivity of the cos8 mutants. To determine if pFB57 contained COS8, the 1.8-kb HindIII fragment was subcloned into the low-copy vector pRS316. The resulting plasmid, pFB161, complemented all of the cos8 mutant phenotypes. The 1.8-kb HindIII fragment was physically mapped to a region of chromosome XVI (RILES et al. 1993 and references therein) that contained a single, complete open reading frame, VPS4/END13.

The 1.8-kb HindIII COS8 fragment was inserted into pUC8 (NEB catalogue) to generate plasmid pFB217. Two different derivatives of pFB217 were used to disrupt the chromosomal COS8 gene. In the first construct, the URA3 marker was inserted into a unique MscI site in COS8 to generate plasmid pFB277. The cos8::URA3 disruption was liberated from plasmid pFB277 with KpnI, and transplaced into the yeast genome. To confirm the cos8::URA3 disruption from pFB277 represented the complete loss of Cos8 function, a cos8::URA3 deletion was constructed from pFB217 by replacing the EcoRV-SpeI region of COS8 with URA3 to generate plasmid pFB275. The cos8::URA3 disruption was liberated with HindIII and transplaced into the yeast genome. Although transformation frequencies using the second construct were reduced by the limited flanking homology, transformants derived from either construct exhibited identical phenotypes.

Cloning COS9 :
The recessive, cold-sensitive (18°) growth defect of the cos9-26(Cs) revertant was exploited to select COS9 clones from a low-copy yeast genomic library (ROSE et al. 1987 ). Two plasmids, pFB188 and pFB190, containing overlapping regions complemented the conditional growth defect of the cos9-26(Cs) strain and reversed the temperature-resistant growth of a cdc1-1(Ts) cos9-26(Cs) double mutant at 30°. A 1.0-kb EcoRI-XbaI complementing fragment from plasmid pFB188 was subcloned into pRS316 to generate plasmid pFB220. This region was physically linked to the ARF1 region on chromosome IVL (RILES et al. 1993 and references therein). A frameshift mutation was introduced into the ARF1 gene by digesting pFB220 with Bgl II and filling in the staggered ends to generate plasmid pFB254. The arf1::HIS3 construct was obtained from DR. R. KAHN (Emory University, Atlanta).

Cloning COS16:
The COS16 gene was cloned from a low-copy yeast genomic library (ROSE et al. 1987 ) by screening for temperature-sensitive transformants of a cdc1-1(Ts) cos16-57 mutant. Two plasmids (pFB163 and pFB164) with overlapping inserts were recovered. Plasmids pFB168 and pFB169 were derived from the low-copy LEU2 vector pRS315 (SIKORSKI and HIETER 1989 ) and contained 2.7-kb PstI-BamHI and 3.2-kb HindIII-PstI fragments, respectively, from pFB163. Plasmid pFB179 was generated by inserting a complementing 2.1-kb Bgl II-SpeI (the SpeI site was from the MCS) fragment from pFB168 between the BamHI and XbaI sites of pRS316. Physical analysis of the 2.1-kb fragment (RILES et al. 1993 and references therein) showed it contained a single, complete open reading frame, YCR44c. A frameshift mutation was introduced into YCR44c by linearizing pFB179 at the unique XbaI site within YCR44c, filling in the staggered ends, and religating to generate plasmid pFB182.

The cos16::HIS3 disruption was generated in several steps. First, the 2.7-kb PstI-BamHI insert from pFB168 was inserted into the XbaI and XhoI sites of pRS306, using the SpeI and Sal I sites of the pFB168 MCS. This left unique XhoI and XbaI sites in the insert of the resulting plasmid pFB211. The 0.6-kb XhoI-XbaI fragment within the COS16 coding region of pFB211 was replaced with a XhoI-XbaI HIS3 fragment to generate plasmid pFB218. The cos16::HIS3 allele was transplaced into the yeast genome after liberating it from plasmid pFB218 with EcoRI. The cos16::LEU2 allele was constructed by replacing the XhoI-SnaBI region of YCR44c in plasmid pFB211 with a Sal I-HpaI LEU2 fragment to generate plasmid pFB371. The cos16::LEU2 allele was liberated from pFB371 with BamHI and Bgl II.

The COS16-lacZ fusion was generated by inserting the MfeI fragment carrying the COS16 open reading frame into the EcoRI site of YEp356R (MYERS et al. 1986 ) to generate plasmid pFB325. This fusion lacked the carboxy-terminal residue of the Cos16 protein. An integrating version (pFB333) of the COS16-lacZ fusion plasmid was constructed by inserting the AatII COS16-lacZ fusion fragment from pFB325 in place of the homologous AatII fragment of YIp356R (MYERS et al. 1986 ). Plasmid pFB333 was integrated at the URA3 locus after linearizing with StuI.

Vacuolar staining and carboxypeptidase Y (CPY) secretion:
Log-phase cells were stained with the vital vacuolar dye CDCFDA (Molecular Probes, Eugene, OR), using a procedure described previously (MANOLSON et al. 1992 ), except that CDCFDA was used instead of CFDA. CPY secretion was detected by a colony hybridization assay (ROBERTS et al. 1991 ).

Isolation of vacuoles and Cos16 localization:
Crude fractionation of cell extracts was performed as described (COWLES et al. 1994 ), except that the ³⁵S-labeling step was omitted. Vacuoles were purified as described (CARDENAS et al. 1995; OHSUMI and ANRAKU 1981 ), with the exception that the SW40Ti rotor was used for ultracentrifugation (rpms were adjusted to generate the required g force). Protein content of extracts was measured by the Bradford assay (Bio-Rad, Richmond, CA) with BSA standards. For antigen detection and quantitation, extracts were diluted into sample buffer, heated at 65° for 10 min, separated on 10% SDS-PAGE, and blotted to nitrocellulose (Shleicher and Schuell, Keene, NH). The Cos16-LacZ fusion protein was detected with -ß-Gal antiserum (Cappel Research Products). CPY and Vph1 were detected with monoclonal antibodies from Molecular Probes (Eugene, OR). Kar2 and Pma1 were detected using antisera obtained from M. D. ROSE (Princeton University, Princeton, NJ) and M. CARDENAS (Duke University Medical Center, Durham, NC), respectively. HRP-conjugated secondary antibodies (Sigma Chemical, St. Louis; Promega, Madison, WI) were used for detection with the ECL system (Amersham, Arlington Heights, IL).

Mn²⁺ estimation:
The cells were prepared as previously described (LAPINSKAS et al. 1995 ), and concentrated by centrifugation to 20–80 A₆₀₀ units/ml. The cell slurry (25 µl, 1–2 A₆₀₀ units) was digested with 50% nitric acid for 2 hr at 95° (until solution cleared) in an acid-rinsed Eppendorf tube, cooled to room temperature, and diluted to 1 ml with deionized water. Atomic absorption spectroscopy was performed with a L'vov platform furnace in a Z3030 (Perkin-Elmer, Norwalk, CT) instrument with 25 µg Mg(NO₃)₂ (Sigma Chemical, ACS grade) matrix modifier. The furnace temperature program was: Dry 140° (10 sec ramp, 60 sec hold); Char 1400° (8 sec ramp, 25 sec hold); Atomize and reading 2200° (0 sec ramp, 5 sec hold); Clean 2650° (1 sec ramp, 4 sec hold); Cool 20° (1 sec ramp, 30 sec hold). Argon gas flow was maintained at 300 ml/min, except during atomization (0 ml/min). Each digested sample was analyzed twice and averaged. The results of two independent experiments are shown. The Mn²⁺ content in each sample was normalized to the A₆₀₀.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Suppressors of the cdc1-1(Ts) temperature-sensitive growth defect:
Suppressors of the cdc1-1(Ts) growth defect were isolated by selecting spontaneous, temperature-resistant revertants of MATa cdc1-1(Ts) and MAT cdc1-1(Ts) strains (see MATERIALS AND METHODS). Fifty-four independent revertants were isolated at 30°, and two revertants were recovered at 36°. None of the revertants isolated at 30° grew at 36°, suggesting that the growth defect at 30° was less severe than that at 36° (Figure 1). Three of the 54 revertants isolated at 30°, and both revertants isolated at 36°, contained dominant suppressors. Temperature-resistant growth segregated 2:2 in all cases, indicating that each revertant carried a single nuclear suppressor mutation.

View larger version (98K): In this window In a new window Download PPT slide   Figure 1. —Suppressors of the cdc1-1(Ts) growth defect. The parent cdc1-1(Ts) strain and seven spontaneous revertants were streaked onto Y EPD agar and incubated at 23°, 30°, and 36°. Strains were: C DC1 (FY 70 = FY 11 C DC1); cos2 (FY 11 cos2-2); cos11 (FY 11 cos11-29); cos4 (FY 11 cos4-8); cos16 (FY 11 cos16-57); cos3 (FY 11 cos3-3); cos1 (FY 11 cos1-1); cdc1-1 (FY 11).

Intragenic suppressors were identified by crossing each cdc1-1(Ts) cos (cdcone suppressor) strain to a wild-type CDC1 COS strain, and examining tetrads for temperature-sensitive recombinants [cdc1-1(Ts) COS]. Suppressors in both revertants isolated at 36°, as well as one recessive revertant isolated at 30°, failed to recombine with the cdc1-1(Ts) lesion (>30 tetrads examined). The cdc1-1(Ts) allele resulted from a single C to T transition [Pro₃₅₁ (CCT) to Leu₃₅₁ (CTT)]. The CDC1 gene from one of the revertants (FY70) isolated at 36° contained the wild-type CCT sequence at codon 351; the recessive intragenic suppressor, cdc1-130, retained the original cdc1-1(Ts) lesion and a second C to T transition [Thr₂₃₁ (ACA) to Ile₂₃₁ (ATA)].

Aberrant segregation of the ade8 marker suggested two recessive revertants were disomic for chromosome IV. Because CDC1 is located on chromosome IV, it seemed likely that cdc1-1(Ts) duplication was sufficient to allow growth at 30°. Consistent with this idea, a cdc1-1(Ts) haploid strain carrying an extra copy of the cdc1-1(Ts) allele on an integrating plasmid grew at 30° (data not shown). By contrast, a diploid strain containing four copies of the cdc1-1(Ts) allele failed to grow at 30° (data not shown). Thus, diploid cells need disproportionately more Cdc1 activity than haploid cells. This may explain why one of the intragenic suppressors (cdc1-130) and the chromosome IV duplications were recessive.

Recessive suppressors define 15 genes that fall into four phenotypic groups:
Forty-eight of the 54 suppressors isolated at 30° were recessive, and unlinked to CDC1. These were placed into complementation groups by testing diploids from pairwise matings of the cdc1(Ts) cos isolates for growth at 30°. By this analysis, 40 of the 48 suppressors were assigned to 15 different complementation groups (Table 1).

Several cdc1-1(Ts) cos double mutants were sensitive to salts at 23°. To examine the salt sensitivity in more detail, we transformed a representative mutant from each complementation group with a YIp5-CDC1 plasmid, and tested growth on medium containing 200 mM CaCl₂, 200 mM MgCl₂, and 100 mM LiCl. Seven CDC1 cos mutants (cos4, cos5, cos15, cos6, cos11, cos8, and cos9) exhibited varying degrees of sensitivity to all of the salts tested (Figure 2 and data not shown; cos6 and cos11 mutants were less sensitive than cos4 mutants to all salts, and their sensitivity to 200 mM MgCl₂ is not readily apparent in the figure). Interestingly, three cos mutants (cos1, cos3, and cos14) were sensitive only to CaCl₂ and LiCl, two cos mutants (cos13 and cos16) were specifically sensitive to MgCl₂ (Figure 2 and data not shown), and three cos mutants (cos2, cos7, and cos10) did not exhibit measurable sensitivity to any of the salts tested (Figure 2 and data not shown). In all cases, salt sensitivity and cdc1-1(Ts) suppression were tightly linked (no recombination in >10 tetrads), implying a single mutation was responsible for both phenotypes. Finally, sensitivity to specific cations probably reflects differences in the cellular functions of particular COS genes because all of the mutants within each of the latter complementation groups (cos1, cos3, cos14, cos13, cos16, cos2, cos7, and cos10) exhibited sensitivity to the same range of cations as the representative isolate (Figure 2 and data not shown). Accordingly, the CDC1 cos mutants were divided into four broad phenotypic groups, I to IV (Table 2). Mutations in three group III genes, COS4, COS5, and COS15, also blocked accumulation of red pigment in an ade1 background and conferred a temperature-sensitive (36°) growth defect in a CDC1 strain (data not shown). Thus, group III genes were subclassified as group IIIA (COS4, COS5, and COS15), and group IIIB (COS6, COS8, COS9, and COS11).

View larger version (70K): In this window In a new window Download PPT slide   Figure 2. —Salt tolerance of the cos mutants. The indicated strains were transformed with the Y Ip5-C DC1 plasmid, and then streaked onto YEPD agar, YEPD pH 5.5 agar containing 200 mm CaCl2, Y EPD agar containing 200 mm MgCl2, or Y EPD agar containing 100 mm LiCl, and incubated at 30° for 3–5 days. Strains were: cos2 (FY 11 cos2-2); cos11 (FY 11 cos11-29 ); cos6 (FY 11 cos6-18); cos4 (FY 11 cos4-8); cos16 (FY 11 cos16-57); cos3 (FY 11 cos3-3); cos1 (FY 11 cos1-1); + (FY 11).

Group IIIA COS genes are identical to class C VPS genes:
The ade1 pigment-accumulation defect, temperature sensitivity, and salt sensitivity of the group IIIA cos mutants resembled phenotypes exhibited by a subset of vacuolar protein sorting (vps/pep) mutants (BANTA et al. 1988 ). The group IIIA cos mutants also displayed several additional phenotypes exhibited by a subset of the vps/pep mutants, including sensitivity to 500 mM NaCl, 4 mM ZnCl₂, and 8 mM MnCl₂, as well as failure to sporulate as homozygous diploids (data not shown). To determine if group IIIA COS genes were related to known VPS genes, we isolated the COS4, COS5, and COS15 genes from a yeast genomic library by complementing the temperature-sensitive growth defect of CDC1 cos4, CDC1 cos5, and CDC1 cos15 strains, respectively (see MATERIALS AND METHODS). We also examined the ability of cos4, cos5, and cos15 mutants to complement the phenotypes of known vps/pep mutants. These studies showed (Table 1) that COS4 is identical to VPS18/PEP3 (PRESTON et al. 1991 ; ROBINSON et al. 1991 ), COS5 is identical to VPS16/VPH4 (HORAZDOVSKY and EMR 1993 ; PRESTON et al. 1992 ), and COS15 is identical to VPS11/PEP5 (WOOLFORD et al. 1990 ). Finally, these assignments were confirmed by showing that mutations inactivating VPS18 (pep3::LEU2), VPS16 (vph4-5) or VPS11 (pep5::URA3), as well as a fourth class C VPS gene VPS33 (pep14-5) (BANTA et al. 1990 ) suppressed the growth defect of a cdc1-1(Ts) strain at 30° (data not shown). Thus, mutations in all four class C VPS genes suppressed the cdc1-1(Ts) growth defect at 30°.

Group IIIB cos mutants mislocalize the vacuolar protein CPY:
Group IIIB cos mutants shared several phenotypes with the class C vps mutants, including general salt sensitivity and a sporulation defect, with the class C vps mutants (although cos9 mutants did not exhibit a sporulation defect). To determine if group IIIB COS genes were also involved in vacuolar protein sorting, we examined localization of the vacuolar protein CPY. Wild-type strains efficiently target CPY to the vacuole. Mutants with defects in the vacuolar protein sorting pathway, by contrast, secrete significant amounts of CPY (RAYMOND et al. 1992 ). Similar to the class C vps18 (cos4-8) mutant, all of the group IIIB cos mutants (cos6, cos8, cos9, and cos11 mutants) secreted significantly more CPY than the wild-type strain (Figure 3B and Figure C). The sorting defect was less apparent, but nevertheless obvious, in the cos9 mutant (Figure 3C). Thus, group IIIB COS genes are necessary for efficient vacuolar protein sorting. By contrast, cos mutants from groups II and IV did not exhibit significant CPY secretion (Figure 3), and group I cos mutants displayed a slight defect in CPY secretion (Figure 3A) that became more severe in older colonies.

View larger version (61K): In this window In a new window Download PPT slide   Figure 3. —CPY secretion. Patches of the indicated strains were replica plated to YEPD agar, overlaid with nitrocellulose membrane, and incubated at 23°. After 12 hr, membranes were rinsed in water and probed with -CPY antibodies. All strains in rows A–C carried a YIp5-CDC1 plasmid. Strains were, from left to right, row A: FY11 cos16-57, FY11 cos3-3, FY11 cos1-1, and FY11; row B: FY11 cos2-13, FY11 cos11-29, FY11 cos6-18, and FY11 pep3-8; row C: FY11 arf1-32, FY11 pep3-8, FY11 vps4-21, FY11; row D: FY70 [pRS202-COS4], FY70 [pRS202], FY11 [pRS202-COS4], FY11 [pRS202].

Mutations in class D VPS genes suppress the cdc1-1(Ts) growth defect:
To determine if group IIIB cos mutations affected vacuolar biogenesis, we examined vacuolar morphology using the vital stain CDCFDA. In contrast to the multilobed vacuolar structure of wild-type cells (Figure 4), cos6 and cos11 mutant cells displayed a single vacuolar lobe (Figure 4 and data not shown) that was similar to the abnormal vacuolar morphology of class D vps mutants (RAYMOND et al. 1992 ). Like class D vps mutants, cos6 and cos11 cells failed to form vacuolar segregation structures and exhibited a defect in vacuolar inheritance (Figure 4 and data not shown). Finally, disruption of the class D VPS gene, VPS19/PEP7/VAC1 (WEISMAN and WICKNER 1992 ), alleviated the growth defect of a cdc1-1(Ts) mutant at 30° (data not shown). Thus, loss of class D Vps function suppressed the cdc1(Ts) growth defect. COS6 and COS11 were distinct from PEP7 and might define other class D VPS genes.

View larger version (69K): In this window In a new window Download PPT slide   Figure 4. —Vacuolar morphology of cos mutants visualized by CDCFDA. Exponentially growing cells were stained with the vital dye CDCFDA and visualized by differential interference contrast (DIC) or fluorescence (CDCFDA) microscopy. Strains: COS+ (FY 11 [YIp5-C DC1]); cos4 (vps18) (FY 11 cos4-8 [YIp5-C DC1]); cos6 (FY 11 cos6-18 [Y Ip5-C DC1]); cos8 (vps4) (FY 11 cos8-21 [Y Ip5-CDC1]); cos9 (arf1) (FY 11 cos9-32 [Y Ip5-C DC1]).

COS8 is identical to VPS4/END13, a class E VPS gene:
cos8 cells exhibited a single, prominent vacuolar structure surrounded by one or more small, CDCFDA-staining vesicles (Figure 4). Although this vacuolar morphology resembled that of wild-type cells (Figure 4), class A and class E vps mutants also contain normal-looking vacuoles (RAYMOND et al. 1992 ). Indeed, molecular characterization (see MATERIALS AND METHODS) showed that COS8 was identical to VPS4/END13, a class E VPS gene implicated in vacuolar protein sorting (MUNN and RIEZMAN 1994 ). Furthermore, a vps4::URA3 disruption suppressed the cdc1-1(Ts) growth defect at 30° (data not shown), indicating that loss of class E Vps function suppressed the cdc1-1(Ts) growth defect.

Because VPS4 was also identified as a high-copy suppressor of the cdc1-1(Ts) growth defect (see MATERIALS AND METHODS), we determined if Vps4 overproduction blocked vacuolar protein sorting. As shown in Figure 3D, an increase in VPS4 dosage caused CDC1 and cdc1-1(Ts) cells to secrete more CPY, consistent with the idea that Vps4 overproduction suppressed the cdc1-1(Ts) growth defect by impeding vacuolar protein sorting.

Loss of Arf1 function suppresses the cdc1-1(Ts) growth defect:
The last group IIIB mutant, cos9, displayed numerous (>15) CDCFDA-staining vesicles, indicative of vacuolar fragmentation (Figure 4; this phenotype was clearer under Nomarski optics because vesicles outside the focal plane interfered with visualization of the CDCFDA fluorescence). Interestingly, cos9 mutants also exhibited abnormalities in cell size and shape (Figure 4), suggesting Cos9 might affect several cellular processes. We cloned COS9 and found it was identical to the ARF1 gene (see MATERIALS AND METHODS), which encodes a GTPase implicated in transport between the endoplasmic veticulum (ER) and Golgi compartments (STEARNS et al. 1990 ). As expected, the arf1::HIS3 disruption allowed a cdc1-1(Ts) mutant to grow at 30° (data not shown). Although ARF1 has not been previously described as a VPS gene, the arf1::HIS3 mutant secreted CPY, displayed abnormal vacuolar morphology, and exhibited decreased tolerance to salt. Because Arf1 function is required for normal vacuolar biogenesis, the cos9 mutations probably suppress the cdc1-1(Ts) growth defect by debilitating vacuolar function.

Proteinase A deficiency does not suppress the cdc1-1(Ts) growth defect:
Because the group III cos mutations blocked formation of a normal vacuole, the cdc1(Ts) growth defect was probably alleviated by the absence of a normal vacuole, rather than a defect at a specific stage of vacuolar protein sorting. The yeast vacuole plays a major role in protein turnover. Several vacuolar hydrolases, including proteinase B and CPY, are activated by the vacuolar Pep4 proteinase (proteinase A). As a result, loss of Pep4 function results in a 90–95% reduction in vacuolar hydrolase activity ( JONES et al. 1982 ). Nevertheless, a cdc1-1(Ts) pep4::LEU2 strain failed to grow at 30° (data not shown). Thus, neither the loss of Pep4 nor the accompanying reduction in vacuolar hydrolase activity could account for cdc1(Ts) suppression by defects in vacuolar biogenesis.

Group II gene COS16:
The group II cos mutants exhibited a subset of the group III cos mutant phenotypes, including Mg²⁺ sensitivity (Figure 2), failure to sporulate, and a slight growth defect at 36° (data not shown). However, group II cos mutants did not secrete CPY (Figure 3). We cloned the COS16 gene from a yeast genomic library by complementation of the temperature-resistant growth of the cdc1-1(Ts) cos16-57 mutant. The complementing fragment was mapped to a region of chromosome IIIR (data not shown), which contained a single 1071-bp open reading frame, YCR44c. A frameshift mutation introduced into the YCR44c coding region abolished the ability of the 2.1-kb fragment to complement the cos16-57 mutation (data not shown). In addition, a YCR44c disruption suppressed the cdc1(Ts) temperature-sensitive growth defect (data not shown) and conferred Mg²⁺ sensitivity and temperature-sensitive growth (at 36°) to a CDC1 strain (data not shown). Finally, all of the spontaneous cos16 alleles were tightly linked (<2.5 cM with at least 20 tetrads) to the MAT locus, as predicted by the physical proximity of YCR44c to MAT. Thus, COS16 is identical to the open reading frame YCR44c.

Subcellular localization of Cos16:
The predicted primary structure of the Cos16 protein did not exhibit significant homology to proteins in the available databases. Nevertheless, hydropathy analysis (KYTE and DOOLITTLE 1982 ) showed that Cos16 contained 8 putative membrane-spanning helices, suggesting that it was an integral membrane protein. To examine Cos16 localization, we constructed a hybrid gene between COS16 and lacZ. A single copy of the COS16-lacZ fusion was integrated at the URA3 locus in a cos16::HIS3 mutant. Most (>80%) of the Cos16-LacZ fusion protein was detected in the particulate fraction (100,000 g for 1 hr), and could be solubilized with Triton X-100, but not 1 M NaCl or 2 M urea (data not shown). Thus, the Cos16-LacZ fusion protein is probably an integral membrane protein.

The subcellular localization of the Cos16-LacZ fusion protein was initially examined by crude fractionation. Cell lysates were fractionated by sequential centrifugation (COWLES et al. 1994 ), and the fusion protein in each fraction was detected by Western blotting. Greater than 90% of the Cos16-LacZ fusion protein cofractionated with the vacuolar membrane marker, Vph1, as well as the ER lumenal marker, Kar2, in the P13 fraction (data not shown). To determine if COS16 encoded a vacuolar membrane protein, vacuoles were purified by flotation on a Ficoll gradient (CARDENAS and HEITMAN 1995 ), and analyzed for presence of the Cos16-LacZ fusion protein. The intact Cos16-LacZ fusion was enriched in the vacuolar extract to an extent that was comparable to the vacuolar lumenal marker, CPY (Figure 5). By contrast, the ER protein, Kar2, and the plasma-membrane protein, Pma1, were not significantly enriched in the vacuolar fraction (Figure 5). Similar studies using vacuolar vesicles purified by a different method (OHSUMI and ANRAKU 1981 ) yielded identical results (data not shown). These results suggest that Cos16 is an integral membrane protein of the vacuole.

View larger version (62K): In this window In a new window Download PPT slide   Figure 5. —Subcellular localization of Cos16-ß-gal fusion protein. Protein from whole-cell lysates (Total) or purified vacuoles (Vacuole) were separated by SDS-PAGE and transferred to nitrocellulose membrane. The ß-gal lane contains purified ß-galactosidase. Vacuoles were purified as previously described (CARDENAS and HEITMAN 1995 ). The membrane was probed with antibodies against bacterial ß-gal protein, as well as the yeast proteins CPY, Kar2, and Pma1.

Misfolded membrane proteins default to the vacuole (STACK and EMR 1993 ). However, that seems an unlikely explanation for the vacuolar localization of the Cos16-LacZ fusion protein, because single-copy COS16-lacZ complemented all of the cos16 phenotypes (data not shown). Moreover, a different epitope-tagged COS16 fusion, in which the myc epitope was fused to the C terminus of Cos16, was functional, and exhibited a similar localization pattern (data not shown). Thus, the vacuolar localization of the Cos16-LacZ fusion protein probably reflects the subcellular location of the native Cos16 protein.

Cos16 and Mn²⁺ homeostasis:
Because the vacuole sequesters Mn²⁺ from the cytosol (OKOROKOV et al. 1977 ), we examined if Cos16 was also involved in Mn²⁺ homeostasis. The cos16::HIS3 deletion had a modest effect, similar to that of the vps4 mutation, on the Mn²⁺ sensitivity of an otherwise wild-type strain (data not shown). Mutations in the Golgi Mn²⁺ transporter gene, PMR1, increase sensitivity to Mn²⁺ as the result of Mn²⁺ accumulation in the cytosol (LAPINSKAS et al. 1995 ). The cos16::HIS3 deletion exacerbated this sensitivity (Figure 6A). Thus, deletion of COS16 altered tolerance to Mn²⁺.

View larger version (251K): In this window In a new window Download PPT slide   Figure 6. —COS16 and Mn2+ homeostasis. (A) cos16 exacerbates the Mn2+ sensitivity of a pmr1 mutant. The indicated strains were streaked onto YEPD pH 5.5 agar, supplemented with 0 mm, 2 mm, or 4 mm MnCl2, and incubated at 30° for 3–5 days. Strains were: cos16 pmr1 (FY 70 cos16::HIS3 pmr1::LEU2); COS16 pmr1 (FY70 pmr1::HIS3), cos16 PMR1 (FY70 cos16::HIS3); COS16 PMR1 (FY70). (B) cos16 suppresses the EGTA sensitivity of a pmr1 mutant. The strains of panel A were streaked onto YEPD agar, supplemented with 0 mm or 0.5 mm EGTA, and incubated at 30° for 3–5 days. (C) cos16 suppresses the EGTA sensitivity of a smf1 mutant. 10-fold dilutions of exponentially growing cultures of the indicated strains were spotted onto YEPD agar or YEPD agar containing 4 mm EGTA, and incubated at 30° for 3 days. Strains were: + (FY70), cos16 (FY70 cos16::HIS3), smf1 (FY70 smf1::URA3), and cos16 smf1 (FY 70 cos16::HIS3 smf1::URA3). (D) cos16 suppresses the EGTA sensitivity of a cdc1-1(Ts) mutant. The indicated strains were streaked onto YEPD agar or YEPD agar containing 1.5 mm EGTA and incubated at 30° for 3 days. Strains were: cdc1 cos16 (FY11 cos16::HIS3); cdc1 (FY11); CDC1 (FY70); cos16 (FY70 cos16::HIS3).

We also examined the effect of COS16 on the chelator sensitivities of several Mn²⁺-homeostasis mutants. Strains lacking Pmr1 function are sensitive to EGTA, and this sensitivity can be ameliorated by increasing Mn²⁺ influx across the plasma membrane (PAIDHUNGAT and GARRETT 1998 ). As shown in Figure 6B, the cos16::HIS3 deletion allowed a pmr1::LEU2 strain to grow on medium containing 0.5 mM EGTA. Mutations in SMF1 also cause cells to become sensitive to EGTA, in this case as the result of the loss of a high-affinity Mn²⁺ uptake system. This EGTA sensitivity is alleviated by Mn²⁺ supplement (SUPEK et al. 1996 ), as well as overproduction of Cdc1 (PAIDHUNGAT and GARRETT 1998 ). Again, only the cos16::HIS3 smf1::URA3 double mutant tolerated 4 mM EGTA in the medium (Figure 6C). Finally, the EGTA sensitivity (1.5 mM) of the cdc1-1(Ts) strain was alleviated by the cos16::HIS3 deletion (Figure 6D). These findings implicate Cos16 in Mn²⁺ homeostasis and suggest Cos16 is involved in the sequestration of Mn²⁺ into the vacuole. The last result is also consistent with the notion that the cos16 mutations alleviate the cdc1(Ts) conditional growth defect through their effect on Mn²⁺ homeostasis.

Mn²⁺ content of cdc1 and cos16 mutants:
Several genes that affect Mn²⁺ homeostasis, including PMR1 and ATX2, alter intracellular Mn²⁺ content (LAPINSKAS et al. 1995 ; LIN and CULOTTA 1996 ). By contrast, intracellular Mn²⁺ levels are unaffected by overexpression of CCC1, a high-copy suppressor of the pmr1 Mn²⁺ defect (LAPINSKAS et al. 1996 ). Because our results suggested that Cdc1 and Cos16 affected the intracellular distribution of Mn²⁺, we examined if mutations in either CDC1 or COS16 altered intracellular Mn²⁺ levels as judged by atomic absorption spectroscopy. The Mn²⁺ content of cdc1-1(Ts) (38.9 ± 8.3 pmols/A₆₀₀) and cos16::HIS3 (41.4 ± 12.3 pmols/A₆₀₀) mutants were similar to the Mn²⁺ content of the isogenic wild-type (41.2 ± 7.0 pmols/A₆₀₀) strain, suggesting that depletion of neither Cdc1 nor Cos16 had a significant effect on whole-cell Mn²⁺ content. Cellular Mn²⁺ levels were also unaffected by CDC1 overexpression (data not shown). Thus, both Cdc1 and Cos16 seem to affect the intracellular distribution of Mn²⁺ without changing the overall cellular content.

Cdc1 is dispensable in cos16 mutants:
Because Cos16 seemed to be involved in vacuolar Mn²⁺ homeostasis, we determined if a cdc1 cos16 double mutant could grow in the presence of exogenous Mn²⁺. Isogenic diploids (cdc1::HIS3/CDC1 COS16/COS16 and cdc1::HIS3/CDC1 cos16::LEU2/COS16) were sporulated and dissected onto YEPD medium agar with, or without, Mn²⁺ supplement. Two viable His^- (CDC1⁺) segregants were recovered from each tetrad of the cdc1::HIS3/CDC1 COS16/COS16 diploid (Figure 7), and the 2:0 segregation was not affected by Mn²⁺ (data not shown). By contrast, tetrads from the cdc1::HIS3/CDC1 cos16::LEU2/COS16 diploid yielded 60% of the expected cdc1::HIS3 cos16::LEU2 segregants (Figure 7). Thus, Cdc1 is not essential in strains lacking Cos16 function. Moreover, >90% of the expected cdc1::HIS3 cos16::LEU2 progeny formed healthy colonies when supplemented with 2 mM or 4 mM Mn²⁺ (Figure 7). Thus, Mn²⁺ augments the ability of the cos16 deletion to bypass Cdc1 function.

View larger version (95K): In this window In a new window Download PPT slide   Figure 7. —Cdc1 is dispensable in a cos16 mutant. Heterozygous cdc1::HIS3/C DC1 COS16/COS16 (FY 1) or cdc1::HIS3/C DC1 cos16::LEU2/COS16 (FY 1 cos16::LEU2) diploids were sporulated and dissected onto rich yeast medium with, or without, Mn2+ supplement (see materials and methods). Tiny (or small) colonies were cdc1::HIS3 cos16::LEU2 as determined by their growth on minimal medium agar lacking histidine and leucine. (a, b, c, and d refer to individual spores of a tedrad.)

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Previous studies implicated Cdc1 in the regulation of intracellular Mn²⁺ (LOUKIN and KUNG 1995 ; PAIDHUNGAT and GARRETT 1998 ). That hypothesis was based on several observations, including: (1) Mn²⁺ supplement partially alleviated the conditional cdc1(Ts) growth defect; and (2) CDC1 overexpression ameliorated the EGTA sensitivity of two, unrelated Mn²⁺-homeostasis mutants. We show here that, in certain genetic backgrounds, Mn²⁺ supplement completely bypasses the essential Cdc1 requirement. Indeed, a cdc1 cos16 double mutant exhibits robust growth when the medium is supplemented with Mn²⁺ (Figure 7). Because COS16 is also involved in intracellular Mn²⁺ homeostasis, these results are best explained by a model in which Mn²⁺ regulation is the single, essential function of Cdc1. These results also rule out the possibility that the Mn²⁺ dependence of cdc1 strains reflects the fact that Cdc1 is a Mn²⁺-dependent enzyme (SUPEK et al. 1996 ; PAIDHUNGAT and GARRETT 1998 ).

Although the mechanism by which Cdc1 affects intracellular Mn²⁺ levels is not clear, the evidence favors a model in which Cdc1 functions to maintain cytosolic Mn²⁺. Mn²⁺ depletion from the Golgi elicits a protein glycosylation defect not observed in the cdc1(Ts) mutants, and the cdc1(Ts) growth defect is exacerbated, not alleviated, by alterations that increase Mn²⁺ flux from the cytosol to the Golgi (PAIDHUNGAT and GARRETT 1998 ). Thus, we suggested that the cytosol, rather than the Golgi apparatus, was the more likely site of cdc1(Ts) Mn²⁺ depletion (PAIDHUNGAT and GARRETT 1998 ). Another major intracellular Mn²⁺ store is the vacuole (OKOROKOV et al. 1977 ). Although the vacuole is known to play a role in Mn²⁺ detoxification, nothing is known about the cellular function(s) of the vacuolar Mn²⁺ pool. However, mutations that inhibit vacuolar biogenesis (and function) at several distinct steps suppress, rather than exacerbate, the cdc1(Ts) growth defect. Suppression of conditional growth probably results from the block in vacuolar Mn²⁺ sequestration because the same vacuolar biogenesis mutations also relieve the EGTA-sensitive phenotype of the cdc1(Ts) mutants (data not shown). These results eliminate the vacuole as the essential, Cdc1-dependent Mn²⁺ store and suggest that the vacuolar biogenesis mutations suppress the cdc1(Ts) growth defect by attenuating the depletion of cytosolic Mn²⁺.

Lesions in vacuole biogenesis (VPS) genes affect many vacuolar functions, including proteolysis, Ca²⁺ accumulation, proton uptake and Mn²⁺ sequestration. However, the conditional cdc1(Ts) growth defect is suppressed only by mutations that affect the last of these processes. In particular, the cdc1(Ts) growth defect is suppressed by the inactivation of a protein, Cos16, that appears to be involved in Mn²⁺ sequestration into the vacuole. Several lines of evidence support this assertion. First, Cos16 is an integral membrane protein of the vacuole, as judged by the subcellular localization pattern of a functional Cos16-LacZ fusion protein (Figure 5). This observation is consistent with the presence of eight putative transmembrane domains within the predicted coding region. Thus, Cos16 is appropriately positioned to affect transport between the cytosol and vacuole. Second, a cos16 deletion suppresses the chelator sensitivity of a strain (smf1) compromised for Mn²⁺ uptake into the cytosol (Figure 6C), suggesting that loss of Cos16 function compensates for the low influx into the cytosol. cos16 mutations also relieve the chelator sensitivity of a mutant (pmr1) that can be efficiently suppressed by genetic manipulations that raise cytosolic Mn²⁺ (Figure 6B; PAIDHUNGAT and GARRETT 1998 ). Finally, Cos16 inactivation exacerbates the Mn²⁺ sensitivity of both wild-type (data not shown) and pmr1 mutant strains (Figure 6A), presumably by aggravating the accumulation of cytosolic Mn²⁺. Thus, Cos16 appears to antagonize the cdc1(Ts) growth defect by sequestering cytosolic Mn²⁺ into the vacuole.

If COS16 and CDC1 are involved in the homeostasis of intracellular Mn²⁺, why do intracellular Mn²⁺ levels not vary with changes in Cos16 or Cdc1 function? A previous report (OKOROKOV et al. 1977 ) suggested that most intracellular Mn²⁺ is found in the vacuole. According to that report, a defect in cytosolic Mn²⁺ retention, as predicted for cdc1(Ts) mutants, might not affect total intracellular Mn²⁺ content. However, it is harder to reconcile the fact that the cos16 deletion did not alter the level of intracellular Mn²⁺. We can think of several possible explanations for this apparent paradox. First, Cos16 might either be involved in the regulation of other ions, or affect an organelle that might copurify with the vacuole. Although we have not formally ruled out these possibilities, they do not easily account for the observation that cos16 mutations affect the growth of Mn²⁺ homeostasis mutants. Alternatively, Cos16 inactivation might block cytosolic to vacuole Mn²⁺ transfer without having a measurable effect on vacuolar Mn²⁺ content.

Finally, along with COS16 and the vacuolar biogenesis genes characterized in this study, we identified several other suppressors of the cdc1(Ts) growth defect. One suppressor, cos13, falls within the same phenotypic and epistasis group as cos16 and might, therefore, identify a component of a complex or pathway involved in vacuolar Mn²⁺ homeostasis. By contrast, the remaining recessive suppressors (cos1, cos3, cos14, cos2, cos7, and cos10) confer phenotypes that are not obviously related to Mn²⁺ regulation. Thus, the genes corresponding to these suppressors might specify receptors of the Cdc1 (and Mn²⁺?)-dependent growth process. Future studies of these genes might help elucidate the role of Cdc1 in Mn²⁺ homeostasis and growth.

	FOOTNOTES

¹ Present address: Department of Biochemistry, University of Connecticut Health Center, Farmington, CT 06030.

	ACKNOWLEDGMENTS

We thank M. CARDENAS, V. CULOTTA, L. DAVIS, K. DOLINSKY, J. HEITMAN, M. HILLER, E. W. JONES, R. KAHN and M. D. ROSE for yeast strains, plasmids and antisera. M.P. also thanks T. GRAF and J. JOHNSON for help and instruction with atomic absorption spectroscopy.

Manuscript received August 14, 1997; Accepted for publication December 3, 1997.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

BANTA, L. M., J. S. ROBINSON, D. J. KLIONSKY, and S. D. EMR, 1988  Organelle assembly in yeast: characterization of yeast mutants defective in vacuolar biogenesis and protein sorting. J. Cell Biol. 107:1369-1383[Abstract/Free Full Text].

BANTA, L. M., T. A. VIDA, P. K. HERMAN, and S. D. EMR, 1990  Characterization of yeast Vps33p, a protein required for vacuolar protein sorting and vacuole biogenesis. Mol. Cell. Biol. 10:4638-4649[Abstract/Free Full Text].

CARDENAS, M. E. and J. HEITMAN, 1995  FKBP12-rapamycin target TOR2 is a vacuolar protein with an associated phosphatidylinositol-4 kinase activity. EMBO J. 14:5892-5907[Medline].

CASADABAN, M. J., A. MARTINEZ-ARIAS, S. K. SHAPIRO, and J. CHOU, 1983  ß-Galactosidase gene fusions for analyzing gene expression in Escherichia coli and yeast. Methods Enzymol. 100:293-308[Medline].

COWLES, C. R., S. D. EMR, and B. F. HORAZDOVSKY, 1994  Mutations in the VPS45 gene, a SEC1 homologue, result in vacuolar protein sorting defects and accumulation of membrane vesicles. J. Cell Sci. 107:3449-3459[Abstract].

EIDE, D. and M. L. GUERINOT, 1997  Metal ion uptake in eukaryotes. ASM News 63:199-205.

GARRETT, S. and J. BROACH, 1989  Loss of Ras activity in Saccharomyces cerevisiae is suppressed by disruption of a new kinase gene, YAK1, whose product may act downstream of the cAMP-dependent protein kinase. Genes Dev. 3:1336-1348[Abstract/Free Full Text].

HARTWELL, L., J. CULOTTI, and B. J. REID, 1970  Genetic control of the cell division cycle in yeast, I. Detection of mutants. Proc. Natl. Acad. Sci. USA 66:352-359[Abstract/Free Full Text].

HORAZDOVSKY, B. F. and S. D. EMR, 1993  The VPS16 gene product associates with a sedimentable protein complex and is essential for vacuolar proteins sorting in yeast. J. Biol. Chem. 268:4953-4962[Abstract/Free Full Text].

IIDA, H., S. SAKAGUCHI, Y. YAGAWA, and Y. ANRAKU, 1990  Cell cycle control by Ca²⁺ in Saccharomyces cerevisiae. J. Biol. Chem. 265:21216-21222[Abstract/Free Full Text].

JONES, E. W., G. S. ZUBENKO, and R. R. PARKER, 1982  PEP4 gene function is required for expression of several vacuolar hydrolases in Saccharomyces cerevisiae. Genetics 102:665-677[Abstract/Free Full Text].

KYTE, J. and R. F. DOOLITTLE, 1982  A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157:105-132[Medline].

LAPINSKAS, P. J., K. W. CUNNINGHAM, X. F. LIU, G. R. FINK, and V. C. CULOTTA, 1995  Mutations in PMR1 suppress oxidative damage in yeast cells lacking superoxide dismutase. Mol. Cell. Biol. 15:1382-1388[Abstract].

LAPINSKAS, P. J., S. LIN, and V. C. CULOTTA, 1996  The role of the Saccharomyces cerevisiae CCC1 gene in the homeostasis of manganese ions. Mol. Microbiol. 21:519-528[Medline].

LIN, S.-J. and V. C. CULOTTA, 1996  Suppression of oxidative damage by Saccharomyces cerevisiae ATX2, which encodes a manganese-trafficking protein that localizes to Golgi-like vesicles. Mol. Cell. Biol. 16:6303-6312[Abstract].

LOUKIN, S. and C. KUNG, 1995  Manganese effectively supports yeast cell-cycle progression in place of calcium. J. Cell Biol. 131:1025-1037[Abstract/Free Full Text].

MANOLSON, M. F., D. PROTEAU, R. A. PRESTON, A. STENBIT, and B. T. ROBERTS et al., 1992  The VPH1 gene encodes a 95-kDA integral membrane polypeptide required for in vivo assembly and activity of the yeast vacular H⁺-ATPase. J. Biol. Chem. 267:14294-14303[Abstract/Free Full Text].

MUNN, A. L. and H. RIEZMAN, 1994  Endocytosis is required for the growth of vacuolar H⁺-ATPase-defective yeast: identification of six new END genes. J. Cell Biol. 127:373-386[Abstract/Free Full Text].

MYERS, A. M., A. TZAGOLOFF, D. M. KINNEY, and C. J. LUSTY, 1986  Yeast shuttle and integrative vectors with multiple cloning sites suitable for construction of lacZ fusions. Gene 45:299-310[Medline].

NAGASU, T., Y.-I. SHIMMA, Y. NAKANISHI, J. KUROMITSU, and K. IWAMA et al., 1992  Isolation of new temperature-sensitive mutants of Saccharomyces cerevisiae deficient in mannose outer chain elongation. Yeast 8:535-547[Medline].

OHSUMI, Y. and Y. ANRAKU, 1981  Active transport of basic amino acids driven by a proton motive force in vacuolar membrane vesicles of Saccharomyces cerevisiae. J. Biol. Chem. 256:2079-2082[Abstract/Free Full Text].

OKOROKOV, L. A., L. P. LICHKO, V. M. KADOMTSEVA, V. P. KHOLODENKO, and V. T. TITOVSKY et al., 1977  Energy-dependent transport of manganese into yeast cells and distribution of accumulated ions. Eur. J. Biochem. 75:373-377[Medline].

PAIDHUNGAT, M. and S. GARRETT, 1998  Cdc1 is required for growth and Mn²⁺ regulation in Saccharomyces cerevisiae. Genetics 148:1777-1786[Abstract/Free Full Text].

PRESTON, R. A., P. S. REINAGEL, and E. W. JONES, 1992  Genes required for vacuolar acidity in Saccaromyces cerevisiae. Genetics 131:551-558[Abstract].

PRESTON, R. A., M. F. MANOLSON, K. BECHERER, E. WEIDENHAMMER, and D. KIRKPATRICK et al., 1991  Isolation and characterization of PEP3, a gene required for vacuolar biogenesis in Saccharomyces cerevisiae. Mol. Cell. Biol. 11:5801-5812[Abstract/Free Full Text].

RAYMOND, C. K., I. HOWALD-STEVENSON, C. A. VATER, and T. H. STEVENS, 1992  Morphological classification of the yeast vacuolar protein sorting mutants: evidence for a prevacuolar compartment in class E vps mutants. Mol. Biol. Cell 3:1389-1402[Abstract].

RILES, L., J. E. DUTCHIK, A. BAKTHA, B. K. MCCAULEY, and E. C. THAYER et al., 1993  Physical maps of the six smallest chromosomes of Saccharomyces cerevisiae at a resolution of 2.6 kilobase pairs. Genetics 134:81-150[Abstract].

ROBERTS, C. J., C. K. RAYMOND, C. T. YAMASHIRO and T. H. STEVENS, 1991 Methods for studying the yeast vacuole, in Guide to Yeast Genetics and Molecular Biology. Academic Press, San Diego.

ROBINSON, J. S., T. R. GRAHAM, and S. D. EMR, 1991  A putative zinc finger protein, Saccharomyces cerevisiae Vps18p, affects late Golgi functions required for vacuolar protein sorting and efficient -factor prohormone maturation. Mol. Cell. Biol. 12:5813-5824.

ROSE, M. D., P. NOVICK, J. H. THOMAS, D. BOTSTEIN, and G. FINK, 1987  A Saccharomyces cerevisiae genomic plasmid bank based on a centromere-containing shuttle vector. Gene 60:237-243[Medline].

SIKORSKI, R. S. and P. HIETER, 1989  A system of shuttle vectors and yeast host strains designed for efficient manipulation of DNA in Saccharomyces cerevisiae. Genetics 122:19-27[Abstract/Free Full Text].

STACK, J. H. and S. D. EMR, 1993  Genetic and biochemical studies of protein sorting to the yeast vacuole. Curr. Opin. Cell Biol. 5:641-646[Medline].

STEARNS, T., M. C. WILLINGHAM, D. BOTSTEIN, and R. A. KAHN, 1990  ADP-ribosylation factor is functionally and physically associated with the Golgi complex. Proc. Natl. Acad. Sci. USA 87:1238-1242[Abstract/Free Full Text].

SUPEK, F., L. SUPEKOVA, H. NELSON, and N. NELSON, 1996  A yeast manganese transporter related to the macrophage protein involved in conferring resistance to mycobacteria. Proc. Natl. Acad. Sci. USA 93:5105-5110[Abstract/Free Full Text].

WEDLER, F. C., 1994 Biochemical and nutritional role of manganese: an overview, pp. 1–38 in Manganese in Health and Disease, edited by D. J. KLIMIS-TAVANTZIS. CRC Press, Inc., Boca Raton, FL.

WEISMAN, L. S. and W. WICKNER, 1992  Molecular characterization of VAC1, a gene required for vacuole inheritance and vacuole protein sorting. J. Biol. Chem. 267:618-623[Abstract/Free Full Text].

WOODCOCK, D. M., P. J. CROWTHER, J. DOHERTY, S. JEFFERSON, and E. DECRUZ et al., 1989  Quantitative evaluation of Escherichia coli host strains for tolerance to cytosine methylation in plasmid and phage recombinants. Nucleic Acids Res. 17:3469-3478[Abstract/Free Full Text].

WOOLFORD, C. A., C. K. DIXON, M. F. MANOLSON, R. WRIGHT, and E. W. JONES, 1990  Isolation and characterization of PEP5, a gene essential for vacuolar biogenesis in Saccharomyces cerevisiae. Genetics 125:739-752[Abstract].

YOUATT, J. and I. MCKINNON, 1993  Manganese (Mn²⁺) reverses the inhibition of fungal growth by EGTA. Microbios 74:181-196.

This article has been cited by other articles:

				E. Losev, E. Papanikou, O. W. Rossanese, and B. S. Glick Cdc1p Is an Endoplasmic Reticulum-Localized Putative Lipid Phosphatase That Affects Golgi Inheritance and Actin Polarization by Activating Ca2+ Signaling Mol. Cell. Biol., May 15, 2008; 28(10): 3336 - 3343. [Abstract] [Full Text] [PDF]

				S. P. Bai, L. Lu, X. G. Luo, and B. Liu Cloning, Sequencing, Characterization, and Expressions of Divalent Metal Transporter One in the Small Intestine of Broilers Poult. Sci., April 1, 2008; 87(4): 768 - 776. [Abstract] [Full Text] [PDF]

				M. Fujita, M. Umemura, T. Yoko-o, and Y. Jigami PER1 Is Required for GPI-Phospholipase A2 Activity and Involved in Lipid Remodeling of GPI-anchored Proteins Mol. Biol. Cell, December 1, 2006; 17(12): 5253 - 5264. [Abstract] [Full Text] [PDF]

				V. C. Culotta, M. Yang, and M. D. Hall Manganese Transport and Trafficking: Lessons Learned from Saccharomyces cerevisiae Eukaryot. Cell, July 1, 2005; 4(7): 1159 - 1165. [Full Text] [PDF]

				L. Eguez, Y.-S. Chung, A. Kuchibhatla, M. Paidhungat, and S. Garrett Yeast Mn2+ Transporter, Smf1p, Is Regulated by Ubiquitin-Dependent Vacuolar Protein Sorting Genetics, May 1, 2004; 167(1): 107 - 117. [Abstract] [Full Text] [PDF]

				E. S. Seeley, M. Kato, N. Margolis, W. Wickner, and G. Eitzen Genomic Analysis of Homotypic Vacuole Fusion Mol. Biol. Cell, March 1, 2002; 13(3): 782 - 794. [Abstract] [Full Text] [PDF]

				E. E-C. Luk and V. C. Culotta Manganese Superoxide Dismutase in Saccharomyces cerevisiae Acquires Its Metal Co-factor through a Pathway Involving the Nramp Metal Transporter, Smf2p J. Biol. Chem., December 7, 2001; 276(50): 47556 - 47562. [Abstract] [Full Text] [PDF]

				L. Ni and M. Snyder A Genomic Study of the Bipolar Bud Site Selection Pattern in Saccharomyces cerevisiae Mol. Biol. Cell, July 1, 2001; 12(7): 2147 - 2170. [Abstract] [Full Text] [PDF]

				O. W. Rossanese, C. A. Reinke, B. J. Bevis, A. T. Hammond, I. B. Sears, J. O'Connor, and B. S. Glick A Role for Actin, Cdc1p, and Myo2p in the Inheritance of Late Golgi Elements in Saccharomyces cerevisiae J. Cell Biol., March 26, 2001; 153(1): 47 - 62. [Abstract] [Full Text] [PDF]

				K. B. Lengeler, R. C. Davidson, C. D'souza, T. Harashima, W.-C. Shen, P. Wang, X. Pan, M. Waugh, and J. Heitman Signal Transduction Cascades Regulating Fungal Development and Virulence Microbiol. Mol. Biol. Rev., December 1, 2000; 64(4): 746 - 785. [Abstract] [Full Text] [PDF]

				M. Paidhungat and S. Garrett Cdc1 Is Required for Growth and Mn2+ Regulation in Saccharomyces cerevisiae Genetics, April 1, 1998; 148(4): 1777 - 1786. [Abstract] [Full Text] [PDF]