The Golgi complex is involved in the post-translational modification of glycoproteins and in the sorting of these proteins to their correct destination. Each of the individual Golgi cisternae is biochemically and functionally distinct, differing in both protein and lipid composition. In the case of the glycosyltransferases that mediate glycoprotein modifications, immunocytological and biochemical studies have clearly demonstrated that these enzymes are compartmentalized within particular cisternae. Successful glycan synthesis is dependent upon the compartmentalization and regulation of the glycosyltransferases that participate in these stepwise reactions.

Following the initial glycosylation steps in the endoplasmic reticulum (ER), ()yeast glycans are elongated by Golgi-localized mannosyltransferases to form glycoproteins with extended outer chains of 50 or more mannose units (for review, see (2) and (3) ) The outer chain, consisting of an 1,6-linked mannose backbone, is normally highly branched with 1,2- and 1,3-linked mannoses(2, 3, 4) . Several groups of yeast mutants with defects in glycosylation have been isolated, most notably the sec (secretion), alg (asparagine-linked glycosylation), and mnn (mannan) mutants. The sec mutants are conditional mutants with defects in transport steps through out the secretory pathway(5) . The alg mutants are affected primarily in the synthesis of the core oligosaccharide that is added in the ER(6) . The mnn mutants are blocked at various stages of outer-chain carbohydrate elongation that occur in the Golgi complex(1, 4, 7) . Many of the mnn mutants do not appear to contain lesions in genes that encode glycosyltransferases(1, 4, 7, 8, 9) . Rather, they likely affect other cellular functions associated with the secretory pathway that affect glycosylation in the Golgi(9) .

As a first step toward identifying factors that participate in the correct localization of resident Golgi proteins, and therefore contribute to normal Golgi biogenesis, our efforts have been directed toward the characterization of mutants that have defects in Golgi-specific functions, specifically in glycosylation. The underlying prediction was that these glycosylation mutants would fall into two general classes: those containing defects in genes encoding the glycosyltransferases themselves, and those encoding proteins that regulate the activity or localization of these enzymes. In this report, we describe one glycosylation mutant that falls into the latter class. This mutant is allelic to vrg4(1) (also known as van2; see (10) and (11) ), a previously identified vanadate-resistant glycosylation mutant. Here, we present a phenotypic analysis of the vrg4 mutant and a molecular analysis of the VRG4 gene.

MATERIALS AND METHODS

Media and Yeast Strains

All yeast strains used in this study are listed in Table 1. Yeast transformations were performed using the lithium acetate protocol, as described previously(13) .

Isolation of Vanadate-resistant Mutants

Cloning and Sequencing the Wild Type VRG4 Gene

A 2.1-kb EcoRI/HindIII fragment capable of complementing the hygromycin B sensitivity of vrg4 was sequenced by the dideoxy method (14) generating a nested deletion series using the ExoIII/ExoVII method(15) . Both DNA strands were sequenced. DNA and predicted protein sequence comparisons against data bases were made using the BLAST algorithm (16) and analyzed using the GCG programs.

Plasmid Constructions

The disruption plasmid pG5::LEU was constructed by inserting a SmaI/SalI fragment (blunt-ended with Klenow) containing the LEU2 gene into the unique HpaI site that lies within the VRG4 gene.

The integrative plasmid pG5i was constructed by cloning the HindIII/EcoRI fragment containing the entire VRG4 gene into the URA3-containing pRS306. The plasmid was linearized at a unique HpaI site in the VRG4 gene and transformed into strain NDY5.

Western Immunoblotting and Immunoprecipitations

Analysis of CPY by immunoprecipitation was carried out as described (21) , except that cells were labeled with 200 µCi of [S]methionine and cysteine (ExpreSS, DuPont NEN), and chased with the addition of a 10-fold chase solution of 50 mM methionine and 10 mM cysteine to a final concentration of 5 mM methionine and 1 mM cysteine.

In Situ Invertase Gel Assay

DiOC Staining of Yeast Cells

Electron Microscopy

RESULTS

Isolation of the vrg4 Mutant

To select mutants with defects in glycosylation, yeast were grown on medium supplemented with 10 mM sodium vanadate. Ballou et al. have found that resistance to sodium vanadate enriches for mutants defective in Golgi-specific glycosylation(1) . They identified five complementation groups, all of which are defective in glycoprotein modifications that occur in the Golgi complex. Three of these are allelic to mnn8, mnn9, and mnn10, which have well characterized defects in glycoprotein outer chain structure(4, 8) .

Spontaneous mutants that were resistant to 10 mM sodium vanadate arose at a frequency of approximately 10. As expected from previous studies(1, 24) , all exhibited varying degrees of sensitivity to the aminoglycoside, hygromycin B. Mutants were isolated from both MATa and MAT strains. To assay for dominance of the vanadate resistance/hygromycin B sensitivity phenotype, the mutants were crossed to the parental strain of the appropriate mating type. By this criterion, all of the mutants we isolated were recessive. Mutants were also tested for temperature-sensitive growth and none were found to be temperature-sensitive.

Complementation analysis was performed by crossing mutants of opposite mating types and testing for vanadate resistance and hygromycin B sensitivity. This analysis indicated that the 28 isolates we obtained represent a minimum of eight different genes. Complementation analyses with a partial set of previously identified vanadate-resistant glycosylation mutants were performed. Among the eight vanadate-resistant mutants we isolated, we found mutants allelic to vrg4(1) , vrg7(5) (also known as van1; see (10) ), and mnn9(7, 25) . mnn9 and van1 are well characterized mutants, known to be defective in outer chain glycoprotein modifications that occur in the Golgi complex(1, 4) . Little else is known about vrg4. Since we have not exhausted our search, it is unclear if other previously identified vanadate-resistant mutants (10, 26) are represented in the collection of mutants we isolated.

The glycosylation pattern of the secreted, periplasmic form of invertase was examined in the mutants to determine if all had defects in glycosylation (Fig. 1). Invertase exists in two states, a cytoplasmic, non-glycosylated form, and a secreted form. The secreted form is a highly glycosylated protein whose rate of migration on native gels reflects the size and number of N-linked oligosaccharide chains it contains. Using an in situ gel assay(4) , the electrophoretic mobility of invertase was visualized by an activity stain that monitors sucrose hydrolysis by invertase. A glycosylation defect can be detected by comparing the average rate of migration of invertase in extracts prepared from mutant and wild type cells.

Figure 1: Analysis of invertase glycosylation in vanadate-resistant mutants. Protein was isolated from wild type (lane 4) or vanadate-resistant isolates (lanes 1-3 and 5) and electrophoresed on 5% nondenaturing polyacrylamide gels to detect native invertase using an activity stain as described under ``Materials and Methods.'' The position of glycosylated secreted invertase is indicated by brackets. The arrow denotes the position of nonglycosylated, cytoplasmic invertase. Lane 1 contains extracts from strain NDY10, which contains an uncharacterized glycosylation defect. Strain NDY4 (lane 2) contains the mnn9 mutation, strain NDY7.4 contains the van1 mutation (lane 3), and strain NDY5 contains the vrg4 mutation (lane 5).

All of the mutants expressed the periplasmic form of invertase that migrated with an increased electrophoretic mobility compared to wild type. A subset of these are shown in Fig. 1. All of the mutants also expressed the cytoplasmic, nonglycosylated form of invertase that comigrated with wild type cytoplasmic invertase (see arrow in Fig. 1). Since both forms of invertase are derived from the same gene, we infer that the change in mobility of the secreted invertase in each of these mutants is due to a decrease in carbohydrate modifications rather than differences in protein structure. The electrophoretic resolution of this procedure is insufficient to detect differences in individual carbohydrate modifications. Despite this, several general classes of defects were apparent. The most severe glycosylation phenotype is exemplified by the mutation in isolate NDY5, which carries the vrg4 mutation (Fig. 1, lane 5). Invertase in this mutant migrates with a mobility similar to that of the mnn9 mutant (Fig. 1, lane 2), that lacks an outer chain and contains an oligosaccharide consisting of 10-14 mannoses(4) . The mobility of invertase was somewhat variable, and in some experiments we observed a small fraction of invertase in the vrg4-2 mutant that co-migrated with the fully glycosylated form (for example see Fig. 6B, lane 3). The source of this variability is unknown. However, in all cases, the bulk of invertase that is secreted in this mutant migrates with an increased mobility. This result suggests that, like mnn9, vrg4 severely impairs elongation of outer chain carbohydrates.

Figure 6: The cloned VRG4 gene rescues both the glycosylation and ER retention defects of vrg4. Panel A, Western immunoblot of proteins in culture supernatants from wild type (lane 2), erd1 (lane 1), vrg4 mutant (lane 4), and vrg4 mutant harboring a plasmid bearing the VRG4 gene (lane 3). The mutant vrg4 harboring a plasmid bearing the cloned gene no longer secretes BiP into the culture media. Panel B, the vrg4 mutant harboring a plasmid bearing the wild type VRG4 gene recovers the ability to glycosylate invertase normally. Shown is a native in situ invertase assay (performed as in Fig. 1) comparing the glycosylation state of invertase from wild type (lane 1), vrg4 mutant (lane 3), and mutant harboring the complementing clone (lane 2).

The vrg4 Mutant Accumulates the Golgi Intermediate of Carboxypeptidase Y

Protein extracts from mutant and parental cells were subjected to immunoblot analyses using antiserum raised against CPY. Only one mutant, vrg4, showed an accumulation of the ER and Golgi form of CPY (Fig. 2A, lane 6). Most of the other mutants accumulated mature CPY that was indistinguishable from that of the parental, wild type cells. The two exceptions were mutants that accumulated forms of CPY that migrated even faster than mature CPY (Fig. 2A, compare lanes 1 and 2 with lane 7). The increased mobility in these two glycosylation mutants was due to a reduction in the number or size of core oligosaccharides that are added in the ER. ()This increased electrophoretic mobility of CPY is diagnostic of early glycosylation mutants with lesions in the ER(28) . This increased mobility is not observed in vrg4, suggesting that the glycosylation defect caused by the vrg4 mutation is not due to an ER glycosylation defect. The accumulation of the ER and Golgi forms of CPY in the vrg4 mutant was not a consequence of its glycosylation defect, since none of the other glycosylation mutants accumulated these forms of CPY (Fig. 2A, lanes 1-5).

Figure 2: The vrg4 mutant accumulates the ER and Golgi intermediates of carboxypeptidase Y. In panels A-C, the ER, Golgi and mature forms of CPY are denoted by p1, p2, and m, respectively. Panel A, Western blot analysis of CPY in vanadate-resistant mutants (lanes 1-6) and wild type cells (lane 7). Protein was prepared and analyzed as described under ``Materials and Methods.'' Strain NDY13.4 (lane 1) contains an uncharacterized alg-like mutation; strain NDY1.4 (lane 2) contains a mutation in the OST4 gene; see legend in Fig. 1for strain description of other mutants in lanes 3-6. Note that among these mutants, only vrg4 (lane 6) accumulates the Golgi and ER form. All the other mutants except NDY13.4 (lane 1) and NDY1.4 (lane 2) are indistinguishable from wild type (lane 7). Panel B, immunoprecipitation of CPY from mutant and wild type cells. Proteins in vrg4 mutant (lane 1), NDY10 (lane 2), and wild type cells (lane 3) were labeled with [S]methionine for 10 min and chased for 15 min. CPY was immunoprecipitated and subjected to SDS-PAGE, as described under ``Materials and Methods.'' Panel C, kinetic analysis of CPY in vrg4 mutant cells. Proteins in vrg4 mutant were labeled with [S]methionine for 20 min and chased for 0 min (lane 1), 30 min (lane 2), and 60 min (lane 3) with cold methionine/cysteine. CPY was immunoprecipitated and subjected to SDS-PAGE, as described under ``Materials and Methods.''

To further confirm the presence of the ER and Golgi forms of CPY in the vrg4 mutant, immunoprecipitation studies using anti-CPY antibodies were performed on metabolically labeled cells. CPY was immunoprecipitated from cells that were pulse-labeled with [S]methionine and chased for variable lengths of time. In the vrg4 mutant there was an accumulation of both the ER and Golgi form after a 15-min chase, with a detectable increase in the Golgi form (Fig. 2B, lane 1). In the parental strain, after 15 min, all of the CPY had chased into the mature form (Fig. 2B, lane 3). In the vrg4 mutant, most of the ER and Golgi CPY intermediates eventually chased into the mature form, but only after 60 min (Fig. 2C, lane 3). Some experimental variability in the kinetics of CPY transport to the vacuole in the vrg4-2 mutant was observed in pulse-chase experiments. This variability may be a result of differences in the growth stage of mutant cells, which grow poorly in the synthetic medium used to deplete the intracellular pools of methionine. However, when analyzed at steady state, the accumulation of the p1 and p2 forms of CPY was always seen in the vrg4 mutant. No secreted CPY was ever observed in the medium (data not shown), suggesting that the effect of this mutation is not due to a general perturbation of the secretory pathway. Since most of the CPY was correctly delivered to the vacuole, these results suggest that transport through primarily the Golgi is delayed, but not blocked in this mutant.

The VRG4 Gene Is Required for the Retention of BiP in the ER

Proteins in culture supernatants from mutant and wild type cells were precipitated and assayed for the presence of the resident ER protein, BiP, by Western immunoblot analyses. Wild type cells do not secrete significant levels of BiP into the culture supernatant, as BiP is efficiently retained in the ER (Fig. 3, lane 5). By this assay, we found that among the different vanadate-resistant mutants, only vrg4 had an ER retention defect, and secreted BiP at levels comparable to the ER retention mutant, erd1 (ER retention defect) (30, 31) (Fig. 3, compare lanes 4 and 6). The intracellular level of BiP is induced as part of the unfolded protein response. To examine if the secretion of BiP in vrg4 was an indirect result of increased BiP synthesis, which in turn saturates HDEL retention, we compared the intracellular level of BiP in the vrg4 mutant to that of wild type cells by immunoblot analysis of protein extracts. An equal number of mutant and wild type cells were washed and intracellular proteins were extracted, precipitated, and assayed for the presence of the resident ER protein, BiP, by Western immunoblot analyses. The result of this experiment demonstrated that the level of intracellular BiP in vrg4 and wild type cells was indistinguishable (Fig. 3B, compare lanes 1 and 2). None of the other vanadate-resistant glycosylation mutants secreted more BiP than did wild type cells. However, several mutants did contain markedly increased intracellular levels of BiP, presumably due to increased levels of misfolded proteins as a result of glycosylation defects (data not shown). These results demonstrate that the secretion of BiP in vrg4 was not an indirect effect of elevated levels of misfolded proteins due to inappropriate glycosylation. Rather, these results suggest that the vrg4 mutation affects the receptor-mediated retrieval of BiP from the early Golgi.

Figure 3: VRG4 is required for the retention of BiP in the ER. Panel A, Western immunoblot of BiP in culture supernatants from mutant and wild type cells. Equivalent amounts of protein in media from four different vanadate-resistant mutants were analyzed as described under ``Materials and Methods'' using antibodies against the C-terminal HDEL peptide. These are compared with supernatants derived from erd1 (lane 4) and parental strains (lane 5). Note that of the vanadate-resistant mutants, only vrg4 (lane 6) fails to retain BiP intracellularly. Panel B, Western immunoblot of intracellular BiP in wild type and vrg4 mutant cells. Cultures (2 A units) were spun down to separate cells from culture supernatants and total intracellular protein extracted from washed, cell pellets as described under ``Materials and Methods.'' Equivalent amounts of protein from each sample were subjected to SDS-PAGE and immunoblotted with anti-HDEL antibodies.

The VRG4 Gene Is Required for the Maintenance of Normal Membrane Morphology

A comparison of the staining pattern revealed striking differences in the endomembrane system of wild type and vrg4 mutant cells. Wild type cells showed a characteristic network of ER membranes just below the plasma membranes and peripheral to the nucleus (Fig. 4, panels B and D). This was in contrast to the highly vesiculated appearance of vrg4. While the mutant also stained membranes below the plasma membrane and peripheral to the nucleus, the most evident difference in mutant cells was a fragmented, vesiculated appearance of stained membranes (Fig. 4, panels A and C).

Figure 4: DiOC staining of wild type, vrg4, and mnn10 cells. Mutant and wild type cells (10 cells) were treated with 10 µg/ml of DiOC (as described under ``Materials and Methods'') and visualized by confocal microscopy. Shown are vrg4 (panels A and C), wild type (panels B and D), and mnn10 (panel E) cells. All cells were viewed with a 60 oil immersion lens, but in panels A and B, an additional 2.5-fold magnification was provided through the zoom function in the software. Bar represents 5 µm.

A feature of vrg4 that was usefully highlighted by DiOC staining was the heterogeneity of mutant cells. (Fig. 4, panel C). Wild type cells are of a uniform size and shape, while vrg4 cells are variable and formed large, clumped aggregates. The tendency to form aggregates was a general feature of the vanadate-resistant mutants we isolated and is a phenotype described previously(1) . The vesiculation of intracellular membrane in vrg4 was unlikely to be a consequence of this aggregation phenotype. While other glycosylation mutants also exhibited this phenotype, the staining patterns were very distinct from vrg4. This is exemplified by the staining pattern of mnn10, whose membranes had a more punctate-like appearance (Fig. 4, panel E).

We observed other phenotypes that were common to these vanadate-resistant mutants including osmotic sensitivity, reduced sporulation frequency, poor spore viability, reduced spheroplasting frequency, and 2-8-fold decreased growth rate (data not shown). Presumably, these were a consequence of cell wall defects due to defects in glycosylation. Because the entry or accumulation of DiOC in mutant cells may be influenced by these defects, the concentration of DiOC for both wild type and vrg4 cells was carefully titrated and staining patterns were examined at different dye concentrations. Since the vrg4 mutant grows with a decreased growth rate, we also examined the staining patterns at different growth stages. In all cases, we observed the same increased, fragmented membrane morphology as well as heterogeneity in vrg4 cells, which was never observed in wild type cells (data not shown).

The morphology of intracellular membrane structures in vrg4 was examined at higher resolution by thin-section electron microscopy. Cells were rapidly fixed in glutaraldehyde and membranes stained with potassium permanganate (Fig. 5). Again, a marked difference between vrg4 and isogenic wild type parental cells was observed. Wild type cells exhibit highly contrasted, intensely staining membranous structures. Mutant membranes lack this contrast and appeared to be devoid of membrane-bound organelles entirely (Fig. 5, compare panels A and B), although a faintly stained nucleus was always observed. This phenotype was seemingly different from that of DiOC staining, where the mutant membranes appear to be fragmented and highly vesiculated, but relatively abundant. Based upon these differences, it seems likely that, unlike wild type membranes, the mutant membranes were somehow altered in their permanganate staining properties. At higher magnifications, membranes and vesicles become more apparent, but clearly were distinct in their staining properties (Fig. 5, compare panels C and D). Unlike wild type cells, which typically contain a single vacuole, the vrg4 mutant had small fragmented vacuolar-like structures that contain a higher amount of electron dense material. From these analyses we conclude that the VRG4 gene is required for a normal endomembrane system.

Figure 5: Thin-section electron micrographs of vrg4 and wild type yeast cells. vrg4 mutant cells (panels B and D) or wild type (panels A and C) cells were stained with potassium permanganate and prepared as described under ``Materials and Methods.'' Bars represent 2 µm (panels A and B) and 500 nm (panels C and D).

Cloning and Analysis of the VRG4 Gene

Figure 9: VRG4 encodes a protein required for viability. Panel A is a schematic representations of the restriction map of the region surrounding the VRG4 gene, on chromosome XV. Panel B is a schematic diagram of the strategy used to create the vrg4::LEU2 disruption plasmid that was used to replace one copy of the chromosomal wild type VRG4 allele in a diploid. B, H, Hp, and E refer to BamHI, HindIII, HpaI, and EcoRI restriction sites. Panel C, tetrad analysis of diploid strains heterozygous for the VRG4 disrupted allele. Tetrads obtained from the sporulation were dissected on YPD plates, with four spores on one column. These were incubated for 3 days at 30 °C. Each column is labeled numerically.

To confirm that the cloned fragment contained the VRG4 locus, the EcoRI/HindIII fragment was cloned in an integrative plasmid (pRS306) that contains the selectable marker, URA3. The plasmid was linearized at a unique site within the VRG4 portion to allow homologous recombination at the vrg4 locus and used to transform vrg4 ura3 cells. Ura transformants were then crossed to a VRG4 ura3 strain. The resulting diploid was sporulated and tetrads dissected. This analysis demonstrated a 2:2 segregation pattern for Ura/Ura and a 4:0 pattern for hygromycin resistance/hygromycin sensitivity, indicating that the cloned fragment is tightly linked to VRG4 and most likely does contain the VRG4 locus (data not shown).

DNA sequence analysis of the 2.1-kb fragment revealed the presence of two open reading frames (Fig. 9A). Further analyses mapped the complementing activity to the larger open reading frame, within a 1.6-kb HindIII/EcoRV fragment. The nucleotide and predicted amino acid sequence of this region is shown in Fig. 7.

Figure 7: The nucleotide and predicted amino acid sequence of the VRG4 gene and its relation to other genes. Panel A, the five potential glycosylation sites, at amino acid positions 81, 119, 242, 246, and 249, are denoted by asterisks. Four potential membrane-spanning domains, comprised of at least 20 uncharged residues and flanked by charged residues are underlined (and correspond to the black bars in Fig. 8). Recently, this sequence was found to be the same as that of the VAN2 gene (GenBank(TM) accession no. U15599 (11) . Panel B, summary of percent amino acid identities and similarities shared between the yeast Vrg4 protein and other members of this family (gap alignment program used the Needleman and Wunsch algorithm on the GCG program). The accession numbers for those sequences in the data base are as follows: yeast homologue, U18796 (gene YER039c); Leishmania, U26175; rice 1, D24450; rice 2, D24744; Arabidopsis, T45513.

Figure 8: Hydropathy plot of the predicted Vrg4p. Hydrophobicity was calculated according to the method of Kyte and Doolitle(19) , using a window of 11 amino acids. The black bar represents the potential transmembrane domains, underlined in Fig. 7.

The VRG4 DNA sequence encodes a predicted protein of 36.9 kDa. There are five potential recognition sites for N-linked glycosylation (indicated by asterisks in Fig. 7A). Hydrophobicity analysis (33) (Fig. 8) suggests that the protein is hydrophobic, containing multiple membrane-spanning domains.

VRG4 Encodes a Protein Required for Viability

Diploids were sporulated and dissected tetrads were analyzed for cell viability (Fig. 9). The resulting tetrad analysis demonstrated that the VRG4 gene encodes a protein that is essential for cell viability. Only those Leu segregants carrying the wild type copy of VRG4 were viable. Haploid segregants carrying the null allele were inviable. Spores carrying the vrg4 disruption were able to germinate since microcolonies containing about 20 cells were formed. The conclusion from this experiment is that this gene product is required for the vegetative growth of these cells.

VRG4 Defines a Related Family of Proteins

The second VRG4-related gene is the Leishmania homologue, LPG2. The Vrg4p and Lpg2p proteins are 28% identical and 54% similar. Like vrg4, the Leishmania lpg2 mutant is defective in mediating modifications that are specific to the Golgi(35) . The VRG4 gene does not rescue a Leishmania lpg2 mutant(35) , and likewise the LPG2 gene fails to complement the vrg4 mutant. ()It is unknown whether or not this lack of complementation reflects the evolutionary divergence between these two organisms, or whether yet another VRG4-like gene exists that is more functionally related to LPG2 .

A search of the data base of expressed sequence tags (DBEST) has identified an additional set of cDNAs that encode putative proteins with a high degree of homology to VRG4, though no functions have been ascribed to these. These include one Arabidopsis (Arabidopsis thaliana) and two rice (Oryza sativa) genes. In the case of these three genes, sequence alignments suggest that these three cDNAs represent partial sequences. A summary of the similarity between VRG4 and these related genes is shown in Fig. 7B. The identification of these related genes suggests that the VRG4 gene product serves a highly conserved function, both in Saccharomyces cerevisiae and in other distantly related eukaryotes.

DISCUSSION

The analysis of oligosaccharide modification and the extent of glycosylation has been a major tool for the analysis of glycoprotein localization within the secretory pathway. Extending this idea, we have screened yeast mutants with Golgi-specific glycosylation defects to identify any that affect protein transport in the Golgi. We have identified one mutant in this screen, vrg4, that has a severe glycosylation defect and in addition, affects transport processes that occur specifically in the Golgi complex. Recently, Kanik-Ennulat et al.(11) have identified the VRG4 gene as VAN2, a gene that can mutate to confer vanadate resistance. The van2-93 mutant was isolated as a vanadate-resistant isolate and, similarly, was shown to be required for viability and to secrete underglycosylated invertase. The vrg4 mutant is defective in the retrieval of HDEL-bearing ER proteins, the transport of CPY through the Golgi and, morphologically, exhibits an aberrant endomembrane system. Together, these phenotypes suggest that this protein has an important role in regulating normal Golgi function.

Like vrg4, several other mutants, including pmr1 and erd1, have been identified that affect the secretory pathway and are also defective in glycosylation. Both pmr1 and erd1 are resistant to vanadate, although not to the same degree as vrg4. The PMR1 (plasma membrane ATPase-related) gene encodes a Golgi-localized Ca ATPase and leads to the same underglycosylation of invertase as seen in mnn9(36, 37) . While the mechanism for the effect of pmr1 on the secretory pathway is not understood, it is likely to be the result of its affect on Ca flux into or out of the Golgi(36, 37) . The erd1 mutation affects the retention of HDEL-bearing ER proteins and also leads to the underglycosylation of invertase(30, 31) . Like PMR1, ERD1 may play a role in maintaining some aspect of Golgi structure or ionic environment that, when perturbed, results in pleiotropic effects on both glycosylation and secretion(30) . The deleterious effect of the vrg4 mutation on Golgi functions argues that Vrg4p may perform a similar function. The Leishmania VRG4 homologue, LPG2, immunolocalizes to the Golgi apparatus(35) , suggesting that Vrg4p is resident in that compartment and may play a direct role in mediating Golgi functions. Given its affect on the sorting of ER proteins and on the outer chain glycosylation of invertase, processes that are thought to occur in an early yeast Golgi compartment(21) , it is likely that Vrg4p affects an early Golgi compartment, although it may, in addition, influence later compartments as well.

What is the essential function of Vrg4p? Although necessary for normal glycosylation, the protein is probably not a glycosyltransferase. Thus far only four yeast genes that encode Golgi-specific glycosyltransferases have been isolated. These include MNN1(25, 38) , MNT1 (also known as KRE2)(39, 40) , OCH1(41) , and MNN10(42) . Like Golgi-localized glycosyltransferases in higher eukaryotes, these gene products share certain structural features. All are Type II membrane proteins, with single transmembrane domains and large C-terminal lumenal domains. The predicted Vrg4 protein sequence does not match this consensus. Furthermore, yeast can survive, albeit poorly, in the absence of outer chain glycosylation. Since the VRG4 gene is required for viability, it is unlikely that its essential function is carbohydrate modification.

The aberrant morphology of mutant membranes, most clearly seen by thin-section electron microscopy, suggests at least one important function in which the VRG4 gene does play a role. The mutants lack the high staining contrast of wild type permanganate-stained membranes. The mechanism by which permanganate stains membranes is not well understood. It is likely due to the deposition of MnO at the polar ends of lipids in the membrane(43) . The clarity of permanganate-stained membrane is also thought to be due, in part, to the loss of protein components in the membrane as a result of oxidative cleavage of proteins by KMnO(43) . In either case, the altered staining characteristics of mutant membranes reflects an alteration in their lipid or protein composition, suggesting that the essential role of VRG4 is to establish or maintain the normal lipid/protein ratio of these membranes.

The combined effects of VRG4 on a number of Golgi-specific functions, coupled to its effect on membrane morphology, suggest that Vrg4p plays an important role in establishing or maintaining the organization of this organelle. Analyses of this protein, as well as those defective in other vanadate-resistant mutants that indirectly affect the Golgi complex, will further our understanding of the factors that regulate the structure and function of this organelle.

FOOTNOTES

*

This work was supported in part by Grant GM-48467 from the National Institutes of Health and a grant from the Council for Tobacco Research (to N. D.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore by hereby marked ``advertisement'' in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The nucleotide sequence(s) reported in this paper has been submitted to the GenBank(TM)/EMBL Data Bank with accession number(s) L33915[GenBank].

§

To whom correspondence should be addressed. Tel.: 516-632-9309; Fax: 516-632-8575.

()

The abbreviations used are: ER, endoplasmic reticulum; kb, kilobase pair(s); BiP, binding protein; CPY, carboxypeptidase Y; PAGE, polymacrylamide gel electrophoresis.

()

N. Dean, unpublished data.

()

J. B. Poster and N. Dean, unpublished observations.

ACKNOWLEDGEMENTS

Note Added in Proof-We have recently analyzed the localization of epitope-tagged Vrg4 protein by indirect immunofluorescence. Like the Leishmania homologue, yeast Vrg4p is localized in the Golgi complex, supporting our conclusion that its affect on Golgi functions is direct.

REFERENCES

1. Ballou, L., Hitzeman, R. A., Lewis, M. S., and Ballou, C. E. (1991) Proc. Natl. Acad. Sci. U. S. A. 88, 3209-3212 [Abstract/Free Full Text]
2. Herscovics, A., and Orlean, P. (1993) FASEB J. 7, 540-550 [Abstract]
3. Kukuruzinska, M. A., Bergh, M. L. E., and Jackson, B. J. (1987) Annu. Rev. Biochem. 56, 915-944 [CrossRef][Medline] [Order article via Infotrieve]
4. Ballou, C. E. (1990) Methods Enzymol. 185, 440-470 [Medline] [Order article via Infotrieve]
5. Novick, P., Field, C., and Schekman, R. (1980) Cell 21, 205-215 [CrossRef][Medline] [Order article via Infotrieve]
6. Huffaker, T. C., and Robbins, P. W. (1983) Proc. Natl. Acad. Sci. U. S. A. 80, 7466-7470 [Abstract/Free Full Text]
7. Raschke, W. C., Kern, K. A., Antalis, C., and Ballou, C. E. (1973) J. Biol. Chem. 248, 4660-4666 [Abstract/Free Full Text]
8. Ballou, L., Cohen, R. E., and Ballou, C. E. (1980) J. Biol. Chem. 255, 5986-5991 [Free Full Text]
9. Lewis, M. S., and Ballou, C. E. (1991) J. Biol. Chem. 266, 8255-8261 [Abstract/Free Full Text]
10. Kanik-Ennulat, C., and Neff, N. (1990) Mol. Cell. Biol. 10, 898-909 [Abstract/Free Full Text]
11. Kanik-Enulat, C., Montalvo, E., and Neff, N. (1995) Genetics 140, 933-943 [Abstract]
12. Guthrie, C., and Fink, G. R. (1991) Methods Enzymol. 194, 3-20 [CrossRef][Medline] [Order article via Infotrieve]
13. Ito, H. Y., Fukuda, Y., Murata, K., and Kimura, J. (1983) J. Bacteriol. 153, 163-168 [Abstract/Free Full Text]
14. Sanger, F., Nicklen, S., and Coulson, A. R. (1977) Proc. Natl. Acad. Sci. U. S. A. 74, 5463-5467 [Abstract/Free Full Text]
15. Yannisch-Perron, C., Vieira, J., and Messing, J. (1985) Gene (Amst.) 33, 103-119
16. Altschul, S. F., Gish, W., Miller, W., Myers, E. W., and Lipman, D. J. (1990) J. Mol. Biol. 215, 403-410 [CrossRef][Medline] [Order article via Infotrieve]
17. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual , Cold Spring Harbor Laboratory, Cold Spring Harbor, NY
18. Sikorski, R., S, and Hieter, P. (1989) Genetics 122, 19-27 [Abstract/Free Full Text]
19. Carle, G. F., and Olsen, M. V. (1985) Proc. Natl. Acad. Sci. U. S. A. 82, 3756-3760 [Abstract/Free Full Text]
20. Laemmli, U. K. (1970) Nature 227, 680-685 [CrossRef][Medline] [Order article via Infotrieve]
21. Dean, N., and Pelham, H. R. B. (1990) J. Cell Biol. 111, 369-377 [Abstract/Free Full Text]
22. Koning, A. J., Lum, P. Y., Williams, J. M., and Wright, R. (1993) Cell Motil. Cytoskel. 25, 111-128 [CrossRef][Medline] [Order article via Infotrieve]
23. Kaiser, C., and Schekman, R. (1990) Cell 61, 723-733 [CrossRef][Medline] [Order article via Infotrieve]
24. Dean, N. (1995) Proc. Natl. Acad. Sci. U. S. A. 92, 1287-1291 [Abstract/Free Full Text]
25. Yip, C. L., Welch, S. K., Klebl, F., Gilbert, T., Seidel, P., Grant, F. J., O'Hara, P. J., and MacKay, V. L. (1994) Proc. Natl. Acad. Sci. U. S. A. 91, 2723-2727 [Abstract/Free Full Text]
26. Willsky, G. R., Leung, J. O., Offerman, P. V., Plotnick, E. K., and Dosch, S. F. (1985) J. Bacteriol. 164, 611-617 [Abstract/Free Full Text]
27. Stevens, T., Esmon, B., and Schekman, R. (1982) Cell 30, 439-448 [CrossRef][Medline] [Order article via Infotrieve]
28. te Heesen, S., Janetsky, B., Lehle, L., and Aebi, M. (1992) EMBO J. 11, 2071-2075 [Medline] [Order article via Infotrieve]
29. Pelham, H. R. B. (1989) Annu. Rev. Cell Biol. 5, 1-23 [CrossRef]
30. Hardwick, K. G., Lewis, M. J., Semenza, J., Dean, N., and Pelham, H. R. B. (1990) EMBO J. 9, 623-630 [Medline] [Order article via Infotrieve]
31. Pelham, H. R. B., Hardwick, K. G., and Lewis, M. J. (1988) EMBO J. 7, 1757-1762 [Medline] [Order article via Infotrieve]
32. Terasaki, M. (1989) Methods Cell Biol. 29, 125-135 [Medline] [Order article via Infotrieve]
33. Kyte, J., and Doolittle, R. F. (1982) J. Mol. Biol. 157, 105-132 [CrossRef][Medline] [Order article via Infotrieve]
34. Rothstein, R. J. (1983) Methods Enzymol. 101, 202-211 [Medline] [Order article via Infotrieve]
35. Descouteaux, A., Luo, Y., Turco, S. J., and Beverley, S. M. (1995) Science 269, 1869-1872 [Abstract/Free Full Text]
36. Antebi, A., and Fink, G. R. (1992) Mol. Biol. Cell 3, 633-654 [Abstract]
37. Rudolph, H. K., Antebi, A., Fink, G. R., Buckley, C. M., Dorman, T. E., LeVitre, J., Davidow, L. S., Mao, J., and Moir, D. T. (1989) Cell 58, 133-145 [CrossRef][Medline] [Order article via Infotrieve]
38. Graham, T. R., Verostek, M. F., Mackay, V., Trimble, R., and Emr, S. D. (1992) Yeast 8, S458
39. Hausler, A., and Robbins, P. W. (1992) Glycobiol. 2, 77-84 [Abstract/Free Full Text]
40. Hill, K., Boone, C., Goebl, M., Puccia, R., Sdicu, A. M., and Bussey, H. (1992) Genetics 130, 273-283 [Abstract]
41. Nakayama, K., Nagasu, T., Shimma, Y., Kuromitsu, J., and Jigami, Y. (1992) EMBO J. 11, 2511-2519 [Medline] [Order article via Infotrieve]
42. Dean, N., and Poster, J. (1996) Glycobiology , in press
43. Hayat, M. A. (1981) Fixation for Electron Microscopy , pp. 184-193, Academic Press, Inc., New York

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				T. R. Cottrell, C. L. Griffith, H. Liu, A. A. Nenninger, and T. L. Doering The Pathogenic Fungus Cryptococcus neoformans Expresses Two Functional GDP-Mannose Transporters with Distinct Expression Patterns and Roles in Capsule Synthesis Eukaryot. Cell, May 1, 2007; 6(5): 776 - 785. [Abstract] [Full Text] [PDF]

				S. Bates, D. M. MacCallum, G. Bertram, C. A. Munro, H. B. Hughes, E. T. Buurman, A. J. P. Brown, F. C. Odds, and N. A. R. Gow Candida albicans Pmr1p, a Secretory Pathway P-type Ca2+/Mn2+-ATPase, Is Required for Glycosylation and Virulence J. Biol. Chem., June 17, 2005; 280(24): 23408 - 23415. [Abstract] [Full Text] [PDF]

				M. Abe, Y. Noda, H. Adachi, and K. Yoda Localization of GDP-mannose transporter in the Golgi requires retrieval to the endoplasmic reticulum depending on its cytoplasmic tail and coatomer J. Cell Sci., November 1, 2004; 117(23): 5687 - 5696. [Abstract] [Full Text] [PDF]

				X.-D. Gao, A. Nishikawa, and N. Dean Physical interactions between the Alg1, Alg2, and Alg11 mannosyltransferases of the endoplasmic reticulum Glycobiology, June 1, 2004; 14(6): 559 - 570. [Abstract] [Full Text] [PDF]

				L. Norambuena, L. Marchant, P. Berninsone, C. B. Hirschberg, H. Silva, and A. Orellana Transport of UDP-galactose in Plants. IDENTIFICATION AND FUNCTIONAL CHARACTERIZATION OF AtUTr1, AN ARABIDOPSIS THALIANA UDP-GALACTOSE/UDP-GLUCOSE TRANSPORTER J. Biol. Chem., August 30, 2002; 277(36): 32923 - 32929. [Abstract] [Full Text] [PDF]

				A. Nishikawa, J. B. Poster, Y. Jigami, and N. Dean Molecular and Phenotypic Analysis of CaVRG4, Encoding an Essential Golgi Apparatus GDP-Mannose Transporter J. Bacteriol., January 1, 2002; 184(1): 29 - 42. [Abstract] [Full Text] [PDF]

				T. C. Baldwin, M. G. Handford, M.-I. Yuseff, A. Orellana, and P. Dupree Identification and Characterization of GONST1, a Golgi-Localized GDP-Mannose Transporter in Arabidopsis PLANT CELL, October 1, 2001; 13(10): 2283 - 2295. [Abstract] [Full Text] [PDF]

				X.-D. Gao, V. Kaigorodov, and Y. Jigami YND1, a Homologue of GDA1, Encodes Membrane-bound Apyrase Required for Golgi N- and O-Glycosylation in Saccharomyces cerevisiae J. Biol. Chem., July 23, 1999; 274(30): 21450 - 21456. [Abstract] [Full Text] [PDF]

				N. Dean, Y. B. Zhang, and J. B. Poster The VRG4 Gene Is Required for GDP-mannose Transport into the Lumen of the Golgi in the Yeast, Saccharomyces cerevisiae J. Biol. Chem., December 12, 1997; 272(50): 31908 - 31914. [Abstract] [Full Text] [PDF]

K. Fiedler and J. E. Rothman Sorting Determinants in the Transmembrane Domain of p24 Proteins J. Biol. Chem., October 3, 1997; 272(40): 24739 - 24742. [Abstract] [Full Text] [PDF]