In higher eukaryotes a quality control system monitoring the folding state of glycoproteins is located in the ER and is composed of the proteins calnexin, calreticulin, glucosidase II, and UDP-glucose: glycoprotein glucosyltransferase. It is believed that the innermost glucose residue of the N-linked oligosaccharide of a glycoprotein serves as a tag in this control system and therefore performs an important function in the protein folding pathway. To address this function, we constructed Saccharomyces cerevisiae strains which contain nonglucosylated (G0), monoglucosylated (G1), or diglucosylated (G2) glycoproteins in the ER and used these strains to study the role of glucose residues in the ER processing of glycoproteins. These alterations of the oligosaccharide structure did not result in a growth phenotype, but the induction of the unfolded protein response upon treatment with DTT was much higher in G0 and G2 strains as compared to wild-type and G1 strains. Our results provide in vivo evidence that the G1 oligosaccharide is an active oligosaccharide structure in the ER glycoprotein processing pathway of S.cerevisiae. Furthermore, by analyzing N-linked oligosaccharides of the constructed strains we can directly show that no general glycoprotein glucosyltransferase exists in S.cerevisiae. Key words: protein folding/protein glycosylation/protein transport/quality control/yeast
In higher eukaryotes, there is growing evidence of a quality control system monitoring the folding status of glycoproteins in the endoplasmic reticulum (ER). This system ensures that misfolded glycoproteins will be retained in the ER and either be subject to refolding or degradation. Correctly folded glycoproteins then exit the ER for further processing in the Golgi apparatus and are sorted to their final destinations. N-Linked glycoproteins are either co- or post-translationally translocated into the lumen of the ER whereby the oligosaccharyltransferase transfers a preassembled oligosaccharide Glc3Man9GlcNAc2 (G3) from dolichyl-pyrophosphate to asparagine residues in the sequon Asn - X - Ser/Thr (Kornfeld and Kornfeld, 1985; Roth, 1995). Once transferred to protein the oligosaccharide is subject to trimming by the glucosidase I removing the terminal [alpha]1,2-linked glucose residue (Glc2Man9GlcNAc2; G2), followed by the action of glucosidase II which cleaves sequentially the remaining [alpha]1,3-linked glucose residues (Glc1Man9GlcNAc2, G1; Man9GlcNAc2, G0). Finally, in S.cerevisiae, [alpha]-mannosidase I selectively removes one [alpha]1,2-linked mannose residue (Man8GlcNAc2) before the glycoproteins exit the ER (Helenius, 1994; Moremen et al., 1994). By using the glucosidase inhibitors castanospermine and 1-deoxynojirimycin, Hammond et al., (1994) and Hebert et al. (1995) demonstrated in mammalian cells the importance of monoglucosylated N-linked oligosaccharides for binding to calnexin. Moreover, UDP-glucose: glycoprotein glucosyltransferase, present in S.pombe (Fernandez et al., 1994), D.melanogaster (Parker et al., 1995) and rat (Trombetta and Parodi, 1992), is able to reglucosylate G0 oligosaccharides on N-linked glycoproteins which expose hydrophobic patches (Fernandez et al., 1994). The misfolded reglucosylated glycoprotein can be recognized and bound by calnexin and thereby retained in the folding environment of the ER. The correctly folded glycoprotein would then become a substrate for glucosidase II and be able to exit the ER. This quality control model first postulated by A NAME=bbr15_1 HREF=#bb15>Hammond and Helenius (1994) for mammalian cells defines a central role of monoglucosylated glycoproteins in the control of protein folding in the ER. To address the role of glucosylated glycoproteins in vivo, we used yeast genetic techniques and constructed strains which contain nonglucosylated or mono- or diglucosylated glycoproteins in the ER and studied possible effects on glycoprotein processing. Our data are compatible with the model that only the monoglucosylated and not the non- or diglucosylated N-linked oligosaccharides are the positive signal for protein folding in the ER in vivo. Sequence homologies of the ORF YBR229c to pig liver and human lymphocyte glucosidase II
Using the amino acid sequence derived from the pig liver glucosidase II (Flura et al., 1997), we performed a database search in the nonredundant protein database and identified an open reading frame of S.cerevisiae with a sequence similarity of 34% (GenEMBL accession no. Z36098). The yeast amino acid sequence showed a sequence identity of 46% in the more conserved C-terminal regions (amino acids 234-954; data not shown). This ORF YBR229c, now termed GLS2, was replaced with the geneticin resistance gene by homologous recombination. In order to biochemically confirm that this locus encodes glucosidase II, membrane fractions of wild-type, gls1-1 and [Delta]gls2 cells were assayed for glucosidase II activity. The membrane-enriched protein preparation of the GLS2-deficient cells displayed a 5-fold reduced enzymatic activity as compared to both wild-type and gls1-1 cells in two independent [Delta]gls2 strains (Table II). We do not know the nature of the residual substrate-hydrolizing activity in the [Delta]gls2 preparations since the substrate was stable in samples incubated without crude extract. Furthermore, the mobility of mature vacuolar CPY of the [Delta]gls2 strains was slightly reduced due to the inability to trim the two glucose residues of each oligosaccharide (see below). These findings in addition to the amino acid sequence similarity confirmed that the locus ORF YBR229c encoded glucosidase II. Table 1
Glycobiology Pages
Genetic tailoring of N-linked oligosaccharides: the role of glucose residues in glycoprotein processing of Saccharomyces cerevisiae in vivo
Introduction
Results
Discussion
Materials and methods
Acknowledgments
Abbreviations
References
Genetic tailoring of N-linked oligosaccharides: the role of glucose residues in glycoprotein processing of Saccharomyces cerevisiae in vivo
Introduction
Results
Strain
Genotype
Reference
SS328
MAT[alpha] ade2-101 ura3-52 his3[Delta]200 lys2-801
(Vijayraghavan et al., 1989)
MLY1601
MATa leu2-3, 112 [Delta]pep4::URA3 gls1-1
(Latterich and Schekman, 1994)
YG592
MATa ade2-101 ura3-52 his3[Delta]200 lys2-801 [Delta]alg6::HIS3
(Reiss et al., 1996)
YG126
MAT[alpha] ade2-101 ura3-52 his3[Delta]200 lys2-801 [Delta]alg8::HIS3
(Stagljar et al., 1994)
YG428
MAT[alpha] ade2-101 ura3-52 his3[Delta]200 tyr1 [Delta]alg10::KanMX
This study
YG427
MAT[alpha] ade2-101 ura3-52 his3[Delta]200 lys2-801 [Delta]gls2::KanMX
This study
YG590
MATa ade2-101 ura3-52 his3[Delta]200 lys2-801 [Delta]gls2::KanMX [Delta]alg6::HIS3
This study
YG424
MATa ade2-101 ura3-52 his3[Delta]200 lys2-801 [Delta]gls2::KanMX [Delta]alg8::HIS3
This study
YG491
MATa ade2-101 ura3-52 his3[Delta]200 lys2-801 [Delta]gls2::KanMX [Delta]alg10::KanMX
This study
YG571
MAT[alpha] ade2-101 ura3-52 his3[Delta]200 lys2-801 ura3-52::UPRE-lacZ-URA3
This study
YG572
MAT[alpha] ade2-101 ura3-52 his3[Delta]200 lys2-801 [Delta]gls2::KanMX ura3-52::UPRE-lacZ-URA3
This study
YG595
MATa ade2-101 ura3-52 his3[Delta]200 lys2-801 [Delta]alg6::HIS3 ura3-52::UPRE-lacZ-URA3
This study
YG567
MAT[alpha] ade2-101 ura3-52 his3[Delta]200 lys2-801 [Delta]alg8::HIS3 ura3-52::UPRE-lacZ-URA3
This study
YG568
MAT[alpha] ade2-101 ura3-52 his3[Delta]200 lys2-801 [Delta]alg10::KanMX ura3-52::UPRE-lacZ-URA3
This study
YG596
MATa ade2-101 ura3-52 his3[Delta]200 lys2-801 [Delta]gls2::KanMX [Delta]alg6::HIS3 ura3-52::UPRE-lacZ- URA3
This study
YG569
MATa ade2-101 ura3-52 his3[Delta]200 lys2-801 [Delta]gls2::KanMX [Delta]alg8::HIS3 ura3-52::UPRE-lacZ-URA3
This study
YG570
MATa ade2-101 ura3-52 his3[Delta]200 lys2-801 [Delta]gls2::KanMX [Delta]alg10::KanMX ura3-52::UPRE-lacZ-URA3
This study
YG556
MAT[alpha] ade2-101 ura3-52 his3[Delta]200 lys2-801 leu2 sec18-50
This study
YG603
MAT[alpha] ade2-101 ura3-52 his3[Delta]200 lys2-801 leu2 [Delta]gls2::KanMX [Delta]alg6::HIS3 sec18-50
This study
YG562
MAT[alpha] ade2-101 ura3-52 his3[Delta]200 lys2-801 leu2 [Delta]gls2::KanMX [Delta]alg8::HIS3 sec18-50
This study
YG563
MATa ade2-101 ura3-52 his3[Delta]200 lys2-801 leu2 [Delta]gls2::KanMX [Delta]alg10::KanMX sec18-50
This study
| Strain | Glucosidase II activity | Relative GLS2p activity |
| Wild-type | 287 U mg protein-1 | 1.0 |
| gls1-1 | 270 U mg protein-1 | 0.9 |
| [Delta]gls2 (clone 3) | 57 U mg protein-1 | 0.2 |
| [Delta]gls2 (clone 7) | 50 U mg protein-1 | 0.2 |
Genetic tailoring of oligosaccharide structures
In order to obtain N-linked oligosaccharides of defined structure on glycoproteins in the ER of S.cerevisiae, the [Delta]gls2 mutation was crossed with several yeast strains harboring deletions in the lipid-linked oligosaccharide biosynthetic pathway. Incomplete lipid-linked oligosaccharide can be transferred to protein, albeit with a reduced efficiency (Stagljar et al., 1994). In [Delta]alg6 cells nonglucosylated oligosaccharide (G0; Reiss et al., 1996) is transferred, in [Delta]alg8 strains monoglucosylated (G1; Stagljar et al., 1994) and in [Delta]alg10 strains diglucosylated oligosaccharide (G2; this study and Burda et al., unpublished observations) as depicted in Figure 1. In the absence of glucosidase II activity, glucose residues should remain on the protein-bound oligosaccharides. Therefore, the combination of specific alg deletions with the [Delta]gls2 mutation should make it possible to genetically tailor the structure of the oligosaccharide on glycoproteins in the ER (Figure 1, Table III). To test this hypothesis, appropriate [Delta]algX[Delta]gls2 strains were constructed and characterized. The growth rates of all strains were equal to wild-type cells and no temperature-dependent growth phenotype was observed (data not shown). Nonlethal mutations in the alg genes lead to an incomplete assembly of the dolichol-linked oligosaccharide and thereby form a suboptimal substrate for the oligosaccharyltransferase (Sharma et al., 1981). The effect of this suboptimal substrate in vivo is apparent by the occurrence of underglycosylated glycoproteins such as CPY and Wbp1p. Protein extracts of exponential growing yeast strains were subjected to Western blot analysis using anti-CPY and anti-Wbp1p sera. As expected the [Delta]gls2 mutant showed no underglycosylation of both glycoproteins CPY and Wbp1p (Figure 2, lane 2), because the completely assembled dolichol-linked oligosaccharide is transferred to the nascent polypeptide chains. Furthermore, the combination of [Delta]gls2 with deletions in alg genes did not modulate the extent of underglycosylation of CPY and Wbp1p (Figure 2, lanes 3-8): the pattern of underglycosylation of both, CPY and Wbp1p, was not altered in the [Delta]alg cells by addition of the [Delta]gls2 mutation. However, we observed a different molecular weight of the resulting glycoproteins due to the altered biosynthesis and trimming of the oligosaccharides. This alteration is visualized best in the case of the fully glycosylated Wbp1p which showed the slowest mobility in [Delta]gls2 and [Delta]alg10[Delta]gls2 cells (Figure 2B, lanes 2 and 8), and an intermediate mobility in [Delta]alg8[Delta]gls2 cells (Figure 2B, lane 7). In [Delta]alg6[Delta]gls2 cells, the fully glycosylated form of Wbp1p had the same molecular weight as the wild-type cells (Figure 2B, lanes 6 and 1, respectively). A similar change in mobility was observed for the different CPY glycoforms. These altered mobilities showed that in the absence of glucosidase II, [alpha]1,3-linked glucose residues were not removed from glycoproteins and were still found on vacuolar proteins (e.g., mature CPY). Most significantly, these glucose residues were also retained on the ER-resident protein Wbp1p. This data shows by differences in electrophoretic mobility that the combination of alg and gls2 mutations makes it possible to tailor the glucosylation state of glycoproteins in vivo.
Figure N-Linked oligosaccharide composition of glycoproteins of the tailored strains
In order to verify the differences in protein-linked oligosaccharide composition as observed by SDS-PAGE (Figure 2), lipid-linked oligosaccharides (LLO) and N-linked oligosaccharides (NLO) were labeled with [3H]mannose, extracted, released with EndoH, and analyzed by HPLC. To prevent modification by Golgi mannosyltransferases the temperature-conditional sec18-50 mutation (Esmon et al., 1981; Novick et al., 1981; Verostek et al., 1991, 1993) was introduced into all analyzed strains. The cells were shifted to nonpermissive temperature for 30 min prior to labeling to block the fusion of ER vesicles to the Golgi. As expected, the [Delta]alg10[Delta]gls2 mutation lead to the accumulation of the incomplete LLO Glc2Man9GlcNAc2 (G2M9), the [Delta]alg8[Delta]gls2 mutations to the Glc1Man9GlcNAc2 (G1M9) and the [Delta]alg6[Delta]gls2 mutations to the Man9GlcNAc2 (M9) LLO (Figure 3; LLO). The identity of the EndoH-released NLO signals (Figure 3; NLO) was determined by mixing appropriate LLO and NLO extracts and HPLC analysis (Figure 3; LLO+NLO). In the wild-type strain, the NLO consisted of Man9GlcNAc2 and Man8GlcNAc2 oligosaccharides. In the [Delta]alg6[Delta]gls2 mutant strain the same oligosaccharide composition as in the wild-type strain was found (Man9GlcNAc2 and Man8GlcNAc2). The Man9GlcNAc2 and Man8GlcNAc2 signals completely disappeared upon digestion with [alpha]-mannosidase (data not shown). In the [Delta]alg8[Delta]gls2 double mutant the Glc1Man9GlcNAc2 was identified by mixing experiments, the other occurring peak appears to be Glc1Man8GlcNAc2 (G1M8, see Discussion) since only one oligosaccharide species (Glc1Man4GlcNAc2) was detectable after [alpha]-mannosidase treatment (data not shown). Likewise, in the case of the [Delta]alg10[Delta]gls2 double mutant the NLO structure Glc2Man9GlcNAc2 was confirmed and presumably the Glc2Man8GlcNAc2 (G2M8, see discussion) also exists. Upon [alpha]-mannosidase digestion of these NLO only one single oligosaccharide (Glc2Man4GlcNAc2) was detectable (data not shown). Hence, by HPLC analysis of the N-linked oligosaccharides, we could directly show that the oligosaccharide structures as predicted by genetics occurred in our strains in vivo.
Absence of a general glycoprotein glucosyltransferase activity
Since the [Delta]alg6[Delta]gls2 strain is glucosidase II deficient, glucose residues transferred to protein-bound oligosaccharides by a general glycoprotein glucosyltransferase ought to be visible by analyzing the NLO profile of the [Delta]alg6[Delta]gls2 strain. The LLO of the [Delta]alg6[Delta]gls2 mutant strain consisted mainly of Man9GlcNAc2 (Figure 3, LLO). No glucose-containing LLO were detected. Likewise, no NLO of a larger size than Man9GlcNAc2 of the same strain were observed (Figure 3, NLO) confirming the absence of a general glycoprotein glucosyltransferase in S.cerevisiae (Fernandez et al., 1994).
Table III.
Disturbing the protein folding and transport of CPY by DTT
Figure
Figure
Figure
The absence of any detectable growth phenotype in [Delta]algX[Delta]gls2 strains demonstrate that a potential quality control/folding system requiring correctly glucosylated glycoproteins was not limiting for growth in these strains. In order to assay for the role of glucose residues on glycoprotein folding, we have chosen two experimental approaches: first, reducing the efficiency of protein folding by the addition of DTT (Simons et al., 1995), and second, determining the degree of induction of the unfolded protein response (UPR; Mori et al., 1992) as a measure for the extent of misfolded proteins in vivo. Reducing agents such as DTT modulate the redox potential in the ER. In yeast, it has been shown that proteins containing disulfides are retained in the ER, whereas the transport of proteins devoid of disulfide bonds such as [alpha]-factor is not affected (Jamsa et al., 1994; Simons et al., 1995). We examined the concentration-dependent interference of DTT on the transport of CPY by monitoring the proteolytic digestion of the protein during its maturation. Concentrations of higher than 0.5 mM DTT reversibly blocked the export of CPY out of the ER irrespective of the oligosaccharide structure (Figure 4). Furthermore, an underglycosylated form of the ER-resident p1CPY (Figure 4A-C, p1CPY*) occurred. Both, the p1CPY and the underglycosylated p1CPY* displayed identical molecular weight upon treatment with endoglycosidase H (Figure 4B). Upon diluting DTT to 0.5 mM, the transport and maturation of CPY reoccurred demonstrating the reversibility of the DTT-induced transport block. Moreover, upon reinitiating the protein transport, the p1CPY* then disappeared, either due to glycosylation and export or degradation (Figure 4C). In a pulse chase experiment, we were able to demonstrate that CPY, trapped in the ER by DTT, maturated normally and was targeted to the vacuole upon dilution of the reducing agent (Figure 4D). Furthermore, the transport of CPY was independent of the glucose composition (wild-type, G1 or G2) of the N-linked oligosaccharides (Figure 4D). We determined the transport rate of CPY by quantifying the bands of the proCPY (ER and Golgi form) and the CPY (vacuolar form) using a Phosphoimager. The transport rate of CPY of the wild-type and [Delta]gls2 strains were identical (T50 = 18 min). In the other strains, CPY was processed at a slightly slower rate (T50 = 23 and >30 min, respectively), however, no glucose-dependent transport rate was observed. CPY bearing G1 glycoproteins was transported from ER to vacuole at the same rate as CPY exhibiting the G2 oligosaccharide (Figure 4D). Likewise, the CPY transport from ER to Golgi was not influenced by the glucose composition of the oligosaccharide (data not shown). A reduced transport rate of underglycosylated CPY has been reported previously (Winther et al., 1991; te Heesen and Aebi, 1994).
Induction of unfolded protein response
In yeast, the presence of misfolded proteins in the ER is transmitted from the ER lumen into the nucleus via a signal transduction pathway requiring the IRE1 protein kinase (Cox et al., 1993; Mori et al., 1993; Nikawa et al., 1993). Specific transcription factors such as Hac1p (Cox and Walter, 1996) bind to specific promoter segments, the unfolded protein response elements (UPRE), present and highly conserved in the promoters of KAR2/BiP, PDI1, EUG1, and FKB2 (Shamu et al., 1994) and activate the appropriate genes. Such a regulatory system allows for a direct measurement of the unfolded protein content in the ER in vivo (Mori et al., 1992). We used a reporter system based on the E.coli ß-galactosidase (LacZ) gene under the control of the UPRE of the yeast KAR2/BiP promoter. We ligated the UPRE upstream of the ß-galactosidase gene according to Mori et al. (1992) and integrated it as a single copy into the ura3-52 locus. This single copy reporter construct, placed at the identical location in all strains tested, allowed a direct comparison of the UPR in the different strains used. By the sensitive ß-galactosidase read-out, we quantified the extent of misfolded proteins in the ER of the various yeast strains with genetically tailored N-linked oligosaccharides. Tunicamycin is known to strongly induce the unfolded protein response (Mori et al., 1992). Upon tunicamycin treatment, the ß-galactosidase activity was strongly stimulated in all strains as compared to untreated cells (40- to 50-fold induction in single experiment, data not shown). When incubating the cells with 0.5 mM DTT, a concentration interfering with protein folding but not abolishing the ER to Golgi transport (Figure 4A,D), we observed a differential response. Wild-type cells did not induce ß-galactosidase activity upon DTT treatment, whereas [Delta]gls2 cells did (Figure 5). The deletion of GLS2 leads to conservation of the N-linked oligosaccharides in the diglucosylated G2 form (Table III). The [Delta]alg6 mutation prevents the addition of the innermost [alpha]1,3-linked glucose residue and thereby renders the N-linked oligosaccharide structure in the unglucosylated G0 form. In [Delta]alg6 cells, the UPR was also stimulated upon treatment with DTT. Furthermore, the additional loss of glucosidase II activity in [Delta]alg6[Delta]gls2 cells did not change this response to DTT (Figure 5). The [Delta]alg8 mutation leads to the transfer of monoglucosylated oligosaccharides (G1) and in combination with the [Delta]gls2 deletion the single glucose remains on the oligosaccharide of glycoproteins. In contrast to [Delta]alg6, [Delta]alg6[Delta]gls2, and [Delta]gls2 cells, the UPR was not induced neither in [Delta]alg8 nor in [Delta]alg8[Delta]gls2 cells. Alg10p transfers the outermost glucose residue to complete the synthesis of the dolichol-linked oligosaccharide precursor. Upon inactivation of this locus ([Delta]alg10) the lipid-linked oligosaccharide is transferred in the diglucosylated form (G2) to glycoproteins. The trimming of the glucose residues was not impaired. In the [Delta]alg10 strain the UPR was the same as in the wild-type, [Delta]alg8 and [Delta]alg8[Delta]gls2 strains. However, upon combination of [Delta]alg10 with [Delta]gls2 and thereby preserving the N-linked oligosaccharide structure in the G2 form, the unfolded protein response was induced by DTT. We conclude that with respect to the induction of the UPR by DTT the [Delta]alg8 mutation was epistatic over [Delta]gls2, but [Delta]alg10 was not.
Figure
Using the combined approach of the complete yeast genome information and yeast genetic techniques, we were able to obtain yeast strains with tailored oligosaccharides not only on lipid-linked precursors but also linked to proteins. By HPLC analysis we confirmed the structures of the oligosaccharides suggested by genetics. Interestingly, the NLO in the wild-type and in the [Delta]alg6[Delta]gls2 strains consisted of two different oligosaccharides (Man9GlcNAc2 and Man8GlcNAc2). In the [Delta]alg8[Delta]gls2 and [Delta]alg10[Delta]gls2 strains one signal did not coelute with marker oligosaccharides. The NLO Man9GlcNAc2 is converted to Man8GlcNAc2 by the action of the ER mannosidase I (Grondin and Herscovics, 1992; Lipari and Herscovics, 1994). In analogy to the mannosidase I-trimmed Man9GlcNAc2 structures in the wild-type and in the [Delta]alg6[Delta]gls2 strains, we suppose that the second oligosaccharide species represents the corresponding mannosidase I-trimmed products ([Delta]alg8[Delta]gls2, Glc1Man8GlcNAc2; [Delta]alg10[Delta]gls2, Glc2Man8GlcNAc2). The two oligosaccharide species can be converted to only one by digesting with [alpha]-mannosidase confirming the identity of Glc1Man9GlcNAc2/Glc1Man8GlcNAc2 and Glc2Man8GlcNAc2/Glc2Man8GlcNAc2 in the [Delta]alg8[Delta]gls2 and [Delta]alg10[Delta]gls2 strains, respectively. The presence of these two oligosaccharides suggests that glucose trimming is not a prerequisite for mannosidase I activity. Interestingly, trimming of the lipid-linked oligosaccharide by mannosidase I seems not to occur.
These strains with tailored NLO provided the tools to investigate the involvement of the innermost [alpha]1,3-linked glucose residue of N-linked oligosaccharides of S.cerevisiae in coping cellular stress and its involvement in protein folding in vivo. In higher eukaryotic cells the innermost glucose residue of N-linked oligosaccharides is involved in binding to calnexin and calreticulin as shown by experiments using the glucosidase inhibitors castanospermine and 1-deoxynojirimycin (Hammond et al., 1994; Hebert et al., 1996), glucosidase II-deficient mouse lymphoma cells (Ora and Helenius, 1995) and in vitro reglucosylated ribonuclease B (Rodan et al., 1996; Zapun et al., 1997). According to the model of Hammond and Helenius (1994) monoglucosylated oligosaccharides of N-linked glycoproteins play a central role in the protein refolding process. However, the in vivo importance of this glucose residue for the overall folding process of proteins in the ER has not been demonstrated directly in any organism. In S.cerevisiae, using a reporter construct under the control of the UPRE of KAR2/BiP (Mori et al., 1992), we were able to demonstrate the involvement of the monoglucosylated N-linked oligosaccharides in glycoprotein folding. In our experiments we sought for mild nonlethal stress conditions interfering with protein folding. Since we wanted to study the influence of N-linked oligosaccharides it was not feasible to employ tunicamycin, a drug blocking N-linked glycosylation and known to strongly induce the unfolded protein response (UPR). We used a low concentration of DTT instead shown to still allow protein transport albeit at a reduced rate. Glycoproteins with fixed nonglucosylated and diglucosylated N-linked oligosaccharides induced the UPR under these mild reducing conditions whereas the response of the strain with monoglucosylated oligosaccharides was the same as in wild-type cells. We postulate that under these conditions of stress the monoglucosylated oligosaccharide-dependent folding process may be required in wild-type cells. Inactivation of this system leads to a higher degree of misfolded proteins in the ER and a subsequent induction of the UPR. Apparently, the G0 and the G2 oligosaccharides failed to interact with elements of the folding system, e.g., Cne1p, the calnexin analogue (de Virgilio et al., 1993; Parlati et al., 1995) and hence, the cells have a reduced ability to handle misfolded proteins. However, in S.cerevisiae, a direct involvement of Cne1p in the refolding network has never been demonstrated. In higher eukaryotes though, Ware et al. (1995) and Spiro et al. (1996) showed that calnexin and calreticulin, respectively, bound to oligosaccharides of the structure Glc1Man5-9GlcNAc2. More recently, Rodan et al. (1996) and Zapun et al. (1997) were able to show that the single monoglucosylated oligosaccharide of ribonuclease B was responsible for the binding to dog pancreas calnexin and calreticulin.
We note that the unfolded protein response of [Delta]gls2 strain (G2) was lower than the response of the [Delta]alg10[Delta]gls2 (G2), the [Delta]alg6 (G0) or the [Delta]alg6[Delta]gls2 (G0) strains. We propose that this difference is due to the underglycosylation of glycoproteins in the [Delta]alg cells (Huffaker and Robbins, 1981, 1983; Stagljar et al., 1994; te Heesen et al., 1994; Burda et al., 1996). Underglycosylation may lead to a reduced folding of proteins in the ER (Hickman et al., 1977; Copeland et al., 1988; Marquardt and Helenius, 1992; Imperiali and Rickert, 1995). The underglycosylation of glycoproteins and the impaired glucose trimming add up to the cumulative effect in stimulating the UPR in the [Delta]alg6 (G0), [Delta]alg6[Delta]gls2 (G0), and[Delta]alg10[Delta]gls2 (G2).
In eukaryotes, the UDP-glucose: glycoprotein glucosyltransferase reglucosylates proteins displaying hydrophobic patches on their surface (Fernandez et al., 1994). In S.cerevisiae, KRE5, a glucosyltransferase probably involved in cell wall biosynthesis (Meaden et al., 1990) is the only locus displaying significant sequence homology to the UDP-glucose: glycoprotein glucosyltransferase protein. However, no glycoprotein glucosyltransferase activity has been determined in S.cerevisiae. Our NLO analysis data and the UPRE-LacZ results support the view that a general glycoprotein glucosyltransferase is absent in S.cerevisiae (Fernandez et al., 1994): the [Delta]alg6 single and [Delta]alg6[Delta]gls2 double mutant cells contain no glucose residues on their lipid-linked oligosaccharides and the [Delta]gls2 mutation would prevent the trimming of putatively reglucosylated N-linked glycoproteins. We analyzed both, the LLO and NLO of the [Delta]alg6[Delta]gls2 strain. As expected the LLO consisted mainly of Man9GlcNAc2. The NLO profile showed the presence of Man9GlcNAc2 and Man8GlcNAc2 structures, but there was no evidence for glucose-containing oligosaccharides in this strain. However, a protein-specific glucosyltransferase reglucosylating N-linked high-mannose oligosaccharides of only a minor portion of glycoproteins cannot completely be excluded by our HPLC experiments since such an oligosaccharide would be below the limit of detection. Furthermore, using the UPRE-lacZ reporter under mild reducing conditions, both, the [Delta]alg6 and the [Delta]alg6[Delta]gls2 strains, induced the UPR to a similar extent, indicating that such a general glycoprotein glucosyltransferase in S.cerevisiae is very unlikely to exist. The lack of this glucosyltransferase makes a quality control system apparently leaky. The question remains how S.cerevisiae copes with such an open system. There are several possibilities to be considered. Recently, Zapun et al. (1997) have shown that the calnexin-oligosaccharide interaction prevented the glucose cleavage by glucosidase II. By such means, a G1 oligosaccharide-binding protein in the ER of S.cerevisiae (e.g., Cne1p) may reduce the cleavage rate of glucosidase II by monitoring the folding state of a bound glycoprotein. Further, a chaperone-like function of glucosidase II may also be possible. Here, glucosidase II would discriminate between folded and misfolded glycoproteins. The trimming of the innermost glucose residue would only occur if a protein has attained its correct three-dimensional structure. However, such a property needs to be demonstrated.
Our experimental system made it possible to assign a specific function of the monoglucosylated oligosaccharide in the processing of glycoproteins in the ER. However, this function is not essential, because a [Delta]gls2 strains has no detectable growth phenotype. For specific glycoproteins, e.g., CPY, folding, export, and correct targeting is not dependent on the oligosaccharide (Schwaiger et al., 1982; Winther et al., 1991) suggesting back-up systems in the process of protein maturation. However, the degradation of incorrectly folded CPY molecules in the ER requires the oligosaccharide structures (Knop et al., 1996). The genetic tailoring of these structures makes it possible to address their function in this process.
Yeast strains and manipulations
Yeast strains used are listed in Table I. Standard protocols (Guthrie and Fink, 1991) were followed for growth of yeast, mating, sporulation, and ascus dissection. If not otherwise stated, the cells were grown at 30°C in either YPD medium (2% Bacto-Peptone, 1% yeast extract, 2% glucose) or for metabolic labeling experiments overnight in MV medium (0.67% yeast nitrogen base, 2% glucose supplemented with the appropriate amino acids).
Construction of strains
Disruption of the GLS2 locus. The GLS2 locus (GenEMBL accession no. Z36098, ORF YBR229c) showed high sequence similarities to the peptide sequence of pig liver glucosidase II (Flura et al., 1997). A large portion of the N-terminal sequence of this ORF was removed by homologous recombination using a PCR product containing the complete kanamycin resistance (Wach et al., 1994) gene flanked by GLS2-specific regions. The sequence of the kanamycin resistance gene was amplified by using the pFA6a-KanMX4 plasmid (Wach et al., 1994) and by using the primers 5[prime]-ctgcgtatcttaaaatagcggtctcgaatcaaccgtatgcaactcgatgaattcgagctc and 5[prime]-tctaaaaaatgggtaccataagcccgcttggtaccaacgtgccgtacgctgcaggtcgac (boldface letters represent GLS2 sequence). The resulting PCR fragment was used to transform the haploid yeast strain SS328 selecting for resistance to G418 (200 µg/ml) (Wach et al., 1994). Transformants were analyzed for correct integration by whole cell PCR (Sathe et al., 1991) using KanMX- (5[prime]-gtattgatgttggacgag-3') and GLS2-specific primers (5[prime]-gagagctataactcaatg-3[prime], 5[prime]-gaattgtggaatactggt-3[prime]). Metabolic labeling and immunoprecipitations
The metabolic labeling was performed as described previously (Simons et al., 1995). After labeling the cells were resuspended in 50 mM Tris-HCl, pH 7.5, 1% SDS, 2 mM PMSF, disrupted by vortexing with glass beads and boiled for 5 min. The immunoprecipitation of carboxypeptidase Y (CPY) was performed as described previously (Franzusoff et al., 1991; te Heesen et al., 1992). One unit of endoglycosidase H was added where indicated and incubated at 37°C for 3 h before SDS-PAGE.
Glucosidase II assay
Logarithmically growing cells were disrupted in 0.25 M sucrose, 10 mM Tris-HCl (pH 7.5), 1.5 mM AEBSF (Calbiochem, San Diego, CA) using glass beads. The membranes were enriched by subsequent centrifugation steps discarding the supernatant (600 × g, 10 min; 8500 × g, 10 min; 100,000 × g, 1 h). The pellet was resuspended in 50 mM sodium phosphate (pH 7.0) and homogenized by passing through a G24 needle. Finally, Triton X-100 was added to 1% final concentration and the solution was subjected to ultracentrifugation 100,000 × g, 1 h). The glucosidase II activity was assayed from the resulting supernatant according to Brada and Dubach (1984) using 4-methyl-umbelliferyl-[alpha]-d-glucopyranosid as artificial substrate. The specific enzyme activity was calculated from the range of maximal velocity (incubation < 5 min). One unit enzyme activity was defined as the release of 1 µmol 4-methyl-umbelliferon per min. The protein concentrations were determined by the Bio-Rad protein assay.
Labeling, extraction and analysis of LLO and NLO
The labeling and extraction of the LLO was performed according to Zufferey et al. (1995). However, the cells were preincubated at 37°C for 30 min prior to the labeling. The labeling reaction was terminated after 30 min. The hydrolyzed LLO were digested with EndoH (2 U; Boehringer Mannheim, Germany) in 25 mM sodium citrate, pH 5.5 at 37°C for 3 h and then desalted by applying onto spin columns (Bio-Rad, Bio-Spin; Hercules, CA) filled with each 0.5 ml equilibrated Serdolit Red and Blue (Serva, Heidelberg, Germany) ion-exchanger matrix. The run-through was filtered through a 0.45 µm filter (Millipore UFC3OHV00) and stored frozen at -20°C. The HPLC analysis was performed as described previously (Cacan et al., 1993; Zufferey et al., 1995). For the extraction of the NLO, the protein pellet obtained from the LLO extraction was digested with trypsin (TPCK-treated; 1 mg/ml; Sigma, St. Louis, MO) in 0.1 M ammonium bicarbonate, pH 7.5, over night at 37°C. The digest was boiled for 10 min to destroy the trypsin. The labeled N-linked oligosaccharides were released by EndoH cleavage and processed as described above. To identify the nature of the oligosaccharide species, NLO were incubated with each 50 mU of [alpha]-mannosidase (Canavalia ensiformis, Boehringer Mannheim, Germany) in 25 mM sodium citrate (pH 5.5) at ambient temperature for 2 h. The digestion was terminated by extracting the oligosaccharides once with chloroform/methanol (5:1, v/v). The aqueous phase was dried, resuspended, filtered, and subjected to HPLC analysis.
Integration of the UPRE-LacZ reporter construct
From the plasmid pLG-[Delta]178 UPRE-Y containing the E.coli LacZ gene under the control of the unfolded protein response element of yeast KAR2/BiP (Mori et al., 1992) the 2µ element was excised using HindIII, the plasmid religated and then linearized with the StuI restriction endonuclease before integrating the plasmid into the ura3-52 locus of various strains. The correct integration of the reporter gene was tested by incubating the cells with 5 µg/ml tunicamycin for 2 h and then performing the ß-galactosidase assay (see below).
Beta-galactosidase assay
The ß-galactosidase assays were performed according to Guarente (1983) using o-nitrophenyl-ß-d-galactopyranoside (Sigma Chemicals, St. Louis, MO) as a substrate. The assays were performed in duplicates in two separate experiments.
This work was supported by the Swiss National Science Foundation grants 3100-040350 (to M.A.) and 31-40754.94 (to J.R.).
CPY, carboxypeptidase Y; EndoH, endoglycosidase H; ER, endoplasmic reticulum; LLO, lipid-linked oligosaccharide; NLO, protein N-linked oligosaccharide.
Genotype
LLO structurea
Final NLO structure2
Underglycosylation
Wild-type
Glc3Man9GlcNAc2
Man9GlcNAc2
No
[Delta]alg6
Man9GlcNAc2
Man9GlcNAc2
Yes
[Delta]alg8
Glc1Man9GlcNAc2
Man9GlcNAc2
Yes
[Delta]alg10
Glc2Man9GlcNAc2
Man9GlcNAc2
Yes
[Delta]gls2
Glc3Man9GlcNAc2
Glc2Man9GlcNAc2 (G2)
No
[Delta]alg6 [Delta]gls2
Man9GlcNAc2
Man9GlcNAc2 (G0)
Yes
[Delta]alg8 [Delta]gls2
Glc1Man9GlcNAc2
Glc1Man9GlcNAc2 (G1)
Yes
[Delta]alg10 [Delta]gls2
Glc2Man9GlcNAc2
Glc2Man9GlcNAc2 (G2)
Yes
Discussion
Materials and methods
Acknowledgments
Abbreviations
References
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 27 Jan 1998
Copyright© Oxford University Press, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  R. P Quinn, S. J Mahoney, B. M Wilkinson, D. J Thornton, and C. J Stirling A novel role for Gtb1p in glucose trimming of N-linked glycans Glycobiology, December 1, 2009; 19(12): 1408 - 1416. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Watanabe, K. Totani, I. Matsuo, J.-i. Maruyama, K. Kitamoto, and Y. Ito Genetic analysis of glucosidase II {beta}-subunit in trimming of high-mannose-type glycans Glycobiology, August 1, 2009; 19(8): 834 - 840. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Clerc, C. Hirsch, D. M. Oggier, P. Deprez, C. Jakob, T. Sommer, and M. Aebi Htm1 protein generates the N-glycan signal for glycoprotein degradation in the endoplasmic reticulum J. Cell Biol., January 12, 2009; 184(1): 159 - 172. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. N. Hebert and M. Molinari In and Out of the ER: Protein Folding, Quality Control, Degradation, and Related Human Diseases Physiol Rev, October 1, 2007; 87(4): 1377 - 1408. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. M. Wilkinson, J. Purswani, and C. J. Stirling Yeast GTB1 Encodes a Subunit of Glucosidase II Required for Glycoprotein Processing in the Endoplasmic Reticulum J. Biol. Chem., March 10, 2006; 281(10): 6325 - 6333. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. D'Alessio, J. J. Caramelo, and A. J. Parodi Absence of Nucleoside Diphosphatase Activities in the Yeast Secretory Pathway Does Not Abolish Nucleotide Sugar-dependent Protein Glycosylation J. Biol. Chem., December 9, 2005; 280(49): 40417 - 40427. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Faridmoayer and C. H. Scaman Binding residues and catalytic domain of soluble Saccharomyces cerevisiae processing alpha-glucosidase I Glycobiology, December 1, 2005; 15(12): 1341 - 1348. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. G. Frank and M. Aebi ALG9 mannosyltransferase is involved in two different steps of lipid-linked oligosaccharide biosynthesis Glycobiology, November 1, 2005; 15(11): 1156 - 1163. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. A. Wearsch, C. A. Jakob, A. Vallin, R. A. Dwek, P. M. Rudd, and P. Cresswell Major Histocompatibility Complex Class I Molecules Expressed with Monoglucosylated N-Linked Glycans Bind Calreticulin Independently of Their Assembly Status J. Biol. Chem., June 11, 2004; 279(24): 25112 - 25121. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Chavan, T. Suzuki, M. Rekowicz, and W. Lennarz Genetic, biochemical, and morphological evidence for the involvement of N-glycosylation in biosynthesis of the cell wall {beta}1,6-glucan of Saccharomyces cerevisiae PNAS, December 23, 2003; 100(26): 15381 - 15386. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Hirano, C. Zuber, J. Roth, and M. Ziak The Proteasome Is Involved in the Degradation of Different Aquaporin-2 Mutants Causing Nephrogenic Diabetes Insipidus Am. J. Pathol., July 1, 2003; 163(1): 111 - 120. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. D'Alessio, E. S. Trombetta, and A. J. Parodi Nucleoside Diphosphatase and Glycosyltransferase Activities Can Localize to Different Subcellular Compartments in Schizosaccharomyces pombe J. Biol. Chem., June 13, 2003; 278(25): 22379 - 22387. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Miseta, M. Tokes-Fuzesi, D. P. Aiello, and D. M. Bedwell A Saccharomyces cerevisiae Mutant Unable To Convert Glucose to Glucose-6-Phosphate Accumulates Excessive Glucose in the Endoplasmic Reticulum due to Core Oligosaccharide Trimming Eukaryot. Cell, June 1, 2003; 2(3): 534 - 541. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Hirano, M. Ziak, K. Kamoshita, Y. Sukenaga, S. Kametani, Y. Shiga, J. Roth, and H. Akanuma N-linked oligosaccharide processing enzyme glucosidase II produces 1,5-anhydrofructose as a side product Glycobiology, December 1, 2000; 10(12): 1283 - 1289. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Zuber, M. J. Spiro, B. Guhl, R. G. Spiro, and J. Roth Golgi Apparatus Immunolocalization of Endomannosidase Suggests Post-Endoplasmic Reticulum Glucose Trimming: Implications for Quality Control Mol. Biol. Cell, December 1, 2000; 11(12): 4227 - 4240. [Abstract] [Full Text]
|
|
|
|
|
|
M. F. Pelletier, A. Marcil, G. Sevigny, C. A. Jakob, D. C. Tessier, E. Chevet, R. Menard, J. J.M. Bergeron, and D. Y. Thomas The heterodimeric structure of glucosidase II is required for its activity, solubility, and localization in vivo Glycobiology, August 1, 2000; 10(8): 815 - 827. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. C. Tessier, D. Dignard, A. Zapun, A. Radominska-Pandya, A. J. Parodi, J. J.M. Bergeron, and D. Y. Thomas Cloning and characterization of mammalian UDP-glucose glycoprotein: glucosyltransferase and the development of a specific substrate for this enzyme Glycobiology, April 1, 2000; 10(4): 403 - 412. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. F. Cipollo and R. B. Trimble The Accumulation of Man6GlcNAc2-PP-dolichol in the Saccharomyces cerevisiae Delta alg9 Mutant Reveals a Regulatory Role for the Alg3p alpha 1,3-Man Middle-arm Addition in Downstream Oligosaccharide-lipid and Glycoprotein Glycan Processing J. Biol. Chem., February 11, 2000; 275(6): 4267 - 4277. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. D'Alessio, F. Fernandez, E. S. Trombetta, and A. J. Parodi Genetic Evidence for the Heterodimeric Structure of Glucosidase II. THE EFFECT OF DISRUPTING THE SUBUNIT-ENCODING GENES ON GLYCOPROTEIN FOLDING J. Biol. Chem., September 3, 1999; 274(36): 25899 - 25905. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Castro, L. Y. Chen, A. J. Parodi, and C. Abeijón Uridine Diphosphate-Glucose Transport into the Endoplasmic Reticulum of Saccharomyces cerevisiae: In Vivo and In Vitro Evidence Mol. Biol. Cell, April 1, 1999; 10(4): 1019 - 1030. [Abstract] [Full Text]
|
|
|
|
|
|
S. Fanchiotti, F. Fernandez, C. D'Alessio, and A. J. Parodi The UDP-Glc:Glycoprotein Glucosyltransferase Is Essential for Schizosaccharomyces pombe Viability under Conditions of Extreme Endoplasmic Reticulum Stress J. Cell Biol., November 2, 1998; 143(3): 625 - 635. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. A. Jakob, P. Burda, J. Roth, and M. Aebi Degradation of Misfolded Endoplasmic Reticulum Glycoproteins in Saccharomyces cerevisiae Is Determined by a Specific Oligosaccharide Structure J. Cell Biol., September 7, 1998; 142(5): 1223 - 1233. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Shahinian, G. J. P. Dijkgraaf, A.-M. Sdicu, D. Y. Thomas, C. A. Jakob, M. Aebi, and H. Bussey Involvement of Protein N-Glycosyl Chain Glucosylation and Processing in the Biosynthesis of Cell Wall ss-1,6-Glucan of Saccharomyces cerevisiae Genetics, June 1, 1998; 149(2): 843 - 856. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Zuber, J.-y. Fan, B. Guhl, A. Parodi, J. H. Fessler, C. Parker, and J. Roth Immunolocalization of UDP-glucose:glycoprotein glucosyltransferase indicates involvement of pre-Golgi intermediates in protein quality control PNAS, September 11, 2001; 98(19): 10710 - 10715. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||