HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by D'Alessio, C. Articles by Parodi, A. J. Search for Related Content PubMed PubMed Citation Articles by D'Alessio, C. Articles by Parodi, A. J. Social Bookmarking                      What's this?

J Biol Chem, Vol. 274, Issue 36, 25899-25905, September 3, 1999

Genetic Evidence for the Heterodimeric Structure of Glucosidase II
THE EFFECT OF DISRUPTING THE SUBUNIT-ENCODING GENES ON GLYCOPROTEIN FOLDING^*

Cecilia D'Alessio^§, Fabiana Fernández^¶, E. Sergio Trombetta^**, and Armando J. Parodi

From the  Instituto de Investigaciones Bioquímicas Fundación Campomar, Antonio Machado 151, 1405 Buenos Aires, Argentina and the  Department of Cell Biology, Yale University School of Medicine, New Haven, Connecticut 06520-8002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

It has been proposed that in rat and murine tissues glucosidase II (GII) is formed by two subunits, GII and GII, respectively, responsible for the catalytic activity and the retention of the enzyme in the endoplasmic reticulum (ER). To test this proposal we disrupted genes (gls2⁺ and gls2⁺) encoding GII and GII homologs in Schizosaccharomyces pombe. Both mutant cells (gls2 and gls2) were completely devoid of GII activity in cell-free assays. Nevertheless, N-oligosaccharides formed in intact gls2 cells were identified as Glc₂Man₉GlcNAc₂ and Glc₂Man₈GlcNAc₂, whereas gls2 cells formed, in addition, small amounts of Glc₁Man₉GlcNAc₂. It is suggested that this last compound was formed by GII transiently present in the ER. Monoglucosylated oligosaccharides facilitated glycoprotein folding in S. pombe as mutants, in which formation of monoglucosylated glycoproteins was completely (gls2) or severely (gls2 and UDP-Glc:glycoprotein:glucosyltransferase null) diminished, showed ER accumulation of misfolded glycoproteins when grown in the absence of exogenous stress as revealed by (a) induction of binding protein-encoding mRNA and (b) accumulation of glycoproteins bearing ER-specific oligosaccharides. Moreover, the same as in mammalian cell systems, formation of monoglucosylated oligosaccharides decreased the folding rate and increased the folding efficiency of glycoproteins as pulse-chase experiments revealed that carboxypeptidase Y arrived at a higher rate but in decreased amounts to the vacuoles of gls2 than to those of wild type cells.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Glucosidase II (GII)¹ plays a pivotal role in the processing of N-oligosaccharides in the endoplasmic reticulum (ER). The oligosaccharide transferred from the dolichol-P-P-derivative to Asn residues in nascent polypeptide chains (Glc₃Man₉GlcNAc₂) is first deglucosylated by glucosidase I (GI), which removes the more external glucose unit, and then by glucosidase II (GII), which excises both remaining glucose units. The UDP-Glc:glycoprotein glucosyltransferase (GT) then transfers a single glucose unit to glucose-free oligosaccharides (1, 2). According to the model proposed for the quality control of glycoprotein folding in mammalian cells, monoglucosylated oligosaccharides formed either by partial deglucosylation of the transferred compound or by GT-mediated reglucosylation are recognized by two ER lectins, membrane-bound calnexin or its soluble homolog calreticulin (3). Further deglucosylation of the oligosaccharides by GII liberates the glycoproteins from their lectin anchors. The oligosaccharides are then reglucosylated by GT and thus recognized again by the lectins, only when linked to misfolded protein moieties as it has been found that GT only glucosylates glycoproteins not displaying their proper tertiary structures (2, 4). The deglucosylation-reglucosylation cycle continues until proper glycoprotein folding is achieved.

The calnexin/calreticulin-monoglucosylated glycoprotein interaction was called quality control of glycoprotein folding as it was found to be one of the several existing alternative mechanisms by which cells retain misfolded species in the ER and eventually transport them to the cytosol where they are degraded in the proteasomes (5). In addition, the above mentioned interaction facilitates glycoprotein folding by preventing aggregation and formation of non-native disulfide bonds (6, 7). As a result of this interaction, a decrease in glycoprotein folding rate but an increase in folding efficiency were observed (6).

All evidence for the proposed mechanism of quality control of glycoprotein folding was derived from experiments performed in mammalian cell systems (3), but presumably a similar mechanism is operative in Schizosaccharomyces pombe as this yeast expresses a GT activity and a calnexin homolog (8-11). In addition, it has been reported that the oligosaccharide transferred to nascent polypeptides in S. pombe is Glc₃Man₉GlcNAc₂, the same as in most eukaryotic cells, and that this compound is processed in the ER to Man₉GlcNAc₂, thus indicating the presence of both GI and GII activities (12). We have recently reported that GT-mediated formation of monoglucosylated oligosaccharides is essential for S. pombe viability under conditions of severe ER stress such as underglycosylation of glycoproteins caused by the alg6 mutation and high temperature (Man₉GlcNAc₂ and not Glc₃Man₉GlcNAc₂ is transferred in alg6 mutants) (13). It was proposed that folding of a glycoprotein involved in cell wall formation was affected in gpt1/alg6 double mutants as the wild type phenotype could be restored not only upon transfection with a GT-encoding expression vector but also in a hyperosmotic growth medium (1 M sorbitol). No evidence was presented, however, indicating that monoglucosylated oligosaccharides actually facilitated glycoprotein folding in this yeast.

Purification of GII from rat and murine tissues yielded two tightly bound polypeptides (named GII and GII) that could not be separated by procedures commonly used for enzyme purification without loss of the enzymatic activity. Subunit GII (a 104-kDa soluble protein) had a certain homology to other glucosidases and no ER retrieval signal at its C terminus, whereas this last feature was found in subunit GII, a 58-kDa soluble polypeptide with no sequence similarity to other proteins (14, 15). It was proposed that GII was GII catalytic subunit, whereas GII was responsible for its ER retention, but no actual evidence for this proposal was presented. The proposal remained controversial, however, as no such dimeric structure was found in GII purified from pig liver and kidney and from plant tissues (16-18). Moreover, although a GII homolog devoid of an ER retrieval sequence is encoded in the Saccharomyces cerevisiae genome, no protein with significant homology to GII and having an ER retrieval sequence at its C terminus is encoded in the DNA of this budding yeast (14).

The purpose of work reported here is to present genetic support for the proposed heterodimeric structure of GII as well as for the occurrence of monoglucosylated oligosaccharide-mediated glycoprotein folding facilitation in low eukaryote cells.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Strains and Medium-- Escherichia coli DH5 was used for cloning procedures. E. coli XL1-Blue MRF' and XLOLR (Stratagene) were used for phage screening and plasmid excision, respectively. The S. pombe wild type strains used for disrupting gls2⁺ and gls2⁺ genes were h⁹⁰, ura4-D18, ade6-M216, leu1-32, and h, ura4-D18, ade6-M210, leu1-32, respectively. S. pombe gpt1 mutant was described previously (9). Bacteria were grown in LB medium, 0.5% NaCl, 1% tryptone (Difco), 0.5% yeast extract (Difco), and 100 µg/ml ampicillin and 40 µg/ml kanamycin if necessary, or in the same medium supplemented with 0.2% maltose and 10 mM MgSO₄. S. pombe was grown in YE medium (3% glucose, 0.5% yeast extract supplemented with adenine). The minimal medium used, supplemented with Ade and Leu, was described before (19).

Identification and Disruption of the GII- and GII-encoding Genes (gls2⁺ and gls2⁺, Respectively)-- Identification and disruption of the gls2⁺ gene was performed, and the resulting mutant strain (gls2) was genetically characterized as described before for construction of the gls2/alg6 double mutant (13, 20). To sequence gls2⁺ gene to its C-terminal end, the 998-bp fragment used for gls2⁺ disruption (20) was used as probe for screening S. pombe genomic DNA library constructed in ZAP (10). Five independent phages were isolated, and insert-containing plasmids (pBK-CMV) were recovered from them. All plasmids had inserts of about 4,500 bp. Sequencing them revealed that GII ended with PQLFLV at its C terminus.

For disrupting the gls2 gene, exact primers (Bls, sense: 5'AGATGAAGTTCAGTCAATGG3' and Bla, antisense: 5'ATATTATAACTCATCGACAG3') were designed according to positions 27-46 and 1533-1552 of an S. pombe cDNA sequence encoding a protein homologous to rat and mouse GII (GenBank^TM accession number D89245). A PCR reaction using S. pombe genomic DNA as template yielded a fragment of approximately 1,800 bp, which was larger than the expected size of 1,526 bp. The 1,800-bp fragment was cloned in pGEMT vector, and approximately 200 bp from both ends were sequenced. The sequences corresponded to those of the D89245 sequence. This indicated the presence of at least one intron. Reverse transcriptase-PCR was performed on total S. pombe RNA using primer Bla for reverse transcription and the same primer and Bls for PCR. A single 1,526-bp band was obtained, thus confirming the presence of introns. To facilitate introduction of the ura4⁺ marker gene in the 1,526-bp fragment, two primers were designed according to the D89245 sequence (BaHindIII, positions 595-614, antisense: 5'AATCCGTTGAAAGCTTCATC3' and BsHindIII, positions 592-611, sense: 5'GAAGATGAAGCTTTCAACGG3'). These primers introduce a HindIII site through a G-T substitution. Two PCRs were then performed using S. pombe cDNA as template and either primers Bls and BaHindIII or Bs HindIII and Bla to obtain fragments N (588 bp) and C (961 bp), respectively. Both were separately cloned in pGEMT. Band N was liberated from the vector with SacII (site in the vector) and HindIII (site in the insert). The C-containing vector (pGEMT-C) was linearized with the same enzymes. Fragment N was then ligated to linearized pGEMT-C to form pGEMT-NC. This vector has a point mutation in position 577 of the insert that introduces the single HindIII site in the construct. This was linearized with HindIII and ligated to the ura4⁺ gene. The ura4⁺ had been liberated from pBluescript with the same enzyme. The DNA fragment containing the GII-encoding fragment interrupted with ura4⁺ (3,290 bp) was liberated from the vector with NotI and transfected into strain SpAD: h, ura4-D18, leu1-32, ade6-M210, kindly provided by Angel Durán, University of Salamanca, Spain. Transformants (ura4⁺) were selected in plates with minimal medium supplemented with Ade and Leu. Resulting strain was called SpADII. Its genotype was h, ura4-D18, leu1-32, ade6-M210, gls2::ura4⁺. PCRs performed on 5 independent transformant colonies using sense and antisense primers, respectively, complementary to the 1,526-bp gls2⁺ fragment (primer Bls) and to ura4⁺ (primer UraAN, 5'TTTTCATCCCCTCAGCTC3') yielded, in 4 colonies, fragments that were larger than the expected 1,180-bp fragment. This indicated that in those cells homologous recombination had occurred downstream of the intron in genomic DNA. The fifth colony and the construct yielded fragments of the expected size. This indicated that in this single colony recombination had occurred upstream of the intron in genomic DNA or, alternatively, that a non-homologous recombination had occurred. No fragment was synthesized when wild type DNA was used as template. The existence of the intron in 4 colonies and its absence in one of them was confirmed by Southern blotting analysis. Genomic DNAs from wild type cells, from the putative intron-containing mutants, and from the putative intron-free mutant were digested with BglI and NcoI. The probe used for Southern analysis consisted of two 192- and 260-bp gls2⁺ fragments that were contiguous in wild type DNA but that were flanking the 5' and 3' ends of ura4⁺, respectively, in the construct used for gene disruption. They were generated by digestion of pGEMT-NC with XhoI and HindIII. The expected fragment sizes in wild type, intron-containing, and intron-free mutant DNAs were about 1000 (746 plus intron), 2780 (2510 plus intron), and 2510 bp, respectively. Those were precisely the sizes found.

RNA Procedures-- Cells were grown under normal conditions (YE medium at 28 °C), and RNA was extracted from cells in exponential phase of growth at the same optical density and submitted to Northern blotting as performed previously (10). Probes used were DNA fragments synthesized by PCR using a genomic DNA as template and had 1004 and 973 base pairs from BiP- and actin-encoding genes, respectively.

DNA Procedures-- Standard DNA manipulations and hybridization conditions were carried out as described (21).

Materials-- [¹⁴C]Glucose (250 Ci/mol) and [³⁵S]Met plus [³⁵S]Cys (>1000 Ci/mmol, EasyTag Express protein labeling mix) were from NEN Life Science Products. Jack bean -mannosidase, dithiothreitol, protein A-Sepharose, and endo--N-acetylglucosaminidase H (Endo H) were from Sigma. Anti-S. pombe carboxypeptidase Y (CPY) serum was a generous gift from Dr. K. Takegawa, Kagawa University, Japan.

Substrates and Standards-- [glucose-¹⁴C]Glc₁Man₇GlcNAc, [glucose-¹⁴C]Glc₁Man₈GlcNAc,[glucose-¹⁴C]Glc₁Man₉GlcNAc, [glucose-¹⁴C]Glc₃Man₉GlcNAc, [glucose-¹⁴C]Glc₂Man₉GlcNAc, [glucose-¹⁴C]Glc₁Man₄GlcNAc, [glucose-¹⁴C]Glc₁Man₅GlcNAc, [¹⁴C]Man₈GlcNAc, and [¹⁴C]Man₉GlcNAc were prepared as described previously (8). Treatment of [glucose-¹⁴C]Glc₂Man₉GlcNAc with jack bean -mannosidase generated [glucose-¹⁴C]Glc₂Man₄GlcNAc.

Methods-- Strong acid hydrolysis and treatment of oligosaccharides with jack bean -mannosidase were as described previously (8). Short term (15 min) in vivo labeling of S. pombe cells with [¹⁴C]glucose and purification of labeled Endo H-sensitive oligosaccharides were performed as described previously (8) for S. cerevisiae cells, but 100 µCi of [¹⁴C]glucose and no 1-deoxynojirimycin were used. For pulse-chase labeling, cells were incubated with [¹⁴C]glucose for 60 min after which 1 M unlabeled glucose up to a 0.1 M final concentration was added, and incubation was prolonged for an additional 30 min. Where indicated dithiothreitol up to a 5 mM final concentration was added 5 min before addition of [¹⁴C]glucose. Pulse-chase labeling of cells with [³⁵S]Met and [³⁵S]Cys as well as CPY immunoprecipitations were performed as already described (22). Whatman 1 papers were used for chromatographies (8). Solvents employed were as follows: solvent A, 1-propanol/nitromethane/water (5:2:4), or solvent B, 1-butanol/pyridine/water (10:3:3). GI and GII activities were assayed as described previously (23) using [glucose-¹⁴C]Glc₃Man₉GlcNAc and [glucose-¹⁴C]Glc₁Man_7-9GlcNAc as substrates, respectively.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Construction and Genetic Analysis of gls2 and gls2 Mutant Strains-- As mentioned above, it was proposed that in mammalian tissues GII is composed by two subunits, called GII and GII, respectively, responsible for the catalytic activity and the ER retention of the first subunit (14, 15). To verify this proposal the genes that are homologous to those encoding GII and GII subunits in mammalian cells were identified and individually disrupted in S. pombe. The GII-encoding gene (here called gls2⁺) was disrupted, and the mutant was genetically characterized as described previously (13, 20) for construction of the gls2/alg6 double mutant. We have now sequenced the 3' end of gls2⁺ and verified that S. pombe gls2p (GII) lacks, as its mammalian counterpart, an ER retrieval signal as it ends with PQLFLV (see "Experimental Procedures"). The GII-encoding gene (here called gls2⁺) was disrupted and genetically characterized as described under "Experimental Procedures." It is worth mentioning that gls2p displays a signal peptide and an ER retrieval sequence (VDEL) at its N and C termini, respectively.

Cell-free Biochemical Characterization of gls2 and gls2 Cells-- Microsomes prepared from wild type and mutant cells were incubated with either [glucose-¹⁴C]Glc₃Man₉GlcNAc or [glucose-¹⁴C]Glc₁Man_7-9GlcNAc to probe for GI and GII activities, respectively. Results depicted in Fig. 1, A and B, show that labeled glucose units were liberated from Glc₃Man₉GlcNAc by wild type, gls2, and gls2 microsomes. On the other hand, microsomes from wild type but not from mutant strains liberated glucose from Glc₁Man_7-9GlcNAc (Fig. 1C). These results showed, therefore, that cells from both mutant strains displayed GI but not GII activity.

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Cell-free determination of GI and GII activities. Microsomes from wild type (open circles), gls2 (filled circles), and gls2 (gray circles) cells were incubated with [glucose-14C]Glc3Man9GlcNAc (A and B) or with [glucose-14C]Glc1Man7-9GlcNAc (C). Ordinates represent the percentages of glucose liberated. One hundred percent corresponds to 1200 cpm.

In Vivo Biochemical Characterization of Mutants-- Wild type and both gls2 and gls2 mutant cells were pulsed for 15 min with [¹⁴C]glucose. Total cell proteins were degraded with an unspecific protease, and Endo H-sensitive oligosaccharides were released from glycopeptides thus obtained. The wild type strain mainly yielded substances that migrated on paper chromatography as Man₈GlcNAc and Man₉GlcNAc standards (Fig. 2A). Strong acid hydrolysis of substances migrating as either one of the standards only yielded labeled mannoses (not shown).

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Patterns of protein-linked oligosaccharides. Wild type (WT) (A), gls2 (B), and gls2 (C) cells were incubated with [14C]glucose for 15 min. Total cell proteins were degraded with an unspecific protease, and Endo H-sensitive oligosaccharides were liberated from glycopeptides thus formed. Oligosaccharides were then run on paper chromatography with solvent A. Standards are as follows: 2, Glc2Man9GlcNAc; 1, Glc1Man9GlcNAc; 9, Man9GlcNAc; and 8, Man8GlcNAc.

A different oligosaccharide pattern was obtained in the case of the mutant strains; only two compounds that migrated as Glc₂Man₉GlcNAc and Glc₁Man₉GlcNAc standards appeared in the chromatogram. A higher proportion of the faster migrating compound was observed in gls2 than in gls2 cells (Fig. 2, B and C). Strong acid hydrolysis of either one of oligosaccharides synthesized by both mutant cells yielded glucose and mannose units. The monosaccharide patterns yielded by substances that migrated as the Glc₂Man₉GlcNAc standard in Fig. 2, B and C, are depicted in Fig. 3, A and B, respectively. Exhaustive -mannosidase treatment of compounds that migrated as Glc₂Man₉GlcNAc and Glc₁Man₉GlcNAc standards synthesized by gls2 cells yielded, for both oligosaccharides, a compound that migrated as a Glc₂Man₄GlcNAc standard, differently from both Glc₁Man₄GlcNAc and Glc₁Man₅GlcNAc standards (Fig. 3, C and D, respectively). This indicated that the compound that migrated as a Glc₁Man₉GlcNAc standard in Fig. 2B was Glc₂Man₈GlcNAc. Results shown in Figs. 2 and 3 indicate, therefore, that in gls2 cells only the external glucose unit was removed from the transferred oligosaccharide (Glc₃Man₉GlcNAc₂) and that in some cases a mannose residue was also excised. The mutant cells were, therefore, totally unable to form in vivo monoglucosylated oligosaccharides either by partial deglucosylation of Glc₃Man₉GlcNAc₂ or by reglucosylation of glucose-free compounds.

View larger version (28K): [in this window] [in a new window]   Fig. 3.   Characterization of oligosaccharides, monosaccharide composition, and -mannosidase degradation. Saccharides that migrated as the Glc2Man9GlcNAc standard in Fig. 2B (gls2 cells) (A) and in Fig. 2C (gls2 cells) (B) were submitted to strong acid hydrolysis followed by paper chromatography with solvent B. Standards are as follows: 1, mannose; 2, glucose; and 3, galactose. Saccharides that migrated as the Glc2Man9GlcNAc standard in Fig. 2B (gls2 cells) (C) and Fig. 2C (gls2 cells) (E) were treated with -mannosidase and run on paper chromatography with solvent A. Saccharides that migrated as the Glc1Man9GlcNAc standard in Fig. 2B (gls2 cells) (D) and Fig. 2C (gls2 cells) (F) were similarly treated. Standards are as follows: 4, Glc1Man4GlcNAc; 5, Glc2Man4GlcNAc; and 6, Glc1Man5GlcNAc.

On the other hand, an -mannosidase treatment of the oligosaccharide synthesized by gls2 cells that migrated as a Glc₂Man₉GlcNAc standard (Fig. 2C) yielded the same as that synthesized by gls2 cells, a compound that migrated as a Glc₂Man₄GlcNAc standard (Fig. 3E), but the compound that migrated as a Glc₁Man₉GlcNAc standard in Fig. 2C yielded, with the same enzymatic treatment, compounds that migrated as Glc₂Man₄GlcNAc and Glc₁Man₄GlcNAc standards (Fig. 3F). This indicated that the compound that migrated as a Glc₁Man₉GlcNAc standard in Fig. 2C was a mixture of Glc₂Man₈GlcNAc and Glc₁Man₉GlcNAc.

From results shown in Figs. 1-3, it was concluded that gls2p (GII) is the catalytic subunit of GII as absolutely no GII activity was detected in both in vitro and in vivo assays in gls2 mutants. On the other hand, gls2p (GII) represents a subunit that retains gls2p (GII) in the ER, as the extremely low GII activity that could be detected in in vivo but not in in vitro assays in gls2 cells probably represented gls2p that was transiently present in the ER before secretion.

No differences in the growth rates of wild type and gls2 cells at 18, 28, and 39 °C were observed. In addition, no morphological differences were observed between both strains under a light microscope.

Cells Defective in Monoglucosylated Oligosaccharide Formation Accumulate Glycoproteins Bearing ER-specific Oligosaccharides-- It has been reported that in S. pombe wild type cells Glc₃Man₉GlcNAc₂ is processed in the ER mainly to Man₉GlcNAc₂ (but formation of small amounts of Man₈GlcNAc₂ was also observed) and further elongated in the Golgi by the addition of mannose and galactose units (12, 13, 20). It may be speculated that if monoglucosylated oligosaccharides facilitated glycoprotein folding, mutants in which formation of those compounds were totally (gls2) or partially (UDP-Glc:glycoprotein glucosyltransferase null, called gpt1) prevented would show an accumulation of misfolded glycoproteins bearing ER-specific oligosaccharides. Wild type and mutant cells were incubated for 60 min in the presence of minute amounts of [¹⁴C]glucose followed by a 30-min chase with an excess of the unlabeled monosaccharide to allow Golgi elongation of secreted (properly folded) glycoprotein molecules. Different patterns of protein-linked oligosaccharides were obtained in wild type and mutant cells. Whereas in the former cells Man₉GlcNAc₂ and Man₈GlcNAc₂ represented a small proportion of the protein-linked oligosaccharides (Fig. 4A), in both gls2 and gpt1 mutants the ER-specific oligosaccharides (Glc₂Man₉GlcNAc₂ and Glc₂Man₈GlcNAc₂ in gls2 and Man₉GlcNAc₂ and Man₈GlcNAc₂ in gpt1) were the main compounds present.

View larger version (23K): [in this window] [in a new window]   Fig. 4.   Pulse-chase incubations of wild type (WT), gls2, and gpt1 cells. Wild type (A and D), gls2 (B), and gpt1 (C) cells were incubated with [14C]glucose for 60 min followed by a 30-min chase with an excess of the unlabeled monosaccharide. Endo H-liberated oligosaccharides were run on paper chromatography with solvent A. D, 5 mM dithiothreitol (DTT) was added 5 min before the label. Standards are as follows: 2, Glc2Man9GlcNAc; 1, Glc1Man9GlcNAc; 9, Man9GlcNAc; and 8, Man8GlcNAc.

To confirm that indeed misfolding of glycoproteins resulted in the accumulation of protein species bearing ER-specific oligosaccharides, wild type cells were incubated as above but in the presence of 5 mM dithiothreitol. It has been shown that this reagent prevents correct folding of disulfide bond-containing glycoproteins (20, 24, 25). Contrary to what was found on incubation in the absence of dithiothreitol (Fig. 4A), in the presence of the drug wild type cells accumulated Man₉GlcNAc₂- and Man₈GlcNAc₂-containing glycoproteins (Fig. 4D). Three oligosaccharides larger than the above mentioned compounds were observed. They probably correspond to glycoproteins not having disulfide linkages.

The Unfolded Protein Response in Wild Type, gls2, gls2, and gpt1 Mutant Strains-- Accumulation of misfolded proteins in the S. pombe ER leads to the so called unfolded protein response, that is to the induction of mRNAs coding for chaperones and other proteins that facilitate proper folding of recently synthesized species (9-11, 26). If monoglucosylated oligosaccharides were involved in glycoprotein folding facilitation as indicated by the model proposed for the quality control of glycoprotein folding, it would be expected that an increased synthesis of folding facilitating proteins would occur in gls2, gls2, and gpt1 cells grown under non-stressed conditions.

We determined by Northern blotting analysis the amounts of the mRNAs coding for BiP, the more abundant ER chaperone, and for a constitutively expressed protein (actin) present in wild type, gls2, gls2, and gpt1 cells grown in the absence of exogenous stress (28 °C in a rich medium). It has been reported that in S. pombe, accumulation of misfolded species in the ER triggers synthesis of two BiP mRNAs, one of them having the same size as the only one present in wild type cells grown under normal conditions and the other having a slightly smaller size (26). Northern blotting analysis of total mRNA using a portion of the BiP-encoding gene as probe revealed two bands in gls2 , gls2, and gpt1 but only one in wild type cells (Fig. 5A). Results were scanned, and intensities obtained for BiP mRNAs were normalized with those obtained for actin mRNA. Ratios obtained for wild type cells were taken as 1. In three independent experiments the ratios of BiP/actin mRNAs were 1.8, 2.6, and 2.2 for gls2 and 2.4, 2.7, and 2.1 for gpt1 cells. In two experiments, the ratios obtained for gls2 cells were 2.1 and 2.2 (Fig. 5B).

View larger version (40K): [in this window] [in a new window]   Fig. 5.   Expression of BiP-encoding mRNA in wild type and mutant cells. A, wild type (WT), gls2, gls2, and gpt1 cells were grown in YE medium at 28 °C, and mRNA was extracted and submitted to Northern blotting analysis. Both larger signals correspond to BiP mRNAs and the smaller ones to actin (Act) mRNAs. B, results shown in A were scanned in less exposed photos in order to ensure the proportionality of the intensities. The ratios of BiP/actin mRNAs intensities obtained are represented in the ordinates. The value obtained for wild type cells was taken as 1.

It may be concluded, therefore, that under normal growth conditions misfolded proteins accumulated in the ER of S. pombe cells in which absolutely no monoglucosylated oligosaccharides (gls2 mutants) or reduced levels of them (gls2 mutants) were formed or in cells in which monoglucosylated oligosaccharides were not formed by GT-mediated reglucosylation (gpt1 mutants).

Processing of Carboxypeptidase Y (CPY) in gls2 Cells-- It has been reported that interaction with calnexin/calreticulin delays folding velocity but increases folding efficiency of hemagglutinin translated in a rabbit reticulocyte-dog pancreas microsome system (6). To test if the same occurs in S. pombe, intact wild type and gls2 cells were pulsed with [³⁵S]Met and [³⁵S]Cys for 5 min. Samples were then withdrawn after different chase periods, and S. pombe CPY was immunoprecipitated from them. It has been reported that the ER form of S. pombe CPY is a 110-kDa protein that is proteolytically processed in the vacuole to a 51-kDa species (22). This in turn is formed by two polypeptides of 19 and 32 kDa held together by a disulfide bridge. The antiserum specifically recognized the latter fragment. Immunoprecipitates were submitted to SDS-polyacrylamide gel electrophoresis under reducing conditions followed by autoradiography. The gels were scanned and intensities of the 32-kDa fragment were normalized relative to amount of label precipitated by 10% trichloroacetic acid from each sample.

In Fig. 6A, the relative intensities of the CPY fragment of the 30-min chase samples of either wild type or gls2 cells were taken as 100%. It may be observed that CPY arrived faster to the vacuole in gls2 than in wild type cells. On the other hand, in Fig. 6B the relative intensity of the 30-min sample of wild type cells was taken as 100% for samples obtained from either wild type or gls2 cells. It may be observed that the amount of CPY that arrived to the vacuole in wild type cells was approximately double than that in gls2 mutants. It may be concluded, therefore, that as described for mammalian cells also in intact S. pombe cells formation of monoglucosylated oligosaccharides reduces the folding rate and increases folding efficiency (6).

View larger version (13K): [in this window] [in a new window]   Fig. 6.   CPY processing in wild type and gls2 cells. Wild type and gls2 cells were pulsed for 5 min with [35S]Met and [35S]Cys and chased for the indicated times. CPY was immunoprecipitated and run on 10% SDS-polyacrylamide gel electrophoresis under reducing conditions. Autoradiographies were scanned and intensities normalized respective to label precipitated by 10% trichloroacetic acid from each sample. A, the relative intensities of the 30-min chase samples of either wild type (filled circles) or gls2 (open circles) cells were taken as 100%. B, the relative intensity of the wild type cell 30-min chase sample was taken as 100% for all samples.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

As mentioned above, it has been proposed that GII is a heterodimer composed by catalytic (GII) and ER retention (GII) subunits (14, 15). This proposal remained controversial as such structure was detected in mouse T-lymphoma cell and rat liver enzyme preparations but not in GII purified to homogeneity from plants and pig liver and kidney (16-18). In addition, although the GII homolog in S. cerevisiae also lacks an ER retrieval sequence and shows significant similarity to its mammalian counterpart, no protein with significant homology to GII and having an ER retrieval sequence was found in the budding yeast genome (14). The possibility exists, therefore, that the heterodimeric structure could be restricted to mouse and rat tissues or, alternatively, that it could be an artifact created during enzyme purification.

We herein present genetic evidence for the heterodimeric structure of GII. Genes coding for either GII or GII homologs in S. pombe were disrupted, and no GII activity was detected in in vitro assays performed with microsomal fractions isolated from the respective mutants. On the other hand, GI activity in both mutant microsomal fractions had levels similar to those found in wild type cells. Despite the total absence of GII activity revealed by in vitro assays, structural analysis of protein-linked oligosaccharides synthesized in intact wild type and mutant cells pulsed with [¹⁴C]glucose confirmed that GII is indeed responsible for catalysis and GII for GII ER retention; although Glc₂Man₉GlcNAc₂ and Glc₂Man₈GlcNAc₂ were the only compounds synthesized in the gls2 mutant, formation of small amounts of Glc₁Man₉GlcNAc₂ was detected in gls2 cells. It may be speculated that the latter compound was formed by gls2p transiently present in the ER before secretion. The notion that gls2p (GII) is the catalytic subunit agrees with the fact that it shows significant similarity to the so-called family 31 and 9 glucosidases (27). Although GII displays an ER retrieval sequence (VDEL) and as such its primary role is probably that of retaining GII in the ER, it cannot be discarded that it might also contribute to the stability of the latter subunit. It is worth mentioning that a GII homolog has been detected also in the human genome (14, 15). This fact as well as results presented in the present report strongly suggest that GII is a heterodimer in most if not all eukaryotic cells. The presence of ER-retaining subunits has been already described for prolyl 4-hydroxylase and the triglyceride transfer proteins that lack ER retrieval signals (28, 29). In both cases the retaining protein (protein disulfide isomerase) is essential for cell viability, thus precluding confirmation of its retaining role by gene disruption.

Retention of two glucose units in protein-linked oligosaccharides is not expected to cause misfolding of glycoproteins per se as it has been determined that oligosaccharides are mainly processed in S. pombe ER to Man₉GlcNAc₂ and that the tertiary structure of this compound is identical whether it has one, two, or no glucoses (12, 30). In addition, GII and GII are minor lumenal ER proteins, so their absence is not expected to trigger the unfolded protein response for unspecific reasons. Nevertheless, gls2 and gls2 mutant cells grown in the absence of exogenous stress accumulated misfolded proteins in the ER as judged by the induction of BiP-encoding mRNA. That misfolding of glycoproteins (and therefore BiP induction) in gls2 and gls2 mutants was not caused by a deleterious effect of the two retained glucoses on the tertiary structure of the protein moieties but by a deficiency in the formation of monoglucosylated oligosaccharides was also indicated by the fact that the unfolded protein response was observed in gpt1 mutants, that is in cells that lacked GT-mediated reglucosylation. No diglucosylated oligosaccharides accumulated in those mutant cells. Induction of BiP-encoding mRNA in the absence of exogenous stress was shown to occur also in GII-defective mammalian cells (31). However, it cannot be concluded that BiP mRNA induction in the latter cells was an exclusive consequence of the lack of GII because those mutants had been obtained by selecting cells resistant to Phaseolus vulgaris leukoagglutinin after treatment with chemical mutagens (32, 33). On the contrary, Southern blotting analysis of all mutants employed in the present report (gls2, gls2, and gpt1) showed that the marker gene employed (ura4⁺) had always only disrupted the intended target. Misfolding glycoproteins in cells deficient in monoglucosylated oligosaccharide formation was also revealed by the much higher proportion of ER-specific, protein-linked oligosaccharides detected in gls2 and gpt1 mutants when compared with wild type cells.

Formation of monoglucosylated oligosaccharides decreased the folding rate but increased the folding efficiency in S. pombe, the same as in mammalian cells, because monitoring of CPY processing indicated that this glycoprotein arrived early but in diminished amounts to the vacuoles of gls2 than to those of wild type cells. This result and the already mentioned accumulation of misfolded glycoproteins in the ER of cells grown under non-stressed conditions but having impaired formation of monoglucosylated oligosaccharides confirm that those compounds are indeed involved in glycoprotein folding facilitation in S. pombe. The absence of a discernible phenotype in gls2, gls2, and gpt1 cells may be explained by the up-regulation of BiP and probably of other chaperones and folding-assisting proteins that compensate for the null or lower amount of monoglucosylated oligosaccharides formed. As mentioned above, only under extreme stress conditions (glycoprotein underglycosylation and high temperature) the formation of monoglucosylated glycoproteins was found to be required for S. pombe viability (13).

No induction of BiP mRNA was observed in S. cerevisiae GII-deficient mutants grown under non-stressed conditions (34, 35). This result, as well as the absence of GT, the sensor of misfolded protein conformations, from those cells and the existence of a calnexin homolog with striking structural variations in the domain that recognizes monoglucosylated oligosaccharides (P-domain) (8, 34, 36, 37) casts doubts on the occurrence of a quality control mechanism of glycoprotein folding in S. cerevisiae similar to that described for mammalian cells.

	FOOTNOTES

^* This work was supported by National Institutes of Health Grant GM44500, by Howard Hughes Medical Institute Grant 75197-553502, by the University of Buenos Aires, and by the Argentine Federal Government (Consejo Nacional de Investigaciones Cientificas y Técnias and Agencia Nacional de Promocion Cientifica y Tecnologia).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^§ Doctoral Fellow.

^¶ Post-doctoral Fellow.

^** A Jane Childs Coffin Post-doctoral Fellow.

Career Investigator of the National Research Council (Argentina) and a Howard Hughes Medical Institute International Research Scholar. To whom correspondence should be addressed: Instituto de Investigaciones Bioquímicas Fundación Campomar, Antonio Machado 151, 1405 Buenos Aires, Argentina. Tel.: 54-11 4863-4011; Fax: 54-11 4865-2246; E-mail: aparodi@iib.uba.ar.

	ABBREVIATIONS

The abbreviations used are: GII, glucosidase II; BiP, binding protein; CPY, carboxypeptidase Y; Endo H, endo--N-acetylglucosaminidase H; ER, endoplasmic reticulum; GI, glucosidase I; GT, UDP-Glc:glycoprotein glucosyltransferase; PCR, polymerase chain reaction; bp, base pair.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				J. J. Caramelo and A. J. Parodi Getting In and Out from Calnexin/Calreticulin Cycles J. Biol. Chem., April 18, 2008; 283(16): 10221 - 10225. [Full Text] [PDF]

				C.-M. Cruciat, C. Hassler, and C. Niehrs The MRH Protein Erlectin Is a Member of the Endoplasmic Reticulum Synexpression Group and Functions in N-Glycan Recognition J. Biol. Chem., May 5, 2006; 281(18): 12986 - 12993. [Abstract] [Full Text] [PDF]

				F. Reyes, L. Marchant, L. Norambuena, R. Nilo, H. Silva, and A. Orellana AtUTr1, a UDP-glucose/UDP-galactose Transporter from Arabidopsis thaliana, Is Located in the Endoplasmic Reticulum and Up-regulated by the Unfolded Protein Response J. Biol. Chem., April 7, 2006; 281(14): 9145 - 9151. [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]

J. Schirawski, H. U. Bohnert, G. Steinberg, K. Snetselaar, L. Adamikowa, and R. Kahmann Endoplasmic Reticulum Glucosidase II Is Required for Pathogenicity of Ustilago maydis PLANT CELL, December 1, 2005; 17(12): 3532 - 3543. [Abstract] [Full Text] [PDF]