| Glycobiology | Pages |
Recycling cell surface glycoproteins undergo limited oligosaccharide reprocessing in LEC1 mutant Chinese hamster ovary cells
Introduction
Results
Discussion
Materials and methods
Acknowledgments
Abbreviations
References
Recycling cell surface glycoproteins undergo limited oligosaccharide reprocessing in LEC1 mutant Chinese hamster ovary cells
Introduction
As newly synthesized glycoproteins traverse the cisternae of the Golgi apparatus, the trimming and processing of complex N-linked oligosaccharides takes place in a highly ordered fashion (reviewed in Kornfeld and Kornfeld, 1985). Fully processed glycoproteins are packaged in lipid vesicles and sorted from the trans-Golgi network (TGN) to lysosomes, secretory granules, or the plasma membrane (Orci et al., 1987). The TGN is also the focal point for the return of recycling cell surface glycoproteins contained in endocytic vesicles within which molecules returning to the cell surface are separated from those destined to enter maturing lysosomes (Griffiths and Simons, 1986; Gruenberg and Maxfield, 1995). The resialylation of sialidase-treated cell surface glycoproteins has been cited as biochemical evidence that the endocytic pathway includes the return to the trans-Golgi network. Using this approach, constitutive recycling between the cell surface and the TGN has been shown in several cell types for both the 46 and 215 kDa mannose-6-phosphate receptors (M6PR), low density lipoprotein receptor (LDLR), transferrin, and transferrin receptor (TfR) (Regoeczi et al., 1984; Snider and Rogers, 1985; Duncan and Kornfeld, 1988; Reichner et al., 1988; Bos et al., 1995). Elucidation of this pathway suggests that fusion of recycling vesicles with sialyltransferase-containing Golgi compartments is a normal aspect of the recycling itinerary and may permit the reprocessing of oligosaccharides on some cell surface glycoproteins.
Efforts to determine the extent to which cell surface proteins participate in an oligosaccharide reprocessing pathway and to identify additional Golgi compartments into which they might return have been controversial. Initial reports of evidence for the return of transferrin receptor (Ottosen et al., 1980; Snider and Rogers, 1986; Woods et al., 1986) to the cis-Golgi compartment were not confirmed in other studies (Volz et al., 1995; Neefjes et al., 1988). Huang and Snider (1993) found at least 10 glycoproteins return to the galactosyltransferase-containing compartment of the Golgi apparatus in an ldlD mutant of Chinese hamster ovary cells, including M6PR which was shown to be transported subsequently to a sialyltransferase-containing compartment. In contrast, using the same CHO cell line as well as a murine lymphoma cell line, Duncan and Kornfeld (1988) did not detect return of mannose 6-phosphate receptors to galactosyltransferase-containing compartments.
The present study was designed to identify endocytosed cell surface glycoproteins that are exposed to mannosidase II, a resident enzyme of the medial-Golgi compartment, and to determine the extent to which participation in this pathway is selective. These endocytosed glycoproteins were further traced to determine whether they reenter the oligosaccharide processing pathway traversed by newly synthesized glycoproteins.
The use of oligosaccharides as tracers of glycoprotein movement within a cell has several significant advantages over other techniques used to monitor intracellular trafficking of cellular proteins. Oligosaccharides are covalent modifications of cell surface proteins and as such are resistant to dissociation in acidified compartments. Furthermore, they do not perturb the recycling process (Snider and Rogers, 1985; Reichner et al., 1988; Volz et al., 1995). The use of glycosyltransferases to introduce a covalently bound monosaccharide has additional advantages including well-defined substrate specificity, absence of cell toxicity, and membrane impermeance of both sugar nucleotide donor and enzyme (Whiteheart et al., 1989). The exquisite specificity of glycosyltransferases can also be used to provide oligosaccharide structural information. Analysis of oligosaccharide structure on recycling glycoproteins can indicate subcellular compartments which are encountered during the recycling itinerary.
CHO mutants have been characterized with defects throughout the glycosylation process (Stanley and Ioffe, 1995). Lec1 mutants are defective in UDP-N-acetylglucosamine:[alpha]1,3-mannoside-[beta]1,2-N-acetylglucosaminyltransferase I (GlcNAc-TI; EC 2.4.1.101) activity (Chaney and Stanley, 1986). This structural defect has been shown to arise from a point mutation in the enzyme (Kumar et al., 1990). In these cells, all N-linked complex oligosaccharides are incompletely processed to Man5GlcNAc2Asn structures during transit through the Golgi following de novo synthesis (Robertson et al., 1978). Thus, we have used here N-acetylglucos-aminyltransferase I from porcine liver to exogenously place [3H]GlcNAc from donor UDP-[3H]GlcNAc onto acceptor cell surface glycoproteins which carry the truncated Man5GlcNAc2 structure. If cell surface glycoproteins indeed recycle to Golgi compartments, oligosaccharides with the added GlcNAc residue should then be appropriate substrates for further processing beginning with the loss of two mannose residues through the action of mannosidase II. Further transit through the medial and trans-cisternae would be readily detectable by elongation of this structure with sequential addition of N-acetylglucosamine, galactose and sialic acid residues (Kornfeld and Kornfeld, 1985).
Upon reculturing of GlcNAc-TI-treated cells, all cell surface glycoproteins that acquired [3H]GlcNAc were rapidly acted upon by intracellular mannosidase II. In contrast, no further oligosaccharide processing was evident. We conclude that the endocytic return of cell surface glycoproteins to a mannosidase II-containing compartment may be a significant aspect of the life cycle of membrane glycoproteins but reentry into the remaining oligosaccharide processing pathway may not occur to a significant extent.
Results
Phenotypic properties of wild-type (Pro-5) and mutant (Lec1) CHO cells
Altered oligosaccharide structure in the Lec1 CHO cell line has been demonstrated to be due to reduced expression of GlcNAc-TI activity (Stanley et al., 1975) and was confirmed prior to use in our studies. Sensitivity to plant lectins was determined for both cell types as described by Stanley, (1983). Pro-5 cells were resistant to Con A and sensitive to WGA whereas Lec1 mutant cells were conversely sensitive to Con A and resistant to WGA (data not shown). Endogenous GlcNAc-TI activity towards Man5GlcNAc2Asn was determined as described in Materials and methods, and found to be 8 nmol/h/mg and <0.2 nmol/h/mg for wild type and mutant cell lines, respectively.
Exogenous radiolabeling of wild-type and mutant CHO cells with porcine GlcNAc-TI
GlcNAc-TI is associated with the medial-Golgi such that a defect in action of this enzyme results in incompletely processed oligosaccharides with the structure Man5GlcNAc2Asn (Robertson et al., 1978). The patterns of lectin sensitivity as well as endogenous enzyme activity in the parental and Lec1 mutant CHO cells indicate that cell surface glycoproteins on the mutant cells bear this aborted structure. In addition, these cell surface glycoproteins should be substrates for exogenous radiolabeling with purified GlcNAc-TI, whereas most glycoproteins on the surface of wild-type cells should not. The data in Figure
Figure 1. Time course of radiolabeling with porcine N-acetylglucosaminyltransferase I (GlcNAc-TI). Cells were incubated with 3 mU porcine GlcNAc-TI and 1.39 nM UDP-[3H]GlcNAc at 4°C for the times indicated as described in Materials and methods. [3H]GlcNAc transferred to macromolecules was determined by exclusion on Sephadex G-50-80 (left). Radiolabeled macromolecules were fractionated by 5-12.5% SDS-PAGE and visualized by autofluorography (right). Each lane received an equivalent amount of total cellular protein. Lec1 cell surface glycoproteins are reprocessed in a mannosidase II-containing intracellular compartment
Glycoproteins on the Lec1 lectin-resistant CHO cell line accumulate Man5GlcNAc2 oligosaccharides since mannosidase II cannot act prior to the addition of GlcNAc to the [alpha]1,3 mannose by GlcNAc-TI (Kornfeld and Kornfeld, 1985). Some cell surface glycoproteins were made substrates for mannosidase II by exoglycosylation with porcine GlcNAc-TI and UDP-[3H]GlcNAc (as described above) at 4°C to arrest endocytosis. Radiolabeled cells were then washed and recultured at 37°C. Figure
Figure 2. Characterization of [3H]GlcNAc-labeled oligosaccharides obtained from Lec1 cells during reculturing. Cells were incubated for 3 h at 4°C with 1.39 nM UDP-[3H]GlcNAc and 3mU porcine GlcNAc-TI prior to reculturing at 37°C. Bio-Gel P-4 chromatography of oligosaccharides released by PNGase from Lec1 glycoproteins as described under Materials and methods. Arrows, Elution positions of [3H]GlcNAcMan5GlcNAc2 (right) and [3H]GlcNAcMan3GlcNAc2 (left) standards.
In addition, aliquots of the cell lysates containing radiolabeled macromolecules were treated, or not, with Endoglycosidase H (see Materials and methods) prior to analysis by SDS-PAGE (Figure
Figure 3. [3H]GlcNAc-labeled Lec1 glycoproteins become resistant to Endoglycosidase H during reculturing. Samples of lysates prepared from the radiolabeled cells used in Figure 2 were equalized on the basis of c.p.m. and were either exposed to Endo H, or not, fractionated by 5-12.5% SDS-PAGE and visualized by autofluorography (see Materials and methods). The effects of physiological inhibitors on reprocessing
In order to characterize the reprocessing pathway taken by Lec1 cell surface glycoproteins, cells were recultured under various conditions prior to analysis of total oligosaccharide molecular weight and sensitivity to Endo H digestion. Treatment of cells with the lysosomotropic amine chloroquine (Figures
Figure 4. Characterization of [3H]GlcNAc-labeled oligosaccharides obtained from Lec1 cells during reculturing in the presence of lysosomotropic amines. Cells were incubated for 3 h at 4°C with 1.39 nM UDP-[3H]GlcNAc and 3mU porcine GlcNAc-TI prior to reculturing at 37°C. Bio-Gel P-4 chromatography of oligosaccharides released by PNGase from Lec1 glycoproteins as described in Figure 2. Arrows, Elution positions of [3H]GlcNAcMan5GlcNAc2 (right) and [3H]GlcNAcMan3GlcNAc2 (left) standards.
Figure 5. [3H]GlcNAc-labeled Lec1 glycoproteins become resistant to Endoglycosidase H during reculturing in the presence of lysosomotropic amines. Samples of lysates prepared from the radiolabeled cells used in Figure 4 were equalized on the basis of c.p.m. and were either exposed to Endo H or not, fractionated by 5-12.5% SDS-PAGE, and visualized by autofluorography.
The treatment of Lec1 cells with 20 µM monensin not only permitted the trimming of N-linked oligosaccharides, but also the kinetics were somewhat accelerated (Figures
The enzymatic nature of the reprocessing of cell surface oligosaccharides was demonstrated when cells were incubated in the presence of 29 µM swainsonine, a potent inhibitor of Golgi mannosidase II (Tulsiani et al., 1982). The molecular weight profile of isolated cellular oligosaccharides remained unchanged and no increase in resistance to Endo H digestion was seen (data not shown).
Reduced temperature has been shown to delay or block processes of vesicle transport and fusion (Haylett and Thilo, 1991). In order to demonstrate that transport to an intracellular compartment was necessary for exposure of cell surface glycoproteins to mannosidase II, cells were recultured at 15°C and at 0-4°C, i.e., on ice. Both temperatures effectively blocked the trimming of total oligosaccharides as well as the acqusition of resistance to Endo H of cellular macromolecules (data not shown). This is not due to the inhibition of mannosidase II activity at lower temperatures. Lysates of radiolabeled cells were maintained at on ice either in the presence or absence of swainsonine for 30 min prior to isolation of N-linked oligosaccharides and analysis by P-4 gel filtration (see Materials and methods). Significant manno-sidase II activity was evident on ice within this short incubation period (data not shown). Additionally, Snider and Rogers (1985) demonstrated the sialylation of newly synthesized Tfr at 18°C, indicative of Golgi transport and activity of Golgi enzymes at reduced temperature. Therefore, exposure of cell surfaceglycoproteins to mannosidase II requires internalization and a membrane fusion event. Mannosidase II activity is not extracellular
Reculturing radiolabeled Lec1 cells at temperatures known to inhibit endocytosis and fusion of lipid vesicles prevented exposure of cell surface glycoproteins to the action of mannosidase II. To confirm the intracellular localization of this enzyme experiments were designed to detect activity present on the cell surface or released into the culture medium. Ovalbumin was radiolabeled with GlcNAc-TI and [3H]GlcNAc, dried onto nitrocellulose, and incubated overnight with either whole cells, cell lysate, culture medium, or culture medium conditioned from incubation with viable cells. Following incubation, the ovalbumin was harvested by trypsin digestion and oligosaccharide sizing analysis carried out as described above for cellular oligosaccharides. The data presented in Figure
Figure 6. Mannosidase II activity is not extracellular. [3H]GlcNAc-Ovalbumin was dried onto nitrocellulose and incubated for 16 h at 37°C with either (A) intact Lec1 cells, (B) culture medium, (C) lysate of Lec1 cells, or (D) culture medium conditioned by intact Lec1 cells. Shown are Bio-Gel P-4 profiles of released oligosaccharides.
Elongation of Lec1 oligosaccharides
Following the removal of two mannose residues, the Golgi-mediated processing of complex oligosaccharides would then predict the addition of N-acetylglucosamine by GlcNAc-TII followed by penultimate galactose and terminal sialic acid additions. However, these processing events are not evident based on the size distribution of total cell surface oligosaccharides seen by analysis on Bio-Gel P-4. The capacity of Lec1 cells to further process recycled oligosaccharides beyond the action of mannosidase II was examined. As described in Materials and methods, the experiment shown in Figure
Figure 7. Elongation of [3H]GlcNAc-labeled Lec1 cell surface glycoproteins. All samples were prepared as described under Materials and methods. Shown are Bio-Gel P-4 profiles of oligosaccharides isolated from lysates of [3H]GlcNAc-labeled Lec1 cells that were incubated at 37°C for the times indicated. Elution positions of [3H]-galactosylated GlcNAc polymers from [3H]Gal-GlcNAc1 to [3H]Gal-GlcNAc6 are indicated at the top of the figure.
This study was prompted by findings reported by us and others that plasma membrane glycoproteins can recycle to elements of the Golgi apparatus and that this pathway can be detected by acquired alterations in oligosaccharide structure (Regoeczi et al., 1984; Snider and Rogers, 1985; Duncan and Kornfeld, 1988; Reichner et al., 1988; Bos et al., 1995). The main objective of the present study was to determine whether recycling glycoproteins which return to the Golgi can reenter the same oligosaccharide processing pathway followed by newly synthesized glycoproteins. We designed a system which would first detect exposure of cell surface glycoproteins to mannosidase II, a component of the medial aspects of the Golgi, and then demonstrate further structural alterations consistent with transport through the later cisternae. Findings reported herein demonstrate an endocytic pathway from the plasma membrane to a mannosidase II-containing compartment with no further processing of N-linked oligosaccharides.
Multiple lines of evidence demonstrate that following addition of [3H]GlcNAc by GlcNAc-TI, radiolabeled cell surface glycoproteins are internalized into a mannosidase II-containing compartment. Evidence was provided by inhibition of the conversion of glyco-proteins to Endo H-resistant forms by swainsonine and by comigration of isolated cellular oligosaccharides with oligosaccharide standards on BioGel P-4 columns consistent with the loss of two hexose units (Figure
One plausible explanation for the absence of oligosaccharide processing beyond the action of mannosidase II could include a defect in the ability of Lec1 cells to translocate UDP-GlcNAc into the Golgi for GlcNAc-TII function. The best evidence to demonstrate that Lec1 cells are competent for UDP-GlcNAc translocation is that transfection of Lec1 cells with human cDNA containing the GlcNAc-TI gene conferred expression of GlcNAc-TI activity and binding of L-PHA/sRBC (Kumar and Stanley, 1989).
The data in Figure
The encounter of recycling molecules with mannosidase II may have taken place in an intracellular compartment other than the medial-Golgi such that additional oligosaccharide processing could not have taken place. Precedence for an intracellular structure which contains Golgi-derived components has been described by Griffiths et al. (1988) and Brown et al., (1986). Griffiths demonstrated the transient colocalization of the endocytic marker [alpha]2-macroglobulin along with M6PR and a lysosomal membrane glycoprotein (lgp) in normal rat kidney cells. Since [alpha]2-macroglobulin and lgp are detectable lysosomal components and M6PR is not, the structure in which they were colocalized was hypothesized to be a specialized endosome that serves as an intermediate compartment into which endocytic vesicles discharge their contents upon fusion. Whether this compartment is a common destination of recycling plasma membrane glycoproteins was not determined, but such a Golgi-derived structure may be analogous to the mannosidase II-containing compartment described in the present report. Furthermore, Brown et al. (1986) found accumulation of M6PR in endosomes of clone 9 hepatocytes upon treatment with weak base. Upon removal of the base, M6PR reappeared in the Golgi indicating that part of the intracellular trafficking of the M6PR includes recycling between the Golgi and endosomes. Together with our finding that mannosidase II trimming continued in the presence of weak base, these data further support the coincident trafficking of M6PR and mannosidase II between Golgi and endosomes. Pfeffer and coworkers have confirmed that M6PR indeed recycles between late endosomes and the TGN in vivo and that this pathway is central to lysosome biogenesis (Riederer et al., 1994). Whereas the kinetics of exposure to mannosidase II described herein is relatively slow as compared to that expected of sequential transfer from the cell surface to the endosome, the extra-Golgi localization of mannosidase II is a plausible explanation of our findings when considered as a nonvectorial salvage pathway of glycoproteins. Finally, cell surface glycoproteins bearing high mannose-type oligosaccharides were shown to return to an intracellular compartment containing mannosidase I activity with subsequent return to the cell surface (Porwoll et al., 1998). In findings similar to those reported herein, these investigators found no further oligosaccharide processing following trimming, suggesting that these glycoproteins return to intracellular compartments peripheral to the medial-Golgi.
The consideration that endocytosed cell surface glycoproteins were exposed to a unique isoform of mannosidase II gains interest in light of such a finding in mice lacking the mannosidase II gene (Chui et al., 1997). Nonerythroid cells in these mice were found to express hybrid oligosaccharides despite absence of the Golgi mannosidase II, a process attributed to the newly described [alpha]-mannosidase III. That this is not identical to the enzymatic activity shown by the Lec1 cells in the present study is evident by the fact that [alpha] mannosidase activity was relatively swainsonine resistant and Man5GlcNAc but not GlcNAcMan5GlcNAc served as substrates. However, that an analogous variant of mannosidase II is expressed by Lec1 cells cannot be ruled out at this time.
It is conceivable that following transit through endosomes the glycoproteins followed in this study may have been degraded in lysosomes and the mannosidase II activity being shown may actually be located in mature lysosomes. A direct lysosomal pathway is unlikely based upon the data shown. Whereas more thorough degradation by lysosomal glycosidases would be expected to reveal smaller oligosaccharide structures shown by P-4 profiles over time in culture, lysosomal dismantling might not be detected if [beta]-hexosaminidase acts to remove the [3H]GlcNAc tracer. However, in this latter case we would have seen a strong quantitative loss of labeled oligosaccharides which was not observed. The use of lysosomotropic reagents including monensin, chloroquine, and ammonium chloride all permitted mannosidase II activity with kinetics that were either unaltered or somewhat accelerated. Since the optimal pH of lysosomal [alpha]-d-mannosidase is 4.6, increased intracellular pH should have a demonstrable effect on the activity of this enzyme (Opheim and Touster, 1978). In contrast, Golgi mannosidase II has an optimum pH of 5.5-6.0 and could function adequately under treatment with the drugs used here which would be expected to raise the pH of intracellular vesicles to ~6.0. However, the lack of effect of lysosomotropic agents taken together with the maintenance of distinct bands on SDS-PAGE (Figure
In summary, numerous studies in a variety of cell types have traced the endocytic return of cell surface glycoproteins to the trans-Golgi network, thereby demonstrating that this is a fundamental recycling itinerary which can result in intermingling components of the endocytic and biosynthetic pathways. Issues which remain more controversial include whether this pathway is selective for some but not all cell surface glycoproteins, the precise Golgi compartments into which these proteins return, whether oligosaccharide reprocessing defines reentry into the biosynthetic machinery, and the extent to which these issues are specific to individual cell types. The present study employed a novel approach to studying this pathway by combining the specificity of glycosyltransferase probes to radiolabel cell surface glycoproteins with a CHO mutant cell line to trace the reentry of cell surface glycoproteins into and through the medial-Golgi. Cell culture
Parental (Pro-5) and lectin-resistant mutant (Pro-5 PhaR1 1-1; Lec1) Chinese hamster ovary cell lines (Stanley et al., 1975) were obtained from Dr. Pamela Stanley (Department of Cell Biology, Albert Einstein College of Medicine, Bronx, NY) by Dr. Sharon Krag (Department of Biochemistry, School of Hygiene, Johns Hopkins University, Baltimore, MD) who provided them to our laboratory with permission. Cells were maintained as monolayers in DMEM (GIBCO, Grand Island, NY) supplemented with 10% heat-inactivated fetal bovine serum (FBS, GIBCO) and incubated at 37°C, 97% humidity, and 5% CO2. The sensitivity of cells to the lectins wheat germ agglutinin (WGA; agglutinin from Triticum vulgaris; Sigma Chemical Co., St. Louis, MO) and concanavalin A (Con A; agglutinin from Canavalia ensiformis, Pharmacia) was determined exactly as described by Stanley (1983). Endogenous GlcNAc-TI activity in Pro-5 and Lec1 mutant CHO cells was determined as described previously (Chaney and Stanley, 1986). Preparation of enzyme
An [alpha]1-3 mannoside [beta]1-2 N-acetylglucosaminyl-transferase (GlcNAc-TI) was purified from porcine liver as described previously (Oppenheimer and Hill, 1981; Oppenheimer et al., 1981). The enzyme was stored at -20°C in 50% glycerol and 0.5% Triton X-100. In order to remove detergent prior to incubation with intact cells, the enzyme was diluted 1:4 with labeling buffer (25 mM Hepes, 0.075 M NaCl, 0.1 M glucose, 10 mg/ml bovine serum albumin fraction V, fatty acid free, and 1 mM MnCl2, pH 6.5) containing 0.1% n-octyl [beta]-d-glucopyranoside (OG) essentially as described previously (Whiteheart and Hart, 1987). The enzyme was then applied to a 0.3 × 25 cm column containing G-50 Sephadex resin overlaid with 1 cm of UDP-Sepharose affinity resin (11 µmol/ml). The column was washed with buffer and the enzyme eluted in the void volume with 3 M NaCl and 0.1% OG. Enzyme assay
GlcNAc-TI activity was assayed using a modification of the method described by Oppenheimer and Hill, (1981). A solution of 60 mg/ml ovalbumin (Sigma) was prepared in GlcNAc-TI assay buffer (stored as a 5× buffer stock, 0.25 M sodium cacodylate, pH 6.5, 2.5% Triton X-100, and 50 mM MnCl2) for use as the acceptor. Ten microliters of enzyme (diluted 1:5 in ovalbumin solution) was added to 30 µl ovalbumin solution and warmed to 37°C. The reaction was initiated by the addition of 10 µl UDP-N-acetyl-d-[3H]glucosamine (UDP-[3H]GlcNAc; 5 mM; 37 c.p.m./pmol; New England Nuclear) which was also prepared in ovalbumin solution. The reaction was stopped with 10 µl of 20% SDS, 50 mM EDTA, 5 mg/ml blue dextran, and samples were applied to Sephadex G-50 minicolumns prepared in 1 ml plastic tuberculin syringe barrels. Columns were washed through with 50 mM ammonium formate, 0.1% SDS; 80 µl fractions were collected; and the void volume was quantitated by scintillation counting. One unit of enzyme is defined by the transfer of 1 µmol UDP-GlcNAc to ovalbumin/min incubation at saturatingconcentrations of substrates. Radiolabeling of cells
The vectorial labeling of Lec1 cells was accomplished using 5 × 106/sample which were suspended in 100 µl labeling buffer containing 3 mU purified GlcNAc-TI and 4 µCi UDP-[3H]GlcNAc and kept at 4°C throughout the incubation. Cells were washed three times in PBS/ 5 mM glucose/0.25% BSA, solubilized in lysis buffer (10 mM Tris-HCl, pH 7.4, 1.5 mM MnCl2, 150 mM NaCl, 0.5% Nonidet P-40, 0.1 mM phenylmethylsulfonyl fluoride, 1% aprotinin, and 8 µg/ml swainsonine) for 30 min on ice and clarified by centrifugation at 1500 × g for 20 min. Swainsonine (Boehringer Mannheim Biochemicals, Indianapolis, IN) was added to preventmannosidase II activity during the lysis process. Preliminary experiments showed this activity to be detectable by P-4 analysis (data not shown). Supernatants were made 1% SDS, boiled, and stored at -80°C. Radiolabeled macromolecules were collected in the void volume of Sephadex G-50-80 columns (1.1 × 31 cm; Sigma) in 50 mM ammonium formate with 0.1% SDS. Aliquots of each collected fraction were suspended in scintillation fluid (Formula 963, Du Pont-New England Nuclear) and counted on a Packard (Prias model) liquid scintillation counter. Remaining material was pooled and concentrated by lyophilization. Resolubilized samples to be analyzed by SDS-PAGE were precipitated in cold acetone at -20°C, resuspended in 100 µl SDS-PAGE sample buffer, boiled, and applied to linear 5-12.5% SDS-polyacrylamide gradients. Gels were fixed in 50% methanol, treated with EN3HANCE (Du Pont-New England Nuclear), and dried. Fluorography was carried out on Kodak X-Omat AR film (Eastman Kodak, Rochester, NY) at -80°C. Reculturing of radiolabeled cells
Lec1 cells were radiolabeled (as described above) and washed three times in ice cold PBS, 5 mM glucose, 0.25% BSA. Aliquots of 5 × 106 cells were resuspended in 10 ml DMEM containing 10% FBS and returned to culture in 100 mm tissue culture plates at 37°C, 5% CO2. When inhibitors were used they were added to the culture medium prior to addition of the cells. Swainsonine was added from a 2 mg/ml stock solution and used at a final concentration of 29 µM; choroquine (125 µM, Sigma) and ammonium chloride (20 mM, Sigma) were added directly to the culture medium. Monensin (20 µM, Sigma) was added from a 10 mM stock solution in 95% EtOH. Cell viability was maintained at >90% in the presence of these inhibitors at the doses used. The doses of monensin, chloroquine, and ammonium chloride were selected on the basis of previous reports which showed complete alkalinization of intracellular acid compartments at these or lesser concentrations (Poole and Ohkuma, 1981; Maxfield, 1982; Wileman et al., 1984). Cells to be recultured at lower temperatures were placed in 75 mm Falcon tissue culture flasks and exposed to blood gas mixture, and the sealed flasks were maintained either on ice or in a 15°C recirculating water bath. Oligosaccharide analysis
In some experiments, radiolabeled macromolecules were treated with Endoglycosidase H (Miles Scientific, Naperville, IL) prior to SDS-PAGE. Following precipitation in cold acetone at -20°C, these samples were aliquoted on the basis of radioactivity and were resuspended in 50 µl Endo H buffer (0.1 M citric acid/ NaOH, pH 6.0, 0.5% SDS, 0.5% [beta]-mercaptoethanol, NaN3) and treated with 10 µl of either Endoglycosidase H (from 1 U/ ml stock) or Endo H buffer in a 37°C water bath for 16 h. An equal volume of 2× SDS-PAGE sample buffer was added to stop the reaction, the samples were boiled for 10 min and applied to SDS-PAGE as described above.
Sizing analysis of N-linked oligosaccharides isolated from radiolabeled macromolecules was performed as follows.Lyophilized macromolecules (see above) were resuspended in water and precipitated in 6 volumes of 95% EtOH at -20°C and pelleted by centrifugation. Pellets were solubilized in 50 µl peptide:N-glycosidase F (PNGase F) buffer A (3% SDS and 3% [beta]-mercaptoethanol) and boiled for 10 min. Two volumes of PNGase F buffer B (150 mM NaPO4, pH 9.3, 75 mM EDTA and 15% NP-40) were added in order to bind excess SDS. PNGase F was prepared from cultures of Flavobacterium meningosepticum exactly as described previously (Tarentino et al., 1985). Two additions of 1 unit PNGase F were used per sample during the course of incubation at 37°C for 18 h. Samples were boiled for 10 min to terminate the reaction, 50 µl 20% SDS was added to aid solubilization, and fractionation was performed on Sephadex G-50-80 in column buffer (50 mM ammonium formate, 0.1% SDS, NaN3). Radioactivity in aliquots of each fraction was determined by liquid scintillation counting, and remaining material that was in the included volume was pooled and lyophilized. At least 85% of the radioactivity in macromolecular material was released into the included volume by the action of PNGase F. Oligosaccharides were resuspended in dH2O and excess SDS removed by precipitation on ice prior to application on a Bio-Gel P-4 column (-400 mesh, 1 × 200 cm). Dextran and galactose were added to each sample as indicators of V0 and Vi, respectively. The location of the standards was determined by the phenol sulfuric acid assay (Chaplin, 1986). Elongation of [3H]GlcNAc-labeled Lec1 cell surface glycoproteins
Lec1 cells were harvested, washed and radiolabeled on ice with porcine GlcNAc-TI and UDP-[3H]GlcNAc exactly as described above. Cells were washed three times in ice cold PBS/5 mM glucose/0.25% BSA, resuspended in tissue culture medium and recultured for 10 h at 37°C, 5% CO2. Cells were then collected and lysed in lysis buffer containing 1% NP-40, pH 7.4 at 108 cells/ml, and postnuclear supernatants were prepared. Aliquots containing 107 cell equivalents were combined with 10 µl of a 10× glycosyltransferase substrate cocktail (50 mM MnCl2, 30 mM CMP-N-acetylneuraminic acid, 200 mM 5[prime]-AMP, and 50 mM UDP-galactose) prepared in cell lysis buffer. To some samples UDP-GlcNAc was added to 10 mM final concentration. Samples were incubated in a 37°C water bath for 1 or 2 h, made 1% SDS to stop the reaction and boiled for 10 min. Oligosaccharide sizing analysis was performed as described above. Experiments were conducted at pH 7.4 in order to minimize glycosidase activity released in cell lysates. Preparation of radiolabeled oligosaccharides from ovalbumin
Radiolabeling of 2 mg ovalbumin (Sigma) was performed in the presence of 10 µCi dried UDP-[3H]GlcNAc and 10 µl GlcNAc-TI taken directly from stock. Radiolabeled macromolecules were separated on a column of G-25 Sephadex (Sigma) in 0.1 M ammonium acetate, 20% EtOH, pH 5.0. The void volume was pooled, lyophilized, and treated with 500 µg trypsin (Sigma) for 5 h at 37°C followed by boiling to inactivate the enzyme. Peptides were treated with PNGase F as described above, and oligo-saccharides were collected from the included volume on a column of Sephadex G-50 and lyophilized. These oligosaccharides migrate on Bio-Gel P-4 with a Kav = 0.42 and are presumed to consist of [3H]GlcNAcMan5GlcNAc2 based on both the migration on P-4 (Kobata et al., 1987) and the specificity of the GlcNAc-TI (Harpaz and Schachter, 1980). An aliquot of these oligosaccharides was treated with 26 µl jack bean [alpha]-mannosidase (Boehringer) for 2 h at 37°C. This generated a predominant peak on P-4 analysis with a Kav = 0.49, which is consistent with the loss of two mannose residues and is therefore presumed to consist of [3H]GlcNAcMan3GlcNAc2. Other methods
Protein was determined by the bicinchoninic acid assay (Pierce Chemical Co.). [3H]-Galactosylated GlcNAc polymers used as standards for Bio-Gel P-4 analysis were prepared as described previously (Bangs et al., 1988).
We thank Drs. Pamela Stanley and Wally Whiteheart for helpful discussions and critical reading of the manuscript. This work was supported in part by NIH Grants CA 42486 (G.W.H.) and GM51493 (J.S.R); J.S.R. received funds allocated to the Department of Surgery by the Rhode Island Hospital.
CHO, Chinese hamster ovary cell line; Endo H, endo-[beta]-N-acetylglucosaminidase; DMEM, Dulbecco's modified Eagle's medium; SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gel electrophoresis; HPLC, high pressure liquid chromatography; PNGase F, peptide:N-glycosidase F; GlcNAc-TII, UDP-N-acetyl-glucosamine:[alpha]1,3-mannoside-[beta]2-N-acetylglucosaminyltransferase II; M6PR, mannose-6 phosphate receptor; TfR, transferrin receptor; ASGPR, asialoglycoprotein receptor; MHC, major histocompatibility complex.
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: 19 Dec 1998
Copyright©Oxford University Press, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
This Article 
Abstract
FREE Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (2)
Request Permissions
Disclaimer
Google Scholar
Articles by Reichner, J. S.
Articles by Hart, G. W.
Search for Related Content
PubMed
PubMed Citation
Articles by Reichner, J. S.
Articles by Hart, G. W.
Social Bookmarking
What's this?