SEARCH

SEARCH BY CITATION

Keywords:

  • Gangliosides;
  • Glycosyltransferases;
  • α2,8-Sialyltransferase;
  • GD3 synthase;
  • Golgi compartments;
  • N-Glycosylation

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. SDS-polyacrylamide gel electrophoresis (PAGE) and immunoblotting
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

Abstract: GD3 synthase (Sial-T2) is a key enzyme of ganglioside synthesis that, in concert with GM2 synthase (GAlNAc-T), regulates the ratio of a- and b-pathway gangliosides. In this work, we study the sub-Golgi location of an epitope-tagged version of chicken Sial-T2 transfected to CHO-K1 cells. The expressed protein was enzymatically active both in vitro and in vivo and showed a molecular mass of ∼47 or ∼95 kDa on sodium dodecyl sulfate-polyacrylamide gel electrophoresis in the presence or absence of, respectively, β-mercaptoethanol. The 95-kDa form of Sial-T2 was also detected if the protein was retained in the endoplasmic reticulum (ER) due to impaired glycosylation, indicating that it was formed in the ER. Confocal immunofluorescence microscopy showed Sial-T2 localized to the Golgi complex and, within the organelle, partially co-localizing with the mannose-6-phosphate receptor, a marker of the trans-Golgi network (TGN). In cells treated with brefeldin A, a major fraction of Sial-T2 redistributed to the ER, even under controlled expression to control for mislocalization due to protein overloading. In experiments of incorporation of sugars into endogenous acceptors of Golgi membranes in vitro, GD3 molecules formed by incubation with CMP-NeuAc were converted to GD2 upon incubation with UDP-GalNAc. These results indicate that Sial-T2 localizes mainly to the proximal Golgi, although a fraction is located in the TGN functionally coupled to GalNAc-T. Consistent with this, most of the enzyme was in an endoglycosidase H (Endo-H)-sensitive, neuraminidase (NANase)-insensitive form. A minor secreted form lacking ∼40 amino acids was Endo-H-resistant and NANase-sensitive, indicating that the cells were able to process N-glycans to Endo-H-resistant forms. Taken together, the results of these biochemical and immunocytochemical experiments indicate that in CHO-K1 cells, most Sial-T2 localizes in the proximal Golgi and that a functional fraction is also present in the TGN.

Ganglioside synthesis is regulated during cell differentiation, development, and oncogenic transformation (Panzetta et al., 1980; Hakomori, 1981; van Echten and Sandhoff, 1993; Zeng et al., 1995). The relative activities of two transferases, GD3 synthase (Sial-T2) and GM2 synthase (GalNAc-T), are considered relevant during these processes. Sial-T2 catalyzes the addition of sialic acid in α 2[RIGHTWARDS ARROW]8 linkage to the sialic acid moiety of the ganglioside GM3 to form ganglioside GD3. GalNAc-T catalyzes the incorporation of GalNAc in β 1[RIGHTWARDS ARROW]4 linkage to the galactose moiety of GM3, GD3, and LacCer to form GM2, GD2, and GA2, respectively. The expression of the genes coding for these transferases is under transcriptional (Yamamoto et al., 1996; Daniotti et al., 1997b) and posttranscriptional (Takamiya et al., 1995; Daniotti et al., 1997a; Bieberich et al., 1998) regulation. These enzymes are membrane-bound residents of the Golgi complex, which work in succession on the transfer of sugars from soluble sugar nucleotide donors to membrane-bound growing glycolipid acceptors (for review, see Maccioni et al., 1999). The assemblage of the machinery for synthesis is still under investigation, and the possibility that the topological organization may also participate in the modulation of ganglioside expression at the cell surface is worth considering.

The localization of ganglioside glycosyltransferases along the Golgi complex subcompartments has been examined using pharmacological and biochemical criteria. In vivo studies on chick embryo neural retina cells indicated that synthesis of GM3, GD3, and GT3 occurs in the proximal Golgi compartment (Rosales Fritz and Maccioni, 1995), with GM3 and GD3 synthesis displaced toward the cis and medial compartments and GT3 synthesis toward the trans compartment (Rosales Fritz et al., 1996). On the other hand, in vitro studies in intact Golgi membrane preparations from the same cells (Maxzúd et al., 1995; Maxzúd and Maccioni, 1997) found GD3 synthesis functionally coupled to GM2 synthesis by GalNAc-T, an activity mapped to the trans-Golgi network (TGN) in neural (van Echten et al., 1990) and nonneural (Young et al., 1990; Giraudo et al., 1999) cells. Determination of ganglioside glycosyltransferase activities in liver subcellular fractions obtained by centrifugation in sucrose gradients led to the proposal that they are ordered in the order in which they act (Trinchera and Ghidoni, 1989; Trinchera et al., 1990). However, similar experiments but using different in vitro assay conditions led to the proposition that all these transferases reside in the late Golgi (Lannert et al., 1998).

In the present work, we study the sub-Golgi location of chicken Sial-T2 transfected to the cell line CHO-K1. We examined the immunocytochemical co-localization of an epitope-tagged version of the enzyme with the TGN marker mannose-6-phosphate receptor (M6PR), its functional co-localization with the TGN-located GalNAc-T, and its N-glycan-processing status. Results indicate that most of the enzyme localizes in the proximal Golgi and that a functional fraction is also present in the TGN.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. SDS-polyacrylamide gel electrophoresis (PAGE) and immunoblotting
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

Expression plasmids

The plasmid pCEFL-GD3 synthase was prepared as indicated (Daniotti et al., 1997b). To express the carboxy-terminally epitope-tagged Sial-T2 with the nonapeptide epitope of the viral hemagglutinin (HA), a 1.1-kb DNA fragment including the chick Sial-T2 coding region was amplified by PCR using primer oligonucleotides (sense 5′-CGCGGATCCGCCGCCACCATGGCGGGGCTGGCG-3′ and antisense 5′-TCTGTCGACGAATTCGGCAGAAGACTGGGTTAA-3′) and pCEFL-GD3 synthase as template. The sense primer contains a BamHI restriction site and Kozak initiation sequence (Kozak, 1987) upstream of the initiation codon for proper initiation of translation. At the antisense primer, the stop codon TAA was deleted and a SalI site was incorporated in-frame with the SalI site upstream of the HA sequence. The fragment was purified from 1% agarose gels using the Genclean II (Bio 101) method, digested with the appropriate restriction enzyme, and ligated using T4 ligase into pCEFL-HA, a modified pCEFL vector encoding the HA nonapeptide epitope YPYDVPDYA (Sells and Chernoff, 1995), to generate pCEFL—Sial-T2—HA. The plasmid pIND—Sial-T2—HA, which drives the expression of the HA-tagged enzyme using an ecdysone-inducible promoter, was engineered by subcloning into the plasmid pIND (Invitrogen, CA, U.S.A.) the fragment BamHI—NotI taken from pCEFL—Sial-T2—HA, which codes for the tagged enzyme. The identity of each construct was confirmed by restriction mapping and the final construct by DNA sequencing (Medigene Sequencing Services, Germany).

Cell transfection and generation of stably transfected cell clones

CHO-K1 cells (Chinese hamster ovary cells; ATCC, U.S.A.) grown in Dulbecco's modified Eagle's medium (DMEM)—10% fetal calf serum (FCS) (GibcoBRL) during 12 h at 37°C in 5% CO2 were transfected with 1 μg/dish pCEFL—HA or pCEFL—Sial-T2—HA using Lipofectamine (GibcoBRL). Twelve to 15 h later, the cells were washed with phosphate-buffered saline (PBS) and either fixed with methanol to —20°C for immunostaining or harvested for enzyme activity determinations using 10 mM Tris-HCl buffer (pH 7.2) containing 0.25 M sucrose (buffer A).

To generate stable transfectants, pCINeo human GalNAc-T (Giraudo et al., 1999) and pCEFL—Sial-T2—HA were independently transfected into CHO-K1 cells using Lipofectamine. Following 48 h of expression, the cells were cultured in DMEM containing 10% FCS and 1 mg/ml geneticin (G418). Colonies of stable transfectants were screened for GM1 expression (cholera toxin binding) or GD3 ganglioside expression and maintained in 0.5 mg/ml G418. To generate stable CHO-K1 cells expressing Sial-T2 under the control of an ecdysone-inducible promoter, we first transfected the cells with the plasmid pVgRXR (Invitrogen), which encodes the hormone receptor. Following 48 h of expression, the cells were expanded in DMEM containing the antibiotic zeocin. Zeocin-resistant clones were screened for inductive response by transient transfection with the plasmid pIND—Sial-T2—HA, as reporter plasmid, and later induction with 2 μM of the ecdysone analogue ponasterone. Cells stably expressing the hormone receptor were used to prepare stable cell lines expressing Sial-T2 by transfection with the plasmid pIND—Sial-T2—HA and selection in G418. Colonies of stable transfectants were isolated and screened for Sial-T2 protein expression by western blot, enzyme activity, and immunolocalization of the product GD3 at the cell surface. A clone (clone IST2A) having the lowest basal enzyme activity and the highest inducibility with ponasterone was selected.

Labeling of gangliosides in vivo and in vitro

For metabolic labeling of gangliosides, 20 μCi/ml [3H]galactose was added to the culture medium of CHO-K1 cells 1 day after transfection, and the culture was extended for 12 h. For labeling of endogenous gangliosides of Golgi membranes in vitro, CHO-K1 cells were suspended in buffer A, washed twice in the same solution, and homogenized. A microsomal membrane fraction (containing Golgi membranes) was collected by centrifugation between 1,000 g× 10 min and 100,000 g× 1 h. Two-step labeling of endogenous ganglioside acceptors (Maxzúd et al., 1995) was carried out as follows: CHO-K1 cell membranes (15 μg of protein) were incubated for 1 h in an incubation system that contained, in a final volume of 20 μl, 15 mM MnCl2, 10 μM CMP-[3H]NeuAc (7,500 cpm/pmol) or 10 μM CMP-NeuAc (as will be indicated), 25 mM HEPES—KOH (pH 7.0), 25 mM KCl, and 2.5 mM magnesium acetate. Membranes were then freed from incubation medium ingredients by layering onto 0.6 M sucrose and centrifugation at 13,000 g for 6 min at 4°C. The pellet was resuspended to the original incubation volume with the ingredients lacking CMP-NeuAc but containing 10 μM UDP-GalNAc or 10 μM UDP-[3H]GalNAc (as will be indicated) and incubated for 2 h at 37°C.

Chromatographic analysis of radioactive gangliosides

Membranes after metabolic labeling of cells with [3H]galactose or incubation systems after labeling in vitro of membrane-bound endogenous acceptors were extracted with 1 ml of chloroform/methanol/water (3:48:47 by volume), and glycolipids were purified using Sep-Pak C18 cartridges (Waters Corp., U.S.A.). Total radioactivity incorporated was determined in an aliquot by liquid scintillation spectrometry. For chromatographic analysis, the lipid extract was supplemented with GM3 (from canine spleen), GM2 (from Tay Sachs brain), and bovine brain gangliosides as internal standards and run on HPTLC using chloroform/methanol/0.25% CaCl2 (45:45:10 by volume) as solvent (Maxzúd et al., 1995). Positions of the internal standards were determined by exposure of the plate to iodine vapors. Fluorography was carried out by dipping the plate in 2-methylnaphthalene containing 0.4% 2,5-diphenyloxazole and further exposure of the plate to an x-ray film for 4-8 days at -70°C.

Immunoprecipitation and glycosidase digestions

Stably transfected cells expressing Sial-T2—HA were lysed for 60 min in ice with buffer B (50 mM Tris—HCl, pH 7.2, 0.6% Triton X-100, 140 mM NaCl, 3 μg/ml leupeptin, 1 mM phenylmethylsulfonyl fluoride, 3 μg/ml aprotinin, 1 mM EDTA, 0.05% sodium azide). Lysates were preabsorbed with protein G—agarose beads (75% suspension, washed with buffer B prior to use; Pharmacia-Biotech, Sweden) for 60 min at room temperature and then were incubated overnight on a rotating wheel at 4°C with monoclonal antibody anti-HA (1:50) and with 50 μl of protein G—agarose beads (Martina et al., 1998). Beads were pelleted by centrifugation at 2,500 g for 10 s, washed five times at 4°C with buffer B, washed three times with buffer C (100 mM phosphate buffer, pH 7.2, 10 mM EDTA), and resuspended in 50 μl of buffer C. Immunoprecipitates of Sial-T2—HA released by cells to the culture medium were obtained following the same procedures starting from a 100,000-g, 2-h culture medium supernatant concentrated 10- to 15-fold using Centricon 10 (Amicon). For N-Glycanase digestion, immunoprecipitates were incubated for 10 h at 37°C in the presence or absence of 8 U/ml N-Glycanase in 100 mM phosphate buffer (pH 7.2), 10 mM EDTA, and 5% methanol in a total volume of 100 μl. For digestion with neuraminidase (NANase), the immunoprecipitates were incubated in the presence or absence of 300 mU/ml NANase from Vibrio cholerae for 15 h at 37°C in acetate buffer (50 mM, pH 5.5). For digestion with endoglycosidase H (Endo-H), immunoprecipitates were incubated in the presence or absence of 250 mU/ml Endo-H in citrate buffer (0.1 M, pH 5.6) and sodium dodecyl sulfate (SDS; 0.2%) for 18 h at 37°C. The incubates were cooled in ice, and the beads were washed with buffer B prior to western blot analysis or four times with 100 mM HCl/cacodylate buffer (pH 6.5) prior to sialyltransferase activity determinations according to Daniotti et al. (1994).

SDS-polyacrylamide gel electrophoresis (PAGE) and immunoblotting

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. SDS-polyacrylamide gel electrophoresis (PAGE) and immunoblotting
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

CHO-K1 cell homogenates and immunoprecipitates were resolved by electrophoresis through 10% SDS-polyacrylamide gels (Laemmli, 1970) both in the presence and absence of β-mercaptoethanol (βME). Proteins were electrophoretically transferred to nitrocellulose membranes (Towbin et al., 1979) for 1 h at 300 mA. For immunoblotting, nonspecific binding sites on the nitrocellulose were blocked with 2.5% bovine serum albumin (BSA)/2.5% polyvinylpyrrolidone 40 in Tris-buffered saline (400 mM NaCl, 100 mM Tris-HCl, pH 7.5). The anti-HA murine monoclonal antibody 12CA5 (Babco, U.S.A.) was used at a dilution of 1:1,000. Blots were developed with a 1:30,000 dilution of anti-IgG mouse polyclonal antibody coupled to horseradish peroxidase using the enhanced chemiluminescence (ECL) detection kit (Amersham Corp., Arlington Heights, IL, U.S.A.) and Kodak Biomax MS films. The molecular masses were calculated based on calibrated standards (GibcoBRL, Gaithersburg, MD, U.S.A.) run in every gel. Protein bands in nitrocellulose membranes were visualized by Ponceau S staining. The relative contribution of individual bands was determined by densitometric scanning of the film in a CS 930 Shimadzu UV/vis scanner.

Immunofluorescence

Cells grown in coverslips were rinsed three times in PBS (pH 7.4), fixed, permeabilized in methanol at -20°C for 8 min, and incubated with PBS containing 3% BSA (PBS-BSA) during 1 h. Coverslips were then incubated with a 1:200 dilution of specific monoclonal anti-HA antibody 12CA5 (Babco) together with a 1:150 dilution of rabbit polyclonal anti-M6PR (a gift of Dr. Alfredo Caceres, Instituto Mercedes y Martin Ferreyra, Córdoba, Argentina) in PBS-BSA. After washing in PBS, the coverslips were incubated with a 1:250 dilution of rhodamine-labeled donkey anti-mouse IgG (Jackson ImmunoResearch, West Grove, PA, U.S.A.) and a 1:200 dilution of fluorescein isothiocyanate-conjugated donkey anti-rabbit IgG (Jackson ImmunoResearch) in PBS-BSA; after several washes with PBS and a last one with water, coverslips were mounted in Fluorosave (Calbiochem, San Diego, CA, U.S.A.) and viewed in a Carl-Zeiss Axiovert 135M laser scanning confocal microscope with 488- and 543-nm lasers (Centro de Equipo Mayor, Santiago, Chile). Images were analyzed using LSM version 3.80 and Metamorph Imaging software. For GD3 immunostaining, cells were fixed in acetone at -20°C for 7 min and the procedure was followed as above using a 1:150 dilution of the monoclonal anti-GD3 antibody R24 in PBS-BSA and a 1:200 dilution of fluorescein isothiocyanate-conjugated donkey antimouse IgG.

For double detection of Sial-T2-HA and mannosidase II (Man II), cells processed as indicated above were immunostained with a 1:200 dilution of the monoclonal anti-HA antibody and a 1:500 dilution of the polyclonal rabbit anti-Man II antibody (from K. Moremen, University of Georgia, Athens, GA, U.S.A.). After washing, cells were incubated for 60 min at 37°C with a mixture of a 1:1,000 dilution of rhodamine-conjugated donkey anti-mouse IgG (Jackson ImmunoResearch) to visualize Sial-T2—HA and a 1:1,000 dilution of fluorescein-conjugated donkey anti-rabbit IgG (Jackson ImmunoResearch) to visualize Man II. Specimens were mounted with Fluorosave, observed in a Zeiss Axioplan fluorescence microscope equipped with a 63× 1.4 NA oil immersion objective, and photographed with a Princeton Instrument Micromax camera controlled with Metamorph Imaging software (Universal Imaging Corp., U.S.A.).

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. SDS-polyacrylamide gel electrophoresis (PAGE) and immunoblotting
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

Enzyme activity of expressed Sial-T2

To study the protein product of Sial-T2 cDNA and to learn its subcellular localization, the plasmid pCEFL-Sial-T2-HA was generated. This plasmid directs the expression of chick Sial-T2 tagged with a 9-amino acid antigenic epitope from the viral HA at its C-terminus. The HA epitope did not affect Sial-T2 activity either in vitro or in vivo. Homogenates from transiently transfected CHO-K1 cells showed Sial-T2 activity values of 9.5 nmol/mg of protein/h (Fig. 1). In experiments not shown, metabolic labeling of transfected cells with [3H]galactose resulted in synthesis of radioactive GD3, and immunostaining with monoclonal antibody R24, specific for GD3, showed the expression of this ganglioside at the cell surface in a conspicuous dotted pattern.

image

Figure 1. Western blot of expressed Sial-T2. Postnuclear lysates of CHO-K1 cells (70 μg of protein) transiently transfected with the expression vector alone (-) or vector with insert (+) were run in 10% SDS-PAGE, immunoblotted with anti-HA antibody, and visualized by the ECL technique. The position of recombinant Sial-T2 (47 kDa) is indicated. The specific activity (nmol of NeuAc transferred/mg of protein/h) of Sial-T2 was measured in homogenates using GM3 as acceptor, as indicated (Daniotti et al., 1994).

Download figure to PowerPoint

Characterization of expressed Sial-T2

Extracts from Sial-T2-transfected CHO-K1 cells analyzed by western blotting with anti-HA antibodies showed Sial-T2 migrating in SDS-PAGE with the expected molecular mass of ∼47,000 Da (Fig. 1), corresponding to the molecular mass predicted from the cDNA sequence plus the three N-linked oligosaccharides (Martina et al., 1998). No immunostained band was observed in extracts from mock-transfected CHO-K1 cells. In gels run in the absence of βME, a major band migrating at ∼95 kDa appeared, which accounts for ∼80% of the total enzyme (Fig. 2). In the presence of 10% βME, the 95-kDa form disappeared and the concomitant appearance of the 47-kDa form was observed. This showed that the 95-kDa form contains the 47-kDa form and strongly suggested that the 95-kDa form is formed by a dimeric association, via disulfide bridges, of the 47-kDa form.

image

Figure 2. Disulfide-bonded species of Sial-T2: effect of CST. Transiently transfected CHO-K1 cells were incubated in the absence (-) or presence (+) of 75 μg/ml CST during the last 8 h of the 15-h transfection period. Fifty micrograms of protein of cell homogenates was run in 10% SDS-PAGE gels with or without 10% βME in the sample buffer, immunoblotted with anti-HA antibody, and visualized by the ECL technique. Positions and sizes of the different Sial-T2 forms are indicated on the right.

Download figure to PowerPoint

The subcellular site of formation of the 95-kDa form was investigated taking advantage of previous knowledge that castanospermine (CST), an inhibitor of endoplasmic reticulum (ER) glucosidase I and II, leads to retention of a slightly higher molecular mass Sial-T2 in the ER (Martina et al., 1998). Figure 2 shows that a single Sial-T2 band of slightly lower mobility than in CST-untreated cells was observed at ∼47 kDa in extracts of CST-treated cells run in the presence of βME. The major species in the absence of βME ran at ∼95 kDa, in both CST-treated and -untreated cells, indicating that formation of the 95-kDa form occurred mainly in the ER, even in the absence of normal processing of N-linked oligosaccharides.

Sub-Golgi localization

Double immunostaining of the epitope-tagged Sial-T2 and of M6PR was carried out. Sial-T2 was found predominantly located in a region near the cell nucleus (Fig. 3A), co-localizing partially with the condensed and rather uniform area of immunostaining of the M6PR, a marker of the TGN that extends also in a dotted pattern characteristic of the small vesicles and tubules of the endosomal system (Goda and Pfeffer, 1988; Griffiths et al., 1988) (Fig. 3B). Merging of the confocal images of Fig. 3A and B clearly shows areas of co-localization of both Sial-T2 and M6PR (Fig. 3C). In different optical sections (not shown), most of these areas correspond to the condensed M6PR labeling.

image

Figure 3. Double immunofluorescence for Sial-T2 and M6PR. CHO-K1 cells 12 h after transfection (A-C) or treated with BFA during the last 30 min of culture (D-F) were fixed with cold methanol, stained with either anti-HA (A and D) or anti-M6PR (B and E) followed by their appropriate second antibodies, and examined by confocal laser microscopy. The co-localization of Sial-T2 (red) and M6PR (green) after superposition of confocalized images is shown in yellow (Merge; C and F). Confocal sections were taken every 0.5 μm parallel to the coverslip. Images are representative of the majority of the observed transfected cells.

Download figure to PowerPoint

To further study the co-localization of Sial-T2 and M6PR, cells were treated with brefeldin A (BFA), a fungal macrocyclic lactone that reversibly blocks intra-Golgi vesicular transport, causing redistribution of proximal Golgi membranes (cis, medial, and trans), but not TGN membranes, into the ER (Donaldson and Klausner, 1994). BFA led to a dispersion of most Sial-T2 immunoreactivity throughout the ER (Fig. 3D), indicating that a substantial fraction of the enzyme localizes to some compartment of the proximal Golgi. M6PR immunoreactivity, on the other hand, did not redistribute but remained near the nucleus, showing a more condensed appearance (Fig. 3E). The co-localization remains even after 30 min of exposure to BFA, as shown in Fig. 3F, resulting from the merging of Fig. 3D and E, thus confirming the Sial-T2 presence in membranes of the TGN.

It may be argued that the presence of Sial-T2 in the proximal Golgi was due to mislocalization of overexpressed protein in the transiently transfected cells, as was shown to occur for the case of GalNAc-T (Giraudo et al., 1999; see also Young et al., 1999). To investigate this possibility, we examined the localization of Sial-T2 in CHO-K1 cells stably expressing HA-tagged Sial-T2 under the control of an ecdysone-inducible promoter. Figure 4 is a western blot showing that expression of Sial-T2 is at the limit of detection in extracts from transfected cells grown in the absence of the ecdysone analogue ponasterone. The Sial-T2 band was observed in cells induced with 0.5 μM ponasterone, and its intensity increased with 5 μM ponasterone. Sial-T2 activity was 0.07, 1.1, and 3.9 nmol of NeuAc incorporated/mg of protein/h in, respectively, extracts from cells induced with 0, 0.5, and 5 μM ponasterone. No Sial-T2 band or activity was detected in mock-transfected cells induced with 0.5 μM ponasterone.

image

Figure 4. Induction of Sial-T2 by the ecdysone analogue ponasterone. Cells from clone IST2A were grown in the absence and presence of the indicated concentrations of ponasterone during 24 h. Homogenates were western blotted with anti-HA antibody, as indicated in Fig. 1. Mock, cells transfected with empty pIND.

Download figure to PowerPoint

Figure 5 shows the immunolocalization of Sial-T2 in comparison with that of the medial Golgi marker Man II in cells induced with 0.5 μM ponasterone and treated with BFA during the last 30 min of culture. Whereas Sial-T2 immunoreactivity was at the limit of detection in uninduced cells, it was clearly seen as a dotted perinuclear rim in ponasterone-induced cells, similar to the pattern of immunostaining of the medial Golgi Man II. Induced cells treated with BFA showed most Sial-T2 redistributed to the ER, as was the immunostaining of Man II. These results confirm those of Fig. 3 and discard that the presence of Sial-T2 in proximal Golgi compartments was the consequence of overloading the cell with expressed protein. Similarly, the results of Fig. 6, showing that co-localization of Sial-T2 with M6PR is still present in cells expressing controlled levels of Sial-T2, support that Sial-T2 presence in the TGN was not due to overexpression of the enzyme.

image

Figure 5. Double immunofluorescence for Sial-T2 and Man II. CHO-K1 inducible cells (clone IST2A) were induced to express Sial-T2 by exposure during 24 h to the indicated concentrations of the inductor ponasterone. Cells were then processed for Sial-T2 (A, C, and E; red fluorescence) and Man II (B, D, and F; green fluorescence) immunodetection, as indicated in Materials and Methods. E and F show the redistribution of both Sial-T2 (E) and Man II (F) after 30 min of exposure of cells to BFA.

Download figure to PowerPoint

image

Figure 6. Double immunofluorescence for Sial-T2 and M6PR. Cells as in Fig. 5 were processed for Sial-T2 (A and C; red fluorescence) and M6PR (B and D; green fluorescence) immunodetection, as indicated in Materials and Methods, in the absence (A and B) and presence (C and D) of BFA added to the culture medium 30 min before fixing. Arrows indicate the co-localization of Sial-T2 and M6PR.

Download figure to PowerPoint

Functional co-localization with GalNAc-T

To investigate if the fraction of Sial-T2 detected in the distal Golgi was functionally coupled to the distal Golgi-located GalNAc-T (van Echten et al., 1990; Rosales Fritz and Maccioni, 1995; Giraudo et al., 1999), a CHO-K1 cell clone stably expressing a TGN-located c-myc-tagged form of GalNAc-T (clone 1; Giraudo et al., 1999) was transiently transfected with the plasmid pCEFL-Sial-T2-HA. Twelve hours after transfection, microsomal membranes (containing Golgi membranes) were collected from these cells and incubated first with 15 μM CMP-NeuAc during 2 h to complete the endogenous acceptors of sialic acid (mainly, LacCer, GM3, and GM1) to their respective products (GM3, GD3, and GD1a). The synthesis of these products was confirmed in parallel experiments in which the membranes were incubated with CMP-[3H]NeuAc (not shown). The membranes were then washed and incubated for 2 additional h with 10 μM UDP-[3H]GalNAc to examine if GD3 molecules synthesized by transfected Sial-T2 were accessible as substrate for GalNAc-T and converted into radioactive GD2. Figure 7 shows that whereas ganglioside GM2 and GA2 were the main radioactive products synthesized in membranes from the GalNAc-T stably transfected cells, radioactive GD2 was clearly detected in membranes from GalNAc-T stable transfectants transiently transfected with Sial-T2. As the experimental conditions do not support vesicular coupling between compartments (Rothman, 1994), these results show that at least part of the endogenous GD3 synthesized in the first incubation step was accessible to the distal Golgi GalNAc-T and, together with results of the immunocytochemistry (Fig. 3), indicate that the distal Golgi-located Sial-T2 was functionally coupled to the distal (TGN) GalNAc-T. The fraction of GD3 coupled to synthesis of GD2 by GalNAc-T is difficult to calculate because we do not know either the size of the endogenous pool of preexisting GD3 or the percentage of GalNAc-T-expressing cells that were transfected with Sial-T2.

image

Figure 7. Radioactive labeling pattern of gangliosides from CHO-K1 cells. Membranes from CHO-K1 cells stably expressing human GalNAc-T or from the same cells transiently transfected with Sial-T2 (GalNAc-T/Sial-T2) were incubated first with unlabeled CMP-NeuAc and then with UDP-[3H]GalNAc and the glycolipids purified and run on HPTLC as indicated in Materials and Methods. The positions of GA2, GM2, and GD2 glycolipids are indicated.

Download figure to PowerPoint

N-Glycan processing status

It has been already shown that Sial-T2 contains three potential N-glycosylation sites that were occupied by N-glycans (Martina et al., 1998). As an alternative approach to investigate the sub-Golgi location of Sial-T2, we examined the processing status of these N-glycans. It is accepted that glycoproteins moving along the secretory pathway and progressing beyond the medial Golgi suffer the successive action of the processing enzymes GlcNAc-T1/Man II, acquiring Endo-H resistance and terminal glycosylation of their N-glycans (Kornfeld and Kornfeld, 1985). Sial-T2 immunoprecipitated from cell homogenates (Fig. 8A, Control) shows the major band of 47 kDa and a band of ∼44 kDa that correspond to a partially deglycosylated form (two N-glycans) detectable in this overloaded gel. Partial digestion with N-Glyanase, which cleaves the asparagine-oligosaccharide bond of N-linked oligosaccharides, produced two additional bands corresponding to polypeptides with 1 (∼42 kDa) and no (∼40 kDa) N-glycans (Fig. 8A, N-Glycanase). These N-glycans were Endo-H-sensitive, as shown by the decrease of the molecular mass of each polypeptide (Fig. 8A, Endo-H), and NANase-insensitive (Fig. 8A, NANase). This suggested that most Sial-T2 did not progress beyond the medial Golgi.

image

Figure 8. Western blot analysis of cellular and released Sial-T2. CHO-K1 cells stably expressing Sial-T2 were grown for 48 h. Sial-T2 was immunoprecipitated both from the cellular homogenate (A; 150 μg of protein) and from the culture medium (B; from 10 ml collected after 48 h of culture), treated with N-Glycanase, Endo-H, or NANase, and western blotted following the procedures indicated in Materials and Methods.

Download figure to PowerPoint

To discount that the lack of processing of Sial-T2 was due to failure of the cell line used to properly process N-glycans, we looked for the presence of Sial-T2 released to the culture medium (Nara et al., 1994) to examine its N-glycan-processing status. Secreted Sial-T2 was found in the culture medium of ∼7 × 106 cells grown during 48 h (10 ml), which amounted to ∼ 7-8% (63 pmol/h) of the total activity in vitro (840 pmol/h, released plus cellular). The released Sial-T2 migrated slightly slower than the cellular form (Fig. 8B, Control), and when deglycosylated with N-Glycanase, it generated a major band at 35 kDa corresponding to the fully deglycosylated polypeptide and a minor band at 40 kDa corresponding probably to one N-glycan-bearing polypeptide (Fig. 8B, N-Glycanase). The N-glycans of this form were Endo-H-resistant (Fig. 8B, Endo-H) and decreased its molecular mass to ∼3 kDa after NANase treatment (Fig. 8B, NANase). These results indicate that the Endo-H sensitivity of membrane-bound Sial-T2 was not due to failure of the cells for proper N-glycan processing or to the intrinsic properties of Sial-T2 (i.e., bearing hybrid-type N-glycans) impeding proper processing and support the conclusion that it was located mainly in early Golgi compartments.

Comparison of N-Glycanase lanes of Fig. 8A and B evidences that the molecular mass of the major deglycosylated polypeptide of the released form (∼35 kDa) was slightly lower than the corresponding deglycosylated polypeptide obtained from the cellular form (40 kDa). This strongly suggests that the released form is a cleaved C-terminal fragment bearing the HA tag that acquired Endo-H resistance and sialic acid while moving through the distal Golgi before secretion. Cleavage of glycosyl-transferases has been already observed in transfected cells (Nara et al., 1994; Jaskiewicz et al., 1996; Costa et al., 1997; Borsig et al., 1998). An amino acid sequence analysis of Sial-T2 showed a putative cathepsin D-like protease site in the stem region.

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. SDS-polyacrylamide gel electrophoresis (PAGE) and immunoblotting
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

CHO-K1 cells stably transfected with a cDNA coding for an epitope-tagged version of chicken Sial-T2 (Daniotti et al., 1997b) expressed a protein with an estimated molecular mass (∼47 kDa) very close to that predicted from its cDNA sequence plus three N-glycan chains (Martina et al., 1998). The protein was enzymatically active in vitro and in vivo and was expressed mainly as a disulfide-bonded 95-kDa product, whose formation persisted when exit of the protein from the ER was impaired by inhibiting ER glucosidases I and II with CST, indicating that formation of homodimer occurs mainly in the ER. Disulfide-bonded homodimers of GalNAc-T formed in the ER (Jaskiewicz et al., 1996) and of inactive α2,6-sialyltransferase formed in the Golgi (Ma and Colley, 1996) have been described.

Confocal immunofluorescence microscopy revealed a broad distribution of Sial-T2 along the Golgi complex including the TGN in transiently transfected cells, as was evidenced by co-localization with the M6PR, a typical marker of TGN and late endosomes (Goda and Pfeffer, 1988; Griffiths et al., 1988). In cells treated with BFA, a major fraction of the enzyme redistributed to the ER, indicative of a proximal Golgi location. It has recently been shown that overloading the TGN with expressed GalNAc-T results in the presence of the protein also in the proximal Golgi (Giraudo et al., 1999). For the case of Sial-T2, it should be noticed that a specific activity of 9.5 nmol/mg of protein/h in the transient transfectants, as shown in Fig. 1, may mean a 10-fold higher value per transfected cell if an efficiency of transfection of 10% is considered. However, the proximal Golgi location of Sial-T2 was not due to overloading of more distal compartments, as the same localization was found in cells stably transfected with Sial-T2 under the control of an ecdysone-inducible promoter, which express ∼80-fold less Sial-T2 (1.1 nmol/mg of protein/h). The Sial-T2 activity in the induced cells was in the order of the activities measured in chick brain and retina at developmental periods of highest activity (Daniotti et al., 1997b). Additionally, the presence of Sial-T2 in the TGN of cells expressing controlled levels of Sial-T2 favors the notion that overexpression was not the cause of a minor fraction of this enzyme localizing in the TGN.

The Endo-H-sensitive and NANase-insensitive nature of Sial-T2 N-glycans was consistent with the immunocytochemical data indicating that the major fraction of the enzyme did not progress beyond the medial Golgi. An Endo-H-resistant fraction was detectable in highly exposed blots, and a fraction still remains associated with M6PR-bearing membranes in BFA-treated cells, indicative of the presence of the enzyme also in the TGN. When stable GalNAc-T transfectants, in which GalNAc-T concentrates in the TGN (Giraudo et al., 1999), were transfected with Sial-T2, the synthesis of GD2 was observed, indicating that the TGN-located fraction of Sial-T2 was functionally active.

Evidence has accumulated in the last few years to indicate that Golgi glycosyltransferases not only diffuse laterally along Golgi cisternae of different stacks, but also recycle between adjacent cisternae as a requirement for proper Golgi function (for review, see Allan and Balch, 1999). Despite different possibilities of subcompartmentation used for different cell types (Roth et al., 1986; Sjoberg and Varki, 1993), the present results show that most Sial-T2 localizes in the proximal (cis, medial, or trans) Golgi and that a minor functional fraction is also present in the TGN. The presence of Sial-T2 in the TGN is consistent with previous findings in Golgi membranes from chick neural retina cells in vitro, evidencing considerable functional co-localization of transfer steps for synthesis of LacCer, GM3, and GD3 with the transfer step catalyzed by the TGN-located GalNAc-T (Maxzúd et al., 1995; Maxzúd and Maccioni, 1997). Biochemical studies carried out in liver cells also indicated a displacement of ganglioside glycosyltransferase activities toward the late Golgi, with GM3 synthase (and Sial-T2) distributed in subcellular fractions enriched in markers of the trans-Golgi and of the TGN (Lannert et al., 1998). Whether the findings reported here reflect the steady-state distribution of the Sial-T2 recycling between these subcompartments and whether in vivo the amount of enzyme in a subcompartment correlates directly with the efficiency of that transfer step in the subcompartment remain to be established.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. SDS-polyacrylamide gel electrophoresis (PAGE) and immunoblotting
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

This work was supported in part by grants 3590/95 from Consejo de Investigaciones Científicas y Tecnológicas de la Provincia de Córdoba (CONICOR) of Argentina, 89/96 from Secretaría de Ciencia y Tecnología de la Universidad Nacional de Córdoba, PMT-PICT-0181 from the Consejo de Investigaciones Cientificas y Tecnologicas (CONICET) of Argentina, 4218R1 from the Council for Tobacco Research (U.S.A.), and HHMI 75197 554001. H.J.F.M. and J.L.D. are career investigators and J.A.M., C.G., and A.R.Z. are fellows of CONICET.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. SDS-polyacrylamide gel electrophoresis (PAGE) and immunoblotting
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES
  • 1
    Allan B.B. & Balch W.E. (1999) Protein sorting by directed maturation of Golgi compartments.Science 285,6366.
  • 2
    Bieberich E., Freischütz B., Liour S., Yu R.K. (1998) Regulation of ganglioside metabolism by phosphorylation and dephosphorylation.J. Neurochem. 71,972979.
  • 3
    Borsig L., Katopodis A.G., Bowen B.R., Berger E.G. (1998) Trafficking and localization studies of recombinant alpha 1, 3-fucosyltransferase VI stably expressed in CHO cells.Glycobiology 8,259268.
  • 4
    Costa J., Grabenhorst E., Nimtz M., Conradt H.S. (1997) Stable expression of the Golgi form and secretory variants of human fucosyltransferase III from BHK-21 cells. Purification and characterization of an engineered truncated form from the culture medium.J. Biol. Chem. 272,1161311621.
  • 5
    Daniotti J.L., Landa C.A., Maccioni H.J.F. (1994) Regulation of ganglioside composition and synthesis is different in developing chick retinal pigment epithelium and neural retina.J. Neurochem. 62,11311136.
  • 6
    Daniotti J.L., Rosales Fritz V.M., Martina J.A., Furukawa K., Maccioni H.J.F. (1997a) Expression of β1-4 N-acetylgalactosaminyltransferase gene in the developing rat brain and retina: mRNA, protein immunoreactivity and enzyme activity. Neurochem. Int. 31,1119.
  • 7
    Daniotti J.L., Rosales Fritz V.M., Kunda P., Nishi T., Maccioni H.J.F. (1997b) Cloning, characterization and developmental expression of α2,8 sialyltransferase (GD3 synthase, ST8Sia I) gene in chick brain and retina. Int. J. Dev. Neurosci. 15,767776.
  • 8
    Donaldson J.G. & Klausner R.D. (1994) ARF: a key regulatory switch in membrane traffic and organelle structure.Curr. Opin. Cell Biol. 6,527532.
  • 9
    Giraudo C., Rosales Fritz V.M., Maccioni H.J.F. (1999) GA2/GM2/GD2 synthase localises to the trans Golgi network of CHO-K1 cells.Biochem. J. 342,633640.
  • 10
    Goda Y. & Pfeffer S.R. (1988) Selective recycling of the mannose 6-phosphate/IGF-II receptor to the trans Golgi network in vitro.Cell 55,309320.
  • 11
    Griffiths G., Hoflack B., Simons K., Mellman I., Kornfeld S. (1988) The mannose 6-phosphate receptor and the biogenesis of lysosomes.Cell 52,329341.
  • 12
    Hakomori S.H. (1981) Glycosphingolipids in cellular interaction, differentiation and oncogenesis.Annu. Rev. Biochem. 50,733764.
  • 13
    Jaskiewicz E., Zhu G., Bassi R., Darling D.S., Young W.W.J r. (1996) β1,4-N-Acetylgalactosaminyltransferase (GM2 synthase) is released from Golgi membranes as a neuraminidase-sensitive, disulfide-bonded dimer by a cathepsin D-like protease. J. Biol. Chem. 271,2639526403.
  • 14
    Kornfeld R. & Kornfeld S. (1985) Assembly of asparagine-linked oligosaccharides.Annu. Rev. Biochem. 54,631664.
  • 15
    Kozak M. (1987) An analysis of 5′-noncoding sequences from 699 vertebrate messenger RNAs. Nucleic Acids Res. 15,81258148.
  • 16
    Laemmli U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4.Nature 227,680685.
  • 17
    Lannert H., Gorgas K., Meibner I., Wieland F.T., Jeckel D. (1998) Functional organization of the Golgi apparatus in glycosphingo-lipid biosynthesis.J. Biol. Chem. 273,29392946.
  • 18
    Ma J. & Colley K.J. (1996) A disulfide-bonded dimer of the Golgi β-galactoside α2,6-sialyltransferase is catalytically inactive yet still retains the ability to bind galactose. J. Biol. Chem. 271,77587766.
  • 19
    Maccioni H.J.F., Daniotti J.L., Martina J.A. (1999) Organisation of ganglioside synthesis in the Golgi apparatus.Biochim. Biophys. Acta 1437,101118.
  • 20
    Martina J.A., Daniotti J.L., Maccioni H.J.F. (1998) Influence of N-glycosylation and N-glycan trimming on the activity and intra-cellular traffic of GD3 synthase (ST8Sia I). J. Biol. Chem. 273,37253731.
  • 21
    Maxzúd M.K. & Maccioni H.J.F. (1997) Compartmental organization of the synthesis of GM3, GD3 and GM2 in Golgi membranes from neural retina cells.Neurochem. Res. 22,455461.
  • 22
    Maxzúd M.K., Daniotti J.L., Maccioni H.J.F. (1995) Functional coupling of glycosyl transfer steps for synthesis of gangliosides in Golgi membranes from neural retina cells.J. Biol. Chem. 270,2020720214.
  • 23
    Nara K., Watanabe Y., Maruyama K., Kasahara K., Nagai Y., Sanai Y. (1994) Expression cloning of a CMP-NeuAc NeuAcα2-3Galβ1-4Glcβ1-1′Cer α2,8-sialyltransferase (GD3 synthase) from human melanoma cells. Proc. Natl. Acad. Sci. USA 91,79527956.
  • 24
    Panzetta P., Maccioni H.J.F., Caputto R. (1980) Synthesis of retinal gangliosides during chick embryonic development.J. Neurochem. 35,100108.
  • 25
    Rosales Fritz V.M. & Maccioni H.J.F. (1995) Effects of brefeldin A on synthesis and intracellular transport of ganglioside GT3 by chick embryo retina cells.J. Neurochem. 65,18591864.
  • 26
    Rosales Fritz V.M., Maxzúd M.K., Maccioni H.J.F. (1996) GT3 synthesis in the proximal Golgi occurs in a compartment different from those for GD3 and GM3 synthesis.J. Neurochem. 67,13931400.
  • 27
    Roth J., Taatjes D.J., Weinstein J., Paulson J.C., Greenwell P., Watkins W.M. (1986) Differential subcompartmentation of terminal glycosylation in the Golgi apparatus of intestinal absorptive and goblet cells.J. Biol. Chem. 261,1430714312.
  • 28
    Rothman J.E. (1994) Mechanisms of intracellular protein transport.Nature 372,5563.
  • 29
    Sells M.A. & Chernoff J. (1995) Epitope-tag vectors for eukaryotic protein production.Gene 152,187189.
  • 30
    Sjoberg E.R. & Varki A. (1993) Kinetic and spatial interrelationships between ganglioside glycosyltransferase and O-acetyltransferase(s) in human melanoma cells. J. Biol. Chem. 268,1018510196.
  • 31
    Svennerholm L. (1963) Chromatographic separation of human brain gangliosides.J. Neurochem. 10,613623.
  • 32
    Takamiya K., Yamamoto A., Yamashiro S., Furukawa K., Haraguchi M., Okada M., Ikeda T., Shiku H., Furukawa K. (1995) T cell receptor-mediated stimulation of mouse thymocytes induces up-regulation of the GM2/GD2 synthase gene.FEBS Lett. 358,7983.
  • 33
    Towbin H., Staehelin T., Gordon J. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications.Proc. Natl. Acad. Sci. USA 76,43504354.
  • 34
    Trinchera M. & Ghidoni R. (1989) Two glycosphingolipid sialyltransferases are localized in different sub-Golgi compartments in rat liver.J. Biol. Chem. 264,1576615769.
  • 35
    Trinchera M., Pirovano B., Ghidoni R. (1990) Sub-Golgi distribution in rat liver of CMP-NeuAc GM3 and CMP-NeuAc GT1b α2[RIGHTWARDS ARROW]8 sialyltransferases and comparison with the distribution of the other glycosyltransferase activities involved in ganglioside biosynthesis. J. Biol. Chem. 265,1824218247.
  • 36
    Van Echten G. & Sandhoff K. (1993) Ganglioside metabolism. Enzymology, topology, and regulation.J. Biol. Chem. 268,53415344.
  • 37
    Van Echten G., Iber H., Stotz H., Takatsuki A., Sandhoff K. (1990) Uncoupling of ganglioside biosynthesis by brefeldin A.Eur. J. Cell Biol. 51,135139.
  • 38
    Yamamoto A., Haraguchi M., Yamashiro S., Fukumoto S., Furukawa K., Takamiya K., Atsuta M., Shiku H., Furukawa K. (1996) Heterogeneity in the expression pattern of two ganglioside synthase genes during mouse brain development.J. Neurochem. 66,2634.
  • 39
    Young W.W.J, Lutz M.S., Mills S.E., Leschler-Osborn S. (1990) Use of brefeldin A to define sites of glycosphingolipid synthesis: GA2/GM2/GD2 synthase is trans to the brefeldin A block.Proc. Natl. Acad. Sci. USA 87,68386842.
  • 40
    Young W.W.J, Allende M.L., Jaskiewicz E. (1999) Reevaluating the effect of brefeldin A (BFA) on ganglioside synthesis: the location of GM2 synthase cannot be deduced from the inhibition of GM2 synthesis by BFA.Glycobiology 9,689695.
  • 41
    Zeng G., Ariga T., Gu X., Yu R.K. (1995) Regulation of glycolipid synthesis in HL-60 cells by antisense oligodeoxynucleotides to glycosyltransferase sequence: effect on cellular differentiation.Proc. Natl. Acad. Sci. USA 92,86708674.