Evidence supporting a late Golgi location for lactosylceramide to ganglioside GM3 conversion

Ganglioside GM2 synthase and other enzymes required for complex ganglioside synthesis were localized recently to the trans Golgi network (TGN). However, there are conflicting reports as to the location of GM3 synthase; originally this enzyme was detected in the early Golgi of rat liver but a recent report localized it to the late Golgi. We have used chimeric forms of ganglioside GM2 synthase to determine if the location of lactosylceramide (LacCer) to GM3 conversion in Chinese hamster ovary (CHO) cells was the early or late Golgi. Our approach tested whether GM3 could be utilized as a substrate by GM2 synthase chimeras which were targeted to compartments earlier than the trans Golgi, i.e., GM3 produced in the cis Golgi should be utilized by GM2 synthase located anywhere in the Golgi whereas GM3 produced in the trans Golgi should only be used by GM2 synthase located in the trans Golgi or TGN. Comparison of cell lines stably expressing these chimeras revealed that the in vivo functional activity of GM2 synthase decreased progressively as the enzyme was targeted to earlier compartments; specifically, the percentage of GM3 converted to GM2 was 83-86% for wild type enzyme, 70% for the medial Golgi targeted enzyme, 13% for the ER and cis Golgi targeted enzyme, and only 1.7% for the ER targeted enzyme. Thus, these data are consistent with a late Golgi location for LacCer to GM3 conversion in these cells.     

Alteration of ganglioside synthesis by GM3 synthase knockout in murine embryonic fibroblasts

To probe the functions of membrane gangliosides, the availability of ganglioside-depleted cells would be a valuable resource. To attempt to identify a useful genetic model of ganglioside depletion, we assessed ganglioside metabolism in murine GM3 synthase (GM3S)-/- knockout primary embryonic fibroblasts (MEF), because normal fibroblast gangliosides (GM3, GM2, GM1, and GD1a), all downstream products of GM3S, should be absent. We found that heterozygote MEF (GM3S+/-) did have a 36% reduced content of qualitatively normal gangliosides (7.0+/-0.8 nmol LBSA/mg cell protein; control: 11+/-1.6 nmol). However, two unexpected findings characterized the homozygous (GM3-/-) MEF. Despite complete knockout of GM3S, (i) GM3-/- MEF retained substantial ganglioside content (21% of normal or 2.3+/-1.1 nmol) and (ii) these gangliosides were entirely different from those of wild type MEF by HPTLC. Mass spectrometry identified them as GM1b, GalNAc-GM1b, and GD1alpha, containing both N-acetyl and N-glycolylneuraminic acid and diverse ceramide structures. All are products of the 0 pathway of ganglioside synthesis, not normally expressed in fibroblasts. The results suggest that complete, but not partial, inhibition of GM3 synthesis results in robust activation of an alternate pathway that may compensate for the complete absence of the products of GM3S.     

Incorporation and metabolism of ganglioside GM2 in skin fibroblasts from normal and GM2 gangliosidosis subjects

Ganglioside GM2, 3H-labeled in the sphingoid base, was added to the culture medium of normal and GM2 gangliosidosis fibroblasts. Ganglioside was found to adsorb rapidly to the cell surface, most of it could however be removed by trypsination. The trypsin-resistant incorporation was about 10 nmol/mg cell protein, after 48 h. The rates of adsorption and incorporation depended strongly on the concentration of fetal calf serum in the medium, higher serum concentrations being inhibitory. After various incubation times, the lipids were extracted, separated by thin-layer chromatography and visualized by fluorography. In normal cells a variety of degradation products as well as sphingomyelin was found whereas in GM2 gangliosidosis cells, only trace amounts of such products (mainly GA2) were found. In contrast, the higher gangliosides GM1 and GD1a were formed in comparable amounts (2.2-3.6% of total radioactivity after 92 h) in normal and pathologic cell lines. Supplementation of cells from GM2 gangliosidosis, variant AB, with purified GM2-activator protein restored ganglioside GM2 degradation to almost normal rates but had no effect on its glycosylation to gangliosides GM1 and GD1a. From these results we conclude that the synthesis of higher gangliosides from incorporated GM2 can occur by direct glycosylation and not only via lysosomal degradation and resynthesis from [3H]sphinganine-containing degradation products. Preliminary studies with subcellular fractionation after various times of [3H]ganglioside incorporation indicated biphasic kinetics for the net transport of membrane-inserted ganglioside to lysosomes, compatible with the notion that a portion of the glycolipids can also escape from secondary lysosomes and migrate to Golgi compartment or cell surface.     

Expression machinery of GM4: the excess amounts of GM3/GM4S synthase (ST3GAL5) are necessary for GM4 synthesis in mammalian cells

The ganglioside GM4 is a sialic acid-containing glycosphingolipid mainly expressed in mammalian brain and erythrocytes. GM4 is synthesized by the sialylation of galactosylceramide (GalCer), while the ganglioside GM3 is synthesized by the sialylation of lactosylceramide (LacCer). Recently, the enzyme GM3 synthase was found to be responsible for the synthesis of GM4 in vitro and in vivo, yet the mechanism behind GM4 expression in cells remains unclear. In this study, we attempted to establish GM4-reconstituted cells to reveal the regulation of GM4 synthesis. Interestingly, GM4 was not detected in RPMI 1846 cells expressing LacCer, GalCer, and GM3. Similarly, GM4 was not detected in CHO-K1 cells, even when such cells expressing LacCer and GM3 were stably transfected with the GalCer synthase (GalCerS) gene. GM4 became detectable only when the GM3/GM4 synthase (GM3/GM4S, ST3GAL5) gene was overexpressed in either RPMI 1846 or CHO-K1/GalCerS cells. A mutant of the B16 melanoma cell line, GM-95, lacks GlcCer and LacCer, due to an absence of GlcCer synthase, but carries endogenous LacCer synthase and GM3/GM4S. GalCer became detectable after transfection of GalCerS into GM95 cells, but the GM95/GalCerS reconstituted cells did not express GM4, indicating that competition between the substrates LacCer and GalCer for GM3/GM4S does not cause the failure of GM4 synthesis. These results suggest that the expression machinery of GM4 under physiological conditions is independent from that of GM3.     

Compartmental Organization of the Synthesis of GM3, GD3, and GM2 in Golgi Membranes from Neural Retina Cells

The relationship among lactosylceramide-(LacCer), GD3- and GM2-synthases and between the two last transferases and their common GM3 acceptor was investigated in intact Golgi membrane from chick embryo neural retina cells at early (8-days) and late (14 days) stages of the embryonic development. [³H]Gal was incorporated into endogenous glucosylceramide by incubation of Golgi membranes with UDP-[³H]Gal. Conversion of the synthesized [³H]Gal-LacCer into GM3, and of the latter into GD3, GM2 and GD2 was examined after a second incubation step with unlabeled CMP-NeuAc and/or UDP-GalNAc. With CMP-NeuAc, most [³H]Gal-LacCer was converted into GM3 in either 8- or 14- day membranes. However, while about 90% of GM3 was converted into GD3 in 8-day membranes, only about 25% followed this route in 14-day membranes. With CMP-NeuAc and UDP-GalNAc, about 90% of GM3 was used for synthesis of GM2 in 14-day membranes, while in 8-day membranes about 80% followed the route to GD3, and a part to GD2. Performing the second incubation step in the presence of increasing detergent concentrations showed that conversion of GM3 to GM2 was inhibited at concentrations lower than those required for inhibition of LacCer to GM3 conversion. Taken together, results indicate that transfer steps leading to synthesis of GM3, GD3, GM2 and GD2 from LacCer are functionally coupled in the Golgi membranes, and that GD3- and GM2-synthases compete in a common compartment for using a fraction of GM3 as substrate. In this competition, the relative activities of the transferases and their relative saturation with the respective donor sugar nucleotides, are important factors influencing conversion of GM3 toward either GD3 or GM2.     

Ganglioside biosynthesis in Golgi apparatus of rat liver. Stimulation by phosphatidylglycerol and inhibition by tunicamycin

Golgi vesicles were isolated and purified from rat liver, in which the specific activities of glycosyltransferases (e.g. GM3:CMP-NeuAc sialyltransferase, GD3 synthase; GM3:UDP-GalNAc galactosaminyltransferase, GM2 synthase) were 50-60-times enriched relative to microsomes or total homogenate. Synthesis of gangliosides GM2 and GM1 in such Golgi vesicles is, in the absence of any detergents, stimulated 6-fold and 20-fold respectively by phosphatidylglycerol. Other phospholipids like phosphatidylethanolamine and phosphatidylserine are also significantly stimulatory. With 50 micrograms Golgi protein and 1 nmol UDP-GalNAc, optimal stimulation of GM2 synthase was obtained with 20 micrograms of phosphatidylglycerol and 7.5 nmol of the lipid acceptor GM3. Under the same experimental conditions this stimulation exceeds (by about 40%) that obtained with optimal amount (200 micrograms) of the detergent octylglucoside. Phosphatidylglycerol, on the other hand, has virtually no stimulatory activity on the synthesis of ganglioside GD3 either in the presence of Mg2+ or Mn2+, indicating that facilitation by phospholipid of GM3 transport into Golgi vesicles was not the basis of stimulation of GM2 synthesis. Tunicamycin inhibits the synthesis of gangliosides GM2 and GM1 in isolated Golgi vesicles, but only in the absence of detergents. In the presence of phosphatidylglycerol, GM2 synthesis, for example, was inhibited by 60% by 2 micrograms tunicamycin and more than 85% by 10 micrograms tunicamycin, per 50 micrograms Golgi membrane protein. The inhibition was stronger on GM1 synthesis: 85% with 2.5 micrograms of the antibiotic. The dependence on phosphatidylglycerol and the degree of inhibition by tunicamycin of the synthetic activities are strictly dependent on the intactness of the Golgi vesicles: both phenomena become increasingly less evident when the vesicles are pelleted, and frozen and thawed several times, and completely disappear when the vesicles are solubilized by detergents or disrupted by ultrasonication. Furthermore, tunicamycin inhibition is reversible by increased concentration of phosphatidylglycerol. All these results indicate that phosphatidylglycerol does not stimulate, and tunicamycin does not inhibit, the transferases themselves; rather, the two opposing effects might relate to carrier-mediated transport, e.g. of nucleotide sugars, across Golgi vesicles.     

Oligosialogangliosides inhibit GM2- and GD3-synthesis in isolated Golgi vesicles from rat liver

The effect of end-product gangliosides (GD1a, GT1b, GQ1b) on the activities of two key enzymes in ganglioside biosynthesis, namely GM2-synthase and GD3-synthase in rat liver Golgi apparatus, has been investigated in detergent-free as well as in detergent-containing assays. In detergent-free intact Golgi vesicles, phosphatidylglycerol was used as a stimulant. This phospholipid was earlier shown to stimulate the activity of GM2-synthase without disrupting the vesicular intactness; it has, however, no effect on GD3-synthase (Yusuf, H.K.M., Pohlentz, G., Schwarzmann, G. & Sandhoff, K. (1983) Eur. J. Biochem. 134, 47-54). In the presence of this stimulant, all higher gangliosides inhibited the activity of GM2-synthase, the inhibition being more profound with increasing negative charge of the inhibiting gangliosides. These inhibitions are unspecific, but they do not exclude an end-product regulation of ganglioside biosynthesis. In detergent-solubilized Golgi membranes, on the other hand, the inhibition pattern was completely different. Here, ganglioside GD1a was the strongest inhibitor of GM2-synthase, followed by GM1 and GM2, but GT1b also inhibited this enzyme appreciably, in fact more strongly than GM1 or GM2. On the other hand, GQ1b had no effect at all. Conversely, GD3-synthase activity was most strongly inhibited by GQ1b, followed by GT1b, but GD1a also inhibited this enzyme almost as strongly as GT1b. These latter findings indicate that feed-back control of the a- and the b-series pathways of ganglioside biosynthesis is probably not specific, but the pathways appear to be inhibited more preferably by their respective end-products than by any other gangliosides of the same of the other series.     

Enhancement of ganglioside GM3 synthesis in okadaic-acid-treated granulosa cells.

Okadaic acid is a potent inhibitor of type-2A (PP2A) and type-1 (PP1) protein phosphatases and has been proved to be a valuable tool for studies on the protein phosphorylation. We have investigated the effects of okadaic acid on rat granulosa cells in order to determine whether the regulation of ganglioside synthesis involves protein phosphorylation via inhibition of PP2A and PP1. Granulosa cells expressed luteinizing hormone (LH) receptors, measured as the binding of 125I-deglycosylated human chorionic gonadotropin (hCG) to intact cells, and synthesized the gangliosides NeuAc alpha 2-->3Gal beta 1-->4Glc beta 1-->1Cer (GM3) and Gal beta 1-->3GalNAc beta 1-->4[NeuAc alpha 2-->3]Gal beta 1-->4Glc beta 1-->1Cer (GM1), demonstrated by metabolic labeling of glycosphingolipids with [3H]galactose, in response to follicle-stimulating hormone (FSH). When FSH-stimulated granulosa cells were treated with 10 nM okadaic acid for 15 h, down-regulation of LH receptors, dissociation of LH receptor-effector coupling and significant decreases of intracellular and extracellular 3',5'-cyclic adenosine monophosphate (cAMP) levels were observed. The okadaic acid-induced desensitization to gonadotropin in granulosa cells was accompanied by increased ganglioside synthesis. The amount of 3H-labeled ganglioside GM3, the major ganglioside (about 95% of the total) synthesized by mature granulosa cells, was enhanced in okadaic acid-desensitized cells (to 215% of the control value) and in those desensitized by hCG (by 354%), forskolin (by 190%) and 12-O-tetradecanoylphorbol 13-acetate (by 143%). The results of this study suggest that an increase in the phosphorylation state of cells is accompanied by enhancement of ganglioside synthesis.     

Substrate specificity of GM2 and GD3 synthase of Golgi vesicles derived from rat liver

Several GM3 derivatives have been synthesized. Among them were lyso-GM3 derivatives and GM3 analogues with modifications in the sialic acid moiety. They were used as glycolipid acceptors in assays for GM2 and GD3 synthase of rat liver Golgi. Analysis of the resulting enzyme activities and of the reaction products revealed different substrate specificities for GM2 and GD3 synthase although the normal glycolipid acceptor for both transferases is ganglioside GM3. Specificity of GD3 synthase is strongly determined by the substrate's negative charge and the acyl residue in amide bond to the amino group of neuraminic acid, while GM2 synthase reacts quite indifferently to these changes in the sialic moiety of the substrate. Both enzymes seem to be sensitive to the spatial extension at the neuraminic acid's carboxylic group.     

The total synthesis of ganglioside GM3

Previous syntheses of ganglioside GM3 (NeuAc alpha3Gal beta4Glc beta1Cer) are reviewed, and both chemoenzymatic and chemical total synthetic approaches were investigated. In a chemoenzymatic approach, (2S,3R,4E)-5'-acetyl-alpha-neuraminyl-(2' --> 3')-beta-galactopyranosyl-(1' --> 4')-beta-glucopyranosyl-(1' <--> 1)-2-azido-4-octadecene-1,3-diol (azidoGM3) was readily prepared utilizing recombinant beta-Gal-(1' --> 3'/4')-GlcNAc alpha-(2' --> 3')-sialyltransferase enzyme, and was evaluated as a synthetic intermediate to ganglioside GM3. The chemical total synthesis of ganglioside GM3 was performed on one of the largest scales yet reported. The highlights of this synthesis include minimizing the steps necessary to prepare the lactosyl acceptor as a useful anomeric mixture, which was present in excess for the highly regioselective and fairly stereoselective sialylation with a known neuraminyl donor to give the protected GM3 trisaccharide. The synthetic methodology maximized convergence by a subsequent glycosidic coupling of the well-characterized GM3 trisaccharide trichloroacetimidate derivative with protected ceramide. The ganglioside GM3 was nearly homogeneous as the two glycosidic couplings utilized preparative HLPC purifications, and variations in the sphingosine base and fatty acyl group were under 0.1 and 0.2%, respectively.     

Ganglioside depletion and EGF responses of human GM3 synthase-deficient fibroblasts

Recognition of important roles of gangliosides in normal and abnormal cell function has motivated pharmacological modification of cellular ganglioside content. However, constitutive depletion of gangliosides in untransformed human cells has not been reported. In this context, the recent identification of a kindred carrying a point mutation in the GM3 synthase [ST3Gal5, Siat9] gene (Simpson MA, Cross H, Proukakis C, Priestman DA, Neville DC, Reinkensmeier G, Wang H, Wiznitzer M, Gurtz K, Verganelaki A, Pryde A, Patton MA, Dwek RA, Butters TD, Platt FM, Crosby AH. 2004. Infantile-onset symptomatic epilepsy syndrome caused by a homozygous loss-of-function mutation of GM3 synthase. Nat Genet. 36:1225-1229) provided an opportunity to explore this possibility. We established primary cultures of skin fibroblasts of three patients homozygous for this autosomal recessive defect. They exhibited a 93% reduction in ganglioside content (0.8 +/- 0.2 nmol lipid-bound sialic acid per 10(7) cells versus 12.7 +/- 1.3 nmol per 10(7) normal fibroblasts). Importantly, this marked reduction was not compensated by the activation of an alternate pathway of ganglioside synthesis, as occurs in murine GM3 synthase knockout fibroblasts. Cell morphology appeared unaffected, but under stringent conditions EGF-induced proliferation and migration of the mutant fibroblasts were reduced by 80% and 60%, respectively. Probing potential explanations, we found that EGF binding (effective membrane EGF receptor (EGFR) number) was reduced by 52% (to 6.2 +/- 1.9 from 12.8 +/- 2.0 pmol/10(8) normal fibroblasts, P < 0.01), despite normal total EGFR protein. EGFR activation was likewise reduced as was EGF-induced Rho/Rac1 phosphorylation, which is associated with cell migration. We conclude that this GM3 synthase point mutation almost completely depletes human fibroblast cellular gangliosides, dampens membrane EGFR activation, and modulates related critical cell functions such as proliferation and migration. These cells offer a valuable model for the study of ganglioside modulation of cell function.     

Modulation of GalT1 and SialT1 sub-Golgi localization by SialT2 expression reveals an organellar level of glycolipid synthesis control

Ganglioside glycosyltransferases organize as multienzyme complexes that localize in different sub-Golgi compartments. Here we studied whether in CHO-K1 cells lacking CMP-NeuAc: GM3 sialyltransferase (SialT2), the sub-Golgi localization of UDP-Gal:glucosylceramide beta-1,4-galactosyltransferase (GalT1) and CMP-NeuAc:lactosylceramide sialyltransferase (SialT1) complex is affected when SialT2, another member of this complex, is coexpressed. GalT1 and SialT1 sub-Golgi localization was determined by studying the effect of brefeldin A (BFA) and monensin on the synthesis of glycolipids and on the sub-Golgi localization of GalT1(1-52)-CFP (cyan fluorescent protein) and SialT1(1-54)-YFP (yellow fluorescent protein) chimeras by single cell fluorescence microscopy and by isopycnic subfractionation. We found that BFA, and also monensin, impair the synthesis of glycolipids beyond GM3 ganglioside in wild type (WT) cells but beyond GlcCer in SialT2(+) cells. Although BFA redistributed GalT1-CFP and SialT1-YFP to the endoplasmic reticulum in WT cells, a fraction of these chimeras remained associated with a distal Golgi compartment, enriched in trans Golgi network, and recycling endosome markers in SialT2(+) cells. In BFA-treated cells, the percentage of GalT1-CFP and SialT1-YFP associated with Golgi-like membrane fractions separated by isopycnic subfractionation was higher in SialT2(+) cells than in WT cells. These effects were reverted by knocking down the expression of SialT2 with specific siRNA. Results indicate that sub-Golgi localization of glycosyltransferase complexes may change according to the relative levels of the expression of participating enzymes and reveal a capacity of the organelle to adapt the topology of the glycolipid synthesis machinery to functional states of the cell.     

Design and efficient synthesis of novel GM2 analogues with respect to the elucidation of the function of GM2 activator

To elucidate the mechanism underlying the hydrolysis of the GalNAcβ1→4Gal linkage in ganglioside GM2 [GalNAcβ1→4(NeuAcα2→3)Galβ1→4Glcβ1→1′ Cer] by β-hexosaminidase A (Hex A) with GM2 activator protein, we designed and synthesized two kinds of GM2 linkage analogues—6′-NeuAc-GM2 and α-GalNAc-GM2. In this paper, the efficient and systematic synthesis of these GM2 analogues was described. The highlight of our synthesis process is that the key intermediates, newly developed sialyllactose derivatives, were efficiently prepared in sufficient quantities; these derivatives directly served as highly reactive glycosyl acceptors and coupled with GalNTroc donors to furnish the assembly of GM2 tetrasaccharides in large quantities.     

Stable Transfection of GM1 Synthase Gene into GM1-Deficient NG108-15 Cells,CR-72 Cells,Rescues the Responsiveness of TRK-Neurotrophin Receptor to Its Ligand,NGF

Previous studies from this laboratory and others have suggested the evidences that acidic glycosphingolipid, ganglioside GM1 (GM1), is an endogenous regulator of high affinity nerve growth factor receptor, Trk, which is an essential factor for the normal development and differentiation of neuronal cells by forming a complex with Trk. The present study was aimed to examine whether Trk expressed in cells that are deficient in endogenous GM1 due to the mutation of GM1 synthase gene (NG-CR72 cells) is responsive to its ligand nerve growth factor and how genetic restoration of GM1 synthase gene by a stable transfection of the gene affects the function of the Trk protein. The data clearly showed that (1) confocal lazor microscopic studies disclosed NG-CR72 cells are really deficient in GM1, (2) stable transfection of GM1 synthase cDNA into these cells (NG-CR72G cells) restores the expression of GM1 in the cells, and (3) Trk protein is expressed in NG-CR72 cells but its location seemed not to be on the plasma membrane, whereas we clearly observed that the Trk protein is expressed on the plasma membrane in NG-CR72G cells. (4) NGF did not elicit the autophosphorylation of the Trk protein in GM1 deficient NG-CR72 cells but did elicit the activation of the Trk protein in NG-CR72G cells with an activation of mitogen activated protein kinase. These studies strongly suggested that GM1 is necessary for the normal expression of the Trk protein function and for normal targeting of the Trk protein to the plasma membrane.     

In Cellulo Examination of a Beta-Alpha Hybrid Construct of Beta-Hexosaminidase A Subunits,Reported to Interact with the GM2 Activator Protein and Hydrolyze GM2 Ganglioside

The hydrolysis in lysosomes of GM2 ganglioside to GM3 ganglioside requires the correct synthesis, intracellular assembly and transport of three separate gene products; i.e., the alpha and beta subunits of heterodimeric beta-hexosaminidase A, E.C. # 3.2.1.52 (encoded by the HEXA and HEXB genes, respectively), and the GM2-activator protein (GM2AP, encoded by the GM2A gene). Mutations in any one of these genes can result in one of three neurodegenerative diseases collectively known as GM2 gangliosidosis (HEXA, Tay-Sachs disease, MIM # 272800; HEXB, Sandhoff disease, MIM # 268800; and GM2A, AB-variant form, MIM # 272750). Elements of both of the hexosaminidase A subunits are needed to productively interact with the GM2 ganglioside-GM2AP complex in the lysosome. Some of these elements have been predicted from the crystal structures of hexosaminidase and the activator. Recently a hybrid of the two subunits has been constructed and reported to be capable of forming homodimers that can perform this reaction in vivo, which could greatly simplify vector-mediated gene transfer approaches for Tay-Sachs or Sandhoff diseases. A cDNA encoding a hybrid hexosaminidase subunit capable of dimerizing and hydrolyzing GM2 ganglioside could be incorporated into a single vector, whereas packaging both subunits of hexosaminidase A into vectors, such as adeno-associated virus, would be impractical due to size constraints. In this report we examine the previously published hybrid construct (H1) and a new more extensive hybrid (H2), with our documented in cellulo (live cell- based) assay utilizing a fluorescent GM2 ganglioside derivative. Unfortunately when Tay-Sachs cells were transfected with either the H1 or H2 hybrid construct and then were fed the GM2 derivative, no significant increase in its turnover was detected. In vitro assays with the isolated H1 or H2 homodimers confirmed that neither was capable of human GM2AP-dependent hydrolysis of GM2 ganglioside.     

GM3 alpha2,8-sialyltransferase (GD3 synthase): protein characterization and sub-golgi location in CHO-K1 cells

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 approximately 47 or approximately 95 kDa on sodium dodecyl sulfate-polyacrylamide gel electrophoresis in the presence or absence of, respectively, beta-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 approximately 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.     

Induction of GM1a/GD1b synthase triggers complex ganglioside expression and alters neuroblastoma cell behavior; a new tumor cell model of ganglioside function

Neuroblastoma is the most common extracranial solid tumor in children and tumor ganglioside composition has been linked to its biological and clinical behavior. We recently found that high expression of complex gangliosides that are products of the enzyme GM1a/GD1b synthase predicts a more favorable outcome in human neuroblastoma, and others have shown that complex gangliosides such as GD1a inhibit metastasis of murine tumors. To determine how a switch from structurally simple to structurally complex ganglioside expression affects neuroblastoma cell behavior, we engineered IMR32 human neuroblastoma cells, which contain almost exclusively (89%) the simple gangliosides (SG) GM2, GD2, GM3, and GD3, to overexpress the complex gangliosides (CG) GM1, GD1a, GD1b and GT1b, by stable retroviral-mediated transduction of the cDNA encoding GM1a/GD1b synthase. This strikingly altered cellular ganglioside composition without affecting total ganglioside content: There was a 23-fold increase in the ratio of complex to simple gangliosides in GM1a/GD1b synthase-transduced cells (IMR32-CG) vs. wild type (IMR32) or vector-transfected (IMR32-V) cells with essentially no expression of the clinical neuroblastoma marker, GD2, confirming effectiveness of this molecular switch from simple to complex ganglioside synthesis. Probing for consequences of the switch, we found that among functional properties of IMR32-CG cells, cell migration was inhibited and Rho/Rac1 activities were altered, while proliferation kinetics and cell differentiation were unaffected. These findings further implicate cellular ganglioside composition in determining cell migration characteristics of tumor cells. This IMR32 model system should be useful in delineating the impact of ganglioside composition on tumor cell function.     

Uncoupling of ganglioside biosynthesis by Brefeldin A

We have studied the effect of Brefeldin A (BFA), an antiviral antibiotic, on glycosphingolipid metabolism in primary cultured cerebellar cells. Cells were labeled metabolically with [14C]galactose, or pulse-labeled with precursors of glycosphingolipid biosynthesis; i.e., [14]serine, [3H]palmitic acid or [3H]sphingosine. In all cases BFA (1 microgram/ml) strongly inhibited (75-95%) ganglioside biosynthesis beyond the stage of GM3 and GD3, that is the formation of GM1, GD1a, GT1b and GQ1b. Simultaneously an accumulation of GlcCer, LacCer, GM3 and GD3 was observed (up to 2000%). These effects could be reversed fully by removal of the BFA from the culture medium. These results indicate that the LacCer-, GM3- and GD3-synthases of murine cerebellar cells are localized together on the proximal site of the Golgi apparatus, probably in the cis-Golgi compartment. It is probable that sphingomyelin synthase and some of the other glycosyltransferases involved in ganglioside biosynthesis are localized in distinct compartments beyond the cis Golgi.     

A photoreactive derivative of radiolabeled GM1 ganglioside: preparation and use to establish the involvement of specific proteins in GM1 uptake by human fibroblasts in culture

A new procedure was used to synthesize a derivative of ganglioside GM1 containing a photoreactive nitrophenyl azide group at the end of the fatty acyl moiety, using deAc-deAcyl-GM1 obtained by deacetylation of the sialic acid and deacylation of the ceramide portion of GM1. This deAc-deAcyl-GM1 was first acylated at the long chain base amino group with 12-aminododecanoic acid, which has the amino group protected by a fluorenyl residue, and tritium labeled at the sialic acid amino group with [3H]acetic anhydride of very high specific radioactivity. The fluorenyl group removed by ammonia treatment was substituted by a nitrophenyl azide group. Cultured human fibroblasts were exposed to mixtures of radioactive photolabeled GM1 and cold natural GM1 (1:10 by mol) for different times and then illuminated and the radioactive protein patterns studied by SDS-PAGE. After 2h of exposure, the photolabeled GM1 was stably associated to the cells and underwent almost no metabolic processing, behaving exactly as the underivatized natural GM1. Under these conditions very few proteins became radioactive: one, of about 30 kDa, interacted with the ganglioside molecules inserted into the outer membrane layer; three, in the region of 46 kDa, interacted with the portion of associated ganglioside able to be released by trypsin treatment. Thus, it is evident that the ganglioside binding to fibroblasts and insertion into the outer layer of the plasma membrane involve few individual proteins.(ABSTRACT TRUNCATED AT 250 WORDS)     

Neobiosynthesis of Glycosphingolipids by Plasma Membrane-associated Glycosyltransferases

Gangliosides, complex glycosphingolipids containing sialic acids, are synthesized in the endoplasmic reticulum and in the Golgi complex. These neobiosynthesized gangliosides move via vesicular transport to the plasma membrane, becoming components of the external leaflet. Gangliosides can undergo endocytosis followed by recycling to the cell surface or sorting to the Golgi complex or lysosomes for remodeling and catabolism. Recently, glycosphingolipid catabolic enzymes (glycohydrolases) have been found to be associated with the plasma membrane, where they display activity on the membrane components. In this work, we demonstrated that ecto-ganglioside glycosyltransferases may catalyze ganglioside synthesis outside the Golgi compartment, particularly at the cell surface. Specifically, we report the first direct evidence of expression and activity of CMP-NeuAc:GM3 sialyltransferase (Sial-T2) at the cell surface of epithelial and melanoma cells, with membrane-integrated ecto-Sial-T2 being able to sialylate endogenously synthesized GM3 ganglioside as well as exogenously incorporated substrate. Interestingly, we also showed that ecto-Sial-T2 was able to synthesize GD3 ganglioside at the cell surface using the endogenously synthesized cytidine monophospho-N-acetylneuraminic acid (CMP-NeuAc) available at the extracellular milieu. In addition, the expression of UDP-GalNAc:LacCer/GM3/GD3 N-acetylgalactosaminyltransferase (GalNAc-T) was also detected at the cell surface of epithelial cells, whose catalytic activity was only observed after feeding the cells with exogenous GM3 substrate. Thus, the relative interplay between the plasma membrane-associated glycosyltransferase and glycohydrolase activities, even when acting on a common substrate, emerges as a potential level of regulation of the local glycosphingolipid composition in response to different external and internal stimuli.