Glycolipids in mouse thymocytes stimulated by concanavalin A. Characterisation of glycolipids and rate of lipid biosynthesis during stimulation

Simultaneous determination of sugar incorporation into glycosphingolipids and glycoproteins

We assessed inhibitors of glycosylation by simultaneous determination of [14C]Gal incorporated into glycosphingolipids and glycoproteins as well as of [3H]Leu incorporated into proteins of intact cells. After metabolic labeling in 96-well plates in the presence or absence of a test substance, cells were collected on glass-fiber filters. The lipid components were extracted from the filter and radioactivities of both extract and filter determined. The reliability of the procedure was tested with different drugs. Using the glucocerebroside synthetase inhibitor 1-phenyl-2-decanoylamino-3-morpholino-1-propanol (PDMP; 5 microM), glycolipid biosynthesis was shown to be reduced to 50% in the murine T-cell EL-4 6.1 line, whereas glycosylation of proteins was not disturbed. With 0.5 microM tunicamycin, the glycosylation of proteins was 50% of that in the control. The procedure was also able to detect various specific effects: the inhibition of protein glycosylation with D-glucosamine and castanospermine, the inhibition of glycosphingolipid biosynthesis with L-cycloserine, and a slight enhancement of glycosphingolipid biosynthesis with conduritol B epoxide and castanospermine. Within a series of N-acyl homologs of PDMP the inhibitory potency increased with chain length. In contrast, these homologs were equipotent by enzymatic in vitro assays.

Glycolipid changes in murine myelogenous leukemias: neutral glycolipids as markers for specific populations of leukemias

We have studied the glycolipid composition of six different murine myelogenous leukemias as well as that of T-cell leukemias and normal spleen cells. Neutral and acidic lipid fractions were isolated by column chromatography on DEAE-Sephadex and analyzed by high-performance thin-layer chromatography (HPTLC) and an HPTLC overlay method. Murine myelogenous leukemias were found to contain globo- and ganglio-series neutral glycolipids, e.g., glucosylceramide (Glc-cer), lactosylceramide (Lac-cer), globotriaosylceramide (Gb3), globoside (Gb4), Forssman glycolipid (Gb5), and asialo-GM1 (GA1). Monoblastic leukemia cells contained increased proportions of Gb3, Gb4, Gb5, and GA1. Monocytic and myelomonocytic leukemia cells contained increased proportions of Glc-cer and Lac-cer. Especially, Glc-cer accounted for approximately 60% of the total neutral glycolipids in monocytic leukemia cells. Gb3 was the major neutral glycolipid in reticulum cell neoplasm type A, and it accounted for approximately 75% of the neutral glycolipids. GA1 was the major neutral glycolipid in myeloblastic and granulocytic leukemia cells as well as T-cell leukemias. Especially, granulocytic leukemia cells contained predominantly GA1, and it accounted for approximately 80% of the total neutral glycolipids. The pattern of gangliosides in myelogenous leukemias was more complex when compared with that of the neutral glycolipids; murine myelogenous leukemias contained at least 13 gangliosides, including such major gangliosides as GM1, GM1b containing N-acetyl neuraminic acid and N-glycolyl neuraminic acid, and Ga1NAc-GM1b. Alterations of glycolipid composition in murine myeloid leukemias may be associated with cellular differentiation and maturation, and therefore these characteristic glycolipid species may be regarded as markers for specific populations of leukemia cells.

Structural characterization of gangliosides from murine T lymphocytes

Mouse spleen cells were prepared from CBA/J mice, and T lymphocytes were selectively stimulated with the T cell mitogen concanavalin A and further propagated in the presence of the T cell growth factor interleukin-2. The T cells were metabolically labeled with D-[1-14C]galactose and D[1-14C]glucosamine, and the gangliosides were extracted and purified by DEAE-Sepharose column chromatography. Carbohydrate backbone structures of the asialogangliosides, prepared by mild acid hydrolysis, were determined by high-performance liquid chromatography, treatment with exoglycosidases and immunostaining. Monosialylated gangliosides were isolated by gradient elution from DEAE-Sepharose and further separated by preparative high-performance thin-layer chromatography in two solvent systems. Isolated fractions were characterized by preparation of asialogangliosides by mild acid hydrolysis, the action of Vibrio cholerae neuraminidase, and fast-atombombardment mass spectrometry. The following structures were identified: IVNeuAc-GgOse4Cer; IVNeuGc-GgOse4Cer; IVNeuAc-GgOse5Cer; and IVNeu-Gc-GgOse5Cer. The latter two gangliosides were not detected on B lymphoblasts and may be T-cell-specific structures. All gangliosides were heterogeneous in their ceramide moieties, being substituted with C16:0, C24:0, and C24:1 fatty acids. A preliminary study of several other mouse strains showed no strain-specific genetic variations in the T cell gangliosides. The possible role of these gangliosides is discussed.

[Neutral glycosphingolipids of Fabry's disease lymphoblastoid lines established by Epstein-Barr virus transformation]

Human lymphoid cell lines established by Epstein-Barr viral transformation of peripheral B-lymphocytes from normal subjects and from Fabry patients, were investigated for their ability to biosynthesize neutral glycosphingolipids from [14C]galactose and [14C]glucose as precursors. Galactose was taken up in the presence of high concentrations of glucose and selectively utilised by the cells in the synthesis of galactosphingolipids. The pattern of neutral glycosphingolipids labelled from [14C]galactose was slightly modified with time of labelling in either lymphoid cell line: the first labelled glycosphingolipid was lactosylceramide (LacCer) in the normal line and globotetraosylceramide (GbOse4Cer) in the Fabry line. After labelling for 96 h, a steady state was reached and the percentage of every type of labelled glycosphingolipid was stable in each cell line; however, differences in the neutral sphingolipid composition appeared between the various cell lines. When using radiolabelled glucose as precursor, the major part of the radioactivity was incorporated into neutral lipids and phospholipids; neutral sphingolipids were much less labelled than when using galactose. Catabolism of endogeneous labelled glycosphingolipids (synthesized by the cells during the 'pulse') was studied after cultivating the cells without radiolabelled precursors ('chase'). In the cells from normal subjects, all the neutral glycosphingolipids were slowly degraded (half-life time around 15-25 days for LacCer and GbOse3Cer). In contrast, in a lymphoid line from a Fabry patient, no appreciable degradation of GbOse3Cer occurred during 30 days. This block in the catabolism of GbOse3Cer is in good agreement with the previously reported deficiency of alpha-galactosidase A activity in this Fabry lymphoid cell line [Salvayre, R. et al. (1981) Biochim. Biophys. Acta 659, 445-456] and demonstrates that alpha-galactosidase B does not hydrolyze GbOse3Cer in the living cell (in contrast to the situation in vitro).

Murine leukemia. Proposed role for gangliosides in immune suppression

Glycolipid-bound sialic acid levels were elevated 2 to 4-fold in the sera of two strains of mice bearing thymic lymphoma produced either spontaneously (AKR/J) or due to chemical carcinogenesis [Swiss mice injected with 5-(3,3-dimethyl-1-triazeno)imidazole-4-carboximide]. The serum glycolipid-bound sialic acid level reflected the tumor burden of the AKR/J mice during the early stages of leukemogenesis. Furthermore, the elevation was found to coincide with the ontogenesis of thymic lymphoma and not to be simply an age-dependent phenomenon. TLC analysis of Florisil column-purified gangliosides from the sera of AKR/J and Swiss mice suggested presence of gangliosides with mobilities very close to GM2 and GM3 standards respectively. On the premise that the elevated glycolipid levels in circulation might interfere with normal lymphocyte functions, the immunoinhibitory properties of exogenously added mixed gangliosides were examined on tests of in vitro correlates of the immune response. Gangliosides inhibited concanavalin A and lipopolysaccharide-induced [3H]-thymidine, [14C]-leucine and [3H]-lysine uptake by normal AKR/J mouse thymocytes and spleen cells. Mixed gangliosides also suppressed the two-way mixed lymphocyte reaction of AKR/J X Swiss and AKR/J X DBA/2 spleen cells. These and other results strongly suggest a general immunologically relevant role for gangliosides in the ontogeny of thymic lymphoma of mice.

Ganglioside composition of rabbit thymus

Gangliosides of rabbit thymus were extracted and analyzed by the ganglioside-mapping procedure developed previously. 1 g of thymus contained 205.1 nmol of lipid-bound sialic acid and the relative amounts of monosialoganglioside and di- and trisialoganglioside fractions were 76.6 and 23.3%, respectively, as based on lipid-bound sialic acid. No ganglioside containing N-acetylgalactosamine was detected in rabbit thymus. The predominant component was N-glycoloylneuraminosyl lacto-N-norhexaosyl ceramide, NeuGc(alpha, 2-3)Gal(beta, 1-4)GlcNAc(beta, 1-3)Gal(beta, 1-4)GlcNAc(beta, 1-3)Gal(beta, 1-4)Glc(beta, 1-1) ceramide, which constituted 38.4% of the total gangliosides. The other major gangliosides were N-glycoloylneuraminosyl lacto-N-neotetraosyl ceramide (31.3%), GD3 containing N-glycoloylneuraminic acid (11.8%), GM3 containing N-glycoloylneuraminic acid (10.6%) and GM3 containing N-acetylneuraminic acid (6.4%). C-18 sphingosine was the only long-chain base in all the gangliosides. Palmitic acid was the major fatty acid of thymus gangliosides and chromatographically more polar gangliosides contained higher proportions of palmitic acid: 46.3% in GM3, 47.5% in GD3, 60.2% in N-glycoloylneuraminosyl lacto-N-neotetraosyl ceramide and 89.6% in N-glycoloylneuraminosyl lacto-N-norhexaosyl ceramide.

Comparative study on ganglioside compositions of various rabbit tissues. Tissue-specificity in ganglioside molecular species of rabbit thymus

Ganglioside compositions of various organs of rabbit (NIBS strain, male, 10 months old) were studied. Organs examined contained lipid-bound sialic acid at various concentrations but the amounts in extraneural tissues were less than one-fifth of that in brain. The gangliosides of various tissues were analyzed by ganglioside-mapping and by isolating individual components and determining their structures chemically or enzymatically. According to their backbone asialocarbohydrate chain, the major gangliosides of various tissues were classified into three groups: (1) lactose and ganglio-N-triose (lung, stomach, liver, intestine, kidney, testis and muscle); (2) ganglio-N-tetraose (brain); (3) lacto-N-neotetraose (thymus). 70% of all thymus gangliosides had a lacto-N-neotetraose backbone, which was tissue-specific. In marked contrast to the case in other tissues, in thymus N-glycoloylneuraminic acid constituted 90% of the total lipid-bound sialic acid, and all molecular species of thymus gangliosides contained N-glycoloylneuraminic acid. Palmitic acid was a major fatty acid of thymus gangliosides. Distinct differences were found in the fatty acid compositions of gangliosides with longer carbohydrate chains in various tissues.

Carbohydrate-labelled glycoproteins as markers of human lymphocyte subsets stimulated with mitogen and alloantigen

Human lymphocytes from thymus, spleen, and blood were stimulated by concanavalin A, phytohemagglutinin or pokeweed mitogen to incorporate radioactive thymidine and various carbohydrates. Peripheral-blood lymphocytes were also activated by alloantigen in a mixed lymphocyte reaction. Incorporation of thymidine and carbohydrates was parallel in stimulated cells. Carbohydrate-labelled cells were extracted with Triton X-100 buffer and the extracts subjected to sodium dodecylsulfate/polyacrylamide gel electrophoresis and subsequent autoradiography. From a comparison of the resulting glycoprotein patterns the following conclusions could be drawn. (a) The same population of thymocytes responds to all three mitogens. (b) Pokeweed-mitogen-stimulated blood lymphocytes are similar to this mitogen-responsive thymocyte population. (c) The glycoprotein pattern of pokeweed-mitogen-responsive spleen cells shows two characteristic glycoproteins, GP 50 and GP 78, and resembles that of murine B cells. (d) Whereas concanavalin A and phytohemagglutinin activate identical subpopulations of T cells in the spleen, this is not the case in the blood. (e) Concanavalin-A-activated and alloantigen-activated blood lymphocytes express a glycoprotein GP 185 which is not found on phytohemagglutinin-responsive cells. (f) The kinetic relationship during alloantigen stimulation in a mixed lymphocyte reaction between the appearance of glycoprotein GP 185 and subsequent maximal killing activity in an assay using release of 51Cr makes it likely that glycoprotein GP 185 is a marker for cytotoxic T lymphocytes.

Gangliosides of calf thymus and of normal and leukemic bovine lymphocytes

The gangliosides of calf thymus and of lymphocytes from blood, lymph nodes, lymph and spleen of normal and leukemic cows were investigated in an attempt to determine whether there exists a relation between the ganglioside composition and the maturity of the lymphocytes. With all normal peripheral lymphocytes studied the largely prevailing (up to 97%) ganglioside component was found to be N-glycoloylneuraminosyllactosylceramide. The ganglioside spectrum of calf thymus was much more complex and included at least six different components. By thin-layer chromatography, methylation analysis and neuraminidase treatment they were identified as N-glycoloylneuraminosyl and N-acetylneuraminosyl lactosylceramides, N-glycoloylneuraminosylneolactotetraosylceramide, di-(N-glycoloylneuraminosyl), di-(N-acetylneuraminosyl) lactosylceramides and N-glycoloylneuraminosyl-N-acetylneuraminosyllactosylceramide. The ganglioside spectra of leukemic peripheral lymphocytes were different from those of normal peripheral lymphocytes and resembled the ganglioside profile of the thymus. The data obtained indicate that, in the normal animal, the development of thymocytes into mature peripheral lymphocytes is accompanied by loss of disialosyl-gangliosides and sialosylneolactotetraosylceramide. It is concluded that both the structure and the composition of the lymphocyte ganglioside become more simple during maturation and ageing of the lymphocyte.

Studies on glycosphingolipid biosynthesis by lectin-stimulated human lymphocytes

1. Human peripheral lymphocytes, normal and stimulated by phytohemagglutinin or concanavalin A. were investigated with respect to their ability to biosynthesize neutral and acidic glycosphingolipids from D-[U-14C]glucose, D-[U-14C]galactose and D-[U-14C]glucosamine as precursors. 2. Galactose and glucosamie are taken up selectively, in the presence of excessive glucose concentrations. Labeling of total neutral glycosphingolipids from D-[U-14]galactose in normal cells decreases after the first 6 h while in stimulated cells there is a fourfold and a sevenfold increase after 6 and 12 h of incubation respectively. Under similar conditions stimulation with concanavalin A gives a fourfold increase after 12 h of incubation. 3. Analysis of individual glycosphingolipids biosynthesized from [U-14C]galactose shows that lactosylceramide is the major radioactive neutral glycosphingolipid and that stimulation by phytohemagglutinin yields an almost sevenfold and 13-fold increase in the radioactivity incorporated within 6 and 12 h of incubation respectively. Glucosylceramide shows about an eightfold increase, globotetraosylceramide a threefold increase and globotriaosylceramide a fourfold increase. The rate of incorporation into glucosylceramide of stimulated lymphocytes declines after 6 h of incubation, accompanied by a concomitant increase of incorporation into lactosylceramide. 4. At 12 h of incubation the ratios of radioactivities incorporated into neutral glycosphingolipids of phytohemagglutinin-stimulated cells compared to normal cells were 1.0 for D-[U-14C]glucose 7.0 for D-[U-14C]galactose and 2.5 for D-[U-14C]glucosamine. Respective ratios for lactosylceramide are 1.0 for [U-14C]glucose and 13.0 for [U-14C]galactose. These differences did not arise from changes of the uptake of the glycosyl precursor by the cell due to stimulation. 5. Incorporation of D-[U-14C]galactose into II3-N-acetylneuraminosyllactosylceramide by cells stimulated by phytohemagglutinin and concanavalin A is enhanced 14-times and 15-times respectively. With [U-14C]glucosamine as precursor, this increase in the labeling is much more impressive, 80-fold after 12 h of incubation by the phytohemagglutinin-stimulated lymphocytes. Neuraminidase treatment and gas radiochromatographic analysis of the labeled compound derived from [U-14C]-galactose as precursor indicate that 89% of the radioactivity was incorporated into the glucosyl and galactosyl moieties, in a ratio 1:1. With [U-14C]glucosamine as precursor, a selective labeling of the sialyl moiety of the II3-N-acetylneuraminosyllactosylceramide was indicated. 6. The pattern of complex gangliosides (more complex than II3-N-acetylneuraminosyllactosylceramide) which are biosynthesized after phytohemagglutinin stimulation of the cells, show no significant differences when compared to the patterns obtained from normal cells.

Cell cycle dependent rate of labelling of cellular and secreted glycosaminoglycans in mouse embryonic fibroblasts

Cultures of embryonic fibroblasts from Balb/c or CBA/J mice were given 12-h pulses of 14C-galactose, or were double-labelled with 3H-galactose and 35H-sulfate. The time course of the rates of labelling of glycosaminoglycans--galactose label was found in the uronic acid moiety--was studied in synchronously and asynchronously growing cultures. Partial synchrony was achieved by trypsinising quiescent, confluent cells and subsequent transfer of cells to new cultures with fresh medium. Synchrony was monitored by measurement of thymidine uptake in parallel cultures. The distribution of label in the hyaluronic acid, chondroitin sulfate, and heparan sulfate fractions from cells and culture media was determined at each time point. Peaks of DNA synthesis were accompanied by or followed 12 h later by a maximal rate of labelling with galactose of secreted glycosaminoglycans, and with the exception of hyaluronic acid--also of cellular glycosaminoglycans. The rate of labelling with galactose of glycosphingolipids in parallel cultures followed a different time course. In double-label experiments the rates of labelling of glycosaminoglycan sulfates with 3H-galactose and 35S-sulfate did not go parallel. In older, quiescent cultures the labelling rate with galactose decreased while the sulfation rate increased. It is discussed that the labelling rate with galactose is indicative of the biosynthetic rate of the glycosaminoglycans. The conclusion is reached that glycosaminoglycans are preferentially synthesized and secreted after the S phase of the cell cycle.

Biphasic rate of synthesis of glycoconjugates, phospholipids and DNA in concanavalin A-stimulated mouse thymocytes. Involvement of cortisone-sensitive and -resistant subpopulations

The time course of the rate of labeling of membrane components (phospholipids, glycolipids and glycoproteins) and DNA was followed in concanavalin A-stimulated CBA/J mouse thymocyte cultures. Two peaks of stimulated biosynthetic activity were noted, the first at the beginning of the cultivation and the second about 25 h later. Both early and late peaks of biosynthesis of membrane components were accompanied by blast transformation and were unimpeded by suppression of DNA synthesis by hydroxyurea. Cortisone-sensitive and cortisone-resistant thymocytes were prepared by selective agglutination of the cortisone-sensitive cells with peanut agglutinin (Reisner et al. Cell. Immunol. 1976. 25: 129) or cortisone treatment of the animals. Cortisone-sensitive cells responded early, while the cortisone-resistant population gave only the late response. The autoradiographic patterns from sodium dodecyl sulfate polyacrylamide gels of [3H]fucose or [3H]galactose-labeled glycoproteins from early and late labeling cells, and cortisone-resistant cells, were compared. Late-labeling and cortisone-resistant cells gave indistinguishable patterns, but differed significantly in their patterns from early-labeling cells. It is concluded that the two peaks of biosynthetic activity during the course of concanavalin A stimulation of thymocytes are caused by two different cell populations which require different times for maximal response and react independently of one another.

Metabolic carbohydrate-labelling of glycolipids from mouse splenocytes. Mitogen-stimulated B and T cells show different labelling patterns

Splenic lymphocytes from CBA/J, AKR/A/J, BALB/c/A, C57/BL/6J, C3H/HeJ and C3H/Tif nu/nu mice and B lymphocyte or T lymphocyte preparations derived from CBA/J mouse spleen were cultivated in the presence of either concanavalin A, phytohemagglutinin, Salmonella minnesota R595 lipopolysaccharide or Proteus mirabilis soluble lipoprotein. The mitogens stimulated the incorporation of [14C]galactose into acid-insoluble cell material with the same specificity for B or T cells as that known for thymidine incorporation. The glycolipids extracted from mitogen-activated, carbohydrate-labelled B or T cells were compared by thin-layer chromatography and characteristic differences between B and T cells were noted in the ganglioside as well as in the neutral glycolipid fractions. In addition, subsets of B or T cells, namely lipopolysaccharide-responsive or lipoprotein-responsive B-cell populations or nylon-purified T cells may be recognized by characteristic neutral glycolipid bands.