Gangliosides and glycophorin inhibit T-lymphocyte activation.

Gangliosides and glycophorin inhibit T-lymphocyte activation FRANCES J. SHAROM,'ANITAL. H. CHIU,and T. ELAINEROSS

Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by San Diego (UCSD) on 12/31/14 For personal use only.

Guelph- Waterloo Centre for Graduate Work in Chemistry, Department of Chemistry and Biochemistry, University of Guelph. Guelph, Ont., Canada NIG 2W1 Received August 25, 1989 SHAROM, F. J., CHIU,A. L. H., and Ross, T. E. 1990. Gangliosides and glycophorin inhibit T-lymphocyte activation. Biochem. Cell Biol. 68: 735-744. Increased levels of gangliosides in the serum have been linked to tumour-induced immunosuppression in vivo. Both bovine brain gangliosides and human erythrocyte glycophorin were potent inhibitors of concanavalin A, periodate, and phorbol ester - ionomycin induced activation of murine T-lymphocytes. Structurally complex gangliosides were more inhibitory, while simpler glycolipids caused less inhibition. Lymphocytes exposed to these molecules for up to 24 h could still proliferate after washing. Substantial inhibition was observed only when gangliosides and glycophorin were present during the first 18 h of activation. Studies using Quin-2 showed that gangliosides did not block the initial rapid rise in cytoplasmic c a 2 + following mitogen stimulation. Interleukin-2 (IL-2) production by ganglioside- and glycophorin-treated lymphocytes was unchanged. After treatment with gangliosides for 24 h, lymphocytes proliferated normally in response to added IL-2.These results suggest that the first round of signal transduction in response to mitogen was unaffected by gangliosides. Addition of pa~lgllasidesto activated lymphocytes in the presence of IL-2 resulted in complete inhibition o f proliferation. lrnmunosuppression by gangliosides and glycophorin thus appears to occur at Ihe Il-2dependent stage of proliferation and may be partially due to IL-2 binding to these molecules. However, high levels of IL-2 failed to reverse inhibition and IL-2-dependent cell lines were much less sensitive to ganglioside inhibition than T-iymphwytes, suggesring that more t h a n one mechanism of inhibition likely exists. Key words: gangliosides, glycophorin, T-lymphocyte, interleukin-2, interleukin-2 receptor.

F. J., CHIU,A. L. H., et Ross, T. E. 1990. Gangliosides and glycophorin inhibit T-lymphocyte activation. SHAROM, Biochem. Cell Biol. 68 : 735-744. Les taux accrus de gangliosides dans le serum sont relies a l'immunosuppression in vivo induite par une tumeur. Les gangliosides du cerveau de boeuf et la glycophorine erythrocytaire humaine sont de puissants inhibiteurs de l'activation des lymphocytes T murins induite par la concanavaline A, le periodate et le phorbol ester - ionomycine. Les gangliosides de structure complexe sont plus inhibiteurs et les glycolipides plus simples produisent une inhibition moindre. Apres 24 h d'exposition a ces molecules, les lymphocytes peuvent encore prolifkrer apres lavage. L'inhibition n'est substantielle que si les gangliosides et glycophorine sont prksents durant les 18 premitres heures de l'activation. Les ktudes avec le Quin-2 montrent que les gangliosides ne bloquent pas 1'61kvation initiale rapide du c a Z +cytoplasmique a p r b stimulation par un mitogene. La production d'interleukine-2 (IL-2) par les lymphocytes traitks avec des gangliosides ou la glycophorine demeure inchangee. A p r b traitement avec les gangliosides durant 24 h, les lymphocytes proliftrent normalement en rkponse a I'addition de IL-2. Ces rksultats suggtrent que la premibre ronde de transduction des signaux en reponse a un mitogtne n'est pas affectke par les gangliosides. L'addition de gangliosides aux lymphocytes actives en presence de IL-2 engendre une complete inhibition de la proliferation. L'irnmunosuppression par les gangliosides et glycophorine se produirait donc a un stade de prolifkration dependant de IL-2 et elle serait partiellement due a la liaison de IL-2 a ces molkcules. Cependant, des t a w BevQ de IL-2 ne peuvent renverser I'inhibition et les lignkes cellulaires dependantes de IL-2 sont beaucoup moins sensibles a l'inhibition par les gangliosides que les lymphocytes T, preuve de I'existence probable de plus d'un mecanisme d'inhibition. Mots clPs : gangliosides, glycophorine, lymphocytes T, interleukine-2, recepteur d'interleukine-2. [Traduit par la revue]

Introduction Gangliosides are a family of sialic-acid-containing glycosphingolipids present in the plasma membrane of marnmalian cells. They are known to play an important role in cell surface recognition processes such as contact inhibition and adhesion, and have recently been proposed to act as bimodal regulators of cell growth, capable of transducing both positive and negative growth signals (Spiegel and

Fishman 1987). Gangliosides also appear to modulate various immune processes and have been observed to augment some responses and inhibit others. For example, the ganglioside GMl has been implicated in the activation of thymocytes (Spiegel et al. 1985) and gangliosides inserted into the membrane of intact lymphocytes are able to stimulate a mixed lymphocyte reaction in autologous cells (Sela 1980). Gangliosides can also act as potent inhibitors of lymphoproliferative responses. They inhibit the ability of human ABBREVIATIONS: IL-2, interleukin-2; LPS, lipopolysaccharide; monocytes to secrete IL-1 following LPS stimulation, by a TPA, 12-0-tetradecanoylphorbol-13-acetate; LIS, lithium diiodomechanism which apparently involves ganglioside blocking salicylate; SDS-PAGE, sodium dodecyl sulfate - polyacrylamide of specific LPS binding to endotoxin receptors in the gel electrophoresis; ConA, concanavalin A; rIL-2, recombinant monocyte membrane (Cavaillon et al. 1987). Gangliosides interleukin-2; Quin-2-AM. 2-([bis(carboxymethyl)amino-5have also been shown to inhibit the in vitro proliferation methylphenoxy]methyl~-6-metho~y-~-bis(carboxymethyl~ of a variety of B- and T-lymphocytes stimulated with soluble aminoquinotine tetrakis(acetoxymethy1)ester; DMSO, dirnethyl antigens and lectins (Lengle et al. 1979; Whisler and Yates su1Foxide; kDa, kilodalton(s); IL-2r, interleukin-:! receptor; EGF, 1980; Tsuru et al. 1983; Ladisch et al. 1984; Marcus et al. epidermal growth factor; PDGF, platelet-derived growth factor; gangl, gangljoside. 1987; Jackson et al. 1987), as well as inhibiting the function 'Author l o whom a11 correspondence should be addressed. of T-helper cells (Offner and Vandenbark 1985) and the

Printed in Canada / Imprim6 au Canada

BIOCHEM. CELL BIOL. VOL. 68,


736

cytotoxic activity of natural killer cells (Dyatlovitskaya et al. 1986). The immunoregulatory properties of gangliosides and other glycosphingolipids have been reviewed by Marcus (1984) and Dyatlovitskaya and Bergelson (1987). Levels of circulating gangliosides up t o 50-fold normal have been found in the serum of cancer patients and tumourbearing animals, and serum ganglioside concentrations appear to correlate with progress of the disease, declining in response t o therapy or tumour removal and reappearing in patients whose tumours recur (Kloppel et al. 1977; Ladisch et al. 1987). It, therefore, seems likely that these gangliosides originate from the tumour itself. This proposal has been lent credence by the observation that tumour cells actively shed gangliosides from their plasma membrane in relatively large quantities and that the rate of shedding appears t o increase with the tumorigenic and metastatic potential of the cells (Ladisch et a/. 1983; Young et a/. 1986; Kasparayan and Yogeeswaran 1987). Tumour-derived circulating gangliosides may be responsible for the generalized immunosuppression seen in many cancer patients (Lengle et al. 1979; Krishnaraj et al. 1982; Tsuru et al. 1983) and ganglioside shedding could thus contribute t o the escape of tumour cells from destruction by the immune system of the host. I n this study, we have investigated the ability of gangliosides t o inhibit the proliferation responses of T-lymphocytes t o both primary mitogens (such as lectins, periodate, and TPA-ionomycin) and the growth factor IL-2, which is necessary for proliferation of mitogen-stimulated lymphocytes. We have shown that human erythrocyte glycophorin, a n integral membrane sialoglycoprotein, is also capable of inhibiting the immune response of T-cells in a very similar fashion t o gangliosides. Various aspects of the mechanism by which gangliosides and glycophorin exert their immunosuppressive effects are examined and it is suggested that, although these molecules appear to act primarily at the IL-2dependent stage of lymphocyte activation, multiple mechanisms of inhibition probably exist.

Materials and methods Glycolipids Mixed bovine brain gangliosides and purified GM,, GD,,, GTlb, asialo-GM,, and globoside were obtained from Supelco Canada (Oakville, Ont.). Gangliosides modified on sialic acid residues were prepared by treating mixed gangliosides with 2.5 mM sodium periodate at 4°C in the dark for 15 min, followed by quenching with ethylene glycol, reduction with sodium borohydride, dialysis, and lyophilization. Intact sialyl oligosaccharide moieties were cleaved from bovine brain gangliosides by ozonolysis and base hydrolysis according to the method of Smith et al. (1981). Cerebrosideswere purchased from Sigma Chemical Co. (St. Louis, MO). All glycolipids were stored in CHC1,-methanol solutions at - 20°C. For addition to lymphocyte cultures, glycolipids were dried down in a stream of nitrogen and pumped extensively under vacuum to remove traces of organic solvent. Gangliosides were added to lymphocytes in 96-well microplate culture dishes as micellar solutions in culture medium. Other glycolipids were vortexed vigorously in 37°C culture medium, using small diameter glass beads, until a homogeneous dispersion was obtained. Preparation of glycophorin Human erythrocyte ghosts were prepared according to Dodge et al. (1963). Glycophorin was isolated from ghosts by either the method of Marchesi and Andrews (1971), using extraction with LIS (Sigma) followed by extensive ethanol washing, or by sodium

1990

deoxycholate extraction and subsequent removal of residual detergent on a Sephadex G-50 gel filtration column (Segrest et al. 1979). The final products were homogeneous as shown by SDSPAGE. Glycophorin was desialylated by treatment with 0.1 M HCl at 80°C for 1 h (Lee and Grant 1980), followed by neutralization, dialysis, and lyophilization. An assay for sialic acid (Massamiri et al. 1979) showed that more than 92% of the sialic acid was removed by this treatment. Glycophorin, asialoglycophorin, and calf fetuin (Sigma) were added to cell cultures as solutions in culture medium.

Lymphocyte cultures Mouse splenocytes were obtained from 6- to 10-week-old female Swiss-Webster mice. Fresh aseptically removed spleens were passed through a tissue sieve into sterile RPMI 1640 - Hepes (Gibco Canada, Burlington, Ont.), followed by filtration through a fine wire mesh and harvesting of the cells by centrifugationat 1000 x g for 10 min. Lymphocytes were then purified by centrifugation through Ficoll-Isopaque (Davidson and Parish 1975), followed by washing in RPMI 1640 - Hepes and resuspension in culture medium. Cells were cultured in 96-well microplates at 1.25 x lo6 cells/mL in RPMI 1640 - 10% fetal calf serum - 50 pg gentamycid mL, in a humidified 37°C atmosphere of 5% COz. ConA (Sigma) was added at 2 pg/mL for lectin activation studies. For periodate activation, cells were incubated with 1 mM sodium periodate for 15 min at 4°C in the dark, and then washed and cultured as above. For stimulation by a phorbol ester - ca2' ionophore combination, lymphocytes were treated with 10 nM TPA (Sigma) and 0.3 pM ionomycin (Calbiochem, San Diego, CA). Monosaccharides were added as solutions in culture medium. Following a 48-h incubation period, lymphocytes were pulsed with 1 pCi t 3 ~ ] thymidine/well (6.7 Ci/mmol, ICN Canada, Montrtal, Que; 1 Ci = 37 GBq) and harvested 18 h later using a Titertek automatic cell harvester. Filter paper discs were dried and counted using an anhydrous scintillant. The IL-2-dependent murine T-cell lines HT-2 and CTLL-2 were maintained in RPMI 1640 - 10% fetal calf serum containing 5-30 U/mL of Jurkat-derived IL-2, in a humidified atmosphere of 5% CO, at 37°C. For measurement of inhibition of proliferation, cells were washed and cultured with 0.2 or 1 U/well (1-5 U/mL) of human rIL-2 (Boehringer-Mannheim Canada, Dorval, Que.), in the presence of various concentrations of gangliosides and glycophorin. For experiments involving the addition of high levels of IL-2, lyophilized human rIL-2 (from Cetus, a kind gift from Dr. G. Mills) was dissolved at a concentration of 8 x lo6 U/mL before addition to cell cultures. For measurement of IL-2 concentration, cells were washed and incubated with various dilutions of the solution to be tested, as well as known amounts of rIL-2 as a positive control. After incubation for 18 h (HT-2) or 42 h (CTLL-2), cells were pulsed for 6 h with 1 pCi [3~]thymidine/welland harvested, and the filters were counted as above. Time course of inhibition For determination of the periods during the activation process when gangliosides and glycophorin were capable of causing inhibition, lymphocytes were activated at time 0 with ConA and either mixed gangliosides (10 pg/well) or glyco horin (50 pg/well) were added at various times after initiation. [P HIThymidine incorporation into the cells was measured as described above. Measurement of IL-2 production 1L-2 production by ganglioside- and glycophorin-inhibited T-lymphocytes was measured using the IL-2-dependent cell lines HT-2 and CTLL-2. Mouse splenocytes were activated with ConA in the presence of increasing amounts of mixed gangliosides and glycophorin. After a 24-h incubation, the supernatant containing secreted 1L-2 was removed and various dilutions were added to the IL-2-dependent test cells, followed by determination of [ , ~ ] t h ~ m i d i nincorporation e as above. To determine the effects

737

SHAROM ET AL.

TABLE1. Concentrations of glycosphingolipids and glycoproteins causing 50% inhibition of ConA-stimulated murine T-lymphocyte activation

Species added


.-s

-2

0

2

4

6

8

10

a,

s

p g p e r well gangliosides

Concentration required for 50% inhibition, udmL

Glycosphingolipids GM, GDl, mixed GT,, Asialo-GM, Globoside Cerebrosides Sialyloligosaccharidesa Glycoproteins Glycophorin Asialoglycophorin Fetuinb 'No significant inhibition was seen at 500 pg/mL (-300 pM). b ~ e s than s 20% inhibition was seen at 1250 pg/mL.

p g p e r well glycoprotein

FIG. 1. Inhibition of ConA-stimulated proliferation of mouse splenic T-lymphocytes by gangliosides (A) and glycoproteins (B). Purified gangliosides and glycoproteins (see Materials and methods) were dissolved in culture medium and incubated with lymphocytes in the presence of 2 pg ConA/mL. Proliferation was measured as [3~]thymidine incorporation into cellular DNA. Results are presented as the mean f SE (n = 3) from a representative experiment.

at 37°C. Fluorescence measurements were performed using a Perkin-Elmer model LS-5 luminescence spectrometer with a quartz cuvette at 37"C, with excitation at 339 nm and emission at 495 nrn. Immediately prior to the start of each run, cells were washed rapidly, resuspended in buffer, and transferred to the cuvette. The stable resting fluorescence was monitored for several minutes prior to addition of ConA (2 pg/mL), or mixed gangliosides (50 pg/mL) followed by ConA. The system was calibrated by the addition of 0.5 pM ionomycin (F,,), followed by 2 mM ~ n ' +(F,,), and changes in [ca2+]were estimated as described previously (Tsien et al. 1982).

Results Inhibition of ConA-stimulated T-lymphocyte activation by glycolipids and glycoproteins Gangliosides are potent inhibitors of ConA-stimulated activation when added to the culture medium of mouse splenocytes, as shown in Fig. 1A. The dose-response to indiof carry-over of gangliosides and glycophorin on the HT-2 and vidual ganglioside species depended on the complexity of CTLL-2 cultures, an amount of each equivalent to that transferred the glycolipid oligosaccharide structure, with GTlb > in the IL-2 supernatant was mixed with rIL-2 and [3~]thymidine GDIa = mixed > GMI. Concentrations required to proincorporation in the test cells was assessed. duce 50% inhibition of activation were in the range of 3-6.5 pg/well(15-32.5 pg/mL, 10-15 pM). Modification of Determination of IL-2 responsiveness Mouse spleen lymphocytes were cultured as described above with the sialic acid residues of mixed brain gangliosides by 2 pg ConA/mL in the presence or absence of 100 pg mixed oxidative cleavage of the side chain with periodate did not gangliosides/mL. After 24 h, the cells were incubated with 50 mM reduce their inhibitory potency. Neutral glycolipids were also a-methyl mannoside for 30 min at 37"C, harvested by centrifugatested for their ability to inhibit lectin-mediated activation tion, and washed with 50 mM phosphate-buffered saline. Washed (see Table 1). Globoside and cerebrosides were 10-fold lymphocytes were then incubated with exogenous IL-2, in the prespoorer than gangliosides in their ability to inhibit activation, ence or absence of 100 pg mixed gangliosides/mL, for a further while asialo-GM1 was as effective an inhibitor as GM, 24 h, pulsed with [3~]thymidinefor 24 h. and harvested as itself. described above. Human erythrocyte glycophorin is a highly sialylated Determination of intracellular calcium concentration 31-kDa integral membrane glycoprotein, which is readily Free cytoplasmic calcium concentrations were measured using soluble in aqueous solutions, where it exists as an aggregated the fluorescent calcium chelator Quin-2, as described by Tsien et al. structure. Glycophorin also possesses the ability to suppress (1982). Murine splenocytes at 5 x lo6 cells/mL in RPMI 1640 ConA-mediated T-lymphocyte activation (see Fig. lB), with Hepes - 10% fetal calf serum were loaded with Quin-2-AM 50% inhibition seen at -60 pg/well (300 pg/mL, 10 pM). (Calbiochem) by adding the appropriate volume of a 2 mM stock Glycophorin thus appears to be as effective an inhibitor on solution in DMSO, to give a final concentration of 5 p M . Following a molar basis as gangliosides; however, since each glycoa 1-h incubation at 37"C, the cells were washed twice at 400 x g and resuspended in buffer (140 mM NaCl, 3 mM KCI, 1.8 mM phorin molecule possesses 16 short 0-linked sugar chains, c a 2 + , 1 mM M ~ ' + ,10 mM glucose, and 10 mM Hepes, pH 7.2) the effectiveness is lower if considered on the basis of

738

BIOCHEM. CELL BIOL. VOL. 68, 1990

10 p g gangliosides

TABLE2. Effects of various treatments on inhibition of murine T-lymphocyte activation by 50 pg mixed gangliosides/mL

50 p g glycophorin

Treatment


None 24 h preincubation and wash None None 50 nM TPA

time of addition, h o u r s

FIG. 2. Inhibition of proliferation of murine splenic lymphocytes by gangliosides (10 pg/well, 50 pg/mL) and glycophorin (50 pg/well, 250 pg/mL) added to cultures at various times following ConA stimulation at time 0. Proliferation was measured as [3~]thymidineincorporation into cellular DNA. Results are presented as the mean + SE (n = 3) from a representative experiment.

-

oligosaccharides. Glycophorin purified using LIS is known to retain several molecules of the detergent in a tightly bound state (Romans and Segrest 1978). To rule out the possibility that this bound detergent could interfere with lymphocyte proliferation, glycophorin was isolated using sodium deoxycholate, a procedure which results in an almost detergentfree final product (Segrest et al. 1979). Deoxycholatepurified glycophorin was found to have identical inhibitory properties to that of the LIS preparation and was used in all further experiments. When glycophorin was stripped of more than 92070 of its sialic acid residues, the resulting asialoglycophorin proved to be a more effective inhibitor (on a weight basis) than glycophorin itself. On a molar basis (correcting for the loss of sialic acid which makes up about 32% of the glycophorin molecule), asialoglycophorin showed similar immunosuppressivepotency to glycophorin. The ability to suppress lymphocyte proliferation in this fashion was not a property of sialoglycoproteins in general. Fetuin, a soluble serum sialoglycoprotein, produced little inhibition at concentrations as high as 250 /.~g/well (1.25 mg/mL), as shown in Fig. 1B. Thus our data suggest that the specific structure and sequence of glycolipid and glycoprotein oligosaccharides, rather than the presence of sialic acid per se, are important for inhibition. The ability of gangliosides and glycophorin to inhibit lymphocyte proliferation is not caused by binding to the mitogen. Neither molecule has been reported to bind ConA and we have determined that they do not alter the doseresponse of lymphocytes to lectin. Toxicity also does not seem to be a factor; neither gangliosides nor glycophorin affected lymphocyte viability (as assessed by trypan blue exclusion) in the concentration range which inhibited cell proliferation. Several monosaccharides commonly found in glycolipid and glycoprotein oligosaccharidechains were tested for their ability to both inhibit ConA-stimulated proliferation and reverse ganglioside-mediated inhibition of proliferation.

Mitogen

%

inhibition

ConA

> 98

ConA Periodate TPA-ionomycin ConA

>98

18 88

>98

Galactose, N-acetylglucosamine, and N-acetylgalactosamine at concentrations of 25 mM had no effect on cell proliferation and did not alter the dose-response of ganglioside inhibition. N-Acetylneuraminic acid inhibited proliferation at concentrations higher than 7.5 mM (complete inhibition was seen at 25 mM), but 5 mM concentrations did not affect inhibition by gangliosides. Thus simple monosaccharides are unable to either inhibit lymphocyte proliferation or reverse inhibition, at concentrations >500-fold higher than those of the gangliosides themselves. Intact sialyl oligosaccharides, prepared from mixed gangliosides by ozonolysis and base cleavage, had no deleterious effects on lymphocyte activation at concentrations as high as 500 pg/mL (-300 pM, see Table 1) and were unable to reverse ganglioside-mediated inhibition.

Time course of inhibition To determine the point of action of gangliosides and glycophorin, lymphocytes were activated with ConA and the glycoconjugates were added to the culture medium at various times follawing initiation. As shown in Fig. 2, both gangliosides and glycophorin were most effective inhibitors of cell proliferation when added at the same time as the initiating lectin. They could be added 8 h after lectin stimulation and still have maximal inhibitory activity, and addition as late as 18 h after initiation still resulted in substantial inhibition. Their inhibitory potency was greatly reduced if they were added later than 32 h following initiation. Gangliosides, glycophorin, and asialoglycophorin all showed identical kinetics of inhibition. These results suggest that gangliosides and glycophorin do not interrupt the process of lymphocyte proliferation at a very early step in the activation sequence. When lymphocytes were preincubated with an inhibitory concentration of mixed gangliosides for 24 h, followed by washing to remove exogenous material, the response of the cells to ConA was normal (see Table 2), indicating that this treatment had not affected the ability of the cells to respond to mitogens. We have previously shown that micellar ganghosides bind rapidly to the lymphocyte membrane, with the bulk of the association occurring during the first 20-30 min, followed by a slower uptake phase which levels off after 2 h (Sharom and Ross 1986). Yet even a 24-h preincubation with gangliosides, which would result in sustained maximum surface binding, is not sufficient to cause inhibition after removal of the remaining soluble glycolipids. Inhibition of proliferation by these molecules is thus unlikely to be mediated by insertion of the molecules into the outer leaflet of the lymphocyte plasma membrane. Gangliosides labelled with 3~ on sialic acid residues were added to the culture medium and the cells were recovered at various times up to

739

SHAROM ET AL. gangliosides p e r well 0O P ~ EXXI 5I.1g 'OI.19 15/47 2 0 Pg

ezd

s

.-0


.a~

.-s

0

25 50 p g p e r well g a n g l i o s i d e s

2

I

g l y c o p h o r i n p e r well 0O C L ~ EXXI 25 50 ~ g Qa 7 5 ~ 9

75

5.

L .+.

I

HT-2 CTLL-2

e 0

25

50 75 p g p e r well g l y c o p h o r i n

100

FIG. 3. Inhibition of IL-2-stimulated proliferation of the lymphocyte cell lines HT-2 and CTLL-2 by gangliosides (A) and glycophorin (B). Purified gangliosides and glycoproteins were dissolved in culture medium and incubated with lymphocytes in the presence of 0.2 or 1.0 U/well(l-5 U/mL) of rIL-2, as described in Materials and methods. Proliferation was measured by [3~]thymidine incorporation into cellular DNA. Results are presented as the mean k SE (n = 3) from a representativeexperiment.

48 h following ConA stimulation. Only a small fraction (a few percent) of the labelled gangliosides was associated with the cells; the vast bulk of the glycolipid remained in the culture supernatant, where it probably forms complexes with serum proteins.

Effects on ZL-2-dependent lymphocyte proliferation Gangliosides and glycophorin were also found to inhibit IL-2-induced proliferation of the IL-2-dependent lymphocyte T-cell lines HT-2 and CTLL-2, as shown in Figs. 3A and 3B. However, considerably higher ganglioside concentrations were needed to inhibit proliferation in these cell lines relative to murine splenocytes; 50% inhibition values were in the range of 25-40 pg/well (125-200 pg/mL; 62.5-100 pM). Glycophorin remained inhibitory in the same concentration range that was effective for murine lymphocytes (35-70 pg/well; 140-280 pg/mL; 4.5-9 pM).

Effects of gangliosides and glycophorin on ZL-2 production Gangliosides and glycophorin could potentially interfere with lymphocyte activation by blocking the production of the lyrnphokine IL-2, which is required for the second stage of the activation process following signal initiation by mitogens. We thus measured IL-2 concentrations in the 24-h

FIG. 4. Secretion of IL-2 into the culture supernatant of mouse splenic lymphocytes was measured in the presence of gangliosides and glycophorin. Murine splenic lymphocytes were cultured with 2 pg ConA/mL and (a) 0-20 pg/well (0-100 pg/mL) of mixed gangliosides or (c) 0-75 pg/well (0-375 pg/mL) of human erythrocyte glycophorin. After 24 h, the culture supernatant was removed and assayed for IL-2 activity using the IL-2-dependent cell line HT-2. Parallel experiments measuring ['H]thymidine uptake showed that proliferation of the murine lymphocytes was almost completely abolished at 10 pg gangliosides/well and 75 pg glycophorin/well. The effects of carry-over of gangliosides or glycophorin in the spleen cell supernatant on the HT-2 test cells are shown in b and d, respectively. Results are expressed as a percentage of control (untreated cells) k SE (n = 3).

culture supernatants of ganglioside- and glycophorininhibited lymphocytes using the IL-2-dependent cell lines HT-2 and CTLL-2, which grow and divide only in response to the growth factor. Using such culture supernatants involves addition of some gangliosides and glycophorin to test cell lines, which in itself causes inhibition of IL-2-mediated proliferation. This caused fewer problems than anticipated because of the relative resistance of these cell lines to ganglioside inhibition compared with murine lymphocytes. Amounts of gangliosides and glycophorin equivalent to those transferred in the culture supernatant were added directly to the test cells, to assess the profile of inhibition caused by carry-over of the inhibitory molecules. As shown in Fig. 4, supernatants from mouse spleen lymphocytes treated with 10 pg/well gangliosides (which causes total inhibition of proliferation) contain the same levels of IL-2 as untreated lymphocytes, once the carry-over inhibition of the IL-2-dependent cells is accounted for. Even treatment with much higher levels of gangliosides (up to 20 pg/well) had only small effects on IL-2 production. Similarly, mouse lymphocytes treated with very inhibitory concentrations of glycophorin (50 pg/well) produced amounts of IL-2, which were only slightly reduced relative to untreated cells (see Fig. 4). It was not possible to test the effect of suprainhibitory levels of glycophorin on IL-2 secretion, since the IL-2-dependent test cells displayed a glycophorin sensitivity similar to that of mouse splenocytes. Addition of high concentrations of exogenous ZL-2 It has been suggested that gangliosides may interfere with lymphocyte proliferation by binding and sequestering the


740


units rlL-2 added x 1 o - ~ FIG. 5. Effect of high concentrations of exogenous rIL-2 on immunosuppression of ganglioside-treated T-lymphocytes. Mouse splenocytes were stimulated with ConA in the presence of 10 pg/well (SO (rg/mL) of mixed gmgliosides, as described in Materials and methods. Increasing amounts of rIL-2 (2 x 10' - 1 x 10' U/well, I x lo3 - 5 x 10' U / m L ) were added and cell proliferation was assessed by [%]thymidine uptake in the usual way. Controls with no g a n g h i d e s added showed that high levels of rIL-2 had no negative effects on the ability of the lymphocytes to proliferate. Proliferation of the control cultures reached a maximum at around 1.0 U/well (5 U/mL) or rIL-2.

growth factor IL-2 in the extracellular medium (Robb 1986). To determine whether such an IL-2 binding mechanism could be totally or partially responsible for ganglioside and glycophorin inhibition of mouse T-lymphocytes, large amounts of rIL-2 were added to inhibited cultures of mouse spleen lymphocytes. As shown in Fig. 5, addition of up to 50 000 U/mL (10 000 U/well) of rIL-2 failed to reverse inhibition by gangliosides, and glycophorin inhibition was similarly unaffected (results not shown). High levels of natural IL-2 were also unable to reverse ganglioside or glycophorin inhibition of T-lymphocyte activation. Lymphocytes stimulated with a combination of TPA and c a 2 + ionophores have been reported to secrete very high levels of IL-2 relative to cells stimulated with ConA alone (Mohr et al. 1986). Using the IL-2-dependent cell lines HT-2 and CTLL-2 to assay secreted IL-2 levels, we have confirmed that mouse spleen lymphocytes stimulated with TPA-ionomycin secrete at least 20-fold higher levels of IL-2 than those treated with ConA. Despite this large increase in IL-2 concentration, the inhibitory effects of gangliosides on the cells remained unchanged (see Table 2).

Effects of gangliosides on IL-2 responsiveness It is also possible that gangliosides and glycophorin inhibit T-lymphocyte proliferation by preventing the expression of high-affinity IL-2r on the cell surface. IL-2r expression is dependent on signal transduction following binding of the

a b c

d e f g

h i j k

controls

no g a n g l

+ gangl

FIG. 6. Responsiveness of ganglioside-treated lymphoblasts to L-2 following ganglioside treatmmt. Lymphocytes wwe stimulated with 2 pg C o r W m t in the presence (h-k) or absence (d-g) of mixed gangliosides. After 24 h, the lymphoblasts were washed to remove gangliosides, bound lectin, and endogenous IL2. and were cultured for a Further 24 h in the presence (e, g, i,k) or absence (d, f,h, j ) of exogenous IL-2. f, g, j, and k were also incubated with gangliosides during 24-h period of IL-2 stimulation. Controls did not receive the 24-h wash and were as follows: unstimulated cells (a), and cells stimulated with ConA in the absence (b) and presence (c) of mixed gangliosides. Proliferation was measured as [3~]thymidine incorporation into cellular DNA. Results are presented as the mean + SE (n = 3) from a representative experiment.

initial mitogen and is absolutely necessary for proliferation in response to JL-2 in the second stage of the activation process. Induced TL-2r appear on the cell surface 18-24 h following stimulation with ConA. If ganglioside-treated lymphocytes still express high-affinity receptors for IL2,then they should be able to proliferate in response to exogenous IL-2 at this point in time. Lymphocytes were stimulated with C o d in the presence or absence of gangliosides and after 24 h, the cells were washed with rr-methyl mannoside and then buffer. These wash steps were sufficient to remove both bound ConA and IL-2 secreted into the culture supernatant. Cells treated in this way were only able to proliferate if exogenous TL-2was provided (compare d and e of Fig. 6), confirming that they expressed IL-2r. After treatment with gangjiosides for the first 24 h after ConA stimulation, lymphoblasts remained competent to respond to IL-2 (Fig. 69. Interestingly, if gangliosides were present only during the period of exogenous IL-2 incubation, proliferation was completely abolished (Fig. 6, g and k). Thus gangliosides appear to have very potent inhibitory effects on the IL-2-mediated steps of lymphocyte proliferation.

Effects on activation by other mitogens The effects of gangliosides and gtycophorin on lymphocyte proliferation were not restricted to the use of ConA as the rnitogen. The dose-response curves for inhibition by gangliosides {and also glycophorin and asialoglycophorin, data not shown) were virtually identical when lymphocytes were activated by treatment with 1 mM sodium periodate, with complete inhibition observed at 10 ,ug/well (Table 2). Cell stimulation with a combination of the caZ' ionophorc ionomycin and the phorbol ester tumour promoter TPA bypasses the requirement for antigen or lectin-induced signal


SHAROM ET AL.

74 1

showed a small amount of autofluorescence when added to buffer alone, which was proportional to ganglioside concentration. Both crude and highly purified gangliosides showed this increase in background fluorescence, which could be due to either a minor fluorescent contaminant or light scattering by ganglioside micelles. However, the addition of 50 pg gangliosides/mL to Quin-2-loaded cells prior to stimulation did not prevent the ConA-stimulated c a 2 + increase (Fig. 7B), which was of a similar magnitude in both cases. Thus gangliosides do not inhibit mitogen-stimulated lymphocyte activation by preventing early increases in intracellular c a 2 + . In addition to an increase in intracellular c a 2 + , activation of protein kinase C is also a consequence of initial signal transduction by mitogens. Ganghosides and other sphingosine derivatives have been reported to act as inhibitors of protein kinase C in vitro (Kreutter et a/. 1987). However, lymphocytes treated with 50 nM TPA, a potent activator of protein kinase C, were still inhibited by gangliosides (Table 2). Thus gangliosides are still able to inhibit proliferation in cells where protein kinase C has already been maximally activated. 50

o

5

10

15

20

time, min

FIG. 7. Measurement of changes in intracellular c a Z +concentration with time in mouse splenic lymphocytes loaded with the fluorescentdye Quin-2, Following stimdation with 2 pg ConA/mL in the absence (A) or prexnc-(B) of 50 Hg gangtiasides/ml. The vertical bar shows the approximate change in the intracelIular free ca concentration as calculated by calibrating the systm with ionomycin and ~ n and " are from a typical experiment. +

at the onset of activation, by increasing intracellular c a 2 + and activating protein kinase C (Truneh et al. 1985). This method of cell stimulation is independent of accessory cells such as macrophages (Grier and Mastro 1986). Since gangliosides have been reported to suppress immune responses to soluble antigens by reversible inhibition of the function of adherent accessory cells (Ladisch et al. 1984), the effect of gangliosides and glycophorin on TPA-ionomycin stimulation was determined. The ganglioside inhibition response of lymphocytes stimulated in this fashion was identical to that seen for activation by the other mitogens (see Table 2). Thus it seems unlikely that the mechanism of ganglioside- and glycophorin-mediated inhibition of proliferation involves negative effects on accessory cells.

Effects on mitogen-induced increases in intracellular calcium An increase in the concentration of intracellular c a 2 + is involved in the transmission of many signals across cell membranes and is known to be important in transduction of the primary signal for lymphocyte activation. Gangliosides have been reposted to inhibit the mitogen-induced influx of %a2' into rnurine lymphocytes and also to greatly reduce ca2 uptake via the ionophore A23 187 (Krishnaraj et a/. 1983). We have pursued this question Further by monitoring the levels of intracellular c a b in murine splenocytes using the fluorescent probe Quin-2. The addition of mitogenic concentrations of ConA to murine !ymphocytespreloaded with Quin-2-AM resulted in a small, 3ut reproducible, transient increase in the intracellular ?a2+ concentration (Fig. 7A). Gangliosides themselves +

Discussion We have shown that activation of mouse T-lymphocytes by a variety of polyclonal mitogens is inhibited in a dosedependent fashion by membrane glycolipids such as gangliasides. The glycolipid concentration required for 50% inhibition depends on the structural complexity of the glycolipid, with a complex ganglioside such as GTlb being over 20-fold more inhibitory than cerebroside, a simple neutral glycolipid. The sialoglycoprotein glycophorin, from the human erythrocyte membrane, is also able to inhibit T-lymphocyte The various characteristics of - inhibition indicated by our results are so similar in both cases, that it seems likely the two types of glycosylated molecule act by a similar mechanism. This negative effect on immune responses is not due to toxicity and the same dose-response is seen for several different polyclonal mitogens, such as ConA, periodate, and TPA-ionomycin. The presence of sialic acid per se does not appear to be an absolute requirement for immunosuppression; asialo-GM1 and asialoglycophorin are as effective inhibitors as the sialylated parent molecules. Rather, the specific oligosaccharide structure and sequence appear to determine the inhibitory potency. Inhibition of T-cell activation is not a general property of sialoglycoproteins, since other species such as fetuin fail to cause inhibition even at very high concentrations. Other glycoproteins, such as the serum component al-acid glycoprotein (orosomucoid), have been reported to be capable of negatively affecting immune responses (Cheresh et al. 1984), and many uncharacterized "suppressive factors" found both in vitro and in vivo appear to be glycoproteins. Thus the modulation of T-lymphocyte activation by glycoproteins may be a generalized mechanism for regulation of the immune system. If the rapid turnover of the tumour cell plasma membrane results in the shedding of gangliosides in relatively large amounts, it seems entirely possible that glycoproteins may be shed in a similar fashion. Our results suggest that such shed glycoproteins could also contribute to suppression of immune responses. Glycophorin has an advantage over other glycoproteins in the exploration of the factors regulating immune suppression, in that it is


742


well characterized at the molecular level and a good deal is known about its behaviour, both in aqueous solution and in biological membranes. Gangliosides are well known for their unique ability to transfer from micelles in aqueous solution to both lipid bilayer vesicles (Felgner et al. 1983) and the plasma membrane of intact cells (Sharom and Ross 1986), where they appear to retain their function as receptors. Glycoproteins such as glycophorin are not known to possess this property. Our results suggest that inhibition of T-cell proliferation by gangliosides and glycophorin is unlikely to be mediated by membrane insertion. If lymphocytes are washed free of gangliosides or glycophorin after 24 h, they can still be activated normally by mitogens such as ConA, despite the fact that maximal membrane association of gangliosides would have been maintained for a considerable period of time. The process of mitogen-induced T-lymphocyte activation is known to occur in two distinct stages. The first stage involves the generation of the primary transmembrane signals, with binding of the initial mitogen to the cell surface resulting in signal transduction via the phosphoinositide pathway, i.e., an increase in intracellular c a 2 + concentration and the activation of protein kinase C. The point at which gangliosides and glycophorin exert their inhibitory effects seems unlikely to be at these first signal transduction steps, since the addition of either of these molecules as late as 18 h after initiation still results in substantial inhibition of proliferation. This assumption is borne out by the results of other experiments designed to focus on the possible mechanisms of inhibition. Experiments using a fluorescent indicator of intracellular c a 2 + , Quin-2, showed that gangliosides failed to abolish the transient pulse of Ca2+ which occurs within minutes of initiation by ConA. In addition, lymphocytes in which both branches of the early signal transduction pathway were activated directly, using a combination of phorbol ester and c a 2 + ionophore, were still unable to proliferate in the presence of gangliosides. Since this set of stimulation signals does not require accessory cells, it seems unlikely that ganglioside inhibition is mediated via effects on such cells. The result of these initial signals is the production of the lymphokine IL-2, which is released into the culture supernatant, and the expression of high-affinity receptors for this growth factor on the lymphocyte cell surface, usually within 12-24 h of initiation. We have determined that the production of IL-2 by ganglioside and glycophorin-inhibited lymphocytes is essentially normal. Even ganglioside concentrations much higher than those necessary to abolish proliferation have little effect on the levels of growth factor secreted into the culture medium. Thus neither gangliosides nor glycophorin prevent the production of IL-2 by stimulated lymphocytes. Ganghoside-inhibited lymphoblasts are also able to proliferate normally in response to exogenous IL-2, implying that their expression of IL-2r is unimpaired. However, if activated lymphoblasts are coincubated with both IL-2 and gangliosides, proliferation is completely abolished. Thus the point at which these glycoconjugates act appears to be at the second phase of the cascade of events leading to cellular proliferation. Marcus et al. (1987) have determined that gangliosides block the entry of lymphocytes into S phase, with the cells being arrested at late GI.

The second stage of the activation process involves binding of IL-2 to its specific membrane receptor, resulting in the transduction of a second, as yet unknown, set of signals that move the cell into S phase. Interference with binding of IL-2 to high-affinity IL-2r will also result in suppression of the overall proliferative response. Results presented here show that proliferation of the IL-2-dependent lymphocyte cell lines CTLL-2 and HT-2 in response to the lyrnphokine is also inhibited by gangliosides and glycophorin. Other IL-2-dependent lines are apparently inhibited by gangliosides in a similar fashion (Merritt et al. 1984; Parker et al. 1984). Results obtained in our laboratory also indicate that gangliosides prevent the binding of IL-2 to high-affinity IL-2r on both CTLL-2 and HT-2. There is some indirect evidence that gangliosides may bind to IL-2 (Parker et al. 1984; Robb 1986). We have recently demonstrated direct binding of IL-2 to ganglioside micelles in aqueous solution (Sharom et al. 1989; J.W.K. Chu and F.J. Sharom, in preparation). Aqueous aggregates of glycophorin are also able to bind IL-2 (A.L.H. Chiu and F.J. Sharom, unpublished data). If IL-2 binding is solely responsible for ganglioside inhibition, addition of large amounts of 1L-2 to inhibited lymphocyte cultures should reverse the inhibition by saturating all available IL-2 binding sites on the ganglioside micelles. This appears to be the case for the IL-2-dependent cell line HT-2, where substantial reversal of ganglioside inhibition of proliferation was seen at > 2500 U rIL-2/mL (J.W.K. Chu and F.J. Sharom, in preparation; Robb 1986). In this study, however, very high levels of IL-2 were unable to overcome ganglioside inhibition of ConA-stimulated mouse splenocytes. This suggests that additional mechanisms of ganglioside inhibition may be the dominant factor in the case of T-lymphocytes. This hypothesis is supported by the observation that the ganglioside concentration necessary to reduce proliferation of IL-2-dependent cell lines by 50% is substantially higher than that needed to inhibit activation of T-lymphocytes (in this study, >62.5 versus 15 pM). Interestingly, glycophorin inhibition is seen in the same concentration range for both T-lymphocytes and IL-2-dependent cell lines. This may indicate that gangliosides possess an additional effect(s) on T-cell proliferation that is not shared by glycophorin. Although the inhibitory effect of gangliosides on T-lymphocyte activation may be partially due to binding of IL-2, these glycolipids have additional effects that cannot be overcome by addition of exogenous IL-2. It is possible that gangliosides and glycophorin block signal transduction via the IL-2r. The molecular mechanism whereby the IL-2r transduces a growth signal has been the subject of much disagreement over the last few years. It is now apparent that neither the CAMP nor the phosphoinositide pathway are involved and it has been suggested that the p75 subunit of the IL-2r possesses tyrosine kinase activity (Tigges et al. 1989). In this regard, it is interesting to note that gangliosides can regulate the activity of many protein kinases both in vivo and in vitro, including the EGF and PDGF receptor tyrosine kinases (Bremer et al. 1984, 1986), protein kinase C (Kreutter et al. 1987), myelin basic protein kinase (Chan 1987), and a protein kinase in brain (Chan 1988). Further studies are needed to determine the additional mechanisms by which gangliosides and glycophorin inhibit the IL-2-induced growth of T-lymphocytes and the possible relevance of this

SHAROM ET

inhibition t o tumour-induced immunosuppression in vivo.

Acknowledgement This research was supported by a grant from the National Cancer Institute of Canada.


BREMER, E.G., HAKOMORI, S.-I., BOWEN-POPE, D.F., RAINES,E., and Ross, R. 1984. Ganglioside-mediated modulation of cell growth, growth factor binding, and receptor phosphorylation. J. Biol. Chem. 259: 6818-6825. BREMER,E.G., SCHLESSINGER, J., and HAKOMORI, S.-I. 1986. Ganglioside-mediated modulation of cell growth. J. Biol. Chem. 261: 2434-2440. CAVAILLON, J.-M., FITTING,C., HAUTTECOEUR, B., and HAEFFNER-CAVAILLON, N. 1987. Inhibition by gangliosides of the specific binding of lipopolysaccharide (LPS) to human monocytes prevents LPS-induced interleukin-1 production. Cell. Immunol. 106: 293-303. CHAN,K.-F. J. 1987. Ganglioside-modulated protein phosphorylation in myelin. J. Biol. Chem. 262: 2415-2422. -1988. Ganglioside-modulatedprotein phosphorylation: partial purification and characterization of a ganglioside-inhibited protein kinase in brain. J. Biol. Chem. 263: 568-574. CHERESH,D.A., HAYNES,D.H., and DISTASIO,J.A. 1984. Interaction of the acute phase reactant a,-acid glycoprotein (orosomucoid) with the lymphoid cell surface: a model for nonspecific immune suppression. Immunology, 51: 541-548. DAVIDSON,W.F., and PARISH,C.R. 1975. A procedure for removing red cells and dead cells from lymphoid cell suspensions. J. Immunol. Methods, 7: 291-300. DODGE,J.T., MITCHELL,C., and HANAHAN, D.J. 1963. The preparation and chemical characteristics of hemoglobin-free ghosts of human erythrocytes. Arch. Biochem. Biophys. 101: 119-130. E.V., and BERGELSON, L.D. 1987. GlycoDYATLOVITSKAYA, sphingolipids and antitumor immunity. Biochim. Biophys. Acta, 907: 125-143. D Y A T L O V ~E.V.. Y AK, L w ~ A T.E., , MATVEEVA, V.A., S I N ~ Y NE.V., A , AKHMEBZADE, A.S., L E W O V S K A Y A , A.F.. FOMINA-AGEEVA, E.V., and BERGELSON, L.D. 1986. Influence of gangliosides on the cytotoxic activity of natural killer cells of Syrian hamsters. Biokhimiya (Moscow), 50: 1298-1300. FELGNER. P.L., THOMPSON, T.E., BARENHQLZ,Y., and LICHTENBERG, D. 1983. Kinetics of transfer of gangtiosides from their micelles to dipalmitoyl phosphatidylcholine vesicles. Biochemistry, 22: 1670-1674. CRIER,C.E., and MASTRO,A.M. 1986. Phorbol ester circumvents the need for macrophages as well as for mitogenic lectins in the stimulation of lymphocytes with wheat germ agglutinin or the calcium ionophores A23187 and ionomycin. J. Leukocyte Biol. 40: 51 1-523. JACKSON, K.M., YATES,A. J., OROSZ,C.G., and WHITACRE, C.C. 1987. Gangliosides suppress the proliferation of autoreactive cells in experimental allergic encephalomyelitis: ganglioside effects on IL-2 activity. Cell. Immunol. 104: 169-181. KASPARAYAN, N.G., and YOGEESWARAN, G. 1987. Comparison of glycolipid shedding in non-transformed and virus-transformed non-producer mouse sarcoma cell lines. Fed. Proc. Fed. Am. Soc. Exp. Biol. 46: 1926. KLOPPEL, T.M., KEENAN,T.W., FREEMAN,M.J., and MOORE,D.J. 1977. Glycolipid-bound sialic acid in serum: increased levels in mice and humans bearing mammary carcinomas. Proc. Natl. Acad. Sci. U.S.A. 74: 3011-3013. K R E ~ RD.,, KIM, J.Y.H., GOLDENRING, J.R., RASMUSSEN, H., OKOMADU, C., DELORENZO, R.J., and Yu, R.K. 1987. Regulation of protein kinase C activity by gangliosides. J. Biol. Chem. 262: 1633-1637. CRISHNARAJ, R., LENGLE,E.E., and KEMP,R.G. 1982. Murine

AL.

743

leukemia. Proposed role for gangliosides in immune suppression. Eur. J. Cancer Clin. Oncol. 18: 89-98. KRISHNARAJ, R., LIN, J., and KEMP, R.G. 1983. Lectin and ionophore-stimulated c a 2 + influx in murine lymphocytes: inhibition by disialoganglioside. Cell. Immunol. 78: 152- 160. LADISCH,S., GILLARD, B., WONG,C., and ULSH,L. 1983. Shedding and immunoregulatory activity of YAC-1 lymphoma cell gangliosides. Cancer Res. 43: 3808-38 13. LADISCH,S., ULSH, L., GILLARD,B., and WONG,C. 1984. Modulation of the immune response by gangliosides. J. Clin. Invest. 74: 2074-2081. E., LADISCH.S., WU, Z.-L., FEIG, S., ULSH, L., SCHWARTZ, FLOUTSIS, G., WILEY,F., LENARSKY, C., and SEEGER,R. 1987. Shedding of GD, ganglioside by human neuroblastoma. Int. J. Cancer, 39: 73-76. LEE, P.M., and GRANT,C.W.M. 1980. Headgroup oligosaccharide dynamics of a transmembrane glycoprotein. Can. J. Biochem. 58: 1197-1205. LENGLE,E.E., KRISHNARAJ, R., and KEMP,R.G. 1979. Inhibition of the lectin-induced mitogenic response of thymocytes by glycolipids. Cancer Res. 39: 817-822. MARCHESI, V.T., and ANDREWS, E.P. 1971. Glycoproteins: isolation from cell membranes with lithium diiodosalicylate. Science (Washington, D.C.), 174: 1247-1248. MARCUS,D.M. 1984. A review of the immunogenic and immunomodulatory properties of glycosphingolipids. Mol. Immunol. 21: 1083-1091. MARCUS,D.M., DUSTIRA,A., DIEGO, I., OSOVITZ,S., and LEWIS, D.E. 1987. Studies of the mechanism by which gangliosides inhibit the proliferative response of murine splenocytes to concanavalin A. Cell. Immunol. 104: 71-78. Y., DURAND,G., RICHARD,A., FEGER,G.. and MASSAMIRI, AGNERAY, J. 1979. Determination of erythrocyte surface sialic acid residues by a new colorimetric method. Anal. Biochem. 97: 346-35 1. MERRITT,W.D., BAILEY,J.M., and PLUZNIK, D.H. 1984. Inhibition of interleukin-2-dependent cytotoxic T-lymphocyte growth by gangliosides. Cell. Immunol. 89: 1-10. MOHR, H., MONNER,D., and PLESSING,A. 1986. Calcium ionophore A23187 in the presence of phorbol ester PMA: a potent inducer of interleukin-2 and interferon-y synthesis by human blood cells. Immunobiology, 171: 195-204. OFFNER,H., and VANDENBARK, A.A. 1985. Gangliosides inhibit phenotypic and functional properties of an encephalitogenic T-helper lymphocyte line. Cell. Immunol. 95: 84-94. PARKER,J., CALDINI, G., KRISHNAMURTI, C., AHRENS,P.B., and ANKEL,H. 1984. Binding of interleukin-2 to gangliosides. FEBS Lett. 170: 391-395. ROBB,R.J. 1986. The suppressive effect of ganglioside upon IL 2-dependent proliferation as a function of inhibition of IL 2-receptor association. J. Immunol. 136: 97 1-976. ROMANS,A.Y., and SEGREST,J.P. 1978. Binding of lithium diiodosalicylate to glycophorin. Biochim. Biophys. Acta, 511: 297-301. SEGREST,J.P., WILKINSON, T.M., and SHENG,L. 1979. Isolation of glycophorin with deoxycholate. Biochim. Biophys. Acta, 554: 533-537. SELA, B.-A. 1980. Splenocytes incorporated with exogenous lymphocytes induce a mixed lymphocyte reaction in autologous lymphocytes. Cell. Immunol. 49: 196-201. SHAROM, F.J., and ROSS,T.E. 1986. Association of gangliosides with the lymphocyte plasma membrane studied using radiolabels and spin labels. Biochim. Biophys. Acta, 854: 287-297. SHAROM,F. J., CHU, J.W.K., and CHIU,A.L.H. 1989. Interaction of gangliosides with interleukin-2and the interleukin-2receptor. J. Cell Biol. 107: 446a. V. 1981. GlycoSMITH,D.F., MAGNANI,J.L., and GINSBURG, lipids of cell surfaces: Isolation of specific sugar sequences using

744


carbohydrate-binding proteins. Arch. Biochem. Biophys. 209:

bypassed by synergy between calcium ionophores and phorbol ester. Nature (London), 313: 318-320. SPIEGEL, S., and FISHMAN, P.H. 1987. Gangliosides as bimodal TSIEN, R.Y., POZZAN,T., and RINK, T.J. 1982. Calcium regulators of cell growth. Proc. Natl. Acad. Sci. U.S.A. 84: homeostasis in intact lymphocytes: cytoplasmic free calcium 141-145. monitored with a new intracellularly trapped fluorescent SPIEGEL,S., FISHMAN, P.H., and WEBER,R.J. 1985. Direct indicator. J. Cell Biol. 94: 325-334. evidence that endogenous GM, ganglioside can mediate T s u ~ uS., , NOMOTO, K., AISO,S., OGATA,T., and ZINNAKA, Y. thymocyte proliferation. Science (Washington, D.C.), 230: 1983. Regulation of lymphocyte responses in cancer patients. Int. 1285-1287. Arch. Allergy Appl. Immunol. 71: 88-92. TIGGES,M.A., CASEY,L.S., and KOSHLAND, M.E. 1989. MechYOUNG,W.W., BORGMAN,C.A., and WOLOCK,D.M. 1986. anism of interleukin-2signalling: mediation of different outcomes Modes of shedding of glycosphingolipids from mouse lymphoma by a single receptor and transduction pathway. Science cells. J. Biol. Chem. 261: 2279-2283. (Washington, D.C.), 243: 781-786. WHISLER, R.L., and YATES,A.J. 1980. Regulation of lymphocyte TRUNEH,A., ALBERT,F., GOLDSTEIN, P., and SCHMITTresponses by human gangliosides. J. Immunol. 125: 2106-2111. VERHULST, A.-M. 1985. Early steps of lymphocyte activation


52-56.

Gangliosides inhibit platelet-derived growth-factor-stimulated increases in intracellular calcium in Swiss 3T3 cells.

Gangliosides and neurological diseases.

[Gangliosides and nociception].

Distinct selectivity of gangliosides required for CD4⁺ T and CD8⁺ T cell activation.

Immune complexes inhibit IL-1 secretion and inflammasome activation.

Korean red ginseng extracts inhibit NLRP3 and AIM2 inflammasome activation.

Protein-lipid interactions glycophorin and dipalmitolyphosphatidylcholine.

Species specificity in mouse glycophorin.

Gangliosides and neuronal-astrocytic interactions.

Diffusion of glycophorin A in human erythrocytes.

Gangliosides of human melanoma.

Conformation of human erythrocyte glycophorin A and its constituent peptides.

Sestrins inhibit mTORC1 kinase activation through the GATOR complex.

Low density lipoproteins inhibit endotoxin activation of monocytes.

Enteropathogenic Escherichia coli Uses NleA to Inhibit NLRP3 Inflammasome Activation.

Drugs that disrupt microtubuli do not inhibit lymphocyte activation.

GM1 and GM2 gangliosides: recent developments.

Gangliosides of active and inactive neuroblastoma clones.

Binding and transport of gangliosides by prosaposin.

Surface behaviour of gangliosides and related glycosphingolipids.

Bovine gangliosides and acute motor polyneuropathy.

Thin-layer chromatography of gangliosides.

Coisolation of glycophorin A and polyphosphoinositides from human erythrocyte membranes.

Gangliosides of the bovine neurohypophysis.