Synthetic Glycolipid Research Articles

Carbohydrate-specific cell adhesion is mediated by immobilized glycolipids.

We describe a technique for examining the ability of one important class of cell surface complex carbohydrates, glycosphingolipids, to mediate carbohydrate-specific cell recognition and adhesion. Analogs of natural glycosphingolipids were synthesized, consisting of 1-glycosyl derivatives of 3-deoxyceramide (N-palmitoyl-2-aminostearol) radiolabeled in the fatty acid portion. Methods were developed to efficiently adsorb both these synthetic glycolipids and natural glycosphingolipids (including gangliosides) from aqueous ethanol solution onto plastic wells. The glycolipids remained firmly attached to the surface in aqueous solutions, but could be recovered using detergents or organic solvents. The ability of the adsorbed glycolipids to elicit specific adhesion of intact hepatocytes was tested using specific adhesion of intact hepatocytes was tested using a cell adhesion assay based on that of McClay, D. R., Wessel, G. M., and Marchase, R. B. (1981) Proc. Natl. Acad. Sci. U. S. A. 78, 4975-4979. When otherwise nonadhesive plastic surfaces were adsorbed with N-acetylglucosaminyl 3-deoxyceramide, they supported adhesion of 80-95% of the chicken hepatocytes added to the well. No adhesion above background levels (10-25%) was observed to surfaces adsorbed with other synthetic glycolipids including glucosyl, galactosyl, mannosyl, or lactosyl 3-deoxyceramide, 3-deoxyceramide, or to the naturally occurring glycosphingolipids, lactosyl ceramide or ganglioside GM1. Chicken hepatocyte adhesion to surfaces adsorbed with N-acetylglucosaminyl 3-deoxyceramide was inhibited by soluble N-acetylglucosamine (IC50 = 3 m M), but not by other soluble sugars. Rat hepatocytes adhered preferentially to surfaces adsorbed with lactosyl 3-deoxyceramide, but not to surfaces adsorbed with the N-acetylglucosaminyl derivative. These studies demonstrate the ability of adsorbed glucolipids to mediate carbohydrate- and cell-specific adhesion from intact cells. Using these techniques, the ability of naturally occurring complex glycosphingolipids to elicit specific cellular responses from a variety of cell types can be examined.

Threshold effects on the lectin-mediated aggregation of liposomes: Influence of the diameter of the liposomes

It had previously been found that small unilamellar liposomes of ca. 0.03 μm diameter which bear synthetic cholesterol-containing glycolipids may be aggregated by an appropriate lectin [8]. Where studied, threshold effects have been observed in that the amount of glycolipid incorporated in the liposomes must exceed a certain minimum concentration in order for aggregation to occur [3, 8, 9, 13, 14]. Threshold effects of this type may be important in mediating cell-cell and virus-cell interactions. However, before studies with small unilamellar liposomes are useful as a model for these recognition and binding phenomena, it must be shown that the observed threshold effects are not associated with the very small radius of curvature of these liposomes. This article reports that larger liposomes of average diameter 0.26 and 0.45 μm which contain the synthetic glycolipidl also show threshold effects when aggregated with the galactose binding lectin ricin agglutinin. Under conditions where more than 1% (mole) glycolipid is required to support the aggregation of the smallest liposomes, those of intermediate size require only 0.18% (mole) while the largest liposomes examined require between 0.095 and 0.15% (mole) depending on the method of preparation.

Effects of structural variations in synthetic glycolipids upon mitogenicity for spleen lymphocytes, adjuvancy for humoral immune response and on anti-tumour potential.

Synthetic glycolipids prepared by esterification of various sugars and sorbitol, and containing various numbers of saturated or unsaturated fatty acid residues as well as bacterial lipid A and lipopolysaccharide, were tested for mitogenicity of splenic cells of Fischer rats and Swiss mice and for the augmentation of humoral immune response against sheep red blood cells in these species. Subsequently a few of the humoral immune-response-enhancing glycolipids were compared with non-enhancers in their anti-tumour activity against 13762 rat mammary carcinoma in inbred Fischer 344 rats and Ehrlich tumour in Swiss mice. They were given systemically after tumour inoculation and intratumourally in squalene and Tween emulsion after intradermal MAC tumour development. It was observed that certain structural characteristics in glycolipids with respect to the type of sugar, the type and number of fatty-acid residues were needed for their adjuvant action of the humoral arm of the immune response. Although humoral immune-response enhancers were somewhat superior to non-enhancers in their anti-tumour activity, the correlation coefficient demonstrated a lack of significant concordance. It is concluded that glycolipids selected for their ability to augment humoral immune responses against standard antigens need not be suspect as tumour-enhancers on the grounds that they would elicit blocking antibodies in vivo against tumour-associated antigens.

Transference of the high molecular weight carbohydrate sequences of fetuin to the surface of carrot embryo protoplasts

A method is described for the removal of the carbohydrate sequences of glycoproteins, and their covalent attachment to hydrocarbon chains. These synthetic membrane components may then be incorporated into liposome and cell membranes. Pronase-liberated glycopeptides derived from fetuin were linked by a reduced Schiff's base linkage to tetradecyl aldehyde. The resulting glycolipid was incorporated by external addition, into phosphatidylcholine liposomes. Glycolipid transfer to these liposomes rendered them suseptible to agglutination by wheat germ lectin, which binds N-acetylneuraminic acid, the terminal carbohydrate of the high molecular weight fetuin sugar sequence. Sequential removal of the terminal sugars, and subsequent agglutination behaviour towards various lectins, suggests that the carbohydrate sequence had been transfered intact. The glycolipid was incorporated into plant protoplast membranes by incubation with glycolipid-containing liposomes for 2 h at 37°C. These synthetic glycolipids may find a use in the study of carbohydrate-based recognition systems in animal and plant membranes. In addition they may prove useful in the development of cell and membrane tagging and handling techniques, by the insertion of sugar groups not normally present in these membranes.

Agglutination of glycolipid—phospholipid vesicles by concanavalin A: Evidence for steric modulation of lectin binding by phospholipid head groups

Lectin binding to a glycolipid inserted in phospholipid vesicles seems to require that the carbohydrate group extends some distance from the hydrocarbonwater interface. Thus, vesicles containing a diglycosylceramide, but not a monoglycosylceramide, could be agglutinated by Ricinis communis agglutinin [ 11. The same lectin could agglutinate glycolipid-phosphatidylcholine vesicles if the glycolipid contained a 13or 22-member, but not a 4-member, spacer arm between the hydrophobic part and the carbohydrate group [2]. The interpretation favoured in [ 1,2] was that the lectin can only bind carbohydrate that extends beyond phospholipid head groups on the vesicle surface. If so, lectin binding to vesicles in which the lectin-binding group is properly located with respect to the hydrocarbon region might be sterically regulated by the (hydrated size) of surrounding phospholipid head groups. Fig.1. Structure of the synthetic glycolipid octadecylmaltobionamide.

Monolayers from synthetic glycolipids

Synthetic glycolipids have been prepared by coupling of (a) aliphatic amines with lactone derivatives, or (b) fatty acids with amino derivatives of mono- and disaccharides via amide linkage. The aliphatic chain was varied with respect to the chain length and to its chemical composition (fluorination, functional end group). The influence of the glycolipid structure on the monolayer properties was investigated. Stable films are obtained with most of the products mainly due to strong interactions by hydrogen bonds in the subphase. Polymeric films may be produced by polycondensation in the subphase using particular crosslinking agents for the carbohydrate head group.

Modified in vivo behavior of liposomes containing synthetic glycolipids

Liposomes with synthetic saccharide determinants were prepared from synthetic cholesterol conjugates of D-mannose and 6-amino-6-deoxy- D-mannose and labeled with [ 51Cr]chromate. The kinetics and tissue distribution of label in mice were determined after footpad and subcutaneous injection. Liposomes bearing either of these saccharide determinants greatly increased retention of label at the injection sites compared to control liposomes, which contain no glycolipid, and to free [ 51Cr]chromate. Draining lymph nodes contained small fractions of the injected radioactivity but in some cases this retention was saccharide-dependent and highly concentrated. These results show that incorporation of synthetic glycolipids can substantially alter the in vivo lifetime and distribution of liposomes outside the bloodstream. Such surface-modified liposomes may be useful for sustained release or selective delivery of therapeutic or diagnostic agents.

Phagocytosis of carbohydrate-modified phospholipid vesicles by macrophage.

Modification of the surface of distearoyl phosphatidylcholine vesicles with synthetic glycolipids dramatically affects the rate of uptake of these vesicles by mouse peritoneal macrophage. The high rate of uptake of 6-aminomannose-modified vesicles is effectively inhibited by cytochalasin B and chloroquine but not by colchicine, indicating that the mechanisms of vesicle uptake is phagocytosis. Other modified vesicles appear to have some effect on the rate of uptake of 6-aminomannose-modified vesicles suggesting that the various vesicle types compete for the same initial binding sites. Analysis of 6-aminomannose-modified vesicles by gamma-ray perturbed angular correlation spectroscopy shows that the rotational correlation time of the encapsulated 111In3+ does not change when the vesicles associate with macrophage. This result is consistent with transmission electron microscopy, which indicates that the aminomannose-modified vesicles remain intact after phagocytosis as aggregates of fused and intact vesicles surrounded by a single bilayer membrane structure.

Synthetic glycolipids: Interaction with galactose-binding lectin and hepatic cells

Lactosyl- and melibiosyl-phosphatidylethanolamine prepared by reductive animation with sodium cyanoborohydride were incorporated into small unilamellar liposomes. Lactosyland melibiosyl-phosphatidylethanolamine liposomes are aggregated by Ricinus communis agglutinin whereas Banderiaea simplicifolia isolectin I aggregates only melibiosyl-phosphatidylethanolamine liposomes. The association constant ( K a ) values of interactions of R. communis agglutinin and glycolipids were 5 × 10 5 and 1.2 × 10 5 m −1 for lactosyl- and melibiosyl-phosphatidylethanolamine, respectively, whereas the K a for the interaction of B. simplicifolia isolectin I for melibiosyl-phosphatidylethanolamine was found to be 6 × 10 5 m −1. The rates of aggregation of these liposomes are strikingly influenced by the amount of glycolipid incorporated into them. In vivo studies indicate that lactosyl-phosphatidyl-ethanolamine-containing liposomes are rapidly taken up by hepatic cells due to binding of their β- d-galactopyranosyl residues by the hepatic galactose-binding lectin.

Lectin-mediated aggregation of liposomes containing glycolipids with variable hydrophilic spacer arms.

Synthetic glycolipids containing a cholesterol anchor group attached via a spacer group to a sugar moiety can be incorporated into small unilamellar liposomes, rendering them susceptible to agglutination by the appropriate multivalent lectin [Rando, R. R., Orr, G. A., & Bangerter, F. W. (1979) J. Biol. Chem. 254, 8318-8323; Rando, R. R., & Bangerter, F. W. (1979) J. Supramol. Struct. 11, 295-309]. We report here on the role of spacer arm length in rendering these liposomes susceptible to agglutination. In order to eliminate the ambiguities inherent in using hydrophobic or charged spacer groups, we have synthesized a hydrophilic, ethylene glycol based amino acid (8-amino-3,6-dioxaoctanoic acid) for these studies. The Ricinus communis agglutinin (ricin) mediated agglutination of these beta-galactoside-containing glycolipids was studied. A spacer arm length of four atoms will not support agglutination under any conditions. A seven-atom spacer will support agglutination, but only at high phospholipid concentrations (0.24 mumol/mL) with a pseudo-first-order rate of agglutination of 0.0079 min-1. With 13 and 22 atom spacer groups, the pseudo-first-order rate constants were 0.4 min-1 and 1.3 min-1, respectively, at a phospholipid concentration of 0.06 mumol/mL. Under conditions where the liposomes containing the glycolipid with the 13-atom hydrophilic spacer arm were completely agglutinated by ricin, 9.3% of the total available sugar moieties of the liposome were bound to ricin. This means that, on the average, 38 interliposomal bonds were formed in an aggregate.

Functional incorporation of synthetic glycolipids into cells.

Synthetic glycolipids containing an alpha-mannoside group linked by a hydrophilic spacer arm to cholesterol were incorporated into bovine erythrocytes by exchange from glycolipid-containing liposomes. When the distance between the sugar and the cholesterol moieties was approximately 26 A, functional incorporation of these glycolipids could be easily detected, as revealed by the concanavalin A-mediated agglutination of these cells. Bovine erythrocytes are not themselves susceptible to concanavalin A-mediated agglutination. The minimal concentration of concanavalin A required for agglutination of modified erythrocytes, containing 9.15 x 10(6) glycolipid molecules per cell, was 4 microgram/ml. Under these conditions, only approximately 4% of the membrane-bound cholesterol had been exchanged for the synthetic glycolipid. The observed aggregation was reversible in the presence of alpha-methyl mannoside and did not occur when beta-galactosyl-containing glycolipids were used in place of their alpha-mannoside isomers. These studies demonstrate a technique of sugar incorporation into cell membranes which should be of great advantage in studies on the roles of cell surface sugars in biological recognition. Furthermore, they demonstrate that the sugars need only be a short distance (26 A) from the membrane in order to functionally bind concanavalin A.

Regression of a murine fibrosarcoma after intralesional injection of a synthetic C39 glycolipid related to cord factor

Intratumoral injection of ultrasonically prepared emulsions of the synthetic glycolipid methly 6-O-(2-tetradecyl-3-hydroxyoctadecanoyl)-alpha-D-glucopyranoside (designated C39) induced complete regression of transplants of a syngeneic murine fibrosarcoma in most of the treated animals as did 6,6'-di-O(2-tetradecyl-3-hydroxyoctadecanoyl)-alpha,alpha,-trehalose (designated C76) in a previous study. The C76 compound, about twice the molecular weight of C39, was more effective therapeutically than the smaller molecule. Ultrasonically prepared emulsions of C39 and C76 were not toxic when given intravenously. Intravenously administered emulsions of C39 prepared by mechanical grinding were more toxic, but less granulomagenic, than those containing C76. Squalane and squalene, but not peanut oil, were effective substitutes for mineral oil as carriers of C39 in the treatment of the tumor.

Combined immunostimulation in the prevention of tumor take in mice using endotoxins, their derivatives, and other immune adjuvants

In this study a variety of immunostimulatory agents was tested alone or in combination with other agents to measure their ability to enhance nonspecific resistance to challenge with the TA3-Ha transplantable murine ascites tumor. Lipopolysaccharides (LPS) were effective alone in protecting mice from tumor. Their lipid-free, nontoxic, polysaccharide-rich hydrolytic products (PS) showed a much reduced, but still significant effect in anti-TA3-Ha resistance induction. Other agents that stimulated tumor resistance included nontoxic ‘native hapten’ and trehalose-6,6′-dimycolate (P3) of Ribi, synthetic glycolipids, cord factor, and killed BCG preparations. Simultaneous pretreatment with a combination of any of these agents and either LPS or PS resulted in a significantly higher level of tumor resistance. Administration of LPS, PS, or native hapten to mice that had been previously infected with viable BCG resulted in the strongest antitumor effect. These studies demonstrate that nonspecific resistance can be enhanced in an additive or synergistic manner by using combinations of agents, which presumably stimulate different cell populations of the host's immune system.

Synthetic glycolipids and the lectin-mediated aggregation of liposomes.

Synthetic mannose-containing glycolipids utilizing the cholesterol nucleus as a lipid anchor, and either the 6-aminohexyl- or the 6-(6-aminohexanamido)hexyl-1-thio-alpha-D-mannopyranosides as the carbohydrate ligands, have been synthesized and incorporated into small unilamellar liposomes. Incorporation of these cholesterol-mannoside derivatives at concentrations up to 14 mol% apparently does not affect the physical characteristics of the liposomes. Addition of concanavalin A to a suspension of liposomes containing the long chain cholesterol-mannose derivative causes an increase in light-scattering at 360 nm. As the increase in absorbance is completely reversed by the addition of alpha-methylmannoside, aggregation rather than fusion of the liposomes appears to be occurring. Liposomes containing 14 mol % of the short chain (6-aminohexyl-) derivative are aggregated by concanavalin A indicating that the lectin can approach to within 10 A of the lipid bilayer. Preliminary results suggest that the aggregation of vesicles containing either the long or short chain derivatives is highly dependent on the density of the sugar in the membrane.

Threshold effects on the lectin-mediated aggregation of synthetic glycolipid-containing liposomes.

Cholesterol analogs containing sugar residues linked by spacer groups to the cholesterol O can be incorporated into egg yolk lecithin small unilamellar liposomes. The synthetic glycolipid analogs distribute evenly on both sides of the bilayer. These liposomes are aggregated by the appropriate lectin. For example, when the sugar residue is a beta-galactoside the liposomes are aggregated by ricin and when it is an alpha-mannoside they are aggregated by Con A. The lectin-mediated aggregation of these liposomes is reversed by the addition of the appropriate sugar. The rates but not the extents of aggregation of these liposomes are highly sensitive to the amount of glycolipid incorporated. Below approximately 5% glycolipid incorporation the rate of the lectin-mediated aggregation of these liposomes is exceedingly slow, whereas above this level rapid aggregation proceeds. At all concentrations studied the synthetic glycolipids are incorporated in a unimodal fashion so that the observed threshold effects cannot be based on possible differences in the manner in which the glycolipids are incorporated at different concentrations. This conclusion is based on 1) studies with galactose oxidase that show that the percentage of galactose oxidation in a liposome prepared from a galactosyl-containing glycolipid is independent of glycolipid concentration, and 2) studies on the aggregation of liposomes containing mixed glycolipids in which the glycolipids are shown to behave independently. The importance of a critical density of membrane-bound receptors in order for aggregation to occur is discussed.

Relation between structure of bacterial lipopolysaccharide and the stimulation of B lymphocytes into colony formation

AbstractLipopolysaccharide (LPS) was analyzed in order to determine which component, lipid A or polysaccharide (PS), is able to stimulate B lymphocytes from ICR lymph nodes and spleen cells from nude (nu/nu) mice into forming colonies in soft agar culture. Lipid A, obtained by acid hydrolysis of LPS and solubilized by complex‐formation with bovine serum albumin, was found to be the active moiety of LPS capable of stimulating colony growth of lymphoid cells in soft agar culture. The PS portion exhibited no significant activity at the concentrations used. Glycolipids from mutant strains of S. minnesota which contain the intact lipid moiety but are deficient in PS content, were as potent as S. abortus equi LPS in stimulating B cells into colony growth. Alkaline hydrolysis of LPS which cleaves ester‐linked fatty acids, substantially decreased the number of lymphocyte colonies formed. This indicates that the intact lipid moiety is required for stimulating lymphocytes into colony formation. The synthetic glycolipid, N‐palmitoyl‐D‐glucosamine (NPG), whose structure is similar to some components of lipid A, was also able to induce B lymphocyte colony development. In summary, our data point to lipid A as the active moiety of the endotoxin which induces B lymphocytes to grow and develop into colonies in the 2‐layer soft agar culture system.

Specific interaction of concanavalin a with glycolipid monolayers

The effect of 131I-labelled concanavalin A on the surface pressure and surface radioactivity of monolayers formed from phospholipids and from natural and synthetic glycolipids has been studied. The lectin binds to and penetrates dipalmitoyl phosphatidylcholine monolayers at a surface pressure of 15 dynes/cm and this interaction is inhibited by the presence of α-methyl mannose int he subphase. At surface pressures of 25 dynes/cm or higher, concanavalin A will interact with monoglucosyl diglyceride or diglucosyl diglyceride from Acholeplasma laidlawii and with synthetic glycolipids containing 2 or 3 α1 → 4- linked D-glucose residues in the headgroup, but not with phosphatidylcholine, phosphatidylethanolamine, phosphatidylglycerol, or with the ganglioside II 3NeuAc-GgOse 4-Cer. The binding to the glycolipid sugar group and penetration of the hydrocarbon region seem to occur simultaneously, as the time courses for the development of surface pressure and surface radioactivity coincide.

Synthetic Glycolipid Adjuvants

In the course of the organic synthesis of model compounds similar in some features to the lipid moiety of endotoxic lipopolysaccharide (LPS), Nacylated-D-glucosamine derivatives were prepared. One of these, N-palmitoyl-D-glucosamine, has been previously found to be mitogenic for athymic nude mouse B cells. This and other N-acylated homologs were tested for adjuvant activity in the immune response to human gamma-globulin (HGG) and sheep red blood cells (SRBC) in mice. Comparable or superior enhancement of the immune response was obtained for these glycolipids when compared to LPS in assays measuring anti-SRBC or HGG hemagglutinin titers. In the determination of hemolytic plaque formation, considerable adjuvant effect was shown by the lauroyl derivative, and less but still significant enhancement was achieved by the N-palmitoyl-D-glucosamine. In the rosette formation assay, in addition to the above two glycolipids, N-oleyl-D-glucosamine showed good adjuvant effect. In the latter two assays, the LPS was a superior adjuvant as compared to the synthetic glycolipids. The radiation protective effect of some of the better synthetic adjuvants was also investigated in mice. It was found that although LPS was more effective in this assay, the N-myristoyl-D-glucosamine and N-decanoyl-D-glucosamine compounds gave a definite protection, since up to 40% of the lethally irradiated (700 R) mice survived.

A synthetic glycolipid with B-cell mitogenic activity.

A synthetic glycolipid, N-palmitoyl-D-glucosamine, (NPG) was found to be a potent B-cell mitogen, that stimulated spleen cells from nu/nu as well as from normal mice. The proper dispersion of the water insoluble preparation is critical for the elicitation of this mitogenic effect. Limulus lysate clotting assay indicated that the NPG preparation either contains only 0.001% endotoxin contamination, or that NPG itself is 10(-5) times less active in this assay than purified endotoxic LPS. Since such low levels of endotoxin concentration are not mitogenic, it is concluded that synthetic N-palmitoyl-D-glucosamine, when properly dispersed, is itself a B-cell mitogen.