Donald M. Marcus

Glycosphingolipids with Lewis Blood Group Activity: Uptake by Human Erythrocytes

The Lewis blood group antigens of the human erythrocyte are acquired from plasma and not synthesized in situ. Although assumed previously to be glycoproteins, tile Lewis antigens in plasma are glycosphingolipids Which are taken up by the the erythrocyte membrane from lipoproteins or from aqueous dispersions.

Antibodies to glycosphingolipids in patients with multiple sclerosis and SLE.

We used a liposome lysis assay to measure antibodies against a panel of glycolipids. Antibodies to one or more compounds were detected in 34 of 46 patients with multiple sclerosis, 19 of 31 patients with systemic lupus erythematosus (SLE), and in the majority of patients with cranial trauma or cerebrovascular accidents. Antibodies against ganglioside GM1 and asialo GM1 were found most commonly, and they were frequently present in the same sera. The specificity of the antibodies was tested in four sera that contained antibodies to both glycolipids. The anti-GM1 antibodies cross-reacted with asialo GM1, but the converse was not true. Among patients whose sera contained antibodies to glycolipids, anti-asialo GM1 alone was more common in patients with SLE (7 of 17) than in multiple sclerosis (2 of 34; p = 0.004). Anti-GM1 alone was found in 9 of 34 patients with multiple sclerosis and 1 of 17 patients with SLE, a difference that was not statistically significant (0.135). No correlation was observed between the presence of anti-glycolipid antibodies and symptoms related to the nervous system in patients with SLE. Because of our inability to detect these antibodies by a solid phase immunoassay (ELISA), a comparison was made of the titers obtained with three monoclonal anti-glycolipid antibodies in the liposome lysis assay and ELISA. The ELISA was less sensitive in all instances, requiring from four to 1000 times as much antibody as the liposome lysis assay to give a positive test. We conclude that antibodies to glycolipids occur frequently in patients with multiple sclerosis, SLE, major cranial trauma, and cerebrovascular accidents. Their role in the initiation or perpetuation of inflammatory disease of the central nervous system has yet to be determined.

Demyelination in vitro

Myelinated cultures of mouse spinal cord have been exposed to sera raised in rabbits against whole white matter (anti-WM), myelin basic protein (anti-MBP) and galactocerebroside (anti-GC), the major glycolipid of CNS myelin, to determine which factor in central nervous system (CNS) tissue in vitro is the target of serum demyelinating and myelin swelling antibodies. The sera were tested by radioimmunoassay for activity against MBP and against GC and were also specifically absorbed with MBP, GC and control antigens. Studies were also performed with and without active complement. The findings show that demyelination and myelin swelling in vitro are caused by antibodies against GC and not against MBP. Ultrastructurally, the effects of anti-WM and anti-GC sera with and without complement were indistinguishable. This study demonstrates that GC is a major target in antibody-mediated demyelination.

Human erythrocyte P and Pk blood group antigens: Identification as glycosphingolipids☆

Abstract The human erythrocyte P blood group system consists of three known antigens, P1, P and Pk. We have identified the P antigen as the glycosphingo-lipid globoside, βGalNAc(1→3)αGal(1→4)βGal(1→4)Glc-cer, and the Pk antigen as ceramide trihexoside, αGal(1→4)βGal(1→4)Glc-cer. These data suggest, in contrast to previous hypotheses, that the Pk antigen is a biosynthetic precursor of P, and that neither P nor Pk is a precursor of P1. These findings also provide an explanation for the apparent recessive inheritance of the Pk antigen, and for the nature of the biochemical abnormality in individuals of the rare Pk and p phenotypes.

A review of the immunogenic and immuno-modulatory properties of glycosphingolipids.

This review summarizes recent data concerning the immunogenicity and immunomodulatory properties of glycosphingolipids. Many murine monoclonal antibodies that react with glycosphingolipids have been described recently. Most of these antibodies have been elicited by immunization with tumor cells and they may also bind to glycoproteins that contain similar carbohydrate sequences. Immunization with a variety of tissues, murine teratocarcinomas, myeloid leukemia, and carcinomas of the human lung, colon and stomach, has elicited antibodies that react with the sugar sequence Gal beta 1-4[Fuc alpha 1-3]GlcNAc beta 1-3Gal----. The suppression of lymphocyte responses to mitogens and antigens by gangliosides in vitro has led to suggestions that these glycolipids possess immunodulatory properties in vivo. The in vitro studies were performed by incubating mononuclear cells with either dispersions of pure gangliosides or ganglioside-containing liposomes. In vivo gangliosides are found only in cell membranes or in lipoproteins, where they represent a small mole percent of total lipids, and there is little information about the transfer of gangliosides from lipoproteins to cells in vivo. A role for gangliosides as modulators of the immune response is an interesting possibility that is not supported by physiologically relevant data at present.

Immunochemical properties of glycolipids and phospholipids.

Publisher Summary Glycolipids offer unique experimental advantages for studies of the architecture and functional properties of cell membranes. This chapter discusses the immunochemical properties of glycolipids and phospholipids. The problems associated with lipid antigens are also considered. The immunological properties of complex lipid antigens and their growing importance in mammalian and microbial immunology are discussed. The structure, biosynthesis, distribution, immunological properties, and biological aspects of glycolipids and phospholipids are described. Three types of phospholipids reviewed are––cardiolipin, phosphatidyl inositol, and sphingomyelin. Glycolipids are the only components of cell membranes that are readily isolated and possess a single antigenic determinant. Antisera to these determinants can be used for research purposes: (1) to identify cells that are not readily distinguished by morphological differences, (2) to obtain data on the accessibility of specific antigenic determinants to antibodies during different physiological states, (3) to determine the cellular and subcellular distribution of glycolipids, (4) to prepare affinity columns for fractionation of cells and macromolecules, and (5) to study model membranes containing glycolipid antigens. Some of the other glycolipids discussed are glycosphingolipids, glycosyl glycerides, and lipoeichoic acids. The importance of glycolipids and their study by immunological procedures as specific moieties of the cell membrane is emphasized.

Do no harm: avoidance of herbal medicines during pregnancy.

Herbal medicines are regarded by the public and some health care providers as gentle and safe, but there is no scientific basis for that belief. The active ingredients of plant extracts are chemicals that are similar to those in purified medications, and they have the same potential to cause serious adverse effects. This commentary summarizes recent data on the poor quality control and toxicity of herbal remedies and on the pharmacologic activities of ginger, which is used for treatment of morning sickness. There are no rigorous scientific studies of the safety of dietary supplements during pregnancy, and the Teratology Society has stated that it should not be assumed that they are safe for the embryo or fetus. Obstetricians should advise women not to expose their fetuses to the risks of herbal medicines.

Isolation of ferritin from human mammary and pancreatic carcinomas by means of antibody immunoadsorbents.

Abstract In the course of a study of tumor antigens we prepared an absorbed antiserum to a breast tumor that reacted strongly with breast tumor but not with normal breast. The antigen was purified by adsorption to an antibody immunoadsorbent prepared from this antiserum, elution with 2 m potassium thiocyanate at neutral pH, and passage through an immunoadsorbent containing antibodies to human serum. The purified antigen was identified as ferritin by electrophoretic, chemical, and immunological criteria. Isoelectric focusing in acrylamide gels revealed that tumor ferritin contained six bands seen in normal liver ferritin plus a variable number of acidic components not detected in normal liver ferritin. The acidic components were concentrated by chromatography on a DEAE-cellulose column. Similar acidic components described previously in ferritins isolated from cultured human tumor cells, hepatomas, and fetal liver have been designated as “carcino-fetal” ferritins.

Glycosphingolipids of human plasma

A number of glycosphingolipids, including 10 gangliosides, not previously identified in human plasma have been characterized. The plasma contains 2 micrograms of lipid-bound sialic acid/ml plasma and 54% of the gangliosides are monosialo, 30% disialo, 10% trisialo, and 6% tetrasialo. Individual glycosphingolipids were purified by high-performance liquid chromatography and thin-layer chromatography, and were characterized on the basis of their chromatographic mobility, carbohydrate composition, hydrolysis by glycosidases, methylation analysis, and immunostaining with anti-glycosphingolipid antibodies. The monosialogangliosides were identified as GM3, GM2, sialosyl(2-3)- and sialosyl(2-6)lactoneotetraosylceramides, sialosyllacto-N-nor-hexaosylceramide, and sialosyllacto-N-isooctaosylceramide. The major gangliosides in the polysialo fractions contained a ganglio-N-tetraose backbone and were identified as GD3, GD1a, GD1b, and GQ1b. The most abundant neutral glycosphingolipids were glucosyl, lactosyl, globotriaosyl, globotetraosyl and lactoneotetraosylceramides. The other neutral glycosphingolipids, tentatively identified by immunostaining with monoclonal antibodies, contained H1, Lea, Leb, and lacto-N-fucopentose III (X hapten) structures.

Measurement and significance of antibodies against GM1 ganglioside Report of a workshop, 18 April 1989, Chicago, IL, U.S.A.

Twelve laboratories from the United States, Canada, France, Italy and Switzerland participated in a workshop to compare assays used to measure anti-GM1 antibodies, and to discuss the clinical significance of these antibodies. A panel of test samples containing varying amounts of anti-GM1 antibody was prepared by mixing varied proportions of normal serum with a serum containing a monoclonal IgM antibody that bound GM1 ganglioside. Enzyme-linked immunosorbent assay (ELISA) data were supplied by eight laboratories and ten laboratories classified the sera as negative, weakly or strongly positive. Most laboratories correctly identified the two samples that contained the highest quantities of antibody, but there was considerable disagreement on the classification of the three samples with moderate or small amounts of antibody. The sensitivity of the assays varied considerably. The more sensitive assays did not use detergent in the washing buffers, and incubated the human serum with the antigen at 4 degrees C overnight. Several investigators have identified a subset of patients with lower motor neuron disease or multifocal neuropathy who have high titers of anti-GM1 antibodies. Many patients with neurological and non-neurological diseases have low to moderate levels of anti-GM1 antibodies, and the significance of these antibodies is unclear. There was general agreement that standardization of the ELISA assays is urgently required, and that distribution of a reference high-titered antiserum would facilitate this process.