Radioimmunoassay of glycosphingolipids: application for the detection of forssman glycolipid in tissue extracts and cell membranes

Characterization of the human Forssman synthetase gene. An evolving association between glycolipid synthesis and host-microbial interactions

Differences in glycolipid expression between species contribute to the tropism of many infectious pathogens for their hosts. For example, we demonstrate that cultured human and monkey urinary epithelial cells fail to bind a canine Escherichia coli uropathogenic isolate; however, transfection of these cells with the canine Forssman synthetase (FS) cDNA enables abundant adherence by the same pathogen, indicating that addition of a single sugar residue to a glycolipid receptor has marked effects on microbial attachment. Given the contribution of glycolipids to host-microbial interactions, we sought to determine why human tissues do not express Forssman glycolipid. Query of the GenBank(TM) data base yielded a human sequence with high identity to the canine FS cDNA. Reverse transcription polymerase chain reaction and Northern blotting demonstrated the presence of FS mRNA in all tissues examined. A human FS cDNA was characterized, revealing identities with the canine FS gene of 86 and 83% at the nucleotide and predicted amino acid sequences, respectively. In contrast to the canine FS cDNA, transfection of COS-1 cells with the human FS cDNA resulted in no detectable FS enzyme activity. These results suggest that variability in glycolipid synthesis between species is an important determinant of microbial tropism. Evolutionary pressure from pathogenic organisms may have contributed to diversity in glycolipid expression among species.

Biosynthesis in vitro of a globoside containing a 2-acetamido-2-deoxy-beta-D-galactopyranosyl group (1----3)-linked and Forssman glycolipid by two N-acetylgalactosaminyltransferases from chemically transformed guinea pig cells

Two N-acetylgalactosaminyltransferase activities (GalNAcT-2 and GalNAcT-3) have been characterized in chemically transformed, cultured guinea-pig cell lines (104C1 and 106B). Line 104C1 is a benz[a]pyrene-transformed tumorigenic variant, whereas line 106B is a 7,12-dimethylbenz[a]anthracene-transformed nontumorigenic variant obtained from fetal guinea-pig cells at 43 days of gestation. The GalNAcT-2 (UDP-GalNAc:GbOse3Cer beta-N-acetylgalactosaminyltransferase) isolated from both 104C1 and 106B cells catalyzed the transfer of Gal-NAc from UDP-GalNAc to the 3H-labeled terminal galactose group of Gb3 [( 6-3H]Gal alpha 1----4Gal beta 1----4Glc----Cer). The 3H-labeled globoside was purified and then subjected to exhaustive methylation. After acetolysis, the partially methylated sugars were separated by two-dimensional, thin-layer chromatography. 3H-Label was detected in two major areas, 2,4,6-tri-O-Me-Gal (40%) and 2,3,4,6-tetra-O-Me-Gal (46%). In a separate experiment, 80% of the GalNAc was released when labeled GbOse4Cer [( 3H]GalNAc----Gal alpha 1----4Gal beta 1----4Glc----Cer) was treated with purified clam beta-hexosaminidase. The present results establish the formation of a beta-D-GalpNAc-(1----3) linkage in the terminal region of the biosynthesized globoside. GalNAcT-3 activity (UDP-GalNAc:GbOse4Cer alpha-GalNAc-transferase), which catalyzes the transfer of GalNAc from UDP-[14C]- or -[3H]GalNAc to GbOse4Cer (GalNAc beta 1----3Gal alpha 1----4Gal beta 1----4Glc----Cer), was three times higher in 106B cells than in 104C1 cells. The isolated, purified radioactive product formed an immunoprecipitin line against rabbit anti-Forssman antibody.

A potential radioimmunoassay system for detection of Hanganutziu-Deicher type heterophile antigen(s) and antibodies in tissues and fluids

A relatively simple, specific and sensitive radioimmunoassay system has been developed for the detection of heterophile Hanganutziu-Deicher (H-D) antigen(s) and antibodies. The 125I-labeled H-D antigen-active molecule used for the assay is a bovine erythrocyte major glycoprotein previously found to have a strong H-D antigen potency. The antigen-antibody complex was precipitated with normal human serum as the carrier protein, followed by the addition of rabbit anti-human IgG F(ab')2 serum. With this method, different H-D antigen-active molecules were compared for heterophile H-D antigen potency with reasonable sensitivity detecting about 0.3 ng of cold glycoprotein. 8 different lung cancer tissues were assayed for H-D antigen. The sera from the 8 lung cancer patients were also screened by ELISA and RIA in an attempt to correlate expression of H-D antigen on tissues with elevation of H-D antibodies. The results showed that all patients' tissues expressed the antigen(s) but only 3 of them had abnormal levels of H-D antibodies. This could have been due to excess antigens in circulation or immune complexes.

A quantitative analysis of cell surface glycosphingolipid with a fluorescence activated cell sorter

The fluorescence activated cell sorter (FACS) was used with an indirect membrane immunofluorescence technique to detect antibody against the Forssman antigen, a glycosphingolipid. Sheep erythrocytes, which contain Forssman antigen as a major membrane glycosphingolipid, were used as the target antigen. Detection of the anti-Forssman antibody on the sheep erythrocytes was done with specific fluorescein-conjugated second antibody and analyzed on a FACS. Compared to other available methods, analysis with the FACS was simple, sensitive, reproducible and quantitative. More than 250 pg of antibody could be detected. In addition, as little as 1 ng of Forssman antigen could be estimated by a binding inhibition experiment.

Immunochemical determination of Forssman and blood group A-active glycolipids in human gastric mucosa by inhibition assay of liposome lysis

A simple liposome immunoassay, liposome immune-lysis inhibition (LILI) assay, is described for quantitative determination of individual glycolipid antigens. Liposomes containing fluorogenic marker, 4-methylumbelliferyl phosphate, were prepared from sphingomyelin, cholesterol, dicetylphosphate and standard glycolipid. Release of trapped markers from these liposomes by antibody and complement (liposome lysis) was inhibited by preincubating the antibody with test glycolipid incorporated into inhibitor liposomes. Based on the competitive inhibition, it was possible to quantitate each glycolipid antigen in less than picomolar amounts. The sensitivity and specificity of the assay were examined with purified glycolipid standards. LILI assay has been applied for the determination of Forssman glycolipid and blood group A-active glycolipid in human gastric mucosa and cancer tissues.