Monogamous bivalency of IgG antibodies. I. Deficiency of branched ABHI-active oligosaccharide chains on red cells of infants causes the weak antiglobulin reactions in hemolytic disease of the newborn due to ABO incompatibility.

This study shows that monogamous bivalency of human IgG anti-A and anti-B is necessary to maintain sensitization of red cells in the antiglobulin test. On red cells of infants and of adults of i phenotype, the known deficiency of branched ABHI-active oligosaccharide chains limits monogamous bivalency, and these accounts for the relatively weak direct antiglobulin reactions in hemolytic disease of the newborn due to ABO incompatbility. One corollary of this finding is that there is strong biologic selection against the transition on fetal red cells from the straight-chain i phenotype to branched-chain I phenotype, since such branching of cell-surface oligosaccharide chains would compromise ABO-incompatible pregnancies.

Evidence that blood group A antigen on lymphocytes is derived from the plasma.

Serum from group O volunteers, who had been injected with porcine A blood group substance, was used in lymphocytotoxicity tests. Positive reactions were obtained only with lymphocytes of group A secretors; the strongest reactors were Le(a--b--). The same group O sera reacted with group O lymphocytes which had been exposed to a glycosphingolipid fraction prepared from the plasma of A,Le(a--b--) secretors. These reactions were specifically inhibited by A substance. It is suggested that, unlike the A antigen on red cells, the A antigen detected in lymphocytotoxicity tests is entirely derived from the plasma.

Human blood-group A- and H-specified glycosyltransferase levels in the sera of newborn infants and their mothers.

The level of blood-group A1-specified alpha,3'-N-acetyl-D-galactosaminyl-transferase in the serum of recently-delivered women was found to be appreciably lower than the level of this enzyme in the serum of non-pregnant adults and of newborn infants; a similar but less striking decrease was observed in the levels of the A2-specified alpha,3'-N-acetyl-D-galactosaminyltransferase and the H-specified alpha,2'-L-fucosyltransferase. Although the red cells of newborn infants are known to have relatively few A and H antigen sites, the serum of neonates was found to have a level of A1- and A2-dependent N-acetylgalactosaminyltransferases and H-dependent fucosyltransferase as high as, if not higher than, the serum of non-pregnant adults. This finding is compatible with the fact that the haemopoietic tissue contributes only about 20% of the serum transferase level.

A further example of human blood group chimaerism.

Blood group chimaerism was detected in a healthy fertile woman, not known to be a twin. Her peripheral lymphocytes had a male karyotype (46/XY); fibroblasts cultured from her skin had a female karyotype (46/XX). The mechanism of chimaerism could not be established.

Conversion of incomplete antibodies to direct agglutinins by mild reduction: evidence for segmental flexibility within the Fc fragment of immunoglobulin G.

Reduction of interchain disulfide bonds converted some IgG incomplete antibodies to direct hemagglutinins. This conversion occurred whether antibody was free in solution or bound to the red-cell surface. Reduced antibody permitted to reoxidize in air no longer behaved as a direct agglutinin; reversion to an incomplete antibody did not occur when reoxidation was prevented by S-alkylation. These results suggest that mild reduction of the antibody imparts sufficient freedom to permit bridging between cells and are interpreted as evidence that the interheavy-chain disulfide bonds restrict segmental flexibility within the Fc fragment of IgG.

Red blood cell survival studies in patients with anti-Cha, anti-Yka, anti-Ge, and anti-Vel.

The survival of red blood cells, which were strongly incompatible in vitro, was measured in five patients whose serum contained an antibody to a high-frequency antigen. In the two patients with anti-Cha, and in the patient with anti-Yka, the cells survived normally. In the patient with anti-Ge, a small proportion of the cells was destroyed at an increased rate during the first 24 hours, but the remaining cells survived normally. In the patient with anti-Vel, the injected cells were rapidly destroyed.

A and B and A1Leb substances in glycosphingolipid fractions of human serum.

A and B and A1Leb substances were adsorbed onto red cells exposed to glycosphingolipid fractions prepared from the serum of group A and B and A1,Le(a minus b plus) donors. Group O cells exposed to fractions prepared from the serum of group A or B donors were agglutinated by an IgM cross-reacting antibody present in some group O sera. Cells exposed to fractions from A1,Le(a minus b plus) serum were agglutinated by anti-A1Leb. The amount of A substance in the fractions was related to the A subtype (A1 or A2) and to the Lewis and secretor phenotype of the donor. The uptake of blood-group substances from the lipid fractions was inhibited by the addition of whole serum to the fractions.

Qualitative differences in the N-acetyl-D-galactosaminyltransferases produced by human A1 and A2 genes.

This study describes the kinetic properties of N-acetyl-D-galactosaminyltransferase in serum from subjects with blood groups A(1) and A(2). When the A(1) and A(2) enzymes were compared, with lacto-N-fucopentaose I and 2'-fucosyllactose as acceptors, the enzymes differed in their cation requirements, pH optima, and K(m) values. The two acceptors competed for the same transferase. Mixing experiments showed that the lower activity of the A(2) enzyme could not be attributed to a modifier or inhibitor in serum. It was concluded that the A(1) and A(2) enzymes differ qualitatively.