Genetic mechanism of blood group (ABO)-expression.

It has generally been believed that human blood group ABO is controlled by allelic ABO genes. However, this hypothesis has not yet been experimentally proven, and other possibilities such as the non-allelic gene model and the regulatory gene model for ABO locus have also been proposed. The genetic mechanisms of many unusual blood group expressions remain unanswered. Purification of human blood group N-acetylgalactosyltransferase (A-enzyme) which synthesizes A-substance, and blood group galactosyltransferase which is responsible for synthesis of B-substance, allows us to resolve these problems from an immuno-biochemical approach. It was found that rabbit antibody against-A-enzyme completely neutralized not only A-enzyme but also B-enzyme activity. Moreover, plasma from blood type O subjects contained an enzymatically inactive but immunologically cross-reactive material (CRM). Plasma from heterozygous AO and BO subjects also contained CRM, but plasma from homozygous AA and BB subjects did not contain CRM. These facts led us to conclude that the ABO genes are allelic in the strict sense, refuting other genetic models for ABO locus. Genotypes of phenotype A and B subjects can be unequivocally determined by examining the presence or absence of CRM in their plasma. Mechanism of the unusual blood group inheritance of Cis-AB (i.e., AB and/or O childbirth from AB X O parent) was elucidated by examining properties of the A and B enzymes, CRM in their plasma, and separation of active enzymes and CRM by affinity chromatography. It became clear that Cis-AB expressions in one family was due to unequal chromosomal crossing-over producing a single chromosome with the genes for A and B enzymes. In contrast, in the other two unrelated families, the Cis-AB expression was due to a structural mutation in A or B gene producing a single abnormal enzyme which was capable of transferring both GalNAc and Gal to H-substance. Mechanism of very weak B expression in a family with A1Bm character was studied. Plasma enzyme activity and kinetic characteristics of B-enzyme from the subjects was not different from that of normal. However, the A1Bm red cells contained a large amount of unoccupied H-sites which can be galactosylated in vitro and become B active. Examination of membrane components by isoelectric focussing revealed that blood group components of the A1Bm membranes were distinctively different from that of the usual membranes. Consequently, the weak B expression is not due to direct mutation of ABO locus, but due to a secondary consequence of genetic abnormality of a membrane component (or components) associated with blood group substances.