Red cells and serum from two B(h) variants (B+H-cells) have been investigated for
B and H blood group glycosyltransferases. The H enzyme could not be detected using either
type 1 or type 2 chain acceptors. The B enzyme was present in normal amount when 2'-fucosyllactose
was used as substrate, neither 6'-fucosyllactose nor 6'-fucosyllactosamine could act as
acceptors for the B enzyme.
Upon treatment of the B(h) red cells by the B-degrading enzyme from Trichomonas foetus the
B antigen was destroyed while H determinants were uncovered (B-H+ cells). The cells thus treated
could be further converted into A+H- red cells by the action of the A transferase from human
blood group A serum. Previous treatment of the B-H+cells by the H-degrading enzyme from
T. foetus, however, led to B-H- erythrocytes and prevented their conversion into A red blood
cells by the A enzyme. The results clearly demonstrate that, as found in normal B individuals,
the B antigen from B(h) cells is built up from the H precursor and provide additional evidence
that H is not a completely silent gene in B(h) individuals.