Two group-O children of a group-AB father: paternity, cis-AB or Bombay phenotype?
A couple from a town near Jaipur is referred for genetic counselling. The father's red cells are agglutinated by both anti-A and anti-B sera; his serum agglutinates A cells and B cells but not his own. The mother's cells react with anti-A only. Their two healthy children, aged 4 and 2, were each typed twice in different laboratories as group O. The elder child's serum agglutinates A cells, B cells and also pooled group O cells at room temperature. There is no history of transfusion or bone-marrow transplant, and the couple denies consanguinity. The parents want to know whether the children are really theirs and whether future pregnancies are at risk. The local laboratory holds anti-A, anti-B and anti-AB sera but no anti-H lectin and has no facility for saliva testing. The family can afford one visit to a referral centre in Jaipur.
1. Standard ABO logic says an AB father cannot have an O child. List the explanations that remain consistent with these results, and identify the single further test that would discriminate among them most sharply.
Read the worked answer
Why the obvious answer fails. The father is group AB, so his genotype is IAIB; he has no i allele to give. An ordinary group O child (ii) therefore needs an i allele from each parent, which is impossible unless one of the assumptions is wrong. The candidate explanations are:
- Bombay phenotype (hh). The children may carry IA or IB from the father but lack the H antigen that A and B transferases act on, so no A or B antigen is expressed on red cells. They type as O but are genotypically not O.
- cis-AB allele in the father. A single allele encoding an enzyme with both A and B activity, paired with an i allele, gives an AB phenotype but can transmit i, allowing an ii child.
- Weak subgroups or technical error (for example weak A or B expression, or a mistyped sample).
- Non-paternity or a sample mix-up - still possible and must be discussed with care.
The discriminating clue is already in the data. The elder child's serum agglutinates pooled group O cells. Ordinary group O serum contains anti-A and anti-B but not a strong anti-H, so it does not agglutinate O cells. Agglutination of O cells means an anti-H antibody, which is the serological signature of the Bombay phenotype. The confirming test at the referral centre is agglutination with Ulex europaeus anti-H lectin: Bombay (hh) cells are not agglutinated, whereas cis-AB or technical-error cells are. Saliva testing for H substance and family studies support the result.
2. If the Bombay phenotype is confirmed in both children, explain the recurrence risk for future pregnancies, the transfusion hazard, and whether the mother is at risk of haemolytic disease of the newborn.
Read the worked answer
Recurrence risk. Both parents express A or B antigens, so both must carry at least one H allele; because each child is hh, both parents must be Hh. Each future pregnancy therefore has a 1 in 4 chance of an hh child. Each child has a 1 in 2 chance of carrying h, so the parents' siblings should be screened if they may need transfusion.
Transfusion hazard. An hh person has anti-H in addition to anti-A and anti-B. Anti-H is a naturally occurring IgM antibody that haemolyses any red cell carrying H antigen, which includes ordinary group O cells - the very cells a laboratory would instinctively choose. Transfusion of group O blood could cause an acute haemolytic reaction. The child must receive blood only from another hh (Bombay) donor, arranged in advance through a rare-donor registry, or autologous blood for elective surgery. Cross-matching must include a 37 °C and antihuman-globulin phase with a Bombay donor panel, and the patient should carry a card stating the diagnosis.
Haemolytic disease of the newborn. Anti-H in Bombay individuals is usually IgM and does not cross the placenta efficiently, and ABO antibodies in general cause only mild disease, so severe haemolytic disease is not expected. The mother should still be monitored, and the diagnosis recorded, because an H-negative mother with any IgG anti-H would change management.