Abstract

In the ABO blood group system any allele which does not give rise to A or B antigen expression is an O allele. The most common O alleles, O1 (O01) and O1v (O02) encode enzymatically inactive truncated proteins due to a single nucleotide polymorphism (SNP) compared to the consensus A1 allele, 261delG. Other O alleles containing the consensus nucleotide at residue 261 have been identified. The most common of these non-deletional O alleles is O2 (O03). It contains other SNPs which render the resulting protein non-functional. The most significant SNP is 802G>A (G268R); the arginine residue blocks the donor sugar’s access to the active site preventing its transfer. Recently O2 alleles were proposed to be the most common cause of ABO discrepancies (anti-A lacking or weakened in plasma despite apparent group O phenotype) in automated blood grouping (

Transfusion
2005
;
45
:
1331
). Other clinical or biochemical reports have suggested that O2 may produce small amounts of A antigen on the RBC surface (
Transfusion
2005
;
45
:
70
, 2005;45:359,
J Biol Chem
2005
;
280
:
525
). To establish the effect of the O2 allele on automated ABO typing we extracted DNA from 562 randomly selected group O RBC units that were available for transfusion. Automated forward and reverse ABO grouping was performed on an Olympus PK7231 using A1 and B cells for reverse typing (Pittsburgh) and the IBG Multisampler Plato 3000 SI system using A1, A2 and B cells (Lund). A PCR-RFLP assay (
Vox Sang
1995
;
69
:
242
) involving amplification of exon 6 of the ABO gene followed by cleavage with KpnI to detect the presence of the 261delG SNP was performed and 34/562 (6%) of these units were heterozygous for a deletional and a non-deletional O allele. All 34 of these donors demonstrated an identical digestion pattern with BstEII which cleaves exon 6 in the presence of the consensus sequence at residue 261, confirming their heterozygosity for a non-deletional O allele. Allele-specific PCR to detect the O2-specific 802G>A polymorphism was performed on 32 of these donors for which additional DNA was available; the O2 allele was uniformly detected along with either an O1 or O1v allele. No homozygous O2O2 donors were identified. A sensitive flow cytometric analysis of the RBCs from 7 donors with the O2 allele was performed using monoclonal anti-A (ES-15) and a PE-labelled rat-anti-mouse kappa Ig secondary antibody. None of these RBCs demonstrated fluorescence levels in significant excess of the control O (O1O1) cells whilst A antigen on group B cells was demonstrable. Manual immediate spin and gel card forward typing was performed on these donors using a variety of murine monoclonal, polyclonal and experimental anti-A reagents; no A antigen was detected on their RBCs. Adsorption-elution studies were performed on the RBCs of 3 donors heterozygous for O2 using human-source polyclonal anti-A but no A antigen was detected. All 34 group O RBC units were transfused to group O recipients without reported hemolytic reactions. Our results suggest that in its heterozygous state the O2 allele does not necessarily induce ABO discrepancies in automated testing. Transfusing group O RBCs from donors with an O2 allele to recipients with anti-A did not result in reports of symptoms associated with hemolytic events in the recipients.

Disclosure: No relevant conflicts of interest to declare.

Author notes

*

Corresponding author