Identification of two new HLA alleles: B*3546* and B*5611*. How reliable are the published HLA-B intron 2 sequences?

Polymerase chain reaction-sequence-specific primer (PCR-SSP) typing for human leukocyte antigen (HLA)-B in a male 25-year-old Caucasian individual of Iranian origin and in a 42-year-old German Caucasian bone marrow donor revealed reaction patterns that did not agree with any known HLA specificity, thus suggesting in both cases the existence of a novel allele. Sequence-based typing (SBT) after allelic separation revealed the sequences of the new alleles HLA-B*5611 and B*3546. The sequence patterns of both new alleles might have been generated as the results of double crossing over, possibly over several generations. During the analysis of the HLA-B*3546 intron 2 sequence for possible crossing over points, a base insert, an additional G after position 700, was found. This insert was analyzed using SBT and PCR-SSP and was found to be present not only in all samples carrying B*35, but also in all HLA-B specificities tested. It appears that all known HLA-B alleles may contain a G insert at position 700 of intron 2, and that the published intron 2 sequence alignments of the HLA-B locus may contain errors at this position.