PRIMING THE Bw4 SEQUENCE MOTIF AS A SUPPLEMENT TO THE InnoLiPA HLA-A and HLA-B typing SYSTEMS: ENHANCED RESOLUTION AND EVIDENCE THAT SECONDARY STRUCTURE IN TARGET AMPLICONS INTERFERES WITH PROBE BINDING.
Wayne Shumway and William M LeFor, LifeLink Foundation, Inc., Tampa, FL.
The InnoLiPA HLA-A and HLA-B typing systems are reverse line blot, SSOPH assays that probe exons 2 and 3 and provide intermediate to high resolution typing results at the allele level. When interpreted at the antigen (serologic) level, the HLA-A assay has 3 ambiguous probe patterns, while the HLA-B assay has 70-90 (depending on how one accounts for alleles like B*4012, which sometimes cannot be distinguished from B*4803, but may be mistaken for B48 by serology). In an effort to overcome the most blatant discrepancies, we designed primers to amplify alleles bearing the Bw4 sequence motif. Using the original InnoLiPA amplicons as templates, nested amplification of alleles bearing the Bw4 sequence motif will resolve 1 A-locus and 44 B-locus ambiguities, all of which involve obvious serologic differences (A25 vs A32, B38 vs B39, B49 vs B50, etc.). The primer pair used includes a 5 generic primer that binds to the second exon of HLA-A, B and C alleles at codons 1-8 (GCTCCCACTCCATGAGGTATTTC). The 3 primer (CGCTCTGGTTGTAGTAGCGGA) is 5 biotinylated and recognizes codon 83 (TCC), which is unique to the Bw4 alleles. The Bw4 amplicons are hybridized with InnoLiPA HLA-A or HLA-B probe strips under normal assay conditions. Some probes give much stronger signal when hybridized with the Bw4 amplicons as opposed to the original amplicons spanning exons 2 and 3. Most striking is HLA-B probe #48, which recognizes the Bw4 region of B*37 and some B*27 alleles. This probe is at best weakly positive and sometimes false negative in the standard assay (a fact acknowledged in the product insert). When binding to the Bw4 amplicons however, the signal from this probe is equal to or greater than that of any other. Dilutional studies prove that this is not due to differences in the quantity of target amplicons.
During initial testing of the standard InnoLiPA HLA-A and HLA-B assays, we had made another unexpected observation. Hybridization and stringent wash temperatures below the optimum resulted in some probes having weaker instead of stronger or false positive signals.
We believe the most likely explanation for these observations is the formation of secondary structure within target amplicons which reduces or blocks access to certain probe sites. With the HLA-A and HLA-B amplicons each being about 1000 nucleotides long, the potential for these molecules to fold over on themselves is significant as compared to the Bw4 amplicon, which has a length of less than 300 nucleotides.
Specific amplification of alleles bearing the Bw4 sequence motif can greatly improve the resolution of the InnoLiPA HLA-A and HLA-B typing systems. This modification also provides evidence that, when hybridizing amplicons generated in the routine assay, secondary structure may interfere with the binding of some probes. If this is the case, amplification of exons 2 and 3 separately, without the intervening intron, may improve overall performance.