No abstract

We describe a novel high resolution DNA based typing approach for HLA class I alleles, which identifies the recombinational motifs present in exons 2 and 3 of the HLA class I genes. Unique identification patterns for 201 known HLA-A, HLA-B, and HLA-Cw alleles were generated by the use of only 40 probes, which were targeted at these common motifs. The unambiguous identification of the alleles was achieved by the development of a new and powerful allelic separation technique that allows isolation of single alleles after amplification. To validate the method, we have used locus-specific primers to amplify exons 2 and 3 of HLA-A, HLA-B, and HLA-Cw loci from 22 heterozygous and 41 homozygous cell lines. After amplification, the allelic fragments from each locus were separated, blotted, and hybridized with the 40 probes. In all cases, the allelic products could be separated and 81 different class I alleles, 33 HLA-A, 30 HLA-B, and 18 HLA-Cw, were identified according to the predicted probe hybridization patterns.HLA-A, HLA-B, and HLA-Cw antigens are polymorphic cell surface glycoproteins, which are involved in cytotoxic T-cell recognition during the immune response. HLA matching is essential for kidney and bone marrow transplantation (1-4) and new molecular biological methods (5) for detection of HLA polymorphism currently provide an opportunity to improve matching of the patients and donors, as well as a research tool to investigate the relationship between HLA disparity and transplant complications. These molecular typing methods include sequence-specific primer (SSP) amplification (6, 7), hybridization with sequence-specific oligonucleotide (SSO) probes (8, 9), heteroduplex analysis (10-12), single-strand conformation polymorphism (13, 14), and direct nucleotide sequencing (14,15). SSP uses group and/or allele specific sequences for primer design (7) and has been proposed as a single technique for HLA class I and II typing (16); SSO is based on the identification of allele-specific sequences, which are either unique or in certain combinations will allow the identification of the allele (9). The resolution of each method could be increased by using a larger number of primer mixes for SSP and more probes in the case of SSO. These two methods have also been used for class I typing, but their application to high-resolution class I allele analysis has been hindered due to the extensive sequence homology between the HLA-A, -B and -Cw loci (17), between class I classical and nonclassical genes and the reported pseudogenes (18). Direct sequencing is at present used for high-resolution analysis of the class I alleles (14, 15) and, to avoid ambiguous combinations, prior typing is required, which is then followed by allelespecific PCR amplification to generate single allelic fragments. Although this approach achieves the highest resolution, it is expensive and difficult to perform routinely. Heterozygous combinations of certain alleles frequently produce ambiguous

We describe a method, complementary strand analysis (CSA), for separating alleles potentially from any heterozygous genetic locus. Locus specific PCR is performed generating two allelic products. The antisense strands are isolated and hybridised with a sense reference strand to form a chimeric DNA duplex for each allele which is then separated by non-denaturing PAGE. We demonstrate the application of CSA for separation of highly polymorphic HLA-A, -B and -Cw alleles and characterisation of HLA identity in related bone marrow donors and patients. CSA is capable of resolving one nucleotide differences in a DNA fragment nearly as large as a kilobase in length.

No abstract