HLA-DR typing using restriction fragment length polymorphism (RFLP) with one enzyme and two probes.

A panel of 43 homozygous and 36 heterozygous highly selected cells, representing the most common DR-specificities, were investigated with the DNA hybridization technique. By using a single restriction enzyme, TaqI, and two probes, DR beta and DQ alpha, it was possible to construct assignment criteria giving a reasonable definition of DR1, 3, 4, 5, 7 + w9, w8, w10 and w11. The criteria sometimes require that certain bands must not be present. Therefore, in certain genotypic combinations, the presence or absence of the particular specificity on one haplotype cannot be decided. This is a problem only for DR2 and DRw6, which for this reason cannot be assigned in about 1/3 to 1/4 of the cases. The association between RFLP assignment and serological assignment is not absolute, the correlation coefficients ranking from 0.62 to 1.0. In the case of false negative RFLP assignment, this may be due to genetic heterogeneity, as in the case of a DR2 individual who proved to be Dw12 and not Dw2 associated. It is often stated that interpretation of the RFLP pattern is particularly difficult in random or heterozygous individuals compared to proven homozygotes. This is not the case in the present study, where in fact correlation coefficients between RFLP and serologically determined DR specificities were higher in the heterozygotes (range 0.79-1.00).