P096 HLA-DPB1 level of expression, determined by SNP analysis, and serologic epitopes

Abstract

Aim SNPs rs2281389 and rs9277534 in the 3 ′ UTR region of the DPB1 gene have been found to be associated with the levels of expression of DPB1 molecules and with graft-versus-host disease in haematopoietic stem cell transplantation. (New Engl J Med 2015 373:599) HLA-DPB1 alleles show 2 serologic epitope dimorphisms that classify DPB1 alleles into 4 serologic groups: DP1, DP2, DP3 and DP4. (Hum Immunol 2009 70:839) The purpose of this study is to measure the relationship between these 3’UTR expression-linked SNPs (EL-SNP) and the previously described serologic groups. Methods A typing set of 4,587 DPB1 allele assignments for which EL-SNP status is known was used. 2 × 2 contingency tables were built with the following pairs of variables: (DP2 & DP4) versus EL-SNP haplotype AA, DP1 versus EL-SNP hap. GA, DP3 versus EL-SNP hap. GG. Results The chi square values for these 2 × 2 tables were as follows: (DP2 & DP4) versus EL-SNP haplotype AA: chi sqr 4046 DP1 versus EL-SNP hap. GA: chi sqr 4192 DP3 versus EL-SNP hap. GG: chi sqr 4137 Conclusions The chi square in all cases is very high showing a very strong association. Any observation, like GVHD in DP mismatched bone marrow transplants, made regarding EL-SNP cannot be separated from DPB1 serologic groups. It is a mistake to claim that any difference in GVHD in two groups with different EL-SNP haplotypes is due to the effect of these SNPs on levels of expressions of DP molecules without considering the more straightforward possibility that the effect is due to a mismatch between well-known serologic groups. T cells do recognise immunological differences also detected by antibodies.