Objective: Celiac disease is closely correlated with certain human lymphocyte antigen (HLA) alleles. The aim of this study was to compare linkage disequilibrium parameters and the frequencies of the two loci haplotypes: HLA A/B, A/C, A/DR, A/DQ; HLA B/C, B/DR, B/DQ; HLA C/DR, C/DQ and HLA DR/DQ in children with celiac disease and in a control population within the same geographical area.
Methods: Thirty-eight children with celiac disease, aged 5months to 18years at diagnosis, were HLA typed by microlymphocytotoxicity assay using T and Bcells separated by monoclonal antibody labeled immunomagnetic particles. The frequency of each haplotype of two loci (Hij) depends on the frequency of each allele (pi and pj) and on a correction factor delta, according to the formula: Hij5 Dij1(pi3pj). The existence of a correction factor delta, or linkage disequilibrium, was assessed by a chi square test using 2 X 2contingency tables for gametic association.
Results: Among children with celiac disease the most frequent and significant haplotypes were A1/B8, A9/B5, A19/B12, A28/B22, A28/Cw1, A9/DQ3, B8/Cw7, B18/Cw5, B22/Cw1, B5/DR5, B8/DR3, B12/DR7, B5/DQ3, DR3/DQ2, DR4/DQ8 (3) and DR5/DQ3, showing a positive linkage disequilibrium. Negative linkage disequilibrium was found between B18/Cw7, B12/DR3, Cw4/DR3 and DR3/DQ3.
Conclusions: Our findings show that the frequency of A1/B8,A19/B12, B8/DR3,B12/DR7 and DR3/DQ2 haplotypes is higher in children with celiac disease than in the control population and suggest that these two loci haplotypes confer susceptibility to celiac disease.