Celiac disease (CD) is a malabsorptive disorder precipitated in genetically susceptible individuals by ingestion ofgluten . A significant part of the disease predisposition maps to the HLA class II region, but the primary HLA association is not known (1) . DR3 seems to be a susceptibility allele together with any other DR allele while DR7 is associated to CD almost only in combination with DR3 or DR5 (2, 3).The HLA-DR3DQw2 haplotype demonstrates the strongest association to CD (1-5) . We investigated whether DR3DQw2 negative patients might share parts of the genetic information ofthis HLA haplotype but encoded in trans position. When we compared amino acid sequences of different DQa and DQß chains we found that the DQa chain encoded by the DR3DQw2 haplotype is identical to that encoded by DR5DQw7 (6, 7), while the DQß chain of the DR3DQw2 haplotype is identical to that of DR7DQw2 except for a single amino acid difference in the second domain (8, 9; Lee, J. S., personal communication) . The pairing of a and ß chains seems to be determined by amino acid residues located in the first domains (10). Thus, individuals wao are DR5DQw7/DR7DQw2 heterozygous may by transcomplementation express the same (or almost the same) DQ a/0 heterodimers as those formed by DQAl and DQBI genes in cis position on the DR3DQw2 haplotype (see Fig . 1). As a first step to confirm this hypothesis we typed 94 CD children with synthetic DQA1 and DQB1 allele-specific oligonucleotide (ASO) probes. Here we report that all except one ofthese CD patients carry DQAI and DQB1 genes that may encode the same DQ a/ß heterodimer.

The HLA-associated susceptibility to develop celiac disease (CD) seems mainly to be conferred by a particular HLA-DQ heterodimer encoded by the DQA1*0501 and DQB1*0201 genes either in cis or in trans position. To study the possible influence of DRB1 or other DQA1 and DQB1 alleles on the CD susceptibility conferred by these DQ genes, we performed genomic HLA typing of 94 CD patients, selected those who carried at least one copy of the DRB1*0301-DQA1*0501-DQB1*0201 haplotype (N = 89) and compared them to 47 random, healthy Norwegians matched with the patients to carry at least one copy of the above haplotype. We found an excess of DQB1*0201 homozygosity in the patients. This was due to an increased frequency of the DRB1*0301-DQA1*0501-DQB1*0201 and DRB1*0701-DQA1*0201-DQB1*0201 haplotypes present on the other chromosome. We propose that, in individuals carrying the DQA1*0501 and DQB1*0201 alleles, the presence of a second copy of the DQB1*0201 allele increases susceptibility to CD.

Celiac disease (CD) is a common chronic inflammatory disorder of the small intestine with a multifactorial aetiology. HLA is a well-known risk factor, but other genetic factors also influence disease susceptibility. To identify the genes involved in this disorder, we performed a genome-wide scan on 106 well-defined Swedish and Norwegian families with at least two affected siblings. We investigated familial segregation of 398 microsatellite markers, and utilised non-parametric linkage analysis. The strongest linkage with disease was found to the HLA locus (6p) (P50.000006). There were eight regions besides HLA with a point wise P value below 0.05. Among these eight regions were 11q and 5q, both of which have been suggested in several linkage studies of independent celiac disease families. We also performed a stratification analysis of families according to their HLA genotypes. This resulted in significant differences on chromosome 2q. These results indicate that 11q, 5q and possibly also 2q are true susceptibility regions in CD. European Journal of Human Genetics (2001) 9, 938 ± 944.

Coeliac disease (CD) is a T-cell mediated immunological disease of the small intestine which is precipitated in susceptible individuals by ingestion of gluten. We recently reported that gliadin-specific T cells can be found in the small intestinal mucosa of CD patients, and that a preponderance of these T cells was restricted by the CD-associated DQ(alpha 1*0501,beta 1*0201) heterodimer. Here we report studies on whether the same is found for gliadin specific T cells in the peripheral blood of CD patients. T-cell responses towards gluten antigens in vitro were found for both most CD patients and healthy controls. Gluten-specific T-cell clones (TCC) were established from four CD patients. Although a large proportion of these TCC were restricted by DQ molecules, including the CD-associated DQ(alpha 1*0501,beta 1*0201) heterodimer, several were restricted instead by DR or DP molecules. Thus, gluten-derived peptides can be presented to T cells by several different HLA class-II molecules, and the preferential DQ(alpha 1*0501,beta 1*0201) restriction of gluten-specific T cells in the small intestinal mucosa of CD patients is less pronounced than for similar T cells in the peripheral blood.