T-cell response to gluten in patients with HLA-DQ2.2 reveals requirement of peptide-MHC stability in celiac disease

Abstract

Background & aims: Celiac disease is a diet-induced, T cell-mediated enteropathy. The HLA variant DQ2.5 increases risk of the disease, and the homologous DQ2.2 confers a lower level of risk. As many as 5% of patients with celiac disease carry DQ2.2 without any other risk alleles. Epitopes commonly recognized by T cells of patients with HLA-DQ2.5 bind stably to DQ2.5 but unstably to DQ2.2. We investigated the response to gluten in patients with HLA-DQ2.2.

Methods: We generated intestinal T-cell lines and clones from 7 patients with HLA-DQ2.2 (but not DQ2.5) and characterized the responses of the cells to gluten. The epitope off-rate was evaluated by gel filtration and T cell-based assays. Peptide binding to DQ2.2 was studied with peptide substitutes and DQ2 mutants.

Results: Patients with DQ2.2 and no other risk alleles had gluten-reactive T cells that did not respond to the common DQ2.5-restricted T-cell epitopes. Instead, many of the T cells responded to a distinct epitope that was not recognized by those from patients with HLA-DQ2.5. This immunodominant epitope bound stably to DQ2.2. A serine residue at P3 was required for the stable binding. The effect of this residue related to a polymorphism at DQα22 that was previously shown to determine stable binding of peptides to DQ2.5.

Conclusions: High levels of kinetic stability of peptide-major histocompatibility complexes are required to generate T-cell responses to gluten in celiac disease; the lower risk from DQ2.2 relates to constraints imposed on gluten peptides to stably bind this HLA molecule. These observations increase our understanding of the role of the major histocompatibility complex in determining T-cell responses in patients with celiac disease and are important for peptide-based vaccination strategies.