He has uncovered key pathogenic mechanisms of the widespread condition
celiac disease. Celiac disease develops in genetically predisposed individuals on exposure to dietary cereal
gluten proteins.
Human leukocyte antigen is as a major genetic risk factor for celiac disease, and he identified the HLA allotype HLA-DQ2.5 as the major culprit HLA molecule. This HLA molecule is carried by most celiac disease patients, and the remaining patients carry HLA-DQ2.2 or HLA-DQ8. He and his coworkers demonstrated that celiac disease patients who express HLA-DQ2.5 have gluten-specific CD4+ T cells in their gut mucosa that recognize gluten antigen in the context of this HLA molecule. Later he showed that celiac disease patients who express HLA-DQ8 and HLA-DQ2.2 have gluten-specific T cells that recognize gluten antigen in context of these HLA-DQ molecules. Biochemical peptide binding studies revealed that HLA-DQ2.5 has preference for binding antigenic peptides with negatively charged residues, yet gluten proteins have few negatively charged amino acids. An answer to this conundrum came with the observation that transglutaminase 2, an enzyme to which celiac disease patients generate autoantibodies, can posttranslationally modify gluten peptides by converting certain
glutamine residues to negatively charged glutamate residues in a process called
deamidation. A little later, the group of Frits Koning published similar results. The research group of Sollid identified the sequence of immunodominant gluten peptides recognized by T cells of celiac disease patients. Together with the group of
Chaitan Khosla he solved the x-ray crystal structure of HLA-DQ2.5 with a bound deamidated gluten T-cell epitope. More recently he has transitioned his research to studies of the antibody response in celiac disease. His research aimed to answer the question why exposure to a foreign antigen (
gluten) causes production of autoantibodies. His group has established human monoclonal antibodies from single plasma cells of the celiac gut lesion, both transglutaminase 2-specific as well as gluten-specific antibodies. This work has allowed detailed analysis of how celiac antibodies recognize the autoantigen, and it has enabled the establishment of a mouse model to study autoantibody formation to transglutaminase 2. Surprisingly, there is no B-cell tolerance to transglutaminase 2 in mice that have been made knock-in for a celiac antibody that recognizes both human and mouse transglutaminase 2. In such mice the generation of autoantibodies seems to be contingent on T-cell help from gluten-specific T cells – a type of help that he early suggested to be mediated by such T cells via involvement of hapten-carrier like TG2-gluten complexes. Collectively, his work gives mechanistic insights into how the disease-predisposing HLA-DQ molecules, via presentation of posttranslationally modified gluten peptides, are connected to the generation of autoantibodies to transglutaminase 2 in celiac disease, as reviewed. == Awards and honors ==