Abstract Body (Do not enter title and authors here): Background: Type 1 diabetes (T1D) is an autoimmune disorder resulting from destruction of beta cells in the pancreas by autoreactive CD8 T cells. AI4 is a highly pathogenic CD8 T cell clone, which was previously isolated from the islets of a young diabetes-prone non-obese diabetic (NOD) mouse. The AI4 T cell receptor (TCR) recognizes multiple natural and synthetic peptide ligands presented by the class I MHC molecule H2-D^b. The highly diabetogenic nature of AI4 is likely linked to this promiscuity, or cross-reactivity, which can trigger autoimmunity. However, the underlying mechanisms by which this autoreactive AI4 TCR accommodates diverse ligands remain to be determined.
Hypothesis: Defining the chemical and structural determinants of H2-D^b/peptide interactions and autoreactive AI4 TCR cross-reactivity will aid in understanding the recognition mechanism in T1D.
Methods: H2-D^b/peptide complexes recognized by AI4 were refolded from inclusion bodies and purified by size exclusion chromatography. The complexes were crystallized and structures determined by X-ray crystallography.
Results: Three crystal structures of H2-D^b with AI4 ligands (YQLENYCAL, derived from insulin; FQDENYLYL, derived from dystrophia myotonica kinase; and YFIENYLEL, a synthetic mimotope peptide) were determined. The structures of these ligand complexes highlight a common network of interactions. The side chain of Asn5 of the peptide is a key anchoring residue, fitting inside the C-pocket and participating in polar interactions with the side chains of Gln94 and Gln121 of the H2-D^b heavy chain. In all three complexes, Glu4 and Tyr6 side chains are solvent-exposed to serve as potential TCR contact points. In the complexes of FQDENYLYL and YFIENYLEL, Tyr8 and Glu8, respectively, are also solvent-exposed and may serve as additional TCR contact points. Because of its shorter side chain, Ala8 in YQLENYCAL does not seem to be solvent-exposed.
Conclusion: Conserved Glu4 and Tyr6 of the peptides are solvent-exposed and available for contact with the TCR in the three complexes, suggesting they play important roles in the recognition by the AI4 TCR. This hypothesis will be verified by obtaining the structures of the AI4 TCR bound to these H2-D^b complexes. The promiscuous and pathogenic nature of the autoreactive AI4 TCR can be understood by elucidating its recognition mechanism. This information could be used to block promiscuous TCRs to prevent T1D, a major risk factor for cardiovascular disease.