Profiling associations of interactive ligand-receptors (HLA class I and KIR gene products) with the progression to type 1 diabetes among seroconverted participants.

Abstract

AIMS/HYPOTHESIS The aim of this work was to explore associations between type 1 diabetes progression from stages 1 or 2 to stage 3 and interacting ligand-receptor complexes of HLA class I (HLA-I) and KIR gene products. METHODS Applying next-generation sequencing technology to genotype HLA-I genes (HLA-A, -B, -C) and KIR genes (KIR2DL1, KIR2DL2, KIR2DL3, KIR2DL4, KIR2DL5, KIR2DS1, KIR2DS2, KIR2DS3, KIR2DS4, KIR2DS5, KIR3DL1, KIR3DL3, KIR3DS1, KIR2DP1, KIR3DP1) from 1215 participants in the Diabetes Prevention Trial-Type 1 (DPT-1) and the Diabetes Prevention Trial (TN07), we systematically explored associations of HLA-I-KIR ligand-receptor interactions (LRIs) with disease progression via a Cox regression model. We investigated the structural properties of identified LRI complexes. RESULTS KIR and HLA-I genes had no or sporadic associations with disease progression. Out of all possible LRIs, nine HLA-A ligands and 14 HLA-B ligands with corresponding receptors had modest associations with progression (p<0.05). As an example, carriers of A*03:01-KIR2DS4 had slower progression (HR 0.36, p=3.06 × 10-2), as did B*07:02-KIR2DL3 carriers (HR 0.26, p=7.76 × 10-3). Structural investigations of KIR-HLA-I complexes via homology modelling based on already-solved respective complex structures suggested that the respective electrostatic and van der Waals interactions encoded in the protein sequences result in strong biophysical LRIs, which could alter the progression of type 1 diabetes. CONCLUSIONS/INTERPRETATION These results reveal that LRIs of KIR-HLA-I gene products, rather than individual genes, contribute to type 1 diabetes progression, and such interactions are likely to be stabilised by electrostatic and van der Waals forces. As the KIR-HLA-I interactions involve part of the C-terminus of the antigen-binding groove of HLA-I, but may be affected by the respective bound peptide, this suggests a new mechanism for type 1 diabetes pathogenesis. DATA AVAILABILITY Clinical data on participants in DPT-1 and TN07 can be obtained from the NIDDK-Central Repository ( https://repository.niddk.nih.gov/home ) following the formal approval process.