Genome-Wide Analysis of Genetic Risk Factors for Rheumatic Heart Disease in Aboriginal Australians Provides Support for Pathogenic Molecular Mimicry

Authors

Lesley-Ann Gray, University of Melbourne
Heather D'Antoine, Charles Darwin University
Steven Tong, Charles Darwin University, University of Melbourne
Melita McKinnon, Charles Darwin University
Dawn Bessarab, University of Western Australia
Ngiare J. Brown, University of WollongongFollow
Bo Remenyi, Charles Darwin University
Andrew C. Steer, Murdoch Children's Research Institute, University of Melbourne, Royal Children's Hospital Melbourne
Genevieve Syn, University of Western Australia
Jenefer Blackwell, University of Western Australia
Michael Inouye, University of Melbourne
Jonathan R. Carapetis, University of Western Australia, Charles Darwin University

RIS ID

118240

Publication Details

Gray, L., D'Antoine, H. A., Tong, S. Y. C., McKinnon, M., Bessarab, D., Brown, N., Remenyi, B., Steer, A., Syn, G., Blackwell, J. M., Inouye, M. & Carapetis, J. R. (2017). Genome-Wide Analysis of Genetic Risk Factors for Rheumatic Heart Disease in Aboriginal Australians Provides Support for Pathogenic Molecular Mimicry. Journal of Infectious Diseases, 216 (11), 1460-1470.

Abstract

Background. Rheumatic heart disease (RHD) after group A streptococcus (GAS) infections is heritable and prevalent in Indigenous populations. Molecular mimicry between human and GAS proteins triggers proinflammatory cardiac valve-reactive T cells. Methods. Genome-wide genetic analysis was undertaken in 1263 Aboriginal Australians (398 RHD cases; 865 controls). Singlenucleotide polymorphisms were genotyped using Illumina HumanCoreExome BeadChips. Direct typing and imputation was used to fine-map the human leukocyte antigen (HLA) region. Epitope binding affinities were mapped for human cross-reactive GAS proteins, including M5 and M6. Results. The strongest genetic association was intronic to HLA-DQA1 (rs9272622; P = 1.86 x 10−7). Conditional analyses showed rs9272622 and/or DQA1*AA16 account for the HLA signal. HLA-DQA1*0101_DQB1*0503 (odds ratio [OR], 1.44; 95% confidence interval [CI], 1.09-1.90; P = 9.56 x 10−3) and HLA-DQA1*0103_DQB1*0601 (OR, 1.27; 95% CI, 1.07-1.52; P = 7.15 x 10−3) were risk haplotypes; HLA_DQA1*0301-DQB1*0402 (OR 0.30, 95%CI 0.14-0.65, P = 2.36 x 10−3) was protective. Human myosin cross-reactive N-terminal and B repeat epitopes of GAS M5/M6 bind with higher affinity to DQA1/DQB1 alpha/beta dimers for the 2-risk haplotypes than the protective haplotype. Conclusions. Variation at HLA_DQA1-DQB1 is the major genetic risk factor for RHD in Aboriginal Australians studied here. Cross-reactive epitopes bind with higher affinity to alpha/beta dimers formed by risk haplotypes, supporting molecular mimicry as the key mechanism of RHD pathogenesis.