Open AccessLowMICAGene Expression Confers an Increased Risk of Graves' Disease: A Mendelian Randomization Study
Author(s) -
Yoichi Sutoh,
Shohei Komaki,
Taiki Yamaji,
Shiori Suzuki,
Ryoko Katagiri,
Norie Sawada,
Kanako Ono,
Hideki Ohmomo,
Tsuyoshi Hachiya,
Yayoi Otsuka-Yamasaki,
Akira Takashima,
So Umekage,
Motoki Iwasaki,
Atsushi Shimizu
Publication year - 2022
Publication title -
thyroid
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.918
H-Index - 142
eISSN - 1557-9077
pISSN - 1050-7256
DOI - 10.1089/thy.2021.0417
Subject(s) - mendelian randomization , expression quantitative trait loci , nkg2d , biology , gene expression , immunology , odds ratio , locus (genetics) , linkage disequilibrium , confidence interval , allele , cytotoxic t cell , oncology , gene , genetics , medicine , genotype , single nucleotide polymorphism , haplotype , in vitro , genetic variants
Background: Expression of natural killer group 2 member D (NKG2D) ligand (NKG2DL) plays a major role as a "danger signal" on stressed cells to promote removal of the latter by NKG2D-expressing cytotoxic lymphocytes. NKG2DL expression has been found in peripheral immune cells as well, such as in macrophages; however, the effect of this expression is yet to be determined. Methods: We determined instrumental variables (IVs; R 2 <0.01 in linkage disequilibrium), explaining the major variance in major histocompatibility complex class I chain-related protein A (MICA) and B (MICB) gene expression levels from the expression-quantitative trait locus (eQTL) of NKG2DLs based on the RNA-seq analysis of peripheral blood mononuclear cells (PBMCs) from 381 Japanese. Simultaneously, the target outcomes were filtered by PheWAS from 58 health risks, using a community-based cohort study composed of 44,739 Japanese residents. Finally, we estimated the causal effect of gene expression levels on the outcomes using the Mendelian randomization approach. Results: We determined nine and four IVs, explaining 87.6% and 33.0% of MICA and MICB gene expression levels, respectively. In the association test, we identified 10 or 13 significant outcomes associated with the MICA or MICB eQTLs, respectively, as well as the causal effect of MICA expression on Graves' disease (GD) ( p = 4.2 × 10 -3 ; odds ratio per 1 S.D. difference in the expression: 0.983 [confidence interval: 0.971-0.995]), using the weighted median estimator, without significant pleiotropy ( p > 0.05), and the results were consistent across the sensitivity analyses. Conclusions: Our study provide novel evidence associating NKG2DL expression with GD, an autoimmune thyroiditis; direction of the effect indicated the immunoregulatory role of MICA expression in PBMCs, suggesting the importance of further functional assays in inflammatory diseases.