Association of MYH9 and lupus nephritis in multiple populations

Abstract

Objectives: Systemic lupus erythematosus (SLE) has many debilitating or even fatal manifestations. Lupus nephritis is among the most severe of these and affects African Americans (AA) three times more than European Americans (EA) and women more than men. Genetic determinants of SLE susceptibility and specific SLE manifestations are supported by high twin concordance rates, linkage and association study findings and the identification of variants associated with disease. Given the threefold increased risk of lupus renal disease in certain populations we chose in this study to investigate a recently identified gene for non-diabetic end stage renal disease, MYH9 (Myosin, heavy Chain 9), shown to be associated in African Americans. To this end, we performed genetic association tests in and around MYH9 in six different populations of lupus cases shown to have renal disease. Methods: We investigated MYH9 in independent cases and controls respectively, from six population groups: AA (635 / 1734), EA (1119 / 3546), Asian (531 / 1270), Native American (111 / 193), Hispanic (548 / 619) and Gullah (70 / 122). Cases were defined, from among SLE cases as individuals reported as positive for the American College of Rheumatology renal criteria. Controls were defined as either SLE cases without renal involvement or healthy individuals. Results were similar and therefore only the latter presented. Seventy-three single nucleotide polymorphisms (SNPs) in moderate linkage disequilibrium (LD) (r2 > 0.80) were selected. We then performed single SNP tests for association with renal disease status using logistic regression adjusted for global African, European and Asian ancestry age and sex assuming either an additive or dominant genetic model. Results: In EA, the association peak centered around 35,040,000 base pairs with multiple SNPs showing association. We identified three SNPs rs5750250, rs2413396 and rs4820232 within MYH9 that were strongly associated with renal disease (1null 10-4(less-than or equal to) p (less-than or equal to) 9 null10-4) with high LD (D' (greater-than or equal to) 0.99).The best associated SNP was rs5750250 ( P = 2.32 null 10 -4 ). In the Gullah, significance of association centered at 22q35.05. No statistically significant associations were seen in populations of other ancestral backgrounds. Conclusions: While we did not replicate previously reported findings in AA, a significant association in the Gullah population, despite a sample size of only 192, confirms the elevated risk of lupus nephritis in populations of African ancestry. The presence of our strongest effect in EA is a novel finding and suggests the role of MYH9 in lupus nephritis in diverse populations.