Interaction of HLA Class II rs9272219 and TMPO rs17028450 (Arg690Cys) Variants Affects Neuromyelitis Optica Spectrum Disorder Susceptibility in an Admixed Mexican Population.

Abstract Neuromyelitis Optica Spectrum Disorder (NMOSD) is a demyelinating autoimmune disease of the central nervous system. It has been consistently suggested that NMOSD is more frequent in non-European populations, and Native American ancestry was recently found to contribute to NMOSD susceptibility in the admixed Mexican population. As many other autoimmune diseases, NMOSD is a multifactorial disorder resulting from complex interactions between genetic and environmental factors. A GWAS identified that HLA class II rs9272219 is associated with NMOSD risk in the Mexican population. Moreover, experimental evidence in Xenopus and humans strongly suggests that thymopoietin is involved in T-cell differentiation and is highly expressed in immune cells. The TMPO rs17028450 missense variant (Arg690Cys) is frequent in Latin Americans but very rare in other continental populations. We thus aimed to explore whether this variant is associated with NMOSD, and whether there is an interaction between this variant and the HLA class II rs9272219 affecting NMOSD risk. A total of 119 NMOSD patients and 1208 controls were included in the study. HLA rs9272219 frequency ranges from 21% to 68% in Native Mexican populations; the minor rs17028450 allele frequency is less than 2% in Asian, African and European populations, 11% in Mexican American population and up to 18% in Native Mexicans. Both SNPs showed a significant association with NMOSD risk, rs9272219 (OR=2.5, P=8x10-10) and rs17028450 (OR=1.9, P=0.002). In addition, we identified an interaction between both SNPs that led to an estimated 3.9-fold increased NMOSD risk for those individuals having both risk alleles (P=0.05). Gene-gene interaction effects have not been fully explored in most complex diseases because of methodological difficulties. HLA class II and TMPO genes are not only biologically meaningful for autoimmune diseases, but they are also meaningful for the Latin America population because of their high frequency. To our knowledge, this is the first study that evaluates the interaction of specific susceptibility gene variants that might contribute to the genetic architecture of NMOSD in admixed Latin American populations.