DNA repair genes versus age at onset and expanded repeat instabilities in Machado-Joseph disease

Abstract Growing evidence links DNA repair to CAG repeat expansion diseases, though little of this has been studied in Machado-Joseph disease, also known as spinocerebellar ataxia type 3 (SCA3/MJD). We aimed to evaluate if DNA mismatch repair would influence this disease, using a cross-sectional design where functional variants in genes related to this system were tested as independent variables, and intergenerational instability, age at onset of symptoms and velocity of disease progression were the outcomes, comparing also with the effects of FAN1 and ATXN2 variants. Symptomatic carriers with <81 CAG repeats were genotyped for rs1799977 (MLH1), rs175080 (MLH3), rs2303425 (MSH2), rs2250063 (MSH3), rs1042821 (MSH6), rs3791767 (PMS1), rs1805323 (PMS2), rs3512 (FAN1), and the longer normal CAG allele at ATXN2. Best models of inheritances were used for the analyses. Gene–gene interaction effects were further explored among significant loci. Variants in MLH3 and PMS1 were significantly associated with the intergenerational instability of CAG repeat expansions, while variants in MLH1, MSH3 and ATXN2 modified the age at onset. Underdominant effects related MLH3 heterozygotes to larger CAG repeat expansion instabilities and MSH3 heterozygotes to earlier ages at onset, while overdominant effects related MLH1 heterozygotes to older ages at onset. The strong effect of ATXN2 large normal alleles over age at onset made the analysis of interaction difficult. The other analyses did not show significant effects. Our results proposed that mismatch repair genes impact over CAG repeat expansion germline instabilities in SCA3/MJD. Moreover, we showed that they impacted on the age at onset, probably through additional somatic instabilities. The occurrence of over and underdominance in mismatch repair gene variants was also evident and needs to be considered in future studies.