DNA methylation age acceleration is associated with age of onset in Chinese spinocerebellar ataxia type 3 patients

• Highly variable age of onset was observed in a Chinese SCA3 family and SCA3 cohort • The genome-wide DNA methylation study estimated DNAm age (epigenetic clock), a biomarker of biological aging status • DNAm age acceleration is significantly associated with SCA3 age of onset Spinocerebellar ataxia type 3 (SCA3), also known as Machado Joseph disease (MJD), is a common dominantly inherited ataxia, and has heterogeneous clinical features and variable age of onset, ranging from 10 to 78 years. Repeats variability of ATXN3, HTT, ATN1 and ATXN2 can explain partially but not fully SCA3 age of onset heterogeneity. Aging is a reported modifier of SCA3 severity and closely linked to DNA methylation (DNAm). DNAm age acceleration was associated with disease risk and/or variable disease phenotypes in several repeat associated neurodegenerative diseases (such as Huntington's disease and Amyotrophic lateral sclerosis). To understand if DNAm age acceleration is associated with SCA3 age of onset, we performed a genome-wide DNAm study of a Chinese SCA3 family with variable age of onset and clinical presentations. All patients showed unsteady gait, deterioration of extremities coordination, speech (dysarthria) and swallowing problems (dysphagia, choking on eating and/or drinking) and oculomotor abnormalities, with variable age of onset ranging from 27 to 52 years. We found that DNAm age acceleration is associated with age of onset ( p -value = 0.0023, B = -1.26), suggesting that every 5 year increase in DNAm-age acceleration is corresponding to a 6.3 year earlier disease onset. This association remains significant after the adjustment to ATXN3 CAG repeats (adjusted p -value = 0.037, adjusted B = -1.0). In an independent SCA3 cohort (n = 40), we also observed the association between DNAm age acceleration and age of onset (adjusted p -value = 0.007, adjusted B = -0.69). Of note, we found no significant association between DNAm of single-CpG locus and/or CpG-SNPs and SCA3 age of onset in the current family or the SCA3 cohort. Our findings suggested that DNAm age acceleration might be a SCA3 age of onset modifier, and encourage further investigations in extended SCA3 cohorts to clarify the role of epigenetic aging in modifying disease onset.