Bidirectional two-sample Mendelian randomization analysis identifies causal associations between oxidative stress and Parkinson's disease.

Abstract

Observational studies have shown that oxidative stress (OS) is associated with Parkinson's disease (PD). However, whether such observations reflect cause-effect remains largely unknown. To test this, we performed a two-sample bidirectional Mendelian randomization (MR) analysis to investigate the causal-effects between OS biomarkers and PD. We selected summary statistics data for single-nucleotide polymorphisms (SNPs) associated with catalase (n = 13), glutathione peroxidases (n = 12), superoxide dismutase (n = 13), vitamin A (n = 7), vitamin C (n = 10), vitamin E (n = 12), vitamin B12 (n = 8), folate (n = 14), copper (n = 6), Zinc (n = 7), and iron (n = 23) levels, and the corresponding data for PD from the International Parkinson Disease Genomics Consortium (IPDGC, 33,674 cases and 449,056 controls). Inverse-variance weighted (IVW) MR analyses were conducted to estimate associations of OS with PD. Reverse MR analysis was further performed to predict the causal effects of PD on the above OS biomarkers. As for PD, the IVW method suggested that the Zinc (Zn) levels was significantly associated with PD (OR = 1.107, 95% CI 1.013-1.211; p = 0.025), which is consistent with results from the weighted median analyses. Moreover, the results remained consistent and robust in the sensitivity analysis. However, there were no significant associations of catalase, glutathione peroxidases, superoxide dismutase, vitamin A, vitamin C, vitamin E, vitamin B12, folate, copper, or iron with PD. As for OS, our reverse MR analysis also did not support a causal effect of liability to PD on OS. The MR study supported the causal effect of Zn on PD. These findings may inform prevention strategies and interventions directed toward OS and PD.