Oxidative stress and neurodegenerative diseases: a bidirectional Mendelian randomization study.

Abstract

Introduction: Oxidative stress (OS) has been linked to neurodegenerative diseases in numerous epidemiological studies; however, whether it is a pathogenesis or a downstream factor remains controversial.

Methods: A two-sample bidirectional Mendelian randomization (MR) analysis was implemented to examine evidence of causality of 15 OS injury markers with 3 major neurodegenerative diseases using available genome-wide association studies statistics. As a main approach, inverse-variance weighted (IVW) analysis was performed. The weighted-median (WM) analysis was used to validate the relationship. In order to investigate the existence of horizontal pleiotropy and correct the IVW estimate, the Radial MR approach was applied. To gauge the consistency and robustness of the findings, several sensitivity and pleiotropy analyses were used. For this analysis, p < 0.05 indicates a nominally causal association; according to the Bonferroni correction test, p < 0.0011 indicates a statistically significant causal association.

Results: Via IVW and WM, in directional MR, it was genetically predicted that zinc was nominally causally correlated with the risk of Parkinson's disease but not after Bonferroni correction test; alpha-tocopherol was nominally causally correlated with the risk of Amyotrophic lateral sclerosis (ALS) but not after Bonferroni correction test; furthermore, in reverse MR, it was genetically predicted that Alzheimer's disease was causally correlated with uric acid but not after Bonferroni correction test. These above findings were stable across sensitivity and pleiotropy analyses.

Conclusions: Based on the current study, there is no authentic genetic causal association between OS biomarkers and neurodegenerative diseases. The complex relationship is required to be confirmed in future experimental research.