Mendelian Randomization to Evaluate the Effect of Plasma Vitamin C Levels on the Risk of Alzheimer’s Disease

Liu Haijie,Zhang Yan,Hu Yang,Zhang Haihua,Chinese Institute for Brain Research,Han Zhifa,Gao Shan,Wang Longcai

DOI: https://doi.org/10.1186/s12263-021-00700-9

2021-01-01

Genes & Nutrition

Abstract:Until now, observational studies have explored the impact of vitamin C intake on Alzheimer’s disease (AD) risk, however, reported ambiguous findings. To develop effective therapies or prevention, the causal link between vitamin C levels and AD should be established. Here, we selected 11 plasma vitamin C genetic variants from a large-scale plasma vitamin C GWAS dataset (N = 52,018) as the potential instrumental variables. We extracted their corresponding summary statistics from large-scale IGAP clinically diagnosed AD GWAS dataset (N = 63,926) and UK Biobank AD proxy phenotype GWAS dataset (N = 314,278), as well as two UK Biobank subgroups including the maternal AD group (27,696 cases of maternal AD and 260,980 controls) and paternal AD group (14,338 cases of paternal AD and 245,941 controls). We then performed a Mendelian randomization (MR) study to evaluate the causal association between plasma vitamin C levels and the risk of AD and AD proxy phenotype. Meanwhile, we further verified these findings using a large-scale cognitive performance GWAS dataset (N = 257,841). In IGAP, we found no significant causal association between plasma vitamin C levels and the risk of AD. In UK Biobank, we found that per 1 SD increase in plasma vitamin C levels (about 20.2 μmol/l) was significantly associated with the reduced risk of AD proxy phenotype (OR = 0.93, 95% CI 0.88–0.98, P = 7.00E−03). A subgroup MR analysis in UK Biobank indicated that per 1 SD increase in plasma vitamin C levels could significantly reduce the risk of AD proxy phenotype in the maternal AD group (OR = 0.89, 95% CI 0.84–0.94, P = 7.29E−05), but not in the paternal AD group (OR = 1.02, 95% CI 0.92–1.12, P = 7.59E−01). The leave-one-out permutation further showed that the SLC23A1 rs33972313 variant largely changed the precision of the overall MR estimates in all these four GWAS datasets. Meanwhile, we did not observe any significant causal effect of plasma vitamin C levels on the cognitive performance. We demonstrated that there may be no causal association between plasma vitamin C levels and the risk of AD in people of European descent. The insistent findings in clinically diagnosed AD and AD proxy phenotype may be caused by the phenotypic heterogeneity.