Abstract
Epidemiological evidence suggests that cerebrospinal fluid (CSF) metabolites are associated with multiple sclerosis (MS). However, the causality remains unclear as a consequence of confounding or reverse causation. As a result, we evaluated the causal relationship between CSF metabolites and MS. A 2-sample Mendelian randomization analysis was performed for 338 CSF metabolites and MS, with causal relationship elucidated based on genome-wide association studies (GWAS) data. Summary statistics for 338 CSF metabolites were retrieved from a GWAS of 689 participants, and summary statistics for MS consisted of GWAS data including 115,803 participants (47,429 patients and 68,374 controls). The inverse-variance weighted method has been applied to perform primary analysis, and extensive sensitivity analyses were completed to guarantee robustness. The inverse-variance weighted analysis revealed a suggestive causal correlation between higher beta-alanine (odds ratio [OR] = 1.248, 95% confidence interval [CI] 1.092-1.427, P = 1.16 × 10-3), 1-stearoyl-2-arachidonoyl-gpc (18:0/20:4) (OR = 1.090, 95% CI 1.026-1.159, P = 5.48 × 10-3), allantoin (OR = 1.123, 95% CI 1.021-1.236, P = 1.68 × 10-2), and heightened susceptibility of MS, an increased risk of MS. By comparison, Wald ratio analysis indicated that N,N,N-trimethyl-l-alanyl-l-proline betaine (OR = 0.934, 95% CI 0.894-0.976, P = 2.57 × 10-3), butyrate (4:0) (OR = 0.908, 95% CI 0.840-0.982, P = 1.61 × 10-2), and N-acetylglycine (OR = 0.909, 95% CI 0.836-0.987, P = 2.29 × 10-2) were significantly connected to a reduced MS risk. In addition, sensitivity analyses her eliminated potential biases arising from heterogeneity and horizontal pleiotropy. We present indicative genetic data that support a causal correlation between CSF metabolites and MS, pointing out the possibility of intervening CSF metabolites to prevent MS.