The Causal Association Analysis between Depression and Cerebrospinal Fluid: From the Perspective of Mendelian Randomization

Yu Zhang,1,2,* Ping Wu,3,* Zhuo Liu1,4 1Hunan Provincial Hospital of Integrated Traditional Chinese and Western Medicine (The Affiliated Hospital of Hunan Academy of Traditional Chinese Medicine), Changsha, Hunan, 410006, People’s Republic of China; 2Integrated Traditional Chinese and Western Medicine College of Hunan University of Chinese Medicine, Changsha, Hunan, 410208, People’s Republic of China; 3Changde Hospital, Xiangya School of Medicine, Central South University (The First People’s Hospital of Changde City), Changde, Hunan, 415099, People’s Republic of China; 4College of Traditional Chinese Medicine, Changsha Medical University, Changsha, Hunan, 410219, People’s Republic of China*These authors contributed equally to this workCorrespondence: Zhuo Liu, Hunan Provincial Hospital of Integrated Traditional Chinese and Western Medicine (The Affiliated Hospital of Hunan Academy of Traditional Chinese Medicine, No. 58 Lushan Road, Yuelu District, Changsha City, Hunan Province, 412000, People’s Republic of China, Email [email protected]: Major depressive disorder (MDD) leads to significant distress and disruption across social, occupational, and other functional domains. Although cerebrospinal fluid (CSF) biomarkers have been identified as potential indicators and therapeutic targets for depression, their causal relationship with MDD remains unclear.Methods: We analyzed publicly available CSF metabolomics and genotype data, quantifying 338 distinct metabolites. Among these, 296 were chemically validated and classified into eight major metabolic groups, while 38 remained undefined. To assess the genetic association with depression, we used summary statistics from a GWAS (F5_DEPRESSIO dataset, including 53,313 diagnosed cases and 394,756 controls from Finland). An integrated approach combining Mendelian randomization (MR), inverse variance weighting (IVW), and linkage disequilibrium score regression (LDSC) was applied to explore the causal impact of CSF metabolites on depression risk.Results: Our analysis identified 62 metabolites significantly associated with depression (p < 0.05). Sensitivity tests revealed heterogeneity in five metabolites: 5-hydroxyindoleacetic acid, X-19438, ethylmalonic acid, γ-glutamylglutamine, and β-alanine. A focused analysis on 14 metabolites further supported a potential causal link with depression. LDSC confirmed significant genetic heritability for metabolites such as creatinine, arginine succinate, N-acetylisourea, 3-amino-2-piperidone, and carboxyethyl-GABA. Systematic leave-one-out analyses demonstrated that these associations are driven by multiple interacting SNPs rather than a single variant.Conclusion: This study provides novel insights into the potential causal relationship between CSF metabolites and depression, highlighting 14 key metabolites with significant associations. The robustness of these findings is supported by MR and sensitivity analyses. Further longitudinal studies are warranted to confirm the clinical relevance of these CSF biomarkers in MDD.Keywords: depression, cerebrospinal fluid, Mendelian randomization, LDSC, causal association