Causal Relationship Between Gut Microbiome and Infectious Mononucleosis: Bidirectional Mendelian Randomization Reveals Infectious Mononucleosis-Driven Gut Dysbiosis

Background To assess bidirectionality between gut microbiota and Epstein-Barr virus (EBV)-driven infectious mononucleosis (IM), we conducted two-sample Mendelian randomization (MR). Given IM’s heterogeneous symptoms and evidence linking microbiota to viral infection, this may inform novel prevention or treatment strategies. Methods We employed a bidirectional two-sample MR framework using summary data from 207 gut microbial taxa and 205 metabolic pathways (Dutch Microbiome Project, n = 7,738) and the FinnGen consortium. Causality was assessed via inverse variance weighting (IVW), MR-Egger, weighted median, and mode-based estimation. Statistical significance was set at P <0.05 (Bonferroni-corrected), with instruments validated by F-statistics >10. Heterogeneity and pleiotropy were evaluated using Cochran’s Q, MR-Egger intercepts, and leave-one-out analyses. Results Forward MR: Several bacterial pathways and taxa were positively associated with IM risk. These include the de novo purine nucleotide biosynthesis II superpathway (odds ratio (OR) = 1.246, 95% confidence interval (CI): 1.026-1.514, P = 0.027), anhydromuropeptide recycling (OR = 1.24, 95% CI: 1.014-1.517, P = 0.036), the superpathway of unsaturated fatty acid biosynthesis (Escherichia coli) (OR = 1.194, 95% CI: 1.019-1.4, P = 0.028), Lactobacillaceae (OR = 1.109, 95% CI: 1.02-1.206, P = 0.016), and Lactobacillus (OR = 1.108, 95% CI: 1.017-1.207, P = 0.019). Conversely, several taxa and pathways exhibited protective effects. These include the glucose-1-phosphate degradation pathway (G1P-DP) (OR = 0.852, 95% CI: 0.731-0.994, P = 0.042), heme biosynthesis from glutamate (OR = 0.809, 95% CI: 0.676-0.969, P = 0.032), the superpathway of L-tyrosine biosynthesis (OR = 0.9, 95% CI: 0.811-0.998, P = 0.046), the flavin biosynthesis I pathway (OR = 0.817, 95% CI: 0.672-0.993, P = 0.042), Streptococcaceae (OR = 0.869, 95% CI: 0.779-0.968, P = 0.011), and Streptococcus (OR = 0.844, 95% CI: 0.731-0.937, P = 0.020). Reverse MR: IM was found to causally alter gut microbiome composition. IM was associated with a decrease of beneficial genera such as Roseburia (OR = 0.901, 95% CI: 0.837-0.982, P = 0.016) and Bacteroides ovatus (OR = 0.915, 95% CI: 0.841-0.995, P = 0.038), as well as Streptococcus (OR = 0.876, 95% CI: 0.775-0.991, P = 0.035). Conversely, IM increased the risk of enrichment for Prevotellaceae (OR = 1.107, 95% CI: 1.016-1.210, P = 0.020) and Prevotella copri (OR = 1.096, 95% CI: 1.000-1.200, P = 0.048). Regarding metabolic pathways, IM increased the risk of polyamine biosynthesis II (OR = 1.123, 95% CI: 1.016-1.243, P = 0.024), L-lysine biosynthesis II (OR = 1.094, 95% CI: 1.008-1.118, P = 0.031), and L-lysine biosynthesis VI (OR = 1.083, 95% CI: 1.000-1.172, P = 0.048), while showing a protective association with L-rhamnose degradation I (OR = 0.922, 95% CI: 0.851-0.999, P = 0.046). Conclusion This study provides genetic evidence of a bidirectional causal relationship between the gut microbiome and IM. These findings suggest that IM may influence gut microbial ecosystem structure, characterized by a reduction in beneficial symbionts (e.g., Roseburia) and an enrichment of potentially pro-inflammatory taxa (e.g., P. copri). These findings may inform future microbiota-targeted interventions or risk stratification strategies for EBV-related diseases. Limitations include the European ancestry of study populations and the need for mechanistic validation.