Gut microbiota causally affects ulcerative colitis by potential mediation of plasma metabolites: A Mendelian randomization study

Abstract

Ulcerative colitis (UC) is a chronic inflammatory bowel disease with a multifactorial etiology, including genetic, immunological, and environmental factors, as well as alterations in the gut microbiome and plasma metabolites. The interplay between these factors is complex and not fully elucidated, particularly regarding the potential mediation of metabolites in the relationship between gut microbiota and UC. We performed a Mendelian randomization (MR) study to investigate the causal associations between gut microbiota, plasma metabolites, and UC. The study utilized a two-sample MR approach to discern causal relationships among these factors. Genetic variants from genome-wide association studies served as instrumental variables in the MR analyses, conducted using the "TwoSampleMR" package in R software. We adhered to the fundamental assumptions of MR analyses, ensuring the validity of our causal inferences. Additionally, we incorporated a mediation analysis to assess the potential mediating role of plasma metabolites in the relationship between gut microbiota and UC. Our current study found the substantial relationship between certain gut microbial taxa and the development of UC. Indeed, we have identified 6 microbial taxa, including Genus Dorea, Phylum Proteobacteria, Species Streptococcus parasanguinis, Species Ruminococcus obeum, Species Roseburia intestinalis, and Order Lactobacillales, which were found to be causally related to UC. Seventy-three metabolites and metabolite ratios of were also causally associated with UC, and a mediation analysis revealed that metabolites such as stearoylcarnitine, 3-hydroxyoctanoylcarnitine, 1-arachidonoyl-GPE (20:4n6), 3-(3-hydroxyphenyl)propionate sulfate, and thioproline mediated the effects of gut microbiota on UC and hence might play roles in disease pathogenesis. This microbiota-UC-specific MR study provides evidence for causal associations between specific gut microbiota and UC, potentially mediated through plasma metabolites. The findings give new perspectives on the causal nexus of the gut microbiota and plasma metabolites with UC, highlighting potential intervention targets for the disease. These findings call for confirmation in further research, together with investigation of the underlying mechanisms.

Keywords: Mendelian randomization (MR); causal relationship; gut microbiota; plasma metabolites; ulcerative colitis (UC).

Figure 1.

The study design. A two-step MR study of gut microbiota on ulcerative colitis mediated by plasma metabolites. MR = Mendelian randomization.

Figure 2.

Forest plots of causal relationship of gut microbiota and UC. CI = confidence interval; OR = odds ratio; nSNP = number of single nucleotide polymorphism; Pval = P-values, UC = ulcerative colitis.

Figure 3.

Forest plots of causal relationship of metabolites and metabolite ratios and UC. CI = confidence interval; OR = odds ratio; nSNP = number of single nucleotide polymorphism; Pval = P-values, UC = ulcerative colitis.

Figure 4.

Enrichment analysis results of the causal effect of plasma metabolites on UC. (A) Using Small Molecule Pathway Database (SMPDB); (B) using the Kyoto Encyclopedia of Genes and Genomes (KEGG) database. UC = ulcerative colitis.

Figure 5.

Forest plots of causal relationship of gut microbiota and metabolites and metabolite ratios. CI = confidence interval; OR = odds ratio; nSNP = number of single nucleotide polymorphism; Pval = P-values.

Figure 6.

Mediation effects of the selected human plasma metabolites between gut microbiota and UC. (A) Mediation effect of Stearoylcarnitine levels between Genus Dorea and UC. (B) Mediation effect of 3-hydroxyoctanoylcarnitine levels between Genus Dorea and UC. (C) Mediation effect of 3-(3-hydroxyphenyl)propionate sulfate levels between Species Ruminococcus obeum and UC. (D) Mediation effect of Thioproline levels between Species Ruminococcus obeum and UC. (E) Mediation effect of Stearoylcarnitine levels between Species Roseburia intestinalis and UC. (F) Mediation effect of Glyco-beta-muricholate levels between Species Roseburia intestinalis and UC. (G) Mediation effect of 1-arachidonoyl-GPE (20:4n6) levels between Phylum Proteobacteria and UC. UC = ulcerative colitis.