Impact of the gut microbiota and associated metabolites on cardiometabolic traits, chronic diseases and human longevity: a Mendelian randomization study

Abstract

Features of the gut microbiota have been associated with several chronic diseases and longevity in preclinical models as well as in observational studies. Whether these relations underlie causal effects in humans remains to be established. We aimed to determine whether the gut microbiota influences cardiometabolic traits as well as the risk of chronic diseases and human longevity using a comprehensive 2-Sample Mendelian randomization approach. We included as exposures 10 gut-associated metabolites and pathways and 57 microbial taxa abundance. We included as outcomes nine cardiometabolic traits (fasting glucose, fasting insulin, systolic blood pressure, diastolic blood pressure, HDL cholesterol, LDL cholesterol, triglycerides, estimated glomerular filtration rate, body mass index [BMI]), eight chronic diseases previously linked with the gut microbiota in observational studies (Alzheimer's disease, depression, type 2 diabetes, non-alcoholic fatty liver disease, coronary artery disease (CAD), stroke, osteoporosis and chronic kidney disease), as well as parental lifespan and longevity. We found 7 associations with evidence of causality before and after sensitivity analyses, but not after multiple testing correction (1198 tests). Most effect sizes (4/7) were small. The two largest exposure-outcome effects were markedly attenuated towards the null upon inclusion of BMI or alcohol intake frequency in multivariable MR analyses. While finding robust genetic instruments for microbiota features is challenging hence potentially inflating type 2 errors, these results do not support a large causal impact of human gut microbita features on cardiometabolic traits, chronic diseases or longevity. These results also suggest that the previously documented associations between gut microbiota and human health outcomes may not always underly causal relations.

Fig. 1

Overview of the Mendelian randomization framework used to investigate the causal effect of gut microbiome features (blood and gut-derived metabolites and microbial taxa abundance) on cardiometabolic traits, chronic diseases and human longevity

Fig. 2

Balloon plot of the association of microbial fecal metabolites, microbial pathway and plasma metabolites with all 19 health outcomes. LDL cholesterol is included as positive control. Non-available (NA) associations stem from a lack of overlapping SNPs or proxies between exposure and outcome data resulting in fewer than three genetic instruments in the harmonized data set. Associations at P-value  > 0.05 are depicted with crosses. For readability, the effect of LDL on LDL was forced to be non-available

Fig. 3

Balloon plot of the association of gut microbial taxa abundance with all 19 health outcomes. LDL cholesterol is included as positive control. Non-available (NA) associations stem from a lack of overlapping SNPs or proxies between exposure and outcome data resulting in fewer than three genetic instruments in the harmonized data set. Associations at P-value  > 0.05 are depicted with crosses. For readability, the effect of LDL on LDL was forced to be non-available

Fig. 4

Forest plot of the associations that were consistent across robust MR analyses. Dichotomous traits are reported on a log(OR) scale. Continuous are reported on 1-SD scale. Dots depicts the point estimate. Horizontal bars depicts 95% confidence interval (CI)

Fig. 5

IVW-MR results before and after correcting for BMI and alcohol intake frequency using multivariable MR framework