Mendelian-randomization study reveals causal relationships between nitrogen dioxide and gut microbiota

Abstract

Exposure to nitrogen dioxide might potentially change the makeup and operation of gut microbes. Nitrogen dioxide data was procured from the IEU Open GWAS (N = 456 380). Subsequently, a two-sample Mendelian randomization study was executed, utilizing summary statistics of gut microbiota sourced from the most expansive available genome-wide association study meta-analysis, conducted by the MiBioGen consortium (N = 13 266). The causal relationship between nitrogen dioxide and gut microbiota was determined using inverse variance weighted, maximum likelihood, MR-Egger, Weighted Median, Weighted Model, Mendelian randomization pleiotropy residual sum and outlier, and constrained maximum likelihood and model averaging and Bayesian information criterion. The level of heterogeneity of instrumental variables was quantified by utilizing Cochran's Q statistic. The colocalization analysis was used to examine whether nitrogen dioxide and the identified gut microbiota shared casual variants. Inverse variance weighted estimate suggested that nitrogen dioxide was causally associated with Akkermansia (β = -1.088, 95% CI: -1.909 to -0.267, P = 0.009). In addition, nitrogen dioxide presented a potential association with Bacteroides (β = -0.938, 95% CI: -1.592 to -0.284, P = 0.005), Barnesiella (β = -0.797, 95% CI: -1.538 to -0.055, P = 0.035), Coprococcus 3 (β = 1.108, 95% CI: 0.048-2.167, P = 0.040), Eubacterium hallii group (E. hallii) (β = 0.776, 95% CI: 0.090-1.463, P = 0.027), Holdemania (β = -1.354, 95% CI: -2.336 to -0.372, P = 0.007), Howardella (β = 1.698, 95% CI: 0.257-3.139, P = 0.021), Olsenella (β = 1.599, 95% CI: 0.151-3.048, P = 0.030) and Sellimonas (β = -1.647, 95% CI: -3.209 to -0.086, P = 0.039). No significant heterogeneity of instrumental variables or horizontal pleiotropy was found. The associations of nitrogen dioxide with Akkermansia (P_H4 = 0.836) and E. hallii (P_H4 = 0.816) were supported by colocalization analysis. This two-sample Mendelian randomization study found that increased exposure to nitrogen dioxide had the potential to impact the human gut microbiota.

Keywords: Causal inference; Gut microbiota; Mendelian randomization study; Nitrogen dioxide.