Causal Effects of Gut Microbiome on Systemic Lupus Erythematosus: A Two-Sample Mendelian Randomization Study

Abstract

The observational association between gut microbiome and systemic lupus erythematosus (SLE) has been well documented. However, whether the association is causal remains unclear. The present study used publicly available genome-wide association study (GWAS) summary data to perform two-sample Mendelian randomization (MR), aiming to examine the causal links between gut microbiome and SLE. Two sets of MR analyses were conducted. A group of single nucleotide polymorphisms (SNPs) that less than the genome-wide statistical significance threshold (5 × 10^-8) served as instrumental variables. To obtain a comprehensive conclusion, the other group where SNPs were smaller than the locus-wide significance level (1 × 10^-5) were selected as instrumental variables. Based on the locus-wide significance level, the results indicated that there were causal effects of gut microbiome components on SLE risk. The inverse variance weighted (IVW) method suggested that <i>Bacilli</i> and <i>Lactobacillales</i> were positively correlated with the risk of SLE and <i>Bacillales</i>, <i>Coprobacter</i> and <i>Lachnospira</i> were negatively correlated with SLE risk. The results of weighted median method supported that <i>Bacilli</i>, <i>Lactobacillales</i>, and <i>Eggerthella</i> were risk factors for SLE and <i>Bacillales</i> and <i>Coprobacter</i> served as protective factors for SLE. The estimates of MR Egger suggested that genetically predicted <i>Ruminiclostridium6</i> was negatively associated with SLE. Based on the genome-wide statistical significance threshold, the results showed that <i>Actinobacteria</i> might reduce the SLE risk. However, Mendelian randomization pleiotropy residual sum and outlier (MR-PRESSO) detected significant horizontal pleiotropy between the instrumental variables of <i>Ruminiclostridium6</i> and outcome. This study support that there are beneficial or detrimental causal effects of gut microbiome components on SLE risk.