Dissecting Causal Relationships Between Gut Microbiota, Blood Metabolites, and Stroke: A Mendelian Randomization Study

Qi Wang; Huajie Dai; Tianzhichao Hou; Yanan Hou; Tiange Wang; Hong Lin; Zhiyun Zhao; Mian Li; Ruizhi Zheng; Shuangyuan Wang; Jieli Lu; Yu Xu; Ruixin Liu; Guang Ning; Weiqing Wang; Yufang Bi; Jie Zheng; Min Xu

doi:10.5853/jos.2023.00381

Dissecting Causal Relationships Between Gut Microbiota, Blood Metabolites, and Stroke: A Mendelian Randomization Study

J Stroke. 2023 Sep;25(3):350-360. doi: 10.5853/jos.2023.00381. Epub 2023 Sep 26.

Authors

Qi Wang^{1

2}, Huajie Dai^{1

2}, Tianzhichao Hou^{1

2}, Yanan Hou^{1

2}, Tiange Wang^{1

2}, Hong Lin^{1

2}, Zhiyun Zhao^{1

2}, Mian Li^{1

2}, Ruizhi Zheng^{1

2}, Shuangyuan Wang^{1

2}, Jieli Lu^{1

2}, Yu Xu^{1

2}, Ruixin Liu^{1

2}, Guang Ning^{1

2}, Weiqing Wang^{1

2}, Yufang Bi^{1

2}, Jie Zheng^{1

2

3}, Min Xu^{1

2}

Affiliations

¹ Department of Endocrine and Metabolic Diseases, Shanghai Institute of Endocrine and Metabolic Diseases, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China.
² Shanghai National Clinical Research Center for Metabolic Diseases, Key Laboratory for Endocrine and Metabolic Diseases of the National Health Commission of the PR China, Shanghai Key Laboratory for Endocrine Tumor, State Key Laboratory of Medical Genomics, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China.
³ MRC Integrative Epidemiology Unit, Bristol Medical School, University of Bristol, Bristol, UK.

Abstract

Background and purpose: We investigated the causal relationships between the gut microbiota (GM), stroke, and potential metabolite mediators using Mendelian randomization (MR).

Methods: We leveraged the summary statistics of GM (n=18,340 in the MiBioGen consortium), blood metabolites (n=115,078 in the UK Biobank), and stroke (cases n=60,176 and controls n=1,310,725 in the Global Biobank Meta-Analysis Initiative) from the largest genome-wide association studies to date. We performed bidirectional MR analyses to explore the causal relationships between the GM and stroke, and two mediation analyses, two-step MR and multivariable MR, to discover potential mediating metabolites.

Results: Ten taxa were causally associated with stroke, and stroke led to changes in 27 taxa. In the two-step MR, Bifidobacteriales order, Bifidobacteriaceae family, Desulfovibrio genus, apolipoprotein A1 (ApoA1), phospholipids in high-density lipoprotein (HDL_PL), and the ratio of apolipoprotein B to ApoA1 (ApoB/ApoA1) were causally associated with stroke (all P<0.044). The causal associations between Bifidobacteriales order, Bifidobacteriaceae family and stroke were validated using the weighted median method in an independent cohort. The three GM taxa were all positively associated with ApoA1 and HDL_PL, whereas Desulfovibrio genus was negatively associated with ApoB/ApoA1 (all P<0.010). Additionally, the causal associations between the three GM taxa and ApoA1 remained significant after correcting for the false discovery rate (all q-values <0.027). Multivariable MR showed that the associations between Bifidobacteriales order, Bifidobacteriaceae family and stroke were mediated by ApoA1 and HDL_PL, each accounting for 6.5% (P=0.028) and 4.6% (P=0.033); the association between Desulfovibrio genus and stroke was mediated by ApoA1, HDL_PL, and ApoB/ApoA1, with mediated proportions of 7.6% (P=0.019), 4.2% (P=0.035), and 9.1% (P=0.013), respectively.

Conclusion: The current MR study provides evidence supporting the causal relationships between several specific GM taxa and stroke and potential mediating metabolites.

Keywords: Gastrointestinal microbiome; Mendelian randomization analysis; Metabolomics; Stroke.

Abstract

Grants and funding