Dissecting the causal effect between gut microbiota, DHA, and urate metabolism: A large-scale bidirectional Mendelian randomization

Abstract

Objectives: Our aim was to investigate the interactive causal effects between gut microbiota and host urate metabolism and explore the underlying mechanism using genetic methods.

Methods: We extracted summary statistics from the abundance of 211 microbiota taxa from the MiBioGen (N =18,340), 205 microbiota metabolism pathways from the Dutch Microbiome Project (N =7738), gout from the Global Biobank Meta-analysis Initiative (N =1,448,128), urate from CKDGen (N =288,649), and replication datasets from the Global Urate Genetics Consortium (N gout =69,374; N urate =110,347). We used linkage disequilibrium score regression and bidirectional Mendelian randomization (MR) to detect genetic causality between microbiota and gout/urate. Mediation MR and colocalization were performed to investigate potential mediators in the association between microbiota and urate metabolism.

Results: Two taxa had a common causal effect on both gout and urate, whereas the Victivallaceae family was replicable. Six taxa were commonly affected by both gout and urate, whereas the Ruminococcus gnavus group genus was replicable. Genetic correlation supported significant results in MR. Two microbiota metabolic pathways were commonly affected by gout and urate. Mediation analysis indicated that the Bifidobacteriales order and Bifidobacteriaceae family had protective effects on urate mediated by increasing docosahexaenoic acid. These two bacteria shared a common causal variant rs182549 with both docosahexaenoic acid and urate, which was located within MCM6/LCT locus.

Conclusions: Gut microbiota and host urate metabolism had a bidirectional causal association, implicating the critical role of host-microbiota crosstalk in hyperuricemic patients. Changes in gut microbiota can not only ameliorate host urate metabolism but also become a foreboding indicator of urate metabolic diseases.

Keywords: docosahexaenoic acids; gout; gut microbiome; mediation; mendelian randomization; uric acid.

Figure 1

Study design and flowchart. MR, Mendelian randomization; LDSC, Linkage disequilibrium score regression.

Figure 2

Dendrograms showing the taxonomy structure of the association between gut microbiota and host gout/urate level. Four dendrograms compared with the MR results of inferior and superior taxa according to the taxonomy structure. We included 203 taxa in the figures and excluded eight taxa named unknown. Taxa with significant MR results were listed at the bottom of the figure: the first column showed eight common taxa in both gout and urate level; the other four columns showed significant taxa in the respective figures. (A). MR results from microbiota to gout; (B). MR results from microbiota to urate level; (C). MR results from gout to microbiota; (D). MR results from urate level to microbiota.

Figure 3

Heatmap showing bidirectional causality of 35 families and 131 genera. Four heatmaps comparing the results of gout with the urate level of each taxon in 35 families and 131 genera. The outer ring represents urate level and the inner ring represents gout. (A). 35 microbiota families on gout and urate level; (B). 131 microbiota genera on gout and urate level; (C). Gout and urate level on 35 microbiota families; (D). Gout and urate level on 131 microbiota genera. * means significant results.

Figure 4

Bidirectional MR suggestive causal association between microbiota metabolism pathways and gout/urate level. (A). Causal effect of microbiota metabolism pathways on gout and urate level. (B). Causal effect of gout and urate level on microbiota metabolism pathways. We included “suggestive” significant pathways in Figure 4 (P <0.05 in gout or urate level and P <0.1 with the same direction in the other phenotype).