Genetic Evidence for the Causal Relationship Between Gut Microbiota and Diabetic Kidney Disease: A Bidirectional, Two-Sample Mendelian Randomisation Study

Abstract

Aims: According to the gut-kidney axis theory, gut microbiota (GM) has bidirectional crosstalk with the development of diabetic kidney disease (DKD). However, empirical results have been inconsistent, and the causal associations remain unclear. This study was aimed at exploring the causal relationship between GM and DKD as well as the glomerular filtration rate (GFR) and urinary albumin-to-creatinine ratio (UACR). Materials and Methods: Two-sample Mendelian randomisation (MR) analysis was performed with inverse-variance weighting as the primary method, together with four additional modes (MR-Egger regression, simple mode, weighted mode, and weighted median). We utilised summary-level genome-wide association study statistics from public databases for this MR analysis. Genetic associations with DKD were downloaded from the IEU Open GWAS project or CKDGen consortium, and associations with GM (196 taxa from five levels) were downloaded from the MiBioGen repository. Results: In forward MR analysis, we identified 13 taxa associated with DKD, most of which were duplicated in Type 2 diabetes with renal complications but not in Type 1 diabetes. We observed a causal association between genetic signature contributing to the relative abundance of Erysipelotrichaceae UCG003 and that for both DKD and GFR. Similarly, host genetic signature defining the abundance of Ruminococcaceae UCG014 was found to be simultaneously associated with DKD and UACR. In reverse MR analysis, the abundance of 14 other GM taxa was affected by DKD, including the phylum Proteobacteria, which remained significant after false discovery rate correction. Sensitivity analyses revealed no evidence of outliers, heterogeneity, or horizontal pleiotropy. Conclusion: Our findings provide compelling causal genetic evidence for the bidirectional crosstalk between specific GM taxa and DKD development, contributing valuable insights for a comprehensive understanding of the pathological mechanisms of DKD and highlighting the possibility of prevention and management of DKD by targeting GM.

Keywords: Mendelian randomisation; diabetes; diabetic kidney disease; gut; gut microbiota; kidney axis.

Figure 1

Flowchart of the bidirectional MR study to explore the causal crosstalk between GM and diabetic nephropathy development. BMI: body mass index; DKD: diabetic kidney disease; GFR: glomerular filtration rate; GM: gut microbiota; GWAS: genome-wide association study; MR: Mendelian randomisation; SNPs: single-nucleotide polymorphisms; UACR: urinary albumin-to-creatinine ratio.

Figure 2

MR-estimated causal effect of gut microbiota and diabetic kidney disease development using the IVW statistical model. IVW: inverse-variance weighted; Nsnp: number of single-nucleotide polymorphisms selected as instrumental variables.

Figure 3

MR-estimated causal effect of diabetic kidney disease on the gut microbiota using the IVW statistical model. IVW: inverse-variance weighted; Nsnp: number of single-nucleotide polymorphisms selected as instrumental variables.