CAUSAL PATHWAYS LINKING GUT MICROBIOTA, METABOLITES, AND DIABETIC NEPHROPATHY: A MENDELIAN RANDOMIZATION STUDY

The specific pathogenesis of diabetic nephropathy had not yet been elucidated. Previous studies have shown that the gut microbiota and their metabolites played important roles in body’s metabolism and immune regulation which can be a new therapeutic target for improving diabetic nephropathy (DN). However, the specific mechanisms underlying the causal relationship between gut microbiota and diabetic nephropathy are not fully understood. This study was aimed to investigate whether the gut microbiota regulates the occurrence and progression of diabetic nephropathy by influencing metabolite levels through Mendelian randomization and mediation analysis.

An overview of the study design is shown in Figure 1.The Genome-wide association study (GWAS) data for gut microbiota were obtained from the MiBioGen Consortium, which included data from 18,340 individuals across 24 cohorts. The data for diabetic nephropathy were obtained from the FinnGen database, which was a statistical data mainly from European studies worldwide, including 3676 cases of diabetic nephropathy and 283,456 healthy controls, for a total of 10,485,575 SNPs. The metabolomics data, published in Nature in 2023 by J Brent Richards and his colleagues, encompassed a large-scale study of 8299 participants. This database included 1091 metabolites and 309 metabolite ratios. No new ethical applications were made for our MR analyses because no new data were added.The inverse variance weighted (IVW) model was employed as the primary analytical method, supplemented by various sensitivity analyses. Heterogeneity was assessed using Cochran's Q test, horizontal pleiotropy was examined using MR-Egger regression and MR-PRESSO global test.

The study conducted a MR using gut microbiota and diabetic nephropathy as exposure and outcome, respectively, and results revealed the screening of 15 gut microbiota with significant differences. Subsequently, using diabetic nephropathy as the exposure and 15 different gut microbiota as the outcome, a reverse MR was conducted, resulting in the identification of 14 disease-related gut microbiota(P < 0.05)（Figure 2）.Using metabolites and diabetic nephropathy as exposure and outcome, a forward Mendelian randomization was conducted, resulting in the identification of 61 metabolites with significant differences. After excluding 11 unnamed metabolites, a total of 50 metabolites with significant differences were ultimately retained. Subsequently, diabetic nephropathy was used as the exposure, with 50 different metabolites as the outcome for reverse MR, resulting in the identification of 46 disease-related metabolites.(P < 0.05)（Figure 3）. The 14 types of gut microbiota and 46 metabolites selected above were subjected to Mendelian randomization, resulting in the identification of 26 combinations of gut microbiota and metabolites with a causal effect. The mediation MR analysis showed that sphingomyelin mediated the causal effects of genus Parasutterella.id.2892 on diabetic nephropathy (proportion mediated = 13.4%, 95% CI = 0.439% ~ 26.4%, P value = 0.043)(Figure 4).

Our study clarified the causal relationship between Parasutterella, sphingomyelin and DN, and we first found that Parasutterella may be increase the risk of DN via influence the sphingomyelin level. This prompted us to determine how Parasutterella and sphingomyelin affects diabetic nephropathy. Previous studies demonstrated that Parasutterella and sphingomyelin play an important role in diseases via metabolism disorders, inflammation states or insulin receptor signaling. Therefore, we plan to perform a series of experiments to explore the mechanism of Parasutterella an sphingomyelin with the DN in the future.