Significant Association Between Abundance of Gut Microbiota and Plasma Levels of microRNAs in Individuals with Metabolic Syndrome and Their Potential as Biomarkers for Metabolic Syndrome: A Pilot Study.

Background/Objectives: The relationship between gut microbiota (GM) and microRNAs (miRs) related to lipid metabolism in individuals with metabolic syndrome (MetS) remains unclear. This pilot study examined the relationship between Bacteroidetes and Firmicutes abundance at the phylum level and the plasma levels of miR-122 and miR-370, both of which are associated with lipid metabolism, in Korean individuals with MetS and in healthy controls. We also evaluated the potential of these miRs as biomarkers for MetS. Methods: This study enrolled 7 individuals with MetS and 8 controls. The abundance of GM was analyzed by 16S rRNA amplicon sequencing. To evaluate the relationship between the dominant phyla in the 2 groups, the log ratio of Firmicutes to Bacteroidetes (F/B) was calculated using a centered log-ratio (CLR) transformation. The abundance of the 2 plasma miRs was also quantified by real-time quantitative PCR (RT-qPCR). Pearson's and Spearman's correlation analyses were then performed to evaluate the relationship between Bacteroidetes and Firmicutes abundance, the clinical parameters, and plasma levels of the 2 miRs. Additionally, the area under the curve (AUC) value of the receiver operating characteristic (ROC) curve was calculated to evaluate the potential of the 2 miRs as MetS biomarkers. Results: The 2 most abundant phyla were Bacteroidetes and Firmicutes. Bacteroidetes made up an average of 24.7% in the MetS group and 69.7% in the control group. Meanwhile, the average abundance of Firmicutes was 69.8% in the MetS group and 26.5% in the control group. The log F/B ratios in the MetS and control groups were 0.7 ± 0.5 and -0.4 ± 0.1 (p < 0.001), respectively. FDR analysis revealed significant correlations between Bacteroidetes abundance and BMI, DBP, FBG, total chol, insulin and HOMA-IR (FDR-adjusted p < 0.05), as well as between Firmicutes abundance and BMI, FBG, total chol, insulin and HOMA-IR (FDR-adjusted p < 0.05). Plasma levels of the 2 miRs differed significantly between the MetS and control groups: miR-122 (1.43 vs. 0.73; p = 0.0065) and miR-370 (1.39 vs. 0.83; p = 0.0089). The AUC values for miR-122 and miR-370 were 0.946 (p < 0.001) and 0.964 (p < 0.001), respectively. Pearson's and Spearman's correlation analyses revealed significant negative correlations between Bacteroidetes abundance and levels of miR-122 (p = 0.0048 and p = 0.0045, respectively) and miR-370 (p = 0.0003 and p < 0.0001, respectively), as well as significant positive correlations between Firmicutes abundance and levels of miR-122 (p = 0.0038 and p = 0.0027, respectively) and miR-370 (p = 0.0004 and p < 0.0001, respectively). However, as our exploratory findings were based on a small sample size, the high correlation results may partly reflect the separation between the MetS and control groups. Conclusions: Our exploratory findings suggest that the GM abundances of individuals with MetS may be significantly associated with plasma levels of miR-122 and miR-370, which are related to lipid metabolism. These miRs may therefore serve as potential MetS biomarkers.