Analysis of the potential mechanism of stem cells in the treatment of diabetic nephropathy based on 16S, metabolome and transcriptome

Abstract

Background

Diabetic nephropathy (DN) has become a major cause of end-stage renal failure. The therapeutic mechanism of mesenchymal stem cells (MSCs) in DN is not fully understood.

In this study, we used transcriptome sequencing, 16S rRNA sequencing, and metabolomics sequencing to perform a combined multi-omics analysis to investigate the potential mechanisms of MSCs for DN.

Methods

First, DN mouse model was established. Kidneys, feces, and blood were collected from 6 control, 6 model, and 6 intervention (MSCs) groups for transcriptome sequencing, 16S RNA sequencing, and metabolome sequencing, respectively. Then, candidate genes between the 3 groups were identified and enriched using transcriptomic analysis. Next, with the help of metabolomics analysis, differential metabolites were screened by OPLS-DA analysis for control and model groups, as well as model and MSCs groups, respectively. Similarly, differential microorganisms and candidate microorganisms were selected by 16S rRNA gene sequencing data. Subsequently, the correlations between candidate genes and candidate metabolites, candidate genes and candidate microorganisms, as well as candidate metabolites and candidate microorganisms were explored by Spearman correlation analysis, respectively. Finally, a microbe-metabolite-gene network was constructed to identify key genes, key metabolites and key microbes, and their expression levels were analyzed.

Results

There were differences in genes, microorganisms, and metabolites among the samples in the control, model, and MSCs groups. Candidate genes enriched in the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways included adhesion molecules and 2-oxocarboxylic acid metabolism. GDP-mannose biosynthesis and purine ribonucleoside degradation were significantly enriched by different microorganisms. The KEGG pathways mainly enriched for differential metabolites were PPAR signaling pathway, arachidonic acid metabolism, and Rap1 signaling pathway. A microorganisms-metabolite-gene network containing 25 nodes and 53 edges was constructed with interactions including Sorangium-neg-M501T271 and Tmem238l-pos-M373T270, among others. In addition, 10 key genes, 5 key microorganisms and 10 key metabolites were significantly expressed in both the MSCs group and the control group.

Conclusion

This study identified 10 key genes, 10 key metabolites and 5 key microorganisms and a correlation network diagram was constructed. It provided a theoretical reference for exploring the molecular mechanisms of MSCs for DN treatment.