Multi-Omics Analysis to Examine Gene Expression and Metabolites From Multisite Adipose-Derived Mesenchymal Stem Cells

• Main Authors: Chuanxi Yang, Jing Zhang, Tingting Wu, Kun Zhao, Xiaoguang Wu, Jing Shi, Wei Sun, Xiangqing Kong
• Format: Article
• Language: English
• Published: Frontiers Media S.A. 2021-02-01
• Series: Frontiers in Genetics
• Subjects: adipose-derived mesenchymal stem cells; gene expression; metabolites; RNA-seq; metabolomics
• Online Access: https://www.frontiersin.org/articles/10.3389/fgene.2021.627347/full

This study aimed at exploring the gene expression and metabolites among multisite adipose-derived mesenchymal stem cells (ASCs) and investigate the metabolic pathway using a multi-omics analysis. Subcutaneous adipose-derived mesenchymal stem cells (SASCs), perirenal adipose-derived mesenchymal stem cells (PASCs), and epididymal adipose-derived mesenchymal stem cells (EASCs) were isolated from Sprague Dawley rats. RNA and metabolites were extracted and sequenced using transcriptomics and metabolomics analyses, respectively. There were 720 differentially expressed genes (DEGs) in EASCs and 688 DEGs in PASCs compared with SASCs; there were 166 unique DEGs in EASCs, 134 unique DEGs in PASCs, and 554 common DEGs between EASCs and PASCs. Furthermore, there were 226 differential metabolites in EASCs, 255 differential metabolites in PASCs, 83 unique differential metabolites in EASCs, 112 unique differential metabolites in PASCs, and 143 common differential metabolites between EASCs and PASCs. The transcriptomics and metabolomics analyses identified four hub genes, one in EASCs and three in PASCs. There are functional differences among multisite ASCs that may be related to the hub genes Atac2, Rrm1, Rrm2, and Gla. The relevant signaling pathways are the Ras signaling pathway, HIF-1 signaling pathway, and the p53 signaling pathway. In conclusion, compared with SASCs, our multi-omics analysis identified that EASCs with higher Acat2 expression may be more correlated to fat metabolism and insulin resistance, while PASCs with abnormal expression of Rrm1/2 and Gla may be more correlated with some malignant tumors and cardiac-cerebral vascular disease.