脂肪干细胞成骨分化的转录因子-miRNA-mRNA网络分析

Transcription factor-miRNA-mRNA network analysis of osteogenic differentiation of adipose-derived stem cells

背景:脂肪干细胞来源广泛、易于获取,人脂肪组织来源的干细胞进入分化骨细胞涉及基因表达的变化主要受miRNA和转录因子调节,但调节脂肪干细胞成骨分化的转录因子-miRNA-mRNA调控网络尚未建立。目的:筛选脂肪干细胞成骨分化过程中的差异表达基因、差异表达miRNA及转录因子,探讨脂肪干细胞成骨分化过程中的转录因子-miRNA-mRNA调控网络及其潜在的靶基因、miRNA和转录因子。方法:从GEO数据库获取3张芯片(GSE37329,GSE63754 mRNA数据芯片2张,GSE72429 miRNA数据芯片1张),筛选差异表达基因和差异表达miRNA;差异表达基因进行PPI网络、GO功能和KEGG信号通路分析;随后差异表达基因预测上游miRNA,差异表达miRNA预测转录因子,通过Cytoscape3.7.1软件构建转录因子-miRNA-mRNA网络图。结果与结论:筛选出86个差异表达基因(26个上调,60个下调)、16个差异表达miRNA(10个上调,6个下调);PPI网络中蛋白互作共表达的占76.52%,协同定位的占8.69%;GO功能富集分析主要在对经典Wnt信号通路的负调控、软骨发育的负调控等功能相关,KEGG信号通路涉及酪氨酸代谢、脂肪酸降解等信号通路;差异表达基因预测到11381个miRNA及差异miRNA预测到55个转录因子,构建转录因子-miRNAmRNA调控网络。脂肪干细胞成骨分化过程中的关键基因可能是PODXL、SEMA3D、ADGRG6、LGR4、LPL、CADM3、GRIA1、GPM6B、RERG、APCDD1和NRCAM,重要的miRNA有hsa-miR-502-3p、hsa-miR-762和hsa-miR-1275,核心转录因子是CTCF、TAL1、STAT2、STAT1和TCF3。结果显示,通过生物信息学技术和数据库挖掘等方法构建的转录因子-miRNA-mRNA调控网络,为脂肪干细胞成骨分化的深入研究提供潜在的治疗靶标。

BACKGROUND:Adipose stem cells have a wide range of sources and are easy to obtain.The changes in gene expression involved in the entry of human adipose tissue-derived stem cells into differentiated bone cells are mainly regulated by miRNA and transcription factor,but the specific molecular mechanisms that regulate the osteogenic differentiation of adipose stem cells and their potential transcription factor-miRNA-mRNA regulatory network have not yet been established.OBJECTIVE:To screen the differentially expressed genes,differentially expressed miRNAs and transcription factors during the osteogenic differentiation of adipose-derived stem cells,and to explore the potential target genes,miRNAs,transcription factors,and transcription factor-miRNA-mRNA regulatory network during the osteogenic differentiation of adipose-derived stem cells.METHODS:GEO database was used to acquire three chips(GSE37329,GSE63754 two miRNA gene chips and GSE72429 one miRNA data chip)to screen differentially expressed genes and differential miRNAs.Differentially expressed genes were analyzed for PPI network,GO function,and KEGG signaling pathway.Differentially expressed genes were utilized to predict upstream miRNAs and differential miRNAs pre-transcription factors.Transcription factor-miRNA-mRNA network diagrams were constructed through Cytoscape 3.7.1 software.RESULTS AND CONCLUSION:Eighty-six differentially expressed genes(26 up-regulated genes,60 down-regulated genes)and 16 differential miRNAs(10 up-regulated miRNAs,6 down-regulated miRNAs)were screened out.There were 76.52%of protein interactions and co-expression and 8.69%of synergistic localization in the PPI network.GO function enrichment analysis was mainly related to the negative regulation of the classical Wnt signaling pathway and the negative regulation of cartilage development.The KEGG signaling pathway involved signaling pathways,such as tyrosine metabolism and fatty acid degradation.Differentially expressed genes predicted 11381 miRNAs and differential miRNAs predicted 55 transcription factors.Transcription factor-miRNA-mRNA regulatory network was constructed.The key genes in the osteogenic differentiation of adipose-derived stem cells may be PODXL,SEMA3D,ADGRG6,LGR4,LPL,CADM3,GRIA1,GPM6B,RERG,APCDD1,and NRCAM.Important miRNAs were:hsa-miR-762,hsa-miR-502-3p,and hsa-miR-1275.The core transcription factors were CTCF,TAL1,STAT2,STAT1,and TCF3.The results show that the transcription factor-miRNA-mRNA regulatory network constructed through bioinformatics technology and database mining methods provides potential therapeutic targets for the in-depth study of osteogenic differentiation of adipose-derived stem cells.