Comprehensive analysis of lncRNA-miRNA-mRNA networks during osteogenic differentiation of bone marrow mesenchymal stem cells

Abstract

Abstract Background Long non-coding RNA (lncRNA) plays crucial role in osteogenic differentiation of bone marrow mesenchymal stem cells (BMMSCs), involving in regulation of competing endogenous RNA (ceRNA) mechanisms and conduction of signaling pathways. However, its mechanisms are poorly understood. This study aimed to investigate lncRNAs, miRNAs and mRNAs expression profiles in rat BMMSCs (rBMMSCs) osteogenic differentiation, screen the potential key lncRNA-miRNA-mRNA networks, explore the putative functions and identify the key molecules, as the basis of studying potential mechanism of rBMMSCs osteogenic differentiation driven by lncRNA, providing molecular targets for the management of bone defect. Methods High-throughput RNA sequencing (RNA-seq) was used to determine lncRNAs, miRNAs, and mRNAs expression profiles at 14-day rBMMSCs osteogenesis. The pivotal lncRNA-miRNA and miRNA-mRNA networks were predicted from sequencing data and bioinformatic analysis, and the results were exported by Cytoscape 3.9.0 software. Gene Ontology (GO) analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis were used for functional exploration. Real-time quantitative reverse transcription-polymerase chain reaction (qRT-PCR) was performed to validate lncRNAs, miRNAs and mRNAs. Results rBMMSCs were identified, and the osteogenic and adipogenic differentiation ability were detected. A total of 8634 lncRNAs were detected by RNA-seq, and 1524 differential expressed lncRNAs, of which 812 up-regulated and 712 down-regulated in osteo-inductive groups compared with control groups. 30 up-regulated and 61 down-regulated miRNAs, 91 miRNAs were differentially expressed in total. 2453 differentially expressed mRNAs including 1272 up-expressed and 1181 down-expressed were detected. 10 up-regulated lncRNAs were chosen to predict 21 down-regulated miRNAs and 650 up-regulated mRNAs. 49 lncRNA-miRNA and 1515 miRNA–mRNA interactive networks were constructed. GO analysis showed the most important enrichment in cell component and molecular function were “cytoplasm” and “protein binding”, respectively. Biological process related to osteogenic differentiation such as “cell proliferation”, “wound healing”, “cell migration”, “osteoblast differentiation”, “extracellular matrix organization” and “response to hypoxia” were enriched. KEGG analysis showed differentially expressed genes were mainly enriched in “PI3K-Akt signaling pathway”, “Signaling pathway regulating pluripotency of stem cells”, “cGMP-PKG signaling pathway”, “Axon guidance” and “Calcium signaling pathway”. qRT-PCR verified that lncRNA Tug1, lncRNA AABR07011996.1, rno-miR-93-5p, rno-miR-322-5p, Sgk1 and Fzd4 were consistent with the sequencing results, and 4 lncRNA-miRNA-mRNA networks based on validations were constructed, and enrichment pathways were closely related to “PI3K-Akt signaling pathway”, “Signaling pathway regulating pluripotency of stem cells” and “Wnt signaling pathway”. Conclusions lncRNAs, miRNAs and mRNAs expression profiles provide clues for future studies on their roles for BMMSCs osteogenic differentiation. Furthermore, lncRNA–miRNA–mRNA networks give more information on potential new mechanisms and targets for management on bone defect.

Keywords

• Bone marrow mesenchymal stem cells
• Expression profile
• High-throughput RNA sequencing
• Long noncoding RNA
• Osteogenic differentiation