Abstract
Fracture healing is a complex biological process. This study aimed to investigate the key molecules involved in fracture healing and their potential competing endogenous RNA (ceRNA) regulatory mechanisms within the first 28 days post-fracture using bioinformatics methods. The experiment was conducted on 15 adult male SD rats, with tibia callus tissue samples collected at days 0, 3, 7, 14, and 28 (n = 3) post-fracture. RNA-Seq was used for high-throughput transcriptome sequencing, followed by differential expression analysis to identify differentially expressed genes (DEGs), long non-coding RNA (DELs), and microRNA (DEMs) at different stages. Protein-protein interaction (PPI) networks were constructed using the STRING database and visualized with Cytoscape. GO and KEGG enrichment analyses were performed to explore potential biological mechanisms. miRNA-mRNA interactions were predicted using TargetScan, miRWalk, and miRDB, while RNA22 v2 was used for lncRNA-miRNA interactions. These interactions were integrated into ceRNA networks. Finally, qRT-PCR validated key molecules within the ceRNA network. We identified 4,997 DEGs, 315 DELs, and 89 DEMs at day 3; 5,087 DEGs, 300 DELs, and 84 DEMs at day 7; 3,073 DEGs, 235 DELs, and 68 DEMs at day 14; and 2,609 DEGs, 197 DELs, and 90 DEMs at day 28. Further analysis revealed hub osteogenic genes and their ceRNA regulatory networks at each time point. The networks consisted of 2 mRNAs, 3 miRNAs, and 9 lncRNAs at day 3; 2 mRNAs, 3 miRNAs, and 8 lncRNAs at days 7 and 14; and 1 mRNA, 3 miRNAs, and 10 lncRNAs at day 28. We validated two key lncRNAs (AABR07030366.1 and AABR07057997.1) along with their interacting miRNAs and mRNAs: rno-miR-9a-5p/Col9a1 (day 3), rno-miR-181c-5p/Comp (day 7), rno-miR-423-5p/Col1a1 (day 14), and rno-miR-185-5p/Ctsk (day 28). In summary, our study leveraged bioinformatics to construct ceRNA networks involved in osteogenesis post-fracture, offering insights into their dynamic regulatory role in healing and underlying molecular mechanisms.