Study of Retinoic Acid-Induced Osteoarthritis: Integrating RNA-sequencing, Network Pharmacology, Molecular Docking, and Experimental Validation

Osteoarthritis (OA) is a debilitating joint disorder characterized by cartilage degradation and chondrocyte homeostasis disruption. Although retinoic acid (RA) has been applied in the model preparation of OA, its target of action and signaling pathway are not clear. Leveraging a translational framework that integrates RNA-sequencing transcriptomics, network pharmacology prediction, computational ligand-receptor molecular docking, and biological experimental validation, this study systematically deciphers RA's disease-modifying targets and associated signaling circuitry in OA pathogenesis. RNA-sequencing of RA-treated chondrocytes revealed 656 differentially expressed genes (DEGs). Protein-protein interaction (PPI) network analysis and functional enrichment (GO/KEGG) highlighted key pathways, including extracellular matrix reorganization and PI3K-Akt-mediated mechanotransduction and others. Additionally, molecular docking results ranked binding affinities in descending order as follows: ACAN > SFRP2 > KANSL2 (NSL2), and so on. Network pharmacology identified 42 RA-OA shared targets, with protein-protein interaction (PPI) analysis and functionally enriched (GO/KEGG) including the Renin-angiotensin system, Neuroactive ligand-receptor interaction, and others. The molecular docking results showed a total score in descending order of MAPK14(p38α), PTGER3(PGE2-R), CA2(CACNA2D2), and so on. Five intersecting targets (CA2, ACE, PTGS1(COX-1), PGR, and EDNRA(ETAR) ) from RNA-sequencing and network pharmacology were obtained, and molecular docking was used to dock the intersecting targets to RA. The docking results demonstrated strong binding affinities, which were further validated through western blot and RT-qPCR experiments, confirming the RA-induced upregulation of the intersecting targets. These findings were accompanied by immunofluorescent evidence showing elevated levels of MMP13 and suppressed expression of COL2A1, collectively reflecting the phenotypic characteristics of OA. Our findings position CA2 and ACE as key hubs for multi-targeting. This study not only reveals possible mechanistic studies of RA in OA but also provides a drug reference for the development of drugs against OA. In our experiments, we chose 24-hour-old specific pathogen-free (SPF) female Sprague-Dawley rats provided by Shanghai XipurBiak Laboratory Animal Co. Ltd (located in Shanghai, China.) with the laboratory animal license number SCXK(Shanghai)2023-0009 as experimental animals, the quality certificate number of this batch of animals is NO.20230009011756. Following the method described by Wang et al. 0.1% type II collagenase (BS164, Biosharp) was to digest the cartilage tissues at 37°C for a continuous period of 2-3 h. Subsequently, chondrocytes were cultured in Dulbecco's Modified Eagle's Medium (DMEM, Hyclone, UT) containing 10% fetal bovine serum (FB-1058, Biosera) and 1% penicillin-streptomycin(ST488S, Beyotime). The chondrocytes used in this experiment were all at passage 1 or passage 2 after isolation. Before RNA sequencing, control cells were cultured in a high-glucose DMEM medium containing 10% FBS for 24 h, while cells in the RA group were cultured in the same medium with 1 μM retinoic acid(S1653, Bimake) for 24 h. The retinoic acid isomer used in this experiment was all-trans retinoic acid (ATRA), purchased from Selleckchem (Item No. NSC 122758). In the RNA-sequencing (RNA-Seq), RNA was extracted from three independent samples for each group and sequenced by the Illumina HiSeq 4000 sequencing platform from Shanghai Life Genetics Technology Co. RNA quality was assessed by the RNA Nano 6000 assay kit(R0027, Beyotime) from Agilent Technologies and the Bioanalyzer 2100 instrument for assessment. Subsequently, sequencing libraries were constructed using NEB's NEBNext® Ultra™ RNA Library Preparation Kit and fragment purification was performed by the AMPure XP system to ensure the selection of fragments of appropriate length and quality. Differentially expressed genes (DEGs) were identified based on a P-value of less than 0.05, and the DEGs were functionally annotated using the Gene Ontology (GO) database. Fold change (FC) of gene expression between groups was calculated by specific formulas such as FC (RP/RS) =Log2[RP (FPKM)/RS (FPKM)] or FC (RP/RPT) = Log2[RP (FPKM)/RPT (FPKM)] to assess the effect of different treatment conditions on gene expression.