The present study is the first to use bioinformatics analysis combined with subsequent experimental verification to investigate the mechanism of resveratrol in the treatment of osteoarthritis. Based on this strategy, we obtained 15 genes, which were the DEGs closely related to OA pathology and resveratrol treatment. The hub genes, such as IL6, PTGS2, HIF1A, CXCL1, MMP3, also the inflammatory response, extracellular matrix disassembly, TNF signaling pathway and HIF-1 signaling pathway may be involved in the biological activity in the resveratrol treatment process. After analysis of these 15 genes by network and functional enrichment analysis, six genes, including CXCL1, HIF1A, IL-6, MMP3, NOX4, and PTGS2, were chosen for experimental validation. We found that all six genes were significantly changed in human OA chondrocytes. Meanwhile, all six genes were markedly changed after resveratrol treatment, suggesting that they may serve as the crucial targets in the progression of OA and resveratrol treatment.
Chemokine (CXC motif) ligand 1 (CXCL1), also known as growth-regulated protein a (GROa), belongs to the ELR-family of chemokines and modulates angiogenesis, inflammation, tumorigenesis [20, 21]. Previous studies have reported that CXCL1 contributes to the ingress of neutrophils into rheumatoid arthritis (RA) joint , and during cartilage development, CXCL1 induces chondrocyte hypertrophy and apoptosis . In general, CXCL1 levels are low under normal physiological conditions but appear to be elevated in patients with RA and OA . In this study, we predicted and verified that the level of CXCL1 was significantly increased in the joint tissues of OA model mice and human OA chondrocytes, which is consistent with the previous studies. Moreover, resveratrol treatment can inhibit the expression of CXCL1, indicating that CXCL1 is the target of resveratrol treatment in patients with OA. According to this evidence, CXCL1 may play a pivotal role in the pathogenesis and treatment of OA.
Hypoxia-inducible factor 1 (HIF-1) belongs to the family of basic-helix–loop–helix-containing PAS domain transcription factors and consists of the subunit HIF-1α and HIF-1β, produced in response to hypoxia . HIF-1α is well known as one of the major regulators of the hypoxic response , and controls hypoxic expression of erythropoietin, as well as the expression of genes with metabolic functions , which is essential in tumor genesis, inflammation . Our previous study showed that HIF‑1α is up-regulated in bone tissue of rats treated with hind limb occlusion, which plays an essential role in fracture healing . It has been demonstrated that HIF-1α is expressed in OA articular cartilage  and plays a significant role in chondrocyte survival . HIF-1α could alleviate apoptosis and senescence via mitophagy in chondrocytes under hypoxia conditions, which could also ameliorate surgery-induced cartilage degradation in mice OA model . However, the association between HIF-1α and OA has not been sufficiently explored, and HIF-1α regulation in OA progression is poorly understood. In the present study, our results demonstrated that HIF-1α expression is increased in the joint tissues of OA model mice and human OA chondrocytes and can be enhanced by resveratrol treatment. These findings suggest that the resveratrol may stimulate the HIF-1α to promote the matrix accumulation and decrease degradation of human OA chondrocytes, indicated that HIF-1α may serve as a promising strategy for OA treatment.
Interleukin-6 (IL-6) is known as a mediator of inflammation, immune response, and hematopoiesis . Due to the pathological effects of IL-6 under many adverse conditions, targeting IL-6 has become important in drug development. IL-6 is detected in synovial fluid and expressed in osteoarthritic cartilage and has an essential role in the pathogenesis of OA , which makes its inhibition a potential target in the OA treatment . A previous review has confirmed the critical role of IL-6 in the progression of OA, which is supported by the finding that IL-6 expression is elevated in OA human synovial fluid and sera . CXCL1/CXCR2 activation promoted IL-6 expression in a time-dependent manner , suggesting IL-6 is an immediate-early gene in OA pathogenesis. In the present study, IL-6 was significantly increased in the joint tissues of OA model mice and human OA chondrocytes, which can be inhibited after resveratrol treatment. This indicates that IL-6 has the function to promote the OA process as the target for resveratrol treatment. Although the function of IL-6 in OA has been proposed, the role of IL-6 in OA still requires further studied.
Matrix metalloproteinases (MMPs) are extracellular zinc-dependent proteases. It not only has a crucial role in extracellular decomposition processes  but also provokes focal destruction of the vascular extracellular matrix (ECM) through proteolysis . MMP3 is a member of the MMPs family and a joint modulator of ECM involved in disease morphogenesis, wound healing, tissue repair, and remodeling . Ma et al. showed that the expression levels of serum MMP3 and inflammatory mediators in rheumatoid arthritis were positively correlated with the severity of the disease . Also, MMP3 was found to be overexpressed in OA synovial cells, and high expression of MMP3 would promote cell proliferation and inhibit cell apoptosis, increasing OA inflammation . Based on the important function in bone-related diseases, MMP3 was chosen as the candidate genes to investigate the OA progress and resveratrol treatment. In the present study, MMP3 was significantly increased in the joint tissues of OA model mice and human OA chondrocytes, which can be reversed by resveratrol treatment. The present study not only provided a treatment strategy for OA but also proved that MMP3 might be a new biomarker for judging the prognosis of OA.
NADPH Oxidase (NOX) 4, a member of the NOX family, is the source of reactive oxygen species (ROS) production . Temporal and spatially directed specific inhibition of abnormal ROS signals in the inflammatory response after joint injury may have the potential to protect normal joint tissues and reduce early changes associated with the development of osteoarthritis . NOX4 constitutively produces hydrogen peroxide (H2O2)  and participates in the development and homeostasis of tissues that make up normal joints . NOX4 plays a significant role in the acute phase after joint injury, and targeted inhibition of inflammation caused by NOX4 may help prevent early joint changes in the pathogenesis of post-traumatic osteoarthritis . However, the role of NOX4 in the OA progress and resveratrol treatment remains blank. In this study, we firstly demonstrated that NOX4 was significantly increased human OA chondrocytes, and it was also the target of resveratrol treatment. An in-depth study of the role of NOX4 in the diagnosis and treatment of OA patients is essential and has aroused great interest.
Prostaglandin-endoperoxide synthase 2 (PTGS2) encodes cyclooxygenase-2, which is and highly polymorphic and has a type of single nucleotide polymorphism located in its regulatory regions . Previous studies have also identified high expression of PTGS2 in human OA chondrocytes and synovium samples  and showed increased expression in both mouse and human OA cartilage . Down-regulation of PTGS2 reduces the expression of IL-6, MMP-3 , and the down-regulation of MMP-3 and PTGS2 have been shown to help inhibit synovial fibroblast proliferation and joint inflammation in rheumatoid arthritis . In the present study, PTGS2 was significantly increased in the joint tissues of OA model mice and human OA chondrocytes, which is consistent with the previous research. Moreover, resveratrol treatment can inhibit PTGS2 expression, which indicates that PTGS2 has the function to promote the OA process as a target for resveratrol treatment.
Our study provides valuable mechanism information of the resveratrol in the treatment of osteoarthritis, which is the first study to use such an approach for predicting resveratrol treatment OA targets. The limitation of our study is only used human OA chondrocytes to simulate the OA status. Alternative methods, such as using OA animal models, would further support our findings, and the predicted genes need to be validated on large-scale samples in the future.