The ECM is a complex network of large molecules such as polysaccharides and proteins around cells. This structure and special cells make up the cartilage (Theocharis et al., 2019). Many diseases are associated with changes in the composition and properties of ECM, such as OA. In healthy articular cartilage, the synthesis and metabolism of ECM should always maintain a dynamic balance to maintain homeostasis. The secretion and operation of some components of ECM are completed by articular cartilage (Rahmati et al., 2017). When the activities of synthesis and catabolism of articular cartilage are imbalanced, the details and homeostasis of ECM are also affected (Guo et al., 2002). Considering that the degradation of ECM leads to the loss of cartilage tissue, ECM also plays a role in maintaining the function of chondrocytes. The continuous degradation of ECM will induce OA.

Based on the summarizing of articles related to the pathogenesis of OA and the function of chondrocytes, Goldring and Goldring (2007) found that the degradation of ECM and decomposition of collagen could lead to the gradual loss of the shape and position of articular cartilage to induce OA. Setayeshmehr et al. (2019) summarized the functions and effects of various ECM scaffolds through summarizing and analyzing the related articles on hybrid and composite scaffolds derived from chondrocyte ECM. They concluded that the degradation of chondrocyte ECM could lead to cartilage degeneration. Malemud (2017) analyzed the role of matrix metalloproteinases in OA. They found that in the ECM of the cartilage, some essential proteins would degrade. Matrix metalloproteinases can promote or inhibit the degradation of these proteins, change the biomechanical properties of tissues, and destroy articular cartilage. In addition, based on the analysis of the action mechanism of ECM and its degrading enzymes in OA, the ECM of cartilage in OA patients can release certain fibronectin, thereby inducing the expression of related proteases and degrading the components of the ECM (Pérez-García et al., 2019).

Kapoor et al. (2011) and Theocharis et al. (2019) also found that the tissue protein would be hydrolyzed gradually during the development of OA. Then, some catabolic mediators will be produced, which will induce the expression of some related cytokines and proteases, and finally lead to ECM degradation. This situation will lead to prolonged OA. Based on the analysis of the changes of some macromolecules in the ECM of early OA patients, the content of proteins related to ECM was changed, such as collagen and cartilage protein. Hence, the degradation of ECM and even the deterioration of OA disease occurred (Lorenzo et al., 2004).

Apoptosis is an active gene-determined process that automatically ends life, and it is often called programmed cell death, a basic biological phenomenon of cells (Musumeci et al., 2011). Apoptosis involves endogenous and exogenous pathways, which play an essential role in maintaining the function of various tissues in the body (Elmore, 2007). A correlation has been observed between the degree of cartilage injury and apoptosis, which is a vital mechanism of cartilage injury (Hwang and Kim, 2015). Moreover, the apoptosis of OA chondrocytes is related to cartilage degradation (Kim et al., 2003). When OA occurs, it will produce matrix-degrading enzymes, leading to the degradation of the ECM and the destruction of cell homeostasis. Cell stress induces an oxidation reaction, leading to the apoptosis of chondrocytes and regulating the pathological changes of cartilage tissues (Bauer et al., 2006).

Histologically, Musumeci et al. (2011) studied the articular cartilage of normal individuals and OA patients by staining apoptotic cells. They found that the articular cartilage of OA patients changed in structure. The pro-apoptotic receptor on the cell surface could induce apoptosis through an exogenous pathway in the damaged articular cartilage. Aigner et al. (2004) summarized the incidence, induction mechanism, and morphology of apoptosis in OA patients and found that apoptosis occurred in patients with OA, but the incidence was not high. The body induced apoptosis through specific proteoglycans, signaling molecules, and other pathways to regulate the pathological changes of articular cartilage and accelerate or delay the progression of OA. Based on the synthesis of related articles on OA and apoptosis, some articles indicate that apoptosis induces OA, while other articles indicate that OA induces apoptosis. These two views are disputed, and more reports believe that these conditions affect and promote each other and have a correlation (Zamli and Sharif, 2011). Chondrocytes in the knee joints of patients with OA and the apoptosis of chondrocytes and its relationship with cartilage degradation have been studied. Notably, the findings indicate that first, in some populations, the morphology and function of cartilage are normal (Hashimoto et al., 1998). Still, many apoptotic cells appear, indicating that apoptosis may lead to degenerative changes in articular cartilage, thus inducing OA. Furthermore, the number of apoptotic articular cartilage cells in the OA population increased, and the number of apoptotic cells was related to cartilage degradation. Finally, when all subjects, including the average population and OA patients, were analyzed, a correlation was found between age and apoptosis of chondrocytes. Similarly, rabbit chondrocytes have been studied and compared with mature rabbits. Results show that the cell density of each layer of articular cartilage in rabbits decreased, and the expression level of pro-apoptotic genes increased. These studies further proved the correlation between age and chondrocyte apoptosis at the animal level and the phenomenon of apoptosis that existed in the early stage of OA (Todd Allen et al., 2004). In another study involving horses, a positive correlation was found between the severity of chondrocyte injury and apoptosis. The expression level of pro-apoptotic factors is higher in horses that often suffer from OA. Apoptosis participates in the pathogenesis of OA. Apoptosis has been observed on human femoral head cartilage (Thomas et al., 2011). In comparison with the average population, the expression of receptors and the degree of apoptosis in OA patients’ articular cartilage was much higher. In the late stages of OA, the degree of apoptosis become even worse (Héraud et al., 2000).

The inflammatory reaction is a basic pathological process mainly involving a defensive response when the body is stimulated. Inflammation is involved in the pathological process of many diseases, such as cancer (Gianni et al., 2016), tendon or ligament injury (Gracey et al., 2020). As early as the middle of the 19th century, studies pointed out that inflammation was closely related to OA. Articular chondrocytes and synovial cells all expressed inflammatory mediators (Liu-Bryan, 2013). Inflammatory factors such as IL-1β and TNFα participate in the inflammatory response in articular chondrocytes and synovial cells (Malfait, 2016). The cells mainly involved in the inflammatory response are macrophages and monocytes. The degree of inflammation is related to the degree of joint dysfunction and inflammatory factors (Scanzello et al., 2011). In the early stage of OA, the expression of monocytes and inflammatory mediators increased, whereas with the continuous development of OA, the expression of monocytes and inflammatory mediators was gradually decreased (Benito et al., 2005).

Rosshirt et al. (2019) used the knee joints of 55 OA patients as the research object, described the activation state of T cells, and found that a large proportion of T cells were activated to participate in OA inflammation, which mainly affected the joint itself the most. Inflammation can promote catabolism and participate in the pathogenesis of the disease. van den Bosch (2019) found a large number of inflammatory mediators in the tissues of patients with OA. These inflammatory mediators can induce the generation of degrading enzymes in articular cartilage, leading to the degradation of the ECM and the destruction of cartilage tissue. Lieberthal et al. (2015) established a post-traumatic OA model and observed the relationship between inflammation and OA progression. They found that an inflammatory reaction occurred in the early stage of OA. In patients with, multiple inflammatory pathways were activated and produced multiple inflammatory mediators such as cytokines and chemokines, which played an essential role in the pathogenesis of OA (Scanzello, 2017a). The genomic expression profiles of OA patients have been collected and subjected to meta-analysis, and the results show that MAP kinase, NF-κB activation, and oxidative phosphorylation can induce inflammatory signals and participate in the pathological change mechanism of OA (Li et al., 2014). Macrophages are immune cells that play an essential role in the inflammatory response (Bondeson et al., 2010). Woodell-May and Sommerfeld (2020) explored the role of immune cells such as macrophages in the synovial membrane of patients with OA. They found that in OA, macrophages could release oxygen free radicals, proteases, and inflammatory factors, thus affecting the microenvironment of inflammation. In the wound healing stage, macrophages also release IL-1, IL-6, and other inflammatory factors to regulate the inflammatory response. By stimulating protease, inflammation can cause cartilage degeneration and eventually induce OA. Based on the relationship between mechanical injury and OA, mechanical damage can lead to mechanical inflammation, which activates NF-kB and inflammatory mitogen-activated protein kinase, thereby causing a series of functional problems of the join (Vincent, 2019). If the rash persists for a long time, it will affect the repair of cartilage tissue. Based on the summary of articles related to inflammation in OA (Goldring and Otero, 2011), when some risk factors inducing OA appear, the expression of pro-inflammatory factors and various related enzymes are upregulated in articular cartilage and synovial tissue through specific signaling pathways. Inflammatory elements, which are essential for cartilage damage and repair, can change adjacent joint tissues and form a vicious circle (Houard et al., 2013). The knee joint is damaged under severe mechanical stimulation, causing catabolism and stress response of chondrocytes, and finally inducing inflammation, leading to joint pain. Inflammation is one of the important mechanisms that lead to the pathological changes of joint cartilage and synovial membrane (Scanzello, 2017b).

During autophagy, cells degrade their damaged organelles and macromolecular substances by using lysosomes to regulate autophagy-related genes, which can achieve the renewal of organelles and is essential for cell metabolism (Guo et al., 2021). Autophagy is involved in the occurrence of OA. Through cellular autophagy, the function of damaged articular cartilage can be restored, thereby alleviating the pathological process of OA. Excess ROS will lead to cartilage degradation and inhibit the synthesis of ECM. Kongara and Karantza (2012) showed that autophagy could maintain the typical morphology of cartilage and the dynamic balance of ECM synthesis and metabolism by regulating the body’s ROS.

Barranco (2015) found that the inhibition of the Akt-mTOR signaling pathway in chondrocytes can promote the autophagy of articular chondrocytes and regulate oxidative stress (Xue et al., 2017), thus participating in the development process of OA. If autophagy is activated, the damaged mitochondria and peroxidase bodies are removed, thus inhibiting the production of reactive oxygen species in the body and protecting the articular cartilage from pathological changes. Based on the summary of the roles of autophagy in OA, in the early stage of OA, moderate autophagy contributes to the survival of chondrocytes and is essential for preventing and delaying OA (Duan et al., 2020). Li et al. (2016) found that autophagy was mainly committed in the chondrocytes of patients with OA. By regulating oxidative stress response and apoptosis, the pathological changes of the knee joint can be alleviated, and chondrocytes can be protected from various stimulations. With the growth of age, the incidence of OA increases gradually, and this phenomenon is related to autophagy. Autophagy gradually weakens with age. Therefore, the structure and function of articular cartilage decreases with age, leading to pathological changes in the knee joint (Mathew et al., 2009). Caramés et al. (2010) that in OA animal models, the expression of autophagy regulatory factors is downregulated, and the steady-state of cartilage is damaged, indicating that autophagy can protect articular cartilage and play an essential role in maintaining cartilage steady state. The decrease of autophagy will induce OA. Taking articular cartilage of OA and non-OA patients as the research object, we can find that autophagy regulates the expression of OA-related genes. In comparison with patients without OA, autophagy is expressed in the chondrocytes of OA patients at a higher level, and the manifestation of autophagy markers is also upregulated (Sasaki et al., 2012). Similarly, Chang et al. (2013) took human chondrocytes as the research object and found that many autophagy-related proteins are expressed in the articular cartilage of patients with OA. In the pathogenesis of OA, autophagy plays a role in protecting chondrocytes and promoting metabolism. However, excessive autophagy causes a large number of chondrocyte deaths.

Non-coding RNA (ncRNA) is an RNA that does not code for protein, mainly including miRNA, lncRNA, and circRNA. These RNAs can exert biological functions at the RNA level without being translated into proteins. ncRNA plays an essential role in the pathological process of many diseases, such as cardiovascular disease (Jusic and Devaux, 2020), cancer (Karreth and Pandolfi, 2013), diabetes (Beltrami et al., 2015). ncRNA also plays a vital role in the process of OA. It can promote or inhibit cartilage formation by regulating the degradation of cell-matrix, cell proliferation, apoptosis, and inflammatory response, thereby inducing or treating OA (Hong and Reddi, 2012).

Changes in the expression levels of many miRNAs can induce pathological changes in OA. MiR-204/-211 and miR-29b-3p are common miRNAs, which are differentially expressed in patients with OA. Huang et al. (2019) established an OA mouse model and found that miR-204/-211 was missing, and Runx2 was increased. Mesenchymal progenitor cells proliferated abnormally at the same time by Micro-CT and histological determination. Akt signal North is activated, thus inducing the dysfunction of various components in the joint and finally leading to OA. Therefore, miR-204/-211 can maintain intra-articular homeostasis and ensure that articular chondrocytes and synovial cells function usually. Chen et al. (2017) used rat chondrocytes as the research object. Through luciferase reporter gene detection, they found that miR-29b-3p in patients with OA was upregulated, and miR-29b-3p ultimately led to the degeneration of articular cartilage tissue by targeting and inhibiting the expression of rat GRN mRNA. In addition, microarray technology was used to detect the expression of miRNA in chondrocytes. They found that miRNA-140, miRNA-455, miR-146a, miR-155, and miR125b differed in OA patients and the average population, thus inducing pathological changes of articular cartilage, synovial membrane, and subchondral bone (Swingler et al., 2012). lncRNA also plays a vital role in OA. Throughout the cartilage development, different lncRNA is regulated, and the regular expression of lncRNA can prevent cartilage differentiation disorders. During cartilage degeneration, other lncRNA plays various roles (Zhu et al., 2019). Ye et al. (2018) studied the articular cartilage of patients with OA. They found that the expression of ZFAS1 in OA chondrocytes was downregulated, and ZFAS1 might reduce the activity of chondrocytes and induce pathological changes of articular cartilage tissues by targeting the Wnt3a signaling pathway. Similarly, the role of circa in OA has been widely studied. Cyclic RNA plays a multi-faceted regulatory role in the progression of OA (Zhang W. et al., 2021). Chen et al. (2020) found that in OA tissues, the expression levels of circRNA-UBE2G1 and HIF-1a were significantly increased, while the expression of miR-373 was downregulated. Functional testing showed that circRNA-UBE2G1 could bind to miR-373, thereby inhibiting the interaction of miR-373 and HIF-1a, damaging the chondrocytes, and leading to a series of pathological changes. In conclusion, ncRNA is closely related to OA. The differential expression of miRNA, lncRNA and circRNA in healthy people and OA patients can lead to pathological changes in articular cartilage, synovial membrane, and subchondral bone and affect the pathological process of OA.

Methylation is the catalyzed transfer of methyl groups from an active compound to another compound. This process can form various methyl compounds or result in chemical modification of particular proteins or nucleic acids to start methylation products (Urnov, 2002). Methylation mainly includes DNA methylation and m6a methylation (Dai et al., 2021).

DNA methylation is a chemical modification of DNA that alters genetic behavior without altering the DNA sequence. Its primary process is under the action of DNA methyltransferase. DNA methylation can cause changes in chromatin structure, DNA conformation, DNA stability, and the way DNA interacts with the protein, thereby controlling gene expression (Reynard, 2017). When CtBP1 and CtBP2 are overexpressed, the pro-inflammatory factors are increased, while the NLRP3 signaling pathway is activated to induce OA. DNA methylation in the CtBPs promoter reduces the expression level of CtBPs in OA tissues and regulate CtBP-mediated signal transduction, finally participating in the pathogenesis of OA (Sun et al., 2020). In addition, DNA methylation can participate in the pathogenesis of OA by affecting the expression of genes such as matrix metalloproteinase-3, matrix metalloproteinase-9, and type II collagen (Cui and Xu, 2018).

N6- methyladenine (m6a) is one of the most abundant chemical modifications of eukaryotic messenger RNA. m6a modifications mainly include m6a methyltransferase catalysis, m6a demethylase removal, and m6a binding protein recognition (Ma and Ji, 2020). This process is widely involved in regulating various life cycle stages such as mRNA splicing, processing, translation, and degradation. It is related to osteosarcoma, rheumatoid arthritis, osteoporosis, OA, and abnormal physiological functions (Zhang W. et al., 2020). The strange expression of m6a-related gene and protein in OA can trigger the imbalance of m6a methylation, regulate the expression of OA-related genes to participate in the occurrence and development of OA, and is closely related to the poor prognosis of patients (Yang et al., 2021). METTL3 is a methylated gene of m6a. He et al. (2021) studied mice in vivo and in vitro and induced the OA model with inflammatory stimuli such as TNF-α. The results showed that the expression of METTL3 was decreased in the OA model. Bcl2 is a downstream target gene of METTL3, and the presentation of METTL3 can promote m6A methylation of Bcl2 mRNA, thereby inhibiting chondrocyte apoptosis and autophagy. The detection of synovial tissue in patients with OA revealed that METTL3 could also induce autophagy by inhibiting ATG7 (Chen X. et al., 2021). In addition, METTL3 may participate in OA by regulating the inflammatory response (Sang et al., 2021). In summary, methylation is also an essential mechanism for pathological changes in OA. Furthermore, the epigenetic mechanism plays an essential role in the pathogenesis of OA.

Exercise can delay the pathological process of OA by inhibiting the degradation of the ECM. Articular cartilage degradation is an essential pathological change of OA. Blazek et al. (2016) used OA rats as the research object. Exercise and non-exercise were used as intervention means to compare the whole gene expression of the transcriptome of articular cartilage in rats. Microarray analysis showed 644 differentially expressed genes in the articular cartilage of exercise rats. Therefore, exercise can prevent OA by changing genes related to OA’s pathogenesis and regulating the degradation and synthesis of ECM by changing metabolic pathways to alleviate the pathological changes of articular cartilage with sound prevention effects on OA. An OA rat model has been established to evaluate the expression of IL-10, TGF-β, and collagen type I and II in articular cartilage. The presentation of the above biochemical indicators increased in the articular cartilage of mice after exercise, indicating that exercise is very beneficial to articular cartilage. Moreover, considering that collagen type II is an essential component of ECM, exercise may delay the progression of OA by inhibiting the degradation of ECM and promoting cartilage synthesis (Assis et al., 2018). Vasilceac et al. (2021) conducted resistance training for eight weeks for OA rats. They analyzed whether ELISA changed the activity of MMP-2 in the tendons around the knee joint and finally found that exercise could reduce the activity of MMP-2 in the quadriceps tendon of OA rats and greatly relieve the adverse effects of OA on the knee joint. Results show that exercise may reduce the degradation of the ECM by inhibiting the activity of MMP-2 from achieving the purpose of treating OA. Exercise in different stages of the course of OA produces other effects. Type II collagen (CoII) and matrix metalloproteinase-13 (MMP-13) are essential components related to the degradation of the ECM. Hsieh and Yang (2018) established an OA rat model, in which the samples were allowed to swim in different stages of the course of OA. They found that swimming training in the early stage of OA could increase CoII. The level of MMP-13 can be reduced to maintain a balance between the degradation and anabolism of ECM and prevent cartilage damage through exercise training in the early stage of OA rather than that in the late stage.

Overall, the degradation of the ECM is the pathological change mechanism of OA. Exercise can protect articular cartilage and delay the progression of OA by inhibiting this mechanism, and it is a crucial intervention to prevent and treat OA.

Apoptosis is one of the mechanisms for the pathological changes of OA. Exercise delays the pathological process of OA by inhibiting apoptosis. Taking articular cartilage of OA rats as a research sample, Yang et al. (2020) detected the expression of relevant genes in articular cartilage after exercise through bioinformatics analysis. Results show that the levels of TRAIL, NF-κB p65, and NLRP3 in cartilage after activity were decreased. Exercise could inhibit apoptosis and prevent OA by regulating the TRAIL/NF-κB/NLRP3 signaling pathway of OA. By comparing OA rats with mild, moderate, and high-intensity training and comparing the expression of caspase three and Hsp70, we found that mild and moderate exercise could inhibit apoptosis and protect the articular cartilage (Galois et al., 2004). Qian et al. (2014) established an OA rat model, and the experimental group was subjected to passive motion. Based on the measurements of the proteoglycan content of the cartilage matrix, the number of type II collagen fibers, and apoptotic chondrocytes in the experimental and control group, the changes in biochemical signals caused by passive exercise in the early and middle stages of OA differed. In comparison with the control group, the cartilage matrix proteoglycan content and type II collagen fiber level remarkably increased three weeks after passive motion in the early stage of OA. The number of apoptotic cells was significantly reduced. However, three weeks after passive exercise in the middle stage of OA, the number of apoptotic cells was not significantly changed. Therefore, passive exercise at the early stage of OA may delay articular cartilage degeneration by inhibiting apoptosis, preventing, and treating OA. After intermittent training for OA rats, the pathological changes of their subchondral bones were assessed by immunolabeling with cleaved caspase-3 in the cortical subchondral bone. Intermittent aerobic training may prevent the OA-induced reduction in bone mineral density by reducing apoptosis (Boudenot et al., 2014). Studies (Iijima et al., 2015) have established the OA rat model and analyzed the pathological conditions of rat articular cartilage and subchondral bone before and after four weeks of exercise. They found that exercise might inhibit apoptosis and protect cartilage tissue, thus delaying OA progression. Musumeci et al. (2013) established the OA rat model to study the effect of exercise on articular cartilage. First, drugs were used to induce pathological changes in the articular cartilage. Then, exercise intervention was performed on rats to observe the morphological changes of articular cartilage. The results show that exercise might induce the expression of lubricating oil and inhibit the activity of caspase-3, thereby reducing apoptosis and delaying or even reversing pathological changes of articular cartilage. In some studies (Zhang J. et al., 2021) concerning establishing the OA rat model, the experimental group was intervened by swimming for four weeks. The morphological changes of cartilage were analyzed by hematoxylin-eosin (H&E) staining. The expression of caspase-3 was analyzed by Western blot and qPCR. The results showed that after four weeks of swimming, the abnormal morphology of articular cartilage was improved, and the level of caspase-3 protein decreased, indicating that exercise might reduce apoptosis by inhibiting the level of caspase-3 protein, thereby protecting articular cartilage and preventing and treating OA.

Exercise can inhibit the inflammatory response by reducing pro-inflammatory factors, thereby delaying the pathological changes of OA, which is the most common mechanism of activity in OA treatment. The inflammatory response mechanism of exercise therapy for OA has been widely studied, and many related pathways regulate the inflammatory response of OA, such as PI3k/Akt, NF-κB p65, JNK/NF-kB, HDAC3/NF-kappaB, and AMPK/NF-κB signaling pathways. Lu et al. (2021) established the OA rat model and allowed them to exercise on the treadmill. They found that after running exercise, the mice produced maresin-1, and the increased level of maresin-1 activated the PI3k/Akt pathway and inhibited the NF-κB p65 pathway, thus playing an anti-inflammatory role and delaying the pathological changes of OA. Griffin et al. (2012) established the OA mouse model by inducing a high-fat diet and then letting the mice run to observe the expression of inflammation-related factors. Finally, they found that exercise could regulate the expression of pro-inflammatory factors, promote joint health, and reduce the severity of pathological changes in joints of OA mice. To study the signal transduction of JNK/NF-kB in KOA patients, we established the 0A rat model and conducted a controlled intervention experiment. Finally, the knee joint diameter in the exercise group is lower than that in the OA group, indicating that exercise is conducive to KOA recovery. The IL-1b, IL-6, and TNF-a levels decreased, indicating that exercise can reduce the inflammatory response by regulating the JNK/NF-kB signaling pathway, which ultimately can delay the pathological changes of OA (Chen et al., 2020). Osteoarthritis rats were subjected to moderate-intensity exercise on the treadmill. The H&E staining and toluidine blue O staining were used to detect cartilage injury. The expression levels of some biochemical signals in the articular cartilage were examined via immunohistochemistry and other methods. The results showed that moderate-intensity treadmill exercise could reduce the inflammatory response and protect the articular cartilage by inhibiting the HDAC3/NF-kappaB pathway (Zhang H. et al., 2019). Similarly, Yang et al. (2019) intervened OA rats with treadmill exercise and used the articular cartilage as the research object. Observation and analysis results show that moderate-intensity exercise can reduce the sensitivity of articular cartilage and chondrocytes to inflammatory response through the AMPK/NF-κB signaling pathway, and it is an essential mechanism for reversing the pathological changes of articular cartilage. The analysis of biochemical indicators of OA rats after aerobic exercise show that the expression levels of IL-1β, caspase-3, and MMP-13 in OA rats decreased after exercise, and aerobic exercise could inhibit the inflammatory response of OA and prevent and treat the pathological changes of cartilage (Assis et al., 2016). After water sports, OA rats’ articular cartilage and inflammatory mediators were changed. After eight weeks of the experiment, the cell damage degree of the rat was relieved, and the expression levels of IL-1β and caspase-3 were reduced. Therefore, sports can regulate OA inflammatory factors and inhibit inflammatory reactions to treat OA (Milares et al., 2016).

In OA, autophagy is a cellular protective response. Exercise can protect the knee joints of KOA patients by inducing autophagy, which is mainly regulated by oxidative stress. Relatively few studies have demonstrated the specific mechanism pathways through which exercise protects the knee joint through autophagy. Zhang X. et al. (2019) used sodium iodoacetate to induce OA rat model, and the experimental group was subjected to treadmill exercise for four weeks. The study found that serum IL-1β was decreased, IL-4 was increased, and the expression of type II collagen in articular cartilage was increased in rats undergoing treadmill exercise compared with those in OA rats that did not experience any activity. Therefore, treadmill exercise may protect the knee joints of OA patients by promoting the autophagy of articular cartilage. Baur et al. (2011) established an OA mouse model to study the effect of running on articular cartilage and bone of OA patients and observed the changes of biochemical signals and articular cartilage of running mice after eight weeks. They found that the content of reactive oxygen species in mice increased after exercise. Exercise can promote autophagy and protect the articular cartilage and bone of OA patients by regulating oxidative stress. Another study (Cifuentes et al., 2010) has established an OA rat model. The experimental group was exposed to treadmill exercise for 8 weeks to observe the pathological changes of articular cartilage. In comparison with OA rats that did not undergo exercise, the protein and polysaccharide contents on the surface of the articular cartilage of rats after eight weeks of the exercise was relatively high. This experiment confirmed that training can enhance the oxidative stress mechanism and protect articular cartilage. Moderate-intensity exercise plays different roles in different stages of OA progression. In the early stage of OA, the average power of movement can delay the progression of OA by promoting autophagy. In the late stage of OA, moderate-intensity exercise can over-activate purinergic receptor P2X ligand-gated ion channel 7 and increase the number of apoptotic cells, thus aggravating the progression of OA. Throughout the OA progression, the body may promote autophagy or apoptosis through the IRE1-mTOR-PERK signal axis to delay or accelerate the pathological process of OA (Li Z. et al., 2021).

Exercise can alleviate the pathological changes of OA by regulating the expression of ncRNA. The practice affect the face of miRNA, lncRNA, and circRNA in articular cartilage, synovial membrane, and subchondral bone of OA patients.

miRNA is related to maintaining the typical morphology of articular cartilage. The detection of miRNA expression in OA rats revealed that compared with the non-exercise rats, 394 differentially expressed miRNAs were detected in the exercise rats (Yang et al., 2020). The analysis of articular cartilage from different animals revealed that exercise could change the miRNA expression and improve the pathological changes of articular cartilage through a series of mechanisms. Dunn et al. (2009) analyzed the expression of miRNA in bovine articular cartilage. They found that the expression patterns of miR-221 and miR-222 in weight-bearing medial anterior condylar articular cartilage was higher than those in non-weight-bearing medial posterior condylar articular cartilage. When the articular cartilage was loaded with weight, miR-221 downregulated type II collagen and Sox9, and miR-222 downregulated HDAC4 and MMP-13, maintaining the typical morphology of articular cartilage. Guan et al. (2011) subjected chicken chondrocyte samples to mechanical stimulation to analyze the roles of relevant miRNA in the mechanical stimulation using microarray technology. Results show that mechanical stimulation could upregulate the miR-365 expression in chondrocytes. miR-365 stimulated the differentiation of chondrocytes by targeting histone deacetylase 4 (HDAC4) and finally alleviated the pathological process of cartilage tissue. In addition, the expression levels of many lncRNA changed during exercise. Zhou et al. (2021) established an OA mouse model. Mice with moderate-intensity exercise intervention had an upregulated lncRNA H19 expression, thickened articular cartilage, and resolved OA compared with mice without exercise intervention. Under the stimulation of mechanical stress, lncRNA-MSR was activated and then competitively bound to miR-152, thereby promoting the expression of TMSB4 and leading to the degradation of the ECM (Liu Q. et al., 2016). The regulatory role of circRNA during exercise is also crucial. A total of 104 circRNA were differentially expressed (44 upregulated and 60 downregulated), with increased circRNA-MSR expression, in damaged articular cartilage compared with intact articular cartilage. circRNA-MSR expression was downregulated after mechanical stress was applied to the cartilage. The downregulation of circRNA-MSR can inhibit the expression of TNF-α and promote the degradation of the extracellular chondrocyte matrix, thus enhancing the structure and function of cartilage (Liu et al., 2017). In addition, circRNA-MSR can directly target miR-643 to u-regulate MAP2K6. The downregulation of circRNA-MSR expression can promote cell proliferation and reduce cartilage damage (Jia and Wei, 2021). qRT-PCR and Western blot have been used to analyze the genes related to cartilage differentiation in cartilage tissues. The results show that the expression of circUNK was upregulated, which improved the cartilage injury in OA rabbit and promoted the expression of molecules related to cartilage differentiation, such as PCNA, SOX9, Col II, Aggrecan, and protein-polysaccharide. These molecules had sound effects on the proliferation and differentiation of articular cartilage. In conclusion, exercise or mechanical stimulation can lead to changes in the expression of ncRNA, which prevents and improves the pathological changes of articular cartilage, synovial membrane, and subchondral bone through a series of mechanisms, thereby delaying the progression of OA (Fang et al., 2021).