MicroRNAs' Involvement in Osteoarthritis and the Prospects for Treatments

Osteoarthritis (OA) is a chronic disease and its etiology is complex. With increasing OA incidence, more and more people are facing heavy financial and social burdens from the disease. Genetics-related aspects of OA pathogenesis are not well understood. Recent reports have examined the molecular mechanisms and genes related to OA. It has been realized that genetic changes in articular cartilage and bone may contribute to OA's development. Osteoclasts, osteoblasts, osteocytes, and chondrocytes in joints must express appropriate genes to achieve tissue homeostasis, and errors in this can cause OA. MicroRNAs (miRNAs) are small noncoding RNAs that have been discovered to be overarching regulators of gene expression. Their ability to repress many target genes and their target-binding specificity indicate a complex network of interactions, which is still being defined. Many studies have focused on the role of miRNAs in bone and cartilage and have identified numbers of miRNAs that play important roles in regulating bone and cartilage homeostasis. Those miRNAs may also be involved in the pathology of OA, which is the focus of this review. Future studies on the role of miRNAs in OA will provide important clues leading to a better understanding of the mechanism(s) of OA and, more particularly, to the development of therapeutic targets for OA.


Introduction
OA is a joint disease of relatively frequent occurrence that is characterized by progressive damage to the articular cartilage and subchondral bone, causing pain and disability in older adults [1,2]. Common features of OA include joint pain, turgidity, synovial inflammation, and loss of function. The most common risk factors for OA include age, gender, obesity, genetic predisposition, prior joint trauma, mechanical factors, malformation, and hypoplasia. Several tissues of the joint, including the cartilage, synovial membrane, and subchondral bone, play significant roles in the development of OA pathology. OA is associated with degradation of cartilage, ultimately leading to the complete loss of cartilage, due to degeneration and then apoptosis of chondrocytes, loss of bone, with increased bone remodeling followed by slow reversal. Alterations beneath the cartilage at the osteochondral junction have drawn interest as mediators of the structural progression of OA. Osteochondral alterations happen early during the development of OA and can aggravate the pathological changes elsewhere in the joint [3]. In this process, inflammatory factors have a significant effect, but genetic factors may be the biggest issue. Chondrocytes and bone cells must express appropriate genes to achieve joint homeostasis and if this is upset, this can lead to OA. The genes expressed and the transcription factors that control chondrogenesis and osteogenesis have been widely studied.
MicroRNAs (miRNAs) are a large family of small (21-25-nucleotide) noncoding RNAs. There have been many publications describing miRNA synthesis and function, and miRNAs have emerged as important posttranscriptional regulators of gene expression that function by binding to specific sequences within target mRNAs. Each miRNA recognizes target sequences in many genes and each gene contains target sequences for the binding of many miRNAs, leading to translational repression or mRNA degradation. miRNAs regulate gene expression by binding to the 3 -untranslated region (UTR) of their target mRNAs. Many miRNAs exhibit a tissuespecific or developmental stage-specific expression pattern, which has been reported to be associated with diseases, such as cancer, leukemia, viral infections, and bone disease [4][5][6][7]. Recent evidence indicates that miRNAs have important roles in regulating osteogenic and chondrogenic differentiation and proliferation, eventually influencing the catabolism and anabolism of bone and cartilage. This review focuses on miRNA in OA. Recently, miRNA has been described as a novel therapeutic strategy for OA and is expected to provide targets for clinical treatment and biomarkers for diagnosis and prognosis for OA [8]. Thus, a more complete understanding of all facets of miRNA function in the joint is necessary.

MicroRNA Formation and Biological Function
Many miRNAs are located within introns of protein-coding genes, and a small proportion originate from exons or noncoding mRNA-like regions [9]. Despite the distinct biological differences between miRNAs and mRNAs, present evidence suggests that their transcripts share common mechanisms of transcriptional regulation [10]. miRNAs are transcribed by RNA polymerase II and processed in the nucleus by a protein complex that contains RNase III, Drosha, and DGCR8 ( Figure 1). Exportin 5, a GTP-dependent and dsRNA-binding protein, mediates the process that transports the resulting 60-80-nucleotide pre-miRNA cleavage product to the cytoplasm. Another protein complex, Dicer, in the cytoplasm, containing the RNase III enzyme, cleaves pre-miRNA at the stem-loop structure to generate an approximately 22nucleotide imperfect double-stranded miRNA duplex [11]. One strand of the duplex is designated as the guide miRNA and remains stably associated with the miRNA-induced silencing complex (mRISC). The other strand, known as the passenger strand, can be degraded rapidly. It is possible that both strands could be used differently in relation to extracellular or intracellular effects, indicating that the two strands may regulate a more diverse set of protein-coding genes as needed, or the selection of strand could be tissuespecific. Mature miRNA guides the mRISCs to the 3 -UTR of its target mRNA, which is partially or fully complementary to the seed sequence of the mature miRNA [12]. Then, miRNA binding to the 3 -UTR can promote or repress the translation of the target mRNA ( Figure 1). This may be an evolutionary strategy to diversify miRNA-based gene silencing.

Role of miRNA in Bone Development
The precise role and expression of miRNA in bone development have been reported and established.
In addition, miRNAs promote osteoblastogenesis by affecting other signaling factors. miR-210 acts as a positive regulator of osteoblast differentiation and suppresses the TGF-/activin signaling pathway through inhibition of ACVR1B [23]. miR-26a modulates late osteoblast differentiation by targeting the SMAD1 transcription factor [24]. miR-96 simulates osteoblast differentiation of hMSCs by increasing the expression of FABP4, and miR-199a can promote osteoblast differentiation of hMSCs by downregulating the expression of transcription factor SOX9 and upregulating the expression of aggrecan [25].
In consideration of the previous studies, we know that change in the subchondral bone is one of the factors of OA, and when the functions of these miRNAs mutated in osteoblasts or osteoclasts, they can result in the changes in the subchondral bone, which may induce the progression of OA.

Role of miRNAs in Chondrogenesis
Dicer is an essential component for miRNA biogenesis, a deficiency of which in chondrocytes can lead to a reduction in the number of proliferating chondrocytes via accelerated differentiation and decreased proliferation into hypertrophic chondrocytes [74,75]. Recently, the important role of miRNAs in chondrogenesis and cartilage homeostasis has received much attention (Table 1).

miRNAs in Articular Cartilage
Articular cartilage provides an enduring friction-free tissue for the movement of limbs. This significant function of articular cartilage is achieved by a cartilage-specific extracellular matrix (ECM). Aggrecans and type II collagen are two prime components of the ECM, and MMPs, which include matrix-degrading enzymes, and ADAMTSs are associated with the degradation of ECM. The balance of these catabolic and anabolic processes is essential for cartilage homeostasis. Many miRNAs have been recognized as vital factors for the development, maintenance, and destruction of articular cartilage.
Miyaki et al. showed that miR-140 regulated cartilage development and homeostasis, and it can be reduced in OA cartilage and in response to IL-1 , resulting from a chondrocyte differentiation-related expression pattern, and this reduction may contribute to abnormal gene expression in OA [64]. miR-27b can inhibit the IL-1 -induced expression of MMP-13 protein in chondrocytes to reduce the degradation of cartilage [60]. Díaz-Prado et al. observed that seven miRNAs were differentially expressed in normal and OA chondrocytes [61]. Among them, miR-483-5p was upregulated in OA chondrocytes, and miR-149, miR-582-3P, miR-1227, miR-634, miR-576-5p, and miR-641 were downregulated in OA chondrocytes but upregulated in normal chondrocytes, indicating that these seven miRNAs were involved in the development, maintenance, and destruction of articular cartilage. miR-125b can target ADAMTS-4 to inhibit normal chondrocyte ECM degradation [66]. miR-127-5p inhibits normal cartilage degradation via suppressing IL-1 -induced MMP-13 production [67]. miR-146a increases the levels of VEGF and damages the TGF-signaling pathway by targeting inhibition of Smad4, contributing to the mechanical injury and apoptosis of chondrocytes [62]. miR-199a can directly prevent the activity of cyclooxygenase-2 (COX-2) and inhibit IL-1 -induced COX-2 protein expression in chondrocytes [68]. miR-558 directly suppresses the activity of COX-2 mRNA and inhibits IL-1 -induced catabolic effects in chondrocytes to protect the homeostasis of cartilage [69]. miR-675 can upregulate ECM molecules, such as type II collagen [70]. Yang et al. [56] and Martinez-Sanchez et al. [63] found that overexpression of miR-145 suppressed SOX9 in chondrocytes, and expression of matrix genes in articular cartilage-COL2A1, ACAN, COMP, COL9A2, and COL11A1-was inhibited and that of hypertrophic phenotype genes-RUNX2 and MMP-13-was stimulated.

Role of miRNAs in OA
Osteoarthritis is one of the most common chronic diseases; it results in the chondrocyte phenotype and changes in cartilage homeostasis, as well as metabolic changes in subchondral  bone, which is an integral part and active component of the OA disease process [76]. The joint disorder is caused mostly by excessive mechanical loading and movement; ultimately, cartilage and bone receive the stress and dissipate it simultaneously. Thus, the cartilage and bone are continuously challenged biomechanically. It is accepted that the balance between catabolic and anabolic rates in bone and cartilage, including the synthesis and degradation of cartilage, is perturbed in OA. Alterations in differentiation, proliferation, activity, and apoptosis of bone cells and chondrocytes are also associated with the development of OA [1]. Recently, several studies have shown that miRNAs play important roles in these processes in OA (Figure 2). Moreover, various correlations between miRNAs and pathological conditions in bone and cartilage add more complexity in OA development. Iliopoulos et al. tested the expression of 365 miRNAs using microarrays and identified 16 that were expressed differentially in OA cartilage versus normal controls: seven miRNAs (miR-29a, miR-140, miR-25, miR-337, miR-210, miR-26a, and miR-373) were downregulated and 9 (miR-483, miR-22, miR-377, miR-103, miR-16, miR-223, miR-30b, miR-23b, and miR-509) were upregulated [77]. Jones et al. also investigated the expression of 157 miRNAs and showed that 17 were differentially expressed, with 4-fold variation or more when comparing normal and late-stage OA cartilage [78].

miRNAs Related to Aging.
Age is an important risk factor for the development of OA, making the joint more vulnerable to OA risk factors, due to age-related dysregulation of matrix, and catabolic and anabolic processes in cartilage and bone generally. Although the relationship between miRNAs and aging is not fully understood, many studies have provided evidence to demonstrate that miRNAs are associated with aging. Miyaki et al. showed that miR-140 was a regulator of cartilage development and homeostasis, loss of which could contribute to the development of age-related OA [65]. Ukai et al. showed that miR-199a-3p and miR-193b were upregulated with age and associated with Col-II, aggrecan, and SOX9 downregulation, and miR-320c was downregulated with aging and associated with ADAMTS5 upregulation, 8 Evidence-Based Complementary and Alternative Medicine indicating that miR-199a-3p and miR-193b were involved in senescent chondrocytes, and miR-320c was involved in the juvenile chondrocytes in regulating cartilage metabolism [80]. miR-21 has also been shown to increase in aging cartilage [81].

miRNAs Related to Mechanical Loading.
Appropriate mechanical stress is required for cartilage and bone homeostasis. However, excessive impact energy or chronic significant mechanical loading can lead to chondrocyte damage and OA. Guan et al. identified that miR-365 was necessary for the mechanical stimulation of chondrocyte proliferation under cyclic loading conditions; this was shown in culturing primary chicken chondrocytes in a three-dimensional collagen scaffold as a mechanoresponsive miRNA, directly targeting HDAC4 and stimulating RUNX2 and Ihh (Indian hedgehog) expression [82]. Dunn et al. showed that miR-222 expression in the weight-bearing anterior medial condyle was higher than that in the posterior non-weight-bearing medial condyle in the articular cartilage [83], indicating that miR-222 is a potential regulator of the articular cartilage mechanotransduction pathway. miR-146a is overexpressed in the mechanical injury of human chondrocytes, increasing the levels of VEGF and inhibiting Smad4 to damage the TGFsignaling pathway, contributing to the pathogenesis of OA [62]. Thus, the identification of mechanisms of regulation by miRNAs in the mechanical loading of cartilage is important for developing effective treatments for OA.

miRNAs Related to Pain.
The joint degradation associated with OA is a common phenomenon and causes chronic and severe pain, especially in the knee joints. As a key debilitating symptom of OA, pain is associated with many changes in gene expression in damaged peripheral tissues and neurons. It has been reported that chronic pain in OA is caused by inflammatory responses that modulate gene expression [84]. Li et al. found that miR-146a regulated knee joint homeostasis and OA-related algesia by balancing inflammation in the cartilage and synovium with pain-related factors in glial cells [85]. miR-146a and/or the miR-183 clusters have also been identified as being closely associated with the stimulation of inflammatory pain mediators and could be a powerful therapeutic target for OA [86].

miRNAs Related to Inflammation.
Osteoarthritis is a complex disease in which inflammation plays a significant role, pertaining to development and progression, as a result of inflammatory mediators such as cytokines and prostaglandins released by cartilage, bone, and synovium, involving soluble mediators from local or systemic sources. Recently, some miRNAs have emerged as important controllers of Toll-like receptor (TLR) pathways, which can activate NF-B and induce the expression of many genes in establishing the inflammatory response in OA and of the proinflammatory cytokine IL-1. miR-155 has a negative effect on the TLR/IL-1 inflammatory pathway by preventing the activation of TGF--activated kinase 1 and NF-B and MAPK [87]. miR-146a inhibits the activation of the TLR/IL-1 inflammatory pathway through targeting IL-1 receptorassociated kinase 1 and TNF receptor-associated factor-6 [88]. Interestingly, the expression of miR-146a is elevated markedly by IL-1 and TNF-stimulation [89], suggesting a feedback loop involving miR-146a and TLR signaling. Several miRNAs have been reported to be induced by IL-1 , such as miR-34a [90], miR194 [91], and miR-27b [60]. Expression of miR-34a was significantly induced by IL-1 , promoting chondrocyte apoptosis, and downregulated the expression of type II collagen [90]. miR-27b regulates MMP-13 expression by interacting with the 3 -UTR of MMP-13 mRNA at the posttranscriptional level [60]. miR-140 was identified as being suppressed by IL-1 [64], and the expression of miR-140 was found to be NF-B-dependent [92]. miR-101 participates in IL-1 -induced chondrocyte ECM degradation, and IL-1induced ECM degradation can be prevented when miR-101 is downregulated [93]. miRNA-210 can decrease inflammation in joint cavity in OA, by targeting the death receptor 6 (DR6) and inhibiting NF-B signaling pathway [94]. It was found that miRNA-130a regulates an underlying mechanism of OA, by regulating the expression of TNF-, and it can be a novel target in OA [95]. miRNA-25 can induce COX-2 overexpressed since miRNA-25 is upregulated by WFA, which can influence the WFA-induced inflammation in OA [96]. It was proved that miRNA-149 is downregulated in OA chondrocytes, which seems to be correlated to overexpression of TNF-, IL-1 , and IL-6 [97]. Onju Ham et al. showed that inhibition of protein kinase A (PKA) signaling in synovial fluid-derived mesenchymal stem cells (SFMSCs) using miRNA-23b can be a useful method for the treatment of degenerative arthritis [98]. miRNA-9, miRNA-98, and miRNA-146 were identified by functional analysis in primary chondrocytes that they can regulate the expression of TNF-, and miRNA-9 was upregulated in OA tissue and can inhibit secretion of MMP-13 in chondrocytes [99]. miRNA-488 inhibited MMP-13 activity to promote chondrocyte differentiation/cartilage development through targeting ZIP-8, and suppression of ZIP-8 in OA can reduce cartilage degradation [100].

miRNAs in the Diagnosis and
Therapy of OA 7.1. Diagnosis. Although it is still unclear whether measurement of miRNAs in serum may be a useful tool for the diagnosis of OA, we believe that miRNAs can be powerful diagnostic biomarkers for the development of OA; extracellular miRNAs can be detected in almost all body tissues and fluids [101,102]. miR-21 was the first discovered miRNA biomarker in serum [103]. Subsequently, it has been reported that the serum miRNAs are useful for the diagnosis, and determining the prognosis, of solid cancers and leukemia [102]. Murata et al. found many miRNAs in plasma, and some of them differed significantly between OA and normal tissues [104]. The measurement of miRNAs in peripheral blood mononuclear cells (PBMCs) is also useful in developing a biomarker for OA. If a subject is suffering from OA, circulating PBMCs could accumulate in the synovium and produce proinflammatory Evidence-Based Complementary and Alternative Medicine 9 cytokines and proteinases associated with OA progression. The high expression of miR-155, miR-146A, miR-181a, and miR-223 in PBMCs from OA patients, compared with normal controls, may be connected with the pathogenesis of OA [105].

7.2.
Therapy. There has been more than 100 clinical trials worldwide based on miRNA regulation to treat diseases, including cancers and cardiovascular condition, but none for OA yet [106]. On the basis of the important roles of miRNAs in bone and cartilage homeostasis, there could be a novel therapeutic opportunity for the treatment of OA by targeting the expression and activity of miRNAs. It has been reported that the silencing of miR-34a by locked nucleic acid-(LNA-) modified antisense oligonucleotides could effectively reduce IL-1 -induced chondrocyte apoptosis [90]. Nagata et al. showed an intra-articular injection of synthetic doublestranded miR-15a resulted in cell apoptosis through preventing action of the target gene BCL-2 in the synovium of osteoarthritic mice. Labeled miR-15a that was injected into the articular cavity was detected in synovium cells, but not in chondrocytes [107]. It has been demonstrated that miR-146a regulates cytokine signaling through a negative feedback loop, and miR-140 regulates cartilage homeostasis. These findings suggest that the administration of miR-146a and miR-140 might be candidate targets for new treatments for the early stage of OA and that several miRNAs combined could be more effective than a single miRNA. However, the key to miRNA's application is overcoming the barrier of delivering miRNA into the cells.
Recently, there have been many tissue engineering strategies that have combined cell-based therapies with scaffolds and different growth factors, morphogens, and differentiation signals to achieve functional tissue-engineered cartilage. The purpose of cartilage tissue engineering is to generate an ECM with biomechanical properties that mimic those of native cartilage. However, stimulating complete regeneration and repair of cartilage has still not been achieved. Studies of miRNAs and their function in regulating expression patterns of different genes in different tissues may revolutionize strategies for cartilage tissue engineering. Combinations of cells with miRNA gene therapy in tissue engineering may bring considerable advances in the treatment of arthritis. Osteogenic differentiation of hMSCs in a three-dimensional scaffold was promoted by an miR-148b mimic and an miR-489 inhibitor [108]. Through overexpressing miR-1 and miR-206 and inhibiting the expression of miR-133, skeletal muscle cell differentiation was improved in three-dimensional bioartificial muscle constructs [109].

Conclusions and Perspective
As a progressive degenerative joint disorder, OA is characterized by cartilage damage, changes in the subchondral bone, osteophyte formation, muscle weakness, and inflammation of the synovium tissue and tendon. Based on the functions of miRNAs in bone and cartilage, and regulation in the development of OA, miRNAs may provide a novel and efficient method for the regulation of gene expression. Understanding the mechanism of expression and dynamic regulation of miRNAs could be key to promoting chondrogenesis and osteogenesis and to maintaining the balance of bone and cartilage homeostasis to prevent OA progressing, which, in turn, could lead to treating OA directly by preventing tissue degradation and stimulating repair.
It is clear that miRNAs have profound effects on the development of OA, and altered expression of miRNAs in the different stages of OA can be identified in vitro, so miRNAs may be useful in the diagnosis of OA as a biomarker. This may be the key to a radical therapy for OA, by regulating the expression of miRNAs. All of these ideas need further advanced studies, more evidence of the networks of miRNAs in arthritis, and more knowledge about gene interference technology in vivo to provide new strategies for OA treatment.
Studies on miRNAs have opened a new and exciting vista of gene regulation and offered a new understanding of the mechanisms controlling bone and cartilage development, homeostasis, degradation, and remodeling. Thus, the combination of biological reagents and antiarthritis drugs, together with miRNA treatment, may be more effective and beneficial than traditional treatments. However, the regulation of miRNAs might have carcinogenic effects, given the role of miRNAs as oncogenes. Thus, the side effects of miRNA use in the clinical setting must be considered, and sufficient examinations of miRNA therapy should be performed before clinical use.
Quan-Yi Guo, Jiang Peng, Ai-Yuan Wang, and Shi-Bi Lu approved the final version to be published.