OA is the most common disease of joints in adults around the world [41
], being the knee OA the most common type. About one-third of all adults have radiological signs of OA. The likelihood of developing OA increases with age. The prevalence of OA of the knee is higher among 70- to 74-year-old. Epidemiological studies have revealed that there are both endogenous and exogenous risk factors for OA. The pathogenesis of OA is characterized by the progressive destruction of articular cartilage. Many factors contribute to the overall degradation of cartilage observed in OA, either directly or indirectly by modulating anabolic factors [14
]. The capacity of articular cartilage to repair is very limited, mainly because it is an avascular tissue [42
]. Currently, there are no effective pharmacological treatments to treat OA although some drugs slow its progression. Surgical treatments are not a way to treat OA even so they still constitute an important tool in cartilage repair of those injuries which if not treated or repaired properly could inevitably end up producing a secondary OA.
Some miRNA-related studies on cancers and genetic diseases are based on their different expression profile in pathologic compared to normal tissues [19
]. Most of these studies postulate that differential expression of candidate miRNAs is a possible approach to investigate the function of miRNAs in human diseases. In attempting to understand the biological pathways and processes that underlie the pathogenesis and progression of OA, genomic approaches have identified miRNAs genes associated with cartilage development and homeostasis, extracellular matrix components and IL-1 signal transduction pathway among others [11
]. Tissue miRNAs have been noted not only as key molecules in intracellular regulatory networks for gene expression, but also as biomarkers for various pathological conditions [45
]. For example, the expression of miR-146 has been shown to be intensely expressed in low grade OA cartilage, indicating that it might play a role in OA cartilage pathogenesis [12
]. In addition, it has been demonstrated that miR-27b regulates the expression of MMP13 (Matrix Metalloproteinase 13) in human OA chondrocytes, suggesting that up-regulation of miR-27b in vivo would represent a novel therapeutic approach in OA [44
]. In recent years, a putative role for miR-140 in the pathogenesis of OA was evidenced [11
], since its expression is significantly reduced in OA tissue and that in vitro treatment of chondrocytes with IL-1ß, a cytokine involved in the pathogenesis of OA, suppresses miR-140 expression [46
]. Recent studies have also demonstrated that several miRNAs might play a role in OA pathogenesis [9
In the present study, to better understand the molecular mechanisms involved in the pathogenesis of OA and to investigate a possible role of miRNAs in cartilage-related genes regulation and OA development, we comprehensively isolated and analyzed miRNAs of normal and OA chondrocyte micropellets, using miRNA microarray analysis. From a technical point of view, it is difficult to obtain a high number of chondrocytes from articular cartilage explants. Chondrocytes from a healthy cartilage represent only the 2% of the total volume of the cartilage. Therefore it is also complicated obtain large amounts of RNA, enriched in miRNAs, of excellent quality to performed the microarrays. This difficulty is even higher when using OA cartilage samples, since OA cartilage has a smaller number of cells than the healthy one. The micropellet model is different to the tissue, since it allows maintaining the cells in a three-dimensional position, these cells can synthesize extracellular matrix and allows obtaining a greater number of chondrocytes. For this reason chondrocytes micropellets, and not the cartilage explants, were employed for miRNAs isolation.
This study identified and characterized the expression profiles of 723 human miRNAs from normal and OA chondrocytes, of which 1 miRNA up-regulated in OA chondrocyte and 6 were up-regulated in normal chondrocyte micropellets. Unsupervised clustering performed by using processed data from miRNA microarray analysis highlighted differential expression profiles of 48 miRNAs, interestingly clustering the samples into 2 groups, OA versus normal chondrocyte micropellets. Therefore these miRNAs could represent valid markers in discriminating normal versus OA chondrocyte samples although further studies focused in a large number of samples should be performed to determine the potential of these miRNAs for clinical application in the diagnosis of OA pathology.
Our profiling results were further validated by the detection of some selected miRNAs by qPCR. Some of these selected miRNAs (e.g. hsa-miR-145 and hsa-miR-483-5p) have already been described in the literature [15
]. Of particular interest was the finding that hsa-miR-483-5p was up-regulated in OA chondrocyte micropellets as previously described Iliopoulus et al., [15
]. In this regard, Iliopoulos et al. [15
] reported their finding of a 16-miRNA OA gene signature from their studies comparing osteoarthritic and nondiseased human cartilage. These authors found that hsa-miR-483-5p was upregulated in OA cartilage, not only by miRNA microarray analysis but also by qPCR techniques. These findings are in agreement with our miRNA microarray and qPCR results since we observed an upregulation of hsa-miR-483-5p in OA chondrocyte micropellets with the highest fold (8.45) obtained by qPCR. On the other hand, Zuntini et al. [40
] also verified that hsa-miR-145 and hsa-miR-483 are both upregulated in osteochondromas when they are compared to normal cartilage. A recent study postulated that aberrant expression miR-483-5p together with miR-195 allow the identification of a subset of poorer prognosis adrenocortical carcinomas [49
]. Moreover, Patterson et al. [50
] found that the high expression of miR-483-5p appears to be a defining characteristic of adrenocortical malignancies, indicating that it can thus be used to accurately distinguish between benign and malignant adrenocortical tumors. However Dunn et al. [48
], profiling miRNA expression in bovine articular cartilage, found that hsa-miR-145 were downregulated in monolayers of tissue cultured chondrocytes as compared with levels determined directly from intact native cartilage. Our microarray analyses showed that the relative expression levels for hsa-miR-145 were −2.87 for healthy and 1.85 for OA samples. Moreover, qPCR experiments showed that this miRNA was also up regulated in OA donors, in particular 4.4 fold. However neither the microarray nor the qPCR results achieve the statistical significance previously published in the literature. Perhaps it could be due to the use of different microarray technologies, or to the use of cultured cell instead of tissue samples.
In our study miR-149* was down regulated in OA chondrocyte micropellets, in agreement with a recent study published by Jones et al. [9
]. These authors, studying the expression profiles of 157 human miRNA, identified 17 differentially expressed miRNAs in human OA in comparison to normal cartilage and they determined their relevance to chondrocyte function. In this sense, they postulated that miR-149 was downregulated in OA cartilage, this result is in agreement with our miRNA microarray analysis regarding miR-149, which was also downregulated in OA chondrocyte micropellets.
In previous reports hsa-miR-140 was down regulated and hsa-miR-146 was up regulated in OA cartilage [9
]. In our study, the expression levels of hsa-miR-140 and hsa-miR-146 in the microarray analyses showed no statistical significant differences when comparing healthy and OA samples. However, hsa-miR-140 showed a tendency to be down regulated in OA and hsa-miR-146 showed a tendency to be up regulated in this pathology. Such discrepancies found among our results and those published on the literature could be due to the use of different microarray technologies, or to the use of cultured cell instead of tissue samples, or to the use of OA samples obtained from the different zones of the cartilage.
It is noteworthy that some of the miRNAs differentially expressed in chondrocyte that we identified in our study are novel as compared with those identified and published in the literature, e.g. hsa-miR-576-5p, hsa-miR-582-3p, hsa-miR-634, hsa-miR-641, hsa-miR-1227, suggesting that they may therefore represent new targets in articular cartilage.
The key molecular pathways potentially altered by the miRNAs differentially expressed in normal and OA chondrocyte micropellets, as predicted by the DIANA-mirPath web-based computational tool, include TGF-beta, Wnt, MAPK and mTOR signalling, focal adhesion and regulation of actin cytoskeleton among others. These results should be considered since Wnt signalling pathway has a role in OA pathology [51
]. In particular these pathways are key inducers and regulators of joint development, and are involved in formation of bone, cartilage and also synovium [53
]. For these reasons Wnt-family of proteins and signalling pathways are attractive targets for OA therapy. In this sense, products of the Wnt, frizzled, secreted frizzled-related protein (sFRP), Dickkopf and LDL-receptor-related protein gene families have crucial roles in the development and maintenance of bone, cartilage and joints [54
]. In this sense, genes of the Wnt pathway are upregulated in the OA bone, suggesting their involvement not only in cartilage distortion but also in subchondral bone changes [55
]. On the other hand, TGF-beta 1 induces cartilage extracellular matrix synthesis and tissue inhibitor of metalloproteinases-3 (TIMP-3), an important natural inhibitor of matrix metalloproteinases, aggrecanasses and TNF-alpha-converting enzyme, which are implicated in cartilage degradation and joint inflammation [56
]. Moreover, genes belonging to the TGF-beta signalling pathway, which are supposedly targeted by the miRNAs differentially expressed in our work, regulate the chondrocyte differentiation and potentially the OA development [57
]. Also, TGF-beta pathway regulates the expression of the superficial zone protein (SZP) in the superficial zone chondrocytes, protein implicated in the lubrication of the articular cartilage surface [58