Osteoarthritis is a multifactorial disease characterized by destruction of the articular cartilage due to genetic, mechanical and environmental components affecting more than 100 million individuals all over the world. Despite the high prevalence of the disease, the absence of large-scale molecular studies limits our ability to understand the molecular pathobiology of osteoathritis and identify targets for drug development. In this study we integrated genetic, bioinformatic and proteomic approaches in order to identify new genes and their collaborative networks involved in osteoarthritis pathogenesis. MicroRNA profiling of patient-derived osteoarthritic cartilage in comparison to normal cartilage, revealed a 16 microRNA osteoarthritis gene signature. Using reverse-phase protein arrays in the same tissues we detected 76 differentially expressed proteins between osteoarthritic and normal chondrocytes. Proteins such as SOX11, FGF23, KLF6, WWOX and GDF15 not implicated previously in the genesis of osteoarthritis were identified. Integration of microRNA and proteomic data with microRNA gene-target prediction algorithms, generated a potential “interactome” network consisting of 11 microRNAs and 58 proteins linked by 414 potential functional associations. Comparison of the molecular and clinical data, revealed specific microRNAs (miR-22, miR-103) and proteins (PPARA, BMP7, IL1B) to be highly correlated with Body Mass Index (BMI). Experimental validation revealed that miR-22 regulated PPARA and BMP7 expression and its inhibition blocked inflammatory and catabolic changes in osteoarthritic chondrocytes. Our findings indicate that obesity and inflammation are related to osteoarthritis, a metabolic disease affected by microRNA deregulation. Gene network approaches provide new insights for elucidating the complexity of diseases such as osteoarthritis. The integration of microRNA, proteomic and clinical data provides a detailed picture of how a network state is correlated with disease and furthermore leads to the development of new treatments. This strategy will help to improve the understanding of the pathogenesis of multifactorial diseases such as osteoarthritis and provide possible novel therapeutic targets.

Objective: To investigate whether epigenetic mechanisms can regulate leptin’s expression and affect its downstream targets as metalloproteinases 3,9,13 in osteoarthritic chondrocytes. Methods: DNA methylation in leptin promoter was measured by DNA bisulfite sequencing, and mRNA expression levels were measured by real-time quantitative PCR in osteoarthritic as well as in normal cartilage. Osteoarthritic articular cartilage samples were obtained from two distinct locations of the knee (n = 15); from the main defective area of maximum load (advanced osteoarthritis (OA)) and from adjacent macroscopically intact regions (minimal OA). Using small interference RNA, we tested if leptin downregulation would affect matrix metalloproteinase (MMP)-13 activity. We also evaluated the effect of the demethylating agent, 5′-Aza-2-deoxycytidine (AZA) and of the histone deacetylase inhibitor trichostatin A (TSA) on leptin expression in chondrocyte cultures. Furthermore, we performed chromatin immunoprecipitation in leptin’s promoter area. Results: We found a correlation between leptin expression and DNA methylation and also that leptin controls MMP-13 activity in chondrocytes. Leptin’s downregulation with small interference RNA inhibited MMP-13 expression dramatically. After 5-AZA application in normal chondrocytes, leptin’s methylation was decreased, while its expression was upregulated, and MMP-13 was activated. Furthermore, TSA application in normal chondrocyte cultures increased leptin’s expression. Also, chromatin immunoprecipitation in leptin’s promoter after TSA treatment revealed that histone H3 lysines 9 and 14 were acetylated. Conclusion: We found that epigenetic mechanisms regulate leptin’s expression in chondrocytes affecting its downstream target MMP-13. Small interference RNA against leptin deactivated directly MMP-13, which was upregulated after leptin’s epigenetic reactivation, raising the issue of leptin’s therapeutic potential for osteoarthritis.

MicroRNAs (miRNAs) in circulation have emerged as promising biomarkers. In this study, we aimed to identify a circulating miRNA signature for osteoarthritis (OA) patients and in combination with bioinformatics analysis to evaluate the utility of selected differentially expressed miRNAs in the serum as potential OA biomarkers. Serum samples were collected from 12 primary OA patients, and 12 healthy individuals were screened using the Agilent Human miRNA Microarray platform interrogating 2549 miRNAs. Receiver Operating Characteristic (ROC) curves were constructed to evaluate the diagnostic performance of the deregulated miRNAs. Expression levels of selected miRNAs were validated by quantitative real-time PCR (qRT-PCR) in all serum and in articular cartilage samples from OA patients (n = 12) and healthy individuals (n = 7). Bioinformatics analysis was used to investigate the involved pathways and target genes for the above miRNAs. We identified 279 differentially expressed miRNAs in the serum of OA patients compared to controls. Two hundred and five miRNAs (73.5%) were upregulated and 74 (26.5%) downregulated. ROC analysis revealed that 77 miRNAs had area under the curve (AUC) > 0.8 and p < 0.05. Bioinformatics analysis in the 77 miRNAs revealed that their target genes were involved in multiple signaling pathways associated with OA, among which FoxO, mTOR, Wnt, pI3K/akt, TGF-β signaling pathways, ECM-receptor interaction, and fatty acid biosynthesis. qRT-PCR validation in seven selected out of the 77 miRNAs revealed 3 significantly downregulated miRNAs (hsa-miR-33b-3p, hsa-miR-671-3p, and hsa-miR-140-3p) in the serum of OA patients, which were in silico predicted to be enriched in pathways involved in metabolic processes. Target-gene analysis of hsa-miR-140-3p, hsa-miR-33b-3p, and hsa-miR-671-3p revealed that InsR and IGFR1 were common targets of all three miRNAs, highlighting their involvement in regulation of metabolic processes that contribute to OA pathology. Hsa-miR-140-3p and hsa-miR-671-3p expression levels were consistently downregulated in articular cartilage of OA patients compared to healthy individuals. A serum miRNA signature was established for the first time using high density resolution miR-arrays in OA patients. We identified a three-miRNA signature, hsa-miR-140-3p, hsa-miR-671-3p, and hsa-miR-33b-3p, in the serum of OA patients, predicted to regulate metabolic processes, which could serve as a potential biomarker for the evaluation of OA risk and progression.

Altered lipid metabolism has been implicated as a critical player in osteoarthritis (OA). Our study aimed to investigate the expression of genes regulating cholesterol efflux in human chondrocytes and to study the effect of an LXR agonist on cholesterol efflux and lipid accumulation in osteoarthritic chondrocytes. ATP‐binding‐cassette transporter A1 (ABCA1), apolipoprotein A1 (ApoA1), and liver X receptors (LXRα and LXRβ) mRNA expression levels were evaluated using real‐time polymerase chain reaction (PCR) and ApoA1 protein levels by Western blot analysis in normal and osteoarthritic articular cartilage samples. Cholesterol efflux was evaluated in osteoarthritic chondrocytes radiolabeled with [1,2(n)‐³H] cholesterol after LXR treatment, while intracellular lipid accumulation was studied after Oil‐red‐O staining. Cholesterol efflux gene expressions were significantly lower in osteoarthritic cartilage compared to normal. Treatment of osteoarthritic chondrocytes with the LXR agonist TO‐901317 significantly increased ApoA1 and ABCA1 expression levels, as well as cholesterol efflux. Additionally, osteoarthritic chondrocytes presented intracellular lipids deposits, while no deposits were found after treatment with TO‐901317. Our findings suggest that impaired expression of genes regulating cholesterol efflux may be a critical player in osteoarthritis, while the ability of the LXR agonist to facilitate cholesterol efflux suggests that it may be a target for therapeutic intervention in osteoarthritis. © 2010 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 28:1033–1039, 2010

Introduction: Events normally taking place in the terminal chondrocyte differentiation in the growth plate are also observed during osteoarthritis (OA) development, suggesting that molecules, such as Wnts and bone morphogenetic proteins (BMPs) regulating chondrocyte activity in the growth plate, may play a key role in osteoarthritis pathogenesis. The aim of the study was to investigate the possible cross-talk between BMP-2 and Wnt/β-catenin pathways in OA progression. Methods: Low-density-lipoprotein receptor-related protein 5 (LRP-5) and 6, BMP-2, -4, and -7, bone morphogenetic protein receptor-IA and IB (BMPR-IA and BMPR-IA), lymphoid enhancer factor-1 (LEF-1), and transcription factor 4 (TCF-4) expression levels were investigated in normal and osteoarthritic chondrocytes. LRP-5, β-catenin (phospho and active form), matrix metalloproteinases (MMPs) 7, 9, 13, 14, ADAMTS-4, 5, as well as collagen X (COL10A1) expression levels were evaluated after LRP-5 silencing in BMP-2-treated chondrocytes. The investigation of Smad1/5/8 binding to LRP-5 promoter was assessed with chromatin immunoprecipitation (ChIP). Furthermore, we evaluated the effect of experimental activation of the Wnt/β-catenin pathway with LiCl and LEF-1 silencing, in LiCl-treated chondrocytes, on matrix metalloproteinases (MMPs) 7, 9, 13, 14, ADAMTS-4, 5, and collagen X (COL10A1) expression, as well as possible interactions between LEF-1 and MMPs and COL10A1 promoters by using a ChIP assay. Results: LRP-5, BMP-2, BMP-4, BMPR-IA, and LEF-1 mRNA and protein expression levels were found to be significantly upregulated in osteoarthritic chondrocytes compared with normal. We showed that treatment of cultured chondrocytes with BMP-2 resulted in increased β-catenin nuclear translocation and LRP-5 expression and that the BMP-2-induced LRP-5 upregulation is mediated through Smad1/5/8 binding on LRP-5 promoter. LRP-5 silencing reduced nuclear β-catenin protein levels, MMPs and collagen X expression, whereas increased phospho-β-catenin protein levels in BMP-2-treated chondrocyte. Furthermore, we demonstrated that activation of the Wnt/β-catenin signaling pathway by LiCl and LEF-1 downregulation by using siRNA regulates MMP-9, 13, 14, ADAMTS-5, and COL10A1 expression, evidenced by the observed strong binding of LEF-1 to MMP-9, 13, 14, ADAMTS-5 and COL10A promoters. Conclusions: Our findings suggest, for the first time to our knowledge, that BMP-2-induced Wnt/β-catenin signaling activation through LRP-5 may contribute to chondrocyte hypertrophy and cartilage degradation in osteoarthritis.