Pharmacologic iron chelation reduces markers of chondrocyte hypertrophy and osteoarthritis-associated cartilage lesions in an animal model of idiopathic disease

Abstract

Purpose: It has previously been documented that chondrocytes from joints afflicted with osteoarthritis (OA) undergo improper dedifferentiation into hypertrophic chondrocytes; thus, it is postulated that this process may contribute to the progression of OA. Relative to unaffected adult chondrocytes, hypertrophic chondrocytes mimic cellular processes for endochondral ossification observed during initial development of long bones. In particular, they express markers of hypertrophic differentiation such as runx family transcription factor 2 (runx2), collagen type x, and the matrix degrading enzyme matrix metalloproteinase-13 (MMP-13). In addition to these markers, hypertrophic chondrocytes exhibit increased apoptosis, decreased autophagy, and promote cartilage vascularization and mineralization of the extracellular matrix. One theory for why this hypertrophic state occurs is that oxidative stress from reactive oxygen species (ROS) contributes to this shift in chondrocyte phenotype. We had previously demonstrated the ability of systemic iron reduction, achieved by an iron deficient diet, to reduce ROS production and thereby decrease structural changes of OA in an animal model of idiopathic/spontaneous disease. Given the success of this study, we hypothesized that systemic iron chelation - achieved by administration of the pharmacologic iron chelator deferoxamine (DFO) - would reduce markers of chondrocyte hypertrophy and decrease the development of knee OA in this animal model. Methods: All procedures received University approval and were performed in accordance with the NIH Guide for the Care and Use of Laboratory Animals. Three-month-old male Dunkin-Hartley guinea pigs were randomly assigned to receive either DFO or vehicle control (n=8 per group). Animals within the DFO group received 46 mg/kg of DFO injected subcutaneously twice daily while animals within the control group received an equivalent dose of saline. Animals were subjected to 10-minute sessions of open field behavior monitoring once per month for the duration of the study, with baseline activity levels collected prior to initiating treatment. The study was terminated when animals were 8-months-old, at which time body weights were recorded. Hind limbs were removed and the right tibia was measured using calipers. Articular cartilage was collected from the right knee joint for gene expression analysis using Nanostring nCounter technology. Femoral head cartilage and liver tissue were collected for iron quantification by atomic absorption spectroscopy. Left knee joints were formalin-fixed, paraffin-embedded, and stained with toluidine blue. For histologic analysis, all joint compartments were blindly scored by two assessors using OARSI published guidelines. Data were compared using non-parametric Mann-Whitney U tests, Welch’s t-tests, or unpaired student’s t-tests as appropriate given the normality and heteroscedasticity of the data. All statistical analyses were performed with Prism 8.0; statistical significance was set at P≤0.05. Results: No significant differences in body weight (p=0.576) or tibia length (p=0.628) were present between groups. As anticipated, animals treated with DFO had lower levels of iron within the liver (p=0.0218) and articular cartilage (p=0.0005) tissue than control animals. Histologic grading of knee joints confirmed that animals treated with DFO had lower total joint OA scores than control animals (p<0.0001). Gene transcript analysis demonstrated that, relative to controls, animals treated with DFO had lower expression of runx2 (p=0.0010), MMP-13 (p=0.0151), vascular endothelial growth factor (VEGF; p=0.0140), β- catenin (p<0.0001), wnt (p=0.0008), mammalian target of rapamycin (mTOR; p=0.0018), and the serine/threonine protein kinase AKT (p=0.0041). Animals treated with DFO also had increased expression of b cell lymphoma 2 (BCL-2; p<0.0001), a prototypical anti-apoptotic marker. There was no significant difference in the gene expression of collagen type x between groups (p=0.3163). Open field behavior monitoring indicated that DFO treated animals maintained their baseline activity level throughout the study, while the movement of control animals declined throughout the study. Conclusions: Administration of a pharmacologic iron chelator was successful in reducing iron levels both systemically and within articular cartilage tissue, which was associated with a reduced development of OA-associated cartilage lesions and maintained mobility in these animals. Gene transcript analysis of cartilage tissue revealed that DFO treated animals had a decreased expression of: select markers of chondrocyte hypertrophy (runx2 and MMP-13), negative regulators of autophagy (AKT and mTOR), the angiogenesis promoter VEGF, and genes associated with chondrocyte dedifferentiation (β- catenin and wnt); while displaying increased expression of the anti-apoptotic protein BCL-2. These results indicate that removal of excess iron may help prevent chondrocyte hypertrophy and development of OA-related changes within the cartilage of knee joints.