Endogenous adenosine maintains cartilage homeostasis and exogenous adenosine inhibits osteoarthritis progression

Abstract

Osteoarthritis (OA) is characterized by cartilage destruction and chondrocytes have a central role in this process. With age and inflammation chondrocytes have reduced capacity to synthesize and maintain ATP, a molecule important for cartilage homeostasis. Here we show that concentrations of ATP and adenosine, its metabolite, fall after treatment of mouse chondrocytes and rat tibia explants with IL-1β, an inflammatory mediator thought to participate in OA pathogenesis. Mice lacking A2A adenosine receptor (A2AR) or ecto-5'nucleotidase (an enzyme that converts extracellular AMP to adenosine) develop spontaneous OA and chondrocytes lacking A2AR develop an 'OA phenotype' with increased expression of Mmp13 and Col10a1. Adenosine replacement by intra-articular injection of liposomal suspensions containing adenosine prevents development of OA in rats. These results support the hypothesis that maintaining extracellular adenosine levels is an important homeostatic mechanism, loss of which contributes to the development of OA; targeting adenosine A2A receptors might treat or prevent OA.

Figure 1. Osteoarthritis in A2AR KO mice.

(a) Open field and rotarod tests were carried out in 25-week-old WT and A2AR KO mice (n=5 for each group) and the digital record analysed, as described in Materials and Methods. Shown are the mean (±s.e.m.) values obtained during each 10 min interval for distance moved, immobility time, rearing frequency, velocity of motion and time spent in the border and the latency to fall in the rotarod test. (b) Immunohistologic staining for Collagen X and MMP-13 in a representative section of tibia from a WT and A2AR KO mouse (12 week old mice, original magnification 400 X). Sections from the same mice stained with Safranin O and hematoxylin and eosin (H&E) are shown in the right-hand panels (original magnification 100 X). In the lower panel is plotted the OARSI scores obtained on safranin-O-stained knees, as described in Materials and Methods. Each data point represents the mean (±s.e.m.) of the blinded scores obtained on 4–5 different mice. *P<0.05, **P<0.01, ***P<0.001 versus WT (Student T-test or one-way analysis of variance followed by Bonferroni post hoc test, as appropriate).

Figure 2. Expression of osteoarthritis markers in primary A2AR-KO and WT chondrocytes.

(a) Primary chondrocytes were isolated from neonatal WT and A2AR-KO mice. RNA was isolated, reverse transcribed and analysed by real time PCR. Shown are expression levels normalized to GAPDH for Mmp13 and Col10a1 levels when compared to levels in WT cells. (b) Representative western blots are shown for MMP-13 and collagen X, and quantification of protein bands in the bar graphs below (n=4). (c) Staining of cell layers shows increases in MMP-13 and collagen X in the A2AR KO chondrocytes compared to the WT. (n=4; data are represented as means±s.e.m.). *P<0.05; **P<0.01 versus WT; Student's T-test.

Figure 3. ATP and adenosine decrease after treatment with IL-1β.

Treatment of primary murine chondrocytes with IL-1β (5 ng ml⁻¹ for 24 h) decreases: (a) The intracellular ATP content and the concentration of ATP in the supernatant medium. (b) The concentration of adenosine in the supernatant medium. Data are expressed as mean±s.e.m. of 9 experiments performed in duplicate for the adenosine determination and 3 experiments perfomed in duplicate for the ATP assay. Student's T-test was performed; *P<0.05, **P<0.01). (c) Adenosine levels in supernatants of control and IL-1β-treated rat tibial plateau explants (left) and following mechanical loading of the tibial plateau explants (right, data were analysed for statistical significance by two-way analysis of variance followed by Bonferroni post hoc test *P<0.05). The data are expressed as the mean±s.e.m. of five experiments for adenosine release and the adenosine concentration was normalized to the weight of the tibial explant. (d) ATP concentration in supernates of tibial explants was measured as described. Each point represents the mean±s.e.m. of three separate determinations.

Figure 4. NT5E decreases after IL-1β treatment and NT5E KO mice manifest mild OA.

(a) 24 h treatment with IL-1β decreases NT5E expression in the total protein content and NT5E membrane expression on the cell membrane. Representative NT5E western blot and protein quantification is shown on the graph on the left side of the figure and on the right are shown representative immunofluorescence photomicrographs of NT5E (green fluorescence) and phalloidin staining (red fluorescence). (b) Shown are representative images of knees from 1-year-old NT5E KO mice with cartilage fraying, minimal loss of proteoglycan and increased MMP-13 expression by immunohistochemistry. The black arrow in the safranin-O-stained section indicates an osteophyte.

Figure 5. Adenosine prevents and treats OA in a PTOA rat model via A2AR ligation.

(a) Shown is the experimental design with anterior cruciate ligament disruption on day 0 followed by injection of liposomal adenosine at the indicated times. (b) Representative photographs of the gross appearance of the knees of the rats at the time of killing (top row femur, bottom row tibia) and on the bottom is shown knee size measured with a caliper immediately before injection. (c) Representative μCT images of hexabrix-imaged cartilage (Top row, femur; bottom row, tibia). The cartilage is pink in this image and underlying bone is grey. In the panel beneath is shown the mean (±s.e.m.) volume of cartilage present in the affected knee expressed as the percentage of the volume of cartilage in the unaffected knee. S, saline-injected; L, empty liposome-injected; A, adenosine-liposome injected. (d) Shown are representative safranin-O-stained sections and immunohistologic sections for MMP-13 expression in rat tibial plateaus following ACL rupture. Graphs show the OARSI scores of the knees of the rats studied here. (e) Representative gross photograph and photomicrographs of rat knee injected with liposome formulation containing adenosine plus ZM241385 and respective safranin-O staining and MMP-13 immuhohistochemistry. Graphs show the OARSI scores for rat knees treated with adenosine plus adenosine receptor antagonists. Data are expressed as mean±s.e.m. of 5–6 animals for each group and data were analysed for statistical significance by one-way analysis of variance followed by Bonferroni post hoc test of differences among various treatments (*P<0.05; **P<0.01, ***P<0.001).

Figure 6. Endogenous adenosine maintains cartilage homeostasis.

Here we provide a conceptual model of the role of adenosine production and adenosine receptor ligation in inflammation and in the pathogenesis of OA. Adenosine is generated which ligates A2AR maintaining homeostasis in cartilage after mechanical loading and other stresses. A2AR ligation exerts an anti-inflammatory and chondroprotective effect by blocking NF-κB nuclear translocation and promoting production of growth factors. During OA progression a decrease of adenosine in the extracellular space occurs due partially to IL-1β effect on ATP and adenosine transporters and the enzymes involved in ATP/ADP catabolism. Moreover, decreased mitochondrial number, observed in OA chondrocytes, contributes to diminished cellular ATP levels. Moreover, production and release of adenosine deaminase by inflammatory cells can exacerbate the progression of OA by further reduction of adenosine levels (unpublished figure created by C.C.).