Celecoxib modulates release of different prostanoid subtypes by intra-articular tissues: towards celecoxib based intra-articular treatment for OA modulation?

Abstract

Purpose: Celecoxib is currently used as an analgesic in knee osteoarthritis (OA) patients with proven efficacy. Besides solely being a drug with analgesic properties, evidence on the OA disease modifying properties of celecoxib is accumulating. On the other hand, other studies do not confirm OA disease modifying properties of celecoxib. However, these studies evaluated effects of orally administered celecoxib, which not surprisingly will result in lower local concentrations of celecoxib in the knee joint. In this study we hypothesized that celecoxib can be used as an OA-modulatory drug when injected intra-articularly in the knee joint. Since intra-articular administration of celecoxib means that the drug will interact with different intra-articular tissues such as the cartilage, synovium, Hoffa's Fat Pad (HFP) and meniscus we first evaluated the ability of celecoxib to modulate prostanoid release from these intra-articular tissues. Furthermore, we investigated whether prostanoid-reducing actions of celecoxib on different intra-articular tissues is accompanied by a modulation of the expression of aggrecanases, which are important proteolytic enzymes involved in OA pathophysiology. Finally, we sought to determine potential chondroprotective effects of a single intra-articular celecoxib bolus injection in the rat ACLT/pMMx model for OA. Methods: Upon medical ethical approval, cartilage, synovium, HFP and meniscus were obtained after total knee replacement from nine patients with knee OA. Explants were generated and cultured at a ratio of 100 mg tissue per mL culture medium for 24 hours in the presence or absence of 10 μM celecoxib. Presence of prostaglandin subtypes PGE2, PGF2α, PGD2 and (thromboxane B2) TXB2 was determined quantitatively by specific ELISAs. Gene expression levels of ADAMTS4 and ADAMTS5 was evaluated in the intra-articular tissues by means of RT-qPCR. For in vivo experiments OA was induced by performing anterior cruciate ligament transection (ACLT) + partial medial menisectomy (pMMx) in the right knees of 14 male Lewis rats (12-weeks old). Four weeks after OA induction, both the OA-induced and contralateral healthy knees were injected with 25 μL of 0.9 % NaCl or 9.7 nM celecoxib (92.25 ng celecoxib). Twelve weeks after injection, rats were sacrificed and OA severity was assesed by two blinded observers using the OARSI histopathology initiative for the rat (n = 7 per experimental group). Results: Average concentration ranges for the different prostanoids were 26216-385039 pg/mL for PGE2, 5496-69,574 pg/mL for PGF2α, 3848-43830 for PGD2 and 388-9102 pg/mL for TXB2. Celecoxib, at a dose of 10 μM, was able to significantly reduce PGE2, PGF2α, PGD2 and TXB2 release in cartilage, synovium, HFP and meniscus. Average fold reduction in prostanoid release for different intra-articular tissues tested ranged from 25- to 62-fold for PGE2, 9-fold to 20-fold for PGF2α, 8-fold to 23-fold for PGD2 and 2- to 5-fold for TXB2. Celecoxib significantly reduced gene expression levels of both ADAMTS4 and ADAMTS5 in cartilage, synovium, HFP and meniscus. OA levels based upon OARSI scoring were significantly reduced in OA-induced knees treated with bolus celecoxib (P = 0.009 by observer 1 and P = 0.032 by observer 2) compared to OA-induced knees injected with 0.9% NaCl (Fig. 1). Conclusions: Different intra-articular tissues secrete various prostanoid subtypes. These prostanoid subtypes are involved in catabolic processes and celecoxib is able to reduce the release of these prostanoids from cartilage, synovium, HFP and meniscus, but is also able to reduce the expression of aggrecanases which are important enzymes involved in cartilage matrix degradation. A single bolus injection with celecoxib protects against OA progression in vivo and therefore this study reinforces the potential of celecoxib as a potential OA disease-modifying drug when used intra-articularly.