GSK2193874

Sensitization of transient receptor potential vanilloid 4 and increasing its endogenous ligand 5,6-epoxyeicosatrienoic acid in rats with monoiodoacetate-induced osteoarthritis

ABSTRACT
Transient receptor potential vanilloid 4 (TRPV4) receptor modulates pain, and this has been noted in several animal models. However, the involvement of TRPV4 in osteoarthritic (OA) pain remains poorly understood. The present study assessed the functional changes in TRPV4 and the expression of its endogenous ligand 5,6-epoxyeicosatrienoic acid (5,6-EET) in a rat monoiodoacetate (MIA)-induced OA pain model (MIA rats). MIA-treated rats showed reduced grip strength as compared with sham-treated rats, and this loss in function could be recovered by the intra-articular administration of a TRPV4 antagonist (HC067047 or GSK2193874). In contrast, the intra-articular administration of the TRPV4 agonist, GSK1016790A, increased the pain-related behaviors in MIA rats but not in sham rats. TRPV4 expression was not increased in knee joints of MIA rats; however, the levels of phosphorylated TRPV4 at Ser824 were increased in dorsal root ganglion (DRG) neurons. Additionally, 5,6-EET was increased in lavage fluids from the knee joints of MIA rats, and in meniscectomy-induced OA pain model rats. 5,6-EET and its metabolite were also detected in synovial fluids from OA patients. In conclusion, TRPV4 was sensitized in the knee joints of MIA rats through phosphorylation in DRG neurons, along with an increase in the levels of its endogenous ligand 5,6-EET. The analgesic effects of the TRPV4 antagonist in the OA pain model rats suggest that TRPV4 may be a potent target for OA pain relief.

INTRODUCTION
Osteoarthritis (OA) of the knee is one of the most common diseases in older people, and is characterized by cartilage loss and structural changes in the subchondral bone [18,40]. The main symptom of knee OA is joint pain, which causes locomotor impairment and reduced quality of life. Several underlying mechanisms for knee joint pain have been proposed, including increases in pro-inflammatory cytokines [37], lipid mediators [24,28], or nerve growth factor (NGF) [23], and subchondral bone attrition associated with bone marrow lesions [50]. Although current therapies for pain management can relieve knee joint pain in patients with OA, the analgesic effects are limited. Anti-NGF antibody is a promising drug for OA pain but safety issues have beenraised, limiting the use of this therapy [12]. As such, novel target molecules and new therapies are required for the better management of knee joint pain in patients with OA. Transient receptor potential vanilloid 4 (TRPV4) is a nonselective cation channel belonging to the TRPV family, and is gated by mechanical stimuli [30,44], warm temperatures [49], hypoosmotic stress [29], and arachidonic acid metabolites like 5,6-epoxyeicosatrienoic acid (5,6-EET) [48]. TRPV4 is expressed ubiquitously, including in the dorsal root ganglion (DRG) [16], kidney [46], bladder [5], vascular endothelium [32], and cartilage [14]. TRPV4 has been implicated in some types of pain, such as sunburn pain [33], temporomandibular arthritic pain [13], and neuropathic pain[2] or in the maintenance of the cartilage [14], after suggestion of its physiological roles in the first time [29].

However, the involvement of TRPV4 in OA pain has not been examined. Mon iodoacetate (MIA)-induced OA pain model rats (MIA rats) exhibit similar clinical features to knee OA, including increased cytokine expression [36], joint degeneration and cartilage loss [27], and some features of joint pain, such as movement-induced pain, identified by reduced grip strength [11], and resting pain, identified by reduced weight bearing on the ipsilateral hind limb [6]. Recently, we found that protein kinase C (PKC) was activated in DRG neurons of MIA rats and sensitized TRPV1 by promoting its phosphorylation, resulting in increased pain [27]. PKCfunctionally activates TRPV4 through its increased phosphorylation [17], and coexpression of TRPV4 and TRPV1 in DRG neurons has been reported [9,26]. Furthermore, patients with irritable bowel syndrome show increased expression of the TRPV4 endogenous ligand, 5,6-EET, in colon biopsies, and this expression is positively correlated with abdominal pain scores [10]. These findings motivated us to explore the function of TRPV4 in OA pain.In the present study, we investigated the function of TRPV4 in knee joint pain using MIA rats.

We first examined the effects of an agonist and two antagonists of TRPV4 in MIA rats versus sham rats, and then investigated changes in the phosphorylation of TRPV4 in the DRG. We further measured changes in the expression of its endogenous ligand 5,6-EET in the knee joints of OA model rats and in the synovial fluid samples from patients with OA.Animal experiments were performed using male Sprague–Dawley rats (CLEA, Tokyo, Japan) weighing 250 to 350 g. A total of 153 rats were used. The rats were housed in groups of three in plastic cages under controlled temperature and lighting (12-h/12-h light/dark cycle), and provided with food and water ad libitum. All animal behavioral tests were conducted during the light period. Protocols for the animal procedures were reviewed and approved by the Animal Care and Use Committee of Shionogi Pharmaceutical Research Center (Osaka, Japan), and met the Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC) guidelines. The results of the experiments are reported in accordance with the ARRIVE guidelines [25].Stable TRPV4-expressing CHO (TRPV4-CHO) cells were established by transfection of a rat TRPV4 cDNA (GenBank accession number NM_023970.1) in the pcDNA3.1(−) vector (Invitrogen, Carlsbad, CA) into CHO-K1 cells. The cells were maintained in DMEM (high-glucose) with GlutaMAX (Sigma-Aldrich, St. Louis, MO) supplementedwith 10% fetal bovine serum (Invitrogen), 0.1 mM MEM Non-Essential Amino Acid Solution (Gibco, Life Technologies, Carlsbad, CA), 25 mM HEPES, 1% Penicillin-Streptomycin Mixed Solution (Nacalai Tesque Inc., Kyoto, Japan), and 10 µg/mL Blasticidin (Gibco).Two TRPV4 antagonists, HC067047 and GSK2193874 [45], were synthesized in our laboratory and dissolved at 300 µM and 3 µM in 30% DMSO in saline, respectively.

The TRPV4 agonist, GSK1016790A (Sigma-Aldrich), was dissolved in 10% DMSO in saline.Collection of synovial fluids from patients with knee OA was conducted by Asterand Bioscience (Detroit, MI). The participants provided informed consent according to the Declaration of Helsinki. The study was approved by the Research Ethics Committee of Shionogi Pharmaceutical Research Center. Synovial fluids from the knees of patients with OA were immediately aliquoted and mixed with an equal volume of methanol, and then stored at −80°C. The inclusion criteria for the patients were: clinical diagnosis ofactive knee OA with chronic pain for >5 days/week for 3 months; no history of use of nonsteroidal anti-inflammatory drugs, selective COX2 inhibitors, or paracetamol in the previous 7 days; and no active communicable diseases.MIA rats were prepared according to our previously described method [27]. Rats received an intra-articular injection of MIA (Sigma-Aldrich) through the infrapatellar ligament of the right knee joint at a dose of 2 mg in 50 µL of saline. Sham-operated rats received an intra-articular injection of 50 µL of saline. The left knee joints remained untreated in all rats. All rats were anesthetized with isoflurane during the procedure. Pain development was assessed by the grip strength test [11,27]. Briefly, rats were allowed to grasp the wire mesh frame of the grip strength meter (San Diego Instruments, San Diego, CA) with their hind limbs and then moved in a rostral-to-caudal direction until the grip was released.

Grip strength measurements were conducted in a blinded manner, and the mean grip strength was averaged from three readings. The selection criterion for MIA rats was <950 grip strength (g)/body weight (kg). Rats were randomly allocated by their grip strengths. Pain-related behaviors induced by GSK1016790A orvehicle injection were measured by the duration of right leg withdrawal and lifting behavior over a period of 5 min in a Plexiglas cylinder (25-cm diameter; 30-cm height) by a blinded investigator. Meniscectomy-induced OA pain model rats (MNx rats) were prepared by removal of the medial meniscus and transection of the medial collateral ligament from the right knee joint [47]. After surgery, the joint capsule and skin were sutured. Sham-operated rats received only medial collateral ligament transection. Rats were used for experiments at 6 weeks after surgery.Rats were decapitated under deep anesthesia. The knee joint cavity was exposed carefully by dissecting the patellar ligament, and lavaged four times with 10 µL of ice-cold methanol. The lavage fluids were collected and stored at −80°C until analysis.Rats were decapitated under deep anesthesia. The knee joints were dissected, immersed in 4% paraformaldehyde in phosphate-buffered saline (PBS) (Nacalai Tesque Inc.) overnight at 4°C, embedded in optimal cutting temperature (OCT) compound (Sakura Finetek, Torrance, CA), and quickly frozen in liquid nitrogen. Sections (16-µm thickness) were cut using a cryostat (CM1850; Leica, Nussloch, Germany) and thaw-mounted onto glass slides (Matsunami Glass, Osaka, Japan). The sections were then incubated with 0.3% hydrogen peroxide in methanol for 30 min, blocked with 5% goat serum in 0.3% Triton X-100/PBS for 30 min, incubated with a rabbit anti-TRPV4 antibody (0.4 µg/ml; Enzo Life Sciences Inc., New York, NY) in blocking solution overnight at 4°C, and developed by the ABC method (VECTASTAIN ABC Kit; Vector Laboratories, Burlingame, CA) using diaminobenzidine as the substrate. After IHC staining, hematoxylin staining was performed.TRPV4-CHO cells cultured in 96-well plates were loaded with 5 µM Fluo-4 AM (Dojindo, Kumamoto, Japan) and 0.1% Pluronic F-127 (Molecular Probes, OR) in assay buffer (pH 7.4) comprising Hanks’ balanced salt solution (Nissui Pharmaceutical, Tokyo, Japan), 20 mM HEPES (Sigma-Aldrich), and 2.5 mM probenecid for 1 h at 37°C. Toexamine the effects of phorbol 12-myristate 13-acetate (PMA) (Invitrogen), the cells were washed with assay buffer and incubated with 3 µM PMA in assay buffer for 1 h at 37°C. Cells were washed twice with assay buffer, and then incubated in assay buffer for 10 min at 30°C. A FDSS 7000 functional drug screening system (Hamamatsu Photonics, Shizuoka, Japan) was used to measure calcium influx, with 480-nm excitation and 540-nm emission wavelengths. GSK1016790A was dissolved in 1% DMSO and added to the cells at a final concentration of 0.1% DMSO. The EC50 of the GSK1016790A-induced calcium flux was calculated from the concentration-response curve using GraphPad Prism 4.0 (GraphPad Software, San Diego, CA). DRG lysates were prepared by homogenizing the right L3 and L4 DRGs in cell lysis buffer (Cell Signaling Technology, Beverly, MA) containing a Complete Mini EDTA-free tablet and a PhosSTOP tablet (Roche, Indianapolis, IN). The lysates were centrifuged at 15,000×g for 5 min at 4°C, and the supernatants were collected. The protein concentrations were determined using a BCA Protein Assay-Reducing Agent Compatible Kit (Thermo Fisher Scientific, Yokohama, Japan).A polyclonal antibody against TRPV4 phosphorylated at Ser824 was raised in rabbits using KLH-conjugated CGRLRRDRWS(pS)VVPRVVE as the antigen (Sigma-Aldrich). Sera from each rabbit were processed by positive affinity purification (using the phosphorylated peptide), followed by negative affinity purification (using the nonphosphorylated peptide). Antibody specificity was confirmed by titration with phosphorylated or nonphosphorylated peptide-coated plates. A 96-well plate was coated with anti-TRPV4-phospho-Ser824 antibody (5 µg/mL) or anti-total-TRPV4-middle region antibody (5 µg/mL; Aviva Systems Biology Corporation, San Diego, CA) overnight at 4°C, and then blocked with 1% Block Ace (DS Pharma Biomedical, Suita, Japan) for 2 h at room temperature. After three washes with PBS containing 0.05% Tween-20 (PBST), 50 µL of protein lysate was added and incubated overnight at 4°C. After washing with PBST, 50 µL of Streptavidin-Poly-HRP (500 ng/mL; Thermo Scientific, Waltham, MA) and biotinylated anti-TRPV4 antibody (500 ng/mL; Enzo Life Sciences), prepared with NHS-PEG4-Biotin (Thermo Scientific), were added to each well and incubated for 2 h at room temperature. After washing with PBST, 50 µL of TMB (Dako, Glostrup, Denmark) substrate solution was added. The reaction was stopped by the addition of 0.5 N sulfuric acid, and the optical absorbance was measuredat 450 nm using an Envision 2102 Multilabel Reader (PerkinElmer, Waltham, MA).LC-MS/MS analysis of arachidonic acid metabolites in synovial fluids was performed as described previously [41]. An internal standard was added to the stored methanol mixtures of synovial fluid. The mixtures were vortex-mixed vigorously and centrifuged to obtain the upper layer. Arachidonic acid metabolites were extracted with a MonoSpin C18-AX column (GL Sciences Inc., Tokyo, Japan) and introduced into the LC-MS/MS system. The arachidonic acid metabolites were separated using an Acquity UPLC System (Waters Corp., Milford, MA) with an Acquity UPLC BEH C18 column (100 mm × 2.1 mm i.d., 1.7 µm; Waters Corp.). Mobile phases A and B were composed of water/5 mM phosphoric acid/acetic acid (1000:1:1, v/v/v) and acetonitrile/isopropanol/acetic acid (500:500:1, v/v/v), respectively. MS analysis was performed with an API5000 Mass Spectrometer (AB Sciex, Foster City, CA) equipped with an electrospray ionization (ESI) source in the scheduled multiple reaction monitoring (MRM) mode. Arachidonic acid metabolites were detected in the negative ESI mode. Acquisition and processing of data were carried out with Analyst v1.6.2 (AB Sciex). The MRM conditions for each molecule are summarized in SupplementaryData are expressed as mean ± SEM. Comparisons of two unpaired groups were carried out by an F test followed by a t-test. Dunnett’s test was used to compare pain-related behaviors after injection of GSK1016790A into the knees of sham or MIA rats. A paired t-test was used to compare the amounts of arachidonic acid metabolites in synovial fluids between the ipsilateral and contralateral knee joints of OA model rats. Values of P< 0.05 were considered significant. RESULTS Male MIA rats were used as an experimental OA pain model, with grip strength measurements (grip strength [g]/body weight [kg]) used to confirm knee joint pain. After an intra-articular injection of MIA, we found that the grip strength of the hind limbs was reduced, as we previously reported [27] (Fig. 1A, 1B). The body weights did not differ significantly between the sham rats and MIA rats (Supplementary Table 1, available at http://links.lww.com/PAIN/A537). MIA and sham rats were then treatedwith an intra-articular injection of a TRPV4 antagonist, HC067047 or GSK2193874, to examine the role of TRPV4 in mediating this reduced grip strength. We found that both antagonists could significantly improve the reduction in grip strength in MIA rats within 30 min of their administration as compared with the sham rats (Fig. 1A, 1B). These reversal effects of GSK2193874 were also observed in female rats (Supplementary Fig. 1; Supplemental Table 1, available at http://links.lww.com/PAIN/A537). These findings suggest the involvement of TRPV4 in the development of knee joint pain in OA rats.We next examined the functional changes in TRPV4 expression in the knee joints of MIA and sham rats following an intra-articular injection with a TRPV4 agonist, GSK1016790A. Sensitization in TRPV4 was measured in terms of pain-related behaviors in MIA and sham rats, with pain estimated by changes in right leg withdrawal and lifting behavior. MIA rats showed significantly longer durations of right leg withdrawal and lifting behavior following delivery of the agonist (48 ± 5.6 vs. 12 ± 2.8 sec at baseline levels, P < 0.001; Fig. 2A) but no significant increase was seen in the sham rats (9.6 ± 2.2 vs. 0.78 ± 0.36 sec; Fig. 2A). These findings suggest that TRPV4 becomes sensitized in the knee joints of MIA rats.To clarify the mechanism of TRPV4 sensitization in the knee joints of MIA rats, we first sought to examine the expression of TRPV4 by IHC analysis. TRPV4 antibody specificity was confirmed by western blot analysis, absorption tests, and IHC analysis using bladder tissue as a positive control (Supplementary Figs. 2, 3; available at http://links.lww.com/PAIN/A537). We detected TRPV4-immunopositive signals in the articular cartilage of the knee joints in sham rats however there was no positive signal for TRPV4 in the knee joints of MIA rats. This loss in TRPV4 expression was accompanied by a deterioration in the structural architecture of the articular cartilage in MIA rats (Fig. 2B, Supplementary Fig. 3; available at http://links.lww.com/PAIN/A537). No obvious TRPV4-immunopositive signals were detected in the synovium of either the sham or MIA rats (Fig. 2B). These findings suggest that the sensitization of TRPV4 in the knee joints of MIA rats was not caused by an increased expression of TRPV4 in the knee joint articular cartilage and synovium. Pain signals are transmitted by sensory nerves in the knee joint. Recently, we found that PKC was activated in the DRG neurons of MIA rats [27]. As PKC can phosphorylate and sensitize TRPV4 at Ser824 [17,38], we next examined whether TRPV4 sensitization occurred in the DRG neurons. TRPV4-immunopositive signals were detected in most DRG neurons examined in the IHC analysis (Supplementary Fig. 4, available at http://links.lww.com/PAIN/A537), as previously reported [9]. We generated ELISA systems for quantitative measurement of total and phosphorylated TRPV4 with the development of a novel antibody against TRPV4 phosphorylated at Ser824. The created antibody was not suitable for western blotting or immunohistology (data not shown), suggesting the antibody only recognized the three-dimensional structure of the phosphorylated TRPV4 protein. The ELISA systems revealed specific and dose-dependent detection of total and phosphorylated TRPV4 in both TRPV4-CHO cells and the DRG (Supplementary Figs. 5 and 6, available at http://links.lww.com/PAIN/A537). Analysis of total and phosphorylated TRPV4 levels in the DRG of MIA rats revealed that TRPV4 phosphorylated at Ser824 was significantly increased in MIA rats compared with sham rats. However, total TRPV4expression remained unchanged between the sham and MIA rats. This suggests that TRPV4 sensitization was associated with an increase in the phosphorylation of TRPV4 at Ser824 in the DRG of MIA rats.We next sought to confirm that TRPV4 was sensitized by its phosphorylation. Phorbol 12-myristate 13-acetate (PMA) is a highly potent activator of PKC, which phosphorylates and sensitizes TRPV4 [17]. Thus, we surmised that PMA would indirectly stimulate the TRPV4 channel. Indeed, we found that PMA increased the proportion of phosphorylated TRPV4 in TRPV4-expressing CHO cells (Supplementary Fig. 6C, available at http://links.lww.com/PAIN/A537), as previously reported [17]. Because TRPV4 is a nonselective cationic channel, we next tested whether a TRPV4 agonist, GSK1016790A, could also stimulate an increase in calcium flux in a concentration-dependent manner in both PMA-treated TRPV4-CHO cells and vehicle-treated TRPV4-CHO cells. However, the ED50 of the calcium flux induced by GSK1016790A was significantly decreased in PMA-treated TRPV4-CHO cells compared with vehicle-treated TRPV4-CHO cells (Fig. 3B, Supplementary Fig. 7, available at http://links.lww.com/PAIN/A537; PMA-treated: 0.064 ± 0.010 µM vs.vehicle: 0.15 ± 0.02 µM, F = 2.12, P = 0.004). These findings confirm that TRPV4 was sensitized by its phosphorylationWe recently developed a novel lipid profiling system [41] and identified the arachidonic acid metabolite 5,6-EET, an endogenous TRPV4 ligand, and other arachidonic acid metabolites in the knee joint cavities of MIA rats. Here, we collected lavage fluids from the knee joint capsules of MIA and sham rats and tested the profile changes using our lipid profiling system. We found a significant increase in the amounts of 5,6-EET, arachidonic acid, and prostaglandin E2 (PGE2) in the lavage fluids of MIA-treated knee joints as compared with the contralateral (untreated) knee joints (Fig. 4, Supplementary Table 3, available at http://links.lww.com/PAIN/A537). To verify this increase in 5,6-EET, we also examined the profile of arachidonic acid metabolites in lavage fluids from MNx rats, which develop similar clinical features to knee OA [47], and show knee joint pain in dynamic weight-bearing tests (Supplementary Fig. 8, available at http://links.lww.com/PAIN/A537). Significant increases in 5,6-EET, as well as arachidonic acid and PGE2, were observed in lavage fluids from MNx-treated knee joints compared with the contralateral knee joints (Fig. 4, Supplementary Table 4;To examine the existence of 5,6-EET in clinical synovial fluids, we investigated the profile of arachidonic acid metabolites in synovial fluids from OA patients. Synovial fluids were collected from 12 Caucasian patients diagnosed with OA. In the analysis, 5,6-EET was detected in four specimens (0.39±0.29 ng/mL) and its metabolite dihydroxyeicosatrienoic acid (5,6-DHET) was detected in nine specimens (0.19±0.07 ng/mL). Arachidonic acid was detected in all specimens (975.55±149.26 ng/mL), while PGE2 was detected in only three specimens (0.03±0.02 ng/mL) (Fig. 5, Supplementary Table 5, available at http://links.lww.com/PAIN/A537). DISCUSSION In the present study, we investigated the involvement of TRPV4 in knee OA pain by examining the effects of a TRPV4 agonist and two antagonists in the knee joints of OA pain model rats. We found that the two TRPV4 antagonists offered analgesic effects in MIA rats, and clarified that TRPV4 was sensitized in the knee joint. Further, we showed an increase in TRPV4 phosphorylation at Ser824 without any increase in the total TRPV4 expression in the DRGs of MIA rats. These findings suggest that activated PKC increases the phosphorylation of TRPV4 in the DRG of MIA rats, similar to the case for TRPV1 [27]. Others reported an increase in TRPV4 expression was reported in the DRGs in chronic NGF-overexpressing mice [20]. However, our current findings suggest that the levels of NGF in knee joint of MIA rats would be not sufficient to induce TRPV4 expression. A previous study showed an increase in TRPV4 expression in the trigeminal ganglion in a rat model of temporomandibular joint pain, but this expression was transient and returned to normal levels by day 13 [13]. A temporal increase in the presence of immune cells in the infrapatellar fat pad in MIA rats was also reported, and these levels also returned baseline by day 14 [15]. It is plausible that the increase in TRPV4 in the DRGs of MIA rats in our study might be also increased temporarily at an earlier time point after injection and have returned to normal levels by the assay time point; this could be tested in future experiments. This is the first study to detect changes in TRPV4 phosphorylation at Ser824 directly in vivo and in an animal pain model. TRPV4 phosphorylated at Ser824 shows increased sensitivity to hypotonic solution, which acts like an endogenous TRPV4 agonist [38]. Likewise, we confirmed the increased sensitivity of TRPV4 to the chemically synthesized TRPV4 agonist GSK1016790A in cells with phosphorylated TRPV4. The decrease in the ED50 and the decrease in the time required to reach the maximum response to GSK1016790A in cells with phosphorylated TRPV4 suggest that TRPV4 is sensitized by its phosphorylation. Interestingly, although GSK1016790A showed a dose-dependent calcium influx in control TRPV4- and phosphorylated TRPV4-expressing CHO cells, the intra-articular injection of GSK1016790A did not induce significant pain-related behaviors in sham rats. These findings suggest that, in addition to calcium influx, other signaling molecules or cellular processes are required to produce pain signals. Indeed, others have shown that an intraplantar injection of hypotonic solution can induce TRPV4-dependent, pain-related behaviors in normal mice only in the presence of an inflammatory mediator [1,3]. The authors proposed that, in addition to the PKC pathway, protein kinase A (PKA)-dependent cyclic AMP accumulation may promote TRPV4-mediated pain signals. Protease-activated receptor 2 (PAR2) also causes pain in a TRPV4-dependent manner [21,42] by activating pathways such as PKC, PKA, and extracellular signal-regulated kinases (ERK) [53]. PAR2 cleavage enzymes are expressed in the OA knee joint [34,52], suggesting a role for PAR2 in TRPV4-mediated pain in knee OA. Further studies focusing on the relationship between TRPV4 sensitization and PKA and PAR2 activation in the OA knee joint are required. In the present study, we first demonstrated that an endogenous TRPV4 ligand, 5,6-EET, was increased in the knee joint of MIA rats as well as MNx rats using a recently-developed lipid profiling system. We also found increases in the profile of other arachidonic acid metabolites, variably across all samples. Although MIA rats and MNx rats both exhibit the clinical features of OA pain [22], the development of OA differs between the two rat models. Interestingly, though, the profiles of the arachidonic acid metabolites in the knee joint were almost identical between the two models. In addition to the increases in arachidonic acid and PGE2, which are known to be clinically increased in OA patients [28], increases in lipoxygenase products (12-HETE, 15-HETE) were observed in both models. In contrast, leukotrienes were not increased in either model, suggesting that the changes in arachidonic metabolites were enzymatically controlled in the knee joints of OA model rats. Other EETs were also detected, indicating the existence of a cytochrome P-450 pathway in the knee joints of OA model rats. Arachidonic acid and some EETs gate TRPV4 [48], suggesting potential synergic effects with 5,6-EET in stimulating TRPV4. A recent lipidomics study showed no differences in the profile of eicosanoids in the knee joints of sham and MIA rats except for an increase in 12-HETE and the absence of EETs in the knee joint of MIA rats [51]. It is possible that these results differ to ours because of the tissues examined: in the previous study, lipids were extracted from whole knee joints, including bone, connective tissues, and synovium, whereas we used lavage fluids from the knee joint cavity, which would contain all of the substances released into the knee joint cavity; this allowed us to focus on the function of the molecules produced around the knee joint. We did not investigate how 5,6-EET is produced in this study. Previous work using colon biopsies has shown that 5,6-EET is correlated with abdominal pain in patients with irritable bowel syndrome [10]. In the OA knee joint, endothelial cells [39], macrophages [8], and peripheral sensory nerves [7] may be responsible for the production of 5,6-EET; however, given the significant cartilage disruption seen in our OA models, it is unlikely that 5,6-EET is produced by chondrocytes. 5,6-EET gates TRPV4 intracellularly [4]; however, we did not determine the concentration of 5,6-EET that would be physiologically relevant for this role. 5,6-EET is also reported to have functions other than TRPV4 activation [43]. Further studies are required to clarify the production and function of 5,6-EET in the OA knee joint, especially through modulation of its production in vivo and to examine its function in genetically TRPV4-deficient animals. Although this is the first study to confirm the existence of 5,6-EET in synovial fluids from OA patients, we could not conduct analyses in healthy volunteers because of ethical limitations, or perform correlation analyses for the pain scores linked with the clinical samples. Unlike in the animal experiments, we detected significant amounts of 5,6-DHET, a 5,6-EET metabolite, and other DHETs in the synovial fluid samples. The abundance of the 5,6-DHET metabolite likely reflects both the presence and instability of 5,6-EET [19], and the probability that 5,6-EET is degraded during the accumulation process of synovial fluids. This is also consistent with the finding that fewer samples expressed PGE2. Further clinical studies are needed to clarify the relationship between 5,6-EET and clinical OA pain in the future. TRPV4 is involved in cartilage and bone homeostasis [31]; yet, the function of TRPV4 in cartilage and bone homeostasis seems to differ depending on the stage of development: male TRPV4-knockout mice show accelerated age-related OA [14], whereas conditional knockdown mice show a prevention of age-related OA progression [35]. This suggests that late-stage inhibition of TRPV4 may provide benefits for OA patients. Overlapping of the binding pocket of 5,6-EET and hypotonic induced activation of TRPV4 [4] suggest that 5,6-EET would function as a co-activator of mechanical stimulation and potentiate the pain in the OA knee. Our results suggest that TRPV4 inhibition would prevent not only OA progression, but also prevent OA pain without sex differences. Further studies using genetically TRPV4-deficient animals are required to confirm these suppositions. In conclusion, we show that TRPV4 was sensitized in the knee joints of MIA rats with increased TRPV4 phosphorylation in the DRGs and the elevated expression of its endogenous ligand for TRPV4, 5,6-EET. Furthermore, the intra-articular administration of a TRPV4 antagonist could suppress the pain-related behaviors in MIA rats. These findings will GSK2193874 enhance our understanding of TRPV4 involvement in OA pain mechanisms in the knee joint, and aid in the discovery of relevant TRPV4 antagonists as potent novel analgesics for OA pain.