- Email: [email protected]
Deficits in spontaneous burrowing behavior in the rat bilateral monosodium iodoacetate model of osteoarthritis: an objective measure of pain-related behavior and analgesic efficacy
Osteoarthritis and Cartilage 23 (2015) 1605e1612
Deficits in spontaneous burrowing behavior in the rat bilateral monosodium iodoacetate model of osteoarthritis: an objective measure of pain-related behavior and analgesic efficacy L.A. Bryden, J.R. Nicholson, H. Doods, A. Pekcec* Boehringer Ingelheim Pharma GmbH & Co. KG, Div. Research Germany, Birkendorfer Strasse 65, 88397 Biberach an der Riss, Germany
a r t i c l e i n f o
s u m m a r y
Article history: Received 8 October 2014 Accepted 1 May 2015
Objective: To characterize deficits in burrowing behavior e an ethologically-relevant rodent behavior e in the monosodium iodoacetate (MIA) rat model of osteoarthritis (OA), and the sensitivity of these deficits to reversal by analgesic drugs of both prototypical and novel mechanisms of action. A second objective was to compare the burrowing assay to a spontaneous locomotor activity (sLA) assay. Method: Male Wistar Han rats (200e220 g) received intrarticular (i.a.) injections of MIA or saline for sham animals. A deficit in the amount of sand burrowed from steel tubes filled with 2.5 kg of sand was used as a measure of pain-related behavior, and sensitivity to reversal of these deficits by analgesic drugs was assessed in bilaterally MIA-injected rats. Results: Bilateral MIA injections induced a significant impairment of burrowing behavior, which was concentration-dependent. The temporal pattern of the deficits was biphasic: a large deficit at 3 days post-injection, resolving by day 14 and returning at the 21 and 28 day time points. At the 3 day time point ibuprofen, celecoxib and an anti-nerve growth factor (NGF) monoclonal antibody (mAb) were able to significantly reinstate burrowing behavior, whereas the fatty acid amide hydrolase (FAAH) inhibitor PF-04457845 and morphine displayed no reversal effect. Morphine impaired burrowing behavior at 3 mg/kg in sham animals. Deficits in rearing frequency in the locomotor activity assay proved irreversible by analgesics. Conclusion: Burrowing behavior provides an objective, non-reflexive read-out for pain-related behavior in the MIA model that has predictive validity in detecting analgesic efficacy of nonsteroidal antiinflammatory drugs (NSAIDs) and an anti-NGF mAb. © 2015 Osteoarthritis Research Society International. Published by Elsevier Ltd. All rights reserved.
Keywords: Burrowing Monosodium iodoacetate Spontaneous behavior Pain Rat
Introduction The prevalence of osteoarthritis (OA) is estimated to be 50% in people aged 65 and older1 and with an aging population the cost and health impact of this disease will continue to rise. The intraarticular (i.a.) injection of monosodium iodoacetate (MIA) into the rat knee joint produces histological changes representative of those seen in human OA2. Concomitant with the early and late
* Address correspondence and reprint requests to: A. Pekcec, Boehringer Ingelheim Pharma GmbH & Co. KG, Div. Research Germany, Birkendorfer Strasse 65, 88397 Biberach an der Riss, Germany. Tel: 49-7351-54-92312; Fax: 49-7351-545128. E-mail addresses: [email protected] (L.A. Bryden), janet. [email protected] (J.R. Nicholson), [email protected] (H. Doods), [email protected] (A. Pekcec).
histological changes in the MIA model are pain-related behaviors3, making it a useful model for assessing the analgesic efficacy of drugs. Pain is the main clinical manifestation of OA. The current lack of effective disease modifying agents to target the aetiology of OA4 means the use of analgesic drugs is the mainstay treatment for alleviating the impact of the disease on daily life, namely nonsteroidal anti-inflammatory drugs (NSAIDs) and weak opioids5. The main techniques for assessing pain-like behaviors in the MIA model are measures of mechanical hypersensitivity: shifts in weight bearing (WB) from the ipsilateral affected to contralateral unaffected knee, distal mechanical hyperalgesia and tactile allodynia6e8. A pressing issue in the pain field is the lack of translation between preclinical and clinical findings, which has resulted in relatively few safe and effective analgesics being developed (reviewed by Blackburn-Munro 20049 and Mogil, 200910).
http://dx.doi.org/10.1016/j.joca.2015.05.001 1063-4584/© 2015 Osteoarthritis Research Society International. Published by Elsevier Ltd. All rights reserved.
1606
L.A. Bryden et al. / Osteoarthritis and Cartilage 23 (2015) 1605e1612
Although there is evidence that current pain assays have some predictive validity from rat to human11, it is becoming increasingly apparent that assays utilizing spontaneous rather than evoked/reflexive measures are needed to assess the global impact of pain beyond hypersensitivity12,13. In addition to sharp pain evoked by movement, patients with OA rank constant and aching pain among the most distressing features of the disease14. How well the current evoked/reflexive assays of pain-like behaviors in rodents account for this persistent pain is unclear9. Furthermore, clinical assessments of chronic pain are focusing increasingly on the multifaceted nature of pain, such as the effect on emotion and physical function15, which are not assessed preclinically by reflexive assays. Burrowing is an innate rodent behavior indicative of animal wellbeing that is conserved across various strains of rat16e19 and mice20e22. Deficits in burrowing behavior occur in preclinical rat models of inflammatory and neuropathic pain and can be reversed by analgesics16,18,19,23,24. This behavior, therefore, is a useful preclinical measure of non-evoked pain. Burrowing may encompass the supraspinal mechanisms that contribute to pain phenotypes, both sensory and affective, as well as being more ethologically relevant than the commonly used assays measuring hypersensitivity25. Here we show deficits in burrowing behavior in the bilateral MIA model of OA and sensitivity of these deficits to reversal by prototypical and novel analgesics. To our knowledge, this is the first published article demonstrating deficits in burrowing behavior in the MIA model. Additionally, burrowing behavior was compared to another non-reflexive readout: spontaneous locomotor activity (sLA). This has been shown previously to be impaired in various preclinical animal models and to be sensitive to pharmacological modulation26e29. Methods Animals 594 male Wistar Han rats (200e220 g; Charles River Laboratories, Germany) were used for all experiments and were housed in groups of four with food and water ad libitum with a 12 h light/dark cycle. All animal experimental protocols were authorized by the Local Animal Care and Use Committee and carried out according to the local animal care guidelines, AAALAC regulations, and the USDA Animal Welfare Act. MIA injections Rats were briefly anaesthetized with 5% isoflurane (Abbott Laboratories, Wiesbaden, Germany) followed by 3% maintenance, after which an i.a. injection of 3 mg of MIA (SigmaeAldrich, Steinheim, Germany) dissolved in 50 mL of 0.9% physiological saline was performed into the femorotibial joint. Sham rats received 50 mL injections of physiological saline. All injections for pharmacology experiments were bilateral, except for one model conditions studies in which rats received either uni- or bilateral injections. Drugs and drug administration The analgesic drugs tested were: morphine (Caelo, Germany), ibuprofen (Sigma Aldrich, Steinheim, Germany), celecoxib (LKT Laboratories Inc., St. Paul, MN, USA) and the fatty acid amide hydrolase (FAAH) inhibitor PF-04457845 (synthesized by Boehringer Ingelheim). All drugs were administered orally (p.o.) with 0.5% (w/v) natrosol and 0.1% (v/v) tween-80 (9:1 ratio) as vehicle, except morphine which was injected subcutaneously (s.c.) in the interscapular area with 0.9% physiological saline as vehicle. All
drugs were administered in a volume of 2 mL/kg except PF04457845, which was administered in a volume of 4 mL/kg. For the generation of the anti-nerve growth factor (NGF), monoclonal antibody (mAb) variable domains were extracted from the patent application WO 2004/058184 A2 (applicant: Rinat Neuroscience Corporation) and processed as described by Hezareh et al., 200130. The antibody was administered s.c., with phosphate buffered saline as vehicle. Burrowing training and burrowing experiments For all burrowing experiments steel tubes (32 cm in length and 10 cm in diameter) were filled with 2.5 kg of quartz sand and placed in Plexiglas cages (600 340 200 mm). The open-end of the tube was elevated 6 cm from the floor of the cage. Training for each experiment was carried in 2 phases: social facilitation (SF) and individual training (IT). For SF rats were placed in pairs in a cage for 2 h on two consecutive days. The amount of sand burrowed by each pair was measured and if a pair burrowed less than 1500 g of sand one of the pair was swapped with a rat from a pair that had burrowed greater than 1500 g for the second SF day. For IT rats were placed alone in the burrowing set-up for 30 min per day and the average amount burrowed over 3 days was calculated to attain a baseline burrowing performance value. Before each animal was given MIA or sham intrarticular injections they were assigned to groups to ensure that each treatment group for an experiment had a comparable baseline burrowing value. For the model conditions study rats were allowed to individually burrow for 30 min 3 days after MIA or sham injections were performed and the amount of sand burrowed was recorded. For the time-course experiment burrowing performance was assessed 3, 14, 21 and 28 days post-injection. For all pharmacology experiments burrowing behavior was measured 3 days after MIA injection, and animals were placed in the burrowing set-up after the appropriate pre-treatment time. Exclusions During training any animal that has a baseline burrowing value of less than 1000 g or a standard deviation of burrowing (SD) greater than 450 g was excluded to ensure high and stable baselines. To ensure that all rats used in pharmacology experiments were in a comparable pain state burrowing values were also measured 1 day after MIA injection and any animal with a burrowing value greater than 1000 g was excluded from pharmacology at day 3. In total, exclusions after training and after day 1 accounted for around 5% of rats per study. Locomotor activity sLA was measured for 30 min with an automated monitoring system (TruScan Activity Monitor version 2.0, Coulbourn Instruments, Allentown, PA, USA). Each monitoring system was an enclosed 43 cm2 arena with two levels of sensory photobeams; one level elevated 3 cm from the floor of the arena and the other 14 cm, measuring horizontal and vertical activity respectively. Each level of the detection system was equipped with 16 photobeams per wall of the arena, spaced 2.5 cm apart. The parameters measured by the system were: ambulatory horizontal distance moved (cm), rearing frequency and rearing time (s). Prior to testing, all rats were habituated in an annexe to the testing room. Statistical analysis All statistical processing was performed in GraphPad Prism version 6.0 (San Diego, CA, USA). For all analyses P < 0.05 was
L.A. Bryden et al. / Osteoarthritis and Cartilage 23 (2015) 1605e1612
1607
considered statistically significant. To control for familywise type I error rates a Bonferroni correction was applied where appropriate. A 1-way analysis of variance (ANOVA) with Bonferroni's post hoc was used to analyze the model conditions studies investigating whether uni- or bilateral MIA injections were most appropriate and the concentration-dependency of burrowing deficits. A Bonferroni cut-off of P < 0.025 was applied: 0.05/(number of comparisons per family (2)/number of families(1)). The time-course study was analyzed using 2-way repeated measures ANOVA. Significant differences between sham animals and MIA-injected animals on the same day were ascertained using Bonferroni's post hoc. A Bonferroni cut-off of P < 0.025 was applied: 0.05/(number of comparisons per family (4)/number of families(2)). The statistical significance of deficits in locomotor activity parameters was analyzed using an unpaired Student's t test. For all pharmacology studies (burrowing and locomotor activity) data were analyzed using a 2-way ANOVA. To determine that the MIA vehicle group was significantly lower than the sham vehicle group and to determine which doses of compound were efficacious in reversing burrowing deficits a Dunnett's post hoc multiple comparisons was used, comparing with the MIA vehicle control group. All data are expressed as means and corresponding 95% confidence intervals (CI). Gaussian distribution was analyzed using a ShapiroeWilk normality test. Results Establishment of model conditions To determine whether unilateral or bilateral MIA injections were most appropriate for the burrowing assay an experiment was conducted for comparison. Injection of MIA bilaterally resulted in a 60% reduction in burrowing performance compared with the sham bilateral control group 3 days post-injection, whereas unilaterally MIA-injected rats displayed no deficit in burrowing performance at day 3 [Fig. 1(A)], or at later time points of 14, 21 and 28 days [Fig. 1(C)]. We investigated the effect of saline sham injections on burrowing performance and found that neither unilateral nor bilateral injections of saline influenced burrowing behavior [Fig. 1(A)]. Next we wanted to establish the concentrationresponsiveness of MIA-induced burrowing deficits. Concentrations of 0.3, 1 and 3 mg/knee were selected based on previous findings in WB6. The 0.3 mg/knee concentration of MIA did not significantly impair burrowing performance, whereas 1 and 3 mg/ knee induced significant deficits [Fig. 1(B)]. Although 1 and 3 mg/ knee concentrations produced comparable deficits, we decided on 3 mg/knee for pharmacological experiments as this concentration produced the most robust deficit with the least variance. To establish time points for pharmacological experiments burrowing performance was assessed 3, 14, 21 and 28 days after the bilateral injection of MIA [Fig. 1(C)]. As seen in [Fig. 1(C)], a robust deficit occurred at 3 days post-injection, which resolved back to baseline values by the 14 day time point. However, the deficits at the 21 and 28 day time points in the pharmacology experiments were found not to be large enough to provide an assay window to titrate a reversal effect. Therefore, it was decided to focus on the 3 day time point for all pharmacology. After establishing the model conditions for the burrowing assay, the same conditions were chosen for locomotor activity studies to allow for comparisons between these two assays. Three days after bilateral injection of MIA (3 mg/knee) deficits in distance moved [Fig. 2(A)], rearing frequency [Fig. 2(B)] and rearing time [Fig. 2(C)] occurred. We decided to use rearing frequency as the parameter for pharmacology studies as it had the largest deficit; rearing frequency was 48% lower in MIA injected animals than in sham animals.
Fig. 1. Effect of MIA on burrowing behavior. (A) Amount of sand burrowed by naïve rats and 3 days after intra-articular (i.a.) of saline or MIA (3 mg/knee). Rats either received unilateral (right knee) or bilateral injections of saline/MIA. Data analyzed with 1-way ANOVA followed by Bonferroni's post hoc multiple comparisons (****P < 0.0001 vs sham bilateral group, n ¼ 12 per group). (B) Concentration-response to bilateral i.a. injections of MIA 3 days after injection. Data analyzed with 1-way ANOVA followed by Bonferroni's post hoc multiple comparisons (*P < 0.05 and **P < 0.01 vs sham group, n ¼ 10e11 per group). (C) Time-course of burrowing performance after unilateral and bilateral i.a. MIA injections (3 mg/knee). Data were analyzed with 2-way repeated measures ANOVA (MIA time interaction: F4,48 ¼ 12.93, P < 0.0001), followed by Bonferroni's post hoc multiple comparisons (significant differences compared with sham group on same day: **P < 0.01,****P < 0.0001, n ¼ 12 per group). BL ¼ baseline.
Rearing time was similarly reduced by MIA injection, whereas the deficit in distance moved was more modest (21% lower than sham group). At day 28 there was no statistically significant effect of MIA on any locomotor activity parameter (Supplementary Table 2). Effect of NSAIDs, morphine and gabapentin on deficits in burrowing performance Nonsteroidal anti-inflammatory drugs (NSAIDs) and opioids are categories of drugs that are currently used to treat pain in OA
1608
L.A. Bryden et al. / Osteoarthritis and Cartilage 23 (2015) 1605e1612
Effect of an anti-NGF mAb and the FAAH inhibitor PF-04457845 on deficits in burrowing performance An anti-NGF mAb was tested in the burrowing in light of recent phase III clinical trials with an antibody of the same mechanism of action31. When given 24 h before measurement of burrowing performance the antibody had no effect on sham animals and was able to reverse the deficit in burrowing behavior induced by MIA to a level similar to that of the respective sham group, at a dose of 9 mg/ kg [Fig. 4(A)]. The FAAH inhibitor PF-04457845 is another analgesic with a novel mechanism of action tested in burrowing because of recent negative phase II clinical trials in patients with knee OA32. At all doses tested (10, 30 and 100 mg/kg) there was no reversal of burrowing deficits observed in MIA injected rats and also no effect on burrowing performance in the sham animals [Fig. 4(B)]. Locomotor activity pharmacology Supplementary Table 1 shows that in our hands no analgesic effect of any drugs tested was detectable in the rearing frequency parameter of the locomotor activity assay, which is in contrast to the pharmacological sensitivity evident in the burrowing assay. The 10 mg/kg dose of morphine caused a significant reduction in rearing frequency compared with the sham vehicle group; however, unlike for burrowing, there was no significant reduction in rearing frequency induced by the 3 mg/kg dose (P ¼ 0.18) compared with sham vehicle. Discussion
Fig. 2. Locomotor activity 3 days after bilateral injection of saline or 3 mg/knee MIA. Distance moved (A), rearing frequency (B) and rearing time (C) was monitored over a 30 min period. Data were analyzed with unpaired Student's t test (*P < 0.05 vs sham group, n ¼ 8 per group).
clinically5, so were assessed to determine their efficacy in reversing deficits in burrowing performance 3 days after the injection of MIA. Both the non-selective cyclooxygenase (COX) inhibitor ibuprofen and the COX-2 selective inhibitor celecoxib were efficacious in reinstating deficits in burrowing performance at doses of 10 and 30 mg/kg for ibuprofen, and 3, 10 and 30 mg/kg for celecoxib [Fig. 3(A) and (B), respectively]. The increase in burrowing performance elicited by ibuprofen and celecoxib at these doses was a minimum of 50% when compared with the MIA vehicle group. Neither morphine or gabapentin reversed burrowing deficits at any dose tested [Fig. 3(C) and (D)], and morphine impaired burrowing performance at 3 mg/kg in the sham group.
The initial aim of this study was to investigate whether burrowing was impaired by i.a. injection of MIA and could therefore provide a measure of pain-related behavior in this pre-clinical model of OA. Our model conditions studies showed that injecting MIA into only one knee joint was insufficient to cause depression of burrowing behavior, however, bilateral injection resulted in a robust deficit in performance. One possibility for this is that when MIA injections are given unilaterally compensation occurs through use of the three unaffected limbs to burrow. In contrast, previous studies investigating the effect of intraplantar (i.pl) complete Freund's adjuvant (CFA), i.a. CFA and nerve ligation models of neuropathic pain found unilateral injury to be sufficient to depress burrowing behavior16,18,19,24. It could be inferred that the MIA model is less severe in its pain phenotype and effect on animal well-being, therefore bilateral injury is needed to induce a deficit. However, unilateral injection of MIA is sufficient to induce deficits in wheel-running33 and locomotor activity34 in rats, perhaps suggesting burrowing is less sensitive to MIA-induced joint pain than these assays. Depression of locomotor activity after unilateral MIA injection e shown by More et al. (2013)34 e is in contrast to studies in our laboratory where no deficit occurred at the early (day 3) or late (day 28) time points (data not shown). The temporal pattern of MIA-induced deficits in burrowing performance was biphasic, which is similar to reported findings in WB6 and biotelemetric assessment of mobility (BAM)27. It has been suggested from behavioral and histological findings that the early phase of the MIA model (day 3 in our study) is predominantly pain mediated by inflammation6,7, whereas the later stages of the model represent pain of a different aetiology, perhaps neuropathic in nature35. This is inferred from observed upregulation of nerve injury markers at the late phase time points36e38 and sensitivity to drugs used to treat neuropathic pain37. However, it is important to point out that previous studies mentioned here used unilateral MIA injections, whereas we
L.A. Bryden et al. / Osteoarthritis and Cartilage 23 (2015) 1605e1612
1609
Fig. 3. The effect of NSAIDs, morphine and gabapentin on burrowing performance 3 days after bilateral 3 mg/knee MIA or sham injections. Data analyzed with 2-Way ANOVA followed by Dunnett's post hoc multiple comparisons. (A) Ibuprofen (p.o.) administered 2 h before burrowing performance was measured (MIA dose interaction: F3,66 ¼ 2.785, P < 0.05; ##P < 0.01 vs sham vehicle, *P < 0.05, **P < 0.01 vs MIA vehicle). N ¼ 9e10 per group. (B) Celecoxib (p.o.) administered 2 h before burrowing performance was measured (MIA dose interaction: F3,68 ¼ 2.82, P < 0.05; ####P < 0.0001 vs sham vehicle, *P < 0.05,**P < 0.01 vs MIA vehicle). N ¼ 8e11 per group. (C) Morphine (s.c.) administered 1 h before burrowing performance was measured (#P < 0.05, ##P < 0.01 vs sham vehicle). N ¼ 8e9 per group. (D) Gabapentin (p.o.) administered 2 h before burrowing performance was measured (main effect of MIA: F1,62 ¼ 48.21, ****P < 0.0001). N ¼ 8e9 per group.
injected bilaterally e differences in pathophysiology between unilateral and bilateral injections should therefore be borne in mind. Taking this into account we suggest that the early phase deficits in burrowing represent inflammatory pain, which resolves by day 14 and a return of deficits in the late phase could be chronic/ neuropathic in nature. We tested the neuropathic pain drug gabapentin at day 3 but found no effect at any dose tested, which is supportive of primarily inflammatory pain in the early phase of the MIA model. Unfortunately, the assay window for pharmacology at day 21 and 28 was found to be insufficient, therefore it was not possible to use gabapentin to discern whether this late phase was neuropathic in nature. It was estimated that 20e25 rats per treatment group would be required to determine a pharmacological effect at day 28. These characterization studies show that deficits in burrowing behavior can be used as an objective measure of pain-related behavior in the MIA model of OA. Burrowing is a self-rewarding behavior39 that is not appetite-driven and is indicative of animal well-being17. It is suggested that burrowing is able to measure the negative impact of pain on motivation (affective aspect) in addition to pain evoked by the act of burrowing (sensory aspect). Recent reviews have highlighted the need to assess the global impact of pain in preclinical models of pain beyond measures of hypersensitivity9,10,12,13 through use of more ethological rodent behaviors25. Burrowing may provide one means of achieving this. However, further characterization in rodent models is needed to disinter what aspects of pain burrowing measures compared with other assays and whether results from burrowing can be differentially interpreted. To investigate whether MIA-induced deficits in burrowing performance were sensitive to reversal by analgesics we tested a range of compounds that work by mechanisms of action currently used to
treat pain associated with OA clinically5. Both the non-selective COX inhibitor ibuprofen and the COX-2 selective inhibitor celecoxib (375fold selectivity for COX-240) were efficacious in reversing deficits in burrowing performance at low doses of 10 and 3 mg/kg, respectively. Similarly, previous published findings found that a relatively low dose of ibuprofen (30 mg/kg) was efficacious in reversing deficits in burrowing induced by i.pl CFA16. Both our study and this previous work demonstrate that burrowing may have higher sensitivity to the analgesic effect of ibuprofen than evoked endpoints. In the carrageenan model of inflammatory pain the minimally effective dose (MED) of ibuprofen to reverse mechanical allodynia was 300 mg/kg and to reverse thermal hyperalgesia 1000 mg/kg41, which is considerably higher than the MED in burrowing. However, such comparisons should be approached with caution due to the differences in the underlying neurophysiological mechanisms between pain models and the strain of rat used for experimentation. Morphine did not significantly reverse deficits in burrowing performance at any dose tested, and 3 mg/kg caused a significant reduction in burrowing performance in sham animals, presumably due to sedative side effects. This is in contrast to a previous published study which showed morphine to be efficacious in reversing burrowing deficits induced by i.a. CFA at 1 and 3.16 mg/kg, without reduction in burrowing performance in naïve rats19. However, the route of administration used was intraperitoneal (i.p.) in this previous study, whereas in this study morphine was administered s.c. The sedative side effects of morphine at 3 mg/kg was reflected in previous sLA experiments28 and in our sLA studies significant reduction in rearing frequency occurred at 10 mg/kg. In WB, the efficacy of morphine in normalizing deficits induced by MIA is observed at 3 mg/kg8,42 e a dose which caused a reduction in burrowing performance e and reversal of mechanical hyperalgesia occurs at 6 mg/kg7. There is the possibility of false
1610
L.A. Bryden et al. / Osteoarthritis and Cartilage 23 (2015) 1605e1612
Fig. 4. The effect of the anti-NGF mAb and the FAAH inhibitor PF-04457845 on burrowing performance 3 days after bilateral 3 mg/knee MIA or sham injections. Data analyzed with 2-Way ANOVA followed by Dunnett's post hoc multiple comparisons. (A) anti-NGF mAb administered 24 h before burrowing performance was measured (MIA dose interaction: F1,28 ¼ 7.273, P < 0.05; ###P < 0.001 vs sham vehicle, **P < 0.01 vs MIA vehicle). N ¼ 8 per group. (B) PF-04457845 (p.o.) administered 2 h before burrowing performance was measured (main effect of MIA: F1,65 ¼ 54.65, ****P < 0.0001). N ¼ 8e10 per group.
positives when morphine is tested at these doses because sedation may impair motor function and tone. This impairment could cause a shift in weight onto the ipsilateral injected knee when WB is measured, in addition to any analgesia-mediated effect, and may also inhibit withdrawal reflexes. This highlights an advantage of assays that rely on the reinstatement of pain-depressed behaviors by analgesics: there is no risk of detecting false positives when drugs that potentially impair motor functioning are tested25. The anti-NGF mAb tanezumab has been shown to be efficacious in phase III clinical trials for OA pain, improving pain scores (including global pain assessment) and physical functioning32. In our study an anti-NGF mAb with the same mechanism of action was effective in reversing deficits in burrowing performance in the MIA model. This further demonstrates the predictive validity of the burrowing assay. PF-04457845, a FAAH inhibitor, showed no efficacy at any dose tested in burrowing or sLA assays. Although no published data demonstrating an effect of this compound at the 3 day time point in the MIA model are available there is data showing a positive effect in ameliorating mechanical hyperalgesia in the i.pl CFA model of inflammatory pain, and at the 14 day time point in the MIA model43. However, results from a clinical trial with PF-04457845 failed to show effective analgesia despite raised levels of endocannabinoids32 indicative of FAAH inhibition. The authors suggest a lack of translation between animal models and clinical data in light of the studies showing a potent effect of the compound preclinically. Whether this lack of translation is a result of a fault in the
assessment of pain in preclinical models is not ascertainable at this stage. Parallel studies investigating both spontaneous and evoked endpoints in the MIA model would allow for more comprehensive comparisons. In contrast to burrowing, no analgesic effect of any drug was observed in sLA. Previous pharmacological characterization of MIAinduced deficits in rearing frequency demonstrated an acute analgesic effect of analgesics such as celecoxib and morphine29. In our hands, reversal of burrowing deficits provides a more sensitive read-out for analgesic drug effect, which is in line with the findings comparing i.a. CFA-induced burrowing deficits and locomotor activity18. Furthermore, deficits in burrowing performance are independent of motor impairment, as measured by the locomotor activity assay; we found no correlation between burrowing behavior or any locomotor activity parameters 3 days after i.a. MIA injections (data not shown). The rationale behind the development of nonreflexive and ethologically relevant behavioral paradigms is to improve translation of preclinical to clinical findings. To complement this, the underlying pathophysiology of the animal model used should be as close to the human condition as possible. In that regard, the late phase (28 day time point) of the MIA model is perhaps more translational as it is suggested to model the chronic, noninflammatory pain that is more relevant to OA when it presents clinically. However, it must be noted that denotating OA as ‘noninflammatory’ may refer to a subset of patients, as there is evidence for inflammatory markers in some cases44. This highlights the complexity and heterogeneity of OA as a disease. Nevertheless, further studies are warranted to optimize the burrowing paradigm to enable pharmacological testing at the 28 day time point. In conclusion, these data demonstrate that burrowing provides an objective read-out for pain in the early phase of the MIA model and is sensitive to reversal by NSAIDs and an anti-NGF mAb. As well as having the advantage of being objective, the lack of experimenter presence during the test period in the burrowing assay reduces the influence of stress on rodent behavior. This is illustrated by a recent study showing that rodent pain responses are inhibited by the presence of a male experimenter due to stress-induced by axillary secretions45. We also found that burrowing has superior sensitivity to pharmacological modulation over sLA, which is in line with findings in the i.a. CFA model of rheumatoid arthritis19, further supporting the utility of burrowing as a preclinical pain assay. Contributions HD and JN contributed by conception and design of the studies, interpretation of data and critical revision of the article. LB generated, analyzed and interpreted the data and drafted the article. AP contributed by conception and design of the studies, interpretation of data and writing the manuscript. Role of the funding source All funding was provided by Boehringer Ingelheim Pharma GmbH & Co KG. Competing interests All authors were all employees of Boehringer Ingelheim Pharma GmbH & Co KG at the time of data collection and manuscript writing. Acknowledgments The authors would like to thank Andreas Sterner for technical assistance with locomotor activity studies, Anna-Lena Martin for advisory support with statistical analysis and Torsten Lamla for
L.A. Bryden et al. / Osteoarthritis and Cartilage 23 (2015) 1605e1612
providing the anti-NGF mAb. The current study was performed as part of a contribution of Boehringer Ingelheim Pharma GmbH & Co. KG to the “EUROPAIN” project within the European Consortium “Innovative Medicine Initiative”.
17. 18.
Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.joca.2015.05.001.
19.
References 1. Hinton R, Moody RL, Davis AW, Thomas SF. Osteoarthritis: diagnosis and therapeutic considerations. Am Fam Physician 2002;65:841e8. 2. Guzman R, Evans MG, Bove S, Morenko B, Kilgore K. Monoiodoacetate-induced changes in subchondral bone and articular cartilage of rat femorotibial joints: an animal model of osteoarthritis. Toxicol Pathol 2003;31:619e24. 3. Liu P, Okun A, Ren J, Guo RC, Ossipov MH, Xie J, et al. Ongoing pain in the MIA model of osteoarthritis 2011;493:72e5. 4. Le Graverand-Gastineau MP. Disease modifying osteoarthritis drugs: facing development challenges and choosing molecular targets. Curr Drug Targets 2010;11:528e35. 5. Zhang W, Nuki G, Moskowitz RW, Abramson S, Altman RD, Arden NK, et al. OARSI recommendations for the management of hip and knee osteoarthritis: part III: changes in evidence following systematic cumulative update of research published through January 2009. Osteoarthritis Cartilage 2010;18: 476e99. 6. Bove SE, Calcaterra SL, Brooker RM, Huber CM, Guzman RE, Juneau PL, et al. Weight bearing as a measure of disease progression and efficacy of anti-inflammatory compounds in a model of monosodium iodoacetate-induced osteoarthritis. Osteoarthritis Cartilage 2003;11:821e30. 7. Fernihough J, Gentry C, Malcangio M, Fox A, Rediske J, Pellas T, et al. Pain related behaviour in two models of osteoarthritis in the rat knee. Pain 2004;112:83e93. 8. Pomonis JD, Boulet JM, Gottshall SL, Phillips S, Sellers R, Bunton T, et al. Development and pharmacological characterization of a rat model of osteoarthritis pain. Pain 2005;114: 339e46. 9. Blackburn-Munro G. Pain-like behaviours in animals e how human are they? Trends Pharmacol Sci 2004;25:299e305. 10. Mogil JS. Animal models of pain: progress and challenges. Nat Rev Neurosci 2009;10:283e94. 11. Whiteside GT, Adedoyin A, Leventhal L. Predictive validity of animal pain models? A comparison of the pharmacokineticpharmacodynamic relationship for pain drugs in rats and humans. Neuropharmacology 2008;54:767e75. 12. Mogil JS, Crager SE. What should we be measuring in behavioral studies of chronic pain in animals? Pain 2004;112:12e5. 13. Vierck CJ, Hansson PT, Yezierski RP. Clinical and pre-clinical pain assessment: are we measuring the same thing? Pain 2008;135:7e10. 14. Hawker GA, Stewart L, French MR, Cibere J, Jordan JM, March L, et al. Understanding the pain experience in hip and knee osteoarthritisean OARSI/OMERACT initiative. Osteoarthritis Cartilage 2008;16:415e22. 15. Dworkin RH, Turk DC, Farrar JT, Haythornthwaite JA, Jensen MP, Katz NP, et al. Core outcome measures for chronic pain clinical trials: IMMPACT recommendations. Pain 2005; 113:9e19. 16. Andrews N, Legg E, Lisak D, Issop Y, Richardson D, Harper S, et al. Spontaneous burrowing behaviour in the rat is reduced
20.
21.
22. 23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
1611
by peripheral nerve injury or inflammation associated pain. Eur J Pain 2012;16:485e95. Deacon RM. Burrowing in rodents: a sensitive method for detecting behavioral dysfunction. Nat Protoc 2006;1:118e21. Rutten K, Schiene K, Robens A, Leipelt A, Pasqualon T, Read SJ, et al. Burrowing as a non-reflex behavioural readout for analgesic action in a rat model of sub-chronic knee joint inflammation. Eur J Pain 2014;18:204e12. Rutten K, Robens A, Read SJ, Christoph T. Pharmacological validation of a refined burrowing paradigm for prediction of analgesic efficacy in a rat model of sub-chronic knee joint inflammation. Eur J Pain 2014;18:213e22. Jirkof P, Cesarovic N, Rettich A, Nicholls F, Seifert B, Arras M. Burrowing behavior as an indicator of post-laparotomy pain in mice. Front Behav Neurosci 2010;4:165. Jirkof P, Leucht K, Cesarovic N, Caj M, Nicholls F, Rogler G, et al. Burrowing is a sensitive behavioural assay for monitoring general wellbeing during dextran sulfate sodium colitis in laboratory mice. Lab Anim 2013;47:274e83. Jirkof P. Burrowing and nest building behavior as indicators of well-being in mice. J Neurosci Methods 2014;234:139e46. Huang W, Calvo M, Karu K, Olausen HR, Bathgate G, Okuse K, et al. A clinically relevant rodent model of the HIV antiretroviral drug stavudine induced painful peripheral neuropathy. Pain 2013;154:560e75. Lau W, Dykstra C, Thevarkunnel S, Silenieks LB, de Lannoy IA, Lee DK, et al. A back translation of pregabalin and carbamazepine against evoked and non-evoked endpoints in the rat spared nerve injury model of neuropathic pain. Neuropharmacology 2013;73:204e15. Andrews N, Harper S, Issop Y, Rice AS. Novel, nonreflex tests detect analgesic action in rodents at clinically relevant concentrations. Ann N Y Acad Sci 2011;1245:11e3. Cobos EJ, Ghasemlou N, Araldi D, Segal D, Duong K, Woolf CJ. Inflammation-induced decrease in voluntary wheel running in mice: a nonreflexive test for evaluating inflammatory pain and analgesia. Pain 2012;153:876e84. Guingamp C, Gegout-Pottie P, Philippe L, Terlain B, Netter P, Gillet P. Mono-iodoacetate-induced experimental osteoarthritis: a dose-response study of loss of mobility, morphology, and biochemistry. Arthritis Rheum 1997;40:1670e9. Matson DJ, Broom DC, Carson SR, Baldassari J, Kehne J, Cortright DN. Inflammation-induced reduction of spontaneous activity by adjuvant: a novel model to study the effect of analgesics in rats. J Pharmacol Exp Ther 2007;320:194e201. Nagase H, Kumakura S, Shimada K. Establishment of a novel objective and quantitative method to assess pain-related behavior in monosodium iodoacetate-induced osteoarthritis in rat knee. J Pharmacol Toxicol Methods 2012;65:29e36. Hezareh M, Hessell AJ, Jensen RC, van de Winkel JG, Parren PW. Effector function activities of a panel of mutants of a broadly neutralizing antibody against human immunodeficiency virus type 1. J Virol 2001;75:12161e8. Brown MT, Murphy FT, Radin DM, Davignon I, Smith MD, West CR. Tanezumab reduces osteoarthritic hip pain: results of a randomized, double-blind, placebo-controlled phase III trial. Arthritis Rheum 2013;65:1795e803. Huggins JP, Smart TS, Langman S, Taylor L, Young T. An efficient randomised, placebo-controlled clinical trial with the irreversible fatty acid amide hydrolase-1 inhibitor PF04457845, which modulates endocannabinoids but fails to induce effective analgesia in patients with pain due to osteoarthritis of the knee. Pain 2012;153:1837e46. Stevenson GW, Mercer H, Cormier J, Dunbar C, Benoit L, Adams C, et al. Monosodium iodoacetate-induced osteoarthritis
1612
34.
35.
36.
37.
38.
L.A. Bryden et al. / Osteoarthritis and Cartilage 23 (2015) 1605e1612
produces pain-depressed wheel running in rats: implications for preclinical behavioral assessment of chronic pain. Pharmacol Biochem Behav 2011;98:35e42. More AS, Kumari RR, Gupta G, Lingaraju MC, Balaganur V, Pathak NN, et al. Effect of iNOS inhibitor S-methylisothiourea in monosodium iodoacetate-induced osteoathritic pain: implication for osteoarthritis therapy. Pharmacol Biochem Behav 2013;103:764e72. Thakur M, Rahman W, Hobbs C, Dickenson AH, Bennett DL. Characterisation of a peripheral neuropathic component of the rat monoiodoacetate model of osteoarthritis. PLoS One 2012; 7:e33730. ~es S, Sousa RM, Mendonça M, CastroFerreira-Gomes J, Ada Lopes JM. Dose-dependent expression of neuronal injury markers during experimental osteoarthritis induced by monoiodoacetate in the rat. Mol Pain 2012;8:50. Ivanavicius SP, Ball AD, Heapy CG, Westwood FR, Murray F, Read SJ. Structural pathology in a rodent model of osteoarthritis is associated with neuropathic pain: increased expression of ATF-3 and pharmacological characterisation. Pain 2007;128:272e82. Orita S, Ishikawa T, Miyagi M, Ochiai N, Inoue G, Eguchi Y, et al. Pain-related sensory innervation in monoiodoacetate-induced osteoarthritis in rat knees that gradually develops neuronal injury in addition to inflammatory pain. BMC Musculoskelet Disord 2011;12:134.
39. Sherwin CM, Huag E, Terkelsen N, Vadgma M. Studies on the motivation for burrowing by laboratory mice. Appl Anim Behav Sci 2004;88:343e58. 40. Antoniou K, Malamas M, Drosos A. Clinical pharmacology of celecoxib, a COX-2 selective inhibitor. Expert Opin Pharmacother 2007;8:1719e32. 41. Jones CK, Peters SC, Shannon HE. Efficacy of duloxetine, a potent and balanced serotonergic and noradrenergic reuptake inhibitor, in inflammatory and acute pain models in rodents. J Pharmacol Exp Ther 2005;312:726e32. 42. Combe R, Bramwell S, Field MJ. The monosodium iodoacetate model of osteoarthritis: a model of chronic nociceptive pain in rats? Neurosci Lett 2004;370:236e40. 43. Ahn K, Smith SE, Liimatta MB, Beidler D, Sadagopan N, Dudley DT, et al. Mechanistic and pharmacological characterization of PF-04457845: a highly potent and selective fatty acid amide hydrolase inhibitor that reduces inflammatory and noninflammatory pain. J Pharmacol Exp Ther 2011;338: 114e24. 44. Sokolove J, Lepus CM. Role of inflammation in the pathogenesis of osteoarthritis: latest findings and interpretations. Ther Adv Musculoskelet Dis 2013;5:77e94. 45. Sorge RE, Martin LJ, Isbester KA, Sotocinal SG, Rosen S, Tuttle AH, et al. Olfactory exposure to males, including men, causes stress and related analgesia in rodents. Nat Methods 2014;11:629e32.
Deficits in spontaneous burrowing behavior in the rat bilateral monosodium iodoacetate model of osteoarthritis: an objective measure of pain-related behavior and analgesic efficacy L.A. Bryden, J.R. Nicholson, H. Doods, A. Pekcec* Boehringer Ingelheim Pharma GmbH & Co. KG, Div. Research Germany, Birkendorfer Strasse 65, 88397 Biberach an der Riss, Germany
a r t i c l e i n f o
s u m m a r y
Article history: Received 8 October 2014 Accepted 1 May 2015
Objective: To characterize deficits in burrowing behavior e an ethologically-relevant rodent behavior e in the monosodium iodoacetate (MIA) rat model of osteoarthritis (OA), and the sensitivity of these deficits to reversal by analgesic drugs of both prototypical and novel mechanisms of action. A second objective was to compare the burrowing assay to a spontaneous locomotor activity (sLA) assay. Method: Male Wistar Han rats (200e220 g) received intrarticular (i.a.) injections of MIA or saline for sham animals. A deficit in the amount of sand burrowed from steel tubes filled with 2.5 kg of sand was used as a measure of pain-related behavior, and sensitivity to reversal of these deficits by analgesic drugs was assessed in bilaterally MIA-injected rats. Results: Bilateral MIA injections induced a significant impairment of burrowing behavior, which was concentration-dependent. The temporal pattern of the deficits was biphasic: a large deficit at 3 days post-injection, resolving by day 14 and returning at the 21 and 28 day time points. At the 3 day time point ibuprofen, celecoxib and an anti-nerve growth factor (NGF) monoclonal antibody (mAb) were able to significantly reinstate burrowing behavior, whereas the fatty acid amide hydrolase (FAAH) inhibitor PF-04457845 and morphine displayed no reversal effect. Morphine impaired burrowing behavior at 3 mg/kg in sham animals. Deficits in rearing frequency in the locomotor activity assay proved irreversible by analgesics. Conclusion: Burrowing behavior provides an objective, non-reflexive read-out for pain-related behavior in the MIA model that has predictive validity in detecting analgesic efficacy of nonsteroidal antiinflammatory drugs (NSAIDs) and an anti-NGF mAb. © 2015 Osteoarthritis Research Society International. Published by Elsevier Ltd. All rights reserved.
Keywords: Burrowing Monosodium iodoacetate Spontaneous behavior Pain Rat
Introduction The prevalence of osteoarthritis (OA) is estimated to be 50% in people aged 65 and older1 and with an aging population the cost and health impact of this disease will continue to rise. The intraarticular (i.a.) injection of monosodium iodoacetate (MIA) into the rat knee joint produces histological changes representative of those seen in human OA2. Concomitant with the early and late
* Address correspondence and reprint requests to: A. Pekcec, Boehringer Ingelheim Pharma GmbH & Co. KG, Div. Research Germany, Birkendorfer Strasse 65, 88397 Biberach an der Riss, Germany. Tel: 49-7351-54-92312; Fax: 49-7351-545128. E-mail addresses: [email protected] (L.A. Bryden), janet. [email protected] (J.R. Nicholson), [email protected] (H. Doods), [email protected] (A. Pekcec).
histological changes in the MIA model are pain-related behaviors3, making it a useful model for assessing the analgesic efficacy of drugs. Pain is the main clinical manifestation of OA. The current lack of effective disease modifying agents to target the aetiology of OA4 means the use of analgesic drugs is the mainstay treatment for alleviating the impact of the disease on daily life, namely nonsteroidal anti-inflammatory drugs (NSAIDs) and weak opioids5. The main techniques for assessing pain-like behaviors in the MIA model are measures of mechanical hypersensitivity: shifts in weight bearing (WB) from the ipsilateral affected to contralateral unaffected knee, distal mechanical hyperalgesia and tactile allodynia6e8. A pressing issue in the pain field is the lack of translation between preclinical and clinical findings, which has resulted in relatively few safe and effective analgesics being developed (reviewed by Blackburn-Munro 20049 and Mogil, 200910).
http://dx.doi.org/10.1016/j.joca.2015.05.001 1063-4584/© 2015 Osteoarthritis Research Society International. Published by Elsevier Ltd. All rights reserved.
1606
L.A. Bryden et al. / Osteoarthritis and Cartilage 23 (2015) 1605e1612
Although there is evidence that current pain assays have some predictive validity from rat to human11, it is becoming increasingly apparent that assays utilizing spontaneous rather than evoked/reflexive measures are needed to assess the global impact of pain beyond hypersensitivity12,13. In addition to sharp pain evoked by movement, patients with OA rank constant and aching pain among the most distressing features of the disease14. How well the current evoked/reflexive assays of pain-like behaviors in rodents account for this persistent pain is unclear9. Furthermore, clinical assessments of chronic pain are focusing increasingly on the multifaceted nature of pain, such as the effect on emotion and physical function15, which are not assessed preclinically by reflexive assays. Burrowing is an innate rodent behavior indicative of animal wellbeing that is conserved across various strains of rat16e19 and mice20e22. Deficits in burrowing behavior occur in preclinical rat models of inflammatory and neuropathic pain and can be reversed by analgesics16,18,19,23,24. This behavior, therefore, is a useful preclinical measure of non-evoked pain. Burrowing may encompass the supraspinal mechanisms that contribute to pain phenotypes, both sensory and affective, as well as being more ethologically relevant than the commonly used assays measuring hypersensitivity25. Here we show deficits in burrowing behavior in the bilateral MIA model of OA and sensitivity of these deficits to reversal by prototypical and novel analgesics. To our knowledge, this is the first published article demonstrating deficits in burrowing behavior in the MIA model. Additionally, burrowing behavior was compared to another non-reflexive readout: spontaneous locomotor activity (sLA). This has been shown previously to be impaired in various preclinical animal models and to be sensitive to pharmacological modulation26e29. Methods Animals 594 male Wistar Han rats (200e220 g; Charles River Laboratories, Germany) were used for all experiments and were housed in groups of four with food and water ad libitum with a 12 h light/dark cycle. All animal experimental protocols were authorized by the Local Animal Care and Use Committee and carried out according to the local animal care guidelines, AAALAC regulations, and the USDA Animal Welfare Act. MIA injections Rats were briefly anaesthetized with 5% isoflurane (Abbott Laboratories, Wiesbaden, Germany) followed by 3% maintenance, after which an i.a. injection of 3 mg of MIA (SigmaeAldrich, Steinheim, Germany) dissolved in 50 mL of 0.9% physiological saline was performed into the femorotibial joint. Sham rats received 50 mL injections of physiological saline. All injections for pharmacology experiments were bilateral, except for one model conditions studies in which rats received either uni- or bilateral injections. Drugs and drug administration The analgesic drugs tested were: morphine (Caelo, Germany), ibuprofen (Sigma Aldrich, Steinheim, Germany), celecoxib (LKT Laboratories Inc., St. Paul, MN, USA) and the fatty acid amide hydrolase (FAAH) inhibitor PF-04457845 (synthesized by Boehringer Ingelheim). All drugs were administered orally (p.o.) with 0.5% (w/v) natrosol and 0.1% (v/v) tween-80 (9:1 ratio) as vehicle, except morphine which was injected subcutaneously (s.c.) in the interscapular area with 0.9% physiological saline as vehicle. All
drugs were administered in a volume of 2 mL/kg except PF04457845, which was administered in a volume of 4 mL/kg. For the generation of the anti-nerve growth factor (NGF), monoclonal antibody (mAb) variable domains were extracted from the patent application WO 2004/058184 A2 (applicant: Rinat Neuroscience Corporation) and processed as described by Hezareh et al., 200130. The antibody was administered s.c., with phosphate buffered saline as vehicle. Burrowing training and burrowing experiments For all burrowing experiments steel tubes (32 cm in length and 10 cm in diameter) were filled with 2.5 kg of quartz sand and placed in Plexiglas cages (600 340 200 mm). The open-end of the tube was elevated 6 cm from the floor of the cage. Training for each experiment was carried in 2 phases: social facilitation (SF) and individual training (IT). For SF rats were placed in pairs in a cage for 2 h on two consecutive days. The amount of sand burrowed by each pair was measured and if a pair burrowed less than 1500 g of sand one of the pair was swapped with a rat from a pair that had burrowed greater than 1500 g for the second SF day. For IT rats were placed alone in the burrowing set-up for 30 min per day and the average amount burrowed over 3 days was calculated to attain a baseline burrowing performance value. Before each animal was given MIA or sham intrarticular injections they were assigned to groups to ensure that each treatment group for an experiment had a comparable baseline burrowing value. For the model conditions study rats were allowed to individually burrow for 30 min 3 days after MIA or sham injections were performed and the amount of sand burrowed was recorded. For the time-course experiment burrowing performance was assessed 3, 14, 21 and 28 days post-injection. For all pharmacology experiments burrowing behavior was measured 3 days after MIA injection, and animals were placed in the burrowing set-up after the appropriate pre-treatment time. Exclusions During training any animal that has a baseline burrowing value of less than 1000 g or a standard deviation of burrowing (SD) greater than 450 g was excluded to ensure high and stable baselines. To ensure that all rats used in pharmacology experiments were in a comparable pain state burrowing values were also measured 1 day after MIA injection and any animal with a burrowing value greater than 1000 g was excluded from pharmacology at day 3. In total, exclusions after training and after day 1 accounted for around 5% of rats per study. Locomotor activity sLA was measured for 30 min with an automated monitoring system (TruScan Activity Monitor version 2.0, Coulbourn Instruments, Allentown, PA, USA). Each monitoring system was an enclosed 43 cm2 arena with two levels of sensory photobeams; one level elevated 3 cm from the floor of the arena and the other 14 cm, measuring horizontal and vertical activity respectively. Each level of the detection system was equipped with 16 photobeams per wall of the arena, spaced 2.5 cm apart. The parameters measured by the system were: ambulatory horizontal distance moved (cm), rearing frequency and rearing time (s). Prior to testing, all rats were habituated in an annexe to the testing room. Statistical analysis All statistical processing was performed in GraphPad Prism version 6.0 (San Diego, CA, USA). For all analyses P < 0.05 was
L.A. Bryden et al. / Osteoarthritis and Cartilage 23 (2015) 1605e1612
1607
considered statistically significant. To control for familywise type I error rates a Bonferroni correction was applied where appropriate. A 1-way analysis of variance (ANOVA) with Bonferroni's post hoc was used to analyze the model conditions studies investigating whether uni- or bilateral MIA injections were most appropriate and the concentration-dependency of burrowing deficits. A Bonferroni cut-off of P < 0.025 was applied: 0.05/(number of comparisons per family (2)/number of families(1)). The time-course study was analyzed using 2-way repeated measures ANOVA. Significant differences between sham animals and MIA-injected animals on the same day were ascertained using Bonferroni's post hoc. A Bonferroni cut-off of P < 0.025 was applied: 0.05/(number of comparisons per family (4)/number of families(2)). The statistical significance of deficits in locomotor activity parameters was analyzed using an unpaired Student's t test. For all pharmacology studies (burrowing and locomotor activity) data were analyzed using a 2-way ANOVA. To determine that the MIA vehicle group was significantly lower than the sham vehicle group and to determine which doses of compound were efficacious in reversing burrowing deficits a Dunnett's post hoc multiple comparisons was used, comparing with the MIA vehicle control group. All data are expressed as means and corresponding 95% confidence intervals (CI). Gaussian distribution was analyzed using a ShapiroeWilk normality test. Results Establishment of model conditions To determine whether unilateral or bilateral MIA injections were most appropriate for the burrowing assay an experiment was conducted for comparison. Injection of MIA bilaterally resulted in a 60% reduction in burrowing performance compared with the sham bilateral control group 3 days post-injection, whereas unilaterally MIA-injected rats displayed no deficit in burrowing performance at day 3 [Fig. 1(A)], or at later time points of 14, 21 and 28 days [Fig. 1(C)]. We investigated the effect of saline sham injections on burrowing performance and found that neither unilateral nor bilateral injections of saline influenced burrowing behavior [Fig. 1(A)]. Next we wanted to establish the concentrationresponsiveness of MIA-induced burrowing deficits. Concentrations of 0.3, 1 and 3 mg/knee were selected based on previous findings in WB6. The 0.3 mg/knee concentration of MIA did not significantly impair burrowing performance, whereas 1 and 3 mg/ knee induced significant deficits [Fig. 1(B)]. Although 1 and 3 mg/ knee concentrations produced comparable deficits, we decided on 3 mg/knee for pharmacological experiments as this concentration produced the most robust deficit with the least variance. To establish time points for pharmacological experiments burrowing performance was assessed 3, 14, 21 and 28 days after the bilateral injection of MIA [Fig. 1(C)]. As seen in [Fig. 1(C)], a robust deficit occurred at 3 days post-injection, which resolved back to baseline values by the 14 day time point. However, the deficits at the 21 and 28 day time points in the pharmacology experiments were found not to be large enough to provide an assay window to titrate a reversal effect. Therefore, it was decided to focus on the 3 day time point for all pharmacology. After establishing the model conditions for the burrowing assay, the same conditions were chosen for locomotor activity studies to allow for comparisons between these two assays. Three days after bilateral injection of MIA (3 mg/knee) deficits in distance moved [Fig. 2(A)], rearing frequency [Fig. 2(B)] and rearing time [Fig. 2(C)] occurred. We decided to use rearing frequency as the parameter for pharmacology studies as it had the largest deficit; rearing frequency was 48% lower in MIA injected animals than in sham animals.
Fig. 1. Effect of MIA on burrowing behavior. (A) Amount of sand burrowed by naïve rats and 3 days after intra-articular (i.a.) of saline or MIA (3 mg/knee). Rats either received unilateral (right knee) or bilateral injections of saline/MIA. Data analyzed with 1-way ANOVA followed by Bonferroni's post hoc multiple comparisons (****P < 0.0001 vs sham bilateral group, n ¼ 12 per group). (B) Concentration-response to bilateral i.a. injections of MIA 3 days after injection. Data analyzed with 1-way ANOVA followed by Bonferroni's post hoc multiple comparisons (*P < 0.05 and **P < 0.01 vs sham group, n ¼ 10e11 per group). (C) Time-course of burrowing performance after unilateral and bilateral i.a. MIA injections (3 mg/knee). Data were analyzed with 2-way repeated measures ANOVA (MIA time interaction: F4,48 ¼ 12.93, P < 0.0001), followed by Bonferroni's post hoc multiple comparisons (significant differences compared with sham group on same day: **P < 0.01,****P < 0.0001, n ¼ 12 per group). BL ¼ baseline.
Rearing time was similarly reduced by MIA injection, whereas the deficit in distance moved was more modest (21% lower than sham group). At day 28 there was no statistically significant effect of MIA on any locomotor activity parameter (Supplementary Table 2). Effect of NSAIDs, morphine and gabapentin on deficits in burrowing performance Nonsteroidal anti-inflammatory drugs (NSAIDs) and opioids are categories of drugs that are currently used to treat pain in OA
1608
L.A. Bryden et al. / Osteoarthritis and Cartilage 23 (2015) 1605e1612
Effect of an anti-NGF mAb and the FAAH inhibitor PF-04457845 on deficits in burrowing performance An anti-NGF mAb was tested in the burrowing in light of recent phase III clinical trials with an antibody of the same mechanism of action31. When given 24 h before measurement of burrowing performance the antibody had no effect on sham animals and was able to reverse the deficit in burrowing behavior induced by MIA to a level similar to that of the respective sham group, at a dose of 9 mg/ kg [Fig. 4(A)]. The FAAH inhibitor PF-04457845 is another analgesic with a novel mechanism of action tested in burrowing because of recent negative phase II clinical trials in patients with knee OA32. At all doses tested (10, 30 and 100 mg/kg) there was no reversal of burrowing deficits observed in MIA injected rats and also no effect on burrowing performance in the sham animals [Fig. 4(B)]. Locomotor activity pharmacology Supplementary Table 1 shows that in our hands no analgesic effect of any drugs tested was detectable in the rearing frequency parameter of the locomotor activity assay, which is in contrast to the pharmacological sensitivity evident in the burrowing assay. The 10 mg/kg dose of morphine caused a significant reduction in rearing frequency compared with the sham vehicle group; however, unlike for burrowing, there was no significant reduction in rearing frequency induced by the 3 mg/kg dose (P ¼ 0.18) compared with sham vehicle. Discussion
Fig. 2. Locomotor activity 3 days after bilateral injection of saline or 3 mg/knee MIA. Distance moved (A), rearing frequency (B) and rearing time (C) was monitored over a 30 min period. Data were analyzed with unpaired Student's t test (*P < 0.05 vs sham group, n ¼ 8 per group).
clinically5, so were assessed to determine their efficacy in reversing deficits in burrowing performance 3 days after the injection of MIA. Both the non-selective cyclooxygenase (COX) inhibitor ibuprofen and the COX-2 selective inhibitor celecoxib were efficacious in reinstating deficits in burrowing performance at doses of 10 and 30 mg/kg for ibuprofen, and 3, 10 and 30 mg/kg for celecoxib [Fig. 3(A) and (B), respectively]. The increase in burrowing performance elicited by ibuprofen and celecoxib at these doses was a minimum of 50% when compared with the MIA vehicle group. Neither morphine or gabapentin reversed burrowing deficits at any dose tested [Fig. 3(C) and (D)], and morphine impaired burrowing performance at 3 mg/kg in the sham group.
The initial aim of this study was to investigate whether burrowing was impaired by i.a. injection of MIA and could therefore provide a measure of pain-related behavior in this pre-clinical model of OA. Our model conditions studies showed that injecting MIA into only one knee joint was insufficient to cause depression of burrowing behavior, however, bilateral injection resulted in a robust deficit in performance. One possibility for this is that when MIA injections are given unilaterally compensation occurs through use of the three unaffected limbs to burrow. In contrast, previous studies investigating the effect of intraplantar (i.pl) complete Freund's adjuvant (CFA), i.a. CFA and nerve ligation models of neuropathic pain found unilateral injury to be sufficient to depress burrowing behavior16,18,19,24. It could be inferred that the MIA model is less severe in its pain phenotype and effect on animal well-being, therefore bilateral injury is needed to induce a deficit. However, unilateral injection of MIA is sufficient to induce deficits in wheel-running33 and locomotor activity34 in rats, perhaps suggesting burrowing is less sensitive to MIA-induced joint pain than these assays. Depression of locomotor activity after unilateral MIA injection e shown by More et al. (2013)34 e is in contrast to studies in our laboratory where no deficit occurred at the early (day 3) or late (day 28) time points (data not shown). The temporal pattern of MIA-induced deficits in burrowing performance was biphasic, which is similar to reported findings in WB6 and biotelemetric assessment of mobility (BAM)27. It has been suggested from behavioral and histological findings that the early phase of the MIA model (day 3 in our study) is predominantly pain mediated by inflammation6,7, whereas the later stages of the model represent pain of a different aetiology, perhaps neuropathic in nature35. This is inferred from observed upregulation of nerve injury markers at the late phase time points36e38 and sensitivity to drugs used to treat neuropathic pain37. However, it is important to point out that previous studies mentioned here used unilateral MIA injections, whereas we
L.A. Bryden et al. / Osteoarthritis and Cartilage 23 (2015) 1605e1612
1609
Fig. 3. The effect of NSAIDs, morphine and gabapentin on burrowing performance 3 days after bilateral 3 mg/knee MIA or sham injections. Data analyzed with 2-Way ANOVA followed by Dunnett's post hoc multiple comparisons. (A) Ibuprofen (p.o.) administered 2 h before burrowing performance was measured (MIA dose interaction: F3,66 ¼ 2.785, P < 0.05; ##P < 0.01 vs sham vehicle, *P < 0.05, **P < 0.01 vs MIA vehicle). N ¼ 9e10 per group. (B) Celecoxib (p.o.) administered 2 h before burrowing performance was measured (MIA dose interaction: F3,68 ¼ 2.82, P < 0.05; ####P < 0.0001 vs sham vehicle, *P < 0.05,**P < 0.01 vs MIA vehicle). N ¼ 8e11 per group. (C) Morphine (s.c.) administered 1 h before burrowing performance was measured (#P < 0.05, ##P < 0.01 vs sham vehicle). N ¼ 8e9 per group. (D) Gabapentin (p.o.) administered 2 h before burrowing performance was measured (main effect of MIA: F1,62 ¼ 48.21, ****P < 0.0001). N ¼ 8e9 per group.
injected bilaterally e differences in pathophysiology between unilateral and bilateral injections should therefore be borne in mind. Taking this into account we suggest that the early phase deficits in burrowing represent inflammatory pain, which resolves by day 14 and a return of deficits in the late phase could be chronic/ neuropathic in nature. We tested the neuropathic pain drug gabapentin at day 3 but found no effect at any dose tested, which is supportive of primarily inflammatory pain in the early phase of the MIA model. Unfortunately, the assay window for pharmacology at day 21 and 28 was found to be insufficient, therefore it was not possible to use gabapentin to discern whether this late phase was neuropathic in nature. It was estimated that 20e25 rats per treatment group would be required to determine a pharmacological effect at day 28. These characterization studies show that deficits in burrowing behavior can be used as an objective measure of pain-related behavior in the MIA model of OA. Burrowing is a self-rewarding behavior39 that is not appetite-driven and is indicative of animal well-being17. It is suggested that burrowing is able to measure the negative impact of pain on motivation (affective aspect) in addition to pain evoked by the act of burrowing (sensory aspect). Recent reviews have highlighted the need to assess the global impact of pain in preclinical models of pain beyond measures of hypersensitivity9,10,12,13 through use of more ethological rodent behaviors25. Burrowing may provide one means of achieving this. However, further characterization in rodent models is needed to disinter what aspects of pain burrowing measures compared with other assays and whether results from burrowing can be differentially interpreted. To investigate whether MIA-induced deficits in burrowing performance were sensitive to reversal by analgesics we tested a range of compounds that work by mechanisms of action currently used to
treat pain associated with OA clinically5. Both the non-selective COX inhibitor ibuprofen and the COX-2 selective inhibitor celecoxib (375fold selectivity for COX-240) were efficacious in reversing deficits in burrowing performance at low doses of 10 and 3 mg/kg, respectively. Similarly, previous published findings found that a relatively low dose of ibuprofen (30 mg/kg) was efficacious in reversing deficits in burrowing induced by i.pl CFA16. Both our study and this previous work demonstrate that burrowing may have higher sensitivity to the analgesic effect of ibuprofen than evoked endpoints. In the carrageenan model of inflammatory pain the minimally effective dose (MED) of ibuprofen to reverse mechanical allodynia was 300 mg/kg and to reverse thermal hyperalgesia 1000 mg/kg41, which is considerably higher than the MED in burrowing. However, such comparisons should be approached with caution due to the differences in the underlying neurophysiological mechanisms between pain models and the strain of rat used for experimentation. Morphine did not significantly reverse deficits in burrowing performance at any dose tested, and 3 mg/kg caused a significant reduction in burrowing performance in sham animals, presumably due to sedative side effects. This is in contrast to a previous published study which showed morphine to be efficacious in reversing burrowing deficits induced by i.a. CFA at 1 and 3.16 mg/kg, without reduction in burrowing performance in naïve rats19. However, the route of administration used was intraperitoneal (i.p.) in this previous study, whereas in this study morphine was administered s.c. The sedative side effects of morphine at 3 mg/kg was reflected in previous sLA experiments28 and in our sLA studies significant reduction in rearing frequency occurred at 10 mg/kg. In WB, the efficacy of morphine in normalizing deficits induced by MIA is observed at 3 mg/kg8,42 e a dose which caused a reduction in burrowing performance e and reversal of mechanical hyperalgesia occurs at 6 mg/kg7. There is the possibility of false
1610
L.A. Bryden et al. / Osteoarthritis and Cartilage 23 (2015) 1605e1612
Fig. 4. The effect of the anti-NGF mAb and the FAAH inhibitor PF-04457845 on burrowing performance 3 days after bilateral 3 mg/knee MIA or sham injections. Data analyzed with 2-Way ANOVA followed by Dunnett's post hoc multiple comparisons. (A) anti-NGF mAb administered 24 h before burrowing performance was measured (MIA dose interaction: F1,28 ¼ 7.273, P < 0.05; ###P < 0.001 vs sham vehicle, **P < 0.01 vs MIA vehicle). N ¼ 8 per group. (B) PF-04457845 (p.o.) administered 2 h before burrowing performance was measured (main effect of MIA: F1,65 ¼ 54.65, ****P < 0.0001). N ¼ 8e10 per group.
positives when morphine is tested at these doses because sedation may impair motor function and tone. This impairment could cause a shift in weight onto the ipsilateral injected knee when WB is measured, in addition to any analgesia-mediated effect, and may also inhibit withdrawal reflexes. This highlights an advantage of assays that rely on the reinstatement of pain-depressed behaviors by analgesics: there is no risk of detecting false positives when drugs that potentially impair motor functioning are tested25. The anti-NGF mAb tanezumab has been shown to be efficacious in phase III clinical trials for OA pain, improving pain scores (including global pain assessment) and physical functioning32. In our study an anti-NGF mAb with the same mechanism of action was effective in reversing deficits in burrowing performance in the MIA model. This further demonstrates the predictive validity of the burrowing assay. PF-04457845, a FAAH inhibitor, showed no efficacy at any dose tested in burrowing or sLA assays. Although no published data demonstrating an effect of this compound at the 3 day time point in the MIA model are available there is data showing a positive effect in ameliorating mechanical hyperalgesia in the i.pl CFA model of inflammatory pain, and at the 14 day time point in the MIA model43. However, results from a clinical trial with PF-04457845 failed to show effective analgesia despite raised levels of endocannabinoids32 indicative of FAAH inhibition. The authors suggest a lack of translation between animal models and clinical data in light of the studies showing a potent effect of the compound preclinically. Whether this lack of translation is a result of a fault in the
assessment of pain in preclinical models is not ascertainable at this stage. Parallel studies investigating both spontaneous and evoked endpoints in the MIA model would allow for more comprehensive comparisons. In contrast to burrowing, no analgesic effect of any drug was observed in sLA. Previous pharmacological characterization of MIAinduced deficits in rearing frequency demonstrated an acute analgesic effect of analgesics such as celecoxib and morphine29. In our hands, reversal of burrowing deficits provides a more sensitive read-out for analgesic drug effect, which is in line with the findings comparing i.a. CFA-induced burrowing deficits and locomotor activity18. Furthermore, deficits in burrowing performance are independent of motor impairment, as measured by the locomotor activity assay; we found no correlation between burrowing behavior or any locomotor activity parameters 3 days after i.a. MIA injections (data not shown). The rationale behind the development of nonreflexive and ethologically relevant behavioral paradigms is to improve translation of preclinical to clinical findings. To complement this, the underlying pathophysiology of the animal model used should be as close to the human condition as possible. In that regard, the late phase (28 day time point) of the MIA model is perhaps more translational as it is suggested to model the chronic, noninflammatory pain that is more relevant to OA when it presents clinically. However, it must be noted that denotating OA as ‘noninflammatory’ may refer to a subset of patients, as there is evidence for inflammatory markers in some cases44. This highlights the complexity and heterogeneity of OA as a disease. Nevertheless, further studies are warranted to optimize the burrowing paradigm to enable pharmacological testing at the 28 day time point. In conclusion, these data demonstrate that burrowing provides an objective read-out for pain in the early phase of the MIA model and is sensitive to reversal by NSAIDs and an anti-NGF mAb. As well as having the advantage of being objective, the lack of experimenter presence during the test period in the burrowing assay reduces the influence of stress on rodent behavior. This is illustrated by a recent study showing that rodent pain responses are inhibited by the presence of a male experimenter due to stress-induced by axillary secretions45. We also found that burrowing has superior sensitivity to pharmacological modulation over sLA, which is in line with findings in the i.a. CFA model of rheumatoid arthritis19, further supporting the utility of burrowing as a preclinical pain assay. Contributions HD and JN contributed by conception and design of the studies, interpretation of data and critical revision of the article. LB generated, analyzed and interpreted the data and drafted the article. AP contributed by conception and design of the studies, interpretation of data and writing the manuscript. Role of the funding source All funding was provided by Boehringer Ingelheim Pharma GmbH & Co KG. Competing interests All authors were all employees of Boehringer Ingelheim Pharma GmbH & Co KG at the time of data collection and manuscript writing. Acknowledgments The authors would like to thank Andreas Sterner for technical assistance with locomotor activity studies, Anna-Lena Martin for advisory support with statistical analysis and Torsten Lamla for
L.A. Bryden et al. / Osteoarthritis and Cartilage 23 (2015) 1605e1612
providing the anti-NGF mAb. The current study was performed as part of a contribution of Boehringer Ingelheim Pharma GmbH & Co. KG to the “EUROPAIN” project within the European Consortium “Innovative Medicine Initiative”.
17. 18.
Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.joca.2015.05.001.
19.
References 1. Hinton R, Moody RL, Davis AW, Thomas SF. Osteoarthritis: diagnosis and therapeutic considerations. Am Fam Physician 2002;65:841e8. 2. Guzman R, Evans MG, Bove S, Morenko B, Kilgore K. Monoiodoacetate-induced changes in subchondral bone and articular cartilage of rat femorotibial joints: an animal model of osteoarthritis. Toxicol Pathol 2003;31:619e24. 3. Liu P, Okun A, Ren J, Guo RC, Ossipov MH, Xie J, et al. Ongoing pain in the MIA model of osteoarthritis 2011;493:72e5. 4. Le Graverand-Gastineau MP. Disease modifying osteoarthritis drugs: facing development challenges and choosing molecular targets. Curr Drug Targets 2010;11:528e35. 5. Zhang W, Nuki G, Moskowitz RW, Abramson S, Altman RD, Arden NK, et al. OARSI recommendations for the management of hip and knee osteoarthritis: part III: changes in evidence following systematic cumulative update of research published through January 2009. Osteoarthritis Cartilage 2010;18: 476e99. 6. Bove SE, Calcaterra SL, Brooker RM, Huber CM, Guzman RE, Juneau PL, et al. Weight bearing as a measure of disease progression and efficacy of anti-inflammatory compounds in a model of monosodium iodoacetate-induced osteoarthritis. Osteoarthritis Cartilage 2003;11:821e30. 7. Fernihough J, Gentry C, Malcangio M, Fox A, Rediske J, Pellas T, et al. Pain related behaviour in two models of osteoarthritis in the rat knee. Pain 2004;112:83e93. 8. Pomonis JD, Boulet JM, Gottshall SL, Phillips S, Sellers R, Bunton T, et al. Development and pharmacological characterization of a rat model of osteoarthritis pain. Pain 2005;114: 339e46. 9. Blackburn-Munro G. Pain-like behaviours in animals e how human are they? Trends Pharmacol Sci 2004;25:299e305. 10. Mogil JS. Animal models of pain: progress and challenges. Nat Rev Neurosci 2009;10:283e94. 11. Whiteside GT, Adedoyin A, Leventhal L. Predictive validity of animal pain models? A comparison of the pharmacokineticpharmacodynamic relationship for pain drugs in rats and humans. Neuropharmacology 2008;54:767e75. 12. Mogil JS, Crager SE. What should we be measuring in behavioral studies of chronic pain in animals? Pain 2004;112:12e5. 13. Vierck CJ, Hansson PT, Yezierski RP. Clinical and pre-clinical pain assessment: are we measuring the same thing? Pain 2008;135:7e10. 14. Hawker GA, Stewart L, French MR, Cibere J, Jordan JM, March L, et al. Understanding the pain experience in hip and knee osteoarthritisean OARSI/OMERACT initiative. Osteoarthritis Cartilage 2008;16:415e22. 15. Dworkin RH, Turk DC, Farrar JT, Haythornthwaite JA, Jensen MP, Katz NP, et al. Core outcome measures for chronic pain clinical trials: IMMPACT recommendations. Pain 2005; 113:9e19. 16. Andrews N, Legg E, Lisak D, Issop Y, Richardson D, Harper S, et al. Spontaneous burrowing behaviour in the rat is reduced
20.
21.
22. 23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
1611
by peripheral nerve injury or inflammation associated pain. Eur J Pain 2012;16:485e95. Deacon RM. Burrowing in rodents: a sensitive method for detecting behavioral dysfunction. Nat Protoc 2006;1:118e21. Rutten K, Schiene K, Robens A, Leipelt A, Pasqualon T, Read SJ, et al. Burrowing as a non-reflex behavioural readout for analgesic action in a rat model of sub-chronic knee joint inflammation. Eur J Pain 2014;18:204e12. Rutten K, Robens A, Read SJ, Christoph T. Pharmacological validation of a refined burrowing paradigm for prediction of analgesic efficacy in a rat model of sub-chronic knee joint inflammation. Eur J Pain 2014;18:213e22. Jirkof P, Cesarovic N, Rettich A, Nicholls F, Seifert B, Arras M. Burrowing behavior as an indicator of post-laparotomy pain in mice. Front Behav Neurosci 2010;4:165. Jirkof P, Leucht K, Cesarovic N, Caj M, Nicholls F, Rogler G, et al. Burrowing is a sensitive behavioural assay for monitoring general wellbeing during dextran sulfate sodium colitis in laboratory mice. Lab Anim 2013;47:274e83. Jirkof P. Burrowing and nest building behavior as indicators of well-being in mice. J Neurosci Methods 2014;234:139e46. Huang W, Calvo M, Karu K, Olausen HR, Bathgate G, Okuse K, et al. A clinically relevant rodent model of the HIV antiretroviral drug stavudine induced painful peripheral neuropathy. Pain 2013;154:560e75. Lau W, Dykstra C, Thevarkunnel S, Silenieks LB, de Lannoy IA, Lee DK, et al. A back translation of pregabalin and carbamazepine against evoked and non-evoked endpoints in the rat spared nerve injury model of neuropathic pain. Neuropharmacology 2013;73:204e15. Andrews N, Harper S, Issop Y, Rice AS. Novel, nonreflex tests detect analgesic action in rodents at clinically relevant concentrations. Ann N Y Acad Sci 2011;1245:11e3. Cobos EJ, Ghasemlou N, Araldi D, Segal D, Duong K, Woolf CJ. Inflammation-induced decrease in voluntary wheel running in mice: a nonreflexive test for evaluating inflammatory pain and analgesia. Pain 2012;153:876e84. Guingamp C, Gegout-Pottie P, Philippe L, Terlain B, Netter P, Gillet P. Mono-iodoacetate-induced experimental osteoarthritis: a dose-response study of loss of mobility, morphology, and biochemistry. Arthritis Rheum 1997;40:1670e9. Matson DJ, Broom DC, Carson SR, Baldassari J, Kehne J, Cortright DN. Inflammation-induced reduction of spontaneous activity by adjuvant: a novel model to study the effect of analgesics in rats. J Pharmacol Exp Ther 2007;320:194e201. Nagase H, Kumakura S, Shimada K. Establishment of a novel objective and quantitative method to assess pain-related behavior in monosodium iodoacetate-induced osteoarthritis in rat knee. J Pharmacol Toxicol Methods 2012;65:29e36. Hezareh M, Hessell AJ, Jensen RC, van de Winkel JG, Parren PW. Effector function activities of a panel of mutants of a broadly neutralizing antibody against human immunodeficiency virus type 1. J Virol 2001;75:12161e8. Brown MT, Murphy FT, Radin DM, Davignon I, Smith MD, West CR. Tanezumab reduces osteoarthritic hip pain: results of a randomized, double-blind, placebo-controlled phase III trial. Arthritis Rheum 2013;65:1795e803. Huggins JP, Smart TS, Langman S, Taylor L, Young T. An efficient randomised, placebo-controlled clinical trial with the irreversible fatty acid amide hydrolase-1 inhibitor PF04457845, which modulates endocannabinoids but fails to induce effective analgesia in patients with pain due to osteoarthritis of the knee. Pain 2012;153:1837e46. Stevenson GW, Mercer H, Cormier J, Dunbar C, Benoit L, Adams C, et al. Monosodium iodoacetate-induced osteoarthritis
1612
34.
35.
36.
37.
38.
L.A. Bryden et al. / Osteoarthritis and Cartilage 23 (2015) 1605e1612
produces pain-depressed wheel running in rats: implications for preclinical behavioral assessment of chronic pain. Pharmacol Biochem Behav 2011;98:35e42. More AS, Kumari RR, Gupta G, Lingaraju MC, Balaganur V, Pathak NN, et al. Effect of iNOS inhibitor S-methylisothiourea in monosodium iodoacetate-induced osteoathritic pain: implication for osteoarthritis therapy. Pharmacol Biochem Behav 2013;103:764e72. Thakur M, Rahman W, Hobbs C, Dickenson AH, Bennett DL. Characterisation of a peripheral neuropathic component of the rat monoiodoacetate model of osteoarthritis. PLoS One 2012; 7:e33730. ~es S, Sousa RM, Mendonça M, CastroFerreira-Gomes J, Ada Lopes JM. Dose-dependent expression of neuronal injury markers during experimental osteoarthritis induced by monoiodoacetate in the rat. Mol Pain 2012;8:50. Ivanavicius SP, Ball AD, Heapy CG, Westwood FR, Murray F, Read SJ. Structural pathology in a rodent model of osteoarthritis is associated with neuropathic pain: increased expression of ATF-3 and pharmacological characterisation. Pain 2007;128:272e82. Orita S, Ishikawa T, Miyagi M, Ochiai N, Inoue G, Eguchi Y, et al. Pain-related sensory innervation in monoiodoacetate-induced osteoarthritis in rat knees that gradually develops neuronal injury in addition to inflammatory pain. BMC Musculoskelet Disord 2011;12:134.
39. Sherwin CM, Huag E, Terkelsen N, Vadgma M. Studies on the motivation for burrowing by laboratory mice. Appl Anim Behav Sci 2004;88:343e58. 40. Antoniou K, Malamas M, Drosos A. Clinical pharmacology of celecoxib, a COX-2 selective inhibitor. Expert Opin Pharmacother 2007;8:1719e32. 41. Jones CK, Peters SC, Shannon HE. Efficacy of duloxetine, a potent and balanced serotonergic and noradrenergic reuptake inhibitor, in inflammatory and acute pain models in rodents. J Pharmacol Exp Ther 2005;312:726e32. 42. Combe R, Bramwell S, Field MJ. The monosodium iodoacetate model of osteoarthritis: a model of chronic nociceptive pain in rats? Neurosci Lett 2004;370:236e40. 43. Ahn K, Smith SE, Liimatta MB, Beidler D, Sadagopan N, Dudley DT, et al. Mechanistic and pharmacological characterization of PF-04457845: a highly potent and selective fatty acid amide hydrolase inhibitor that reduces inflammatory and noninflammatory pain. J Pharmacol Exp Ther 2011;338: 114e24. 44. Sokolove J, Lepus CM. Role of inflammation in the pathogenesis of osteoarthritis: latest findings and interpretations. Ther Adv Musculoskelet Dis 2013;5:77e94. 45. Sorge RE, Martin LJ, Isbester KA, Sotocinal SG, Rosen S, Tuttle AH, et al. Olfactory exposure to males, including men, causes stress and related analgesia in rodents. Nat Methods 2014;11:629e32.