Evolution of mirror-image pain in temporomandibular joint osteoarthritis mouse model

Abstract Mirror-image pain is a kind of pain that occurs on the contralateral side, but its pathogenesis remains unclear. Objective To develop an osteoarthritis mouse model for investigating mirror-image pain through observing nocifensive behaviors, histological changes, and nociceptive activity at days 3, 7, 14, 21, and 28 after the chemical induction of unilateral temporomandibular joint (TMJ) osteoarthritis. Methodology We randomly divided 6-week-old mice into sham and complete Freund adjuvant groups. To induce nocifensive behaviors, we applied 0.04 g of von Frey filament, 10 psi of air puff, and cold acetone on both sides of whisker pads at different days. The histology of TMJ on both sides was observed by hematoxylin/eosin staining and microcomputed tomography scanning. Furthermore, the nociceptive activity was evaluated using the phosphorylated cyclic AMP response element binding protein (pCREB) and a microglia marker at different days in the trigeminal subnucleus caudalis. Results Nocifensive behaviors against mechanical and temperature stimuli on the contralateral side became stronger than the baseline on day 28, in agreement with the elevation of the pCREB and the microglia marker in the trigeminal subnucleus caudalis. Thus, hypernociception on the contralateral side occurred at day 28. Conclusions Clearly, the TMJ model with unilateral osteoarthritis exhibited mirror-image pain. Therefore, this model is useful in investigating the pathogenesis of pain and in developing treatments.


Introduction
Mirror-image pain is the pain found on the opposite side after acquiring peripheral nerve lesion. 1,2 This type of pain is generally characterized by hypersensitivity to mechanical or thermal painful stimulus that responds even to a light touch or low-threshold stimulus. 3 Mirror-image pain occurs in 8%-15% of patients with chronic temporomandibular joint (TMJ) osteoarthritis. 4 However, the pathogenesis of mirror-image pain on TMJ osteoarthritis is insufficiently understood. Recent evidence of pain associated with mirror-image pain was mainly found on neuropathic animal models. 5,6 In chronic constriction injury of the spinal nerve or the infraorbital nerve, pain hypersensitivity on the contralateral side appeared 3 weeks after the nerve lesion. 7,8 The expression of proinflammatory cytokines (TGFβ1, IL-1β, TNFα, and IL-10) in the contralateral nerve, which was not directly injured, significantly increased 2 weeks after a nerve injury. 9 For the pathophysiology of the pain pathway, the contralateral dorsal root ganglia or trigeminal ganglia upregulate the proinflammatory cytokines 3 weeks after the nerve injury. 8,10 Furthermore, the number of glia cells in the contralateral dorsal horn or trigeminal subnucleus caudalis increases 3 weeks after nerve damage. 10,11 All of the evidence about mirror-image pain condition in neuropathic pain models suggested that proinflammatory cytokines in the peripheral nerve injury were carried by the cerebrospinal fluid to the contralateral dorsal horn, consequently activating glia cells in that horn. 12,13 The activated glia cells might potentially stimulate other glia cells and increase the excitability of pain signals in the contralateral dorsal horn. 5,14 Meanwhile, the association between the structural changes of contralateral nerves, expression of proinflammatory cytokines, and nociceptive activities in the pain mechanisms remains unclear. 7 Therefore, an animal model induced by osteoarthritis pain has been proposed to investigate the changes of contralateral structures and the pathophysiology of osteoarthritisinduced mirror-image pain condition.
Several studies on osteoarthritis-induced mirrorimage pain condition in animal models obtained consistent results. In a knee joint with osteoarthritis induced by a unilateral complete Freund adjuvant, a proinflammatory cytokine (IL-1β) was highly expressed in the contralateral synovial joint after 3 weeks of induction, but the contralateral knee joint did not morphologically change. 15 (9 cm×3 cm×3 cm), with their head and front paws exposed outside and their tail firmly fixed with a rubber, to allow them to acclimate in the restrainer for at least 30 min before the test. Nocifensive behaviors were performed on pre-injected mice at day 0 and also on mice on days 3, 7, 14, 21, and 28 after being injected with complete Freund adjuvant. We recorded their responses 12 times by randomly starting and changing either ipsilateral or contralateral side of the whisker pad and taking approximately 10 min per mouse to avoid bias and minimize stress behaviors. 27 Subsequently, the responses were scored according to the following criteria: no response=0, head withdrawal=0.25, single face grooming=1, and face grooming more than 3 times=1.5. 27 After 12 applications of the three behavioral tests, the pain response scores from each test were summed to achieve the total response pain score. An investigator blinded to the animal group assignment performed all the behavioral tests. Table   1 shows the number of mice at each time point.

TMJ structure
After nocifensive behaviors, animals were deeply anesthetized and transcardially perfused with 250 ml of ice-cold phosphate-buffered saline with pH 7.4. Then, they were decapitated. The skulls were immersed in After micro-CT scanning, we removed the skin on the head and opened the cervical bone. Thereafter, we dissected the skull that covers the cerebrum and removed the brain and the spinal cord. In the dissected brain and spinal cord, only the trigeminal subnucleus caudalis was removed. Next, the TMJ was dissected into two separate sides, namely, ipsilateral side and contralateral side. The TMJs were then dissected and as the mean ± standard error of the mean. The normal distribution of data was examined using the Kolmogorov-Smirnov test. We compared two independent groups in terms of the body weight, nocifensive behaviors, and nociceptive activity by using the independent t-test. In comparing TMJ degeneration between more than two independent groups, we used the one-way analysis of variance (ANOVA) and Dunnett's test (post hoc test) sequentially. Moreover, the correlations of two independent groups in terms    Figures 5A and 6A). The mean number of pCREB and microglia on the ipsilateral side in the complete Freund adjuvant group was significantly higher than that in the sham group on days 14, 21, and 28, respectively (n=6, p<0.05) (Figures 5B and 6B).
Likewise, the mean number of pCREB and microglia on the contralateral side was significantly higher in the complete Freund adjuvant group than in the sham group on day 28 (n=6, p<0.05) (Figures 5B and 6B).
The correlation between contralateral nocifensive and nociceptive activity was estimated as Pearson's correlation coefficient (r)≥0.9. The contralateral nociceptive activity on the trigeminal subnucleus caudalis induced by the injection of complete Freund   In brief, our study provides an insight on the possible development of TMJ osteoarthritis that induced mirror-image pain in a mouse model to achieve contralateral nocifensive behaviors after ipsilateral pain occurrence. These findings are useful in future pathogenesis studies of mirror-image pain models.
Evolution of mirror-image pain in temporomandibular joint osteoarthritis mouse model