The effects of acute restraint stress on nociceptive responses evoked by the injection of formalin into the temporomandibular joint of female rats

Abstract

The temporomandibular joint (TMJ) formalin test was used to evaluate the effects of acute restraint stress on the nociceptive behavioral responses of female rats during proestrus and estrus phases of the estrous cycle. Rats were subjected to one session of restraint stress (15, 30 min or 1 h). They were then either immediately killed to allow the collection of blood for hormonal radioimmunoassay determinations or subjected to TMJ formalin test to evaluate nociception. All stress protocols significantly raised the plasma concentrations of corticosterone. The performance of rats subjected to 15 and 30 min of restraint stress was similar to that of control rats, whereas rats that were stressed for 1 h showed a decrease in nociceptive responses, during both proestrus and estrus phases. The stress-induced analgesia (SIA) was greater in the proestrus phase. To evaluate the role of κ-opioid receptors, the selective receptor κ-opioid antagonist nor-binaltorphimine (nor-BNI; 200 μg or saline) was injected into the TMJ 24 h prior to the 1 h stress period and the TMJ formalin test. The local administration of nor-BNI partially reversed the SIA during the proestrus phase. These findings suggest that (1) acute stress for 1 h can produce analgesia both during proestrus and estrus phases; this effect is greater during the proestrus phase and (2) κ-opioid receptor activation is involved in the SIA observed in the proestrus phase.

Keywords::

• analgesia
• estrous cycle
• κ-opioid
• rats
• restraint stress
• temporomandibular joint

Introduction

Temporomandibular dysfunction is a painful condition of the masticatory muscles and temporomandibular joint (TMJ; Dworkin and LeResche 1992; Denucci et al. 1996), in which spreading of pain to adjacent regions of the head and neck is a frequent occurrence (Bereiter 2001). The prevalence, severity, and duration of this condition are greater in women than in men (LeResche 1997; Fischer et al. 2007). This disparity between the genders might be attributed to a pronociceptive effect of ovarian hormones on TMJ pain modulation (LeResche 1997; Cairns et al. 2001; Craft et al. 2004). Using rats it was previously demonstrated that, in response to the injection of formalin into the TMJ, a significantly lower behavioral nociceptive response was obtained in males than in females (Clemente et al. 2004).

One possible explanation for the sexual dimorphism in the TMJ nociceptive responses is related to specific endogenous opioid mechanisms. As previously described, functional κ-opioid receptors are located within the TMJ of rats. Although activation of these receptors suppresses formalin-induced TMJ nociceptive behavior in both males and females, the suppression is significantly greater in females (Clemente et al. 2004). When κ-opioid receptor agonists are used, the analgesic system in women has been found to be more sensitive than in men (Arthuri et al. 2005). It is possible that a male-related hormone, such as testosterone, interacts negatively with κ-opioid agonists; testosterone is present in both sexes, although the circulating testosterone levels in female subjects are typically about 10% of those observed in male subjects (Fischer et al. 2007). Alternatively, female-related hormones, such as progesterone or estrogen, may potentiate the action of κ-opioid (Gear et al. 1996).

Recently, it was demonstrated in rats that female sex hormones can modulate the function of the adrenal medulla; this results in a sexually dimorphic response in both the baseline mechanical nociceptive threshold and in epinephrine-induced hyperalgesia (Khasar et al. 2005). Considering that alterations in the activity of endogenous opioid systems underlie both the mechanisms that regulate the stress-induced changes in nociception (Przewlocki et al. 1987; Amit and Galina 1988; Yamada and Nabeshima 1995; Gear et al. 1996) and the hyperalgesia that is produced in rats in response to repeated (chronic) stress (Gameiro et al. 2005), we hypothesized that the differential activation of stress response systems in male and female rats could be involved in the sexually dimorphic modulation of TMJ nociception.

Chronic generalized pain, which is characterized by a diffuse lowered pain threshold (such as fibromyalgia and irritable bowel syndromes), disproportionately affects women (Buskila 2001; Yunus 2002). In many cases, stress may precede or be co-morbid with symptoms of such generalized pain syndromes (Davis et al. 2001; Raphael et al. 2004). However, despite the fact that TMJ is more prevalent in women (LeResche 1997), little is known about the effect of female sex hormones in the orofacial region and the possible involvement of stress activation on TMJ nociception. As the nociceptive behavioral responses that are elicited by the injection of formalin into the TMJ are a valid and reliable model of deep orofacial pain (Roveroni et al. 2001), this study aimed to evaluate the effects of acute restraint stress on the nociceptive behavioral responses induced by the TMJ formalin test in female rats during both the estrus and proestrus phases.

Methods

Animals

This study was carried out in 3-month-old female Wistar rats obtained from Centro Multidisciplinar para Investigação Biológica (CEMIB), UNICAMP, Piracicaba, Brazil. The rats were housed in groups of five and maintained in a temperature-controlled room (23 ± 1°C) with a 12 h/12 h light–dark cycle (lights on at 6:00 am), and food and water were available ad libitum. Rats were handled for at least 1 week prior to the experiments. This research was approved by the Committee on Animal Research of the University of Campinas (protocol 938-1) and conformed to International Association for the Study of Pain (IASP, Redwood City, CA, USA) guidelines for the study of pain in animals (Zimmermann 1983). Procedures were performed between 8:00 am and 1:00 pm.

Estrous phase determination

Proestrus and estrus phases were determined by daily microscopic examination of vaginal smears that were taken by gentle lavage between 8:00 and 10:00 am. Proestrus was identified by the predominance (>70%) of nucleated epithelial cells, and the occurrence of estrus was marked by the presence of non-nucleated cornified cells in rats with at least two consecutive regular 4-day cycles (Smith et al. 1975; Marcondes et al. 2002). These phases were chosen because they represent phases of both high and low ovarian hormone levels (Butcher et al. 1974).

Stress exposure

All rats were stressed by restraint for periods of 15, 30 min or 1 h for each exposure (Gameiro et al. 2006). Restraint was carried out by placing the animal in a transparent plastic restraint device, adjustable in size depending on the rat's weight (dimension 7 × 7 × 25 cm). The size of the tube could be adjusted individually and the tube was held firmly in place with adhesive tape on the outside so that rat was unable to move. There was a 1-cm hole in the far end to allow breathing. The control group was not submitted to restraint. The restraint procedure was carried out in a separate quiet room between 9:00 and 11:00 am.

Hormone assay

Blood of rats in proestrus and estrus was collected after conscious decapitation, which occurred either under basal conditions (within 30 s of removal from the home cage; also referred to as the 0 min time point) or at the end of each period of restraint (within 30 s of removal from the restrainer). Trunk blood was collected in tubes containing heparin, for determination of plasma corticosterone. The blood was centrifuged at 3000g, 4°C for 10 min and then the plasma was collected and frozen at − 20°C until use. Plasma corticosterone levels were determined by radioimmunoassay (anti-corticosterone—C-8784, Sigma, St Louis, MO, USA) after plasma extraction with ethanol. Assay sensitivity, intra- and interassay coefficients of variations for costicosterone were 0.4 μg dl^{− 1}, 5.1 and 8.4%, respectively. All samples from each group were assayed in duplicate in the same assay, as previously described (Castro et al. 1995).

Testing procedure for TMJ pain

The design of this study follows that used by Roveroni et al. (2001). Testing sessions took place between 10:00 am and 12:00 noon in a quiet room maintained at 23 ± 1°C. Immediately after the period of stress, each rat was placed in a glass container (10 × 10 × 18 cm) containing one cotton ball soaked with 5 ml of halothane (CRISTÁLIA) until the rat fell over (within ca. 10 s). This procedure was used for the TMJ injection (Roveroni et al. 2001). The formalin solution was prepared from a commercially available (Sigma) formalin stock (an aqueous solution of 37% formaldehyde). This solution was further diluted in 0.9% NaCl (saline) to a concentration of 1.5%. Rats received a 50-μl injection of formalin solution into the left TMJ region. The injections were performed via a 30-gauge needle introduced into the TMJ capsule. A cannula consisting of polyethylene tubing (30 cm) was connected to the needle and also to a Hamilton syringe (50 μl) that had previously been filled with formalin (1.5%). Following the TMJ injection, the rat was placed in a test chamber (30 × 30 × 30 cm mirrored-wood chamber with glass at the front side) and nociceptive behavioral responses, characterized by rubbing the orofacial region (amount of time in seconds) and flinching the head (number of head flinches), were quantified for 30 min (10 blocks of 3 min). Considering that the flinching of the head behavior followed a uniform pattern of 1 s in duration, each flinching was expressed as 1 s. The combination (sum) of both behaviors provides a better measure of pain intensity than any single behavior (Roveroni et al. 2001; Gameiro et al. 2003). To confirm the TMJ injection site at postmortem, the rats were anesthetized by an intraperitoneal injection of xylazine hydrochloride (Sigma) (10 mg/kg) followed by an intramuscular injection of ketamine hydrochloride (Sigma) (50 mg/kg) at the end of each experiment and Evans blue dye (0.1%, 5 mg/kg; Sigma) was injected systemically (intracardially) as previously described (Haas et al. 1992). The formalin-induced plasma extravasation of the Evans blue dye bound to plasma protein was then examined visually. This procedure allows confirmation of injection at the correct site by finding that the plasma extravasation induced by formalin is restricted to the immediate TMJ region. If the Evans blue dye was observed outside the TMJ capsule, the rats are excluded from the experiment.

Assessing the role of the endogenous opioid system in stress-induced anti-nociception

In the case of stressed rats that exhibited a significant reduction in nociception that indicated analgesia, the sensitivity of a specific opioid antagonist to the κ-receptor was tested. The κ-opioid antagonist nor-binaltorphimine (nor-BNI; Binder et al. 2001) (200 μg/25 μl; Sigma) was dissolved in saline. Because it has been reported that nor-BNI may not be selective for κ-opioid receptors until several hours after its administration (Schmidt et al. 2002), nor-BNI was administered into the left TMJ region one day (24 h) prior to the experiment.

Statistical analysis

Statistical analysis of the plasma corticosterone data was made using a two-way ANOVA based on ranks. The sum of rubbing and flinching responses exhibited by each rat was calculated. The data were analyzed by a two-way ANOVA on ranks followed by either Tukey or Student–Newman–Keuls post hoc tests, as appropriate. All values are given as mean ± SEM. A level of 5% was taken as evidence of statistical significance. All statistical analyses were performed using Sigma STAT version 3.0 for Windows—licensed to the University of Campinas.

Results

Effects of the stress procedures on plasma corticosterone

This experiment was carried out to define the efficacy of different acute restraint protocols (15, 30 min and 1 h) at inducing stress-like hormonal modifications. The plasma corticosterone concentrations in female rats during proestrus and estrus phases of the estrous cycle are shown in Figure 1. Compared to the controls, there was a significant increase in the plasma concentrations of corticosterone after administration of the various stress protocols—two-way ANOVA [F(3,66) = 28.68; p = 0.000]. There was no significant difference in the concentrations of corticosterone at the different phases of the estrous cycle (p>0.05).

Figure 1.  Basal and stress-induced plasma corticosterone concentrations after the 15, 30 min and 1 h stress procedures in the proestrus and estrus phases of the estrous cycle. Each column represents the mean ± SEM. Number of rats was set as n = 8/group. *p < 0.05 vs. control (Student–Newman–Keuls).

Effect of acute stress on nociceptive behavioral responses

Exposure to a single restraint session for either 15 or 30 min did not affect the nociceptive responses evoked by injection of formalin (1.5%) into the TMJ of rats (p>0.05; Figure 2). However, there was a significant difference between estrus vs. proestrus in control (unstressed) rats—two-way ANOVA [F(3,23) = 16.98; p < 0.001]. Exposure to a single restraint session for a period of 1 h did have an effect on the nociceptive responses. An analgesic effect (i.e. a significant decrease in the nociceptive behavioral responses) was observed in the 1-h stressed rats when compared to the control rats (Figure 2). The nociceptive responses of the stressed (1 h) proestrus group were significantly lower than those of the estrus group, two-way ANOVA [F(2,23) = 15.42; p < 0.001].

Figure 2.  Sum of flinching and facial rubbing behaviors recorded in rats given TMJ injection of formalin (50 μl, 1.5%) previously submitted to 15, 30 min or 1 h of restraint (n = 6/group) or left undisturbed in their home cage (n = 6/group) during the estrus and proestrus phases. Each column represents the mean. Error bars indicate the SEM. ⁺p < 0.05, estrus vs. proestrus control rats; *p < 0.05, controls vs. 1 h stressed groups; ^#p < 0.05, estrus vs. proestrus in 1 h stressed group (two-way ANOVA).

Effect of nor-BNI on nociception in rats subjected to acute restraint stress

The local pre-administration (24 h before the stress and TMJ formalin test) of nor-BNI partially reversed the stress-induced analgesia (SIA) in proestrus females—two-way ANOVA [F(2,35) = 19.77; p < 0.001]. This effect was not observed in estrus females (p>0.05; Figure 3).

Figure 3.  Nociceptive behavior in control (unstressed), stressed (1 h), saline-treated (+1 h stress), and nor-BNI-treated (+1 h stress) rats after formalin injection (50 μl, 1.5%) into the TMJ in estrus or proestrus phases (n = 6/group). Each column represents the mean ± SEM. *p < 0.05 vs. unstressed control rats (two-way ANOVA). ^#p < 0.05 estrus vs. proestrus (Tukey test). ^$p < 0.05, nor-BNI vs. saline and 1 h stressed groups (Tukey test).

Effect of nor-BNI on nociception in non-stressed rats

Estrous cycle phase affected formalin-evoked (1.5%) nociceptive responses in rats that were not exposed to restraint stress. The sum of nociceptive behaviors (flinching+rubbing) was statistically significantly less in proestrus vs. estrus females—two-way ANOVA [F(1,23) = 54.95; p < 0.001]. However, the local pre-administration of nor-BNI did not affect the nociceptive responses in either estrus or proestrus females (p>0.05; Figure 4).

Figure 4.  Nociceptive behavior in unstressed rats in estrus or proestrus given formalin injection (50 μl, 1.5%) and nor-BNI into the TMJ. Control: formalin only, saline: saline+formalin, nor-BNI: nor-BNI+formalin; n = 6/group. Each column represents the mean ± SEM. ^#p < 0.05 vs. the estrus groups (two-way ANOVA).

Discussion

Numerous animal studies have demonstrated that pain-like behavior evoked by cutaneous (Martinez-Gomez et al. 1994), deep or visceral tissue stimulation (Kayser et al. 1996; Giamberardino et al. 1997; Ness et al. 2001) varies throughout the estrous cycle. LeResche et al. (2003), using a human clinical assay, suggested that temporomandibular pain in women is highest at times of lowest estrogen. During the estrous cycle in rats, prolactin, LH, and FSH remain low but increase throughout the afternoon of the proestrus phase. Estradiol levels begin to increase at metestrus, reaching peak levels during proestrus and returning to baseline at estrus. The secretion of progesterone also increases during metestrus and diestrus, but decreases afterwards. The progesterone levels then rise to reach a second peak towards the end of proestrus (Spornitz et al. 1999; Marcondes et al. 2002). Although studies in non-humans generally support the view that gonadal hormones influence nociception in females, the factors involved in mediating this effect are not well understood (Terner et al. 2005). Moreover, there are no previous experimental studies that examine the effects of stress on the modulation of nociceptive input from articular tissue in female rats during the estrus and proestrus phases.

A variety of environmental and/or stressful stimuli have been shown to elicit analgesia—a phenomenon often referred to as SIA (Watkins et al. 1982; Furuta et al. 2003; Gameiro et al. 2005; King et al. 2007). For example, some evidence suggests that female rats are most sensitive to thermal nociceptive stimuli in proestrus (Kayser et al. 1996; Vincler et al. 2001) when estrogen and progesterone levels peak. In contrast, others suggest that peak sensitivity occurs during estrus (Martinez-Gomez et al. 1994; Kayser et al. 1996; Stoffel et al. 2003) when estrogen and progesterone levels are relatively low. These discrepancies might, in part be related to the type of nociceptive stimulus used in the experiments. For example, lower nociceptive thresholds are typically observed in response to mechanical and electrical nociceptive stimuli during estrus (Kayser et al. 1996), whereas no differences in thresholds are seen across phases in response to chemical stimuli (Vincler et al. 2001) and higher thresholds are observed with visceral stimuli during estrus and proestrus (Bradshaw et al. 1999; Terner et al. 2005). In the present study, a single exposure (1 h) to restraint stress reduced the nociceptive behavioral responses evoked by nociceptive chemical stimulation (formalin 1.5%) of the female rat's TMJ in both the estrus and proestrus phases; however, the SIA was higher in the proestrus phase. This finding is in accordance with findings in another study (Ryan and Maier 1988), in which female rats with high hormonal levels exhibited higher SIA in the tailshocks test.

The evidence which indicated that gonadal hormones influence pain sensitivity, came from studies that demonstrated a trend for higher pain thresholds and tolerance levels during the follicular phase in which estrogen levels peak (Riley et al. 1999; Terner et al. 2005). These results support the hypothesis suggested by other authors that elevated female sex hormone levels are responsible for reduced pain sensitivity (Medina et al. 1993; Liu and Gintzler 2000; Gupta et al. 2001; Tall and Crisp 2004; Arthuri et al. 2005). The higher susceptibility to pain behavior in the present study was also observed in female rats with lower hormonal levels (i.e. estrus).

In the present study, the relationship between hormonal variations and nociceptive responses induced by stress was evaluated. A significant increase in plasma corticosterone levels was observed after acute (15, 30 min, 1 h) restraint stress sessions, although only the 1-h stress period was able to alter the nociception evoked by the TMJ formalin test. It is well established that not only the type of stressor, its intensity, and duration affects the stress-induced changes to pain modulation, but also the type of nociceptive model used has an effect as well (Gameiro et al. 2006). We have already demonstrated that restraint stress can release endogenous opioids (Gameiro et al. 2005), but this effect was not capable of inducing analgesia in male rats that were submitted to acute restraint stress before the TMJ formalin test. As female rats (in both estrus and proestrus phases) exhibited decreased nociceptive responses to the TMJ formalin test and this analgesic effect was higher in the proestrus females, the present results suggest that female sex hormones can modulate SIA. We suggest that female rats are more prone to exhibit stress-induced changes that are both hormonal and behavioral. The finding of higher levels of corticosterone levels after the stress procedures, when compared with those that we previously found in males, also supports this hypothesis.

Stress and opioid agents act throughout the neuraxis to modulate sensory, motor, autonomic, motivational and emotional responses to nociceptive stimulation (Houshyar et al. 2001; Hebb et al. 2005; King et al. 2007). Studies examining the involvement of exogenous opioids in the peripheral modulation of TMJ pain (Bakke et al. 1998; Cai et al. 2001), support the presence of peripheral opioid receptors in the TMJ that may have a role in modulating nociceptive responses (Clemente et al. 2004; Arthuri et al. 2005). In normal cycling females, morphine and buprenorphine (opioid agonists) were generally most potent in metestrus and proestrus and least potent in estrus (Terner et al. 2005). In the last experiment, we tested the nociceptive responses in controls and in rats exposed to a single 1 h restraint stress that were previously injected with nor-BNI (200 μg/25 μl) (κ-opioid antagonist) in the TMJ formalin test. Our results demonstrated that the local pre-administration of nor-BNI reduced the SIA in proestrus females. The effect of nor-BNI in the estrus females was not significant; this indicated that higher female sex hormonal levels may increase SIA via opioid mechanisms. Although the induction of hyperalgesia after several hours of nor-BNI administration has been reported (Schmidt et al. 2002), this finding was not observed in our study as the application of nor-BNI to unstressed rats did not evoke a hyperalgesic effect. This result is also supported by the findings of Arthuri et al. (2005).

The present data indicate that κ-receptors are partly involved in mediating SIA in female rats. The administration of nor-BNI before formalin to unstressed rats did not alter the nociceptive responses to the TMJ formalin test; this suggested that the κ-opioid antagonist selectivity of nor-BNI observed in the present results was related to the gonadal hormones.

In conclusion, acute restraint stress (1 h) produced analgesia in proestrus and estrus female rats, but this effect was greater in the proestrus phase. Moreover, this study demonstrated that κ-opioid receptor activation is involved in the SIA that is observed during the proestrus phase. The data presented in this study could be of clinical value and might be important for understanding the neurobiological mechanisms concerning temporomandibular disorders and the relationship among sex hormones, nociception, and stress.

Acknowledgements

The authors thank Carlos Alberto Feliciano for technical assistance and Professor Margaret de Castro (FMRP-USP) for radioimmunoassay assistance. This work was supported by CNPq, Brazil.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.