- Email: [email protected]
Formalin Injection Into Knee Joints of Rats: Pharmacologic Characterization of a Deep Somatic Nociceptive Model
The Journal of Pain, Vol 7, No 2 (February), 2006: pp 100-107 Available online at www.sciencedirect.com
Formalin Injection Into Knee Joints of Rats: Pharmacologic Characterization of a Deep Somatic Nociceptive Model Maria Alcina Martins, Lúcia de Castro Bastos, and Carlos Rogério Tonussi Department of Pharmacology, Federal University of Santa Catarina, Campus Universitário–Trindade, Florianópolis, Santa Catarina.
Abstract: Formalin (0.25, 0.5, 3, and 5%) injected into the knee joint of rats induced a dosedependent nociception that was featured by 2 phases of intense guarding behavior of the affected limb, interposed by a period of quasinormal gait (quiescent phase). The guarding behavior during a period of forced gait was measured by the total time the paw of the affected limb was not in contact with the surface of a revolving cylinder (paw elevation time [PET]). Pretreatment with morphine (4 mg/kg, subcutaneously) reduced PET in both nocifensive phases, and naloxone (1 mg/kg, subcutaneously) antagonized morphine’s effect. The cyclooxygenase inhibitor diclofenac (5 mg/kg, intraperitoneal) reduced only the second phase of nocifensive responses. A low dose of the benzodiazepine midazolam (0.25 mg/kg, intraperitoneal) significantly reduced only the second phase of response, but a higher dose (1 mg/kg, intraperitoneal) had no effect. A subconvulsant, anxiogenic dose of pentylenetetrazol (30 mg/kg, intraperitoneal) also did not affect the PET increase induced by formalin. The antihistamine meclizine (2.5 mg/kg, intraperitoneal) caused an increase of the response in the second phase, but a higher dose (7.5 mg/kg, intraperitoneal) caused inhibition. The peripheral antihistamine loratadine (5 and 10 mg/kg, intraperitoneal) also caused an increase of the second phase. Neither antihistamine altered the first phase of PET. These results reproduced previous findings with classical analgesics in formalin-induced nociception. However, the pronociceptive effect of antihistamines, and the antinociceptive effect of midazolam observed here suggest that formalin-induced incapacitation introduces new characterists of nociceptive system that may complement the study of analgesics. Perspectives: Anxiety is thought to influence pain experience in an opposing manner depending on nociception originates in cutaneous or deep somatic/visceral tissues. The present formalin-induced nociceptive test may help to predict more reliably the pain-killing effect of new pharmacologic strategies, with classical or nonclassical mechanisms, for the treatment of clinically relevant pains, which are generally originated in deep structures. © 2006 by the American Pain Society Key words: Articular pain, arthritis, benzodiazepines, anxiety, opioids, NSAID, antihistamines.
F
ormalin has long been used as a nociceptive stimulus in different animal models. The most used approach has been the subcutaneous injection into the rat or mouse footpad, which elicits a well-known behavioral response characterized by intense flinches of the injected paw, augmented attention to the paw generally
Received February 15, 2005; Revised June 23, 2005; Rerevised September 5, 2005; Accepted September 8, 2005. Address reprint requests to Carlos Rogério Tonussi, Department of Pharmacology, Federal University of Santa Catarina, Campus Universitário – Trindade, Florianópolis, SC, 88040-900. E-mail: [email protected] 1526-5900/$32.00 © 2006 by the American Pain Society doi:10.1016/j.jpain.2005.09.002
100
featuring the licking of the paw, and also by a less dramatic response characterized by a repeated lifting of the injected paw from the floor. The best known characterization of the behavioral nocifensive responses elicited by the injection of formalin into the paw comes from Dubuisson and Dennis’s work.11 Several subsequent modifications were applied to the original idea in order to improve the quantification of the inferred nociception or to propose different regional models of persistent nociception by injecting formalin in orofacial, temporomandibular, tail, muscle, and intracolonic tissues.2,6,7,17,22,25,29,45,48 The development of models based on the injection of formalin in tissues other than skin is very useful for the better understanding of noci-
ORIGINAL REPORT/Martins et al ceptive transmission and modulation, because there is a consensus that nociceptive inputs from deep tissues are processed by different sensory, affective, and motivational substrates.23,31,50 Notwithstanding, in all of the above approaches, the quantification of formalin-induced responses remained dependent on the subjective human observation, which clearly introduces a significant bias into the results from different observers and/or research groups. Attempts to implement unbiased, automated recording systems for the registering of the formalin-induced nociception have been made in dogs and rats.19,44,56 The most recent implementations, however, would be sensitive to the action of sedative, analeptic drugs, because these are passive models and do not require the execution of a task by the animals. In this work, we present a model to study formalin-induced nociception in rats based on an apparatus previously described for the measurement of articular inflammatory incapacitation, together with its pharmacologic characterization.
Material and Methods Animals The experiments were performed on 200- to 300-g male Wistar rats housed in temperature-controlled room (22 ⫾ 2°C) under a 12-12 hour light/dark cycle with free access to water and food. All behavioral testing were performed between 7:00 AM and 2:00 PM. Animal care and handling procedures were in accordance with the ethical guidelines of the International Association for the Study of Pain18 and also approved by the Ethics Committee for Animal Research of Federal University of Santa Catarina. The animals were euthanized in a CO2 chamber right after the experimental sessions.
Drugs and Vehicles Morphine and midazolam (Laboratórios Cristália, Itapira, SP, Brazil), diclofenac (Novartis, São Paulo, SP, Brazil), meclizine, naloxone, and pentylenetetrazol (Sigma, St. Louis, Mo), and loratadine (Aventis Pharma, São Paulo, SP, Brazil) were used. Loratadine was dissolved in polyoxyethylenesorbitan monoleate (Tween 80), not exceeding 1% of the total volume in physiologic saline, and meclizine was dissolved in sunflower oil. The other drugs were dissolved in physiologic saline. All treatments were administered by intraperitoneal route in a volume of 0.1 mL/100 g of body weight, except morphine and naloxone, which were injected subcutaneously. Morphine, diclofenac, loratadine, and meclizine were injected 60 min before, while midazolam and pentylenetetrazol were injected 30 min before formalin. Naloxone was given 70 min before formalin, ie, 10 min before morphine. Formalin (formaldehyde 37% - Merck AG, Germany) was diluted in saline to 0.25, 0.5, 3, and 5%, considering the initial concentration as 100%. Control group treatments were carried out with the vehicle of the respective test drug.
101
Knee Joint Incapacitation Test Induced by Formalin The rat knee joint incapacitation test is described in detail elsewhere.53 In this test, rats are placed on a revolving cylinder (30 cm diameter; 3 rpm) for 1-min periods and a computer-assisted device measures the total time that a specific hind paw was not in contact with the cylinder surface (ie, paw elevation time [PET]). Normally, control animals display a PET of approximately 10 s, whereas algogenic substances injected into the knee joint increase this value only in the affected limb. In order to induce incapacitation, 50 L of formalin solution (5%), diluted in sterile saline, was injected into the right knee joints of rats. The injection site was first shaved and treated with an iodine alcohol antiseptic solution. The animals were gently restrained in a supine position by the hands of the experimenter, and intraarticular injections were quickly performed with 27-gauge needles. The PET was measured each 5 min after formalin injection, as described above. As the nocifensive behavior was scored automatically, the experimenters were not blind to the treatment protocols.
Motor Impairment Assessment When the cylinder started revolving, the animals promptly and spontaneously walked to keep themselves on top, without assistance from the experimenter. Each experiment comprised 13 periods of gait registering for each animal. Any repeated delay to react to the cylinder movement or an animal fall was considered motor impairment. A treatment was identified as causing no motor impairment if the above-mentioned events were not observed.
Statistical Analysis All statistical analyses were carried out using Graphpad Prism version 3 (http://www.graphpad.com). Results are expressed as mean ⫾ SE mean in the figures. Comparisons of experimental curves were made with paired, two-tailed Student t test or one-way analysis of variance (ANOVA) for repeated measures and, when a significance level of at least P ⬍ .05 was detected, analysis was followed by the Tukey multiple comparison test. Quiescent phase analysis was performed with simple one-way ANOVA.
Results Dose-Response Relationship of FormalinInduced Nocifensive Responses Injection of 0.5, 3, and 5% formalin solution into 1 knee joint of rats induced a dose-related nocifensive response characterized by the expression of guarding behavior that was registered as the increase of the PET compared with pre-formalin values (Fig 1). At a concentration of 0.25% the PET registered was not different from the physiologic saline control. There was also no difference between the mean PET of 3% and 5% groups, although the variability of response was greater in the
102
Formalin Injection Into Knee Joints of Rats
Figure 2. Antinociceptive effect of morphine. Morphine (4 mg/ Figure 1. Dose response relationship of the formalin-induced incapacitation. Formalin (0.25, 0.5, 3, and 5 %) was injected into the knee joint in a volume of 50 L, and the paw elevation time (PET) was measured immediately after. C represents PET before formalin injection (0 min). Control group received only physiologic saline into the knee joint, n indicates the number of animals in each group. * indicates statistical significance (P ⬍ .05) when compared to interphase control value (one-way ANOVA followed by Tukey test).
3% group. The time course of the response presented a first, short-lasting phase (0-5 min interval) immediately after the injection followed by a second, long-lasting phase (10-60 min interval). The PET registered immediately after formalin injection were 45.4 ⫾ 2.5 s for 5%, 48.9 ⫾ 3.2 s for 3%, 34.8 ⫾ 4.4 s for 0.5%, and 16.3 ⫾ 1.8 s for 0.25%. The second phase also presented a characteristic peaking response that was usually reached 25 min after formalin injection. The PET registered at the peak of the second phase were 46.5 ⫾ 4.0 s for 5%, and 49.3 ⫾ 7.2 s for 3%. Both phases were interposed by a period of apparently lower nociceptive inputs from the joint (quiescent phase, 5-10 min interval). The group injected with 0.5% formalin peaked earlier at 10 min after injection (32.9 ⫾ 4.3 s), and presented a quiescent phase (24.0 ⫾ 2.3 s) significantly greater than the control (13.9 ⫾ 0.8 s) (P ⬍ .05). The guarding behavior expressed by the animals after injection of formalin was obvious even when the animals were not walking on the cylinder, but after 60 min there was no signs of nociception. Also after this period, we observed that the higher formalin concentration produced an intense synovial sweeling. For the subsequent experimental purposes the 5% formalin solution was used as a standard nociceptive stimulus.
Antinociceptive Effect of Morphine and Diclofenac in Knee Joint Incapacitation Test Induced by Formalin Morphine (4.0 mg/kg, sc, 60 min before) inhibited the nocifensive response to 5% formalin in both nocifensive phases (P ⬍ .01; Fig 2). The PET registered immediately after formalin injection were 42.6 ⫾ 1.2 s for the salinetreated group and 23.5 ⫾ 2.5 s for the morphine-treated
kg, sc, 60 min before formalin) inhibited PET in both phases (P ⬍ .01). Naloxone (1.0 mg/kg, sc, 10 min before morphine) reversed morphine’s effect in both phases (P ⬍ .01). C represents PET before formalin injection (0 min). Control group received only physiologic saline by subcutaneous route, n indicates the number of animals in each group. Statistical differences were detected by a one-way ANOVA for repeated measures followed by Tukey test.
group. The PET registered at the peak of the second phase were 40.4 ⫾ 5.5 s for saline- and 22.7 ⫾ 3.7 s for morphine-treated groups. This inhibitory effect produced by morphine was totally reversed by naloxone (1.0 mg/kg, subcutaneously, 10 min before morphine) (P ⬍ 0.01; Fig 2). Interestingly, naloxone caused the anticipation of the peak response in the second phase—ie, a deviation to the left of the time-course curve. At the dose above, morphine did not impair the normal locomotion of the animals. For comparison, a dose of 50 g/kg, sc, of
Figure 3. Antinociceptive effect of diclofenac. Diclofenac (5.0 mg/kg, intraperitoneal, 60 min before formalin) significantly reduced the PET in the second phase of the formalin-induced response (P ⬍ .001) compared to the control group. Diclofenac (0.5 mg/kg, intraperitoneal) did not have a significant effect on the PET. C represents PET before formalin injection (0 min). Control group received only physiologic saline by intraperitoneal route. n indicates the number of animals in each group. Statistical differences were detected by a one-way ANOVA for repeated measures followed by Tukey test.
ORIGINAL REPORT/Martins et al
103 54
atogenic effects, respectively. These treatments did not affect the first peak of response (PETsal ⫽ 42.6 ⫾ 1.2 s, PET0.5mg ⫽ 39.4 ⫾ 3.8 s and PET5mg ⫽ 37.1 ⫾ 3.8 s) nor the quiescent phase (PETsal ⫽ 15.5 ⫾ 1.3 s, PET0.5mg ⫽ 13.8 ⫾ 0.8 and PET5mg ⫽ 10.7 ⫾ 1.8 s).
Dual Effect of H1 Histamine Receptor Antagonists
Figure 4. Dual effect of H1 histamine receptor antagonists. (A) Meclizine (2.5 mg/kg, intraperitoneal, 60 min before formalin) significantly increased PET (P ⬍ .05), but the higher dose (7.5 mg/kg, intraperitoneal) significantly decreased PET (P ⬍ .001) in the second phase of the formalin-induced response compared to the control group (one-way ANOVA for repeated measures followed by Tukey test) (B) Loratadine (5 and 10 mg/kg, intraperitoneal, 30 min before formalin) induced a significant increase in the PET in the second phase of the formalin-induced response (P ⬍ .001), compared to the respective control group (paired Student t test). C represents PET before formalin injection (0 min). The control group for meclizine received only sunflower oil, whereas that for loratadine received only physiological saline with Tween 80 (1%) by intraperitoneal route. n indicates the number of animals in each group.
fentanyl impaired the animals’ spontaneous positioning on the top of the cylinder, causing frequent falls (data not shown). The cyclooxygenase inhibitor diclofenac (5 mg/kg, intraperitoneal, 60 min before) also induced a significant antinociceptive effect, but only in the second phase (P ⬍ .001; Fig 3). At a smaller dose (0.5 mg/kg), diclofenac did not affect the formalin-induced nocifensive response (Fig 3). The lower and higher doses were chosen for comparison with our previous data in carrageenan and LPS models of inflammatory knee joint incapacitation, when they presented maximal antinociceptive and antiedem-
Formalin-induced nociceptive test has been demonstrated to be sensitive to antihistamines. Pretreating the animals with the blood brain barrier permeable, H1 receptor antagonist, meclizine (1, 2.5, and 7.5 mg / kg, intraperitoneal, 60 min before; Fig 4A) caused a dual effect on the nocifensive behavior exhibited by the rats after formalin injection into knee joints. The dose of 2.5 mg/kg caused the increase of the paw elevation time in the second phase period (P ⬍ .05), but the higher dose (7.5 mg/kg) decreased this response (P ⬍ .001), when compared to the control time-course curve. However, it was not observed any effects of the drugs in the first phase (PETveh ⫽ 49.9 ⫾ 2.1 s, PET1mg ⫽ 39.3 ⫾ 5.9 s, PET2.5mg ⫽ 38.6 ⫾ 7.2 s and PET7.5mg ⫽ 40.8 ⫾ 4.5 s) (P ⫽ .42, one-way ANOVA). These treatments did not affect the quiescent phase either (PETveh ⫽ 20.9 ⫾ 3.9 s, PET1mg ⫽ 18.4 ⫾ 5.0 s, PET2.5mg ⫽ 19.6 ⫾ 3.5 s and PET7.5mg ⫽ 16.6 ⫾ 3.6 s), nor was there motor impairment with the higher dose. Pretreatment with the peripheral H1 receptor antagonist, loratadine (5.0 and 10.0 mg/kg, intraperitoneal, 60 min before; Fig 4B) increased the second phase of nocifensive responses induced by formalin (P ⬍ .001). The lower dose tested (2.5 mg/kg) did not produce a statistically significant effect. These treatments also did not affect the first peak of response (PETveh ⫽ 49.5 ⫾ 4.1 s, PETlor ⫽ 43.1 ⫾ 3.4 s) nor the quiescent phase (PETveh ⫽ 16.7 ⫾ 1.5 s, PETlor ⫽ 19.0 ⫾ 3.4 s).
Effect of Midazolam and Pentylenetetrazol Benzodiazepine anxiolitic drugs or anxiogenic conditions have been demonstrated to cause pronociceptive and antinociceptive effects in cutaneous models of nociception, respectively. Pretreatment with the GABAA receptor benzodiazepine site ligand midazolam (0.25 mg/ kg, intraperitoneal, 30 min before; Fig 5A) caused an intense inhibition of the second phase nocifensive response (PETsa l⫽ 53.0 ⫾ 3.8 s, PETmdz ⫽ 30.0 ⫾ 6.8 s; P ⬍ .001) but only a tendency (P ⫽ .06) to inhibit the first phase (PETsal ⫽ 52.3 ⫾ 3.4 s, PETmdz ⫽ 40.6 ⫾ 4.7 s), and did not affect the quiescent phase (PETsal ⫽ 20.7 ⫾ 2.3 s, PETmdz ⫽ 18.4 ⫾ 2.3 s). A 4-fold higher dose (1.0 mg/kg, intraperitoneal, 30 min before; Fig 5A ), however, showed no effect in first phase (PETsal ⫽ 45.0 ⫾ 4.8 s, PETmdz ⫽ 45.2 ⫾ 2.1 s) and second phase (PETsal ⫽ 39.6 ⫾ 2.3 s, PETmdz ⫽ 37.5 ⫾ 6.3 s), but did exhibit a tendency to inhibit the quiescent phase (PETsal ⫽ 18.4 ⫾ 3.6 s, PETmdz ⫽ 26.8 ⫾ 1.6 s). Pretreatment with the GABAA chloride channel blocker pentylenetetrazol in an anxiogenic, subconvulsant dose (30 mg/kg, intraperitoneal, 30 min be-
104
Formalin Injection Into Knee Joints of Rats sive response interposed by a period of quiescence. This is important for comparative purposes because it suggests that the neural apparatus at the articulation is similar to that found in subcutaneous tissues, in terms of its capacity to respond to formalin. Furthermore, formalin produced a dose-related nocifensive response, although the dose producing intermediate effects (0.5%) also produced a less marked quiescent phase. The quiescent phase is recognized as being due to an active inhibition of the nociceptive transmission, integrating supraspinal as well as spinal circuitries.21,27 If this inhibition is also modulating articular nociception induced by formalin, our observation suggests that the activation of this inhibitory system is proportional to the intensity of the nociception in the early phase. In addition, this may be specific to, or at least more evident in, this model because, in experiments using subcutaneous injection of formalin, the quiescent phase was the same regardless of the formalin concentration used.21,52 By extending this point, the earlier peak of incapacitation in the second phase of the response elicited by 0.5% formalin may be likely due to a lesser inhibitory tone in the quiescent phase. Whatever the case, we think that the quiescent phase in any formalin-based test should not be neglected because it can account for important and specific changes in the nociceptive system.
The Sensitivity to Classical Analgesics Figure 5. Effect of midazolam and pentylenetetrazol. (A) Midazolam (0.25 mg/kg, intraperitoneal, 30 min before formalin) significantly reduced paw elevation time in the second phase of the formalin-induced response (P ⬍ .001) compared to the control group (paired Student t test). (B) Midazolam (1.0 mg/kg, intraperitoneal) and pentylenetetrazol (30 mg/kg, intraperitoneal) did not have a significant effect on the PET (paired Student t test), but a tendency to increase PET in the interphase. C represents PET before formalin injection (0 min).
fore; Fig 5B) did not induce changes in either phase (PET0min ⫽ 42.4 ⫾ 3.4 s, PET25min ⫽ 40.0 ⫾ 4.4 s), but also showed a tendency to inhibit the quiescent phase (PETq ⫽ 28.7 ⫾ 4.3 s). Dose and times of pretreatment with midazolam and pentylenetetrazol were according previous reports from studies in anxiety models.5,20 These treatments produced neither motor incoordination, in the case of midazolam, nor hyperexcitability and convulsion, in the case of pentylenetetrazol.
Discussion It was shown in the present study that formalin injection into rat knee joints induces a guarding behavior of the limb, which can be registered as the paw elevation time during the forced walk on a revolving cylinder, using the same apparatus we previously used for inflammatory articular nociception quantification.53 As is generally observed in models using formalin as a nociceptive stimulus, the animals exhibited two periods of nocifen-
The analgesic efficacy of morphine and its antagonism by naloxone emphasizes the sensitivity of the model to the analgesic action of opioids. As a -opioid receptor acting analgesic, morphine inhibited both phases of the nocifensive response. In addition, besides the sensitivity to opiate analgesics, the enforced walk on the cylinder allows the discrimination of pure analgesic effect from other effects on locomotor activity (eg, sedation, incoordination, catalepsy). This feature could be observed in the treatment with fentanyl, which at a dose generally used as analgesic, caused the recurrent fall of the animals from the cylinder. Because normal animals try actively to walk on the top of the cylinder, any difficulty in maintaining this position is considered to represent impairment of motor control. In contrast, the cyclooxygenase inhibitor diclofenac inhibited only the second phase of the formalin-induced response. The intensity of the inhibitory effects of diclofenac in the second phase of responses suggests that also in the articular tissues the production of prostaglandins is an important component of the nociception at this stage52 and that this model can be used as a sensitive test for detection of aspirin-like antiinflammatories. Two doses of diclofenac was used here. The lower dose was previously observed to produce maximal antinociceptive effect in inflammatory models of carrageenan- and LPSinduced arthritis,30,54 whereas the higher dose was observed to produce maximal antiedematogenic effect in a model of LPS-induced arthritis (unpublished data). The high effectiveness of the lower dose of diclofenac was previously seen to be mediated by a nitric oxide/cGMPdependent mechanism,54 whereas the higher dose
ORIGINAL REPORT/Martins et al seems to be mediated by the inhibition of prostaglandin synthesis, as suggested by the antiedematogenic effect. What this result means is that, despite the possible activation of the nitric oxide pathway, it was not enough to produce an antinociceptive effect in the present model. The possibility that by using enhancers of the nitric oxide pathway, such as an inhibitor of the cGMP phosphodiesterase combined with diclofenac,3,39 we might also observe an antinociceptive effect with low doses of diclofenac remains to be evaluated. However, the effect of opioid and nonsteroidal antiinflammatory drugs in this model is in line with classical formalin tests52 and reinforces the validity of the method as a confident tool for nociception studies.
The Antinociceptive Effect of Midazolam The effect of benzodiazepines given systemically in nociceptive tests has been inconsistent. These drugs were found to produce moderate antinociception, no effect , or even a pronociceptive effect, with this last result being corroborated by the antinociceptive effect induced by an inverse benzodiazepine agonist.13,36,37,49,51,53,57 Furthermore, several authors reported an antagonistic effect of the morphine-induced analgesia as well as nonnarcotic analgesia by benzodiazepines, while others have reported a potentiation of a nonnarcotic analgesic with alprazolam.9,10,35,36 It has been recognized that nociception elicited in cutaneous and deep tissues activates different regions in the periaqueductal gray area and amygdala, structures that participate in anxiety/fear behavior and are involved in the activation of descending modulatory projections to the spinal cord.24,31 Anxiety/ fear may alter animal and human nociception in both directions, depending on their origin. For example, cutaneous elicited nociception was attenuated by anxiety/ fear producing paradigms, whereas nociception of deep origin may be enhanced or unaffected.8,12,15,16,33,40 Benzodiazepines can reverse stress-induced analgesia, probably by attenuating the anxiogenic/fear drive for the activation of endogenous antinociceptive circuitries, which may explain the hyperalgesic effect of the drug in cutaneous pain models.12,16 Conversely, the anxiety facilitation of persistent nociception elicited in deep tissues could be attenuated by benzodiazepines resulting in antinociception.31 In line with this idea, midazolam was seen to produce antinociception in the writhing test in mice, a model of visceral nociception.34 This may be one of the explanations for this remarkable difference between our results with midazolam and the previous findings in the literature, but it does not explain the lack of antinociceptive effect with the higher dose of midazolam. GABAA agonist injection in the rostral ventromedial medulla or intracerebroventricular injection of benzodiazepines strongly increases nociception.14,51 On the contrary, spinally injected benzodiazepines were seen to produce antinociception in several models.32,57 In view of these data, it is not difficult to argue that much of the contradictory data about the effects of systemically given benzodiazepines on nociceptive models may arise
105 from a differential action of the drug on supraspinal or spinal circuitries. We could, therefore, propose that the low dose of midazolam would be facilitating a spinal GABAergic mechanism of antinociception, while the higher dose of midazolam would be inhibiting a supraspinal inhibitory system, as proposed by Gilbert and Franklin.14 This inhibition of a descending inhibitory system is consistent with the enhanced nociception observed in the quiescent phase when the animals were treated with the higher dose of midazolam and also may explain why we did not observe an antinociceptive effect with diazepam in a previous study using carrageenaninduced articular nociception.53 Pentylenetetrazol, an inhibitor of the Cl- channel associated with the GABAA receptor, acted similarly to the higher dose of midazolam and also showed a tendency to increase the nociception in the quiescent phase. This result contradicts previous reports showing that subconvulsant, anxiogenic doses of pentylenetetrazol induce antinociception in thermally and mechanically elicited cutaneous nociception.8,43 However, as discussed above, the finding does support the notion that anxiety/fear enhances the sensitivity to persistent nociception from deep tissues.
The Effect of Antihistamines Another peculiarity of the present model may be our results with meclizine, which produced a significant increase and decrease of the nocifensive responses. Systemically given H1 antihistamine drugs were demonstrated to be antinociceptive in several models and also clinically.46,47 The antinociceptive action of antihistamines has been attributed to a central action on histaminergic pathways, although several of them may also possess a local anesthetic-like mechanism of action.41,42 Our result with the higher dose of meclizine, although small, was consistent with the antinociceptive effect of antihistamines previously reported by others. However, the hyperalgesic effect observed with the lower dose was unexpected. A possible explanation for this result may be derived from the observation with loratadine treatment, which caused hyperalgesic effects. Loratadine is only weakly permeant to the blood brain barrier, which precludes a central action, and also does not appear to possess a local anesthetic-like effect.1,4,28 On this basis, we can only suppose a local action by the blockade of H1 receptors. Local application of antihistamines in the formalin test was recently readdressed by Parada et al,38 confirming an antinociceptive effect. However, in that study the authors only used antihistamines with recognized local anesthetic-like effect, which did not answer the question of how locally given antihistamines without local anesthetic effect could affect the formalininduced response. Our observation of an enhancement of the nocifensive responses after treating the animals with loratadine may be due to an inhibitory effect over vasodilation. Neurogenic vasodilation and plasma exudation in rat knee joints is substantially mediated by histamine and serotonin released from mast-cells.26,55 Formalin induces an intense swelling into synovia. Antagonizing the vasodilation and plasma leakage by H1
106 antagonism would delay the clearance of formalin from the synovial space and/or delay its dilution due to joint sweeling, which would increase its nociceptive action. In conclusion, it must be born in mind that relevant clinical pains are of deep somatic/visceral origin not only because of their pathologic features, but also because they elicit more intense emotional and autonomic responses.24,50 In this context, cutaneous nociceptive models may be of minor significance in predicting the effectiveness of new analgesic or adjuvant therapies. On the other hand, models of visceral or deep somatic pain are generally less practical, or do not give clear-cut results about the efficiency of a drug as an analgesic. Formalin-
References 1. Algemissen B, Hermes B, Henz BM U, Berlien HP: Laserinduced weal and flare reactions: clinical aspects and pharmacological modulation. Br J Dermatol 146:863-868, 2002 2. Aloisi AM, Decchi B, Carli G: Importance of site of injection in the formalin test. Med Sci Res 23:603-610, 1995 3. Asomoza-Espinosa R, Alonso-Lopez R, Mixcoatil-Zecuatil T, Aguirre-Banuelos P, Torres-Lopez JE, Granados-Soto V: Sildenafil increase diclofenac antinociception in the formalin test. Eur J Pharmacol 418:195-200, 2001 4. Barnett A, Lorio LC, Kreutner W, Tozzi S, Ahn HS, Gulbenkian A: Evaluation of the CNS properties of SCH 29851, a potencial non-sedating antihistamine. Agents Actions 14: 590-597, 1984 5. Bertoglio LJ, Anzini C, Lino-de-Oliveira C, Carobrez AP: Enhanced dorsolateral periaqueductal gray activity counteracts the anxiolytic response to midazolam on the elevated plus-maze Trial 2 in rats. Behav Brain Res 162:99-107, 2005 6. Clavelou P, Dallel R, Orliaguet T, Woda A, Raboisson P: The orofacial formalin test in rats: effects of different formalin concentration. Pain 62:295-301, 1995 7. Coderre TJ, Fundytus ME, McKenna JE, Dalal S, Melzack R: The formalin test: a validation of the weighted-scores method of behavioral pain rating. Pain 54:43-50, 1993 8. Conceição IM, Maiolini Junior M, Mattia N, Vital MA, Santos BR, Smaili S, Frussa-Filho R: Anxiety-induced antinociception in the mouse. Braz J Med Biol Res 25:831-834, 1992 9. Daghero AM, Bradley EL Jr, Kissin I: Midazolam antagonizes the analgesic effect of morphine in rats. Anesth Analg 66:944-947, 1987 10. Dolezal T, Krsiak M: Augmentation of analgesic effect of ibuprofen by alprazolam in experimental model of pain. Physiol Res 52:179-184, 2002 11. Dubuisson D, Dennis SC: The formalin test: a quantitative study of analgesic effects of morphine, meperidine, and brain stem stimulation in rats and cats. Pain 4:161-174, 1977
Formalin Injection Into Knee Joints of Rats induced incapacitation may be a useful tool for deep somatic nociceptive studies. The model features an automated registering system with reliable sensitivity and specificity to classical analgesics. With nonclassical analgesics, such as benzodiazepines and sedative antihistamines, the model may help to predict to what extent analgesia may be expected.
Acknowledgments The authors wish to thank the Coordenação para o Aperfeiçoamento de Pessoal do Ensino Superior (CAPES) for support in the form of M.Sc. studentships for MAM and LCB.
14. Gilbert AK, Franklin KB: GABAergic modulation of descending inhbitory system from the rostral ventromedial medulla (RVM). Dose-response analysis of nociception and neurological deficits. Pain 90:25-36, 2001 15. Gunter WD, Shepard JD, Foreman RD, Meyers DA, Greenwood-Van Meerveld B: Evidence for visceral hypersensitivity in high anxiety rats. Physiol Behav 69:379-382, 2000 16. Helmstetter FJ: Stress-induced hypoalgesia and defensive freezing are attenuated by application of diazepam to the amygdala. Pharmacol Biochem Behav 44:433-438, 1993 17. Hunskaar S, Berge OG, Hole K: Dissociation between antinociception and anti-inflammatory effects of acetylsalicylic acid and indomethacin in the formalin test. Pain 25: 125-132, 1986 18. IASP (International Association for Study of Pain): Ethical guidelines for investigation of experimental pain in conscius animals. Pain 16:09-110, 1983 19. Jett MF, Michelson S: The formalin test in rat: validation of an automated system. Pain 64:19-25, 1996 20. Jones N, Duxon MS, King SM: Ethopharmacological analysis of the unstable elevated exposed plus maze, a novel model of extreme anxiety: predictive validity and sensitivity to anxiogenic agents. Psychopharmacology 161:314-323, 2002 21. Kaneko M, Hammond DL: Role of spinal gamma-aminobutyric acid A receptors in formalin-induced nociception in the rat. J Pharmacol Exp Ther 282:928-938, 1997 22. Keay KA, Bandler R: Deep and superficial noxius stimulation increased Fos-like immunoreactivity in different regions of the midbrain periaqueductal grey of the rat. Neurosci Lett 14 154:23-26, 1993 23. Keay KA, Bandler R: Distinct central representations of inescapable and escapable pain: observations and speculation. Exp Physiol 87:275-279, 2002 24. Keay KA, Li QF, Blander R: Muscle pain activates a direct projection from ventrolateral medulla in rats. Neurosci Lett 1 290:157-160, 2000
12. Fanselow MS, Helmstetter FJ: Conditional analgesia, defensive freezing, and benzodiazepines. Behav Neurosci 102: 233-243, 1988
25. Kolesnikov Y, Cristea M, Oksman G, Torosjan A, Wilson R: Evaluation of the tail formalin test in mice as a new model to assess local analgesic effects. Brain Res 17 1029:217-223, 2004
13. Gatch MB: Effects of benzodiazepines on acute and chronic athanol-induced nociception in rats. Alcohol Clin Exp Res 23:1736-1743, 1999
26. Lam FY, Ferrell WR: Mediators of substance P-induced inflammation in the rat knee joint. Agents Actions 31:298307, 1990
ORIGINAL REPORT/Martins et al 27. Matthies BK, Franklin KB: Formalin pain is expressed in decebrated rats but not attenuated by morphine. Pain 51: 199-206, 1992 28. Mattila MJ, Paakkari I: Variations among non-sedating antihistamines: are there real differences? Eur J Clin Pharmacol 55:85-93, 1999 29. Miampamba M, Chery-Croze S, Gorry F, Berger F, Chayvialle JA: Inflammation of the colonic wall induced by formalin as a model of acute visceral pain. Pain 57:327-334, 1994 30. Motta AF, Borges Junior NG, da Fonseca JC, Tonussi CR: The antinociceptive effect of iontophoretic direct application of diclofenac to arthritic knee joints of rats. Life Sci 73:1995-2004, 2003
107 anesthestics on excess tachycardia in conscius dogs. J Pharmacol Exp 238:367-371, 1986 43. Rocha L, Cano A, Cruz Z, Omana-Zapata I, Villalobos R, Maidment NT: Opioid peptide systems following a subconvulsant dose of pentylenetetrazol in rats. Epilepsy Res 37: 141-150, 1999 44. Rodriguez R, Pardo EG: Reversal by acetylsalicylic acid of pain induced functional impairment. Life Sci 5:775-781, 1966 45. Roveroni RC, Parada CA, Cecilia M, Veiga FA, Tambeli CH: Development of a behavioral model of TMJ pain in rats: the TMJ formalin test. Pain 94:185-191, 2001 46. Rumore MM, Schlichting DA: Analgesic affects of antihistamines. Life Sci 36:403-416, 1985.
31. Neugebauer V, Li W, Bird GC, Han JS: The amygdala and persistent pain. Neuroscientist 10:221-234, 2004
47. Rumore MM, Schlichting DA: Clinical efficacy of antihistamines as analgesics. Pain 25:7-22, 1986
32. Niv D, Davidovich S, Geller E, Urca G: Analgesic and hyperalgesic effects of midazolam: dependence on route of administration. Anesth Analg 67:1169-1173,1988
48. Shibata M, Onkubo T, Takahashi H, Inoki R: Modified formalin test: characteristic biphasic pain response. Pain 38: 347-352, 1989
33. Norton PJ, Asmundson GJ: Anxiety sensitivity, fear, and avoidance behavior in headache pain. Pain 111:218-223, 2004
49. Sieve AN, King TE, Ferguson AR, Grau JW, Meagher MW: Pain and negative affect: evidence the inversive benzodiazepine agonist DMCM inhibits pain and learning in rats. Psychopharmacol (Berl) 153:180-190, 2001
34. Nunes-de-Souza RL, Canto-de-Souza A, da-Costa M, Fornari RV, Graeff FG, Pela IR: Anxiety-induced antinociception in mice: effects of systemic and intra-amygdala administration of 8-OH-DPAT and midazolam. Psychopharmacol 150: 300-310, 2000
50. Strigo IA, Bushnell, MC, Boivin, M, Duncan, GH: Psychophysical analysis of visceral and cutaneous pain in human subjects. Pain 97:235-246, 2002
35. Pakulska W, Czarnecka E: Effect of diazepam and midazolam on the antinociceptive effect of morphine, metamizol and indomethacin in mice. Phamazie 56:89-91, 2001
51. Tatsuo MA, Salgado JV, Yokoro CM, Duarte ID, Francischi JN: Midazolam-induced hyperalgesia in rats: modulation via GABA(A) receptors at supraspinal level. Eur J Pharmacol 370:9-15, 1999
36. Palaoglu O, Ayhan IH: The possible role of benzodiazepine receptors in morphine analgesia. Pharmacol Biochem Behav 25:215-217, 1986
52. Tjolsen A, Berge OG, Hunskaar S, Rosland JH, Hole K: The formalin test: an evaluation of the method. Pain 51:517, 1992
37. Pang CS, Tsang SF, Yang JC: Effects of melatonin, morphine and diazepam on formalin-induced nociception in mice. Life Sci 12 68:943-951, 2001
53. Tonussi CR, Ferreira SH: Rat knee joint carrageenin incapacitation test: an objective screen for central and peripheral analgesics. Pain 48:421-427,1992
38. Parada CA, Tambeli CH, Cunha FQ, Ferreira SH: The major role of peripheral release of histamine and 5-hydroxytryptamine in formalin-induced nociception. Neurosci 102: 937-944, 2001
54. Tonussi CR, Ferreira SH: Mechanisms of diclofenac analgesia: direct blockade of inflammatory sensitization. Eur J Pharmacol 251:173-179, 1994
39. Patil CS, Jain NK, Singh A, Kulkarni SK: Modulatory effect of ciclooxygenase inhibitors on sildenafil-induced antinociception. Pharmacol 69:183-189, 2003 40. Rhudy JL. Grimes JS, Meagher MW: Fear-induced hypoalgesia in humans: effects on low intensity thermal stimulation and finger temperature. J Pain 5:458-468, 2004 41. Riccioppo Neto F: The local anesthetic effect of cyproheptadine on mammalian nerve fibers. Eur J Pharmacol 54: 203-207; 1979 42. Rigel DF, Katona PG: Effects of antihistamines and local
55. Xie G, Wang Y, Sharma M, Gabriel A, Mitchell J, Xing Y, Meuser T, Palmer PP: 5-Hydroxytryptamine-induced plasma extravasation in rat knee joint is mediated by multiple prostaglandins. Inflamm Res 52:32-38, 2003 56. Yaksh TL, Ozaki G, McCumber D, Rathbun M, Svensson C, Malkmus S, Yaksh MC: An automated flinch detecting system for use in the formalin nociceptive biossay. J Appl Physiol 90:2386-2402, 2001 57. Zambotti F, Zonta N, Tammiso R, Conci F, Hafner B, Zacca L, Ferrario P, Mantegazza P: Effects of diazepam on nociception in rats. Naunyn Schmiedebergs Pharmacol 344: 84-89, 1991
Formalin Injection Into Knee Joints of Rats: Pharmacologic Characterization of a Deep Somatic Nociceptive Model Maria Alcina Martins, Lúcia de Castro Bastos, and Carlos Rogério Tonussi Department of Pharmacology, Federal University of Santa Catarina, Campus Universitário–Trindade, Florianópolis, Santa Catarina.
Abstract: Formalin (0.25, 0.5, 3, and 5%) injected into the knee joint of rats induced a dosedependent nociception that was featured by 2 phases of intense guarding behavior of the affected limb, interposed by a period of quasinormal gait (quiescent phase). The guarding behavior during a period of forced gait was measured by the total time the paw of the affected limb was not in contact with the surface of a revolving cylinder (paw elevation time [PET]). Pretreatment with morphine (4 mg/kg, subcutaneously) reduced PET in both nocifensive phases, and naloxone (1 mg/kg, subcutaneously) antagonized morphine’s effect. The cyclooxygenase inhibitor diclofenac (5 mg/kg, intraperitoneal) reduced only the second phase of nocifensive responses. A low dose of the benzodiazepine midazolam (0.25 mg/kg, intraperitoneal) significantly reduced only the second phase of response, but a higher dose (1 mg/kg, intraperitoneal) had no effect. A subconvulsant, anxiogenic dose of pentylenetetrazol (30 mg/kg, intraperitoneal) also did not affect the PET increase induced by formalin. The antihistamine meclizine (2.5 mg/kg, intraperitoneal) caused an increase of the response in the second phase, but a higher dose (7.5 mg/kg, intraperitoneal) caused inhibition. The peripheral antihistamine loratadine (5 and 10 mg/kg, intraperitoneal) also caused an increase of the second phase. Neither antihistamine altered the first phase of PET. These results reproduced previous findings with classical analgesics in formalin-induced nociception. However, the pronociceptive effect of antihistamines, and the antinociceptive effect of midazolam observed here suggest that formalin-induced incapacitation introduces new characterists of nociceptive system that may complement the study of analgesics. Perspectives: Anxiety is thought to influence pain experience in an opposing manner depending on nociception originates in cutaneous or deep somatic/visceral tissues. The present formalin-induced nociceptive test may help to predict more reliably the pain-killing effect of new pharmacologic strategies, with classical or nonclassical mechanisms, for the treatment of clinically relevant pains, which are generally originated in deep structures. © 2006 by the American Pain Society Key words: Articular pain, arthritis, benzodiazepines, anxiety, opioids, NSAID, antihistamines.
F
ormalin has long been used as a nociceptive stimulus in different animal models. The most used approach has been the subcutaneous injection into the rat or mouse footpad, which elicits a well-known behavioral response characterized by intense flinches of the injected paw, augmented attention to the paw generally
Received February 15, 2005; Revised June 23, 2005; Rerevised September 5, 2005; Accepted September 8, 2005. Address reprint requests to Carlos Rogério Tonussi, Department of Pharmacology, Federal University of Santa Catarina, Campus Universitário – Trindade, Florianópolis, SC, 88040-900. E-mail: [email protected] 1526-5900/$32.00 © 2006 by the American Pain Society doi:10.1016/j.jpain.2005.09.002
100
featuring the licking of the paw, and also by a less dramatic response characterized by a repeated lifting of the injected paw from the floor. The best known characterization of the behavioral nocifensive responses elicited by the injection of formalin into the paw comes from Dubuisson and Dennis’s work.11 Several subsequent modifications were applied to the original idea in order to improve the quantification of the inferred nociception or to propose different regional models of persistent nociception by injecting formalin in orofacial, temporomandibular, tail, muscle, and intracolonic tissues.2,6,7,17,22,25,29,45,48 The development of models based on the injection of formalin in tissues other than skin is very useful for the better understanding of noci-
ORIGINAL REPORT/Martins et al ceptive transmission and modulation, because there is a consensus that nociceptive inputs from deep tissues are processed by different sensory, affective, and motivational substrates.23,31,50 Notwithstanding, in all of the above approaches, the quantification of formalin-induced responses remained dependent on the subjective human observation, which clearly introduces a significant bias into the results from different observers and/or research groups. Attempts to implement unbiased, automated recording systems for the registering of the formalin-induced nociception have been made in dogs and rats.19,44,56 The most recent implementations, however, would be sensitive to the action of sedative, analeptic drugs, because these are passive models and do not require the execution of a task by the animals. In this work, we present a model to study formalin-induced nociception in rats based on an apparatus previously described for the measurement of articular inflammatory incapacitation, together with its pharmacologic characterization.
Material and Methods Animals The experiments were performed on 200- to 300-g male Wistar rats housed in temperature-controlled room (22 ⫾ 2°C) under a 12-12 hour light/dark cycle with free access to water and food. All behavioral testing were performed between 7:00 AM and 2:00 PM. Animal care and handling procedures were in accordance with the ethical guidelines of the International Association for the Study of Pain18 and also approved by the Ethics Committee for Animal Research of Federal University of Santa Catarina. The animals were euthanized in a CO2 chamber right after the experimental sessions.
Drugs and Vehicles Morphine and midazolam (Laboratórios Cristália, Itapira, SP, Brazil), diclofenac (Novartis, São Paulo, SP, Brazil), meclizine, naloxone, and pentylenetetrazol (Sigma, St. Louis, Mo), and loratadine (Aventis Pharma, São Paulo, SP, Brazil) were used. Loratadine was dissolved in polyoxyethylenesorbitan monoleate (Tween 80), not exceeding 1% of the total volume in physiologic saline, and meclizine was dissolved in sunflower oil. The other drugs were dissolved in physiologic saline. All treatments were administered by intraperitoneal route in a volume of 0.1 mL/100 g of body weight, except morphine and naloxone, which were injected subcutaneously. Morphine, diclofenac, loratadine, and meclizine were injected 60 min before, while midazolam and pentylenetetrazol were injected 30 min before formalin. Naloxone was given 70 min before formalin, ie, 10 min before morphine. Formalin (formaldehyde 37% - Merck AG, Germany) was diluted in saline to 0.25, 0.5, 3, and 5%, considering the initial concentration as 100%. Control group treatments were carried out with the vehicle of the respective test drug.
101
Knee Joint Incapacitation Test Induced by Formalin The rat knee joint incapacitation test is described in detail elsewhere.53 In this test, rats are placed on a revolving cylinder (30 cm diameter; 3 rpm) for 1-min periods and a computer-assisted device measures the total time that a specific hind paw was not in contact with the cylinder surface (ie, paw elevation time [PET]). Normally, control animals display a PET of approximately 10 s, whereas algogenic substances injected into the knee joint increase this value only in the affected limb. In order to induce incapacitation, 50 L of formalin solution (5%), diluted in sterile saline, was injected into the right knee joints of rats. The injection site was first shaved and treated with an iodine alcohol antiseptic solution. The animals were gently restrained in a supine position by the hands of the experimenter, and intraarticular injections were quickly performed with 27-gauge needles. The PET was measured each 5 min after formalin injection, as described above. As the nocifensive behavior was scored automatically, the experimenters were not blind to the treatment protocols.
Motor Impairment Assessment When the cylinder started revolving, the animals promptly and spontaneously walked to keep themselves on top, without assistance from the experimenter. Each experiment comprised 13 periods of gait registering for each animal. Any repeated delay to react to the cylinder movement or an animal fall was considered motor impairment. A treatment was identified as causing no motor impairment if the above-mentioned events were not observed.
Statistical Analysis All statistical analyses were carried out using Graphpad Prism version 3 (http://www.graphpad.com). Results are expressed as mean ⫾ SE mean in the figures. Comparisons of experimental curves were made with paired, two-tailed Student t test or one-way analysis of variance (ANOVA) for repeated measures and, when a significance level of at least P ⬍ .05 was detected, analysis was followed by the Tukey multiple comparison test. Quiescent phase analysis was performed with simple one-way ANOVA.
Results Dose-Response Relationship of FormalinInduced Nocifensive Responses Injection of 0.5, 3, and 5% formalin solution into 1 knee joint of rats induced a dose-related nocifensive response characterized by the expression of guarding behavior that was registered as the increase of the PET compared with pre-formalin values (Fig 1). At a concentration of 0.25% the PET registered was not different from the physiologic saline control. There was also no difference between the mean PET of 3% and 5% groups, although the variability of response was greater in the
102
Formalin Injection Into Knee Joints of Rats
Figure 2. Antinociceptive effect of morphine. Morphine (4 mg/ Figure 1. Dose response relationship of the formalin-induced incapacitation. Formalin (0.25, 0.5, 3, and 5 %) was injected into the knee joint in a volume of 50 L, and the paw elevation time (PET) was measured immediately after. C represents PET before formalin injection (0 min). Control group received only physiologic saline into the knee joint, n indicates the number of animals in each group. * indicates statistical significance (P ⬍ .05) when compared to interphase control value (one-way ANOVA followed by Tukey test).
3% group. The time course of the response presented a first, short-lasting phase (0-5 min interval) immediately after the injection followed by a second, long-lasting phase (10-60 min interval). The PET registered immediately after formalin injection were 45.4 ⫾ 2.5 s for 5%, 48.9 ⫾ 3.2 s for 3%, 34.8 ⫾ 4.4 s for 0.5%, and 16.3 ⫾ 1.8 s for 0.25%. The second phase also presented a characteristic peaking response that was usually reached 25 min after formalin injection. The PET registered at the peak of the second phase were 46.5 ⫾ 4.0 s for 5%, and 49.3 ⫾ 7.2 s for 3%. Both phases were interposed by a period of apparently lower nociceptive inputs from the joint (quiescent phase, 5-10 min interval). The group injected with 0.5% formalin peaked earlier at 10 min after injection (32.9 ⫾ 4.3 s), and presented a quiescent phase (24.0 ⫾ 2.3 s) significantly greater than the control (13.9 ⫾ 0.8 s) (P ⬍ .05). The guarding behavior expressed by the animals after injection of formalin was obvious even when the animals were not walking on the cylinder, but after 60 min there was no signs of nociception. Also after this period, we observed that the higher formalin concentration produced an intense synovial sweeling. For the subsequent experimental purposes the 5% formalin solution was used as a standard nociceptive stimulus.
Antinociceptive Effect of Morphine and Diclofenac in Knee Joint Incapacitation Test Induced by Formalin Morphine (4.0 mg/kg, sc, 60 min before) inhibited the nocifensive response to 5% formalin in both nocifensive phases (P ⬍ .01; Fig 2). The PET registered immediately after formalin injection were 42.6 ⫾ 1.2 s for the salinetreated group and 23.5 ⫾ 2.5 s for the morphine-treated
kg, sc, 60 min before formalin) inhibited PET in both phases (P ⬍ .01). Naloxone (1.0 mg/kg, sc, 10 min before morphine) reversed morphine’s effect in both phases (P ⬍ .01). C represents PET before formalin injection (0 min). Control group received only physiologic saline by subcutaneous route, n indicates the number of animals in each group. Statistical differences were detected by a one-way ANOVA for repeated measures followed by Tukey test.
group. The PET registered at the peak of the second phase were 40.4 ⫾ 5.5 s for saline- and 22.7 ⫾ 3.7 s for morphine-treated groups. This inhibitory effect produced by morphine was totally reversed by naloxone (1.0 mg/kg, subcutaneously, 10 min before morphine) (P ⬍ 0.01; Fig 2). Interestingly, naloxone caused the anticipation of the peak response in the second phase—ie, a deviation to the left of the time-course curve. At the dose above, morphine did not impair the normal locomotion of the animals. For comparison, a dose of 50 g/kg, sc, of
Figure 3. Antinociceptive effect of diclofenac. Diclofenac (5.0 mg/kg, intraperitoneal, 60 min before formalin) significantly reduced the PET in the second phase of the formalin-induced response (P ⬍ .001) compared to the control group. Diclofenac (0.5 mg/kg, intraperitoneal) did not have a significant effect on the PET. C represents PET before formalin injection (0 min). Control group received only physiologic saline by intraperitoneal route. n indicates the number of animals in each group. Statistical differences were detected by a one-way ANOVA for repeated measures followed by Tukey test.
ORIGINAL REPORT/Martins et al
103 54
atogenic effects, respectively. These treatments did not affect the first peak of response (PETsal ⫽ 42.6 ⫾ 1.2 s, PET0.5mg ⫽ 39.4 ⫾ 3.8 s and PET5mg ⫽ 37.1 ⫾ 3.8 s) nor the quiescent phase (PETsal ⫽ 15.5 ⫾ 1.3 s, PET0.5mg ⫽ 13.8 ⫾ 0.8 and PET5mg ⫽ 10.7 ⫾ 1.8 s).
Dual Effect of H1 Histamine Receptor Antagonists
Figure 4. Dual effect of H1 histamine receptor antagonists. (A) Meclizine (2.5 mg/kg, intraperitoneal, 60 min before formalin) significantly increased PET (P ⬍ .05), but the higher dose (7.5 mg/kg, intraperitoneal) significantly decreased PET (P ⬍ .001) in the second phase of the formalin-induced response compared to the control group (one-way ANOVA for repeated measures followed by Tukey test) (B) Loratadine (5 and 10 mg/kg, intraperitoneal, 30 min before formalin) induced a significant increase in the PET in the second phase of the formalin-induced response (P ⬍ .001), compared to the respective control group (paired Student t test). C represents PET before formalin injection (0 min). The control group for meclizine received only sunflower oil, whereas that for loratadine received only physiological saline with Tween 80 (1%) by intraperitoneal route. n indicates the number of animals in each group.
fentanyl impaired the animals’ spontaneous positioning on the top of the cylinder, causing frequent falls (data not shown). The cyclooxygenase inhibitor diclofenac (5 mg/kg, intraperitoneal, 60 min before) also induced a significant antinociceptive effect, but only in the second phase (P ⬍ .001; Fig 3). At a smaller dose (0.5 mg/kg), diclofenac did not affect the formalin-induced nocifensive response (Fig 3). The lower and higher doses were chosen for comparison with our previous data in carrageenan and LPS models of inflammatory knee joint incapacitation, when they presented maximal antinociceptive and antiedem-
Formalin-induced nociceptive test has been demonstrated to be sensitive to antihistamines. Pretreating the animals with the blood brain barrier permeable, H1 receptor antagonist, meclizine (1, 2.5, and 7.5 mg / kg, intraperitoneal, 60 min before; Fig 4A) caused a dual effect on the nocifensive behavior exhibited by the rats after formalin injection into knee joints. The dose of 2.5 mg/kg caused the increase of the paw elevation time in the second phase period (P ⬍ .05), but the higher dose (7.5 mg/kg) decreased this response (P ⬍ .001), when compared to the control time-course curve. However, it was not observed any effects of the drugs in the first phase (PETveh ⫽ 49.9 ⫾ 2.1 s, PET1mg ⫽ 39.3 ⫾ 5.9 s, PET2.5mg ⫽ 38.6 ⫾ 7.2 s and PET7.5mg ⫽ 40.8 ⫾ 4.5 s) (P ⫽ .42, one-way ANOVA). These treatments did not affect the quiescent phase either (PETveh ⫽ 20.9 ⫾ 3.9 s, PET1mg ⫽ 18.4 ⫾ 5.0 s, PET2.5mg ⫽ 19.6 ⫾ 3.5 s and PET7.5mg ⫽ 16.6 ⫾ 3.6 s), nor was there motor impairment with the higher dose. Pretreatment with the peripheral H1 receptor antagonist, loratadine (5.0 and 10.0 mg/kg, intraperitoneal, 60 min before; Fig 4B) increased the second phase of nocifensive responses induced by formalin (P ⬍ .001). The lower dose tested (2.5 mg/kg) did not produce a statistically significant effect. These treatments also did not affect the first peak of response (PETveh ⫽ 49.5 ⫾ 4.1 s, PETlor ⫽ 43.1 ⫾ 3.4 s) nor the quiescent phase (PETveh ⫽ 16.7 ⫾ 1.5 s, PETlor ⫽ 19.0 ⫾ 3.4 s).
Effect of Midazolam and Pentylenetetrazol Benzodiazepine anxiolitic drugs or anxiogenic conditions have been demonstrated to cause pronociceptive and antinociceptive effects in cutaneous models of nociception, respectively. Pretreatment with the GABAA receptor benzodiazepine site ligand midazolam (0.25 mg/ kg, intraperitoneal, 30 min before; Fig 5A) caused an intense inhibition of the second phase nocifensive response (PETsa l⫽ 53.0 ⫾ 3.8 s, PETmdz ⫽ 30.0 ⫾ 6.8 s; P ⬍ .001) but only a tendency (P ⫽ .06) to inhibit the first phase (PETsal ⫽ 52.3 ⫾ 3.4 s, PETmdz ⫽ 40.6 ⫾ 4.7 s), and did not affect the quiescent phase (PETsal ⫽ 20.7 ⫾ 2.3 s, PETmdz ⫽ 18.4 ⫾ 2.3 s). A 4-fold higher dose (1.0 mg/kg, intraperitoneal, 30 min before; Fig 5A ), however, showed no effect in first phase (PETsal ⫽ 45.0 ⫾ 4.8 s, PETmdz ⫽ 45.2 ⫾ 2.1 s) and second phase (PETsal ⫽ 39.6 ⫾ 2.3 s, PETmdz ⫽ 37.5 ⫾ 6.3 s), but did exhibit a tendency to inhibit the quiescent phase (PETsal ⫽ 18.4 ⫾ 3.6 s, PETmdz ⫽ 26.8 ⫾ 1.6 s). Pretreatment with the GABAA chloride channel blocker pentylenetetrazol in an anxiogenic, subconvulsant dose (30 mg/kg, intraperitoneal, 30 min be-
104
Formalin Injection Into Knee Joints of Rats sive response interposed by a period of quiescence. This is important for comparative purposes because it suggests that the neural apparatus at the articulation is similar to that found in subcutaneous tissues, in terms of its capacity to respond to formalin. Furthermore, formalin produced a dose-related nocifensive response, although the dose producing intermediate effects (0.5%) also produced a less marked quiescent phase. The quiescent phase is recognized as being due to an active inhibition of the nociceptive transmission, integrating supraspinal as well as spinal circuitries.21,27 If this inhibition is also modulating articular nociception induced by formalin, our observation suggests that the activation of this inhibitory system is proportional to the intensity of the nociception in the early phase. In addition, this may be specific to, or at least more evident in, this model because, in experiments using subcutaneous injection of formalin, the quiescent phase was the same regardless of the formalin concentration used.21,52 By extending this point, the earlier peak of incapacitation in the second phase of the response elicited by 0.5% formalin may be likely due to a lesser inhibitory tone in the quiescent phase. Whatever the case, we think that the quiescent phase in any formalin-based test should not be neglected because it can account for important and specific changes in the nociceptive system.
The Sensitivity to Classical Analgesics Figure 5. Effect of midazolam and pentylenetetrazol. (A) Midazolam (0.25 mg/kg, intraperitoneal, 30 min before formalin) significantly reduced paw elevation time in the second phase of the formalin-induced response (P ⬍ .001) compared to the control group (paired Student t test). (B) Midazolam (1.0 mg/kg, intraperitoneal) and pentylenetetrazol (30 mg/kg, intraperitoneal) did not have a significant effect on the PET (paired Student t test), but a tendency to increase PET in the interphase. C represents PET before formalin injection (0 min).
fore; Fig 5B) did not induce changes in either phase (PET0min ⫽ 42.4 ⫾ 3.4 s, PET25min ⫽ 40.0 ⫾ 4.4 s), but also showed a tendency to inhibit the quiescent phase (PETq ⫽ 28.7 ⫾ 4.3 s). Dose and times of pretreatment with midazolam and pentylenetetrazol were according previous reports from studies in anxiety models.5,20 These treatments produced neither motor incoordination, in the case of midazolam, nor hyperexcitability and convulsion, in the case of pentylenetetrazol.
Discussion It was shown in the present study that formalin injection into rat knee joints induces a guarding behavior of the limb, which can be registered as the paw elevation time during the forced walk on a revolving cylinder, using the same apparatus we previously used for inflammatory articular nociception quantification.53 As is generally observed in models using formalin as a nociceptive stimulus, the animals exhibited two periods of nocifen-
The analgesic efficacy of morphine and its antagonism by naloxone emphasizes the sensitivity of the model to the analgesic action of opioids. As a -opioid receptor acting analgesic, morphine inhibited both phases of the nocifensive response. In addition, besides the sensitivity to opiate analgesics, the enforced walk on the cylinder allows the discrimination of pure analgesic effect from other effects on locomotor activity (eg, sedation, incoordination, catalepsy). This feature could be observed in the treatment with fentanyl, which at a dose generally used as analgesic, caused the recurrent fall of the animals from the cylinder. Because normal animals try actively to walk on the top of the cylinder, any difficulty in maintaining this position is considered to represent impairment of motor control. In contrast, the cyclooxygenase inhibitor diclofenac inhibited only the second phase of the formalin-induced response. The intensity of the inhibitory effects of diclofenac in the second phase of responses suggests that also in the articular tissues the production of prostaglandins is an important component of the nociception at this stage52 and that this model can be used as a sensitive test for detection of aspirin-like antiinflammatories. Two doses of diclofenac was used here. The lower dose was previously observed to produce maximal antinociceptive effect in inflammatory models of carrageenan- and LPSinduced arthritis,30,54 whereas the higher dose was observed to produce maximal antiedematogenic effect in a model of LPS-induced arthritis (unpublished data). The high effectiveness of the lower dose of diclofenac was previously seen to be mediated by a nitric oxide/cGMPdependent mechanism,54 whereas the higher dose
ORIGINAL REPORT/Martins et al seems to be mediated by the inhibition of prostaglandin synthesis, as suggested by the antiedematogenic effect. What this result means is that, despite the possible activation of the nitric oxide pathway, it was not enough to produce an antinociceptive effect in the present model. The possibility that by using enhancers of the nitric oxide pathway, such as an inhibitor of the cGMP phosphodiesterase combined with diclofenac,3,39 we might also observe an antinociceptive effect with low doses of diclofenac remains to be evaluated. However, the effect of opioid and nonsteroidal antiinflammatory drugs in this model is in line with classical formalin tests52 and reinforces the validity of the method as a confident tool for nociception studies.
The Antinociceptive Effect of Midazolam The effect of benzodiazepines given systemically in nociceptive tests has been inconsistent. These drugs were found to produce moderate antinociception, no effect , or even a pronociceptive effect, with this last result being corroborated by the antinociceptive effect induced by an inverse benzodiazepine agonist.13,36,37,49,51,53,57 Furthermore, several authors reported an antagonistic effect of the morphine-induced analgesia as well as nonnarcotic analgesia by benzodiazepines, while others have reported a potentiation of a nonnarcotic analgesic with alprazolam.9,10,35,36 It has been recognized that nociception elicited in cutaneous and deep tissues activates different regions in the periaqueductal gray area and amygdala, structures that participate in anxiety/fear behavior and are involved in the activation of descending modulatory projections to the spinal cord.24,31 Anxiety/ fear may alter animal and human nociception in both directions, depending on their origin. For example, cutaneous elicited nociception was attenuated by anxiety/ fear producing paradigms, whereas nociception of deep origin may be enhanced or unaffected.8,12,15,16,33,40 Benzodiazepines can reverse stress-induced analgesia, probably by attenuating the anxiogenic/fear drive for the activation of endogenous antinociceptive circuitries, which may explain the hyperalgesic effect of the drug in cutaneous pain models.12,16 Conversely, the anxiety facilitation of persistent nociception elicited in deep tissues could be attenuated by benzodiazepines resulting in antinociception.31 In line with this idea, midazolam was seen to produce antinociception in the writhing test in mice, a model of visceral nociception.34 This may be one of the explanations for this remarkable difference between our results with midazolam and the previous findings in the literature, but it does not explain the lack of antinociceptive effect with the higher dose of midazolam. GABAA agonist injection in the rostral ventromedial medulla or intracerebroventricular injection of benzodiazepines strongly increases nociception.14,51 On the contrary, spinally injected benzodiazepines were seen to produce antinociception in several models.32,57 In view of these data, it is not difficult to argue that much of the contradictory data about the effects of systemically given benzodiazepines on nociceptive models may arise
105 from a differential action of the drug on supraspinal or spinal circuitries. We could, therefore, propose that the low dose of midazolam would be facilitating a spinal GABAergic mechanism of antinociception, while the higher dose of midazolam would be inhibiting a supraspinal inhibitory system, as proposed by Gilbert and Franklin.14 This inhibition of a descending inhibitory system is consistent with the enhanced nociception observed in the quiescent phase when the animals were treated with the higher dose of midazolam and also may explain why we did not observe an antinociceptive effect with diazepam in a previous study using carrageenaninduced articular nociception.53 Pentylenetetrazol, an inhibitor of the Cl- channel associated with the GABAA receptor, acted similarly to the higher dose of midazolam and also showed a tendency to increase the nociception in the quiescent phase. This result contradicts previous reports showing that subconvulsant, anxiogenic doses of pentylenetetrazol induce antinociception in thermally and mechanically elicited cutaneous nociception.8,43 However, as discussed above, the finding does support the notion that anxiety/fear enhances the sensitivity to persistent nociception from deep tissues.
The Effect of Antihistamines Another peculiarity of the present model may be our results with meclizine, which produced a significant increase and decrease of the nocifensive responses. Systemically given H1 antihistamine drugs were demonstrated to be antinociceptive in several models and also clinically.46,47 The antinociceptive action of antihistamines has been attributed to a central action on histaminergic pathways, although several of them may also possess a local anesthetic-like mechanism of action.41,42 Our result with the higher dose of meclizine, although small, was consistent with the antinociceptive effect of antihistamines previously reported by others. However, the hyperalgesic effect observed with the lower dose was unexpected. A possible explanation for this result may be derived from the observation with loratadine treatment, which caused hyperalgesic effects. Loratadine is only weakly permeant to the blood brain barrier, which precludes a central action, and also does not appear to possess a local anesthetic-like effect.1,4,28 On this basis, we can only suppose a local action by the blockade of H1 receptors. Local application of antihistamines in the formalin test was recently readdressed by Parada et al,38 confirming an antinociceptive effect. However, in that study the authors only used antihistamines with recognized local anesthetic-like effect, which did not answer the question of how locally given antihistamines without local anesthetic effect could affect the formalininduced response. Our observation of an enhancement of the nocifensive responses after treating the animals with loratadine may be due to an inhibitory effect over vasodilation. Neurogenic vasodilation and plasma exudation in rat knee joints is substantially mediated by histamine and serotonin released from mast-cells.26,55 Formalin induces an intense swelling into synovia. Antagonizing the vasodilation and plasma leakage by H1
106 antagonism would delay the clearance of formalin from the synovial space and/or delay its dilution due to joint sweeling, which would increase its nociceptive action. In conclusion, it must be born in mind that relevant clinical pains are of deep somatic/visceral origin not only because of their pathologic features, but also because they elicit more intense emotional and autonomic responses.24,50 In this context, cutaneous nociceptive models may be of minor significance in predicting the effectiveness of new analgesic or adjuvant therapies. On the other hand, models of visceral or deep somatic pain are generally less practical, or do not give clear-cut results about the efficiency of a drug as an analgesic. Formalin-
References 1. Algemissen B, Hermes B, Henz BM U, Berlien HP: Laserinduced weal and flare reactions: clinical aspects and pharmacological modulation. Br J Dermatol 146:863-868, 2002 2. Aloisi AM, Decchi B, Carli G: Importance of site of injection in the formalin test. Med Sci Res 23:603-610, 1995 3. Asomoza-Espinosa R, Alonso-Lopez R, Mixcoatil-Zecuatil T, Aguirre-Banuelos P, Torres-Lopez JE, Granados-Soto V: Sildenafil increase diclofenac antinociception in the formalin test. Eur J Pharmacol 418:195-200, 2001 4. Barnett A, Lorio LC, Kreutner W, Tozzi S, Ahn HS, Gulbenkian A: Evaluation of the CNS properties of SCH 29851, a potencial non-sedating antihistamine. Agents Actions 14: 590-597, 1984 5. Bertoglio LJ, Anzini C, Lino-de-Oliveira C, Carobrez AP: Enhanced dorsolateral periaqueductal gray activity counteracts the anxiolytic response to midazolam on the elevated plus-maze Trial 2 in rats. Behav Brain Res 162:99-107, 2005 6. Clavelou P, Dallel R, Orliaguet T, Woda A, Raboisson P: The orofacial formalin test in rats: effects of different formalin concentration. Pain 62:295-301, 1995 7. Coderre TJ, Fundytus ME, McKenna JE, Dalal S, Melzack R: The formalin test: a validation of the weighted-scores method of behavioral pain rating. Pain 54:43-50, 1993 8. Conceição IM, Maiolini Junior M, Mattia N, Vital MA, Santos BR, Smaili S, Frussa-Filho R: Anxiety-induced antinociception in the mouse. Braz J Med Biol Res 25:831-834, 1992 9. Daghero AM, Bradley EL Jr, Kissin I: Midazolam antagonizes the analgesic effect of morphine in rats. Anesth Analg 66:944-947, 1987 10. Dolezal T, Krsiak M: Augmentation of analgesic effect of ibuprofen by alprazolam in experimental model of pain. Physiol Res 52:179-184, 2002 11. Dubuisson D, Dennis SC: The formalin test: a quantitative study of analgesic effects of morphine, meperidine, and brain stem stimulation in rats and cats. Pain 4:161-174, 1977
Formalin Injection Into Knee Joints of Rats induced incapacitation may be a useful tool for deep somatic nociceptive studies. The model features an automated registering system with reliable sensitivity and specificity to classical analgesics. With nonclassical analgesics, such as benzodiazepines and sedative antihistamines, the model may help to predict to what extent analgesia may be expected.
Acknowledgments The authors wish to thank the Coordenação para o Aperfeiçoamento de Pessoal do Ensino Superior (CAPES) for support in the form of M.Sc. studentships for MAM and LCB.
14. Gilbert AK, Franklin KB: GABAergic modulation of descending inhbitory system from the rostral ventromedial medulla (RVM). Dose-response analysis of nociception and neurological deficits. Pain 90:25-36, 2001 15. Gunter WD, Shepard JD, Foreman RD, Meyers DA, Greenwood-Van Meerveld B: Evidence for visceral hypersensitivity in high anxiety rats. Physiol Behav 69:379-382, 2000 16. Helmstetter FJ: Stress-induced hypoalgesia and defensive freezing are attenuated by application of diazepam to the amygdala. Pharmacol Biochem Behav 44:433-438, 1993 17. Hunskaar S, Berge OG, Hole K: Dissociation between antinociception and anti-inflammatory effects of acetylsalicylic acid and indomethacin in the formalin test. Pain 25: 125-132, 1986 18. IASP (International Association for Study of Pain): Ethical guidelines for investigation of experimental pain in conscius animals. Pain 16:09-110, 1983 19. Jett MF, Michelson S: The formalin test in rat: validation of an automated system. Pain 64:19-25, 1996 20. Jones N, Duxon MS, King SM: Ethopharmacological analysis of the unstable elevated exposed plus maze, a novel model of extreme anxiety: predictive validity and sensitivity to anxiogenic agents. Psychopharmacology 161:314-323, 2002 21. Kaneko M, Hammond DL: Role of spinal gamma-aminobutyric acid A receptors in formalin-induced nociception in the rat. J Pharmacol Exp Ther 282:928-938, 1997 22. Keay KA, Bandler R: Deep and superficial noxius stimulation increased Fos-like immunoreactivity in different regions of the midbrain periaqueductal grey of the rat. Neurosci Lett 14 154:23-26, 1993 23. Keay KA, Bandler R: Distinct central representations of inescapable and escapable pain: observations and speculation. Exp Physiol 87:275-279, 2002 24. Keay KA, Li QF, Blander R: Muscle pain activates a direct projection from ventrolateral medulla in rats. Neurosci Lett 1 290:157-160, 2000
12. Fanselow MS, Helmstetter FJ: Conditional analgesia, defensive freezing, and benzodiazepines. Behav Neurosci 102: 233-243, 1988
25. Kolesnikov Y, Cristea M, Oksman G, Torosjan A, Wilson R: Evaluation of the tail formalin test in mice as a new model to assess local analgesic effects. Brain Res 17 1029:217-223, 2004
13. Gatch MB: Effects of benzodiazepines on acute and chronic athanol-induced nociception in rats. Alcohol Clin Exp Res 23:1736-1743, 1999
26. Lam FY, Ferrell WR: Mediators of substance P-induced inflammation in the rat knee joint. Agents Actions 31:298307, 1990
ORIGINAL REPORT/Martins et al 27. Matthies BK, Franklin KB: Formalin pain is expressed in decebrated rats but not attenuated by morphine. Pain 51: 199-206, 1992 28. Mattila MJ, Paakkari I: Variations among non-sedating antihistamines: are there real differences? Eur J Clin Pharmacol 55:85-93, 1999 29. Miampamba M, Chery-Croze S, Gorry F, Berger F, Chayvialle JA: Inflammation of the colonic wall induced by formalin as a model of acute visceral pain. Pain 57:327-334, 1994 30. Motta AF, Borges Junior NG, da Fonseca JC, Tonussi CR: The antinociceptive effect of iontophoretic direct application of diclofenac to arthritic knee joints of rats. Life Sci 73:1995-2004, 2003
107 anesthestics on excess tachycardia in conscius dogs. J Pharmacol Exp 238:367-371, 1986 43. Rocha L, Cano A, Cruz Z, Omana-Zapata I, Villalobos R, Maidment NT: Opioid peptide systems following a subconvulsant dose of pentylenetetrazol in rats. Epilepsy Res 37: 141-150, 1999 44. Rodriguez R, Pardo EG: Reversal by acetylsalicylic acid of pain induced functional impairment. Life Sci 5:775-781, 1966 45. Roveroni RC, Parada CA, Cecilia M, Veiga FA, Tambeli CH: Development of a behavioral model of TMJ pain in rats: the TMJ formalin test. Pain 94:185-191, 2001 46. Rumore MM, Schlichting DA: Analgesic affects of antihistamines. Life Sci 36:403-416, 1985.
31. Neugebauer V, Li W, Bird GC, Han JS: The amygdala and persistent pain. Neuroscientist 10:221-234, 2004
47. Rumore MM, Schlichting DA: Clinical efficacy of antihistamines as analgesics. Pain 25:7-22, 1986
32. Niv D, Davidovich S, Geller E, Urca G: Analgesic and hyperalgesic effects of midazolam: dependence on route of administration. Anesth Analg 67:1169-1173,1988
48. Shibata M, Onkubo T, Takahashi H, Inoki R: Modified formalin test: characteristic biphasic pain response. Pain 38: 347-352, 1989
33. Norton PJ, Asmundson GJ: Anxiety sensitivity, fear, and avoidance behavior in headache pain. Pain 111:218-223, 2004
49. Sieve AN, King TE, Ferguson AR, Grau JW, Meagher MW: Pain and negative affect: evidence the inversive benzodiazepine agonist DMCM inhibits pain and learning in rats. Psychopharmacol (Berl) 153:180-190, 2001
34. Nunes-de-Souza RL, Canto-de-Souza A, da-Costa M, Fornari RV, Graeff FG, Pela IR: Anxiety-induced antinociception in mice: effects of systemic and intra-amygdala administration of 8-OH-DPAT and midazolam. Psychopharmacol 150: 300-310, 2000
50. Strigo IA, Bushnell, MC, Boivin, M, Duncan, GH: Psychophysical analysis of visceral and cutaneous pain in human subjects. Pain 97:235-246, 2002
35. Pakulska W, Czarnecka E: Effect of diazepam and midazolam on the antinociceptive effect of morphine, metamizol and indomethacin in mice. Phamazie 56:89-91, 2001
51. Tatsuo MA, Salgado JV, Yokoro CM, Duarte ID, Francischi JN: Midazolam-induced hyperalgesia in rats: modulation via GABA(A) receptors at supraspinal level. Eur J Pharmacol 370:9-15, 1999
36. Palaoglu O, Ayhan IH: The possible role of benzodiazepine receptors in morphine analgesia. Pharmacol Biochem Behav 25:215-217, 1986
52. Tjolsen A, Berge OG, Hunskaar S, Rosland JH, Hole K: The formalin test: an evaluation of the method. Pain 51:517, 1992
37. Pang CS, Tsang SF, Yang JC: Effects of melatonin, morphine and diazepam on formalin-induced nociception in mice. Life Sci 12 68:943-951, 2001
53. Tonussi CR, Ferreira SH: Rat knee joint carrageenin incapacitation test: an objective screen for central and peripheral analgesics. Pain 48:421-427,1992
38. Parada CA, Tambeli CH, Cunha FQ, Ferreira SH: The major role of peripheral release of histamine and 5-hydroxytryptamine in formalin-induced nociception. Neurosci 102: 937-944, 2001
54. Tonussi CR, Ferreira SH: Mechanisms of diclofenac analgesia: direct blockade of inflammatory sensitization. Eur J Pharmacol 251:173-179, 1994
39. Patil CS, Jain NK, Singh A, Kulkarni SK: Modulatory effect of ciclooxygenase inhibitors on sildenafil-induced antinociception. Pharmacol 69:183-189, 2003 40. Rhudy JL. Grimes JS, Meagher MW: Fear-induced hypoalgesia in humans: effects on low intensity thermal stimulation and finger temperature. J Pain 5:458-468, 2004 41. Riccioppo Neto F: The local anesthetic effect of cyproheptadine on mammalian nerve fibers. Eur J Pharmacol 54: 203-207; 1979 42. Rigel DF, Katona PG: Effects of antihistamines and local
55. Xie G, Wang Y, Sharma M, Gabriel A, Mitchell J, Xing Y, Meuser T, Palmer PP: 5-Hydroxytryptamine-induced plasma extravasation in rat knee joint is mediated by multiple prostaglandins. Inflamm Res 52:32-38, 2003 56. Yaksh TL, Ozaki G, McCumber D, Rathbun M, Svensson C, Malkmus S, Yaksh MC: An automated flinch detecting system for use in the formalin nociceptive biossay. J Appl Physiol 90:2386-2402, 2001 57. Zambotti F, Zonta N, Tammiso R, Conci F, Hafner B, Zacca L, Ferrario P, Mantegazza P: Effects of diazepam on nociception in rats. Naunyn Schmiedebergs Pharmacol 344: 84-89, 1991