- Email: [email protected]
The effect of medetomidine, an α2-adrenoceptor agonist, in various pain tests
European Journal of Pharmacology, 179 (1990) 323-328
323
Elsevier UP 51276
The effect of medetomidine, an a2-adrenoceptor agonist, in various pain tests Antti Pertovaara, Timo Kauppila and Timo Tukeva Department of Physiology. Unit:ersity of Helsmki. Siltavuorenpenger 20 J. 00l 70 ttelsmki. Finland
Received 4 September 1989. revised MS received 16 November 1989. accepted 6 February 1990
Medetomidine, a new a2-adrenoceptor agonist produced dose-dependent (30-100 /tg/kg i.p.) analgesia in the formalin test in rats, and this effect was reversed by atipamezole (1 mg/kg), a new a2-adrenoceptor antagonist. However, medetomidine at the dose of 100 p.g/kg did not influence tail flick latencies or latencies of the biting response to mechanical pinch stimuli. Moreover, medetomidine produced sedation and a decrease in locomotor activity. In comparison, the non-sedative monoaminergic agent, cocaine (25 mg/kg), produced highly significant analgesic effects in the formalin and mechanical pain tests. The cocaine effect in the formalin test was not reversed by atipamezole (1 mg/kg). It is concluded that the analgesic effect of medetomidine in the formalin test is due to supraspinal mechanisms related to sedation and is mediated by ~2-adrenoceptors. The a2-adrenoceptors are not involved in cocaine-induced anagesia. Analgesia; Medetomidine; Atipamezole; Cocaine; Formalin test: Tail flick test; Mechanical pain: (Rat)
1. Introduction Several groups have reported that clonidine, an a2-adrenoceptor agonist, given systemically can produce analgesia (Paalzow, 1974; Dennis et al., 1980; Aceto and Harris, 1981; Fielding et al., 1981; Chan and Lai, 1982; Skingle et al., 1982; Curtis and Marwah, 1987). Medetomidine has been shown to be a highly selective and potent a2adrenoceptor agonist (Virtanen et al., 1988). In a recent study it was found that dexmedetomidine, a stereoisomeric form of medetomidine, prolonged the latency of the behavioral response to tail clamping under halothane anesthesia, and this analgesic effect of dexmedetomidine could be reversed by idazoxan, an a2-adrenoceptor antagonist (Segal et al., 1988). In the current study we wished to determine the effect of medetomidine in various
Correspondence to: A. Pertovaara, Department of Physiology, University of Helsinki, Siltavuorenpenger 20 J, 00170 Helsinki. Finland.
pain tests in awake rats. Since sedation can be a complicating factor in a2-adrenoceptor studies (Tasker and Melzack, 1989) we also evaluated the level of m o t o r activity following medetomidine and compared the analgesic effect of medetomidine with that produced by a non-sedative monoaminergic agent, cocaine (Lin et al., 1989). An attempt was made to reverse the analgesic effects of medetomidine and cocaine by using a novel a2-adrenoceptor antagonist, atipamezole, which has been shown to antagonize medetomidine-induced sedative, hypothermic and neurochemical changes (Scheinin et al., 1988; M a c D o n a l d et al., 1988; Virtanen et al., 1989).
2. Materials and methods Adult male Wistar rats, weight range 200-450 g, were obtained from the Finnish National Laboratory Animal Centre. They had free access to pelleted food chow and water and were on a 12-h
0014-2999/90/$03.50 '~ 1990 Elsevier Science Publishers B.V. (Biomedical Division)
324
light-dark cycle. Medetomidine and atipamezole (both synthetized by the Farmos Group Ltd., Turku, Finland), cocaine and 0.9% saline ( = control) were administered i.p. Analgesia was evaluated by the formalin test, heat-induced tail flick test and mechanically induced tail biting response. In the formalin test (Dubuisson and Dennis, 1977), 0.05 ml of 5% formalin was injected s.c. into the dorsum of the one forepaw. Pain behavior was graded continuously throughout 3-rain observation periods for 30 rain according to the proportion of the time the paw was held up and licked (grade 3), held fully elevated (grade 2), partially weight bearing (grade 1), or fully weight bearing (grade 0). In the heatinduced tail flick test, the rat was placed in a clear plexiglass cylinder and radiant heat was applied to the tail until the tail was moving as detected by a light-emitting sensor. The lacency to the first movement of the tail was used as a parameter of pain sensitivity. The heat source was adjusted to give an average baseline latency of about 1.5 s in untreated rats. A cut-off time of 5 s was used to avoid excessive pain. Since medetomidine is known to produce a decrease in skin temperature (MacDonald et al., 1988) which could produce an artefactual latency increase in the tail flick test (Berge et al., 1988), the tail was kept in a water bath at constant temperature (39 °C) throughout the session, except during actual testing when the tail was first dried then stimulated using radiant heat. This served to eliminate the artefactual increase of tail flick latency due to medetomidineinduced hypothermia. Two trials of the tail flick test were run at 2-min intervals. In the mechanically induced pain test, a hemostatic clamp was applied to the tail and the latency to biting of the tail or clamp was observed. A cut-off time of 5 min was used in the tail biting test. The sedative effect of medetomidine was evaluated using the open field test as described earlier (Hilakivi et al., 1988). Briefly, the open field was a circular arena (83 cm in diameter) covered with lacquered cork painted white. The walls of the arena, 50 cm high, were also white. The floor had three concentric black circles. These circles were subdivided by radiating black lines into 19 segments of approximately equal area. For
testing, each animal was placed on a starting-point in the innermost circle of the field and observed for 2 min. The number of marked floor segments which a rat entered with at least both its front paws, was recorded and used as a measure of locomotion/sedation. Each animal was used only once for each test condition and was killed after the session by an overdose of pentobarbitol. All drugs or drug combinations were given 15 min before the actual testing (in the formalin test, 15 min before the application of formalin) except when the time course of the drug effects was to be tested (see Results for details). The significance of difference in pain scores of the formalin test was evaluated by computing the 95% confidence intervals about each mean. Compared means were considered significantly different when one fell outside the confidence limit of the other. The differences in the tail flick or tail biting latencies and the open field scores were evaluated using a t-test (two-tailed. The P < 0.05 level was considered to represent a significant difference.
3. Results Medetomidine produced dose-dependent analgesia in the formalin test (fig. 1A). At a dose of 30 btg/kg the analgesic effect was significant only from 12 to 21 rain after formalin application. At higher doses the analgesic effect was significant throughout the observation period. At the maximally effective dose (100 g g / k g ) the rats appeared sedated and spent a considerable part of the test time lying with eyes closed. However, motor behavior did not appear to be impaired when they were awake (e.g. they could climb and run). Atipamezole (1 m g / k g ) completely reversed the analgesic effect of medetomidine (100 /.tg/kg) in the formalin test (fig. 1B). Atipamezole alone (1 m g / k g ) did not have any significant effect in the formalin test (fig. 1B). For studying the time course of drug effects, medetomidine (100 /.tg/kg) was applied 3 min after the formalin injection, and atipamezole (1 m g / k g ) 15 min after the formalin injection ( = 12 min after the administration of medetomidine). There was a highly significant de-
325
Pain intensity A
3-
rating *30 A55 pg/kg medet.
~saline
B 3"
• I00 T
~r
2-
~,
,J
j\
0,
i
r
:L
~atipam.
"r
r/L
y=
•
L
z-
T
~-i-i-i-i'
i
I~*.~m.._e d e t.
1; C :3-
0
~--t-~o-- e--e--o-- .--.
3;
2?
,; 3-
,+
• cocaine ~, cocaine+atipam.
T
,,
21-
3;
2?
D
medet.
~T
I ~a~i~pam.
I 0
O,
,~
2'0
0-0-0=0-0=0=0=0=0=0
3~
I~
Time
2~
3~
(min)
Fig. 1. (A) Effects of different i.p. doses of medetomidine (30-100 /xg/kg) on averaged pain intensity ratings in the formalin test. Formalin (5%, 0.05 ml) was injected s.c. into one forepaw at 0 time on the abscissa and the drugs were administered 15 min before 0 time; n = 6 in saline and 100/~g/kg medetomidine groups, n = 4 in other groups. (B) Effects of atipamezole alone (n = 4; l mg/kg), or of 100 # g / k g medetomidine alone (n = 6) or with 1 m g / k g of atipamezole (n = 4) in the formaline test. (C) The time course of the effect of medetomidine (100 itg/kg) and its reversal by atipamezole (1 m g / k g ) in the formalin test (n = 4). Formalin was applied at time 0, medetomidine 3 min later, and atipamezole 15 min following the formalin application. (D) Effects of 25 m g / k g cocaine alone (n = 4) or with 1 m g / k g of atipamezole (n = 4) in the formalin test. The vertical bars represent + S.E.
A Tail flick Latency
B Pinch
C Open field
(s)
1.8-
Rating of motor activity
3oo.3.
301.6-
20-
40 1.4.
20
10-
1.2. sal
med
sal
reed coc
1 sal
med
Fig. 2. (A) The average latencies of heat-induced tail flick responses 15 rain after the i.p. administration of saline (sal) or medetomidine (reed; 100/~g/kg). (B) The average latencies of biting responses induced by mechanical stimulation of the tail 15 min after the i.p. administration of saline (sal), medetomidine (med; 100 p g / k g ) , or cocaine (coc; 25 mg/kg). (C) Average ratings of locomotor activity in the open field test 15 min after the administration of saline (sal) or medetomidine (reed; 100 p g / k g ) ; n = 6 in each group. The error bars represent S.E. * * P < 0.01 (t-test, two-tailed; reference: saline-treated rats).
326
crease in pain scores within 9 min following medetomidine injection (fig. 1C). The effect of medetomidine was reversed 6 min following the injection of atipamezole (fig. IC). in comparison, cocaine (25 mg/kg), produced a maximal analgesic effect in the formalin test (fig. 1D), with no signs of sedation. The analgesic effect of cocaine was not reversed by atipamezole (1 mg/kg: fig. 1D). The latencies in the heat-induced tail flick test were not significantly increased 15 rain following the administration of medetomidine (100 ~ g / k g ) when compared with saline-treated controls (fig. 2A). In the test with mechanically induced pain, the latency to biting of the tail or the tail-clamp was not significantly increased 15 rain following the administration of medetomidine (100 ~ g / k g ) when compared with saline-treated controls (fig. 2B). However, 15 min following cocaine administration (25 mg/kg), the latency to tail biting was at least 10 times greater than that in the control session (fig. 2B). None of the cocaine-treated animals had bitten its tail or the clamp before the removal of the clamp (5 min cut-off time). Consistent with earlier observations, the cocainetreated animals in this study, rather than having motor impairments (Lin et al., |989), appeared to show, if anything, locomotor hyperactivity (Elliot et al., 1987). Medetomidine (100 p,g/kg) produced a significant reduction in scores in the open field test, a test of sedation/locomotor activity (fig. 2C), indicating a decrease in locomotor activity.
4. Discussion
The results of this study indicated that medetomidine can produce dose-dependent analgesia in the formalin test and that the maximal analgesia can be completely reversed by atipamezole, a highly selective a2-adrenoceptor antagonist. The mcdetomidine-induced analgesia in the formalin test is consistent with the previously reported effect of clonidine (Dennis et al., 1980; Skingle et ai., 1982). However, a recent study suggested that clonidine-induced analgesia in the formalin test is mediated by a~-adrenoceptors as revealed by its
reversal with prazosin but not yohimbine (Tasker and Melzack, 1989). Medetomidine did not produce significant analgesic effects in the tail flick and mechanically induced pain test in this study. This result differs from that obtained using clonidine in the heat-induced tail flick test (Fielding et al., 1981; Skingle et al., 1982: Tasker et al., 1989) and in mechanical pain tests (Fielding et al., 1981; Skingle et al.. 1982). Also dexmedetomidine has been reported to increase latencies to mechically induced nociceptive responses in halothane-anesthetized rats (Segal et al., 1988). et-Adrem~eptor-induced hypothermia as an artefactual cause of increased latencies to heat-induced tail flick responses (Berge et al., 1988) was not excluded in previous studies with clonidine and could provide an explanation for the difference in the effects of medetomidine and clonidine in the tail flick test. Differences in test conditions could also explain some of the variability in results (e.g. in the above cited dexmedetomidine study the rats were under halothane anesthesia). Finally, different compounds may activate partly different receptor populations. Medetomidine has been shown to produce dose-dependent sedation with the sedative effect significant at the lowest analgesic dose used in the current study (30 p,g/kg; MacDonald et al., 1988; Virtanen et al., 1989). The sedative/locomotordecreasing activity of medetomidine was also confirmed in the current study. Simple motor impairment cannot explain the analgesic effect in the formalin test since the rats were periodically climbing and running even after the maximally analgesic dose and since the biting response induced by mechanical stimulation of the tail was within normal limits. These findings indicate that the analgesic effect of medetomidine could be related to sedation but not to impairment of motor activity per se. Neurophysiological studies showed that a decrease in the state of arousal could suppress responses in the somatosensory neurons of the thalamus based on supramedullary mechanisms (Morrow and Casey, 1989). Accordingly, the medetomidine-induced analgesic effect seen in the current study could be due to supraspinal sedation-related mechanisms. This hypothesis is supported by the finding that the spinally medi-
327
ated tail flick response was not affected by medetomidine. However, as shown earlier with other drugs and tests (Dennis et al., 1980), the type of pain is also of importance since the mechanically induced biting response, which obviously involves supraspinal neuronal circuits, was not significantly influenced by medetomidine in the present study. The powerful analgesic effect of a non-sedative, monoaminergic agent, cocaine, in the formalin test (Lin et al., 1989) was confirmed in this study. Moreover, cocaine produced a very strong analgesic effect in the test with mechanically induced pain. Since atipamezole did not attenuate cocaine-induced analgesia, different mechanisms underlie the analgesic effects of cocaine and medetomidine. The results of a previous neurophysiological study indicate that the antinociceprive effects of systemic cocaine at the currently used dose (25 mg/kg) cannot be explained by a local anesthetic effect (Pertovaara et al., 1988). A recent behavioral study showed specific dopamine receptor antagonists to attenuate cocaine-induced analgesia, indicating an important role of dopamine in the analgesic effect of cocaine (Lin et al., 1989). In conclusion, the results of this study indicate that medetomidine produces dose-dependent analgesia in a chemically induced pain test involving highly organized behavior (formalin test) but not in a mechanically induced pain test involving supraspinally organized behavior nor in the spinally mediated heat-induced tail flick test. Furthermore, it is suggested that the medetomidineinduced analgesia in the formalin test is based on supraspinal mechanisms related to sedation and involving a2-adrenoceptors. However, a2-adrenoceptors are not involved in the non-sedative cocaine-induced analgesia.
Acknowledgements This study was supported by grants from the Paulo Foundation, Helsinki, Finland.
References Aceto, M.D. and L.S. Harris, 1981, Antinociceptive mechanisms and acute and chronic behavioral effects of clonidine, in: Psychopharmacology of Clonidine, eds. H. Lal and S. Fielding (Alan R. Liss, Inc., New York) p. 243. Berge, O.-G., I. Garcia-Cabrera and K. Hole, 1988, Response latencies in the tail-flick test depend on tail skin temperature, Neurosci. Lett. 86. 284. Chan, S.H.H. and Y.-Y. i,ai, 1982, Effects of aging on pain responses and analgesic efficacy of morphine and clonidine in rats, Exp. Neurol. 75, 112. Curtis, A.L. and J. Marwah, 1987, The nociceptive]aw-opening reflex: evidence for a-2 adrenoceptor involvement, Pharmacol. Biochem. Behav. 26, 437. Dennis, S.G., R. Melzack, S. G u t m a n and F. Boucher, 1980. Pain modulation by adrenergic agents and morphine as measured by three pain tests, Life Sci. 26, 1247. Dubuisson, D. and S.G. Dennis, 1977, The formalin test: a quantitative study of the analgesic effects of morhpine, meperidine, and brain stem stimulation in rats and cats, Pain 4, 161. Elliot, P.J., G.M. Rosen and C.B. Nemeroff, 1987, A comparison of cocaine and its metabolite norcocaine: effects on locomotor activity, Pharmacol. Biochem. Behav. 26, 573. Fielding, S., T.C. Spaulding and H. Lal, 1981. Antinociceptive actions of clonidine, in: Psychopharmacology of Clonidine, eds. H. Lal and S. Fielding (Alan R. Liss, Inc., New York) p. 225. Hilakivi, L.A., T. Taira and I. Hilakivi, 1988, Early postnatal deprivation of active sleep with desipramine or zimeldine impairs later behavioural reactivity to auditory stimuli in rats, Acta Physiol. Scand. 132, 191. Lin, Y., T.J. Morrow, J.A. Kiritsy-Roy. L. Cass Terry and K.L. Casey, 1989, Cocaine: evidence for supraspinal, dopaminemediated, non-opiate analgesia, Brain Res. 479, 306. MacDonald, E., H. Scheinin and M. Scheinin, 1988, Behavioural and neurochemical effects of medetomidine, a novel veterinary sedative, European J. Pharmacol. 158, 119. Morrow, T.J. and K.L. Casey, 1989, Modulation of the spontaneous and evoked discharges of ventral posterior thalamic neurons during shifts in arousal, Brain Res. Bull. 21,433. Paalzow, G., 1974, Analgesia produced by clonidine in mice and rats, J. Pharm. Pharmacol. 26, 361. Pertovaara, A.. C.R. Belczynski, T.J. Morrow and K.L. Casey, 1988, The effect of systemic cocaine on spinal nociceptive reflex activity in the rat, Brain Res. 438, 286. Scheinin, H., E. MacDonald and M. Scheinin, 1988. Behavioural and neurochemical effects of atipamezole, a novel a 2adrenoceptor antagonist, European J. Pharmacol. 157, 243. Segal. I.S.. R.G. Vickery, J.K. Walton, V.A. Doze and M. Maze. 1988, Dexmedetomidine diminishes halothane anesthetic requirements in rats through a postsynaptic a-2adrenergic receptor, Anesthesiology 69, 818. Skingle, M., A.G. Hayes and M.B. Tyers, 1982, Antinociceptive activity of chmidine in the mouse, rat and dog, Life Sci. 31, 1123.
328 Tasker, R.A.R. and R. Melzack, 1989, Different a-receptor subtypes are involved in clonidine-prtxluced analgesia in different pain tests, Life Sci. 44, 9. Virtanen, R., J.-M. Savola and V. Saano, 1989, Highly selective and specific antagonism of central and peripheral a-2adrenoceptors by atipamezole, Arch. Int. Pharmacodyn. Ther. 297, 190.
Virtanen, R., J.-M. Savola, V. Saano and L. Nyman, 1988, Characterization of the selectivity, specificity and potency of medetomidine as an a2-adrenoceptor agonist, European J. Pharmacol. 150, 9.
323
Elsevier UP 51276
The effect of medetomidine, an a2-adrenoceptor agonist, in various pain tests Antti Pertovaara, Timo Kauppila and Timo Tukeva Department of Physiology. Unit:ersity of Helsmki. Siltavuorenpenger 20 J. 00l 70 ttelsmki. Finland
Received 4 September 1989. revised MS received 16 November 1989. accepted 6 February 1990
Medetomidine, a new a2-adrenoceptor agonist produced dose-dependent (30-100 /tg/kg i.p.) analgesia in the formalin test in rats, and this effect was reversed by atipamezole (1 mg/kg), a new a2-adrenoceptor antagonist. However, medetomidine at the dose of 100 p.g/kg did not influence tail flick latencies or latencies of the biting response to mechanical pinch stimuli. Moreover, medetomidine produced sedation and a decrease in locomotor activity. In comparison, the non-sedative monoaminergic agent, cocaine (25 mg/kg), produced highly significant analgesic effects in the formalin and mechanical pain tests. The cocaine effect in the formalin test was not reversed by atipamezole (1 mg/kg). It is concluded that the analgesic effect of medetomidine in the formalin test is due to supraspinal mechanisms related to sedation and is mediated by ~2-adrenoceptors. The a2-adrenoceptors are not involved in cocaine-induced anagesia. Analgesia; Medetomidine; Atipamezole; Cocaine; Formalin test: Tail flick test; Mechanical pain: (Rat)
1. Introduction Several groups have reported that clonidine, an a2-adrenoceptor agonist, given systemically can produce analgesia (Paalzow, 1974; Dennis et al., 1980; Aceto and Harris, 1981; Fielding et al., 1981; Chan and Lai, 1982; Skingle et al., 1982; Curtis and Marwah, 1987). Medetomidine has been shown to be a highly selective and potent a2adrenoceptor agonist (Virtanen et al., 1988). In a recent study it was found that dexmedetomidine, a stereoisomeric form of medetomidine, prolonged the latency of the behavioral response to tail clamping under halothane anesthesia, and this analgesic effect of dexmedetomidine could be reversed by idazoxan, an a2-adrenoceptor antagonist (Segal et al., 1988). In the current study we wished to determine the effect of medetomidine in various
Correspondence to: A. Pertovaara, Department of Physiology, University of Helsinki, Siltavuorenpenger 20 J, 00170 Helsinki. Finland.
pain tests in awake rats. Since sedation can be a complicating factor in a2-adrenoceptor studies (Tasker and Melzack, 1989) we also evaluated the level of m o t o r activity following medetomidine and compared the analgesic effect of medetomidine with that produced by a non-sedative monoaminergic agent, cocaine (Lin et al., 1989). An attempt was made to reverse the analgesic effects of medetomidine and cocaine by using a novel a2-adrenoceptor antagonist, atipamezole, which has been shown to antagonize medetomidine-induced sedative, hypothermic and neurochemical changes (Scheinin et al., 1988; M a c D o n a l d et al., 1988; Virtanen et al., 1989).
2. Materials and methods Adult male Wistar rats, weight range 200-450 g, were obtained from the Finnish National Laboratory Animal Centre. They had free access to pelleted food chow and water and were on a 12-h
0014-2999/90/$03.50 '~ 1990 Elsevier Science Publishers B.V. (Biomedical Division)
324
light-dark cycle. Medetomidine and atipamezole (both synthetized by the Farmos Group Ltd., Turku, Finland), cocaine and 0.9% saline ( = control) were administered i.p. Analgesia was evaluated by the formalin test, heat-induced tail flick test and mechanically induced tail biting response. In the formalin test (Dubuisson and Dennis, 1977), 0.05 ml of 5% formalin was injected s.c. into the dorsum of the one forepaw. Pain behavior was graded continuously throughout 3-rain observation periods for 30 rain according to the proportion of the time the paw was held up and licked (grade 3), held fully elevated (grade 2), partially weight bearing (grade 1), or fully weight bearing (grade 0). In the heatinduced tail flick test, the rat was placed in a clear plexiglass cylinder and radiant heat was applied to the tail until the tail was moving as detected by a light-emitting sensor. The lacency to the first movement of the tail was used as a parameter of pain sensitivity. The heat source was adjusted to give an average baseline latency of about 1.5 s in untreated rats. A cut-off time of 5 s was used to avoid excessive pain. Since medetomidine is known to produce a decrease in skin temperature (MacDonald et al., 1988) which could produce an artefactual latency increase in the tail flick test (Berge et al., 1988), the tail was kept in a water bath at constant temperature (39 °C) throughout the session, except during actual testing when the tail was first dried then stimulated using radiant heat. This served to eliminate the artefactual increase of tail flick latency due to medetomidineinduced hypothermia. Two trials of the tail flick test were run at 2-min intervals. In the mechanically induced pain test, a hemostatic clamp was applied to the tail and the latency to biting of the tail or clamp was observed. A cut-off time of 5 min was used in the tail biting test. The sedative effect of medetomidine was evaluated using the open field test as described earlier (Hilakivi et al., 1988). Briefly, the open field was a circular arena (83 cm in diameter) covered with lacquered cork painted white. The walls of the arena, 50 cm high, were also white. The floor had three concentric black circles. These circles were subdivided by radiating black lines into 19 segments of approximately equal area. For
testing, each animal was placed on a starting-point in the innermost circle of the field and observed for 2 min. The number of marked floor segments which a rat entered with at least both its front paws, was recorded and used as a measure of locomotion/sedation. Each animal was used only once for each test condition and was killed after the session by an overdose of pentobarbitol. All drugs or drug combinations were given 15 min before the actual testing (in the formalin test, 15 min before the application of formalin) except when the time course of the drug effects was to be tested (see Results for details). The significance of difference in pain scores of the formalin test was evaluated by computing the 95% confidence intervals about each mean. Compared means were considered significantly different when one fell outside the confidence limit of the other. The differences in the tail flick or tail biting latencies and the open field scores were evaluated using a t-test (two-tailed. The P < 0.05 level was considered to represent a significant difference.
3. Results Medetomidine produced dose-dependent analgesia in the formalin test (fig. 1A). At a dose of 30 btg/kg the analgesic effect was significant only from 12 to 21 rain after formalin application. At higher doses the analgesic effect was significant throughout the observation period. At the maximally effective dose (100 g g / k g ) the rats appeared sedated and spent a considerable part of the test time lying with eyes closed. However, motor behavior did not appear to be impaired when they were awake (e.g. they could climb and run). Atipamezole (1 m g / k g ) completely reversed the analgesic effect of medetomidine (100 /.tg/kg) in the formalin test (fig. 1B). Atipamezole alone (1 m g / k g ) did not have any significant effect in the formalin test (fig. 1B). For studying the time course of drug effects, medetomidine (100 /.tg/kg) was applied 3 min after the formalin injection, and atipamezole (1 m g / k g ) 15 min after the formalin injection ( = 12 min after the administration of medetomidine). There was a highly significant de-
325
Pain intensity A
3-
rating *30 A55 pg/kg medet.
~saline
B 3"
• I00 T
~r
2-
~,
,J
j\
0,
i
r
:L
~atipam.
"r
r/L
y=
•
L
z-
T
~-i-i-i-i'
i
I~*.~m.._e d e t.
1; C :3-
0
~--t-~o-- e--e--o-- .--.
3;
2?
,; 3-
,+
• cocaine ~, cocaine+atipam.
T
,,
21-
3;
2?
D
medet.
~T
I ~a~i~pam.
I 0
O,
,~
2'0
0-0-0=0-0=0=0=0=0=0
3~
I~
Time
2~
3~
(min)
Fig. 1. (A) Effects of different i.p. doses of medetomidine (30-100 /xg/kg) on averaged pain intensity ratings in the formalin test. Formalin (5%, 0.05 ml) was injected s.c. into one forepaw at 0 time on the abscissa and the drugs were administered 15 min before 0 time; n = 6 in saline and 100/~g/kg medetomidine groups, n = 4 in other groups. (B) Effects of atipamezole alone (n = 4; l mg/kg), or of 100 # g / k g medetomidine alone (n = 6) or with 1 m g / k g of atipamezole (n = 4) in the formaline test. (C) The time course of the effect of medetomidine (100 itg/kg) and its reversal by atipamezole (1 m g / k g ) in the formalin test (n = 4). Formalin was applied at time 0, medetomidine 3 min later, and atipamezole 15 min following the formalin application. (D) Effects of 25 m g / k g cocaine alone (n = 4) or with 1 m g / k g of atipamezole (n = 4) in the formalin test. The vertical bars represent + S.E.
A Tail flick Latency
B Pinch
C Open field
(s)
1.8-
Rating of motor activity
3oo.3.
301.6-
20-
40 1.4.
20
10-
1.2. sal
med
sal
reed coc
1 sal
med
Fig. 2. (A) The average latencies of heat-induced tail flick responses 15 rain after the i.p. administration of saline (sal) or medetomidine (reed; 100/~g/kg). (B) The average latencies of biting responses induced by mechanical stimulation of the tail 15 min after the i.p. administration of saline (sal), medetomidine (med; 100 p g / k g ) , or cocaine (coc; 25 mg/kg). (C) Average ratings of locomotor activity in the open field test 15 min after the administration of saline (sal) or medetomidine (reed; 100 p g / k g ) ; n = 6 in each group. The error bars represent S.E. * * P < 0.01 (t-test, two-tailed; reference: saline-treated rats).
326
crease in pain scores within 9 min following medetomidine injection (fig. 1C). The effect of medetomidine was reversed 6 min following the injection of atipamezole (fig. IC). in comparison, cocaine (25 mg/kg), produced a maximal analgesic effect in the formalin test (fig. 1D), with no signs of sedation. The analgesic effect of cocaine was not reversed by atipamezole (1 mg/kg: fig. 1D). The latencies in the heat-induced tail flick test were not significantly increased 15 rain following the administration of medetomidine (100 ~ g / k g ) when compared with saline-treated controls (fig. 2A). In the test with mechanically induced pain, the latency to biting of the tail or the tail-clamp was not significantly increased 15 rain following the administration of medetomidine (100 ~ g / k g ) when compared with saline-treated controls (fig. 2B). However, 15 min following cocaine administration (25 mg/kg), the latency to tail biting was at least 10 times greater than that in the control session (fig. 2B). None of the cocaine-treated animals had bitten its tail or the clamp before the removal of the clamp (5 min cut-off time). Consistent with earlier observations, the cocainetreated animals in this study, rather than having motor impairments (Lin et al., |989), appeared to show, if anything, locomotor hyperactivity (Elliot et al., 1987). Medetomidine (100 p,g/kg) produced a significant reduction in scores in the open field test, a test of sedation/locomotor activity (fig. 2C), indicating a decrease in locomotor activity.
4. Discussion
The results of this study indicated that medetomidine can produce dose-dependent analgesia in the formalin test and that the maximal analgesia can be completely reversed by atipamezole, a highly selective a2-adrenoceptor antagonist. The mcdetomidine-induced analgesia in the formalin test is consistent with the previously reported effect of clonidine (Dennis et al., 1980; Skingle et ai., 1982). However, a recent study suggested that clonidine-induced analgesia in the formalin test is mediated by a~-adrenoceptors as revealed by its
reversal with prazosin but not yohimbine (Tasker and Melzack, 1989). Medetomidine did not produce significant analgesic effects in the tail flick and mechanically induced pain test in this study. This result differs from that obtained using clonidine in the heat-induced tail flick test (Fielding et al., 1981; Skingle et al., 1982: Tasker et al., 1989) and in mechanical pain tests (Fielding et al., 1981; Skingle et al.. 1982). Also dexmedetomidine has been reported to increase latencies to mechically induced nociceptive responses in halothane-anesthetized rats (Segal et al., 1988). et-Adrem~eptor-induced hypothermia as an artefactual cause of increased latencies to heat-induced tail flick responses (Berge et al., 1988) was not excluded in previous studies with clonidine and could provide an explanation for the difference in the effects of medetomidine and clonidine in the tail flick test. Differences in test conditions could also explain some of the variability in results (e.g. in the above cited dexmedetomidine study the rats were under halothane anesthesia). Finally, different compounds may activate partly different receptor populations. Medetomidine has been shown to produce dose-dependent sedation with the sedative effect significant at the lowest analgesic dose used in the current study (30 p,g/kg; MacDonald et al., 1988; Virtanen et al., 1989). The sedative/locomotordecreasing activity of medetomidine was also confirmed in the current study. Simple motor impairment cannot explain the analgesic effect in the formalin test since the rats were periodically climbing and running even after the maximally analgesic dose and since the biting response induced by mechanical stimulation of the tail was within normal limits. These findings indicate that the analgesic effect of medetomidine could be related to sedation but not to impairment of motor activity per se. Neurophysiological studies showed that a decrease in the state of arousal could suppress responses in the somatosensory neurons of the thalamus based on supramedullary mechanisms (Morrow and Casey, 1989). Accordingly, the medetomidine-induced analgesic effect seen in the current study could be due to supraspinal sedation-related mechanisms. This hypothesis is supported by the finding that the spinally medi-
327
ated tail flick response was not affected by medetomidine. However, as shown earlier with other drugs and tests (Dennis et al., 1980), the type of pain is also of importance since the mechanically induced biting response, which obviously involves supraspinal neuronal circuits, was not significantly influenced by medetomidine in the present study. The powerful analgesic effect of a non-sedative, monoaminergic agent, cocaine, in the formalin test (Lin et al., 1989) was confirmed in this study. Moreover, cocaine produced a very strong analgesic effect in the test with mechanically induced pain. Since atipamezole did not attenuate cocaine-induced analgesia, different mechanisms underlie the analgesic effects of cocaine and medetomidine. The results of a previous neurophysiological study indicate that the antinociceprive effects of systemic cocaine at the currently used dose (25 mg/kg) cannot be explained by a local anesthetic effect (Pertovaara et al., 1988). A recent behavioral study showed specific dopamine receptor antagonists to attenuate cocaine-induced analgesia, indicating an important role of dopamine in the analgesic effect of cocaine (Lin et al., 1989). In conclusion, the results of this study indicate that medetomidine produces dose-dependent analgesia in a chemically induced pain test involving highly organized behavior (formalin test) but not in a mechanically induced pain test involving supraspinally organized behavior nor in the spinally mediated heat-induced tail flick test. Furthermore, it is suggested that the medetomidineinduced analgesia in the formalin test is based on supraspinal mechanisms related to sedation and involving a2-adrenoceptors. However, a2-adrenoceptors are not involved in the non-sedative cocaine-induced analgesia.
Acknowledgements This study was supported by grants from the Paulo Foundation, Helsinki, Finland.
References Aceto, M.D. and L.S. Harris, 1981, Antinociceptive mechanisms and acute and chronic behavioral effects of clonidine, in: Psychopharmacology of Clonidine, eds. H. Lal and S. Fielding (Alan R. Liss, Inc., New York) p. 243. Berge, O.-G., I. Garcia-Cabrera and K. Hole, 1988, Response latencies in the tail-flick test depend on tail skin temperature, Neurosci. Lett. 86. 284. Chan, S.H.H. and Y.-Y. i,ai, 1982, Effects of aging on pain responses and analgesic efficacy of morphine and clonidine in rats, Exp. Neurol. 75, 112. Curtis, A.L. and J. Marwah, 1987, The nociceptive]aw-opening reflex: evidence for a-2 adrenoceptor involvement, Pharmacol. Biochem. Behav. 26, 437. Dennis, S.G., R. Melzack, S. G u t m a n and F. Boucher, 1980. Pain modulation by adrenergic agents and morphine as measured by three pain tests, Life Sci. 26, 1247. Dubuisson, D. and S.G. Dennis, 1977, The formalin test: a quantitative study of the analgesic effects of morhpine, meperidine, and brain stem stimulation in rats and cats, Pain 4, 161. Elliot, P.J., G.M. Rosen and C.B. Nemeroff, 1987, A comparison of cocaine and its metabolite norcocaine: effects on locomotor activity, Pharmacol. Biochem. Behav. 26, 573. Fielding, S., T.C. Spaulding and H. Lal, 1981. Antinociceptive actions of clonidine, in: Psychopharmacology of Clonidine, eds. H. Lal and S. Fielding (Alan R. Liss, Inc., New York) p. 225. Hilakivi, L.A., T. Taira and I. Hilakivi, 1988, Early postnatal deprivation of active sleep with desipramine or zimeldine impairs later behavioural reactivity to auditory stimuli in rats, Acta Physiol. Scand. 132, 191. Lin, Y., T.J. Morrow, J.A. Kiritsy-Roy. L. Cass Terry and K.L. Casey, 1989, Cocaine: evidence for supraspinal, dopaminemediated, non-opiate analgesia, Brain Res. 479, 306. MacDonald, E., H. Scheinin and M. Scheinin, 1988, Behavioural and neurochemical effects of medetomidine, a novel veterinary sedative, European J. Pharmacol. 158, 119. Morrow, T.J. and K.L. Casey, 1989, Modulation of the spontaneous and evoked discharges of ventral posterior thalamic neurons during shifts in arousal, Brain Res. Bull. 21,433. Paalzow, G., 1974, Analgesia produced by clonidine in mice and rats, J. Pharm. Pharmacol. 26, 361. Pertovaara, A.. C.R. Belczynski, T.J. Morrow and K.L. Casey, 1988, The effect of systemic cocaine on spinal nociceptive reflex activity in the rat, Brain Res. 438, 286. Scheinin, H., E. MacDonald and M. Scheinin, 1988. Behavioural and neurochemical effects of atipamezole, a novel a 2adrenoceptor antagonist, European J. Pharmacol. 157, 243. Segal. I.S.. R.G. Vickery, J.K. Walton, V.A. Doze and M. Maze. 1988, Dexmedetomidine diminishes halothane anesthetic requirements in rats through a postsynaptic a-2adrenergic receptor, Anesthesiology 69, 818. Skingle, M., A.G. Hayes and M.B. Tyers, 1982, Antinociceptive activity of chmidine in the mouse, rat and dog, Life Sci. 31, 1123.
328 Tasker, R.A.R. and R. Melzack, 1989, Different a-receptor subtypes are involved in clonidine-prtxluced analgesia in different pain tests, Life Sci. 44, 9. Virtanen, R., J.-M. Savola and V. Saano, 1989, Highly selective and specific antagonism of central and peripheral a-2adrenoceptors by atipamezole, Arch. Int. Pharmacodyn. Ther. 297, 190.
Virtanen, R., J.-M. Savola, V. Saano and L. Nyman, 1988, Characterization of the selectivity, specificity and potency of medetomidine as an a2-adrenoceptor agonist, European J. Pharmacol. 150, 9.