- Email: [email protected]
Attenuation of ethanol intoxication by alpha-2 adrenoceptor antagonists
Life Sciences, Vol. 44, pp. 111-119 Printed in the U.S.A.
Pergamon Press
ATTENUATION OF ETHANOL INTOXICATION BY ALPHA-2 ADRENOCEPTOR ANTAGONISTS Richard G. Lister, Michael J. Durcan, David J. Nutt and Markku Linnoila Laboratory of Clinical Studies, National Institute on Alcohol Abuse and Alcoholism, DICBR, Building I0 Room 3C218, 9000 Rockville Pike, Bethesda, Maryland 20892, USA. (Received in final form November 15, 1988)
The interaction of a highly potent and selective alpha-2 adrenoceptor antagonist, atlpamezole with ethanol was investigated in tests assessing a number of ethanol's behavioral effects. Atlpamezole antagonized ethanol's effects on directed exploration in a holeboard test, reduced observer-rated intoxication and also reduced the duration of loss of righting reflex caused by ethanol. Similar effects were produced by another alpha-2 adrenoceptor antagonist idazoxan. The magnitude of the effects was comparable to that produced in the same animal models by the imidazodiazepine Ro 15-4513, which antagonizes ethanol by an action at central benzodlazepine receptors. Whereas Ro 15-4513 possesses marked behavioral effects on its own, atlpamezole is comparatively inactive in all paradigms so far tested. The data suggest that alpha-2 adrenoceptors can play an important role in modulating the intoxicating effects of ethanol. The role of noradrenerglc mechanisms in the behavioral effects of ethanol has been unclear, in part due to the relative lack of specificity of the agents available for study. Atlpamezole is a recently developed alpha-2 adrenoceptor antagonist which is highly potent and with an alpha-2/alpha-i receptor selectivity ratio more than 100 fold greater than idazoxan (1,2), the drug currently considered to be the optimal alpha-2 adrenoceptor antagonist. The aim of the present study was to examine the effects of atipamezole on the behavioral effects of ethanol in four different behavioral paradigms: a holeboard test, which assesses an animal's directed exploration (head-dlppfng) independently of its locomotor activity (3); observer-rated tests of intoxication and loss of righting reflex, and a test sensitive to ethanol's anticonvulsant effect. Comparisons are made with Idazoxan, and with the effects of Ro 15-4513, a benzodiazepine receptor partial inverse agonlst that has been shown to reverse a number of ethanol's behavioral effects (4-6), but which possesses properties that make it unsuitable for use in human populations (7-9).
0024-3205/89 $3.00 + .00
112
Alpha-2 Adrenoceptors and Alcohol
Vol. 44, No. 2, 1989
Methods Drugs. Atipamezole (Farmos) and Idazoxan (Reckltt and Colman) were dissolved in distilled water in all experiments, except the one comparing their effects to that of Ro 15-4513. In that study they were dissolved in 0.3% Tween 20. Ro 15-4513 was suspended in 0.3% TWeen 20. These drugs were all given i.p. in a volume of i0 ml/kg. Animals. In all experiments NIH Swiss mice were used. They weighed approximately 23 g and were housed in groups of i0 and maintained on a 12/12 h light/dark cycle with ad lib access to food and water. Apparatus. The holeboard apparatus was made of Plexiglas (40 x 40 x 30 cm) and had four holes 3 cm in diameter equally spaced in the floor. Infra-red photocells in the walls of the box and directly beneath each hole provided automated measures of locomotor activity (number of beam interruptions) and of the number of head-dips made. Holeboard experiment The doses of atlpamezole used in the holeboard experiments were chosen on the basis of the drug's ability to antagonize the effects of the selective alpha-2 adrenoceptor agonist medetomldine in the same test (10). Atipamezole (0, 1 or 3 mg/kg) and ethanol (0 or 2 g/kg) were administered i.p. contralaterally 30 min before an 8 mln holeboard test. To examine the role of pharmacokinetic factors in the interaction of atipamezole with ethanol, separate groups of mice (n=10 per group) were killed by decapitation 30 mln after treatment with ethanol (2 g/k g) in combination with atipamezole (3 mg/kg) or its vehicle. Blood alcohol concentrations were determined using Sigma diagnostics procedure 332-UV. Intoxication stud~ In the test of intoxication, NIH Swiss mice were treated with ethanol (2.4 g/kg) and 5 mln later their intoxication was rated using a variation of the scale of MaJchrowlcz (ii). The following scores were assigned: 0=no observable effect, l=mild ataxia, 2:moderate ataxia 3:severe ataxia 4=very severe ataxia, barely able to recover righting reflex 5=loss of righting reflex. Immediately after the first rating, mice received atipamezole (0, 1 or 3 mg/kg) and were rated every 5 min thereafter for a further 20 min by an observer who was unaware of the drug treatment each animal had received. The timing between ethanol administration and testing was shorter in this study than in the holeboard experiment because of the rapid change in intoxication scores that takes place within 30 mln of ethanol treatment (see Fig 2). In a second experiment, the effect of atipamezole was compared with that of another alpha-2 antagonist (Idazoxan 1 mg/kg)) and with the benzodiazeplne receptor partial inverse agonist Ro 15-4513 (3 mg/kg), which attenuates ethanol intoxication (6). The dose of Ro 15-4513 chosen was one that has been found to produce a maximal attenuation of ethanol's behavioral effects in the NIH Swiss mice (12,13).
Vol. 44, No. 2, 1989
Alpha-2 Adrenoceptors and Alcohol
113
Loss of riahting reflex • he effect'of pretreating mice with atipamezole or idazoxan on the duration of loss of righting reflex caused by a high dose of ethanol (3.5 g/kg) was also examined. Mice received atipamezole or idazoxan i0 min before ethanol. A mouse was taken to have regained its righting reflex when it was able to right itself on three occasions within a 60 s period. Seizure threshold In the final experiment the effects of atipamezole and idazoxan on the anticonvulsant effect of ethanol were examined. Separate groups of mice received atipamazole or idazoxan 5 min after treatment with ethanol (0 or 2.4 g/kg) to parallel the treatment schedule used in the intoxication studies. Fifteen minutes after the second injection seizure threshold to bicuculline was determined by infusion through the tail vein (12). Seizure threshold was calculated from the latency to the onset of repeated myoclonic Jerking of head and forelimbs. Results Holeboard exp~rimen~ There was a significant ethanol x atipamezole interaction in the analysis of the number of head-dips (F(2,46)=4.9, P<0.05). Fig 1 shows that ethanol (2 g/kg) reduced exploratory head-dipplng (p<0.01). Atipamezole, which alone failed to alter exploration, reversed the decrease in exploration caused by ethanol (p<0.01).
[] ATIPAMEZOLE+0G/KGETHANOL • ATIPAMEZOLE+2G/KG ETHANOL ~OF HEAD-DIPS
MOTORACTIVITY •,
80 60
""
45 3
J
## 5
##
##
~
25O
4O
150
20
5O 0 mg~g mg~g 3 mg~g A~PAIVIEZOLE
0 rng~g 1 rng~g 3 mg~g ATIPAIVEZOLE
Figure i The effect of ethanol (0 or 2 g/kg) in combination with atipamezole (0, 1 or 3 mg/kg) on the exploratory headdipping (left) and the locomotor activities (right) of NIH Swiss mice in an 8 min holeboard test. Ethanol and atipamezole were both administered i.p. 30 min before testing. Scores are means ± SEM, n=8-10 per group. ## significant effect of ethanol p(0.01, ** significant reversal of the effect of ethanol p<0.01 (Tukey's test).
114
~pha~
Adrenoceptors and Alcohol
A)
r~
VEH
[]
Vol. 44, No. 2, 1989
AT (I MG/KG)
I
AT (3MG/KG)
o z
3
- -
2
-
**
**
X
o
I0 15 20 TIME AFTER ETHANOL (MIN)
5
r~ VEH
[] AT (I MG/KG)
[ ] ID (I MG/KG)
I
B) 4
RO (3 MG/KG)
- -
0
2X
o
I 5
I0 15 TIME AFTER ETHANOL(MIN)
20
Figure 2 The effect of (A) atipamezole (AT 0, 1 or 3 mg/kg) and of (B) atlpamezole (AT 1 mglkg), idazoxan (ID 1 mg/kg) and Ro 15-4513 (R0 3 mg/kg) or a waterlTween vehicle (VEH) on the intoxicating effect of ethanol assessed using a modification of an observer rating of intoxication (ii). Mice were injected with ethanol (2.4 g/kg) and individually rated every 5 mln for 20 mln by an observer ignorant of the treatment each animal had received. Drug treatment was given immediately after the first (5 mln) rating. Values are means ~ SEM, n=16-20 per group. Significantly different from vehlcle-treated mice *p<0.05, ** p<0.01, Dunnett's test.
Vol. 44, No. 2, 1989
~pha-2 Adrenoceptors ~ d Alcohol
115
T~le 1 The duration of loss of righting reflex of different groups of NIH Swiss mice following treatment with 3.5 g/kg ethanol. Mice were treated ~ t h a) atipamezole (0, 1.0 or 3.0 mg/kg) or b) atipamezole (0.3 or 1.0 mg/kg), idazoxan (0.3 or 1.0 mg/kg) or the vehicle 10 min before receiving the e t ~ n o l . Values are means ~ SEM.
a)
b)
n
~ r a t i o n of loss of righting reflex (min)
Vehicle Atipamezole Atlpamezole
(i.0 mg/kg) (3.0 mglkg)
22 20 23
38.6 ± 2.4 25.7 ± 1.8"* 26.9 ± 2.3**
Vehicle Atip~ezole Atipamezole Idazoxan Idazoxan
(0.3 (1.0 (0.3 (i.0
12 ii 12 12 10
39.3 32.5 28.5 24.3 29.8
mg/kg) mg/kg) mg/kg) mg/kg)
± ± ± ± ±
2.6 3.7 2.7* 2.2** 2.6*
s i ~ i f i c a n t l y different from vehlcle-treated a n i ~ I s , ** p<0.01 (Dunnett's test)
1.2-
•
*p<0.05,
E T H A ~ L (2.4 G/KG)
1.o, o
0.8"
0.4" W
"2 2 0.2 0.0
VEHICLE
1.0 3.0 ATIPAMEZOLE (MG/KG)
9 1.0 IDAZOXAN (MG/KG)
Figure 3 The effect of ethanol (0 or 2.4 g/kg) in c o ~ i n a t i o n with atipamezole (0, 1 or 3 mg/kg) or idazoxan (I mg/kg) on seizure threshold to bicuculline (0.05 mg/ml in pH 3 saline) infused through the tall vein. E t ~ n o l was administered 20 min before test, and atipamezole or idazoxan 15 min before test. Values are m e ~ s ~ SEM, n=9-12 per group.
116
Alpha-2 Adrenoceptors and Alcohol
Vol. 44, No. 2, 1989
Ethanol increased locomotor activity (F(1,46)=71.2, p<0.001), but atlpamezole dld not alter this effect. Atlpamezole did not alter blood ethanol concentration 30 min after treatment (BACs of mlce receiving ethanol + vehicle were 199 t 15 mg/100ml (means ± SEM) and of mlce receiving ethanol in combination wlth atlpamezole were 213 ± 12 mg/100ml). The antagonism of ethanol's effect on exploration was, therefore, pharmacodynamlc rather than pharmacoklnetlc. In Fig 2 It can be seen that although the groups dld not differ In their sensitivity to ethanol prior to drug treatment, atlpamezole (i and 3 mg/kg) reduced the intoxicating effects of ethanol. Fig 2 also shows that in experiment 2 atlpamezole, idazoxan and Ro 15-4513 all reduced the intoxicating effects of ethanol, producing approximately equivalent effects. Lo~s Qf ~i~htlnqre~f~]~ P~hile neither alpha-2 adrenoceptor antagonist prevented the loss of righting reflex caused by the ethanol, both compounds significantly reduced the period during which the righting reflex was absent (see Table I). Seizure thres~gld Ethanol exerted a significant antlconvulsant action, raising seizure threshold to bicuculline (F(I,75)=581, p(0.0001). Neither atlpamezole nor idazoxan alone showed any indication of lowering seizure threshold to blcuculline, and neither drug reduced the antlconvulsant action of ethanol, see Flg 3. Shortly after these experiments were completed a report appeared in which several alpha-2 adrenoceptor antagonists were proconvulsant against blcuculline-lnduced seizures (26). He therefore examined the effects of stlll higher doses of atipamezole and Idazoxan on the seizure threshold to bicuculllne in the present paradigm. Three groups of mlce received atlpamezole (10 mg/kg), idazoxan (i0 mg/kg) or the vehicle 15 mln prior to determining seizure threshold to blcuculline. As can be seen in Table 2, this hlgh dose of atipamezole caused a very slight but significant reduction in seizure threshold to blcuculline (p<0.05). The reduction in seizure threshold caused by idazoxan failed to reach significance. Table 2 The seizure threshold to blcuculllne (mg//kg) of NIH Swiss mice 15 min after treatment with atipamezole, idazoxan or the distilled water vehicle. Values are means t SEM. N Seizure Threshold Vehicle 8 0.448 f 0.013 Atipamezole (I0 mg/kg) 8 0.401 ± 0.012, Idazoxan (I0 mg/kg) I0 0.415 i 0.013 ~slgnlflcantly different from vehlcle-treated mice, p<0.05, Dunnett
Vol. 44, No. 2, 1989
Alpha-2 Adrenoceptors and Alcohol
117
It is interesting to compare the effects of atlpamezole with those of the benzodlazeplne receptor partial inverse agonlst Ro 15-4513 whlch has been the focus of much recent alcohol research. Both atipamezole and Ro 15-4513 reduce the effects of ethanol on directed exploratlon in a holeboard test (13,14), although Ro 15-%513, in contrast to atlpamezole, reduces exploration on its own. Both drugs can also reduce the intoxicating effects of ethanol as assessed by observer ratings (6), and shorten the duration of loss of righting reflex caused by ethanol (15). However, neither Ro 15-4513 nor atIpamezole is capable of preventing the loss of righting reflex caused by high doses of ethanol. Further, neither compound has much effect on the locomotor stimulant action of ethanol (10,16). Clearly neither compound is capable of 'normallzlng' the behavior of ethanol-treated animals, and the cllnlcal significance of the findings is as yet unclear. The reduction of ethanol's effects in the holeboard test did not appear to have a pharmacoklnetlc basis since blood ethanol concentrations at the time of testing were not altered by the drug. The posslbillty that blood ethanol concentrations were altered by atlpamezole in the experiments in which the drug was given before or after ethanol seems unlikely but cannot be excluded by the present data. Despite the parallels between the effects of atipamezole and those of Ro 15-4513, the receptor sites mediating the actions of each are quite different. Ro 15-4513's effects are mediated by central benzodiazepine receptors since they are reversed by benzodiazeplne receptor antagonists (6,14). Atlpamezole In contrast does not bind to benzodlazeplne receptors (i), but is a selective and potent alpha-2 adrenoceptor antagonist. Its ethanol-antagonlzing effects are not reversed by benzodlazepine receptor antagonists (unpublished observations). A common neuroanatomical site, however, might underly the similarities between Ro 15-4513 and atlpamezole. One posslbillty is the locus coeruleus. It has been suggested that drugs acting at alpha-2 adrenoceptors may modulate sleep and arousal mechanisms by an action in the locus coeruleus (17). There is a high density of alpha-2 receptors in this area. They are thought to mediate auto-inhlbltlon of locus coeruleus neurc-'s (18,19). Further, benzodiazepine receptor ligands have be, shown to modulate the firing of locus coeruleus neurons (20). Since alpha-2 adrenoceptor antagonists increase noradrenaline release by reducing autolnhlbltion, drugs that act as agonlsts at postsynaptic adrenoceptors might be expected to interact with ethanol in a manner similar to the alpha-2 antagonists. Indeed, recent studies have found the llpophillc alpha-i adrenoceptor agonist St 587 to be capable of antagonizing some of the effects of ethanol in one inbred strain of mouse (21,22). However It should be noted that thls drug is not completely selectlve for alpha-1 receptors and also appears to act at alpha-2 adrenoceptors (23), and its ethanol-antagonlzlng effects were not observed in some strains of mice. Previous studies have implicated noradrenergic neurones in the depressant effects of barbiturates and other anaesthetics (24).
118
Alpha-2 Adrenoceptors and Alcohol
Vol. 44, No. 2, 1989
The present report suggests a clear role for alpha-2 adrenoceptors in the intoxicating effects of ethanol. The observation that atipamezole reduces ethanol's ataxic but not its motor stimulant and anticonvulsant effects, adds further support to the notion that ethanol's different behavioral effects are mediated by different neuroblologlcal mechanisms. In the present study neither atlpamezole nor idazoxan showed any indication of proconvulsant activity in NIH Swiss mice except at doses considerably higher than those required to reduce ethanol intoxication (i.e. i0 mg/kg). It is questionable whether this proconvulsant action is mediated by alpha-2 receptors. It should be noted that at these high doses there was a marked change in the appearance of the mice. They appeared large and fluffy due to piloerection. It is unlikely that an inability to detect a proconvulsant action at the lower doses reflects a low sensitivity of the test, since the identical paradigm has proven extremely sensitive to the proconvulsant action of other drugs (27). Alpha-2 adrenoceptor antagonists may exert more potent proconvulsant effects in other strains of mice (26). The clinical utility of alcohol antagonists is questionable (25). A drug that reduces the effects of ethanol without altering blood ethanol concentrations (as both Ro 15-4513 and alpha-2 adrenoceptor antagonists do) poses a number of ethical problems. While the drugs may reduce the acute intoxicating effects of ethanol, it would appear unlikely that they would reduce the long-term toxicity of ethanol. Further, any clinical potential for Ro 15-4513 is compromised both by its short duration of action (8) and its marked proconvulsant effects (6-8). Atlpamezole and idazoxan have a longer duration of action (2) and a much less clear effect on seizure threshold. Their longer duration of action may make these two drugs more suitable for use in humans, but their potential proconvulsant properties will need further clarification. Acknowledgements We are grateful to Farmos for the gift of atlpamezole, to Reckltt and Colman for the gift of idazoxan, and to Hoffmann-La Roche for the gift of Ro 15-4513. References I. R. VIRTANEN, J.-M. SAVOLA and V. SAAN0, Br. J. Pharmacol., in press. 2. H. SCHEININ, E. MACDONALD and M. SCHEININ, Eur. J. Pharmacol., 151 35-42 (1988). 3. S.E. FILE and A.G. WARDILL, Psychopharmacology 44 53-59 (1975). 4. E.P. BONETTI, W.P. BURKARD, M. GABL and H. MOHLER, Br. J. Pharmacol. 86 463P (1985). 5. P. POLC, Br. J. Pharmacol. 86 465P (1985). 6. P.D. SUZDAK, J.R. G L 0 ~ , J.N. CRAk~E¥, R.D. SCHWARTZ, P. SKOLNICK and S.M. PAUL, Science 234 1243-1247 (1986). 7. R.G. LISTER and J. KARANIAN, Alcohol 4 409-411 (1987). 8. K. MICZEK and E.M. ~k~RTS, Science 235 1127 (1987). 9. R.G. LISTER and D.J. NUTT, Br. J. Pharmacol. 93 210-214 (1988). 10. M.J. DURCAN, R.G. LISTER and M.LINNOILA, Neuropharmacology, in press.
Vol. 44, No. 2, 1989
ii. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27.
Alpha-2 Adrenoceptors and Alcohol
119
E. MAJCHROHICZ, Psychopharmacologia 43 245-254 (1975). D.J. ~ and R.G. LISTER, Brain Res. 413, 193-196 (1987). R.G. LISTER, Life Sci. 41 1481-1489 (1987). R.G. LISTER, Pharmacol. Biochem. Behav. 28 75-79 (1987). P. SYAPIN, K.W. GEE and R.L. ALKANA, Brain Res. Bull. 19 603-605 (1987). R.G. LISTER, Pharmacol. Biochem. Behav., in press. G.B. DE SARR0, C. ASCIOTI, F. FROIO, V. LIBRI and G. NISTIC0, Br. J. Pharmacol. 90 675-685 (1987). S.L. F00TE, F.E. BLOOM and G. ASTON-JONES, Physiol. Rev. 63 844-914 (1983). J.R. UNNERSTALL, T.A. KOPAJTIC and M.J. KUHAR, Brain Res. Rev. 7 69-101 (1984). K. RASMUBBEN and B.L. JAC0BS, Brain Res. 371 335-344 (1986). M.K. MENON and C.K. KODAMA, Life Sci 37 2091-2098 (1985). M.K. MENON and C.K. KODAMA, Life Sci 41 1599-1610 (1987). L. PICHLER and W. KOBINGER, Arznel. Forsch. 35 201-205 (1985). S.T. MASON and A. ANGEL, Eur. J. Pharmacol. 91 29-39 (1983). R.G. LISTER and D.J. NUTT, Trends in Neurosci. i0, 223-225 (1987). A. FLETCHER and E.A. FORSTER, Eur. J. Pharmacol. 151 27-34 (1988). D.J. NUTT and R.G. LISTER, Pharmacol. Biochem. Behav., in press.
Pergamon Press
ATTENUATION OF ETHANOL INTOXICATION BY ALPHA-2 ADRENOCEPTOR ANTAGONISTS Richard G. Lister, Michael J. Durcan, David J. Nutt and Markku Linnoila Laboratory of Clinical Studies, National Institute on Alcohol Abuse and Alcoholism, DICBR, Building I0 Room 3C218, 9000 Rockville Pike, Bethesda, Maryland 20892, USA. (Received in final form November 15, 1988)
The interaction of a highly potent and selective alpha-2 adrenoceptor antagonist, atlpamezole with ethanol was investigated in tests assessing a number of ethanol's behavioral effects. Atlpamezole antagonized ethanol's effects on directed exploration in a holeboard test, reduced observer-rated intoxication and also reduced the duration of loss of righting reflex caused by ethanol. Similar effects were produced by another alpha-2 adrenoceptor antagonist idazoxan. The magnitude of the effects was comparable to that produced in the same animal models by the imidazodiazepine Ro 15-4513, which antagonizes ethanol by an action at central benzodlazepine receptors. Whereas Ro 15-4513 possesses marked behavioral effects on its own, atlpamezole is comparatively inactive in all paradigms so far tested. The data suggest that alpha-2 adrenoceptors can play an important role in modulating the intoxicating effects of ethanol. The role of noradrenerglc mechanisms in the behavioral effects of ethanol has been unclear, in part due to the relative lack of specificity of the agents available for study. Atlpamezole is a recently developed alpha-2 adrenoceptor antagonist which is highly potent and with an alpha-2/alpha-i receptor selectivity ratio more than 100 fold greater than idazoxan (1,2), the drug currently considered to be the optimal alpha-2 adrenoceptor antagonist. The aim of the present study was to examine the effects of atipamezole on the behavioral effects of ethanol in four different behavioral paradigms: a holeboard test, which assesses an animal's directed exploration (head-dlppfng) independently of its locomotor activity (3); observer-rated tests of intoxication and loss of righting reflex, and a test sensitive to ethanol's anticonvulsant effect. Comparisons are made with Idazoxan, and with the effects of Ro 15-4513, a benzodiazepine receptor partial inverse agonlst that has been shown to reverse a number of ethanol's behavioral effects (4-6), but which possesses properties that make it unsuitable for use in human populations (7-9).
0024-3205/89 $3.00 + .00
112
Alpha-2 Adrenoceptors and Alcohol
Vol. 44, No. 2, 1989
Methods Drugs. Atipamezole (Farmos) and Idazoxan (Reckltt and Colman) were dissolved in distilled water in all experiments, except the one comparing their effects to that of Ro 15-4513. In that study they were dissolved in 0.3% Tween 20. Ro 15-4513 was suspended in 0.3% TWeen 20. These drugs were all given i.p. in a volume of i0 ml/kg. Animals. In all experiments NIH Swiss mice were used. They weighed approximately 23 g and were housed in groups of i0 and maintained on a 12/12 h light/dark cycle with ad lib access to food and water. Apparatus. The holeboard apparatus was made of Plexiglas (40 x 40 x 30 cm) and had four holes 3 cm in diameter equally spaced in the floor. Infra-red photocells in the walls of the box and directly beneath each hole provided automated measures of locomotor activity (number of beam interruptions) and of the number of head-dips made. Holeboard experiment The doses of atlpamezole used in the holeboard experiments were chosen on the basis of the drug's ability to antagonize the effects of the selective alpha-2 adrenoceptor agonist medetomldine in the same test (10). Atipamezole (0, 1 or 3 mg/kg) and ethanol (0 or 2 g/kg) were administered i.p. contralaterally 30 min before an 8 mln holeboard test. To examine the role of pharmacokinetic factors in the interaction of atipamezole with ethanol, separate groups of mice (n=10 per group) were killed by decapitation 30 mln after treatment with ethanol (2 g/k g) in combination with atipamezole (3 mg/kg) or its vehicle. Blood alcohol concentrations were determined using Sigma diagnostics procedure 332-UV. Intoxication stud~ In the test of intoxication, NIH Swiss mice were treated with ethanol (2.4 g/kg) and 5 mln later their intoxication was rated using a variation of the scale of MaJchrowlcz (ii). The following scores were assigned: 0=no observable effect, l=mild ataxia, 2:moderate ataxia 3:severe ataxia 4=very severe ataxia, barely able to recover righting reflex 5=loss of righting reflex. Immediately after the first rating, mice received atipamezole (0, 1 or 3 mg/kg) and were rated every 5 min thereafter for a further 20 min by an observer who was unaware of the drug treatment each animal had received. The timing between ethanol administration and testing was shorter in this study than in the holeboard experiment because of the rapid change in intoxication scores that takes place within 30 mln of ethanol treatment (see Fig 2). In a second experiment, the effect of atipamezole was compared with that of another alpha-2 antagonist (Idazoxan 1 mg/kg)) and with the benzodiazeplne receptor partial inverse agonist Ro 15-4513 (3 mg/kg), which attenuates ethanol intoxication (6). The dose of Ro 15-4513 chosen was one that has been found to produce a maximal attenuation of ethanol's behavioral effects in the NIH Swiss mice (12,13).
Vol. 44, No. 2, 1989
Alpha-2 Adrenoceptors and Alcohol
113
Loss of riahting reflex • he effect'of pretreating mice with atipamezole or idazoxan on the duration of loss of righting reflex caused by a high dose of ethanol (3.5 g/kg) was also examined. Mice received atipamezole or idazoxan i0 min before ethanol. A mouse was taken to have regained its righting reflex when it was able to right itself on three occasions within a 60 s period. Seizure threshold In the final experiment the effects of atipamezole and idazoxan on the anticonvulsant effect of ethanol were examined. Separate groups of mice received atipamazole or idazoxan 5 min after treatment with ethanol (0 or 2.4 g/kg) to parallel the treatment schedule used in the intoxication studies. Fifteen minutes after the second injection seizure threshold to bicuculline was determined by infusion through the tail vein (12). Seizure threshold was calculated from the latency to the onset of repeated myoclonic Jerking of head and forelimbs. Results Holeboard exp~rimen~ There was a significant ethanol x atipamezole interaction in the analysis of the number of head-dips (F(2,46)=4.9, P<0.05). Fig 1 shows that ethanol (2 g/kg) reduced exploratory head-dipplng (p<0.01). Atipamezole, which alone failed to alter exploration, reversed the decrease in exploration caused by ethanol (p<0.01).
[] ATIPAMEZOLE+0G/KGETHANOL • ATIPAMEZOLE+2G/KG ETHANOL ~OF HEAD-DIPS
MOTORACTIVITY •,
80 60
""
45 3
J
## 5
##
##
~
25O
4O
150
20
5O 0 mg~g mg~g 3 mg~g A~PAIVIEZOLE
0 rng~g 1 rng~g 3 mg~g ATIPAIVEZOLE
Figure i The effect of ethanol (0 or 2 g/kg) in combination with atipamezole (0, 1 or 3 mg/kg) on the exploratory headdipping (left) and the locomotor activities (right) of NIH Swiss mice in an 8 min holeboard test. Ethanol and atipamezole were both administered i.p. 30 min before testing. Scores are means ± SEM, n=8-10 per group. ## significant effect of ethanol p(0.01, ** significant reversal of the effect of ethanol p<0.01 (Tukey's test).
114
~pha~
Adrenoceptors and Alcohol
A)
r~
VEH
[]
Vol. 44, No. 2, 1989
AT (I MG/KG)
I
AT (3MG/KG)
o z
3
- -
2
-
**
**
X
o
I0 15 20 TIME AFTER ETHANOL (MIN)
5
r~ VEH
[] AT (I MG/KG)
[ ] ID (I MG/KG)
I
B) 4
RO (3 MG/KG)
- -
0
2X
o
I 5
I0 15 TIME AFTER ETHANOL(MIN)
20
Figure 2 The effect of (A) atipamezole (AT 0, 1 or 3 mg/kg) and of (B) atlpamezole (AT 1 mglkg), idazoxan (ID 1 mg/kg) and Ro 15-4513 (R0 3 mg/kg) or a waterlTween vehicle (VEH) on the intoxicating effect of ethanol assessed using a modification of an observer rating of intoxication (ii). Mice were injected with ethanol (2.4 g/kg) and individually rated every 5 mln for 20 mln by an observer ignorant of the treatment each animal had received. Drug treatment was given immediately after the first (5 mln) rating. Values are means ~ SEM, n=16-20 per group. Significantly different from vehlcle-treated mice *p<0.05, ** p<0.01, Dunnett's test.
Vol. 44, No. 2, 1989
~pha-2 Adrenoceptors ~ d Alcohol
115
T~le 1 The duration of loss of righting reflex of different groups of NIH Swiss mice following treatment with 3.5 g/kg ethanol. Mice were treated ~ t h a) atipamezole (0, 1.0 or 3.0 mg/kg) or b) atipamezole (0.3 or 1.0 mg/kg), idazoxan (0.3 or 1.0 mg/kg) or the vehicle 10 min before receiving the e t ~ n o l . Values are means ~ SEM.
a)
b)
n
~ r a t i o n of loss of righting reflex (min)
Vehicle Atipamezole Atlpamezole
(i.0 mg/kg) (3.0 mglkg)
22 20 23
38.6 ± 2.4 25.7 ± 1.8"* 26.9 ± 2.3**
Vehicle Atip~ezole Atipamezole Idazoxan Idazoxan
(0.3 (1.0 (0.3 (i.0
12 ii 12 12 10
39.3 32.5 28.5 24.3 29.8
mg/kg) mg/kg) mg/kg) mg/kg)
± ± ± ± ±
2.6 3.7 2.7* 2.2** 2.6*
s i ~ i f i c a n t l y different from vehlcle-treated a n i ~ I s , ** p<0.01 (Dunnett's test)
1.2-
•
*p<0.05,
E T H A ~ L (2.4 G/KG)
1.o, o
0.8"
0.4" W
"2 2 0.2 0.0
VEHICLE
1.0 3.0 ATIPAMEZOLE (MG/KG)
9 1.0 IDAZOXAN (MG/KG)
Figure 3 The effect of ethanol (0 or 2.4 g/kg) in c o ~ i n a t i o n with atipamezole (0, 1 or 3 mg/kg) or idazoxan (I mg/kg) on seizure threshold to bicuculline (0.05 mg/ml in pH 3 saline) infused through the tall vein. E t ~ n o l was administered 20 min before test, and atipamezole or idazoxan 15 min before test. Values are m e ~ s ~ SEM, n=9-12 per group.
116
Alpha-2 Adrenoceptors and Alcohol
Vol. 44, No. 2, 1989
Ethanol increased locomotor activity (F(1,46)=71.2, p<0.001), but atlpamezole dld not alter this effect. Atlpamezole did not alter blood ethanol concentration 30 min after treatment (BACs of mlce receiving ethanol + vehicle were 199 t 15 mg/100ml (means ± SEM) and of mlce receiving ethanol in combination wlth atlpamezole were 213 ± 12 mg/100ml). The antagonism of ethanol's effect on exploration was, therefore, pharmacodynamlc rather than pharmacoklnetlc. In Fig 2 It can be seen that although the groups dld not differ In their sensitivity to ethanol prior to drug treatment, atlpamezole (i and 3 mg/kg) reduced the intoxicating effects of ethanol. Fig 2 also shows that in experiment 2 atlpamezole, idazoxan and Ro 15-4513 all reduced the intoxicating effects of ethanol, producing approximately equivalent effects. Lo~s Qf ~i~htlnqre~f~]~ P~hile neither alpha-2 adrenoceptor antagonist prevented the loss of righting reflex caused by the ethanol, both compounds significantly reduced the period during which the righting reflex was absent (see Table I). Seizure thres~gld Ethanol exerted a significant antlconvulsant action, raising seizure threshold to bicuculline (F(I,75)=581, p(0.0001). Neither atlpamezole nor idazoxan alone showed any indication of lowering seizure threshold to blcuculline, and neither drug reduced the antlconvulsant action of ethanol, see Flg 3. Shortly after these experiments were completed a report appeared in which several alpha-2 adrenoceptor antagonists were proconvulsant against blcuculline-lnduced seizures (26). He therefore examined the effects of stlll higher doses of atipamezole and Idazoxan on the seizure threshold to bicuculllne in the present paradigm. Three groups of mlce received atlpamezole (10 mg/kg), idazoxan (i0 mg/kg) or the vehicle 15 mln prior to determining seizure threshold to blcuculline. As can be seen in Table 2, this hlgh dose of atipamezole caused a very slight but significant reduction in seizure threshold to blcuculline (p<0.05). The reduction in seizure threshold caused by idazoxan failed to reach significance. Table 2 The seizure threshold to blcuculllne (mg//kg) of NIH Swiss mice 15 min after treatment with atipamezole, idazoxan or the distilled water vehicle. Values are means t SEM. N Seizure Threshold Vehicle 8 0.448 f 0.013 Atipamezole (I0 mg/kg) 8 0.401 ± 0.012, Idazoxan (I0 mg/kg) I0 0.415 i 0.013 ~slgnlflcantly different from vehlcle-treated mice, p<0.05, Dunnett
Vol. 44, No. 2, 1989
Alpha-2 Adrenoceptors and Alcohol
117
It is interesting to compare the effects of atlpamezole with those of the benzodlazeplne receptor partial inverse agonlst Ro 15-4513 whlch has been the focus of much recent alcohol research. Both atipamezole and Ro 15-4513 reduce the effects of ethanol on directed exploratlon in a holeboard test (13,14), although Ro 15-%513, in contrast to atlpamezole, reduces exploration on its own. Both drugs can also reduce the intoxicating effects of ethanol as assessed by observer ratings (6), and shorten the duration of loss of righting reflex caused by ethanol (15). However, neither Ro 15-4513 nor atIpamezole is capable of preventing the loss of righting reflex caused by high doses of ethanol. Further, neither compound has much effect on the locomotor stimulant action of ethanol (10,16). Clearly neither compound is capable of 'normallzlng' the behavior of ethanol-treated animals, and the cllnlcal significance of the findings is as yet unclear. The reduction of ethanol's effects in the holeboard test did not appear to have a pharmacoklnetlc basis since blood ethanol concentrations at the time of testing were not altered by the drug. The posslbillty that blood ethanol concentrations were altered by atlpamezole in the experiments in which the drug was given before or after ethanol seems unlikely but cannot be excluded by the present data. Despite the parallels between the effects of atipamezole and those of Ro 15-4513, the receptor sites mediating the actions of each are quite different. Ro 15-4513's effects are mediated by central benzodiazepine receptors since they are reversed by benzodiazeplne receptor antagonists (6,14). Atlpamezole In contrast does not bind to benzodlazeplne receptors (i), but is a selective and potent alpha-2 adrenoceptor antagonist. Its ethanol-antagonlzing effects are not reversed by benzodlazepine receptor antagonists (unpublished observations). A common neuroanatomical site, however, might underly the similarities between Ro 15-4513 and atlpamezole. One posslbillty is the locus coeruleus. It has been suggested that drugs acting at alpha-2 adrenoceptors may modulate sleep and arousal mechanisms by an action in the locus coeruleus (17). There is a high density of alpha-2 receptors in this area. They are thought to mediate auto-inhlbltlon of locus coeruleus neurc-'s (18,19). Further, benzodiazepine receptor ligands have be, shown to modulate the firing of locus coeruleus neurons (20). Since alpha-2 adrenoceptor antagonists increase noradrenaline release by reducing autolnhlbltion, drugs that act as agonlsts at postsynaptic adrenoceptors might be expected to interact with ethanol in a manner similar to the alpha-2 antagonists. Indeed, recent studies have found the llpophillc alpha-i adrenoceptor agonist St 587 to be capable of antagonizing some of the effects of ethanol in one inbred strain of mouse (21,22). However It should be noted that thls drug is not completely selectlve for alpha-1 receptors and also appears to act at alpha-2 adrenoceptors (23), and its ethanol-antagonlzlng effects were not observed in some strains of mice. Previous studies have implicated noradrenergic neurones in the depressant effects of barbiturates and other anaesthetics (24).
118
Alpha-2 Adrenoceptors and Alcohol
Vol. 44, No. 2, 1989
The present report suggests a clear role for alpha-2 adrenoceptors in the intoxicating effects of ethanol. The observation that atipamezole reduces ethanol's ataxic but not its motor stimulant and anticonvulsant effects, adds further support to the notion that ethanol's different behavioral effects are mediated by different neuroblologlcal mechanisms. In the present study neither atlpamezole nor idazoxan showed any indication of proconvulsant activity in NIH Swiss mice except at doses considerably higher than those required to reduce ethanol intoxication (i.e. i0 mg/kg). It is questionable whether this proconvulsant action is mediated by alpha-2 receptors. It should be noted that at these high doses there was a marked change in the appearance of the mice. They appeared large and fluffy due to piloerection. It is unlikely that an inability to detect a proconvulsant action at the lower doses reflects a low sensitivity of the test, since the identical paradigm has proven extremely sensitive to the proconvulsant action of other drugs (27). Alpha-2 adrenoceptor antagonists may exert more potent proconvulsant effects in other strains of mice (26). The clinical utility of alcohol antagonists is questionable (25). A drug that reduces the effects of ethanol without altering blood ethanol concentrations (as both Ro 15-4513 and alpha-2 adrenoceptor antagonists do) poses a number of ethical problems. While the drugs may reduce the acute intoxicating effects of ethanol, it would appear unlikely that they would reduce the long-term toxicity of ethanol. Further, any clinical potential for Ro 15-4513 is compromised both by its short duration of action (8) and its marked proconvulsant effects (6-8). Atlpamezole and idazoxan have a longer duration of action (2) and a much less clear effect on seizure threshold. Their longer duration of action may make these two drugs more suitable for use in humans, but their potential proconvulsant properties will need further clarification. Acknowledgements We are grateful to Farmos for the gift of atlpamezole, to Reckltt and Colman for the gift of idazoxan, and to Hoffmann-La Roche for the gift of Ro 15-4513. References I. R. VIRTANEN, J.-M. SAVOLA and V. SAAN0, Br. J. Pharmacol., in press. 2. H. SCHEININ, E. MACDONALD and M. SCHEININ, Eur. J. Pharmacol., 151 35-42 (1988). 3. S.E. FILE and A.G. WARDILL, Psychopharmacology 44 53-59 (1975). 4. E.P. BONETTI, W.P. BURKARD, M. GABL and H. MOHLER, Br. J. Pharmacol. 86 463P (1985). 5. P. POLC, Br. J. Pharmacol. 86 465P (1985). 6. P.D. SUZDAK, J.R. G L 0 ~ , J.N. CRAk~E¥, R.D. SCHWARTZ, P. SKOLNICK and S.M. PAUL, Science 234 1243-1247 (1986). 7. R.G. LISTER and J. KARANIAN, Alcohol 4 409-411 (1987). 8. K. MICZEK and E.M. ~k~RTS, Science 235 1127 (1987). 9. R.G. LISTER and D.J. NUTT, Br. J. Pharmacol. 93 210-214 (1988). 10. M.J. DURCAN, R.G. LISTER and M.LINNOILA, Neuropharmacology, in press.
Vol. 44, No. 2, 1989
ii. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27.
Alpha-2 Adrenoceptors and Alcohol
119
E. MAJCHROHICZ, Psychopharmacologia 43 245-254 (1975). D.J. ~ and R.G. LISTER, Brain Res. 413, 193-196 (1987). R.G. LISTER, Life Sci. 41 1481-1489 (1987). R.G. LISTER, Pharmacol. Biochem. Behav. 28 75-79 (1987). P. SYAPIN, K.W. GEE and R.L. ALKANA, Brain Res. Bull. 19 603-605 (1987). R.G. LISTER, Pharmacol. Biochem. Behav., in press. G.B. DE SARR0, C. ASCIOTI, F. FROIO, V. LIBRI and G. NISTIC0, Br. J. Pharmacol. 90 675-685 (1987). S.L. F00TE, F.E. BLOOM and G. ASTON-JONES, Physiol. Rev. 63 844-914 (1983). J.R. UNNERSTALL, T.A. KOPAJTIC and M.J. KUHAR, Brain Res. Rev. 7 69-101 (1984). K. RASMUBBEN and B.L. JAC0BS, Brain Res. 371 335-344 (1986). M.K. MENON and C.K. KODAMA, Life Sci 37 2091-2098 (1985). M.K. MENON and C.K. KODAMA, Life Sci 41 1599-1610 (1987). L. PICHLER and W. KOBINGER, Arznel. Forsch. 35 201-205 (1985). S.T. MASON and A. ANGEL, Eur. J. Pharmacol. 91 29-39 (1983). R.G. LISTER and D.J. NUTT, Trends in Neurosci. i0, 223-225 (1987). A. FLETCHER and E.A. FORSTER, Eur. J. Pharmacol. 151 27-34 (1988). D.J. NUTT and R.G. LISTER, Pharmacol. Biochem. Behav., in press.