- Email: [email protected]
Proconvulsant and anticonvulsant effects in mice of acute and chronic treatment with cocaine
Neurophormacology Vol. 28, No. 7, pp. 709-714, Printed in Great Britain. All rights resewed
1989 Copyright
0
0028-3908/89 $3.00 + 0.00 1989 Pergamon Press plc
PROCONVULSANT AND ANTICONVULSANT EFFECTS IN MICE OF ACUTE AND CHRONIC TREATMENT WITH COCAINE R. KARLER, C. PETTY, L. CALDER and S. A. TURKANIS Departments of Pharmacology and Anesthesiology, University of Utah School of Medicine, Salt Lake City, Utah 84132, U.S.A. (Accepted 21 November 1988) Summary-The proconvulsant and anticonvulsant effects of acute and chronic exposure to cocaine were investigated in adult, male, CF-I mice. The proconvulsant effects of cocaine in mice only manifested themselves after daily exposure to motor-stimulant doses. Although daily treatment decreased electroshock convulsion threshold, no motor convulsions were observed. Animals in the proconvulsant state. however, kindled to electrically-induced convulsions more rapidly than did controls. Furthermore, daily treatment with cocaine and electroshock also enhanced the development of electrical kindling. These results illustrate that the excitatory properties of cocaine in the CNS can enhance phenomena which cause a persistent increase in excitability of the CNS. In contrast to the proconvulsant activity after chronic exposure, cocaine, administered acutely, in motor-stimulant doses, was anticonvulsant in a variety of tests using electroshock and chemically-induced convulsions. The drug elevated electroshock thresholds for both minimal and maximal convulsions and these responses were not blocked by haloperidol. In tests for minimal chemically-induced convulsions, cocaine elevated the threshold to N-methyl-DL-aspartate, but not to bicuculline; against maximal convulsions, the drug was anticonvulsant against both N-methylDL-aspartate and bicuculline. Cocaine did not affect convulsion thresholds for strychnine, arecoline or aminophylline; these data suggest that the anticonvulsant action of cocaine is relatively selective for the y-aminobutyric acid (GABA) and glutamate systems.
Key words-cocaine,
The pharmacological
anticonvulsants, convulsants, mice, kindling, neurotransmitters.
properties
of cocaine,
in part,
resemble those of the psychomotor stimulants, such as amphetamine, and those of the local anesthetics (Post, Weiss, Pert and Uhde, 1987). Although both classes of drug exhibit stimulant and depressant properties in the CNS, only the stimulant properties of cocaine have been studied extensively. If the properties of cocaine include the depressant effects, characteristic of amphetamine or the local anesthetics (Covino and Vassallo, 1975), cocaine should have anticonvulsant activity as well as its reputed proconvulsant activity. Some previous reports suggest that cocaine possesses anticonvulsant activity. Burright, Donovick, Michels, Fanelli and Dolinsky (1982) observed that cocaine inhibited electricallyinduced seizures in both lead-treated and control mice; and Russell and Stripling (1985) and Stripling and Russell (1987) demonstrated in rats that cocaine exerted a variety of anticonvulsant effects on the afterdischarge and on cortically-kindled seizures. lpern (1979) claimed Earlier, however, Greer and that cocaine was proconvu rp sant against flurothylinduced convulsions in adult mice. Preliminary data from this laboratory in mice indicated that cocaine did exert anticonvulsant effects in a variety of conventional electroshock tests and the following study was undertaken to describe in detail the characteristics of the anticonvulsant activity, as well
as some proconvulsant effects, and to investigate possible mechanisms of these actions. METHODS Experimental animals and preparation of drugs
The experiments were conducted on male, CF-1 mice that were initially 5-6 weeks old (20-25 g). The hydrochloride salts of cocaine (NIDA), bicuculline (Sigma) and haloperidol (Sigma), the sulfate salts of oL-amphetamine (Sigma) and strychnine (Mallinckrodt), and N-methyl-DL-aspartic acid (Sigma), arecoline hydrobromide (Sigma) and aminophylline (Lyphomed) were prepared in isotonic solutions of NaCl; all preparations of drug and vehicle were administered intraperitoneally or intravenously, except for haloperidol, which was given subcutaneously. The doses were calculated as the salts. Experimental procedures
Electrical convulsions were induced by applying stimuli with conventional silver cornea1 electrodes. For 60-Hz stimulation, a constant-current stimulator (Wahlquist Instrument Co., Salt Lake City, Utah) was used to generate sinusoidal stimuli of 0.2 or 2 set duration (Sangdee, Turkanis and Karler, 1982). For ~-HZ stimulation, a constant-voltage stimulator 709
R. KARLER et al.
710
(Wahlquist Instrument Co., Salt Lake City, Utah) was used to generate a 3-set train of 200-msec rectangular monophasic pulses. The end-point for minimal convulsions was either jaw or front-limb clonus; for maximal convulsions, hind-limb tonic extension. The thresholds for minimal and maximal convulsions for the chemically-induced seizures were determined by intravenous infusion via the tail-vein, using a Harvard Apparatus infusion pump. The infusion rates were: bicuculline (0.1 mg/ml), 0.2 ml/min; strychnine (0.05 mg/ml), 0.28 ml/min; N-methyl-DL-aspartate (20 mg/ml), 0.54 ml/min; arecoline (3 mg/ml), 0.1 ml/min; and aminophylline (19.7 mg/ml), 0.39 ml/min. Heparin (10 units/ml) was added to the infusion solutions in order to prevent clotting. The end-point for minimal convulsions was at least 3 set of persistent clonic convulsive activity; for maximal convulsions, hind-limb tonic extension. Locomotor activity was assessed using 12 clear plastic cages, each containing an infrared photocell beam, placed 2 cm above the floor. Immediately after injection, the mice were put into the activity cages. Motor activity was quantified by totaling interruptions of the photocell beam, which were stored automatically in a computer and the mean values and their standard deviations were printed out at 15-min intervals during the test period. Data reported below represent mean values obtained from 6 cages with 3 mice/cage, measured over a 60- or 90-min period immediately after treatment with drug. All test sessions were carried out between 1000 and 1500 hr. RESULTS
Locomotor stimulant and proconvulsant studies
Figure 1 represents a dose-effect curve for cocaine as a locomotor stimulant. These data serve as a reference point for the doses used in the subsequent proconvulsant and anticonvulsant studies. The data in Figure 2 illustrate the effect of oncedaily treatment with cocaine on the threshold for electroshock convulsions. Initially, the drug raised the threshold for convulsions, which represents anticonvulsant activity (shown in the acute experiments described in Fig. 7). With repeated treatment, this property of cocaine changed and after 16 days of treatment the drug lowered the threshold for convulsions; therefore, it became proconvulsant but did not evoke convulsions. The data in Figure 3 demonstrate that the proconvulsant effect of cocaine in chronically-treated animals was dose-related. Figure 4 shows the persistence of the proconvulsant effect after withdrawal from a chronic pretreatment regimen with cocaine. In this experiment, the controls represent the threshold for electroshock convulsions in animals that had received only chronic treatment with saline; the acute effects of cocaine were the control responses with cocaine in chronically saline-treated animals. A comparison of
I
I
IO
20
I
40
COCAINE,mg/kg
Fig. 1. Dose-effect relationship for cocaine and motor activity. Locomotor activity of mice was measured electronically, with an infrared beam and a detector in each activity cage, plus a computer. Each data point = the mean number of counts over a period of 60-min from 12 cages with 3 mice/cage. l = Data from cocaine-treated mice; n = control. The slope of the dose-response curve is significantly different from zero, as determined by a chisquare test (P < 0.05; Spiegel, 1961).
these two groups again illustrates the anticonvulsant activity, characteristic of the acute administration of cocaine. The data for chronic administration of cocaine were the effects of cocaine in animals withdrawn from the chronic pretreatment with cocaine; these results indicate that cocaine was proconvulsant on day 1 of withdrawal, but, by day 7, the anticonvulsant activity again was manifested. Figure 5 illustrates the influence of concomitant treatment with cocaine and electroshock on the development of electrical kindling. These animals were given a daily electrical stimulus and kindling was expressed as the percentage of animals in each group that exhibited a minimal convulsion. The group given
-I
6
16
TREATMENT
24
32
DAYS
Fig. 2. Influence of daily treatment with cocaine on electroshock minimal convulsion threshold. Each data point represents the median threshold obtained from 35 mice. Mice were treated once daily with either 40 mg/kg cocaine (i.p.) or with saline for 32 days. On the days indicated, convulsion thresholds to a 60-Hz, 2-set stimulus were determined 15 min after each treatment. Each value for the drug is significantly different from its comparable control, as determined by a relative potency test (J- d 0.05; Litchfield and Wilcoxon, 1949).
711
Cocaine pro- and anticonvulsant activity
4
EJo-(7 I
2
345678
36 DAYS
20
40
60
COCAINE,mg/
kg
Fig. 3. Dose-effect relationship for proconvulsant activity of cocaine in mice treated chronically with cocaine. All animals were treated with either 4Omg/kg cocaine or vehicle once daily (i.p.) for 23 days, On day 24, the animals pretreated with cocaine were given varying doses of cocaine and the mean minimal convulsion threshold for groups of 10 mice was determined 15 min after the drug. Parameters for the electrical stimuli were 60 Hz and 2 set; the mean control threshold for the saline-pretreated mice was 3.2 mA. The values for drug are significantly different from the controls, as determined by a many-one rank statistic (P < 0.05; Steel, 1959).
cocaine received treatment with cocaine 15 min (peak anticonvulsant time) prior to the daily electrical stimulation. On the first two days of treatment, there were fewer animals pretreated with cocaine that exhibited convulsions, which demonstrated the anticonvulsant activity of cocaine. By day 4, however, the anticonvulsant effect had given way to a proconvulsant effect which persisted throughout the remainder of the experiment. At the end of the ~
3.6
COCAINE
E
4
(acute)
3.4
8 z
3.24
COCAINE
WITHDRAWAL
khronicl
Fig. 5. Enhancement of electrical kindling to minimal convulsions by concomitant exposure to cocaine. l = Data from cocaine-treated mice; 0 = data from saline-treated mice. Fifty mice were given 40 mg/kg cocaine once daily for 8 days; another 50 mice were given saline in an identical schedule. Fifteen min after treatment with drug or saline, a kindling stimulus (60 Hz, 2 set, 2.6 mA) was administered to each animal. To ascertain the persistence of the kindling, the animals were not treated with saline or drug after day 8 but were rechallenged with the kindling stimulus on day 36. The curve obtained-from the drug-treated animals is significantly different from the control curve, as determined by a chisquare test (P < 0.05; Spiegel, 1961).
treatment period, day 8, more than twice as many animals in the group given cocaine, as compared with the controls, were exhibiting seizures; therefore, pretreatment with cocaine enhanced the development of kindled convulsions. After day 8, both groups of animals were withdrawn from treatment but were retested for the persistence of the kindling effect on day 36,28 days after withdrawal of treatment: as can be seen, the kindling stimulus alone still elicited convulsions in about the same number of mice as it did on day 8. Figure 6 illustrates another effect of cocaine on the development of electrical kindling. In this experiment, the group given cocaine was pretreated once daily for 20 days with cocaine; as shown in Figure 2, cocaine was proconvulsant after this period of re-
DAYS
Fig. 4. Persistence of the proconvulsant cocaine effect after withdrawal from chronic treatment. Each data point is the mean minimal convulsive threshold of 10 mice after receiving a test dose of cocaine (40 mg/kg, i.p.). The electroshock parameters were 60Hz and 2sec. Animals were pretreated once daily with either cocaine (40 mg/kg, i.p.) or isotonic saline solution for 26 days. Thresholds were determined 15 min after cocaine treatment in separate groups of animals on the withdrawal days indicated. l = Values from the acute cocaine controls: 0 = chronic cocaine treatment; m = saline controls. As determined by a Mann-Whitney test, the chronic cocaine value on day 1 of withdrawal was significantly different from the acute cocaine response (P < 0.05; Snedecor and Cochran, 1967). As determined by a many-one rank statistic (Steel, 1959), all drug values are significantly different from their respective controls (P < 0.05).
DAYS
Fig. 6. Enhancement of electrical kindling to minimal convulsions by pretreatment with cocaine. l = Data from cocaine-treated mice; 0 = data from saline-treated mice.. Twenty-five mice were pretreated with 40 mg/kg of cocaine once daily for 20 days; another 25 mice were given saline in an identical schedule. Twenty-four hr after the last day of treatment, a kindling stimulus (60Hz, 2sec, 2.6mA) was administered to each mouse once daily for 10 days. The curve obtained from the mice pretreated with drug is significantly different from the control curve, as determined by a chi-square test (P < 0.05; Spiegel, 1961).
R. KARLERet al.
chronically with cocaine kindled much more rapidly than did the controls. Anticonvulsant studies
7
’; ;
’
Cocaine
;
lb
;o ’ 410
(mg/kg)
Fig. 7. Anticonvulsant activity of cocaine against electrically-induced minimal convulsions. l = Data from cocaine-treated and n = data from saline-treated mice; all tests, 15 min after treatment. (A) Relationship between the dose of cocaine and mean minimal convulsion threshold. Each data point represents the results from 10 mice; parameters for the electrical stimuli were 60 Hz and 2 sec. Data obtained with 1 mg/kg were significantly different from those obtained with 40/kg, as determined by a Mann-Whitney test (P ~0.05; Siegel, 1956). (B) Relation between the dose of cocaine and the percentage of mice that elicited a minimal convulsion in response to a constant electroshock stimulus (60 Hz, 2 set, 3 mA). Each data point represents the results from 35 mice. The slope of the dose-effect curve is significantly different from zero, as determined by a chi-square test (P < 0.05; Spiegel, 1961).
exposure. After withdrawal from the treatment with cocaine, animals were subjected to a daily electrical kindling stimulus and the development of kindled convulsions was plotted for a period of lOdays. It can be seen that the group pretreated peated
Table
The data shown in Figure 2 suggest that cocaine, acutely administered, had anticonvulsant activity, which was investigated in more detail by the results of the study illustrated in Figure 7 (A and B). The data shown in Figure 7(A) indicate that cocaine raised the minimal convulsion threshold in the 60-Hz electroshock test in a dose-related manner. Although the increases were statistically significant, the functional significance of the changes is shown in Figure 7(B). In this experiment, the effect of cocaine on the percentage of animals that displayed a convulsion to a constant electrical stimulus was determined. The electrical stimulus employed caused a minimal convulsion in about 50% of the controls; and the data for cocaine showed that the greater the dose, the greater the number of animals which were protected from a convulsion. These data support the conclusion that the increases in threshold for convulsions produced by cocaine were functionally significant. Table 1 summarizes the anticonvulsant properties of cocaine, as determined in a variety of tests with electroshock and chemically-induced convulsions. In all three electroshock tests, cocaine elevated the thresholds for convulsions; however, in the chemical tests, only the minimal convulsion threshold to Nmethyl-DL-aspartate and the maximal convulsion thresholds to bicuculline and N-methyl-DL-aspartate were elevated. The drug did not affect the convulsion thresholds to strychnine, arecoline or aminophylline. The data in Table 2 represent the influence of haloperidol on the anticonvulsant and locomotorstimulant activities of cocaine, amphetamine and tetracaine. The data indicate that haloperidol blocked
1. Anticonvulsant urooerties of cocaine Convulsive
threshold
Minimal Control Electroshock
threshold
2.9 (2.7-3.1) 15(13-17)
&Hz (V) temb
conclusion
(mg/kg,
Control
Cocaine
testsa
60-Hz (mA) Chemical
Maximal Cocaine
4.2 (3.9-4.5) 48 (42-54)
10(9-11) -
> 100~
506573 1.0+0.3 0.5 50.1
> 828’ >I.50 0.5 f 0.1
threshold
i.v.)
N-methyl-DL-aspartate Bicuculline Strychnine Arecoline
498f 49 0.6iO.1 7.2 k 1.1
618f 57’ 0.6 k 0.1 -
Aminoehvlline
318+51
315 +33
7.3 i
1.5
-
-
369 i 63
369 f 43
Mice were treated intraperitoneally 15 min before tests with a solution of either saline or cocaine (40 mg/kg). ‘Electroshock data are median thresholds and their 95% confidence limits; each median value was obtained with 35 mice; the duration of stimulus was 2 sec. bChemical convulsion data are mean convulsion thresholds and their standard deviations; each mean value was obtained with 15 mice. value significantly different from that of control, as determined by a relative potency test (P C 0.05; Litchfield and Wilcoxon, 1949). dAll drug-treated animals were completely protected from a stimulus of lOOmA. ‘Value significantly different from that of control, as determined by a f-test (P < 0.05; Snedecor and Cochran, 1967). ‘All drug-treated animals were completely protected against maximal convulsions up to a dose of N-methyl-oL-aspartate of 828 mg/kg. sAlI drug-treated animals were completely protected against maximal convulsions; dose of bicuculline greater than 1.5 mg/kg, (iv.) caused death.
Cocaine pro- and anticonvulsant activity Table 2. Influenca of haloperidol on the locomotor and anticonvulsant activities of cocaine and amnhetamine Locomotor activity’ (counts/90 min) Cocaine (20 mg/kg, i.p.) Saline Cocaine Cocaine + haloperidol
448 (Eal6) 2118 (98&2926)d 198 (32-869)
Convulsion thresholdb (mA) 5.4 (4.8-6.0) 6.6 (6.0-7.2) 6.6 (6.iX7.2)’
(1mg/kd Amphefamke (6 mg/kg, i.p.) Saline Amphetamine Amphetamine + haloperidol (0.25 mg/kg)c Tetracaine (30 mg/kg, i.p.) Saline Tetracainc Tetracaine + haloperidol (1 ma/kn. s.c.Y
6.4 (5.8-7.0) 403 (238-480) 3466 (2i351-3975)d 8.0 (7.5-8.5) 312(21&417) 6.8 (6.4-7.2)
-
5.3 (4.8-5.8) 6.3 (5.8-6.8) 6.5 (6X-7.0)’
‘Locomotor activity represents means and ranges (N = 6) obtained immediately after test drugs. bConvulsion thresholds (60-Hz) represent the median convulsive currents and their 95% confidence limits (N = 40) obtained by the “staircase” method (Litchfield and Wilcoxon, 1949; Finney, 1971). The electroshock stimulus duration was 0.2 set; test carried out 30 min after test drug. ‘Haloperidol administered subcutaneously 30 min prior to the test drugs; haloperidol alone exerted no effect on control locomotor activity and convulsion threshold. %gnificantly different from control (P < 0.05; many-one rank statistic; Steel. 1959). CSignigcantIy different from control, as determined by a relative potency test (P < 0.05; Litchficld and Wilcoxon, 1949).
the motor-stimulant effect of cocaine, but had no effect on the anticonvulsant activity of cocaine in the 60-Hz test, even in doses as large as 5 mg/kg. In contrast, haloperidol blocked both the motorstimulant and anticonvulsant activity of amphetamine. Finally, although tetracaine is not a motor stimulant, it has anticonvulsant activity which, like that of cocaine, was not blocked by haloperidol. DISCUSSION
Local anesthetics generally possess both excitatory and depressant properties in the CNS; specifically, large doses cause convulsions and small doses are anticonvulsant (Covino and Vassallo, 1975). Similarly, cocaine is both convulsant and anticonvulsant, even though only the excitatory properties have generally been studied. For example, cocaine, given acutely to rats can cause convulsions (Stripling and Hendricks, 1981) and, when administered repeatedly in non-convulsive, doses can produce kindling (Post and Kopanda, 1975). There appear, however, to be species differences because, in the mouse, cocaine administered acutely does not precipitate convulsions. even in doses up to 200mg/kg, (iv.) (unpublished data). Similarly, neither lidocaine nor tetracaine given intravenously, causes convulsions in the mouse (unpublished data). Furthermore, chronically-administered cocaine does not cause kindling, even though the threshold for convulsions decreases. The decrease in threshold with repeated treatment, however, was relatively short-lived (Fig. 4), which distinguishes the effect from the
713
long-term change in excitability, characteristic of kindling (Sangdee, Turkanis and Karler, 1982). Although mice did not kindle with repeated exposure to cocaine, the data presented illustrate that chronic exposure to cocaine could facilitate electrical kindling (Figs 5 and 6). The observation that cocaine enhanced kindling is in contrast to several previous reports of the absence of any effect of cocaine and electrical kindling (Stripling, 1983). The most conspicuous differences between the previous and present studies lie in the method of electrical kindling and in the species. The earlier results were derived from the rat and the kindling was accomplished with indwelling electrodes in the brain, whereas the present experiments were effected on the mouse and the kindling was produced by means of transcorneal electrical stimulation. Whether the mechanism of kindling differs with different methods of stimulation is not known, but species differences in response to cocaine were manifested clearly in the present studies by the failure of cocaine, administered acutely, to elicit convulsions and, administered chronically, to kindle convulsions. Nevertheless, the ability of cocaine either to kindle or to facilitate kindling may be of toxicological significance in the abuse of cocaine. In a study reported by Finkle and McCloskey (1977), a large fraction of human deaths, related to abuse of cocaine, was the result of convulsions followed by respiratory arrest. In the studies described above of the excitatory and proconvulsant activity of cocaine, the anticonvulsant activity of cocaine was evident in the range of doses associated with excitation of the CNS. The detailed studies of the anticonvulsant activity clearly illustrated that cocaine, administered acutely, can raise thresholds for convulsions in a variety of tests with electroshock and chemically-induced convulsions (Table 1). Although the data show that cocaine elevated thresholds for convulsions in all the electroshock tests, they do not provide information on potential mechanisms of the anticonvulsant activity; they do, however, suggest that the anticonvulsant activity of cocaine is relatively non-selective, compared with drugs such as phenytoin, which, for example, is effective in the maximal electroshock tests and the ~-HZ test, but not in the 60-Hz electroshock minimal-convulsive threshold test. The purpose of the tests with chemically-induced convulsions was to define the anticonvulsant selectivity of cocaine and to identify potential mechanisms of its anticonvulsant activity. The results of these tests illustrate the selectivity of cocaine against both minimal and maximal convulsions induced chemically (Table 1). For example, cocaine did not elevate the minimal convulsion threshold to bicuculline, arecoline or aminophylline, but it did raise the minimal threshold against N-methyl-DL-aspartate. Some selectivity was also illustrated by the results in the maximal convulsion threshold tests. Although the drug raised this
714
R. KARLER et al.
threshold for bicuculline and N-methyl-DL-aspartate, it did not affect the threshold for either strychnine or aminophylline. This relative selectivity in the various chemical tests points to possible mechanisms of action for the anticonvulsant activity of cocaine: because the drug failed to elevate minimal convulsive thresholds for the y -aminobutyric acid (GABA) antagonist, bicuculline, the cholinergic agonist, arecoline and the adenosine antagonist, aminophylline, these various potential mechanisms do not appear to be involved in the anticonvulsant action of cocaine. The only minimal convulsion threshold affected by cocaine was the threshold for N-methyl-DL-aspartate, which suggests that this subtype of glutamate receptor is either directly or indirectly affected by the action of cocaine. Such selectivity did not extend to the maximal convulsion tests: in this case, cocaine was anticonvulsant against both the GABA antagonist and the glutamate agonist; the activity did not, however, include the glycine antagonist, strychnine, or the adenosine antagonist, aminophylline. The ability of the drug to block maximal convulsions, originating from both the excitatory and inhibitory components of the CNS contrasts with its more selective action against minimal convulsions induced by N-methyl-DL-aspartate. These results were repeated in other experiments and the reason for the difference in selectivity between the minimal and maximal convulsions is not understood. The haloperidol-cocaine studies illustrate another facet of the mechanism of action of cocaine (Table 2). These studies were designed to determine if the cocaine was a activity of anticonvulsant dopaminergically-mediated effect. Because of the pharmacological similarities between cocaine and both amphetamine and tetracaine, these latter two drugs were included in the study with haloperidol for comparative purposes. The results indicate that haloperidol could block the locomotor-stimulant effect of both cocaine and amphetamine and could block the anticonvulsant effect of amphetamine, but not that of either cocaine or tetracaine. The failure of haloperidol to block the anticonvulsant effects of cocaine in mice is consistent with similar previous observations in the rat (Russell and Stripling, 1985). The data suggest that the anticonvulsant activity of amphetamine involves a dopamine system, but that of cocaine some other mechanism related, possibly, to its local-anesthetic activity. The data from the various tests with chemically-induced convulsions,
described above, indicate that the drug may be anticonvulsant, at least in part, by virtue of an action affecting the N-methyl-DL-asparatate component of the glutamate system. work was supported by NIDA
Acknowledgement-This
research grant DA-00346. REFERENCES
Burright R. G., Donovick P. J., Michels K., Fanelli R. J. and Dolinskv Z. (1982) . , Effect of amohetamine and cocaine on seizure on lead treated mice. Pharmac. Biothem. Behav. 16: 631-635.
Covino B. G. and Vassallo H. G. (1975) Local Anesthetics. Grune and Stratton, New York. Finkle B. S. and McCloskey K. L. (1977) The forensic toxicology of cocaine. In: Cocaine: 1977 (Peterson R. C., Ed.), NIDA Research Monograph 13. U.S. Govt. Printing Office, Washington, D.C. _ Finnev D. J. (1971) Probit Analvsis. 3rd Edn. Cambridge Umversity Press,‘Cambridge. . Greer C. A. and Alpern H. P. (1979) Maturational changes related to dopamine in the effects of d-amphetamine, cocaine, nicotine, and strychnine on seizure susceptibility. _ _ Psychopharmacology
64-255-260.
Litchfield J. T. Jr and Wilcoxon F. (1949) A simolified method for evaluating dose-effect experiments. J. ‘Pharmat. exp. Ther. %: 99113. Post R. M. and Kopanda R. T. (1975) Cocaine, kindling, and reverse tolerance. Luncet i: 409410. Post R. M., Weiss S. R. B., Pert A. and Uhde T. W. (1987) Chronic cocaine administration: Sensitization and kindling effects. In: Cocaine: Clinical and Biobehavioral Aspects (Fisher S., Raskin A. and Uhlenhuth E. H., Eds) Oxford University Press, New York. Russell R. D. and Stripling J. S. (1985) Monoaminergic and local anesthetic components of cocaine’s effects on kindled seizure expression. Pharmac. Biochem. Behav. 22: 427-434.
Sangdee P., Turkanis S. A. and Karler R. (1982) Kindlinglike effect induced by repeated cornea1 electroshock in mice. Epilepsia 23: 471-479. Siegel S. (1956) Nonparametric Statistics. McGraw-Hill, New York. Snedecor G. W. and Cochran W. G. (1967) In: Statistical Methods, 6th Edn. Ames, Iowa. Sniegal M. R. (1961) Theorv and Problems of Statistics. ‘M&aw-Hill,X New York.. Steel R. G. D. (1959) A multiple comparison rank sum test: treatments versus control. Biometrics 15: 560-572. Stripling J. S. (1983) Cocaine and “pharmacological kindling” in the rat..Expl Neural. 82: 499-503. Strinlina J. S. and Hendricks C. (1981) Facilitation of k;ndlLg by convulsions induced by cocaine or lidocaine but not pentylenetetrazol. Pharmac. Biochem. Behav. 15: 793-798. Stripling J. S. and Russell R. D. (1985) Effect of cocaine and pentylenetetrazol on cortical kindling. Pharmac. Biochem. Behav. 23: 573-581.
1989 Copyright
0
0028-3908/89 $3.00 + 0.00 1989 Pergamon Press plc
PROCONVULSANT AND ANTICONVULSANT EFFECTS IN MICE OF ACUTE AND CHRONIC TREATMENT WITH COCAINE R. KARLER, C. PETTY, L. CALDER and S. A. TURKANIS Departments of Pharmacology and Anesthesiology, University of Utah School of Medicine, Salt Lake City, Utah 84132, U.S.A. (Accepted 21 November 1988) Summary-The proconvulsant and anticonvulsant effects of acute and chronic exposure to cocaine were investigated in adult, male, CF-I mice. The proconvulsant effects of cocaine in mice only manifested themselves after daily exposure to motor-stimulant doses. Although daily treatment decreased electroshock convulsion threshold, no motor convulsions were observed. Animals in the proconvulsant state. however, kindled to electrically-induced convulsions more rapidly than did controls. Furthermore, daily treatment with cocaine and electroshock also enhanced the development of electrical kindling. These results illustrate that the excitatory properties of cocaine in the CNS can enhance phenomena which cause a persistent increase in excitability of the CNS. In contrast to the proconvulsant activity after chronic exposure, cocaine, administered acutely, in motor-stimulant doses, was anticonvulsant in a variety of tests using electroshock and chemically-induced convulsions. The drug elevated electroshock thresholds for both minimal and maximal convulsions and these responses were not blocked by haloperidol. In tests for minimal chemically-induced convulsions, cocaine elevated the threshold to N-methyl-DL-aspartate, but not to bicuculline; against maximal convulsions, the drug was anticonvulsant against both N-methylDL-aspartate and bicuculline. Cocaine did not affect convulsion thresholds for strychnine, arecoline or aminophylline; these data suggest that the anticonvulsant action of cocaine is relatively selective for the y-aminobutyric acid (GABA) and glutamate systems.
Key words-cocaine,
The pharmacological
anticonvulsants, convulsants, mice, kindling, neurotransmitters.
properties
of cocaine,
in part,
resemble those of the psychomotor stimulants, such as amphetamine, and those of the local anesthetics (Post, Weiss, Pert and Uhde, 1987). Although both classes of drug exhibit stimulant and depressant properties in the CNS, only the stimulant properties of cocaine have been studied extensively. If the properties of cocaine include the depressant effects, characteristic of amphetamine or the local anesthetics (Covino and Vassallo, 1975), cocaine should have anticonvulsant activity as well as its reputed proconvulsant activity. Some previous reports suggest that cocaine possesses anticonvulsant activity. Burright, Donovick, Michels, Fanelli and Dolinsky (1982) observed that cocaine inhibited electricallyinduced seizures in both lead-treated and control mice; and Russell and Stripling (1985) and Stripling and Russell (1987) demonstrated in rats that cocaine exerted a variety of anticonvulsant effects on the afterdischarge and on cortically-kindled seizures. lpern (1979) claimed Earlier, however, Greer and that cocaine was proconvu rp sant against flurothylinduced convulsions in adult mice. Preliminary data from this laboratory in mice indicated that cocaine did exert anticonvulsant effects in a variety of conventional electroshock tests and the following study was undertaken to describe in detail the characteristics of the anticonvulsant activity, as well
as some proconvulsant effects, and to investigate possible mechanisms of these actions. METHODS Experimental animals and preparation of drugs
The experiments were conducted on male, CF-1 mice that were initially 5-6 weeks old (20-25 g). The hydrochloride salts of cocaine (NIDA), bicuculline (Sigma) and haloperidol (Sigma), the sulfate salts of oL-amphetamine (Sigma) and strychnine (Mallinckrodt), and N-methyl-DL-aspartic acid (Sigma), arecoline hydrobromide (Sigma) and aminophylline (Lyphomed) were prepared in isotonic solutions of NaCl; all preparations of drug and vehicle were administered intraperitoneally or intravenously, except for haloperidol, which was given subcutaneously. The doses were calculated as the salts. Experimental procedures
Electrical convulsions were induced by applying stimuli with conventional silver cornea1 electrodes. For 60-Hz stimulation, a constant-current stimulator (Wahlquist Instrument Co., Salt Lake City, Utah) was used to generate sinusoidal stimuli of 0.2 or 2 set duration (Sangdee, Turkanis and Karler, 1982). For ~-HZ stimulation, a constant-voltage stimulator 709
R. KARLER et al.
710
(Wahlquist Instrument Co., Salt Lake City, Utah) was used to generate a 3-set train of 200-msec rectangular monophasic pulses. The end-point for minimal convulsions was either jaw or front-limb clonus; for maximal convulsions, hind-limb tonic extension. The thresholds for minimal and maximal convulsions for the chemically-induced seizures were determined by intravenous infusion via the tail-vein, using a Harvard Apparatus infusion pump. The infusion rates were: bicuculline (0.1 mg/ml), 0.2 ml/min; strychnine (0.05 mg/ml), 0.28 ml/min; N-methyl-DL-aspartate (20 mg/ml), 0.54 ml/min; arecoline (3 mg/ml), 0.1 ml/min; and aminophylline (19.7 mg/ml), 0.39 ml/min. Heparin (10 units/ml) was added to the infusion solutions in order to prevent clotting. The end-point for minimal convulsions was at least 3 set of persistent clonic convulsive activity; for maximal convulsions, hind-limb tonic extension. Locomotor activity was assessed using 12 clear plastic cages, each containing an infrared photocell beam, placed 2 cm above the floor. Immediately after injection, the mice were put into the activity cages. Motor activity was quantified by totaling interruptions of the photocell beam, which were stored automatically in a computer and the mean values and their standard deviations were printed out at 15-min intervals during the test period. Data reported below represent mean values obtained from 6 cages with 3 mice/cage, measured over a 60- or 90-min period immediately after treatment with drug. All test sessions were carried out between 1000 and 1500 hr. RESULTS
Locomotor stimulant and proconvulsant studies
Figure 1 represents a dose-effect curve for cocaine as a locomotor stimulant. These data serve as a reference point for the doses used in the subsequent proconvulsant and anticonvulsant studies. The data in Figure 2 illustrate the effect of oncedaily treatment with cocaine on the threshold for electroshock convulsions. Initially, the drug raised the threshold for convulsions, which represents anticonvulsant activity (shown in the acute experiments described in Fig. 7). With repeated treatment, this property of cocaine changed and after 16 days of treatment the drug lowered the threshold for convulsions; therefore, it became proconvulsant but did not evoke convulsions. The data in Figure 3 demonstrate that the proconvulsant effect of cocaine in chronically-treated animals was dose-related. Figure 4 shows the persistence of the proconvulsant effect after withdrawal from a chronic pretreatment regimen with cocaine. In this experiment, the controls represent the threshold for electroshock convulsions in animals that had received only chronic treatment with saline; the acute effects of cocaine were the control responses with cocaine in chronically saline-treated animals. A comparison of
I
I
IO
20
I
40
COCAINE,mg/kg
Fig. 1. Dose-effect relationship for cocaine and motor activity. Locomotor activity of mice was measured electronically, with an infrared beam and a detector in each activity cage, plus a computer. Each data point = the mean number of counts over a period of 60-min from 12 cages with 3 mice/cage. l = Data from cocaine-treated mice; n = control. The slope of the dose-response curve is significantly different from zero, as determined by a chisquare test (P < 0.05; Spiegel, 1961).
these two groups again illustrates the anticonvulsant activity, characteristic of the acute administration of cocaine. The data for chronic administration of cocaine were the effects of cocaine in animals withdrawn from the chronic pretreatment with cocaine; these results indicate that cocaine was proconvulsant on day 1 of withdrawal, but, by day 7, the anticonvulsant activity again was manifested. Figure 5 illustrates the influence of concomitant treatment with cocaine and electroshock on the development of electrical kindling. These animals were given a daily electrical stimulus and kindling was expressed as the percentage of animals in each group that exhibited a minimal convulsion. The group given
-I
6
16
TREATMENT
24
32
DAYS
Fig. 2. Influence of daily treatment with cocaine on electroshock minimal convulsion threshold. Each data point represents the median threshold obtained from 35 mice. Mice were treated once daily with either 40 mg/kg cocaine (i.p.) or with saline for 32 days. On the days indicated, convulsion thresholds to a 60-Hz, 2-set stimulus were determined 15 min after each treatment. Each value for the drug is significantly different from its comparable control, as determined by a relative potency test (J- d 0.05; Litchfield and Wilcoxon, 1949).
711
Cocaine pro- and anticonvulsant activity
4
EJo-(7 I
2
345678
36 DAYS
20
40
60
COCAINE,mg/
kg
Fig. 3. Dose-effect relationship for proconvulsant activity of cocaine in mice treated chronically with cocaine. All animals were treated with either 4Omg/kg cocaine or vehicle once daily (i.p.) for 23 days, On day 24, the animals pretreated with cocaine were given varying doses of cocaine and the mean minimal convulsion threshold for groups of 10 mice was determined 15 min after the drug. Parameters for the electrical stimuli were 60 Hz and 2 set; the mean control threshold for the saline-pretreated mice was 3.2 mA. The values for drug are significantly different from the controls, as determined by a many-one rank statistic (P < 0.05; Steel, 1959).
cocaine received treatment with cocaine 15 min (peak anticonvulsant time) prior to the daily electrical stimulation. On the first two days of treatment, there were fewer animals pretreated with cocaine that exhibited convulsions, which demonstrated the anticonvulsant activity of cocaine. By day 4, however, the anticonvulsant effect had given way to a proconvulsant effect which persisted throughout the remainder of the experiment. At the end of the ~
3.6
COCAINE
E
4
(acute)
3.4
8 z
3.24
COCAINE
WITHDRAWAL
khronicl
Fig. 5. Enhancement of electrical kindling to minimal convulsions by concomitant exposure to cocaine. l = Data from cocaine-treated mice; 0 = data from saline-treated mice. Fifty mice were given 40 mg/kg cocaine once daily for 8 days; another 50 mice were given saline in an identical schedule. Fifteen min after treatment with drug or saline, a kindling stimulus (60 Hz, 2 set, 2.6 mA) was administered to each animal. To ascertain the persistence of the kindling, the animals were not treated with saline or drug after day 8 but were rechallenged with the kindling stimulus on day 36. The curve obtained-from the drug-treated animals is significantly different from the control curve, as determined by a chisquare test (P < 0.05; Spiegel, 1961).
treatment period, day 8, more than twice as many animals in the group given cocaine, as compared with the controls, were exhibiting seizures; therefore, pretreatment with cocaine enhanced the development of kindled convulsions. After day 8, both groups of animals were withdrawn from treatment but were retested for the persistence of the kindling effect on day 36,28 days after withdrawal of treatment: as can be seen, the kindling stimulus alone still elicited convulsions in about the same number of mice as it did on day 8. Figure 6 illustrates another effect of cocaine on the development of electrical kindling. In this experiment, the group given cocaine was pretreated once daily for 20 days with cocaine; as shown in Figure 2, cocaine was proconvulsant after this period of re-
DAYS
Fig. 4. Persistence of the proconvulsant cocaine effect after withdrawal from chronic treatment. Each data point is the mean minimal convulsive threshold of 10 mice after receiving a test dose of cocaine (40 mg/kg, i.p.). The electroshock parameters were 60Hz and 2sec. Animals were pretreated once daily with either cocaine (40 mg/kg, i.p.) or isotonic saline solution for 26 days. Thresholds were determined 15 min after cocaine treatment in separate groups of animals on the withdrawal days indicated. l = Values from the acute cocaine controls: 0 = chronic cocaine treatment; m = saline controls. As determined by a Mann-Whitney test, the chronic cocaine value on day 1 of withdrawal was significantly different from the acute cocaine response (P < 0.05; Snedecor and Cochran, 1967). As determined by a many-one rank statistic (Steel, 1959), all drug values are significantly different from their respective controls (P < 0.05).
DAYS
Fig. 6. Enhancement of electrical kindling to minimal convulsions by pretreatment with cocaine. l = Data from cocaine-treated mice; 0 = data from saline-treated mice.. Twenty-five mice were pretreated with 40 mg/kg of cocaine once daily for 20 days; another 25 mice were given saline in an identical schedule. Twenty-four hr after the last day of treatment, a kindling stimulus (60Hz, 2sec, 2.6mA) was administered to each mouse once daily for 10 days. The curve obtained from the mice pretreated with drug is significantly different from the control curve, as determined by a chi-square test (P < 0.05; Spiegel, 1961).
R. KARLERet al.
chronically with cocaine kindled much more rapidly than did the controls. Anticonvulsant studies
7
’; ;
’
Cocaine
;
lb
;o ’ 410
(mg/kg)
Fig. 7. Anticonvulsant activity of cocaine against electrically-induced minimal convulsions. l = Data from cocaine-treated and n = data from saline-treated mice; all tests, 15 min after treatment. (A) Relationship between the dose of cocaine and mean minimal convulsion threshold. Each data point represents the results from 10 mice; parameters for the electrical stimuli were 60 Hz and 2 sec. Data obtained with 1 mg/kg were significantly different from those obtained with 40/kg, as determined by a Mann-Whitney test (P ~0.05; Siegel, 1956). (B) Relation between the dose of cocaine and the percentage of mice that elicited a minimal convulsion in response to a constant electroshock stimulus (60 Hz, 2 set, 3 mA). Each data point represents the results from 35 mice. The slope of the dose-effect curve is significantly different from zero, as determined by a chi-square test (P < 0.05; Spiegel, 1961).
exposure. After withdrawal from the treatment with cocaine, animals were subjected to a daily electrical kindling stimulus and the development of kindled convulsions was plotted for a period of lOdays. It can be seen that the group pretreated peated
Table
The data shown in Figure 2 suggest that cocaine, acutely administered, had anticonvulsant activity, which was investigated in more detail by the results of the study illustrated in Figure 7 (A and B). The data shown in Figure 7(A) indicate that cocaine raised the minimal convulsion threshold in the 60-Hz electroshock test in a dose-related manner. Although the increases were statistically significant, the functional significance of the changes is shown in Figure 7(B). In this experiment, the effect of cocaine on the percentage of animals that displayed a convulsion to a constant electrical stimulus was determined. The electrical stimulus employed caused a minimal convulsion in about 50% of the controls; and the data for cocaine showed that the greater the dose, the greater the number of animals which were protected from a convulsion. These data support the conclusion that the increases in threshold for convulsions produced by cocaine were functionally significant. Table 1 summarizes the anticonvulsant properties of cocaine, as determined in a variety of tests with electroshock and chemically-induced convulsions. In all three electroshock tests, cocaine elevated the thresholds for convulsions; however, in the chemical tests, only the minimal convulsion threshold to Nmethyl-DL-aspartate and the maximal convulsion thresholds to bicuculline and N-methyl-DL-aspartate were elevated. The drug did not affect the convulsion thresholds to strychnine, arecoline or aminophylline. The data in Table 2 represent the influence of haloperidol on the anticonvulsant and locomotorstimulant activities of cocaine, amphetamine and tetracaine. The data indicate that haloperidol blocked
1. Anticonvulsant urooerties of cocaine Convulsive
threshold
Minimal Control Electroshock
threshold
2.9 (2.7-3.1) 15(13-17)
&Hz (V) temb
conclusion
(mg/kg,
Control
Cocaine
testsa
60-Hz (mA) Chemical
Maximal Cocaine
4.2 (3.9-4.5) 48 (42-54)
10(9-11) -
> 100~
506573 1.0+0.3 0.5 50.1
> 828’ >I.50 0.5 f 0.1
threshold
i.v.)
N-methyl-DL-aspartate Bicuculline Strychnine Arecoline
498f 49 0.6iO.1 7.2 k 1.1
618f 57’ 0.6 k 0.1 -
Aminoehvlline
318+51
315 +33
7.3 i
1.5
-
-
369 i 63
369 f 43
Mice were treated intraperitoneally 15 min before tests with a solution of either saline or cocaine (40 mg/kg). ‘Electroshock data are median thresholds and their 95% confidence limits; each median value was obtained with 35 mice; the duration of stimulus was 2 sec. bChemical convulsion data are mean convulsion thresholds and their standard deviations; each mean value was obtained with 15 mice. value significantly different from that of control, as determined by a relative potency test (P C 0.05; Litchfield and Wilcoxon, 1949). dAll drug-treated animals were completely protected from a stimulus of lOOmA. ‘Value significantly different from that of control, as determined by a f-test (P < 0.05; Snedecor and Cochran, 1967). ‘All drug-treated animals were completely protected against maximal convulsions up to a dose of N-methyl-oL-aspartate of 828 mg/kg. sAlI drug-treated animals were completely protected against maximal convulsions; dose of bicuculline greater than 1.5 mg/kg, (iv.) caused death.
Cocaine pro- and anticonvulsant activity Table 2. Influenca of haloperidol on the locomotor and anticonvulsant activities of cocaine and amnhetamine Locomotor activity’ (counts/90 min) Cocaine (20 mg/kg, i.p.) Saline Cocaine Cocaine + haloperidol
448 (Eal6) 2118 (98&2926)d 198 (32-869)
Convulsion thresholdb (mA) 5.4 (4.8-6.0) 6.6 (6.0-7.2) 6.6 (6.iX7.2)’
(1mg/kd Amphefamke (6 mg/kg, i.p.) Saline Amphetamine Amphetamine + haloperidol (0.25 mg/kg)c Tetracaine (30 mg/kg, i.p.) Saline Tetracainc Tetracaine + haloperidol (1 ma/kn. s.c.Y
6.4 (5.8-7.0) 403 (238-480) 3466 (2i351-3975)d 8.0 (7.5-8.5) 312(21&417) 6.8 (6.4-7.2)
-
5.3 (4.8-5.8) 6.3 (5.8-6.8) 6.5 (6X-7.0)’
‘Locomotor activity represents means and ranges (N = 6) obtained immediately after test drugs. bConvulsion thresholds (60-Hz) represent the median convulsive currents and their 95% confidence limits (N = 40) obtained by the “staircase” method (Litchfield and Wilcoxon, 1949; Finney, 1971). The electroshock stimulus duration was 0.2 set; test carried out 30 min after test drug. ‘Haloperidol administered subcutaneously 30 min prior to the test drugs; haloperidol alone exerted no effect on control locomotor activity and convulsion threshold. %gnificantly different from control (P < 0.05; many-one rank statistic; Steel. 1959). CSignigcantIy different from control, as determined by a relative potency test (P < 0.05; Litchficld and Wilcoxon, 1949).
the motor-stimulant effect of cocaine, but had no effect on the anticonvulsant activity of cocaine in the 60-Hz test, even in doses as large as 5 mg/kg. In contrast, haloperidol blocked both the motorstimulant and anticonvulsant activity of amphetamine. Finally, although tetracaine is not a motor stimulant, it has anticonvulsant activity which, like that of cocaine, was not blocked by haloperidol. DISCUSSION
Local anesthetics generally possess both excitatory and depressant properties in the CNS; specifically, large doses cause convulsions and small doses are anticonvulsant (Covino and Vassallo, 1975). Similarly, cocaine is both convulsant and anticonvulsant, even though only the excitatory properties have generally been studied. For example, cocaine, given acutely to rats can cause convulsions (Stripling and Hendricks, 1981) and, when administered repeatedly in non-convulsive, doses can produce kindling (Post and Kopanda, 1975). There appear, however, to be species differences because, in the mouse, cocaine administered acutely does not precipitate convulsions. even in doses up to 200mg/kg, (iv.) (unpublished data). Similarly, neither lidocaine nor tetracaine given intravenously, causes convulsions in the mouse (unpublished data). Furthermore, chronically-administered cocaine does not cause kindling, even though the threshold for convulsions decreases. The decrease in threshold with repeated treatment, however, was relatively short-lived (Fig. 4), which distinguishes the effect from the
713
long-term change in excitability, characteristic of kindling (Sangdee, Turkanis and Karler, 1982). Although mice did not kindle with repeated exposure to cocaine, the data presented illustrate that chronic exposure to cocaine could facilitate electrical kindling (Figs 5 and 6). The observation that cocaine enhanced kindling is in contrast to several previous reports of the absence of any effect of cocaine and electrical kindling (Stripling, 1983). The most conspicuous differences between the previous and present studies lie in the method of electrical kindling and in the species. The earlier results were derived from the rat and the kindling was accomplished with indwelling electrodes in the brain, whereas the present experiments were effected on the mouse and the kindling was produced by means of transcorneal electrical stimulation. Whether the mechanism of kindling differs with different methods of stimulation is not known, but species differences in response to cocaine were manifested clearly in the present studies by the failure of cocaine, administered acutely, to elicit convulsions and, administered chronically, to kindle convulsions. Nevertheless, the ability of cocaine either to kindle or to facilitate kindling may be of toxicological significance in the abuse of cocaine. In a study reported by Finkle and McCloskey (1977), a large fraction of human deaths, related to abuse of cocaine, was the result of convulsions followed by respiratory arrest. In the studies described above of the excitatory and proconvulsant activity of cocaine, the anticonvulsant activity of cocaine was evident in the range of doses associated with excitation of the CNS. The detailed studies of the anticonvulsant activity clearly illustrated that cocaine, administered acutely, can raise thresholds for convulsions in a variety of tests with electroshock and chemically-induced convulsions (Table 1). Although the data show that cocaine elevated thresholds for convulsions in all the electroshock tests, they do not provide information on potential mechanisms of the anticonvulsant activity; they do, however, suggest that the anticonvulsant activity of cocaine is relatively non-selective, compared with drugs such as phenytoin, which, for example, is effective in the maximal electroshock tests and the ~-HZ test, but not in the 60-Hz electroshock minimal-convulsive threshold test. The purpose of the tests with chemically-induced convulsions was to define the anticonvulsant selectivity of cocaine and to identify potential mechanisms of its anticonvulsant activity. The results of these tests illustrate the selectivity of cocaine against both minimal and maximal convulsions induced chemically (Table 1). For example, cocaine did not elevate the minimal convulsion threshold to bicuculline, arecoline or aminophylline, but it did raise the minimal threshold against N-methyl-DL-aspartate. Some selectivity was also illustrated by the results in the maximal convulsion threshold tests. Although the drug raised this
714
R. KARLER et al.
threshold for bicuculline and N-methyl-DL-aspartate, it did not affect the threshold for either strychnine or aminophylline. This relative selectivity in the various chemical tests points to possible mechanisms of action for the anticonvulsant activity of cocaine: because the drug failed to elevate minimal convulsive thresholds for the y -aminobutyric acid (GABA) antagonist, bicuculline, the cholinergic agonist, arecoline and the adenosine antagonist, aminophylline, these various potential mechanisms do not appear to be involved in the anticonvulsant action of cocaine. The only minimal convulsion threshold affected by cocaine was the threshold for N-methyl-DL-aspartate, which suggests that this subtype of glutamate receptor is either directly or indirectly affected by the action of cocaine. Such selectivity did not extend to the maximal convulsion tests: in this case, cocaine was anticonvulsant against both the GABA antagonist and the glutamate agonist; the activity did not, however, include the glycine antagonist, strychnine, or the adenosine antagonist, aminophylline. The ability of the drug to block maximal convulsions, originating from both the excitatory and inhibitory components of the CNS contrasts with its more selective action against minimal convulsions induced by N-methyl-DL-aspartate. These results were repeated in other experiments and the reason for the difference in selectivity between the minimal and maximal convulsions is not understood. The haloperidol-cocaine studies illustrate another facet of the mechanism of action of cocaine (Table 2). These studies were designed to determine if the cocaine was a activity of anticonvulsant dopaminergically-mediated effect. Because of the pharmacological similarities between cocaine and both amphetamine and tetracaine, these latter two drugs were included in the study with haloperidol for comparative purposes. The results indicate that haloperidol could block the locomotor-stimulant effect of both cocaine and amphetamine and could block the anticonvulsant effect of amphetamine, but not that of either cocaine or tetracaine. The failure of haloperidol to block the anticonvulsant effects of cocaine in mice is consistent with similar previous observations in the rat (Russell and Stripling, 1985). The data suggest that the anticonvulsant activity of amphetamine involves a dopamine system, but that of cocaine some other mechanism related, possibly, to its local-anesthetic activity. The data from the various tests with chemically-induced convulsions,
described above, indicate that the drug may be anticonvulsant, at least in part, by virtue of an action affecting the N-methyl-DL-asparatate component of the glutamate system. work was supported by NIDA
Acknowledgement-This
research grant DA-00346. REFERENCES
Burright R. G., Donovick P. J., Michels K., Fanelli R. J. and Dolinskv Z. (1982) . , Effect of amohetamine and cocaine on seizure on lead treated mice. Pharmac. Biothem. Behav. 16: 631-635.
Covino B. G. and Vassallo H. G. (1975) Local Anesthetics. Grune and Stratton, New York. Finkle B. S. and McCloskey K. L. (1977) The forensic toxicology of cocaine. In: Cocaine: 1977 (Peterson R. C., Ed.), NIDA Research Monograph 13. U.S. Govt. Printing Office, Washington, D.C. _ Finnev D. J. (1971) Probit Analvsis. 3rd Edn. Cambridge Umversity Press,‘Cambridge. . Greer C. A. and Alpern H. P. (1979) Maturational changes related to dopamine in the effects of d-amphetamine, cocaine, nicotine, and strychnine on seizure susceptibility. _ _ Psychopharmacology
64-255-260.
Litchfield J. T. Jr and Wilcoxon F. (1949) A simolified method for evaluating dose-effect experiments. J. ‘Pharmat. exp. Ther. %: 99113. Post R. M. and Kopanda R. T. (1975) Cocaine, kindling, and reverse tolerance. Luncet i: 409410. Post R. M., Weiss S. R. B., Pert A. and Uhde T. W. (1987) Chronic cocaine administration: Sensitization and kindling effects. In: Cocaine: Clinical and Biobehavioral Aspects (Fisher S., Raskin A. and Uhlenhuth E. H., Eds) Oxford University Press, New York. Russell R. D. and Stripling J. S. (1985) Monoaminergic and local anesthetic components of cocaine’s effects on kindled seizure expression. Pharmac. Biochem. Behav. 22: 427-434.
Sangdee P., Turkanis S. A. and Karler R. (1982) Kindlinglike effect induced by repeated cornea1 electroshock in mice. Epilepsia 23: 471-479. Siegel S. (1956) Nonparametric Statistics. McGraw-Hill, New York. Snedecor G. W. and Cochran W. G. (1967) In: Statistical Methods, 6th Edn. Ames, Iowa. Sniegal M. R. (1961) Theorv and Problems of Statistics. ‘M&aw-Hill,X New York.. Steel R. G. D. (1959) A multiple comparison rank sum test: treatments versus control. Biometrics 15: 560-572. Stripling J. S. (1983) Cocaine and “pharmacological kindling” in the rat..Expl Neural. 82: 499-503. Strinlina J. S. and Hendricks C. (1981) Facilitation of k;ndlLg by convulsions induced by cocaine or lidocaine but not pentylenetetrazol. Pharmac. Biochem. Behav. 15: 793-798. Stripling J. S. and Russell R. D. (1985) Effect of cocaine and pentylenetetrazol on cortical kindling. Pharmac. Biochem. Behav. 23: 573-581.