Psychopharmacologia (Berl.) 44, 241- 246 (1975) - © by Springer-Verlag 1975

Generalization of Behaviorally Augmented Tolerance to Ethanol, and Its Relation to Physical Dependence A. E. LEBLANC, R. J. GIBBINS, and H. KALANT Addiction Research Foundation of Ontario and Department of Pharmacology, University of Toronto Received May 12, 1975

Abstract. Adult rats, required to perform a motor coordination task while under the influence of ethanol (2.2 g/kg) daily for 12 days, developed significant tolerance to the effects of ethanol, not only on this test but also on a food-motivated maze task. At the same time, they showed reduced shock threshold for production of a startle response, and reduced open field test scores 23 hrs after the last dose of ethanol. Other animals receiving the same dose of ethanol immediately

after each training session on the coordination task, and others performing the task and receiving no ethanol, showed no tolerance on either test and no change in shock threshold or open field behaxdor. The results suggest that behaviorally augmented tolerance depends on basic neuronal adaptive changes indistinguishable from those accompanying physiological tolerance and dependence.

Key words: Behavior - Ethanol - Tolerance - Generalization - Dependence.

The rate of development of tolerance to ethanol and various other drugs has been shown to depend not only on the degree of drug exposure, but also on the circumstances under which it is given (Kalant et at., 1971). I f alcohol exposure is held constant, the rate of development of tolerance increases as a function of the frequency of task performance under the influence of alcohol (LeBlanc, I972; LeBlanc et al., 1973). We have referred to this phenomenon as behavioral augmentation of tolerance (LeBlanc et at., 1973). I f behaviorally augmented tolerance is identical with tolerance produced by more conventional pharmacological techniques of administration of large doses, rather than a task-specific form of learning, then it should generalize to unrelated tasks for which no transfer o f learning is demonstrable from that used in the development of tolerance. Some evidence in support o f this prediction is provided by studies of the development of tolerance to barbiturates in the rat, as measured by a change in the threshold dose required to produce one-second silence in the electroencephalogram (WahlstrSm, 1968). Rats treated chronically with small doses of barbiturates during their normal waking hours became tolerant, while animals receiving the same dosage during their sleeping hours did not. In addition, it has been noted repeatedly that the development of tolerance to ethanol and numerous other drugs is accompanied by signs of physical dependence, as indicated by a withdrawal syndrome

on suspension of drug treatment (Kalant et al., 1971). One might, therefore, expect behaviorally augmented tolerance to be attended by signs of physical dependence also. The present report presents experimental evidence supporting both these predictions.

Methods SuNects for this study were male Wistar rats (Woodlyn Farms, Guelph, Ontario) obtained at an initial weight of 80-100 g. Each animal was given a distinctive code number. They were housed singly, with standard laboratory chow and tap water continuously available except that when the controls reached a weight of 300 g, they were kept at this weight by appropriate rationing of chow. Animals trained on the circular maze (see below) were subjected to stricter rationing to hold their growth rate and final body weight at 80 ~ of those of the controls. The moving belt test (Gibbins et al., 1968) was employed in this study in a modified form incorporating a number of technical improvements (LeBlanc, 1972). In this test animals are obliged to remain on a motor-driven belt which moves continuously over a shock-grid. If the animal puts one or more paws on the grid it receives shock and activates a cumulative timer. The effect of ethanol, pentobarbital and other drugs is seen as a monotonic dose-dependent increase in time off belt. The criterion level of performance for fuUy trained rats is less than 1 ~o (J .2 sec) off belt during a standard 2-rain test, on each of six tests in a session. The circular maze (Chen, 1967) consists of a circular runway of about 45 cm path length, with a. straight side arm running radially outward from it for a distance of 51 cm. At the far end of the arm is a Noyes food pellet dispenser triggered by an electric eye. The animals are trained to move from the end of the side arm to the circulm runway, proceed twice

242

Psychopharmacologia (Bed.), Vol. 44, Fasc. 3 (1975) Type R Dynogroph

I Ch.7 Ch.8 ,

II

Perforated film programmer

Stepper

Input Couplers

Shock generator and control unit

I

Electrocraft movement sensing unlt model MTIO0

]

Pre-

amplifiers

~Po

"~ Scrambler

2 8 V OC Power suppl

Ch.8 Dynograph

p e n s /~~

,Event

"~ markerI Fig. 1, Schematic diagram of automated startle response threshold apparatus. For detailed description, see text

around the circuit in a counter-clockwise direction, and then return to the end of the side arm to obtain a food reward. This constituted the only correct response. Any other excursion past the junction of side arm and the circular runway, followed by a return to the food dispenser, constituted an incorrect response. The criterion of training is 90 ~ correct responses during six successive 2-min trials. The animal was removed from the maze at the end of each 2-min trial, no matter what point it had reached in its run. Startle responses to electric foot shock were measured with an automated version of an apparatus described previously (Gibbins et al., 1971). Startle responses were detected by an Electrocraft model MTI00 movement transducer located in the roof of the test chamber. A Beckman Dynograph pen recorder, type R, was used for recording the time and intensity of each shock administered, as well as the detected response. A schematic diagram of the apparatus is shown in Fig. 1; further details have been given elsewhere (LeBlanc, 1972). In a preliminary experiment, the original technique of visual scoring and the new automated technique showed a correlation coefficient of 0.94 to 0.97, depending on the person doing the visual scoring. The new apparatus did not materially alter the technique as originally described, except to make it more convenient. The open field test is a commonly used measure of emotionality in rodentS. It was carried out,with an apparatus modelled on that described by Eysenck and Broadhurst (1964). Bolus counts were ignored because the test animals were kept on restricted diet as described in the experimental section, and had very low fecal output at all times. Activity scores were obtained as recommended by Eysenck and Broadhurst. Cross-Learning Experiment Three groups of 16 male Wistar strain rats per group were used in this experiment. One group served as body weight control. The other two groups received restricted rations and were used as the experimental groups. One experimental

group was trained on the moving belt test and the other on the maze. When 14 of the 16 animals in each group had reached the respective criterion for the test concerned, training on the alternate test was begun, while daily maintenance sessions on the first task were continued. The experiment was terminated when 12 out of 14 animals in each group reached criterion on the second task. This procedure was followed because in the subsequent experiments it was intended to study the ethanol effect on performance of both tests by the same animals. For this purpose it was necessary to have both performances at pre-drug criterion level simultaneously, and the present experiment was conducted to test the feasibility of the procedure. The data were calculated for the 12 animals in each group which had been successfully trained on both tests. M e a s u r e m e n t o f A l c o h o l W i t h d r a w a l Effect o n O p e n Field Test Four groups of 6 animals per group were randomly drawn from a pool of 111 animals previously trained to criterion on the moving belt test. All groups were tested in the open field for 4 min. A dose of ethanol (2.2 g/kg i.p., as an 8 ~ w/v solution in isotonic saline) was then administered and the moving belt test was carried out. The chronic treatments for the various groups were started. Group 1 was designated t h e alcohol group; Group 2, the water controls ; Group 3, sucrose controls; and Goup 4, handling controls. Alcohol was administered by stomach tube in a dose of 6 g/kg, as a 20 w/v solution in tap water. Water was given in equal volume, and sucrose in an equicaloric dose of equal volume. Group 4 were handled daily but received no fluid by intubation. These treatments were continued for a period of 2 weeks. Twentyfour hours after the final gavage, all rats were again tested on the moving belt apparatus after a dose of ethanol (2.2 g/kg i.p.). Twenty-four hours later they were again tested in the open field.

A. E. LeBlanc et aI. : Generalization of Behaviorally Augmented Tolerance to Ethanol A

243 B

84

I00

50

o

k.J #J r

30

80

~-

6O

o w n~ nO

t~ 0

z

0 o. o~

u

ZO

o

L~

40

l-Z W

I--

I0

0

o

n. w D.

~

,

,

S I 82834

20

I_

6

8

to

~2

14

is

~8

sfszs34

6

8

t'o

1'2

m'4'" 16

18

DAYS Fig. 2, Learning curves for moving belt test (A) and circular maze test (B) for animals sequentially trained on the two tasks. The open circles indicate animals trained on the moving belt test first and then on the maze; the solid triangles indicate the reverse. S~ _ a indicate unscored shaping days. Training days were consecutive

Test for Behavioral A u g m e n t a t i o n o f E t h a n o l Effect Fifteen rats were trained first on the moving belt test and then on the circular maze, by the procedm'es for the cross-learning experiment. The maximum body weight was held at 240 g for all animals during the experiment. The animals were assigned individual numbers, and were all tested in random order in the open field and then on the startle response apparatus. The next day, they were all tested under ethanol (2.2 g/kg i.p.), again in random order, on the moving belt apparatus, Each animal was given six consecutive 2-rain trials, beginning 2.5 min after the ethanol injection and separated from each other by 2.5-min rest intervals. The alcoholinduced impairment was measured by the maximum time off belt during any single trial; this occurred most commonly during the second or third trial. Twerity-four hours after the first alcohol test the rats were tested on the maze, under ethanol (1.4 g/kg i.p.). Each animal was given six 2-rain trial periods, on the same schedule as for the moving belt test. The alcohol effect was measured by the change in percentage of correct runs during the total six trial periods. With the dose of ethanol used in this work, there was no effect,on the running speed or on the total number of runs completed during each trial period (average about 10 runs per 2 min). The animals were ranked on the basis of their scores on the moving belt test, and successive groups of 3 were randomly assigned to three treatmem schedules, designated as Group 1, 2 or 3. On each of i2 consecutive days they received stabilization runs on the maze, consisting of a 4-min warmup and six 2-min training trials. Immediately on completion of this, the animals were given six 2-rain trials on the moving belt test, preceded and tbllowed by an i.p, injection of either saline or ethanol. Group 1 received saline for both injections. Group 2 received saline before training and alcohol (2.2 g/kg) after, and Group 3 received the reverse sequence. At the end of tile 12 days, they were again tested on the moving belt apparatus under a 2.2 g/kg dose of ethanol. Twenty-four hours later they were re-tested on the maze under

a dose of 1.4 g/kg. Twenty-three hours after this dose, they were tested in the open field and the startle response apparatus.

Results Cross-Learning Experiment The acquisition curves on the m o v i n g belt test and the maze test in the two g r o u p s are shown in Fig. 2. As can be seen, the curves for the two groups o n each test are completely superimposable, regardless o f the order in which the tests were learned. Therefore, it can be concluded that there was n o cross-learning f r o m one test to the other. Effect o f A l c o h o l W i t h d r a w a l State o n O p e n Field Behavior The results o f the second preliminary experiment are s h o w n in Fig. 3. As expected, only G r o u p 1 developed significant tolerance (t --- 17.71, P < 0.001), as shown by a reduction in the error score on the m o v i n g belt test at the end o f the chronic treatment period. The other three groups showed n o change. In the parallel tests on the open field, G r o u p 1 again showed a striking reduction in score (t = 11.86, P < 0.001), while the other three groups were n o t significantly changed. Previous w o r k had shown that physical dependence, as defined by withdrawalinduced lowering o f f o o t shock threshold fbr the startle response, developed concurrently with tolerance to e t h a n o l (LeBlanc et aL, 1969; Gibbins eta/.,

244

Psychopharmacologia (Berl.), Vol. 44, Fasc. 3 (1975) A 80 -

I00

-.[,

T

~

80

7"7

r //',

fl

~'/

I

g

/I

// //

60

// 50

S=

//,,

o= " ~ w

-

30

--

, ,'///I

r

'.,';"

g" =~~

// ,..,

//

// t/

///I

,,

v,,

l/

,,

I 0

!

IROUPI

//

/.' v., // //

,

GROUP2

70

//

I/l

I-

20

.



///

//

s~ M 40 w l.u z 30 ,,,

O-

0 20

GROUP3

//

GROUP4

/,,

7/ /1

//

//

z

,,/

//

IT

~

//

t/

II

""

1

90

0

I/

GROUPI

GROUP2

/I

/I

"J

GROUP3

GROUP4

Fig. 3. Effect of ethanol on moving belt test (A) and open field counts (B) for four groups of animals on different chronic treatment schedules. The left bar of each pair gives the pre-treatment values; the right, the post-treatment. Vertical bars indicate S.E.M. Group I received ethanol treatment (6 g/kg daily) for 14 days; Group 2, water; Group 3, sucrose (10 g/kg daily); Group 4, no treatment

1971). Since emotional lability is one of the symptoms of alcohol withdrawal, the present results in the open field test, 23 hrs after the end of alcohol treatment, are also consistent with the presence of withdrawal symptoms.

field score on re-testing, presumably because of loss of novelty. However, the reduction in Group 3 (81% decrease) was much greater than in Groups 1 and 2 (17 and 19%, respectively), and this difference in degree of change was highly significant (P < 0.001 in both cases).

Test for Behavioral Augmentation of Ethanol Tolerance

Discussion

The results of the four tests before and after treatment in each group are shown in Figs. 4 and 5. Group 3, which received alcohol before each daily session on the moving belt apparatus, developed tolerance as expected (t = 4.39, P < 0.01). Group 2, which received the same total dose of ethanol after the sessions, and Group I, which received no ethanol, did ffot become tolerant (Fig.4). These findings confirmed the earlier observations related to the behavioral augmentation of tolerance (LeBlanc et al., 1973). At the same time, Group 3 also developed tolerance as shown by an increase in the accuracy of performance of the maze test under alcohol (t = 14.41, P < 0.001). The other two groups showed no change in maze performance. Group 3 showed a marked reduction in the startle threshold at the end of chronic treatment, as compared with the pre-treatment scores (t = 3.76, P < 0.01), while Groups 1 and 2 showed no significant change (Fig.5). On the open field test all groups showed significant reductions from pre- to post-treatment scores (P < 0.005 in all cases); this is not surprising, since all animals tend to show a reduction in open

The results of the main experiment (Fig. 4) confirm the earlier finding (Chen, 1968; LeBlanc et al., 1973) that tolerance to ethanol develops more rapidly in animals which are repeatedly required to perform a training procedure under the influence of ethanol than in those which receive the ethanol after the training sessions. Comparable observations by other investigators, using other drugs, have been cited elsewhere (Kalant et al., 1971). In addition, further support has recently been reported on the basis of studies of tolerance to tetrahydrocannabinol (Carder and Olson, 1973). We have shown previously that this "behaviorally augmented tolerance" (BAT) can not be explained by change in the disposition of ethanol, as reflected by the concentration in the blood at the time of behavioral testing (LeBlanc et al., 1973). The blood samples tested in that work were obtained from the cut tip of the tail, and were therefore essentially arteriolar and capillary blood. We have recently reported (LeBlanc et al., 1975) that ethanol concentrations in arterial blood and brain did not differ significantly, when measured as soon as 30 rain after an i.p.

A. E. LeBlanc

et al.

: Generalization of Behaviorally Augmented Tolerance to Ethanol A

245 B

70 I00 60

90

8O

r LLt 5 0 o3

70

~,6o

40 LLI n"t

e~

D

LL Z,O LL 0 LU

50

~4o

20

I'

30

I--

20

Io IOi

0 GROUP I

GROUP

2

GROUP

3

GROUP

I

GROUP

2

GROUP

3

Fig. 4. Effect of ethanol on moving belt test (A) and circular maze performance (B), before and after chronic treatment. For each group, the left-hand column indicates the pre-treatment, the right-hand column the post-treatment, score. Vertical bars indicate S.E.M. As the chronic treatment, Group i received saline injections before and after each training session on the moving belt test, daily for 12 days; Group 2 received saline before and ethanol (2.2 g/kg i.p.) after; Group 3 received ethanol before and saline after A

B

120

0.4

110

I00 90 o3 I.-z

0-3

80

I

[

o

E

70

c3 ._1

0 g:l -J LU LL

TI

60

o 0-2

50

o3 u3

40

='t" F-

ere

LU

30

.J

0

0.1

I--,

20 I,o3

tO 0 GROUP

I

GROUP

2

GROUP

3

GROUP

1

GROUP

2

GROUP

3

Fig. 5. Open field scores and startle thresholds measured in the absence of ethanol, before and 23 hrs after chronic treatment. Left-hand column of each pair, pre-treatment; right-hand column, post-treatment. Vertical bars indicate S.E.M. Chronic treatments were the same as for Fig. 4

injection of 1.9 g/kg. Although a reduction in delivery of ethanol to the brain, during the first few minutes after injection, can not be ruled out, it seems more probable that BAT represents a true change in sensitivity of the central nervous system to ethanol. The present work indicates that BAT is not a task-specific learned compensation for impairment of particular motor or sensory capabilities. If maze

performance depended upon the same skills as those required for the moving belt test, prior learning of one test should have facilitated learning of the other. This was not the case (Fig. 2). Therefore, the transfer of BAT from the moving belt test to the maze suggests that BAT, like conventionally developed tolerance, depends on some basic neuronal adaptive change in the central nervous system.

246 Reduction in startle threshold has been shown previously to be a useful index of physical dependence on ethanol (Gibbins et al., 1971). The finding of reduced threshold after 23 hrs of alcohol deprivation in the present study also confirms our previous finding that BAT, like conventionally produced tolerance to ethanol, is accompanied by signs of physical dependence (LeBlanc et al., 1973). Exploratory behavior in the open field test is known to be reduced by emotionality or by excessive sensory input. Therefore, it is not surprising that the animals in Group 3, which showed reduced startle threshold, also showed marked reduction in open field score (Fig. 5). Since exploratory behavior in the open field shows progressive and rapid reduction on repeat testing, this is probably not a suitable index for following the course of development of physical dependence on alcohol. However, the single re-test used in the present experiment was able to differentiate among the treatment groups in the same way as the evidence of physical dependence in the tolerant animals. These findings, in conjunction with those reported earlier, are consistent with the hypothesis that neuronal adaptive changes, responsible for tolerance to ethanol or other drugs, are influenced by the functional demand imposed upon the central nervous system during the period of drug effect. The stimulus to development of tolerance, according to this hypothesis, would not be the presence of the drug itself, but the degree of impairment of required neuronal functions which the drug produces. The present findings support the earlier conclusion (LeBlanc et al., 1973) that BAT is indistinguishable from conventionally defined physiological tolerance when analysed by the present methods. However, they do not wholly exclude the possibility that there might be two different processes involved. It will be possible to establish the presence or absence of a

Psychopharmacologia (Berl.), Vol. 44, Fasc. 3 (1975) difference only by the use of a finer grained kinetic analysis, or clarification of the cellular mechanism involved in both.

Acknowledgments. The authors are grateful to L. Currin, C. E. Fahlgren, and A. Wilson for their technical assistance. References Carder, B., Olson, J. : Learned behavioral tolerance to marihuana in rats. Pharmacol. Bioehem. Behav. 1, 73-76 (1973) Chen, C.-S. : A study of the alcohol tolerance effect and an introduction of a new behavioral technique. Psychopharmacologia (Berl.) 12, 433-440 (1968) Eysenck, H. J., Broadhurst, P. L. : In : Experiments in motivation, H. J. Eysenck (ed.), p. 285. New York: MacMillan 1964 Gibbins, R. J., Kalant, H., LeBlanc, A. E. : Technique for accurate measurement of moderate degrees of alcohol intoxication in small animals. J. Pharmacol. exp. Ther. 159, 236-242 (1968) Gibbins, R. J., Kalant, H., LeBlanc, A. E., Clark, J. W. : The effects of chronic administration of ethanol on startle thresholds in rats. Psychopharmacologia (Berl.) 19, 95-104 (1971) Kalant, H., LeBlanc, A. E., Gibbins, R. J. : Tolerance to, and dependence on, some non-opiate psychotropic drugs. Pharmacol. Rev. 23, 135-191 (1971) LeBlanc, A. E. : Behavioral and pharmacological variables in the development of ethanol tolerance. Ph.D.Thesis, University of Toronto, Toronto, Canada (1972) LeBlanc, A. E., Gibbins, R. J., Kalant, H. : Behavioral augmentation of tolerance to ethanol in the rat. Psychopharmacologia (Berl.) 30, 117-122 (1973) LeBlanc, A. E., Kalant, H., Gibbins, R. J. : Acute tolerance to ethanol in the rat. Psychopharmacologia (Berl.) 41, 43 - 46 (1975) LeBlanc, A. E., Kalant, H., Gibbins, R. J., Berman, N. D. : Acquisition and loss of tolerance to ethanol by the rat. J. Pharmacol. exp. Ther. 168, 244-250 (1969) Wahlstr6m, G.: Differences in tolerance to hexobarbital (enhexymalum NFN) after barbital (diemalum NFN) pre-treatment during activity or rest. Acta pharmacol. (Kbh.) 26, 92-104 (1968)

Dr. A. E. LeBlanc, Addiction Research Foundation, 33 Russell Street, Toronto, Ontario, Canada M5S 2S1

Generalization of behaviorally augmented tolerance to ethanol, and its relation to physical dependence.

Adult rats, required to perform a motor coordination task while under the influence of ethanol (2.2 g/kg) daily for 12 days, developed significant tol...
619KB Sizes 0 Downloads 0 Views