Decreased experimental anxiety and voluntary ethanol consumption in rats following central but not basolateral amygdala lesions

Brain Research 760 Ž1997. 94–101

Research report

Decreased experimental anxiety and voluntary ethanol consumption in rats following central but not basolateral amygdala lesions Christian Moller, Lisa Wiklund, Wolfgang Sommer, Annika Thorsell, Markus Heilig ¨

)

Magnus Huss Clinic, Karolinska Hospital,Stockholm Sweden Accepted 18 February 1997

Abstract A long-debated ‘tension reduction’ hypothesis postulates anti-anxiety effects to be important for ethanol reward, and states that elevated anxiety levels might predispose for ethanol consumption and addiction. Human data are contradictory, possibly due to heterogeneity of patient samples. In rats, baseline levels of experimental anxiety have been reported to correlate with voluntary ethanol consumption. Here, we addressed the possibility that mechanisms underlying experimental anxiety might be causally related to regulation of voluntary ethanol intake. Rats were bilaterally lesioned in central amygdala using microinjections of ibotenic acid. This resulted in a robust release of punished drinking in a modified Vogel conflict test, an effect typically seen with anxiety reducing drugs. This effect was specific, as unpunished drinking was unaffected by the lesion. On the elevated plus-maze, central amygdala lesions did not affect behaviour under baseline conditions, but attenuated the anxiogenic effect of restraint stress. Measures of locomotor activity were not affected. Voluntary ethanol consumption was examined in a two-bottle, free choice paradigm. Ethanol intake was markedly decreased in the lesion group. Total fluid intake was not affected. Basolateral amygdala lesions, which did not affect conflict behaviour, also left ethanol intake unaffected. These results are consistent with previous reports of an important role for central amygdala in anxiety related behaviours, and suggest that cell bodies located in central amygdala might be important in this context. Further, our results support a relation between experimental anxiety and voluntary ethanol consumption. Keywords: Anxiety; Alcoholism; Amygdala; Stress; Emotionality

1. Introduction Specific reinforcing properties have been shown to initiate and maintain self administration of addictive drugs in experimental animals. For major classes of addictive substances such as central stimulants and opiates, consistent evidence points to the mesolimbic dopamine ŽDA. pathway as a critical substrate of reinforcement. However, the pharmacodynamic actions of ethanol are much more diverse w20x. The relative contribution of different components within ethanol’s complex effect spectrum to the reinforcement produced by this drug is not well established. Although ethanol is clearly capable of affecting mesolimbic DA transmission in a manner similar to that described for other addictive drugs w5,19,45x, this action of the drug does not appear to be necessary for ethanol self-administration in rats w34x.

) Corresponding author. [email protected]

Fax:

q 46

Ž8 .

326369;

e-mail:

A clinically prominent acute action of ethanol is to suppress anxiety. In experimental animals, ethanol potentiates GABA-ergic transmission w1x, and reduces experimental anxiety w4,26,41x. A long held ‘tension reduction’ or ‘stress reduction’ hypothesis w7,32x states that elevated levels of anxiety or stress might increase the reinforcement value of ethanol, and predispose an individual to increasing voluntary ethanol consumption and developing addiction. This hypothesis has been highly controversial. Epidemiological associations between certain anxiety states and ethanol addiction have been reported w8,35x, but this type of finding is complicated by the fact that ethanol use in itself frequently produces symptoms of anxiety. This has led leading workers in the field to question a ‘true’ association between anxiety disorders and alcoholism w38x. The issue is further complicated by the fact that clinical populations of alcoholics are heterogeneous, and a differential association between anxiety and alcoholism might be expected in different subgroups of alcoholics on the basis of this heterogeneity w9x. Recent studies have examined the relation between

0006-8993r97r$17.00 Copyright q 1997 Elsevier Science B.V. All rights reserved. PII S 0 0 0 6 - 8 9 9 3 Ž 9 7 . 0 0 3 0 8 - 9

C. Moller ¨ et al.r Brain Research 760 (1997) 94–101

anxiety and voluntary ethanol consumption in experimental animals. Using heterogeneous rats, it was recently reported that individual differences in experimental anxiety constituted a powerful predictor of voluntary ethanol consumption in a free-choice, two-bottle procedure: individuals within the lowest ‘anxiety’ quartile with regard to behaviour on the elevated plus-maze consumed significantly less ethanol than those within the uppermost quartile w40x. A causal relation between experimental anxiety and ethanol consumption would predict that a manipulation leading to decreased levels of the former would also decrease the latter. The amygdala complex, and particularly central amygdala, is a critical component of brain networks mediating fear and anxiety w13,16,23,27x. Lesions of this region have i.a. been reported to result in reduced experimental anxiety as measured in a conflict test based on fear-suppression of behaviour w22x. In the present study, we therefore produced ibotenic acid lesions of central amygdala, and examined the consequences for experimental anxiety and voluntary ethanol consumption in heterogeneous male rats. For comparison, effects were also studied of lesions in basolateral amygdala, a structure for which a different role in anxiety mechanisms has been postulated w23,24x.

95

surface; for basolateral amygdala: 2.4 mm posterior and 4.6 mm lateral to bregma, 8.8 mm ventral to skull surface. Sham operated controls were subjected to identical procedures, but the injection needle was only lowered to the dura mater. Animals were allowed a 10-day recovery period prior to continued experiments, and were handled during that time. The general well being of all subjects was carefully monitored postoperatively. 2.3. Histology At the end of experiments, animals were decapitated. Brains were removed, cut into blocks using a Kopf Brain Slicer, frozen using isopentane Žy408C., and stored at y708C. Blocks were mounted on a cryostat, and 30 mm coronal cryosections were obtained at the level of the lesions under guidance of an atlas w29x. Sections were Nissl stained, and examined for extent of lesions by an experienced observer unaware of the behavioural results. 2.4. Punished drinking test

Male SD rats ŽBee Kay, Sollentuna, Sweden, 225–250 g. were kept in individual cages under controlled light-dark conditions Žlight on at 07.00 h and off at 19.00 h., at a controlled temperature Ž21–238C.. Standard laboratory chow and tap water were available ad libitum, except during the Vogel test. Subjects were allowed to adapt to the animal facility for at least 7 days prior to any procedures. Two separate sets of experiments were performed, one for central amygdala lesions, and one for basolateral lesions. In each of these, lesioned and sham operated animals Ž12 in each group. were operated and tested in parallel.

A modified Vogel’s drinking test was used w43x. Following 24 h water deprivation, animals were adapted for 12 min to an operant chamber equipped with a grid floor of stainless steel bars and a drinking bottle containing 5% Žwrv. glucose solution. After an additional 24 h period of water deprivation, animals were adapted once more to the same test chamber for a further 12 min during which free access was given to the glucose solution. Finally, after a further 24 h period of water deprivation, animals were returned to the operant chamber for the testing session. Behavioural testing was under the control of a computerized system ŽMedAssociates, St. Albans, VT.. Contact with the drinking spout was detected electronically. To establish drinking behaviour, subjects were allowed three drinking episodes before the timer started. This was followed by a 4-min unpunished drinking component, during which drinking episodes were detected, but no shock was administered. Finally, an 8-min conflict component followed, during which electric shock Ž0.24 mA, 2 s. was delivered upon every drinking episode, and the number of punished drinking episodes was recorded.

2.2. Surgery

2.5. EleÕated plus-maze

Animals were anaesthetized with ketaminerxylazine Ž0.1 mgrkg resp. 0.5 mgrkg. and placed in a stereotactic apparatus. Bilateral lesions were obtained by microinjecting ibotenic acid Ž10 mgrml; 0.1 mlrmin; injection volume 0.2 ml. through a 31 G needle using a Hamilton syringe fitted into a Kopf microinjection pump. Following injection, the needle was left in place for an additional two minutes. Tooth bar was at 3.3 mm below interaural line. For central amygdala, coordinates were: 2.3 mm posterior and 4.3 mm lateral to bregma, 8.3 mm ventral to skull

To further examine effects of central amygdala lesions observed in the Vogel test, elevated plus maze testing was undertaken with these subjects. Plus-maze testing was essentially as originally described w30,31x. The apparatus was made of black plastic with a rubber floor and consisted of two open arms measuring 50 = 10 cm, and two arms which were of the same size but had 40 cm high end and side walls. The arms were connected by a central area measuring 10 = 10 cm and the maze was 50 cm above the floor. Testing was under dimmed red light. Behaviour was

2. Materials and methods 2.1. Subjects

96

C. Moller ¨ et al.r Brain Research 760 (1997) 94–101

scored by a trained observer blind to treatment conditions. Prior to testing, subjects were placed in a novel environment Žan empty operant chamber. for 5 min; this has been shown to increase exploratory activity and improve the reliability of subsequent plus-maze testing. Rats were then placed in the central area of the maze facing one of the open arms. The number of entries into open and closed arms and the time spent in open and closed arms were recorded over 5 min testing. An arm entry was defined as all four paws into an arm. Animals were tested once under baseline conditions, and then re-tested one week later, at which time plus maze testing was performed 1 h following the completion of a 1 h restraint stress. In agreement with the original description of the model w30x, work in our laboratory has shown plus maze behaviour to be stable with regard to both percent open-arm time and entries in male Sprague Dawley rats over at least three repeated sessions under the testing conditions used, while the total number of entries, an overall activity parameter, slightly decreases between session 1 and 2 ŽHeilig, unpublished results.. 2.6. Ethanol preference procedure A modification of the procedure used by Spanagel et al. w40x was used. Animals were placed in individual cages, and two bottles containing water or increasing concentrations of ethanol were made available continuously as a free choice to the animals. Ethanol concentration was increased

every 5th day as follows: day 1–5: 2%; day 6–10: 4%; and day 11–15: 6% Žvrv solutions.. Taste aversion was avoided by adding 0.2% saccharin Žwrv. to both the water and the ethanol solutions. Bottle positions were changed randomly daily so that no position habit would develop. Consumption of ethanol and water was measured daily at the same time. Body weights were recorded daily.

3. Results All subjects recovered successfully from surgery, and within 3–5 postoperative days no general signs of ill-being were seen in any subject. The inclusion of the intended target area within the ibotenic acid lesion was confirmed histologically in 9r12 central amygdala lesioned subjects. The remaining three subjects were excluded from analysis. The extent of central amygdala lesions in the animals used for analysis is shown in Fig. 1 Žleft panel.. Basolateral lesions were verified in 8r12 subjects, and the extent of lesions in these subjects is shown in Fig. 1 Žright panel.. Only these eight subjects were included in the analysis of functional results. In the Vogel punished drinking paradigm one way ANOVA with respect to treatment was separately performed for unpunished and punished drinking, respectively. Unpunished drinking was virtually identical in central amygdala lesioned animals and sham controls Ž F1,19 s 0.02; P s 0.88, n.s.; Fig. 2.. In contrast, a robust release of

Fig. 1. Left: inclusion of the central nucleus of the amygdala within the lesion was confirmed histologically in 9r12 animals. Only these animals were used for data analysis. Extent of lesions in these subjects is shown. Solid area indicates minimum, and hatched area maximum extent of lesion. Right: histological confirmation of basolateral lesions. 8r12 animals were retained following histological analysis. Solid area indicates minimum, and hatched area maximum extent of lesion in the included subjects.

C. Moller ¨ et al.r Brain Research 760 (1997) 94–101

97

Table 1 Significant suppression of open arm exploration Ž% openrŽopenqclosed.; mean"S.E.M.; see Section 3. by restraint stress in sham-operated controls, but not in central amygdala-lesioned animals

TimeŽ%. Control Lesioned EntriesŽ%. Control Lesioned EntriesŽnr. Control Lesioned Fig. 2. Unaffected unpunished drinking, and significant release of punished drinking in central amygdala lesioned rats Ž P - 0.01; for statistics, see Section 3.. Number of drinking episodes is shown Žmean"S.E.M.; ns12 and ns9 for controls and for lesioned subjects, respectively. during a 4-min unpunished, and an 8-min conflict component, during which each drinking episode resulted in a 0.24 mA shock between the drinking spout and the grid floor.

punished drinking was seen in the lesion group Ž F1,19 s 8.1; P - 0.01; Fig. 2.. The magnitude of this effect, an approximate doubling of punished drinking, was equivalent to that seen with established anti-anxiety compounds such as diazepam under these conditions Žnot shown.. Basolateral amygdala lesions affected neither unpunished nor punished drinking Žunpunished: 26.8 " 3.2 vs. 33.0 " 6.6; F1,18 s 0.88, P s 0.36, n.s.; conflict: 20.9 " 4.2 vs. 17.8 " 1.6; F1,18 s 0.36, P s 0.56, n.s.; mean nr. of drinking episodes " S.E.M. for sham operated controls, and basolateral lesion group, respectively.. In the plus maze, repeated measures two-way ANOVA’s with respect to the lesion effect and the effect of restraint were performed separately for the percent time spent on and the percent entries made into the open alleys, as well as for the total number of entries; this was followed by Tukey’s HSD post hoc test to evaluate individual group comparisons. Overall, central amygdala lesions did not influence the anxiety related behavioural parameters Žpercent open arm time: F18 f 0.0, P s 0.99; percent open arm entries: F18 f 0.0, P s 0.96.. However, restraint stress significantly lowered the percent open arm time Ž F18 s 7.1, P s 0.016., and Tukey’s HSD test revealed that that this effect of restraint was present in the sham control group Ž P s 0.03., but was entirely absent in the lesion group Ž P s 0.82.. A similar trend for an effect of restraint was present in the less sensitive percent open arm entries measure Ž F18 s 3.5, P s 0.08.. Finally, locomotor activity as reflected in the total number of entries made during a session was also unaffected by central amygdala lesions Ž F18 s 0.02, P s 0.90., but was suppressed by restraint Ž F18 s 17.8, P - 0.001.. For this measure, however, the effect of restraint was similar in both the sham operated control Ž P s 0.02. and the central amygdala lesioned group Ž P s 0.06.. Plus maze data are summarized in Table 1.

Baseline

Restraint stress

53.6"6.1 47.9"6.5

33.8"5.6 42.5"6.2

45.8"4.0 44.2"4.4

36.4"5.1 41.9"3.5

18.6"1.3 18.6"1.0

14.1"1.1 14.8"1.4

P s 0.03 P s 0.82

The correlated but less sensitive corresponding measure for open arm entries showed a similar tendency, but was not significant. Total locomotor activity, as measured by the total number of entries made into any alley, was suppressed in a virtually identical manner in both groups. For detailed statistics, see Section 3.

Central amygdala lesions have been reported to affect food intake w14x. Over the course of the ethanol preference procedure, the weight of the animals was therefore monitored, and mean group weights are given in Table 2. A two way ANOVA with respect to time and treatment was performed on the daily weight values. Both groups significantly gained weight over the 15 days of the procedure Ž F14,285 s 15.2; P - 0.001.. Body weights of central amygdala lesioned animals were significantly lower than those of sham lesioned controls Ž F1,285 s 14.5; P - 0.001.. However, the two weight gain curves were essentially parallel, and there was no significant difference in weight gain over the 15 days of the self administration procedure Ž88.6 " 4.7 vs. 80.0 " 3.2 g, mean " S.E.M.; F1,22 s 2.0; P s 0.18, n.s... Due to the group difference in body weights, both ethanol consumption and total liquid intake in the two bottle free choice procedure were expressed per unit body weight. A two way ANOVA with respect to time and group effects was performed separately for ethanol consumption, and total liquid intake, respectively. Independent of group, there was a significant increase in the amount ethanol consumed over time Ž F14,285 s 1.8; P s 0.04.. Central amygdala lesions markedly decreased ethanol consumption Ž F1,285 s 11.2; P - 0.001; Fig. 3a.; during the final phase of the procedure, when a 6% ethanol

Table 2 Body weights Žmean"S.E.M.. of experimental groups at the start Žday 1. and end Žday 15. of the ethanol preference procedure

Central amygdala lesions Sham-operated controls Basolateral amygdala lesions Sham-operated controls

Day 1

Day 15

358.1"5.8 366.9"8.3 362.5"4.1 391.3"5.5

438.2"8.2 455.5"12.5 446.0"7.1 486.2"7.3

For detailed statistics, see Section 3.

98

C. Moller ¨ et al.r Brain Research 760 (1997) 94–101

F1,18 s 6.8; P s 0.018.. Ethanol and total fluid intake were corrected for body weights. There was a trend toward a decrease of total fluid consumption in basolateral lesioned subjects Žtreatment effect: F1,270 s 3.3; P s 0.07, n.s.; time effect: F14,270 s 3.8; P - 0.001.. Despite this fact, there was no decrease in ethanol intake in lesioned subjects, while ethanol intake significantly increased over time in both groups Žtreatment effect: F1,270 s 0.5; P s 0.47, n.s.; time effect: F14,270 s 2.1; P s 0.015..

4. Discussion

Fig. 3. a: decreased voluntary ethanol consumption in central amygdala lesioned rats Ž ns9. compared to sham operated controls Ž ns12; error bars represent S.E.M.; P - 0.001; for detailed statistics, see Section 3. over a 15-day step up procedure Žethanol concentrations available: day 1–5, 2%; 6–10, 4%; 11–15, 6%.. b: virtually identical total liquid intake in central amygdala lesioned subjects and sham operated controls ŽS.E.M. was for all data points between 7.2 and 20.1 mlrkg; P ) 0.98; for detailed statistics, see Section 3..

solution was available, intake was 2.3 " 0.2 and 1.4 " 0.2 Žgrkg and day; mean " S.E.M.. for control and lesioned animals, respectively ŽFig. 3a.. There was a trend for a negative correlation between individual subjects ethanol intake, and its response count during the punished component of the Vogel test. Total liquid intake was virtually identical between the two groups Ž F1,285 s 0.0; P ) 0.98; Fig. 3b.. Mean group weights for basolateral lesioned subjects and sham operated controls are given in Table 2. Although it has been reported that excitotoxic basolateral amygdala lesions did not affect food intake w2x, significantly lower body weights were seen in our study in the basolateral amygdala lesion group compared to sham operated controls over the time of the drinking procedure Žtreatment effect: F1,270 s 192.0; P - 0.001; time effect: F14,270 s 35.9; P - 0.001., similar to what was seen for central amygdala lesioned subjects. In addition, weight gain over the 15 days of the self administration procedure was significantly lower in the basolateral lesion group than in controls Ž83.5 " 4.2 vs. 94.8 " 2.2 g, mean " S.E.M.;

In the present study, ibotenic acid lesions of central amygdala produced a robust anticonflict effect in the Vogel procedure. Since unpunished drinking was identical in lesioned and sham operated controls, it appears that the anticonflict effect was not due to effects on thirst or other appetitive motivation, or motor performance deficits. The anticonflict effect of central amygdala lesions was qualitatively and quantitatively similar to that of established anti-anxiety drugs such as the benzodiazepines w43x, and may be interpreted as a reduction in experimental anxiety. Furthermore, the anticonflict effect was anatomically specific, since it was not reproduced by lesions in the proximally located basolateral amygdala. Our result is in agreement with present perception of a key role for central amygdala in a wide range of fear and stress responses w6,12,13,16,18,22x. Previous studies w6,18,22x have demonstrated the role of this structure in conditioned fear. Here, we found this role to be equally important in an unconditioned fear suppression of behaviour. Together, these observations are consistent with the notion that central amygdala is a final common relay in the activation of fear responses, which is independent of the nature of the fear stimulus. Furthermore, cell bodies located within the central nucleus appear to be important for the role of this structure in fear, since both the effects seen in the fear potentiated startle model w6x and those observed in the present study were obtained using the axon sparing toxin ibotenic acid. The neurochemical identity of the cell bodies involved is presently unknown. In contrast, it has been postulated that basolateral amygdala is involved in fear responses by serving as a link between cortical and subcortical sensory areas on one hand, and the central nucleus on the other. This link appears to be essential for conditioned fear responses w6,23,24,37x. From our present data, on the other hand, basolateral amygdala does not seem to be involved in the unconditioned fear suppression of behaviour underlying the Vogel model. This finding supports the notion that basolateral amygdala may be more important for integrating sensory data and memory content in order to extract their aversive andror fearful content w23,24,37x, but less so in the final activation of coordinated behavioural, endocrine and autonomic fear programs, which appears to be

C. Moller ¨ et al.r Brain Research 760 (1997) 94–101

mediated by amygdofugal projections of the central nucleus w23,25x. The effects of central amygdala lesions were less clear in the elevated plus maze, where behaviour of lesioned and control animals was virtually identical under baseline conditions. Although both anxiety models used in the present study have been extensively validated as being sensitive and specific for anxiety-promoting and suppressing treatments, it might be speculated that they partly reflect different emotional processes. Indeed, it has been proposed that the plus maze probes a special kind of anxiety w15x. However, a considerably simpler interpretation may be offered. The stress level inherent in the plus maze procedure is low under baseline conditions, and the extensive handling involved in, e.g., the preceding surgical procedures appears to have reduced the fearfulness of the plusmaze further to virtually none, witnessed by the fact that there was an equal distribution between open- and closedarm exploration in both control and central amygdala lesioned subjects. Pre-stressing the animals prior to testing, however, significantly suppressed open arm exploration in normal subjects in an expected manner, while no such effect was seen in central amygdala lesioned animals. This result may be parallel to the observation that the CRH-receptor antagonist, alpha-helical CRH, is inert in the plusmaze under baseline conditions w3x, while it effectively blocks the effect of various pre-stressors both upon i.c.v. and central amygdala administration w3,17,21,33x. Thus, central amygdaloid nucleus-dependent, anxiety-promoting signalling may not be tonically active, but rather be activated by various stressors. We next examined the effects of central amygdala lesions on the initiation phase of voluntary ethanol consumption. Ethanol self administration procedures might be limited by the fact that amounts of the drug sufficient to produce pharmacological effects are not easily consumed voluntarily by heterogeneous rats under normal conditions, making it difficult to relate changes in intake to rewarding actions of ethanol. However, procedures of the type employed here as well as by others w40x seem to partly overcome this limitation by gradually stepping up ethanol concentration, and also, in our case, by masking taste aversion with a low concentration of a non-caloric sweetener, saccharin. Under these conditions, 2.3 grkg ethanol was finally consumed daily on an average by sham operated controls. This is only appr. 30% less than consumption under identical conditions by the selectively bred Alko AA ethanol preferring line w28,39x. Due to an uneven distribution of intake over time, blood ethanol concentration at a given sampling time is highly variable with this type of unlimited access paradigm, and ranges between 0–37 mgrdl at a daily consumption of 2.2 grkg ŽHeilig, unpublished results.. Despite this fact, consumption in this range has been shown to produce blood ethanol concentrations sufficient to reduce experimental anxiety w40x. In our paradigm, voluntary ethanol intake was markedly

99

reduced by central amygdala lesions, even when ethanol intake was corrected for the lower body weight of the lesioned group. During the final phase of testing, when the 6% ethanol solution was available, consumption per unit body weight was more than 65% higher in control animals than in lesioned subjects. This effect appears to be unrelated to changes in thirst or motor performance, since total liquid intake was virtually identical in both groups. Secondly, central amygdala lesions have been shown to affect food intake w14x, and the ethanol solution available in the free choice procedure might have been consumed partly for its caloric content. Therefore, the possibility has to be considered that the difference in ethanol intake in the present study may have been related to lesion effects on appetite. However, we consider this possibility unlikely for two reasons. Although weight gain in central amygdala lesioned subjects was slower during the immediate post operative recovery period, resulting in significantly lower body weights in this group, weight gain over the 15 days of the self administration procedure was not significantly different between the lesion group and controls, indicating similar caloric intake in both groups over the course of the procedure. Secondly, a similar difference in body weights was seen in the basolateral lesion group, and in this group lower caloric intake was also indicated by a significantly slower weight gain over the time of the procedure. Despite this observation, ethanol intake was virtually identical in basolateral lesioned subjects and controls. On the other hand, a clear difference between central and basolateral lesions was that the former but not the latter resulted in a decrease in experimental anxiety in the Vogel paradigm. It is therefore our hypothesis that the decrease in ethanol consumption caused by the central amygdala lesion may have been related to the reduction in experimental anxiety produced by this treatment. Such an interpretation is in agreement with the finding of Spanagel et al. w40x, in which individual variability in experimental anxiety among a heterogeneous rat population correlated with voluntary ethanol intake, supporting the notion that anxiety reduction by ethanol may provide an important component of this drug’s reinforcement value. While our data support such a hypothesis for the initiation phase of a selfadministration procedure, the same may be true in dependent individuals, where local activation of GABA-A receptors within the amygdala was recently reported to suppresses ethanol selfadministration w36x. More variable relations between experimental anxiety and ethanol intake have been found in rat lines bred for ethanol preference. Thus, for male rats of the selectively bred ethanol-preferring P line, lower baseline measures of experimental anxiety and lower sensitivity to the antianxiety actions of ethanol were reported compared to heterogeneous Wistar rats w4x. Recently, similar results were obtained in males of the Alko AA ethanol preferring line w28x. In contrast, female P rats were reported to display elevated measures of experimental anxiety in sev-

100

C. Moller ¨ et al.r Brain Research 760 (1997) 94–101

eral paradigms, including fear suppression of responding, and the elevated plus maze, as well as an increased sensitivity to the anti-anxiety actions of ethanol w41x. Similar results were reported for the Sardinian ethanol-preferring line w11x. Clearly, depending on genetic background as well as chance events, different behavioural traits may co-segregate with ethanol preference during a breeding procedure, to create constellations different from those present in a heterogeneous population. Human alcoholism is a heterogeneous disorder, with differential genetic loading, clinical course and neurochemistry w9,10,42,44x. Among the main two subtypes characterized, late onset type I alcoholism has little genetic loading, and has been postulated to correlate with anxious personality traits. In contrast, early onset type II alcoholism has a strong genetic loading, and is characterized i.a. by disinhibited, impulsive behaviours. Our present finding, together with the result of Spanagel et al. w40x, is compatible with the hypothesis that antianxiety effects may contribute to the reward value of ethanol in a subtype of alcoholic patients, presumably those of type I. In contrast, antianxiety effects of ethanol may be less important in other subgroups, such as type II, where e.g. euphorogenic effects of potentiated DA-transmission in mesolimbic pathways may dominate instead.

Acknowledgements Supported by grants from the Swedish Medical Research Council and the Swedish Alcohol Monopoly Research Fund to MH. The authors wish to express their gratitude to Dr. Petri Hyytia¨ and Prof. U. Rydberg for valuable contributions.

References w1x A.M. Allan, J.P. Huidobro-Toro, V. Bleck, R.A. Harris, Alcohol and the GABA receptor-chloride channel complex of brain, Alcohol and Alcoholism, Suppl. 1 Ž1987. 643–646. w2x C. Ambrogi Lorenzini, E. Baldi, C. Bucherelli, A. Giachetti, G. Tassoni, Effects of nucleus basolateralis amygdalae neurotoxic lesions on some spontaneous activities in the rat, Physiol. Behav. 50 Ž1991. 1215–1219. w3x H.A. Baldwin, S. Rassnick, J. Rivier, G.F. Koob, T.K. Britton, CRF antagonist reverses the ‘anxiogenic’ response to ethanol withdrawal in the rat, Psychopharmacology 103 Ž1991. 227–232. w4x H.A. Baldwin, T.L. Wall, M.A. Schuckit, G.F. Koob, Differential effects of ethanol on punished responding in the P and NP rats, Alcohol. Clin. Exp. Res. 15 Ž1991. 700–704. w5x A.D. Campbell, W.J. McBride, Serotonin-3 receptor and ethanolstimulated dopamine release in the nucleus accumbens, Pharmacol. Biochem. Behav. 51 Ž1995. 835–842. w6x S. Campeau, M. Davis, Involvement of the central nucleus and basolateral complex of the amygdala in fear conditioning measured with fear-potentiated startle in rats trained concurrently with auditory and visual conditioned stimuli, J. Neurosci. 15 Ž1995. 2301– 2311.

w7x H. Cappell, C.P. Herman, Alcohol and tension reduction. A review, Q. J. Stud. Alcohol 33 Ž1972. 33–64. w8x K.A. Christie, J.D. Burke Jr., D.A. Regier, D.S. Rae, J.H. Boyd, B.Z. Locke, Epidemiologic evidence for early onset of mental disorders and higher risk of drug abuse in young adults, Am. J. Psychiatry 145 Ž1988. 971–975. w9x C.R. Cloninger, Neurogenetic adaptive mechanisms in alcoholism, Science 236 Ž1987. 410–416. w10x C.R. Cloninger, M. Bohman, S. Sigvardsson, Inheritance of alcohol abuse. Cross-fostering analysis of adopted men, Arch. Gen. Psychiatry 38 Ž1981. 861–868. w11x G. Colombo, R. Agabio, C. Lobina, R. Reali, A. Zocchi, F. Fadda, G.L. Gessa, Sardinian alcohol-preferring rats: a genetic animal model of anxiety, Physiol. Behav. 57 Ž1995. 1181–1185. w12x M. Davis, The role of the amygdala in fear and anxiety, Annu. Rev. Neurosci. 15 Ž1992. 353–375. w13x M. Davis, D. Rainnie, M. Cassell, Neurotransmission in the rat amygdala related to fear and anxiety, Trends. Neurosci. 17 Ž1994. 208–214. w14x A. Hajnal, P. Sandor, G. Jando, I. Vida, A. Czurko, Z. Karadi, L. Lenard, Feeding disturbances and EEG activity changes after amygdaloid kainate lesions in the rat, Brain Res. Bull. 29 Ž1992. 909–916. w15x S.L. Handley, J.W. McBlane, An assessment of the elevated X-maze for studying anxiety and anxiety-modulating drugs, J. Pharmacol. Toxicol. Methods 29 Ž1993. 129–138. w16x M. Heilig, G.F. Koob, R. Ekman, K.T. Britton, Corticotropin-releasing factor and neuropeptide Y: role in emotional integration, Trends. Neurosci. 17 Ž1994. 80–85. w17x S.C. Heinrichs, E.M. Pich, K.A. Miczek, K.T. Britton, G.F. Koob, Corticotropin-releasing factor antagonist reduces emotionality in socially defeated rats via direct neurotropic action, Brain Res. 581 Ž1992. 190–197. w18x J. Hitchcock, M. Davis, Lesions of the amygdala, but not of the cerebellum or red nucleus, block conditioned fear as measured with the potentiated startle paradigm, Behav. Neurosci. 100 Ž1. Ž1986. 11–22. w19x K. Kiianmaa, M. Nurmi, I. Nykanen, J.D. Sinclair, Effect of ethanol on extracellular dopamine in the nucleus accumbens of alcohol-preferring AA and alcohol-avoiding ANA rats, Pharmacol. Biochem. Behav. 52 Ž1995. 29–34. w20x G.F. Koob, Drugs of abuse: anatomy, pharmacology and function of reward pathways, Trends. Pharmacol. Sci. 13 Ž1992. 177–184. w21x G.F. Koob, S.C. Heinrichs, E.M. Pich, F. Menzaghi, H. Baldwin, K. Miczek, K.T. Britton, The role of corticotropin-releasing factor in behavioural responses to stress, Ciba. Found. Symp. 172 Ž1993. 277–289. w22x K.L. Kopchia, H.J. Altman, R.L. Commissaris, Effects of lesions of the central nucleus of the amygdala on anxiety-like behaviors in the rat, Pharmacol. Biochem. Behav. 43 Ž1992. 453–461. w23x J.E. LeDoux, Brain mechanisms of emotion and emotional learning, Curr. Opin. Neurobiol. 2 Ž1992. 191–197. w24x J.E. LeDoux, P. Cicchetti, A. Xagoraris, L.M. Romanski, The lateral amygdaloid nucleus: sensory interface of the amygdala in fear conditioning, J. Neurosci. 10 Ž1990. 1062–1069. w25x J.E. LeDoux, J. Iwata, P. Cicchetti, D.J. Reis, Different projections of the central amygdaloid nucleus mediate autonomic and behavioral correlates of conditioned fear, J. Neurosci. 8 Ž1988. 2517–2529. w26x S. Liljequist, J.A. Engel, The effects of GABA and benzodiazepine receptor antagonists on the anti-conflict actions of diazepam or ethanol, Pharmacol. Biochem. Behav. 21 Ž1984. 521–525. w27x C. Moller, O. Bing, M. Heilig, C-fos expression in the amygdala: in ¨ vivo antisense modulation and role in anxiety, Cell. Mol. Neurobiol. 14 Ž1994. 415–423. w28x C. Moller, L., Wiklund, P., Hyytia, M. Heilig, Decreased measures ¨ of experimental anxiety in rats bred for spontaneous alcohol preference, Alcohol Clin. Exp. Res. Ž1997. in press.

C. Moller ¨ et al.r Brain Research 760 (1997) 94–101 w29x G. Paxinos, C. Watson, The Rat Brain in Stereotaxic Coordinates, Academic Press, San Diego, 1986. w30x S. Pellow, P. Chopin, S.E. File, M. Briley, Validation of open: closed arm entries in an elevated plus-maze as a measure of anxiety in the rat, J. Neurosci. Methods 14 Ž1985. 149–167. w31x S. Pellow, S.E. File, Anxiolytic and anxiogenic drug effects on exploratory activity in an elevated plus-maze: a novel test of anxiety in the rat, Pharmacol. Biochem. Behav. 24 Ž1986. 525–529. w32x R.J. Powers, I.L. Kutash, Stress and alcohol, Intern. J. Addictions 20 Ž1985. 461–482. w33x S. Rassnick, S.C. Heinrichs, K.T. Britton, G.F. Koob, Microinjection of a corticotropin-releasing factor antagonist into the central nucleus of the amygdala reverses anxiogenic-like effects of ethanol withdrawal, Brain Res. 605 Ž1993. 25–32. w34x S. Rassnick, L. Stinus, G.F. Koob, The effects of 6-hydroxydopamine lesions of the nucleus accumbens and the mesolimbic dopamine system on oral self-administration of ethanol in the rat, Brain Res. 623 Ž1993. 16–24. w35x D.A. Regier, M.E. Farmer, D.S. Rae, B.Z. Locke, S.J. Keith, L.L. Judd, F.K. Goodwin, Comorbidity of mental disorders with alcohol and other drug abuse. Results from the Epidemiologic Catchment Area ŽECA. study, J. Am. Med. Assoc. 264 Ž1990. 2511–2518. w36x A.J. Roberts, M. Cole, G.F. Koob, Intra-amygdala muscimol decreases operant ethanol self-administration in dependent rats, Alcohol. Clin. Exp. Res. 20 Ž1996. 1289–1298. w37x C.B. Sananes, M. Davis, N-methyl-D-aspartate lesions of the lateral and basolateral nuclei of the amygdala block fear-potentiated startle and shock sensitization of startle, Behav. Neurosci. 106 Ž1992. 72–80.

101

w38x M.A. Schuckit, V. Hesselbrock, Alcohol dependence and anxiety disorders: what is the relationship?, Am. J. Psychiatry 151 Ž1994. 1723–1734. w39x J.D. Sinclair, A.D. Le, K. Kiianmaa, The AA and ANA rat lines, selected for differences in voluntary alcohol consumption, Experientia 45 Ž1989. 798–805. w40x R. Spanagel, A. Montkowski, K. Allingham, T. Stohr, M. Shoaib, F. Holsboer, R. Landgraf, Anxiety — a potential predictor of vulnerability to the initiation of ethanol self administration in rats, Psychopharmacology 122 Ž1995. 369–373. w41x R.B. Stewart, G.J. Gatto, L. Lumeng, T.K. Li, J.M. Murphy, Comparison of alcohol-preferring ŽP. and nonpreferring ŽNP. rats on tests of anxiety and for the anxiolytic effects of ethanol, Alcohol 10 Ž1993. 1–10. w42x J. Tiihonen, J. Kuikka, K. Bergstrom, P. Hakola, J. Karhu, O.P. Ryynanen, J. Fohr, Altered striatal dopamine re-uptake site densities in habitually violent and non-violent alcoholics, Nature Med. 1 Ž1995. 654–657. w43x J.R. Vogel, B. Beer, D.E. Clody, A simple and reliable conflict procedure for testing anti-anxiety agents, Psychopharmacologia 21 Ž1971. 1–7. w44x A.L. von Knorring, M. Bohman, L. von Knorring, L. Oreland, Platelet MAO activity as a biological marker in subgroups of alcoholism, Acta Psychiatr. Scand. 72 Ž1985. 51–58. w45x F. Weiss, M.T. Lorang, F.E. Bloom, G.F. Koob, Oral alcohol self-administration stimulates dopamine release in the rat nucleus accumbens: genetic and motivational determinants, J. Pharmacol. Exp. Ther. 267 Ž1993. 250–258.