- Email: [email protected]
Angiotensin II reduces alcohol intake and choice in water- or food-restricted rats
Alcohol. Vol. 13, No. 4, pp. 359-363. 1996 Copyright (91996Elsevier Science Inc. Printed in the USA. All rights reserved I)741-8329/96$15.00 + .00 ELSEVIER
Pll S0741-8329(96)00009-2
Angiotensin II Reduces Alcohol Intake and Choice in Water- or Food-Restricted Rats P A U L J. K U L K O S K Y ,
TANYA
G. A L L I S O N
AND BRIAN A. CARR
Department o f Psychology, University o f Southern Colorado, Pueblo, CO 81001-4901 R e c e i v e d 20 J u l y 1995; A c c e p t e d 17 N o v e m b e r 1995 KULKOSKY, P. J., T. G. ALLISON AND B. A. CARR. Angiotensin H reduces alcohol intake and choice in water- or foodrestricted rats. ALCOHOL 13(4) 359-363, 1996.--Two experiments were conducted to assess the effects of administration of the neuropeptide and hormone angiotensin II (AII) on ethanol intake and choice. First, 18 male Wistar rats were water deprived for 23 h and given access to 5% w/v ethanol for 30 min, followed by 30 rain of access to water; food was ad lib. Following adaptation to this schedule, rats were randomly assigned to receive an IP injection of 0, 100, or 200 ~g/kg of AII at either - 3 0 or 0 min prior to ethanol access. Each AII injection decreased ethanol intake only if injected immediately before access; water and food intake were unaffected. Secondly, rats were given food daily at 2% of body weight with ad lib water and randomly assigned to receive either only water or 4% w/v ethanol ad lib on alternate days. Following adaptation, rats were randomly assigned to receive IP saline or 200 txg/kg of AII prior to presentation of a choice of ethanol or water for 1 h. AII reduced ethanol intake and increased water intake at 0-30 rain after injection. Results confirm previous reports of inhibition of alcohol consumption by peripheral AII, and indicate a temporal constraint on AII's effect, which is consistent with a role as a short-term satiety factor. Angiotensin II
Ethanol intake and choice
Water intake
A controversy exits concerning the main effect of the neuropeptide and h o r m o n e angiotensin II (AI D on ethanol intake. A n early brief report showed that when A I I was injected into the septum of the rat brain, alcohol was selected as if the animals were deprived of water (24). Later reports demonstrated that peripheral injections of A l l selectively decreased alcohol intake (12), but chronic intraventricular infusion of A l l increased alcohol consumption (4). However, another study showed no effect of lateral or third ventricular infusion of A l l or A I I I on alcohol intake (21). The effects of A l l are interesting because they may represent alcohol satiation or desatiation actions, which could prove useful in neuropeptidebased therapies for alcoholism (27,30). For example, an opioid neuropeptide receptor blocker, naltrexone, was recently approved for use in the treatment of h u m a n alcoholism. In the following experiments, we performed two criterial tests of A I I as an alcohol satiety signal. First, we examined the effect of delay between peptide injection and alcohol access. A true satiety signal characteristically inhibits intake with a temporal constraint; close contiguity of signaling and consumption is required (26). Satiation occurs promptly, within the length of an ingestive bout, and thus preabsorptive signals
Food intake
Alcohol satiation
of satiety must work quickly. Secondly, a genuine signal of satiation is relatively commodity specific; ingestive acts that release the signal are inhibited, but other intakes are less affected. We examined the specificity of angiotensin's actions by administration of the peptide to nonfluid-deprived but hungry rats with a choice of alcohol and water. In this design, called "nutritive expectancy" (3), we also evaluated the effect of prior experience with the postingestive effects of ethanol on the action of AII. Results should allow further characterization of the effects of peripherally administered angiotensin on alcohol intake in two different motivational contexts and provide critical tests of this neuropeptide as a putative alcohol satiety or appetite stimulus.
METHOD
Animals Eighteen experimentally naive male Wistar rats [Charles River, Crl:(WI)] weighing 254-372 g were individually housed and tested in stainless steel wire-mesh cages. Experiments were conducted in a room maintained at approximately 23°C
Requests for reprints should be addressed to Paul J. Kulkosky, Department of Psychology, University of Southern Colorado. Pueblo, CO 81001-4901. 359
360 and on a 12:12 light:dark lighting cycle (0700 h on). All rats had ad lib access to Rodent Laboratory Chow (Purina 5001) in stainless steel hoppers and to deionized water in polycarbonate tubes fitted with rubber stoppers and valveless, stainless steel spouts, except as described below.
KULKOSKY, A L L I S O N A N D C A R R anol) × 2 (within-subjects, Time: 0-30 or 30-60 min) × 2 (within-subjects, Fluid: ethanol or water) x 2 (within-subjects, Dose: A I I or saline) mixed factorial A N O V A , as described above. RESULTS
Procedures Water Deprivation. All rats were deprived of water but continued on ad lib access to food. Twenty-three hours later (at 1500 h), rats received 30-min access to 5% w/v ethanol (from U.S.P. 95%), followed by 30-rain access to water. Fluid intake was measured to the nearest 0.5 ml. F o o d intake was measured to the nearest 0.05 g at 30 and 60 min, and corrected for spillage caught on sheets placed beneath the cages. This procedure was repeated for 7 days. On the seventh day of this baseline phase, rats were randomly assigned into two groups to receive a 1 ml/kg IP injection of 0.9% NaC1 (saline) at either 30 or 0 min prior to presentation of ethanol (N = 9). Following this adaptation phase, all rats were randomly assigned to receive IP injection of either 0, 100, or 200 Ixg/kg of A I I (Bachem, human octapeptide), at 0 or - 3 0 rain before ethanol, as previously assigned. Across a total of 3 consecutive test days, each rat received each dose of A I I in a randomized and counterbalanced sequence. Fluid and food intake data were analyzed with 2 (between-subjects, Delay: 0 and - 3 0 rain) x 3 (within-subjects, Dose: 0, 100, and 200 ~xg/kg) x 2 [within-subjects, Measurement Interval: 0-30 and 30-60 min (food intake only)] mixed factorial analyses of variance ( A N O V A ) , followed by Duncan's multiple range test, at an alpha significance level o f p < 0.05. Food Restriction. This procedure was adapted from that of Fedorchak and Bolles (3), as previously described (28). This method, called "nutritive expectancy," was developed to assess whether putative satiety factors such as cholecystokinin act unconditionally or require the prior acquisition of an expectation of caloric repletion from ethanol (3). Seventeen days following the water deprivation experiment, all rats were deprived of food, but continued on ad lib access to water. Twenty-three hours later, at 1500 h, rats were randomly assigned to receive 1 h of access to either 4% w/v ethanol (N = 9) or water (N = 8). Fluid intake was measured at 30 and 60 min after presentation. Following this measurement period, rats received 2% of their body weight in Purina Chow and continued on ad lib access to assigned ethanol solution or water. On the following day, all rats received access to water only at 1500 h for 1 h, followed by daily food ration and ad lib water. Rats assigned to receive ethanol continued on this alternatedays access to first 4% ethanol, and then water for a total of 10 days, whereas the other rats received only water throughout this 10-day adaptation phase. On the final 2 days of this adaptation period, all rats received a 1 ml/kg IP injection of saline immediately (within 1 min) prior to fluid access. On the following 2 days, all rats received a randomly assigned IP injection of either 200 txg/kg of A I I or saline prior to access to a two-bottle choice of 4% w/v ethanol and water. Right/left position of fluids was randomized on the first day and reversed on the second day, as was injection treatment. A d lib water and daily food ration were provided following the test period. If angiotensin inhibits ethanol intake unconditionally, prior experience with ethanol's caloric repletion effects should have no effect on the suppressive ability of AII, whereas if A I I acts on an acquired nutritive expectancy of ethanol's effects, then prior exposure should enhance A I I ' s effect. Data were analyzed with a 2 (between-subjects, Baseline: prior water or eth-
Water Deprivation Figure 1 shows mean --- SE intake of 5% ethanol by rats (Ns = 9) injected at - 3 0 or 0 min prior to ethanol access, as a function of IP dose of AII. Analysis revealed significant main effects on intake of injection delay, F(1, 16) = 4.65, and dose of AII, F(2, 32) -- 3.53, and a significant interaction of these factors, F(2, 32) = 3.44, ps < 0.05. Comparisons of means with Duncan's test indicated that rats injected - 3 0 min prior to ethanol consumed more alcohol than rats injected immediately prior to ethanol, and both doses of A I I reliably decreased ethanol intake in the latter rats (ps < 0.05). Figure 2 shows mean -+ SE intake of water by these rats at 30-450 min after presentation of ethanol solution, Analysis indicated no statistically significant main effects on intake of time, F(1, 16) = 4.488, p = 0.05, or dose, F(2, 32) = 2.41, ps > 0.05, and that the interaction of these factors was also not significant, F(2, 32) = 1.68, p > 0.05. However, an inspection of means with Duncan's test gave an increase in water consumption after 200 p~g/kg of A I I in rats injected - 3 0 rain before ethanol access. There was a significant effect of measurement interval on food intake, F(1, 16) = 9.91, and a significant interaction of injection delay and measurement interval, F(1, 16) = 17.02, ps < 0.05, but the main effects of delay and A I I dose, and all interactions with dose, were not statistically significant (ps > 0.05). Rats ate 67.2% more chow at 0-30 min than at 30--60 min (5.2 vs. 3.1 g/kg), hut the difference in intake was significant only in rats injected at - 3 0 min (6.81 vs. 1.94 g/kg). Inspection of means with Duncan's test showed a 172.7% increase in food intake after 200/xg/kg of A I I in rats injected
Injection Delay 70
50
-
40
'"
30
o
0 min. -30 min.
* p<0.05
60
g
f--I I
:~ 2o
10
0
100
200
ip. Dose (pg/kg)
FIG. 1. Mean ± SE 5% w/v ethanol intake of water-deprived rats after IP injection of 0, 100, or 200 p,g/kg of angiotensin II at -30 or 0 rain (Ns = 9) prior to ethanol access.
ANGIOTENSIN AND ALCOHOL INTAKE
361
Injection Delay
70
0 min. -30 rain.
~ l
* p<0.05
60 -~ 50
40
~ 3o
ence with fluid type, F(1, 105) = 8.97, fluid type and dose of All, F(1,105) = 29.4, and measurement period, fluid type, and dose of AII, F(1, 105) -- 19.19, ps < 0.05. All other main effects and interactions were not statistically significant (ps > 0.05). Post hoc comparison of means with Duncan's test indicated that AII significantly suppressed intake of ethanol and increased intake of water during the 0-30-min measurement period (ps < 0.05). There was significantly less intake of fluid at the 30-60-min measurement period (ps < 0.05) and no significant effects of AII at that time (ps > 0.05). Rats with prior baseline exposure to ethanol consumed more ethanol than water (p < 0.05), but rats that received only water during baseline did not consume more ethanol than water (p > 0.05).
IlJ
20
DISCUSSION
10
0
100
200
i.p. Dose (pg/kg)
FIG. 2. Mean _+ SE water intake of water-deprived rats after injection and access to ethanol solution for the preceding 30 min.
immediately prior to ethanol access (6.05 vs. 2.22 g/kg), at 30-60 min after access.
Food Restriction Mean -+ SE intake of 4% ethanol at 0-30 min after injection of saline or AII in rats with baseline exposure to ethanol or water is shown in Fig. 3; mean intake of water for these rats at this time period is shown in Fig. 4. A N O V A revealed a significant main effect on intake of measurement period, F(1, 105) = 36.36, and significant interactions of baseline experi-
[~
14
Our data confirm a specific suppressive effect of peripheral injection of AI1 on ethanol intake and choice in the male Wistar rat. Results of the water deprivation design further indicate that this inhibitory action of AII on alcohol ingestion satisfies the criterion of temporal constraint for a signal of alcohol satiation. A rapidly released, short-term, preabsorptive satiety signal acts promptly and briefly to inhibit intake before postabsorptive events can occur to elicit long-term satiation (26). Thus, close contiguity of ingestion and signal is typically required for inhibition of intake. We observed that the suppressive effect of AII is abolished by a 30-rain delay between injection and access to ethanol solution. This effect on ingestion appeared brief, persisting less than 30 min in both experiments, and relatively specific, as intake of food and water was typically unaffected by AII in the water deprivation design. A n increase in water intake was observed after the higher dose of AII in rats injected - 3 0 rain before ethanol, but this marginal effect was not preceded by inhibition of ethanol intake. A later increase in food intake did accompany inhibition of ethanol intake by the higher dose of AII; this marginal effect might reflect a compensatory increase in caloric intake after an initial depression of alcohol drinking and related feed-
prior water prior ethanol
E_-B prior water prior ethanol
14
* p<0.05
~12 E
e
F~ l o
,~10 E
6
6
8
8
t~
= 6
'"
4
0
~ 0
200
Dose of Angiotensin II (pg/kg)
FIG. 3. M e a n _+ SE intake o f 4% w/v ethanol by food-restricted rats
with prior baseline experience with ethanol (N = 9) or water (N = 8), for 30 min after IP injection of AII or saline.
0
200
Dose of Angiotensin II (pg/kg)
FIG. 4. Mean _+ SE intake of water by food-restricted rats with prior baseline exposure to ethanol or water, for 30 rain after IP injection of AII or saline.
362
KULKOSKY, A L L I S O N A N D C A R R
ing by AII. However, intake of water did increase in our food restriction design, a result that is compatible with response competition as an alternative explanation for A I I ' s effect on alcohol intake. The relative lack of effects of A I I on water intake in the water deprivation design is puzzling, but may reflect a ceiling effect on water intake. Water consumption was much lower in the nutritive expectancy design wherein A I I did strongly stimulate water intake. Nevertheless, Grupp and colleagues have shown no effect of A l l on glucose solution intake, which strengthens the claim that A I I functions specifically to satiate ethanol (8,12). Therefore, current evidence begins to satisfy the criteria of temporal and commodity specificity for an authentic signal of alcohol satiation. Our data and others' reports indicate the effect of A I I is robust, as it can be demonstrated in water-deprived, food-restricted, or ad lib-fed and -watered rats in the "limited access procedure" (12). In the latter procedure, peripheral A I I in this dose range inhibited ethanol intake but did not similarly affect water intake, a further demonstration of the specificity of the peptide's effect. In the limited access procedure, ethanol intakes and blood ethanol levels are typically lower than in the water deprivation procedure (6,29), so our design reveals that A I I functions when blood ethanol levels are at a higher pharmacological plane. This is important in view of potential therapeutic applications of angiotensin-based interventions, wherein blood ethanol levels are by definition abnormally high and persistent. Impressive evidence has been presented to support the hypothesis that the ability of A I I to inhibit alcohol choice reflects a genuine physiological action of the renin-angiotensin system. Receptor dependence, nervous system site of action, and neurotransmitter interactions have all been described in a thorough analysis of this neuropeptide effect (13-16,20,23,35). Comprehensive reviews of this novel action of A l l have been presented (7,8,17,18). We have previously obtained results consistent with the hypothesis that increases in the renin-angiotensin system function endogenously to satiate ethanol consumption. Addition of NaC1 to a saccharin, glucose, and ethanol solution maximized free-choice ethanol intake and blood ethanol levels in Wistar rats (25). This finding may be interpreted as the result of a salt inhibition of the renin-angiotensin system, which releases or disinhibits ethanol intake. There is extensive experimental evidence of a positive relation of sodium and ethanol intakes (1,9-11). Further, we found that injection of caffeine reduces ethanol intake in Wistar rats (2). This result could be explained in terms of the known stimulation of renin release by caffeine (34) and a consequent AII-induced inhibition of ethanol intake. However, several criteria for verification of a humoral factor as a signal of satiety remain to be satisfied for angiotensin [cf. (26)]. Whether these doses of A l l mimic a normal physiological response to alcohol ingestion is unclear. In our hands, an IP dose of 200 ~g/kg of A l l , relative to saline, produced a dramatic +58% increase in mean blood pressure within 5 min; however, pressure returned to basal values by 20 min after injection (unpublished observations). It seems unlikely alcohol drinking would produce such a pressor event, although the time course of the hypertensive effect of this large dose of A I I parallels the observed time course of inhibition of ethanol in-
take. Nevertheless, pressor responses and putative satiety effects of A I I have been clearly dissociated in several previous experiments reviewed by Grupp (7,8,17,18). Yet, injected "satiating" doses of a putative satiety factor should produce circulating levels within the range produced by ingestion. Further, rapid renin release should be provoked by self-selected oral intakes of ethanol solutions in the rat. Other criteria to be satisfied include the elicitation of a normal behavioral sequence and human self-report of alcohol satiety and support of dose- and repletion-dependent conditioning of satiety (26). A n initial application in humans of the angiotensin hypothesis of alcohol satiety has been unsuccessful (31,32), although further work is in progress. A I I induces an area postrema-dependent conditioned taste aversion robustly in cats and rats in preference designs (19,33). We found that prior experience with A I I and ethanol did not result in aversion to alcohol in the subsequent nutritive expectancy design, and that further experience with ethanol in the nutritive expectancy baseline was not necessary for the expression of AfI's inhibitory effect. However, such baseline experience did relatively augment alcohol choice, so prior ethanol intake during water deprivation and A I I injection may have induced a slight, easily extinguishable reduction in choice of alcohol, although not a frank aversion. Also, an angiotensin-converting enzyme inhibitor was briefly reported to inhibit liver aldehyde dehydrogenase activity (22), so a disulfiram-like aversive effect may be involved in mediation of inhibition of alcohol intake by angiotensin-based interventions. Further criticism of the notion that A I I is an alcohol satiety factor stems from demonstrations that intraventricular infusion of A l l increases intake of ethanol in male Long-Evans rats (4), although this result was not replicated (21). The intake stimulation effect of peripherally administered angiotensin converting enzyme inhibitors was blocked by lesion of circumventricular organs (5), as was the inhibitory effect of peripherally injected A l l (15). The controversy generated by these disparate results can be clarified by further replication and analysis of the mechanism of these effects in critical experiments. For example, strain differences (Long-Evans vs. Wistar) could be examined, as well as sex differences--no studies have yet explored the effects of A l l on ethanol consumption in female rats. In an initial study we found no reliable effect of A I I on alcohol intake in a strain of hooded rats, in direct contrast to Wistar rats (manuscript in preparation). Therefore, extensive critical experimentation remains to resolve controversy in the interpretation of the role of angiotensin in the control of alcohol ingestion. We found a specific, temporally constrained inhibitory effect of peripherally injected angiotensin II on alcohol intake and choice in waterdeprived and food-restricted male Wistar rats. These results encourage further examination of the behavioral actions of elements of the renin-angiotensin system in the regulation of alcohol consumption and the disregulation of alcoholism. ACKNOWLEDGEMENTS This research was supported by Minority Biomedical Research Support Grant 5 SO6 GM08197-12 from the National Institutes of
Health. A preliminary communication of these findings was presented at the 65th annual meeting of The Rocky Mountain Psychological Association in Boulder, CO, April, 1995.
REFERENCES
1. DiBattista, D. Dietary salt and voluntary ethanol consumption in golden hamsters. Alcohol 11:505-512; 1994.
2. Dietze, M. A4 Kulkosky, P. J. Effects of caffeine and bombesin on ethanol and food intake. Life Sci. 48:1837-1844; 1991.
ANGIOTENSIN AND ALCOHOL INTAKE 3. Fedorchak, P. M.; Bolles, R. C. Nutritive expectancies mediate cholecystokinin's suppression-of-intake effect. Behav. Neurosci. 102:451-455; 1988. 4. Fitts, D. A. Angiotensin and captopril increase alcohol intake. Pharmacol. Biochem. Behav. 45:35-43; 1993. 5. Fitts, D. A. Forebrain circumventricular organs mediate captopril-enhanced ethanol intake in rats. Pharmacol. Biochem. Bebav. 45:811-816; 1993. 6. Glazner, G. W.; Cannon, R. L.; Kulkosky, P. J. Effect of bombesin on behaviors associated with ethanol satiation and blood ethanol levels. Alcohol 5:325-330; 1988. 7. Grupp, L. A. Alcohol satiety, hypertension and the renin-angiotensin system. Med. Hypoth. 24:11-19; 1987. 8. Grupp, L. A. The renin-angiotensin system as a regulator of alcohol consumption: A review and some new insights. In: Zakhari, S., ed. N I A A A Research Monograph No. 23: Alcohol and the endocrine system. Bethesda, MD: National Institutes of Health; 1993:37-65. 9. Grupp, L. A.; Perlanski, E.; Stewart, R. B. Dietary salt and Docasalt treatments modify ethanol self-selection in rats. Behav. Neural Biol. 40:239-250; 1984. 10. Grupp, L. A.; Stewart, R. B.; Perlanski, E. Salt restriction and the voluntary intake of ethanol in rats. Physiol. Psychol. 12:242-246; 1984. 11. Grupp, L. A.; Perlanski, E.; Stewart, R. B. Diet and diuretics in the reduction of voluntary alcohol drinking in rats. Alcohol Alcohol. 21:75-79: 1986. 12. Grupp, L. A.; Killian, M.; Perlanski, E.; Stewart, R. B. Angiotensin I1 reduces voluntary alcohol intake in the rat. Pharmacol. Biochem. Behav. 29:479-482; 1988. 13. Grupp, L. A.; Perlanski, E.; Stewart, R. B. Attenuation of alcohol intake by a serotonin uptake inhibitor: Evidence for mediation through the renin-angiotensin system. Pharmacol. Biochem. Behav. 30:823-827; 1988. 14. Grupp, L.A.; Perlanski, E.; Stewart, R.B. Angiotensin II-induced suppression of alcohol intake and its reversal by the angiotensin antagonist Sar-1 Thr-8 angiotensin II. Pharmcol. Biochem. Behav. 31:813-816; 1988. 15. Grupp, L. A.: Perlanski, E.; Stewart, R. B. Systemic angiotensin II acts at the subfornical organ to suppress voluntary alcohol consumption. Pharmacol. Biochem. Behav. 34:201-205; 1989. 16. Grupp, L. A.; Perlanski, E.; Stewart, R. B. Systemic angiotensin II does not act at the area postrema to suppress alcohol intake. Alcohol 8:165-167; 1991. 17. Grupp~ L. A.; Perlanski, E.; Stewart, R. B. Regulation of alcohol consumption by the renin-angiotensin system: A review of recent findings and a possible mechanism of action. Neurosci. Biobehav. Rev. 15:265-275; 1991. 18. Grupp, L. A.; Spinosa, G.; Lingham, T. Management of alcohol consumption with angiotensin converting enzyme inhibitors: A review of the animal findings. In: Naranjo, C. A.; Sellers, E. M., eds. Novel pharmacological interventions for alcoholism. New York: Springer-Verlag; 1992:201-214. 19. Grupp, L. A.; Chow, S. Y.-M. Angiotensin II mediates a condi-
363
20.
21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31.
32.
33.
34.
35.
tioned taste aversion in water-replete rats. Pharmacol. Biochem. Behav. 44:985-988; 1993. Grupp, L. A.; Harding, S. The reduction in alcohol drinking by peripherally injected angiotensin II is selectively mediated by the AT 1 receptor subtype. Pharmacol. Biochem. Behav. 47:385-392; 1994. Grupp, L. A.; Harding, S. Intracerebroventricularly infused angiotensin II or lII do not alter voluntary alcohol intake in rats. Pharmacol. Biochem. Behav. 51:593-599; 1995. Guivernau, M.; Quintanilla, M. E.: Tampier, L. Effect of dipyridamole and captropril on prostacyclin production and ALDH activity in rats. Alcohol. Clin. Exp. Res. 14:295; 1990. Hubbell, C. L.; Chrisbach, G. A.; Bilsky, E. J.; Reid, L. D. Manipulations of the renin-angiotensin system and intake of a sweetened alcoholic beverage among rats. Alcohol 9:53-61; 1992. Johnson, D. A.; Anderson, R. P. Alcohol ingestion following intracerebral angiotensin administration and water deprivation. Pharmacol. Biochem. Behav. 1:739-741; 1973. Kulkosky, P. J. Effect of addition of ethanol and NaCI on saccharin + glucose polydipsia. Pharmacol. Biochem. Behav. 10:277283; 1979. Kulkosky, P. J. Conditioned food aversions and satiety signals. Ann. NY Acad. Sci. 443:330-347; 1985. Kulkosky, P. J. Brain-gut neuropeptides and the limitation of ethanol consumption. Neurosci. Biobehav. Rev. 9:179-190; 1985. Kulkosky, P. J.; Sanchez, M. R.; Foderaro, M. A.; Chiu, N. Cholecystokinin and satiation with alcohol. Alcohol 6:395-402; 1989. Linseman, M. A. Alcohol consumption in free-feeding rats: Procedural, genetic and pharmacokinetic factors. Psychopharmacology (Berlin) 92:254-261; 1987. Litten, R. Z.; Allen, J. P. Reducing the desire to drink: Pharmacology and neurobiology. In: Galanter, M., ed. Recent developments in alcoholism, vol. 11. New York: Plenum; 1993:325-344. Naranjo, C. A.; Kadlec, K. E.; Sanhueza, P.; Woodley-Remus, D.; Sellers, E. M. Enalapril effects on alcohol intake and other consummatory behaviors in alcoholics. Clin. Pharmacol. Ther. 50:96106; 1991. Naranjo, C. A.; Bremner, K. E.; Sanhueza, P.; Sellers, E. M. Enalapril does not decrease alcohol intake in normotensive alcoholics. In: Naranjo, C. A.; Sellers, E. M., eds. Novel pharmacological interventions for alcoholism. New York: Springer-Verlag; 1992:215-225. Rabin, B. M.; Hunt, W. A.; Bakarich, A. C.; Chedester, A. L.: Lee, J. Angiotensin II-induced taste aversion learning in cats and rats and the role of the area postrema. Physiol. Behav. 36:11731178; 1986. Robertson, D.; Frolich, J. C.; Carr, R. K.; Watson, J. T.; Hollifield, J. W.; Shand, D. G.; Oates, J. A. Effects of caffeine on plasma renin activity, catecholamines and blood pressure. N. Engl. J. Med. 298:181-186; 1978. Ross, A. D.; Perlanski, E.; Grupp, L. A. Vasopressin does not mediate the inhibition of ethanol drinking by the renin-angiotensin system. Pharmacol. Biochem. Behav. 36:761-765: 1990.
Pll S0741-8329(96)00009-2
Angiotensin II Reduces Alcohol Intake and Choice in Water- or Food-Restricted Rats P A U L J. K U L K O S K Y ,
TANYA
G. A L L I S O N
AND BRIAN A. CARR
Department o f Psychology, University o f Southern Colorado, Pueblo, CO 81001-4901 R e c e i v e d 20 J u l y 1995; A c c e p t e d 17 N o v e m b e r 1995 KULKOSKY, P. J., T. G. ALLISON AND B. A. CARR. Angiotensin H reduces alcohol intake and choice in water- or foodrestricted rats. ALCOHOL 13(4) 359-363, 1996.--Two experiments were conducted to assess the effects of administration of the neuropeptide and hormone angiotensin II (AII) on ethanol intake and choice. First, 18 male Wistar rats were water deprived for 23 h and given access to 5% w/v ethanol for 30 min, followed by 30 rain of access to water; food was ad lib. Following adaptation to this schedule, rats were randomly assigned to receive an IP injection of 0, 100, or 200 ~g/kg of AII at either - 3 0 or 0 min prior to ethanol access. Each AII injection decreased ethanol intake only if injected immediately before access; water and food intake were unaffected. Secondly, rats were given food daily at 2% of body weight with ad lib water and randomly assigned to receive either only water or 4% w/v ethanol ad lib on alternate days. Following adaptation, rats were randomly assigned to receive IP saline or 200 txg/kg of AII prior to presentation of a choice of ethanol or water for 1 h. AII reduced ethanol intake and increased water intake at 0-30 rain after injection. Results confirm previous reports of inhibition of alcohol consumption by peripheral AII, and indicate a temporal constraint on AII's effect, which is consistent with a role as a short-term satiety factor. Angiotensin II
Ethanol intake and choice
Water intake
A controversy exits concerning the main effect of the neuropeptide and h o r m o n e angiotensin II (AI D on ethanol intake. A n early brief report showed that when A I I was injected into the septum of the rat brain, alcohol was selected as if the animals were deprived of water (24). Later reports demonstrated that peripheral injections of A l l selectively decreased alcohol intake (12), but chronic intraventricular infusion of A l l increased alcohol consumption (4). However, another study showed no effect of lateral or third ventricular infusion of A l l or A I I I on alcohol intake (21). The effects of A l l are interesting because they may represent alcohol satiation or desatiation actions, which could prove useful in neuropeptidebased therapies for alcoholism (27,30). For example, an opioid neuropeptide receptor blocker, naltrexone, was recently approved for use in the treatment of h u m a n alcoholism. In the following experiments, we performed two criterial tests of A I I as an alcohol satiety signal. First, we examined the effect of delay between peptide injection and alcohol access. A true satiety signal characteristically inhibits intake with a temporal constraint; close contiguity of signaling and consumption is required (26). Satiation occurs promptly, within the length of an ingestive bout, and thus preabsorptive signals
Food intake
Alcohol satiation
of satiety must work quickly. Secondly, a genuine signal of satiation is relatively commodity specific; ingestive acts that release the signal are inhibited, but other intakes are less affected. We examined the specificity of angiotensin's actions by administration of the peptide to nonfluid-deprived but hungry rats with a choice of alcohol and water. In this design, called "nutritive expectancy" (3), we also evaluated the effect of prior experience with the postingestive effects of ethanol on the action of AII. Results should allow further characterization of the effects of peripherally administered angiotensin on alcohol intake in two different motivational contexts and provide critical tests of this neuropeptide as a putative alcohol satiety or appetite stimulus.
METHOD
Animals Eighteen experimentally naive male Wistar rats [Charles River, Crl:(WI)] weighing 254-372 g were individually housed and tested in stainless steel wire-mesh cages. Experiments were conducted in a room maintained at approximately 23°C
Requests for reprints should be addressed to Paul J. Kulkosky, Department of Psychology, University of Southern Colorado. Pueblo, CO 81001-4901. 359
360 and on a 12:12 light:dark lighting cycle (0700 h on). All rats had ad lib access to Rodent Laboratory Chow (Purina 5001) in stainless steel hoppers and to deionized water in polycarbonate tubes fitted with rubber stoppers and valveless, stainless steel spouts, except as described below.
KULKOSKY, A L L I S O N A N D C A R R anol) × 2 (within-subjects, Time: 0-30 or 30-60 min) × 2 (within-subjects, Fluid: ethanol or water) x 2 (within-subjects, Dose: A I I or saline) mixed factorial A N O V A , as described above. RESULTS
Procedures Water Deprivation. All rats were deprived of water but continued on ad lib access to food. Twenty-three hours later (at 1500 h), rats received 30-min access to 5% w/v ethanol (from U.S.P. 95%), followed by 30-rain access to water. Fluid intake was measured to the nearest 0.5 ml. F o o d intake was measured to the nearest 0.05 g at 30 and 60 min, and corrected for spillage caught on sheets placed beneath the cages. This procedure was repeated for 7 days. On the seventh day of this baseline phase, rats were randomly assigned into two groups to receive a 1 ml/kg IP injection of 0.9% NaC1 (saline) at either 30 or 0 min prior to presentation of ethanol (N = 9). Following this adaptation phase, all rats were randomly assigned to receive IP injection of either 0, 100, or 200 Ixg/kg of A I I (Bachem, human octapeptide), at 0 or - 3 0 rain before ethanol, as previously assigned. Across a total of 3 consecutive test days, each rat received each dose of A I I in a randomized and counterbalanced sequence. Fluid and food intake data were analyzed with 2 (between-subjects, Delay: 0 and - 3 0 rain) x 3 (within-subjects, Dose: 0, 100, and 200 ~xg/kg) x 2 [within-subjects, Measurement Interval: 0-30 and 30-60 min (food intake only)] mixed factorial analyses of variance ( A N O V A ) , followed by Duncan's multiple range test, at an alpha significance level o f p < 0.05. Food Restriction. This procedure was adapted from that of Fedorchak and Bolles (3), as previously described (28). This method, called "nutritive expectancy," was developed to assess whether putative satiety factors such as cholecystokinin act unconditionally or require the prior acquisition of an expectation of caloric repletion from ethanol (3). Seventeen days following the water deprivation experiment, all rats were deprived of food, but continued on ad lib access to water. Twenty-three hours later, at 1500 h, rats were randomly assigned to receive 1 h of access to either 4% w/v ethanol (N = 9) or water (N = 8). Fluid intake was measured at 30 and 60 min after presentation. Following this measurement period, rats received 2% of their body weight in Purina Chow and continued on ad lib access to assigned ethanol solution or water. On the following day, all rats received access to water only at 1500 h for 1 h, followed by daily food ration and ad lib water. Rats assigned to receive ethanol continued on this alternatedays access to first 4% ethanol, and then water for a total of 10 days, whereas the other rats received only water throughout this 10-day adaptation phase. On the final 2 days of this adaptation period, all rats received a 1 ml/kg IP injection of saline immediately (within 1 min) prior to fluid access. On the following 2 days, all rats received a randomly assigned IP injection of either 200 txg/kg of A I I or saline prior to access to a two-bottle choice of 4% w/v ethanol and water. Right/left position of fluids was randomized on the first day and reversed on the second day, as was injection treatment. A d lib water and daily food ration were provided following the test period. If angiotensin inhibits ethanol intake unconditionally, prior experience with ethanol's caloric repletion effects should have no effect on the suppressive ability of AII, whereas if A I I acts on an acquired nutritive expectancy of ethanol's effects, then prior exposure should enhance A I I ' s effect. Data were analyzed with a 2 (between-subjects, Baseline: prior water or eth-
Water Deprivation Figure 1 shows mean --- SE intake of 5% ethanol by rats (Ns = 9) injected at - 3 0 or 0 min prior to ethanol access, as a function of IP dose of AII. Analysis revealed significant main effects on intake of injection delay, F(1, 16) = 4.65, and dose of AII, F(2, 32) -- 3.53, and a significant interaction of these factors, F(2, 32) = 3.44, ps < 0.05. Comparisons of means with Duncan's test indicated that rats injected - 3 0 min prior to ethanol consumed more alcohol than rats injected immediately prior to ethanol, and both doses of A I I reliably decreased ethanol intake in the latter rats (ps < 0.05). Figure 2 shows mean -+ SE intake of water by these rats at 30-450 min after presentation of ethanol solution, Analysis indicated no statistically significant main effects on intake of time, F(1, 16) = 4.488, p = 0.05, or dose, F(2, 32) = 2.41, ps > 0.05, and that the interaction of these factors was also not significant, F(2, 32) = 1.68, p > 0.05. However, an inspection of means with Duncan's test gave an increase in water consumption after 200 p~g/kg of A I I in rats injected - 3 0 rain before ethanol access. There was a significant effect of measurement interval on food intake, F(1, 16) = 9.91, and a significant interaction of injection delay and measurement interval, F(1, 16) = 17.02, ps < 0.05, but the main effects of delay and A I I dose, and all interactions with dose, were not statistically significant (ps > 0.05). Rats ate 67.2% more chow at 0-30 min than at 30--60 min (5.2 vs. 3.1 g/kg), hut the difference in intake was significant only in rats injected at - 3 0 min (6.81 vs. 1.94 g/kg). Inspection of means with Duncan's test showed a 172.7% increase in food intake after 200/xg/kg of A I I in rats injected
Injection Delay 70
50
-
40
'"
30
o
0 min. -30 min.
* p<0.05
60
g
f--I I
:~ 2o
10
0
100
200
ip. Dose (pg/kg)
FIG. 1. Mean ± SE 5% w/v ethanol intake of water-deprived rats after IP injection of 0, 100, or 200 p,g/kg of angiotensin II at -30 or 0 rain (Ns = 9) prior to ethanol access.
ANGIOTENSIN AND ALCOHOL INTAKE
361
Injection Delay
70
0 min. -30 rain.
~ l
* p<0.05
60 -~ 50
40
~ 3o
ence with fluid type, F(1, 105) = 8.97, fluid type and dose of All, F(1,105) = 29.4, and measurement period, fluid type, and dose of AII, F(1, 105) -- 19.19, ps < 0.05. All other main effects and interactions were not statistically significant (ps > 0.05). Post hoc comparison of means with Duncan's test indicated that AII significantly suppressed intake of ethanol and increased intake of water during the 0-30-min measurement period (ps < 0.05). There was significantly less intake of fluid at the 30-60-min measurement period (ps < 0.05) and no significant effects of AII at that time (ps > 0.05). Rats with prior baseline exposure to ethanol consumed more ethanol than water (p < 0.05), but rats that received only water during baseline did not consume more ethanol than water (p > 0.05).
IlJ
20
DISCUSSION
10
0
100
200
i.p. Dose (pg/kg)
FIG. 2. Mean _+ SE water intake of water-deprived rats after injection and access to ethanol solution for the preceding 30 min.
immediately prior to ethanol access (6.05 vs. 2.22 g/kg), at 30-60 min after access.
Food Restriction Mean -+ SE intake of 4% ethanol at 0-30 min after injection of saline or AII in rats with baseline exposure to ethanol or water is shown in Fig. 3; mean intake of water for these rats at this time period is shown in Fig. 4. A N O V A revealed a significant main effect on intake of measurement period, F(1, 105) = 36.36, and significant interactions of baseline experi-
[~
14
Our data confirm a specific suppressive effect of peripheral injection of AI1 on ethanol intake and choice in the male Wistar rat. Results of the water deprivation design further indicate that this inhibitory action of AII on alcohol ingestion satisfies the criterion of temporal constraint for a signal of alcohol satiation. A rapidly released, short-term, preabsorptive satiety signal acts promptly and briefly to inhibit intake before postabsorptive events can occur to elicit long-term satiation (26). Thus, close contiguity of ingestion and signal is typically required for inhibition of intake. We observed that the suppressive effect of AII is abolished by a 30-rain delay between injection and access to ethanol solution. This effect on ingestion appeared brief, persisting less than 30 min in both experiments, and relatively specific, as intake of food and water was typically unaffected by AII in the water deprivation design. A n increase in water intake was observed after the higher dose of AII in rats injected - 3 0 rain before ethanol, but this marginal effect was not preceded by inhibition of ethanol intake. A later increase in food intake did accompany inhibition of ethanol intake by the higher dose of AII; this marginal effect might reflect a compensatory increase in caloric intake after an initial depression of alcohol drinking and related feed-
prior water prior ethanol
E_-B prior water prior ethanol
14
* p<0.05
~12 E
e
F~ l o
,~10 E
6
6
8
8
t~
= 6
'"
4
0
~ 0
200
Dose of Angiotensin II (pg/kg)
FIG. 3. M e a n _+ SE intake o f 4% w/v ethanol by food-restricted rats
with prior baseline experience with ethanol (N = 9) or water (N = 8), for 30 min after IP injection of AII or saline.
0
200
Dose of Angiotensin II (pg/kg)
FIG. 4. Mean _+ SE intake of water by food-restricted rats with prior baseline exposure to ethanol or water, for 30 rain after IP injection of AII or saline.
362
KULKOSKY, A L L I S O N A N D C A R R
ing by AII. However, intake of water did increase in our food restriction design, a result that is compatible with response competition as an alternative explanation for A I I ' s effect on alcohol intake. The relative lack of effects of A I I on water intake in the water deprivation design is puzzling, but may reflect a ceiling effect on water intake. Water consumption was much lower in the nutritive expectancy design wherein A I I did strongly stimulate water intake. Nevertheless, Grupp and colleagues have shown no effect of A l l on glucose solution intake, which strengthens the claim that A I I functions specifically to satiate ethanol (8,12). Therefore, current evidence begins to satisfy the criteria of temporal and commodity specificity for an authentic signal of alcohol satiation. Our data and others' reports indicate the effect of A I I is robust, as it can be demonstrated in water-deprived, food-restricted, or ad lib-fed and -watered rats in the "limited access procedure" (12). In the latter procedure, peripheral A I I in this dose range inhibited ethanol intake but did not similarly affect water intake, a further demonstration of the specificity of the peptide's effect. In the limited access procedure, ethanol intakes and blood ethanol levels are typically lower than in the water deprivation procedure (6,29), so our design reveals that A I I functions when blood ethanol levels are at a higher pharmacological plane. This is important in view of potential therapeutic applications of angiotensin-based interventions, wherein blood ethanol levels are by definition abnormally high and persistent. Impressive evidence has been presented to support the hypothesis that the ability of A I I to inhibit alcohol choice reflects a genuine physiological action of the renin-angiotensin system. Receptor dependence, nervous system site of action, and neurotransmitter interactions have all been described in a thorough analysis of this neuropeptide effect (13-16,20,23,35). Comprehensive reviews of this novel action of A l l have been presented (7,8,17,18). We have previously obtained results consistent with the hypothesis that increases in the renin-angiotensin system function endogenously to satiate ethanol consumption. Addition of NaC1 to a saccharin, glucose, and ethanol solution maximized free-choice ethanol intake and blood ethanol levels in Wistar rats (25). This finding may be interpreted as the result of a salt inhibition of the renin-angiotensin system, which releases or disinhibits ethanol intake. There is extensive experimental evidence of a positive relation of sodium and ethanol intakes (1,9-11). Further, we found that injection of caffeine reduces ethanol intake in Wistar rats (2). This result could be explained in terms of the known stimulation of renin release by caffeine (34) and a consequent AII-induced inhibition of ethanol intake. However, several criteria for verification of a humoral factor as a signal of satiety remain to be satisfied for angiotensin [cf. (26)]. Whether these doses of A l l mimic a normal physiological response to alcohol ingestion is unclear. In our hands, an IP dose of 200 ~g/kg of A l l , relative to saline, produced a dramatic +58% increase in mean blood pressure within 5 min; however, pressure returned to basal values by 20 min after injection (unpublished observations). It seems unlikely alcohol drinking would produce such a pressor event, although the time course of the hypertensive effect of this large dose of A I I parallels the observed time course of inhibition of ethanol in-
take. Nevertheless, pressor responses and putative satiety effects of A I I have been clearly dissociated in several previous experiments reviewed by Grupp (7,8,17,18). Yet, injected "satiating" doses of a putative satiety factor should produce circulating levels within the range produced by ingestion. Further, rapid renin release should be provoked by self-selected oral intakes of ethanol solutions in the rat. Other criteria to be satisfied include the elicitation of a normal behavioral sequence and human self-report of alcohol satiety and support of dose- and repletion-dependent conditioning of satiety (26). A n initial application in humans of the angiotensin hypothesis of alcohol satiety has been unsuccessful (31,32), although further work is in progress. A I I induces an area postrema-dependent conditioned taste aversion robustly in cats and rats in preference designs (19,33). We found that prior experience with A I I and ethanol did not result in aversion to alcohol in the subsequent nutritive expectancy design, and that further experience with ethanol in the nutritive expectancy baseline was not necessary for the expression of AfI's inhibitory effect. However, such baseline experience did relatively augment alcohol choice, so prior ethanol intake during water deprivation and A I I injection may have induced a slight, easily extinguishable reduction in choice of alcohol, although not a frank aversion. Also, an angiotensin-converting enzyme inhibitor was briefly reported to inhibit liver aldehyde dehydrogenase activity (22), so a disulfiram-like aversive effect may be involved in mediation of inhibition of alcohol intake by angiotensin-based interventions. Further criticism of the notion that A I I is an alcohol satiety factor stems from demonstrations that intraventricular infusion of A l l increases intake of ethanol in male Long-Evans rats (4), although this result was not replicated (21). The intake stimulation effect of peripherally administered angiotensin converting enzyme inhibitors was blocked by lesion of circumventricular organs (5), as was the inhibitory effect of peripherally injected A l l (15). The controversy generated by these disparate results can be clarified by further replication and analysis of the mechanism of these effects in critical experiments. For example, strain differences (Long-Evans vs. Wistar) could be examined, as well as sex differences--no studies have yet explored the effects of A l l on ethanol consumption in female rats. In an initial study we found no reliable effect of A I I on alcohol intake in a strain of hooded rats, in direct contrast to Wistar rats (manuscript in preparation). Therefore, extensive critical experimentation remains to resolve controversy in the interpretation of the role of angiotensin in the control of alcohol ingestion. We found a specific, temporally constrained inhibitory effect of peripherally injected angiotensin II on alcohol intake and choice in waterdeprived and food-restricted male Wistar rats. These results encourage further examination of the behavioral actions of elements of the renin-angiotensin system in the regulation of alcohol consumption and the disregulation of alcoholism. ACKNOWLEDGEMENTS This research was supported by Minority Biomedical Research Support Grant 5 SO6 GM08197-12 from the National Institutes of
Health. A preliminary communication of these findings was presented at the 65th annual meeting of The Rocky Mountain Psychological Association in Boulder, CO, April, 1995.
REFERENCES
1. DiBattista, D. Dietary salt and voluntary ethanol consumption in golden hamsters. Alcohol 11:505-512; 1994.
2. Dietze, M. A4 Kulkosky, P. J. Effects of caffeine and bombesin on ethanol and food intake. Life Sci. 48:1837-1844; 1991.
ANGIOTENSIN AND ALCOHOL INTAKE 3. Fedorchak, P. M.; Bolles, R. C. Nutritive expectancies mediate cholecystokinin's suppression-of-intake effect. Behav. Neurosci. 102:451-455; 1988. 4. Fitts, D. A. Angiotensin and captopril increase alcohol intake. Pharmacol. Biochem. Behav. 45:35-43; 1993. 5. Fitts, D. A. Forebrain circumventricular organs mediate captopril-enhanced ethanol intake in rats. Pharmacol. Biochem. Bebav. 45:811-816; 1993. 6. Glazner, G. W.; Cannon, R. L.; Kulkosky, P. J. Effect of bombesin on behaviors associated with ethanol satiation and blood ethanol levels. Alcohol 5:325-330; 1988. 7. Grupp, L. A. Alcohol satiety, hypertension and the renin-angiotensin system. Med. Hypoth. 24:11-19; 1987. 8. Grupp, L. A. The renin-angiotensin system as a regulator of alcohol consumption: A review and some new insights. In: Zakhari, S., ed. N I A A A Research Monograph No. 23: Alcohol and the endocrine system. Bethesda, MD: National Institutes of Health; 1993:37-65. 9. Grupp, L. A.; Perlanski, E.; Stewart, R. B. Dietary salt and Docasalt treatments modify ethanol self-selection in rats. Behav. Neural Biol. 40:239-250; 1984. 10. Grupp, L. A.; Stewart, R. B.; Perlanski, E. Salt restriction and the voluntary intake of ethanol in rats. Physiol. Psychol. 12:242-246; 1984. 11. Grupp, L. A.; Perlanski, E.; Stewart, R. B. Diet and diuretics in the reduction of voluntary alcohol drinking in rats. Alcohol Alcohol. 21:75-79: 1986. 12. Grupp, L. A.; Killian, M.; Perlanski, E.; Stewart, R. B. Angiotensin I1 reduces voluntary alcohol intake in the rat. Pharmacol. Biochem. Behav. 29:479-482; 1988. 13. Grupp, L. A.; Perlanski, E.; Stewart, R. B. Attenuation of alcohol intake by a serotonin uptake inhibitor: Evidence for mediation through the renin-angiotensin system. Pharmacol. Biochem. Behav. 30:823-827; 1988. 14. Grupp, L.A.; Perlanski, E.; Stewart, R.B. Angiotensin II-induced suppression of alcohol intake and its reversal by the angiotensin antagonist Sar-1 Thr-8 angiotensin II. Pharmcol. Biochem. Behav. 31:813-816; 1988. 15. Grupp, L. A.: Perlanski, E.; Stewart, R. B. Systemic angiotensin II acts at the subfornical organ to suppress voluntary alcohol consumption. Pharmacol. Biochem. Behav. 34:201-205; 1989. 16. Grupp, L. A.; Perlanski, E.; Stewart, R. B. Systemic angiotensin II does not act at the area postrema to suppress alcohol intake. Alcohol 8:165-167; 1991. 17. Grupp~ L. A.; Perlanski, E.; Stewart, R. B. Regulation of alcohol consumption by the renin-angiotensin system: A review of recent findings and a possible mechanism of action. Neurosci. Biobehav. Rev. 15:265-275; 1991. 18. Grupp, L. A.; Spinosa, G.; Lingham, T. Management of alcohol consumption with angiotensin converting enzyme inhibitors: A review of the animal findings. In: Naranjo, C. A.; Sellers, E. M., eds. Novel pharmacological interventions for alcoholism. New York: Springer-Verlag; 1992:201-214. 19. Grupp, L. A.; Chow, S. Y.-M. Angiotensin II mediates a condi-
363
20.
21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31.
32.
33.
34.
35.
tioned taste aversion in water-replete rats. Pharmacol. Biochem. Behav. 44:985-988; 1993. Grupp, L. A.; Harding, S. The reduction in alcohol drinking by peripherally injected angiotensin II is selectively mediated by the AT 1 receptor subtype. Pharmacol. Biochem. Behav. 47:385-392; 1994. Grupp, L. A.; Harding, S. Intracerebroventricularly infused angiotensin II or lII do not alter voluntary alcohol intake in rats. Pharmacol. Biochem. Behav. 51:593-599; 1995. Guivernau, M.; Quintanilla, M. E.: Tampier, L. Effect of dipyridamole and captropril on prostacyclin production and ALDH activity in rats. Alcohol. Clin. Exp. Res. 14:295; 1990. Hubbell, C. L.; Chrisbach, G. A.; Bilsky, E. J.; Reid, L. D. Manipulations of the renin-angiotensin system and intake of a sweetened alcoholic beverage among rats. Alcohol 9:53-61; 1992. Johnson, D. A.; Anderson, R. P. Alcohol ingestion following intracerebral angiotensin administration and water deprivation. Pharmacol. Biochem. Behav. 1:739-741; 1973. Kulkosky, P. J. Effect of addition of ethanol and NaCI on saccharin + glucose polydipsia. Pharmacol. Biochem. Behav. 10:277283; 1979. Kulkosky, P. J. Conditioned food aversions and satiety signals. Ann. NY Acad. Sci. 443:330-347; 1985. Kulkosky, P. J. Brain-gut neuropeptides and the limitation of ethanol consumption. Neurosci. Biobehav. Rev. 9:179-190; 1985. Kulkosky, P. J.; Sanchez, M. R.; Foderaro, M. A.; Chiu, N. Cholecystokinin and satiation with alcohol. Alcohol 6:395-402; 1989. Linseman, M. A. Alcohol consumption in free-feeding rats: Procedural, genetic and pharmacokinetic factors. Psychopharmacology (Berlin) 92:254-261; 1987. Litten, R. Z.; Allen, J. P. Reducing the desire to drink: Pharmacology and neurobiology. In: Galanter, M., ed. Recent developments in alcoholism, vol. 11. New York: Plenum; 1993:325-344. Naranjo, C. A.; Kadlec, K. E.; Sanhueza, P.; Woodley-Remus, D.; Sellers, E. M. Enalapril effects on alcohol intake and other consummatory behaviors in alcoholics. Clin. Pharmacol. Ther. 50:96106; 1991. Naranjo, C. A.; Bremner, K. E.; Sanhueza, P.; Sellers, E. M. Enalapril does not decrease alcohol intake in normotensive alcoholics. In: Naranjo, C. A.; Sellers, E. M., eds. Novel pharmacological interventions for alcoholism. New York: Springer-Verlag; 1992:215-225. Rabin, B. M.; Hunt, W. A.; Bakarich, A. C.; Chedester, A. L.: Lee, J. Angiotensin II-induced taste aversion learning in cats and rats and the role of the area postrema. Physiol. Behav. 36:11731178; 1986. Robertson, D.; Frolich, J. C.; Carr, R. K.; Watson, J. T.; Hollifield, J. W.; Shand, D. G.; Oates, J. A. Effects of caffeine on plasma renin activity, catecholamines and blood pressure. N. Engl. J. Med. 298:181-186; 1978. Ross, A. D.; Perlanski, E.; Grupp, L. A. Vasopressin does not mediate the inhibition of ethanol drinking by the renin-angiotensin system. Pharmacol. Biochem. Behav. 36:761-765: 1990.