Effects of caffeine and bombesin on ethanol and food intake

Effects of caffeine and bombesin on ethanol and food intake

Life Sciences, Vol. 48, pp. 1837-1844 Printed in the U.S.A. Pergamon Press EFFECTS OF CAFFEINE AND BOMBESIN ON ETHANOL AND FOOD INTAKE Max A. Dietze...

472KB Sizes 11 Downloads 76 Views

Life Sciences, Vol. 48, pp. 1837-1844 Printed in the U.S.A.

Pergamon Press

EFFECTS OF CAFFEINE AND BOMBESIN ON ETHANOL AND FOOD INTAKE Max A. Dietze and Paul J. Kulkosky Department of Psychology University of Southern Colorado Pueblo, CO 81001 U.S.A. (Received in final form March i, 1991)

Summary The methylxanthine caffeine and ethyl alcohol are widely used and powerful psychotropic drugs, but their interactions are not well understood. Bombesin is a brain-gut neuropeptide which is thought to function as a neurochemical factor in the inhibitory control of voluntary alcohol ingestion. We assessed the effects of combinations of intraperitoneal (i.p.) doses of caffeine (CAF, 0.1-50 mg/kg) and bombesin (BBS, 1-10/zg/kg) on 5% w/v ethanol solution and food intake in deprived rats. Deprived male and female Wistar rats received access to 5% ethanol or Purina chow for 30 minutes after i.p. injections. In single doses, CAF and BBS significantly decreased both ethanol and food consumption, at 50 mg/kg and 10/~g/kg, respectively. CAF and BBS combinations produced infra-additive, or lessthan-expected inhibitory effects on ethanol intake, but simple additive inhibitory effects on food intake. This experimental evidence suggests a reciprocal blocking of effects of CAF and BBS on ethanol intake but not food intake. Caffeine, when interacting with bombesin, increases alcohol consumption beyond expected values. Caffeine could affect the operation of endogenous satiety signals for alcohol consumption. The methylxanthine caffeine and ethyl alcohol are powerful psychotropic drugs which are widely used, although their combined effects are not well understood. A "research front" has recently been identified, by the Institute for Scientific Information, in the study of ethanol/caffeine interactions with nutrition. There are apparently conflicting reports of the main effect of caffeine on ethanol (ETOH) intake. Caffeine (CAF) administration has been shown to augment ethanol intake in marginally nourished and malnourished rats (1,2,3,4). In contrast, others reported that intraperitoneal injection of a high dose of CAF decreased ethanol intake (5). CAF also increases operant responding for food reinforcement at moderate doses, and decreases responding at high doses in animals (6,7,8,9). Previous studies of the effects of combinations of CAF and ETOH in various experimental paradigms have shown a variety of results. A high dose (25 mg/kg) of caffeine increases the stimulation of locomotion by ethanol (1-2 g/kg) in female mice, while a higher dose

Corresponding Author: Paul J. Kulkosky, Department of Psychology, University of Southern Colorado, Pueblo, CO 81001-490l. 0024-3205/91 $3.00 + .00 Copyright (c) 1991 Pergamon Press plc

1838

Caffeine, Bombesin and Ethanol

Vol. 48, No. 19, 1991

(100 mg/kg) abolished the ethanol-induced stimulation (10). Both CAF (25,50 mg/kg) and ETOH (0.5, 1 g/kg) separately and dose-dependently decrease measures of responding for water reinforcement in thirsty male rats, and in combination, CAF decreased and increased the effects of ETOH, depending on the specific measure of operant performance (11). A h;gh dose of CAF (60 mg/kg) reduced social behavior and locomotor activity, and antagonized ETOH-induced (2 g/kg) increases in locomotor activity in male mice (12). CAF (1 mM) and ETOH (0.95%) combinations were synergistically toxic in a motor performance task in guppies (!3). CAF (45,95 mg/kg) administered in drinking water increased ethanol-induced (1.5 g/kg) motor discoordination in male mice (14), and an injected high dose of CAF (62.5 mg/kg) increased ethanol-induced (1.75 g/kg) ataxia in male mice (15). CAF at moderate doses (I-10 mg/kg) attenuated ethanol's (0.5-3.0 g/kg) response-decreasing properties in a multiple-schedule operant in male pigeons (16). A large dose of CAF (100 mg/kg) had no significant effect on ethanolinduced (>1.25 g/kg) decreases in active avoidance behaviors in male rats (17), but high doses of CAF (100,15C mg/kg) potentiated the decrease of conditioned and unconditioned avoidance responses by alcohol (1 g/kg) in male rats (18). In conclusion, there are disparate reports on caffeine's effects on voluntary, behaviors. Caffeine's ~ctions are complexly dose-dependent and paradigm specific, and the direction of the effects produced by caffeine often reverse with increases in doses. Apparently nonspecific inhibitou; effects on behaviors, e.g., debilitation, have often been ,eported at relatively large doses ( > 25 mg/kg). Bombe~in (BBS) is a brain-gut peptide and a candidate ne.lromodulator which has been shown to inhibit intake of both food and ethanol (19-24). Little is known of potential interactions of CAF and BBS, although ~heophylline, a methylxanthine metabolite of caffeine, doubles Me secretion rate of bombesin from small ce.ll lung cancer cells in vitro (25). That finding suggests a possible interaction of caffeine witl. an endogenous neuropeptide which ,'nay function to regulate voluntar~ intake of ethanol. Caffeine consumption could affect a potential neuroendocrine c,mtrol of alcohol consumption, and thereby influ,mce the intake of alcohol. The following experiments attempt to clarify, firstly, the direct actions of CAF on ethanol and food intake, and secondly, the interactions of CAF with bombesin in the control of ethanol and food consumption. Obtained data should provide evidence on whether combinations of doses of CAF and BBS produce infra-additive, additive or supra-additive effects on ingestive behaviors. This informalion will help to determine if an interaction of caffeine with an endogenous neuropeptide can provide an explanation of some of caffeine's actions on ethanolrelated behavior. Method Animal~. Six male and nineteen female, experimentally naive, Wistar rats (Charles River, Crl: (WI)BR) were used in this experiment. The rats' weights ranged from 270-630 g and they were individually housed and maintained in wire-mesh, steel cages, in a room with ambient temperature of approximately 23 ° C, relative humidity of approximately 55 %, and a 12:12 light:dark lighting cycle (0700 on). The rats were provided with PurL-m Rodent Laboratory Chow (5001) and deionized water lid libitum, except as specified. Procedure. Phase 1: Ethanol Consumption. Eleven female and six male rats were deprived of water and continued on ad libitum access to food. Twenty-three hours later (1400), the rats were weighed and given access to a calibrated cylinder of 5% w/v ethanol for 30 minutes. At the end of thirty minutes, the ethanol bottles were taken off and measured for amount drunk. A calibrated tube of water was given for another thirty minutes and measured.

Vol. 48, No. 19, 1991

Caffeine, Bombesin and Ethanol

1839

This adaptation procedure was repeated for a total of 18 days, with two 1 ml/kg i.p. injections of 0.9% w/v NaCI (.saline) given immediately (within 1 minute) prior to presentation of ethanol on the last 5 days. At that point, mean ethanol consumption of the rats did not differ significantly (12> 0.05) across three consecutive days in a repeated measures analysis of variance. Following this baseline phase, rats were then randomly assigned to be injected i.p. with one dose of CAF (anhydrous, Sigma; 0.0, 1.0, 10.0, and 50.0 mg/kg) and one dose of BBS (tetradecapeptide, Sigma; 0.0, 1.0, and 10.0 ~g/kg). This procedure was repeated for a total of twelve days, until each rat had received all dose combinations. Data on ethanol consumption were analyzed with a 3 x 4 repeated measures analysis of variance, followed by Duncan's multiple range test, at an alpha significance level of 12<0.05. Phase 2: Food Consumption. Eight female rats were deprived of food and continued on ad libitum access to water. Twenty-three hours later (1400), the rats were weighed and given thirty minutes of access to five pellets of food weighing approximately 23 g. At the end of thirty minutes, the pellets were weighed and then returned to the rats for another thirty minutes. Apparent food consumption was corrected for spillage. This adaptation procedure was repeated for a total of 16 days, with two 1 ml/kg i.p. injections of saline given immediately prior to presentation of food on the last 9 days. At that point, mean food consumption of the rats did not differ significantly (12> 0.05) across three consecutive days in a repeated measures analysis of variance. Following this baseline phase, rats were then randomly assigned to be injected i.p. with one dose of CAF (0.0, 0.1, 1.0, 10.0, or 50.0 mg/kg) and then a dose of BBS (0.0, 1.0, or 10.0 ~g/kg). This procedure was repeated for fifteen days, until each rat had received all dose combinations. Data on food consumption were analyzed with a 3 x 5 repeated measures analysis of variance, followed by Duncan's multiple range test, at an alpha significance level of la<0.05, R¢~ults Phase 1. Figure 1 illustrates 5% ethanol intake of rats after i.p. injections of single doses and combinations of caffeine and bombesin. Preliminary analysis revealed no significant interaction of doses of CAF and BBS with sex of rats, F(i I, 165) = 1.09, p > 0.05; therefore data from both genders were combined in all subsequent analyses. Analysis revealed significant main effects of caffeine, F(3,48)=27.22, 12<0.05, and bombesin, F(2,32)=15.29, 12<0.05, on ethanol intake. Inspection of Fig. 1 shows that, in single doses, caffeine (50 mg/kg) and bombesin (10 vg/kg) suppressed mean alcohol consumption by 74.8 % and 65.8%, respectively (ps <0.05). Ethanol intake was correlated with single doses of caffeine, r(66)= -0.475, 12<0.05, and with bombesin, r(49)= -0.476, 12<0.05. The regression equation predicting ethanol intake (y, in ml/kg) from dose of caffeine (x, in mg/kg) is: y = -0.40x +27.26. The regression equation predicting ethanol intake from dose of bombesin (in ug/kg) is: y = -1.6Ix +25.73. Figure 2 represents the actual intake of ethanol (+95 % confidence interval) after combinations of doses of caffeine and bombesin, compared with the expected value of the intake. The expected value of intake was computed assuming that doses of CAF and BBS can substitute for one another in proportion to their relative potencies as determined from single-dose results (cf. 26,27). The combinations of CAF and BBS produced infra-dose-additive effects at all pairs of doses except 10+1 CAF+BBS. That is, suppression of ethanol intake by CAF+BBS combination was significantly less than predicted from the results of single doses of CAF and BBS.

1840

Vol. 48, No. 19, 1991

Caffeine, Bombesln and Ethanol



p
1.o

lo

vs.

#

4-0

35

30

25

,..q '-d

20

15

10

5

0

I.o

1.o

IO

0

lO

o

Bombesln

(/~g/kg)

1.o

1.0

to

o

10

50

Cedfeine ( m g / k g )

FIG. 1 Mean ( + standard error, SE) intake of 5% ethanol by rats after i.p. injections of doses of caffeine and bombesin.

[---7 E x p e c t e d 30

Actual

25 A 2o

1.0

~0

1.0

1.0 Bombesln

10

1.0

10

(/~g/kg)

10

.50

Car feir'.e ( m g / k g )

FIG. 2 Mean ( + 9 5 % confidence interval) intake of 5 % ethanol by rats after doses of caffeine and bombesin, compared with the expected value of the intake.

Vol. 48, No. 19, 1991

Caffeine, Bombesln and Ethanol

1841

Phase 2. Figure 3 displays food intake (g/kg) of rats after i.p. injections of single doses and combinations of caffeine and bombesin. Analysis showed significant main effects of caffeine F(4,28) = 9.93, 12< 0.05, and bombesin, F(2,14) = 14.13, p < 0.05, on food intake. As with ethanol intake, single doses of CAF (50 mg/kg) and BBS (10 #g/kg) suppressed mean intake of solid food by 35.3 % and 44.5 %, respectively (ps < 0.05). Food intakes were correlated with single doses of caffeine, r(38) = -0.385, 12<0.05, and with hombesin, r(22) = -0.400, p=0.05. The regression equation predicting food intake from dose of caffeine is: y -- -0.228x +32.52. The regression equation predicting food intake from dose of bombesin is: y = -1.37x +31.41. Figure 4 shows the actual intake of food after combinations of doses of caffeine and bombesin, compared with the expected value of the intake. The expected value of intake was computed assuming that the effects of doses of CAF and BBS were dose-additive (cf. 26,27). The combinations of CAF and BBS produced simple dose-additive effects at all pairs of doses. Suppression of food intake by each CAF + BBS combination was predictable, when suppressions produced by single doses of CAF and BBS were simply added together.

• p < 0 . 0 5 vs. #

4O

#

30

25

20 ,=

15

10

5

0

lo

1.o

1.o

lO

Bombesln (#g/kg) 0

0.1

1.0

10

50

Caffeine (mg/kg)

FIG. 3 Mean (__+SE) intake of solid food by rats after i.p. injection of doses of caffeine and bombesin.

1842

Caffeine,

Bombesin

Vol.

and E t h a n o l

48, No.

19, 1991

I-'-IExpected 50

Actual

35 30

25 20 15 10

5 0

1.0

10

1.O

10

1.0

10

10

1.0

Bombesln (#Mi/kg) 0.1

1.0

10

50

Caffeine(mg/kg)

FIG. 4 Mean (__+_95 % confidence interval) intake of solid food by rats after doses of caffeine and bombesin, compared with the expected value of the intake. Discussion The results show that both the methylxanthine caffeine and the brain-gut neuropeptide bombesin significantly reduce ethanol and food intake in deprived rats. These four main effects on ingestive behaviors are in accord with previous reports that high doses of caffeine reduce ethanol intake (5) and food intake (7) and that bombesin reduces ethanol intake (20-24) and food intake (19). These inhibitory main effects would appear consistent with the pharmacological use of caffeine in dietetic formulations (1) and the hypothesis that bombesin-like peptides are satiety signals in the control of caloric intake (19-24). However, the dose of CAF shown to suppress food intake (50.0 mg/kg) is within the range that produces toxicological effects (1) and the anorectic effect may be nonspecific. Nevertheless, this dose of CAF is within the range of doses which thirsty rats self-administer when caffeine is added (0.5 mg/ml) to tile available fluid (28), and is approached in humans by heavy coffee drinking ( > 15 mg/kg/day) (1,29,30). Further, we have obtained novel information on the behavioral effects of combinations of doses of caffeine and bombesin. Caffeine and bombesin doses typically combine in a infradose-additive manner to inhibit ethanol intake, but combine in a simple dose-additive manner to inhibit food intake. That is, pairs of doses of CAF and BBS inhibit alcohol consumption less

Vol. 48, No. 19, 1991

Caffeine, Bombesin and Ethanol

1843

than expected from results of single doses (26,27). Across a range of doses, less inhibition of alcohol intake is observed by CAF+BBS combinations than would be predicted from addition of the effects of single doses. This unexpected infra-dose-additivity of the homergic, inhibitory effects of CAF and BBS on ethanol intake may reflect a final common mechanism of action of these chemicals in the control of alcohol ingestion. Such a link between these chemicals is suggested by the finding that a methylxanthine, theophylline, a major metabolite of caffeine (29), can increase the rate of bombesin secretion from small cell lung cancer cells in vitro (25). However, that finding suggests an enhancement of the effects of BBS by CAF and not the observed mutual blocking in the present results. Therefore, further experiments on the biochemical mechanisms of CAF and BBS's effects on cellular activity are needed to clarify the nature of the interaction of CAF and BBS during alcohol intake (31). It is not clear why food intake was not similarly affected by combinations of CAF and BBS, although this could be related to the much higher caloric intake during the feeding on chow. Also, it is not known how fluid intakes other than ethanol solution intake would be affected by these combinations. Recent research findings suggest a further physiological process for the reduction of alcohol intake by caffeine. It has been shown that caffeine stimulates systemic release of renin (32), and activation of the renin-angiotensin system reduces voluntary ethanol intake in the rat (33,34). For example, angiotensin I1 acts specifically at the subfornical organ to reduce alcohol intake (35). Taken together with the present results, these findings indicate multiple potential interactions of caffeine with neuropeptide controls of ethanol intake. Our results indicate that CAF may influence the ability of BBS to produce decreases in alcohol intake. Thus, caffeine administration could conceivably result in greater-thananticipated alcohol intake, given that satiety signals such as BBS are operating, or exogenous bombesio is administered, or a bombesin secretagogue is consumed. The notion of caffeine as an antidote for alcohol over-consumption (cf. 36) is not compatible with our finding of an infradose-additive inhibitory interaction of CAF and BBS in the control of consumption of ethanol solution. Our results and others (37) indicate that caffeine can exert an influence on ingestive behaviors by interacting with endogenous peptide controls of intake. Acknowled~,ements This research was supported by Grant 2 S06 RR08197-08A1 funded by the Division of Research Resources and the National Institute on Alcohol Abuse and Alcoholism. The authors thank R. Boice, H.D. Moore, M.A. Foderaro, D.A. Marrinan and S.L. Sandoval for their helpful suggestions. Address reprint requests to P.J.K. References 1.

2. 3. 4. 5.

R.M. GILBERT, _Research Advances in Alcohol and Drua Problems, Vol. 3, R.J. Gibbins, Y. Israel, H. Kalant, R.E. Popham, W. Schmidt, and R.G. Smart (eds), 49-176, Wiley, NY (1979). R.M. GILBERT, J. Stud. Alcohol 37 11-18 (1976). R.M. GILBERT, J. Stud. Alcohol 40 159-162 (1979). U.D. REGISTER, S.R. MARSH, C.T. THURSTON, B.J. FIELDS, M.C. HORNING, M.G. HARDINGE and A. SANCItEZ, J. Am. Diet. Assoc. 61 159-162 (1972). A. HEDERRA, J. ALDUNATE, N. SEGOVIA-RIQUELME and J. MARDONES, The Effects of Centrally Active Dru~s on Voluntary Alcohol Consumption, J.D. Sinclair and K. Kiianmaa (eds), 9-13, Finnish Foundation for Alcohol Studies, Helsinki (1975).

1844

6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25.

26. 27.

28. 29.

Caffeine, Bombesin and Ethanol

Vol. 48, No. 19, 1991

M.E. KOSMAN and K.R. UNNA, Clin. Pharmacol. Ther. 9 240-254 (1967). W.A. McKIM, Psychopharmacology 68 135-138 (1980). D.E. McMILLAN, J. Pharmacol. Exp. Ther. 163 172-187 (1968). B.F. SKINNER and W.T. HERON, Psychol. Rec. 1 340-346 (1937). B. WALDEK, Psychopharmacologia 36 209-220 (1974). J. ELSNER, S, ALDERS and G. ZBINDEN, Psychopharmacology 96 194-205 (1988). L.A. HILAKIVI, M.J. DURCAN and R.G. LISTER, Life Sci. 44 543-553 (1989). M.E. POLLARD, Biochem. Arch. 4 117-124 (1988). M.S. DAR and W.R. WOLLES, Life Sci. 39 1429-1437 (1986). M.S. DAR, M. JONES, G. CLOSE, S.J. MUSTAFA and W.R. WOLLES, Psychopharmacology 91 1-4 (1987). M.L. HEALEY and L.A. DYKSTRA, Psychopharmacology 62 141-144 (1979). L.F. ALLEN, H.C. FERGUSON and G.R. McK/NNEY, Eur. J. Pharmacol 15 371-374 (1971). F.W. HUGHES and R.B. FORNEY, Proc. Soc. Exp. Biol. Med. 108 157-159 (1961). J. GIBBS, D.J. FAUSER, A.E. ROWE, B.J. ROLLS, E.T. ROLLS and S.P. MADDISON, Nature 282 208-210 (1979). G.W. GLAZNER, R.L. CANNON and P.J. KULKOSKY, Alcohol 5 325-329 (1988). P.J. KULKOSKY, Neurosci. Biobehav. Rev. 9 179-190 (1985). P.J. KULKOSKY and G.W. GLAZNER, Alcoholism: Clin. Exp. Res. 12 177-181 (1988). P.J. KULKOSKY, M. ROQUE and M.R. SANCHEZ, Peptides 6 103-106 (1985). P.J. KULKOSKY, M.R. SANCHEZ, N. CHIU and G.W. GLAZNER, Neuropharmacology 26 1211-1216 (1987). T.W. MOODY, T.L. O'DONOHUE, B.M. CHRONWALL, F. CUTTITTA, C.D. LINDEN, R.L. GETZ and S.S. WOLF, Peptide Hormones: Effects and Mechanisms of Action, Vol. II, A. Negro-Vilar and M. Conn (eds), 3-40, CRC Press, Boca Raton, FL (1988). W.D. WESSINGER, Neurosci. Biobehav. Rev. 10 102-112 (1986). W.L. WOOLVERTON, Neurobehavioral Pharmacology. Advances in Behavioral Pharmacology, T. Thompson, P.B. Dews, and J.E. BaITeR (eds), 275-302, Lawrence Earlbaum Assocs., Hillsdale, NJ (1987). P.J. KULKOSKY, E.W. HOLST, W.G. SMITH and M.A. DIETZE, Bull. Psychon. Soc. (in press). N.L. BENOWITZ, Annu. Rev. Med. 41 277-288 (1990).

30. R.M. GILBERT, J.A. MARSHMAN, M. SCHWIEDER and R. BERG, Can. Med. Assoc. J. 114 205-208 (1976). 31. S.H. SNYDER and P. SKLAR, J. Psychiat. Res. 18 91-106 (1984). 32. D. ROBERTSON, J.C. FROLICH, R.K. CARR, J.T. WATSON, J.W. HOLLIFIELD, D.G. SHAND and J.A. OATES, N. Eng. J. Med. 298 181-186 (1978). 33. L.A. GRUPP, Med. Hypoth. 24 11-19 (1987). 34. L.A. GRUPP, M. KILLIAN, E. PERLANSKI and R.B. STEWART, Pharmacol. Biochem. Behav. 29 479-482 (1988). 35. L.A. GRUPP, E. PERLANSKI and R.B. STEWART, Pharmacol. Biochem. Behav. 34 201205 (1989). 36. H. NASH, Alcohol and Caffeine: A Studv of their Psvchological Effects, Thomas, Springfield, IL (1962). 37. H.D. MOORE, M.S. thesis, Colorado State University (1990).