- Email: [email protected]
Differential Effects of Monoaminergic Agonists on Alcohol Intake in Rats Fed a Tryptophan-Enhanced Diet
Alcohol, Vol. 18, No. 1, pp. 55–64, 1999 © 1999 Elsevier Science Inc. Printed in the USA. All rights reserved 0741-8329/99/$–see front matter
PII S0741-8329(98)00068-8
Differential Effects of Monoaminergic Agonists on Alcohol Intake in Rats Fed a Tryptophan-Enhanced Diet A. K. HALLADAY,* G. C. WAGNER,* T. HSU,* A. SEKOWSKI† AND H. FISHER† Departments of *Psychology and †Nutritional Sciences, Rutgers University, New Brunswick, NJ Received 17 April 1998; Accepted 21 October 1998 HALLADAY, A. K., G. C. WAGNER, T. HSU, A. SEKOWSKI AND H. FISHER. Differential effects of monoaminergic agonists on alcohol intake in rats fed a tryptophan-enhanced diet. ALCOHOL 18(1) 55–64, 1999.—The goal of the present study was to determine if enhancement of tryptophan levels in a nutritionally balanced liquid diet would affect alcohol intake in a two-bottle choice procedure. Furthermore, the monoaminergic agonists amphetamine, phentermine (dopaminergic- and noradrenergic-releasing drugs), and fenfluramine (a serotonin releaser) were administered to determine if these drugs reduced alcohol intake in animals fed the tryptophan-enhanced diet compared to those fed an alcohol-containing diet without added tryptophan. Amphetamine 0.5 and 2 mg/kg and phentermine 4 mg/kg selectively reduced alcohol intake in animals fed the tryptophan-enhanced diet; higher doses also reduced alcohol intake in animals fed the control alcohol diet. Three hours after drug administration, phentermine 2 and 4 mg/kg produced increases in consumption of the nonalcoholic diet in animals fed the control diet without affecting consumption in animals fed the tryptophan-enhanced diet. Finally, animals in the tryptophan-enhanced group gained less weight than those animals fed an identical diet without the added tryptophan. Neurochemical analysis revealed that the tryptophan-fed groups showed increased 5-HIAA concentrations and serotonin turnover in the striatum, hypothalamus, and frontal cortex compared to animals fed the control diet. The tryptophan-alcohol group also showed almost double the tryptophan levels in the hypothalamus compared to the tryptophan-isocaloric group. These results indicate that, whereas increasing tryptophan levels by itself was not sufficient to alter consumption of an alcohol-containing diet, the administration of monoaminergic agonists significantly interacted with tryptophan in a dose-dependent manner to reduce intake of an alcohol-containing diet without reducing intake of an isocaloric diet. © 1999 Elsevier Science Inc. All rights reserved. Monoaminergic agonists
Tryptophan
Alcohol intake
tonin and dopamine. Phentermine, a noradrenergic and dopaminergic agonist, has shown to be an effective anorectic agent (9,15) as well as an antagonist of the cocaine-induced rise in mesolimbic dopamine (21). Given the interaction of dopamine and serotonin on alcohol intake and preference, we sought to determine what added benefit a dopaminergic agonist would have on animals fed a tryptophan-enhanced diet. Therefore, the purpose of the present study was to determine if a tryptophan-enhanced diet would increase the effectiveness of phentermine, amphetamine, and fenfluramine at reducing alcohol intake in a two-bottle choice test. These effects were examined in both animals rendered alcohol dependent and in those fed a nonalcoholic, isocaloric control diet. Finally, neurochemical analysis of brain tissue was performed
RECENT evidence has suggested that pharmacological interventions that increase both central serotonin and dopamine levels may have clinical relevance for the treatment of alcoholism (10–12,22,23). The combination of both amphetamine plus fenfluramine (25) and phentermine plus fenfluramine (8) has been shown to decrease alcohol intake in alcohol-dependent animals. However, the combination of phentermine plus fenfluramine exerted a greater than additive effect in reducing the intake of a nonalcoholic isocaloric control diet but not of an alcohol-containing diet (9). The recent decision of makers of d,l-fenfluramine and dexfenfluramine to stop the sale of the drug, due to reports of valvular heart damage, increased the demand for an alternative pharmacological treatment of both obesity and alcoholism that would increase central sero-
Requests for reprints should be addressed to Dr. H. Fisher, Department of Nutritional Sciences, Rutgers University, New Brunswick, NJ 08904. Tel: (732) 932-9825; Fax: (732) 932-6837.
55
56
HALLADAY ET AL.
following an 8-h withdrawal period to determine what enhanced dietary tryptophan produced with respect to neurochemical changes commonly associated with alcohol withdrawal.
METHOD
Sixty-four Long–Evans rats weighing 120–130 g at the beginning of the study were housed in suspended wire cages in a colony room under temperature and humidity control. The colony room was maintained on a 12-h light:12-h dark cycle. Rats were divided into four groups with 16 rats per group. Thirty-two rats were rendered physically dependent on alcohol with a continuous 8-day period of feeding a 6% alcoholcontaining liquid diet. Sixteen of these animals received an identical diet where 0.5% maltodextrin was replaced by the addition of 0.5% l-tryptophan. The remaining 32 rats received a control, isocaloric liquid diet (alcohol was replaced with maltodextrin for caloric balance) under the same regimen. Sixteen of these animals received a diet where, again, 0.5% maltodextrin was replaced by the addition of 0.5% l-tryptophan. Two-Bottle Choice On the eighth day, at 1130 h, each animal received an IP injection of saline. Subsequently, the animal’s respective diets were replaced with two identical 100-ml plastic tubes containing either the 6% alcohol-containing diet or the isocaloric control diet. Animals placed on the tryptophan-enhanced diet received the tryptophan-alcohol diet and the tryptophan isocaloric diet; those placed on the control diet received the control alcohol diet and the control isocaloric diet. Lights were turned off at 1200 h, and consumption measurements of both diets were made 3, 5.5, and 24 h following drug administration. This process was repeated every fourth day, whereupon 0.5, 1.0, and 2.0 mg/kg d-amphetamine sulfate (Sigma, St.
Louis, MO), 2.0, 4.0, and 8.0 mg/kg phentermine hydrochloride (Sigma), 1.0 mg/kg d,l-fenfluramine hydrochloride (Sigma), and a second saline injection were administered. Drugs were dissolved in 0.9% saline and injected at a volume of 1 ml/kg. Bottle placement was alternated compared to the previous choice period. Animals always had continuous access to liquid diet throughout the entire experiment. Between drug administration days, animals were placed back on their respective diets until baseline consumption was reached. Two days following the last injection, the second saline administration, animals rendered alcohol dependent were withdrawn from the alcohol-containing diet and given their respective nonalcoholic isocaloric control diet for a period of 8 h. The animals in the nonalcohol groups were maintained on their respective diets during this time. Half of the animals from each group were then injected with saline and sacrificed 1 h later. The brains were removed and dissected on a cold block and portions of the striatum, frontal cortex, and hypothalamus removed and placed in liquid nitrogen. These brain areas were selected for their high concentrations of dopamine, serotonin, and their metabolites; in addition, it has been previously reported that changes in monoamine levels occur in these regions after alcohol administration and withdrawal (9,17). HPLC Methods Tissue was homogenized in 0.5 ml (0.3 ml for hypothalamus) of 0.4 N perchloric acid with 0.1 mM ethylenediaminetetraacetic acid added to inhibit biochemical degradation. Samples were centrifuged at 14,000 3 g for 20 min at 48C and the supernatant assayed for dopamine, serotonin, their metabolites dihydroxyphenylacetic acid (DOPAC), homovanillic acid (HVA), and 5-hydroxyindoleacetic acid (5-HIAA), and hypothalamic tryptophan by high pressure liquid chromatography (Bioanalytical Systems, West Lafayette, IN). Samples were delivered through a high-pressure (Rheodyne) valve fit-
FIG. 1. Body weight over 36 days on diets. 1 Significantly different from tryptophan isocaloric-fed animals. ! Significantly different from control isocaloric-fed animals, using Fisher’s PLSD, p , 0.05. n 5 16 per group.
MONOAMINERGIC AGONIST EFFECTS ON ALCOHOL INTAKE ted with a 20-ml sample loop onto a Biophase ODS C-18 reverse-phase column (5 mm, 250 3 4.6 mm i.d.), and oxidized with a 10.72 V potential between the glassy carbon electrode and the Ag/AgCl reference electrode. For determination of tryptophan the potential was increased to 10.90 V. The mobile phase consisted of 0.1 M sodium phosphate (dibasic), 0.1 M citric acid, 5.0 mg EDTA, and 13% methanol; flow rate was 0.7 ml/min. Frontal cortex samples were run with a mobile phase of 0.1375 M sodium phosphate (dibasic) 0.0625 M citric acid, and 14% methanol. Quantification was against external standards injected between every six samples. Statistical Analysis Body weight for the full 36-day period and consumption during the first 8 days was analyzed using a two-factor repeated measures ANOVA with diet condition and day as the two main factors. During the two-bottle choice procedure, each separate time period was analyzed using a repeated measures ANOVA with diet and drug condition as the two factors. Analyses of alcohol diet intake and isocaloric diet intake were made separately for each time period. A factorial ANOVA was used to determine differences in monoamine and monoamine metabolite concentrations among the diet conditions. Fisher’s PLSD post hoc test was used to determine differences between groups.
57 RESULTS
Body Weight A repeated measures ANOVA revealed a significant overall effect of group, F(3, 59) 5 56.94, p , 0.0001, and day, F(11, 33) 5 1672.3, p , 0.0001, as well as an interaction of group and day, F(33, 649) 5 20.67, p , 0.0001, on body weight gain (Fig. 1). Both groups consuming the alcohol diets weighed less than their isocaloric diet controls; by day 3, only those animals fed the nonalcoholic diet showed significant body weight gain. Also, animals consuming the tryptophan-enhanced nonalcoholic diet weighed less than those on the control nonalcoholic diet by day 9. Tryptophan did not affect body weight gain in the groups receiving the alcohol-containing diets. Eight-Day Consumption Consumption measurements were taken by weight on the first 8 days of the study, prior to drug administration. There was a significant effect of group, F(3, 58) 5 58.84, p , 0.0001, and day, F(7, 21) 5 98.39, p , 0.0001, as well as an interaction of group and day on diet consumption, F(21, 406) 5 6.02, p , 0.0001. Animals on the tryptophan-enhanced alcohol diet consumed consistently more diet than those fed the control alcohol diet after the third day. This was not the case for the tryptophan-enhanced isocaloric diet group. These animals showed no significant differences in consumption compared
FIG. 2. Consumption of alcohol diet during two-bottle choice administration during hours 0–3 after drug administration and diet presentation. *Significantly different from saline-treated controls, using Fisher’s PLSD, p , 0.05. n 5 16 for tryptophan 1 EtOH, tryptophan 1 isocaloric, and control 1 isocaloric. n 5 13 for control 1 EtOH.
58
HALLADAY ET AL.
to the control isocaloric diet group, with the exception of days 3, 6, and 8, when they ate significantly less than the controls. Furthermore, those animals fed the alcohol-containing diets consumed less of their diets than those fed the nonalcoholic isocaloric control diets. Two-bottle Choice Consumption of the alcohol-containing and nonalcohol isocaloric diets during the two-bottle choice periods were analyzed separately. Two animals were excluded from the control plus alcohol group after failing Dixon’s test for outliers. The two saline measurements were averaged together and drug treatments were compared to each group’s saline-treated control. Each time period was measured separately using a twofactor repeated measures ANOVA. Hours 0–3. Consumption of the alcohol-containing diet. During the first 3 h of the two-bottle choice period, there was a significant effect of drug treatment, F(7, 21), p , 0.0001, and an interaction of diet condition and drug treatment, F(21, 399) 5 2.1, p , 0.004, on consumption of the alcohol-containing diets (Fig. 2). Amphetamine at 0.5 and 1 mg/kg reduced consumption of the alcohol-containing diet groups rendered alcohol dependent; however, at 2 mg/kg this drug selectively reduced alcohol intake in animals fed the tryptophan-enhanced alcohol diet. During this same period, phentermine 2 mg/kg also selectively reduced consumption of alcohol in those animals fed the tryptophan-enhanced diet. A higher dose of 8 mg/kg reduced consumption in both groups of animals rendered alcohol depen-
dent. Finally, fenfluramine at 1.0 mg/kg reduced intake of the alcohol-containing diet in both groups receiving the tryptophan-enhanced diet only. Consumption of the nonalcoholic isocaloric diet. There was a significant effect of diet, F(3, 58) 5 14.1, p , 0.0001, of drug treatment, F(7, 21) 5 10.6, p , 0.0001, and an interaction, F(21, 413) 5 2.0, p , 0.004 (Fig. 3). Amphetamine at 0.5 mg/ kg increased consumption of the nonalcoholic diet in the tryptophan-enhanced alcohol group; a higher dose of 1 mg/kg reduced consumption in both groups rendered alcohol dependent. Phentermine 8 mg/kg reduced consumption in all groups except the tryptophan-enhanced alcohol group. Fenfluramine, 1 mg/kg, was only effective in reducing intake of the isocaloric diet in the tryptophan control diets. Hours 3–5.5. Consumption of the alcohol-containing diet. During hours 3 to 5.5 of the two-bottle choice period, there was only a significant interaction of diet condition and drug treatment, F(21, 399) 5 1.85, p , 0.01, on consumption of the alcohol-containing diets (Fig. 4). Amphetamine at 0.5 mg/kg reduced consumption of the alcohol-containing diet in the tryptophan-enhanced alcohol diet group; however, at 1 mg/kg amphetamine significantly increased consumption of the alcohol diet in this group. During this same period, phentermine and fenfluramine did not alter consumption of the alcohol diet in any group. Consumption of the nonalcoholic isocaloric diet. During this period, there was a significant effect of diet, F(3, 59) 5 2.7, p , 0.05, and drug treatment, F(7, 21) 5 4.62, p,0.0001, with no significant interaction (Fig. 5). Amphetamine at 0.5 mg/kg and phentermine at 8 mg/kg increased consumption of the iso-
FIG. 3. Consumption of isocaloric diet during two-bottle choice administration during hours 0–3 after drug administration and diet presentation. *Significantly different from saline-treated controls, using Fisher’s PLSD, p , 0.05. n 5 16 for tryptophan 1 EtOH, tryptophan 1 isocaloric, and control 1 isocaloric. n 5 13 for control 1 EtOH.
MONOAMINERGIC AGONIST EFFECTS ON ALCOHOL INTAKE caloric diet in animals rendered alcohol dependent on the tryptophan-enhanced diet. Phentermine at 2 and 4 mg/kg significantly increased consumption of the isocaloric diet in the animals receiving the nontryptophan control alcohol diet without affecting consumption of the nonalcoholic isocaloric diet in any other group. Hours 5.5–24. Consumption of the alcohol-containing diet. During the last period (hours 5.5–24) of the two-bottle choice period, there was a significant effect of drug treatment, F(7, 21) 5 5.53, p , 0.0001, and an interaction of diet condition and drug treatment, F(21, 406) 5 1.86, p , 0.01, on consumption of the alcohol-containing diets (Fig. 6). Amphetamine at 0.5 mg/kg reduced consumption of the alcohol-containing diet in the nontryptophan control group; however, 1 mg/kg significantly increased consumption of the alcohol diet in all animals receiving the tryptophan-enhanced diets. During this same period, phentermine at 2 and 8 mg/kg reduced consumption of the alcohol diet in both groups rendered alcohol dependent. Phentermine at 4 mg/kg was only effective in reducing consumption in animals fed the nontryptophan-enhanced control alcohol diet. Finally, fenfluramine at 1.0 mg/kg reduced intake of the alcohol containing diet in both groups receiving the tryptophan-enhanced diet as well as those receiving the nontryptophan alcohol diet. Consumption of the nonalcoholic isocaloric diet. There was a significant effect of diet, F(3, 58) 5 5.93, p , 0.0013, of drug treatment, F(7, 21) 5 2.37, p , 0.02, and an interaction, F(21, 406) 5 1.79, p , 0.02 (Fig. 7). Amphetamine at 0.5 mg/kg re-
59
duced consumption of the isocaloric diet in animals receiving the nontryptophan-enhanced control isocaloric diet. Phentermine, 2 mg/kg, reduced consumption of the isocaloric diet in this same group. However, higher doses (4 and 8 mg/kg) significantly increased consumption of the nonalcohol diet in animals receiving the nontryptophan-enhanced alcohol diet without affecting those animals in the tryptophan-enhanced alcohol group. Fenfluramine, 1 mg/kg, was only effective in reducing intake of the isocaloric diet in the group receiving the nontryptophan control diet. Preference ratio. Analysis of the preference ratio during the first 90 min of consumption was not possible because intake values were too low following drug administration. During hours 1.5–3, no significant effects were found of diet, drug treatment, nor an interaction. During hours 3–5.5, there was a significant interaction of diet condition by drug treatment, F(21, 343) 5 1.682, p 5 0.03. At 1 mg/kg amphetamine, animals in the tryptophan alcohol group showed a higher preference for alcohol than the nontryptophan-enhanced alcohol group; however, under saline conditions, animals in the tryptophanenhanced alcohol-dependent group showed a lower preference ratio compared to the nontryptophan alcohol-fed animals. During hours 5–24, there was an effect of drug treatment, F(7, 21) 5 3.4, p 5 0.001, and an interaction, F(21, 399) 5 1.7, p 5 0.02. Again, amphetamine, 1 mg/kg, increased preference for alcohol in animals fed the tryptophan-enhanced alcohol diet; however, again the baseline saline preference measurement was lower in those fed the tryptophan-enhanced alcohol diet.
FIG. 4. Consumption of alcohol diet during two-bottle choice administration during hours 3–5.5 after drug administration and diet presentation. *Significantly different from saline-treated controls, using Fisher’s PLSD, p , 0.05. n 5 15 for tryptophan 1 EtOH and tryptophan 1 isocaloric. n 5 16 for control 1 isocaloric. n 5 14 for control 1 EtOH.
60 Neurochemistry Hypothalamus. There were significant differences in dopamine, F(3, 26) 5 3.23, p , 0.03, DOPAC, F(3, 21) 5 2.92, p , 0.05, and 5-HIAA, F(3, 23) 5 13.21, p , 0.0001, across the diet groups. Animals in the tryptophan alcohol group showed increased hypothalamic dopamine compared to their tryptophan isocaloric controls, but not compared to the nontryptophan alcohol group (Table 1). Furthermore, this group showed a decreased DOPAC concentration compared to the nontryptophan alcohol group, but not to the tryptophan isocaloric group. Both tryptophan groups showed higher levels of 5-HIAA. There was a significant effect of diet on tryptophan concentrations, F(3, 26) 5 45.589, p , 0.0001. The animals in the tryptophan alcohol group showed higher levels of tryptophan than the tryptophan isocaloric group, and almost five times the levels of the control groups. Striatum. As seen from Table 2, 5-HIAA levels were also increased in the striatum of animals fed the tryptophan-enhanced diet, F(3, 20) 5 11.07, p , 0.0002. In addition, levels in the tryptophan alcohol group were elevated compared to the tryptophan isocaloric group. Both groups receiving the tryptophan-enhanced diet showed enhanced serotonin turnover in the striatum compared to those fed a control, nontryptophanenhanced diet, F(3, 20) 5 6.68, p , 0.002. Tryptophan was not detected in the striatum or frontal cortex. Frontal cortex. There were no significant differences between groups on dopamine levels or dopamine metabolites (Table 3). However, there was a significant elevation in serotonin levels, F(3, 28) 5 12.94, p , 0.0001, and 5-HIAA concentrations, F(2, 28) 5 7,73, p , 0.006, in both groups fed the
HALLADAY ET AL. tryptophan-enhanced diet compared to those fed the nontryptophan diet. Furthermore, there was a significant drop in serotonin levels in the nontryptophan alcohol-fed animals compared to the nontryptophan isocaloric animals. There was also a significant increase in serotonin turnover, F(3, 28) 5 5.46, p , 0.0044, in the tryptophan alcohol group compared to both groups on the control diet.
DISCUSSION
Previous studies have confirmed the efficacy of dopaminergic and serotonergic agonists in the reduction of alcohol consumption (9,12,25). The search for a new serotonergic agonist to replace fenfluramine led us to try an enhanced tryptophan diet. Results indicate that the tryptophan-enhanced diet combined with a low dose of phentermine or amphetamine reduced the intake of alcohol while not affecting consumption of a nonalcoholic isocaloric diet. Phentermine was more effective than amphetamine in producing selective decreases in alcohol diet consumption without affecting consumption of the nonalcoholic control diet in animals on the tryptophanenhanced alcohol diet. There have been conflicting reports on the effect of tryptophan on alcohol consumption. Dietary tryptophan causing an increased preference for alcohol was reported in Long– Evans and Royal–Victoria rats but not for Sprague–Dawley rats (18). However, Lu et al. (16) reported that, in the absence of pharmacological manipulation, a tryptophan-enhanced diet did not significantly affect baseline alcohol intake in rats. In humans, tryptophan administration has been found to en-
FIG. 5. Consumption of isocaloric diet during two-bottle choice administration during hours 3–5.5 after drug administration and diet presentation. *Significantly different from saline-treated controls, using Fisher’s PLSD, p , 0.05. n 5 15 for tryptophan 1 EtOH and tryptophan 1 isocaloric. n 5 16 for control 1 isocaloric. n 5 14 for control 1 EtOH.
MONOAMINERGIC AGONIST EFFECTS ON ALCOHOL INTAKE
61
FIG. 6. Consumption of alcohol diet during two-bottle choice administration during hours 5.5–24 after drug administration and diet presentation. *Significantly different from saline-treated controls, using Fisher’s PLSD, p , 0.05. n 5 15 for tryptophan 1 EtOH. n 5 16 for tryptophan 1 isocaloric and control 1 isocaloric. n 5 14 for control 1 EtOH.
FIG. 7. Consumption of isocaloric diet during two-bottle choice administration during hours 5.5–24 after drug administration and diet presentation. *Significantly different from saline-treated controls, using Fisher’s PLSD, p , 0.05. n 5 15 for tryptophan 1 EtOH. n 5 16 for tryptophan 1 isocaloric and control 1 isocaloric. n 5 14 for control 1 EtOH.
62
HALLADAY ET AL. TABLE 1 NEUROTRANSMITTER CONCENTRATION IN THE RAT HYPOTHALAMUS
Group
Tryptophan alcohol Tryptophan isocaloric Control alcohol Conrol isocaloric
Dopamine
DOPAC
5-HT
5-HIAA
DOPAC/DA
5-HIAA/5-HT
Tryptophan
2.79 (0.344)† 2.21 (0.208) 2.29 (0.171) 1.84 (0.143)
0.057 (0.004) 0.060 (0.014) 0.162 (0.055)*† 0.093 (0.024)
0.737 (0.054) 0.710 (0.041) 0.390 (0.074) 0.556 (0.057)
0.561 (0.047) 0.469 (0.038) 0.270 (0.017)*† 0.299 (0.029)*†
0.019 (0.002) 0.031 (0.011) 0.072 (0.02)*† 0.055 (0.013)
0.767 (0.053) 0.664 (0.042) 0.467 (0.069)*† 0.553 (0.071)
15.82 (0.812) 8.806 (1.33)*†‡ 3.997 (0.243)* 3.929 (0.325)*
Animals were withdrawn from alcohol for 9 h but received an identical nonalcohol-containing isocaloric diet until sacrifice. Values are expressed as mg/g wet tissue. Values in parentheses are SEM. *Significantly different from tryptophan alcohol-fed animals. †Significantly different from tryptophan isocaloric-fed animals. ‡Significantly different from control isocaloric-fed animals, using Fisher’s PLSD, p , 0.05.
hance dysphoria resulting from alcohol in normal male volunteers (6). In the present study, the tryptophan-enhanced diet reduced body weight gain in control animals, an effect that was not apparent in the alcohol-fed groups; however, rats in both alcohol groups did not gain as much weight over the course of the study as their counterparts on the nonalcohol diets. Because weight gain may alter drug sensitivity, a partial replication of the present study is being conducted incorporating pair-fed control groups. A phentermine-induced increase in consumption of both an alcoholic and nonalcoholic diet 1–24 h after drug administration has been reported (9). In the present study, phentermine increased consumption of the nonalcoholic diet only in animals rendered alcohol dependent, but this effect was not seen in the groups receiving the tryptophan-enhanced diets. The effect of the tryptophan may be to block the phentermine-induced increase in consumption, though this was not evident at all time points. Amphetamine decreased consumption of an alcohol-containing diet while increasing consumption of an isocaloric diet in animals rendered alcohol dependent and receiving the tryptophan-enhanced diet. Although dopaminergic agonists were more effective in reducing intake of alcohol in animals fed a tryptophan-enhanced diet, the combined effects of serotonergic and dopaminergic agonists were clearly more selective for alcohol intake over intake of a nonalcohol-containing diet. This is consistent with other studies showing that tryptophan antagonized the increase in motor activity and stereotyped behavior consequent to d-amphetamine (13,24). Competing behaviors such as stereotypy were not measured or observed in the present study; the increase in consumption of the nonal-
coholic diet after amphetamine or phentermine indicates that, at these doses, such behaviors did not interfere with consumption levels. The present results are in agreement with previous studies in our lab demonstrating that a tryptophan-enriched diet enhanced the fenfluramine-induced decrease in alcohol consumption (16). During the first time point, fenfluramine was only effective in reducing intake in animals that received added dietary tryptophan. However, in animals rendered alcohol dependent, the drug selectively reduced alcohol intake. Additionally, reduction in intake of a control isocaloric diet during the last time period in animals not rendered alcohol dependent during the last time period displays this drug’s potent anorectic effects regardless of dietary condition. Neurochemical analysis of dopamine, serotonin, their major metabolites, and hypothalamic tryptophan following alcohol withdrawal revealed no differences in animals fed the control alcohol diet compared to those fed the control isocaloric diet, except in the frontal cortex where there was a decrease in serotonin. Interestingly, the addition of tryptophan to an alcohol-containing diet had a synergistic effect on hypothalamic tryptophan in that tryptophan levels in animals fed the alcohol-containing diet had almost five times the amount of tryptophan as those fed the control alcohol diet, whereas those fed the tryptophan isocaloric diet showed only double the concentration of serotonin. Other studies have linked tryptophan to enhanced brain serotonin concentrations in alcohol-fed animals, attributing this effect to decreased tryptophan pyrrolase activity consequent to chronic alcohol administration (1,2). These authors suggested that tryptophan may alter alcohol preference by altering catecholamine synthesis.
TABLE 2 NEUROTRANSMITTER CONCENTRATION IN THE RAT STRIATUM Group
Trytophan alcohol Tryptophan isocaloric Control alcohol Control isocaloric
Dopamine
DOPAC
HVA
5-HT
5-HIAA
DOPAC/DA
5-HIAA/5-HT
6.833 (1.11) 7.512 (1.24) 8.635 (0.946) 8.786 (0.758)
0.700 (0.126) 0.986 (0.227) 0.904 (0.136) 0.952 (0.128)
0.404 (0.077) 0.492 (0.084) 0.517 (0.086) 0.483 (0.068)
0.994 (0.172) 0.854 (0.090) 0.650 (0.056) 0.805 (0.066)
0.700 (0.060) 0.510 (0.087)*‡ 0.264 (0.034)*† 0.272 (0.022)*†
0.128 (0.046) 0.109 (0.019) 0.103 (0.007) 0.109 (0.012)
0.676 (0.069) 0.581 (0.073) 0.363 (0.053)*† 0.339 (0.044)*†
Animals were withdrawn from alcohol for 9 h but received an identical nonalcohol-containing isocaloric diet until sacrifice. Values are expressed as mg/g wet tissue. Values in parentheses are SEM. *Significantly different from tryptophan alcohol-fed animals. †Significantly different from tryptophan isocaloric-fed animals. ‡Significantly different from control isocaloric-fed animals, using Fisher’s PLSD, p , 0.05.
MONOAMINERGIC AGONIST EFFECTS ON ALCOHOL INTAKE
63
TABLE 3 NEUROTRANSMITTER CONCENTRATION IN THE RAT FRONTAL CORTEX Group
Tryptophan alcohol Tryptophan isocaloric Control alcohol Control isocaloric
Dopamine
DOPAC
HVA
5-HT
5-HIAA
DOPAC/DA
5-HIAA/5-HT
0.277 (0.48) 0.488 (0.136) 0.660 (0.299) 0.302 (0.091)
0.031 (0.007) 0.070 (0.016) 0.073 (0.026) 0.060 (0.014)
0.132 (0.029) 0.123 (0.015) 0.126 (0.025) 0.131 (0.013)
0.446 (0.017) 0.422 (0.008) 0.335 (0.017)*†‡ 0.374 (0.009)*†
0.283 (0.031) 0.215 (0.014) 0.146 (0.033)*† 0.137 (0.012)*†
0.137 (0.011) 0.165 (0.027) 0.142 (0.020) 0.388 (0.158)
0.635 (0.058) 0.509 (0.031) 0.422 (0.070)* 0.365 (0.030)*
Animals were withdrawn from alcohol for 9 h but received an identical nonalcohol-containing isocaloric diet until sacrifice. Values are expressed as mg/g wet tissue. Values in parentheses are SEM. *Significantly different from tryptophan alcohol-fed animals. †Significantly different from tryptophan isocaloric-fed animals. ‡Significantly different from control isocaloric-fed animals, using Fisher’s PLSD, p , 0.05.
In contrast to these findings, Bano et al. (4) reported a dramatic increase in hepatic tryptophan pyrrolase activity leading to a drop in brain tryptophan and 5-HIAA levels following a 7-h withdrawal period from chronic alcohol exposure. The authors suggest that this increase in enzyme activity as well as mRNA expression could account for the decrease in serotonergic concentration following alcohol withdrawal. The differences in our findings could be attributed to the discrete brain areas examined in the present investigation, as well as the tryptophan-enhanced diet administered during alcohol administration and withdrawal. This dietary supplement may have provided the animals with tryptophan sufficient to protect the animals against serotonergic deficiency, increases in tryptophan pyrrolase activity, and withdrawal symptoms. The results of the current study further the argument that both serotonergic and dopaminergic manipulations are needed to selectively reduce alcohol intake. Clinical studies have demonstrated the efficacy of serotonin agonists such as fenfluramine, selective serotonin uptake inhibitors (19), as well as serotonin receptor agonists (24) in the treatment of alcohol abuse. However, the above agents exert unpleasant side ef-
fects such as sweating, nausea, and somnolence (3,20). Clinical studies have reported only a transient effect of fluoxetine alone on reduction of alcohol intake (7), or a failure of fluoxetine alone to prevent relapse in alcoholics (14). Phentermine has been reported to reduce the aversive effects of fenfluramine in healthy volunteers, and the positive and arousing effects of phentermine were offset by fenfluramine (5). Along these lines, phentermine plus fenfluramine have an additivetype relationship in reducing alcohol intake while working synergistically to abolish alcohol withdrawal seizures (9). Tryptophan has not been used clinically in the treatment of alcoholism, and animals on the tryptophan-enhanced alcohol diet did not consume less of the alcohol-containing diet in the choice procedure compared to those fed the control alcohol diet. Tryptophan suppressed the increase in consumption of an isocaloric diet subsequent to phentermine administration in alcohol-fed animals. Finally, although dopaminergic agonists suppressed intake of an alcohol-containing diet, this effect was nonspecific; only in cases where there was both enhanced serotonergic and dopaminergic functioning was there a selective effect on reduction of alcohol intake.
REFERENCES 1. Badaway, A. A.-B.; Punjani, N. F.; Evans, M.: Enhancement of rat brain tryptophan metabolism by chronic ethanol administration and possible involvement of decreased liver tryptophan pyrrolase activity. Biochem. J. 178:575–580; 1979. 2. Badaway, A. A.-B.; Evans, M.; Punjani, M.: Enhancement of rat brain metabolism of a tryptohan load by chronic ethanol administration. Br. J. Pharmacol. 68:22–24; 1980. 3. Bergeron, R.; Blier, P.: Cipraside for the treatment of nausea produced by selective serotonin reuptake inhibitors. Am. J. Psychiatry 151:1084–1086; 1994. 4. Bano, S.; Oretti, R. G.; Morgan, C. J.; Badaway, A. A-B.; Buckland, P. R.; McGuffin, P.: Effects of chronic administration and subsequent withdrawal of ethanol-containing liquid diet on rat liver tryptophan pyrrolase and tryptophan metabolism. Alcohol Alcohol. 31:205–215; 1996. 5. Brauer, L. H.; Johanson, C-E.; Schuster, C. R.; Rothman, R. B.; deWit, H.: Evaluation of phentermine and fenfluramine alone and in combination, in normal, healthy volunteers. Neuropsychopharmacology 14:233–241; 1996. 6. Clayton, C. J.; Hicks, R. E.: Ethanol, monoamines, and affect. J. Neural Transm. 98:169–195; 1994. 7. Gorelick, D. A.; Paredes, A.: Effect of fluoxetine on alcohol consumption in male alcoholics. Alcohol. Clin. Exp. Res. 16:261–265; 1992. 8. Halladay, A. K.; Fisher, H.; Wagner, G. C.: Behavioral and neu-
9.
10.
11. 12. 13.
14.
15.
rochemical effects of phentermine plus fenfluramine. Soc. Neurosci. Abstr. 23:551.9; 1997. Halladay, A. K.: Effects of phentermine plus fenfluramine on alcohol consumption and alcohol-withdrawal seizures in rats. Unpublished master’s thesis: Rutgers University, New Brunswick, NJ; 1998. Higgins, G. A.; Tomkins, D. M.; Fletcher, P. J.; Sellers, E. M.: Effect of drugs influencing 5-HT function on ethanol drinking and feeding behavior in rats: studies using a drinkometer system. Neurosci. Biobehav. Rev. 16:535–552; 1992. Hitzig, P.: Combined dopamine and serotonin agonists: A synergistic approach to alcoholism and other addictive behaviors. Md Med. J. 42:153–156; 1993. Hitzig, P.: Combined serotonin and dopamine indirect agonists correct alcohol craving and alcohol-associated neuroses. J. Subst. Abuse Treat. 11:489–490; 1994. Hollister, A. S.; Breese, G. R.; Kuhn, C. M.; Cooper, B. R.; Schanberg, S. M.: An inhibitory role for brain serotonin containing systems in the locomotor effects of d-amphetamine. J. Pharmacol. Exp. Ther. 198:12–22; 1976. Kabel, D. K.; Petty, F.: A placebo-controlled, double-blind study of fluoxetine in severe alcohol dependence: Adjunctive pharmacotherapy during and after inpatient treatment. Alcohol. Clin. Exp. Res. 20:780–784; 1996. Lawlor, R. B.; Trivedi, M. C.; Yelnosky, J.: A determination of
64
16. 17.
18. 19.
HALLADAY ET AL. the anorexigenic potential of dl-amphetamine, d-amphetamine, l-amphetamine and phentermine. Arch. Int. Pharmacodyn. 179:401– 407; 1969. Lu, M.-R.; Wagner, G. C.; Fisher, H.: Ethanol consumption following acute fenfluramine, fluoxetine, and dietary tryptophan. Pharmacol. Biochem. Behav. 44:931–937; 1993. Mirovsky, Y.; Wagner, G. C.; Sekowski, A; Goldberg, M.; Fisher, H.: Simultaneous changes in striatal dopamine, serotonin and metabolites after withdrawal seizures in rats from dependence on alcohol. Alcohol 12:251–256; 1995. Myers, R. D.; Melchior C. L.: Dietary tryptophan and the selection of ethyl alcohol in different strains of rats. Psychopharmacologia 42:109–115; 1975. Naranjo, C. A.; Sellers, E. M.: Novel pharmacological interventions for alcoholism. New York: Springer Verlag; 1992.
20. Nelson, J. C.: Safety and tolerability of the new antidepressants. J. Clin. Psychiatry. 58(Suppl. 6):26–31; 1997. 21. Rothman, R. B.; Ayestas, M.; Baumann, M. H.: Phentermine pretreatment antagonizes the cociane-induced rise in mesolimbic dopamine. Neuroreport 8:7–9; 1996. 22. Rothman, R. B.; Gendron, R.; Hitzig, P.: Letter to the editor. J. Subst. Abuse Treat. 11:273–275; 1994. 23. Sellers, E. M.; Higgins, G. A.; Sobell, M. B.: 5-HT and alcohol abuse. Trends Pharmacol. Sci. 13:69–75; 1992. 24. Taylor, M.: Dietary modification of amphetamine stereotyped behavior: The action of tryptophan, methionine and lysine. Psychopharmacology (Berlin) 61:81–83; 1979. 25. Yu, Y.-L.; Fisher, H.; Sekowski, A.; Wagner, G. C.: Amphetamine and fenfluramine suppress ethanol intake in ethanoldependent rats. Alcohol 14:45–48; 1997.
PII S0741-8329(98)00068-8
Differential Effects of Monoaminergic Agonists on Alcohol Intake in Rats Fed a Tryptophan-Enhanced Diet A. K. HALLADAY,* G. C. WAGNER,* T. HSU,* A. SEKOWSKI† AND H. FISHER† Departments of *Psychology and †Nutritional Sciences, Rutgers University, New Brunswick, NJ Received 17 April 1998; Accepted 21 October 1998 HALLADAY, A. K., G. C. WAGNER, T. HSU, A. SEKOWSKI AND H. FISHER. Differential effects of monoaminergic agonists on alcohol intake in rats fed a tryptophan-enhanced diet. ALCOHOL 18(1) 55–64, 1999.—The goal of the present study was to determine if enhancement of tryptophan levels in a nutritionally balanced liquid diet would affect alcohol intake in a two-bottle choice procedure. Furthermore, the monoaminergic agonists amphetamine, phentermine (dopaminergic- and noradrenergic-releasing drugs), and fenfluramine (a serotonin releaser) were administered to determine if these drugs reduced alcohol intake in animals fed the tryptophan-enhanced diet compared to those fed an alcohol-containing diet without added tryptophan. Amphetamine 0.5 and 2 mg/kg and phentermine 4 mg/kg selectively reduced alcohol intake in animals fed the tryptophan-enhanced diet; higher doses also reduced alcohol intake in animals fed the control alcohol diet. Three hours after drug administration, phentermine 2 and 4 mg/kg produced increases in consumption of the nonalcoholic diet in animals fed the control diet without affecting consumption in animals fed the tryptophan-enhanced diet. Finally, animals in the tryptophan-enhanced group gained less weight than those animals fed an identical diet without the added tryptophan. Neurochemical analysis revealed that the tryptophan-fed groups showed increased 5-HIAA concentrations and serotonin turnover in the striatum, hypothalamus, and frontal cortex compared to animals fed the control diet. The tryptophan-alcohol group also showed almost double the tryptophan levels in the hypothalamus compared to the tryptophan-isocaloric group. These results indicate that, whereas increasing tryptophan levels by itself was not sufficient to alter consumption of an alcohol-containing diet, the administration of monoaminergic agonists significantly interacted with tryptophan in a dose-dependent manner to reduce intake of an alcohol-containing diet without reducing intake of an isocaloric diet. © 1999 Elsevier Science Inc. All rights reserved. Monoaminergic agonists
Tryptophan
Alcohol intake
tonin and dopamine. Phentermine, a noradrenergic and dopaminergic agonist, has shown to be an effective anorectic agent (9,15) as well as an antagonist of the cocaine-induced rise in mesolimbic dopamine (21). Given the interaction of dopamine and serotonin on alcohol intake and preference, we sought to determine what added benefit a dopaminergic agonist would have on animals fed a tryptophan-enhanced diet. Therefore, the purpose of the present study was to determine if a tryptophan-enhanced diet would increase the effectiveness of phentermine, amphetamine, and fenfluramine at reducing alcohol intake in a two-bottle choice test. These effects were examined in both animals rendered alcohol dependent and in those fed a nonalcoholic, isocaloric control diet. Finally, neurochemical analysis of brain tissue was performed
RECENT evidence has suggested that pharmacological interventions that increase both central serotonin and dopamine levels may have clinical relevance for the treatment of alcoholism (10–12,22,23). The combination of both amphetamine plus fenfluramine (25) and phentermine plus fenfluramine (8) has been shown to decrease alcohol intake in alcohol-dependent animals. However, the combination of phentermine plus fenfluramine exerted a greater than additive effect in reducing the intake of a nonalcoholic isocaloric control diet but not of an alcohol-containing diet (9). The recent decision of makers of d,l-fenfluramine and dexfenfluramine to stop the sale of the drug, due to reports of valvular heart damage, increased the demand for an alternative pharmacological treatment of both obesity and alcoholism that would increase central sero-
Requests for reprints should be addressed to Dr. H. Fisher, Department of Nutritional Sciences, Rutgers University, New Brunswick, NJ 08904. Tel: (732) 932-9825; Fax: (732) 932-6837.
55
56
HALLADAY ET AL.
following an 8-h withdrawal period to determine what enhanced dietary tryptophan produced with respect to neurochemical changes commonly associated with alcohol withdrawal.
METHOD
Sixty-four Long–Evans rats weighing 120–130 g at the beginning of the study were housed in suspended wire cages in a colony room under temperature and humidity control. The colony room was maintained on a 12-h light:12-h dark cycle. Rats were divided into four groups with 16 rats per group. Thirty-two rats were rendered physically dependent on alcohol with a continuous 8-day period of feeding a 6% alcoholcontaining liquid diet. Sixteen of these animals received an identical diet where 0.5% maltodextrin was replaced by the addition of 0.5% l-tryptophan. The remaining 32 rats received a control, isocaloric liquid diet (alcohol was replaced with maltodextrin for caloric balance) under the same regimen. Sixteen of these animals received a diet where, again, 0.5% maltodextrin was replaced by the addition of 0.5% l-tryptophan. Two-Bottle Choice On the eighth day, at 1130 h, each animal received an IP injection of saline. Subsequently, the animal’s respective diets were replaced with two identical 100-ml plastic tubes containing either the 6% alcohol-containing diet or the isocaloric control diet. Animals placed on the tryptophan-enhanced diet received the tryptophan-alcohol diet and the tryptophan isocaloric diet; those placed on the control diet received the control alcohol diet and the control isocaloric diet. Lights were turned off at 1200 h, and consumption measurements of both diets were made 3, 5.5, and 24 h following drug administration. This process was repeated every fourth day, whereupon 0.5, 1.0, and 2.0 mg/kg d-amphetamine sulfate (Sigma, St.
Louis, MO), 2.0, 4.0, and 8.0 mg/kg phentermine hydrochloride (Sigma), 1.0 mg/kg d,l-fenfluramine hydrochloride (Sigma), and a second saline injection were administered. Drugs were dissolved in 0.9% saline and injected at a volume of 1 ml/kg. Bottle placement was alternated compared to the previous choice period. Animals always had continuous access to liquid diet throughout the entire experiment. Between drug administration days, animals were placed back on their respective diets until baseline consumption was reached. Two days following the last injection, the second saline administration, animals rendered alcohol dependent were withdrawn from the alcohol-containing diet and given their respective nonalcoholic isocaloric control diet for a period of 8 h. The animals in the nonalcohol groups were maintained on their respective diets during this time. Half of the animals from each group were then injected with saline and sacrificed 1 h later. The brains were removed and dissected on a cold block and portions of the striatum, frontal cortex, and hypothalamus removed and placed in liquid nitrogen. These brain areas were selected for their high concentrations of dopamine, serotonin, and their metabolites; in addition, it has been previously reported that changes in monoamine levels occur in these regions after alcohol administration and withdrawal (9,17). HPLC Methods Tissue was homogenized in 0.5 ml (0.3 ml for hypothalamus) of 0.4 N perchloric acid with 0.1 mM ethylenediaminetetraacetic acid added to inhibit biochemical degradation. Samples were centrifuged at 14,000 3 g for 20 min at 48C and the supernatant assayed for dopamine, serotonin, their metabolites dihydroxyphenylacetic acid (DOPAC), homovanillic acid (HVA), and 5-hydroxyindoleacetic acid (5-HIAA), and hypothalamic tryptophan by high pressure liquid chromatography (Bioanalytical Systems, West Lafayette, IN). Samples were delivered through a high-pressure (Rheodyne) valve fit-
FIG. 1. Body weight over 36 days on diets. 1 Significantly different from tryptophan isocaloric-fed animals. ! Significantly different from control isocaloric-fed animals, using Fisher’s PLSD, p , 0.05. n 5 16 per group.
MONOAMINERGIC AGONIST EFFECTS ON ALCOHOL INTAKE ted with a 20-ml sample loop onto a Biophase ODS C-18 reverse-phase column (5 mm, 250 3 4.6 mm i.d.), and oxidized with a 10.72 V potential between the glassy carbon electrode and the Ag/AgCl reference electrode. For determination of tryptophan the potential was increased to 10.90 V. The mobile phase consisted of 0.1 M sodium phosphate (dibasic), 0.1 M citric acid, 5.0 mg EDTA, and 13% methanol; flow rate was 0.7 ml/min. Frontal cortex samples were run with a mobile phase of 0.1375 M sodium phosphate (dibasic) 0.0625 M citric acid, and 14% methanol. Quantification was against external standards injected between every six samples. Statistical Analysis Body weight for the full 36-day period and consumption during the first 8 days was analyzed using a two-factor repeated measures ANOVA with diet condition and day as the two main factors. During the two-bottle choice procedure, each separate time period was analyzed using a repeated measures ANOVA with diet and drug condition as the two factors. Analyses of alcohol diet intake and isocaloric diet intake were made separately for each time period. A factorial ANOVA was used to determine differences in monoamine and monoamine metabolite concentrations among the diet conditions. Fisher’s PLSD post hoc test was used to determine differences between groups.
57 RESULTS
Body Weight A repeated measures ANOVA revealed a significant overall effect of group, F(3, 59) 5 56.94, p , 0.0001, and day, F(11, 33) 5 1672.3, p , 0.0001, as well as an interaction of group and day, F(33, 649) 5 20.67, p , 0.0001, on body weight gain (Fig. 1). Both groups consuming the alcohol diets weighed less than their isocaloric diet controls; by day 3, only those animals fed the nonalcoholic diet showed significant body weight gain. Also, animals consuming the tryptophan-enhanced nonalcoholic diet weighed less than those on the control nonalcoholic diet by day 9. Tryptophan did not affect body weight gain in the groups receiving the alcohol-containing diets. Eight-Day Consumption Consumption measurements were taken by weight on the first 8 days of the study, prior to drug administration. There was a significant effect of group, F(3, 58) 5 58.84, p , 0.0001, and day, F(7, 21) 5 98.39, p , 0.0001, as well as an interaction of group and day on diet consumption, F(21, 406) 5 6.02, p , 0.0001. Animals on the tryptophan-enhanced alcohol diet consumed consistently more diet than those fed the control alcohol diet after the third day. This was not the case for the tryptophan-enhanced isocaloric diet group. These animals showed no significant differences in consumption compared
FIG. 2. Consumption of alcohol diet during two-bottle choice administration during hours 0–3 after drug administration and diet presentation. *Significantly different from saline-treated controls, using Fisher’s PLSD, p , 0.05. n 5 16 for tryptophan 1 EtOH, tryptophan 1 isocaloric, and control 1 isocaloric. n 5 13 for control 1 EtOH.
58
HALLADAY ET AL.
to the control isocaloric diet group, with the exception of days 3, 6, and 8, when they ate significantly less than the controls. Furthermore, those animals fed the alcohol-containing diets consumed less of their diets than those fed the nonalcoholic isocaloric control diets. Two-bottle Choice Consumption of the alcohol-containing and nonalcohol isocaloric diets during the two-bottle choice periods were analyzed separately. Two animals were excluded from the control plus alcohol group after failing Dixon’s test for outliers. The two saline measurements were averaged together and drug treatments were compared to each group’s saline-treated control. Each time period was measured separately using a twofactor repeated measures ANOVA. Hours 0–3. Consumption of the alcohol-containing diet. During the first 3 h of the two-bottle choice period, there was a significant effect of drug treatment, F(7, 21), p , 0.0001, and an interaction of diet condition and drug treatment, F(21, 399) 5 2.1, p , 0.004, on consumption of the alcohol-containing diets (Fig. 2). Amphetamine at 0.5 and 1 mg/kg reduced consumption of the alcohol-containing diet groups rendered alcohol dependent; however, at 2 mg/kg this drug selectively reduced alcohol intake in animals fed the tryptophan-enhanced alcohol diet. During this same period, phentermine 2 mg/kg also selectively reduced consumption of alcohol in those animals fed the tryptophan-enhanced diet. A higher dose of 8 mg/kg reduced consumption in both groups of animals rendered alcohol depen-
dent. Finally, fenfluramine at 1.0 mg/kg reduced intake of the alcohol-containing diet in both groups receiving the tryptophan-enhanced diet only. Consumption of the nonalcoholic isocaloric diet. There was a significant effect of diet, F(3, 58) 5 14.1, p , 0.0001, of drug treatment, F(7, 21) 5 10.6, p , 0.0001, and an interaction, F(21, 413) 5 2.0, p , 0.004 (Fig. 3). Amphetamine at 0.5 mg/ kg increased consumption of the nonalcoholic diet in the tryptophan-enhanced alcohol group; a higher dose of 1 mg/kg reduced consumption in both groups rendered alcohol dependent. Phentermine 8 mg/kg reduced consumption in all groups except the tryptophan-enhanced alcohol group. Fenfluramine, 1 mg/kg, was only effective in reducing intake of the isocaloric diet in the tryptophan control diets. Hours 3–5.5. Consumption of the alcohol-containing diet. During hours 3 to 5.5 of the two-bottle choice period, there was only a significant interaction of diet condition and drug treatment, F(21, 399) 5 1.85, p , 0.01, on consumption of the alcohol-containing diets (Fig. 4). Amphetamine at 0.5 mg/kg reduced consumption of the alcohol-containing diet in the tryptophan-enhanced alcohol diet group; however, at 1 mg/kg amphetamine significantly increased consumption of the alcohol diet in this group. During this same period, phentermine and fenfluramine did not alter consumption of the alcohol diet in any group. Consumption of the nonalcoholic isocaloric diet. During this period, there was a significant effect of diet, F(3, 59) 5 2.7, p , 0.05, and drug treatment, F(7, 21) 5 4.62, p,0.0001, with no significant interaction (Fig. 5). Amphetamine at 0.5 mg/kg and phentermine at 8 mg/kg increased consumption of the iso-
FIG. 3. Consumption of isocaloric diet during two-bottle choice administration during hours 0–3 after drug administration and diet presentation. *Significantly different from saline-treated controls, using Fisher’s PLSD, p , 0.05. n 5 16 for tryptophan 1 EtOH, tryptophan 1 isocaloric, and control 1 isocaloric. n 5 13 for control 1 EtOH.
MONOAMINERGIC AGONIST EFFECTS ON ALCOHOL INTAKE caloric diet in animals rendered alcohol dependent on the tryptophan-enhanced diet. Phentermine at 2 and 4 mg/kg significantly increased consumption of the isocaloric diet in the animals receiving the nontryptophan control alcohol diet without affecting consumption of the nonalcoholic isocaloric diet in any other group. Hours 5.5–24. Consumption of the alcohol-containing diet. During the last period (hours 5.5–24) of the two-bottle choice period, there was a significant effect of drug treatment, F(7, 21) 5 5.53, p , 0.0001, and an interaction of diet condition and drug treatment, F(21, 406) 5 1.86, p , 0.01, on consumption of the alcohol-containing diets (Fig. 6). Amphetamine at 0.5 mg/kg reduced consumption of the alcohol-containing diet in the nontryptophan control group; however, 1 mg/kg significantly increased consumption of the alcohol diet in all animals receiving the tryptophan-enhanced diets. During this same period, phentermine at 2 and 8 mg/kg reduced consumption of the alcohol diet in both groups rendered alcohol dependent. Phentermine at 4 mg/kg was only effective in reducing consumption in animals fed the nontryptophan-enhanced control alcohol diet. Finally, fenfluramine at 1.0 mg/kg reduced intake of the alcohol containing diet in both groups receiving the tryptophan-enhanced diet as well as those receiving the nontryptophan alcohol diet. Consumption of the nonalcoholic isocaloric diet. There was a significant effect of diet, F(3, 58) 5 5.93, p , 0.0013, of drug treatment, F(7, 21) 5 2.37, p , 0.02, and an interaction, F(21, 406) 5 1.79, p , 0.02 (Fig. 7). Amphetamine at 0.5 mg/kg re-
59
duced consumption of the isocaloric diet in animals receiving the nontryptophan-enhanced control isocaloric diet. Phentermine, 2 mg/kg, reduced consumption of the isocaloric diet in this same group. However, higher doses (4 and 8 mg/kg) significantly increased consumption of the nonalcohol diet in animals receiving the nontryptophan-enhanced alcohol diet without affecting those animals in the tryptophan-enhanced alcohol group. Fenfluramine, 1 mg/kg, was only effective in reducing intake of the isocaloric diet in the group receiving the nontryptophan control diet. Preference ratio. Analysis of the preference ratio during the first 90 min of consumption was not possible because intake values were too low following drug administration. During hours 1.5–3, no significant effects were found of diet, drug treatment, nor an interaction. During hours 3–5.5, there was a significant interaction of diet condition by drug treatment, F(21, 343) 5 1.682, p 5 0.03. At 1 mg/kg amphetamine, animals in the tryptophan alcohol group showed a higher preference for alcohol than the nontryptophan-enhanced alcohol group; however, under saline conditions, animals in the tryptophanenhanced alcohol-dependent group showed a lower preference ratio compared to the nontryptophan alcohol-fed animals. During hours 5–24, there was an effect of drug treatment, F(7, 21) 5 3.4, p 5 0.001, and an interaction, F(21, 399) 5 1.7, p 5 0.02. Again, amphetamine, 1 mg/kg, increased preference for alcohol in animals fed the tryptophan-enhanced alcohol diet; however, again the baseline saline preference measurement was lower in those fed the tryptophan-enhanced alcohol diet.
FIG. 4. Consumption of alcohol diet during two-bottle choice administration during hours 3–5.5 after drug administration and diet presentation. *Significantly different from saline-treated controls, using Fisher’s PLSD, p , 0.05. n 5 15 for tryptophan 1 EtOH and tryptophan 1 isocaloric. n 5 16 for control 1 isocaloric. n 5 14 for control 1 EtOH.
60 Neurochemistry Hypothalamus. There were significant differences in dopamine, F(3, 26) 5 3.23, p , 0.03, DOPAC, F(3, 21) 5 2.92, p , 0.05, and 5-HIAA, F(3, 23) 5 13.21, p , 0.0001, across the diet groups. Animals in the tryptophan alcohol group showed increased hypothalamic dopamine compared to their tryptophan isocaloric controls, but not compared to the nontryptophan alcohol group (Table 1). Furthermore, this group showed a decreased DOPAC concentration compared to the nontryptophan alcohol group, but not to the tryptophan isocaloric group. Both tryptophan groups showed higher levels of 5-HIAA. There was a significant effect of diet on tryptophan concentrations, F(3, 26) 5 45.589, p , 0.0001. The animals in the tryptophan alcohol group showed higher levels of tryptophan than the tryptophan isocaloric group, and almost five times the levels of the control groups. Striatum. As seen from Table 2, 5-HIAA levels were also increased in the striatum of animals fed the tryptophan-enhanced diet, F(3, 20) 5 11.07, p , 0.0002. In addition, levels in the tryptophan alcohol group were elevated compared to the tryptophan isocaloric group. Both groups receiving the tryptophan-enhanced diet showed enhanced serotonin turnover in the striatum compared to those fed a control, nontryptophanenhanced diet, F(3, 20) 5 6.68, p , 0.002. Tryptophan was not detected in the striatum or frontal cortex. Frontal cortex. There were no significant differences between groups on dopamine levels or dopamine metabolites (Table 3). However, there was a significant elevation in serotonin levels, F(3, 28) 5 12.94, p , 0.0001, and 5-HIAA concentrations, F(2, 28) 5 7,73, p , 0.006, in both groups fed the
HALLADAY ET AL. tryptophan-enhanced diet compared to those fed the nontryptophan diet. Furthermore, there was a significant drop in serotonin levels in the nontryptophan alcohol-fed animals compared to the nontryptophan isocaloric animals. There was also a significant increase in serotonin turnover, F(3, 28) 5 5.46, p , 0.0044, in the tryptophan alcohol group compared to both groups on the control diet.
DISCUSSION
Previous studies have confirmed the efficacy of dopaminergic and serotonergic agonists in the reduction of alcohol consumption (9,12,25). The search for a new serotonergic agonist to replace fenfluramine led us to try an enhanced tryptophan diet. Results indicate that the tryptophan-enhanced diet combined with a low dose of phentermine or amphetamine reduced the intake of alcohol while not affecting consumption of a nonalcoholic isocaloric diet. Phentermine was more effective than amphetamine in producing selective decreases in alcohol diet consumption without affecting consumption of the nonalcoholic control diet in animals on the tryptophanenhanced alcohol diet. There have been conflicting reports on the effect of tryptophan on alcohol consumption. Dietary tryptophan causing an increased preference for alcohol was reported in Long– Evans and Royal–Victoria rats but not for Sprague–Dawley rats (18). However, Lu et al. (16) reported that, in the absence of pharmacological manipulation, a tryptophan-enhanced diet did not significantly affect baseline alcohol intake in rats. In humans, tryptophan administration has been found to en-
FIG. 5. Consumption of isocaloric diet during two-bottle choice administration during hours 3–5.5 after drug administration and diet presentation. *Significantly different from saline-treated controls, using Fisher’s PLSD, p , 0.05. n 5 15 for tryptophan 1 EtOH and tryptophan 1 isocaloric. n 5 16 for control 1 isocaloric. n 5 14 for control 1 EtOH.
MONOAMINERGIC AGONIST EFFECTS ON ALCOHOL INTAKE
61
FIG. 6. Consumption of alcohol diet during two-bottle choice administration during hours 5.5–24 after drug administration and diet presentation. *Significantly different from saline-treated controls, using Fisher’s PLSD, p , 0.05. n 5 15 for tryptophan 1 EtOH. n 5 16 for tryptophan 1 isocaloric and control 1 isocaloric. n 5 14 for control 1 EtOH.
FIG. 7. Consumption of isocaloric diet during two-bottle choice administration during hours 5.5–24 after drug administration and diet presentation. *Significantly different from saline-treated controls, using Fisher’s PLSD, p , 0.05. n 5 15 for tryptophan 1 EtOH. n 5 16 for tryptophan 1 isocaloric and control 1 isocaloric. n 5 14 for control 1 EtOH.
62
HALLADAY ET AL. TABLE 1 NEUROTRANSMITTER CONCENTRATION IN THE RAT HYPOTHALAMUS
Group
Tryptophan alcohol Tryptophan isocaloric Control alcohol Conrol isocaloric
Dopamine
DOPAC
5-HT
5-HIAA
DOPAC/DA
5-HIAA/5-HT
Tryptophan
2.79 (0.344)† 2.21 (0.208) 2.29 (0.171) 1.84 (0.143)
0.057 (0.004) 0.060 (0.014) 0.162 (0.055)*† 0.093 (0.024)
0.737 (0.054) 0.710 (0.041) 0.390 (0.074) 0.556 (0.057)
0.561 (0.047) 0.469 (0.038) 0.270 (0.017)*† 0.299 (0.029)*†
0.019 (0.002) 0.031 (0.011) 0.072 (0.02)*† 0.055 (0.013)
0.767 (0.053) 0.664 (0.042) 0.467 (0.069)*† 0.553 (0.071)
15.82 (0.812) 8.806 (1.33)*†‡ 3.997 (0.243)* 3.929 (0.325)*
Animals were withdrawn from alcohol for 9 h but received an identical nonalcohol-containing isocaloric diet until sacrifice. Values are expressed as mg/g wet tissue. Values in parentheses are SEM. *Significantly different from tryptophan alcohol-fed animals. †Significantly different from tryptophan isocaloric-fed animals. ‡Significantly different from control isocaloric-fed animals, using Fisher’s PLSD, p , 0.05.
hance dysphoria resulting from alcohol in normal male volunteers (6). In the present study, the tryptophan-enhanced diet reduced body weight gain in control animals, an effect that was not apparent in the alcohol-fed groups; however, rats in both alcohol groups did not gain as much weight over the course of the study as their counterparts on the nonalcohol diets. Because weight gain may alter drug sensitivity, a partial replication of the present study is being conducted incorporating pair-fed control groups. A phentermine-induced increase in consumption of both an alcoholic and nonalcoholic diet 1–24 h after drug administration has been reported (9). In the present study, phentermine increased consumption of the nonalcoholic diet only in animals rendered alcohol dependent, but this effect was not seen in the groups receiving the tryptophan-enhanced diets. The effect of the tryptophan may be to block the phentermine-induced increase in consumption, though this was not evident at all time points. Amphetamine decreased consumption of an alcohol-containing diet while increasing consumption of an isocaloric diet in animals rendered alcohol dependent and receiving the tryptophan-enhanced diet. Although dopaminergic agonists were more effective in reducing intake of alcohol in animals fed a tryptophan-enhanced diet, the combined effects of serotonergic and dopaminergic agonists were clearly more selective for alcohol intake over intake of a nonalcohol-containing diet. This is consistent with other studies showing that tryptophan antagonized the increase in motor activity and stereotyped behavior consequent to d-amphetamine (13,24). Competing behaviors such as stereotypy were not measured or observed in the present study; the increase in consumption of the nonal-
coholic diet after amphetamine or phentermine indicates that, at these doses, such behaviors did not interfere with consumption levels. The present results are in agreement with previous studies in our lab demonstrating that a tryptophan-enriched diet enhanced the fenfluramine-induced decrease in alcohol consumption (16). During the first time point, fenfluramine was only effective in reducing intake in animals that received added dietary tryptophan. However, in animals rendered alcohol dependent, the drug selectively reduced alcohol intake. Additionally, reduction in intake of a control isocaloric diet during the last time period in animals not rendered alcohol dependent during the last time period displays this drug’s potent anorectic effects regardless of dietary condition. Neurochemical analysis of dopamine, serotonin, their major metabolites, and hypothalamic tryptophan following alcohol withdrawal revealed no differences in animals fed the control alcohol diet compared to those fed the control isocaloric diet, except in the frontal cortex where there was a decrease in serotonin. Interestingly, the addition of tryptophan to an alcohol-containing diet had a synergistic effect on hypothalamic tryptophan in that tryptophan levels in animals fed the alcohol-containing diet had almost five times the amount of tryptophan as those fed the control alcohol diet, whereas those fed the tryptophan isocaloric diet showed only double the concentration of serotonin. Other studies have linked tryptophan to enhanced brain serotonin concentrations in alcohol-fed animals, attributing this effect to decreased tryptophan pyrrolase activity consequent to chronic alcohol administration (1,2). These authors suggested that tryptophan may alter alcohol preference by altering catecholamine synthesis.
TABLE 2 NEUROTRANSMITTER CONCENTRATION IN THE RAT STRIATUM Group
Trytophan alcohol Tryptophan isocaloric Control alcohol Control isocaloric
Dopamine
DOPAC
HVA
5-HT
5-HIAA
DOPAC/DA
5-HIAA/5-HT
6.833 (1.11) 7.512 (1.24) 8.635 (0.946) 8.786 (0.758)
0.700 (0.126) 0.986 (0.227) 0.904 (0.136) 0.952 (0.128)
0.404 (0.077) 0.492 (0.084) 0.517 (0.086) 0.483 (0.068)
0.994 (0.172) 0.854 (0.090) 0.650 (0.056) 0.805 (0.066)
0.700 (0.060) 0.510 (0.087)*‡ 0.264 (0.034)*† 0.272 (0.022)*†
0.128 (0.046) 0.109 (0.019) 0.103 (0.007) 0.109 (0.012)
0.676 (0.069) 0.581 (0.073) 0.363 (0.053)*† 0.339 (0.044)*†
Animals were withdrawn from alcohol for 9 h but received an identical nonalcohol-containing isocaloric diet until sacrifice. Values are expressed as mg/g wet tissue. Values in parentheses are SEM. *Significantly different from tryptophan alcohol-fed animals. †Significantly different from tryptophan isocaloric-fed animals. ‡Significantly different from control isocaloric-fed animals, using Fisher’s PLSD, p , 0.05.
MONOAMINERGIC AGONIST EFFECTS ON ALCOHOL INTAKE
63
TABLE 3 NEUROTRANSMITTER CONCENTRATION IN THE RAT FRONTAL CORTEX Group
Tryptophan alcohol Tryptophan isocaloric Control alcohol Control isocaloric
Dopamine
DOPAC
HVA
5-HT
5-HIAA
DOPAC/DA
5-HIAA/5-HT
0.277 (0.48) 0.488 (0.136) 0.660 (0.299) 0.302 (0.091)
0.031 (0.007) 0.070 (0.016) 0.073 (0.026) 0.060 (0.014)
0.132 (0.029) 0.123 (0.015) 0.126 (0.025) 0.131 (0.013)
0.446 (0.017) 0.422 (0.008) 0.335 (0.017)*†‡ 0.374 (0.009)*†
0.283 (0.031) 0.215 (0.014) 0.146 (0.033)*† 0.137 (0.012)*†
0.137 (0.011) 0.165 (0.027) 0.142 (0.020) 0.388 (0.158)
0.635 (0.058) 0.509 (0.031) 0.422 (0.070)* 0.365 (0.030)*
Animals were withdrawn from alcohol for 9 h but received an identical nonalcohol-containing isocaloric diet until sacrifice. Values are expressed as mg/g wet tissue. Values in parentheses are SEM. *Significantly different from tryptophan alcohol-fed animals. †Significantly different from tryptophan isocaloric-fed animals. ‡Significantly different from control isocaloric-fed animals, using Fisher’s PLSD, p , 0.05.
In contrast to these findings, Bano et al. (4) reported a dramatic increase in hepatic tryptophan pyrrolase activity leading to a drop in brain tryptophan and 5-HIAA levels following a 7-h withdrawal period from chronic alcohol exposure. The authors suggest that this increase in enzyme activity as well as mRNA expression could account for the decrease in serotonergic concentration following alcohol withdrawal. The differences in our findings could be attributed to the discrete brain areas examined in the present investigation, as well as the tryptophan-enhanced diet administered during alcohol administration and withdrawal. This dietary supplement may have provided the animals with tryptophan sufficient to protect the animals against serotonergic deficiency, increases in tryptophan pyrrolase activity, and withdrawal symptoms. The results of the current study further the argument that both serotonergic and dopaminergic manipulations are needed to selectively reduce alcohol intake. Clinical studies have demonstrated the efficacy of serotonin agonists such as fenfluramine, selective serotonin uptake inhibitors (19), as well as serotonin receptor agonists (24) in the treatment of alcohol abuse. However, the above agents exert unpleasant side ef-
fects such as sweating, nausea, and somnolence (3,20). Clinical studies have reported only a transient effect of fluoxetine alone on reduction of alcohol intake (7), or a failure of fluoxetine alone to prevent relapse in alcoholics (14). Phentermine has been reported to reduce the aversive effects of fenfluramine in healthy volunteers, and the positive and arousing effects of phentermine were offset by fenfluramine (5). Along these lines, phentermine plus fenfluramine have an additivetype relationship in reducing alcohol intake while working synergistically to abolish alcohol withdrawal seizures (9). Tryptophan has not been used clinically in the treatment of alcoholism, and animals on the tryptophan-enhanced alcohol diet did not consume less of the alcohol-containing diet in the choice procedure compared to those fed the control alcohol diet. Tryptophan suppressed the increase in consumption of an isocaloric diet subsequent to phentermine administration in alcohol-fed animals. Finally, although dopaminergic agonists suppressed intake of an alcohol-containing diet, this effect was nonspecific; only in cases where there was both enhanced serotonergic and dopaminergic functioning was there a selective effect on reduction of alcohol intake.
REFERENCES 1. Badaway, A. A.-B.; Punjani, N. F.; Evans, M.: Enhancement of rat brain tryptophan metabolism by chronic ethanol administration and possible involvement of decreased liver tryptophan pyrrolase activity. Biochem. J. 178:575–580; 1979. 2. Badaway, A. A.-B.; Evans, M.; Punjani, M.: Enhancement of rat brain metabolism of a tryptohan load by chronic ethanol administration. Br. J. Pharmacol. 68:22–24; 1980. 3. Bergeron, R.; Blier, P.: Cipraside for the treatment of nausea produced by selective serotonin reuptake inhibitors. Am. J. Psychiatry 151:1084–1086; 1994. 4. Bano, S.; Oretti, R. G.; Morgan, C. J.; Badaway, A. A-B.; Buckland, P. R.; McGuffin, P.: Effects of chronic administration and subsequent withdrawal of ethanol-containing liquid diet on rat liver tryptophan pyrrolase and tryptophan metabolism. Alcohol Alcohol. 31:205–215; 1996. 5. Brauer, L. H.; Johanson, C-E.; Schuster, C. R.; Rothman, R. B.; deWit, H.: Evaluation of phentermine and fenfluramine alone and in combination, in normal, healthy volunteers. Neuropsychopharmacology 14:233–241; 1996. 6. Clayton, C. J.; Hicks, R. E.: Ethanol, monoamines, and affect. J. Neural Transm. 98:169–195; 1994. 7. Gorelick, D. A.; Paredes, A.: Effect of fluoxetine on alcohol consumption in male alcoholics. Alcohol. Clin. Exp. Res. 16:261–265; 1992. 8. Halladay, A. K.; Fisher, H.; Wagner, G. C.: Behavioral and neu-
9.
10.
11. 12. 13.
14.
15.
rochemical effects of phentermine plus fenfluramine. Soc. Neurosci. Abstr. 23:551.9; 1997. Halladay, A. K.: Effects of phentermine plus fenfluramine on alcohol consumption and alcohol-withdrawal seizures in rats. Unpublished master’s thesis: Rutgers University, New Brunswick, NJ; 1998. Higgins, G. A.; Tomkins, D. M.; Fletcher, P. J.; Sellers, E. M.: Effect of drugs influencing 5-HT function on ethanol drinking and feeding behavior in rats: studies using a drinkometer system. Neurosci. Biobehav. Rev. 16:535–552; 1992. Hitzig, P.: Combined dopamine and serotonin agonists: A synergistic approach to alcoholism and other addictive behaviors. Md Med. J. 42:153–156; 1993. Hitzig, P.: Combined serotonin and dopamine indirect agonists correct alcohol craving and alcohol-associated neuroses. J. Subst. Abuse Treat. 11:489–490; 1994. Hollister, A. S.; Breese, G. R.; Kuhn, C. M.; Cooper, B. R.; Schanberg, S. M.: An inhibitory role for brain serotonin containing systems in the locomotor effects of d-amphetamine. J. Pharmacol. Exp. Ther. 198:12–22; 1976. Kabel, D. K.; Petty, F.: A placebo-controlled, double-blind study of fluoxetine in severe alcohol dependence: Adjunctive pharmacotherapy during and after inpatient treatment. Alcohol. Clin. Exp. Res. 20:780–784; 1996. Lawlor, R. B.; Trivedi, M. C.; Yelnosky, J.: A determination of
64
16. 17.
18. 19.
HALLADAY ET AL. the anorexigenic potential of dl-amphetamine, d-amphetamine, l-amphetamine and phentermine. Arch. Int. Pharmacodyn. 179:401– 407; 1969. Lu, M.-R.; Wagner, G. C.; Fisher, H.: Ethanol consumption following acute fenfluramine, fluoxetine, and dietary tryptophan. Pharmacol. Biochem. Behav. 44:931–937; 1993. Mirovsky, Y.; Wagner, G. C.; Sekowski, A; Goldberg, M.; Fisher, H.: Simultaneous changes in striatal dopamine, serotonin and metabolites after withdrawal seizures in rats from dependence on alcohol. Alcohol 12:251–256; 1995. Myers, R. D.; Melchior C. L.: Dietary tryptophan and the selection of ethyl alcohol in different strains of rats. Psychopharmacologia 42:109–115; 1975. Naranjo, C. A.; Sellers, E. M.: Novel pharmacological interventions for alcoholism. New York: Springer Verlag; 1992.
20. Nelson, J. C.: Safety and tolerability of the new antidepressants. J. Clin. Psychiatry. 58(Suppl. 6):26–31; 1997. 21. Rothman, R. B.; Ayestas, M.; Baumann, M. H.: Phentermine pretreatment antagonizes the cociane-induced rise in mesolimbic dopamine. Neuroreport 8:7–9; 1996. 22. Rothman, R. B.; Gendron, R.; Hitzig, P.: Letter to the editor. J. Subst. Abuse Treat. 11:273–275; 1994. 23. Sellers, E. M.; Higgins, G. A.; Sobell, M. B.: 5-HT and alcohol abuse. Trends Pharmacol. Sci. 13:69–75; 1992. 24. Taylor, M.: Dietary modification of amphetamine stereotyped behavior: The action of tryptophan, methionine and lysine. Psychopharmacology (Berlin) 61:81–83; 1979. 25. Yu, Y.-L.; Fisher, H.; Sekowski, A.; Wagner, G. C.: Amphetamine and fenfluramine suppress ethanol intake in ethanoldependent rats. Alcohol 14:45–48; 1997.