Departments of
Medicine (J.C.F., N.E.B.-E., R.W.Z., D.E.M.) and
Physiology/Biophysics (J.C.F.) and
Indiana Program in Medical
Neurobiology (J.C.F., R.W.Z.), Indiana University School of Medicine,
Indianapolis, Indiana; and
Department of Medicinal Chemistry,
University of Minnesota College of Pharmacy, Minneapolis, Minnesota
(P.S.P.)
 |
Introduction |
The
effects of alcohol are biphasic, rewarding during the rising portion of
the BAC curve, and aversive or dysphoric during the falling portion of
the curve (Pohorecky, 1977
; Risinger and Cunningham 1992). A genetic
propensity toward high alcohol drinking is determined by multiple genes
that regulate many predisposing neurobiological factors (Froehlich,
1995; Froehlich and Wand, 1997). One of these factors may be increased
sensitivity to, or lower threshold for, alcohol's reinforcing effects
that occur during the rising portion of the BAC curve (Froehlich and
Li, 1993; 1994; Krishnan-Sarin et al., 1998; Li et
al., 1998). Another factor may be decreased sensitivity to, or
higher threshold for, alcohol's aversive effects that occur during the
falling portion of the BAC curve. Activation of the endogenous opioid
system, as a consequence of alcohol ingestion, may influence both of
these processes (Froehlich and Wand, 1996).
Endorphin and enkephalin peptides induce feelings of well-being,
euphoria and pleasantness (Belluzzi and Stein, 1977; Bozarth
and Wise, 1984) and function as positive reinforcers in a variety of
behavioral paradigms (Goeders et al., 1984; Almaric et
al., 1987; Shippenberg et al., 1992). We have
previously postulated that alcohol-induced activation of the endogenous
opioid system during the rising portion of the BAC curve enhances the
hedonic value of alcohol and contribute to the positively reinforcing properties of the drug (Froehlich et al., 1991;
Krishnan-Sarin et al., 1995a, 1995b; Froehlich, 1997). This
may serve to support continued drinking during a single drinking bout
and to increase the probability of subsequent drinking. Repeated bouts
of alcohol drinking characterize rodents selectively bred for alcohol
preference (P line), but not rodents bred for alcohol nonpreference (NP
line) (Files et al., 1992). We now postulate that
alcohol-induced activation of the endogenous opioid system during the
falling portion of the BAC curve also contributes to drinking by
counteracting some of the aversive consequences of alcohol ingestion,
such as dysphoria or malaise, that can begin when blood alcohol levels
are falling and extend long after blood alcohol levels have reached
zero. In humans, a session of intoxication may be followed by malaise, nausea, headache and hyperirritability which are commonly referred to
as "hangover" and are clearly aversive. Whether something akin to
"hangover" occurs in the rat after alcohol exposure is less clear
but rebound hyperthermia, hyperactivity and increased vocalization during handling have been reported at 16-24 hr after a single intraperitoneal (i.p.) injection of alcohol (2.5 g/kg) and these aftereffects of acute intoxication have been described as hangover signs (Sinclair and Gustafsson, 1987; Sinclair and Tiara, 1988).
We postulate that an elevated threshold for, or reduced sensitivity to,
the aversive effects of alcohol may contribute, in part, to high
alcohol drinking in rats selectively bred for alcohol preference (P
line). This view is supported by our prior finding that P rats have a
higher alcohol aversion threshold than do NP rats as measured in a
conditioned taste aversion (CTA) test (Froehlich et al.,
1988). We further postulate that this line difference in alcohol
aversion may be due to differences in sensitivity of the opioid system
to the postingestive effects of alcohol. Specifically, the aversive
effects of alcohol, experienced after an initial exposure to an
intoxicating dose of alcohol, may be attenuated by alcohol-induced
activation of the opioid system in P rats. This view is supported by
our recent finding that an intragastric (i.g.) infusion of alcohol (2.5 g/kg BW) elevates genetic message for beta-endorphin synthesis (POMC
mRNA levels) in the pituitary of P but not NP rats at 8 hr after onset
of the infusion, at a time when blood alcohol content (BAC) has
returned to near preinfusion levels (Froehlich, 1995). While dysphoric
symptoms may arise after initial exposure to an intoxicating dose of
alcohol in both P and NP rats, greater alcohol-induced activation of
the opioid system in P rats may serve to attenuate dysphoria and
subsequent aversion to alcohol and contribute to high alcohol drinking
in this rat line. If so, one might expect that blocking the action of
endogenous opioid peptides, by pretreatment with an opioid receptor
antagonist, would lower the alcohol aversion threshold in P rats. One
approach to measuring alcohol aversion threshold is to use a CTA test
which involves pairing the consumption of a novel and detectable taste
solution with the administration of alcohol. Aversion to alcohol is
indicated by a reduction of intake of the taste solution in subsequent
exposures (Froehlich et al., 1988). A CTA test was used in
the present study to examine whether administration of the selective
delta opioid receptor antagonist, naltrindole (NTI), would
lower the alcohol aversion threshold in P rats.
| |
Methods |
Subjects.
Alcohol-naive female rats that were selectively
bred for alcohol preference (P line) served as subjects. The P rat
line, from which these subjects were taken, was derived from a
foundation stock of outbred Wistar rats. Rats of the P line are
selected for breeding based on the results of a 4-week oral alcohol
preference test (Li et al., 1981; Lumeng et al.,
1977). Each generation of rats is given alcohol as the sole source of
fluid for 4 days followed by 4 weeks of ad libitum
free-choice between a 10% (v/v) alcohol solution and water with food
freely available. Alcohol intake is calculated for each rat during the
four weeks of preference testing and alcohol intake is expressed as g
alcohol/kg BW/day. Rats selected for breeding in each generation are
those that consume more than 5 g alcohol/kg BW/day and demonstrate
a greater than 2:1 preference ratio of alcohol to water. Rats of the P
line represent an animal model that is useful for examinations of the
biological mechanisms underlying alcohol drinking behavior.
The P rats which served as subjects in the present studies were taken
from generations 39 to 41 of selection for alcohol preference but were
not tested for alcohol preference, and hence were alcohol-naive. All
rats from the 39 to 41 generations of selection that did not serve as
subjects in the present study were tested for alcohol preference. Rats
from the 39th generation of selection consumed an average of 7.1 g
alcohol/kg BW/day ± 0.23 (S.E.) with a range of 3.7-11.87 g/kg
BW. Rats from the 40th generation consumed an average of 7.0 g
alcohol/kg BW/day ± 0.63 (S.E.) with a range of 2.6-13.7 g/kg
BW. Rats from the 41st generation consumed an average of 6.9 g
alcohol/kg BW/day ± 0.31 (S.E.) with a range of 2.0-11.2 g/kg
BW. All rats were housed individually in a temperature and humidity
controlled room with a 12 hr light/dark cycle (lights on at 0700 hrs)
with food and water available ad libitum except as dictated
by the experimental design.
Drugs.
Naltrindole hydrochloride or NTI
(17-cyclopropylmethyl-6,7-dehydro-4,5a-epoxy-3,
14-dihydroxy-6,7,2',3'-indolomorphinan) is a nonpeptide, highly potent
antagonist with high affinity and selectivity for delta
opioid receptors (Portoghese et al., 1988; 1990). Both
beta-endorphin and the enkephalins, but not dynorphin, evidence a high
affinity for the delta opioid receptor subtype (Kosterlitz
and Paterson, 1985; Raynor et al., 1994). NTI has a
prolonged duration of action owing to its resistance to enzymatic degradation by peptidases (Portoghese et al., 1988; 1990;
Sofuoglu et al., 1991).
NTI (MW = 450.6) was dissolved in saline containing enough NaOH to
solubilize the compound and the pH of the solution was adjusted to
6.9-7.2 by titrating with 0.1 N HCl before i.p. injection. The
solution containing NTI was heated in order to get the drug into
solution and to keep it in solution. NTI was administered in doses of
2.5 mg/2.5 ml NaCl/kg BW, 5.0 mg/5.0 ml NaCl/kg BW, 10 mg/5.0 ml
NaCl/kg BW or 20.0 mg/10 ml NaCl/kg BW.
Experiment 1: Establishing a dose-response curve for aversion to
alcohol.
Prior to initiation of experiment 1, a concentration of a
banana-flavored solution that was both detectable and neutral was identified. Two separate groups of alcohol-naive, female P rats from
the 39th generation of selection for alcohol preference served as
subjects (n = 9/grp). Rats were given a 24-hr
free-choice between water and one of two different concentrations of a
banana-flavored solution (0.5% or 0.025%) for 5 to 7 days. Of the 9 rats given access to water and the 0.5% banana-flavored solution, 7 rats drank more water than banana solution, 1 rat drank more
banana-flavored solution than water and 1 rat consumed equal amounts of
water and banana. Hence, the 0.5% banana-flavored solution was not
neutral. In contrast, of the 9 rats given access to water and the
0.025% banana-flavored solution, 5 rats drank more water than banana, 3 rats drank more banana than water and 1 rat consumed equal amounts of
water and banana. Consequently, the 0.025% concentration of banana-flavored solution was considered "neutral" in terms of overall taste preference. In order to ensure that the banana-flavored solution was detectable to every rat, the position of the bottle containing banana-flavored solution was alternated daily which forced
the rats to switch drinking sides daily in order to "track" either
banana or water. All rats tracked one solution or the other every day
except the 1 rat which consumed equal volumes of water and the
banana-flavored solution. Therefore, the 0.025% concentration of
banana solution was considered detectable and neutral and was used as
the CS in the CTA conditioning trials.
A dose-response curve for alcohol aversion in P rats was established
using a CTA paradigm. Establishing a CTA to alcohol involves presenting
conditioning trials over several days, each of which consists of
pairing consumption of a neutral yet distinctive flavor, such as
banana, with administration of alcohol or vehicle. During postconditioning rats are given a free-choice between the flavor and
water and aversion to alcohol is inferred from a decrease in intake of
the flavored solution.
Thirty-six female P rats from the 39th generation of selection for
alcohol preference served as subjects. All rats weighed 250 to 300 g at the start of the experiment. During preconditioning, access to
water was limited to 20 min/day for 8 days until water intake
stabilized. Water was presented in one of two 100 ml calibrated drinking tubes with the other tube empty. The positions of the two
tubes were rotated daily to control for the effects of a potential side
preference. On the last day of preconditioning, rats were counterbalanced on water intake over the 8 days of preconditioning, and
were assigned to one of four groups (0.5, 1.0, 1.5 g/kg EtOH or saline
control; n = 9/group) in a manner which ensured that the groups did not differ in average daily water intake.
During conditioning, access to banana-flavored solution was substituted
for access to water during the 20 min/day fluid access period. Five
conditioning trials were given, one every other day, during which
access to the banana solution (McCormick's; 0.025% v/v) was paired
with an intraperitoneal (i.p.) injection of one of 3 doses of alcohol
(0.5 g/10 ml/kg, 1.0 g/10 ml/kg or 1.5 g/15 ml/kg BW) or an equal
volume of saline (control). Alcohol was administered 30 min after
termination of access to the banana solution. The volume of alcohol
injected i.p. was altered as a function of dose so that the
concentration injected would not exceed 13% (v/v) in order to minimize
tissue irritation at the site of injection (Barry and Walgren, 1968;
Linakis and Cunningham, 1979). On conditioning days, the banana
solution was presented in one calibrated drinking tube, with the other
tube empty, and the two tubes were rotated on each conditioning day. On
intervening days, water alone was presented for 20 min per day in one
of two calibrated drinking tubes, with the other tube empty, in the
absence of injections. In the CTA paradigm, development of aversion to alcohol is reflected by a decrease in consumption of the flavored solution which is paired with alcohol. Given that aversion can develop
during conditioning, access to water on intervening days is necessary
to prevent dehydration. Consumption of the banana-flavored solution and
water were recorded on conditioning and intervening days respectively.
On conditioning days, all rats were weighed and injected with saline
(0.8 mls/100 g BW) before onset of the conditioning trial in order to
mimic the conditions required for drug administration in experiment 2.
During postconditioning rats were presented with a choice between water
in one drinking tube and banana solution (0.025%) in the other tube
for 20 min/day without injections. One postconditioning trial was given
per day for 10 consecutive days and the positions of the tubes
containing water and banana solution were rotated daily. Intake of
water and banana solution was recorded daily, during both conditioning
and postconditioning, and rats were weighed daily.
At the conclusion of experiment 1, all rats were maintained in their
home cages and were provided with food and water ad libitum for 6 weeks. A subset of 24 rats were then randomly selected to serve
as subjects in an additional study which was designed to determine the
BAC that was produced by each of the three doses of alcohol used in
experiment 1. In order to reestablish the fluid access conditions used
in experiment 1, rats were deprived of food and water for 15 hr and
were then given access to water for 20 min immediately before receiving
an injection of alcohol in one of the three doses used in experiment 1 (0.5, 1.0 and 1.5 g//kg, i.p.).
Blood samples (50-100 µl) were collected from the tail of each rat
into heparinized capillary tubes at 15, 30, 60 and 180 min after
alcohol injection. Blood was centrifuged and the plasma stored at
-20°C until assayed for plasma alcohol concentration by high
performance liquid chromatography. Plasma alcohol content was analyzed
by direct injection of 1.0 µl of plasma into a Hewlett-Packard 5730A
gas chromatograph equipped with a flame ionization detector and a 3380A
integrator. The glass columns were packed with Poropack Q (80/100 mesh)
and the oven temperature was 150°C. Isopropanol was used as the
internal standard.
Experiment 2: Effect of NTI on conditioned taste aversion to
alcohol.
Experiment 2 was designed to determine whether blocking
delta opioid receptors would serve to make alcohol more
aversive and hence shift the alcohol aversion threshold to the left.
Seventy-two alcohol-naive female P rats from the 40-41st generations
of selection for alcohol preference served as subjects. All rats
weighed 250-300 g at the start of the experiment. The dose of alcohol
that was found to be nonaversive in experiment 1, 0.5 g/kg, was paired with consumption of the banana-flavored solution (0.025%) in separate groups of P rats. Rats were pretreated with either saline or one of 4 doses of the delta opioid receptor antagonist, NTI. As in experiment 1, aversion to alcohol was inferred from a decrease in
intake of the banana-flavored solution.
During preconditioning, the same paradigm was used as in the
preconditioning phase of experiment 1. On the last day of
preconditioning, the rats were assigned to groups and were pretreated
with either saline or one of four doses of NTI (2.5, 5.0, 10.0 and 20.0 mg/kg) before injection of the nonaversive dose of alcohol (0.05 g/kg BW) or an equal amount of saline. Rats were counterbalanced and assigned to groups (n = 8/group) based on their average
daily water consumption (20 min/day) during the 8 days of
preconditioning.
During conditioning, 5 conditioning trials were given, one every other
day. On conditioning days, all rats were given 20 min access to
banana-flavored solution (McCormick's, 0.025%) in one drinking tube,
with the other tube empty. Ten minutes after termination of access to
the banana-flavored solution, rats received an i.p. injection of one of
four doses of NTI (2.5 mg/2.5 ml/kg, 5.0 mg/5.0 ml/kg, 10.0 mg/5 ml/kg
or 20.0 mg/10 ml/kg BW) or an equal volume of saline, which was
followed, 20 min later, by an injection of the nonaversive dose of
alcohol (0.5 g/10 ml/kg BW i.p.) or an equal volume of saline. We have
previously reported that NTI is physiologically active within 20 min
after i.p. administration in rats (Krishnan-Sarin et al.,
1995a). As in experiment 1, water was presented alone for 20 min per
day in one of two calibrated drinking tubes, with the other tube empty,
on intervening days in the absence of injections in order to prevent
potential dehydration. The position of the tubes containing banana
solution and water were rotated daily during conditioning and fluid
intake and body weight were recorded daily.
During postconditioning, the rats were presented with a choice between
water and the 0.025% banana solution for 20 min a day without
injections. One postconditioning trial was given per day for 10 consecutive days, and the positions of the tubes containing water and
banana solution were rotated daily. Intake of water and banana solution
was recorded daily during both conditioning and postconditioning and
rats were weighed daily.
At the conclusion of experiment 2, all rats were maintained in their
home cages and were provided with food and water ad libitum for 4 weeks. A subset of 12 rats were randomly selected to serve as
subjects in an additional study which was designed to determine the
effect of NTI on BAC. In order to reestablish the fluid access conditions used in experiment 2, rats were deprived of food and water
for 15 hr and were then given access to water for 20 min followed, 10 min later, by an injection of NTI (10 mg/5 ml/kg BW i.p.) or an equal
volume of saline which, in turn, was followed 20 min later by an
injection of alcohol (0.05/10 ml/kg BW, i.p.). Blood samples were
collected and alcohol concentration in plasma was analyzed as described
in experiment 1.
Data analysis.
In order to determine which doses of alcohol
were aversive in experiment 1, comparisons were made between alcohol vs
saline for each of 3 doses of alcohol (0.5, 1.0 and 1.5 g/kg BW) using separate two-way repeated measures analysis of variance (ANOVA) in the
conditioning and postconditioning phases of the experiment. Specifically, three separate two-way, repeated measures ANOVAs, one for
each dose of alcohol vs saline, were used to analyze the data from the
conditioning phase of the study. An additional three separate two-way,
repeated measures ANOVAs, one for each dose of alcohol vs saline, were
used to anlayze the data from the postconditioning phase of the study.
In order to determine whether combining different doses of NTI with a
nonaversive dose of alcohol makes alcohol aversive, the following
comparisons were made using separate two-way repeated measures ANOVAs
for conditioning: alcohol (0.5 g/kg BW) vs. NTI for each of 4 doses of
NTI (2.5, 5.0, 10.0 and 20.0 mg/kg BW); alcohol (0.5 g/kg BW) vs. NTI + alcohol for each of 4 doses of NTI (2.5, 5.0, 10.0 and 20.0 mg/kg BW);
NTI vs. NTI + alcohol for each of 4 doses of NTI (2.5, 5.0, 10.0 and
20.0 mg/kg BW). The same analyses were used to analyze the data from
the postconditioning phase.
In order to determine the blood alcohol concentration (BAC) produced by
various doses of alcohol, BACs were analyzed at each time point using a
one-way ANOVA. Post-hoc comparisons of the differences between the
means at a given time point, for each of the alcohol doses, were made
using Bonferroni's t tests for all pairwise multiple
comparisons. Blood alcohol elimination rates were calculated from the
slope of the falling portion of the BAC curve using linear regression
analyses for each rat and comparisons of alcohol elimination rate
between groups were made using an unpaired student's t
test. Area under the BAC curve was calculated using AUC version 1.1, and comparisons of area under the curve between groups were made using
unpaired students t tests.
| |
Results |
Experiment 1: Establishing a dose-response curve for aversion to
alcohol.
Figure 1 illustrates intake
of the banana solution during conditioning and postconditioning in
experiment 1. During conditioning, there were no differences in intake
of the banana solution between the groups receiving alcohol in doses of
0.5, 1.0 or 1.5 g/kg BW compared to those receiving saline. Alcohol in
a dose of 0.5g/kg was not aversive to P rats as evidence by the fact
this dose, when compared with saline, did not suppress intake of the
banana solution during postconditioning. In contrast, alcohol in a dose of 1.0 g./kg BW was aversive as evidence by the fact that this dose, when compared with saline, significantly suppressed intake of the
banana solution during postconditioning, F(1,16)=39.184, P < .001. The highest dose of alcohol tested, 1.5 g/kg BW, produced the
strongest aversion in comparison with saline, F(1,16)=166.87, P < .001.
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 1.
Intake of banana-flavored solution in P rats
receiving an i.p. injection of saline or alcohol in a dose of 0.5, 1.0 or 1.5 g/kg paired with banana solution during conditioning (panel 1).
Intake of banana-flavored solution in P rats when banana solution is
presented concurrently with water during postconditioning in the
absence of drug treatment (panel 2). Each point represents the
mean ± S.E.M. and N indicates the number of rats per group.
|
|
Total fluid intake (banana-flavored solution + water) did not differ
when the various treatment groups (0.5, 1.0 and 1.5 g/kg BW) were
compared to each other or to the saline control group.
BAC.
Analysis of the blood alcohol dose-response curves showed
that BAC was dose-dependent and that peak blood alcohol levels were achieved at 30 min after administration of each of the three doses of
alcohol (fig. 2). There was a significant
difference between the blood alcohol concentrations produced by each of
the three alcohol doses (0.05, 1.0 and 1.5 g/15 ml/kg BW i.p.) at all
time points: 15 min, F(2,21)=18.85, P < .001; 30 min,
F(2,21)=58.46, P < .001; 60 min, F(2,21)=67.08, P < .001;
and 180 min, F(2,21)=92.22, P < .001. Post-hoc comparisons
demonstrated that the BAC was significantly lower after administration
of the 0.5 g/kg BW compared with the 1.0 and 1.5 g/kg doses at all time
points. BAC levels were lower in the 1.0 g/kg group compared with the
1.5 g/kg group at 30, 60 and 180 min. BAC levels returned to base line
by 180 min after administration of alcohol in a dose of 0.5 g/kg but
BAC was still elevated above base line at 180 min after administration
of the other two alcohol doses (1.0 and 1.5 g/kg BW). Area under the BAC curve after administration of the 0.5 g/kg dose of alcohol was
significantly lower than that produced by 1.0 g/kg alcohol, t(14)=
16.58, P < .001 or 1.5 g/kg alcohol, t(14)= 11.70, P < .001. Area under the curve differed between the two highest alcohol
doses (1.0 and 1.5 g/kg) as well t(14) = 5.15, P < .001. Alcohol elimination rate, based on the slope of the linear regression analysis, did not differ when the groups receiving alcohol (0.5, 1.0 and 1.5 g/kg BW) were compared.
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 2.
Blood alcohol concentrations (BACs) in P rats after
an i.p. injection of alcohol in a dose of 0.5, 1.0 or 1.5 g
alcohol/kg BW. Each point represents the mean ± S.E.M. and N
indicates the number of rats per group.
|
|
Experiment 2: Effect of NTI on conditioned taste aversion to
alcohol.
As illustrated in figure 3,
NTI in a dose of 2.5 mg/kg, when administered either alone or in
combination with a nonaversive dose of alcohol, did not alter intake of
the banana solution or total fluid intake during conditioning or
postconditioning when compared with intake in the group receiving
alcohol alone. Hence, this dose of NTI (2.5 mg/kg) is not aversive when
administered alone or when combined with a nonaversive dose of alcohol.
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 3.
Intake of banana-flavored solution in P rats
receiving alcohol (0.5g/kg, i.p.) paired with banana solution, or NTI
(2.5 mg/kg, i.p.) + saline paired with banana solution, or NTI (2.5 mg/kg, i.p.) + alcohol (0.5g/kg, i.p.) paired with banana solution,
during conditioning (panel 1). Intake of banana-flavored solution in P
rats when banana solution is presented concurrently with water during
postconditioning in the absence of drug treatment (panel 2). Each point
represents the mean ± S.E.M. and N indicates the number of rats
per group.
|
|
As illustrated in figure 4, NTI in a dose
of 5.0 mg/kg BW, was not aversive when administered alone as evidenced
by the fact that this dose did not alter intake of the banana solution
when compared with intake in the group receiving a nonaversive dose of
alcohol alone during either conditioning [F (1,13)=3.60, P = .08] or postconditioning [F(1,13)=.015, P = .91]. In contrast, combining the nonaversive dose of NTI (5.0 mg/kg) with the nonaversive dose of alcohol (0.5 g/kg) resulted in the formation of a significant aversion to alcohol as evidenced by a reduction in intake of the banana-flavored solution during postconditioning when the group receiving NTI + alcohol was compared with the group receiving alcohol
alone, F(1,13)=14.56, P < .002 or the group receiving NTI alone
F(1,14)=5.64, P < .032 (fig. 4). Aversion to alcohol in the group
receiving NTI and alcohol was first evident during the conditioning
phase when intake in this group was compared with intake in the group
receiving NTI without alcohol, F(1,14)=7.90, P < .014. Total
fluid intake (banana-flavored solution + water) did not differ when the
treatment group (NTI + alcohol) was compared to the NTI alone or
alcohol alone control groups.
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 4.
Intake of banana-flavored solution in P rats
receiving alcohol (0.5g/kg, i.p.) paired with banana solution, or NTI
(5.0 mg/kg, i.p.) + saline paired with banana solution, or NTI (5.0 mg/kg, i.p.) + alcohol (0.5g/kg, i.p.) paired with banana solution,
during conditioning (panel 1). Intake of banana-flavored solution in P
rats when banana solution is presented concurrently with water during
postconditioning in the absence of drug treatment (panel 2). Each point
represents the mean ± S.E.M. and N indicates the number of rats
per group.
|
|
As illustrated in figure 5, NTI in a dose
of 10.0 mg/kg BW, was not aversive when administered alone as evidenced
by the fact that this dose did not alter intake of the banana solution
when compared with intake in the group receiving alcohol alone during either conditioning [F (1,13)=.32, P = .58], or
postconditioning, [F(1,13)=1.05, P = .32]. However, as was seen
with the 5.0 mg/kg dose of NTI, when a dose of 10.0 mg/kg of NTI was
combined with the nonaversive dose of alcohol (0.5 g/kg), a significant
aversion to alcohol was apparent as evidenced by a reduction in intake of the banana-flavored solution during postconditioning when the group
receiving NTI and alcohol was compared with the group receiving alcohol
alone, F(1,13)=6.06, P < .029 (fig. 5). Total fluid intake (banana-flavored solution + water) did not differ when the treatment group (NTI + alcohol) was compared to the NTI alone or alcohol alone
control groups.
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 5.
Intake of banana-flavored solution in P rats
receiving alcohol (0.5g/kg, i.p.) paired with banana solution, or NTI
(10.0 mg/kg, i.p.) + saline paired with banana solution, or NTI (10.0 mg/kg, i.p.) + alcohol (0.5g/kg, i.p.) paired with banana solution,
during conditioning (panel 1). Intake of banana-flavored solution in P
rats when banana solution is presented concurrently with water during
postconditioning in the absence of drug treatment (panel 2). Each point
represents the mean ± S.E.M. and N indicates the number of rats
per group.
|
|
As illustrated in figure 6, the highest
dose of NTI tested, 20 mg/kg BW, was aversive when administered either
alone, F(1,13)=7.29, P < .018 or in combination with the
nonaversive dose of alcohol (0.5 mg/kg), F(1,11)=8.23, P < .015 during postconditioning. Total fluid intake (banana-flavored solution + water) did not differ when the treatment group (NTI + alcohol) was
compared to the NTI alone or alcohol alone control groups.
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 6.
Intake of banana-flavored solution in P rats
receiving alcohol (0.5g/kg, i.p.) paired with banana solution, or NTI
(20.0 mg/kg, i.p.) + saline paired with banana solution, or NTI (20.0 mg/kg, i.p.) + alcohol (0.5g/kg, i.p.) paired with banana solution,
during conditioning (panel 1). Intake of banana-flavored solution in P
rats when banana solution is presented concurrently with water during
postconditioning in the absence of drug treatment (panel 2). Each point
represents the mean ± S.E.M. and N indicates the number of rats
per group.
|
|
Effect of NTI on BAC.
As illustrated in figure
7, combining NTI (10.0 mg/kg) with
alcohol (0.5g/kg BW i.p.) did not alter peak BAC, area under the BAC
curve or rate of alcohol elimination when compared with a group
receiving alcohol alone.
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 7.
BAC in P rats pretreated with NTI (10.0 mg/kg BW,
i.p.), or an equal volume of saline, before i.p. injection of alcohol
in a dose of 0.5 g/kg BW. Each point represents the mean ± S.E.M.
and N indicates the number of rats per group.
|
|
| |
Discussion |
We have previously shown that rats selectively bred for
alcohol-preference (P line) show an increased responsiveness of the opioid system to alcohol in comparison to rats bred for
alcohol-nonpreference (Li et al., 1998; Froehlich and Wand,
1996; Froehlich, 1997). Other investigators have also noted a positive
association between alcohol preference and increased sensitivity of the
opioid system to alcohol in mice (De Waele et al., 1992). In
addition, increased responsiveness of the opioid system to alcohol has
been reported in humans at high risk for the future development of
alcoholism compared with those at low risk (Gianoulakis et
al., 1996). We postulate that alcohol-induced activation of the
opioid system serves to attenuate the aversive properties of alcohol
ingestion and increase the probability of subsequent drinking in rats
of the P line. If this view is correct, blocking the action of opioid peptides, via administration of an opioid receptor
antagonist, would be expected to lower the alcohol aversion threshold
in P rats. The present study examined the effects of the selective delta opioid receptor antagonist, NTI, on alcohol aversion
threshold. A dose-response curve for alcohol aversion was established
in female P rats in experiment 1. Alcohol, in doses of 1.0 and 1.5 g/kg
BW, was aversive to P rats while a dose of 0.5 g/kg BW was not. These
results confirm prior reports that alcohol in a dose of 0.5 g/kg BW is
not aversive to P rats while higher doses, in excess of 1.0 g/kg BW,
are aversive (Froehlich et al., 1988; Stewart et
al., 1991). A nonaversive dose of alcohol (0.5 g/kg BW) is the
optimal dose to use in combination with an opioid receptor antagonist
in order to investigate whether blocking the action of opioid peptides
makes alcohol more aversive.
In experiment 2, the effect of the selective delta opioid
receptor antagonist NTI on alcohol aversion was examined. The lowest dose of NTI (2.5 mg/kg BW) was not aversive when administered alone or
when administered in combination with the nonaversive dose of alcohol
(0.5 g/kg BW). In contrast, higher doses of NTI (5.0 or 10.0 mg/kg BW)
were not aversive when administered alone, but when combined with
alcohol, both doses converted a nonaversive dose of alcohol into an
aversive dose in P rats. The highest dose of NTI tested (20.0 mg/kg)
was aversive when administered either alone or in combination with
alcohol. It appears that pretreating P rats with a nonaversive dose of
NTI, before administration of a nonaversive dose of alcohol, shifted
the alcohol aversion threshold to the left. These results suggest that
the endogenous ligands with high affinity for the delta
opioid receptor subtype, namely beta-endorphin and the enkephalins,
normally play a role in regulating alcohol aversion threshold. It
should be noted that the doses of NTI used in the present study are
similar to those that we have previously found to be effective in
decreasing alcohol drinking in rats of the P line (Krishnan-Sarin
et al., 1995). It should also be noted that the NTI-induced
alteration of aversion threshold seen in P rats in the present study
was not due to an NTI-induced change in alcohol pharmacokinetics since
blood alcohol concentration at peak BAC, rate of alcohol disappearance
and area under the BAC curve did not differ in rats pretreated with NTI
vs those pretreated with saline. It is possible that the decrease in
intake of the banana-flavored solution seen in rats receiving
nonaversive doses of NTI and alcohol in combination may have resulted
from a summation of the effects of alcohol and NTI since neither of the
drugs were aversive when administered alone in the doses used in the
combination, yet both of the drugs were aversive when administered alone at higher doses. However, summation of the effects of alcohol and
NTI is not likely in the present study. Although the alcohol dose used
in the combination (0.5 g/kg) was probably just below threshold since
this dose produced no aversion but doubling the dose to 1.0 g/kg
produced aversion, the 5.0 mg/kg dose of NTI used in the combination
was not near threshold since doubling the dose to 10.0 mg/kg still did
not produce aversion when administered alone. Hence the decrease in
intake of the banana-flavored solution seen in rats treated with NTI
and alcohol in combination is probably not due to summation of
subthreshold effects of alcohol and NTI.
A genetic difference in alcohol aversion, which may be mediated
via the endogenous opioid system, has also been noted in
mice by Cunningham and colleagues (Broadbent et al., 1996).
Neither alcohol alone, nor naloxone alone, produced a CTA to alcohol in the alcohol-preferring C57BL/6J mice but pretreating the C57 mice with
naloxone, before alcohol, resulted in a CTA. In contrast, a comparable
dose of alcohol alone was sufficient to produce a CTA in
alcohol-avoiding DBA/2J mice (Broadbent et al., 1996). These
results suggest that alcohol is less aversive for the
alcohol-preferring C57 mice compared to the alcohol-avoiding DBA mice
and that this strain difference in aversion to alcohol may be due, in
part, to greater alcohol-induced activation of the opioid system in the
alcohol-preferring C57 mice.
The results of the present study suggest that alcohol-induced
activation of the endorphin and enkephalin systems may serve to reduce
aversion to alcohol and hence increase the probability of subsequent
alcohol drinking. The importance of these finding lies in the
elucidation of the biological basis of motivation for alcohol drinking
in humans. Several clinical studies have demonstrated that the
nonselective opioid receptor antagonist, naltrexone, decreases alcohol
drinking in detoxified outpatient alcoholics (O'Malley et
al., 1992; Volpicelli et al., 1992). Specifically, naltrexone decreases mean number of drinking days per week, frequency of relapse, and desire to drink or subjective craving for alcohol (O'Malley et al., 1992; Volpicelli et al., 1992;
O'Malley et al., 1996). The results of these clinical
trials led to FDA approval in 1994 of naltrexone as a
pharmacotherapeutic agent for the treatment of alcohol dependence. One
of the most interesting effects of naltrexone was found in subjects who
drank alcohol or "slipped" while taking naltrexone. Naltrexone was
found to be effective in decreasing subsequent drinking once drinking
had occurred. Relapse, or significant resumption of drinking was seen
in 54% of the subjects receiving placebo but in only 23% of the
subjects receiving naltrexone (Volpicelli et al., 1992).
Similar results have been reported by O'Malley (O'Malley et
al., 1996). The effect of naltrexone in these studies has been
attributed, in part, to a naltrexone-induced decrease in the
reinforcing or euphoriant effects of alcohol (Volpicelli et
al., 1995; O'Malley et al., 1996). However, the
results of the present study suggest that opioid antagonists, such as
naltrexone, may also decrease the probability of subsequent alcohol
drinking by blocking the action of opioid peptides which may normally
serve to diminish the aversive effects of alcohol and protect against
hangover. The results of a recent study, conducted in social drinkers,
are of interest with regard to this hypothesis. Swift and colleagues
investigated the effects of naltrexone vs placebo on several subjective
and objective measures of alcohol intoxication in social drinkers (Swift et al., 1994). Subjects were given naltrexone (50 mg
p.o.) or placebo on two different occasions, each time followed by a standard, intoxicating dose of alcohol. As a control for naltrexone effects, additional subjects received naltrexone or placebo followed by
a nonintoxicating "placebo" dose of alcohol. Subjects receiving naltrexone followed by alcohol not only found that the positively reinforcing stimulant effects of alcohol were reduced, but also that
the dysphoric effects were augmented after alcohol consumption. No
dysphoria was reported in response to naltrexone alone or alcohol alone. These results again suggest that alcohol-induced release of
endogenous opioid peptides may normally serve to attenuate the aversive
properties of alcohol and hence, when the action of endogenous opioids
are blocked, via co-administration of naltrexone and
alcohol, the aftereffects of alcohol may be more aversive. Consistent
with this interpretation, subjects in the Swift study reported that
they liked the placebo-alcohol session more than the naltrexone-alcohol
session. It should be noted that naltrexone did not alter alcohol
pharmacokinetics or the magnitude of alcohol-induced performance
deficits in this study (Swift, 1994).
The results of the present study indicate that alcohol-induced
activation of the opioid system may make the postingestional effects of
alcohol less aversive in rats selectively bred for alcohol preference.
Genetic differences in sensitivity of the opioid system to alcohol may
be part of a biological mechanism underlying individual differences in
aversion to the postingestional effects of alcohol commonly referred to
as "hangover." It is interesting to speculate that increased
sensitivity of the endorphin and enkephalin systems to alcohol may
represent a potential biological "marker" that could be used to
identify individuals at risk for the future development of alcoholism.
We thank Dr. Ting-Kai Li for supplying us with selectively bred
rats from the alcohol-preferring (P) line.
CTA, conditioned taste aversion;
NTI, naltrindole;
BAC, blood alcohol concentration;
P rats, alcohol-preferring rats;
ANOVA, analysis of variance;
CS, conditioned
stimulus;
i.p., intraperitoneal;
i.g., intragastric.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Top
Abstract
Introduction
