Persistent High Alcohol Consumption in Alcohol-Preferring (P) Rats Results from a Lack of Normal
Aversion to Alcohol
Amir H. Rezvani*, Hannah Sexton and Edward D. Levin
Department of Psychiatry and Behavioral Sciences, Duke University School of Medicine, Durham, NC, USA
*Corresponding author: Department of Psychiatry and Behavioral Sciences, Duke University Medical Center, Box 3412, Durham NC 27710, USA.
Tel.: +1-919-668-1880; Fax: +1-919-681-3416; E-mail: Azadi@duke.edu
(Received 3 December 2009; in revised form 2 February 2010; accepted 15 March 2010)
ABSTRACT — Aims: In this study, we tested the impact of pretreatment with alcohol on subsequent alcohol drinking in outbred
Sprague–Dawley and selectively bred alcohol-preferring (P) rats. Methods: As a pretreatment, male Sprague–Dawley and P rats were
given a passive oral administration of either alcohol (1.0 g/kg) or tap water. Then, they were given free choice of drinking alcohol (5%
v/v) or water in their home cages, which was measured over 4 weeks. Results: Without alcohol pretreatment, there was no significant
strain difference in alcohol preference; both strains preferred 5% (v/v) alcohol solution. The strain difference was only apparent in the
groups given alcohol pretreatment. This arose from the fact that alcohol pretreatment significantly reduced alcohol preference in the
Sprague–Dawley rats to a level well below 50%, while it did not alter drinking behavior in P rats. The same effects were seen with total
alcohol consumption (g/kg/day). These effects persisted throughout the 4 weeks of the study. Conclusions: The principal difference
between the Sprague–Dawley and P rats was that the P rats did not show the normal aversion to alcohol after forced exposure to alcohol
that the Sprague–Dawley rats showed. One of the potential contributors to high alcohol intake and preference in P rats may be lack of
sensitivity to aversive effects of alcohol.
Usually the first experience with drugs including alcohol is an
important determinant factor in subsequent use. Experiencing
aversion to the initial drug episode may discourage the use of
a drug while experiencing reward and euphoria may encour-
age it. The first experience, reward or aversion, depends on
several factors including the environment setting, the onset
of the effect, prediction of the effects, psychological condi-
tions of the subject at the time of the use and, very
importantly, the genetic makeup of the subject. It is estimated
that about 60% of alcoholism risk is genetic and about 40%
stems from environmental factors. The genetic component of
alcoholism has been modeled in rats by selectively breeding
rats with high levels of alcohol preference (Lumeng et al.,
1982a, b; Li et al., 1993). The rat models can be useful for
determining the neurobehavioral bases for alcoholism to help
direct development of new efficacious approaches to thera-
peutic treatment. These genetic models can help determine
which aspects of the biobehavioral effects of alcohol are
key in determining heavy consumption and alcoholism.
In humans, it has been conclusively demonstrated that a
low level of initial behavioral or subjective response to alco-
hol ingestion predicts a higher rate of alcohol use disorder
onset over time (Schuckit and Smith, 1996; Schuckit and
Gold, 1998; Schuckit and Smith, 2000, 2001; Schuckit et
al., 2005, 2006, 2009; Schuckit 2009; Trim et al., 2009). In
all of these studies, a lower level of response early in life pre-
dicted later heavy drinking and alcohol use disorder. These
findings suggest that low level of response to alcohol is a
unique risk factor for alcohol use disorders across adulthood.
Similar findings have been reported in animal models of
drinking behaviors (Froehlich et al., 1988; Kurtz et al.,
1996; Stewart et al., 1992, 1996). It has also been shown that
alcohol-naïve, alcohol-preferring (P) rats compared with se-
lectively bred alcohol non-preferring (NP) rats are less
sensitive to the sedative/hypnotic effect of alcohol. Compared
with NP rats, P rats took longer to lose righting reflex, had a
shorter recovery time and recovered at a higher blood alcohol
concentration following a single sedative/hypnotic dose of al-
cohol given systemically (Kurtz et al., 1996). Using a
conditioned taste aversion paradigm, Froehlich et al. (1988)
demonstrated that repeated pairing of saccharin and alcohol
produced stronger and prolonged taste aversion to saccharin
in the NP rats compared with the P rats. These findings sug-
gest that genetic factors are important in determining whether
a drug will be perceived as aversive or rewarding, and conse-
quently, among other factors, genetic factors will determine
whether an organism will avoid a particular drug.
A number of studies have shown that both selectively bred
P rats and non-selectively bred (i.e. Fawn-Hooded) rats P rats
differ from Wistar and Sprague–Dawley rats substantially in
the amount of alcohol consumed and preference for alcohol
(Eriksson, 1968; Lumeng et al., 1982a, b; Mardones and Se-
govia-Riquelme, 1983; Rezvani et al., 1990; Li et al., 1993;
Colombo, 1997), suggesting the influence of genetics in
drinking behaviors. However, it is not clear if the main driving
force underlying the increased alcohol consumption in P rats
is that they find alcohol more rewarding or less aversive than
other rats. The present study was designed to determine and
compare the level of aversion to alcohol in P and Sprague–
Dawley rats. It is hypothesized that P rats drink more alcohol
than Sprague–Dawley rats because alcohol is less aversive in
P rats than Sprague–Dawley rats.
The animals used in these studies were adult male Sprague–
Dawley rats (Taconic Farms, Germantown, NY, USA) and rats
preferring P line at Indiana University, Indianapolis. Rats were
housed individually in a standard laboratory maintained at 21 ±
© The Author 2010. Published by Oxford University Press on behalf of the Medical Council on Alcohol. All rights reserved
Alcohol & Alcoholism Vol. 45, No. 3, pp. 219–222, 2010
Advance Access Publication 30 March 2010
1°C and relative humidity at 50 ± 10% and reversed light cycle
(lights off: 0700–1900). The P and Sprague–Dawley rats
weighed an average of 373 ± 4.3 and 303 ± 4.3 g, respectively,
at the beginning of the experiment. The treatment and care of
the animals was carried out under an approved protocol of the
Animal Care and Use Committee of Duke University in an As-
sociation for Assessment and Accreditation of Laboratory
Animal Care-approved facility.
At about 80 days of age, the rats were handled briefly for
10 min/day for several days. Then, rats were given either
one single dose of 1 g/kg alcohol (16% v/v) or an equal vol-
ume of tap water by gavage once and only at the beginning of
the experiment. To avoid possible taste aversion, all solutions
were directly delivered into the stomach and not into the
mouth. Fifteen minutes later, rats were placed in specialized
polycarbonate cages that were fitted with two 100-mL gradu-
ated Richter drinking tubes for the recording of water and
alcohol (5% v/v). The reason for choosing 5% alcohol rather
than a higher concentration was that alcohol at this concentra-
tion is more palatable for outbred strains of rats such as
Sprague–Dawley. Furthermore, with this concentration of al-
cohol, there is less variability in drinking in outbred strains.
Rats had a choice between water and alcohol for the remain-
der of the study. Rats had continual access to water and
alcohol for four consecutive weeks, and their water and
alcohol intake levels were measured during the day at approxi-
mately the same time. Procedures were similar to those
routinely used in our laboratory (Rezvani and Grady, 1994;
Rezvani et al., 1995; Overstreet et al., 1996; Rezvani et al.,
1997, 1999, 2000, 2007, 2009).
Preparation of alcohol solutions
A solution of 5% (v/v) alcohol was prepared twice weekly
from a solution of 100% ethanol mixed with tap water.
Statistical analysis of data
Alcohol intake was calculated as grams per kilogram from a
volume of 5% (v/v) alcohol consumed. Alcohol preference
was calculated as a percentage of alcohol consumed over total
fluid intake (alcohol + water). Means were calculated and sub-
jected to ANOVAwith strain and pre-testing alcohol treatment
as between-subjects factors and weekly averages of percent
alcohol vs water preference and grams per kilogram per day
of alcohol intake. Significant interaction effects were followed
up by tests of the simple main effects of each strain at each
pre-testing alcohol treatment condition to determine which in-
dividual strain and treatment groups differed from each other.
A P-value of 0.05 (two-tailed) was used as the threshold for
For amount of alcohol consumed (g/kg/day), there was not a
significant main effect of strain (P = 0.098). However, there
was clearly a significant interaction of strain × alcohol pretreat-
ment (F(1,27) = 7.27, P < 0.025). As displayed in Fig. 1A
(mean alcohol consumption over 4 weeks), the tests of the sim-
ple main effects showed significantly (P < 0.005) greater
alcohol consumption in P rats than Sprague–Dawley rats when
there was alcohol pretreatment, but not when the pretreatment
was only water. This did not result from the induction of alco-
hol drinking by alcohol pretreatment in P rats. They did not
significantly change their consumption with alcohol pretreat-
ment. Rather, it was the Sprague–Dawley rats that showed a
significant (P < 0.005) decrease in alcohol consumption after
alcohol pretreatment by nearly 50%. Fig. 2A shows the de-
tailed week-by-week levels of alcohol consumption. There
was a significant main effect of week (F(3,81) = 13.30, P <
0.0005) with an overall increase in alcohol consumption over
weeks. There was no significant interaction of weeks with ei-
ther strain or alcohol pretreatment. The lower amount of
alcohol consumption in Sprague–Dawley rats with pre-expo-
sure vs Sprague–Dawley rats without alcohol pre-exposure as
well as relative to P rats pre-exposed to alcohol was maintained
throughout the 4-week study. Sprague–Dawley and P rats
without alcohol pre-exposure did not differ in alcohol consum-
mation at any point in the 4-week study.
The percent alcohol preference measure showed similar re-
sults. With percent preference, there was a significant main
effect of strain (F(1,27) = 14.80, P < 0.001) with the P rats
having a higher preference for alcohol than the Sprague–
Dawley rats (81.1 ± 3.6% vs 52.0 ± 7.3%); however, this re-
sulted completely from the lower alcohol preference in the
Sprague–Dawley rats given alcohol pretreatment (Fig. 1B).
There was no significant difference in the preference between
the strains when no alcohol pretreatment was given. As with
Effect of Alcohol Pretreatment on Alcohol Consumption
in Sprague-Dawley and P-rats: Average over Four Weeks
Effect of Alcohol Pretreatment on Alcohol Preference
in Sprague-Dawley and P-rats: Average over Four Weeks
Fig. 1. (A) Effects of an oral dose (1 g/kg) of alcohol administration on sub-
sequent alcohol intake (g/kg/day) in alcohol-preferring P and Sprague–
Dawley rats. Data represent the means ± SEM of 4 weeks. N = 8 for each
group of Sprague–Dawley rats, 9 for vehicle-treated P rats and 6 for alcohol-
treated P rats. (B) Effects of an oral dose (1 g/kg) of alcohol administration on
subsequent alcohol preference in alcohol-preferring P and Sprague–Dawley
rats.Data represent the means ± SEM of 4 weeks.N = 8 for each group of Spra-
gue–Dawley rats, 9 for vehicle-treated P rats and 6 for alcohol-treated P rats.
Rezvani et al.220
alcohol consumption, there was a significant strain × alcohol
pretreatment interaction (F(1,27) = 7.41, P < 0.025). Follow-
up tests of the simple main effects showed that alcohol
pretreatment significantly (P < 0.005) reduced alcohol
preference in the Sprague–Dawley rats while the P rats did
not show a change. There was a significantly (P < 0.0005)
greater alcohol preference in the P rats pretreated with alcohol
vs pretreated Sprague–Dawley rats. Sprague–Dawley rats
showed aversion to alcohol with only 36.0% preference for
the alcohol-containing bottle. This was well below the 50%
chance level and the 68.1% level shown by Sprague–Dawley
rats not given alcohol pretreatment. This is the same pattern of
effects as shown with the grams per kilogram per day alcohol
down in alcohol preference by the groups. As with the
consumption, the preference data showed a significant main
effect of weeks (F(3,81) = 14.03, P < 0.0005) with overall
increase in percent preference over weeks. There were no
significant interactions of weeks with either strain or alcohol
pretreatment was clearly seen during the fourth week of the
The results of this study demonstrate that selectively bred al-
cohol-preferring P rats, compared with Sprague–Dawley rats,
are less sensitive to the aversive effects of alcohol. Sprague–
Dawley rats showed a long-lasting decrease in alcohol con-
sumption and preference for 4 weeks after a single alcohol
pretreatment. On the other hand, the P rats given the same al-
cohol pretreatment showed no such effect. Without alcohol
pretreatment, there were no discernible differences between
Sprague–Dawley and P rats in either the amount of alcohol
consumed and the percent alcohol preference. The lack of dif-
ference in alcohol preference between Sprague–Dawley and P
rats before being treated with alcohol is likely contributed to
the palatability of 5% (v/v) alcohol. It is likely that at higher
concentrations of alcohol, the P rats will drink significantly
more alcohol than Sprague–Dawley rats. The increased alco-
hol consumption seen in P rats seems to be entirely due to a
lack of the normal aversion to initial alcohol treatment rather
than increased preference. It has previously been shown that
the P rats self-administer significant quantities of alcohol in-
tragastrically, indicating that post-absorptive effects of
alcohol, rather than its taste or smell, are reinforcing for the
P rats (Waller et al., 1983a). It has also been demonstrated that
P rats find the post-ingestional effects of high dose of alcohol
less aversive and low dose of alcohol more rewarding than
their control counterpart, the NP rats (Froehlich et al.,
1988). The possible effects of taste in the current experiment
should be ruled out because alcohol solution was delivered
directly into the stomach and not into the mouth. Genetic dif-
ferences in sensitization and tolerance development to alcohol
between the P and NP rats have also been reported. Kurtz and
colleagues (1996) studied the initial sensitivity to alcohol and
the development of tolerance comparing the P with the NP
rats. They demonstrated that P rats were less sensitive to the
behaviorally impairing effects of alcohol than were NP rats, as
evidenced by longer latency to lose righting reflex and a
shorter time to recover following an acute dose of alcohol.
It has also been shown that within-session tolerance to alcohol
is developed in P rats (Waller et al., 1983b). Thus, it can be
speculated that the combination of lack or less sensitivity to
alcohol, less aversion to alcohol and development of acute tol-
erance to alcohol may serve to increase alcohol intake in P
rats. As mentioned in the Introduction, in the conditioned taste
aversion (CTA) paradigm (Froehlich et al., 1988) and in the
conditioned place preference (CPP) paradigm (Stewart et al.,
1996), both P and NP rats showed aversion to alcohol. How-
ever, the magnitude of aversion to alcohol was significantly
less in P rats compared with NP rats. In one study, no differ-
ence was found between P and NP rats using the CPP
paradigm (Schechter, 1992). Our results should not be directly
compared with CTA and CPP results because measuring
drinking alcohol is the most appropriate test for testing aver-
sion to alcohol in rats that have been selectively bred to drink
Thefactthat onesingleintragastricadministration ofalcohol
impaired acquisition of alcohol drinking in Sprague–Dawley
rats but not in P rats suggests that this selectively bred line
might have less aversion to alcohol than the outbred Spra-
gue–Dawley line. One can speculate that the diminished
aversion to alcohol may partly explain the high level of drink-
ing in selectively bred P rats.
Interestingly, similar findings have been reported in humans
regarding the importance of diminished aversive effects of al-
cohol in alcoholics. A lower level of response, i.e. lower
sensitivity to subjective and intoxicating effects of alcohol,
Effect of Alcohol Pretreatment on Alcohol Consumption
in Sprague-Dawley and P-rats: Weekly Intake
Strain and Pretreatment
Effect of Alcohol Pretreatment on Alcohol Preference
in Sprague-Dawley and P-rats: Weekly Choice
P-rats/ No Alcohol
Strain and Pretreatment
Fig. 2. (A) Effects of one single oral dose (1 g/kg) of alcohol administration
on subsequent alcohol intake (g/kg/day) in alcohol-preferring P and Sprague–
8 for each group of Sprague–Dawley rats, 9 for vehicle-treated P rats and 6 for
alcohol-treated P rats. (B) Effects of one single oral dose (1 g/kg) of alcohol
administration on subsequent preference for alcohol in alcohol-preferring P
and Sprague–Dawley rats. Data represent the means ± SEM of weekly intake
P rats and 6 for alcohol-treated P rats.
Lack of Aversion to Alcohol in P Rats221
has been repeatedly demonstrated in adult children of alco-
holics (Schuckit and Smith, 1996; Schuckit and Gold, 1998;
Schuckit and Smith, 2000, 2001; Schuckit et al., 2005, 2006,
2009; Trim et al., 2009). Indeed, low level of response to al-
cohol is shown to be a unique risk factor for future heavy
drinking (Trim et al., 2009). Clearly, genetic factors play a
major role in lower sensitivity to alcohol, which may lead
to heavy drinking.
The current results clearly call for further research, espe-
cially regarding the relative roles of pharmacokinetic and
pharmacodynamic effects underlying the diminished aversive
effects of alcohol in P rats. However, since P rats and its con-
trol counterpart, NP rats, metabolize alcohol at the same rate
(Waller et al., 1983b), it is unlikely that the lack of aversion to
alcohol in P rats is of pharmacokinetic nature. Determining
critical pharmacodynamic factors underlying this effect will
be key not only for understanding the basic biology of addic-
tion but also for developing new treatments to combat
alcoholism. This study demonstrates the importance of aver-
sion to initial alcohol exposure in the control of later voluntary
consumption. The lack of normal aversion can be a key factor
leading to heavy alcohol consumption and increased risk for
Acknowledgements—We are grateful to Drs Lawrence Lumeng and Richard Bell of the
Indiana University School of Medicine for generously providing alcohol-preferring P rats
(supported by NIAAA; R24 AA015512-02).
Conflict of interest statement. None declared.
Colombo G. (1997) Ethanol drinking behavior in Sardinian alcohol
preferring rats. Alcohol Alcohol 32:443–53.
Eriksson K. (1968) Genetic section for voluntary alcohol consump-
tion in the Albino rat. Science 159:739–41.
Froehlich JC, Harts J, Lumeng L, Li TK. (1988) Differences in re-
sponse to the aversive properties of ethanol in rats selectively bred
for oral ethanol preference. Pharm Biochem Behav 31:215–22.
Kurtz DL, Stewart RB, Zweifel M et al. (1996) Genetic difference in
tolerance and sensitization to the sedative/hypnotic effects of al-
cohol. Pharm Biochem Behav 53:585–91.
Li TK, Lumeng L, Doolittle DP. (1993) Selective breeding for alco-
hol preference and associated responses. Behav Genet 23:163–70.
Lumeng L, Waller MB, McBride WJ, Li TK. (1982a) Different sen-
sitivities in alcohol-preferring and -nonpreferring rats. Pharm
Biochem Behav 16:124–30.
Lumeng L, Waller MB, McBride WJ, Li TK. (1982b) Different sen-
sitivities to alcohol in alcohol preferring and -nonpreferring rats.
Pharm Biochem Behav 13:777–80.
Mardones J, Segovia-Riquelme N. (1983) Thirty-two years of selec-
tion of rats by ethanol preference: UCh and UChB strains.
Neurobehav Toxicol Teratol 5:171–8.
Overstreet DH, Lee YW, Rezvani AH. (1996) Suppression of alcohol
intake following injection of the Chinese herbal medicine, XJL
(NPI-028). Alcohol Clin Exp Res 20:221–7.
Rezvani AH, Overstreet DH, Janowsky DS. (1990) Genetic serotonin
deficiency and alcohol preference in the Fawn-Hooded rats. Alco-
hol Alcohol 25:573–5.
Rezvani AH, Grady DR. (1994) Suppression of alcohol consumption
by fenfluramine in Fawn-Hooded rats with serotonin dysfunction.
Pharmacol Biochem Behav 48:105–10.
Rezvani AH, Peek AE, Grady DR, Pucilowski O. (1995) Inhibition
of nitric oxide (NO) synthesis attenuated alcohol consumption in
two strains of alcohol preferring rats. Pharmacol Biochem Behav
Rezvani AH, Overstreet DH, Ying Y et al. (1997) Attenuation of
alcohol consumption by a novel nontoxic ibogaine analogue
(18-methoxycoronaridine) in alcohol-preferring rats. Pharmacol
Biochem Behav 58:615–9.
Rezvani AH, Overstreet DH, Ying Y, Clark E Jr. (1999) Attenuation
of alcohol intake by the extract of Hypericum perforatum (St.
John’s Wort) in two strains of alcohol preferring rats. Alcohol Al-
Rezvani AH, Overstreet DH, Mason GA et al. (2000) Combination
pharmacotherapy: a mixture of small doses of naltrexone, fluox-
etine, and a TRH analog reduces alcohol intake in three strains of
alcohol preferring rats. Alcohol Alcohol 35:76–83.
Rezvani AH, Overstreet DH, Levin ED et al. (2007) The effects of
atypical anxiolytic N-phenyl-2-[1-[3-(2-pyridinylethynyl)benzo-
yl]-4-piperidine]acetamide (JNJ-5234801) on alcohol intake in
alcohol-preferring P rats: comparison with naltrexone. Alcohol
Clin Exp Res 31:57–63.
Rezvani AH, Overstreet DH, Vaidya AH et al. (2009) Carisba-
mate, a novel antiepileptic candidate compound, attenuates
alcohol intake in alcohol preferring rats. Alcohol Clin Exp
Schechter MD. (1992) Locomotor activity but not conditioned place
preference is differentially affected by a moderate dose of ethanol
administered to P and NP rats. Alcohol 9:185–8.
Schuckit MA. (2009) An overview of genetic influences in alcohol-
ism. J Subs Abuse Treat 36:s5–14.
Schuckit MA, Gold EO. (1998) A simultaneous evaluation of multi-
ple markers of ethanol/placebo challenges in sons of alcoholics
and controls. Arch Gen Psychiatry 45:211–6.
Schuckit MA, Smith TL. (1996) An 8-year follow-up of 450
sons of alcoholic and control subjects. Arch Gen Psychiatry
Schuckit MA, Smith TL. (2000) The relationships of a family history
of alcohol dependence, a low level of response to alcohol and six
domains of life functioning to the development of alcohol use dis-
orders. J Stud Alcohol 61:827–35.
Schuckit MA, Smith TL. (2001) The clinical course of alcohol de-
pendence associated with low level of response to alcohol.
Schuckit MA, Smith TL, Danko GP et al. (2009) An evaluation of
the full level of response to alcohol model of heavy drinking in
COGA offspring. J Stud Alcohol Drugs 70:436–45.
Schuckit MA, Smith TL, Pierson J et al. (2006) Relationships among
the level of response to alcohol and the number of alcoholic rel-
ative in predicting alcohol-related outcomes. Alcohol Clin Exp
Schuckit MA, Wilhelmsen K, Smith TL et al. (2005) Autosomal
linkage analysis for the level of response to alcohol. Alcohol Clin
Exp Res 29:1976–82.
Stewart RB, Kurtz DL, Zweifel M et al. (1992) Difference in the hy-
pothermic response to ethanol in rats selectively bred for oral
ethanol preference and nonpreference. Psychopharmacol 106:
Stewart RB, Murphy JM, McBride WJ et al. (1996) Place condition-
ing with alcohol in alcohol-preferring and -nonpreferring rats.
Pharmacol Biochem Behav 53:487–91.
Trim RS, Schuckit MA, Smith TL. (2009) The relationships of the
level of response to alcohol and additional characteristics to alco-
hol use disorder across adulthood: a discrete-time survival
analysis. Alcohol Clin Exp Res 33:1562–70.
Waller MB, McBride WJ, Gatto GJ et al. (1983a) Intragastric self-
infusion of ethanol by ethanol-preferring and -nonpreferring lines
of rats. Science 225:78–80.
Waller MB, McBride WJ, Lumeng L, Li TK. (1983b) Initial sensitiv-
ity and acute tolerance to alcohol in the P and NP lines of rats.
Pharmacol Biochem Behav 19:683–6.
Rezvani et al. 222