Content uploaded by Raphael Mechoulam
Author content
All content in this area was uploaded by Raphael Mechoulam on Nov 18, 2014
Content may be subject to copyright.
Psychopharmacology (2004) 175:360–366
DOI 10.1007/s00213-004-1825-7
ORIGINAL INVESTIGATION
Linda A. Parker · Page Burton · Robert E. Sorge ·
Christine Yakiwchuk · Raphael Mechoulam
Effect of low doses of D
9
-tetrahydrocannabinol and cannabidiol
on the extinction of cocaine-induced and amphetamine-induced
conditioned place preference learning in rats
Received: 21 October 2003 / Accepted: 25 January 2004 / Published online: 5 May 2004
Springer-Verlag 2004
Abstract Rationale: Using the place-preference condi-
tioning paradigm, we evaluated the potential of the two
most prominent cannabinoids found in marijuana, the
psychoactive component D
9
-tetrahydrocannabinol (D
9
-
THC) and the nonpsychoactive component cannabidiol
(CBD), to potentiate extinction of a cocaine-induced and
an amphetamine-induced conditioned place preference in
rats. Methods: To determine the effects of pretreatment
with D
9
-THC or CBD on extinction, a cocaine-induced
and amphetamine-induced place preference was first es-
tablished. Rats were then given an extinction trial, during
which they were confined to the treatment-paired floor for
15 min. Thirty minutes prior to the extinction trial, they
were injected with a low dose of D
9
-THC (0.5 mg/kg),
CBD (5 mg/kg) or vehicle. The potential of the CB
1
re-
ceptor antagonist, SR141716, to reverse the effects of D
9
-
THC or CBD was also evaluated. To determine the he-
donic effects of CBD, one distinctive floor was paired
with CBD (5 mg/kg) and another with saline. Finally, to
determine the effect of D
9
-THC or CBD on the estab-
lishment and/or the expression of a place preference
during four cycles of conditioning trials, rats were in-
jected with D
9
-THC (0.25–1 mg/kg), CBD (5 mg/kg) or
vehicle 25 min prior to receiving an injection of am-
phetamine followed by placement on the treatment floor;
on alternate days, they received injections of vehicle
followed by saline and placement on the nontreatment
floor. The rats then received two test trials; on one trial
they were pretreated with the cannabinoid and on the
other trial with vehicle. Results: D
9
-THC and CBD po-
tentiated the extinction of both cocaine-induced and am-
phetamine-induced conditioned place preference learning,
and this effect was not reversed by SR141716. The
cannabinoids did not affect learning or retrieval, and CBD
was not hedonic on its own. Conclusions: These results
are the first to show that both D
9
-THC, which acts on both
CB1 and CB2 receptors, and CBD, which does not bind to
CB
1
or CB
2
receptors, potentiate the extinction of con-
ditioned incentive learning.
Keywords Memory · Cannabinoids ·
D
9
-Tetrahydrocannabinol · Cannabidiol · Extinction ·
Cocaine · Amphetamine · Conditioned place-preference
learning · Incentive · Classical conditioning
For organisms to gain maximum benefit from their en-
vironment, they must not only learn new associations, but
also extinguish associations that are no longer useful.
Recent evidence implicates the endogenous cannabinoid
system in the extinction process (Marsicano et al. 2002;
Varvel and Lichtman 2002). Marsicano et al. (2002) re-
ported that CB
1
knockout mice show impaired extinction,
in auditory fear-conditioning tests, with unaffected
memory acquisition and consolidation. Furthermore,
treatment of wild-type mice with the CB
1
antagonist,
SR141716 (SR), mimicked the response of CB
1
deficient
mice, suggesting that activation of the CB
1
receptor is
involved in extinction of learned memories. Using a very
different task, the Morris water maze, Varvel and Licht-
man (2002) reported that CB
1
knockout mice and wild-
type mice exhibit identical acquisition rates in learning to
swim to a fixed platform; however, the CB
1
-deficient
mice demonstrate considerable deficits in a reversal task
in which the location of the hidden platform is moved to
the opposite side of the tank. The difference in perfor-
mance was attributed to disruption of extinction of the
L. A. Parker (
)
) · P. Burton · C. Yakiwchuk
Department of Psychology,
Wilfrid Laurier University,
Waterloo, ON, Canada, N2L 3C5
e-mail: lparker@wlu.ca
Tel.: +1-519-884-0710
Fax: +1-519-746-7605
R. E. Sorge
Center for Studies in Behavioural Neuroscience,
Concordia University,
Montreal, QC, Canada
R. Mechoulam
Department of Medicinal Chemistry and Natural Products,
Hebrew University of Jerusalem,
Jerusalem, Israel
learned response. These results suggest that the endoge-
nous cannabinoid system modulates extinction of learned
responses.
Since animals deficient in CB
1
receptors show a failure
to extinguish learned behavior, cannabinoid agonists
might be expected to facilitate extinction of learned be-
haviors in nondeficient animals. The most prevalent
cannabinoids found in marijuana are the psychoactive
component D
9
-tetrahydrocannabinol (D
9
-THC) and the
nonpsychoactive component (Mechoulam et al. 2002)
cannabidiol (CBD). Unlike D
9
-THC, CBD does not bind
to CB
1
or CB
2
receptors; instead it may act by blocking
the reuptake of anandamide or act on an unknown re-
ceptor (Mechoulam et al. 2002). In mice and rats re-
spectively, CBD is a highly effective anti-inflammatory
agent (Malfait et al. 2000) and plays a role in the sup-
pression of nausea (Parker et al. 2002). The experiments
described here evaluated the potential of low doses of D
9
-
THC and CBD to modulate extinction of cocaine and
amphetamine incentive learning using the place-prefer-
ence procedure. Conditioned incentive learning has been
suggested to play a role in craving (Robinson and Ber-
ridge 1993; DeWit and Stewart 1981); therefore, treat-
ments that potentiate extinction of conditioned incentive
learning may be useful in the control of craving.
Place-preference learning is a commonly used measure
of positive incentive learning (Cunningham 1998). In the
typical procedure, a drug is paired with one distinctive
environment and vehicle is paired with another. Follow-
ing conditioning, the animal is given access to both
contexts in a drug-free state. A conditioned place pref-
erence is demonstrated if the animal spends a greater
amount of time in the drug-paired environment than the
vehicle-paired environment. The preference demonstrates
that the conditioned stimulus (CS) environment associ-
ated with the drug effects has acquired conditioned in-
centive properties (Robinson and Berridge 1993; Cun-
ningham 1998). Place-preference learning follows the
principles of classical conditioning, including extinction.
In extinction, animals display a reduced approach to the
CS context following its experience in the absence of the
unconditioned stimulus (UCS) drug (Bardo et al. 1986;
Calcagnetti and Schechter 1993; Tzschentke and Schmidt
1995; Cunningham et al. 1998; Parker and McDonald
2000; Meuller et al. 2000; Itzhak and Martin 2002).
There are two sets of experiments reported here. The
first set evaluated the effects of nonhedonic doses of D
9
-
THC and CBD to promote extinction of place-preference
learning and the potential of SR to reverse this effect. The
second set evaluated the effect of D
9
-THC and CBD on
the acquisition and/or initial expression of place-prefer-
ence learning. The doses of D
9
-THC were selected on the
basis of prior work indicating that they are neither re-
warding nor aversive (Parker and Gillies 1995; Mallet and
Beninger 1998b; Robinson et al. 2003), yet they have a
physiological effect of suppression of nausea (Limebeer
and Parker 1999; Parker et al. 2003). The dose of CBD
(5 mg/kg) was selected on the basis of prior work that has
shown it to be effective in suppressing nausea in rats
(Parker et al. 2003), inflammatory pain in mice (Malfait et
al. 2000), and vomiting in shrews (Parker et al. 2004;
Kwiatkowska et al. 2004). Additionally, the potential of
CBD to produce place conditioning on its own was
evaluated.
Materials and methods
Subjects
The subjects were experimentally naive male Sprague-Dawley rats
weighing between 250 g and 310 g on the first conditioning day.
They were pair-housed in Plexiglas cages (472620 cm) with
unrestricted access to food and water, except as specified. The room
was illuminated on a 12-h/12-h light/dark schedule with the lights
on at 0800 hours, and all procedures occurred in the light. Upon
arrival in the laboratory, the rats were handled daily for 1 week
prior to the conditioning trials. All procedures were approved by
the Institutional Animal Care Committee according to the guide-
lines of the Canadian Council for Animal Care.
Drugs
All drugs were injected intraperitoneally (i.p.). D
9
-THC [supplied
by the National Institute on Drug Abuse (NIDA)], CBD (supplied
by R. Mechoulam) and SR (supplied by NIDA) were prepared in a
solution of 1 ml ethanol/1 ml Cremaphor (Sigma)/18 ml saline. D
9
-
THC was prepared as a 0.5-mg/ml solution of the vehicle and was
administered at a volume of 0.5, 1.0 or 2.0 ml/kg (0.25, 0.5, 1.0 mg/
kg), depending on the experiment. CBD was prepared as a 5-mg/ml
solution of the vehicle and was administered in a volume of 1 ml/kg
(5 mg/kg). SR was prepared as a 2.5-mg/ml solution of vehicle and
was administered at a volume of 1 ml/kg (2.5 mg/kg). Vehicle was
administered in a volume of 1 ml/kg. Cocaine HCl (supplied by
NIDA) was prepared in a solution of 5 mg/ml (expressed as the salt)
and administered in a volume of 3 ml/kg (15 mg/kg). Amphetamine
(supplied by NIDA) was prepared in a solution of 1 mg/ml and
injected in a volume of 3 ml/kg (3 mg/kg).
Apparatus
The place-preference apparatus consisted of a black Plexiglas
rectangular box (602525 cm) with a mesh top. The black
Plexiglas floors of the box were interchangeable. During condi-
tioning, the entire floor of the apparatus consisted of either small
holes (1 cm in diameter arranged 1 cm apart from one another) or
wire mesh (1-cm squares) attached to the black Plexiglas floor. For
half of the rats, the hole floor was the treatment floor and for the
other half of the rats the mesh floor was the treatment floor.
During testing, three floors were inserted into the apparatus: one
made of holes (25 cm25 cm), one made of mesh (25 cm25 cm)
and a center floor made of smooth black Plexiglas (9 cm25 cm).
The tactile stimulus properties of the hole and the mesh floors were
the same as those used in conditioning. The novel smooth black
Plexiglas floor was placed in the center of the chamber and the hole
and mesh floors were placed on either side (with the side coun-
terbalanced among the groups). The amount of time that each rat
spent on each of the three floors was automatically recorded using a
camera mounted to the ceiling that sent the signal to a computer.
The signal was recorded and subsequently analyzed by the Noldus
Etho-Vision videotracking system (Noldus Information Technolo-
gy, Sterling, VA).
Prior to the experiments described below, the rats were given a
single 15-min pretest in the testing apparatus. In none of these tests
was there a significant difference between the time spent on the
mesh floor and on the holes floor indicating that the apparatus
provides an unbiased test of place conditioning (e.g., Cocaine US:
361
mesh, mean=430 s, SEM=12.4; holes, mean=416 s, SEM=12.4;
t
47
=0.9).
Procedure
Experiment 1: effect of nonhedonic doses of CBD
and THC on extinction of place-preference learning
Experiment 1a: evaluation of the potential of CBD to produce place
conditioning. On each of four conditioning trial cycles, rats in
group CBD (n=12) were injected with 5 mg/kg CBD or vehicle
(1 ml/kg), 5 min prior to placement in a chamber with a distinctive
floor (holes or mesh) for 30 min. A conditioning cycle consisted of
one CBD trial and one vehicle trial separated by 24 h. The floors
were removed and washed following each trial. The order of the
CBD trial within a cycle and the floor paired with CBD were
counterbalanced. Rats in group VEH (n=12) were injected with
vehicle on each trial; the floor designated as the treatment-paired
floor was randomly determined. Each of the cycles of conditioning
was separated by 48–72 h. Seventy-two hours after the final con-
ditioning trial, the rats all received a 15-min place-preference test in
the test apparatus, and the amount of time spent on each floor was
monitored.
Experiment 1b: effect of THC and CBD on extinction
of cocaine-induced and amphetamine-induced place preference.
Pretest
All rats received a 15-min drug-free pretest in the testing ap-
paratus to determine baseline preference for the floors. The activity
of the rat was automatically recorded. Following each pretest,
conditioning trial and test trial, the floors were removed and wa-
shed.
Conditioning
For cocaine, rats were given two (n=36) or four (n=48) condi-
tioning cycles, during which they were injected with cocaine or
saline 5 min prior to placement in a chamber with a distinctive floor
(holes or mesh) for 30 min. A conditioning cycle consisted of one
cocaine trial and one saline trial separated by 24 h. The order of the
cocaine trial within a cycle and the floor paired with cocaine were
counterbalanced. Each of the cycles of conditioning was separated
by 48–72 h. For amphetamine, the rats were given four condi-
tioning cycles (n=36), during which they were injected with am-
phetamine or saline 5 min prior to placement in the conditioning
apparatus, as described above. For saline, an additional 24 rats were
given four conditioning cycles, as described above, except that they
received saline injections 5 min prior to placement in each of the
floors. One floor was designated as the treatment-paired floor and
the other was designated as the nontreatment-paired floor by ran-
dom assignment.
Extinction
Seventy-two hours after the final conditioning trial, the rats
(n=8–12 per group) were injected with 0.5 mg/kg D
9
-THC, 5 mg/kg
CBD or vehicle (1 ml/kg), 30 min prior to receiving a single ex-
tinction trial during which they were confined to the treatment-
paired floor for 15 min. Among the cocaine and amphetamine
conditioned groups, an additional group of rats (no extinction; n=8–
12 per group) were not given the extinction trial; they remained in
their home cage on the day of extinction.
Testing
All rats received a place-preference test 48 h following the
extinction trial. They were placed in the test apparatus for 15 min,
and the amount of time spent on each floor was automatically
recorded.
Experiment 1c: effect of SR on D9-THC and CBD modulation of
extinction. The potential of the CB
1
receptor antagonist, SR to
reverse modulation of extinction by D
9
-THC (0.5 mg/kg) and CBD
(5 mg/kg) was evaluated in two replications that produced a similar
pattern of results and were, therefore, pooled. The dose of 2.5 mg/
kg i.p. was selected on the basis of findings that this dose, but not
1 mg/kg, reversed the effects of the synthetic cannabinoid, HU-210,
on lithium-induced conditioned gaping in rats (Parker et al. 2003).
The rats received a total of four cycles of conditioning with am-
phetamine using an identical conditioning and testing procedure as
described above. Seventy-two hours after the final conditioning
trial, the rats received a single 15-min extinction trial that began
30 min after they were injected i.p. (with 5 min between injections)
with two solutions: SR-D
9
-THC (n=21), SR-CBD (n=21), SR-VEH
(n=21), VEH–VEH (n=21).
Experiment 2: evaluation of effect of D
9
-THC and CBD
on the establishment and the expression
of amphetamine-induced place-preference learning
D9-THC. On each of four conditioning cycles (separated by 48–
72 h), 48 rats received a 30-min treatment trial following an in-
jection of 3 mg/kg amphetamine and a 30-min nontreatment trial
following an injection of saline. The treatment and nontreatment
trials were separated by 24 h. On the treatment trial, various groups
(n=12 per group) were injected with vehicle, 0.25 mg/kg, 0.5 mg/kg
or 1.0 mg/kg D
9
-THC, 30 min prior to the injection of amphet-
amine. On the nontreatment trial, all rats were injected with vehi-
cle, 30 min prior to the injection of saline. The floor paired with the
treatment and the order of the trials was counterbalanced. In order
to evaluate the effect of D
9
-THC on the expression of a previously
established place preference, the rats received two test trials, the
vehicle test trial occurred 30 min following an injection of vehicle
and the D
9
-THC test trial occurred 30 min following an injection of
D
9
-THC. The dose of D
9
-THC administered on the D
9
-THC test
trial was the same dose given during conditioning; group VEH
received a dose of 0.5 mg/kg D
9
-THC on the test trial. The order of
D
9
-THC and VEH test trials were counterbalanced.
Cannabidiol. The effect of CBD on the establishment and the ex-
pression of amphetamine place-preference learning was evaluated.
During each of four conditioning cycles, on the treatment trial, the
rats received an injection of 5 mg/kg CBD (n=9) or vehicle (n=9)
30 min prior to an injection of amphetamine 5 min prior to
placement on one floor for 30 min. On the nontreatment trial, they
received an injection of vehicle 30 min prior to an injection of
saline 5 min prior to placement on the other floor for 30 min. The
order of trials within a cycle and the floor paired with amphetamine
were counterbalanced. The test trials began 72 h later. On one test
trial, rats were pretreated with 5 mg/kg CBD and, on the other test
trial, they were pretreated with vehicle (5 ml/kg) 30 min prior to a
15-min place-preference test. The trials were separated by 24 h, and
the order was counterbalanced across the groups.
Data analysis
The data were converted to difference scores in the following
manner: mean seconds on the treatment-paired floor minus the
nontreatment-paired floor. Therefore, a value of 0.0 represents
neither a preference nor aversion to the floor. Groups were com-
pared using factorial analyses of variance (ANOVAs) with pairwise
least significant difference (LSD) tests when appropriate. Addi-
tionally, single sample t-tests compared the mean difference score
of each group with a value of 0.0.
Results
Experiment 1a: CBD and place conditioning
CBD produced neither a place preference nor aversion.
During the test trial, the mean seconds spent on the
362
treatment-paired floor minus the nontreatment-paired
floor by group CBD (mean=120 s, SEM=84.0) did not
significantly differ from that of group VEH (mean=53 s,
SEM=61.1). Neither of these scores differed significantly
from a value of 0.0 assessed by single sample t-tests.
Experiment 1b: effect of D
9
-THC and CBD
on extinction of cocaine-induced
and amphetamine-induced place preference
Cocaine
When administered during extinction training, both D
9
-
THC and CBD potentiated extinction of cocaine-induced
place-preference learning, regardless of the number of
conditioning trials. The mean (€SEM) number of seconds
that the rats in each group spent on the cocaine-paired
minus the saline-paired floor is presented in Fig. 1. A 42
ANOVA, with the between group factors of groups (no
extinction, VEH, CBD and D
9
-THC) and number of
conditioning trials (2 or 4), revealed only a significant
main effect of groups (F
3, 76
=4.6; P<0.01). Both groups
given the cannabinoid (D
9
-THC or CBD) during extinc-
tion spent less time on the cocaine-paired floor than either
the group given the vehicle during extinction (P val-
ues<0.05) or than the group given no extinction trial (P
values<0.05). Single-sample t-tests, pooled across the
factor of number of conditioning trials, revealed that
groups no extinction (t
19
=4.0; P<0.001) and vehicle
[t
(21)3.9
; P<0.001] displayed a place preference defined as
a significant difference from 0.0; however groups D
9
-
THC and CBD displayed neither a place preference nor a
place aversion.
Amphetamine
When administered during extinction training, D
9
-THC
and CBD both potentiated the extinction of an amphet-
amine-induced place preference. Figure 2 presents the
mean (€SEM) seconds on the amphetamine-paired floor
minus the saline-paired floor for the groups pretreated
with VEH, CBD, D
9
-THC and those that received no
extinction. A single-factor ANOVA indicated that the
groups significantly differed (F
3, 32
=4.7; P<0.01). Subse-
quent LSD comparison tests revealed that those rats re-
ceiving either no extinction or vehicle displayed a greater
preference for the amphetamine-paired floor than those
rats pretreated with CBD or D
9
-THC prior to a single
extinction trial (P values<0.05). Single-sample t-tests re-
vealed that groups VEH (t
8
=3.0; P<0.025) and no ex-
tinction [t
(8)4.4
; P<0.01] displayed a difference score
greater than 0.0, but groups CBD and D
9
-THC did not
differ significantly from a value of 0.0.
Saline
When administered during extinction training, neither
CBD nor D
9
-THC changed baseline floor preference; the
effect of extinction condition was not significant.
Fig. 1 Mean (€SEM) seconds on the cocaine-paired floor minus
the saline-paired floor by the various extinction groups previously
given two or four conditioning trial cycles. Asterisks (*P<0.05;
**P<0.01) indicate means that differed significantly from a value
of 0.0
Fig. 2 Mean (€SEM) seconds on the amphetamine-paired floor
minus the saline-paired floor by the various extinction groups.
Asterisks (*P<0.05; **P<0.01) indicate means that differed sig-
nificantly from a value of 0.0
363
Experiment 1c: effect of SR on D
9
-THC
and CBD modulation of extinction
As is apparent in Fig. 3, SR did not reverse the potentiated
extinction produced by either D
9
-THC or CBD. A single-
factor ANOVA revealed a significant extinction drug
effect (F
3, 80
=3.4; P=0.021). Subsequent pairwise com-
parison tests revealed that group VEH–VEH displayed a
significantly greater place preference than groups SR-D
9
-
THC or SR-CBD (P values<0.05), but did not differ from
group SR–VEH. However, group SR–VEH did not sig-
nificantly differ from groups SR-D
9
-THC or SR-CBD.
Single sample t-tests indicated that groups VEH–VEH
(P<0.001) and SR–VEH (P<0.05), but not groups SR–D
9
-
THC or SR–CBD, differed significantly from a value of
0.0.
Experiment 2. Effect of D
9
-THC or CBD
on the establishment and expression
of amphetamine place-preference conditioning
Neither D
9
-THC nor CBD affected either the establish-
ment or the expression of amphetamine-induced place-
preference learning. For D
9
-THC, the 42 mixed-factors
ANOVA with the between-groups factor of conditioning
pretreatment (0, 0.25, 0.5 or 1.0 mg/kg D
9
-THC) and the
within-groups factor of test pretreatment (VEH or D
9
-
THC) revealed no significant effects; however, a single
sample t-test for each test trial pooled across groups re-
vealed that rats displayed a place preference (as defined
as a difference from 0.0) on both the VEH test trial
[t
47
=7.0; P<0.001] and on the D
9
-THC test trial [t 47=7.5;
P<0.001].
For CBD, a 22 mixed-factors ANOVA with the be-
tween-groups factor of conditioning pretreatment (CBD
or VEH) and within-groups factor of test pretreatment
(CBD or VEH) revealed no significant effects; but a
single sample t-test pooled across conditioning pretreat-
ment groups revealed that rats displayed a place prefer-
ence (differed from 0.0) on both the VEH test trial
(t
17
=2.80; P<0.025) and the CBD test trial (t
17
=3.06;
P<0.01).
General discussion
Low doses of the principal cannabinoids found in mari-
juana—the psychoactive component D
9
-THC and the
nonpsychoactive component CBD—both potentiated the
extinction of cocaine place-preference and amphetamine
place-preference learning. This effect was not simply a
summation of the hedonic properties of the cannabinoids
and the incentive properties of the stimulant-paired place,
because rats conditioned with saline on both trials of each
cycle and extinguished with the cannabinoids did not
avoid the floor paired with the cannabinoid during the
extinction trial. Furthermore, when administered over
four conditioning cycles, the dose of CBD that potentiated
extinction of cocaine-induced and amphetamine-induced
place-preference learning produced neither a place aver-
sion nor a place preference. However, given that the
variability was high in the CBD conditioned groups, it is
possible that further training would have produced a place
aversion. Previous work indicated that, at the low doses
employed in these experiments, D
9
-THC produces neither
a place preference nor place aversion (Parker and Gillies
1995; Mallet and Beninger 1998b, Robinson et al. 2003).
Therefore, as has been shown with fear conditioning
(Marsicano et al. 2002) and spatial learning (Varvel and
Lichtman 2002), the cannabinoid system appears to play a
role in the extinction of conditioned incentive learning.
Two findings from this first set of experiments suggest
that the cannabinoid potentiation of extinction is not
mediated by action on known cannabinoid receptors: (1)
not only D
9
-THC, but also CBD which does not bind to
the CB
1
or CB
2
receptor (Mechoulam et al. 2002), pro-
moted extinction, and (2) the CB1 receptor antagonist, SR
(2.5 mg/kg), did not reverse the potentiation of extinction
by D
9
-THC. These results suggest that the effect of these
agents on extinction of positive incentive learning may
not be CB
1
mediated. Indeed, there is recent evidence that
the synthetic cannabinoid agonist WIN55,212-2 inhibits
glutamatergic transmission in CB1 (/) mice, indicating
that a novel cannabinoid-sensitive receptor may be re-
sponsible for this inhibition (Hajos et al. 2001).
Although a dose of 2.5 mg/kg SR is well above the
threshold to reverse the effect of CB1 agonists, it is
possible that a higher dose may have been more effective.
It is also possible that SR may have produced effects on
its own in this paradigm. Rats pretreated with SR-VEH
did not differ from those pretreated with SR-THC or SR-
CBD during extinction training. Therefore, it is conceiv-
able that SR may have produced a motivational effect
during extinction training. It has been shown that when
administered on multiple conditioning trials, SR produces
a place preference at doses ranging from 0.25 mg/kg to
3.0 mg/kg (Cheer et al. 2000); therefore, it is possible that
the exposure to the antagonist during extinction training
Fig. 3 Mean (+SEM) seconds on the amphetamine-paired floor
minus the saline-paired floor among the groups that varied by ex-
tinction experience. Asterisks (*P<0.05; **P<0.01) indicate means
that differed significantly from a value of 0.0
364
could have resulted in apparent reversal of the effects of
CBD or D
9
-THC on extinction because of the re-estab-
lishment of a preference.
Although the cannabinoids potentiated extinction,
neither D
9
-THC nor CBD affected the establishment or
the expression of amphetamine place preference. This is
consistent with reports that CB
1
knockout mice were
unimpaired in the acquisition or the initial expression of
conditioned fear (Marsicano et al. 2002) or spatial
learning in the Morris water maze (Varvel and Lichtman
2002). The finding that cannabinoids have a different
effect on acquisition and extinction suggests that these
processes are mediated by different mechanisms. Indeed,
recent evidence indicates that the cellular changes that
occur in extinction of cocaine self-administration differ
from those during acquisition or consolidation (Sutton et
al. 2003).
Although considerable evidence suggests that D
9
-THC
interferes with working memory (Lichtman and Martin
1996; Molina-Holgado et al. 1995; Nakamura et al. 1991;
Mallet and Beninger 1998a; Heyser et al. 1993; Varvel et
al. 2001) and to a lesser extent with reference memory
(Brodkin and Moerschbaecher 1997; Nava et al. 2000;
Varvel et al. 2001), the doses (1.5–18 mg/kg) employed in
such studies are considerably higher than the dose em-
ployed in the present experiments (0.5 mg/kg). In fact, at
the doses that interfere with learning and memory, D
9
-
THC also produces a conditioned place aversion (Parker
and Gillies 1995; Mallet and Beninger 1998b; Robinson
et al. 2003), suggesting that nonspecific aversive prop-
erties of D
9
-THC may interfere with performance.
Recent evidence suggests that cannabinoids may pro-
mote relapse to drug self-administration. DeVries et al.
(2001) demonstrated that the potent cannabinoid agonist
HU-210 reinstates cocaine self-administration on its own,
although Schenk and Partridge (1999) reported that THC
does not promote cocaine self-administration. Since the
phenomenon of reinstatement may be related to relapse to
the previously reinforced behavior, one might expect that
a treatment that potentiates extinction should attenuate
rather than promote reinstatement. However, different
processes are called upon during extinction training than
during priming. During extinction, the rat is exposed to
the CS in the absence of the US and thus learns a CS–no
US new association; it has been argued that the endoge-
nous cannabinoid system plays a role in this type of
learning (Marsicano et al. 2002). However, during prim-
ing, re-exposure to the US is argued to retrieve the in-
centive value of the previously extinguished CS (DeWit
and Stewart 1981). It is therefore, possible that the en-
dogenous cannabinoid system plays a different role in
extinction and in priming. Additionally, the dose of HU-
210 (0.02 mg/kg) required to promote relapse to cocaine-
seeking behavior (DeVries et al. 2001) was considerably
higher than the equivalent dose of THC (0.5 mg/kg) that
promoted extinction here. HU-210 is 100–1000 times
more potent than THC (Martin et al. 1999); therefore, the
dose of 0.02 mg/kg HU-210 would be the equivalent of
that between 2 mg/kg and 20 mg/kg THC. It is clear that
cannabinoids produce biphasic effects on systems such as
anxiety and locomotor activity at low and high doses
(Iversen 2000); therefore, lower doses of cannabinoids
may actually prevent reinstatement.
It has been suggested that the place-preference learn-
ing is a measure of “wanting” or “craving” (Robinson and
Berridge 1993). That is, cues previously paired with the
neurochemical consequences of a rewarding drug come to
elicit approach responses. It has been proposed that
“craving” resulting in drug-seeking behaviors may be
elicited by drug-associated cues (O’Brien et al. 1992;
Stewart and Wise 1992). Extinction of drug-associated
cues (“cue re-exposure”) is a method of extinguishing
craving and relapse (O’Brien et al. 1992). Strategies that
would accelerate extinction would potentially reduce the
likelihood of relapse. Our results suggest that low doses
of the cannabinoids, D
9
-THC and CBD, may facilitate
such extinction processes.
Acknowledgements The research in Canada was supported by the
Natural Sciences and Engineering Research Council of Canada and
the Canadian Institutes of Health Research; the Research in Jer-
usalem was supported by the Israel Science Foundation and the
National Institute on Drug Abuse (US).
References
Bardo MT, Neisewander JL, Miller JS (1986) Repeated testing
attenuates conditioned place preference with cocaine. Psycho-
pharmacology 89:239–243
Brodkin J, Moerschbaecher JM (1997) SR 141716A antagonizes
the disruptive effects of cannabinoid ligands on learning in rats.
J Pharm Exp Ther 282:1526–1532
Calcagnetti DJ, Schechter MD (1993) Extinction of cocaine-in-
duced place approach in rats: a validation of the “biases”
conditioning procedure. Brain Res Bull 30:696–700
Cheer JF, Kendall DA, Marsden CA (2000) Cannabinoid receptors
and reward in the rat: a conditioned place preference study.
Psychopharm 151:25–30
Cunningham CL (1998) Drug conditioning and drug-seeking be-
havior. In: O’Donohue W (ed) Learning and Behavior Therapy.
Allyn & Bacon, Boston, MA, pp 518–44
Cunningham CL, Henderson CM, Bormann NM (1998) Extinction
of ethanol-induced conditioned place preference and condi-
tioned place aversion: effects of naloxone. Psychopharmacol-
ogy 139:62–70
DeVries TJ, Shaham Y, Homberg JR, Crombag C, Schuurman K,
Diegen J, Vanderschuren LJMJ, Schoffelmeer ANM (2001) A
cannabinoid mechanism in relapse to cocaine seeking. Nature
Med 7:1151–1154
DeWit H, Stewart J (1981) Reinstatement of cocaine-reinforced
responding in the rat. Psychopharm 75:134–174
Hajos N, Ledent C, Freund TF (2001) Novel cannabinoid-sensitive
receptor mediates inhibition of glutamatergic synaptic trans-
mission in the hippocampus. Neuroscience 106:1–4
Heyser CJ, Hampson RE, Deadwyler SA (1993) Effects of delta-9-
tetrahydrocannabinol on delayed match to sample performance
in rats: alterations in short-term memory associated with
changes in task specific firing of hippocampal cells. J Pharm
Exp Ther 264:294–307
Itzhak Y, Martin JL (2001) Cocaine-induced conditioned place
preference in mice: induction, extinction and reinstatement by
related psychostimulants. Neuropsychopharm 26:130–134
Iversen LL (2000) The Science of Marijuana. Oxford University
Press, NY
365
Kwiatkowska M, Parker LA, Burton P, Mechoulam R (2004) A
comparative analysis of the potential of cannabinoids and on-
dansetron to suppress cisplatin-induced emesis in the Suncus
murinus (house musk shrew). Psychopharmacology (in press)
Lichtman AH, Martin BR (1996) D-9-Tetrahydrocannabinol im-
pairs spatial memory through a cannabinoid receptor mecha-
nism. Psychopharmacology 126:125–131
Limebeer CL, Parker LA (1999) Delta-9-tetrahydrocannabinol in-
terferes with the establishment and the expression of condi-
tioned rejection reactions produced by cyclophosphamide: a rat
model of nausea. Neuroreport 10:3769–3772
Malfait AM, Gallily R, Sumariwalla PF, Malik AS, Andreakos E,
Mechoulam R, Feldmann M (2000) The nonpsychoactive
cannabis constituent cannabidiol is an oral anti-arthritic thera-
peutic in murine collagen-induced arthritis. Proc Nat Acad Sci
U S A 97:9561–9566
Mallet PE, Beninger RJ (1998a) The cannabinoid CB1 receptor
antagonist SR 141716A attenuates the memory impairment
produced by D
9
-tetrahydrocannabinol or anandamide. Psycho-
pharmacology 140:11–19
Mallet PE, Beninger RJ (1998b) D(9)-tetrahydrocannabinol, but not
the endogenous cannabinoid receptor ligand, anandamide,
produces conditioned place avoidance. Life Sci 62:2431–2439
Marsicano G, Wotjak CT, Azad SC, Bisogno T, Rammes G, Cascio
MG, Hermann H, Tang J, Hofmann C, Zieglgansberger W,
DiMarzo V, Lutz B (2002) The endogenous cannabinoid sys-
tem controls extinction of aversive memories. Nature 418:530–
534
Martin BR, Mechoulam R, Razdan RK (1999) Discovery and
characterization of endogenous cannabinoids. Life Sci 65:573–
95
Mechoulam R, Parker LA, Galilly R (2002) Cannabidiol: an
overview of some pharmacological aspects. J Clin Pharmacol
42:11S–19S
Meuller D, Perdikaris D, Stewart J (2000) Persistence and drug-
induced reinsttement of a morphine-induced conditioned place
preference. Behav Brain Res 136:389–397
Molina-Holgado F, Gonzalez MI, Leret ML (1995) Effects of D
9
-
tetrahydrocannabinol on short-term memory in the rat. Physiol
Behav 57:177–179
Nakamura EM, da Silva EA, Concilio GV, Wilkinson DA, Masur J
(1991) Reversible effects of acute and long-term administration
of D
9
-tetrahydrocannabinol (THC) on memory in the rat. Drug
Alcohol Depend 28:167–175
Nava F, Carta G, Battasi AM, Bessa GL (2000) D2 dopamine
receptors enable delta-9-tetrahydroicannabinol induced memo-
ry impairment and reduction of hippocampal extracellular
acetylcholine concentration. Br J Pharmacol 130:1201–1210
O’Brien CP, Childress AR, McLellan AT, Ehrman R (1992)
Classical conditioning in drug-dependent humans. Ann NY
Acad Sci 654:400–415
Parker LA, Gillies T (1995) THC-induced place and taste aversions
in Lewis and Sprague-Dawley rats. Behav Neurosci 109:71–78
Parker LA, McDonald RV (2000) Reinstatement of both a condi-
tioned place preference and a conditioned place aversion with
drug primes. Pharmacol Biochem Behav 66:559–61
Parker LA, Mechoulam R, Schievert C, Abbott LA (2002) Can-
nabidiol, a non-psychoactive component of cannabis and its
synthetic dimethylheptyl homolog suppress nausea in an ex-
perimental model with rats. NeuroReport 13:567–570
Parker LA, Mechoulam R, Fudge M, Schlievert C, Burton PD,
Abbott L (2003) Cannabinoid agonists interfere with and SR-
141716A potentiates lithium-induced conditioned rejection re-
actions: a rat measure of nausea. Psychopharmacology
166:156–162
Parker LA, Kwiatkowska M, Burton P, Mechoulam R (2004) Ef-
fects of cannabinoids on lithium-induced vomiting in the Sun-
cus murinus (house musk shrew). Psychopharmacology
171:156–161
Robinson TE, Berridge KC (1993) The neural basis of drug crav-
ing: an incentive-sensitization theory of addiction. Brain Res
Rev 18:247–291
Robinson L, Hinder L, Pertwee RG, Riedel G (2003) Effects of D
9
-
THC and WIN-55,212,2 on place preference in the water maze
in rats. Psychopharmacology 166:40–50
Schenk S, Partridge B (1999) Cocaine-seeking produced by ex-
perimenter-administered drug injections: dose-effect relation-
ships in rats. Psychopharmacology 147:285–290
Stewart J, Wise RA (1992) Reinstatement of heroin self-adminis-
tration habits: morphine prompts and naltrexone discourages
renewed responding after extinction. Psychopharmacology
108:79–84
Sutton MA, Schmidt EF, Choi KH, Schad CA, Whisler K, Sim-
mons D, Daranian DA, Moteggia LM, Neve RL, Self DW
(2003) Extinction-induced upregulation in AMPA receptors
reduces cocaine-seeking behaviour. Nature 421:70–75
Tzschentke TM, Schmidt WJ (1995) N-Methyl-d-aspartic-acid-re-
ceptor antagonists block morphine-induced conditioned place
preference in rats. Neurosci Lett 193:37–40
Varvel SA, Lichtman AH (2002) Evaluation of CB
1
receptor
knockout mice in the Morris water maze. J Pharm Exp Ther
301:915–924
Varvel SA, Hamm RJ, Martin BR, Lichtman AH (2001) Differ-
ential effects of D
9
-THC on spatial reference and working
memory in mice. Psychopharmacology 157:142–150
366
