| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Department of Neuropsychopharmacology and Hospital Pharmacy, Nagoya University Graduate School of Medicine, Nagoya, Japan; and
Laboratory of Neuropsychopharmacology, Kanazawa University Graduate School of Natural Science and Technology, Kanazawa, Japan
1Correspondence: Department of Neuropsychopharmacology and Hospital Pharmacy, Nagoya University Graduate School of Medicine, Showa-ku, Nagoya 466-8560, Japan. E-mail: tnabeshi{at}med.nagoya-u.ac.jp
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: GDNF mutant mice • METH self-administration • relapse
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2
3
4
5
6
7)
. In animal models, vulnerability to self-administration and reinstatement in the taking of different addictive substances has been suggested to share common genetic determinants (8)
. By using a model of drug self-administration in mutant mice, several specific genes or proteins have been identified as potentially involved in the development of drug dependence (9
10
11
12
13
14)
. However, a good model of relapse in mutant mice has yet to be established. Thus, few lines of direct evidence have been obtained for an association between specific genes and vulnerability to relapse of drug-seeking behavior, which is a major challenge in the clinical treatment of addiction (15
, 16)
.
Glial cell line-derived neurotrophic factor (GDNF) was originally purified from a rat glioma cell-line supernatant as a trophic factor for embryonic midbrain dopamine neurons (17)
. As a potential therapeutic agent for the treatment of Parkinson’s disease, GDNF has been widely tested (18
, 19)
. It is well established that dopaminergic transmission in the cortico-limbic system is crucial for the development of drug dependence/addiction (20
21
22
23
24
25)
. Given that GDNF is considered an important modulator for dopaminergic neuronal function (17
, 26)
, it is reasonable to postulate that GDNF may be involved in drug addiction. Although direct evidence of a clinical association between GDNF and drug dependence/addiction has yet to be obtained, GDNF has been identified in the development of drug dependence in animal models (27
28
29
30)
. Manipulations that modulate GDNF content in the brain affected cocaine-induced conditioned place preference and cocaine or ethanol self-administration in rats (27
28
29
30)
. GDNF (+/–) heterozygous knockout mice [GDNF (+/–) mice] showed greater morphine, cocaine, and methamphetamine (METH) conditioned place preference (27
, 31)
. However, the role of GDNF in vulnerability to relapse of drug-seeking behavior remains unclear. Using animal models of drug self-administration and relapsing behavior recently established in our laboratory (32
, 33)
, which represent drug-taking and relapse of drug-seeking behavior in addicts (34)
, we provided evidence that a partial loss of GDNF expression not only facilitated the acquisition of METH self-administration, resulted in an upward shift in the dose-response curve, and increased motivation to take METH, but also led to increased vulnerability to METH-primed reinstatement and enduring cue-induced reinstatement of extinguished drug-seeking behavior.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. GDNF (–/–) homozygous knockout mice die shortly after birth, but GDNF (+/–) mice are viable. After genomic DNA was purified from a 0.5–1.0 cm segment of tail, the mice were genotyped by PCR utilizing three sets of primers selective of the neomycin cassette: primer 1 (5'-GAC TGG CTT GGT TCT TTG CAT GCA TCC –3'); primer 2 (5'-ACC AAA GAA CGG AGC CGG TTG GCG C-3'), and primer 3 (5'-GAG AGG AAT CGG CAG GCT GCA GCT G-3'). To characterize the influence of the GDNF expression on the operant behavior, a colony of GDNF (+/–) mice was employed in the present study. In this colony, the levels of GDNF expression in corticolimbic areas of the brain are reduced to 54–66% of those in wild-type littermates, at the age of 8 wk (Supplemental Fig. 1). Wild-type littermates were used as a control of the GDNF (+/–) mice. GDNF (+/–) and wild-type mice were bred locally in the Laboratory Animal Center, Nagoya University Graduate School of Medicine in Japan. Male GDNF (+/–) and wild-type mice were 8-wk-old and weighed 25–30 g at the beginning of the experiments. All mice were kept in a regulated environment (23 ± 0.5°C; 50 ± 0.5% humidity) with a reversed 12-h light/dark cycle (lights on at 9:00 AM). Both water and food were available ad libitum throughout the experiments unless otherwise noted. All procedures followed the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals and were approved by the Nagoya University School of Medicine Animal Care and Use Committee.
METH hydrochloride (Dainippon Pharmaceutical Ltd, Osaka, Japan) was dissolved in sterile saline and self-administered at a dose of 0.1 mg/kg/infusion over 5 s (infusion volume=2.1 µl). The unit dose for METH self-administration is based on our previous report (32)
.
Food-reinforced operant behavior and reinstatement of food-seeking behavior
Food-reinforced operant behavior and motivation
Food-reinforced operant behavior and motivation were tested in standard mouse operant conditioning chambers as described previously (32)
. Briefly, the chamber was equipped with two nose-poke sensors (ENV-313M, Med Associates) in two holes, two cue-lamps in and above each hole, and a food pellet dispenser (ENV-203–20, Med Associates, Georgia, VT, USA) connected to a rectangular opening (2.25 cm x 2.25 cm) between the two holes. The bottom of the opening was 5 mm above the chamber floor and was equidistant from the holes. A house light was located at the top of the chamber opposite the holes. During the tests for food-reinforced operant behavior and motivation, one hole was defined as active, and the other, as inactive. Nose-poke responses in the active hole resulted in the delivery of a single food pellet (dustless precision pellets 20 mg, A Holton Industries Co., Frenchtown, NJ, USA) to the opening by the dispenser (ENV302M, Med Associates) and inactivation of the cue-lamp and hole-lamp for 5 s followed by a 5 s timeout period. Nose-poke responses in the active hole during the timeout period and in the inactive hole had no programmed consequences but were recorded by the software MED-PC for Windows (Med Associates).
Naive GDNF (+/–) and wild-type mice (n=7 for each genotype) were deprived of food for 20 h (water remained available ad libitum throughout the experiments). From the next day, both genotypes were daily subjected to nose-poke responding for food pellets in the standard operant chambers as mentioned above. During this phase, the mice were returned to their home cages and given unlimited amounts of food for 2 h immediately after each session of nose-poke responding for food pellets. The daily 3 h sessions of food-reinforced nose-poke responding in GDNF (+/–) and wild-type mice were initially performed under a fixed ratio (FR) 1 schedule. Once the mice showed stable nose-poke responding for food pellets (deviations of <15% of the mean of active responses in 3 consecutive training sessions), the reinforcement schedule was changed to FR2 until the same criterion as above was achieved. The same groups of mice were then subjected to nose-poke responding for food pellets under a progressive ratio (PR) schedule, in which the number of active nose-poke responses required to obtain a single food pellet escalates according to the following series: 1, 2, 4, 6, 9, 12, 15, 20, 25, 32, 40, 50, 62, 77, 95, 118, 145,178, 219, 268, 328, 402, 492, 603, 737, etc. (36)
. This "breaking point," expressed as the final ratio (the number of active nose-poke responses needed to earn the last single food pellet), reflects the intensity of motivation for earning food pellets. Each session under the PR schedule lasted for 5 h or until mice failed to respond within 1 h. After 2–4 daily sessions, GDNF (+/–) and wild-type mice demonstrated stable active nose-poke responses for food pellets (deviations of <15% of the mean of total active responses in 2 consecutive sessions).
Extinction and reinstatement of food-seeking behavior
During this phase, both food and water were available ad libitum in the home cages. After the test for motivation to take food pellets under the PR schedule, the same groups of GDNF (+/–) and wild-type mice were then subjected to 6–10 daily 3 h sessions of extinction. Throughout the extinction session, the house light was on. The food-associated cue- and hole-lamps, and the system that delivers food pellets were turned off. Therefore, the nose-poke responses into the previously active hole resulted in neither the delivery of food pellets nor food-associated cues (cue- and hole-lamps). Once the mice met the criterion of extinction (<15 active responses or 25% of active responses in the stable phase of self-administration in 2 consecutive sessions), they were subjected to a 3 h session of the food-priming reinstatement test under the same conditions as in the extinction sessions (without either food-associated cues or the delivery of food pellets). As priming of food pellets, 12 food pellets were placed into the rectangular opening between the 2 holes before the food-priming reinstatement test. Nose-poke responses in the previously active or inactive hole were counted as active and inactive, respectively.
After the food-primed reinstatement test, the same groups of mice were subjected to 3–6 daily 3 h sessions of extinction immediately, and 3 months, after withdrawal from responding for food pellets. Once the mice met the extinction criterion as mentioned above, they were subjected to daily 3 h sessions of food-associated cue-induced reinstatement tests immediately, and 3 months, after the withdrawal. The food-associated cue-induced reinstatement tests were performed under the same conditions as the food-reinforced operant behavioral test under the FR2 schedule, except that there was no delivery of food pellets after the nose-poke responses in a previously active hole. Nose-poke responses in the previously active or inactive hole were counted as active and inactive, respectively.
Surgery and apparatus for METH self-administration
Catheter implantation
New groups of naive GDNF (+/–) and wild-type mice were deprived of food for 20 h (water remained available ad libitum) and then trained to make nose-poke responses under the FR 1 schedule for food pellets in the operant chambers as mentioned above, except that both nose-poke holes were defined as active. Once a mouse had earned 30 food pellets, the session for nose-poke training ended (for 2–8 h, no difference between GDNF (+/–) and wild-type mice). After the training session, the mice were returned to their home cages, where both food and water were available ad libitum throughout the subsequent experiments. Two days later, the mice were anesthetized with pentobarbital sodium (50 mg/kg ip). Indwelling catheters were constructed of microsilicone tubing (inner diameter, 0.50 mm; outer diameter, 0.7 mm; IMG, Imamura Co., Ltd., Tokyo, Japan) and polyethylene tubing (inner diameter, 0.50 mm; outer diameter, 0.8 mm). Incisions were made on the skin of the head and ventral neck, and the right jugular vein was externalized. The end of the catheter was inserted into the jugular vein via a small incision and was secured to the vein and surrounding tissue with silk sutures. The exit port of the catheter passed subcutaneously to the top of the skull where it was attached to a modified 24-gauge cannula, which was secured to the mouse’s skull with quick self-curing acrylic resin (Shofu Inc., Tokyo, Japan). To extend catheter patency, the catheters were flushed immediately after surgery, and in the morning and evening of the following days, with 0.03 ml of an antibiotic solution of cefmetazole sodium (20.0 mg/ml; Sankyo Co., Ltd., Tokyo, Japan) dissolved in heparinized saline (70 U/ml; Leo Pharmaceutical Products, Ltd., Tokyo, Japan). The patency of the catheter was usually confirmed once a week before operant behavior tests by infusion of a pentobarbital sodium solution (6.0 mg/ml, 0.15 ml/mouse) into the jugular vein. If the mice could not be knocked down within 5 s, the corresponding data were excluded from the statistical analysis.
Apparatus for METH self-administration
METH self-administration was conducted in standard mouse operant conditioning chambers (ENV-307A, Med Associates) located within ventilated sound attenuation cubicles as described previously (32)
. Briefly, the chambers were equipped with nose-poke sensors (ENV-313M, Med Associates) in two holes located on one side of the chamber 1.0 cm above the floor, cue- and hole-lamps located, respectively, above and in each hole, and a red house light located on the top of the chamber opposite the holes. During the self-administration, one hole was defined as active, and the other, as inactive. Nose-poke responses in the active hole resulted in activation of the infusion pump (PHM-100, Med Associates) and inactivation of the cue-lamp and hole-lamp. Nose-poke responses in the inactive hole, and in the active hole during the timeout period, had no programmed consequences but were recorded. The components of the infusion line were connected to each other from the injector to the exit port of the mouse’s catheter by joint FEP tubing (inner diameter=0.25 mm; outer diameter=0.55 mm; Eicom Co., Ltd., Japan), which was encased in steel spring leashes (Instech, Plymouth Meeting, PA). Swivels were suspended above the chamber. One pump/syringe set was used for each chamber located inside of the cubicle. The infusion pump/syringe set was outside of the chambers but inside of the cubicles.
METH self-administration and reinstatement of METH-seeking behavior
Outline
After recovering from the surgery to implant the catheter, GDNF (+/–) (n=28) and wild-type mice (n=26) were subjected to METH self-administration, extinction, and reinstatement of extinguished METH-seeking behavior according to the workflow shown in Table 1
. During METH self-administration, nose-poke responses in the active hole resulted in an infusion of METH at a dose of 0.1 mg/kg/infusion over 5 s (infusion volume=2.1 µl) followed by a 5 s timeout period. Nose-poke responses in the inactive hole, and in the active hole during the timeout period, had no programmed consequences but were recorded.
|
Acquisition of METH self-administration under an FR schedule
METH self-administration was initially under the FR1 schedule. Once the mice could make a minimum of 60% nose-poke responses in the active hole and received no >10 infusions of METH <2 consecutive sessions (at least for 4 sessions), the METH reinforcement schedule was changed to FR2. Under the FR2 schedule, the mice gradually acquired stable METH self-administration behavior (deviations of <15% of the mean of active responses in 3 consecutive training sessions). After acquiring stable self-administration behavior, GDNF (+/–) and wild-type mice were each counterbalance-separated into two subgroups. One subgroup of GDNF (+/–) and wild-type mice were subjected to the test for dose responses. The others were subjected to METH self-administration under the PR schedule.
Dose responses for METH self-administration under an FR2 schedule
After acquiring stable self-administration behavior, one subgroup of GDNF (+/–) and wild-type mice were subjected to METH self-administration under the FR2 schedule of reinforcement in the dose range 0.003–0.1 mg/kg/infusion from the higher to lower dose. Each mouse was subjected to two to four daily 3 h sessions of METH self-administration at one dose until it demonstrated stable active nose-poke responses (deviations of <15% of the mean of total active responses in 2 consecutive sessions).
Motivation for METH self-administration under a PR schedule
After stable self-administration behavior was acquired, the other subgroup of GDNF (+/–) and wild-type mice were subjected to METH self-administration under the PR schedule. The "breaking point" is defined as the final ratio (the number of active nose-poke responses needed to earn the last infusion of METH) and reflects the intensity of motivation for taking the drug tested. Each session lasted for 5 h or until mice failed to respond within 1 h. Each mouse was subjected to two to five sessions of METH self-administration. Both genotypes of mice demonstrated stable active nose-poke responses for METH infusion (as described in the section of dose response) during the two to five sessions.
Extinction
After the self-administration under the FR or PR schedule, the two subgroups of GDNF (+/–) and wild-type mice were subjected to METH (0.1 mg/kg/infusion) self-administration under the FR2 schedule until both genotypes showed stable (as described above) active nose-poke responses once again, and took approximately the same amount of METH. The mice were then subjected to 6–10 daily 3 h sessions of extinction before the METH-primed reinstatement test or 3–6 daily 3 h sessions of extinction before the cue-induced reinstatement test until they met the extinction criterion (<15 active responses or 25% of active responses in the stable phase of self-administration in 2 consecutive sessions). Throughout the extinction session, the house light was on. The METH-associated cue- and hole-lamps, and the pump for METH infusion, were turned off. Therefore, nose-poke responses into the previously active hole resulted in neither an infusion of METH nor METH-associated cues (cue- and hole-lamps, and pump noise for METH infusion).
METH-primed reinstatement
Once the extinction criterion was met, the GDNF (+/–) and wild-type mice were firstly subjected to a 3 h session of the operant test 30 min after the injection (ip) of saline as a control for the METH-primed reinstatement. From the next day, the mice were consecutively subjected to METH-primed reinstatement tests 30 min after the intraperitoneal injection with increasing doses of METH (0.2, 0.4, 1.0, 1.5, or 3.0 mg/kg, each dose for 1 daily 3 h session). The METH-primed reinstatement tests were conducted under the same conditions as in the extinction sessions in which neither METH infusions nor METH-associated cues were available after nose-poke responses into a previously active hole. Nose-poke responses in the previously active or inactive hole were counted as active and inactive, respectively.
Cue-induced reinstatement
Once the extinction criterion was met, the same groups of mice were subjected to the cue-induced reinstatement tests immediately, 3 months, and 6 months after withdrawal from METH self-administration. The cue-induced reinstatement tests were conducted under the same conditions as the METH self-administration under the FR2 schedule, except that METH was unavailable throughout the testing session. Nose-poke responses in the previously active or inactive hole were counted as active and inactive, respectively.
Data analysis
All data are ± SE. A one- or two-way ANOVA with (or without) repeated measures was performed for the difference in locomotor activity and nose-poke responses between the two genotypes of mice during the self-administration training, dose-response function, and METH-primed and cue-induced reinstatement of drug-seeking behavior, followed post hoc by the Bonferroni/Dunn test. The Mann-Whitney test was used to analyze the breaking points under the PR schedule, whereas Student’s t test was used to analyze the other two sets of data. In all cases, a significant difference was set at P < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A). Also, there was no significant difference in the number of training sessions to acquire stable operant behavior between GDNF (+/–) and wild-type mice (Fig. 1B
). Furthermore, GDNF (+/–) and wild-type mice showed similar breaking points under the PR schedule (Fig. 1C
). These findings indicate that a partial loss of GDNF expression does not affect food-reinforced operant behavior and motivation in mice.
|
Reinstatement of food-seeking behavior in GDNF (+/–) and wild-type mice
To evaluate the reinstatement of food-seeking behavior in the mutant animals, the same groups of GDNF (+/–) and wild-type mice were subjected to extinction training after the tests for the food-reinforced operant behavior. After similar daily 3 h sessions of extinction training, GDNF (+/–) and wild-type mice achieved the extinction criterion (Fig. 2
A; F(3,48)=27.83; P<0.001). However, the priming of food pellets failed to reinstate food-seeking behavior in either genotype (Fig. 2A
). The same groups of mice were then subjected to extinction training once again. Once the extinction criterion was met, the food-associated cue-induced reinstatement tests were conducted immediately and 3 months after the end of tests for the food-reinforced operant behavior. Food-associated cues reliably triggered reinstatement of food-seeking behavior in both GDNF (+/–) and wild-type mice immediately after withdrawal (Fig. 2B
; F(1,24)=33.44; P<0.001). However, no significant difference in cue-induced reinstatement behavior was observed between GDNF (+/–) and wild-type mice (Fig. 2B
). Importantly, the food cue-induced reinstatement of food-seeking behavior disappeared within the period of a 3 month withdrawal in both genotypes of animals (Fig. 2B
). These findings suggest that reinstatement of food-seeking behavior in both genotypes of animals is weak or transient and that the partial loss of GDNF expression does not affect extinction behavior, reinstatement of food-seeking behavior, or duration of food-associated cue-induced reinstatement behavior in mice.
|
Facilitated acquisition of METH self-administration behavior in GDNF (+/–) mice
To investigate whether a partial loss of GDNF expression affects drug self-administration behavior in animals, separate groups of GDNF (+/–) and wild-type mice were subjected to METH self-administration training. GDNF (+/–) mice took less time than wild-type littermates to acquire stable METH self-administration behavior (Fig. 3
A, P<0.01). However, there was no significant difference in total METH intake during the period of METH self-administration training between wild-type (30.7±2.0 mg/kg) and GDNF (+/–) (28.5±1.5 mg/kg) mice (Fig. 3B
). In the early phase of METH self-administration under the FR1 schedule (Fig. 3C
, day 1–4), neither genotype could discriminate active (METH-associated) from inactive (without METH infusion) nose-poke responses. Accordingly, there was no significant difference in active nose-poke responses for METH self-administration between GDNF (+/–) and wild-type mice. The mice gradually demonstrated stable METH self-administration behavior under the FR2 schedule. Accordingly, they could discriminate active from inactive nose-poke responses to METH reinforcement (Fig. 3C
, the last day 1–6 (L1-L6), P<0.001). However, there was no significant difference in active nose-poke responses for METH-taking between wild-type and GDNF (+/–) mice. These findings indicate that GDNF (+/–) mice are capable of METH self-administration.
|
Upward shifted dose responses and increased motivation to take METH in GDNF (+/–) mice
GDNF (+/–) and wild-type mice showed significantly different dose responses to self-administer METH (Fig. 4
A; F(1,55)=12.43, P<0.001). In the dose range of 0.01–0.03 mg/kg/infusion, the number of active nose-poke responses for METH-taking was significantly higher in GDNF (+/–) mice than in wild-type littermates (P<0.05 and 0.001). There was no significant difference in active nose-poke responses to self-administer METH at 0.1 and 0.003 mg/kg/infusion between GDNF (+/–) and wild-type mice. When saline was substituted for METH, no significant difference was observed in self-administration behavior between GDNF (+/–) and wild-type mice. The upward shift of the dose-response function for METH self-administration in GDNF (+/–) mice suggests that the partial loss of GDNF expression may increase sensitivity to METH self-administration in mutant animals during the maintenance phase.
|
To further support this idea, the other subgroups of GDNF (+/–) and wild-type mice were subjected to METH (0.1 mg/kg/infusion) self-administration under the PR schedule. GDNF (+/–) mice demonstrated a significantly increased breaking point compared with wild-type littermates (Fig. 4B
; P<0.01), suggesting that the partial loss of GDNF expression leads to greater motivation to take METH. Representative curves for the two genotypes of animals are illustrated in Fig. 4C
.
Increased vulnerability to METH-primed reinstatement of drug-seeking behavior in GDNF (+/–) mice
We further investigated the performance of GDNF (+/–) mice in the reinstatement of extinguished METH-seeking behavior. After once again acquiring stable METH self-administration behavior, during which wild-type and GDNF (+/–) mice had taken similar amounts of METH (43.9±3.2 and 50.1±4.1 mg/kg, respectively), the two genotypes were exposed to extinction training for 6–10 daily 3 h sessions. There was no significant difference between GDNF (+/–) and wild-type mice in the number of nose-poke responses into previously active holes [METH-associated; Fig. 5
A, session 1–3 and the last session 1–3 (L1-L3)] or the number of extinction training sessions to achieve the extinction criterion (data not shown). During the last two extinction sessions, neither group of animals could discriminate active (previously associated with METH self-administration) from inactive (previously without METH self-administration) nose-poke responses, similar to the early stage of METH self-administration. These findings suggested that in GDNF (+/–) and wild-type mice, purposely active nose-poke responses acquired during METH self-administration had been extinguished.
|
Once the mice met the extinction criterion (Fig. 5A
), a drug-primed reinstatement test was carried out 30 min after treatment with either saline or a different dose of METH. Wild-type and GDNF (+/–) mice showed different active nose-poke responses (Fig. 5B
; F(1, 77)=12.72; P<0.001), although there was no significant difference in inactive nose-poke responses between GDNF (+/–) and wild-type mice. In wild-type littermates, both lower (0.2 and 0.4 mg/kg) and higher (1.5 and 3.0 mg/kg) doses of METH-priming failed to reinstate drug-seeking behavior. However, a moderate dose of METH (1.0 mg/kg) reliably triggered the reinstatement behavior. In contrast, both lower and moderate doses of METH (0.4 mg/kg and 1.0 mg/kg) reliably triggered the reinstatement of extinguished drug-seeking behavior in GDNF (+/–) mice (P<0.05), although higher doses of METH did not evoke the reinstatement behavior. The leftward shift of the dose-response curve for METH-primed reinstatement behavior suggests that the partial loss of GDNF expression may affect vulnerability to the reinstatement of extinguished METH-seeking behavior in mice.
Prolonged persistence of cue-induced reinstatement of drug-seeking behavior in GDNF (+/–) mice
To investigate the enduring vulnerability to cue-induced reinstatement of METH-seeking behavior in GDNF (+/–) mice, the same groups of mice were subjected to three to six extinction sessions, followed by three cue-induced reinstatement tests, which were conducted immediately, 3 months, and 6 months after METH withdrawal. During the extinction sessions, neither genotype showed any significant difference in nose-poke responses or in the number of training sessions needed to achieve the extinction criterion (Fig. 5A
). Once the extinction criterion was achieved, the mice were subjected to cue-induced reinstatement tests. GDNF (+/–) and wild-type mice initially demonstrated a cue-triggered reinstatement of METH-seeking behavior (Fig. 6
; 1st test; P<0.001). Importantly, there was a clear tendency for GDNF (+/–) mice to show more active nose-poke responses than wild-type littermates when exposed to the METH-associated cues (Fig. 6
; F(1, 26)=3.99; P=0.056). With a prolonged withdrawal, GDNF (+/–) and wild-type mice showed significantly different responses (Fig. 6
, F(1, 150)=26.1; P<0.001). In wild-type littermates, the cue-induced reinstatement behavior was still observed 3 months after the withdrawal (Fig. 6
; 2nd test; P<0.05) but disappeared after a 6 month withdrawal (3rd test). In contrast, GDNF (+/–) mice maintained the cue-induced reinstatement behavior even after a 6-month withdrawal (Fig. 6
; 3rd test; P<0.01). In addition, there was no significant difference in inactive nose-poke responses during any of the tests for cue-induced reinstatement behavior between GDNF (+/–) and wild-type mice.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Acquisition and maintenance of METH self-administration behavior
It has been documented that GDNF (+/–) mice demonstrate increased morphine, cocaine, and METH-conditioned place preference (27
, 31)
. In the present study, GDNF (+/–) mice took a shorter period of time to acquire stable METH self-administration behavior, with an upward shift of dose responses to METH-taking, compared with wild-type littermates. Furthermore, GDNF (+/–) mice showed greater motivation (breaking point) for METH self-administration. In contrast, there was no significant difference in food-reinforced operant behavior (including acquisition and maintenance) and motivation between GDNF (+/–) and wild-type mice. Thus, GDNF (+/–) mice may represent a phenotype susceptible to the rewarding and reinforcing effects of addictive drugs although the precise molecular mechanisms underlying this phenomenon remain unclear. It has been documented that the extracellular dopamine concentration, FosB levels, and deltaFosB expression are elevated in the nucleus accumbens and striatum of GDNF (+/–) mice as compared with wild-type littermates (37)
. In addition, it is well established that GDNF is an important modulator for dopaminergic neuronal function (e.g., refs. 17
, 26
). Thus, it seems reasonable to postulate that the reduced GDNF content causes an increase in the extracellular dopamine concentration, FosB levels, and deltaFosB expression, leading to greater morphine-, cocaine-, or METH-conditioned place preference (27
, 31
, 37)
, facilitated acquisition of METH self-administration behavior, upward shifted dose responses, and enhanced motivation to take METH (in the present study).
It seems unlikely that the differences in acquisition of self-administration behavior, dose responses, and motivation to take METH between GDNF (+/–) and wild-type mice are due to nonspecific physiological adaptations or compensatory effects during the development of mutant animals. Firstly, in our colony of GDNF (+/–) mice, the levels of GDNF expression in cortico-limbic areas of the brain were reduced 34–46% (see Supplemental Fig. 1). This is consistent with previous reports that striatal GDNF contents are reduced in GDNF (+/–) mice (37
, 38)
. Secondly, both cocaine-conditioned place preference and cocaine- or ethanol-reinforced self-administration are reduced by an increased level of GDNF in the animal brain (27
, 28
, 30)
. Thirdly, cocaine-conditioned place preference and ethanol self-administration are potentiated by a decrease in the amount of GDNF in the brain through local delivery of anti-GDNF neutralizing antibodies (27
, 30)
. A previous report has demonstrated an impairment of water-maze learning in GDNF (+/–) mice (39)
. In the present study, there was no significant difference in the acquisition of, retention of, and motivation for food-reinforced operant behavior between GDNF (+/–) and wild-type mice. Thus, it seems difficult to explain the alterations in METH-reinforced self-administration behavior and motivation, based on the learning and memory deficits in GDNF (+/–) mice. It has been suggested that responses to novelty in animals are associated with the propensity for drug self-administration (40)
. Given that there was no significant difference in exploratory behavior in the open field test, locomotor activities during habituation to the testing box environment, and locomotor responses to METH between GDNF (+/–) and wild-type mice (see Supplemental Figs. 2, 3), it is unlikely that differences in novelty responses or locomotor activity during METH self-administration contribute to alterations in METH-reinforced self-administration and motivation between GDNF (+/–) and wild-type mice.
Vulnerability to METH-primed reinstatement of drug-seeking behavior in GDNF (+/–) mice
No significant difference was observed in active or inactive nose-poke responses (Fig. 5A
), and the number of extinction training sessions needed to achieve the extinction criterion (data not shown) between GDNF (+/–) and wild-type mice during the period of extinction training. This phenomenon indicates that a primary reinforcer (METH) or secondary reinforcer (METH-associated cues) may be necessary for the effects of GDNF on the development of METH-reinforced self-administration behavior. Indeed, GDNF (+/–) mutant mice showed a leftward and upward shifted dose-response curve for reinstatement of extinguished drug-seeking behavior after a priming injection of the primary reinforcer METH, whereas neither genotype showed reinstatement of food-seeking behavior after priming with food pellets. It has been shown that drug-primed reinstatement and drug self-administration share similar anatomical neural substrates (cortico-limbic system) and neural transmission (dopamine) in the brain (22
, 41)
. Thus, possible mechanisms underlying enhanced METH-reinforced self-administration and motivation may contribute to the vulnerability to METH-primed reinstatement behavior in GDNF (+/–) mice. It seems unlikely that the vulnerability to METH-primed reinstatement behavior is due to the different experiences of METH self-administration. First, wild-type littermates took longer to acquire stable METH self-administration behavior than GDNF (+/–) mice (Fig. 5A
), whereas there was no significant difference in active nose-poke responses for METH during the early and stable phases of METH self-administration between GDNF (+/–) and wild-type mice (Fig. 3C
). Second, there was no significant difference in total METH intake before the test for METH-primed reinstatement between wild-type and GDNF (+/–) mice (43.9±3.2 and 50.1±4.1 mg/kg, respectively). In addition, there was no significant difference in novelty seeking behavior and in METH-stimulated hyperlocomotion and locomotor sensitization between GDNF (+/–) and wild-type mice (see Supplemental Figs. 2, 3). This phenomenon is consistent with a previous report that the effects of acute and repeated treatment of cocaine on locomotor activity are similar between GDNF (+/–) and wild-type mice (37)
. For similar reasons to those mentioned above, it is unlikely that vulnerability to METH-primed reinstatement behavior in GDNF (+/–) mice reflects nonspecific increases in motor activity or novelty responses.
Prolonged cue-induced reinstatement of METH-seeking behavior in GDNF (+/–) mice
In the present study, GDNF (+/–) mice demonstrated a stronger and more persistent cue-induced reinstatement of extinguished METH-seeking behavior than did wild-type littermates. Moreover, cue-induced reinstatement behavior in GDNF (+/–) mice could be observed even after a 6 month withdrawal when the cue-induced reinstatement in wild-type littermates had disappeared. In contrast, there was no significant difference in the transient cue-induced reinstatement of food-seeking behavior between GDNF (+/–) and wild-type mice (Fig. 2B
). The more severe and persistent cue-induced reinstatement of extinguished METH-seeking behavior in GDNF (+/–) mice suggests that the partial loss of GDNF expression may lead to vulnerability to and persistence of cue-induced reinstatement of drug-seeking behavior, without affecting food-seeking behavior. It has been reported that striatal synaptic plasticity is crucial for the formation of an addictive habit or cue-controlled drug-seeking behavior (42
43
44)
and that deltaFosB, once expressed, persists in the brain for a relatively long period of time in the absence of further drug exposure and acts as a sustained molecular switch for addiction (45
46
47)
. Thus, the enduring vulnerability to cue-induced reinstatement may be attributable to higher levels of deltaFosB in striatal brain areas of GDNF (+/–) mice (37)
. In addition, the occurrence of drug-seeking behavior after a delay of several weeks in rats (48)
seems inconsistent with our finding that cue-induced reinstatement of drug-seeking behavior was reduced with the time of withdrawal from METH self-administration in GDNF (+/–) and wild-type mice. This discrepancy may be because the cue-induced reinstatement behavior in the present study was examined after repeated cycles of extinction training (a with-subjects design), since repeated extinction training decreases the propensity for a relapse of extinguished drug-seeking behavior (49)
.
The present series of experiments demonstrated an association between specific genes or proteins, for example, the expression of the GDNF gene, and vulnerability to relapse of drug-seeking behavior, suggesting that GDNF may be critically involved in the acquisition and maintenance of METH self-administration, vulnerability to METH-primed reinstatement, and persistent cue-induced reinstatement of extinguished drug-seeking behavior. In line with previous reports (27
28
29
30)
and our present findings, GDNF may be a potential target of therapeutic agents not only for the prevention of drug dependence but also for the control of relapse of drug-seeking behavior.
|
ACKNOWLEDGMENTS |
|---|
Received for publication November 29, 2006. Accepted for publication February 1, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
¿ protects against methamphetamine-induced rewarding effects and sensitization. (In press)Biol. Psychiatry This article has been cited by other articles:
|
|
 |
|
  S. Carnicella, V. Kharazia, J. Jeanblanc, P. H. Janak, and D. Ron GDNF is a fast-acting potent inhibitor of alcohol consumption and relapse PNAS, June 10, 2008; 105(23): 8114 - 8119. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||
P < 0.01 vs. first session of extinction in same genotype. *P < 0.05 vs. saline treatment in same genotype. #P < 0.05 vs. wild-type littermates during same reinstatement test. Primed-RLP, METH-primed reinstatement; 1st Cue (S+), first test for METH-associated cue-induced reinstatement immediately after withdrawal; 2nd Cue (S+), second test for METH-associated cue-induced reinstatement 3 months after withdrawal; 3rd Cue (S+), third test for METH-associated cue-induced reinstatement 6 months after withdrawal.