Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

The characterization of factors that contribute to the onset of compulsive drug-seeking behavior is crucial to the formulation of effective treatment strategies for drug abuse. It is estimated that out of the total population of individuals who initially try cocaine, only 10% to 15% will eventually become addicted (Gawin, 1991). Others are able to use cocaine casually over extended periods of time without ever progressing to compulsive use (Siegel, 1984). Although there are no reliable predictors of individual vulnerability to cocaine addiction in humans (Gawin, 1991), it is likely that environmental variables are critical determinants. Exposure to stressful environmental stimuli has been reported to enhance individual sensitivity to the behavioral (Antelman et al., 1980; MacLennan and Maier, 1983) and neurochemical (Sorg and Kalivas, 1991) effects of psychostimulants in rats. These observations have been extended to include the effects of stressors on the reinforcing properties of psychostimulants. It has been demonstrated that isolation housing (Schenk et al., 1987), repeated tailpinch (Piazza et al., 1990), social stress (Miczek and Mutschler, 1996), daily exposure to uncontrollable electric footshock (Goeders and Guerin, 1994), and the "stress" of witnessing another rat being subjected to shock (Ramsey and Van Ree, 1993) all facilitate the acquisition of intravenous cocaine or amphetamine self-administration in rats. In consideration of these findings, it can be hypothesized that the interaction between stressors and cocaine occurs as a result of the activation of one or more common pharmacological effector systems.

The HPA axis is commonly activated by cocaine and stressors. Acute cocaine administration stimulates the release of -endorphin, ACTH and corticosterone in rats (Moldow and Fischman, 1987). Similar to the stressor-induced secretion of ACTH and corticosterone (Rivier and Plotsky, 1986), this response appears to be dependent on CRF (Rivier and Vale, 1987). Cocaine also stimulates the release of cortisol in humans (Baumann et al., 1995), apparently through effects on the CRF-driven pulsatile secretion of ACTH (Teoh et al., 1994). Individual variability in sensitivity to many of the behavioral and neurochemical effects of psychostimulants appears to be directly related to differences in corticosterone before and at the time of drug exposure (see Piazza and Le Moal, 1996, for review). Rats classified as HR based on their locomotor responses to a novel environment display a higher locomotor response to an amphetamine challenge and are more predisposed to develop amphetamine self-administration when compared with LR rats (Piazza et al., 1989). These differences between HR and LR rats are eliminated by the administration of corticosterone (Piazza et al., 1991). It has also been demonstrated that ADX rats will not develop cocaine self-administration over a range of doses (Goeders and Guerin, 1996a; Deroche et al., 1997), while drugs that block the synthesis of corticosterone, such as metyrapone (Goeders and Guerin, 1996a) and ketoconazole (Goeders and Guerin, 1997), attenuate cocaine self-administration once it has been established. Thus, it is apparent that the stimulation of corticosterone secretion by cocaine is important, if not necessary, for its establishment as a positive reinforcer.

It is also likely that corticosterone provides a substrate through which exposure to stressors can influence individual predisposition to self-administer cocaine. A highly significant positive correlation has been reported between the facilitation of cocaine self-administration by uncontrollable electric footshock and the effects of footshock on plasma corticosterone (Goeders and Guerin, 1996b). In these experiments, plasma corticosterone was always elevated above a critical threshold in self-administering rats, suggesting that the effects of stressors on cocaine sensitivity may have been attributable to their effects on corticosterone secretion. In the present experiment, the ability of CORT administration to mimic the effects of uncontrollable stress on the acquisition of cocaine-seeking behavior was investigated using an ascending dose response curve for intravenous cocaine self-administration in rats. The potential roles of type I MR and type II GR in the effects of corticosterone on self-administration were assessed using the MR agonist, ALDO and the GR agonist DEX.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Subjects. Ninety adult, male Wistar rats (Harlan Sprague Dawley, Indianapolis, IN) 80 to 100 days old at the start of the experiments were used. Fifty rats were used to investigate the effects of daily drug pretreatments on ascending dose-response curves for intravenous cocaine self-administration. The remaining 40 rats were used to determine the effects of these pretreatments on the plasma corticosterone response to intraperitoneal injections of cocaine. All rats were housed individually in cages equipped with a laminar flow unit and air filter in a temperature- and humidity-controlled, AAALAC-accredited animal care facility on a reversed 12-hr light-dark cycle (lights on at 6:00 p.m.). Rats were maintained at 85% to 90% of their free-feeding body weights by presentations of food pellets (P. J. Noyes, Lancaster, NH; 45 mg) during the behavioral sessions when applicable and/or by supplemental feeding (Purina Rat Chow) and had access to water ad libitum. All procedures were carried out in accordance with the NIH "Principles of Laboratory Animal Care" (NIH publication No. 85-23).

Venous catheterization and drug delivery. For the cocaine self-administration experiments, each rat was implanted with a chronic indwelling jugular catheter under sodium pentobarbital anesthesia (50 mg/kg i.p.) with methylatropine nitrate pretreatment (10 mg/kg i.p.) using previously reported procedures (Koob and Goeders, 1989; Goeders and Guerin, 1996b). A silicon catheter (0.20 mm o.d. × 0.037 mm i.d.) was inserted into the right posterior facial vein and pushed down into the jugular vein until it terminated outside the right atrium. The catheter was sutured to tissue surrounding the vein and continued subcutaneously to the back where it exited just posterior to the scapulae via a Marlex mesh/dental acrylic/22-guage guide cannula (Plastic One, Roanoke, VA) assembly. This assembly was anchored to tissue under the skin for attachment of a stainless steel spring leash (Plastics One, Inc.) which was connected to a 20 ml syringe in a motor-driven pump (Razel, Stamford, CT) via a leak-proof fluid swivel suspended above the chamber to allow the delivery of drug solutions. The swivel and leash assembly was counterbalanced to permit relatively unrestrained movement of the animal. The animals were injected with sterile penicillin G procaine suspension (75,000 units i.m.) immediately before surgery, and they were allowed a minimum of 5 days to recover after surgery. The swivel and leash assembly was always connected during the experimental sessions. At the end of each session the leash was disconnected, the catheter was filled with Urokinase (250,000 i.u.) to eliminate blood clots, and a dummy cannula was inserted into the guide before rats were returned to their home cages. Catheter patency was tested regularly by obtaining blood from the catheter. If blood could not be obtained, methohexital sodium (1.5 mg) was infused through the catheter. In this case, a functional catheter was indicated by immediate, light anesthesia.

Experimental apparati. Modified plastic and stainless steel operant conditioning chambers contained in sound-attenuating cubicles (Med-Associates, Lafayette, IN) were used for the self-administration experiments. The operant chambers were equipped with two retractable response levers with stimulus lights located above each lever. One lever was mounted on the front wall of the chamber next to a food pellet dispenser. The other lever was mounted in the center of the back wall. The cubicles were also equipped with an exhaust fan that supplied ventilation and white noise to mask extraneous sound. Programming and data collection were performed using Med-PC software and an IBM-compatible personal computer and interface system (Med-Associates).

Measurement of locomotor responses to novelty and cocaine. Rats were initially screened for their locomotor responses to a novel environment and to an acute intraperitoneal injection of cocaine. It has been reported that the behavioral responses to novelty of individual rats can predict their propensity to self-administer amphetamine (Piazza et al., 1989). To control for this potential source of variability, rats were assigned to the drug treatment groups so that the mean locomotor responses to novelty and cocaine did not significantly differ between groups. Locomotor testing was conducted in 48 cm long × 25 cm wide Plexiglas chambers contained within a photocell apparatus (Coulbourn, Allentown, PA) consisting of two photocell beams evenly spaced across the length of each chamber. The number of individual breaks of the front and back beams, consecutive beam breaks (crossovers), and total locomotor counts were recorded over 60-min sessions. On the first day of locomotor testing (novelty), animals were placed into the chambers and activity was measured immediately for 60 min. On day 2, the locomotor response to an injection of saline (0.9% NaCl, 1 ml/kg i.p.) was recorded after a 60-min acclimation period. On day3, the locomotor response to an acute injection of cocaine HCl (15 mg/kg i.p., National Institute on Drug Abuse) was recorded after acclimation.

Food-reinforced responding. Rats were initially trained to respond under a fixed-ratio 1 (FR1) schedule of food reinforcement during daily sessions. At the start of these sessions, the food (front) lever was extended and the corresponding stimulus light was illuminated. Each response on the food lever resulted in the presentation of a single food pellet (45 mg) into the dispenser. This was immediately followed by a 25-sec timeout period during which the lever was retracted and the stimulus light extinguished. Sessions were terminated after 60 min or when 100 food pellets were delivered. Once stable patterns of food-reinforced responding were observed (i.e., 100 food pellets obtained in 10 min or less for ~1 week), food sessions were only conducted once weekly to control for potential nonspecific effects of the drug treatments on operant behavior.

Acquisition of intravenous cocaine self-administration. The acquisition of intravenous cocaine self-administration was determined using an ascending dose-response curve as previously reported (Goeders and Guerin, 1994). Rats were allowed to self-administer cocaine or vehicle (heparinized 0.9% NaCl bacteriostatic saline) under an FR1 schedule of reinforcement during daily 1-hr sessions. During these sessions, only the drug (back) lever was extended and the corresponding stimulus light illuminated. Each depression of the lever resulted in an intravenous infusion (200 µl over 5.6 sec) of cocaine or vehicle solution followed by a 20-sec timeout period during which the lever was retracted and the stimulus light extinguished. During the initial 2 weeks of the study, a base line for responding on the drug lever was established with only vehicle available for self-administration. This was followed by the initiation of the daily drug treatment regimen (see below). Drugs were administered daily for 2 weeks before the determination of the dose-response curve. During this 2-week period, no behavioral experiments were performed. These treatments persisted daily throughout the course of the experiment. Rats were then tested for the acquisition of intravenous cocaine self-administration using an ascending cocaine dose-response curve. During dose-response testing, food sessions were conducted on Mondays and self-administration sessions were conducted on Tuesday through Friday. Control rates of responding were established with only vehicle available for self-administration during the first week. Rats were then tested with a very low dose of cocaine (0.03125 mg/kg/inf) that is not normally self-administered by rats in our laboratory. This dose was doubled weekly so that each rat was tested with 0.0, 0.03125, 0.0625, 0.125, 0.25, 0.5 and 1.0 mg/kg/inf cocaine. Subsequent to dose-response determination in each rat, cocaine self-administration was extinguished by replacing the cocaine solution with vehicle for 4 consecutive sessions.

Drug treatments. Rats were divided into 6 treatment groups. All treatments were administered daily, starting 2 weeks before dose-response testing and continuing through the remainder of the experiment. Rats received daily i.p. injections of VEH (bacteriostatic saline 0.9% NaCl, n = 10), CORT (2.0 mg/kg, suspended in saline, n = 7; Sigma, St. Louis), ALDO (100 µg/kg dissolved in saline, n = 7; Sigma), DEX (10 or 100 µg/kg dissolved in saline, n = 6 for each; Sigma) or a combination of DEX and ALDO (10 µg/kg DEX and 100 µg/kg ALDO, n = 6). During dose-response testing, the presession treatment time for each of the drugs was 15 min except for DEX, which was administered 60 min before the sessions.

Plasma corticosterone responses to self-administered infusions of cocaine. The plasma corticosterone response to single self-administered infusions of cocaine was determined in some rats as follows. Rats were placed in the self-administration chambers as described above. After the self-administration of single infusions of vehicle, 0.0625, 0.25 or 1.0 mg/kg/inf cocaine, the session was terminated and blood for plasma corticosterone determination was obtained 15 min postinfusion from the catheter (when possible) or from the tail vein under light methohexital sodium (5.0 mg/kg i.v.) anesthesia (Goeders and Guerin, 1996b). Blood (~500 µl) was collected into preheparinized tubes and placed on ice. Blood was centrifuged to allow separation of plasma, which was collected and frozen at -20°C until needed. Plasma corticosterone was measured using the Immunochem^TM Double Antibody Corticosterone assay kit (ICN Biomedical, Irvine, CA).

Plasma corticosterone responses to intraperitoneal injections of cocaine. The effects of drug treatments on the plasma corticosterone response to i.p. injections of cocaine was measured in an additional 40 rats. As above, these rats received daily i.p. injections of VEH, CORT, ALDO or one of the two doses of DEX (n = 8/group). For these experiments, blood for plasma corticosterone determination was obtained from the tail vein of conscious rats after 2 weeks of habituation to the blood sampling procedure as described previously (Simar et al., 1997). Rats were wrapped in a hand towel, and 1 to 2 mm was cut from the tip of the tail. Corticosterone was measured as described above. All blood sampling was performed between 9:00 a.m. and noon. On the first day of drug treatments (acute) the corticosterone response to an i.p. injection of 20 mg/kg cocaine was determined 20 min postinjection. This was followed by 2 weeks of drug treatments, during which no sampling was performed. A dose-response curve for the plasma corticosterone response to i.p. injections of cocaine after chronic drug treatments was then determined. Rats were tested for their corticosterone responses to 0.0, 5.0, 10.0 and 20.0 mg/kg cocaine (i.p., 20 min postinjection). The sequence of cocaine dosing was randomized and blood samples were obtained no more than twice weekly for 3 weeks. The plasma corticosterone response to 20 mg/kg cocaine was also tested after a 3-day treatment washout to determine the persistence of the effects of the drug treatments on the corticosterone response to cocaine. All drugs were administered 45 min before cocaine injections except for DEX, which was administered 90 min before cocaine. These pretreatment times were chosen so that they would correspond with points in the middle of the 60-min self-administration sessions from the previous experiments. At the end of the experiment, these rats were killed by decapitation and thymus and adrenals were dissected so that their weights could be determined.

Statistical analysis. Data collected during the self-administration experiments included the mean number of infusions self-administered per session for each dose of cocaine and the mean rates of food-reinforced responding during the weekly food sessions. Self-administration sessions were conducted Tuesday through Friday, and the means were determined from data from the final 3 sessions (i.e., Wednesday through Friday) for each dose. The acquisition of cocaine self-administration was defined as the lowest dose at which the mean number infusions/session was significantly greater than the base-line measurements (P < .05). The significance of the differences between drug and vehicle treatment conditions and between doses of cocaine within treatment groups was determined using repeated measures analysis of variance (ANOVA). Post-hoc analyses were performed using the Fisher's Protected Least Significant Difference test (Fisher's PLSD). The significance of differences in the mean corticosterone responses (ng/ml) to cocaine were determined as described above. One-way ANOVA was used to determine any significance of differences in body, thymus and adrenal weights between groups.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

The 6 treatment groups did not significantly differ in their mean body weights or locomotor responses to novelty or acute injections of cocaine (15 mg/kg i.p.) at the start of the experiment (data not shown).

Effects on food-reinforced responding. Table 1 shows the base-line rates of food-reinforced responding (responses/min ± S.E.M.) and the effects of the various treatments on food reinforcement at timepoints corresponding to the doses of cocaine that were available for self-administration each week. A significant effect of the 100 µg/kg dose of DEX on food-reinforced responding was observed over time [Time effect, F(7,173) = 5.861, P < .0001; Time × Treatment interaction, F(7,2) = 4.103, P < .0001]. The rate of responding was significantly decreased from base line at the timepoints corresponding to weeks during which 0.25 and 1.0 mg/kg/inf cocaine were available for self-administration (P < .05 for each). At the 1.0 mg/kg/inf timepoint, the response rate in 100 µg/kg DEX-treated rats was also significantly less than in VEH controls (P < .05). This effect on food-reinforcement was not apparent in rats treated with the 10 µg/kg dose of DEX. No other significant differences in food-reinforced responding within or between treatment groups were observed.

Effects on the acquisition of intravenous cocaine self-administration. The effect of daily pretreatment with CORT (2.0 mg/kg i.p.) on the ascending dose-response curve for iv cocaine self-administration is illustrated in figure 1. The acquisition of cocaine self-administration by both VEH- and CORT-treated rats was indicated by a significant effect of dose [Dose effect, F(6,109) = 6.41, P < .0001]. CORT-treated rats acquired cocaine self-administration at a lower dose (0.0625 mg/kg/inf; P < .05 vs. base line) than VEH-treated controls (0.125 mg/kg/inf; P < .0001). A significant treatment effect of CORT was also observed [Treatment effect, F(1,109) = 5.699, P < .05]. The mean number of infusions/session at the 0.0625 mg/kg/inf dose of cocaine in CORT-treated rats was significantly greater than in VEH-treated rats (P < .05). The effects of daily pretreatments with ALDO (100 µg/kg i.p.) or DEX (10 or 100 µg/kg i.p.) are shown in figure 2A. In ALDO-treated rats, a significant effect of dose was observed [Dose effect, F(6,114) = 3.495, P < .001]. Similar to the VEH controls, rats from the ALDO group acquired cocaine self-administration at the 0.125 mg/kg/infusion dose (P < .05). No significant differences between ALDO- and VEH-treated rats were observed. Although there was no significant effect of treatment in rats treated daily with the 10 or 100 µg/kg doses of DEX compared with VEH, a significant interaction between dose and treatment was observed [Treatment × Dose, F(2,6) = 2.350, P < .05]. In contrast to VEH-treated rats, significant acquisition of cocaine self-administration did not occur at any dose of cocaine tested in rats treated with either dose of DEX. A small but statistically non-significant trend toward acquisition was observed in the 10 µg/kg DEX group at the 0.5 mg/kg/inf dose of cocaine. Figure 2B shows the effects of coadministration of 100 µg/kg ALDO on the 10 µg/kg DEX-induced suppression of cocaine self-administration. No significant differences in self-administration were observed at any dose between rats treated with the DEX/ALDO combination and those treated with 10 µg/kg dose of DEX alone. However, a significant effect of dose was observed [Dose effect, F(1,83) = 3.498, P < .01]. In contrast to rats treated with 10 µg/kg DEX alone, rats treated with the DEX/ALDO combination did acquire self-administration at the 0.125 mg/kg/infusion dose as defined by a significantly greater number of infusions/session at this dose compared with base line (P < .05). Significant differences from base line were also observed at the 0.25 and 0.5 mg/kg/infusion doses of cocaine in this group.

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Ascending dose-response curves for intravenous cocaine self-administration in rats pretreated daily with CORT (2.0 mg/kg; n = 7) or VEH (n = 10). Data points represent the mean number of infusions per session (± S.E.M.) for each dose of cocaine. The significance of the differences between treatments or doses was determined using repeated measures analysis of variance followed by post-hoc testing using the Fisher's Least Significant Difference test. The acquisition of cocaine self-administration was defined by the lowest dose of cocaine in each treatment group at which the mean number of infusions/session was significantly greater than base line responding. * P < .05 vs. base line. ** P < .05 CORT vs. VEH.

View larger version (28K): [in this window] [in a new window]   Fig. 2.   Ascending dose-response curves for intravenous cocaine self-administration in rats pretreated daily with ALDO (100 µg/kg; n = 7), DEX (10 or 100 µg/kg; n = 6 for each dose) or VEH (n = 10; fig. 2A) or in rats treated daily with VEH, DEX (10 µg/kg) alone or DEX in combination with ALDO (100 µg/kg; n = 6; fig. 2B). Data points represent the mean number of infusions per session (± S.E.M.) for each dose of cocaine. The significance of the differences between treatments or doses was determined using repeated measures analysis of variance followed by post-hoc testing using the Fisher's Least Significant Difference test. The acquisition of cocaine self-administration was defined by the lowest dose of cocaine in each treatment group at which the mean number of infusions/session was significantly greater than base line responding. * P < .05 vs. base line. ** P < .001 vs. base line.

Extinction of cocaine-reinforced responding. In VEH-treated rats, replacement of the cocaine solution with saline for 4 consecutive days after the determination of the dose-response curves produced a pattern of extinction characterized by a high level of responding on day 1 (51.63 ± 6.57 responses per session) followed by a progressive decline in responding from days 2 to 4. On day 4, the mean number of infusions per session (15.13 ± 3.16) approached that observed under base-line conditions. Similar extinction patterns were observed in the CORT and ALDO treatment groups (data not shown). During extinction, there was a significant effect of day [F(3,139) = 34.18, P < .0001] and an interaction between extinction day and drug treatment [Day × Treatment, F(3,4) = 3.20, P < .01]. On day 1, the mean numbers of responses in the 10 and 100 µg DEX groups (24.75 ± 11.35 and 18.4 ± 6.2 respectively) were significantly less than the number of responses in VEH-treated rats (P < .05 for each). No significant differences in responding were observed between groups on extinction days 2 through 4.

Corticosterone responses to self-administered infusions of cocaine. The effects of daily pretreatments with VEH, DEX (10 or 100 µg/kg) or the combination of DEX (10 µg/kg) and ALDO (100 µg/kg) on the plasma corticosterone response to single self-administered iv infusions of cocaine are presented in figure 3. Despite a trend toward increases in plasma corticosterone with increasing doses of cocaine in VEH-treated rats, no significant effect of dose was observed. However, there was a highly significant effect of treatment [F(3,67) = 61.790, P < .0001]. DEX dose-dependently suppressed the plasma corticosterone responses to infusions of saline and to the 0.0625, 0.25, and 1.0 mg/kg/infusion doses of cocaine. Significant decreases compared with VEH-treated controls were observed in both the 10 µg/kg (P < .05 for all) and 100 µg/kg (P < .01 for all) treatment groups. In the 100 µg/kg DEX group, plasma corticosterone was almost unmeasurable (i.e., <5 ng/ml) after the infusion of each dose of cocaine. Coadministration of ALDO (100 µg/kg) with 10 µg/kg DEX failed to reverse the inhibitory effects of DEX on the plasma corticosterone response to cocaine.

View larger version (28K): [in this window] [in a new window]   Fig. 3.   Effects of daily pretreatment with VEH, DEX (10 or 100 µg/kg), or a combination of DEX (10 µg/kg) and ALDO (100 µg/kg) on the plasma corticosterone responses to single self-administered i.v. infusions of cocaine. Data points represent mean plasma corticosterone concentrations (ng/ml ± S.E.M.). Corticosterone was measured 15 min after the self-administration of single infusions of cocaine or saline. The significance of the differences between treatments or doses was determined using repeated measures analysis of variance followed by post-hoc testing using the Fisher's Least Significant Difference test. * P < .05 vs. VEH-treated rats. ** P < .01 vs. VEH-treated rats.

Corticosterone responses to intraperitoneal injections of cocaine. The effects of daily pretreatments with VEH, CORT, ALDO, and DEX (10 and 100 µg/kg) on the plasma corticosterone responses to i.p. injections of cocaine are illustrated in figures 4 and 5. Figure 4 shows the effects of chronic treatment with these drugs on the plasma corticosterone response to a randomized sequence of i.p. injections with various doses of cocaine (0, 5, 10, and 20 mg/kg). Significant effects of cocaine dose (Dose effect, F(3,146) = 2.916, P < .05) and drug treatment (Treatment effect, F(4,146) = 80.280, P < .0001) were observed. In VEH-treated rats, cocaine produced a dose-dependent increase in plasma corticosterone with significant increases compared with the 0 mg/kg dose after the administration of 10 (P < .01) and 20 (P < .001) mg/kg cocaine. Plasma corticosterone after injection with 20 mg/kg cocaine was also significant when compared with the 5 mg/kg dose of cocaine. As expected, CORT pretreatment resulted in increases in plasma corticosterone which were significantly greater than those seen in VEH-treated rats after administration of 0 (P < .0001) and 5 (P < .05) mg/kg cocaine. At the 0 mg/kg cocaine dose, administration of CORT (2.0 mg/kg i.p.) produced almost a 200% increase in the plasma concentration of the hormone compared with VEH-treated controls. These findings indicate that i.p. injections with CORT resulted in significant elevations in plasma corticosterone concentrations 45 min after administration, a timepoint which corresponds to the middle of the self-administration sessions in the previous experiments. The differences between CORT- and VEH-treated rats were attenuated with the administration of 10 and 20 mg/kg cocaine. No dose-related effects of cocaine were observed in CORT-treated rats, indicating that the elevated corticosterone may have prevented further cocaine-induced increases of the hormone via feedback mechanisms. Daily ALDO pretreatment did not affect the corticosterone response to i.p. injections of cocaine compared with VEH treatment. Daily pretreatment with DEX (10 and 100 µg/kg) dose-dependently suppressed the plasma corticosterone responses to i.p. injections of cocaine in a manner similar to that observed with self-administered iv infusions of cocaine. After injections of saline (0 mg/kg cocaine), plasma corticosterone was significantly decreased in the 10 (P < .001) and 100 (P < .01) µg/kg DEX groups compared with VEH-treated controls. Likewise, the plasma corticosterone responses to 5, 10 and 20 mg/kg cocaine were significantly decreased in both DEX treatment groups compared with VEH treatment (P < .0001 for all). Under all conditions, corticosterone in the 100 µg/kg DEX-treated rats was virtually undetectable (<10 ng/ml), while the mean plasma corticosterone concentration in the 10 µg/kg DEX group remained at ~50 ng/ml.

View larger version (29K): [in this window] [in a new window]   Fig. 4.   Effects of daily pretreatment with CORT (2.0 mg/kg), ALDO (100 µg/kg), DEX (10 or 100 µg/kg) or VEH on the plasma corticosterone responses to i.p. injections of cocaine (n = 8/dose). Data represent the mean plasma corticosterone response (ng/ml ± S.E.M.) 20 min after injection with cocaine or saline after at least 14 days of daily drug treatments. Drug pretreatments were administered 45 min before cocaine injections, except for DEX, which was administered 90 min before cocaine injections. The significance of the differences between treatments or doses was determined using repeated measures analysis of variance followed by post-hoc testing using the Fisher's Least Significant Difference test. * P < .01 DEX vs. VEH. ** P < .01 vs. 0 mg/kg (base line).  P < .05 vs. VEH

View larger version (31K): [in this window] [in a new window]   Fig. 5.   Plasma corticosterone responses to cocaine after acute or chronic pretreatment with CORT, ALDO, DEX, or VEH, or after a 3-day drug treatment washout period. Data represent the mean plasma corticosterone concentrations (ng/ml ± S.E.M.) 20 min after i.p. injections with cocaine (20 mg/kg). Drug pretreatments were administered 45 min before cocaine injections, except for DEX, which was administered 90 min before cocaine injections. The effects of acute pretreatment were measured on day 1 of the dosing regimen. The effects of chronic pretreatment were measured after 15 days of drug administration. The effects of treatment washout were determined by discontinuing drug treatment for 3 days before the cocaine injection. The significance of the differences between treatments or doses was determined using repeated measures analysis of variance followed by post-hoc testing using the Fisher's Least Significant Difference test. * P < .01 vs. acute treatment. ** P < .05 vs. VEH.  P < .0001 vs. chronic and acute treatment

Effects on body, thymus and adrenal weights. The effects of the various treatments on mean body, thymus and adrenal weights and on the ratios of adrenal and thymus to body weight at the end of the experiment are represented in table 2. Significant effects of treatments were observed on body weight [F(4,39) = 17.006, P < .0001], thymus weights [F(4,39) = 114.486, P < .0001], adrenal weight [F(4,39) = 18.848, P < .0001], the ratio of thymus/body weights [F(4,35) = 109.951, P < .0001] and the ratio of adrenal/body weight [F(4,39) = 13.516, P < .0001]. Significant decreases in thymus weight (P < .05) and thymus/body weight (P < .01) were observed in CORT-treated rats compared with VEH-treated controls. An inverse proportion exists between the activation of type II receptors in the thymus and thymus weight, making this variable a useful measure of peripheral type II receptor occupation (Akana et al., 1985). CORT treatment did not significantly affect total body or adrenal weights. Significant dose-related effects of DEX on all of the above measures were observed. Daily administration of 100, but not 10, µg/kg DEX resulted in a significant decrease in body weight at the end of the experiment compared with VEH-treated controls (P < .0001). Significant reductions in thymus weight and thymus/body weight were also observed in rats treated with 10 and 100 µg/kg DEX (P < .0001 for all comparisons). In the 100 µg/kg group, the thymus was indiscernable (thymus weights of 0 were used for statistical analysis). Significant decreases in adrenal weights and adrenal/body weight ratios compared with VEH-treated rats (P < .0001 for both) were observed in rats treated with 100, but not 10, µg/kg DEX. This difference probably reflects atrophy of the adrenals as a result of negative feedback of DEX on ACTH secretion.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The results of these experiments demonstrate that daily pretreatment with CORT (2.0 mg/kg i.p.) facilitates the acquisition of cocaine-seeking behavior in rats, as observed by a leftward and upward shift in the ascending dose-response curve for intravenous cocaine self-administration. CORT-treated rats acquired self-administration at a lower dose of cocaine than did VEH-treated controls (0.0625 vs. 0.125 mg/kg/infusion). In contrast, no significant differences in food-maintained responding were observed between these treatment groups. The present results are consistent with findings that corticosterone is necessary for the establishment and maintenance of cocaine self-administration in rats (Goeders and Guerin, 1996a; 1996b; Deroche et al., 1997) and with reports that CORT-treated rats are more sensitive to the reinforcing (Piazza et al., 1991) and locomotor activating (Marinelli et al., 1997; Cador et al., 1993) effects of psychostimulants.

Previous studies from this laboratory have demonstrated that daily preexposure to uncontrollable EFS facilitates the acquisition of cocaine self-administration in rats (Goeders and Guerin, 1994). In these experiments, a significant positive correlation was observed between the effects of EFS on plasma corticosterone and on self-administration (Goeders and Guerin, 1996a), suggesting that the facilitation of self-administration by EFS may have been mediated by corticosterone. In the present experiment, the CORT dosing regimen was designed to mimic the schedule of EFS exposure used in these previous studies. CORT treatment increased plasma corticosterone concentrations by almost 200% (e.g., >300 ng/ml). Although these concentrations are higher than those observed after EFS in the previous experiments (e.g., 200 ng/ml; Goeders and Guerin, 1996a), they are still within the EFS stressor-induced range under certain conditions (unpublished observations). The finding that corticosterone and EFS produce similar effects on cocaine self-administration provides further evidence that corticosterone may be a substrate through which environmental variables interact with the reinforcing effects of cocaine.

Corticosterone produces many of its effects by binding to two types of cytosolic receptors (Jöels and De Kloet, 1994). The type I (mineralocorticoid) receptor has a higher affinity for corticosterone and is usually occupied at basal concentrations of the hormone. In contrast, the type II (glucocorticoid) receptor has a lower affinity for corticosterone and is more likely to be occupied when corticosterone concentrations are elevated (e.g. during "stress" or after cocaine administration). To determine the roles of type I and type II receptors in the effects of corticosterone on cocaine self-administration, rats were treated daily with the type I receptor agonist, ALDO, or the type II receptor agonist, DEX, and ascending dose-response curves for cocaine self-administration were determined.

Daily administration of either ALDO or DEX failed to produce CORT-like effects on the acquisition of cocaine self-administration. Interestingly, however, DEX-treated rats did not acquire self-administration at any dose. This effect was associated with significant attenuations of food-maintained responding and weight loss in rats treated with 100 µg/kg but not 10 µg/kg DEX. In rats treated with 10 µg/kg DEX, no significant differences from base line were observed in self-administration despite a trend toward acquisition at the higher doses of cocaine. Compared with the other treatment groups, DEX-treated rats also displayed a significantly attenuated extinction response pattern when the cocaine was replaced with saline.

The finding that DEX-treated rats did not acquire cocaine self-administration was unexpected. However, these results are consistent with reports that DEX pretreatment blocks the locomotor stimulating effects of cocaine and amphetamine in mice (Capasso et al., 1996). We hypothesized that the effects of DEX may have been a result of feedback inhibition of the HPA response to cocaine, since the dose-response curves for cocaine self-administration in the DEX-treated groups resembled those observed in ADX rats (Goeders and Guerin, 1996a; Deroche et al., 1997). In addition to mediating many of the physiological effects of glucocorticoids, type II receptors are also important for negative feedback on the HPA axis in response to elevated corticosterone (Dallman et al., 1994). Thus, administration of DEX has been reported to attenuate the plasma corticosterone responses to stressors (Donald, 1966) and cocaine (Simar et al., 1997). In the present study, treatment with DEX produced dose-dependent decreases in the plasma corticosterone responses to either i.p. or single self-administered iv infusions of cocaine. In fact, in both DEX-treated groups, plasma corticosterone was also significantly decreased after saline administration, suggesting that basal concentrations of the hormone were reduced. Cocaine-induced increases in corticosterone were restored after DEX treatment was discontinued for 3 days in the 10 µg/kg but not the 100 µg/kg DEX group. Thus, daily pretreatment with 10 µg/kg DEX blocked cocaine self-administration without producing long-term suppression of the HPA axis. In both the 10 and 100 µg/kg DEX groups, significant decreases in the plasma corticosterone response to cocaine were observed after a single DEX pretreatment. For the 100 µg/kg dose of DEX, the corticosterone response to cocaine was further diminished after chronic DEX treatment compared with its acute administration, suggesting a cumulative effect of the DEX treatment.

Based on the decreases in thymus weight observed in the DEX-treated rats, it is probable that DEX was occupying type II receptors peripherally. Peripheral type II receptors, most likely in the anterior pituitary, are believed to be important for the feedback inhibitory effects of DEX on the HPA axis (De Kloet et al., 1974). However, in the present study, the extent to which type II receptors within the brain were occupied is unknown. DEX penetrates the brain relatively poorly (Rees, 1974) and may even be actively transported out the central nervous system (De Kloet, 1997). Western blot analysis of brain tissue from the DEX-treated rats indicated that no significant down-regulation of type II receptors occurred in a number brain regions (unpublished observation). Thus, it is possible that the present findings reflect the unusual pharmacodynamic profile of DEX itself, rather than the effects of central type II receptor occupation.

Under basal corticosterone concentrations, type I receptors are predominantly occupied, while type II receptors remain largely unbound (Joels and De Kloet, 1994). In the present study, plasma corticosterone was 60 ng/ml or less in rats treated with 10 µg/kg DEX and was virtually undetectable in rats treated with 100 µg/kg DEX under all conditions. In either case, it is likely that plasma concentrations of corticosterone were low enough that type I receptors were not fully occupied. Thus, DEX treatment may have resulted in a physiologically abnormal state in which type II receptors were bound but type I receptor occupation was submaximal. Although it had no effects on cocaine self-administration by itself, coadministration of the type I receptor agonist ALDO (100 µg/kg) with DEX (10 µg/kg) restored self-administration to levels observed in VEH-treated rats. This DEX/ALDO combination, at the doses used, did not produce CORT-like effects on self-administration (e.g., facilitation of acquisition vs. VEH-treated rats). The restoration of self-administration by ALDO was not accompanied by a reversal of the DEX-induced suppression of the corticosterone response to cocaine. Thus, type I receptor occupation by basal corticosterone concentrations may be necessary for the acquisition of cocaine self-administration. These findings are consistent with reports that the occupation of type I receptors is required for the locomotor stimulating effects of cocaine (Marinelli et al., 1997).

It has been demonstrated that exposure to stress or treatment with CORT, but not DEX, facilitates ethanol self-administration in rats (Fählke et al., 1995). In these experiments, the effects of CORT were not blocked by antagonists at type I (RU 28318) or type II (RU 38486) receptors, indicating that CORT may have been acting independently of these receptors. A non-MR/GR effect of CORT could explain the ability of DEX to prevent the acquisition of cocaine self-administration despite its occupation of type II receptors. In such a scenario, the type II receptor-mediated feedback inhibition of the HPA axis by DEX could block any non-GR effects of corticosterone related to reinforcement by preventing the release of the hormone in response to cocaine.

An important role for the mesocorticolimbic dopaminergic (DA) system in drug reinforcement has been clearly established (Koob, 1992). It has been demonstrated that the facilitation of the acquisition of cocaine self-administration by social stress is associated with an enhancement of DA transmission within this system (Tidey and Miczek, 1997). It has also been reported that corticosterone in the range of stress-induced concentrations can stimulate the release of DA in the nucleus accumbens (Piazza et al., 1996). Thus, corticosterone, and its effects on mesocorticolimbic DA transmission, may be an interface through which "stress" and cocaine reinforcement interact.

In summary, the results from these experiments demonstrate that daily pretreatment with corticosterone facilitates the acquisition of intravenous cocaine self-administration in rats. These findings suggest that corticosterone may provide a substrate through which environmental factors interact with the reinforcing effects of cocaine and other drugs of abuse. Differences in plasma corticosterone before or at the time of cocaine exposure may be critical determinants of whether or not individuals will progress to compulsive drug use and may provide a target for clinical intervention once self-administration has been established. Finally, the findings that daily treatment with dexamethasone blocks the acquisition of cocaine self-administration may be relevant to the development of novel pharmacotherapeutic approaches to the treatment of cocaine addiction.