Introduction

Top Abstract Introduction Methods Results Discussion References

Cocaine's actions have been attributed to binding at DA, 5-HT and NE transporters, with a resultant inhibition of monoamine reuptake (e.g., Blackburn et al., 1967; Coyle and Snyder, 1969; Ross and Renyi, 1969; Javitch et al., 1984; Madras et al., 1989; Ritz and Kuhar, 1989). Of these neuropharmacological mechanisms, the reinforcing effects of cocaine are believed to be primarily mediated through the dopaminergic system. For example, in cocaine self-administration studies, dopamine antagonists have been shown to shift the cocaine dose-response curve to the right (e.g., Bergman et al., 1990), and direct-acting DA agonists have been shown to function as reinforcers when substituted for cocaine (e.g., Caine and Koob, 1993; Weed and Woolverton, 1995; Nader and Mach, 1996). Using indirect DA agonists, there is clear evidence of a positive correlation between binding affinity at the DA transporter and potency in drug self-administration studies (Ritz et al., 1987). In contrast, there does not appear to be a significant correlation between binding to 5-HT or NE transporters and drug self-administration (Ritz et al., 1987; Woolverton, 1987).

Several laboratories have synthesized cocaine-like derivatives that differ in potency and selectivity for cocaine binding sites, in an effort to better understand the neurobiological actions of cocaine at each monoamine transporter. For example, Carroll and colleagues (Boja et al., 1990; Carroll et al., 1991, 1992a, 1992b, 1993) have synthesized cocaine analogs that bind with 100-fold higher affinity for the DA transporter compared to cocaine. Some of these phenyl tropanes, e.g., -CIT or RTI-55, and its fluorinated analog (-CFT), have been shown to function as reinforcers and have cocaine-like discriminative stimulus effects in non-human primates (Spealman et al., 1991a, 1991b; Weed et al., 1995). One of the limitations of the synthetic pathway used to develop these cocaine analogs is the use of (-) cocaine as the starting material, which restricts synthetic flexibility.

An alternative synthetic strategy, attempting to overcome the limitations of cocaine-based synthesis of tropane analogs, uses the reaction of vinylcarbenoids with pyrroles (Davies et al., 1991). This approach has led to the synthesis of several tropane analogs, some of which bind with extremely high potency to DA and/or 5-HT transporters (Davies et al., 1993, 1994; Bennett et al., 1995). One of these compounds, PTT is 20 times more potent at binding to the DA transporter compared to cocaine (8.2 vs. 173 nM) and is approximately 50 times more potent at inhibiting DA uptake (cf. table 1, Bennett et al., 1995). PTT is also more selective in binding to DA transporters than cocaine, because it is 20 to 100 times less potent at 5-HT than at DA transporters (Davies et al., 1994). In addition, PTT lacks the ester linkage of cocaine, thereby increasing its metabolic stability (Davies et al., 1993). In behavioral studies, PTT is approximately 20 to 30 times more potent than cocaine in increasing locomotor activity and stereotypy, and its duration of action was longer than that elicited by the highest cocaine doses (Hemby et al., 1995; Porrino et al., 1994, 1995).

The purpose of our studies was 2-fold. First, the ability of PTT pretreatments to decrease cocaine self-administration was assessed. One prediction was that treatment with a dopamine agonist would decrease cocaine self-administration in a manner analogous to an opiate agonist decreasing heroin or alfentanil self-administration (e.g., Mello et al., 1983; Winger et al., 1992). Recent studies with the indirect DA agonist GBR 12909 have reported decreases in cocaine self-administration, under some conditions (Skjoldager et al., 1993; Glowa et al., 1995a,b).

A second purpose was to assess the abuse liability of PTT in two animal models of cocaine abuse, drug self-administration and drug discrimination. Because PTT has a higher affinity at the DA transporter compared to cocaine, it is possible that PTT would function as a positive reinforcer with greater potency than cocaine, and such an outcome would be consistent with previous studies using phenyltropanes (Spealman and Kelleher, 1981; Spealman et al., 1991a, 1991b; Weed et al., 1995). However, the long duration of action of PTT may, in fact, decrease rates of PTT self-administration. The drug discrimination paradigm was utilized to determine whether PTT produced cocaine-like discriminative stimulus effects, and to determine the time-course for these actions.

	Methods

Top Abstract Introduction Methods Results Discussion References

Self-Administration

Subjects. Six individually housed adult rhesus monkeys (Macaca mulatta), five male and one female (1079), served as subjects. All monkeys had a history of cocaine self-administration under fixed-interval schedules (Nader and Reboussin, 1994; Nader and Bowen, 1995) and two monkeys had a history of self-administering the dopamine D₃ agonist 7-hydroxy-N, N-di-n-propyl-2-aminotetralin (Nader and Mach, 1996). Monkeys weighed between 6.0 and 10 kg under free-feeding conditions. Their body weights were maintained at approximately 90% of free-feeding weights by supplemental feeding of Purina Monkey Chow (Richmond, IN) no sooner than 30 min postsession. Monkeys were weighed approximately once a month and, if necessary, their diet was adjusted to maintain stable weights. In addition, monkeys were given a chewable multiple vitamin tablet 3 days per week and occasionally received fresh fruit. Monkeys lived in a temperature and humidity controlled colony room; lighting was maintained on a 0600:2000 on:off schedule.

Surgery. Each monkey was anesthetized with a combination of ketamine (15 mg/kg, i.m.) and butorphanol (0.03 mg/kg, i.m.) and a chronic indwelling venous catheter was surgically implanted under sterile conditions. The proximal end of the silicone catheter (0.08 cm inside diameter, Ronsil Rubber Products, Blackstone, VA) was inserted into a major vein (internal or external jugular, femoral) or brachial vein (0.02 cm inside diameter of catheter), terminating in the vena cava. The distal end of the catheter was threaded s.c. and exited through a small incision in the back of the animal. Monkeys were given 1 to 2 days to recover from surgery, prior to returning to the experiment. Antibiotics (Kefzol; Cefazolin sodium, Marsam Pharmaceuticals, Inc., Cherry Hill, NJ) were administered prophylactically for 5 to 7 days after surgery.

Apparatus. Monkeys were individually housed in sound attenuating cubicles (91 cm wide × 91 cm deep × 91 cm high; Plas Labs, Lansing, MI); the front wall of each cubicle was Plexiglas to allow the monkey visual access to the laboratory. During experiments, the front wall was covered with a drape. Each cubicle was equipped with two response levers (BRS/LVE, PRL-001, Beltsville, MD) and a peristaltic infusion pump (7531-10, Cole-Parmer Co., Chicago, IL) for delivering drug injections at a rate of approximately 1 ml/10 sec. Above each lever were four stimulus lights, two covered with white lens caps and two covered with red lens caps. Sessions began with illumination of the white lights above the left lever; only responding on the left lever had scheduled consequences. During drug injections, the white lights were extinguished and the red lights were illuminated for 10-sec. Each monkey was fitted with a stainless-steel restraint harness and spring arm (Restorations Unlimited, Chicago, IL) which attached to the rear of the cubicle. Experimental events were controlled and counted by a Macintosh II and associated interfaces, located in an adjacent room.

Procedure. Experiment 1: Presession Administration of PTT and Cocaine. For three monkeys (9119, 5664 and 9125), responding was maintained by cocaine (0.03 or 0.1 mg/kg/injection) presentation under an FI 5-min schedule, during daily 4-hr sessions. Under this schedule, the first response after 5 min produced a 10-sec cocaine injection; responses during the injection had no scheduled consequence and were not counted. When responding was stable (±20% of the mean for three consecutive sessions, with no trends in responding), monkeys were tested with either PTT (0.03-0.3 mg/kg) or cocaine (0.3-3.0 mg/kg); at least two base-line sessions separated test sessions. On the day of a test session, a dose of drug (approximately 1 ml/10 kg, in volume) was administered through the i.v. line, followed by a flush of the catheter with the dose of cocaine available for self-administration. The session began approximately 1 to 2 min after administration of the drug. Typically, dose-response curves for one drug (e.g., PTT) were completed before evaluating the second drug (e.g., cocaine). Whether PTT or cocaine was evaluated first was randomly determined between monkeys. Each dose of PTT or cocaine was typically tested twice in each monkey. After completion of cocaine and PTT dose-response curves, on rates of cocaine-maintained responding, the dose of cocaine available for self-administration (0.03 or 0.1 mg/kg/injection) was changed. Responding at the new cocaine dose was allowed to stabilize for at least five sessions before the start of presession testing of PTT or cocaine.

Data analysis. The primary dependent variables were total responses, response rate (responses/minute), total drug intake (mg/kg/session) and QL, which is an index of pattern of responding under FI schedules (Catania and Reynolds, 1968). QL values represent the proportion of the FI elapsed when 25% of the responses in that interval had been emitted. Therefore, values of more than 0.25 indicate a positively accelerating response rate across the interval. For the pretreatment study (experiment 1), separate analyses of variance were performed for PTT and cocaine, at the two self-administered doses (0.03 and 0.1 mg/kg/injection cocaine). Data from each pretreatment session were compared to the preceding control session. ED₅₀ values for PTT and cocaine were calculated on log-transformed data and were based on the linear portion of the mean-effect curve for each hour of the 4-hr session. For analysis of dose-response curves in the substitution experiment (experiment 2), mean data from the last three sessions for a particular dose were included in the analysis, for each monkey. The analysis compared the overall mean rate across all doses to rates of responding maintained by saline injections. For cocaine and PTT, data were analyzed separately by one-way repeated measures analysis of variance, with dose as the main effect. To determine if there were differences between cocaine and PTT, difference scores (i.e., Coc-PTT) were calculated and statistically compared to the value of 0 by individual t tests. Repeated measures analysis included only those dose levels for which data were obtained in at least three monkeys. For all analyses, P < .05 was considered statistically significant. For the base-line dose of cocaine (0.03 mg/kg/injection), the three sessions preceding the start of the PTT substitution experiment were included in the analysis.

Drugs. (-)Cocaine HCl, provided by the National Institute on Drug Abuse (Rockville, MD), was dissolved in sterile saline. (±)PTT was synthesized according to the procedure described by Davies et al. (1991) and dissolved in a vehicle consisting of 95% EtOH:sterile water, in a ratio of 4:1, to a concentration of 15 mg/ml PTT. All drug concentrations available for self-administration were prepared in 250 ml of sterile saline. During sessions, responding on the left lever delivered approximately 1.0 ml of drug solution over a 10-sec period. Different doses were studied by changing the drug concentration. Before the beginning of each session, catheters were flushed for approximately 30-sec with the concentration of drug available for self-administration. Because each catheter was filled with heparinized saline, the total amount of drug solution injected into the animal before the start of the session was approximately 1 ml. At the end of each session, catheters were flushed with approximately 3 ml of heparinized saline (100 U/ml), to help prevent clotting.

Drug Discrimination

Subjects. Three individually housed adult male rhesus monkeys (M. mulatta) served as subjects. All monkeys had a history of cocaine self-administration (Nader and Reboussin, 1994; Nader and Mach, 1996). One monkey (9119) was used in the self-administration experiments, described above. None of the subjects had any previous training under the discrimination paradigm, before the start of this study. Their body weights were maintained at approximately 90% of free-feeding weights by supplemental feeding of Purina Monkey Chow no sooner than 30 min postsession. Monkeys were weighed approximately once a month and, if necessary, their diet was adjusted to maintain stable weights. In addition, monkeys were given a chewable multiple vitamin tablet 3 days per week and occasionally received fresh fruit.

Apparatus. Monkeys were individually housed in sound attenuating cubicles, identical to those described above, with the addition of a food-pellet dispenser (G5210, model A, Gerbrands Corp., Arlington, MA) located on the front wall. Sessions began with illumination of the white lights above both levers. During food presentation, the white lights above the correct lever were extinguished and the red lights were illuminated for 2 sec. Each monkey was fitted with a stainless-steel restraint harness and spring arm (Restorations Unlimited, Chicago, IL) that attached to the rear of the cubicle.

Training procedure. The monkeys were trained in a two-lever, food-reinforced, drug discrimination paradigm to discriminate 0.2 mg/kg cocaine (C) from 0.5 ml saline (S). All injections were given i.m. immediately before the session. Experimental sessions began with a 10-min TO; at the end of the TO, both sets of white lever lights were illuminated and food (1 g banana-flavored pellets) was available under a FR 50 schedule for 15 min. Sessions were conducted at approximately the same time each day, typically 7 days per week. During training sessions, 50 consecutive responses on the correct lever resulted in food delivery; incorrect responses reset the FR value to 50. The presession injection determined the correct response lever. For two monkeys, the right lever was correct after cocaine administration and the left lever after saline; the order was reversed for the third monkey. Cocaine and saline were given in a fixed daily sequence (SSCCSCCSSC). Training continued until performance met the following criteria of stimulus control: 1) at least 80% of the first 50 responses were on the injection-appropriate lever and 2) at least 90% of the total responses were on the injection-appropriate lever, for five consecutive sessions.

Testing procedure. After the discrimination was acquired, test sessions were conducted no more than every three sessions, as long as performance in the intervening training sessions remained at or above criteria for stimulus control and with an intervening C and S training session between each test session. If a monkey's performance fell below criteria, the animal was returned to the training sequence until discrimination was at or above criteria for three consecutive sessions. Test sessions were identical to training sessions, except that different doses of cocaine or PTT (see below) were administered before the session and 50 consecutive responses on either lever resulted in food presentation.

Data analysis. The primary dependent variables were the percentage of total responses on the cocaine-appropriate lever, rates of responding (total responses/15 min) and reinforcement frequency (food presentations/session) in test sessions. Dose-response curves for cocaine and PTT were determined at least twice and are represented as the mean of all determinations, for each monkey. PTT was considered to have substituted for the discriminative stimulus effects of cocaine if, after PTT administration, at least 80% of the total session responses occurred on the cocaine-appropriate lever.

Drugs. Cocaine HCl was dissolved to a final concentration of 5.0 mg/ml; PTT concentration varied from 0.1 to 1.0 mg/ml. Injections were administered in a volume of 0.5 ml/10 kg. Monkeys never received cocaine on more than three consecutive sessions.

	Results

Top Abstract Introduction Methods Results Discussion References

Self-Administration

Experiment 1: presession administration of PTT and cocaine. The effects of presession administration of PTT and cocaine were studied in combination with two self-administered doses of cocaine (0.03 and 0.1 mg/kg/injection). The mean (±S.E.) rate of responding when 0.03 mg/kg/injection cocaine was available was 1.96 (0.55) responses/min; response rates were slightly higher when responding was maintained by 0.1 mg/kg/injection cocaine (2.25 ± 0.97 responses/min). Mean cocaine intake when responding was maintained by 0.03 and 0.1 mg/kg/injection cocaine was 1.30 (0.02) and 4.20 (0.09) mg/kg/session, respectively. Pattern of responding was characteristic of FI performance with QL values of 0.73 (0.02) and 0.61 (0.11) when responding was maintained by 0.03 and 0.1 mg/kg/injection cocaine, respectively.

View larger version (50K): [in this window] [in a new window]   Fig. 1.   Effects of PTT (A and B) and cocaine (C and D) pretreatments on response rates (upper panels) and total cocaine intake (lower panels), as a percentage of base-line levels, in monkeys self-administering cocaine (0.03-0.1 mg/kg/injection). Each point represents mean data from three monkeys and vertical bars represent 1 S.E.; each histogram represents a different pretreatment dose.

Experiment 2: substitution of PTT for cocaine

Cocaine-maintained response rates varied as a function of dose, with mean rates of responding significantly higher than saline [F(3, 44) = 7.80; P < .001] and were characterized by an inverted-U shaped function of dose in all monkeys (fig. 2, open circles). Cocaine intake increased in a dose-related manner, with the highest intakes occurring when 0.3 mg/kg/injection was available for self-administration (table 2). When substituted for cocaine, PTT did not maintain response rates significantly higher than saline [F(1, 16) = .03; P = .86]; the dose-response curve was relatively flat as a function of dose (fig. 2, closed circles). Daily session intake gradually increased, as a function of PTT dose, in all monkeys (table 2).

View larger version (27K): [in this window] [in a new window]   Fig. 2.   Effects of cocaine (open circles) or PTT (filled circles) dose on rates of responding in monkeys responding under a fixed-interval 5-min schedule, during daily 4-hr sessions. Each point is the mean (±S.D.) of the last three sessions a particular dose was available for self-administration.

Repeated measures comparisons between drugs revealed that rates of responding were significantly lower when PTT was self-administered compared to cocaine-maintained response rates [F(1, 32) = 36.01; P < .001]. In addition, total session intake was significantly lower for PTT compared to cocaine [F(1, 32) = 379.27; P < .001].

To better determine whether behaviorally active doses of PTT were studied in the substitution paradigm, average interinjection intervals, across the 4-hr session were determined for each dose of cocaine and PTT, as well as saline (fig. 3). When the base-line dose of cocaine (0.03 mg/kg/injection) was available, the mean inter-injection interval was slightly more than 5 min (the minimum interinjection interval). At this dose of cocaine, the mean interinjection interval did not substantially increase from hour 1 to 4 (data not shown). When saline was available for self-administration, the mean interinjection interval was approximately 28 min (fig. 3). At the highest (0.3 mg/kg/injection) and the lowest (0.003 mg/kg/injection) cocaine doses, the mean interinjection interval was approximately 8 and 10 min, respectively. When 0.003 mg/kg/injection PTT was self-administered the interinjection interval was similar to that observed when saline was available (approximately 27 min). A one-half log unit increase in PTT dose resulted in a decrease in the mean interinjection interval to approximately 13 minutes. Higher doses of PTT resulted in large increases in mean interinjection intervals (fig. 3).

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Average interinjection interval (min) as a function of cocaine (open circles) or PTT (filled circles) dose for monkeys responding under an FI 5-min schedule of drug presentation. Each point is the mean of the last three sessions averaged across four monkeys. Vertical lines represent 1 S.E. The dashed line represents the minimum interinjection interval possible under the FI 5-min schedule.

Drug Discrimination

Criteria performance was obtained in each animal within 90 to 120 training sessions. When tested with saline, all monkeys responded exclusively on the saline-appropriate lever (fig. 4), with a mean session rate of responding of 3.2 to 3.8 responses/second (see table 4). When the training dose of cocaine (0.2 mg/kg) was tested, 90 to 100% of the total responses occurred on the cocaine-appropriate lever. This dose of cocaine had minimal effects on response rates, with mean rates of responding of 3.6 to 4.6 responses/second (see table 4). Other doses of cocaine (0.03-0.1 mg/kg) occasioned dose-dependent increases in cocaine-appropriate responding (fig. 4, open circles), with no significant effect on rates of responding (data not shown). Monkeys typically received between 70 to 100 food pellets per session, at all cocaine doses.

View larger version (18K): [in this window] [in a new window]   Fig. 4.   The discriminative stimulus effects (% of total responses that occurred on the cocaine-appropriate lever) of cocaine (open symbols) and PTT (filled symbols) in monkeys trained to discriminate cocaine (0.2 mg/kg, i.m.) from saline (0.5 ml, i.m.). Cocaine was administered 10 min before testing and PTT was tested 1 and 5 hr after administration. Symbols above Veh represent saline (open symbols), tested 10 min after injection and PTT vehicle (closed symbols), tested 1 hr after administration. Each point represents the mean of at least two determinations.

When tested 1 hr after drug administration, PTT (0.003-0.1 mg/kg) occasioned dose-related increases in cocaine-appropriate responding, in all three monkeys (fig. 4, filled circles). PTT was 0.5 to 1.0 log units more potent than cocaine at producing cocaine-like discriminative stimulus effects (table 3). At these doses of PTT, there were no significant effects on rates of responding (data not shown). Monkeys typically received between 70 to 100 food pellets per session, at all PTT doses. When tested 5 hr after administration, PTT dose-dependently occasioned cocaine-appropriate responding in each monkey (fig. 4, filled squares), although the potency of PTT was decreased compared to the pretreatment time of 1 hr (table 3).

The cocaine-like discriminative stimulus effects of 0.1 mg/kg PTT were evaluated at several different pretreatment times, in an effort to determine the time course of its discriminative stimulus effects (table 4). For all three animals, 0.1 mg/kg PTT resulted in >75% cocaine-appropriate responding within 10 to 30 min after administration. These effects were still apparent 8 hr after administration, in all three subjects. For monkey 9125, 0.1 mg/kg PTT still occasioned nearly 100% cocaine-appropriate responding 24 hr after administration. After 0.1 mg/kg PTT, response rates were significantly different as a function of pretreatment time [F(5, 12) = 4.47; P < .02], which was attributable to the lower rates of responding observed 5 hr after PTT administration. Saline, the PTT vehicle, and the training dose of cocaine (0.2 mg/kg) were tested 5 hr after administration and resulted in primarily saline-appropriate responding (table 4). Response rates after 0.2 mg/kg cocaine were not significantly affected by pretreatment time.

	Discussion

Top Abstract Introduction Methods Results Discussion References

One purpose of our study was to assess the effects of PTT and cocaine pretreatments on rates and intake of cocaine self-administration. We predicted that treatment with a dopamine agonist would decrease cocaine self-administration in a manner analogous to an opiate agonist decreasing heroin or alfentanil self-administration (e.g., Mello et al., 1983; Winger et al., 1992). When studied as a pretreatment to cocaine self-administration, PTT was at least 1 log unit more potent than cocaine at decreasing rates of cocaine-maintained responding and total session intake. This potency was achieved using the racemic mixture of PTT; an even higher potency would be expected if the active enantiomer were used in these studies (Bennett et al., 1995). In self-administration studies, PTT-maintained response rates were not different from saline-maintained response rates and were significantly lower than responding maintained by cocaine. Finally, in an animal model of subjective drug effects, PTT produced cocaine-like effects that persisted for at least 8 hr, and up to 24 hr in one subject. These results suggest that, on the one hand, PTT has abuse liability since it produces discriminative stimulus effects similar to cocaine; on the other hand, direct measures of the reinforcing effects of PTT in self-administration studies suggest that consumption and drug seeking may be substantially lower for PTT compared to cocaine.

Two general strategies have been implemented in searching for potential pharmacotherapies, one involving antagonists, the other using agonists. Regarding the former, there is clear evidence that dopamine D₁ and/or D₂ antagonists can block the reinforcing effects of cocaine, shifting the cocaine dose-response curve to the right (e.g., Bergman et al., 1990). However, there are some limitations to this approach, primarily that compliance is poor (e.g., Sherer et al., 1989) and that the presence of antagonists may, in fact, increase consumption. However, the use of agonists should decrease rates of cocaine self-administration, especially when studied on the descending limb of the cocaine dose-response curve. To date, agonist treatment of cocaine abuse has had limited clinical efficacy (e.g., Gawin et al., 1985; Jaffe et al., 1989; Preston et al., 1993).

In our study, PTT dose-dependently decreased cocaine-maintained response rates and total cocaine intake. These results with PTT are consistent with previous findings using the indirect DA agonists mazindol and GBR 12909 (Kleven and Woolverton, 1993; Mansbach and Balster, 1993; Skjoldager et al., 1993; Tella, 1995). There are several potential mechanisms for the reductions in cocaine-maintained responding by coadministration of other indirect DA agonists. For example, in a study of cocaine pharmacokinetics, Tella and Goldberg (1993) reported that pretreatment with GBR 12909, a compound that binds with high affinity to the DA transporter, enhanced plasma levels of cocaine following a 3.0 mg/kg i.v. bolus injection, although this effect was relatively brief. Rothman et al. (1991) reported that i.p. pretreatment with GBR 12909 produced a dose-dependent decrease in [³H]cocaine binding in rat caudate nuclei, and in microdialysis studies, GBR 12909 pretreatment resulted in a 50% reduction in cocaine-induced increases in extracellular DA, as measured in the striatum. Our findings demonstrate that the pretreatment administration of the indirect DA agonist PTT resulted in significant reductions in the rates of cocaine self-administration, consistent with a shift downward in the cocaine dose-response curve.

When substituted for cocaine, PTT did not function as a positive reinforcer. These results are in contrast with previous findings using the tropane analogs -CIT and -CFT (Spealman et al., 1991a; Weed et al., 1995), where high rates of self-administration were maintained. One possibility for the different outcomes is the base-line schedule of cocaine self-administration. Weed et al. (1995) used an FR 10 schedule of cocaine presentation, in which the minimum interinjection interval was not controlled by the investigator. In that study, -CIT injections maintained response rates that were approximately 50% lower than peak rates maintained by cocaine. However, Spealman and Kelleher (1981), using an FI 5-min schedule, reported that various cocaine analogs maintained response rates similar to those of cocaine. Thus, it appears that restricting the interinjection interval can result in increased rates of drug-maintained responding. These results are consistent with data reported by Winger (1993), in which increasing the interinjection interval resulted in increases in rates of drug self-administration. Thus, our findings showing that PTT does not function as a positive reinforcer when substituted for cocaine are not due to the use of an FI base-line.

It is possible that the low rates of PTT self-administration were due to the relatively long duration of action of PTT at inhibiting DA uptake. Porrino et al. (1994, 1995) have observed behavioral activation after PTT that persisted for more than 5 hr and Hemby et al. (1995), using in vivo microdialysis, reported significant elevation in levels of DA in the nucleus accumbens that persisted for more than 6 hr after 1.0 and 3.0 mg/kg PTT administration. The long duration of action for PTT raises the issue of whether the monkeys were titrating their intake. Thus, it is possible that PTT was highly reinforcing but because of its long duration of action and increased potency compared to cocaine, only low doses were needed to produce cocaine-like reinforcing effects. Future studies using discrete-trial choice, concurrent schedules or progressive-ratio schedules with cocaine and PTT will be necessary to specifically address the issue of reinforcing efficacy of PTT.

In the drug discrimination paradigm, PTT produced cocaine-like discriminative stimulus effects, and was approximately 0.5 to 1.0 log units more potent than cocaine. These results are consistent with those reported for other tropanes (Spealman et al., 1991a; Weed et al., 1995). In addition, consistent with its purported long duration of action, PTT occasioned cocaine-appropriate responding up to 24 hr after administration. This duration of action is three times longer than the maximal effect observed with -CIT, in which cocaine-appropriate responding was observed 8 hr after -CIT administration (Weed et al., 1995).

These results suggest that PTT has a unique profile of action relative to other indirect DA agonists, in that it produces cocaine-like discriminative stimulus effects but does not function as a reinforcer in substitution studies. For indirect DA agonists such as GBR 12909, -CFT and d-amphetamine, Spealman (1992) reported a high correlation between relative potency in self-administration studies and drug discrimination studies. However, he also noted that mazindol was an exception to this relationship, because it shares discriminative stimulus effects with cocaine, but is an equivocal reinforcer (cf. Spealman, 1992). In clinical trials, mazindol has been largely ineffective in maintaining cocaine abstinence (Diakogiannis et al., 1991). Thus, other agonists with unique profiles must continue to be evaluated. The use of indirect DA agonists may be a reasonable strategy in light of the growing evidence of the importance of the DA transporter in the behavioral effects of cocaine (Giros et al., 1996). Our results, using a compound that binds with high affinity and selectivity to the DA transporter and has a long duration of action, suggest that administration of PTT would produce cocaine-like subjective effects, as well as decrease cocaine self-administration in humans. Importantly, it appears that PTT has low abuse liability, indicating that it may be an excellent candidate for therapeutic intervention of cocaine dependence.