Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Cocaine abuse remains a significant public health problem in the United States. At present there are no effective medications for the treatment of cocaine dependence. However, growing understanding of the neurobiological mechanisms of cocaine's behavioral and subjective effects may facilitate the development of a successful pharmacotherapy for cocaine addiction (Johnson and Vocci, 1993; Rothman and Glowa, 1995). Considerable evidence has implicated mesolimbic dopamine (DA) neurotransmission in the psychomotor stimulant, reinforcing, and subjective effects of cocaine (Wise, 1984; Johanson and Fischman, 1989; Woolverton and Johnson, 1992). By binding to a site on the DA transporter to inhibit the neuronal reuptake of DA (Kennedy and Hanbauer, 1983; Reith et al., 1986; Madras et al., 1989; Carroll et al., 1992), cocaine elevates synaptic DA levels in both terminal field regions [nucleus accumbens (NAc) and prefrontal cortex] and cell body regions (ventral tegmental area) of mesolimbic DA neurons (Di Chiara and Imperato, 1988; Bradberry and Roth, 1989; Hurd and Ungerstedt, 1989; Mogaddam and Bunney, 1989). This action, particularly in the NAc, is thought to underlie acute cocaine-induced locomotor stimulation (Roberts et al., 1975; Kelly and Iversen, 1976; Di Chiara and Imperato, 1988; Giros et al., 1996) and cocaine self-administration in animals (Roberts et al., 1977; Wise, 1984; Ritz et al., 1987). In addition, recent studies have associated DA transporter occupancy (Volkow et al., 1997) and dopaminergic activity (Breiter et al., 1997) with the euphorigenic and other subjective effects of cocaine in humans. Because cocaine's action at the DA transporter may be central to its abuse liability, modulation of cocaine binding to the transporter represents one potentially important strategy in the development of a medication for cocaine dependence (Rothman, 1990; Witkin, 1994; Rothman and Glowa, 1995).

To this end, Rothman and colleagues have proposed the use of high-affinity DA uptake inhibitors as cocaine antagonists or substitution agents in human cocaine abusers (Rothman, 1990; Rothman et al., 1991; Baumann et al., 1994; Rothman and Glowa, 1995). Although the DA uptake inhibitors methylphenidate (Gawin et al., 1985) and mazindol (Preston et al., 1993; Stine et al., 1995) are of no reliable therapeutic benefit in cocaine addicts, recent preclinical studies suggest that the selective DA transporter inhibitor GBR 12909 can modulate the effects of cocaine on mesolimbic DA transmission in the rat in vivo (Rothman et al., 1991; Baumann et al., 1994; but see Gifford et al., 1993) and GBR 12909 or its derivatives can attenuate or delay cocaine self-administration in rats and primates (Glowa et al., 1996; Tella et al., 1996). Like most compounds that inhibit DA uptake with high affinity in vitro, GBR 12909 exhibits a cocaine-like behavioral profile in animals when administered alone (Heikkila and Manzino, 1984; Bergman et al., 1989; Spealman et al., 1989). Thus, it is perhaps not surprising that the closely related analog GBR 12935 can potentiate other effects of cocaine, including the locomotor stimulant, discriminative stimulus, and convulsive effects in rodents (Acri et al., 1996).

In contrast to cocaine, GBR 12909, and most other drugs that inhibit DA uptake with high affinity in vitro, recent studies of benztropine and structurally related phenyltropane analogs have identified at least one compound that binds to the DA transporter with high affinity in vitro but lacks cocaine-like behavioral effects in vivo (Newman et al., 1994, 1995; Acri et al., 1996; Kline et al., 1997). More potent at the DA transporter in vitro than cocaine (Newman et al., 1994) or its parent compound benztropine (van der Zee and Hespe, 1978; van der Zee et al., 1980), the 4'-substituted analog 4'-chloro-3-(diphenylmethoxy)tropane (4-Cl-BZT) is less efficacious as a locomotor stimulant than cocaine and does not substitute for cocaine as a discriminative stimulus across a range of doses (Newman et al., 1994; Kline et al., 1997). It is currently unknown whether the lack of cocaine-like behavioral efficacy for this compound results from a reduced ability to elevate mesolimbic DA in vivo via its interaction with the DA transporter (pharmacodynamic factors), from diminished or delayed uptake into brain (pharmacokinetic or dispositional factors), or from a potential functional antagonism exerted by additional site(s) of action.

In an attempt to better characterize the actions of 4-Cl-BZT in vivo, the present study was designed to address the following issues. First, after full dose-response characterization of the acute locomotor-stimulant effects of 4-Cl-BZT and cocaine, the locomotor response and the NAc DA response to selected doses of 4-Cl-BZT were evaluated simultaneously in freely moving rats and compared with responses to the same doses of cocaine. Because DA uptake inhibitors are known to vary widely in their rates of entry into the brain (Stathis et al., 1995), a variable that can have a profound influence on the behavioral and subjective effects of abused drugs (Sellers et al., 1989), the time course of both behavioral activation and neurochemical effects for each drug was carefully characterized and compared. Second, to address potential dispositional factors in the differences between 4-Cl-BZT and cocaine, the locomotor and NAc DA responses to systemic administration of each drug were compared with those after local administration of either drug directly into the NAc. Finally, the present study evaluated whether systemic 4-Cl-BZT administration could antagonize the neurochemical effects, the locomotor-stimulant effects, or the discriminative stimulus effects of a subsequent systemic cocaine challenge.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals and Drugs

Animals used in microdialysis, microinjection, and locomotor dose-response studies were housed at the San Francisco Veterans Affairs Medical Center animal care facility. Male Sprague-Dawley rats (Simonsen, Gilroy CA) weighing between 200 and 350 g were housed in pairs on a 12-h light/12-h dark cycle (lights on at 6:00 AM) and received food and water ad libitum. Rats used for microinjection and microdialysis experiments were housed individually after surgery. For cocaine discrimination studies, male Sprague-Dawley rats (Charles River, Wilmington, MA) weighing 310 to 385 g were housed individually at the National Institute on Drug Abuse Intramural Research facility with free access to water under a 12-h light/12-h dark cycle (lights on at 7:00 AM). Rats were fed daily 15 g of Purina chow 30 min after testing. Cocaine hydrochloride (Sigma, St. Louis, MO) was dissolved in isotonic saline for i.p. injections and in artificial cerebrospinal fluid (aCSF: 125 mM NaCl, 0.5 mM NaH₂PO₄, 2.5 mM Na₂HPO₄, 2.5 mM KCl, 1.2 mM CaCl₂, 1 mM MgCl₂, pH 7.4) for microinjections and microdialysis studies. 4-Chlorobenztropine (van der Zee et al., 1980; Newman et al., 1994) was dissolved with extensive sonication in saline for i.p. injections and in aCSF for microinjection/microdialysis studies.

Locomotor Activity

Test chambers (Opto-Varimex Minor, Columbus Instruments, Columbus, OH) measured 50 × 50 cm and were equipped with 15 photocell beams in each direction located 4 cm off the cage floor. Interruptions of photocell beams were cumulatively computer-registered in 10-min intervals (15-min in microdialysis groups). In all test groups, rats were allowed to habituate to the test chamber for 1 h while being monitored for baseline locomotor activity. At the end of this period, rats received drug treatments (see Experimental Design section below) and were monitored further for locomotor activity and stereotypy for at least 2 h. Frequency and intensity of stereotyped behavior was scored by a trained observer (often, but not in all cases, blind to the treatment conditions) for 10 s every 10 min according to the following scale (Steketee and Kalivas, 1991): 1) asleep or still; 2) inactive, grooming, mild licking; 3) locomotion, rearing, or sniffing; 4) any combination of two of locomotion, rearing, or sniffing; 5) continuous sniffing for 10 s without locomotion or rearing; 6) continuous sniffing for 10 s with locomotion or rearing; 7) patterned sniffing for 5 s; 8) patterned sniffing for 10 s; 9) continuous gnawing; 10) bizarre diskinetic movements or seizures. Scored behavior is presented as the percentage of animals exhibiting stereotypy as defined for a given experiment (scores of 5 or higher during any 10-min interval for single drug treatments, and scores of 8 or higher during any 10-min interval for drug combination studies employing a high challenge dose of cocaine). The threshold defining stereotypy was raised in drug combination studies to allow comparison of pretreatment effects; such comparisons were impossible using the lower-score criteria because all animals exceeded scores of 5 regardless of pretreatment in these experiments.

Cocaine Discrimination

Rats were tested in operant-conditioning chambers (modified Med Associates model ENV 007; St. Albans, VT) housed within light- and sound-attenuating enclosures with white noise present throughout testing. Ambient illumination was provided by lamps mounted at the top of the front panel. Two response keys (levers) were set 17 cm apart, with three stimulus lights above each. A force of 0.4 N through 1 mm was required to register a response, and each response produced an audible click from a relay mounted behind the front panel of the chamber. Reinforced responses produced one 45-mg pellet (BioServe, Frenchtown, NJ) delivered from a dispenser mounted behind the front panel into a food tray located centrally between the response keys. On-line experimental control and data collection were by MS-DOS computers operating Med Associates software (Med Associates).

Rats initially were trained to press both keys under a fixed-ratio schedule of food reinforcement. Responding on each key was trained separately in a mixed order, with the active key on a given training session indicated by illumination of the lamps directly above it. Rats subsequently were trained to discriminate i.p. injections of cocaine (10 mg/kg) from i.p. injections of saline. After cocaine injections, responses on only one key were reinforced; after saline injections, responses on the alternate key were reinforced. The assignment of cocaine- and saline-appropriate keys was counterbalanced across rats. Immediately after injection, rats were placed inside the experimental chambers and a 5-min time-out period was initiated, during which all stimulus lamps were extinguished and responding produced feedback clicks but had no other scheduled consequences. All lamps then were illuminated and responses on the appropriate key were reinforced. The fixed-ratio (FR) value was increased to 10 or 20 (depending on the experiment; see figure legends) over several training sessions. Responses on the inappropriate key reset the FR-response requirement on the appropriate key. Each food presentation was followed by a 20-sec time-out period during which all lamps were off, and responding had no scheduled consequences other than the feedback clicks. Sessions ended after 20 food presentations or 15 min, whichever occurred first.

As the FR value reached 20, training sessions for which cocaine (C) and saline (S) injections were administered were ordered in a CSSCCS. . . sequence, with test sessions conducted after consecutive SC or CS training sessions. On test sessions, different doses of cocaine, 4-Cl-BZT, or their combination were substituted for cocaine or saline. A test session was conducted if the subject achieved criteria on both of the immediately preceding saline and cocaine training sessions. The criteria were at least 80% cocaine- or saline-appropriate responding overall and during the first FR of the session. Test sessions were identical with training sessions, with the exception that 20 consecutive responses on either key were reinforced.

Microdialysis

All animals were experimentally naive, and microdialysis probes (CMA/12; CMA/Microdialysis, Acton MA) were unused before surgical procedures and testing. Rats were anesthetized with ketamine/xylazine and placed in a stereotaxic instrument (Kopf, Tujunga, CA). A new 2-mm microdialysis probe was implanted into the left NAc [A/P, +1.0 mm; M/L: +1.3 mm; D/V, 8.3 mm from bregma according to the atlas of Paxinos and Watson (1986)] and cemented in place using three sterile skull screws and dental acrylic. On the following day (18-24 h after surgery) rats were fitted with collars and tethers, placed in the test chambers described above, and connected to a perfusion line that delivered aCSF at 2 µl/min using a syringe pump (CMA 100; CMA/Microdialysis). Rats were allowed to move freely while perfusate was collected through a two-channel liquid swivel attached to a free-standing pendulum and then into an automated sample collector (CMA 140; CMA/Microdialysis). Animals were preperfused for at least 30 min before testing began; 30 to 90 min after being connected to the perfusion lines, collection of baseline behavioral data and corresponding perfusate samples began. Behavioral data and perfusate samples were collected simultaneously in 15-min intervals throughout the experiment as described below. Comparisons of behavioral and microdialysis time courses took into consideration a 15- to 20-min time lag in DA response (or in drug delivery for studies of drugs perfused locally through the microdialysis probe) because of dead space in the collection tubing. Perfusates were stored at 20°C until assayed (within 1 week of collection). All rats were euthanized for histological analysis of the probe and tract locations upon completion of the experiment.

HPLC

Perfusate sample aliquots (20 µl) were analyzed for DA content by reverse-phase HPLC connected to an electrochemical detector (LC-4B; Bioanalytical Systems, West Lafayette, IN) equipped with a dual glassy carbon electrode set at 0.65 V. The mobile phase (0.06 M Na₂HPO₄, 0.09 M EDTA, and 1.3 mM 1-octanesulfonic acid in an 18% methanol solution adjusted to pH 2.95 with phosphoric acid) was pumped (model 510; Waters, Milford, MA) at 0.6 ml/min through a reversed-phase, ion-pairing column (100 × 3.2 mm) prepacked with Phase II octadecasilane 3-µm particulate at 0.6 ml/min (model 6213, Bioanalytical Systems). Retention times and concentrations of DA, 3,4-dihydrophenylacetic acid, and 5-hydroxyindoleacetic acid in dialysates were determined by comparisons with individual standards and standard mixtures of known concentration run daily before and after microdialysis samples. Linearity of standard curves was observed over a wide range of concentrations (0.01-200 pmol). Chromatographic data were recorded using a dual pen chart recorder (BD 41; Kipp and Zonen, Bohemia, NY) with signal amplitudes set at 0.5 and 5 V. The limit of detection for DA was approximately 10 fmol.

Intracranial Microinjections

Rats were anesthetized with ketamine/xylazine and placed in a stereotaxic instrument. Bilateral guide cannulas (22-gauge; Plastics One, Roanoke, VA) were implanted into the NAc (A/P, +1.0 mm; M/liter, ±1.25 mm; D/V, 6.2 mm) and cemented in place using dental cement. Bilateral 26-gauge obturators were inserted into the guide cannulas, and rats were allowed to recover for at least 1 week with daily habituation to handling. On the test day rats received bilateral microinjections and were monitored for locomotor activity as described below (see Experimental Design). All rats were euthanized for histological analysis of tract locations upon completion of the experiment.

Histology

Upon completion of microinjection/microdialysis experiments, animals were overdosed with ketamine (100 mg/kg), perfused via the ascending aorta with 4% paraformaldehyde in 0.1 M phosphate buffer, and decapitated. Whole brains were removed and stored in phosphate buffer until sectioned using a vibratome (Lancer, St. Louis, MO). Coronal sections (100 µm) were mounted on gel-coated slides, stained with 2% cresyl violet, and examined under a light microscope. Microinjection tract location and probe placement were determined according to the atlas of Paxinos and Watson (1986).

Experimental Design

Five experiments were designed and conducted as follows.

Experiment 1. The effects of i.p. 4-Cl-BZT, cocaine, or the combination of the two drugs on locomotor activity were evaluated as follows. For studies of each drug alone, rats were habituated to the test chambers for 1 h and then were injected with saline (1 ml/kg), cocaine (1, 10, 20, 40, or 50 mg/kg i.p.), or 4-Cl-BZT (1, 10, 20, 40, or 60 mg/kg i.p.) and monitored for locomotor activity for 2 h. For combination groups, after 1 h habituation rats were injected with saline or 4-Cl-BZT (20 or 40 mg/kg i.p.) and returned to the test chamber. Thirty minutes later, all rats were challenged with 40 mg/kg cocaine and were monitored for locomotion, stereotypy, and convulsions for 2 h.

Experiment 2. The effects of i.p. 4-Cl-BZT, cocaine, and the combination of the two drugs on NAc DA were evaluated simultaneously with locomotor activity in freely moving rats. Baseline dialysate samples were collected every 15 min while rats were habituated to the test chambers for 1 h. After i.p. injection of saline, 4-Cl-BZT (10 or 40 mg/kg), or cocaine (40 mg/kg) dialysates and behavioral measures were collected every 15 min for 2 h. All rats then were challenged with 10 mg/kg cocaine, and dialysates and behavioral measures were collected for another 2 h.

Experiment 3. The effects of unilateral intra-accumbens 4-Cl-BZT or cocaine perfusion through the microdialysis probe on extracellular DA were evaluated simultaneously with locomotor activity in freely moving rats. Baseline perfusate samples were collected every 15 min while rats were habituated to the test chambers for 1 h. After habituation, cocaine or 4-Cl-BZT (1, 10, or 100 µM in aCSF) was perfused locally through the microdialysis probe for 1 h while samples were collected every 15 min. Artificial CSF alone then again was perfused and samples were collected every 15 min for 3 to 5 h.

Experiment 4. The effects of bilateral microinjections of 4-Cl-BZT or cocaine into the NAc on locomotor activity were compared with respect to total locomotion and time course of activity. Doses were chosen based on previous studies of intra-accumbens cocaine (Delfs et al., 1990; Jones et al., 1994). On the test day rats were habituated to the test chamber for 1 h, obturators were removed, and 26-gauge injection cannulas were inserted into the guide cannulas. Microinjections of aCSF (0.5 µl/side), cocaine (30 or 300 nmol/0.5 µl/side), or 4-Cl-BZT (30 or 300 nmol/0.5 µl/side) were made over 60 s using a bilateral polyethylene connector (Plastics One) and two 1-µl Hamilton syringes. An additional 60 s was allowed for drug diffusion, injection cannulas were removed, and obturators were replaced before monitoring locomotor activity for 2 h.

Experiment 5. The effects of 4-Cl-BZT, cocaine, or the combination of the two drugs on rats discriminating cocaine injections were evaluated as follows. Daily sessions were conducted and performances were evaluated to assess whether subjects met the testing criteria. Rats were administered the drug i.p. and placed in the darkened experimental chamber. Responses during an initial 5-min time-out had no scheduled consequences. After the time-out, completion of the FR requirement on either lever produced a food pellet followed by the 20-s time-out period. Test sessions were conducted with saline (1 ml/kg i.p.), cocaine (1.0, 3.0, 10.0 mg/kg i.p.), 4-Cl-BZT (1.0, 2.5, 5.0, 10.0, 25.0 mg/kg i.p.), or cocaine (0.1, 0.3, 0.56, 1.0, 3.0, 10.0 mg/kg i.p.) preceded by 4-Cl-BZT (3.0 and 10.0 mg/kg i.p.). The distribution of responses on the two levers was monitored until 20 food pellets were presented or for 15 min, whichever occurred first.

Statistical Analysis

Effects of drug and dose on cumulative locomotor responses were analyzed by one-way and two-way between-subject ANOVA followed by Newman-Keuls and Dunnett's post hoc tests. Time course locomotor and neurochemical data were analyzed by mixed factorial (between subject = drug or dose, within subject = time) ANOVA. Secondary analysis of time course interactions compared the two drugs for onset and duration of action. Onset of action was analyzed by one-way ANOVA comparisons of time to peak effects, and duration of action was quantified for behaviorally active doses of each drug as mean time required for dissipation to 50% of peak effects. Stereotypy data were analyzed by Kruskal-Wallis and Mann-Whitney U tests and confirmed by one-way ANOVA. Basal DA data were analyzed as raw values (fmol/20 µl). Because of interindividual differences in basal DA efflux within treatment groups, microdialysis data were converted to percentages of control values, defined as the mean extracellular DA content measured in the four samples before drug administration. Statistical comparisons for drug effects on neurochemical measures were made using percent control values and subsequently were confirmed using raw data (fmol/20 µl).

For cocaine discriminative stimulus studies, the overall response rate and the percentage of responses occurring on the cocaine-appropriate lever were calculated for each subject and mean values were determined for each measure at each drug dose. Data from any rat that failed to produce at least 20 responses were not included in the calculation of mean cocaine-appropriate responding at that dose. If fewer than three rats met the response rate requirement, no mean value was calculated for percentage of cocaine-appropriate responding at that dose. Standard ANOVA and linear regression techniques were used to calculate mean effective dose (ED₅₀) values and their 95% confidence limits (Snedecor and Cochran, 1967). A significant difference in ED₅₀ values is indicated when the 95% confidence limits do not overlap. To assess relative potency of cocaine in 4-Cl-BZT-treated subjects, the dose-effect data also were analyzed by standard parallel-line bioassay techniques as described by Finney (1964). This analysis involves a one-way ANOVA that determines whether the slopes of the two dose-response curves are different from parallel and fits a common slope to the two dose-response curves. It then compares the ratio of doses for a given effect (in this case a 50% substitution for the training dose of cocaine) to provide a value for relative potency. This value represents the dose of cocaine in mg/kg for 4-Cl-BZT-treated rats equal to 1 mg/kg cocaine in subjects not pretreated. A significant relative potency difference is indicated when the 95% confidence limits for that ratio do not include 1.0.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Locomotor Activity: Intraperitoneal Dose Response. Although both cocaine [F(5,30) = 19.24, p < .0001] and 4-Cl-BZT [F(6,33) = 4.19, p < .005] dose dependently stimulated locomotor activity when administered i.p., cocaine exhibited a lower threshold dose for behavioral activity [drug × dose interaction F(3,43) = 4.74, p < .007] and was clearly more efficacious as a locomotor stimulant than 4-Cl-BZT across the broad range of doses tested [drug effect F(1,53) = 7.29, p < .01; Fig. 1A]. At 80 mg/kg, twice the maximally active dose of cocaine, 4-Cl-BZT produced marked locomotor stimulation in only one of six rats; the remaining five animals in this dose group exhibited significantly less locomotion than the mean response to 40 mg/kg of cocaine. Analysis of scored behavior revealed no significant differences between cocaine and 4-Cl-BZT in the induction of stereotypy [Mann-Whitney U test, p > .05; F(1,56) = .40, p = 0.53; Fig. 1B]] The two drugs displayed markedly different kinetic profiles with respect to their locomotor stimulant effects when injected i.p. [drug × time interaction: F(34,1139) = 7.30, p < .0001]. Analysis of this interaction revealed a significantly longer time to peak stimulant effect for 4-Cl-BZT [58 ± 7 min versus 26 ± 3 min for cocaine across all active doses; F(1,45) = 16.81, p < .0002; Fig. 1C]. This was true at all doses [no dose effect, F(5,39) = 1.58, p = 0.19 or drug × dose interaction, F(1,39) = .20, p = 0.66]. In addition, the duration of action of 4-Cl-BZT was longer than that of cocaine. Whereas the locomotor stimulant effect of cocaine rapidly dissipated to 51% of peak effect by 60 min postinjection, 4-Cl-BZT continued to stimulate locomotion to 83% of its peak effect up to 120 min postinjection, when testing ended (Fig. 1C).

View larger version (27K): [in this window] [in a new window]   Fig. 1.   Effects of five doses of cocaine and six doses of 4-Cl-BZT on locomotor activity and scored behavior in 2 h after i.p. injection. Data are presented as mean ± S.E.M., n = 5 to 6 each dose for A-C. A, log dose response for cumulative locomotion totaled over 2 h after injection. Cocaine produced significant locomotor stimulation at 10, 20, 40, and 50 mg/kg (*p < .05 relative to saline, Dunnett's). Significant locomotor stimulation was produced by 4-Cl-BZT at 20, 40, 60, and 80 mg/kg (*p < .05, Dunnett's). B, percentage of rats that exhibited stereotypy [interval score of 5 or higher using the scale of Steketee and Kalivas (1992)] in the same experimental groups described in A. No significant differences between cocaine and 4-Cl-BZT as determined by Mann-Whitney U test (p > .05) or ANOVA (p = 0.48). C, time course of locomotor stimulation induced over 12 10-min intervals by maximally active doses of i.p. cocaine or 4-Cl-BZT in groups depicted in A. Significant drug × time interaction (p < .0001).

NAc DA: Intraperitoneal Administration. The above behavioral results of i.p. 4-Cl-BZT and cocaine administration were paralleled by the effects of i.p. administration of the two drugs on extracellular DA in the NAc. Whereas basal DA did not differ among i.p. treatment groups [F(3,14) = .13, p = 0.94; see Fig. 2 legend for absolute values], a significant effect of drug challenge on NAc DA over 2 h was present [F(3,144) = 13.91, p < .0001; Fig. 2A, points after the first injection]. Secondary analysis revealed that 4-Cl-BZT significantly elevated NAc DA relative to saline at 40 mg/kg i.p. [F(1,72) = 25.14, p < .0007; Fig. 2A () after the first injection] but not at 10 mg/kg [F(1,72) = .25, p = 0.63; Fig. 2A () after the first injection]. Within-subject comparison revealed that 10 mg/kg 4-Cl-BZT also had no significant effect on DA relative to baseline [F(8,32) = 1.13, p = 0.37; Fig. 2A, compare  before and after the first injection]. Although cocaine was not administered at 10 mg/kg as a first injection in the i.p. microdialysis experiment, this dose of cocaine did elevate NAc DA significantly above baseline (190% control) in rats that had been treated 2 h previously with i.p. saline [F(8,32) = 7.72, p < .001; Fig. 2A () after the second injection]. At 40 mg/kg, cocaine produced a pronounced and rapid elevation of NAc DA relative to saline [F(1,80) = 24.83, p < .0006; Fig. 2A, compare  and  after the first injection]. Averaged over the full 2-h period after injection, the NAc DA response to 40 mg/kg cocaine was not significantly different from that of 40 mg/kg 4-Cl-BZT (Newman-Keuls, p > .05). However, the time course of action of cocaine was significantly different from that of 4-Cl-BZT [drug × time interaction F(24,144) = 8.12, p < .0001], and the peak DA overflow after 4-Cl-BZT (245%) did not reach that of cocaine (370%, Fig. 2A, compare  and  after the first injection).

View larger version (35K): [in this window] [in a new window]   Fig. 2.   Effects of i.p. 4-Cl-BZT or cocaine, alone and in combination, on extracellular DA levels in the left NAc as measured by in vivo microdialysis (A) and on locomotor activity in the same rats (B). All data are presented as mean ± S.E.M., n = 5 to 6 for each group, for each 15-min collection interval. A, arrows 1 and 2 represent the first and second injections, respectively. For the first injection, rats received saline, 40 mg/kg cocaine, or 4-Cl-BZT at 10 or 40 mg/kg. The second injection was a 10-mg/kg cocaine challenge administered to all four groups 2 h after the first injection. Percentage of baseline data were calculated at all time points for each animal using the four 15-min baseline DA samples collected before the first injection. No differences in basal DA were found among treatment groups (saline group, mean = 100 ± 7 fmol/µl; 10 mg/kg 4-Cl-BZT, 121 ± 22 fmol/µl; 40 mg/kg 4-Cl-BZT, 109 ± 15 fmol/µl; 40 mg/kg cocaine, 94 ± 7 fmol/µl). Significant elevation of NAc DA levels after the first injection was found for 40 mg/kg cocaine (p < .001) and 4-Cl-BZT at 40 mg/kg (p < .001) but not at 10 mg/kg (p = 0.63) relative to saline. Upon second injection, NAc DA levels were elevated more in rats pretreated with 40 mg/kg 4-Cl-BZT than in all other pretreatment groups (Newman-Keuls, p < .05) in response to 10 mg/kg cocaine. B, Time course of locomotor response to the two injections described above. No differences in basal locomotion among groups were found (p = 0.30). Only 40 mg/kg cocaine (Newman-Keuls, p < .05) and 40 mg/kg 4-Cl-BZT (Newman-Keuls, p < .05) were able to stimulate locomotor activity after the first injection. Pretreatment with 40 mg/kg 4-Cl-BZT significantly potentiated the locomotor response to cocaine challenge relative to saline (Dunnett's, p < .05).

NAc DA: Local Administration. No differences in basal DA levels among the NAc drug treatment groups were detected [F(2,35) = .68, p = 0.51]. Perfused locally through the microdialysis probe, both 4-Cl-BZT [F(3,304) = 3.95, p < .03] and cocaine [F(3,266) = 9.32, p < .005] dose dependently elevated DA levels in the left NAc (Fig. 3, A and B). Secondary analysis revealed that local 4-Cl-BZT stimulated NAc DA overflow at 10 µM [F(1,152) = 284.77, p < .0001] and at 100 µM [F(1,171) = 5.34, p < .05], but not at 1 µM [F(1,171) = 1.21, p = 0.30]. Similarly, locally perfused cocaine elevated DA at 10 µM [F(1,171) = 10.88, p < .01] and at 100 µM [F(1,190) = 20.49, p < .002], but not at 1 µM [F(1,110) = 1.19, p = 0.31]. Post hoc comparisons at each active dose revealed significantly greater elevation of accumbens DA by 4-Cl-BZT than by cocaine at the 100-µM but not at the 10-µM dose (Newman-Keuls, p < .05). As was the case with peripheral administration, the onset and duration of local cocaine's effects on NAc DA were significantly shorter than those of locally perfused 4-Cl-BZT [drug × time interaction: F(19,437) = 3.79, p < .0001]. Whereas cocaine-induced DA overflow peaked by the third sample after beginning infusion and remained elevated for only 1 h after removal of cocaine, local 4-Cl-BZT was maximally active 2 h from the start of infusion and NAc DA levels remained elevated above baseline for more than 3 h after cessation of 4-Cl-BZT infusion (Fig. 3, A and B). Neither cocaine nor 4-Cl-BZT induced locomotor activity at any dose when perfused locally through the microdialysis probe [F(2,360) = .953, p = 0.40; Fig. 3, C and D].

View larger version (35K): [in this window] [in a new window]   Fig. 3.   Effects of three doses of 4-Cl-BZT or cocaine on NAc DA levels (A and B) and locomotor activity (C and D) when perfused locally into the left NAc through the microdialysis probe. The bar represents the time of drug administration. All data are presented as mean ± S.E.M., n = 5 to 7 for each group. No differences in basal DA were found among any of the seven treatment groups (p = 0.51). Although both cocaine (A) and 4-Cl-BZT (B) dose dependently elevated NAc DA at 10 and 100 µM, the maximal effect of 4-Cl-BZT was nearly 5-fold greater than that of cocaine (note the different ordinate scales in A and B representing maximal DA responses) at 100 µM but not at 10 µM (Newman-Keuls, p < .05). In addition, the extracellular DA response to 4-Cl-BZT was significantly prolonged relative to cocaine [drug × time interaction F(19,437) = 3.79, p < .0001]. Neither cocaine (C) nor 4-Cl-BZT (D) stimulated locomotor activity at any dose when perfused locally through the probe.

Bilateral NAc Microinjections. No significant differences in basal locomotion were found among drug treatment groups [F(2,28) = 2.054, p = 0.15; Fig. 4A, preinjection data points]. A significant effect of drug was present [F(2,28) = 3.79, p < .05], with microinjections of either 4-Cl-BZT [F(2,15) = 5.23, p < .02] or cocaine [F(2,14) = 4.11, p < .05] eliciting locomotor stimulation (Fig. 4A). Post hoc analysis revealed no differences between doses of either cocaine or 4-Cl-BZT (Newman-Keuls, p > .05). Secondary comparisons of 4-Cl-BZT and cocaine revealed no differences between drugs in cumulative locomotor stimulation at either the 30-nmol/side [F(1,10) = 1.43, p = 0.26] or the 300-nmol/side [F(1,12) = .323, p = 0.58] doses. However, as was the case with peripheral administration, intra-accumbens cocaine and 4-Cl-BZT exhibited significantly different time courses of action at both doses [drug × time interaction: F(22,308) = 9.55, p < .0001]. Secondary analysis revealed that the locomotor response to either dose of cocaine was markedly higher than to either dose of 4-Cl-BZT in the first 30 min after microinjection (Dunnett's, p < .05). No significant differences among all treatment groups were uncovered at any other time point (Fig. 4B).

View larger version (30K): [in this window] [in a new window]   Fig. 4.   Time course (A) and cumulative (B) effects of bilateral microinjection of 30 or 300 nmol cocaine or 4-Cl-BZT into the NAc on locomotor activity. Data are presented as mean ± S.E.M., n = 6 for each group. Time course data were collected in 10-min bins for 1-h preinjection and 2-h postinjection. No differences in basal locomotion among treatment groups was observed (p = 0.15). Compared across 2 h, no differences in locomotor efficacy were found between cocaine and 4-Cl-BZT at either the 30-nmol/side (p = 0.26) or the 300-nmol/side (p = 0.58) doses. However, the two drugs exhibited strikingly different temporal profiles of action [drug × time interaction: F(22,308) = 9.55, p < .0001].

Drug Combination Studies. In rats simultaneously assessed for locomotor activity and extracellular DA, 4-Cl-BZT pretreatment potentiated the NAc DA and locomotor responses to a low dose (10 mg/kg i.p.) of cocaine administered 2 h later (Fig. 2, points after the second injection). Post hoc analysis of the significant effect of pretreatment on the DA response to 10 mg/kg cocaine [F(3,120) = 8.17, p < .002] revealed that pretreatment 2 h earlier with 40 mg/kg 4-Cl-BZT resulted in significantly higher NAc DA levels upon cocaine injection relative to pretreatment with saline (Fig. 2A, compare  with  after the second injection), 10 mg/kg 4-Cl-BZT (), or 40 mg/kg cocaine () (Newman-Keuls, p < .05). Pretreatment with 10 mg/kg 4-Cl-BZT or cocaine did not alter the NAc DA response to 10 mg/kg cocaine relative to saline pretreatment (Newman-Keuls, p > .05).

View this table: [in this window] [in a new window]   TABLE 1 Extracellular DA and locomotor responses to 10 mg/kg i.p. cocaine challenge in subjects pretreated with 4-Cl-BZT or cocaine Effects of i.p. pretreatment (2 hr before) on extracellular DA and locomotor response to a challenge injection of 10 mg/kg i.p. cocaine. Values for extracellular DA concentrations in dialysate fractions from NAc are means (±S.E.M.) and are expressed as fmol/20-µl sample. Values for locomotor activity are mean (±S.E.M.) photocell counts/15 min. Final preinjection values (column A) were collected during last time period before cocaine challenge and are reproduced from Fig. 2. Peak postinjection data (column B) represent highest absolute values after the 10-mg/kg cocaine injection (generally 15-45 min postinjection). Absolute change values (column C) represent difference between final preinjection and peak postinjection values. Percent change values (column D) were calculated for each rat as (postinjection  preinjection)/preinjection × 100. For all values, calculations were made for each individual animal and then averaged within each treatment group. The () sign associated with locomotor change in cocaine-pretreated rats indicates an average net lower locomotion after 10 mg/kg cocaine challenge than immediately before.

View this table: [in this window] [in a new window]   TABLE 2 Effects of 4-Cl-BZT pretreatment on behavioral and toxic properties of cocaine Locomotor data are presented as mean ± S.E.M. cumulative scores over 2 h after injection of 40 mg/kg cocaine. Percent stereotypy data represent percentage of animals that were scored as level 8 or higher according to scale of Steketee and Kalivas (1991) in any one 10-min interval over 2 h after cocaine injection. Convulsion data indicate ratio of rats exhibiting seizures after 40 mg/kg cocaine.

Cocaine Discrimination. As has been shown previously, subjects trained to discriminate cocaine (10.0 mg/kg) readily acquired the discriminative performance with greater than 80% of responses on the cocaine-appropriate lever after an injection of the training dose of cocaine and less than 20% of the responses on the cocaine-appropriate lever after saline injection (Fig. 5, unconnected points above C and S, respectively). Substitutions of varying doses of cocaine produced a dose-related increase in the percentage of responses on the cocaine-appropriate lever up to full substitution for cocaine at doses of 5.6 and 10.0 mg/kg (Fig. 6, ). In contrast to the dose-related increase in cocaine-appropriate responding with cocaine, none of the doses of 4-Cl-BZT produced a maximum level of drug-appropriate responding that exceeded saline levels (Fig. 5, upper graph, connected points). This lack of substitution for cocaine occurred despite the study of a range of doses of 4-Cl-BZT, from those having no effect to those that substantially decreased response rates. Similar results have been reported previously for this compound (Newman et al., 1994). When studied in combination with cocaine, 4-Cl-BZT shifted the cocaine dose-effect curve to the left (Fig. 6, ). The shift in the cocaine dose-effect curve was dose-related, with the higher dose of 4-Cl-BZT shifting the curve further to the left. The change in potency of cocaine with 4-Cl-BZT pretreatments is quantified further with dose-related changes in the ED₅₀ value for cocaine (Table 3). As can be seen in the table, doses of 3.0 and 10.0 mg/kg 4-Cl-BZT significantly decreased the ED₅₀ value from 3.5 mg/kg to 1.7 and 0.15 mg/kg, respectively. These changes in ED₅₀ values represent 1.7- and 16.2-fold increases in the potency of cocaine (Table 3).

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Effects of 4-Cl-BZT in rats trained to discriminate injections of cocaine from saline under a fixed-ratio 10-response schedule of reinforcement. Abscissae: drug dose in mg/kg (log scale). Ordinate (upper): percentage of responses emitted on the lever on which rats were trained to respond after injections of cocaine. Each point represents the effect in four to six rats. Ordinate (lower): rates at which responses were emitted as a percentage of response rate after saline administration. Each point represents the effect in six rats. Points above S represent control data after saline injections. Points above C represent control data after cocaine injections.

View larger version (19K): [in this window] [in a new window]   Fig. 6.   Effects of 4-Cl-BZT on the discriminative-stimulus effects of cocaine in rats trained to discriminate injections of cocaine from saline. Abscissae: drug dose in mg/kg (log scale). Ordinate (upper): percentage of responses emitted on the lever on which rats were trained to respond after injections of cocaine. Each point represents the effect in four to six rats. Ordinate (lower): rates at which responses were emitted as a percentage of response rate after saline administration. Each point represents the effect in six rats responding under a fixed-ratio 20-response schedule of reinforcement. , effects of cocaine when administered alone. , effects of cocaine after administration of 3.0 mg/kg 4-Cl-BZT. , effects of cocaine after administration of 10.0 mg/kg 4-Cl-BZT.

View this table: [in this window] [in a new window]   TABLE 3 Potency of cocaine for production of discriminative stimulus effects when administered alone and in combination with 4-Cl-BZT Effects of pretreatment with 4-Cl-BZT (with both cocaine and 4-Cl-BZT administered 5 min before testing) on discriminative stimulus effects of cocaine in rats trained to discriminate injections of 10 mg/kg cocaine from saline. Values are ED50 values and their 95% confidence limits for percentage of responses occurring on cocaine-appropriate lever and were calculated using standard ANOVA and linear regression techniques (Snedecor and Cochran, 1967; see Materials and Methods). A significant difference in ED50 values is indicated when 95% confidence limits do not overlap. Relative potency of cocaine in 4-Cl-BZT subjects was assessed by standard parallel-line bioassay techniques as described by Finney (1964) (see Materials and Methods). This value represents dose of cocaine in mg/kg for 4-Cl-BZT-treated rats equal to 1 mg/kg cocaine in subjects not pretreated.

Histology. Representative tract placement and tissue damage associated with NAc microdialysis or microinjections are depicted in Fig. 7. Tissue damage was confined primarily to the regions injured mechanically during the surgical procedures. As determined by microscopic examination and counting of neurons in 100-µm coronal sections stained with cresyl violet, no obvious neurotoxicity was observed in rats infused with either cocaine or 4-Cl-BZT relative to saline or aCSF controls (data not shown).

View larger version (60K): [in this window] [in a new window]   Fig. 7.   A, tract locations of microdialysis probes (vertical bars) and microinjection cannulas () in the NAc of selected rats used in local administration studies. Probe tracts are from rats that received aCSF or 100 µM 4-Cl-BZT and are representative of those of accurate placement in other treatment groups. Microinjection tracts are from rats that received aCSF or 300 nmol/side 4-Cl-BZT and are representative of those of accurate placement in other treatment groups. Animals with tract placements outside the NAc were excluded from the behavioral analysis. Although tract placements are represented schematically as occurring in a single anterior-posterior plane, tracts that met medial-lateral and dorsal-ventral criteria were accepted as accurate from approximately 1.7 to 0.7 mm from bregma according to the atlas of Paxinos and Watson (1986). Photomicrographs (×40) of tract locations and tissue damage associated with a 2-mm microdialysis probe delivering 100 µM 4-Cl-BZT (B) or microinjection cannulas delivering aCSF (C) or 4-Cl-BZT (D) at 300 nmol/side. In each case, tissue was harvested 18-24 h after behavioral testing.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The current results confirm and extend previous reports that 4-Cl-BZT lacks a cocaine-like behavioral profile despite its high affinity for the DA transporter and its potency in inhibiting DA reuptake in vitro (Newman et al., 1994, 1995; Kline et al., 1997). As in previous studies demonstrating weak locomotor stimulant efficacy at early time points (0-30 min) in mice (Newman et al., 1994; Kline et al., 1997), 4-Cl-BZT was found in the current study to induce only modest locomotor activation in rats across a wide range of doses for up to 2 h postinjection (Fig. 1A). Although this behavioral activation appeared to differ from that of cocaine quantitatively rather than qualitatively, the temporal profiles of the two drugs' locomotor stimulant effects were clearly distinct. Unlike the rapid onset and dissipation of pronounced locomotor stimulation by cocaine, the response to active doses of 4-Cl-BZT remained virtually unchanged throughout the 2-h test period (Fig. 1B). The lower locomotor stimulant efficacy of 4-Cl-BZT was not the result of greater induction of stereotypy relative to cocaine (Fig. 1C); like its parent compound benztropine (Scheel-Kruger, 1972), 4-Cl-BZT produced less stereotypy than equal doses of cocaine across all time points of the 2-h test session (data not shown). Thus, other mechanisms must account for the low behavioral efficacy of systemic 4-Cl-BZT. Previously, it has been unknown whether the ability of 4-Cl-BZT to inhibit DA uptake in vitro is conserved in vivo, whether the interaction of 4-Cl-BZT with the DA transporter is analogous to that of a partial agonist (as defined by its effects on locomotor activity), or whether penetration of 4-Cl-BZT into the brain may be poor or delayed relative to cocaine, resulting in insufficient brain levels for neurochemical and behavioral stimulant effects. Each of these mechanisms/factors is addressed by the present results.

Systemic versus Local Administration. As depicted in Fig. 2, peripherally administered 4-Cl-BZT clearly is able to inhibit DA uptake in vivo, as demonstrated by its dose-dependent stimulation of NAc DA overflow in freely moving rats. In fact, the locomotor stimulation produced by 4-Cl-BZT is highly concordant with its stimulation of DA overflow in the NAc (Fig. 2); only the higher (40 mg/kg i.p.) dose that elevates NAc DA elicits locomotor activation, and both the neurochemical and behavioral effect are similarly blunted relative to those of cocaine at the same dose. Moreover, allowing for the 15- to 20-min time lag in DA response due to dead space in the collection tubing, the time course of locomotor stimulation correlates well with the respective time course for stimulation of NAc DA overflow for each drug, although peak locomotor stimulation by 4-Cl-BZT appeared to slightly precede peak elevation of NAc DA (Fig. 2). As is the case for locomotor stimulation (Figs. 1B and 2B), the rapid and short-lived peak in NAc DA overflow after cocaine injection contrasts with the slow and persistent elevation of extracellular DA by 40 mg/kg 4-Cl-BZT (Fig. 2A). Although 4-Cl-BZT thus clearly penetrates into the brain after systemic administration, the degree and rate at which this occurs relative to cocaine, the distribution of the drug within the brain, and the relative influence of these factors on the behavioral effects of the drugs remain unknown.

Drug Combination Studies. Although the effects of pretreatment with 4-Cl-BZT on responses to low dose (10 mg/kg) and high dose (40 mg/kg) cocaine were studied under different conditions, dose-dependent additivity/potentiation of cocaine's effects by 4-Cl-BZT were observed in both cases. In microdialysis experiments, the elevation of NAc DA produced by 10 mg/kg cocaine was additive with that induced by administration of 40 mg/kg 4-Cl-BZT 2 h before and was prolonged relative to saline-pretreated controls (Fig. 2A). Furthermore, 4-Cl-BZT significantly shifted the cocaine dose-effect curve for discriminative-stimulus effects to the left, despite its own lack of efficacy as a cocaine-like stimulus. Thus, 4-Cl-BZT does not antagonize the actions of cocaine at the DA transporter in the NAc in vivo, whereas such an antagonism has been suggested for the selective DA uptake inhibitor GBR 12909 (Rothman et al., 1991; Baumann et al., 1994). Interestingly, the closely related analog GBR 12935 has been shown to enhance the locomotor stimulant effects of cocaine, even at doses of GBR 12935 that do not stimulate locomotion when given alone (Acri et al., 1996). In the current study, cocaine-induced locomotion in the rats undergoing microdialysis was clearly potentiated by 4-Cl-BZT pretreatment 2 h before, although this potentiation was observed only at a dose of 4-Cl-BZT that had stimulant effects of its own (Fig. 2B).