Abstract
The serotonin2C (5-hydroxytryptamine2C; 5-HT2C) receptor (5-HT2CR) is found in abundance in dopamine (DA) mesocorticolimbic pathways and is one of the important target proteins that modulates the behavioral effects of cocaine. In the present study, the hypothesis was tested that the 5-HT2CR in the prefrontal cortex (PFC) may control either spontaneous or cocaine-evoked locomotor activity as well as the discriminative stimulus properties of cocaine. In male Sprague-Dawley rats implanted with bilateral cannulae aimed at the PFC, local microinjections of the preferential 5-HT2CR agonist 6-chloro-2-(1-piperazinyl)pyrazine hydrochloride (MK 212) (0.05–0.5 μg/side) did not alter spontaneous activity, but dose-dependently decreased horizontal hyperactivity evoked by cocaine (10 mg/kg i.p.). Given alone, the selective 5-HT2CR antagonist 8-[5-(2,4-dimethoxy-5-(4-trifluorophenylsulfonamido)phenyl-5-oxopentyl]-1,3,8-triazo-spiro[4.5]decane-2,4-dione hydrochloride (RS 102221) (5 μg/side) increased basal locomotor activity of rats expressed in the vertical plane. Microinjections of RS 102221 (5 μg/side, but not 0.15–1.5 μg/side) significantly enhanced the horizontal activity induced by cocaine (10 mg/kg). In rats trained to discriminate cocaine (10 mg/kg i.p.) from saline (i.p.) in a two-lever, water-reinforced fixed ratio 20 task, intra-PFC microinjections of MK 212 (0.05 and 0.5 μg/side) did not substitute for cocaine, but attenuated the stimulus effects of cocaine. However, intra-PFC microinjections of RS 102221 (1.5 and 5 μg/side) evoked 13 and 40% cocaine-lever responding when tested alone and enhanced the recognition of cocaine. These data indicate that the PFC is a brain site at which the 5-HT2CR exerts an inhibitory control over the hyperactive and discriminative stimulus effects of cocaine known to be dependent upon activation of the DA mesoaccumbens circuit.
Augmented dopamine (DA) neurotransmission and indirect activation of DA D1- and D2-like receptors have been established to play a central role in the in vivo effects of cocaine. A sizeable body of literature supports the importance of the mesoaccumbens DA pathway, which originates in DA cell bodies in the ventral tegmental area (VTA) and terminates in the nucleus accumbens (NAc) in mediating the behavioral effects of cocaine. In addition to the pronounced involvement of DA in its in vivo effects, cocaine also inhibits serotonin (5-hydroxytryptamine; 5-HT) reuptake and enhances 5-HT availability for interaction with potentially all sixteen 5-HT receptors found in the brain (Hoyer et al., 2002). One such receptor, the 5-HT2C receptor (5-HT2CR), is densely localized in brain DA circuits (Lopez-Gimenez et al., 2001), and neurochemical studies implicate a tonic inhibitory control of the 5-HT2CR in brain DA pathways (De Deurwaerdere and Spampinato, 1999). In keeping with an inhibitory role for 5-HT2CR, systemic pretreatment with 5-HT2CR agonists (MK 212 or RO 60-0175) has been reported to suppress the hypermotive (Grottick et al., 2000), discriminative stimulus (Callahan and Cunningham, 1995), and reinforcing effects of cocaine as well as reinstatement of cocaine-seeking behavior following extinction (Grottick et al., 2000). However, brain-penetrant 5-HT2CR antagonists (SB 242084, SDZ SER 082) have been shown to potentiate these behavioral effects of cocaine (McCreary and Cunningham, 1999; Fletcher et al., 2002). These data suggest that the 5-HT2CR may be a functionally important regulator of the neural substrates that control responsiveness to cocaine.
Two recent microinjection studies tested the hypothesis that one site of action for the 5-HT2CR to modulate cocaine-induced behaviors was the NAc. We demonstrated that intra-NAc (but not intra-VTA) shell microinjection of the 5-HT2CR antagonist RS 102221 blocked the hypermotive (McMahon et al., 2001) and discriminative stimulus effects of cocaine (Filip and Cunningham, 2002), whereas intra-NAc shell pretreatment with the preferential 5-HT2CR agonists MK 212 or RO 60-0175 enhanced the behavioral actions of cocaine (Filip and Cunningham, 2002). However, based upon systemic studies with 5-HT2CR agonists (above), we would have predicted diametrically opposite outcomes if the NAc were the common site of action for cocaine and the 5-HT2CR ligands. These data suggest that the influence of systemically administered 5-HT2CR ligands on cocaine-induced behaviors may represent actions at 5-HT2CRs differentially localized to multiple brain nuclei.
The mesocortical DA pathway, which originates in the VTA and terminates in several cortical areas, including the prefrontal cortex (PFC), has been postulated as a forebrain circuit important in the systemic effects of cocaine. In fact, systemic administration of cocaine has been shown to enhance DA release in PFC (Maisonneuve et al., 1990). Interestingly, intra-PFC microinjection of cocaine does not mimic the hyperactivity or discriminative stimulus effects of systemically administered cocaine (Wood and Emmett-Oglesby, 1989; Delfs et al., 1990), but does support self-administration (Goeders and Smith, 1983) and the reinstatement of cocaine-seeking behavior (McFarland and Kalivas, 2001). Moreover, recent neuroimaging studies in human cocaine abusers show that injections of cocaine, recent withdrawal from cocaine, and exposure to cocaine-associated cues are associated with activation of glucose metabolism or an increase in regional blood flow in the PFC (Volkow et al., 1991; Childress et al., 1999). As an important part of a complex neuroanatomical network implicated in the regulation of subcortical DA neurotransmission and the behavioral effects of cocaine, the PFC sends direct excitatory glutamate projections to the NAc and VTA (Sesack and Pickel, 1992). Functional studies of these connections suggest that DA in the PFC inhibits excitatory glutamate pyramidal cells to negatively control DA function in the mesoaccumbens pathway with a net outcome of reduced mesoaccumbens-dependent behavioral output (Taber et al., 1995).
To better elucidate the neurobiological substrates of action for 5-HT2CR to control basal and cocaine-induced behaviors, we have extended our previous findings (McMahon et al., 2001; Filip and Cunningham, 2002) to analyze the manner in which specific behaviors induced by cocaine are controlled by the moderate levels of 5-HT2CR located in the PFC (Lopez-Gimenez et al., 2001). We analyzed the ability of intra-PFC microinjection of a 5-HT2CR agonist or antagonist to mimic or alter spontaneous motor activity as well as the hyperactive or discriminative stimulus effects of cocaine. This study used the preferential 5-HT2CR agonist MK 212 (Kennett, 1993; Porter et al., 1999) and the selective 5-HT2CR antagonist RS 102221 (Bonhaus et al., 1997).
The PFC itself is an anatomically heterogeneous region composed of ventral and dorsal divisions. The ventral PFC is a component of a ventral “limbic” circuit with connectivity to the NAc shell, medial ventral pallidum, amygdala, and VTA (Zahm and Brog, 1992; McFarland and Kalivas, 2001). The functional influence of the ventral PFC has recently been shown to be important to the behavioral effects of cocaine. For example, selective quinolinic acid lesions of the ventral PFC attenuated cocaine-evoked rearing and conditioned place preference as well as the development of behavioral sensitization to cocaine (Tzschentke and Schmidt, 1999). Therefore, given the prominent involvement of the limbic forebrain in the behavioral effects of cocaine (Delfs et al., 1990; Callahan et al., 1994; Filip and Siwanowicz, 2001; Rodd-Henricks et al., 2002), the important linkage between the ventral PFC and the limbic forebrain, and the key control of ventral PFC over the behavioral effects of cocaine, we chose to target our microinfusions of 5-HT2CR ligands to the ventral PFC.
Materials and Methods
Animals. Male Sprague-Dawley rats (n = 204; Collegium Medicum, Jagiellonian University, Krakow, Poland) weighing 250 to 330 g at the beginning of the experiment were used. The rats were housed two or three per cage in standard plastic rodent cages in a colony room maintained at 21 ± 2°C and at 40–50% humidity under a 12 h light/dark cycle (lights on at 0700 h). Rats surgically fitted with indwelling bilateral guide cannulae were housed individually. Rats assigned to locomotor activity assays (n = 180) were provided with continuous access to tap water and rodent chow except during experimental sessions. In drug discrimination assays (n = 24), rodent chow was available ad libitum; the amount of water each animal received was restricted to that given during daily training sessions, after test sessions (10–15 min), and on weekends (36 h). All experiments were conducted during the light phase of the light/dark cycle (between 9:00 AM and 2:00 PM) and were carried out in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and approval from the Institutional Animal Care and Use Committee.
Surgical Implantation of Cannulae and Microinjection Protocols. Rats underwent surgical implantation of 26-gauge stainless steel bilateral guide cannulae (Small Parts Inc., Miami Lakes, FL). Each rat was anesthetized using an intramuscular (i.m.) injection of 65 mg/kg ketamine and 15 mg/kg xylazine in 0.9% NaCl. With the upper incisor bar of a stereotaxic instrument (Kopf Instruments, Tujunga, CA) positioned at –3.8 mm below the interaural line and using the intersection of the bregma and longitudinal sutures as the origin, the ventral surfaces of the bilateral guide cannulae were positioned 2 mm above the PFC (AP = +2.7 mm, ML = ±0.75 mm, and DV = –3 mm; Paxinos and Watson, 1998). The guide cannulae were fastened to the skull with stainless steel screws (Small Parts) and cranioplastic cement (Plastics One, Inc., Roanoke, VA) and were fitted with 28-gauge stainless steel bilateral obturators (Small Parts). Rats received two injections of penicillin (10,000 units/kg i.m.) after surgery, and were allowed a 1-week recovery period during which rats were handled and weighed daily. Following the initial 1-week recovery period, each rat was habituated to the brief confinement associated with intracranial microinjections by removing the 28-gauge internal obturators, gently restraining the rat for approximately 3 min, and replacing the obturators. For bilateral intra-PFC microinjections, the obturators were removed and two internal cannulae (Plastics One) were positioned to extend 1 mm below the tips of the bilateral guide cannulae. The bilateral internal cannulae were attached to two 5-μl syringes (Hamilton Co., Reno, NV) via PE-50 tubing (Clay Adams, Parsippany, NJ). A microsyringe drive (Baby Bee, BAS Bioanalytical Systems Inc., West Lafayette, IN) driven by a programmable controller (Bee Hive Controller, BAS Bioanalytical Systems Inc.) delivered a volume of 0.2 μl/side at a rate of 0.1 μl/min. After completion of microinjections, the injection cannulae remained in place for an additional 1 min to allow for diffusion away from cannula tips; the obturators were then replaced.
Drugs. All drugs were dissolved in sterile saline (0.9% NaCl). Cocaine HCl (Merck, Darmstadt, Germany) was injected intraperitoneally (i.p.) in a volume of 1 ml/kg. MK 212 and RS 102221 (Tocris Cookson Inc., Bristol, UK) were injected intracranially in a volume of 0.2 μl/side. All solutions injected centrally were adjusted to pH 7.2 except the solution of RS 102221, which was adjusted to pH 6–7; control vehicle at pH 6–7 did not alter basal or cocaine-stimulated locomotor activity or the stimulus effects of cocaine (10 mg/kg; data not shown).
Measurement of Locomotor Activity
Apparatus. Motor activity was monitored and quantified in clear Plexiglas chambers (43 × 43 × 25 cm) housed inside Optovarimex activity monitors surrounded with a 15 × 15 array of photocell beams located 3 cm from the floor surface (Columbus Instruments, Columbus, OH). Interruptions of these photobeams resulted in horizontal activity defined as distance traveled (expressed in centimeters). A second set of 15 photocell beams was located 14 cm above the floor surface, and interruptions of these photobeams provided for monitoring of rearing (expressed as activity counts). Separate records of horizontal activity and rearing were made by the control software (Columbus Instruments) for subsequent statistical evaluation.
Locomotor Activity and Microinfusion Protocols. Surgically implanted rats were habituated to the test environment for 2 h/day on each of the 2 days before the start of the experiment, and on the test day for 1 h before the administration of drugs. Rats were assigned to two individual groups according to 5-HT2CR ligand treatment (MK 212 or RS 102221) and each rat underwent only one test session. One group of rats (n = 80, divided into eight subgroups, 10 animals/subgroup) received bilateral intra-PFC microinjections of either sterile saline (0.2 μl/side) or a different dose of MK 212 (0.05, 0.15, or 0.5 μg/0.2 μl/side); bilateral microinjections were followed immediately by an i.p. injection of either saline (1 ml/kg) or cocaine (10 mg/kg). Another group of animals (n = 100, divided into 10 subgroups; 10 animals/subgroup) received bilateral intra-PFC microinjections of either saline (0.2 μl/side) or a different dose of RS 102221 (0.15, 0.5, 1.5, or 5 μg/0.2 μl/side); microinjections were immediately followed by an i.p. injection of either saline (1 ml/kg) or cocaine (10 mg/kg). Measurements of locomotor activity began immediately after the systemic injection and lasted 60 min.
Drug Discrimination Experiments
Apparatus. The procedures were conducted in commercially available, two-lever operant chambers (MED Associates, St. Albans, VT). Each chamber was equipped with a water-filled dispenser mounted equidistant between two response levers on one wall and housed in a light- and sound-attenuating cubicle (MED Associates). A 28-V house light provided illumination; a blower supplied ventilation and masking noise. An interface (MED Associates) connected the chambers to a computer that controlled and recorded all experimental events using MedState software.
Discrimination Training and Test Protocols. Standard two-lever, water-reinforced drug discrimination procedures were utilized (Callahan and Cunningham, 1995; Filip and Cunningham, 2002). Rats were injected intraperitoneally with cocaine (10 mg/kg) or saline (1 ml/kg) 15 min before daily (Monday through Friday) 30-min sessions (n = 24). During the initial “errorless training” phase, only the stimulus-appropriate (drug or saline) lever was present. Training began under a fixed ratio 1 (FR 1) schedule of water reinforcement and the FR requirement was incremented until all animals were responding reliably under an FR 20 schedule for each experimental condition. For half of the rats left-lever responses were reinforced after cocaine administration, whereas right-lever responses were reinforced after saline administration; conditions were reversed for the remaining rats. During this phase of training, cocaine and saline were administered irregularly with the restriction that neither condition prevailed for more than three consecutive sessions.
After responding stabilized, both levers were presented simultaneously during 15-min sessions. The rats were required to respond on the stimulus-appropriate (correct) lever to obtain water reinforcement, and there were no programmed consequences for responding on the incorrect lever. This phase of training continued until the performance of all rats attained criterion (defined as mean accuracies of at least 80% correct for 10 consecutive sessions).
Pharmacological Tests and Microinfusion Protocols. When rats achieved the criterion for accuracy, test sessions were initiated and training sessions were run during the intervening days to maintain discrimination accuracy. Rats were required to maintain accuracies of at least 80% correct for the saline and cocaine maintenance sessions that immediately preceded a test. During a test session the rat was placed in the chamber and, upon completion of 20 responses on either lever, a single reinforcer was delivered and the house light was turned off. The rat was removed from the chamber, returned to the colony, and allowed free access to water for 10 min beginning 15 to 30 min after the end of each test. A test session was terminated after 15 min if the rat did not complete 20 responses on either lever. After achieving the criterion for accuracy, the group of trained rats was separated into two subgroups: one group was tested with intra-PFC infusions of MK 212 (n = 12), the other with intra-PFC infusions of RS 102221 (n = 12).
Several pharmacological manipulations were performed during test sessions. A systemic dose-response curve for cocaine was established before surgical implantation of cannulae; rats were tested 15 min after an injection of cocaine (1.25–10 mg/kg i.p.). Following recovery from surgery, discrimination training was reinstated. After several weeks, the systemic dose-response curve for cocaine was reestablished and did not differ from that established before surgery (data not shown); the postsurgical dose-response curve served as control in the present experiment.
In intracranial substitution tests, lever selection was assessed 10 min after bilateral intracranial infusions of sterile saline (0.9% NaCl; 0.2 μl/side), MK 212 (0.05 or 0.5 μg/side), or RS 102221 (1.5 or 5 μg/side) paired with a systemic injection of saline (1 ml/kg i.p.). Control tests were also conducted in which rats were assessed for lever selection 10 min following administration of either saline or cocaine (10 mg/kg i.p.) which had been immediately preceded by bilateral intracranial injections of saline (0.2 μl/side). In combination tests, bilateral microinjections of MK 212 (0.05 or 0.5 μg/side) or RS 102221 (1.5 or 5 μg/side) immediately preceded an injection of cocaine (1.25–10 mg/kg i.p.); rats were tested for lever selection 10 min later. The order of drug tests in the group of rats treated with intra-PFC infusions of MK 212 was as follows: saline (0.2 μl/side) + saline (i.p.), cocaine (2.5 mg/kg i.p.), saline (0.2 μl/side) + cocaine (10 mg/kg i.p.), cocaine (1.25 mg/kg i.p.), MK 212 (0.05 μg/side) + cocaine (2.5, 5, and 10 mg/kg i.p.), cocaine (5 mg/kg i.p.), MK 212 (0.5 μg/side) + cocaine (2.5, 5, and 10 mg/kg i.p.), cocaine (10 mg/kg i.p.), MK 212 (0.05 μg/side) + saline (i.p.), and MK 212 (0.5 μg/side) + saline (i.p.). The order of drug tests in the group of rats treated with intra-PFC infusions of RS 102221 was as follows: saline (0.2 μl/side) + saline (i.p.), cocaine (5 mg/kg i.p.), saline (0.2 μl/side) + cocaine (10 mg/kg i.p.), cocaine (1.25 mg/kg i.p.), RS 102221 (1.5 μg/side) + cocaine (2.5, 5, and 10 mg/kg i.p.), cocaine (10 mg/kg i.p.), RS 102221 (5 μg/side) + cocaine (2.5, 5, and 10 mg/kg i.p.), cocaine (2.5 mg/kg i.p.), RS 102221 (1.5 μg/side) + saline (i.p.), and RS 102221 (5 μg/side) + saline (i.p.). Test sessions in which intracranial microinjections were performed were spaced 5 to 7 days apart and the time frame during which microinjection analyses were conducted was approximately 2 months. Each rat received a total of 10 bilateral microinfusions.
Histology. At the completion of the study rats were overdosed with chloral hydrate (800 mg/kg i.p.) and the brains were removed and stored in a 20% sucrose/10% formalin solution for at least 3 days before the sectioning. Brain sections (50 μm) were mounted onto gelatin-coated glass slides. The brain sections were defatted, stained with cresyl violet, cleared with xylene, and placed under coverslips. The cannula placements were verified using a light microscope. Only those animals whose cannulae were within the ventral PFC were included for statistical analysis. No significant tissue damage was evident upon histological examination of sections.
Statistical Analyses. For motor activity assays, the dependent measures were horizontal activity (mean total distance traveled, in centimeters) (±S.E.M.) and rearing (mean total counts of photobeam breaks) (±S.E.M.) observed during the 60-min test session. Each experiment was subjected to a one-way analysis of variance with levels of the treatment factor corresponding to the drug combinations administered to that group. Planned, pairwise comparisons of the treatment means were made with a least-significant-difference test (SAS for Windows, Version 8.1), which were conducted with an experimentwise error rate of α = 0.05.
During drug discrimination training sessions, accuracy was defined as the percentage of correct responses to total responses before the delivery of the first reinforcer; during test sessions, performance was expressed as the percentage of drug-appropriate responses to total responses before the delivery of the first reinforcer. Response rates (responses per second) were also evaluated during training and test sessions as a measure of behavioral disruption. For training sessions, the response rate was calculated as the total number of responses emitted on either lever before completion of the first FR 20 divided by the number of seconds taken to complete that FR 20. During test sessions, the response rate was calculated as the total number of responses before the completion of 20 responses on either lever divided by the number of seconds taken to complete that FR 20. Only data (percentage of drug-lever responding and response rates) from animals that completed the FR 20 during test sessions were used in analyses.
Student's t test for repeated measures was used to compare the percentage of drug-lever responding and response rate during test sessions with the corresponding values for either the previous drug session (substitution tests) or the training dose tested alone (combination tests). All comparisons were made with an experimentwise type I error rate (α) set at 0.05.
Results
Histology
For each animal included in the analyses below, the injection cannulae projected bilaterally past the outer guide cannulae into the ventral PFC. Examples of bilateral placements identified in motor activity (Fig. 1, left) and drug discrimination studies (Fig. 1, right) are illustrated. Inspection of brain tissue revealed slight evidence of gliosis at the site of injection, although surrounding tissue was generally intact.
Locomotor Activity Assays
Intra-PFC Microinjection of Saline. Microinjections of saline into the ventral PFC followed by a systemic injection of saline resulted in levels of activity (Figs. 2 and 3) similar to that reported following a systemic injection of saline tested alone (McCreary and Cunningham, 1999) or following intra-NAc microinfusions of saline using equivalent test protocols (McMahon et al., 2001; Filip and Cunningham, 2002). Microinjections of saline into the PFC followed by a systemic injection of cocaine (10 mg/kg i.p.) resulted in significant increases in horizontal activity (654–812%) and rearing (323–600%) as compared with saline-saline control values (Figs. 2 and 3).
Intra-PFC microinjection of MK 212. Eighty rats received bilateral microinjections of saline or the 5-HT2CR agonist MK 212 (0.05, 0.15, or 0.5 μg/side) followed by a systemic injection of saline or cocaine (10 mg/kg). Of these, 69 rats exhibited cannula placement bilaterally positioned in the ventral PFC at +2.2 to +3.2 mm posterior to the bregma. A main effect of treatment was observed for horizontal activity (F7,55 = 10.64, p < 0.001) and rearing (F7,55 = 2.74, p < 0.05). Intra-PFC pretreatment with MK 212 (0.05–0.5 μg/side) before a systemic injection of saline did not alter basal motor activity (p > 0.05; Fig. 2). However, intra-PFC microinfusions of MK 212 significantly decreased cocaine-evoked horizontal activity (0.15 and 0.5 μg/side; p < 0.05; Fig. 2, top) in the absence of a significant effect on rearing (p > 0.05; Fig. 2, bottom).
Intra-PFC Microinjection of RS 102221. Of the 100 rats originally cannulated and tested, 76 rats exhibited cannulae placements bilaterally positioned in the ventral PFC at +2.2 to +3.2 posterior to the bregma. For these 76 rats, a main effect of treatment was observed for horizontal activity (F9,65 = 9.39, p < 0.001) and rearing (F9,65 = 3.68, p < 0.05). Intra-PFC microinfusions of RS 102221 (0.15–5 μg/side) before a systemic injection of saline did not alter horizontal activity (p > 0.05; Fig. 3, top), but did significantly increase rearing at a dose of 5 μg/side (p < 0.05; Fig. 3, bottom). Intra-PFC microinfusions of RS 102221 at 5 μg/side also significantly increased cocaine-evoked horizontal activity (p < 0.05; Fig. 3, top), but not cocaine-evoked rearing (p < 0.05; Fig. 3, bottom).
Drug Discrimination Experiments
Cocaine-Saline Discrimination and Dose-Response Relationship for Cocaine. Acquisition of the cocaine (10 mg/kg) versus saline discrimination was met in an average of 33 sessions (range: 24–41). After recovery from surgery, the performance criterion was reestablished in 15 sessions (range: 13–18). Systemic administration of cocaine (1.25–10 mg/kg) produced a dose-dependent increase in cocaine-appropriate responding before (data not shown) and after surgical implantation of cannulae in both subgroups of trained rats (Figs. 4 and 5); no differences were observed between the pre- and postsurgical dose-response curves for cocaine in either subgroup of trained rats (data not shown). Drug-lever responding after 1.25 and 2.5 mg/kg cocaine in both subgroups of rats was significantly different from the previous cocaine training session (p < 0.05); response rates for all test doses of cocaine did not differ from those observed during the immediately previous cocaine maintenance session for either subgroup of trained rats (p > 0.05).
Control tests were also conducted to ensure that the microinjection procedure did not interfere with the discrimination between cocaine and saline. Systemic administration of saline engendered <10% drug-lever responding (data not shown), as has been observed in our numerous drug discrimination studies in rats implanted with intracranial cannulae (e.g., Filip and Cunningham, 2002). Intra-PFC microinjections of saline did not alter the low levels of drug-lever responses seen after a systemic injection of saline (Figs. 4 and 5, far left panels); response rates did not vary between the control tests and the previous maintenance saline sessions. Intra-PFC microinjections of saline did not alter cocaine-lever responding (97.9 ± 0.5%) or response rates (0.67 ± 0.08 responses/s) seen after systemic injection of cocaine (10 mg/kg) for the group of rats that were tested with MK 212. Likewise, in the group of rats tested with RS 102221, intra-PFC microinjections of saline did not alter cocaine-lever responding (95.4 ± 4.2%) or response rates (0.59 ± 0.07 responses/s) seen after systemic injection of cocaine (10 mg/kg). These data indicate that the microinjection protocols themselves did not alter the ability of rats to appropriately recognize either a systemic injection of saline or cocaine (10 mg/kg).
Intra-PFC Microinjection of MK 212. Of the 12 rats originally cannulated and tested, 8 rats exhibited cannula placements bilaterally positioned in the ventral PFC at +2.2 to +3.2 posterior to the bregma (see Fig. 1, right, for examples of placements). In substitution tests in these animals, intra-PFC microinfusions 0.05 μg/side or 0.5 μg/side of MK 212 evoked 0% and 13% drug-lever responding, respectively; these values were significantly different (p < 0.05) from the previous cocaine training session; response rates were unaltered (Fig. 4, left).
A statistically significant suppression of drug-lever responding was observed for the combination of MK 212 (0.05 μg/side or 0.5 μg/side) plus either 5 or 10 mg/kg cocaine (p < 0.05; Fig. 4, top right) as compared with the same dose of cocaine tested alone. Intra-PFC microinjections of MK 212 (0.05 μg/side) plus cocaine (5 mg/kg) resulted in a significantly higher response rate relative to the same dose of cocaine tested alone (p < 0.05; Fig. 4, bottom right).
Intra-PFC Microinjection of RS 102221. Of the 12 rats originally cannulated and tested, 7 rats exhibited cannula placements bilaterally positioned in the ventral PFC at +2.2 to +3.2 posterior to the bregma (see Fig. 1, right, for examples of placements). In substitution tests in these rats, 1.5 and 5 μg/side of RS 102221 evoked 11 and 40% drug-lever responding, respectively; these values were significantly different from the previous cocaine training sessions (p < 0.05). Response rates were unaltered (Fig. 5, left). Pretreatment with intra-PFC microinfusions of RS 102221 (1.5 and 5 μg/side) dose-dependently increased drug-lever responding observed at submaximal doses of cocaine (1.25 and 2.5 mg/kg), which alone elicited 12 and 34% drug-lever responding, respectively (p < 0.05; Fig. 5, top right). Pretreatment with RS 102221 did not affect response rates seen at any tested dose of cocaine (Fig. 5, bottom right).
Discussion
Our findings strongly indicate that the 5-HT2CR within the ventral PFC is important in the regulation of expression of the behavioral effects of cocaine. We found that intra-PFC microinjections of the 5-HT2CR agonist MK 212 into the ventral PFC decreased, and the 5-HT2CR antagonist RS 102221 increased, the hyperactivity induced by cocaine. We also found that this population of PFC 5-HT2CR controlled expression of the discriminative stimulus effects of cocaine, which are thought to model the subjective effects of cocaine in humans (Schuster and Johanson, 1988). The perfectly oppositional effects of these two ligands and the extensive control studies support the concept that the observed outcomes are a consequence of the respective abilities of MK 212 and RS 102221 to stimulate and block the 5-HT2CR, and are not an outcome of lesions induced following implantation of the cannulae or infusions of the drugs.
The choice of drugs in the present study was guided by their selectivity and affinity for the 5-HT2CR and the efficacy of MK 212 in neuropharmacological analyses. Neurochemical studies of MK 212 have demonstrated that this piperazine analog exhibits its highest affinity (Ki ≈ 32–490 nM) for 5-HT2CR and displays full efficacy to stimulate the 5-HT2CR (Kennett, 1993; Porter et al., 1999; Cussac et al., 2002). In vivo analyses indicate that systemic administration of MK 212 evokes hyperthermia and hypophagia (Clineschmidt, 1979), oral dyskinesias (Eberle-Wang et al., 1996), penile erections (Berendsen et al., 1990), hypomotility (Lucki et al., 1989), and discriminative stimulus effects (Cunningham et al., 1986), all of which are blocked preferentially by 5-HT2CR antagonists. In contrast, MK 212 has a much lower affinity for 5-HT2AR (Ki ≈ 17,400 nM; Kennett, 1993; Porter et al., 1999) and does not evoke behavioral effects consistent with an efficacy to stimulate 5-HT2AR (e.g., Lucki et al., 1989; Fiorella et al., 1995). Although MK 212 does exhibit affinity and partial agonist actions at the 5-HT2BR (Porter et al., 1999; Cussac et al., 2002), the 5-HT2BR is unlikely to transduce the effects observed here due to its absence in the PFC (Duxon et al., 1997).
MK 212 binds to the 5-HT3R (Glennon et al., 1989), although its action as either an agonist or antagonist is not well defined. If MK 212 has efficacy as a 5-HT3R agonist, intra-PFC microinfusion would be expected to increase extracellular DA levels in PFC and suppress the firing of PFC neurons similar to that seen following intra-PFC microinfusion of selective 5-HT3R agonists (Ashby et al., 1989; Chen et al., 1992; Gobbi and Janiri, 1999). Such an agonist action at 5-HT3R might culminate in a functional suppression of stimulant-induced behaviors similar to that seen following intra-PFC microinfusion of DA agonists (Sokolowski and Salamone, 1994; Karler et al., 1998). However, although MK 212 is reported to block the actions of the 5-HT3R agonist 2-methyl-5-HT in dorsal root ganglion cells (Todorovic and Anderson, 1990), we have been unable to locate evidence to suggest that MK 212 acts as a 5-HT3R agonist in vivo. In drug discrimination studies, MK 212 substitutes for preferential 5-HT2CR agonist m-chlorophenylpiperazine (mCPP; Callahan and Cunningham, 1994) and the selective 5-HT2CR agonist RO 60-0175 (Dekeyne et al., 1999). Neither 5-HT3R agonists nor antagonists mimicked these 5-HT2CR-dependent cues and 5-HT3R antagonists did not block the mCPP- or RO 60-0175-mediated cues (Callahan and Cunningham, 1994; Dekeyne et al., 1999). Furthermore, 5-HT3R antagonists were not effective in blocking the ability of MK 212 to enhance adrenocorticotropin hormone release in male rats (Jorgensen et al., 2002). Although indirect, these studies offer evidence to exclude an agonist action of MK 212 to act at 5-HT3R in vivo.
The 5-HT2CR antagonist utilized in the present study was RS 102221, which displays high affinity (3.8–7.4 nM) for 5-HT2CR and 100-fold selectivity compared with 5-HT2AR and 5-HT2BR; the affinity of RS 102221 for all other assayed receptors is low (pKi < 6.5; Bonhaus et al., 1997). Consistent with its ability to function as a full antagonist in vitro (Bonhaus et al., 1997), RS 102221 lacks efficacy at h5-HT2CR expressed in Chinese hamster ovary cells (Cussac et al., 2002). Although brain distribution is limited, systemic administration of RS 102221 does increase food intake and weight gain and block a well characterized effect of 5-HT2CR stimulation to induce hypophagia (Bonhaus et al., 1997). After local application, RS 102221 has been shown to block the inhibitory effects of MK 212 on neurons in the rat nucleus tractus solitarius (Sevoz-Couche et al., 2000). These data suggest that the ability of intra-PFC infusion of RS 102221 to block the behavioral effects of cocaine is most likely related to antagonism of 5-HT2CR.
Serotonin neurons of the dorsal raphe nuclei densely innervate the ventral PFC (Lidov et al., 1980) and 5-HT2CR mRNA and protein have been localized in this region (Clemett et al., 2000; Lopez-Gimenez et al., 2001). Furthermore, the ventral PFC (over the dorsal PFC) provides significant afferent input to raphe and regulation of raphe 5-HT neurons (Hajos et al., 1998; Varga et al., 2001). In the ventral PFC, afferent 5-HT terminals primarily contact interneurons (Smiley and Goldman-Rakic, 1996), which are likely to be γ-aminobutyric acid (GABA) interneurons (Lopez-Gimenez et al., 2001), although a possible serotonergic innervation of glutamate pyramidal neurons has not been ruled out (Smiley and Goldman-Rakic, 1996; Lopez-Gimenez et al., 2001). The ventral PFC is also a primary cortical target of the DA pathway originating in the VTA; afferent DA terminals synapse on both glutamate pyramidal neurons and GABA interneurons (Sesack et al., 1995). A DA-dependent suppression of activity of PFC pyramidal neurons and the subsequent reduction of excitatory glutamate output appear to have a net depressive effect on DA function in the mesoaccumbens pathway and on expression of behaviors evoked by psychostimulants (Sokolowski and Salamone, 1994; Karler et al., 1998). Stimulation of raphe nuclei (Mantz et al., 1990) or iontophoretic application of 5-HT ligands (Bergqvist et al., 1999) suppresses spontaneous and/or glutamate-activated firing of PFC neurons in a 5-HT2CR-dependent manner (Bergqvist et al., 1999), suggesting that the 5-HT2CR limits the excitability of cortical pyramidal neurons (Carr et al., 2002). Given the parallel between the behavioral profile observed upon intra-PFC microinjections of the 5-HT2CR ligands with that seen following DA ligands, 5-HT2CR in ventral PFC may influence the output of cocaine-evoked behaviors via modification of PFC output to the DA mesoaccumbens circuit.
Intra-PFC application of 5-HT2CR ligands might act at a receptor protein localized in several different cellular populations in the PFC, either at the level of the 5-HT [see (1) on Fig. 6] or DA axon terminals [see (2) on Fig. 6], on excitatory glutamate neurons [see (3) on Fig. 6], or on GABA interneurons [see (4) on Fig. 6]. If 5-HT2CRs localize to 5-HT terminals in PFC, these receptors could presynaptically control 5-HT release in this region [see (1) on Fig. 6]. The PFC receives a dense, possibly exclusive, innervation from dorsal raphe 5-HT neurons (O'Hearn and Molliver, 1984). However, the 5-HT2CR appears to be localized somatodendritically in PFC (Lopez-Gimenez et al., 2001). Furthermore, although the influence of direct application of 5-HT2CR ligands on PFC 5-HT efflux has not yet been established, neither the selective 5-HT2CR agonist RO 60-0175 nor the selective antagonist SB 242084 affected 5-HT release in the frontal cortex upon systemic administration (Millan et al., 1998). The 5-HT2CR might also localize to DA terminals in PFC, and 5-HT2CR activation could presynaptically control DA function in this region after local infusion of 5-HT2CR ligands [see (2) on Fig. 6]. However, although the 5-HT2CR transcript has been localized to regions containing DA cell bodies, expression of 5-HT2CR mRNA does not appear to colocalize with tyrosine hydroxylase in VTA neurons (Eberle-Wang et al., 1997), suggesting that membranes of DA terminals in the PFC are unlikely to possess 5-HT2CR. Thus, these data suggest that the modulatory influence of intra-PFC 5-HT2CR ligands on the behavioral effects of cocaine are not likely to be related to the actions of the 5-HT2CR localized to either 5-HT or DA terminals in PFC.
The majority of PFC pyramidal neurons express 5-HT2CR mRNA, as shown by single-cell polymerase chain reaction (Carr et al., 2002), and a direct effect of locally infused ligands at a somatodendritic 5-HT2CR (Lopez-Gimenez et al., 2001) may account for observed effects upon behavior. In the present experiments, inhibition of reuptake by cocaine would result in increased DA and 5-HT efflux, resulting in enhanced, indirect stimulation of DA D2R (Karler et al., 1998) and 5-HT2CR (Carr et al., 2002), respectively, and a subsequent dampening of excitability of PFC pyramidal neurons [see (3) on Fig. 6]. Costimulation of 5-HT2CR with MK 212 might be expected to further depress excitability of this pathway. In contrast, RS 102221 may block the actions of 5-HT at 5-HT2CR, perhaps enhancing excitability to some degree and further potentiating the behavioral effects of cocaine.
Interneurons in the PFC appear to be a primary target of 5-HT terminals, many of which are GABA interneurons (Smiley and Goldman-Rakic, 1996). Stimulation of 5-HT2CR found within GABA interneurons [see (4) on Fig. 6] would be expected to control PFC projection neurons, and could provide a mechanism through which locally administered 5-HT2CR ligands could affect behavioral responses to cocaine. In keeping with this hypothesis, the 5-HT2A/2CR agonist 1-(2,5-dimethoxy-4-iodophenyl)-2-aminopropane has been shown to increase extracellular GABA in rat brain (Abi-Saab et al., 1999) and electrophysiological studies indicate that stimulation of 5-HT2CR excites the activity of GABA neurons in the VTA (e.g., Di Giovanni et al., 2001). If this is the case for GABA interneurons in PFC, GABA released consequent to 5-HT2CR stimulation would be expected to reduce excitatory output to medium spiny neurons in the NAc and/or the DA cell bodies in the VTA (Sheldon and Aghajanian, 1990; Kalivas et al., 1993) and decrease stimulant-evoked behaviors. In fact, the GABAAR agonist gaboxadol injected intracortically has been shown to decrease cocaine-induced stereotypy (Karler et al., 1998). However, the preferential 5-HT2CR agonist mCPP inhibits the activity of pyramidal neurons, an effect not blocked by the GABAAR antagonist bicuculline (Bergqvist et al., 1999). These findings cast doubt on a mechanism of action for our observed effects based on a 5-HT2CR control of GABA function in the PFC
Our present findings help clarify the discrepancy between the influence of systemic and intracranial application of 5-HT2CR ligands on cocaine-evoked behaviors (see Introduction). The modulation of cocaine-evoked behaviors by systemic administration of the 5-HT2CR ligands is directionally identical with the influence of the 5-HT2CR ligands after local application to the ventral PFC. Extending our previous observations (McMahon et al., 2001; Filip and Cunningham, 2002), the present findings suggest that separate populations of 5-HT2R within the PFC, NAc, and VTA differentially control the mesoaccumbens DA pathway and that the ventral PFC is a specific brain site at which the 5-HT2CR exerts an inhibitory control over behavioral responses to cocaine.
The localization of an inhibitory influence of the 5-HT2CR to the ventral PFC is important to our understanding of the acute behavioral effects of cocaine and the processes involved in the development of cocaine use disorders. When taken in the context of studies of systemic administration of the 5-HT2CR ligands in animals and the brain imaging studies of human cocaine abusers (see above), the present findings suggest the therapeutic potential of 5-HT2CR manipulations in the treatment of cocaine dependence, the maintenance of abstinence, and/or the reduction of craving. In addition, abnormal post-transcriptional regulation of the 5-HT2CR has been linked to depression and suicide (Niswender et al., 2001; Gurevich et al., 2002) and 5-HT2CR ligands might ultimately be useful for the treatment of such major psychiatric disorders, some of which may be pathological states dependent upon limbic forebrain malfunction.
Acknowledgments
We gratefully acknowledge the technical assistance of Ewa Nowak and the secretarial assistance of Laurie Mitchell and Teri Tarrant.
Footnotes
-
This research was supported by the U.S.-Poland Joint Commission Maria Sklodowska-Curie Fund (M.F. and K.A.C.); a National Science Foundation-North Atlantic Treaty Organization Visiting Scientist Fellowship (M.F.); and by National Institute on Drug Abuse Grants DA 05708 and DA 06511 (K.A.C.).
-
Portions of these data were presented at the 62nd annual meeting of the College on Problems of Drug Dependence (San Juan, Puerto Rico, 2000) and at the XXIInd Collegium Internationale Neuro-Psychopharmacologicum Congress (Brussels, Belgium, 2000).
-
DOI: 10.1124/jpet.102.045716.
-
ABBREVIATIONS: 5-HT, 5-hydroxytryptamine (serotonin); 5-HT2CR, 5-HT2C receptor; DA, dopamine; PFC, prefrontal cortex; MK 212, 6-chloro-2-(1-piperazinyl)pyrazine hydrochloride; RS 102221, 8-[5-(2,4-dimethoxy-5-(4-trifluorophenylsulfonamido)phenyl-5-oxopentyl]-1,3,8-triazo-spiro[4.5]decane-2,4-dione hydrochloride; VTA, ventral tegmental area; NAc, nucleus accumbens; FR, fixed ratio; mCPP, m-chlorophenylpiperazine; RO 60-0175, (S)-2-(6-chloro-5-fluroindol-1-yl)-1-methylethylamine fumarate; SB 242084, 6-chloro-2,3-dihydro-5-methyl-N-[6-[(2-methyl-3-pyridinyl)oxy]-3-pyridinyl]-1H-indole-1-carboxyamide dihydrochloride; SDZ SER 082, (+)-cis-4,5,7a,8,9,10,11,11a-octahydro-7H-10-methylindolo(1,7BC)(2,6)naphthyridine.
- Received October 21, 2002.
- Accepted April 18, 2003.
- The American Society for Pharmacology and Experimental Therapeutics