Acute administration of phencyclidine induces tonic activation of medial prefrontal cortex neurons in freely moving rats

Acute administration of phencyclidine induces tonic activation of medial prefrontal cortex neurons in freely moving rats

PII: S 0 3 0 6 - 4 5 2 2 ( 0 2 ) 0 0 2 9 8 - 1 Neuroscience Vol. 114, No. 3, pp. 769^779, 2002 D 2002 IBRO. Published by Elsevier Science Ltd All rig...

243KB Sizes 3 Downloads 32 Views

PII: S 0 3 0 6 - 4 5 2 2 ( 0 2 ) 0 0 2 9 8 - 1

Neuroscience Vol. 114, No. 3, pp. 769^779, 2002 D 2002 IBRO. Published by Elsevier Science Ltd All rights reserved. Printed in Great Britain 0306-4522 / 02 $22.00+0.00

www.neuroscience-ibro.com

ACUTE ADMINISTRATION OF PHENCYCLIDINE INDUCES TONIC ACTIVATION OF MEDIAL PREFRONTAL CORTEX NEURONS IN FREELY MOVING RATS Y. SUZUKI,a;b E. JODO,a S. TAKEUCHI,b S. NIWAb and Y. KAYAMAa a

Department of Physiology, Fukushima Medical University School of Medicine, 1 Hikari-ga-oka, Fukushima 960-1295, Japan b

Department of Neuropsychiatry, Fukushima Medical University School of Medicine, Fukushima, Japan

Abstract7Recent studies have reported that acute administration of the psychotomimetic drug phencyclidine results in considerable increases in the amounts of both extracellular glutamate and dopamine in the medial prefrontal cortex (mPFC). However, the e¡ect of phencyclidine on the ¢ring activity of mPFC neurons remains unknown. Here, we report the ¢rst data on phencyclidine-induced activation of mPFC neurons in freely moving rats. Unanesthetized rats received an intraperitoneal injection of either phencyclidine (5 mg/kg) or physiological saline (0.5 ml/kg) in order to investigate the impulse activity of mPFC neurons and behavioral activity. The phencyclidine injection induced a remarkable increase (two-fold or more) in the spontaneous discharge rate of the majority of mPFC neurons (20/23), and this increase lasted for more than 70 min. In addition, a considerable augmentation of behavioral activity was observed that nearly paralleled that of the mPFC neuronal activation. In contrast, microiontophoretically applied phencyclidine exerted little in£uence on the spontaneous ¢ring activity of most mPFC neurons (25/29) in anesthetized rats, although systemically applied phencyclidine produced activation of mPFC neurons even under general anesthesia. These results suggest that the behavioral abnormalities induced by acute administration of phencyclidine may be caused by hyperactivation of mPFC neurons, and that this hyperactivation is elicited through excitatory inputs from brain regions outside the mPFC. D 2002 IBRO. Published by Elsevier Science Ltd. All rights reserved. Key words: phencyclidine, prelimbic cortex, unit recording, electrophysiology.

area implicated in schizophrenia. PCP can thus elicit schizophrenic-like symptoms, and it may a¡ect brain regions that are disturbed in schizophrenia, including the PFC. Therefore, PCP-administered animals are now considered to be the most reliable pharmacological model of schizophrenia. Acute injection of PCP in rats induces behavioral abnormalities that include hyperlocomotion and stereotypic behaviors (Castellani and Adams, 1981; Haggerty et al., 1984), as well as impaired learning and memory processes (Verma and Moghaddam, 1996). Numerous pharmacological studies have indicated that the behavioral and cognitive disturbances caused by PCP are associated with abnormalities in dopamine neurotransmission (Hertel et al., 1996; Verma and Moghaddam, 1996). Some researchers, however, have proposed an alternative hypothesis: that disturbance of NMDA receptor-mediated glutamate neurotransmission is a major factor involved in PCP-induced behavioral abnormalities (Javitt and Zukkin, 1991). Recently, this glutamate hypothesis was supported by the results of a study demonstrating that normalization of increased glutamate release in the PFC ameliorated the disruptive e¡ects of PCP on stereotypy and locomotion without altering dopamine neurotransmission (Moghaddam and Adams, 1998). Therefore, an increase in glutamate e¥ux in the PFC may be primarily responsible for behavioral and cognitive dysfunctions produced by PCP. However, the elevation of

Numerous clinical studies have reported that phencyclidine [1-(1-phenylcyclohexyl)piperidine, PCP], which is known to be a non-competitive N-methyl-D-aspartate (NMDA) receptor antagonist, induces psychotic symptoms that resemble schizophrenia in healthy individuals (Luby et al., 1959; Domino and Luby, 1981; Javitt and Zukkin, 1991). PCP-induced psychosis has both positive and negative schizophrenia-like symptoms (Javitt and Zukkin, 1991). In addition, the administration of PCP to stabilized schizophrenic patients leads to an exacerbation of preexisting symptoms (Luby et al., 1959; Domino and Luby, 1981), and low-dose administration of PCP not only disturbs selective attention and paired-associate learning in schizophrenic patients, but also does so in normal subjects (Bakker and Amini, 1961). Moreover, patients who undergo prefrontal lobotomy are relatively resistant to the psychotomimetic e¡ects of PCP (Itil et al., 1967), suggesting that PCP may primarily disturb function of the prefrontal cortex (PFC), which is the

*Corresponding author. Tel. : +81-24-548-2111; fax: +81-24-5482571. E-mail address: [email protected] (E. Jodo). Abbreviations : ACh, acetylcholine ; MANOVA, multivariate analysis of variance ; MK-801, dizocilpine maleate; mPFC, medial prefrontal cortex; NAC, nucleus accumbens; NMDA, N-methyl-Daspartate ; PCP, phencyclidine ; PFC, prefrontal cortex. 769

NSC 5731 22-8-02

770

Y. Suzuki et al.

glutamate e¥ux by PCP in the mPFC does not necessarily indicate a correspondent augmentation of the ¢ring activity of mPFC neurons, given that PCP has an antagonistic action against NMDA receptors. Since the PFC has direct and indirect connections to various brain regions, such as the hippocampus (Jay and Witter, 1991; Jay et al., 1992), nucleus accumbens (NAC; Gorelova and Yang, 1999), locus coeruleus (Jodo et al., 1998) and thalamus (Groenewegen et al., 1990; Krettek and Price, 1977; Rose and Woolsey, 1948), it is important to know how PCP a¡ects the ¢ring activity of PFC neurons. However, no study has reported on the e¡ects of PCP on the ¢ring activity of PFC neurons in unanesthetized animals. In the present study, we simultaneously recorded the single-unit activity of medial prefrontal cortex (mPFC) neurons and the behavioral activity of freely moving rats, and examined the responses to systemic administration of PCP. In addition, we microiontophoretically applied PCP in anesthetized rats to examine the direct in£uence of PCP on mPFC neurons.

EXPERIMENTAL PROCEDURES

Animal preparation Adult male Sprague Dawley rats (250^350 g, n = 22) were used in two experiments. Animals were housed in the Fukushima Medical University Animal Facility under conditions of constant temperature and humidity, with food and water available ad libitum. All e¡orts were made to minimize animal su¡ering and to reduce the number of animals used. All procedures adopted in this study were approved by, and conducted under control of, the Fukushima Medical University Animal Care and Use Committee. Preparation of freely moving rats and single-unit recording Rats were surgically anesthetized with pentobarbital (50 mg/ kg, i.p.). In brief, rats were positioned in a stereotaxic frame with blunt ear bars according to the atlas of Paxinos and Watson (1997), and a 30^35 mm incision of the skin over the skull was made under sterile conditions. The wound margin was in¢ltrated with 2% lidocaine. A commercial micromanipulator with a tungsten microelectrode (Unique Medical; 150 Wm P, 250 Wm/turn, 20^30 k6 at 1000 Hz) was used to record the single-unit activity of mPFC neurons. The recording electrode was lowered unilaterally into the left mPFC (coordinates, 3.0 mm anterior to the bregma, 0.8 mm lateral to midline, and 2.3^2.7 mm ventral to the surface of the brain), and the micromanipulator was then ¢xed onto the cranium using dental cement. A miniature stainless steel bolt (1.4 mm diameter) that was screwed to the skull near the insertion point of the needle electrode to contact the dura mater acted as a reference. All lead wires from the electrodes were soldered onto a miniature connector, which was anchored onto the cranium using dental cement and skull screws (1.4 mm diameter). Rats received antibiotics after surgery. Recording began at least seven days after their recovery from surgery. The signal from a recording electrode was passed into a bioelectric ampli¢er with a bandpass ¢lter of 500^3000 Hz via a miniature preampli¢er (JB220J; Nihon Kohden, Tokyo, Japan) directly attached to the socket on the animal’s head. All signals were stored on magnetic tapes using a DAT (digital audiotape) recorder (PCM5881; NF Electronic Instruments, Yokohama, Japan). Unit action potentials were separated from the back-

ground noise using a laboratory interface and spike detecting software (CED1401, SPIKE II; Cambridge Electronic Design, Cambridge, UK). Averaged spike shapes of each neuron were constructed o¥ine from 100 traces to measure the widths of action potentials. Drug injection in freely moving rats Animals received an intraperitoneal injection of a subanesthetic dose of 5 mg/kg PCP or 0.5ml/kg physiological saline 30 min after the start of recording. The recording session lasted for 90 min after drug injection. When a rat was injected with both PCP and saline on the same day, the recording for the PCP injection was started at least 120 min after the saline injection. When a rat was injected with PCP more than once, injections were made at an interval of more than seven days. Five of 12 rats underwent PCP injection only once, six rats did so twice, and one rat received three injections. In the preliminary experiment, a series of at least three injections exerted little in£uence on response pro¢les of mPFC neurons. Behavioral activity rating The rat was placed in a transparent plastic box (W 30, L 30, H 35 cm) during recording sessions. Intersecting lines were drawn on the £oor of the box. All behavioral activities were recorded by a video camcorder for 120 min. The ‘general activity score’ was de¢ned as the number of times the muzzle crossed a line on the £oor of the box. This score was considered to be an index that included both locomotor activity and stereotypy. The score was recorded every 2 min. The recording session started at least 2 h after the animals were placed in the experimental box, to give them time to adapt to the experimental environment. Microiontophoresis of PCP We examined the direct e¡ect of microiontophoretically applied PCP on mPFC neurons in an additional group of 10 rats anesthetized with pentobarbital and urethane. In brief, these animals were ¢rst anesthetized with pentobarbital (50 mg/kg), and then mounted in a stereotaxic apparatus. Supplemental doses of pentobarbital (14 mg/kg, i.p.) were administered when necessary during surgery. A burr hole was then made in the skull over the PFC, and the dura was incised to insert microelectrodes to the mPFC at the same coordinates as described above. A glass microelectrode containing 2% Pontamine Sky Blue in 0.5 M sodium acetate was used to record the impulse activity of mPFC neurons. The recording electrode had a tip diameter of about 1^2 Wm, and its impedance ranged from 6 to 9 M6. Single neuronal activity was passed to an ordinary microelectrode ampli¢er as described previously (Jodo et al., 1998). Iontophoretic application of drugs was administered through six-barreled glass micropipets glued to the recording electrode. The tip of the recording electrode protruded about 8 Wm beyond the tips of the injection pipets. The injection pipets had an overall tip diameter of about 8 Wm. Two of the six barrels were ¢lled with 3 M NaCl to balance the current. Two additional barrels were ¢lled with 0.1 M PCP in 200 mM NaCl (pH 5.5). The remaining two barrels were ¢lled with 50 mM NMDA in 200 mM NaCl (pH 8.1^8.5). Iontophoretic drug currents ranged between 10 and 40 nA. PCP was retained with 5 nA of negative current and ejected with positive current. NMDA was retained with 5 nA of positive current and ejected with negative current. The ejection of drugs was accomplished using an iontophoresis circuit employing automatic current neutralization (DPI-30B; Dia Medical System, Tokyo, Japan). The duration of ejection was 20 s for PCP, and 10 s for NMDA. Before each ejection, spontaneous ¢ring activity was recorded for at least 3 min as a control (baseline activity). We made no recordings until the pentobarbital anesthesia began to wear o¡, as indicated by sporadic whisker movement. Urethane (0.4^0.5 g/kg, i.p.) was then administered, about 30

NSC 5731 22-8-02

Hyperactivation of prefrontal cortex neurons by PCP

771

Fig. 1. (A) Histological locations of recording sites in the coronal section including the mPFC. Filled circles represent sites of recorded neurons; which showed the increase of ¢ring rate after PCP injection. Open rectangles denote those of neurons exhibiting no signi¢cant changes after PCP injection. A ¢lled triangle denotes the site of a neuron; which showed a decrease of ¢ring rate after PCP injection. The distance of each section from the bregma is indicated in mm (Paxinos and Watson, 1997). (B) Typical spike shape of mPFC neuron on the oscilloscope.

min before the start of iontophoresis, to maintain a moderate but stable state of anesthesia during the entire recording period. Intraperitoneal PCP injection (5^10 mg/kg) was administered at the last recording in each experiment to con¢rm that systemic PCP can produce activation of mPFC neurons even under anesthesia. Before PCP injection, baseline ¢ring activity was recorded for at least 15 min. Histology Animals were deeply anesthetized with pentobarbital upon completion of the experiments. For those animals in which the unit recording occurred under a freely moving state, the last position of the recording electrode was marked by passing a positive current (30 WA, 20 s). The electrode was then moved to the position of the ¢rst neuron encountered in the track, and the position was marked again. In animals who received microiontophoresis of PCP, dye was passed from the recording barrel by a constant-current source (10 WA for 4 min) to deposit two blue spots at the points where the ¢rst and the last neurons were recorded. The animals were then perfused with 50 ml of 10% formalin-phosphate-bu¡ered saline solution. The brain was

removed and sectioned coronally (50 Wm thickness) using a frozen microtome. Sections were stained with neutral red to facilitate the detection of marks created by the recording electrode. Data analysis To analyze the ¢ring activity of mPFC neurons in freely moving rats, we generated peristimulus time histograms (2-min bin width) for each mPFC neuron recorded from 30 min before to 90 min after drug injection. The baseline period was de¢ned as the 30-min period preceding the injection. The mean and S.D. of counts per baseline bin was determined. The onset of the signi¢cant increase in ¢ring rate was de¢ned as the ¢rst point at which ¢ve consecutive bins had a mean value exceeding the mean baseline activity by 2 S.D.). We excluded data within the ¢rst 10 min after injection from the statistical analyses, due to the possible in£uence of the direct e¡ect of pain upon needle insertion, which otherwise could have been mistakenly interpreted as the e¡ect of the drug itself. A signi¢cant decrease of ¢ring rate was scored when the mean value of ¢ve consecutive bins of mPFC activity fell below mean baseline activity by 2 S.D.

NSC 5731 22-8-02

772

Y. Suzuki et al.

Fig. 2. E¡ects of PCP or saline injection on ¢ring activity of mPFC neurons. Typical samples of three individual mPFC neurons showing an increase (A), a decrease (B) or no signi¢cant change (C) in spontaneous ¢ring rate, after PCP injection (5 mg/kg, i.p.). (D) Averaged relative ¢ring rates of mPFC neurons per 2-min bin among all neurons tested for PCP or saline injection show marked long-lasting increase after PCP injection, but no marked change after saline injection. PCP or saline was injected at point 0 on the time axis. Asterisks indicate signi¢cant di¡erences between PCP and saline groups at the same postinjection time (P 6 0.05). Line on each plot denotes standard error of the mean. F : PCP (n = 23). R: Saline (n = 22).

Fig. 3. Averaged general activity scores across all recording sessions for PCP or saline injection. PCP or saline was injected at point 0 on the time axis. Asterisks indicate signi¢cant di¡erences between PCP and saline groups at the same postinjection time (P 6 0.05). Line on each plot denotes standard error of the mean. F : PCP (n = 23). R: Saline (n = 22).

NSC 5731 22-8-02

Hyperactivation of prefrontal cortex neurons by PCP

General behavioral activity scores (the number of lines crossed) were counted every 2 min; the mean and S.D. were calculated for a control period 30 min before the injection, and signi¢cant activation and suppression were de¢ned in the same manner as for the ¢ring rate. We calculated the relative ¢ring rate in each neuron, which was de¢ned as the ratio of ¢ring rate in each bin to the average ¢ring rate during the baseline period. This value was used to determine the averaged e¡ects of PCP (or saline) in mPFC neurons with di¡erent ¢ring levels. Multivariate analyses of variance (MANOVAs) were performed on the relative ¢ring rate and general activity score for the factors of each group (PCP, saline) and time. The Tukey method was used for post hoc pairwise comparisons. In other cases, the two-tailed t-test was used to determine statistical signi¢cance between the two groups, which was established at the P 6 0.05 level. In the iontophoresis experiment, a neuron was considered responsive to the drug if it satis¢ed the following criteria (French, 1986): (1) the ¢ring rate changed by 50% or more

773

during the ¢rst 1 min after the start of ejection; (2) the neuron returned to the baseline ¢ring rate; (3) the response could be repeated at least twice. In the experiment involving systemic injection of PCP under anesthesia, signi¢cant activation of mPFC neurons was de¢ned in the same manner as for recordings in freely moving rats, but with a di¡erent bin width (5 s) and control period (15 min).

RESULTS

E¡ects of PCP on mPFC neurons in freely moving rats The tips of the recording electrodes were all located histologically in the mPFC upon completion of the experiment in 12 rats (see Fig. 1A). In all 12 rats, the single-unit activity was recorded with a peak voltage

Fig. 4. Typical responses of two mPFC neurons to iontophoretically applied PCP and NMDA. (A) mPFC neuron showing no signi¢cant response to locally applied PCP. (B) A di¡erent mPFC neuron, exhibiting long-latency suppression after PCP ejection. Each thick horizontal bar denotes the duration of drug application (PCP, 20 s; NMDA, 10 s). Numbers over bars represent ejection current (nA).

NSC 5731 22-8-02

774

Y. Suzuki et al.

exceeding the background activity or noise by at least three times (see Fig. 1B). Recorded spikes typically consisted of negative and positive strokes. The width of the ¢rst negative stroke was 0.52 V 0.01 (mean V S.E.M.) for PCP injection, 0.53 V 0.01 ms for saline injection. The width of the positive one was 0.86 V 0.02 for PCP, 0.90 V 0.02 ms for saline. There was no signi¢cant di¡erence in spike width between two injections. In these 12 rats, 45 neurons were recorded in total; the e¡ect of PCP was examined in 23 neurons, and the other 22 neurons provided control data after saline injection. The average ¢ring rate in the preinjection period was 2.91 V 0.37 (mean V S.E.M.) before PCP, and 3.62 V 0.33 spikes/s before saline. The results of statistical analysis demonstrated no signi¢cant di¡erence in baseline ¢ring rate between the two groups (t = -1.50, d.f. = 43, P 6 0.154). Following PCP injection, 20 of the 23 neurons (87.0%) exhibited the increase of ¢ring rate (Fig. 2A), two neurons (8.7%) showed no signi¢cant change (Fig. 2C), and only one (4.3%) showed a decrease in ¢ring rate (Fig. 2B). For the saline injection, 18 of the 22 neurons (81.8%) showed no change and four neurons (18.2%) showed a decrease. The averaged relative ¢ring rates for all neurons following PCP or saline injection are shown in Fig. 2D. The results of MANOVA revealed a signi¢cant interaction between GROUP and TIME (F(1,44) = 4.81, P 6 0.001), indicating that the relative ¢ring rate of mPFC neurons signi¢cantly increased after PCP injection (lasting for about 80 min), whereas there was little change after saline injection. E¡ects of PCP on behavioral activity (general activity score) PCP was seen to induce abnormal behavior including hyperlocomotion and stereotypy in all rats. The averaged general activity scores before and after PCP or saline injection are presented in Fig. 3. The results of MANOVA revealed a signi¢cant interaction between GROUP and TIME (F(1,44) = 13.32, P 6 0.001), indicating that the general activity score was signi¢cantly increased after PCP injection (lasting for about 70 min), whereas there was little change after saline injection.

a typical example of this is shown in Fig. 4B. To examine the long-latency e¡ects of PCP, the mean values of interspike intervals for each neuron were calculated for four consecutive 1-min periods (PRE, the 1-min period immediately before PCP application; POST1, the 1-min period immediately after the beginning of PCP application; POST2, the 1-min period between 1 and 2 min after the beginning of PCP application; POST3, the 1-min period between 2 and 3 min after the beginning of PCP application), and these were then converted to ¢ring rates for each period by taking their reciprocals. Figure 5 shows the average ¢ring rates of all mPFC neurons tested during these four consecutive periods. MANOVAs were performed on the ¢ring rates for the factor of TIME (PRE, POST1, POST2, POST3). The results of these MANOVAs showed a signi¢cant main e¡ect of TIME (F(3,84) = 3.25, P 6 0.02). Post hoc analyses revealed that the average ¢ring rates signi¢cantly decreased 3 min after the start of PCP ejection. In order to determine whether PCP had been ejected as designed, PCP was also iontophoretically applied in 24 of the 29 neurons, immediately followed by an approximately 10-s period of iontophoretic NMDA application. PCP was applied for a 20-s duration before the start of NMDA application. The 24 neurons were all activated by iontophoretic administration of NMDA. Preapplication of PCP suppressed the NMDA-induced activation in all mPFC neurons tested (see Fig. 4). We systemically administered PCP for the last recorded neuron in each experiment. In half of the rats (5/10), the ¢ring rate of recorded neurons was markedly increased about 10 to 15 min after systemic injection of PCP, and this activation usually lasted until the recorded neuron moved away from the tip of the recording electrode (see Fig. 6). In these rats, PCP was microiontophoretically applied to 19 neurons. In the remaining rats, recorded neurons moved away from the tip of the recording electrode a few minutes after PCP injection,

Microiontophoresis of PCP in the mPFC Neuronal activity of the mPFC (prelimbic area) was recorded in an additional group of 10 rats under anesthesia. As shown in Fig. 4A, microiontophoretically applied PCP exerted no signi¢cant in£uence on the spontaneous ¢ring activity of most neurons tested (25/29, 86.2%) for at least 1 min after the start of PCP ejection. Three neurons (10.3%) exhibited more than a 50% decrease in ¢ring rate shortly after local application of PCP, and only one neuron (3.4%) exhibited more than a 50% increase in ¢ring rate after local application of PCP. However, some of the neurons that showed no signi¢cant change shortly after PCP application exhibited a longlatency suppression (9/29, 31.0%) of the spontaneous ¢ring rate more than 1 min after cessation of PCP ejection;

Fig. 5. Mean ¢ring rates of mPFC neurons (n = 29) in four consecutive 1-min periods (PRE, 1-min period immediately before local PCP application; POST1, 1-min period immediately after the beginning of PCP application; POST2, 1-min period between 1 and 2 min after the beginning of PCP application ; POST3, 1-min period between 2 and 3 min after the beginning of PCP application). The vertical line on each bar represents standard error of the mean.

NSC 5731 22-8-02

Hyperactivation of prefrontal cortex neurons by PCP

775

Fig. 6. A typical response of a mPFC neuron to systemically applied PCP (10 mg/kg, i.p.) in an anesthetized rat. Panel B shows a peristimulus time histogram of the unit discharges shown in panel A. About 1480 s after PCP injection, this neuron moved away from the recording electrode, and unit activity disappeared.

and so we could not record impulse activity of neurons until systemic PCP exerted its e¡ects.

¢ring activity in anesthetized rats in which systemic PCP clearly activated mPFC neurons. mPFC activity and behavior

DISCUSSION

Acute administration of PCP at a subanesthetic dose produced long-lasting activation of mPFC neurons in freely moving rats, in addition to marked augmentation of behavioral anomalies. This is the ¢rst report to directly demonstrate the excitatory in£uence of PCP administration on the ¢ring activity of mPFC neurons in unanesthetized rats. The responses of mPFC neurons to PCP roughly agree with previously reported variation in extracellular glutamate in the mPFC (Moghaddam and Adams, 1998). The present results indicate that PCP itself does not directly activate mPFC neurons, because microiontophoretic application of PCP on a single mPFC neuron exerted no excitatory in£uence on its

The onset time and duration of activation of ¢ring activity of mPFC neurons by PCP were almost identical to the time course of changes in behavioral activity. The question arises whether the activation of mPFC neurons is a cause of hyperlocomotive states or only a secondary phenomenon due to alteration of the arousal level by PCP-induced accelerated motor activity. The latter possibility is less likely, because the systemic administration of PCP increased the ¢ring activity of mPFC neurons (a response typically observed in freely moving rats) even in immobilized unconscious rats under general anesthesia. Therefore, our results seem to indicate that at least some part of behavioral abnormalities may be caused by tonic activation of the mPFC neurons induced by PCP.

NSC 5731 22-8-02

776

Y. Suzuki et al.

In a human PET study with [18 F]£uorodeoxyglucose, a subanesthetic dose of ketamine, a dissociative anesthetic agent (White et al., 1982) that acts as a low-potency NMDA antagonist, administered to healthy volunteers produced bilateral focal increases in metabolic activity in the PFC (Breier et al., 1997). In the above study, ketamine also induced an acute psychotic state, and the conceptual disorganization induced by ketamine was positively correlated with prefrontal metabolic activity. These results support our ¢nding that a potent NMDA antagonist, PCP, produced long-lasting activation of the PFC neurons. This prolonged activation may have a pivotal role in inducing abnormal behaviors. Mechanisms of PCP-induced activation of mPFC neurons Anomalous dopamine neurotransmission, especially in the NAC, is known to be a key factor eliciting abnormal behaviors, such as hyperlocomotion and stereotypic behaviors (Creese and Iversen, 1975; Fibiger et al., 1973; Jackson and Moghaddam, 2001; Kelly et al., 1975). In particular, several studies have reported that systemically administered PCP increases dopamine e¥ux in addition to glutamate in the mPFC and NAC (Adams and Moghaddam, 1998; Moghaddam and Adams, 1998). The possibility is thus suggested that the increase in dopamine in the mPFC and/or NAC is a major factor in the behavioral abnormalities produced by systemically administered PCP. However, several studies have found that the dopaminergic system has a limited role in PCPinduced behavioral abnormalities; locomotor e¡ects of PCP are blocked only at high cataleptic doses of dopamine antagonists (Castellani and Adams, 1981; Ogren and Goldstein, 1994), and disturbed performance by ketamine or PCP in working memory-related tasks is only partially ameliorated with D2 receptor antagonists, and is little a¡ected by other dopamine antagonists (Krystal et al., 1995; Verma and Moghaddam, 1996). Furthermore, the normalization of glutamate release in the mPFC by pretreatment with a metabotropic glutamate receptor agonist attenuates the disruptive e¡ects of PCP on both stereotypy and locomotion, despite the sustained increase in dopamine e¥ux (Adams and Moghaddam, 1998; Moghaddam and Adams, 1998). Therefore, other neurotransmitter systems should be investigated to explore their role in the devastating action of PCP on behavior. As stated above, systemically administered PCP increases glutamate e¥ux in the mPFC. This previously reported increase in extracellular glutamate (Moghaddam and Adams, 1998) is very similar in terms of time course to the observed increase in the ¢ring of mPFC neurons in the present study following systemic administration of PCP. Although the mechanism of increase in extracellular glutamate by PCP is presently unknown, these ¢ndings indicate that increased glutamate release in the mPFC may have induced the activation of mPFC neurons in the present study, and that this activation may be related to the disruptive behavioral e¡ects of PCP. The increased dopamine e¥ux in the NAC may also result from long-lasting activation of the mPFC by systemic administration of PCP, since it

has been shown that PFC exerts excitatory control over dopamine release in the NAC (Taber and Fibiger, 1995; You et al., 1998; Karreman and Moghaddam, 1996). However, PCP itself is expected to have a suppressive e¡ect on both spontaneous activity and glutamateinduced activation of neurons, since it has been shown to be a NMDA receptor antagonist (Javitt and Zukkin, 1991). French (1986) reported that iontophoretically applied PCP had a suppressive e¡ect on putative dopaminergic neurons in the ventral tegmental area. The suppressive e¡ect of locally applied PCP on mPFC neurons was also reported by Gratton et al. (1987). In the study by Gratton et al., the spontaneous ¢ring activity of almost all mPFC neurons was suppressed by local application of PCP. This result seems to contradict our ¢nding that more than half of neurons tested exhibited no signi¢cant change after local application of PCP, although it produced short- or long-latency suppression of ¢ring in about 40% of mPFC neurons. This apparent contradiction may be due to di¡erences in experimental procedures. One notable di¡erence is the recording method. Gratton et al. used a recording barrel ¢lled with an electrolyte containing 10 mM glutamate to augment neural ¢ring. Local application of PCP under such conditions may suppress mainly glutamate-induced activation of neurons, and tend to produce apparent suppression of ¢ring. Therefore, the results of Gratton et al. may not necessarily contradict ours. To examine whether PCP directly activates mPFC neurons, local application of PCP was performed with rats placed under anesthesia. Generally, it may be di⁄cult to directly extend the results obtained under anesthesia to unanesthetized animals, given that PCP strongly interacts with anesthetic agents (Balster and Wessinger, 1983). However, in the present study, we con¢rmed that systemic application of PCP-induced activation of mPFC neurons in both anesthetized and unanesthetized rats. This result indicates that the neural mechanism responsible for PCP-induced activation of mPFC neurons in unanesthetized animals may also work in anesthetized animals. In addition, we con¢rmed that locally applied PCP could exert its pharmacological action as a NMDA receptor antagonist under anesthesia. Thus, the interaction between locally applied PCP and systemically administered anesthetics appears to be negligible, at least in the mPFC. If so, the results of our local injection study suggest that the excitation of mPFC neurons by systemically applied PCP in freely moving rats is not caused by a direct pharmacological action on the mPFC neurons, because few neurons exhibited excitatory responses to locally applied PCP in our study. What then is the neural mechanism involved in PCPinduced activation of mPFC neurons? Several possibilities can be considered. The most likely is that PCP indirectly activates a group of neurons providing glutamatergic input to the mPFC by disinhibition or an unknown mechanism. There are three candidate projections to the mPFC; one projection originates in the ventral hippocampus (Jay and Witter, 1991; Jay et al., 1995), namely the CA1 region of the ventral hippocam-

NSC 5731 22-8-02

Hyperactivation of prefrontal cortex neurons by PCP

pus (VCA1) and the ventral subiculum (VS), and the second projection originates from the mediodorsal nucleus of the thalamus (Groenewegen et al., 1990; Krettek and Price, 1977; Rose and Woolsey, 1948). The third candidate projection originates from the basolateral amygdala (Groenewegen and Berendse, 1990; Kelly et al., 1982). Stimulation of either the VCA1 or VS region activates mPFC neurons mainly via nonNMDA receptors (Jay et al., 1992). Furthermore, pretreatment with non-NMDA receptor antagonists ameliorates impaired performance caused by the NMDA antagonist ketamine in a spatial delayed alteration task (Moghaddam et al., 1997). Thus, at present, it seems reasonable to hypothesize that the activation of mPFC neurons after PCP injection may be caused by the activation of a glutamatergic pathway to the mPFC via non-NMDA receptors. Cholinergic projections are another possible origin of the PCP-induced activation of mPFC neurons, given that acetylcholine (ACh) is a neurotransmitter known to induce potent excitation of cortical neurons (Mednikova et al., 1998; Yang and Mogenson, 1990). Some reports suggest the involvement of cholinergic projections in PCP action; the extracellular concentration of ACh increases in both the mPFC (Jentsch et al., 1998) and the hippocampus (Giovannini et al., 1994) following administration of PCP or MK-801 (a more selective NMDA antagonist than PCP), whereas these drugs reduce striatal ACh out£ow (Zocchi and Pert, 1994). Furthermore, the mPFC receives projections from two major cholinergic systems: the basal nucleus of Meynert in the basal forebrain (Luiten et al., 1987; Rye et al., 1984) and the laterodorsal tegmental nucleus in the brainstem (Satoh and Fibiger, 1986). However, to date, there are no reports describing the excitatory e¡ects of PCP on these cholinergic neurons. Furthermore, PCPinduced behavioral symptoms are markedly ameliorated not by controlling the cholinergic systems, but by modulating the glutamatergic systems in the mPFC (Moghaddam and Adams, 1998). Therefore, although we cannot rule out the possibility of modulation of the mPFC by PCP via the cholinergic pathways, the cholinergic systems appear to play a minor role in the malfunctioning of the mPFC caused by PCP injection. The possibility that pharmacological action of PCP within the mPFC is involved in PCP-induced activation of mPFC neurons also cannot be ruled out. The ¢rst action that we should consider is the disinhibition of mPFC neurons through decreased GABA release in the mPFC, since local perfusion of PCP (or MK-801) in freely moving rats reportedly reduces the amount of extracellular GABA in the mPFC (Yonezawa et al., 1998). The sustained PCP-induced suppression of

777

GABA release in the mPFC may activate the mPFC neurons, but it does not appear to be a major factor in activation in light of the following ¢ndings: iontophoretically applied PCP, which may spread to nearby inhibitory neurons, activated almost none of the mPFC neurons in the present study; and the local application of PCP to brain slices does not induce an increase in excitatory postsynaptic current in mPFC neurons (Aghajanian and Marek, 2000). Our present results, nevertheless, do not completely rule out the possibility of disinhibition, because PCP microiontophoretically applied for only 20 s may not be able to defuse in adequate amounts to disfacilitate local GABA neurons at substantial distances from the injection pipet. Therefore, this possibility should be examined in future studies using a pressure ejection technique. Another possibility to be considered is that PCP metabolites, and not PCP itself, may activate the mPFC neurons. Although some PCP metabolites are more potent inhibitors of serotonin re-uptake than PCP itself (Hori et al., 1996), PCP metabolites have a weaker ability to elicit abnormal behavior than PCP (Baba et al., 1994). Therefore, it is unlikely that PCP metabolites are a key factor in the PCP-induced activation of mPFC neurons. Conclusions The present ¢ndings demonstrate that systemic administration of PCP produces prolonged activation of mPFC neurons, which parallels behavioral activation in freely moving rats, and that this activation may not be due to the direct action of PCP on the mPFC neurons, but rather may be mediated by excitatory input from a remote structure. A better understanding of the psychotomimetic e¡ects of drugs in humans may be achieved by clarifying the mechanism of PCP-induced activation in the mPFC. Since PCP-induced psychosis resembles schizophrenia, further studies of the e¡ects of PCP on cortical neurons or neural networks may also lead to a better understanding of the neuropathology of schizophrenia, and hopefully will suggest a new paradigm for its treatment.

Acknowledgements(Preliminary results of this study were presented at the 29th Annual Meeting of the Society for Neuroscience, 1999. This work was supported by a Grant-in-Aid for Scienti¢c Research (No. 11670958) from Japan Society for the Promotion of Science, and by a grant to E.J. from the Fukushima Society for the Promotion of Medicine. We wish to thank Dainippon Pharmaceutical Corporation for their generous donation of PCP. We also wish to thank Nobuko Anzai for her excellent technical assistance.

REFERENCES

Adams, B., Moghaddam, B., 1998. Corticolimbic dopamine neurotransmission is temporally dissociated from the cognitive and locomotor e¡ects of phencyclidine. J. Neurosci. 18, 5545^5554. Aghajanian, G.K., Marek, G.J., 2000. Serotonin model of schizophrenia: emerging role of glutamate mechanisms. Brain Res. Rev. 31, 302^312. Baba, A., Yamamoto, T., Kawai, N., Yamamoto, H., Suzuki, T., Moroji, T., 1994. Behavioral e¡ects of phencyclidine and its major metabolite, (trans)4-phenyl-4-(1-piperidinyl)cyclohexanol, in mice. Behav. Brain Res. 65, 75^81.

NSC 5731 22-8-02

778

Y. Suzuki et al.

Bakker, C.B., Amini, F.B., 1961. Observations on the psychotomimetic e¡ects of Sernyl. Comp. Psychiatry 2, 269^280. Balster, R.L., Wessinger, W.D., 1983. Central nervous system depressant e¡ects of phencyclidine. In: Kamenka, J.M., Domino, E.F., Geneste, P. (Eds.), Phencyclidine and Related Arylcyclohexylamines: Present and Future Applications. NPP Books, Ann Arbor, MI, pp. 291^309. Breier, A., Malhotra, A.K., Pinals, D.A., Weisenfeld, N.I., Pickar, D., 1997. Association of ketamine-induced psychosis with focal activation of the prefrontal cortex in healthy volunteers. Am. J. Psychiatry 154, 805^811. Castellani, S., Adams, P.M., 1981. E¡ects of dopaminergic drugs on phencyclidine-induced behavior in the rat. Neuropharmacology 20, 371^374. Creese, I., Iversen, S.D., 1975. The pharmacological and anatomical substrates of the amphetamine response in the rat. Brain Res. 83, 419^436. Domino, E.F., Luby, E., 1981. Abnormal mental states induced by phencyclidine as a model of schizophrenia. In: Domino, E.F. (Eds.), PCP (Phencyclidine) : Historical and Current Perspectives. NPP Books, Ann Arbor, MI, pp. 401^418. Fibiger, H.C., Fibiger, H.P., Zis, A.P., 1973. Attenuation of amphetamine-induced motor stimulation and streotypy by 6-hydroxydopamine in the rat. Br. J. Pharmacol. 47, 683^692. French, E.D., 1986. E¡ects of phencyclidine on ventral tegmental A10 dopamine neurons in the rat. Neuropharmacology 25, 241^248. Giovannini, M.G., Mutolo, D., Bianchi, L., Michelassi, A., Pepeu, G., 1994. NMDA receptor antagonists decrease GABA out£ow from the septum and increase acetylcholine out£ow from the hippocampus : a microdialysis study. J. Neurosci. 14, 1358^1365. Gorelova, N., Yang, C.R., 1999. The course of neural projection from the prefrontal cortex to the nucleus accumbens in the rat. Neuroscience 76, 689^706. Gratton, A., Ho¡er, B.J., Freedman, R., 1987. Electrophysiological e¡ects of phencyclidine on the medial prefrontal cortex of the rat. Neuropharmacology 26, 1275^1283. Groenewegen, H.J., Berendse, H., 1990. Connections of the subthalamic nucleus with ventral striatopallidal parts of the basal ganglia in the rat. J. Comp. Neurol. 294, 607^622. Groenewegen, H.J., Berendsf, H.W., Wolters, J.G., Lohman, A.H.M., 1990. The anatomical relationship of the prefrontal cortex with the striatopallidal system, the thalamus and the amygdala: evidence for a parallel organisation. Prog. Brain Res. 85, 95^118. Haggerty, G.C., Forney, R.B., Johnson, J.M., 1984. The e¡ect of a single administration of phencyclidine on behavior in the rat over a 21-day period. Toxicol. Appl. Pharmacol. 75, 444^453. Hertel, P., Mathe, J.M., NomiKos, G.G., Mathe, A.A., Svensson, T.H., 1996. E¡ects of D-amphetamine and phencyclide on behavior and extracellular concentrations of neurotensin and dopamine in the ventral striatum and the medial prefrontal cortex of the rat. Behav. Brain Res. 22, 103^114. Hori, T., Suzuki, T., Baba, A., Abe, S., Yamamoto, T., Moroji, T., Shiraishi, H., 1996. E¡ects of phencyclidine metabolites on serotonin uptake in rat brain. Neurosci. Lett. 209, 153^156. Itil, T., Keskiner, A., Kiremicti, N., Holden, J.M.C., 1967. E¡ect of phencyclidine in chronic schizophrenia. Can. Psychiatr. Assoc. J. 12, 209^212. Jackson, M.E., Moghaddam, B., 2001. Amygdala regulation of nucleus accumbens dopamine output is governed by the prefrontal cortex. J. Neurosci. 21, 676^681. Javitt, D.C., Zukkin, S.R., 1991. Recent advances in the phencyclidine model of schizophrenia. Am. J. Psychiatry 148, 1301^1308. Jay, T.M., Glowinski, J., Thierry, A.M., 1995. Inhibition of hippocampo-prefrontal cortex excitatory responses by the mesocortical DA system. NeuroReport 6, 1845^1848. Jay, T.M., Thierry, A.M., Wiklund, L., Glowinski, J., 1992. Excitatory amino acid pathway from the hippocampus to the prefrontal cortex: contribution of AMPA receptor in hippocampo-prefrontal cortex transmission. Eur. J. Neurosci. 4, 1285^1295. Jay, T.M., Witter, M.P., 1991. Distribution of hippocampal CA1 and subicular e¡erents in the prefrontal cortex of the rat studied by means of anterograde transport of phaseolus vulgaris-leucoagglutinin. J. Comp. Neurol. 313, 574^586. Jentsch, J.D., Dazzi, L., Chhatwal, J.P., Verrico, C.D., Roth, R.H., 1998. Reduced prefrontal cortical dopamine, but not acetylcholine, release in vivo after repeated, intermittent phencyclidine administration to rats. Neurosci. Lett. 258, 175^178. Jodo, E., Chiang, C., Aston-Jones, G., 1998. Potent excitatory in£uences of prefrontal cortex activity on noradrenergic locus coeruleus neurons. Neuroscience 83, 63^79. Karreman, M., Moghaddam, B., 1996. The prefrontal cortex regulates the basal release of dopamine in the limbic striatum : an e¡ect mediated by ventral tegmental area. J. Neurosci. 66, 589^598. Kelly, A., Domesick, V., Mauta, W., 1982. The amygdalostriatal projection in the rat: an anatomical study by anterograde and retrograde tracing methods. Neuroscience 7, 615^630. Kelly, P.H., Seviour, P.W., Iversen, S.D., 1975. Amphetamine and apomorphine responses in the rat following 6-OHDA lesions of the nucleus accumbens and the corpus striatum. Brain Res. 94, 507^522. Krettek, J.E., Price, J.L., 1977. The cortical projections of the mediodorsal nucleus and adjacent thalamic nuclei in the rat. J. Comp. Neurol. 171, 157^192. Krystal, J., Karper, L., Bennett, A., Abi-Saab, D., Souza, C., Abi-Dargham, A., Charney, D., 1995. Modulating ketamine-induced thought disorder with lorazepam and haloperidol in humans. Schizophr. Res. 15, 156. Luby, E.D., Cohen, B.D., Rosenbaum, G., Gottlieb, J.S., Kelley, R., 1959. Study of a new schizophrenomimetic drug ^ Sernyl. Am. Med. Assoc. Arch. Neurol. Psychiatry 81, 363^369. Luiten, P.G.M., Gaykema, R.P.A., Traber, J., Spencer, D.G., Jr., 1987. Cortical projection patterns of magnocellular basal nucleus subdivision as revealed by anterogradely transported Phaseolus vulgaris leucoagglutinin. Brain Res. 413, 229^250. Mednikova, Y., Loseva, E.V., Karnup, S.V., Zhadin, M.N., 1998. Responses of cortical neurons to microiontophoretic application of acetylcholine to their dendrites. Neurosci. Behav. Physiol. 28, 107^115. Moghaddam, B., Adams, B., Verma, A., Daly, D., 1997. Activation of glutamatergic neurotransmission by ketamine: a novel step in the pathway from NMDA receptor blockade to dopaminergic and cognitive disruptions associated with the prefrontal cortex. J. Neurosci. 17, 2921^2927. Moghaddam, B., Adams, B.W., 1998. Reversal of phencyclidine e¡ects by a group II metabotropic glutamate receptor agonist in rats. Science 281, 1349^1352. Ogren, S.O., Goldstein, M., 1994. Phencyclidine and dizociloine-induced hyperlocomotion are di¡erentially mediated. Neuropsychopharmacology 11, 167^177. Paxinos, G., Watson, C., 1997. The Rat Brain in Stereotaxic Coordinates. Academic Press, London. Rose, J.E., Woolsey, C.N., 1948. Structures and relation of limbic cortex and anterior thalamic nuclei in rabbit and cat. J. Comp. Neurol. 89, 279^ 347. Rye, D.B., Wainer, B.H., Mesulam, M.-M., Mufson, E.J., Saper, C.B., 1984. Cortical projections arising from the basal forebrain: a study of cholinergic and noncholinergic components employing combined retrograde tracing and immunohistochemical localization of choline acetyltransferase. Neuroscience 13, 627^643. Satoh, K., Fibiger, H.C., 1986. Cholinergic neurons of the laterodorsal tegmental nucleus: e¡erent and a¡erent connections. J. Comp. Neurol. 253, 277^302.

NSC 5731 22-8-02

Hyperactivation of prefrontal cortex neurons by PCP

779

Taber, M., Fibiger, H., 1995. Electrical stimulation of the prefrontal cortex increases dopamine release in the nucleus accumbens of the rat: modulation by metabotropic glutamate receptors. J. Neurosci. 15, 3896^3904. Verma, A., Moghaddam, B., 1996. NMDA receptor antagonists impair prefrontal cortex function as assessed via spatial delayed alternation performance in rats: modulation by dopamine. J. Neurosci. 16, 373^379. White, P.F., Way, W.L., Trevor, A.J., 1982. Ketamine ^ its pharmacology and therapeutic uses. Anesthesiology 56, 119^136. Yang, C.R., Mogenson, G.J., 1990. Dopaminergic modulation of cholinergic responses in rat medial prefrontal cortex: an electrophysiological study. Brain Res. 524, 271^281. Yonezawa, Y., Kuroki, T., Kawahara, T., Tashiro, N., Uchimura, H., 1998. Involvement of gamma-aminobutyric acid neurotransmission in phencyclidine-induced dopamine release in the medial prefrontal cortex. Eur. J. Pharmacol. 341, 45^56. You, Z., Tzschentke, T., Brodin, E., Wise, R., 1998. Electrical stimulation of the prefrontal cortex increases cholecystokinin glutamate, and dopamine release in the nucleus accumbens: an in vivo microdialysis study in freely moving rats. J. Neurosci. 18, 6492^6500. Zocchi, A., Pert, A., 1994. Alterations in striatal acetylcholine over£ow by cocaine, morphine, and MK-801: relationship to locomotor output. Psychopharmacology 115, 297^304. (Accepted 20 March 2002)

NSC 5731 22-8-02