Abstract

We previously reported synchronization of pyramidal neurons within prefrontal cortex of rats repeatedly exposed to amphetamine (AMPH). To test the hypothesis that cortical synchronization may be related to changes in local GABA signaling, we used antibodies specific for parvalbumin (PV), calbindin D28k (CB) and calretinin (CR) as selective labels for three distinct GABA interneuron classes in the anterior cingulate cortex (ACC) of similarly treated rats. We observed a selective increase in the density of PV-immunoreactive (ir), but not CB-ir or CR-ir, neurons in the ACC of AMPH-treated rats at both 1 day and 7 day withdrawal. Increased density of PV-ir GABA interneurons in the ACC at 1 day withdrawal was reproduced in rats repeatedly injected with apomorphine or with SKF-38393. Thus, the critical role of DA receptors during AMPH exposure is evident. However, DA receptor activation did not appear to account for the PV up-regulation in AMPH-treated rats at 7 day withdrawal. Significantly higher numbers of pericellular basket-like puncta immunoreactive for corticotropin-releasing factor (CRF) were observed in the ACC of AMPH rats at 7 day withdrawal. Combined dual immunofluorescence and confocal microscopy further revealed that CRF-ir puncta made possible pericellular contacts on PV-ir (not CB-, CR- or glutamate-ir) cell bodies. A potential cellular mechanism seems to emerge that CRF-ir terminals, that may be underdetected under normal conditions due to low activity levels, may be functionally activated during psychostimulant withdrawal, thereby altering local GABAergic signaling.

Introduction

In our recent in vivo study of amphetamine (AMPH)-withdrawn rats (Onn and Grace, 2000), we found increased synchronization of prefrontal cortical output neurons via augmentation of membrane oscillations. Similar repeated exposure of monkeys to low doses of AMPH induces enduring cognitive and emotional deficits similar to that identified in chronic AMPH abusers (Castner and Goldman-Rakic, 1999; Rogers et al., 1999). Cortical deficits, and particularly a basic dysfunction of prefrontal-accumbens circuits, are critical to the cognitive dysfunction, negative affect and abnormal reward-related behavior that characterize drug abuse (Jentch and Taylor, 1999). It has been frequently proposed that a deficit in intrinsic cortical GABA activity underlies the etiology of drug dependence (Onn and Grace, 2000; Wang et al., 1995; Zhang et al., 2000). GABA local circuit interneurons in neocortex are heterogeneous on the basis of their distinct dendritic and axonal arborization patterns as well as in terms of their co-localization with different calcium-binding proteins (Conde et al., 1994; DeFelipe, 1997; Gabbott et al., 1997; Kawaguchi and Kubota, 1997), i.e. parvalbumin (PV), calbindin D28k (CB) and calretinin (CR). The PV-containing GABA interneuron class, corresponding morphologically to wide arbor (basket) and chandelier cells, constitutes 40–50% of total cortical intrinsic neurons (Gabbott et al., 1997), and is known for its pattern of firing at an extremely high frequency and in bursts (Kawaguchi and Kubota, 1993; Cauli et al., 1997). The CB- and CR-containing classes, each constituting 20–30% of GABA interneurons (Gabbott et al., 1997), are characterized by regular spiking or burst spiking firing patterns (Kawaguchi and Kubota, 1993; Cauli et al., 1997). The PV-containing GABA neurons, which make pericellular synaptic contacts with both pyramidal neurons (Lewis and Lund, 1990; DeFelipe et al., 1989), receive direct DA synapses (Sesack et al., 1995) and exhibit high concentrations of DA D1/D2 receptors (Gaspar et al., 1995; Vincent et al., 1995; Meador-Woodruff et al., 1997; Le Moine and Gaspar, 1998). In contrast, the CB- and CR- containing GABA neurons, which make synapses on dendrites or dendritic spines of pyramidal neurons (DeFelipe and Jones, 1992; del Rio and DeFelipe, 1997; Kawaguchi and Kubota, 1997), exhibit few DA receptors and receive less (to CB-cells) or little (to CR-cells) DA input (Sesack et al., 1995; Le Moine and Gasper, 1998; Muly et al., 1998). Changes in PV immunoreactivity in limbic cortices have been observed to correspond with changes in GABA immunoreactivity under several pathophysiological conditions, e.g. prenatal cocaine exposure (Wang et al., 1995; Stanwood et al., 2001), schizophrenia (Kalus et al., 1997; Woo et al., 1997; Beasley et al., 2002; Reynolds et al., 2002) and epilepsy (Ribak, 1985; DeFelipe et al., 1993; Marco et al., 1996; Silva et al., 2002). Very little is known about the long-term effects of chronic exposure to amphetamine on local GABA signaling in limbic cortices.

In this study, we used immunohistochemistry to evaluate the effects of chronic AMPH exposure in adult rats on prefrontal cortical GABA interneurons identified by the calcium-binding proteins PV, CB and CR. We compared density of cells expressing PV, CB or CR in vehicle- versus AMPH-treated animals. Two time points were chosen for examining cortical intrinsic circuits to better associate alterations with drug-induced actions (i.e. 1 day withdrawal with the presence of AMPH in the brain system) versus alterations with the withdrawal process (i.e. 7 day withdrawal) (Onn and Grace, 2000). To further evaluate the contribution of DA receptors to the cellular changes induced by AMPH, we compared changes in rats treated repeatedly with the DA D1/D2 agonist apomorphine, and with the DA D1 agonist SKF-38393. Given the high levels of corticotropin-releasing factor (CRF) present in withdrawn animals (Richter et al., 1995; Erb et al., 1998; Richter and Weiss, 1999; Zorrilla et al., 2001), we characterized changes in terms of CRF-immunoreactive (ir) puncta in the ACC of AMPH-treated rats, at 1 day or 7 day withdrawal. In addition, using double immunofluorescence and confocal microscopy, we examined CRF-ir puncta in relation to PV-, CB- or CR-ir GABA interneurons as well as glutamate (Glu)-ir pyramidal neurons. Portions of these data were presented in abstract form (Mohila and Onn, 2001, 2002).

Materials and Methods

Animals and Drug Treatment Program

Male Sprague–Dawley rats (n = 56) weighing 150 g at the beginning of treatment were used in this study. Animals were housed two per cage on a 12 h light–dark cycle in a climate-controlled colony with free access to food and water. Rats arriving at the colony were divided into three groups. Rats in the first group of animals (n = 20) were given escalating i.p. doses of AMPH (1.0–7.0 mg/kg) or an equal volume of vehicle (0.85% saline) in their home cages for 10–14 consecutive days to simulate the abuse pattern of drug addicts (Robinson and Becker, 1986; Onn and Grace, 2000). The initial dose of 1.0 mg/kg of AMPH was increased each day by 0.5 mg/kg until a final dose of 7.0 mg/kg was reached. Rats in the second group (n = 16) were given one daily i.p. injection of apomorphine (0.5 mg/kg; 0.02% ascorbic acid) or an equal volume of vehicle solution (0.02% ascorbic acid in 0.85% saline) in their home cages for 10–14 consecutive days. Rats in the third group (n = 20) were given one daily i.p.injection of SKF-38393 (10 mg/kg in 0.85% saline) or an equal volume of vehicle solution (0.85% saline) in their home cage for 10 consecutive days. The injection sites alternated between left and right sides of the abdomen to minimize pain and inflammation. Animals were observed behaviorally after each injection to verify drug effects in which locomotion and stereotypy were observed in rats injected with AMPH, apomorphine or SKF-38393, but not in rats injected with vehicle. All experimental procedures were approved by Drexel University's institutional animal welfare committee in accordance with the NIH Guide for the Care and Use of Laboratory Animals.

Immunoperoxidase and Double Immunofluorescence Staining

At two time points, 1 day withdrawal from the termination of drug treatment and 7 day withdrawal from treatment, a group of 4–6 rats consisting of drug-treated and vehicle-treated controls were deeply anesthetized with 8% chloral hydrate (400 mg/kg, i.p.) and transcardially perfused with ice-cold saline, followed by 4% buffered paraformaldehyde. The brains were removed, postfixed in the same solution for 4 h, and then cryoprotected in 25% sucrose overnight. Tissue sections obtained from control and treatment groups were processed simultaneously for all immunocytochemical staining. Serial coronal sections (60 μm thick) were cut on a freezing microtome and collected in 0.1 M phosphate buffer (PB) solution, pH 7.4. Free-floating sections were washed with PB (3 × 10 min), pretreated with 0.5% H2O2 for 10 min, and then blocked with 5% normal serum diluted in PB containing 0.3% Triton X-100 for 1 h at room temperature. Every sixth section was incubated at 4°C overnight with mouse anti-PV (1:1000; Sigma, St Louis, MO), mouse anti-CB (1:5000; Sigma), rabbit anti-CR (1:2000; Sigma) or goat anti-CRF (1:100; Santa Cruz Biotechnology, Santa Cruz, CA) antibody diluted accordingly in PB and 5% normal serum. After several rinses in cold PB, sections were incubated for 2 h at room temperature with biotinylated secondary antibodies diluted 1:200 in PB and 2% normal serum. Sections were then rinsed three times in cold PB and incubated for 1 h at room temperature in avidin–biotin complex (Vector Laboratories, Burlingame, CA). The immunoreaction product was visualized using 0.005% H2O2 and 0.05% diaminobenzidine (DAB) (Vector Laboratories) as a chromogen. Sections were mounted on gelatin-coated slides, air dried, dehydrated with ethanol, cleared in xylenes and coverslipped using Pro-Texx (Lerner Laboratories, Pittsburgh, PA).

Using double immunofluorescence, we examined CRF immunoreactivity in combination with PV, CB or Glu immunoreactivity. Free-floating sections were incubated at 4°C overnight with goat anti-CRF antibody diluted 1:100 in PB and 5% normal donkey serum. Following several rinses with PB, sections were incubated for 4 h at room temperature with biotinylated donkey anti-goat antibody diluted 1:200 in PB and 2% normal donkey serum. After several rinses with PB, sections were subsequently incubated at 4°C overnight with avidin-Texas Red (Vector Laboratories) diluted 1:100 in PB. Sections were then incubated at 4°C overnight with mouse anti-PV (1:1000), mouse anti-CB (1:5000), rabbit anti-CR (1:2000) or mouse anti-Glu (1:500; Sigma) antibody in PB and 5% normal horse serum. Sections were subsequently incubated for 4 h at room temperature with biotinylated horse anti-mouse antibody diluted 1:200 with PB and 2% normal horse serum. After several rinses with PB, sections were incubated at 4°C overnight with avidin–fluorescein (FITC) (Vector Laboratories) diluted 1:100 in PB. Although two ABC protocols for two distinct primary antibodies were mostly used in this study, an indirect immunofluorescence method was also used to process double labeling for the two individual primary antibodies and yielded similar results. For double-labeling experiments using the two ABC protocols, sections were treated with the avidin–biotin blocking kit (Vector Laboratories) between two individual primary antibodies. To assess nonspecific staining, sections were processed with omission of primary antibody or the secondary bridging serum and showed an absence of immunoreactivity. For double-immunofluorescent labeling experiments, we also verified the absence of any cross-reactivity of the secondary antibodies in control sections by incubating with a single primary antibody and then with the secondary antibody appropriate to the other primary, which yielded no specific staining. Such use of a system for detection of false positives was routinely performed to reliably exclude any potential overlap in detection of the individual fluorophores. Sections were mounted on gelatinized slides and coverslipped with Vectashield (Vector Laboratories). All other chemicals and reagents, unless otherwise indicated, were purchased from Sigma.

Quantitative Data Analysis

Quantitative analyses on the ACC and the parietal cortex were performed by an investigator who was blind to the treatment of the animal from which the particular brain tissue was taken. Immunoperoxidase-stained sections were examined/identified using a Leitz Laborlux S microscope. Digital images of the ACC were captured using a Polaroid DCM camera (Polaroid) and processed using Polaraoid DCM2 and Adobe Photoshop software packages. Images spanning cortical layers II–VI for PV-ir and CR-ir neurons and layers III–VI for CB-ir neurons in the ACC were taken bilaterally using a 10× objective on three corresponding anatomical levels as depicted in Figure 1. The genu of corpus callosum was used as an initial reference point for aligning the counting frame (2.0 mm × 1.5 mm at the 10× objective lens). The frame was then adjusted medially to span layers II–VI of the ACC. Similarly, images of parietal cortices on three anatomical levels were taken bilaterally (using a 4× objective lens) to span layers II–VI within the counting frame (5.0 mm × 3.75 mm at 4× objective lens). Manual counting of immunoreactive neurons was done bilaterally at a final magnification of 350× in digitized color images projected on a high-resolution color monitor. PV-ir, CB-ir and CR-ir neurons were identified if they displayed clear, intense immunoreactivity in the cytoplasm and possessed a distinct nucleus. In CB-immunoreacted sections, layer II, the outer granular cell layer, had both intensely and weakly stained cell bodies that are heterogeneous in two apparent cell types, with weakly-stained cell bodies possibly co-localized with PV (Kawaguchi and Kubota, 1997). Thus, CB-ir perikarya in layer II were excluded from CB cell counts. The density of immunoreactive cells for an area (2.0 mm × 1.5 mm) was then scored. In addition, a second method of cell counting as commonly employed by Chapman and Zahm (1996) was applied to compare with the method indicated above. Each neuron was first identified under a microscope and then traced onto a clear acetate sheet with the aid of a drawing tube attached to the microscope. The border of the ACC, an agranular cortex, can be defined accurately by referring to adjacent Nissl-stained sections. Neurons drawn on acetate sheets, which was later superimposed upon diagrams traced from serial adjacent Nissl-stained sections, were then counted with a grid of same size of counting frame as shown in Figure 1. Both counting methods revealed nearly identical numbers of neurons scored from the same area (i.e. ±1.4 cells; NS).

Figure 1.

Schematic drawing of coronal brain sections containing the cingulate cortex (Cg1–Cg3) and the somatosensory cortex (S1) at three corresponding rostral–caudal levels: 2.2 mm, 1.7 mm and 1.2 mm rostral to the bregma. Shaded boxes were placed over cortical layers II–VI in the Cg1/Cg3 (2.0 mm × 1.5 mm), and in the SI (5.0 mm × 3.75 mm) for quantitative data analyses. Cg1, cingulate cortex area 1; Cg 3, cingulate cortex area 3; Fr2, frontal cortex area 2; SI, primary somatosensory cortex. (Adapted from Paxinos and Watson, 1998.)

Figure 1.

Schematic drawing of coronal brain sections containing the cingulate cortex (Cg1–Cg3) and the somatosensory cortex (S1) at three corresponding rostral–caudal levels: 2.2 mm, 1.7 mm and 1.2 mm rostral to the bregma. Shaded boxes were placed over cortical layers II–VI in the Cg1/Cg3 (2.0 mm × 1.5 mm), and in the SI (5.0 mm × 3.75 mm) for quantitative data analyses. Cg1, cingulate cortex area 1; Cg 3, cingulate cortex area 3; Fr2, frontal cortex area 2; SI, primary somatosensory cortex. (Adapted from Paxinos and Watson, 1998.)

Double immunofluorescence-stained sections were examined using a Leica DMRBE microscope equipped with epifluorescence to visualize Texas Red and FITC fluorescence separately. The digital images were then captured using a Photometric Sensys KAF-1400 CCD camera (Photometric, Inc.) and processed using IP Lab Spectrum (Scanalytics) and Adobe Photoshop software programs.

Confocal Laser Microscopy

Dual immunofluorescent sections were also examined on a confocal laser scanning microscope (Leica TSC SP2) equipped with a Leica DMR microscope to reveal the relationship of CRF-ir puncta to PV-ir perikarya. Using lasers that excite at wavelengths of 488 and 543 nm respectively, FITC (green) and Texas Red (red) fluorophores were visualized. The pinhole apertures were set constant (1 airy unit) for all scans. For each section, the gain and black parameters of the confocal microscope were corrected to ensure that the entire dynamic range of each sample was captured and were kept approximately the same across samples. Z-series were scanned at 0.2 μm increments using a 40× oil immersion lens to examine the relationship between CRF-ir puncta and PV-ir perikarya. Thus, the images shown were the results of optical sections taken at 2 μm intervals and processed using Leica Confocal and Adobe Photoshop software programs.

Quantification of Tissue Area and Volume

To estimate tissue volume in the ACC of vehicle-treated controls (n = 4) and AMPH-treated rats (n = 4), measurements of tissue area (mm2) were performed on Nissl-stained sections of three corresponding anatomical levels (Fig. 1) for each brain. Measurements were taken from the midline to the genu of the corpus callosum (the medial–lateral axis; M-L) and from the genu of the corpus callosum to the pial surface (the dorsal–ventral axis; D-V). M-L and D-V axes at 90° were averaged for each group respectively, to indirectly reflect cross-sectional areas. StereoInvestigator software (MicroBrightField Inc., Williston, VT) was employed to estimate tissue area (mm2) and volume (mm3) on these Nissl-stained sections of vehicle- and AMPH-treated rats (see Notes). Tissue thickness was measured at 4–8 random points within the ACC and then averaged to obtain a mean of tissue thickness for the ACC in each section. The ACC was then outlined and the area and thickness measurements were used to calculate tissue volume, using a modified Cavalieri method (Bothwell et al., 2001).

Statistical Analysis

For all drug treatments, sections at three corresponding anatomical levels (Fig/ 1) of the ACC were analyzed bilaterally; thus six data points were sampled for each rat. The density of cells was expressed as the number of immunoreactive somata per mm2. Data are presented as mean ± standard error of the mean (SEM). Data on tissue volume were tested for normality using a Kolmogorov–Smirnov (KS) test and results that met the criterion of a normal distribution were compared using a Student's t-test as done by Bothwell et al. (2001). Unpaired Student's t-test was used for between-subject comparisons (i.e. between the means of drug- and vehicle-treated groups). Two-way (3 experimental vehicle groups × 2 post-treatment time points) analysis of variance (ANOVA) was used to compare the baseline density of PV-ir cells among different vehicle-treated control groups and post hoc comparisons were used to further identify the source of possible interactions. Differences were considered to be statistically significant at P < 0.05.

Results

Calcium-binding Proteins in the Anterior Cingulate Cortex

Parvalbumin Immunoreactivity

In vehicle-treated adult rats, multipolar-shaped PV-ir cell bodies were scattered throughout cortical layers II–VI (Fig. 2AD). There were no apparent differences in the morphology of PV-ir neurons between vehicle- and AMPH-treated rats. Quantitative analysis revealed an average of 19.79 ± 1.08 (mean ± SE) PV-ir somata/mm2 in the ACC of vehicle-treated control rats at 1 day withdrawal (Fig. 2A; Table 1). Significantly higher densities of PV-positive neurons (24.97 ± 1.47 somata/mm2; P = 0.0061; df = 58) were found in AMPH-treated rats at 1 day withdrawal (Fig. 2C) than in vehicle-treated controls (Fig. 2A). In addition, the density of PV-ir neurons in AMPH-treated rats (25.62 ± 1.25 somata/mm2) remained significantly elevated (P < 0.0001; df = 58) at 7 day withdrawal (Fig. 2D) when compared with the density of PV-ir neurons in vehicle-treated rats (19.63 ± 0.61 somata/mm2) (Fig. 2B). Therefore, significantly higher numbers of PV-positive ACC neurons were present in AMPH-treated rats at both 1 day and 7 day withdrawal than were present in vehicle-treated rats at either of these time points (Fig. 2E; Table 1). When densities of PV-ir neurons were compared in the parietal cortex (a non-limbic cortex), similar numbers were observed between vehicle- and AMPH-treated rats, both at 1 day withdrawal (AMPH: 6.03 ± 0.77 somata/mm2 versus vehicle: 7.95 ± 0.57 somata/mm2; P = 0.055; df = 22; NS) and at 7 day withdrawal (AMPH: 9.97 ± 0.77 somata/mm2 versus vehicle: 9.28 ± 0.66 somata/mm2; P = 0.50; df = 22; NS). Therefore, compared to respective vehicle-treated controls, a significant increase in the number of PV-ir neurons was observed in limbic ACC, but not in the parietal cortex, of AMPH-treated rats at both time points examined.

Figure 2.

Photomicrographs of coronal sections showing PV-ir neurons in the ACC of vehicle-treated rats (A: 1 day withdrawal; B: 7 day withdrawal) and of AMPH-treated rats (C: 1 day withdrawal; D: 7 day withdrawal). Layer II is located to the far left and layer VI to the far right edge of each picture. Higher densities of PV-ir neurons were observed in the ACC of rats treated with AMPH (C: at 1 day withdrawal; D: at 7 day withdrawal), relative to their respective vehicle-treated controls. These qualitative observations were confirmed by quantitative analysis (E) on three anatomical levels as depicted in Figure 1. Asterisks indicate statistical significance as determined by unpaired Student's t-test. (E) AMPH: at 1 day withdrawal (P = 0.0061) and at 7 day withdrawal (P < 0.0001). A-D refer to scale bar in D, 0.5 mm.

Figure 2.

Photomicrographs of coronal sections showing PV-ir neurons in the ACC of vehicle-treated rats (A: 1 day withdrawal; B: 7 day withdrawal) and of AMPH-treated rats (C: 1 day withdrawal; D: 7 day withdrawal). Layer II is located to the far left and layer VI to the far right edge of each picture. Higher densities of PV-ir neurons were observed in the ACC of rats treated with AMPH (C: at 1 day withdrawal; D: at 7 day withdrawal), relative to their respective vehicle-treated controls. These qualitative observations were confirmed by quantitative analysis (E) on three anatomical levels as depicted in Figure 1. Asterisks indicate statistical significance as determined by unpaired Student's t-test. (E) AMPH: at 1 day withdrawal (P = 0.0061) and at 7 day withdrawal (P < 0.0001). A-D refer to scale bar in D, 0.5 mm.

Table 1

Changes in calcium-binding proteins parvalbumin, calbindin and calretinin in the ACC of rats subjected to repeated exposure to AMPH, apomorphine or SKF-38393 at 1 day or at 7 day withdrawal from drug treatment

Calcium-binding proteins
 
Drug treatment groups
 
Withdrawal (days)
 
Positive ir somata/mm2 (percent of vehicle control group)
 
 
PV Vehicle (n = 5) 19.79 ± 1.08 (100 ± 5.5) 
 AMPH (n = 5) 24.97 ± 1.47 (126 ± 7.4)a 
 Vehicle (n = 3)b 18.48 ± 0.83 (100 ± 4.5) 
 Apomorphine (n = 4) 40.79 ± 1.36 (221 ± 7.4)a 
 Vehicle (n = 6)b 21.30 ± 0.75 (100 ± 2.9) 
 SKF-38393 (n = 6) 28.89 ± 0.63 (136 ± 3.5)a 
 Vehicle (n = 5) 19.63 ± 0.61 (100 ± 3.1) 
 AMPH (n = 5) 25.62 ± 1.25 (127 ± 5.7)a 
 Vehicle (n = 4)b 18.63 ± 0.82 (100 ± 4.4) 
 Apomorphine (n = 5) 20.31 ± 0.75 (109 ± 4.0) 
 Vehicle (n = 4)b 21.27 ± 0.48 (100 ± 2.2) 
 SKF-38393 (n = 4) 22.60 ± 0.55 (107 ± 2.6) 
CB Vehicle (n = 57.67 ± 0.31 (100 ± 4.1) 
 AMPH (n = 5) 7.78 ± 0.40 (101 ± 5.2) 
 Vehicle (n = 5) 7.59 ± 0.24 (100 ± 3.2) 
 AMPH (n = 5) 7.07 ± 0.26 (93 ± 3.4) 
CR Vehicle (n = 4) 9.50 ± 0.42 (100 ± 4.2) 
 AMPH (n = 4) 10.42 ± 0.44 (91 ± 4.0) 
 Vehicle (n = 4) 11.29 ± 0.40 (100 ± 3.1) 

 
AMPH (n = 4)
 
7
 
10.90 ± 0.35
 
(103 ± 3.7)
 
Calcium-binding proteins
 
Drug treatment groups
 
Withdrawal (days)
 
Positive ir somata/mm2 (percent of vehicle control group)
 
 
PV Vehicle (n = 5) 19.79 ± 1.08 (100 ± 5.5) 
 AMPH (n = 5) 24.97 ± 1.47 (126 ± 7.4)a 
 Vehicle (n = 3)b 18.48 ± 0.83 (100 ± 4.5) 
 Apomorphine (n = 4) 40.79 ± 1.36 (221 ± 7.4)a 
 Vehicle (n = 6)b 21.30 ± 0.75 (100 ± 2.9) 
 SKF-38393 (n = 6) 28.89 ± 0.63 (136 ± 3.5)a 
 Vehicle (n = 5) 19.63 ± 0.61 (100 ± 3.1) 
 AMPH (n = 5) 25.62 ± 1.25 (127 ± 5.7)a 
 Vehicle (n = 4)b 18.63 ± 0.82 (100 ± 4.4) 
 Apomorphine (n = 5) 20.31 ± 0.75 (109 ± 4.0) 
 Vehicle (n = 4)b 21.27 ± 0.48 (100 ± 2.2) 
 SKF-38393 (n = 4) 22.60 ± 0.55 (107 ± 2.6) 
CB Vehicle (n = 57.67 ± 0.31 (100 ± 4.1) 
 AMPH (n = 5) 7.78 ± 0.40 (101 ± 5.2) 
 Vehicle (n = 5) 7.59 ± 0.24 (100 ± 3.2) 
 AMPH (n = 5) 7.07 ± 0.26 (93 ± 3.4) 
CR Vehicle (n = 4) 9.50 ± 0.42 (100 ± 4.2) 
 AMPH (n = 4) 10.42 ± 0.44 (91 ± 4.0) 
 Vehicle (n = 4) 11.29 ± 0.40 (100 ± 3.1) 

 
AMPH (n = 4)
 
7
 
10.90 ± 0.35
 
(103 ± 3.7)
 

Values are expressed as mean percent (± SEM) of paired vehicle-treated controls [vehicle = 0.85% saline (AMPH- and SKF-38393-groups) or 0.02% ascorbic acid in 0.85% saline (apomorphine-group)].

a

P ≤ 0.05 (Student's t-test) relative to paired vehicle-treated control group.

b

P < 0.05 (one-way ANOVA) comparing vehicle-treated controls of SKF-38393- and apomorphine-group at 1 day or at 7 day withdrawal.

Calbindin and Calretinin Immunoreactivities

To determine whether or not the increased density of ACC neurons expressing PV in AMPH-treated rats is a result of drug-specific actions on this PV-containing subclass of GABA interneurons, we immunostained adjacent brain sections with antibodies specific for two other calcium-binding proteins, CB or CR, which co-localize in two other GABA interneuron classes (DeFelipe, 1997) and have little overlap with PV-ir neurons in deep layers (III–VI) (Kawaguchi and Kubota, 1997). CB-positive non-pyramidal neurons were intensely labeled in both vehicle- and AMPH-treated rats and were mostly bipolar- or multipolar-shaped somata that were easily distinguished from occasional lightly labeled pyramidal-shaped cell bodies (Fig. 3A,C). Quantitative analysis (Fig. 3E) on the density of CB-ir non-pyramidal-shaped cell bodies in the ACC revealed similar densities of CB-ir cells in AMPH-treated versus vehicle-treated control rats at 1 day withdrawal (AMPH: 7.78 ± 0.40 somata/mm2 versus vehicle: 7.67 ± 0.31 somata/mm2; P = 0.828; df = 58; NS) and at 7 day withdrawal (AMPH: 7.07 ± 0.26 somata/mm2 versus vehicle: 7.59 ± 0.24 somata/mm2; P = 0.143; df = 58; NS). CR-positive ACC neurons in both vehicle- and AMPH-treated rats were mostly bipolar-shaped somata and were intensely labeled across most cortical layers II–VI (Fig. 3B, D), with a few cells scattered in layer I. Quantitative analysis (Fig. 3F) also revealed similar densities of CR-ir neurons in the ACC of AMPH-treated versus of vehicle-treated rats, at 1 day withdrawal (AMPH: 10.42 ± 0.44 somata/mm2 versus vehicle: 9.50 ± 0.42 somata/mm2; P = 0.161; df = 46) and at 7 day withdrawal (AMPH: 10.90 ± 0.35 somata/mm2 versus 11.29 ± 0.40 somata/mm2; P = 0.466; df = 46). Therefore, in contrast to the up-regulation in the density of PV cells in the ACC of AMPH-treated rats, the density of CB- or CR-ir GABA interneurons in the ACC was not altered by repeated exposure to AMPH (Table 1).

Figure 3.

Photomicrographs of coronal sections showing CB-ir (A and C) and CR-ir (B and D) neurons in layers II–VI of the ACC in vehicle-treated (A and B) and in AMPH-treated rats (C and D) at 1 day withdrawal. Layer II is located at far left and layer VI to the far right edge of each picture. Similar densities of CB-ir and CR-ir neurons were observed in the ACC of rats treated with AMPH at 1 day withdrawal (C and D respectively) compared with rats treated with vehicle (A and B respectively). Quantitative analysis of CB-ir (E) and CR-ir neurons (F) in the ACC on three coronal sections revealed no significant difference between the numbers of CB-ir neurons in the ACC of AMPH- and vehicle-treated rats (P = 0.828, NS), nor was any difference in the density of CR-ir neurons observed between AMPH- and vehicle-treated rats (P = 0.161, NS) at 1 day withdrawal. Similarly, at 7 day withdrawal no apparent changes were noted for both CB-ir (P = 0.143; NS) and CR-ir neurons (P = 0.466; NS) at 7 day withdrawal. AD refer to scale bar in D, 0.5 mm.

Figure 3.

Photomicrographs of coronal sections showing CB-ir (A and C) and CR-ir (B and D) neurons in layers II–VI of the ACC in vehicle-treated (A and B) and in AMPH-treated rats (C and D) at 1 day withdrawal. Layer II is located at far left and layer VI to the far right edge of each picture. Similar densities of CB-ir and CR-ir neurons were observed in the ACC of rats treated with AMPH at 1 day withdrawal (C and D respectively) compared with rats treated with vehicle (A and B respectively). Quantitative analysis of CB-ir (E) and CR-ir neurons (F) in the ACC on three coronal sections revealed no significant difference between the numbers of CB-ir neurons in the ACC of AMPH- and vehicle-treated rats (P = 0.828, NS), nor was any difference in the density of CR-ir neurons observed between AMPH- and vehicle-treated rats (P = 0.161, NS) at 1 day withdrawal. Similarly, at 7 day withdrawal no apparent changes were noted for both CB-ir (P = 0.143; NS) and CR-ir neurons (P = 0.466; NS) at 7 day withdrawal. AD refer to scale bar in D, 0.5 mm.

Tissue Volume in AMPH-treated Rats

To determine whether the increased density of PV-containing neurons in AMPH rats was a result of drug-induced tissue shrinkage, M-L versus D-V measurements were taken at comparable anatomical sections from AMPH-treated rats. There was no statistically significant difference between groups in area measurements. This was further confirmed by area and volume calculations (StereoInvestigator). When ACC tissue area was compared between AMPH-treated and vehicle-treated control rats, similar values were revealed at both 1 day withdrawal (AMPH: 2.93 ± 0.15 mm2 versus vehicle: 2.62 ± 0.22 mm2; P = 0.27; df = 10; NS) and 7 day withdrawal (AMPH: 3.04 ± 0.09 mm2 versus vehicle: 2.77 ± 0.13 mm2; P = 0.13; df = 10; NS) (Fig. 4A). Similar volumes of the ACC were also derived in AMPH-treated and vehicle-treated control rats at 1 day withdrawal (AMPH: 0.046 ± 0.003 mm3 versus vehicle: 0.044 ± 0.002 mm3; P = 0.53; df = 10; NS) and 7 day withdrawal (AMPH: 0.052 ± 0.003 mm3 versus vehicle: 0.052 ± 0.001 mm3; P = 0.92; df = 10; NS) (Fig. 4B).

Figure 4.

Bar graphs comparing ACC tissue area and volume in vehicle- and AMPH-treated rats at 1 day and 7 day withdrawal. Quantitative analyses revealed no significant differences between the mean ACC tissue area (A) of AMPH- and vehicle-treated rats at either 1 day withdrawal (P = 0.27, NS) or 7 day withdrawal (P = 0.13, NS). No differences were observed when the mean ACC tissue volume was compared (B), at either 1 day (P = 0.53, NS) or 7 day withdrawal (P = 0.92, NS).

Figure 4.

Bar graphs comparing ACC tissue area and volume in vehicle- and AMPH-treated rats at 1 day and 7 day withdrawal. Quantitative analyses revealed no significant differences between the mean ACC tissue area (A) of AMPH- and vehicle-treated rats at either 1 day withdrawal (P = 0.27, NS) or 7 day withdrawal (P = 0.13, NS). No differences were observed when the mean ACC tissue volume was compared (B), at either 1 day (P = 0.53, NS) or 7 day withdrawal (P = 0.92, NS).

Dopamine Receptor Agonists and Parvalbumin Immunoreactivity

PV- and CB-ir GABA interneurons express both DA D1 and D2 receptor subtypes, with a higher density on PV- than on CB-ir cell bodies (Gaspar et al., 1995; Vincent et al., 1995; Meador-Woodruff et al., 1997; Le Moine and Gaspar, 1998). We therefore examined the effects of repeated exposure to apomorphine (a D1/D2 mixed agonist) or to SKF-38393 (a D1 agonist) on PV expression in ACC neurons. Quantitative analysis revealed a significantly higher density of PV-ir neurons at 1 day withdrawal in rats treated with apomorphine (40.79 ± 1.36 somata/mm2 versus vehicle: 18.48 ± 0.83 somata/mm2; P < 0.0001, df = 40) or in rats treated with SKF-38393 (28.89 ± 0.63 somata/mm2 versus vehicle: 21.30 ± 0.75 somata/mm2; P < 0.0001, df = 70). However, unlike in AMPH-treated rats, the increase in PV-ir neuron density at 1 day withdrawal in rats treated with apomorphine or SKF-38393 disappeared as withdrawal continued into day 7 (Table 1). Thus, at 7 day withdrawal in apomorphine-treated rats the average density of PV-ir neurons was 20.31 ± 0.75 somata/mm2, that was similar to that in vehicle-treated rats (18.63 ± 0.82 somata/mm2; P = 0.136; df = 52; NS). Similarly, in SKF-38393-treated rats at 7 day withdrawal, the average density of PV-ir neurons was 22.60 ± 0.55 somata/mm2, that was of no difference from 21.27 ± 0.48 somata/mm2 (P = 0.058; df = 46; NS) in vehicle-treated rats. Thus, unlike in AMPH-treated rats, repeated administration with either DA agonist did not produce lasting effects on PV expression (Table 1).

Two-way (3 experimental vehicle groups × 2 post-treatment time points) ANOVA revealed no effect of post-treatment time or interaction between factors on the density of PV-ir neurons; however, the effect of vehicle treatments reached statistical significance [F(2,156) = 5.51, P = 0.005]. Post hoc analyses further revealed differences in vehicle-treated rats between SKF-38393 group and apomorphine-group, at both 1 day [F(1,52) = 6.89, P = 0.0113] and 7 day withdrawal [F(1,46) = 7.36, P = 0.0094] (Table 1); this may be attributed to the ascorbic acid (an antioxidant) contained in the vehicle for apomorphine treatment. There were no significant differences among other comparisons. Thus, the density of PV-ir somata in vehicle-treated rats of AMPH- versus apomorphine-group was of no difference, either at 1-day [F(1,46) = 0.72, P = 0.399; NS] or at 7 day [F(1,52) = 1.01, P = 0.319; NS] withdrawal. No differences were revealed in vehicle-treated rats between the AMPH and SKF-38393 groups, either at 1 day [F(1,64) = 1.56, P = 0.216, NS] or at 7 day withdrawal [F(1,52) = 3.73, P = 0.059; NS].

Corticotropin-releasing Factor Immunoreactivity

Neurochemical studies have shown elevated concentrations of CRF in rats during psychostimulant withdrawal (Richter et al., 1995; Nemeroff, 1998; Richter and Weiss, 1999). In this study, we compared CRF immunoreactivity in ACC neurons of vehicle- and AMPH-treated rats. CRF-ir cell bodies were faintly labeled in adult ACC (Yan et al., 1998) and cytoplasmic CRF staining appeared homogeneously light brown right up to the rim of nucleus. However, intensely labeled CRF-ir terminal-like puncta were frequently found in the neuropil of the ACC, surrounding somatic profiles in both vehicle- (Fig. 5A) and AMPH-treated rats (Fig. 5B). Occasionally, CRF-ir puncta, in particular in AMPH-treated rat ACC (Fig. 5B3), were observed to cover significant outer portions of cytoplasm, but never abut the rim of nucleus (see Fig. 6), suggesting they derive from nerve endings making possible contacts with the cell body (Lowry et al., 2000). The estimated density of CRF-ir puncta was 4.50 ± 0.24 somata/mm2 in the ACC of AMPH-treated rats at 1 day withdrawal, by counting the number of somata that were apposed by CRF-ir puncta (Fig. 5C). This was not significantly different from the average 4.97 ± 0.29 somata/mm2 in the ACC of vehicle-treated control rats (P = 0.3173; df = 82; NS). However, at 7 day withdrawal of AMPH-treated rats, a significantly higher density of somata was apposed by CRF-ir puncta (9.02 ± 0.33 somata/mm2) relative to the average of 5.44 ± 0.30 cells/mm2 in vehicle-treated controls (P < 0.0001; df = 82; Fig. 5C).

Figure 5.

Photomicrographs of coronal sections showing CRF-ir puncta scattered throughout layers II–VI of the ACC in vehicle- (A) and AMPH-treated (B) rats at 7 day withdrawal. CRF-ir puncta appeared higher in density in AMPH-treated rats at 7 day withdrawal (B1: low power; B2 and B3: high power) than in vehicle-treated controls (A). Higher-power images (B2 and B3: 25×) of a region in B1 illustrate the morphology of CRF-ir puncta in close apposition with CRF-containing PV-ir cell bodies (arrows). (C) Bar graph compares the average number of cell bodies encircled by CRF-ir puncta in the ACC of vehicle- versus AMPH-treated rats. A significant increase in CRF-ir puncta was observed in AMPH-rats compared to vehicle-treated controls at 7 day withdrawal (P < 0.0001), but not at 1 day withdrawal (P = 0.317, NS). Asterisks indicate statistical significance as determined by unpaired Student's t-test. AB1 refer to scale bar in B1, 0.5 mm; scale bar in B2, 0.2 mm; B3, 0.1 mm.

Figure 5.

Photomicrographs of coronal sections showing CRF-ir puncta scattered throughout layers II–VI of the ACC in vehicle- (A) and AMPH-treated (B) rats at 7 day withdrawal. CRF-ir puncta appeared higher in density in AMPH-treated rats at 7 day withdrawal (B1: low power; B2 and B3: high power) than in vehicle-treated controls (A). Higher-power images (B2 and B3: 25×) of a region in B1 illustrate the morphology of CRF-ir puncta in close apposition with CRF-containing PV-ir cell bodies (arrows). (C) Bar graph compares the average number of cell bodies encircled by CRF-ir puncta in the ACC of vehicle- versus AMPH-treated rats. A significant increase in CRF-ir puncta was observed in AMPH-rats compared to vehicle-treated controls at 7 day withdrawal (P < 0.0001), but not at 1 day withdrawal (P = 0.317, NS). Asterisks indicate statistical significance as determined by unpaired Student's t-test. AB1 refer to scale bar in B1, 0.5 mm; scale bar in B2, 0.2 mm; B3, 0.1 mm.

Figure 6.

Photomicrographs showing a close relation between CRF-ir puncta and PV-immunopositive (B) or −negative (C) somata in the ACC of a vehicle-treated rat (AC) and an AMPH-treated rat (DG) at 7 day withdrawal. PV-ir neurons and processes were marked by FITC fluorescence (green) whereas CRF-ir puncta were labeled with Texas Red fluorescence (red) in the neuropil of the ACC. Arrowheads in (A) indicate dual-labeled CRF-ir puncta and PV-positive cell bodies, whereas arrows mark CRF-ir puncta in apposition to PV-negative somata. (B1, B2, B3) High-power views of the neuron marked with asterisk in (A) showing punctate CRF-ir varicosities in apposition to a PV-ir cell body and its proximal dendrites. In addition, CRF-ir was also detected in the PV-ir cell body, i.e. CRF co-localized with PV in the cytoplasm (yellow in B3). This is an example of CRF-ir puncta (non-PV containing) making possible pericellular contacts on a GABA interneuron containing both CRF and PV (Lewis and Lund, 1990; DeFelipe, 1997). (C1, C2, C3) High-power views of neurons marked with arrows in (A) showing CRF-ir puncta apposed cell bodies that were negatively-labeled with PV immunoreactivity. (DG) Photomicrographs showing dual labeling for CRF-ir puncta and PV-ir cell bodies (D1–D3), CB-ir cell bodies (E), CR-ir cell bodies (F) or Glu-ir cell bodies (G) in the ACC of AMPH-treated rats at 7 day withdrawal. (B1, C1, D1) FITC-labeled PV-ir cell bodies (green). (B2, C2, D2) Texas Red-labeled CRF-ir puncta (orange red). (B3, C3, D3) Dual FITC and Texas Red fluorescence, indirectly marking PV-ir (green) and CRF-ir (red), respectively. The yellow color indicates the merging of FITC and Texas Red fluorescence. Texas Red, although mostly red color, has a significant orange hue (Vector Laboratories, Inc), and therefore may appear orange-red (Gonchar et al., 2002) as a result of intense fluorescence labeling. Scale bars: A, 50 μm; B1–B3 refer to scale bar in B3, 25 μm. C1–C3 refer to scale bar in C3, 25 μm. D1–D3 refer to scale bar in D3, 25 μm. E–G, 25 μm.

Figure 6.

Photomicrographs showing a close relation between CRF-ir puncta and PV-immunopositive (B) or −negative (C) somata in the ACC of a vehicle-treated rat (AC) and an AMPH-treated rat (DG) at 7 day withdrawal. PV-ir neurons and processes were marked by FITC fluorescence (green) whereas CRF-ir puncta were labeled with Texas Red fluorescence (red) in the neuropil of the ACC. Arrowheads in (A) indicate dual-labeled CRF-ir puncta and PV-positive cell bodies, whereas arrows mark CRF-ir puncta in apposition to PV-negative somata. (B1, B2, B3) High-power views of the neuron marked with asterisk in (A) showing punctate CRF-ir varicosities in apposition to a PV-ir cell body and its proximal dendrites. In addition, CRF-ir was also detected in the PV-ir cell body, i.e. CRF co-localized with PV in the cytoplasm (yellow in B3). This is an example of CRF-ir puncta (non-PV containing) making possible pericellular contacts on a GABA interneuron containing both CRF and PV (Lewis and Lund, 1990; DeFelipe, 1997). (C1, C2, C3) High-power views of neurons marked with arrows in (A) showing CRF-ir puncta apposed cell bodies that were negatively-labeled with PV immunoreactivity. (DG) Photomicrographs showing dual labeling for CRF-ir puncta and PV-ir cell bodies (D1–D3), CB-ir cell bodies (E), CR-ir cell bodies (F) or Glu-ir cell bodies (G) in the ACC of AMPH-treated rats at 7 day withdrawal. (B1, C1, D1) FITC-labeled PV-ir cell bodies (green). (B2, C2, D2) Texas Red-labeled CRF-ir puncta (orange red). (B3, C3, D3) Dual FITC and Texas Red fluorescence, indirectly marking PV-ir (green) and CRF-ir (red), respectively. The yellow color indicates the merging of FITC and Texas Red fluorescence. Texas Red, although mostly red color, has a significant orange hue (Vector Laboratories, Inc), and therefore may appear orange-red (Gonchar et al., 2002) as a result of intense fluorescence labeling. Scale bars: A, 50 μm; B1–B3 refer to scale bar in B3, 25 μm. C1–C3 refer to scale bar in C3, 25 μm. D1–D3 refer to scale bar in D3, 25 μm. E–G, 25 μm.

In an effort to identify those unlabeled somatic profiles surrounded by CRF-ir puncta, we undertook double immunofluorescence using antibodies specific for PV, CB or Glu and antibody for CRF-ir puncta. Double immunofluorescence for PV and CRF (Fig. 6) confirmed similar levels of increases in CRF-ir puncta, as quantified on brain sections stained only for CRF (Fig. 5). Both PV-immunopositive (Fig. 6A,B,D) and PV-immunonegative neurons (Fig. 6C) in the ACC were apposed by CRF-ir puncta in both vehicle- (Fig. 6AC) and AMPH-treated rats (Fig. 6DG). In addition, dual immunofluorescence for CRF and CB, CR or Glu revealed no cases of apposition between CRF-ir puncta and CB- (Fig. 6E), CR- (Fig. 6F) or Glu-ir cell bodies (Fig. 6G). On a few occasions, CRF-ir puncta were, however, observed to coexist with Glu-ir puncta that surrounded unlabeled somatic profiles (Fig. 6G). Confocal microscopic images (Fig. 7) illustrated the three-dimensional views of CRF-ir varicosities and PV-ir soma by rotating these two structures every 20° between images 1–6 (with each image stacked through several optical sections scanned at 2 μm intervals). The three-dimensional confocal images of CRF-ir varicosities (red) and PV-ir soma (green) revealed the apparent lack of coexistence between these two antigens (CRF and PV) in CRF-ir puncta (red), and cytoplasmic co-localization (yellow) of CRF and PV. Therefore, this is an example showing that CRF-ir puncta were apposed to a CRF-/PV-positive soma, a CRF-containing PV-ir GABA interneuron (Lewis and Lund, 1990; DeFelipe, 1997).

Figure 7.

Confocal microscopy images of a neuron double-labeled with both CRF and PV in the ACC of a vehicle-treated control rat. The images shown (1–6) are the stacked image of several optical sections (scanned at 2 μm intervals apart) of the neuron shown in Figure 6A (marked by an arrowhead with an asterisk) or in Figure 6B1–3. A rotation of 200 was applied between images 1–6. The three-dimensional views of CRF-ir varicosities and PV-ir soma by rotating these structures reliably revealed their interrelationship. The presence of a close relation (but the lack of coexistence) was established between CRF-ir puncta and PV-ir somata and proximal dendrites, but the presence of co-localization of CRF and PV was present in the cytoplasm. (A1–6) FITC-labeled PV-ir cell body and processes. (B1–6) Texas Red-labeled CRF-ir puncta. (C1–6) Dual-fluorescence labeling for both PV and CRF immunoreactivity. The yellow color indicates the merging of both green (PV-ir) and red (CRF-ir) fluorescence in the cytoplasm. Scale bar, 20 μm.

Figure 7.

Confocal microscopy images of a neuron double-labeled with both CRF and PV in the ACC of a vehicle-treated control rat. The images shown (1–6) are the stacked image of several optical sections (scanned at 2 μm intervals apart) of the neuron shown in Figure 6A (marked by an arrowhead with an asterisk) or in Figure 6B1–3. A rotation of 200 was applied between images 1–6. The three-dimensional views of CRF-ir varicosities and PV-ir soma by rotating these structures reliably revealed their interrelationship. The presence of a close relation (but the lack of coexistence) was established between CRF-ir puncta and PV-ir somata and proximal dendrites, but the presence of co-localization of CRF and PV was present in the cytoplasm. (A1–6) FITC-labeled PV-ir cell body and processes. (B1–6) Texas Red-labeled CRF-ir puncta. (C1–6) Dual-fluorescence labeling for both PV and CRF immunoreactivity. The yellow color indicates the merging of both green (PV-ir) and red (CRF-ir) fluorescence in the cytoplasm. Scale bar, 20 μm.

Discussion

We labeled three distinct classes of GABA interneurons in the ACC using antibodies specific for calcium-binding proteins. We observed a selective increase in the density of PV-ir (but not CB-ir or CR-ir) neurons in the ACC of rats repeatedly exposed to AMPH. The increase in the density of PV-ir neurons was not attributed to shrinkage of brain tissue volume, because similar tissue volumes of ACC were measured in vehicle- and AMPH-treated rats. Instead, this increase appears to result from increased PV expression, and therefore detectability of PV-ir, in the ACC of AMPH-treated rats. This result suggests that dominant effects of AMPH in the ACC may be channeled via PV-containing interneurons, which may be due to the presence of high levels of DA receptors on PV-interneurons (see Introduction). A critical role of DA receptors during AMPH exposure is evident, as activation of these receptors by apomorphine or by SKF-38393 mimicked the preferential influences of AMPH on PV-containing GABA interneurons at 1 day withdrawal. The sustained PV up-regulation in AMPH-treated rats at 7 day withdrawal is accompanied by an increase in pericellular labeling for CRF. Confocal microscopic examinations revealed a close relation (but no coexistence) between CRF-ir puncta and PV-ir somata, suggesting an association between increased CRF terminal activity and sustained PV expression at 7 day AMPH withdrawal. The present findings have implications for understanding the primary role of DA receptors and/or the secondary role of CRF peptides in changes associated with PV-containing GABA local circuit neurons during AMPH withdrawal.

PV calcium-binding protein evolved as a functionally distinct, physiologically relevant modulator of intracellular calcium transients that is critical for synaptic plasticity, as PV-knockout mice fail to maintain synaptic transmission during high-frequency firing (Caillard et al., 2000). Cortical GABA neurons co-localized with PV, often termed fast-spiking (FS) neurons, are known to fire at an extremely high frequency (Baimbridge et al., 1992; Kawaguchi and Kubota, 1993). The increase in PV after AMPH treatment may be protective against prolonged bursting activity in this subclass of GABA interneurons, as an increase in the level of PV in GABA neurons can be induced by bursting activity (Chard et al., 1993). Furthermore, GAD activity in GABA neurons is regulated by levels of internal calcium and is increased after tetanic stimulation (Erecinska et al., 1996). Thus, increased PV expression in this study is believed to reflect increased GABA neuron activity. A corresponding increase in both PV and GABA immunoreactivities was reported in the ACC of rabbits subjected to prenatal (intrauterine) cocaine exposure (Wang et al., 1995). The PV-containing GABA interneurons, corresponding morphologically to wide arbor (basket) and chandelier cells, have their axonal arbors elaborating mostly horizontally/vertically to make chemical synapses on somata/axon initial segments of pyramidal neurons (DeFelipe et al., 1989; DeFelipe, 1997; Lewis and Lund, 1990; Cruz et al., 2003) and strongly regulate cortical output activity. However, PV is only present in a subpopulation of chandelier axon cartridges (Lewis and Lund, 1990; Del Rio and DeFelipe, 1997; Cruz et al., 2003), thus heterogeneity is present among basket and chandelier neurons. CB and somatostatin were recently identified in axo-axonic terminals in making synapses on many axon initial segments of pyramidal neurons in human temporal cortex (DeFelipe, 1997) and rat visual cortex (Gonchar et al., 2002), respectively. Based on our present findings, it is justified to assume that increased PV neuron density in AMPH rats is a result of drug-induced increases in its cellular expression and consequent detectability in previously undetected PV-neurons, and probably not due to emergence of a new set of neurons. PV-containing FS GABA interneurons can contact tens of PV-containing GABA neurons (Galarreta and Hestrin, 1999; Gibson et al., 1999; Gonchar and Burkhalter, 1999; Amitai et al., 2002) that could lead to disinhibition of pyramidal neurons. These complex inhibitory/disinhibitory processes within GABA interneuronal networks can induce cortical oscillatory activities (Cobb et al., 1995) and synchronize gamma oscillations (Traub et al., 2001). A differentiated pattern of changes in different calcium binding proteins (i.e. PV as opposed to no changes in calbindin and calretinin) and its functional significance in altering dendritic branching pattern and spine density (Robinson and Kolb, 1997) and neuronal synchronization in pyramidal neurons of AMPH-withdrawn rats (Onn and Grace, 2000) should encourage further studies into detailing cortical microcircuits in amphetamine addicts.

Two time points were chosen in the present study to associate cellular changes in PV either with drug treatment (i.e. at 1 day withdrawal) or with the withdrawal process (i.e. at 7 day withdrawal). Repeated exposure to apomorphine (a DA D1/D2 agonist) or to SKF-38393 (a D1 agonist) can also induce PV up-regulation during the treatment period (i.e. at 1 day withdrawal), suggesting that the PV increases at 1 day withdrawal of AMPH-treated rats may result from repeated stimulation of DA receptors. A larger increase in the density of PV-ir neurons was noted in rats repeatedly exposed to apomorphine than in rats exposed to SKF-38393. The increase in the former group is reminiscent of what was observed in AMPH rats. This may be explained by the fact that D1 and D2 receptors are preferentially co-localized to PV-containing GABA interneurons (see Introduction). The involvement of amines other than DA cannot be excluded, as AMPH releases DA and other amines (e.g. norepinephrine and serotonin). However, the lack of PV changes in the parietal cortex of AMPH-treated rats (present study) speaks for the critical involvement of the mesolimbic/mesocortical DA system in PV changes, because parietal cortex, a non-limbic cortex, contains low levels of DA fibers (but high levels of noradrenergic fibers: Stanwood et al., 2001). Similarly, psychostimulants (AMPH and cocaine) increase c-fos activity, i.e. increased neuronal activity (Sharp et al., 1993; Chapman and Zahm, 1996), only in cortical structures (e.g. the prefrontal cortex) that tie closely to the mesolimbic/mesocortical DA system) and thereby the ventral basal ganglia loop linked to drug-seeking behavior (McFarland and Kalivas, 2001).

Compelling evidence exists for a regulatory role of DA receptors on GABA activity within the prefrontal cortex (PFC). Neurochemically, activation of DA D2 receptors within the PFC by D2 agonists increases spontaneous GABA release (Retaux et al., 1991; Grobin and Deutch, 1998) and blockade of these D2 receptors decreases GABA levels (Bourdelais and Deutch, 1994). Physiologically, DA excites cortical GABA interneurons (Penit-Soria et al., 1987; Zhou and Hablitz, 1999), probably via D1 receptors (Seamans et al., 2001). Thus, activation of DA D1 receptors within the PFC can increase membrane excitability of GABA interneurons, thereby enhancing GABA release (Seamans et al., 2001). On the other hand, Gao et al. (2003) have recently shown that DA via presynaptic D1 receptors reduced inhibitory postsynaptic currents arising only from axons of FS interneurons (PV-containing), but not from axons of non-FS interneurons (e.g. CB- and CR-containing). This non-uniform DA action, preferentially targeting axons of FS-GABA interneurons in the ACC, is consistent with our postmortem observations that showed the preferential up-regulation of PV expression in the ACC of AMPH-treated rats.

DA receptor activation however did not appear to account for the up-regulation of PV observed in AMPH-treated rats at 7 day drug withdrawal: different cellular mechanisms may account for this persistent PV up-regulation. Our present study showed increases in densities of both CRF-ir puncta and PV-ir cell bodies in AMPH-treated rats at 7 day withdrawal. Interestingly, we observed no cases of CB-, CR- or Glu-containing neurons that were apposed by CRF-ir puncta. The crisp appearance of CRF-varicosities could only be obtained by incubating CRF-primary alone (followed later by PV-primary), not by the cocktail method (incubating both CRF-primary and PV-primary together). This may be caused by the fact that CRF peptides are delicate and in low concentrations in axons compared to the level and resilience of calcium-binding proteins in cell bodies. Nevertheless, sequential incubations of CRF-ir puncta with PV (or CB, CR, Glu) antibodies in double-labeling experiments revealed similar increases in CRF-ir puncta by single immunoperoxidase stains on brain sections reacted only for CRF (Fig. 5). As CRF and PV coexist in some basket/chandelier cells (Lewis and Lund, 1990; DeFelipe, 1997), we observed a few cases of cytoplasmic co-localization of CRF and PV (Fig. 6B). In those cases, cytoplasmic CRF-ir staining appeared yellow in the merged images (of two individual fluorophores red and green; Fig. 7) and can be differentiated from CRF-varicosities in red (rarely co-localized with PV). These interrelated pre-/postsynaptic structures between CRF-varicosities and PV-ir cell bodies can be easily discerned in three-dimensional views by rotating these structures every 20°, as exemplified in Figure 7. Increased CRF-ir puncta and their exclusive association with PV-ir neurons in the ACC suggest a causal relationship between increased CRF release (Richter et al., 1995; Erb et al., 1998; Richter and Weiss, 1999; Zorrilla et al., 2001) and sustained PV expression during AMPH withdrawal (present study).

CRF-ir puncta within neocortex could arise from CRF-containing local circuit neurons (see Fig. 6B2) that overlap to some extent with PV-ir neurons (Sakanaka et al., 1987; Lewis et al., 1989; Yan et al., 1998). CRF-ir GABA interneurons (Lewis et al., 1989), were mostly lightly stained in the adult ACC (Yan et al., 1998). This was attributed to low levels of CRF peptides in cell bodies in the mature neocortex, as CRF-ir neurons in adult rat neocortex are often undetected, and thereby appear less numerous than those stained in immature rat cortex (Yan et al., 1998). Colchicine pretreatment is commonly used to block axonal transport to increase detectability of CRF-ir cell bodies in adult neocortex (Sakanaka et al., 1987; Yan et al., 1998). This technical difficulty has prevented us from evaluating changes in CRF-ir local circuit neurons to further associate them with the increased CRF-ir puncta during withdrawal. However, an extrinsic source for CRF-ir puncta in the ACC could arise from the amygdala or hypothalamus (Bloom et al., 1982; Swanson et al., 1983; Daikoku et al., 1985; Richter et al., 1995) that uses Glu as the neurotransmitter. Our occasional observation (Fig. 6G) of the coexistence between CRF-ir and Glu-ir puncta in AMPH-withdrawn rats favors such a possibility. CRF from an extrinsic source (i.e. from amydgala and hypothalamus) has been shown to change firing activity in both serotonergic and noradrenergic neurons by making possible pericellular contacts with their cell bodies (Siggins et al., 1985; Valentino and Foote, 1988; Lowry et al., 2000). It is highly possible that the increase in CRF-ir puncta in the ACC during withdrawal may exert significant influence over cortical inhibitory tone by directly modulating PV expression. CRF peptides, present at extremely low levels in axon terminals, may go undetected in normal (vehicle-treated) animals. Increased levels of CRF in axon terminals during withdrawal are presumably due to axonal transport from the somata, where CRF is synthesized, and may act to cope with the increased demand of terminal release during psychostimulant withdrawal. Increased CRF release in limbic cortices of AMPH-withdrawn rats may be a mechanism whereby altered GABA inhibitory tone during withdrawal leads to relapse to drug seeking (Richter et al., 1995; Erb et al., 1998; Richter and Weiss, 1999; Zorrilla et al., 2001).

The authors would like to thank Drs Gloria Meredith and Adrian Dervan at Finch University of Health Sciences/The Chicago Medical School for their evaluation of cortical tissue volume and Dr Gloria Meredith for fruitful discussions on cellular changes in calcium-binding protein expression and Dr David Lewis at University of Pittsburgh for his excellent advice on heterogeneity among basket and chandelier cells. The authors would also like to thank Drs Jed Shumsky and Chris Tobias for their assistance in statistical analyses and Ms Ivy Z. Li for her assistance in quantitative analyses. This work was supported by MH 63498 and by MCPHU and Drexel University Synergy Grants.

References

Amitai Y, Gibson JR, Beierlein M, Patrick SL, Ho AM, Connors BW, Golomb D (
2002
) The spatial dimensions of electrically coupled networks of interneurons in the neocortex.
J Neurosci
 
22
:
4142
–4152.
Baimbridge KG, Celio MR, Rogers JH (
1992
) Calcium-binding proteins in the nervous system.
Trends Neurosci
 
15
:
303
–308.
Beasley CL, Zhang ZJ, Patten I, Reynolds GP (
2002
) Selective deficits in prefrontal cortical GABAergic neurons in schizophrenia defined by the presence of calcium-binding proteins.
Biol Psychiatry
 
52
:
708
–715.
Bloom FE, Battenberg EL, Rivier J, Vale W (
1982
) Corticotropin releasing factor (CRF): immunoreactive neurones and fibers in rat hypothalamus.
Regul Pept
 
4
:
43
–48.
Bothwell S, Meredith GE, Phillips J, Staunton H, Doherty C, Grigorenko E, Glazier S, Deadwyler SA, O'Donovan CA, Farrell M (
2001
) Neuronal hypertrophy in the neocortex of patients with temporal lobe epilepsy.
J Neurosci
 
21
:
4789
–4800.
Bourdelais AJ, Deutch AY (
1994
) The effects of haloperidol and clozapine on extracellular GABA levels in the prefrontal cortex of the rat: an in vivo microdialysis study.
Cereb Cortex
 
4
:
69
–77.
Caillard O, Moreno H, Schwaller B, Llano I, Celio MR, Marty A (
2000
) Role of the calcium-binding protein parvalbumin in short-term synaptic plasticity.
Proc Natl Acad Sci U S A
 
97
:
13372
–13377.
Castner SA, Goldman-Rakic PS (
1999
) Long-lasting psychotomimetic consequences of repeated low-dose amphetamine exposure in rhesus monkeys.
Neuropsychopharmacology
 
20
:
10
–28.
Cauli B, Audinat E, Lambolez B, Angulo MC, Ropert N, Tsuzuki K, Hestrin S, Rossier J (
1997
) Molecular and physiological diversity of cortical nonpyramidal cells.
J Neurosci
 
17
:
3894
–3906.
Chapman MA, Zahm DS (
1996
) Altered Fos-like immunoreactivity in terminal regions of the mesotelencephalic dopamine system is accompanied by partial reappearance of TH immunoreactivity at the sites of focal 6-hydroxydopamine lesions in the nucleus accumbens.
Brain Res.
 
736
:
270
–279.
Chard PS, Bleakman D, Christakos S, Fullmer CS, Miller RJ (
1993
) Calcium buffering properties of calbindin D28k and parvalbumin in rat sensory neurones.
J Physiol
 
472
:
341
–357.
Cobb SR, Buhl EH, Halasy K, Paulsen O, Somogyi P (
1995
) Synchronization of neuronal activity in hippocampus by individual GABAergic interneurons.
Nature
 
378
:
75
–78.
Conde F, Lund JS, Jacobowitz DM, Baimbridge KG, Lewis DA (
1994
) Local circuit neurons immunoreactive for calretinin, calbindin D-28k or parvalbumin in monkey prefrontal cortex: distribution and morphology.
J Comp Neurol
 
341
:
95
–116.
Cruz DA, Eggan SM, Lewis DA (
2003
) Postnatal development of pre- and postsynaptic GABA markers at chandelier cell connections with pyramidal neurons in monkey prefrontal cortex.
J Comp Neurol
 
465
:
385
–400.
Daikoku S, Okamura Y, Kawano H, Tsuruo Y, Maegawa M, Shibasaki T (
1985
) CRF-containing neurons of the rat hypothalamus.
Cell Tissue Res
 
240
:
575
–584.
DeFelipe J (
1997
) Types of neurons, synaptic connections and chemical characteristics of cells immunoreactive for calbindin-D28K, parvalbumin and calretinin in the neocortex.
J Chem Neuroanat
 
14
:
1
–19.
DeFelipe J, Hendry SHC, Jones EG (
1989
) Visualization of chandelier cell axons by parvalbumin immunoreactivity in monkey cerebral cortex.
Proc Natl Acad Sci USA
 
86
:
2093
–2097.
DeFelipe J, Garcia Sola R, Marco P, del Rio MR, Pulido P, Ramon y Cajal S (
1993
) Selective changes in the microorganization of the human epileptogenic neocortex revealed by parvalbumin immunoreactivity.
Cereb Cortex
 
3
:
39
–48.
Del Rio MR, DeFelipe J (
1997
) Colocalization of parvalbumin and calbindin D-28K in neurons, including chandelier cells of human temporal neocortex.
J Comp Neurol
 
342
:
389
–408.
Erb S, Shaham Y, Stewart J (
1998
) The role of corticotropin-releasing factor and corticosterone in stress- and cocaine-induced relapse to cocaine seeking in rats.
J Neurosci
 
18
:
5529
–5536.
Erecinska M, Nelson D, Daikhin Y, Yudkoff M (
1996
) Regulation of GABA level in rat brain synaptosomes: fluxes through enzymes of the GABA shunt and effects of glutamate, calcium, and ketone bodies.
J Neurochem
 
67
:
2325
–2334.
Gabbott PL, Dickie BG, Vaid RR, Headlam AJ, Bacon SJ (
1997
) Local-circuit neurones in the medial prefrontal cortex (areas 25, 32 and 24b) in the rat: morphology and quantitative distribution.
J Comp Neurol
 
377
:
465
–499.
Galarreta M, Hestrin S (
1999
) A network of fast-spiking cells in the neocortex connected by electrical synapses.
Nature
 
402
:
72
–75.
Gao WJ, Wang Y, Goldman-Rakic PS (
2003
) Dopamine modulation of perisomatic and peridendritic inhibition in prefrontal cortex.
J Neurosci
 
23
:
1622
–1630.
Gaspar P, Bloch B, Le Moine C (
1995
) D1 and D2 receptor gene expression in the rat frontal cortex: cellular localization in different classes of efferent neurons.
Eur J Neurosci
 
7
:
1050
–1063.
Gibson JR, Beierlein M, Connors BW (
1999
) Two networks of electrically coupled inhibitory neurons in neocortex.
Nature
 
402
:
75
–79.
Gonchar Y, Burkhalter A (
1999
) Differential subcellular localization of forward and feedback interareal inputs to parvalbumin expressing GABAergic neurons in rat visual cortex.
J Comp Neurol
 
406
:
346
–360.
Gonchar Y, Turney S, Price JL, Burkhalter A (
2002
) Axo-axonic synapses formed by somatostatin-expressing GABAergic neurons in rat and monkey visual cortex.
J Comp Neurol
 
443
(1):
1
–14.
Grobin AC, Deutch AY (
1998
) Dopaminergic regulation of extracellular gamma-aminobutyric acid levels in the prefrontal cortex of the rat.
J Pharmacol Exp Ther
 
285
:
350
–357.
Jentsch JD, Taylor JR (
1999
) Impulsivity resulting from frontostriatal dysfunction in drug abuse: implications for the control of behavior by reward-related behaviors.
Psychopharmacology
 
146
:
373
–390.
Kalus P, Senitz D, Beckmann H (
1997
) Altered distribution of parvalbumin-immunoreactive local circuit neurons in the anterior cingulate cortex of schizophrenic patients.
Psychiatry Res
 
75
:
49
–59.
Kawaguchi Y, Kubota Y (
1993
) Correlation of physiological subgroupings of nonpyramidal cells with parvalbumin- and calbindinD28k-immunoreactive neurons in layer V of rat frontal cortex.
J Neurophysiol
 
70
:
387
–396.
Kawaguchi Y, Kubota Y (
1997
) GABAergic cell subtypes and their synaptic connections in rat frontal cortex.
Cereb Cortex
 
7
:
476
–486.
Le Moine C, Gaspar P (
1998
) Subpopulations of cortical GABAergic interneurons differ by their expression of D1 and D2 dopamine receptor subtypes.
Brain Res Mol Brain Res
 
58
:
231
–236.
Lewis DA, Lund JS (
1990
) Heterogeneity of chandelier neurons in monkey neocortex: corticotropin-releasing factor and parvalbumin immunoreactive populations.
J Comp Neurol
 
293
:
599
–615.
Lewis DA, Foote SL, Cha CI (
1989
) Corticotropin-releasing factor immunoreactivity in monkey neocortex: an immunohistochemical analysis.
J Comp Neurol
 
290
:
599
–613.
Lowry CA, Rodda JE, Lightman SL, Ingram CD (
2000
) Corticotropin-releasing factor increases in vitro firing rates of serotonergic neurons in the rat dorsal raphe nucleus: evidence for activation of a topographically organized mesolimbocortical serotonergic system.
J Neurosci
 
20
:
7728
–7736.
Marco P, Sola RG, Pulido P, Alijarde MT, Sanchez A, Ramon y Cajal S, DeFelipe J (
1996
) Inhibitory neurons in the human epileptogenic temporal neocortex. An immunocytochemical study.
Brain
 
119
:
1327
–1347.
McFarland K, Kalivas P (
2001
) The circuitry mediating cocaine-induced reinstatement of drug-seeking behavior.
J Neurosci.
  2001 Nov 1;
21
(21):
8655
–63.
Meador-Woodruff JH, Haroutunian V, Powchik P, Davidson M, Davis KL, Watson SJ (
1997
) Dopamine receptor transcript expression in striatum and prefrontal and occipital cortex. Focal abnormalities in orbitofrontal cortex in schizophrenia.
Arch Gen Psychiatry
 
54
:
1089
–1095.
Mohila CA, Onn S-P (
2001
) Immunohistochemical characterization of GABA intrinsic circuits in prefrontal cortex of amphetamine-withdrawn rats.
Soc Neurosci Abstr
 
31
:728.710.
Mohila CA, Onn S-P (
2002
) Selective increases in parvalbumin expression in GABA interneurons in anterior cingulate cortex (ACC) of amphetamine rats.
Soc Neurosci Abstr
 
31
:430.410.
Muly EC, 3rd, Szigeti K, Goldman-Rakic PS (
1998
) D1 receptor in interneurons of macaque prefrontal cortex: distribution and subcellular localization.
J Neurosci
 
18
:
10553
–10565.
Nemeroff CB (
1998
) Psychopharmacology of affective disorders in the 21st century.
Biol Psychiatry
 
44
:
517
–525.
Onn SP, Grace AA (
2000
) Amphetamine withdrawal alters bistable states and cellular coupling in rat prefrontal cortex and nucleus accumbens neurons recorded in vivo.
J Neurosci
 
20
:
2332
–2345.
Penit-Soria J, Audinat E, Crepel F (
1987
) Excitation of rat prefrontal cortical neurons by dopamine: an in vitro electrophysiological study.
Brain Res
 
425
:
263
–274.
Retaux S, Besson MJ, Penit-Soria J (
1991
) Opposing effects of dopamine D2 receptor stimulation on the spontaneous and the electrically evoked release of [3H]GABA on rat prefrontal cortex slices.
Neuroscience
 
42
:
61
–71.
Reynolds GP, Beasley CL, Zhang ZJ (
2002
) Understanding the neurotransmitter pathology of schizophrenia: selective deficits of subtypes of cortical GABAergic neurons.
J Neural Transm
 
109
:
881
–889.
Ribak CE (
1985
) Axon terminals of GABAergic chandelier cells are lost at epileptic foci.
Brain Res
 
326
:
251
–260.
Richter RM, Weiss F (
1999
) In vivo CRF release in rat amygdala is increased during cocaine withdrawal in self-administering rats.
Synapse
 
32
:
254
–261.
Richter RM, Pich EM, Koob GF, Weiss F (
1995
) Sensitization of cocaine-stimulated increase in extracellular levels of corticotropin-releasing factor from the rat amygdala after repeated administration as determined by intracranial microdialysis.
Neurosci Lett
 
187
:
169
–172.
Robinson TE, Becker JB (
1986
) Enduring changes in brain and behavior produced by chronic amphetamine administration: a review and evaluation of animal models of amphetamine psychosis.
Brain Res Rev
 
11
:
157
–198.
Robinson TE, Kolb B (
1997
) Persistent structural modifications in nucleus accumbens and prefrontal cortex neurons produced by previous experience with amphetamine.
J Neurosci
 
17
:
8491
–8497.
Rogers RD, Everitt BJ, Baldacchino A, Blackshaw AJ, Swainson R, Wynne K, Baker NB, Hunter J, Carthy T, Booker E, London M, Deakin JF, Sahakian BJ, Robbins TW (
1999
) Dissociable deficits in the decision-making cognition of chronic amphetamine abusers, opiate abusers, patients with focal damage to prefrontal cortex, and tryptophan-depleted normal volunteers: evidence for monoaminergic mechanisms.
Neuropsychopharmacology
 
20
:
322
–339.
Sakanaka M, Shibasaki T, Lederis K (
1987
) Corticotropin releasing factor-like immunoreactivity in the rat brain as revealed by a modified cobalt–glucose oxidase–diaminobenzidine method.
J Comp Neurol
 
260
:
256
–298.
Seamans JK, Gorelova N, Durstewitz D, Yang CR (
2001
) Bidirectional dopamine modulation of GABAergic inhibition in prefrontal cortical pyramidal neurons.
J Neurosci
 
21
:
3628
–3638.
Sesack SR, Snyder CL, Lewis DA (
1995
) Axon terminals immunolabeled for dopamine or tyrosine hydroxylase synapse on GABA-immunoreactive dendrites in rat and monkey cortex.
J Comp Neurol
 
363
:
264
–280.
Sharp FR, Sagar SM, Swanson RA (
1993
) Metabolic mapping with cellular resolution: c-fos versus 2-deoxyglucose.
Crit Rev Neurobiol
 
7
:
205
–228.
Siggins GR, Gruol D, Aldenhoff J, Pittman Q (
1985
) Electrophysiological actions of corticotropin-releasing factor in the central nervous system.
Fed Proc
 
44
:
237
–242.
Silva AV, Sanabria ER, Cavalheiro EA, Spreafico R (
2002
) Alterations of the neocortical GABAergic system in the pilocarpine model of temporal lobe epilepsy: neuronal damage and immunocytochemical changes in chronic epileptic rats.
Brain Res Bull
 
58
:
417
–421.
Stanwood GD, Washington RA, Shumsky JS, Levitt P (
2001
) Prenatal cocaine exposure produces consistent developmental alterations in dopamine-rich regions of the cerebral cortex.
Neuroscience
 
106
:
5
–14.
Swanson LW, Sawchenko PE, Rivier J, Vale WW (
1983
) Organization of ovine corticotropin-releasing factor immunoreactive cells and fibers in the rat brain: an immunohistochemical study.
Neuroendocrinology
 
36
:
165
–186.
Traub RD, Kopell N, Bibbig A, Buhl EH, LeBeau FE, Whittington MA (
2001
) Gap junctions between interneuron dendrites can enhance synchrony of gamma oscillations in distributed networks.
J Neurosci
 
21
:
9478
–9486.
Valentino RJ, Foote SL (
1988
) Corticotropin-releasing hormone increases tonic but not sensory-evoked activity of noradrenergic locus coeruleus neurons in unanesthetized rats.
J Neurosci
 
8
:
1016
–1025.
Vincent SL, Khan Y, Benes FM (
1995
) Cellular colocalization of dopamine D1 and D2 receptors in rat medial prefrontal cortex.
Synapse
 
19
:
112
–120.
Wang XH, Levitt P, Grayson DR, Murphy EH (
1995
) Intrauterine cocaine exposure of rabbits: persistent elevation of GABA-immunoreactive neurons in anterior cingulate cortex but not visual cortex.
Brain Res
 
689
:
32
–46.
Woo TU, Miller JL, Lewis DA (
1997
) Schizophrenia and the parvalbumin-containing class of cortical local circuit neurons.
Am J Psychiatry
 
154
:
1013
–1015.
Yan XX, Baram TZ, Gerth A, Schultz L, Ribak CE (
1998
) Co-localization of corticotropin-releasing hormone with glutamate decarboxylase and calcium-binding proteins in infant rat neocortical interneurons.
Exp Brain Res
 
123
:
334
–340.
Zhang K, Tarazi FI, Campbell A, Baldessarini RJ (
2000
) GABA(B) receptors: altered coupling to G-proteins in rats sensitized to amphetamine.
Neuroscience
 
101
:
5
–10.
Zhou FM, Hablitz JJ (
1999
) Dopamine modulation of membrane and synaptic properties of interneurons in rat cerebral cortex.
J Neurophysiol
 
81
:
967
–976.
Zorrilla EP, Valdez GR, Weiss F (
2001
) Changes in levels of regional CRF-like-immunoreactivity and plasma corticosterone during protracted drug withdrawal in dependent rats.
Psychopharmacology (Berl)
 
158
:
374
–381.