Abstract

A compromised γ-aminobutyric acid (GABA)ergic system is hypothesized to be part of the underlying pathophysiology of schizophrenia. N-methyl-d-aspartate (NMDA) receptor hypofunction during neurodevelopment is proposed to disrupt maturation of interneurons causing an impaired GABAergic transmission in adulthood. The present study examines prefrontal GABAergic transmission in adult rats administered with the NMDA receptor channel blocker, phencyclidine (PCP), for 3 days during the second postnatal week. Whole-cell patch-clamp recordings from pyramidal cells in PCP-treated rats showed a 22% reduction in the frequency of miniature inhibitory postsynaptic currents in layer II/III, but not in layer V pyramidal neurons of the prefrontal cortex. Furthermore, early postnatal PCP treatment caused insensitivity toward effects of the GABA transporter 1 (GAT-1) inhibitor, 1,2,5,6-tetrahydro-1-[2-[[(diphenyl–methylene)amino]oxy]ethyl]-3-pyridinecarboxylic acid, and also diminished currents passed by δ-subunit-containing GABAA receptors in layer II/III pyramidal neurons. The observed impairments in GABAergic function are compatible with the alteration of GABAergic markers as well as cognitive dysfunction observed in early postnatal PCP-treated rats and support the hypothesis that PCP administration during neurodevelopment affects the functionality of interneurons in later life.

Introduction

The pathophysiology of schizophrenia, particularly in the prefrontal cortex (PFC), involves an aberrant functionality of γ-aminobutyric acid (GABA)-mediated neurotransmission. Analysis of postmortem brains of patients with schizophrenia shows a region-specific reduction in glutamic acid decarboxylase 67 (GAD67), up-regulation of GABAA receptor subunits, and reduced expression of the GABA transporter 1 (GAT-1) in the PFC (Akbarian et al. 1995; Dean et al. 1999; Benes and Berretta 2001; Schleimer et al. 2004). These alterations have been proposed to involve subpopulations of interneurons that express parvalbumin (PV) and cholecystokinin (Curley and Lewis 2012).

The underlying mechanisms mediating these alterations in the GABAergic system are presently unknown, but pharmacological intervention suggests hypofunction of N-methyl-d-aspartate (NMDA) receptor transmission to be involved. For instance, acute administration of the NMDA receptor channel blockers, phencyclidine (PCP) and ketamine, induces schizophrenia-like symptoms in healthy volunteers, including transient psychosis and cognitive deficits, and exacerbate specific symptoms in schizophrenic patients (Krystal et al. 1994; Adler et al. 1999; Lahti et al. 2001). In addition, exposure to PCP and other types of developmental insults during the late second trimester increases the risk of the fetus developing schizophrenia later in life (Deutsch et al. 1998; Green et al. 2004; Brown and Derkits 2010). In terms of neurodevelopment, this corresponds to the first 2 postnatal weeks in rodents (Bayer et al. 1993), where NMDA receptors are highly expressed on prefrontal GABAergic interneurons (Li et al. 2002; Wang and Gao 2009).

These results provide the basis for a rodent model of schizophrenia-like cognitive deficits, in which treatment with PCP in the second postnatal week induces long-lasting behavioral changes (Wang et al. 2001; Broberg et al. 2008). In this model, adult animals display a reduction in the cortical levels of the GABAergic marker, PV (Nakatani-Pawlak et al. 2009; Powell et al. 2012; Kaalund et al. 2013), and increases in the level of GABAA receptors (du Bois et al. 2009). However, very little is known about the consequences, this has for GABAergic synaptic transmission. Early postnatal PCP treatment is a novel potential animal model for certain anatomical and behavioral traits observed in schizophrenia, so further characterization of the functional inhibitory system could provide important insights into the pathophysiology of schizophrenia.

The goal of this study was to provide a link between the observed reductions in GABAergic markers with the behavioral phenotype of this model. We hypothesized that a functional deficit is present in the prefrontal GABAergic input to pyramidal neurons in rats treated with PCP at early postnatal age. To examine this hypothesis, we have performed standard electrophysiological experiments to assess possible changes in pre- and postsynaptic GABAergic transmission.

Materials and Methods

Animals

All animal experiments were carried out in accordance with the European Parliament and the Council of the European Union directive of 22 September 2010 (2010/63/EU) and approved by the Danish State Research Inspectorate (J. No. 2009/561-1596).

Animals was obtained from Charles River (Germany) and housed pairwise (pregnant dams were housed singly) in cages with standard sawdust bedding and environmental enrichment (plastic house and wooden chew blocks) under standard laboratory conditions, with a 12-h light/12-h dark cycle and ad libitum access to food and water.

Early postnatal PCP treatment was performed as previously described (Broberg et al. 2008). Briefly, timed pregnant Lister hooded rats were obtained at gestational day 15, and the day of parturition was counted as postnatal day (PND) 0. On PND 5, pups were cross-fostered and randomly assigned to a lactating dam in litters of 7–10. On PND 7, 9, and 11, pups were treated subcutaneously with either vehicle (0.9% isotonic saline) or PCP (20 mg kg−1) in a 10 mL kg−1 dose volume. The animals were weaned at PND 24 and sacrificed for electrophysiological recordings (below) after reaching adulthood on PNDs 56–80. Besides acting as a NMDA receptor channel blocker, PCP also acts on other neurotransmitter systems. However, since the main effect of PCP at clinical doses is on NMDA receptors (Morris et al. 2005), we have chosen not to speculate about the additional effects of PCP in this study.

Slice Preparation and Electrophysiology

Early postnatal vehicle- (n = 17) and PCP- (n = 19) treated rats were decapitated and the head immediately immersed in ice-cold cutting solution for 6 min at −20°C. The cutting solution was a carbogenated (95% O2 /5% CO2) artificial cerebrospinal fluid (ACSF) containing 3 mM kynurenic acid (Sigma, Munich, Germany). ACSF was composed of (in mM): NaCl (126), KCl (2.5), CaCl2 (2), MgCl2 (2), NaHCO3 (26), NaH2PO4 (1.25), d-glucose (10), ascorbic acid (0.3), and pyruvic acid (1), adjusted to an osmolality of 310 ± 5 mOsm kg−1 (all chemicals from Sigma). The brain was rapidly dissected out and glued to a platform of the Leica VT 1200S microtome (Leica, Wetzlar, Germany). Coronal slices (350 µm thick) of the regions 4.5–2.2-mm rostral of bregma (Paxinos and Watson 1998) were cut on the microtome, hemisected and transferred to a chamber containing carbogenated ACSF at 30°C, and allowed to rest for at least 60 min before recording.

Slices in the recording chamber were perfused with carbogenated ACSF (including drugs as described below) at 34–35°C at a rate of 2.5 mL min−1. Recording electrodes were prepared on a horizontal puller (Flaming/Brown Micropipette Puller P-97, Sutter Instruments, Novato, CA, USA) using borosilicate glass capillaries (outer diameter, 1.5 mm; inner diameter, 1.1 mm, Sutter Instruments). The intracellular solution consisted of (in mM): CsCl (135), MgCl2 (2), ethylene glycol-bis(2-aminoethylether)-N,N,N′,N′-tetraacetic acid (0.05), 4-(2-hydroxyethyl)piperazine-1-ethanesulfonic acid (10), adenosine 5′-triphosphate (2), guanosine 5′-triphosphate (0.5), tetraethylammonium (5), N-(2,6-dimethylphenylcarbamoylmethyl)triethylammonium (5), creatine phosphate disodium (10) and 0.5% (w/v) biocytin. pH was adjusted to 7.35 at 4°C with CsOH and the osmolality to 293 ± 2 mOsm kg−1 (all chemicals from Sigma). Electrode resistance using this solution was 4–6 MΩ. Neurons were visualized using a differential interference contrast video microscope (Leica). Pyramidal neurons in layer II/III or V of the PFC [defined as infralimbic, prelimbic, and frontal cingulate cortex (Paxinos and Watson 1998)] were patched and held at a potential of −70 mV (Vh) using a Multiclamp 700A patch-clamp amplifier (Molecular Devices, Sunnyvale, CA, USA). Recordings were digitized at 20 kHz using a digidata 1320 (Molecular Devices) and low-pass filtered (4-pole Bessel) at 2 kHz. For recordings involving electrically evoked inhibitory postsynaptic currents (eIPSCs), stimulations (single or trains) were delivered as a 50-µs pulse using an A365 constant current isolation unit (World Precision Instruments, Hitchin, UK) driven by the digitizer.

Recordings began when baseline holding current was stable over 30 s, which was usually achieved 5–8 min after establishment of the whole-cell configuration. Whole-cell capacitance and series resistance (RS) were noted every second minute and RS compensated by 70%. The whole-cell mode was kept for at least 20 min to ensure sufficient time for biocytin diffusion into distal dendrites and axons. Recordings were discarded if RS or cell capacitance deviated >20% from initial values.

Miniature IPSCs (mIPSCs) was pharmacologically isolated by recording in ACSF containing 3 mM kynurenic acid (Sigma) and 1 µM tetrodotoxin (TTX) (Tocris, Bristol, UK).

In a different set of recordings under similar conditions as described for mIPSCs (3 mM kynurenic acid and 1 µM TTX), the level of δ-containing extrasynaptic GABAA receptors was assessed by 5 min perfusion into the recording chamber of 2 μM 4,5,6,7-tetrahydroisoxazolo[5,4-c]pyridin-3-ol (THIP) hydrochloride (synthesized at the Department of Medicinal Chemistry, H. Lundbeck A/S, Valby, Denmark). The current induced by activation of δ-containing GABAA receptors was measured as the change in holding current, following addition of 20 μL volume of SR95531 stock (25 mM, Sigma) directly to the chamber resulting in an immediate >100 µM bath concentration.

In a final set of recordings, evoked GABAA receptor-mediated currents were recorded in ACSF containing 3 mM kynurenic acid. GABAA synaptic events were electrically evoked in layer II/III neurons employing 2 different stimulation paradigms: In setting 1, paired-pulse events were evoked by placing a bipolar stimulation electrode in layer I in the vicinity of the recorded neuron and by adjusting the stimulation intensity to yield an eIPSC 2–3 times the size of an average spontaneous IPSC (sIPSC). An average of 5–6 paired eIPSCs with an interstimulus interval (ISI) of 50, 500, and 1000 ms was used to analyze the paired-pulse ratio (PPR); In setting 2, events were evoked at high frequency as a train of 40 pulses delivered at 50 Hz. For these high-frequency stimulations (HFSs), the stimulation electrode (bipolar borosilicate glass with an opening of 5 µm) was placed 50 μm from the patched neuron, and the stimulation intensity adjusted to yield a stable eIPSC 2–3 times the size of an average sIPSC event. Subsequently, the electrode was moved to a location <5 µm from the patched soma. HFS was repeated 2–3 times with 2 min between each train. The GABAB receptor antagonist, CGP54626 (1 µM, Tocris), was included in the ACSF for all recordings involving HFS. The effect of inhibiting GAT-1 in the HFS stimulation experiment was assessed by inclusion of 10 μM 1,2,5,6-tetrahydro-1-[2-[[(diphenylmethylene)amino]oxy]ethyl]-3-pyridinecarboxylic acid (NNC-711) (Tocris) to the ACSF. In addition, the effect of decreasing the extracellular ACSF Ca2+/Mg2+ ratio from 2/2 to 1/3 (mM) on HFS parameters was evaluated.

Analysis of GABAergic Currents

Mini Analysis (Synaptosoft, version 6.0.7) was used to detect and analyze synaptic events. Events were detected using a threshold value of 4 × root-mean-square noise (=standard deviation) of a 60-ms event-free period for each neuron. All detected events during a continuous period of 5 min were used to assess mIPSC frequency, and sIPSC frequency was analyzed in the window between 3 successive HFSs for a total period of 2 × 2 min. The event amplitude, 10–90% rise time, and monoexponential fit for decay-time constants were assessed for the averaged noncontaminated event.

eIPSCs and THIP-induced extrasynaptic current were analyzed in ClampFit (Molecular Devices/Axon, version 9). The THIP-induced current was measured as the difference in holding current for two 5-s windows beginning 10 s before and 10 s after SR95531 addition to the bath. The PPR of 2 successive eIPSCs was calculated as the ratio of amplitudes, eIPSC2/eIPSC1.

With the stimulation electrode close to the recorded neurons soma (within 5 µm), we employed HFS to obtain an estimate for the readily releasable pool (RRP) of GABA. An RRP was estimated from plots of accumulated peak eIPSC amplitudes against stimulus number during the HFS train (baseline was set as the holding current at the starting level of each individual eIPSC). We used the interception with the y-axis of a fitted straight line through the accumulated eIPSC peaks for stimulation number 20–40 as an estimate of the RRP, which in this setting is defined as the cumulative eIPSC amplitudes in the absence of pool replenishment (Schneggenburger et al. 1999; Kirischuk and Grantyn 2003; Abidin et al. 2008). The slope of the fitted straight line allows an estimate of the level of pool replenishment. The total area of the HFS-induced current was calculated, and the decay phase after the last eIPSC in the train back to baseline was characterized for all cells by weight of tau (area/peak).

The average conductance of GABAA receptors contributing to the synaptic current was assessed by applying nonstationary fluctuation analysis (NSFA) on selected sIPSCs (range: 120–190 events per cell recording), as described previously (Nusser et al. 2001). Exclusively for NSFA, events with characteristics indicating a minimized electrotonic distortion were selected, that is, a rise time (10–90%) <1 ms and an amplitude ≥mean sIPSC amplitude. The calculations were run using routines written in OriginPro 8.1 (OriginLab Corporation, Northampton, MA, USA). Briefly, the mean of selected events was scaled to each individual event and subtracted. The variance around the mean in the period 1–30 ms after the mean peak for all peak-scaled individual events was divided into 40 bins and plotted against the mean single-event current and fitted with the relation, $σ2=i⋅I−I2/N+σBL2$, in which σ2 is the bin variance of traces around the mean, $σBL2$ the baseline variance, i the single-channel current, and I the mean current at the peak passed by N channels. The conductance is given by γ = i/(Vm− Vrev), in which Vm is the membrane (holding) potential and Vrev the reversal potential of GABA receptors in this preparation (∼0 mV).

It should be emphasized that the calculations of average conductance of GABAA receptors are based on the recordings of sIPSCs that involve inhibitory synapses that innervate both somatic and dendritic regions of pyramidal neurons as well as activate receptors with different subunit compositions.

Histological Examination of Recorded Neurons

Prior to patching, pyramidal cells from layers II/III and V in the PFC were identified based on their triangular somatic structure and the presence of a surface-projecting apical dendrite. Following recording, brain slices were fixed in phosphate-buffered saline with 4% paraformaldehyde, and biocytin-labeled neurons were labeled with fluorescence using a streptavidin fluor Alexa 488 conjugate (Molecular Probes, Grand Island, NY, USA). The identity of pyramidal neurons was confirmed by the presence of spines on their apical and basal dendrites. Cells exhibiting a large local axonal plexus and the presence of multiple boutons were assumed to be interneurons and excluded from analysis.

Statistics

The D'Agostino and Pearson normality test was used to assess normal distribution of data. A 2-tailed, 2-sample equal variance t-test was used for comparing vehicle- and PCP-treated groups. For the comparison of means before and following drug infusion, a paired Student's t-test was employed. One sample t-tests were used to test whether relative changes following drug treatment were significantly different from 1. The significance level was set to P < 0.05, and data were expressed as means ± SEM, unless where noted, with n indicating the number of cells.

Results

PCP-Treated Rats Have Reduced mIPSC Frequencies in Layer II/III of PFC

To evaluate the effect of early postnatal PCP treatment on inhibitory synaptic transmission, mIPSCs were recorded from pyramidal neurons in layers II/III and V of the PFC (for more detail see Fig. 1A). Layer IV is very thin and not included in Figure 1A. Application of the GABAA antagonist, SR95531 (10 µM), abolished all mIPSCs (data not shown). The mean RS for all recorded cells was 12.7 ± 0.6 (range: 6–30 MΩ) and the capacitance 18.0 ± 0.6 (range: 8–30 pF). There was no difference in the mean capacitance and RS of recorded neurons between vehicle and PCP groups. All detected events within a 5-min period were included in the analysis. The frequency of mIPSCs in layer II/III of vehicle-treated rats was 10.2 ± 0.7 Hz (n = 19) compared with 8.0 ± 0.7 Hz (n = 21) in PCP-treated rats (P = 0.027), indicating a treatment-induced reduction in mIPSC frequency of ∼22% (Fig. 1C and Table 1; power of test = 0.91; Cohen's d = 0.74). The amplitude, 10–90% rise time, and decay-time constant of the average noncontaminated events were similar between PCP- and saline-treated animals in layer II/III pyramidal neurons (Table 1).

Table 1

mIPSC frequency and kinetics

Layer II/III

Layer V

Vehicle (N = 19) PCP (N = 21) Vehicle (N = 10) PCP (N = 9)
Frequency (Hz) 10.2 ± 0.7 8.0 ± 0.7* 9.8 ± 1.2 9.3 ± 1.2
Amplitude (pA) 25.5 ± 1.6 25.5 ± 1.8 21.9 ± 1.2 24.4 ± 1.7
Rise time (10–90%) (ms) 0.74 ± 0.05 0.80 ± 0.05 1.02 ± 0.09 0.94 ± 0.15
Decay time constant (ms) 9.3 ± 0.7 9.6 ± 0.4 10.9 ± 0.4 10.4 ± 0.4
Layer II/III

Layer V

Vehicle (N = 19) PCP (N = 21) Vehicle (N = 10) PCP (N = 9)
Frequency (Hz) 10.2 ± 0.7 8.0 ± 0.7* 9.8 ± 1.2 9.3 ± 1.2
Amplitude (pA) 25.5 ± 1.6 25.5 ± 1.8 21.9 ± 1.2 24.4 ± 1.7
Rise time (10–90%) (ms) 0.74 ± 0.05 0.80 ± 0.05 1.02 ± 0.09 0.94 ± 0.15
Decay time constant (ms) 9.3 ± 0.7 9.6 ± 0.4 10.9 ± 0.4 10.4 ± 0.4

Note: mIPSC frequency, amplitude, 10–90% rise time, and decay time constant of the average noncontaminated event in layers II/III and V of vehicle- and PCP-treated animals are shown as mean ± standard error of the mean (SEM). mIPSC, miniature inhibitory postsynaptic current.

*A significant difference (P < 0.05) in mIPSC frequency between vehicle and PCP animals in layer II/III.

Figure 1.

The frequency of miniature inhibitory postsynaptic currents (mIPSCs) in layer II/III is decreased in early postnatal PCP-treated animals. (A) mIPSCs was recorded in medial PFC defined as frontal cingulate (Cg1), prelimbic (PrL), and infralimbic (IL) cortices (Paxinos and Watson 1998). Recordings from cells in the medial orbital cortex (MO) were discarded. Recorded cells were loaded with biocytin in order to confirm identity as pyramidal neurons (upper left) and were subsequently divided according to location in middle (approximately layer II/III) or deeper layers (approximately layer V). (B) Representative mIPSC traces of layer II/III pyramidal neurons of vehicle- and PCP-treated animals. (C) The cumulative probability plot indicates a right shift in the curve of the interevent interval of mIPSCs in layer II/III of PCP-treated animals. The summary plot of the mIPSC frequencies shows a lower frequency of mIPSCs in PCP animals (P < 0.05). To the right is shown the average noncontaminated event for vehicle and PCP animals, which reveal no difference in the kinetic characteristics (Table 1). (D) In layer V, no differences in mIPSC frequency and kinetics (Table 1) of the average noncontaminated event are evident between PCP and vehicle animals. *P < 0.05. Layer II/III: vehicle, n = 19; PCP, n = 21. Layer V: vehicle, n = 10; PCP, n = 9.

Figure 1.

The frequency of miniature inhibitory postsynaptic currents (mIPSCs) in layer II/III is decreased in early postnatal PCP-treated animals. (A) mIPSCs was recorded in medial PFC defined as frontal cingulate (Cg1), prelimbic (PrL), and infralimbic (IL) cortices (Paxinos and Watson 1998). Recordings from cells in the medial orbital cortex (MO) were discarded. Recorded cells were loaded with biocytin in order to confirm identity as pyramidal neurons (upper left) and were subsequently divided according to location in middle (approximately layer II/III) or deeper layers (approximately layer V). (B) Representative mIPSC traces of layer II/III pyramidal neurons of vehicle- and PCP-treated animals. (C) The cumulative probability plot indicates a right shift in the curve of the interevent interval of mIPSCs in layer II/III of PCP-treated animals. The summary plot of the mIPSC frequencies shows a lower frequency of mIPSCs in PCP animals (P < 0.05). To the right is shown the average noncontaminated event for vehicle and PCP animals, which reveal no difference in the kinetic characteristics (Table 1). (D) In layer V, no differences in mIPSC frequency and kinetics (Table 1) of the average noncontaminated event are evident between PCP and vehicle animals. *P < 0.05. Layer II/III: vehicle, n = 19; PCP, n = 21. Layer V: vehicle, n = 10; PCP, n = 9.

We saw no difference in the frequency of mIPSCs in layer V pyramidal neurons between PCP- (9.3 ± 1.2 Hz, n = 9) and vehicle- (9.8 ± 1.2, n = 10) treated animals, and the average noncontaminated events were similar in amplitude, 10–90% rise time and decay time (Fig. 1D and Table 1).

Diminished Evoked GABAergic Output in PCP Animals

The reduced frequency of mIPSCs observed following early postnatal PCP treatment suggests a change in properties mediating presynaptic release of GABA. To determine whether one or both of these parameters were affected, estimates based on measurements of the RRP during HFS were calculated. An RRP is the pool of neurotransmitter immediately released from the presynaptic terminal following activation and was measured by exhausting the docked vesicles by repetitive stimulation.

IPSCs were recorded during HFS (40 pulses at 50 Hz). Normalization of stimulus intensity was done by, at first, positioning the stimulation electrodes 50 µm from the soma to stably induce eIPSCs (approximately 2–3 times the average sIPSC amplitude without any failures) in the patched pyramidal neuron in layer II/III (Fig. 2A). Stimulation intensity for vehicle and PCP were 28.3 ± 3.5 and 33.0 ± 3.6 mA (n.s.), respectively. The stimulation electrode was then moved very close to the soma (∼5 µm) for HFS (Fig. 2A). The magnitude of the stimulation artifact normalized to stimulation intensity was similar for vehicle and PCP animals (vehicle, 19.6 ± 3.8; PCP, 19.2 ± 2.0, n.s.), indicating the same distance between the stimulation electrode and patched cell within both treatment groups. This normalization approach was undertaken, since adjusting stimulation intensity close to the cell had to be abandoned due to very variable eIPSC responses (data not shown). Placing the stimulation electrode, ∼50 µm from the soma could induce reliable small eIPSC of similar size between cells and with similar stimulation intensity between treatment groups. By moving the electrode close to the soma (<5 µm), very large events (typically >1 nA) were generated.

Figure 2.

Experimental set-up during HFS. (A) Top: stimulation intensity was normalized by placing the stimulation electrode 50 μm from the patched pyramidal neuron (PY) and by adjusting the intensity of stimulation to yield an eIPSC of 2–3 times the size of the average sIPSC amplitude. Bottom: the electrode was then moved to a location <5 μm from the patched soma and recording was initiated. (B) Stimulation consisted of a train of 40 pulses, each of 50 µs duration, in 0.8 s (50 Hz). eIPSCs were recorded and followed by measurement of sIPSCs 5 s after the last eIPSC of each HFS train.

Figure 2.

Experimental set-up during HFS. (A) Top: stimulation intensity was normalized by placing the stimulation electrode 50 μm from the patched pyramidal neuron (PY) and by adjusting the intensity of stimulation to yield an eIPSC of 2–3 times the size of the average sIPSC amplitude. Bottom: the electrode was then moved to a location <5 μm from the patched soma and recording was initiated. (B) Stimulation consisted of a train of 40 pulses, each of 50 µs duration, in 0.8 s (50 Hz). eIPSCs were recorded and followed by measurement of sIPSCs 5 s after the last eIPSC of each HFS train.

HFS was conducted in slices from vehicle- (n = 24) and PCP-treated (n = 37) rats. HFS resulted in a gradual reduction in the amplitude of eIPSC until a steady-state level of eIPSC amplitudes was reached, as illustrated by the linear part of the cumulative amplitude plot (Fig. 3A). From these plots, the RRP was calculated by back-extrapolating a linear fit to the cumulative amplitude plot of eIPSC stimulation 20–40 to its interception with the y-axis (Fig. 3A). The RRP was significantly reduced in PCP animals (P = 0.0246, Fig. 3A and Table 2). Apparent was also a reduction in eIPSC1 of the stimulation train in PCP-treated rats (P = 0.0075, Fig. 3B and Table 2).

Table 2

Measurements and estimates from HFS protocol

Vehicle

N PCP

N
HFS measurements
RRP (pA) 7210 ± 963  24 4597 ± 674*  37
eIPSC1 amplitude (pA) 1234 ± 158  24 763 ± 91**  37
eIPSC2/eIPSC1 1.06 ± 0.13  24 0.97 ± 0.12  37
Control NNC-711  Control NNC-711
Slope (pulse 20–40) (pA/pulse) 636 ± 148 395 ± 92# 297 ± 68* 227 ± 53# 12
Total area (pC) 1488 ± 283 3101 ± 624## 386 ± 88*** 1201 ± 209### 12
Weight of tau (area/peak) (ms) 146 ± 21 314 ± 47## 89 ± 13* 184 ± 14## 12
eIPSC1 amplitude (pA) 1715 ± 262 1235 ± 183# 650 ± 70*** 608 ± 148 12
Vehicle

N PCP

N
HFS measurements
RRP (pA) 7210 ± 963  24 4597 ± 674*  37
eIPSC1 amplitude (pA) 1234 ± 158  24 763 ± 91**  37
eIPSC2/eIPSC1 1.06 ± 0.13  24 0.97 ± 0.12  37
Control NNC-711  Control NNC-711
Slope (pulse 20–40) (pA/pulse) 636 ± 148 395 ± 92# 297 ± 68* 227 ± 53# 12
Total area (pC) 1488 ± 283 3101 ± 624## 386 ± 88*** 1201 ± 209### 12
Weight of tau (area/peak) (ms) 146 ± 21 314 ± 47## 89 ± 13* 184 ± 14## 12
eIPSC1 amplitude (pA) 1715 ± 262 1235 ± 183# 650 ± 70*** 608 ± 148 12

Note: Mean ± standard error of the mean (SEM) and number of recordings (N) are shown for measurements and estimates made from analysis of the HFS experiments conducted on pyramidal neurons in layer II/III of vehicle- and PCP-treated rats. Values before and after administration of the GABA transporter 1 (GAT-1) inhibitor, NNC-711 (10 µM), are also displayed. RRP, readily releasable pool; eIPSC, evoked IPSC; NNC-711, 1,2,5,6-tetrahydro-1-[2-[[(diphenylmethylene)amino]oxy]ethyl]-3-pyridinecarboxylic acid.

*, **, and *** denote significant differences between vehicle and PCP animals (P < 0.05, <0.01, <0.001, respectively).

#, ##, and ### denote significant differences between control and NNC-711 conditions (P < 0.05, <0.01, <0.001, respectively).

Figure 3.

High-frequency stimulations (HFS) disclose a reduction in the RRP, and the first eIPSC1. (A) Plot of the cumulative amplitude of eIPSCs versus the number of stimuli. A straight line was fitted to the linear part of the graph of the cumulated amplitudes (stimulation 20–40 in the train) and back-extrapolated to the y-axis to obtain the estimate for the RRP. The RRP was reduced in PCP animals (P < 0.05). (B) Representative current trace of the first eIPSC (eIPSC1) in vehicle and PCP rats. eIPSC1 was reduced in PCP animals (P < 0.01). (D) The ratio between eIPSC2 and eIPSC1 of the HFS train (interstimulus interval of 20 ms) was not significant different from 1 in either vehicle- or PCP-treated animals. *P < 0.05; **P < 0.01; vehicle, n = 24; PCP, n = 37.

Figure 3.

High-frequency stimulations (HFS) disclose a reduction in the RRP, and the first eIPSC1. (A) Plot of the cumulative amplitude of eIPSCs versus the number of stimuli. A straight line was fitted to the linear part of the graph of the cumulated amplitudes (stimulation 20–40 in the train) and back-extrapolated to the y-axis to obtain the estimate for the RRP. The RRP was reduced in PCP animals (P < 0.05). (B) Representative current trace of the first eIPSC (eIPSC1) in vehicle and PCP rats. eIPSC1 was reduced in PCP animals (P < 0.01). (D) The ratio between eIPSC2 and eIPSC1 of the HFS train (interstimulus interval of 20 ms) was not significant different from 1 in either vehicle- or PCP-treated animals. *P < 0.05; **P < 0.01; vehicle, n = 24; PCP, n = 37.

No Difference in GABAA Receptor Conductance Between PCP and Control Animals

We further assessed synaptic GABAA receptor conductance by NSFA on populations of sIPSCs obtained from cells used for HFS recordings (11 cells were chosen from each group of vehicle- and PCP-treated rats; Fig. 4). Events for NSFA were chosen as described in Materials and methods section. The average of the calculated single-channel current, i, for vehicle- and PCP-treated animals were −2.54 ± 0.16 (n = 11) and −2.71 ± 0.28 pA (n = 11), respectively, which corresponds to an average single-channel conductance of 36.3 ± 2.0 and 38.7 ± 4.0 pS (n.s.), in vehicle- and PCP-treated rats, respectively (driving force, VhVrev, was −70 mV; Fig. 4). From these calculations, was also obtained the average number of channels, N, open at the peak of the constructed event (15.6 ± 0.7 and 14.0 ± 1.5 in vehicle- and PCP-treated rats, respectively, n.s.; Fig. 4).

Figure 4.

Nonstationary fluctuation analysis (NSFA). It shows NSFA from a representative cell performed on sIPSCs recordings obtained from the HFS experiments. The displayed summary graphs show no difference between vehicle and PCP animals in the amount of single-channel current (i), the number of open channels (N), or in conductance (y). Vehicle, n = 11; PCP, n = 11.

Figure 4.

Nonstationary fluctuation analysis (NSFA). It shows NSFA from a representative cell performed on sIPSCs recordings obtained from the HFS experiments. The displayed summary graphs show no difference between vehicle and PCP animals in the amount of single-channel current (i), the number of open channels (N), or in conductance (y). Vehicle, n = 11; PCP, n = 11.

At Low-Release Probability, the Readily Releasable Pool Is Equally Reduced in Vehicle and PCP Animals

Further characterization of presynaptic GABAergic release properties was done by altering the Ca2+:Mg2+ ratio in the ACSF from 2:2 to 1:3, and observing the effect on eIPSC amplitudes during HFS (vehicle, n = 5 and PCP: n = 8; Fig. 5). RRP relative to control was reduced 78% in vehicle (P = 0.0005) and 63% in PCP animals (P = 0.0006; Fig. 5), with no difference between vehicle and PCP groups. In comparison, the average amplitude of the eIPSC on the steady-state part of the curve (eIPSC20–40) was only reduced 33% (P = 0.026) and 34% (P = 0.0012) in vehicle and PCP animals, respectively, suggesting that presynaptic GABA release is more sensitive to Ca2+ changes during release of the RRP than in the recycling phase.

Figure 5.

PCP animals display no difference in the response to low calcium in the extracellular medium. Top: A plot showing the effect of reducing the Ca2+/Mg2+ ratio from 2:2 to 1:3 on the accumulated eIPSC amplitude in vehicle animals. Bottom: in response to low Ca2+, a relative reduction was observed in RRP (vehicle, P < 0.001 and PCP, P < 0.001) and amplitude of the average eIPSC 20–40 (vehicle, P < 0.05 and PCP, P < 0.01). *P < 0.05, **P < 0.01, ***P < 0.001. Vehicle: n = 5, PCP: n = 8.

Figure 5.

PCP animals display no difference in the response to low calcium in the extracellular medium. Top: A plot showing the effect of reducing the Ca2+/Mg2+ ratio from 2:2 to 1:3 on the accumulated eIPSC amplitude in vehicle animals. Bottom: in response to low Ca2+, a relative reduction was observed in RRP (vehicle, P < 0.001 and PCP, P < 0.001) and amplitude of the average eIPSC 20–40 (vehicle, P < 0.05 and PCP, P < 0.01). *P < 0.05, **P < 0.01, ***P < 0.001. Vehicle: n = 5, PCP: n = 8.

Presynaptic short-term plasticity was further characterized by the PPR between 2 successive eIPSCs (vehicle, n = 19 and PCP: n = 15) evoked by stimulation in layer I. At ISI of 50 ms, PPR was 1.36 ± 0.10 in vehicles and 1.56 ± 0.16 in PCP animals (n.s., Fig. 6). Increasing the ISI to 500 or 1000 ms gave PPR ratios very close to of 1 in both vehicle and PCP animals (Fig. 6). The PPR between the first 2 pulses in the HFS train (eIPSC2/eIPSC1; ISI: 20 ms; vehicle, n = 24 and PCP: n = 37) rats displayed no significant deviation from 1 in either vehicle or PCP animals (Fig. 3C and Table 2).

Figure 6.

Paired-pulse facilitation was observed in both vehicle and PCP animals. Top: a representative trace displaying eIPSCs in layer II/III pyramidal neurons following paired stimulations with an ISI of 50 ms in vehicle and PCP animals. The eIPSC2/eIPSC1 ratio was increased at a 50-ms ISI (P < 0.01), but displayed no significant deviation from 1 at longer ISIs in either vehicle or PCP. **P < 0.01. Vehicle, n = 19, PCP, n = 15.

Figure 6.

Paired-pulse facilitation was observed in both vehicle and PCP animals. Top: a representative trace displaying eIPSCs in layer II/III pyramidal neurons following paired stimulations with an ISI of 50 ms in vehicle and PCP animals. The eIPSC2/eIPSC1 ratio was increased at a 50-ms ISI (P < 0.01), but displayed no significant deviation from 1 at longer ISIs in either vehicle or PCP. **P < 0.01. Vehicle, n = 19, PCP, n = 15.

Compromised GABA Recycling and Extrasynaptic GABAA Receptor Activation in PCP Animals

It is possible that the diminished accumulated eIPSC response in PCP-treated rats involves a GABA recycling deficit. GABA uptake is a crucial step in GABA recycling, and we therefore examined the effect of inhibiting GABA uptake on HFS by wash-in of 10 μM NNC-711, a selective inhibitor of GAT-1 (vehicle, n = 9 and PCP, n = 12; Fig. 7A). The slope of the fitted lines to the steady-state part of the cumulative amplitude plot (pulse 20–40) before wash-in of NNC-711 was reduced in PCP animals in comparison to the slope in vehicle animals (P = 0.04, Fig. 7B and Table 2). Blocking GAT-1 in vehicle- and PCP-treated animals reduced the slope of the eIPSC linear fit in the cumulative amplitude plot (vehicle, P = 0.036 and PCP, P = 0.038; Fig. 7B and Table 2) by 38% and 24%, respectively. When blocking GAT-1 transporters, the amplitude of eIPSC1 was reduced in vehicle-treated rats (P = 0.0109, Table 2) corresponding to a relative reduction of 22% (Fig. 7C). However, we saw no change in the amplitude of eIPSC1 in PCP-treated rats in response to 10 μM NNC-711 (n.s., P = 0.367; Fig. 7C and Table 2).

Figure 7.

GABA recycling and extrasynaptic GABAA-mediated current assessed during HFS are affected following PCP treatment. (A) Representative traces showing eIPSCs during HFS are displayed for vehicle- and PCP-treated animals, and before and after wash-in of 1,2,5,6-tetrahydro-1-[2-[[(diphenylmethylene)amino]oxy]ethyl]-3-pyridinecarboxylic acid (NNC-711) in vehicle animals. (B) Left: in response to wash-in of the inhibitor of the GABA transporter 1 (GAT-1), NNC-711, the slope of the linear part of the cumulative amplitude plot (pulses 20–40) is reduced in both vehicle (P < 0.05) and PCP (P < 0.05) rats. Middle: The slopes before NNC-711 wash-in are significantly different between vehicle and PCP animals (P < 0.05). Right: The NNC-711 slope values relative to control are significantly different from 1 in vehicle (P < 0.05) and with a strong trend in PCP animals (P = 0.0982). (C) The amplitude of the first eIPSC1 is reduced in response to NNC-711 in vehicle (P < 0.05), but not in PCP, animals. (D) The total area of HFS-induced current and the weight of tau for the decay phase (ratio between decay-phase area and peak after last eIPSC) were calculated as illustrated for vehicle. (E) The total area is reduced in PCP animals (P < 0.001) in comparison to vehicle. Upon NNC-711 wash-in, a pronounced relative increase in total area is evident for both vehicle (P < 0.001) and PCP (P < 0.001) animals. Furthermore, PCP animals display a reduction in weight of tau (P < 0.05). The addition of NNC-711 produces a relative increase in weight of tau in both vehicle (P < 0.05) and PCP (P < 0.01) animals. *P < 0.05; **P < 0.01; ***P < 0.001. Vehicle, n = 9; PCP, n = 12.

Figure 7.

GABA recycling and extrasynaptic GABAA-mediated current assessed during HFS are affected following PCP treatment. (A) Representative traces showing eIPSCs during HFS are displayed for vehicle- and PCP-treated animals, and before and after wash-in of 1,2,5,6-tetrahydro-1-[2-[[(diphenylmethylene)amino]oxy]ethyl]-3-pyridinecarboxylic acid (NNC-711) in vehicle animals. (B) Left: in response to wash-in of the inhibitor of the GABA transporter 1 (GAT-1), NNC-711, the slope of the linear part of the cumulative amplitude plot (pulses 20–40) is reduced in both vehicle (P < 0.05) and PCP (P < 0.05) rats. Middle: The slopes before NNC-711 wash-in are significantly different between vehicle and PCP animals (P < 0.05). Right: The NNC-711 slope values relative to control are significantly different from 1 in vehicle (P < 0.05) and with a strong trend in PCP animals (P = 0.0982). (C) The amplitude of the first eIPSC1 is reduced in response to NNC-711 in vehicle (P < 0.05), but not in PCP, animals. (D) The total area of HFS-induced current and the weight of tau for the decay phase (ratio between decay-phase area and peak after last eIPSC) were calculated as illustrated for vehicle. (E) The total area is reduced in PCP animals (P < 0.001) in comparison to vehicle. Upon NNC-711 wash-in, a pronounced relative increase in total area is evident for both vehicle (P < 0.001) and PCP (P < 0.001) animals. Furthermore, PCP animals display a reduction in weight of tau (P < 0.05). The addition of NNC-711 produces a relative increase in weight of tau in both vehicle (P < 0.05) and PCP (P < 0.01) animals. *P < 0.05; **P < 0.01; ***P < 0.001. Vehicle, n = 9; PCP, n = 12.

Calculating the area of the total evoked current produced by the HFS train (Fig. 7D) provides information about the collected postsynaptic and extrasynaptic currents activated by the GABA spill-over. The total charge of the HFS events including decay was reduced in PCP-treated rats (P = 0.0007; Fig. 7E and Table 2). Following inhibition of GAT-1-mediated GABA uptake by 10 μM NNC-711, these values were significantly increased for both vehicle (P = 0.0019) and PCP (P = 0.0006) animals, about 123% and 265%, respectively (Fig. 7E and Table 2). As an estimate of the extrasynaptic component, we characterized the decay phase after the last eIPSC back to baseline by its weight of tau (area/peak) (Fig. 7D). The weight of tau was decreased in PCP-treated rats (P = 0.0234; Fig. 7E and Table 2), suggesting a reduction in extrasynaptic GABAA receptor activation. Application of NNC-711 increased these values of weight of tau (vehicle, P = 0.0082 and PCP, P = 0.0013; Table 2) to a similar extent in both vehicle and PCP animals of almost 200% (Fig. 7E).

THIP-Mediated GABAA Receptor Current Is Decreased in PCP Animals

The reduction in magnitude of the extrasynaptic current (weight of tau) from the HFS experiment observed in PCP compared with vehicle rats is a convoluted effect of differences in GABA release, uptake, and activation of extrasynaptic GABAA receptors. To distinguish between possibilities for differences in GABA uptake, the amount of current induced by activation of extrasynaptic δ-subunit-containing GABAA receptors was examined independently from synaptic spill-over by the measurement of extrasynaptic current induced by the δ-subunit agonist, THIP. Extrasynaptic GABAA receptors activated by 2 μM THIP perfused into the recording chamber induced a characteristic baseline shift in the holding current (Fig. 8A), which was stable after 5 min bath perfusion of THIP. The THIP-induced current was then assessed by determination of the jump in holding current following addition of SR95531 to the bath (Fig. 8A,B) and was shown to be significantly smaller in PCP-treated rats (PCP, 30.0 ± 4.8 pA, n = 19 and vehicle, 47.3 ± 7.0 pA, n = 12; P = 0.0451, Fig. 8B).

Figure 8.

Extrasynaptic GABAA-mediated current is reduced in PCP-treated animals. (A) Representative current trace (sampling interval: 50 ms) showing 5-min perfusion with 2 μM of the δ-containing extrasynaptic GABAA receptor agonist, THIP, followed by the addition of the GABAA receptor antagonist, SR95531. The change in holding current was measured for assessment of THIP-induced current. (B) Representative traces showing shift in holding current in response to SR95531 in vehicle- and PCP-treated animals. THIP-induced GABAA-mediated current was significantly reduced in PCP animals (P < 0.05). *P < 0.05. Vehicle: n = 12; PCP: n = 19.

Figure 8.

Extrasynaptic GABAA-mediated current is reduced in PCP-treated animals. (A) Representative current trace (sampling interval: 50 ms) showing 5-min perfusion with 2 μM of the δ-containing extrasynaptic GABAA receptor agonist, THIP, followed by the addition of the GABAA receptor antagonist, SR95531. The change in holding current was measured for assessment of THIP-induced current. (B) Representative traces showing shift in holding current in response to SR95531 in vehicle- and PCP-treated animals. THIP-induced GABAA-mediated current was significantly reduced in PCP animals (P < 0.05). *P < 0.05. Vehicle: n = 12; PCP: n = 19.

Discussion

In this study, we have performed a functional characterization of different components of the GABAergic input to pyramidal neurons in the PFC of early postnatal PCP-treated rats, an animal model of cognitive impairment associated with schizophrenia. One of the primary findings is that the frequency of mIPSC is reduced in superficial layers II/III, but not the deeper layer V in adult PCP-treated rats. This suggests that the inhibitory synaptic input to layer II/III pyramidal neurons is compromised in PCP- versus vehicle-treated animals with no apparent difference in synaptic GABAA receptor kinetics, including channel conductance. HFS depresses and short-term depletes the RRP from the population of stimulated synapses, and our estimates of RRP showed a ∼37% reduction in PCP-treated animals; a finding possibly directly related to the reduction in the frequency of mIPSC. Factors governing presynaptic release were assessed by paired-pulse stimulations and by recordings of RRP with a low Ca2+/Mg2+ ratio, and we saw no differences in these parameters between vehicle and PCP animals. Finally, other parts of the GABAergic system also displayed deficits, including a reduced δ-containing GABAA-mediated extrasynaptic current, and insensitivity toward effects of GAT-1 inhibition on eIPSC amplitude.

mIPSC Frequency Is Reduced in Prefrontal Layer II/III, but not in Layer V Pyramidal Neurons of PCP Animals

The initial observation that frequency of mIPSC was reduced in the superficial layer II/III, but not in the deeper layer V pyramidal neurons of PCP-treated rats, suggests an anatomical or physiological difference in, among others, GABAergic transmission between these layers. It is well known that the midbrain dopaminergic projection to the PFC innervates preferentially deeper layers (Santana et al. 2009), and that the excitatory thalamocortical projection preferentially makes asymmetric synapses on pyramidal neurons in layers III and V (Kuroda et al. 1998). Although not demonstrated specifically for the PFC, the interneuronal GABAergic innervation also differs between superficial and deeper layers (Gupta et al. 2000; Thomson and Bannister 2003), and the sIPSC properties recorded in layer V pyramidal neurons are kinetically more diverse than those recorded in layer II/III pyramidal neurons (Woodhall et al. 2005). These anatomical and physiological data support the notion of separate neuronal origin of IPSCs in cortical layers, and our observation can therefore be explained by a differential layer-specific response to systemic insults. For this reason, we have chosen to focus only on describing the compromised GABAergic synaptic transmission in PFC layer II/III in this animal model.

The Estimated Readily Releasable Pool Is Reduced in PCP Animals

The reduced mIPSC frequency in layer II/III of PCP animals demonstrates an impaired synaptic GABAergic transmission, which is further supported by a reduction in the estimate of RRP made from the cumulative amplitude plot during HFS. The plot of accumulated eIPSC amplitudes is clearly biphasic owing to a shift to a reduced rate of vesicular release from the presynaptic terminal after ∼15 stimulations (Fig. 3). Following this rapid decrease, the eIPSC amplitudes reach a steady-state level maintained by a recycling pool of vesicles (Rizzoli and Betz 2005) represented by the linear component in the cumulative amplitude plot. We estimated the RRP as the value of accumulated eIPSC amplitudes, where a linear fit to eIPSC20–40 intercepts the y-axis (Schneggenburger et al. 1999) (Fig. 3). This method of estimating the RRP assumes that the change in eIPSC amplitude during HFS is caused by depletion of the stimulated pool of GABAergic terminals. However, it should be emphasized that this estimate—despite our efforts to stimulate very close to soma—is an average composed of contributions from different GABAergic terminals. The terminals of these interneuronal populations show different levels of synaptic depression upon HFS (Xiang et al. 2002; Kapfer et al. 2007; Ma et al. 2012). Furthermore, the eIPSC response during HFS is convolved with components serving to adapt both the pre- and the postsynaptic responses, such as receptor desensitization (Overstreet et al. 2000), presynaptic Ca2+ channel inactivation (Patil et al. 1998), possible release of glutamate activating presynaptic mGluRs on GABAergic terminals, and negative feedback through autoinhibitory GABAB receptors (Deisz and Prince 1989). We excluded influence from GABAB receptors by including the GABAB receptor selective antagonist, CGP54626. In addition, presynaptic mGluRs on GABAergic terminals, at least in the hippocampus, appear only on synapses onto other interneurons (Kogo et al. 2004), but we cannot exclude that mGluRs present on GABAergic synapses in PFC could affect release parameters.

The PCP-mediated reduction in RRP might be due to a reduction in the number of GABA release sites in PCP animals or a reduction in the size of the vesicular pool within axon terminals. The first scenario is supported by our observation of a reduced frequency of mIPSCs and might be linked to a disturbed synaptogenesis of interneurons, whereas the latter could be coupled to a diminished (GAD-mediated) GABA synthesis. A reduced GAD activity may also result in a diminished vesicle filling that could reduce quantal size if the synapse is not saturated upon single-vesicle release. We observed unchanged mIPSC kinetics between groups in both layers II/III and V and pyramidal neurons (Table 1). Since GABAergic synapses on layer II/III pyramidal neurons, but not on layer V pyramidal neurons, appear saturated upon vesicular release (Hajos et al. 2000), it cannot be excluded that alterations in GAD could go unnoticed in our analysis of layer II/III GABA-mediated synaptic physiology.

An additional limitation of the estimates provided here is their dependence on the stimulation intensity, distance of stimulation electrode to activated synapses, and the volume transmission through the tissue. It was not possible to generate meaningful input–output relations close to the patched cell, either because the number of presynaptic contacts reached by the stimulation or the probability of release was too variable. We addressed these issues by ensuring that our stimulation protocol produced similar levels of activation at a distance of 50 µm from the soma of the recorded cells. Stimulation intensity was similar between treatment groups, indicating that the normalization procedure was independently of any phenotypic differences. Subsequent movement of the electrode to close proximity of the soma generated very large events most likely due to activation of particularly perisomatic synapses. The magnitude of the stimulation artifact normalized to stimulation intensity was used as an indicator of the distance between recording and stimulation electrode, and no differences between groups were observed. These procedures were done to ensure that stimulation parameters were identical between groups and suggest that the described differences in synaptic properties are truly phenotype-related.

Ca2+-Mediated Short-Term Plasticity Is Not Changed in PCP Animals

Lowering the external ACSF Ca2+/Mg2+ ratio reduces the probability of Ca2+-mediated vesicle release. During HFS, the size of RRP and the average amplitude of eIPSC20–40 were significantly reduced in both vehicle and PCP, following wash-in of low Ca2+ ACSF. We did not see any difference in the relative response to low Ca2+ in PCP- versus vehicle-treated animals, suggesting that presynaptic Ca2+-mediated GABA release is not affected by postnatal PCP treatment. To further characterize Ca2+-dependent short-term plasticity, we measured PPRs in a variety of settings. All PPRs were similar between groups, indicating that release probability responds similar to paired pulses in PCP- and vehicle-treated animals. However, different settings gave different PPRs: 2 pulse stimulation in layer I with an ISI of 50 ms produced paired-pulse facilitation, whereas the ratio between the first 2 events in the HFS experiment (ISI, 20 ms) was equal to 1. This difference is most likely due to stimulation of GABAergic terminals from different types of interneurons; we expect layer I stimulation to recruit both perisomatic and dendritic inhibitory inputs, and HFS to predominately stimulate perisomatic synapses.

Extrasynaptic GABAA-Mediated Current Is Reduced in PCP Animals

The diminished charge transferred in the decay phase of the HFS (weight of tau) in PCP animals could be due to decreased synaptic spill-over related to a reduction in the number of GABAergic release sites or, alternatively, a compromised content of extrasynaptic GABAA receptors in the recorded neuron. To further investigate the level of extrasynaptic GABAA receptor current in PCP animals in a manner decoupled from synaptic over-spill, the level of δ-containing GABAA receptor-mediated conductance was measured following bath perfusion with THIP. THIP induced a significantly reduced GABAA receptor-mediated current in PCP compared with vehicle animals, most likely due to a reduced expression of THIP-activated GABAA receptors. In a recent study, it has been shown that acute administration with THIP reversed PCP-induced deficits in recognition memory in rats (Damgaard et al. 2011). It is possible that a reduced level of extrasynaptic GABAA receptors underlies some of the observed GABAergic deficits in these animal models. Our results suggest that the early postnatal PCP model could be a useful model for future investigations within this matter.

Responses to GABA Reuptake Inhibition Are Compromised in PCP Animals

Inhibition of GAT-1 by NNC-711 decreased the amplitude of the eIPSC1 in vehicle animals, an effect that has been explained by presynaptic inhibition of GABA release by autoinhibitory GABAB receptors (Deisz and Prince 1989). However, since GABAB receptors were blocked during our experiments, the amplitude reduction might involve slow desensitization of GABAA receptors in response to the presence of low levels of ambient GABA (Overstreet et al. 2000). Interestingly, PCP animals displayed insensitivity toward the effects of NNC-711 on eIPSC1 amplitude, suggesting a PCP-induced deficit in GAT-1. This is supported by a study showing reduced GAT-1 expression following dosage of adult rats with low levels of PCP over 1 month (Bullock et al. 2009). However, extrasynaptic current in PCP animals as assessed by weight of tau showed the same level of up-regulation following GABA reuptake inhibition as vehicle animals, which do not point toward a PCP-induced deficit in GAT-1.

Conclusion

The presented results indicate a reduced GABAergic synaptic input to pyramidal neurons in layer II/III of the PFC in early postnatal PCP-treated rats. PCP treatment also reduced extrasynaptic inhibitory current and seemed to introduce deficits in GAT-1 function. These functional deficits make early postnatal PCP treatment a promising model for future investigations of schizophrenia focusing on GABAergic deficits and cognitive dysfunctions.

Funding

The work was supported by the Faculty of Pharmaceutical Sciences at University of Copenhagen, H. Lundbeck A/S and the Danish Agency for Science, Technology and Innovation (Drug Research Academy grant no. 08-342-191).

Notes

We thank Lis Glasdam Jensen for laboratory assistance. Conflict of Interest: The authors affiliated at Lundbeck have declared no conflict of interest.

References

Abidin
I
Eysel
UT
Lessmann
V
Mittmann
T
Impaired GABAergic inhibition in the visual cortex of brain-derived neurotrophic factor heterozygous knockout mice
J Physiol
,
2008
, vol.
586
(pg.
1885
-
1901
)
CM
Malhotra
AK
Elman
I
Goldberg
T
Egan
M
Pickar
D
Breier
A
Comparison of ketamine-induced thought disorder in healthy volunteers and thought disorder in schizophrenia
Am J Psychiatry
,
1999
, vol.
156
(pg.
1646
-
1649
)
Akbarian
S
Huntsman
MM
Kim
JJ
Tafazzoli
A
Potkin
SG
Bunney
WE
Jr
Jones
EG
GABAA receptor subunit gene expression in human prefrontal cortex: comparison of schizophrenics and controls
Cereb Cortex
,
1995
, vol.
5
(pg.
550
-
560
)
Bayer
SA
Altman
J
Russo
RJ
Zhang
X
Timetables of neurogenesis in the human brain based on experimentally determined patterns in the rat
Neurotoxicology
,
1993
, vol.
14
(pg.
83
-
144
)
Benes
FM
Berretta
S
GABAergic interneurons: implications for understanding schizophrenia and bipolar disorder
Neuropsychopharmacology
,
2001
, vol.
25
(pg.
1
-
27
)
Broberg
BV
Dias
R
Glenthoj
BY
Olsen
CK
Evaluation of a neurodevelopmental model of schizophrenia—early postnatal PCP treatment in attentional set-shifting
Behav Brain Res
,
2008
, vol.
190
(pg.
160
-
163
)
Brown
AS
Derkits
EJ
Prenatal infection and schizophrenia: a review of epidemiologic and translational studies
Am J Psychiatry
,
2010
, vol.
167
(pg.
261
-
280
)
Bullock
WM
Bolognani
F
Botta
P
Valenzuela
CF
Perrone-Bizzozero
NI
Schizophrenia-like GABAergic gene expression deficits in cerebellar Golgi cells from rats chronically exposed to low-dose phencyclidine
Neurochem Int
,
2009
, vol.
55
(pg.
775
-
782
)
Curley
AA
Lewis
DA
Cortical basket cell dysfunction in schizophrenia
J Physiol
,
2012
, vol.
590
(pg.
715
-
724
)
Damgaard
T
Plath
N
Neill
JC
Hansen
SL
Extrasynaptic GABAA receptor activation reverses recognition memory deficits in an animal model of schizophrenia
Psychopharmacology (Berl)
,
2011
, vol.
214
(pg.
403
-
413
)
Dean
B
Hussain
T
Hayes
W
Scarr
E
Kitsoulis
S
Hill
C
Opeskin
K
Copolov
DL
Changes in serotonin2A and GABA(A) receptors in schizophrenia: studies on the human dorsolateral prefrontal cortex
J Neurochem
,
1999
, vol.
72
(pg.
1593
-
1599
)
Deisz
RA
Prince
DA
Frequency-dependent depression of inhibition in guinea-pig neocortex in vitro by GABAB receptor feed-back on GABA release
J Physiol
,
1989
, vol.
412
(pg.
513
-
541
)
Deutsch
SI
Mastropaolo
J
Rosse
RB
Neurodevelopmental consequences of early exposure to phencyclidine and related drugs
Clin Neuropharmacol
,
1998
, vol.
21
(pg.
320
-
332
)
du Bois
TM
Deng
C
Han
M
Newell
KA
Huang
XF
Excitatory and inhibitory neurotransmission is chronically altered following perinatal NMDA receptor blockade
Eur Neuropsychopharmacol
,
2009
, vol.
19
(pg.
256
-
265
)
Green
MF
Kern
RS
Heaton
RK
Longitudinal studies of cognition and functional outcome in schizophrenia: implications for MATRICS
Schizophr Res
,
2004
, vol.
72
(pg.
41
-
51
)
Gupta
A
Wang
Y
Markram
H
Organizing principles for a diversity of GABAergic interneurons and synapses in the neocortex
Science
,
2000
, vol.
287
(pg.
273
-
278
)
Hajos
N
Nusser
Z
Rancz
EA
Freund
TF
Mody
I
Cell type- and synapse-specific variability in synaptic GABAA receptor occupancy
Eur J Neurosci
,
2000
, vol.
12
(pg.
810
-
818
)
Kaalund
SS
Riise
J
Broberg
BV
Fabricius
K
Karlsen
AS
Secher
T
Plath
N
Pakkenberg
B
Differential expression of parvalbumin in neonatal phencyclidine-treated rats and socially isolated rats
J Neurochem
,
2013
, vol.
124
(pg.
548
-
557
)
Kapfer
C
Glickfeld
LL
Atallah
BV
Scanziani
M
Supralinear increase of recurrent inhibition during sparse activity in the somatosensory cortex
Nat Neurosci
,
2007
, vol.
10
(pg.
743
-
753
)
Kirischuk
S
Grantyn
R
Intraterminal Ca2+ concentration and asynchronous transmitter release at single GABAergic boutons in rat collicular cultures
J Physiol
,
2003
, vol.
548
(pg.
753
-
764
)
Kogo
N
Dalezios
Y
Capogna
M
Ferraguti
F
Shigemoto
R
Somogyi
P
Depression of GABAergic input to identified hippocampal neurons by group III metabotropic glutamate receptors in the rat
Eur J Neurosci
,
2004
, vol.
19
(pg.
2727
-
2740
)
Krystal
JH
Karper
LP
Seibyl
JP
Freeman
GK
Delaney
R
Bremner
JD
Heninger
GR
Bowers
MB
Jr
Charney
DS
Subanesthetic effects of the noncompetitive NMDA antagonist, ketamine, in humans. Psychotomimetic, perceptual, cognitive, and neuroendocrine responses
Arch Gen Psychiatry
,
1994
, vol.
51
(pg.
199
-
214
)
Kuroda
M
Yokofujita
J
Murakami
K
An ultrastructural study of the neural circuit between the prefrontal cortex and the mediodorsal nucleus of the thalamus
Prog Neurobiol
,
1998
, vol.
54
(pg.
417
-
458
)
Lahti
AC
Weiler
MA
Tamara Michaelidis
BA
Parwani
A
Tamminga
CA
Effects of ketamine in normal and schizophrenic volunteers
Neuropsychopharmacology
,
2001
, vol.
25
(pg.
455
-
467
)
Li
Q
Clark
S
Lewis
DV
Wilson
WA
NMDA receptor antagonists disinhibit rat posterior cingulate and retrosplenial cortices: a potential mechanism of neurotoxicity
J Neurosci
,
2002
, vol.
22
(pg.
3070
-
3080
)
Ma
Y
Hu
H
Agmon
A
Short-term plasticity of unitary inhibitory-to-inhibitory synapses depends on the presynaptic interneuron subtype
J Neurosci
,
2012
, vol.
32
(pg.
983
-
988
)
Morris
BJ
Cochran
SM
Pratt
JA
PCP: from pharmacology to modelling schizophrenia
Curr Opin Pharmacol
,
2005
, vol.
5
(pg.
101
-
106
)
Nakatani-Pawlak
A
Yamaguchi
K
Tatsumi
Y
Mizoguchi
H
Yoneda
Y
Neonatal phencyclidine treatment in mice induces behavioral, histological and neurochemical abnormalities in adulthood
Biol Pharm Bull
,
2009
, vol.
32
(pg.
1576
-
1583
)
Nusser
Z
Naylor
D
Mody
I
Synapse-specific contribution of the variation of transmitter concentration to the decay of inhibitory postsynaptic currents
Biophys J
,
2001
, vol.
80
(pg.
1251
-
1261
)
Overstreet
LS
Jones
MV
Westbrook
GL
Slow desensitization regulates the availability of synaptic GABA(A) receptors
J Neurosci
,
2000
, vol.
20
(pg.
7914
-
7921
)
Patil
PG
Brody
DL
Yue
DT
Preferential closed-state inactivation of neuronal calcium channels
Neuron
,
1998
, vol.
20
(pg.
1027
-
1038
)
Paxinos
G
Watson
G
The rat brain in stereotaxic coordinates
,
1998
San Diego, CA, USA: Academic Press
Powell
SB
Sejnowski
TJ
Behrens
MM
Behavioral and neurochemical consequences of cortical oxidative stress on parvalbumin-interneuron maturation in rodent models of schizophrenia
Neuropharmacology
,
2012
, vol.
62
(pg.
1322
-
1331
)
Rizzoli
SO
Betz
WJ
Synaptic vesicle pools
Nat Rev Neurosci
,
2005
, vol.
6
(pg.
57
-
69
)
Santana
N
Mengod
G
Artigas
F
Quantitative analysis of the expression of dopamine D1 and D2 receptors in pyramidal and GABAergic neurons of the rat prefrontal cortex
Cereb Cortex
,
2009
, vol.
19
(pg.
849
-
860
)
Schleimer
SB
Hinton
T
Dixon
G
Johnston
GA
GABA transporters GAT-1 and GAT-3 in the human dorsolateral prefrontal cortex in schizophrenia
Neuropsychobiology
,
2004
, vol.
50
(pg.
226
-
230
)
Schneggenburger
R
Meyer
AC
Neher
E
Released fraction and total size of a pool of immediately available transmitter quanta at a calyx synapse
Neuron
,
1999
, vol.
23
(pg.
399
-
409
)
Thomson
AM
Bannister
AP
Interlaminar connections in the neocortex
Cereb Cortex
,
2003
, vol.
13
(pg.
5
-
14
)
Wang
C
McInnis
J
Ross-Sanchez
M
Shinnick-Gallagher
P
Wiley
JL
Johnson
KM
Long-term behavioral and neurodegenerative effects of perinatal phencyclidine administration: implications for schizophrenia
Neuroscience
,
2001
, vol.
107
(pg.
535
-
550
)
Wang
HX
Gao
WJ
Cell type-specific development of NMDA receptors in the interneurons of rat prefrontal cortex
Neuropsychopharmacology
,
2009
, vol.
34
(pg.
2028
-
2040
)
Woodhall
GL
Bailey
SJ
Thompson
SE
Evans
DI
Jones
RS
Fundamental differences in spontaneous synaptic inhibition between deep and superficial layers of the rat entorhinal cortex
Hippocampus
,
2005
, vol.
15
(pg.
232
-
245
)
Xiang
Z
Huguenard
JR
Prince
DA
Synaptic inhibition of pyramidal cells evoked by different interneuronal subtypes in layer v of rat visual cortex
J Neurophysiol
,
2002
, vol.
88
(pg.
740
-
750
)