Effects of neonatal ventral hippocampal lesion in rats on stress-induced acetylcholine release in the prefrontal cortex

Authors

  • François Laplante,

    1. Douglas Hospital Research Centre, Department of Psychiatry, McGill University, Montreal, Québec, Canada
    2. Department of Pharmacology/Therapeutics, McGill University, Montréal, Québec, Canada
    Search for more papers by this author
  • Carl W. Stevenson,

    1. Douglas Hospital Research Centre, Department of Psychiatry, McGill University, Montreal, Québec, Canada
    Search for more papers by this author
  • Alain Gratton,

    1. Douglas Hospital Research Centre, Department of Psychiatry, McGill University, Montreal, Québec, Canada
    Search for more papers by this author
  • Lalit K. Srivastava,

    1. Douglas Hospital Research Centre, Department of Psychiatry, McGill University, Montreal, Québec, Canada
    Search for more papers by this author
  • Rémi Quirion

    1. Douglas Hospital Research Centre, Department of Psychiatry, McGill University, Montreal, Québec, Canada
    2. Department of Pharmacology/Therapeutics, McGill University, Montréal, Québec, Canada
    Search for more papers by this author

Address correspondence and reprint requests to Rémi Quirion, Douglas Hospital Research Centre, 6875 Boulevard LaSalle, Verdun, Québec, Canada H4H 1R3. E-mail: remi.quirion@douglas.mcgill.ca

Abstract

Excitotoxic neonatal ventral hippocampus (NVH) lesions in rats result in characteristic post-pubertal hyper-responsiveness to stress and cognitive abnormalities analogous to those described in schizophrenia and suggestive of alterations in dopamine (DA) neurotransmission. Converging lines of evidence also point to dysfunctions in the cortical cholinergic system in neuropsychiatric disorders. In previous studies, we observed alterations in dopaminergic modulation of acetylcholine (Ach) release in the prefrontal cortex (PFC) in post-pubertal NVH-lesioned rats. These two neurotransmitter systems are involved in the stress response as PFC release of DA and Ach is enhanced in response to some stressful stimuli. As adult NVH-lesioned rats are behaviorally more reactive to stress, we investigated the effects of NVH lesions on tail-pinch stress-induced Ach and DA release in the PFC. Using in vivo microdialysis, we observed that tail-pinch stress resulted in significantly greater increases in prefrontal cortical Ach release in post-pubertal NVH-lesioned rats (220% baseline) compared with sham-operated controls (135% baseline). Systemic administration of the D1-like receptor antagonist SCH 23390 (0.5 mg/kg i.p.) or the D2-like receptor antagonist haloperidol (0.2 mg/kg i.p.), as well as intra-PFC administration of the D2-like antagonist sulpiride (100 µm), reduced stress-induced Ach release in PFC of adult NVH-lesioned rats. By contrast, intra-PFC administration of SCH 23390 (100 µm) failed to affect stress-induced Ach release in PFC of NVH-lesioned rats. Interestingly, using in vivo voltammetry, stress-induced stimulation of PFC DA release was found to be attenuated in adult NVH-lesioned rats. Taken together, these data suggest developmentally specific reorganization of prefrontal cortical cholinergic innervation notably regarding its regulation by DA neurotransmission.

Abbreviations used
AA

ascorbic acid

Ach

acetylcholine

DA

dopamine

DOPAC

3,4-dihydroxyphenylacetic acid

NVH

neonatal ventral hippocampal

PFC

prefrontal cortex

SCH

23390, R-(+)-7-chloro-8-hydroxy-3-methyl-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine hydrochloride

VTA

ventral tegmental area

Excitotoxic neonatal ventral hippocampus (NVH) lesions in rats result in characteristic neurochemical and behavioral abnormalities which emerge post puberty and are thought to bear resemblances to those described in schizophrenia (for reviews, see Lipska and Weinberger 2000 and Marcotte et al. 2001). NVH-lesioned animals display hyper-reactivity to stress and amphetamine, deficit in latent inhibition and pre-pulse inhibition of startle and impaired social behavior and working memory (Lipska et al. 1993, 1995, 2002; Chambers et al. 1996; Becker et al. 1999). These phenomena are suggestive of dysfunctions of the mesocorticolimbic dopamine (DA) neurotransmission believed to play key roles in many behavioral manifestations of schizophrenia (Weinberger 1987; Willner 1997).

Accumulating evidence suggests that alterations in the activity of cortical cholinergic neurons originating mainly from the nucleus basalis, substantia innominata and also the horizontal limb of the diagonal band (Luiten et al. 1987; Woolf 1991) may be involved in cognitive symptoms seen in major neuropsychiatric disorders (Sarter and Bruno 1999). Accordingly, studies have reported alterations in cortical and subcortical muscarinic and nicotinic receptor-binding sites in schizophrenic brains (Hyde and Crook 2001) and atypical anti-psychotics such as clozapine and olanzapine are potent muscarinic receptor antagonists (Bolden et al. 1991; Bymaster et al. 1996). Cortical cholinergic innervation plays a key role in learning and memory, attention and sensory gating (Sarter and Bruno 1997) and has been shown to be activated in response to stressful stimuli such as handling (Thiel et al. 1998) or restraint (Mark et al. 1996). Interestingly, these cognitive functions, as well as stress responses, are reported to be impaired in schizophrenia (Breier 1999; Jansen et al. 2000).

DA neurons arising from the ventral tegmental area (VTA) and projecting notably to the nucleus accumbens and prefrontal cortex (PFC) are also known to be activated in response to a variety of stressors (Thierry et al. 1976; Abercrombie et al. 1989; Doherty and Gratton 1992, 1996; Finlay and Zigmond 1997) resulting in enhanced DA release in these two regions. Experimental data also indicate that the mesocorticolimbic DA system exerts an excitatory influence on acetylcholine (Ach) release in the PFC (Casamenti et al. 1986; Day and Fibiger 1992), through the activation of D1-like receptors (Day and Fibiger 1993; Acquas et al. 1994). We have recently shown that activation of these receptors markedly enhanced PFC Ach release in post-pubertal NVH-lesioned rats (Laplante et al. 2004).

As NVH-lesioned animals are behaviorally more reactive to stress (Lipska et al. 1993; Flores et al. 1996; Brake et al. 1999), prefrontal cortical stress-induced release of neurotransmitters can be altered in this model. In this study, in vivo microdialysis and in vivo voltammetry were used to, respectively, investigate the effect of NVH lesions on post-pubertal Ach and DA release in the PFC in response to acute stress induced by a 20 min tail pinch. Given that DA modulates prefrontal cortical Ach function, we also examined the effect of DA D1-like and D2-like receptor antagonists on PFC Ach release in response to stress. Our results show that post-pubertal NVH-lesioned animals display greater increases in prefrontal cortical Ach release in response to tail-pinch stressors, which are blocked by both D1-like and D2-like receptor antagonists. Interestingly, stress-induced DA release is apparently attenuated in post-pubertal NVH-lesioned animals.

Materials and methods

Neonatal ventral hippocampal lesions

Lesions of the ventral hippocampus in pups were performed as previously described (Flores et al. 1996). Pregnant Sprague–Dawley rats at gestational day 15 were obtained from Charles River Canada (St-Constant, Québec, Canada), housed individually in 12-h light/dark cycle rooms and fed ad libidum. Animal care and surgery were carried out according to guidelines approved by the McGill University Animal Care Committee and the Canadian Council for Animal Care.

On postnatal day 7 (PD 7) male pups (14–17 g) within each litter (4–9 males/litter, 14 litters used) were randomly divided to sham-operated or lesion groups. Pups were anaesthetized by hypothermia by placing them on ice for 10–20 min and were immobilized on a platform fixed on a stereotaxic frame. An incision in the skin overlaying the skull was made and two 1 mm holes were drilled. A needle connected to an infusion pump through a Hamilton syringe was lowered into the ventral hippocampus at coordinates: AP −3.0 mm, ML ±3.5 mm relative to bregma and −5.0 mm relative to the surface of the skull (Sherwood and Timiras 1970). Ibotenic acid (0.3 µL, 10 µg/µL; Sigma Chemical Co, St. Louis, MO, USA) in 0.15 m phosphate-buffered saline pH 7.4 was infused bilaterally at a flow rate of 0.15 µL/min. Sham-operated animals received the same volume of phosphate-buffered saline. The needle was withdrawn 2 min after completion of the infusion. Pups were placed under a warming lamp and then returned to their mothers. On PD 21, rats were weaned and grouped 2–3 per cage, each cage containing sham-operated and NVH-lesioned rats. Experiments were performed on pre-pubertal (between PD 32 and 40) and post-pubertal (between PD 56 and 70) animals.

In vivo microdialysis

Microdialysis measurements of Ach were determined as described previously with some modifications (Quirion et al. 1994; Day et al. 2001; Laplante et al. 2004) Concentric I-shape microdialysis probes were made from AN69 Hospal fibers (molecular mass cut-off < 60 000 Da, i.d. = 220 µm, o.d. = 310 µm; Hospal, Lyon, France) with an open area of 4 mm. Rats were anaesthetized with a mix of ketamine (50 mg/kg; Vetrepharm, Bellville, ON, Canada), xylazine (5 mg/kg; Novopharm, Toronto, ON, Canada) and acepromazine (0.5 mg/kg; Ayerst, Montréal, PQ, Canada) and immobilized into a stereotaxic frame. The scalp was opened to uncover the skull and a 2 mm diameter hole was drilled to access the brain. The microdialysis probe was lowered into the medial prefrontal cortex according to the following coordinates for post-pubertal and pre-pubertal, respectively: AP +2.8 mm, ML +0.5 mm, DV −5 mm and AP +2.2 mm, ML +0.5 mm, DV −4 mm according to bregma (Paxinos and Watson 1982). The dialysis probe was secured with dental cement. Following surgery, animals were housed individually and allowed to recover for 2 days prior to their use in in vivo dialysis experiments.

At the beginning of each dialysis experiment, animals were placed in a lidless cage and probes were connected to a microliter syringe pump (Harvard Apparatus Inc., South Natick, MA, USA) in a manner that allowed free movement within the cage. The probe was perfused with a CSF-like solution containing (in mm) NaCl 123, KCl 3, CaCl2 1.3, MgCl2 1, NaHCO3 23 and sodium phosphate buffer 1 pH 7.4. Neostigmine bromine (100 nm) (Sigma RBI, Oakville, ON, Canada), an acetylcholinesterase inhibitor, was added to the solution. Probes were perfused at a flow rate of 5 µL/min. After a 1 h wash-out period for stabilization, 10 min dialysate fractions were collected. After noticeable stabilization of the baseline level of Ach release, which usually happens within 1 h after the beginning of measurements, tail-pinch stress was induced for 20 min. Tail-pinch stress consisted of placing a wooden clothes pin 1–2 cm from the base of the animal's tail. In response to this stressor, rats typically alternate between remaining immobile and making repeated attempts to remove the clothes pin (i.e. chewing).

In a second series of experiments, we measured the effects of DA antagonists on stress-induced stimulation of Ach release. The D1-like antagonist R-(+)-7-chloro-8-hydroxy-3-methyl-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine hydrochloride (SCH 23390; 0.5 mg/kg i.p.) or the D2-like antagonist haloperidol hydrochloride (0.2 mg/kg i.p.; Tocris. Ballwin, MO, USA) were injected 20 min prior of the tail pinch. The doses used here were in the same range as those previously shown to be effective in blocking the effects of amphetamine on PFC Ach release (Day and Fibiger 1992). Drugs were dissolved in a 0.9% sodium chloride saline solution (Abbott Laboratory, Montreal, PQ, Canada). SCH 23390 and S(–)-sulpiride (Sigma RBI) were also tested by intraprobe administration into the perfusion solution (100 µm). Each animal was dialysed once and received only one pharmacological treatment.

Ach assay

Ach content of the dialysate fraction was determined using HPLC separation, post-column enzymatic reaction and electrochemical detection (Day et al. 2001). Ach was separated from other molecules in the dialysate on a reverse-phase column (75 × 2.1 mm) pre-treated with sodium lauryl sulfate (Sigma Chemical Co). From this column, the eluate passes through an enzyme reactor (10 × 2.1 mm) containing acetylcholinesterase (AchE; EC 3.1.1.7; type VI-S; Sigma Chemical Co) and choline oxidase (1.1.3.17; Sigma Chemical Co) covalently bound to glutaraldehyde-activated Lichrosorb NH2 (10 µm; Merck, Darmstadt, Germany). All column hardware and packing materials were purchased from Chrompack (Raritan, MA, USA). The separated Ach reacts with the enzymes to give stoichiometric yield of hydrogen peroxide, which is electrochemically detected with a platinum electrode at a potential of +500 mV versus an Ag/AgCl reference electrode (Antec VT-03/Decade; Leiden, The Netherlands). The mobile phase consists of a 0.2 m aqueous potassium phosphate buffer, pH 8.0, containing 1 mm tetramethylammonium hydroxide (Sigma Chemical Co). The mobile phase is degassed online (CMA 260; Cargenie Medicin, Stockholm, Sweden) and delivered at 0.35–0.45 mL/min by a dual piston pump (ESA 580; ESA, Chelmford, MA, USA). Ach elutes at ˜ 4min, and the best detection limit of the assay is ˜ 10 fmol/injection. Sample concentrations were calculated by comparison with known standards. Data are expressed as percent of baseline. The baseline was calculated from the average of four samples preceding either tail pinch or drug injection.

In vivo electrochemistry

Electrochemical measurements of DA were performed in post-pubertal (PD 60) sham-operated and NVH-lesioned rats as described by Stevenson and Gratton (2003) with minor modifications. Animals were anaesthetized as described above and were implanted with a voltammetric electrode in the PFC according to these coordinates: AP +3.2, ML −0.8, DV −4.3 relative to bregma (Paxinos and Watson 1982). Animals were also implanted with an Ag/AgCl reference electrode and a stainless steel ground-wire in the contralateral and ipsilateral parietal cortices, respectively. Miniature pin connectors soldered to the voltammetric and reference electrodes and ground-wire were inserted into a Carleton connector (Ginder Scientific, Ottawa ON, Canada). This assembly was then secured with acrylic dental cement to three stainless steel screws threaded into the cranium.

Voltammetric electrodes, consisted of a bundle of three 30 µm diameter carbon fibres (Textron Systems, Wilmington, MA, USA) that extended 50–100 µm beyond the sealed tip of a pulled glass capillary (o.d. = 0.5 mm). The exposed fibres were repeatedly coated with a 5% solution of Nafion (Aldrich, Milwaukee, WI, USA), a perfluorinated ionomer which promotes the exchange of cations such as DA, while impeding the exchange of interfering anionic species such as ascorbic acid (AA) and 3,4-dihydroxyphenylacetic acid (DOPAC). Each electrode was calibrated prior to implantation to determine its sensitivity to DA and its selectivity for DA compared with AA. Calibrations were performed in 0.1 m phosphate-buffered saline (pH 7.4) that contained 250 µm AA to mimic brain extracellular conditions. Only electrodes with a highly linear response (r = 0.997) to increasing concentrations of DA and a DA-to-AA selectivity ratio of at least 1000 : 1 were used. It has been shown previously that these Nafion-coated carbon fibre electrodes will retain their sensitivity for DA and their selectivity for DA versus AA (and DOPAC) for several days following implantation (Doherty and Gratton 1997).

Electrochemical recordings were performed using a computer-controlled, high-speed chronoamperometric apparatus (Quanteon, Lexington, KY, USA). An oxidation potential of +0.55 mV (with respect to the reference electrode) was applied to the electrode for 100 ms at a rate of 5 Hz. The oxidation current was digitally integrated during the last 80 ms of each pulse. The sums of every 10 digitized oxidative cycles of the chronoamperometric waveform were automatically converted into equivalent values of DA concentration using the in vitro calibration factor. Values were displayed graphically on a video monitor at 2 s intervals. The reduction current generated when the potential was returned to resting level (0.0 V for 100 ms) was digitized and summed in the same manner and served as an index to identify the main electroactive species contributing to the stress-induced increases in electrochemical signals. With Nafion-coated electrodes and a sampling rate of 5 Hz, the magnitude of the increase in reduction current elicited by an elevation in DA concentration is typically 60–80% of the corresponding increase in oxidation current, i.e. the reduction-to-oxidation ratio (red : ox) = 0.6–0.8 (Gerhardt et al. 1988; Gratton et al. 1989; Pentney and Gratton 1991; Doherty and Gratton 1992; Mitchell and Gratton 1992). Previous studies have also indicated that the oxidation of AA is virtually irreversible (red : ox = 0–0.1), whereas that of DOPAC is almost entirely reversible (red : ox = 1.0), whereas the reduction to oxidation ratios for norepinephrine and serotonin are 0.4–0.5 and 0.1–0.3, respectively (Gerhardt et al. 1988).

Electrochemical recordings

Electrochemical recordings began 4–5 days after surgery. Immediately before a recording session, the in vitro calibration factor (the slope of the function relating increases in oxidation current to increases in DA concentration) for each animal's electrode was entered in the data acquisition software. This allowed on-line conversion of an increase in oxidation current to a value equivalent to the change in DA concentration that was required to produce an equal change in signal in vitro. Each animal was placed in a recording chamber and connected to the chronoamperometric instrument via a shielded cable and a low-impedance multichannel commutator (Airflyte, Bayonne, NJ, USA). Electrical interference was minimized by connecting a pre-amplifier, configured as a current-to-voltage converter (gain = 1 × 108), directly into the animal's head assembly. Experiments began only after obtaining 60 min of stable baseline recordings.

Electrochemical data format

Because of the inherent differences in sensitivity between Nafion-coated electrodes, in vivo changes in oxidation current recorded with different electrodes (i.e. in different animals) cannot be assumed to be equivalent. Thus, valid comparisons are possible only if the sensitivity of each electrode is calibrated against a standard and the electrochemical data are expressed as standard equivalent values. Because DA was used as the standard to calibrate electrode sensitivity, in vivo changes in oxidation current are expressed as nm equivalent values of DA concentration. Data are presented as changes in electrochemical signal (nm DA equivalents) relative to the signal level recorded immediately prior to the tail pinch (time 0). Because the recording at time 0 was the point of comparison for changes in electrochemical signal that followed, it was arbitrarily given a value of 0 µm. Therefore, a value of 0 µm is not meant to correspond to the concentration of extracellular DA; electrochemical data represent relative changes in extracellular DA levels elicited by a stimulus such as stress.

Histology

After in vivo microdialysis or voltammetry experiments, animals were sacrificed by decapitation and brains were removed and frozen in 2-methylbutane at −40°C and stored at −80°C until sectioning using a cryostat. Coronal sections (20 µm) were mounted onto gelatin-coated slides and stained with cresyl violet. Confirmation of the location of the dialysis probe or voltammetric electrode into the medial PFC (Fig. 1a,b) and lesion size of the ventral hippocampus (Fig. 1c) were performed under light microscopy.

Figure 1.

Histological representation of coronal sections of an adult rat brain stained with cresyl violet demonstrating (a) microdialysis probe location into the medial prefrontal cortex (arrow), (b) voltammetric electrode into the medial prefrontal cortex (arrow) and (c) the structural damage of the ventral hippocampus resulting of the neonatal lesion (arrow) compared with (d) sham-operated control rats. Scale bar: 2 mm.

Statistical analysis

For in vivo microdialysis experiments, repeated-measure three-way analysis of variance (anova) was performed for the analysis of the main effects of lesion, age and time as repeated measures, and interactions among these factors. To study the drug effects on stress-induced Ach release in post-pubertal rats, repeated measure two-way anova was performed for the analysis of the main effect of time as repeated measures, main effect of drug administration and interaction between these two factors. For the voltammetry experiments, the effect of the tail-pinch stressor on DA release was analyzed using two-way anova for the analysis of the main effect of time as repeated measures, main effect of lesion and interaction between these two factors. The basal level of Ach from the different experimental groups was analysed using one-way anova. Post-hoc Bonferonni tests were conducted where appropriate. A value of p < 0.05 was considered significant.

Results

The average basal efflux of Ach in PFC dialysates was 357 ± 90 fmol/50 µL (n = 6) for pre-pubertal sham-operated rats and 397 ± 86 fmol/50 µL (n = 7) for NVH-lesioned animals. In post-pubertal animals, values were of 229 ± 23 fmol/50 µL (n = 18) and 250 ± 35 fmol/50 µL (n = 24) for sham-operated and NVH-lesioned animals, respectively. In accordance with data reported previously (Laplante et al. 2004), there was no significant difference between the experimental groups in prefrontal cortical Ach release under basal conditions (F3,51 = 2.41; p = 0.07).

In response to tail-pinch stress, prefrontal cortical Ach release increased in all four experimental groups. Three-way anova revealed a significant main effect of time (F15,315 =12.78; p < 0.0001), significant main effect of lesion (F1,315 = 11.41; p = 0.0028), non-significant main effect of age (F1,315 = 0.41; p = 0.52), significant lesion × time interaction (F15,315 = 3.24; p < 0.0001), significant age × time interaction (F15,315 = 2.47; p = 0.002) and non-significant lesion × age × time interaction (F15,315 = 0.93; p =0.53). In pre-pubertal rats (Fig. 2a) post-hoc analysis failed to reveal significant difference in stress-induced PFC Ach release between sham-operated and NVH-lesioned animals. However, in post-pubertal rats (Fig. 2b), post-hoc analysis revealed significant differences in stress-induced PFC Ach release between NVH-lesioned rats (reaching a maximum of 220% baseline) and their sham-operated littermates (reaching a maximum of 135% baseline).

Figure 2.

Ach release in the prefrontal cortex of (a) pre-pubertal and (b) post-pubertal rats in response to 20 min tail-pinch stress. Data represent mean ± SEM (n = 6–7 in each group). Dialysate content is expressed as percent of baseline. Baseline was calculated from the average of four samples preceding tail pinch (black bar). Post-hoc analysis revealed significant differences of the stress effect on Ach release between sham-operated and NVH-lesioned rats *p < 0.05; **p < 0.001.

The D1-like receptor antagonist, SCH 23390 (0.5 mg/kg) injected 20 min before tail pinch, failed to affect stress-induced Ach release in sham-operated animals (Fig. 3a). However, SCH 23390 significantly reduced stress-induced enhancement of Ach release in post-pubertal NVH-lesioned rats compared with saline injection (Fig. 3b; F15,150 = 2.19; p < 0.009 for interaction). The D2-like receptor antagonist, haloperidol (0.2 mg/kg) injected 20 min before tail-pinch stress also reduced stress-induced stimulation of Ach release in post-pubertal NVH-lesioned rats compared with saline injection (Fig. 3b; F15,150 = 3.02; p = 0.0003 for interaction).

Figure 3.

Ach release in the prefrontal cortex of post-pubertal rats in response to the administration of D1-like receptor antagonist SCH 23390 (0.5 mg/kg) or the D2-like receptor antagonist haloperidol (0.2 mg/kg) 20 min prior to a 20 min tail-pinch stress (black bar) in (a) sham-operated and (b) NVH-lesioned rats. Data represent mean ± SEM (n = 6 in each group). Dialysate content is expressed as percent of baseline. Baseline was calculated from the average of four samples preceding drug administration (arrowhead).

Addition of SCH 23390 (100 µm) or sulpiride (100 µm) to the perfusate solution produces average basal effluxes of Ach in PFC dialysates of 206 ± 54 and 325 ± 69 fmol/50 µL, respectively (n = 6 in both groups) in post-pubertal NVH-lesioned animals. Local administration of these drugs did not significantly affect basal efflux of Ach (F2,33 = 0.81; p =0.45). Intraprobe administration of SCH 23390 (100 µm) did not significantly affect stress-induced stimulation of prefrontal cortical Ach release in post-pubertal NVH-lesioned rats (Fig. 4a; F1,150 = 2.05; p = 0.18 for lesion effect and F15,150 = 1.12; p = 0.34 for interaction). By contrast, intraprobe administration of sulpiride (100 µm) significantly reduced stress-induced stimulation of prefrontal cortical Ach release in post-pubertal NVH-lesioned rats (Fig. 4b; F15,150 = 2.89; p = 0.0005 for interaction).

Figure 4.

Ach release in the prefrontal cortex of post-pubertal NVH-lesioned rats in response to 20 min tail-pinch stress when (a) D1-like antagonist SCH 23390 (100 µm) and (b) D2-like antagonist sulpiride (100 µm) is included into the perfusate. Data represent mean ± SEM (n = 6 in each group). Dialysate content is expressed as percent of baseline. Baseline was calculated from the average of four samples preceding tail pinch (black bar).

In vivo electrochemical measurements of DA in PFC revealed that DA release was increased in response to the stressor. However, in post-pubertal NVH-lesioned rats, this enhanced response was significantly lower than in sham-operated animals (Fig. 5; F1,48 = 8.53; p = 0.0053 for lesion effect and F7,48 = 2.38; p = 0.036 for time × lesion interaction). It was not possible to determine the basal DA efflux with the voltammetry approach used here, however, a previous study suggested no change in frontal cortical basal level of DA in this model (Chrapusta et al. 2003).

Figure 5.

Dopamine release in the prefrontal cortex of post-pubertal rats following 20 min tail pinch (black bar). Data are presented as changes in electrochemical signal (nm DA equivalents) relative to the signal level recorded immediately prior to the onset of stress (time 0). The x-axis is considered to be the basal level. Data represent mean ± SEM (n = 7 in each group) Post-hoc analysis revealed significant differences of the stress effect on DA release between sham-operated and NVH-lesioned rats. *p < 0.05; **p < 0.01.

Discussion

We previously demonstrated that prefrontal cortical Ach release was increased to a greater extent after systemic and intra-PFC administration of the D1-like agonist SKF 81297 in post-pubertal NVH-lesioned rats (Laplante et al. 2004). However, the systemic administration of the D2-like agonist quinpirole produced no significant effect on Ach release in either sham-operated or NVH-lesioned rats (Laplante et al. 2004). In this study, tail-pinch stress elicited greater increases in PFC Ach release in post-pubertal NVH-lesioned rats than in sham-operated controls. Consequently, the heightened prefrontal cortical Ach release seen after pharmacological stimulation of D1-like receptors (Laplante et al. 2004) can be reproduced by an environmental stimulus known to activate the mesolimbic dopaminergic system and to increase DA release into the nucleus accumbens and PFC (Abercrombie et al. 1989; Doherty and Gratton 1992, 1996; Sullivan and Gratton 1998).

Injection of a D1-like antagonist systemically or of D2-like antagonists systemically or directly through the dialysis probe, normalized stress-induced stimulation of prefrontal cortical Ach release in NVH-lesioned rats, suggesting that the hyper-responsiveness of the cholinergic system may be mediated by dopaminergic innervation and both DA receptor families may be involved in that process. Both D1- and D2-like antagonists injected systemically have also been shown to reverse amphetamine-induced increase of prefrontal cortical Ach release without affecting basal levels (Day and Fibiger 1992), in accordance with our data. However, systemic injections of D1-like receptor agonists increased Ach release in the PFC, whereas the administration of D2-like agonists produced no significant effect (Day and Fibiger 1993; Laplante et al. 2004). Taken together, these studies suggest that DA stimulates prefrontal cortical Ach release through D1-like receptors (D1 and/or D5 molecular subtypes) but this stimulation is not present if D2-like receptors (D2, D3 and/or D4 molecular subtypes) are inhibited.

However, tail pinch increased PFC DA release to a lesser extent in post-pubertal NVH-lesioned rats compared with sham-operated controls. These data are consistent with previous studies in which increases in nucleus accumbens DA elicited by tail pinch, restraint stress or amphetamine were reported to be reduced or relatively unchanged in NVH-lesioned animals (Wan et al. 1996; Brake et al. 1999; Lillrank et al. 1999). The apparent hypo-responsiveness of mesocorticolimbic DA neurons to stressor demonstrated previously (Brake et al. 1999; Lillrank et al. 1999) and in our study suggests that heightened Ach release in the PFC of post-pubertal NVH-lesioned rats may be due to post-synaptic changes in mesocorticolimbic DA neurotransmission. Indeed, the hyper-phosphorylation of the cAMP responsive element binding protein, known to be involved in D1-like receptor signalling (Das et al. 1997), has been reported in the PFC of post-pubertal NVH-lesioned animals (Bhardwaj et al. 2001). Moreover, the expression of the immediate-early genes c-fos and nerve growth factor inducible-B is altered in the PFC in response to the administration of amphetamine in post-pubertal NVH-lesioned animals (Lillrank et al. 1996; Bhardwaj et al. 2003). In vivo intracellular recording also revealed that pyramidal neurons in the PFC and nucleus accumbens neurons exhibited excessive firing in response to electrical stimulation of the VTA in NVH-lesioned rats (O'Donnell et al. 2002; Goto and O'Donnell 2002).

In our experiments, DA release was measured using in vivo voltammetry. In vivo dialysis could have been used instead, especially as this was the method chosen to investigate Ach release. Although both methods have been widely used to investigate DA release in behaving animals, they differ in many ways including spatial and temporal resolution, susceptibility to diffusion and volume transmission (Kuhr et al. 1984; Westerink et al. 1987). Having access to both methods, we opted for in vivo voltammetry to measure DA release in our model because this method is more sensitive to monitor tail-pinch stress-induced DA release in the PFC (Finlay et al. 1995; Stevenson and Gratton 2003). A similar voltammetry approach is inexistent to monitor Ach release.

Regarding the mechanism(s) by which DA stimulates Ach release in the PFC, a previous study showed that local administration of amphetamine through the dialysis probe failed to increase Ach release in the PFC (Day and Fibiger 1992). This result suggests that cortical DA receptors were not directly involved in this stimulatory effect. Indeed, cortical DA receptors are not directly associated with cholinergic nerve terminals. In the PFC, both D1 and D5 receptors are expressed on pyramidal glutamatergic neurons and GABA inhibitory interneurons (Bergson et al. 1995; Muly et al. 1998; Ciliax et al. 2000). However, D2, D3 and D4 receptors are mainly expressed in GABA interneurons (Khan et al. 1998; Khan et al. 2001).

In the basal forebrain, anatomical evidence revealed the presence of catecholamine-containing nerve terminals in the substantia innominata (Jones and Cuello 1989; Zaborszky and Cullinan 1996). Moreover, basal forebrain cholinergic neurons receive direct inputs from the substantia nigra/ventral tegmental area (Gaykema and Zaborszky 1996). In addition, in the rat brain, D5 receptors are found in the basal forebrain cholinergic nuclei including nucleus basalis, diagonal band and medial septum (Ciliax et al. 2000). D2, D3 and D4 receptors seem absent in basal forebrain cholinergic neurons (Khan et al. 1998). It is thus possible that DA can stimulate Ach release in the PFC via direct action on basal forebrain cholinergic neurons most likely via the activation of D5 receptors. Arnold et al. (2001) reported that infusion of DA antagonists in the nucleus basalis failed to attenuate the ability of amphetamine to increase Ach release in PFC. This result could be explained by the restricted diffusion and penetration of the drug through dialysis probes and brain tissue. Therefore, the antagonists may interact with only a limited number of cholinergic neurons, especially considering that cholinergic neurons projecting to PFC also originate from other basal forebrain nuclei including the horizontal limb of the diagonal band (Luiten et al. 1987; Woolf 1991).

In a previous study, we observed increases in PFC Ach release following local perfusion of a D1-like agonist through the dialysis probe (Laplante et al. 2004). Here, the local perfusion of SCH 23390 failed to reduce stress-induced increases of PFC Ach release. This may be taken as evidence of the rather limited role of cortical D1-like receptors in Ach release and suggest the existence of other regulatory functions for this receptor family in this brain region. However, although the intracortical infusion of a rather high concentration of SCH 23390 (100 µm) failed to affect stress-induced PFC Ach release, it is possible that even higher doses of this antagonist could elicit an effect.

Beside its stimulatory action, DA can also indirectly increase PFC Ach release by inhibiting GABAergic neurons exerting a tonic inhibitory input to cholinergic neurons. Such mechanisms involving the inhibition by DA of GABAergic afferents originating from the nucleus accumbens and innervating basal forebrain cholinergic neurons was proposed (Casamenti et al. 1986; Moore et al. 1999) and appear to be mediated through D2-like receptors (Moore et al. 1999). The effect of local administration of sulpiride could also be explained by an indirect action via GABAergic neurons. Prefrontal cortical D2-like receptors are mainly associated with GABA interneurons (Khan et al. 2001) where they may exert an inhibitory action on cortical GABA release (Seamans et al. 2001). Interestingly, GABA also exerts a local inhibitory action on Ach release in the PFC (Giorgetti et al. 2000). Accordingly, sulpiride could potentiate the release of GABA, which in turn would prevent the heightened stress-induced Ach release. Interestingly, disruption in cortical GABAergic neurotransmission has been suggested to occur in NVH-lesioned animals (Bhardwaj et al. 2000; Lipska et al. 2003).

It was shown that glutamatergic neurotransmission in the PFC mediates mesolimbic DA neurotransmission during stress responses (Moghaddam 2002). In that regard, an earlier study revealed that stress-induced DA release in the PFC was reduced by the local activation of ionotropic glutamate receptors (Del Arco and Mora 2001). Interestingly, post-pubertal NVH-lesioned rats exhibit alterations in prefrontal cortical excitatory neurotransmission, as evidenced by increased cortical levels of glutamatergic receptor binding sites (Schroeder et al. 1999) and altered behavioral responses to glutamatergic drugs (Al Amin et al. 2000; Hori et al. 2000).

Alterations in prefrontal cortical circuitry as a result of neonatal lesions of the ventral hippocampi have been proposed to be responsible for the emergence of at least some of the physiological and behavioral abnormalities reported in the NVH animal model. Combined excitotoxic neonatal hippocampal and adult prefrontal cortical lesions were shown to normalize both the hyperlocomotion induced by novelty or amphetamine (Lipska et al. 1998) and the altered electrophysiological properties of nucleus accumbens neurons following stimulation of the VTA (Goto and O'Donnell 2004). On the basis of our and earlier results (Laplante et al. 2004), we propose that alterations in prefrontal cortical cholinergic innervation and its modulation by DA likely plays a role in behavioral abnormalities reported in the NVH model. This is of interest in the context of the purported roles of cortical cholinergic system in schizophrenia (Hyde and Crook 2001) and its modulation by mesocorticolimbic DA innervation, which is well known to be affected in this neuropsychiatric disorder.

Acknowledgements

This work was supported by grants from the Canadian Institutes of Health Research (CIHR). LS and AG are ‘Chercheur National’ of the ‘Fonds de la Recherche en Santé du Québec’. FL holds a studentship from CIHR.

Ancillary