Food Restriction Paradigm

Rats with a body mass of 200 to 230 g were assigned either to a control group that received food and water ad libitum or to the FR group. As described previously [23], FR animals were allowed to access to their daily meal (rat food pellets; Standard Diet GLP, Mucedola, Italy) for only 2 h, from 1100 to 1300 hours, whereas water was available ad libitum. Body weight and food consumption were measured daily during the 3-week period in which the FR regimen was applied. The amount of food consumed by FR animals during the 2 h in which the food was available, increased gradually to achieve a constant level (21.7±0.5 g/2 h) within 10 to 14 days, being slightly but significantly (P<0.05) lower than what observed in rats fed ad libitum during 24 h (24.6±0.3 g/24 h).

The amount of food eaten by FR as well as rats fed ad libitum did not significantly change after neither acute administration of WIN 55,212-2 (22.2±0.8 g/2 h) nor SR141716 (20.7±0.7 g/2 h).

Microdialysis and Drug Treatment

Rats were anesthetized by intraperitoneal (i.p.) injection of chloral hydrate (0.4 g/kg), and a concentric dialysis probe was inserted at the level of the mPFC (A +3.2, ML +0.8, V –5.3 relative to the bregma) or NAcc shell (A +2.2, ML +1, V −7.8), as shown in Fig. 1, according to the Paxinos atlas [24]. The active length of the dialysis membrane (Hospal Dasco, Bologna, Italy) was restricted to 4 mm for mPFC and 2 mm for NAcc. In the mPFC the lenght of the dialitycal membrane allowed to sample from both infralimbic and prelimbic cortices. Experiments were performed in freely moving rats ~24 h after probe implantation to allow recovery from surgery procedures. Ringer’s solution [3 mM KCl, 125 mM NaCl, 1.3 mM CaCl2, 1 mM MgCl2, 23 mM NaHCO3, 1.5 mM potassium phosphate (pH 7.3)] was pumped through the dialysis probe at a constant rate of 2 µl/min. Samples of dialysate were collected every 20 min from 830 to 1500 hours and immediately analyzed for dopamine by HPLC with electrochemical detection, as previously described [25]; the detection limit for dopamine was 2 fmol per injection. The average neurotransmitter concentration in the first two samples was taken as 100%, and all subsequent values were expressed as means ± SEM relative to the basal value. The mean in vitro recovery of the probes was 15±3%. All probes were tested before implantation, and those with a recovery value outside of this range were not used. The absolute concentration of dopamine was not corrected for this value. At the end of each experiment, the placement of the probe was verified histologically. All rats in which the probe was located outside of the target region were excluded from the analysis. Rats were subjected to acute or long-term (twice a day for 21 days) treatment with SR141716 (1 mg/kg, i.p.) dissolved in ethanol/Tween 80/saline (11: 16, v/v/v). In a separate group of FR rats, SR141716 was administered locally in the mPFC through the dialysis probe via inverse dialysis. In this case a 1 mM solution of SR141716 was prepared in ethanol/Tween 80/Ringer (11: 16, v/v/v) then diluted to 10 µM to be perfused. Forty min before food presentation basal Ringer solution was replaced with the solution containing SR141716. As the probe recovery was 15±3%, we assume that the amount of SR141716 reaching the brain was approximately 1.2–1.7 µM. WIN 55,212-2 was dissolved in distilled water containing one drop of Tween 80 per milliliter and administered acutely at a dose of 5 mg/kg (i.p.). WIN 55,212-2 and SR141716 were administered in a volume of 3 ml/kg, and an identical volume of vehicle was administered as a control.

Figure 1

Schematic representation of the localization of the dialysis probe in the mPFC and NAcc shell.

Recording of Spontaneous Exploratory Behavior

Spontaneous exploratory behavior was measured by a motility meter (Omnitech Electronics Inc.) in FR rats at different time points following the three-week period of FR regimen: during the anticipatory phase (1 h and 5 min before food presentation), at the end of feeding and 2 h after food removal. A group of rats fed ad libitum were tested as controls. Eight rats were used for each group. On the test day, rats were brought into the test room under standardized environmental conditions and acclimated for 1 h before the beginning of the experiment. They were then individually placed in the test cage. Each cage consisted of a 41×41×30 cm specially designed sound-attenuating shells made of polypropylene and an expanded PVC sheet equipped with two sets of photocells located at the bottom of both cage sides, projecting horizontal infrared beams. Animals were isolated from noise of the recorder and printer used to acquire the data by placing these devices in a different room. Horizontal activity (number of horizontal infrared beams interruptions), total distance traveled (centimeters), locomotion time (seconds), and rest time (seconds) were acquired with a recorder (Omnitech Electronics, Inc.). Recordings were performed every 5 min, starting immediately after placing the rat in the cage, over a period of 30 min.

Preparation of mPFC and VTA Slices and Whole-cell Patch-clamp Recordings

Coronal and horizontal slices containing the mPFC and the VTA, respectively, were prepared from rats fed ad libitum and FR animals at different time points relative to the 2-h feeding period (60 and 5 min before food presentation; at the end of the feeding period; and 1 h after food removal) as previously described for the hippocampus [26]–[27].

Whole-cell patch-clamp recordings from mPFC neurons were performed as previously described [26]–[27]. In brief, spontaneous GABA type A (GABAA) receptor–mediated ISPCs (sIPSCs) were recorded in the presence of the ionotropic glutamate receptor antagonist kynurenic acid (1 mM) with the use of an Axopatch 200-B amplifier (Axon Instruments, Sunnyvale, CA, USA), and they were filtered at 2 kHz and digitized at 5 kHz. Off-line analysis of sIPSCs was performed with Minianalysis 6.0 software (Synaptosoft, Decatur, GA, USA). For application of the depolarization-induced suppression of inhibition (DSI) protocol, we recorded sIPSCs under control conditions for 3 min and then imposed a depolarization step from –65 to 0 mV on the postsynaptic neuron for 5 s. DSI was calculated from the sIPSC amplitudes and frequencies measured before (baseline) and after the depolarization step. In experiments evaluating the effect of various drugs perfusion, the last 180 s of each recording were used for statistical comparison with baseline. In a different set of experiments, neuronal excitability was studied in the current-clamp mode by evaluating spontaneous and evoked action potential (AP) firing rate in VTA and mPFC neurons, respectively. Dopamine neurons in the VTA were identified by the presence of an Ih current, as previously reported [35], which was evoked by incrementally hyperpolarizing (steps of 10 mV) the membrane from an holding potential of −70 mV. The resting potential in spontaneously firing dopamine neurons was measured by single cell patch-clamp recordings in the current-clamp mode.

Immunoblot Analysis

At the end of the food restriction protocol, rats were killed by decapitation for immunoblot analysis. The brain was rapidly removed, and the mPFC was dissected on ice, weighed, and stored at –80°C until analysis. Total protein was extracted from the tissue with the use of a kit (Membrane and Cytoplasmatic Protein Extraction kit, Bio Basic Inc, Markham Ontario, Canada), and the protein concentration of the extract was determined with the use of a DC Protein Assay kit (Bio-Rad, Milan, Italy). The extract (40 µg of protein in 15 µl) was incubated for 10 min at 70°C and then fractionated by SDS-polyacrylamide gel electrophoresis (NuPAGE Novex 4–12% Bis-Tris Midi Gel, Life Technologies, Monza, Italy). The separated proteins were transferred to a polyvinylidene difluoride membrane (Immobilon-P; Millipore, Milan, Italy) with the use of a Criterion Blotter (Bio-Rad), after which the membrane was exposed to 5% nonfat dried milk in Tris-buffered saline containing 1% Tween 20 (TBS-T) before an over-night incubation for at 4°C with rabbit polyclonal antibodies to CB1 receptors (1500) (Calbiochem, Merk-Millipore, Milan, Italy) and mouse monoclonal antibodies to glyceraldehyde-3-phosphate dehydrogenase (GAPDH, 15000) (Millipore) diluted in TBS-T containing 5% nonfat dried milk. The membrane was then incubated for 1 h at room temperature with horseradish peroxidase–conjugated goat secondary antibodies (anti-rabbit, Sigma, Milan, Italy; anti-mouse, Millipore, Milan, Italy) in TBS containing 5% nonfat dried milk and 0.5% Tween 20, after which immune complexes were detected with Luminata Crescendo and Forte substrates (Millipore). The intensity of immunoreactive bands was quantitated with the use of a Geliance 600 Chemi Imaging System (Perkin Elmer, Monza, Italy) and GeneTools software (Perkin Elmer). The intensity of CB1 receptor bands was normalized by that of the corresponding GAPDH band.

Antibodies

The following primary antibodies were used in this study: rabbit anti-CB1 (1500; Millipore, Billerica, MA), rabbit anti-CCK (1500; Millipore, Billerica, MA), mouse anti-parvalbumin (12000; Millipore, Billerica, MA), guinea pig anti-GAD65 (11500; Synaptic System, Goettingen Germany).

Tissue Preparation and Immunofluorescence

Rats were deeply anesthetized with Equithesin (1 g sodium pentobarbital, 4.251 g choral hydrate, 2.125 g MgSO4, 12 ml EtOH, 43.6 ml propylene glycol, adjusted to a total volume of 100 ml with distilled water) at 0.3 ml/g and transcardially perfused with 4% paraformaldehyde in 0.1 M phosphate buffer, pH 7.3, at 1 h and 5 min before food presentation, during eating phase and 1 h after food removal. Brains were then quickly removed and post-fixed for 4 h with 4% paraformaldehyde. After thorough rinsing, brains were cryoprotected in 20% sucrose overnight at 4°C. The day after sagittal sections of 50 µm thickness were obtained with a vibratome (Leica Microsystems, Milan, Italy) and maintained in antifreeze solution (30% ethylene glycol, 20% glycerol and 50% 0.05 M phosphate buffer v/v) at −20°C. Indirect immunohistochemistry was performed with free-floating sections as follows: (Method 1, double labeling): the sections were washed with PBS, permeabilized for 1 h with 0.1% Triton X-100 in PBS (PBS-T), incubated for 1 h with 10% normal donkey serum (Jackson ImmunoResearch, West Grove, PA, USA) in PBS-T and incubated overnight with CCK primary antibody. The sections were then rinsed in PBS-T and incubated with CY3-conjugated polyclonal donkey anti-rabbit antibody for 2 h at RT (11500 in PBS-T; Millipore). The sections were washed several time in PBS-T, re-incubated for 1 h with 10% of goat serum in PBS-T and incubated overnight at 4°C with CB1 receptor primary antibody.

The day after the sections were washed several times and incubated for 1 h with CY2-conjugated polyclonal goat anti-rabbit antibody for 2 h a RT (1750 in PBS-T; Millipore). After several washing in PBS-T, nuclei were counterstained with 1 µg/ml Hoechst 33342 in PBS for 30 min, rinsed again in PBS and coverslipped with gelatin-glicerol mounting medium (Sigma-Aldrich, Milan, Italy). (Method 2, triple labeling): serial sections were used and processed as described for method 1. After permeabilization with 0.1% Triton X-100 and incubation with normal donkey serum, the sections were incubated with a combinations of three primary antibodies (anti CB1 receptor, anti-parvalbumin and anti GAD 65) raised in different species. The sections were then rinsed in PBS-T, incubated with the appropriate secondary antibodies, raise either in goat or donkey, conjugated to one of the following fluorophores cyanine-derived Cy2 (1750 in PBS-T; Millipore), Cy5 (11500 in PBS-T; Millipore) and Dylight 594 (11500 in PBS-T; Jackson Immunoreasearch, West Grove, PA), incubated with Hoescht 33342 and coverslipped with gelatin glycerol mounting medium (Sigma-Aldrich, Milan, Italy). Control sections not exposed to primary antibodies did not yield positive staining (results not shown).

Confocal Microscopy and Data Analysis

Acquisitions were carried out using a Leica TCS SP5X Inverted Supercontinuum Confocal Laser Scanning microscope (Leica Microsystems, Heidelberg, Germany) equipped with a diode 405 laser and a white laser and using the multichannel acquisition mode to avoid fluorescence crosstalk. Images (2048×2048 pixels) were obtained with a Plan Apo 63X oil immersion objective NA 1.4. For quantification, confocal settings were chosen to ensure that the signal obtained from co-localization of CB1/CCK and CB1/PV/GAD35 were fully resolved within the dynamic range of detection (8 bit, 0–255) and no less or saturation of signal did occur. Acquisitions were performed in the section III and IV of the prefrontal cortex for co-localization study of CB1/PV/GAD65 and CB1/CCK, respectively. Same setting was used for all images captured consisting of identical detector gain, amplifier gain, amplifier offset, pinhole diameter (1 Airy unit), laser power excitation, scan mode and speed, frame size (2048×2048 pixels, 8 bit) and magnification of 1.4×10−2 µm2/pixel. In our experiments for excitation of the Hoescht 33342, the 405 nm wavelength was used and the emission was detected in the 417–486 nm range; for excitation of the Cy2, the 492 nm wavelength was used and the emission was detected in the 497–548 nm range; for the Cy3 the 550 nm wavelength and the 660–1250 nm range were used; for the Cy5 the 650 nm wavelength and the 655–730 nm range were used; and for the DyLight 594 the 594 nm wavelength and the 604–650 nm range were used respectively for excitation and emission. The images were processed with the image-analysis program images NIH Image J software (http://rsb.info.nih.gov/nih-image). Control images were obtained using sections incubated with secondary antibody alone and/or without antibodies.

To determine the density of CB1 receptor and their co-localization with CCK or GAD65 proteins, images were first segmented using a threshold that maximized the selection of immunofluorescent puncta over background labeling and then processed with the “co-localization” module, in which a mask is generated from the comparison of two different confocal channels. The number and density of puncta were then calculated with ImageJ (version 1.46a). It should be noted that the ratio between the intensity of CB1 receptor immunofluorescent puncta and diffuse extrasynaptic labeling is considerably lower in young animals compared to adult, making it impossible to select an intensity threshold that allows detection of all clusters without also including some background fluorescence. Because of this the number of immunopositive puncta will be either overestimated, due to the inclusion of background labeling, or underestimated, by elimination of the less intensely fluorescent clusters. This bias can be corrected in part by selecting clusters according to an appropriate size threshold during the automated counting procedure. In the present analysis we found that selecting clusters larger than 0.084 µm2 (corresponding to five pixels) resulted in counts that approximated the number of clusters determined by Leica software. Another important issue concerns the analysis of co-localization. Because the overlap between individual CCK/CB1 and CB1/GAD65 receptor positive clusters is only partial, the resulting co-localized clusters (i.e., pixels expressing both fluorescence channel signals) will have a considerably smaller size. To take this into account, we set a smaller size threshold (250 nm, corresponding to two pixels) for co-localized clusters. The co-apposition of CB1 receptor clusters with CCK or GAD-65 positive boutons was estimated visually in the three orthogonal planes. For presentation, digital micrographs were processed with software Leica and only minimal contrast adjustments were made. Files were then imported into Adobe Photoshop (Adobe Systems, San Jose, CA), where images were cropped.

Chronic Administration of SR141716 Prevents the Increase in Dopamine Output in the mPFC Induced by Food Anticipation and Consumption in FR Rats

It has been shown that after chronic administration, tolerance develops to some of the behavioral and biochemical effects of SR141716 [30] but not to its anorexic effect [31]. To evaluate whether tolerance would develop also to the ability of this drug to antagonize the increase in dopamine output in the mPFC induced by food restriction, we evaluated the effect of chronic administration of SR141716 (1 mg/kg, i.p., twice a day for 21 days) in FR rats. The basal extracellular concentration of dopamine was not significantly affected by such long-term treatment (11.87±2.68 fmol per 40 µl sample; P=0.175 vs rats fed ad libitum). In animals chronically treated with SR141716, the increase in dopamine output during the anticipatory phase was abolished, whereas that during the consummatory phase was markedly reduced from a maximum of +350% of basal to +80% and delayed (Fig. 2C). ANOVA revealed a significant main effect of treatment [F(1, 189)=1.947, P<0.001], a significant main effect of repeated measures [F(18, 189)=0.763, P<0.001], and a significant interaction between factors [F(18, 189)=4.951, P<0.001].

Effect of Food Anticipation and Consumption on Extracellular Dopamine Concentration in the NAcc and its Modulation by CB1 Receptor Ligands in FR Rats

Consistent with previous observations [14], mesolimbic dopaminergic neurons projecting to the NAcc shell did not respond significantly to anticipation of a familiar food, and food presentation induced only a small and transient increase (to +40% of basal values) in dopamine output in this brain area of FR rats (Fig. 3). In contrast, FR rats subjected to acute administration of WIN 55,212-2 (5 mg/kg, i.p.), at 40 min before food presentation and at a dose which per se induced a small (+28%) and transient (40 min) increase in dopamine output in rats fed ad libitum (data not shown), showed a significant increase in the extracellular concentration of dopamine during both the anticipatory (up to +60% of basal) and consummatory (+130%) phases (Fig. 3). Moreover, as expected [13], the concomitant administration of SR141716, which per se did not significantly modify basal extracellular dopamine concentration, completely antagonized the effect of WIN55,212-2 on this parameter.

Figure 3

Changes in extracellular dopamine concentration in NAcc shell during anticipation and consumption of food in FR rats: modulation by CB1 receptor ligands.

ANOVA revealed a significant main effect of treatment [F(2,284 )=123.20, P<0.01], a significant main effect of repeated measures [F(18, 284)=164.51, P<0.01], and a significant interaction between factors [F(36, 284)=182.27, P<0.01].

Changes in Exploratory Behavior in Rats Exposed to Food Restriction

Because previous studies have shown that food restriction increases arousal in many mammalian species [32], we wanted to test whether in our FR animals, anticipation and consumption of food were associated with an altered spontaneous exploratory behavior assessed in a motility meter. The data illustrated in Fig. 4indicate that FR rats tested 5 min before food consumption displayed an increased horizontal and vertical activity, and traveled a greater distance, compared to animals fed ad libitum, as assessed during the initial 5 min of testing (Fig. 4A–F; one-way ANOVA; horizontal activity, F(5,49)=5.267, aP<0.05 vs rats fed ad libitum; vertical activity, F(5,49)=6.232, aP<0.05 vs rats fed ad libitum; total distance, F(5,49)=5.640, aP<0.05 vs rats fed ad libitum ). Moreover, FR animals tested 2 h after or 2 h before food consumption, when dopamine levels are similar to those observed in animals fed ad libitum, showed a significant decrease in exploratory activity compared to FR rats tested 5 min before food consumption (Fig. 4A–F; bP<0.05 vs FR 5 min before, post-hoc analysis with Newman-Keuls)).

Figure 4

Spontaneous exploratory activity in FR and animals fed ad libitum.

Changes in CB1 Receptor-mediated Regulation of GABAergic Synapses in the mPFC of Rats Exposed to Food Restriction

The activity of GABAergic interneurons with their inhibitory synapses can regulate the excitability of mPFC glutamatergic neurons that project onto VTA and that, in turn, control the excitability of mesocortical dopaminergic neurons. Given that CB1 receptors are located in these inhibitory synapses [33]and that FR can alter the function of the eCB system [7], we next analyzed the basal properties of GABAA receptor–mediated sIPSCs recorded in voltage-clamped (–65 mV) pyramidal neurons in slices of the mPFC obtained from FR and fed ad libitum rats. Both the frequency and amplitude of sIPSCs were significantly reduced at 60 and 5 min before food presentation, during the consummatory phase (at the end of the 2-h period of food availability), as well as at 1 h after food removal in FR rats compared with values measured in neurons from rats fed ad libitum that had free access to food at all time. Two-way ANOVA revealed a significant main effect of treatment (fed ad libitum vs FR) [amplitude, F(1, 265)=18.49; P<0.0001; frequency, F(1, 265)=38.54, P<0.0001] (Fig. 5A–C). To determine whether the reduced sIPSC frequency detected in FR rats was attributable to a change in CB1 receptor function as well as CB1-mediated regulation of presynaptic GABA release, we studied depolarization-induced suppression of inhibition (DSI) that involves the release of eCBs in response to an increase in intracellular Ca2+ concentration triggered by postsynaptic depolarization. The eCBs released, acting via a retrograde mechanism, activate the presynaptic CB1 receptors, inducing a transient suppression of GABA release from presynaptic terminals. As expected from previous observations [19], [34], administration of a brief depolarizing pulse (from –65 to 0 mV for 5 s) to pyramidal neurons from rats fed ad libitum resulted in a transient and significant reduction in sIPSC frequency (Fig. 5D), with no changes in current amplitude (data not shown), an effect likely dependent by a decreased probability of GABA release from presynaptic terminals. This scenario was supported by the observation that the DSI was blocked by bath application of SR141716 (Fig. 5E). DSI was not appreciable in brain slices isolated from FR rats sacrificed during the anticipatory phase (5 min before food presentation)(Fig. 5F) or during the consummatory phase, but was again detectable 1 h after food removal. Two-way ANOVA revealed a significant main effect of treatment (fed ad libitum vs FR) [F(1,131)=29.82; P<0.0001]; a significant main effect of time [F(3,131)=3.27; P<0.05], and a significant interaction between factors [F(3,131)=3.27; P<0.05] (Fig. 4G).

Figure 5

Changes in basal GABAergic sIPSCs and endocannabinoid-dependent DSI in mPFC principal neurons of FR rats during anticipatory and consummatory phase.

To further evaluate the change in CB1 receptor-mediated regulation of presynaptic GABA release observed in FR rats, we also tested the effect of WIN 55212-2. As reported by previous reports [33], in mPFC slices from animals fed ad libitum, 10 min bath perfusion with 5 µM WIN 55,212-2 resulted in a marked decrease in sIPSC frequency, an effect that was antagonized by the subsequent co-application of 1 µM SR141716 (Fig. 6A). The WIN 55,212-2-induced effect was greatly attenuated in mPFC neurons from FR rats sacrificed 5 min before food presentation (Fig. 6B) or during the 2-h period of food consumption, but it returned to control values 1 h after food removal (Fig. 6C). Two-way ANOVA revealed a significant main effect of treatment (fed ad libitum vs FR) [F(1,135)=21.84; P<0.0001] (Fig. 6C).

Figure 6

Changes in suppression of sIPSC frequency induced by WIN 55212-2, baclofen, or quinpirole in mPFC neurons of FR rats during anticipatory and consummatory phase.

The observation that the apparent sensitivity of CB1 receptors to endogenous and exogenous agonist stimulation was reduced in FR rats would be consistent with a reduced inhibitory control of presynaptic GABA release and with a consequent increase in the probability of neurotransmitter exocytosis. However, this was not the case, as the basal frequency of sIPSCs in FR rats was reduced compared with that in rats fed ad libitum (Fig. 5A–C). To investigate this discrepancy and to better understand whether a presynaptic additional mechanism might contribute to the modulation of GABA release, we examined a possible role for GABAB and D2 dopamine receptors, both involved in the control of GABA release and thought to be localized at the same GABAergic synapses at the mPFC [33]. To evaluate the possible changes in the function of these presynaptic receptors, we recorded in glutamatergic mPFC neurons the sIPSCs in the presence of baclofen and quinpirole, agonists of GABAB and D2 receptors, respectively. Bath application of baclofen (10 µM) or quinpirole (1 µM) induced a significant decrease in sIPSC frequency in mPFC slices from rats fed ad libitum, but the effect of each agonist was significantly (T-test, P<0.05) more pronounced in the FR animals sacrificed 5 min before food consumption, the time point in which we observed the marked alteration in CB1 receptor-mediated modulation of sIPSC frequency (Fig. 6D–F).

Changes in Neuronal Excitability in mPFC and VTA Projecting Neurons in Rats Subjected to FR

The reorganization of presynaptic receptors involved in the control of inhibitory transmission onto mPFC GABAergic synapses could result, in turn, in an alterated excitability of the whole mesocortical circuitry. In order to test this hypothesis, we performed current-clamp recordings in single mPFC glutamatergic and VTA dopaminergic neurons. As shown in Fig. 7, in mPFC glutamatergic projecting neurons of FR rats sacrificed 5 min before food consumption, injection of depolarizing current steps (from 0 to 200 pA) evoked firing of action potentials (APs) whose frequency was markedly increased [F(1, 435)=4.83, P<0.0001] compared to animals fed ad libitum (Fig. 7A, B). The resting membrane potential of cells from the two groups of animals was not significantly different (fed ad libitum: −63±3.7 mV; FR: −59±2.3 mV) (Fig. 7C).

Figure 7

Changes in mPFC pyramidal neuron excitability in FR rats.

Due to the their well established neurophysiological properties [35], VTA dopaminergic projecting neurons were easily identified on the basis of the presence of a typical Ih current sag that could be evoked in response to hyperpolarizing potentials (Fig. 8A). In agreement with their pace-maker properties [36], these neurons show a spontaneous firing of APs (Fig. 8B). The frequency of spontaneous APs resulted significantly increased (T-test, P<0.0001) in neurons of FR rats (3.8±0.32 Hz), tested 5 min before food consumption, compared to what observed in animals fed ad libitum (1.1±0.25 Hz) (Fig. 8B, C). The resting membrane potential of VTA dopamine neurons from FR rats (−44.3±1.6 mV) resulted significantly (T-test, P<0.05) lower than that found in rats fed ad libitum (−49±1.8 mV) (Fig. 6D).

Figure 8

Changes in VTA dopaminergic neurons excitability in FR rats.

Antagonism of Acute SR141716 Treatment on FR-induced Effects

In order to understand whether the FR-induced changes observed in both mPFC and VTA principal neurons were consequent of CB1 receptor activation before food consumption, we treated a subset of animals with a single intraperitoneal injection of SR141716 (1 mg/kg, i.p) 1 h before food presentation. Current-clamp experiment performed in both mPFC and VTA revealed that the increase in spontaneous firing rate of VTA dopaminergic neurons as well as the depolarization-induced AP discharge in mPFC glutamatergic neurons induced by FR were completely antagonized in FR rats injected with SR141716 (Fig. 9A, B) [one-way ANOVA and Newman-Keuls test: F(3,11)=55.24, P<0.0001; one-way ANOVA and Newman-Keuls test: F(3,24)=4.683, P<0.05]. The injection of SR141716 1 h before food presentation was also able to restore in FR rats the inhibitory efficacy of WIN 55,212-2 on sIPSCs frequency (Fig. 9C) [one-way ANOVA and Newman-Keuls test: F(3,15)=8.1, P<0.05 vs FR injected with vehicle] as well as the increased efficacy observed with baclofen perfusion in slices (Fig. 9D) [one-way ANOVA and Newman-Keuls test: F(3,24)=4.683, P<0.05 vs FR injected with vehicle].

Figure 9

Effects of the systemic administration of SR141716 in fed ad libitum and FR rats on mPFC and VTA neurons excitability and inhibitory activity of WIN 55212-2 and baclofen on sIPSC frequency in mPFC neurons.