Further evidence for involvement of the dorsal hippocampus serotonergic and γ-aminobutyric acid (GABA)ergic pathways in the expression of contextual fear conditioning in rats

Abstract

Intra-dorsal hippocampus (DH) injections of 8-hydroxy-2-(di-n-propylamino) tetralin (8-OH-DPAT), a serotonin-1A (5-hydroxytryptamine (5-HT)-1A) receptor agonist, were previously shown to inhibit the expression of contextual fear when administered six hours after conditioning. However, further understanding of the consolidation and expression of aversive memories requires investigations of these and other mechanisms at distinct time points and the regions of the brain to which they are transferred. Thus, the purpose of the present study was to investigate the role of DH serotonergic and γ-aminobutyric acid (GABA)ergic mechanisms in the expression of contextual fear 24 h after conditioning, reflected by fear-potentiated startle (FPS) and freezing behavior. The recruitment of the amygdala and medial prefrontal cortex (mPFC) in these processes was also evaluated by measuring Fos protein immunoreactivity. Although intra-DH injections of 8-OH-DPAT did not produce behavioral changes, muscimol reduced both FPS and the freezing response. Fos protein immunoreactivity revealed that contextual fear promoted wide activation of the mPFC, which was significantly reduced after intra-DH infusions of muscimol. The present findings, together with previous data, indicate that in contrast to 5-HT, which appears to play a role during the early phases of contextual aversive memory consolidation, longer-lasting GABA-mediated mechanisms are recruited during the expression of contextual fear memories.

Keywords

Contextual fear serotonin GABA dorsal hippocampus medial prefrontal cortex

Introduction

In 1970, Graeff and Schoenfeld found that methysergide, a serotonin (5-hydroxytryptamine [5-HT]) blocker, markedly increased responding suppressed by punishment in pigeons, a property characteristic of anxiolytic-like drugs. Soon after, Wise et al. (1972) added in vitro neurochemical evidence that supports the involvement of 5-HT in anxiety. These authors proposed that ascending 5-HT pathways facilitate the effects of punishment by acting on structures localized in both the forebrain and midbrain, the function of which is to suppress ongoing behavior.

In contrast to the theoretical model proposed by Wise et al. (1972), however, 5-HT in the periaqueductal gray matter (PAG), a mesencephalic area associated with defensive behavior (Brandão et al., 2005), was shown to inhibit aversion (Graeff and Rawlins, 1980; Graeff et al., 1996; Kiser and Lebovitz, 1975; Kiser et al., 1980; Schenberg and Graeff, 1978). Thus, to reconcile these contradictory findings, a dual 5-HT-defense hypothesis was proposed, suggesting that 5-HT enhances conditioned anxiety/fear by acting in the forebrain and inhibits unconditioned fear through actions in the dorsal PAG (Deakin and Graeff, 1991).

Fear conditioning has been reliably evaluated by assessing freezing responses and fear-potentiated startle (FPS) in rodents when they return to the background context in which they previously received footshocks (Avanzi and Brandão, 2001; Avanzi et al., 1998; Borelli et al., 2005; Fanselow, 2000; Silva et al., 2004). Moreover, evidence indicates that the hippocampus is a key structure involved in stimulus selection during the learning (Goosens, 2011; Winocur et al., 1987), acquisition (Phillips and LeDoux, 1992; Weitemier and Ryabinin, 2004), and expression (Kim and Fanselow, 1992; Milanovic et al., 1998; Strekalova et al., 2003) of associative fear memories evoked by contextual stimuli. The dorsal hippocampus (DH) receives an ascending projection from the median raphe nucleus (MRN), whose 5-HT neurons appear to be crucial for the expression of contextual freezing and FPS (Almada et al., 2009; Avanzi and Brandão, 2001; Avanzi et al., 1998; Borelli et al., 2005; Silva et al., 2004).

Freezing and FPS have been shown to be conveyed by distinct 5-HT-mediated circuits of the MRN (Silva et al., 2002, 2004). In a recent study, we showed that the reduction of 5-HT transmission in the MRN by locally infusing the 5-HT_1A receptor agonist 8-hydroxy-2-(di-n-propylamino)-tetralin (8-OH-DPAT) six hours after training decreased contextual freezing but did not alter FPS (Almada et al., 2009). This result has led us to determine whether 5-HT₁ or 5-HT₂ receptors in the DH are involved in the expression of contextual fear. We expected that local injections of 8-OH-DPAT into the DH would produce an ‘anxiogenic-like’ effect and that similar injections of ketanserin would produce the opposite ( i.e. an ‘anxiolytic-like’ effect). Intra-DH injections of 8-OH-DPAT inhibited both contextual freezing and FPS, although ketanserin treatment did not alter these responses. These data suggest the existence of an inhibitory mechanism between the incoming 5-HT projection from the MRN to DH (Almada et al., 2009). In the present study we extend this pharmacological analysis to the serotonergic mediation of the expression of contextual fear 24 h after conditioning.

Among other functions, the neurotransmitter γ-aminobutyric acid (GABA) has been associated with the modulation of the neural substrates involved in the expression of defensive reactions (Brandão et al., 2005). Serotonin modifies the excitability of GABA interneurons localized in the DH, especially the CA1 region (Katsurabayashi et al., 2003), and GABA_A receptors appear to modulate post-training processes that underlie aversive memory consolidation (Ammassari-Teule et al., 1991; Brioni and McGaugh, 1988). Furthermore, pyramidal neurons and inhibitory GABA interneurons are densely innervated by 5-HT neurons that originate in the MRN (Azmitia and Segal, 1978; Katsurabayashi et al., 2003). Thus, a reasonable assumption is that the expression of defensive responses elicited in a threatening environment depends on both serotonergic and GABAergic systems or their interaction at the hippocampal level.

The elucidation of these mechanisms in the DH may represent an important step toward identifying and understanding the neural basis of anxiety disorders. The main purpose of the present study was to assess the involvement of DH serotonergic and GABAergic neurotransmission in the expression of contextual fear conditioning in rats using the FPS paradigm. The effects of intra-DH GABAergic drugs on the freezing response exhibited by rats that were reexposed to the aversive context were then examined, followed by measurements of Fos protein immunoreactivity in the amygdala and medial prefrontal cortex (mPFC), structures anatomically and functionally connected to the DH (Barker and Warburton, 2011; Ferino et al., 1987; Gross and Canteras, 2012; Jay and Witter, 1991; Swanson, 1981).

Materials and methods

Animals

A total of 156 male Wistar rats from the animal house of the University of São Paulo, Ribeirão Preto campus, weighing 250–300 g, were used. During the experimental period they were housed in groups of four per cage with food and water available ad libitum in a temperature-controlled room (23±1ºC) under a 12 h/12 h light/dark cycle (lights on at 07:00). The rats were transported to the experimental room in their home cages and left undisturbed for one hour prior to testing. All efforts were made to minimize animal suffering and reduce the number of rats used. The experiments reported in this article were performed in accordance with the recommendations of the Brazilian Society of Neuroscience and Behavior and complied with the United States National Institutes of Health Guide for Care and Use of Laboratory Animals. The procedures were approved by the Committee for Animal Care and Use, University of São Paulo (no. 11.1.308.53.9).

Surgical procedures

The rats were anesthetized with ketamine/xylazine (100/7.5 mg/kg, intraperitoneal [i. p.]) and fixed in a stereotaxic apparatus (David Kopf, Tujunga, California, USA). The upper incisor bar was set 3.0 mm below the interaural line so that the skull was horizontal between bregma and lambda. After scalp anesthesia with 2% lidocaine, the skull was surgically exposed, and stainless-steel guide cannulae (10 mm length, 0.6 mm outer diameter, 0.4 mm inner diameter) were unilaterally (Experiment 1) or bilaterally (Experiment 2) implanted in the DH (coordinates: anterior/posterior, –4.2 mm; medial/lateral, ±2.6 mm; dorsal/ventral: –2.8 mm). In Experiment 1, the cannulae were implanted alternately between the right and the left DH. The cannulae were fixed to the skull with dental cement and two stainless-steel screws. At the end of surgery, each guide cannula was sealed with a stainless steel wire to prevent obstruction. The rats then received an intramuscular injection of penicillin G benzathin (Pentabiotic, 600,000 IU, 0.2 mL; Fort Dodge, Campinas, Brazil) and a subcutaneous injection of the antiinflammatory and analgesic Banamine (flunixin meglumine, 2.5 mg/kg (10 mg/mL, 0.2 mL); Schering-Plough, São Paulo, Brazil). After surgery, the rats were returned to their home cages in groups of four. The pharmacological approach used to assess the 5-HT mediation of context fear conditioning in Experiment 1 in the present study is a continuation of a study published previously (Almada et al., 2009), in which the injections were performed unilaterally. For comparison purposes, we kept the general methodology the same across both studies. In Experiment 2, we thought that the addition of the analysis of Fos distribution in the brain along with bilateral injections of GABAergic drugs would produce clearer results with bilateral labeling.

Drugs

The following drugs were used: the 5-HT_1A receptor agonist 8-OH-DPAT (Sigma-Aldrich, St. Louis, Missouri, USA), the 5-HT_2A/2C receptor antagonist ketanserin tartrate (Research Biochemicals International, Natick, Massachusetts, USA), the GABA_A receptor agonist muscimol (Sigma-Aldrich), and the GABA_A receptor antagonist bicuculline (Sigma-Aldrich). 8-OH-DPAT, muscimol, and bicuculline were dissolved in saline solution (0.9%) to final concentrations of 1 µg/0.2 µL, 0.5 µg/0.5 µL and 0.1 µg/0.5 µL, respectively. Ketanserin was dissolved in saline (0.9%) with Tween (0.2%) to a concentration of 5.5 µg/0.2 µL. Physiological saline alone served as the control for 8-OH-DPAT, muscimol and bicuculline. Physiological saline combined with Tween served as the control for ketanserin. The rats received the injections of vehicle or drugs 10 min before the test session. Previous reports were used as a reference for the selection of doses, the volume of injections, and time of testing for the serotonergic (Almada et al., 2009; Brandão et al., 2005; Holt and Maren, 1999) and GABAergic (Martinez et al., 2006) drugs.

Microinjection procedure

Infusions were delivered using an infusion pump (Harvard Apparatus, Holliston, Massachusetts, USA) in a volume of 0.2 μL during one minute (8-OH-DPAT and ketanserin) or 0.5 μL during 90 s (muscimol and bicuculline). A thin dental needle (0.3 mm outer diameter) attached by polyethylene tubing to a 5 μL Hamilton syringe was introduced through the guide cannulae. The injection needle extended 1 mm below the ventral tip of the implanted guide cannula. The displacement of an air bubble inside the polyethylene tubing that connected the syringe to the injection needle was used to monitor the microinjections. The injection needles were left in place for one minute after the end of the infusion to allow for diffusion. In Experiment 2, bilateral infusions were administered simultaneously.

Experiment 1: Involvement of DH serotonergic and GABAergic mechanisms in the expression of contextual FPS

The acoustic startle reflex is elicited by sudden, unexpected, and intense auditory stimulation and involves a series of rapid and phasic contractions of most skeletal muscles throughout the body. The FPS test has proven to be useful to analyze the neural systems involved in fear and anxiety. This test measures conditioned fear, reflected by an increase in the amplitude of the acoustic startle reflex in the presence of an explicit or contextual cue previously paired with footshock (Davis et al., 1993; Grillon, 2002). In Experiment 1, 92 rats were tested to reveal the involvement of DH serotonergic and GABAergic mechanisms in the expression of contextual fear, reflected by FPS, assessed 24 h after conditioning.

Apparatus

The test cage was a wire-grid cage (16.5×7.5×7.5 cm) fixed to a response platform by four thumb screws. The floor consisted of six 3.0 mm diameter stainless-steel bars spaced 1.5 cm apart. The cage and response platform were located inside a ventilated, sound-attenuating plywood chamber (64×60×40 cm). A loudspeaker located 10 cm behind the test cage delivered both the startle stimulus (100 dB, 50 ms burst of white noise) and continuous background noise (55 dB). The startle reaction of the rats generated pressure on the platform, and the analog signals were amplified, digitized and analyzed using Startle Reflex software (Med Associates, St. Albans, Vermont, USA). The startle reaction was recorded within a time window of 100 ms after the onset of the startle stimulus. Calibration procedures were conducted before the experiments to ensure equivalent sensitivities of the response platform.

Procedure

Matching: On the first two days, the rats were placed in the test cage for a five minute habituation period and afterward received a total of 30 startle stimuli with an interstimulus interval of 30 s. Each matching session lasted 20 min. The rats were assigned to different control and drug groups such that each group had similar average startle amplitudes based on the last matching day. This mean startle amplitude was taken as the baseline condition before training began. Training: Five days after surgery, the rats were conditioned to the background context in the experimental cage described above. The rats were individually placed in the cage and five minutes later received 10 footshocks (0.3 mA, 1 s) with a variable intertrial interval of 60–180 s (Almada et al., 2009; Borelli et al., 2005; Reimer et al., 2008; Santos et al., 2005). The footshocks were delivered through the training cage floor by a constant-current generator built with a scrambler (Albarsch Instruments, Porto Alegre, Brazil). Stimulus presentation was controlled by a microprocessor and input/output board (Insight Equipment, Ribeirão Preto, Brazil). Each animal was removed five minutes after the last shock and returned to its home cage. The duration of each training session was approximately 25 min. Testing: The test sessions were conducted without footshock presentation in the same cage used for matching and training. Twenty-four hours after training, the rats received an intra-DH injection of vehicle or drug and 10 min later were placed in the startle test cage. After five minutes of habituation, the rats received 30 startle stimuli (i.e. noise bursts) with a 30 s interstimulus interval. All of the experimental steps were performed in the morning between 08:00–11:00.

Histology

After the completion of the tests, the rats were deeply anesthetized with urethane (3 g/kg, i.p.; Sigma-Aldrich) and transcardially perfused with 0.9% saline followed by 10% formalin. The brains were removed from the skulls, maintained in formalin solution for two hours, and cryoprotected in 30% sucrose for 72 h. Serial 60 µm coronal brain sections were cut using a cryostat (−19°C), mounted on gelatin-coated slides, and stained with Cresyl violet (5%; Sigma-Aldrich) to localize the microinjection sites by microscopic examination according to the Paxinos and Watson (2007) rat brain atlas.

Statistical analysis

The software used for all of the statistical analyses was Statistica version 6.0. The data are expressed as mean±standard error of the mean (SEM). A two-way repeated-measures analysis of variance (ANOVA) was used to assess the effects of 8-OH-DPAT, muscimol and ketanserin injections in the DH on startle amplitude, with treatment (vehicle and drug) and condition (baseline and test) as the factors. Significant comparisons were followed by the Newman-Keuls post-hoc test. Values of p<0.05 were considered statistically significant.

Experiment 2: Involvement of DH GABAergic mechanisms in the contextual freezing response and neural activation of the amygdala and mPFC

Many studies have shown that rodents acquire fear by being exposed to an environment where they had been previously presented with aversive unconditioned stimuli, such as footshock. In this situation, the most prominent behavioral outcome is the freezing response, operationally defined as the total absence of movement, with the exception of respiration (Albrechet-Souza et al., 2011; Almada et al., 2009; Bouton and Bolles, 1980; Fanselow, 1984). In Experiment 2, 64 rats were tested to reveal the involvement of DH GABAergic mechanisms in the expression of contextual fear, reflected by the freezing response, assessed 24 h after conditioning, followed by measuring Fos protein expression in the amygdala and mPFC.

Apparatus

Two distinct chambers, chamber A and chamber B, were used for the fear conditioning procedures. Chamber A (48×26×25 cm) was used for the conditioning sessions. The side and back walls were made of gray acrylic, and the ceiling and front door were made of transparent Plexiglas. The grid consisted of 36 stainless-steel rods spaced 1.5 cm apart through which computer-controlled footshocks could be delivered. The chamber had a loudspeaker and 15 W red lamp mounted on the rear panel. A sound generator produced a 1 kHz tone (72 dB; Insight Instruments). The chamber was cleaned with 20% alcohol after each session. Chamber B (31×21×21 cm) was used as a different context during the test sessions. The side and back walls were made of steel, and the ceiling and front door were made of transparent Plexiglas. The grid was covered with durable plastic. The chamber had a sound generator, loudspeaker, and 15 W white lamp mounted in a different position compared with chamber A (Insight Instruments). The chamber was cleaned with 20% alcohol and scented with 0.5% acetic acid after each session. Both chambers were enclosed in wooden sound-attenuating boxes.

Procedure

Training: Individual rats were placed in chamber A. Following a five minute acclimation period, the rat received 10 footshocks (0.6 mA, 1 s) with an intertrial interval that varied randomly between 30–120 s. The rat was removed three minutes after the last shock and returned to its home cage. The conditioning session lasted approximately 15 min. Testing: After 24 h, the test session was conducted similarly to the training session but without the presentation of footshocks. The ‘same context’ group was tested in chamber A, and the ‘different context’ group was tested in chamber B. The experiments occurred during the morning between 09:00–11:00. The criterion used to assess fear conditioning was the duration of the freezing response during 15 min. All of the experiments were monitored by well-trained investigators and experienced with these behavioral recordings. The measurements were monitored live and recorded later through a video camera positioned inside the experimental boxes.

Fos protein immunohistochemistry

Two hours after the beginning of the test session, the rats were deeply anesthetized with urethane (3 g/kg, i.p.; Sigma-Aldrich) and intracardially perfused with 0.1 M phosphate-buffered saline (PBS) followed by 4% paraformaldehyde in 0.1 M PBS (pH 7.4). The brains were removed from the skulls, immersed for two hours in paraformaldehyde, and stored for 72 h in 30% sucrose in 0.1 M PBS for cryoprotection. The brains were quickly frozen in isopentane (−40°C) and sliced in a cryostat (−19°C).

Brain slices (40 µm) were collected in antifreeze solution, and Fos protein immunoreactivity was detected using standard techniques as previously described (Albrechet-Souza et al., 2011; Almada et al., 2009). Briefly, the sections were first treated with 1% hydrogen peroxide in 0.1 M PBS for 10 min and incubated with primary c-Fos antibody (1:4000, rabbit polyclonal; Santa Cruz Biotechnology, Santa Cruz, California, USA) in 0.1 M PBS enriched with 0.2% Triton-X and 0.1% bovine serum albumin overnight at 23±1ºC. The sections were then incubated in biotinylated goat anti-rabbit secondary antibody (1:400; Vector Laboratories, Burlingame, California, USA) and avidin-biotin complex (1:200; Vector Laboratories) for one hour each. Fos immunoreactivity was revealed by the addition of the chromogen 3,3’-diaminobenzidine (DAB; 0.02%; Sigma-Aldrich) in 0.1 M PBS to which 0.04% peroxide hydrogen was added before use. After 7–8 min, the tissue sections were rinsed with 0.1 M PBS, mounted on gelatin-coated slides, and dehydrated. Fos-positive neurons were visualized and counted under a 10× objective using a bright-field microscope (Olympus BX-50) equipped with a video camera module (Leica DFC320). The analyzed brain areas were the mPFC, comprising the prelimbic (PrL) and infralimbic (IL) cortices (anterior/posterior coordinates from bregma: 3.24–3.00 mm) and cingulate cortex, areas 1 and 2 (Cg1 and Cg2; 1.44–1.08 mm), basolateral nucleus of the amygdala (BLA), and central nucleus of the amygdala (CeA; −2.40 to −2.64 mm) according to Paxinos and Watson (2007). Dark objects with areas between 10–80 µm² were identified and bilaterally counted using a computerized image analysis system (Image Pro Plus 6.2, Media Cybernetics, Bethesda, Maryland, USA). Nuclei were counted individually and are expressed as the mean number of Fos-positive cells per 0.1 mm² of tissue.

Statistical analysis

The data are expressed as mean+SEM. Two-way ANOVA was used to analyze the duration of the freezing response and number of Fos-positive cells, with condition (same and different context) and treatment (saline and drug) as the factors. Significant comparisons were followed by the Newman-Keuls post-hoc test. Values of p<0.05 were considered statistically significant.

Results

A representative photomicrograph that shows the injection sites in the DH is presented in Figure 1(a). Diagrammatic representations of the vehicle and drugs injection sites in the DH are shown in Figure 1(b). Rats were excluded from the analysis if they had obstructed or loose cannulae, infections or problems with the histology.

Experiment 1

The mean startle amplitudes for rats that received intra-DH saline (n=10) or 8-OH-DPAT (n=14) before the test session are shown in Figure 2(a). The two-way ANOVA revealed a significant effect of condition (F _1,27=10.89, p<0.05) but no effect of treatment (F _1,27=1.94, p>0.05) and no treatment × condition interaction (F _1,27=1.11, p>0.05). The mean startle amplitudes for rats that received intra-DH vehicle (n=15) or ketanserin (n=15) before the test session are shown in Figure 2(b). The two-way ANOVA revealed a significant effect of condition (F _1,28=18.07, p<0.05) but no effect of treatment (F _1,28=1.08, p>0.05) and no treatment × condition interaction (F _1,28=0.21, p>0.05). The mean startle amplitudes for rats that received intra-DH saline (n=12) or muscimol (n=12) before the test session are shown in Figure 2(c). The two-way ANOVA revealed significant effects of condition (F _1,22=4.21, p<0.05) and treatment (F _1,22=3.96, p<0.05) and a significant treatment × condition interaction (F _1,22=11.45, p<0.05).

Experiment 2

The mean durations of freezing exhibited by rats that received intra-DH saline (n=10 and 10, respectively), muscimol (n=10 and 10, respectively), or bicuculline (n=8 and 10, respectively) before reexposure to a different or the same context are illustrated in Figure 3. The two-way ANOVA revealed significant effects of condition (F _2,52=55.93, p<0.05) and treatment (F _2,52=11.34, p<0.05) and a significant treatment × condition interaction (F _2,52=5.65, p<0.05). The Newman-Keuls post-hoc test revealed that intra-DH injections of muscimol in rats that were reexposed to the same conditioning context significantly decreased the freezing response compared with the control group.

Histograms of the mean number of Fos-immunoreactive cells in the mPFC and amygdala in rats that received intra-DH saline or muscimol before reexposure to a different or the same context are presented in Figure 4. The two-way ANOVA revealed significant effects of condition and treatment and a significant condition×treatment interaction in the Cg2 (F _1,23=3.31, 5.54, and 5.54, respectively, p<0.05 in all cases) (n=6–7 per group) and significant effects of condition and treatment in the PrL (F _1,23=7.91 and 6.75, respectively, p<0.05 in both cases) (n=6–7 per group), IL (F _1,23=5.97 and 4.38, respectively, p<0.05 in both cases) (n=6–7 per group), and Cg1 (F _1,23=4.26 and 7.18, respectively, p<0.05 in both cases) (n=6–7 per group). No significant effects were found in the BLA (F _1,19=3.98, 0.38, and 0.23, for condition, treatment, and condition×treatment interaction, respectively, p>0.05 in all cases) (n=5–6 per group) or CeA (F _1,19=0.30, 0.54, and 0.21, for condition, treatment, and condition×treatment interaction, respectively, p>0.05 in all cases) (n=5–6 per group). However, there was a trend for significance in the Fos immunoreactivity in the BLA for the factor condition (F _1,19=3.98, p=0.07). The Newman-Keuls post-hoc test showed that contextual fear conditioning significantly increased Fos protein expression in all of the mPFC subregions, and muscimol treatment decreased the activation of these areas. Representative photomicrographs of Fos immunoreactivity in the PrL and Cg2 are shown in Figure 5.

Discussion

We previously showed that the reduction of 5-HT activity in the median raphe nucleus-DH pathway induced by intra-MRN injections of 8-OH-DPAT caused an anxiolytic-like effect in rats subjected to contextual fear conditioning. In Experiment 1, we sought to determine whether 5-HT_1A or 5-HT₂ receptors in the DH are involved in the expression of contextual fear 24 h after conditioning. We expected that local injections of 8-OH-DPAT into the DH would produce an anxiogenic-like effect and that similar injections of ketanserin would produce the opposite effect (i.e. an anxiolytic-like effect). At least some 5-HT₁ receptors are presynaptic, whereas 5-HT₂ receptors are likely postsynaptic in the DH (Fischette et al., 1987; Mendelson and McEwen, 1991). Surprisingly, intra-DH injections of 8-OH-DPAT and ketanserin 24 h after training did not alter FPS, indicating that, at least at this time point, 5-HT_1A and 5-HT_2A/2C receptors are not involved in the expression of this response. In contrast to these findings, however, we previously demonstrated that similar intra-DH injections of 8-OH-DPAT but not ketanserin reduced FPS and freezing when administered six hours after contextual fear conditioning (Almada et al., 2009). To explain these effects, we propose the existence of an inhibitory mechanism mediated by 5-HT_1A receptors/GABAergic interneurons in the DH. The activation of these 5-HT receptors by 8-OH-DPAT injected into the DH six hours after training has an indirect consequence of activating GABAergic output neurons in the DH to other structures, leading to a reduction of contextual fear responses. Altogether, these findings suggest that this inhibitory mechanism time-dependently operates at the hippocampal level, such that presynaptic 5-HT_1A receptors in the DH may be involved in the initial consolidation of contextual aversive information. This possibility appears reasonable when we consider that these receptors have been shown to be numerous in the DH (Fischette et al., 1987; Mendelson and McEwen, 1991).

In support of the above mentioned possibility, Bevilaqua et al. (1997) demonstrated that intra-hippocampus injections of 8-OH-DPAT produced retrograde amnesia when administered immediately and three hours after training in the one-trial avoidance task but was ineffective when injected after nine hours. Thus, our data are consistent with the hypothesis that a hippocampal cyclic adenosine monophosphate/protein kinase A pathway is involved in memory consolidation at three and six hours (Bevilaqua et al., 1997) and further suggests the involvement of 5-HT_1A receptors in the regulation of this signaling. Moreover, intra-DH injections of 8-OH-DPAT 24 h after context learning were recently shown to have no effect on the contextual freezing response (Chang and Liang, 2012).

The effects of intra-DH muscimol on the expression of FPS and the freezing response 24 h after contextual conditioning were also evaluated in this study. The GABA_A agonist reduced both types of conditioned responses, demonstrating the crucial importance of the DH in the expression of contextual fear (Fanselow, 2000; Maren, 2001). These findings are consistent with previous reports, in which inactivation of the DH with the local anesthetic lidocaine or inhibition of GABAergic neurotransmission produced deficits in learning and the expression of contextual fear conditioning (Daumas et al., 2004; Holt and Maren, 1999; McEown and Treit, 2010; Matus-Amat et al., 2007). Moreover, muscimol was shown to induce anxiolysis when injected into the central nucleus of the amygdala (Amaral et al., 2007; Moreira et al., 2007; Reimer et al., 2008; Rossato et al., 2004). In contrast, rats that received intra-DH injections of the GABA_A antagonist bicuculline did not exhibit significant changes in the duration of the freezing response, indicating that this area is not under tonic GABAergic control, which occurs in lower structures, such as the PAG, in which this neurotransmitter exerts persistent control over the defensive reaction (Brandão et al., 2005).

In the present study, rats subjected to the same context where they received footshocks exhibited a widespread increase in Fos-like immunoreactivity in the mPFC (PrL, IL, Cg1 and Cg2). This mPFC activation pattern in rats subjected to contextual fear was also found in previous studies (Albrechet-Souza et al., 2012; Almada et al., 2009). In fact, contextual fear is considered to model essential features of anxiety, in which the trigger of the response is a set of nonspecific environmental stimuli, and anxiety is known to not be triggered or suppressed by explicit cues (Grillon, 2002). Thus, the significant activation of the mPFC during the expression of contextual fear is consistent with the idea that anxiety involves the recruitment of higher neural levels (McNaughton and Corr, 2004; Mobbs et al., 2007).

Intra-DH injections of muscimol, in addition to reducing FPS and freezing behavior, also decreased mPFC activation. In fact, neurons of the hippocampal CA1 subfield and subiculum have been shown to project directly to the mPFC, particularly the PrL, IL, and medial orbital cortex (Ferino et al., 1987; Jay and Witter, 1991; Swanson, 1981). These projections appear to be unidirectional, ipsilateral and glutamatergic (Jay et al., 1992). Moreover, disconnection studies demonstrated the importance of interactions between the hippocampus and mPFC for memory acquisition and retrieval (Barker and Warburton, 2011). A key direction for future studies would be to further explore the role of discrete mPFC subregions in the acquisition, consolidation and retrieval of conditioned aversive memories.

Among the amygdaloid nuclei, the BLA has been the focus of the majority of studies on fear conditioning. Although not reaching statistical significance, the notable Fos labeling in the BLA induced by contextual fear conditioning 24 h after conditioning suggests that this region is indeed recruited as reported in our previous study (Almada et al., 2009). The activation of the BLA over time is plausible insofar as this structure provides the primary sensory interface for the CS and encodes the aversive memory of learned fear behavior (Goosens et al., 2003; Roozendaal et al., 2009). This specific time window associated with the recruitment of the BLA in the present study is reasonable because this amygdala nucleus appears to be primarily involved in the modulation of memory processed elsewhere (McGaugh and Izquierdo, 2000). The importance of these parameters of the fear conditioning procedure for the study of the neural substrates of fear memory have been emphasized in many previous reports (Beck and Fibiger, 1995; Berlau and McGaugh, 2006; Ferry et al., 1999; Matus-Amat et al., 2007). Fos labeling in the CeA did not change within 24 h after contextual fear conditioning. Importantly, divergent results have been reported when sensory and output nuclei of the amygdala (e.g. CeA) are considered (for review, see Roosendaal et al., 2009).

Data from this study support the hypothesis that the mPFC is necessary for the retrieval of remote memories, suggesting that information stored in this neocortical structure may be necessary for memory recall beginning in the first hours after training (Blum et al., 2006; Zhao et al., 2005). In fact, inhibition of the N-methyl-D-aspartate (NMDA) NR2B receptor subunit in the mPFC has been shown to interrupt the expression of recently acquired memories (Teixeira et al., 2006; Zhao et al., 2005). Moreover, hippocampus-dependent learning appears to be associated with upregulation of immediate early genes in the PrL, and pharmacological interventions targeted at this area can prevent both new learning and the recall of remotely and even recently consolidated information (Teixeira et al., 2006). These findings challenge the concept of the specialization of the hippocampus in rapid-learning systems and neocortex in slow-learning systems (Tse et al., 2011).

In conclusion, we previously demonstrated the involvement of DH serotonergic mechanisms in the expression of contextual fear six hours after conditioning (Almada et al., 2009), and the present study showed that intra-DH injections of 8-OH-DPAT and ketanserin administered 24 h after conditioning did not affect the expression of contextual fear. However, intra-DH injections of muscimol reduced both FPS and the conditioned freezing response. These data suggest that a time-dependent process mediated by distinct neurotransmitters in the DH may participate in the consolidation and expression of contextual fear memories and further indicate the involvement of the mPFC in these processes. Serotonin appears to participate in the earlier phases, whereas GABAergic mechanisms are recruited later in the consolidation of contextual aversive memories. More studies are still needed for further understanding of the consolidation and expression of aversive memories mediated by 5-HT, GABA and other mechanisms at distinct time points after the conditioning.

Footnotes

Conflict of interest

The authors declare no conflict of interest.

Funding

This work was supported by Fundação de Amparo à Pesquisa do Estado de São Paulo (grant no. 11/00041-3) and Conselho Nacional de Desenvolvimento Científico e Tecnológico (grant no. 471325/2011-2). RC Almada holds a Doctor scholarship from Conselho Nacional de Desenvolvimento Científico e Tecnológico (Proc no. 143024/2009-4). L Albrechet-Souza holds a Post-Doctoral fellowship from Conselho Nacional de Desenvolvimento Científico e Tecnológico (Proc no. 159411/2010-6).

References

Albrechet-Souza

Borelli

Almada

. (2011) Midazolam reduces the selective activation of the rhinal cortex by contextual fear stimuli. Behav Brain Res 216: 631–638.

Albrechet-Souza

Carvalho

Brandão

(2012) D1-like receptors in the nucleus accumbens shell regulate the expression of contextual fear conditioning and activity of the anterior cingulate cortex in rats. Int J Neuropsychopharmacol. DOI: 10.1017/S146114571200082X

Almada

Borelli

Albrechet-Souza

. (2009) Serotonergic mechanisms of the median raphe nucleus-dorsal hippocampus in conditioned fear: Output circuit involves the prefrontal cortex and amygdala. Behav Brain Res 203: 279–287.

Amaral

Luft

Cammarota

. (2007) Temporary inactivation of the dorsal hippocampus induces a transient impairment in retrieval of aversive memory. Behav Brain Res 180: 113–118.

Ammassari-Teule

Pavone

Castellano

. (1991) Amygdala and dorsal hippocampus lesions block the effects of GABAergic drugs on memory storage. Brain Res 551: 104–109.

Avanzi

Brandão

(2001) Activation of somatodendritic 5-HT_1A autoreceptors in the median raphe nucleus disrupts the contextual conditioning in rats. Behav Brain Res 126: 175–184.

Avanzi

Castilho

de Andrade

. (1998) Regulation of contextual conditioning by the median raphe nucleus. Brain Res 790: 178–184.

Azmitia

Segal

(1978) An autoradiographic analysis of the differential ascending projections of the dorsal and median raphe nuclei in the rat. J Comp Neurol 179: 641–667.

Barker

GRI

Warburton

(2011) When is the hippocampus involved in recognition memory? J Neurosci 31: 10721–10731.

10.

Beck

Fibiger

(1995) Conditioned fear-induced changes in behavior and in the expression of the immediate early gene c-fos: With and without diazepam pretreatment. J Neurosci 15: 709–720.

11.

Berlau

McGaugh

(2006) Enhancement of extinction memory consolidation: The role of the noradrenergic and GABAergic systems within the basolateral amygdala. Neurobiol Learn Mem 86: 123–132.

12.

Bevilaqua

Ardenghi

Schröder

. (1997) Drugs acting upon the cyclic adenosine monophosphate/protein kinase: A signalling pathway modulate memory consolidation when given late after training into rat hippocampus but not amygdala. Behav Pharmacol 8: 331–338.

13.

Blum

Hebert

Dash

(2006) A role for the prefrontal cortex in recall of recent and remote memories. Neuroreport 17: 341–344.

14.

Borelli

Gárgaro

dos Santos

. (2005) Effects of inactivation of serotonergic neurons of the median raphe nucleus on learning and performance of contextual fear conditioning. Neurosci Lett 387: 105–110.

15.

Bouton

Bolles

(1980) Conditioned fear assessed by freezing and by the suppression of three different baselines. Anim Learn Behav 8: 429–434.

16.

Brandão

Borelli

Nobre

. (2005) Gabargic regulation of the neural organization of fear in the midbrain tectum. Neurosci Biobehav Rev 29: 1299–1311.

17.

Brioni

McGaugh

(1988) Post-training administration of GABAergic antagonists enhances retention of aversively motivated tasks. Psychopharmacology (Berl) 96: 505–510.

18.

Chang

Liang

(2012) Roles of hippocampal GABA_A and muscarinic receptors in consolidation of context memory and context-shock association in contextual fear conditioning: A double dissociation study. Neurobiol Learn Mem 98: 17–24.

19.

Davis

Falls

Campeau

. (1993) Fear-potentiated startle: A neural and pharmacological analysis. Behav Brain Res 58: 175–198.

20.

Daumas

Halley

Francés

. (2004) Encoding, consolidation, and retrieval of contextual memory: Differential involvement of dorsal CA3 and CA1 hippocampal subregions. Learn Mem 12: 375–382.

21.

Deakin

Graeff

(1991) 5-HT and mechanisms of defence. J Psychopharmacol 5: 305–315.

22.

Fanselow

(1984) What is conditioned fear? Trends Neurosci 7: 460–462.

23.

Fanselow

(2000) Contextual fear, gestalt memories, and the hippocampus. Behav Brain Res 110: 73–81.

24.

Ferino

Thierry

Glowinski

(1987) Anatomical and electrophysiological evidence for a direct projection from Ammon’s horn to the medial prefrontal cortex in the rat. Exp Brain Res 65: 421–426.

25.

Ferry

Roozendaal

McGaugh

(1999) Basolateral amygdala noradrenergic influences on memory storage are mediated by an interaction between β- and α₁-adrenoceptors. J Neurosci 19: 5119–5123.

26.

Fischette

Nock

Renner

(1987) Effects of 5,7-dihydroxytryptamine on serotonin1 and sertotonin2 receptors throughout the rat central nervous system using quantitative autoradiography. Brain Res 421: 263–279.

27.

Goosens

(2011) Hippocampal regulation of aversive memories. Curr Opin Neurobiol 21: 460–466.

28.

Goosens

Hobin

Maren

(2003) Auditory-evoked spike firing in the lateral amygdala and Pavlovian fear conditioning: Mnemonic coding of fear bias? Neuron 40: 1013–1022.

29.

Graeff

Rawlins

(1980) Dorsal periaqueductal gray punishment, septal lesions and the mode of action of minor tranquilizers. Pharmacol Biochem Behavior 12: 41–45.

30.

Graeff

Schoenfeld

(1970) Tryptaminergic mechanisms in punished and nonpunished behavior. J Pharmacol Exp Ther 173:277–283

31.

Graeff

Guimarães

De Andrade

. (1996) Role of 5-HT in stress, anxiety, and depression. Pharmacol Biochem Behav 54: 129–141.

32.

Grillon

(2002) Startle reactivity and anxiety disorders: Aversive conditioning, context, and neurobiology. Biol Psychiatry 52: 958–975.

33.

Gross

Canteras

(2012) The many paths to fear. Nat Rev Neurosci 13: 651–658.

34.

Holt

Maren

(1999) Muscimol inactivation of the dorsal hippocampus impairs contextual retrieval of fear memory. J Neurosci 19: 9054–9062.

35.

Jay

Witter

(1991) Distribution of hippocampal CA1 and subicular efferents in the prefrontal cortex of the rat studied by means of anterograde transport of Phaseolus vulgaris-leucoagglutinin. J Comp Neurol 313: 574–586.

36.

Jay

Thierry

Wiklund

. (1992) Excitatory amino-acid pathway from the hippocampus to the prefrontal cortex: Contribution of AMPA receptors in hippocampo-prefrontal cortex transmission. Eur J Neurosci 4: 1285–1295.

37.

Katsurabayashi

Kubota

Tokutomi

. (2003) A distinct distribution of functional presynaptic 5-HT receptor subtypes on GABAergic nerve terminals projecting to single hippocampal CA1 pyramidal neurons. Neuropharmacology 44: 1022–1030.

38.

Kim

Fanselow

(1992) Modality-specific retrograde amnesia of fear. Science 256: 675–677.

39.

Kiser

Lebovitz

(1975) Monoaminergic mechanisms in aversive brain stimulation. Physiol Behav 15: 47–53.

40.

Kiser

Brown

Sanghera

. (1980) Dorsal raphe nucleus stimulation reduces centrally-elicited fearlike behavior. Brain Res 191: 265–272.

41.

McEown

Treit

(2010) Inactivation of the dorsal or ventral hippocampus with muscimol differentially affects fear and memory. Brain Res 1353: 145–151.

42.

McGaugh

Izquierdo

(2000) The contribution of pharmacology to research on the mechanisms of memory formation. Trends Pharmacol Sci 21: 208–210.

43.

McNaughton

Corr

(2004) A two-dimensional neuropsychology of defense: Fear/anxiety and defensive distance. Neurosci Biobehav Rev 28: 285–305.

44.

Maren

(2001) Neurobiology of Pavlovian fear conditioning. Annu Rev Neurosci 24: 897–931.

45.

Martinez

RCB

De Oliveira

Brandão

(2006) Conditioned and unconditioned fear organized in the periaqueductal gray are differentially sensitive to injections of muscimol. Neurobiol Learn Mem 85: 58–65.

46.

Matus-Amat

Higgins

Sprunger

. (2007) The role of dorsal hippocampus and basolateral amygdala NMDA receptors in the acquisition and retrieval of context and contextual fear memories. Behav Neurosci 121: 721–731.

47.

Mendelson

McEwen

(1991) Autoradiographic analyses of the effects of the restraint-induced stress on 5-HT1A, 5-HT1c and 5-HT2 receptors in the dorsal hippocampus of male and female rats. Neuroendocrinology 54: 454–461.

48.

Milanovic

Radulovic

Laban

. (1998) Production of the Fos protein after contextual fear conditioning of C57BL/6N mice. Brain Res 784: 37–47.

49.

Mobbs

Petrovic

Marchant

. (2007) When fear is near: Threat imminence elicits prefrontal-periaqueductal gray shifts in humans. Science 317: 1079–1083.

50.

Moreira

Masson

Carvalho

. (2007) Exploratory behaviour of rats in the elevated plus-maze is differentially sensitive to inactivation of the basolateral and central amygdaloid nuclei. Brain Res Bull 71: 466–474.

51.

Paxinos

Watson

(2007) The rat brain in stereotaxic coordinates, 6th edition. San Diego: Academic Press.

52.

Phillips

LeDoux

(1992) Differential contribution of amygdala and hippocampus to cued and contextual fear conditioning. Behav Neurosci 106: 274–285.

53.

Reimer

Oliveira

Brandão

(2008) Selective involvement of GABAergic mechanisms of the dorsal periaqueductal gray and inferior colliculus on the memory of the contextual fear as assessed by the fear potentiated startle test. Brain Res Bull 76: 545–550.

54.

Roozendaal

McEwen

Chattarji

(2009) Stress, memory and the amygdala. Nature Reviews 10: 423–433.

55.

Rossato

Bonini

Coitinho

. (2004) Retrograde amnesia induced by drugs acting on different molecular systems. Behav Neurosci 118: 563–568.

56.

Santos

Gárgaro

Oliveira

. (2005) Pharmacological dissociation of moderate and high contextual fear as assessed by freezing behavior and fear-potentiated startle. Eur Neuropsychopharmacol 15: 239–246.

57.

Schenberg

Graeff

(1978) Role of the periaqueductal gray substance in the antianxiety action of benzodiazepines. Pharmacol Biochem Behav 9: 287–295.

58.

Silva

RCB

Cruz

APM

Avanzi

. (2002) Distinct contributions of median raphe nucleus to contextual fear conditioning and fear-potentiated startle. Neural Plast 9: 233–247.

59.

Silva

Gárgaro

Brandão

(2004) Differential regulation of the expression of contextual freezing and fear-potentiated startle by 5-HT mechanisms of the median raphe nucleus. Behav Brain Res 151: 93–101.

60.

Strekalova

Zörner

Zacher

. (2003) Memory retrieval after contextual fear conditioning induces c-Fos and JunB expression in CA1 hippocampus. Genes Brain Behav 2: 3–10.

61.

Swanson

(1981) A direct projection from Ammon’s horn to prefrontal cortex in the rat. Brain Res 217: 150–154.

62.

Teixeira

Pomedli

Maei

. (2006) Involvement of the anterior cingulate cortex in the expression of remote spatial memory. J Neurosci 26: 7555–7564.

63.

Tse

Takeuchi

Kakeyama

. (2011) Schema-dependent gene activation and memory encoding in neocortex. Science 333: 891–895.

64.

Weitemier

Ryabinin

(2004) Subregion-specific differences in hippocampal activity between Delay and Trace fear conditioning: An immunohistochemical analysis. Brain Res 995: 55–65.

65.

Winocur

Rawlins

Gray

(1987) The hippocampus and conditioning to contextual cues. Behav Neurosci 101: 617–625.

66.

Wise

Berger

Stein

(1972) Benzodiazepines: Anxiety-reducing activity by reduction of serotonin turnover in the brain. Science 177: 180–183.

67.

Zhao

Toyoda

Lee

. (2005) Roles of NMDA NR2B subtype receptor in prefrontal long-term potentiation and contextual fear memory. Neuron 47: 859–872.