Carbachol injections into the nucleus accumbens disrupt acquisition and expression of fear-potentiated startle and freezing in rats

Carbachol injections into the nucleus accumbens disrupt acquisition and expression of fear-potentiated startle and freezing in rats

Neuroscience 140 (2006) 769 –778 CARBACHOL INJECTIONS INTO THE NUCLEUS ACCUMBENS DISRUPT ACQUISITION AND EXPRESSION OF FEAR-POTENTIATED STARTLE AND F...

483KB Sizes 0 Downloads 69 Views

Neuroscience 140 (2006) 769 –778

CARBACHOL INJECTIONS INTO THE NUCLEUS ACCUMBENS DISRUPT ACQUISITION AND EXPRESSION OF FEAR-POTENTIATED STARTLE AND FREEZING IN RATS I. SCHWIENBACHER,a1 H.-U. SCHNITZLER,a R. F. WESTBROOK,b R. RICHARDSONb AND M. FENDTa2*

The nucleus accumbens (NAC) has been shown to be involved in both appetitive learning and behavior (Cardinal et al., 2002; Wise and Rompre, 1989), and in fear learning (Haralambous and Westbrook, 1999; Jongen-Relo et al., 2003; Parkinson et al., 1999; Salamone, 1994). This latter finding has been supported by a recent study from our laboratory, which showed that temporary inactivation of the NAC, by local infusion of tetrodotoxin (TTX), blocked both the acquisition and the expression of conditioned fear, as measured by fear-potentiated startle (FPS) (Schwienbacher et al., 2004). It is possible that this effect of temporary inactivation of the NAC on acquisition and expression of conditioned fear is due to alterations in a neurotransmitter system involved in reward (Olds, 1958; Konorski, 1967; Dickinson and Pearce, 1977; Dickinson and Dearing, 1979; Grauer and Thomas, 1982; Brandao et al., 1991; Anderson et al., 1995; Lang, 1995; Schmid et al., 1995). A first plausible candidate system would be the dopaminergic system, given that a number of studies have demonstrated that intra-NAC dopamine (DA) is one of the key elements in reward-related processes (summarized in Wise and Rompre, 1989). Surprisingly, intra-NAC infusions of amphetamine, a DA agonist, do not appear to affect either the acquisition or the expression of Pavlovian conditioned fear, as measured by FPS (Schwienbacher et al., 2005); similar results have also been reported by Josselyn et al. (2005). Taken together, these two studies clearly indicate that intra-NAC DA is not critically involved in conditioned fear. A second, possible neurotransmitter system that could mediate the disruption of acquisition and expression of conditioned fear caused by the temporary inactivation of the NAC (Schwienbacher et al., 2004) is acetylcholine. That is, several studies have suggested that intra-NAC acetylcholine is involved in approach behavior and other reward-related processes. For example, Ikemoto and colleagues (1998) have shown that rats self-administer the unselective cholinergic agonist carbachol directly into the NAC. Further, intra-NAC infusions of the muscarinic cholinergic antagonist atropine disrupt acquisition and expression of an operant approach response for food (Ikemoto and Panksepp, 1996; Pratt and Kelley, 2004). In addition, a recent study from our laboratory demonstrated that intraNAC infusions of carbachol elicit conditioned place preference and 50 kHz ultrasonic vocalizations, which are thought to reflect activation of an appetitive motivational system, in rats (Fendt, et al., in press). Finally, some studies have suggested that acetylcholine within the NAC may modulate aversive processes, such as active avoid-

a Tierphysiologie, Universität Tübingen, Auf der Morgenstelle 28, D-72076 Tübingen, Germany b School of Psychology, University of New South Wales, Sydney, Australia 2052

Abstract—The nucleus accumbens is involved in different types of emotional learning, ranging from appetitive instrumental learning to Pavlovian fear conditioning. In previous studies, we found that temporary inactivation of the nucleus accumbens blocked both the acquisition and expression of conditioned fear. This was not due to altered dopaminergic activity as we have also found that intra-nucleus accumbens infusions of the dopamine agonist amphetamine do not affect either the acquisition or the expression of conditioned fear. Therefore, in the present study we examined whether cholinergic activity in the nucleus accumbens is involved in the acquisition and expression of conditioned fear. Specifically, the effect of intra-nucleus accumbens infusions of the unselective cholinergic agonist carbachol on the acquisition and expression of conditioned fear was assessed. Across several experiments, we measured fear to visual and acoustic conditioned stimuli and to the experimental context. Further, two different measures of conditioned fear were recorded: fear potentiation of startle and freezing. Intra-nucleus accumbens carbachol infusions disrupted acquisition as well as expression of conditioned fear, regardless of the modality of the fear-eliciting stimulus or of the specific measure of conditioned fear. This disruption of conditioned fear was not simply a by-product of enhanced motor activity which also occurred after intra-nucleus accumbens carbachol infusions. Interestingly, despite the substantial effect of intra-nucleus accumbens carbachol on expression of conditioned fear, the results of the final experiment suggest that these rats extinguish similarly to control rats. Taken together, the present results indicate that acetylcholine within the nucleus accumbens is important for the learning and retrieval of conditioned fear. © 2006 Published by Elsevier Ltd on behalf of IBRO. Key words: acetylcholine, dopamine, reward, learned fear, extinction. 1

Present address: Boehringer Ingelheim Pharma GmbH & Co. KG, Birkendorfer Strasse 65, D-88397 Biberach/Riss, Germany. 2 Present address: Novartis Pharma AG, Neuroscience Research, WSJ-386.3.44, CH-4002 Basel, Switzerland. *Corresponding author. Tel: ⫹41-61-3241042; fax: ⫹41-61-3244502. E-mail address: [email protected] (M. Fendt). Abbreviations: ANOVA, analysis of variance; ASR, acoustic startle response; CS, conditioned stimulus; DA, dopamine; FPS, fear-potentiated startle; ISI, interstimulus interval; ITI, intertrial interval; NAC, nucleus accumbens; TTX, tetrodotoxin; US, unconditioned stimulus. 0306-4522/06$30.00⫹0.00 © 2006 Published by Elsevier Ltd on behalf of IBRO. doi:10.1016/j.neuroscience.2006.02.052

769

770

I. Schwienbacher et al. / Neuroscience 140 (2006) 769 –778

ance learning and stress (Mark et al., 1995; Pfister et al., 1994; Rada and Hoebel, 2001). Based on the evidence reviewed above, the present study examined whether intra-NAC infusions of the unselective acetylcholine receptor agonist carbachol affect acquisition/expression of Pavlovian conditioned fear. First, we measured the effects of carbachol-injections into the NAC on acquisition and expression of FPS using a discrete visual conditioned stimulus (CS; a light). Second, we measured the effects of intra-NAC carbachol on acquisition and expression of conditioned freezing to an acoustic CS as well as to the training context. As a part of these experiments on freezing, we also examined the effects of carbachol on the extinction learning produced by testing the CS and training context in the absence of the US. Third, since intra-NAC carbachol infusions also enhance motor activity, we tested whether the measured changes in fear expression were a by-product of the enhanced motor activity.

EXPERIMENTAL PROCEDURES Experiments 1A, 1B, 3A, and 3B were done at the University of Tübingen, Germany. Experiments 2A, 2B, and 2C were done at the University of New South Wales, Sydney, Australia.

Subjects Tübingen. A total of 89 male Sprague–Dawley rats (Charles River, Sulzfeld, Germany), weighing between 220 and 280 g at the time of surgery, were used in experiments 1A and 1B; 83 rats were used in experiments 3A and 3B. All rats were housed in groups of five to six rats/cage in a temperature-controlled colony room with a 12-h light/dark cycle. They were fed 14 g standard rat chow/ animal/day, and tap water was available ad libitum. Each rat was handled daily before and after surgery, with handling starting 7 days before surgery. The experiments were performed in accordance with the U.S. National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the local council of animal care (Regierungspräsidium Tübingen, ZP 4/02). All attempts were made to minimize the number of animals used in this study and their suffering. Sydney. A total of 46 male Wistar rats (Gore Hill, Sydney, AU), weighing between 210 and 260 g at the time of surgery, were used in experiments 2A, 2B, and 2C. Rats were housed in groups of seven to eight rats/cage. Food and water were available ad libitum. Each rat was handled daily before and after surgery, with handling starting 7 days before surgery. The experiments were performed in accordance with international ethical guidelines for the care and use of experimental animals, and all procedures were approved by the Animal Care and Ethics Committee at the University of New South Wales.

stylets (diameter 0.4 mm) in order to maintain patency. The behavioral tests commenced following full recovery (5– 6 days post surgery).

Experiments 1A, 1B, 3A, and 3B (Tübingen) Fear conditioning. Rats were trained in two identical, unilluminated boxes (38 cm⫻60 cm⫻28 cm). The CS was produced by turning on a 15 W white light bulb located on the ceiling of the box. The unconditioned stimulus (US) was a 0.6 mA foot shock produced by a custom-made shock generator, and delivered through a floor made of stainless steel grids (diameter 5 mm) spaced approximately 20 mm apart. Animals were placed into the conditioning boxes and allowed to adapt for 5 min. After this, rats received either five or 10 CS–US pairings (see behavioral procedures below), with a mean intertrial interval (ITI) of two minutes (range: 1–3 min). The US was presented during the last 0.5 s of the 3.7 s light CS. Measurement of startle response and locomotor activity. The magnitude of the acoustic startle response (ASR) was measured in two identical sound-attenuating test chambers (100 cm⫻ 80 cm⫻60 cm). These chambers contained custom-made wire mesh test cages (20 cm⫻10 cm⫻12 cm) with steel floors that were placed onto piezoelectric accelerometers. Movements of the rats within the cages caused changes of the accelerometer voltage output; these changes were amplified, digitized, and analyzed by a PC. The ASR magnitude was computed as the difference of the peak-to-peak voltage output of the accelerometer 80 ms before and 80 ms after the onset of the acoustic startle stimulus (100 dB SPL, 10 kHz tone pulse, 20 ms duration including 0.4 ms rise and fall times). The acoustic startle stimulus was generated by a function-synthesizer (Hortmann, Neckartenzlingen, Germany) and delivered through loudspeakers mounted 40 cm away from the test cages. A continuous background white noise (55 dB SPL) was present throughout the test session. All noise intensity measurements were done with a 0.5 inch condenser microphone and a measuring amplifier (Brüel and Kjaer Type 2606, Copenhagen, Denmark) after bandpass (0.25– 80 kHz) filtering. In addition to the ASR, spontaneous motor activity was also measured in the test cages. Spontaneous motor activity was calculated as the root mean square value of the accelerometer output, measured in a 28 s time window between startle stimulus presentations. Test of FPS. Rats were placed in the wire mesh test cages, and given a 5 min adaptation period. Following this, 10 initial startle stimuli (30s interstimulus interval, ISI) were presented in order to induce a stable baseline ASR. Then, 20 test trials occurred. On 10 test trials the startle stimulus was presented 3.2 s after onset of the light CS (light-tone trials), and on the other 10 test trials the startle stimulus was presented by itself. Test trial type occurred in a pseudo-randomized order (30 s ISI). The tonealone trials provide a measure of baseline startle, and the difference between the tone-alone and the light-tone trials was used as an operational measure of conditioned fear (e.g. Brown et al., 1951; Davis and Astrachan, 1978; Fendt and Fanselow, 1999).

Surgery Rats were anesthetized with ketamine/xylazine (9:1, 100 mg/kg i.p.) and placed into a stereotaxic frame. The skull was exposed and stainless steel guide cannulas (Tübingen: 0.7 mm diameter, custom made; Sydney: 22-gauge, Plastic One Inc., Roanoke, VA, USA) were implanted bilaterally aiming at the NAC (toothbar ⫹5 mm above the interaural plane; distance from Bregma: rostrocaudal ⫹3.4 mm, mediolateral ⫾1.5 mm, dorsoventral ⫺7.2 mm; coordinates from Pellegrino et al. (1979)). The guide cannulas were fixed to the skull with acrylic cement (Kulzer, Wehrheim, Germany, and Vertex, Sydney, Australia, respectively) and three anchoring screws. After surgery, the cannulas were fitted with

Experiments 2A, 2B, and 2C (Sydney) Fear conditioning. Two identical cages (20 cm⫻21 cm⫻23 cm) with a grid floor (stainless steel grids, 2 mm in diameter spaced 10 mm apart, center to center) were used for training. The side and front walls, as well as the lid, of the cages were constructed of clear Perspex. The back walls of each cage were made of stainless steel. Each cage was located in separate compartments of a wood cabinet; this cabinet did not have doors in order to permit observation of the rats. Each cage was placed above a tray of paper bedding (Fibercycle, Mudgeeraba, Australia) that was changed between rats, and the chambers also were wiped with

I. Schwienbacher et al. / Neuroscience 140 (2006) 769 –778 1% acetic acid between rats. A discrete auditory CS, consisting of an 81 dB clicker (10 s duration, 10 Hz spike; rise time ⬍10 ␮s, decay time 250 ␮s), delivered from a speaker located on the ceiling of the experimental room was used in each of these experiments. A continuous background white noise (69 dB SPL) was present throughout the test session. All noise intensities were measured with a sound level meter (A scale; type 2235, BrüelKjaer Instruments, Marlbourough, MA, USA) with the microphone placed in the center of each cage. The shock used in training was 1.0 s in duration and 0.8 mA in intensity (unscrambled AC 50 Hz; custom-made, constant current generator), and was administered during the last second of the CS. After an initial 2-minute adaptation period, rats received two clicker-shock pairings, with a 30 s ITI. Rats were removed from the conditioning context 30 s after the second clicker-shock pairing. Test for context freezing. One or two days after conditioning, rats were returned to the conditioning context for a 10 min period. No shocks were administered during this time, and the rat’s behavior was videotaped for later scoring of freezing. Test for CS-elicited freezing. To provide a measure of conditioning to the CS independent of conditioning to the context where the CS–US pairings occurred, CS tests occurred in a different (novel) context. This context consisted of a second set of two cages (16 cm⫻40 cm⫻26 cm). The front of each cage was made of clear Perspex. The walls and the floor were made of white plastic and the top consisted of a stainless steel grid. These chambers were wiped with a 1% vanilla solution between each rat. These cages were located on the top of the wood cabinet that housed the training cages. For the CS test, rats were placed into the novel context, and after a 2 min adaptation period, the clicker was repeatedly turned on (for 50 s) and off (for 10 s) for 10 min. No shocks were presented during this test, and the rat’s behavior was videotaped for later scoring of freezing during the clicker CS. Scoring of freezing. Freezing was defined as the absence of all movements, except of those related to breathing. Rats were observed every two seconds, and their behavior was scored as either freezing or not. A percentage score was then calculated for the total time each rat spent freezing across the total observation period.

Drugs Carbachol (carbamylcholine chloride 99%, Acros, Geel, Belgium; 1, 2, 4 ␮g) was dissolved in 0.5 ␮l saline and was administered intra-accumbally with a velocity of 0.5 ␮l/min. The injection cannulas (Tübingen: 0.4 mm diameter, custom-made; Sydney: 28gauge, Plastic One Inc.) were left in place for an additional 2 min to allow diffusion of the solution away from the cannulas. The animals tested in experiment 3A and B received a combined infusion of (1) either saline or 5 ␮g haloperidol (D2 DA receptor antagonist; Janssen-Cilag, Neuss, Germany) and (2) either saline or 4 ␮g carbachol. The infusion volume in each case was 0.5 ␮l and a 1 min interval separated the two infusions.

Behavioral procedures Experiments 1A and 1B: effects of intra-NAC carbachol on acquisition and expression of conditioned FPS to a light CS. Acquisition of fear. Groups of rats received either saline (n⫽15), 1 ␮g (n⫽17), 2 ␮g (n⫽18), or 4 ␮g (n⫽19) carbachol into the NAC immediately before fear conditioning in experiment 1A. Two days after conditioning, rats were tested for FPS in a drug-free state. Expression of fear. A group of 18 rats was fear conditioned without any injections in experiment 1B. Rats were then tested for FPS on each of the following four days (receiving saline, 1 ␮g, 2 ␮g, or 4 ␮g of carbachol; test order determined by a Latin-

771

square design). To reduce potential effects of extinction on later test days, rats were given an additional five CS–US pairings on each test day (4 h prior to testing). Experiments 2A, 2B, and 2C: effects of intra-NAC carbachol on acquisition and expression of conditioned freezing to a clicker CS and to a context. Acquisition of fear. Eleven rats received saline and eight rats received 4 ␮g carbachol (the most effective dose in experiments 1A and 1B) into the NAC immediately before fear conditioning in experiment 2A. Two days after conditioning, rats were tested for freezing to the training context and to the clicker (in a novel cage) in a drug-free state. Test sessions were counterbalanced for order (i.e. half tested to clicker first, half tested to context first), and separated by a few minutes. Expression of fear. In experiment 2B, the 11 saline-infused rats from experiment 2A were retrained and four additional naive rats were also trained in a drug-free state. The next day, eight rats received saline and seven rats received 4 ␮g carbachol into the NAC immediately before testing (two naive rats in each condition). Each rat was tested to the clicker and the context in the same manner as described above. Effects of carbachol on expression of conditioned freezing and on extinction of conditioned fear: This experiment was intended to replicate the findings of experiment 2B, but with two groups of naive rats. Further, this experiment also examined what effect intra-NAC carbachol had on subsequent responding to the fear-eliciting CS (i.e. whether it affected extinction learning). In experiment 2C, two groups of experimentally naive rats were trained and tested with the same general procedures as were used in experiment 2A. On the day following conditioning, 11 rats were infused with saline and 12 were infused with 4 ␮g carbachol into the NAC immediately before testing. Only the clicker CS was tested in this experiment. On the day following test, all rats were tested for freezing to the clicker CS in a drug-free test (i.e. in order to assess any extinction that may have resulted from the first test). Experiments 3A and 3B: effects of co-administration of carbachol and haloperidol into the NAC on acquisition and expression of conditioned FPS to a light CS. The intention of this experiment was to show that the effects of intra-NAC carbachol infusions seen in experiments 1 and 2 were not a by-product of enhanced motor activity induced by intra-NAC carbachol. Coadministration of the D2 DA receptor antagonist haloperidol should block the motor-enhancing effects of intra-NAC carbachol (cf. Boye et al., 2001) but not the fear-blocking effects because these latter effects are independent of DA (cf. Schwienbacher et al., 2005). Acquisition of fear. Groups of rats received either saline/ saline (n⫽14), haloperidol/saline (n⫽16), saline/carbachol (n⫽14), or haloperidol/carbachol (n⫽16) into the NAC immediately before fear conditioning. Two days after conditioning, rats were tested for FPS in a drug-free state. Expression of fear. A group of 14 rats was fear conditioned without any injections. Rats were then tested for FPS on each of the following four days (receiving saline/saline, haloperidol/saline, saline/carbachol, or haloperidol/carbachol; test order determined by a Latin-square design). To reduce potential effects of extinction on later test days, rats were given an additional five CS–US pairings on each test day (4 h prior to testing).

Histology After test, rats were killed with an overdose of pentobarbital (Nembutal). Their brains were removed and immersion-fixed in 8% paraformaldehyde and 20% sucrose. Frontal sections (50 ␮m) were cut on a freezing microtome and Nissl stained with thionine. The injection sites were localized and the extent of tissue lesions due to cannulation was examined under a light microscope. The

772

I. Schwienbacher et al. / Neuroscience 140 (2006) 769 –778

injection sites were drawn onto plates taken from a rat brain atlas (Paxinos and Watson, 1997).

Statistical analyses For most statistical analyses, multi-factorial analysis of variance (ANOVA), with subsequent pairwise comparisons being made with either the Tukey procedure or with t-tests, was used. In some cases, the Kruskal-Wallis procedure was used for non-parametric statistical analysis. A P⬍0.05 was considered to be statistically significant.

RESULTS Experiments 1A and 1B: effects of intra-NAC carbachol on acquisition and expression of conditioned FPS to a light CS Histology. All injection sites from rats that were included in the statistical analysis (n⫽73) were bilaterally localized in the core region of the NAC, including a few sites located at the border to the medial shell region of the NAC (Fig. 1A). Fourteen further injection sites were bilaterally located in neighboring areas of the NAC, mainly in the dorsal striatum, and these rats were included in the “misplaced injection” group. The remaining animals (n⫽2) had either only unilateral injection sites within the NAC or had suffered substantial damage to their NAC by the injection cannulae, and were therefore excluded from further analysis. Experiment 1A: acquisition of FPS. As can be seen in Fig. 2A, rats given intra-NAC infusions of carbachol prior to training failed to exhibit an increased amplitude ASR on light-noise trials at test (i.e. they did not exhibit FPS;

n’s⫽16/15/14 for the different carbachol concentrations). In contrast, those rats given intra-NAC infusions of saline prior to conditioning exhibited an increased amplitude ASR on light-noise trials (i.e. they exhibited FPS; n⫽12). Statistical analysis confirmed these interpretations of the data. Specifically, a multi-factorial ANOVA, with trial type (tone alone, light-tone) as a within-subject factor and treatment (carbachol concentration) as a between-subject factor, yielded a significant interaction between trial type and treatment [F(3, 53)⫽4.86; P⫽0.005]. The main effects of trial type and treatment were both non-significant (both Fs⬍1.0) Post hoc Tukey-tests of the difference scores revealed rats receiving any dose of carbachol exhibited significantly less FPS than the saline-infused rats. Experiment 1B: expression of conditioned FPS. IntraNAC infusions of carbachol prior to test markedly reduced expression of conditioned fear (see Fig. 2B; n⫽14). Statistical analysis confirmed these interpretations of the data. Specifically, a repeated-measures ANOVA, with trial type and treatment both as within-subject factors, yielded a significant main effect of trial type [F(1, 13)⫽8.72; P⬍ 0.011], of treatment [F(3, 39)⫽12.55; P⬍0.001], and a significant interaction of trial type and treatment [F(3, 39)⫽7.83; P⬍0.001]. Subsequent pairwise comparisons indicated that rats receiving any dose of carbachol exhibited significantly reduced FPS compared with the salineinfused controls. Because the statistical analysis also revealed an effect of pre-test intra-NAC carbachol on baseline startle amplitude [F(3, 39)⫽12.55; P⬍0.001] we also calculated a percent fear-potentiation startle effect (which takes baseline into account; see inset in Fig. 2B) and found

Fig. 1. Serial drawings of frontal sections of the NAC depicting injection sites of carbachol injection within the NAC for experiment 1A and 1B (FPS, panel A), experiment 2A, 2B and 2C (freezing, panel B) and experiment 3A and 3B (FPS, panel C). Abbreviations: AcbC, nucleus accumbens core; AcbSh, nucleus accumbens shell.

I. Schwienbacher et al. / Neuroscience 140 (2006) 769 –778

773

Fig. 2. Effects of carbachol injections into the NAC on fear-potentiation of the ASR. Bar diagrams show the mean ASR magnitude in the absence (black bars) or presence (white bars) of the light-CS. Hatched bars represent the difference between tone-alone and light-tone trials which was taken as an index of fear. Carbachol injected into the NAC immediately before training reduced the acquisition of fear significantly (panel A) and also blocked the expression of FPS when injected immediately before the test (panel B: FPS and inlay: percent FPS) (post hoc Tukey-test after ANOVAs: * P⬍0.05, ** P⬍0.01, *** P⬍0.001).

exactly the same pattern of results. Finally, locomotor activity during the test session was increased by pre-test intra-NAC infusions of carbachol [F(3, 39)⫽12.31; P⬍0.001]. All doses of carbachol had this effect (data not shown). Experiments 2A, 2B, and 2C: effects of intra-NAC carbachol on acquisition and expression of conditioned freezing to a context and to a clicker CS Histology. All injection sites were localized in the core region of the NAC (Fig. 1B). Experiment 2A: acquisition of fear to a clicker CS and to a context. Pre-training injections of carbachol into the NAC completely disrupted fear conditioning to the clicker CS [t(17)⫽7.91, P⬍0.001; Fig. 3A] as well as to the conditioning context [t(17)⫽5.20, P⬍0.001; Fig. 3C] when measured by freezing (n⫽11 for saline, n⫽8 for carbachol). Experiment 2B: expression of conditioned freezing to a clicker CS and to a context. Pre-test infusions of carbachol into the NAC, in rats trained in a drug-free state, completely disrupted expression of learned fear to both the clicker CS [t(13)⫽7.44, P⬍0.001; Fig. 3B] and the conditioning context [t(13)⫽5.57, P⬍0.001; Fig. 3D] when measured by freezing (n’s⫽8 and 7, for the saline and carbachol groups, respectively). Experiment 2C: effects of intra-NAC carbachol on expression of conditioned fear and on subsequent extinction. The results of this experiment, with two groups of experimentally naive rats, replicated those of experiment 2B (see Fig. 4; n⫽11 for saline, n⫽12 for carbachol). That is, pre-test infusions of 4 ␮g of carbachol into the NAC completely blocked expression of learned fear. Interestingly, however, this disruption in the expression of conditioned fear did not affect performance on the test the following day (in a drug-free state). The two groups exhibited a comparable level of extinction on this second test. Statistical analysis supported these interpretations of the data. Specifically, a mixed-design ANOVA, with group as a be-

tween-groups factor and test day as a within-subject factor, yielded a significant main effect of group [F(1, 21)⫽35.88, P⬍0.001] and a significant interaction of group and test day [F(1, 21)⫽21.57, P⬍0.001]. The main effect of test day was not significant [F(1, 21)⫽1.35, P⫽0.26). The significant interaction was due to the two groups differing on the first test day, when the drug was present [t(21)⫽6.45, P⬍0.001], but not on the second test day, when the drug was absent [t(21)⫽.58, P⫽0.57]. A post hoc, dependentsample t-test revealed that the saline-infused rats did indeed exhibit some extinction of learned fear across days (i.e. their performance on the second test was less than their performance on the first; t(10)⫽3.11, P⫽0.011). Critically, the amount of extinction learning accruing from the initial (non-shocked) test session was unaffected by whether that session had occurred under an infusion of saline or carbachol. Experiments 3A, and 3B: effects of co-administration of carbachol and haloperidol into the NAC on acquisition and expression of conditioned FPS to a light CS Histology. Fig. 1C shows the injection sites for those rats included in the statistical analysis in this experiment; all were located within the core region of the NAC. Seven rats in the acquisition experiment and one rat in the expression experiment had misplaced injections (mainly into the dorsal striatum) and were not included in the statistical analysis. Experiment 3A: acquisition of fear. Conditioned fear (i.e. FPS) was present under control conditions (saline/ saline; n⫽13), indicated by a significant effect of trial type (tone alone vs. light tone, paired t-test: t(12)⫽5.99, P⬍0.001, Fig. 5A). Subsequent analysis of the test data used ANOVA and post hoc Tukey-tests. The ANOVA revealed a significant group difference in the acquisition of conditioned fear [F(3,49)⫽17.38, P⬍0.001]. Haloperidol

774

I. Schwienbacher et al. / Neuroscience 140 (2006) 769 –778

Fig. 3. Effects of carbachol injections into the NAC on freezing to a cue (clicker) and a context. Bar diagrams show the mean percent of freezing with saline (black bars) and carbachol (hatched bars). Carbachol injected into the NAC blocked the acquisition (panel A) and expression (panel B) of conditioned freezing to a cue (t-test: *** P⬍0.001) as well as the acquisition (panel C) and expression (panel D) of conditioned freezing to a context.

did not affect test performance (comparison between saline/saline vs. haloperidol/saline (n⫽14), Tukey-test: P⫽ 0.99), whereas carbachol injections disrupted test performance comparison between saline/saline vs. saline/carba-

Fig. 4. Effects of carbachol injections into the NAC on freezing during an extinction training (test 1) and during a retention test on extinction (test 2, without treatment). Freezing during the retention test was totally blocked by intra-NAC carbachol infusions (*** P⬍0.001) but there were no difference in the freezing response in the retention test. Inset: percent FPS.

chol (n⫽12), Tukey-test: P⬍0.001, replicating the results of experiments 1A and 2A. Co-administration of haloperidol did not block the carbachol effect on learned fear (comparison between saline/carbachol vs. haloperidol/carbachol (n⫽14), Tukey-test: P⫽0.97). Furthermore, baseline startle amplitudes and spontaneous motor activity at test were not affected by the pre-training manipulations (ANOVAs: F’s⬍1.86, P’s⬎0.15). Experiment 3B: expression of fear. As in experiment 1B, rats (n⫽13) in the present experiment exhibited a significant fear-potentiation of the startle response in the presence of the light CS when infused with saline (tone alone vs. light tone, paired t-test: t(12)⫽3.41, P⫽0.005, Fig. 5B). Furthermore, ANOVA revealed a significant group difference in test performance [F(3,36)⫽12.05, P⬍0.001]. Haloperidol, when administered alone, did not affect test performance (comparison between saline/saline vs. haloperidol/saline, Tukey-test: P⫽0.96). But, as was found in experiments 1B and 2B, intra-NAC infusions of carbachol blocked expression of learned fear (saline/saline vs. saline/carbachol, Tukey-test: P⬍0.001). Importantly, co-administration of haloperidol did not alleviate the impairment on expression of learned fear produced by intraNAC carbachol (saline/carbachol vs. haloperidol/carbachol, Tukey-test: P⫽0.97).

I. Schwienbacher et al. / Neuroscience 140 (2006) 769 –778

775

Fig. 6. Effects of combined haloperidol/carbachol injections into the NAC on spontaneous motor activity. Carbachol injections increase motor activity. This effect was blocked by co-administration of haloperidol which had no effects injected alone. Post hoc Tukey-test after ANOVAs: comparison with sal/sal and carb/sal: ** P⬍0.01, comparison between carb/sal and carb/halo: ⫹ P⬍0.05. Abbreviations: see legend of Fig. 5.

by the carbachol (comparison between saline/carbachol vs. haloperidol/carbachol, P⫽0.02; saline/saline vs. haloperidol/carbachol, P⫽0.81). Misplaced injections Fig. 5. Effects of combined haloperidol/carbachol injections into the NAC on FPS. Bar diagrams show the mean ASR magnitude in the absence (black bars) or presence (white bars) of the light-CS. Hatched bars represent the difference between tone-alone and light-tone trials which was taken as an index of fear. Carbachol injected into the NAC immediately before fear-conditioning blocks the acquisition of conditioned fear (panel A); haloperidol had no effects injected alone and did also affect the carbachol effect on acquisition. The same pattern of effects was seen after haloperidol and carbachol immediately before testing on expression of conditioned fear (panel B). Post hoc Tukeytest after ANOVAs: ** P⬍0.01, *** P⬍0.001. Abbreviations: carb, carbachol; halo, haloperidol; sal, saline.

There were group differences in baseline startle magnitude (ANOVA: F(3,36)⫽10.62, P⬍0.001). Haloperidol alone had no effect (comparison saline/saline vs. haloperidol/saline, Tukey-test: P⫽0.99), but carbachol attenuated baseline startle in a similar manner as was observed in experiment 1B (saline/saline vs. saline/carbachol, Tukeytest: P⬍0.001). Co-administration of haloperidol did not affect this consequence of intra-NAC infusions of carbachol (saline/carbachol vs. haloperidol/carbachol, Tukeytest: P⫽0.96). As was found in experiment 1B, spontaneous motor activity was significantly enhanced by intra-NAC carbachol. An initial ANOVA yielded a significant group difference [F(3, 36)⫽8.34, P⬍0.001; Fig. 6]. Subsequent pairwise comparisons, with the Tukey procedure showed that rats in the saline/carbachol group were significantly more active than those in the saline/saline group (P⬍0.001). Haloperidol alone had no effects on motor activity (comparison between saline/saline vs. haloperidol/saline, P⫽ 0.96), but when given in combination with carbachol it did significantly reduce the heightened motor activity caused

In experiment 1A and 1B, 14 rats had misplaced injection sites of saline or carbachol. Because of the small sample sizes, we used a non-parametric statistic (Kruskal-Wallis test) for statistical analysis. Misplaced carbachol injections (mainly in the dorsal striatum) did not affect acquisition or expression of conditioned fear measured with FPS (H’s⬍1.22, P’s⬎0.75). Furthermore, baseline ASR amplitude was not changed (H’s⬍3.109, P’s⬎0.38). Surprisingly, there was a tendency for enhanced spontaneous motor activity at test in those rats with misplaced carbachol injections during training (i.e. Experiment 1A; H⫽7.04, P⫽0.07) but not in those given carbachol at test (i.e. Experiment 1B; H⫽5.691, P⫽0.128). In the other experiments, there were too few misplaced injections for a serious analysis of the effects of misplaced injections.

DISCUSSION The present series of experiments studied the effects of intra-NAC infusions of the unselective acetylcholine receptor agonist carbachol on the acquisition and expression of conditioned fear. We chose carbachol concentrations for which an activating effect of the appetitive system has previously been shown (Ikemoto et al., 1998). The results clearly show that carbachol infusions into the NAC completely block both the learning and the expression of conditioned fear. This occurred whether FPS or freezing was used as the measure of conditioned fear, and whether a context, a visual, or an auditory stimulus was used as the CS for the shock US. Besides these effects of intra-NAC carbachol on acquisition and expression of conditioned fear, an enhancement of spontaneous motor activity was

776

I. Schwienbacher et al. / Neuroscience 140 (2006) 769 –778

also observed (cf. Austin and Kalivas, 1988). To test whether the effects of carbachol on learned fear were a by-product of its effects on motor activity, we co-administered haloperidol in experiments 3A and 3B. Co-administration of haloperidol into the NAC should antagonize the motor-enhancing effects of intra-NAC carbachol (Austin and Kalivas, 1988). Indeed, co-administration of haloperidol blocked the effects of carbachol effects on spontaneous motor activity but did not affect the blockade of acquisition and expression of conditioned fear induced by intraNAC carbachol. These findings show that the effects of intra-NAC carbachol on the acquisition and expression of conditioned fear were independent of any motoric effects produced by that infusion. Recently, we demonstrated that temporary inactivation of the NAC, with local infusions of TTX, causes a disruption in the acquisition and expression of FPS (Schwienbacher et al., 2004). In a subsequent study, we explored whether this disruption might be due to an alteration in the intraNAC dopaminergic system. Surprisingly, the results of that study showed that intra-accumbal infusions of the DA agonist amphetamine had no effect on either the acquisition or the expression of conditioned fear (Josselyn et al., 2005; Schwienbacher et al., 2005). We explained these latter findings by suggesting that DA within the NAC is more important for motivation (Berridge and Robinson, 1998; Salamone and Correa, 2002) than it is for acquisition and expression of learning. Regardless of the interpretation of those findings, it was still unclear as to what neurotransmitter system within the NAC mediated the disruption in fear acquisition/expression produced by temporary inactivation of the NAC. Therefore, in the present study we examined the possibility that intra-NAC cholinergic activity might be involved in this disruption. Several studies have suggested that intra-NAC acetylcholine activates the appetitive system: (1) rats self-administer carbachol into the NAC (Ikemoto et al., 1998), (2) carbachol infused into the NAC elicits 50 kHz ultrasonic vocalizations in rats (which are emitted in appetitive behavior; summarized in Brudzynski, 2005) and produces a conditioned place preference (Schwienbacher et al., submitted for publication), and (3) local infusions of muscarinic cholinergic antagonists disrupt acquisition and expression of an instrumental response for food reward (Ikemoto and Panksepp, 1996; Pratt and Kelley, 2004). Other studies have also shown that acetylcholine release within the NAC is increased during the learning or expression of an aversive response, such as avoidance or taste aversion (Mark et al., 1995; Rada and Hoebel, 2001). Based on this evidence, it is possible that intra-NAC cholinergic activity is involved in either the acquisition or the expression of conditioned fear. Consistent with this possibility, the present data show that injections of carbachol into the NAC block acquisition and expression of conditioned freezing and FPS. It is likely that these effects are mediated by effects within the NAC because (1) misplaced injections (mainly in the dorsal striatum) had no effects on either the acquisition or the expression of conditioned fear, and (2) carbachol injections into the ventricle (which is located very near to the

NAC), and into a nearby brain site, the septum, actually have been found to increase fear responses (Bihari et al., 2003; Brudzynski and Bihari, 1990; Johansson et al., 1979). One particularly intriguing finding in the present study was that although pre-test infusions of carbachol into the NAC markedly disrupted expression of conditioned fear, essentially a total blockade, this had no effect on extinction learning. That is, even though rats given a pre-test infusion of carbachol failed to exhibit any conditioned fear response at test, they exhibited the same level of extinction on a test the following day as did saline-treated rats (experiment 2C). This finding is relevant to theoretical debates about the necessity of eliciting the conditioned response in order to extinguish it. As reviewed recently by Bouton (2004), there are very few data relevant to this issue in the case of aversive conditioning. Further, there are potential procedural concerns about some of the previously published data on this issue as well. The results of the present study, however, clearly show that it is not necessary to elicit the learned behavioral fear response in order to extinguish it. It is questionable whether this observation reflects a general mechanism in extinction learning since studies using the nictitating membrane response in the rabbit clearly showed that the expression of the conditioned response is necessary to extinguish it (Krupa and Thompson, 2003; Robleto et al., 2004). In the present study, carbachol injections into the NAC decreased baseline ASR amplitude (experiment 1B, 3B). Several other studies have found an attenuated ASR amplitude during appetitive arousal (Lang et al., 1998; Schmid et al., 1995; Steidl et al., 2001). Interestingly, DA within the NAC is not involved in the expression of this effect (Koch et al., 2000). This is supported by a recent study from our laboratory showing that infusions of the indirect DA agonist amphetamine into the NAC do not decrease baseline ASR (Schwienbacher et al., 2005). Taken together, the data described above suggest that intra-NAC acetylcholine is, in contrast to intra-NAC DA, involved in the reduction of ASR amplitude observed during activations of the appetitive system. On the other hand, intra-NAC acetylcholine and DA both appear to be involved in the potentiation of locomotor activity which is usually observed after stimulation of the appetitive system. That is, injections of amphetamine (Schwienbacher et al., 2005) or carbachol (present study; Austin and Kalivas, 1988) into the NAC enhance locomotor activity. Austin and Kalivas (1988) showed that the locomotor activity stimulating effect they observed was based on a carbachol induced increased DA transmission within the NAC. Co-administration of the D2 DA receptor antagonist haloperidol into the NAC blocks the locomotor activity stimulating effects of carbachol (Austin and Kalivas, 1988). We used this observation to demonstrate that the effects of carbachol on acquisition and expression of conditioned fear are not a by-product of carbachol’s motorenhancing effects (experiments 3A and 3B). This control experiment was important because several studies have demonstrated that motor activity can affect both baseline startle magnitude as well as conditioned potentiation of the

I. Schwienbacher et al. / Neuroscience 140 (2006) 769 –778

startle response (Leaton and Borszcz, 1985; Wecker and Ison, 1986). Obviously, enhanced motor activity could also affect the expression of conditioned freezing behavior (cf. discussion in Anagnostaras et al., 1999) although it is not clear how such an enhancement would consistently affect its acquisition across the range of stimuli (context, visual and auditory CSs) used here. Another critical issue which should be discussed here is the possibility that the observed effects are due to a state dependent process. That is, intra-NAC infusions of carbachol prior to acquisition do not disrupt the learning of the CS–US association but rather lead to that learning being encoded in the drug state induced by the carbachol; retrieval of that association is difficult in a drug-free state. The same logic applies in the expression experiments: rats that had acquired the CS–US association in a drug-free state have difficulty in retrieving that association when in the carbachol-induced drug state at test. Although there is no doubt that state-dependent effects can occur, there is at least one piece of indirect evidence arguing against this interpretation of the present results. Specifically, extinction learning (i.e. CS–no US) is much more susceptible to changes in context/state than is excitatory learning (i.e. CS–US; see Bouton et al., in press). Therefore, if intraNAC carbachol produces state-dependent learning, then one would particularly expect to see such an effect when testing for retention of extinction. However, that is not what was found in the present study (experiment 2C). In that experiment, rats infused with intra-NAC carbachol prior to test exhibited comparable levels of extinction as did salineinfused rats on a subsequent, drug-free test. The rats given carbachol should have exhibited substantially less extinction retention on this drug-free test if intra-NAC carbachol induced a state-dependent effect. Nevertheless, the question of whether a state dependent process might contribute to the effects reported in this study remains open. Taken together, the experiments in the present study demonstrate that injections of the unselective acetylcholine agonist carbachol into the NAC completely disrupt acquisition and expression of conditioned fear to different, discrete CSs and to a context, measured by either FPS or freezing. Further, this disruption in expression of conditioned fear does not appear to affect long-term extinction learning. One possible explanation of these findings is that acetylcholine transmission within the NAC encodes hedonic state, and that during positive hedonic states the learning and expression of conditioned fear are inhibited.

REFERENCES Anagnostaras SG, Maren S, Sage JR, Goodrich S, Fanselow MS (1999) Scopolamine and Pavlovian fear conditioning in rats: doseeffect analysis. Neuropsychopharmacology 21:731–744. Anderson R, Diotte M, Miliaressis E (1995) The bidirectional interaction between ventral tegmental rewarding and hindbrain aversive stimulation effects in the rat. Brain Res 688:15–20. Austin MC, Kalivas PW (1988) The effect of cholinergic stimulation in the nucleus accumbens on locomotor behavior. Brain Res 441: 209 –214.

777

Berridge KC, Robinson TE (1998) What is the role of dopamine in reward: hedonic impact, reward learning, or incentive salience? Brain Res Brain Res Rev 28:309 –369. Bihari A, Hrycyshyn AW, Brudzynski SM (2003) Role of the mesolimbic cholinergic projection to the septum in the production of 22 kHz alarm calls in rats. Brain Res Bull 60:263–274. Bouton ME (2004) Context and behavioral processes in extinction. Learn Mem 11:485– 494. Bouton ME, Westbrook RF, Corcoran K, Maren S. Contextual and temporal modulation of extinction: behavioural and biological mechanisms. Biol Psychiatry, in press. Boye SM, Grant RJ, Clarke PB (2001) Disruption of dopaminergic neurotransmission in nucleus accumbens core inhibits the locomotor stimulant effects of nicotine and D-amphetamine in rats. Neuropharmacology 40:792– 805. Brandao ML, Rees H, Witt S, Roberts MH (1991) Central antiaversive and antinociceptive effects of anterior pretectal nucleus stimulation: attenuation of autonomic and aversive effects of medial hypothalamic stimulation. Brain Res 542:266–272. Brown JS, Kalish HI, Farber IE (1951) Conditioned fear as revealed by magnitude of startle response to an auditory stimulus. J Exp Psychol 41:317–328. Brudzynski SM (2005) Principles of rat communication: quantitative parameters of ultrasonic calls in rats. Behav Genet 35:85–92. Brudzynski SM, Bihari F (1990) Ultrasonic vocalization in rats produced by cholinergic stimulation of the brain. Neurosci Lett 109:222–226. Cardinal RN, Parkinson JA, Hall J, Everitt BJ (2002) Emotion and motivation: the role of the amygdala, ventral striatum, and prefrontal cortex. Neurosci Biobehav Rev 23:321–352. Davis M, Astrachan DI (1978) Conditioned fear and startle magnitude: effects of different footshock or backshock intensities used in training. J Exp Psychol Anim Behav Proc 4:95–103. Dickinson A, Dearing MF (1979) Appetitive-aversive interactions and inhibitory processes. In: Mechanisms of learning and motivation (Dickinson A, Boakes RA, eds), pp 203–231. Hillsdale, NJ: Erlbaum. Dickinson A, Pearce JM (1977) Inhibitory interactions between appetitive and aversive stimuli. Psychol Bull 84:711. Fendt M, Fanselow MS (1999) The neuroanatomical and neurochemical basis of conditioned fear. Neurosci Biobehav Rev 23:743–760. Fendt M, Schwienbacher I, Schnitzler HU (2006) Cabochol injections into the nucleus accumbens induce 50 kHz calls in rats. Neurosci Lett, in press. Grauer E, Thomas E (1982) Conditioned suppression of medial forebrain bundle and septal intracranial self-stimulation in the rat: evidence for a fear-relief mechanism of the septum. J Comp Physiol Psychol 96:61–70. Haralambous T, Westbrook RF (1999) An infusion of bupivacaine into the nucleus accumbens disrupts the acquisition but not the expression of contextual fear conditioning. Behav Neurosci 113:925–940. Ikemoto S, Glazier BS, Murphy JM, McBride WJ (1998) Rats selfadminister carbachol directly into the nucleus accumbens. Physiol Behav 63:811– 814. Ikemoto S, Panksepp J (1996) Dissociations between appetitive and consummatory responses by pharmacological manipulations of reward-relevant brain regions. Behav Neurosci 110:331–345. Johansson GG, Rosset S, Sandstrom M, Soini H (1979) Increased fear reactions in cats after carbachol administered intraventricularly. Physiol Behav 22:803– 807. Jongen-Relo AL, Kaufmann S, Feldon J (2003) A differential involvement of the shell and core subterritories of the nucleus accumbens of rats in memory processes. Behav Neurosci 117:150 –168. Josselyn SA, Falls WA, Gewirtz JC, Pistell P, Davis M (2005) The nucleus accumbens is not critically involved in mediating the effects of a safety signal on behavior. Neuropsychopharmacology 30:17–26.

778

I. Schwienbacher et al. / Neuroscience 140 (2006) 769 –778

Koch M, Schmid A, Schnitzler H-U (2000) Role of nucleus accumbens dopamine D1 and D2 receptors in instrumental and Pavlovian paradigms of conditioned reward. Psychopharmacology 152:67–73. Konorski J (1967) Integrative activity of the brain. An interdisciplinary approach. Chicago: University of Chicago Press. Krupa DJ, Thompson RF (2003) Inhibiting the expression of a classically conditioned behavior prevents its extinction. J Neurosci 23:10577–10584. Lang PJ (1995) The emotion probe. Studies of motivation and attention. Am Psychol 50:385. Lang PJ, Bradley MM, Cuthbert BN (1998) Emotion, motivation, and anxiety: brain mechanisms and psychophysiology. Biol Psychiatry 44:1248 –1263. Leaton RN, Borszcz GS (1985) Potentiated startle: its relation to freezing and shock intensity in rats. J Exp Psychol Anim Behav Proc 11:421– 428. Mark GP, Weinberg JB, Rada PV, Hoebel BG (1995) Extracellular acetylcholine is increased in the nucleus accumbens following the presentation of an aversively conditioned taste stimulus. Brain Res 688:184 –188. Olds J (1958) Self-stimulation of the brain: its use to study local effects of hunger, sex, and drugs. Science 127:315–324. Parkinson JA, Robbins TW, Everitt BJ (1999) Selective excitotoxic lesions of the nucleus accumbens core and shell differentially affect aversive pavlovian conditioning to discrete and contextual cues. Psychobiology 27:256 –266. Paxinos G, Watson C (1997) The rat brain in stereotaxic coordinates. San Diego: Academic Press. Pellegrino LJ, Pellegrino AS, Cushman AJ (1979) A stereotaxic atlas of the rat brain. New York: Plenum Press. Pfister M, Boix F, Huston JP, Schwarting RK (1994) Different effects of scopolamine on extracellular acetylcholine levels in neostriatum and nucleus accumbens measured in vivo: possible interaction with aversive stimulation. J Neural Transm Gen Sect 97:13–25.

Pratt WE, Kelley AE (2004) Nucleus accumbens acetylcholine regulates appetitive learning and motivation for food via activation of muscarinic receptors. Behav Neurosci 118:730 –739. Rada PV, Hoebel BG (2001) Aversive hypothalamic stimulation releases acetylcholine in the nucleus accumbens, and stimulationescape decreases it. Brain Res 888:60 – 65. Robleto K, Poulos AM, Thompson RF (2004) Effects of a corneal anesthetic on extinction of the classically conditioned nictitating membrane response in the rabbit (Oryctolagus cuniculus). Behav Neurosci 118:1433–1438. Salamone JD (1994) The involvement of nucleus accumbens dopamine in appetitive and aversive motivation. Behav Brain Res 61:117–133. Salamone JD, Correa M (2002) Motivational views of reinforcement: implications for understanding the behavioral functions of nucleus accumbens dopamine. Behav Brain Res 137:3–25. Schmid A, Koch M, Schnitzler H-U (1995) Conditioned pleasure attenuates the startle response in rats. Neurobiol Learn Memory 64:1–3. Schwienbacher I, Fendt M, Richardson R, Schnitzler HU (2004) Temporary inactivation of the nucleus accumbens disrupts acquisition and expression of fear-potentiated startle in rats. Brain Res 1027:87–93. Schwienbacher I, Fendt M, Schnitzler HU (2005) Amphetamine injections into the nucleus accumbens affect neither acquisition/expression of conditioned fear nor baseline startle response. Exp Brain Res 160:538 –544. Steidl S, Li L, Yeomans JS (2001) Conditioned brain stimulation reward attenuates the acoustic startle reflex in rats. Behav Neurosci, in press. Wecker JR, Ison JR (1986) Effects of motor activity on the elicitation and modification of the startle reflex in rats. Anim Learn Behav 14:287–292. Wise RA, Rompre PP (1989) Brain dopamine and reward. Annu Rev Psychol 40:191–225.

(Accepted 23 February 2006) (Available online 3 April 2006)