Facilitation of Conditioned Fear Extinction

Neuroscience 134 (2005) 247–260 FACILITATION OF CONDITIONED FEAR EXTINCTION BY D-CYCLOSERINE IS MEDIATED BY MITOGEN-ACTIVATED PROTEIN KINASE AND PHOSPHATIDYLINOSITOL 3-KINASE CASCADES AND REQUIRES DE NOVO PROTEIN SYNTHESIS IN BASOLATERAL NUCLEUS OF AMYGDALA Y. L. YANGa AND K. T. LUb* Institute of Biotechnology, Department of Molecular Biology and Biochemistry, National Chia-Yi University, 300 University Road, Chia-Yi, Taiwan b Department of Life Science, National Taiwan Normal University, 88 Ming-Chow Road, Sec 4, Taipei, Taiwan a
Key words: extinction, D-cycloserine, MAPK, PI-3 kinase, amygdala. Abstract—Recent results showed that either systemic or intra-amygdala administration of D-cycloserine, a partial agonist at the glycine modulatory site on the glutamate N-methylD-aspartate receptor facilitates the extinction of conditioned fear. Here we evaluated the role of mitogen-activated protein kinase and phosphatidylinositol 3-kinase in the basolateral nucleus of amygdala on the effect of D-cycloserine.
The facilitation effect of D-cycloserine on fear extinction and mitogen-activated protein kinase activation was completely blocked by intra-amygdala administration of mitogen-activated protein kinase inhibitor PD98059 (500 ng/side, bilaterally) or U0-126 (20 M/side, bilaterally). Furthermore, phosphatidylinositol 3-kinase inhibitor (wortmannin, 5. 0 g/side, bilaterally) infused into the basolateral nucleus of amygdala signi? cantly reduced both facilitation effect of D-cycloserine and phosphatidylinositol 3-kinase activation.

Intra-amygdala administration of a transcription inhibitor (actinomycin D, 10 g dissolved in 1. 6 l vehicle; 0. 8 l per side) and a translation inhibitor (anisomycin, 125 g dissolved in 1. 6 l vehicle; 0. 8 l per side) completely blocked the facilitation effect of D-cycloserine. Control experiments indicated the blockage by actinomycin D or anisomycin were not due to lasting damage to the basolateral nucleus of amygdala or state dependency. In addition, none of the active drugs used here altered the expression of conditioned fear.
These results suggested that phosphatidylinositol 3-kinase and mitogenactivated protein kinase-dependent signaling cascades and new protein synthesis within the basolateral nucleus of amygdala played important roles in the D-cycloserine facilitation of the extinction of conditioned fear. © 2005 Published by Elsevier Ltd on behalf of IBRO. *Corresponding author. Tel: 886-2-29333149×234; fax: 886-229312904. E-mail address: [email protected] ntnu. edu. tw (K. -T. Lu).
Abbreviations: ACT DCS, actinomycin D D-cycloserine; ACT SAL, actinomycin D saline; ANI DCS, anisomycin D-cycloserine; ANI SAL, anisomycin saline; BLA, basolateral nucleus of the amygdala; CS, conditioned stimulus; DCS, D-cycloserine; EDTA, ethylenediaminetetraacetic acid; ISI, interstimulus interval; MAPK, mitogen-activated protein kinase; NMDA, N-methyl-D-aspartate; PD DCS, PD98059 D-cycloserine; PD SAL, PD98059 saline; PI-3K, phosphatidylinositol 3-kinase; US, unconditioned stimulus; U0 DCS, U0-126 D-cycloserine; U0 SAL, U0126 saline; VEH DCS, vehicle D-cycloserine; VEH SAL, vehicle saline; WH DCS, wortmannin D-cycloserine; WH SAL, wortmannin saline. 0306-4522/05$30. 00 0. 00 © 2005 Published by Elsevier Ltd on behalf of IBRO. doi:10. 1016/j. neuroscience. 2005. 04. 003 Fear conditioning occurs when a previously neutral stimulus (conditioned stimulus) is paired with an aversive stimulus (McAllister and McAllister, 1971).
Following such pairing the conditioned stimulus is thought to elicit a state of conditioned fear. This is de? ned in animals by their behavior: freezing, autonomic reactivity, and fear-potentiated startle. A large literature indicates that the basolateral nucleus of the amygdala (BLA) is critically involved in both the acquisition and the expression of conditioned fear (Davis, 2000). Neurotoxic lesions or intra-amygdala infusion of glutamate antagonists into the BLA blocks the expression of conditioned fear. In addition, local infusion of N-methyl-D-aspartate (NMDA) speci? c antagonists blocks the acquisition of conditioned fear (Miserendino et al. , 1990; Kim et al. , 1991; Maren et al. , 1996; Gewirtz and Davis, 1997).
Synaptic plasticity in this area is thought to underlie the learning process when afferent sensory information elicited by the conditioned stimulus is paired with afferent pain information elicited by the unconditioned stimulus (Fanselow and LeDoux, 1999). Extinction is de? ned as a reduction in conditioned fear when the conditioned stimulus (CS) is presented repeatedly in the absence of the unconditioned stimulus (US). Many studies show that extinction is not the result of forgetting or memory erasure but results from formation of new associations that compete with prior fear-conditioned associations (Falls and Davis, 1995; Davis et al. , 2000). Similar to acquisition, extinction is also blocked by glutamate NMDA receptor antagonists either given systemically (Cox and Westbrook, 1994; Baker and Azorlosa, 1996; Kehoe et al. 1996) or locally infused into the BLA (Falls and Davis, 1992). The glycine modulatory site of the NMDA receptor provides a critical regulatory role. Whereas direct NMDA agonists may be neurotoxic due to unregulated calcium entry, partial agonists can facilitate glutamatergic NMDA activity in a more limited fashion (Lawler and Davis, 1992; Olney, 1994). Recent results showed that partial agonists acting at the glycine modulatory site of the NMDA receptor, such as D-cycloserine (DCS), enhance learning and memory in several animal models (Thompson and Disterhoft, 1997; Pussinen et al. , 1997; Matsuoka and Aigner, 1996; Land and Riccio, 1999; Walker et al. , 2002; 247 248 Y. L. Yang and K. T.
Lu / Neuroscience 134 (2005) 247–260 extinction test, an extinction training and a post-extinction test (see Fig. 1A). Acclimation. On each of 3 consecutive days, rats were placed in the test chambers for 10 min and then returned to their home cages. Baseline startle test. On each of the next 2 consecutive days, animals were placed in the test chambers and presented with 30 95-dB startle stimuli at a 30-s interstimulus interval (ISI). Animals whose baseline startle response was 1% of the measurable level were not included in later analysis. Fear conditioning. Twenty-four hours later, rats were returned to the test chambers and after 5 min were given the ? rst of 10 light-footshock pairings.
The shock (US) was delivered during the last 0. 5 s of the 3. 7 s light (CS). The average intertrial interval was 4 min (range 3–5 min) and the shock intensity was 0. 6 mA. Pre-extinction test. Twenty-four hours after fear conditioning, rats were returned to the test chambers and 5 min later presented with 30 startle-eliciting noise bursts (95 dB, 30 s ISI). These initial startle stimuli were used to habituate the startle response to a stable baseline prior to the light-noise test trials that followed. Thirty seconds later a total of 20 startle-eliciting noise bursts were presented, 10 in darkness (noise alone) and 10 3. 2 s after onset of the 3. s light (light-noise) in a balanced, irregular order at a 30-s ISI. Percent fear-potentiated startle was computed as [(startle amplitude on light-noise noise-alone trials)/noisealone trials] 100. Rats were then divided into equal size groups of comparable mean levels of percent fear-potentiated startle. Rats with less than 50% fear-potentiated startle during the pre-extinction test were not used. Extinction training. Extinction training (cue exposure) is de? ned as the repetitive exposure to the CS cue (light) in the absence of the US (shock). Twenty-four hours after the preextinction test, rats were returned to the test chamber. After 5 min, they were presented with 30, 3. s light exposures at a 30-s ISI. Context control groups (context exposure) remained in the test cages for the same amount of time but did not receive light presentations. Extinction training was performed for varying numbers of consecutive days (2 days for experiment 1 and 1 day for subsequent experiments). Post-extinction test-1. Twenty-four hours after the last extinction training, rats were returned to the test chamber. After 5 min, they were presented with 30 95-dB leader stimuli for a habituated startle baseline. This was followed by a total of 60 startle-eliciting noise bursts, 30 in darkness (noise alone) and 30 presented 3. 2 s after onset of the 3. s light (light-noise) in a balanced, irregular order at a 30-s ISI. Results were evaluated the same way as pre-extinction test. Post-extinction test-2. Twenty-four hours after the extend extinction training period, rats were returned to the test chamber and process the post-extinction test described above. Fear-potentiated startle test. Twenty-four hours after fear conditioning, rats were returned to the test chamber and testing for fear-potentiated startle using the post-extinction test-1 described above. Ledgerwood et al. , 2003; Richardson et al. , 2004). In addition, ( )-HA966, a competitive antagonist at the glycine regulatory site on the NMDA receptor, reversed the DCS effect (Walker et al. , 2002).
Clinical studies have shown that DCS can sometimes enhance implicit memory and improve cognition in patients with Alzheimer’s disease (Schwartz et al. , 1996; Tsai et al. , 1998, 1999). It can also counter cognitive de? cits in schizophrenia (Javitt et al. , 1994; Goff et al. , 1999). Furthermore, systemic administration of DCS also proved to facilitate extinction of conditioned fear (Walker et al. , 2002; Ledgerwood et al. , 2003, 2004; Ressler et al. , 2004). Numerous signaling cascades including mitogen-activated protein kinase (MAPK), phosphatidylinositol 3-kinase (PI-3K) and calcineurin, are involved in the extinction of conditioned fear (Lu et al. , 2001; Lin et al. , 2003).
Similar mechanisms may also be involved in the facilitation effect of DCS. This study was designed to clarify the relationship between amygdaloid NMDA receptors, MAPK and PI-3K signal cascades on the extinction of conditioned fear. EXPERIMENTAL PROCEDURES Animals Adult male Sprague–Dawley (SD) rats (obtained from the animal center of National Taiwan University Taipei, Taiwan) weighing between 250 and 350 g were used. Animals were housed in groups of four rats in a temperature (24 °C) -controlled animal colony with continuous access to food and water. Rats were kept on a 12-h light/dark cycle with lights on at 07:00 h. All behavioral procedures took place during the light cycle.
All procedures were conducted in accordance with the National Institutes of Health Guide for Care and Use of Laboratory Animals and the guidelines set forth by the Institutional Animal Care and Use Committee at the National Taiwan Normal University. In all experimental procedures involving animals, all efforts were made to minimize pain and the number of animals used. Surgery All surgeries were carried out under sodium pentobarbital (50 mg/ kg, i. p. ) anesthesia. Once anesthetized, the rat was placed in a Kopf stereotaxic instrument, the skull was exposed, and 22 gauge guide cannula (model C313G, Plastic-one Products, Roanoke, VA, USA) were implanted bilaterally into the BLA (AP, 2. ; DV, 9. 0, ML, 5. 0 from bregma (Paxinos and Watson, 1997)). Size 0 insect pins (Carolina Biological Supply, Burlington, NC, USA) were inserted into each cannula to prevent clogging. These extended about 2 mm past the end of the guide cannula. Screws were anchored to the skull and the assembly was cemented in place using dental cement (Plastic-one Products). Rats received an antibiotic (penicillin) once every day for the ? rst 3 days after the surgery to reduce the risk of infection. General behavioral procedures Fear conditioning was measured using the potentiated startle paradigm (Cassella and Davis, 1986; Lu et al. , 2001; Walker et al. , 2002).
The rats were trained and tested in a startle chamber (San Diego Instruments, San Diego, CA, USA) in which cage movement resulted in the displacement of an accelerometer. Startle amplitude was de? ned as peak accelerometer voltage within 200 ms after startle stimulus onset. The behavioral procedures common to all experiments consisted of an acclimation phase, a baseline startle test phase, a fear conditioning phase, a pre- Drug injection DCS (Sigma) was freshly dissolved in saline. DCS (15 mg/kg, i. p. ) or saline was injected intraperitoneally 15 min prior to extinction training with a 26 gauge injection needle connected to a 1 ml syringe (Walker et al. , 2002; Ledgerwood et al. , 2003) (experiments 1– 8).
MAPK inhibitor PD98059 (500 ng in 1 l of 20% DMSO; Calbiochem) (Lu et al. , 2001) or U0-126 (50 ng/side; Sigma) (Lin et al. , 2003) or 20% DMSO was infused into the BLA Y. L. Yang and K. T. Lu / Neuroscience 134 (2005) 247–260 249 Fig. 1. Systemic administration of DCS accelerated extinction of conditioned fear. (A) Timeline of behavioral procedures for experiment 1. (B) Percent fear-potentiated startle measured 24 h before (pre-extinction test) and 24 h after extinction training (post-extinction test). Rats in each group were treated with either DCS or saline prior to a single session of extinction training. (C) To test for toxicity, after 24 h all animals of experiment 1 were retrained.
They were tested for fear-potentiated startle response in the absence of drugs 24 h later (fear-potentiated startle test) (values are mean SEM, * P 0. 05 versus control group; # P 0. 05 versus the group with 1 day extinction and saline injection). 250 Y. L. Yang and K. T. Lu / Neuroscience 134 (2005) 247–260 10 min prior to saline or DCS injection (experiments 2 and 8). PI-3K inhibitor (wortmannin, 5 g/side) (Lin et al. , 2003) or vehicle was administrated to the BLA 10 min prior to saline or DCS injection (experiment 3). Transcription inhibitor actinomycin D (10 g dissolved in 1. 6 l vehicle; 0. 8 l per side) or translation inhibitor (anisomycin, 125 g dissolved in 1. 6 l vehicle; 0. 8 l per side) or vehicle (Lin et al. 2003) was administrated to the BLA 10 min prior to DCS or saline injection (experiment 4) or 25 min prior to fear-potentiated startle test (experiment 6). In the control experiment, PD98059, U0-126, wortmannin, actinomycin D, and anisomycin were injected 25 min prior to the fear-potentiated startle test. Injections were made through 28-gauge injection cannula (model C313I, Plastic-one Products) connected to a Hamilton micro-syringe via polyethylene tubing. Infusion speed was 0. 25 l/ min. The total volume of injection was 0. 8 l per side. Western blot analysis Animals were killed by decapitation 10 min after extinction training. The lateral and basolateral subregions of the amygdala were collected and sonicated brie? y in ice-cold buffer: 50 mM Tris–HCl (pH 7. ), 50 mM NaCl, 10 mM EGTA, 5 mM EDTA, 2 mM sodium pyrophosphate, 4 mM para-nitrophenylphosphate, 1 mM sodium orthovanadate, 1 mM phenylmethylsulfonyl ? uoride (PMSF), 20 ng/ml leupeptin, and 4 ng/ml aprotinin. Following sonication, the soluble extract was obtained after pelleting the crude membrane fraction in a centrifuge at 50,000 g at 4 °C. Protein concentration in the soluble fraction was then measured using a Bradford assay with bovine serum albumin as the standard. Equivalent amounts of protein for each sample were resolved in 10% sodium dodecyl sulfate (SDS)–polyacrylamide gels, blotted electrophoretically to PVDF membranes and blocked overnight in 5% skim milk (Cell Signaling Technology, Inc. , USA).
Blots were incubated with antiphospho-ERK antibody (New England Biolabs, USA), anti-ERK antibody (BD Transduction Laboratories, USA), anti-phospho-Akt antibody (New England Biolabs) or anti-pan-Akt (BD Transduction Laboratories). Band detection was performed with an enhanced chemiluminescence Western blotting analysis system (RPN 2108; Amersham International, Amersham, UK). fear-potentiated startle during the pre-extinction test. The ? nal 30 rats were assigned into ? ve groups of six animals based on their level of fear-potentiated startle in the preextinction test. Twenty-four hours after the pre-extinction test, each rat received 1 or 2 consecutive days of extinction training with DCS (15 mg/kg, i. p. ) or saline. Saline or DCS was injected 15 min prior to the extinction training.
An additional control group was tested 2 days after the pre-extinction training without intervening exposures to visual CS. Fig. 1B shows that DCS accelerated extinction of conditioned fear. A two way ANOVA for differences in treatment (DCS vs saline) and days (one or two extinction sessions) between-subjects indicated a signi? cant treatment effect (F(1,20) 9. 02) and a signi? cant treatment days interaction (F(2,20) 6. 68). Thus, the reduction of fear-potentiated startle after 1 day of extinction training was greater in the group that received DCS than in the group that received saline (Fig. 1B). Individual comparisons between DCS- and saline-treated groups indicated signi? ant differences after 1 day of extinction sessions (t(10) 3. 86). Previous studies have shown that lesions of the BLA block expression of fear-potentiated startle (Campeau and Davis, 1995). DCS may have toxic effect on BLA, and resulting misinterpretation of its facilitation effects on extinction. To test for toxicity, all animals of experiment 1 were retrained and tested 24 h later. Under these conditions, animals previously injected with DCS or saline showed a signi? cant fear-potentiated startle (Fig. 1C). Thus, the facilitation effect of DCS observed during the post-extinction test 1 appeared to result from the acute drug effect rather than from a more permanent, perhaps toxic, action of DCS.
Experiment 2: intra-amygdala infusion of MAPK inhibitors blocked the facilitation of extinction by DCS To test the possible role of MAPK-dependent signaling cascade in the DCS-enhanced effect on the extinction of condition fear, 48 rats received fear conditioning, extinction training, and testing for fear-potentiated startle. Initially, 58 rats were used, but 10 of them were excluded. Rats were randomly assigned to six different groups and received one of the following treatments: vehicle saline (VEH SAL), vehicle DCS (VEH DCS), PD98059 DCS (PD DCS), U0-126 DCS (U0 DCS), PD98059 saline (PD SAL) or U0-126 saline (U0 SAL). The MAPK inhibitors, PD98059, and U0-126 (or vehicle) were administrated to the BLA 10 min prior to the injection with DCS or saline. Animals were then returned to their cage.
Fifteen minutes after injection, animals were subjected to a single session of extinction training. Previously, we show that a single day of extinction training with cue exposure led to about 35% decrease in fear-potentiated startle, whereas 2–3 days of extinction training led to near complete extinction (Lu et al. , 2001; Walker et al. , 2002). We concluded that the acceleration of extinction is best detected after a single session of extinction training. As shown in Fig. 2, DMSO, PD98059 (500 ng/side, bilaterally), or U0-126 (20 nM/per side, bilaterally) was given 10 min prior to saline or DCS injection; rats were returned to their cages for 30 min before a single Histology
Rats were overdosed with chloral hydrate and perfused intracardially with 0. 9% saline followed by 10% formalin. The brains were removed and immersed in a 30% sucrose-formalin solution for at least 3 day. Coronal sections (30 M) were cut through the area of interest, stained with Cresyl Violet, and examined under light microscope for cannula placement. Animals with misplaced cannula were not included in later analysis. Statistical analysis The mean startle amplitude across the 30 noise alone and 30 light-noise trials during the pre- and post-extinction tests was calculated for each animal. All results were analyzed using a score of percent fear-potentiated startle, as de? ned in the post-extinction tests above.
ANOVA on scores was the primary statistical measure. Between-group comparisons were made using two-tailed t-tests for independent samples. The criterion for signi? cance for all comparisons was P 0. 05. RESULTS Experiment 1: systemic administration of DCS accelerated extinction of conditioned fear This experiment assessed the facilitation effect of DCS on different amounts of extinction training. Initially, 35 rats were used. Five were excluded for showing less than 50% Y. L. Yang and K. T. Lu / Neuroscience 134 (2005) 247–260 251 Fig. 2. Intra-amygdala infusion of MAPK inhibitors blocked facilitation effect of DCS on extinction. (A) Timeline of behavioral procedures for experiment 2. B) Cannula was placed in the BLA. Percent fear-potentiated startle measured 24 h before (pre-extinction test) and 24 h after extinction training (post-extinction test). Rats in each group underwent VEH SAL, VEH DCS, PD DCS, U0 DCS, PD SAL, or U0 SAL prior to a single session of extinction training. Twenty-four hours later, animals were tested for fear-potentiated startle in the absence of drugs (values are mean SEM, * P 0. 05 versus VEH SAL group; # P 0. 05 versus VEH DCS group). (C) Cannula tip placements transcribed onto atlas plates adapted from Paxinos and Watson (1997). 252 Y. L. Yang and K. T. Lu / Neuroscience 134 (2005) 247–260 session of extinction training.
Twenty-four hours later, animals were tested for fear-potentiated startle in the absence of drugs. Results showed that there was a signi? cant overall difference between treatments (F(5,42) 11. 81). Fig. 2 shows that administration of DCS facilitated extinction of conditioned fear (VEH DCS) compared with the control group (VEH SAL) (t(14) 3. 12, P 0. 05). This effect was blocked by co-administration of MAPK inhibitor PD98059 (PD DCS) or U0-126 (U0 DCS) (t(14) 3. 08, P 0. 05 and t(14) 3. 29, P 0. 05, respectively) compared with the control (VEH DCS), treated with PD98059 only (PD SAL) or U0-126 only (U0 SAL) did not affect extinction (t(14) 0. 7 and t(14) 0. 36, respectively). These results indicated that the MAPK dependent signaling cascade most likely mediated the facilitation effect of DCS. Experiment 3: intra-amygdala infusion of the PI-3K inhibitor blocked facilitation of extinction by DCS Previous ? ndings have shown that PI-3K inhibitors retard acquisition in a variety of learning paradigms (Lin et al. , 2003). To evaluate the possible role of PI-3K signaling cascade in the DCS enhancement of extinction of conditioned fear, 32 rats received fear conditioning, extinction training, and testing for fear-potentiated startle. Although 38 rats were used initially, six were excluded.
They were then randomly assigned to four different groups and received one of the following treatments: VEH SAL, VEH DCS, wortmannin DCS (WH DCS) and wortmannin saline (WH SAL). The PI-3K inhibitor (wortmannin, 5 g/side, bilaterally) was infused to the BLA 10 min prior to the injection of saline or DCS. Then rats were returned to their cages for 15 min before being subjected to a single session of extinction training. Twenty-four hours later, animals were tested for fear-potentiated startle in the absence of drugs. Results showed that there was a signi? cant overall difference between treatments (F(4,28) 12. 17). As shown in Fig. 3, the facilitation effect of DCS (VEH DCS) on extinction was blocked by co-administration of PI-3K inhibitor (WH DCS) (t(14) 2. 98, P 0. 05).
With the single extinction training session used in this experiment, this dose of wortmannin alone (WH SAL) at this dose had no effect on the extinction of fear-potentiated startle compared with control group (VEH SAL) (t(14) 0. 19). These results suggest that the PI-3K signaling cascade was involved in the DCS facilitation of extinction. Experiment 4: DCS enhanced the extinction training induced MAPK and PI-3K phosphorylation According to the results of the above experiments, the DCS facilitation effect on extinction was prevented by coadministration of MAPK or PI-3K inhibitor. Previous studies have shown that infusion of these same inhibitors blocks extinction (Lu et al. , 2001; Lin et al. , 2003). Therefore, these treatments in conjunction with DCS must result in no extinction and resulting misinterpretation of its blockage effects on DCS.
To show the MAPK and PI-3K signaling pathways are essential for the facilitation effect of DCS, we used Western blot to evaluate the DCS effect on the extinction training induced MAPK and PI-3K phosphorylation. Additional amygdala-cannulated rats received fear conditioning, extinction training, and testing for fear-potentiated startle. Then PD98059 or wortmannin was infused to the BLA 10 min prior to the injection of saline or DCS. Rats were returned to their cages. Fifteen minutes after DCS or saline injection, animals were subjected to a single session of extinction training. Animals were killed by decapitation 10 min after extinction training.
The lateral and basolateral sub-regions of the amygdala were tested with Western blot analysis. Compared with control group, MAPK phosphorylation was signi? cantly elevated in BLA after extinction training (Fig. 4A, lane 2). Administration of DCS enhanced the effect of extinction training on MAPK phosphorylation (Fig. 4A, lane 3). The MAPK inhibitor PD98059 blocked the effect of DCS (Fig. 4A, lane 4). In addition, we measured the state of Akt phosphorylation as an index of PI-3K activity (Lin et al. , 2001). Fig. 4B showed that administration of DCS enhanced the effect of extinction training on Akt phosphorylation (Fig. 4B, lane 3). The PI-3K inhibitor, wortmannin, blocked the enhancement effect of DCS (Fig. 4b, lane 4).
These results raise the possibility that DCS enhancement effect of extinction of conditioned fear is mediated by the amygdaloid MAPK and PI-3K dependent signaling cascades. Experiment 5: intra-amygdala infusion of the transcription inhibitor or translation inhibitor blocked DCS facilitation of extinction The MAPK pathway participates in the synthesis of proteins important for the long-term stabilization and storage of fear memories. According to the result of experiment 2, the facilitation effect of DCS on extinction is mediated by the MAPK dependent signaling cascade. We predicted that the facilitation effect of DCS required new protein synthesis in the BLA.
To test this hypothesis, 48 rats received fear conditioning, extinction training, and testing for fear-potentiated startle. Initially, 56 rats were used but eight of them were excluded. Rats were then randomly assigned to six different groups and received one of the following treatments: VEH SAL, VEH DCS, actinomycin D DCS (ACT DCS), anisomycin DCS (ANI DCS), actinomycin D saline (ACT SAL), and anisomycin saline (ANI SAL). Transcription inhibitor (actinomycin D, 10 g dissolved in 1. 6 l vehicle; 0. 8 l per side) and translation inhibitor (anisomycin, 125 g dissolved in 1. 6 l vehicle; 0. 8 l per side) were administered to the BLA 10 min prior to saline or DCS injection. Then rats were returned to their cages. Fifteen minutes later, nimals were subjected to a single session of extinction training. Twenty-four hours later, animals were tested for fear-potentiated startle in the absence of drugs. Results showed that there was a significant overall difference between treatments (F(5,42) 10. 17). As shown in Fig. 5, actinomycin D and anisomycin completely blocked the facilitation effect of DCS (t(14) 3. 11 and t(14) 2. 96, respectively) compared with the VEH DCS group. With a single extinction training session used in this experiment, actinomycin alone (ACT SAL) or anisomycin alone (ANI SAL) did not affect the extinction of fear-potentiated startle compared with control Y. L. Yang and K. T.
Lu / Neuroscience 134 (2005) 247–260 253 Fig. 3. Intra-amygdala infusion of the PI-3K inhibitor blocked the facilitation effect of DCS on extinction. (A) Timeline of behavioral procedures for experiment 3. (B) Cannula was placed in the BLA. Percent fear-potentiated startle measured 24 h before (pre-extinction test) and 24 h after (post-extinction test) extinction training. Rats in each group were treated with VEH SAL, VEH DCS, WH DCS, or WH SAL prior to a single session of extinction training. Twenty-four hours later, animals were tested for fear-potentiated startle in the absence of drugs (values are mean SEM, * P 0. 05 versus VEH SAL group). C) Cannula tip placements transcribed onto atlas plates adapted from Paxinos and Watson (1997). 254 Y. L. Yang and K. T. Lu / Neuroscience 134 (2005) 247–260 (VEH SAL) (t(14) 0. 88 and t(14) 0. 48, respectively). These results suggest that new protein synthesis within the BLA played an important role in DCS facilitation of extinction of conditioned fear. Experiment 6: the disruptive effect of intra-amygdala infusion of actinomycin D and anisomycin was not attributed to lasting damage to the amygdala The active drugs used in the above experiments may have toxic effect within the amygdala. Previous work shows that infusion of PD98095 (Lu et al. , 2001) or wortmannin (Lin et al. 2003) into BLA did not appear to cause permanent impairment of amygdala function. To test for possible toxic effects of actinomycin D and anisomycin on the BLA, all animals of experiment 5 received an additional 2 days of drug free extinction training followed 24 h later by testing. Under these conditions, rats previously treated with actinomycin D (ACT DCS-treated group and ACT SALtreated group) or anisomycin (ANI DCS- and ANI SALtreated group) showed a signi? cant reduction of fearpotentiated startle between post-extinction test 1 and post-extinction test 2 (t(7) 3. 08 and t(7) 3. 32 for the ACT DCS-treated group and ACT SAL-treated group respectively) and (t(7) 2. 96 and t(7) 3. 1 for the ANI DCStreated group and ANI SAL-treated group respectively) (Fig. 6B). Thus, the blockage of extinction observed during post-extinction test 1 appeared to result from an acute drug effect rather than from a more permanent and perhaps toxic action, of actinomycin D or anisomycin. Previous studies have shown that lesions of the BLA block fear-potentiated startle (Campeau and Davis, 1995), an outcome opposite from that obtained with infusion of actinomycin D or anisomycin. It is also important to note that actinomycin D or anisomycin may have long-term toxicity within the BLA. The extinction of fear would look the same as a gradual loss of ability to express or relearn fear.
Experiment 7: the disruptive effect of intra-amygdala infusion of actinomycin D and anisomycin was not attributed to state dependency To evaluate the contribution of state-dependency effects to the results obtained in experiment 6, additional amygdala-cannulated rats were tested for extinction in a drug-free state and after receiving the same compound that they had received during extinction training. Results showed that there was a signi? cant overall difference between treatments in post-extinction test 2 (F(2,21) 32. 16). These results are shown in Fig. 7. Rats infused with actinomycin or anisomycin before postextinction test 2 showed a slight, but non-signi? cant, decrease in fear-potentiated startle from post-extinction test 1 to post-extinction test 2. For control rats (n 8), fear-potentiated startle was signi? cantly lower during post-extinction test 2 than post-extinction test 1 (t(7) 2. 455; P 0. 05). The lost of fear-potentiated startle in both groups probably re? cted incidental extinction produced by the 30 non-reinforced CS presentations of post-extinction test 1. The failure of rats infused before Fig. 4. MAPK and PI-3K inhibitors blocked extinction training activation of MAPK and PI-3K. (A) Representative Western blots and densitometric analysis of the activation of MAPK in the BLA under different treatments (values are mean SEM, * P 0. 05 versus VEH SAL group). (B) Representative Western blots and densitometric analysis Akt phosphorylation as an index of PI-3K activity in the BLA under different treatments (values are mean SEM, * P 0. 05 versus VEH DCS group). Y. L. Yang and K. T. Lu / Neuroscience 134 (2005) 247–260 255 Fig. 5.
Intra-amygdala infusion of the transcription inhibitor or translation inhibitor blocks the facilitation effect of DCS on extinction of conditioned fear. (A) Timeline of behavioral procedures for experiment 5. (B) Cannula was placed in the BLA. Percent fear-potentiated startle measured 24 h before (pre-extinction test) and 24 h after (post-extinction test 1) extinction training. Rats underwent treatment with VEH SAL, VEH DCS, ACT DCS, ANI DCS, ACT SAL, or ANI SAL prior to a single session of extinction training. Twenty-four hours later, animals were tested for fear-potentiated startle in the absence of drugs (values are mean SEM, * P 0. 05 comparing with the VEH SAL group; # P 0. 05 compared with the VEH DCS group). C) Cannula tip placements transcribed onto atlas plates adapted from Paxinos and Watson (1997). 256 Y. L. Yang and K. T. Lu / Neuroscience 134 (2005) 247–260 Fig. 6. The disruptive effects of intra-amygdala infusion of actinomycin D and anisomycin on extinction were not attributed to lasting damage to the BLA. (A) Timeline of behavioral procedures for experiment 6. The same animals used in experiment 3 were subjected for two more trials of extinction training. (B) Twenty-four hours after the last extinction training, animals were tested for fear-potentiated startle in the absence of drugs (post-extinction test-2) (values are mean SEM, * P 0. 05 versus the corresponding post-extinction test-2). esting with the transcription and translation inhibitors before testing to show a loss of fear-incidental extinction suggested that state dependency was not a major factor in the effects of actinomycin D and anisomycin. Experiment 8: effect of pretest PD98059, U0-126, wortmannin, actinomycin, and anisomycin administration on fear-potentiated startle This experiment was designed to evaluate whether the effect of the active drugs used has had a secondary effect on fear itself or on CS processing. For example, if MAPK inhibitor U0-126 reduced CS-elicited fear, this might attenuate extinction by decreasing the discrepancy between CS predictions and what actually occurred. If actinomycin D or anisomycin interfered with visual processing, this might block extinction produced by non-reinforced exposures to the visual CS.
To evaluate these possibilities, 42 amygdala-cannulated rats received acclimation, baseline startle test, and fear conditioning. Initially, 50 rats were used, but eight of them were excluded. After 24 h, rats were infused with PD98059, U0-126, wortmannin, actinomycin, and anisomycin. At 25 min after the infusions, rats were tested for fear-potentiated startle. As shown in Fig. 8, none of the active drugs we used here signi? cantly in? uenced fearpotentiated startle (F(6,35) 0. 993). Thus, it is unlikely that these drugs in? uenced extinction by increasing fear or by disrupting CS processing. Y. L. Yang and K. T. Lu / Neuroscience 134 (2005) 247–260 57 Fig. 7. The disruptive effect of intra-amygdala infusion of actinomycin D and anisomycin were not attributed to state dependency. (A) Timeline of behavioral procedures for experiment 7. (B) Cannula was placed in the BLA. Percent fear-potentiated startle measured 24 h before (pre-extinction test), 24 h after (post-extinction test 1), and 48 h after (post-extinction test 2) extinction training. Rats in each group underwent VEH SAL, ACT DCS, or ANI DCS prior to a single session of extinction training and prior to post-extinction test 2. Animals were tested for fear-potentiated startle in the absence of drugs (values are mean SEM, * P 0. 05). C) Cannula tip placements transcribed onto atlas plates adapted from Paxinos and Watson (1997). 258 Y. L. Yang and K. T. Lu / Neuroscience 134 (2005) 247–260 Fig. 8. Effect of pretest PD98059, U0-126, wortmannin, actinomycin, and anisomycin administration on fear-potentiated startle. (A) Timeline of behavioral procedures for experiment 8. (B) Cannula was placed in the BLA. Percent fear-potentiated startle was measured 24 h after fear conditioning. Rats were treated with DMSO, PD98059 (PD), U0-126 (U0), wortmannin (WH), vehicle (VEH), actinomycin (ACT), or anisomycin (ANI) 25 min prior to the fear-potentiated startle test (values are mean SEM). C) Cannula tip placements transcribed onto atlas plates adapted from Paxinos and Watson (1997). DISCUSSION We build on the previous ? ndings that DCS facilitated extinction of conditioned fear (Walker et al. , 2002; Ledgerwood et al. , 2003, 2004; Ressler et al. , 2004). Here, we show for the ? rst time that the DCS effect was prevented by co-administration of MAPK, PI-3K, transcription, and translation inhibitors. Control experiments indicated that the blocking effects of actinomycin D and anisomycin on extinction were not due to lasting damage to the BLA or state dependency. In addition, none of active drugs we used in this study altered the expression of conditioned fear.
These results suggest that PI-3K and MAPK-dependent signaling cascades and de novo protein synthesis within the BLA were important for DCS facilitation. Early behavioral studies by Pavlov (1927) and Konorski (1948) de? ned extinction as an active process involving formation of new inhibitory associations as opposed to forgetting previously conditioned associations. Numerous studies since have con? rmed and elaborated these early ? ndings (reviewed in Falls and Davis, 1995; Davis et al. , 2000). It is now well accepted that extinction occurs with repeated presentation of a CS in the absence of the pre- Y. L. Yang and K. T. Lu / Neuroscience 134 (2005) 247–260 259 viously paired US.
This reduces the conditioned response elicited by the CS. In contrast to forgetting which implies the passive loss of memory, extinction implies active formation of new inhibitory associations competing with and overpowering original excitatory associations. Evidence is growing that extinction may involve circuits and use mechanisms of synaptic plasticity similar to those of conditioned fear learning (Falls and Davis, 1992; Cox and Westbrook, 1994; Baker and Azorlosa, 1996; Davis et al. , 2000). NMDA-dependent synaptic plasticity appears to mediate many forms of active learning (Morris, 1989; Staubli et al. , 1989; Flood et al. , 1990; Collinridge and Bliss, 1995).
It is likely that conditioned fear learning depends on CS–US pairing mediated by NMDA receptors within the BLA (Miserendino et al. , 1990; Fanselow and LeDoux, 1999). Extinction also appears to require active, NMDA-dependent learning within the amygdala. This was demonstrated by blockage of extinction by microinjections of APV into the BLA in both fear-potentiated startle (Falls and Davis, 1992) and freezing paradigms (Lee and Kim, 1998). Furthermore, systemic administration of a different NMDA antagonist, MK801, blocks the extinction process in a range of different learning paradigms (Cox and Westbrook, 1994; Baker and Azorlosa, 1996; Kehoe et al. , 1996).
Recently, DCS, a partial agonist acting at the strychnine-insensitive glycine-recognition site of the NMDA receptor complex, has repeatedly been shown to facilitate learning in various cue and context association paradigms (Monahan et al. , 1989; Flood et al. , 1992; Thompson and Disterhoft, 1997). Walker et al. (2002) reported the ? rst evidence that DCS facilitates extinction of learned fear. Since then, further studies con? rmed and elaborated this early ? nding (Ledgerwood et al. , 2003, 2004; Ressler et al. , 2004). These studies reported that DCS is more effective at facilitating extinction when given after extinction training, rather than before. They interpret these ? dings as evidence that DCS facilitates the consolidation of a new memory acquired during extinction. It is important to note that although some studies have shown DCS to be effective in improving memory impairment due to Alzheimer’s disease (Schwartz et al. , 1996; Tsai et al. , 1999) and schizophrenia (Javitt et al. , 1994; Goff et al. , 1999), other studies found little or no improvement (Tsai et al. , 1998; van Berckel et al. , 1999). This may be related to the fact that acute treatment with DCS may have a more pronounced facilitation than chronic treatment (Quartermain et al. , 1994; Ledgerwood et al. , 2003; Richardson et al. , 2004). Ledgerwood et al. (2003, 2004) reported that DCStreated animals fail to exhibit reinstatement effects.
That DCS enhances extinction may be through some processes different from extinction induced by repeat representation of CS. Lin et al. (2003) investigated the similarities and differences between consolidation of conditioning and consolidation of extinction. They found that both processes depend on activation of NMDA receptors, PI-3K, MAPK, and require synthesis of new proteins within the amygdala. They also found that different characteristics show differential sensitivity to the transcription inhibitor actinomycin D. Our results were consistent with the model that the extinc- tion process involved active learning of new inhibitory associations.
Here we showed that DCS facilitation of extinction could be blocked by actinomycin D and anisomycin. These seemingly con? icting results could be attributable to our extinction protocol. Our protocol resembled betweensession extinction, presumably corresponding to long-term extinction memory. In addition, we used DCS to facilitate the extinction process and tested the animals in a drug free condition. Acquisition or consolidation of long-term memory requires activation of protein kinase, transcription of genes, new protein synthesis, and synapse formation (Schafe and LeDoux, 2000). Similar mechanisms were involved in the DCS facilitation of extinction. The DCS activated NMDA receptors, resulted in Ca2 in? x into the cell, and led to the PI-3K and MAPK activation. Activated MAPK can translocate to the nucleus, subsequently activating transcription factors to promote gene transcription and new protein synthesis. Thus, combinations of drugs and extinction training may weaken or erase original memory. There is increasing evidence that learning of CS–US associations involves synaptic plasticity within the BLA, leading to differential activation of this circuit by sensory afferents (Davis, 1997; Rogan et al. , 1997; Lee and Kim, 1998; Fanselow and LeDoux, 1999). Our results suggested that the extinction of conditioned fear also involved NMDA-dependent plasticity, but speci? inhibitory circuits may be activated by extinction learning. We hypothesize that this newly activated inhibitory circuit would oppose conditioned excitatory pathways normally leading to activation of the central nucleus of the amygdala, resulting in the elicitation of fear responses. CONCLUSION This may be the ? rst study to show that PI-3K and MAPKdependent signaling cascade and de novo protein synthesis within the BLA were essential to the DCS facilitation of the extinction of conditioned fear. Acknowledgments—The work was supported by grants from the National Science Council (NSC 90-2320-B-003-007, NSC 902320-B-006-038, NSC 93-2320-B-003-003).
Our gratitude (also) goes to the Academic Paper Editing Clinic, NTNU. REFERENCES Baker J, Azorlosa J (1996) The NMDA antagonist MK-801 blocks the extinction of Pavlovian fear conditioning. Behav Neurosci 110:618–620. Campeau S, Davis M (1995) Involvement of the central nucleus and basolateral complex of the amygdala in fear conditioning measured with fear-potentiated startle in rats trained concurrently with auditory and visual conditioned stimuli. J Neurosci 15:2301–2311. Cassella J, Davis M (1986) The design and calibration of a startle measurement system. Physiol Behav 36:377–383. Collinridge G, Bliss T (1995) Memories of NMDA receptors and LTP. Trends Neurosci 18:54 –56.
Cox J, Westbrook R (1994) The NMDA receptor antagonist MK-801 blocks acquisition and extinction of conditioned hypoalgesia responses in the rat. Q J Exp Psychol 47B:187–210. Davis M (1997) Neurobiology of fear responses: The role of the amygdala. J Neuropsychol Clin Neurosci 9:382– 402. Davis M (2000) The role of the amygdala in conditioned and unconditioned fear and anxiety. In: The amygdala, Vol. 2 (Aggleton JP, ed), pp 213–287. Oxford, UK: Oxford University Press. 260 Y. L. Yang and K. T. Lu / Neuroscience 134 (2005) 247–260 visual recognition memory in rhesus monkeys. J Pharmacol Exp Ther 278:891– 897. McAllister WR, McAllister DE (1971) Behavioral measurements of conditioned fear. In: Aversive conditioning and learning (Brush FR, ed), pp 105–179. New York: Academic Press.
Miserendino MJD, Sananes CB, Melia KR, Davis M (1990) Blocking of acquisition but not expression of conditioned fear-potentiated startle by NMDA antagonists in the amygdala. Nature 345:716–718. Monahan J, Handelmann G, Hood W, Cordi A (1989) D-cycloserine, a positive modulator of the N-methyl-D-aspartate receptor, enhances performance of learning tasks in rats. Pharmacol Biochem Behav 34:649 – 653. Morris R (1989) Synaptic plasticity and learning: selective impairment of learning in rats and blockade of long-term potentiation in vivo by the N-methyl-D-aspartate antagonist AP5. J Neurosci 9:3040–3057. Olney JW (1994) New mechanisms of excitatory transmitter neurotoxicity.
J Neural Transm Suppl 43:47–51. Pavlov I (1927) Conditioned re? exes. Oxford: Oxford University Press. Paxinos G, Watson C (1997) The rat brain in stereotaxic coordinates, 3rd ed. New York: Academic. Pussinen R, Niememinen S, Koivisto E, Haapalinna A, Riekkinen S, Sirvio J (1997) Enhancement of intermediate-term memory by an alpha-1 agonist or a partial agonist at the glycine site of the NMDA receptor. Neurobiol Learn Mem 67:69 –74. Quartermain D, Mower J, Rafferty M, Herting R, Lanthorn T (1994) Acute but not chronic activation of the NMDA-coupled glycine receptor with D-Cycloserine facilitates learning and retention. Eur J Pharmacol 257:7–12.
Ressler KJ, Rothbaum BO, Tannebaum L, Anderson P, Graap K, Zimand E, Hodges L, Davis M (2004) Cognitive enhancers as adjuncts to psychotherapy: Use of D-Cycloserine in phobics to facilitate extinction of fear. Arch Gen Psychiatry 61:1136 –1144. Richardson R, Ledgerwood L, Cranney J (2004) Facilitation of fear extinction by D-cycloserine: Theoretical and clinical implications. Learn Mem 11:510 –516. Rogan MT, Staubli UV, LeDoux JE (1997) Fear conditioning induces associative long-term potentiation in the amygdala. Nature 390:604–607. Schafe GE, LeDoux JE (2000) Memory consolidation of auditory Pavlovian fear conditioning requires protein synthesis and protein kinase A in the amygdala. J Neurosci 20:8177– 8187. Schwartz B, Hashtroudi S, Herting R, Schwartz P, Deutsch S (1996) D-Cycloserine enhances implicit memory in Alzheimer patients.
Neurology 46:420 – 424. Staubli U, Thibaoult O, DiLorenzo M, Lynch G (1989) Antagonism of NMDA receptors impairs acquisition but not retention of olfactory memory. Behav Neurosci 103:54 – 60. Thompson L, Disterhoft J (1997) Age- and dose-dependent facilitation of associative eyeblink conditioning by D-cycloserine in rabbits. Behav Neurosci 111:1303–1312. Tsai G, Falk W, Gunther J (1998) A preliminary study of D-cycloserine treatment in Alzheimer’s disease. J Neuropsychiatry Clin Neurosci 10:224–226. Tsai G, Falk W, Gunther J, Coyle J (1999) Improved cognition in Alzheimer’s disease with short-term D-cycloserine treatment. Am J Psychiatry 156:467– 469. an Berckel B, Evenblij C, van Loon B, Maas M, van der Geld M, Wynne J, van Ree J, Kahn R (1999) D-cycloserine increases positive symptoms in chronic schizophrenic patients when administered in addition to antipsychotics: a double-blind, parallel, placebo-controlled study. Neuropsychopharmacology 21:203–210. Walker DL, Ressler KJ, Lu KT, Davis M (2002) Facilitation of conditioned fear extinction by systemic administration or intra-amygdala infusion of D-cycloserine as assessed with fear-potentiated startle. J Neurosci 22:2343–2351. Davis M, Falls WA, Gewirtz J (2000) Neural systems involved in fear inhibition: Extinction and conditioned inhibition. In: Contemporary issues in modeling psychopathology (Myslobodsky M, Weiner I, eds), pp 113–142. Boston: Kluwer Academic Publishers.
Falls W, Davis M (1992) Extinction of fear-potentiated startle: blockade by infusion of an NMDA antagonist into the amygdala. J Neurosci 12:854 – 863. Falls W, Davis M (1995) Behavioral and physiological analysis of fear inhibition: extinction and conditioned inhibition. In: Neurobiological and clinical consequences of stress: From normal adaptation to PTSD (Friedman MJ, Charney DS, Deutch AY, eds), pp 177–202. Philadelphia: Lippincott-Raven. Fanselow MS, LeDoux JE (1999) Why we think plasticity underlying pavlovian fear conditioning occurs in the basolateral amygdala. Neuron 23:229 –232. Flood J, Baker M, Davis J (1990) Modulation of memory processing by glutamic acid receptor antagonists. Brain Res 521:197–202.
Flood J, Morley J, Lanthorn T (1992) Effect on memory processing by D-cycloserine, an agonist of the NMDA/glycine receptor. Neurosci Lett 146:215–218. Gewirtz JC, Davis M (1997) Second-order fear conditioning prevented by blocking NMDA receptors in amygdala. Nature 388:471– 474. Goff D, Tsai G, Levitt J, Amico E, Manoach D, Schoenfeld D, Hayden D, McCarley R, Coyle J (1999) A placebo-control trial of D-cycloserine added to conventional neuroleptics in patients with schizophrenia. Arch Gen Psychiatry 56:21–27. Javitt D, Aylberman I, Zukin S, Heresco L, Lindenmayer J (1994) Amelioration of negative symptoms in schizophrenia by glycine. Am J Psychiatry 151:1234 –1236.
Kehoe E, Macrae M, Hutchinson C (1996) MK-801 protects conditioned response from extinction in the rabbit nictitating membrane preparation. Psychobiology 24:127–135. Kim J, DeCola J, Landeira-Fernandez J, Fanselow M (1991) N-methylD-aspartate receptor antagonist APV blocks acquisition but not expression of fear conditioning. Behav Neurosci 105:126 –133. Konorski J (1948) Conditioned re? exes and neuronal organization. London: Cambridge University Press. Land C, Riccio D (1999) D-cycloserine: Effects on long-term retention of a conditioned response and on memory for contextual attributes. Neurobiol Learn Mem 72:158 –168. Lawler BA, Davis KL (1992) Does modulation of glutamatergic function represent a viable therapeutic strategy in Alzheimer’s disease?
Biol Psychiatry 31:337–350. Ledgerwood L, Richardson R, Cranney J (2003) Effects of D-cycloserine on extinction of conditioned freezing. Behav Neurosci 117:341–349. Ledgerwood L, Richardson R, Cranney J (2004) D-cycloserine and the facilitation of extinction of conditioned fear. Consequences for reinstatement. Behav Neurosci 118:505–513. Lee H, Kim J (1998) Amygdalar NMDA receptors are critical for new fear learning in previously fear-conditioned rats. J Neurosci 18:8444–8454. Lin CH, Yeh SH, Lin CH, Lu KT, Leu TH, Cheung WC, Gean PW (2001) A role for the PI-3 kinase signaling pathway in fear conditioning and synaptic plasticity in the amygdala. Neuron 31:845–851.
Lin CH, Yeh SH, Lu HY, Gean PW (2003) The similarities and diversities of signal pathways leading to consolidation of conditioning and consolidation of extinction of fear memory. J Neurosci 23:8310–8317. Lu KT, Walker DL, Davis M (2001) Mitogen-activated protein kinase cascade in the basolateral nucleus of amygdala is involved in extinction of fear-potentiated startle. J Neurosci 21(RC162):1–5. Maren S, Aharonov G, Stote D, Fanselow M (1996) N-methyl-Daspartate receptors in the basolateral amygdala are required for both acquisition and expression of conditional fear in rats. Behav Neurosci 110:1365–1374. Matsuoka N, Aigner T (1996) D-cycloserine, a partial agonist at the glycine site coupled to N-methyl-D-aspartate receptors, improves (Accepted 1 April 2005)