Complex effects of NMDA receptor antagonist APV in the basolateral amygdala on acquisition of two-way avoidance reaction and long-term fear memory

Abstract

Although much has been learned about the role of the amygdala in Pavlovian fear conditioning, relatively little is known about an involvement of this structure in more complex aversive learning, such as acquisition of an active avoidance reaction. In the present study, rats with a pretraining injection of the N-methyl-D-aspartate (NMDA) receptor antagonist, 2-amino-5-phosphonopentanoic acid (APV), into the basolateral amygdala (BLA) were found to be impaired in two-way active avoidance learning. During multitrial training in a shuttle box, the APV-injected rats were not different from the controls in sensitivity to shock or in acquisition of freezing to contextual cues. However, APV injection led to impaired retention of contextual fear when tested 48 h later, along with an attenuation of c-Fos expression in the amygdala. These results are consistent with the role of NMDA receptors of the BLA in long-term memory of fear, previously documented in Pavlovian conditioning paradigms. The APV-induced impairment in the active avoidance learning coincided with deficits in directionality of the escape reaction and in attention to conditioned stimuli. These data indicate that normal functioning of NMDA receptors in the basolateral amygdala is required during acquisition of adaptive instrumental responses in a shuttle box but is not necessary for acquisition of short-term contextual fear in this situation.

Figure 1

Cannula placements. (A) Photomicrograph of Nissl-stained section of left amygdala in rat injected with APV. (B) Schematic representation of the same section. Inserted squares represent the areas shown on Figure 8 for examples of c-Fos immunostaining. Scale bar, 1 mm; (to) tractus opticus; (ec) external capsule; (ast) amigdalostriatal area. (C) Cannula tip placements from rats infused with vehicle (open arrows) or 5 μg of APV (filled arrows). Only rats with cannula tips at or within the boundaries of basolateral nuclei were included in the data analysis.

Figure 2

An acquisition of the two-way active avoidance reaction in control (top panel), vehicle (middle panel), and APV group (low panel). Thick lines represent the number of avoidance reactions to acoustic (Noise, N) or visual (Darkness, D) conditioned stimuli in consecutive blocks of trials. Thin lines represent the number of nondirectional responses that were observed as a chaotic jumping between the walls or to the ceiling of the shocked compartment before entering the opposite safe compartment. A decrease in the number of nondirectional responses to the end of the session (as it can be seen in controls but not in APV group) indicates an acquisition of escape reaction with an appropriate directionality and facilitates the learning of the avoidance reaction. Data are expressed as means and SEM.

Figure 3

Effect of APV on the latency of (A) directional and (B) nondirectional escape responses. In all groups of rats, the directional escape (defined as an escape from the shocked compartment with the shortest trajectory) was characterized by significantly shorter latencies than the nondirectional escape response (Kruskal-Wallis ANOVA, H_{(1, 1122)} = 764.28, p < 0.0001). APV-injected rats were significantly different from the control groups only in the latency of nondirectional escape reactions (see text). Because there was no significant effect of modality of CS on escape latency, data are expressed as mean escape latencies for both conditioned stimuli.

Figure 4

Effect of APV injection on the preparatory reaction during (A) Darkness or (B) Noise conditioned stimuli or during (C) the intertrial interval. The reaction of orientation of the body or head to the opening was significantly suppressed in the rats of the APV group during presentation of the acoustic CS as well as during the intertrial interval. Values represent the mean ± SEM. (*) Significant differences from the APV group in separate blocks of trials as a result of Newman-Keuls post hoc test, p < 0.05.

Figure 5

Attention reaction to (A) Darkness or (B) Noise conditioned stimuli was suppressed after APV injection in the basolateral nucleus of the amygdala. The effect of APV was observed for conditioned stimuli of both modalities and became more pronounced in the course of the session.

Figure 6

Freezing to contextual cues measured before, during (1–5 blocks of trials), and 48 h after training. APV-injected rats showed a deficit in freezing only when tested 48 h after the training.

Figure 7

Freezing during Darkness or Noise conditioned stimuli in (A) control, (B) vehicle, and (C) APV groups. Freezing during the CS is expressed relative to the level of freezing during the intertrial interval (taken as 100% in every block of trials). Only trials that were not terminated by avoidance reaction were analyzed (see text).

Figure 8

Representative examples of c-Fos immunostaining in the lateral, central, and medial nuclei of amygdala in control rats (naive, handled, and handled and injected with vehicle) and in trained rats injected with vehicle or APV. For schematic location of areas shown, see Figure 1B. (to) Tractus opticus; (ast) amygdalostriatal area; (ec) external capsule.

Figure 9

c-Fos expression in basolateral complex (BLC), central (CeA), cortical (CoA), and medial (MeA) amygdalar nuclei in control and experimental groups of rats. The data represent the average number of c-Fos immunostained cells of left and right nuclei except the handled group, where the data from right (Handled) and left (Handled_Vehicle) nuclei were presented separately. (*) Significant differences between right and left (vehicle-injected) sides in group of handled rats; (#) significant differences from naive rats; (+) significant differences from Trained-Vehicle group, p < 0.05 in each case (Newman-Keuls post hoc tests; see text for results of ANOVAs).