Single-unit activity in the medial prefrontal cortex during immediate and delayed extinction of fear in rats

Abstract

Delivering extinction trials minutes after fear conditioning yields only a short-term fear suppression that fully recovers the following day. Because extinction has been reported to increase CS-evoked spike firing and spontaneous bursting in the infralimbic (IL) division of the medial prefrontal cortex (mPFC), we explored the possibility that this immediate extinction deficit is related to altered mPFC function. Single-units were simultaneously recorded in rats from neurons in IL and the prelimbic (PrL) division of the mPFC during an extinction session conducted 10 minutes (immediate) or 24 hours (delayed) after auditory fear conditioning. In contrast to previous reports, IL neurons exhibited CS-evoked responses early in extinction training in both immediate and delayed conditions and these responses decreased in magnitude over the course of extinction training. During the retention test, CS-evoked firing in IL was significantly greater in animals that failed to acquire extinction. Spontaneous bursting during the extinction and test sessions was also different in the immediate and delayed groups. There were no group differences in PrL activity during extinction or retention testing. Alterations in both spontaneous and CS-evoked neuronal activity in the IL may contribute to the immediate extinction deficit.

Figure 1. Anatomical placement of tetrodes.

(A) Coronal sections representing all of the tetrode placements included in the data analysis. (B) Serial sections from one DELAY animal showing tetrodes in both the IL and PrL; these placements are shown as filled gray circles in (A).

Figure 2. Percentage of freezing (mean ± SEM) during the baseline (BL), extinction (EXT), and test (TEST) sessions in rats undergoing immediate (IMMED) or delayed (DELAY) extinction.

Freezing was significantly higher in IMMED rats compared to their DELAY controls prior to extinction (pre-period in EXT session) and during the retention test (p<0.05). Data were averaged across the 10-min pre-CS period or 10 CS trials (first and last 10 trials for early and late EXT periods).

Figure 3. Peri-event time histograms illustrating CS-evoked activity in the IL (A) and PrL (B) during behavioral training.

The number of neurons contributing to each average (Z>0 within 200 ms) is indicated in the panels. (A) For the IL, there was no difference between the IMMED and DELAY groups in firing to tones during the baseline (BL) session (p>0.05) and early extinction (EXT). During the extinction session, only DELAY rats decreased their firing to the CS (p<0.05). Firing to tone CSs was significantly higher in IMMED rats compared to DELAY rats during the test session (TEST, p<0.05). (B) For the PrL, there was no difference between groups in any of the behavioral phases. The 2-sec tone CS period is shaded in gray.

Figure 4. Neuronal bursting in the IL (A) and the PrL (B) divisions of the medial prefrontal cortex.

(A) The frequency of IL bursting (mean ± SEM) during the 2-s CS (filled circles) and the 1-min ITI (moving average of 5 sec) is shown for rats in the IMMED and DELAY groups; the number of neurons showing bursting in each session is indicated in the panels. During extinction, bursting was greater in the DELAY rats. (B) There was no significant difference in PrL bursting between the IMMED and DELAY groups for trial-related bursting across different behavioral phases (all ps>0.05).