Infralimbic activity during REM sleep facilitates fear extinction memory

Abstract

Rapid eye movement (REM) sleep is known to facilitate fear extinction and play a protective role against fearful memories.^1,2 Consequently, disruption of REM sleep after a traumatic event may increase the risk for developing PTSD.^3,4 However, the underlying mechanisms by which REM sleep promotes extinction of aversive memories remain largely unknown. The infralimbic cortex (IL) is a key brain structure for the consolidation of extinction memory.⁵ Using calcium imaging, we found in mice that most IL pyramidal neurons are intensively activated during REM sleep. Optogenetically suppressing the IL specifically during REM sleep within a 4-h window after auditory-cued fear conditioning impaired extinction memory consolidation. In contrast, REM-specific IL inhibition after extinction learning did not affect the extinction memory. Whole-cell patch-clamp recordings demonstrated that inactivating IL neurons during REM sleep depresses their excitability. Together, our findings suggest that REM sleep after fear conditioning facilitates fear extinction by enhancing IL excitability and highlight the importance of REM sleep in the aftermath of traumatic events for protecting against traumatic memories.

Keywords: Fear extinction; REM sleep; closed-loop stimulation; infralimbic cortex; memory consolidation.

Figure 1.. Closed-loop inhibition of IL pyramidal neurons during REM sleep after fear conditioning impairs extinction memory.

(A) Schematic illustrating optogenetic inhibition of IL pyramidal neurons. (B) Experimental protocol for fear conditioning, extinction, and recall with closed-loop inhibition after fear conditioning. Following fear conditioning, mice were transferred to their homecage and subjected for 8 hours to closed-loop inhibition of IL neurons specifically during REM sleep. (C) Top, amount of each brain state and overlap with laser application during sleep recordings after fear conditioning, represented as a pie chart. Bottom left, power spectral density (PSD) of the EEG during REM sleep. Dashed lines indicate the $\theta$ frequency range. Shadings, 95% confidence intervals (CIs). Bottom right, frequency of phasic $\theta$ events during REM sleep episodes with closed-loop inhibition and episodes without inhibition (in baseline recordings without laser). Error bars, 95% CIs; lines, individual mice. Mixed ANOVA, laser, F(1, 21) = 11.3010, P = 0.0030; virus × laser interaction, F(2, 21) = 9.6629, P = 0.0011; pairwise t-tests, eYFP without vs with laser, T(9) = −0.0604, P = 1.0000; iC++ without vs with laser, T(7) = 6.4670, P = 0.0010; iC++ delay without vs with laser, T(5) = 0.4293, P = 1.0000; eYFP, n = 10; iC++, n = 8; iC++ delay, n = 6 mice. **P<0.001. (D) Freezing during auditory CS in fear conditioning, extinction, and recall. A block in extinction and recall comprises 3 consecutive tones. Error bars, ± s.e.m. Conditioning: mixed ANOVA, virus, F(2,21) = 2.1084, P = 0.1464; virus × trials interaction, F(10, 105) = 0.5501, P = 0.8506; extinction: mixed ANOVA, virus, F(2,21) = 0.1935, P = 0.8256; virus × blocks interaction, F(30, 315) = 1.2496, P = 0.1781; recall: one-way ANOVA, F(2, 21) = 5.3869, P = 0.0129; eYFP vs iC++, T(15.0284) = −2.9044, P = 0.0109; eYFP vs iC++ delay, T(10.1088) = 0.6982, P = 0.5008; iC++ vs iC++ delay, T(7.3724) = 3.0606, P = 0.0172; eYFP, n = 10; iC++, n = 8; iC++ delay, n = 6 mice. *P<0.05. See also Figure S1, S2, Video S1 and Table S1.

Figure 2.. Closed-loop inhibition of IL neurons 4 hours after fear conditioning does not impair the extinction memory.

(A) Experimental protocol for fear conditioning, extinction, and recall with closed-loop inhibition during REM sleep starting 4 hours after fear conditioning. (B) Top, amount of each brain state and overlap with laser application during sleep recordings starting 4 hours after fear conditioning. Bottom left, PSD of the EEG during REM sleep. Shadings, 95% CIs. Bottom right, frequency of phasic $\theta$ events during REM sleep episodes with closed-loop inhibition and episodes without inhibition (in baseline recordings without laser). Error bars, 95% CIs; lines, individual mice. Mixed ANOVA, laser, F(1, 14) = 1.7343, P = 0.2090; virus × laser interaction, F(1, 14) = 4.6903, P = 0.0481; pairwise t-tests, eYFP without vs with laser, T(6) = −0.5311, P = 0.6144; iC++ without vs with laser, T(8) = 4.9243, P = 0.0023; eYFP, n = 7; iC++, n = 9 mice. **P<0.01. (C) Freezing during auditory CS in fear conditioning, extinction, and recall. A block in extinction and recall comprises 3 consecutive tones. Error bars, ± s.e.m. Conditioning: mixed ANOVA, virus, F(1, 14) = 0.2306, P = 0.6385; virus × trials interaction, F(5, 70) = 0.5184, P = 0.7615; extinction: mixed ANOVA, virus, F(1, 14) = 1.9228, P = 0.1872; virus × blocks interaction, F(15, 210) = 1.3723, P = 0.1631; recall: t-test, T(10.2230) = 0.5716, P = 0.5799; eYFP, n = 7; iC++, n = 9 mice. See also Figure S3 and Table S1.

Figure 3.. Closed-loop inhibition of IL neurons during REM sleep after fear conditioning reduces neural excitability.

(A) Experimental protocol to assess the impact of REM sleep-dependent inhibition after fear conditioning on the excitability of IL neurons. (B) Freezing during each auditory CS in fear conditioning. Error bars, ± s.e.m. Mixed ANOVA, virus, F(1, 5) = 0.0511, P = 0.8301; virus × trials interaction, F(5, 25) = 0.8744, P = 0.5123; eYFP, n = 3; iC++, n = 4 mice. (C) Location of IL cells recorded in eYFP (left) and iC++ mice (right). (D) Resting membrane potential of recorded cells. Error bars, 95% CIs. T-test, T(66.5855) = −3.4713, P = 0.0009; eYFP, n = 31 cells from 3 mice; iC++, n = 48 cells from 4 mice. **P<0.001. (E) Input resistance of recorded cells. Error bars, 95% CIs. T-test, T(66.1903) = 1.9038, P = 0.0613; eYFP, n = 31 cells from 3 mice; iC++, n = 48 cells from 4 mice. (F) Representative voltage traces of IL pyramidal neurons expressing eYFP or iC++ in response to depolarizing current pulses. (G) Number of action potentials evoked in response to current steps with different intensity. Error bars, ± s.e.m. Mixed ANOVA, virus, F(1, 77) = 10.5011, P = 0.0018; virus × current interaction, F(15, 1155) = 6.7495, P < 0.0001; pairwise t-tests, iC++ vs eYFP, *P<0.05; eYFP, n = 31 cells from 3 mice; iC++, n = 48 cells from 4 mice. See also Table S1.

Figure 4.. Closed-loop inhibition of IL neurons after fear extinction does not impair extinction memory.

(A) Experimental protocol for fear conditioning, extinction, and recall with closed-loop inhibition of IL neurons after fear extinction. (B) Top, amount of each brain state and overlap with laser application during sleep recordings after fear extinction. Bottom left, PSD of the EEG during REM sleep after fear extinction. Shadings, 95% CIs. Bottom right, frequency of phasic $\theta$ events during REM sleep episodes with closed-loop inhibition and episodes without inhibition (in baseline recordings without laser). Error bars, 95% CIs; lines, individual mice. Mixed ANOVA, laser, F(1, 11) = 15.9511, P = 0.0021; virus × laser interaction, F(1, 11) = 6.9256, P = 0.0233; pairwise t-tests, eYFP without vs with laser, T(5) = 0.6775, P = 0.5282; iC++ without vs with laser, T(6) = 5.5515, P = 0.0029; eYFP, n = 6; iC++, n = 7 mice. **P<0.01. (C) Freezing during auditory CS in fear conditioning, extinction, and recall. A block during extinction and recall comprises 3 consecutive tones. Error bars, ± s.e.m. Conditioning: mixed ANOVA, virus, F(1, 11) = 1.6495, P = 0.2254; virus × trials interaction, F(5, 55) = 1.5676, P = 0.1846; extinction: mixed ANOVA, virus, F(1, 11) = 0.7654, P = 0.4003; virus × blocks interaction, F(15, 165) = 0.6362, P = 0.8418; recall: t-test, T(8.7247) = −0.6479, P = 0.5337; eYFP, n = 6; iC++, n = 7 mice. See also Figure S4 and Table S1.