RNA editing of the 5-HT2C receptor in the central nucleus of the amygdala is involved in resilience behavior

Abstract

Post-traumatic-stress-disorder (PTSD) is a stress-related condition that may develop after exposure to a severe trauma-event. One of the core brain areas that is considered to be a key regulatory region of PTSD is the amygdala. Specifically, the central amygdala (CeA) is involved in emotion processing and associative fear learning memory, two main circuits involved in PTSD. Long term dysregulation of trauma-related emotional processing may be caused by neuroadaptations that affect gene expression. The adenosine-(A) to-inosine (I) RNA editing machinery is a post-transcriptional process that converts a genomic encoded A to I and is critical for normal brain function and development. Such editing has the potential to increase the transcriptome diversity, and disruption of this process has been linked to various central nervous system disorders. Here, we employed a unique animal model to examine the possibility that the RNA editing machinery is involved in PTSD. Detection of RNA editing specifically in the CeA revealed changes in the editing pattern of the 5-HT2C serotonin receptor (5-HT2CR) transcript accompanied by dynamic changes in the expression levels of the ADAR family enzymes (ADAR and ADARb1). Deamination by ADAR and ADARb1 enzymes induces conformational changes in the 5-HT2CR that decrease the G-protein-coupling activity, agonist affinity, and thus serotonin signaling. Significantly, a single intra-CeA administration of a 5-HT2CR pharmacological antagonist produced a robust alleviation of PTSD-like behaviors (that was maintained for three weeks) as well as single systemic treatment. This work may suggest the way to a new avenue in the understanding of PTSD regulation.

Fig. 1. Behavioral response in the PTSD model.

(A) Schematic depiction of the experimental model. The animals began the experimental procedure after two weeks of acclimation and habituation to the home cage and the open field. Baseline response to the three behavioral scenarios was monitored on day 1. Day 7: Exposure to ‘trauma’ (Initial exposure). Day 14: first exposure to the ‘trauma’s’ reminder (‘first reminder’). Day 35: second exposure to the ‘trauma’s reminder (‘second reminder’). Day 56: third exposure to the ‘trauma’s reminder (‘third reminder’). Day 77: fourth exposure to the ‘trauma’s reminder (‘fourth reminder’). (B) Freezing behavior during: exploration, (C) social interaction, and (D) hyperarousal tests. After exposure to trauma and the reminders, susceptible but not resilient animals showed an increase from baseline in freezing behavior, i.e., incubation of fear over time in all three behavioral tests (exploration: *p < 0.05 and ***p < 0.001 susceptible vs. resilient; social interaction: ***p < 0.001 susceptible vs. resilient and hyperarousal: *p < 0.05 and ***p < 0.001 susceptible vs resilient). Data presented as mean ± SEM; n = 26–35 per group. (E–G) Pearson’s product-moment correlation between the following tests: (E) exploration×social interaction, (F) hyperarousal×exploration, and (G) hyperarousal×social interaction (r > 0.33; *p < 0.005).

Fig. 2. Targeted sequencing by mmPCR revealed differential RNA editing of the 5-HT2CR in the CeA of susceptible compared to the resilient group.

(A) The ten RNA editing sites common to both experimental groups that passed the defined criteria. No significant differences were found between the experimental groups (p > 0.05) in RNA editing levels in the sites of the 5-HT2CR, Grik2, CACNA1D, and Cog3. (B) Relative abundance of the 5-HT2CR mRNA variants produced by RNA editing. Data show a relatively higher abundance of the VNV isoform (*p < 0.05) in the resilient animals than in the susceptible group. Data presented as mean + SEM; n = 7–13 per group. (C) mRNA expression levels of the 5-HT2CR did not reveal a significant change between the experimental groups (p > 0.05). (D, F) qPCR analysis revealed significantly higher expression levels of ADAR (D) and ADARb1 (F) in the resilient group compared to the susceptible group (*p < 0.05). Data presented as mean + SEM; n = 6 7 per group. (E, G–I) Protein expression of ADAR (E, H) and ADARb1 (F, I) analyzed by western blot with specific antibodies against ADAR, ADARb1 and β-Actin as a loading control revealed a significantly elevated expression of these proteins in the resilient animals compared to the susceptible group (*p < 0.05). Data presented as mean + SEM; n = 3 per group. (J) Pearson’s correlation between ADAR proteins and freezing behavior at the second reminder time point presenting a high correlation between: ADAR and exploration (r = −0.7, p = 0.04), ADAR and hyperarousal (r = −0.7 and p = 0.02), and ADARb1 and social interaction (r = −0.6 and p = 0.07).

Fig. 3. Effect of RS-102221 antagonist treatment on PTSD-like behaviors.

Susceptible and resilient animals received one injection of RS-102221 or control vehicle 15 min before the third trauma reminder. Freezing behavior was assessed following the third reminder and fourth reminder in (A) exploration, (B) social interaction, and (C) hyperarousal tests. Bars represent freezing levels after the third and the fourth reminders showing a significant attenuation in freezing behavior of susceptible-RS-treated animals vs. susceptible- control-treated and all of the resilient groups (exploration: third reminder: **p < 0.01 for susceptible-control-treated vs. susceptible-RS-treated and resilient-control-treated animals, *p < 0.05 susceptible-control-treated vs. resilient-RS-treated animals and *p < 0.05 for susceptible-RA-treated vs. resilient-SAM and resilient- RA- treated groups. Fourth reminder: *p < 0.05 for susceptible-RA-treated vs. all other experimental animals; social interaction: third reminder: **p < 0.001 for susceptible-control-treated vs. susceptible-RS and resilient-RS groups and ^#p < 0.05 for susceptible-control-treated vs. resilient-RS treated group. Fourth reminder: **p < 0.01 for susceptible-control-treated vs. all resilient groups and ^#p < 0.05 for susceptible-RA-treated vs. susceptible-control-treated animals; hyperarousal: third reminder: ***p < 0.001 for susceptible-control-treated vs. susceptible-RS-treated and resilient-control-treated groups, ^##p < 0.01 susceptible-control-treated vs. resilient-RS-treated animals. Fourth reminder: **p < 0.01 for susceptible-control-treated vs. all other experimental animals. Data presented as mean ± SEM; n = 4-5 per group. (D–F) Fold change of freezing behavior in the susceptible-treated-groups in the third and fourth reminders compared to the freezing levels in the second reminder in (D) exploration, (E) social interaction, and (F) hyperarousal behavioral test. (G–I) Pearson’s product-moment correlation between the third and the fourth reminders in the following tests: (G) exploration (H) social interaction and (I) hyperarousal (r > 0.45;*p < 0.05). (J) Left: Image showing PI staining within the CeA. The overlay of the amygdala sub-nuclei demonstrates that the cannula placement is specific to the CeA. The arrow points to the tip of the injection needle. Right: Corresponding section from rat stereotaxic atlas (Paxinos and Watson). (K–M) Susceptible animals received one systemic injection of RS-102221 or control vehicle 30 min before the third trauma reminder. Freezing behavior was assessed following the third reminder in (K) exploration (***p < 0.001) (L) social interaction (**p < 0.01) (M) hyperarousal (*p < 0.05) tests comparing susceptible group treated with RS-102221 to susceptible group treated with vehicle control. Data presented as mean + SEM; n = 4-5 per group.