Hippocampal fear engrams modulate ethanol-induced maladaptive contextual generalization in mice

Abstract

The compounding symptomatology of alcohol use disorder (AUD) and co-occurring mental health disorders gives rise to interactions of maladaptive neurobiological processes, the etiology of which are elusive. Here, we devised an optogenetic strategy aimed at rescuing maladaptive fear processing in male c57BL/6 mice that underwent a chronic ethanol administration and forced abstinence paradigm. In the first experiment, we confirmed that fear acquisition and maladaptive contextual generalization was potentiated in ethanol-exposed mice during fear conditioning and exposure to a novel environment, respectively. In the second experiment, using an activity-dependent tet-tag system, we labeled the neural ensemble selectively activated by contextual fear conditioning in the dorsal hippocampus with an inhibitory opsin to attenuate behavioral dysfunctions resulting from ethanol exposure. We found that acute optogenetic perturbations during exposure to a novel environment suppressed maladaptive generalization in ethanol-exposed mice. These results provide further evidence for a crucial link between ethanol exposure and impaired fear memory processing by providing cellular and behavioral insights into the neural circuitry underlying AUD and maladaptive fear processing.

Keywords: addiction; alcohol; engram; fear conditioning; generalization; hippocampus; optogenetics; withdrawal.

Figure 1 –. Ethanol-induced maladaptive contextual generalization.

(a) Schematic of viral strategy. A viral cocktail of AAV9-c-Fos-tTA and AAV9-TRE-eYFP was infused into the dorsal dentate gyrus (dDG) for activity-dependent transcription of eYFP. Schematic created with BioRender. (b) Behavioral design for ethanol (EtOH) or saline exposure, forced abstinence, fear conditioning, and recall or generalization tests. Schematic created with BioRender. (c) Representative histology for activity-dependent tagging of contextual fear engrams. Scale bar = 100 μm. (d) Fear acquisition in Context A. No group differences in freezing during fear acquisition were observed (two-way RM ANOVA, treatment F_(1,174)= 3.648, P=0.0661). Main effect of time (two-way RM ANOVA, time F_(6,174)=99.15, P<0.001). Time x Treatment interaction (two-way RM ANOVA, time x treatment F_(6,174)=3.440, P=0.0031). (e) Recall or generalization tests in neutral Context B. Main effects and interaction of context and treatment (two-way ANOVA, treatment F_(1,27)= 13.51, P=0.0010, context F_(1,27)= 4.907, P=0.0354, context x treatment F_(1,27)= 4.416, P=0.0451). EtOH-exposed mice froze significantly higher than the saline control mice in novel Context B (Tukey’s HSD p=0.0023). Saline-control mice froze significantly less in Context B relative to Context A (Tukey’s HSD p=0.0276), while the EtOH-exposed mice showed no difference across contexts (Tukey’s HSD p=0.9998). (f) Quantified normalized overlaps. See main text for normalization process. No significant differences across treatment groups or contexts in terms of percent of reactivated cells from the tagged putative engram (two-way ANOVA, context F_(1,16)=0.8946, p=0.3585, treatment F_(1,16)=0.1045, p=0.7507). (g) Representative histology for dual-labelled engram cells reactivated during recall or generalization. Scale bar = 100 μm.

Figure 2 –. ArchT inhibition of a contextual fear memory prevents ethanol-induced maladaptive generalization.

a) Schematic of viral strategy. A viral cocktail of AAV9-c-Fos-tTA and AAV9-TRE-ArchT or AAV9-TRE-eYFP was infused into the dorsal dentate gyrus (dDG) for activity-dependent transcription of ArchT or eYFP. Schematic created with BioRender. (b) Behavioral design for ethanol (EtOH) or saline exposure, forced abstinence, fear conditioning, and recall or generalization tests. Schematic created with BioRender. (c) Representative histology for activity-dependent tagging of contextual fear engrams. Scale bar = 100 μm. (d) Optogenetic inhibition during recall or generalization tests. Main effects of context, virus, and treatment (three-way ANOVA, context F_(1,64)=174.4, p<0.0001; virus F_{(1, 64)}=6.963, p=0.0104; treatment F_(1,64)=7.3, p=0.0085). Strong trending interactions of context x virus (F_(1,64)=3.860, p=0.0538), and context x virus x treatment (F_(1,64)=3.428, p=0.0687). No significant difference in freezing between ArchT EtOH-exposed mice and ArchT saline-control mice in Context B (Tukey’s post-hoc p=0.9997), and ArchT EtOH-exposed mice froze at a significantly lower level than the eYFP EtOH-exposed mice in Context B (Tukey’s post-hoc, p=0.0089), indicating optogenetic inhibition was able to significantly decrease freezing in EtOH-exposed mice, preventing mice from overgeneralizing to the neutral context. (e) Quantified normalized overlaps. See main text for normalization process. Analysis revealed a main effect of context (three-way ANOVA, Context F_{(1, 42)}=5.180, p=0.0280). No significant interactions or individual group differences. (f) Representative histology for dual-labelled engram cells reactivated during recall or generalization. Scale bar = 100 μm.