Dorsal raphe to basolateral amygdala corticotropin-releasing factor circuit regulates cocaine-memory reconsolidation

Abstract

Environmental stimuli elicit drug craving and relapse in cocaine users by triggering the retrieval of strong cocaine-related contextual memories. Retrieval can also destabilize drug memories, requiring reconsolidation, a protein synthesis-dependent storage process, to maintain memory strength. Corticotropin-releasing factor (CRF) signaling in the basolateral amygdala (BLA) is necessary for cocaine-memory reconsolidation. We have hypothesized that a critical source of CRF in the BLA is the dorsal raphe nucleus (DR) based on its neurochemistry, anatomical connectivity, and requisite involvement in cocaine-memory reconsolidation. To test this hypothesis, male and female Sprague-Dawley rats received adeno-associated viruses to express Gi-coupled designer receptors exclusively activated by designer drugs (DREADDs) selectively in CRF neurons of the DR and injection cannulae directed at the BLA. The rats were trained to self-administer cocaine in a distinct environmental context then received extinction training in a different context. Next, they were briefly re-exposed to the cocaine-predictive context to destabilize (reactivate) cocaine memories. Intra-BLA infusions of the DREADD agonist deschloroclozapine (DCZ; 0.1 mM, 0.5 µL/hemisphere) immediately after memory reactivation attenuated cocaine-memory strength, relative to vehicle infusion. This was indicated by a selective, DCZ-induced and memory reactivation-dependent decrease in drug-seeking behavior in the cocaine-predictive context in DREADD-expressing males and females at test compared to respective controls. Notably, BLA-projecting DR CRF neurons that exhibited increased c-Fos expression during memory reconsolidation co-expressed the glutamatergic neuronal marker, vesicular glutamate transporter 3. Together, these findings suggest that the DR_CRF → BLA circuit is engaged to maintain cocaine-memory strength after memory destabilization, and this phenomenon may be mediated by DR CRF and/or glutamate release in the BLA.

Fig. 1. Chemogenetic DR_CRF → BLA circuit inhibition prior to memory reconsolidation weakens cocaine-memory strength.

A Experiment 1 timeline. The Gi-coupled DREADD hM4Di fused with mCherry (Gi) or mCherry alone (control; Ctrl) was expressed in dorsal raphe (DR) corticotropin-releasing factor (CRF) neurons, and guide cannulae were directed bilaterally at the basolateral amygdala (BLA). Rats then received cocaine self-administration training in a salient environmental context and extinction training in a different context. Next, they received 15-min re-exposure to the cocaine-predictive context (i.e., memory reactivation) followed by bilateral intra-BLA microinjections of vehicle (VEH) [(Gi) VEH; n = 10 males, 11 females)] or the DREADD agonist deschloroclozapine (DCZ) [(Ctrl) DCZ: n = 8 males, 8 females; (Gi) DCZ: n = 10 males, 8 females)]. Non-reinforced lever responding in the extinction and cocaine-predictive contexts was assessed 24 and 72 h later, respectively. B Schematic depicting the cell- and circuit-specific chemogenetic approach. C Representative images of hM4Di-mCherry expression in the DR (left), overlap between hM4Di-mCherry expression and CRF-immunoreactivity in DR cell bodies (middle), and hM4Di-mCherry expression in DR CRF terminals within the BLA and adjacent central amygdaloid nucleus (CeA). D Schematic depicting injection cannula placements in the BLA. The numbers between the schematics indicate AP distance from Bregma in millimeters. E Cocaine infusions and active- and inactive-lever responses (mean/2 h ±SEM) during cocaine self-administration training (last 10 sessions) and extinction training in males and females. F Active-lever responses during the memory-reactivation session (mean/15 min ±SEM) immediately prior to intra-BLA treatment. G Active-lever responses (mean/2 h ±SEM) upon first re-exposure to the extinction and cocaine-predictive contexts after intra-BLA treatment. H Active-lever response latency (mean ± SEM) in the cocaine-predictive context at test. I Time course of active-lever responses (mean/20 min ±SEM) in the cocaine-predictive context at test. Symbols: ANOVA ^‡time main (underlined) or simple main effect (see details for pairwise comparisons in Results); ^#context simple main effect; *treatment group simple main effect. All ps ≤ 0.05. Aq cerebral aqueduct.

Fig. 2. Chemogenetic inhibition of the DR_CRF → BLA circuit six hours after memory retrieval does not alter cocaine-memory strength.

A Experiment 2 timeline. The Gi-coupled DREADD hM4Di fused with mCherry (Gi) was expressed in dorsal raphe (DR) corticotropin-releasing factor (CRF) neurons and guide cannulae were directed at the basolateral amygdala (BLA). Rats then received cocaine self-administration training in a distinct context and extinction training in a different context. Next, they received 15-min re-exposure to the cocaine-predictive context (i.e., memory reactivation). Six hours later, the rats received bilateral intra-BLA injections of vehicle (VEH) [(Gi) VEH, n = 6 males, 5 females)] or the DREADD agonist deschloroclozapine (DCZ) [(Gi) DCZ; n = 5 males, 5 females)]. Non-reinforced lever responding in the extinction and cocaine-predictive contexts was assessed 24 and 72 h later, respectively. B Schematic depicting injection cannula placements in the BLA. The numbers between the schematics indicate AP distance from Bregma in millimeters. C Cocaine infusions and active- and inactive-lever responses (mean/2 h ±SEM) during cocaine self-administration training (last 10 sessions) and extinction training in males and females. D Active-lever responses during the memory-reactivation session (mean/15 min ±SEM) 6 h prior to intra-BLA treatment. E Active-lever responses (mean/2 h ±SEM) upon first re-exposure to the extinction and cocaine-predictive contexts after delayed intra-BLA treatment. F Active-lever response latency (mean ± SEM) in the cocaine-predictive context at test. G Time course of active-lever responses (mean/20 min ±SEM) in the cocaine-predictive context at test. Symbols: ^‡time main (underlined) or simple main effect (see details for pairwise comparisons in Results); ^#context main effect; All ps < 0.05.

Fig. 3. Chemogenetic inhibition of the DR_CRF → BLA circuit attenuates Zif268 expression in the BLA.

A Experiment 3 timeline. A subset of control rats expressing hM4Di fused with mCherry (Gi) or mCherry alone (Ctrl) in the DR received a 15-min memory-reactivation session ~6 days after their last test session in Experiment 1 or 2. Immediately after the memory-reactivation session, vehicle (VEH) or the DREADD agonist deschloroclozapine (DCZ) treatment was infused bilaterally into the BLA [(Gi) VEH: n = 6 males, 7 females; (Ctrl) DCZ: n = 2 males, 7 females; (Gi) DCZ: n = 4 males, 5 females)]. Brain tissue was collected 2 h later to assess Zif268 expression in the BLA and adjacent CeA during reconsolidation. B Representative images of hM4Di-mCherry-expressing DR CRF terminals and Zif268-immunoractivity in the BLA and adjacent central amygdaloid nucleus (CeA). Analyses were limited to images of the BLA and CeA without significant tissue damage. C Representative images of Zif268-immunoreactive BLA nuclei in the proximity of mCherry-expressing DR terminals illustrating group effects in the BLA. D Representative images of Zif268-immunoreactive CeA nuclei in the proximity of mCherry-expressing DR terminals illustrating sex effects in the CeA. E Active-lever responses during the memory-reactivation session (mean/15 min ± SEM) prior to intra-BLA treatment. F Density of Zif268-immunoreactive nuclei (mean number of nuclei/mm² ± SEM) in the BLA collapsed across sex and in the CeA in males and females. Symbols: ANOVA *treatment group main effect; sex main effect (underlined). All ps < 0.05.

Fig. 4. BLA-projecting DR CRF neurons activated during cocaine-memory reconsolidation co-express vGlut3.

A Experiment 4 timeline. GFP was expressed in basolateral amygdala-projecting (BLA) corticotropin-releasing factor (CRF) cell bodies in the dorsal raphe (DR) using a combinatorial viral approach. Rats then received cocaine self-administration training in a distinct context and extinction in a different context. Next, they received 15-min re-exposure to the cocaine-predictive context to reactive cocaine memories (MR; n = 4 males, 4 females) or remained in their home cages (No-MR; n = 4 males, 4 females). Brain tissue was collected 2 h later to capture c-Fos expression approximately 30 min into memory reconsolidation. B Cocaine infusions and active- and inactive-lever responses (mean/2 h ± SEM) during cocaine self-administration training (last 10 sessions), extinction training sessions, and memory reactivation, collapsed across sex (but see Fig. S5). C Density of GFP-expressing (BLA-projecting CRF) cell bodies in the DR (mean number of cell bodies/mm² ± SEM) in the No-MR and MR groups. D Representative images of GFP- expression in BLA-projecting DR CRF neurons and in their axon terminals in the BLA from the same subject. D, inset Schematic representation of the DR subregions in which cell types were analyzed. E Representative image of overlap between GFP-expression and CRF-immunoreactivity in DR neurons. F, upper panel Density of GFP-expressing DR neuronal populations that were c-Fos-immunoreactive following No-MR or MR (mean number of cell bodies/mm² ± SEM) collapsed across DR subregions and sex where sex differences were not observed. F, inset Pie charts illustrating the proportion of GFP-expressing cell types that were c-Fos-immunoreactive following No-MR and MR. F, lower panel Representative images of c-Fos-immunoreactive GFP-expressing cell types observed in the DR. G, upper panel Density of GFP-non-expressing (GFP^-) cell types that were c-Fos-immunoreactive following No-MR and MR (mean number of cell bodies/mm² ± SEM) collapsed across DR subregions and sex where sex differences were not observed. G, inset Total c-Fos expression (mean number of cell bodies/mm² ± SEM) following No-MR or MR. Black segment represents the density of c-Fos-immunoreactive GFP-expressing neurons collapsed across cell type; other segments represent the density of c-Fos-immunoreactive GFP-non-expressing cell types. (G, lower panel) Representative images of c-Fos-immunoreactive GFP-non-expressing cell types observed in the DR. Symbols: ANOVA ^‡time main effect; *group main effect; sex main effect; All p < 0.05. Aq cerebral aqueduct, dDR dorsal subregion of dorsal raphe, lDR lateral subregion of dorsal raphe, vDR ventral subregion of dorsal raphe, PAG periaqueductal gray.