Cannabidiol modulates spontaneous recovery through distinct and conserved transcriptomic signatures in mPFC subregions

Minji Son; Inwoong Song; Prajitha Pradeep; Bomi Chang; Matthew J Girgenti; Cha Young Kim; Jae Cheol Jeong; Boyoung Lee

doi:10.1038/s42003-026-10111-4

. 2026 Apr 28;9:1269. doi: 10.1038/s42003-026-10111-4

Cannabidiol modulates spontaneous recovery through distinct and conserved transcriptomic signatures in mPFC subregions

Minji Son ^1,², Inwoong Song ¹, Prajitha Pradeep ^1,³, Bomi Chang ¹, Matthew J Girgenti ^4,⁵, Cha Young Kim ⁶, Jae Cheol Jeong ^6,^✉, Boyoung Lee ^1,^3,^✉

PMCID: PMC13624265 PMID: 42050066

Abstract

Cannabidiol (CBD) has shown therapeutic potential for post-traumatic stress disorder (PTSD) by reducing spontaneous recovery (SR), yet the underlying mechanisms remain unclear. We examined the effects of CBD on SR in male and female mice and found greater behavioral efficacy in males. To uncover the molecular basis of these effects, we performed transcriptomic profiling of the prelimbic (PL) and infralimbic (IL) subregions of the medial prefrontal cortex (mPFC). We identified distinct and overlapping SR-associated gene sets in each subregion, and CBD shifted the expression of a significant subset of these genes toward non-stressed control levels, including genes involved in retrograde endocannabinoid signaling. Cross-species comparisons with human PTSD transcriptomic datasets revealed conserved gene alterations within each subregion, suggesting shared molecular signatures between mouse SR and human PTSD. These findings highlight region-specific and conserved pathways through which CBD may exert therapeutic effects and provide mechanistic insight to advance CBD-based interventions for PTSD-associated spontaneous recovery.

Subject terms: Transcriptomics, Diseases of the nervous system

Cannabidiol attenuates spontaneous recovery by reshaping distinct and conserved transcriptomic signatures across mPFC subregions, providing mechanistic insight into PTSD-relevant pathways.

Introduction

PTSD is a chronic and debilitating psychiatric condition characterized by intrusive memories, heightened anxiety, sleep disturbances, and maladaptive fear responses^1–4. A core pathological feature of PTSD is the persistence and relapse of fear memories despite extinction-based learning. One prominent manifestation of this relapse is spontaneous recovery (SR), defined as the re-emergence of an extinguished fear response after the passage of time^5,6. SR represents a major obstacle to the long-term efficacy of exposure-based therapies and highlights a critical unmet need for treatments that stabilize extinction memory and prevent fear relapse.

Cannabidiol (CBD), a non-psychoactive phytocannabinoid derived from Cannabis sativa L., has emerged as a promising candidate for the treatment of anxiety-related disorders, including PTSD. This is due to its ability to modulate fear and stress responses without the psychoactive effects of Δ⁹−tetrahydrocannabinol (THC)^7–9. Clinical evidence remains limited but suggests potential benefits of CBD in reducing anxiety and intrusive symptoms in social anxiety disorder and obsessive–compulsive disorder^10–13. Small-scale PTSD studies further indicate that CBD may alleviate nightmares, improve sleep quality, and reduce anxiety and mood disturbances, although these findings are based on very small cohorts¹⁴. In healthy human subjects, CBD has also been shown to modulate aversive memory extinction and fear expression¹⁵. Together, these studies support the therapeutic promise of CBD, while also emphasizing the need for mechanistic investigations.

Preclinical studies provide strong and consistent evidence that CBD directly modulates fear-related behaviors^9,16. Systemic administration of CBD in rats attenuates fear memory retrieval. Moreover, fear conditioning significantly increased the number of c-Fos-positive neurons in the prelimbic (PL) medial prefrontal cortex (mPFC), an effect that was markedly reduced by CBD treatment, implicating PL mPFC activity as a key neural substrate underlying the effects of CBD on fear memory retrieval¹⁶. More interestingly, a local infusion study found opposite effects when CBD was infused into the PL versus the infralimbic (IL) mPFC: PL infusion reduced freezing, while IL infusion increased freezing during retrieval, suggesting differential responses to CBD across mPFC subregions. This differential modulation of fear memory retrieval in the PL and IL was further shown to depend on 5HT1A-mediated neurotransmission and previous stressful experience^17–19. Despite limited evidence regarding CBD’s effects on extinction, intracerebroventricular administration has been shown to facilitate extinction learning²⁰. The pharmacological actions of CBD are not fully understood, but proposed mechanisms include its interaction with multiple components of the endocannabinoid (eCB) system²¹. Together, these findings suggest that CBD may normalize dysfunctional fear circuits underlying PTSD-like behaviors, particularly through modulation of mPFC subregions.

Regarding the effect of CBD on SR, a recent study indicates that it can suppress SR, though the underlying molecular mechanisms are unknown²². Our previous work demonstrated that pharmacological modulation during extinction can bias mPFC circuits toward long-term fear suppression¹. The mPFC is a central regulator of fear expression, extinction, and SR, and its dysfunction is consistently implicated in PTSD^1,23. Within the mPFC, the PL and IL subregions exert opposing influences on fear behavior: the PL promotes the expression of conditioned fear, with heightened PL activity associated with elevated freezing and impaired extinction^24,25, whereas the IL supports extinction learning and consolidation, and its activation is required for long-term suppression of fear, while inhibition of IL disrupts extinction recall^25,26. This PL–IL functional dichotomy provides a circuit framework for understanding how fear responses either persist or are inhibited following extinction, highlighting the importance of subregion-specific mechanisms in regulating SR. Whether CBD engages similar or distinct molecular programs within the PL and IL to regulate SR remains unknown. Given the opposing roles of PL and IL in fear expression and suppression, and the distinct effects of CBD on each subregion’s fear response, simultaneous profiling of subregion-specific transcriptional programs during SR is essential for defining how CBD differentially stabilizes extinction memory versus suppresses fear expression and, ultimately, prevents SR.

In the present study, we investigated the effects of CBD on SR in both male and female mice to assess potential sex-dependent differences in treatment efficacy. Under our experimental conditions, CBD more effectively suppressed SR in male mice, and therefore subsequent molecular analyses focused on males. To define the underlying mechanisms, we performed bulk transcriptomic profiling of the PL and IL subregions of the mPFC following SR. This approach allowed us to identify subregion-specific and shared gene networks associated with SR and their modulation by CBD. To enhance translational relevance, we further compared these molecular signatures with transcriptomic data obtained from postmortem dorsolateral prefrontal cortex samples of individuals diagnosed with PTSD. Through this cross-species strategy, we aimed to identify conserved molecular programs underlying fear relapse and its suppression by CBD. Together, this study provides mechanistic insight into how CBD regulates SR at the level of prefrontal subregion-specific transcriptional programs and identifies molecular targets that may be leveraged to improve long-term PTSD treatment outcomes.

Results

Cannabidiol reduces spontaneous recovery after extinction in male mice

To investigate the effects of CBD on SR in a mouse model of PTSD, we employed a modified version of a well-established auditory fear conditioning protocol¹ (Fig. 1a). Male and female mice were tested in separate experiments. During the conditioning phase, no significant differences in freezing behavior were observed between vehicle-treated (SR_Veh) and CBD-treated (SR_CBD) groups in either sex (Fig. 1b, males, p = 0.9232; Fig. 1c, females, p = 0.9520). As described in a previous study²², either vehicle or CBD (5 mg/kg or 10 mg/kg) was administered 30 min before the first extinction session. Mice then underwent extinction training over four consecutive days, during which freezing behavior did not differ significantly between SR_Veh and SR_CBD groups in either males (Fig. 1b) or females (Fig. 1c). Seven days after the final extinction session, mice were re-exposed to Context B and presented with the tone to assess SR. CBD-treated male mice showed a significant reduction in freezing behavior during SR testing only at 5 mg/kg (Fig. 1b), not 10 mg/kg (Supplementary Fig. 1a). whereas no such effect was observed in females (Fig. 1c). These findings indicate that a 5 mg/kg dose of CBD, in conjunction with serial extinction sessions, more effectively attenuates SR in male mice. A control group (Con_Veh), which received context exposure and vehicle injection without shocks, was included to account for gene expression changes arising solely from exposure to Contexts A and B or from injection procedures (Fig. 1b, c).

Fig. 1 — a Illustration of the experimental design for behavioral testing. b Male behavior results: Fear conditioning, Extinction (Day 1-4), and SR. Two-way ANOVA (Time × Group): fear conditioning, F(5, 140) = 0.2804, p = 0.9232; extinction Day 1, F(5, 140) = 0.2556, p = 0.9364; Day 2, F(5, 140) = 1.493, p = 0.1960; Day 3, F(5, 140) = 0.7516, p = 0.5862; Day 4, F(5, 140) = 0.7421, p = 0.5932. SR (unpaired Student’s t-test), t(28) = 2.908, p = 0.0071. Red circles indicate mouse samples used for RNA-seq analysis. **p < 0.01 for indicated comparisons. Error bars represent S.E.M. (Con_Veh: n = 10, SR_Veh: n = 14, SR_CBD: n = 16). c Female behavior results: Fear conditioning, Extinction (Day 1-4), and SR. Two-way ANOVA (Time × Group): fear conditioning, F(5, 135) = 0.2233, p = 0.9520; extinction Day 1, F(5, 135) = 0.7964, p = 0.5541; Day 2, F(5, 135) = 1.063, p = 0.3989; Day 3, F(5, 135) = 0.3565, p = 0.8773; Day 4, F(5, 135) = 1.393, p = 0.2306. SR (unpaired Student’s t-test), t(27) = 0.05783, p = 0.9543. Error bars represent S.E.M. (Con_Veh: n = 10, SR_Veh: n = 13, SR_CBD: n = 16).

The spontaneous recovery drives distinct gene expression changes in PL mPFC of male mice

To investigate transcriptomic changes associated with SR, we isolated the PL subregion of the mPFC 24 hours after the SR test from three groups: Con_Veh, SR_Veh, and SR_CBD. Differentially expressed genes (DEGs) were identified using DESeq2, and expression patterns were visualized by heatmaps. Notably, gene expression profiles in the SR_Veh group clustered distinctly from both the Con_Veh and SR_CBD groups (Fig. 2a). In contrast, SR_CBD samples clustered closely with Con_Veh samples, despite undergoing fear conditioning and SR, suggesting that CBD may normalize fear-induced transcriptional alterations. Together, these findings indicate that CBD not only reduces SR at the behavioral level but also modulates gene expression in the PL at the molecular level.

Fig. 2 — a Heatmap of the normalized counts from each group (Con_Veh, SR_Veh, and SR_CBD). Three mice from each group were used for RNA-seq. b KEGG pathway analysis of DEGs from the SR_Veh vs. Con_Veh comparison. c Gene Ontology (GO) enrichment analysis of DEGs from the Veh vs. Con comparison. d KEGG pathway analysis of DEGs from the SR_CBD vs. SR_Veh comparison. e Gene Ontology (GO) enrichment analysis of DEGs from the SR_CBD vs. SR_Veh comparison.

To identify gene sets modulated by the fear response in the PL mPFC, we first compared DEGs between the SR_Veh and Con_Veh groups (PL: SR_Veh vs. Con_Veh) using a significance threshold of p < 0.05. This analysis revealed 924 upregulated genes and 1,090 downregulated genes (Supplementary Fig. S2a; Supplementary Data 1a). Among the top downregulated genes was Npas4²⁷, which has been implicated in stress- or anxiety-related behaviors, supporting the validity of our RNA-seq results. Conversely, notable upregulated genes included Sh2d1a²⁸, Isl1²⁹, Cplx3³⁰, and Nxph4³¹, previously linked to anxiety- or PTSD-like phenotypes, further reinforcing the reliability and biological relevance of our findings (Supplementary Fig. S2e).

KEGG pathway analysis of DEGs revealed axon guidance and ribosome as the most significantly modulated pathways in the PL. Several synapse-related pathways were also prominent among the top fifteen including glutamatergic, cholinergic, and oxytocin signaling (Fig. 2b, Supplementary Data 1b). Notably, SR also affected the retrograde endocannabinoid signaling pathway, suggesting that SR may influence endocannabinoid-mediated regulation in the PL (Fig. 2b, Supplementary Fig. 3a). Gene Ontology (GO) enrichment analysis further highlighted significant alterations in biological processes related to cellular processes, with molecular function categories enriched for protein and ion binding (Fig. 2c, Supplementary Data 1c). For cellular components, DEGs were enriched in terms associated with cellular anatomical entity, organelle, cell junction, and synapse (Fig. 2c, Supplementary Data 1c). Taken together, these results indicate that in the PL mPFC, SR strongly modulates gene sets associated with synaptic function and ribosomal pathways.

CBD induces distinct gene expression changes in the PL mPFC of male mice following spontaneous recovery

To determine whether CBD modulates gene expression associated with reduced SR in the PL region of the mPFC, we compared CBD-treated and vehicle-treated groups (PL: SR_CBD vs. SR_Veh). DESeq2 analysis identified 562 downregulated and 815 upregulated genes using a significance threshold of p < 0.05 (Supplementary Fig. S2b, Supplementary Data 1d). Among the top downregulated genes, Sh2d1a²⁸ and Isl1²⁹, both previously implicated in anxiety- or stress-related behaviors, were notable (Supplementary Fig. S2f). These genes were significantly upregulated in the PL: SR_Veh vs. Con_Veh comparison, indicating that CBD may counteract SR-induced transcriptional alterations (Supplementary Fig. S2e).

KEGG pathway analysis for PL: SR_CBD vs. SR_Veh highlighted ribosome, glutamatergic synapse, and axon guidance pathways, patterns that were also prominent in the DEGs from PL: SR_Veh vs. Con_Veh, reinforcing the idea that CBD attenuates SR-induced transcriptional changes within these pathways (Fig. 2b, d, Supplementary Data 1b, e). Notably, CBD treatment also produced significant alterations in genes associated with retrograde endocannabinoid signaling (Fig. 2d), suggesting that CBD-induced modulation of endocannabinoid signaling in the PL mPFC may serve as a central mechanism driving downstream gene set changes. Collectively, these findings support a broad regulatory role for CBD in the PL mPFC, influencing pathways related to cannabinoid signaling, protein synthesis, and synaptic function. GO enrichment analysis of DEGs in the PL: SR_CBD vs. SR_Veh comparison showed similar biological processes, molecular functions, and cellular components to those altered in the PL: SR_Veh vs. Con_Veh (Fig. 2e, Supplementary Data 1f). These results suggest that in the PL mPFC, CBD modifies gene sets in a manner closely resembling those influenced by SR alone (PL: SR_Veh vs. Con_Veh), albeit likely with opposing expression profiles (Fig. 2a, Supplementary Data 1).

CBD suppresses the gene expression changes induced by the spontaneous recovery in PL mPFC in male mice

To further validate and identify CBD-responsive genes showing expression shifts opposite to stress-induced changes in a PTSD model, we focused on DEGs that displayed opposite expression patterns between the SR_Veh vs. Con_Veh and SR_CBD vs. SR_Veh comparisons in PL region of the mPFC. We organized these genes into two distinct categories (Fig. 3a, c, Supplementary Data 2a, c): the first included genes significantly upregulated in the SR_Veh compared to Con_Veh and subsequently downregulated in SR_CBD compared to SR_Veh, while the second contained genes significantly downregulated in SR_Veh vs. Con_Veh but upregulated in SR_CBD vs. SR_Veh (Fig. 3a, c, Supplementary Figs. S4a, c, Supplementary Data 2a, c).

a Venn diagram of distinct and shared genes from the increased DEGs in the SR_Veh vs. Con_Veh comparison and the decreased DEGs in the SR_CBD vs. SR_Veh comparison in the PL mPFC. b KEGG pathway analysis for shared DEGs from Fig. 3a. c Venn diagram of distinct and shared genes from the decreased DEGs in the SR_Veh vs. Con_Veh comparison and the increased DEGs in the SR_CBD vs. SR_Veh comparison in the PL mPFC. d KEGG pathway analysis for shared DEGs from Fig. 3c.

For the 315 shared DEGs upregulated in SR_Veh vs. Con_Veh and downregulated in SR_CBD vs. SR_Veh, KEGG pathway analysis indicated that the ribosome pathway was the most significantly enriched (Fig. 3b, Supplementary Data 2b), with retrograde endocannabinoid signaling also ranking among the top 15 pathways (Fig. 3b, Supplementary Data 2b). GO analysis of these genes underscored cytoplasmic translation as a major biological process, the structural constituent of the ribosome as a key molecular function, and cytosolic ribosome as a primary cellular component, suggesting that these DEGs may be closely related to protein translation (Supplementary Fig. S4b, Supplementary Data 6a).

In contrast, for the 494 DEGs downregulated in SR_Veh vs. Con_Veh and upregulated in SR_CBD vs. SR_Veh, KEGG pathway analysis revealed enrichment in pathways related to the circadian entrainment and glutamatergic synapse (Fig. 3d, Supplementary Data 2d). GO analysis highlighted roles in nervous system development (biological processes), protein binding (molecular functions), and synapse (cellular components), suggesting these genes are potentially involved in nervous system development, synapse formation, and signaling transmission (Supplementary Fig. S4d, Supplementary Data 6b). For validation, Sh2d1a, one of the genes showing the most pronounced change in the opposite direction of stress-induced alterations following CBD treatment, was examined by RT-qPCR. Consistent with the RNA-seq findings, Sh2d1a expression was significantly increased in the PL mPFC following SR (SR_Veh) and was markedly reduced by CBD treatment, nearly returning to control levels (SR_CBD) (Supplementary Fig. S4e). We further assessed the relationship between Sh2d1a expression and freezing level during SR and found a significant positive correlation (Supplementary Fig. S4f), indicating that higher Sh2d1a expression is strongly associated with greater SR.

In summary, in PL region of the mPFC, the shared DEGs upregulated by SR (and downregulated by CBD) differ distinctly from those downregulated by SR (and upregulated by CBD) at the level of gene lists (Supplementary Figs. S4a, c), KEGG pathways (Fig. 3b, d, Supplementary Data 2b, d), and GO enrichment analysis (Supplementary Figs. S4b, d, Supplementary Data 6a, b).

The spontaneous recovery drives distinct gene expression changes in IL mPFC of male mice

To examine transcriptomic changes linked to SR in IL region of the mPFC, we analyzed IL mPFC tissue 24 hours post-recovery across three experimental groups: Con_Veh, SR_Veh, and SR_CBD. Using DESeq2 analysis, we identified DEGs and created heatmaps to visualize gene expression profiles in the IL mPFC (Fig. 4a). Similar to the PL mPFC, gene expression in the vehicle group clustered separately from both the Con_Veh and SR_CBD (Fig. 4a), suggesting that CBD may also normalize fear-induced transcriptional alterations in the IL mPFC.

Fig. 4 — a Heatmap of the normalized counts from each group (Con_Veh, SR_Veh, and SR_CBD). Three mice from each group were used for RNA-seq. b KEGG pathway analysis of DEGs from the SR_Veh vs. Con_Veh comparison. c Gene Ontology (GO) enrichment analysis of DEGs from the SR_Veh vs. Con_Veh comparison. d KEGG pathway analysis of DEGs from the SR_CBD vs. SR_Veh comparison. e Gene Ontology (GO) enrichment analysis of DEGs from the SR_CBD vs. SR_Veh comparison.

To identify the gene sets modulated by the fear response in the IL mPFC, we first examined the DEGs between the SR_Veh and Con_Veh (IL: SR_Veh vs. Con_Veh) using a significance threshold of p < 0.05. This analysis revealed 588 downregulated and 460 upregulated genes (Supplementary Fig. S2c, Supplementary Data 3a). Among the top ten downregulated genes were Aox2 which has been suggested to play a role in stress- or anxiety-like behaviors (Supplementary Fig. S2g). Top upregulated genes included Cxcl10³², Iigp1³³, Ifit3³⁴, Pagr1a³⁵, Sh2d1a²⁸, and Slc6a5³⁶, previously associated with stress or anxiety-related behaviors, further confirming the robustness of our findings (Supplementary Fig. S2g).

KEGG pathway analysis revealed significant enrichment in pathways related to ribosome and neurodegenerative diseases, including amyotrophic lateral sclerosis and Huntington disease, suggesting that major gene changes in the IL mPFC may involve protein translation processes and neurodegenerative mechanisms (Fig. 4b, Supplementary Data 3b). Notably, retrograde endocannabinoid signaling also ranked within the top 15 pathways, indicating that endocannabinoid pathways are similarly affected by SR in the IL mPFC (Fig. 4b, Supplementary Fig. S3b), consistent with the patterns observed in the PL mPFC (Supplementary Fig. S3a). GO enrichment analysis of these DEGs revealed notable alterations in cellular process and cellular component organization or biogenesis (Fig. 4c, Supplementary Data 3c). In terms of molecular function, DEGs were predominantly associated with protein and ion binding, as well as structural constituent of ribosome (Fig. 4c, Supplementary Data 3c). For cellular component, enriched categories included organelle, cellular anatomical entity, and synapse (Fig. 4c, Supplementary Data 3c). Taken together, the DEGs identified in the IL: SR_Veh vs. Con_Veh comparison are strongly associated with protein translation, particularly ribosomal function and ribosome structural components, as well as protein binding, intracellular organelles, and synaptic processes. Unexpectedly, although the top upregulated and downregulated genes in the IL following SR (IL: SR_Veh vs. Con_Veh) (Supplementary Fig. S2g) differed from those identified in the PL (PL: SR_Veh vs. Con_Veh) (Supplementary Fig. S2e), the shared GO terms across these regions suggest common underlying mechanisms associated with SR in both the PL and IL mPFC.

CBD drives distinct gene expression changes in IL mPFC of male mice following the spontaneous recovery

To examine whether CBD influences gene expression changes related to attenuating SR in IL region of mPFC, we compared the SR_CBD and SR_Veh (IL: SR_CBD vs. SR_Veh). DESeq2 analysis revealed 449 downregulated and 510 upregulated genes, using a p-value threshold of <0.05 (Supplementary Fig. S2d, Supplementary Data 3d). Among the top ten downregulated genes, Cxcl10³², Sertad1³⁷, and Sh2d1a²⁸, Slc6a5³⁶ have known associations with stress or anxiety-like behaviors (Supplementary Fig. S2h). In the top ten upregulated genes, we found genes, such as Nox1³⁸, Ulk4^39,40 and Oprm1^41–43, Tlr7^44,45, which are similarly linked to anxiety or stress responses, providing additional validation for our model (Supplementary Fig. S2h).

KEGG pathway analysis for IL: SR_CBD vs. SR_Veh also highlighted pathways, such as ribosome, amyotrophic lateral sclerosis, and Huntington disease, indicating that both SR and CBD treatment influence similar KEGG pathways (Fig. 4b, d, Supplementary Data 3b, e). CBD treatment also significantly affected the retrograde endocannabinoid signaling pathway in the IL mPFC, and although it was not listed within the top fifteen pathways, it was ranked twenty sixth (Fig. 4d, Supplementary Data 3e). GO enrichment analysis of DEGs in the IL: SR_CBD vs. SR_Veh comparison revealed similar biological processes, molecular functions, and cellular components as those seen in the IL: SR_Veh vs. Con_Veh comparison (Fig. 4c, e, Supplementary Data 3c, f). These data suggest that the modulation of gene expression by CBD in the IL (IL: SR_CBD vs. SR_Veh) aligns with the gene expression changes observed in SR alone (IL: SR_Veh vs. Con_Veh), potentially in opposite directions.

CBD suppresses the gene expression changes induced by the spontaneous recovery in IL mPFC in male mice

To further confirm and identify CBD-responsive genes in the PTSD-like animal model in the IL region of the mPFC, we focused on DEGs that exhibited expression changes in the opposite direction between the SR_Veh vs. Con_Veh and SR_CBD vs. SR_Veh conditions. These genes were categorized into two groups (Fig. 5a, c, Supplementary Data 4a, c): one group included genes significantly upregulated in SR_Veh vs. Con_Veh and subsequently downregulated with SR_CBD vs. SR_Veh (Fig. 5a, Supplementary Fig. S5a, Supplementary Data 4a), while the other comprised genes significantly downregulated in SR_Veh vs. Con_Veh and upregulated with SR_CBD vs. SR_Veh (Fig. 5c, Supplementary Fig. S5c, Supplementary Data 4c).

a Venn diagram of distinct and shared genes from the increased DEGs in the SR_Veh vs. Con_Veh comparison and the decreased DEGs in the SR_CBD vs. SR_Veh comparison in the IL mPFC. b KEGG pathway analysis for shared DEGs from Fig. 5a. c Venn diagram of distinct and shared genes from the decreased DEGs in the SR_Veh vs. Con_Veh comparison and the increased DEGs in the SR_CBD vs. SR_Veh comparison in the IL mPFC. d KEGG pathway analysis for shared DEGs from Fig. 5c.

KEGG pathway analysis of the 158 shared differentially expressed genes that were upregulated in the SR_Veh and downregulated in the SR_CBD identified the ribosome pathway and neurodegeneration related diseases, including Parkinson disease and Huntington disease, as the most significantly enriched (Fig. 5b, Supplementary Data 4b). Interestingly, the retrograde endocannabinoid signaling pathway was also significantly enriched in this gene group (Fig. 5b, Supplementary Data 4b). GO analysis highlighted oxidative phosphorylation and cytoplasmic translation as key biological processes, structural constituent of ribosome as the major molecular function, and the ribosomal subunit as the primary cellular component (Supplementary Fig. S5b, Supplementary Data 6c). Together, these findings suggest that genes upregulated in SR_Veh vs. Con_Veh and downregulated in SR_CBD vs. SR_Veh in the IL mPFC are closely linked to protein translation and neurodegeneration, consistent with the patterns observed in the PL mPFC.

For genes that were downregulated in SR_Veh vs. Con_Veh and upregulated in SR_CBD vs. SR_Veh in the IL mPFC, we identified 200 shared DEGs with opposing expression patterns (Fig. 5c, Supplementary Fig. S5c, Supplementary Data 4c). KEGG pathway analysis of this group revealed enrichment in axon guidance, glutamatergic synapse, and nicotine addiction pathways (Fig. 5d, Supplementary Data 4d). GO analysis indicated major terms related to regulation of biological process and nervous system development for the biological process, ion binding and protein binding for the molecular function, and neuron projection and synapse, particularly glutamatergic synapse, for the cellular component. These findings suggest involvement in developmental processes, synapse formation, and synaptic transmission, which is consistent with the patterns observed in the PL mPFC (Supplementary Fig. S5d, Supplementary Data 6d).

For validation, Cxcl10, one of the genes showing the most prominent expression change in the opposite direction following CBD treatment in the IL mPFC, was examined by RT–qPCR. Consistent with the RNA-seq findings, Cxcl10 expression tended to be increased in the IL mPFC following SR (SR_Veh) and was markedly reduced by CBD treatment, nearly returning to control levels (SR_CBD) (Supplementary Fig. S5e). We further assessed the relationship between Cxcl10 expression and freezing level during SR and found a significant positive correlation (Supplementary Fig. S5f), indicating that higher Cxcl10 expression is strongly associated with greater SR.

Distinct PL and IL mPFC gene expression signatures emerge following spontaneous recovery

When examining KEGG pathways and GO enrichment within the DEGs of each mPFC subregion, we unexpectedly found that the PL and IL regions shared many similar pathways and GO terms. Although these analyses revealed considerable overlap, given the distinct roles of each subregion in fear modulation, we reasoned that each area might also possess unique gene sets that respond specifically to SR and CBD treatment. Thus, we analyzed KEGG pathways generated exclusively by DEGs altered in PL: SR_Veh vs. Con_Veh but not in IL: SR_Veh vs. Con_Veh, and vice versa (Supplementary Figs. S6a, b, Supplementary Data 7a, 7b). In the PL mPFC, SR uniquely affected pathways, such as axon guidance, cholinergic synapse, regulation of actin cytoskeleton and circadian entrainment (Supplementary Fig. S6a, Supplementary Data 7a). In contrast, pathways uniquely altered in the IL mPFC included Parkinson disease, non-alcoholic fatty liver disease, chemical carcinogenesis-reactive oxygen species, and oxidative phosphorylation (Supplementary Fig. S6b, Supplementary Data 7b). These findings indicate that SR recruits subregion-specific molecular programs within the mPFC (Supplementary Fig. S6c).

Additionally, we identified several pathways uniquely altered in the PL: SR_CBD vs. SR_Veh comparison, including diabetic cardiomyopathy, retrograde endocannabinoid signaling, and Huntington disease (Supplementary Fig. S6d, Supplementary Data 7d). In contrast, pathways uniquely altered in the IL mPFC included RNA polymerase and Choline metabolism in cancer (Supplementary Fig. S6e, Supplementary Data 7e). Taken together, these differences suggest that CBD also engages distinct molecular programs in each subregion, even though certain pathways are shared between both areas (Supplementary Fig. S6f).

Cross-species transcriptomic analysis reveals shared DEGs between functionally corresponding human DLPFC and mouse mPFC

To identify genes with conserved expression changes between our mouse model and human PTSD subjects, we analyzed DEGs from postmortem brains of PTSD patients reported by Girgenti et al. (2021)⁴⁶. We focused specifically on DEGs from the human dorsolateral prefrontal cortex (DLPFC), a region functionally implicated in PTSD-related symptoms and considered to correspond to the PL and IL subdivisions of the rodent mPFC^46,47. Although the human DLPFC is not an anatomical homolog of the rodent PL or IL, its roles in cognitive control, executive regulation, and fear responsiveness align closely with key functional domains of the rodent mPFC. For this reason, several previous studies from other groups have compared the human DLPFC with rodent mPFC subregions using transcriptomic or proteomic approaches, and in the present study we extend this line of work by conducting similar cross-species comparisons^48,49. We identified 22 DEGs common to the human DLPFC (PTSD vs. Control) and mouse PL (SR_Veh vs. Con_Veh) and 17 DEGs common to the human DLPFC and mouse IL (SR_Veh vs. Con_Veh). Nine DEGs were shared across all three groups (Fig. 6a).

a Venn diagram of distinct and shared DEGs from human DLPFC (PTSD vs. Control), mouse PL (SR_Veh vs. Con_Veh) and mouse IL (SR_Veh vs. Con_Veh). b GO enrichment analysis for the 22 shared DEGs between human DLPFC and mouse PL from Fig. 6a. c GO enrichment analysis for the 17 shared DEGs between human DLPFC and mouse IL from Fig. 6a. d Potential CBD targets in the PL region. Log₂ fold change for five shared DEGs between human PTSD and mouse PL, showing expression shifts in the opposite direction following CBD treatment. e Potential CBD targets in the IL region. Log₂ fold change for 2 shared DEGs between human PTSD and mouse IL, showing expression shifts in the opposite direction following CBD treatment. f Nine shared DEGs among all regions. Log₂ fold change for nine shared DEGs across all regions (human DLPFC, mouse PL, mouse IL). (Lower right) Log₂ fold change for KLF9.

Although the number of overlapping genes was limited, we performed GO enrichment analysis on the 22 genes shared between the human DLPFC and mouse PL datasets (Fig. 6a, b, Supplementary Data 5a). This analysis revealed enrichment for molecular function regulator activity and hydrolase activity under molecular function, and for neuron projection and cell junction under cellular components (Fig. 6b; Supplementary Data 5a). For the 17 DEGs shared between the human DLPFC and mouse IL (Fig. 6a; Supplementary Data 5b, 8b), GO analysis identified cellular response to growth factor stimulus, cellular response to endogenous stimulus, and negative regulation of signaling as the most significantly enriched biological processes (Fig. 6c, Supplementary Data 5b). Among the 22 overlapping genes, ANP32B, HMGB2, MCM6, and RELL1 were upregulated, and ELFN1 was downregulated in both human DLPFC and mouse PL, with CBD shifting their expression in the opposite direction of stress-associated changes in the mouse PL, suggesting these genes as potential CBD-responsive targets in the PL mPFC (Fig. 6d, Supplementary Data 8a).

Among the 17 shared IL genes, NOS1 and DNMT1 were downregulated in both human DLPFC and mouse IL, and CBD similarly shifted their expression in the opposite direction of stress-associated changes in the IL, identifying them as potential CBD-responsive targets within the IL mPFC (Fig. 6e, Supplementary Data 8b). Of the nine DEGs shared across all three datasets, KLF9 was the only gene exhibiting a fully conserved expression pattern, with increased expression in the human DLPFC and in both the mouse PL and IL. Notably, CBD treatment shifted Klf9 expression in the opposite direction of stress-associated changes in the PL mPFC (Fig. 6f, Supplementary Data 5c). Collectively, these results highlight the utility of cross-species analysis in identifying PL-specific, IL-specific, and shared DEGs. Because CBD modulates the expression of a subset of these genes in a subregion-dependent manner, such analyses may be essential for pinpointing the functional targets involved in CBD-mediated regulation of SR.

Discussion

Our study demonstrates that CBD significantly attenuates the SR of fear in male mice, supporting its potential therapeutic value for mitigating SR, a phenomenon in PTSD characterized by the re-emergence of fear after extinction. These behavioral observations align with prior reports showing the beneficial effects of CBD on PTSD-like symptoms in both human patients^8,14 and rodent models^22,50. Importantly, our work extends these findings by identifying subregion-specific molecular signatures within the mPFC and revealing potential CBD-responsive targets that may contribute to the suppression of PTSD-like behaviors, particularly those related to SR.

The mPFC, specifically its PL and IL regions, is essential for regulating fear responses, with each region showing distinct physiological and molecular profiles. As previous studies have established, the PL and IL subregions of the mPFC play dissociable roles in fear behavior, with the PL primarily supporting fear expression and the IL facilitating extinction learning and consolidation^51,52. In line with this functional distinction, our prior work demonstrated that a single administration of the NMDAR positive allosteric modulator NYX-783 enhances GluN2B-containing NMDAR activity and increases BDNF expression specifically in the IL mPFC, producing long-lasting suppression of SR¹. These findings highlight that molecular interventions delivered during extinction training can induce persistent neuroplastic changes in the IL that strengthen extinction memory leading to inhibit the SR. Consistent with this framework, Papagianni et al. (2022)²² and our study used a similar experimental design in which CBD was administered prior to the extinction session, and observed significant reductions in freezing during SR, which was conducted several days after treatment. These findings raise the possibility that CBD administered during extinction may induce enduring molecular and circuit-level adaptations that persist beyond the extinction session, potentially involving the IL mPFC. In addition, CBD was administered prior to the first extinction session rather than immediately before the SR test. Although no significant differences in freezing behavior were observed between groups during extinction training, CBD administration before extinction may influence processes occurring during extinction learning that subsequently affect behavior during the SR test. Therefore, the observed behavioral effects may reflect enhanced consolidation or long-term stabilization of extinction memory rather than a selective modulation of SR mechanisms. Future studies will be required to determine the extent to which CBD directly influences SR versus extinction memory stabilization.

However, several studies have also implicated the PL mPFC in fear extinction. Persistent PL activity has been associated with sustained freezing behavior and impaired extinction learning, which may influence later fear responses, including SR²⁴. Moreover, coordinated synchronization between the PL and IL subregions has been suggested to be required for proper regulation of fear extinction⁵³. Together, these findings indicate that both PL and IL subregions may contribute to the regulation of fear extinction and its long-term behavioral outcomes. Despite these advances, molecular alterations within the PL and IL subregions have rarely been compared simultaneously using high-throughput approaches.

Our transcriptomic analyses revealed both shared and distinct CBD-induced molecular effects across the PL and IL regions, providing insight into how CBD modulates prefrontal circuitry to suppress SR. SR itself induced substantial transcriptional remodeling in both regions, with a greater number of unique DEGs in the PL than in the IL, suggesting that SR engages broader PL-specific molecular changes than previously appreciated. This finding highlights that, beyond its canonical role in fear expression, the PL undergoes significant molecular adaptation during SR, contributing to relapse-like processes. KEGG pathway analyses further identified retrograde endocannabinoid signaling as a prominently altered pathway in the PL, where CBD reduced the expression of key genes, such as Cnr1. Although the magnitude of downregulation was modest, this molecular shift is consistent with the PL’s established role in driving fear expression and SR when extinction memory weakens over time^1,54. Conversely, the IL region is well established as a critical locus for extinction memory formation and retrieval, functioning to inhibit conditioned fear responses and prevent the return of fear after extinction²⁶. In the IL mPFC, we identified CBD-induced transcriptional changes in genes associated with neurodegeneration-related pathways, including Huntington disease, as well as pathways related to RNA polymerase and ribosome. Because Huntington’s disease is characterized by deficits in fear extinction⁵⁵, and de novo protein synthesis is essential for the consolidation of extinction memory⁵⁶, these CBD-induced molecular changes may enhance the stability of extinction-related circuits, contributing to the sustained suppression of fear responses observed during SR.

Despite these subregion-specific DEGs, we also observed substantial overlap in the pathways altered by SR and modulated by CBD across both regions (Figs. 2 and 4; Supplementary Fig. 6). Importantly, the KEGG pathways and GO terms affected by CBD closely mirrored those dysregulated during SR within each subregion, suggesting that CBD may counteract key components of the SR-associated molecular program in both the PL and IL mPFC. Notably, the ribosome pathway emerged as one of the top-ranked pathways in both regions following SR and was shifted toward non-shocked control levels by CBD, highlighting a shared mechanistic route through which CBD may influence SR-related molecular processes. This convergence is consistent with emerging evidence that PL–IL functional synchronization is essential for extinction learning⁵³, and our transcriptomic findings further raise the possibility that coordinated molecular regulation across these subregions contributes to the regulation of SR-related behavioral outcomes. Nonetheless, interpretation of these patterns will require further mechanistic investigation.

Building on this, the results of this study underscore the value of cross-species transcriptomic analyses for identifying conserved gene expression patterns relevant to PTSD. By examining DEGs from the DLPFC of human PTSD patients and comparing them with DEGs from the PL and IL subregions of the mouse mPFC, we identified both shared and region-specific molecular signatures. Although the number of overlapping genes was relatively small, reflecting inherent species differences and anatomical distinctions, the identification of 22 and 17 common DEGs between the human DLPFC and mouse PL and IL, respectively, highlights conserved transcriptional responses to fear across species. These shared DEGs are particularly informative, as they may represent core molecular contributors to stress-related pathology (Fig. 6, Supplementary Fig. S7, Supplementary Data 5 and 8). Several genes shared between the human DLPFC and mouse mPFC subregions exhibited stress-associated expression patterns that were modulated by CBD. In the PL, ANP32B, HMGB2, MCM6, RELL1, and ELFN1 showed SR–related alterations mitigated by CBD. In the IL, NOS1 and DNMT1 demonstrated similar CBD-responsive patterns. Notably, many of these genes, including HMGB2⁵⁷, MCM6⁵⁸, ANP32B^59,60, and NOS1⁶¹, have previously been implicated in anxiety- or PTSD-like phenotypes. Additionally, KLF9 emerged as the only gene consistently elevated across the PL, IL, and human DLPFC datasets, with CBD modulating its expression specifically in the PL. Given its established roles in stress regulation and neuroplasticity^62,63, KLF9 represents a strong candidate conserved molecular regulator that may contribute to synchronization between the PL and IL subregions underlying fear modulation. Nonetheless, interpretation of these patterns will require further mechanistic investigation. Notably, while common genes identified via cross-species analysis, including Klf9, did not rank among the primary DEGs within the rodent subregions alone, they emerged as translationally significant candidates upon integration with human datasets. This finding underscores the necessity of cross-species analysis as a strategy for uncovering high-priority targets that might otherwise remain obscured within single-species datasets, although the latter remain valuable for elucidating the mechanisms underlying SR and CBD.

Our study identified gene sets that are uniquely or commonly associated with each mPFC subregion in relation to SR and CBD treatment; however, several limitations should be acknowledged. First, DEGs were defined using a nominal p < 0.05 threshold as an exploratory approach, which increases the risk of false positives at the individual gene level. Accordingly, individual gene-level findings should be interpreted with caution, whereas pathway-level enrichment and cross-species convergence may provide more robust biological insight. Second, although CBD treatment significantly reduced SR, residual freezing levels remained relatively high. Consistent with this behavioral observation, a substantial proportion of SR-associated DEGs were not fully shifted toward the levels observed in non-shocked controls following CBD treatment. This incomplete molecular normalization may contribute to the persistence of freezing behavior. In addition, extinction training did not completely suppress freezing behavior, with freezing remaining at approximately 40% during the final extinction session. This pattern has been reported in previous studies and represents a recognized limitation in modeling fear extinction in mice⁶⁴. Because extinction was incomplete, freezing observed during the SR test may partly reflect residual fear memory rather than the re-emergence of a fully extinguished memory. Third, the original RNA-seq design did not include a Con_CBD group, which limits the ability to fully dissociate baseline transcriptional effects of CBD from its SR-specific actions. To partially address this limitation, we included a Con_CBD group during validation of the most significant DEGs in the PL and IL regions and observed minimal CBD-induced changes in these selected genes under control conditions. Fourth, tissue samples were collected 24 h after the SR test following a behavioral protocol that included conditioning and extinction training. Therefore, the observed transcriptional signatures likely reflect a composite molecular state capturing not only processes related to SR but also residual traces of extinction learning, retrieval-induced plasticity during the SR test, and stress- or arousal-related responses associated with behavioral testing. Accordingly, the gene sets identified here should be interpreted as SR-associated transcriptional states within the mPFC rather than mechanisms exclusively specific to SR.

In addition, increased variability was observed within the CBD-treated group during the SR test. This variability may reflect individual differences in drug distribution or sensitivity to CBD. Importantly, the overall effect remained statistically significant, supporting the robustness of the finding. The present study also focused exclusively on freezing behavior as the primary index of conditioned fear and SR and did not quantify darting behavior, which has been reported as an additional fear-related response, particularly in female animals. This limitation should be considered when interpreting potential sex-dependent effects of CBD. The transcriptomic analyses were conducted exclusively in male mice. Therefore, the molecular findings described in this study apply specifically to males under the present experimental conditions and should not be generalized to females. Finally, we tested only single injections of 5 mg/kg and 10 mg/kg CBD, and a significant effect was observed only in male mice at 5 mg/kg. These findings highlight the need for a more comprehensive dose–response assessment. Future studies employing a broader dose range, larger sample sizes, or refined delivery strategies will be necessary to fully evaluate CBD efficacy and to clarify the sources of variability observed in the present study.

In summary, our study provides converging behavioral, molecular, and cross-species evidence that CBD acts as a potent modulator of SR, a core feature of PTSD. We show that CBD robustly attenuates SR in male mice and shifts SR-induced transcriptional alterations in both the PL and IL subregions of the mPFC toward non-shocked control levels. Despite region-specific DEGs, the underlying pathways and CBD-responsive gene networks are highly overlapping and functionally synchronized. Notably, retrograde endocannabinoid signaling emerged as a shared pathway modulated by CBD, suggesting a potential mechanistic route through which CBD may influence the SR after extinction. Cross-species analysis further identified conserved stress-responsive genes whose expression was shifted by CBD in the opposite direction toward non-stressed control levels, underscoring their potential roles as key molecular mediators of CBD’s therapeutic effects. Together, these findings position CBD as a promising upstream regulator capable of restoring stress-disrupted molecular states within the mPFC and highlight conserved gene networks that may serve as targets for preventing the re-emergence of traumatic fear memories in PTSD.

Materials and Methods

Animals

Experiments were conducted on 11- to 12-week-old male and female C57BL/6NTac mice (DBL, South Korea). Upon arrival, mice were housed in groups of five per cage and acclimated to their environment for two weeks prior to behavioral testing. They were kept in individually ventilated cages under a 12-hour light/dark cycle (lights on at 08:00) with unrestricted access to food and water. Humane endpoints, such as severe body weight loss (>20%) and changes in physical appearance were monitored regularly; however, no animals met these criteria and no samples were excluded. All procedures took place during the light cycle and were conducted in accordance with the Institute for Basic Science (IBS) Institutional Animal Care and Use Committee (IACUC). Approval for this study was obtained under protocol number (IBS-2024-002), and all protocols were designed to minimize animal discomfort in line with the IACUC guidelines.

Pharmacological agents

CBD (Catalog No. THC-303, Lot No. 303.1B25.1; Lipomed AG, Arlesheim, Switzerland) was suspended in 2% Tween 80 in sterile saline and freshly prepared each day. A dose of 5 mg/kg or 10 mg/kg was administered intraperitoneally 30 min before extinction training on Day 1. Control and vehicle groups received an equivalent vehicle solution (2% Tween 80 in sterile saline). The dose and timing of CBD administration were selected based on a previously published protocol by Papagianni et al. (2022)²².

Fear conditioning

For behavioral experiments, sample size was determined using the resource equation method (E = N − B − T), where N represents the total number of animals, B the number of experimental groups, and T the number of treatments. The resulting E value was maintained within the recommended range of 10–20⁶⁵.

The fear conditioning protocol was conducted according to a previously established protocol with slight modifications¹. Two days before fear conditioning, mice were habituated to Context A for 10 min. On the following day, mice were additionally habituated to Context B for 10 min to reduce novelty-induced anxiety during subsequent behavioral procedures. Fear conditioning was then conducted on the next day. Mice were then placed in a training chamber (Context A) and given a 180-second acclimation period, followed by five pairings of a conditioned stimulus (CS) tone (84 dB, 2 kHz, 5 ms rise/fall time, 30 s duration) with an unconditioned stimulus (US) foot shock (1 second, 0.7 mA). After each tone, there was a 30-second interval. Chambers were cleaned with 70% ethanol between subjects. Freezing behavior during each CS tone presentation was recorded and analyzed to determine baseline freezing levels. Mice were then randomly assigned to vehicle or CBD groups, balanced based on freezing levels. Behavioral data were analyzed using Video Freeze software (Version 2.7.1.107, Med Associates Inc., Fairfax, VT, USA).

Extinction training

The extinction and SR protocols were conducted according to a previously established method with slight modifications¹. Briefly, thirty minutes post-treatment, mice in the CBD and vehicle groups were placed in a novel environment (Context B) for extinction training. Context B was distinct from Context A, differing in lighting, background, and floor texture. Following a 180-second acclimation period, mice were presented with 12 non-reinforced conditioned stimulus (CS) tones (30 s, 84 dB) with random inter-tone intervals ranging from 60 to 80 s. This extinction training was repeated daily for four consecutive days. Chambers were sanitized with 70% ethanol between animals to prevent olfactory cues. One week after the final extinction training session, SR was assessed by reintroducing the mice to Context B. Following a 180-second acclimation period, mice were presented with single non-reinforced conditioned stimulus (CS) tone (30 s, 84 dB). Behavioral data were analyzed using Video Freeze software (Version 2.7.1.107, Med Associates Inc., Fairfax, VT, USA).

mPFC dissection and RNA purification

Twenty-four hours after SR, mice were rapidly decapitated, and their brains were collected. In each group, three mice with freezing levels close to the group mean were randomly selected to represent typical responders for gene expression profiling. Postbehavioral tissue sampling was performed under blinded conditions. The whole brain was placed in a stainless-steel mouse brain matrix, and bregma was identified. The mPFC was dissected using a razor blade between bregma +1.0 mm and +2.0 mm. The slice was placed on a glass slide and snap-frozen on dry ice. A horizontal cut at the ventral edge of the forceps minor, corresponding to the tip of the corpus callosum, was used as a practical anatomical landmark to separate the prelimbic (PL, dorsal) and infralimbic (IL, ventral) subregions. Tissue from a single mouse was used for each sample, and no pooling was performed. RNA isolation and purification were performed using the Maxwell RSC 16 system with the RSC RNA FFPE Kit (Promega, USA). RNA quality was assessed using a Bioanalyzer 2100 (Agilent Technologies, Wilmington, DE, USA). RNA library preparation and sequencing were conducted by Inocras (San Diego, CA, USA) using the KAPA mRNA Hyper Prep Kit (Illumina, San Diego, CA, USA), followed by sequencing on an Illumina NovaSeq 6000 platform (Illumina, San Diego, CA, USA).

RNA-seq data analysis

For RNA-seq analysis, three animals per group were selected based on their behavioral performance during the SR test. In general, animals whose freezing levels were close to the group average were prioritized in order to represent typical behavioral responses within each group. However, the SR_CBD group exhibited greater inter-individual variability in freezing behavior compared with the other groups. Therefore, animals in this group were selected to reflect the observed behavioral distribution rather than strictly selecting those closest to the arithmetic mean. All selected animals fell within the behavioral range observed for their respective groups. This sampling strategy allowed representation of the overall behavioral patterns while maintaining comparability across groups. RNA-seq data analysis was conducted using a modified pipeline implemented in the Partek Genomics Suite (Flow version 10.0.21.0328; Partek, St. Louis, MO, USA), with methodological adjustments guided by prior work⁶⁶. Prior to library preparation, RNA integrity was confirmed for all samples, with RNA integrity numbers (RIN) exceeding 8.0 (Supplemental Data 9). Pre-alignment quality control verified high-quality sequencing reads, with average base quality scores above 39. Reads were aligned to the mouse whole genome (mm39) using STAR (version 2.7.8a). All samples had more than 50 million aligned reads, with comparable numbers of detected genes and uniform gene body coverage (Supplemental Data 9). Differential expression analysis was conducted in DESeq2 using the median ratio method for normalization. Genes with p < 0.05 were subjected to KEGG pathway and Gene Ontology (GO) enrichment analyses. Comparative analyses of differentially expressed genes (DEGs) between human and mouse datasets were further evaluated using KEGG and GO enrichment in gProfiler (https://biit.cs.ut.ee/gprofiler/gost). We used postmortem human brain RNA-seq data from individuals with PTSD published by Girgenti et al. (2021)⁴⁶. Statistical analyses were conducted using GraphPad Prism 10.3.1. (GraphPad Software, 225 Franklin Street, Fl. 26, Boston, MA 02110).

Statistics and reproducibility

Group differences in fear conditioning and extinction were evaluated through two-way ANOVA, followed by Sidak’s post hoc tests. SR was analyzed with a two-tailed unpaired Student’s t-test comparing the Veh and CBD groups. Statistical significance for behavioral data and RNA-seq results was set at p < 0.05, with specific details for each test outlined in the figure legends. All experiments were performed with at least three independent biological replicates unless otherwise stated. Sample sizes (n) for each experiment are indicated in the figure legends. Animals were randomly assigned to experimental groups.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Supplementary information

Supplementary Information^{(1.3MB, pdf)}

42003_2026_10111_MOESM2_ESM.docx^{(13.4KB, docx)}

Description of Additional Supplementary Files

Supplemental Data^{(6.9MB, xlsx)}

Reporting Summary^{(1.9MB, pdf)}

Transparent Peer Review file^{(581.3KB, pdf)}

Acknowledgements

This work was supported by the Korea Research Institute of Bioscience and Biotechnology Research Initiative Programs (grant number KGM5382622 to JCJ), the National Research Council of Science & Technology (NST) grant funded by the Korean government (MSIT) (grant number CAP23053-000 to JCJ), and the Institute for Basic Science (IBS), Center for Memory and Glioscience (IBS-R001-D2 to BL).

Author contributions

J.C.J. and B.L. conceived the original idea, with contributions to the study concept from M.J.G. and C.Y.K. M.S., J.C.J., and B.L. designed the experiments. M.S., I.S., P.P., and B.C. performed the experiments and analyzed the data. M.S., J.C.J., and B.L. wrote the manuscript. All authors contributed to the interpretation of the results and provided critical feedback on the manuscript.

Peer review

Peer review information

: Communications Biology thanks the anonymous reviewers for their contribution to the peer review of this work. Primary Handling Editors: Joao Valente. A peer review file is available.

Data availability

The DEGs, as well as the GO and KEGG enrichment results for the PL and IL regions of the mouse brain and the DLPFC of the human postmortem brain, are included in the Supplementary Data. Summary statistics are provided in Supplementary Data 1–10. RNA-seq data have been deposited in the NCBI Sequence Read Archive (SRA) under BioProject accession PRJNA1449752 (SRR37964749–SRR37964758). Numerical source data underlying graphs in the manuscript can be found in the Supplementary Data file.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Contributor Information

Jae Cheol Jeong, Email: jcjeong@kribb.re.kr.

Boyoung Lee, Email: blee@ibs.re.kr.

Supplementary information

The online version contains supplementary material available at https://doi.org/10.1038/s42003-026-10111-4.

References

1.Lee, B. et al. Positive modulation of N-methyl-D-aspartate receptors in the mPFC reduces the spontaneous recovery of fear. Mol. Psychiatry27, 2580–2589 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Pitman, R. K. et al. Biological studies of post-traumatic stress disorder. Nat. Rev. Neurosci.13, 769–787 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Maren, S. & Holmes, A. Stress and fear extinction. Neuropsychopharmacology41, 58–79 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Koenen, K. C. et al. Posttraumatic stress disorder in the World Mental Health Surveys. Psychol. Med47, 2260–2274 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Bouton, M. E. Context and behavioral processes in extinction. Learn Mem.11, 485–494 (2004). [DOI] [PubMed] [Google Scholar]
6.Myers, K. M. & Davis, M. Behavioral and neural analysis of extinction. Neuron36, 567–584 (2002). [DOI] [PubMed] [Google Scholar]
7.Stella, N. THC and CBD: Similarities and differences between siblings. Neuron111, 302–327 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Bitencourt, R. M. & Takahashi, R. N. Cannabidiol as a therapeutic alternative for post-traumatic stress disorder: from bench research to confirmation in human trials. Front Neurosci.12, 502 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Simei, J. L. Q. et al. Cannabidiol in anxiety disorders: current and future perspectives. Int Rev. Neurobiol.177, 205–234 (2024). [DOI] [PubMed] [Google Scholar]
10.Blessing, E. M. et al. Cannabidiol as a potential treatment for anxiety disorders. Neurotherapeutics12, 825–836 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Crippa, J. A. et al. Neural basis of anxiolytic effects of cannabidiol (CBD) in generalized social anxiety disorder: a preliminary report. J. Psychopharmacol.25, 121–130 (2011). [DOI] [PubMed] [Google Scholar]
12.Bergamaschi, M. M. et al. Cannabidiol reduces the anxiety induced by simulated public speaking in treatment-naïve social phobia patients. Neuropsychopharmacology36, 1219–1226 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
13.De Morais, E. R. et al. Open-label: the clinical effects of adding cannbidiol to usual care of patients with residual symptoms in the diagnosis of obsessive compulsive disorder. Eur. Psychiatry66, S225–S226 (2023). [Google Scholar]
14.Elms, L. et al. Cannabidiol in the treatment of post-traumatic stress disorder: a case series. J. Alter. Complement Med.25, 392–397 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Das, R. K. et al. Cannabidiol enhances consolidation of explicit fear extinction in humans. Psychopharmacology226, 781–792 (2013). [DOI] [PubMed] [Google Scholar]
16.Lemos, J. I., Resstel, L. B. & Guimarães, F. S. Involvement of the prelimbic prefrontal cortex on cannabidiol-induced attenuation of contextual conditioned fear in rats. Behav. Brain Res.207, 105–111 (2010). [DOI] [PubMed] [Google Scholar]
17.Fogaça, M. V. et al. Effects of intra-prelimbic prefrontal cortex injection of cannabidiol on anxiety-like behavior: involvement of 5HT1A receptors and previous stressful experience. Eur. Neuropsychopharmacol.24, 410–419 (2014). [DOI] [PubMed] [Google Scholar]
18.Marinho, A. L. et al. Effects of intra-infralimbic prefrontal cortex injections of cannabidiol in the modulation of emotional behaviors in rats: contribution of 5HT₁A receptors and stressful experiences. Behav. Brain Res286, 49–56 (2015). [DOI] [PubMed] [Google Scholar]
19.Jurkus, R. et al. Cannabidiol regulation of learned fear: implications for treating anxiety-related disorders. Front Pharm.7, 454 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Bitencourt, R. M., Pamplona, F. A. & Takahashi, R. N. Facilitation of contextual fear memory extinction and anti-anxiogenic effects of AM404 and cannabidiol in conditioned rats. Eur. Neuropsychopharmacol.18, 849–859 (2008). [DOI] [PubMed] [Google Scholar]
21.Bisogno, T. et al. Molecular targets for cannabidiol and its synthetic analogues: effect on vanilloid VR1 receptors and on the cellular uptake and enzymatic hydrolysis of anandamide. Br. J. Pharm.134, 845–852 (2001). [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Papagianni, E. P. et al. Cannabidiol prevents spontaneous fear recovery after extinction and ameliorates stress-induced extinction resistance. Int J. Mol. Sci.23, 9333 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Marek, R. et al. The amygdala and medial prefrontal cortex: partners in the fear circuit. J. Physiol.591, 2381–2391 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Burgos-Robles, A., Vidal-Gonzalez, I. & Quirk, G. J. Sustained conditioned responses in prelimbic prefrontal neurons are correlated with fear expression and extinction failure. J. Neurosci.29, 8474–8482 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Sierra-Mercado, D., Padilla-Coreano, N. & Quirk, G. J. Dissociable roles of prelimbic and infralimbic cortices, ventral hippocampus, and basolateral amygdala in the expression and extinction of conditioned fear. Neuropsychopharmacology36, 529–538 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Quirk, G. J. & Mueller, D. Neural mechanisms of extinction learning and retrieval. Neuropsychopharmacology33, 56–72 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Jaehne, E. J. et al. Effects of Npas4 deficiency on anxiety, depression-like, cognition and sociability behaviour. Behav. Brain Res.281, 276–282 (2015). [DOI] [PubMed] [Google Scholar]
28.Zhang, H. et al. Construction of diagnostic model and subtype analysis of major depressive disorder based on PANoptosis key genes. BMC Psychiatry25, 929 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Shen, E. Y. et al. Neuronal deletion of Kmt2a/Mll1 histone methyltransferase in ventral striatum is associated with defective spike-timing-dependent striatal synaptic plasticity, altered response to dopaminergic drugs, and increased anxiety. Neuropsychopharmacology41, 3103–3113 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Guo, Y. P. et al. Global gene knockout of Kcnip3 enhances pain sensitivity and exacerbates negative emotions in rats. Front. Mol. Neurosci.12, 5 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Meng, X. et al. Neurexophilin4 is a selectively expressed α-neurexin ligand that modulates specific cerebellar synapses and motor functions. Elife8, e46773 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Camacho-Arroyo, I. et al. Chemokine profile in women with moderate to severe anxiety and depression during pregnancy. BMC Pregnancy Childbirth21, 807 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Depke, M. et al. Altered hepatic mRNA expression of immune response and apoptosis-associated genes after acute and chronic psychological stress in mice. Mol. Immunol.46, 3018–3028 (2009). [DOI] [PubMed] [Google Scholar]
34.Rimmerman, N. et al. The hippocampal transcriptomic signature of stress resilience in mice with microglial fractalkine receptor (CX3CR1) deficiency. Brain Behav. Immun.61, 184–196 (2017). [DOI] [PubMed] [Google Scholar]
35.Miyako, K. et al. Identification of the antidepressant effect of electroconvulsive stimulation-related genes in hippocampal astrocyte. J. Psychiatr. Res.170, 318–327 (2024). [DOI] [PubMed] [Google Scholar]
36.Zoubovsky, S. P. et al. Neurobehavioral abnormalities following prenatal psychosocial stress are differentially modulated by maternal environment. Transl. Psychiatry12, 22 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Yang, C. et al. Co-Expression network modeling identifies specific inflammation and neurological disease-related genes mRNA modules in mood disorder. Front Genet13, 865015 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Ibi, M. et al. Depressive-like behaviors are regulated by NOX1/NADPH oxidase by redox modification of NMDA receptor 1. J. Neurosci.37, 4200–4212 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Lang, B. et al. Recurrent deletions of ULK4 in schizophrenia: a gene crucial for neuritogenesis and neuronal motility. J. Cell Sci.127, 630–640 (2014). [DOI] [PubMed] [Google Scholar]
40.Liu, M. et al. Ulk4 regulates GABAergic signaling and anxiety-related behavior. Transl. Psychiatry8, 43 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Lovallo, W. R. et al. Cortisol stress response in men and women modulated differentially by the mu-opioid receptor gene polymorphism OPRM1 A118G. Neuropsychopharmacology40, 2546–2554 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Collins, D. et al. Oprm1 A112G, a single nucleotide polymorphism, alters expression of stress-responsive genes in multiple brain regions in male and female mice. Psychopharmacology235, 2703–2711 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Swann, G. et al. Effect of OPRM1 and stressful life events on symptoms of major depression in African American adolescents. J. Affect Disord.162, 12–19 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Hung, Y. F. & Hsueh, Y. P. TLR7 and IL-6 differentially regulate the effects of rotarod exercise on the transcriptomic profile and neurogenesis to influence anxiety and memory. iScience24, 102384 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
45.Dunstan, I. K. et al. Unique pathways downstream of TLR-4 and TLR-7 activation: sex-dependent behavioural, cytokine, and metabolic consequences. Front. Cell Neurosci.18, 1345441 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
46.Girgenti, M. J. et al. Transcriptomic organization of the human brain in post-traumatic stress disorder. Nat. Neurosci.24, 24–33 (2021). [DOI] [PubMed] [Google Scholar]
47.Alexandra Kredlow, M. et al. Prefrontal cortex, amygdala, and threat processing: implications for PTSD. Neuropsychopharmacology47, 247–259 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
48.Phelps, E. A. et al. Extinction learning in humans: role of the amygdala and vmPFC. Neuron43, 897–905 (2004). [DOI] [PubMed] [Google Scholar]
49.Hartley, C. A. & Phelps, E. A. Changing fear: the neurocircuitry of emotion regulation. Neuropsychopharmacology35, 136–146 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
50.Xie, G. et al. Cannabidiol ameliorates PTSD-like symptoms by inhibiting neuroinflammation through its action on CB2 receptors in the brain of male mice. Brain Behav. Immun.119, 945–964 (2024). [DOI] [PubMed] [Google Scholar]
51.Sotres-Bayon, F. & Quirk, G. J. Prefrontal control of fear: more than just extinction. Curr. Opin. Neurobiol.20, 231–235 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
52.Senn, V. et al. Long-range connectivity defines behavioral specificity of amygdala neurons. Neuron81, 428–437 (2014). [DOI] [PubMed] [Google Scholar]
53.Giustino, T. F. & Maren, S. The role of the medial prefrontal cortex in the conditioning and extinction of fear. Front Behav. Neurosci.9, 298 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
54.Corcoran, K. A. & Quirk, G. J. Activity in prelimbic cortex is necessary for the expression of learned, but not innate, fears. J. Neurosci.27, 840–844 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]
55.Walker, A. G., Ummel, J. R. & Rebec, G. V. Reduced expression of conditioned fear in the R6/2 mouse model of Huntington’s disease is related to abnormal activity in prelimbic cortex. Neurobiol. Dis.43, 379–387 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
56.Santini, E. et al. Consolidation of fear extinction requires protein synthesis in the medial prefrontal cortex. J. Neurosci.24, 5704–5710 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]
57.Daws, S. E. et al. MicroRNA regulation of persistent stress-enhanced memory. Mol. Psychiatry25, 965–976 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
58.Sokoloff, G. et al. Anxiety and fear in a cross of C57BL/6J and DBA/2J mice: mapping overlapping and independent QTL for related traits. Genes Brain Behav.10, 604–614 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
59.Paban, V. et al. Molecular gene expression following blunt and rotational models of traumatic brain injury parallel injuries associated with stroke and depression. J. Transl. Sci.2, 330–339 (2016). [Google Scholar]
60.Zhou, J. & Zhu, L. Shared genetic links between hypothyroidism and psychiatric disorders: evidence from a comprehensive genetic analysis. Front. Endocrinol.15, 1370019 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
61.Bruenig, D. et al. Nitric oxide pathway genes (NOS1AP and NOS1) are involved in PTSD severity, depression, anxiety, stress and resilience. Gene625, 42–48 (2017). [DOI] [PubMed] [Google Scholar]
62.Besnard, A. et al. Targeting Kruppel-like Factor 9 in excitatory neurons protects against chronic stress-induced impairments in dendritic spines and fear responses. Cell Rep.23, 3183–3196 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
63.Gans, I. et al. Klf9 is a key feedforward regulator of the transcriptomic response to glucocorticoid receptor activity. Sci. Rep.10, 11415 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
64.Kamprath, K. et al. Short-term adaptation of conditioned fear responses through endocannabinoid signaling in the central amygdala. Neuropsychopharmacology36, 652–663 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
65.Arifin, W. N. & Zahiruddin, W. M. Sample size calculation in animal studies using resource equation approach. Malays. J. Med. Sci.24, 101–105 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
66.Ju, Y. H. et al. Astrocytic urea cycle detoxifies Aβ-derived ammonia while impairing memory in Alzheimer’s disease. Cell Metab.34, 1104–1120 (2022). [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Information^{(1.3MB, pdf)}

42003_2026_10111_MOESM2_ESM.docx^{(13.4KB, docx)}

Description of Additional Supplementary Files

Supplemental Data^{(6.9MB, xlsx)}

Reporting Summary^{(1.9MB, pdf)}

Transparent Peer Review file^{(581.3KB, pdf)}

Data Availability Statement

[CR1] 1.Lee, B. et al. Positive modulation of N-methyl-D-aspartate receptors in the mPFC reduces the spontaneous recovery of fear. Mol. Psychiatry27, 2580–2589 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR2] 2.Pitman, R. K. et al. Biological studies of post-traumatic stress disorder. Nat. Rev. Neurosci.13, 769–787 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR3] 3.Maren, S. & Holmes, A. Stress and fear extinction. Neuropsychopharmacology41, 58–79 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR4] 4.Koenen, K. C. et al. Posttraumatic stress disorder in the World Mental Health Surveys. Psychol. Med47, 2260–2274 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR5] 5.Bouton, M. E. Context and behavioral processes in extinction. Learn Mem.11, 485–494 (2004). [DOI] [PubMed] [Google Scholar]

[CR6] 6.Myers, K. M. & Davis, M. Behavioral and neural analysis of extinction. Neuron36, 567–584 (2002). [DOI] [PubMed] [Google Scholar]

[CR7] 7.Stella, N. THC and CBD: Similarities and differences between siblings. Neuron111, 302–327 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR8] 8.Bitencourt, R. M. & Takahashi, R. N. Cannabidiol as a therapeutic alternative for post-traumatic stress disorder: from bench research to confirmation in human trials. Front Neurosci.12, 502 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR9] 9.Simei, J. L. Q. et al. Cannabidiol in anxiety disorders: current and future perspectives. Int Rev. Neurobiol.177, 205–234 (2024). [DOI] [PubMed] [Google Scholar]

[CR10] 10.Blessing, E. M. et al. Cannabidiol as a potential treatment for anxiety disorders. Neurotherapeutics12, 825–836 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR11] 11.Crippa, J. A. et al. Neural basis of anxiolytic effects of cannabidiol (CBD) in generalized social anxiety disorder: a preliminary report. J. Psychopharmacol.25, 121–130 (2011). [DOI] [PubMed] [Google Scholar]

[CR12] 12.Bergamaschi, M. M. et al. Cannabidiol reduces the anxiety induced by simulated public speaking in treatment-naïve social phobia patients. Neuropsychopharmacology36, 1219–1226 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR13] 13.De Morais, E. R. et al. Open-label: the clinical effects of adding cannbidiol to usual care of patients with residual symptoms in the diagnosis of obsessive compulsive disorder. Eur. Psychiatry66, S225–S226 (2023). [Google Scholar]

[CR14] 14.Elms, L. et al. Cannabidiol in the treatment of post-traumatic stress disorder: a case series. J. Alter. Complement Med.25, 392–397 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR15] 15.Das, R. K. et al. Cannabidiol enhances consolidation of explicit fear extinction in humans. Psychopharmacology226, 781–792 (2013). [DOI] [PubMed] [Google Scholar]

[CR16] 16.Lemos, J. I., Resstel, L. B. & Guimarães, F. S. Involvement of the prelimbic prefrontal cortex on cannabidiol-induced attenuation of contextual conditioned fear in rats. Behav. Brain Res.207, 105–111 (2010). [DOI] [PubMed] [Google Scholar]

[CR17] 17.Fogaça, M. V. et al. Effects of intra-prelimbic prefrontal cortex injection of cannabidiol on anxiety-like behavior: involvement of 5HT1A receptors and previous stressful experience. Eur. Neuropsychopharmacol.24, 410–419 (2014). [DOI] [PubMed] [Google Scholar]

[CR18] 18.Marinho, A. L. et al. Effects of intra-infralimbic prefrontal cortex injections of cannabidiol in the modulation of emotional behaviors in rats: contribution of 5HT₁A receptors and stressful experiences. Behav. Brain Res286, 49–56 (2015). [DOI] [PubMed] [Google Scholar]

[CR19] 19.Jurkus, R. et al. Cannabidiol regulation of learned fear: implications for treating anxiety-related disorders. Front Pharm.7, 454 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR20] 20.Bitencourt, R. M., Pamplona, F. A. & Takahashi, R. N. Facilitation of contextual fear memory extinction and anti-anxiogenic effects of AM404 and cannabidiol in conditioned rats. Eur. Neuropsychopharmacol.18, 849–859 (2008). [DOI] [PubMed] [Google Scholar]

[CR21] 21.Bisogno, T. et al. Molecular targets for cannabidiol and its synthetic analogues: effect on vanilloid VR1 receptors and on the cellular uptake and enzymatic hydrolysis of anandamide. Br. J. Pharm.134, 845–852 (2001). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR22] 22.Papagianni, E. P. et al. Cannabidiol prevents spontaneous fear recovery after extinction and ameliorates stress-induced extinction resistance. Int J. Mol. Sci.23, 9333 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR23] 23.Marek, R. et al. The amygdala and medial prefrontal cortex: partners in the fear circuit. J. Physiol.591, 2381–2391 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR24] 24.Burgos-Robles, A., Vidal-Gonzalez, I. & Quirk, G. J. Sustained conditioned responses in prelimbic prefrontal neurons are correlated with fear expression and extinction failure. J. Neurosci.29, 8474–8482 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR25] 25.Sierra-Mercado, D., Padilla-Coreano, N. & Quirk, G. J. Dissociable roles of prelimbic and infralimbic cortices, ventral hippocampus, and basolateral amygdala in the expression and extinction of conditioned fear. Neuropsychopharmacology36, 529–538 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR26] 26.Quirk, G. J. & Mueller, D. Neural mechanisms of extinction learning and retrieval. Neuropsychopharmacology33, 56–72 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR27] 27.Jaehne, E. J. et al. Effects of Npas4 deficiency on anxiety, depression-like, cognition and sociability behaviour. Behav. Brain Res.281, 276–282 (2015). [DOI] [PubMed] [Google Scholar]

[CR28] 28.Zhang, H. et al. Construction of diagnostic model and subtype analysis of major depressive disorder based on PANoptosis key genes. BMC Psychiatry25, 929 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR29] 29.Shen, E. Y. et al. Neuronal deletion of Kmt2a/Mll1 histone methyltransferase in ventral striatum is associated with defective spike-timing-dependent striatal synaptic plasticity, altered response to dopaminergic drugs, and increased anxiety. Neuropsychopharmacology41, 3103–3113 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR30] 30.Guo, Y. P. et al. Global gene knockout of Kcnip3 enhances pain sensitivity and exacerbates negative emotions in rats. Front. Mol. Neurosci.12, 5 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR31] 31.Meng, X. et al. Neurexophilin4 is a selectively expressed α-neurexin ligand that modulates specific cerebellar synapses and motor functions. Elife8, e46773 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR32] 32.Camacho-Arroyo, I. et al. Chemokine profile in women with moderate to severe anxiety and depression during pregnancy. BMC Pregnancy Childbirth21, 807 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR33] 33.Depke, M. et al. Altered hepatic mRNA expression of immune response and apoptosis-associated genes after acute and chronic psychological stress in mice. Mol. Immunol.46, 3018–3028 (2009). [DOI] [PubMed] [Google Scholar]

[CR34] 34.Rimmerman, N. et al. The hippocampal transcriptomic signature of stress resilience in mice with microglial fractalkine receptor (CX3CR1) deficiency. Brain Behav. Immun.61, 184–196 (2017). [DOI] [PubMed] [Google Scholar]

[CR35] 35.Miyako, K. et al. Identification of the antidepressant effect of electroconvulsive stimulation-related genes in hippocampal astrocyte. J. Psychiatr. Res.170, 318–327 (2024). [DOI] [PubMed] [Google Scholar]

[CR36] 36.Zoubovsky, S. P. et al. Neurobehavioral abnormalities following prenatal psychosocial stress are differentially modulated by maternal environment. Transl. Psychiatry12, 22 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR37] 37.Yang, C. et al. Co-Expression network modeling identifies specific inflammation and neurological disease-related genes mRNA modules in mood disorder. Front Genet13, 865015 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR38] 38.Ibi, M. et al. Depressive-like behaviors are regulated by NOX1/NADPH oxidase by redox modification of NMDA receptor 1. J. Neurosci.37, 4200–4212 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR39] 39.Lang, B. et al. Recurrent deletions of ULK4 in schizophrenia: a gene crucial for neuritogenesis and neuronal motility. J. Cell Sci.127, 630–640 (2014). [DOI] [PubMed] [Google Scholar]

[CR40] 40.Liu, M. et al. Ulk4 regulates GABAergic signaling and anxiety-related behavior. Transl. Psychiatry8, 43 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR41] 41.Lovallo, W. R. et al. Cortisol stress response in men and women modulated differentially by the mu-opioid receptor gene polymorphism OPRM1 A118G. Neuropsychopharmacology40, 2546–2554 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR42] 42.Collins, D. et al. Oprm1 A112G, a single nucleotide polymorphism, alters expression of stress-responsive genes in multiple brain regions in male and female mice. Psychopharmacology235, 2703–2711 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR43] 43.Swann, G. et al. Effect of OPRM1 and stressful life events on symptoms of major depression in African American adolescents. J. Affect Disord.162, 12–19 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR44] 44.Hung, Y. F. & Hsueh, Y. P. TLR7 and IL-6 differentially regulate the effects of rotarod exercise on the transcriptomic profile and neurogenesis to influence anxiety and memory. iScience24, 102384 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR45] 45.Dunstan, I. K. et al. Unique pathways downstream of TLR-4 and TLR-7 activation: sex-dependent behavioural, cytokine, and metabolic consequences. Front. Cell Neurosci.18, 1345441 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR46] 46.Girgenti, M. J. et al. Transcriptomic organization of the human brain in post-traumatic stress disorder. Nat. Neurosci.24, 24–33 (2021). [DOI] [PubMed] [Google Scholar]

[CR47] 47.Alexandra Kredlow, M. et al. Prefrontal cortex, amygdala, and threat processing: implications for PTSD. Neuropsychopharmacology47, 247–259 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR48] 48.Phelps, E. A. et al. Extinction learning in humans: role of the amygdala and vmPFC. Neuron43, 897–905 (2004). [DOI] [PubMed] [Google Scholar]

[CR49] 49.Hartley, C. A. & Phelps, E. A. Changing fear: the neurocircuitry of emotion regulation. Neuropsychopharmacology35, 136–146 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR50] 50.Xie, G. et al. Cannabidiol ameliorates PTSD-like symptoms by inhibiting neuroinflammation through its action on CB2 receptors in the brain of male mice. Brain Behav. Immun.119, 945–964 (2024). [DOI] [PubMed] [Google Scholar]

[CR51] 51.Sotres-Bayon, F. & Quirk, G. J. Prefrontal control of fear: more than just extinction. Curr. Opin. Neurobiol.20, 231–235 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR52] 52.Senn, V. et al. Long-range connectivity defines behavioral specificity of amygdala neurons. Neuron81, 428–437 (2014). [DOI] [PubMed] [Google Scholar]

[CR53] 53.Giustino, T. F. & Maren, S. The role of the medial prefrontal cortex in the conditioning and extinction of fear. Front Behav. Neurosci.9, 298 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR54] 54.Corcoran, K. A. & Quirk, G. J. Activity in prelimbic cortex is necessary for the expression of learned, but not innate, fears. J. Neurosci.27, 840–844 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR55] 55.Walker, A. G., Ummel, J. R. & Rebec, G. V. Reduced expression of conditioned fear in the R6/2 mouse model of Huntington’s disease is related to abnormal activity in prelimbic cortex. Neurobiol. Dis.43, 379–387 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR56] 56.Santini, E. et al. Consolidation of fear extinction requires protein synthesis in the medial prefrontal cortex. J. Neurosci.24, 5704–5710 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR57] 57.Daws, S. E. et al. MicroRNA regulation of persistent stress-enhanced memory. Mol. Psychiatry25, 965–976 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR58] 58.Sokoloff, G. et al. Anxiety and fear in a cross of C57BL/6J and DBA/2J mice: mapping overlapping and independent QTL for related traits. Genes Brain Behav.10, 604–614 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR59] 59.Paban, V. et al. Molecular gene expression following blunt and rotational models of traumatic brain injury parallel injuries associated with stroke and depression. J. Transl. Sci.2, 330–339 (2016). [Google Scholar]

[CR60] 60.Zhou, J. & Zhu, L. Shared genetic links between hypothyroidism and psychiatric disorders: evidence from a comprehensive genetic analysis. Front. Endocrinol.15, 1370019 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR61] 61.Bruenig, D. et al. Nitric oxide pathway genes (NOS1AP and NOS1) are involved in PTSD severity, depression, anxiety, stress and resilience. Gene625, 42–48 (2017). [DOI] [PubMed] [Google Scholar]

[CR62] 62.Besnard, A. et al. Targeting Kruppel-like Factor 9 in excitatory neurons protects against chronic stress-induced impairments in dendritic spines and fear responses. Cell Rep.23, 3183–3196 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR63] 63.Gans, I. et al. Klf9 is a key feedforward regulator of the transcriptomic response to glucocorticoid receptor activity. Sci. Rep.10, 11415 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR64] 64.Kamprath, K. et al. Short-term adaptation of conditioned fear responses through endocannabinoid signaling in the central amygdala. Neuropsychopharmacology36, 652–663 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR65] 65.Arifin, W. N. & Zahiruddin, W. M. Sample size calculation in animal studies using resource equation approach. Malays. J. Med. Sci.24, 101–105 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR66] 66.Ju, Y. H. et al. Astrocytic urea cycle detoxifies Aβ-derived ammonia while impairing memory in Alzheimer’s disease. Cell Metab.34, 1104–1120 (2022). [DOI] [PubMed] [Google Scholar]

PERMALINK

Cannabidiol modulates spontaneous recovery through distinct and conserved transcriptomic signatures in mPFC subregions

Minji Son

Inwoong Song

Prajitha Pradeep

Bomi Chang

Matthew J Girgenti

Cha Young Kim

Jae Cheol Jeong

Boyoung Lee

Abstract

Introduction

Results

Cannabidiol reduces spontaneous recovery after extinction in male mice

Fig. 1. Cannabidiol reduces spontaneous recovery after extinction in male mice.

The spontaneous recovery drives distinct gene expression changes in PL mPFC of male mice

Fig. 2. CBD treatment suppresses gene expression changes following spontaneous recovery in the PL region of the mPFC in male mice.

CBD induces distinct gene expression changes in the PL mPFC of male mice following spontaneous recovery

CBD suppresses the gene expression changes induced by the spontaneous recovery in PL mPFC in male mice

Fig. 3. CBD treatment similarly upregulates and downregulates specific pathways in the PL regions of the mPFC in male mice.

The spontaneous recovery drives distinct gene expression changes in IL mPFC of male mice

Fig. 4. CBD treatment suppresses gene expression changes following spontaneous recovery in the IL region of the mPFC in male mice.

CBD drives distinct gene expression changes in IL mPFC of male mice following the spontaneous recovery

CBD suppresses the gene expression changes induced by the spontaneous recovery in IL mPFC in male mice

Fig. 5. CBD treatment similarly upregulates and downregulates specific pathways in the IL regions of the mPFC in male mice.

Distinct PL and IL mPFC gene expression signatures emerge following spontaneous recovery

Cross-species transcriptomic analysis reveals shared DEGs between functionally corresponding human DLPFC and mouse mPFC

Fig. 6. Transcriptomics analysis identifies common DEGs in human and mouse PTSD models.

Discussion

Materials and Methods

Animals

Pharmacological agents

Fear conditioning

Extinction training

mPFC dissection and RNA purification

RNA-seq data analysis

Statistics and reproducibility

Reporting summary

Supplementary information

Acknowledgements

Author contributions

Peer review

Peer review information

Data availability

Competing interests

Footnotes

Contributor Information

Supplementary information

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases