Acute pretrauma ethanol exacerbates PTSD-like phenotype in rats and is reversed by early intranasal ketamine Authors: Bar Eilat Yogev, Gal Levi, Noa Efroni, Amnon C Sintov, Doron Todder, Joseph Zohar, Hagit Cohen Published: 2026 ABSTRACT AbstractAlcohol consumption before trauma is a prevalent but understudied risk factor for posttraumatic stress disorder (PTSD). This study investigated whether acute ethanol exposure prior to trauma modulates PTSD-like symptom development in rats, identified hippocampal mechanisms involved, and tested an early post-trauma intervention. Adult male Sprague-Dawley rats received ethanol (1.6 g/kg, 40% v/v, intraperitoneal) or saline 4 h or 30 min before predator scent stress (PSS) or sham-PSS exposure. Seven days post-exposure, anxiety-like behavior (elevated plus-maze), acoustic startle response, and cue-induced freezing were assessed using validated cut-off behavioral criteria to classify PTSD-like phenotypes. Hippocampal CA1 dendritic morphology was examined in relation to behavioral outcomes. In parallel cohorts, immunofluorescence quantified hippocampal hyperpolarization-activated cyclic nucleotide-gated channel 1 (HCN1), neuropeptide Y (NPY), NPY-Y1 receptor (NPY-Y1-R), and brain-derived neurotrophic factor (BDNF) 1 h post-PSS. A separate ethanol-pretreated group received subanesthetic intranasal ketamine (0.6 mg/kg) via amylolipid nanovesicles (ketamine-ALN) 1 h after PSS. Rats administered ethanol 4 h—but not 30 min—before PSS showed a higher prevalence of PTSD-like phenotypes at 7 days and significantly greater CA1 dendritic retraction. Combined ethanol and PSS exposure was associated with reduced BDNF and NPY levels, increased HCN1, and loss of NPY-Y1-R immunoreactivity, consistent with a hypoexcitable, plasticity-resistant state in the hippocampus. Rats that received early post-PSS ketamine-ALN showed lower cue-induced freezing, a more resilient behavioral profile, and less dendritic atrophy than unloaded-ALN-treated rats. Pre-trauma timed ethanol exposure was associated with a hippocampal “double-hit” involving HCN1 and NPY-Y1-R circuits, together with greater PTSD-like vulnerability. A single early post-trauma intranasal subanesthetic dose of ketamine was associated with a lower prevalence of PTSD-like phenotype, suggesting a promising preventive strategy for alcohol-exposed trauma survivors.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1038/s41598-026-56757-2.Keywords: Posttraumatic stress disorder (PTSD), Animal model, Ethanol, Hyperpolarization-activated cyclic nucleotide-gated channel 1 (HCN1), Neuropeptide Y (NPY), Ketamine, Amylolipid nanovesicles, Dendritic morphology, Dendritic atrophySubject terms: Diseases, Medical research, Neuroscience Introduction The relationship between posttraumatic stress disorder (PTSD) and alcohol use is bidirectional and complex, with high comorbidity between PTSD and alcohol use disorder1. However, whether acute alcohol consumption before trauma exposure increases vulnerability to PTSD remains unclear. In our previous clinical study, we examined how pre-trauma recreational drug use influenced acute trauma-related symptoms among survivors of the terrorist attack at the Nova Music Festival on October 7, 2023. Among approximately 4,000 young adult attendees, many were under the influence of substances such as alcohol, cannabis, MDMA, LSD, cocaine, 3-methylmethcathinone, and polydrug “DOSA” mixtures (a common recreational drug cocktail). Notably, only alcohol—whether consumed alone or in combination with other substances—was associated with significantly higher levels of peritraumatic dissociation, anxiety, depression, and acute stress symptoms2. Clinically, the impact of pre-trauma alcohol consumption on PTSD remains controversial. Some evidence suggests a protective effect3,4, anxiolysis attenuating peri-traumatic anxiety or disruption of traumatic memory consolidation through inhibition of noradrenergic activation and NMDA-mediated currents in the amygdala3,5–8. Others report no association or even increased PTSD rates with pre-trauma alcohol use9,10, consistent with our findings. For example, McFarlane et al.11 found that moderate pre-trauma alcohol consumption predicted lower PTSD symptoms at 1 week and 3 months, whereas low or no consumption was associated with higher symptoms levels at 3 months. These conflicting results underscore pre-trauma alcohol consumption as a potentially modifiable risk factor for PTSD and highlight the need for preventive strategies in at-risk populations. If acute alcohol consumption prior to trauma increases the risk of developing PTSD-related psychopathology, it represents a key modifiable risk factor with important implications for prevention. However, controlled experimental research is needed to determine causal mechanisms. Given the extreme severity and duration of the attack, we used a prospective, controlled rodent model to examine whether acute ethanol exposure before trauma modulates the development of PTSD-like symptoms in rats. Rats were exposed to predator scent stress (PSS) or sham-PSS, with or without prior ethanol administration, and classified according to validated Cut-off Behavioral Criteria (CBC), into extreme (EBR), partial (PBR), and minimal (MBR) behavioral responders12–15. To explore underlying mechanisms, we quantified hippocampal hyperpolarization-activated cyclic nucleotide-gated channel 1 (HCN1) levels, along with brain-derived neurotrophic factor (BDNF), neuropeptide Y (NPY), and NPY-Y1 receptor (NPY-Y1-R), which are key regulators of stress resilience and HCN1–NPY–BDNF circuitry16–19. HCN1, the predominant isoform in the hippocampus, neocortex, and cerebellar cortex, is a hyperpolarization-activated cation channel that opens at membrane voltage more negative than the typical resting potential20,21. Permeable to both Na⁺ and K⁺ and modulated by cyclic nucleotides, HCN1 generates a non-inactivating inward current (Iₕ) at rest22. This current depolarizes the resting membrane potential, lowers membrane resistance, and powerfully regulates intrinsic excitability, dendritic synaptic integration, and plasticity23. Its importance in synaptic integration is further emphasized by a steep gradient of increasing HCN1 expression along apical dendrites, rising with distance from the soma24. HCN1 is implicated in the neurobiology of PTSD25,26 and ethanol/alcohol dependence27,28. In PTSD rodent models, elevated prefrontal/hippocampal HCN1 contributes to anxiety-, depression-, and fear-related behaviors, while HCN1 inhibition (pharmacological or genetic) exerts anxiolytic/antidepressant effects via enhanced BDNF-mTOR signaling and synaptic plasticity25,26. In alcohol-related processes, HCN1 modulates preference, withdrawal-induced anxiety, motivation, and reward; its upregulation occurs in withdrawal states, and inhibition reduces alcohol-seeking and related behaviors27,28. This convergence highlights HCN1 as a candidate link in PTSD–alcohol use disorder comorbidity. We further investigated whether a single subanesthetic intranasal dose of ketamine encapsulated in Amylolipid nanovesicles (ketamine-ALN), administered 1 h post-trauma, would be associated with reduced ethanol-induced PTSD-like phenotypes in ethanol-pre-exposed rats. Ketamine was selected based on the results of experiment 2 and its pleiotropic actions at subanesthetic doses. Beyond blocking NMDA receptors and enhancing AMPA-mediated transmission, ketamine promotes neuroplasticity through upregulation of BDNF signaling and exerts inhibitory effects on HCN1 channels activity29–32. ALNs facilitate direct nose-to-brain delivery, bypassing the blood–brain barrier while enabling sustained drug release33,34. We hypothesized that pre-trauma ethanol exposure would be associated with greater vulnerability to PTSD-like responses, whereas early post-trauma ketamine-ALN administration would be associated with less severe behavioral and molecular abnormalities involving HCN1–NPY–BDNF pathways. Experimental procedures All procedures were approved by the Institutional Animal Care Committee of Ben-Gurion University and complied with the National Institutes of Health Guide for the Care and Use of Laboratory Animals (BGU-358-08-2024E). Experimental procedures and results are reported in accordance with the ARRIVE guidelines35. Animals One-hundred and sixteen adult male Sprague–Dawley rats (150–200 g, Envigo, Israel) were pair-housed under a 12:12 h light/dark cycle (lights on 07:00) at 22 ± 2 °C with ad libitum access to food and water. Rats were acclimated for ≥ 7 days and handled daily. All testing occurred during the light phase. Experimental design Three independent experiments were conducted, designed based on the results of the preceding one. All procedures were performed in a randomized and blinded manner (see Supplementary Information 1.1). Experiment 1: Time-dependent effects of pre-trauma ethanol (n = 78). Rats were randomly allocated into the following groups: Saline + sham-PSS (n = 15; saline administered 30 min–4 h before sham-PSS, n = 7 and 8, respectively. Groups were merged due to no timing-dependent differences). Saline + sham-PSS (n = 15; saline administered 30 min–4 h before sham-PSS, n = 7 and 8, respectively. Groups were merged due to no timing-dependent differences). Ethanol (1.6 g/kg i.p.) 30 min before sham-PSS (n = 10). Ethanol (1.6 g/kg i.p.) 30 min before sham-PSS (n = 10). Ethanol (1.6 g/kg i.p.) 4 h before sham-PSS (n = 8). Ethanol (1.6 g/kg i.p.) 4 h before sham-PSS (n = 8). Saline + PSS (n = 16; saline administered 30 min–4 h before PSS, n = 8 each. Groups were merged due to no timing-dependent differences). Saline + PSS (n = 16; saline administered 30 min–4 h before PSS, n = 8 each. Groups were merged due to no timing-dependent differences). Ethanol (1.6 g/kg i.p.) 30 min before PSS (n = 10). Ethanol (1.6 g/kg i.p.) 30 min before PSS (n = 10). Ethanol (1.6 g/kg i.p.) 4 h before PSS (n = 15). Ethanol (1.6 g/kg i.p.) 4 h before PSS (n = 15). The 30-min and 4-h time points were selected based on previous evidence of rapid, transient NPY fluctuations after acute ethanol36. Following injection, rats underwent a 10-min PSS or sham-PSS exposure. Behavioral testing (elevated plus maze (EPM) and acoustic startle response (ASR) was performed 7 days later. Behavioral responses were retrospectively classified as extreme (EBR), partial (PBR), or minimal (MBR) using validated cut-off behavioral criteria (see Supplementary Information 1.2). Trauma-cue freezing was measured on day 8 using clean cat litter as the cue, and brains were collected on day 9 for Golgi-Cox morphological analysis. The experimental design is depicted in Fig. 1. Time-dependent exacerbation of PTSD-like phenotype by pre-trauma ethanol: (1) Experimental timeline. Rats received ethanol (1.6 g/kg i.p.) or saline either 30 min–4 h before predator-scent stress (PSS) or sham-PSS [(1) saline + sham-PSS n = 15; (2) ethanol 30 min before sham-PSS n = 10; (3) ethanol 30 min before sham-PSS n = 8; (4) saline + PSS n = 16; (5) ethanol 30 min before PSS n = 10; and (6) ethanol 4 h before PSS, n = 15]. Behavioral testing was performed 7 days later using the elevated plus-maze (EPM) and acoustic startle response (ASR) paradigms. On day 8, freezing to a trauma cue was assessed by exposure to clean, unused cat litter in an open-field arena. (A) Anxiety Index calculated from EPM performance. (B) Representative EPM exploration tracks (closed arms, black; open arms, blue). (C) Mean startle amplitude (arbitrary units). (D) Startle response duration (ms). (E–G) Prevalence rates (%) of extreme (EBR), partial (PBR), and minimal (MBR) behavioral response groups according to validated cut-off behavioral criteria (CBC). (H) Percentage freezing to trauma cue in the open field. (I) Representative cumulative locomotor activity patterns across groups. Data Presentation: Panels A, C, D and H show individual data points with mean ± SEM; panels E-G show percentages. Experiment 2: Acute molecular changes (n = 22). Since timing of ethanol administration relative to PSS proved critical in Experiment 1, we hypothesized that exposure to PSS under the influence of ethanol drives the altered behavioral response. To examine the underlying molecular mechanisms, rats received saline or ethanol (1.6 g/kg i.p.) 4 h before PSS or sham-PSS (n = 5–6/group). Brains were collected 1 h post-PSS or sham-PSS (5 h post-administration) for immunohistochemical assessment of hippocampal HCN1, BDNF, NPY, and NPY-Y1-R immunoreactivity. The experimental design is depicted in Fig. 3. Effects of pre-trauma ethanol on CA1 acute molecular changes: Rats received saline or ethanol (1.6 g/kg i.p.) 4 h before PSS or sham-PSS (n = 5/group). One hour post-exposure, brains were collected and hippocampal expression of HCN1, BDNF, NPY, and NPY-Y1-R was quantified by immunohistochemistry. (1) Experimental timeline. Rats received saline or ethanol (1.6 g/kg i.p.) 4 h before PSS or sham-PSS. One hour post-exposure, brains were collected and hippocampal expression of HCN1, BDNF, NPY, and NPY-Y1-R was quantified by immunohistochemistry. (2) Representative images of HCN1 immunoreactivity in the hippocampus which represents the difference analysis that used: (A) Quantitative morphometric analysis of HCN1 density in CA1 sublayers. (B) Quantitative morphometric analysis of HCN1 density in the stratum lacunosum molecular (SLM). (C) Representative images of HCN1 immunoreactivity in the hippocampus across experimental groups. HCN1-positive cells in red; scale bar: 500 μm. (D) Quantitative morphometric analysis of NPY-immunoreactivity in the CA1 region. (E) Representative images of NPY-immunoreactivity in the CA1 across all experimental groups. NPY-positive cells in green; scale bar: 200 μm. (F) Quantitative morphometric analysis of NPY-Y1-R-immunoreactivity in the CA1 region. (G) Representative images of NPY-Y1-R-immunoreactivity in the CA1 across all experimental groups. NPY-positive cells in red; scale bar: 200 μm. (H) Quantitative morphometric analysis of BDNF-immunoreactivity in the CA1 region. Data Presentation: Panels B, D, F and H show individual data points with mean ± SEM. A show mean ± SEM. Experiment 3: Early ketamine-ALN intervention (n = 20). All rats received ethanol (1.6 g/kg i.p.) 4 h before PSS. One hour post-PSS, rats were treated intranasally with either ketamine-ALN (0.6 mg/kg; n = 10) or unloaded-ALN vehicle (n = 10). This dose was chosen based on previous findings suggesting that it normalizes the hippocampal HCN1–BDNF–NPY balance and is associated with greater stress resilience31. Behavioral assessments (EPM, ASR, trauma-cue freezing, and classification into EBR, MBR, and PBR) was performed 7 days later, followed by Golgi-Cox staining on day 9. The experimental timeline is shown in Fig. 4. Effects of ketamine-ALN in ethanol pre-exposed PSS rats: (1) Experimental timeline. All rats received ethanol (1.6 g/kg i.p.) 4 h before PSS. One hour post-PSS, rats were randomly assigned to receive intranasal ketamine-ALN (0.6 mg/kg; n = 10) or unloaded-ALN vehicle (n = 10). Behavioral testing was performed 7 days later using the elevated plus-maze (EPM) and acoustic startle response (ASR) paradigms. On day 8, freezing to a trauma cue was assessed by exposure to clean, unused cat litter in an open-field arena. On day 9, animals were euthanized and brains collected for Golgi–Cox staining. (A) Anxiety Index calculated from EPM performance. (B) Representative EPM exploration tracks (closed arms, black; open arms, blue). (C) Mean startle amplitude (arbitrary units). (D) Startle response duration (ms). (E–G) Prevalence rates (%) of extreme (EBR), partial (PBR), and minimal (MBR) behavioral response groups according to validated cut-off behavioral criteria (CBC). (H) Percentage freezing to trauma cue in the open field. (I) Representative cumulative locomotor activity patterns across groups. Data Presentation: Panels A, C, D and H show individual data points with mean ± SEM; panels E-G show percentages. Drugs Ethanol (40% v/v in saline) was administered intraperitoneally (i.p.) at 1.6 g/kg. This dose is capable of inducing mild behavioral effects in naive rats36,37. Ketamine-ALN (0.6 mg/kg) was prepared as starch–lipid hybrid nanoparticles (see supplementary Information 1.3) and administered intranasally 1 h post-PSS31,34. Intranasal delivery (15 µL/nostril; total 30 µL) was performed under brief isoflurane anesthesia (1–2% via nose cone, 15–30 s), with the head elevated for 15 s to ensure retention31. Predator scent stress Rats were placed individually in a 30 × 30 × 40 cm chamber containing well-soiled cat litter (used for 2 days; feces removed) for 10 min. Sham-PSS rats were exposed to clean, unused litter13,38,39. Behavioral measurements: Tests were conducted between 09:00–14:00 in a sound-attenuated room and recorded using ETHOVISION XT (Noldus, The Netherlands). Elevated plus maze: The maze consisted of a plus-shaped platform with two open and two closed arms (14-cm-high opaque walls on three sides). Rats were placed on the central platform, facing an open arm, and allowed to explore freely for 5 min. Behaviors included time spent in open and closed arms, and central platform, plus arm entries (defined as all four paws entering an arm), and total exploration. An Anxiety Index was calculated by integrating EPM measures as follows: Apparatus was cleaned with 70% ethanol between trials. Acoustic startle response: Startle responses were measured using SR-LAB chambers (San-Diego Instruments, San-Diego, CA, USA). Each rat was habituated to 68 dB background white noise for 5 min, followed by 30 startle signals (110 dB, 40 ms, 30–45 s inter-trial interval). Startle amplitude (arbitrary units) and duration (ms) were averaged across all 30 signals. Trauma-cue–induced freezing: Rats were exposed to unused cat litter for 10 min to simulate the PSS context. Freezing behavior was quantified as the absence of movement excluding respiration for ≥ 1 s and was scored by observers blinded to group allocation. Golgi-Cox staining Tissue Processing and Golgi-Cox Staining: Twenty-four hours post-behavioral testing, animals were deeply anesthetized and transcardially perfused with 0.9% saline. Brains were immediately extracted and processed according to the Golgi-Cox protocol described by Zaqout and Kaindl40(see Supplementary Information 1.4). Neuronal Reconstruction and Morphometric Analysis: Neurons were selected for quantitative analysis based on stringent criteria41: (1) untruncated dendrites, (2) consistent and dense impregnation along the entire dendritic arbor, (3) isolation from neighboring impregnated neurons to avoid analytical interference, and (4) CA1 pyramidal neurons were sampled from the septotemporal axis of the dorsal hippocampus (between 3.7 and 2.2 mm anterior to lambda or between 2.3 and 3.8 mm caudal to bregma). Dendritic branching was quantified from both somatic and dendritic origins. Primary dendrites were defined as direct somatic extensions ≥ 10 μm in length. Only neurons with at least one primary dendrite meeting this criterion were included. Immunohistochemistry Brain slice preparation. Rats were anesthetized and transcardially perfused with ice-cold 0.9% saline, followed by 4% paraformaldehyde in 0.1 M phosphate buffer. Brains were postfixed 12 h at 4 °C, cryoprotected in 30% sucrose for 24 h, frozen and stored at −80 °C. Coronal Sect. (10 μm) were cut using a cryostat (Leica CM 1850; Leica Microsystems, Wetzlar, Germany) and mounted onto coated slides. Staining: Sections were fixed with ethanol, rinsed in phosphate-buffered saline (PBS), blocked with normal goat serum in PBS for 2 h, incubated overnight at 4 °C with primary antibodies against BDNF (1:300, ANT-010, Alomone Labs, Israel), NPY(1–36) (1:100, AB221145, Abcam, Israel), NPY-Y1-R (1:500, NBP1-59008, Novus Biologicals), or HCN1 (1:750, APC-056, Alomone Labs, Israel). After PBS washes, sections were incubated for 2 h with Alexa Fluor 594-conjugated Goat Anti-Rabbit IgG (1:500 and 1:1,000, AB150080, Abcam, UK). For HCN1, sections were permeabilized with 0.3% Triton X-100 in PBS, blocked with BSA, and processed as above. Quantitative Analysis: Immunoreactivity for anti-BDNF, anti-NPY, anti-NPY-Y1-R, anti-HCN1 in the hippocampal CA1 (AP: −3.24 mm, ML: ±1.5 mm, DV: −3.0 mm) was analyzed using a Leica DM4500B microscope. Fiber and cell densities (NPY, HCN1) were quantified using ImageJ (version 1.54f) with standardized settings. BDNF- and NPY-Y1-R-expressing cells were counted using the Cell Counter plugin. HCN1 expression in the stratum lacunosum-moleculare (SLM) of CA1 was quantified by outlining the SLM at 4× magnification and measuring fiber density, or by subdividing CA1 subregions (stratum oriens (SO), stratum pyramidale (SP), stratum radiatum (SR), and SLM) into ten ROIs for mean staining density. Subdividing CA1 into 10 well-defined ROIs along the radial dendritic axis significantly improves spatial resolution, enabling precise quantification of HCN1 distribution and increasing the interpretability of how localized changes in HCN1 density may influence dendritic integration and neuronal excitability (see Supplementary Information 1.5). In each animal, 5–9 tissue samples per region were quantified. Statistical analysis Statistical analyses were performed using Statistica 14.1.0 (TIBCO Software Inc.). Behavioral and morphological measures were analyzed using two-way analysis of variance (ANOVA) (experiment 1) or one-way ANOVA (experiment 3) with post-hoc Bonferroni tests. Sholl analysis and immunohistochemistry used two-way ANOVA. When the interaction term was significant, synergy was confirmed using the Bliss Independence model with one-sample one-tailed t-test against the predicted additive effect42. HCN1 staining intensity in CA1 sublayers was assessed with repeated-measures (RM) ANOVA. Behavioral response prevalence was analyzed using cross-tabulation and Chi-squared tests on raw data. Results Experiment 1: Time-dependent effects of pre-trauma ethanol Elevated plus maze: Two-way ANOVA revealed a significant main effects of PSS (F(1,68) = 30.0, p < 0.0001) and ethanol treatment: (F(2,68) = 3.4, p < 0.000) on the anxiety index, with no significant interaction (Fig. 1A + B). Post-hoc Bonferroni analysis showed that saline + sham-PSS had a significantly lower anxiety index than saline + PSS rats (p < 0.00025). Rats treated with ethanol 4 h before PSS also displayed a significantly higher anxiety index than those given ethanol 30-min before PSS (p < 0.03) or ethanol 4 h before sham-PSS (p < 0.0003). Detailed EPM parameters are provided in Supplementary Information 2.1. Acoustic startle response Acoustic startle response Startle amplitude: Two-way ANOVA revealed a significant main effects of PSS (F(1,68) = 275.6, p < 0.0001) and PSS-ethanol treatment interaction: (F(2,68) = 6.58, p < 0.0025) (Fig. 1D). Post-hoc Bonferroni analysis showed that all PSS-exposed rats displayed significantly heightened startle amplitude relative to their sham controls (saline and ethanol, p < 0.0001). Rats treated with ethanol 4 h before PSS exhibited significantly higher startle amplitude than saline + PSS (p < 0.003). Startle duration: Two-way ANOVA revealed a significant main effects of PSS (F(1,68) = 502, p < 0.0001), ethanol treatment (F(2,68) = 7.6, p < 0.0015), and PSS-ethanol interaction: (F(2,68) = 7.7, p < 0.001) (Fig. 1D). Al PSS-exposed rats displayed significantly heightened startle duration relative to their sham controls (saline and ethanol, p < 0.0001). Rats treated with ethanol 4 h before PSS exhibited significantly higher startle duration than saline + PSS (p < 0.0045) and ethanol-30 min + PSS (p < 0.0001), indicating that rats given ethanol 4 h before trauma showed a more pronounced startle response. Cut-off behavioral criteria: Significant group differences were found in extreme behavioral response (EBR) prevalence (χ² = 21.9, df = 5, p < 0.0006, Fig. 1E). Ethanol 4 h before PSS produced the highest EBR rate (53.3%), significantly higher than ethanol-30 min + PSS (χ² = 4.89, p < 0.03), saline + PSS (χ² = 4.04, p < 0.045), and ethanol-4 h + sham-PSS (χ² = 6.54, p < 0.015). No differences were observed in MBR or PBR prevalence (Fig. 1F and G). For PBR, a near-significant trend emerged (χ² = 10.8, df = 5, p = 0.055), with the ethanol 4 h + PSS group showing 40% PBR, significantly lower than other PSS-exposed groups (p < 0.05). These findings indicate that rats given ethanol 4 h before PSS showed more severe PTSD-like behavioral disturbances than the other groups. Trauma-cue induced freezing: Two-way ANOVA revealed a significant main effects of PSS (F(1,68) = 74.3, p < 0.0001) and PSS-ethanol treatment interaction: (F(2,68) = 3.3, p < 0.0045) (Fig. 1H + I). Post-hoc Bonferroni analysis showed that all PSS-exposed rats displayed significantly higher immobility relative to their sham controls (saline and ethanol, p < 0.005). Additionally, ethanol-4 h + PSS rats showed significantly greater freezing than saline-treated PSS rats (p < 0.05), consistent with greater fear generalization. Morphological analyses of CA1 pyramidal neurons. Guided by the behavioral findings, we examined Golgi-stained CA1 pyramidal neurons from sham-PSS controls (saline-treated) and PSS-exposed rats treated with either saline or ethanol (1.6 g/kg, 4 h before PSS). No group differences were observed in soma diameter. Basal dendrites: Both total basal dendritic number (F(3, 18) = 32.0, p < 0.00015; Fig. 2A + G) and total length (F(3, 18) = 48.7, p < 0.0001; Fig. 2B + G) were significantly reduced in PSS-exposed groups (saline and ethanol) compared with saline + sham-PSS controls (number: p < 0.045 and p < 0.0001, respectively; length: p < 0.0001 for both). Ethanol-treated PSS rats exhibited the greatest reductions, displaying fewer basal dendrites than saline-treated PSS (p < 0.00025) and ethanol-treated sham-PSS groups (p < 0.0001). Sholl analysis revealed significant effects of group (F(3,540) = 236.12, p < 0.0001), radius (F(29,540) = 57.8, p < 0.0001), and their interaction (F(87,540) = 5.3, p < 0.0001; Fig. 2C), with ethanol-treated PSS rats showing the greatest reduction in dendritic intersections. Post-hoc Bonferroni tests confirmed that sham-PSS rats (both saline and ethanol) had more intersections between 90 and 240 μm from the soma than both PSS groups (p < 0.05), and saline-treated PSS rats had more intersections at 130–190 μm than ethanol-treated PSS-exposed rats (p < 0.05). Effects of pre-trauma ethanol on morphological characteristics of CA1 pyramidal neuron dendrites: (A) Quantitative analysis of total basal dendritic number in CA1 pyramidal neurons. (B) Quantitative analysis of total basal dendritic length (µm) in CA1 pyramidal neurons. (C) Sholl-analysis of basal dendritic intersections per 10-µm radial unit distance in CA1 pyramidal neurons. (D) Quantitative analysis of total apical dendritic number in CA1 pyramidal neurons. (E) Quantitative analysis of total apical dendritic length (µm) in CA1 pyramidal neurons. (F) Sholl-analysis of apical dendritic intersections per 10-µm radial unit distance in apical CA1 pyramidal neurons. (G) Computer-generated reconstructions of dendritic trees from CA1 pyramidal neurons across experimental groups. Data Presentation: Panels A–B and D-E show individual data points with mean ± SEM. D–F show mean ± SEM. Apical dendrites: Total dendritic number (F(3, 18) = 11.3, p < 0.00025; Fig. 2D + G) and length (F(3, 18) = 37.7, p < 0.0001; Fig. 2E + G) were also significantly reduced in both PSS groups relative to saline sham-PSS controls (number: p < 0.05 and p < 0.0015, respectively, length: p < 0.0001 for both). Ethanol-treated PSS rats further exhibited lower number of apical dendrites than ethanol-treated sham-PSS rats (p < 0.005). Ethanol-treated PSS rats further exhibited shorter apical dendrites than ethanol-treated PSS rats (p < 0.04) and ethanol-treated sham-PSS rats (p < 0.0001). Sholl analysis showed significant effects of group (F(3,850) = 131.5, p < 0.0001), radius (F(49,850) = 36.85, p < 0.0001), and interaction (F(147,850) = 2.1, p < 0.0001; Fig. 2F). Sham rats (saline and ethanol) displayed more intersections at 90–220 μm than both PSS groups (p < 0.05). Compared with saline-treated PSS rats, ethanol-treated PSS rats showed less intersections at 230–280 μm but fewer at 290–320 μm (p < 0.05 each). Overall, rats given ethanol 4 h before PSS showed more marked morphological alterations in CA1 pyramidal neurons than rats exposed to PSS alone. Experiment 2: Immunohistochemical analyses HCN1immunoreactivity (HCN1-ir) in CA1 sublayers (Fig. 3): HCN1 labeling intensity was quantified as a function of distance from the soma. RM-ANOVA revealed significant main effects of group (F(3, 16) = 8.4, p < 0.0015), sublayer (F(9, 144) = 51.6, p < 0.0001), and a group-sublayer interaction (F(27, 144) = 3.3, p < 0.0001; Fig. 3A-C). HCN1-ir was strongest in the stratum lacunosum-moleculare (SLM), followed by the pyramidal cell layer. Post-hoc Bonferroni tests showed significantly higher HCN1-ir density in SLM sublayers (SLM1–3) in ethanol-treated PSS-exposed rats compared with all other groups (all p < 0.05). Focusing on the SLM (Fig. 3B-C), two-way ANOVA showed significant main effects of PSS (F(1,16) = 48.8, p < 0.0001), ethanol: (F(1,16) = 57.3, p < 0.0001), and a PSS-ethanol interaction (F(1,16) = 5.9, p < 0.0001). Bliss independence analysis confirmed synergistic upregulation (observed: 424% of control vs. expected additive: 200%; one-sample t(4) = 10.04, p < 0.001). Post-hoc Bonferroni tests revealed higher HCN1-ir in ethanol-treated sham rats (p < 0.02), saline-treated PSS rats (p < 0.035), and especially ethanol-treated PSS rats (p < 0.0001) compared with saline sham controls. Ethanol + PSS group exhibited significantly higher density than both saline + PSS (p < 0.0001) and ethanol + sham (p < 0.0001) groups. Thus, rats given ethanol 4 h before PSS showed greater HCN1 upregulation in the SLM than rats exposed to PSS alone. NPY-ir: Two-way ANOVA in CA1 revealed main effects of PSS (F(1,14) = 34.6, p < 0.0001), and ethanol: (F(1,14) = 5.6, p < 0.035), without interaction (Fig. 3D + E). Post-hoc Bonferroni tests showed significantly reduced NPY-ir in saline-treated PSS compared with saline-treated sham controls (p < 0.04) and in ethanol-treated PSS compared with ethanol-treated sham controls (p < 0.0001). Thus, PSS-exposure rats showed lower NPY levels in CA1 regardless of ethanol treatment. NPY-Y1-R ir: Two-way ANOVA showed strong main effects of PSS (F(1,15) = 188.9, p < 0.0001), ethanol: (F(1,15) = 117.3, p < 0.0001), and a significant interaction (F(1,15) = 4.9, p < 0.045; Fig. 3F + G). Bliss Independence confirmed synergistic loss: ethanol + PSS reduced NPY-Y1-R-ir to 22.7 ± 3.2% of control, far below the predicted additive 46.7% (t(4) = −3.34, one-tailed p < 0.01). Post-hoc tests confirmed significant reductions in saline-treated PSS group and ethanol-treated sham group versus saline-treated sham controls (p < 0.0001 for both), with the greatest loss in the ethanol + PSS group compared with both saline-treated PSS and ethanol-treated sham groups (p < 0.0001 for both). BDNF-ir: Two-way ANOVA revealed main effects of PSS (F(1,18) = 7.4, p < 0.015), ethanol: (F(1,18) = 10.1, p < 0.006), and a significant interaction (F(1,18) = 8.8, p < 0.009; Fig. 3H). Post-hoc tests showed significantly reduced BDNF-ir ethanol-treated sham group (p < 0.0025), saline-treated PSS (p < 0.0075), and ethanol-treated PSS (p < 0.0035) compared with saline-treated sham controls. The combined ethanol + PSS group, however, showed significantly less suppression than predicted by the Bliss Independence model (predicted = 48.1% of control; observed = 73.6%; one-sample one-tailed t(5) = 3.44, p = 0.009), indicating antagonistic (sub-additive) interaction on BDNF expression. Experiment 3: Effects of ketamine-ALN in ethanol pre-exposed PSS rats Behavior. In ethanol-pre-exposed PSS rats (1.6 g/kg, 4 h before PSS), those treated with post-PSS intranasal ketamine-ALN (0.6 mg/kg, 1 h after PSS) showed a significantly lower anxiety index than unloaded-ALN rats (F(1, 18) = 5.07, p < 0.04; Fig. 4A + B). Detailed EPM parameters are reported in Supplementary Information 2.2. Ketamine-ALN-treated rats also showed lower acoustic startle amplitude (F(1,18) = 6.2, p < 0.025; Fig. 4C) and shorter startle duration (F(1,18) = 10.4, p < 0.005; Fig. 4D) than unloaded-ALN-treated rats, consistent with less pronounced hyperarousal and anxiety-like behavior. Prevalence rates according to cut-off behavioral criteria. Unloaded-ALN-treated rats showed 50% EBR and 0% MBR, whereas ketamine-ALN-treated rats showed 0% EBR and 50% MBR (χ² = 6.67, p < 0.01 for both; Fig. 4E + F). No difference was observed in partial response prevalence (Fig. 4G). Trauma-cue-induced freezing. In ethanol-pre-exposed PSS rats, ketamine-ALN-treated rats showed significantly less immobility than unloaded-ALN-treated rats (F(1,18) = 9.9, p < 0.006; Fig. 4H + I), consistent with lower freezing behavior. Basal dendrites: ketamine-ALN-treated rats showed significantly greater dendritic number (F(1,8) = 5.7, p < 0.045; Fig. 5A + G) and length (F(1,8) = 29.5, p < 0.001; Fig. 5B + G) than unloaded-ALN-treated rats. Sholl analysis revealed main effects of group (F(1,328) = 17.5, p < 0.0001), radius (F(40,328) = 49.3, p < 0.0001), and interaction (F(40,328) = 1.6, p < 0.02; Fig. 5C). Bonferroni tests showed more dendritic intersections at 80–120 μm in the ketamine-ALN group (p < 0.05). Effects of ketamine-ALN in ethanol pre-exposed PSS rats on CA1 pyramidal dendrites: (A) Quantitative analysis of total basal dendritic number in CA1 pyramidal neurons. (B) Quantitative analysis of total basal dendritic length (µm) in CA1 pyramidal neurons. (C) Sholl-analysis of basal dendritic intersections per 10-µm radial unit distance in CA1 pyramidal neurons. (D) Quantitative analysis of total apical dendritic number in CA1 pyramidal neurons. (E) Quantitative analysis of total apical dendritic length (µm) in CA1 pyramidal neurons. (F) Sholl-analysis of apical dendritic intersections per 10-µm radial unit distance in apical CA1 pyramidal neurons. (G) Computer-generated reconstructions of dendritic trees from CA1 pyramidal neurons across experimental groups. Data Presentation: Panels A–B and D-E show individual data points with mean ± SEM. D–F show mean ± SEM. Apical dendrites: ketamine-ALN-treated rats showed significantly greater dendritic number (F(1,8) = 14.7, p < 0.0055; Fig. 5D + G) and length (F(1,8) = 26.0, p < 0.001; Fig. 5E + G) than unloaded-ALN-treated rats. Sholl analysis revealed main effects of group (F(1,400) = 149.4, p < 0.0001), radius (F(49,400) = 62.9, p < 0.0001), and interaction (F(49,400) = 2.7, p < 0.0001; Fig. 5F). Bonferroni post-hoc tests showed more dendritic intersections at 70–160 μm (p < 0.05) in the ketamine-ALN group. Thus, compared with unloaded-ALN, ketamine-ALN-treated rats showed less CA1 dendritic retraction following ethanol + PSS exposure. Discussion This prospective, controlled rodent study indicates a pronounced time-dependent relationship between pre-trauma ethanol exposure on long-term PTSD-like behavioral disturbances. A single moderate-to-high dose of ethanol (1.6 g/kg i.p., 40% v/v) administered 4 h—but not 30 min—before PSS was associated with more severe sequelae. Specifically, rats given ethanol 4 h before PSS showed a higher prevalence of extreme behavioral responders (EBR; 53%) compared with 18.8% in the saline-PSS group and 10.0% in the ethanol-30 min PSS groups, and greater freezing responses to trauma-related cues (70% versus 54% in the saline-PSS controls). Notably, ethanol did not induce EBR in sham-PSS animals, indicating that acute ethanol exposure per se does not confer vulnerability to PTSD-like outcomes. These behavioral changes were paralleled by morphological alterations in the hippocampus. Ethanol alone induced no detectable changes in CA1 pyramidal neuron morphology, whereas PSS alone caused significant dendritic retraction in CA1 neurons41. Pre-trauma ethanol was associated with more pronounced atrophy, including further reduction in total dendritic length, branching, and Sholl intersections, thereby compromising the structural integrity of stress-resilient hippocampal circuits. One hour after PSS (5 h post ethanol), both ethanol and PSS independently elevated somatodendritic HCN1-ir, with maximal increases in the SLM. Their combination produced synergistic HCN1-ir upregulation. Elevated HCN1 levels would be expected to enhance Iₕ conductance, reduce dendritic input resistance (Rₙ), and suppress dendritic excitability, thereby impairing synaptic integration and long-term potentiation (LTP) at temporoammonic–CA1 synapses20,21,24,43. Such changes may impose a distal dendritic “excitability clamp” that hinders LTP induction at entorhino–CA1 synapses and facilitates fear generalization and resistance to extinction—hallmark features of PTSD-like states. Concurrently, PSS and ethanol independently reduce hippocampal NPY-ir and BDNF-ir. Their combination elicited pronounced synergistic downregulation of NPY-Y1-R-ir in CA1. This loss disrupts a key anxiolytic mechanism whereby NPY-Y1-R on CA1 pyramidal neurons inhibits cAMP/PKA signaling and suppresses HCN and voltage-gated Ca²⁺ channels44,45, thereby exacerbating dysfunction in anxiety- and fear-related circuits. We propose that the synergistic upregulation of HCN1 channels and downregulation of NPY-Y1-R in CA1 neurons may represent a central neurobiological mechanism associated with the greater PTSD-like vulnerability and structural pathology observed after pre-PSS ethanol exposure. This targeted “double hit” to HCN1- and NPY-Y1-R-expressing CA1 neurons provides a plausible explanation for the more severe trauma-related psychopathology associated with ethanol exposure despite its acute anxiolytic properties. The resulting dysregulation initiates a maladaptive positive-feedback loop: excessive HCN1 activity shunts dendritic excitation and elevates the threshold for synaptic plasticity, while loss of NPY-Y1-R signaling abolishes the principal endogenous pathway that physiologically suppresses HCN channels via cAMP inhibition. Consequently, the CA1 network becomes hypoexcitable yet severely impaired in plasticity, strongly favoring the formation and persistence of generalized fear memories—a molecular profile that mirrors the clinical severity of PTSD. Similar mechanistically results have been reported: Lewis et al.46 showed that genetic deletion of TRIP8b, HCN1, or HCN2 in mice reduces Ih and confers antidepressant-like phenotypes, suggesting that inhibition of Ih function may underlie antidepressant activity. Kim et al.25 reported the similar mechanistically results: lentiviral shRNA-mediated HCN1 knockdown in dorsal CA1 increased neuronal excitability, activated BDNF–mTOR signaling, and produced robust anxiolytic- and antidepressant-like effects. Similarly, Giesbrecht et al.47 showed in the basolateral amygdala (BLA) that NPY, via Y1 receptors, inhibits pyramidal neuron excitability by suppressing postsynaptic Ih, whereas corticotropin-releasing factor (CRF) enhances resting Ih and increases excitability. Given that pre-trauma ethanol was associated with greater PTSD-like vulnerability in this model, we evaluated whether a single intranasal subanesthetic dose of ketamine-ALN (0.6 mg/kg) would be associated with attenuation of these effects. Compared with unloaded-ALN controls, ketamine-ALN-treated rats showed less severe behavioral and morphological abnormalities. Retrospective analysis showed that ketamine-ALN-treated rats had 0% EBR and 50% MBR, whereas unloaded-ALN-treated rats had 50% EBR and 0% MBR, indicating a shift toward adaptive stress responses. Ketamine-ALN-treated rats also showed less trauma-cue-induced freezing than unloaded-ALN-treated rats (46.4% vs. 56.6%). Morphologically, ketamine-ALN-treated rats showed greater CA1 dendritic complexity, with more proximal intersections (closer to the soma), suggesting enhanced local connectivity and arborization. In apical dendrites, ketamine-ALN-treated rats showed greater complexity in the SLM (Sholl radii of 300–420 μm), corresponding spatially to the observed molecular changes. Limitations These initial findings were obtained in male rats only. To address this and to explore potential sex differences, we have initiated a parallel study in females. This extension is particularly relevant given the higher PTSD prevalence in women and emerging evidence of sex-specific variations in behavioral responses (e.g., ethanol intake, aversion resistance, stress reactivity) and underlying molecular mechanisms following ethanol exposure. In summary, our study identified four key findings: (1) pre-trauma ethanol (4 h before PSS) was associated, in a time-dependent manner, with more severe PTSD-like behaviors and trauma vulnerability. (2) it was associated with greater PSS-related dendritic retraction in CA1 pyramidal neurons, consistent with reduced integrity of stress-adaptive hippocampal circuits. (3) it was associated with two converging pathological changes in CA1 distal dendrites—an overactive HCN1-mediated excitability clamp restricting synaptic integration and plasticity, combined with loss of NPY-Y1-R signaling that eliminates the primary neuropeptide buffer against anxiety and stress. And (4) post-trauma intranasal ketamine-ALN was associated with less severe behavioral and morphological abnormalities, consistent with partial restoration of neural plasticity. These results position pre-trauma alcohol consumption as a critical modifiable risk factor for PTSD vulnerability and identify intranasal ketamine-ALN as a promising early intervention for trauma-related disorders, opening avenues for targeted preventive in high-risk populations. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Author contributions B.E.Y, G.L, A.C.S, J.Z, D.T and H.C contributed to conception and design of the study; B.E.Y organized the database; B.E.Y, N.E, G.L conducted the experiments; B.E.Y and H.C performed the statistical analysis; B.E.Y, GL and HC wrote the first draft of the manuscript; B.E.Y, G.L, N.E, A.C.S, J.Z, D.T and H.C contributed to manuscript revision and read and approved the submitted version. Funding This research did not receive funding. Data availability All data associated with this study are present in the paper. Raw data are available upon request. In case of request please contact (corresponding author): hagitc@bgu.ac.il. Declarations 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. References Associated Data Data Availability Statement All data associated with this study are present in the paper. Raw data are available upon request. In case of request please contact (corresponding author): hagitc@bgu.ac.il.