Unraveling the epigenomic and transcriptomic interplay during alcohol-induced anxiolysis

The study addresses how acute ethanol induces anxiolysis through epigenomic remodeling that leads to transcriptomic changes in the amygdala. Prior clinical and preclinical work indicates that acute alcohol’s anxiolytic and euphoric effects promote consumption and contribute to AUD development, with the extended amygdala implicated in anxiety regulation and ethanol’s effects. Epigenetic mechanisms (e.g., histone acetylation, HDAC inhibition, and histone methylation changes) are known to mediate acute and chronic ethanol effects. The specific genome-wide chromatin loci and linked gene expression changes in the amygdala following a low dose of ethanol remained unknown. The study’s purpose is to map ethanol-induced chromatin accessibility changes (ATAC-seq) and integrate them with transcriptomic changes (RNA-seq) to identify candidate genes and mechanisms underlying anxiolysis following acute ethanol exposure.

The authors summarize evidence that acute ethanol reduces anxiety-like behavior and that individuals with anxiety are at higher risk for rapid progression to AUD. The extended amygdala regulates anxiety and mediates ethanol’s behavioral and molecular effects. Prior work showed acute ethanol increases histone acetylation (H3K9ac, H4K8ac) via HDAC inhibition and decreases H3K9me2/G9a in the amygdala, with similar increases in histone acetylation observed in other brain regions and liver. These changes undergo neuroadaptation with tolerance and are reversed following withdrawal after chronic ethanol exposure. MicroRNA-494 signaling impacts CBP/p300 and CITED2 and is required for ethanol-induced anxiolysis. The literature supports a central role for epigenetic remodeling (histone acetylation, DNA methylation changes) and transcription factor pathways (e.g., glucocorticoid receptor/NR3C1, CREB) in alcohol-related behaviors, motivating genome-wide mapping of chromatin accessibility and gene expression in the amygdala after acute ethanol.

• Animals and acute ethanol paradigm: Adult male and female Sprague-Dawley rats were group housed with ad libitum food/water on a 12/12-h light/dark cycle. Acute ethanol (1 g/kg, intraperitoneal) or saline was administered. Anxiety-like behavior was assessed 1 h post-injection using the elevated plus maze (EPM) for 5 min. Immediately after behavior or 1 h post-injection (no behavior), rats were anesthetized and brain regions (amygdala including central, medial, basolateral nuclei; BNST; dorsal hippocampus) were dissected and frozen. Blood ethanol levels were measured (males: 89.3 ± 2.9 mg%, n=40; females: 74.5 ± 7.6 mg%, n=8). Unless otherwise noted, experiments were in adult males.
• ATAC-seq: Amygdala tissue from six control and six ethanol-treated rats underwent tagmentation (Illumina Nextera), library amplification, QC (Bioanalyzer), and paired-end sequencing on Illumina platforms. Bioinformatics processing is in supplementary methods.
• RNA-seq: Total RNA from separate cohorts (n=6 control, n=6 ethanol) was extracted (QIAGEN RNeasy). RNA quality (RIN) was high (Control 8.7 ± 0.118; Ethanol 8.9 ± 0.119). RNA from two animals was pooled per library, yielding n=3 per group. Libraries were prepared with TruSeq Stranded mRNA kit and sequenced (100 nt single-end) on HiSeq2500 across three lanes. Bioinformatics methods are in supplementary materials.
• Integration of ATAC-seq and RNA-seq: ATAC peaks were annotated to genes by proximity (promoter ±2 kb of TSS) and distal association (up to 200 kb). Thirty-one genes with promoter-proximal differential peaks (FDR < 0.2) were cross-referenced with RNA-seq differential expression (FDR < 0.2).
• Candidate gene assays: In amygdala, BNST, and hippocampus, qPCR validated mRNA changes for selected genes. ChIP-qPCR assessed active histone marks (H3K9/14ac, H3K27ac) and HIF3A occupancy at specific gene regions (e.g., within ATAC peaks at Hif3a and Slc10a6 promoters). DNA methylation (5mC) at Hif3a and Slc10a6 promoters was measured using MethylMiner followed by qPCR. Fold changes were calculated by ΔΔCt.
• CeA Hif3a knockdown: Rats were bilaterally cannulated targeting the central nucleus of amygdala (CeA). One week later, Hif3a siRNA or control siRNA (1 µg/0.5 µl/side) was infused. About 23 h later, rats received ethanol (1 g/kg) or saline; 1 h later EPM was performed, followed by tissue collection for mRNA.
• Chronic ethanol and withdrawal: Rats received Lieber-DeCarli ethanol or control diet for 15–16 days. Amygdala Hif3a mRNA was measured at 0 h and 24 h withdrawal.
• Statistics: Student’s two-tailed unpaired t test for two-group comparisons; two-way ANOVA with Tukey post hoc for siRNA experiments; Kruskal–Wallis with Tukey for chronic ethanol/withdrawal; outliers by IQR. Whole-genome analyses used methods detailed in supplements.

• Behavior: Acute ethanol produced anxiolytic-like effects in males (increased percent open-arm entries and time on EPM; Fig. S1) and similarly in females (Fig. S3A).
• ATAC-seq (amygdala, acute ethanol vs control): 118,836 total peaks identified. There was a global shift toward increased chromatin accessibility after ethanol. Of 164 significant differential peaks at FDR < 0.05, 148 were more open in ethanol and 16 were more closed. About 12% of all peaks (14,707) were promoter-proximal (±2 kb of TSS). Thirty-one genes had differential promoter peaks at FDR < 0.2, including Hif3a (two promoter peaks) and Slc10a6 (one promoter peak); Hif3a showed the most significant promoter peak. Motif analysis and footprinting of differential peaks (FDR < 0.2; 345 peaks) indicated enrichment of transcription factor motifs including NR3C1 (glucocorticoid receptor), AR, and NR3C2, suggesting TF occupancy changes with ethanol.
• RNA-seq: Sequencing depth 62.6–70.8 million reads/sample. At FDR < 0.2, 91 genes were differentially expressed (heatmap Fig. 2A). Pathway analysis implicated networks in organismal injury, cell death/survival, cellular organization, and tissue development. qPCR validated increased mRNAs (Hif3a, Syt5, Tbr1, Trim54, Robo2, Nptx2, Slc10a6, Dusp1; and Sgk1) and decreased mRNAs (Kcnj13, Mapk13; and P2rx6). Hif3a mRNA was also increased in female amygdala (Fig. S3B) and in BNST and hippocampus (Fig. S2) after acute ethanol.
• Epigenetic validation at Hif3a and Slc10a6 promoters (ATAC peak regions): ChIP-qPCR showed increased H3K27ac at both Hif3a and Slc10a6; increased H3K9/14ac at Hif3a but not at Slc10a6. DNA methylation (5mC) decreased at both promoters, consistent with permissive chromatin facilitating transcription.
• Functional role of Hif3a in CeA: CeA infusion of Hif3a siRNA reduced Hif3a mRNA and blocked ethanol-induced anxiolysis. Hif3a knockdown in ethanol-naïve rats decreased Hif3a mRNA and increased anxiety-like behavior. Slc10a6 mRNA was unchanged by Hif3a siRNA. Anxiety measures two-way ANOVA examples (EPM): percent open arm entries treatment effect F1,37=72.397, p<0.001; group effect F1,37=71.445, p<0.001; interaction F1,37=3.078, p=0.088. Percent time in open arms treatment effect F1,37=97.836, p<0.001; group effect F1,37=90.159, p<0.001; interaction F1,37=10.175, p<0.01.
• Chronic ethanol and withdrawal: Hif3a mRNA in amygdala decreased significantly at 24 h withdrawal compared to control (Kruskal–Wallis H2=7.3, p<0.05), but was unchanged at 0 h withdrawal.
• Downstream target mechanism: Acute ethanol increased HIF3A binding (ChIP) at a predicted hypoxia response element in the Npy1r promoter and increased Npy1r mRNA in amygdala, suggesting HIF3A may drive anxiolysis via NPY1R signaling.

The study demonstrates that a single acute ethanol exposure induces a largely open chromatin state in the amygdala, facilitating DNA-protein interactions and altering gene expression relevant to anxiolysis. Integration of ATAC-seq and RNA-seq pinpointed Hif3a and Slc10a6 as genes with increased promoter accessibility and upregulated transcripts, with locus-specific increases in activating histone acetylation and decreases in DNA methylation. Functional knockdown of Hif3a in the CeA abolished ethanol-induced anxiolysis and increased baseline anxiety-like behavior, indicating that ethanol-driven induction of Hif3a is necessary for the anxiolytic phenotype. The data further link HIF3A to increased Npy1r expression through promoter binding, providing a mechanistic pathway whereby HIF3A may enhance NPY1R signaling to produce anxiolysis. Changes in Hif3a reverse with chronic exposure and withdrawal (decreased expression), paralleling known epigenetic shifts (increased HDAC activity, reduced CBP, reduced histone acetylation) and the emergence of anxiety-like behavior, suggesting neuroadaptation from acute to chronic states. The findings support a model in which acute ethanol’s epigenomic remodeling primes transcriptional programs (including Hif3a) that modulate anxiety circuits within the extended amygdala, with potential broader circuit contributions from BNST and hippocampus and similar effects observed in both sexes.

By combining ATAC-seq and RNA-seq in rat amygdala, the study identifies genome-wide chromatin accessibility and transcriptional changes induced by acute ethanol and establishes Hif3a as an epigenetically regulated driver of ethanol-induced anxiolysis. Increased histone acetylation and decreased DNA methylation at the Hif3a promoter accompany elevated Hif3a mRNA, and Hif3a knockdown in CeA blocks anxiolysis. HIF3A appears to promote Npy1r expression, suggesting a pathway for anxiolytic effects. With chronic ethanol withdrawal, Hif3a expression decreases, aligning with increased anxiety-like behavior. These data highlight HIF3A signaling as a potential therapeutic target for AUD comorbid with anxiety and suggest that acute ethanol-induced epigenomic signatures may forecast genes vulnerable to dysregulation with chronic exposure. Future directions include: mapping long-range enhancer-promoter interactions (e.g., 3C/Hi-C) to link distal ATAC peaks to target genes; cell-type–specific ATAC/RNA-seq within amygdala nuclei; mechanistic dissection of HIF3A isoforms and co-regulators (CBP/p300, CREB); testing pharmacologic modulation of HIF pathway in ethanol-induced behaviors; and extending findings to females and additional brain circuits (BNST, hippocampus).

• Most experiments used adult male rats; female validations were limited to select measures, which may limit sex-specific generalizability.
• RNA-seq libraries pooled RNA from two animals, yielding n=3 per group, which reduces resolution of inter-individual variability.
• Experiments were not performed blindly, potentially introducing bias.
• ATAC-seq and RNA-seq used relatively liberal thresholds for some integrative analyses (e.g., FDR < 0.2), increasing false-positive risk.
• Motif enrichment and footprinting infer TF occupancy but do not conclusively demonstrate specific TF binding without ChIP-seq for each factor.
• CeA tissue punches included surrounding medial and basolateral regions, potentially diluting region specificity.
• Lack of direct functional manipulation of Slc10a6 leaves its behavioral role unresolved.
• Findings from rodents may not directly translate to humans.