Psilocybin sex-dependently reduces alcohol consumption in C57BL/6J mice

Classical psychedelics, particularly LSD and psilocybin, have been associated with prolonged reductions in alcohol use in observational and controlled clinical studies, and there is growing interest in their therapeutic potential for alcohol use disorder (AUD). The authors note that preclinical animal models can aid target identification and mechanistic investigation; prior rodent studies of psychedelics or 5-HT2A receptor (5-HT2AR) agonist analogs have reported reduced ethanol consumption or preference in at least one experimental condition, with dose-related effects in some cases. The clinical literature shows persistent decreases in drinking that outlast the drugs' metabolic half-lives, motivating preclinical work to probe duration, dose-dependence and mechanisms. Alper and colleagues set out to evaluate the effect of a single intraperitoneal dose of psilocybin on voluntary ethanol consumption and preference in adult male and female C57BL/6J mice. The study aimed to characterise dose–response relationships, the persistence of any effect beyond metabolic clearance, and sex differences, using a continuous two-bottle choice ethanol drinking paradigm that included a short withdrawal and reintroduction phase to test durability of the effect.

Adult male and female C57BL/6J mice (10–18 weeks old) were used; the extracted text does not clearly report the exact group sizes per dose or sex. Animals were singly housed, maintained on a reverse light–dark cycle and given ad libitum chow. Ethanol exposure used a two-bottle choice paradigm: mice were given continuous access to a 20% ethanol bottle and a water bottle for 24 h starting at the dark phase, Monday to Friday, with both bottles water-filled on weekends (2 days withdrawal). Bottle positions were alternated daily to avoid place preference, and evaporation/spillage was measured with control cages lacking mice. Mice underwent 4 weeks of ethanol exposure to acquire consumption before drug administration. On a test Wednesday, mice were assigned to five groups balanced for prior ethanol intake and received a single intraperitoneal injection of saline (vehicle) or psilocybin at 0.1, 0.5, 1.0, or 2.0 mg/kg. Twenty minutes after dosing, ethanol and water bottles were presented and consumption was recorded at 24-h intervals for three consecutive days (days 1–3). Ethanol was then withheld for 2 days (weekend), and ethanol access resumed for two further days (days 4–5) to assess persistence after withdrawal. Two additional assays evaluated potential confounds. A quinine preference two-bottle test (0.06 mM quinine versus water) was performed after a week of quinine habituation; mice received vehicle or psilocybin (0.5, 1.0, 2.0 mg/kg i.p.) and intake was measured for 3 days to assess taste-mediated avoidance. Locomotor activity was measured in alcohol-naïve mice over 24 h beginning at the dark cycle after administration of saline or psilocybin (0.5, 1.0, 2.0 mg/kg) using an infrared beam-based activity sensor, recording ambulatory counts and distance travelled. Statistical analysis was conducted in SAS Version 9.4. Univariate comparisons used the Satterthwaite t-test. For repeated measures, the investigators used multilevel mixed-effects models (MLM) via the MIXED procedure, including fixed effects of time, psilocybin treatment and their interaction, and modelling nested random effects of individual mice within dose groups. The Durbin–Watson test revealed significant first- and second-order autocorrelation, so autoregressive structure was controlled for in the MLMs. Model fit was assessed with AIC, BIC and Cox–Snell adjusted R-square; no violations of MLM assumptions (normality of residuals, heterogeneity, nonlinearity) were reported.

Ethanol consumption and preference: In males, univariate comparisons across the 3-day interval immediately following a single psilocybin dose showed significant reductions in ethanol consumption and preference for 0.5, 1.0 and 2.0 mg/kg; the 0.1 mg/kg dose did not reach significance. Multilevel mixed-model analyses confirmed a significant, dose-related effect of psilocybin on ethanol consumption and preference in male mice that was sustained across days 1–3. Analyses restricted to days 2–3 remained significant in males, indicating the effect was not driven solely by day 1. After the 2-day withdrawal and reintroduction of ethanol, no significant psilocybin effect on consumption or preference was detected in males during the resumed-access interval (two subsequent days). In females, neither univariate nor MLM analyses found significant effects of psilocybin on ethanol consumption or preference at any tested dose or time point. Baseline consumption differed by sex: control (vehicle-treated) females consumed more ethanol than control males averaged across days 1–3 (mean 15.52 ± 3.90 g/kg/day for females versus 10.80 ± 2.15 g/kg/day for males, t = 6.10, p < 0.0001). Ethanol preference did not differ significantly between control sexes (42.08 ± 13.55% in females versus 39.50 ± 7.40% in males, p = 0.3435). Total fluid consumption: Psilocybin effects on total fluid intake were limited. In males, some increases in total fluid consumption were observed at particular doses and time points, but the overall effect across all doses for days 1–3 was not significant; only the 0.5 mg/kg and 1.0 mg/kg doses across days 1–3 reached significance, with the 1.0 mg/kg effect appearing driven by day 1. No decreases in total fluid intake were observed in either sex at any time, which argues against a nonspecific suppression of consummatory behaviour accounting for reduced ethanol intake in males. Quinine preference and locomotor activity: Psilocybin did not alter quinine preference in males or females (males: F = 0.43, p = 0.734; females: F = 0.50, p = 0.685), suggesting taste aversion did not mediate reduced ethanol intake. Ambulatory counts were not significantly affected in males (F = 0.158, p = 0.923) but were significant in females (F = 4.316, p = 0.013), apparently driven by larger counts for the 0.5 mg/kg group during the first 8 h of the 24-h monitoring window. Distance travelled did not differ by treatment in either sex. The authors conclude that motor effects do not account for the male-specific reduction in ethanol consumption.

Alper and colleagues interpret their findings as evidence that a single intraperitoneal dose of psilocybin produces a dose-dependent reduction in voluntary ethanol consumption and preference in male C57BL/6J mice that persists for 3 days but is not evident after a 2-day ethanol withdrawal and reintroduction. The effect was sex-dependent: psilocybin had no detectable impact on ethanol drinking in female mice at the doses and measurement schedule used. The investigators highlight that the persistent reduction in males appears to extend beyond the drug's metabolic clearance in mice, given the reported short half-life of psilocin in mice (~30 minutes). The authors position these results as broadly consistent with other preclinical reports that classical psychedelics reduce alcohol consumption in rodents and with clinical evidence for persistent therapeutic effects of psychedelics in AUD and other substance use disorders. They note, however, differences between studies: for example, prior work found longer-lasting effects for LSD over repeated withdrawal cycles, and a different rat relapse study of psilocybin/LSD found only limited and short-lived effects; methodological differences in species, dosing schedules and ethanol access paradigms (continuous access versus intermittent or relapse models) may explain divergent outcomes. Continuous-access ethanol administration in this study is argued to model aspects of AUD where alcohol is continuously available. Sex differences receive particular attention. Female control mice consumed more ethanol than males and appeared resistant to the treatment effect; the authors suggest this aligns with other reports that female mice often consume greater ethanol quantities and can be less responsive to interventions. Possible explanations discussed include sex differences in psilocybin pharmacokinetics (a cited DOI study reported lower plasma and brain concentrations in female mice), receptor number or signalling differences (including 5-HT2AR and interactions with other receptors), and downstream signalling affecting efficacy. The authors acknowledge that their 24-h interval sampling might have missed transient early effects in females and recommend pharmacokinetic analyses and higher dosing in future work. Mechanistically, the discussion outlines hypotheses rather than definitive conclusions. The investigators reference the concept of classical psychedelics as “neuroplastogens” that may induce persistent changes via neurotrophins, transcriptional and epigenetic mechanisms. They emphasise the relevance of 5-HT2AR signalling and its interactions with metabotropic glutamate receptor 2 (mGlu2R), noting prior reports that psilocybin increases mGlu2R gene expression and that mGlu2R antagonism increases ethanol consumption in rats. The similarity between doses producing mouse head-twitch response (HTR) and doses affecting ethanol consumption is noted as suggestive of shared targets across species. Limitations acknowledged by the authors include the need for independent replication, the absence of pharmacokinetic measurements in the present study, the lack of mechanistic experiments to identify causal targets, and the potential for the 24-h sampling cadence to miss short-lived effects. They also flag the need to test other AUD-relevant models such as vapor-induced dependence or continuous heavy-drinking paradigms. Finally, the authors propose that the sex-dependent result supports the utility of the C57BL/6J strain for studying sex differences in AUD vulnerability and for future neurobiological and drug-discovery studies aimed at developing psychedelic-based pharmacotherapies for substance use disorders.