Ayahuasca, a psychedelic beverage, modulates neuroplasticity induced by ethanol in mice

Alcohol use disorder (AUD) is a major public‑health problem with high morbidity and mortality and limited effective treatments; relapse rates exceed 50% and are higher when anxiety disorders co‑occur. Previous clinical and preclinical work has suggested that psychedelic substances, including ayahuasca (AYA), may have therapeutic effects in depression, anxiety and substance use disorders. Ayahuasca is a traditional Amazonian decoction combining Psychotria viridis leaves, which contain N,N‑dimethyltryptamine (DMT), and Banisteriopsis caapi vines, which provide β‑carbolines (harmine, harmaline, tetrahydroharmine) that inhibit monoamine oxidase A and permit oral DMT activity. Ethanol affects dopaminergic, serotonergic and opioid systems and produces neuroplastic adaptations; dynorphin (and its precursor prodynorphin) and κ‑opioid receptors have been implicated in anxiety, depression and addiction, and ethanol‑induced behavioural sensitization (EIBS) is a model of ethanol‑related neuroadaptation measured as progressively enhanced locomotor activation following repeated exposure. Almeida and colleagues set out to evaluate whether oral ayahuasca administration modifies ethanol‑induced behavioural sensitization and withdrawal‑related anxiety in mice, and whether AYA alters molecular markers in striatum and hippocampus related to dopamine (D1, D2), serotonin (5‑HT1a, 5‑HT2a) and the dynorphin/prodynorphin system. The study therefore combined behavioural testing (open field locomotion for sensitization and elevated plus maze for anxiety) with biochemical assays (UPLC‑ESI‑MS/MS quantification of AYA alkaloids and Western blotting of receptor and peptide levels) to probe potential neuroplastic effects of AYA on ethanol responses.

Seventy‑two adult male Swiss mice (6–7 weeks old) were used; they were group‑housed under controlled temperature and a 12:12 light:dark cycle. One animal died during the protocol (in the SAL/AYA/ET group). Experimental procedures were approved by the institutional ethics committee. Ethanol was prepared as 15% v/v in saline and administered intraperitoneally at 2.2 g/kg. Lyophilised ayahuasca tea was reconstituted and dosed orally by gavage to deliver 1.76 mg DMT/kg per day. Alkaloid content of the lyophilised AYA was quantified by UPLC‑ESI‑MS/MS using a Waters UPLC system with tandem MS in positive ion mode and multiple reaction monitoring; samples were diluted in ammonium formate/formic acid buffer and injected (5 μL) with DMT‑d6 as internal standard. Sample preparation followed a validated multi‑step dilution procedure. Behavioral apparatus included an open field (40 cm diameter) with automated tracking (EthoVision) for locomotion and an elevated plus maze (two open and two closed arms) for anxiety‑related measures. Grooming and head‑dipping were scored manually from recorded videos by a trained, blinded observer. The behavioural protocol comprised three phases. During acquisition mice received IP injections of ethanol (2.2 g/kg) or saline every other day over nine days (D1, D3, D5, D7, D9) and locomotion was recorded for 10 min after a 5‑minute post‑injection home‑cage interval. During an eight‑day withdrawal/treatment phase mice were given AYA (1.76 mg DMT/kg) or water by gavage daily; on day 7 of this phase animals were assessed on the elevated plus maze (five minutes, six hours after AYA). In the expression phase animals were challenged with a single injection of ethanol or saline and placed in the open field for 10 min to test sensitization expression. Groups followed a factorial design (examples include SAL/SAL/ET and SAL/SAL/SAL) with nominal group sizes around n = 10 reported for some groups. Blood ethanol concentrations were measured by headspace extraction and GC‑FID. For Western blotting, striatum and hippocampus were dissected after euthanasia, homogenised, and protein quantified; specified protein loads (e.g. 75–200 μg depending on target) were separated by SDS‑PAGE, transferred to PVDF membranes and probed with antibodies to D1, D2, 5‑HT1a, 5‑HT2a, dynorphin A and prodynorphin, with β‑actin normalisation. Statistical analysis used repeated‑measures ANOVA for acquisition, factorial ANOVA for expression and elevated plus maze data (with Bonferroni post‑hoc tests), and one‑way ANOVA with Tukey post‑hoc for Western blots. Significance was set at p < 0.05 and data are reported as mean ± SEM.

Alkaloid quantification confirmed the presence of DMT and β‑carbolines in the lyophilised ayahuasca (table values reported in the paper). Habituation locomotor data showed no group differences. During the acquisition phase repeated‑measures ANOVA identified a significant main effect of time (F4,236 = 5.60, p < 0.001) and a treatment × time interaction (F4,236 = 3.69, p < 0.001). Mice repeatedly given ethanol showed significantly greater locomotor activity than saline controls on all acquisition days (p < 0.05) and activity in ethanol‑treated animals increased from Day 1 to Day 9 (p < 0.001), indicating development of behavioural sensitization. Following the withdrawal/treatment period, challenge‑day locomotor activity (expression phase) yielded a significant treatment effect by factorial ANOVA (F1,57 = 6.33, p < 0.001). Specifically, mice that had received ethanol in acquisition and ayahuasca during withdrawal/treatment (ET‑AYA‑ET group) showed significantly lower locomotor activity on challenge than ethanol‑sensitised mice that received vehicle during withdrawal (ET‑SAL‑ET; p < 0.05), indicating attenuation of EIBS by AYA. Plasma ethanol concentrations measured after the final test did not differ among ethanol‑treated groups. Elevated plus maze data showed an acquisition phase × withdrawal interaction for number of open‑arm entries (F1,57 = 5.38, p < 0.01); the ET‑AYA‑ET group had increased entries into open arms compared with SAL‑SAL‑ET, SAL‑AYA‑ET and ET‑SAL‑ET groups (p < 0.01), interpreted by the authors as reduced anxiety‑like behaviour during withdrawal after AYA. Grooming frequency exhibited significant treatment effects during acquisition and withdrawal (both F1,57 = 10.11, p < 0.01) and an acquisition × withdrawal interaction (F1,57 = 9.05, p < 0.01); the ET‑SAL‑ET group showed increased grooming relative to the other groups (p < 0.001). No group differences were observed for head‑dipping. Western blot analyses revealed region‑ and marker‑specific changes. In striatum, D1 receptor levels were reduced in the SAL‑SAL‑ET, SAL‑AYA‑ET, ET‑SAL‑ET and ET‑AYA‑ET groups relative to the SAL‑SAL‑SAL baseline group (p < 0.05), whereas D2 levels showed no differences. Dynorphin and prodynorphin absolute levels did not differ across groups in the striatum, but the dynorphin/prodynorphin (DYN/pDYN) ratio was decreased in SAL‑SAL‑ET, SAL‑AYA‑ET, ET‑SAL‑ET and ET‑AYA‑ET compared with SAL‑SAL‑SAL (p < 0.001). Notably, ET‑AYA‑ET animals showed an increased DYN/pDYN ratio relative to ET‑SAL‑ET (p < 0.001), suggesting partial reversal of ethanol‑induced change by AYA. In the hippocampus, the ET‑SAL‑ET group had elevated 5‑HT1a receptor levels compared with SAL‑SAL‑SAL, SAL‑AYA‑ET and ET‑AYA‑ET groups (p < 0.01). Prodynorphin levels were lower in ET‑SAL‑ET versus SAL‑SAL‑SAL (p < 0.01). The hippocampal DYN/pDYN ratio decreased in SAL‑AYA‑ET (p < 0.01) and ET‑SAL‑ET (p < 0.001) groups compared with SAL‑SAL‑ET. Overall, AYA treatment during withdrawal prevented the ethanol‑associated increase in hippocampal 5‑HT1a and the prodynorphin change, and modified dynorphin/prodynorphin balance in striatum and hippocampus in ways the authors highlight as relevant to anxiety and sensitization.

Almeida and colleagues interpret their findings to indicate that eight days of oral ayahuasca (1.76 mg DMT/kg) attenuated the expression of ethanol‑induced behavioural sensitization in mice and reduced anxiety‑like behaviour during ethanol withdrawal. The authors note that AYA prevented ethanol‑induced upregulation of hippocampal 5‑HT1a receptors and opposed ethanol effects on prodynorphin in the hippocampus and on the dynorphin/prodynorphin ratio in the striatum. They therefore propose that modulation of serotonergic and dynorphinergic systems may underlie the behavioural effects observed. Dose selection and route of administration are discussed: the AYA dose was calibrated to DMT content and chosen to approximate doses used in human clinical studies, and gavage was used to match oral ritual use of ayahuasca. The authors acknowledge that AYA composition can vary by plant source and preparation, justifying quantification of alkaloids in the lyophilised material. They also consider confounding effects of stress: repeated saline injections can produce cross‑sensitization to ethanol, and the authors observed enhanced challenge‑day locomotion in some saline‑treated animals, suggesting stress contributions to sensitization; ayahuasca did not affect this cross‑sensitization but did attenuate ethanol‑induced sensitization itself. The possibility that reduced EIBS after AYA reflects loss of context–drug contingency (a context‑dependent component of sensitization) is raised and not excluded. The discussion situates results relative to prior work, noting concordance with other preclinical reports of AYA or DMT‑related anxiolytic and antidepressant effects and with one study that found inhibition of ethanol sensitization after AYA. The authors highlight that, despite ethanol‑related reductions in striatal D1 receptor levels, AYA did not prevent this D1 decrease, suggesting AYA’s effects on EIBS are not mediated via restoring D1 expression. They emphasise the relevance of dynorphin/κ‑opioid signalling and serotonergic changes to addiction‑related neuroplasticity and withdrawal‑related anxiety, referencing evidence that dynorphin/κ‑opioid upregulation influences anxiety, sensitization and abstinence. Finally, the authors propose that AYA’s capacity to prevent specific ethanol‑induced molecular alterations in hippocampus and to modulate dynorphin/prodynorphin balance in striatum may underlie its behavioural effects. Key limitations acknowledged in the text include variability in AYA composition, potential stress‑related confounds, and the need for further preclinical studies to clarify mechanisms and translational relevance.