Prophylactic action of ayahuasca in a non-human primate model of depressive-like behavior

Major depressive disorder (MDD) is closely linked to chronic stress and to individual vulnerability versus resilience in stress-response systems. The introduction outlines resilience as an acquired or innate capacity to recover homeostasis after adversity and notes that interventions such as exercise, psychotherapy and some pharmacological agents may promote resilience. Classic serotonergic psychedelics, including ayahuasca, have shown antidepressant effects in clinical and experimental settings, but prior work has focused on treatment rather than prevention. By contrast, some studies have investigated ketamine as an acute prophylactic against stress-induced psychopathology in rodents and humans, but no experimental studies have previously tested classic psychedelics for prophylactic use in depression. This study, therefore, set out to test whether repeated, long‑term administration of ayahuasca before and during a chronic social isolation protocol would increase resilience and prevent depressive‑like outcomes in a validated non‑human primate model. Specifically, Sanna and colleagues hypothesised that prophylactic ayahuasca would prevent the emergence of behavioural signs of depression (notably anhedonia) and blunt the cortisol alterations typically induced by prolonged social isolation in juvenile common marmosets (Callithrix jacchus).

Eighteen juvenile male common marmosets (Callithrix jacchus), aged about 7–9 months, were housed at a university primatology centre under standard husbandry conditions. The study received institutional ethical approval. Animals were randomly assigned to three groups: an Ayahuasca Group (AG, n = 6), an Isolated Group (IG, n = 5) and a Family Group (FG, n = 7). All groups underwent a 4‑week baseline during which behaviour and fecal cortisol were sampled; AG and IG then experienced 9 weeks of social isolation to induce depressive‑like behaviour, while FG remained in family groups for the full 13 weeks. AG received three prophylactic administrations of ayahuasca by gavage: once in the last baseline week (3 days before isolation) and twice during isolation (25 and 50 days after the first gavage). IG received no ayahuasca, and no sham gavages were performed (the authors attribute this change to COVID‑19 related restrictions). Behavioural observations were carried out on alternate days during baseline and daily in the final isolation week; fecal samples for cortisol assays were collected at the same morning time window (around 07:00–08:00) to reduce circadian variation. The ayahuasca brew (same batch for all administrations) consisted of 50% Banisteriopsis caapi and 50% Psychotria viridis boiled and stored refrigerated. Mass spectrometry quantified key constituents (mean concentrations reported as DMT 0.36 mg/mL; harmine 1.86 mg/mL; harmaline 0.24 mg/mL; tetrahydroharmine 0.20 mg/mL). Dose scaling used an allometric approach adapted from human ceremonial use, yielding an administered dose of 1.67 mL per 300 g of marmoset body weight. Gavage was performed by trained personnel. Behaviour was recorded by continuous focal sampling in 30‑minute sessions, using an ethogram adapted to stress contexts. Primary behavioural endpoints included time spent ingesting a 4.16% sucrose solution (used as a translational measure of anhedonia), autogrooming, scratching (anxiety indicator), feeding, locomotion, piloerection and scent‑marking. Fecal cortisol was measured by steroid extraction followed by a competitive ELISA; assay intra‑ and inter‑assay coefficients of variation were 1.69% and 9.96%, respectively. Because of inter‑individual variability, the authors calculated “reactivity” as the percentage change from baseline to the final week (FW – BP) * 100 / BP, using medians and interquartile ranges for descriptive statistics. Variable selection was performed with the Boruta algorithm (a random‑forest based method) to identify behaviours most informative for group discrimination. For variables selected by Boruta (and for cortisol), non‑parametric tests were applied: Kruskal–Wallis to compare the three groups, followed by Mann–Whitney post‑hoc tests. Effect sizes were reported as η2 for Kruskal–Wallis and r for Mann–Whitney comparisons. A two‑tailed p ≤ 0.05 threshold was used for significance.

The Boruta selection process identified time spent ingesting the sucrose solution as the single most important behaviour for discriminating between groups; autogrooming, scratching and food ingestion were placed in an indecision shadow and other behaviours were classified as noise and excluded from further analysis. For the primary behavioural outcome, sucrose ingestion (anhedonia proxy), group differences were statistically significant (Kruskal–Wallis H (2,17) = 12.09, p = 0.002, η2 = 0.673). Post‑hoc comparisons showed that AG animals maintained higher sucrose intake reactivity than IG (U = 0.00, p = 0.006, r = 0.826), while AG and FG did not differ significantly (U = 8.00, p = 0.063, r = 0.515). IG had significantly lower sucrose ingestion than FG (U = 0.00, p = 0.004, r = 0.820). These results indicate that isolated controls developed anhedonia whereas ayahuasca‑treated animals resembled family‑housed controls on this measure. Autogrooming reactivity differed across groups (H (2,17) = 8.33, p = 0.015, η2 = 0.423). AG showed lower autogrooming reactivity than IG (U = 1.00, p = 0.011, r = 0.771) and was similar to FG (U = 15.50, p = 0.431, r = 0.218). Scratching, an index of anxiety, showed group differences (H (2,17) = 6.41, p = 0.040, η2 = 0.295); IG exhibited higher scratching reactivity than FG (U = 1.00, p = 0.007, r = 0.774), while AG did not differ significantly from either IG (U = 8.00, p = 0.201, r = 0.385) or FG (U = 15.00, p = 0.391, r = 0.238). For fecal cortisol reactivity, groups differed (H (2,15) = 6.44, p = 0.04, η2 = 0.342). AG showed greater cortisol reactivity than IG (U(10) = 2.00; p = 0.018, r = 0.716) and did not differ significantly from FG (U(10) = 6.00, p = 0.100, r = 0.495). FG and IG did not differ significantly in cortisol reactivity (U(9) = 8.00, p = 0.347, r = 0.297); nevertheless, the authors note a large effect size for cortisol levels at the final week between FG and IG (reported r = 0.628), suggesting meaningful biological differences despite the lack of statistical significance in that comparison. Overall, AG animals showed cortisol and behavioural reactivity profiles more similar to family‑housed controls than to isolated untreated animals.

Sanna and colleagues interpret the pattern of results as evidence that prophylactic ayahuasca administration increased resilience during a chronic social isolation protocol in juvenile male marmosets. The key finding is that ayahuasca‑treated animals avoided the pronounced reduction in sucrose intake seen in isolated controls, indicating prevention of anhedonia; autogrooming levels in AG were lower than in isolated controls and similar to family‑housed animals, which the authors take as additional indication of a stress‑buffering effect. Scratching was elevated in isolated controls compared with family animals, while AG did not differ significantly from either group; the authors describe this as partial or inconclusive protection for anxiety‑related behaviours and note that prior work found acute ayahuasca reduced scratching. Physiologically, ayahuasca enhanced cortisol reactivity compared with isolated controls, preventing the hypocortisolaemia (HPA axis blunting) that typically follows prolonged isolation in this model; importantly, AG cortisol at the final week resembled family controls rather than indicating sustained hypercortisolaemia. The investigators link these endocrine effects to prior reports that ayahuasca can acutely raise cortisol, modulate HPA function, promote neuroplasticity and reduce inflammation — mechanisms plausibly related to antidepressant and resilience effects. The authors situate their findings relative to prophylactic ketamine studies in rodents, noting methodological differences (for example, many ketamine prophylactic studies did not assess anhedonia). They highlight the translational relevance of using juvenile animals given adolescence as a critical window for social and neural plasticity and the rising incidence of depressive disorders in youth; observational data in adolescent ayahuasca users are cited as consistent with a protective effect. Limitations acknowledged by the investigators include the absence of female subjects, the lack of sham gavages in the stressed control group (a methodological change attributed to SARS‑CoV‑2 pandemic constraints), the modest sample size typical of non‑human primate research, potential competition effects on sucrose access in non‑isolated animals, individual differences in coping strategies, developmental changes within the juvenile stage, and the use of a single dosing/frequency schedule. They recommend future work with larger, sex‑inclusive samples, sham‑control procedures, varying doses and schedules, and finer age stratification. Finally, the authors note that, to their knowledge, this is the first experimental demonstration of a classic psychedelic used prophylactically to attenuate the onset of stress‑related depressive‑like behaviours, and they propose that these preclinical results could motivate exploratory human research in prophylactic applications of psychedelics.