Metabolism and disposition of N,N-dimethyltryptamine and harmala alkaloids after oral administration of ayahuasca

Ayahuasca is a traditional Amazonian plant brew prepared from Banisteriopsis caapi, which contains reversible monoamine-oxidase A (MAO-A) inhibitors (harmine, harmaline, tetrahydroharmine), together with leaves of Psychotria viridis or Diplopterys cabrerana that provide N,N-dimethyltryptamine (DMT). Earlier clinical and preclinical work established that MAO-mediated oxidative deamination converts DMT into indole-3-acetic acid (IAA), explaining why DMT is ordinarily inactive orally unless MAO is inhibited. However, in vitro and animal studies have also reported alternative DMT pathways (N-oxidation, N-demethylation, cyclization), and preliminary human observations suggested DMT-N-oxide and O-demethylated harmala metabolites can appear after ayahuasca intake. Despite increasing global use, the combined metabolism and urinary disposition of DMT plus harmala alkaloids after ayahuasca ingestion had not been characterised systematically in humans. Riba and colleagues set out to map the urinary metabolites and recovery of DMT and the primary harmala alkaloids following controlled administration of encapsulated freeze-dried ayahuasca to healthy volunteers. The study aimed to quantify unmetabolised parent compounds and specific metabolites (for DMT: IAA, DMT-N-oxide, N-methyltryptamine (NMT), 2-methyltetrahydro-β-carboline (2MTHBC); for harmalas: harmol, harmalol, tetrahydroharmol), to assess the relative contribution of MAO-dependent and MAO-independent metabolic routes and to evaluate conjugation (glucuronidation/sulfation) of harmala metabolites.

Ten healthy male volunteers (mean age 29.0 years, mean weight 67.0 kg), all experienced with psychedelics but naive to ayahuasca, were recruited. Psychiatric screening excluded current or past Axis-I disorders and substance dependence; general health was confirmed by history, laboratory tests and ECG. The study had ethical approval and participants provided informed consent. Ayahuasca was administered as an encapsulated lyophilised preparation derived from a Brazilian batch. The lyophilisate contained per gram: 8.33 mg DMT, 14.13 mg harmine, 0.96 mg harmaline and 11.36 mg tetrahydroharmine (THH); small amounts of harmol (0.30 mg/g) and harmalol (0.07 mg/g) were also present. Dosing was 1.0 mg DMT/kg body weight. The clinical trial used a double-blind, balanced crossover design with three sessions per participant (placebo, 20 mg d-amphetamine, and ayahuasca); this report covers only urine data from the ayahuasca session. Twenty-four-hour urine collections were made after each session and partitioned into intervals 0–4 h, 4–8 h, 8–16 h and 16–24 h. Collected volumes were recorded, pooled, aliquoted and stored at −80 °C. Analyses were performed on samples with and without enzymatic hydrolysis (β-glucuronidase/sulfatase) to assess conjugated forms. Urine (100 µl diluted to 1.0 ml) was analysed by a validated high-performance liquid chromatography–electrospray ionisation tandem mass spectrometry (HPLC–ESI–MS/MS) method using selected reaction monitoring; the method targeted DMT, IAA, DMT-N-oxide, NMT, 2MTHBC and a panel of harmala and related compounds. The proven limit of quantitation (LOQ) was 5 ng/ml for all compounds; limits of detection (LODs) ranged down to 0.07 ng/ml for DMT-N-oxide. Descriptive statistics (mean and SD) were used to report amounts. Percentage recoveries were calculated relative to administered parent compound. Paired-samples t-tests compared enzymatically treated versus non-treated samples. Pearson correlation coefficients explored linear relationships. Statistical significance was set at p < 0.05.

Urine collection volumes averaged 1535 ml (SD 366) after ayahuasca and 1632 ml (SD 519) after placebo, a non-significant difference. Creatinine excretion showed higher mean values after ayahuasca but this difference did not reach significance (p = 0.074). After placebo, all measured compounds were below the LOD except for IAA, which is known to be present physiologically; placebo IAA amounts were subtracted from ayahuasca values to control for baseline. DMT and metabolites: Less than 1% of the administered DMT dose was recovered as unchanged parent compound in 24-h urine. Recovery of parent DMT decreased further after enzymatic treatment, possibly owing to degradation during the hydrolysis step. The predominant DMT-related metabolite measured was IAA (the MAO oxidative deamination product). When expressed as a proportion of the measured DMT-related substances in urine, IAA comprised roughly 80% on average (individual range 43–88%). DMT-N-oxide (DMT-NO) represented about 10% of the administered DMT dose and approximately 20% of the measured tryptamine derivatives (individual range 10–50%). Minor fractions were accounted for by 2MTHBC (cyclization product, 0.13–0.16% of dose) and NMT (together with 2MTHBC about 0.2% of dose); enzymatic treatment increased NMT amounts by about 1.5-fold. Excretion of DMT and its metabolites was heavily front-loaded: 95–97% of all measured DMT (free/total) and the majority of DMT-NO (88%) and IAA (≈66–68%) were recovered within the first 8 h. Harmala alkaloids and metabolites: Tetrahydroharmine (THH) was the most abundant harmala detected in urine, followed by harmaline and harmine. Overall recoveries of parent harmala plus their O-demethylated metabolites varied markedly: harmaline plus harmalol accounted for ≈65% of the harmaline dose, harmine plus harmol ≈28% of the harmine dose, and THH plus tetrahydroharmol only ≈9% of the THH dose. Enzymatic hydrolysis revealed extensive conjugation: harmol increased nearly 50-fold after β-glucuronidase/sulfatase treatment, harmalol increased ≈3-fold and tetrahydroharmol ≈1.5-fold. Free fractions prior to hydrolysis were approximately 2% for harmol, 36% for harmalol and 68% for tetrahydroharmol. Calculations comparing ingested and excreted levels indicated that urine contained far more harmol and harmalol than were present in the administered tea (approximately 1,028% and 516% of the amounts in the lyophilisate), implying formation via metabolism of harmine and harmaline. The temporal profile of harmala excretion was more distributed over 24 h than for DMT: 45–53% of harmine, 35–39% of harmaline and 32–33% of THH (free/total) were excreted in the first 8 h. Significant positive correlations were observed among the metabolic ratios harmol/harmine, harmalol/harmaline and tetrahydroharmol/THH (r values 0.800–0.967, p < 0.01 to p < 0.001), but no significant correlations were found between harmala measures and DMT or DMT-metabolite measures.

The investigators interpret their findings as confirming that MAO-mediated oxidative deamination is a prominent route for DMT metabolism in humans, given the large IAA excretion observed even when DMT was taken with harmala MAO-A inhibitors. At the same time, the presence of substantial amounts of DMT-N-oxide and smaller amounts of N-demethylation and cyclization products demonstrates that MAO-independent pathways substantially contribute to DMT degradation when DMT and harmalas are co-administered. The early appearance of IAA in urine suggests that MAO inhibition by harmala alkaloids is partial or transient, and that this partial inhibition nonetheless is sufficient to permit psychoactive effects of orally administered DMT. Comparing with prior in vitro and animal data, the authors note consistency with reports that N-oxidation is an important compensatory pathway when MAO is inhibited; in this study DMT-NO represented about 20% of measured tryptamine derivatives. They cite evidence that MAO inhibition can shift DMT metabolism toward N-oxidation and related pathways, and propose that future studies administering oral DMT without harmalas would help quantify the degree of metabolic shift induced by MAO inhibition. For the harmala alkaloids, the results indicate extensive O-demethylation followed by conjugation (glucuronidation/sulfation), with large inter-metabolite differences in the degree of conjugation. The study reports for the first time in humans that tetrahydroharmol (the O-demethylation product of THH) is formed in vivo. Nonetheless, total recoveries of harmala parent plus O-demethylated metabolites were lower than expected for some compounds, particularly THH. The authors suggest possible explanations including intense first-pass metabolism of THH limiting systemic availability, alternative metabolic routes (for example hydroxylation products), and interindividual variability; despite these uncertainties, the strong correlations among O-demethylation ratios point to a shared enzymatic pathway for O-demethylation of the harmalas. The authors acknowledge limitations implicit in urinary recovery studies: analytical and hydrolysis conditions can affect measured parent compound and metabolite amounts (the enzymatic hydrolysis step appeared to reduce measured parent DMT), the sample size was limited (N = 10), and the design involved co-administration of harmalas so cannot isolate DMT metabolism under physiological (uninhibited MAO) conditions. They recommend targeted studies of oral DMT alone to determine baseline metabolic partitioning and to quantify how harmala co-administration shifts metabolism.

Riba and colleagues conclude that when DMT is administered orally together with harmala alkaloids in ayahuasca, N-oxidation is a major degradation pathway in humans in addition to MAO-catalysed oxidative deamination. For the harmala alkaloids, O-demethylation followed by conjugation is an important metabolic route, though other degradation pathways likely contribute as well. The authors propose future investigations of oral DMT without harmalas to characterise the baseline metabolic profile and quantify the degree of metabolic 'shift' induced by harmala-mediated MAO inhibition.