Determination of Tryptamines and β-Carbolines in Ayahuasca Beverage Consumed During Brazilian Religious Ceremonies

Ayahuasca is a traditional hallucinogenic beverage typically prepared as a decoction of Banisteriopsis caapi (providing reversible monoamine oxidase inhibitors: harmine, harmaline, tetrahydroharmine) and Psychotria viridis (leaves containing N,N-dimethyltryptamine, DMT). The reversible MAO inhibitors in B. caapi render orally ingested DMT active. In Brazil, consumption of ayahuasca is permitted in a religious context under specific national regulations, which also highlight the need for multidisciplinary research including chemical characterisation and analytical methods for tryptamines and β-carbolines. Such characterisation is important for monitoring use, informing clinical or forensic investigations, and understanding risks from combinations with other psychoactive substances. Santos and colleagues set out to develop and validate a simple, low-cost analytical method to quantify major alkaloids in ayahuasca—DMT, tryptamine, harmine, harmaline, harmalol, and tetrahydroharmine—using solid-phase extraction (SPE) on silica cartridges coupled with liquid chromatography and UV diode-array detection (LC-UV/DAD). The study applied the method to 20 samples taken throughout a two-stage ayahuasca cooking process from a religious group in Fortaleza, Ceará, Brazil, and included structural characterisation of tetrahydroharmine isolated during the work.

The investigators obtained reference standards for tryptamine, harmalol, harmine and harmaline from commercial sources; DMT and tetrahydroharmine were prepared according to cited methods. Stock standard solutions were prepared in methanol (2 mg/mL) and working standards were prepared by dilution. The extracted text contains inconsistent concentration units reported for working standards and calibration ranges; the Method Validation section reports calibration over 0.001–3.0 mg/L. Ayahuasca samples were produced by members of a religious group from freshly harvested B. caapi stems and P. viridis leaves. The cooking process lasted 20 h in the first step, during which hourly samples produced 16 fractions. These 16 fractions were subdivided into five sets, recombined and subjected to a second 4 h cooking step that produced four further hourly ‘‘apuro’’ fractions (samples 17–20). A total of 20 aqueous samples were bottled and sent for analysis; samples 1–16 were processed directly, while the viscous apuro fractions (17–20) were diluted 1:100 prior to SPE to avoid cartridge clogging. Solid-phase extraction used silica cartridges preconditioned with methanol and 1 mol/L HCl. A 20 mL aliquot of ayahuasca (adjusted to pH 8 with 0.01 mol/L NaOH) was percolated at ~0.5 mL/min, dried, then eluted with acidified methanol (pH 3). The eluent was concentrated under nitrogen to 1 mL and 20 μL injections were analysed by HPLC-UV/DAD. SPE here refers to concentrating and cleaning analytes on a solid sorbent prior to analysis. Chromatography employed a reversed-phase Zorbax Eclipse Plus C8 column (150 × 4.6 mm, 5 μm) with a binary gradient of 0.1% formic acid in water (A) and 0.1% formic acid in methanol (B) at 1.5 mL/min. The gradient started at 5% B for 2 min, ramped linearly to 50% B at 30 min, and returned to initial conditions over 10 min. Spectral data were collected across 190–800 nm. The authors report characteristic UV absorption maxima used for identification: 278 nm for DMT and tryptamine, 246 nm for harmine, 320 nm for tetrahydroharmine, and 372 nm for harmalol and harmaline. Tetrahydroharmine isolated in the study was characterised by melting point (199.3 ± 0.2°C), GC–MS (major peak at 18.600 min, molecular ion m/z 216, base peak m/z 201, 89% similarity to library spectrum) and 1H/13C NMR (recorded in CDCl3 on a 400 MHz instrument), supporting its identity. Method optimisation compared different sorbents and solvents and found silica and methanol produced cleaner chromatograms. The pH of extraction was adjusted because indole alkaloids are basic; washing the sorbent with methanol and acidic aqueous solution decreased ion suppression. Acidified methanol provided average recoveries of 71.2–82.5% in optimisation experiments and was selected as the eluent. Validation steps included assessment of linearity, recovery, repeatability (intra-assay precision) and intermediate (between-day) precision, and determination of limits of detection (LOD) and quantification (LOQ). The extracted text reports specific calibration regression equations and correlation coefficients for each compound and provides recovery and precision summaries (see Results).

Linearity of calibration curves was reported across 0.001–3.0 mg/L, with correlation coefficients (r) ≥ 0.9902 for all analytes: harmine r = 0.9905, harmaline r = 0.9904, tetrahydroharmine r = 0.9902, harmalol r = 0.9929, tryptamine r = 0.9943, and DMT r = 0.9919. Regression equations for peak area versus concentration were provided for each compound. Recovery experiments used fortified aqueous samples at 1, 5, 10, 50, 100 and 250 μg/mL with five replicates per level. For most alkaloids, recoveries ranged from 71.7% to 107.4% with RSDs between 1.1% and 9.8%. Harmalol showed lower recoveries (45.0–58.4%) at the 1 and 5 μg/mL fortification levels with RSDs of 2.3–5.5% at those levels. Repeatability (intra-assay) assessments at 100 μg/mL gave RSDs for retention times lower than 0.8%. Intermediate precision over three alternate days (n = 5 per day) reported RSD < 0.3% for all compounds studied, indicating good precision. Limits of detection (LOD) and quantification (LOQ) were calculated using signal-to-noise ratios of 3 and 10, respectively. Reported LODs ranged from 6.8 to 18.8 μg/mL (compound-specific values: tryptamine 6.8 μg/mL, DMT 18.8 μg/mL, harmalol 13.8 μg/mL, harmine 11.6 μg/mL, harmaline 6.8 μg/mL, tetrahydroharmine 17.5 μg/mL). LOQs ranged from 20.6 to 57.1 μg/mL. Application to the 20 ayahuasca samples from the cooking process showed substantial concentrations of the target alkaloids. Reported concentrations for harmine, harmaline, tetrahydroharmine, harmalol and DMT ranged from 0.3 to 36.7 g/L across samples. The four ‘‘apuro’’ fractions (samples 17–20) exhibited increased concentrations, which the authors attribute to recombining fractions 1–16 into the apuro and thereby concentrating DMT and β-carbolines. Tryptamine was not detected in the analysed samples; the authors suggest this absence may reflect complete bioconversion of tryptamine to DMT in the plant material. Structural characterisation of the isolated tetrahydroharmine was consistent with literature values: melting point approximately 199°C, GC–MS spectrum showing molecular ion at m/z 216 and a base peak at m/z 201, and corroborative 1H and 13C NMR data.

Santos and colleagues conclude that the SPE on silica coupled with HPLC-UV/DAD provides acceptable accuracy, precision and selectivity for determination of major DMT and β-carboline alkaloids in ayahuasca prepared by the studied religious group. The authors interpret the marked increases in alkaloid concentrations in the apuro fractions as a consequence of combining earlier cooking fractions, which concentrates active constituents. The study situates its method as a simple, low-cost option suitable for chemical characterisation of ritual ayahuasca and for use in contexts where documentation of alkaloid content is valuable, such as forensic investigation of adverse events, monitoring of ceremonial preparations, or research on pharmacology and toxicology. The absence of detectable tryptamine is discussed as plausibly reflecting plant bioconversion to DMT rather than analytical failure. The authors acknowledge method limitations implicitly through the validation data: harmalol showed lower recovery at low fortification levels, and LOD/LOQ values are compound-dependent. The extracted text contains some inconsistent reporting of concentration units for standards and calibration ranges, which the paper does not explicitly reconcile in the available extraction. The investigators emphasise that analytical characterisation is important given growing use of ayahuasca in religious, recreational and clinical research contexts, and that reliable measurement methods can inform safety assessments and multidisciplinary studies.

The authors state that the proposed SPE–HPLC-UV/DAD method demonstrates acceptable accuracy, precision and selectivity for measuring DMT and major β-carbolines in ayahuasca samples from the described cooking process. They present the method as suitable for routine analysis of these alkaloids in ritual ayahuasca preparations.