Gas chromatographic analysis of dimethyltryptamine and beta-carboline alkaloids in ayahuasca, an Amazonian psychoactive plant beverages

Ayahuasca is a traditional Amazonian psychoactive tea prepared from Banisteriopsis caapi and Psychotria viridis. Psychotria viridis is a source of the psychedelic indole N,N-dimethyltryptamine (DMT), while B. caapi supplies β-carboline alkaloids (primarily harmine, harmaline and tetrahydroharmine) that act as monoamine‑oxidase (MAO) inhibitors. Because peripheral MAO normally inactivates orally ingested DMT, the β‑carbolines permit oral activity by inhibiting MAO; this synergistic combination underlies ayahuasca’s psychotropic effects. Use of the beverage has extended from indigenous ritual contexts into organised religious groups in Brazil and to communities outside South America, prompting growing pharmacological and toxicological interest. Paula and colleagues note that previous analyses often quantified DMT and β‑carbolines by separate techniques (commonly GC-NPD for DMT and HPLC for β‑carbolines), and that reports of simultaneous determination are limited and not widely standardised. The study therefore set out to develop and validate a single analytical method capable of simultaneous extraction and measurement of DMT plus the principal β‑carbolines in ayahuasca preparations, using solid‑phase extraction (C18) followed by gas chromatography with a nitrogen‑phosphorus detector (GC‑NPD). The authors position this as a practical tool for estimating administered doses in animal and human studies for subsequent pharmacological and toxicological work.

This was an analytical methods development and validation study. Reagents included methanol, acetonitrile and standard alkaloids (diphenhydramine as internal standard, tryptamine, harmine hydrochloride and harmaline hydrochloride). Eight different ayahuasca preparations were obtained from a religious group in Araçoiaba da Serra, São Paulo, Brazil and used to test the method. Where commercial standards were not available, tetrahydroharmine and dimethyltryptamine were synthesised by reduction/dimethylation procedures; structures were confirmed by 1H NMR and mass spectrometry. Stock standard solutions (10 mg/mL in methanol) and working solutions (1 mg/mL) were prepared and shown to be stable for at least 30 days at −20°C. Sample extraction used C18 solid‑phase extraction (SPE): 0.5 mL ayahuasca was mixed with borate buffer (pH 9.0, 3.0 mL) and 100 μL internal standard (diphenhydramine, 1.0 mg/mL), then loaded on conditioned Sep‑Pak C18 cartridges. Cartridges were washed (water and acetonitrile‑water 1:9), dried under vacuum, and analytes eluted with 2.0 mL methanol; 1 μL of the eluate was injected into the GC system. Chromatographic analysis employed an Agilent 6890 gas chromatograph with a nitrogen‑phosphorus detector and an HP Ultra‑2 fused‑silica capillary column (25 m × 0.2 mm × 0.33 μm). Ultra‑pure nitrogen was used as carrier gas at 0.6 mL/min in constant flow mode; injections (1 μL) were in split mode (1:20). Injector and detector temperatures were 200°C and 250°C respectively. The oven programme held 150°C for 1 min, ramped at 10°C/min to 250°C, and held at 250°C for 7 min. Validation experiments included determination of limit of detection (LOD) and lower limit of quantification (LLOQ) by analysing serially diluted aqueous samples such that LOD was defined as the lowest concentration with RSD ≤ 20% and LLOQ as the lowest concentration with RSD ≤ 10% across six replicates. Recovery was assessed by comparing processed samples (SPE applied before spiking; set A) with unprocessed samples (standards added after SPE; set B) at three concentrations (0.3, 1.5 and 3.0 mg/mL) in six replicates. Linearity was tested with aqueous calibrators at 0.02, 0.1, 0.5, 1.0, 2.0 and 4.0 mg/mL (triplicates). Precision (intra‑ and inter‑assay) was evaluated across three days using six replicates per concentration and reported as relative standard deviation (RSD or CV). Accuracy measured the percentage of nominal concentration recovered. Stability of analytes was tested in an ayahuasca sample and a spiked water sample (1.5 mg/mL each analyte) after 24 h at room temperature, and extract stability was evaluated on the autosampler carousel over 8 h.

The SPE–GC‑NPD method achieved linear calibration across 0.02–4.0 mg/mL for all four analytes. Reported linear regression equations and coefficients of determination were: DMT y = 8.9651x − 0.6738 (r2 = 0.9971); harmine y = 3.9346x − 0.5288 (r2 = 0.9946); harmaline y = 1.7309x + 0.0057 (r2 = 0.9956); tetrahydroharmine y = 10.052x − 1.5445 (r2 = 0.9941), where y is the peak area ratio (compound/internal standard) and x the calibration concentration. Sensitivity and precision metrics were acceptable for the intended application. The lower limit of quantification (LLOQ) was 0.02 mg/mL for all analytes (determined by the empirical RSD criterion); the LOD values are reported in a table in the original text but are not explicitly stated in this extract. Precision across the studied concentration range was reported as CV < 10%. Recovery from the SPE procedure exceeded 68% for the analytes, and accuracy results (expressed as measured/nominal × 100) were consistent with the validation criteria. Stock standards were stable for at least 30 days at −20°C. Stability tests showed analyte losses < 10% after 24 h at room temperature in both an ayahuasca sample and a spiked water sample, and extracts remained stable on the autosampler carousel for at least 8 h. The method was applied to the eight ayahuasca preparations; chromatograms from a 0.5 mg/mL standard mix and from one sample (sample 2) are described. Substantial variation in absolute alkaloid concentrations was observed across batches obtained from the same origin. The authors attribute this variability to differences in preparation methods, relative amounts and proportions of the source plants, and natural variation in plant alkaloid content. Finally, the authors note a practical limitation: although the method quantifies these alkaloids reliably in ayahuasca preparations, the LLOQ is not low enough to permit quantification in human plasma following ayahuasca ingestion.

Paula and colleagues interpret their findings as demonstrating that a single SPE–GC‑NPD procedure can simultaneously extract and quantify the principal psychoactive alkaloids of ayahuasca (DMT, harmine, harmaline and tetrahydroharmine) with satisfactory linearity, precision and recovery for analyses of beverage samples. They highlight that DMT can be co‑extracted under conditions appropriate for β‑carbolines and measured by GC‑NPD, enabling a simpler workflow than separate GC and HPLC assays. Calibration performance was strong across a broad concentration range, and the method required only a small sample volume (0.5 mL). In positioning this work relative to prior reports, the investigators note that while some studies have used GC–MS or LC‑ESI‑MS to detect mixtures of tryptamines and β‑carbolines, a straightforward validated GC‑NPD approach for simultaneous quantification in ayahuasca preparations had not been detailed. The authors emphasise the practical utility of the method for estimating administered doses in animal and human pharmacological or toxicological studies involving the beverage. Key limitations acknowledged by the study team include insufficient sensitivity to quantify these alkaloids in human plasma after ingestion and variability in alkaloid concentrations across ayahuasca batches, which complicates dose estimation based solely on preparation volume. The LOD values are presented in a table in the paper (not reproduced in the extracted text). The authors suggest the method is fit for purpose for compositional analysis of preparations but would need greater sensitivity (for example via mass spectrometric detection) to support pharmacokinetic measurements in biological fluids. Overall, the authors present the SPE–GC‑NPD method as a validated, practical analytical option for laboratories needing to quantify the main active constituents in ayahuasca teas prior to or alongside further pharmacological and toxicological investigations.