Toxicometabolomics Characterization of Two N1-Sulfonated Dimethyltryptamine Derivatives in Zebrafish Larvae and Human Liver S9 Fractions Using Liquid Chromatography–High-Resolution Mass Spectrometry

TL;DR

This study uses advanced chemical analysis to explore how two new psychoactive substances are processed in zebrafish and human liver samples, revealing important metabolic changes.

Contribution

The study introduces a toxicometabolomics approach to identify novel psychoactive substance metabolites and their effects in zebrafish and human liver fractions.

Findings

Metabolic pathways and key biotransformations of two N1-sulfonated dimethyltryptamine derivatives were identified in zebrafish and human liver S9 fractions.

Exposure to the compounds caused significant downregulation of L-threonine, indicating potential biochemical impacts.

Abstract

Introduction: The availability of toxicokinetic data is critical for detecting and monitoring the intake of psychoactive substances. Timely characterization of novel psychoactive substances (NPS) is particularly important to assess their abuse potential and inform public health responses. Methods: Toxicometabolomics offers a powerful approach to characterize xenobiotic metabolism through high-resolution profiling of biochemical transformations. It thus allows the finding of exogenous biomarkers, such as new drug metabolites, and endogenous biomarkers, which could be indications of acute drug ingestions or sample manipulation, as well as offering information on the mode of action of drugs. In this study, we applied a liquid chromatography–high-resolution mass spectrometry workflow to investigate the toxicometabolomics of two N1-sulfonated N,N-dimethyltryptamine derivatives with potential for both therapeutic use and recreational abuse. Results: Zebrafish (Danio rerio), an increasingly valuable model for preclinical pharmacology and toxicology studies, along with pooled human liver S9 fractions were used to elucidate metabolic pathways and identify key phase I and phase II biotransformations. Furthermore, untargeted metabolomics revealed significant downregulation of L-threonine associated with compound exposure. Conclusions: These findings advance the current understanding of tryptamine metabolism and underscore the utility of toxicometabolomics in the analytical evaluation of NPS.

Figures

Full text

1. Introduction

Between 2005 and 2024, the European Monitoring Centre for Drugs and Drug Addiction (EMCDDA) reported approximately 950 new psychoactive substances (NPS), including over 50 synthetic tryptamines [1]. Due to the structural diversity of NPS and the lack of toxicokinetic information including metabolic fate, the detection of an intake is an analytical challenge. Furthermore, knowledge about the detailed mode of action and possible off target effects of drugs (of abuse) and NPS is often missing. In clinical and forensic toxicology as well as doping control, knowledge about the toxicometabolomics of new drugs (of abuse) is important for their risk assessment in general and for monitoring their ingestion [2]. Toxicometabolomics, as a subdiscipline of metabolomics, has increasingly gained interest in toxicological studies of drugs/drugs of abuse [3,4,5,6,7,8,9,10]. The application of untargeted toxicometabolomics allows the identification of exogenous biomarkers, such as new drug metabolites, and endogenous biomarkers, which could be indications of acute drug ingestions or sample manipulation, as well as offer information on the mode of action of drugs [11,12,13]. Studying the metabolism allows the evaluation of the chemical modifications in their ability to enable prodrug functionality and toxicokinetic studies [14].

Danio rerio is a model organism well suited for such preclinical studies. It offers the unique advantage of being an in vivo model for which regulations on animal experimentation do not apply (under European Union Directive 2010/63/EU) up to five days post fertilization (dpf). Compared to rodent or mouse models, the ease of handling, low maintenance costs, and compliance with the 3Rs principle make it an appealing model. At three dpf, the zebrafish larvae (ZFL) develop a functional liver capable of performing phase I and phase II biotransformation, which makes it a suitable model especially for metabolism studies [15]. Moreover, it has been successfully used as a model to investigate metabolism in various studies with small molecules in the preclinical stages [16,17,18,19,20]. In addition to its metabolic system, its physiological and genetic similarity to humans makes it highly suitable for toxicometabolomics studies, especially when combined with comprehensive analyses using human liver S9 fractions [14].

N,N-dimethyltryptamine (DMT), a substituted tryptamine, is a hallucinogenic substance which has been in consumption for thousands of years, produced mainly through botanical preparations for traditional medicine or spiritual practices [21]. It is also considered to be a psychoactive compound for the treatment of various conditions in the anxiety spectrum [22]. Its simple molecular structure, low molecular weight, and hydrophobic nature allow DMT to cross the blood–brain barrier (BBB) [23]. However, DMT is not orally bioavailable due to rapid deamination by intestinal and hepatic monoamine oxidase-A (MAO-A) [24,25]. Administration in conjunction with MAO-A inhibitors is associated with negative side effects such as high blood pressure, nausea, and hyperthermia [22,26]. Therefore, structural modifications to DMT were proposed to improve the therapeutic properties. Furthermore, DMT is classified as a scheduled drug in most countries due to its adverse psychedelic effects, meaning that the possession, sale, transport, and cultivation of DMT is illegal. However, prodrugs of the molecule are typically not regulated. Therefore, these molecules have the potential to be abused for recreational use, and studying the metabolism of these molecules should allow to identify the ingestion of these compounds in clinical and forensic toxicology as well as doping control [1].

The current study should give insights into the in vivo toxicometabolomics and in vitro metabolism of two DMT derivatives—DMT esylate and DMT mesylate (Figure 1), hereafter referred to as DMT-E and DMT-M respectively, which were designed for their improved pharmacokinetic properties compared to DMT, and to potentially act as DMT prodrugs. The compounds are N_1_-sulfonyltryptamine derivatives, which are known to be antagonists of the human 5-HT_6_ receptor, which has been linked to various neuropsychiatric disorders [27]. It is a well-known strategy to utilize the sulfonamide moiety to impart metabolic stability to drug candidates, since sulfonamides are generally not hydrolyzed by common mammalian metabolic enzymes such as esterases or amidases, and their chemistry makes them inherently stable [28]. Therefore, it is envisioned that they are not subject to the kind of first pass metabolism observed for DMT. Cleavage of the sulfonamide group is rare, but nevertheless still enzymatically possible via a Glutathione-S-transferase mediated pathway [29], which may potentially impart prodrug character to the compounds. Although N_1_-sulfonated DMTs are not expected to function as psychedelic substances due to the blocking of the N_1_ amine group key to inducing a psychedelic response [30,31], they are known to have promising therapeutic effects in cognition enhancement and memory modulation as a result of 5-HT_6_ binding [32,33], which leads to their potential as substances of abuse.

2. Materials and Methods

2.1. Chemicals and Reagents

DMT-E and DMT-M were provided by MiHKAL GmbH (Basel, Switzerland). Dimethyl sulfoxide (DMSO) and tricaine (3-amino-benzoic acid ethyl ester) were obtained from Sigma-Aldrich (Taufkirchen, Germany). Methanol (LC-MS grade), acetonitrile (LC-MS grade), and formic acid (LC-MS grade) were from VWR (Darmstadt, Germany). NaCl, KCl, MgSO_4_, Ca(NO_3_)2, and HEPES were obtained from Carl Roth (Karlsruhe, Germany). The working solutions of the compounds were freshly prepared prior to each experiment. The 24-well plates and 6-well plates were purchased from Sarstedt (Nümbrecht, Germany). ZFL of the AB wild-type line were initially obtained from the Luxembourg Center for Systems Biomedicine (Belvaux, Luxembourg). Dry small granulate food was purchased from SDS Deutschland (Limburgerhof, Germany), and Artemia cysts (>230,000 nauplii per gram) were obtained from Coralsands (Wiesbaden, Germany). NADP-Na_2_, acetonitrile (LC-MS grade), and methanol (LC-MS grade) were obtained from VWR (Darmstadt, Germany), 3′-phosphoadenosine-5′-phosphosulfate (PAPS), S-(5′-adenosyl)-l-methionine (SAM), dithiothreitol (DTT), reduced glutathione (GSH), acetylcarnitine transferase (AcT), acetylcarnitine, acetyl coenzyme A (AcCoA), MgCl_2_, K_2_HPO_4_, KH_2_PO_4_, superoxide dismutase, isocitrate dehydrogenase, isocitrate, ammonium formate, and formic acid from Sigma (Taufkirchen, Germany). Water was purified using a Millipore filtration unit (18.2 Ω × cm water resistance) from Sigma (Taufkirchen, Germany) Pooled human liver S9 fraction (pHLS9, 20 mg protein/mL, from 30 individual donors), UGT reaction mix solution A (25 mM UDP-glucuronic acid), and UGT reaction mix solution B (250 mM Tris-HCl, 40 mM MgCl_2_, and 0.125 mg/mL alamethicin) were obtained from Corning (Amsterdam, The Netherlands). After delivery, the pHLS9 were thawed at 37 °C, aliquoted, snap-frozen in liquid nitrogen, and stored at −80 °C until use.

2.2. In Vitro Metabolism Using Pooled Liver S9 (pHLS9) Fraction

2.2.1. Sample Preparation

Investigation of the in vitro metabolism using pHLS9 was based on previous publications [34,35,36]. Briefly, pHLS9 was used at a final protein concentration of 2 mg/mL. The incubation was performed after a 10 min preincubation at 37 °C with 25 µg/mL alamethicin (UGT reaction mix solution B), 90 mM phosphate buffer (pH 7.4), 2.5 mM Mg^2+^, 2.5 mM isocitrate, 0.6 mM NADP+, 0.8 U/mL isocitrate dehydrogenase, 100 U/mL superoxide dismutase, 0.1 mM AcCoA, 2.3 mM acetyl carnitine, and 8 U/mL carnitine acetyltransferase. Subsequently, 2.5 mM UDP-glucuronic acid (UGT reaction mix solution A), 40 µM aqueous PAPS, 1.2 mM SAM, 1 mM DTT, 10 mM GSH, and 25 µM substrate in distilled water were added. The final volume of the incubation mixture was 150 µL. All given concentrations are final concentrations. Reactions were started by the addition of substrate and incubated for a maximum of 480 min. After 60 min, 60 µL of the mixture was transferred into a reaction tube and reactions were terminated by addition of 20 µL ice-cold acetonitrile containing 5 µM alpha-PVP as an internal standard. The remaining mixture was incubated for additional 7 h and then stopped by the addition of 30 µL ice-cold acetonitrile containing 5 µM alpha-PVP. Additionally, blank samples not containing pHLS9, but the according amount of phosphate buffer were incubated to identify compounds of non-metabolic origin. Blank samples not containing substrate were incubated as well. The solutions were cooled for 30 min at −18 °C, centrifuged for 2 min at 14,000× g, and finally 60 µL of the supernatants were transferred into an autosampler vial. One µL supernatant was injected onto the LC-HR-MS/MS system as described below. All incubations were done in duplicates.

2.2.2. Data Acquisition (LC-HRMS/MS Conditions) and Evaluation for Identification of Metabolites

The analysis was performed using a Thermo Fisher Scientific (TF, Dreieich, Germany) Dionex UltiMate 3000 RS pump consisting of a degasser, a quaternary pump, and an UltiMate Autosampler, coupled to a TF Q Exactive Plus system equipped with a heated electrospray ionization HESI-II source. Mass calibration was done prior to analysis according to the manufacturer’s recommendations using external mass calibration. The conditions were set according to published procedures [34,37,38]. Gradient elution was performed on a TF Accucore Phenyl-Hexyl column (100 mm × 2.1 mm, 2.6 µm). The polar mobile phase (eluent A) consisted of 2 mM aqueous ammonium formate, containing acetonitrile (1%, v/v) and formic acid (0.1%, v/v, pH 3). Eluent B consisted of 2 mM ammonium formate dissolved in acetonitrile: methanol (1:1, v/v) containing water (1%, v/v) and formic acid (0.1%, v/v). The flow rate was set from 1 to 10 min to 500 µL/min and from 10 to 13.5 min to 800 µL/min using the following gradient: 0–1.0 min 99% A, 1–10 min to 1% A, 10–11.5 min hold 1% A, and 11.5–13.5 min hold 99% A. Samples were also analyzed using hydrophilic interaction chromatography (HILIC) as described under 2.3.4. The HESI-II source conditions were as follows: ionization mode, positive; sheath gas, 53 AU; auxiliary gas, 14 AU; sweep gas, 3 AU; spray voltage, 3.50 kV; heater temperature, 438 °C; ion transfer capillary temperature, 320 °C; and S-lens RF level, 50.0. Mass spectrometry was performed using full scan (FS) data and subsequent data-dependent acquisition (DDA). The settings for FS data acquisition were as follows: resolution, 35,000 at m/z 200; microscans, 1; automatic gain control (AGC) target, 1 × 10^6^; maximum injection time, 120 ms; and scan range, m/z 50–750. The settings for the DDA mode were as follows: dynamic exclusion, not used; resolution, 17,500 at m/z 200; microscans, 1; loop count, 5; AGC target, 2 × 10^4^; maximum injection time, 250 ms; isolation window, m/z 1.0; high collision dissociation (HCD) with stepped normalized collision energy (NCE), 17.5, 35, and 52.5; spectrum data type, profile; and underfill ratio, 1%. TF Xcalibur Qual Browser software version 3.0.63 was used for data handling.

2.3. In Vivo Toxicometabolomics Using ZFL

2.3.1. ZF Husbandry

ZF husbandry and all experiments with ZFL were executed according to EU Directive 2010/63/EU and the German Animal Welfare Act (§11 Abs. 1 TierSchG), in which all works were accomplished by following internal standard operating procedures (SOPs) based on published standard methods. Adult ZF were kept in an automated aquatic ecosystem (PENTAIR, Apopka, FL, USA), which is a continuous and real-time monitoring system under the following conditions: temperature (27 ± 0.5 °C), pH (7.0 ± 0.1), conductivity (800 ± 50 µS), and light–dark cycle (14 h/10 h). Fish were fed twice a day with dry, small granulate food and freshly hatched live Artemia cysts once per day. The ZFL medium (0.3 × Danieau’s solution) was composed of 17 mM NaCl, 2 mM KCl, 0.12 mM MgSO_4_, 1.8 mM Ca(NO_3_)2, 1.5 mM HEPES, pH 7.1–7.3, and 1.2 µM methylene blue. To reproduce ZFL, the ZF pairs were kept overnight in standard mating cages, separated by gender. The following morning, the adult ZF started spawning as soon as the separators were removed. All fertilized eggs of ZF were collected and sorted using a Zeiss Stemi 508 stereo microscope (Carl Zeiss Microscopy GmbH, Jena, Germany). Experiments were carried out with wild-type AB embryos within the first 120 h post fertilization (hpf) as these early life stages are not considered as animal experiments according to the EU Directive 2010/63/EU. At a maximum of 120 hpf, embryos were euthanized by submersion in ice water. All ZFL were raised in an incubator at 28 °C, with a daily medium change to clean the ZFL cultures. The ZFL at 4 dpf were used for drug metabolism studies.

2.3.2. Maximum Tolerated Concentration (MTC) Determination and Bioavailability Estimation

Each compound was treated with 10 ZFL (4 dpf) at 50, 100, and 500 µM concentrations. Naloxone at 300 µM was also done as a treatment as a positive control. Naloxone is known to be tolerated by ZFL upon medium treatment at 300 µM, and its spatial distribution within ZFL has also been studied to validate its bioavailability [19]. This makes it a suitable candidate to be used as a positive control to validate sample preparation steps (described below). Treatment solution contained the compound in 0.3× Danieau’s solution with a maximum of 1% DMSO. Treatment was done in a 24-well plate with 1 mL solutions. The ZFL were incubated at 28 °C for 24 h and observed for deaths, malformations, or any visible phenotypic changes under a Zeiss Stemi 508 stereo microscope. The maximum tolerated concentration among the treated concentrations was identified and the ZFL at this concentration were pooled for estimating bioavailability.

For the extraction, the ZFL were washed with 0.3× Danieau’s and transferred to an Eppendorf tube. Excess medium was removed using a pipette, the ZFL snap-frozen, and lyophilized for at least 4 h. A volume of 80 µL methanol was then added and the tube was vigorously vortexed for 2 min, followed by centrifugation at 15,000× g for 15 min. The supernatant was transferred to a new Eppendorf tube and then sampled for LC-HRMS/MS analysis. All supernatant was transferred to an autosampler vial and 5 µL was injected onto the LC-HRMS/MS system consisting of a Dionex Ultimate 3000 RSLC system (Thermo Fisher Scientific, Germering, Germany) coupled to a maXis 4G HR-QTOF mass spectrometer (Bruker Daltonics, Bremen, Germany) with an Apollo II ESI source. Separation was carried out on a Waters ACQUITY BEH C_18_ column (100 × 2.1 mm, 1.7 µm) equipped with a Waters VanGuard BEH C_18_ 1.7 µm guard column at 45 °C using 0.1% formic acid in water (v/v, eluent A) and 0.1% formic acid in acetonitrile (v/v, eluent B) at a flow rate of 600 µL/min. The linear gradient mode was as follows; 0–0.5 min, 5% eluent B; 0.5–18.5 min, 5–95% eluent B; 18.5–20.5 min, 95% eluent B; 20.5–21 min, 95–5% eluent B; and 21–22.5 min, 5% eluent B. A 5 µM pure compound solution in methanol was also run to gauge the ionization behavior of the pure compound.

2.3.3. Experimental Setup for Identification of Metabolites

For each compound, 20 ZFL at 4 dpf were incubated with 3 mL of 100 µM compound solution in a six-well plate. Three such treatments were done in parallel, to have three biological replicates. The solutions were prepared in 0.3× Danieau’s with 1% DMSO, which was used as a control incubation with ZFL. The experiment was performed in triplicates. After 24 h incubation, the ZFL were extracted with the same procedure described above and measured on a TF Q Exactive Plus coupled to a Dionex Ultimate 3000 LC system using data dependent acquisition (DDA) mode with an inclusion list of potential metabolites. The conditions used are summarized under Section 2.2. The inclusion lists used are given in the Supplementary Materials.

Three samples were analyzed in triplicates for each compound—ZF extract and surrounding medium each after 24 h of incubation, and a pure compound solution (100 µM) in 0.3× Danieau’s as a control for spontaneous conversion into metabolites upon solvation. Fragment annotation of metabolites was done by comparing their MS^2^ spectra with those of the parent compounds using TF XCalibur software. Mass accuracy tolerance of 5 ppm was generally used for characteristic fragment matching, increasing to 10 ppm for low-mass fragments (<100 m/z).

All metabolites detected can only be described as tentatively identified based on their MS^2^ spectra, and synthesized authentic standards are required to fully validate and confirm the presence of these compounds. Also, absolute quantification was not performed due to the lack of reference standards for novel metabolites. Relative abundances were assessed semi-quantitatively based on peak area integration to identify major metabolic pathways (see Figures S2 and S3).

2.3.4. Experimental Setup and Analysis Method for Untargeted Metabolomics

For each compound, 20 ZFL at 4 dpf were incubated with 3 mL of 100 µM compound solution in a 6-well plate. The solutions were prepared in 0.3× Danieau’s with 1% DMSO, which was used as a control incubation with ZFL. The experiment was performed in triplicates. After 24 h incubation, the ZFL were extracted with the same procedure described above and analyzed on a TF Orbitrap Q Exactive Plus coupled to a Dionex Ultimate 3000 LC system. The elution conditions for reverse phase chromatography were the same as detailed in Section 2.2. HILIC elution was applied using a Merck (Darmstadt, Germany) SeQuant ZIC HILIC (150 mm × 2.1 mm, 3.5 µm). The gradient elution for HILIC was performed using aqueous ammonium acetate (200 mM, eluent C) and acetonitrile containing formic acid (0.1%, v/v, eluent D). The flow rate was set to 500 µL/min using the following gradient: 0–1 min hold 2% C, 1–5 min to 20% C, 5–8.5 min to 60% C, 8.5–10 min hold 60% C, and 10–12 min hold 2% C. The HESI-II source conditions were as follows: ionization mode, positive; sheath gas, 53 AU; auxiliary gas, 14 AU; sweep gas, 3 AU; spray voltage, 3.50 kV; heater temperature, 438 °C; ion transfer capillary temperature, 320 °C; and S-lens RF level, 50.0. Mass spectrometry was performed using full scan (FS) data and subsequent data-dependent acquisition (DDA). The settings for FS data acquisition were as follows: resolution, 140,000 at m/z 200; microscans, 1; automatic gain control (AGC) target, 5 × 10^5^; maximum injection time, 120 ms; and scan range, m/z 50–750. The settings for the DDA mode were as follows: dynamic exclusion, not used; resolution, 35,000 at m/z 200; microscans, 1; loop count, 5; AGC target, 2 × 10^5^; maximum injection time, 100 ms; isolation window, m/z 1.0; high collision dissociation (HCD) with stepped normalized collision energy (NCE), 10, 20, and 40; spectrum data type, profile; and underfill ratio, 1%.

Feature detection was done using TF Compound Discoverer 3.2 (Thermo Fisher Scientific, Waltham, MA, USA) with a predefined untargeted metabolomics workflow—‘Untargeted Metabolomics with Statistics Detect Unknowns with ID Using Online Databases and mzLogic’. Among the features observed, the parent compound and its metabolites were also present, and the boundary of p-value for the selection of significant features was set as the highest p-value observed among the parent compound and two of its most abundant metabolites. This resulted in the p-value boundary conditions of 0.004 for DMT-E and 0.006 for DMT-M. Fold change > 2 was used as the fold change criteria. Principal component analysis (PCA) was used as a further tool to inspect the grouping of significant features. Features with biological group Coefficient of Variance (CV) < 40 in both groups (control and sample) were excluded. Minimum QC coverage of 50% and maximum allowed relative standard deviation of 30% for areas across QC samples were used to exclude features based on QC.

A DDA run was then performed with an inclusion list containing the identified significant features to obtain their MS^2^ spectra. Profile data was converted into centroid data using MSConvert 3.0 [39]. Mzmine 4.0 [40] was used to load HMDB [41] and NIST 23 Mass Spectral Library for the MS^2^ annotation of significant features with the following settings—scans for matching: MS^2^ (all scans); precursor and fragment m/z tolerance: 5 ppm; minimum matched signals: 4; and minimum cosine similarity: 0.7 (without precursor).

3. Results

3.1. In Vitro Metabolism Using pHLS9 Fraction

The proposed metabolic pathways for the two compounds in pHLS9 (and ZFL) are shown in Figure 2. DMT-E metabolic steps included N-demethylation and the hydroxylation of the alkyl chain. A total of three in vitro metabolites were observed for DMT-M and included N-demethylation, as well as alkyl chain hydroxylation and N-oxide formation. Key fragments used to annotate the metabolites are discussed in the later section. The above-described metabolites were detected in incubations containing pHLS9, as well as in low amounts (<10 fold) in neat samples and control samples without pHLS9, suggesting the occurrence of spontaneous conversion into metabolites to an extent upon solvation. Other potential reasons for the occurrence of metabolites in neat/control samples could be due to synthesis byproducts, build-up over storage, or HRMS artifacts. However, since the peak areas increased over time in pHLS9 incubations, their formation was concluded to also be results of enzymatic activity. Neither of the compounds were metabolized to DMT at least in detectable amounts. Nevertheless, certain amounts of DMT (near 1/6th of pure compound for DMT-M, and near 1/500th for DMT-E, based on peak areas) were already present in both stock solutions and areas did not increase after incubations.

3.2. MTC and Bioavailability in ZFL

Among the three tested concentrations, 50 and 100 µM concentrations caused no visible morphological changes such as twisted tail or development of an edema (Table 1). However, at the 500 µM concentration, lethality was observed in both compound treatments. Therefore, 100 µM was fixed as the treatment concentration for all studies.

Results from the bioavailability estimation experiments are shown in Figure 3. Both compounds were detected at high intensities (compared to a 5 µM pure compound solution) in the ZFL extract after treatment at their MTC, suggesting favorable oral bioavailability and ionization behavior. The ratio of intensities between ZFL extract and pure compound solution was seen to be much higher in the case of both compounds as compared to Naloxone, which is a compound known to be bioavailable from previous studies [19].

3.3. In Vivo Metabolism Using ZFL

The proposed metabolic pathways for the two compounds in ZFL (and pHLS9) are shown in Figure 2. The peak areas, retention times, and calculated elemental compositions of the proposed molecules are shown in the Supporting Information (Tables S1 and S2). The MS^2^ spectra for DMT-E metabolites are exemplified in Figure 4. The data for DMT-M metabolites are shown in the Supplementary Materials (Table S2).

In total, four phase I and one phase II metabolite could be found in ZFL. Hydroxylation and demethylation resulted in metabolites with high relative intensities, and glucuronidation after hydroxylation was identified as the only phase II metabolite. As was the case for pHLS9 incubations, some metabolites could be detected in low amounts in neat samples and control samples (Figures S2, S3, S6 and S7), suggesting the occurrence of spontaneous conversion into metabolites to an extent upon solvation. As mentioned in Section 3.1, other potential reasons for the occurrence of some metabolites in neat/control samples could be due to synthesis byproducts, build-up over storage, or HRMS artifacts. However, since the peak areas increased over time in ZFL incubations, their formation was concluded to also be results of enzymatic activity.

In both parent compounds (DMT-E and DMT-M), cleavage of the C-C bond resulted in a fragment corresponding to an iminium cation (m/z = 58.0651) that was seen in high relative intensities in accordance with reported spectra from DMT and many of its derivatives (available on NIST 23 mass spectral database). This cleavage resulted in a stable N cation and a resonance stabilized counter ion, explaining the favorable formation. The lack of this peak at high intensity indicated N-demethylation, and/or hydroxylation that prevents the formation of this fragment ion.

3.4. Untargeted Toxicometabolomics Using ZFL

Nine upregulated features and one downregulated feature were observed upon ZFL DMT-E exposure. The m/z and putative classifications of the compounds are given in the Supplementary Materials, along with principal component analysis (PCA) plots and volcano plots. The upregulated features mostly consist of the parent compound and its metabolites. The only endogenous compound that was well annotated was L-threonine (or one of its closely related isomers), which was observed to be downregulated. The MS^2^ spectrum is shown in Figure 5.

ZFL DMT-M treatment resulted in 15 significant features, of which most were upregulated. Most of the upregulated features were once again identified to be the metabolites of the parent compound and the remaining few could not be identified yet. Threonine was once again identified as a downregulated feature.

4. Discussion

4.1. In Vitro Metabolism Using pHLS9 Fraction

Incubation with pHLS9 fraction resulted in phase I metabolites, with hydroxylation and demethylation being the most favored pathways. Conjugation with glucuronide or sulfate could not be detected for any of the investigated substances, suggesting the lack of phase II metabolism. This contrasts with what was observed in the ZFL system (as discussed later), where a metabolite formed after glucuronidation was observed. The absence of this metabolite, and a relatively lower overall number of metabolites in the pHLS9 system as compared to the ZFL system is in line with the observations from previous studies [16].

4.2. MTC and Bioavailability in ZFL

Given the lack of any lethality or morphological effects, 100 µM was an easy choice for subsequent treatments, although it is likely that there is a concentration between 100 and 500 µM, which is also well tolerated, further treatment concentrations were not tested given the observation of a favorable bioavailability even at 100 µM, as discussed below. Upon DMT-M treatment, ZFL were observed to move slower as compared to untreated larvae, while DMT-E treatment did not induce such an effect, suggesting a difference in the pathways being affected upon uptake.

In the bioavailability study, the ratio of intensities between ZFL extract and pure compound solution was seen to be much higher in the case of both compounds as compared to naloxone, which is a compound known to be bioavailable from previous studies [19]. Although more robust analyses are required to quantitatively compare bioavailability across compounds, the graph nonetheless suggests a very favorable bioavailability for the two compounds, preventing the need to perform a microinjection for treatment. This can be justified by the lack of free charges, and that the compounds contain almost no hydrogen bond donors or acceptors [42]. Moreover, they are small molecules which makes them more likely to penetrate membranes.

4.3. In Vivo Metabolism Using ZFL

All metabolic transformations observed in pHLS9 incubations were also detected in the ZFL extracts, along with additional transformations such as glucuronidation. HPLC chromatograms (depicted in the Supplementary Materials) clearly suggest that there are multiple isomers generated via hydroxylation and glucuronidation pathways, some of which have not been described due to the lack of reliable MS^2^ structure elucidation for those isomers. The total number of metabolites detected is understandably higher in ZFL extracts as compared to pHLS9 incubations, since such a difference due to the model systems was also reported for other psychoactive compounds in the past [16], and highlights the value of ZFL as a model to predict metabolism.

The most abundant metabolite observed was the N-oxide following a hydroxylation at the tertiary amine. Although N-oxide formation was observed in controls, the abundance after ZFL exposure was > 10-fold higher, confirming a relevant enzymatic contribution (shown in detail in the Supplementary Materials). All other metabolites were observed in relatively high amounts in the ZFL extracts, suggesting that they were metabolized by the ZFL. Most of the phase I metabolites detected are expected to have been generated by Cytochrome P450 (CYP) enzymes [43], and ZFL are known to possess orthologs of human CYP enzymes [44]. It was observed in the case of DMT (and other tryptamines such as 5-MeO-DMT) that CYP2D6 was the main CYP enzyme responsible for its metabolism [45,46,47], but studies investigating the enzymes responsible for metabolizing sulfonated tryptamines are lacking. Given that sulfonated tryptamines are known to have deviating enzymatic preferences compared to DMT, which is reflected in their ability to bind to the 5HT_6_ receptor instead of 5HT_2_A [30,31], it is possible that they are metabolized by a different set of CYP enzymes. In vitro inhibition studies and phenotypic screens have been described for tryptamines [45,46], and similar approaches can be applied to investigate the key enzymes involved in metabolizing the compounds in this study.

Conversion of the tertiary amine into an acetic acid moiety was not observed for either of the compounds, which is rather interesting given that indole acetic acid (IAA) is the primary metabolite of DMT [48]. The conversion of DMT into IAA is performed by monoamine oxidase (MAO) enzymes rapidly upon uptake, which hinders the therapeutic potential of DMT [22]. Although studies exploring DMT metabolism in zebrafish are lacking, the existence of a functionally active MAO in zebrafish has been well documented, and the principal characterizing domains are known to be conserved [49]. The lack of conversion of the compounds into IAA suggests that MAO is inactive on the compounds, thereby suggesting the absence of the kind of first pass metabolism known to be an issue in the case of DMT. The inactivity of MAO on the compound would explain the lack of formation of the IAA metabolite, thus leading to the preferential formation of hydroxylated and demethylated metabolites as the main phase I metabolites. However, neither of the compounds went through a conversion into DMT, which hampers the therapeutic potential of the compounds in their use as DMT prodrugs.

The metabolism observed for the DMT-M derivative (see Table S2 in the Supplementary Materials) closely resembles what was also observed for DMT-E, which suggests that the presence of an additional methyl group does not affect the binding of the molecule to its metabolizing enzymes. Given the structural similarity, it is understandable that both derivatives undergo similar biotransformation. Although difficult to predict, it is also possible that some of the metabolites retain a therapeutic potential, either by 5-HT_6_ binding like the parent compound, or by other unexplored mechanisms. The main binding moiety for the 5-HT_6_ receptor, in the case of sulfonyl tryptamines, is a protonated amine [50], which was not significantly altered in most of the biotransformations observed. Further studies are necessary to investigate if the observed metabolites retain binding ability and therapeutic potential.

4.4. Untargeted Toxicometabolomics Using ZFL

Untargeted toxicometabolomics studies allow the detection and identification of endogenous and exogenous biomarkers after drug exposure. This enables the determination of metabolites (as discussed in the previous section) of the investigated substance in order to detect an intake by patients, as well as the identification of biomarkers that provide information on the (off target) effects in only one experiment [51,52].

Threonine is an essential amino acid that is known to play a key role in diverse physiological processes including protein synthesis, lipid metabolism, embryonic stem cell differentiation, and intestinal gut health regulation [53]. The downregulation of threonine could occur because of increased synthesis of threonine-rich proteins, or because of reduced biosynthesis, or due to a more complicated mechanism that requires further investigation.

Threonine was identified as a downregulated feature in both compound treatments, suggesting that the two derivatives have a similar mode of action. Interestingly, ayahuasca—the natural DMT-containing substance through which DMT is widely consumed, is known to alter the levels of large neutral amino acids, but not threonine [54].

5. Conclusions

This study offers valuable insights into the biotransformation and metabolomic response following treatment with two novel sulfonated derivatives of DMT, utilizing pHLS9 and ZFL as model systems. For each compound, five phase I and one phase II metabolite was identified in ZFL, while fewer phase I metabolites were detected after pHLS9 incubation. The two main metabolic reactions were identified to be hydroxylation and demethylation. Glucuronidation upon hydroxylation on the ring was the only phase II reaction observed. Notably, the generation of an acetic acid moiety—a major biotransformation pathway for DMT and a key limitation to its therapeutic use—was absent. The loss of the sulfonation to yield DMT was also not observed neither in pHLS9 nor on ZFL, suggesting that sulfonation does not act as a prodrug mechanism for the compounds in these model systems. Further investigation using advanced metabolism models, including human microdosing, is warranted to elucidate this mechanism. However, these spectral data provide the necessary foundation for forensic laboratories and doping laboratories to identify these NPS in biological samples.

Threonine was the only identified endogenous metabolite to be significantly altered upon treatment with both compounds. Some significant features were identified as metabolites of the parent compound supporting prior observations from the targeted approach, but several remain unidentified due to the limited availability of a dedicated ZFL metabolite library. Identification of endogenous ZFL metabolites is a challenge given the lack of an existing library specifically containing ZFL metabolites. Moreover, the lack of more endogenous biomarkers suggests a possible scope in further optimization of sample preparation steps such as ZFL pool size, extraction and injection volumes, etc. Nevertheless, the metabolites and biomarkers identified in this study can be used in a targeted approach to analyze intake of these synthetic tryptamine derivatives. Overall, this study advances the understanding of synthetic tryptamine metabolism and their effects on the metabolome. Nonetheless, further research is essential to validate these findings and assess their relevance to human physiology.