Analytical profile, in vitro metabolism and behavioral properties of the lysergamide 1P-AL-LAD.

Lysergic acid diethylamide (LSD) is known to induce powerful psychoactive effects in humans, which cemented its status as an important tool for clinical research. A range of analogues and derivatives has been investigated over the years, including those classified as new psychoactive substances. This study presents the characterization of the novel lysergamide N,N-diethyl-1-propanoyl-6-(prop-2-en-1-yl)-9,10-didehydroergoline-8β-carboxamide (1P-AL-LAD) using various mass spectrometric, gas- and liquid chromatographic and spectroscopic methods. In vitro metabolism studies using pooled human liver microsomes (pHLM) confirmed that 1P-AL-LAD converted to AL-LAD as the most abundant metabolite consistent with the hypothesis that 1P-AL-LAD may act as a prodrug. Fourteen metabolites were detected in total; metabolic reactions included hydroxylation of the core lysergamide ring structure or the N6 -allyl group, formation of dihydrodiol metabolites, N-dealkylation, N1 -deacylation, dehydrogenation, and combinations thereof. The in vivo behavioral activity of 1P-AL-LAD was evaluated using the mouse head twitch response (HTR), a 5-HT2A -mediated head movement that serves as a behavioral proxy in rodents for human hallucinogenic effects. 1P-AL-LAD induced a dose-dependent increase in HTR counts with an inverted U-shaped dose-response function, similar to lysergic acid diethylamide (LSD), psilocybin, and other psychedelics. Following intraperitoneal injection, the median effective dose (ED50 ) for 1P-AL-LAD was 491 nmol/kg, making it almost three times less potent than AL-LAD (174.9 nmol/kg). Previous studies have shown that N1 -substitution disrupts the ability of lysergamides to activate the 5-HT2A receptor; based on the in vitro metabolism data, 1P-AL-LAD may induce the HTR because it acts as a prodrug and is metabolized to AL-LAD after administration to mice.

INTRODUCTION

The classical psychedelic lysergic acid diethylamide (LSD, Figure 1) has long been known for its ability to induce complex, non‐ordinary states of consciousness in humans. , , Over the last decade, psychedelic drugs such as LSD and psilocybin have been explored as potential treatments for a variety of psychiatric disorders, including anxiety, treatment‐resistant depression, anorexia, drug and alcohol dependence, cluster headaches, and obsessive–compulsive disorder. , , , , At the same time, recreational use of LSD and other psychedelic drugs continues to occur though the prevalence of use varies widely across countries and age groups. ,

FIGURE 1

Chemical structures of lysergic acid diethylamide (LSD) and N 6‐modified analogs and their N 1‐propanoyl derivatives

Extensive modifications of the LSD structural scaffold have been investigated. Studies characterizing the structure–activity relationships for LSD have frequently focused on the amide substituents although some exceptions exist. The majority of changes made to the N,N‐diethyl group were found to reduce potency. , Another structural modification that has been explored is the replacement of the LSD N 6‐methyl group with other substituents. For example, Niwaguchi and co‐workers reported the synthesis of N 6‐ethyl‐nor‐LSD (ETH‐LAD), N 6‐n‐propyl‐nor‐LSD (PRO‐LAD), and N 6‐allyl‐nor‐LSD (AL‐LAD) (Figure 1 for ETH‐LAD and AL‐LAD) in 1976. All three lysergamides showed greater oxytocic activity than LSD in rat uterine tissues. Likewise, AL‐LAD was also shown to induce a stronger hyperthermic response than LSD in rabbits. In 1985, Hoffman and Nichols published on the behavioral evaluation of ETH‐LAD, PRO‐LAD, AL‐LAD, as well as three other N 6‐substituted lysergamides and tested them in a drug discrimination assay in which rats were trained to discriminate 0.08 mg/kg LSD from saline. The LSD derivatives containing N 6‐ethyl (ETH‐LAD), N 6‐n‐propyl (PRO‐LAD), N 6‐allyl (AL‐LAD), N 6‐isopropyl (IPR‐LAD), and N 6‐n‐butyl (BU‐LAD) substituents produced full substitution in LSD‐trained rats. Notably, while PRO‐LAD had about the same potency as LSD, ETH‐LAD and AL‐LAD were more potent than LSD in the drug discrimination paradigm.

Anecdotal reports suggest that oral administrations of ETH‐LAD, PRO‐LAD, and AL‐LAD produce psychedelic‐like effects in humans. Consistent with the rat drug discrimination data, ETH‐LAD was slightly more potent than LSD in humans whereas AL‐LAD was approximately equipotent. , In recent years, both AL‐LAD and ETH‐LAD became available for purchase from online retailers, often distributed on small pieces of absorbent paper (“blotters”). , Reports describing their detection were first notified to the European Monitoring Centre for Drugs and Drug Addiction (EMCDDA) in 2015 and 2016, respectively. , Since then, other publications have appeared describing the detection of AL‐LAD and ETH‐LAD by researchers located outside of Europe. ,

Investigations carried out in vitro where ETH‐LAD and AL‐LAD were incubated with pooled human liver S9 microsomal fraction detected five ETH‐LAD metabolites and 11 AL‐LAD metabolites. Furthermore, studies using selective inhibitors of cytochrome P450 (CYP) enzymes showed that CYP3A4 was involved in N 6‐dealkylation and N‐deethylation reactions (ETH‐LAD and AL‐LAD), whereas CYP2D6 and CYP3A4 (ETH‐LAD) and CYP1A2 and CYP3A4 (AL‐LAD) were responsible for various hydroxylations; phase 2 glucuronide metabolites were also detected.

In addition to the N,N‐diethylamide group of LSD and the N 6‐position, various substitutions have been made to the indole nitrogen (position N 1). Among the earliest examples were 1‐methyl‐LSD (MLD‐41) and 1‐acetyl‐LSD (ALD‐52). More recently, a number of novel 1‐acyl‐substituted lysergamides have emerged on the market (Figure 1). 1‐Propanoyl‐LSD (1P‐LSD) first appeared in 2016, followed by 1‐butanoyl‐LSD (1B‐LSD), 1‐valeroyl‐LSD (1V‐LSD), and 1‐cyclopropanoyl‐LSD (1cP‐LSD). , , , , One complexity with these molecules is that substitution on the indole nitrogen reduces the affinity of ergolines for the 5‐HT2A receptor, , the site responsible for mediating the characteristic effects of psychedelic drugs. However, the 1‐acyl group appears to be rapidly hydrolyzed in vivo, so the activity of these lysergamides is likely dependent on their metabolism to LSD.

In addition to LSD, the 1‐acyl derivatives of other lysergamides have also been synthesized. 1‐Propanoyl‐ETH‐LAD (1P‐ETH‐LAD), the first known example, appeared online in 2016 and its detection was first notified to the EMCDDA in 2017.

The present investigation continues the research on novel lysergamides and reports on the analytical properties of N,N‐diethyl‐1‐propanoyl‐6‐(prop‐2‐en‐1‐yl)‐9,10‐didehydroergoline‐8β‐carboxamide (1‐propanoyl‐AL‐LAD; 1P‐AL‐LAD), the N 1‐propanoyl derivative of the psychedelic drug 6‐allyl‐norlysergic acid diethylamide (AL‐LAD) (Figure 1). There is no indication that this compound is currently available for sale by Internet retailers but it was considered prudent that its analytical profile should be made available to stakeholders involved in psychoactive drug research. Because 1P‐AL‐LAD is expected to act as a prodrug for AL‐LAD, experiments were conducted to determine whether 1P‐AL‐LAD is N 1‐deacylated in vitro after incubation with pooled human liver microsomes (pHLM). Finally, 1P‐AL‐LAD was evaluated in the mouse head‐twitch response (HTR) assay to test whether it has an LSD‐like behavioral profile. , ,

EXPERIMENTAL

Materials

Formic acid (Rotipuran® ≥ 98%, p.a.) and potassium hydrogen phosphate (≥ 99%, p.a.) were obtained from Carl Roth (Karlsruhe, Germany); acetonitrile (ACN) (LC–MS grade), aqueous ammonium formate solution (10 M, ≥99.995% trace metals basis), and potassium hydroxide [puriss. p.a. ≥ 86% (T) pellets] from Sigma‐Aldrich (Steinheim, Germany); pHLM (150 donors, 20 mg/ml protein in 250 mM sucrose), nicotinamide adenine dinucleotide phosphate (NADPH) regenerating solutions A and B (reductase activity 0.43 μmol/min/ml), and potassium phosphate buffer 0.5 M (pH 7.5) from Corning (Corning, NY, USA). Other chemicals and solvents were of analytical and HPLC grade and obtained from Aldrich (Dorset, UK). 1P‐AL‐LAD hemitartrate (2:1) (>95%) powder was supplied by Synex Synthetics BV, Maastricht, The Netherlands.

Instrumentation

Gas chromatography‐electron ionization‐mass spectrometry (GC‐EI‐MS)

Approximately 1 mg of the sample was dissolved in 10 ml of methanol and diluted 1:10. A Finnigan TSQ 8000 triple stage quadrupole mass spectrometer coupled to a Trace GC Ultra gas chromatograph (Thermo Fisher, Waltham, USA) and equipped with a fused silica DB‐1 column (30 m x 0.32 mm i.d., 0.25 μm film thickness) (Agilent Technologies, Santa Clara, USA) was used for GC‐EI‐MS analysis. Sample solutions were introduced by a CTC CombiPAL (CTC Analytics, Zwingen, Switzerland) autosampler.

The following GC parameters were employed: injection volume: 0.5 μl, splitless; injector temperature: 280°C; carrier gas: helium; flow rate: 1.2 ml/min. Initially, the oven temperature was kept at 80°C for 2 min, ramped to 310°C at 20°C/min and subsequently maintained at the final temperature of 310°C for 23 min. MS parameters were set as follows: ionization mode: EI at 70 eV; emission current: 50 μA; ion source temperature: 220°C; MS transfer line temperature: 300°C; scan time: 1 s; scan range: m/z 29–600.

Data analysis was conducted using Xcalibur 4.0 Qual Browser (Thermo Fisher) and the National Institute of Standards and Technology (NIST) MS search program (version 2.3) (NIST, MD, US). EI mass spectra were compared with EI mass spectral libraries provided by the NIST (release 2020), the European Network of Forensic Science Institutes (ENFSI, release 2018), Scientific Working Group for the Analysis of Seized Drugs (SWGDRUG, release 2021), as well as the designer drug library 2021 (DigiLab, SH, DE) and libraries built in‐house. Retention indices (RI) were calculated from the measurement of retention times obtained from the constituents of an n‐alkane mixture. The temperature program is specified above. For calculation, logarithmic interpolation was applied between two consecutive n‐alkanes.

Gas chromatography‐solid‐phase‐infrared analysis (GC‐sIR)

A GC‐solid phase‐IR‐system consisting of an Agilent GC 7890B (Agilent Technologies) equipped with a fused silica capillary DB‐1 column (30 m x 0.32 mm internal diameter, 0.25 μm film thickness), an Agilent G4567A probe sampler (Agilent Technologies) and a DiscovIR‐GC (Spectra Analysis, Marlborough, MA, US) were used for the acquisition of solid transmission IR spectra. The eluting substances were cryogenically accumulated on a spirally rotating ZnSe disk cooled by liquid nitrogen to −40°C. IR spectra were recorded through the IR‐transparent ZnSe disk using a nitrogen‐cooled mercury cadmium telluride (MCT) detector.

The GC parameters were as follows: injection volume 1 μl; splitless mode; injection port temperature 240°C; carrier gas: helium; flow rate 2.5 ml/min. Chromatographic conditions were as follows: oven temperature program: 80°C for 2 min, ramped to 310°C at 20°C/min, and maintained for 20 min; transfer line: 280°C. Infrared conditions: oven temperature 300°C; restrictor temperature 300°C; disc temperature −40°C; dewar cap temperature 35°C; vacuum 0.2 mTorr; disc speed 3 mm/min; spiral separation 1 mm; wavelength resolution 4 cm−1; IR range 650–4000 cm−1; acquisition time: 0.6 s/file and 64 scans per spectrum. Data were processed using GRAMS/AI Ver. 9.1 (Grams Spectroscopy Software Suite, Thermo Fisher) followed by OMNIC Software, Ver. 7.4.127 (Thermo Fisher).

High performance liquid chromatography electrospray ionization quadrupole time‐of‐flight mass spectrometry (HPLC‐ESI‐QTOF‐MS) experiments

HPLC‐ESI‐QTOF‐MS analysis was performed on an impact II™ QTOF instrument coupled with an Elute HPLC system (both from Bruker Daltonik, Bremen, Germany). Chromatographic separation was achieved on a Kinetex® Biphenyl column (100 × 2.1 mm, 2.6 μm particle size, Phenomenex, Aschaffenburg, Germany) equipped with a corresponding guard column (SecurityGuard™ ULTRA Cartridges UHPLC Biphenyl for columns with an internal diameter of 2.1 mm, Phenomenex, Aschaffenburg, Germany). HPLC mobile phase A consisted of deionized water (979 ml), formic acid (1 ml), ammonia formate (200 mM, 10 ml, freshly prepared from aqueous 10 M solution) and ACN (10 ml). Mobile phase B consisted of acetonitrile (989 ml), formic acid (1 ml) and ammonia formate (200 mM, 10 ml, freshly prepared from aqueous 10 M solution). Mobile phase A and B were varied in a linear program (Tmin/A:B; T0/90:10; T10/20:80; T10.5–12.5/5:95; T12.7–14/90:10) with LC flow set at 0.3 ml/min and column oven temperature at 40°C. The autosampler was cooled to 5°C. The injection volume was 10 μl. HyStar™ version 3.2 and DataAnalysis version 4.2 (both from Bruker Daltonik) were used for data acquisition and processing, respectively. The QTOF‐MS was operated in positive electrospray ionization mode acquiring spectra in the range of m/z 50–500 (acquisition rate of 4.0 Hz). Acquisition was performed in full scan/broadband collision induced dissociation (bbCID) mode (data independent) and in a second run in full scan/AutoMS/MS mode (data dependent) to obtain cleaner fragment spectra. The collision energy applied for bbCID and Auto‐MS/MS was 30 ± 6 eV. The dry gas temperature was set to 200°C with a dry gas flow of 8.0 L/min. The nebulizer gas pressure was 200 kPa. Nitrogen was used as collision gas. The voltages for the capillary and end plate offset were 2500 and 500 V, respectively. External and internal mass calibrations were performed using sodium formate/acetate clusters and high precision calibration (HPC) mode. Metabolites generated in the pHLM assay were tentatively identified and characterized in manual data processing with the following criteria: mass error of the precursor ion < 5 ppm, signal‐to‐noise ratio > 3:1, and mass tolerance for fragment ions ± 10 ppm.

Pooled human liver microsome (pHLM) assay

Total assay volume of 100 μl consisted of 5 μl pHLM solution, 1 μl 1P‐AL‐LAD stock solution (1 mg/ml in ACN), 5 μl NADPH regenerating solution A, 1 μl NADPH regenerating solution B, 20 μl phosphate buffer, and 58 μl deionized water. Incubation was conducted for 30 min at 37°C. The incubation was quenched by adding 300 μl of ice‐cold acetonitrile. Ammonium formate (50 μl, 10 M) was added for improved phase separation and after centrifugation the organic layer was transferred into a separate vial and stored at −20°C. Incubations were performed in triplicate. Additionally, two negative control samples were processed in the same way. One was performed with 5 μl phosphate buffer instead of pHLM to identify compounds not formed by metabolism, and a second with 1 μl ACN instead of the substrate to confirm the absence of interfering compounds. For LC‐ESI‐QTOF‐MS analysis, 100 μl supernatant was evaporated dryness under a gentle stream of nitrogen and reconstituted in 50 μl mobile phase A/B (70:30, v/v).

Animal pharmacology

Male C57BL/6J mice (6–8 weeks old) from Jackson Labs (Bar Harbor, ME, USA) were used for the behavioral experiments. The mice were housed on a reversed light–dark cycle (lights on at 1900 h, off at 0700 h,) in a vivarium approved by the Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC) at the University of California San Diego. Mice were housed up to four per cage in a climate‐controlled room and with food and water provided ad libitum except during behavioral testing. Testing was performed between 1000 and 1800 h (during the dark phase of the light–dark cycle). The studies were conducted in accordance with National Institutes Health (NIH) guidelines and were approved by the University of California San Diego Institutional Animal Care and Use Committee. 1P‐AL‐LAD hemitartrate (2:1) was dissolved in isotonic saline for intraperitoneal (IP) injection (5 ml/kg).

The head‐twitch response (HTR) was assessed using a head‐mounted neodymium magnet and a magnetometer detection coil. The magnet was attached as described previously and then the mice were allowed to recover from the surgeries for at least 1 week prior to behavioral testing. HTR experiments were conducted in a well‐lit room, and the mice were allowed to habituate to the room for at least 1 h prior to testing. Mice (n = 5–6/group, 32 total) were injected IP with vehicle or 1P‐AL‐LAD and then placed in a 12‐cm diameter glass cylinder surrounded by a magnetometer coil; activity was recorded continuously for 30 minutes. Coil voltage was low‐pass filtered (1 kHz), amplified, and digitized (20‐kHz sampling rate) using a Powerlab/8SP with LabChart v 7.3.2 (ADInstruments, Colorado Springs, CO, USA). Head twitches were identified in the recordings using artificial intelligence. HTR counts were analyzed using a one‐way Welch ANOVA. Dunnett's T3 multiple comparisons test was used for post hoc comparisons. Significance was demonstrated by surpassing an alpha level of 0.05. Median effective doses (ED50 values) and 95% confidence intervals for dose–response experiments were calculated by nonlinear regression (Prism 9.01, GraphPad Software, San Diego, CA, USA).

RESULTS AND DISCUSSION

Analytical features

The electron ionization (EI) mass spectrum of 1P‐AL‐LAD is shown in Figure 2a with suggested fragmentation pathways included in the supporting information. In comparison with AL‐LAD reported previously, and other LSD‐related compounds, a group of key ions were detectable in the form of fragment clusters including m/z 151–156, m/z 161–169, m/z 178–182, m/z 191–197, and m/z 205–208. Similar to other lysergamides with a N,N‐diethylamide group, iminium ions were detected at m/z 72 together with m/z 100 and m/z 128. , , , , , Compared with AL‐LAD, the presence of the propanoyl group shifted some of the ions by 56 Da including the molecular ion at m/z 405 (AL‐LAD: m/z 349), the retro‐Diels‐Alder fragment following the loss of N‐allylmethanimine at m/z 336 (AL‐LAD: m/z 280), or the species reflecting the loss of the N 6‐allyl group at m/z 364 (AL‐LAD: m/z 308). A characteristic oxonium ion reflecting the presence of the N 1‐propanoyl substituent was also be observed at m/z 57, which was comparable to other N 1‐propanoyl lysergamides such as 1P‐LSD and 1P‐ETH‐LAD. An extracted ion chromatogram (EIC; m/z 405) obtained from the GC–MS analysis of a methanolic solution of 1P‐AL‐LAD is shown in the supporting information to consider the detection of 1P‐AL‐LAD isomers including iso‐1P‐AL‐LAD. Three minor isomers were detected under the conditions used with 1P‐AL‐LAD being labeled as isomer III. The EI mass spectral data for isomers I, II, and IV are shown in the supporting information. The epimeric iso‐1P‐AL‐LAD was expected to show a comparable electron ionization mass spectrum but it appeared that the recorded spectra showed distinct differences (for example the presence of m/z 333 at high abundance especially for isomers I and II), which led to the tentative suggestion that the detected isomers might not have reflected the detection of iso‐1P‐AL‐LAD but rather GC‐induced by‐products (not detectable under LC–MS conditions, see below). However, these suggested structures were based on mass spectral grounds alone and in the absence of suitable reference material these proposals must remain speculative. As shown below, however, analysis of 1P‐AL‐LAD by LC–MS suggested the detection of iso‐1P‐AL‐LAD at low abundance.

FIGURE 2

(a) Electron ionization mass spectrum of 1P‐AL‐LAD (isomer III). (b) Electrospray ionization QTOF tandem mass spectrum of 1P‐AL‐LAD

A full scan GC–MS trace of this analysis has also been provided as supporting information and further inspections revealed another peak eluting at 23.08 min, labeled as 1P‐AL‐LAD‐A (–C3H8). The molecular ion was detected at m/z 361 and together with the information shown in the mass spectrum, it was thought that this compound may have represented the N 6‐dealkylated, fully aromatic (D‐ring) derivative (supporting information). Similar to the minor isomers, it was hypothesized that the formation of 1P‐AL‐LAD‐A (‐C3H8) was also GC‐induced. Finally, the TIC trace also showed a peak at 18.00 min, which was identified as AL‐LAD consistent with the information obtained from LC–MS analysis described below. Interestingly, LSD formation during GC–MS analysis of 1P‐LSD has been described by Tanaka and colleagues, which suggests that the potential for such artificial formations might need to be considered in cases where the N 1‐acylated form is not a controlled substance. In the laboratories of the authors, however, this artificial formation has not yet been observed. Basification followed by extraction into diethyl ether led to hydrolysis to AL‐LAD when subjected to GC–MS analysis. The three minor isomers and 1P‐AL‐LAD‐A (–C3H8) were also detected (full scan, supporting information).

The ESI‐QTOF tandem mass spectrum recorded for 1P‐AL‐LAD after incubation with pHLM is shown in Figure 2b (proposed fragmentation pathways in the supporting information). The base peak was detected at m/z 365.2097 consistent with a homolytic loss of the N 6‐allyl group and formation of a radical cation (C22H27N3O2 •+), similar to a previous report with AL‐LAD where this loss also occurred to a significant extent. For ETH‐LAD and 1P‐ETH‐LAD, the corresponding loss of the N 6‐ethyl group was observed as well although the resulting ions did not reach base peak abundance. A neutral loss of N,N‐diethylformamide from the m/z 365.2097 ion might have led to a further dissociation to C17H18N2O•+ (m/z 264.1256) followed by detection of C14H12N2 •+ (m/z 208.0994), reflecting the subsequent cleavage of the N 1‐acyl group. The species at m/z 337.1910 (C21H25N2O2 +) might have reflected the involvement of a retro‐Diels‐Alder mechanism (neutral loss of N‐allylmethanimine) and was also thought to be the same ion formed in the spectra of 1P‐LSD and 1P‐ETH‐LAD. The mass spectral resolution was also sufficient to detect another m/z 208 species (m/z 208.0760, C14H10NO+) possibly representing an oxonium species following an alternative pathway (supporting information), which then might have resulted in the formation of m/z 180.0806 (C13H10NO+). Two m/z 249 ions were also observed with minor abundance (Figure 2b). One of them was detected at m/z 249.1384 (C17H17N2 +) that might have resulted from 1P‐AL‐LAD following a consecutive loss of the propanoyl group and N,N‐diethylformamide. The other ion was observed at m/z 249.1024 (C17H17N2 +) that could have arisen from the m/z 264.1256 radical cation after losing a methyl radical. Fragment ions that might have formed from the carboxamide moiety included m/z 128.1071 (C7H14NO+), 100.0759 (C9H10NO+), and 74.0966 (C4H12N+). The presence of the N 6‐allyl substituent might have been reflected by the ion at m/z 70.0654 (protonated N‐allylmethanimine, C4H8N+). Additional QTOF tandem mass spectra recorded from 1P‐AL‐LAD on a different instrument (Agilent 6530B) can also be found in the supporting information. In addition, multistage MSn data recorded from a direct infusion using a linear ion trap (LIT) mass spectrometer have also been added in the supporting information.

The GC–MS results described above suggested that iso‐1P‐AL‐LAD remained undetectable based on mass spectral considerations. The implementation of LC‐ESI‐QTOF‐MS analysis however indicated that the epimer was detectable both in a 1P‐AL‐LAD solution and in a negative pHLM mixture. Based on a comparison of signal responses involving peak heights, the percentage value for the iso‐form was estimated to be around 1.12% (supporting information). As described above, the GC–MS analysis also revealed the detection of AL‐LAD (supporting information), which raised the question as to whether it was potentially GC‐induced or present as an impurity, or both. LC‐ESI‐QTOF‐MS analysis confirmed that AL‐LAD could be detected as an impurity in the 1P‐AL‐LAD sample estimated to represent around 0.64%, which was considered low. This estimation was based on a comparison with the signal response recorded for AL‐LAD at a 1 μg/ml concentration (supporting information). For the purpose of the in vitro metabolism study described below, analysis of the incubation mixture without addition of pHLM also confirmed the detection of iso‐1P‐AL‐LAD and AL‐LAD. In the latter case, however, it could not be established whether some proportion of the detected peak was also formed after non‐enzymatic hydrolysis (supporting information). An alternative LC‐ESI‐linear ion trap‐MS/MS method was also employed and it could be confirmed that AL‐LAD was detectable in the 1P‐AL‐LAD sample as well (supporting information).

When analyzing solutions of 1P‐AL‐LAD and negative pHLM, the EIC at m/z 422.2434 (C25H32N3O3 +) revealed the detection of additional peaks (supporting information). Though the signal intensity seemed low (~0.28%), the QTOF‐MS/MS data of the peak at 6.9 min suggested the presence of a hydroxylated 1P‐AL‐LAD species and further inspection of the mass spectral data suggested that the hydroxylation might have occurred on the 7‐position (Figure 3). The majority of product ions appeared to be comparable to those detected with 1P‐AL‐LAD (Figure 2) but some noticeable differences were also observed. The detection of the retro‐Diels Alder fragment (m/z 337.1909 (similar to 1P‐AL‐LAD, Figure 2b and supporting information) initially suggested that the hydroxylation must have occurred on the part of the molecule that would have been lost during the formation of m/z 337.1909 (C21H25N2O2 +). An important key indicator for the tentative identification of 7‐HO‐1P‐AL‐LAD arose from the detection of m/z 381.2047 (C22H27N3O3 •+) representing the loss of the N 6‐allyl radical, which in turn suggested that the hydroxylation could not have occurred on the allyl group. As shown in Figure 2b, the radical loss of the N 6‐allyl group attached to 1P‐AL‐LAD was seen at m/z 365.2097 (C22H27N3O2 •+) whereas in the hydroxylated impurity, this was not detected. Instead, a mass shift consistent with an additional hydroxyl group gave rise to m/z 381.2047 albeit at low abundance. The presence of the hydroxyl group that supported the provisional identification of 7‐HO‐1P‐AL‐LAD was also consistent with m/z 86.0599 (C4H8NO+) where the addition of the hydroxyl group led to a corresponding mass shift from m/z 70.0654 (1P‐AL‐LAD, C4H8N+) (Figure 2b). Overall, this indicated that 7‐HO‐1P‐AL‐LAD might have been a synthesis by‐product or formed during storage. The same impurity was also observed in the negative pHLM solution but it was not detected as a 1P‐AL‐LAD metabolite in the positive pHLM incubation mixture. As described below, its metabolite however was detected (7‐HO‐AL‐LAD) following the hydrolysis of the propanoyl group.

FIGURE 3

Tentative identification of the 7‐HO‐1P‐AL‐LAD impurity [Colour figure can be viewed at wileyonlinelibrary.com]

The infrared spectrum recorded during the implementation of GC‐sIR can be found as supporting information. Similar to other N 1‐propanoyl lysergamides (1P‐LSD and 1P‐ETH‐LAD) investigated previously, , indole NH bands were absent and two carbonyl stretches were detectable at 1704 and 1639 cm−1. In 1P‐LSD, these were recorded at 1703 and 1638 cm−1 whereas in 1P‐ETH‐LAD both carbonyl stretches were observed at 1704 and 1640 cm−1. In the GC‐sIR spectrum recorded from the N 1‐unsubstituted AL‐LAD, the NH band was visible at around 3280 cm−1 together with only one carbonyl stretch at 1626 cm−1. A spectrum obtained from the 1P‐AL‐LAD hemitartrate (2:1) salt using attenuated total reflectance Fourier transform infrared spectroscopy (ATR‐FT‐IR) has also been added in the supporting information.

Nuclear magnetic resonance (NMR) spectroscopy information is shown in Table 1 with full 1D/2D NMR spectra provided in the supporting information. Assignments were aided by 2D experiments (1H/1H correlation spectroscopy, COSY; 1H/13C heteronuclear single quantum coherence spectroscopy, HSQC; 1H/13C heteronuclear multiple bond correlation spectroscopy, HMBC) and essentially comparable to the spectra recorded for AL‐LAD during previous investigations with the main difference being the presence of the additional N 1‐propanoyl group in 1P‐AL‐LAD. For visual comparisons, AL‐LAD proton and carbon NMR spectra together with 1P‐AL‐LAD spectra can be found in the supporting information as well. The estimated 1.12% iso‐1P‐AL‐LAD impurity found in 1P‐AL‐LAD by LC‐ESI‐QTOF‐MS was not observed in the proton NMR spectrum under the conditions used.

TABLE 1

1H and 13C NMR data for 1P‐AL‐LAD hemitartrate in d6‐DMSO at 600/150 MHz

chemical structure image
No.	13C [δ/ppm]	1H [δ/ppm]
2	119.90	7.56 (d, J = 1.9 Hz, 1 H)
3	116.14	–
4	26.13	2.48–2.41 (m, H‐4α, 1 H) 3.52 (dd, J = 15.2, 5.3 Hz, H‐4β, 1 H)
5	58.82	3.35–3.32 (m, H‐5β, 1 H) * partially overlapping with H‐21 (2 H)
6		–
7	51.51	3.07 (dd, J = 11.3, 4.6 Hz, H‐7α, 1 H) 2.58 (t, J = 10.7 Hz, H‐7β, 1 H)
8	39.08	3.77–3.71 (m, 8α, 1 H)
9	122.31	6.34 (s, 1 H)
10	134.08	–
11	128.17	–
12	116.52	7.34–7.28 (m, 2 H) * partially overlapping with H‐13
13	125.89	7.34–7.28 (m, 2 H) * partially overlapping with H‐12
14	114.75	8.00 (d, J = 7.6 Hz, 1 H)
15	133.11	–
16	127.58	–
17	56.25	3.63 (dd, J = 14.6, 4.9 Hz, 1 H) 3.16 (dd, J = 14.6, 8.0 Hz, 1 H)
18	134.69	5.97 (dddd, J = 17.1, 10.2, 7.9, 4.9 Hz, 1 H)
19	118.03	5.30 (d, J = 17.1 Hz, 1 H); H‐19 trans to H‐18
19	118.03	5.20 (d, J = 10.1 Hz, 1 H); H‐19 cis to H‐18
20	170.52	–
21	41.56	3.42 (AB qq, J = 13.9, 7.1 Hz, 2 H).
21	39.72	3.32–3.27 (m, 2 H) *peaks are partially overlapping with H‐5β
22	14.82	1.20–1.14 (m, 6 H) *peaks are overlapping with H‐25 (3 H)
22	13.07	1.05 (t, J = 7.1 Hz, 3 H)
23	172.49	–
24	28.17	3.00 (AB qq, J = 14.0, 7.2 Hz, 2 H)
25	8.57	1.20–1.14 (m, 6 H) *peaks are overlapping with H‐22 (3 H)
TA	173.16	–
TA	72.09	4.29 (s, ~1.3 H)

Abbreviation: TA, tartaric acid.

Microsomal phase I metabolism of 1P‐AL‐LAD

In the pHLM assay, 14 phase I metabolites were detected and characterized. The metabolic reactions in vitro included hydroxylation of the core lysergamide ring or the N 6‐allyl group, formation of dihydrodiol metabolites, N‐dealkylation, N 1‐deacylation, dehydrogenation, and combinations thereof. An overview of the QTOF‐MS data of all detected and proposed in vitro metabolites of 1P‐AL‐LAD, their most abundant product ions and exact masses are presented in Table 2. Extracted ion chromatograms (EIC) using the protonated molecules are shown in Figure 4. An overview of the proposed metabolic pathways for 1P‐AL‐LAD is shown in Figure 5.

TABLE 2

1P‐AL‐LAD and its phase I metabolites along with their identification numbers (ID), retention times (RT), peak heights of the signal of the protonated molecule in MS1 (PH), accurate and exact mass to charge ratios, the respective error and the elemental compositions of the protonated molecule [M + H]+ and its three most abundant fragment ions (FI A‐C) with their relative ion intensities (rel. Ion. Int.)

ID	Compound	RT [min]	Ion	PH [cps]	Rel. Ion Int.	Accurate mass [m/z]	Elemental composition	Exact mass [m/z]	Error [ppm]	Rank
P	1P‐AL‐LAD	6.1	[M + H]+	9.34E+05	‐	406.2488	C25H32N3O2	406.2489	−0.25	‐
			FI A	‐	100%	365.2097	C22H27N3O2	365.2098	−0.27
			FI B	‐	51%	208.0994	C14H12N2	208.0995	−0.48
			FI C	‐	50%	264.1256	C17H16N2O	264.1257	−0.38
M1	1‐Depropanoyl‐di‐hydrodiol‐P1	3.1	[M + H]+	1.43E+06	‐	384.2282	C22H30N3O3	384.2282	0.0	3
			FI A	‐	100%	325.1783	C19H23N3O2	325.1785	−0.62
			FI B	‐	53%	224.0939	C14H12N2O	224.0944	−2.23
			FI C	‐	34%	223.0863	C14H11N2O	223.0866	−1.34
M2	1‐Depropanoyl‐N‐deallyl‐N‐deethyl‐P	3.2	[M + H]+	9.54E+04	‐	282.1601	C17H20N3O	282.1601	0.0	13
			FI A	‐	100%	209.1072	C14H13N2	209.1073	−0.48
			FI B	‐	25%	237.1016	C15H13N2O	237.1022	−2.53
			FI C	‐	17%	183.0921	C12H11N2	183.0917	2.18
M3	1‐Depropanoyl‐di‐hydrodiol‐P2	3.2	[M + H]+	1.70E+05	‐	384.2822	C22H30N3O3	384.2822	0.0	10
			FI A	‐	100%	224.0707	C14H10NO2	224.0706	0.47
			FI B	‐	83%	343.1891	C19H25N3O3	343.1891	0.0
			FI C	‐	57%	315.1703	C18H23N2O3	315.1703	0.0
M4	1‐Depropanoyl‐hydroxy‐P1	3.8	[M + H]+	3.32E+05	‐	366.2172	C22H28N3O2	366.2176	−1.09	6
			FI A	‐	100%	325.1783	C19H23N3O2	325.1785	−0.62
			FI B	‐	44%	225.1017	C14H13N2O	225.1022	−2.22
			FI C	‐	39%	224.0941	C14H12N2O	224.0944	−1.34
M5	1‐Depropanoyl‐N‐deethyl‐P	3.9	[M + H]+	2.75E+06	‐	322.1917	C29H24N3O	322.1914	0.93	2
			FI A	‐	100%	281.1523	C17H19N3O	281.1523	0.0
			FI B	‐	81%	208.0995	C14H12N2	208.0995	0.0
			FI C	‐	24%	182.0840	C12H10N2	182.0838	1.10
M6	1‐Depropanoyl‐hydroxy‐P2	3.9	[M + H]+	1.66E+05	‐	366.2177	C22H28N3O2	366.2176	0.27	11
			FI A	‐	100%	325.1783	C19H23N3O2	325.1785	−0.62
			FI B	‐	44%	225.1017	C14H13N2O	225.1022	−2.22
			FI C	‐	39%	224.0941	C14H12N2O	224.0944	−1.34
M7	1‐Depropanoyl‐di‐hydrodiol‐P3	4.2	[M + H]+	1.43E+05	‐	384.2281	C22H30N3O3	384.2282	−0.26	12
			FI A	‐	71%	283.1441	C17H19N2O2	283.1441	0.0
			FI B	‐	30%	208.0995	C14H12N2	208.0995	0.0
			FI C	‐	29%	309.1836	C19H23N3O	309.1836	0.0
M8	1‐Depropanoyl‐hydroxy‐P3	4.2	[M + H]+	1.79E+05	‐	366.2176	C22H28N3O2	366.2176	0	8
			FI A	‐	100%	325.1783	C19H23N3O2	325.1785	0.2
			FI B	‐	53%	224.0943	C14H12N2O	224.0944	0.1
			FI C	‐	23%	207.0917	C14H11N2	207.0917	0
M9	1‐Depropanoyl‐hydroxy‐P4	4.3	[M + H+]	1.74E+05	‐	366.2176	C22H28N3O2	366.2176	0	9
			FI A	‐	100%	325.1780	C19H23N3O2	325.1785	0.4
			FI B	‐	73%	224.0941	C14H12N2O	224.0944	0.3
			FI C	‐	60%	207.0916	C14H11N2	207.0917	0.1
M10	1‐Depropanoyl‐N‐deallyl‐P (Nor‐LSD)	4.3	[M + H]+	9.45E+05	‐	310.1913	C19H24N3O	310.1914	−0.32	4
			FI A	‐	100%	209.1072	C14H13N2	209.1073	−0.48
			FI B	‐	67%	74.0966	C4H12N	74.0964	2.70
			FI C	‐	24%	237.1022	C15H13N2O	237.1022	0.00
M11	1‐Depropanoyl‐dehydro‐P1	5.0	[M + H]+	8.50E+04	100%	348.2069	C22H26N3O	348.2070	−0.29	14
			FI A	‐	21%	306.1600	C19H20N3O	306.1601	−0.33
			FI B	‐	19%	235.0863	C15H11N2O	235.0866	−1.28
			FI C	‐	6%	277.1337	C18H17N2O	277.1335	0.72
M12	AL‐LAD (1‐depropanoyl‐P)	5.1	[M + H]+	5.10E+06	‐	350.2225	C22H28N3O	350.2227	−0.57	1
			FI A	‐	100%	208.0994	C14H12N2	208.0995	−0.48
			FI B	‐	81%	309.1835	C19H23N3O	309.1836	−0.32
			FI C	‐	18%	207.0916	C14H11N2	207.0917	−0.48
M13	N‐Deallyl‐P (1P‐Nor‐LSD)	5.3	[M + H]+	1.91E+05	‐	366.2177	C22H28N3O2	366.2176	0.27	7
			FI A	‐	100%	74.0966	C4H12N	74.0964	2.70
			FI B	‐	75%	265.1334	C17H17N2O	265.1335	−0.38
			FI C	‐	46%	209.1073	C14H13N2	209.1073	0.00
M14	1‐Depropanoyl‐dehydro‐P2	5.5	[M + H]+	5.13E+05	‐	348.2071	C22H26N3O	348.2070	0.29	5
			FI A	‐	100%	208.0994	C14H12N2	208.0995	−0.48
			FI B	‐	51%	207.0918	C14H11N2	207.0917	0.48
			FI C	‐	43%	307.1678	C19H21N3O	307.1679	−0.32

Note: Metabolites were ranked from 1 (for most abundant) to 14 (for least abundant).

FIGURE 4

Extracted ion chromatograms using the protonated molecules of 1P‐AL‐LAD and its metabolites monitored by HPLC‐ESI‐QTOF‐MS. numbering according to Table 2

FIGURE 5

Postulated phase I metabolic pathways of 1P‐AL‐LAD studied by in vitro incubations with pooled human liver microsomes. Numbering according to Table 2

The main metabolite (peak height 5.10E+06 cps) could be unambiguously identified as AL‐LAD (M12, C22H27N3O, [M + H]+, m/z 350.2223 Da), following hydrolysis of the N 1‐acyl substituent. The detection of AL‐LAD in the pHLM‐negative control sample (peak height 4.15E+05 cps) (supporting information) was hypothesized to be a result of a spontaneous hydrolysis, which has also been reported previously for 1P‐LSD. , However, as described above, analysis of a 1P‐AL‐LAD solution also revealed the detection of about 0.64% of AL‐LAD, which suggested that at least some proportion was present as an impurity. As shown in the supporting information, the detection of AL‐LAD in the pHLM‐positive incubation mixture confirmed its formation to a significant extent far beyond the very low abundance found in the pHLM‐negative sample.

The second most abundant metabolite detected from the in vitro assay was the 1‐depropanoyl‐N‐deethyl product (N‐deethyl‐AL‐LAD, M5, C20H23N3O, [M + H]+, m/z 322.19139 Da, peak height 2.75E+06 cps) following cleavage of both the 1‐propanoyl group and one of the ethyl groups from the N,N‐diethylamide moiety. Detection of the N 1‐deacylated dihydrodiol metabolite M1 (C22H29N3O3, [M + H]+, m/z 384.2282 Da) showed a relatively high abundance (peak height 1.43E+06 cps). Although the exact position of the hydroxyl functions could not be assigned, the product ion spectrum of M1 suggested that the metabolic oxidations might have taken place on the phenyl part of the indole ring and on ring C (see supporting information for all ESI‐QTOF‐MS/MS data of metabolites). In comparison, metabolite M3 was another dihydrodiol isomer of M1 detected in a much lower abundance (peak height 1.70E+05 cps). By contrast, in M7 (C22H29N3O3, [M + H]+, m/z 384.2282 Da, peak height 1.43E+05 cps) the dihydrodiol group was located on the former N 6‐allyl side chain. Four mono hydroxylated metabolites (M4, M6, M8, M9, C22H27N3O2, [M + H]+, m/z 366.2176 Da) were formed after hydrolysis of the 1‐propanoyl group. Oxidation/dehydration at the diethylamide moiety after N 1‐deacylation led to the formation of metabolite M14 (C22H25N3O, [M + H]+, m/z 348.2070 Da). The metabolic cleavage of the N 6‐allyl group, which has also been described for AL‐LAD previously, was also represented by metabolites M2 (C17H19N3O, [M + H]+, m/z 282.1601 Da), M10 (C19H23N3O, [M + H]+, m/z 310.1913 Da), and M13 (C22H29N3O2, [M + H]+, m/z 366.2176 Da). However, estimation of abundance based on signal responses alone must be viewed with caution because the ionization efficiency for each metabolite might have been different and reference material would be needed in order to get more detailed insights.

As shown in Figure 4, a number of peaks were detected under EIC conditions in addition to the proposed metabolites. For example, in the M2 trace (m/z 282.1601), another peak labeled as ‘artifact’ was detected at 3.9 min that represented a fragment ion of M5 (13C isotope of m/z 281.1523). The EIC trace presenting m/z 366.2176 (M4, M6, M8, M9, M13) also contained a peak at 4.9 min and labeled as an impurity (“imp”). As described above (Figure 3), one minor impurity identified both in the negative pHLM mixture and 1P‐AL‐LAD solution was hypothesized to be 7‐HO‐1P‐AL‐LAD. Correspondingly, the positive pHLM incubation solution revealed the presence of the metabolic N 1‐deacylation product of this impurity to form 7‐HO‐AL‐LAD. The mass spectral data recorded for this metabolite is shown in Figure 6. The tentative identification process (including similarities with the tandem mass spectrum of AL‐LAD; M12, supporting information), was similar to the one used above for 7‐HO‐1P‐AL‐LAD. The ‘artifact’ peak detected in the EIC trace of m/z 310.1914 (M10) at 5.1 min was identified as a fragment ion of M11 (13C isotope of m/z 309.1834). The peak eluting at 5.0 min in the trace presenting M11 (m/z 348.2070) might have reflected an isomer of M14 where another dehydro metabolite (double bond possibly between C4‐C5, supporting information) might have been generated. The EIC of m/z 382.2125 (C22H28N3O3, [M + H]+), showed four peaks in the pHLM positive incubations. The peak at 4.2 min could also be identified in the pHLM negative control and was interpreted as N‐deethyl‐hydroxy‐1P‐AL‐LAD. Three further peaks eluting at 3.4, 3.6, and 3.8 min could indicate the enzymatically‐catalyzed formation of 1‐depropanoyl‐di‐hydroxy metabolites. However, the product ions detected for any of the signals (supporting information) seemed inconsistent with the data for di‐hydroxy‐AL‐LAD, described by Wagmann et al. Because mass spectral interpretation did not allow for a clear differentiation between dihydroxy and oxo‐hydroxy metabolites, and signal intensities were relatively low (peak height <6.0E+04 cps), these potential metabolites were not considered identifiable. However, they should be kept in mind, when biological samples are obtained, especially because the 2‐oxo‐3‐hydroxy‐LSD metabolite appeared to be a major metabolite of LSD.

FIGURE 6

Tentative identification of 7‐HO‐AL‐LAD labelled as “imp.” in the EIC trace at m/z 366.2176 (Figure 4) [Colour figure can be viewed at wileyonlinelibrary.com]

Head‐twitch response

The 5‐HT2A receptor is the primary target for LSD and other psychedelic drugs in the brain. , In mice, psychedelic drugs induce a rapid side‐to‐side head shaking known as the head‐twitch response (HTR). The HTR serves as a rodent behavioral proxy for human psychedelic effects and can be used to distinguish hallucinogenic and non‐hallucinogenic 5‐HT2A agonists. HTR experiments were conducted with 1P‐AL‐LAD in male C57BL/6J mice to determine whether it has an LSD‐like behavioral profile. After administration of 1P‐AL‐LAD to mice, there was a dose‐dependent increase in HTR counts (W(5,11.49) = 13.85, p = 0.0002). The response induced by 1P‐AL‐LAD had an inverted U‐shaped dose–response function (Figure 7), which is also the case for LSD and other psychedelic drugs). The maximal response occurred after administration of 0.5 mg/kg 1P‐AL‐LAD (p < 0.01, Dunnett's T3 multiple comparisons test). The median effective dose (ED50 value) for 1P‐AL‐LAD was 236 (95% CI: 155–322) μg/kg. Based on its molecular weight, 1P‐AL‐LAD induced the HTR with ED50 = 491 (323–670) nmol/kg. The reported ED50 for AL‐LAD is 174.9 nmol/kg, making it about three times as potent as 1P‐AL‐LAD. In recent studies, 1P‐LSD and 1cP‐LSD were reported to induce the HTR with ED50 values of 349.6 nmol/kg and 430 nmol/kg, making them slightly more potent than 1P‐AL‐LAD. ,

FIGURE 7

Dose‐dependent increase of head‐twitch response counts [Colour figure can be viewed at wileyonlinelibrary.com]

Previous studies have shown that N 1‐acyl substitution reduces the affinity of LSD for the 5‐HT2A receptor and markedly dampens its agonist efficacy. Nevertheless, N 1‐acyl‐substituted derivatives of LSD such as ALD‐52 and 1P‐LSD act as psychedelic drugs in humans and induce the HTR in mice, indicating those molecules retain the ability to activate the 5‐HT2A receptor after in vivo administration. Because there is evidence from both in vitro and in vivo studies that ALD‐52 and 1P‐LSD are metabolized to LSD, , , those molecules are believed to serve as prodrugs for LSD. Although AL‐LAD has high affinity for the 5‐HT2A receptor (K i = 8.1 nM for [3H]ketanserin‐labeled sites in rat frontal cortex) and acts as a 5‐HT2A agonist in calcium flux assays, the activity of 1P‐AL‐LAD at the 5‐HT2A receptor is likely to be relatively weak because it contains an N 1‐acyl group. Based on the pHLM data, 1P‐AL‐LAD is likely to be hydrolyzed to AL‐LAD after administration to mice and therefore acts as a prodrug, similar to ALD‐52 and 1P‐LSD.

For LSD and a large series of psychedelic drugs, there is a robust correlation (r = 0.9448) between HTR ED50 values in C57BL/6J mice and hallucinogenic potencies in humans. When tested previously in HTR experiments, LSD (ED50 = 132.8 nmol/kg) and AL‐LAD (ED50 = 174.9 nmol/kg) had three to four times higher potency than 1P‐AL‐LAD. Although it is tempting to speculate based on the HTR data that 1P‐AL‐LAD may also have 3–4‐fold lower potency than AL‐LAD and LSD in humans, the relationship between psychedelic drug potencies in mice and humans may not extend to molecules acting as prodrugs. The identity and tissue expression of the enzymes responsible for the hydrolysis of 1P‐AL‐LAD to AL‐LAD may not be uniform across different species, potentially resulting in considerable cross‐species variations in potency. Therefore, additional studies are warranted to understand the activity of 1P‐AL‐LAD in humans and to compare its potency relative to LSD and AL‐LAD.

CONCLUSION

Much of the interest in the properties of novel lysergamides reflects both the accumulating evidence of their therapeutic properties as well as their use and distribution as recreational drugs. 1P‐AL‐LAD is a relatively new addition to the lysergamide family. A comprehensive analytical characterization of 1P‐AL‐LAD was performed involving several forms of mass spectrometry, gas‐ and liquid chromatography, nuclear magnetic resonance spectroscopy and gas chromatography solid‐phase infrared analysis aimed at supporting stakeholders interested in psychoactive drug research. Incubations with pooled human liver microsomes confirmed that 1P‐AL‐LAD is hydrolyzed to AL‐LAD as the most abundant metabolite together with 13 other metabolites of varying signal response, which is consistent with the hypothesis that 1P‐AL‐LAD acts as a prodrug. The ability of 1P‐AL‐LAD to induce the HTR in mice shows it has a behavioral profile reminiscent of LSD and other serotonergic psychedelic drugs. Ultimately, human trials with 1P‐AL‐LAD are required to define its effects and abuse potential in humans.

Data S1. Supporting Information

Click here for additional data file.