Identification of LSD analogs, 1cP-AL-LAD, 1cP-MIPLA, 1V-LSD and LSZ in sheet products

HPLC-grade acetonitrile and methanol were purchased from Kanto Chemical Co., Inc. (Tokyo, Japan). Methanol-d₄ (99.8 at. % D) was purchased from Isotec, Inc. (Miamisburg, OH, USA). LSD was purchased from Cerilliant Corporation (Round Rock, TX, USA). (2’S,4’S)-Lysergic acid 2,4-dimethylazetidide (LSZ) was purchased from Chiron (Trondheim, Norway). Bond Elut C18, 500 mg, 3 mL (Agilent, Santa Clara, CA, USA) was used for purification.

The four products (A to D) analyzed in this study were obtained in Japan between August 2021 and March 2022. Each sheet was a square with an overall size of 4.5 × 4.5 cm and perforated in a grid pattern with approximately 9 mm on each side. The structural formula of the LSD analog suggested to be contained was printed on one side of the sheet, and the name of the compound was printed on the reverse side.

Eight mg of each sheet cut into 2 mm squares was extracted with 1 mL of methanol in 15 min under sonication. Methanol was removed from the extract under a nitrogen stream and redissolved in 1 mL of acetonitrile for analysis by gas chromatography–mass spectrometry (GC–MS), liquid chromatography–photodiode array–mass spectrometry (LC–PDA–MS) and liquid chromatography with hybrid quadrupole time-of-flight mass spectrometry (LC–Q-TOF-MS). For nuclear magnetic resonance (NMR) spectroscopy, each piece of a cut sheet was sonicated in 1.0 mL of methanol at room temperature for 5 min. The extraction procedure was repeated three times. The extracts were combined and the solvent was removed using an evaporator. Each residue was purified with Bond Elut C18 eluted with water–methanol, dissolved in 0.3 mL of methanol-d₄, and then subjected to NMR spectroscopy.

The LC–PDA–MS analysis was performed on an ACQUITY UPLC system with a mass detector and a photodiode array (PDA) detector (Waters, Milford, MA, USA). Chromatographic separation was performed using an ACQUITY HSS T3 (2.1 mm i.d. × 100 mm, 1.8 μm particle size, Waters) UPLC column protected by Van Guard HSS T3 (2.1 mm i.d. × 5 mm, 1.7 μm particle size, Waters) at 40 °C. For the LC–MS analysis, we employed a binary mobile phase of solvent A (0.1% formic acid in water) and solvent B (0.1% formic acid in acetonitrile). The samples were analyzed by the following elution program: 5–20% B (0–20 min) and then up to 80% B (20–30 min, 10-min hold) at a flow rate of 0.3 mL/min. The injection volume was 1 μL and the wavelength of the PDA detector for screening was set from 210 to 450 nm. The MS conditions for the LC–ESI-MS were as follows: positive and negative ionization, nitrogen desolvation gas (flow rate 650 L/h at 350 °C), capillary and cone voltages of 2500 V and 30 V, respectively, and mass spectral range m/z 120–650 [9, 10].

The GC–MS was performed on an Agilent 6890 N GC system with a 5975 mass-selective detector (Agilent Technologies, Santa Clara, CA, USA) using a capillary column (DB-1HT capillary, 15 m × 0.25 mm i.d., 0.10-μm film thickness; Agilent Technologies) with helium-gas carrier flowing at 1.0 mL/min. The conditions were as follows: electron energy, 70 eV; injector temperature, 200˚C; injection mode, splitless mode for 1.0 min; transfer line temperature, 280˚C; scan range, m/z 40–550. The oven temperature was held at 120˚C for 1 min, and then increased at 15˚C/min to 280˚C, where it was held for 5 min [9, 10].

The HR-MS analysis was carried out on a TripleTOF^® 6600 LC/MS/MS system (AB SCIEX, Framingham, MA, USA) and a Nexera X2 system (Shimadzu, Kyoto, Japan). Chromatographic separation was performed in an ACQUITY HSS T3 (2.1 mm i.d. × 100 mm, 1.8 μm particle size, Waters) UPLC column protected by Van Guard HSS T3 (2.1 mm i.d. × 5 mm, 1.7 μm particle size, Waters) at 40 °C. The mobile phase was a binary phase of solvent A (0.1% formic acid in water) and solvent B (0.1% formic acid in acetonitrile) with a gradient program of A/B 95/5–5/95 (10 min, 2 min hold). The flow rate was 0.3 mL/min, and the elution was monitored at 210–450 nm. The mass spectrometer was operated in ESI mode with an ion-spray voltage of 5500 V (positive mode), a source temperature of 550 °C, an ion-source gas of nitrogen, a pressure of 50 psi for ion-source gases 1 and 2, a curtain gas pressure of 25 psi, and a declustering potential of 80 V. The samples were analyzed in TOF-MS scan mode (m/z 100–650) [9, 10].

The ¹H-NMR and ¹³C-NMR spectra were measured using a JEOL JMN-ECA800 or ECZ800 spectrometer (JEOL, Tokyo, Japan). The chemical shifts were presented with reference to the residual deuterated methanol (CD₃OD, δ_H 3.33 ppm and δ_C 49.0 ppm) in the NMR. Compound identification was performed using ¹H-NMR, ¹³C-NMR, heteronuclear multiple quantum coherence (HMQC), heteronuclear multiple bond correlation (HMBC), H–H correlation spectroscopy (H–H COSY) and nuclear Overhauser effect spectroscopy (NOESY).

In the LC–PDA–MS analysis of product A, the peak for compound 1 was detected at 16.0 min with a protonated molecular ion ([M + H]⁺) at m/z 418 (Fig. 2a and b). In the GC–MS analysis, the peak at 13.4 min indicated a molecular ion ([M]⁺) at m/z 417 (Fig. 3a, b). The accurate mass spectrum of compound 1 displayed an ion peak at m/z 418.2490, and the estimated composition of the protonated molecular formula was C₂₆H₃₂N₃O₂ (calc 418.2489 (0.1 mDa)). The ¹³C-NMR spectra indicated the existence of another carbonyl (δ_C 174.4 ppm) in addition to the amide portion of the LSD. Two methylene signals [(δ_H 1.10 ppm, δ_C 10.1 ppm), (δ_H 1.19 ppm, δ_C 10.1 ppm)] and one methine signal (δ_H 2.50 ppm, δ_C 14.4 ppm) were also observed, and 2D NMR correlations revealed the presence of cyclopropanecarbonyl at the N1 position of LSD (Fig. 4). The ¹H-NMR and ¹³C-NMR shift values of compound 1 were almost the same as those of 1cP-LSD, except for positions 5 and 7. The molecular weight of compound 1 was 26 more than that of 1cP-LSD, and the presence of a partial structure [CH = CH₂ (δ_H 6.01 ppm, δ_C 134.7 ppm), CH = CH₂ (δ_H 5.28, 5.35 ppm, δ_C 119.9 ppm)], which appeared to be a terminal olefin, suggested the presence of an allyl group (Tables 1, 2). The methyl group at N6 was not observed, and correlations were observed between the methylene protons of the allyl group and the carbons at positions 5 and 7, as shown in Fig. 4, indicating that there is an allyl group at position 6 rather than a methyl group. Thus, the structure of compound 1 was determined as 4-(cyclopropanecarbonyl)-N,N-diethyl-7-(prop-2-en-1-yl)-4,6,6a,7,8,9-hexahydroindolo[4,3-fg]quinoline-9-carboxamide (1cP-AL-LAD (1)), as shown in Fig. 1.

In the LC–PDA–MS analysis of product B, a peak of compound 2 was detected at 13.7 min with a protonated molecular ion ([M + H]⁺) at m/z 392 (Fig. 5a, b). In the GC–MS analysis, a peak at 13.0 min revealed a molecular ion ([M]⁺) at m/z 391 (Fig. 6a, b). The accurate mass spectrum of compound 2 revealed an ion peak at m/z 392.2333, and the estimated composition of the protonated molecular formula was C₂₄H₃₀N₃O₂ [calc 392.2333 (0 mDa)]. The NMR spectroscopic data showed the presence of another carbonyl (δ_C 174.4 ppm), two methylenes [(δ_H 1.10 ppm, δ_C 10.1 ppm), (δ_H 1.19 ppm, δ_C 10.1 ppm)] and one methine (δ_H 2.50 ppm, δ_C 14.4 ppm) in addition to the LSD moiety, and they were correlated in 2D NMR spectrum, as shown in Fig. 4. Thus, compound 3 is presumed to be cyclopropionylation at the N1 position, as is 1cP-AL-LAD (1). Two doublet methyl [(δ_H 1.15 ppm, δ_C 19.5 ppm), (δ_H 1.19 ppm, δ_C 19.5 ppm)] and methine (δ_H 4.82 ppm, δ_C 46.2 ppm) signals were observed, suggesting the presence of isopropyl groups (Table 1 and 2). One singlet methyl (δ_H 3.05 ppm, δ_C 29.1 ppm) was observed, but no signal derived from the diethyl group at the N18 of LSD was observed. The correlation shown in Fig. 4 indicates that the N18 position of LSD is not a diethyl group but methyl and isopropyl groups. Thus, the structure of compound 2 was determined as 4-(cyclopropanecarbonyl)-N-methyl-N-isopropyl-7-methyl-4,6,6a,7,8,9-hexahydroindolo-[4,3-fg]quinoline-9-carboxamide (1cP-MIPLA (2)), as shown in Fig. 1.

In the ¹H-NMR and ¹³C-NMR spectra, there are minor signals at positions 7–9 and at the N-methyl-N-isopropyl moiety. As the ratios of these minor signals were changed in different solvents and the same phenomenon was observed for MIPLA, it was presumed that these minor signals were derived from the rotational isomer of the N18 amide moiety of 1cP-MIPLA.

In the LC–PDA–MS analysis, a peak of 2b with the same m/z 392 [M + H]⁺ as 1cP-MIPLA was detected at 15.8 min (Fig. 5b), in addition to the peak of 1cP-MIPLA. The UV spectrum of 2b was not similar to that of 1cP-MIPLA. In addition, the signals presumed to be derived from minor compounds other than 1cP-MIPLA were also observed in the ¹H-NMR and ¹³C-NMR spectra. It has been reported that the epimerization of LSD at position 8 occurs to produce iso-LSD [13, 14]. As epimerization at position 8 also occurs during the synthesis of LSD analogs [15], this minor compound was presumed to be iso-1cP-MIPLA.

In the LC–PDA–MS analysis of product C, a peak of compound 3 was detected at 18.0 min with a protonated molecular ion ([M + H]⁺) at m/z 408 (Fig. 7a, b). In the GC–MS analysis, the peak at 13.0 min revealed a molecular ion ([M]⁺) at m/z 407 (Fig. 8a, b). The accurate mass spectrum of compound 3 contained an ion peak at m/z 408.2647, and the estimated composition of the protonated molecule formula was C₂₆H₃₂N₃O₂ (calc 408.2646 (0.1 mDa)). The NMR spectroscopic data showed the presence of another carbonyl (δ_C 173.8 ppm), three methylenes [(δ_H 2.97 ppm, δ_C 36.2 ppm), (δ_H 1.78 ppm, δ_C 28.0 ppm), (δ_H 1.48 ppm, δ_C 23.4 ppm)], and one methyl (δ_H 0.99 ppm, δ_C 14.2 ppm) in addition to the LSD moiety, and they were correlated in the 2D NMR spectra, as shown in Fig. 4. The data suggested the presence of a pentanoyl group. The chemical shift values of ¹H-NMR and ¹³C-NMR spectra were in good agreement with those of 1B-LSD (Tables 1 and 2). As the estimated compositional formula of compound 3 contains more CH₂ than that of 1B-LSD, and the correlation was observed, as shown in Fig. 4, the N1 position of LSD was presumed to be pentanoylated. Thus, the structure of compound 3 was determined as N,N-diethyl-7-methyl-4-pentanoyl-4,6,6a,7,8,9-hexahydroindolo[4,3-fg]quinoline-9-carboxamide (1V-LSD (3)) [12], as shown in Fig. 1.

In the LC–PDA–MS analysis, a peak of 3b with the same m/z 392 [M + H]⁺ as 1V-LSD was detected at 20.7 min (Fig. 7b) in addition to the peak of 1V-LSD. The UV spectrum of 3b was not similar to that of 1V-LSD. Moreover, the signals presumed to be derived from minor compound other than 1V-LSD were also observed in the ¹H-NMR and ¹³C-NMR spectra. This minor compound was estimated to be iso-1V-LSD, C8-epimerized 1V-LSD.

In the LC–PDA–MS analysis of sheet D, a peak of compound 4 was detected at 6.8 min with a protonated molecular ion ([M + H]⁺) at m/z 336 (Fig. S1). In the GC–MS analysis, a peak at 11.4 min showed a molecular ion ([M]⁺) at m/z 335 (Fig. S2a and S2b). After comparing the data from LC–PDA–MS and GC–MS analyses with those of the authentic compound, this compound was identified as LSZ (4) [5, 16], a compound in which the diethyl moiety at position N18 of LSD has been converted to 2,4-dimethylazetidin.