Forensic Toxicology

, Volume 29, Issue 1, pp 25–37 | Cite as

Identification and quantitation of two cannabimimetic phenylacetylindoles JWH-251 and JWH-250, and four cannabimimetic naphthoylindoles JWH-081, JWH-015, JWH-200, and JWH-073 as designer drugs in illegal products

  • Nahoko Uchiyama
  • Maiko Kawamura
  • Ruri Kikura-Hanajiri
  • Yukihiro Goda
Original Article

Abstract

Six cannabimimetic indoles have been identified as adulterants in herbal or chemical products being sold illegally in Japan, with four of the compounds being new as adulterants to our knowledge. The identifications were based on analyses using gas chromatography–mass spectrometry, liquid chromatography–mass spectrometry, high-resolution mass spectrometry, and nuclear magnetic resonance spectroscopy. The first two compounds were identified as phenylacetyl indoles JWH-251 (2-(2-methylphenyl)-1-(1-pentyl-1H-indol-3-yl)ethanone; 1) and its demethyl-methoxylated analog JWH-250 (2-(2-methoxyphenyl)-1-(1-pentyl-1H-indol-3-yl)ethanone; 2). Compound 2 was identical to that found as an adulterant in the UK and in Germany in 2009. The third compound was naphthoylindole JWH-081 (1-(4-methoxynaphthalenyl)-(1-pentyl-1H-indol-3-yl)methanone; 3), and the fourth was JWH-073 (1-naphthalenyl(1-butyl-1H-indol-3-yl)methanone; 4), which had been identified as an adulterant in our previous study. Two additional compounds were JWH-015 (1-naphthalenyl(2-methyl-1-propyl-1H-indol-3-yl)methanone; 5) and JWH-200 (1-naphthalenyl(1-(2-(4-morpholinyl)ethyl)-1H-indol-3-yl)methanone; 6). Compounds 14 and 6 were reported to be synthetic cannabinoids with selective affinity for cannabinoid CB1 receptors, while compound 5 was reported to be a selective CB2 receptor agonist causing immunosuppressive effects without psychotropic affects. One product contained both CB1 and CB2 receptor agonists in our collection. Quantitative analyses of the six cannabimimetic compounds in 20 products revealed that there was large variation in concentrations of the detected compounds among products; for herbal cutting products, the total amounts of these cannabinoids ranged from 26 to 100 mg.

Keywords

JWH-251 JWH-081 JWH-015 JWH-200 Synthetic cannabinoid Designer drug 

Introduction

Recently, a number of psychotropic herbal products have been marketed on the Internet under brand names such as “Spice” and “herbal blends” [1, 2, 3, 4, 5]. We have reported that two types of synthetic cannabinoids are present as psychoactive ingredients in herbal products [1, 2, 4]. German scientists have also found these compounds in some herbal products [6, 7]. The first drug group consists of cyclohexylphenols such as cannabicyclohexanol (CCH; (1RS,3SR)-3-[2-hydroxy-4-(2-methylnonan-2-yl)phenyl]cyclohexan-1-ol) and CP 47497 ((1RS,3SR)-3-[2-hydroxy-4-(2-methyloctan-2-yl)phenyl]cyclohexan-1-ol); these compounds exert potent cannabimimetic actions [8, 9, 10, 11, 12]. The second group consists of naphthoylindoles, such as JWH-018 (1-naphthalenyl-(1-pentyl-1H-indol-3-yl)methanone) and JWH-073 (1-naphthalenyl-(1-butyl-1H-indol-3-yl)methanone, 4). Although the chemical structures of JWH-018 and JWH-073 are greatly different from those of natural psychoactive cannabinoids such as Δ9-tetrahydrocannabinol (Δ9-THC), the former compounds have higher affinities for cannabinoid CB1 receptors and activities that are comparable with or more potent than Δ9-THC [13, 14, 15, 16]. Since January 2009, JWH-018, CP 47497, and three homologs of CP 47497, including CCH, have been controlled in Germany [17], followed by Austria, France, Sweden, and other countries [18]. In Japan, CCH, CP 47497, and JWH-018 have been controlled as designated substances (Shitei-Yakubutsu) under the Pharmaceutical Affairs Law since November 2009. Despite such control measures, new synthetic cannabinoids are continuing to appear throughout the world. The classical cannabinoid HU-210 ((6aR,10aR)-3-(1,1-dimethylheptyl)-6a,7,10,10a-tetrahydro-1-hydroxy-6,6-dimethyl-6H-dibenzo[b,d]pyran-9-methanol), which is a potent cannabinoid receptor CB1 agonist, was identified in herbal products in the USA and the UK [18]. In addition, two novel synthetic JWH cannabinoids, JWH-398 (4-chloro-1-naphthalenyl-(1-pentyl-1H-indol-3-yl)methanone) and JWH-250 (2-(2-methoxyphenyl)-1-(1-pentyl-1H-indol-3-yl)ethanone, 2), were found in the UK and Germany in October 2009 [18]. Both compounds were reported as cannabinoid receptor agonists [19, 20]. However, no scientific reports with details of identification and isolation of these compounds from herbal or chemical products have been published. In the present study, we describe identification and quantitative analyses of six cannabimimetic aminoalkylindoles, including four new ones as designer drugs, in herbal or chemical products commercially available in Japan. The structures of the six compounds dealt with in this study are shown in Fig. 1.
Fig. 1

Structures of cannabimimetic indoles detected in the present study

Materials and methods

Chemicals and reagents

Betamethasone valerate (internal standard, IS) and JWH-015 (5) were purchased from Wako (Osaka, Japan); JWH-250 (2), JWH-073 (4), and JWH-200 (6) from Cayman Chemical Company (Ann Arbor, MI, USA). All other common chemicals and solvents were of analytical reagent grade or high-performance liquid chromatography (HPLC) grade.

Samples for analysis

Eighteen herbal products and two powder products being sold in Japan for their expected cannabis-like effects were purchased via the Internet from December 2009 to April 2010. All products had different names and were packaged differently. Sixteen of the herbal products were in the form of dried leaves and the remaining two were in the form of solids (resins). The labels on the packages indicated that the products contained 1–3 g of a mixture of plants. The two powder products were pale yellow powders, and the labels on the packages indicated that the products contained 200 and 250 mg, respectively. On the basis of the advertisements on the website, the single dosage amounts were estimated to be between 20 and 25 mg for powder products and between 200 and 300 mg for dried leaf products being finely cut.

Analytical conditions

The sample solutions were qualitatively and quantitatively analyzed by liquid chromatography-mass spectrometry (LC–MS) with positive electrospray ionization (ESI). The instrument consisted of an ACQUITY UPLC system, a mass detector, and a photodiode array (PDA) detector (Waters, Milford, MA, USA). The sample solutions were separated with an ACQUITY UPLC HSS T3 column (100 mm × 2.1 mm i.d., particle size 1.8 μm; Waters) protected by a Van Guard column (5 mm × 2.1 mm i.d., 1.8 μm; Waters) at 40°C. Each analysis was carried out with a binary mobile phase consisting of solvent A (0.1% formic acid in water) and solvent B (0.1% formic acid in acetonitrile). An elution program (1) with a linear gradient was: 50% B (3-min hold) to 70% B (3–5 min), and 70% B with 7-min hold (5–12 min) at a flow rate of 0.3 ml/min. Another elution program (2) was also used for quantitation of 6 as follows: 30% B (3-min hold) to 70% B (3–5 min), and 70% B with 7-min hold (5–12 min). The injection volume was 1 μl for both programs. The wavelength of the PDA detector for screening was set from 190 to 500 nm.

The MS conditions for the LC–ESI–MS were: ionization, positive; desolvation gas, nitrogen at a flow rate of 650 l/h at 350°C; capillary and cone voltages, 3000 and 30 V, respectively; mass spectral range, m/z 50–500. For qualitative analysis of JWH-251 (1), 2, JWH-081 (3), 4, 5, and 6, the protonated molecular peaks ([M + H]+) of the compounds and IS were monitored in the scan mode. The monitoring ions were: 1, m/z 320; 2, m/z 336; 3, m/z 372; 4 and 5, m/z 328; 6, m/z 385; and betamethasone valerate (IS), m/z 477.

Gas chromatography-mass spectrometry (GC–MS) analysis was also performed in electron ionization (EI) mode at 70 eV of electron energy according to our previous report [21]. It was performed on a Hewlett-Packard 6890N GC with a 5975 mass selective detector using a capillary column (HP1-MS capillary, 30 m × 0.25 mm i.d., 0.25 μm film thickness; Hewlett-Packard, Palo Alto, CA, USA) with helium gas as a carrier at 0.7 ml/min. The conditions were: injector temperature, 200°C; injection, splitless mode for 1.0 min; oven temperature program, 80°C (1-min hold) and increase at a rate of 5°C/min to 190°C (15-min hold) followed by increase at 10°C/min up to 310°C (15-min hold); mass selective detector temperature, 280°C; scan range, m/z 40–550.

The accurate mass spectrum of the target compound was measured using a direct analysis in real time (DART) ion source coupled to a time-of-flight (TOF) mass spectrometer (AccuTOF JMS-100LC; JEOL, Tokyo, Japan) operated in positive ion mode. The measurement conditions were: ion guide peak voltage, 500 V; reflectron voltage, 950 V; orifice 1 voltage, 15 V; orifice 2 voltage, 5 V; ring lens voltage, 5 V; orifice 1 temperature, 80°C; mass range, m/z 100–500. The conditions of the DART ion source were: helium gas flow rate, 2.0 l/min; gas heater temperature, 250°C; discharge electrode needle voltage, 3200 V; voltages of electrodes 1 and 2, 100 and 250 V, respectively. Internal mass number calibration was achieved using PEG600, and diphenhydramine (C17H21NO) and verapamil (C27H38N2O4) were also used as ISs for each accurate mass analysis. The product itself or an extract was directly exposed to the vicinity of the DART ion source.

For nuclear magnetic resonance (NMR) analysis, CDCl3 (99.96%) was purchased from the ISOTEC division of Sigma–Aldrich (St. Louis, MO, USA). The NMR spectra were obtained on ECA-600 spectrometers (JEOL). Assignments were made via 1H NMR, 13C NMR spectra, heteronuclear multiple quantum coherence (HMQC) spectra, heteronuclear multiple-bond correlation (HMBC) spectra, double quantum filtered correlation spectroscopy (DQF-COSY) spectra, and rotating frame nuclear Overhauser effect (ROE) spectra.

For isolation of compounds, a recycling preparative HPLC (Japan Analytical Industry, Tokyo, Japan) was used with a JAIGEL-GS310 column (500 mm × 20 mm i.d.; Japan Analytical Industry) and ultraviolet (UV) absorbance and refractive index (RI) detectors.

Isolation of compounds 1 and 3

A 3-g sample of a herbal product was extracted with 100 ml of chloroform by ultrasonication for 1 h. The extractions were repeated three times, the supernatant fractions were combined and evaporated to dryness. The extract was placed on a preparative silica gel thin-layer chromatography (TLC) plate (Silica Gel 60, 20 × 20 cm, 2 mm; Merck, Darmdstadt, Germany), which was then developed using hexane/ethyl acetate (4:1). A portion of the silica gel in the TLC plate was scraped and the target compound was eluted with chloroform/methanol (3:1) to get fractions 1 and 2. Each fraction was further purified by recycling preparative HPLC with chloroform/methanol (1:1) to give compound 1 (41 mg) as a pale yellow oil and compound 3 (111 mg) as a white solid, respectively.

Isolation of compound 2

A 60-mg sample of the pale yellow powder was dissolved in 2 ml of chloroform/methanol (1:1) and passed through a centrifugal filter (Ultrafree-MC, 0.45 μm filter unit; Millipore, Bedford, MA, USA). The solution was subjected to recycling preparative HPLC with chloroform/methanol (1:1) to obtain compound 2 (31 mg) as a white solid.

Standard solutions

For qualitative and quantitative analysis, standard solutions were prepared for each compound (16) at a concentration of 1.0 or 0.1 mg/ml in methanol.

Calibration curves

The concentrations of compounds 16 in the samples were calculated using the peak area ratios of 1 and 2 at 302 nm versus IS at 240 nm, and those of 36 at 314 nm versus IS at 240 nm, respectively. Compounds 16 were diluted with methanol to prepare calibration solutions containing 10, 25, 50, 100, 250, and 500 μg/ml (14) or 10, 25, 50, 100, and 250 μg/ml (5 and 6). The solutions also included IS (betamethasone valerate) at 100 μg/ml.

Precision and accuracy of the method

The precision and accuracy of the method were evaluated by analyzing triplicates of the standard solutions containing 10, 50, and 500 μg/ml (14) or 10, 50, and 250 μg/ml (2, 5 and 6) for each compound. Accuracy, expressed as bias, was calculated as the percent difference between the amounts of each compound added and recovered.

Sample extraction procedure before instrumental analyses

For quantitative and qualitative analyses, the herbal product (5–10 mg) after being crushed into powder or the powdery product (5 mg) was extracted with 1 ml of methanol including IS (100 μg/ml) under ultrasonication for 10 min. After centrifugation (5 min, 3,000 rpm), the supernatant solution was passed through a centrifugal filter (Ultrafree-MC, 0.45 μm filter unit; Millipore). If necessary, the solution was diluted with methanol to a suitable concentration before instrumental analyses.

Results and discussion

Identification of unknown peaks 13

Three unknown peaks 13 were observed in the total ion chromatogram (TIC) obtained by GC–MS for product No. 11 (Fig. 2a). Unknown peak 1 at 48.2 min showed a mass spectrum with four major ion signals at m/z (% relative intensity) 319 (5), 214 (100), 144 (17), and 105 (3) as shown in Fig. 2b. LC–MS analysis showed the corresponding three unknown peaks both in the PDA detection and TIC (Fig. 3a, e). Peak 1 at 9.0 min showed a major ion signal at m/z 320 [M + H]+ and absorbance maxima at 246 and 303 nm in the UV spectrum (Fig. 3b, f). In the accurate mass spectrum obtained by DART–TOF–MS with direct exposure of the sample extract to the ion source, the major ion peak showed a protonated molecular ion signal ([M + H]+) at m/z 320.19966 in the positive mode, suggesting that the molecular formula of 1 was C22H26NO. The error between the mass number observed and the theoretical mass number of [M + H]+ was −1.78 mDa.
Fig. 2

Gas chromatography–mass spectrometry (GC–MS) analysis of product No. 11. Total ion chromatogram (TIC) (a), electron ionization (EI) mass spectra of the detected peaks at the retention times (RTs) 48.2 (b, 1), 49.1 (c, 2), and 53.7 min (d, 3), and the EI mass spectrum of the standard of JWH-250 (RT 49.1 min, e)

Fig. 3

Liquid chromatography (LC)-ultraviolet (UV) detection (a) and LC–MS analysis for the extract of product No. 11. Mass chromatograms obtained at m/z 320 (b, 1), 336 (c, 2), and 372 (d, 3), and the TIC (e); UV spectra and electrospray ionization (ESI) mass spectra for each peak (fh) and those of the standard of JWH-250 (i)

To elucidate the exact chemical structure of compound 1 by NMR analysis, we isolated 1 from the extract by TLC and preparative HPLC as described before. The 1H NMR spectrum of 1 showed 25 nonexchangeable protons, including two methyl signals at δ 0.90 (3H, t, J = 7.2 Hz) and 2.33 (3H, s), and nine aromatic proton signals at δ 7.04, 7.16, 7.18, 7.19, 7.22, 8.41 (each 1H, m), 7.28 (1H, td, J = 6.9, 1.7 Hz), 7.30 (1H, td, J = 6.9, 1.7 Hz), and 7.74 (1H, s) as shown in Table 1. The spectrum also showed three methylene proton signals at δ 1.30, 1.35 (each 2H, m) and 1.87 (2H, q, J = 7.2 Hz), as well as two characteristic methylene signals connected to a nitrogen atom at δ 4.14 (2H, t, J = 7.2 Hz) and next to a carbonyl group at δ 4.20 (2H, s). The 13C NMR spectrum of 1 showed 22 carbon signals, suggesting the presence of 2 methyls (δ 13.9 and 20.0), 5 methylenes with a nitrogenated carbon (δ 47.1) and a carbon adjacent to a carbonyl group (δ 44.9), 9 aromatic carbons (δ 109.8, 122.6, 122.8, 123.3, 126.0 126.9, 130.3, 130.3, and 134.4), 5 aromatic quaternary carbons (δ 116.2, 126.7, 134.6, 136.6, and 137.0), and 1 carbonyl carbon (δ 192.7). The analyses of DQF-COSY, HMQC, and HMBC spectra indicated that 1 has three partial structures (1,3-substituted indole group; 1,2-substituted phenyl group; and n-pentyl group), as shown in Table 1 and Fig. 4. The connectivity of these groups and the carbonyl methyl group was deduced from the HMBC spectrum. An aromatic proton at δ 7.74 (H-2′) of the indole group correlated to the carbonyl carbon at δ 192.7 (C-1) and the methylene carbon of the n-pentyl group at δ 47.1 (C-1″). A methylene proton next to a carbonyl group at δ 4.20 (H-2) showed correlations to the carbon at δ 116.2 (C-3′) of the indole group and the carbons of the phenyl group at δ 134.6 (C-2″′) and 130.3 (C-6″′). In addition, a methyl proton singlet resonance at δ 2.33 (Me-2″′) correlated to the carbons of the phenyl group at δ 137.0 (C-1″′) and 130.3 (C-3″′). Furthermore, the irradiation of the methylene proton at δ 4.14 (H-1″) of the n-pentyl group resulted in ROE correlations on the aromatic protons (H-2′ and H-7′) as shown in Fig. 4. The other methylene proton (H-2) also showed ROE correlations to the aromatic protons (H-2′ and H-6″′) and methyl proton (Me-2″′). On the basis of these mass and NMR spectral data (Figs. 2, 3, 4; Table 1), the structure of compound 1 was finally elucidated as 2-(2-methylphenyl)-1-(1-pentyl-1H-indol-3-yl)ethanone. The deduced compound had been synthesized as a potent cannabinoid receptor agonist and was named JWH-251 by Huffman et al. [20]. However, the present study is the first to detect compound 1 as a designer drug and an adulterant in illegal products. It should be noted that compound 1 is one of the cannabimimetic indoles with a phenylacetyl substituent, which are different from naphthoylindoles such as JWH-018 and JWH-073 (4) (Fig. 1).
Table 1

Nuclear magnetic resonance (NMR) data for compounds 1 and 3 in CDCl3

No.

JWH-251 (1)a

Referenceb

JWH-081 (3)a

13C

1H

HMBCc

13C

13C

1H

HMBCc

1

192.7

191.7

191.8

2

44.9

4.20, 2H, s

1, 3′, 1″′, 2″′, 6″′

2′

134.4

7.74, 1H, s

1, 3′, 3′a, 7′a, 1″

137.4

137.4

7.41, 1H, s, overlapped

1, 3′, 3′a, 7′a, 1″

3′

116.2

117.6

117.7

3′a

126.7

127.1

127.2

4′

122.8

8.41, 1H, m

3′, 3′a, 5′, 6′, 7′a

122.7

122.9

8.46, 1H, m

3′, 3′a, 5′, 6′, 7′a

5′

122.6

7.28, 1H, td, J = 6.9, 1.7 Hz, overlapped

3′a, 4′, 6′, 7′

123.3

123.4

7.35, 1H, m, overlapped

3′a, 4′, 7′

6′

123.3

7.30, 1H, td, J = 6.9, 1.7 Hz, overlapped

4′, 7′, 7′a

122.5

122.6

7.34, 1H, m, overlapped

4′, 7′, 7′a

7′

109.8

7.04, 1H, m

3′a, 5′, 7′a

109.9

109.9

7.40, 1H, m

3′a, 5′, 6′

7′a

136.6

136.9

136.9

1″

47.1

4.14, 2H, t, J = 7.2 Hz

2′, 7′a, 2″, 3″

47.0

47.1

4.08, 2H, t, J = 7.3 Hz, overlapped

2′, 7′a, 2″, 3″

2″

29.5

1.87, 2H, q, J = 7.2 Hz

1″, 3″, 4″

29.4

29.5

1.82, 2H, q, J = 7.3 Hz

1″, 3″, 4″

3″

29.0

1.30, 2H, m. overlapped

1″, 2″, 4″, 5″

28.8

28.9

1.25, 2H, m, overlapped

1″, 2″, 4″

4″

22.2

1.35, 2H, m, overlapped

3″, 5″

22.1

22.2

1.31, 2H, m, overlapped

2″, 3″, 5″

5″

13.9

0.90, 3H, t, J = 7.2 Hz

3″, 4″

13.8

13.9

0.86, 3H, t, J = 7.0 Hz

3″, 4″

1″′

137.0

125.5/125.6

125.7

2″′

134.6

127.8

127.8

7.66, 1H, d, J = 7.9 Hz

1, 1″′, 3″′, 8″′a

3″′

130.3

7.19, 1H, m, overlapped

1″′, 4″′, Me

102.1

102.1

6.84, 1H, d, J = 7.9 Hz

1″′, 4″′, 4″′a

4″′

126.0

7.16, 1H, m, overlapped

2″′, 3″′, 6″′

156.9

157.0

4″′a

131.3

131.4

5″′

126.9

7.18, 1H, m, overlapped

1″′, 3″′, 6″′

122.0

122.0

8.34, 1H, m

4″′, 6″′, 8″′a

6″′

130.3

7.22, 1H, m, overlapped

2, 1″′, 2″′, 5″′

127.2

127.4

7.52, 1H, m, overlapped

5″′, 8″′

7″′

125.5/125.6

125.7

7.50, 1H, m, overlapped

5″′, 8″′, 8″′a

8″′

125.7

125.8

8.30, 1H, m

1″′, 4″′a, 7″′

8″′a

132.1

132.1

Me

20.0

2.33, 3H, s

1″′, 2″′, 3″′

OMe

55.6

55.7

4.08, 3H, s, overlapped

4″′

aRecorded in CDCl3 at 600 MHz (1H) and 150 MHz (13C), respectively; data in δ ppm (J in Hz)

bRecorded in CDCl3 at 300 MHz (1H) and 75.5 MHz (13C), respectively; data in δ ppm (J in Hz) for JWH-081 [22]

cJ = 8 Hz; the proton signal correlated with the indicated carbons

Fig. 4

Double quantum filtered correlation spectroscopy (DQF-COSY), heteronuclear multiple-bond correlation (HMBC) and selected rotating frame nuclear Overhauser effect (ROE) correlations of JWH-251 (1) and JWH-081 (3)

An unknown peak 2 detected at 49.1 min in Fig. 2a showed a specific major GC–MS signal at m/z 214 (Fig. 2c), which was also observed in the mass spectrum of compound 1 (Fig. 2b), with a putative molecular ion signal at m/z 335. In the LC-PDA and LC–MS chromatograms as shown in Fig. 3a, e, the corresponding peak 2 at 8.0 min showed a protonated molecular ion signal at m/z 336 and a UV spectrum very similar to that of compound 1 (Fig. 3c, g). Using the isolated preparation of peak 2, we measured its NMR spectra; the 1H and 13C NMR spectra of 2 showed one methoxy signal at δH 3.83 (3H, s) and δC 55.4 (data not shown). These data suggest that compound 2 has one methoxy group in place of a methyl group of compound 1. Therefore, we purchased authentic JWH-250 and analyzed it to compare the data with those of unknown peak 2. The GC–MS and LC–MS data (Figs. 2e, 3i) combined with the NMR results revealed that compound 2 was JWH-250, namely 2-(2-methoxyphenyl)-1-(1-pentyl-1H-indol-3-yl)ethanone. Compound 2 was also reported to be a potent cannabinoid receptor agonist, along with JWH-251 (1) [20]. In 2009, JWH-250 (2) was first detected as an adulterant in herbal products available on the UK and German markets [18].

The third unknown peak 3 detected at 53.7 min in Fig. 2a showed five major ion signals at m/z (% relative intensity) 371 (100), 314 (43), 214 (40), 185 (27), and 157 (10) by GC–MS analysis (Fig. 2d). In the LC-PDA analysis, the corresponding peak was detected at 10.6 min as shown in Fig. 3a; the peak showed a protonated ion signal at m/z 372 [M + H]+ with an absorbance maximum at 316 nm in the UV spectrum, as shown in Fig. 3d, h. Although these characteristics were completely different from those of compounds 1 and 2 (Figs. 2b, c, 3f, g), they were similar to those of naphthoylindoles such as JWH-018 and JWH-073 (4) (Fig. 5c) [2, 4]. The accurate mass spectrum measured by DART–TOF–MS in the positive mode showed the protonated ion peak at m/z 372.19494, suggesting that the molecular formula of compound 3 was C25H26NO2. The error between the observed and theoretical mass numbers was −1.41 mDa. The 1H NMR spectrum of compound 3 showed 25 nonexchangeable protons, including one methyl signal at δ 0.86 (3H, t, J = 7.0 Hz), one methoxy signal at δ 4.08 (3H, s), nine aromatic proton signals, AB-type aromatic proton signals, and one methylene proton signal connected to a nitrogen atom with three methylene proton signals as shown in Table 1. The 13C NMR spectrum of compound 3 showed 25 carbon signals, containing 1 methyl signal, 1 methoxy signal (δ 55.7), 4 methylenes with a nitrogenated carbon (δ 47.1), 11 aromatic carbons, 7 aromatic quaternary carbons and a carbonyl carbon (δ 191.8). Following the two-dimensional NMR analyses shown in Table 1 and Fig. 4, compound 3 was identified as an aminoalkyl naphthoylindole. On the basis of the above mass and NMR data, the structure of compound 3 was deduced to be 1-(4-methoxynaphthalenyl)-(1-pentyl-1H-indol-3-yl)methanone. This compound has been reported as JWH-081 by Huffman et al. [22] and as a cannabinoid receptor agonist [14]. The 13C NMR spectral data reported for JWH-081 [22] were identical to those of compound 3 (Table 1).
Fig. 5

GC-MS analysis for product No. 19. TIC (a), and EI mass spectra of peaks detected at 50.5 min (b, 5), and 50.9 min (c, 4) and of the peak of the standard of JWH-015 (d, RT 50.5 min)

Identification of unknown peaks 46

An unknown peak 5 was detected with two other peaks 2 (JWH-250) and 4 (JWH-073) in the GC–MS and LC–MS chromatograms of product No. 19 (Figs. 5, 6). In the GC–MS chromatogram, the unknown peak 5 at 50.5 min in Fig. 5a showed five major fragment signals at m/z (% relative intensity) 310 (38), 270 (31), 200 (29), 155 (23), and 127 (38), with a putative molecular ion signal at m/z 327 (100) as shown in Fig. 5b. In the LC–MS chromatogram, the corresponding peak found at 7.6 min in Fig. 6a showed a protonated ion signal at m/z 328 [M + H]+ and an absorbance maximum at 318 nm in the UV spectrum as shown in Fig. 6b, e. Although the putative molecular weight of compound 5 was the same as that of compound 4, their characteristics differed (Figs. 5a–c, 6a, b, e). At the present time, several types of synthetic cannabinoids are available as commercial reagents. In order to rapidly identify compounds that are illegally added as designer drugs, we purchased a number of these compounds. Two of the purchased compounds were JWH-073 (1-naphthalenyl(1-butyl-1H-indol-3-yl)methanone) and JWH-015 (1-naphthalenyl(2-methyl-1-propyl-1H-indol-3-yl)methanone), both of which have a molecular weight of 327. Therefore, we measured their mass spectra by GC–MS and LC–MS. The compounds for peaks 4 and 5 were completely identical to JWH-073 and JWH-015, respectively (Figs. 5, 6). It is of interest that compound 5 was reported as a selective CB2 receptor ligand, but compounds 14 were CB1 receptor agonists [14].
Fig. 6

LC-UV detection (a) and LC–MS analysis for the extract of product No. 19. Mass chromatograms at m/z 328 (b, 4 and 5) and 336 (c, 2), the TIC (d), and UV spectra and ESI mass spectra for peak 5 (e) and for the standard of JWH-015 (f)

As shown in Figs. 7 and 8, an unknown peak 6 was detected together with peak 2 (JWH-250) in the GC–MS and LC–MS chromatograms of product No. 20. By GC–MS, the unknown peak 6 at 55.9 min in Fig. 7 showed five major EI-MS signals at m/z (% relative intensity) 384 (6), 207 (10), 155 (4), 127 (7), and 100 (100). In the LC–MS chromatogram, the corresponding peak appearing at 4.8 min in Fig. 8a showed a protonated ion signal at m/z 385 [M + H]+ and an absorbance maximum at 312 nm in the UV spectrum as shown in Fig. 8b, e. As in the case of compounds 4 and 5, both GC–MS and LC–MS spectra of the authentic (purchased) JWH-200, the molecular weight of which was 384, were measured (Figs. 7c, 8f); compound 6 was found to be identical to JWH-200 (1-naphthalenyl(1-(2-(4-morpholinyl)ethyl)-1H-indol-3-yl)methanone). Compound 6 has been reported as a synthetic cannabinoid possessing high affinity for CB1 receptors and exhibiting Δ9-THC-like pharmacological effects [16, 23].
Fig. 7

GC-MS analysis for product No. 20. TIC (a), EI mass spectra of the peaks detected at 55.9 min (b, 6) and of the peak of the standard of JWH-200 (RT 55.9 min, c)

Fig. 8

LC-UV detection (a) and LC–MS analysis for the extract of product No. 20. Mass chromatograms at m/z 385 (b, 6 and 5) and 336 (c, 2), TIC (d), UV spectra and ESI mass spectra of peak 6 (e), and those of the standard of JWH-200 (f). The analysis was performed under the gradient program 2

Quantitation of the cannabimimetic compounds

As shown in Table 2, the calibration curves for LC-UV detection for solutions prepared by dilution of each standard solution using the gradient program 1 were linear over the concentration range 10–500 μg/ml (with calibration points at six different concentrations) (14) or 100–250 μg/ml (five concentrations) (5) with good correlation coefficients of r2 ≥ 0.993. Intraassay precision and accuracy were also measured for the diluted standard solutions. The precision of the compounds ranged from 0.1% to 4.9%, and the accuracy ranged from −2.6% to 1.7% (Table 2). The calibration curves for compounds 2 and 6 obtained with the gradient program 2 were also linear over the concentration range 10–250 μg/ml (five concentrations) with good correlation coefficients of r2 ≥ 0.992 (Table 3). The precision of the two compounds ranged from 1.2% to 3.9%, and the accuracy ranged from −2.4% to 2.1%.
Table 2

Linearity, precision, and accuracy obtained by the LC-UV analysis under gradient program 1 for cannabimimetic compounds 15

Compound

Linear range (μg/ml)

Linearity

Concentration (μg/ml)

Precision (%)

Accuracy (%)

JWH-251 (1)

10–500

y = 0.033x − 0.1838

r2 = 0.9985

10

4.3

−1.8

50

0.6

0.4

500

0.3

1.0

JWH-250 (2)

10–500

y = 0.0122x − 0.0185

r2 = 0.9993

10

4.9

−1.9

50

2.6

−2.6

500

0.4

1.1

JWH-081 (3)

10–500

y = 0.0147x − 0.1045

r2 = 0.9934

10

1.3

−0.0

50

2.0

−1.3

500

2.3

−0.2

JWH-073 (4)

10–500

y = 0.0149x + 0.094

r2 = 0.9931

10

1.5

−2.3

50

0.5

−0.4

500

0.2

1.7

JWH-015 (5)

10–250

y = 0.0152x + 0.0383

r2 = 0.9984

10

0.1

0.2

50

0.1

0.2

250

0.6

1.2

Table 3

Linearity, precision, and accuracy obtained by LC-UV analysis under gradient program 2 for cannabimimetic compounds 2 and 6

Compound

Linear range (μg/ml)

Linearity

Concentration (μg/ml)

Precision (%)

Accuracy (%)

JWH-250 (2)

10–250

y = 0.0137x − 0.0837

r2 = 0.9926

10

2.7

−1.5

50

1.7

2.1

250

3.9

−0.1

JWH-200 (6)

10–250

y = 0.013x − 0.0788

r2 = 0.9921

10

3.3

−1.0

50

1.2

−2.4

250

2.1

−0.3

Eighteen herbal products and 2 powder products currently being sold in Japan for their expected cannabis-like effects were purchased via the Internet (Table 4). The concentrations of compounds 16 in the 20 products were measured using calibration curves (Tables 2, 3). The results are summarized in Table 4. The concentrations of compounds 14 in these products were in the ranges of 3.65–18.9, 16.9–340, 27.6–83.6, and 24.7–107 mg/g, respectively. The concentrations of compounds 5 and 6 in the products were 9.56 and 21.7 mg/g, respectively (Table 4). These results indicate that there is great variation in the concentrations of illegal compounds added to these products. Two powder products (Nos. 8 and 9) and two resin products (Nos. 2 and 3) contained high contents of JWH-250 and JWH-073, respectively. In 9 out of 16 dried leaf (cutting) products, multiple drugs could be detected (Table 4); there are possibilities that a herbal material was adulterated with multiple drugs and that multiple herbal materials, each adulterated with a single or multiple adulterants, were combined together.
Table 4

Concentrations of detected cannabimimetic compounds in the tested products

Product

No.

Form

JWH-251 (1) (mg/g)

JWH-250 (2) (mg/g)

JWH-081 (3) (mg/g)

JWH-073 (4) (mg/g)

JWH-015 (5) (mg/g)

JWH-200 (6) (mg/g)

1

Dried leaf (cutting)

n.d.

n.d.

n.d.

37.3 ± 3.25

n.d.

n.d.

2

Solid (resin)

n.d.

n.d.

n.d.

91.5 ± 10.9

n.d.

n.d.

3

Solid (resin)

n.d.

n.d.

n.d.

107 ± 5.33

n.d.

n.d.

4

Dried leaf (cutting)

n.d.

n.d.

n.d.

79.8 ± 3.98

n.d.

n.d.

5

Dried leaf (cutting)

n.d.

25.9 ± 0.64

n.d.

n.d.

n.d.

n.d.

6

Dried leaf (cutting)

n.d.

51.5 ± 2.88

n.d.

n.d.

n.d.

n.d.

7

Dried leaf (cutting)

n.d.

16.9 ± 1.79

83.6 ± 0.94

n.d.

n.d.

n.d.

8

Powder

n.d.

321 ± 8.97

n.d.

n.d.

n.d.

n.d.

9

Powder

n.d.

340 ± 28.0

n.d.

n.d.

n.d.

n.d.

10

Dried leaf (cutting)

18.9 ± 4.64

n.d.

32.1 ± 7.49

n.d.

n.d.

n.d.

11

Dried leaf (cutting)

3.65 ± 0.81

41.7 ± 8.07

34.8 ± 7.18

n.d.

n.d.

n.d.

12

Dried leaf (cutting)

n.d.

30.5 ± 0.55

40.1 ± 1.77

29.6 ± 0.78

n.d.

n.d.

13

Dried leaf (cutting)

n.d.

n.d.

47.4 ± 7.51

n.d.

n.d.

n.d.

14

Dried leaf (cutting)

n.d.

n.d.

27.6 ± 4.08

26.7 ± 3.39

n.d.

n.d.

15

Dried leaf (cutting)

n.d.

n.d.

44.1 ± 6.72

n.d.

n.d.

n.d.

16

Dried leaf (cutting)

n.d.

n.d.

30.5 ± 7.76

24.7 ± 5.56

n.d.

n.d.

17

Dried leaf (cutting)

n.d.

n.d.

33.1 ± 3.10

27.6 ± 2.24

n.d.

n.d.

18

Dried leaf (cutting)

n.d.

n.d.

n.d.

88.3 ± 7.37

n.d.

n.d.

19

Dried leaf (cutting)

n.d.

23.6 ± 1.52

n.d.

45.2 ± 4.42

9.56 ± 0.55

n.d.

20

Dried leaf (cutting)

n.d.

41.7 ± 0.75

n.d.

n.d.

n.d.

21.7 ± 0.47

Data are given as the mean ± standard deviation, n = 3

n.d. not detected

Conclusions

In the present study, we identified six synthetic cannabinoids in herbal or chemical products collected via the Internet from December 2009 to April 2010. Among the six compounds, four compounds (1, 3, 5 and, 6) are new as adulterants to our knowledge. Although “Spice” and other similar herbal products are known to have been sold on the Internet since as early as 2006 [18], the synthetic cannabinoids were first reported as adulterants in these herbal products in early 2009 [1, 2, 6]. CCH, CP 47497, and JWH-018 were most frequently detected as adulterants in the products available in Japan from June 2008 to June 2009 [4]. However, new synthetic cannabinoids, especially indole derivatives belonging to the JWH series of cannabinoids, began to appear shortly thereafter. The compounds 16 detected in this study had been synthesized as cannabimimetic substances and were reported to have affinity actions on cannabinoid receptors. In addition, compounds 36 were reported to have in vivo pharmacological effects [13, 16, 23, 24, 25]. Compounds 14 showed affinity for CB1 receptors three- to tenfold higher than for CB2 receptors unlike JWH-018 [14, 20], whereas JWH-015 (6) showed affinity for the CB2 receptor that was 24-fold that for the CB1 receptor [14]. In the present study, it should be pointed out that product No. 19 contained not only CB1 cannabinoid receptor agonists (compounds 2 and 4) but also one CB2 receptor agonist (compound 5). This is the first case in which several synthetic cannabinoids possessing different types of activity were detected in a single product.

There is little information on pharmacology, toxicology, and safety of the cannabimimetic adulterants for humans, and there are possibilities of serious health damage for their abusers. For rapid identification of various new synthetic cannabinoids included as designer drugs in illegal products, we are collecting the GC–MS and LC–MS data of many cannabimimetic compounds to construct a database. In view of the worldwide trend to adulterating herbal or chemical products with designer drugs, an international system of cooperation to share the analytical information of such compounds is needed to prevent their worldwide spread.

Notes

Acknowledgments

Part of this work was supported by a Health and Labor Sciences Research Grant from the Ministry of Health, Labour, and Welfare, Japan.

References

  1. 1.
    Uchiyama N, Kikura-Hanajiri R, Kawahara N, Haishima Y, Goda Y (2009) Identification of a cannabinoid analog as a new type of designer drug in a herbal product. Chem Pharm Bull 57:439–441CrossRefPubMedGoogle Scholar
  2. 2.
    Uchiyama N, Kikura-Hanajiri R, Kawahara N, Goda Y (2009) Identification of a cannabimimetic indole as a designer drug in a herbal product. Forensic Toxicol 27:61–66CrossRefGoogle Scholar
  3. 3.
    Kikura-Hanajiri R, Kawamura M, Maruyama T, Kitajima M, Takayama H, Goda Y (2009) Simultaneous analysis of mitragynine, 7-hydroxymitragynine, and other alkaloids in the psychotropic plant “kratom” (Mitragyna speciosa) by LC-ESI-MS. Forensic Toxicol 27:67–74CrossRefGoogle Scholar
  4. 4.
    Uchiyama N, Kikura-Hanajiri R, Ogata J, Goda Y (2010) Chemical analysis of synthetic cannabinoids as designer drugs in herbal products. Forensic Sci Int 198:31–38CrossRefPubMedGoogle Scholar
  5. 5.
    Kikuchi H, Uchiyama N, Ogata J, Kikura-Hanajiri R, Goda Y (2010) Chemical constituents and DNA sequence analysis of a psychotropic herbal product. Forensic Toxicol 28:77–83CrossRefGoogle Scholar
  6. 6.
    Auwärter V, Dresen S, Weinmann W, Müller M, Pütz M, Ferreirós N (2009) ‘Spice’ and other herbal blends: harmless incense or cannabinoid designer drugs? J Mass Spectrom 44:832–837CrossRefPubMedGoogle Scholar
  7. 7.
    Lindigkeit R, Boehme A, Eiserloh I, Luebbecke M, Wiggermann M, Ernst L, Beuerle T (2009) Spice: a never ending story? Forensic Sci Int 191:58–63CrossRefPubMedGoogle Scholar
  8. 8.
    Compton DR, Rice KC, De Costa BR, Razdan RK, Melvin LS, Johnson MR, Martin BR (1993) Cannabinoid structure–activity relationships: correlation of receptor binding and in vivo activities. J Pharmacol Exp Ther 265:218–226PubMedGoogle Scholar
  9. 9.
    Compton DR, Johnson MR, Melvin LS, Martin BR (1992) Pharmacological profile of a series of bicyclic cannabinoid analogs: classification as cannabimimetic agents. J Pharmacol Exp Ther 260:201–209PubMedGoogle Scholar
  10. 10.
    Melvin LS, Milne GM, Johnson MR, Subramaniam B, Wilken GH, Howlett AC (1993) Structure–activity relationships for cannabinoid receptor-binding and analgesic activity: studies of bicyclic cannabinoid analogs. Mol Pharmacol 44:1008–1015PubMedGoogle Scholar
  11. 11.
    Thomas BF, Compton DR, Martin BR (1990) Characterization of the lipophilicity of natural and synthetic analogs of delta 9-tetrahydrocannabinol and its relationship to pharmacological potency. J Pharmacol Exp Ther 255:624–630PubMedGoogle Scholar
  12. 12.
    Martin BR, Compton DR, Thomas BF, Prescott WR, Little PJ, Razdan RK, Johnson MR, Melvin LS, Mechoulam R, Ward SJ (1991) Behavioral, biochemical, and molecular modeling evaluations of cannabinoid analogs. Pharmacol Biochem Behav 40:471–478CrossRefPubMedGoogle Scholar
  13. 13.
    Wiley JL, Compton DR, Dai D, Lainton JA, Phillips M, Huffman JW, Martin BR (1998) Structure–activity relationships of indole- and pyrrole-derived cannabinoids. J Pharmacol Exp Ther 285:995–1004PubMedGoogle Scholar
  14. 14.
    Aung MM, Griffin G, Huffman JW, Wu M, Keel C, Yang B, Showalter VM, Abood ME, Martin BR (2000) Influence of the N-1 alkyl chain length of cannabimimetic indoles upon CB1 and CB2 receptor binding. Drug Alcohol Depend 60:133–140CrossRefPubMedGoogle Scholar
  15. 15.
    Huffman JW (1999) Cannabimimetic indoles, pyrroles and indenes. Curr Med Chem 6:705–720PubMedGoogle Scholar
  16. 16.
    Huffman JW, Mabon R, Wu MJ, Lu J, Hart R, Hurst DP, Reggio PH, Wiley JL, Martin BR (2003) 3-Indolyl-1-naphthylmethanes: new cannabimimetic indoles provide evidence for aromatic stacking interactions with the CB1 cannabinoid receptor. Bioorg Med Chem 11:539–549CrossRefPubMedGoogle Scholar
  17. 17.
    Zweiundzwanzigste Verordnung, zur Änderung betäubungsmittelrechtlicher Vorschriften (2009) Germany. BGBl I Nr. 3 vom 21.01.2009, 22. BtMÄndV vom 19. January 2009, S. 49–50. http://www.bgblportal.de/BGBL/bgbl1f/bgbl109s0049.pdf. Accessed January 2009
  18. 18.
    EMCDDA (2009) EMCDDA action on new drugs briefing paper: understanding the ‘Spice’ phenomenon (a report from an EMCDDA expert meeting, 6 March 2009, Lisbon). http://www.emcdda.europa.eu/attachements.cfm/att_80086_EN_Spice%20Thematic%20paper%20—%20final%20version.pdf. Accessed April 2010
  19. 19.
    Huffman JW (2009) Cannabimimetic indoles, pyrroles, and indenes: structure–activity relationships and receptor interactions. In: Reggio PH (ed) The cannabinoid receptors. Humana Press, New York, pp 49–94CrossRefGoogle Scholar
  20. 20.
    Huffman JW, Szklennik PV, Almond A, Bushell K, Selley DE, He H, Cassidy MP, Wiley JL, Martin BR (2005) 1-Pentyl-3-phenylacetylindoles, a new class of cannabimimetic indoles. Bioorg Med Chem Lett 15:4110–4113CrossRefPubMedGoogle Scholar
  21. 21.
    Kikura-Hanajiri R, Kawamura M, Uchiyama N, Ogata J, Kamakura H, Saisho K, Goda Y (2008) Analytical data of designated substances (Shitei-Yakubutsu) controlled by the Pharmaceutical Affairs Law in Japan, part I: GC-MS and LC-MS. Yakugaku Zasshi 128:971–979CrossRefPubMedGoogle Scholar
  22. 22.
    Huffman JW, Wu M, Lu J (1998) A very facile SNAr reaction with elimination of methoxide. J Org Chem 63:4510–4514CrossRefGoogle Scholar
  23. 23.
    Compton DR, Gold LH, Ward SJ, Balster RL, Martin BR (1992) Aminoalkylindole analogs: cannabimimetic activity of a class of compounds structurally distinct from delta 9-tetrahydrocannabinol. J Pharmacol Exp Ther 263:1118–1126PubMedGoogle Scholar
  24. 24.
    Showalter VM, Compton DR, Martin BR, Abood ME (1996) Evaluation of binding in a transfected cell line expressing a peripheral cannabinoid receptor (CB2): identification of cannabinoid receptor subtype selective ligands. J Pharmacol Exp Ther 278:989–999PubMedGoogle Scholar
  25. 25.
    Huffman JW, Padgett LW (2005) Recent developments in the medicinal chemistry of cannabimimetic indoles, pyrroles and indenes. Curr Med Chem 121:1395–1411CrossRefGoogle Scholar

Copyright information

© Japanese Association of Forensic Toxicology and Springer 2010

Authors and Affiliations

  • Nahoko Uchiyama
    • 1
  • Maiko Kawamura
    • 1
  • Ruri Kikura-Hanajiri
    • 1
  • Yukihiro Goda
    • 1
  1. 1.National Institute of Health SciencesTokyoJapan

Personalised recommendations