The determination of α- and β-thujone in human serum – Simple analysis of absinthe congener substance

Forensic Science International 259 (2016) 188–192

Contents lists available at ScienceDirect

Forensic Science International journal homepage: www.elsevier.com/locate/forsciint

The determination of a- and b-thujone in human serum – Simple analysis of absinthe congener substance Michal P. Dybowski *, Andrzej L. Dawidowicz Faculty of Chemistry, Maria Curie Sklodowska University, Pl. Marii Curie Sklodowskiej 3, 20-031 Lublin, Poland

A R T I C L E I N F O

A B S T R A C T

Article history: Received 7 September 2015 Received in revised form 27 November 2015 Accepted 11 December 2015 Available online 19 December 2015

Absinthe is a strong spirit beverage, mostly green in color, containing besides ethyl alcohol (main component), alcoholic macerate of wormwood and other plants such as star anise and fennel seed. Due to the potential risks associated with the presence of a- and b-thujone many countries have implemented strict rules limiting the content of these congeners in alcohol products. The presented paper describes a simple and sensitive method for the determination of a- and b-thujone in human serum using Solid Phase Extraction as a sample preparation method combined with GC/MS analysis. The procedure involves the protein precipitation process, which generally degrades the protein–analyte complex, and SPE isolation of thujone from the examined materials. The described method is characterized by a low LOD and a very high recovery of the analytes. The present method for the estimation of a- and b-thujone concentration in human fluids after the consumption of alcoholic beverages and other foods containing the substance is applicable in forensic and clinical toxicology because of its simplicity and rapidness with high sensitivity. ß 2015 Elsevier Ireland Ltd. All rights reserved.

Keywords: a- and b-Thujone Human serum SPE Alcohol congener analysis Gas chromatography

1. Introduction Absinthe is a strong spirit beverage, a mostly green colored, containing besides alcohol (main component), alcoholic macerate of wormwood (Artemisia absinthium L.) and other plants such as star anise (Illicium verum L.), fennel seed (Feoniculum vulgare L.) and hyssop (Hyssopus officinalis L.) [1,2]. In the late 19th century, absinthe, in the meantime called ‘‘green fairy’’ (‘‘fe´e verte’’), was the most popular spirit drink in Europe [3,4], achieving great recognition among artist and writers. Primarily due to its biomedical active chemical, thujone, the primary constituent of essential oils derived from a variety of plants including wormwood. Thujone, a monoterpenoid ketone, occurs in nature as a mixture of a-( )- and b-(+)-diastereoisomers [5,6]. In animal studies, thujone inhibits the gammaaminobutyric acid-A (GABAA) receptor causing excitation and convulsion in a dose-dependent manner [7– 9]. The toxicology of absinthe in humans is reviewed with regard to the cause of a syndrome called ‘‘absinthism’’, which was described, in the 19th century, after chronic abuse of the thujone containing spirit [10]. The fear of absinthism in connection with the medical

* Corresponding author. Tel.: +48 81 537 55 69; fax: +48 81 533 33 48. E-mail address: [email protected] (M.P. Dybowski). http://dx.doi.org/10.1016/j.forsciint.2015.12.015 0379-0738/ß 2015 Elsevier Ireland Ltd. All rights reserved.

use of wormwood is also still evinced by the recently published monograph of European Medicines Agency (EMEA) [11]. EMEA has proposed maximum daily intakes for wormwood based on the presence of its constituent thujone on 3 mg/person for maximum duration of 2 weeks. Studies demonstrate that the a-( )-thujone isomer has a greater ability to elicit seizures than the b-(+)-isomer and it is more likely that the convulsive effect of thujone involves a specific receptor system [12]. It has been also proposed that athujone might act through the activation of CB1 cannabinoid receptors [13], but subsequent studies have shown, that although a-thujone binds to this receptor, it is unable to activate it [14]. Due to the potential risks associated with the presence of a- and bthujone, an absinthe congener substance (when wormwood extract is used as an absinthe ingredient), many countries have implemented strict rules limiting their content in alcohol products. The term ‘congener substances’ was introduced by Machata and Prokop [15] and summarizes all other compounds in alcoholic beverage, other than water and ethanol [16–19] that assist in the distinctive aroma, flavor and appearance of the beverage [16,20]. An example of such substances might be: 1-propanol, 1-butanol, iso-butanol, isoamyl alcohol [21], anethole in anised alcohols [22,23], limonene in citrus flavor beverages [24] or thujone in absinthe [25], etc. Although many methods of a- and bthujone estimation in complex matrices (e.g. in food products) is described in the literature [25–33], there is very few of reports

M.P. Dybowski, A.L. Dawidowicz / Forensic Science International 259 (2016) 188–192

concerning the determination of these substances in human plasma [34]. The presented paper describes a simple and sensitive method for determination of a- and b-thujone in human serum using Solid Phase Extraction (SPE) as a sample preparation method combined with gas chromatography mass spectrometry (GC/MS) analysis. This analytical approach is proposed as a method for the estimation of thujones concentration in human fluids after the consumption of alcoholic beverages and other foods containing these substances. 2. Experimental

189

plasma sample in 4 mL glass. Then, after addition of internal standard solution (10 mL), whole mixture was vortexed for 5 min. Prepared sample was centrifuged for 15 min at 950  g. After centrifugation 1 mL of supernatant was transferred to 1.5 mL of water and mixed. Two milliliters of the obtained solution was placed into SPE cartridge filled with 0.2 g of SepraC18-E. Cartridge was washed with 5 mL of n-hexane before loading and vacuumdrying for 5 min. The remaining components (after SPE bed drying) were eluted into a calibrated 2 mL flask using n-hexane and analyzed by GC. The SPE-12G (J.T. Baker, USA) vacuum chamber was used to carry out the experiment. The eluent velocity was set on 1 drop s 1.

2.1. Materials and chemicals Cyclodecanone as an internal standard (IS) was obtained from Sigma–Aldrich (Steinheim, Germany), a- and b-thujone mixture was supplied by Fluka (Buchs, Switzerland), a-thujone standard was supplied by Fluka (Buchs, Switzerland). Ethanol and n-hexane were obtained from Chemical Plant POCH S.A. (Gliwice, Poland). All of the reagents were of the highest purity available. The Sepra C18E sorbent used for SPE procedure (50 mm, 65 A˚) was supplied by Phenomenex (Torrance, CA, USA). Blood samples both for validation procedure and samples containing a- and b-thujones were supplied by volunteers. Water was purified on a Milli-Q system from Millipore (Millipore, Bedford, MA, USA). 2.2. Blood samples – collecting and storage All blood samples (5 mL each) were collected from volunteers by a qualified person using Sarstedt Monovette single closed system containing a coagulation activator. Sample collection was carried out twice – on an empty stomach (reference/blank samples) and 1 h after consuming of 20 g (15 mg/g) a- and bthujone alcoholic solution (which corresponds to approximately 15 mL of alcoholic solution). The blood samples were than subjected to centrifugation to obtain plasma samples and stored at 5 8C (3 8C) until SPE procedure. 2.3. Estimation of a- and b-thujone free form fractions in thujonehuman serum albumin system In order to estimate the degree of a- and b-thujone binding with human serum protein, individual portions of plasma samples were spiked with appropriate volumes of thujones standard to obtain 100 and 150 mg/g total concentration of a- and b-thujone, respectively. The unbound a- and b-thujone fraction were isolated by ultrafiltration on Amicon MPS (Millipore, Bedford, MA, USA) units, utilizing the YM-10 membranes (product no. 40424, Millipore, Bedford, MA, USA) of 10 kDa molecular mass cutoff. The ultrafiltration process was performed in a thermostatic centrifuge (MPW-350-RH, MPW Med. Instruments, Warsaw, Poland) at 15 8C. One milliliter of each solution was put into a sample compartment of the ultrafiltration unit. After the attachment of an ultrafiltrate collection container, the unit was centrifuged at 2300 rpm (ca. 800  g) until 400 mL of ultrafiltrate was obtained (for 25 min). To each ultrafiltrate sample (300 mL) hexane (300 mL) was added. The mixtures were vigorously shaken for 10 min at 200 rpm. After centrifugation (3000 rpm  1100  g for 5 min) hexane layer was transferred to a clean tube and subjected to GC analysis.

2.4.2. a- and b-thujone isolation by liquid–liquid extraction procedure Hexane (1 mL) was added to the plasma sample (1 mL) and vigorously shaken for 10 min at 2000 rpm. After phase separation by centrifugation the hexane layer was subjected to GC analysis. 2.5. Chromatographic analysis 2.5.1. GC–MS conditions The GC–MS system used for identification of a- and b-thujone was a GC/MS QP2010 series with a mass selective detector (Shimadzu, Kyoto, Japan). ZB5-MSi fused-silica capillary column (30 m  0.25 mm i.d., 0.25 mm film thickness; Phenomenex, USA) was used. Helium (grade 5.0) was used as carrier gas. One microliter of the sample was injected by an AOC-20i type autosampler (Shimadzu, Kyoto, Japan). The injector temperature was 310 8C. The following temperature program was applied: 1 min at 50 8C subsequently a linear temperature increase up to 110 8C at the rate of 6 8C/min, eventually reach a temperature 285 8C at the rate of 25 8C/min. The mass spectrometer was operated in electron ionization (EI) mode at 70 eV. The mass spectra were measured in the range 35– 360 amu. Qualitative analysis was carried out comparing the retention indices and MS spectra for the obtained peaks with the analogous data from NIST’14 and Adams (4th ed.) databases. 2.5.2. GC-FID conditions For quantification of a- and b-thujone from SPE extracts GC 2010 series gas chromatograph with flame ionization detector was used (Shimadzu, Kyoto, Japan). 1 mL of the sample by AOC-20i type autosampler into a ZB5-MSi fused-silica capillary column (30 m  0.25 mm i.d., 0.25 mm film thickness) (Phenomenex, USA). The temperature program during GC-FID separation was the same as for GC/MS. Due to the lack of commercially available b-thujone standard, the concentration of a- and b-thujone in a standard mixture was evaluated in relation to analytical standard of a-thujone. The applied GC-FID equipment was calibrated using a- and b-thujone standard mixture solutions and the internal standard. The working solutions were obtained by serial dilutions of the stock solution with hexane to obtain the following concentration of analytes (the number of analytical procedure repetitions – n = 5):  a-thujone: 10 ng/g, 100 ng/g, 1 mg/g, 10 mg/g, 25 mg/g, 50 mg/g, 75 mg/g, 100 mg/g, 250 mg/g.  b-thujone: 10 ng/g, 100 ng/g, 1 mg/g, 10 mg/g, 25 mg/g, 50 mg/g.

2.4. Sample preparation procedure 2.4.1. SPE procedure for isolation of a- and b-thujone To eliminate eventual loss of analytes caused by its binding with plasma proteins, 1.5 mL of ethanol was added to 1 mL of

a- and b-thujone peaks identification was carried out by comparing the GC retention index with this from GC/MS and with the retention data for analytical standards. The retention indices of a- and b-thujone were determined under the standard method

190

M.P. Dybowski, A.L. Dawidowicz / Forensic Science International 259 (2016) 188–192

conditions for the target compound using a series of n-alkane standards. 2.6. Statistical analysis of validation parameters Specificity of the analytical procedure was evaluated by analyzing blank whole blood samples from five different volunteers. To provide accuracy of presented SPE procedure, the recoveries were calculated as percent or the true value for a- and b-thujone. The internal standard multipoint calibration technique was used for calibration of chromatographic system. The quantification of a- and b-thujone were establish using the ratio of the analyte and cyclodecanone (the internal standard) response. The method linearity was determine injecting standard solutions of different concentration to the chromatographic system. The detection limit (LOD) and quantitation limit (LOQ) were define as the height values with signal-to-noise ratios of 3 and 10, respectively using extracts from plasma samples spiked with aand b-thujone analytical standard. The estimation of signal-tonoise ratio, for LODs and LOQs determination, was evaluated comparing signals from samples of known low analyte concentrations with blank samples. The repeatability and trueness of the SPE step were evaluated by five-time repetition of the applied procedure for the analysis of samples differing in analyte concentration spanning the range of calibration curve. The intra- and interday precision were evaluated by analyzing five samples tested compounds on the same day and on different days (on the next day and the day after tomorrow). In order to determine there is a significant difference between recoveries depending on the analytes concentration level (samples prepared according to 2.5. paragraph) the ANOVA (analysis of variance) was performed. Two types of variances were compared: the variance within each sample and the variance between different samples. The significance of difference was expressed by p- and F-values.

3. Results and discussion According to performed studies a- and b-thujone binds with plasma proteins, creating a very strong protein–thujone complexes. The percentage of thujone bonded with serum proteins exceeds 90%. It should be noted, however, that strong binding of the analyzed substance with plasma proteins can cause difficulties with precise analyte determination. For this reason, the presented SPE procedure involves protein–thujone complex cleavage before their chromatographic analysis. Fig. 1 presents chromatograms: of a-, b-thujone and cyclodecanone analytical standards (Fig. 1A); SPE extracts from reference plasma sample (Fig. 1B), and plasma sample spiked with alcoholic solution of a-, b-thujone mixture (Fig. 1C) – samples obtained according to the presented SPE procedure. Fig. 1D represents chromatogram of LLE extract from blank/reference plasma sample. Fig. 1E depicts the additional chromatogram of SPE extract from plasma of a volunteer who consumed a- and b-thujone alcoholic solution. As appears from the figure, the applied chromatographic conditions allow for a sufficient resolution of those analytes from sample matrix peaks, both in the SPE and the LLE procedures. However, the employment of SPE seems better due to the lack of peaks occurring in the vicinity of a- and b-thujone retention in this procedure. In the case of LLE, the extract contains some impurities e.g. cholesteric derivatives which might cause contamination of chromatographic equipment – especially its injection system and/ or column (Fig. 1D).

Fig. 1. GC chromatograms of: (A) hexanolic solution of a- and b-thujone with IS. (B) SPE extract from blank plasma sample. (C) SPE extract from plasma sample spiked with alcoholic solution of a-, b-thujone and IS. (D) LLE extract from blank plasma. (E) SPE extract from plasma of a voluntary who consumed alcoholic solution of aand b-thujone.

M.P. Dybowski, A.L. Dawidowicz / Forensic Science International 259 (2016) 188–192 Table 1A Validation data obtained by analysis of spiked samples using the described method (n = 5). RSDb [%]

Recovery [%]

250  5.85 100  2.27 75  2.12 50  1.39 25  0.66 10  0.25 1  0.031 0.1  0.0032 0.01  0.0003

2.34 2.27 2.83 2.78 2.64 2.50 3.10 3.20 3.00

99.03 99.21 98.27 98.14 98.31 97.14 97.21 97.33 97.01

Average

2.74

97.96

a-Thujone concentration  SDa [mg/g]

a b

Source of variability

SD – standard deviation. RSD – relative standard deviation.

RSDc [%]

Recovery [%]

50  1.25 25  0.58 10  0.27 1  0.03 0.1  0.0033 0.01  0.00034

2.50 2.32 2.70 3.01 3.30 3.40

97.85 97.42 97.61 96.32 97.01 96.27

Average

2.87

97.08

c

Source of variability

Sum of squares (SS)

Mean square (MS)

F-value

8

9.6450

1.2060

1.5250

36

28.4600

0.7905

–

44

38.1000

–

–

Sum of squares (SS)

Mean square (MS)

F-value

5

4.311

0.862

0.6792

24

30.465

1.269

–

29

34.776

–

–

Degrees of freedom (df)

Table 2B The ANOVA F-test table for ß-thujone (a = 0.05).

b-Thujone concentration  SDb [mg/g]

b

Table 2A The ANOVA F-test table for a-thujone (a = 0.05).

Treatment (between groups) Error (within groups) Total

Table 1B Validation data obtained by analysis of spiked samples using the described method (na = 5).

a

191

n – number of repetitions. SD – standard deviation. RSD – relative standard deviation.

The identity and purity of the peaks were confirmed by the MS spectra and the value of the retention index, which are 1102 for and a-thujone and 1116 for b-thujone, exactly the same as in the literature [35]. The resultant calibration curves show in both cases excellent linearity with R2 = 0.9996 for a-thujone and R2 = 0.9993 for bthujone. The obtained values exceed the Food and Drug Administration (FDA) requirements of R2  0.999 over in a minimum of 5 observations [36,37]. The repeatability and trueness of the SPE step of the applied procedure are exhibited in Tables 1A and 1B. As results from the data, the examined method is characterized by very high SPE recovery [mean recovery equals 97.96% for a-thujone (see Table 1A) and 97.08% for b-thujone (see Table 1B)] and exhibits excellent precision in all cases. The precision of the SPE step determined as mean relative standard deviation (RSD) equals: 2.74% for a-thujone and 2.87% for b-thujone.

Treatment (between groups) Error (within groups) Total

Degrees of freedom (df)

The results of one-way ANOVA F-test show that there is no significant difference between recoveries depending on the concentration level both for a- and b-thujone (see Tables 2A and 2B). The estimated F-value (F = 1.5250 for a-thujone and F = 0.6792 for b-thujone) is significantly lower than Fcritical-value (Fcrtical = 2.620 for a-thujone and Fcrtical = 2.220 for b-thujone). The estimated limit of detection for a- and b-thujone (LOD) equals 2.5 ng/g and 4.29 ng/g, respectively. The limit of quantification (LOQ) equals: 8.33 ng/g for a-thujone and 14.27 ng/g for bthujone. It is worth mentioning at this moment the work of Kro¨ner et al. [34] who has presented a pilot studies concerning the determination of a- and b-thujone in human plasma by HS-SPME/GC/MS. In these studies, blood samples collected from two persons were examined 2 h after ingestion of 110 mL absinthe (thujone concentration 35 mg/L). Despite of the meticulous optimization of presented procedure [34] and achieving a very low detection limit (0.34 ng/mL), the authors have not been able to determine aand b-thujone in the examined blood samples. It can results not only from relatively quick metabolism of a- and b-thujones but also from their strong binding with plasma proteins. The SPE/GC method described in this paper involves an analyte–protein complex cleavage step, before chromatographic analysis. Table 3 combines the a- and b-thujone concentration in human plasma samples taken from volunteers who consumed the alcoholic solution of those analytical standards 1 h before sampling. The observed differences in thujone plasma concentration obviously result from different metabolism of the volunteers

Table 3 Concentration of a- and b-thujone in deferent human serum samples (n = 5). Test subject

1 2 3 4 5 a b c

Weight [kg]

95 82 74 71 55

Age [years]

24 49 45 49 23

Sexa

M M F M F

M – male, F – female. KBW – kilogram body weight. SD – standard deviation; RSD – relative standard deviation.

Ingested dose of aand b-thujone [mg/KBWb]

3.2 3.7 4.1 4.2 5.5

The quantity of detected analytes (total content of a- and b-thujone) in Intra-day analysis [ng/g  SD (RSD%)]c

Inter-day analysis [ng/g  SD]

22.26  1.00 28.28  0.85 29.43  1.18 30.80  1.03 37.63  1.09

22.26  1.02 28.28  0.91 29.43  1.22 30.80  1.09 37.63  1.13

(4.5) (3.0) (4.0) (3.4) (2.9)

(4.6) (3.2) (4.2) (3.5) (3.0)

192

M.P. Dybowski, A.L. Dawidowicz / Forensic Science International 259 (2016) 188–192

connected with their age, sex, weight, etc. The determined a- and b-thujone concentrations in human plasma are much higher than LOQ in all cases, although the administered doses were significantly lower than the amount causing negative symptoms in humans. 4. Concluding remarks The presented study describes simple and rapid analysis of aand b-thujone in human serum after consumption of those compounds in alcoholic beverages (characteristic congeners of Absinthe). According to performed studies a- and b-thujone strongly binds with plasma proteins (more than 90%), creating a very strong protein–thujone complexes. This is why, the presented SPE procedure involves protein–thujone complex cleavage before chromatographic analysis. The method described in this paper is characterized by low detection limit very high recovery and great linearity After a slight tuning, the presented SPE procedure conditions for thujones isolation may be adapted for extraction a- and b-thujone from other biological fluids, such as cerebral biofluids and urine samples. References [1] D.W. Lachenmeier, S.G. Walch, S.A. Padosch, L.U. Kro¨ner, Absinthe – a review, Crit. Rev. Food Sci. Nutr. 46 (2006) 365–377. [2] A.L. Dawidowicz, M.P. Dybowski, SPE isolation of low-molecular oxygen compounds from essential oils, J. Sep. Sci. 33 (2010) 3213–3220. [3] D.W. Lachenmeier, J. Emmert, T. Kuballa, G. Sartor, Thujone – Cause of absinthism? Forensic Sci. Int. 158 (2006) 1–8. [4] C. Gambelunghe, P. Melai, Absinthe: enjoying a new popularity among young people? Forensic Sci. Int. 130 (2002) 183–186. [5] A. Arceusz, A. Occhipinti, A. Capuzzo, M.E. Maffei, Comparison of different extraction methods for the determination of a- and b-thujone in sage (Salvia officinalis L.) herbal tea, J. Sep. Sci. 36 (2013) 3130–3134. [6] D. El Montassir, A. Aamouche, N. Vanthuyne, M. Jean, P. Vanloot, M. Taourirte, N. Dupuy, Ch. Roussel, Attempts to separate (–)-a-thujone, (+)-b-thujone epimers from camphor enantiomers by enantioselective HPLC with polarimetric detection, J. Sep. Sci. 36 (2013) 832–839. [7] O. Pelkonen, K. Abass, J. Wiesner, Thujone and thujone-containing herbal medicinal and botanical products: toxicological assessment, Regul. Toxicol. Pharmacol. 65 (2013) 100–107. [8] D.W. Lachenmeier, M. Uebelacker, Risk assessment of thujone in food and medicines containing sage and wormwood – evidence for a need of regulatory changes? Regul. Toxicol. Pharmacol. 58 (2010) 437–443. [9] O.H. Drummer, Book review – Medical toxicology of drug abuse – synthesized chemicals and psychoactive plants. Donald G. Barceloux with contributions by Robert B. Palmer (Wiley 2012), Forensic Sci. Int. 231 (2013) 270, http://dx.doi.org/ 10.1016/j.forsciint.2013.06.002. [10] S.A. Padosch, D.W. Lachenmeier, L.U. Kro¨ner, Absinthism: a fictitious 19th century syndrome with present impact, Subst. Abuse Treat Prev. Policy 1 (2006) 14. [11] D.W. Lachenmeier, Wormwood (Artemisia absinthium L.)—a curious plant with both neurotoxic and neuroprotective properties? J. Ethnopharmacol. 131 (2010) 224–227. [12] J.P. Meschler, A.C. Howlett, Thujone exhibits low affinity for cannabinoid receptors but fails to evoke cannabimimetic responses, Pharmacol. Biochem. Behav. 62 (1999) 473–480. [13] S. Yamaori, Ch. Maeda, I. Yamamoto, K. Watanabe, Differential inhibition of human cytochrome P450 2A6 and 2B6 by major phytocannabinoids, Forensic Toxicol. 29 (2011) 117–124.

[14] M.M. Czyzewska, J.W. Mozrzymas, Monoterpene a-thujene exerts a differential inhibitory action on GABAA receptors implicated in phasic and tonic GABAergic inhibition, Eur. J. Pharmacol. 702 (2013) 38–43. [15] G. Machata, O. Prokop, Uber Begleitsubstanzen alkoholischer Getranke im Blut, Blutalkohol 8 (1971) 349–353. [16] L.N. Rodda, J. Beyer, D. Gerostamoulos, O.H. Drummer, Alcohol congener analysis and the source of alcohol: a review, Forensic Sci. Med. Pathol. 9 (2013) 194–207. [17] B.H. McAnalley, E.H. Aguayo, The chemistry of alcoholic beverages, in: J.C. Garriott (Ed.), Medicolegal aspects of alcohol, 5th ed., Lawyers and Judges Publishing Co., Tuscan, 2008, pp. 1–23. [18] K. Schulz, K. Schlenz, S. Malt, R. Metasch, W. Ro¨mhild, J. Dreßler, D.W. Lachenmeier, Headspace solid-phase microextraction–gas chromatography–mass spectrometry for the quantitative determination of the characteristic flavouring agent eugenol in serum samples after enzymatic cleavage to validate post-offence alcohol drinking claims, J. Chromatogr. A. 1211 (2008) 113–119. [19] W.T. Lowry, J.C. Garriott, Controlled and noncontrolled but commonly abused substances, Forensic Toxicol. (1979) 117–410. [20] M.C. Santillan-Valverde, M. Garcia-Garibay, Congeners biosynthesis during alcoholic fennentations, Rev. Latinoam Microbiol. 15 (1998) 108–124. [21] W. Bonte, Congener analysis, in: Encyclopedia of Forensic Sciences, Academic Press, Duesseldorf, 2000, pp. 93–102. [22] J.M. Jurado, A. Alca´zar, F. Pablos, M.J. Martı´n, LC determination of anethole in aniseed drinks, Chromatographia 64 (2006) 223–226. [23] K. Schulz, K. Schlenz, S. Malt, R. Metasch, W. Ro¨mhild, J. Dreßler, Determination of anethole in serum samples by headspace solid-phase microextraction–gas chromatography–mass spectrometry for congener analysis, J. Chromatogr. A 1200 (2008) 235–241. [24] K. Hajime, Comprehensive two-dimensional gas chromatographic analysis of commercial lemon-flavored beverages, J. Sep. Sci. 29 (2006) 2350–2356. [25] A.L. Dawidowicz, M.P. Dybowski, Fast determination of a- and b-thujone in alcoholic beverages using solid-phase extraction and gas chromatography, Food Contam. 25 (2012) 197–201. [26] J. Emmert, G. Sartor, F. Sporer, J. Gummersbach, Determination of a-/ß-thujone and related terpenes in absinthe using solid phase extraction and gas chromatography, Deut. Lebensm. Rundsch. 100 (2004) 9–13. [27] C.L. Galli, G. Galli, E. Tragni, D. Caruso, A. Fiecchi, Quantilalive analysis of a,ßthujone, pulegone. safrole, coumarin and ß-asarone in alcoholic beverages by selected-ion monitoring, J. Appl. Toxicol. 4 (1984) 213–216. [28] D.W. Lachenmeier, Assessing the authenticity of absinthe using sensory evaluation and HPTLC analysis of the bitter principle absinthin, Food Res. Int. 40 (2007) 167–175. [29] D.W. Lachenmeier, D. Nathan-Maister, T.A. Breaux, E.M. Sohnius, K. Schoeberl, T. Kuballa, Chemical composition of vintage preban absinthe with special reference to thujone, fenchone, pinocamphone, methanol, copper, and antymony concentrations, J. Agric. Food Chem. 56 (2008) 3073–3081. [30] D.W. Lachenmeier, N. Scha¨fer, Bestimmung von a- und ß-Thujon in Spirituosen mittels Festphasenmikroextraktion (HS-SPME): Vergleich mit konventionellen Extraktionsmethoden, Deut. Lebensm. Rundsch. 101 (2005) 534–539. [31] P.M. Scott, G.A. Lawrence, B.P.Y. Lau, Determination of thujone by derivatization with dansylhydrazine and liquid chromatography, J. Liq. Chromatogr. Rel. Technol. 27 (2004) 1071–1081. [32] M. Scha¨ffer, T. Gro¨ger, M. Pu¨tz, R. Zimmermann, Assessment of the presence of damiana in herbal blends of forensic interest based on comprehensive twodimensional gas chromatography, Forensic Toxicol. 31 (2013) 251–262. [33] Y.B. Monakhova, T. Kuballa, D.W. Lachenmeier, Rapid determination of total thujone in absinthe using 1H NMR Spectroscopy, Int. J. Spectrosc. 2011 (2011), http://dx.doi.org/10.1155/2011/171684, Article ID 171684. [34] L.U. Kroner, S.A. Padosch, D.W. Lachenmeier, B. Madea, Investigations on the medico-legal relevance of spirits containing thujone with special regard to toxicological-analytical aspects, Blutalkohol 42 (2005) 263–271. [35] R.P. Adams, Identification of Essential Oil Components by Gas Chromatography/ Mass Spectrometry, Allured Publishing Corporation, Carol Stream, IL, 1995. [36] Center for Drug Evaluation and Research (CDER), US Food and Drug Administration (FDA), Guide for Industry. Q2B Validation of Analytical Methods Protocol. Metrology, 1996 http://www.fda.gov./CDER/guidance/index.htm. [37] Center for Drug Evaluation and Research (CDER), US Food and Drug Administration (FDA), Reviewer Guidance. Validation of Chromatographic Methods, 1994 www.fda.gov/CDER.