IJFS#1252_bozza Ital. J. Food Sci., vol. 30, 2018 - 752 PAPER GC-C-IRMS CHARACTERIZATION OF SYNTHETIC BIS(METHYL-THIO)METHANE IN TRUFFLE FLAVORINGS M. BONONI*1, F. TATEO1, F. BENEVELLI2, A. PENNETTA3 and G. DE BENEDETTO3 1Department of Agricultural and Environmental Sciences, University of Milan, Via Celoria 2, 20133 Milan, Italy 2Bruker Italia S.r.l., Viale Lancetti, 43, 20158 Milan, Italy 3Department of Cultural Heritage, University of Salento, S.P. 6, Monteroni, Lecce, Italy *Corresponding author: Tel.: +39 0250316538 E-mail address: monica.bononi@unimi.it ABSTRACT Bis(methyl-thio)methane (BMTM), the molecule which provides “white truffle-like” flavor was characterized by physico-chemical methods. Analysis by GC-C-IRMS of eight samples of synthetic BMTM from various raw material suppliers allowed the investigation of the δ13C values. More, ten samples purchased on the Italian flavoring market, declared as synthetic BMTM principal component diluted in olive oil were analyzed by GC-C-IRMS. The results of both sample groups allowed us to define the range of δ13C values of synthetic BMTM. We verified if the simple proposed extraction method allows to measure the δ13C value of BMTM also identified in seasonings of the Italian market declared on label as “white truffle flavored olive oil”. In all twenty samples purchased on the market, the data strictly corresponded with synthetic BMTM as the principal component. Measurements by 1H NMR made on synthetic BMTM and BMTM extracted from “white truffle-like flavor” confirmed that the adopted extraction method using methanol-d4 determined the isotopic distribution of 13C/12C ratio in two characteristic sites of this molecule. Keywords: white truffle flavoring, BMTM, GC-C-IRMS, 1H NMR Ital. J. Food Sci., vol. 30, 2018 - 753 1. INTRODUCTION Tuber magnatum (white truffle), Tuber aestivum (summer truffle), and Tuber melanosporum (black truffle) are the most well-known species belonging to the genus Tuber F.H. These species have been the object of numerous studies using various analytical systems for the identification of the compounds that provide the distinctive aroma of these fungi (BELLESIA et al., 1996; DÍAZ et al., 2002; DÍAZ et al., 2003; TALOU et al., 1989). Several reports have considered the constituents responsible for the typical aroma and have also studied the quantitative and qualitative fluctuations of these compounds, depending upon truffle type and geographical origin (COSTA et al., 2015; FIECCHI et al., 1967; GIOACCHINI et al., 2008; MAURIELLO et al., 2004; PELUSIO et al., 1995). Presently, pure natural BMTM derived from truffles is not available on the raw material flavor market because the levels of BMTM in truffles are too low for satisfactory isolation of a significant quantity of this molecule (SCHMIDBERGER and SCHIEBERLE, 2017). Also, the commercial cost of the natural BMTM from the white truffle would be unsustainable because the raw material is very expensive (BORSINO DEL TARTUFO, 2018). Therefore, due to the non-feasibility of BMTM isolation from the natural matrix, no measurements using Gas Chromatography-Combustion-Isotope Ratio Mass Spectrometry (GC-C-IRMS) have been previously carried on. The European Union (EU) Regulation 1334/2008 allows the citation of the flavoring source (in our case “white truffle”) only if the flavoring components are obtained exclusively or by at least 95% (w/w) from the source material referred to and the other maximum 5% can derive only from other natural sources. Therefore, the citation in the label “natural flavoring”, without the citation of a defined natural source, may only be used if all the flavoring components derive from natural sources. If one or more synthetic compounds are present in a flavoring formulation, the label must report only the term “flavor” without any reference to a food, food category or a vegetable or animal flavoring source (REG. EU 1334, 2008). The corresponding naturally-occurring identical compound, easily synthesized by the oil industry and supported in olive oil (PACIONI et al., 2014), is used as a flavoring agent for truffle flavored food products. Among the “white truffle-like” flavored foods, extra virgin olive oil, flavored with bis(methyl-thio)methane (BMTM), used as a seasoning, occupies the most important position in the market. Using a simple extraction method and the GC-C-IRMS analytical technique, we aimed to ascertain the reliability of characterizations of the synthetic BMTM molecule present in “white truffle-like” flavors and seasonings. Following the EU Regulation 1334/2008 before cited, the identification of synthetic BMTM in the formulation of flavors and seasonings does not allow the use on the label the term “natural white truffle flavor”, or “natural flavor” or “white truffle flavor”, but only the term “flavor”. Synthetic BMTM characterization by GC-C-IRMS, confirmed by 1H-NMR, allowed us to demonstrate the identity of the BMTM molecule extracted from “white truffle-like” flavors purchased on the Italian flavoring market and from olive oil seasonings available on the Italian consumer market. 2. MATERIALS AND METHODS 2.1. Solvents Methanol (anhydrous) (99.8) and methanol d-4 were purchased from Sigma Aldrich (Milan, Italy). Ital. J. Food Sci., vol. 30, 2018 - 754 2.2. Samples analyzed by GC-C-IRMS - Eight samples of BMTM (C2H8S2) declared to be synthetically derived were obtained. The first purchased was a standard (> 99%) from Sigma Aldrich (Italy) and the others were purchased on the flavor raw material market (FCI, Frutarom, Moellhausen, Penta, Treatt, Sterling, Sigma Aldrich). - Ten samples of “white truffle-like” flavors purchased on the Italian flavorings market and declared as synthetic BMTM diluted in olive oil. - Twenty “white truffle-like” seasonings purchased on the Italian market and consisting of flavored extra virgin olive oil, or olive oil, or sunflower oil. All the seasonings were declared to contain olive oil or extra-virgin olive oil except the seasoning n. 4, which contains sunflower oil. The product n. 8 reported on the label “natural flavor”. The product n. 13 reported on the label “white truffle natural flavor”. The products n. 6, 11, 16, and 19 reported on the label “white truffle flavor”. The other products reported on the label “flavor”. 2.3. Samples analysis by 1H NMR Two samples were analyzed, one corresponding to the standard BMTM (sample n.1 in Table 1) and another corresponding to a “white truffle-like” flavoring (sample n.1 in Table 2). 2.4. Samples preparation For GC-C-IRMS analysis, samples of synthetic BMTM (Table 1) were dissolved in anhydrous methanol (99.8) at a concentration of 60 µL mL-1. The mix was vortexed for 1 min. For samples of “white truffle-like” flavoring (Table 2) an aliquot of 3 mL was vortexed with 1 mL of methanol for 5 min and then centrifuged at 5000 rpm for 5 min. The methanol layer was isolated by a Pasteur pipette and utilized for the analysis. For samples of seasonings, an aliquot of 5-10 mL was vortexed with 1 mL of methanol for 5 min and then centrifuged at 5000 rpm for 5 min. The methanol layer was isolated by a Pasteur pipette and utilized for the analysis. For NMR analysis, a sample of standard BMTM (sample n.1 in Table 1) was dissolved in methanol d-4 at a concentration of 60 µL mL-1. The mix was vortexed for 1 min. For sample of a “white truffle-like” flavoring (sample n.1 in Table 2), an aliquot of 3 mL was vortexed with 1 mL of methanol d-4 for 5 min and then centrifuged at 5000 rpm for 5 min. The methanol layer, isolated by a Pasteur pipette, was utilized for the analysis. 2.5. GC-C-IRMS analysis The system consisted of an Agilent Technology 7890A gas chromatograph equipped with a G4513A autosampler (Agilent Technology, Germany) and coupled to an IsoPrime stable IRMS GC5 (Isoprime, Cheadle, UK) via a combustion interface under a continuous flow of helium. The combustion interface consisted of a ceramic furnace with a copper oxide and platinum catalyst at 850°C. The carbon stable isotope ratio was determined by referring to the international standard Vienna PeeDee Belemnite (δ13CVPDB) with a defined 13C content. The CRM used for the GC-C-IRMS multipoint calibration were n-undecane (δ13C: -26.11‰, Chiron C0414.11-150-CY), n-pentadecane (δ13C: -30.22‰, Chiron C0418.15-150-CY) and n- hexadecane (δ13C: -34.87‰). The hexadecane delta value was obtained by EA-IRMS using the following primary standards: glucose (δ13C: -10.76‰, Sigma Aldrich BCR657), Ital. J. Food Sci., vol. 30, 2018 - 755 polyethylene (δ13C: -32.15‰, IAEA IAEA-CH-7) and lithium carbonate (δ13C: -46.60‰, NIST RM 8545). Isotope ratios were expressed as values (‰) and calculated on the basis of the following equation: (i R SA – i R REF) δi E = i R REF where “i” is the mass number of the heavier isotope of element E (in this case 13C), R SA is the respective isotope ratio of the sample and RREF is the relevant internationally recognized reference material. The delta values were multiplied by 1000 and expressed in units “per mill” (‰) (COPLEN, 2011). The GC was operated using a HP-5 capillary column, 30 m x 0.32 mm i.d., 0.25 µm film thickness (Agilent – Italy). Helium was used as the carrier gas at a flow rate of 1.2 mL. The oven temperature program was initially 50°C (held for 1 min), then increased to 150°C at a rate of 5°C min-1, then increased to 250°C at a rate of 20°C min-1 (held for 1 min). The injector temperature was 220°C and the injection volume was 1 mL (split 1:10). Data were collected in triplicate. 2.6. NMR analysis The 1H NMR spectra were measured on a Bruker spectrometer AVIII400 equipped with a SampleXpress sample changer and using a BBIz probe. The spectra were acquired using a 30° excitation pulse width of 2.66 µs, relaxation delay of 40 sec, and acquisition time of 5 s at 298 K. T1 relaxation times of the quantified peaks were measured with inversion- recovery experiments to assure that no bias from a short D1 would arise. The longest T1 measured for the standard BMTM (sample n.1 in Table 1) was 6.96 s, found for the CH2 peak, while for the “white truffle-like” flavoring (sample n.1 in Table 2) we observed 6.02 sec for the same peak as the longest T1. Thus, the chosen D1 was sufficient to ensure a complete relaxation of the magnetization and to give fully quantitative results. The spectral width was 40 ppm centered at 6.5 ppm to ensure that the baseline was perfectly flat, as this is a precondition to correct integration and quantification of the peaks. For each spectrum, 64 scans were summed. The acquisitions were performed in a fully automated way. The spectra were processed with 260 K points and an exponential multiplication of 0.3 Hz. The data were acquired, processed, and analyzed using the software program Topspin 3.5 from Bruker Biospin. Phasing and baseline correction were performed manually. In the case of the standard BMTM (sample n.1 in Table 1), the two main peaks of the H1-C12 were considered together with both satellites for each peak. However, in the case of the “white truffle-like” flavoring (sample n.1 in Table 2), only one satellite per peak was considered due to the superposition with other signals in the matrix. The intensity of the peak was doubled for the calculation of the isotopic ratio. 3. RESULTS 3.1. GC-C-IRMS The samples of commercially available synthetic BMTM could be readily prepared for GC- C-IRMS analysis by simple dilution in anhydrous methanol (Table 1). Ital. J. Food Sci., vol. 30, 2018 - 756 Similarly, the extraction method adopted for samples reported in Tables 2 and 3 allowed us to isolate BMTM as a methanolic extract for use in GC-C-IRMS measurements. For samples reported in Tables 1 and 2, corresponding to BMTM declared from synthesis and “white truffle-like” flavors purchased on the Italian flavorings market, respectively, the δ13C values varied in the tight range of -42.24 and -43.40‰. The above cited range, deduced from analysis of a sufficiently large number of samples (standards and flavorings) produced with “BMTM” certified from synthesis, was very useful for characterization of the “synthetic” authenticity of the molecule subjected to δ13C measure. In addition, the same data, as expected, was not consistent with the carbon isotope ratio natural abundance ranges, as documented in literature data (VAN LEEUWEN et al., 2014). Data obtained for seasonings of Table 3 are also included in the range between -42.34 and -43.26‰, and was not significantly different from the values corresponding to the synthetic samples cited above. Table 1. GC-C-IRMS data produced by eight samples declared as BMTM from synthesis and available on the flavoring market. Sample δ13C ‰ (mean) s 1 -43.35 0.43 2 -42.30 0.35 3 -43.05 0.34 4 -42.47 0.32 5 -42.24 0.35 6 -43.40 0.34 7 -42.26 0.34 8 -43.14 0.41 Table 2. GC-C-IRMS data produced by ten samples, all declared as “white truffle-like” flavors, purchased on the Italian flavorings market and consisting of synthetic BMTM as the principal component diluted in olive oil. Sample d13C ‰ (mean) s 1 -42.50 0.58 2 -42.43 0.53 3 -43.12 0.41 4 -42.40 0.34 5 -42.55 0.38 6 -42.64 0.47 7 -43.18 0.52 8 -42.38 0.45 9 -42.46 0.37 10 -42.52 0.46 Ital. J. Food Sci., vol. 30, 2018 - 757 Table 3. GC-C-IRMS data produced by twenty samples of white truffle flavored oils purchased on the Italian market (glass pack 40-250 mL). Seasoning δ13C ‰ (mean) s 1 -42.34 0.42 2 - 43.08 0.34 3 -42.81 0.52 4 -42.63 0.36 5 -43.16 0.38 6 -42.86 0.48 7 -43.02 0.35 8 n.d. ---- 9 -42.53 0.46 10 -43.12 0.43 11 -43.06 0.37 12 -43.18 0.51 13 -42.68 0.45 14 -42.48 0.39 15 -43.04 0.37 16 -43.10 0.41 17 -42.66 0.56 18 -42.74 0.46 19 -43.26 0.38 20 -42.45 0.42 3.2. 1H NMR characterization of BMTM The aim of our study was to develop an approach using NMR methodology to characterize the BMTM synthetic molecule of a standard sample and to verify the possibility of using the extraction method referred in “2.4 Sample preparation” to characterize the synthetic BMTM present in a “white truffle-like” flavor. We measured the values 13C/12C for the two sites corresponding to CH2 and CH3 in the BMTM molecule extracted from the two samples. Our data showed the structural isotopic distribution of 13C/12C in the characteristic sites of these two molecules. 3.2.1 BMTM from synthesis (sample 1, Table 1) The 1H NMR spectra of the synthetic standard BMTM were acquired using two aliquots of the same sample diluted in methanol-d4 and in replicates to evaluate the experimental CV% (coefficient of variation %) from 8 trials. The BMTM 1H NMR spectrum in methanol-d4 displayed two singlets, resonating at 3.68 and at 2.15 ppm. The signals are attributed to the CH2 (3.68 ppm) and the CH3 (2.15 ppm). The 13C satellites of the two peaks were easily recognizable and they appeared as doublets with a heteronuclear J coupling of 149.9 Hz for the CH2 and of 138.5 Hz for the CH3. The doublets were centered at the isotropic resonances of the peaks. As expected, the intensities of the satellites were significantly lower than those of the main peaks, which were attributed to the 1H attached to the 12C. Ital. J. Food Sci., vol. 30, 2018 - 758 The ratio between the intensities of satellites due to the 1H - 13C heteronuclear coupling and that of the main peak allowed us to extract the integral ratio between 13C and 12C in a site specific manner. The 1H NMR spectrum (Fig. 1) and the results (Table 4) indicated that the ratio of 13C/12C gave the same value for both sites (CH2 and CH3). This value is in agreement with the expected 13C/12C global ratio. Figure 1. 1H spectrum of the BMTM from synthesis, available on the flavor raw material market (sample 1, Table 1). The asterisk indicates the 13C satellites. The dotted lines indicate the integral regions. Table 4. Data %13C measured by 1H NMR spectra in replicates of two aliquots for each sample: synthetic BMTM standard (sample 1 in Table 1), synthetic BMTM from “white truffle-like” flavor (sample 1 in Table 2). % 13CH BMTM synth. BMTM synth. std. flav. replicates -CH2- -CH3 -CH2- -CH3 1 1.02 1.06 1.01 0.99 2 1.03 1.02 1.07 0.99 3 1.05 1.05 1.03 1.06 1 1.05 1.05 0.99 1.02 2 1.02 1.00 0.99 0.99 3 1.00 1.01 0.99 1.04 4 1.01 1.00 0.99 1.02 5 1.03 0.99 1.03 1.05 mean 1.03 1.02 1.01 1.02 CV% 0.02 0.03 0.03 0.03 al iq uo t 1 al iq uo t 2 Ital. J. Food Sci., vol. 30, 2018 - 759 3.2.2 BMTM extracted from synthetic white truffle-like flavoring (sample1, Table 2) The same approach described for the synthetic BMTM standard was applied to the synthetic BMTM in “white truffle-like” flavoring. As in the case of the molecule BMTM from synthesis, the spectrum was acquired in 8 replicates derived from two extracts produced from two aliquots of synthetic BMTM from “white truffle-like” flavoring. The 1H spectra of the two aliquots extracted with methanol-d4 showed similar features to that of the synthetic BMTM standard, with the two singlets in the same position (3.67 ppm for CH2 and 2.14 ppm for CH3) and the same heteronuclear J couplings (149.9 Hz for the CH2 and of 138.5 Hz for the CH3). The 0.01 ppm shift is attributable to a matrix effect and it is reproducible in all the replicates. The spectrum acquired is reported in Fig. 2 and also in this case, as reported in Table 4, the ratio of 13C/12C gave the same value for both sites (CH2 and CH3). As shown in Fig. 2, the presence of a potentially interfering substance often found in “white truffle” flavor (identifiable by GC/MS as methylsulfinyl(methylthio)-methane) did not hinder the identification of 13C satellites and the % 13C values were strictly comparable to data obtained for synthetic BMTM standard. To clarify, the 1H NMR spectrum of the sulfoxide is easily recognizable by these features: one singlet at 2.71 ppm attributable to one terminal methyl; one singlet at 2.33 ppm attributable to the second terminal methyl; two doublets centered at 3.98 ppm and at 3.83 ppm with a J=13.8 Hz attributable to the CH2 protons that became diastereotopic upon asymmetric oxidation. Figure 2. 1H spectrum of the “white truffle” flavor purchased on the Italian flavor market and consisting of synthetic BMTM as the principal component diluted in olive oil. The asterisk indicates the 13C satellites. The dotted lines indicate the integral regions. The inset shows the expansion of the 4.05-3.75. These peaks are attributed to the CH2 protons in a molecule identified as methylsulfinyl(methylthio)-methane. Ital. J. Food Sci., vol. 30, 2018 - 760 4. DISCUSSION The simple extraction method, adopting methanol as the solvent, is useful for application to provide data by GC-C-IRMS of a synthetic BMTM standard, in flavoring from the raw material market (see Table 1), and in Italian flavoring from the commercial market (see Table 2). The δ13C‰ data collected for all these matrices ranged between -42.24 (σ 0.35) and -43.40‰ (σ 0.34). Clearly, the range is external to the natural abundance of the carbon isotope ratio, and allows one to identify the synthetic origin of BMTM if this molecule is used to produce seasonings made by the addition to olive oil or other vegetable oils. The δ13C‰ data obtained with the same simple extraction method on twenty seasonings purchased on the Italian market allowed us to deduce that the synthetic BMTM molecule is clearly identifiable. In fact, the δ13C‰ values reported in Table 3, except for the sample n. 8, which does not contain BMTM, exhibited data in the range -42.34 and -43.26‰, identified as characteristic of synthetic BMTM. Most seasoning samples – representing about 75% of the total seasoning samples examined (namely, n. 1-5, 7, 9, 10, 12, 14-15, 17-18 and 20 in Table 3) are compliant with the EU Regulation 1334/2008 because they contain synthetic BMTM and on the label only the term “flavor” is correctly used. Samples n. 6, 11, 13, 16, and 19 in Table 3 are not compliant with the EU Regulation 1334/2008 because they contain synthetic BMTM and report on label the reference to the term “truffle”. Specifically, the term “white truffle flavor” was used for samples 6, 11, 16, 19, while “natural white truffle flavor” for sample 13. For the sample n. 8 in Table 3, the only one resulting as not containing BMTM, and reported on label as produced with “natural flavor”, the judgment of compliance or otherwise with the EU Regulation 1334/2008 does not depend from the BMTM identification (natural or synthetic), but from the origin of all the molecules constituting the flavor. In fact, a natural flavor used for a seasoning can be realized totally with natural molecules. In this case, the final judgment is not linked to the BMTM identification, but from the naturalness of all the components of the flavor used. There was no evidence of data that did not fall within the range characteristic of synthetic BMTM, and it is justified by being anyhow unavoidable due to the lack of commercial availability of natural BMTM in the raw material flavor market. Adopting the same extraction method, but using methanol d-4 as the solvent, we demonstrated the feasibility of 1H NMR measures to calculate % 13C/12C in two characteristic sites -CH2- and -CH3. This investigation, applied to a declared synthetic BMTM standard (sample 1 in Table 1) characterizes the integral ratio between 13C and 12C in a site specific manner. In fact, the 1H NMR spectrum (Fig. 1) and the results (Table 4) indicated that the ratio of 13C/12C gave the same value for both sites (CH2 and CH3), in agreement with the expected 13C/12C global ratio. Data produced from a synthetic BMTM standard were not statistically different from the corresponding data produced for a “white truffle-like” flavor, as shown in Table 4. 5. CONCLUSIONS The data reported in this paper are the first GC-C-IRMS and 1H NMR contributions to the characterization of synthetic BMTM available on the flavoring market and in some “white truffle-like” flavors. All the analyzed seasoning (except one sample not containing BMTM) produced GC-C-IRMS data of BMTM in agreement with the values of the synthetic molecule. Ital. J. Food Sci., vol. 30, 2018 - 761 REFERENCES Bellesia F., Pinetti A., Bianchi A. and Tirillini B. 1996.Volatile compounds of the white truffle (Tuber magnatum Pico) from middle Italy. Flavour Frag. J. 11:239-243 Borsino del tartufo, 2018. www.tuber.it/en/borsino-del-tartufo.php. (most recent access date 12 April 2018). Coplen T.B. 2011. Guidelines and recommended terms for expression of stable-isotope-ratio and gas-ratio measurement results. Rapid Commun. Mass Spectrom. 25:2538-2560. Costa R., Fanali C., Pennazza G., Tedone L., Dugo L., Santonico M., Sciarrone D., Cacciola F., Cucchiarini L., Dachà M. and Mondello L. 2015. Screening of volatile compounds composition of white truffle during storage by GCxGC- (FID/MS) and gas sensor array analyses. LWT Food Sci. Technol. 60:905-913. Díaz P., Señoráns F.J., Reglero G. and Ibáñez E. 2002. Truffle aroma analysis by headspace solid phase micoextraction. J. Agr. Food Chem. 50:6468-6472. Díaz P., Ibáñez E., Señoráns F.J. and Reglero G. 2003. Truffle aroma characterization by headspace solid-phase microextraction. J. Chromatogr. A 1017:207-214. European Union (EU) Regulation (EC) No 1334/2008 of the European Parliament and of the Council of 16 December 2008 on flavourings and certain food ingredients with flavouring properties for use in and on foods and Amending Council Regulation (EEC) No 1601/91, Regulations (EC) No 2232/96 and (EC) No 110/2008. Off. J. Eur. Union, L: Legis. 2008, 51, 34-50. Fiecchi A., Galli Kienle M., Scala A. and Cabella P. 1967. Bis-methylthiomethane, an odorous substance from white truffle, Tuber magnatum pico. Tetrahedron Lett. 8:1681-1682. Gioacchini A.M., Menotta M., Guescini M., Saltarelli R., Ceccaroli P., Amicucci A., Barbieri E., Giomaro G. and Stocchi V. 2008. Geographical traceability of Italian white truffle (Tuber magnatum Pico) by the analysis of volatile organic compounds. Rapid Commun. Mass Sp. 22:3147-3153. Mauriello G., Marino R., D’Auria M., Cerone G. and G.L. Rana. 2004. Determination of volatile organic compounds from truffle via SPME-GC-MS. J. Chromatogr. Sci. 42:299-305. Pacioni G., Cerretani L., Procida G. and Cichelli A. 2014. Composition of commercial truffle flavores oils with GC-MS analysis and discrimination with an electronic nose. Food Chem. 146:30-35. Pelusio F., Nilsson T., Montanarella L., Tilio R., Larsen B., Facchetti S. and Madsen J. 1995. Headspace solid-phase microextraction analysis of volatile organic sulfur compounds in black and white truffle aroma. J. Agr. Food Chem. 43:2138-2143. Schmidberger P.C. and Schieberle P. 2017. Characterization of the key aroma compounds in white Alba truffle (Tuber magnatum pico) and Burgundy truffle (Tuber uncinatum) by means of the sensomics approach. J. Agr. Food Chem. 65:9287-9296. Talou T., Delmas M. and Gaset A. 1989. The volatile components of tinned black perigord truffles Tuber melanosporum Vitt. Flavour Frag. J. 4:109-112. van Leeuwen K.A., Prenzler P.D., Ryan D. and Camin F. 2014. Gas Chromatography-Combustion-Isotope Ratio Mass Spectrometry for traceability and authenticity in foods and beverages. Comprehensive Rev. Food Sci. F.13:814-83. Paper Received April 30, 2018 Accepted July 25, 2018