If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Truffles are the fruit bodies of ascomycetous fungi belonging to the genus Tuber which form ectomycorrhizae on the roots of angiosperms and gymnosperms.1, 2 The mycorrhiza is a specialized symbiotic structure, where the exchange of nutrients and metabolites takes place, providing mutual benefits to both partners. It is probably after this interaction that the fruit body is developed.
Truffle species are found worldwide; their culinary and commercial value is mainly due to their organoleptic properties such as their aroma, the quality of which clearly provides the economic value of such edible fungi.
Due to the economic interest in truffles, mainly in countries such as Spain, France and Italy (and in others such as New Zealand, Australia, and the United States),3 it is important to develop methods that allow an objective evaluation of truffle aroma.
In the past few decades, volatile organic compounds (VOCs) in truffle aromas have been analyzed using several methods. Some research has been devoted to the identification of truffle aroma compounds and to the study of the effect of processing on the original aroma of different Tuber species.4–9
The most used analytical techniques to concentrate the VOCs of food aroma have obviously been those based on headspace analysis.10 For truffle aroma, techniques such as dynamic headspace coupled to gas chromatography/mass spectrometry (GC/MS)11, 12 and purge-and-trap GC/MS13 have been used to detect black Perigord truffle and Italian white truffle aromas, respectively. Headspace solid-phase microextraction (HS-SPME) combined with GC/MS8, 14, 15 has been used to detect the volatile sulfur compounds in the aroma of white and black truffles (T. magnatum Pico and T. melanosporum, respectively).
SPME is a technique, developed in 1990 by Pawliszyn,16–20 that shows clear advantages over traditional techniques: high sensitivity and reproducibility; low cost; relative simplicity. Unlike other conventional methods, which require extensive sample preparation, SPME is a one-step extraction procedure in which the compounds of interest are absorbed by a thin polymer film or by porous carbonaceous materials bonded to a fused-silica fiber. The HS-SPME technique combined with ion trap mass spectrometry (ITMS),21 capable of producing full-scan mass spectra at very low concentration levels, allowed a detailed analysis of VOCs.
The study of truffle aroma has also been suggested as a way of authenticating different truffle species, for example, to detect the presence of Tuber borchii used as an adulterant for the more highly prized truffle species, T. magnatum.
The aim of this study was to develop a new method for the characterization of six species of truffles by solid-phase microextraction gas chromatography/mass spectrometry (SPME-GC/MS).
Six different species of white and black truffles, with similar morphological characteristics but different organoleptic qualities and economic value, were selected and analyzed. The analyzed species were: Tuber magnatum Pico, T. borchii Vittad, T. dryophilum Tul., T. aestivum Vittad., T. mesentericum Vittad., and T. brumale Vittad. Tuber magnatum is the most prized fruit body having particular organoleptic properties as well as commercial value in Mediterranean areas.
VOCs were extracted from the various species under study using a 2-cm 50/30 µm DVB/CAR/PDMS fiber placed for 10 min at 20°C in the headspace of the truffle sample.
The fragment ions in each spectrum (40 ≤ m/z ≤ 150) were considered as potential descriptors of the composition of the VOCs in the truffles' headspace. We have developed a PCA (principal component analysis) model to distinguish the six truffle species under study, by the composition of the VOCs forming the headspace of samples. The model was used to interrogate MS data of unknown truffle samples, and these were correctly classified. The proposed study of truffle aroma represents a new way of identification and authentication of the different Tuber species.
Truffle species identification
Truffle species were identified, before the GC/MS analysis, by molecular methods, as described below.
DNA from fruit bodies was extracted from about 200 mg of tissue using the method of Lee and Taylor.22
The amplification experiments were performed in a PTC-200 DNA Engine (Genenco) Peltier thermal cycler. The specific PCR (polymerase chain reaction) was carried out using Tuber species-specific ITS (internal transcribed spacer) primers for the identification of T. magnatum, T. dryophilum, T. borchii and T. brumale.23, 24Tuber mesentericum and T. aestivum, for which the specific primers are not known, were identified by the sequence analyses of the ribosomal DNA (rDNA) of the ITS regions. The ITS regions were amplified with the primers ITS1/ITS4.25
The ITS/PCR products were purified using the QIAquik purification kit (Qiagen, SpA, Milan, Italy) and directly sequenced using an ABI prism cycle-sequencing kit (dRhodamine terminator cycle-sequencing kit with AmpliTaq DNA polymerase FS; Perkin–Elmer/Cetus, Branchburg, NJ, USA).
The truffles were collected in a natural truffle ground in central Italy, kept at 4°C and placed in 50 mL vials (Kimble Glass Inc., Vineland, NJ, USA) sealed with butyl-Teflon septum caps (Kimble Glass Inc.). The samples were analyzed within 24 h by SPME-GC/MS. The weight of ascocarps collected was about 1–3 mg.
SPME extraction was performed with Supelco fibers coated with three different stationary phases: polydimethylsiloxane (PDMS, thickness 100 µm), polydimethylsiloxane/divinylbenzene (PDMS/DVB, thickness 65 µm) and divinylbenzene/Carboxen/polydimethylsiloxane (DVB/CAR/PDMS, thickness 50/30 µm). The fibers were supplied by Supelco (Bellafonte, PA, USA). The 50/30 µm DVB/CAR/PDMS fiber, the most suitable, was chosen for further method development.
The method included inserting a new 2-cm 50/30 µm DVB/CAR/PDMS fiber in a manual injection holder followed by preconditioning before the day's analyses by performing two blank injections, at a temperature of 270°C. The volatile components were extracted by the static headspace method. During this step, each of the fibers was exposed for 10 min in the headspace of the truffle with the vial maintained at 20°C (in a thermostatically controlled analysis room). The adsorbed molecules were desorbed by introducing the SPME fiber into the injector of a 3800 gas chromatograph (Varian, Inc., Palo Alto, CA, USA).
The injector, in splitless mode for 2 min, was set at 260°C. The volatile compounds were separated on a CP-Sil 8 CB low-bleed/MS capillary column with a 5% phenyl/95% dimethylpolysiloxane stationary phase (30 m long, i.d. 0.25 mm, film thickness 0.25 µm; Chrompack Varian, Inc., Palo Alto, CA, USA). The carrier gas was helium and the column flow was constant (1 mL/min).
The GC oven temperature program was 30°C, hold 1 min, increase at 1°C/min to 40°C, hold for 2 min, increase at 1°C/min to 60°C, at 8°C/min to 200°C, then at 20°C/min to 250°C, at which temperature it was held for 2 min (total run 55 min).
The analyses were performed using a Saturn 2200 (Varian, Inc.) GC/MS instrument operating in electron ionization mode (EI, internal ionization source; conditions 70 eV, 20 µA, ion trap temperature 180°C) with the ion trap operating in scan mode (scan range from m/z 40–650 at a scan rate of 1 scan/s). Mass calibration was performed using perfluorotributylamine.
Peak identification was based on mass spectral interpretation and on the standard library NIST ′98 data bank (NIST/EPA/NIH Mass Spectral Library, version 1.6; Gaithersburg, MD, USA).
According to their peak resolution, the areas were either calculated from the total ion current (TIC) or estimated from the integrations performed on selected ions. The resulting peak areas were expressed in arbitrary units of area.
The evaluation of the performance of the fibers was assayed using T. magnatum, the most appreciated fruit body, and thus known to be very rich in flavor.
Choice of SPME fiber
Three different fibers were employed, namely polydimethylsiloxane (PDMS) 100 µm, polydimethylsiloxane/divinylbenzene (PDMS/DVB) 65 µm, and divinylbenzene/Carboxen/polydimethylsiloxane 50/30 µm (DVB/CAR/PDMS). Each analysis was carried out three times. The comparison of the performance of the three fibers was made using T. magnatum samples. The volatile components were extracted by the static headspace method (SHS-SPME). During this step, each of the fibers was exposed for 10 min in the headspace of the truffle, maintaining the vial at 20°C (thermostatically controlled analysis room). The results obtained, at rigorously reproduced temperature and exposure time conditions, are reported in Fig. 1(a). The absorption kinetics were determined for six exposure times (3, 5, 10, 15, 20, 30 min) of the fiber in the headspace. The extraction was carried out at 20°C (controlled temperature). Each measurement was repeated five times. Some samples of T. magnatum were also tested at different heating temperatures (20, 25, 30, 40, 50 and 100°C). The fiber exposure time was 10 min and each measurement was repeated five times.
The efficiency was measured for the terpene compounds extracted from the matrix under study. These results (Fig. 1(b)) show that, for the highest temperature, a decrease in peak areas was observed. Furthermore, for extractions performed at 20°C, better chromatographic reproducibility was achieved, and therefore extraction at this temperature was employed.
In order to determine the optimum sampling time, fiber exposures of 3, 5, 10, 15, 20, and 30 min were tested. As shown in Fig. 1(c), no significant differences were observed for an exposure time longer than 10 min, indicating that, for this exposure time, an equilibrium of analytes is reached among the three system phases, i.e. the coated fiber, the headspace and the sample solution. Consequently, the extraction time was fixed at 10 min.
The spectra obtained from the chromatograms of all samples under study were utilized in order to carry out a discriminant principal component analysis (PCA). Each chromatogram file was converted into an MS file and subsequently processed by Wsearch32 software, in order to obtain an average spectrum of the entire chromatographic run. Only fragment ions in the range m/z 40–150 that stood out from the 5% of total relative abundances of the mass spectra were considered to be carrying useful information. The mass range was extended to m/z 150 because some species possess ions over m/z 100 that could be useful for characterization. A second set of data was used as input in discriminant analysis that of all ions that stood out from the 5% of relative abundance in the range m/z 40–50 (ions characteristic of compounds containing sulfur). The set 5% threshold allowed us to obtain a useful dataset for the subsequent analyses, highlighting only characteristic ions of each species, and removing the common background.
A subset of significant fragment ions (entry F, value 3.84 and removal F value, 2.71 for inclusion and rejection) was obtained by stepwise discriminant analysis (Mahlanobis distance procedure in SPSS software). For each discriminant PCA, a cross-validation test was performed, and the fruit bodies of four unknown species were used to evaluate the discriminating power of the SPME-GC/MS method.
RESULTS AND DISCUSSION
Among the fibers tested, the dual-layer DVB/CAR/PDMS was chosen for our experiments. The DVB/CAR/PDMS fiber is a new type of coating in which porous materials such as Carboxen 1006 (a porous carbon with a surface area of 1200 m2/g) and DVB are suspended in the PDMS polymer. The outer DVB coating captures large molecules, while smaller and more volatile compounds diffuse through the DVB layer and are trapped by the inner Carboxen/PDMS layer. Thus, the dual-layer fiber can efficiently extract a greater range of analytes than other fibers, which is an important factor when analyzing unknowns such as the headspace components of different species of truffle fruit bodies.
Blank runs were conducted, between extractions, with the chosen fiber, to check for carry-over which would cause memory effects and misinterpretation of results. The effects of various physical-chemical parameters on extraction efficiency were studied.
Using the DVB/CAR/PDMS fiber, we observed that an exposure time of 10 min was sufficient to obtain a high signal. We investigated the influence of fiber exposure temperature (20–100°C) on the other two materials. For this reason, an extraction time of 10 min at a temperature of 20°C was selected to estimate the VOCs by SPME-MS (see Figs. 1(b) and 1(c)).
VOCs extracted under the experimental conditions set previously were identified by SPME-GC/ITMS analysis. Figures 2(a)–2(f) show the average spectra of the six analyzed species of Tuber. Nearly 36 VOCs (alkanes, alcohols, esters, aldheydes, ketones, terpenes, etc.) of widely ranging polarity and molecular weight were identified for T. magnatum, 29 VOCs for T. borchii, 34 VOCs for T. driophyllum, 46 VOCs for T. brumale, 40 VOCs for T. mesentericum, and 66 VOCs for T. aestivum. Among these substances the presence of the sulfur-containing compounds can be noted. These compounds play an important role in the aroma of truffles and especially in that of T. magnatum. The average spectra of the chromatograms obtained by SPME-GC/MS show the fingerprints of each species of Tuber under study. The fragment ions at m/z 47, 61, 79 and 83 are characteristic of sulfur-containing compounds, such as dimethyl sulfide, methanethiol, dimethyl disulfide, etc. The fragmentation of VOCs for all compounds of the single species was therefore globally very similar. This is an important finding because the recombination of ionic species from different volatile compounds was limited in the developed method.
The results of the discrimination test are shown in Fig. 3. The sample weight did not significantly influence either the EI mass spectra or the PCA, since the latter was carried out on relative ion abundances.
Using an ion range of m/z 40–50 (Fig. 3(a)) the samples were correctly classified by using all available variables to compute the canonical discriminant function (in entering order: ion 41, 47, 42, 40, 50, 44, 45, 46 and 43). A greater Mahlanobis distance was obtained by using the ion range of m/z 40–150 of the spectra (Fig. 3(b)). In fact, the stepwise discriminant analysis, carried out on this data set, selected the fragment ions 79, 53, 43, 149, 67, 55, 58, 112, 68, 83, 135, 82, 123, 57, 61 and 46 as the best variables to estimate the canonical discriminant function. The molecular origin of these fragment ions can be found in several compounds identified in the headspace by SPME-GC/MS. The fragment ion at m/z 57 is preponderant in the spectra of alcohols, esters and hydrocarbons and m/z 43 is characteristic of aldehydes, ketones and hydrocarbons. The fragment ion at m/z 58 characterizes compounds such as 3-methylbutanal, pentanal and hexan-2-one. The fragment ion at m/z 55 is observed in the spectra of octen-3-one, pentan-2-one, and 3-methylbutanol. The fragment ions at m/z 61, 79 and 83 are characteristic of sulfur-containing compounds. Fragment ions at m/z 67, 68 and 53 are present in the spectra of terpenes. Finally, the ion at m/z 123 is present in the spectra of benzene and phenol derivatives.
In both cases the cross-validation test correctly classified the fruit bodies of the six species under study with a value of 100%.
The PCA discriminant model, obtained as mentioned above, was used to interrogate the EI mass spectra of four different fruit bodies of unidentified species of the genus Tuber. The results of this process are shown in Fig. 3(c), and permitted us to classify the four unknown fruit bodies. Two of the fruit bodies are classified as T. borchii species and the other two as T. aestivum species. These samples, analyzed by GC/MS, were then identified by molecular methods. The specificity of both species has been confirmed, validating the method developed in this study. Tuber borchii was identified by specific PCR (Fig. 4), and T. aestivum species by the sequence of the ITS regions.
From the obtained results, we can ascertain that SPME-MS together with a PCA represents a new solution toward fruit body classification of truffle species.
The static headspace solid-phase microextraction gas chromatography/mass spectrometry method together with multivariate analysis offers a new and effective means for characterizing truffles through the analysis of their volatile fraction. The method described makes it possible to distinguish between different species of truffles. The data analysis is simple and needs no statistical pre-treatment. Also, the method chosen for the analysis by SPME-GC/MS reduces thermal, mechanical, and chemical modification of the samples.
The method developed could be extended to other truffle species and to different ripening stages, permitting the development of a new tool useful in quality control in the trade of these goods.