Component features, odor‐active volatiles, and acute oral toxicity of novel white‐colored truffle Tuber japonicum native to Japan

Abstract Component analysis of a novel white‐colored truffle native to Japan, Tuber japonicum, was performed to determine its characteristic features. The analysis of odor‐active volatile compound showed a high contribution of 1‐octen‐3‐ol and 3‐methyl‐2,4‐dithiapentane to the odor of T. japonicum. Although 2,4‐dithiapentane is a key odorant of well‐known white truffle T. magnatum, 3‐methyl‐2,4‐dithiapentane was detected from the ripe T. japonicum. The chemical components of T. japonicum showed no clear difference with those of edible truffles T. magnatum and T. melanosporum. It was rich in crude protein, crude fiber, and minerals (especially potassium), and low in crude fat. Glutamine and glutamic acid were detected in T. japonicum as free amino acids, while T. magnatum contained a large amount of alanine. Acute oral toxicity tests showed no abnormality, with an LD50 value of over 2000 mg/kg under the test conditions. The present study may support future market distribution of T. japonicum as a high‐class foodstuff.

As T. japonicum is little known in Japan, this white-colored truffle has a limited culinary history, having been tasted by only a few people in a limited area. A large proportion of theoretically edible mushroom species have low consumption values, with many having an unverified edibility status (Kotowski, 2016;Sitta & Davoli, 2012).
Although many species belonging to genus Tuber are considered edible, no information is available on T. japonicum. Japan has a tradition of picking edible wild mushrooms, with many mushrooms enthusiasts entering mountainous areas to collect wild mushrooms in the season. White-colored truffle T. japonicum has attracted attention as a potential food material with commercial value along with increasing interest in distinctive local ingredients. To achieve future utilization as a food material, the characterization and safety confirmation of T. japonicum will be indispensable.
In this study, the chemical composition of Japanese white-colored truffle T. japonicum was compared with those of two wellknown edible truffles, namely white truffle T. magnatum and black truffle T. melanosporum. To demonstrate its characteristic features, the free amino acid compositions and aroma compounds of Japanese white truffle collected from different locations were compared with those of T. magnatum. Furthermore, an acute oral toxicity test was conducted as the first step in safety confirmation for the future market distribution of T. japonicum.

| Samples
Fresh ascomata samples of T. magnatum from Italy and T. melanosporum from France were purchased in November 2017. Species of these market-purchased samples was confirmed by analyzing the nucleotide sequences of the internal transcribed spacer (ITS) region of rDNA, as follows. Genomic DNA from each sample was extracted using a DNeasy Plant Mini Kit (Qiagen) according to the manufacturer protocol. The polymerase chain reaction (PCR) amplification was performed with the T. melanosporum -specific T.mel and T.mel_rev primer pair (Bonito, 2009), T. magnatum -specific P7 and M3 primer pair (Mello, Garnero, & Bonfante, 1999), and fungal universal ITS1F (Gardes & Bruns, 1993) and ITS4 (White, Bruns, Lee, & Taylor, 1990)

| Mineral element analysis
Mineral measurements were conducted using the facility service at the Japan Food Research Laboratory. Mineral analysis of Na and K was conducted on samples digested with 1% hydrochloric acid by atomic absorption photometric analysis using a SpectrAA 240FS spectrometer (Agilent Technologies). For P, Fe, Ca, Mg, Zn, and Mn determination, the ash was hydrolyzed with 20% hydrochloric acid, and contents were analyzed by inductively coupled plasma (ICP) emission spectrometry using an Agilent 5100VDV spectrometer.
Each mineral content was measured twice within the experimental error range of 5%, and an average value was obtained.

| Odor-active volatile analysis using GColfactometry (GC-O)/MS
Odor-active volatile analysis was conducted using the facility service at the Hitachi Chemical Techno Service Co., Ltd. using a GERSTEL MPS-2 autosampler (Gerstel) with a headspace and dynamic headspace system (DHS) option, a 7890 GC equipped with a 5977A Mass Selective Detector (Agilent Technologies), and a GERSTEL ODP3 olfactory detection port for GC-O. Each sample (0.14-0.18 g) was sealed in a 10-ml vial bottle. For DHS analysis conditions, Tenax TA was used as an adsorbent at a trap temperature of 30°C. Each sample vial was incubated at 40°C for 5 min, purging with nitrogen (10 ml) at 10 ml/min, followed by a dry purge of nitrogen (100 ml, 50 ml/min) to reduce water content.
Desorption of the compound was conducted using a thermal desorption unit (TDU) and programmable temperature vaporizing (PTV) injector inlet system. TDU analysis conditions were programmed to hold the temperature at 20°C for 0.5 min and then ramp to 250°C at a rate of 720°C/min, with a final time of 5 min. The PTV system was programmed to hold the temperature at 10°C for 0.5 min and then ramp to 250°C at a rate of 12°C/min, with a final time of 5 min. Compounds were separated using a capillary column (DB-WAX, 60 m × 0.25 mm i.d., 0.25-µm film thickness, J&W Scientific). The carrier gas was helium at a flow rate of 2.11 ml/min. The oven temperature was held at 40°C for 5 min and raised to 250°C at a rate of 12°C/min, with a final time of 5 min. The compounds were identified by matching their mass spectra with NIST 14 database software. Odor intensity was scored at six levels (from 0 to 5) by a trained panelist, as follows: 0, odorless; 1, perceptible odor; 2, weak smell with distinguishable odor; 3, easily perceptible odor; 4, strong odor; and 5, intense odor.

| Volatile compounds analysis by solid-phase microextraction (SPME)-GC/MS
Volatile compounds in truffle samples were extracted using a SPME fiber coated with 50/30 µm divinylbenzene/carboxen/polydimethylsiloxane (Supelco Co.). Frozen truffle sample was crushed by a mill, and headspace gas of 0.4 g of pulverized sample was extracted at 50°C for 20 min. The SPME fiber was then injected to the GC at 250°C in split mode (1:20 split ratio). The analysis of volatile compounds was car-  Values are averages of two determinations within the error range of 5%, in g/100 g dry weight basis.
thickness, J&W Scientific). Helium was used as the carrier gas at a flow rate of 2.11 ml/min, and the column temperature was held at 40°C for 5 min and raised to 250°C at a rate of 5°C/min, with a final time of 5 min. The injector and detector temperatures were set at 250 and 230°C, respectively. The compounds were identified by matching their mass spectra with NIST 14 and FFNSC3 database libraries. Retention indices were calculated after analyzing C8-C20 n-alkane series (Merck) under the same chromatographic conditions.

| Acute oral toxicity
Acute oral toxicity tests against rats were conducted with reference to the OECD guideline for testing of chemicals 420 ( Each test substance was prepared at the FFPRI and stored in a deep freezer immediately after preparation, as described previously.
Test substances were sent to BSRC in a frozen state. Female Slc: Wistar [SPF] rats (7 weeks old; body weight range, 110-150 g) were purchased from Japan SLC Inc. and housed in a climate-and light-controlled room with 12 hr each of light and dark. The test substance was administered to 8-to 9-week old rats (body weight range, 138-148 g) that had been fasted overnight before administration. Starting doses of 300 mg/kg and 2000 mg/kg were selected for the sighting study owing to the lack of safety information and according to the guideline.
A total of five animals were used in the main study. The test suspension was administered using a teflon gastric tube. For all animals, the general condition and body weight were observed for 14 days after administration, and all surviving animals were subjected to gross necropsy.

| Chemical composition
The proximate compositions of the three kinds of truffles are shown in   study were close to that reported for Tir. nivea (13.0%). The crude fat contents of the four truffle samples herein were in the range of 0.95%-1.34%, which were lower than the lipid content of T. melanosporum (7.8%) reported by Harki et al. (2006) using the method of Weete, Ssancholle, and Motant (1983). The crude fat contents of Tir.
nivea from Saudi Arabia, as reported by Sawaya et al. (1985), ranged from 2.81% to 7.42%, despite using the same solvent, diethyl ether, as outlined in AOAC (1980) to compare the contents with that of Tir.
nivea. The analytical results of fat contents in T. latisporum, T. subglobosum, and T. pseudohimalayense were 2.23%-2.55% with petroleum ether using a Soxhlet apparatus (Yan, Wang, Sang, & Fan, 2017). To confirm the crude fat contents of the samples used in this study, the three truffle samples were processed by Soxhlet extraction with petroleum ether. The resulting crude fat contents were 1.20% (T. magnatum), 1.11% (T. melanosporum), and 0.97% (T. japonicum Hyogo).
As a result, the samples used in this study showed lower crude fat contents than those of truffles reported previously. Table 3 shows the minerals contents in the four truffle samples examined. Among these, the potassium content was markedly high, followed by the phosphorus content, as previously reported for other truffles (Harki et al., 2006;Sawaya et al., 1985). The potassium contents of T. japonicum were 683.9 and 650.1 mg/100 g dry weight, and those were lower than that of T. magnatum and T. melanosporum in this study. The potassium contents in this study were considerably lower than that of white truffle Tir. nivea (1734 mg/100 g dry weight, Sawaya et al., 1985). According to Harki et al. (2006), the potassium contents, expressed as % dry weight, of T. melanosporum ranged from 3.17% to 4.43%, measured using energy dispersive spectrometry. Libyan truffle Terfezia boudieri (Chatin) had a potassium content of 9,960 mg/kg on a moisture-free basis (Ahmed, Mohamed, & Hami, 1981) by flame photometry. One reason for the lower potassium content in this study seemed to be the experimental conditions. Among other minerals, the sodium contents of T. japonicum were higher than that of T. magnatum and T. melanosporum, while its calcium contents were low.
The chemical composition of T. japonicum was found to be rich in crude protein, crude fiber, minerals, and low in crude fat, with no clear differences compared with the two controls. Although the nutritional value might vary with growth conditions, the characteristics of T. japonicum seemed not to deviate from those of the other two edible truffles of genus Tuber tested in this study.

| Amino acid composition
Free amino acids detected in two samples of T. japonicum (Hyogo and Tochigi) and one sample of T magnatum are shown in Table 4.
The amino acid components that exceeded contents of 10 µmol/g dry weight are listed. The most abundant amino acid detected from T. japonicum was glutamine, followed by glutamic acid. Alanine and asparagine were also detected in the extract of T. japonicum. In contrast, a significantly larger amount of alanine, over six times higher than that in T. japonicum, was detected from T. magnatum.
For T. melanosporum at maturation stage V, glutamine was reported to be the most abundant free amino acid after cysteine, while alanine was the most abundant at ripening stage IV (Harki et al., 2006). Alanine and glutamine are known to be related to sweetness, while glutamic acid is the source of umami taste (Bachmanov et al., 2016;Schiffman & Sennewald, 1981). This combination of free amino acids might affect the characteristic flavor of each truffle.

| Volatile compounds
Based on a library search identification of their mass spectra, 14, 26, and 19 compounds were detected in T. japonicum Hyogo, Tochigi 1, and Osaka samples by GC-olfactometry/MS analyses, respectively (Table 5). In the three T. japonicum samples, 1-octen-3-ol was strongly detected. This component was also marked as level 5 in the sensory test, having the most intense perceptible odor. In addition to 1-octen-3-ol, 3-methyl-2,4-dithiapentane (1,1-bis(methylthio)ethane), a sulfur volatile component, was also detected in the Tochigi 1 sample at an odor intensity of level 5 (Table 5). 3-Methyl-2,4-dithiapentane was also detected at a level 3 odor intensity from the Osaka sample but not detected from the Hyogo sample. The ascomata samples from Tochigi 1 and Osaka were collected in December, while the Hyogo sample was collected in November. Delaying the collection time might cause advanced ripening and change the composition of volatile components. Other sulfur compounds, such as dimethyl sulfide, dimethyl disulfide, and dimethyl trisulfide, were also detected from the Tochigi 1 and Osaka samples. These sulfur compounds are found in many truffles and are considered to be characteristic aroma components especially in black truffles (Rubini, Belfiori, Riccioni, & Paolocci, 2012;Spivallo, Ottonello, Mello, & Karlovsky, 2011). The most characteristic detected odor of white truffle T. magnatum was 2,4-dithiapentane, which had a level 5 odor intensity in this study as shown in Table 5.

| Acute oral toxicity
The acute oral toxicity of T. japonicum was investigated using a fixed-dose procedure. As truffles are a food ingredient that can be eaten raw, the raw truffle was pulverized in sterilized water using a homogenizer and frozen until administered. In the sighting study, no abnormalities were observed in individuals with doses of both 300 mg/kg and 2000 mg/kg, that let four additional test animals at 2000 mg/kg for the main study.

| CON CLUS ION
Distinctive characteristics of Japanese white-colored truffle T. japonicum, accompanied by the generality of its chemical composition. Regional foodstuffs need generality as an ingredient and distinguishable qualities from others. White-colored truffle T. japonicum seems to satisfy these demands and has a potentially high market value that may drive future market distribution.

ACK N OWLED G M ENTS
This work was supported by a grant from the Ministry of Agriculture, Forestry, and Fisheries of Japan, entitled "Technology development for the optimal use of forest resources." The authors thank the open facility service of University of Tsukuba for support with amino acid detection. We thank Simon Partridge, PhD, from Edanz Group (www.edanz editi ng.com/ac) for editing a draft of this manuscript.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interest.

E TH I C A L A PPROVA L
This study involving the rat procedure was approved by the Animal