TY - JOUR
AU - Touhara,, Kazushige
AB - Abstract In this study, we examined the mode of metabolism of food odorant molecules in the human nasal/oral cavity in vitro and in vivo. We selected 4 odorants, 2-furfurylthiol (2-FT), hexanal, benzyl acetate, and methyl raspberry ketone, which are potentially important for designing food flavors. In vitro metabolic assays of odorants with saliva/nasal mucus analyzed by gas chromatography mass spectrometry revealed that human saliva and nasal mucus exhibit the following 3 enzymatic activities: (i) methylation of 2-FT into furfuryl methylsulfide (FMS); (ii) reduction of hexanal into hexanol; and (iii) hydrolysis of benzyl acetate into benzyl alcohol. However, (iv) demethylation of methyl raspberry ketone was not observed. Real-time in vivo analysis using proton transfer reaction-mass spectrometry demonstrated that the application of 2-FT and hexanal through 3 different pathways via the nostril or through the mouth generated the metabolites FMS and hexanol within a few seconds. The concentration of FMS and hexanol in the exhaled air was above the perception threshold. A cross-adaptation study based on the activation pattern of human odorant receptors suggested that this metabolism affects odor perception. These results suggest that some odorants in food are metabolized in the human nasal mucus/saliva, and the resulting metabolites are perceived as part of the odor quality of the substrates. Our results help improve the understanding of the mechanism of food odor perception and may enable improved design and development of foods in relation to odor. cross adaptation, enzyme, metabolism, mucus, odorant receptors, olfaction Introduction Odor is an essential element of food, contributing to its palatability along with taste and texture. More than 10 000 compounds have been identified as volatile odorants in regularly consumed foods (Nijssen et al. 2016). Among them, a relatively small number (266) of key food odorants (KFOs) are thought to constitute the framework of food odors (Dunkel et al. 2014). Therefore, analyses focusing on the reception of these KFOs in the human nasal cavity are important for understanding how we perceive and recognize food odors. Odorant reception occurs because of the binding of odorants to approximately 400 types of human olfactory receptors (ORs) expressed by olfactory neurons (Niimura and Nei 2003; Niimura et al. 2018). However, prior to binding, odorants must dissolve in the nasal mucus (Poivet et al. 2016). The Bowman’s gland, a secretory gland in the olfactory mucosa, contains metabolic enzymes, which are involved in metabolizing exogenous substances (Nef et al. 1989; Zupko et al. 1991; Ben-Arie et al. 1993; Ling et al. 2004; Casado et al. 2005; Débat et al. 2007; Mörtstedt et al. 2013; Thiebaud et al. 2013; Yoshikawa et al. 2018). It is also known that the nasal respiratory mucosa containing secretory glands and goblet cells, as well as the mucus secreted by these components, contain numerous metabolic enzymes such as cytochrome P450, esterase, reductase, and oxidase (Dahl and Hadley 1991; Kaliner 1991). Recent in vivo and in vitro animal studies have demonstrated that these metabolic enzymes metabolize odorant molecules and could affect the perception of olfactory stimulus (for review, Rychlik 2017; Schilling 2017). In our previous study, the differences between in vitro OR activity and in vivo olfactory bulb firing pattern in mice indicated that modification of odor components occurs in the nasal mucus layer (Oka et al. 2006). The enzymatic activity of the mucus in metabolizing aldehydes into alcohols and hydrolyzing esters into alcohols and carbonates has also been demonstrated (Nagashima and Touhara 2010). Murine olfactory bulb firing patterns and behavioral studies indicated that metabolic structural transformation affects odor perception (Nagashima and Touhara 2010). Robert-Hazotte et al. established ex vivo assay system in which the odorant metabolism and production of volatile metabolites in the explant of olfactory mucosa can be monitored in real time (Robert-Hazotte et al. 2019b). Using this system, they demonstrated that there was a rapid metabolism of mammary pheromone (MP) in the mucosa (Robert-Hazotte et al. 2019a). Furthermore, the reduction of the MP metabolism by in vivo mucus washing modulates the newborn rabbit behavioral responsiveness to the MP, suggesting that the odorant metabolism in the nasal mucus can have a direct influence on olfactory perception and associated behavior (Robert-Hazotte et al. 2019a). In in vitro studies, Asakawa et al. showed that coexpression of CYP1a2 in OR-expressing cell culture modified the ligand-OR response, especially the response of MOR161-2 by acetophenone (Asakawa et al. 2017). Similarly, Kida et al. demonstrated that coexpression of carboxyl esterase 1d together with the ORs in culture changed the activation profiles of ORs by specific odorants (Kida et al. 2018). We hypothesized that odorants in food are similarly metabolized in human nasal mucus and that each food odor is recognized as a mixture of the original odor component and its metabolites. When humans eat, they detect not only orthonasal odors entering from the nostrils but also retronasal odors, which enter the nasal cavity through the oral cavity and pharynx (Buettner et al. 2001; Linforth 2002; Hodgson 2003; Frasnelli 2005). Retronasal odors contribute to the palatability and flavor of food; thus, the metabolism of the odorants in the intraoral saliva can also affect the odor of the food content reaching the olfactory mucosa (Linforth 2002; Buettener 2017). Previous studies examined the metabolism of odorants in the human nasal cavity and/or oral cavity, and the following 4 reaction types were reported (Buettner 2002a, 2002b, Itobe et al. 2009; Schilling 2010): (i) methylation of thiols; (ii) reduction of aldehydes; (ii) hydrolysis of esters; and (iv) demethylation of methoxy groups. However, the mode of odorant metabolism in the human nose remains unclear. Furthermore, whether this metabolism affects human odor reception is unknown. In the present study, we examined the metabolic profile of odorants in the human nasal cavity and oral cavity. The in vitro metabolic activity of odorants in the saliva and nasal mucus was analyzed by gas chromatography mass spectrometry (GC/MS). Additionally, real-time analysis of metabolites in vivo was carried out by proton transfer reaction-mass spectrometry (PTR-MS) to evaluate whether the metabolism of odorants by the saliva/nasal mucus is rapid enough to affect the perception of inhaled odors. To evaluate whether the concentration of metabolites in the exhaled air is above the perception threshold, we prepared gas samples of the metabolite molecules at several concentrations near the concentration found in the exhaled air and tested if the subjects could perceive the odors in these gas samples. Next, we examined how each odorant and its metabolites activated human ORs. To achieve this, we measured the activity of all human ORs against an odorant (2-furfurylthiol [2-FT]), its metabolite (furfuryl methylsulfide [FMS]), and other related odor compounds in an in vitro assay and selected a set of ORs activated by 2-FT or FMS or both, as well as a set of odorants by which these ORs were activated. Finally, we performed a cross-adaptation study by presenting these sets of odorants in a specific order to test if FMS generated from 2-FT affected the perception of the 2-FT odor by modulating the OR activation profile. The goal of this study was to understand how the intranasal metabolism of odors is involved in human olfactory recognition. Materials and methods This study was approved by the Study Review Committee at Ajinomoto Co., Inc. (Tokyo, Japan; no. 2015–028) and Institutional Review Board at the University of Tokyo Hospital (no. 11114) and conducted according to the Declaration of Helsinki. All participants were recruited by Ajinomoto Co., Inc. and provided informed consent for this study. Odorants and reagents We selected the following 4 odorants among KFOs for metabolism analysis: 2-FT, hexanal, benzyl acetate, and methyl raspberry ketone (4-(4-methoxyphenyl) butane-2-one). 2-FT, which has a coffee-like odor, is the most frequently detected thiol compound in foods (Dunkel 2014). Its detection threshold is as low as 0.04 ppb (Fors 1988). Hexanal is an aldehyde compound with a grassy odor. It is the fourth most frequently detected food odorant molecules in the aldehyde group after methional, 2- and 3-methylbutanal, and (E, E)-3, 4-decadienal (Dunkel 2014). In a previous study, aldehydes were shown to undergo reduction to alcohols by metabolism in the murine olfactory mucosa (Nagashima and Touhara 2010). In the ester group, benzyl acetate is important for designing food flavors as a fruit and/or sweet flavoring agent (Mosciano 1990; McGinty 2012; Murnane 2013) and was appropriate for use in our study design because of its high sensitivity in GC/MS. Methyl raspberry ketone was selected because it has been suggested to undergo metabolism to generate demethylated metabolites (Schilling 2010). The details of the odorants used are listed in Table 1. In addition to these, the following reagents were used: Tris (2-carboxyethyl) phosphine hydrochloride (TCEP) purchased from Tokyo Chemical Industry Co., Ltd (T1656); S-(5′-adenosyl)-L-methionine chloride dihydrochloride (SAM) purchased from Sigma-Aldrich (A7007); Tris buffer purchased from Nacalai Tesque, Inc. (35436-01); and the Gentest NADPH Regenerating System (451220) purchased from Corning, Inc. (Corning). Table 1. Test compounds Compound Product no. CAS no. Molecular weight Manufacturer 2-FT W249300 98-02-2 114.17 Sigma-Aldrich Hexanal W255726 66-25-1 100.16 Sigma-Aldrich Benzyl acetate W213500 140-11-4 150.17 Sigma-Aldrich 4-(4-methoxyphenyl)-2-butanone M0756 104-20-1 178.23 Tokyo Chemical Industry Co., Ltd FMS F0245 1438-91-1 128.19 Tokyo Chemical Industry Co., Ltd DMS 132–05913 75-18-3 62.13 FUJIFILM Wako Pure Chemical Corporation 2-PP A2278 2294-76-0 149.24 Tokyo Chemical Industry Co., Ltd Furfural W248924 98-01-1 96.09 Sigma-Aldrich Compound Product no. CAS no. Molecular weight Manufacturer 2-FT W249300 98-02-2 114.17 Sigma-Aldrich Hexanal W255726 66-25-1 100.16 Sigma-Aldrich Benzyl acetate W213500 140-11-4 150.17 Sigma-Aldrich 4-(4-methoxyphenyl)-2-butanone M0756 104-20-1 178.23 Tokyo Chemical Industry Co., Ltd FMS F0245 1438-91-1 128.19 Tokyo Chemical Industry Co., Ltd DMS 132–05913 75-18-3 62.13 FUJIFILM Wako Pure Chemical Corporation 2-PP A2278 2294-76-0 149.24 Tokyo Chemical Industry Co., Ltd Furfural W248924 98-01-1 96.09 Sigma-Aldrich Open in new tab Table 1. Test compounds Compound Product no. CAS no. Molecular weight Manufacturer 2-FT W249300 98-02-2 114.17 Sigma-Aldrich Hexanal W255726 66-25-1 100.16 Sigma-Aldrich Benzyl acetate W213500 140-11-4 150.17 Sigma-Aldrich 4-(4-methoxyphenyl)-2-butanone M0756 104-20-1 178.23 Tokyo Chemical Industry Co., Ltd FMS F0245 1438-91-1 128.19 Tokyo Chemical Industry Co., Ltd DMS 132–05913 75-18-3 62.13 FUJIFILM Wako Pure Chemical Corporation 2-PP A2278 2294-76-0 149.24 Tokyo Chemical Industry Co., Ltd Furfural W248924 98-01-1 96.09 Sigma-Aldrich Compound Product no. CAS no. Molecular weight Manufacturer 2-FT W249300 98-02-2 114.17 Sigma-Aldrich Hexanal W255726 66-25-1 100.16 Sigma-Aldrich Benzyl acetate W213500 140-11-4 150.17 Sigma-Aldrich 4-(4-methoxyphenyl)-2-butanone M0756 104-20-1 178.23 Tokyo Chemical Industry Co., Ltd FMS F0245 1438-91-1 128.19 Tokyo Chemical Industry Co., Ltd DMS 132–05913 75-18-3 62.13 FUJIFILM Wako Pure Chemical Corporation 2-PP A2278 2294-76-0 149.24 Tokyo Chemical Industry Co., Ltd Furfural W248924 98-01-1 96.09 Sigma-Aldrich Open in new tab Collection and preparation of human saliva and mucus Saliva The participants rinsed their mouths with 20 cc of water 3 times. They then spat the saliva into a plastic tube. The collected saliva was centrifuged for deforming. Nasal mucus (superior part) To obtain mucus from the superior part of the nasal cavity, we irrigated the olfactory cleft with saline. The participants were placed in the prone position, with their knees bent, and the top of the forehead on the floor (Figure 1). One milliliter of saline at 37 °C was introduced into the nostril on one side and left for 10 min. The participant then stood up, and the liquid that exited the nostril was collected into a tube and then immediately frozen in liquid nitrogen for storage. This sample is considered to be the mixture of olfactory and respiratory mucus. Figure 1. Open in new tabDownload slide Head position of the subject when the nasal mucus (superior part) was collected. Participants took the prone position, with knees bent, and the top of the forehead on the floor. Figure 1. Open in new tabDownload slide Head position of the subject when the nasal mucus (superior part) was collected. Participants took the prone position, with knees bent, and the top of the forehead on the floor. Nasal mucus (inferior part) Surgical sponges (Merocel, Medtronic Japan), 1 mm in diameter and 2-cm long, were placed for 10 min in both the right and left nostrils to absorb mucus from the respiratory mucosa. Immediately after removal, the sponges were frozen in liquid nitrogen for storage. At the time of use, the sponges were thawed on ice and the mucus was collected into a tube. This sample is considered to be almost pure respiratory mucus. Proteomics See Supplementary Materials and methods. Odorant compound solution For GC/MS, each odorant compound solution was prepared as follows. 2-Furfurylthiol 2-FT was prepared as 1-M stocks in ethanol, as 2-FT forms dimer in dimethyl sulfoxide (DMSO). The 2-FT reaction mixture was adjusted to the following concentrations: 100 mM Tris-HCl (pH 7.6), 5 mM TCEP, 1 mM SAM, and 500 μM 2-FT. TCEP is necessary to prevent the dimerization of 2-FT. SAM functions as a methyl donor. Hexanal Hexanal was prepared as 1-M stocks in DMSO. The hexanal reaction mixture was prepared in the following concentrations: 2.6 mM NADP+, 6.6 mM glucose-6-phosphate, 6.6 mM MgCl2, 0.8 U/mL glucose-6-phosphate dehydrogenase, 0.1 mM sodium citrate, and 100 μM hexanal. Our unpublished observation revealed that addition of NADP+ promoted the reduction of aldehyde (data not shown). Benzyl acetate Benzyl acetate was prepared as 1-M stocks in DMSO. To prepare the reaction mixture, this stock solution was diluted to 100 μM with saline. Methyl raspberry ketone Methyl raspberry ketone was prepared as 1-M stock in DMSO. The reaction mixture was diluted with saline to a concentration of 100 μM. Analysis of enzymatic activity Ten microliters of each odorant compound solution were mixed with 10 μL of (i) saline, (ii) saliva, nasal mucus (superior part), or nasal mucus (inferior part), which had been heat-deactivated at 100°C for 10 min, and (iii) the same saliva or mucus. The resulting mixture was placed in a 0.1-mL inactive glass insert vial (1030–51122; GL Sciences, Inc.). The tube was sealed with Parafilm and the contents were allowed to react in a warm bath at 37 °C for 30 min. After the reaction, the tube was cooled on ice and the contents were transferred to a glass GC/MS vial for subsequent analysis. GC/MS analysis methods A gas chromatograph 6890N with a mass selective detector 5973C (Agilent Technologies, Inc.) equipped with modular accelerated column heating (MACH) system (Gerstel K.K) using a low thermal mass column (Agilent Technologies) was used for the GC-MS measurements. The head space-solid phase microextraction experiment was performed using a 50/30-μm divinylbenzene/carboxen/polydimethylsiloxane (solid phase microextraction) fiber from Sigma-Aldrich. The sample was transferred into a 20-mL headspace vial with a silicon/polytetrafluoroethylene septum and steel cap and introduced to the GC under the following head space-solid phase microextraction conditions: preheating for 1 min at 50 °C; headspace adsorption for 30 min at 50 °C without agitation; desorption for 90 s at 200 °C in the injection port as split mode (5:1). A DB-Wax column (low thermal mass column, 10 m length × 0.18 mm ID, 0.3 μm film thickness; Agilent Technologies) and helium as the carrier gas were used at a constant linear velocity of 70 cm/s. The oven temperature was programmed to increase from 50 to 230 °C at a rate of 25 °C/min with an initial hold time of 0.6 min and final hold time of 4.2 min. Mass chromatogram measurement was performed in scan mode (m/z 35 350) with an electrospray ionization source (230 °C, 70 eV). The quadrupole temperature was 150 °C. The retention times under these conditions and m/z values chosen to detect the extracted ion chromatogram are shown in each figure. The identified compounds were quantified by calculating the relative intensity of peak areas of each compound with those of the respective standard solution (Supplementary Figure 1). Real-time analysis of exhaled air using PTR-MS Measurements were carried out using a commercial PTR-QiTOF instrument (Ionicon Analytik GmbH). The Nosespace Air Sampling Extension (N.A.S.E Ionicon Analytik) apparatus was directly attached to the PTR-ToF-MS. All measurements were carried out under drift tube conditions of 500 V, drift pressure of 3.8 mbar, temperature of 80 °C, and E/N value of 70 Td. This relatively low E/N was chosen to minimize fragmentation of the target compounds. Data acquisition was carried out at 2 spectra per second. Before measuring the odor components, measurements of the air in the laboratory were carried out using calibrated mass equipment. All analytical data were corrected with the background value measured immediately before the measurements. The m/z of each compound was set to have no influence of the isotopic contribution: Measurement of 2-FT, FMS, hexanal, and hexanol was made at the peak of 115, 129, 101, and 86 m/z, respectively. Additionally, exhaled air was monitored by acetone detection. Calibration curves were obtained using the signal intensities of standard gas samples by PTR-MS (Supplementary Figure 2). Concentrations of odorants in in vivo study were calculated using this calibration curve. Our preliminary experiment demonstrated that under our in vivo metabolism conditions, only one major metabolite from each substrate was detected by GC-MS (Supplementary Figure 3, i.e., FMS from 2-FT and hexanol from hexanal). Similar results were reported previously (Itobe 2009). Samples of the exhaled air were collected in the following manner (n = 5 for each condition). Air sample passed through the nasal cavity alone (condition (i); Figure 3A) Odor-containing air was prepared by placing 10 mL of the prepared odor solution (200 mg/L for 2-FT and 4000 mg/L for hexanal) in a 50-mL syringe and leaving it to stand for 2–3 min. A total of 45 mL of the headspace air was injected into the left nostril at 9 mL/s for 5 s. Exhaled air from the right nostril was injected directly into the PTR-MS device, and the substrate and metabolites were measured. While the air was being injected, the subject phonated “eeee” so that the soft palate completely separated between the oral and nasal cavities (Rubesin et al. 1988). The time point when the air injection was started was defined as time 0 s and the measurement continued until 15 s. Exhaled air after self-inhalation of the odor-containing air from the nose (condition (ii); Figure 3D) Odor-containing air was prepared by placing 10 mL of the prepared odor solution (200mg/L for 2-FT and 1000 mg/L for hexanal) in a 50-mL syringe and leaving it to stand for 2–3 min. A total of 45 mL of the headspace air was injected into one nostril at once while inhaling, and the subjects held their breath for 5 s until the start of measurement. At time 0 s, the subjects started an exhalation–inhalation cycle through their nose. Inhalation and exhalation periods were set for 3 s each with the aid of metronome. The exhaled air was injected into the PTR-MS device, and the substrate and metabolites were measured. Measurement continued until the end of the 10th cycle. Exhaled air after swallowing odorant solution and nasal breathing (condition (iii); Figure 3G) The subjects held 5 mL of the prepared odorant solution (100 mg/L for 2-FT and 1000 mg/L for hexanal) in their mouth for 5 s, after which they swallowed it. The subject then immediately started an exhalation–inhalation cycle through their nose. The start of the first exhalation was set as time 0 s. Inhalation and exhalation periods were set for 3 s each with the aid of metronome. The subsequent exhaled air of 10 natural nasal breaths was injected directly into the PTR-MS device, and the substrate and metabolites were measured. Evaluation of metabolite odor recognition To determine whether the concentration of metabolites in the exhaled air was above the detection and/or recognition threshold, gas samples of FMS and hexanol, at 20 and 60 ppb and 250 and 750 ppb, respectively, were presented to the subjects (n = 20). These concentrations were determined based on the average peak concentration detected in exhaled air (Figures 2 and 3). The ability or inability to recognize the odor was then recorded using a 4-grade evaluation scale as follows: − (no odor), + (supradetection threshold [odor quality cannot be stated]), ++ (suprarecognition threshold [odor quality can be stated]), and +++ (definite odor [odor quality can be stated definitely]). Figure 2. Open in new tabDownload slide Metabolic transformation of 2-FT, hexanal, benzyl acetate, and methyl raspberry ketone in vitro. (A) Molecular formula of 2-FT (substrate) and FMS (its methylated metabolite). (B)–(D) Metabolic transformation of 2-FT by saliva (B), nasal mucus (superior part) (C), and nasal mucus (inferior part) (D). The peak areas for 2-FT (retention time [RT] = 3.0 min) and FMS (RT = 3.2 min) were obtained from an extraction ion chromatogram (m/z 81). (E) Molecular formula of hexanal (substrate) and hexanol (its reduced metabolite). (F)–(H) Metabolic transformation of hexanal by saliva (F), nasal mucus (superior part) (G), and nasal mucus (inferior part) (H). The peaks for hexanal (RT = 1.2 min) and hexanol (RT = 2.7 min) were identified on the extraction ion chromatogram at m/z 56 and the peak areas were calculated. (I) Molecular formula of benzyl acetate (substrate) and benzyl alcohol (its hydrolyzed metabolite). (J)–(L) Metabolic transformation of benzyl acetate by saliva (J), nasal mucus (superior part) (K), and nasal mucus (inferior part) (L). The peaks for benzyl acetate and benzyl alcohol were identified at m/z 108 from an extraction ion chromatogram at RT of 4.5 and 5.6 min, respectively, and each of the GC/MS peak areas was calculated. (M) Molecular formula of methyl raspberry ketone (substrate) and raspberry ketone (its demethylated metabolite). (N)–(P) Metabolic transformation of methyl raspberry ketone by saliva (N), nasal mucus (superior part) (O), and nasal mucus (inferior part) (P). The peaks for methyl raspberry ketone (RT = 5.7 min) and raspberry ketone (RT = 7.7 min) were identified on the extraction ion chromatogram at m/z 121 and the peak areas were calculated. The concentration of each compound was calculated based on the signal intensities obtained by GC/MS using standard solution samples and are shown on the y axes. Three columns in each graph indicate the following reaction conditions: (i) mixture of test compound and saline; (ii) for each mucus type, mixture of test compound and heat-treated mucus; (iii) for each mucus type, mixture of test compound and mucus. The tests were carried out on 4 subjects. Error bars represent the standard deviation of the mean. Each data point is shown as a dot. Asterisks indicate a significant difference (*P < 0.05, **P < 0.01, ***P < 0.001) compared with the saline control group (Dunnett post hoc test). N.S. indicate not significant (P > 0.05). Figure 2. Open in new tabDownload slide Metabolic transformation of 2-FT, hexanal, benzyl acetate, and methyl raspberry ketone in vitro. (A) Molecular formula of 2-FT (substrate) and FMS (its methylated metabolite). (B)–(D) Metabolic transformation of 2-FT by saliva (B), nasal mucus (superior part) (C), and nasal mucus (inferior part) (D). The peak areas for 2-FT (retention time [RT] = 3.0 min) and FMS (RT = 3.2 min) were obtained from an extraction ion chromatogram (m/z 81). (E) Molecular formula of hexanal (substrate) and hexanol (its reduced metabolite). (F)–(H) Metabolic transformation of hexanal by saliva (F), nasal mucus (superior part) (G), and nasal mucus (inferior part) (H). The peaks for hexanal (RT = 1.2 min) and hexanol (RT = 2.7 min) were identified on the extraction ion chromatogram at m/z 56 and the peak areas were calculated. (I) Molecular formula of benzyl acetate (substrate) and benzyl alcohol (its hydrolyzed metabolite). (J)–(L) Metabolic transformation of benzyl acetate by saliva (J), nasal mucus (superior part) (K), and nasal mucus (inferior part) (L). The peaks for benzyl acetate and benzyl alcohol were identified at m/z 108 from an extraction ion chromatogram at RT of 4.5 and 5.6 min, respectively, and each of the GC/MS peak areas was calculated. (M) Molecular formula of methyl raspberry ketone (substrate) and raspberry ketone (its demethylated metabolite). (N)–(P) Metabolic transformation of methyl raspberry ketone by saliva (N), nasal mucus (superior part) (O), and nasal mucus (inferior part) (P). The peaks for methyl raspberry ketone (RT = 5.7 min) and raspberry ketone (RT = 7.7 min) were identified on the extraction ion chromatogram at m/z 121 and the peak areas were calculated. The concentration of each compound was calculated based on the signal intensities obtained by GC/MS using standard solution samples and are shown on the y axes. Three columns in each graph indicate the following reaction conditions: (i) mixture of test compound and saline; (ii) for each mucus type, mixture of test compound and heat-treated mucus; (iii) for each mucus type, mixture of test compound and mucus. The tests were carried out on 4 subjects. Error bars represent the standard deviation of the mean. Each data point is shown as a dot. Asterisks indicate a significant difference (*P < 0.05, **P < 0.01, ***P < 0.001) compared with the saline control group (Dunnett post hoc test). N.S. indicate not significant (P > 0.05). Figure 3. Open in new tabDownload slide Open in new tabDownload slide Real-time analysis of 2-FT and FMS concentrations in the exhaled air. (A) A schematic view of condition (i), in which the air containing the substrate was passed solely through the nasal cavity. Broken line indicates the injected air into the right nostril and solid line indicates the flow-out air from the left nostril. (B), (C) 2-FT and FMS concentration after 2-FT injection in condition (i). PTR-MS charts are shown for 2-FT (B1–B5) and metabolite FMS (C1–C5). Measurements were made for 2-FT with the peak at m/z 115 and for FMS with the peak at m/z 129, and the values were converted to values of concentrations. The data were collected from 5 subjects. Profiles B1 and C1, B2 and C2, B3 and C3, B4 and C4, and B5 and C5 are from subject no. 1, 2, 3, 4, and 5, respectively. The x axes show time (s) and y axes show concentration (ppb). Time zero in the x axis (broken line) indicates the start of 2-FT injection. (D) A schematic view of condition (ii), in which the air containing the substrate was inhaled from the nose and exhaled through the nose. Broken line indicates the inhaled air from the both nostrils and solid line indicates exhaled air from the nostrils. (E)–(F) 2-FT and FMS concentration in the exhaled air in condition (ii). PTR-MS charts show the test compound 2-FT (E1–E5) and the metabolite FMS (F1–F5). The 2-FT peak was measured at m/z 115 and the FMS peak was measured at m/z 129; the values were then converted to values of concentrations. The data were collected from 5 subjects. Profiles E1 and F1, E2 and F2, E3 and F3, E4 and F4, and E5 and F5 are from subject no. 1, 2, 3, 4, and 5, respectively. The x axes show time (s) and y axes show concentration (ppb). Time zero in the x axis (broken line) indicates the start of the first exhalation. (G) A schematic view of condition (iii), in which the solution containing the substrate was swallowed and air was exhaled through the nose. The subjects held the prepared odorant solution in the mouth for 5 s and subsequently swallowed it (broken line). Next, the subjects exhaled the air through the nostrils (solid line). (H)–(I) 2-FT and FMS concentration in the exhaled air in condition (iii). The PTR-MS charts for the test compound 2-FT (H1–H5) and the metabolite FMS (I1–I5) are shown. The 2-FT peak was measured at m/z 115 and the FMS peak was measured at m/z 129, followed by conversion of the values to values of concentrations. The data were collected from 5 subjects. Profiles H1 and I1, H2 and I2, H3 and I3, H4 and I4, and H5 and I5 are from subject no. 1, 2, 3, 4, and 5, respectively. The x axes show time (s) and y axes show concentration (ppb). Time zero in the x axis (broken line) indicates the start of the first exhalation. The signal count for 2-FT and FMS in background room air was 112.1 ± 19.8 and 137.6 ± 25.9, respectively, and the signal count for 2-FT and FMS in the exhaled air of the subjects without administration of substrate was 185.8 ± 22.5 and 188.2 ± 24.7, respectively. Figure 3. Open in new tabDownload slide Open in new tabDownload slide Real-time analysis of 2-FT and FMS concentrations in the exhaled air. (A) A schematic view of condition (i), in which the air containing the substrate was passed solely through the nasal cavity. Broken line indicates the injected air into the right nostril and solid line indicates the flow-out air from the left nostril. (B), (C) 2-FT and FMS concentration after 2-FT injection in condition (i). PTR-MS charts are shown for 2-FT (B1–B5) and metabolite FMS (C1–C5). Measurements were made for 2-FT with the peak at m/z 115 and for FMS with the peak at m/z 129, and the values were converted to values of concentrations. The data were collected from 5 subjects. Profiles B1 and C1, B2 and C2, B3 and C3, B4 and C4, and B5 and C5 are from subject no. 1, 2, 3, 4, and 5, respectively. The x axes show time (s) and y axes show concentration (ppb). Time zero in the x axis (broken line) indicates the start of 2-FT injection. (D) A schematic view of condition (ii), in which the air containing the substrate was inhaled from the nose and exhaled through the nose. Broken line indicates the inhaled air from the both nostrils and solid line indicates exhaled air from the nostrils. (E)–(F) 2-FT and FMS concentration in the exhaled air in condition (ii). PTR-MS charts show the test compound 2-FT (E1–E5) and the metabolite FMS (F1–F5). The 2-FT peak was measured at m/z 115 and the FMS peak was measured at m/z 129; the values were then converted to values of concentrations. The data were collected from 5 subjects. Profiles E1 and F1, E2 and F2, E3 and F3, E4 and F4, and E5 and F5 are from subject no. 1, 2, 3, 4, and 5, respectively. The x axes show time (s) and y axes show concentration (ppb). Time zero in the x axis (broken line) indicates the start of the first exhalation. (G) A schematic view of condition (iii), in which the solution containing the substrate was swallowed and air was exhaled through the nose. The subjects held the prepared odorant solution in the mouth for 5 s and subsequently swallowed it (broken line). Next, the subjects exhaled the air through the nostrils (solid line). (H)–(I) 2-FT and FMS concentration in the exhaled air in condition (iii). The PTR-MS charts for the test compound 2-FT (H1–H5) and the metabolite FMS (I1–I5) are shown. The 2-FT peak was measured at m/z 115 and the FMS peak was measured at m/z 129, followed by conversion of the values to values of concentrations. The data were collected from 5 subjects. Profiles H1 and I1, H2 and I2, H3 and I3, H4 and I4, and H5 and I5 are from subject no. 1, 2, 3, 4, and 5, respectively. The x axes show time (s) and y axes show concentration (ppb). Time zero in the x axis (broken line) indicates the start of the first exhalation. The signal count for 2-FT and FMS in background room air was 112.1 ± 19.8 and 137.6 ± 25.9, respectively, and the signal count for 2-FT and FMS in the exhaled air of the subjects without administration of substrate was 185.8 ± 22.5 and 188.2 ± 24.7, respectively. Measurement of OR activities Construction of a human OR library An OR library was prepared to examine the activity of each of the 403 human ORs toward odorant molecules. In a study by Niimura et al. (2018), 398 putatively functional OR genes with intact coding sequences were identified from the latest version of the human genome (GRCh38). Among them, 3 OR genes had identical amino acid sequences with other OR genes and, therefore, 395 OR genes had unique amino acid sequences. In this study, we used the sequence data of 395 OR genes together with 8 OR genes provided by GenBank. Among the 403 OR clones, 345 were purchased from the TrueClone cDNA collection (OriGene) and Flexi ORF Clone collection (Promega). Of these OR clones, 33 were not amplified by PCR or contained one or several mutations that had not previously been reported. This may be because amplification was inhibited by GC-rich or complex secondary structures in the target DNA. These 33 clones and the remaining 58 clones are not commercially marketed and, thus, were synthesized using GenScript gene synthesis services. The sequences of synthesized ORs were adopted from those with the highest frequency of appearance as reported by Olender et al. (2012). Additional clones were adopted based on sequences containing single-nucleotide polymorphisms (SNPs) with the highest frequency in the 1000 genomes project (The 1000 Genomes Project Consortium 2015). Sixty-three SNPs differing from the reference sequences in GenBank (http://ncbi.nlm.nih.gov/genbank/) in the ORs used in this study are summarized in Supplementary Table 1. Although 4 of the 63 genes shown in the table are annotated as pseudogenes in Genbank, these genes have intact coding sequences (Supplementary Table 1). Therefore, we included them in our analyses to avoid missing any possibly functional ORs. Each OR gene was inserted into the pME18S expression vector with a tag incorporating the first 20 amino acids of human rhodopsin (Rho-pMEs18S; Katada et al. 2005) using Gibson Assembly Master Mix (New England Biolabs) according to the instruction manual. Cell culture and the luciferase assay The luciferase reporter gene assay was performed as previously described (Zhang and Matsunami 2008; Shirasu et al. 2014). The Dual-Glo Luciferase Assay System (Promega) was used to measure firefly and Renilla luciferase activities. A cyclic adenosine monophosphate (cAMP) response element promoter-containing firefly luciferase vector (CRE/luc2P-pGL4.29; Promega) was used to measure the cAMP content dependent on receptor activation. A thymidine kinase promoter-containing Renilla luciferase vector (TK/Rluc-pGL4.74; Promega) was used as an internal control to evaluate cell viability and transfection efficiency. A short form of microbat RTP1, human Ric8B, and human Golf were cotransfected as accessory molecules to promote the functional expression of ORs. To identify ORs responsive to an odorant, HEK293T cells were transfected with 12.5 ng tagged OR vector, 2.5 ng CRE/luc2P, 1.25 ng TK/Rluc, 2.5 ng RTP1S-pcDNA3.1 vector, 1.25 ng Ric8B-pcDNA3.1 vector, and 1.25 ng Golf-pcDNA3.1 using 0.0425 μL Lipofectamine 2000 (Invitrogen) per well. At 24 h after transfection, the medium was replaced with CD293 (Invitrogen) containing 1/10 000 v/v GlutaMAX (Thermo Fisher Scientific) and the cells were stimulated with 100 μM odorant for 4 h at 37 °C and 5% CO2. Luminescence was detected using the FDSS7000EX Functional Drug Screening System (Hamamatsu Photonics). Fold increases were calculated as Luc/hRLuc (Luc [luminescence of firefly luciferase] divided by hRLuc [luminescence of Renilla luciferase]) stimulated with an odorant divided by Luc/hRLuc upon no stimulation. ORs showing an increased response of more than 4-fold of the baseline luminescence were defined as responsive. These responsive ORs were further assayed for dose-dependent analysis. At 24 h after transfection, the cells were stimulated with 1–1000 μM of odorant, except for 2-FT, which was used at a concentration of 1–300 μM because of its cytotoxicity. Methods for calculating normalized values are described in each figure. The relationship between odorant concentration and receptor response was fitted to the Hill equation (eq. 1) using R (version 3.4.4, R Foundation for Statistical Computing) and package drc (Ritz et al. 2015). Intensity=Bottom+Top−Bottom1+eSlope(logConcentration−logEC50) (1) We adopted the criteria established by Mainland et al. (2015). It was defined that a receptor showed dose-dependent response against an odorant if all of the following 4 conditions were satisfied: (i) the 95% confidence intervals of the top and bottom parameters did not overlap, (ii) the standard error of the fitted log EC50 was less than 1 log unit, (iii) an approximate F-test confirmed that the dose-response curve of the receptor was significantly different from that of the control, which was transfected with an empty vector, and (iv) the 95% confidence intervals of the fitted log EC50 of the receptor and the control did not overlap (Ritz et al. 2015; Mainland et al. 2015). Cross-adaptation study Cross adaptation is a physiological process in which if an odorant B is presented after smelling odorant A, the odor of odorant B is weaker than, and/or has a different quality from, the odor of odorant B when presented without prior presentation of odorant A (Tonosaki 1993). It is thought that cross adaptation occurs when odorants A and B partially share activated ORs; if an OR is activated by odorant A and also by odorant B, the OR is adapted and temporarily inactivated by the presentation of odorant A, leading to changes in the activation profile by subsequent odorant B presentation. The cross-adaptation study was carried out on 10 odor sensory evaluation experts, and each test was conducted twice with each panel. Each test compound concentration was set so that its sensory intensity was equivalent to that of 100 mg/L (v/v) aqueous solutions of 2-FT (Table 2). Samples were prepared by placing 40-mL aqueous solutions of each test compound into 50-mL brown glass bottles and leaving them to stand for at least 5 h. We used the terms “pungent, garlic-like odor” and “egg-like sulfurous odor” to describe the odor quality of 2-FT. Each panel member smelled substance A shown in Table 2 for 1 min, held his/her breath, and then smelled substance B, which was 100 ppm 2-FT. Panel members provided answers regarding the intensity of each of the 3 odor qualities (overall odor intensity; intensity of pungent, garlic-like odor; and intensity of egg-like, sulfurous odor) on a visual analog scale (VAS). Each panel was first presented with test no. 1 as shown in Table 2, followed by test nos. 2–5 in a random order for each subject and each time the test was carried out. To allow recovery from the postadaptation state, an interval of at least 2 min was allowed between tests. The bottles presented to the panel members were labeled only with the single letters shown in columns A and B in Table 2. Table 2. Combination of test compounds in the cross- adaptation study Test no. A B 1 M Water C 100 ppm 2-FT 2 P 500 ppm FMS H 100 ppm 2-FT 3 O 500 ppm DMS K 100 ppm 2-FT 4 N 500 ppm 2-PP T 100 ppm 2-FT 5 R 1000 ppm furfural Y 100 ppm 2-FT Test no. A B 1 M Water C 100 ppm 2-FT 2 P 500 ppm FMS H 100 ppm 2-FT 3 O 500 ppm DMS K 100 ppm 2-FT 4 N 500 ppm 2-PP T 100 ppm 2-FT 5 R 1000 ppm furfural Y 100 ppm 2-FT Open in new tab Table 2. Combination of test compounds in the cross- adaptation study Test no. A B 1 M Water C 100 ppm 2-FT 2 P 500 ppm FMS H 100 ppm 2-FT 3 O 500 ppm DMS K 100 ppm 2-FT 4 N 500 ppm 2-PP T 100 ppm 2-FT 5 R 1000 ppm furfural Y 100 ppm 2-FT Test no. A B 1 M Water C 100 ppm 2-FT 2 P 500 ppm FMS H 100 ppm 2-FT 3 O 500 ppm DMS K 100 ppm 2-FT 4 N 500 ppm 2-PP T 100 ppm 2-FT 5 R 1000 ppm furfural Y 100 ppm 2-FT Open in new tab Statistical analysis The results are presented as the mean ± standard deviation for all samples. The concentration of each compound in the in vitro metabolic assay of odorants with saliva/nasal mucus and scores in the cross-adaptation study were statistically analyzed by one-way analysis of variance (ANOVA), followed by Dunnett post hoc test with correction for multiple tests, using R software to compare each subgroup to the corresponding control group. Differences associated with P values <0.05 were considered as statistically significant. Results In vitro analysis of metabolism using saliva and nasal mucus To examine metabolic activity in the saliva, nasal mucus (superior part), and nasal mucus (inferior part), 10 μL of each odorant compound solution was mixed with 10 μL of either (i) saline, or (ii) one of the mucus types that had been heat-deactivated at 100°C for 10 min, or (iii) the same mucus type. The mixed solution was then incubated in a warm bath at 37 °C for 30 min, followed by GC/MS analysis. In the proteome analysis, keratins were not detected in the nasal mucus (inferior part; Supplementary Table 2), suggesting that contamination of the respiratory epithelial cells in the mucus was minimal or absent. 2-Furfurylthiol We examined whether 2-FT (roasted coffee odor) was converted to its methylated form, FMS (onion/garlic odor), in the reaction with each mucus type (Figure 2A–D). This conversion was not observed in saline or heat-treated mucus (Figure 2B–D), suggesting that the reaction was mediated by metabolic enzymes. Under our in vitro assay conditions with saliva, the concentration of 2-FT was decreased to approximately 17% of the initial concentration and FMS was detected in the mixture at a concentration of lower than 1 μM (0.2–0.7 μM; Figure 2B). In the reaction with nasal mucus (superior part), the 2-FT concentration was decreased to approximately 22% of the initial concentration, but FMS was detected in the mixture in only one of the 4 subjects at a concentration of less than 0.1 μM (Figure 2C). In the reaction with nasal mucus (inferior part), the 2-FT concentration was decreased to approximately 15% of the initial concentration and FMS was detected in the mixture in 3 of 4 subjects (Figure 2D). The FMS concentration in the mixture was less than 0.1 μM as observed in the nasal mucus (superior part). ANOVA followed by Dunnett post hoc test showed that the concentration of 2-FT in the saliva, nasal mucus (superior part), and nasal mucus (inferior part) was significantly decreased compared with that in saline control group (P < 0.001). The concentration of FMS in the saliva was significantly increased compared with that in the saline control group (P < 0.01) but was not significantly different from the control group in the nasal mucus (superior part) and nasal mucus (inferior part). Hexanal We tested whether hexanal (grassy odor) was converted to hexanol (herbal/ethereal odor) in the reactions using each mucus type (Figure 2E–H). This transformation was not observed with saline or heat-treated mucus. Under our in vitro assay conditions with saliva, the concentration of hexanal decreased to approximately 4% of the initial concentration, and hexanol was detected in the mixture at concentrations of 1.9–2.5 μM (Figure 2F). In the reaction with nasal mucus (superior part), the hexanal concentration decreased to approximately 16% of the initial concentration. Hexanol was detected in the mixture at concentrations of 1.6–7.0 μM (Figure 2G). In the reaction with nasal mucus (inferior part), hexanal concentration decreased to approximately 1% of the initial concentration, and hexanol was detected in the mixture at concentrations of 6.6–9.5 μM (Figure 2H). Nearly, all hexanal was consumed during the reaction with each type of mucus, but the amount of hexanol detected was low compared with the extent of consumption of hexanal in each mixture. ANOVA followed by Dunnett post hoc test showed that the concentration of hexanal in saliva, nasal mucus (superior part), and nasal mucus (inferior part) was significantly decreased compared with that in the saline control group (P < 0.001). The concentration of hexanol in the saliva, nasal mucus (superior part), and nasal mucus (inferior part) was significantly increased compared with that in the saline control group (P < 0.001 in saliva and nasal mucus [inferior part], P < 0.01 in nasal mucus [superior part]). Benzyl acetate In the reaction with saliva and nasal mucus (inferior part), benzyl acetate (floral/jasmine odor) was metabolized into benzyl alcohol (rose/balsamic odor) (Figure 2I,J,L). In contrast, when nasal mucus (superior part) was used, benzyl alcohol was not formed and no decrease in the substrate, benzyl acetate, was observed (Figure 2K). In the reaction with saliva, the benzyl acetate concentration decreased to approximately 20% of the initial concentration. Benzyl alcohol was detected in the mixture in 3 of 4 subjects and its concentration was 22.4–27.6 μM (Figure 2J). In the reaction with nasal mucus (inferior part), the benzyl acetate concentration was decreased to approximately 38% of the initial concentration and benzyl alcohol was detected in the mixture at the concentration of 15.7–27.9 μM (Figure 2L). This transformation was not observed in saline or heat-treated mucus. ANOVA followed by Dunnett post hoc test showed that the concentration of benzyl acetate in the saliva and nasal mucus (inferior part) was significantly decreased compared with that in the saline control group (P < 0.001). The concentration of benzyl alcohol in the saliva and nasal mucus (inferior part) was significantly increased compared with that in the saline control group (P < 0.01 in saliva and P < 0.001 in nasal mucus [inferior part]). Methyl raspberry ketone The metabolite raspberry ketone was not detected in assays with any of the mucus types (Figure 2M–P) despite the different assay conditions tested. In vivo analysis of exhaled air using PTR-MS If metabolic transformation of this type occurs quickly in the nasal cavity or oral cavity, odor perception can be affected. To test this possibility, in vivo analysis of exhaled air was carried out using PTR-MS, which can be used to measure components in real time. As described in Materials and methods: Real-time analysis of exhaled air using PTR-MS, exhaled air samples under the following 3 different conditions were analyzed: (i) Air samples passed through the nasal cavity alone (Figures 3A, 4A); (ii) Exhaled air after self-inhalation of the odor-containing air from the nose (Figures 3D, 4D); and (iii) Exhaled air after swallowing odorant solution and nasal breathing (Figures 3G, 4G). Figure 4. Open in new tabDownload slide Open in new tabDownload slide Real-time analysis of hexanal and hexanol concentrations in the exhaled air. (A) Schematic view of condition (i), in which the air containing the substrate was passed solely through the nasal cavity. Broken line indicates the injected air into the right nostril and solid line indicates the flow-out air from the left nostril. (B), (C) Hexanal and hexanol concentration after hexanol injection in condition (i). PTR-MS charts for hexanal (B1–B5) and hexanol (F1–F5) are shown. The hexanal peak was detected at m/z 101 and the hexanol peak was detected at m/z 86, and the values were converted to values of concentrations. The data were collected from 5 subjects. Profiles B1 and C1, B2 and C2, B3 and C3, B4 and C4, and B5 and C5 are from subject no. 1, 2, 3, 4, and 5, respectively. The x axes show time (s) and y axes show concentration (ppb). Time zero in the x axis (broken line) indicates the start of hexanal injection. (D) Schematic view of condition (ii), in which the air containing the substrate was inhaled from the nose and exhaled through the nose. Broken line indicates the inhaled air from both nostrils and solid line indicates the exhaled air from the nostrils. (E)–(F) Hexanal and hexanol concentration in the exhaled air in condition (ii). PTR-MS charts of hexanal (E1–E5) and hexanol (L–N) are shown. The peak for hexanal was measured at m/z 101 and that for hexanol was measured at m/z 86, and the values were converted to concentrations. The data were collected from 5 subjects. Profiles E1 and F1, E2 and F2, E3 and F3, E4 and F4, and E5 and F5 are from subject no. 1, 2, 3, 4, and 5, respectively. The x axes of the graphs show time (s) and y axes show concentrations (ppb) for hexanal and hexanol. Time zero in the x axis (broken line) indicates the start of the first exhalation. (G) Schematic view of condition (iii), in which the solution containing the substrate was swallowed and air was exhaled through the nose. The subjects held the prepared odorant solution in the mouth for 5 s and subsequently swallowed it (broken line). Next, the subjects exhaled the air through the nostrils (solid line). (H)–(I) Hexanal and hexanol concentration in exhaled air under condition (iii). PTR-MS charts of the test compound, hexanal (H1–H5), and the metabolite, hexanol (I1–I5), are shown. The hexanal peak was measured at m/z 101 and the hexanol peak was measured at m/z 86, and the measurements were converted to concentrations. The data were collected from 5 subjects. Profiles H1 and I1, H2 and I2, H3 and I3, H4 and I4, and H5 and I5 are from subject no. 1, 2, 3, 4, and 5, respectively. The x axes show time (s) and y axes show concentrations (ppb) of hexanal and hexanol. Time zero in the x axis (broken line) indicates the start of the first exhalation. The signal count for hexanal and hexanol in background room air was 447.2 ± 38.4 and 156.8 ± 21.6, respectively, and the signal count for hexanal and hexanol in the exhaled air of the subjects without administration of substrate was 525.6 ± 12.7 and 123.2 ± 24.3, respectively. Figure 4. Open in new tabDownload slide Open in new tabDownload slide Real-time analysis of hexanal and hexanol concentrations in the exhaled air. (A) Schematic view of condition (i), in which the air containing the substrate was passed solely through the nasal cavity. Broken line indicates the injected air into the right nostril and solid line indicates the flow-out air from the left nostril. (B), (C) Hexanal and hexanol concentration after hexanol injection in condition (i). PTR-MS charts for hexanal (B1–B5) and hexanol (F1–F5) are shown. The hexanal peak was detected at m/z 101 and the hexanol peak was detected at m/z 86, and the values were converted to values of concentrations. The data were collected from 5 subjects. Profiles B1 and C1, B2 and C2, B3 and C3, B4 and C4, and B5 and C5 are from subject no. 1, 2, 3, 4, and 5, respectively. The x axes show time (s) and y axes show concentration (ppb). Time zero in the x axis (broken line) indicates the start of hexanal injection. (D) Schematic view of condition (ii), in which the air containing the substrate was inhaled from the nose and exhaled through the nose. Broken line indicates the inhaled air from both nostrils and solid line indicates the exhaled air from the nostrils. (E)–(F) Hexanal and hexanol concentration in the exhaled air in condition (ii). PTR-MS charts of hexanal (E1–E5) and hexanol (L–N) are shown. The peak for hexanal was measured at m/z 101 and that for hexanol was measured at m/z 86, and the values were converted to concentrations. The data were collected from 5 subjects. Profiles E1 and F1, E2 and F2, E3 and F3, E4 and F4, and E5 and F5 are from subject no. 1, 2, 3, 4, and 5, respectively. The x axes of the graphs show time (s) and y axes show concentrations (ppb) for hexanal and hexanol. Time zero in the x axis (broken line) indicates the start of the first exhalation. (G) Schematic view of condition (iii), in which the solution containing the substrate was swallowed and air was exhaled through the nose. The subjects held the prepared odorant solution in the mouth for 5 s and subsequently swallowed it (broken line). Next, the subjects exhaled the air through the nostrils (solid line). (H)–(I) Hexanal and hexanol concentration in exhaled air under condition (iii). PTR-MS charts of the test compound, hexanal (H1–H5), and the metabolite, hexanol (I1–I5), are shown. The hexanal peak was measured at m/z 101 and the hexanol peak was measured at m/z 86, and the measurements were converted to concentrations. The data were collected from 5 subjects. Profiles H1 and I1, H2 and I2, H3 and I3, H4 and I4, and H5 and I5 are from subject no. 1, 2, 3, 4, and 5, respectively. The x axes show time (s) and y axes show concentrations (ppb) of hexanal and hexanol. Time zero in the x axis (broken line) indicates the start of the first exhalation. The signal count for hexanal and hexanol in background room air was 447.2 ± 38.4 and 156.8 ± 21.6, respectively, and the signal count for hexanal and hexanol in the exhaled air of the subjects without administration of substrate was 525.6 ± 12.7 and 123.2 ± 24.3, respectively. 2-Furfurylthiol The results of the measurement of 2-FT and FMS under condition (i) are shown in Figure 3B,C. By injecting 2-FT-containing air into one nostril, FMS was detected in the flow-out air from the other nostril. FMS was detected in the exhaled air almost simultaneously with the 2-FT detection start time. The intersubject variabilities of the mean maximum concentration of exhaled 2-FT and FMS were approximately 4- and 4-fold, respectively. The results of the measurements of 2-FT and FMS under condition (ii) are shown in Figure 3E,F. In this assay, after inhaling 2-FT containing air through the nose, FMS was detected in the exhaled air from the nose. The maximum concentration of 2-FT in the exhaled air was observed in the first exhaled breath. The concentration then decreased rapidly and the mean concentration in the second exhaled breath was 19.5% ± 9.5% of that in the first exhaled breath (Figure 3E1–E5). It then continued to decrease gradually, although 2-FT was detectible until the tenth exhaled breath. FMS was detected in the first exhaled breath. The average level in the second exhaled breath decreased to 82.6% ± 25.0% of that in the first exhaled breath (Figure 3F1–F5). FMS was detectable until the tenth exhaled breath. The intersubject variabilities for the maximum concentration of 2-FT and FMS were approximately 3- and 2-fold, respectively. Under condition (iii) (Figure 3H,I), the conversion of 2-FT into FMS was also observed. The concentration of 2-FT in the exhaled air reached a peak in the first exhaled breath, and then decreased rapidly. FMS was detected in the first exhaled breath. The waveform was identified up to 10 exhaled breaths in all subjects. The intersubject variabilities for the maximum concentrations of 2-FT and FMS were approximately 3- and 3-fold, respectively. Hexanal Real-time metabolic analysis of hexanal using PTR-MS was carried out in a manner similar to that used for 2-FT (Figure 3). As with 2-FT, by injecting hexanal from the nostril under condition (i), hexanol was detected in the flow-out air (Figure 4B,C). Although the timing when the concentration of hexanal in the exhaled air reached a peak varied depending on the subjects, the appearance of hexanol in the exhaled air was almost simultaneous to that of hexanal. In one of the subjects (subject 2), hexanol was not detected (Figure 4C2). The intersubject variabilities of the exhaled hexanal and hexanol concentrations were approximately 4- and 5-fold (except subject 2), respectively. Under condition (ii), the conversion of hexanal into hexanol was also observed (Figure 4E,F). The maximum concentration of hexanal in the exhaled air was observed in the first exhaled breath, and no obvious peaks were found in the second and later exhaled breaths. Intersubject variabilities of the maximum hexanal and hexanol concentrations were approximately 29- and 3-fold, respectively. Although the level of hexanol decreased rapidly from the second exhaled breath onward, a waveform was observed until the 10th breath. The conversion of hexanal into hexanol was also observed under condition (iii) (Figure 4H,I). Hexanal at the highest level was observed in the first exhaled breath and decreased rapidly, but a waveform for hexanol was observed until the 10th breath as observed for condition (ii). Intersubject variabilities of the maximum concentrations of hexanal and hexanol were approximately 7- and 5-fold, respectively. Benzyl acetate We performed in vivo metabolite analysis of benzyl acetate similar to that described for 2-FT and hexanal. However, the benzyl alcohol parent peak overlapped with a benzyl acetate fragment ion peak and, thus, the substrate and metabolite could not be distinguished, preventing measurement. The PTR-MS charts of benzyl acetate and benzyl alcohol are shown in Supplementary Figure 4. Methyl raspberry ketone Because we could not detect the metabolism of methyl raspberry ketone into raspberry ketone in vitro, we did not perform in vivo PTR-MS analysis. Metabolite concentrations in exhaled air and human odor perception We examined whether FMS and hexanol at concentrations in the exhaled air were above the perception threshold. Based on the above analysis showing that the peak concentrations of FMS and hexanol in conditions (i), (ii), and (iii) were 20, 100, and 80 ppb and 250, 2000, and 1000 ppb, respectively (Figures 3 and 4); we presented FMS and hexanol at 20 and 60 ppb and 250 and 750 ppb, respectively, to the subjects (Table 3). As for FMS, among the 20 subjects, 14 reported +++ (definite odor; quality of the odor can be described definitely) and 4 reported ++ (suprarecognition threshold; quality of the odor can be described) at 20 ppb, whereas 16 reported +++ and 4 reported ++ at 60 ppb. For hexanol, 3 reported +++ and 12 reported ++ at 250 ppb, whereas 12 reported +++ and 6 reported ++ at 750 ppb. These results indicate that the concentration reached the level for odor perception. Table 3. Odor perception of the metabolite compounds by concentrations Intensity FMS (ppb) Hexanol (ppb) 20 60 250 750 No odor (Under detection threshold) − 0 0 1 0 Supradetection threshold + 2 0 4 2 Suprarecognition threshold ++ 4 4 12 6 Definite odor +++ 14 16 3 12 Intensity FMS (ppb) Hexanol (ppb) 20 60 250 750 No odor (Under detection threshold) − 0 0 1 0 Supradetection threshold + 2 0 4 2 Suprarecognition threshold ++ 4 4 12 6 Definite odor +++ 14 16 3 12 Open in new tab Table 3. Odor perception of the metabolite compounds by concentrations Intensity FMS (ppb) Hexanol (ppb) 20 60 250 750 No odor (Under detection threshold) − 0 0 1 0 Supradetection threshold + 2 0 4 2 Suprarecognition threshold ++ 4 4 12 6 Definite odor +++ 14 16 3 12 Intensity FMS (ppb) Hexanol (ppb) 20 60 250 750 No odor (Under detection threshold) − 0 0 1 0 Supradetection threshold + 2 0 4 2 Suprarecognition threshold ++ 4 4 12 6 Definite odor +++ 14 16 3 12 Open in new tab Relation between OR activities of an odorant and its metabolite To investigate how the odorants and their metabolites are recognized at the receptor level, human ORs that respond to 2-FT or FMS, each at 300 μM, were screened. Of the 403 human ORs analyzed, 2-FT activated 3 ORs and FMS activated 5 ORs. These ORs partially overlapped: OR2C1, OR2W1, and OR1A1 were commonly activated by both compounds (Figure 5A–C, and Supplementary Figure 5). Based on this data, we selected the following odor compounds that fit the following 3 conditions (Figure 5 and Supplementary Figure 5): (i) Dimethyl sulfide (DMS), a compound that activates an OR activated by FMS (OR4S2), but does not activate ORs activated by 2-FT; (ii) 2-Pentylpryidine (2-PP), a compound that activates ORs activated by both of 2-FT and FMS (OR2W1, OR1A1, and OR2L8); and (iii) Furfural, a compound that does not activate ORs activated by 2-FT or FMS but has a molecular structure similar to that of FMS. The dose-dependence of the OR activity toward these compounds was measured. This analysis confirmed the above screening data (Figure 5A–F). Figure 5. Open in new tabDownload slide (A)–(E) Activity of selected human ORs (OR2C1, OR2W1, OR1A1, OR2L8, and OR4S2) against 2-FT at concentration of 0–300 μM, FMS, DMS, 2-PP, and furfural at concentrations of 0–1000 μM. The tests were carried out at n = 3. Decision of whether a receptor showed dose-dependent response against an odorant or not was according to the criteria established by Mainland et al. (see Materials and methods). A magnified view of graph B is shown in B’. (F) Summary of the data in (A)–(E) regarding the magnitude of response of 5 odorant receptors against 2-FT, FMS, DMS, 2-PP, and furfural. The greater the number of “+”, the lower the concentration at which the normalized response of the receptors exceeded 2, that is, ++++: ~1 μM, +++: 1–10 μM, ++: 10–100 μM, and +: 100–1000 μM. (G) Concept of cross-adaptation study using 2-FT. The table shows the predicted magnitude of OR activation when 2-FT is secondarily presented in each of the following 5 odor presentation patterns: (i) Smelling 2-FT after the presentation of water (control). Under this condition, all 5 ORs should be activated by 2-FT presentation. (ii) Smelling 2-FT after presentation of FMS. Under this condition, all ORs activated by FMS should be desensitized and less responsive when the second 2-FT is presented. Therefore, if the subject typically recognizes the odor of a mixture of FMS and 2-FT as a smell of “2-FT,” the subjective overall odor intensity and/or odor quality of “pungent, garlic-like odor,” and “egg-like sulfurous odor,” characteristic description of 2-FT odor quality should be greatly reduced when 2-FT is presented. (iii) Smelling 2-FT after presentation of DMS. Under this condition, only a part of the OR(s) activated by FMS (i.e., OR4S2) should be desensitized and less responsive when the second 2-FT is presented. Therefore, the subjective odor intensity and/or quality of “pungent, garlic-like odor,” “egg-like sulfurous odor” characterized by FMS should be reduced but not greatly as in (ii) or (iv). (iv) Smelling 2-FT after presentation of 2-PP. Under this condition, 3 ORs commonly activated by 2-FT and FMS (i.e., OR2W1, OR1A1, and OR2L8) should be desensitized. Therefore, the odor quality of “pungent, garlic-like odor,” characterized by FMS should be reduced more than in (iii) but less than in (ii). (v) Smelling 2-FT after adaptation to furfural. Prior furfural presentation does not desensitize any ORs activated by 2-FT or FMS. Therefore, the subjective odor quality when 2-FT is presented should be similar to that in (i). (H)–(J) Results of cross-adaptation study. The odor quality when smelling 2-FT under each of the conditions listed below was evaluated using a VAS for (H) overall odor intensity; (I) intensity of pungent, garlic-like odor; (J) intensity of egg-like, sulfurous odor. The experiments were repeated twice for each subject (n = 10). The columns indicate the following presentation patterns corresponding to (i)–(v) in (G): (i) Water (control): smelling 2-FT after the presentation of water. (ii) FMS: smelling 2-FT after the presentation of FMS. (iii) DMS: smelling 2-FT after the presentation of DMS. (iv)2-PP: smelling 2-FT after the presentation of 2-PP. (v) Furfural: smelling 2-FT after the presentation of furfural. The y axes indicate the mean VAS values for the relevant items, whereas the vertical bars represent standard errors. Asterisks indicate a significant difference (*P < 0.05, **P < 0.01) compared with the water control group (Dunnett post hoc test). Figure 5. Open in new tabDownload slide (A)–(E) Activity of selected human ORs (OR2C1, OR2W1, OR1A1, OR2L8, and OR4S2) against 2-FT at concentration of 0–300 μM, FMS, DMS, 2-PP, and furfural at concentrations of 0–1000 μM. The tests were carried out at n = 3. Decision of whether a receptor showed dose-dependent response against an odorant or not was according to the criteria established by Mainland et al. (see Materials and methods). A magnified view of graph B is shown in B’. (F) Summary of the data in (A)–(E) regarding the magnitude of response of 5 odorant receptors against 2-FT, FMS, DMS, 2-PP, and furfural. The greater the number of “+”, the lower the concentration at which the normalized response of the receptors exceeded 2, that is, ++++: ~1 μM, +++: 1–10 μM, ++: 10–100 μM, and +: 100–1000 μM. (G) Concept of cross-adaptation study using 2-FT. The table shows the predicted magnitude of OR activation when 2-FT is secondarily presented in each of the following 5 odor presentation patterns: (i) Smelling 2-FT after the presentation of water (control). Under this condition, all 5 ORs should be activated by 2-FT presentation. (ii) Smelling 2-FT after presentation of FMS. Under this condition, all ORs activated by FMS should be desensitized and less responsive when the second 2-FT is presented. Therefore, if the subject typically recognizes the odor of a mixture of FMS and 2-FT as a smell of “2-FT,” the subjective overall odor intensity and/or odor quality of “pungent, garlic-like odor,” and “egg-like sulfurous odor,” characteristic description of 2-FT odor quality should be greatly reduced when 2-FT is presented. (iii) Smelling 2-FT after presentation of DMS. Under this condition, only a part of the OR(s) activated by FMS (i.e., OR4S2) should be desensitized and less responsive when the second 2-FT is presented. Therefore, the subjective odor intensity and/or quality of “pungent, garlic-like odor,” “egg-like sulfurous odor” characterized by FMS should be reduced but not greatly as in (ii) or (iv). (iv) Smelling 2-FT after presentation of 2-PP. Under this condition, 3 ORs commonly activated by 2-FT and FMS (i.e., OR2W1, OR1A1, and OR2L8) should be desensitized. Therefore, the odor quality of “pungent, garlic-like odor,” characterized by FMS should be reduced more than in (iii) but less than in (ii). (v) Smelling 2-FT after adaptation to furfural. Prior furfural presentation does not desensitize any ORs activated by 2-FT or FMS. Therefore, the subjective odor quality when 2-FT is presented should be similar to that in (i). (H)–(J) Results of cross-adaptation study. The odor quality when smelling 2-FT under each of the conditions listed below was evaluated using a VAS for (H) overall odor intensity; (I) intensity of pungent, garlic-like odor; (J) intensity of egg-like, sulfurous odor. The experiments were repeated twice for each subject (n = 10). The columns indicate the following presentation patterns corresponding to (i)–(v) in (G): (i) Water (control): smelling 2-FT after the presentation of water. (ii) FMS: smelling 2-FT after the presentation of FMS. (iii) DMS: smelling 2-FT after the presentation of DMS. (iv)2-PP: smelling 2-FT after the presentation of 2-PP. (v) Furfural: smelling 2-FT after the presentation of furfural. The y axes indicate the mean VAS values for the relevant items, whereas the vertical bars represent standard errors. Asterisks indicate a significant difference (*P < 0.05, **P < 0.01) compared with the water control group (Dunnett post hoc test). Cross-adaptation study To examine whether metabolism affects odor perception, a cross adaptation using these compounds was designed as summarized in Figure 5G. Some 2-FT is thought to be converted to FMS in the nasal/oral cavity. Therefore, if such FMS affects the perception of the odor of 2-FT, prior presentation of odor compounds that activate OR(s) activated by FMS (i.e., FMS itself, DMS, and 2-PP) would desensitize these OR(s), leading to a reduced intensity and modified quality of 2-FT, that is, the overall intensity of the odor and/or quality of “pungent garlic-like odor” or “egg-like sulfurous odor” may change. The results of cross-adaptation tests (1)–(5) are shown in Figure 5H–J. When the odor quality of 2-FT after the presentation of FMS in (2) was compared with the control in (1), the intensities of both overall odor and “pungent garlic-like odor” were decreased. The odor of 2-FT after smelling DMS in (3), as in the case of (2), showed a decreased intensity of overall odor. However, in (3), there was no significant change in intensity of “pungent garlic-like odor” compared with in (1), suggesting that the adaptation with DMS was weaker than that with FMS. The odor of 2-FT after smelling 2-PP in (4) also showed a decreased overall odor intensity, but not for “pungent garlic-like odor,” suggesting that its effect was weaker than that with FMS. The odor of 2-FT, after smelling furfural in (5), showed no change in intensity for either overall odor or “pungent garlic-like odor,” indicating a lack of cross adaptation. The above results confirmed the cross adaptations that were predicted from the OR activity profiles (Figure 5F,G). This suggests that the FMS formed by the metabolism of 2-FT affects the perception of the odor of 2-FT; that is, when humans smell “2-FT,” the odor recognized may be that of a mixture of FMS and 2-FT. Discussion Our results are summarized as follows: (i) In vitro metabolism assays revealed that the saliva and nasal mucus exhibit enzymatic activity in the metabolism of 2-FT, hexanal, and benzyl acetate; (ii) real-time in vivo analysis using PTR-MS demonstrated that the application of 2-FT and hexanal through 3 different pathways via the nostril or through the mouth generated their metabolites rapidly at concentrations above the detection threshold and, in most cases, above the recognition threshold; (iii) the cross-adaptation test suggested that the metabolism of 2-FT affects odor perception. These results suggest that some odorant compounds in food can be metabolized in the human nasal mucus and that each food odor can be recognized as a mixture of the original odor component and its metabolites. Based on our results, we suggest 3 possibilities for the process of metabolism of odorants in the upper airway: first, odorants inhaled through the orthonasal route may be absorbed by the respiratory nasal mucus, where they are metabolized and revaporized and then reach the olfactory mucosa; second, the odorants may be absorbed directly in the olfactory mucus, where they are metabolized and bind to the ORs; third, the retronasally inhaled odorants may be metabolized in the saliva and/or nasal mucus. Thus, odorant metabolism occurs via both routes, orthonasal and retronasal. Particularly, metabolism is a continuous process, even when food is held in the oral cavity. Methylation of thiols Methylation of thiol compounds is a biological reaction that is well known to occur in mammals and microorganisms, in which methyl groups are transferred by thiol methyltransferases, with S-(5′-adenosyl)-L-methionine chloride dihydrochloride (SAM) as a primary methyl donor. In relation to the oral cavity, numerous reports have described the activity of thiol-S-methyltransferase in rat submaxillary gland microsomes (Yashiro and Takatsu 2001). Itobe et al. reported that air exhaled after swallowing an aqueous solution of 4-methoxy-2-methyl-2-mercaptobutane contains its methylated compound, methoxy-3- methyl-3-(methylthio) butane (Itobe et al. 2009). Buettner also showed that the quantities of the thiol compounds 2-phenyl-ethanethiol, 2-FT, and 3-mercapto-3-methyl-1-butanol decreased when mixed with human saliva (Buettner 2002a, 2002b). In the present in vitro study using 2-FT, human saliva and nasal mucus (inferior part) showed enzymatic activity to methylate 2-FT and generate FMS. To determine whether this metabolic reaction proceeds quickly enough to affect odor recognition in vivo, we performed real-time metabolite analysis of exhaled air using PTR-MS. FMS was detected in the air at the beginning of the flow out from the contralateral nostril (condition (i)) or in the first exhaled air from the nostril (conditions (ii) and (iii)). Moreover, the average of peak concentration of FMS detected under condition (i) was 20 ppb, which was above the recognition threshold. These results suggest that the rate and amount of in vivo odorant metabolism by the mucus is sufficient to affect odorant perception. Although FMS was clearly detected in in vivo studies, the level of FMS detected considerably lower than the amount of 2-FT consumed in vitro. The reason for this finding is unclear, but we speculated the following 2 possibilities: one possibility is that the substrate bound to the native protein in the mucus, which prevented its volatilization and detection in GC-MS. The other possibility is that 2-FT and hexanal may be converted to another compound only under in vitro conditions. We confirmed that peaks detected in GC-MS chart were only those corresponding to FMS and hexanol. However, the possibility that another metabolite compound is present cannot be excluded as some compounds may not be detected by GC/MS. Reduction of aldehydes Itobe et al. reported that when the subjects swallowed aqueous hexanal solution, the reduction product hexanol was detected in the exhaled air (Itobe et al. 2009). Additionally, an in vitro study by Buettner revealed that the concentrations of hexanal, methional, and other aldehydes decreased when mixed with saliva, suggesting that they may be metabolized (Buettner 2002a, 2002b). In a previous study, aldehydes were converted to alcohols in the murine olfactory mucus, and the reaction was partially inhibited by the aldehyde reductase inhibitor 3,5-dichlorosalicylic acid (Nagashima and Touhara 2010). These findings are consistent with those of our in vitro study which showed that hexanal was converted to hexanol by the saliva and nasal mucus. Results of in vivo PTR-MS using hexanal demonstrated the average of peak concentration of hexanol detected under condition (i) was 250 ppb, which was above the recognition threshold, suggesting that the amount of in vivo hexanol production is sufficient to affect odorant perception as in the case of FMS. As observed in the reaction with 2-FT, although hexanol was clearly detected in in vivo studies, the amount of hexanol detected was very low in in vitro studies. Similar possibilities as described above regarding the discrepancy between in vitro 2-FT and FMS concentrations may explain these findings. 2-FT was not metabolized by nasal mucus (superior part), whereas hexanal showed the opposite result. This may be for the following reasons: (i) the metabolic enzymes involved have different distributions and/or (ii) saliva and nasal mucus (inferior part) were collected without dilution, whereas nasal mucus (superior part) was diluted with saline during collection. Therefore, its activity may have been weaker than its in vivo activity. In real-time metabolic analysis, when hexanal-containing headspace air was inhaled followed immediately by nasal breathing, a clear peak was observed for the substrate, hexanal, in the first exhaled breath. However, elimination then progressed rapidly with nearly no peaks being detected in the second and subsequent exhaled breaths. In contrast, peaks corresponding to hexanol were observed in all exhaled breaths. This may be because in the case of aldehyde, the substrate dissolves in the mucus rapidly and undergoes reduction, forming alcohols that are gradually volatilized and thus detected, whereas the water/air distribution coefficient logKwa for hexanol is low, so the volatilization might be continuous. A similar tendency with respect to exhaled air was observed when exhaled air was analyzed after swallowing aqueous hexanal solution. It may be that the substrate in the stomach or that which remained in the oral cavity was metabolized and the metabolite was detected in exhaled air. In one subject (subject 2), hexanol production was not detected in the real-time PTR-MS study of condition (i). PTR-MS study in condition (ii) and (iii) in that subject did detect hexanol, indicating that the subject has metabolic activity for hexanal, at least in the throat/trachea/lung region. The reason for this discrepancy is unclear. Hydrolysis of esters According to a patent published by a flavor and fragrance company in 2008, Givaudan showed that styrallyl acetate and phenethyl acetate are metabolized by hydrolysis in the nasal and oral cavities (Givaudan 2008; Schilling 2010). It was also reported that the levels of ethyl butanoate and other esters were decreased in the saliva, suggesting that they were metabolized in the saliva (Buettner 2002a, 2002b). In a study using murine mucus, acetyl isoeugenol undergoes hydrolysis, which was inhibited by bis-(p-nitrophenyl) phosphate, a carboxylesterase-inhibitor (Nagashima and Touhara 2010). In the present study, when benzyl acetate was incubated with saliva and nasal mucus (inferior part), its hydrolyzed metabolite benzyl alcohol was detected, but this metabolic activity was not observed in the nasal mucus (superior part). Carboxylesterases were not identified even in proteome analysis of human olfactory mucus (Débat et al. 2007). When the nasal mucus was collected from mice, the material collected is a mixture of olfactory mucus and respiratory epithelial mucus, indicating that respiratory mucus exhibits metabolic activity. In our experiment, hydrolysis of benzyl acetate was observed after the reaction of the harvested saliva and nasal mucus (inferior part) in vitro. However, we could not detect metabolite generation, as PTR-MS could not differentiate the peaks of benzyl acetate (substrate) and benzyl alcohol (metabolite). Demethylation of methoxy groups 2-Methoxyacetophenone is demethylated by enzymatic activity in the human nasal cavity, which is inhibited by a volatile inhibitor (Givaudan 2008). Methyl raspberry ketone is similarly demethylated to form raspberry ketone, which has a stronger raspberry odor. Application of an inhibitor into the nose reduces the intensity of the raspberry odor (Givaudan 2008; Schilling 2010). We attempted to detect the metabolite (raspberry ketone) under different in vitro conditions, but detection was unsuccessful. The reason for this discrepancy is unclear. One possibility for this failure is that the enzyme that metabolizes methyl raspberry ketone (probably CYP13A2; Givaudan SA patent no 2008–506958) may be anchored in the cell membrane and, therefore, metabolism is only observed in vivo. Cross-adaptation study We observed that FMS in the exhaled air was above the perception threshold, suggesting that when “2-FT” is smelled, FMS formed by the metabolism of 2-FT contributes to the odor quality of “2-FT.” This possibility was evaluated using a cross-adaptation method. The subject smelled 2-FT after smelling each of the 4 odorants, FMS, DMS, 2-PP, and furfural, which showed a distinct activation capacity toward 2-FT-related ORs. For FMS, the odor of 2-FT decreased in intensity and changed in quality. Prior 2-PP presentation also decreased the intensity of the odor of 2-FT. In contrast, only weak cross adaptation was observed with DMS and no adaptation was found with furfural. These results corresponded well with the changes in odor intensity/quality predicted by the OR activity patterns. Our findings indicate that the odor of the metabolite FMS affects the perception of the odor of 2-FT and that cross adaptation depends on OR activity. Conclusion In conclusion, the present study demonstrated that metabolism of the following odorant molecules occurs in human saliva/nasal mucus in vitro: (i) methylation of 2-FT into FMS; (ii) reduction of hexanal into hexanol; and (iii) hydrolysis of benzyl acetate into benzyl alcohol. Real-time in vivo analysis using PTR-MS demonstrated that application of 2-FT and hexanal through 3 different pathways via the nostril or through the mouth generated their metabolites, FMS, and hexanol within a few seconds. The concentrations of FMS and hexanol in the exhaled air was above the perception threshold. A cross-adaptation study based on the activation pattern of human odorant receptors indicated that this metabolism affects odor perception. These results suggest that some odorants in food are metabolized in the human nasal mucus/saliva, and the resulting metabolites are perceived as a part of the odor quality of substrates. Acknowledgements The authors thank Akira Nakayama, Ajinomoto Co., Inc., for support and advice regarding metabolite measurement methods; Kazutaka Shimbo and Ayaka Shirasawa, Ajinomoto Co., Inc., for proteome analysis; Tetsuya Koyama, EA Pharma Co., Ltd, for advice regarding metabolic enzyme tests; Dr Hiroya Kawasaki, Ajinomoto Co., Inc., for advice regarding cross-adaptation studies; all colleagues at FAR-O, Ajinomoto Co., Inc., for their help with the cross-adaptation study; and Dr Shu Kikuta and Dr Hironobu Nishijima, Department of Otorhinolaryngology, The University of Tokyo, for collecting nasal mucus and drawing of the schematic view of experiments. Funding This work was supported by Ajinomoto Co., Inc. and by ERATO Touhara Chemosensory Signal Project (grant number 501100009024; JPMJER1202) and JST-Mirai Program (grant number 501100002241; JPMJMI17DC) from JST, Japan, to KT. Conflict of interest: C.I., H.W., S.S., Y.I., and Y.N. are employees of Ajinomoto Co., Inc. received research support from Ajinomoto Co., Inc. The remaining authors declare that they have no relevant conflicts of interest. JST had no influence over the study conception, design or interpretation, or decision to publish the data. References Asakawa M , Fukutani Y , Savangsuksa A , Noguchi K , Matsunami H , Yohda M . 2017 . Modification of the response of olfactory receptors to acetophenone by CYP1a2 . Sci Rep . 7 : 10167 . Google Scholar Crossref Search ADS PubMed WorldCat Ben-Arie N , Khen M , Lancet D . 1993 . Glutathione S-transferases in rat olfactory epithelium: purification, molecular properties and odorant biotransformation . Biochem J . 292 ( Pt 2 ): 379 – 384 . Google Scholar Crossref Search ADS PubMed WorldCat Buettener A (editors). 2017 . Springer handbook of odor . New York (NY) : Springer International Publishing . Google Preview WorldCat COPAC Buettner A . 2002a . Influence of human salivary enzymes on odorant concentration changes occurring in vivo. 1. Esters and thiols . J Agric Food Chem . 50 : 3283 – 3289 . Google Scholar Crossref Search ADS WorldCat Buettner A . 2002b . Influence of human saliva on odorant concentrations. 2. aldehydes, alcohols, 3-alkyl-2-methoxypyrazines, methoxyphenols, and 3-hydroxy-4,5-dimethyl-2(5H)-furanone . J Agric Food Chem . 50 : 7105 – 7110 . Google Scholar Crossref Search ADS WorldCat Buettner A , Beer A , Hannig C , Settles M . 2001 . Observation of the swallowing process by application of videofluoroscopy and real-time magnetic resonance imaging-consequences for retronasal aroma stimulation . Chem Senses . 26 : 1211 – 1219 . Google Scholar Crossref Search ADS PubMed WorldCat Casado B , Pannell LK , Iadarola P , Baraniuk JN . 2005 . Identification of human nasal mucous proteins using proteomics . Proteomics . 5 : 2949 – 2959 . Google Scholar Crossref Search ADS PubMed WorldCat Dahl AR , Hadley WM . 1991 . Nasal cavity enzymes involved in xenobiotic metabolism: effects on the toxicity of inhalants . Crit Rev Toxicol . 21 : 345 – 372 . Google Scholar Crossref Search ADS PubMed WorldCat Débat H , Eloit C , Blon F , Sarazin B , Henry C , Huet JC , Trotier D , Pernollet JC . 2007 . Identification of human olfactory cleft mucus proteins using proteomic analysis . J Proteome Res . 6 : 1985 – 1996 . Google Scholar Crossref Search ADS PubMed WorldCat Dunkel A , Steinhaus M , Kotthoff M , Nowak B , Krautwurst D , Schieberle P , Hofmann TA . 2014 . Nature’s Chemical signatures in human olfaction: a foodborne perspective for future biotechnology . Angew Chem Int Ed Engl . 53 : 7124 – 7143 . Google Scholar Crossref Search ADS PubMed WorldCat Fors S . 1988 . The Maillard reaction in foods and nutrition. In: Waller GR , Feather MS , editors. ACS Symposium Series 215. Washington (DC) : ACS . p. 185 – 286 . COPAC Frasnelli J , van Ruth S , Kriukova I , Hummel T . 2005 . Intranasal concentrations of orally administered flavors . Chem Senses . 30 : 575 – 582 . Google Scholar Crossref Search ADS PubMed WorldCat Givaudan SA . 2008 . WO 2006/007752 A1 (Chem. Abstr. 2006, 144, 134740) Hodgson M , Linforth RS , Taylor AJ . 2003 . Simultaneous real-time measurements of mastication, swallowing, nasal airflow, and aroma release . J Agric Food Chem . 51 : 5052 – 5057 . Google Scholar Crossref Search ADS PubMed WorldCat Itobe T , Kumazawa K , Nishimura O . 2009 . New factor characterizing the in-mouth release of odorants (volatile thiols): compositional changes in odorants exhaled from the human nose during drinking . J Agric Food Chem . 57 : 11297 – 11301 . Google Scholar Crossref Search ADS PubMed WorldCat Kaliner MA . 1991 . Human nasal respiratory secretions and host defense . Am Rev Respir Dis . 144 : S52 – S56 . Google Scholar Crossref Search ADS PubMed WorldCat Katada S , Hirokawa T , Oka Y , Suwa M , Touhara K . 2005 . Structural basis for a broad but selective ligand spectrum of a mouse olfactory receptor: mapping the odorant-binding site . J Neurosci . 25 : 1806 – 1815 . Google Scholar Crossref Search ADS PubMed WorldCat Kida H , Fukutani Y , Mainland JD , de March CA , Vihani A , Li YR , Chi Q , Toyama A , Liu L , Kameda M , et al. 2018 . Vapor detection and discrimination with a panel of odorant receptors . Nat Commun . 9 : 4556 . Google Scholar Crossref Search ADS PubMed WorldCat Linforth R , Martin F , Carey M , Davidson J , Taylor AJ . 2002 . Retronasal transport of aroma compounds . J Agric Food Chem . 50 : 1111 – 1117 . Google Scholar Crossref Search ADS PubMed WorldCat Ling G , Gu J , Genter MB , Zhuo X , Ding X . 2004 . Regulation of cytochrome P450 gene expression in the olfactory mucosa . Chem Biol Interact . 147 : 247 – 258 . Google Scholar Crossref Search ADS PubMed WorldCat Mainland JD , Li YR , Zhou T , Liu WL , Matsunami H . 2015 . Human olfactory receptor responses to odorants . Sci Data . 2 : 150002 . Google Scholar Crossref Search ADS PubMed WorldCat McGinty D , Vitale D , Letizia CS , Api AM . 2012 . Fragrance material review on benzyl acetate . Food Chem Toxicol . 50 ( Suppl 2 ): S363 – S384 . Google Scholar Crossref Search ADS PubMed WorldCat Mörtstedt H , Kåredal MH , Jönsson BA , Lindh CH . 2013 . Screening method using selected reaction monitoring for targeted proteomics studies of nasal lavage fluid . J Proteome Res . 12 : 234 – 247 . Google Scholar Crossref Search ADS PubMed WorldCat Mosciano G , Fasano M , Allen BB , Michalski J , Sadural S . 1990 . Organoleptic characteristics of flavor materials . Perfumer and Flavorist . 15 : 59 . WorldCat Murnane SS , Lehocky AH , Owens PD . 2013 . Odor thresholds for chemicals with established health standards . 2 nd ed. Falls Church (VA) : AIHA Publications . Google Preview WorldCat COPAC Nagashima A , Touhara K . 2010 . Enzymatic conversion of odorants in nasal mucus affects olfactory glomerular activation patterns and odor perception . J Neurosci . 30 : 16391 – 16398 . Google Scholar Crossref Search ADS PubMed WorldCat Nef P , Heldman J , Lazard D , Margalit T , Jaye M , Hanukoglu I , Lancet D . 1989 . Olfactory-specific cytochrome P-450. cDNA cloning of a novel neuroepithelial enzyme possibly involved in chemoreception . J Biol Chem . 264 : 6780 – 6785 . Google Scholar PubMed WorldCat Niimura Y , Matsui A , Touhara K . 2018 . Acceleration of olfactory receptor gene loss in primate evolution: possible link to anatomical change in sensory systems and dietary transition . Mol Biol Evol . 35 : 1437 – 1450 . Google Scholar Crossref Search ADS PubMed WorldCat Niimura Y , Nei , M . 2003 . Evolution of olfactory receptor genes in the human genome . Proc Natl Acad Sci USA . 100 : 12235 – 12240 . Google Scholar Crossref Search ADS PubMed WorldCat Nijssen LM , van Ingen-Visscher CA , Donders JJ . 2016 . VCF volatile compounds in food database version 16.3 . Zeist (the Netherlands) : TNO Triskelion . Google Preview WorldCat COPAC Oka Y , Katada S , Omura M , Suwa M , Yoshihara Y , Touhara K . 2006 . Odorant receptor map in the mouse olfactory bulb: in vivo sensitivity and specificity of receptor-defined glomeruli . Neuron . 52 : 857 – 869 . Google Scholar Crossref Search ADS PubMed WorldCat Olender T , Waszak SM , Viavant M , Khen M , Ben-Asher E , Reyes A , Nativ N , Wysocki CJ , Ge D , Lancet D . 2012 . Personal receptor repertoires: olfaction as a model . BMC Genomics . 13 : 414 . Google Scholar Crossref Search ADS PubMed WorldCat Poivet E , Peterlin Z , Tahirova N , Xu L , Altomare C , Paria A , Zou DJ , Firestein S . 2016 . Applying medicinal chemistry strategies to understand odorant discrimination . Nat Commun . 7 : 11157 . Google Scholar Crossref Search ADS PubMed WorldCat Ritz C , Baty F , Streibig JC , Gerhard D . 2015 . Dose-response analysis using R . PLoS One . 10 : e0146021 . Google Scholar Crossref Search ADS PubMed WorldCat Robert-Hazotte A , Faure P , Neiers F , Potin C , Artur Y , Coureaud G , Heydel JM . 2019a . Nasal mucus glutathione transferase activity and impact on olfactory perception and neonatal behavior . Sci Rep . 9 : 3104 . Google Scholar Crossref Search ADS WorldCat Robert-Hazotte A , Schoumacker R , Semon E , Briand L , Guichard E , Le Quéré JL , Faure P , Heydel JM . 2019b . Ex vivo real-time monitoring of volatile metabolites resulting from nasal odorant metabolism . Sci Rep . 9 : 2492 . Google Scholar Crossref Search ADS WorldCat Rubesin SE , Rabischong P , Bilaniuk LT , Laufer I , Levine MS . 1988 . Contrast examination of the soft palate with cross sectional correlation . Radiographics . 8 : 641 – 665 . Google Scholar Crossref Search ADS PubMed WorldCat Rychlik M . 2017 . Metabolism of odorants in humans . In: Buettner A , editor. Springer handbook of odor . Heidelberg (Germany) : Springer . p. 617 – 629 . Google Preview WorldCat COPAC Schilling B . 2017 . Nasal periceptor processes . In: Buettner A , editor. Springer handbook of odor . Heidelberg (Germany) : Springer . p. 605 – 616 . Google Preview WorldCat COPAC Schilling B , Kaiser R , Natsch A , Gautschi M . 2010 . Investigation of odors in the fragrance industry . Chemoecology . 20 : 135 – 147 . Google Scholar Crossref Search ADS WorldCat Shirasu M , Yoshikawa K , Takai Y , Nakashima A , Takeuchi H , Sakano H , Touhara K . 2014 . Olfactory receptor and neural pathway responsible for highly selective sensing of musk odors . Neuron . 81 : 165 – 178 . Google Scholar Crossref Search ADS PubMed WorldCat The 1000 Genome Project Consortium . 2015 . A global reference for human genetic variation . Nature . 526 : 68 – 74 . Crossref Search ADS PubMed WorldCat Thiebaud N , Veloso Da Silva S , Jakob I , Sicard G , Chevalier J , Ménétrier F , Berdeaux O , Artur Y , Heydel JM , Le Bon AM . 2013 . Odorant metabolism catalyzed by olfactory mucosal enzymes influences peripheral olfactory responses in rats . PLoS One . 8 : e59547 . Google Scholar Crossref Search ADS PubMed WorldCat Tonosaki K . 1993 . Cross-adaptation to odor stimulation of olfactory receptor cells in the box turtle, Terrapene carolina . Brain Behav Evol . 41 : 187 – 191 . Google Scholar Crossref Search ADS PubMed WorldCat Yashiro K , Takatsu F . 2001 . Microsomal thiol S-methyltransferase activity in rat salivary glands . Jpn J Oral Biol . 43 : 133 – 139 (Japanese). Google Scholar Crossref Search ADS WorldCat Yoshikawa K , Wang H , Jaen C , Haneoka M , Saito N , Nakamura J , Adappa ND , Cohen NA , Dalton P . 2018 . The human olfactory cleft mucus proteome and its age-related changes . Sci Rep . 8 : 17170 . Google Scholar Crossref Search ADS PubMed WorldCat Zhuang H , Matsunami H . 2008 . Evaluating cell-surface expression and measuring activation of mammalian odorant receptors in heterologous cells . Nat Protoc . 3 : 1402 – 1413 . Google Scholar Crossref Search ADS PubMed WorldCat Zupko K , Poria Y , Lancet D . 1991 . Immunolocalization of cytochromes P-450olf1 and P-450olf2 in rat olfactory mucosa . Eur J Biochem . 196 : 51 – 58 . Google Scholar Crossref Search ADS PubMed WorldCat © The Author(s) 2019. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Metabolism of Odorant Molecules in Human Nasal/Oral Cavity Affects the Odorant Perception
JO - Chemical Senses
DO - 10.1093/chemse/bjz041
DA - 2019-09-07
UR - https://www.deepdyve.com/lp/oxford-university-press/metabolism-of-odorant-molecules-in-human-nasal-oral-cavity-affects-the-QA7BeJOdAT
SP - 465
VL - 44
IS - 7
DP - DeepDyve
ER -